PHARMACOLOGY - RESEARCH, SAFETY TESTING AND REGULATION

DRY POWDER INHALERS

FORMULATION, DEVICE
AND CHARACTERIZATION

No part of this digital document may be reproduced, stored in a retrieval system or transmitted in any form or
by any means. The publisher has taken reasonable care in the preparation of this digital document, but makes no
expressed or implied warranty of any kind and assumes no responsibility for any errors or omissions. No
liability is assumed for incidental or consequential damages in connection with or arising out of information
contained herein. This digital document is sold with the clear understanding that the publisher is not engaged in
rendering legal, medical or any other professional services.
PHARMACOLOGY - RESEARCH, SAFETY
TESTING AND REGULATION

Additional books in this series can be found on Nova’s website

under the Series tab.

Additional e-books in this series can be found on Nova’s website

under the eBook tab.
PHARMACOLOGY - RESEARCH, SAFETY TESTING AND REGULATION

DRY POWDER INHALERS

FORMULATION, DEVICE
AND CHARACTERIZATION

TEERAPOL SRICHANA

New York
Copyright © 2017 by Nova Science Publishers, Inc.

All rights reserved. No part of this book may be reproduced, stored in a retrieval system or transmitted
in any form or by any means: electronic, electrostatic, magnetic, tape, mechanical photocopying,
recording or otherwise without the written permission of the Publisher.

We have partnered with Copyright Clearance Center to make it easy for you to obtain permissions to
reuse content from this publication. Simply navigate to this publication’s page on Nova’s website and
locate the “Get Permission” button below the title description. This button is linked directly to the
title’s permission page on copyright.com. Alternatively, you can visit copyright.com and search by
title, ISBN, or ISSN.

For further questions about using the service on copyright.com, please contact:
Copyright Clearance Center
Phone: +1-(978) 750-8400 Fax: +1-(978) 750-4470 E-mail: [email protected].

NOTICE TO THE READER

The Publisher has taken reasonable care in the preparation of this book, but makes no expressed or
implied warranty of any kind and assumes no responsibility for any errors or omissions. No liability is
assumed for incidental or consequential damages in connection with or arising out of information
contained in this book. The Publisher shall not be liable for any special, consequential, or exemplary
damages resulting, in whole or in part, from the readers’ use of, or reliance upon, this material. Any
parts of this book based on government reports are so indicated and copyright is claimed for those parts
to the extent applicable to compilations of such works.

Independent verification should be sought for any data, advice or recommendations contained in this
book. In addition, no responsibility is assumed by the publisher for any injury and/or damage to
persons or property arising from any methods, products, instructions, ideas or otherwise contained in
this publication.

This publication is designed to provide accurate and authoritative information with regard to the subject
matter covered herein. It is sold with the clear understanding that the Publisher is not engaged in
rendering legal or any other professional services. If legal or any other expert assistance is required, the
services of a competent person should be sought. FROM A DECLARATION OF PARTICIPANTS
JOINTLY ADOPTED BY A COMMITTEE OF THE AMERICAN BAR ASSOCIATION AND A
COMMITTEE OF PUBLISHERS.

Additional color graphics may be available in the e-book version of this book.

Library of Congress Cataloging-in-Publication Data

ISBN:  (eBook)

Library of Congress Control Number: 2016960251

Published by Nova Science Publishers, Inc. † New York

 Contents

Preface vii
General Introduction ix
Chapter 1 Particle Size and Size Characterization of Aerosols 1
Chapter 2 Flow Properties 35
Chapter 3 Forces and Interactions 51
Chapter 4 Formulation Design and Production Technology
in Dry Powder Inhalers 73
Chapter 5 Excipients and Physical Appearances of Dry
Powder Inhalers 107
Chapter 6 Device Design and Delivery Efficiency 139
Chapter 7 In Vitro Quality Control of Dry Powder Inhaler 179
About the Author 199
Index 201
 Preface

This book is a compilation of most works published during 2000-2016 in

the area of dry powder inhalers, especially related to the issues with
formulation and device design. The dry powder inhalers aim to deliver the
medication to the respiratory airways. It is suitable for postgraduate students
and researchers who work in the areas of dry powder inhalers. It provides
some background knowledge in size characterization, flow properties, forces,
and interaction of air and particles. It also ends with in vitro quality control of
dry powder inhalers. This book was finished in Hawaii where the author spent
his time during April 2016. Thank you to Prince of Songkla University for the
financial support. The author had to work very hard for one month without a
holiday. Thank you to Prof. Aran Pattanothai for overseeing this work to make
sure that the researchers were on schedule all the time. Thanks to the author’s
staff at the Graduate School, Prince of Songkla University who had to work
hard and be patient when the author was not in the office. Thanks to the
author’s wife and son who had to wait for him at home in Thailand. Thanks to
Dr. Tan Suwandecha and Dr. Janwit Dechraksa for their great efforts in copy
editing and compiling all the references. The author also thanks Dr. Somchai
Sawatdee and Dr. Dhamodharan Bakkiyaraj for their comments, and Dr.
Padmavathi Alwar Ramanujam for her proofreading. The greatest help was
from Professor Alan Coombes, Professor Pornanong Aramwit, and Dr. Brian
Hodgeson for their criticisms and comments to make this book readable.

Teerapol Srichana, Ph.D.

Prince of Songkla University
Hat Yai, Thailand
 General Introduction

Why Dry Powder Inhalers?

The lungs are an attractive route to deliver therapeutics for both local and
systemic therapy. A drug delivery system as an aerosol offers additional
degrees of manipulation to target the organ and specific cells. Pulmonary
delivery systems have many challenges including poor delivery efficiency,
instability of drugs and their formulations, and poor uniformity. Nevertheless,
these systems offer many advantages over oral routes such as higher
bioavailability and lower dose and less invasiveness in comparison with
injectable dosage forms. The lungs have a large surface area for absorption
and limited proteolytic enzymes. Together with its thin epithelial layer, it gives
high drug permeability and drugs can be delivered specifically to the target
areas. A targeted delivery system can reduce the dose and side effects from
low systemic exposure.
Not only is the local drug delivered to the target organ but also systemic
drug delivery can be done by targeting the alveolar regions (Patton and Byron
2007). A systemic delivery system can achieve a rapid onset and avoid first
pass metabolism. It is also useful for the delivery of biotherapeutics that
cannot be delivered orally. Furthermore, for the treatment of lung tuberculosis,
the therapeutics can be targeted to specific lung cells such as alveolar
macrophages (Changsan, Nilkaeo, et al. 2009).
The lungs can be exposed to various materials from different sources in
the breathing process. These airborne particles deposit along the airways from
the oropharynx, trachea, bronchi, bronchioles down to the respiratory
bronchioles and alveolar sacs. Particles are rapidly cleared by cilia within the
x Teerapol Srichana

mucus layer in the upper airways and by mucociliary escalator in the lower
airways. The pulmonary epithelium acts as a barrier to absorption in the
airways. The epithelium of the lung diminishes towards the lower airways to a
thickness of 0.2 m in the alveoli. Gas exchange occurs in the alveolar region.
The vast surface area of the alveoli with high blood supply from the capillary
arteries provides a great access of therapeutics to a systemic circulation. The
airways are protected by immune cells especially alveolar macrophages to
scavenge foreign materials along the lung surface (Fels and Cohn 1986).
Particles entering the lung are deposited along the airways by inertial
impaction, sedimentation, and diffusion (Hinds 2012). These particles are
characterized by their aerodynamic diameter. Large particles (> 5 m) are
deposited by inertial impaction in the mouth and upper airways, whereas the
smaller particles are deposited deeper in the lungs. Very small particles are
driven by diffusion.
Several technologies exist to deliver therapeutics to the airways.
Nebulizers have been used for a long time to deliver mists of a drug by
ultrasonic or air-jet mechanism. However, these devices tend to limit their
usage only at hospitals or home. Significant advances in medical aerosol
technology began in early 1950 with a focus on delivering asthmatic drugs
directly to the lungs thereby significantly reducing the dose and thus limiting
side effects. The first product was a handheld delivery system, a metered dose
inhaler (MDI), where the drugs were suspended in the compressed liquid gas.
However, these devices varied in particle deposition in the lungs depending on
the device, formulation and inspiratory effort of the patient. Alternatives were
then introduced as dry powder inhalers (DPI).

History of the Dry Powder Inhaler

DPI was established nearly 50 years after the MDI. Since the DPI is in a
solid state, it is suitable for unstable drugs in a liquid dosage form, and it does
not need a propellant gas. However, the MDI is commonly prescribed at the
first visit of an asthmatic or chronic obstructive pulmonary disease patient.
Later on, if the patient finds it difficult to comply with the MDI, the patient
may ask the doctor to change the prescription. Today, it is easy to switch from
the MDI to the DPI as several pharmaceutical manufacturers produce both
types of inhalers as a choice for the patients. The first DPI utilized the
patient’s inspiratory effort to disperse micronized drug particles. Further
 General Introduction xi

development has focused on the device development and particle engineering.

Compressed air may be supplied from the advanced device. Different sizes and
shapes of particles are designed to deposit in the alveolar regions.

Dry Powder Inhaler Formulation

DPI formulations are usually prepared as homogeneous interactive
mixtures comprised of micronized drug particles and a coarse carrier
(Malcolmson and Embleton 1998). The coarse carrier employed in the DPI
formulation is commonly α-lactose monohydrate to improve flow properties
and dose uniformity of highly cohesive drug particles. The entrainment and
subsequent aerosolization of the formulation are achieved using patient
inspiratory effort which is required to elutriate the micronized drug from the
carrier surface.
A critical step in generating a therapeutic aerosol via the DPI device is the
fluidization and entrainment of the bulk powder. Powder fluidization is a
process by which the powder mass is disturbed by a stream of airflow resulting
in the powder bed exhibiting fluid-like properties.
To achieve the goal of delivering drugs to the lung region, it is accepted
that particles must have an aerodynamic size of 1-5 µm (Hickey 2004).
However, particles of this size will have a high surface free energy and
therefore have a tendency to adhere together or to any other surface that they
might encounter to reduce this surface energy. Therefore, the knowledge of
particles, particle–particle interactions, and flow properties are essential for the
production of effective DPIs. The interaction between micronized drugs and
carrier particles has received much attention. Hence the number of previous
studies on this subject is high. In this book, only a few aspects related to the
performance of the powder formulations of the DPIs are discussed. The
adhesion forces between the carrier particles and drug particles consist mainly
of van der Waals forces, electrostatic forces and capillary forces. Van der
Waals forces are the most dominant forces that can determine adhesion or
cohesion in powders that are inhaled. Compared with electrostatic and
capillary forces, van der Waals forces are lower, which is a desirable property
from the viewpoint of dispersion during inhalation. Moreover, they can be
controlled to a certain extent and are more constant over longer periods.
The research areas that involve adhesive mixtures for inhalation have
explored the control of the properties of the carrier surface and measuring the
xii Teerapol Srichana

adhesive forces between the drug and carrier using centrifugal techniques and
atomic force microscopy. Mixing theories have been developed based on the
assumption that competition exists during mixing between drug–drug and
drug–carrier interactions and the equilibrium between them that occurs at any
moment during the mixing process. It has also been postulated that the
equilibrium can be driven in a certain direction by modifying the carrier
surface properties. Much attention has been given to the so-called ‘active sites’
on the carrier surface, onto which the drug particles are attached to higher
adhesive forces than the other sites. The term ‘active site’ has been used for a
multitude of phenomena including surface irregularities, surface rugosity and
adhering fines. Several investigations have attempted to modify and control
the surface rugosity of carrier crystals. The effects of carrier payloads, mixing
conditions, drug-to-carrier interactions, and the type and magnitude of the
removal forces during inhalation have been reported. Lactose monohydrate is
one of the most common carrier materials used in the DPI formulations. Other
sugars such as glucose and mannitol are also acceptable. In addition,
biodegradable polymers, phospholipids, cholesterol derivatives, cholates and
amino acids have been studied (Chuealee, Aramwit, and Srichana 2007,
Changsan, Nilkaeo, et al. 2009, Changsan, Chan, et al. 2009, Chuealee,
Wiedmann, and Srichana 2009, Chuealee et al. 2010, Rojanarat et al. 2011,
Chuealee, Wiedmann, and Srichana 2011, Rojanarat et al. 2012, Rattanupatam
and Srichana 2014, Gangadhar, Adhikari, and Srichana 2014, Ahmad,
Nakpheng, and Srichana 2014, Ahmad, Ungphaiboon, and Srichana 2015,
Kaewjan and Srichana 2016). Mixtures of lactose with the drug are often
called interactive mixtures, which are easier to handle during the
manufacturing process than micronized drugs. The use of a carrier also makes
the handling of small drug doses possible. The drug particles should loosely
adhere to the carrier particles so that they will easily detach from the carrier
particles and become available for deposition in the lungs. The larger carriers
impact tissues in the mouth or at the back of the throat and are swallowed. The
carrier also provides bulk to the formulation, which improves the handling,
dispensing and metering of the drug dose, which is of particular importance
for low dose formulations. To ensure efficient drug delivery, it is critical that
adhesive forces between the drug and carrier are not too strong. Otherwise, the
detachment of the drug from the carrier will be difficult or not possible.
Hence, the balance between adhesive and cohesive forces should be adjusted
to ensure sufficient adhesion between the drug and carrier to provide a
homogeneous blend with a uniform content. It is recognized that the efficiency
of a powder formulation is highly dependent on the carrier, the particle size
 General Introduction xiii

and size distribution, the inhalation flow rate, and the dispersion efficiency of
the respective DPI device.
Three different formulation types exist: spherical pellets, adhesive
mixtures, and particle agglomerates. In general, the micronized drug is
formulated into a powder mixture to improve dose uniformity and filling
process. Drug particles should be distributed homogeneously over the total
surface area of the carrier and should attach to the carrier surface primarily by
van der Waals forces. Carriers in adhesive mixtures consist of appropriate size
fractions. When the drug is distributed in multi-particulate layers around
carrier particles, agglomerates are formed. Carrier size distribution must then
be selected to obtain good flow properties, which is a prerequisite for
reproducible measurements of the dose.
The DPI platform is characterized by dispensing the medication to the
patient as a dry powder using a device specifically designed for that
formulation. In general, the inhalers employ the patient’s inspiratory flow as a
means to disperse and transport the aerosols into the lungs. Therefore we
usually call them breath-actuated devices. However, nowadays active devices
have been made where the device itself produces the aerosol while
disregarding the patient’s effort.

Dry Powder Inhaler Device

As for the design of the inhaler, this is one of the most important technical
aspects to be taken into account when designing and evaluating DPIs. There
must be a particular emphasis on the principles of powder de-agglomeration
and the moment at which the drug in the aerosol is released (Kwok and Chan
2011). The airflow is generated by the patient through the inhaler especially
for breath-actuated devices resulting in fluid and particle dynamics.
Considering the powder de-agglomeration efficiency, the drug particles in
dry powder formulations cannot be released as primary entities (1-2 µm) from
the DPI. By slightly increasing the primary drug particle size, even within this
narrow range, the detached mass fraction of the drug can be increased
substantially, particularly for inhalers with poor de-agglomeration efficiency.
The de-agglomeration system of the inhaler is one of the most important
steps that determines deposition of the drug in the lung. The de-agglomeration
system should break-up the spherical pellets into primary drug particles or
detach the drug particles from the carrier in the mixtures or agglomerates
xiv Teerapol Srichana

during inhalation. The adhesive forces between the drug and carrier particles
or the cohesive forces between drug particles have to be aerosolized to primary
drug particles. Different de-agglomeration systems use a different mechanism
to generate the aerosols. Clearly, the more efficient the force, the higher fine
particle fraction (FPF). Powder de-agglomeration systems in DPIs vary
considerably by their principle of operation. Most DPIs are often breath-
activated systems that utilize the kinetic energy of the inspiratory flow. The
sources of auxiliary energy are electromechanical means and pressurized air.
The efficacy of detachment of the drug particle will increase with an increase
in the inspiratory flow rate through the inhaler. This effect is most pronounced
when the kinetic energy of the air flow is utilized more efficiently. In contrast,
battery- and pressurized air-operated DPIs performance is not dependent on
the inhalation process. However, they are much more complex in design and
expensive. These devices may be inappropriate for use as disposable devices.
The performance of the inhaler obviously relies on many aspects including
the design of the inhaler, the device resistance, the de-agglomeration systems,
the formulation and the inspiratory air flow.
Well-designed DPIs are highly efficient systems, but they are also
complicated. The ideal DPI system should include most or all of the following
attributes (EMEA 2006): simple and comfortable to use; compact and
economical to produce; a stable multiple dose system; a reproducible emitted
dose over a wide range of inspiratory flow rates that are consistent and stable
throughout the inhalers life span; allow for highly reproducible fine particle
dosing; an ability to produce a physically and chemically stable powder; a
minimal extrapulmonary loss of the drug (low oropharyngeal deposition, low
device retention and low exhaled loss); protection of the powder from the
external environment and be usable in all climates; protection against overdose
and a dose delivery indicator; suitability for a wide range of drugs and doses.
DPIs can be single or multi-dose (multiple unit doses and multiple doses),
depending on the design of the powder reservoir and metering components. In
single dose devices, an individual dose is provided. However, they present a
weakness since drug loading is required. DPIs with multiple unit doses contain
a number of individually packaged doses, either as multiple gelatin capsules or
in blisters. In a multiple dose device, bulk powder reservoirs are in the device
from which an individual dose is metered.
The DPI device should help in the generation of very fine particles of the
drug in a way that enables them to avoid any impaction with barriers in the
lung. This mechanism also prevents the ingress of potentially harmful particles
to the airways. Most recently there has been an increasing trend to focus on the
 General Introduction xv

optimization of the powder-device technology to improve the generation of the

aerosol. In general, the inhaler design, particularly the geometry of the
mouthpiece, is critical for patients to produce a sufficient airflow to propel the
drug from the capsule or reservoir and break up the agglomerates in a
turbulent airstream. This device will then deliver a drug dose to the lungs as
therapeutically effective fine particles. Each inhaler will have a device that is
resistant to the airflow (measured as the square root of the pressure drop across
the device divided by the flow rate through the device). Current designs of DPI
devices have specific resistance values that range from about 0.02 to 0.2 (cm
H2O)½/liter per min (LPM) (Newman and Busse 2002). In order to produce a
fine powder aerosol with an increased delivery to the lung, a DPI should have
a low resistance which needs an inspiratory flow of 90 LPM, a medium
resistance DPI requires 50-60 LPM and a high resistance DPI requires 30-40
LPM (Lavorini 2013).

Production Technology
A multitude of techniques has been applied to improve the performance of
dry powder formulations for inhalation as listed below:

 Through the addition of micronized ternary excipients such as

isoleucine or magnesium stearate that decrease the adhesive forces
between the carrier and drug, mainly because the micronized
excipient and the drug compete for the active sites. However, the use
of magnesium stearate for inhalation raises questions related to safety.
 Supercritical fluid technology has been applied to improve
polymorphic purity and the surface properties of the drug substance
that can reduce the adhesive forces between the drug and carrier.
 Large porous particles with a high porosity and low density have been
produced for several reasons; the most important are the improved de-
agglomeration and aerodynamic size. Besides, they are also of interest
from the reduced phagocytosis of the deposited particles in the
alveoli. The macrophage uptake does not occur, so this prolongs the
release of the drug from the particles in the alveolar region. Smaller
porous particles (3-5 μm) have also been used to improve de-
agglomeration and lung deposition.
xvi Teerapol Srichana

 Spray dried, spray freeze-dried drugs and excipients can be used to

incorporate unstable drugs into stabilizing matrices like sugar or
polyol.
 New materials such as deoxycholate derivatives, cholesterol
derivatives, chitosans and their derivatives, and PLGA have been
introduced into dry powder formulations.
 Solid lipid nanoparticles, Trojan microparticles, and dimple-shaped
carriers have been introduced.

Many trends in the design and development of inhalation devices and

formulations can be seen in the literature. Several DPIs were under
development in the year 2000 and more recently even a higher number of
inhalers are in the pipeline. These new developments result in significant
improvements in DPI therapy. Computer simulations have been introduced to
understand the flow properties of the formulations in the inhaler devices. The
fluid dynamics and powder de-aggregation have been under investigation for a
number of years to understand the mechanism of drug particle releases, which
is not possible to visualize in an experiment. This in silico technology helps us
to understand the mechanism behind this technology.
In vitro quality control is essential for DPIs. This is a concern for the
manufacturer to ensure that the inhalation products meet the criteria indicated
in the Pharmacopeia. In this book, all the methods and general requirements
for in vitro tests of dry powders are summarized and grouped for readers with
some examples of certificates of analysis and data comparisons among the
new DPIs.

Next Generation Dry Powder Inhaler

It can be predicted that new formulations and new device technologies for
DPIs will be introduced in the next decade. Also, new biotechnology
developments will stimulate the growth of new DPIs. Other factors are the
possibility of new emerging respiratory diseases and outbreaks of new viral
infections that will also promote study in this area of drug delivery. It is
predicted that new DPIs that arise by making use of advancements in
technology and capability will solve many current problems associated with
the delivery of drugs. However, the technical background of inhalation
technology aspects has to be tuned to obtain the desired performance of dry
 General Introduction xvii

powder inhalers. Possible new advances in technical knowledge, new

developments and the possibilities for further improvements have been
gathered together and have been discussed in the book.

References
Ahmad, Md Iftekhar, Titpawan Nakpheng, and Teerapol Srichana. 2014. “The
safety of ethambutol dihydrochloride dry powder formulations containing
chitosan for the possibility of treating lung tuberculosis.” Inhalation
Toxicology 26 (14):908-917.
Ahmad, Md Iftekhar, Suwipa Ungphaiboon, and Teerapol Srichana. 2015.
“The development of dimple-shaped chitosan carrier for ethambutol
dihydrochloride dry powder inhaler.” Drug Development and Industrial
Pharmacy 41 (5):791-800.
Changsan, Narumon, Hak-Kim Chan, Frances Separovic, and Teerapol
Srichana. 2009. “Physicochemical characterization and stability of
rifampicin liposome dry powder formulations for inhalation.” Journal of
Pharmaceutical Sciences 98 (2):628-639. doi: 10.1002/jps.21441.
Changsan, Narumon, Athip Nilkaeo, Pethchawan Pungrassami, and Teerapol
Srichana. 2009. “Monitoring safety of liposomes containing rifampicin on
respiratory cell lines and in vitro efficacy against Mycobacterium bovis in
alveolar macrophages.” Journal of Drug Targeting 17 (10):751-762. doi:
10.3109/10611860903079462.
Chuealee, Rabkwan, Pornanong Aramwit, and Teerapol Srichana. 2007.
“Characteristics of cholesteryl cetyl carbonate liquid crystals as drug
delivery systems.” Proceedings of the 2nd IEEE International Conference
on Nano/Micro Engineered and Molecular Systems, IEEE NEMS 2007.
Chuealee, Rabkwan, Timothy Scott Wiedmann, and Teerapol Srichana. 2009.
“Thermotropic behavior of sodium cholesteryl carbonate.” Journal of
Materials Research 24 (1):156-163. doi: 10.1557/jmr.2009.0027.
Chuealee, Rabkwan, Timothy Scott Wiedmann, and Teerapol Srichana. 2011.
“Physicochemical properties and antifungal activity of amphotericin b
incorporated in cholesteryl carbonate esters.” Journal of Pharmaceutical
Sciences 100 (5):1727-1735. doi: 10.1002/jps.22398.
xviii Teerapol Srichana

Chuealee, Rabkwan, Timothy Scott Wiedmann, Roongnapa Suedee, and

Teerapol Srichana. 2010. “Interaction of Amphotericin B with cholesteryl
palmityl carbonate ester.” Journal of Pharmaceutical Sciences 99
(11):4593-4602. doi: 10.1002/jps.22176.
EMEA. 2006. Guideline on the pharmaceutical quality of inhalation and nasal
products.
Fels, Anna O., and Zanvil A. Cohn. 1986. “The alveolar macrophage.” Journal
of Applied Physiology 60 (2):353-369.
Gangadhar, Katkam N., Kajiram Adhikari, and Teerapol Srichana. 2014.
“Synthesis and evaluation of sodium deoxycholate sulfate as a lipid drug
carrier to enhance the solubility, stability and safety of an amphotericin B
inhalation formulation.” International Journal of Pharmaceutics 471 (1-
2):430-438. doi: 10.1016/j.ijpharm.2014.05.066.
Hickey, Anthony J. 2004. “Pharmaceutical Inhalation Aerosol Technology.”
American Association of Pharmaceuticals Scientists, 2nd ed Marcel
Dekker, Inc: NewYork.
Hinds, William C. 2012. Aerosol Technology: Properties, Behavior, and
Measurement of Airborne Particles: John Wiley & Sons.
Kaewjan, Kanogwan, and Teerapol Srichana. 2016. “Nano spray-dried
pyrazinamide-l-leucine dry powders, physical properties and feasibility
used as dry powder aerosols.” Pharmaceutical Development and
Technology 21 (1):68-75. doi: 10.3109/10837450.2014.971373.
Kwok, Philip CL, and Hak-Kim Chan. 2011. “Pulmonary delivery of peptide
and proteins.” Peptide and Protein Delivery, Elsevier Inc., London: 23-46.
Lavorini, Federico. 2013. “The challenge of delivering therapeutic aerosols to
asthma patients.” ISRN Allergy 2013.
Malcolmson, Richard J, and Jonathan K Embleton. 1998. “Dry powder
formulations for pulmonary delivery.” Pharmaceutical Science &
Technology Today 1 (9):394-398.
Newman, Simon P, and William W. Busse. 2002. “Evolution of dry powder
inhaler design, formulation, and performance.” Respiratory Medicine 96
(5):293-304.
Patton, John S., and Peter R. Byron. 2007. “Inhaling medicines: delivering
drugs to the body through the lungs.” Nature Reviews Drug Discovery 6
(1):67-74.
Rattanupatam, Teerarat, and Teerapol Srichana. 2014. “Budesonide dry
powder for inhalation: Effects of leucine and mannitol on the efficiency of
delivery.” Drug Delivery 21 (6):397-405. doi: 10.3109/10717544.2013.
868555.
 General Introduction xix

Rojanarat, Wipaporn, Narumon Changsan, Ekawat Tawithong, Sirirat

Pinsuwan, Hak-Kim Chan, and Teerapol Srichana. 2011. “Isoniazid
proliposome powders for inhalation-preparation, characterization and cell
culture studies.” International Journal of Molecular Sciences 12 (7):4414-
4434. doi: 10.3390/ijms12074414.
Rojanarat, Wipaporn, Titpawan Nakpheng, Ekawat Thawithong, Niracha
Yanyium, and Teerapol Srichana. 2012. “Inhaled pyrazinamide
proliposome for targeting alveolar macrophages.” Drug Delivery 19
(7):334-345. doi: 10.3109/10717544.2012.721144.
 Chapter 1

Particle Size and Size

Characterization of Aerosols

1.1. Introduction

Aerosols are ubiquitous in our environment. The physical properties of

aerosol particles and their determination by scientific means have been
developed to predict their behavior. To begin a systematic study of particles, it
is necessary to consider several commonly used definitions of aerosols.
Aerosols are suspensions of solid or liquid particles in a gas, usually air.

1.2. Morphological Properties of Aerosols

It is convenient to think of all aerosol particles as spheres for calculations,
and this also helps to visualize the processes taking place. A particle is
generally imagined to be spherical or nearly spherical. Either the particle
radius or particle diameter can be used to describe a particle’s size. Thus one
should ascertain which definition is being used when the term particle size is
used. Once a choice of diameter or radius is made, there are a number of ways
to define the diameter or radius which reflect particle properties other than
physical size. For a monodisperse aerosol, a single measure describes the
diameter of all particles. However, with polydisperse aerosols a single
diameter is insufficient to describe all particle diameters and certain
2 Teerapol Srichana

presumptions need to be made on the distribution of sizes. Other parameters in

addition to diameter must be used.
Two commonly encountered definitions of particle size are the Feret’s
diameter and Martin’s diameter (Sinko 2006). These refer to estimating the
particle size when determined by viewing the projected images of a number of
irregularly shaped particles. Feret’s diameter is the maximum distance from
the edge to edge of each particle, and Martin’s diameter is the length of the
line that separates each particle into two equal portions. Since these measures
can vary depending on the orientation of the particle, they are valid only if
averaged over a number of particles and if all measurements are made parallel
to one another. Then, by assuming random orientation of the particles, an
average diameter is determined. This measurement problem can be simplified
to an extent by using the projected area diameter instead of the Feret’s
diameter or Martin’s diameter. This is defined as the diameter of a circle
having the same projected area as the particle in question. Figure 1.1 illustrates
these three definitions. In general, the Feret’s diameter will be larger than the
projected area diameter and the projected area diameter will be larger than the
Martin’s diameter.
Sometimes a diameter is defined in terms of particle settling velocity. All
particles having similar settling velocities are considered to be of similar size,
regardless of their actual size or shape (Figure 1.2). Two such definitions are
aerodynamic diameter and Stokes’ diameter.
The aerodynamic diameter is a diameter of a unit density of sphere having
the same aerodynamic properties as the particle in question. This means that
particles of any shape or density will have the same aerodynamic diameter if
their settling velocity is the same (Hickey 2004).
Stokes’ diameter is a diameter of a sphere of the same density as the
particle in question having the same settling velocity. Stokes’ diameter and
aerodynamic diameter differ only in that Stokes’ diameter includes the particle
density whereas the aerodynamic diameter does not.
Aerosol sizes cover a range of about four orders of magnitude from 0.01
μm as a lower limit to approximately 100 μm as the upper limit. Particles
larger than 100 μm do not remain suspended in the air for a sufficient length of
time to be much of interest in aerosol science. Particles between 5 to 10 μm
are usually removed by the upper airways and those smaller than 5 μm can
penetrate deep into the alveoli of the lung. Particle size is the most important
descriptor for predicting aerosol behavior (Hickey 2004). This is apparent
from the above discussion and will become even more apparent in later
chapters.
 Particle Size and Size Characterization of Aerosols 3

Figure 1.1. Martin’s (dM), projected area (dPA) and Feret’s diameter (dF) for an irregular
shaped particle.

Figure 1.2. Particles with similar setting velocities are equivalent spheres.

1.3. Particle Size Distributions

The first step is to divide the entire size range into a series of successive
particle size intervals and determine the number of particles (the count) in each
interval. The intervals must be continuous so that no particles are left out. It
should have 10 size intervals for 1,000 particles. An example of particle size
distribution is shown in Table 1.1. When a particle size is grouped in intervals,
the grouped data will reveal the shape of the size distribution.
4 Teerapol Srichana

Table 1.1. Example of grouped data

Fraction/Size width Particle Cumulative

Size range (µm) Particle Count
(number/µm) Percent Percent
0.0-0.8 18 22.5 1.8 1.8
0.8-1.1 15 50 1.5 3.3
1.1-1.9 33 41.3 3.3 6.6
1.9-2.7 61 76.3 6.1 12.7
2.7-4.0 191 146.9 19.1 31.8
4.0-5.2 223 185.8 22.3 54.1
5.2-6.3 208 189.1 20.8 74.9
6.3-6.9 74 123.3 7.4 82.3
6.9-8.0 106 96.4 10.6 92.9
8.0-10 71 35.5 7.1 100
Total 1000 100.0

One graphical representation of the grouped data is the histogram, shown

in Figure 1.3, where the width of each rectangle represents the size interval
and the height represents the number of particles in the interval.
Thus, doubling an interval’s width results in roughly twice as many
particles falling into that interval. To prevent this distortion, the histogram is
normalized for interval width by dividing the number of particles in each
interval by the width of that interval. The height of each rectangle now equals
the number of particles per unit of size interval (number/µm) and the heights
of intervals with different widths are comparable (Figure 1.4). Furthermore,
the area of each rectangle is proportional to the number or frequency of
particles in that size range.

Figure 1.3. Histogram of frequency versus particle size.

 Particle Size and Size Characterization of Aerosols 5

Figure 1.4. Frequency/µm versus particle size.

Figure 1.5 Mass distribution.

The height of the number divided by the size (y-axis) times with particle
diameter (x-axis) gives an area equal to the number of particles in the interval.
The area of all rectangles is the total number of particles.
The histogram is usually standardized for sample size by dividing the
heights of the rectangles hi (in frequency/µm) by the total number of particles
observed in the sample, giving heights as fraction/µm. The area of each
rectangle in the units of the graph, hi  di is equal to the fraction fi of particles
6 Teerapol Srichana

in that size range, and the total area is equal to 1.0. This change allows for a
direct comparison of histograms obtained from samples of different sizes.
As shown in Figure 1.4, the shape of the distribution is the same as that
shown in Figure 1.3, but the ordinate is now the fraction of the total number of
particles per unit of size interval. Since we describe the frequency function in
terms of the particle size distribution, we can extend the analysis of size
distribution by mass (Figure 1.5).

Figure 1.6. Cumulative mass distributions.

The cumulative distribution enables one to readily determine quantitative

information about the particle size distribution (Figure 1.6). The fraction less
than a given size can be read directly from the graph. The fraction of particles
having diameters between two sizes can be determined directly by subtracting
the cumulative fraction of one size from the other.
As a means for further summarizing the grouped data, a mathematical
distribution function is assumed and parameters are calculated that define this
function for a particular data set. Most distribution functions require two
parameters: one that identifies the location or center of the distribution and one
that characterizes the width or spread of the distribution. These parameters will
be discussed elaborately in connection with specific distribution functions in
subsequent sections.
The most commonly used quantities for defining the location of
distribution are the mean, mode, and median. The mean or arithmetic average
 Particle Size and Size Characterization of Aerosols 7

is simply the sum of all the particle sizes divided by the number of particles.
The mean for the listed data is given by the midpoint size. The midpoint size
can be a geometric mean or an arithmetic mean. The former is often preferred.
The median is defined as the diameter that divides the frequency distribution
curve into equal areas and the diameter corresponding to a cumulative fraction
of 0.5.
The mode is the most frequent size on the frequency function curve. The
mode can be determined by setting the derivative of the frequency function
equal to zero and solving for d. For symmetrical distributions such as the
normal distribution, the mean, median and mode will be the same value which
is the diameter of the axis of symmetry. For an asymmetrical or skewed
distribution, these quantities will have different values. The median is
commonly used with skewed distributions because extreme values in the tail
have less effect on the median than on the mean. Most aerosol size
distributions are skewed, with a long tail to the right. For such a distribution,
mode < median < mean.
The preceding sections have focused on the properties of the distributions
of the size in general without considering any particular type of distribution. In
this section, we describe the characteristics and applications of the log-normal
distribution for aerosol particle size analysis (Figure 1.7). As discussed below,
the normal distribution, although widely used elsewhere, is not suitable for
most aerosol particle size distributions.

Figure 1.7. Frequency distribution curve (logarithmic size).

8 Teerapol Srichana

Other distributions have been applied to the aerosol particle sizes such as
the Rosin-Rammler, power law, and exponential distributions. These
distributions are applied to special situations but have limited application in
aerosol science. The log-normal distributions have been selected empirically to
fit the wide range and skewed shape of most aerosol size distributions. The
normal distribution function is rarely used to describe aerosol particle size
distributions because most aerosols exhibit a skewed distribution function
(Figure 1.8). The normal distribution is, of course, symmetrical. It can be
applied to monodisperse aerosols and to certain pollens and spores.

Figure 1.8. Count and mass distribution (logarithmic size).

The log-normal distribution is the most useful in situations where the

distributed quantity can have only positive values and covers a wide range of
values where the ratio of the largest to the smallest value is greater than 10.
When this range is narrow, the log-normal distribution approximates the
normal distribution.
The log-normal distribution is used extensively for aerosol size
distributions because it fits the observed size distributions reasonably well and
its mathematical form is convenient for dealing with the moment distributions
and moment averages described in the preceding sections.
The standard deviation is replaced by the logarithms of the standard
deviation, called the geometric standard deviation (GSD) or  g . The GSD is a
dimensionless quality with a value equal to or greater than 1.0.
For the count distribution, the geometric mean diameter (dg) is
customarily replaced by the count median diameter (CMD). The geometric
 Particle Size and Size Characterization of Aerosols 9

mean is the arithmetic mean of the distribution of ln d p , which is a

symmetrical normal distribution, and hence, its mean and median are equal.
The median of the distribution of ln d p is also the median of the distribution
of dp, as the order of values does not change in converting to logarithms. Thus,
for a log-normal count distribution, dg = CMD. The frequency function can be
expressed as a cumulative fraction.
Much of the practical application of the log-normal distribution to particle
size analysis is facilitated by using log-probability graphs. In the most
common form of these graphs, the axes are changed and the cumulative
fraction (or percent) scale is converted to a probability scale (Figure 1.9). The
probability scale compresses the percent scale near the median (50%) and
expands the scale near the ends such that a cumulative plot of a log-normal
distribution will yield a straight line. When the particle diameter scale is
logarithmic, the graph is called a log-probability plot. When the size scale is
linear, the cumulative graph will yield a straight line for a normal distribution
and is called a probability plot. In either case, the median size can be read
directly from the graph, as with any cumulative plot.

Figure 1.9. Log-probability graph.

10 Teerapol Srichana

A log-probability plot can be constructed on arithmetic graph paper by

plotting the logarithm of the diameter versus the probits of the cumulative
percentages. The latter, obtained from a table of probits, gives the linear
displacement from the midpoint (50%) in units of standard deviations. The
slope of the straight line is related to the GSD. A line with a steep slope is
associated with a wide distribution and a line with a shallow slope represents a
narrow distribution. A horizontal line in Figure 1.9 characterizes a
monodisperse aerosol in which all particles have the same size.
For other expressions of a probability scale, it can be transformed with a
Z-score (Figure 1.10). The particle size is transformed into a natural logarithm,
and the frequency of data events is transformed to a Z-score, which is equal to
the % cumulative of each data stage minus by the cumulative average and
divided by the standard deviation of all interested events (Srichana, Suedee,
and Srisudjai 2003). The slope of the linear equations is used to define the cut-
off diameter (Z=0), d84 (Z=1), d16 (Z=-1) and GSDs are explained by the
spread of the aerodynamic particle size distribution according to Equation 1.1.

1/2
d 
GSD   84  (1.1)
 d16 

where,
d84 is the diameter at 84% of the aerosol mass

d16 is the diameter at 16% of the aerosol mass

Figure 1.10. Log Z-score distribution.

 Particle Size and Size Characterization of Aerosols 11

Figure 1.11. Cumulative number and cumulative mass distributions on the log-
probability graph.

For any normal distribution, one standard deviation represents the

difference between the size associated with a cumulative count of 84% and the
median size (a cumulative count of 50%) or between the 50% cumulative size
and the 16% cumulative size.
One great advantage of the log-normal distribution is that, for a given
distribution, the GSD remains constant for all moment distributions.
Frequently in aerosol sampling, there is an aerodynamic size above which
particles are aerodynamically unable to enter the sampling apparatus
(Srichana, Martin, and Marriott 2000). This is called the aerodynamic cut-off
size, which means that the cumulative line on the log-probability plot will
curve near its upper end so that it never exceeds the cut-off size.
A similar limit may exist for the lower end of the size distribution if sizing
is done by optical microscopy. Particles less than 0.3 µm in diameter are not
included in the size distribution. This 0.3 µm limit, sometimes called the
optical cut-off, curves the lower end of the size distribution line so that it never
goes below 0.3 µm. These cut-offs are artifacts of the sampling and
measurement system. If the cut-offs affect only a small fraction of the
12 Teerapol Srichana

distribution, it is acceptable to ignore them when fitting a straight line to the

data on a log-probability plot (Srichana, Suedee, and Srisudjai 2003).
Figure 1.11 shows both the number and mass distributions for the data
given in Table 1.1, plotted as cumulative distributions on a log-probability
graph paper. The horizontal axis refers to the cumulative percent of mass for
the mass distribution line and the cumulative percent of the number
distribution line.

1.4. The Hatch-Choate

Conversion Equations
The real power of the log-normal distribution comes from the fact that any
type of average diameter discussed in this chapter can easily be calculated for
any known log-normal distribution that is a distribution for which one average
diameter and the GSD are known. This is useful because it is frequently
necessary to measure one characteristic of the size distribution, such as the
number distribution, when what is really needed is another characteristic, such
as the mass distribution or the diameter of the average mass.

d g  sizeat 50%cumulative frequency (1.2)

dg
g  (1.3)
size at 84% oversize

log dln  log d g  1.151log 2 g (1.4)

log d 'g  log dln  5.757 log 2 g (1.5)

Where,
d g is a geometric mean diameter
 g is a geometric standard deviation
dln is a length-number mean diameter
d 'g is a geometric mean diameter on weight basis
 Particle Size and Size Characterization of Aerosols 13

If the distribution is log-normal, these quantities can be calculated directly

using the log-normal conversion equations originally derived by Hatch and
Choate (1929). These are known as the Hatch-Choate equations. The
transformation equations were used to convert a number distribution to a
weight distribution (Sinko 2006). The logarithm of the particle size can be
plotted against the cumulative percent frequency on a log-probability scale
(Equation 1.5).
Although these equations are written for number distributions, they can be
combined to convert one type of average diameter to any other type of average
diameter. The examples that follow are for the most common conversions
between the number and mass distributions. Analogous equations can be
written for any moment distribution, such as the distribution of surface area or
of settling velocity.

1.5. Instrumentation in
Particle Size Analysis
Drug particle size is recognized to play an important role in defining the
location of the aerosol particle deposition in the respiratory tract (Hickey and
Jones 2000). Therefore, a reliable technique is required in order to measure the
particle size of the inhaled aerosols and to assess the drug deposition profiles
both in terms of the quantity of the drug reaching the respiratory tract and its
depth of penetration. When considering lung deposition, the aerodynamic
diameter (Dae) is of great interest. This is defined as the diameter of a unit
density sphere with the same terminal settling velocity as the particle being
studied (Hickey and Jones 2000, Mitchell and Nagel 2004).
Three main techniques are used to analyze the particle size of aerosols:
direct imaging, optical sizing and inertial impaction (Hickey and Jones 2000).

1.5.1. Direct Imaging Based on Microscopy

The microscopic technique is a direct particle size measurement

technique. The particle’s morphology has been used to describe their powder
properties via a number of approaches and mathematical models. Both
qualitative and quantitative data have been determined by this technique to
present simple to complex phenomena in DPIs. However, the subjective nature
14 Teerapol Srichana

of the dimensions and the largely unrealistic and unrepresentative pattern of

airway deposition are the limitations of this technique (Hickey and Jones
2000). A scanning electron microscope (SEM) is one type of microscope that
produces images of a sample by scanning it with a focused beam of electrons.
The electrons interact with atoms in the sample, producing various signals that
contain information about the sample’s surface topography and composition
when it is equipped with an energy dispersive X-ray. The electron beam is
generally scanned in a raster scan pattern, and the beam’s position is combined
with the detected signal to produce an image. SEM can achieve a resolution of
better than 1 nanometer. Specimens can be observed in a high vacuum, low
vacuum, wet conditions (in environmental SEM) and in a wide range of
cryogenic or elevated temperatures. The most common SEM mode is the
detection of secondary electrons emitted by atoms excited by the electron
beam. The number of secondary electrons that can be detected depends on the
angle at which the beam meets the surface of the specimen (i.e., on specimen
topography). By scanning the sample and collecting the secondary electrons
that are emitted using a special detector, an image displaying the topography
of the surface is created. Here is an example of SEM micrographs obtained
from a DPI formulation (Figure 1.12).

Figure 1.12. The scanning electron microscope (SEM) images of (a) spray dried
mannitol (b) porous mannitol (c–g) Levofloxacin-proliposome formulations (h)
Levofloxacin. (Reprinted from Rojanarat, W. et al., Pharmaceutics, 4(3), 385-412,
2012).
 Particle Size and Size Characterization of Aerosols 15

1.5.2 Optical Sizing

Light scattering (using laser particle sizing equipment) is a non-invasive

technique which provides an estimate of volume median diameter and some
index of polydispersity. The most notable of these methods used phase
Doppler anemometry, time-of-flight laser velocimetry and laser diffraction.
Each of these methods has an advantage of describing particles over a very
broad range of sizes and therefore gives more information about the
distributions in the absence of a chemical detection method. This is a
drawback because the mass of a drug is related to its therapeutic effect. In
addition, this technique does not take the anatomical structure of the human
respiratory system into account (Hickey and Jones 2000).
When an electromagnetic beam is projected through a particle field, some
portion of the beam is transmitted through the field while some portion of the
beam is absorbed and the remaining is scattered by particles in the field. The
ratio of the velocity of light in a vacuum to the velocity of light in a particular
material is the refractive index of that material. For a given wavelength, the
index of refraction depends only on the material, but it varies slightly with the
wavelength of the light used.
In general, the distribution of light scattered by a particle is a function of
the particle size and shape, the incident light and the reflection indices of both
the particle and the surrounding medium (Kokhanovsky and Zege 1997). Both
the scattering and absorption characteristics of a particle can be included by
describing the particle’s optical properties with a complex refractive index
where the real part describes the scattered light characteristics and the
imaginary part describes the absorption.
A laser beam is expanded and then collimated into a diameter of a few
millimeters which passes through the particle cloud (Figure 1.13). Particles in
the beam scatter in all directions, although predominantly in the near-forward
direction. A receiving lens is used to focus both the transmitted beam and the
forward scattered light (predominantly diffracted) onto a detector located at
the back focal plane of the lens. The transmitted light is focused to a point on
the optical axis, while the diffracted light forms a series of concentric rings.
As the receiving lens performs a Fourier transformation on the scattered
light, light scattered at a given angle () by a particle located anywhere in the
illuminated sample volume will be focused at the same radial position in the
transformed plane. Thus the resulting pattern is unaffected by the location of
the particle or its motion.
16 Teerapol Srichana

Laser diffraction offers a convenient and established method to determine

the particle size. It is used to determine a calibrant size either prior to or
subsequent to entering through the impactor (Kwon et al. 2003, Ziegler and
Wachtel 2005). Ziegler and Wachtel (2005) developed the laser diffraction
method for particle size distribution measurements in pharmaceutical aerosols.
An example of size measurement by laser diffraction is shown in Figure 1.14.
It can be seen that the size distribution of the commercial powder is larger than
the milled one. The milled powder has the median size around one micron
whereas the commercial powder before milling has a median size of 4 microns
with a range of 1-80 microns.

Figure 1.13. Schematic diagram of extinction-measuring apparatus.

Figure 1.14. The particle size distribution obtained from laser diffraction of
commercial powder (diamond red line) and milled powder (square blue line).
 Particle Size and Size Characterization of Aerosols 17

The parallel laser beam was introduced into a modified United States
Pharmacopeia (USP) throat to measure particle distribution by a scattering
pattern and intensity of light at different angles. The particle size distributions
seem to be reproducible and have a strict correlation between the impactor and
laser diffraction in the case of aqueous formulations (Ziegler and Wachtel
2005). Kwon et al. (2003) calibrated their 5-stage designed impactor using a
gravimetric or counting method.

1.5.3. Inertial Sampling

This is the most widely used particle-size analysis method for inhaler
output. Various instruments have been used to determine the particle size
distribution of aerosols within a model respiratory tract designed to reproduce
the anatomical dimension of an average healthy human airway. The
instruments vary among inertial samplers, air inlet dimensions, the sampling
airflow rates (12.5 to 120 LPM), the number of collecting stages (i.e. particle
size ranges for collection within the distribution which varies from 2 to 8
stages) and the nature of the collection surface (liquid for impinger and
uncoated or coated solid surfaces for the impactor). The drug is collected and
washed from these stages and analyzed by chromatographic and
spectrophotometric means to determine its mass (Hickey and Jones 2000).
A device with two stages has been developed as a quality control tool to
indicate the proportion of the fine particles present in the distribution. The
most frequently used in vitro method is a multiple-stage impactor because
more information is recovered with regard to the range of particle sizes within
the distribution. The cascade impactor utilizes the relationship between the
velocity and the mass where larger particles with sufficient inertia are
impacted on the upper stages whereas finer particles penetrate to the lower
stages of the separator. Cascade impactors provide a useful aerodynamic
measure of the distribution of particle sizes that can be used to compare
devices and formulations. However, the disadvantage of this method is the
narrow range of discriminated sizes, typically at values of Dae from 0.2 to 10
m, and requires considerable labor to perform the time-consuming analysis,
which is subjected to operator variability and error (Hickey and Jones 2000).
To evaluate the deposition performance of a DPI in vitro, there are two
common techniques, namely, a twin-stage impinger (TSI), which is official in
the British Pharmacopeia (BP) (BP 2015), and an Andersen cascade impactor
18 Teerapol Srichana

(ACI), which has been approved by both the BP and the USP (Figure 1.15).
The operation of the two techniques is described in Chapter 7.

Figure 1.15. Andersen Cascade Impactor (left) and Twin Stage Impinger (right).

Twin Stage Impinger

The TSI is a two stage impinger that uses liquid at the first stage and a
lower stage to capture the aerosol material while air with particles is drawn
through the instrument. Even though the TSI method is easier to operate than
the ACI method, limitations arise from the fact that the TSI divides the total
spray into only two compartments. Hence, this method cannot discriminate
between distributions with specific combinations of a size range and cannot
precisely predict when operating out of the designed flow rate (Broadhead,
Rouan, and Rhodes 1995, Geuns et al. 1997). The data derived from the TSI
are intended to differentiate only good or bad formulations as determined by
the fine particle fraction (FPF, particle size < 6.4 µm). This device is suitable
for the quality control of aerosol products (Snell and Ganderton 1999). There
was an attempt to integrate the oral throat cast to mimic the oropharynx while
the USP describes a 90° sharp bend metal induction port to the cascade
impactor system. Hence the rationale for the design of the in vitro test is an
attempt to simulate the patient use as far as practicable with the hope of
achieving a correlation of the results obtained from such tests with the in vivo
performance (Srichana, Martin, and Marriott 1998).
 Particle Size and Size Characterization of Aerosols 19

Andersen Cascade Impactor

The ACI is a multistage and multi-orifice cascade impactor. The concept
of the ACI evolved on the basis that the human respiratory tract represents an
aerodynamic classifying system for airborne particulates. The ACI is
comprised of eight aluminum stages which are held together by three spring
clamps with o-ring seals. Each impactor stage contains multiple precision
orifices so that when the air is drawn through the sampler, any airborne
particles with high momentum were moved towards an impact with the
collection plate for each stage and those with low momentum adjust to a new
direction of flow and pass around the obstruction (Figure 1.16) (Hickey,
Martonen, and Yang 1996).

Figure 1.16. General principle of inertial sampling through a jet onto a collection plate.

Table 1.2 presents the diameter of the orifices that are progressively
smaller from the top to the bottom stages that range from 0.1004-inch diameter
at stage 0 to 0.0100-inch diameter at stage 7. When operated at 28.3 LPM, the
jet velocity is used to calculate the jet Reynolds number. The range of the
particle sizes collected at each stage depends on the jet velocity of that stage
and the cut-off diameter of the previous stage. Finally, the collected particles
at each stage that are related to the aerodynamic dimension should be used to
predict the deposition of the human lung airborne particles (USP 30-NF 25
2007).
20 Teerapol Srichana

Table 1.2. Jet dimensions for each stage and

ACI (Mark II) parameters operating at 28.3 LPM.
(Adapted from Vaughan, NP., Journal of Aerosol Science, 1989)

Stage Orifice diameter (inches) Number of orifices Jet velocity (cm/s) Rej
0 0.1004 96 96.3 163.7
1 0.0743 96 175.8 221.2
2 0.0360 400 179.8 109.6
3 0.0280 400 297.2 140.9
4 0.0210 400 528.9 187.9
5 0.0135 400 1277.0 292.0
6 0.0100 400 2328.8 394.3
7 0.0100 201 4618.1 782.0
F 0.1100 Filter

This simplified analysis is not accurate enough to characterize the

impactor efficiency fully, but it does show that the Stokes number as defined
by equation 1.6 is the relevant parameter for characterizing impaction.

 p d p2U c
Stk  (1.6)
9 D j

Every in vitro assessment technique must be validated. The performance

of the inhaler device was reported as % FPF and Mass Median Aerodynamic
Diameter (MMAD).
The characteristic dimension for the Stokes number of the impactor should
be the distance between the jet and the impaction plate, but this is not the case.
The streamlines at the nozzle exit are not strongly affected by the spacing
between the nozzle and the plate because the jet of the aerosol expands only
slightly until it reaches within about one jet diameter (Dj) of the impaction
plate. Hence, the characteristic dimension for an impactor is the jet radius or
half the width, rather than the spacing between the nozzle and the plate.
For most impactors, a complete curve of the collection efficiency versus
particle size is not necessary. Impactors that have a “sharp cut-off” curve
approach the ideal from the standpoint of particle size classification) step-
function efficiency curve in which all particles greater than a certain
aerodynamic size are collected and all particles less than that size pass
through. The size in question is called the cut-off size or cut-off diameter (d50).
 Particle Size and Size Characterization of Aerosols 21

As a practical matter, most well-designed impactors can be assumed to be

ideal, and their efficiency curves are characterized by a Stokes number 50%
(Stk50) that provides a 50% collection efficiency. Stk50 is the location of the
ideal cut-off curve that best fits the actual cut-off curve. This is equivalent to
assuming that the mass of particles larger than the cut-off size that get through
(the upper shaded area) equals the mass of particles below the cut-off size that
are collected (the lower shaded area) (Figure 1.17).

Figure 1.17. Actual and ideal impactor cut-off curves. (Adapted from Hinds, W. C.,
John Wiley & Sons, 2012).

All round jet impactors meeting these design criteria will have the same
Stk50 value, regardless of the nozzle diameter. By adding a downstream filter
to the impactor shown in Figure 1.18, one can collect all the particles that
escape impaction. Sampling an aerosol with such an impactor can provide
information about its particle size distribution. The mass of the particles
collected on the impactor plate and of those collected on the filter are
determined by weighing them before and after sampling. The impactor
22 Teerapol Srichana

separates the sampled particulate mass into two particle size ranges that are
contributed by particles larger than the cut-off size (collected on the impaction
plate) and by particles smaller than the cut-off size (collected on the filter). For
example, assume that an impactor with a 5 µm cut-off size collects 30% of the
aerosol mass on the impactor plate and 70% on the filter. Then 30% of the
aerosol mass is from particles greater than 5 µm in aerodynamic diameter and
70% is from particles less than 5 µm. Thus, this measurement provides one
point on the cumulative distribution curve, that 70% of the particulate mass is
associated with particles of less than 5 µm. By operating the impactor at
several flow rates, each corresponding to a different cut-off diameter, several
points on the cumulative mass distribution curve can be obtained. There are
practical limitations on the range of flow rates that can be used and the aerosol
size distribution must remain constant for all the samples. The latter problem
can be overcome by simultaneously operating several impactors with different
cut-off sizes.
The use of several impactors in parallel is, however not common because
of the complexity of controlling multiple flow rates. The more common
approach is to operate several impactors in series, arranged in order of
decreasing cut-off size with the largest cut-off size first. This configuration is
called a cascade impactor (Figure 1.18).
Each separate impactor is called an impactor stage (Figure 1.18). The cut-
off size is reduced at each stage by decreasing the nozzle size. Reducing Dj
increases U, and both serve to reduce d50 according to equation 1.7. Since the
same volume of gas flows through each stage and only one flow needs to be
controlled. Each stage is fitted with a removable impaction plate for
gravimetric (or chemical) determination of the collected particles. The last
stage in a cascade impactor is usually followed by a filter that captures all
particles less than the cut-off size of that stage.

9 D j Stk50
d50  (1.7)
 pU

In its operation, each stage is assumed to capture all particles reaching it

that are larger than its cut-off size. Because the aerosol flows in sequence
through successive stages, the particles captured on the impaction plate of a
given stage represent all particles smaller than the cut-off size of the previous
stage and larger than the cut-off size of the given stage. It is important to keep
 Particle Size and Size Characterization of Aerosols 23

in mind that an impactor stage consists of a nozzle section and the impaction
plate (Figure 1.16).

Figure 1.18. Schematic diagram of cascade impactor with nozzle and impaction plate.

Data reduction is based on the assumption that each stage has the ideal
cut-off characteristics. This assumption imposes an additional requirement that
the cut-off sizes for successive stages be far enough apart that there is a
negligible overlap in their collection curves. If they overlap, a much more
complicated analysis is required, using the exact shape of the collection
efficiency curve. Also, it is assumed that the non-ideal cut-off effects cancel
out; that is, the mass of the oversized particles getting through each stage
equals the mass of undersized ones that are collected. This may not always be
the case and if it is not this will cause a distortion of the size distribution.
The cut-off size ranges given are for typical flow rates and can be changed
by operating the impactor at a different flow rate according to equation 1.7.
The preceding discussion has assumed that the particles stick if they strike the
surface of the impaction plate. For liquid particles, this is nearly always
24 Teerapol Srichana

correct. Solid particles, however, may bounce and blow off when they strike
the impaction plate, or they may adhere. Some of the mass of a certain particle
size ends up on a lower stage and is thus attributed to a smaller particle size
range. Furthermore, once a particle bounces it is likely to continue to bounce
in subsequent stages because the impaction velocity is greater in these stages.
Whether or not a particle bounces depends on its material, its velocity and the
type of impaction surface. Coating the impaction plate with a thin film of oil
or grease reduces bounce (Hinds 2012). Antibounce coatings have been used
successfully including high-vacuum grease, petroleum jelly, and silicone oil.
To obtain a thin film, these materials are usually dissolved in solvents such as
toluene or cyclohexane and are painted or sprayed onto the impaction surface,
and allowed to dry. Even with an antibounce coating, bounce can occur when
the impaction surface becomes overloaded with particles. This is because
incoming particles have an increased probability of impacting on a deposited
particle rather than the coating. Impaction surfaces made up of a porous metal,
or a coarse filter saturated with oil served as a reservoir and allowed the oil to
“wick” through the deposited particles to maintain an effective antibounce
coating. Plastic films or small pore membrane filters do not reduce particle
rebound, although these surfaces may be convenient for analytical purposes.
Particles can be deposited in the passageways between stages of the
cascade impactor. Such deposition is called the interstage loss and represents
another operational problem with cascade impactors. For conventional cascade
impactors, interstage losses are primarily a problem of large particles being
lost in the first two stages. Particles are lost primarily by inertial removal at
bends in the flow path. Since these interstage losses depend on particle size
and are not included in the collected mass, they distort the size distribution
toward smaller sizes. Interstage losses can be reduced by designing the
impactor to minimize sharp bends in the interstage flow path of the first few
stages or by operating the impactor at a lower flow rate. As particles deposit
on the impaction plate, they form a spot that grows into a conical pile of
powder that changes the flow geometry and the cut-off size. This also reduces
the effectiveness of any grease coatings and sets a practical upper limit on the
mass that can be collected on any stage. The lower limit is usually set by the
analytical technique used, such as the sensitivity of the balance used. Some
impactors are equipped with a preseparator to collect large amounts of the
large particles so that accurate measurements can be made of the size
distribution in the smaller particle size range. Preseparators do not need to
have a sharp cut-off, because as long as the preseparator collects particles
larger than the cut-off size at the first stage, the mass collected by the
 Particle Size and Size Characterization of Aerosols 25

preseparator can be combined with that collected on the first stage to give the
total mass that was larger than the first-stage cut-off size. Some impactors
have a collection surface that slowly rotates during operation, to prevent
excessive buildup of deposited particles. Dechraksa et al., (2014) developed a
validated computational fluid dynamic (CFD) model of the ACI and
investigated the effects of the preseparator on the CFD parameters. The flow
field indicated that the preseparator accelerated the airflow velocity at the
induction tube. The CFD model explained the airflow of the preseparator
equipped model by accelerating the airflow along the inlet port to maximize
the trapping of the desirable particles and the generation of a smooth wall
shear stress at the collection plate to reduce the particle reentrainment. Figure
1.19 shows a particle size distribution of isoniazid DPIs evaluated by the ACI
at a flow rate of 60 LPM. The drug particles deposited mainly on the glass
throat (GT) and the preseparator. The drug particles traveled deeper into the
impactor stage -1 to stage 6. It was predicted that the smaller drug particles
were likely to reach the lower airways.

Figure 1.19. The particle size distribution of an isoniazid dry powder inhaler after
aerosolization into the ACI at the flow rate of 60 LPM (GD = Glass Device, GT =
Glass Throat, P = Preseparator) (mean ± SD, n = 3).

Next Generation Impactor (NGI)

The next generation impactor (Figure 1.20) or Apparatus 5 for DPIs in the
USP (USP 30-NF 25 2007) was first designed in 2000 to evaluate
pharmaceutical aerosols. This is highly accurate and easy to operate. There is a
linear correlation between the data obtained from the ACI and NGI (Guo et al.
2008) at 30 LPM. Variations due to manual actuation (operator variance), flow
rates and induction ports, significantly affect the particle size distribution
(PSD) and dose delivery profile. The instrument can be operated at a flow rate
26 Teerapol Srichana

of 15-100 LPM. There are 7 stages (cut-off diameters are 0.54-6.12 μm). The
unit easily collects samples, and the interstage loss is low. Each stage has
500 < Re < 3000.

Figure 1.20. Next generation impactor.

Multistage Liquid Impinger (MSLI)

This impinger has combined the ACI and TSI and reduced the number of
stages to only 4-5 stages (Figure 1.21). It is operated by the concept of aerosol
particles being trapped by the liquid. The instrument can be operated at 30-100
LPM as stated in the USP apparatus 4 for DPI. It can be used either with DPI,
MDI or nebulizer. BP recommended that the MSLI be operated at 60 LPM.
The cut-off diameters of stages 1, 2, 3 and 4 were 13, 6.8, 3.1 and 1.7 m,
respectively. Stage 5 collects particles with sizes less than 1.7 m. This
technique reduces the re-entrainment of aerosol particles.
Figure 1.22 shows the particle distribution of two DPI formulations of
budesonide after aerosolization into the MSLI at 60 LPM into two different
inspiratory volumes of 1 and 2 L. The mass deposition on each stage of the
MSLI was not different between the two different inspiratory volumes. Hence,
the inspiratory volume at the first few seconds is crucial.
 Particle Size and Size Characterization of Aerosols 27

Figure 1.21. Multistage liquid impinger from Copley Scientific.

Figure 1.22. Aerodynamic particle size distributions of budesonide in the test

formulation with inspiratory volumes (IV) of 1 and 2 L.
28 Teerapol Srichana

1.5.4. Time-of-Flight Instruments

Time-of-flight instruments can provide real-time, high-resolution

measurements of aerodynamic particle size and size distribution over a wide
size range. Air is accelerated (> 106 m/s2) in a converging nozzle to a high
velocity (> 100 m/s) at the nozzle exit. Two narrowly focused laser beams are
positioned in the jet about 100 µm apart (Figure 1.23). Particles, focused into
the center of the jet by the clean sheath air, are accelerated by the airflow in
the nozzle. Small particles less than 0.3 µm can keep up with the accelerating
air in the nozzle and exit with approximately the same velocity as the air. As a
particle passes through the first laser beam, it creates a very brief (1µs) pulse
of scattered light that is detected by a photomultiplier tube. A similar pulse is
generated when the particle passes through the second beam. The time interval
between the two pulses is sensed electronically and is used to determine the
average velocity of the particle as it passes through the space between the two
laser beams. The time between these two beams is measured with a precision
of 25 nanoseconds and is termed the time of flight (TOF). Since TOF is
dependent on particle size, it is possible to obtain a particle size distribution of
any powder including, for example, the lactose and drug depositing on the
individual plates of an ACI after a dry powder formulation has been
aerosolized into the impactor. With suitable calibration, the particle’s
aerodynamic diameter can be determined from the magnitude of the TOF. This
is done electronically and the particle size and size distribution are determined
nearly in real time. There are important limitations to these instruments that
arise primarily because particle motion is outside the Stokes region due to the
high velocities used.

The general form of the force equation is given by equation 1.8.

V  Vp 
2
d2 1 dv
a   d 3 p
a
Cd (1.8)
4 2 6 dt

where:

Cd = drag coefficient d = particle diameter (cm)

 Particle Size and Size Characterization of Aerosols 29

 a = density of air (g mL-1)  p = density of particle (g mL-1)

Va = air velocity (dyne s cm3) V p = particle velocity (cm s-1)

Figure 1.23. Time of flight mass spectrometry size analysis. (Adapted from
https://www.osapublishing.org).

d2
The term Cd relates to the drag coefficient and projected area of the
4
V  Vp 
2

a
a
particle, while the term relates to the air density and
2
differential velocity of the air and the particle. The right side of the equation is
30 Teerapol Srichana

1 
simply the particle volume multiplied by the density   d 3  p  and yields
6 
 dv 
the particle mass, which is then multiplied by the particle acceleration  .
 dt 

1.6. Isokinetic Sampling

lsokinetic sampling is a procedure to ensure that a representative sample of
aerosol enters the inlet of a sampling tube when sampling from a moving
aerosol stream. The sampling may be extracted from a windy environment by
a probe. Sampling is isokinetic when the inlet axis of the sampler, a thin-
walled tube or probe, is aligned parallel to the gas streamlines and the gas
velocity entering the probe equals the free-stream velocity approaching the
inlet. This condition is equivalent to taking a sample so that there is no
distortion of the streamlines just upstream of the inlet (Figure 1.24). If
sampling is isokinetic, there is no particle loss at the inlet, regardless of the
particle size or inertia. Isokinetic sampling in no way ensures that there are no
losses between the inlet and the collector; instead, it guarantees that the
concentration and size distribution of the aerosol entering the tube are the
same as that in the flowing stream.

Figure 1.24. Isokinetic sampling.

 Particle Size and Size Characterization of Aerosols 31

In isokinetic sampling, failure to sample isokinetically may result in a

distortion of the size distribution and a biased estimate of the concentration.
These effects arise because of particle inertia in the region of curved
streamlines near the inlet. Depending on the conditions, the sample may
contain an excess or a deficiency of large particles. If sampling is not done
isokinetically, there is no way to determine the true concentration unless the
original particle size distribution is known or can be estimated.

1.7. The In Vivo Assessment Technique

The in vivo assessment technique is commonly used in human
volunteers. There are 2 common techniques including a pharmacokinetic-
pharmacodynamic measurement and gamma scintigraphy.
Pharmacokinetic data (plasma and urine levels) and pharmacodynamic
data (e.g., pharmacological response heart rate or blood pressure) are
sometimes used to infer the delivery of asthma medications to the lungs
(Kempsford et al. 2005). However, only classical bioequivalence testing based
upon equal rates and extents of drug absorption is inappropriate to show the
equivalence of products containing inhaled asthma drugs that act directly on
the airway surface. Further, this technique needs to be validated and is
probably applicable only in a limited range of pharmaceuticals.
Another in vivo assessment method is the non-invasive imaging technique
such as gamma scintigraphy. A method of wide application to radiolabelling
dry powders is by adsorbing the radiolabel on the active particles in a suitable
liquid. This is achieved by wetting the drug particles with a nonsolvent
containing the radiolabel, followed by the evaporation of the solvent, leaving
the radiolabel on the surface of the drug particles. This method gives a
measure of local bioavailability at the site of action in the lungs and the lung
deposition data are strongly correlated with the clinical response to the inhaled
asthma drugs (Newman and Wilding 1998). However, all of the in vivo
assessment techniques are not easy to monitor in product development
processes; therefore, in vitro assessments are still necessary.
32 Teerapol Srichana

References
BP. 2015. British Pharmacopoeia. London: Pharmaceutical Press.
Broadhead, Joanne, Edmond S. K. Rouan, and Christopher T. Rhodes. 1995.
“Dry-powder inhalers: evaluation of testing methodology and effect of
inhaler design.” Pharmaceutica Acta Helvetiae 70 (2):125-31.
Dechraksa, Janwit, Tan Suwandecha, Kittinan Maliwan, and Teerapol
Srichana. 2014. “The comparison of fluid dynamics parameters in an
Andersen cascade impactor equipped with and without a preseparator.”
AAPS PharmSciTech 15 (3):792-801. doi: 10.1208/s12249-014-0102-2.
Geuns, Eduard R. M., Johan S. Toren, Dirk M. Barends, and Auke Bult. 1997.
“Decrease of the stage-2 deposition in the twin impinger during storage of
beclomethasone dipropionate dry powder inhalers in controlled and
uncontrolled humidities.” European Journal of Pharmaceutics and
Biopharmaceutics 44 (2):187-194. doi: 10.1016/S0939-6411(97)00072-6.
Guo, Changning, Stacey R Gillespie, John Kauffman, and William H Doub.
2008. “Comparison of delivery characteristics from a combination
metered-dose inhaler using the Andersen cascade impactor and the next
generation pharmaceutical impactor.” Journal of Pharmaceutical Sciences
97 (8):3321-3334.
Hatch, Theodore, and Sarah P. Choate. 1929. “Statistical description of the
size properties of non uniform particulate substances.” Journal of the
Franklin Institute 207 (3):369-387. doi: 10.1016/S0016-0032(29)91451-4.
Hickey, Anthony J. 2004. Pharmaceutical Inhalation Aerosol Technology. 2nd
ed. Broken Sound Parkway NW: CRC Press.
Hickey, Anthony J., and Latarsha D. Jones. 2000. “Particle-size analysis of
pharmaceutical aerosols.” Pharmaceutical Technology 24 (9):48-50.
Hickey, Anthony J., Ted B. Martonen, and Yadong Yang. 1996. “Theoretical
relationship of lung deposition to the fine particle fraction of inhalation
aerosols.” Pharm Acta Helv 71 (3):185-90.
Hinds, William C. 2012. Aerosol Technology: Properties, Behavior, and
Measurement of Airborne Particles: John Wiley & Sons.
Kempsford, Rodger, Malcom Handel, Rashmi Mehta, Mariza De Silva, and
Peter Daley-Yates. 2005. “Comparison of the systemic pharmacodynamic
effects and pharmacokinetics of salmeterol delivered by CFC propellant
and non-CFC propellant metered dose inhalers in healthy subjects.”
Respiratory Medicine CME 99 Suppl A:S11-9. doi: 10.1016/j.rmed.
2004.11.005.
 Particle Size and Size Characterization of Aerosols 33

Kokhanovsky, Alexander A., and Eleonora P. Zege. 1997. “Optical properties

of aerosol particles: A review of approximate analytical solutions.”
Journal of Aerosol Science 28 (1):1-21. doi: 10.1016/S0021-8502(96)
00058-4.
Kwon, Su B., Keong S. Lim, Ji S. Jung, Gwi N. Bae, and Ki W. Lee. 2003.
“Design and calibration of a 5-stage cascade impactor (K-JIST cascade
impactor).” Journal of Aerosol Science 34 (3):289-300. doi: 10.1016/
S0021-8502(02)00177-5.
Mitchell, Jolyon, and Mark Nagel. 2004. “Particle size analysis of aerosols
from medicinal inhalers.” KONA Powder and Particle Journal 22 (3):32-
65. doi: 10.14356/kona.2004010.
Newman, Stephen P, and Ian R. Wilding. 1998. “Gamma scintigraphy: an in
vivo technique for assessing the equivalence of inhaled products.”
International Journal of Pharmaceutics 170 (1):1-9. doi: 10.1016/S0378-
5173(98)00029-5.
Rojanarat, Wipaporn, Titpawan Nakpheng, Ekawat Thawithong, Niracha
Yanyium, and Teerapol Srichana. 2012. “Levofloxacin-proliposomes:
opportunities for use in lung tuberculosis.” Pharmaceutics 4 (3):385-412.
Sinko, Patrick. 2006. Martin’s Physical Pharmacy and Pharmaceutical
Sciences: Physical Chemical and Biopharmaceutical Principles in the
Pharmaceutical Sciences. Edited by PJ. Sinko. 6th ed. Baltimore, MD:
Lippincott Williams & Wilkin.
Snell, Noel, and David Ganderton. 1999. “Assessing lung deposition of
inhaled medications.” Respiratory Medicine CME 93 (2):123-33.
Srichana, Teerapol, Gary P. Martin, and Christopher M. Marriott. 2000. “A
human oral-throat cast integrated with a twin-stage impinger for
evaluation of dry powder inhalers.” Journal of Pharmacy and
Pharmacology 52 (7):771-778.
Srichana, Teerapol, Gary P. Martin, and Christopher Marriott. 1998.
“Calibration method for the Andersen cascade impactor.” Journal of
Aerosol Science 29 (SUPPL.2):S761-S762.
Srichana, Teerapol, Roongnapa Suedee, and P. Srisudjai. 2003. “Application
of spectrofluorometry for evaluation of dry powder inhalers in vitro.”
Pharmazie 58 (2):125-9.
USP 30-NF 25. 2007. United States Pharmacopeia and National Formulary
(USP 30-NF 25). 39 ed. Vol. 2. Rockville, MD: United States
Pharmacopeia Convention.
Vaughan, Nicholas. 1989. “The Andersen impactor: calibration, wall losses
and numerical simulation.” Journal of Aerosol Science 20 (1):67-90.
34 Teerapol Srichana

Ziegler, Jochen, and Herbert Wachtel. 2005. “Comparison of cascade

impaction and laser diffraction for particle size distribution
measurements.” Journal of Aerosol Medicine 18 (3):311-24. doi:
10.1089/jam.2005.18.311.
 Chapter 2

Flow Properties

2.1. Introduction
In many situations in aerosol technology, there is an interaction between
the particles and the gas, and one must take into account the discontinuous
nature of the gas. That is, the gas cannot be treated as a continuous fluid but
must be considered as an ensemble of rapidly moving molecules colliding
randomly with the particles.
The criterion for using such an approach depends on the particle size
relative to the spacing between the gas molecules. Instead of the average
spacing between molecules, a more useful concept is the mean free path,
which is defined as the average distance traveled by a molecule between
successive collisions. The mean free path of a gas can be determined from the
average number of collisions a particular molecule undergoes in one second
and the average distance it has traveled in that second.

2.2. Reynolds Number

A key to understanding the aerodynamic properties of aerosol particles is
the Reynolds number (Re) which is a dimensionless number that characterizes
the fluid flow through a pipe or around an obstacle such as an aerosol particle.
36 Teerapol Srichana

The Reynolds number has the following properties:

1. It is an index of the flow regime to provide a benchmark that will

determine whether the fluid flow is laminar or turbulent.
2. It is proportional to the ratio of the inertial forces to the viscous forces
acting on each element of the fluid. This ratio is the key to determine
which flow resistance equation is correct in a given situation.
3. Equality of the Reynolds numbers is required for a geometrically
similar flow to occur around geometrically similar objects. This
similarity means that the pattern of the streamlines will be the same
for flow around different-sized objects or even the flow of different
fluids. A streamline is the path traced by a tiny element of fluid as it
flows around an obstacle.

The Reynolds number can be derived by the ratio of the inertial force to a
viscous force acting on an element of fluid in a steady-flow system. The
viscous force is defined by the air viscosity.

Vd
Re  (2.1)


Table 2.1. Property of air at standard conditions:

293K, 101kPa (20°C, 760 mmHg)

Property SI Units cgs Units

Viscosity 1.81×10-4 Pas (N·s/m2) 1.81×10-4 dyne·s/cm2
Density 1.20 Kg/m3 1.20×10-3 g/cm3
Diffusion Coefficient 2×10-5 m2/s 0.20 cm2/s
Mean Free Path 0.066 µm 0.066 µm

The Reynolds number is dimensionless in any consistent system of units.

The quantities associated with inertia are in the numerator and viscosity is in
the denominator. It is important to note that the density is that of the gas and
not the aerosol particles.
The properties of air given in Table 2.1 provide a simpler form of the
Reynolds number for use with SI or cgs units at standard conditions.
Equation 2.1 applies equally well to flow in a pipe or particle motion; the
latter is called the particle Reynolds number.
 Flow Properties 37

The Reynolds number depends only on the relative velocity between an

object such as an aerosol particle and surrounding gas. It is aerodynamically
equivalent for the air to flow past a stationary particle or for the particle to
move through the stationary air at the same velocity. Laminar flow around a
particle occurs at low Reynolds numbers (Re < 1), for which the viscous forces
are much greater than the inertial forces. This kind of flow is characterized by
a smooth pattern of streamlines that are symmetrical on the upstream and
downstream sides of the particle (Figure 2.1(a)). As a Reynolds number
increases above 1.0, eddies are formed downstream.

Figure 2.1. Flow around a sphere. (a) Laminar flow Re = 0.1. (b) Turbulent flow Re =
2. (c) Turbulent flow Re = 250.

The Reynolds number is also important in an in vitro experiment of

aerosol deposition. The complexity of extrathoracic airways leads to different
Reynolds numbers that affect the aerosol deposition pattern (Ma and
Darquenne 2011). Therefore, a correction in the Reynolds number is a critical
step to develop an in vitro model for aerosol assessment (Nicolaou and Zaki
2013).
38 Teerapol Srichana

A flow controller combines a pressure regulator with a metering valve to

maintain a constant flow rate under varying conditions. The regulator
maintains a constant pressure drop across the metering valve.
Unlike the variable-head meter that measures a pressure drop varying with
flow rate, the variable orifice area with the flow rate will maintain a nearly
constant pressure drop. The most common type of variable area meter is the
rotameter (Figure 2.2) that consists of a float that is free to move up and down
in a vertical, tapered tube through which the fluid to be measured passes.
The float rises in the tapered tube until its weight balances the upward
drag force due to the fluid flowing up through the tube. The area between the
float and the tube wall increases as the float rises, and reduces the velocity and
drag force of the fluid.

Figure 2.2. Rotameter.

2.3. Uniform Particle Motion

The most common and perhaps the most important type of particle motion
is steady, straight-line motion. This uniform motion is typically the result of
the action of two forces, a constant external force such as gravity or an
electrical force and the resistance of the gas to particle motion. Analysis of a
uniform particle motion is especially useful for the study of aerosols because
 Flow Properties 39

in most situations aerosol particles come to a constant velocity almost

instantly. The resisting force of the gas depends on the relative velocity
between the particle and the gas and is the same whether the particle moves
through the gas or the gas flows past the particle.
The general equation for the force that resists the motion of a sphere
passing through a gas was derived by Newton as part of a ballistic evaluation
of cannon balls.
Newton’s resistance equation is valid for a wide range of particle motion
but is useful primarily for Reynolds numbers that are greater than 1,000 but
not to aerosol particles. As discussed earlier, Stokes’ law, although derived
from first principles, represents a special case of Newton’s resistance law.
Newton indicated that the resistance encountered by a cannon ball passing
through the air is a result of the breakup of the air that has to be pushed aside
to allow the sphere to travel through it. In one second, a sphere of diameter d
will push aside a volume of gas equal to the projected area of the sphere
multiplied by its velocity (V).
The mass (ṁ) of this amount of gas is,


  g
m d 2V (2.2)
4

Acceleration of the displaced gas is proportional to a relative velocity

between the sphere and the gas, and thus the change of momentum per unit
time is,

change of momentum 
 m V   g d 2V 2 (2.3)
unit time 4

By definition, this rate of change of momentum is equal to the force

required to move the sphere through the gas, the gas resistance or drag force
is given by


FD  K  g d 2V 2 (2.4)
4

Where K is a constant of proportionality. Newton originally thought that K

was independent of the velocity for a given shape. This is true, however, only
40 Teerapol Srichana

when the Reynolds number for particle motion is greater than 1,000. This is
the general form of the resistance equation that is valid for all subsonic
motions of particles. The dimensionless coefficient of drag is constant for
spheres having Re > 1,000.
The derivation of equation 2.4 is based on the inertia of the gas, and it
follows that equation 2.3 is valid only for motion at high Reynolds numbers,
where the inertial forces are much larger than the viscous forces.
Since most aerosol motion problems have either a diameter or a velocity
as one of the unknowns, the Reynolds number cannot be calculated until the
problem is solved.


FD  CD  g d 2V 2 (2.5)
8

24
CD  (1  0.15  Re 0.687 ) (2.6)
Re

Newton’s drag (Re > 1,000) applies to a particle motion for which the
viscous effects of the gas can be neglected compared with the inertial effects.
In 1851, Stokes derived an expression for drag at the other extreme when
inertial forces are negligible compared to viscous forces. As mentioned, the
Reynolds number is a ratio of inertial to viscous forces; consequently, a
condition of negligible inertial forces compared to viscous forces implies a
low Reynolds number and laminar flow. Because of the low velocities and
small particle sizes involved, most aerosol motion occurs at low Reynolds
numbers. Stokes’ law thus has wide application for the study of aerosols, and
it is worthwhile reviewing the derivation of the law and its assumptions and
implications.
Stokes’ law is a solution for the generally unsolvable Navier-Stokes
equations. These equations are the general differential equations that describe
fluid motion. They are derived from the application of Newton’s second law to
a fluid element on which the gravity, pressure, and viscous forces are acting.
The resulting equations are very difficult to solve because they are nonlinear,
partial differential equations.
In general, some simplifying assumptions must be made before they can
be solved. Stokes solution does this by assuming that the inertial forces are
negligible compared to the viscous forces. This assumption eliminates the
 Flow Properties 41

higher order terms in the Navier-Stokes equations and yields linear equations
that can be solved. Further assumptions of Stokes derivation are
incompressible fluid, no walls or other particles nearby, constant motion of the
particle, rigid spheres and zero fluid velocity at the particle’s surface.
Stokes solved the Navier-Stokes equations with the assumptions just
stated to obtain equations for the forces acting at any point in the fluid
surrounding a spherical particle. The net force acting on the particle was
obtained by integrating the normal and tangential forces over the surface of the
particle. The two resulting forces acting in the direction opposite to the particle
motion are the form component,

Fn  Vd (2.7)

and the frictional component.

F  2Vd (2.8)

These components are combined to give the total resisting force on a

spherical particle moving with a velocity V through a fluid. This is Stokes’
law. When the resisting force is experienced by a particle, it can be described
by equation 2.9 and the motion of the particles is said to be in the Stokes
region.

FD  3Vd (2.9)

As a particle moves through a fluid, it deforms the fluid, causing layers of

the fluid in the region around the particle to slide over one another. The
resisting force is the result of the friction of these layers sliding over one
another. The energy spent overcoming this resistance is dissipated as heat
throughout the fluid. In practice, the use of Stokes’ law is restricted to
situations in which the particle Reynolds number is less than 1.0. At a
Reynolds number of 1.0, the error in the drag force calculated by Stokes’ law
is 12% and at a Reynolds number of 0.3 it is 5%.
It is instructive to compare the drag force given by Stokes’ law with that
given by Newton’s law. Stokes’ law contains viscosity, but not factors
42 Teerapol Srichana

associated with inertia. Newton’s law contains g but not viscosity. Solving
equation 2.9 for the coefficient of drag gives

 Re  1
FD  3Vd  CD  g d 2V 2 , for (2.10)
8

This is the equation for the straight-line portion at the left of Figure 2.1.
Because equation 2.10 includes V and d, it changes the functional relationship
of Newton’s equation from an equation having a drag force proportional to V2
and d2 to one with a drag force that is proportional to V and d according to
Stokes’ law (Theerachaisupakij et al. 2002). For conditions between the range
of application of Newton’s and Stokes’ laws, the functional dependence of
drag gradually changes from V2 to V and from d2 to d. This is the curved
portion of Figure 2.1. The frictional component, equation 2.8, represents two-
thirds of the Stokes drag, and the expression is equivalent to equation 2.10 for
16
the flow in tubes where there is no form component and gives C0  , or
Re
two-thirds of equation 2.10.

How do the other assumptions of Stokes’ law limit its application to

aerosol particles? The assumption of an incompressible fluid implies that air is
compressible, but rather that it does not compress significantly near the
particle as the particle moves through it. This is equivalent to assuming that
the relative velocity is much less than the speed of sound, which is nearly
always the case for aerosol particles.
The presence of a wall within 10 diameters of a particle will modify the
drag force on the particle. Because of the small size of the aerosol particles,
only a tiny fraction of the aerosol particles will be within 10 particle diameters
of a wall in any real container or tube.
The correction to Stokes’ law for non-rigid spheres such as water droplets
is generally insignificant. Water droplets settle 0.6% faster than predicted by
the Stokes’ law because of circulations that develop within the droplet caused
by the resisting force at the droplet surface.

24 24
CD   (2.11)
 gVd Re
 Flow Properties 43

In addition to Stokes’ law which is a simplified version of the Navier-

Stokes equations, the advancement in computational power enabled a full
numerical solving of the Navier-Stokes equations using a simulation of the
computational fluid dynamics (CFD). The CFD can deal with the laminar flow
or turbulence flow problems by coupling with a turbulence model such as k-
and k- models. A sophisticated full-scale CFD simulation of the Andersen
cascade impactor (ACI) can visualize flow field and the aerodynamic
parameters that affect the motion of the particles inside the impactor (Figure
2.3) (Dechraksa et al. 2014). The streamline velocity at stage 0 showed a
different clustering pattern for the high velocity where the streamline velocity
was in the traditional induction port (Figure 2.3a, b) and exhibited a higher
airflow velocity than that in the preseparator (Figure 2.3c, d) for both flow
rates (28.3 and 60 LPM).

Figure 2.3. Airflow visualization in the ACI at stage 0 under different inlet profiles. (a)
Stage 0 under a traditional induction port at 28.3 LPM. (b) Stage 0 under a traditional
induction port at 60 LPM. (c) Stage 0 under a preseparator at 28.3 LPM. (d) Stage 0
under preseparator at 60 LPM. (Reprinted from Dechraksa, J. et al., AAPS
PharmSciTech, 15(3), 792-801, 2014. With the permission from Springer).
44 Teerapol Srichana

2.4. Settling Velocity and

Mechanical Mobility
One important application of Stokes’ law is the determination of the
velocity of an aerosol particle undergoing gravitational settling in still air.
When a particle is traveling in air, it reaches the terminal settling velocity very
quickly. Under constant velocity conditions, the drag force on the particle is
equal and opposite to the force of gravity FG.

FD  FG  mg (2.12)

(  P   g ) d 3 g
3Vd  (2.13)
6

Where, g is the acceleration of gravity, P is the density of the particle and

g is the density of the gas. The latter has been included to account for the

buoyancy effect, but this can be neglected because P is much greater

than  g . For example, a water droplet settling in the air has a density ratio

 /  g  800 and neglecting buoyancy introduces an error of only 0.1%.

Solving equation 2.13 for the terminal settling velocity gives:

P d 2 g
VTS  , d  1μm and Re  1.0 (2.14)
18

The terminal settling velocity increases when the particle size increases,
being proportional to the square of the particle diameter. As expected from the
derivation, the settling velocity in the Stokes’ region is inversely proportional
to the viscosity and does not depend on the density of the gas. Aerosol
particles adjust to their terminal settling velocity almost instantly, and VTS is
appropriate for characterizing particle motion in most real situations. Equation
2.14 cannot be used for particles smaller than 1.0 m unless the slip correction
 Flow Properties 45

factor, covered in the next section, is applied (Moshfegh et al. 2010). The
terminal velocity for other kinds of external forces, such as centrifugal force,
can be obtained by derivations similar to that given for gravity. In a centrifugal
force field, the terminal velocity is given by
 P d 2 aC
VTC  (2.15)
18

where, ac is the centrifugal acceleration at the location of the particle.

For a tangential velocity VT and the radius of motion R,

VT2
aC  (2.16)
R

Equation 2.14, is of fundamental importance to aerosol studies. However,

in the form given, it is accurate (±10%) only for determining the settling
velocity of the standard density particles having diameters of 1.5-75 m.
When slip correction is included, it is accurate for particles as small as 0.001
m.
In Stokes’ law, equation 2.9, the resistance force is directly proportional to
the velocity. From this relationship, we can define the particle mobility (B) as
a measure of the relative ease of producing a steady motion for an aerosol
particle.

V 1
B  , d  1 m (2.17)
FD 3 d

Mobility is the ratio of the terminal velocity of a particle to the steady

force producing that velocity. It has units of m/N.s (cm/dyne. s) and is often
called the mechanical mobility to distinguish it from any electrical mobility.
The terminal velocity of an aerosol particle is simply the force times mobility;
for example,

VTS  FG  B (2.18)

Current research has indicated that an aerosol cloud and individual particle
express fairly different settling behaviors. The effect of long-range
46 Teerapol Srichana

interparticle with hydrodynamic interactions in the aerosol cloud results in

their movement being significantly faster than individual particles with the
same magnitude of external forces (Yang et al. 2010).

2.5. Slip Correction Factor

An important assumption of Stokes’ law is that the relative velocity of the
gas right at the surface of the sphere is zero. This assumption is not met for
small particles whose size approaches the mean free path of the gas. Such
particles settle faster than one predicted by Stokes’ law because there is “slip”
at the surface of the particle. At standard conditions, this error becomes
significant for particles with a diameter less than 1 μm. In 1910, Cunningham
derived a correction factor for Stokes’ law to account for the effect of slip. The
factor called the Cunningham correction factor C was always greater than one
and reduced the Stokes drag force by equation 2.19.

3Vd
FD  (2.19)
CC

where,

2.52
CC  1  (2.20)
d

where  is the mean free path and is the particle diameter.

Use of the Cunningham correction factor (equation 2.20), extended the
range of applications of Stokes’ law to particles of 0.1 m in diameter. This
range can be extended to still smaller particles by empirical equations based on
experimental measurements of the slip. This factor is called the slip correction
factor and must be used in the form of equation 2.21 for particles that are less
than 0.1 μm in diameter. The slip-corrected form of the terminal settling
velocity becomes Re < 1.0 (equation 2.22).
Equation 2.22 is valid for all particle sizes when Re < 1.0 and equation
2.21 is used for
 Flow Properties 47

  d 
CC  1   2.34  1.05exp  0.39   (2.21)
d   

The slip correction factor for a 1.0 m particle under standard conditions
is 1.15; that is, the particle settles 15% faster than predicted by equation 2.14,
which is based on the uncorrected form of the Stokes’ law. For the particle
size of less than 1 m, the slip increases rapidly as the size decreases, where
the Cunningham or slip correction factor must be used (Moshfegh et al. 2009).
For accurate work, it should be used for all particles less than 5 or 10 μm. It is
commonly stated that a slip correction is necessary for particles as their size
approaches the mean free path because “the particles are so small they slip
between the molecules” (Hinds 2012). This is incorrect, but a useful way to
remember how to apply the slip correction factor (Moshfegh et al. 2010). Slip
correction increases as the pressure decreases because the mean free path
increases.

 P d 2 gCC
VTS  , for Re  1.0 (2.22)
18

2.6. Nonspherical Particles

The equivalent volume diameter (de) is the diameter of a sphere having the
same volume as that of the irregular particle. This sphere is called the
equivalent volume sphere. The equivalent volume diameter can be imagined as
the diameter of the sphere that would result if the irregular particle melted to
form a droplet.
Dynamic shape factors for particles of various shapes are given in Table
2.2.
Values for the geometric shapes were determined by measuring the
settling velocity of geometric models in liquids. For irregular particles, settling
velocities were measured indirectly using elutriation devices. Values given in
the table are averaged over all orientations, which is the usual situation for the
motion of the aerosol particle (Re < 0.1) because of the Brownian motion of
the particles. There is an exception for certain streamlined shapes in which the
dynamic shape factor is greater than 1.0.
48 Teerapol Srichana

Table 2.2. Dynamic shape factors for particles of various shapes.

(Adapted from Hinds, W. C., John Wiley & Sons, 2012)

Shape Shape factor Dynamic shape factor

Sphere 1.00 Axial ratio
Cube 1.08 2 5 10
Cylinder* 1.09 1.23 1.43
Straight chain 1.10 1.35 1.68
Three spheres 1.15
Four spheres 1.17
*orientation average

This means that non-spherical particles settle more slowly than their
equivalent volume spheres (Mand and Rosendahl 2010).

FD
X (2.23)
3Vde

FD  3Vde X (2.24)

2.7. Aerodynamic Diameter

An equivalent diameter (de) that has found wide applications in aerosol
technology is the aerodynamic diameter da. This is defined as the diameter of
a spherical particle with a density of 1 g/cm3 that has the same settling
velocity. The aerodynamic diameter standardizes for a spherical shape and unit
density (1 g/cm3).
A related equivalent diameter is the Stokes diameter dst which is the
diameter of a sphere that has the same density and settling velocity as the
particle.
Equation 2.25 can be written in terms of these diameters that neglects any
slip correction,
 P de2 g
VTS  (2.25)
18 X
where, 0 is the standard particle density, 1.0 g/cm3.
 Flow Properties 49

The aerodynamic diameter can be defined as the diameter of a water

droplet that will have the same aerodynamic properties as the particle. If a
particle has an aerodynamic diameter of 1 µm, it behaves in an aerodynamic
sense like a 1 µm water droplet regardless of its shape, density or physical
size. Furthermore, it is aerodynamically indistinguishable from other particles
of different size, shape, and density having aerodynamic diameters of 1 µm.
The Stokes’ diameter is usually defined in terms of the normal density of the
bulk material of the particle  b . This definition avoids the problem of defining

the true density of the particle which may be less than b due to porosity,
occlusions or an agglomerated structure.

 P de2 g 0 da2 g b d st2 g

VTS    (2.26)
18 X 18 18

2.8. Settling at High Reynolds Numbers

For particle motion in the Stokes region, the settling velocity can be
determined explicitly if the particle diameter and density are known; the
diameter can be found if the velocity is known. However, for particle motion
with Re > 1.0, this is not the case.

References
Dechraksa, Janwit, Tan Suwandecha, Kittinan Maliwan, and Teerapol
Srichana. 2014. “The comparison of fluid dynamics parameters in an
Andersen cascade impactor equipped with and without a preseparator.”
AAPS PharmSciTech 15 (3):792-801. doi: 10.1208/s12249-014-0102-2.
Hinds, William C. 2012. Aerosol Technology: Properties, Behavior, and
Measurement of Airborne Particles. 2nd ed. Noboken, NJ: John Wiley &
Sons, Inc.
Ma, Baoshun, and Chantal Darquenne. 2011. “Aerosol deposition
characteristics in distal acinar airways under cyclic breathing conditions.”
Journal of Applied Physiology 110 (5):1271-1282. doi: 10.1152/jappl
physiol.00735.2010.
50 Teerapol Srichana

Mand, Matthias, and Lasse Rosendahl. 2010. “On the motion of non-spherical
particles at high Reynolds number.” Powder Technology 202:1-13.
Moshfegh, Abouzar, Mehrzad Shams, Goodarz Ahmadi, and Reza Ebrahimi.
2009. “A novel surface-slip correction for microparticles motion.”
Colloids and Surfaces A: Physicochemical and Engineering Aspects
345:112-120.
Moshfegh, Abouzar, Mehrzad Shams, Goodarz Ahmadi, and Reza Ebrahimi.
2010. “A new expression for spherical aerosol drag in slip flow regime.”
Journal of Aerosol Science 41:384-400.
Nicolaou, Laura, and Tamer A. Zaki. 2013. “Direct numerical simulations of
flow in realistic mouth–throat geometries.” Journal of Aerosol Science
57:71-87. doi: 10.1016/j.jaerosci.2012.10.003.
Theerachaisupakij, Woraporn, Shuji Matsusaka, Y. Akashi, and Hiroaki
Masuda. 2002. “Reentrainment of deposited particles by drag and aerosol
collision.” Journal of Aerosol Science 34:261-274.
Yang, Wei, Keat Theng Chow, Bo Lang, Nathan P. Wiederhold, Keith P.
Johnston, and Robert O. Williams. 2010. “In vitro characterization and
pharmacokinetics in mice following pulmonary delivery of itraconazole as
cyclodextrin solubilized solution.” European Journal of Pharmaceutical
Sciences 39 (5):336-347. doi: 10.1016/j.ejps.2010.01.001.
 Chapter 3

Forces and Interactions

3.1. Introduction
The terminal velocity of a particle moving in the Stokes region is directly
proportional to the net external force F acting on the particle. Here, ‘external
force’ means a force acting remotely on a particle, such as gravity, centrifugal
or electrostatic forces. The drag force is not considered an external force. The
constant of proportionality is the mechanical mobility B that is defined by an
Equation 3.1.

VTS  BF (3.1)

When the external force is the force of gravity, equation 3.1 becomes

VTS  BFG  Bmg (3.2)

The product of particle mass and mobility, mB, frequently occurs in

aerosol mechanics and it is a useful quantity for the analysis of the complex
motion of a particle. This quantity is called the relaxation time of the particle
and is given the symbol  .
The term “relaxation time” is used because it characterizes the time
required for a particle to adjust or “relax” its velocity to a new set of force
conditions. The relaxation time is analogous to a characteristic acceleration
52 Teerapol Srichana

time for a moving object. It depends only on the mass and mobility of the
particle and is not affected by the nature or magnitude of the external forces
acting on the particle. Although it is useful to think of the relaxation time as a
particle property, it includes viscosity and slip correction and is thus affected
by the temperature and pressure of the surrounding gas (Moshfegh et al. 2010,
2009). The use of relaxation time as defined by equation 3.2 is restricted to
particle motion in the Stokes region, where Re < 1. Relaxation time increases
rapidly with particle size because it is proportional to the square of the
diameter as described in equation 3.3.

 Cc   p d Cc 0 d a2Cc

2

  mB   p d  3
  (3.3)
6  3 d  18 18

Relaxation time can be used to simplify the calculation of a particle’s

terminal settling velocity. Substituting τ in equation 3.2 gives

VTS   g (3.4)

The terminal velocity of a particle is simply the product of τ and the

acceleration caused by an external force. For any constant external force F
acting on a particle of mass m, the terminal velocity is

dV t 
F  m  mat  (3.5)
dt

The equations for a particle’s terminal settling velocity that are derived in
the previous chapter ignore the acceleration of the particle considering only
the equilibrium condition when the forces acting on the particle are balanced
and the velocity of the particle is constant. We now consider the acceleration
of a particle that is released with zero initial velocity in still air. The particle
quickly reaches its terminal settling velocity, and we wish to know how long
that takes and the nature of the acceleration process. Newton’s second law of
motion must hold at every instant during the acceleration process.
 Forces and Interactions 53

Figure 3.1. Acceleration of a particle in still air.

dV (t )
FG  FD  mg  3V  t  d  m (3.6)
dt

Multiplying both sides of equation 3.6 by the particle mobility B gives

equation 3.7.
Replacing g with VTS and rearranging, we obtain

dV (t )
mBg  3V  t  Bd  mB (3.7)
dt

dV (t )
 g  V  t   (3.8)
dt

t V (t )
dt dV (t )

0
 
0
VTS  V (t )
(3.9)

  lnVTS  V t   ln VTS
t
(3.10)


VTS  V t 
e t   (3.11)
VTS
54 Teerapol Srichana

3.2. Stopping Distance

The analysis of particle acceleration due to constant external forces can be
carried one step further by replacing V(t) in equation 3.11 with dx/dt and
integrating to get the displacement along the x-axis as a function of time, x(t):
The Stokes number is defined in equation 3.12 where the stopping
distance (S) is the total distance traveled or the inertial range. On a
displacement scale, the stopping distance represents a measure of a particle’s
effective initial momentum which is diminished to zero by air friction over a
distance equal to the stopping distance. As τ is defined as the product of mB,
the stopping distance can be defined in terms of the product of a particle’s
mobility and its initial momentum. Dj is the jet diameter.
In a mathematical sense, it will take an infinite time for a particle to travel
its entire stopping distance, but in a practical sense, the particle travels 95% of
that distance very rapidly in a time equal to 3 τ. The stopping distance
represents the ultimate distance a particle will travel in still air if the external
force acting on the particle is suddenly turned off, as might be the case for an
electrical force. A more important use of the stopping distance is for a particle
moving with an airstream that is abruptly turned 90°. In this case, the stopping
distance represents the distance the particle continues to travel in its original
direction and thus can be thought as a measure of the “persistence” of the
particle. The preceding discussion of stopping distance assumes that the entire
motion of a particle takes place within the Stokes region. The stopping
distance is the most important for large particles with high velocities that
frequently have motion, at least initially outside the Stokes region. This
situation is very difficult to analyze.
The situation is more complicated for flow around an obstacle. Very small
particles with negligible inertia will follow the gas streamlines perfectly. Large
and heavy particles will tend to continue in a straight line, regardless of what
the gas flow does. The particles of interest are those whose motion lies
between these extremes.
To analyze this type of motion, one first defines the flow field and the
pattern of streamlines for the gas flow around the obstacle. This can be a
difficult fluid mechanics problem. Once the flow field is defined, the velocity
and direction of the flow are known for every point near the obstacle. The
actual particle trajectory is determined as it proceeds through the flow field.
The analysis can be done only for very simple geometries, such as flow around
a spherical or cylindrical obstacle. With more complicated shapes, it can be
 Forces and Interactions 55

done numerically by following a series of incremental steps through the flow

field. The forces on the particles are assumed to be constant during each step
and are re-evaluated for each step. Usually, this analysis is done for a large
number of starting positions of the particle relative to the obstacle.

3.3. Stokes Number

Curvilinear motion is characterized by a dimensionless number called the
Stokes number (Stk). It is the ratio of the stopping distance of a particle (S) to
a characteristic dimension of the obstacle (dc). For example, for a flow
perpendicular to a cylinder of diameter dc, the Stokes number is an equation
3.12.

S U 0
Stk   , for R e  1.0 (3.12)
dc dc
Where, Uo is the undisturbed air velocity well away from the cylinder and
Reo = ρg dpU0/η when Re > 1.0. The Stokes number is also the ratio of the
particle relaxation time to the transit time past an obstacle or the ratio of the
time τ it takes for a particle to adjust to the time dcU available for adjustment.
When Stk >> 1, particles continue moving in a straight line when the gas
turns; when Stk << 1, particles follow the gas streamlines perfectly. Because
the characteristic dimension dc in equation 3.12 can be defined differently for
different applications, the definition of Stokes number may be application
specific.
For a geometrically similar particle motion to occur around different-sized
cylinders, two conditions must be met: (1) the flow Reynolds numbers for the
two situations must be equal, and (2) the Stokes numbers must be equal.
Equality of the Reynolds numbers ensures that the gas flows are similar, and
the equality of the Stokes numbers ensures that the particle motion in the flow
fields is also similar. The Stokes number is the ratio of a particle’s
“persistence” to the size of the obstacle.
As the Stokes number approaches zero, particles track the streamlines
perfectly. As the Stokes number increases, the particles resist changing their
direction when the gas streamlines change directions. The Stokes number is
used to characterize the inertial impaction which is the inertial transfer of
particles onto surfaces that is described in the next section.
56 Teerapol Srichana

The concept of an inertial impaction has evolved into a virtual impactor.

The airborne particles do not physically collide with a collection plate as
occurs in a conventional impactor in which the particles are selectively filtered
by the airflow. The virtual impactor was designed to split the airflow into a
major and minor flow (Marple and Olson 2011). The minor flow mimics the
impaction plate in a conventional impactor (Lee et al. 2014). It captures the
particles that are larger than the cut-off size along with the flow. The major
flow separates particles that are smaller than the cut-off of the virtual
impactor. Recently, a ring-shaped nozzle was developed as a novel aerosol
impactor (Son et al. 2015). The collection efficiency curve and Stk50 of an
impactor with an orifice become respectively steeper and smaller than an
impactor without an orifice.

3.4. Inertial Impaction

Impaction is a special case of curvilinear motion of aerosol particles. As a
result of its importance, impaction has been analyzed theoretically and
experimentally more than any other aerosol separation process (Erdal and
Esmen 1990). In the first half of this century, impaction was a common
method for collecting dust for the evaluation of occupational environments.
Since the 1960s, cascade impactor instruments based on impaction have been
used extensively for the measurement of particle size distributions by mass.
All inertial impactors operate on the same principle. An aerosol is passed
through a nozzle and the output stream (jet) directed against a flat plate
(Figure 3.2). The flat plate is called an impaction plate which deflects the flow
abruptly in a 90° bend in the streamlines. Particles whose inertia exceeds a
certain value are unable to follow the streamlines and will impact on the plate.
It is assumed that larger particles will stick to the surface if they hit
whereas the smaller particles will remain airborne, follow the streamlines and
flow out of the impactor. Thus, an impactor separates aerosol particles into
two size ranges; particles larger than a certain aerodynamic size are removed
from the airstream and those smaller than that size remains in the airstream.
Impactor theory seeks to explain the shape of the curve of the collection
efficiency versus the particle size (Figure 3.3). The parameter that governs the
collection efficiency is the Stokes number or the impaction parameter which is
defined for an impactor as the ratio of the particle stopping distance at the
average nozzle exit velocity U to the jet radius Dj/2.
 Forces and Interactions 57

Figure 3.2. Cross-sectional view of an impactor. (Adapted from Hinds, W. C., John
Wiley & Sons, 2012).

Figure 3.3. Typical impactor efficiency.

58 Teerapol Srichana

This Stokes number is slightly different from equation 3.12, being defined
in terms of the nozzle radius Dj/2 rather than a characteristic dimension of an
obstacle. For an impactor with a rectangular nozzle, the jet half-width should
be used in place of the jet radius in equation 3.13.

U  p d p2UCc
Stk   (3.13)
Dj / 2 9 D j
Although the collection efficiency is a function of the Stokes number,
there is no simple relationship between the impactor collection efficiency and
the Stokes number. Theoretical determination of the characteristic efficiency
curve for an impactor requires numerical analysis using a computer (Abouali,
Saadabadi, and Emdad 2011). First, the pattern of streamlines in the vicinity of
the jet is determined by solving the Navier-Stokes equations for a particular
impactor geometry (Dechraksa et al. 2014). Then, for a given particle size,
particle trajectories are determined for each that enters the streamline. The
efficiency for that particle size is determined by the fraction of the trajectories
that intercept the impaction plate. This process is repeated for many particle
sizes in order to generate the characteristic impactor efficiency curve such as
the one shown in Figure 3.3.
The collection efficiency curves for the impactors are often plotted in a
general form as efficiency versus the square root of the Stokes number (Stk0.5)
which is directly proportional to the particle size. Experimental calibration
requires efficient measurements made with a series of monodisperse aerosols
(Srichana, Martin, and Marriott 1998). However, this needs to be done at least
once for each impactor design, because all geometrically similar impactors
that meet the recommended design criteria, given later, will have the same
collection efficiency when operated at the same Stokes number. The simplified
analysis serves to illustrate the process of impaction and the importance of the
relevant parameters. It is necessary to make the simplifying assumption that
the flow velocity is uniform in the jet and the streamlines are arcs of a circle
with their centers at A. Figure 3.4 illustrates a cross-sectional view of a
rectangular jet impactor. Because of the symmetry, only one-half of the
impactor needs to be considered. A particle exiting the nozzle along a
streamline experiences a centrifugal force causing it to move toward the
impaction plate. If these departures are slight, as will be the case for the
limiting condition that separates impaction from no impaction, the particle will
depart from its original streamline with a constant radial velocity (Vr) while
traversing the curved part of the streamline. This velocity is given by
 Forces and Interactions 59

U 2
Vr  a r  (3.14)
r

where, r is the radius of curvature of the streamline.

Figure 3.4. Simplified impactor model. (Adapted from Hinds, W. C., John Wiley &
Sons, 2012).

3.5. Straight-Line Acceleration and

Curvilinear Particle Motion
If the velocity remains constant in the curved-streamline region, the total
radial displacement of the particle from its original streamline is the product of
its radial velocity and the time required to traverse the curved portion (quarter
circle) of the streamline.
60 Teerapol Srichana

At the end of the quarter circle, each particle will have moved a
distance  from its original streamline in the direction of the impaction
surface. Any particle passing through the nozzle within a distance  of the jet
centerline will end up impacting on the plate (Figure 3.4). Particles located
more than a distance  from the centerline will be shifted from their original
streamlines, but not enough to reach the impaction surface. Because Figure 3.4
is a transverse section of a rectangular nozzle, the impaction efficiency (EI) the
fraction of entering particles collected is equal to the ratio of the lengths 
and h.

3.6. Adhesion of Particles

Aerosol particles attach firmly to any surface they contact. This is one of
the characteristics that distinguish them from gas molecules and from
millimeter-sized particles. Whenever aerosol particles contact one another,
they adhere and form agglomerates.
Filtration and other particle collection methods rely on the adhesion of
particles to surfaces. The adhesive forces on micrometer-sized particles exceed
other common forces by orders of magnitude.
Despite its importance, particle adhesion is poorly understood and its
description is partly qualitative. Because it is such a complicated phenomenon,
no complete theory accounts for all the factors that influence adhesion. Much
of the experimental work on particle adhesion has been conducted with ideal
surfaces under special conditions, such as a high vacuum, which have little
relevance to real surfaces of practical interest.

3.7. Adhesive Forces

The main adhesive forces are the van der Waals force, the electrostatic
force and the force arising from the surface tension of adsorbed liquid films.
All of these forces are affected by the material shape, the surface roughness
and the size of the particle, the material roughness and the contamination of
the surface, the relative humidity, temperature, the duration of the contact, and
the initial contact velocity (Cheng, Dunn, and Brach 2002, Karner, Littringer,
and Urbanetz 2014).
 Forces and Interactions 61

First, consider the theoretical description of the adhesive forces. The most
important forces are the London-van der Waals forces which are the long-
range attractive forces that exist between molecules. These forces are long
range in comparison to chemical bonds which are called short-range forces.
Because of the shielding effects of the adsorbed layers of water and organic
molecules, chemical bonds are not important for the adhesion of aerosol
particles under ambient conditions. The van der Waals forces arise because of
the random movement of the electrons in any material that creates momentary
areas of concentrated charge called dipoles. At any instant, these dipoles
induce complementary dipoles in neighboring materials which in turn produce
attractive forces (Figure 3.5). Van der Waals forces decrease rapidly with the
separation distance between the surfaces; consequently, their influence extends
only a few molecular diameters away from a surface.

Figure 3.5 van der Waals adhesive forces.

At a submicroscopic level, most surfaces are irregular with peaks called

asperities, and valleys (Figure 3.6). At least initially, contact between a
particle and a surface occurs only at a few asperities. Most of the material is
separated by an average distance x that depends on the scale of surface
roughness (Figure 3.6). For smooth surfaces, this distance is usually assumed
to be 0.4 nm which is about the size of the molecules involved.
62 Teerapol Srichana

Figure 3.6. Submicroscopic surface contact geometry.

The net effect of the van der Waals forces is determined by integrating the
forces between all pairs of molecules of a spherical particle near a flat surface.
The resulting adhesive force between the particle and a plane surface is

Ad
F (3.15)
12 x 2

where, A is the Hamaker constant which depends on the materials involved

and ranges from 6 x 10-20 to 150 x 10-20 J for common materials. Equation 3.15
applies to hard materials with negligible flattening in the contact area. After
the initial particle contact, the van der Waals and electrostatic forces gradually
deform the asperities to reduce the separation distance and increase the contact
area until the attractive forces balance the forces resisting deformation. The
deformation process may take as long as a few hours. The hardness of the
materials involved controls the size of the ultimate area of contact and
therefore the strength of the adhesive force.
Most particles of size 0.1 µm or larger carry some small net charge q
which induces an equal and opposite charge on the surface. Coulomb’s law is
one basis to describe the electrostatic attraction of charged particles. The
magnitude of the electrostatic force is defined by two factors that are the
amount of the charges that particles carry and the distance between the
charged particles (Jiang, Hansen, and Raymond 2010). This gives an attractive
electrostatic force of
 Forces and Interactions 63

KE q2
FE  (3.16)
xq2

where, KE is a constant of proportionality and xq is the separation distance of

opposite charges which may be different from the separation distance between
the surfaces.
Particles of insulating material at low humidities retain their charge and
are held to surfaces by the electrostatic force. The electrostatic charges during
the aerosolization ranged from 0.1 to 0.3 nC/mg from a 30–90 LPM flow rate.
Thus, electrostatic effect can be omitted and it does not have any significant
role in the aerosolization performance (Suwandecha et al. 2014). The
equilibrium charge carried by particles larger than 0.1µm is approximately
proportional to d , so the electrostatic adhesion force is also proportional to
the first power of the particle diameter.
Under normal conditions, most materials have adsorbed liquid molecules
on their surface. An attractive force between a particle and a surface is created
by the surface tension of the liquid drawn into the capillary space at the point
of contact (Figure 3.7). For relative humidities greater than 90% and ideal
smooth surfaces, this force is

FS  2 d (3.17)

where,  is the surface tension of the liquid. For real surfaces at lower relative
humidities, the force depends on the curvature of the asperities at the points of
contact and not on the particle diameter. This curvature varies greatly from
particle to particle and gives rise to a distribution of adhesive forces for the
same sized particles.
The three adhesive forces just discussed are all proportional to the first
power of the particle diameter, and this is the form of most empirical
expressions for an adhesive force. Except for highly charged particles, van der
Waals and surface tension forces are greater than electrostatic forces.
64 Teerapol Srichana

Figure 3.7. Adhesive forces due to a liquid film.

Experimental measurements of adhesive forces are made by determining

the force required to separate a particle from a surface. They may be direct
measurements using a centrifugal force (Ahmad, Ungphaiboon, and Srichana
2015), or the indirect measurements using vibration or air currents to remove
the particles. For hard materials and clean surfaces a useful empirical
expression for the adhesive force, based on direct measurement of glass and
quartz particles (> 20 µm) at 25°C is given in Newtons for force, in meters for
the particle diameter and in percent for relative humidity (%RH).

3.8. Detachment of Particles

The force required to detach particles from a surface can be measured by
subjecting the particles to a centrifugal force normal to the surface in a
centrifuge and determining the rotational speed required to detach the
particles. These experiments yield a distribution of forces required to remove
monodisperse particles from a surface. About 10 times as much force is
required to remove 98% of the particles as that required for 50% removal
(Ahmad, Ungphaiboon, and Srichana 2015). Particles can also be removed by
air currents directed at the surface.
Adhesive forces are proportional to d, while removal forces are
proportional to d3 for gravitational, vibrational and centrifugal forces, and d2
for air currents. These relationships indicate that as the particle size decreases,
it becomes more difficult to remove particles from surfaces. This fact agrees
without intuition that large, visible particles such as grains of sand can be
removed by shaking or air currents, but smaller ones cannot, although they
may be removed by washing. The important point is that the adhesive force on
a particle size of less than 10 µm is much greater than other forces on such a
particle (Begat et al. 2004).
 Forces and Interactions 65

While individual particles of less than 10 µm are not likely to be removed

by common forces, a thick layer of such particles may be easily dislodged in
large chunks (0.1-10 mm). The particles adhere tightly to each other and form
large agglomerates that can be easily blown or shaken from the surface.

3.9. Resuspension
Closely related to adhesion and detachment of particles is the process of
resuspension which can be defined as the detachment of a particle from a
surface and its transport away from the surface (Selvam et al. 2010).
Resuspension may occur as a result of air jets, mechanical forces, impaction of
other particles or electrostatic forces. Re-entrainment or blow-off, a more
specific term, refers to resuspension by a jet of air. As with other aspects of
adhesion, we understand the nature of the process but are unable to predict
reliably when resuspension will occur. Re-entrainment of deposited particles
may occur by drag and aerosol collision (Theerachaisupakij et al. 2002).
Unlike static adhesion, described in the preceding sections, resuspension
may involve rolling or sliding of the particle before it becomes airborne.
Forces required to remove a particle under these circumstances are
approximately 1% less than those required for static detachment in a
centrifuge. To analyze re-entrainment by an air jet or by airflow in a tube, we
consider two cases: individual particles on a surface (a sparse monolayer) and
an adhering dust layer wherein particles contact each other. While it is
relatively easy to determine the jet or duct velocity and whether a particle has
been removed, it is difficult to determine the air velocity and resulting drag
force on a particle adhering to a surface.
Re-entrainment is a stochastic process in which a given condition of air
velocity permits one to estimate only the fraction of particles of a given size
that will be removed from a surface. The larger the particle and greater the air
velocity, the greater the probability of re-entrainment.
For the usual case of a turbulent airstream, there is a thin layer of laminar
airflow at the surface called the boundary layer or laminar sublayer. Particles
smaller than this layer are partially protected from re-entrainment by being
submerged in the boundary layer. Re-entrainment of particles submerged in
the boundary layer occurs as a result of occasional bursts of turbulent eddies
penetrating through the boundary layer to detach particles. Thus, such re-
entrainment is time dependent.
66 Teerapol Srichana

The situation is different if a layer of particles is present on a surface. Two

processes may occur. First, particles may be re-entrained from the top surface
of the layer as individual particles or small clusters. This is called erosion.
Alternatively, a whole section of the layer can be re-entrained in a process
called denudation. The occurrence of such process depends on the relative
strength of the adhesive forces between the particles and between the particles
and the surface. Erosion increases with the duration of exposure to an air jet,
while denudation is complete in less than a second.

3.10. Particle Bounce

When a solid particle contacts a surface at a low velocity that is less than a
few m/s the particle loses its kinetic energy by deforming itself and the
surface. The greater the velocity, the greater the deformation and better the
adhesion. At high velocities, part of the kinetic energy is dissipated in the
deformation process (plastic deformation) and part is converted elastically to
the kinetic energy of rebound (Wall et al. 1990). If the rebound energy exceeds
the adhesion forces, a particle will bounce away from the surface. Bouncing
can occur in particle sizes that would adhere tightly in a static situation; it does
not occur for droplets or easily deformed materials such as tar. Maximum
rebound velocities occur at intermediate approach velocities.
The problem of particle bounce has been studied for solid particle
collection in impactors and fibrous filters. The harder the particle material, the
larger the particle or the greater its velocity, a bounce will more occur,
although surface roughness plays a significant role. Coating the surfaces with
oil or grease increases the adhesion energy, the deformation, and the
dissipative energy greatly reduces the problem of bounce (Srichana, Martin,
and Marriott 1998, 2000).
There are two approaches to define the conditions at which bounce will
occur. One approach defines the limiting adhesion or kinetic energy and the
other defines a critical velocity Vc for which bounce will occur if that velocity
is exceeded.


Vc  (3.18)
da
 Forces and Interactions 67

The latter forms equation 3.18 where  is a constant that depends on the
materials used and the geometry of the situation. For example,  = 2 x 10-6
m2/s defines an upper limit of velocity for which bounce will not occur on
uncoated metal impaction plates. Measurements of Vc by Wall et al. (1990) for
ammonium fluorescein particles impacting four target materials gave
 values that ranged from 7.4 x 10-6 to 2.9 x 10-5.
The kinetic energy (KEb) required for bounce to occur when a particle
collides with a surface is given by Dahneke (1971) as

KE b 

d p A 1  e2 
2 (3.19)
2 xe

where, x is the separation distance, defined previously, A is the Hamaker

constant, and e is the coefficient of restitution which is equal to the ratio of the
rebound velocity to the approach velocity. Representative values for e range
from 0.73 to 0.81 (Wall et al. 1990). A and e depend only on the particle
material and its surface.
In practice, one relies on the experimental determination of these
constants. A regression by Ellenbecker, Leith, and Price (1980) gives the
probability of bounce (Pb) for irregular fly ash particles with CMD = 0.14 μm
versus their approach kinetic energy in J.
For cgs units, KE is in ergs and 0.000224 is replaced by 0.00958.
Equation (3.20) agrees reasonably well with measurements made by
Suwandecha, Wongpoowarak, and Srichana (2016). The equation gives 0 and
50% probability of bounce at KEs of 2 x 10-16 and 4 x 10-15 J, respectively.
These values correspond to the approach velocities of 0.03 m/s and 0.12 m/s
for 10 μm particles of standard density.

Pb  1  0.000224  KE 
0.233
(3.20)

3.11. Diffusion Coefficient

Brownian motion is the random wiggling motion of an aerosol particle in
the still air caused by random variations in the relentless bombardment of gas
molecules against the particle. Diffusion of aerosol particles is the net transport
68 Teerapol Srichana

of these particles in a concentration gradient. This transport is always from a

region of higher concentration to a region of lower concentration. Both
processes are characterized by the particle diffusion coefficient D. The larger
the value of D, the more vigorous the Brownian motion and more rapid the
mass transfer in a concentration gradient. The diffusion coefficient is the
constant of proportionality that relates the flux J of aerosol particles (the net
number of particles traveling through unit cross section each second) to the
concentration gradient dn/dx. This relationship is called Fick’s first law of
diffusion. In the absence of any external forces, Fick’s law is

J   D dn dx (3.21)

The diffusion coefficient of an aerosol particle can be expressed in terms

of the particle properties by the Stokes-Einstein derivation. In this derivation,
the diffusion force on the particles that causes their net motion down the
concentration gradient, is equated to the force exerted by the gas resisting the
motion of the particles.
Einstein (1905) showed that (1) the observable Brownian motion of an
aerosol particle is equivalent to that of a giant gas molecule; (2) the kinetic
energy of an aerosol particle undergoing Brownian motion is the same as that
of the gas molecules within which it is suspended (KE = 3/2kT); (3) the
diffusion force on a particle is the net osmotic pressure force on that particle.
Osmotic pressure is best understood by considering dissolved molecules
in liquids. A semipermeable membrane permits liquid molecules to pass
through it unimpeded but prevents dissolved molecules from passing through.
The membrane is free to slide to the left or right. If it is moved to a given
position toward the left, it must be applied to hold it in place. This force equals
the net osmotic pressure force acting to the right. The latter can be thought of
as a pressure created by the high concentration of dissolved molecules on the
left side of the membrane or pressure caused by the liquid trying to get to the
region of low liquid concentration (due to the high concentration of dissolved
molecules) and equalize the concentration. The force is directly proportional to
the difference in concentrations on either side of the membrane. These
concepts apply equally well to particles suspended in gasses and to dissolved
molecules in liquids.
 Forces and Interactions 69

References
Abouali, Omid, Saeideh Saadabadi, and Homayoon Emdad. 2011. “Numerical
investigation of the flow field and cut-off characteristics of
supersonic/hypersonic impactors.” Journal of Aerosol Science 42 (2):65-
77. doi: 10.1016/j.jaerosci.2010.11.006.
Ahmad, Md Iftekhar, Suwipa Ungphaiboon, and Teerapol Srichana. 2015.
“The development of dimple-shaped chitosan carrier for ethambutol
dihydrochloride dry powder inhaler.” Drug Development and Industrial
Pharmacy 41 (5):791-800. doi: 10.3109/03639045.2014.903493.
Begat, Philippe, David A. V. Morton, John N. Staniforth, and Robert Price.
2004. “The cohesive-adhesive balances in dry powder inhaler
formulations II: Influence on fine particle delivery characteristics.”
Pharmaceutical Research 21 (10):1826-1834.
Cheng, W., Patrick F. Dunn, and Raymond M. Brach. 2002. “Surface
roughness effects on microparticle adhesion.” The Journal of Adhesion 78
(11):929-965
Dahneke, Barton. 1971. “The capture of aerosol particles by surfaces.” Journal
of Colloid and Interface Science 37 (2):342-353.
Dechraksa, Janwit, Tan Suwandecha, Kittinan Maliwan, and Teerapol
Srichana. 2014. “The comparison of fluid dynamics parameters in an
Andersen cascade impactor equipped with and without a preseparator.”
AAPS PharmSciTech 15 (3):792-801. doi: 10.1208/s12249-014-0102-2.
Einstein, Albert. 1905. “On the movement of small particles suspended in
stationary liquids required by the molecular-kinetic theory of heat.”
Annalen der Physik 17:549-560.
Ellenbecker, Michael J., David Leith, and John M. Price. 1980. “Impaction
and particle bounce at high Stokes numbers.” Journal of the Air Pollution
Control Association 30 (11):1224-1227. doi: 10.1080/00022470.
1980.10465173.
Erdal, Serap, and Nurtan A. Esmen. 1990. “The curvilinear motion of coarse
particles: New thoughts on theory and applications.” Journal of Aerosol
Science 21 (3):431-440. doi: 10.1016/0021-8502(90)90071-5.
Hinds, William C. 2012. Aerosol Technology: Properties, Behavior, and
Measurement of Airborne Particles: John Wiley & Sons.
Jiang, Lu, Nordsiek Hansen, and A. Shaw Raymond. 2010. “Clustering of
settling charged particles in turbulence: theory and experiments.” New
Journal of Physics 12 (12):123030.
70 Teerapol Srichana

Karner, Stefan, Eva Maria Littringer, and Nora Anne Urbanetz. 2014.
“Triboelectrics: The influence of particle surface roughness and shape on
charge acquisition during aerosolization and the DPI performance.”
Powder Technology 262:22-29. doi: 10.1016/j.powtec.2014.04.025.
Lee, Handol, Dae-Hyeon Jo, Won-Geun Kim, Se-Jin Yook, and Kang-Ho
Ahn. 2014. “Effect of an orifice on collection efficiency and wall loss of a
slit virtual impactor.” Aerosol Science and Technology 48 (2):121-127.
doi: 10.1080/02786826.2013.862333.
Marple, Virgil A., and Bernard A. Olson. 2011. “History of Virtual
Impactors.” In Aerosol Science and Technology: History and Reviews,
edited by David S. Ensor. Reston, VA: RTI Press.
Moshfegh, Abouzar, Mehrzad Shams, Goodarz Ahmadi, and Reza Ebrahimi.
2009. “A novel surface-slip correction for microparticles motion.”
Colloids and Surfaces A: Physicochemical and Engineering Aspects
345:112-120.
Moshfegh, Abouzar, Mehrzad Shams, Goodarz Ahmadi, and Reza Ebrahimi.
2010. “A new expression for spherical aerosol drag in slip flow regime.”
Journal of Aerosol Science 41:384-400.
Selvam, Parthiban, Steve Marek, C. Randall Truman, Doug McNair, and Hugh
D. C. Smyth. 2010. “Micronized drug adhesion and detachment from
surfaces: Effect of loading conditions.” Aerosol Science and Technology
45 (1):81-87. doi: 10.1080/02786826.2010.522628.
Son, Minsoo, Seungho Lim, Giwoon Sung, Taesung Kim, Yeonchul Ha,
Kibong Choi, and Weon Gyu Shin. 2015. “Development of a novel
aerosol impactor utilizing inward flow from a ring-shaped nozzle.”
Journal of Aerosol Science 85:1-9. doi: 10.1016/j.jaerosci.2015.02.004.
Srichana, Teerapol, Gary P. Martin, and Christopher M. Marriott. 2000. “A
human oral-throat cast integrated with a twin-stage impinger for
evaluation of dry powder inhalers.” Journal of Pharmacy and
Pharmacology 52 (7):771-778.
Srichana, Teerapol, Gary P. Martin, and Christopher Marriott. 1998.
“Calibration method for the Andersen cascade impactor.” Journal of
Aerosol Science 29 (SUPPL.2):S761-S762.
Suwandecha, Tan, Wibul Wongpoowarak, Kittinan Maliwan, and Teerapol
Srichana. 2014. “Effect of turbulent kinetic energy on dry powder inhaler
performance.” Powder Technology 267:381-391. doi: 10.1016/j.powtec.
2014.07.044.
 Forces and Interactions 71

Suwandecha, Tan, Wibul Wongpoowarak, and Teerapol Srichana. 2016.

“Computer-aided design of dry powder inhalers using computational fluid
dynamics to assess performance.” Pharmaceutical Development and
Technology 21 (1):54-60. doi: 10.3109/10837450.2014.965325.
Theerachaisupakij, Woraporn, Shuji Matsusaka, Y. Akashi, and Hiroaki
Masuda. 2002. “Reentrainment of deposited particles by drag and aerosol
collision.” Journal of Aerosol Science 34:261-274.
Wall, Stephen, Walter John, Hwa-Chi Wang, and Simon L. Goren. 1990.
“Measurements of kinetic energy loss for particles impacting surfaces.”
Aerosol Science and Technology 12 (4):926-946. doi: 10.1080/
02786829008959404.
 Chapter 4

Formulation Design and

Production Technology in Dry
Powder Inhalers

4.1. Introduction
Aerosol inhalation is a preferred route for drug delivery to the respiratory
tract since it confers many distinct advantages over delivery via other routes.
The medication is directly delivered to the desired region of the airway and
thus the inhalation of an antibiotic for the treatment of lung infection provides
an action without wastage through the systemic circulation. Aerosol inhalation
also allows for a rapid and predictable onset of action. For instance, after an
oral dose of salbutamol, it may take two to three hours before the peak plasma
concentration of the drug is achieved and the maximum therapeutic effect
obtained, while the corresponding time after administration of an inhaled dose
may be only 15 to 30 min (Srichana, Suedee, and Reanmongkol 2001). The
first-pass effect is avoided, and even though pulmonary tissues contain a
highly developed cytochrome P-450 system capable of inactivating certain
drugs, the concentration of cytochrome P-450 present in pulmonary tissues is
much less than in the liver. Lower doses can be administered via the
respiratory tract, in comparison to other routes. This minimizes any unwanted
side effects. For example, an inhaled dose of 100 µg salbutamol sulfate is
bioequivalent to 4 mg of the same drug taken orally. Thus, there is a potential
for considerable cost savings. Aerosol inhalation can be employed as an
74 Teerapol Srichana

alternative route to avoid drug interaction when two or more medications are
used concurrently. Conversely, this route does have some disadvantages such
as a low efficiency of drug delivery since a high percentage of the drug often
remains in the device or deposits in the upper airways. Patients must be
reasonably motivated, and often physically adept in manipulating and
employing the delivery devices to ensure maximal efficiency. Drug delivery
via the lungs requires a much higher degree of patient participation in the
delivery process compared to the much more passive role required in
swallowing a tablet or capsule. Co-ordination of drug delivery with the
appropriate stage of inspiration or ensuring a sufficient air flow to actuate
devices can be a problem for a number of patients. In addition, since
corticosteroids are one of the major classes of drugs routinely administered via
this route, associated side effects can include a degree of immune suppression
and more frequently, oral candidiasis.

Table 4.1. Advantages and disadvantages of dry powder inhalers.

Adapted from Ashurst, I. et al., Pharmaceutical
Science & Technology Today, 3(7), 246-256, 2000)

Advantages Disadvantages
 propellant-free  dependency on patient’s inspiratory flow
 less need for patient rate and profile
coordination  device resistance and other design issues
 less potential for  greater potential problem in dose
formulation problems uniformity
 less potential for  less protection from environmental effects
extracting from the and patient abuse
device component  more expensive than pMDI
 not available worldwide

DPIs are versatile delivery systems that may require some degree of
dexterity to operate. Its formulation may consist of the drug alone or of drug
blended with a carrier material which is usually lactose. The advantages and
disadvantages of DPIs are summarized in Table 4.1. DPIs in common use
today are breath actuated, and the energy for dry powder dispersion and
generation of the aerosol are derived from the patients inhalation. This
method alleviates the problem of coordination of actuation and inhalation that
many patients have with pMDIs, but several studies have shown that the
dosing performance of some DPIs, in terms of total dose or fine particle dose
emitted from the device, is dependent on the inspiratory flow rate. One of the
 Formulation Design and Production Technology in Dry Powder… 75

key factors involved in optimizing DPI’s performance is the precise particle

engineering required to produce a powder formulation that delivers an
accurate, consistent and efficient dose of the drug (Telko and Hickey 2005).
Dry powder formulations are activated and driven by the patients’
inspiratory flow. Furthermore, the generation of aerosol often requires a high
inspiratory flow rate to complete the emitted dose. DPIs may cause coughing.
These devices are not recommended for children under the age of five, people
with severe asthma or those suffering a severe attack. All DPIs have four basic
features: a dose metering mechanism, an aerosolization mechanism, a
deaggregation mechanism and an adapter to direct the aerosol into the
patient’s mouth.

4.2. Factors Affecting Dry Powder

Inhaler Performance
From the physiology of the pulmonary system, drug deposition will be
related to pulmonary physiology. In general, an appropriate particle size
should be in the range of 0.5-5 µm. While using a fine particle of the drug, it
will produce more surface free energy. Drug aerosol particles will attach
firmly to surfaces they contact and form agglomerates by the cohesive force.
This is called drug-drug aggregate formulation.
The drug-drug agglomerate formulations produce difficulties in generating
the release of aerosol particle. Filled with an appropriate size of coarser carrier
particle of about 35 µm the drug particles will form weak adhesive forces
(Figure 4.1) that result in an improved percentage of fine drug particle by
reducing the force needed to deagglomerate drug particle and improve a
uniformity of the dosage form (Broadhead, Rouan, and Rhodes 1995). This is
called the drug-carrier aggregates formulation. Sometimes fine particles of a
tertiary component (e.g., leucine, lecithin or magnesium stearate) are used to
adjust the interfacial properties of the carrier particles. They decrease the drug-
carrier adhesion force and lead to producing a greater percent of the fine
particle fraction (%FPF) of the drug (Begat et al. 2004).
76 Teerapol Srichana

Figure 4.1. Relationship between the %FPF (mean ± S.D., n = 5) and the carrier
median diameter. (Adapted from Louey, M. D. et al., International Journal of
Pharmaceutics, 252(1-2), 87-98, 2003).

The important forces in the formulation are the cohesive and adhesive
forces. To improve deposition to the deep lung, fine drug particles must be
generated by reducing the cohesive force and forming a weak adhesive force
(Figure 4.2). The small drug particles have a high surface cohesive force and
form self-aggregates that are hard to deagglomerate. The method to produce
weak adhesive forces between drug-carrier particles by adding adhering
micronized carrier particles on the surface of the coarse lactose particle
(Louey, Razia, and Stewart 2003). The carrier roughness can be reduced by
surface treatment of the carrier particle and/or filled with other excipients.

Figure 4.2. Methodology to reduce the cohesive force and forming a weak adhesive
force.
 Formulation Design and Production Technology in Dry Powder… 77

DPIs can be developed by proper device design and powder technology.

Two factors are important in powder technology; the cohesive forces and the
flowability. Powder dispersibility is controlled by the interparticle cohesive
forces which are proportional to the area of contact and separation distance
between the particles (Weers and Miller 2015). Strong interparticle forces lead
to a poor powder flow. The active drug substance needed in a DPI formulation
is a micronized drug. The smaller the particles, the stronger the cohesive
forces (Ahmad, Ungphaiboon, and Srichana (2015 and drug aggregates cause
poor flowability. To improve flowability, there are two ways to prepare the
formulations; with or without a carrier.

4.3. Formulations without Carrier

Only micronized drug particles form loose aggregates in the formulation.
These loosely aggregated drug particles (spheroids) can be prepared by the
controlled agglomeration of micronized particles. Spheroids have large
particle sizes around 0.5 mm in diameter and thus have appropriate flow
properties. Of course, spheroids exhibit better flow properties than micronized
material. They carry low static charge during handling and operation (Healy et
al. 2014). When spheroids are loaded into the DPI, they break up into primary
particles during inhalation. It has been reported that the major disadvantage of
such systems containing spheroids is the high variability in the emitted dose.
Another technology to prepare spheres is to coat the particles with leucine
(Pilcer and Amighi 2010, Raula, Lähde, and Kauppinen 2009).
Figure 4.3 shows the primary particles forming aggregates and
agglomerates. The size of the primary particles increases as they form
aggregates and when the aggregates bind together they form agglomerates.

4.4. Formulations with a Carrier

The majority of marketed drugs use carriers to improve flowability and
reproducibility of dosing. Commonly used carriers are lactose and glucose.
Other carriers that have been investigated are mannitol, sorbitol, maltitol,
xylitol and arabinose (Steckel and Brandes 2004, Rahimpour, Kouhsoltani,
and Hamishehkar 2014). Coarse carriers can increase the flowability which
leads to an increase in the quantity of the drug reaching the lower airways. The
78 Teerapol Srichana

dispersion properties depend on the performance of the drug aggregates on the

carrier surface (Louey, Razia, and Stewart 2003). Aggregation not only
requires much more appropriate forces for uniform drug dispersion but also an
appropriate strength of the inspiration force to deaggregate the drug from the
carrier. The drug aggregation patterns are of 4 types: drug aggregate on the
carrier, fine drug particles, aggregate of fine drug particles and the aggregate
of fine drug particle and carriers (Srichana, Martin, and Marriott 2000).

Figure 4.3. Features of primary particles, aggregates, and agglomerates.

Figure 4.4. Interaction between particles drug-drug aggregates (a), drug-carrier

aggregates (b), drug-carrier aggregates with “tertiary component” that are shown in
small dark particles (c).
 Formulation Design and Production Technology in Dry Powder… 79

Figure 4.5. Dispersion and deaggregation of the drug from the carrier by turbulent
flow. Orange circle represents the carrier particles, green circle represents drug
particles.

When the inspiratory flow is applied, the turbulent flow of the inspiration
causes drug dispersion and deaggregation of the drug from the carrier (Figure
4.5). The drug will travel to the lower respiratory tract and alveoli while
carriers which have a larger size are deposited on the oropharyngeal regions.
Drug deposition in the lungs ranges from 15-40% of the emitted dose with
current DPI devices (Anderson 2001).
In the development of the dry powder formulations, the following factors
should be considered:

4.4.1. Particle Size of the Drug

Control of particle size may be used as a tool to target aerosol to the site of
action. Particles larger than 10 µm will deposit on the oropharynx while
particles smaller than 0.5 µm can be exhaled. Particle sizes larger than 5 µm
may deposit on the bronchiolar regions, whereas the 1-3 µm particles can
reach the alveoli (Gupta and Hickey 1991, Zanen, Go, and Lammers 1994).
Therefore, the desirable size of an inhaled particle should be less than 5 µm
and a suitable size for bronchodilator drug must be in the range of 0.5-5 µm.
Most drug particles that have decreased sizes due to different processing
technologies will be explained later in this chapter. Also, the drug particles can
be categorized at least into organic and inorganic compounds that behave
differently in terms of their size. Besides the chemical constituents of the
molecule, the crystalline and amorphous states are also the important factors to
80 Teerapol Srichana

be considered. From our experience working with salbutamol sulfate,

budesonide, amphotericin B, rifampicin, isoniazid, pyrazinamide, ethambutol
hydrochloride and levofloxacin, all of these molecules are very different in
their class solubility that affects their physicochemical properties. In all cases,
the size has to be carefully controlled.

4.4.2. Particle Size of the Carrier

A carrier is employed to improve the flowability of the fine drug particle.

Therefore, it must be easy to handle and reproduce in the filling process. The
incorporation of the carrier may cause irritation, coughing and
bronchoconstriction (Timsina et al. 1994). Some studies have shown the
effects of the particle size of a carrier on the efficiency of the formulation
delivery. When the carrier size was decreased, the respirable fraction increased
(Rahimpour and Hamishehkar 2012). However, the exploitation of this method
was limited because of the aggregation of fine carrier particles (Ganderton
1999). The most suitable carrier size was approximately 30-90 µm.
When defining the particle size of the aerosols, it is also important to take
into account the density, shape, surface roughness and geometric diameter,
since all of these parameters affect the aerodynamic diameter of the inhaled
airstream. In view of this, the MMAD can be useful in describing the diameter
of the particles that differ in physical characteristics. Similar to the drug, the
carriers can be grouped into different classes from the very water soluble
sugar, cyclic sugar, phospholipids, amino acids, bile salts to polymer
molecules (chitosan and PLGA) as detailed in Chapter 5. The carrier particle is
also different and difficult to control and requires different comminution
technology to obtain the appropriate size. All carriers behave differently in
terms of their moisture sorption, crystallinity, electrostatic charges and all will
influence the processing condition, handling, and storage.
Particle size is the predominant factor that governs the physical properties
of the powder. The determination and control of the particle size are often an
essential requirement in the formulation of pharmaceuticals. This is
particularly true in the formulation of dry powders for inhalation, since the
particle size of both the carrier and drug particles determines the powder
performance at every stage that affects the flowability, fluidization, separation
and deposition of powders. The importance of particle size in the development
of solid pharmaceutical dosage forms has long been appreciated and
 Formulation Design and Production Technology in Dry Powder… 81

techniques for its measurement are well established. However, the particle size
obtained is dependent upon the principle of measurement (see Chapter 1).

4.4.3. Particle Shape

Particle shape is one of the most intractable and uncontrollable factors in

powder technology. Different generation methods for the same material will
almost always result in the production of particles of different particle shape.
The packing patterns of anisometric particles, such as those with a high
elongation or flatness ratio, are largely dependent on the processing conditions
that the powder has experienced. Powder handling, such as mixing,
compression, and vibration, may introduce enough energy to the powder such
that its component particles tend to orientate themselves to the most stable
conformations.
Particles having a more irregular shape exhibit much higher porosities in
powder beds than spherical particles, both before and after tapping.
Theoretically, irregular particles should exhibit higher adhesion forces than
more spherical particles of a similar mass under the same conditions, if the
microscopic factors are negligible and the particles seek the most stable
orientation on the substrate surface. Particle shape may affect the flow
properties of a powder by changing interparticulate and/or frictional forces
between the particles. In general, if the particle shape can reduce both
interparticulate and frictional forces, then it is said to be favorable for a good
powder flow. The effect of the particle shape on the powder flow is more
complicated than the particle size. Materials of different mechanical properties
would need to possess a unique particle shape for optimized processing and
this may change in different processing conditions.

4.4.4. Drug and Carrier Surface

Generally, dry powder aerosols are made by mixing the micronized drug
and large carrier particle (Sawatdee 2005). The mechanical stability of these
ordered powder mixtures is highly influenced by the surface properties of the
carrier (Lai and Hersey 1987, Staniforth et al. 1982). Only limited areas show
a high binding affinity. Redispersion of the drug from the carrier surface
during inhalation is the most critical factor. Control of the crystallization
conditions to produce a smooth carrier surface has been reported to restrict
82 Teerapol Srichana

adhesion and enhance separation (Ganderton 1999, Zeng et al. 2000, Park,
Park, and Choi 2014).
A carrier surface has a surface porosity. Pore size on the carrier surface
can be characterized into 3 size groups that are nanopores (0.007-1.00 µm),
micropores (1.00-8.06 µm) and macropores (8.06-150 µm) (Ahuja and Pathak
2009, Fan, Zhang, and Wang 2013). Nanopores reduce the effective carrier
contact area and increase the distance between drug and carrier particles.
Nanopores thus reduce the magnitude of interparticulate adhesion forces in an
inhalation mixture. Microporosity carriers do not play important roles during
the aerosolization process. They contribute to the mixing process by increasing
the effectiveness of the cohesive drug agglomerates, de-agglomeration and
distribution of drug particles over the carrier surface. On the other hand,
macroporous carriers apparently have a negative influence on the performance.
When the grooves are introduced on carrier they will prevent drug particles
from the drag and lift from the carrier surface (Shalash, Molokhia, and Elsayed
2015).
Carrier surface coating is an effective technique to improve the carrier
surface characteristics. Lactose carrier particles that were coated with a thin
layer of magnesium stearate by physical mixing enhanced the in vitro
inhalation performance of the formulation when compared to the uncoated
lactose formulations (Iida et al. 2004). Coating leucine or magnesium stearate
on drug particles is also reported (Raula et al. 2012, Zhou et al. 2013). The
particle morphology and surface roughness can be modified by spraying with
bovine serum albumin to create either wrinkled or corrugated particles (Adi et
al. 2008).

4.4.5 Drug Carrier Ratio

The respirable fraction increased when the drug carrier ratio increased
(Kassem 1990). This was due to the increase in the active binding sites as a
higher drug concentration was able to bind to the lower affinity sites or be
free. Recent studies had revealed that when the drug to carrier surface
coverage was more than a monolayer, the aerosol performance was decreased
(Young et al. 2011, Grasmeijer et al. 2013). Although there are wide ranges of
drug contents in various formulations, it has been a common practice to use a
fixed drug to carrier ratio of 1:67.5 in the studies. This ratio was most likely
adopted from earlier carrier based formulations for the Rotahaler®, Diskhaler®
and Cyclohaler®. It is well known that drug content affects the homogeneity
 Formulation Design and Production Technology in Dry Powder… 83

blend and the drug detachment from the carrier. Adhesive mixtures are a
dynamic process of ordering and randomization that determines the outcome
of the blending process. With increased drug content, the equilibrium is
displaced towards randomization which results in a less homogeneous mixture
that is more prone to segregation during handling. This adverse effect may
produce a better drug detachment from the carrier particles when the
separation forces are generated. Several mechanisms have been postulated to
explain the effect of the drug content on the dispersion performance of
adhesive mixtures for inhalation. First, the so-called active sites on the lactose
carrier surface may become saturated with increased drug content. This results
in a decrease in the mass of the drug that adheres strongly to the carrier surface
and improves the detachment of the drug particle. Second, the formation of a
layer or an agglomeration of the drug particles with an increase in the drug
content can cause detachment of the large agglomerates rather than the single
drug particles, if a failure in adhesive bond formation between the inner drug
layer and the carrier surface occurs during inhalation. This too may improve
drug detachment, because it increases the magnitude of the lift and inertial
separation forces to become more than that of the interaction forces of the drug
carrier. Third, a higher drug content increases, drug particles will fill up the
clefts and depressions in the carrier surface. By being transferred through the
powder bed, this enables to act effectively as press-on forces on the drug
particles that would have found shelter from them in carrier surface
irregularities at a much lower drug content. It causes the particle interaction
forces to increase by reducing their separation distance and increasing their
contact surface area, and therefore, it causes drug detachment to be negatively
affected. Lastly, it was hypothesized that with increased drug content,
detached particles might be more likely to collide with neighboring drug
particles to cause an enhanced drug detachment which is a mechanism referred
to as the ‘collision effect’ (Grasmeijer et al. 2013). Collision effects may
marginally contribute to improve drug detachment with an increase in the drug
content.

4.4.6 The Use of Tertiary Component and Mixing Sequence

The addition of the tertiary components affects the deaggregation and

deposition of aerosols. The addition of magnesium stearate reduced the
adhesion between drug and carrier, leading to the loss of homogeneity and
destabilization (Staniforth et al. 1982, Lai and Hersey 1987, Kassem 1990,
84 Teerapol Srichana

Zeng et al. 1999). When the magnesium stearate concentration was increased,
the adhesion between the drug and carrier decreased and resulted in an
increase in the respirable fraction. Furthermore, different mixing sequences
were shown to result in different deposition profiles (Zeng et al. 1999). High-
energy active sites may exist on the coarse carriers for the drug to
preferentially adhere due to stronger interactions (Ahmad, Ungphaiboon, and
Srichana 2015). When fine carriers of a tertiary component are added, they
will compete with the drugs for the active sites, making the drug easier to
detach from the carrier during inhalation. Based on this observation, the active
sites can be saturated. Thus the sequence of mixing the drug, coarse carrier
and tertiary component is expected to be critical. Furthermore, high-energy
sites on the carrier surface were still observed after the addition of the tertiary
component (Chan 2006). A proposed mechanism that may involve the
formation of weak conglomerates between the drug, carrier and tertiary
component was developed. In this case, the drug distribution between the
coarse carrier and tertiary component may cause a disruption of attractive
interactions in the conglomerate.

4.5. Electrostatic Charge

Electrostatic charges also play important roles in dry powder aerosols
(Finlay 2001). Although human studies are limited, an in silico study has
shown that for aerosols with 200 charges per particle, electrostatic forces
become predominant (Balachandran et al. 1997). Byron et al. (1997) found
that a fine particle dose charge that ranged from -400 pC to 200 pC for
Bricanyl® and Pulmicort®. Kwok et al. (2011) also reported the effect of the
relative humidity on the electrostatic charge of DPI using an electrical low-
pressure impactor. Both Bricanyl® and Pulmicort® displayed a different charge
distribution and a bipolar charge across 0.388 μm to 6.06 μm. The effect of the
relative humidity on charging the dry powder aerosol appeared to be
dependent on the drug. A comprehensive review on the role of electrostatic
charge in pharmaceutical aerosol was performed by Wong, Chan, and Kwok
(2013) and Kaialy (2016).
 Formulation Design and Production Technology in Dry Powder… 85

4.6. Particle Production Technology

Comminution technology has been introduced into aerosol science to
prepare powder particles. Milling, high-pressure homogenization, spray
drying, spray freeze- drying, crystallization, supercritical fluid and
adsorption/coacervation technology have been used in aerosol science.
However, the particle size is not well established during these processes. The
size of the drug particles must be reduced to the respirable size range in a
separate unit operation. There are many options and it may be necessary to
find the one that works best for any specific drug. In the dry powder aerosols
production, several techniques are employed to decrease the particle to an
appropriate size (Table 4.2).

Table 4.2. Techniques commonly used in particle

production for dry powder inhalers

Technique Comments References

Jet milling/Wet Commercially established for Irngartinger et al. (2004)
milling numerous inhalation drugs. Telko and Hickey (2005)
No subsequent separation Rasenack (2010)
process required. Nakpheng et al. (2011)
Introducing an amorphous Jetmalani et al. (2012)
content.
Generally, difficult to scale up.
Concerns with respect to
foreign particulate material
from the erosion of the milling
media.
High-pressure Commercially used for several Rabinow (2004)
Homogenization inhalation drugs. Chougule, Padhi, and Misra
Nanosuspension particle. (2007)
Liposome formulation. Xu, Mansour, and Hickey
(2011)
Spray drying Commercially proved White et al. (2005)
Single-step particle formation Chen et al. (2005)
process. White et al. (2005)
Control over size, morphology, Chow et al. (2007)
density and surface Vehring (2008)
composition.
Process development required
to minimize process-induced
degradation.
86 Teerapol Srichana

Technique Comments References

Spray freeze- Two-step process (freezing and Leuenberger (2002)
drying lyophilization). Rogers, Johnston, and
Very low-density, fragile Williams (2003)
particles. Wang et al. (2012)
Energy-intensive, complex Ali and Lamprecht (2014)
process.
Freezing stress, but no thermal
degradation.
Controlled More stable product. Beach et al. (1999)
crystallization Control physical form of active Havelund (2001)
ingredients, polymorphic Rasenack, Steckel, and
forms formation. Muller (2003)

Particle engineering of inhaled Rasenack and Muller (2004)

drugs. Steckel, Rasenack, and
Muller (2003)
Ikegami et al. (2000),
(2002)
Voss and Finlay (2002)
Supercritical Limited drugs soluble in the Steckel, Thies, and Müller
fluid RESS supercritical fluid. (1997)
Drug must be poorly soluble in Steckel and Muller (1998)
Supercritical the supercritical fluid. Velaga, Berger, and
fluid SAS Carlfors (2002)
Shekunov et al. (2003)
Rehman et al. (2004)
Adsorption/ Complex/several critical Alvim and Grosso (2010)
coacervation parameters. Yadidi (2016)

4.6.1. Comminution by Jet Milling/Wet Mill

The common comminution technique is milling. There are three types of

mills that are able to reduce the particle size to 2–5 m. These are jet mill; pin-
mill and ball mill.
Mechanical processing, such as milling has been shown to affect the
crystallinity of the material; this effect must be considered. Jet milling (or air
attrition milling) is the most useful technique; it reduces the particle size via
high-velocity particle–particle collisions. Unmilled particles are introduced
into the milling chamber. High-pressure nitrogen is fed through nozzles and
accelerates the solid particles to sonic velocities. The particles collide and
fracture. While flying around the mill, larger particles are subjected to a higher
 Formulation Design and Production Technology in Dry Powder… 87

centrifugal force and are forced to the outer perimeter of the chamber (Figure
4.6). Small particles exit the mill through the central discharge stream.
Depending on the nitrogen pressure and powder feed rate, particles down to 1
m in diameter can be produced.

Figure 4.6. Air jet mill (Adapted from www.umamicron.com).

A pin mill uses a mechanical impact to grind material, both by particle–

particle and particle-solid collisions. A pin mill is equipped with a series of
concentrically mounted pins located on a spinning rotor and a stationary stator
plate. The powder is fed to the milling chamber and transported through the
milling chamber by centrifugal force. The milled product is collected from the
bottom. The pin mill can produce 1 m particles, but not as small as the jet
mill. On the other hand, the power consumption of a pin mill is lower than that
of the jet mill.
The ball mill is essentially a rotating cylinder loaded with the drug and
“milling media” (i.e., balls that grind the drug between each other as they
tumble inside the mill). The size and material of the milling media can be
varied. Ball milling is very slow and the process is poorly scalable, which is
why tumbling ball mills are used only in the laboratory.
88 Teerapol Srichana

DPI inhalation products contain the micronized drug in either an

agglomerate or blend. Such particles are normally produced by crystallization,
followed by filtering, drying and micronization (Jetmalani et al. 2012). The
particle size can be reduced by attrition, impaction or shear force. Air jet-
milling is well-established to manufacture dry powders for inhalation.
Although milling can be performed on a dry or wet basis, dry grinding is more
commonly employed as it is less labor-intensive. In the jet milling process, the
starting material undergoes many impact events until a quantity of the required
particle size fraction is achieved and separated from the larger particles by
inertial impaction. This will ensure that the particle size required for
respiratory delivery is eventually obtained. The obtained particle shape is
either tabular or rounded. However, this milling process can be time-
consuming and inefficient for organic pharmaceuticals and can adversely alter
the surface and solid-state properties of the materials.
Micronization also generates electrostatic charges and amorphous domains
on the particle surface. This will change the ground material to have both
cohesive and adhesive properties. In some cases, the whole material bulk may
become amorphous. As the amorphous domains are thermodynamically
unstable, such amorphous products will recrystallize leading to crystal growth
on the milled particle surface and formation of solid bridges between the
particles. The material is also prone to chemical decomposition and water
sorption. All of these physical and chemical changes are highly undesirable
and can adversely affect the in vitro performance of the DPI formulations. The
operating condition must be clearly defined such as temperatures and relative
humidities. Since micronization may reduce drug stability, modified
techniques are necessary to solve the problems of thermolabile
biopharmaceuticals (Nakpheng et al. 2011). For example, micronization of the
decapeptide cetrorelix can be done by suspending the drug in a fluid propellant
coupled with a cryostat down to -70 ºC, followed by evaporation of the fluid
propellant to recover the micronized material (Irngartinger et al. 2004). This
milling technique was effective and mild for the peptide, and perhaps better
than spray drying because of the higher respirable fraction of the milled
material.

4.6.2. High-Pressure Homogenization

High-pressure homogenization offers more advantages over traditional

micronization. This technique can produce nanoparticles in sizes between 100
 Formulation Design and Production Technology in Dry Powder… 89

and 700 nm which can afford higher bioavailability and likely to obtain greater
efficient drug delivery and rapid dissolution in the lung. The homogeneity of
nanosuspensions is also superior over the microparticulate suspensions and
leads to a higher delivered dose with better dose uniformity (Figure 4.7)
(Rabinow 2004).

Figure 4.7. High-pressure homogenization (Adapted from https://en.wikipedia.org).

Typically, 20-30% w/w of the surfactant is required to stabilize the

nanosuspensions. This amount is directly proportional to the specific surface
area. The potential problems of this technique are from potential
contamination with the grinding media and adverse effects of the high
temperature on the chemical stability. These problems may be minimized or
overcome by utilizing direct nanoparticle precipitation.

4.6.3. Spray Drying

Spray drying (SD) technology is very popular in the processing of food,

biochemicals, and pharmaceutical materials. The technology is easy to operate,
available to large scale-up, and able to produce composite materials. This
technique can produce particle sizes in the range of 1-5 m. The physical
instability and thermal degradation of the products is a major concern of this
90 Teerapol Srichana

method. This approach is rapidly expanding in a range of applications as the

technology becomes more advanced. A typical SD process consists of four
steps (Chow et al. 2007): (a) atomization of the feed solution; (b) sprayed
droplets contact with hot gas; (c) drying of sprayed droplets and (d) separation
of dried product (Figure 4.8). For each operating step, a variety of process
designs varies depending on specific applications. The atomization can be
generated from rotary atomizers, pressure nozzle or two fluid nozzles while
air/fluid flow inside; the drying chamber can be co-current, counter-current or
a mixed flow type. More recently, four-fluid nozzles with in-line mixing have
been developed for the production of composite particles. The SD process can
also be operated in different modes such as open cycle and closed cycle, semi-
closed cycle with or without aseptic controls. The system can be further
modified on a larger production scale and for better product recovery. For
example, the bag-filter is replaced with a cyclone system or the product is
dried at a lower inlet air temperature for thermolabile materials. The design of
a high-efficiency cyclone separation system is also essential for industrial
scale production.
The spray drying technique can be optimized to obtain suitable conditions
for a particular product. Both product concentration and atomization rate can
be manipulated to produce particles with different surface corrugations. In
some cases, proper humidity control of the drying gas can afford particles with
the desirable densities or aerodynamic diameters for pulmonary delivery
(Chen et al. 2005).

Figure 4.8. Spray drying (Adapted from www.eurotherm.com).

 Formulation Design and Production Technology in Dry Powder… 91

4.6.4. Spray Freeze-Drying

A typical spray freeze-drying (SFD) technique involves the atomization of

an aqueous solution of a drug via an ultrasonic nozzle into a spray chamber
filled with cryogenic liquid nitrogen. The spraying process can be performed
above the surface of the cryogenic liquid or in the liquid depending on the
position of the nozzle (Figure 4.9). Since the cryogenic liquid will decrease
due to evaporation, continuous addition of fresh cryogenic liquid is required,
especially when a longer atomization process or a large spray volume is used.
The droplets solidify rapidly upon contact with the cryogenic liquid because of
the high heat-transfer rate. Stirring of the cryogenic liquid may be required to
prevent the possible aggregation of frozen particles. Once the spraying process
is completed, the whole content can be freeze dried as with conventional
freeze-drying. The sublimation of the spray solvent has also been developed
(Leuenberger 2002, Rogers, Johnston, and Williams 2003). This process
involves drying of the frozen particles by a stream of cold dry air inside an
insulated stainless steel vessel. Changes of the moisture content in the product
can be monitored, without interrupting the drying process.

Figure 4.9. Spray freeze-drying (Adapted from http://powderpro.se).

Spray freeze-dried particles can be engineered to the desired particle size

below 5 m. The most important operating parameter is the mass ratio of the
liquid feed to the atomized nitrogen. A decrease in the particle size can be
92 Teerapol Srichana

achieved by a decrease in the mass ratio, while the addition of excipients may
lead to an increase in the particle size. Further, the spray-freezing process has
been modified; the drug solution is atomized and frozen simultaneously by
mixing with a liquefied gas or supercritical fluid, such as supercritical CO2.
The liquid droplets are first dispersed with a static mixer within the
supercritical CO2 and then frozen. After spray freezing, the frozen solvent is
removed, as in the case of freezing with cryogenic liquids, by vacuum or
atmospheric freeze-drying. The large surface area and loose porous structure
of the powder allow relatively fast drying compared to a standard
lyophilization process.

4.6.5. Controlled Crystallization

Crystallization of hydrophobic drugs can be obtained by antisolvent

precipitation (Rasenack, Steckel, and Muller 2003, Rasenack and Muller
2004). The precipitated crystals (e.g., budesonide, prednisolone, fluticasone
and disodium cromoglycate) have been shown to exhibit a higher FPF than the
jet-milled samples (Rasenack, Steckel, and Muller 2003, Steckel, Rasenack,
and Muller 2003). Also, higher drug concentrations yield smaller particles.
The amorphous content of such drugs is lower than that of the micronized
samples, thus better physical stability can be obtained. Zinc-free insulin
crystals have been prepared in a size range of 0.2-5 m. The precipitated
insulin crystals were more stable than those powders prepared by spray-
drying, freeze-drying or oven drying (Havelund 2001).
Direct crystallization of spherical agglomerates has been utilized in
pulmonary delivery formulations. This technique involves antisolvent
precipitation of a drug solution in an organic solvent, followed by addition of
water or a partially miscible organic solvent with water. For example, the
introduction of ethyl acetate into the water/acetone crystallization medium
resulted in the formation of spherical agglomerates (200-300 m). Primary
crystals were in the respirable range (d50 = 1.3-2.7 m) (Ikegami et al. 2002,
Ikegami et al. 2000). The agglomerated crystals deaggregate into primary
particles upon mixing with lactose carrier. The adhered primary crystals were
easily detached from the lactose during inhalation (Voss and Finlay 2002).
Spherical crystallization can be achieved for certain drugs by quenching of a
hot organic or aqueous drug solution with a cold organic or aqueous organic
solvent. The quench solvent should be miscible with the drug solution. For
instance, spherical microcrystals of salmeterol xinafoate, a long-acting anti-
 Formulation Design and Production Technology in Dry Powder… 93

asthmatic agent, can be readily produced by adding a hot propanol containing

the drug to a chilled quench solvent (Beach et al. 1999). The agglomerates are
free-flowing and readily micronizable to a suitable inhalable material.
The particle size control in all crystallization is the most challenging task
because most molecules tend to form large crystals. The nucleation and growth
mechanisms yield particles in the size range of 10-100 m. Mixing between
the drug solution and non-solvent can be fast agitation, high-velocity mixing
or ultrasonic mixing. In the ultrasonic crystallization, the particle size can be
controlled by the sonic-induced mixing and cavitation on supersaturation and
nucleation. This mechanism renders more uniform particles. The mean particle
size decreases if high concentrations of growth-retarding excipients are
employed.
These growth retardants are compound-specific in terms of their
interaction with the crystal surfaces. Furthermore, the use of additives
designed to inhibit crystal growth is very strict due to purity control and
toxicity issues. Finally, crystalline particles with a narrow size distribution are
feasible with controlled crystallization. A major drawback of such a process is
the need to remove all additives completely. The latter processing steps are not
straightforward. This step may result in powder caking, impure samples and
low dispersible powder.

4.6.6. Supercritical Fluid

A supercritical fluid (SCF) is a compressible liquid having gas transport

properties. SCF exhibits pressure-tunable solubility, which is suitable for
recrystallization operations. The most common SCF uses liquid carbon dioxide
(CO2) as a solvent. The SCF technique can be categorized into rapid expansion
of supercritical solutions (RESS) and antisolvents (SAS). Schiavone et al.
(2004) noted that a product prepared by SCFs yielded smoother budesonide
particles, with less surface area than a milled drug. Particle engineering with
SCFs is the subject of intensive research in the pharmaceutical industry;
excellent reviews on this topic have been published elsewhere (Telko and
Hickey 2005, Tabernero, Martín del Valle, and Galán 2012).
SCFs possess several advantages as solvents (or antisolvents) for
pharmaceutical manufacturing. Supercritical CO2 (SC-CO2) is the most
common SCF because of its low critical temperature (31.1ºC), moderate
pressure (73.8 bars), non-toxic inert nature and low cost. Despite the fact that
SCF technology is often associated with SAS, several existing technologies
94 Teerapol Srichana

differ in the applicability of the material. One of the most important physical
attributes of the SC-CO2 processing is the efficient extraction process. This
often enables the production in a dry form or as an aqueous suspension. It
promotes a clean and recycled precipitation process at low temperatures. The
SCFs can be utilized for plasticization of polymers. SCFs have been used for
the direct production of pure and composite particles with the advantages of
selective precipitation and control of the crystalline forms. In SC-CO2
applications, such as particle engineering of high potency and sensitive drugs,
SC-CO2 can reduce manufacturing complexity, energy, and solvent
requirements.

Figure 4.10. Supercritical fluid antisolvent systems

(Adapted from http://eng.ege.edu.tr).

In all SAS processes, the mechanism is based on the rapid precipitation

when a drug solution comes in contact with a SC-CO2 (Figure 4.10). It is
important to note that the mechanism of SAS changes under different
pressures and temperature conditions. It depends on the solvent composition,
known as the mixtures critical point. In the higher pressure, lower temperature
phase region, the solvents are completely miscible, and the SAS proceeds as a
typical precipitation process. In the lower pressure, higher temperature region,
however, the SAS shifts towards a spraying extraction. Different SAS
modifications are distinguished by mixing between the solution and SC-CO2
feeds. For example, solution enhanced dispersion with SCF utilizes high-
velocity mixing in a multi-component nozzle. Most of the drugs have very
limited solubility in CO2, SAS processes have attracted much attention in
recent years because they offer a method for production of micronized dry
 Formulation Design and Production Technology in Dry Powder… 95

powders. Most of the studies to date have focused on small molecules for
inhalation.
For example, production of steroids has been demonstrated including
budesonide and fluticasone (Steckel and Muller 1998, Steckel, Thies, and
Müller 1997). Coated particles produced with SAS showed a significant
increase in FPF when compared to the jet-milled products. It has been shown
that SAS-produced powders exhibit non-spherical particles with a lower bulk
density than micronized materials. Although the SC-CO2 process produces
particles with larger aerodynamic diameters than the micronized materials,
they have a significantly higher FPF. An analysis of hydrocortisone particles
using a multistage liquid impinger showed that the delivered dose increased by
30-40% for the SAS processed materials relative to the micronized particles
(Velaga, Berger, and Carlfors 2002). The improved properties of the SAS-
processed particles can be explained by the weak surface adhesion of these
particles to the inhaler reservoir. The size reduction process in SAS needs to
be optimized to obtain particles in the respirable range with the required shape.
For such optimization, nozzles may suffer from the inefficient macro-mixing
and periodic blockage, both of which can result in a broadening of the PSD.
To address some of these problems, an SCF technology utilizing a turbulent-
shear mixing system has been developed (Shekunov et al. 2003).
SAS processes possess a very important advantage to control the physical
form of the drug. Controlled production of different polymorphic forms has
been demonstrated by varying the working conditions. The produced materials
are usually crystalline and have a low residual solvent content. Engineering of
the hydrates, solvates and amorphous solids by SAS processing has also been
reported (Steckel, Thies, and Müller 1997, Steckel and Muller 1998, Velaga,
Berger, and Carlfors 2002, Rehman et al. 2004, Shekunov et al. 2003).

4.6.7. Adsorption/Coacervation Particle Formation

Adsorption was employed to produce sodium cromoglycate adsorbed with

fatty acids (Fults, Miller, and Hickey 1997). The coated lauric and stearic
acids led to an increase in the FPFs. The lauric acid appeared to alter the
deposition by changing the particle to a more elongated shape compared to the
untreated sample. Whereas, the stearic acid had an altered particle shape to a
smaller degree but had better FPFs due to the reduced interparticulate
interactions.
96 Teerapol Srichana

Coacervation is a phase separation phenomenon that occurs in a solution

of two liquid phases containing a solute species with polymer-rich and
solvent-rich phases. The process is reported to be scalable, although it requires
a multi-step particle separation: (A) a homogeneous polymer solution, (B)
phase separation and coacervation droplets, (C) deposition of coating and (D)
membrane formation. The physical chemistry of any coacervation process is
complex and has several critical parameters. It usually involves large
molecular weight polymers. Coacervation can be controlled by temperature
change, solution composition and pH. A more complex coacervation may
involve deposition of a shell material around the encapsulated core drug
(Figure 4.11). Coacervation followed by hardening of the dispersed phase
generates solid particles. This technique has been employed in DPI to produce
insulin and α-1-antitrypsin. The technology used a temperature controlled
coacervation in the presence of polyethylene glycol (PEG). Subsequent
dialysis was carried out against an aqueous solution followed by centrifugation
to remove the PEG and then lyophilization. Both drugs were solid spheres
with MMADs of 2.7-2.9 m. This process gained advantages from the
utilization of an aqueous system that was suitable for the production of protein
spheres. Further investigation into its applicability to small molecules and
protein formulations is certainly worthwhile and will be challenging.

Figure 4.11. Coacervation process.

4.7. Processing Conditions

The mixing time for each drug and carrier should be optimized because an
extension of the mixing time increased the drug particle association to the
active surface of the carrier, and consequently, increased the stability and
 Formulation Design and Production Technology in Dry Powder… 97

decreased the dispersibility (Ganderton 1999). The processes involved in a

powder formulation have been extensively reviewed in the pharmaceutical
technology/engineering area. After the drug and excipient(s) have individually
been brought to their desired forms, they are combined in the blending process
(Saleem, Smyth, and Telko 2008). The importance of the blending process can
be easily overlooked. However, it is a critical step for the manufacture of a
DPI product and is in fact subjected to substantial optimization work during
development.
When mixing powders with different properties, particle sizes and ratios,
as is the case with DPI formulations, inadequate mixing can cause poor dose
uniformity. In many cases, inadequate mixing cannot be overcome simply by
increasing the mixing time. Mixer selection, rotation speed, capacity and fill
level are all subjected to optimization as they can all affect the blend
homogeneity. Blending conditions also affect the interparticle forces which are
a primary determinant by the FPF. Different powders may have different
mixing requirements depending on the forces present between the various
particles. For low concentration (drug-carrier ratio) blends, geometric dilutions
are necessary preblending steps. The flow properties of the components of the
powder blend will play an important role in the efficiency of blending and
ultimately in an aerosol dispersion. Powder flow properties have been studied
for some time and methods have been adopted for their characterization
including bulk and tapped density, and the angle of repose.
Powder sampling is an important prerequisite for accurate
characterization. Blending validation is an important activity required by good
manufacturing practices in the United States Code of Federal Regulations.
However, taking blend samples at different times to determine the uniformity
of the blend is associated with several difficulties. New techniques are
emerging that can determine the blend homogeneity without removal of
samples from the mixer; techniques such as near-infrared and Fourier
transform-infrared analysis can determine blend uniformity by the
nondestructive acquisition of infrared spectra. After the formulation has been
blended, it is filled into capsules, multi-dose blisters or reservoirs for use with
the inhaler device. The filling process is automated and depends on the nature
of the metering system.
98 Teerapol Srichana

4.8. Storage Conditions

Adhesion and dispersion of aerosols determine the efficacy of the
formulation. Several parameters are involved in adhesion and dispersion such
as rugosity, shape, surface, charge and moisture. Moisture content is the main
factor that influences the cohesiveness of powder. An increase in cohesiveness
with the rising moisture content is due to the adsorbed water. Hygroscopic
aerosol particles will grow in size until the vapor pressure exerted by the
droplet equals the vapor pressure of the water in the lung. The aerosols
increase in size at high humidity as compared to low humidity (Hiller (1991.
Hygroscopicity is the intrinsic tendency of material to take on moisture from
its surroundings. Hygroscopicity is affected by the crystallinity of the material
and the morphology of the particles. Hygroscopic drugs present a greater risk
of physical and chemical instability. Moisture uptake and loss due to changes
in relative humidity can result in local dissolution and recrystallization and
lead to irreversible aggregation through solid bridge formation which can
adversely affect the aerosol generation and lung deposition. Hygroscopicity
can also alter the adhesive and cohesive properties or, in more extreme
situations, substantially increase the particle size. Hygroscopic growth
involves the uptake of moisture which will reach equilibrium in the droplets as
a function of the water activity of the solutions formed and the surrounding
atmosphere of water vapor. Hygroscopic growth has implications for the
equilibrium moisture content of the particles in the dosage form prior to
generation of the aerosol; it can cause chemical or physical instability of the
product. For aerosols, the physical instability is more important, because
agglomeration may be irreversible and lead to an inability to generate aerosol
particles of respirable size. As aerosol particles enter the lungs, they
experience an environment of high humidity (99.5% relative humidity at
37°C). Although they may not reach equilibrium during transit, susceptible
aerosol particles may be subjected to hygroscopic growth that increases
particle dimensions and affects lung deposition. Hygroscopic growth can be
prevented by coating the drug particles with hydrophobic films. However, no
such approach has been successfully implemented in a marketed formulation.
The equilibrium moisture content of a drug and excipient must be
determined over a range of relative humidities so that storage conditions can
be defined and other protective measures can also be considered. Excipients
that modify the hygroscopic properties of a drug may need to be considered.
 Formulation Design and Production Technology in Dry Powder… 99

References
Adi, Santoso S., Handoko Adi, Patricia Tang, Daniela Traini, Hak-Kim Chan,
and Paul M Young. 2008. “Micro-particle corrugation, adhesion and
inhalation aerosol efficiency.” European Journal of Pharmaceutical
Sciences 35 (1-2):12-18. doi: 10.1016/j.ejps.2008.05.009.
Ahmad, Md Iftekhar, Suwipa Ungphaiboon, and Teerapol Srichana. 2015.
“The development of dimple-shaped chitosan carrier for ethambutol
dihydrochloride dry powder inhaler.” Drug Development and Industrial
Pharmacy 41 (5):791-800. doi: 10.3109/03639045.2014.903493.
Ahuja, Gaurav, and Kamla Pathak. 2009. “Porous carriers for
controlled/modulated drug delivery.” Indian Journal of Pharmaceutical
Sciences 71 (6):599-607. doi: 10.4103/0250-474x.59540.
Ali, Mohamed Ehab Hab, and Alf Lamprecht. 2014. “Spray freeze drying for
dry powder inhalation of nanoparticles.” European Journal of
Pharmaceutics and Biopharmaceutics 87 (3):510-7. doi: 10.1016/
j.ejpb.2014.03.009.
Alvim, Izabela Dutra, and Carlos Raimundo Ferreira Grosso. 2010.
“Microparticles obtained by complex coacervation: influence of the type
of reticulation and the drying process on the release of the core material.”
Food Science and Technology 30:1069-1076.
Anderson, Paula J. 2001. “DElivery options and devices for aerosolized
therapeutics*.” Chest 120 (3_suppl):89S-93S. doi: 10.1378/chest.
120.3_suppl.89S.
Ashurst, Ian, Ann Malton, David Prime, and Barry Sumby. 2000. “Latest
advances in the development of dry powder inhalers.” Pharmaceutical
Science & Technology Today 3 (7):246-256. doi: 10.1016/S1461-
5347(00)00275-3.
Balachandran, Wamadeva, Wojciech W. Machowski, E Gaura, and Chris J.
Hudson. 1997. “Control of drug aerosol in human airways using
electrostatic forces.” Journal of Electrostatics 40:579-584.
Beach, Steve, David Latham, Colin Sidgwick, Mazen Hanna, and Peter York.
1999. “Control of the physical form of salmeterol xinafoate.” Organic
Process Research & Development 3 (5):370-376. doi: 10.1021/
op990160z.
100 Teerapol Srichana

Begat, Philippe, David A. V. Morton, John N. Staniforth, and Robert Price.

2004. “The cohesive-adhesive balances in dry powder inhaler
formulations II: Influence on fine particle delivery characteristics.”
Pharmaceutical Research 21 (10):1826-1834.
Broadhead, Joanne, Edmond S. K. Rouan, and Christopher T. Rhodes. 1995.
“Dry-powder inhalers: evaluation of testing methodology and effect of
inhaler design.” Pharmaceutica Acta Helvetiae 70 (2):125-31.
Byron, Peter R., Joanne Peart, and John N. Staniforth. 1997. “Aerosol
electrostatics. I: Properties of fine powders before and after aerosolization
by dry powder inhalers.” Pharmaceutical Research 14 (6):698-705.
Chan, Hak-Kim. 2006. “Dry powder aerosol drug delivery-Opportunities for
colloid and surface scientists.” Colloids and Surfaces A: Physicochemical
and Engineering Aspects 284-285:50-55. doi: 10.1016/j.colsurfa.
2005.10.091.
Chen, Donghao, Richard P. Batycky, Lloyd Johnston, and Jeffrey Mintzes.
2005. Control of process humidity to produce large, porous particles. US
Patents.
Chougule, Mahavir, Bijay Padhi, and Ambikanandan Misra. 2007. “Nano-
liposomal dry powder inhaler of tacrolimus: Preparation, characterization,
and pulmonary pharmacokinetics.” International Journal of
Nanomedicine 2 (4):675-688.
Chow, Albert, Henry Tong, Pratibhash Chattopadhyay, and Boris Shekunov.
2007. “Particle Engineering for Pulmonary Drug Delivery.”
Pharmaceutical Research 24 (3):411-437. doi: 10.1007/s11095-006-9174-
3.
Fan, Ling, Junping Zhang, and Aiqin Wang. 2013. “In situ generation of
sodium alginate/hydroxyapatite/halloysite nanotubes nanocomposite
hydrogel beads as drug-controlled release matrices.” Journal of Materials
Chemistry B 1 (45):6261-6270. doi: 10.1039/c3tb20971g.
Finlay, Warren H. 2001. “1 - Introduction.” In The Mechanics of Inhaled
Pharmaceutical Aerosols, 1-2. London: Academic Press.
Fults, Kristen A., Irving F. Miller, and Anthony J Hickey. 1997. “Effect of
particle morphology on emitted dose of fatty acid treated disodium
cromoglycate powder aerosols.” Pharmaceutical Development and
Technology 2 (1):67-79. doi: 10.3109/10837459709022610.
Ganderton, David. 1999. “Pulmonary delivery: A testing decade.”
Pharmaceutical Technology Europe 11 (10):52.
 Formulation Design and Production Technology in Dry Powder… 101

Grasmeijer, Floris, Paul Hagedoorn, Henderik W. Frijlink, and Anne H. de

Boer. 2013. “Drug content effects on the dispersion performance of
adhesive mixtures for inhalation.” PLoS ONE 8 (8):e71339. doi:
10.1371/journal.pone.0071339.
Gupta, Pramod K., and Anthony J. Hickey. 1991. “Contemporary approaches
in aerosolized drug delivery to the lung.” Journal of Controlled Release
17 (2):127-147. doi: 10.1016/0168-3659(91)90053-G.
Havelund, Svend. 2001. Pulmonary insulin crystals. Google Patents.
Healy, Anne Marie, Maria Ines Nes Amaro, Krzystof Jan An Paluch, and Lidia
Tajber. 2014. “Dry powders for oral inhalation free of lactose carrier
particles.” Advanced Drug Delivery Reviews 75:32-52. doi: 10.1016/
j.addr.2014.04.005.
Hiller, F. Charles. 1991. “Health Implications of Hygroscopic Particle Growth
in the Human Respiratory Tract.” Journal of Aerosol Medicine 4 (1):1-23.
doi: 10.1089/jam.1991.4.1.
Iida, Kotaro, Yukari Inagaki, Hiroaki Todo, Hirokazu Okamoto, Kazumi
Danjo, and Hans Luenberger. 2004. “Effects of surface processing of
lactose carrier particles on dry powder inhalation properties of salbutamol
sulfate.” Chemical & Pharmaceutical Bulletin 52 (8):938-42.
Ikegami, Kazuhiko, Yoshiaki Kawashima, Hirofumi Takeuchi, Hiromitsu
Yamamoto, Nobuyuki Isshiki, Den-ichi Momose, and Kiyohisa Ouchi.
2002. “Improved inhalation behavior of steroid KSR-592 in vitro with
Jethaler by polymorphic transformation to needle-like crystals (beta-
form).” Pharmaceutical Research 19 (10):1439-45.
Ikegami, Kazuhiko, Yoshiaki Kawashima, Hirofumi Takeuchi, Hiromitsu
Yamamoto, Den-ichi Momose, Noriyasu Saito, and Nobuyuki Isshiki.
2000. “In vitro inhalation behavior of spherically agglomerated steroid
particles with carrier lactose.” Advanced Powder Technology 11 (3):323-
332. doi: 10.1163/156855200750172196.
Irngartinger, Meike, V. Camuglia, Michael Damm, J. Goede, and Henderik W.
Frijlink. 2004. “Pulmonary delivery of therapeutic peptides via dry
powder inhalation: effects of micronisation and manufacturing.” European
Journal of Pharmaceutics and Biopharmaceutics 58 (1):7-14. doi:
10.1016/j.ejpb.2004.03.016.
Jetmalani, Kanika, Paul M Young, T. Smith, Peter J. Stewart, and Daniela
Traini. 2012. “Micronized drug powders in binary mixtures and the effect
of physical properties on aerosolization from combination drug dry
powder inhalers.” Drug Development and Industrial Pharmacy 38
(12):1504-11. doi: 10.3109/03639045.2012.654793.
102 Teerapol Srichana

Kaialy, Waseem. 2016. “A review of factors affecting electrostatic charging of

pharmaceuticals and adhesive mixtures for inhalation.” International
Journal of Pharmaceutics 503 (1-2):262-76. doi: 10.1016/j.ijpharm.
2016.01.076.
Kassem, Nuha Mohammed. 1990. “The influence of carrier surface on the
characteristics of inspirable powder aerosls.” Journal of Pharmacy and
Pharmacology 42 (S1):11P-11P. doi: 10.1111/j.2042-7158.1990.
tb14384.x.
Kwok, Philip, Amolnat Tunsirikongkon, William Glover, and Hak-Kim Chan.
2011. “Formation of protein nano-matrix particles with controlled surface
architecture for respiratory drug delivery.” Pharmaceutical Research 28
(4):788-796. doi: 10.1007/s11095-010-0332-2.
Lai, Felix K. Y., and John A. Hersey. 1987. “Simulated ordered powder
mixture.” International Journal of Pharmaceutics 36 (2):157-164. doi:
10.1016/0378-5173(87)90151-7.
Leuenberger, Hans Georg W. 2002. “Spray freeze-drying – the process of
choice for low water soluble drugs?” Journal of Nanoparticle Research 4
(1):111-119. doi: 10.1023/a:1020135603052.
Louey, Margaret D., Sultana Razia, and Peter J. Stewart. 2003. “Influence of
physico-chemical carrier properties on the in vitro aerosol deposition from
interactive mixtures.” International Journal of Pharmaceutics 252 (1-
2):87-98. doi: 10.1016/S0378-5173(02)00621-X.
Nakpheng, Titpawan, Somchai Sawatdee, Khemmarat Buaking, and Teerapol
Srichana. 2011. “Stabilization of luteinizing hormone-releasing hormone
in a dry powder formulation and its bioactivity.” Asian Biomedicine 5
(2):225-233. doi: 10.5372/1905-7415.0502.029.
Park, Chibeom, Ji Eun Park, and Hee Cheul Choi. 2014. “Crystallization-
induced properties from morphology-controlled organic crystals.”
Accounts of Chemical Research 47 (8):2353-2364. doi: 10.1021/
ar5000874.
Pilcer, Gabrielle, and Karim Amighi. 2010. “Formulation strategy and use of
excipients in pulmonary drug delivery.” International Journal of
Pharmaceutics 392 (1-2):1-19.
Rabinow, Barrett E. 2004. “Nanosuspensions in drug delivery.” Nature
Reviews Drug Discovery 3 (9):785-796. doi: 10.1038/nrd1494.
Rahimpour, Yahya, and Hamed Hamishehkar. 2012. “Lactose engineering for
better performance in dry powder inhalers.” Advanced Pharmaceutical
Bulletin 2 (2):183-187. doi: 10.5681/apb.2012.028.
 Formulation Design and Production Technology in Dry Powder… 103

Rahimpour, Yahya, Maryam Kouhsoltani, and Hamed Hamishehkar. 2014.

“Alternative carriers in dry powder inhaler formulations.” Drug Discovery
Today 19 (5):618-626. doi: 10.1016/j.drudis.2013.11.013.
Rasenack, Norbert. 2010. “Particle engineering for pulmonary dosage forms.”
American Pharmaceutical Review 13 (3):76-80.
Rasenack, Norbert, and Bernd W. Müller. 2004. “Micron-size drug particles:
common and novel micronization techniques.” Pharmaceutical
Development and Technology 9 (1):1-13. doi: 10.1081/pdt-120027417.
Rasenack, Norbert, Hartwig Steckel, and Bernd W. Müller. 2003.
“Micronization of anti-inflammatory drugs for pulmonary delivery by a
controlled crystallization process.” Journal of Pharmaceutical Sciences 92
(1):35-44. doi: 10.1002/jps.10274.
Raula, Janne, Anna Lähde, and Esko I. Kauppinen. 2009. “Aerosolization
behavior of carrier-free l-leucine coated salbutamol sulphate powders.”
International Journal of Pharmaceutics 365 (1-2):18-25. doi:
10.1016/j.ijpharm.2008.08.017.
Raula, Janne, Jukka Veli Seppälä, Jari E M Malm, Maarit J. Karppinen, and
Esko I. Kauppinen. 2012. “Structure and dissolution of l-leucine-coated
salbutamol sulphate aerosol particles.” AAPS PharmSciTech 13 (2):707-
712. doi: 10.1208/s12249-012-9789-0.
Rehman, Mohboob, Boris Y. Shekunov, Peter York, David Lechuga-
Ballesteros, Danforth P. Miller, Trixie Tan, and Paul Colthorpe. 2004.
“Optimisation of powders for pulmonary delivery using supercritical fluid
technology.” European Journal of Pharmaceutical Sciences 22 (1):1-17.
doi: 10.1016/j.ejps.2004.02.001.
Rogers, True L., Keith P. Johnston, and Robert O. Williams, 3rd. 2003.
“Physical stability of micronized powders produced by spray-freezing into
liquid (SFL) to enhance the dissolution of an insoluble drug.”
Pharmaceutical Development and Technology 8 (2):187-97. doi:
10.1081/pdt-120018489.
Saleem, Imran Y., Hugh David C Smyth, and Martin J Telko. 2008.
“Prediction of dry powder inhaler formulation performance from surface
energetics and blending dynamics.” Drug Development and Industrial
Pharmacy 34 (9):1002-1010.
Sawatdee, Somchai. 2005. “Formulation design of antibuberculosis dry
powder inhalers.” M.Pharm. Thesis, Faculty of Pharmaceutical Sciences,
Prince of Songkla University.
104 Teerapol Srichana

Schiavone, Helena, Srinivas Palakodaty, Andrew R Clark, Peter A. York, and

Stelios T. Tzannis. 2004. “Evaluation of SCF-engineered particle-based
lactose blends in passive dry powder inhalers.” International Journal of
Pharmaceutics 281 (1-2):55-66. doi: 10.1016/j.ijpharm.2004.05.029.
Shalash, Ahmed O., Abdulla M. Molokhia, and Mustafa M.A. Elsayed. 2015.
“Insights into the roles of carrier microstructure in adhesive/carrier-based
dry powder inhalation mixtures: Carrier porosity and fine particle
content.” European Journal of Pharmaceutics and Biopharmaceutics
96:291-303. doi: 10.1016/j.ejpb.2015.08.006.
Shekunov, Boris Y., Simon Bristow, Albert H. L. Chow, Lachlan Cranswick,
David J. W. Grant, and Peter York. 2003. “Formation of composite
crystals by precipitation in supercritical CO2.” Crystal Growth & Design
3 (4):603-610. doi: 10.1021/cg034026o.
Srichana, Teerapol, Gary P. Martin, and Christopher M. Marriott. 2000. “A
human oral-throat cast integrated with a twin-stage impinger for
evaluation of dry powder inhalers.” Journal of Pharmacy and
Pharmacology 52 (7):771-778.
Srichana, Teerapol, Roongnapa Suedee, and Wantana Reanmongkol. 2001.
“Cyclodextrin as a potential drug carrier in salbutamol dry powder
aerosols: The in-vitro deposition and toxicity studies of the complexes.”
Respiratory Medicine 95 (6):513-519. doi: 10.1053/rmed.2001.1079.
Staniforth, John N., John E. Rees, Felix K. Lai, and John A. Hersey. 1982.
“Interparticle forces in binary and ternary ordered powder mixes.” Journal
of Pharmacy and Pharmacology 34 (3):141-5.
Steckel, Hartwig, and Heike G. Brandes. 2004. “A novel spray-drying
technique to produce low density particles for pulmonary delivery.”
International Journal of Pharmaceutics 278 (1):187-195. doi:
10.1016/j.ijpharm.2004.03.010.
Steckel, Hartwig, and Bernd W. Müller. 1998. “Metered-dose inhaler
formulations with beclomethasone-17,21-dipropionate using the ozone
friendly propellant R 134a.” European Journal of Pharmaceutics and
Biopharmaceutics 46 (1):77-83.
Steckel, Hartwig, Norbert Rasenack, and Bernd W. Müller. 2003. “In-situ-
micronization of disodium cromoglycate for pulmonary delivery.”
European Journal of Pharmaceutics and Biopharmaceutics 55 (2):173-80.
Steckel, Hartwig, Jochen C. Thies, and Bernd W. Müller. 1997. “Micronizing
of steroids for pulmonary delivery by supercritical carbon dioxide.”
International Journal of Pharmaceutics 152 (1):99-110. doi:
10.1016/S0378-5173(97)00071-9.
 Formulation Design and Production Technology in Dry Powder… 105

Tabernero, Antonio, Eva M. Martín del Valle, and Miguel A. Galán. 2012.
“Supercritical fluids for pharmaceutical particle engineering: Methods,
basic fundamentals and modelling.” Chemical Engineering and
Processing: Process Intensification 60:9-25. doi: 10.1016/j.cep.
2012.06.004.
Telko, Martin J., and Anthony J Hickey. 2005. “Dry powder inhaler
formulation.” Respiratory Care 50 (9):1209-1227.
Timsina, M. P., Gary P. Martin, Christopher M. Marriott, David Ganderton,
and Michael Yianneskis. 1994. “Drug delivery to the respiratory tract
using dry powder inhalers.” International Journal of Pharmaceutics 101
(1):1-13. doi: 10.1016/0378-5173(94)90070-1.
Vehring, Reinhard. 2008. “Pharmaceutical particle engineering via spray
drying.” Pharmaceutical Research 25 (5):999-1022. doi: 10.1007/s11095-
007-9475-1.
Velaga, Sitaram P., Rolf A Berger, and Johan Carlfors. 2002. “Supercritical
fluids crystallization of budesonide and flunisolide.” Pharmaceutical
Research 19 (10):1564-71.
Voss, Austin, and Warren H. Finlay. 2002. “Deagglomeration of dry powder
pharmaceutical aerosols.” International Journal of Pharmaceutics 248 (1-
2):39-50. doi: 10.1016/S0378-5173(02)00319-8.
Wang, Yajie, Katherine Kho, Wean Sin In Cheow, and Kunn Hadinoto. 2012.
“A comparison between spray drying and spray freeze drying for dry
powder inhaler formulation of drug-loaded lipid-polymer hybrid
nanoparticles.” International Journal of Pharmaceutics 424 (1-2):98-106.
doi: 10.1016/j.ijpharm.2011.12.045.
Weers, Jeffry G., and Danforth P. Miller. 2015. “Formulation design of dry
powders for inhalation.” Journal of Pharmaceutical Sciences 104
(10):3259-3288. doi: 10.1002/jps.24574.
White, Steven, David B. Bennett, Scot Cheu, Patrick W. Conley, Donald B.
Guzek, Steven Gray, John Howard, Richard Malcolmson, Joann M.
Parker, Phil Roberts, Negar Sadrzadeh, Jacqueline D. Schumacher,
Sangita Seshadri, Gregory W. Sluggett, Cynthia L. Stevenson, and Nancy
J. Harper. 2005. “EXUBERA®: Pharmaceutical development of a novel
product for pulmonary delivery of insulin.” Diabetes Technology &
Therapeutics 7 (6):896-906. doi: 10.1089/dia.2005.7.896.
Wong, Jennifer, Hak-Kim Chan, and Philip Chi Lip Kwok. 2013.
“Electrostatics in pharmaceutical aerosols for inhalation.” Therapeutic
Delivery 4 (8):981-1002. doi: 10.4155/tde.13.70.
106 Teerapol Srichana

Xu, Zhen, Heidi M. Mansour, and Anthony J Hickey. 2011. “Particle

interactions in dry powder inhaler unit processes: A review.” Journal of
Adhesion Science and Technology 25 (4-5):451-482.
Yadidi, Kambiz. 2016. Dry powder formulations for inhalation. Google
Patents.
Young, Paul M, Owen Wood, Jesslynn Ooi, and Daniela Traini. 2011. “The
influence of drug loading on formulation structure and aerosol
performance in carrier based dry powder inhalers.” International Journal
of Pharmaceutics 416 (1):129-135. doi: 10.1016/j.ijpharm.2011.06.020.
Zanen, Pieter, Liam T. Go, and Jan-Willem J. Lammers. 1994. “The optimal
particle size for β-adrenergic aerosols in mild asthmatics.” International
Journal of Pharmaceutics 107 (3):211-217. doi: 10.1016/0378-
5173(94)90436-7.
Zeng, Xian Ming, Gary P. Martin, Christopher Marriott, and John N.
Pritchard. 2000. “The influence of carrier morphology on drug delivery by
dry powder inhalers.” International Journal of Pharmaceutics 200 (1):93-
106.
Zeng, Xian Ming, Gary P. Martin, Seah Kee Tee, Abeer Abu Bu Ghoush, and
Christopher Marriott. 1999. “Effects of particle size and adding sequence
of fine lactose on the deposition of salbutamol sulphate from a dry powder
formulation.” International Journal of Pharmaceutics 182 (2):133-144.
doi: 10.1016/S0378-5173(99)00021-6.
Zhou, Qitony, Li Qu, Thomas R. Gengenbach, Ian C. Larson, Peter J. Stewart,
and David A. V. Morton. 2013. “Effect of surface coating with
magnesium stearate via mechanical dry powder coating approach on the
aerosol performance of micronized drug powders from dry powder
inhalers.” AAPS PharmSciTech 14 (1):38-44. doi: 10.1208/s12249-012-
9895-z.
 Chapter 5

Excipients and Physical

Appearances of Dry
Powder Inhalers

5.1. Introduction
Many DPI formulations are comprised of a drug formulated with a range
of excipients to produce respirable size for efficient lung delivery. An
excipient should be inert and is directed to improve the physical or chemical
stability and pharmaceutical properties. The use of an excipient is based on its
function(s) in the formulation.
The source of excipients are regulated and approved by the Food and Drug
Administration (FDA) and the European Medicines Agency (EMA). These
authorities issue regulatory guidance; however, they have no list of excipients
for DPIs. Nevertheless, the FDA has a list of materials that are generally
recognized as safe. The choice of excipients is based on its potential use in a
previously approved product. It is a requirement that the manufacturer of a
new dosage form submits full detailed production, safety and toxicology
information when applying for new product approval. The number and
quantities of excipients incorporated in a formulation should be kept at
minimal to obtain regulatory approval.
108 Teerapol Srichana

5.2. Qualified Excipients for Dry Powder

Inhaler Formulations
Excipients of currently approved products are summarized in Table 5.1.
They have been demonstrated to be safe for use as excipients intended for
inhalation.

Table 5.1. Accepted or interesting ingredients for use in DPI formulations.

(Adapted from Pilcer and Amighi, International Journal of
Pharmaceutics, 392(1-2), 1-19, 2010)

Excipients Description Status

Sugars Coarse/fine carrier Approved and used
Lactose Approved (Bronchodual®)
Glucose Approved (Exubera®)
Mannitol Promising alternative
Trehalose
Phospholipids Used in liposomes, matrix, Biocompatible/biodegradable
cholesterol coating Antituberculosis (Changsan,
Proliposomes Chan, et al. 2009, Changsan,
Dry encapsulated material Nilkaeo, et al. 2009, Rojanarat et
al. 2011, Rojanarat et al. 2012a,
Rojanarat et al. 2012b)
Cholesteryl Improved drug stability and Ongoing laboratory works
carbonate esters toxicity and drug (Chuealee, Aramwit, and
bioactivity Srichana 2007, Chuealee,
Dry powder carrier Wiedmann, and Srichana 2009,
Drug encapsulation Chuealee et al. 2010, Chuealee et
materials al. 2011, Chuealee, Wiedmann,
Controlled release materials and Srichana 2011)
Leucine,trileucine Improved aerosol efficiency Endogenous substance but no
Coating on particle surface data on lung toxicity
Carrier (Rabbani and Seville 2005,
Co-carrier with other inert Lechuga-Ballesteros et al. 2008,
carriers Rattanupatam and Srichana 2014,
Kaewjan and Srichana 2016)
Bile salts or their Improved drug stability and Endogenous substances may be
derivatives toxicity accepted but at low dose (2–5%
Absorption enhancer w/w)
Encapsulated drug material (Johansson et al. 2002,
Drug carrier Gangadhar, Adhikari, and
Improve aerosol delivery Srichana 2014, Adhikari et al.
2016, Pilcer and Amighi 2010)
 Excipients and Physical Appearances of Dry Powder Inhalers 109

Excipients Description Status

Hydroxypropylated Drug carrier Promising results (Srichana,
-β-CD, natural γ- Host-guest complex Suedee, and Reanmongkol 2001)
CD Improve aerosol delivery
Chitosan, trimethyl Drug carrier Promising results but toxic in
chitosan Dimple carriers chronic use (Huang et al. 2005,
Improve aerosol delivery Grenha et al. 2008, Johansson et
Enhance endocytosis to al. 2002) Carrier (Ahmad,
macrophage cells Nakpheng, and Srichana 2014,
Ahmad, Ungphaiboon, and
Srichana 2015)
PLGA Used in sustained release Immunogenicity observed
formulations (Sivadas et al. 2008, Dailey et al.
2006)

5.2.1. Sugars

Lactose is the only carrier in DPIs marketed in the United States. Lactose
had long been used as an excipient for oral dosage forms before being
deployed in DPIs. Safety and stability profiles of lactose are well established
and the manufacturing process concerning the purity and physical properties
has been documented. It is easily available and inexpensive. Lactose is highly
crystalline with a smooth surface and satisfactory flow properties for a DPI
carrier. Lactose is less hygroscopic than other sugars. Several manufacturers
offer excipient-grade lactose of various sizes and morphologies. One drawback
of lactose is its incompatibility with primary amine moieties. Other sugars,
such as mannitol and glucose have been shown to be feasible alternatives to
lactose, and it is expected that these sugars will eventually be approved
products. Glucose is already used in DPIs in Europe. Mannitol has been used
as a pharmaceutical excipient. Its potential use as a carrier in DPI has been
reported (Hamishehkar et al. 2010, Steckel and Bolzen 2004). Mannitol is less
hygroscopic than lactose and has proven to be the most promising candidate
for this application in comparison to the more hygroscopic sugar alcohols such
as sorbitol, xylitol, and maltitol. D-Mannitol is currently marketed in some
countries as a pulmonary diagnostic DPI (AridolTM) and as a therapeutic dry
powder for inhalation for the treatment of cystic fibrosis and chronic
bronchitis (BronchitolTM), which were recently approved by the FDA and the
European regulatory committee, respectively (Xu et al. 2010). The spherical
mannitol in AridolTM was produced by a spray-drying technique (Tanga et al.
2009). A DPI of ciprofloxacin hydrochloride was prepared by co-spray-drying
with different percentages of mannitol. The combined formulation that
110 Teerapol Srichana

contained 50% (w/w) mannitol showed the best inhalation performance. It had
good stability and the lowest particle cohesion. It was proposed that the co-
spray-dried mannitol and ciprofloxacin DPI was an attractive approach to
promote mucus clearance in the respiratory tract while simultaneously treating
local infections, such as chronic obstructive pulmonary disease and cystic
fibrosis (Adi et al. 2010). A high proportion of a fine fraction of the drugs was
achieved when mannitol was used as the carrier (Harjunen et al. 2003).
Trehalose dihydrate is a disaccharide and a crystalline hydrate like lactose.
Spray-dried inhalable trehalose microparticulate and/or nanoparticulate
powders with low water content were successfully produced by spray drying
under various conditions and were fully characterized with regard to their
water content, morphology and crystallinity (Li and Mansour 2011). A suitable
particle size with good dispersibility and solid-state properties of salbutamol
sulfate was achieved by co-spray-drying with a trehalose–leucine combination
in comparison with other excipients such as mannitol, glycine, and alanine.
Sorbitol can also play a crucial role in the formulation of a respirable
protein. Sorbitol can also serve as a stability enhancer during the processing. It
was reported that the stability of interferon-β to jet milling was dependent
upon the presence of sorbitol in the formulation (Platz, Winters, and Pitt
1994).

5.2.2. Phospholipids and Cholesterol

Phosphatidylcholine and cholesterol have been used in experimental

liposomal formulations. Several other materials have been included in
experimental DPI formulations with various objectives and varying degree of
success. Liposomes are potential drug carriers for a variety of drugs. They
entrap a quantity of materials both within their aqueous compartment (for
hydrophilic molecules) and within the membrane (for hydrophobic molecules).
Liposomes were used to encapsulate rifampicin (RIF) as an alternative
formulation for delivery to the respiratory tract (Changsan, Separovic, and
Srichana 2008). Factors affecting the stability of liposomes containing RIF
were determined. Four liposome suspensions were prepared, containing
different millimole ratios of cholesterol (CH) and soybean l-α-
phosphatidylcholine (SPC) by the chloroform film method, followed by
freeze-drying. Cryo-transmission electron microscopy, photon correlation
spectroscopy, 2H and 31P solid-state nuclear magnetic resonance (SSNMR)
spectroscopy were used to characterize the liposome suspensions. The
 Excipients and Physical Appearances of Dry Powder Inhalers 111

obtained liposomes were a mixture of 200–300 nm unilamellar and

multilamellar vesicles. A higher CH content in the liposome formulation
resulted in a smaller change in the size distribution with time, and a higher CH
content was associated with an increase in the 2H-NMR splitting, that was
indicative of an increase in the order of the lipid acyl chains. Furthermore, the
SSNMR results indicated that RIF was located between the acyl chains of the
phospholipid bilayer and was associated with the CH molecules. Fifty percent
encapsulation of RIF was obtained when the lipid content was high (SPC 10
mM: CH 10 mM). Mannitol was found to be a suitable cryoprotectant which
was attributed to its crystallinity and use of mannitol gave the particles a
MMAD of less than 5 µm. Cholesterol does not form a bilayer structure by its
own, but it can be incorporated into phospholipid membranes to make major
changes in the properties of these membranes and therefore improve the
bilayer characteristics of the liposomes (Vemuri and Rhodes 1995). The
proliposome was employed to encapsulate several anti-tuberculosis drugs
successfully in a dry powder form with good aerosolized properties and
acceptable stability (Rojanarat et al. 2011, Rojanarat et al. 2012a, Changsan,
Nilkaeo, et al. 2009, Changsan, Chan, et al. 2009). Now the anti-tuberculosis
drug is in the clinical trial phase II.

5.2.3. Cholesteryl Carbonate

The cholesterol derivatives were synthesized and employed as a drug

carrier for amphotericin B (AmB) for fungal lung infection. It was found that
the cholesteryl carbonates had the potential to incorporate AmB into the liquid
crystal system. Besides providing aerosolized properties, it also promotes the
antifungal activities of AmB (Chuealee et al. 2010, Chuealee, Wiedmann, and
Srichana 2011, 2009, Chuealee et al. 2011). The particle size varied inversely
to the liquid crystalline content in the cholesteryl carbonate with observed
MMADs that ranged from 4 to 8 μm. This was consistent with the visual
appearance of the liquid crystals as they possessed low density and were free
flowing at room temperature.

5.2.4. Leucine

There are several drug formulations that have included amino acids to
improve their aerosolization behavior, decrease the hygroscopicity of a
112 Teerapol Srichana

micronized drug, and improve its surface activity and the charge density of
cohesive particles (Aquino et al. 2012, Pilcer and Amighi 2010, Seville et al.
2007, Prota et al. 2011). Among the various amino acids used (L-arginine, L-
aspartic acid, L-leucine, L-phenylalanine and L-threonine), L-leucine has been
shown to be the best excipient in terms of aerosolization (Seville et al. 2007,
Chew et al. 2005). It has been suggested that leucine has the best surfactant-
like properties (Gliński, Chavepeyer, and Platten 2000) and has the capacity to
migrate to the droplet surface during the rapid drying phase of the spray-
drying process, and hence the surface characteristics of leucine influence the
resultant particle size (Seville et al. 2007). Selection of an appropriate solvent
system and an L-leucine concentration will allow for the preparation of a
spray-dried powder with enhanced aerosolization properties (Aquino et al.
2012, Rabbani and Seville 2005). Recently our group prepared budesonide co-
spray dried with leucine and found no interaction between budesonide and L-
leucine. The budesonide dry powders had an aerodynamic diameter of 1.9–2.2
µm at a flow rate of 60 LPM that was suitable for pulmonary delivery
(Rattanupatam and Srichana 2014). The incorporation of L-leucine at 10%
improved the %FPF by nearly twofold compared to the normal spray-dried
pyrazinamide (PZA). Changes in the particle density and morphology were
also observed. The dense solid particles of PZA were completely converted to
bulk hollow particles with a thin shell by increasing the L-leucine content up
to 50% (Kaewjan and Srichana 2016).

5.2.5. Bile Salts and Derivatives

There are many bile salts and derivatives available as absorption

enhancers. They have been used for dry powder aerosols (Gangadhar,
Adhikari, and Srichana 2014, Adhikari et al. 2016). The main mechanisms
behind the absorption enhancement are the production of insulin monomers
and an opening of tight junctions between adjacent airway epithelial cells
(Johansson et al. 2002). Although the results are very promising, the use of
high amounts of bile salts may not be feasible for chronic use since they may
damage the epithelial surface. An attempt has been made to enhance the
solubility and stability of AmB, and to evaluate its bioactivity and safety for
use as an inhaler by using a new excipient, sodium deoxycholate sulfate
(SDCS), with the aim of using it as a drug carrier for AmB. Therefore, SDCS
was formulated together with AmB as a dry powder by lyophilization. AmB-
SDCS was shown to be a potential candidate for the treatment of invasive
 Excipients and Physical Appearances of Dry Powder Inhalers 113

pulmonary aspergillosis (Gangadhar, Adhikari, and Srichana 2014, Adhikari et

al. 2016).

5.2.6. Cyclodextrins and Derivatives

Cyclodextrins (CDs) and their derivatives have been shown to improve the
aerosolization properties by modifying the particle morphology and the
surface of the drugs (Srichana, Suedee, and Reanmongkol 2001, Ungaro et al.
2006). It can be used not only to enhance the absorption of the drug across the
pulmonary epithelium but also to increase the deposition of the drug in rat
lungs (Srichana, Suedee, and Reanmongkol 2001). Two types of cyclodextrin
were chosen (i.e., γ-cyclodextrin [GCD] and dimethyl-β-cyclodextrin
[DMCD]) as carriers in the dry powder formulations. Salbutamol was used as
a model drug. The control formulation contained lactose as a carrier. The TSI
was used to evaluate the delivery efficiency of these dry powder formulations.
The toxicity of the cyclodextrin complexes was investigated in the rat by
monitoring the blood urea nitrogen (BUN) and urinary creatinine, as well as
determining the hemolysis of human red blood cells. The release of salbutamol
from the dry powder formulations was also studied over a period of time.
From the results obtained, the formulation containing GCD enhanced the drug
delivery to the lower stage of the TSI (deposition=65%) that was much greater
than that of both formulations containing DMCD (50%) and the control
formulation (40%). After injecting the GCD complex, the BUN and creatinine
levels in rats were similar to those obtained in the control while those that
received the DMCD complex had higher BUN and creatinine. The hemolysis
of red blood cells incubated with the DMCD complex was higher than that
obtained with the GCD complex. The drug release in both formulations
containing GCD and DMCD was fast (over 70% was released in 5 min) and
nearly all the drug was released within 30 min.

5.2.7. Chitosan and Derivatives

Chitosan and trimethyl chitosan which are cationic polysaccharides have

also been widely used as absorption enhancers for proteins and peptides. A
bioadhesive effect of chitosan particles was observed that might be useful to
enhance drug absorption following inhalation. Inhaled chitosan microparticles
induced a significant pulmonary inflammatory response in a dose-dependent
114 Teerapol Srichana

manner after intratracheal administration (Johansson et al. 2002, Huang et al.

2005, Grenha et al. 2008). The chitosan carrier was spray dried to obtain a
spherical shape with a dimpled surface. The chitosan carrier was developed for
delivering the antituberculosis drug ethambutol dihydrochloride (EDH) from a
DPI to the lungs. The EDH size was 222 nm and the chitosan carrier size was
1.2 μm. The chitosan carrier was spherical in shape with a dimpled surface,
and this provided shallow cavities to which the drug was bound, both within
its grooves as well as on its surface. The MMAD of the EDH was between 2.3
and 2.7 μm with an FPF of 32-42% of the nominal dose (Ahmad, Nakpheng,
and Srichana 2014, Ahmad, Ungphaiboon, and Srichana 2015). The authors
suggested that the EDH mixed with a chitosan carrier was suitable for use in a
DPI to control tuberculosis.

5.2.8. Poly)Lactic-co-Glycolic Acid( or PLGA

The properties of PLGA have been extensively investigated as a drug

carrier for administration via the lung and for sustained drug release. PLGA
microspheres have been used in pulmonary delivery for the controlled release
of a variety of drugs, including antiasthmatic drugs, antibiotics, peptides and
proteins. Due to its extremely slow rate of biodegradation, PLGA is
considered unsuitable for pulmonary drug delivery, especially in cases where
frequent dosing is required. After inhalation, production of the inflammatory
cytokine IL-8 was significantly elevated which indicated its immunogenicity
(Dailey et al. 2006, Sivadas et al. 2008). Moreover, the breakdown of PLGA
leads to the acidic degradation products such as lactic and glycolic acid that
can irritate the lungs.

5.3. Physical Appearances

In a previous chapter, the formulation of a DPI was described together
with the production technology. These formulations and technology lead to
different product outlooks. For example, large porous particles,
nanoporous/microparticles, nanoparticles/microparticles, PulmoSphere®,
agglomerates and Trojan microparticles. Table 5.2 lists all of the physical
appearances of DPI particles and their advantages.
 Excipients and Physical Appearances of Dry Powder Inhalers 115

5.3.1. Large Porous Particles

Large porous particles (LPPs) provide some advantages over conventional

microparticles. Inhalation of LPP formulations can prolong drug action in the
lungs by reducing the alveolar macrophage phagocytosis rate. There are
several preparation techniques of LPPs including a double emulsion approach,
precision particle fabrication technology, high-voltage electrostatic antisolvent
method, supercritical fluid technology and a single emulsion technique. The
preparation techniques for large porous particles are listed in Table 5.3.

Table 5.2. Appearance and advantages of different types of DPI particles

Appearances Advantages References

Large porous Low alveolar macrophage (Koushik et al. 2004, Kwon et al.
particles (LPP) uptake 2007, Nolan et al. 2009, Yang et al.
PLGA microparticles 2009, Ungaro et al. 2010, Rawat,
Peptide delivery Majumder, and Ahsan 2008,
Proliposomes with low- Meenach et al. 2012, Rojanarat et al.
density 2012b, Rojanarat et al. 2012a,
Dhanda et al. 2013, Steckel and
Brandes 2004, Healy et al. 2008,
Gupta, Rawat, and Ahsan 2010).
Nanoporous/ Low aggregation (Nolan et al. 2009, Meenach et al.
microparticles High fine particle fraction 2012, Plumley et al. 2009, Ahmad et
Nano particle agglomeration al. 2012, Healy et al. 2008)
Nanoparticles/
microparticles
Pulmospheres Advanced (Dellamary et al. 2000, Smith et al.
micro/nanospheres specially 2001, Duddu et al. 2002, Newhouse
designed for lung delivery et al. 2003, Geller, Weers, and
Hollow porous particles Heuerding 2011, Weers et al. 2015,
Immunoglobulin delivery Hirst et al. 2002, Tarara et al. 2004,
Bot et al. 2000)
Solid lipid Spherical and hollow with (Hadinoto et al. 2007, Hadinoto,
Nanoparticles low-density microparticles Zhu, and Tan 2007, Almeida and
Peptide and Protein delivery Souto 2007, Li et al. 2010)
Agglomerates Particle aggregates with high (Telko and Hickey 2005, Wong et al.
lung deposition 2010)
Trojan Targeted to an organ (Tewes, Ehrhardt, and Healy 2014,
microparticles Magnetic nanoparticles McBride et al. 2013, Upadhyay et al.
Large porous carrier for 2012, Lübbe et al. 1996, Tsapis et al.
nanoparticles 2002)
Dimple shapes Larger surface area (Ahmad, Ungphaiboon, and Srichana
Grooves and dimples on 2015, Ahmad, Nakpheng, and
surface Srichana 2014)
116 Teerapol Srichana

Porous microparticles are potential candidates for pulmonary drug

delivery. A pore forming agent was employed to obtain porous microparticles.
Bernstein et al. (2002) suggested that pore forming agents in the range of 0.01-
90% w/v are able to increase the matrix porosity and pore-formation during
production. They can be added as solid particles to a polymer solution, to a
melted polymer or added as an aqueous solution. Bendroflumethiazide was
prepared successfully in porous microparticles by spray drying with
ammonium carbonate (Healy et al. 2008). Nolan et al. (2009) prepared
nanoporous microparticles of budesonide using ammonium carbonate. Both
studies have demonstrated that in vitro deposition properties were improved
compared to non-porous particles and have the potential use for drug delivery
by inhalation. Porous mannitol had the macropores, micropores and nanopores
with 3-5 m in size (Rojanarat et al. 2012a) (Figure 5.1). Ammonium
carbonate was observed to produce large hollow particles while ammonium
acetate had smaller pores (Figure 5.1b and 5.1d). Nolan et al. (2009) found
that a spray drying process could be employed to produce nanoporous
budesonide microparticles. The addition of ammonium carbonate was
necessary to obtain the porous particles in a methanol/water solvent condition.
The density of the porous particles was significantly lower and the surface area
significantly higher than that of a non-porous spray-dried preparation. The
porous microparticles had improved in vitro deposition properties.
Both pore forming agents decomposed when they were heated as shown in
equations 5.1 and 5.2. However, in this case, ammonium carbonate
decomposed much easier than ammonium acetate and the decomposition
yielded ammonia, carbon dioxide, and water.

CH3COONH4 

 CH3COOH +NH3 (5.1)

(NH4)2CO3 

 2NH3 + CO2 + H2O (5.2)

Porous mannitol generated by the presence of ammonium carbonate had a

higher porosity than that of ammonium acetate because ammonium carbonate
decomposed at a lower temperature. It was volatilized more easily than
ammonium acetate when heated.
 Excipients and Physical Appearances of Dry Powder Inhalers 117

Table 5.3. Preparation of large porous particles and their properties

(Adapted from Healy, A. M. et al.,
Advance Drug Delivery Reviews, 75, 32-52, 2014)

Technique Porogens Carrier Model drug Properties

Double BSA PLGA Insulin and VEGF Initial burst release,
emulsion sustained for 2 weeks in
vitro
PEI PLGA LMW heparin Initial burst release,
retain constant rate,
increase half-life
PEI PLGA PGE1 increase half-life
Ammonium PLGA Budesonide Sustained release for 24
bicarbonate h
Ammonium PLGA Doxorubicin Gradually released over
bicarbonate 2 weeks
SBE-CD PLGA Lysozyme Zero order release up to7
days
HP-β-CD PLGA Insulin Extended in vivo blood
glucose control in rats
HP-β-CD PLGA Palmityl-acylated Linear release for 5 days,
and NaCl exendin-4 no initial burst
HP-β-CD PLGA Risedronate Sustained release up to
and oils Sodium 15 days
Precision Oils PLGA Ciprofloxacin Controlled released for
particle 2-4 weeks, in vitro
fabrication
High Voltage Ammonium PLLA Methotrexate Sustained release
electrostatic bicarbonate
antisolvent
process
SC-CO2 Ammonium PLLA Methotrexate Sustained release
bicarbonate
None PLGA Celecoxib Increased peak level in
lungs
Single Pluronic PLGA rhGH Low initial release,
emulsion F127 sustained release over 1
month.
Solution Ammonium Mannitol Pyrazinamide High fine particle
Antisolvent bicarbonate Levofloxacin fraction
Spray drying Ammonium Low macrophage uptake
acetate
*BSA = bovine serum albumin, PEI = polyethyleneimine, PLGA = poly (lactic-co-glycolic
acid), SBE-β-CD = sulfobutylether β-cyclodextrin, SC-CO2 = supercritical carbon
dioxide
118 Teerapol Srichana

Figure 5.1. The SEM images of porous mannitol generated with ammonium acetate (a-
b) and with ammonium carbonate (c-d) (bar = 5 µm).

The PLGA LPPs of doxorubicin hydrochloride was prepared by adding

ammonium bicarbonate to an aqueous internal phase (primary phase) prior to
mixing with the secondary aqueous phase (Yang et al. 2009). Particles
presented with MMADs of 4.6-5.7 μm and with FPFs of 16-34%. These
particles were able to reduce macrophage uptake and prolong drug release.
Ungaro et al. (2010) applied LPPs to load rhodamine B isothiocyanate–
dextran. These LPPs were considered to be gas-foamed LPPs. During the
production, ammonium bicarbonate released ammonia and the carbon dioxide
gasses resulted in pore formation in the prepared particles. The LPPs showed
favorable in vitro and in vivo deposition in rats. Gupta et al. (2010) reported
the double emulsion solvent evaporation of PLGA LPPs by incorporating
prostaglandin E1 (PGE1) in the external phase. Particle MMADs varied from 1
to 4 μm and showed a prolonged release of PGE1 after pulmonary
administration. Later, Gupta and Ahsan (2011) reported a modified approach
to PLGA-PGE1 LPP production where the drug was incorporated into the
aqueous internal phase by solubilization in a small quantity of ethanol. The
aqueous internal phase was also supplemented with polyethyleneimine (PEI)
as a pore forming agent and a drug-loading enhancer (Figure 5.2). All reported
MMADs were below 5 μm (Gupta and Ahsan 2011).
 Excipients and Physical Appearances of Dry Powder Inhalers 119

The double emulsion solvent evaporation method is the most widely

exploited method to prepare LPPs. PLGA is commonly used as the carrier
material for LPPs. The porogens such as ammonium bicarbonate, HP-β-CD
and sodium chloride were added during the preparation process (Table 5.3).
This method is suitable for encapsulating water-soluble hydrophilic drugs such
as proteins and peptides. The porogens were broken down into gas bubbles
during the emulsification process to form the porous matrix.

Figure 5.2. Scanning electron micrographs of (a) polylactide glycolide, (b) poly(lactic
acid-co-lysine-graft-lysine), (c) b, (d) bovine serum albumin, (e) camptothecin, (f) low
molecular weight heparin-PLGA-polyethylene imine, (g) double emulsion solvent
evaporation ammonium bicarbonate–PLGA-Placebo, (h) DOTAP, (i) prostaglandin E1-
PLGA-PEI large porous particles. (Reprinted from Healy, A. M. et al., Advanced Drug
Delivery Reviews, 75, 32-52, 2014. With permission from Elsevier).

The LPPs developed by Ungaro et al. (2010), Kwon et al. (2007), Rawat
et al. (2008) and Meenach et al. (2012) presented MMAD values of 3-17 μm
with a high FPF and emitted doses, and extended release profiles. Figure 5.2
shows the SEM images of the LPPs with various porogens, carrier materials
and active pharmaceutical ingredients. PEI is also a suitable pore forming
120 Teerapol Srichana

agent for the conventional double emulsion solvent evaporation production

process. It markedly increased the porosity of the LPPs (Rawat, Majumder,
and Ahsan 2008). The osmotic pressure imbalance between the internal and
external aqueous phases may contribute to the formation of porosity. On the
other hand, electrostatic charge interactions between the negatively charged
low molecular weight heparin (LMW heparin) and PEI played an important
role in the pore formation process because they increased the entrapment of
water in the PEI and LMW heparin droplet structure. Sublimation of water that
was entrapped in the droplet structure during lyophilization resulted in large
pores in the particles.

5.3.2. Modified Production Processes for Large

Porous Particles

Steckel and Brandes (2004) expanded the conventional approach to

solution or suspension spray drying for the production of LPPs. LPPs of
salbutamol sulfate were produced by spray drying of a compressed emulsion.
The emulsion was composed of an oil phase (propellant: Solkane™227) and
an aqueous phase containing salbutamol sulfate, phosphatidylcholine,
poloxamer 188, calcium chloride and HP-β-CD. Koushik et al. (2004)
prepared deslorelin-PLGA-HP-β-CD LPPs using a SC-CO2. PLGA in the
methylene chloride was combined with a methanolic solution of deslorelin
with or without HP-β-CD to produce the oil phase. Then the oil phase was
dispersed in an aqueous phase of polyvinyl acetate. The formed microparticles
were centrifuged, washed and freeze-dried. They were subsequently suspended
in SC-CO2. SCF processing was able to produce LPPs using a low process
temperature of 33°C. A similar approach to produce celecoxib LPPs was
reported by Dhanda et al. (2013). PLGA microparticles were produced by
homogenization of the primary emulsion followed by solvent evaporation. The
oil phase was composed of celecoxib, PLGA and dichloromethane and the
aqueous phase of polyvinyl acetate. The primary emulsion was subsequently
diluted with the aqueous solution and stirred to evaporate the dichloromethane.
Microparticles were isolated, washed, freeze-dried and reprocessed using SC-
CO2 to produce LPPs. Celecoxib–PLGA LPPs were proven to have better
control of drug levels in the lungs and improved the lung accumulation index
following a single administration compared to conventional non-porous
particles.
 Excipients and Physical Appearances of Dry Powder Inhalers 121

5.3.3. Nanoporous/Nanoparticulate Microparticles

Nanoporous/nanoparticulate microparticles (NPMPs) were introduced by

Healy at al. (2008). The production of excipient-free porous microparticles by
spray drying was obtained by a mixed solvent/antisolvent system. The SEM of
NPMPs of bendroflumethiazide, budesonide, p-aminosalicylic acid, sodium
cromoglycate, ambroxol HCl and raffinose are shown in Figure 5.3.

Figure 5.3 .Scanning electron micrographs of selected NPMPs materials: (a)

bendroflumethiazide; (b) budesonide; (c) p-aminosalicylic acid (PAS); (d) sodium
cromoglycate; (e) budesonide ambroxol HCl; (f) raffinose. (Reprinted from Healy, A.
M. et al., Advanced Drug Delivery Reviews, 75, 32-52, 2014. With permission from
Elsevier).

A different co-solvent system (ethanol/water and methanol/water) was

also used in the NPMP production for hydrophobic drug and excipients. In
contrast, the water/methanol/butyl acetate and methanol/butyl acetate were
suitable for more hydrophilic materials. During the atomization of the spray
drying process, droplets were formed and rapid drying of these droplets
occurred on contact with the warm drying gas. The solubility of the solute may
condense out as a nanosized liquid phase within the droplet. As further solvent
loss occurs, the solute phase droplets come closer together, and the solute may
precipitate out as primary nanoparticles leading to the formation of NPMP. All
NPMPs produced particle sizes of about 3 μm and due to their porous nature
the NPMPs had lower bulk densities than the spray dried non-porous particles.
122 Teerapol Srichana

NPMPs were all amorphous in nature with the exception of PAS NPMPs that
were crystalline. The NPMPs improved the FPFs to 50–80% when compared
to the non-porous particles. The improved aerosolization properties of the
NPMPs may be attributed to the reduced interparticulate contact.

5.3.4. PulmoSphere®

SD became the main process for the production of the PulmoSphere®

(Weers et al. 2015). PulmoSphere® microparticles are generated from an
emulsion-based feed. The PulmoSphere® particles are less than 5 μm in size.
The porous particles have reduced particle–particle interaction and cohesion,
and the porous nature of the particles has improved flowability and
aerosolization. Dellamary et al. (2000) produced PulmoSphere® containing
cromolyn sodium, salbutamol sulfate or formoterol fumarate in a preparation
of an emulsion containing the drug and additives in an aqueous solution. The
dispersed phase was composed of fluorocarbon PerflubronTM, while the
aqueous phase contained the emulsifier (e.g., phosphatidylcholine). The
homogenized oil-in-water (o/w) emulsion was mixed with the aqueous
solutions and spray dried. During SD, a solid phase was separated from the
evaporating solvent. The fluorocarbon evaporated from the particle surface
that acted as a pore-former. Bot et al. (2000), Smith et al. (2001), Hirst et al.
(2002) and Tarara et al. (2004) reported on the PulmoSphere® platform that
produced particles using hIgG, gentamicin sulfate, salbutamol sulfate and
budesonide, respectively.
Duddu et al. (2002) modified the PulmoSphere® for DPIs. In contrast to
previous work, budesonide was a microcrystalline form instead of being
dissolved in an aqueous solution. First, microcrystals were combined with a
prepared emulsion of perflubron in water. Second, the suspension-emulsion
was homogenized with an aqueous solution of calcium chloride with lactose
monohydrate and spray dried. The PulmoSphere® can be formulated in either
DPIs or pMDIs (Figure 5.4). Pulmonary deposition of the budesonide
PulmoSphere® was around 60%. The peak plasma budesonide for the
PulmoSphere® was about two times greater than for the Pulmicort®. The tmax
was observed at 5 min for the PulmoSphere® compared to 20 min for the
Pulmicort®.
 Excipients and Physical Appearances of Dry Powder Inhalers 123

Figure 5.4. Scanning electron micrographs of PulmoSphere®: (a) sodium

cromoglycate; (b) budesonide as DPIs; (c) budesonide as pMDIs; (d) tobramycin as
DPIs. (Reprinted from Healy, A. M. et al., Advanced Drug Delivery Reviews, 75, 32-
52, 2014. With permission from Elsevier).

Newhouse et al. (2003) compared tobramycin PulmoSphere® inhaled

through a Turbospin® to a nebulized tobramycin (TOBI®). Whole-lung
deposition of the PulmoSphere® was around 34% compared to 5% for the
TOBI®. Peak plasma tobramycin for PulmoSphere® was about three times
larger than the TOBI®. While the area under the curve (AUC) was about two
times greater. Geller et al. (2011) reported an encapsulated tobramycin
PulmoSphere® (Figure 5.4) inhaled through a Podhaler™. This was compared
to a 300 mg tobramycin solution for inhalation. Serum tobramycin levels were
found to be similar for both preparations. The FDA approved the
TOBI®Podhaler™ to treat a bacterial lung infection in cystic fibrosis in 2013.
Weers et al. (2015) investigated dose emission of placebo PulmoSphere®
particles administered as dry powders with a portable, blister-based Simoon
DPI. In vitro particle depositions were found to be independent on the
inhalation maneuver.
124 Teerapol Srichana

5.3.5. Solid Lipid Nanoparticles

The application of polymeric colloidal drug carriers in pulmonary

formulations is often limited by the unknown toxicity of the carrier in the
lungs. Even biodegradable polymers have not yet undergone any rigorous
toxicity testing for safe delivery via the lungs. It has been suggested that lipids
have a faster biodegradation and higher tolerability in the lungs compared to
particles from polymeric materials. It is feasible that aqueous suspensions and
solid lipid dry powders can be used for pulmonary administration. Lipid
carriers include liposomes, lipid emulsions, lipid complexes and solid lipid
micro- and nanoparticles (SLN). One of the common methods for preparing
SLN is high-pressure homogenization. The drug is dissolved in a molten lipid
and then homogenized in an aqueous medium into 200-500 nm particle sizes
followed by solidification. This method follows the same processing steps as
the lipid emulsions which is a well-established pharmaceutical manufacturing
process and can be easily scaled-up. The major drawback, however, is the
limited drug solubility in the molten lipid and the solidified lipid phase.
Alternatively, SLN can be prepared by precipitation from o/w emulsions
which may produce a smaller particle size with a higher drug loading. This can
also be accomplished as a continuous process using a supercritical fluid
extraction method as detailed previously.

5.3.6. Agglomerates

Suspensions of micronized drug crystals and nanoparticles are spray-dried

to form particulate aggregates. Inhalable particles of water-insoluble drugs
and/or porous or hollow particles are able to improve the aerodynamic
properties of the particles. The aggregates consisted of crystals dispersed in a
matrix of amorphous excipients. Different degrees of drug crystallinity can be
obtained depending on the engineered particles in the suspensions and the SD
conditions. This technique was employed to prepare corticosteroid, disodium
cromoglycate and chitosan nanoparticles. Porous aggregates of budesonide
and albuterol sulfate were prepared using the PulmoSphere® platform. These
particles had improved in physical stability, content uniformity and
aerosolization efficiency compared to conventional micronized materials.
Amorphous solid dispersions of poorly soluble drugs can be prepared by
conventional and modified SD. Drug particles can be co-spray dried with
 Excipients and Physical Appearances of Dry Powder Inhalers 125

nanospheres of hydrophilic polymers to modify the particle surface and to

enhance dispersion.

5.3.7. Trojan Microparticles/Magnetically Targeted Dry Powder

Aerosols

Magnetization as a means of targeting drugs to the lungs was investigated.

Magnetization involves incorporating magnetically active particles to a
chemotherapeutic drug. Such particles can be guided to a specific location in
the body using a strong external magnet. Lübbe et al. (1996) performed the
first clinical trials for epirubicin to treat breast cancer by magnetic carriers.
Epirubicin was ionically bound to a modified carbohydrate layer on iron-oxide
nanoparticles (Upadhyay et al. 2012). They observed the accumulation of
nanoparticles in the target area after exposure to the magnetic field. A release
of heat to the surrounding tissues contributed to tumor cell death through
hyperthermia (Upadhyay et al. 2012). The lipid system presented thermo-
sensitive characteristics of drug release at hyperthermic conditions (45°C).
Superparamagnetic iron oxide nanoparticles (SPION) loaded lipid system
produced an inhalable FPF of 30% using an AerosolizerTM.
Tewes et al. (2014) prepared SPIONs-loaded Trojan microparticles by
spray drying. The resulting particles delivered by a Handihaler® were spherical
with a porous surface and an MMAD of 2.2 ± 0.8 μm. In the presence of a
magnetic field, the lower stage deposition increased significantly. These
Trojan particles were highly sensitive to the magnetic field. The authors
suggested that these particles would be useful to treat localized lung disease
such as cancer or bacterial infection. McBride et al. (2013) prepared
magnetically SPION-loaded Trojan microparticles as a dry powder nano-in-
microparticles (NIMs). The NIMs were prepared from a suspension of lactose,
doxorubicin and Fe3O4 SPIONs by SD. TEM and SEM micrographs
demonstrated the porous nature of the NIMs and the surface localization of
SPIONs. Deposition of the NIMs was performed to mimic the conducting
airway deposition. This in vitro study demonstrated more than twice the spatial
deposition and retention of NIMs, compared to a liquid suspension.
Tsapis et al. (2002) spray dried three different systems to produce Trojan
microparticles: (1) polystyrene nanoparticles in ethanol/water (7:3 v/v)
containing dipalmitoylphosphatidylcholine, 1,2-dimyristoyl-sn-glycero-3-
phosphoethanol-amine and lactose with or without hydroxypropyl cellulose;
(2) silica nanoparticles were added to (1), except that water was replaced by 25
126 Teerapol Srichana

mM Tris buffer (pH 9.25) and (3) polystyrene nanoparticles were added to
bovine serum albumin in phosphate buffer with the addition of ammonium
bicarbonate. The nanoparticles were used to prepare the Trojan particles of 25-
100 nm (Figure 5.5).

Figure 5.5. Polystyrene-DPPC Trojan particles: (a) typical hollow sphere Trojan
particles observed from the spray drying of a solution of polystyrene nanoparticles
(170 nm); (b) a magnified view of the particle’s surface. (Reprinted from Tsapis, N. et
al., Proceedings of the National Academy of Sciences, 99(19), 12001-12005, 2002.
Copyright (2002) National Academy of Sciences, U.S.A.).

Figure 5.6. SEM images of chitosan particles that were taken from spraying of a
chitosan solution at 150 ºC with a feeding rate of 3 mL/min. (Reprinted from Ahmad,
M. I. et al., Drug Development and Industrial Pharmacy, 41(5), 791-800, 2015. With
permission from Taylor & Francis).
 Excipients and Physical Appearances of Dry Powder Inhalers 127

These Trojan microparticles exhibited a much better flow and

aerosolization properties than the typical nanoparticles (NPs). The NPs were
held together in the Trojan particles by van der Waals forces within a matrix of
added ingredients such as biopolymers or phospholipids.

Figure 5.7. The SEM images of chitosan and ethambutol mixtures of 1:2 (a), 1:2.5 (b),
1:3.3 (c), 1: 5 (d), 1:10 (e) weight ratios and carrier alone (f). (Reprinted from Ahmad,
M. I. et al., Drug Development and Industrial Pharmacy, 41(5), 791-800, 2015. With
permission from Taylor & Francis).

5.3.8. Dimple Shaped Carriers

Dimple-shaped chitosan particles were prepared for use as a carrier for

ethambutol. The dimple shape of the chitosan carrier provided grooves over its
whole surface that provided a good space for a drug to bind and help in its
128 Teerapol Srichana

dispersion in the oral cavities. Hence, a dimple-shaped carrier is likely to have

better aerosolization properties (Ahmad, Ungphaiboon, and Srichana 2015,
Ahmad, Nakpheng, and Srichana 2014). Figure 5.6 shows the dimple chitosan
prepared with different conditions to generate dimples on the surface of the
carrier. Figure 5.7 shows the nanosized drug particle after mixing with the
dimple shaped carrier. It was found that the drug particles would sit in the
grooves rather than on the smooth surface.

References
Adhikari, Kajiram, Wilaiporn Buatong, Ekawat Thawithong, Tan
Suwandecha, and Teerapol Srichana. 2016. “Factors affecting enhanced
permeation of amphotericin B across cell membranes and safety of
formulation.” AAPS PharmSciTech 17 (4):820-828. doi: 10.1208/s12249-
015-0406-x.
Adi, Handoko, Paul M Young, Hak-Kim Chan, Helen Agus, and Daniela
Traini. 2010. “Co-spray-dried mannitol–ciprofloxacin dry powder inhaler
formulation for cystic fibrosis and chronic obstructive pulmonary
disease.” European Journal of Pharmaceutical Sciences 40 (3):239-247.
Ahmad, Md Iftekhar, Titpawan Nakpheng, and Teerapol Srichana. 2014. “The
safety of ethambutol dihydrochloride dry powder formulations containing
chitosan for the possibility of treating lung tuberculosis.” Inhalation
Toxicology 26 (14):908-917.
Ahmad, Md Iftekhar, Suwipa Ungphaiboon, and Teerapol Srichana. 2015.
“The development of dimple-shaped chitosan carrier for ethambutol
dihydrochloride dry powder inhaler.” Drug Development and Industrial
Pharmacy 41 (5):791-800.
Ahmad, Sana, Bernard Jousseaume, Thierry Toupance, Odile Babot, Guy
Campet, Christine Labrugere, Joachim Brotz, and Ulrike Kunz. 2012. “A
new route towards nanoporous TiO2 as powders or thin films from the
thermal treatment of titanium-based hybrid materials.” Dalton
Transactions 41 (1):292-299. doi: 10.1039/C1DT11087J.
Almeida, António J., and Eliana Souto. 2007. “Solid lipid nanoparticles as a
drug delivery system for peptides and proteins.” Advanced Drug Delivery
Reviews 59 (6):478-490. doi: 10.1016/j.addr.2007.04.007.
 Excipients and Physical Appearances of Dry Powder Inhalers 129

Aquino, Rita Patrizia Atrizia, Lucia Prota, Giulia Auriemma, Antonietta

Santoro, Teresa Mencherini, Gaia Colombo, and Paola F. Russo. 2012.
“Dry powder inhalers of gentamicin and leucine: formulation parameters,
aerosol performance and in vitro toxicity on CuFi1 cells.” International
Journal of Pharmaceutics 426 (1–2):100-107. doi: 10.1016/j.ijpharm.
2012.01.026.
Bernstein, Howard, Donald Chickering, Sarwat Khattak, and Julie Straub.
2002. Matrices formed of polymer and hydrophobic compounds for use in
drug delivery. Acusphere, Inc.
Bot, Adrian I, Thomas E Tarara, Dan J Smith, Simona R Bot, Catherine M
Woods, and Jeffry G Weers. 2000. “Novel lipid-based hollow-porous
microparticles as a platform for immunoglobulin delivery to the
respiratory tract.” Pharmaceutical Research 17 (3):275-283.
Changsan, Narumon, Hak-Kim Chan, Frances Separovic, and Teerapol
Srichana. 2009. “Physicochemical characterization and stability of
rifampicin liposome dry powder formulations for inhalation.” Journal of
Pharmaceutical Sciences 98 (2):628-639. doi: 10.1002/jps.21441.
Changsan, Narumon, Athip Nilkaeo, Pethchawan Pungrassami, and Teerapol
Srichana. 2009. “Monitoring safety of liposomes containing rifampicin on
respiratory cell lines and in vitro efficacy against Mycobacterium bovis in
alveolar macrophages.” Journal of Drug Targeting 17 (10):751-762. doi:
10.3109/10611860903079462.
Changsan, Narumon, Frances Separovic, and Teerapol Srichana. 2008.
“Determination of rifampicin location in cholesterol-lipid liposomes by
2H and 31P solid-state NMR.” 2008 2nd IEEE International
Nanoelectronics Conference, INEC 2008.
Chew, Nora Y. K., Boris Yu Shekunov, Henry H. Y. Tong, Albert H. L.
Chow, Charles Savage, James Wu, and Hak-Kim Chan. 2005. “Effect of
amino acids on the dispersion of disodium cromoglycate powders.”
Journal of Pharmaceutical Sciences 94 (10):2289-2300. doi:
10.1002/jps.20426.
Chuealee, Rabkwan, Pornanong Aramwit, Kusumarn Noipha, and Teerapol
Srichana. 2011. “Bioactivity and toxicity studies of amphotericin B
incorporated in liquid crystals.” European Journal of Pharmaceutical
Sciences 43 (4):308-317. doi: 10.1016/j.ejps.2011.05.009.
Chuealee, Rabkwan, Pornanong Aramwit, and Teerapol Srichana. 2007.
“Characteristics of cholesteryl cetyl carbonate liquid crystals as drug
delivery systems.” Proceedings of the 2nd IEEE International Conference
on Nano/Micro Engineered and Molecular Systems, IEEE NEMS 2007.
130 Teerapol Srichana

Chuealee, Rabkwan, Timothy Scott Wiedmann, and Teerapol Srichana. 2009.

“Thermotropic behavior of sodium cholesteryl carbonate.” Journal of
Materials Research 24 (1):156-163. doi: 10.1557/jmr.2009.0027.
Chuealee, Rabkwan, Timothy Scott Wiedmann, and Teerapol Srichana. 2011.
“Physicochemical properties and antifungal activity of amphotericin b
incorporated in cholesteryl carbonate esters.” Journal of Pharmaceutical
Sciences 100 (5):1727-1735. doi: 10.1002/jps.22398.
Chuealee, Rabkwan, Timothy Scott Wiedmann, Roongnapa Suedee, and
Teerapol Srichana. 2010. “Interaction of Amphotericin B with cholesteryl
palmityl carbonate ester.” Journal of Pharmaceutical Sciences 99
(11):4593-4602. doi: 10.1002/jps.22176.
Dailey, Lea Ann Nn, Norman Jekel, Ludger Fink, Tobias Gessler, Thomas
Schmehl, Matthias Wittmar, Thomas H. Kissel, and Wernar Seeger. 2006.
“Investigation of the proinflammatory potential of biodegradable
nanoparticle drug delivery systems in the lung.” Toxicology and Applied
Pharmacology 215 (1):100-108. doi: 10.1016/j.taap.2006.01.016.
Dellamary, Luis A., Thomas E. Tarara, Dan J. Smith, Christopher H. Woelk,
Anastasios Adractas, Michael L. Costello, Howard Gill, and Jeffry G.
Weers. 2000. “Hollow porous particles in metered dose inhalers.”
Pharmaceutical Research 17 (2):168-174. doi: 10.1023/a:10075
13213292.
Dhanda, Devender S., Puneet Tyagi, Sidney S. Mirvish, and Uday B.
Kompella. 2013. “Supercritical fluid technology based large porous
celecoxib–PLGA microparticles do not induce pulmonary fibrosis and
sustain drug delivery and efficacy for several weeks following a single
dose.” Journal of Controlled Release 168 (3):239-250. doi: 10.1016/
j.jconrel.2013.03.027.
Duddu, Sarma P., Steven A. Sisk, Yulia H. Walter, Thomas E. Tarara, Kevin
R. Trimble, Andrew R. Clark, Michael A. Eldon, Rebecca C. Elton,
Matthew Pickford, Peter H. Hirst, Stephen P. Newman, and Jeffry G.
Weers. 2002. “Improved lung delivery from a passive dry powder inhaler
using an engineered PulmoSphere® powder.” Pharmaceutical Research
19 (5):689-695. doi: 10.1023/a:1015322616613.
Gangadhar, Katkam N., Kajiram Adhikari, and Teerapol Srichana. 2014.
“Synthesis and evaluation of sodium deoxycholate sulfate as a lipid drug
carrier to enhance the solubility, stability and safety of an amphotericin B
inhalation formulation.” International Journal of Pharmaceutics 471 (1-
2):430-438. doi: 10.1016/j.ijpharm.2014.05.066.
 Excipients and Physical Appearances of Dry Powder Inhalers 131

Geller, David E, Jeffry Weers, and Silvia Heuerding. 2011. “Development of

an inhaled dry-powder formulation of tobramycin using PulmoSphere™
technology.” Journal of Aerosol Medicine and Pulmonary Drug Delivery
24 (4):175-182.
Gliński, Jacek, Guy Chavepeyer, and Jean-Karl Platten. 2000. “Surface
properties of aqueous solutions of L-leucine.” Biophysical Chemistry 84
(2):99-103.
Grenha, Ana, Carmen Remuñán-López, Edison LS Carvalho, and Begona
Seijo. 2008. “Microspheres containing lipid/chitosan nanoparticles
complexes for pulmonary delivery of therapeutic proteins.” European
Journal of Pharmaceutics and Biopharmaceutics 69 (1):83-93.
Gupta, Vivek, and Fakhrul Ahsan. 2011. “Influence of PEI as a core
modifying agent on PLGA microspheres of PGE1, a pulmonary selective
vasodilator.” International Journal of Pharmaceutics 413 (1–2):51-62.
doi: 10.1016/j.ijpharm.2011.04.017.
Gupta, Vivek, Amit Rawat, and Fakhrul Ahsan. 2010. “Feasibility study of
aerosolized prostaglandin E1 microspheres as a noninvasive therapy for
pulmonary arterial hypertension.” Journal of Pharmaceutical Sciences 99
(4):1774-1789.
Hadinoto, Kunn, Ponpan Phanapavudhikul, Zhu Kewu, and Reginald B. H.
Tan. 2007. “Dry powder aerosol delivery of large hollow nanoparticulate
aggregates as prospective carriers of nanoparticulate drugs: Effects of
phospholipids.” International Journal of Pharmaceutics 333 (1-2):187-
198. doi: 10.1016/j.ijpharm.2006.10.009.
Hadinoto, Kunn, Kewu Zhu, and Reginald B. H. Tan. 2007. “Drug release
study of large hollow nanoparticulate aggregates carrier particles for
pulmonary delivery.” International Journal of Pharmaceutics 341 (1-
2):195-206. doi: 10.1016/j.ijpharm.2007.03.035.
Hamishehkar, Hamed, Jaber Emami, Abdolhossein Rouholamini Najafabadi,
Kambiz Gilani, Mohsen Minaiyan, Hamid Mahdavi, and Ali Nokhodchi.
2010. “Effect of carrier morphology and surface characteristics on the
development of respirable PLGA microcapsules for sustained-release
pulmonary delivery of insulin.” International Journal of Pharmaceutics
389 (1-2):74-85. doi: 10.1016/j.ijpharm.2010.01.021.
Harjunen, Päivi, Tapio Lankinen, Heikki Salonen, Vesa-Pekka Lehto, and
Kristiina Järvinen. 2003. “Effects of carriers and storage of formulation on
the lung deposition of a hydrophobic and hydrophilic drug from a DPI.”
International Journal of Pharmaceutics 263 (1):151-163.
132 Teerapol Srichana

Healy, Anne Marie, Maria Ines Nes Amaro, Krzystof Jan An Paluch, and Lidia
Tajber. 2014. “Dry powders for oral inhalation free of lactose carrier
particles.” Advanced Drug Delivery Reviews 75:32-52. doi: 10.1016/
j.addr.2014.04.005.
Healy, Anne Marie, Bernard F. McDonald, Lidia Tajber, and Owen I.
Corrigan. 2008. “Characterisation of excipient-free nanoporous
microparticles (NPMPs) of bendroflumethiazide.” European Journal of
Pharmaceutics and Biopharmaceutics 69 (3):1182-1186. doi: 10.1016/
j.ejpb.2008.04.020.
Hirst, Peter H., Gary R. Pitcairn, Jeff G. Weers, Thomas E. Tarara, Andrew R.
Clark, Luis A. Dellamary, Gail Hall, Jolene Shorr, and Stephen P.
Newman. 2002. “In vivo Lung deposition of hollow porous particles from
a pressurized metered dose inhaler.” Pharmaceutical Research 19 (3):258-
264. doi: 10.1023/a:1014482615914.
Huang, Min, Chee-Wai Fong, Eugene Khor, and Lee-Yong Lim. 2005.
“Transfection efficiency of chitosan vectors: effect of polymer molecular
weight and degree of deacetylation.” Journal of Controlled Release 106
(3):391-406.
Johansson, Fredrik I., Elisabeth Hjertberg, Stefan J. Eirefelt, Ann Tronde, and
Ursula Hultkvist Bengtsson. 2002. “Mechanisms for absorption
enhancement of inhaled insulin by sodium taurocholate.” European
Journal of Pharmaceutical Sciences 17 (1-2):63-71.
Kaewjan, Kanogwan, and Teerapol Srichana. 2016. “Nano spray-dried
pyrazinamide-l-leucine dry powders, physical properties and feasibility
used as dry powder aerosols.” Pharmaceutical Development and
Technology 21 (1):68-75. doi: 10.3109/10837450.2014.971373.
Koushik, Kavitha, Devender S. Dhanda, Narayan P. S. Cheruvu, and Uday B.
Kompella. 2004. “Pulmonary delivery of deslorelin: Large-porous PLGA
particles and HP-β-CD complexes.” Pharmaceutical Research 21
(7):1119-1126. doi: 10.1023/b:pham.0000032997.96823.88.
Kwon, Min Jung, Jun Ho Bae, Jung Ju Kim, Kun Na, and Eun Seong Lee.
2007. “Long acting porous microparticle for pulmonary protein delivery.”
International Journal of Pharmaceutics 333 (1–2):5-9. doi: 10.1016/
j.ijpharm.2007.01.016.
Lechuga-Ballesteros, David, Chatan Charan, Cheryl L M Stults, Cynthia L.
Stevenson, Danforth P. Miller, Reinhard Vehring, Vathana Tep, and Mei
Chang Kuo. 2008. “Trileucine improves aerosol performance and stability
of spray-dried powders for inhalation.” Journal of Pharmaceutical
Sciences 97 (1):287-302. doi: 10.1002/jps.21078.
 Excipients and Physical Appearances of Dry Powder Inhalers 133

Li, Xiaojian, and Heidi M Mansour. 2011. “Physicochemical characterization

and water vapor sorption of organic solution advanced spray-dried
inhalable trehalose microparticles and nanoparticles for targeted dry
powder pulmonary inhalation delivery.” AAPS PharmSciTech 12
(4):1420-1430.
Li, Yan-Zhen, Xun Sun, Tao Gong, Jie Liu, Jiao Zuo, and Zhi-Rong Zhang.
2010. “Inhalable microparticles as carriers for pulmonary delivery of
thymopentin-loaded solid lipid nanoparticles.” Pharmaceutical Research
27 (9):1977-1986. doi: 10.1007/s11095-010-0201-z.
Lübbe, Andreas Stephan, Christian Bergemann, Winfried Huhnt, Thomas
Fricke, Hanno Riess, Jeffery Walter Brock, and Dieter Huhn. 1996.
“Preclinical experiences with magnetic drug targeting: tolerance and
efficacy.” Cancer Research 56 (20):4694-4701.
McBride, Amber A., Dominique N. Price, Loreen R. Lamoureux, Alaa A.
Elmaoued, Jose M. Vargas, Natalie L. Adolphi, and Pavan Muttil. 2013.
“Preparation and characterization of novel magnetic nano-in-
microparticles for site-specific pulmonary drug delivery.” Molecular
Pharmaceutics 10 (10):3574-3581. doi: 10.1021/mp3007264.
Meenach, Samantha A., Yu Jeong Kim, Kevin J. Kauffman, Naveen
Kanthamneni, Eric M. Bachelder, and Kristy M. Ainslie. 2012.
“Synthesis, optimization, and characterization of camptothecin-loaded
acetalated dextran porous microparticles for pulmonary delivery.”
Molecular Pharmaceutics 9 (2):290-298. doi: 10.1021/mp2003785.
Newhouse, Michael T, Peter H Hirst, Sarma P Duddu, Yulia H Walter,
Thomas E Tarara, Andrew R Clark, and Jeffry G Weers. 2003. “Inhalation
of a dry powder tobramycin PulmoSphere formulation in healthy
volunteers.” Chest Journal 124 (1):360-366.
Nolan, Lorraine M., Lidia Tajber, Bernard F. McDonald, Ahmad S. Barham,
Owen I. Corrigan, and Anne Marie Healy. 2009. “Excipient-free
nanoporous microparticles of budesonide for pulmonary delivery.”
European Journal of Pharmaceutical Sciences 37 (5):593-602. doi:
10.1016/j.ejps.2009.05.007.
Pilcer, Gabrielle, and Karim Amighi. 2010. “Formulation strategy and use of
excipients in pulmonary drug delivery.” International Journal of
Pharmaceutics 392 (1–2):1-19. doi: 10.1016/j.ijpharm.2010.03.017.
Platz, Robert M, Mark A Winters, and Colin G Pitt. 1994. Pulmonary
administration of granulocyte colony stimulating factor. Amgen Inc.
134 Teerapol Srichana

Plumley, Carl, Eric M. Gorman, Nashwa El-Gendy, Connor R. Bybee, Eric J.

Munson, and Cory Berkland. 2009. “Nifedipine nanoparticle
agglomeration as a dry powder aerosol formulation strategy.”
International Journal of Pharmaceutics 369 (1–2):136-143. doi: 10.1016/
j.ijpharm.2008.10.016.
Prota, Lucia, Antonietta Santoro, Maurizio Bifulco, Rita P Aquino, Teresa
Mencherini, and Paola Russo. 2011. “Leucine enhances aerosol
performance of Naringin dry powder and its activity on cystic fibrosis
airway epithelial cells.” International Journal of Pharmaceutics 412 (1):8-
19.
Rabbani, Naumana R., and Peter C. Seville. 2005. “The influence of
formulation components on the aerosolisation properties of spray-dried
powders.” Journal of Controlled Release 110 (1):130-40. doi:
10.1016/j.jconrel.2005.09.004.
Rattanupatam, Teerarat, and Teerapol Srichana. 2014. “Budesonide dry
powder for inhalation: Effects of leucine and mannitol on the efficiency of
delivery.” Drug Delivery 21 (6):397-405. doi: 10.3109/10717544.2013.
868555.
Rawat, Amit, Quamrul H. Majumder, and Fakhrul Ahsan. 2008. “Inhalable
large porous microspheres of low molecular weight heparin: In vitro and
in vivo evaluation.” Journal of Controlled Release 128 (3):224-232. doi:
10.1016/j.jconrel.2008.03.013.
Rojanarat, Wipaporn, Narumon Changsan, Ekawat Tawithong, Sirirat
Pinsuwan, Hak-Kim Chan, and Teerapol Srichana. 2011. “Isoniazid
proliposome powders for inhalation-preparation, characterization and cell
culture studies.” International Journal of Molecular Sciences 12 (7):4414-
4434. doi: 10.3390/ijms12074414.
Rojanarat, Wipaporn, Titpawan Nakpheng, Ekawat Thawithong, Niracha
Yanyium, and Teerapol Srichana. 2012a. “Inhaled pyrazinamide
proliposome for targeting alveolar macrophages.” Drug Delivery 19
(7):334-345. doi: 10.3109/10717544.2012.721144.
Rojanarat, Wipaporn, Titpawan Nakpheng, Ekawat Thawithong, Niracha
Yanyium, and Teerapol Srichana. 2012b. “Levofloxacin-proliposomes:
opportunities for use in lung tuberculosis.” Pharmaceutics 4 (3):385-412.
Seville, Peter C., Tristan P. Learoyd, Haoying Li, Ian J. Williamson, and
James C. Birchall. 2007. “Amino acid-modified spray-dried powders with
enhanced aerosolisation properties for pulmonary drug delivery.” Powder
Technology 178 (1):40-50. doi: 10.1016/j.powtec.2007.03.046.
 Excipients and Physical Appearances of Dry Powder Inhalers 135

Sivadas, Neeraj, Desmond O’Rourke, Aoife Tobin, Vivienne Buckley, Zeibun

Ramtoola, John G Kelly, Anthony J Hickey, and Sally-Ann Cryan. 2008.
“A comparative study of a range of polymeric microspheres as potential
carriers for the inhalation of proteins.” International Journal of
Pharmaceutics 358 (1):159-167.
Smith, Dan J., Linda M. Gambone, Thomas Tarara, Diana R. Meays, Luis A.
Dellamary, Catherine M. Woods, and Jeffry Weers. 2001. “Liquid dose
pulmonary instillation of gentamicin PulmoSpheres® formulations: Tissue
distribution and pharmacokinetics in rabbits.” Pharmaceutical Research
18 (11):1556-1561. doi: 10.1023/a:1013078330485.
Srichana, Teerapol, Roongnapa Suedee, and Wantana Reanmongkol. 2001.
“Cyclodextrin as a potential drug carrier in salbutamol dry powder
aerosols: The in-vitro deposition and toxicity studies of the complexes.”
Respiratory Medicine 95 (6):513-519. doi: 10.1053/rmed.2001.1079.
Steckel, Hartwig, and Nina Bolzen. 2004. “Alternative sugars as potential
carriers for dry powder inhalations.” International Journal of
Pharmaceutics 270 (1):297-306.
Steckel, Hartwig, and Heike G. Brandes. 2004. “A novel spray-drying
technique to produce low density particles for pulmonary delivery.”
International Journal of Pharmaceutics 278 (1):187-195. doi:
10.1016/j.ijpharm.2004.03.010.
Tang, Patricia, Hak-Kim Chan, Herbert Chiou, Keiko Ogawa, Matthew D.
Jones, Handoko Adi, Graham Buckton, Robert K. Prud’homme, and Judy
Agnes Raper. 2009. “Characterisation and aerosolisation of mannitol
particles produced via confined liquid impinging jets.” International
Journal of Pharmaceutics 367:51-57.
Tarara, Thomas E, Michael S Hartman, Howard Gill, Alan A Kennedy, and
Jeffry G Weers. 2004. “Characterization of suspension-based metered
dose inhaler formulations composed of spray-dried budesonide
microcrystals dispersed in HFA-134a.” Pharmaceutical Research 21
(9):1607-1614.
Telko, Martin J, and Anthony J Hickey. 2005. “Dry powder inhaler
formulation.” Respiratory Care 50 (9):1209-1227.
Tewes, Frederic, Carsten Ehrhardt, and Anne Marie Healy. 2014.
“Superparamagnetic iron oxide nanoparticles (SPIONs)-loaded Trojan
microparticles for targeted aerosol delivery to the lung.” European
Journal of Pharmaceutics and Biopharmaceutics 86 (1):98-104. doi:
10.1016/j.ejpb.2013.09.004.
136 Teerapol Srichana

Tsapis, Nicolas, David Bennett, Blair Jackson, David A Weitz, and DA

Edwards. 2002. “Trojan particles: large porous carriers of nanoparticles
for drug delivery.” Proceedings of the National Academy of Sciences 99
(19):12001-12005.
Ungaro, Francesca, Giuseppe De Rosa, Agnese Miro, Fabiana Quaglia, and
Maria Immacolata La Rotonda. 2006. “Cyclodextrins in the production of
large porous particles: Development of dry powders for the sustained
release of insulin to the lungs.” European Journal of Pharmaceutical
Sciences 28 (5):423-432. doi: 10.1016/j.ejps.2006.05.005.
Ungaro, Francesca, Concetta Giovino, Ciro Coletta, Raffaella Sorrentino,
Agnese Miro, and Fabiana Quaglia. 2010. “Engineering gas-foamed large
porous particles for efficient local delivery of macromolecules to the
lung.” European Journal of Pharmaceutical Sciences 41 (1):60-70. doi:
10.1016/j.ejps.2010.05.011.
Upadhyay, Dhrumil, Santo Scalia, Robert Vogel, Nial Wheate, Rania O
Salama, Paul M Young, Daniela Traini, and Wojciech Chrzanowski. 2012.
“Magnetised thermo responsive lipid vehicles for targeted and controlled
lung drug delivery.” Pharmaceutical Research 29 (9):2456-2467.
Vemuri, Sriram, and Christopher T. Rhodes. 1995. “Preparation and
characterization of liposomes as therapeutic delivery systems: a review.”
Pharmaceutica Acta Helvetiae 70 (2):95-111.
Weers, Jeffry G, Andrew R Clark, Nagaraja Rao, Keith Ung, Alfred Haynes,
Sanjeev K Khindri, Sheryl A Perry, Surendra Machineni, and Paul
Colthorpe. 2015. “In vitro–in vivo correlations observed with indacaterol-
based formulations delivered with the breezhaler®.” Journal of Aerosol
Medicine and Pulmonary Drug Delivery 28 (4):268-280.
Wong, William, David F. Fletcher, Daniela Traini, Hak-Kim Chan, John
Crapper, and Paul M. Young. 2010. “Particle aerosolisation and break-up
in dry powder inhalers 1: Evaluation and modelling of venturi effects for
agglomerated systems.” Pharmaceutical Research 27 (7):1367-1376. doi:
10.1007/s11095-010-0128-4.
Xu, Zhen, Heidi M Mansour, Tako Mulder, Richard Mclean, John R.
Langridge, and Anthony J Hickey. 2010. “Dry powder aerosols generated
by standardized entrainment tubes from drug blends with lactose
monohydrate: 1. Albuterol sulfate and disodium cromoglycate.” Journal
of Pharmaceutical Sciences 99 (8):3398-3414. doi: 10.1002/jps.22107.
 Excipients and Physical Appearances of Dry Powder Inhalers 137

Yang, Yan, Nimisha Bajaj, Peisheng Xu, Kimberly Ohn, Michael D.

Tsifansky, and Yoon Yeo. 2009. “Development of highly porous large
PLGA microparticles for pulmonary drug delivery.” Biomaterials 30
(10):1947-1953. doi: 10.1016/j.biomaterials.2008.12.044.
 Chapter 6

Device Design and

Delivery Efficiency

6.1. Introduction
Asthma is one example of a respiratory disease where airways develop
hyperresponsiveness and inflammation occur. Inhalation of drugs has been
targeted primarily to inhibit both the release of mediators (e.g., sodium
cromoglycate) and their action (e.g., a histamine/amine derivative antagonist),
and to treat the results of mediator action (e.g., β2 agonists, corticosteroids).
Other uses of aerosol therapy include the use of mucolytics that will control
abnormal mucus secretions and antibiotics for lung infections. Aerosol
technology has been used productively for the treatment of respiratory diseases
and when the prime requirement of that aerosol particle is to deposit the drug
in the lung.
There are several factors that can affect the deposition of pharmaceutical
aerosols in the lung as mentioned in the earlier chapter but all still pertain to
the patient, the delivery device or the formulation. Any proposed delivery
system should provide a high drug deposition both in vitro and in vivo.
The particle size has a major role in determining the site and amount of
drug deposited within the respiratory system. To date, inhalation delivery has
been based around three broad delivery systems: metered dose inhalers;
nebulizers and DPIs. These three inhaler systems that are capable of
140 Teerapol Srichana

dispensing such particles will be considered in terms of their different delivery

mechanisms.

6.2. Classifications of
Dry Powder Inhalers
DPIs can be categorized into three types based on their design. The first
group is comprised of single unit dose systems (usually employing a capsule)
that include Spinhaler®, Rotahaler®, Inhalator® and Cyclohaler® devices. The
second type employs multiple unit dose dispensing systems and examples are
the Diskhaler® and Accuhaler®. The third class of DPI is a group of devices
that contain multiple doses such as the Turbuhaler®, Autohaler®, Easyhaler®
and Chiesi®. In a DPI, the drug is placed between the inlet and outlet of a
passage through which air is inspired by the patient. Difficulties can arise in
producing sufficient air flow through the device to entrain the drug and carry it
as far as possible into the patient’s lung. Air turbulence generated through the
device and oral cavity dictates, in part, the amount of drug reaching the lower
airways. The use of a capsule as a unit container such as the Spinhaler®,
Rotahaler®, Inhalator® and Cyclohaler® is convenient and often does not lead
to a marked loss of the drug through residual formulations that remain in the
container after activation. However, moisture retention by the gelatin capsule
and drug might cause problems with the stability of the drug and the ease of
powder dispersion.
Hence, a subsequent generation of multiple unit dose devices has emerged
to solve such problems by protecting the drug from moisture until the point of
administration. However, in some of these newer devices, the FPF has not
shown much improvement. The dose uniformity in some multiple dose
systems can be poor. Multiple dose inhalers may, therefore, require filling with
more drug than is needed to maintain a dose uniformity until the last dose was
dispensed.
 Device Design and Delivery Efficiency 141

6.3. Dry Powder Inhaler

Device Resistances
The respiratory flow rate at which a patient inhales through a DPI device
significantly affects the amount of drug reaching the lung. The energy
produced by the inspiration disperses the drug powder exiting the device. The
interaction between the device design, the pressure drop and the flow rates
generated by the patient are the prime factors that affect not only the efficacy
of the drug delivery but also influence the in vitro test. The internal resistance
of the DPI device is measured using an in-house designed apparatus. The
apparatus consists of a sealed metal box, approximately 1800 cm3, with a
Perspex® cover housing the device. The device is held between two Teflon
ring seals. One side of the box is connected to a pump. Air is drawn through
the apparatus at various flow rates by adjusting an air flow rotameter. The
experimental set up for a resistance measurement of a DPI is shown in Figure
6.1. The pressure drop across the apparatus containing the device is recorded
by a digital manometer (Extech®, Waltham, MA, USA). The difference in the
pressure drop is determined with and without the device in place, and this is
defined as the specific pressure drop of that device at a particular flow rate.
The device resistance is calculated using the pressure drop and the flow rate as
shown in the equation. The square root of the pressure drop versus the flow
rate on the X-axis is plotted; the slope of the graph is the specific resistance of
the device (RD) as shown in equation 6.1.

PD  RDQ (6.1)

Where,

PD = the pressure drop across the device (mbar)

RD = the specific resistance of the device (mbar1/2 /LPM)
Q = flow rate (LPM)

The specific resistance of the device is related to the pressure drop across
the device (Srichana, Martin, and Marriott 1998). A linear relationship has
been shown to exist between the square root of the pressure drop versus the
142 Teerapol Srichana

flow rate for six inhaler devices in human volunteers (Rotahaler®, Spinhaler®,
Cyclohaler®, Diskhaler®, Turbuhaler® and Inhalator Ingelheim®) as shown by
equation 6.1 (Srichana, Martin, and Marriott 1998). The lowest resistance
device was found to be the Rotahaler® followed by the Spinhaler®,
Cyclohaler®, Diskhaler®, Turbuhaler® and the Inhalator Ingelheim®,
respectively.
Amongst inhalers, DPIs play a major role in aerosol delivery for humans.
However, due to the complexity of the peripheral airways, lungs and the
physical mechanisms that govern aerosol delivery, the development of this
kind of device and associated powder formulations remains mainly empirical.

Figure 6.1. Device resistance measurement apparatus.

A measure of the aerosol delivery efficiency is often done in the

laboratory by employing an apparatus specially designed for the aerodynamic
assessment of fine particles by inertial impaction (e.g., the TSI or the ACI). In
addition, it is assumed that the delivery efficiency is significantly determined
by the FPF in the aerosol. This fraction shows the highest probability of
depositing in the lower airways, a process that depends on numerous variables
including the characteristics of the powder formulation and the specific
features of the inhaler device.
The inhaler device produces a pressure drop as the air passes through it.
The pressure drop is defined as the difference between the values of static
pressure measured at two points in a system. A higher flow resistance leads to
a larger pressure drop and ultimately produces a lower flow rate, for the same
 Device Design and Delivery Efficiency 143

inhalation effort. However, a higher pressure drop in the device, that resulted
from the narrow passage areas for the air flow, may also present a beneficial
effect on the generation of fine particles (Srichana, Martin, and Marriott 1998).
The larger shear stresses that act on the particles and/or the higher local
turbulence lead to a better de-agglomeration of the powders. As a
consequence, the inhalation of smaller-sized individual particles that are able
to reach the lungs was promoted. The opposite effect was observed when the
decrease in the inlet mouthpiece size resulted in an increased particle velocity
release from the device. This caused a drug loss via inertial impaction (Figure
6.2) (DeHaan and Finlay 2004).

Figure 6.2. Oral cavity deposition for a 5 µm diameter aerosol at 32 LPM entering
through various diameter straight tube inlets. (Reprinted from DeHaan, W. H. and
Finlay, W. H., Journal of Aerosol Science, 35(3), 309-331, 2004. With permission
from Elsevier).

Thus, a decrease of the diameter of the device inlet mouthpiece may be

useful to produce a high turbulence air-flow through the device. However, it is
necessary to be aware that it may cause a drug loss in the oral cavity via
inertial impaction.
Olsson and Asking (1994a) have suggested using different flow rates to
determine the dose emission from the devices in vitro. For example, 30 LPM
was suggested to be most suitable for high resistance devices such as the
Inhalator Ingelheim®, 60 LPM for the Turbuhaler® and 100-120 LPM for the
144 Teerapol Srichana

Spinhaler®, Rotahaler® and Diskhaler® (Hindle and Byron 1995). Brindley et

al., (1994) characterized DPI devices using an inhalation simulator that
generated typical in vivo conditions on the grounds that one constant flow rate
could not predict the variation in clinical efficacy. This is because the
inspiratory flow rate generated by an individual patient depends on the air flow
resistance in different devices. The inhalation simulator was operated on the
basis of a pressure drop rather than the flow rate and was designed to mimic
the human inhalation profile of pressure drop versus time.

6.4. Dry Powder Inhaler Technology

The concept of the DPI technology is to transport a drug to the pulmonary
system by use of the self-breathing of patients that aims to provide an
accelerated release of the drug from the container and for the fine particles of
the dry powder drug to reach the deep lung. This method was developed in
1967 to transport a sodium cromoglycate powder for the treatment of asthma.
The advantages of this technique are: use of a small amount of active
compound; less systemic side effects compared to the oral route; an increase in
the efficacy by avoiding first-pass metabolism in the liver; simple to operate
when compared to the pMDI and no need for a propellant. Today this
administration technique is admired in major targeted drug delivery systems
for the treatment of pulmonary disease and plays a major role in a novel
pathway for drug delivery.
Commonly, the degree of drug deposition performance depends on the
forces of interaction within the powder formulation and the mechanical forces
of dispersion from the device and the patient’s inspiration effort (Louey,
Razia, and Stewart 2003).

6.4.1. Device Factors

All DPIs have four basic features to release drug particles from the device:
(1) a dose metering mechanism; (2) an aerosolization mechanism; (3) a
deaggregation mechanism and (4) an adaptor to direct the aerosol into the
patient’s mouth. Drug particles are theoretically stripped from the surface of
the lactose particles on which they are loosely attached during the generation
process. This process is illustrated schematically in Figure 6.3. Thus, the drug
 Device Design and Delivery Efficiency 145

particles are dispersed and can traverse the upper respiratory tract, while the
excipient particles do not pass beyond the mouthpiece of the device or the
mouth and throat of the patient (Tiano and Dalby 1996).

Figure 6.3. Mechanism of drug release from dry powder inhaler. (Redrawn from Tiano
S. L. and Dalby, R. N., Pharmaceutical Development and Technology, 1(3), 261-268,
1996).

The design of the device is a crucial factor for the DPI delivery
performance. In common, the device must be able to generate a force that
results in the de-agglomeration of the particles to generate the fine drug
particles (deaggregation mechanism). The concept on how the powder
interacts with the device during dispersion is generally classified into 3 types
of impaction. Air turbulence and mechanical impaction (particle–particle,
particle–device surface) are generally accepted as mechanisms controlling
powder dispersion in the device (Figure 6.4). As a result, fine drug particles
are generated and are delivered to the deep lung.

Figure 6.4. Mechanisms of dispersion of the powder as aerosol inside an inhaler.

146 Teerapol Srichana

Figure 6.5. Relationship between the device resistance, the breathing force and the
device performance.

An interesting impaction behavior that results from the device design is

the device resistance (Figure 6.5). The device resistance is related to the
breathing force. A high device resistance requires a higher effort from the
patients that directly affects the operating flow rate. However, the high device
resistance provides a high degree of turbulent air-flow inside the device, which
in turn increases the performance of the device. In conclusion, the device
should create a balance between the optimal resistance and the optimal
turbulence of the air-flow (Srichana, Martin, and Marriott 1998).

Table 6.1. Characteristics required for an ideal dry powder inhaler

(Ashurst et al. 2000)

Effective dosing Efficient device Easy to use

 Uniform dose throughout its  Good  Simple operation
life environmental  Dose counter
 Targeted and optimized protection  Dose-ready
delivery  In-process controls indicator
 Controlled respirable fraction for quality  Patient feedback of
 Inhalation of dose-  Compact, portable, dose
independent aerosol cheap and/or administration
generation reusable
 Bolus of aerosol available at  Clear comparative
the beginning of an inhalation data for
 Operatable at low flow rates administration
compliance
 Device Design and Delivery Efficiency 147

For the present competition in the DPI market today, every company must
improve their product design to gain customer satisfaction in addition to the
performance of drug delivery. There are lists of possible requirements for an
ideal DPI, which are summarized in Table 6.1.

6.4.2. Disposable Dry Powder Inhalers

Disposable DPIs provide the pulmonary drug delivery of single-dose

applications (e.g., inhaled vaccination). The design of disposable DPIs is quite
different from that of the multi-dose counterparts. In addition to the general
requirements of dose uniformity and deposition, the disposable DPIs should
have features of simple operation with low costs. The TwinCaps® is an
example of a disposable DPI device for delivery of laninamivir octanoate
(Inavir®), a long-acting neuraminidase inhibitor prodrug for the treatment of
influenza. It is different from the short acting zanamivir that requires twice
daily for five days. A single dose of 40 mg inhaled laninamivir can be
effective for a week, making it a perfect candidate. The design of the
TwinCaps® is simple, with only two assembled plastic parts. Since the Inavir®
TwinCaps® has been approved for children less than 10 years old, an
aerosolization performance at lower flow rates is necessary.
The Twincer™ is a disposable device used to deliver colistin for the
treatment of cystic fibrosis (CF). It has three plastic parts that can be easily
assembled by stacking and clicking three pieces together. The formulation is
packed in a blister-like compartment. The powders are dispersed through a
multiple air classifier. Unlike Inavir® that requires only a single
administration, inhaled colistin therapy for CF patients needs to be
administered twice daily for at least 28 days. The disposable design is able to
reduce the potential risk of bacterial contamination.
Aespira’s resQhaler™ is a recently developed disposable DPI based on
the ActiveMesh™ technology. The doses of resQhaler™ are up to 3 mg,
which may limit its use until further design improvements can accommodate
more powder.

6.4.3. Multiple Dose Dry Powder Inhalers

In multiple dose DPIs, the powder formulation is prefilled in a capsule or

cartridge, which is then loaded into the device. With the storage compartment
148 Teerapol Srichana

being separated from the inhaler, multiple dose inhalers can come with small
dimensions. Since the device can be used many times, the cost of medication
per dose is significantly reduced. In some cases, more than one dosing unit can
be given, making dosing more flexible. Multiple dose devices are currently
employed for high dose antibiotics. Both the TOBI® Podhaler® (tobramycin)
and Colobreathe Turbospin® (colistimethate sodium) are capsule-based
multiple dose DPIs. Passive reloadable devices can achieve satisfactory
aerosol performance with minimized flow rate dependence when the drug
powders are properly engineered.

6.5. Present Design of Commercial Dry

Powder Inhaler Devices
DPI devices can be classified into 3 types based on drug dosing (Table
6.2). The single unit dose and multiple unit dose devices are factory metered
doses whereas the multidose is immediately metered after use. The advantages
of the single unit dose device are that each unit dose has environmental
protection by the in-process control during factory dispensing and all of the
single unit dose devices or even some multiple unit dose devices can be
reused. The disadvantage of the single unit dose device is that the patient
either needs to carry a separate reservoir or the device is not portable. The
multidose devices provide ease of operation and can be portable compared to
the unit dose systems, but the devices cannot be cleaned and reused.
Commercial DPI devices have different structures and delivery
mechanisms. The main objective of the design of the device is to optimize the
device resistance so that patients can inspire while producing good turbulent
air-flow inside the device. Some mechanisms of turbulent air-flow production
are illustrated below.

6.5.1. Spinhaler®

The Spinhaler® works by inserting the capsule into an impeller in the body
of the inhaler. The capsule is pierced by the device pins when the system is
ready to use (Figure 6.6). The patient inhales through the mouthpiece. Sodium
cromoglycate deposition has been studied following delivery of the drug by
inhalation to volunteers using the Spinhaler®. There was a marked difference
 Device Design and Delivery Efficiency 149

between-subject variability in the plasma concentrations of sodium

cromoglycate achieved and in the areas under the plasma concentration-time
curves of the drug (Auty et al. 1987). This reflected the variability between
different subjects on the amount of the drug delivered to the respiratory tract.
Most of this variability was due to differences in the inhalation technique,
particularly with regard to the inspiratory flow rate achieved and the duration
of breath-holding after inhalation.

Table 6.2. Currently marketed dry powder inhaler devices.

(Adapted from Lavorini, F. et al., Respiratory, 88(1), 3-15, 2014 and
Berkenfeld, K. et al., AAPS PharmSciTech, 16(3), 479-490, 2015)

Formulation
Device name Type Company
Storage
Spinhaler® Single dose Capsule Aventis
Rotahaler® Single dose Capsule GSK
Cyclohaler®/Aerolizer® Single dose Capsule Pharmachemie/
Novartis
Handihaler® Single dose Capsule Boehringer-
Ingelheim
Turbuhaler® Multiple dose Reservoir Astra Zeneca
Diskhaler® Multiple unit dose Blister pack GSK
Diskus® Multiple unis dose Blister strip GSK
Aerohaler Single dose Capsule Boehringer-
Ingelheim
Easyhaler® Multiple dose Reservoir Orion
Pulvinal® Multiple dose Reservoir Chiesi
NEXThaler® Multiple dose Reservoir Chiesi
Novolizer® Multiple dose Cartridge MEDA
Turbospin® Single dose Capsule PH&T
Jethaler® Multiple dose Ring tablet Ratiopharm
Taifun Multiple dose Reservoir LAB Pharma
Clickhaler® Multiple dose Reservoir Recipharm
FlexhalerTM Multiple dose Reservoir Astra Zeneca
Twisthaler® Multiple dose Reservoir Schering-Plough
Genuair® Multiple dose Reservoir Almirall
Clickhaler® Multiple dose Reservoir Innovate Biomed
ElliptaTM Multiple unit dose Strip GSK
Airmax Multiple unit dose Reservoir Norton Healthcare
150 Teerapol Srichana

Figure 6.6. Spinhaler®.

6.5.2. Rotahaler®

The drug release from the Rotahaler® (GlaxoSmithKline, UK) is started

by a separation of the body from the cap of the capsule. The release of the
drug from the spinning capsule raises the particle impaction and passes
through a crucial grating after the patient inhales through the Rotahaler®
(Figure 6.7). The bronchodilator effects of the inhaled dry salbutamol powder
(50, 100, 200 and 400 μg) have been compared in ten asthmatic patients with
those of an aerosolized salbutamol (200 μg) delivered from a standard
pressurized inhaler. The dose of the salbutamol powder (400 μg) produced a
greater bronchodilation than the 50, 100 or 200 μg of powder, but there was no
significant difference between the 400 μg of powder and the 200 μg of the
aerosolized salbutamol (Hartley et al. 1977). Powder administered by the
Rotahaler® was well tolerated and offered several advantages over the
pressurized aerosol canister. Recently Sim et al. (2014) studied the powder
fluidization and aerosolization processes in the Rotahaler®. The study focused
on examining the effect of the different characteristics of the lactose carrier
employed and specifically considered the powder fluidization, entrainment and
 Device Design and Delivery Efficiency 151

de-agglomeration mechanisms. The results showed that the powder

fluidization from a dynamic split capsule was substantially different from
those of a static powder bed. Furthermore, the presence of the split capsule
dominated the powder emission mechanisms from the Rotahaler®, and this
was regulated by its impaction on the grid/Rotahaler® wall and the rotational
movement in the entrained air.

Figure 6.7. Rotahaler®.

6.5.3. Handihaler®

The Handihaler® works by inserting the capsule into the chamber and a
button is then pressed to pierce the capsule. The patient then inhales the
medication through the mouthpiece. The dry powder will be released from the
pierced holes by the movement of the capsule like a piston. Many chronic
obstructive pulmonary disease patients use their inhaler ineffectively. Inverse
gas chromatography was used to assess the surface energy of the active
(albuterol and ipratropium bromide) and carrier (lactose monohydrate,
trehalose dihydrate, and mannitol) components of a DPI formulation. Blends
(1%, w/w) of the drug and carrier were prepared and evaluated for the DPIs
performance by cascade impaction. The formulations were tested with either
of two passive DPIs, the Rotahaler® (GlaxoSmithKline) or the Handihaler®
152 Teerapol Srichana

(Boehringer Ingelheim). The in vitro performance of the powder blends was

strongly correlated to the surface energy interactions between the active and
carrier components (Cline and Dalby 2002). Increasing the surface energy
interaction between the drug and carrier resulted in a greater FPF of the drug.
One study compared the preference and the ease of use between Diskus® and
Handihaler® (Van Der Palen et al. 2007). Patients had to state a preference for
Diskus® or Handihaler®. Subsequently, they inhaled through a range of
resistances and scored the acceptability. More patients preferred the Diskus®
than the Handihaler®. There was no difference in the number of instructions
needed for both inhalers to obtain a perfect inhalation technique.

Figure 6.8. Handihaler (Redrawn from www.rpc-bramlage.de/Healthcare-en-01.html).

6.5.4. Diskhaler®

A Diskhaler® is a dry-powder inhaler that holds small blisters, each

containing a dose of medication on a disk. The Diskhaler® punctures each
blister so that its medication can be inhaled. An open, randomized, cross-over
study was performed to compare the efficacy and acceptability of two breath-
actuated inhalers, the Turbohaler® and the Diskhaler® for delivery of beta-
agonists (Brown et al. 1992). Thirty-six adults with chronic asthma requiring
 Device Design and Delivery Efficiency 153

beta-agonists four times daily were treated with terbutaline 500 μg via the
Turbohaler® and salbutamol 400 μg via the Diskhaler® four times daily for
four weeks. The peak expiratory flow (PEF) was recorded in the morning
(before and after the beta-agonist) and the evening. The mean morning PEF
was higher during the first two weeks of using the Turbohaler® (295 LPM)
than the Diskhaler® (281 LPM), but this difference did not persist during the
next two weeks and there were no differences in the post-bronchodilator PEF
or use of the beta-agonist rescue. After four weeks, >90% of the patients used
both inhaler devices efficiently and they were equally acceptable in terms of
ease of use and convenience to carry. The Diskhaler® and Turbohaler®
achieved similar clinical efficacy for the delivery of the beta-agonists.

Figure 6.9. Diskhaler® (Redrawn from www.asthma.ca/adults/treatment/diskhaler.php).

6.5.5. Turbuhaler®

The Tubuhaler® is a cylindrical, multi-dose DPI device. The dosing is

achieved by twisting the turning grip back and forth followed by deep
inhalation. It contains 200 metered doses and is equipped with a dosage
indicator window. The most interesting feature is the number of spiral
channels in the device that can produce good fine particles of the drug even in
154 Teerapol Srichana

the presence of drug-drug aggregate formulations (Figure 6.10). The

Turbuhaler® is manufactured by AstraZeneca. It was one of the first DPIs to
dispense metered doses from a reservoir inside the inhaler. The device is made
up of plastic components and a steel spring, with a reservoir. It contains 50,
100 or 200 doses. Each dose is metered accurately regardless of the remaining
dose. It is impossible for the patient to inhale an overdose. The patient will
notice from the dose indicator when there are 20 doses or fewer remaining.
When the grip at the base is fully twisted in one direction and back again a
single dose is loaded. This action fills powder into conical holes in a rotating
dosing disc, and scrapers then remove any surplus drug to ensure accurate
dosing. Inhalation through the mouthpiece forces air through the holes in the
dosing disc which transfers the powder into the deaggregation zone. This
action occurs in the mouthpieces which are designed to create a turbulent flow.
As the drug deaggregation is airflow dependent, extra air is admitted below the
mouthpiece to reduce the pressure drop. The device presents an airflow
resistance of 0.11 (cmH2O)½/LPM. The Turbohaler® has a protective cover
which contains a desiccant to keep low moisture content in the reservoir for at
least 200 open/close cycles. The loaded drug without carrier comprises soft
aggregates of the micronized drug formed by spheronization. The Turbuhaler®
emits a dose that is dependent on the inspiratory flow rate (moderate to high).
The Turbuhaler® has been approved for budesonide (Pulmicort®, Spirocort®),
formoterol (Oxis®), terbutaline (Aerodura®, Bricanyl®) and a combination of
budesonide and formoterol (Symbicort®).

6.5.6. Diskus®/Accuhaler®

The patient operates the inhaler by sliding a lever that moves the next
dose-containing a blister into place. A ratchet within the inhaler causes the
device to click when the next dose is properly positioned. Priming the device
in this way simultaneously peels the two layers of foil apart exposing the dose
ready for inhalation. The Diskus® also incorporates a dose counter, which
enables the patient to monitor the number of doses remaining in the device,
and also has an integral outer case that serves to keep the device dust free and
also resets the lever ready for the next dose. The Diskus® was designed to be
simple to operate and contains a dose counter. The performance of the Diskus®
has been compared with that of a well-established reservoir powder inhaler
(Fuller 1995). The pharmaceutical assessment of the Diskus® has shown that
 Device Design and Delivery Efficiency 155

Figure 6.10. Turbuhaler® (Redrawn from

www.asthma.ca/images/adults/treatment/turbuhaler.gif).
156 Teerapol Srichana

the delivered dose of salmeterol and fluticasone propionate remains at

approximately 90% of the labeled claim at flow rates of 30-90 LPM. This
contrasts with data for the reservoir powder inhaler which showed that the
delivered dose was lower than the labeled claim and more variable,
particularly at flow rates of 30-60 LPM. The delivered dose from the Diskus®
remains constant at different flow rates, unlike the reservoir powder inhaler, in
which the fine particle mass is more dependent on the flow rate. The doses of
the drug in the Diskus™ are protected from moisture. The FPF of salmeterol
delivered from the Diskus® was unaffected by humidity (30°C/75% RH) as
opposed to the reservoir powder inhaler in which the ingress of moisture was
associated with a decrease in FPF. In clinical studies, salmeterol 50 μg twice
daily and fluticasone propionate 50-500μg twice daily have been shown to be
equally effective and well tolerated when delivered by Diskus® as compared
with the Diskhaler®.

Figure 6.11. Diskus® inhaler (Redrawn from

http://productdesignhub.com/2009/04/every-breath-you-take-dissecting-the-diskus/).
 Device Design and Delivery Efficiency 157

6.5.7. Easyhaler®

This powder device includes seven plastic parts. The powder is packaged
into a powder reservoir from which the drug dose can be accurately measured
by rotating the metering cylinder. This is realized by pushing down on the
overcap of the device. The metered drug dose can then be inhaled through the
mouthpiece which has been designed to redisperse the micronized drug
particles from the surface of the carrier material. Thus the powder inhaler is
constructed in such a way that its mode of use is similar to that of the MDI.
The construction of the Easyhaler® allows high dose reproducibility and a
good in vitro and in vivo deposition of the inhaled drug particles. In clinical
studies with an equal therapeutic efficacy, the safety and tolerability to the
MDI have been documented when salbutamol or beclomethasone dipropionate
was delivered from an Easyhaler® to asthmatic adults or children. The FPF
obtained from the Easyhaler® DPI remained relatively constant over the range
of inspiratory flow rates from 30–60 LPM (Koskela et al. 2000).

Figure 6.12. Easyhaler® (Redrawn from Clin Drug Invest © 2002 Adis International
Limited).
158 Teerapol Srichana

6.5.8. NEXThaler®

This device comprises functional groups of components coupled together.

The dosing mechanisms meter the drug from a reservoir, and the counting
mechanisms include the breath-actuated mechanism that activates the dosing
group under a certain air flow allowing the dose to be taken, and the dose
counter to decrement only after an effective release of the therapeutic dose.
There is a reservoir inside the device that houses the drug and every time a
patient opens the cover and breathes through it, the right dose of the powder
formulation is dispensed. To sense the user’s breath, the device is equipped
with a small vane that moves when the patient breathes and this triggers the
dispensing mechanisms. Nexthaler® has a unique feedback system; a click is
heard as a consequence of the activation of the breath-actuated mechanism and
the dose counter decrements by one count only after the effective release of
the dose.

Figure 6.13. NEXThaler® (Redrawn from

www.medicines.org.uk/emc/medicine/31250).

6.5.9. Exubera®

An insulin powder inhaler was developed by Nektar Therapeutics as

Exubera® (Figure 6.14). A spray dried insulin powder is in unit dose blisters
and a reusable inhaler. The device contains an air pump and valves, small jets
(TransJector) and a mouthpiece. The inspiratory effort is power driven to
deliver small amounts of cohesive powder (1–10 mg). A sonic discharge of air
from the base through the TransJector upon actuation into the chamber de-
 Device Design and Delivery Efficiency 159

agglomerates and disperses the inhalable powder into respirable aerosols. A

clear holding chamber allows the patient to visualize the powder dispersal.
Exubera® was approved by the American and European Drug Agencies in
2006. However, Pfizer announced Exubera’s removal from the market due to a
failure to gain market acceptance in 2007.

Figure 6.14. Exubera® (Redrawn from www.drugs.com/pro/exubera.html).

6.5.10. Podhaler™

The Podhaler™ (Novartis) is a portable, capsule-based, single-dose and

multi-use DPI (Figure 6.15). A drug-containing capsule is loaded into the
device by removing the mouthpiece and inserting the capsule into the
chamber. The mouthpiece is screwed back onto the body. When the button is
depressed to pierce the capsule, the patient inhales the DPI through the
mouthpiece. During inspiration, the capsule rotates rapidly in the chamber
which causes capsule emptying drug.
The Podhaler™ has a low airflow resistance of 0.08 (cmH2O)½/LPM to
produce reliable dose delivery. A patient can empty a capsule with a 30-40
160 Teerapol Srichana

LPM flow rate and this result is an emitted dose of 90%. The Podhaler™ is
marketed as the TOBI® tobramycin formulation using the PulmoSphere®
technology. It is indicated for Pseudomonas aeruginosa infections in cystic
fibrosis patients.
More recently, Bayer has launched phase III clinical trials of a
ciprofloxacin DPI Respire® using the PulmoSphere® technology and a
Podhaler™ for non-cystic fibrosis bronchiectasis patients.

Figure 6.15. Redrawn Podhaler™ (Redrawn from Maltz, DS., and Paboojian, SJ.,
Proceedings of RDD Europe, 55-56, 2011).

6.5.11. Turbospin™

The Turbospin™ is a single-dose, multiuse DPI designed by PH&T for

effective lung delivery. Its shape resembles a pen composed of a cap and a
device for inhalation. The make-up of the plastic device consists of a
mouthpiece and body. The body encloses the pulverization chamber and the
piercing apparatus. The capsule is vertically inserted into the chamber and
pierced by the needles at the bottom. Air is drawn through the designed
chamber slits during inspiration which creates turbulence to empty the capsule.
The drug in dry powder form is protected from moisture by packaging the
capsule in the blister.
Turbospin is marketed as Colobreathe® by Forest Laboratories UK Ltd.,
which contains excipient-free micronized colistimethate sodium. It is active
against P. aeruginosa that is commonly found in cystic fibrosis (Schuster et al.
2013).
 Device Design and Delivery Efficiency 161

6.5.12. Staccato®

Staccato® is a single-use inhaler designed by Alexza Pharmaceuticals

which is based on powder sublimation (Figure 6.16). This is obtained by
rapidly heating a thin film of the drug. The heating process is very quick in
order to prevent thermal decomposition of the drug. It is triggered by the
patient’s inhalation, after sublimation the drug cools rapidly in the air and
condenses into 1 to 3 μm aerosol. The aerosol is drawn into the patient’s
mouth and subsequently into the lungs. The emitted dose is ~90% of the drug
and is independent of the patient’s breathing pattern. As the device presents a
valve that controls the airflow, a patient simply places the device to his or her
lips and takes a deep breath. The Staccato® is marketed as loxapine Adasuve™.
In addition, the device has been studied for the delivery of fentanyl, zaleplon,
alprazolam and prochlorperazine. A multiple dose Staccato® is also under
development.

Figure 6.16. Redrawn Staccato® (Redrawn from Dinh, K. et al., International Journal
of Pharmaceutics, 403(1-2), 101-108, 2011).

6.6. Upcoming Devices

6.6.1. ARCUS® inhaler

The ARCUS® inhaler is a small, simple, portable, capsule-based, breath-

actuated device that allows the delivery of single or multi-doses of a drug
formulation using LPP technology. The device consists of two parts: a
cylindrical chamber with multiple vents and a puncturing mechanism, and a
second part consists of the mouthpiece and device body. Upon loading a
capsule into the chamber, the puncturing mechanism creates two holes in the
162 Teerapol Srichana

capsule. The powder is dispersed into the airways after inhalation. The device
has a resistance of 0.28 (cm H2O)1/2/ LPM.

6.6.2. Cricket™ and Dreamboat™

MannKind Corporation has developed two devices: a single-use,

disposable device called Cricket™ and a 15-day reusable device called
Dreamboat™. The Dreamboat™ is currently used on dry powder insulin. The
patient inhales the powder through the mouthpiece in a single breath. After the
patient uses the device, the empty cartridge is removed and discarded.
The patient activates the Cricket™ by depressing the button and inhales
the powder through the mouthpiece in a single breath.
Both devices present a common flow path: as a patient inhales, whereby
two flow inlet streams converge simultaneously. The first inlet stream lifts the
powder from a containment to deliver into a second by-pass inlet stream
(Figure 6.17). The intersection of these two inlet streams de-agglomerates the
fluidized powder, which then travels into the mouth. The powder dispersion
occurs rapidly in the patient’s inhalation maneuver.
In addition, the inhalers utilize a design that enables low flow rate use,
which in turn reduces powder deposition in the oral cavity and promotes a
deep lung deposition.

Figure 6.17. Design concept of Cricket™ and DreamboatTM (Redrawn from

Berkenfeld, K. et al., AAPS PharmSciTech, 16(3), 479-490, 2015).
 Device Design and Delivery Efficiency 163

6.6.3. 3M Taper™

The 3M Taper™ produced by the 3M Drug Delivery Systems is a multi-

dose inhaler presenting the active drug on a microstructured carrier tape
(MCT). The inhaler uses a “dimpled” tape to provide pre-metered doses. The
dimple design allows for drug loading with the carrier; the loading is based on
a balance between drug retention in the dimples and the drug release upon
dosing. The cohesive drug is vital for this process. The drug delivered in each
dose is determined by the number of dimples on the tape and by the volume of
each dimple and the density of the drug packed into the dimples. Individual
doses in the range of 100 μg to 1 mg are possible.
The device is compact and easy to use by only opening the device and a
fixed length of the MCT is presented into the dosing zone. Upon inhaling, the
air flow releases an impactor that strikes the tape and releases the drug into the
airstream. The de-agglomeration of particles occurs as they pass through the
device before closing. The device also provides a dose counter that is easy to
read. The 3M Taper™ DPI is not currently in the market.

6.6.4. MicroDose

MicroDose Therapeutx, Inc., has developed an electronic DPI that utilizes

a piezo vibrator to aerosolize the drug powders packaged in aluminum or
plastic blisters (Figure 6.18). The device is designed for a single-dose or a
multi-unit dose with disposable blisters.
It is operated in four steps: open cap; advance dose; inhale; close cap.
Blisters are pierced with small needles prior to dosing into the flow channel of
the device. Through breath activation, the piezo transducer converts electrical
energy to vibration, creating an air pressure at the blister holes that levitate and
disperse the powder. The powder emitted from the blister is aerosolized into
the lungs. Because the piezo vibrator generates the energy for powder
aerosolization, the dependent inspiratory flow is eliminated. The device is
capable of delivering over 90% emitted dose, with the high fine particle
fractions.
164 Teerapol Srichana

Figure 6.18. MicroDose DPI (Redrawn from Corcoran, T. E. et al., Journal of Aerosol
Medicine and Pulmonary Drug Delivery, 26(1), 46-55, 2013).

6.6.5. Twincer™

The Twincer™ is a disposable DPI developed at the University of

Groningen in the Netherlands (Figure 6.19). It is composed of three plate-like
parts which constitute the air flow passages and a blister chamber containing
the drug to be delivered. The blister has a long cover foil which connects the
powder channel and the inlet to the blister chamber. Air passing through the
powder channel during inhalation aerosolizes the powder from the blister. This
powder flow is divided between two parallel classifiers. The inertial and shear
forces act on the drug agglomerates resulting in de-agglomeration.
The device was designed for colistin and has been used for pulmonary
delivery of peptides, proteins and vaccines. It was manufactured by Indes,
(Netherlands) for clinical trials in cystic fibrosis. The University of Groningen
continues their research on optimizing the Twincer™ for use in different
formulations.
 Device Design and Delivery Efficiency 165

Figure 6.19. TwincerTM (Redrawn from Friebel, C. and Steckel, H., Expert Opinion on
Drug Delivery, 7(12), 1359-1372, 2010).

6.6.6. Dry Powder Inhaler - The University of Western Ontario

The University of Ontario has developed a multi-dose DPI to deliver very

small doses of drug to the lungs (100 - 500 μg). The powder is pre-metered in
small pockets by a rotating multi-dose disc. The holes drilled the disc are
placed between the air tubule and the compressing chamber in the air passage
for a given dose. The double air flow design is applied to produce complete
dispersion of the drug with the break-up of most agglomerates. After one dose
is delivered, the disc can be rotated to set the next dose in the air passage. The
emitted dose from the inhaler was found to vary between 88-92% with
corresponding FPFs of 65- 69%.

6.7. Innovation in Dry Powder

Inhaler Devices
The fine drug particles delivered to the lungs is determined by the balance
of interparticulate forces within the powders, dispersion forces by airflow and
deposition in the airways. Fine powders generated from micronization have
high surface energy and cohesiveness. An inhaler device is necessary to
166 Teerapol Srichana

disperse such powders into inhalable aerosols. The aerosolization behavior of

a DPI depends on powder formulations and inhaler devices. While much effort
has focused on the formulations in the past, there is a growing need to
understand the device design. Innovation on inhaler design can provide
improvements for powder aerosolization.

6.7.1. Design Modification for Passive Devices

The powder dispersion mechanisms in DPIs depend on the device

resistance, air turbulence and impaction. Recent inhaler designs attempt to
increase powder de-agglomeration by increasing the air turbulence and
particulate collisions. The addition of collision in the airflow path can improve
de-agglomeration. Such collision could be in the form of a 3D array of rods or
an oscillating bead. However, not all the designs can substantially improve the
aerosolization efficiency. For the Clickhaler®, the grid mesh does not increase
the respirable doses of salbutamol sulfate or beclometasone dipropionate. The
modifying grid affects not only the turbulence, but also the flow trajectory. In
contrast, a modified Handihaler® equipped with a 3D array of rods in the
mouthpiece improved the FPF from 87.6 to 97.3% of the emitted dose at 45
LPM. Computational fluid dynamics (CFD) analysis revealed that a much
higher specific dissipation factor contributed to the improved aerosolization
behavior compared to an unmodified device. Conix™ is a passive DPI device
that utilized a reverse-flow cyclone technology to generate a high-velocity
vortex to provide collision for de-agglomerated powders.
Other innovative design concepts go beyond the conventional approaches
of generating shear forces and impactions by inhalation to disperse the
powder. In a prototype device, an aero-elastic film coated with drug particles
is fluttered by the inspiratory flow during operation. For high dose drugs, it is
crucial to ensure sufficient loading on the film and consistent detachment of
the particles from the film. A single-use DPI based on the ActiveMesh™
technology was developed with a mesh-sieving dispersion principle. The
powder formulation was packed into a mesh inside the inhaler. The mesh
package vibrates and forces the powder agglomerates to pass through the mesh
holes to promote de-agglomeration during inhalation. Since the mesh vibration
is a function of the airflow rate, device emptying and de-agglomeration may
be a concern at low flow rates for cohesive powders. A 45 μm size mesh plus a
turbulence-generation insert were shown to significantly improve the
performance of the device (Zhou et al. 2014). In both fluttering and mesh-
 Device Design and Delivery Efficiency 167

sieving designs, the airflow is not the direct source for powder dispersion. The
de-agglomeration processes are still driven by the airflow rate.

6.7.2. Active Devices

Passive devices have been widely used for aerosolizing powder

formulations. However, the dispersion performance of many devices suffers
from being flow dependent. This problem is eliminated in the active devices
that use external energy such as compressed air (e.g., Exubera® and Aspirair®),
electrical vibration (e.g., MicroDose) and heat (e.g., Staccato®). However,
active devices are not as popular as passive ones due to their high costs.
A multi-dose active device using compressed air has been recently
developed to deliver micro-doses. Drug powders are pre-filled in small pocket
holes drilled through a rotating disc, with only one drug pocket positioned in
the air passage at a time. Two different compressed airflows are applied,
where the primary flow disperses the drug powder from the pocket and the
secondary flow fluidizes the powder into the primary flow. The FPF of spray
dried insulin, phenylalanine, nitrendipine and phenylalanine ranged from 50 to
70%. This active device has the potential to deliver small amounts of cohesive
powders which could be challenging.
A piezoelectric system has also been employed to aerosolize drug
powders. Powders are stored in a blister in a MicroDose. When the inhaler is
activated by inspiration, the blister is opened and the powders come in contact
with a piezo vibrator. An atropine sulfate powder formulation aerosolized via
the MicroDose DPI showed a FPF of 57% at a flow rate of 28.3 LPM. In
Staccato®, drugs are coated onto a metallic strip and vaporized with a high
temperature within 0.2 s. The drug vapor condenses and generates solid
aerosols. The emitted doses were above 89% and the FPFs were 80–93% at
flow rates of 15-45 LPM. It is worth noting that although the dispersion energy
of the active devices does not rely on the inspiration flow, the oral deposition
of aerosols is dependent on the patient’s inhalation pattern. Thus, drug
deposition in the oropharynx may vary substantially for patients with different
inhalation capacities.
168 Teerapol Srichana

6.7.3. Future Directions in Dry Powder Inhaler Design

Passive devices based on traditional dispersion mechanisms will continue

to dominate the DPI markets in the near future. In past decades, device
efficiency was less significant as the majority of medications only required
microgram drug doses.
However, this has become a crucial aspect with the increased interest of
delivering high drug doses (>100 mg). It is challenging to deliver large
amounts of powder without compromising the device portability and patient
adherence. Furthermore, deposition of a high-dose powder in the oral cavity
by a low-efficiency device could cause severe side effects such as coughing or
irritation which results in the withdrawal of inhalation therapy. Most passive
DPIs are only prescribed to patients aged > 6 years, leaving nebulisation to be
the only choice for the younger children. Active DPIs can open a new window
for the inhaled pediatric therapies. Another opportunity for active DPIs is in
the critically ill patients. In the past, DPIs were unsuitable for patients who
could not generate the required airflow to disperse the powder. Tang et al.
(2011) have developed a novel powder delivery system that is suitable for
patients with the aid of ventilation.

6.7.4. Computational Modeling

With the rapid advancements of the computational technologies, CFD has

become a useful tool in providing information on flow patterns, turbulence
levels and particle trajectories within the devices (Wong et al. 2012,
Suwandecha, Wongpoowarak, and Srichana 2016). For the different
approaches used for the dispersion of pharmaceutical aerosols, the
corresponding advantages and limitations were reviewed by Wong et al.
(2012) and Ruzycki et al. (2013). CFD modeling has been most prevalent for
DPIs. However, the complicated processes involved in DPI dispersion have
challenges to elucidate the underlying powder de-agglomeration mechanisms
by pure numerical analysis. Therefore, a combination of CFD simulations and
in vitro experiments is now commonly used to provide insights into the
relative contributions of the various dispersion mechanisms (Figure 6.20) on
the aerosolization performance (Suwandecha et al. 2016).
Coates et al. (2005, 2007) were among the first to use coupled CFD to
study the Aerolizer® for a carrier-free formulation. The effects of the grid
design, air inlet dimensions, capsule size, mouthpiece dimensions and air flow
 Device Design and Delivery Efficiency 169

have been reported by Wong et al. (2012) and de Boer et al. (2012). Jiang et al.
(2012) investigated further the effect of the height of the swirling chamber of
the Aerolizer® and found improved aerosolization performance that was
attributed to a lengthened powder–air interaction time. The performance of the
Aerolizer® and HandiHaler® for carrier based formulations was investigated
by Donovan et al. (2012). CFD analysis showed that the frequency of the
carrier particle–inhaler collisions increased as the carrier particle size
increased in the Aerolizer®, but not in the HandiHaler®. Therefore, they
attributed this result to the improved aerosol performance of the larger lactose
carrier formulations in the Aerolizer® to the higher carrier–inhaler collisions
and suggested that impaction was the major de-agglomeration mechanism.

Figure 6.20. Particle dispersion mechanism in the cyclohaler. (Reprinted from

Suwandecha, T. et al., Powder Technology, 267, 381-391, 2014. With permission from
Elsevier).
170 Teerapol Srichana

Shur et al. (2012) also studied the flow characteristics within the
HandiHaler® and Aerolizer® to identify the key features of the inhalers. They
fabricated two modified versions of the Aerolizer® with the same specific
resistances as the HandiHaler® to match the aerosol performance for the
carrier-based Spiriva® formulation. These two modified designs exhibited very
different cyclonic flows and only the one with similar flow properties to that
of the HandiHaler® attained a comparable performance. Son, Y. et al. (2012)
later utilized CFD to investigate the mouthpiece designs of the HandiHaler®.
With increased interest in developing high-dose DPIs, de Boer et al.
(2012) evaluated computationally and experimentally the performance of the
high-dose disposable Twincer™ to improve the product performance. The
CFD results showed that the flow split was independent of the pressure drop
across the inhaler and the bypass channel had little contribution to the swirl in
the classifier. Therefore, they proposed blocking the bypass channel to reduce
the total flow rate to minimize the mouth deposition without affecting the
dispersion efficiency. To achieve quantitative analysis of the de-agglomeration
mechanisms, specific interests have been placed to couple the CFD with the
discrete element method (DEM) models that can take into account the
mechanism of particle–particle, particle-fluid and particle–device interactions
during powder dispersion. A coupled CFD-DEM approach was employed by
Tong et al. (2010) to study the de-agglomeration of a single agglomerate
comprising particles of various sizes and polydispersities in a cyclonic flow
model similar to the Aerolizer®. They found that the breakup of the
agglomerates into smaller fragments was governed by the particle–particle
tensile strength and the particle–wall impact energy. Later, they used a two-
way coupled CFD–DEM model to investigate the mechanism of powder de-
agglomeration on mechanical impaction (Tong et al. 2011). They found that
the first impaction only broke the agglomerate into small fragments with a
weakened structure, while the subsequent impaction disintegrated them into
fine particles. They further proposed that the FPF could be expressed as a
unified error function of the ratio between the impaction energy and the
strength of the agglomerate.
Tong et al. (2011) employed a fully coupled CFD–DEM model to study
the dispersion of multiple agglomerates in the Aerolizer® and identified the
particle–wall impaction as the primary de-agglomeration mechanism instead
of the turbulence flow or particle–particle collisions. Sequential impactions of
agglomerates with the inhaler base and grids led to a significant increase in the
FPF. Though the dispersion mechanisms were elucidated, the information was
specific for the formulation and device studied. Recently, CFD–DEM analysis
 Device Design and Delivery Efficiency 171

coupled with in vitro experiments was applied to identify the critical features
of the Aerolizer®. A CFD–DEM showed that the high-velocity particle–wall
impactions in the inhaler dispersion chamber contributed mainly to the
detachment of fine drug particles from the coarse carriers.

Figure 6.21. Correlation of TKE and probability of deagglomeration in Cyclohaler ®

device with fine (♦, dotted line), medium (■, dash line) and large (●, solid line) lactose
carrier formulations (A), and TKE (bar) and impaction (line) energy along longitudinal
length of the Cyclohaler® at 60 LPM using medium sized lactose carriers (B).
(Reprinted from Suwandecha, T. et al., Powder Technology, 267, 381-391, 2014. With
permission from Elsevier).

CFD–DEM modeling is time-consuming compared to the CFD alone.

However, with the increase in the power of advanced processors, it is
envisaged that advanced computations will play a more pivotal role in inhaler
design in the near future.
172 Teerapol Srichana

Recently, Suwandecha et al. (2016) used a simpler computational method

than the CFD-DEM by the CFD and discrete particle model. It was found that
the particles de-agglomerated in two different ways: by aerodynamic shear
dispersion and mechanical impaction de-agglomeration in that respective
order. The aerodynamic shear force created stress on particle agglomeration
and separated them. It was found that the turbulent kinetic energy (TKE) was a
key parameter that described the aerodynamic shear force. If the TKE
overcomes the interparticulate force, it will break the agglomerated particles
(Figure 6.21).

6.7.5. Emerging Dry Powder Inhaler approaches

DPIs can be grouped into four categories: multi-unit, multi-dose reservoir,

reusable single dose and single-use devices. The choice of the device is
dependent on the dose, the dosing frequency and the powder properties. Multi-
dose devices are popular for the frequently used medicines such as the anti-
asthmatic drugs. There has been an increasing emphasis on simplifying the
operation procedure to reduce incorrect device operation and improve patient
adherence. For example, the NEXThaler® is a multi-dose inhaler designed
with a simple operation procedure. A full dose feedback system was built into
the NEXThaler® using a breath-actuated mechanism. The inhaler is activated
only with a certain inspiratory flow to avoid miscounting if the dose is not
inhaled. In a randomized cross-over study of the NEXThaler®, Diskus® and
Turbuhaler® in 66 subjects, the NEXThaler® was rated as the easiest to use and
the most preferred inhaler with a significantly lower number of usage errors
(Voshaar et al. 2014). Reloadable or reusable inhalers were the most favorable
to administer frequently used medications. Most capsule-based DPIs belong to
this group and are still active in the DPI market. In the last five years, there has
been an increasing trend to explore new applications of DPI to deliver
antimicrobials, vaccines or other single-dose active ingredients. As such,
single-use devices serve the purpose of that particular therapy.
 Device Design and Delivery Efficiency 173

References
Ashurst, Ian, Ann Malton, David Prime, and Barry Sumby. 2000. “Latest
advances in the development of dry powder inhalers.” Pharmaceutical
Science & Technology Today 3 (7):246-256. doi: 10.1016/S1461-
5347(00)00275-3.
Auty, Richard M, Kevan Brown, Michael G Neale, and Phillip D Snashall.
1987. “Respiratory tract deposition of sodium cromoglycate is highly
dependent upon technique of inhalation using the Spinhaler.” British
Journal of Diseases of the Chest 81:371-380.
Berkenfeld, Kai, Alf Lamprecht, and Jason T. McConville. 2015. “Devices for
Dry Powder Drug Delivery to the Lung.” AAPS PharmSciTech 16
(3):479-490. doi: 10.1208/s12249-015-0317-x.
Brindley, Anne, Barry S Sumby, and Ian J Smith. 1994. “The characterisation
of inhalation devices by an inhalation simulator: the Electronic Lung™.”
Journal of Aerosol Medicine 7 (2):197-200.
Brown, Peter H., Jill Lenney, Susan W. Armstrong, AC W S Ning, and
Graham K Crompton. 1992. “Breath-actuated inhalers in chronic asthma:
comparison of Diskhaler and Turbohaler for delivery of beta-agonists.”
European Respiratory Journal 5 (9):1143-1145.
Cline, David, and Richard Dalby. 2002. “Predicting the quality of powders for
inhalation from surface energy and area.” Pharmaceutical Research 19
(9):1274-1277.
Coates, Matthew S, Hak-Kim Chan, David F Fletcher, and Herbert Chiou.
2007. “Influence of mouthpiece geometry on the aerosol delivery
performance of a dry powder inhaler.” Pharmaceutical Research 24
(8):1450-1456.
Coates, Matthew S, Hak-Kim Chan, David F Fletcher, and Judy A Raper.
2005. “Influence of air flow on the performance of a dry powder inhaler
using computational and experimental analyses.” Pharmaceutical
Research 22 (9):1445-1453.
Coates, Matthew S, David F Fletcher, Hak-Kim Chan, and Judy A Raper.
2005. “The role of capsule on the performance of a dry powder inhaler
using computational and experimental analyses.” Pharmaceutical
Research 22 (6):923-932.
174 Teerapol Srichana

Corcoran, Timothy Edward, Raman V. Venkataramanan, Robert M. Hoffman,

Meera Patricia George, Andrej A. Petrov, Thomas J. Richards, Shimin
Zhang, Jiyeon Choi, Y. Y. Gao, C. D. Oakum, R. O. Cook, and Michael P.
Donahoe. 2013. “Systemic Delivery of Atropine Sulfate by the MicroDose
Dry-Powder Inhaler.” Journal of Aerosol Medicine and Pulmonary Drug
Delivery 26 (1):46-55. doi: 10.1089/jamp.2011.0948.
de Boer, Anne H, Paul Hagedoorn, Robert Woolhouse, and Ed Wynn. 2012.
“Computational fluid dynamics (CFD) assisted performance evaluation of
the Twincer™ disposable high-dose dry powder inhaler.” Journal of
Pharmacy and Pharmacology 64 (9):1316-1325.
DeHaan, Wesley H., and Warren H Finlay. 2004. “Predicting extrathoracic
deposition from dry powder inhalers.” Journal of Aerosol Science 35
(3):309-331.
Dinh, Khe, Dan J. Myers, Marc Glazer, Tamara Shmidt, Caitlin Devereaux,
Kathleen Simis, Peter D. Noymer, Min He, Corinna Choosakul, Qiang
Chen, and James V. Cassella. 2011. “In vitro aerosol characterization of
Staccato® Loxapine.” International Journal of Pharmaceutics 403 (1–
2):101-108. doi: 10.1016/j.ijpharm.2010.10.030.
Donovan, Martin J, Sin HYEN Kim, Venkatramanan Raman, and Hugh D
Smyth. 2012. “Dry powder inhaler device influence on carrier particle
performance.” Journal of Pharmaceutical Sciences 101 (3):1097-1107.
Friebel, Christian, and Hartwig Steckel. 2010. “Single-use disposable dry
powder inhalers for pulmonary drug delivery.” Expert Opinion on Drug
Delivery 7 (12):1359-1372. doi: 10.1517/17425247.2010.538379.
Fuller, Rick. 1995. “The DiskusTM: a new multi-dose powder device–efficacy
and comparison with TurbuhalerTM.” Journal of Aerosol Medicine 8
(s2):S-11-S-17.
Hartley, Julie P. R., Stephan G. Nogrady, Owain M. Gibby, and Anthony W.
H. Seaton. 1977. “Bronchodilator effects of dry salbutamol powder
administered by Rotahaler.” British Journal of Clinical Pharmacology 4
(6):673-675.
Hindle, Michael, and Peter R Byron. 1995. “Dose emissions from marketed
dry powder inhalers.” International Journal of Pharmaceutics 116
(2):169-177.
Jiang, Liqun, Yue Tang, Hongjiu Zhang, Xifeng Lu, Xijing Chen, and Jiabi
Zhu. 2012. “Importance of powder residence time for the aerosol delivery
performance of a commercial dry powder inhaler Aerolizer®.” Journal of
Aerosol Medicine and Pulmonary Drug Delivery 25 (5):265-279.
 Device Design and Delivery Efficiency 175

Koskela, Tommi, Kristiina Malmström, Ulla Sairanen, S Peltola, Juha Keski-

Karhu, and Matti S L Silvasti. 2000. “Efficacy of salbutamol via
Easyhaler® unaffected by low inspiratory flow.” Respiratory Medicine 94
(12):1229-1233.
Lavorini, Federico, Giovanni Fontana, and Omar Sharif Usmani. 2014. “New
Inhaler Devices - The Good, the Bad and the Ugly.” Respiration 88 (1):3-
15.
Louey, Margaret D., Sultana Razia, and Peter J. Stewart. 2003. “Influence of
physico-chemical carrier properties on the in vitro aerosol deposition from
interactive mixtures.” International Journal of Pharmaceutics 252 (1-
2):87-98. doi: 10.1016/S0378-5173(02)00621-X.
Maltz, David S., and Stephen J. Paboojian. 2011. “Device engineering insights
into TOBI Podhaler: a development case study of high efficient powder
delivery to cystic fibrosis patients.” Proceedings of RDD Europe:55-65.
Olsson, Bo, and Lars Asking. 1994. “A model for the effect of inhalation
device flow resistance on the peak inspiratory flow rate and its application
in pharmaceutical testing.” Journal of Aerosol Medicine 7 (2):201-204.
Ruzycki, Conor A, Emadeddin Javaheri, and Warren H Finlay. 2013. “The use
of computational fluid dynamics in inhaler design.” Expert Opinion on
Drug Delivery 10 (3):307-323.
Schuster, Antje, Cynthia Haliburn, Gerd Döring, Martin Harris Goldman, and
Freedom Study Group. 2013. “Safety, efficacy and convenience of
colistimethate sodium dry powder for inhalation (Colobreathe DPI) in
patients with cystic fibrosis: a randomised study.” Thorax 68 (4):344-350.
Shur, Jagdeep, Sau Lee, Wallace Adams, Robert Lionberger, James Tibbatts,
and Robert Price. 2012. “Effect of device design on the in vitro
performance and comparability for capsule-based dry powder inhalers.”
The AAPS Journal 14 (4):667-676.
Sim, Sally, Kenneth Margo, Jonathan Parks, Ruth Howell, Gerald A Hebbink,
Laurence Orlando, Ian Larson, Philip Leslie, Louise Ho, and David AV
Morton. 2014. “An insight into powder entrainment and drug delivery
mechanisms from a modified Rotahaler®.” International Journal of
Pharmaceutics 477 (1):351-360.
Son, Yoen-Ju, P Worth Longest, and Michael Hindle. 2012. “Aerosolization
characteristics of dry powder inhaler formulations for the enhanced
excipient growth application: effect of DPI design.” Respiratory Drug
Delivery 3:899-902.
176 Teerapol Srichana

Srichana, Teerapol, Gary P. Martin, and Christopher M. Marriott. 1998. “Dry

powder inhalers: the influence of device resistance and powder
formulation on drug and lactose deposition in vitro.” European Journal of
Pharmaceutical Sciences 7 (1):73-80.
Suwandecha, Tan, Wibul Wongpoowarak, Kittinan Maliwan, and Teerapol
Srichana. 2014. “Effect of turbulent kinetic energy on dry powder inhaler
performance.” Powder Technology 267:381-391. doi: 10.1016/j.powtec.
2014.07.044.
Suwandecha, Tan, Wibul Wongpoowarak, and Teerapol Srichana. 2016.
“Computer-aided design of dry powder inhalers using computational fluid
dynamics to assess performance.” Pharmaceutical Development and
Technology 21 (1):54-60. doi: 10.3109/10837450.2014.965325.
Tang, Patricia, Hak-Kim Chan, Dorrilyn Rajbhandari, and Paul Phipps. 2011.
“Method to introduce mannitol powder to intubated patients to improve
sputum clearance.” Journal of Aerosol Medicine and Pulmonary Drug
Delivery 24 (1):1-9.
Tiano, Susan L, and Richard N Dalby. 1996. “Comparison of a respiratory
suspension aerosolized by an air-jet and an ultrasonic nebulizer.”
Pharmaceutical Development and Technology 1 (3):261-268.
Timsina, M. P., Gary P. Martin, Christopher Marriott, David Ganderton, and
Michael Yianneskis. 1994. “Drug delivery to the respiratory tract using
dry powder inhalers.” International Journal of Pharmaceutics 101 (1):1-
13. doi: 10.1016/0378-5173(94)90070-1.
Tong, Zhenbo, Santoso Adi, Runyu Yang, Hak-Kim Chan, and Aibing Yu.
2011. “Numerical investigation of the de-agglomeration mechanisms of
fine powders on mechanical impaction.” Journal of Aerosol Science 42
(11):811-819.
Tong, Zhenbo, Runyu Yang, Kaiwei Chu, Aibing Yu, Santoso Adi, and Hak-
Kim Chan. 2010. “Numerical study of the effects of particle size and
polydispersity on the agglomerate dispersion in a cyclonic flow.”
Chemical Engineering Journal 164 (2):432-441.
Van Der Palen, Job, Michiel M Eijsvogel, Bart F Kuipers, Maria Schipper, and
Niek A Vermue. 2007. “Comparison of the Diskus® Inhaler and the
Handihaler® regarding preference and ease of use.” Journal of Aerosol
Medicine 20 (1):38-44.
 Device Design and Delivery Efficiency 177

Voshaar, Thomas, Monica Spinola, Patrick Linnane, Alice Campanini, Daniel

Lock, Anthony Lafratta, Mario Scuri, Barbara Ronca, and Andrea S
Melani. 2014. “Comparing usability of NEXThaler® with other inhaled
corticosteroid/long-acting β2-agonist fixed combination dry powder
inhalers in asthma patients.” Journal of Aerosol Medicine and Pulmonary
Drug Delivery 27 (5):363-370.
Wong, William, David F Fletcher, Daniela Traini, Hak-Kim Chan, and Paul M
Young. 2012. “The use of computational approaches in inhaler
development.” Advanced Drug Delivery Reviews 64 (4):312-322.
Zhou, Qi, Patricia Tang, Sharon Shui Yee Leung, John Gar Yan Chan, and
Hak-Kim Chan. 2014. “Emerging inhalation aerosol devices and
strategies: Where are we headed?” Advanced Drug Delivery Reviews 75:3-
17. doi: 10.1016/j.addr.2014.03.006.
 Chapter 7

In Vitro Quality Control of Dry

Powder Inhaler

7.1. Introduction
This chapter contains information on the in vitro quality control of dry
powder inhalers. Most of the information has been abstracted from the British
Pharmacopeia (BP), United States Pharmacopeia (USP), European
Pharmacopeia (EP) and European Medicines Agency (EMA). Some examples
of the quality control data have been provided.
Dry powders for inhalers are intended for lung administration in order to
obtain a local or systemic effect (Figure 7.1). They contain one or more active
substances that are dispersed in a suitable vehicle. Preparations of the dry
powder for inhalation may contain diluents. These diluents do not adversely
affect the functions of the mucosa in the respiratory tract. Preparations for
inhalation are supplied in multidose or single-dose containers.
The size of aerosol particles to be inhaled is precisely controlled so that a
significant fraction is deposited in the deep lung. The fine-particle
characteristics of DPIs are determined by the method for aerodynamic
assessment of fine particles.
In assessing the uniformity of a delivered dose of a multidose inhaler, it is
not sufficient to test a single inhaler. Manufacturers must substitute procedures
which take inter- and intra-inhaler dose uniformity into account. A suitable
procedure based on the intra-inhaler test would be to collect each of the
180 Teerapol Srichana

specified doses at the beginning, middle and end of the number of doses stated
on the label from separate inhalers.

Figure 7.1. Airway generation and sizes from the trachea to alveoli. (Reprinted from
Kleinstreuer, C. et al., Annual Review of Biomedical Engineering, 10, 195-220, 2008.
With permission from Annual Reviews)

7.2. Labeling
For the metered-dose preparations, the delivered dose must be stated on
the label: except for preparations for which the dose has been established as a
predispensed-dose and minimum recommended dose.
The dry powder for an inhaler is presented as either single-dose powder
inhalers or as multidose powder inhalers. To facilitate their use, additional
active substances may be combined with a suitable carrier. For pre-metered
inhalers, the inhaler is loaded with powders pre-dispensed in capsules or other
suitable pharmaceutical forms. For inhalers using a powder reservoir, the dose
is created by a metering mechanism within the inhaler. The delivered dose is
the dose delivered from the inhaler. For some preparations, the dose has been
established as a metered dose or as a premetered dose. The metered dose is
calculated from the delivered dose plus the amount deposited on the inhaler.
 In Vitro Quality Control of Dry Powder Inhaler 181

7.3. Uniformity of the Delivered Dose

The dose collection apparatus must be capable of quantitatively capturing
the delivered dose. A dose collection apparatus may be connected to a flow
system according to the scheme specified in Figure 7.2. The dimensions of the
tube and filter should accommodate the measured flow rate unless otherwise
stated. For determination of the test flow rate and duration using the dose
collection tube, an associated flow system, a suitable differential pressure
meter and a suitable flow meter calibrated for the flow leaving the meter are
required according to the following procedure.
Prepare the inhaler for use and connect it to the inlet of the apparatus
using a mouthpiece adapter to ensure an airtight seal. Connect one port of a
differential pressure meter to the pressure reading point, P1 (Figure 7.2) and
let the other to be open to the atmosphere. Switch on the pump, open the 2-
way solenoid valve and adjust the flow control valve until the pressure drop
across the inhaler is 4.0 kPa (40.8 cm H2O) as indicated by the differential
pressure meter. Remove the inhaler from the mouthpiece adapter without
touching the flow control valve, connect a flowmeter to the inlet of the
sampling apparatus. Use a flowmeter calibrated for the flow leaving the meter
or calculate the volumetric flow leaving the meter (Qout) using the ideal gas
law. For a meter calibrated for the entering volumetric flow (Qin), use the
following equation:

Qin  P0
Qout  (7.1)
P0  P

Where,

P0 = atmospheric pressure

P = pressure drop over the meter.

Select a single inhaler and follow the labeled instructions for loading with
powder into the device. Collect a total of 10 doses, three doses at the
beginning, four in the middle (n/2) – 1 to (n/2) + 2, where n is the number of
minimum recommended doses on the label], and three at the end, of the
182 Teerapol Srichana

labeled contents following the labeled instructions. Prior to collecting each of

the doses to be analyzed, clean the inhaler as directed in the labeling.

Figure 7.2. Apparatus suitable for measuring the uniformity of delivered dose of DPI.
(Redrawn from British Pharmacopoeia, 2010).

For preparations containing more than one active substance, carry out the
test for the uniformity of the delivered dose of each active substance. The
preparation complies with the test if 9 out of 10 results lie between 75% and
125% of the average value and all lie between 65% and 135%. If 2 or 3 values
lie outside the limits of 75% to 125%, repeat the test for 2 more inhalers. It is
acceptable when not more than 3 of the 30 values lie outside the limits of 75%
to 125%, and no value lies outside the limits of 65% to 135%.
In justified and authorized cases, these ranges may be extended, but no
value should be greater than 150 % or less than 50 % of the average value.

7.4. Fine Particle Dose

Using the apparatus and procedure described in Aerodynamic assessment
of fine particles (Apparatus C, D or E), calculate the fine particle dose. Fine
particle mass is the amount of drug that is expected to reach the lower airways
obtained from the multistage impactor or any other instrument that has been
calibrated (USP 35-NF 30, BP 2010). The size of the fine particle is less than 5
 In Vitro Quality Control of Dry Powder Inhaler 183

m. Sometimes it is also reported as a fine particle fraction that is a fraction of

fine particle mass over the total dose delivery.

7.5. Number of Deliveries per Inhaler

for Multidose Inhalers
Check constant discharge doses from the inhaler until empty at the
predetermined flow rate. Record the deliveries discharged. The total number
of doses delivered should not be less than the number stated on the label (this
test may be combined with the test for the uniformity of the delivered dose).

7.6. Aerodynamic Assessment

of Fine Particles
This test is used to determine the fine particle characteristics of the aerosol
clouds generated by preparations for inhalation unless otherwise justified and
authorized (BP 2010) using one of the following apparatuses and test
procedures. Stage mensuration is performed periodically together with
confirmation of other dimensions critical to the effective operation of the
impactor.
Re-entrainment (for apparatus D and E). To ensure efficient particle
capture, each plate is coated with glycerol, silicone oil or a similar high
viscosity liquid, typically deposited from a volatile solvent. Although plate
coating is essential for validation of the method, it may be omitted where
justified and authorized.
Mass balance. The total mass of the active substance is not less than 75%
and not more than 125% of the average delivered dose determined during
testing for uniformity of a delivered dose. This is not a test of the inhaler, but it
serves to ensure that the results are valid.

7.6.1. Apparatus A - Glass Impinger

The glass impinger is a two stage impinger, where stage 1 is the upper
impingement chamber and stage 2 is the lower impingement chamber. This
184 Teerapol Srichana

apparatus is used to determine the fine particle characteristics of the aerosol

clouds generated by preparations for inhalation (Figure 7.3).

Figure 7.3. Apparatus A: glass impinger (Dimensions in millimeters; tolerances ±1 mm

unless otherwise described).

Procedure for Powder Inhalers

Introduce 7 mL and 30 mL of a suitable solvent into the upper and lower
impingement chambers, respectively. Connect all the components. Ensure that
the assembly is vertical and adequately supported and the jet-spacer peg of the
lower jet assembly just touches the bottom of the lower impingement chamber.
Without the inhaler in place, connect a suitable pump to the outlet of the
apparatus. Adjust the air flow through the apparatus, as measured at the inlet
to the throat, to 60 ± 5 LPM. Prepare the inhaler for use and locate the
mouthpiece in the apparatus using a suitable adapter. Switch on the pump for 5
s. Switch off the pump and remove the inhaler. Repeat the discharge sequence.
The number of discharges should be minimized and typically would not be
greater than 10.
Dismantle the apparatus. Wash the inner surface of the inlet tube to the
lower impingement chamber and its outer surface that projects into the
 In Vitro Quality Control of Dry Powder Inhaler 185

chamber with a suitable solvent. Collect the washings in the lower

impingement chamber. Determine the content of active substance in this
solution. Calculate the amount of active substance collected in the lower
impingement chamber per discharge and express the results as a percentage of
the dose stated on the label.

7.6.2. Apparatus C - Multi-Stage Liquid Impinger

Fine Particle Dose and Particle Size Distribution

The multi-stage liquid impinger consists of impaction stages 1 (pre-
separator), 2, 3 and 4 and an integral filter stage (stage 5) (Figure 7.4). An
impaction stage comprises an upper horizontal metal partition wall through
which a metal inlet jet tube with its impaction plate is protruding. A glass
cylinder with sampling port forms the vertical wall of the stage and a lower
horizontal metal partition wall through which the tube connects to the next
lower stage. The tube into stage 4 ends in a multi-jet arrangement. The
impaction plate is secured in a metal frame that is fastened by 2 wires to a
sleeve secured on the jet tube. The horizontal face of the collection plate is
perpendicular to the axis of the jet tube and centrally aligned. The upper
surface of the impaction plate is slightly raised above the edge of the metal
frame. A recess around the perimeter of the horizontal partition wall guides the
position of the glass cylinder. The glass cylinders are sealed against the
horizontal partition walls with gaskets and clamped together by 6 bolts. The
sampling ports are sealed by stoppers. The bottom-side of the lower partition
wall of stage 4 has a concentrical protrusion fitted with a rubber O-ring which
seals against the edge of a filter placed in the filter holder. The filter holder is
constructed as a basin with a concentrical recess in which a perforated filter
support is flush-fitted. The filter holder consists of 76 mm diameter filters. The
assembly of the impaction stages is clamped onto the filter holder by 2 snap-
locks. Connect an induction port (Figure 7.5) onto the stage 1 inlet jet tube of
the impinger. A rubber O-ring on the jet tube provides an airtight connection
to the induction port. A suitable mouthpiece adapter is used to provide an
airtight seal between the inhaler and the induction port.

Procedure for Powder Inhalers

Place a suitable low resistance filter capable of quantitatively collecting
the active substance in stage 5 and assemble the apparatus. Connect the
apparatus to a flow system according to the scheme specified in Figure 7.6.
186 Teerapol Srichana

Unless otherwise defined, conduct the test at the flow rate, Qout, used in
the test for determination of the uniformity of delivered dose, drawing 4 L of
air from the mouthpiece of the inhaler and through the apparatus. Connect a
flowmeter to the induction port. Use a flowmeter calibrated for the volumetric
flow leaving the meter or calculate the volumetric flow leaving the meter
(Qout) using the ideal gas law.
Adjust the flow control valve to achieve a steady flow through the system
at the required rate, Qout (±5%). Switch off the pump and ensure that the
critical flow occurs in the flow control valve by the following procedure. With
the inhaler in place and the test flow rate established, measure the absolute
pressure on both sides of the control valve (pressure reading points P2 and P3
in Figure 7.6). A ratio P3/P2 of less than or equal to 0.5 indicates a critical
flow. Switch to a more powerful pump and re-measure the test flow rate if a
critical flow is not indicated. Dispense 20 mL of a solvent, capable of
dissolving the active substance into each of the 4 upper stages of the apparatus
and replace the stoppers. Tilt the apparatus to wet the stoppers, thereby
neutralizing the electrostatic charge. Place a suitable mouthpiece adapter in
position at the end of the induction port.
Prepare the powder inhaler for use according to the patient instructions.
With the pump running and the 2-way solenoid valve closed, locate the
mouthpiece of the inhaler in the mouthpiece adapter. Discharge the powder
into the apparatus by opening the valve for the required time, T (±5%).
Repeat the procedure. The number of discharges should be minimized and
typically would not be greater than 10. The number of discharges is sufficient
to ensure an accurate and precise determination of fine particle dose.
Dismantle the filter stage of the apparatus. Carefully remove the filter and
extract the active substance into an aliquot of solvent. Remove the induction
port and mouthpiece adapter from the apparatus and extract the active
substance into an aliquot of solvent. If necessary, rinse the inside of the inlet
jet tube to stage 1 with solvent, allowing the solvent to flow into the stage.
Extract the active substance from the inner walls and the collection plate for
each of the 4 upper stages of the apparatus into the solution in the respective
stage by carefully tilting and rotating the apparatus, carefully observing that no
liquid transfer occurs between the stages. Using a suitable method of analysis,
determine the amount of active substance present in each of the aliquots of
solvent.
Calculate the fine particle dose (see Calculations).
 In Vitro Quality Control of Dry Powder Inhaler 187

Figure 7.4. Apparatus C: multi-stage liquid impinger. (Adapted from USP35–NF30,

United States Pharmacopeial Convention, 2010).
188 Teerapol Srichana

Figure 7.5. Induction port. Dimensions in millimeters unless otherwise stated.

(Adapted from British Pharmacopoeia, 2010).
 In Vitro Quality Control of Dry Powder Inhaler 189

Figure 7.6. Experimental set up for testing dry powder inhalers. (Adapted from British
Pharmacopoeia, 2010).

7.6.3. Apparatus D - Andersen Cascade Impactor

The Andersen cascade impactor consists of 8 stages together with a final

filter. The material of construction may be aluminum, stainless steel or other
suitable material. The stages are clamped together and sealed with O-rings.
Critical dimensions are applied by the manufacturer of the apparatus D. In use,
some occlusion and wear of the holes will occur. In-use mensuration
tolerances need to be justified. The entry cone of the impactor is connected to
an induction port (Figure 7.7). A suitable mouthpiece adapter is used to
provide an airtight seal between the inhaler and the induction port. In the
configuration for powder inhalers, a pre-separator is placed above the top stage
to collect large masses of non-respirable powder. To accommodate high flow
rates through the impactor, the outlet nipple, used to connect the impactor to
the vacuum system, is enlarged to have an internal diameter of greater than or
equal to 8 mm.

Procedure for Powder Inhalers

The aerodynamic cut-off diameters of the individual stages of this
apparatus are currently not well-established at flow rates other than 28.3 LPM.
190 Teerapol Srichana

Users must justify and validate the use of the impactor in the chosen
conditions when flow rates that are different from 28.3 LPM are selected.
Assemble the Andersen cascade impactor with the pre-separator and a
suitable filter in place and ensure that the system is airtight. Depending on the
product characteristics, the pre-separator may be omitted where justified.
Stages 6 and 7 may also be omitted at high flow rates if justified. The pre-
separator may be coated in the same way as for the plates or may contain 10
mL of a suitable solvent. Connect the apparatus to a flow system according to
the scheme specified in Figure 7.6.

7.6.4. Apparatus E - Next Generation Impactor

Apparatus E is a cascade impactor with 7 stages and a micro-orifice

collector (MOC). Over the flow rate range of 30 LPM to 100 LPM, the 50
percent efficiency cut-off diameters (D50 values) that range between 0.24 μm
to 11.7 μm will be evenly spaced on a logarithmic scale. In this flow range,
there are always at least 5 stages with D50 values between 0.5 μm and 6.5 μm.
The collection efficiency curves for each stage are sharp and minimize overlap
between stages.
The material of construction may be aluminum, stainless steel or other
suitable material. The configuration of the impactor has removable impaction
cups with all the cups in one plane (Figure 7.8). There are 3 main sections to
the impactor; the bottom frame that holds the impaction cups, the sealed body
that holds the jets and the lid that contains the interstage passage ways.
Multiple nozzles are used except at the first stage (Figure 7.9). The flow
passes through the impactor in a saw-tooth pattern.
In routine operation, the sealed body and lid are held together as a single
assembly. The impaction cups are accessible when this assembly is opened at
the end of an inhaler test. The cups are held in a support tray so that all cups
can be removed from the impactor simultaneously by lifting the tray. An
induction port with internal dimensions (relevant to the airflow path), defined
in Figure 7.7, connects to the impactor inlet. A pre-separator can be added
when required, typically with powder inhalers and its connections between the
induction port and the impactor. A suitable mouthpiece adapter is used to
provide an airtight seal between the inhaler and the induction port.
Apparatus E contains a terminal Micro-Orifice Collector (MOC) that for
most formulations will eliminate the need for a final filter as determined by the
method validation. The MOC is an impactor plate with nominally 4032 holes,
 In Vitro Quality Control of Dry Powder Inhaler 191

each approximately 70 μm in diameter. Most particles that are not captured on

stage 7 of the impactor will be captured on the cup surface below the MOC.
For impactors operated at 60 LPM, the MOC is capable of collecting 80% of
0.14 μm particles. For formulations with a significant fraction of particles not
captured by the MOC, there is an optional filter holder that can replace the
MOC or be placed downstream of the MOC (a glass fiber filter is suitable).

Figure 7.7. Apparatus D: Andersen cascade impactor used for DPI. (Adapted from
British Pharmacopoeia, 2010).
192 Teerapol Srichana

Procedure for Powder Inhalers

Assemble the apparatus with the pre-separator. Depending on the product
characteristics, the pre-separator may be omitted where justified. Place cups
into the apertures in the cup tray. Insert the cup tray into the bottom frame and
lower into place. Close the impactor lid with the sealed body attached and
operate the handle to lock the impactor together so that the system is airtight.

Figure 7.8. Apparatus E shown with the pre-separator in place. (Adapted from British
Pharmacopoeia, 2010).
 In Vitro Quality Control of Dry Powder Inhaler 193

When a pre-separator is used, it should be assembled as follows: assemble

the pre-separator by inserting it into the pre-separator base. Fit the pre-
separator base to the impactor inlet. Add 15 mL of solvent for sample recovery
to the central cup of the pre-separator insert. Place the pre-separator body on
top of this assembly and close the 2 catches.

Figure 7.9. Apparatus E and its components. (Adapted from British Pharmacopoeia,
2010).

Connect an induction port that has internal dimensions defined in Figure

7.5 to the impactor inlet or pre-separator inlet. Place a suitable mouthpiece
adapter in position at the end of the induction port so that the mouthpiece end
of the inhaler, when inserted, lines up along the horizontal axis of the
induction port. When attached to the mouthpiece adapter, the inhaler is
positioned in the same orientation as intended for use. Connect the apparatus
to a flow system according to the scheme specified in Figure 7.8.
Prepare the powder inhaler according to the instructions to the patient.
With the pump running and the 2-way solenoid valve closed, locate the
mouthpiece of the inhaler into the mouthpiece adapter. Discharge the powder
into the apparatus by opening the valve for the required time, T (± 5%).
194 Teerapol Srichana

Repeat the discharge sequence. The number of discharges should be

minimized and typically would not be greater than 10. The number of
discharges is sufficient to ensure an accurate and precise determination of fine
particle dose.
Dismantle the apparatus and recover the active substance as follows:
remove the induction port and mouthpiece adapter from the pre-separator,
when used, and recover the deposited active substance into an aliquot of
solvent. When used, remove the pre-separator from the impactor, being careful
to avoid spilling the cup liquid into the impactor. Recover the active substance
from the pre-separator. Open the impactor by releasing the handle and lifting
the lid.
Remove the cup tray along with the collection cups and recover the active
substance in each cup into an aliquot of solvent. Using a suitable method of
analysis, determine the quantity of active substance present in each of the
solvent aliquots.

7.7. Calculations
From the analysis of the solutions, calculate the mass of the active
substance deposited on each stage per discharge and the mass of the active
substance per discharge deposited in the induction port, mouthpiece adapter
and when used, the pre-separator.
Starting at the final collection site (filter or MOC), derive a table of
cumulative mass versus the cut-off diameter of the respective stage. Calculate
by interpolation the mass of the active substance less than 5 μm. This is the
Fine Particle Dose (FPD).
If necessary and where appropriate (e.g., where there is a log-normal
distribution), plot the cumulative fraction of the active substance versus cut-off
the diameter on a log probability paper and use this plot to determine the
values for the Mass Median Aerodynamic Diameter (MMAD) and Geometric
Standard Deviation (GSD) as appropriate (as detailed in Chapter 1).
Appropriate computational methods may also be used.
 In Vitro Quality Control of Dry Powder Inhaler 195

7.8. Other Quality Control of Finished

Products of Dry Powder Inhaler
The products have to be controlled in all batches as follows.

7.8.1. Moisture Content

Moisture control is specified in the finished products. It is important

because it is one parameter to be controlled assure that the product is stable
over its shelf life.

7.8.2. Microbiological Limits

The microbial limit is specified in the specific monograph.

7.8.3. Simulated Patient Use

The simulation of patient use of an inhaler through the device should be

carried out to obtain the data close to real life use. The patient parameters such
as flow rate, inhalation volume and environmental aspects should mimic real
life as close as possible to obtain realizable data. Therefore, the simulated in
vitro system should be close to the in vivo as much as possible.

7.8.4. Reusable vs Disposable Reliability Testing

The quality control of the dry powder inhaler usually consists of assay,
delivered dose, dose uniformity, content uniformity and fine particle fraction.
Other quality controls are also necessary such as moisture content, microbial
limit and number per container (Table 7.1).
The following tests are applicable to metered-dose inhalers that are
formulated as suspensions or solutions of active drug in propellants and dry
powder inhalers presented as single or multidose units. The following test
methods are specific to the aforementioned inhalers and may require
196 Teerapol Srichana

modification when testing alternative inhalation technologies (e.g., breath-

actuated metered-dose inhalers or dose-metering nebulizers). However,
Pharmacopeial requirements for all dose-metering inhalation dosage forms
require determination of the delivered dose and aerodynamic size distribution.
In all cases, and for all tests, prepare and test the inhaler as directed on the
label and the instructions for use.

Table 7.1. Dry powder inhaler quality control

(Adapted from EMA Guideline on the pharmaceutical
quality of inhalation and nasal products, 2006)

Drug Product Specification Device metered Pre-metered

(a) Description Yes Yes
(b) Assay Yes Yes
(c) Moisture content Yes Yes
(d) Mean delivered dose Yes Yes
(e) Delivered dose uniformity Yes Yes
(f) Content uniformity No No
(g) Fine particle mass Yes Yes
(h) Microbial limit Yes Yes
(i) Number per container Yes No

An example of a quality control report of two products based on an

evaluation of MMAD, GSD, FPF and emitted dose (Table 7.2).

Table 7.2. Quality control of two dry powder inhalers (mean ± sd, n= 10)

Items Product A Product B

MMAD (µm) 3.5±0.1 2.1±0.1
GSD 1.9±0.0 1.7±0.0
FPM (µg) 8.9±0.7 10.5±0.6
Emitted dose (µg) 23.5±0.6 20.1±0.6

The MMADs of product A and product B are significantly different

(P<0.01), where the GSDs of the two products are very similar. The FPM and
emitted dose showed trivial differences. It can be explained that the lung
deposition is expected to be different because the MMAD is different resulting
in different depositions in the airway. The smaller MMAD is expected to
penetrate deeper in the lung.
 In Vitro Quality Control of Dry Powder Inhaler 197

References
BP 2010. 2010. British Pharmacopoeia. London, UK: The Stationery Office
Ltd.
EMEA. 2006. Guideline on the pharmaceutical quality of inhalation and nasal
products.
Kleinstreuer, Clement, Zhe Zhang, and James F. Donohue. 2008. “Targeted
drug-aerosol delivery in the human respiratory system.” Annual Review of
Biomedical Engineering 10:195-220.
Ph. Eur. 8th Edition. 2016. “Preparations for Inhalation: Aerodynamic
Assessment of Fine Particles.” In European Pharmacopoeia Strasbourg,
France: Council of Europe.
USP 35-NF 30. 2010. United States Pharmacopeia and National Formulary
(USP35–NF30). Vol. 2, United States Pharmacopeia—National
Formulary. Rockville, MD: United States Pharmacopeial Convention.
 About the Author

Dr. Teerapol Srichana graduated from Prince of Songkla University in

1989 (Bachelor of Pharmacy), hold a Master of Pharmaceutical Science from
the Gent University, Belgium, and a Ph.D. from the King’s College London,
United Kingdom in 1998. Currently he is an Associate Professor in the
Department of Pharmaceutical Technology, Prince of Songkla University
(PSU), Thailand and also serves as the Dean of the Graduate School at the
PSU. He is currently the Director of Drug Delivery System Excellence Center
and NANOTEC-PSU Center of Excellence on Drug Delivery. In his scientific
career, he has served on numerous academic boards and committees, and
published more than 150 research articles, reviews and book chapters, which
are highly cited.
 Index

air friction, 54
A Air turbulence, 140, 145
airborne particles, ix, 19, 56
Accuhaler®, 154
airway epithelial cells, 112
active site, xii
airways, vii, ix, x, xiv, 2, 25, 37, 50, 74, 77,
active sites, xii, xv, 83, 84
99, 139, 140, 142, 162, 165, 182
ActiveMesh™, 147, 166
alanine, 110
actuation, 25, 74, 158
alveolar macrophage, xviii, 114, 115
additives, 93, 122
alveoli, x, xv, 2, 79, 180
Adhesion, 60, 69, 98, 106
ambroxol, 120
adhesion force, 63, 75
amino acid, xii
adhesive, xi, xiii, xiv, xv, 60, 61, 62, 63, 64,
ammonia, 116, 118
66, 69, 75, 76, 83, 88, 98, 100, 101, 102,
ammonium, 67, 115, 116, 118, 119, 126
104
ammonium acetate, 116, 118
Adhesive mixtures, 83
ammonium carbonate, 115, 116, 118
adsorption, 85
amorphous domains, 88
aerodynamic diameter, x, 2, 13, 22, 28, 48,
anti-asthmatic, 93, 172
49, 80, 112
Aerolizer®, 149, 168, 170, 174 antibiotic, 73
aerosolization, xi, 25, 26, 63, 70, 75, 82, Antibounce, 24
100, 101, 111, 112, 121, 122, 124, 127, antisolvent, 92, 94, 114, 117, 120
128, 144, 147, 150, 163, 166, 168, 169 antitrypsin, 96
aerosols, xiii, xiv, xviii, 1, 8, 13, 16, 17, 25, aqueous solutions, 122
32, 33, 38, 40, 58, 80, 81, 83, 84, 85, 98, aqueous suspension, 94
100, 104, 105, 106, 112, 132, 135, 136, AridolTM, 109
139, 159, 166, 167, 168 arteries, x
agglomerates, xiii, xv, 60, 65, 75, 77, 78, aseptic, 90
82, 83, 92, 114, 159, 162, 164, 165, 166, aspergillosis, 112
170 Aspirair®, 167
aggregation, xvi, 78, 80, 91, 98, 115 Asthma, 139
atmospheric pressure, 181
202 Index

atomic force, xii carrier, xi, xiii, xv, xvii, xviii, 69, 74, 75, 76,
atomization, 90, 91, 121 77, 78, 79, 80, 81, 82, 83, 92, 96, 97, 99,
attrition, 88 101, 102, 103, 104, 106, 108, 109, 111,
Autohaler®, 140 112, 113, 114, 115, 118, 119, 124, 127,
128, 130, 131, 132, 135, 150, 151, 154,
157, 163, 168, 170, 171, 174, 175, 180
B cascade impactor, 17, 18, 19, 22, 23, 24, 32,
33, 43, 49, 56, 69, 70, 189, 190, 191
bacterial infection, 125
cavitation, 93
barrier, x
CFD, 25, 43, 166, 168, 170, 171, 172, 174
Bendroflumethiazide, 115
CFD-DEM, 170, 172
bile salts, 80, 112
Chiesi®, 140
binding affinity, 81
chitosan, xvii, 69, 80, 99, 108, 113, 124,
binding sites, 82
126, 127, 128, 131, 132
bioavailability, ix, 31, 89
chloroform, 110
biodegradation, 114, 124
cholesterol, xii, xvi, 108, 110, 111, 129
biopolymer, 127
chronic obstructive pulmonary, x, 109, 128,
blending, 83, 97, 103
151
blends, 97, 104, 136, 152
cilia, ix
blisters, xiv, 97, 152, 158, 163
circulation, x, 73
bounce, 24, 66, 67, 69
Clickhaler®, 149, 166
bovine insulin, 119
clinical trials, 125, 160, 164
bovine serum albumin, 82, 117, 119, 126
CO2, 92, 93, 94, 95, 104, 116, 117, 120
breathing, ix, 50, 144, 146, 161
coacervation, 85, 86, 96, 99
bronchoconstriction, 80
coating, 24, 82, 96, 98, 106, 108, 183
bronchodilator, 79, 150, 153
cohesive, xi, xii, xiv, 69, 75, 76, 77, 82, 88,
Brownian motion, 47, 67, 68
98, 100, 111, 158, 163, 166, 167
budesonide, 26, 27, 80, 92, 93, 95, 105, 112,
cohesive forces, xii, xiv, 77
115, 120, 121, 122, 123, 124, 133, 135,
cohesiveness, 98, 165
154
colistimethate sodium, 148, 160, 175
buoyancy, 44
colistin, 147, 164
collection efficiency, 20, 23, 56, 58, 70, 190
C collision effect, 83
collisions, 35, 86, 87, 166, 169, 170
calibration, 28, 33, 58 computational fluid dynamics, 43, 71, 175,
camptothecin, 119, 133 176
cancer, 125 contamination, 60, 89, 147
candidiasis, 74 coordination, 74
capillary, x, xi, 63 correlation, 17, 18, 25, 110
capsule, xv, 74, 140, 147, 148, 150, 151, corticosteroids, 74, 139
159, 160, 161, 168, 172, 173, 175 coughing, 75, 80, 168
carbohydrate, 125 count median diameter, 8
carbon dioxide, 93, 104, 116, 117, 118 creatinine, 113
cryogenic liquid, 91
crystal growth, 88, 93
 Index 203

crystalline, 79, 93, 94, 95, 109, 111, 121 Diskhaler®, 82, 140, 142, 144, 149, 152,
crystallinity, 80, 86, 98, 109, 110, 124 153, 156
crystallization, 81, 85, 86, 88, 92, 93, 103, Diskus®, 149, 152, 154, 156, 172, 176
105 dispersion, xi, xiii, 74, 78, 79, 83, 94, 97,
crystals, xii, xvii, 92, 93, 101, 102, 104, 98, 101, 125, 128, 129, 140, 144, 145,
111, 124, 129 162, 165, 166, 167, 168, 170, 172, 176
cumulative percentages, 10 displacement, 10, 54, 59
Cunningham correction factor, 46 distribution functions, 6
cut-off diameter, 10, 19, 20, 22, 194 dosage form, x, 75, 98, 107
cyclodextrin, 50, 112, 117 dose emission, 123, 143
Cyclohaler®, 82, 140, 142, 149, 171 dose metering, 75, 144
cyclone separation, 90 DOTAP, 119
cystic fibrosis, 109, 123, 128, 134, 147, 160, double emulsion, 114, 118, 119
164, 175 doxorubicin, 118, 125
drag force, 38, 39, 41, 42, 44, 46, 51, 65
drug aggregates, 77, 78
D drug carrier, 83
drug interaction, 74
de-agglomeration, xiii, xiv, xv, 82, 143, 145, drug release, 113, 114, 118, 125, 145, 150,
151, 163, 164, 166, 168, 169, 170, 172, 163
176
deaggregation, 75, 79, 83, 144, 145, 154
decomposition, 88, 116, 161 E
deficiency, 31
deformation, 62, 66 Easyhaler®, 140, 149, 157, 175
degradation, 85, 86, 89, 114 electrical force, 38, 54
deposition, x, xii, xiii, xiv, xv, 13, 14, 17, electromagnetic beam, 15
19, 24, 26, 31, 32, 33, 37, 50, 75, 76, 79, electrostatic charges, 63, 80, 84, 88
80, 83, 95, 96, 98, 102, 104, 106, 112, elongation, 81
115, 118, 122, 123, 125, 131, 132, 135, emitted dose, xiv, 75, 77, 79, 100, 160, 161,
139, 143, 144, 147, 148, 157, 162, 165, 163, 165, 166, 196
167, 168, 170, 173, 174, 175, 176, 196 encapsulation, 108, 110
depth, 13 entrapment, 119
derivative, 7, 139 epithelium, x, 112
detachment, xii, xiv, 65, 70, 83, 166, 171 equilibrium, xii, 52, 63, 83, 98
detection, 14, 15 equivalent volume diameter, 47
device design, vii, 77, 141, 146, 166, 175 erosion, 66, 85
device resistance, xiv, 74, 141, 146, 148, evaporation, 31, 88, 91, 118, 119, 120
166, 176 Exubera®, 108, 158, 159, 167
dialysis, 96
differential equations, 40 F
diffusion coefficient, 68
dipoles, 61 fatty acids, 95
direct imaging, 13 fine particle fraction, xiv, 18, 32, 75, 115,
discrete element method, 170 117, 183, 195
204 Index

flow field, 25, 43, 54, 69 hydrodynamic, 46

flow rate, xiii, xiv, xv, 18, 23, 24, 25, 37, hydrophilic material, 121
38, 63, 74, 75, 112, 141, 142, 144, 146, hydroxypropyl cellulose, 125
148, 149, 154, 156, 160, 162, 167, 170, hygroscopic, 98, 109
175, 181, 183, 186, 190, 195
flowability, 77, 80, 122
fluid motion, 40 I
Food and Drug Administration, 107
freezing, 92 IL-8, 114
frequency distribution curve, 7 impact energy, 170
friction, 41 impaction, x, xiv, 13, 20, 21, 22, 23, 24, 33,
55, 56, 58, 60, 65, 67, 88, 142, 143, 145,
146, 150, 151, 166, 169, 170, 171, 172,
G 176, 185, 190
impaction plate, 20, 22, 23, 24, 56, 58, 185
gamma scintigraphy, 31 impinger, 17, 18, 26, 27, 32, 33, 70, 95, 104,
geometric mean, 7, 8, 12, 13 183, 184, 185, 187
geometric mean diameter, 8, 12, 13 incompressible fluid, 41, 42
geometric standard deviation, 8, 12 inertial force, 36
geometry, xv, 24, 58, 62, 67, 173 inertial impaction, x, 88
glucose, xii, 77, 109, 117 inertial sampling, 19
glycerol, 183 Inhalator®, 140
glycine, 110 inspiratory flow, xiii, xiv, xv, 74, 75, 79,
gravity, 38, 40, 44, 45, 51 144, 149, 154, 157, 163, 166, 172, 175
grid, 151, 166, 168 insulin, 92, 96, 101, 105, 112, 131, 132,
grooves, 82, 113, 127 136, 158, 162, 167
growth mechanisms, 93 interaction, vii, xi, 35, 83, 93, 112, 122, 141,
growth retardants, 93 144, 152, 169
interferon-β, 110
H interparticle, 46, 77, 97
interparticle forces, 77, 97
half-life, 116 interstage loss, 24, 26
Handihaler®, 125, 149, 151, 166, 176
Hatch-Choate, 12, 13 J
heparin, 116, 119, 134
heparin-PLGA-polyethylene imine, 119 jet mill, 86, 87
homogeneity, 82, 83, 89, 97
homogeneous, xi, xii, 83, 96
K
HP-β-CD, 117, 118, 120, 132
humidity, 84, 90, 98, 100, 156 kinetic energy, xiv, 66, 67, 68, 70, 71, 172,
hydrocortisone, 95 176
 Index 205

mixing, xii, 81, 82, 84, 90, 92, 93, 94, 95,
L 96, 97, 118, 128
mobility, 45, 51, 52, 53, 54
lactose, xi, xii, 28, 74, 76, 77, 82, 83, 92,
moisture, 80, 91, 98, 140, 154, 156, 160,
101, 104, 106, 109, 113, 122, 125, 132,
195
136, 144, 150, 151, 169, 171, 176
moisture content, 91, 98, 154, 195
laminar, 36, 40, 43, 65
molecules, 35, 47, 60, 61, 62, 63, 67, 68, 80,
laninamivir, 147
93, 95, 96, 110
large porous particles, 114, 116, 119, 136
molten lipid, 124
L-arginine, 111
momentum, 19, 39, 54
laser diffraction, 15, 16, 17, 33 monodisperse aerosol, 1, 10
lauric acid, 95 monolayer, 65, 82
lecithin, 75 mouthpiece, xv, 143, 145, 148, 151, 154,
Leucine, 108, 111, 134 157, 158, 159, 160, 161, 162, 166, 168,
lipids, 124 170, 173, 181, 184, 185, 186, 189, 190,
liposomes, xvii, 108, 110, 124, 129, 136 193, 194
liquid crystal, 111 mucosa, 179
liquid phase, 121 mucus clearance, 109
log-normal distribution, 7, 8, 9, 11, 12, 194 multiple dose, xiv, 140, 147, 161
log-probability plot, 9, 10, 11, 12 multiple unit dose, 140, 148
loose aggregates, 77 Multistage Liquid Impinger, 26
lung deposition, 13, 98, 162
lyophilization, 86, 92, 96, 112, 120
N

M nano-in-microparticles, 125
nanoparticle, 89, 130, 134
macropores, 82, 115 nanoporous, 114, 115, 128, 132, 133
magnesium stearate, xv, 75, 82, 83, 106 Navier-Stokes equations, 40, 41, 42, 58
magnetic field, 125 nebulizers, 139, 195
Magnetically Targeted, 125 NEXThaler®, 149, 158, 172, 177
mannitol, xii, xviii, 14, 77, 109, 110, 115, Nonspherical Particles, 47
116, 118, 128, 134, 135, 151, 176 normal distribution, 7, 8, 9, 11, 12
Mass distribution, 5 nozzle, 20, 21, 22, 23, 28, 56, 58, 60, 70, 90,
mean free path, 35, 46, 47 91, 94
mechanical impaction, 145 NPMP, 121
micronization, 88, 103, 104, 165 nucleation, 93
micronized drug, x, xi, xiii, 77, 81, 88, 106, number distribution, 12, 13
111, 124, 154, 157 numerical analysis, 58, 168
microparticles, xvi, 50, 70, 113, 114, 115,
120, 122, 125, 127, 129, 130, 132, 133,
135, 137 O
micropores, 82, 115
microscope, 14 open cycle, 90
microsphere, 114 optical microscopy, 11
milling, 16, 85, 86, 87, 88, 110 optical properties, 15
206 Index

optical sizing, 13 pressure drop, xv, 37, 38, 141, 142, 144,
optimization, xv, 95, 97, 133 154, 170, 181
oral cavity, 140, 143, 162, 168 probability plot, 9
oropharyngeal regions, 79 probits, 10
oropharynx, ix, 18, 79, 167 projected area, 2, 3, 29, 39
osmotic pressure, 68, 119 proportionality, 39, 51, 63, 68
prostaglandin E1, 118, 119, 131
proteolytic enzymes, ix
P prototype device, 166
Pseudomonas aeruginosa, 160
partial differential equations, 40
purity, xv, 93, 109
Particle dispersion, 169
particle–device, 145, 170
particle–particle, xi, 86, 87, 122, 145, 170 Q
PEG, 96
penetration, 13 quality control, vii, xvi, 17, 18, 179, 195,
peptides, 101, 113, 114, 118, 128, 164 196
permeability, ix quench solvent, 92
phagocytosis, xv, 114
pharmacokinetic, 31 R
phenylalanine, 111, 167
Phosphatidylcholine, 110 recrystallization, 93, 98
phospholipids, xii, 80, 127, 131 red blood cells, 113
plasticization, 94 Re-entrainment, 65, 183
PLGA, xvi, 80, 108, 114, 115, 116, 117,
refractive index, 15
118, 119, 120, 130, 131, 132, 137
relative humidity, 60, 64, 84, 98
Podhaler®, 148
relaxation time, 51, 55
poly(lactic acid-co-lysine-graft-lysine), 119 reservoir, xiv, xv, 24, 95, 148, 154, 157,
polydispersity, 15, 176 158, 172, 180
polyethylene glycol, 96 resistance, xv, 36, 38, 39, 41, 45, 141, 142,
polylactide glycolide, 119 143, 146, 154, 159, 162, 175, 185
polymers, xii, 94, 96, 124, 125 respirable size, 85, 98, 107
polysaccharides, 113 Reynolds number, 19, 35, 36, 37, 39, 40, 41,
polystyrene, 125, 126 50
polyvinyl acetate, 120 Rosin-Rammler, 8
pore forming agent, 115, 118, 119
Rotahaler®, 82, 140, 142, 144, 149, 150,
pore-formation, 115 151, 175
porosity, xv, 49, 82, 104, 115, 116, 119 rotameter, 38, 141
porous microparticles, 115, 120 roughness, 60, 61, 66, 69, 70, 76, 80, 82
porous particle, 116
portability, 168
powder dispersion, 145, 167, 170 S
powder technology, 77, 81
precipitation, 89, 92, 94, 104, 124 sedimentation, x
preseparator, 24, 32, 43, 49, 69 segregation, 83
 Index 207

SEM micrographs, 14, 125 surface tension, 60, 63

separation, 56, 61, 62, 63, 67, 77, 80, 82, 83, surface treatment, 76
85, 90, 96, 150 suspensions, 1, 89, 110, 124, 195
separation forces, 83
settling velocity, 2, 13, 44, 45, 46, 47, 48,
49, 52 T
shear force, 88, 172
TEM, 125
simulation, 33, 42, 195
tensile strength, 170
single dose, xiv, 130, 147, 154, 172
thermal decomposition, 161
skewed distribution, 7, 8
thermal degradation, 89
slip correction factor, 45, 46, 47
tobramycin, 123, 131, 133, 148, 160
sodium cromoglycate, 95, 120, 121, 123,
trachea, ix, 180
139, 144, 149, 173
trajectory, 54, 166
solid bridge, 98
transformation, 13, 15, 101
solubility, xviii, 80, 93, 94, 112, 121, 124,
Trehalose, 108, 109
130
tuberculosis, 113
solvent evaporation, 119
Turbospin®, 123, 148, 149
specific resistance, 141
Turbuhaler®, 140, 142, 143, 149, 153, 154,
spheroids, 77
155, 172
spheronization, 154
turbulence, 43, 69, 143, 146, 160, 166, 168,
Spinhaler®, 140, 142, 144, 148, 149, 150
170
SPION, 125
turbulent, xv, 36, 65, 70, 79, 95, 146, 148,
spray drying, 85, 88, 90, 92, 105, 109, 115,
154, 172, 176
120, 121, 125, 126
Turbulent flow, 37
spray freezing, 92
TwinCaps®, 147
sprayed droplets, 90
Staccato®, 161, 167, 174
static charge, 77 U
stearic acid, 95
Stokes diameter, 48 ultrasonic, x, 91, 93, 176
Stokes number, 20, 21, 54, 55, 56, 58 upper airways, 2
stopping distance, 54, 55, 56 upper respiratory tract, 145
sublimation, 91, 161
supercritical CO2, 92
supercritical fluid, 85, 86, 92, 93, 103, 114, V
124
Superparamagnetic iron oxide van der Waals, xi, xiii, 60, 61, 62, 63, 127
nanoparticles, 125, 135 viscous forces, 36, 37, 40
supersaturation, 93 visualization, 43
surface area, ix, x, xiii, 13, 83, 89, 92, 93,
115, 116
W
surface energy, xi, 151, 165, 173
surface free energy, xi, 75 weight distribution, 13
surface porosity, 82
surface properties, xii, xv, 81
208 Index

X Z

x-axis, 5, 54 zanamivir, 147

Z-score, 10

yield, 9, 64, 92, 93