pubs.acs.

org/JPCB Article

Effect of Confinement on the Translational Diffusivity of Small Dye

Molecules in Thin Polystyrene Films and Its Connection to
Tg‑Confinement and Fragility-Confinement Effects
Published as part of The Journal of Physical Chemistry B special issue “Mark Ediger Festschrift”.
Tian Lan, Tong Wei, Logan M. Fenimore, and John M. Torkelson*
Cite This: J. Phys. Chem. B 2024, 128, 12259−12267 Read Online
See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

ACCESS Metrics & More Article Recommendations *

sı Supporting Information
Downloaded via BOGAZICI UNIV on December 28, 2024 at 13:01:04 (UTC).

ABSTRACT: Using fluorescence, we study the impact of

nanoscale confinement on the translational diffusivity (D) of
trace levels of a small-molecule dye, 9,10-bis(phenylethynyl)-
anthracene (BPEA), in supported polystyrene (PS) films via
Förster resonance energy transfer (FRET). Reductions in BPEA
diffusivity are observed in films thinner than ∼200 nm, with D
decreasing by 80−90% in 100 nm-thick films compared to bulk.
The activation energy of BPEA diffusivity increases from ∼210 kJ/
mol in bulk films to ∼370 kJ/mol in 130 nm-thick films. BPEA
exhibits a greater diffusivity-confinement effect than a larger dye,
decacyclene, in terms of the length scale at which the effects of
confinement become evident and the percentage reduction in
diffusivity. For both BPEA and decacyclene, the diffusivity-
confinement effect in supported PS films occurs at a length scale much larger than that for the glass transition temperature
(Tg)-confinement effect and somewhat larger than that for the fragility-confinement effect. This difference in confinement-effect
length scales can be rationalized as follows: small-molecule dye diffusivity relates predominantly to short times in the α-relaxation
distribution, whereas Tg relates to long times in the α-relaxation distribution, and fragility reflects the overall breadth of this
relaxation time distribution. If confinement results in a narrower relaxation time distribution in PS films with the short-time
relaxations being shifted to longer times and the longest-time relaxation regimes being shifted to shorter times, then Tg, diffusivity,
and fragility all decrease at sufficient levels of confinement. If the narrowing with confinement begins with the shortest relaxation
time regimes, then fragility and small-molecule dye diffusivity are influenced by confinement at larger length scales than Tg.

■ INTRODUCTION
Nanostructured and nanoconfined polymers are utilized in a
properties (e.g., Tg, modulus, and acid diffusivity) of
photoresists is important for obtaining sharp and robust
wide range of technologies, including lithography,1,2 mem- patterns in photolithography,15,33 as well as for other
brane separation,3 organic photovoltaics,4,5 and biomedical applications of polymers close to the nanoscale.
Compared with the well-documented Tg-confinement effect,
devices.6,7 Numerous studies have reported that confinement
limited studies have been reported on the effect of polymer
of polymers at or near the nanoscale can lead to large
confinement on translational diffusivity. Studies of the
deviations from bulk response and behavior, including the glass
diffusivity-confinement effect have largely focused on the
transition temperature (Tg),8−29 breadth of the distribution of
diffusion of polymer chains at temperatures well above Tg. For
α-relaxation times,30 mechanical properties,31−34 phase behav-
example, Frank et al.48 studied the in-plane self-diffusion of PS
ior,35 physical aging,36 segmental dynamics,37 fragility,38−47
chains at 413 K by fluorescence recovery after photobleaching
and diffusivity.48−52 Regarding the effect of nanoscale confine-
(FRAP). Relative to bulk response, they observed a two-fold
ment on the Tg in polymers, reductions in Tg of several degrees
to several tens of degrees have commonly been reported in
sufficiently thin, freestanding films and supported films of Received: September 26, 2024
linear polymers that lack substantial, attractive substrate Revised: November 22, 2024
interactions.8−29,58−60 In lithography, thin polymer films are Accepted: November 26, 2024
used as photoresists for creating nanoscale patterns of Published: December 4, 2024
electronic circuits via a photoacid-catalyzed deprotection
reaction.2,61,62 Understanding how confinement affects the

© 2024 American Chemical Society https://doi.org/10.1021/acs.jpcb.4c06495

12259 J. Phys. Chem. B 2024, 128, 12259−12267
The Journal of Physical Chemistry B pubs.acs.org/JPCB Article

diffusivity reduction in a 60 nm-thick film. Zheng et al.49 investigated the rotational and translational diffusion of
studied the through-plane diffusion of deuterated polystyrene Disperse Red 1 and 4-(dimethylamino)-4′-nitrostilbene in PS
(dPS) in a hydrogenated polystyrene (hPS) matrix by neutron films. They found that in the rubbery state, with 1.1 Tg > T >
scattering. They observed an order-of-magnitude reduction in Tg, dye translational diffusion was much less T-dependent than
the diffusion rate when a dPS tracer layer was placed up to 10 rotational reorientation; however, in a “quenched” glass,
radii of gyration from the polymer−substrate interface. translational and rotational diffusion exhibited similar T
Napolitano et al.50 found via dielectric relaxation spectroscopy dependences. Hall et al. indicated that if the translational
(DRS) that the through-plane diffusivity of a polymeric tracer diffusion of small molecules is reflected by ⟨τ−1⟩ and rotational
decreased to 1% of its bulk value in a 7.5 nm-thick PS film. In diffusion is reflected by ⟨τ⟩−1, where τ is the rotational
contrast, Katzenstein et al.51 investigated the in-plane transla- relaxation time, which is significantly coupled to the α-
tional diffusivity of poly(isobutyl methacrylate) (PiBMA) via relaxation of the polymer matrix,64,74 then “the enhancement
FRAP and observed no change in the self-diffusion coefficient of translation relative to rotation is a consequence of the
in PiBMA films with confinement. Winey and Composto and breadth of the relaxation time distribution and should be
co-workers studied through-plane chain diffusion in polymer− observed at any temperature where the rotational relaxation
nanoparticle composites.52−54 They found that the diffusivity distribution is sufficiently broad that ⟨τ−1⟩ and ⟨τ⟩−1 differ
of PS tracers decreased strongly at a sufficiently small greatly.”74
interparticle spacing. Fragility is a parameter that reflects the packing efficiency of
Torkelson and co-workers55,63−66 used fluorescence to study polymer chains76−78 and relates to the breadth of the α-
the through-plane diffusion of two small-molecule dyes, relaxation distribution.79 The fragility index, m, can be
decacyclene and lophine, in supported, bilayer thin polymer expressed by the following:38−45,67−80
films by Förster resonance energy transfer (FRET), also called
nonradiative energy transfer or NRET, at temperatures close to dlog
m= T = Tg
Tg. They found that the dye-molecule diffusivity decreased dTg /T (1)
with decreasing supported PS film thickness below ∼150
nm.55,64 Notably, there is no significant Tg-confinement effect where τα is the α-relaxation time. Confinement at or near the
in supported PS films until the thickness decreases below ∼50 nanoscale can result in major reductions in m in PS. A range of
nm, and then, Tg decreases with yet greater confine- experimental studies have indicated that the fragility of bulk
ment.10,22−24 These findings make evident the invalid nature PS, mbulk, is ∼150 ± 20.39,42,46,47,78,79,81−83 Torkelson and co-
of the simple, some might call naive, ̈ argument that workers reported that supported PS films exhibit significant
confinement-induced Tg reductions imply an increase in reductions in m from mbulk at thicknesses below ∼100 nm, with
polymer matrix mobility that must lead to an increase in dye reductions by roughly one-half at thicknesses of 20−30
diffusivity. Further supporting the invalid nature of a simple nm.39,46,47 Fukao and Miyamoto40 also reported that m of
connection between Tg-confinement and dye diffusivity- supported PS films decreases substantially with decreasing
confinement effects, they studied dye diffusivity in supported nanoscale thickness. Lan and Torkelson additionally compared
PiBMA and poly(2-vinylpyridine) (P2VP) films55,63,64 and m in confined films of PS, PS doped with 2 wt % of a small-
obtained thickness-independent dye diffusivities even though molecule diluent called 1,10-bis(1-pyrene)decane (BPD), and
PiBMA and P2VP undergo attractive interactions with densely grafted PS brushes. Relative to neat PS, the fragility-
hydroxyl groups on substrate surfaces that can lead to confinement effect was somewhat muted in PS doped with 2
increases in Tg with nanoconfinement.22,28 wt % BPD and eliminated within experimental uncertainty in
Vogt et al.56 studied the through-plane diffusion of water densely grafted PS brushes.39
molecules in thin films of poly(4-ammonium styrenesulfonic Here, we investigate the impact of confinement on the
acid). They reported that the diffusivity of water molecules was translational diffusivity of trace levels (≤0.3 wt %) of the small-
2.7 × 10−11 cm2/s in a 210 nm-thick film but decreased by 4 molecule dye 9,10-bis(phenylethynyl)anthracene (BPEA) in
orders of magnitude in a 3 nm-thick film. In contrast, Tseng et supported thin PS films with bilayer and trilayer geometries.
al.57 reported major increases in the in-plane diffusivity of The through-plane diffusivity of BPEA decreases with
rubrene in confined PS films via FRAP. However, their films decreasing thickness in films thinner than ∼200 nm; relative
(doped with rubrene) were previously annealed in a vacuum to bulk behavior, 80−90% reductions in diffusivity are
oven at 423 K for 2 h, which could cause severe segregation of observed in 100 nm-thick films. The apparent activation
rubrene. Moreover, they reported strikingly large Tg reductions energy for BPEA diffusivity is significantly increased in going
(∼25 K) in 200- to 300 nm-thick films, whereas other studies from bulk films to 130 nm-thick films. We put forward a
have not reported Tg reductions in PS films at similar hypothesis to explain the connection among the diffusivity-,
thicknesses.8−10,12,17 Tg-, and fragility-confinement effects. It is well-known that
Interestingly, researchers have reported an apparent there is major breadth associated with the cooperative
violation of Stokes−Einstein−Debye scaling for translational segmental relaxation regimes in near-Tg polymers;30,74,82,84
and rotational diffusion of small-molecule dyes in bulk, glass- additionally, there is a well-accepted correlation between the
forming systems; this behavior is sometimes called the breadth of the relaxation distribution and fragility for a large
translation−rotation paradox.64,67−74 Ediger and co-work- number of glass-forming liquids, including polymers.79,82 We
ers72,75 reported that the translational diffusion coefficients of hypothesize that the value of the small-molecule dye diffusivity
tetracene, BPEA, and rubrene in bulk PS are significantly is predominantly influenced by the short-time relaxation
enhanced compared to those of rotational diffusion as Tg is regimes inside PS, that the Tg is mainly influenced by the
approached from above, i.e., the dye translational diffusivity has long-time relaxation regimes, and that the fragility reflects the
a significantly weaker temperature (T) dependence than the overall breadth of the relaxation distribution associated with
corresponding rotational correlation times. Hall et al.74 cooperative segmental mobility. A significantly reduced
12260 https://doi.org/10.1021/acs.jpcb.4c06495
J. Phys. Chem. B 2024, 128, 12259−12267
The Journal of Physical Chemistry B pubs.acs.org/JPCB Article

fragility value indicates a narrower α-relaxation distribution in same thickness) was measured via FRET. As shown in Figure
confined, supported PS films. Such a narrowing with 1, the FRET acceptor layer is made of PS doped with a nearly
confinement, with a shift in the short-time side of the
relaxation distribution to substantially longer times and the
long-time side to somewhat shorter times, is consistent with
both the translational diffusivity of the small-molecule dye and
the PS Tg being reduced with sufficient confinement. If the
narrowing with confinement starts with the very shortest
relaxation-time regime, then the small-molecule dye diffusivity
and fragility can exhibit confinement effects at a greater length
scale than Tg.

■ MATERIALS AND METHODS

An anionically polymerized PS standard with a nominal
molecular weight of 400 kg/mol (Tg,onset = 375 K, dispersity <
1.10) was purchased from Pressure Chemical. Pyrene-labeled
PS (Tg,onset = 375 K by differential scanning calorimetry, Mn =
230 kg/mol, dispersity = 1.69) was synthesized by adding trace
levels of 1-pyrenylbutyl methacrylate to the styrene monomer
before synthesis by bulk free-radical polymerization, which was
initiated by benzoyl peroxide. After synthesis, the product was
washed seven times by dissolving in toluene and precipitating
in methanol to remove the unreacted monomer. As
determined by UV−vis absorbance spectroscopy (PerkinElmer
Lambda 35), the pyrene label content was 0.7 mol %.10
For FRET measurements done in a bilayer geometry, a
FRET donor layer was prepared by co-dissolving pyrene-
labeled PS and neat, 400 kg/mol PS standard at a ratio of 1:3
(w/w) in toluene followed by spin coating onto mica. The
FRET donor layer was annealed at 393 K for 2 h and then
floated off mica in a deionized water bath.10 The FRET donor
layer was then layered on top of a film of PS (without the Figure 1. (a) Chemical structures of BPEA (tracer and the FRET
pyrene label), which was doped with 0.3 wt % BPEA and spun acceptor) and 1-pyrenylbutyl methacrylate (the FRET donor when
covalently attached to PS). (b) Normalized steady-state fluorescence
cast on a silica wafer. The bilayer films were left at room spectra of pyrene-labeled PS in bilayer, pyrene-labeled PS/BPEA-
temperature for at least 12 h for water evaporation before doped PS samples (the total thickness is 380 nm with two layers of
being mounted onto a T-controlled sample stage (Instec) for equal thicknesses) annealed at 378 K for t = 0 min (solid curve) and
diffusion measurement at high T. 43 min (dashed curve). Inset: the bilayer film geometry used for
To measure BPEA diffusion in a trilayer geometry, a 30 nm- FRET measurement and illustration of the BPEA concentration
thick PS film doped with 0.3 wt % BPEA was first spun cast profile before (solid curve) and after (dashed curve) diffusion started.
onto a silica wafer. A neat PS film was spun cast onto mica and
then transferred onto the BPEA-doped PS film via floating.
The layered film was left at room T for at least 12 h before a trace level (0.3 wt %) of BPEA, and the FRET donor layer is
third layer of 30 nm-thick, pyrene-labeled PS film was made of pyrene-labeled PS (with 0.7 mol % pyrene label
transferred on top. The trilayer films were held at room T content relative to styrene repeat units). During annealing at a
for at least 12 h to remove residual water before being temperature near the Tg of PS, the BPEA acceptor diffuses into
mounted onto a T-controlled sample stage (Instec) for the FRET donor layer, resulting in a net decrease in the pyrene
diffusion measurement at high T. fluorescence intensity.
Steady-state fluorescence intensity measurements were For our experiments in which a PS layer doped with a nearly
performed using a fluorimeter (Photon Technology Interna- trace level of a small-molecule dye, BPEA, is adjacent to a
tional) equipped with a T-controlled sample stage (Instec). second PS layer labeled with a nearly trace level of another dye,
Samples were excited at 324 nm for pyrene absorbance; the the through-plane diffusion of BPEA in an essentially static PS
fluorescence emission was monitored from 375 to 380 nm as a matrix is expected to follow Fickian behavior. The static nature
function of time. The translational diffusion coefficient was of the PS matrix is clear because polymer self-diffusion
determined using the time dependence of integrated intensity. coefficients are many orders of magnitude smaller than the
The experimental timescales for diffusion measurement were diffusion coefficients of small-molecule dyes.42,85,86 If one
∼8 to ∼240 min, depending on temperature and film assumes that the diffusion of BPEA is Fickian and the diffusion
thickness. time is sufficiently small to satisfy a semi-infinite slab model,

■ RESULTS AND DISCUSSION

Using a method developed by Torkelson and co-work-
then the concentrations of FRET donor pyrene (CD) and
FRET acceptor BPEA (CA) are66,73,74
CDO for w<z<0
ers,63−66,73,74 the translational diffusivity of BPEA in bilayer C D(z , t ) = {
PS films (with FRET acceptor and FRET donor layers of the 0 for 0 < z < w (2)

12261 https://doi.org/10.1021/acs.jpcb.4c06495
J. Phys. Chem. B 2024, 128, 12259−12267
The Journal of Physical Chemistry B pubs.acs.org/JPCB Article

CA0 z
CA(z , t ) = [1 + erf( )]
2 2 Dt (3)
Here, t is diffusion/annealing time, w is the thickness of the
donor layer (also the thickness of the acceptor layer), and D is
the translational diffusivity of BPEA. The solution for CA(z,t) is
limited to short diffusion times (t < w2/(16D)). Equations 2
and 3 can be combined with the mathematical description of
donor fluorescence intensity decay in the presence of acceptor
chromophores based on the FRET process and integrated over
the entire sample thickness:66,73
1 w C D(z , t ) t t
IDN( t , t ) = exp[ ( ) 2 ( )1/2 ]dz
w w CDO D D
(4)
where t ̅ is decay time, IDN(t,̅ t) is the donor intensity decay
profile normalized to a value of 1 at t ̅ = 0, τD is the donor
lifetime in the absence of acceptors, and γ is a function of
diffusion time and defined as CA/AO, where AO is
3000
AO =
2NAR O3 3/2 (5)
Here, RO is the Förster resonance energy transfer radius in cm
(3.53 × 10−7 cm for the FRET donor−acceptor pair of pyrene
and BPEA66) and NA is Avogadro’s number. The energy
transfer efficiency, E(t), is73
t t
I (t ) IDN( , t )d( )
0 D D Figure 2. (a) E(t) vs (K/w)t1/2 for BPEA diffusion in (squares) 800
E(t ) = 1 =1 t t
I(0) IDN( , 0)d( ) nm-thick and (triangles) 130 nm-thick, bilayer PS films (with two
0 D D (6) individual thicknesses of 400 or 65 nm, respectively), measured at 378
By substituting eqs 2, 3, and 5 into eq 4 and then into eq 6, K (Tg + 3 K). (b) E(t) vs (K/w)t1/2 for BPEA diffusion in (squares)
800 nm-thick and (triangles) 130 nm-thick, bilayer PS films (with two
we have65,66,73
individual thicknesses of 400 or 65 nm, respectively), measured at 373
Dt w /(2 Dt ) K (Tg − 2 K).
E(t ) = ( ){2 a exp(a 2)[1 erf(a)]
w 0

dy]}
Dt
= K
w (7)
CAO z
where a = 2A o
[1 erf(y)] and y = 2 Dt
. For the conditions
used in this study, K = 0.632. By plotting E(t) as a function of
(K/w)t1/2, the slope is D1/2.
Figure 2 shows that the BPEA diffusion coefficient can be
determined from the time-dependent fluorescence intensity of
pyrene in bilayer PS films with overall thicknesses of 130 or
800 nm. The E(t) data exhibit linearity with (K/w)t1/2 at
temperatures above and below the Tg of PS; thus, the
translational diffusivity of BPEA exhibits Fickian behavior in
both the glassy and rubbery states of PS. Given that the slope Figure 3. Diffusivity of BPEA in PS films as a function of film
of each data set equals D1/2, the diffusivity of BPEA clearly thickness. Measurements were done at (triangles) 378 K (Tg + 3 K),
decreases with decreasing film thickness (Figure 2). The (circles) 373 K (Tg − 2 K), and (diamonds) 368 K (Tg − 7 K).
average D measured from multiple samples of 800 nm-thick
films is 1.2 ± 0.3 × 10−14 cm2/s at 378 K (Tg + 3 K), in good
agreement with previous reports on the BPEA diffusion (Tg + 3 K), T = 373 K (Tg − 2 K), and T = 368 K (Tg − 7 K).
coefficient in bulk PS films, with D ≈ 1.0 × 10−14 cm2/s In 800 nm-thick, bulk-like PS films, the BPEA diffusivity is 1.2
measured by FRET (at Tg)65 and D ≈ 1.6 × 10−14 cm2/s × 10−14 cm2/s at Tg + 3 K, 6.6 × 10−15 cm2/s at Tg − 2 K, and
measured by FRAP (at Tg).72 In contrast, the 130 nm-thick PS 1.8 × 10−15 cm2/s at Tg − 7 K. Major reductions in diffusivity
films report an average D of 2.8 ± 0.4 × 10−15 cm2/s, roughly a occur with sufficient confinement at all three temperatures. For
factor of five to six reduction relative to bulk PS. example, the diffusivity of BPEA in 100 nm-thick PS films is
Figure 3 shows the BPEA diffusion coefficient as a function 2.0 × 10−15 cm2/s at Tg + 3 K, 4.4 × 10−16 cm2/s at Tg − 2 K,
of film thickness at three annealing temperatures: T = 378 K and 2.4 × 10−16 cm2/s at Tg − 7 K. These respective values are
12262 https://doi.org/10.1021/acs.jpcb.4c06495
J. Phys. Chem. B 2024, 128, 12259−12267
The Journal of Physical Chemistry B pubs.acs.org/JPCB Article

reduced by approximately 80−90% compared to those in 800 breakthrough time (tbreakthrough), which indicates how long it
nm-thick, bulk-like PS films. Clearly, the diffusivity-confine- takes the diffusion front of BPEA to reach the top layer at z =
ment effect of BPEA is much more dramatic than that of PS −d, where the interface between the top and barrier layers is
chain diffusivity, which was reported by Frank et al.48 located. With the range of film thicknesses applied in the
As shown in Figure 4, we also measured the translational current study (0.08 < w/d < 0.3), D can be determined as
diffusivity of BPEA by determining its breakthrough time in a follows (see the Supporting Information for more details):
0.045d2
D
tbreakthrough (9)
In Figure 5, tbreakthrough ≈ 4800 s for the diffusion of BPEA in
a 360 nm-thick film (d = 300 nm), whereas tbreakthrough ≈ 1300 s
for the diffusion of BPEA in a 160 nm-thick film (d = 100 nm).
Table 1 compares the diffusivity of BPEA measured in bilayer

Table 1. Translational Diffusivity of BPEA in PS Films

Measured by Bilayer or Trilayer Methods
Figure 4. Mathematical representation of the trilayer film used to
total diffusivityb (bilayer method, diffusivityb (trilayer method,
determine the tracer diffusivity. thickness 10−14 cm2/s) 10−14 cm2/s)
360 or 380 1.03 ± 0.36 1.04 ± 0.20
trilayer film geometry (supported on silica). In these trilayer nma
films, a 30 nm-thick PS layer doped with 0.3 wt % BPEA 160 nm 0.39 ± 0.18 0.38 ± 0.10
(FRET acceptor) is spun cast on a silica substrate followed by a
Bilayer measurements were done in layered films with a total
the deposition of a neat barrier layer with thickness d and a thickness of 380 nm, with both the FRET donor layer and FRET
third layer of 30 nm-thick, pyrene-labeled PS (FRET donor) acceptor layer of 190 nm. Trilayer measurements were done in
film. At t = 0, all the BPEA molecules are contained in the layered films with a total thickness of 360 nm, with both the FRET
bottom layer (0 < z < w, with concentration of CAO and donor layer and FRET acceptor layer being 30 nm and a 300 nm-thick
thickness of w). At any time during the measurement that the neat PS layer in between. bAverage values calculated from two or
concentration of BPEA is negligible in the top layer to satisfy a three samples with the same thicknesses at 378 K (Tg + 3 K).
semi-infinite slab model (CA(z) ≈ 0 for any z < −d), CA(z)
inside the trilayer film is given by the following (see the and trilayer PS films (T = 378 K), with two to four repeats for
Supporting Information for more details):87 each sample. The diffusivity of BPEA is 1.0 × 10−14 cm2/s as
Ä É measured in both 360 nm-thick trilayer and 380 nm-thick
C ÅÅÅ i 2w z yz i z yzÑÑÑÑ
CA(z) = AO ÅÅÅÅerfjjjj zz + erfjjj zzÑÑ bilayer PS films and reduces to a value of 0.4 × 10−14 cm2/s as
2 ÅÅÇ k 2 Dt z{ j 2 Dt zÑÑ
k {ÑÖ (8) measured in 160 nm-thick bilayer and trilayer PS films. Clearly,
the translational diffusivity of BPEA is reduced by 60% relative
As shown in Figure 5, the fluorescence intensity of the FRET to bulk films when confined in 160 nm-thick PS films.
donor (pyrene) remains constant for a certain time period and Figure 6 shows the T-dependent diffusivity of BPEA in three
then begins to decrease. The time when the pyrene label bilayer PS films measured at five temperatures: T = Tg − 12 K,
fluorescence intensity begins to decrease is identified as the Tg − 7 K, Tg − 2 K, Tg + 3 K, and Tg + 8 K. The total
thicknesses are 800, 160, and 130 nm. As shown in earlier
studies, at relatively narrow T ranges close to Tg, the T-

Figure 5. Time-dependent fluorescence intensity of the pyrene within

a trilayer film geometry: 30 nm-thick, pyrene-labeled PS (FRET
donor layer)/neat PS barrier layer/30 nm-thick PS doped with BPEA
(FRET acceptor layer (0.3 wt %)). The neat PS barrier layer has a
thickness of (circles) 300 or (triangles) 100 nm. Fluorescence
intensity was normalized by the value of the initial intensity plateau. Figure 6. Determination of apparent activation energy for BPEA
Data for the sample with a 100 nm-thick barrier layer have been translational diffusivity in (squares) 800 nm-thick, (circles) 160 nm-
shifted down by 0.10 for ease of comparison. The breakthrough time thick, and (triangles) 130 nm-thick PS films close to Tg. Samples are
is indicated by the arrow. Measurements were taken at 378 K. made of two layers of equal thicknesses.

12263 https://doi.org/10.1021/acs.jpcb.4c06495
J. Phys. Chem. B 2024, 128, 12259−12267
The Journal of Physical Chemistry B pubs.acs.org/JPCB Article

dependent diffusivities can be fitted to an Arrhenius equation reduced. We note that a reported apparent decoupling between
to determine the apparent activation energy (Ea) for each film the Tg-confinement effect characterized by the (pseudo)-
thickness.55,66,73 In 800 nm-thick, bulk PS films, the Ea of thermodynamic measurement of a transition temperature and
BPEA translational diffusivity is 212 kJ/mol, close to the value the (near) absence of confinement effects on cooperative
of 204 kJ/mol in bulk PiBMA66 but smaller than the value of segmental mobility could complicate this hypothesis.89 Future
253 kJ/mol in poly(ethyl methacrylate)73 reported in earlier investigations are warranted to test the validity of our
studies. Significant increases in Ea are observed with confine- hypothesis further and to assess how attractive interactions at
ment: Ea is 312 kJ/mol in 160 nm-thick PS films and 370 kJ/ a substrate interface may impact the relationship among Tg-,
mol in 130 nm-thick PS films. This increase in BPEA fragility-, diffusion- and relaxation distribution-confinement
translational diffusion Ea with confinement indicates that effects in polymer films.
BPEA must overcome larger energy barriers to undergo We compare the diffusivity-confinement effect of BPEA with
translational diffusion in supported PS films with increasing that of decacyclene, which was reported in an earlier study of
confinement at or near the nanoscale. PS films.55 BPEA has a smaller size than decacyclene and a
The observation that the Ea of BPEA translational diffusivity diffusivity in bulk PS films that is four times larger than that of
increases with confinement is in sharp contrast to what is decacyclene (D = 1.2 × 10−14 cm2/s for BPEA vs 3 × 10−15
observed regarding the apparent activation energy of the glass cm2/s for decacyclene, with both measured at Tg + 3 K).55 As
transition (Eg), which decreases with confinement in supported reported in ref 55 for PS films, decacyclene has a smaller onset
PS films.39,88 For example, Fakhraai and Forrest found that as thickness (∼100 nm) of the diffusivity-confinement effect than
the thickness of PS films decreases from 61 to 24 nm, Eg that of BPEA (∼200 nm). Decacyclene also exhibits a smaller
decreases from 1.3 × 105 K (1080 kJ/mol) to 5 × 104 K (420 percentage reduction in D when confined in PS films of similar
kJ/mol).88 In a previous study, Lan and Torkelson measured thickness. For example, the diffusivity of BPEA is reduced from
the fragility index m in confined PS films. Considering that m = bulk response by ∼50% in 160 nm-thick films, whereas the
Eg/(2.303RTg), their results suggest that Eg decreases from diffusivity of decacyclene is reduced by less than 30% in 120
1200 kJ/mol in bulk PS films to 360 kJ/mol in a 20 nm-thick nm-thick films. Although the BPEA diffusion coefficient is four
PS film.39 Importantly, in PS films supported on silica, there is times higher than that of decacylene in bulk-like PS films,
no Tg deviation from Tg,bulk until the thickness becomes less BPEA and decacyclene exhibit similar diffusivity values when
than ∼50 nm,8,10,14 whereas the onset thickness for BPEA confined in ∼100 nm-thick films. These observations can also
diffusivity-confinement effects is ∼200 nm and that of the be explained by the hypothesis described earlier. The smaller
fragility-confinement effect is between 100 and 200 nm.39 tracer, BPEA, can couple to the motion of faster relaxation
How can the different observations with the Tg-confinement regimes in PS films compared to decacyclene. If the
and diffusivity-confinement effects in PS films be understood? distribution of relaxation-time regimes becomes truncated by
At temperatures near Tg, the breadth of the α-relaxation confinement starting from the short-time side of the
distribution in bulk polymer covers many orders of magnitude distribution, then BPEA diffusivity will be affected at greater
in time (with τ ranging from 10−9 to 105 s) due to the presence film thicknesses than decacyclene diffusivity. With the
of relatively fast and relatively slow relaxation regimes; see suppression of those fastest relaxation regimes, the diffusion
Figure S2 (Supporting Information) and accompanying of BPEA depends on the motions of slower relaxation regimes,
description and/or Figure 5 in ref 74. The fast and slow which are also accessible to decacyclene. As a result, BPEA and
relaxation regimes also interchange over time. While Tg reflects decacyclene may exhibit similar diffusivities in sufficiently
the average relaxation time ⟨τ⟩ and, thus, the long-time confined PS films. Assuming the validity of this explanation,
relaxation regimes inside the system, the translational diffusion the apparent activation energy of the diffusivity of BPEA
of small-molecule dyes relates to the “summation over all should increase with confinement, as slower relaxation regimes
(dyes) of the ratio of the square of a diffusive jump length to coincide with more chain segments participating in the
the relaxation time (of) each jump,”74 i.e., it reflects ⟨τ−1⟩, relaxation, thereby increasing the energy barrier. This agrees
which is dominated by and occurs, within experimental well with our observation that Ea for the translational diffusivity
uncertainty, exclusively or nearly exclusively via the fast of BPEA increases substantially in confined PS films.
relaxation regimes (as a tracer remains essentially immobile
until the surrounding PS matrix becomes the fast relaxation
regime). On the other hand, fragility reflects the overall
■ CONCLUSIONS
We studied the impact of nanoscale confinement on the
breadth of the α-relaxation distribution79 inside the entire translational diffusivity of small-molecule dyes in PS films by
matrix, and a smaller fragility value indicates a narrower fluorescence spectroscopy via FRET. A major reduction of
relaxation distribution. With sufficient confinement, the BPEA diffusivity is observed in supported PS films thinner than
fragility index is significantly reduced in PS films, which ∼200 nm, and the diffusivity is reduced by ∼80−90% in 100
indicates that the breadth of the relaxation distribution nm-thick films. The onset thickness for the diffusivity-
becomes narrower.72 If the narrowing of the relaxation confinement effect is much larger than that of the Tg-
distribution occurs predominantly on the short-time side confinement effect and closer to that of the fragility-
down to a thickness of ∼50 nm, it results in a major reduction confinement effect in supported PS films. Additionally, the
in ⟨τ−1⟩, and, hence, the translational diffusivity of small- apparent activation energy for the translational diffusivity of
molecule dyes is significantly reduced. However, ⟨τ⟩ and Tg are BPEA in PS is significantly enhanced with confinement. A
not noticeably affected at length scales where the long-time connection among the fragility-confinement, diffusivity-con-
side of the α-relaxation distribution is substantially un- finement, and Tg-confinement effects is made, which acknowl-
affected.8,10,14,41,74 When the film is confined to a more severe edges that the three parameters reflect different aspects of the
extent (i.e., <50 nm) and the long-time relaxation regimes are overall relaxation distribution. Fragility is a fundamental
suppressed by confinement, ⟨τ⟩ becomes smaller and Tg is parameter that reflects the chain packing efficiency of the
12264 https://doi.org/10.1021/acs.jpcb.4c06495
J. Phys. Chem. B 2024, 128, 12259−12267
The Journal of Physical Chemistry B pubs.acs.org/JPCB Article

polymer matrix and relates to the overall breadth of the α- acknowledge that portions of this manuscript were reported in
relaxation distribution. A reduction in fragility with confine- the Ph.D. dissertations of Tian Lan90 and Tong Wei.91
ment indicates a narrower relaxation distribution in confined
PS films; reductions in Tg and diffusivity with confinement are
consistent with the long-time side of the distribution being
■ REFERENCES
(1) Ruiz, R.; Kang, H.; Detcheverry, F. A.; Dobisz, E.; Kercher, D. S.;
shifted to shorter times and the short-time side of the Albrecht, T. R.; de Pablo, J. J.; Nealey, P. F. Density Multiplication
distribution being shifted to longer times with confinement. and Improved Lithography by Directed Block Copolymer Assembly.
Consequently, both the translational diffusivity of a small- Science 2008, 231, 936−939.
molecule tracer (reflected by ⟨τ−1⟩) and the Tg of the PS (2) Wieberger, F.; Forman, D. C.; Neuber, C.; Gröschel, A. H.;
matrix (reflected by ⟨τ⟩) decrease with confinement, and the Böhm, M.; Müller, A. H. E.; Schmidt, H. W.; Ober, C. K. Tailored
Star-Shaped Statistical Teroligomers via ATRP for Lithographic
small-molecule diffusivity and PS fragility are influenced at Applications. J. Mater. Chem. 2012, 22, 73−79.
larger length scales compared to Tg. We also find that BPEA (3) Ulbricht, M. Advanced Functional Polymer Membranes. Polymer
exhibits a more substantial diffusivity-confinement effect than a 2006, 47, 2217−2262.
larger dye, decacyclene, in terms of both the length scale at (4) Halls, J. J. M.; Walsh, C. A.; Greenham, N. C.; Marseglia, E. A.;
which the confinement effect becomes evident and the Friend, R. H.; Moratti, S. C.; Holmes, A. B. Efficient Photodiodes
percentage reduction in diffusivity. from Interpenetrating Polymer Networks. Nature 1995, 376, 498−
500.
■ ASSOCIATED CONTENT
Data Availability Statement
(5) Yu, G.; Gao, J.; Hummelen, J. C.; Wudl, F.; Heeger, A. J.
Polymer Photovoltaic Cells-Enhanced Efficiencies via a Network of
Internal Donor-Acceptor Heterojunctions. Science 1995, 270, 1789−
The data that support the findings of this study are available 1791.
from the corresponding author upon reasonable request. (6) Chen, J.; Wang, F.; Liu, Q.; Du, J. Antibacterial Polymeric
*
sı Supporting Information Nanostructures for Biomedical Applications. Chem. Commun. 2014,
The Supporting Information is available free of charge at 50, 14482−14493.
(7) He, W. G.; Xiao, G. Y.; Jin, X. B.; Sun, C. H.; Ma, P. X.
https://pubs.acs.org/doi/10.1021/acs.jpcb.4c06495. Electrodeposition on Nanofibrous Polymer Scaffolds: Rapid Miner-
Derivation of dye concentration function, visual of alization, Tunable Calcium Phosphate Composition and Topography.
normalized dye concentration as a function of Dt/d2, Adv. Funct. Mater. 2010, 20, 3568−3576.
and representative visual of narrowing of relaxation time (8) Keddie, J. L.; Jones, R. A. L.; Cory, R. A. Size-Dependent
distribution for polymer samples near Tg (PDF) Depression of the Glass Transition Temperature in Polymer Films.
Europhys. Lett. 1994, 27, 59−64.

■ AUTHOR INFORMATION
Corresponding Author
(9) Dalnoki-Veress, K.; Forrest, J. A.; Murray, C.; Gigault, C.;
Dutcher, J. R. Molecular Weight Dependence of Reductions in the
Glass Transition Temperature of Thin, Freely Standing Polymer
Films. Phys. Rev. E: Stat. Phys., Plasmas, Fluids, Relat. Interdiscip. Top.
John M. Torkelson − Department of Materials Science, 2001, 63, No. 031801.
Engineering Northwestern University, Evanston, Illinois (10) Ellison, C. J.; Torkelson, J. M. The Distribution of Glass
60208, United States; Department of Chemical and Transition Temperatures in Nanoscopically Confined Glass Formers.
Biological Engineering, Northwestern University, Evanston, Nat. Mater. 2003, 2, 695−700.
Illinois 60208, United States; orcid.org/0000-0002- (11) Lang, R. J.; Simmons, D. S. Interfacial Dynamic Length Scales
4875-4827; Email: [email protected] in the Glass Transition of a Model Freestanding Polymer Film and
Their Connection to Cooperative Motion. Macromolecules 2013, 46,
Authors 9818−9825.
Tian Lan − Department of Materials Science, Engineering (12) Ediger, M. D.; Forrest, J. A. Dynamics near Free Surfaces and
Northwestern University, Evanston, Illinois 60208, United the Glass Transition in Thin Polymer Films: A View to the Future.
States Macromolecules 2014, 47, 471−478.
Tong Wei − Department of Chemical and Biological (13) Geng, K.; Tsui, O. K. C. Effects of Polymer Tacticity and
Molecular Weight on The Glass Transition Temperature of
Engineering, Northwestern University, Evanston, Illinois Poly(Methyl Methacrylate) Films on Silica. Macromolecules 2016,
60208, United States 49, 2671−2678.
Logan M. Fenimore − Department of Chemical and (14) Evans, C. M.; Deng, H.; Jager, W. F.; Torkelson, J. M. Fragility
Biological Engineering, Northwestern University, Evanston, is a Key Parameter in Determining the Magnitude of Tg-Confinement
Illinois 60208, United States; orcid.org/0000-0001- Effects in Polymer Films. Macromolecules 2013, 46, 6091−6103.
5151-1255 (15) Mundra, M. K.; Donthu, S. K.; Dravid, V. P.; Torkelson, J. M.
Effect of Spatial Confinement on the Glass Transition Temperature of
Complete contact information is available at: Patterned Polymer Nanostructures. Nano Lett. 2007, 7, 713−718.
https://pubs.acs.org/10.1021/acs.jpcb.4c06495 (16) Jin, K.; Torkelson, J. M. Enhanced Tg-Confinement Effect in
Cross-Linked Polystyrene Compared to Its Linear Precursor: Roles of
Notes Fragility and Chain Architecture. Macromolecules 2016, 49, 5092−
The authors declare no competing financial interest. 5103.

■ ACKNOWLEDGMENTS
This research was supported by discretionary funds from a
(17) Lan, T.; Torkelson, J. M. Methacrylate-Based Polymer Films
Useful in Lithographic Applications Exhibit Different Glass Transition
Temperature-Confinement Effects at High and Low Molecular
Weight. Polymer 2014, 55, 1249−1258.
Walter P. Murphy Professorship (J.M.T.), the National Science (18) Lan, T.; Torkelson, J. M. Substantial Spatial Heterogeneity and
Foundation for a Graduate Research Fellowship (L.M.F.), and Tunability of Glass Transition Temperature Observed with Dense
the McCormick School of Engineering at Northwestern Polymer Brushes Prepared by ARGET ATRP. Polymer 2015, 64,
University for a Terminal Year Fellowship (L.M.F.). We also 183−192.

12265 https://doi.org/10.1021/acs.jpcb.4c06495
J. Phys. Chem. B 2024, 128, 12259−12267
The Journal of Physical Chemistry B pubs.acs.org/JPCB Article

(19) Zhang, L.; Elupula, R.; Grayson, S. M.; Torkelson, J. M. Major (37) Ye, C.; Wiener, C. G.; Tyagi, M.; Uhrig, D.; Orski, S. V.; Soles,
Impact of Cyclic Chain Topology on the Tg-Confinement Effect of C. L.; Vogt, B. D.; Simmons, D. S. Understanding the Decreased
Supported Thin Films of Polystyrene. Macromolecules 2016, 49, 257− Segmental Dynamics of Supported Thin Polymer Films Reported by
268. Incoherent Neutron Scattering. Macromolecules 2015, 48, 801−808.
(20) Xia, W.; Mishra, S.; Keten, S. Substrate vs. Free Surface: (38) Glor, E.; Composto, R. J.; Fakhraai, Z. Glass Transition
Competing Effects on the Glass Transition of Polymer Thin Films. Dynamics and Fragility of Ultrathin Miscible Polymer Blend Films.
Polymer 2013, 54, 5942−5951. Macromolecules 2015, 48, 6682−6689.
(21) Xia, W.; Hsu, D.; Keten, S. Molecular Weight Effects on the (39) Lan, T.; Torkelson, J. M. Fragility-Confinement Effects:
Glass Transition and Confinement Behavior of Polymer Thin Films. Apparent Universality as a Function of Scaled Thickness in Films of
Macromol. Rapid Commun. 2015, 36, 1422−1427. Freely Deposited, Linear Polymer and Its Absence in Densely Grafted
(22) Ellison, C. J.; Kim, S. D.; Hall, D. B.; Torkelson, J. M. Brushes. Macromolecules 2016, 49, 1331−1343.
Confinement and Processing Effects on Glass Transition Temper- (40) Fukao, K.; Miyamoto, Y. Slow Dynamics near Glass Transitions
ature and Physical Aging in Ultrathin Polymer Films: Novel in Thin Polymer Films. Phys. Rev. E 2001, 64, No. 011803.
Fluorescence Measurements. Eur. Phys. J. E 2002, 8, 155−166. (41) Napolitano, S.; Wübbenhorst, M. Structural Relaxation and
(23) Kim, S.; Hewlett, S. A.; Roth, C. B.; Torkelson, J. M. Dynamic Fragility of Freely Standing Polymer Films. Polymer 2010,
Confinement Effects on Glass Transition Temperature, Transition 51, 5309−5312.
Breadth, and Expansivity: Comparison of Ellipsometry and (42) Zhang, C.; Guo, Y.; Shepard, K.; Priestley, R. D. Fragility of an
Fluorescence Measurements on Polystyrene Films. Eur. Phys. J. E Isochorically Confined Polymer Glass. J. Phys. Chem. Lett. 2013, 4,
2009, 30, 83−92. 431−436.
(24) Wang, T.; Hu, S.; Zhang, S.; Peera, A.; Reffner, J.; Torkelson, J. (43) Zhang, C.; Priestley, R. D. Fragility and Glass Transition
M. Eliminating the Tg-Confinement Effect in Polystyrene Films: Temperature of Polymer Confined under Isobaric and Isochoric
Extraordinary Impact of a 2 mol % 2-Ethylhexyl Acrylate Conditions. Soft Matter 2013, 9, 7076−7085.
Comonomer. Macromolecules 2022, 55, 9601−9611. (44) Starr, F. W.; Douglas, J. F. Modifying Fragility and Collective
(25) Serna, S.; Wang, T.; Torkelson, J. M. Eliminating the Tg- Motion in Polymer Melts with Nanoparticles. Phys. Rev. Lett. 2011,
Confinement and Fragility-Confinement Effects in Poly(4-Methyl- 106, No. 115702.
styrene) Films by Incorporation of 3 mol % 2-Ethylhexyl Acrylate (45) Marvin, M. D.; Lang, R. J.; Simmons, D. S. Nanoconfinement
Comonomer. J. Chem. Phys. 2024, 160, No. 034903. Effects on the Fragility of Glass Formation of a Model Freestanding
(26) Kim, S.; Torkelson, J. M. Distribution of Glass Transition Polymer Film. Soft Matter 2014, 10, 3166−3170.
Temperatures in Free-Standing, Nanoconfined Polystyrene Films: A (46) Wang, T.; Peera, A.; Reffner, J.; Torkelson, J. M. Reducing the
Test of de Gennes’ Sliding Motion Mechanism. Macromolecules 2011, Bulk Fragility and Suppressing the Fragility-Confinement Effect in
44, 4546−4553. Polystyrene with Very Low Levels of 2-Ethylhexyl Acrylate
(27) Ellison, C. J.; Torkelson, J. M. Sensing the Glass Transition in
Comonomer. Macromolecules 2023, 56, 3527−3537.
Thin and Ultrathin Polymer Films via Fluorescence Probes and (47) Zhang, L.; Elupula, R.; Grayson, S. M.; Torkelson, J. M.
Labels. J. Polym. Sci., Part B: Polym. Phys. 2002, 40, 2745−2758.
Suppression of the Fragility-Confinement Effect via Low Molecular
(28) Ellison, C. J.; Ruszkowski, R. L.; Fredin, N. J.; Torkelson, J. M.
Weight Cyclic or Ring Polymer Topology. Macromolecules 2017, 50,
Dramatic Reduction of the Effect of Nanoconfinement on the Glass
1147−1154.
Transition of Polymer Films via Addition of Small-Molecule Diluent.
(48) Frank, B.; Gast, A. P.; Russell, T. P.; Brown, H. R.; Hawker, C.
Phys. Rev. Lett. 2004, 92, No. 095702.
Polymer Mobility in Thin Films. Macromolecules 1996, 29, 6531−
(29) Mundra, M. K.; Ellison, C. J.; Rittigstein, P.; Torkelson, J. M.
Fluorescence Studies of Confinement in Polymer Films and 6534.
(49) Zheng, X.; Rafailovich, M. H.; Sokolov, J.; Strzhemechny, Y.;
Nanocomposites: Glass Transition Temperature, Plasticizer Effects,
and Sensitivity to Stress Relaxation and Local Polarity. Eur. Phys. J. Schwarz, S. A.; Sauer, B. B.; Rubinstein, M. Long-Range Effects on
Special Topics 2007, 141, 143−151. Polymer Diffusion Induced by a Bounding Interface. Phys. Rev. Lett.
(30) Hall, D. B.; Hooker, J. C.; Torkelson, J. M. Ultrathin Polymer 1997, 79, 241−244.
Films near the Glass Transition: Effect on the Distribution of α- (50) Napolitano, S.; Rotella, C.; Wübbenhorst, M. Is the Reduction
Relaxation Times as Measured by Second Harmonic Generation. in Tracer Diffusivity under Nanoscopic Confinement related to a
Macromolecules 1997, 30, 667−669. Frustrated Segmental Mobility? Macromol. Rapid Commun. 2011, 32,
(31) Torres, J. M.; Stafford, C. M.; Uhrig, D.; Vogt, B. D. Impact of 844−848.
Chain Architecture (Branching) on the Thermal and Mechanical (51) Katzenstein, J. M.; Janes, D. W.; Hocker, H. E.; Chandler, J. K.;
Behavior of Polystyrene Thin Films. J. Polym. Sci., Part B: Polym. Phys. Ellison, C. J. Nanoconfined Self-Diffusion of Poly(isobutyl meth-
2012, 50, 370−377. acrylate) in Films with a Thickness-Independent Glass Transition.
(32) Chen, F.; Peng, D.; Ogata, Y.; Tanaka, K.; Yang, Z.; Fujii, Y.; Macromolecules 2012, 45, 1544−1552.
Yamada, N. L.; Lam, C.-H.; Tsui, O. K. C. Viscosity and Surface- (52) Choi, J.; Hore, M. J. A.; Clarke, N.; Winey, K. I.; Composto, R.
Promoted Slippage of Thin Polymer Films Supported by a Solid J. Nanoparticle Brush Architecture Controls Polymer Diffusion in
Substrate. Macromolecules 2015, 48, 7719−7726. Nanocomposites. Macromolecules 2014, 47, 2404−2410.
(33) Delcambre, S. P.; Riggleman, R. A.; de Pablo, J. J.; Nealey, P. F. (53) Choi, J.; Hore, M. J. A.; Meth, J. S.; Clarke, N.; Winey, K. I.;
Mechanical Properties of Antiplasticized Polymer Nanostructures. Composto, R. J. Universal Scaling of Polymer Diffusion in
Soft Matter 2010, 6, 2475−2483. Nanocomposites. ACS Macro Lett. 2013, 2, 485−490.
(34) Askar, S.; Evans, C. M.; Torkelson, J. M. Residual Stress (54) Lin, C.-C.; Gam, S.; Meth, J. S.; Clarke, N.; Winey, K. I.;
Relaxation and Stiffness in Spin-Coated Polymer Films: Character- Composto, R. J. Do Attractive Polymer-Nanoparticle Interactions
ization by Ellipsometry and Fluorescence. Polymer 2015, 76, 113− Retard Polymer Diffusion in Nanocomposites? Macromolecules 2013,
122. 46, 4502−4509.
(35) Qiang, Z.; Li, L.; Torkelson, J. M.; Wang, M. Determining (55) Hall, D. B.; Torkelson, J. M. Small Molecule Probe Diffusion in
Order-to-Disorder Transitions in Block Copolymer Thin Films Using Thin and Ultrathin Supported Polymer Films. Macromolecules 1998,
a Self-Referencing Fluorescent Probe. Mol. Syst. Des. Eng. 2020, 5, 31, 8817−8825.
330−338. (56) Vogt, B. D.; Soles, C. L.; Lee, H. J.; Lin, E. K.; Wu, W. L.
(36) Priestley, R. D.; Ellison, C. J.; Broadbelt, L. J.; Torkelson, J. M. Moisture Absorption and Absorption Kinetics in Polyelectrolyte
Structural Relaxation of Polymer Glasses at Surfaces, Interfaces, and Films: Influence of Film Thickness. Langmuir 2014, 30, 14056−
In Between. Science 2005, 309, 456−459. 14061.

12266 https://doi.org/10.1021/acs.jpcb.4c06495
J. Phys. Chem. B 2024, 128, 12259−12267
The Journal of Physical Chemistry B pubs.acs.org/JPCB Article

(57) Tseng, K. C.; Turro, N. J.; Durning, C. J. Molecular Mobility in (78) Kunal, K.; Robertson, C. G.; Pawlus, S.; Hahn, S. F.; Sokolov,
Polymer Thin Films. Phys. Rev. E 2000, 61, 1800−1810. A. P. Role of Chemical Structure in Fragility of Polymers: A
(58) Roth, C. B. Polymers Under Nanoconfinement: Where Are We Qualitative Picture. Macromolecules 2008, 41, 7232−7238.
Now in Understanding Local Property Changes? Chem. Soc. Rev. (79) Böhmer, R.; Ngai, K. L.; Angell, C. A.; Plazek, D. J.
2021, 50, 8050−8065. Nonexponential Relaxations in Strong and Fragile Glass Formers. J.
(59) Li, B.; Zhang, S.; Andre, J. S.; Chen, Z. Relaxation Behavior of Chem. Phys. 1993, 99, 4201−4209.
Polymer Thin Films: Effects of Free Surface, Buried Interface, and (80) Zhang, L.; Marsiglio, J. A.; Lan, T.; Torkelson, J. M. Dramatic
Geometrical Confinement. Prog. Polym. Sci. 2021, 120, No. 101431. Tunability of the Glass Transition Temperature and Fragility of Low
(60) Burroughs, M. J.; Christie, D.; Gray, L. A. G.; Chowdhury, M.; Molecular Weight Polystyrene by Initiator Fragments Located at
Priestley, R. D. 21st Century Advances in Fluorescence Techniques to Chain Ends. Macromolecules 2016, 49, 2387−2398.
Characterize Glass-Forming Polymers at the Nanoscale. Macromol. (81) Santangelo, P. G.; Roland, C. M. Molecular Weight
Chem. Phys. 2018, 219, 1700368. Dependence of Fragility in Polystyrene. Macromolecules 1998, 31,
(61) Cronin, M. F.; Adams, T. G.; Fedynyshyn, T. H.; Georger, J. 4581−4585.
H.; Mori, J. M.; Sinta, R. F.; Thackery, J. W. Investigation of Onium (82) Roland, C. M.; Santangelo, P. G.; Ngai, K. L. The Application
Salt Type Photoacid Generators in Positive DUV Resist Systems. of the Energy Landscape Model in Polymers. J. Chem. Phys. 1999,
Proc. SPIE 1994, 2195, 214−224. 111, 5593−5598.
(62) Vogt, B. D.; Kang, S. H.; Prabhu, V. M.; Lin, E. K.; Satija, S. K.; (83) Robertson, C. G.; Santangelo, P. G.; Roland, C. M. Comparison
Turnquest, K.; Wu, W. L. Measurements of the Reaction-Diffusion of Glass Forming Kinetics and Segmental Relaxation in Polymers. J.
Front of Model Chemically Amplified Photoresists with Varying Non-Cryst. Solids 2000, 275, 153−159.
Photoacid Size. Macromolecules 2006, 39, 8311−8317. (84) Dhinojwala, A.; Wong, G. K.; Torkelson, J. M. Rotational
(63) Dhinojwala, A.; Torkelson, J. M. A Reconsideration of the Reorientation Dynamics of Dispersed Red 1 in Polystyrene: α-
Measurement of Polymer Interdiffusion by Fluorescence Nonradiative Relaxation Dynamics Probed by Second Harmonic Generation. J.
Energy Transfer. Macromolecules 1994, 27, 4817−4824. Chem. Phys. 1994, 100, 6046−6054.
(64) Hall, D. B.; Miller, R. D.; Torkelson, J. M. Molecular Probe (85) Kausch, H. H.; Tirrell, M. Polymer Interdiffusion. Annu. Rev.
Techniques for Studying Diffusion and Relaxation in Thin and Mater. Sci. 1989, 19, 341−377.
Ultrathin Polymer Films. J. Polym. Sci.: Polym. Phys. Ed. 1997, 35, (86) Karim, A.; Felcher, G. P.; Russell, T. P. Interdiffusion of
2795−2802. Polymers at Short Times. Macromolecules 1994, 27, 6973−6799.
(65) Hall, D. B.; Deppe, D. D.; Hamilton, K. E.; Dhinojwala, A.; (87) Crank, J.The Mathematics of Diffusion. Claredon Press: Oxford,
Torkelson, J. M. Probe Translational and Rotational Diffusion in 1975.
Polymers near Tg: Roles of Probe Size, Shape, and Secondary Bonding (88) Fakhraai, Z.; Forrest, J. Probing Slow Dynamics in Supported
in Deviations from Debye-Stokes-Einstein Scaling. J. Non-Cryst. Solids Thin Polymer Films. Phys. Rev. Lett. 2005, 95, No. 025701.
1998, 235, 48−56. (89) Priestley, R. D.; Cangialosi, D.; Napolitano, S. On the
(66) Deppe, D. D.; Miller, R. D.; Torkelson, J. M. Small Molecule Equivalence between the Thermodynamic and Dynamic Measure-
Diffusion in a Rubbery Polymer near Tg: Effects of Probe Size, Shape, ments of the Glass Transition in Confined Polymers. J. Non-Cryst.
and Flexibility. J. Polym. Sci., Polym. Phys. Ed. 1996, 34, 2987−2997. Solids 2015, 407, 288−295.
(67) Ehlich, D.; Sillescu, H. Tracer Diffusion at the Glass Transition. (90) Lan, T.Impact of Confinement and Interfaces on Glass Transition
Phys. Rev. E 1990, 23, 1600−1610. Temperature, Fragility and Small Molecule Diffusivity in Polymers. Ph.D.
(68) Rössler, E. Indications for a Change of Diffusion Mechanism in Dissertation; Northwestern University: Evanston, IL, 2016. https://
Supercooled Liquids. Phys. Rev. Lett. 1990, 65, 1595−1598. www.proquest.com/docview/1826019976?parentSessionId=
(69) Fujara, F.; Geil, B.; Sillescu, H.; Fleischer, G. Translational and qoWywiAiS3XqSmv1law5fVK0RRrM1EguH0pRZQ7Bk1A%3D&pq-
Rotational Diffusion in Supercooled Orthoterphenyl close to the origsite=primo&accountid=12861&sourcetype=
Glass Transition. Z. Phys. B-Condensed Matter 1992, 88, 195−204. Dissertations%20&%20Theses (accessed 2024−11−20).
(70) Stillinger, F. H.; Hodgdon, J. A. Translation-Rotation Paradox (91) Wei, T.Interfacial and Confinement Effects in Polymer and
for Diffusion in Fragile Glass-Forming Liquids. Phys. Rev. E 1994, 50, Polymer Nanocomposites in Absence of Free Surfaces. Ph.D. Dissertation;
2064−2068. Northwestern University: Evanston, IL, 2020. https://www.proquest.
(71) Cicerone, M. T.; Blackburn, F. R.; Ediger, M. D. Anomalous com/docview/2446708756?parentSessionId=
Diffusion of Probe Molecules in Polystyrene: Evidence for Spatially 3eQuazCmMbmmx9ZzIj1kv2iOZk%2FB7vFafWek
Heterogeneous Segmental Dynamics. Macromolecules 1995, 28, WHID6m0%3D&pq-origsite=primo&accountid=12861&sourcetype=
8224−8232. Dissertations%20&%20Theses (accessed 2024−11−20).
(72) Wang, C. Y.; Ediger, M. D. Enhanced Translational Diffusion of
9,10-Bis(phenylethynyl)anthracene (BPEA) in Polystyrene. Macro-
molecules 1997, 30, 4770−4771.
(73) Deppe, D. D.; Dhinojwala, A.; Torkelson, J. M. Small-Molecule
Probe Diffusion in Thin Polymer Films near the Glass Transition: A
Novel Approach Using Fluorescence Nonradiative Energy Transfer.
Macromolecules 1996, 29, 3898−3908.
(74) Hall, D. B.; Dhinojwala, A.; Torkelson, J. M. Translation-
Rotation Paradox for Diffusion in Glass-Forming Polymers: The Role
of the Temperature Dependence of the Relaxation Time Distribution.
Phys. Rev. Lett. 1997, 79, 103−106.
(75) Cicerone, M. T.; Blackburn, F. R.; Ediger, M. D. Anomalous
Diffusion of Probe Molecules in Polystyrene: Evidence for Spatially
Heterogeneous Segmental Dynamics. Macromolecules 1995, 28,
8224−8232.
(76) Dudowicz, J.; Freed, K. F.; Douglas, J. F. The Glass Transition
Temperature of Polymer Melts. J. Phys. Chem. B 2005, 109, 21285−
21292.
(77) Dudowicz, J.; Freed, K. F.; Douglas, J. F. Fragility of Glass
Forming Polymer Liquids. J. Phys. Chem. B 2005, 109, 21350−21356.

12267 https://doi.org/10.1021/acs.jpcb.4c06495
J. Phys. Chem. B 2024, 128, 12259−12267