Module 8: Relative Permeability
Synopsis
What is water-oil relative permeability and why does it matter?
endpoints and curves, fractional flow, what curve shapes mean
Understand the jargon (and impress reservoir engineers) Wettability
water-wet, oil-wet and intermediate
How do we measure it (in the lab)? How do we quality control and refine data?
Page 2
Applications
To predict movement of fluid in the reservoir
e.g velocity of water and oil fronts
To predict and bound ultimate recovery factor Application depends on reservoir type
gas-oil water-oil gas-water
Page 3
Definitions
Absolute Permeability
permeability at 100% saturation of single fluid
e.g. brine permeability, gas permeability
Effective Permeability
permeability to one phase when 2 or more phases present
e.g. ko(eff) at Swi
Relative Permeability
ratio of effective permeability to a base (often absolute) permeability
e.g. ko/ka or ko/ko at Swi
Page 4
Requirements
Gas-Oil Relative Permeability (kg-ko)
solution gas drive gas cap drive
Water-Oil Relative Permeability(kw-ko)
water injection
Water - Gas Relative Permeability (kw-kg)
aquifer influx into gas reservoir
Gas-Water Relative Permeability (kg-kw)
gas storage (gas re-injection into gas reservoir)
Page 5
Jargon Buster!
Relative permeability curves are known as rel perms Endpoints are the (4) points at the ends of the curves The displacing phase is always first, i.e.:
kw-ko is water(w) displacing oil (o) kg-ko is gas (g) displacing oil (o) kg-kw is gas displacing water
Page 6
Why shape is important
Measure air permeability Saturate core in water (brine) Desaturate to Swir
Centrifuge or porous plate Measure oil permeability ko @ Swir endpoint
Ko = 80 mD
ka = 100 mD Swir = 0.20 (20%
So = 1-Swir Swirr
Oil = Sro Sw = 1-Sro
Waterflood collect water volume
Swr = 1-0.25 = 0.75
Sro = 0.25
Measure water permeability kw @Sro endpoint
Page 7
Kw = 24 mD
Endpoints
1.0 0.9 0.8 0.7 0.6
Endpoint- oil kro = ko/ko @ Swir = 80/80 =1 Swir = 0.20 Sro = 0.25
Relative Permeability (-)
0.5 0.4 0.3 0.2 0.1
Endpoint - water krw = kw/ko @ Swir = 24/80 = 0.30
0.0 0.0 0.1 0.2 0.3 0.4 0.5 Water Saturation (-) 0.6 0.7 0.8 0.9 1.0
Page 8
Endpoints
1.0 0.9 0.8 0.7 0.6
Relative Permeability (-)
Swir = 0.20
0.5 0.4 0.3 0.2 0.1 0.0 0.0 0.1 0.2 0.3 0.4 0.5 Water Saturation (-) 0.6 0.7
Sro = 0.25
0.8
0.9
1.0
Page 9
Curves - 1
1.0 0.9 0.8 0.7 0.6
Relative Permeability (-)
Swir = 0.20
0.5 0.4 0.3 0.2 0.1 0.0 0.0 0.1 0.2 0.3 0.4 0.5 Water Saturation (-) 0.6 0.7
Sro = 0.25
0.8
0.9
1.0
Page 10
Curves - 2
1.0 0.9 0.8 0.7 0.6
Relative Permeability (-)
Swir = 0.20
0.5 0.4 0.3 0.2 0.1 0.0 0.0 0.1 0.2 0.3 0.4 0.5 Water Saturation (-) 0.6 0.7
Sro = 0.25
0.8
0.9
1.0
Page 11
Curves - 3
1.0 0.9 0.8 0.7 0.6
Relative Permeability (-)
Swir = 0.20
0.5 0.4 0.3 0.2 0.1 0.0 0.0 0.1 0.2 0.3 0.4 0.5 Water Saturation (-) 0.6 0.7
Sro = 0.25
0.8
0.9
1.0
Page 12
Relative Permeability
1
Non-linear function of Swet Competing forces
gravity forces
Relative Permeability (-)
0.9
0.8
0.7
0.6
minimised in lab tests e.g. water injected from bottom to top
0.5
kro krw
0.4
viscous forces
Darcys Law
0.3
0.2
0.1
capillary forces
low flood rates
Page 13
0 0 0.2 0.4 0.6 0.8 1 Water Saturation (-)
Relative Permeability Curves Key Features
Water-Oil Curves
irreducible water saturation (Swir) endpoint
kro = 1.0 krw = 0.0
residual oil saturation (Sro) endpoint
kro = 0.0 krw = maximum
relative permeability curve shape
Unsteady-state Steady-state Corey exponents:
Page 14
Buckley-Leverett, Welge, JBN Darcy No and Nw
Waterflood Interpretation
Welge
fw=1
fw only after BT
Average Saturation behind flood front
Swf , fw | S
wf
fw
Sw at BT
fw =
1 +
k ro . k rw
w o
Page 15
Swc
Sw
1-Sor
Relative Permeability Interpretation
Welge/Buckley-Leverett fraction flow
gives ratio: kro/krw
fw =
1 +
k ro . k rw
k rw o . M= k ro w
M< 1: piston-like M > 1: unstable
w o
Decouple kro and krw from kro/krw
JBN, Jones and Roszelle, etc
Page 16
JBN Method Outline
Johnson, Bossler, Nauman (JBN)
Based on Buckley-Leverett/Welge W = PV water injected Swa = average (plug) Sw fw2 = 1-fo2
fw = 1+ k ro w . k rw o 1
dS wa = fo2 dW
d( 1 ) f WI r = o2 1 k ro 2 d( ) W
pt =0 Ir = pt =i
Page 17
Injectivity Ratio Waterflood rate, q
Buckley Leverett Assumptions
Fluids are immiscible Fluids are incompressible Flow is linear (1 Dimensional) Flow is uni-directional Porous medium is homogeneous Capillary effects are negligible Most are not met in most core floods
Page 18
Capillary End Effect
If viscous force large (high rate)
Pc effects negligible
If viscous force small (low rate)
Pc effects dominate flood behaviour
Leverett
capillary boundary effects on short cores boundary effects negligible in reservoir
Page 19
End Effect
Pressure Trace for Flood zero p (no injection) start of injection water nears exit
p increases abruptly until Sw(exit) = 1-Sro and Pc nears zero suppresses krw Sw(exit) = 1-Sro, Pc ~0 rate of p increase reduces as krw increases
BT
Page 20
After BT
Scaling Coefficient
Breakthrough Recovery (Rappaport & Leas) Affected by Pc end effects At lengths > 25 cm Little effect on BT recovery (LVw > 1) Hence composite samples or high rates
Page 21
Capillary End Effects
Rapaport and Leas Scaling Coefficient
LVw > 1(cm2/[Link]) : minimal end effect
Overcome by:
flooding at high rate
300 ml/hour +
using longer cores
difficult for reservoir core (limited by core geometry) butt several cores together
using capillary mixing sections
end-point saturations only in USS tests (weigh sample)
Page 22
Composite Core Plug
Capillary end effects adsorbed by Cores 1 and 4
Page 23
Corey Exponents Water/Oil Systems
Define relative permeability curve shapes Based on normalised saturations No guarantee that real rock curves obey Corey
kro = SonNo krw = krw(Swn)Nw krw = end-point krw
1 S w Sro Son = = 1 S wn 1 S wi Sro
S wn
Page 24
S w S wi = 1 S wi S ro
Normalisation
Swn = 1
1 0.9 0.8
Water Relative Permeability (-)
0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 0 0.1 0.2 0.3 0.4 0.5 Water Saturation (-) 0.6 0.7 0.8 0.9 1 krw at Sro krwn = 1
Sample 1 Sample 2
krwn = 1
Page 25
Corey Exponents
Depend on wettability
Wettability Water-Wet Intermediate Wet Oil-Wet No (kro) 2 to 4 3 to 6 6 to 8 Nw (krw) 5 to 8 3 to 5 2 to 3
Uses:
interpolate & extrapolate data lab data quality control
Page 26
Gas-Oil Relative Permeability
Pore-Scale Saturation Distribution
Test performed at Swir
Gas is non wetting takes easiest flow path kro drops rapidly as Sg increases krg higher than krw Srog > Srow in lab tests
end effects
Srog < Srow in field
Page 27
Sgc ~ 2% - 6%
Typical Gas-Oil Curves: Linear
1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 Gas Saturation (fractional)
Relative Permeability (-)
1-(Srog+Swi)
kro krg
Sgc
Page 28
Labs plot kr vs liquid saturation (So+Swi)
Typical Gas-Oil Curves: Semi-Log
1
Relative Permeability (-)
0.1
1-(Srog+Swi)
kro krg
0.01
0.001 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 Gas Saturation (fractional)
Page 29
Gas-Oil Curves
Most lab data are artefacts
due to capillary end effects
Tests should be carried out on long cores
insufficient flood period
Real gas-oil curves
Sgc ~ 3% Srog is low and approaches zero
Due to thin film and gravity drainage
krg = 1 at Srog = 0
Page 30
well defined Corey exponents
Gas-Oil Curves Corey Method
Oil relative permeability
normalised oil saturation
kro = Son No
Son = 1 Sg Swir Srog 1 Swir Srog
Gas relative permeability
normalised gas saturation
Sgc: critical gas saturation
Corey Exponent No Ng
Page 31
krg = Sgn Ng
Sgn = Sg Sgc 1 Swir Srog Sgc
Values 4 to 7 1.3 to 3.0
Corey Gas-Oil Curves
1
0.1
Swir kro krg' Srog Sgc
Kro No = 4 krg Ng = 1.3 kro No = 7 krg Ng = 3.0
0.15 1.00 1.00 0.0000 0.0300
Relative Permeability (-)
0.01
0.001
Sgc = 0.03
0.0001
0.00001 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 Gas Saturation (-)
Page 32
Typical Lab Data - krg
Krg too low
1
Srog too high
0.1
Relative Permeability, krg
0.01
0.001
Composite Gas-Oil Curves Ng : No : Sgc: Srog: krg' : 2.3 4.0 0.03 0.10 1.0
Ng = 2.3; Swir = 0.15 Ng = 2.3; Swir = 0.20 11a-5 # 4 11a-5 # 31 11a-5 # 34 11a-5 #39 11a-7 BEA5 11a-7 BEA7 11a-7 BEB5 11a-7 BEC5
0.0001
0.00001 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 Swi+Sg (fraction)
Page 33
Laboratory Methods
Core Selection
all significant reservoir flow units often constrained by preserved core availability core CT scanning to select plugs
Core Size
at least 25 cm long to overcome end effects butt samples (but several end effects?) flood at high rate to overcome end effects?
Page 34
Test States
Fresh or Preserved State
tested as is (no cleaning) probably too oil wet (e.g OBM, long term storage) Native state term also used (defines bland mud) Some labs fresh state is other labs restored state
Cleaned State
Cleaned (soxhlet or miscible flush) water-wet by definition (but could be oil-wet!!!!!!)
Restored State (reservoir-appropriate wettability)
saturate in crude oil (live or dead) age in oil at P & T to restore native wettability
Page 35
Test State
Fresh-State Tests
too oil wet data unreliable data unreliable
Cleaned-State Tests
too water wet (or oil-wet)
Restored-State Tests
Page 36
native wettability restored data reliable (?) if GOR low can use dead crude ageing (cheaper) if GOR high must use live crude ageing (expensive) if wettability restored - use synthetic fluids at ambient ensure cores water-wet prior to restoration
Compare methods - are there differences?
Irreducible Water Saturation (Swir)
Swir essential for reliable waterflood data Dynamic displacement
flood with viscous oil then test oil rapid and can get primary drainage rel perms Swir too high and can be non-uniform
Centrifuge
faster than others Swir can be non-uniform
Porous Plate
slow, grain loss, loss of capillary contact
Page 37
Swir uniform
Lab Variation in Swir (SPE28826)
30
Dynamic Displacement Porous Plate
25
20 Swi (%)
180 psi ???
15
10
200 psi
0 Lab A Lab B Lab C Lab D
Page 38
Centrifuge Tests
Displaced phase relative permeability only
oil-displacing-brine : krw drainage brine-displacing-oil : kro imbibition assume no hysteresis for krw imbibition
oil-wet or neutral wet rocks?
1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 0.0 0.1 0.2 0.3 0.4 0.5 Water Saturation (-) 0.6 0.7 0.8 0.9 1.0
Good for low kro data (near Sro) Computer simulation used Problems
uncontrolled imbibition at Swirr mobilisation of trapped oil sample fracturing
Page 39
Relative Permeability (-)
e.g. for gravity drainage
Dynamic Displacement Tests
Test Methods
Waterflood (End-Points: ko at Swi, kw at Srow) Unsteady-State (relative permeability curves) Steady-State (relative permeability curves)
Test Conditions
fresh state cleaned state restored state ambient or reservoir conditions
Page 40
Unsteady-State Waterflood
Saturate in brine Desaturate to Swirr Oil permeability at Swirr (Darcy analysis) Waterflood (matched viscosity)
o w o = w res lab
Total Oil Recovery kw at Srow (Darcy analysis)
Page 41
Unsteady-State Relative Permeability
Saturate in brine Desaturate to Swirr Oil permeability at Swirr (Darcy analysis) Waterflood (adverse viscosity)
o o >> w lab w res
Incremental oil recovery measured kw at Srow (Darcy analysis) Relative permeability (JBN Analysis)
Page 42
Unsteady-State Procedures
Water Oil Only oil produced Measure oil volume
Just After Breakthrough Measure oil + water volumes
Increasing Water Collected Continue until 99.x% water
Page 43
Unsteady-State
Rel perm calculations require
fractional flow data at core outlet (JBN) pressure data versus water injected
Labs use high oil/water viscosity ratio
promote viscous fingering provide fractional flow data after BT allow calculation of rel perms
Waterflood (matched viscosity ratio)
little or no oil after BT little or no fractional flow (no rel perms) end points only
Page 44
Effect of Adverse Viscosity Ratio
1.0 0.9 0.8 0.7 Fractional Flow, fw 0.6 0.5 0.4 0.3
o/w = 30:1 Unstable flood front Early BT Prolonged 2 phase flow Oil recovery lower o/w = 3:1 Stable flood front BT delayed Suppressed 2 phase flow Oil recovery higher
0.2 0.1 0.0 0.0 0.1 0.2 0.3 0.4 0.5 Water Saturation (-) 0.6 0.7 0.8 0.9 1.0
Page 45
Unsteady-State Tests
Only post BT data are used for rel perm calculations
Sw range restricted if matched viscosities
Advantages
appropriate Buckley-Leverett shock-front reservoir flow rates possible fast and low throughput (fines)
Disadvantages
inlet and outlet boundary effects at lower rates complex interpretation
Page 46
Steady-State Tests
Intermediate relative permeability curves
Saturate in brine Desaturate to Swir Oil permeability at Swir (Darcy analysis) Inject oil and water simultaneously in steps Determine So and Sw at steady state conditions kw at Srow (Darcy analysis) Relative Permeability (Darcy Analysis)
Page 47
Steady-State Test Equipment
p
Oil in
Water in
Mixing Sections
Coreholder
Oil and water out
Page 48
Steady-State Procedures
Summary
100% Oil:
ko at Swirr ko & kw at Sw(1) ko & kw at Sw(2)
Ratio 1: Ratio 2: . . Ratio n:
ko & kw at Sw(n)
100% Water: kw at Sro
Page 49
Steady-State versus Unsteady-State
Constant rate (SS) vs constant pressure (USS)
fluids usually re-circulated
Generally high flood rates (SS)
end effects minimised, possible fines damage
Easier analysis
Darcy vs JBN
Slower
days versus hours
Endpoints may not be representative Saturation Measurement
gravimetric (volumetric often not reliable) NISM
Page 50
Laboratory Tests
You can choose from:
matched or high oil-water viscosity ratio cleaned state, fresh state, restored-state tests ambient or reservoir condition high rate or low rate USS versus SS
Laboratory variation expected
McPhee and Arthur (SPE 28826) Compared 4 labs using identical test methods
Page 51
Oil Recovery
70 Fixed - 120 ml/hour 60 Oil Recovery (% OIIP) Preferred 360
50
40
120
30
120
20 Bump 10 Lab A Lab B Lab C Lab D
Page 52
Gas-Oil and Gas-Water Relative Permeability
Unsteady-State
adverse mobility ratio (g<<o or w) prolonged two phase flow data after breakthrough drainage tests reliable imbibition tests difficult
Steady-State
kg-ko, kg-kw and kw-kg saturation determination difficult much slower
Gas humidified to prevent mass transfer
Page 53
Drainage Gas-Water Curves (steady-state)
Steady-state test example Log-linear scale (very low krw) Krg > krw Gas saturation increases Krg increases to 1 Krw reduces to close to zero
Page 54
Water-Gas Relative Permeability
Aquifer influx (imbibition) Drainage gas-water curves can be used but
hysteresis expected for non-wetting phase (krg) curve no hysteresis for wetting phase (krw) curve
drainage krw curve same shape as imbibition krw curve
Imbibition tests require
low rate imbibition waterflood kw-kg test
capillary forces dominate
CCI tests for residual gas saturation Hybrid test
Page 55
Imbibition Tests
Waterflood
low rate waterflood from Swi to Sgr obtain krg and krw on imbibition Sgr too low (viscous force dominates)
129.90 g
Counter-Current Imbibition Test
Sgr dominated by capillary forces immerse sample in wetting phase (from Sgi) monitor sample weight during imbibition Determine Sgr from crossplot
Page 56
CCI: Experimental Data
Air-T oluene CCI: Plug 10706: Sgi = 88.8%
70 65 60 Gas Saturation (%) 55 50 45 40 35 30 0 10 20 30 Square R T (se c s) oot ime 40 50 60 Sgr = 33.5%
Page 57
Trapped or Residual Gas Saturation
Sgr vs Sgi North Sea
Low rate waterflood
Page 58
Repeatability of CCI tests
Imbibition Kw-Kg
1
krw@Sgr krg 1-Sgr Swi
Drainage
kr
Imbibition
krw 0
Page 59
Sw
Relative Permeability Controls
Wettability Saturation History Rock Texture (pore size) Viscosity Ratio Flow Rate
Page 60
Wettability
Page 61
Wettability
Page 62
Wettability
Waterflood of Water-Wet Rock
Page 63
front moves at uniform rate oil displaced into larger pores and produced water moves along pore walls oil trapped at centre of large pores - snap-off BT delayed oil production essentially complete at BT water invades smaller pores earlier BT oil remains continuous oil produced at low rate after BT krw higher - fewer water channels blocked by oil
Waterflood of Oil-Wet Rock
Effects of Wettability
Water-Wet
better kro lower krw krw = kro > 50% better flood performance poorer kro higher krw kro = krw < 50% poorer flood performance
Oil-Wet
Page 64
Wettability Effects: Brent Field
Preserved Core Neutral to oil-wet low kro - high krw Extracted Core Water wet high kro - low krw
Page 65
Importance of Wettability - Example
Water Wet
No = 2 Nw = 8 Swir = 0.20 Sro = 0.30, krw = 0.25, ultimate recovery = 0.625 OIIP
Intermediate Wet
No = 4 Nw = 4 Swir = 0.15 Sro = 0.25, krw = 0.5, ultimate recovery = 0.706 OIIP
Oil Wet
No = 8 Nw = 2 Swir = 0.10 Sro = 0.20, krw = 0.75, ultimate recovery = 0.778 OIIP
Page 66
o/w = 3:1
Relative Permeability Curves
1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 Water Saturation (-)
WW kro WW krw
Page 67
Relative Permeability (-)
Relative Permeability Curves
1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 Water Saturation (-)
WW kro WW krw IW kro IW krw
Page 68
Relative Permeability (-)
Relative Permeability Curves
1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 Water Saturation (-)
WW kro WW krw IW kro IW krw OW kro OW krw
Page 69
Relative Permeability (-)
Fractional Flow Curves
1.0 0.9 0.8 0.7 Fractional Flow, fw (-) 0.6 0.5 0.4 0.3 0.2 0.1 0.0 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 Water Saturation (-)
WW fw
Water Wet SOR = 0.33 Recovery = 0.59
Page 70
Fractional Flow Curves
1.0 0.9 0.8 0.7 Fractional Flow, fw (-) 0.6 0.5 0.4 0.3 0.2 0.1 0.0 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 Water Saturation (-)
IW SOR = 0.44 Recovery = 0.482
WW fw IW fw
Page 71
Fractional Flow Curves
1.0 0.9 0.8 0.7 Fractional Flow, fw (-) 0.6 0.5 0.4 0.3 0.2 0.1 0.0 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 Water Saturation (-)
WW fw IW fw OW fw
Oil Wet SOR = 0.63 Recovery = 0.300
Page 72
Costs of Wettability Uncertainty
PV Oil Price Parameter Swi Ultimate Sro Ultimate Recovery Factor SOR Actual Recovery Factor STOIIP (MMbbls) Ultimate Recovery (bbls) Actual Recovery (bbls) "Loss" (MM US$) Water-Wet 0.200 0.300 0.625 0.330 0.588 96 60 56 108 120 MMbbls 30 US$/bbls IW 0.150 0.250 0.706 0.440 0.482 102 72 49 684 Oil wet 0.100 0.200 0.778 0.630 0.300 108 84 32 1548
It is really, really important to get wettability right!!!
Page 73
Rock Texture
Page 74
Viscosity Ratio
krw and kro - no effect ? End-Points - viscosity dependent Hence: use high viscosity ratio for curves use matched for end-points
Not valid for neutral-wet rocks (?)
Page 75
Saturation History
100 %
Primary Drainage
Primary Imbibition
NW
No hysteresis in wetting phase NW
kr
Swi W
kr
Sro
W 0% 0%
Page 76
0%
Sw
100 %
0%
Sw
100 %
Flow Rate
Reservoir Frontal Advance Rate
about 1 ft/day
Typical Laboratory Rates
about 1500 ft/day for 1.5 core samples
Why not use reservoir rates ?
slow and time consuming capillary end effects capillary forces become significant c.f. viscous forces Buckley-Leverett (and JBN) invalidated
Page 77
Flow Parameters
End Effect Capillary Number Flood Capillary Number
Nc end
k o vL
Ncend 2.3 0.07 0.02 0.02 0
Nc =
Rate (ml/h) 4 120 360 400 Reservoir
Nc
Rate (ml/h) 4 120 360 400 Reservoir
1.2 x10-7 103.6 x 10-6 101.1 x 10-5 101.2 x 10-5 10-7
For reservoir-appropriate data Nclab ~ Ncreservoir If Ncend > 0.1 kro and krw decrease as Ncend increases
Page 78
Relative Permeabilities are Rate-Dependent
Bump Flood
1.0 0.9 0.8 0.7 0.6 0.5 High Rate krw ??? 0.4 0.3 0.2 0.1 0.0 0.0 0.1 0.2 0.3 0.4 0.5 Water Saturation (-) 0.6 0.7 0.8 0.9 1.0 Low Rate krw' Bump Flood krw'
Page 79
Relative Permeability (-)
Flow Rate Considerations
Imbibition (waterflood of water-wet rock)
Sro function of Soi: Sro is rate dependent oil production essentially complete at BT krw suppressed by Pcend and rate dependent bump flood does not produce much oil but removes Pcend and krw increases significantly high rates acceptable but only if rock is homogeneous at pore level
Considerations
ensure Swi is representative low rate floods for Sro: bump for krw steady-state tests
Page 80
Flow Rate Considerations
Drainage (Waterflood of Oil-Wet Rock)
end effects present at low rate Sro, krw dependent on capillary/viscous force ratio high rate: significant production after BT reduced recovery at BT compared with water-wet
Considerations
high rate floods (minimum Dp = 50 psid) to minimise end effects steady-state tests with ISSM low rates with ISSM and simulation
Page 81
Flow Rate Considerations
Neutral/Intermediate
Sro and kro & krw are rate dependent bump flood produces oil from throughout sample, not just from ends ISSM necessary to distinguish between end effects and sweep
Recommendations
data acquired at representative rates (e.g. near wellbore, grid block rates)
Page 82
JBN Validity
High Viscosity Ratio
viscous fingering invalidates 1D flow assumption
Low Rate
end effects invalidate JBN
Most USS tests viewed with caution
if Ncend significant if Nc not representative if JBN method used
Use coreflood simulation
Page 83
Test Recommendations
Wettability Conditioning
flood rate selected on basis of wettability Amott and USBM tests required Wettability pre-study
reservoir wettability? fresh-state, cleaned-state, restored-state wettabilities
beware fresh-state tests (often waste of time) reservoir condition tests most representative
but expensive and difficult
Page 84
Wettability Restoration
Hot soxhlet does not make cores water wet! Restored-state cores too oil wet Lose 10% OIIP potential recovery
USBM
0.0 1.0 STRONGLY WATER-WET
STRONGLY OIL-WET -1.0 -1.0 0.0
Original SCAL plugs Hot Sox Cleaned Flush Cleaned
1.0
Amott
Page 85
Key Steps in Test Design
Establishing Swi
must be representative use capillary desaturation if at all possible
remember many labs cant do this correctly
fresh-state Swirr is fixed
Viscosity Ratio
matched viscosity ratio for end-points investigate viscosity dependency for rel perms normalise then denormalise to matched end-points
Page 86
Key Steps In Test Design
Flood Rate
depends on wettability determine rate-appropriate end-points steady-state or Corey exponents for rel perm curves
Saturation Determination
conventional
grain loss, flow processes unknown
NISM
can reveal heterogeneity, end effects, etc
Page 87
Use of NISM
Examples from North Sea Core Laboratories SMAX System
low rate waterflood followed by bump flood X-ray scanning along length of core end-points some plugs scanned during waterflood
Fresh-State Tests
core drilled with oil-based mud
Page 88
X-Ray Scanner
Coreholder (invisible to Xrays)
X-rays detected X-rays emitted
Scanning Bed
X-ray Emitter (Detector Behind)
Page 89
X-ray adsorption
0 %
Sw(NaI)
100%
NISM Flood Scans
SMAX Example 1
uniform Swirr oil-wet(?) end effect bump flood removes end effect some oil removed from body of plug neutral-slightly oil-wet
Page 90
NISM Flood Scans
SMAX Example 2
short sample end effect extends through entire sample length significant oil produced from body of core on bump flood moderate-strongly oil-wet data wholly unreliable due to pre-dominant end effect. Need coreflood simulation
Page 91
NISM Flood Scans
SMAX Example 3
scanned during flood minimal end effect stable flood front until BT
vertical profile
bump flood produces oil from body of core neutral wet data reliable
Page 92
NISM Flood Scans
SMAX Example 4
Sample 175 (fresh-state) scanned during waterflood unstable flood front
oil wetting effects
oil-wet end effect bump produces incremental oil from body of core but does not remove end effect neutral to oil-wet
Page 93
data unreliable
NISM Flood Scans
SMAX Example 5
Sample 175 re-run after cleaning increase in Swirr compared to fresh-state test no/minimal end effects moderate-strongly waterwet
Page 94
NISM Flood Scans
SMAX Example 6
heterogeneous coarse sand variation in Swirr Sro variation parallels Swirr end effect masked by heterogeneity (?) very low recovery at low rate (thiefzones in plug?) bump flood produces significant oil from body of core neutral-wet
Page 95
Key Steps in Test Design
Relative Permeability Interpretation
key Buckley-Leverett assumptions invalidated by most short corefloods
Interpretation Model must allow for:
capillarity viscous instability wettability
Simulation required
e.g. SENDRA, SCORES
Page 96
Simulation Data Input
Flood data (continuous)
injection rates and volumes production rates differential pressure
Fluid properties
viscosity, IFT, density
Imbibition Pc curve (option) ISSM or NISM Scans (option) Beware several non-unique solutions possible
Page 97
History Matching
Pressure and production
1.66 cc/min
800 Differential Pressure (kPa) 700 600 500 400 300 200 100 0 0,1 1,0 10,0 100,0 Time (min) 1000,0 Measured differential pressure Simulated differential pressure Measured oil production Simulated oil production 1,0 2,0 4,0 Oil Production (cc) 6,0
5,0
3,0
0,0 10000,0
Page 98
History Matching
Saturation profiles
0.8 0.7
Water Saturation
0.6 0.5 0.4 0.3 0.2 0.0 0.2 0.4 0.6 0.8 1.0
Page 99
Normalized Core Length
Simulation Example JBN Curves
Relative Permeabilty Curves Pre-Simulation
1 0.9 0.8
Relative Permeability
0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Krw Kro low rate end point high rate end point
Water saturation
Page 100
Simulation Example Simulated Curves
Relative Permeabilty Curves Post Simulation
1 0.9 0.8
Relative Permeability
0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Krw Kro low rate end point high rate end point Krw Simulation Kro Simulation
Water saturation
Page 101
Quality Control
Most abused measurement in core analysis Wide and unacceptable laboratory variation Quality Control essential
test design detailed test specifications and milestones contractor supervision modify test programme if required
Benefits
better data more cost effective
Page 102
Water-Oil Relative Permeability Refining
Key Steps
curve shapes Sro determination and refinement refine krw determine Corey exponents refine measured curves normalise and average
Uses Corey approach
rock curves may not obey Corey behaviour
Page 103
Curve Shapes
1 0.9 0.8 0.7 0.6
Semi-log Good data concave down
Kro Krw
Kr
0.5 0.4 0.3 0.2 0.1 0
Water-Oil Rel. Perms.
1
0 0.2 0.4 Sw 0.6 0.8 1
0.1
Kr
Cartesian Good data convex upwards
0.01
Kro Krw
0.001
0.0001 0 0.2 0.4 Sw 0.6 0.8 1
Page 104
Sro Determination
Compute Son high, medium and low Sro
1
low rate, bump, centrifuge Sro
0.1 Sor = 0.40 Sor = 0.20 Sor = 0.35
Kro
Plot Son vs kro (log-log)
0.01
Sro too low
0.001
curves down Sro too high curves up Sro just right straight line
Page 105
1.000 0.100
Son = (1-Sw-Sor)/(1-Swi-Sor)
0.0001 0.010
Refine krw
Refined krw
Use refined Sro Plot krw versus Swn Fit line to last few points
Krw 0.1 1
least affected by end effects Determine refined krw
0.01 0.1 Swn = 1-Son 1
Page 106
Determine Best Fit Coreys
Use refined Sro and krw Determine instantaneous Coreys
No' & Nw'
3.5 3 2.5 2 1.5 1 0.5 0
log(krw' ) log(krw) Nw* = log(1.0) log(S wn )
No Nw
log(kro ) No* = log(Son )
Plot vs Sw Take No and Nw from flat sections Least influenced by end effects
0.2
0.4 Sw
0.6
0.8
Page 107
Refine Measured Data
Endpoints Refined krw and Sro Corey Exponents No and Nw (stable) Corey Curves
Relative Permeability 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 0 0.1 0.2 0.3 0.4 0.5 Sw 0.6 0.7 0.8 0.9 1 Refined Kro Refined Krw Original Kro Original Krw
kro( refined ) = Son
No
krw( refined ) = krw' Swn
Page 108
Nw
Normalisation Equations
Water-Oil Data
Sw Swi Swn = 1 Swi Srow
k ro n =
k ro k ro end
krw krwn = krwend
Gas - Oil Data
Sgn = Sg Sgc 1Swi SrogSgc
k ro n =
k ro k ro end
krgn =
krg krgend
Page 109
Example - kro Normalisation
1 0.9 0.8
Oil Relative Permeability (-)
0.7 0.6 0.5 Swirr Swn = 0 0.4 0.3 0.2 0.1 0 0 0.1 0.2 0.3 0.4 0.5 Water Saturation (-) 0.6 0.7 0.8 0.9 1 Sw = 1-Sro Swn = 1
Sample 1 Sample 2
Page 110
Example - krw Normalisation
1 0.9 0.8
Water Relative Permeability (-)
0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 0 0.1 0.2 0.3 0.4 0.5 Water Saturation (-) 0.6 0.7 0.8 0.9 1 krw at Sro krwn = 1
Sample 1 Sample 2
Page 111
Normalise and Compare Data - kron
1.0 0.9 Normalised Oil Relative Permeability (-) 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 Normalised Water Saturation (-) Steady State Different Rock Types ? Different Wettabilities?
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Page 112
Normalise and Compare Data - krwn
1.0 0.9 Normalised Water Relative Permeability (-) 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 Normalised Water Saturation (-)
1 2 3 4 5 6 7 8 9 11 12 13 14 15
Page 113
Denormalisation
Group data by zone, HU, lithology etc Determine Swir (e.g. logs, saturation-height model) Determine ultimate Sro
e.g. from centrifuge core tests
Determine krw at ultimate Sro
e.g. from centrifuge core tests
Denormalise to these end-points Truncate denormalised curves at ROS
depends on location in reservoir
Page 114
Denormalisation Equations
Water Oil
S w dn = S wn (1 S wi S ro ) + S wi k rodn = k ro end .k ron k rwdn = k rw end .k rwn
Gas-Oil S = S (1 S S S ) + S g dn gn wi rog gc gc
Denormalised Endpoints Water-Oil Swi kro (@Swi) krw (@1-Srow) From correlations & average data
k rodn = koend .k ron k rgdn = k rg end .k rgn
Page 115
Summary Getting the Best Rel Perms
Ensure samples are representative of poro-perm distribution Ensure Swir representative (e.g. porous plate, centrifuge) Ensure representative wettability (restored-state?) Use ISSM (at least for a few tests) Ensure matched viscosity ratio Low rate then bump flood Centrifuge ultimate Sro and maximum krw
Tail ok kro curve if gravity drainage significant
Use coreflood simulation or Coreys for intermediate kr
Page 116
