Page 1
[email protected] 1
Effects of Solid-State and Pore-Fluid Chemistry and Stress on Permeability Evolution
Derek Elsworth (Penn State) and Josh Taron (USGS)
Basic Observations of Permeability Evolution – EGS and SGRsKey Issues in EGS and SGRs
Spectrum of Behaviors EGS to SGRHomogeneous Permeability Flow Modes
Diagenesis Permeability Evolution
Basin EvolutionStimulation and Production
Scaling Relations in Rocks and ProppantsReinforcing Feedbacks
Induced SeismicityMineralogical Transformations – Seismic -vs- AseismicFirst- and Second-Order Frictional Effects
Key Issues
Page 2
[email protected] 2
Basic Observations of Permeability Evolution
Challenges• Prospecting (characterization) • Accessing (drilling)• Creating reservoir• Sustaining reservoir• Environmental issues
Observation• Stress-sensitive reservoirs• T H M C all influence via effective stress• Effective stresses influence
• Permeability• Reactive surface area• Induced seismicity
Understanding T H M C is key:• Size of relative effects of THMC(B)• Timing of effects• Migration within reservoir• Using them to engineer the reservoir
Permeability
Reactive surface area
Induced seismicity
Resource• Hydrothermal (US:104 EJ) • EGS (US:107 EJ; 100 GW in 50y)
Page 3
[email protected] 3
Key Questions in EGS and SGRs
Needs• Fluid availability
• Native or introduced• H20/CO2 working fluids?
• Fluid transmission • Permeability microD to mD?• Distributed permeability
• Thermal efficiency• Large heat transfer area• Small conduction length
• Long-lived• Maintain mD and HT-area• Chemistry
• Environment• Induced seismicity• Fugitive fluids
• Ubiquitous
[Ingebritsen and Manning, various, in Manga et al., 2012]
Page 4
[email protected] 4
Contrasts Between EGS & SGRs
EGS (Order of Mag.)
Property ESRs (Order of Mag)
Fractured-non-porous General Porous-fractured
<<1%,<1% Porosity, n0 -> nstim ~10-30%, ~same
microD -> mD Permeability, k0 -> kstim
>mD -> >mD
106 Kf/kmatrix 106 ->1
10-100m Heat transfer length, s
1m -> 1cm
>>100/1. >100/1 *Heatsolid/Heatfluid ~10/1-2/1, same
? Chemistry ?
V. Strong TM Perm. Feedbacks Less strong
Moderate, late time TC Perm. Feedbacks Strong?
Page 5
[email protected] 5
Thermal Drawdown EGS –vs- SGRs
Wate
r Tem
p
(at
outl
et)
Rock
Tem
p
(in r
ese
rvoir
)
Thermal Output:
In-Reservoir Water Temperature Distributions:
Page 6
[email protected] 6
Thermal Recovery at Field ScaleParallel Flow Model Spherical Reservoir
Model
Tinjection
Dim
ensi
onle
ss
tem
pera
ture
Dimensionless time Dimensionless time
Trock
[Elsworth, JVGR, 1990]
[Gringarten and Witherspoon, Geothermics,1974] [Elsworth, JGR, 1989]
[Note: not linear in log-time]
Spacing, s, is small
Spacing, s, is large
Page 7
[email protected] 7
What Does This Mean?
This makes the case that:
Permeability needs to be large enough to allow Mdot_sufficient without:
1. Fracturing reservoir during production
2. Large pump costs
Beyond that – issues of heterogeneity are imp:
1. No feedbacks (Rick)
2. Reinforcing feedbacks (Kate/Paul/Golder/Gringarten)
Diagenesis contributes to this:
1. Initial basin evolution [k0,n0]
2. Reservoir stimulation/development [k,n=f(t)]
3. Reinforcing feedbacks [k,n=f(x,t)] for THMC
Page 8
[email protected] 8
Effects of Solid-State and Pore-Fluid Chemistry and Stress on Permeability Evolution
Derek Elsworth (Penn State) and Josh Taron (USGS)
Basic Observations of Permeability Evolution – EGS and SGRsKey Issues in EGS and SGRs
Spectrum of Behaviors EGS to SGRHomogeneous Permeability Flow Modes
Diagenesis Permeability Evolution
Basin EvolutionStimulation and Production
Scaling Relations in Rocks and ProppantsReinforcing Feedbacks
Induced SeismicityMineralogical Transformations – Seismic -vs- AseismicFirst- and Second-Order Frictional Effects
Key Issues
Page 9
[email protected] 9
Controls on Reservoir Evolution
Many processes of vital importance to EGS/SGR are defined by coupled THMC processes.Thermal sweep/fluid residence timeShort circuitingInduced seismicityProlonged sustainability of fluid transmission
Fractures dominate the fluid transfer systemTransmission characterized by:
History of mineral depositionChemo-mechanical creep at contacting asperitiesMechanical compactionShear dilation and the reactivation of relic fractures
Page 10
[email protected] 10
Typical Response of Fractures (Dissolution)
[Polak et al., GRL, 2003]
m
Page 11
[email protected] 11
Typical Response of Fractures (Precipitation)
[Dobson et al., 2001]
Experimental arrangement
Precipitation
Thermal gradient along fracture
Page 12
Dissolution Processes Approaches to Determine k or b
Time
b
precipitation
diffusion
dissolution
grain
grain
Time
b
precipitation
diffusion
dissolution
precipitation
diffusion
dissolution
grain
grain
Time
Page 13
Component Model
•Interface Dissolution
•Interface Diffusion
•Pore Precipitation eqporeporeprec
CCkM
AV
dt
dM
dx
dCDJ b 2
c
m br d
dCJ r D
dr
porec
bdiffm CC
ad
D
dt
dMJ
int
2ln
2
TR
dkV
dd
dt
dM
cgeffm
cgdissdiss
4
3
422
2
TR
dkV
dt
dM cgcamdiss
4
3 22
1
4
mm
cm
TE T
V
[Yasuhara et al., JGR, 2003]
Page 14
Matching Compaction Data
[Experimental data from Elias and Hajash, 1992]
Page 15
System Evolution at 35-70 MPa and 150°C
Observation Extension
70 MPa and 150°C 35 MPa and 150°C
[Experimental data from Elias and Hajash, 1992]
[Yasuhara et al., JGR, 2003]
Page 16
Timescales of Evolution of Granular Systems at 35 MPa and 75-150°C
75°C 150°C
[Yasuhara et al., JGR, 2003]
Page 17
Permeability Evolution in Granular Systems at 35 MPa and 75-300°C
150°C
300°C
75°C
2
Capillary Model :96
nk
20Pore Evolution: ( / 4)pV d
0
Linked Permeability: ~24
pnVk
d[Yasuhara et al., JGR, 2003]
Page 18
[email protected] 18
Fracture/Proppant Diagenesis
Page 19
Do we understand the mechanisms?
Various mechanisms – appear complex but include:• Dissolution/precipitation • Solid and aqueous chemical transformations• Fluid/chemical assisted strength loss of proppant
and proppant collapse
2 4 6 8 10 12 140.000
0.005
0.010
0.015
0.020
0.025
0.030
L (
inch
es)
Time (days)
250F 275F 325F 350F
Experiment
Observation
Characterization
Analysis
[Dae Sung Lee et al., 2009]
Page 20
g3.ems.psu.edu 20
THMC/HPHT Continuum Models
THMC-S – Linked codes Spatial Permeability Evolution
Temporal Permeability Evolution
Page 21
g3.ems.psu.edu 21
Constraint on Fracture Apertures and Fluid Concentrations
(a)
(b)
(c)
max 0[( ) / ]r c cb b b Exp R R a
maxb
rb
Asperity contacts
Local contact
area, Alc
dc
Incre
asin
g f
ractu
re
clo
su
re
Page 22
g3.ems.psu.edu 22
Modeling Results - Novaculite
[Yasuhara et al., JGR, 2004]
K+~x300
Page 23
g3.ems.psu.edu 23 [Yasuhara et al., JGR, 2004]
Projected Response of Fracture
Define projected behavior for varied temperatures
….and mean stress magnitudes
Page 24
Reactive - Hydrodynamic ControlsPeclet No. (Pe)
0Advective flux
Dispersive flux m m
q vbPe
D D
Damkohler No. (Da)
Pe < 1 Dispersion dominated – Perturbations damped
Pe > 1 Advection dominated – Perturbations enhanced
0
2 2Reactive flux
Advective flux
k L k LDa
q vb
Da << 1 Reaction slow - Undersaturated along fracture – Perturbations damped
Da larger << 1 – Reaction fasterSaturated along fracture – Perturbations enhanced
2Reactive flux.
Dispersive flux m
k LPe Da
D
PeDa No. (Removes <q>)
[Sherwood No.]
Page 25
Reactive Hydrodynamics: Role of Damkohler Number (PeDa)
[Detwiler and Rajaram, WRR, 2007]
15 cm x 10cmVoxel = 1 mmAperture:Black (0)-White(0.25mm) Time
High PeDa
Low PeDa
Page 26
[email protected] 26
Effects of Solid-State and Pore-Fluid Chemistry and Stress on Permeability Evolution
Derek Elsworth (Penn State) and Josh Taron (USGS)
Basic Observations of Permeability Evolution – EGS and SGRsKey Issues in EGS and SGRs
Spectrum of Behaviors EGS to SGRHomogeneous Permeability Flow Modes
Diagenesis Permeability Evolution
Basin EvolutionStimulation and Production
Scaling Relations in Rocks and ProppantsReinforcing Feedbacks
Induced SeismicityMineralogical Transformations – Seismic -vs- AseismicFirst- and Second-Order Frictional Effects
Key Issues
Page 27
[email protected] 27
Triggered Seismicity – Key Questions
Principal trigger - change in (effective) stress regime:Fluid pressureThermal stress Chemical creep
How do these processes contribute to:Rates and event size (frequency-magnitude)Spatial distributionTime history (migration)
How can this information be used to: Evaluate seismicityManage/manipulate seismicityLink seismicity to permeability evolution
Reservoir Conditions:
THMC Model:
Page 28
[email protected] 28
Observations of Induced Seismicity (Basel)
[Goertz-Allmann et al, 2011] [Shapiro and Dinske, 2009]
Page 29
[email protected] 29
r-t Plot - Fluid and Thermal Fronts and Induced Seismicity
Parameters utilized in simulation
k0 Permeability[m2] 10-17
Pp Pore Pressure[Mpa] 14.8
Pinj Fluid Pressure[Mpa] 17.8
Tres Reservoir Temperature[°c]
250
Tinj Fluid Temperature[°c] 70
S Fracture Spacing[m] 10 to 500
3
0
2,
12
Qt br k
h S
Q: Flow rate
t : Time
h: Thickness
ϕ: Porosity
b: Aperture
[Izadi and Elsworth, in review, 2013]
Page 30
g3.ems.psu.edu 30
Fault Reactivation (and Control)
FaultInjection well
Controls on Magnitude and Timing:kfault & kmedium
[10-16 – 10-12 m2]Injection temperature dT [50C – 250C]Stress field obliquity [45-60 degrees]
Permeability & Magnitude Timin
g
[Gan and Elsworth, in review, 2013]
Page 31
[email protected] 31
Seismic –vs- Aseismic Events
[Peng and Gomberg, Nature Geosc., 2010]Seismic Moment (N.m) [Magnitude]
Dura
tion (
s) [
secs
->
years
]
Page 32
[email protected] 32
Approaches – Rate-State versus Brittle Behavior
Rate-State Brittle
µ0
(a-b)ln(v/v0)
a ln(v/v0)
DC
Low velocity
High velocity
Low velocity
Displacement
Coe
ffici
ent o
f fric
tion
Syste
m S
tiffn
ess
(Sto
red E
nerg
y)
Failu
re C
riterio
n
(Trig
ger)
-b ln(v/v0)
Page 33
[email protected] 33
Seismic –vs- Aseismic Events
[Ikari et al., Geology, 2011]
Friction
Stability (a-b)
Velocity Weakening (unstable slip)
Velocity Strengthening (stable slip)
Page 34
Scale Effects in Hydrology – Space and Time
[Elkhoury et al., Nature, 2006]
• Remote earthquakes trigger dynamic changes in permeability
• Unusual record transits ~8y• Sharp rise in permeability followed
by slow “healing” to background• Scales of observations:
– Field scale
– Laboratory scale
– Missing intermediate scale with control
Per
mea
bilit
y
Pe
rme
ab
ilit
y
Page 35
[email protected] 35
Role of Wear ProductsSample Holder
Sample
Shear-Permeability Evolution
Dissolution Products
[Faoro et al., JGR, 2009]
Page 36
[email protected] 36
Key Questions in EGS and SGRs Needs• Fluid availability
• Native or introduced – fluid/geochemical compatibility• H20/CO2 working fluids? – arid envts.
• Fluid transmission • Permeability microD to milliD? – high enough?• Distributed permeability
• Characterizing location and magnitude• Defining mechanisms of perm evolution (chem/mech/thermal)• Well configurations for sweep efficiency and isolating short-circuits
• Thermal efficiency• Large heat transfer area – better for SGRs than EGS?• Small conduction length – better for SGRs than EGS?
• Long-lived• Maintain mD and HT-area – better understanding diagenetic effects?• Chemistry - complex
• Environment• Induced seismicity - Event size (max)/timing/processes (THMCB)• Fugitive fluids – Fluid loss on production and environment – seal integrity
• Ubiquitous