laboratory investigation of supercritical co2 use in
TRANSCRIPT
PROCEEDINGS, Thirty-Eighth Workshop on Geothermal Reservoir Engineering
Stanford University, Stanford, California, February 11-13, 2013
SGP-TR-198
LABORATORY INVESTIGATION OF SUPERCRITICAL CO2 USE IN GEOTHERMAL
SYSTEMS
Mario Magliocco1,2
, Steven Glaser1,2
, and Timothy J. Kneafsey2
1Department Civil and Environmental Engineering
University of California, Berkeley, CA 94720 2Lawrence Berkeley National Laboratory
Berkeley, CA 94720
e-mail: [email protected]
ABSTRACT
The use of carbon dioxide (CO2) as a heat transfer
fluid has been proposed as an alternative to water in
enhanced geothermal systems (EGS). Numerical
simulations have shown that under expected EGS
operating conditions, CO2 could achieve more
efficient heat extraction performance than water.
The simulations indicate that the performance
advantage of CO2 over water is a function of
operating parameters, reservoir temperature, and flow
geometry. Since the pore space in formations that
would be likely utilized for EGS are initially
saturated with water or brine, the creation of a CO2
based EGS reservoir would involve a development
period where the resident water would be removed
over time as CO2 is injected into the formation.
Using a specially constructed laboratory apparatus
that is capable of flowing temperature-controlled
supercritical CO2 through a heated porous sample, we
have investigated the behavior of CO2 as a working
fluid in EGS system. In our laboratory based
experiments we have investigated the injection of
CO2 into a heated water saturated sample and have
found that preferential flow paths develop which
bypassed much of the sample pore space.
Additionally we have also explored the behavior of
dry CO2 based heat extraction from a heated sample
and found that the behavior is highly dependent on
the operating conditions. Finally we have explored
the relative performance of dry CO2 versus single-
phase water based heat extraction and found that
under the operating conditions studied that water and
CO2 exhibit similar heat extraction rates.
These experiments are intended create a data set that
can then be used to validate numerical modeling
results and provide insight into the planning,
development, and operation of a CO2 based EGS
reservoir.
INTRODUCTION
The novel concept of using supercritical CO2 (scCO2)
as the working fluid in an enhanced geothermal
system (EGS) for both reservoir creation and heat
extraction was first proposed by Brown (2000). The
advantages of using CO2 instead of water as the
process fluid in a closed loop EGS system include a
much lower viscosity of CO2 resulting in
substantially larger mass flow rates for a given
pressure drop between injection and production
points, and a much larger density difference between
cold fluid in the injection well and hot fluid in the
producer providing increased buoyancy forces for
CO2. As an ancillary benefit, practical operation of a
CO2 system would result in some de facto carbon
sequestration due to fluid loss into the surrounding
formations (Brown, 2000, Pruess 2006). Numerical
simulations of a five-spot well pattern in a hot dry
rock system indicate that (1) CO2 achieves larger heat
extraction rates than water, and (2) the relative
advantage of CO2 increases with decreasing reservoir
temperature, ranging from approximately 50% larger
heat extraction rates at T = 240oC to about 80%
larger rates at T = 120oC (Pruess personal
communication 2009).
The increasing heat extraction efficiency of CO2 at
lower reservoir temperatures can be better understood
by considering the effects of pressure and
temperature on the “fluid mobility” m = ρ/μ (ρ is the
fluid density, μ is the fluid viscosity) (Figure 1, Press
2006).
Figure 1 - Fluid mobility of CO2 (Preuss 2006)
The contour lines in Figure 1 indicate the fluid
mobility for the pressure and temperature shown on
the axes. According to Darcy’s law, for a given
pressure gradient, fluid flux is proportional to the
fluid mobility. CO2 mobility exhibits a much more
dynamic dependence on temperature and pressure
than water, and under the conditions expected to be
present in a geothermal field, the CO2 mobility is
much higher than that of water.
In a water-based 5-spot EGS system the majority of
the driving head loss occurs near the injection well.
This is because the lower mobility resulting from the
colder conditions coupled with the high Darcy flux
due to the radial flow pattern around the well require
a high driving pressure gradient. This is in contrast
to the higher mobility CO2-based system where the
head loss is much more evenly distributed across the
entire flow path.
Typically, the pore space in formations that would be
likely utilized for EGS is initially saturated with
water or brine. The creation of a CO2 based EGS
reservoir in an initially brine-saturated reservoir
would involve a development period where the
resident water would be removed over time as CO2 is
injected into the formation. Simulations have shown
that water removal would occur through two
principal processes: (1) immiscible displacement of
aqueous phase by the CO2-rich phase, and (2)
dissolution of water into the flowing CO2 stream
(Pruess 2010). During the operation of a CO2 based
EGS reservoir, it is expected that the initial
production fluid would be composed of the native
brine until the breakthrough of CO2 occurs, after
which a brine-CO2 mixture would be produced.
In order to verify the results of the numerical
simulations, we have performed laboratory
experiments that have injected cooled CO2 into a
heated water saturated sample to study the reservoir
development stage. We have also injected cooled
CO2 into a heated CO2-saturated sample to study the
behavior of a fully developed, dry reservoir. We
have also injected cooled water into a heated water
saturated sample in order to compare the behavior of
a water-based system to a CO2 based system.
EXPERIMENT DESCRIPTION
Experimental Apparatus
The apparatus consists of a temperature-controlled
pressure vessel filled with porous media through
which temperature controlled fluid can be introduced
by means of high-pressure, high-flow rate pumps
(Figure 2). The pumps can be operated to provide a
constant fluid injection rate, or a constant differential
pressure. The fluid was delivered by a pair of Quizix
C-6000-5K pumps, capable of 5,000psi (345 bar) and
400 ml/min fluid delivery rate. The pumps are
capable of precisely controlled continuous and pulse-
Figure 2 - Simplified schematic of the experimental apparatus.
free flow with a resolution of 27.2 nanoliters. To
ensure that the pumps are filled with high-density
liquid CO2, the injection fluid is passed through a
chiller before entering the pumps. After leaving the
pumps, the fluid can be either chilled by a second
fluid chiller, or heated depending on the desired
experimental parameters. Before entering the vessel,
the injection fluid passes through a Siemens coriolis-
style mass flow meter. The meter could also be
placed at the outlet and used in conjunction with the
pump flow readings to measure fluid accumulation in
the vessel.
The pressure vessel is a hollow stainless steel
cylinder with an inside diameter of 9.1 cm, outside
diameter of 12.7 cm, 50.8 cm distance between the
end caps, and a pressure safety rating of 34.5 MPa
(345 bar, 5000 psi). Instrumentation access to the
interior of the vessel is through three axial passages
through one end cap, and one passage through the
other. The central passages through the end caps are
used as the injection and production ports and the
remaining two passages are used exclusively to pass
thermocouples through. The vessel is oriented
vertically such that the central axis is in line with
gravity. It has been shown that even for small length
scales, buoyant effects of scCO2 can have a large
effect on the dynamics of a scCO2-based system
(Liao and Zhao 2002). For a horizontal flow
arrangement, buoyant forces can result in pressure
gradients that are oriented perpendicular to the vessel
axis, complicating the dynamics. For modeling and
comparison purposes, the experiments were operated
such that the flow path was in the same orientation as
the gravity-induced pressure gradient.
Temperature measurements within the sample were
made with 23 stainless-steel clad type-T
thermocouples, which have a small diameter
(0.79mm) in order to increase the sensor response
time and to minimize disturbance to fluid flow. The
thermocouples are arranged at various elevations and
radii in the sample such that each successive vertical
level is offset angularly to minimize vertical sensor
shadowing (Figure 3). The offset angle we used is
based on the “Golden Angle” (137.5 degrees), found
in plant phyllotaxis that has been shown to minimize
shadowing (King 2004).
At one elevation in the porous medium, two
thermocouples were mirrored so that they were both
at the same radial distance from the central axis of
the vessel. The radially mirrored temperature
measurement was designed to test our assumption of
a radial symmetry in the heat transfer process.
Because the end caps are large, the injection port of
the vessel was lined with a length of nylon tubing
through the end cap in order to provide thermal
insulation for the injected fluid as it passed through
the relatively massive end cap. The injection port
was also fitted with a single thermocouple mounted
where the injected fluid enters the sample space (not
shown in Figure 2).
The sand used in the test sample was prepared from
F95 Ottawa silica sand (U.S. Silica). Sieving and
washing resulted in a narrow grain size distribution
(Figure 4). The mean grain size falls between 147
and 105 microns with no measurable portion below a
grain size of 45 microns.
Figure 3 - Orthographic diagram of thermocouple
placement inside the vessel. Axis units
are in meters. Colored dots index the
different elevations and correspond to the
colors in Figures 5, 6, & 8
Figure 4- Porous medium grain size distribution
The sand was dry placed in the vessel in multiple lifts
with vibratory compaction between lifts. This
method produced a relative density of 84 percent.
The porous core sample properties are listed in Table
1.
Table 1: Design case system properties
The vessel was wrapped with heat tape that extended
around the exterior of the cylinder and both end caps.
The heat tape thermal output was controlled by either
a PID controller using a thermocouple secured on the
vessel exterior or with a pulse width modulated
signal. This allows the vessel boundary to maintain a
constant temperature or a constant heat flux. Finally
the vessel was wrapped in an aerogel insulation
jacket and sealed. The current supplied to the heat
tape was monitored with a true RMS current sensor.
The pressure at the outlet of the vessel was controlled
by a pair of digital backpressure regulators in series.
The fluid exiting the backpressure regulators was
vented to the atmosphere at a safe location outside of
the building. A differential pressure sensor
connecting the inlet and outlet of the vessel was
located at the base of the vessel. The tubing that
connected the differential pressure sensor to the inlet
is encased in a constant temperature water bath so
that the state of the fluid column in the tube could be
determined.
We developed software that incorporates
experimental control and data acquisition. All sensor
readings were collected by a single Labview-based
program that allows for accurate time
synchronization of experimental data. The program
is capable of controlling the pumps, vessel heat input,
and the backpressure regulators. Combining these
functions allows for a tightly integrated experimental
setup, faster data processing, faster experimental
turnaround time, and less chance of experimental
errors.
Table 2 lists the range of operating parameters that
we have used in our experiments. Our current
apparatus is capable of achieving temperatures up to
200C and operating pressure of up to 275 bar.
Table 2: Range of operating parameters
Parameter Min Value Max Value
back pressure 87 bar 138 bar
initial core
temperature 50 C 100C
injection rate 50 ml/min 200 ml/min
Experimental Procedure
The sand-filled vessel was filled with fluid and
pressurized to the experiment pressure, and the vessel
was heated. For the single-phase CO2 experiments,
the vessel was filled with CO2, for the two phase
CO2/water and the single phase water experiment the
vessel was initially filled with water. The pumps and
tubing were then emptied of water if present, filled
with CO2, and pressurized to the vessel pressure. The
backpressure regulators were set to the desired
pressure, and CO2 injected into the bottom of the
vessel at a prescribed volumetric flow rate. The
pumps maintain a constant flow over multiple
injector volumes through computer controlled pumps
switching.
Experimental Challenges
During the course of our experiments we have met
with several practical challenges that are related to
CO2. The density of CO2 can change significantly
with relatively small changes in pressure or
temperature. For example, the density of the CO2 in
the vessel could change by more than a factor of two
within the conditions of a single experimental run.
Practically, this behavior made it difficult to measure
the mass of the CO2 without knowledge of the current
state of the CO2.
When using the first iteration of our apparatus we
inferred the mass flow rate of fluid entering the
vessel by recording the volumetric flow rate of the
0102030405060708090
0.147 0.105 0.075 0.053 0.045
% R
eta
ine
d
Mesh Opening Size (mm)
Porous Media Grain Size Distribution
Porous Core Properties
total core length L = 50.8 cm
cross sectional area A = 6.54x10-3
m2
grain density ρR = 2650 kg/m3
grain specific heat CR = 20 J/ g/ C
rock thermal
conductivity K = 2.51 /m/ C
permeability k = 9.3x10-13
m2
porosity ϕ = 41%
mean grain size d50 d50 ≈ 0.105 mm
pump, measuring the pressure and temperature of the
CO2 exiting the pump, using a lookup table to find
the density of CO2, and finally calculating the mass
flow rate. This method proved difficult as the
pressure and temperature of the fluid in the pump
changed throughout the course of a single
experimental run. To solve this problem we
employed a Siemens corriolis based mass flow meter
near the inlet of the pressure vessel. This device was
capable of measuring the mass flow of the fluid
without requiring knowledge of the current state of
the fluid being measured.
The dynamic density of the CO2 also made measuring
the pressure differential across the vessel difficult.
The differential-pressure sensor we used in our
apparatus required a tubing connection between the
inlet and outlet of the vessel so that the relative
pressure difference could be measured. Since our
vessel was mounted vertically, in line with gravity,
the weight of the fluid in the tubing connecting the
delta-pressure sensor with the vessel outlet resulted in
a hydrostatic pressure component on one side of the
delta pressure sensor. With fluids such as water that
don’t exhibit drastic density changes, the hydrostatic
pressure due to the fluid column in the connection
tube is relatively constant and can simply be
subtracted from the reading. With CO2 we found that
small changes in pressure and temperature in the
column produced relatively large changes in density
that caused the hydrostatic pressure to change
depending on the state of the CO2 in the connection
tube.
In the second iteration of our experimental apparatus
we attempted to impose a constant temperature on the
connection tube by using a small diameter tube and
placing it in a relatively large temperature controlled
water bath. The temperature of the bath was recorded
by a thermocouple throughout the experiment, and a
pressure sensor was located at the top end of the tube.
Using these measurements, the state of the fluid in
the column could be determined and used to calculate
the hydrostatic pressure component and subtract it
from the differential pressure measurement.
The final difficulty we experienced due to the
behavior of CO2 was in our backpressure regulation.
We used a pair of backpressure regulators driven by a
PID control system. As the state of the CO2 exiting
the vessel changed, the density and the viscosity of
the CO2 passing through the backpressure regulator
valve changed. This made it very difficult to identify
PID coefficient values that would provide a
sufficiently stable control tune that would work over
the course of a single experiment. To alleviate this
problem on our final apparatus iteration we placed
the back pressure regulator downstream of a heat
exchangers in order to keep the CO2 passing through
the regulator valve at a more constant temperature.
RESULTS
Single-Phase CO2
The temperature data from twenty-two
thermocouples from a representative single phase
CO2 experimental run is shown in Figure 5. The
thermocouples are numbered primarily in order
increasing radii and secondarily by increasing
elevation in the vessel. Thermocouple one for
example is located on the central axis at the bottom
of the vessel, while thermocouple number twenty-two
is located near the vessel wall at the top of the vessel.
It can be seen from the plot that there is an initial
temperature gradient present in the saturated medium
with a lower temperature at the base. This gradient is
most likely due to gravity and thermal convection.
The temperature front can be seen in the plot as it
passes axially through the sample past the
measurement locations. After the initial sharp
temperature drop, the temperatures gradually
approach equilibrium, and a radial temperature
gradient then develops, indicated by the grouped
lines spreading out. The exterior thermocouple
locations trend towards a higher temperature than
those that are more central (solid lines).
The large spikes can be seen in temperature data for
TC1 at the vessel inlet are due to short interruptions
in flow as the pumps stop during the pump switch
over event. During the brief time when flow was
stopped, heat from the steel end cap increased the
temperature of the cold fluid that was located within
the injection passage
The interplay between convective and conductive
transport can be seen in the shape of the temperature
vs. time curves. A purely convective process would
feature sharp thermal fronts and a near-vertical slope
at the time when the cold fluid slug reached the
thermocouple. A purely conductive process would
generate a gentler slope with smooth transitions. The
experimental run shown in Figure 5 was at a
relatively high flow rate with a calculated bulk Peclet
number of around 1500. The steep temperature front
corresponds to convectively dominated behavior.
Figure 6 shows temperature data from a lower flow
rate experimental run that was less convectively
dominated (only temperature data from the central
axis of the vessel are shown for clarity). The
calculated bulk Peclet number of this experimental
run was approximately 550, or one third that of the
run in Figure 5.
Water & CO2 Compared
In order to compare the performance of CO2 heat
extraction with that of water, we performed a single
phase water run at the same temperature and pressure
as a previous CO2 run (Figure 7). Both experiments
were run with a volumetric fluid injection rate of 150
ml/min, an initial core temperature of 75C, and a
backpressure of approximately 1450 psi. The CO2
and water experiments had approximately the same
pressure differential across the sample for both water
and CO2 (2 bar), and a similar mass flow rate (2.51
g/s for water, and a range of 2.3 to 2.04 g/s for CO2)
which allowed a more straightforward comparison of
the performance of the two fluids. The injected water
and CO2 increased in temperature after the first two
pump volumes due to the fact that we were recycling
the injection fluid. The CO2 plot shows a much
steeper temperature front indicating a more advection
dominated flow than the water experiment.
0 500 1000 1500 200010
20
30
40
50
60
70
80
90
100
Temperature History, 100 (ml/min) Flow Rate, 100C Vessel Temperature, 138 bar Back Pressure
Time (s)
Tem
pe
ratu
re (
C)
TC1
TC4
TC8
TC12
TC13
TC17
TC21
Figure 6 - Temperature vs time from a
experimental run with a CO2 flow rate of
100ml/min, and a bulk Peclet number of
556.
0 100 200 300 400 500 600 700 800 9000
10
20
30
40
50
60
Temperature History, 200 (ml/min) Flow Rate, 60C Vessel Temperature, 83 bar Back Pressure
Time (s)
Tem
pe
ratu
re (
C)
TC1TC2TC3TC4TC5
TC6TC7TC8TC9TC10
TC11TC12TC13TC14TC15
TC16TC17TC18TC19TC20
TC21TC22
Figure 5 - Temperature vs. time data from twenty-two thermocouples from a representative single phase CO2
experimental run. Plot line colors indicate the thermocouple elevation and correspond to the colored
markers shown in the thermocouple diagram (Figure 2)
In order to compare the heat extraction of the two
fluids, we used the experimental data to calculate the
heat extraction rates of the two fluids. Figure 8
shows the heat extraction rate of water (blue line) and
the heat extraction rate of CO2 (red line). Initially the
heat extraction rates of the two fluids are similar,
with a higher rate for water at the start of the
experiment and CO2 overtaking it in as the
experiment progresses. The performance of CO2
stays somewhat stable despite the fact that the mass
Figure 7 - Comparison of temperature history data of water and CO2 as working fluids. Top plot shows a water
run and the bottom plot is a CO2 run. (Thermocouple labeling is not consistent with other figures in
this paper)
0 200 400 600 800 1000 12000
50
100
150
200
250
300Heat Extraction Rate
Time (s)
He
at E
xtr
actio
n R
ate
(W
atts)
H2O
CO2
Figure 8 - Comparison of the heat extraction rate of water and CO2 under similar experimental conditions.
flow rate is decreasing during the experiment due to
the increase in injection temperature. The water heat
extraction rate decreases after the first two pump
volumes due to the increased injection temperature.
Since both the water and CO2 were operated at the
similar pressure and volumetric flow rate, it can be
assumed that the work performed by the pumps in
both experiments was comparable. A complete
thermodynamic characterization of the runs would
require measurements of the fluid accumulation
inside the vessel which was not recorded for these
experiments.
Two Phase Experiments
We performed an experimental run in which cold
CO2 was injected into a heated water-saturated
sample, analogous to the initial development phase of
a CO2 geothermal reservoir. Figure 9 shows the
temperature history.
The temperature drop at the bottom of the vessel,
where the cold CO2 was injected, was very
pronounced, but temperature trends at the other
thermocouple locations were very smooth and
gradual. The temperature drop at the top of the
vessel was not significantly greater than the
temperature drop that would be expected due to
passive cooling to the lab atmosphere. The
temperature spikes seen at the injection location (e.g.
at 1200 seconds) were due to a temporary cessation
of injection flow during the pump switchovers. The
overall temperature change shown in this experiment
is not as drastic as the single phase experiments.
The presence of viscous fingering was indicated by
multiple pieces of experimental evidence. After the
CO2 was vented to the atmosphere, it was found that
approximately 74 percent of the pore volume still
contained the original water. More than one pore
volume of CO2 was injected into the vessel, and if no
fingering occurred the CO2 would act as a solid slug
displacing all of the water originally present in the
vessel. Further evidence of fingering is indicated by
the CO2 breakthrough occurring at approximately 500
seconds after injection began. Breakthrough
occurred after approximately 251 mL of CO2 was
injected into the vessel, corresponding to
approximately 20 percent of the sample pore space.
The volume of CO2 injected at breakthrough, and the
remaining water in the pore space indicate that very
little water was lost after the initial water was
displaced during the CO2 flow path formation.
Compared to our previous results with an initially
CO2 saturated sample, our new results show much
less change in temperature in the vessel during the
course of the experiment. Besides the effects of
0 500 1000 1500 2000 2500 3000 3500 400020
25
30
35
40
45
50
Time (s)
Tem
pe
ratu
re (
C)
TC1TC2TC3TC4TC5
TC6TC7TC8TC9TC10
TC11TC12TC13TC14TC15
TC16TC17TC18TC19TC20
TC21TC22
Figure 9 - Temperature history of cold CO2 injection into a heated water-saturated sample, 45C initial
temperature, 25 ml/min flowrate, 17 MPa (2500 psi) outlet pressure.
viscous fingering, this smaller temperature drop is
most likely the result of the higher heat capacity of
water compared to CO2, the lower flow rate of the
injected CO2, and the higher temperature of the
injected CO2 when compared to our earlier single
phase experiments. At the initial conditions, the
water-saturated sample contained 234 kJ of heat
energy in the pore space compared to 123 kJ for CO2
under the same conditions. The thermal energy of
the water in the vessel was so great and the cooling
capacity of our injected fluid was so low, that we
only saw evidence of the cooling effects of the CO2
flow at the injection end of the sample.
DISCUSSION
The design of our experiment was primarily intended
to produce a data set that could then be used to
validate numerical modeling of the use of CO2 as the
working fluid in an EGS reservoir, and was not
intended to be directly applicable to full-scale EGS
systems. Specifically, our porous medium sample
was not designed to replicate the characteristics of
the flow paths that would be expected in a field-scale
geothermal system. Despite this, our results can be
used to gain insights into the behavior of CO2 as an
EGS working fluid.
Our single-phase CO2 experiments have shown that
the heat extraction behavior of CO2 is sensitive to the
initial and operating conditions; injection rate,
operating pressure, and initial reservoir temperature
(Magliocco 2011). The Peclet number and the shape
of the temperature histories vary greatly depending
on the initial and operating conditions.
We have also shown experimentally that CO2 and
water have comparable behavior under one particular
set of initial and operating conditions. This result
could be viewed as discouraging in the context of
CO2 based EGS, but was in line with previous
modeling results (Preuss 2007). The greater benefits
of CO2 over water are expected to occur in the 5-spot
well geometry that is dominated by radial flow. In
our apparatus, the flow is predominately linear along
the length of the vessel with small portions of radial
flow near the inlet and outlet boundaries. In addition,
the thermal energy stored in the large end caps used
in our vessel may retard the development of cold
zones near the fluid injection point.
Experimental results for the more realistic conditions
of CO2 injection into a water-saturated sample show
convincing evidence of preferential flow of CO2 into
the saturated sample. Much of the sample pore space
appears to have been bypassed by the flowing CO2,
as indicated by the temperature data, the early CO2
breakthrough, and the remaining water in the sample
at the conclusion of the experiment. The effect of
CO2 bypassing large regions of the sample (or
reservoir) will reduce the effective heat transfer to the
working fluid, thus this is an important consideration
warranting further investigation. In future
experiments, a much clearer picture of the system
could be realized by analyzing the produced fluid
over the course of the experiment to determine the
ratio of CO2 to water.
CONCLUSION
Using our laboratory apparatus, we have now
successfully explored three key aspects of using CO2
as an alternative to water in geothermal systems.
During the initial injection of CO2 into a previously
water-saturated sample, we found that preferential
flow patterns strongly affect the heat production of
the system. Our single-phase experiments have
shown that heat production and the temperature
profile of the system is strongly affected by the
injection rate, back pressure, and initial system
temperature. We have also demonstrated that under a
particular set of operation conditions that water and
CO2 have comparable heat extraction rates. We are
currently working on expanding the temperature and
pressure range of our experiments, in order to create
a well curated body of data that can be used to
validate numerical modeling software.
REFERENCES
Brown D. . (2000), “A Hot Dry Roc Geothermal
Energy Concept Utilizing Supercritical CO2
Instead of ater.” In Proceedings of the Twenty-
Fifth Workshop on Geothermal Reservoir
Engineering, Stanford Univ, 233-238
King, S., F. Beck, and U. Lüttge. "On the mystery of
the golden angle in phyllotaxis." Plant, Cell &
Environment 27.6 (2004): 685-695.
Liao S.M. and Zhao T.S. (2002), “An experimental
investigation of convection heat transfer to
supercritical carbon dioxide in miniature tubes”,
Int. J. Heat Mass Transfer, 45, 5025–5034
Magliocco, M., Kneafsey, T., Pruess, K., Glaser S.,
(2011) Laboratory experimental study of heat
extraction from porous media by means of CO2
Proceedings of the Thirty-sixth Workshop on
Geothermal Reservoir Engineering, Stanford
University, Stanford, CA, January 31–February 2
(2011) SGP-TR-191
Magliocco, M., Kneafsey, T., Preuss, K., Glaser, S.
(2012), “Laboratory and Numerical Studies of
Heat Extraction from Hot Porous Media by
Means of Supercritical CO2”, Proceedings of
TOUGH Symposium 2012, Lawrence Berkeley
National Laboratory, Berkeley, California,
September 17-19, 2012.
Pruess, K. (2006) Enhanced Geothermal Systems
(EGS) Using CO2 as Working Fluid – A Novel
Approach for Generating Renewable Energy
with Simultaneous Sequestration of Carbon,
Geothermics, Vol. 35, No. 4, pp. 351–367
Pruess, K. (2007), “Enhanced Geothermal Systems
(EGS) Comparing Water With CO2 as Heat
Transmission Fluids.” In Proceedings, New
Zealand Geothermal Workshop 2007 Auckland,
New Zealand, November 19-21
Pruess, K. (2007), “Enhanced Geothermal Systems
(EGS) Comparing Water With CO2 as Heat
Transmission Fluids.” In Proceedings, New
Zealand Geothermal Workshop 2007 Auckland,
New Zealand, November 19-21
Pruess, K. and Spycher, N. (2010) Enhanced
Geothermal Systems (EGS) with CO2 as Heat
Transmission Fluid - A Scheme for Combining
Recovery of Renewable Energy with Geologic
Storage of CO2, Proceedings, World Geothermal
Congress 2010, Bali/Indonesia, 25-29 April
2010.