American Oil and Gas Reporter - October 2019 - 51

cm2/s. The low incremental recovery factor
(the 0 CO2 diffusion coefficient) corresponds to a 12% increase over the primary
production recovery factor, while the
highest incremental recovery factor (0.001
cm2/s diffusion coefficient) corresponds
to a 31% increase based on primary production recovery factor. Much more CO2
is dissolved for the high diffusion case.
These same CO2 diffusion coefficients
also were simulated using the dual-permeability model. In this case, total oil
recovery was improved very little as the
diffusion coefficient again was incrementally increased from 0 to 0.001 cm2/s.
The oil recovery from the highest gas
diffusion (0.001 cm2/s) case (42.44%) to
no gas diffusion case (40.05%) is only
2.4%. The gas diffusion effect is much
less significant with the dual-permeability
model. Much more CO2 is dissolved in
the reservoir for the high-diffusion case
than the no-diffusion case, but much less
than the single-porosity model cases.
Similar simulations using the dualpermeability model clearly indicated that
gas diffusion helps improve recovery
quality, but not as much as the one in the
single-porosity mode. Because the single-porosity model does not include
natural fractures, the gas diffusion effect
is very pronounced because diffusion is
the only method for CO2 to reach out
into more oil in the matrix and recover it.
In the dual-permeability model, natural
fractures (conduits between matrix and
hydraulic fractures) are more critical, especially for the CO2 HNP because CO2
could quickly transport in the larger
contact area of natural fracture to diffuse
and mix with heavy oil components in
fracture and matrix and recover them.

Neglecting the existence of the natural
fracture network in the single-porosity
model underestimates the primary production performance of the tight oil fractured reservoir and tends to overestimate
the role of gas diffusion for the HNP
process. Rather than the gas diffusion
effect dominating the oil recovery mechanism, transport in natural fracture is the
main oil recovery mechanism for HNP
process using the dual-permeability model.
Consequently, the dual-permeability model
is more accurate for simulating gas HNP
for tight oil fracture reservoirs.
Injection Gas Alternatives
Because of CO2 availability restrictions
and facility limitations, it is realistic to consider alternative gasses for HNP processes,
especially for wells in basins such as the
Eagle Ford, which produce large amounts
of both oil and hydrocarbon gasses. The
research examined three alternative injection
gasses: 100% nitrogen, 93% methane (lean
gas) and 65% methane (rich gas).
At a relatively low injection rate of
600 Mcf/d, all three alternative gases
would recover more oil than CO2 at Eagle
Ford reservoir conditions. Oil recovery
quality of CO2 and rich gas are much
better than nitrogen and lean gas. Rich
gas is the best performer and nitrogen is
the poorest alternative gas. At pressures
and temperatures typical of the Eagle
Ford, CO2 should be better dissolved in
oil phase than any of the alternative gas
components and its volume under this
reservoir condition would be much smaller,
which is why the total CO2 injected
volume (0.084 PV) is much lower than
the alternative gases (0.11-0.12 PV).
However, in terms of cumulative oil

FIGURE 4
Cumulative Oil Production for HNP with Different Injection Gases

Cumulative Oil (bbl)

3.00e+5

2.00e+5

1.00e+5

0.00e+0

Primary Production
N2 HNP
CO2 HNP
C1 65% C2-C5 35% HNP
C1 93% C2-C5 7% HNP

2,000

4,000
Time (days)

6,000

8,000

production at a 600 Mcf/d injection rate
(Figure 4), the ranking of the four gases
would be:
· Rich gas;
· CO2;
· Lean gas; and
· Nitrogen.
From an economic point of view, the
rough economical evaluation shows that
using the rich field gas would be the most
profitable strategy for Eagle Ford operators.
If all the rich gas injected in the HNP
process could come from local Eagle Ford
field production, it would eliminate cost to
acquire a gas supply and significantly
reduce related facility costs. If this fieldproduced gas could be recycled and injected
back into the HNP process, the cost of
HNP projects would be reduced while simultaneously cutting emissions from flaring.
Therefore, using field gas appears to be an
economic and realistic solution for the
HNP process in the Eagle Ford play.
This study only considered a single
well. For applications in multiple horizontal wells on a pad level, interwell interference and complex natural fracture
structures are major issues that need to
be considered. Gas adsorption and trapping
(hysteresis) during HNP operations are
two other very important influence factors
in tight oil reservoirs. Future University
of Texas research projects plan to investigate these and other issues.
❒
Editor's Note: The authors acknowledge Computer Modeling Group Ltd. for
providing the software to simulate the
Eagle Ford HNP process.

LEIZHENG WANG is a research fellow at the University of Texas at Austin.
He has conducted primary production
and CO2 huff-n-puff EOR recovery
process simulations for Bakken and Eagle Ford tight oil reservoirs, and studied
the influencing factors such as reservoir
heterogeneity, permeability correlation
length, injection rate, time, soaking time
and production time. Wang previously
worked in reservoir engineering roles
at ConocoPhillips and Halliburton. He
holds a Ph.D. chemical engineering
from the University of Houston.
WEI YU is research fellow at both
the University of Texas at Austin and
Texas A&M University. His work includes simulation of complex natural
and hydraulic fracture geometries,
and enhanced recovery of shale gas
and tight oil reservoirs. He holds a
Ph.D. in petroleum engineering from
the University of Texas at Austin.
OCTOBER 2019 51

American Oil and Gas Reporter - October 2019

Table of Contents for the Digital Edition of American Oil and Gas Reporter - October 2019