Daniel Purvis
STEP Energy Services
Houston
LPG flooding traditionally has been used for tertiary recovery to mobilize stranded oil after waterflooding. In 1996, Chevron Corp. and Halliburton Co. began studying LPG as a stimulation fluid. The first LPG fracture treatment was performed in Canada in 2008, with the first US treatment performed in 2009. Since then, LPG applications have expanded to include secondary recovery, removing water blockages, and dissolving paraffins.
LPG (primarily propane, butane) is miscible at first contact with oil and soluble with natural gas at virtually all reservoir conditions and pressures. LPG has lower surface tension and lower viscosity than water (see accompanying table). Miscibility is achieved at lower pressures (~400 psi) than typical miscible floods, requiring only low-cost, low-rate frac pumps to inject LPG into the formation below fracture pressure.
For miscible huff-and-puff treatments, LPG improves flowback as opposed to water as the flooding fluid because LPG causes the oil to swell, reducing its viscosity and increasing its mobility. LPG is nearly half the density of water and has a gas-to-liquid expansion ratio of 270:1. If in solution under flowing conditions, LPG lightens the fluid column, improving uplift. If LPG liberates from oil in the tubing, it can provide in situ gas lift. LPG’s hydrostatic gradient is 0.234 psi/ft, which is naturally underbalanced and assists in post-treatment cleanup.
LPG is compatible with water-sensitive formations that have clay swelling issues, sub-irreducible saturation, and imbibition issues. Other applications for LPG are in depleted or low-pressure, low-producing oil wells, condensate wells that have condensate banking-liquid loading issues; and wells with poor frac-load recovery.
Surfactants increase crude recovery by changing wettability of the reservoir and reducing interfacial surface tension (IFT) between crude and water. These surfactants are typically delivered to the reservoir in fracture water or brine (OGJ, July 4, 2022, p. 48).
For LPG injection, STEP Energy Services Co. (STEP) acquired a proprietary surfactant from Universal Chemical Solutions Inc. (UCS) that dissolves in LPG for delivery to the reservoir and subsequent partitioning between LPG and water (connate water, residual fracture water). To prevent emulsification downhole (which could lead to pore blocking) and foaming at surface, demulsifiers and defoamers are added to the surfactant package.
Field application
Located near Eagle Pass in Maverick County, Tex., Sacatosa field, Chittim Ranch was discovered in 1956 and primarily targeted the San Miguel formation, which is a low-permeability (3 md average) sandstone formation with 20% average porosity and shale layers running through the formation. In certain areas, high clay content leads to incompatibility with water. In 1966, a waterflood program made economic recovery of hydrocarbons possible. Since then, over 1,600 wells have been drilled in the field, and there are currently about 289 active producing wells and 322 water-injection wells. Well spacing in parts of the field is as close as 380-400 ft. Wells have relatively shallow (1,200-1,700 ft) total TVD and trend from northwest to southeast. The field produces light oil (38° API gravity) with paraffinic properties. The targeted zone has a 14-ft thickness, 0.5-3 md permeability, and 13-25% porosity. Waterflooding was performed using line drive and five-spot patterns. Line drive appears to be most effective.
Pilot treatment program
In March 2022, Windy Cove Energy II (Windy Cove)—which has an EOR background in the Permian basin, particularly CO2 floods in conventional formations—purchased the Chittim Ranch lease from ConocoPhillips Co., with acquisition of Sacatosa field to be the company’s entrance into unconventional plays and an opportunity to perform miscible LPG-surfactant injection as an alternative to a capital-intensive gas huff-n-puff or CO2 flood.
Goals of the inaugural miscible injection program included:
- Proving low-pressure miscibility.
- Validating surfactant technology.
- Increasing production from a test well, offset wells, and field.
Candidates were selected based on geological and petrophysical characteristics, historical waterflood operations, wellbore integrity, and historical production. Most wells in the field averaged 1-2 bo/d and 10-15 bw/d.
The well selected for the LPP-surfactant process flowed at 1.5 bo/d and 2 bw/d. The well was on rod lift with 20 psi bottom hole pressure (BHP). The main reason for choosing the well was because it had previously been hydraulically fractured in three different zones with a total of 55,000 lb of proppant. This allowed the treatment to be pumped at targeted rates but below fracture initiation pressure.
The final treatment injected 353 bbl total LPG into the well with a predetermined volume of surfactant injected throughout the treatment at an injection rate of 2 bbl/min. The treatment was injected into three separate zones using diverter balls to isolate each zone. At the designed pump rates, expected injection pressures were in the 1,200-1,400-psi range. A pre-job injection test with water showed injection pressures of 1,300 psi at 2 bpm and 1,500 psi at 3 bpm. Based on these data, injection was expected below fracture initiation pressures. During the treatment, offset wells—including both injectors and producers—were monitored for pressure changes due to pressure communication. It was determined pre-job that if pressure communication was observed in offset wells during the treatment, diversion balls would be dropped to cut off pressure communication from the offset well. If pressure communication did not stop, the treatment would end.
Treatment
LPG (104 bbl) was pumped in the first stage of treatment at 2 bpm with an observed injection pressure of 1,177 psi (Fig. 1). This pressure was maintained throughout the entire stage with no observed issues and lower than expected injection pressure. Diverter balls were dropped at the end of the stage, and a pressure increase of 16 psi was observed when the diverter balls landed at the intended perforations, indicating successful diversion.
An additional 52 bbl of LPG was pumped in the second stage of the treatment at 2 bpm with an observed injection pressure of 1,180 psi. At the midpoint of the second stage, a decision was made to increase the injection rate to 3 bpm. At 3 bpm, injection pressures were 1,200 psi, again lower than expected based on pre-job injection tests. Another 52 bbl of LPG were injected at 3 bpm to conclude the second stage, near the end of which pressure communication was observed in an offset injector well. Diverter balls had already been pumped in anticipation of this communication, and upon landing, injection pressure rose 14 psi to 1,214 psi. The pressure in the offset well decreased upon landing of the balls, indicating successful diversion. A total of 104 bbl were injected during second-stage treatment.
A total of 104 bbl of LPG were also scheduled to be injected in the third and final stage of treatment, but due to successful treatment execution up to that point, the decision was made to inject all 145 bbl of LPG remaining in on-site storage, exceeding the originally planned injection volume by 45 bbl. The remaining LPG was injected at 3 bpm at 1,214 psi. Upon completing the treatment, a nitrogen blanket was placed in the wellbore, the well was shut in, and equipment was disconnected from the wellhead. Wellhead shut-in pressure post-job was 1,000 psi. The well, which was shut-in overnight, flowed back the following day.
Flowback, testing
Wellhead pressure maintained at 440 psi the day after treatment. During flowback, oil and water samples were analyzed for miscibility, viscosity, and surface tension changes. After 20 min of initial nitrogen flowback, hydrocarbon gases started to show in the flow stream. After 90 min, hydrocarbon liquids appeared in the flow stream. The well flowed on its own for 5 hr, making 22 bbl total liquids (8 bbl oil, 14 bbl water) compared with the pre-job rate of 3.5 b/d total liquids (1.5 bo/d, 2 bw/d). After 5 hr, wellhead pressure was 30 psi. The well was shut in to re-run rods and pump.
Oil and water samples were obtained pre-job for baseline measurements as well as upon initial flow after treatment, 3 hr after initial flow, 9 days after treatment, and 10 days after treatment.
Fig. 2 shows the C11-C35 oil composition after flowback. Butane (C4) content in the oil drives a dramatic reduction in viscosity and gravity. The original oil gravity was 38-39° API while post-treatment oil was 45-47° API. Post-treatment C5+ component mix indicated increasing weight percent of C12-C35 components and decreasing weight percent of C36+ (data not shown). Mobilization across the C12-C35 spectrum indicates true miscibility as opposed to vaporization of only lighter ends. Miscibility was achieved with butane at low reservoir pressures (about 500 psi).
Fig. 3 shows that surfactant added to LPG successfully partitioned to the formation water, lowering the surface tension of produced water. Post-job samples from Apr. 14 showed 71% of critical micelle concentration (CMC), and a sample from Apr. 16 showed 74% of CMC. CMC is the concentration of surfactant where additional surfactant does not further lower surface tension, and the results indicated that the surfactant effectively partitioned between the oil and water to an extent that yielded close to the minimum water surface tension. The water samples showed no signs of emulsion or foam created by the treatment.
Production analysis
The well produced 22 bbl of total fluids the day following treatment over a 5-hr period, with 8 bbl oil and 14 bbl water. Before treatment, the well produced 1.5 bo/d and 2 bw/d. In the first 5 hour of flowback after treatment, the well produced 9 bbl oil (about 43 bo/d) without water from natural flow without artificial lift assistance. After initial flowback, the well was shut-in for what was originally planned to be 1-2 days to run rods and pump back in the well. Due to unforeseen problems, the well remained shut-in for 8 days before resuming production.
On Day 5 of well shut-in, field stock-tank oil gravity increased to 47° API from 38° API, indicating that injected butane migrated across the field and produced out of a different well or wells. After opening the well 8 days later, total water production had increased 510 bw/d (an increase of about 13%) and total oil production increased 55 bo/d (an increase of about 10%).
Oil production increased 55 bo/d above pre-treatment, to 610 bo/d from 555 b/d. Over 60 days post-treatment, total oil production in the field has increased about 60 bo/d. Twenty-one net wells were added to production at an average of 2 bo/d for an estimated increase of 42 bo/d, leaving potentially 18 bo/d that can be attributed to the LPG injection treatment. The treated well is producing 1.5-2.5 bo/d, with spikes up to 6 bo/d. A secondary benefit of this treatment was dissolving paraffin, as observed post-treatment in collected oil samples.
The pilot program met project goals, as the surfactant effectively partitioned between the LPG-water phases, treated beyond the injected well, and produced long-term EOR.
The Author
Daniel Purvis ([email protected]) is technical sales - LPG stimulation/EOR manager at STEP Energy Services. He holds a BBA from Sam Houston State University (1999) and is a member of the Society of Petroleum Engineers (SPE).