Evaluation of Low Salinity Injection for the Minnelusa Formation, Powder River Basin, Wyoming

Evaluation of Low Salinity

Injection for the Minnelusa

Formation, Powder River Basin,

Wyoming.

Prepared for Enhanced Oil Recovery Institute

Authors:

Geoffrey Thyne

Pubudu Gamage

June 30, 2011

2

DISCLAIMER

THIS MATERIAL WAS PREPARED AS AN ACCOUNT OF WORK PRODUCED BY AN

AGENCY OF THE STATE OF WYOMING. NEITHER THE STATE OF WYOMING, NOR THE

UNIVERSITY OF WYOMING, NOR THE ENHANCED OIL RECOVERY INSTITUTE (EORI),

NOR ANY OF THEIR EMPLOYEES, MAKES ANY WARRANTY, EXPRESS OR IMPLIED, OR

ASSUMES ANY LEGAL LIABILITY OR RESPONSIBILITY FOR THE ACCURACY,

COMPLETENESS, OR USEFULNESS OF ANY INFORMATION, APPARATUS, PRODUCT, OR

PROCESS DISCLOSED, OR REPRESENTS THAT ITS USE WOULD NOT INFRINGE

PRIVATELY OWNED RIGHTS. THE EORI RESERVES THE RIGHT TO CHANGE, AMEND,

ALTER, ADDEND OR REMOVE THE CONTENTS OF THIS REPORT IN FUTURE EDITIONS

BOTH IN PART OR IN ITS ENTIRETY.

THIS DISCLAIMER IS MADE WITH RESPECT TO THE DOCUMENT AS WELL AS THE

ASSOCIATED DATA.

CONFIDENTIALITY NOTICE:

THIS WORK HAS BEEN PREPARED FOR THE SOLE USE OF ENHANCED OIL RECOVERY

INSTITUTE.

THIS REPORT CONTAINS INFORMATION BELONGING TO THE AUTHORS AND THE

ENHANCED OIL RECOVERY INSTITUTE WHICH MAY BE PRIVILEGED, CONFIDENTIAL

AND/OR PROTECTED FROM DISCLOSURE. THIS REPORT IS INTENDED ONLY FOR THE

USE OF THE INDIVIDUALS OR ENTITIES NAMED ABOVE, WITH WHOM A

CONFIDENTIALITY AGREEMENT OR NON-DISCLOSURE AGREEMENT HAS BEEN

FINALIZED. IF YOU ARE NOT THE INTENDED RECIPIENT, YOU ARE HEREBY NOTIFIED

THAT ANY USE, DISSEMINATION, DISTRIBUTION OR COPYING OF THIS REPORT IS

STRICTLY PROHIBITED.

Technical Report: Evaluation of Low Salinity Injection for the Minnelusa

Formation, Powder River Basin, Wyoming.

i

Table of Contents

Executive Summary i

Introduction 1

Low Salinity Background 2

Minnelusa Geology and Production 2

Field Data 5

Water Chemistry 5

Production Data 6

Preliminary Evaluation 9

Controls on Recovery 12

Further Evaluation of Field Data 13

Recovery Factor-Independent Analysis 16

Laboratory Studies 18

Experimental Procedure 18

Brine, Oil and Rock Properties 20

Brines 20

Cores 20

Crude Oils 22

Single Phase Experiment 22

Two-Phase Experiments – Secondary Mode 23

Two-Phase Experiments – Tertiary Mode 25

Summary of Experiments 26

Geochemical Modeling 27

Procedures 27

Model Calibration 27

History Matching the model 29

Field Scale Modeling 30

Discussion 31

References 34

Tables

Table 1. List of fields used in this study with accompanying data. 10

Table 2. Average value for recovery factor for Minnelusa Fields. 12

Table 3. Average value for recovery factor for Minnelusa Fields. 18

Table 4. Brine composition. 20

Table 5. Core properties. 20

Table 6. Crude oil properties. 22

Table 7. Summary of experimental results. 26

Table 8. Parameters used for calibrated model from experiment #25. 27



ii

Figures

Figure 1: Left figure shows the stratigraphic relationships of Pennsylvannian-

Permian reservoir rocks in Wyoming, from Markert and Al-Shaieb (1984). Right

figure shows a typical Minnelusa well log with mineralogy from James (1989). 4

Figure 2: Typical production curve for a Minnelusa field showing recovery factor

and oil cut versus pore volumes (of produced fluid). Most oil production takes

place over 0.4 pore volumes. 4

Figure 3: Map of locations of selected Minnelusa oil fields with salinity contours

in color. Map was used to estimate salinity for those fields lacking water

chemistry analyses (See Table 1). Cartography by B. Reyes. 6

Figure 4: Upper histogram shows the bimodal distribution of recovery factor for

64 Minnelusa fields, while lower histogram shows the distribution for recovery

factor for fields with values greater than 30%. 7

Figure 5: Histogram of the recovery factors for fields with polymer injection and

fields with waterflooding only for fields with greater than 30% recovery. 8

Figure 6: Cumulative oil production curves for 37 Minnelusa Fields versus pore

volumes of produced fluids (oil+water). Pore volumes calculated using oil and

water production. Cool colors are waterfloods with polymer application and

warm colors are waterfloods without polymer. 9

Figure 7: Plots of API gravity, initial oil saturation, OOIP and net pay for

Minnelusa fields with different symbols for fields with polymer injection and

fields with only normal waterflooding.. 13

Figure 8: Salinity ratio versus recovery factor for 52 Minnelusa fields. Dilution

factor of 1 is for fields with no dilution (injection of formation water), shaded area

highlights fields with dilution factors equal or less than 0.1 (10-fold dilution). 14

Figure 9: Oil production versus pore volume for Minnelusa Fields. Cool colors

are saline or mixed waterfloods and warm colors are low-salinity waterfloods for

only cases with 10-fold dilution factors. Shaded area encompasses range of

waterflood recovery factors. 15

Figure 10: Typical curves of normalized cum. oil production versus oil cut (%). 16

Figure 11: XRD spectra of whole rock Minnelusa sample from core material

showing the presence of quartz, anhydrite and dolomite. 21

Figure 12: XRD spectra of clay separate fraction from the Minnelusa core sample

showing the clay is composed of illite. 21

Figure 13: Pressure drop, conductivity and the pH variation during a single phase

experiment, secondary mode experiment (core directly flooded with the low-

salinity brine) with core A. 23

Figure 14: Oil recovery, pressure drop, conductivity and the pH variation during

two phase secondary mode experiment (core directly flooded with the low-salinity

brine) with core plug E. 24


two phase secondary mode experiment (core directly flooded with the low-salinity

brine) with core plug C and Raven Creek oil. 24


two phase, tertiary mode experiment with core plug D and Gibbs oil. 25


two phase tertiary mode experiment with core plug B and Raven Creek oil. 25



iii

Figure 18: Effluent concentrations from experiment #25, single phase, Minnelusa

brine and core plug A. 28

Figure 19: Comparison of predicted (solid line) and measured values for Na, Cl,

Ca and SO4 from the single-phase experimen using Minnelusa brine and rock.. 28

Figure 20: Comparison of predicted (solid line) and measured values for TDS

from the single-phase experimen using Minnelusa brine and rock.. 28

Figure 21: Effluent concentrations during low-salinity brine flood of core plug D

with Gibbs oil. 29

Figure 22: Predicted versus actual TDS for two-phase experiments using

parameters from the calibrated model. Upper line is from experiment tertiary

mode experiment and lower line is for secondary mode experiment. 29

Figure 23: Predicted TDS versus time for low-salinity brine injection over two

years into the North Timber Creek field. Assumptions described in text. 30

Figure 24: Map view of distribution of predicted TDS from low-salinity brine

injection for two years into the North Timber Creek field. Assumptions described

in text. 31



i

EXECUTIVE SUMMARY

Current oil production in Wyoming is primarily from fields under waterflood. Most

major companies are pursuing options that will change the chemistry of injected water to

improve recovery. These efforts involve a range of modifications that include lowering

overall salinity and modifying inorganic solutes. Improving knowledge of these efforts

and how they can be applied to Wyoming producers adheres to the portion of our

mandate on technology transfer.

This report summarizes the history and evaluates the effectiveness of low-salinity

flooding in the Minnelusa Formation in the Powder River Basin of Wyoming. The

Minnelusa sandstone play constitutes a resource of several hundred fields with

cumulative production of more than 600,000,000 barrels of oil. Currently there are 130

Minnelusa fields that are in active waterflood.

Fifty-five are flooded with low-salinity water, 52 with mixed salinity water and 23 with

formation brine. Recovery factors for 51 fields were compiled and used as the primary

metric to evaluate the effectiveness of low-salinity waterflooding. A second metric used

was the comparison of production curves expressed as field pore volume produced versus

recovery factor. The final metric used was normalized production versus % oil cut,

which is independent of recovery factor. None of the metrics showed any increased

production for the fields that used low-salinity injection compared to fields flooded with

reservoir brine or mixed water.

In the field cases studied, the low-salinity injection water was derived from wells in the

shallow Lance and Fox Hills Formations, which have an average salinity of 2100 ppm,

while the Minnelusa fields had initial formation water salinity ranging from 1134 to

261,000 ppm. Some fields studied did not have sufficient salinity reduction because

injected salinity was very similar to formation water salinity. In other cases where there

was sufficient dilution there was still no increase in oil production. Dilution factor was

expressed as salinity ratio to allow direct comparison of all fields. Regardless of dilution,

there was no apparent trend of increasing recovery with lower salinity injection.

To investigate the lack of increased production expected from low-salinity injection,

laboratory studies using Minnelusa oil, synthetic brine and rock were performed. The

studies showed little or no incremental oil production during low-salinity injection.

Geochemical models of water-rock interaction were constructed to help better understand

the low-salinity process in the Minnelusa Formation. These models were calibrated to

the laboratory results and then used to simulate the field scale. The modeling showed the

mineralogy of the formation will increase the salinity of the injection water by mineral

dissolution, however this increase is not significant enough to hinder dilution. It does not

appear that low-salinity injection will increase the effectiveness of waterflooding for the

Minnelusa formation.



1

INTRODUCTION

Low Salinity Background

Low-salinity waterflooding has been widely studied during the last decade by various

research groups as one of the most inexpensive methods of enhanced oil recovery (EOR).

The level of investigation into low-salinity waterflooding has sharply increased in the

past three years as more research groups have become involved (Webb et al., 2008,

Alotaibi and Nasr_el_Din, 2009, Austad et al. 2010, Boussour et al. 2009, Cissokho et al.

2009, Kumar et al. 2010, Lager et al. 2008, Patil et al., 2008, Seccombe et al. 2008, Pu et

al. 2010, Rivet et al. 2010, RezaeiDoust et al. 2010). Laboratory studies with synthetic

formation water, reservoir and outcrop rocks and reservoir oil are injected with water

diluted by a factor ranging from 10 to 100-fold compared to formation waters. Many

studies have confirmed that this method can increase the recovery the lab scale by 2-30%

OOIP depending on the brine composition, crude oil composition and rock type used.

However, while both laboratory and field studies have had successful results, there are

examples in which low-salinity flooding does not create additional production (Sharma

and Filoco, 2000, Rivet et al 2010, Skrettingland et al. 2010).

The fundamental principle of low-salinity flooding is based on observations of laboratory

corefloods by Martin (1959) and Bernard (1967). This work was extended and brought

to wider attention by various workers over the last 15 years (Jadhunanadan and Morrow,

1995, Zhou et al., 1995a, Zhou et al., 1995b, Zhou et al., 1995, Tang and Morrow, 1997,

Yildiz et al., 1999, Morrow et al., 1998, Tang and Morrow, 1999a, Tang and Morrow,

1999b, Maas et al. 2001, Robertson et al., 2003, Lohardo et al, 2008, Morrow et al. 2008,

Pu et al., 2008, Kumar et al., 2010, Pu et al., 2010). The initial work centered on the

observation that significantly diluting the salinity of injected water produced incremental

oil after core had reached residual saturation by conventional waterflooding.

The mechanism(s) is still a matter of debate (Austad et al. 2010, Kumar et al., 2010, Lee

et al, 2010, RezaeiDoust et al. 2010, Sorbie, 2010), however, the proposed mechanisms

can be divided into two general classes. The first class is termed „physical‟ and based on

the observations of associated produced fines and the extensive work on the salinity

shock effect in sandstones, where abrupt reduction in salinity triggers release of clays

from grain surfaces with subsequent migration under flow and permeability reduction as

mobile fines plug pore throats (formation damage). Since most of the laboratory tests

were conducted in Berea Sandstone, the well-documented dilution factors required to

produce migration of fines served as the basis of later experimental protocols. The

„physical‟ mechanisms are 1) redirected flow and displacement of additional or by-passed

oil or micro-conformance (Bernard, 1967), and 2) oil associated with displaced fines

creating incremental production (Tang and Morrow, 1997, Pu et al., 2010). The second

class of mechanism is chemical in nature, and involves some variation on mineral

surface-oil interactions. This class is based on observations of incremental oil without

fines production. This includes changes in wettability, IFT, desorption of oil from



2

mineral surfaces and similar variants (Alotaibi and Nasr-El_Din, 2009, Ligthelm et al.

2009, Rivet et al., 2010, Austad et al., 2010, Lee et al., 2010, Sorbie, 2010).

A limited number of field examples using low-salinity floods have also been reported in

the EOR literature (Webb et al. 2004, McGuire et al. 2005, Robertson, 2007, Seccombe et

al., 2008, Lager et al., 2008, Seccombe et al., 2010, Skrettingland et al., 2010, Vledder et

al., 2010). These field studies of low-salinity water flooding have been confined to single

fields, or several adjacent fields. McGuire et al., (2005) reported that single well

chemical tracer tests performed in Alaska produced favorable results in sandstone

reservoirs with increases between 6 to 12% OOIP. This was expanded to interwell field

experiments in the same interval with low-salinity injection (Lager et al., 2008,

Seccombe et al., 2010). Vledder et al. (2010) reported on successful field scale

application of low-salinity injection based on observations of dual step watercut

development and attributed the effect to changes in wettability. However, Skrettingland

et al. (2010) reported on a series of lab and single well tracer tests in the North Sea that

showed no appreciable increase in recovery with low-salinity injection.

Robinson (2007, 2009) used data from three Minnelusa fields to analyze the recovery at

0.3 pore volumes of injection. He found that the early performance as measured by

recovery factor was related to salinity ratio (ratio of initial to injected salinity) for the

fields studied. He concluded that injection of low-salinity water into Minnelusa fields

slightly increased the recovery factor, but cautioned that his results were tentative due to

limited data on water chemistry and geology. Towler and Griffith (1998) used data

production data from 20 Minnelusa fields of which 19 were flooded with low-salinity

water. They concluded that the fields flooded with low-salinity water had higher

recoveries compared to the single normal waterflood, but cautioned they lacked sufficient

control to perform a meaningful statistical analysis (Towler and Griffith, 1998). Pu et al.

(2010) reported laboratory results for Minnelusa core in which 30-fold dilution of

injected water produced 5.8% OOIP in incremental oil suggesting low-salinity injection

into the Minnelusa fields may provide additional production.

Minnelusa Geology and Production

In Wyoming, the Pennsylvanian-Permian age quartz-rich sandstones of the Minnelusa

Formation have a high degree of geologic similarity between reservoirs. The Minnelusa

production is mostly from numerous small fields (<10MMBO OOIP) that are located in

the northeastern portion of the Powder River Basin. Early data showed the average

recovery factor was 28.4% OOIP under waterflood (Basko and Mulholland, 1976), which

has increased to 53.6% OOIP with continued waterflooding based on 2008 production

data. Many of these fields were flooded with lower salinity water based on low lifting

and conditioning costs (Towler and Griffith, 1998). There was no expectation of

improved production as these fields were waterflooded before low-salinity injection was

recognized as an EOR technique. The Minnelusa formation water has highly variable

salinity, but formation water salinity is generally much higher than the shallower source



3

of injection water (Lance and Fox Hills) and so offers an opportunity to evaluate the

potential for increased recovery by low-salinity waterflooding relevant to Wyoming.

Figure 1 shows the stratigraphic relationship between the Minnelusa and other important

Wyoming reservoir rocks and the typical reservoir geometry with the major reservoir

interval of the B sand interbedded between low porosity dolomite layers. Mack and

Duvall (1984) reported the average Dykstra-Parson coefficient for 15 Minnelusa fields

was 0.75 indicating a high level of heterogeneity in reservoir properties. The reservoir

caprock is the Opeche shale which forms the stratigraphic trap for these reservoirs

(James, 1989). Primary production is small since there is little gas in Minnelusa

reservoirs. Most oil production occurs in the first 0.4 pore volume of fluid production

after which the oil cut falls rapidly (Figure 2). Most Minnelusa fields currently in

production have water cuts above 95%.

Wyoming Oil and Gas Conservation Commission (WOGCC) records include waterflood

application date and the source formation for the injection water. Additional information

can include formation water chemistry, oil gravity, production histories and initial oil

saturation. The data were analyzed to identify and quantify any additional production

compared to fields flooded with reservoir brine. Fields were provisionally classified as

saline (waterflooded with formation water), mixed (combination of formation and low-

salinity water) and low-salinity (injection water from the Lance or Fox Hills formations).

Based on this classification, 55 fields are low-salinity, 52 mixed salinity and 23 saline. In

order to evaluate the impact of low-salinity waterflooding, we calculated the recovery

factor for fields in which oil and water production, total pore volume and original oil in

place (OOIP) were reported (Towler and Griffith, 1990, Hochanadel et al., 1990, Mack

and Duvall, 1990). Fields with recovery factors greater than 80% were removed from

further analysis as questionable leaving 61 fields with sufficient data. The location of the

fields is shown in Figure 3.



4

Figure 1. Left figure shows the stratigraphic relationships of Pennsylvannian-Permian

reservoir rocks in Wyoming, from Markert and Al-Shaieb (1984). Right figure shows a

typical Minnelusa well log with mineralogy from James (1989).

Figure 2. Typical production curve for a Minnelusa field showing recovery factor and oil

cut versus pore volumes (of produced fluid). Most oil production takes place over 0.4

pore volumes.



5

FIELD DATA

Water Chemistry

The USGS and WOGCC databases were queried for the chemistry of Minnelusa, Lance

and Fox Hills Formations water chemistry data (Breit, WOGCC). The samples selected

were either drill stem tests (DST) or produced water for samples that had API well

number and charge balance values of <±10%. Sample temperatures were calculated

based on the perforated interval and the geothermal gradient derived from WOGCC DST

data. When there were multiple samples from a single well, preference was given to

production over DST samples and the earliest (pre-waterflood) production sample over

later production samples. These criteria produced 518 Minnelusa and 95 Lance/Fox Hills

samples from existing fields and wildcat wells. This dataset was reduced to a single

sample per location by choosing the oldest production or DST sample for each API

number. This produced a total of 374 Minnelusa data points that were contoured for

salinity (see Figure 3).

The salinity of Minnelusa formation water ranges from 1134 to 261,982 ppm. The

distribution of salinity is strongly dependent on location with the lower salinity water

found in the northeast portion of the basin and increasing salinity to the southeastern and

deeper portions of the formation. The Minnelusa formation water is primarily composed

of Na and Cl for the higher total dissolved solids (TDS) samples with lower TDS samples

dominated by Ca and SO4. The lower salinity has been attributed to dilution of the

connate saline brine by invasion of meteoric water from the Black Hills to the east. The

other principal control on the water composition is equilibrium with the mineralogy of

the formation. The major minerals are quartz, dolomite and anhydrite and trace amounts

of clay (illite). The most soluble minerals are anhydrite (CaSO4) followed by dolomite

(CaMg(CO3)2).

In contrast, the Lance and Fox Hills Formations, which is the source of the low-salinity

injection water, is composed of Na and HCO3 with much lower TDS, (300 to 6000 ppm)

and variable amounts of Ca, Mg, Cl and SO4. The primary mineral in these formations

are quartz, carbonate cements and small amounts of clay.



6

Figure 3. Map of locations of selected Minnelusa oil fields with salinity contours in

color. Map was used to estimate salinity for those fields lacking water chemistry

analyses (See Table 1). Cartography by B. Reyes.

Production Data

Hochanadel et al. (1990) compared 32 Minnelusa fields with polymer floods to 24 fields

with only waterflood and found that fields flooded with polymer recovered an additional

7.5% OOIP. This figure agrees with Thyne et al. (2009), who used four additional fields

and used 2008 production data to show polymer floods recovered an additional 8%

OOIP. However, as shown in Figure 4, the distribution of the recovery factors is bimodal

with a number of fields having low recovery (<30%) forming a separate group of 9 fields

(1 polymer and 8 regular waterfloods). Inclusion of the fields with lower recovery

produces an average recovery of 41% OOIP for waterflood only and 49% for fields with

polymer application, biasing prior analyses of the impact of polymer flooding. The

reason for low recovery in some fields is unknown at present, but these low recovery

fields were not included in further analysis since they do not appear to represent “typical”

behavior.



7

Removing the low recovery fields left 51 fields (listed in Table 1 and shown in Figure 4),

which have a roughly normal distribution of recovery factors. These fields include cases

of low-salinity and regular flooding with and without polymer treatment. Thirty-one of

the 51 fields had polymer injection with relatively short periods of application (up to 1

year). These fields usually had lower API gravity oil and polymer treatments were

designed to produce a more favorable mobility ratio (Surkalo and Pitts, 1986, Hochanadel

et al., 1990, Brady et al., 1998).

Figure 5 shows the distribution of recovery factors for the fields with greater than 30%

recovery divided into fields that experienced a chemical treatment and those that did not.

The data show there is a slightly higher mean value for fields that had chemical injection

(53.4% versus 50.9% OOIP), but the difference is not significant given the standard

deviations for each group.

Figure 4. Upper histogram shows the bimodal distribution of recovery factor for 60

Minnelusa fields, while lower histogram shows the distribution of recovery factor for the

52 fields with recovery values greater than 30%.

2008 Recovery Factor (% OOIP)

Fre

qu

en

cy

706050403020

12

10

8

6

4

2

0

Mean 45.91

StDev 13.18

N 60

2008 Recovery Factor (% OOIP)

Fre

qu

en

cy

7060504030

12

10

8

6

4

2

0

Mean 52.43

StDev 9.939

N 51



8

Figure 5. Histogram of the recovery factors for 51 fields with polymer injection and

fields with waterflooding only for fields with greater than 30% recovery.

The behavior of oil production expressed as recovery factor versus total fluid production

(as pore volume displaced) was also examined to determine typical patterns and identify

any differences between fields with polymer application and standard waterflood. Using

total produced fluid (water + oil) in terms of pore volumes allows better comparison

between fields by effectively normalizing recovery to field displacement (Towler and

Griffith, 1998). Figure 6 shows the recovery factor versus pore volume for all fields with

complete production histories. The cumulative oil and water production data for 2008

were used (formation volume factor of 1.05 for oil and 1 for produced water) to calculate

pore volume values for each field. Fields that were waterflooded prior to 1974 were not

included since there were no data on early water production making calculation of pore

volumes uncertain.

The 37 Minnelusa fields with complete production histories show differences in the slope

within 0.25 pore volumes. Fields with steeper curves reach larger recovery factors during

continued waterflooding. Most of the oil is produced in the first 0.5 pore volumes with

another 15% between 0.5 and one pore volume. After that point the increase in produced

fluids is small with a fairly flat curve out to as much as 3 pore volumes. The production

curves for this group of fields show no apparent difference between standard waterfloods

and those fields that had application of polymer. Therefore, the two groups of fields

(polymer treated and no-polymer waterfloods) will be combined for purposes of

preliminary analysis.



9

Figure 6. Recovery factor versus pore volumes of produced fluids (oil+water) for 37

Minnelusa Fields. Cool colors are waterfloods with polymer application and warm colors

are waterfloods without polymer.

Preliminary Evaluation

Preliminary analysis was conducted by comparing the average recovery factors of various

groups of fields (Table 2). The average recovery for all fields is 52.4% OOIP, fields with

low salinity injection is 52.2%, and 52.6% OOIP for fields with saline injection (saline +

mixed classes). The data can be further evaluated by breaking the fields into other

groupings. For instance, the data from 20 fields without polymer treatment show there is

no difference between average recovery factor for fields with low-salinity injection

(50.8%) compared to saline injection (51.4%). The difference in average recovery

factors between fields that had polymer treatment and low salinity injection (54.7%)

compared to fields with polymer and saline injection (52.9%) is also very small. Given

the uncertainty in the calculation of recovery factors (discussed in more detail below),

differences of several percent are not considered significant. It appears that regardless of

what groups of fields are considered there is no significant difference in recovery factors

between fields with low salinity injection and those with saline brine injection.



10

Table 1. List of fields used in this study with accompanying data.

Field Discovery WF

permit API WF

type

Inj.

Water

Inj.

TDS

FW

TDS

OOIP

(bbls) PV (bbls)

Source CANDY

DRAW 1985 1992 25 P 1 42,732 6,991,000 10,775,000

A DEADMAN

CREEK 1973 2001 23

P 1 19,500 5,871,000 8,754,000

A

HAMM 1967 1984 20

P 1 81,000 13,547,000 18,590,000

A POWNALL

RANCH 1960 2001 25

P 1 746 18,582 5,270,000 6,201,000

A ROZET

WEST 1967 1984 22

P 1 86,728 22,689,000 29,430,000

B SIMPSON

RANCH 1971 1979 20

P 1 9,920 1,825,000 2,513,000

B STEWART

EAST 1980 1982 21

P 1 81,000 2,009,000 3,313,000

B SWARTZ

DRAW 1981 1987 23

P 1 11,475 4,423,000 7,066,000

A TIMBER

CREEK N. 1978 1982 21

P 1 760 136,313 4,582,000 7,807,000

A

ASH 1987 1992 20

P 2 67,382 2,170,000 2,568,047

C

DEER FLY 1984 1987 26

P 2 19,748 2,961,000 4,457,000

A

EDSEL 1981 1984 21

P 2 18,450 8,673,000 12,886,000

A RAINBOW

RANCH N. 1973 1990 28

P 2 340 50,000 8,130,000 12,428,000

A

SHARP 1975 1990 26

P 2 73,223 2,180,000 3,789,000

B

STEWART 1965 1977 22.5

P 2 86,691 42,250,000 67,883,000

B WAGON

SPOKE 1972 1998 28

P 2 56,000 6,095,000 8,282,000

A

ALPHA 1987 1989 23.8

P 3 4234 4,637 11,449,000 23,183,687

C

BIG MAC 1985 #N/A 22

P 3 38,000 2,646,000 3,970,000

A DILLINGER

RANCH 1964 1979 36

P 3 121,424 35,500,000 51,055,000

B

GLO 1984 1984 20

P 3 81,000 2,075,000 4,073,000

A

GLO NORTH 1985 1984 22

P 3 16,600 4,528,000 6,084,000

A

KIEHL 1972 1984 21

P 3 11,146 9,042,000 15,364,000

A

KIEHL WEST 1985 1982 24

P 3 5,730 1,519,000 2,079,000

A KUEHNE

RANCH 1965 2005 23

P 3 50,161 9,250,000 14,725,000

A KUEHNE

RANCH SE 1967 2005 23

P 3 69,384 9,468,000 13,721,000

A

LILY 1985 1969 21

P 3 5,570 5,205,000 6,974,000

A LONE

CEDAR 1985 2001 26

P 3 105,000 7,724,000 9,034,000

A

OK 1973 1981 30

P 3 7,960 6,300,000 8,565,000

A RIGHT A

WAY 1982 1994 20

P 3 131,170 2,094,000 4,176,000

A

THOLSON 1969 1982 23

P 3 150,000 9,566,000 13,392,000

A

VICTOR 1985 1980 21

P 3 10,000 3,167,000 4,453,000

A



11

Table 1. Continued.

Field Discovery WF

permit API WF

type

Inj.

Water

Inj.

TDS

FW

TDS

OOIP

(bbls) PV (bbls)

Source

BASIN NW 1965 1969 35 WF 1 2,957 4,564,000 7,611,000

A DUVALL

RANCH 1964 1971 29 WF 1 154,243 26,900,000 33,460,000

C

GUTHREY 1963 1968 27 WF 1 38,463 12,500,000 19,900,000

A

HALVERSON 1961 1971 24 WF 1 191,593 40,700,000 53,861,000

C HAWK

POINT 1986 1988 39 WF 1 2740 110,000 10,015,000 13,108,639

C

LITTLE

MITCHELL

CREEK 1967 1969 27 WF 1 46,391 21,000,000 29,480,000

B

MAYSDORF 1976 1991 38 WF 1 128,000 8,648,000 14,128,000

B MELLOTT RANCH 1960 2001 28 WF 1 4,892 15,400,000 18,890,000

B

RAVEN CREEK 1956 1992 33 WF 1 131,971 73,790,000 101,382,000

B

REEL 1962 1987 33 WF 1 87,744 20,602,000 33,968,000

B

RENO 1965 1967 37 WF 1 7,123 23,500,000 36,401,000

B ROURKE GAP 1973 1974 30 WF 1 110,000 8,859,000 10,445,000

B

SPRING HOLE 1983 1987 22 WF 1 136,313 2,661,000 3,877,000

A

WALLACE 1966 1972 29 WF 1 23,619 18,400,000 29,460,000

B

WIDGE 1988 1990 40 WF 1 150,000 3,686,000 5,850,794

C INDIAN TREE 1989 1992 20 WF 2 81,000 3,841,700 5,682,357

C

PICKREL RANCH 1965 1972 36 WF 2 195,992 4,510,000 7,393,000

B

PRAIRIE CREEK S. 1986 1987 20 WF 3 10,000 1,858,000 2,728,000

A

ROBINSON RANCH 1958 1971 26 WF 3 11,966 13,861,000 23,913,264

B

P = polymer, WF = waterflood, 1 = low-salinity, 2 = mixed, 3 = saline

Sources, A= Hochanadel et al., B= Mack and Duvall, C = Towler and Griffith

Values in red are based on interpolation from map. Timber Creek North field is listed as

Spring in cited reference.



12

Table 2. Average value for recovery factor for Minnelusa Fields.

Category %OOIP2008 n

All fields 52.4 51

Low Sal 52.2 25

Saline 52.6 26

All WF (no polymer) 50.9 20

Saline WF 51.4 4

Low sal WF 50.8 16

All polymer 53.4 31

Saline polymer 52.9 22

Low sal polymer 54.7 9

Controls on Recovery

We recognize that using averages of recovery factors for groups of fields is only a

preliminary evaluation since there are many variables may control recovery. The 51

Minnelusa fields have a wide range of oil gravity, OOIP, initial oil saturation and net pay,

all factors that may influence recovery. Figure 7 shows the recovery factor versus the

API gravity, OOIP, initial saturation and net pay for fields with low salinity and saline

waterfloods. The data show that the application of polymer does appear to have some

positive effect on recovery for fields with oil gravity between 20 and 22, but otherwise

there is little effect. The recovery factor does appear slightly higher for smaller fields

(<5MMBO) with polymer application compared to normal waterfloods, however, as with

oil gravity the correlation is weak. Initial oil saturation appears to have no correlation

with recovery, nor does net pay. It does not appear that API oil gravity, OOIP, initial oil

saturation or net pay are correlated with recovery, nor does application of polymer

produce a significantly different response in most fields. Other factors including number

of wells, permeability, porosity, well spacing, pore volumes flooded, duration of

flooding, and depth were also evaluated to see if they influenced the recovery, however

none of these factors appeared to correlate to recovery. The variation in recovery factors

between fields must be due to other factors such as conformance or geological

heterogeneity.



13

Figure 7. Plots of API gravity, initial oil saturation, OOIP and net pay for Minnelusa

fields with different symbols for fields with polymer injection and fields with only

normal waterflooding.

FURTHER EVALUTION OF FIELD DATA

The laboratory and field work cited in the introduction section examined many oil-brine-

rock systems over a range of dilution factors. While no minimum amount of dilution has

been established for increased oil production, the experimental data to date show

recovery can be increased with as little as 2.5-fold dilution (Pu et al., 2010) and that

increasing dilution increases incremental production (Alotaibi and Nasr-El-Din, 2010,

Loahardjo et al. 2007, Yousef et al., 2010). Robertson (2007) used the salinity ratio (SR),

defined as the ratio of salinity of injected water to salinity of formation water, to evaluate

the effect of low-salinity injection on recovery. He was able to calculate the ratio for

three Minnelusa fields, Moran, West Semlek and North Semlek at 0.3 PV. These values

ranged between 0.08 and 0.16 and plotting recovery factor versus salinity ratio showed a

weak trend of increasing recovery with lower salinity ratio (lower salinity of injection

water to formation water).



14

Since some Minnelusa fields have fairly fresh formation water, we use the salinity ratio

to quantify the degree of dilution and better evaluate low salinity injection. These results

are plotted in Figure 8 for 51 fields (Table 1). For calculation purposes, injection water

classified as mixtures assumed injection water salinity was a 50:50 mixture of low-

salinity and formation water. In some cases the low-salinity and formation water

chemistry was known and used to more precisely calculate salinity ratio. If the low-

salinity water chemistry was not known, the average TDS of the Lance and Fox Hills

water analyses (2100 mg/L) was used as the injection salinity.

The data include fields that were waterflooded with formation water (salinity ratio =1),

fields waterflooded with a mixture of low-salinity and formation water (salinity ratio

around 0.3-0.7) and fields waterflooded with low-salinity water (salinity ratio between

0.006 and 0.2). This study had many more saline waterfloods to establish a baseline than

earlier work (Towler and Griffith, 1998). The range of recovery factors is almost exactly

the same for all three groups and the data as a whole show no correlation between degree

of dilution and recovery. Since the laboratory and field data show incremental recovery

at salinity ratios as high as 0.4, it appears there are other factors in the Minnelusa

reservoirs that make low-salinity injection ineffective.

Figure 8. Dilution factor (salinity ratio) versus recovery factor for 51 Minnelusa fields.

Dilution factor of 1 is for fields with no dilution (injection of formation water), shaded

area highlights fields with dilution factors equal or less than 0.1 (10-fold dilution).



15

The low-salinity injection experimental work shows that the effect of low-salinity

injection in tertiary mode is relatively rapid and can produce incremental recovery after

regular waterflooding has reached residual oil saturation. In the Minnelusa fields

injection of low-salinity water was not used as an enhanced oil recovery mechanism, but

instead was employed from the start of waterflooding, or shortly after, making the

identification of incremental oil production impossible. The secondary mode

experiments are more applicable to the field examples. Experimental results in the

secondary mode have shown increased production overall compared to normal salinity

injection or normal salinity plus incremental (Loahardjo et al. 2007, Pu et al., 2010). We

have seen that higher ultimate recovery in Minnelusa fields is apparent in the slope of the

curves when plotting recovery factor versus pore volume.

The oil and water production data from the WOGCC for 2008 were used (formation

volume factor of 1.05 for oil and 1 for produced water) to calculate current pore volume

of produced fluids for each field. Figure 8 shows the production curves for Minnelusa

fields for which most or all of the production history was available and which had at least

10-fold dilution during injection of low-salinity water. We can use both the early slope

and recovery factor at the same pore volume to evaluate the effect of low-salinity

injection. The curves show that the trend of cumulative production with injection

increases as expected, but there is no evident difference based on injection salinity.

Figure 9. Recovery factor versus pore volume for selected Minnelusa Fields with at least

10-fold dilution factors. Cool colors are saline or mixed waterfloods and warm colors are

low-salinity waterfloods. Shaded area encompasses range of waterflood recovery factors.



16

Recovery Factor-Independent Analysis

The recovery factors are based on the reported OOIP, reservoir pore volume and current

production data. All three values have some uncertainty. The largest uncertainty is in the

reported OOIP and reservoir pore volumes. These values were taken from compilations

of values reported to WOGCC during unitization hearings. The values are based on

material balance or volumetric calculations, which have uncertainties in the calculations.

In addition, these hearings are typically held early in the development of a field, and if

later development expands the extent of the field, may represent lower than actual values.

This reserve growth in discovered fields has been well documented (Schmoker and Klett,

2000). The result is that recovery factors based on OOIP and field pore volumes reported

during unitization hearings may be too high. While this study excluded recovery values

above 80%, the uncertainty still remains. Therefore, we used a metric independent of

recovery factor to evaluate low-salinity injection.

The metric chosen was the normalized cumulative production at the point of water

breakthrough. We calculated this point by plotting normalized production over field life

versus % oil cut. This analysis was restricted to fields were the complete production

history was available (22 fields). Figure 10 shows a typical plot. We chose the break

point (A) that represents water breakthrough in terms of normalized production as a

measure of performance. If low-salinity injection produces more oil by increasing oil

mobility, we anticipate the breakthrough points would have a higher value compared to

fields without dilution.



17

Figure 10. Typical curves of normalized cumulative oil production versus oil cut (%).

Some Minnelusa fields have a natural water drive (Mack and Duvall, 1984). Fields with

water drive can have higher early water production since the water drive has essentially

the same effect as water injection wells. Those fields were identified by using the oil cut

at 0.05 normalized production. Fields with values less than 97% oil cut at 0.05

normalized production were not included in comparison analysis. Table 3 lists the

breakthrough points for 19 fields with low-salinity injection (8) and those with mixed or

formation water injection (11).

The data show that there is no increase in the normalized production at break-through for

fields with low-salinity injection compared to those with mixed or saline injection.

Inspection of the 19 water-cut curves (not shown) showed the same patterns for change in

water-cut with production for fields with low-salinity injection and fields with saline

injection. Nor did we observe dual-step water-cut profiles associated with wettability

changes from low-salinity injection (Vledder et al. 2010). An interesting observation

from this analysis is that fields that apparently have a natural water drive defined by the

criteria discussed above have a significantly delayed water breakthrough. These fields

also have a higher average recovery factor compared to the other fields (approximately

5%).



18

Table 3. Normalized production after water breakthrough.

Field

Inj. Water

Type

Norm. Cum. at

Breakthrough

Average

Break.

Pt.

Oil Cut

at 0.05

Hawk Point 1 0.44 99.4

Candy Draw 1 0.25 99.9

Simpson Ranch 1 0.25 99.6

Spring Hole 1 0.25 99.8

Moran 1 0.1 97

Widge 1 0.2 97

Timber Cr. N. 1 0.3

98

Swartz Draw 1 0.1 0.24 96

Ash 2 0.5 99.9

Indian Tree 2 0.25 98

Deer Fly 2 0.05 99.7

Big Mac 3 0.2 99.7

Kiehl W 3 0.6 99.9

Lily 3 0.18 99

Lone Cedar 3 0.2 99.7

Prairie Cr. S. 3 0.5 98.7

Right A Way 3 0.1 99.9

Glo N 3 0.6

96

Victor 3 0.3 0.32 99.8

Fields w/ water drive*

Stewart E. 1 0.65 80

Edsel 2 0.5 85

Alpha 3 0.8 0.65 87

*= based on lower early oil cut

LABORATORY STUDIES

We undertook laboratory studies that used Minnelusa rock, oil and synthetic brine in

order to better understand the results from the field data that showed no increase in

production in Minnelusa reservoirs using low-salinity injection water. The experimental

work used Minnelusa core and crude oil from the formation, together with synthetic

formation brine. These experiments injected synthetic formation water into oil-saturated

core. Both single and two phase experiments were performed. The single phase

experiments provide a baseline to evaluate the two phase experiments.

Experimental Procedure

Crude oils from two Minnelusa fields, Raven Creek and Gibbs, were used. The Raven

Creek reservoir bottomhole temperature is about 75 C while Gibbs is about 68 C. The

two oils cover the range of oil gravities for many Minnelusa fields. Crude oil properties

are listed in Table 1. Crude oil was centrifuged at 6000 rpm for 2 hours and filtered to



19

remove water and sediments and vacuumed for 4 hours to remove the light ends. This

process can increase water wetness in the system benefiting the low-salinity effect (Tang

and Morrow 1997). Crude oil was stored in amber colored bottles in the dark to avoid

photochemical dissociation of the crude oil components. Minnelusa core plugs were

cleaned by soxhelet extraction for a week and dried in an oven at 100°C for 48 hours.

Core plugs were stored in a dry desiccator. Synthetic brine representing average

Minnelusa formation water composition was made from ACS grade chemicals and

distilled water, then vacuumed for two hours to remove dissolved gas before the

experiments.

Both single and two phase experiments were performed. The single phase experiments

provide the baseline to evaluate the two phase experiments. All core plugs were aged

with the synthetic brine (MNB) at room temperature for 7 days. Porosity of the core

plugs was calculated by subtracting dry weight of the core from the weight of the brine

saturated core. Next, the core plug was mounted in a Hassler core holder and synthetic

brine (2-3 PV) was injected to establish a constant pressure drop across the core, then

different flow rates (0.1, 0.2, 0.3 and 0.4 ml/min) were applied and the pressure across

the core used to calculate the brine permeability (Kb). Continuous pressure

measurements were made with Validyne transducer connected to a computer with

Labview software. Next, the cores were flooded with the crude oil (5 PV) to establish the

initial water saturation (Swi). Volume of brine displaced by the oil was used to calculate

the original oil in place (OOIP) and initial water saturation (Swi). Oil permeability was

measured at the Swi by using the same method used to measure the brine permeability.

Core plugs were removed from the core holder and aged in an aging cell for 10 days

(Tang and Morrow 1997). After aging, core plugs were re-mounted in a Hassler core

holder and flooded with fresh crude oil for about 5 PV (core was flooded in the same

direction used to establish Swi.

In the single phase experiment, the core plug was flooded with Minnelusa brine (MNB)

at 0.2 ml/min for more than 20 pore volumes (PV). The core plug was then flooded with

low-salinity brine (1%MNB) for another 10 PV at 0.2 ml/min. In tertiary mode

experiments, the oil-saturated core plug was flooded with MNB at 0.2ml/min to reach

residual saturation and then flooded with low-salinity brine until no more oil was

produced. In the secondary mode experiments, the core plug was flooded directly with

low-salinity brine (0.2ml/min) after preparation. During the experiments, the oil

production was measured at set time intervals. Effluent brine was collected as 8 ml

samples. Conductivity and pH were measured on all samples and chemical analyses were

performed on selected samples.



20

Brine, Oil and Rock properties

Brines

Brines were made from ACS grade chemicals and distilled water. Brine compositions are

listed in Table 4. Synthetic brines were vacuumed for two hours to remove dissolved gas

before the experiments.

Table 4. Brine composition.

Compound Formation Brine

(g/l)

Low Salinity

Brine (mg/l)

NaCl 29.803 298.03

CaCl2 2.106 21.06

Na2SO4 5.903 59.03

MgSO4 0.841 8.41

NaN3 0.100 1.00

TDS 38753 387.53

Core

Core plugs were drilled from a Donkey Creek Field Minnelusa whole core and dried in

an oven at 100°C for 48 hours. Air permeability was measured (confining pressure, 500

psi) and mineralogic compositions were determined by XRD and thin section. Core plugs

were stored in a dry desiccator. The core plug lithologies are quartz-rich sandstones with

minor amounts of anhydrite and dolomite (see Figure 11). There are very small amounts

of clay, identified in XRD as illite (see Figure 12). Core properties are listed in Table 5.

Table 5. Core properties.

Name Length

(cm)

Diameter

(cm)

Permeability

(mD)

Porosity

(%)

PV (ml)

A 7.695 3.775 1.74 7.44 6.40

B 7.664 3.787 12.02 11.17 9.64

C 7.624 3.779 4.18 9.15 7.82

D 7.667 3.766 43.49 13.85 11.83

E 7.637 3.781 10.51 6.74 5.779



21

Figure 11. XRD spectra of whole rock Minnelusa sample from core material showing the

presence of quartz, anhydrite and dolomite.

Figure 12. XRD spectra of clay separate fraction from the Minnelusa core sample

showing the clay is composed of illite.



22

Crude Oils

Raven Creek and Gibbs crude oils were used in all the experiments. The Raven Creek

reservoir bottomhole temperature is about 75 C while Gibbs is about 68 C. The two oils

represent near end-members for the range of oil gravities in Minnelusa fields. Crude oil

properties are listed in Table 6. Crude oil was centrifuged at 6000 rpm for 2 hours to

remove water and sediments. The oil was then filtered and vacuumed for 4 hours to

remove the light ends of the crude oils. This process can increase water wetness in the

system benefiting the low-salinity effect (Tang and Morrow, 1997). Some proposed

chemical mechanisms postulate polar components of the oil binding to mineral surfaces,

so removing non-polar components concentrates these active compounds. Crude oil was

stored in amber colored bottles in the dark to avoid photochemical dissociation of the

crude oil components.

Table 6. Crude oil properties.

Oil Density API Viscosity

(cp)

S.

(%)

A.

(%)

R.

(%)

Asph.

(%)

TBN TAN

Raven

Creek

0.8578 33.5 8.0 80.2 15.8 2.6 1.4 0.92 0.074

Gibbs 0.8834 28.7 11.5 61.5 23.4 3.2 10.4 --- --- S = saturates, A = aromatics, R = resins, Asph. = Asphaltenes, TBN = total base number, TAN = total acid

number

Single Phase Experiment

Single phase experiments were performed to obtain a baseline for the two-phase

experiments. Core was prepared as described in the experimental procedures section and

aged and flooded at 60 C. Core was flooded sequentially with approximately 25 pore

volumes of high salinity brine followed by a similar amount of low-salinity brine to

represent tertiary mode application (Figure 13).

Pressure across the core increased rapidly at the start of the high salinity brine flood and

then decreased during the entire high salinity brine flood. The abrupt pressure change

coincident with change in salinity is an experimental artifact and not related to salinity of

injected water. Low-salinity brine injection resulted in continued pressure decline along

a similar slope compared to the high salinity brine flood. The single phase experiments

show that the initial pH value of about 7.8 changed after injection of low-salinity brine to

about 8. The change in pH and brine salinity, as measured by conductivity, takes place

over about 2 pore volumes. The simultaneous change of pH and brine salinity indicates

that pH is controlled by water-rock reactions. Fines were not observed during high

salinity or low-salinity brine floods.



23

(a) (b)

Figure 13. Pressure drop, conductivity and the pH variation during a single phase

experiment, secondary mode experiment (core directly flooded with the low-salinity

brine) with core A.

Two-Phase Experiments - Secondary Mode

Gibbs and Raven Creek crude oils were used in these experiments. Cores were aged as

described in the experimental procedure. Cores were flooded with 1%MNB directly for

more than 20 PV at 0.2 cm3/min. Oil recovery, pressure drop, pH and the conductivity of

the effluent samples were measured during the floods. Figures 14, 15 and 16 show oil

recovery, pressure drop across the core, pH and the conductivity variation during the

secondary mode experiment with Gibbs and Raven Creek crude oils. Oil recovery was

observed for a longer time period during the low-salinity brine injection. The pH of the

effluent brine was lower than the pH of the effluent brine from the low-salinity flood in

the tertiary mode experiments. Conductivity of the effluent brine decreased slowly with

the low-salinity brine injection. Again, conductivity of the effluent brine is higher than

the conductivity of low-salinity brine, which indicates dissolution of minerals from the

Minnelusa cores.

The results of the secondary mode experiment with Gibbs oil is shown in Figure 14. The

pressure increase and pH change was fairly rapid. Pressure increased rapidly to almost

90 psi, then fell slowly over the course of the experiment. Oil recovery was also rapid,

over the first 2 pore volumes. The initial pH of 7 rapidly increased to about 7.8, then

remained stable. Conductivity fell rapidly over the 4 pore volumes, but did not stabilize

until about 12 pore volumes. The total production was 61.5% OOIP.



24

(a) (b)

Figure 14. Oil recovery, pressure drop, conductivity and the pH variation during two

phase secondary mode experiment (core directly flooded with the low-salinity brine) with

core plug E.

The results from the secondary mode experiment with the lighter Raven Creek oil is

shown in Figure 15. The recovery (68.2% OOIP) is also rapid (2 pore volumes),

concurrent with the change in conductivity and pH. The pressure gradient across the core

was very similar with a slow decline that continued after oil production stopped. The pH

behavior was similar to the Gibbs oil experiment, but pH stabilized at a lower value (7.4).

It appears that the inclusion of oil in the experiments changes pH behavior, presumably

because oil has natural acid and base groups that can act as buffers.

(a) (b)


phase secondary mode experiment (core directly flooded with the low-salinity brine) with

core plug C and Raven Creek oil.



25

Two-Phase Experiments - Tertiary Mode

Gibbs and Raven Creek crude oils were used in the tertiary mode experiments. Cores

were prepared as described in the experimental procedures section and aged and flooded

at 60 C. Cores were flooded sequentially with approximately 25 pore volumes of high

salinity brine followed by a similar amount of low-salinity brine to represent tertiary

mode application. Oil recovery, pressure drop, pH and the conductivity of the effluents

were measured during the floods. Chemical analyses of effluent were performed for

experiments and are discussed in the geochemical modeling section.

Figures 16 and 17 show the oil recovery, pressure drop, pH and the conductivity data

measured during the tertiary mode experiments with Gibbs and Raven Creek crude oil,

respectively. During the low-salinity brine injection very little to no oil recovery (0-1.2%

OOIP) was observed. Also the pH increase during the low-salinity brine injection was

small, similar to the pH increase of the single phase experiment. Conductivity of the

brine collected in the low-salinity flood is higher than that of low-salinity brine. This is

assumed to be due to dissolution of minerals from the Minnelusa cores.

(a) (b)


phase, tertiary mode experiment with core plug D and Gibbs oil.



26

(a) (b)


phase tertiary mode experiment with core plug B and Raven Creek oil.

Summary of Experiments

Table 7 shows the summary of all the experiments. The Minnelusa core plug

experiments produced recovery between 46 and 68.2% OOIP. The range of recovery

values is not unreasonable given the heterogeneity of this reservoir rock (Hochanadel et

al. 1990). These values are also near the higher end of the field recovery values.

Secondary mode experiments generally produced more total oil (average = 64.0% OOIP)

than the tertiary mode experiments (average = 48% OOP). The incremental recovery

using heavier Minnelusa oil (Gibbs) was low (1.2% OOIP), while no incremental

recovery was observed in the experiment using the lighter Minnelusa oil (Raven Creek).

There was no fines production observed during any of the experiments. Fines migration

has been proposed as a mechanism for the low-salinity effect (Tang and Morrow, 1999a,

Pu et al. 2010), and the low incremental production during low-salinity injection is

reasonable given the paucity of clay in the Minnelusa formation.

Table 7. Summary of experimental results.

Core Type Oil Kb (mD) Swi Ro Rot RT

A SP N/A 0.1981 N/A N/A N/A N/A

B TP/TM RC 2.78 30.50 46.23 0 46.23

C TP/SM RC 1.0218 11.76 68.15 N/A 68.15

D TP/TM GBS 7.9004 28.98 49.40 1.20 50.60

E TP/SM GBS 1.136 30.50 61.54 N/A 61.54

RC= Raven Creek, GBS = Gibbs, SP= single phase, TP = two phase, TM = tertiary mode, SM = secondary

mode, Kb = brine permeability, Swi = initial water saturation, Ro = recovery from regular WF, Rot =

recovery from low-salinity injection, RT = total recovery, all recoveries in % OOIP



27

GEOCHEMICAL MODELING

The injection of low-salinity water into the Minnelusa formation to improve recovery

depends on achieving sufficient dilution. While the dilution factor required to increase

production has never been well established, and in fact appears to be field specific, one

factor that may restrict dilution in the Minnelusa is the presence of anhydrite in the

reservoir. Anhydrite is a highly soluble salt and injection of low-salinity water is

anticipated to cause dissolution raising the salinity of the injected water and lowering the

dilution factor. In order to evaluate this effect we used geochemical modeling to

determine how much the salinity of the injected water might increase during water-rock

interaction.

Procedures

Geochemical models of water-rock interaction offer some insight into the low-salinity

process McGuire et al., 2005, Jerauld et al., 2006). The software used was Geochemist‟s

Workbench, which has been used previously for modeling low-salinity waterflooding

(McGuire et al. 2005). The model was first used in the 1D reactive-transport

configuration with the kinetic option for the three major minerals present in the

Minnelusa (quartz, anhydrite and dolomite). The model also includes a cation exchange

surface for which Ca and Na are exchangable.

Model Calibration

The model was calibrated by altering the dissolution rates from initial values (Palandri

and Kharaka, 2004) with the amounts of minerals based on thin section petrography. The

calibration was made to match effluent chemistry for the single-phase Minnelusa

waterflood conducted in tertiary mode (Figure 18). The calibrated kinetic values, mineral

abundance and cation exchange capacity are listed in Table 8. Figure 19 shows the

measured concentrations for dissolved Na, Cl, Ca and SO4 together with the predicted

concentrations (solid line). Chloride can be considered essentially conservative in this

brine-oil-rock system. These four solutes comprise 95+% of the solutes in the Minnelusa

brine. Figure 20 shows the predicted and actual TDS (salinity). The agreement is very

good and since the most important parameter is the salinity, this model is accepted.

Table 8. Parameters used for calibrated model from experiment #25.

Minerals

Volume

(%) K

calibrated

Surface

Area

(cm2/g) CEC (eq)

Quartz 82 1.00E-14 400

Anhydrite 4 5.00E-11 150

Dolomite-ord 5 1.00E-10 400

Cation exchange 6.00E-05



28

Figure 18. Effluent concentrations from experiment #25, single phase, Minnelusa brine

and core plug A.

Figure 19. Comparison of predicted (solid line) and measured values for Na, Cl, Ca and

SO4 from the single-phase experimen using Minnelusa brine and rock.

Figure 20. Comparison of predicted (solid line) and measured values for TDS from the

single-phase experimen using Minnelusa brine and rock.

0 +200 +400 +600 +800 +1000 +1200 +14000

5000

1e4

15000

2e4

25000

3e4

35000

4e4

45000

Time (min)

Dis

solv

ed s

oli

ds

(mg

/kg

), x

= 7

.41 c

m ŸŸŸŸŸŸŸŸŸŸŸŸŸŸŸ

Ÿ

Ÿ

ŸŸŸŸŸŸŸŸ



29

History Matching the Model

Next the calibrated model was used to history match the data from the two-phase

experiments that had analyses of the effluent water chemistry. Figure 21 shows the

measured concentrations for experiment #28 for the major solutes. Figure 22 shows the

actual and predicted concentration for TDS using the parameters from the calibrated

model. The calibrated model is a good match to experimental data, although there does

appear to be a small amount of dispersion in the tail of concentration after switching the

injected brine to low-salinity not explicitly accounted for by the model due to the

presence of two liquid phases. There are small differences between predicted and actual

Ca and SO4 data (not shown) that could be reduced by changing the anhydrite dissolution

kinetics, but the prediction of TDS is very good and is considered the most important

parameter for prediction.

Figure 21. Effluent concentrations during low-salinity brine flood of core plug D with

Gibbs oil.

Figure 22. Predicted versus actual TDS for two-phase experiments using parameters

from the calibrated model. Upper line is from experiment tertiary mode experiment and

lower line is for secondary mode experiment.

0 5 10 15 20 250

500

1000

1500

2000

2500

3000

3500

pore volume

Som

e fl

uid

com

pon

ents

(m

g/k

g)

Experiment # 28

œœœ œ œ œ œ œ œ œ œ œ œ+++ + + + + + + + + + +

,

,

,

, , , , , , , , , ,

Ca++

Mg++

Na+

0 5 10 15 20 250

500

1000

1500

2000

2500

3000

3500

pore volume

Som

e fl

uid

com

pon

ents

(m

g/k

g)

Experiment # 28

!

!! ! ! ! ! ! ! ! ! !

¢¢

¢¢ ¢ ¢ ¢ ¢ ¢ ¢ ¢ ¢ ¢

ƒƒƒ ƒ ƒ ƒ ƒ ƒ ƒ ƒ ƒ ƒ ƒ

Cl-

SO4--

SiO2(aq)



30

Field Scale Modeling

The geochemical simulations show that the most important water-rock reaction

influencing water chemistry is the dissolution of anhydrite (CaSO4). The injected water

is usually undersaturated with respect to anhydrite, while almost all the formation is

saturated. Dissolution of anhydrite will tend to increase the salinity of injected water

from the Fox Hills and Lance Formations. To explore the degree of salinity increase a

simulation was run using the example of the North Timber Creek field (Table 1). The 2-

D simulation was made using the calibrated model values for mineral abundance,

reaction rates and cation exchange surfaces. The formation water and injection water

chemistry was taken from the North Timber Creek field. This field had very saline

formation water (136,000 mg/L) and was injected with very fresh water (766 mg/L).

Using the calibrated single phase model parameters, injection of 28,000 gallons of low-

salinity water per day (average value for 1985-91 injection rate) for two years was

simulated. The results are shown in Figures 23 and 24. Figure 23 shows the change in

TDS during the 2 years of injection at the well node. The TDS of the formation water

initially decreases rapidly over the first 75 days of injection, but then as anhydrite

dissolves the salinity rebounds and stabilizes at about 4000 mg/L. It would take many

years of injection to dissolve all the anhydrite in the single well node. Even if that

anhydrite was eventually dissolved, dissolution of all the anhydrite in the field would

require hundreds of years, so the salinity will remain buffered by dissolution over the

operational life of the field.

Figure 23. Predicted TDS versus time for low-salinity brine injection over two years into

the North Timber Creek field. Assumptions described in text.

Figure 24 shows the results of the injection after two years in map view. The model

domain is 1000X1000 meters with a single injection well fully penetrating the sandstone

0 +100 +200 +300 +400 +500 +600 +700 +800

3000

4000

5000

6000

8000

1e4

2e4

3e4

4e4

5e4

6e4

8e4

1e5

Time (days)

Dis

solv

ed s

oli

ds

(mg/k

g),

x =

475 m

, y =

475 m



31

(10 meters thick). The figure shows that the injection salinity never gets below about

4000 mg/L even after two years of injection and that the lower salinity zone is only about

250 meters in radius from the point of injection.

Based on the results of geochemical modeling it seems unlikely that the salinity in the

field will decline to lower than about 4000 mg/L. The results also suggest that the

salinity ratio calculation may produce too low a value for some of the fields. However,

the overall range of salinity ratios would only be increased by a factor of 2 given the

assumption of 2200 mg/L TDS for most of the injection water. This is not considered a

significant difference since the previous laboratory work has shown incremental oil

production at salinity ratios of 0.4.

Figure 24. Map view of distribution of predicted TDS from low-salinity brine injection

for two years into the North Timber Creek field. Assumptions described in text.

Discussion

The analysis of field data, laboratory experiments and geochemical modeling provide

insight into the effectiveness of low-salinity injection on oil production in the Minnelusa

Formation of Wyoming. Previous work on low-salinity injection in the Minnelusa

Formation included evaluation of three fields (Robertson, 2007), 20 fields (Towler and

44000

44000

4400

0

84000

84000

8400

0

124000

124000

124000

1240

00

124000

Dissolved solids (mg/kg)

4000 1.4e54e4100 m

100 m

Co

lor

ma

sk D

isso

lve

d s

olid

s f

rom

6.5

19

mg

/kg

to

7.2

34

mg

/kg

with

Cya

n

Co

lor

ma

p D

isso

lve

d s

olid

s f

rom

40

00

mg

/kg

(T

urq

uo

ise

) to

1.4

e5

mg

/kg

(R

ed

) m

id 4

e4

mg

/kg

(Y

ello

w)

Co

nto

ur

Dis

so

lve

d s

olid

s f

rom

40

00

mg

/kg

to

1.4

e5

mg

/kg

by in

terv

al o

f 2

e4

Year 2



32

Griffith, 1998) and a single laboratory core flood. This work has added six more core

floods, geochemical modeling constrained by the core flood experiments and an

expanded analysis of recovery factors for 52 fields. We used three separate metrics,

including one that was independent of recovery factors, but found no indication that low-

salinity injection produced more oil during operations in the Minnelusa fields.

The prior quantitative analysis of Minnelusa field data using 3 fields suggested a

maximum increase in recovery of <2% OOIP (Robertson, 2007). The analysis by Towler

and Griffith (1998) used twenty wells, but could not make a statistically valid analysis

because they lacked sufficient controls (saline waterfloods). In our analysis we had

sufficient control to allow meaningful statistical analysis with similar numbers of fields

with high, low and no dilution. We used 2008 recovery factors as the primary metric to

compare different fields. We also used production curves normalized by pore volumes to

compare fields over short and long term. Both analyses showed no discernable increase

in oil recovery with low-salinity injection. Nor was there any difference in water

breakthrough times between the groups of fields. The latter analysis is independent of

recovery factors.

The total recovery of three secondary mode experiments is 64.0% OOIP, while the

tertiary mode experiments produced an average of 48% OOIP. The range in total

recovery in the laboratory experiments was 46.2 to 68.2%, while field results ranged from

32 to 70% with an average of 53%. The results from prior laboratory work on a single

Minnelusa core showed an incremental increase of 5.8% OOIP using Gibbs (heavy) oil

(Pu et al. 2010), after initial production of about 45% OOIP. Our laboratory results

showed initial recovery of about 46-68% OOIP with incremental recovery of 1.2%

(Gibbs oil) in only one of two cores. The variability between experiments may be due to

the lack of homogeneity in the rock samples used. Pu et al. (2010) used core with 14.6%

porosity, 78mD permeability and Swi of 8.2% compared to core D with porosity of

13.8%, permeability of 43.5mD and Swi of 29%. Generally, laboratory floods show

higher recovery for any EOR technique (CO2, chemical, etc.) than the field application.

This was the case on average for the experiments using Minnelusa core.

The high degree of laboratory variability is mirrored by the large range in field

recoveries. This is expected given the high degree of heterogeneity reflected by the

Dykstra-Parsons coefficient for Minnelusa fields that range from 0.6 to 0.8 (Hochanadel

et al., 1990). Given the high degree of variability in recovery factors between fields (30

to 70%) and the anticipated small effect of low-salinity injection, any positive gains in

production may be obscured in the natural variability between fields. These conclusions

are applicable for this brine-oil-rock system only. Other reservoir systems in Wyoming

may benefit from low-salinity injection, but at this time insufficient data are available for

that evaluation.



33

Conclusions

Analysis of recovery factors for fields with and without low-salinity injection

showed no significant increase in recovery due to low-salinity injection.

Comparison of recovery with pore volume showed no significant difference, as

did comparison of water breakthrough time.

Laboratory experiments showed little incremental recovery for the Minnelusa

rock-oil-brine system.

Geochemical modeling showed that while the dissolution of anhydrite found in

Minnelusa rocks will increase the salinity of injection water, the increase is not

sufficient to explain the lack of response to low-salinity injection.

The lack of increased production from low-salinity injection may be related to the

lack of mobile fines, the inherent properties of the Minnelusa brine-oil-rock

system or not apparent due to the natural variability in this reservoir system.



34

References

Alotaibi, M.B. and Nasr-El_Din, H.A., 2010, Effect of Brine Salnity on Reservoir Fluids

Interfacial Tension. SPE 121569.

Austad, T., RezaeiDoust, A. and Puntervold, T., 2010, Chemical Mechanism of Low

Salnity Water Flooding in Sandstone Reservoirs. SPE 129767.

Basko, D. and Mulholland, E.E., 1976, Statistical summaries of the Muddy and

Minnelusa reservoirs in the Powder River Basin, Wyoming. Twenty-eighth Annual Field

Conference, Wyoming Geological Association Fieldbook, pp. 273-275.

Boussour, S., Cissokho, M., Cordier, P., Bertin, H. and Hamon, G., 2009, Oil Recovery

by Low Salnity Brine Injection: Laboratory Results on Outcrop and Reservoir Cores.

SPE 124277.

Brady, C.K. and Lee, L.P. and Lee, S.K., 1998, A comparison of forecast and actual

production after a decade of field operations in the Alpha unit polymer-augmented

waterflood. SPE 39614.

Breit, G., USGS, http://energy.cr.usgs.gov/prov/prodwat/prodwat.htm.

Cissokho, M., Boussour, S. Cordier, Ph. Bertin, H. and Hamon, G., 2009, Low Salinity

Oil Recovery on Clayey Sandstone: Experimental Study. International Symposium of the

Society of Core Analysis, SCA2009-05.

Hochanadel, S., Lunceford, M.L. and Farmer, C.W., 1990, A comparison of 31

Minnelusa polymer floods with 24 Minnelusa waterfloods. SPE 20234.

Jadhunandan, P.P. and N.R. Morrow, 1995, Effect of Wettability on Waterflood recovery

for Crude-Oil/Brine/Rock Systems. SPE Reservoir Engineering, SPE 22597, p. 40-46.

James, S.W., 1989, Diagenetic History and Reservoir Characteristics of a Deep

Minnelusa Reservoir, Hawk Point Field, Powder, River Basin, Wyoming, in E.B.

Coalson, S.S. Kaplan, C.W. Keighin, C.A. Oglesby, and J.W. Robinson, eds.,

Petrogenesis and petrophysics of selected sandstone reservoirs of the Rocky Mountain

region: Rocky Mountain Association of Geologists, Denver, CO, pp. 81-96.

Jerauld, G.R., Lin, C.Y., Webb, K.J. and Seccombe, J.C., 2006, Modeling Low-Salinity

Waterflooding. SPE 102239.

Kumar, M., Fogden, A., Morrow, N.R. and Buckley, J.S., 2010, Mechanisms Of

Improved Oil Recovery from Sandstone by Low Salinity Flooding. International

Symposium of the Society of Core Analysis, SCA2010-25.

http://energy.cr.usgs.gov/prov/prodwat/prodwat.htm



35

Lager, A., Webb, K.J., Collins, I.R., and Richmond, D.M. , 2008, LoSal™ Enhanced Oil

Recovery: Evidence of Enhanced Oil Recovery at the Reservoir Scale. SPE 113976.

Lager, A., Webb, K.J., Black, C.J.J., Singleton, M., and Sorbie, K.S., 2008, Low Salinity

Oil Recovery – An Experimental Investigation. Petrophysics 49(1), pp. 28-35.

Lager, A., Webb, K.J., Collins, I.R., and Richmond, D.M., 2010, LoSal™ enhanced oil

recovery: Evidence of enhanced oil recovery at the reservoir scale. SPE 113976.

Loahardjo, N, Xie, X., Yin, P., and Morrow, N.R., 2007, Low salinity waterflooding of a

reservoir rock. SCA2007-29.

Lee, S.Y., Webb, K.J., Collins, I.R., Lager, A., Clarke, S.M., O‟Sullivan, M.O., Routh,

A.F., and Wang, X., 2010, Low Salinity Oil Recovery – Increased Understanding of the

Underlying Mechanisms. SPE 129722.

Ligthelm, D.J., Gronsveld, J. Hofman, J.P., Brussee, N.J., Marcelis, F., and van der

Linde, H.A., 2009, Novel waterflooding strategy by manipulation of injection brine

composition. SPE 119835.

Maas, J. Wit, K. and Morrow, N.R., 2001, Enhanced oil recovery by dilution of injection

brine: Further interpretation of experimental results. Proceedings SCA Annual Mtg.

Markett, J.C. and Al-Shaieb, Z., 1984, Diagenesis and evolution of secondary porosity in

upper Minnelusa sandstones, Powder River basin, Wyoming, in McDonald, D.A. and

Surdam, R.C., eds. Clastic Diagenesis: AAPGG Memoir 37, pp. 367-389.

Mack, J.C. and Duvall, M.L., 1984, Performance and economics of Minnelusa polymer

floods. SPE 12929.

Morrow, N.R. Tang, G-Q., Valat, M., and Xie, X., 1998, Prospects of improved oil

recovery related to wettability and brine composition. J. Petrol. Sci. Eng. 20, p. 267-276.

Patil, S. Dandekar, A.Y., Patil, S.L. and Khataniar, S., 2008, Low salinity brine injection

for EOR on Alaska North Slope (ANS). IPTC 12004.

Pu, H., Xie, X., Yin, P., and Morrow, N.R., 2008, Application of Coalbed Methane Water

to Oil Recovery by Low Salinity Waterflooding. SPE 113410.

Pu, H., Xie, X., Yin, P. and Morrow, N.R., 2010, Low Salinity Waterflooding and

Mineral Dissolution. SPE 134042.



36

RezaeiDoust, A., Puntervold, T., and Austad, T., 2010, A Discussion of the Low Salinity

EOR Potential for a North Sea Sandstone Field. SPE 134459.

Rivet, S.M., Lake, L.W., and Pope, G.A., 2010, A Coreflood Investigation of Low-

Salinity Enhanced Oil Recovery. SPE 134297.

Robertson, E.P., 2007, Low-Salinity Waterflooding to Improve Oil Recovery - Historical

Field Evidence. SPE 109965.

Robertson, E.P., 2009, Low-Salinity Waterflooding Improves Oil Recovery – Historical

Field Evidence. Journal of Petroleum Technology Jan., pp. 47-49.

Robertson, E.P., 2010, Oil Recovery Increases by Low-Salinity Flooding: Minnelusa and

Green River Formations. SPE 132154.

Schmoker, J.W. and Klett, T.R., 2000, Estimating potential reserve growth of known

(discovered) fields: A component of the USGS world petroleum assessment 2000. USGS

Digital Data Series #60, 37p.

Seccombe, J., Lager, A., Webb, K., Jerauld, G. and Fueg, E., 2008, Improved Waterflood

Recovery: LoSal™ EOR Field Evaluation. SPE 113480.

Sharma, M.M. and Filoco, P.R., 2000, Effect of Brine Salinity and Crude-Oil Properties

on Oil Recovery and Residual Saturations. SPE Journal 5 (3).

Skrettlingland, K., Holt, T., Tweheyo, M. T., and Skjevrak, I., 2010, Snorre low salinity

water injection – core flooding experiments and single well field pilot. SPE 129877.

Sorbie, K., 2010, A Proposed Pore-Scale Mechanism for How Low Salinity

Waterflooding Works. SPE 129833.

Surkalo, H. and Pitts, M.J., 1986, Polyacrylamide vertical conformance process improved

sweep efficiency and oil recovery in the OK Field. SPE 14155.

Tang, G-Q. and Morrow, N.R., 1997, Salinity, Temperature, Oil Composition, and Oil

Recovery by Waterflooding.SPE Reservoir Engineering, November, p. 269-276.

Tang, G-Q. and Morrow, N.R., 1999a, Influence of brine composition and fines

migration on crude oil/brine/rock interactions and oil recovery. J. Petrol. Sci. Eng. 24,

p.99-111.

Tang, G-Q. and Morrow, N.R., 1999b, Oil Recovery by Waterflooding and Imbibition -

Invading Brine Cation Valency and Salinity.Society of Core Analysts, SCA9911.



37

Thyne, G.D., Alvarado, V. and Murrell, G., 2009, Evaluation of Chemical Flooding in

the Minnelusa Formation, Powder River Basin, Wyoming, AAPG National Meeting,

Denver, Colorado, Abstracts, p.60.

Towler, B.F. and Griffith, D. B., 1999, Comparison of 20 Minnelusa Waterfloods in the

Powder River Basin. SPE 55683.

Vledder, P., Fonseca, J. C., Wells, T. Gonzalez, I., Ligthelm, D., 2010, Low salinity

water flooding: Proof of wettability alteration on a field scale. SPE 129564.

Webb, K., Lager, A., and Black, C., 2008, Comparison of High/Low Salinity Water/Oil

Relative Permeability. International Symposium of the Society of Core Analysis,

SCA2008-39.

WOGCC, 2011. http://wogcc.state.wy.us/

Yildiz, H.O., Valat, M., and Morrow, N.R., 1999, Effect of Brine Composition on

Wettability and Oil Recovery of a Prudhoe Bay Crude Oil. J. Canadian Petrol. Technol.

38 (1).

Yousef, A.A., Al-Saleh, S., Al-Kaabi, A. and Al-Jawfi, M., 2010, Laboratory

investigation of novel oil recovery method for carbonate reservoirs. SPE 137634.

Zhou, X. Morrow, N.R. and Ma, S., 1995a, Interrelationship of Wettability, Initial Water

Saturation, Aging Time and Oil Recovery by Spontaneous Imbibition and Waterflooding.

SPE/DOE 35436.

Zhou, X., Morrow, N.R. and Ma, S., 1995b, Interrelationship of Wettability, Initial Water

Saturation, Aging Time, and Oil Recovery by Spontaneous Imbibition and

Waterflooding. SPE 35436.

Zhou, X., Torsæter, O., Xie, X. and Morrow, N.R., 1995, The Effect of Crude-Oil Aging

Time and Temperature on the Rate of Water Imbibition and Long-Term Recovery by

Imbibition. SPE Formation Eval., December 1995.

Evaluation of Low Salinity Injection for the Minnelusa Formation, Powder River Basin, Wyoming

Documents

Transcript of Evaluation of Low Salinity Injection for the Minnelusa Formation, Powder River Basin, Wyoming