Data Analysis and Updated Screening

Criteria for Polymer Flooding Based on

Oilfield Data
Laila Dao Saleh, SPE, Mingzhen Wei, SPE, and Baojun Bai, SPE, Missouri University of Science and Technology

Summary porosity, depth, and temperature (Taber 1983; Taber et al. 1997a).
Enhanced-oil-recovery (EOR) screening is considered the first Screening criteria serve as the first step for evaluating potential
step in evaluating potential EOR techniques for candidate reser- EOR techniques for candidate reservoirs. Screening criteria are
voirs. Therefore, as new technologies are developed, it is impor- critical at the start of an EOR project because most EOR projects
tant to update the screening criteria. Many of the screening require significant financial capital investments and can yield sig-
criteria regarding polymer flooding that have been described in nificant undesirable consequences if they fail (Hite 2004).
the literature were based on data collected from EOR surveys of During the past 20 years, many researchers have developed and
the Oil and Gas Journal. However, the data quality has not been published technical screening criteria for different EOR techni-
addressed in previous research. The data set originally contained ques. Table 1 shows the screening criteria for polymer flooding
481 polymer-flooding projects from around the world, and it con- published by different investigators. The first EOR screening stud-
tained some problems, including outliers and duplicate, inconsis- ies were presented by Brashear and Kuuskraa (1978) by use of data
tent, and missing data. To ensure the quality of the data set before collected from 200 EOR pilot projects in the United States. They
running analyses, boxplots and crossplots were used to detect and analyzed the data from both a technical and an economical perspec-
identify data problems. After detecting outliers and deleting dupli- tive. Carcoana (1982) presented screening criteria for some EOR
cate and severely inconsistent data records, only 250 projects techniques; these criteria were based on the knowledge of reservoir
remained. Both graphical and statistical methods were used to an- properties and the results obtained from commercial applications
alyze and describe the results of the data set. Two major sets of in- of EOR techniques in Romanian oil fields. In this screening, the
formation were given after data cleaning: The first was that the temperature and depth were less than 180 F and 6,562 ft, respec-
majority distribution of each parameter was shown by use of a tively. Taber (1983; Taber et al. 1997b) proposed screening criteria
histogram distribution, and the second was that the range of each on the basis of field data and oil-recovery mechanisms for common
parameter and some of its statistical values were presented by use EOR techniques. This study considered the 1996 Worldwide EOR
of a boxplot. Finally, the screening criteria are presented on the Survey to summarize the criteria. Taber’s criteria that are relevant
basis of these statistics and the defined data parameters. The to polymer flooding include that the maximal oil viscosity should
developed criteria were compared with previously published crite- be less than 150 cp and reservoir permeability should be greater
ria, and their differences were explained. The developed criteria than 10 md. He presented these screening criteria both graphically
show that a polymer-flooding project can be successfully applied and in tables. Goodlett et al. (1986) presented screening criteria on
in a reservoir with a temperature of less than 210 F, an oil viscos- the basis of a summary of previously published screening criteria
ity up to 5,000 cp, and gravity lower to 12 API. for chemical, gas-injection, thermal, and microbial EOR techni-
ques. Al Bahar et al. (2004) illustrated criteria for each EOR tech-
nique on the basis of the literature and his own experience. He used
Introduction software to evaluate the suitability of these criteria for EOR proc-
esses at 81 reservoirs in Kuwait. Dickson et al. (2010) presented
Polymer flooding is one of the most important EOR techniques
screening criteria for various gas-injection, chemical, and thermal
used to improve the mobility ratio and, therefore, sweep efficiency
techniques on the basis of a combination of experience and values
(Alvarado and Manrique 2010). Polymer flooding improves the
published in the literature. Aladasani and Bai (2010) updated EOR
mobility ratio by increasing the viscosity of the displacing fluid
screening criteria on the basis of the raw data of EOR projects
(water) and reducing water permeability in porous media, which
reported in the EOR survey of Oil and Gas Journal from 1998
allows the displaced fluid (oil) to move more freely through the
through 2008. However, a mistake was made when they converted
pores (Sandiford 1964; Pye 1964; Gogarty 1967; Jennings et al.
temperature from  C to  F, which made the temperature criteria up
1971). Polymer flooding has been applied successfully and has
to 237.2 F. In addition, a novel improved hydrocarbon-recovery
improved oil recovery by 5 to15% of the original oil in place (Zaitoun
screening methodology has been developed to identify the appro-
et al. 1998; Wang et al. 2002).
priate process for any number of reservoirs (Note: Table 1 shows
The Worldwide EOR Survey gathers EOR-project information
only the criteria for polymer flooding).
from various operators in different countries, including the United
The objectives of this paper are to show the project distribu-
States, China, Germany, Canada, France, Russia, Argentina,
tion of each parameter that affects EOR selection and to update
India, and Indonesia (Fig. 1). Operators in China and most of the
screening criteria for polymer flooding by use of the data from
Soviet Union did not respond to the EOR survey before 1994
EOR surveys made between 1978 and 2012. To accomplish these
(Moritis 1992, 1994). Currently, China is the largest producer of
objectives, it is critical to ensure that the data are of sufficiently
oil by means of chemical EOR projects (Alvarado and Manrique
high quality to produce reliable results. Therefore, data-cleaning
2010). The largest polymer-flooding project in the world was
methods were applied to identify and remove duplicate, inconsis-
implemented at Daqing field in China in 1996 (Wang et al. 2009).
tent, and missing data. After cleaning the data set, both graphical
The success of a polymer-flooding project for a given field
and statistical methods were used to display and summarize the
depends on the field’s oil properties and reservoir characteristics,
data set to produce the screening criteria.
including oil viscosity, oil gravity, oil saturation, permeability,

Data Collection
Copyright V
C 2014 Society of Petroleum Engineers

A data set was created by collecting information from polymer-

Original SPE manuscript received for review 26 November 2012. Revised manuscript
received for review 11 September 2013. Paper (SPE 168220) peer approved 26 September
flooding projects. The Worldwide EOR Survey published by the
2013. Oil and Gas Journal supplied the data for this data set, which

February 2014 SPE Reservoir Evaluation & Engineering 15

 1% (1)
Argentina
4% (7)
21% (39)
Canada

4% (7) China
3% (6)
3% (6) France
64% (119)
Germany

India

USA

Fig. 1—World polymer projects.

TABLE 1—GENERAL SCREENING GUIDE FOR POLYMER FLOODING

Published Gravity, Oil Viscosity, Oil Saturation, Average Temperature, Depth, SPE
 
Author Year API cp Start, % Permeability, md F ft Paper No.

Brashear and 1978 15 <200 50 >20 <200 NC 6350

Kuuskraa
Carcoana 1982 — 50–80 >50 >50 <180 <6,562 10699
Goodlett et al. 1986 >25 <100 >10 >20 <200 <9,000 15172
Taber et al. 1997b >15 10–100 >50 >10 <200 <9,000 35385
Al-Bahar et al. 2004 — <150 >60 >50 <158 — 88716
Dickson et al. 2010 >15 10–1000 >30 >100 if (10<m <100 ) <170 800–9,000 129768
>1,000 if (100<m <1,000)
Aladasani and Bai 2010 13–42.5 0.4–4000 34–82 1.8–5500 74–237.2 700–9,460 130726

included 481 polymer-flooding projects. Fig. 2 displays the num- approximately 20 projects in China, one field project in India, one
ber of polymer-flooding projects for each year on the basis of our project in Argentina, and two in the US. Because operators in
data set. In 1986, the number of polymer-flooding projects China have not replied with updated EOR information for several
increased by 67.9% over the projects reported in the 1984 survey years, the 2010 survey deleted the projects in China (Koottungal
(Leonard 1986). On the contrary, the number of polymer projects 2010). Therefore, only 2 polymer-flooding projects were included
decreased dramatically between 1986 and 1988 because of the in that survey (Koottungal 2010), and both of them are still pro-
1986 shutdown of 40 polymer projects that had not reported any ducing in 2012. The data set includes successful, unsuccessful,
enhanced production (Aalund 1988). From 1988 to 1996, the and unevaluated projects (Fig. 3).
number of projects continued to decline because of decreased The failure of a polymer-flooding project can be attributed to
crude prices. During that time, polymer-flooding processes were technical or economic problems or to political issues. The techni-
too expensive to implement. In the 2000s, the crude price began cal reasons include, but are not limited to, polymer availability,
to rise, which encouraged companies to pursue additional oil re- polymer being resistant to formation temperature and formation-
covery (Roger et al. 2012). According to the 2008 EOR survey water salinity and hardness, insufficient polymer-slug size, reser-
(Moritis 2008), some polymer-flooding projects are ongoing— voir heterogeneity (unexpected channeling), injectivity problems,
and environmental control. The economics of a polymer-flooding
project depends strongly on investment, rate of return, operating
220 and chemical costs, and crude-oil prices; 48 of the 65 discourag-
200 ing projects were from 1986 and 1988. We believe that the unsuc-
180 Original data set cessful projects were caused mainly by a sudden drop of oil prices
160 at that time, which made the chemical-flooding project unecono-
No. of Projects

140 mical.
120
100
80 Successful Projects
60 400
40 Discouraging Projects
300
20 Unevaluated Projects
0
200
78
80
82
84
86
88
90
92
94
96
98
00
02
04
06
08
10
12
19
19
19
19
19
19
19
19
19
19
19
20
20
20
20
20
20
20

100
Year
0
Fig. 2—Number of polymer-flooding projects, as reported by EOR Successful Projects Discouraging Unevaluated Projects
surveys from 1978 to 2012. Sources—Noran 1978; Shannon 1980; Projects
Alan 1982; Leonard 1984; Leonard 1986; Aalund 1988; Hochanadel
et al. 1990; Moritis 1992, 1994, 1996, 1998, 2000, 2002, 2004, 2006, Fig. 3—Successful evaluation results of polymer-flooding
2008; Koottungal 2010; Anonymous 2012. projects.

16 February 2014 SPE Reservoir Evaluation & Engineering

 314
350 Upper limit Outlier
300

Frequency
250
200
150 82
10 Maximum
100
observation
50
0
Sandstone Carbonate Others 3rd Quartile

Fig. 4—Lithology of polymer-flooding projects.

Mean
Polymer-flooding projects were applied to different formation Median =
types (sandstone, limestone, dolomite, and others). The frequency 2nd Quartile
of application on the basis of the lithology data set was more than
77% in sandstone reservoirs and 20% in carbonate reservoirs 1st
Quartile
(Fig. 4), and less than 3% in other formations (Note: Some proj-
ects did not mention the type of formation).

Data Cleaning
Data quality is critical in ensuring the quality of the results. EOR- Minimum
survey data contain many types of problems that can affect the observation
quality of the data set. In this work, only data from successful
projects were used for statistical analysis. Fig. 5—Schematic of boxplot and outlier.

Data Problems
The most common problems within the data set were duplicate, Data-Problem Detection
missing, and inconsistent data and outliers. A boxplot helps both to summarize a data set and to detect outliers
within the data, as shown in Fig. 5. A summary of the data charac-
Duplicate Data. The duplicate-data problem was observed while teristics is provided by describing the following five numbers: the
collecting data from the worldwide EOR surveys. Many fields lowest value (minimum), the highest value (maximum), the first
were listed numerous times with the same values and in the same quartile (the 25th percentile), the second quartile (the 50th percen-
year or during different years of publications. One possible reason tile), and the third quartile (the 75th percentile). The outlier values
for this duplication is that some countries did not update their are larger than the upper limit of the data and smaller than the
EOR information for several years (Moritis 2002, 2004, 2008), lower limit of the data. The upper limit is calculated as 1.5 times
and the EOR survey did not change their records. However, the the interquartile range plus the third quartile, and the lower limit of
oilfield projects for countries that did not update their information the data as 1.5 times the interquartile range minus the first quartile.
were deleted from the 2010 and 2012 EOR surveys. To solve the Fig. 5 shows two simple rectangles with different colors, repre-
problem with duplicate data from the earlier surveys, duplicate senting the first to the third quartile. The median of the data set
fields were removed from the data set. appears in the center (horizontal line), and the mean of the data set
is indicated by the orange circle. The end of both whiskers repre-
sents the minimal and maximal data-set observations.
Missing Data. Several fields within the data set were missing A crossplot was used to plot a pair of variables from the data set.
one or more pieces of information, including oil saturation (start The plot helps to uncover the relationships between these variables
and end), permeability, depth, and temperature. The missing val- and to detect outliers. The box- and crossplots were combined to
ues as a percentage (the proportion of missing variable data to all yield additional clarity. Fig. 6a shows the crossplot between temper-
variable data) were as follows: ature and depth, whereas Fig. 6b shows the boxplot for temperature.
 Oil saturation (start) data missing for more than 10% of The boxplot shows that data from only one field exceeded the upper
fields data. limit of the data set (orange line). However, this data cannot be con-
 Oil saturation (end) data missing for more than 30% of fields sidered an outlier because the field-temperature value is consistent
data. with other field characteristics, such as depth and porosity, and the
 Permeability missing for more than 10% of permeability polymer-flooding project was successful at this temperature. On the
data. basis of past EOR screenings, polymer flooding is not recom-
 Depth and temperature also missing for 5% of the data. mended for reservoir temperatures more than 200 F. However, this
These missing values were ignored during the analysis. result indicates that successful polymer flooding was achieved
above this temperature. The crossplot also shows that some fields
Inconsistent Data. Inconsistent data contained both discrepan- had temperatures more than 200 F but were not considered outliers
cies and impossible values. Several pieces of information in the for the reasons mentioned previously (Fig. 6a).
data set were inconsistent, such as The porosity of several fields exceeded the upper limit of the
 Oil saturation (end) >oil saturation (start). data set in the boxplot depicted in Fig. 7b. The upper limit is repre-
 Oil saturation (start) <20%. sented by an orange line and is approximately equal to 30%. These
 Average reservoir permeability <1 md. fields also appear in the crossplot in Fig. 7a. Among all the fields,
 Porosity >40%, and those fields are inconsistent with other only two had porosities exceeding 40%; they have been circled in
field characteristics, such as depth. Fig. 7 to indicate their status as outliers. In addition, Fig. 7 indicates
A few fields had light oil (gravity >20 API) with higher oil that one deeper field that is approximately 12,000 ft is considered an
viscosity; for example, one field had oil gravity ¼ 35 API and oil outlier because this value lies far from the majority of the data set.
viscosity ¼ 175 cp. The crossplot in Fig. 8a indicates the relationship between oil
Most of the inconsistent data and outliers were detected by gravity and viscosity. Only one field that lies far from the majority
box- and crossplots for different parameters. of the data set and shows behavior different from that of the trend

February 2014 SPE Reservoir Evaluation & Engineering 17

 (a) (b)
260

210

Temperature, °f
160

110

60

10
100 2,100 4,100 6,100 8,100 10,100 12,100 14,100 Temperature, °f
Depth, ft

Fig. 6—Temperature vs. depth crossplot (a) and boxplot of temperature (b).

(a) (b)
50
45
40
35
Porosity, %

30
25
20
15
10
5
0 Porosity, %
100 2,100 4,100 6,100 8,100 10,100 12,100 14,100

Depth, ft

Fig. 7—Porosity vs. depth crossplot (a) and boxplot of porosity (b).

is circled in the crossplot and marked as an outlier. The oil-gravity Data-Set Analysis
boxplot shows no outliers for this value (Fig. 8b). After the duplicating data sets were removed, the number of poly-
The boxplot of permeability (Fig. 9b) shows that several fields mer-flooding projects decreased from 481 to 250. The detailed
have values exceeding the upper limit (orange line). The crossplot data information can be found on the website www.mst.edu/
between permeability and porosity Fig. 9a shows a few fields lying far baib. Fig. 11 illustrates the number of studied projects by year
from the majority of the data set. These fields have been circled and before and after cleaning the data set.
marked as outliers. The average permeability of the field circled indi- In this paper, two types of methods were used to display data.
vidually (Fig. 9a) was 0.25 md. This field is considered an outlier The first one is the graphic method that includes histogram, box-
because the permeability is very low and requires a polymer of a lower plot, and crossplot. The second type is statistical methods.
molecular weight. A lower molecular weight can be manufactured,
but the amount of polymer required to achieve sufficient viscosity for
a polymer drive makes such projects economically unattractive. Methods for Displaying the Data Set
The oil-saturation boxplot does not indicate any outliers because Histograms. Histograms are used to display the data set graphi-
the upper limit (orange line) of the data set was 100% (Fig. 10b). cally and to depict the sampling distribution of the data set. The
However, the crossplot shows several values situated far from the histogram shows the frequency of the data set on the y-axis and
majority of the data set, circled in Fig. 10a. The values falling the variables being measured on the x-axis. Fig. 12a illustrates the
under the trend line are considered outliers because the oil satura- data set for the distribution of oil gravity. The histogram shows
tion (end) of these values is greater than the oil saturation (start). two peaks within the data set, indicating a bimodal shape (two

(a) (b)
60

50
Oil Gravity, API

40

30

20

10

0
0.1 1 10 100 1,000 Oil Gravity, API
Oil Viscosity, cp

Fig. 8—Oil gravity vs. oil-viscosity crossplot (a) and boxplot of oil gravity (b).

18 February 2014 SPE Reservoir Evaluation & Engineering

 (a) (b)
10,000

Average Permeability, md
1,000

100

10

0.1
0 10 20 30 40 50 Average Permeability, md
Porosity, %

Fig. 9—(a) Crossplot between permeability and porosity and (b) boxplot of average permeability.

(a) (b)
100
90
80
Oil Saturation (start), %

70
60
50
40
30
20
10
0
0 10 20 30 40 50 60 70 80 90 100 Oil Saturation (start), %

Oil Saturation (end), %

Fig. 10—Oil saturation (start) vs. oil saturation (end) crossplot (a) and oil-saturation boxplot (b).

recurring groups of numbers). The first peak includes oil-gravity low). Fig. 12d depicts oil saturation (start), showing that the
values between 20 and 30 API, and the second peak contains val- distribution of the oil-saturation (start) data set is skewed to the
ues between 30 and 45 API (highlighted in yellow) that represent right and contains little variance. Most of the frequency values are
frequency-occurring oil-gravity values. On the basis of this result, between 40 and 70%. Polymer flooding typically produces better
polymer flooding was used mostly when the oil gravity was results if it is begun before the water/oil ratio becomes high
greater than 20 API and less than 40 API (light and medium- (Donaldson et al. 1989; Du and Guan 2004). Polymer flooding is
range oil). Fig. 12b illustrates the multimodal distribution for oil typically intended to improve sweep efficiency rather than to
viscosity. The majority of the data points fall between 1 and 4 cp reduce residual oil saturation to a level below waterflooding, even
and appear in yellow in the figure. The distribution of the oil-vis- though recent research showed that the elasticity of polymer
cosity data set shows that polymer flooding should not be applied might help reduce oil saturation (Wenxiang et al. 2007; Huh and
when the oil viscosity is higher than 130 cp; however, the distri- Pope 2008; Kamaraj et al. 2011).
bution does not include the two polymer-flooding projects in Fig. 13a shows a different distribution of the oil saturation
heavy-oil reservoirs (more than 1,000 cp), which were imple- (end) data set, depicting four peaks, which indicates a multimodal
mented in 2010. We will discuss them as special cases later in this distribution. Each peak has a different range of values, but the
paper. Also, the porosity data set in Fig. 12c shows a multimodal range of the highest peak is between 40 and 55% (highlighted in
distribution. The highest peak is between 14 and 21%, which rep- yellow). The multimodal distribution of average permeability is
resents the majority of the porosity data set (highlighted in yel- shown in Fig. 13b. The data set values most frequently fell
between 90 and 450 md (highlighted in yellow). Also, Fig. 13c
220 shows a multimodal distribution of the depth data set, with the
200 Original data set majority of data points lying between 2,500 and 5,500 ft (high-
180 The data set after cleaning lighted in yellow). Fig. 13d represents the temperature-distribution
No. of Projects

160 histogram, which is characterized by a multimodal distribution. The

140
120 majority of the data points fell between 80 and 130 F. On the basis
100 of that distribution, the temperature should be less than 210 F
80 because, at higher temperatures, the polymer will lose its viscosity,
60 and the molecular weight will decrease because of degradation.
40
20
Also, if depths are less than 10,000 ft, it is better to maintain a con-
0 sistent temperature.
78
80
82
84
86
88
90
92
94
96
98
00
02
04
06
08
10
12
19
19
19
19
19
19
19
19
19
19
19
20
20
20
20
20
20
20

Year Boxplot
The boxplot is used not only to detect outliers, as explained previ-
Fig. 11—Data set of polymer projects before and after pre- ously, but also to display and summarize the data set for each
processing. variable, as shown in Fig. 14. Data-value ranges were provided

February 2014 SPE Reservoir Evaluation & Engineering 19

 (a) (b)
70 40
60 35

50 30

Frequency
Frequency 25
40
20
30
15
20 10
10 5
0 0
15 20 25 30 35 40 45 48 0.3 2 5 8 11 14 17 26 34 46 68 130
Oil Gravity, API Oil Viscosity, cp

(c) (d)
30 50
45
25 40
35

Frequency
20 30
Frequency

25
15
20
10 15
10
5 5
0
0

e
4 9 11 13 15 17 19 21 23 25 27 29 31 33 35

–3

–4

–5

–6

–7

–8

–9

or
–M
20

30

40

50

60

70

80

90
Porosity, % Oil Saturation (start), %

Fig. 12—Histogram representing distribution of (a) oil gravity, (b) oil viscosity, (c) porosity, and (d) oil saturation (start).

for each parameter (minimal and maximal values) after removing Table 2 shows the names of field projects for each low and
any outliers. These ranges are illustrated by the distance between high value of individual screening parameters.
the opposite ends of the whiskers. Boxplots have different scales
because they display different ranges of various data sets. The log
scale is the best choice for displaying a large range of values. Summarizing Screening Criteria
Also, the figure displays additional information, such as the mean Table 3 provides a summary of polymer-flooding criteria on the
and median of the data set. The mean values are indicated in Fig. basis of the preceding statistical analysis of the cleaned data set.
14 by orange circles, and the median values by the horizontal line This summary includes the parameters that contribute to the suc-
between each set of boxes. cess or failure of a polymer-flooding project, including oil gravity,

(a) (b)
30 35
25 30
Frequency

20 25
Frequency

20
15
15
10
10
5
5
0
0
20 30 35 40 45 50 55 60 65 70 75 80
0.6 30 60 90 300 600 1000 4000
Oil Saturation (end), % Average Permeability, md

(c) (d)
35 45
30 40
35
25 30
Frequency
Frequency

20 25
15 20
15
10 10
5 5
0 0
70 90 110 130 150 170 190 210
1000 3000 5000 7000 9000

Depth, ft Temperature, °f

Fig. 13—Histogram represents distribution of (a) oil saturation (end), (b) average permeability, (c) depth, and (d) temperature.

20 February 2014 SPE Reservoir Evaluation & Engineering

 9400
10,000 5500

1,000

210
550
130
100 94 81
48 36.1
65

10 21 20
12

4.1
1

0.6
0.3
0.1
Gravity, API Oil Viscosity, cp Oil Saturation Oil Saturation Average Temperature, °f Porosity, % Depth, ft
(start), % (end), % Permeability, md

Fig. 14—Boxplot of oil gravity, oil viscosity, oil saturation (start), oil saturation (end), average permeability, temperature, porosity,
and depth.

TABLE 2—NAMES OF FIELD PROJECTS

Names of Fields

Parameters Min. Value Max. Value

Gravity,  API Chico Martinez Black Diamond

Oil Viscosity Atlantic Shengli
Oil Saturation (Start) Loudon Stewart Ranch
Oil Saturation (End) Sage Spring Creek—Unit A Vacuum
Average Permeability Levelland Ploen OST
Temperature NPR-3, Teapot Dome Hitts Lake
Porosity Headlee North Cement
Depth NPR-3, Teapot Dome Ploen OST

oil viscosity, porosity, oil saturation (start and end), permeability, temperature than other polymers. This effect occurs because
depth, and temperature. The standard statistics used to describe hydrophobic association, as an entropy-driving endothermal pro-
the criteria are the mean, median, standard deviation, and minimal cess, is favored by high temperatures (Yabin et al. 2001). One lab-
and maximal data-set values. oratory study showed that a polymer can remain stable up to
Fig. 15 compares the updated, improved screening criteria 250 F by modifying the polyacrylamide with various monomers
with previously published criteria. The differences between these (Vermolen et al. 2011). In conclusion, the development of new
criteria can be explained as follows: polymers has made it feasible to use polymers in reservoirs with
 The maximal temperature in our data set was 210 F, whereas higher temperatures than ever before. In addition, at a low brine
other criteria used temperatures less than 200 F. The most fre- salinity and a low concentration of divalent-cation content (hard-
quently used polymer for polymer flooding is hydrolyzed poly- ness), the viscosity of polymer was more stable at a higher
acrylamide (HPAM), the viscosity of which decreases steadily temperature.
because of thermal degradation as the temperature increases.  The polymer flooding can be used in reservoirs with oil vis-
However, this relationship changes when the polymer is hydro- cosity up to 5,000 cp. The viscosity data from heavy-oil fields
phobically associated, which gives the polymer more resistance to were excluded from Fig. 12b for two reasons—there were only

TABLE 3—CRITERIA GUIDE FOR POLYMER FLOODING IN THE DATA SET

Oil Gravity, Oil Porosity, Oil Saturation, Oil Saturation, Average Depth, Temperature,
 
Statistic API Viscosity, cp % Start End Permeability, md ft F

Mean 31.2 12.21 18.15 55.85 46.57 384.88 4,004.21 118.1

Median 32.00 4.00 17.40 53.00 47.00 100.00 3,650 110
Standard Deviation 8.26 19.74 5.4 15.5 13.37 874.55 1,925.8 30.06
Minimum 12 0.3 4.1 21 20 0.6 550 65
Maximum 48 130 (special case, 36.1 94 80.9 5,500 9,400 210
1,000–5,000)

February 2014 SPE Reservoir Evaluation & Engineering 21

 550–9400
800–9000

Depth, ft < 9000

< 9000
< 6562

< 210
< 170
< 158
Temperature, °ft < 200
< 200
< 180
< 200 Our work
0.6–5500 Dickson
> 100
> 50
Permeablity, md > 10 Al-Bahar
> 20
> 50
> 20 Taber
21–85 Goodlett
> 30
> 60
Oil Saturation (Start) > 50 Carcoann
> 10
> 50
50 Brashear
special case

0.3–130 1000–5000
10–1000
< 150
Oil Viscosity, cp < 100
10–100
50–80
< 200

12–48
> 15

Oil Gravity, API > 15

> 25

15

1 10 100 1,000 10,000

Fig. 15—Comparison between different screening criteria.

two heavy-oil fields, and the oil-viscosity values of those fields polymer flooding has higher increased oil recovery in reservoirs
varied significantly from the values of other fields. However, the with a higher permeability (Asghari and Nakutnyy 2008). The
heavy oil fields will be considered as a special case, as shown in lowest permeability limit is 0.6 md in the new criteria. Techni-
Table 3. The two heavy-oil fields are in Canada and have oil-vis- cally, flooding polymers, especially those with high molecular
cosity ranges from 1,000 to 5,000 cp with good permeability (500 weights, cannot freely flow in formations with a permeability of
to 5,000 md). Successfully applying polymer flooding in heavy- less than 10 md because of larger molecular size compared with
oil fields would significantly broaden the possibilities for the pore throat (Chang 1978). However, the permeability of 13 poly-
application of polymer flooding. However, higher oil viscosities mer-flooding reservoirs is less than 10 md. Nine of them are car-
require higher polymer viscosities to improve the mobility ratio bonate reservoirs, including Leveland (0.6 md), Mabee (1.5 md),
and displace more oil. The higher polymer viscosity can be Robertson (1.54 md), Harris (3 md), McElroy (5 md), C-Bar (6
achieved by three conditions—a higher-molecular-weight poly- md), Slaughter (6.02 md), Stephens County Regular field (9 md),
mer to keep concentration low (Zaitoun et al. 1998; Wang et al. and Fitts (6.6 md). The other four reservoirs are sandstone reser-
2008); a higher polymer concentration to keep the same molecular voirs: So. Eunice/Langlie Matrix (1.3 md), Spraberry Deep (5.7
weight (Asghari and Nakutnyy 2008; Hincapie et al. 2011); and md), Elaine (6 md), and Langile mattix (7.2 md). Carbonate reser-
the use of low-salinity makeup water to prepare polymers (Strauss voirs usually have natural fractures and low-permeability matrix.
et al. 2010). Increasing polymer viscosity reduces the injectivity; For example, Wilkes (1981) reported that the Leveland field was
therefore, horizontal wells and hydraulically fracturing the forma- dominated by natural fractures. Qiu et al. (2001) mentioned that
tion near the wellbore if a reservoir has low mobility may be Mabee field was highly heterogeneous (depositional facies, dia-
required to meet the injectivity requirement (Wassmuth et al. genesis, and fractures). Polymer-flooding projects in carbonate
2007, 2009; Seright 2010; Strauss et al. 2010; Chang 2011; Moe reservoirs mainly target the polymer entry and distribution in frac-
Soe Let et al. 2012). Fig. 16 shows the relationship of oil viscosity tures and microfractures; therefore, it is understandable that poly-
and permeability. It can be seen that the two parameters have mer flooding can be successful in these reservoirs even though the
weak correlation (R2 ¼ 0.0168), which means that a higher mobil- reported permeability is low. However, it is important to be aware
ity (k/lo) is not directly related to a higher oil viscosity; but the that detailed information on polymer-flood design for these proj-
figure does show that most projects with permeability higher than ects in low-permeability reservoirs was not found in the literature;
1,000 md have high oil viscosity. therefore, the improved oil-recovery mechanisms from these proj-
 Polymer flooding has been successfully applied in fields with ects lack a full explanation.
permeability up to 5,500 md. Recent laboratory results show that  Polymer flooding has been used in reservoirs with a depth of
9,400 ft. Other screening-criteria data sets showed a limited suc-
cess of polymer flooding at depths of more than 9,000 ft. In fact,
1,000 it is not necessary to consider the depth of a reservoir as a screen-
2 ing criterion of polymer flooding because polymer is sensitive to
R = 0.0168
the formation depth only if temperature is considered as a reflec-
Oil Viscosity, cp

100
tion of reservoir depth, as shown in Fig. 6a.
10  The preceding criteria are the general requirements for a
polymer-flooding project. A recent interest is the use of polymer
1 flooding for offshore oil fields (Wei et al. 2007; Zhou et al. 2008;
Mogbo 2011; Raney et al. 2011). However, several challenges are
0.1 associated with polymer flooding in offshore reservoirs, including
0.1 1 10 100 1,000 10,000 the unavailability of fresh water to makeup polymer solution,
Permeability, md large well spacing, limited platform space, and surface facilities
and environmental requirements. These challenges require the
Fig. 16—Relationship between oil viscosity and permeability. polymers to have high dissolution rates in water, have high

22 February 2014 SPE Reservoir Evaluation & Engineering

resistance to salinity, and be environmentally friendly. For exam- Brashear, J.P. and Kuuskraa, V.A. 1978. The Potential and Economics of
ple, the China Offshore Company requires the polymer used for Enhanced Oil Recovery. J. Pet Tech 30 (9): 1231–1239. http://
flooding to be dissolved in 30 minutes and to be thermostable for dx.doi.org/10.2118/06350-PA.
more than 1 year in seawater (Tang et al. 2005). Carcoana, A.N. 1982. Enhanced Oil Recovery in Rumania. Paper SPE
 It should be mentioned that water salinity and reservoir 10699 presented at the SPE Enhanced Oil Recovery Symposium,
heterogeneity are two important parameters to the success of a Tulsa, Oklahoma, 4–7 April. http://dx.doi.org/10.2118/10699-MS.
polymer-flooding project. Water salinity and divalent-ion concen- Chang, H. 1978. Polymer Flooding Technology Yesterday, Today, and
tration significantly reduce the viscosity and thermostability of Tomorrow. J. Pet Tech 30 (8): 1113–1128. http://dx.doi.org/10.2118/
HPAM polymers (Chang 1978). Moreover, the shear degradation 7043-PA.
of HPAM-polymer solution is more severe in high-salinity forma- Chang, H. 2011. Advances of Polymer Flood in Heavy Oil Recovery. Pa-
tion water (Maerker 1975). Therefore, most polymer solutions per SPE 150384 presented at the SPE Heavy Oil Conference and Exhi-
were prepared with fresh or low-salinity water when they were bition, Kuwait City, Kuwait, 12–14 December. http://dx.doi.org/
injected into fields. The major EOR mechanism of polymer flood- 10.2118/150384-MS.
ing is to reduce mobility ratio and thus improve sweep efficiency. Dickson, J.L., Dios, A.L., and Wylie, P.L. 2010. Development of
However, if a reservoir is very heterogeneous, polymer will have Improved Hydrocarbon Recovery Screening Methodologies. Paper
an early-breakthrough problem. Tyler and Finley (1991) men- SPE 129768 presented at the SPE Improved Oil Recovery Symposium,
tioned that polymer flooding is more successful at low to moder- Tulsa, Oklahoma, 24–28 April. http://dx.doi.org/10.2118/129768-MS.
ate degrees of vertical and horizontal heterogeneity. Du and Guan Donaldson, E.C., Chilingarian, G., and Yen, T. 1989. Enhanced Oil Re-
(2004) mentioned that the Dykstra-Parson variation coefficient of covery, II Processes and Operations. New York City: Elsevier Science
a reservoir should be in the range of 0.28 to 0.80 for a successful Publishing B.V.
polymer-flooding project.
Du, Y. and Guan, L. 2004. Field-Scale Polymer Flooding—Lessons Learnt
 The combination of the following parameters will widen the
and Experiences Gained During Past 40 Years. Paper SPE 91787 pre-
polymer-application envelope.
sented at the SPE International Petroleum Conference in Mexico, Pue-
* Reservoirs with high permeability and high oil viscosity
bla Pue., Mexico, 7–9 November. http://dx.doi.org/10.2118/91787-
can be flooded by use of more-viscous polymers, which increase
MS.
the applicability of polymer flooding to heavy oils.
* Low-salinity and low-hardness formation water allow
Gogarty, W.B. 1967. Mobility Control With Polymer Solutions. SPE J. 7
polymers to be used in higher-temperature reservoirs. (2): 161–173. http://dx.doi.org/10.2118/1566-PA.
Goodlett, G.O., Honarpour, M.M., Chung, F.T. et al. 1986. The Role of
Screening and Laboratory Flow Studies in EOR Process Evaluation.
Conclusions
Paper SPE 15172 presented at the SPE Rocky Mountain Regional
• This paper provided comprehensive methodologies and detailed Meeting, Billings, Montana, 19–21 May. http://dx.doi.org/10.2118/
procedures to clean EOR-project data sets as the first step in 15172-MS.
analyzing EOR data and establishing screening criteria. Hincapie, R., Alvarez, C., and Vargas, A. 2011. Technical Feasibility
• After data cleaning, the histogram plot is preferable to represent of Polymer Injection in Heavy Oil Reservoir BAINF60 and
the distribution of each parameter of successful polymer-flood- BAMED78—Intercampo Norte—Through Predictive Models. Paper
ing projects. SPE 149621presented at the SPE Heavy Oil Conference and Exhibi-
• A boxplot served as a very helpful tool for exploring and dis- tion, Kuwait City, Kuwait, 12–14 December. http://dx.doi.org/10.2118/
playing the screening-criteria data set. 149621-MS.
• New polymer-flooding criteria were presented and compared
Hite, J.R. 2004. Planning EOR Projects. Paper SPE 92006 presented at the
with previous publications, and their differences were ex-
SPE International Petroleum Conference in Mexico, Puebla Pue., Mex-
plained.
ico, 7–9 November. http://dx.doi.org/10.2118/92006-MS.
• The polymer-flooding screening criteria can be summarized as
follows: oil gravity >12 API; oil viscosity <5,000 cp; tempera- Hochanadel, S., Lunceford, M., and Farmer, C. 1990. A Comparison of 31
ture <210 F; porosity >4.1%; start oil saturation >21%; per- Minnelusa Polymer Floods With 24 Minnelusa Waterfloods. Paper
meability >0.6 md (the permeability of target zones should be SPE 20234 presented at the SPE/DOE Enhanced Oil Recovery Sympo-
higher than 10 md, otherwise a high-molecular-weight polymer sium, Tulsa, Oklahoma, 22–25 April. http://dx.doi.org/10.2118/20234-
cannot propagate through the reservoir); formation-type, sand- MS.
stone/carbonate; oil composition not critical; and pay-zone Huh, C. and Pope, G.A. 2008. Residual Oil Saturation From Polymer
thickness not critical; Floods—Laboratory Measurements and Theoretical Interpretation. Pa-
• Specific considerations of polymer flooding were presented. per SPE 113417 presented at the SPE/DOE Symposium on Improved
Oil Recovery, Tulsa, Oklahoma, 20–23 April. http://dx.doi.org/
10.2118/113417-MS.
References
Jennings, R.R., Rogers, J.H., and West, T.J. 1971. Factors Influencing Mo-
Aalund, L.R. 1988. EOR Projects Decline, But CO2 Pushes Up Produc- bility Control by Polymer Solutions. J. Pet Tech 23 (3): 391–401.
tion. Oil & Gas J.: 30–43. http://dx.doi.org/10.2118/2867-PA.
Aladasani, A. and Bai. B. 2010. Recent Developments and Updated
Kamaraj, K., Zhang, G., Liu, Y. et al. 2011. Effect of Residual Oil Satura-
Screening Criteria of Enhanced Oil Recovery Techniques. Interna-
tion on Recovery Efficiency During Polymer Flooding of Viscous
tional Oil and Gas Conference and Exhibition in China, Beijing,
Oils. Paper SPE 22040 presented at the OTC Arctic Technology Con-
China, 8–10 June. http://dx.doi.org/ 10.2118/130726-MS.
ference, Houston, Texas, 7–9 February. http://dx.doi.org/10.2118/
Alan, P.G. 1982. Independents Plunge Into Enhanced Oil Recovery. Oil &
22040-MS.
Gas J.: 1–15.
Koottungal, L. 2010. 2010 Worldwide EOR Survey. Oil & Gas J. 108
Al-Bahar, M.A., Merrill, R., Peake, W. et al. 2004. Evaluation of IOR
(14): 45–60.
Potential Within Kuwait. Paper SPE 88716 MS presented at the Abu
Dhabi International Conference and Exhibition, Abu Dhabi, United Leonard. 1984. EOR Set To Make Significant Contribution. Oil & Gas J.:
Arab Emirates, 10–13 October. http://dx.doi.org/10.2118/88716-MS. 59–80.
Alvarado, V. and Manrique, E. 2010. Enhanced Oil Recovery—An Update Leonard. 1986. Increased Rate of EOR Brightens Outlook. Oil & Gas J.:
Review. Energies 3 (9): 1529–1575. http://dx.doi.org/10.3390/en3091529. 67–86.
Anonymous. 2012. 2012 Worldwide EOR Survey. Oil & Gas J.: 3–4. Maerker, J.M. 1975. Shear Degradation of Partially Hydrolyzed Polyacryl-
Asghari, K. and Nakutnyy, P. 2008. Experimental Results of Polymer amide Solutions. SPE J. 15 (4): 311–322. http://dx.doi.org/10.2118/
Flooding of Heavy Oil Reservoirs. Paper SPE 2008-189 presented at 5101-PA.
the Canadian International Petroleum Conference, Calgary, Alberta, Moe Soe Let, K.P., Manichand, R.N., Suriname, S. et al. 2012. Polymer
Canada, 17–19 June. http://dx.doi.org/10.2118/2008-189-MS. Flooding a 500-cp Oil. Paper SPE 154567 presented at the SPE

February 2014 SPE Reservoir Evaluation & Engineering 23

 Improved Oil Recovery Symposium, Tulsa, Oklahoma, 14–18 April. High-Salinity Reservoirs With Polyacrylamide Based Ter-Polymers.
http://dx.doi.org/10.2118/154567-MS. Paper SPE 141497 presented at the SPE Middle East Oil and Gas
Mogbo, O. 2011. Polymer Flood Simulation in a Heavy Oil Field: Off- Show and Conference, Manama, Bahrain, 25–28 September. http://
shore Niger-Delta Experience. Paper 145027 presented at the SPE dx.doi.org/10.2118/141497-MS.
Enhanced Oil Recovery Conference, Kuala Lumpur, Malaysia, 19–21 Wang, D., Cheng, J., Wu, J. et al. 2002. Producing by Polymer Flooding
July. http://dx.doi.org/ 10.2118/145027-MS. More Than 300 Million Barrels of Oil, What Experiences Have Been
Moritis, G. 1992. EOR Increases 24% Worldwide; Claims 10% of US Pro- Learnt?. Paper SPE 77872 presented at the SPE Asia Pacific Oil and
duction. Oil & Gas J.: 32–44, 89–108. Gas Conference and Exhibition, Melbourne, Australia, 8–10 October.
Moritis, G. 1994. EOR Dips in U.S., but Remains a Significant Factor. Oil http://dx.doi.org/10.2118/77872-MS.
& Gas J.: 27–77. Wang, D., Seright, R.S., Shao, Z., Wang, J., 2008. Key Aspects of Project
Moritis, G. 1996. 1996 Worldwide EOR Survey. Oil & Gas J.: 46–51. Design for Polymer Flooding at the Daqing Oilfield. SPE Res Eval &
Moritis, G. 1998.1998 Worldwide EOR Survey. Oil & Gas J.: 64–65. Eng. 11 (6): 1117-1124. http://dx.doi.org/10.2118/109682-PA
Moritis, G. 2000. 2000 Worldwide EOR Survey. Oil & Gas J.: 50, 51, 58–60. Wang, D., Dong, H., Lv, C. et al. 2009. Review of Practical Experience by
Moritis, G. 2002. 2002 Worldwide EOR Survey. Oil & Gas J.: 76–79. Polymer Flooding at Daqing. SPE Res Eval & Eng 12 (3): 470–476.
Moritis, G. 2004. EOR Continues To Unlock Oil Resources. Oil & Gas J.: http://dx.doi.org/10.2118/114342-PA.
58–61. Wassmuth, F., Green, K., and Arnold, W. 2009. Polymer Flood Applica-
Moritis, G. 2006. 2006 Worldwide EOR Survey. Oil & Gas J.: 52–55. tion To Improve Heavy Oil Recovery at East Bodo. J. Cdn. Pet. Tech.
Moritis, G. 2008. 2008 Worldwide EOR Survey. Oil & Gas J.: 54–57. 48 (2): 55–61. http://dx.doi.org/10.2118/09-02-55-PA.
Noran, D. 1978. Growth Marks Enhanced Oil Recovery. Oil & Gas J. An- Wassmuth, F., Green, K., Hodgins, L. et al. 2007. Polymer Flood Technol-
nual Production Report (113): 1–11. ogy for Heavy Oil Recovery. Paper SPE 2007-182 presented at the Ca-
Pye, D.J. 1964. Improved Secondary Recovery by Control of Water Mo- nadian International Petroleum Conference, Calgary, Alberta, Canada,
bility. J. Pet Tech 16 (8): 911–916. http://dx.doi.org/10.2118/845-PA. 12–14 June. http://dx.doi.org/10.2118/2007-182-MS.
Qiu, Y., Holtz, M., and Yang, A. 2001. Applying Curvature and Fracture Wei, Z., Jian, Z., Ming, H. et al. 2007. Application of Hydrophobically
Analysis to the Placement of Horizontal Wells: Example From the Associating Water-Soluble Polymer for Polymer Flooding in China
Mabee (San Andres) Reservoir, Texas. Paper SPE 70010 presented at Offshore Heavy Oilfield. Paper 11635 presented at the International
the SPE Permian Basin Oil and Gas Recovery Conference, Midland, Petroleum Technology Conference, Dubai, UAE, 4–6 December.
Texas, 15–17 May. http://dx.doi.org/10.2118/70010-MS. http://dx.doi.org/ 10.2523/11635-MS.
Raney, K., Ayirala, S., Chin, R. et al. 2011. Surface and Subsurface Wenxiang, W., Demin, W., and Haifeng, J. 2007. Effect of the Visco-elas-
Requirements for Successful Implementation of Offshore Chemical ticity of Displacing Fluids on the Relationship of Capillary Number
Enhanced Oil Recovery. Paper 21188 presented at the Offshore Tech- and Displacement Efficiency in Weak Oil-Wet Cores. Paper SPE
nology Conference, Houston, Texas, 2–5 May. http://dx.doi.org/ 109228 presented at the Asia Pacific Oil and Gas Conference and
10.4043/21188-MS. Exhibition, Jakarta, Indonesia, 30 October–1 November. http://
Roger, J.H., Blanton, M.L., Kuhlman, M. et al. 2012. Managing Risk in dx.doi.org/10.2118/109228-MS.
EOR Projects. Paper SPE 152700 presented at the SPE Latin America Wilkes, J.H. 1981. Injection Well Workover Program in the Levelland
and Caribbean Petroleum Engineering Conference, Mexico City, Mex- Field: A Case History. Paper SPE 9764 presented at the SPE Produc-
ico, 16–18 April. http://dx.doi.org/10.2118/152700-MS. tion Operations Symposium, Oklahoma City, Oklahoma, 1–3 March.
Sandiford, B.B. 1964. Laboratory and Field Studies of Water Floods Using http://dx.doi.org/10.2118/9764-MS.
Polymer Solutions to Increase Oil Recoveries. J. Pet Tech 16 (8): Yabin, N., Jian, O., Zhuoyan, Z. et al. 2001. Research on Hydrophobically
917–922. http://dx.doi.org/10.2118/844-PA. Associating Water-Soluble Polymer Used for EOR. Paper SPE 65378
Seright, R.S. 2010. Potential for Polymer Flooding Reservoirs With Vis- presented at the SPE International Symposium on Oilfield Chemistry,
cous Oils. Paper SPE 129899 presented at the SPE Improved Oil Re- Houston, Texas, 13–16 February. http://dx.doi.org/10.2118/65378-MS.
covery Symposium, Tulsa, Oklahoma, 24–28 April. http://dx.doi.org/ Zaitoun, A., Tabary, R., Fossey, J.P. et al. 1998. Implementing a Heavy Oil Hor-
10.2118/129899-MS. izontal Well Polymer Flood in Western Canada. http://ceor-alliance.com.
Shannon, L, 1980. EOR Methods Help Ultimate Recovery. Oil & Gas J.: Zhou, W., Zhang, J., Feng, G. et al. 2008. Key Technologies of Polymer
12–16. Flooding in Offshore Oilfield of Bohai Bay. Paper SPE 115240 pre-
Strauss, J.P., Alexander, D.M., Al-Azri, N. et al. 2010. EOR Development sented at the SPE Asia Pacific Oil and Gas Conference and Exhibition,
Screening of a Heterogeneous Heavy Oil Field—Challenges and Solu- Perth, Australia, 20–22 October. http://dx.doi.org/10.2118/115240-MS.
tions. Paper SPE 129157 presented at the SPE EOR Conference at Oil
and Gas West Asia, Muscat, Oman, 11–13 April. http://dx.doi.org/
10.2118/129157-MS. SI Metric Conversion Factors
Taber, J.J. 1983. Technical Screening Guides for the Enhanced Recovery. 
API 141.5/ (131.5þ API) ¼ g/cm3
Paper SPE 12069 presented at the SPE Annual Technical Conference  
F ( F- 32)/ 1.8 ¼ c
and Exhibition, San Francisco, California, 5–8 October. http://
md  9.869233 E04 ¼ cm3
dx.doi.org/10.2118/12069-MS.
Taber, J.J., Martin, F.D., and Seright, R.S. 1997a. EOR Screening Criteria cp  1.0 E03 ¼ Pa .s
Revisited—Part 1—Introduction to Screening Criteria and Enhanced ft  3.048* E01 ¼ m
Recovery Field Projects. SPE Res Eng 12 (3): 189–198. http://
*Conversion factor is exact.
dx.doi.org/10.2118/35385-PA.
Taber, J.J., Martin, F.D., and Seright, R.S. 1997b. EOR Screening Criteria
Revisited—Part 2—Applications and Impact of Oil Prices. SPE Res
Eng 12 (3): 199–206. http://dx.doi.org/10.2118/39234-PA. Laila Dao Saleh, SPE, is a PhD candidate at Missouri University
of Science and Technology majoring in petroleum engineer-
Tang, X., Liu, Y., Bai, B. et al. 2005. Study and Application of Weak Gel
ing. Her research interests are in EOR methods, especially poly-
Used for an Offshore Oilfield. Paper SPE 2005-076 presented at the mer flooding. Saleh holds BS and MS degrees in petroleum
Canadian International Petroleum Conference, Calgary, Alberta, Can- engineering from Tripoli University, Libya.
ada, 7–9 June. http://dx.doi.org/10.2118/2005-076-MS.
Tyler, N. and Finley, R. 1991. Architectural Controls on the Recovery of Mingzhen Wei, SPE, is an assistant professor of petroleum engi-
neering at Missouri University of Science and Technology. She
Hydrocarbons From Sandstone Reservoirs, SEPM Concepts in Sedi-
was an assistant professor in petroleum engineering at China
mentology and Paleontology, Vol. 3, pages 3–7, SEPM Online Publi- University of Geoscience before she started studies at New
cations. http://dx.doi.org/10.2110/csp.91.03.0001. Mexico Institute of Mining and Technology in pursuit of a PhD
Vermolen, E.C., van Haasterecht, M., Masalmeh, S. et al. 2011. Pushing degree in computer science in 2000. Wei holds BS and MS
the Envelope for Polymer Flooding Towards High-Temperature and degrees in petroleum engineering from China University of

24 February 2014 SPE Reservoir Evaluation & Engineering

Petroleum. She has expertise in data management and min- graduate research assistant at the New Mexico Petroleum Re-
ing and in numerical simulation. covery Research Center for EOR projects. Bai has nearly 20
years of experience in the area of EOR, especially in gel treat-
Baojun Bai, SPE, is an associate professor of petroleum engi- ments. He holds PhD degrees in petroleum engineering from
neering at Missouri University of Science and Technology. He New Mexico Institute of Mining and Technology and in petro-
holds the Lester R. Birbeck endowed Chair position. Previously, leum geology from China University of Geoscience-Beijing. Bai
Bai was a reservoir engineer and head of the conformance- has published more than 80 papers in the area of EOR meth-
control team at the Research Institute of Petroleum Explora- ods and application. He serves on the JPT Editorial Committee
tion and Development, PetroChina. He also was a post-doc- and is a technical editor for SPE Journal and SPE Reservoir Eval-
toral scholar at the California Institute of Technology and a uation and Engineering.

February 2014 SPE Reservoir Evaluation & Engineering 25