SPE-179558-MS

The Thermal Stability of Polyacrylamides in EOR Applications

Frederick Swiecinski and Peter Reed, Nalco/An Ecolab Company; William Andrews, TIORCO/A Nalco Champion
Company

Copyright 2016, Society of Petroleum Engineers

This paper was prepared for presentation at the SPE Improved Oil Recovery Conference held in Tulsa, Oklahoma, USA, 11–13 April 2016.

This paper was selected for presentation by an SPE program committee following review of information contained in an abstract submitted by the author(s). Contents
of the paper have not been reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material does not necessarily reflect
any position of the Society of Petroleum Engineers, its officers, or members. Electronic reproduction, distribution, or storage of any part of this paper without the written
consent of the Society of Petroleum Engineers is prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 words; illustrations may
not be copied. The abstract must contain conspicuous acknowledgment of SPE copyright.

Abstract
The prediction of polymer degradation under reservoir conditions is critical to EOR polymer selection.
NMR studies were used to monitor the changes in the anionic content of acrylamide containing polymers.
The monomer sequence distribution of hydrolyzed polyacrylamide polymers (HPAM) was monitored
during lab test to improve our understanding of HPAM stability. The results suggest that accelerated aging
of a polymer at reservoir conditions at elevated temperatures will result in a polymer with structure similar
to the polymer resulting from extended aging at lower temperatures.

Introduction
Polymer flooding is a widely used enhanced oil recovery (EOR) technique that can achieve a good sweep
efficiency of the reservoir because the injected fluid viscosity can be increased to closely match the
viscosity of the oil in the formation. This is achieved by use a polymer that is dissolved in the injection
water. High molecular weight, anionic polyacrylamides are the types of polymers typically employed in
this operation. The injected polymer solutions must maintain their viscosity for months or even years.
Polyacrylamides are subject to degradation during use and this can lower the viscosity of the injected
fluid. Polymer degradation can be caused by mechanical shear, oxidation, or high temperatures. The
primary thermal degradation pathway of polyacrylamides is the hydrolysis of the acrylamide and other
hydrolysable groups on the polymer backbone. Levitt and Pope (2008) provide an excellent summary of
the issue resulting from the chemical and thermal degradation of acrylamide based polymers. The
consequence of acrylamide polymer chemical degradation is an increase in anionic charge, potentially
resulting in precipitation of the polymer in the presence of multivalent cations.
Precipitation of the hydrolyzed EOR polymer is known to be related to the amount of anionic character
of the polymer, but several questions about the relationship between EOR polymer hydrolysis and
precipitation remain unanswered. First of all, it is unclear the role the polymer structure plays in the
precipitation of hydrolyzed acrylamide polymers. That is, is it merely the amount of anionic charge
present that is controlling factor or will the monomer sequencing of the polymer have an effect on when
the polymer will precipitate?
It would be desirable to be able to predict the rate of hydrolysis of EOR polymer at reservoir conditions.
There are several factors that would influence this hydrolysis rate and temperature is one major concern.
2 SPE-179558-MS

It is also unclear how the hydrolysis temperature affects the monomer sequencing of the resulting
hydrolyzed polymer. In this work we study the hydrolysis of an acrylamide/sodium acrylate copolymer
dissolved in synthetic sea water over a temperature range of 60-120°C. Taylor and Nasr-El-Din (1994)
investigated a number of analytical techniques and found that 13C NMR was the most reliable method for
determining the degree of hydrolysis of concentrated polymer samples. Halverson (1985) first used 13C
NMR to characterize the microstructure of hydrolyzed polyacrylamide by determining the distribution of
three monomer blocks or triad units. Other studies have also investigated the structure of acrylamide/
acrylate polymer (Truong et.al. 1986) and the hydrolysis of polyacrylamide (Yasuda et.al 1988) using 13C
NMR.
Experimental Methods
Sample Preparation
For the 60°C and 91°C samples, a solution of 3.5% by weight sea salt was prepared by dissolving sea
salt (Sigma) in 99.9% deuterium oxide (CIL). The NMR samples were prepared by weighing 150 mg of
the dry polymer into a standard 10 mm NMR tube. The 3.5% sea salt deuterium oxide solution was
pipetted over the polymer to create a mixture of about 4% polymer solids. The mixture was allowed to
stand for several hours at room temperature to swell uniformly.
For the 120°C samples, the elevated temperature of the study required the use of a NMR tube capable
of withstanding the expected pressures. A 10 mm precision pressure/vacuum valve NMR sample tube
(Wilmad-LabGlass) was used (Figure 1). The polymer solution was prepared as described above, but in
a small vial rather than directly in the NMR tube. The polymer/sea salt mixture was transferred into the
pressure/vacuum NMR tube. The valve was used to seal the NMR tube and prevent water loss when
heating the sample.

Figure 1—10 mm precision pressure/vacuum valve NMR sample tube (reproduced with the permission of Wilmad-LabGlass)

NMR Sample Heat Treatment

For the 60°C and 91°C samples, the samples in capped NMR tubes were placed in aluminum heating
blocks contain 12 mm diameter holes. The portion of the NMR tube containing the polymer was below
the top of the heating block and insulation was used to fill the void between the NMR tube and the heat
block. Except for the periods that the samples were being analyzed by 13C NMR, the samples remained
in the heating blocks. No solvent loss was observed during the course of the study.
For the 120°C sample, the 10 mm precision pressure/vacuum valve NMR sample tube was heated in
an oven set at 120°C. The sample was heated for the required period and allowed to return to room
temperature before NMR analysis. Between the 120°C heating periods, the sample remained at room
temperature. Again, no solvent loss was observed during the course of the study.
NMR Experiments
The 13C NMR data was collected on either a 300 MHz or 500 MHz Agilent/Varian NMR spectrometer
with a 13C frequency of 75 MHz or 125 MHz, respectively. The spectra were collected using quantitative
SPE-179558-MS 3

conditions which included a 10 s repetition rate, 90° pulse, proton decoupling only during acquisition
(NOE suppressed). The spectra were collected at 60°C to provide narrower NMR lines which would better
resolve the carbonyl signals. Typically, spectra were signal averaged for at least 8 hours, but experiment
times could be significantly longer when the 300 MHz instrument was used.
The data was processed using Mnova software by Mestrelab Research S.L. The data was Fourier
transformed after 5 Hz of exponential line broadening was applied to the FID. The spectra were baseline
corrected using an eighth order polynomial fit.
Triad Distribution Analysis
The polymer triad distribution was determined using the Line Fitting function of the Mnova NMR
software from Mestrelab Research. The chemical shifts of the six triads are separated by approximately
0.5 ppm. The line fitting process kept the peak width and peak shape parameters of all six triad signals
identical and fixed the acrylate and acrylamide triad signal separation at about 0.5 ppm. Although, the
peak width and peak shape of the triad signals were kept constant for each sample, these conditions were
allowed to vary between samples as the viscosity and other sample properties changed upon aging.
Cloud Point Analysis
The cloud point studies were performed on solutions containing 1,000 ppm polymer, 5 weight percent
sodium chloride, and a hardness level varying between 1,000 ppm and 10,000 ppm. For example, the
appropriate amount of an aged 4 weight percent polymer gel obtained from the NMR tube after aging and
NMR analysis was dissolved in the sodium chloride brine and then appropriate amounts calcium chloride
dihydrate and magnesium chloride hexahydrate were added to reach all of the final target concentrations
in the final solution. The total hardness values reported represent the sum of equal weights of calcium and
magnesium ions (i.e., 1,000 ppm hardness represents 500 ppm calcium ions and 500 ppm magnesium
ions).
Cloud points were determined by placing 2.5 grams of the test solution into sealed tubes (12 X 1.75
cm). The sealed tube was then immersed in an oil bath to a point above the solution level in the tube and
then heated at a rate of 1 to1.5 degrees Celsius per minute. An ice/water bath was used for cloud point
measurements below room temperature. The cloud point was noted as the temperature that the solution
first became turbid, in reference to a blank solution containing no polymer. Replicate test results were
averaged and reported as the cloud point.
Viscosity Testing
Hydrolyzed solutions of polymer were prepared by placing the polymer solutions (12 g) in Ace
pressure tubes. The tubes were sealed after deoxygenating for 30 minutes with a nitrogen purge. One tube
per data point was then placed in an oven. At appropriate intervals, a tube was removed and cooled to
room temperature, The viscosities were measured using a Brookfield DV2T viscometer at 39.2 s⫺1 shear
rate.

Results
The chemical degradation of EOR polymers at reservoir conditions has been extensively studied. Truong
et. al., (1986) demonstrated that the polymer sequencing of EOR flooding polymers will be dependent on
whether the polymer is produced by copolymerization of acrylamide and sodium acrylate or by alkaline
hydrolysis of polyacrylamide commonly referred to as hydrolyzed polyacrylamide (HPAM). It is also well
documented that additional hydrolysis of HPAM polymer will occur under reservoir conditions and that
the viscoelastic properties of the HPAM will change and can potentially result in the precipitation of the
polymer in hard brines (Moradi-Araghi and Doe 1987).
It would be desirable to predict the long-term performance of HPAM EOR polymers under reservoir
conditions. The most straightforward approach would be to accelerate the hydrolysis of the polymer by
subjecting the polymer to elevated temperatures. Spildo and Sæ (2015) have recently reported that the
4 SPE-179558-MS

polymer sequencing will influence the viscoelastic properties in calcium containing brines and should be
considered in the EOR polymer selection. The question becomes, will accelerated aging of EOR polymers
at high temperatures result in the same polymer sequence distribution that would result from long-term
aging at lower temperatures.
The chemical degradation of two product were evaluated, EOR980 a 70:30 mole ratio acrylamide/
acrylate copolymer and EOR909 a 75:25 mole ratio acrylamide/sodium acrylamide tertiary butyl sulfonate
(NaATBS) copolymer (Figure 2). Using 13C NMR it is possible to measure the relative molar amount of
acrylamide, acrylate and NaATBS that make up the EOR polymers. In the case of acrylamide/acrylate
copolymers not only can 13C NMR be used to monitor the amount of acrylamide hydrolyzed to acrylate,
but the acrylamide and acrylate signals are influence by neighboring groups so the polymer can be
characterized as combinations of extended structural units.

Figure 2—Structure of acrylamide/acrylate copolymer monomer units and acrylamide/sodium acrylamide tertiary butyl sulfonate
copolymer

Acrylamide/Acrylate Copolymers
Many studies have been conducted to understand the mechanism of the hydrolysis of polyacrylamide and
the resulting polymer structures (Halverson et.al. 1985, Troung et.al. 1986, Yasuda et.al. 1988). The
consensus of this work is that as hydrolysis proceeds an autoretardation mechanism resulting from the
neighboring acrylate group slows the rate of acrylamide hydrolysis. This is illustrated in Figures 3, 4 and
5 where the change in the acrylamide/acrylate ratio of the polymer is shown at temperatures of 60°C, 91°C
and 120°C as a function of time. At all three temperatures, the hydrolysis rate appears to begin to slow
when the acrylate content of the polymer reaches about 45%. This decrease in hydrolysis can be related
to the structural change in the polymer by looking at the polymer sequencing as measured by 13C NMR.
SPE-179558-MS 5

13
Figure 3—Mole percent acrylate and acrylamide composition of EOR980 polymer measured by C NMR over 675 days at 60°C

13
Figure 4 —Mole percent acrylate and acrylamide composition of EOR980 polymer measured by C NMR over 510 days at 91°C
6 SPE-179558-MS

13
Figure 5—Mole percent acrylate and acrylamide composition of EOR980 polymer measured by C NMR over 510 hours at 120°C

The typical approach to understanding the temperature dependence of hydrolysis would be to model
the kinetics of the hydrolysis by fitting the data to a model that would give a good fit at all temperatures.
Unfortunately, no simple kinetic model was found that described the hydrolysis over the entire time period
studied. It may be possible to estimate the kinetics of over a relative limited range of acrylamide
hydrolysis or hydrolysis time, but in order to predict the long-term performance the model must describe
the hydrolysis results over months.
The weakness of these models is that it treats all acrylamide groups identically. The model that will
predict the long-term hydrolysis will need to consider that the acrylamide groups of the polymer are not
equal. The difference between the acrylamide groups of an acrylate/acrylamide copolymer is clearly
illustrated in the 13C NMR spectrum of an acrylamide/acrylate copolymers. The 13C NMR chemical shift
of the carbonyl groups of HPAM is influenced by the neighboring acrylamide or acrylate groups creating
a slight shift of the carbonyl signals of the acrylamide and acrylate groups resulting in separate signals for
each possible three monomer sequences or triad unit. From the quantitative 13C NMR spectrum it is
possible to identify and measure the distribution of the six triads. The carbonyl region of the 13C NMR
spectrum of EOR980 is shown in Figure 6. Acrylamide carbonyl signals afford three broad signals at
centered around 180.5 ppm and the acrylate carbonyl signals are observed at around 184 ppm. The relative
areas of these signals provide the acrylamide/acrylate mole ratio of the polymer. Using line fitting
methods, it is possible to resolve these overlapping signals into three separate signals representing the
relative number of triad groups. In Figure 6, the position of each of the six triad signals is illustrated with
A representing an acrylamide unit and B representing an acrylate unit.
SPE-179558-MS 7

13
Figure 6 —Carbonyl region of C NMR spectrum of acrylamide/acrylate copolymer (EOR980)

In Figure 7, the change in the mole fraction of each of the six monomer triad sequences is shown as
function of the time the HPAM polymer was aged at 120°C. At the early stages of hydrolysis the change
in the triad distribution is dominated by loss of the AAA triad. The expectation is the AAA triads would
be hydrolyzed to ABA triads and these should increase dramatically. Examination of the data indicates
that the triad concentration that increases the fastest while the AAA triad is decreasing are the BAB and
ABB triads. This result is not institutive and illustrates the complex nature of the extended structure of the
polymer. Figure 8 shows the possible triad transitions due to acrylamide hydrolysis. Yasuda et. al. (1988)
modeled the alkaline hydrolysis of polyacrylamide by assigning probabilities to the hydrolysis of the
acrylamide centric triads.
8 SPE-179558-MS

Figure 7—Mole fraction of the six acrylamide/acrylate triads resulting from the hydrolysis of EOR980 versus hours at 120°C

Figure 8 —The possible triad conversion due to acrylamide (A) hydrolysis to acrylate (B)

In this study, a similar approach of assigning probabilities to the triad transitions shown in Figure 8 did
not provide an adequate model of the triad data. The problem encountered was that the model could
reasonable fit the triad curves during the early stages of hydrolysis or in latter stages of hydrolysis but gave
disappointing results for the entire data set.
Without an effective model to predict the hydrolysis rate, an empirical approach was applied to predict
the expected hydrolysis at lower temperatures, based on the hydrolysis data at 120°C. A comparison of
SPE-179558-MS 9

the time required to obtain the same extent of hydrolysis at different temperatures would provide some
insight to the acceleration that the higher temperature provides. Unfortunately, the hydrolysis was
measured at random time intervals and there are no two temperatures where the exact amount of
hydrolysis was measured. The 13C NMR hydrolysis data does provide fairly regularly spaced data points
over the range of hydrolysis, so it was felt that interpolation between data points would provide reasonable
results. For example, for the 120°C hydrolysis at 4 hours an acrylate value of 29 mole percent was
measured and at 8 hours an acrylate value of 31 mole percent was measured. It was calculated that 30 mole
percent hydrolysis was achieved at 6 hours. In Figures 9, the time it takes for the acrylate content to
increase in 5% steps is plotted at hydrolysis temperatures of 60°C and 120°C. The maximum acrylate
obtained in this study at 60°C was 55%. Over this hydrolysis range the plot of the data at 60°C and 120°C
is relatively linear. The slope of the line is ratio of the hydrolysis rates at 60°C and the hydrolysis rate at
120°C (k60C/k120C). The Arrhenius equation can be rearranged to represent the activation energy of
hydrolysis in terms of the ratio of hydrolysis rates. A hydrolysis activation energy of 88 kJ/mol was
obtained for the 60°C and 120°C data.

Figure 9 —Plot the hours required to reach specific levels of mole percent acrylate content in EOR980 polymer when aged at
temperatures of 60°C vs 120°C

In a similar manner the plot of the time to reach 5% increments of acrylate content at temperature of
91°C and 120°C (Figure 10) results in a more complicated relationship. Over the 500 plus days that the
10 SPE-179558-MS

91°C hydrolysis sample was monitored, an acrylate content of over 80% was observed. Figure 10 shows
a definite deviation from a linear relationship after an acrylate content of 60% is achieved. At acrylate
content below 60% the data is relatively linear. Using the slope from the plot of the acrylate content from
30% to 60% to estimate the acrylamide hydrolysis activation energy for the 91°C and 120°C data results
in value of 87 kJ/mol. The activation energy values measured are similar to values obtained for the
acrylamide hydrolysis of other acrylamide copolymers (Al-Muntasheri, et. al. 2008 and Parker and Lezzi
1993). The similarity between the two activation energies measured in this study indicates that the
prediction of acrylamide hydrolysis based on higher temperature hydrolysis data is possible if the amount
of hydrolysis is less than 60%.

Figure 10 —Plot the hours required to reach specific levels of mole percent acrylate content in EOR980 polymer when aged at
temperatures of 91°C vs 120°C

It is unclear why the 91°C and 120°C hydrolysis data deviates from linearity after 60% acrylate content
is obtained. Additional investigation will be required to determine if this deviation is related to the
composition of the polymer or some change in the mechanism of hydrolysis.
The linear relationship between the times require to reach specific levels of HPAM hydrolysis at
different temperatures, as illustrated in Figures 9 and 10, indicates that it should be possible to accelerate
the aging of polymer under reservoir conditions and obtain a polymer that will be representative of the
polymer that will be produced in the formation at lower temperature and longer exposure times. This
accelerated aging is based on obtaining a polymer of the same acrylate content. The questions still exists
if the monomer sequencing of a polymer hydrolyzed at the reservoir temperature will be the same as a
SPE-179558-MS 11

polymer hydrolyzed at an elevated temperature. That is, how structurally similar are polymers that
hydrolyzed to the same acrylate content at different temperatures? It is possible that the rate of the
hydrolysis of some polymer sequences can have different kinetics resulting in polymers with different
sequencing. To answer this question, the mole fraction of each acrylate centered triad is plotted as a
function of the acrylamide content of the polymer using the data obtained by 13C NMR at hydrolysis
temperatures of 60°C, 91°C and 120°C in Figure 11. Similarly, the mole fraction of each acrylamide
centered triad is plotted as a function of the acrylamide content in Figure 12. In both plots, the triad
compositions of the polymers are independent of the hydrolysis temperature. The data illustrates that the
monomer sequencing of the hydrolyzed polymers with the same level of hydrolysis will be the same
regardless of the temperature of hydrolysis. This data indicates that accelerated hydrolysis at higher
temperatures will produce a polymer structure that would be identical to hydrolysis at lower temperatures.

Figure 11—Mole fraction of the three acrylate centric triads versus the acrylamide content of the polymer for the EOR980 samples aged
at 60°C, 91°C and 120°C
12 SPE-179558-MS

Figure 12—Mole fraction of the three acrylamide centric triads versus the acrylamide content of the polymer for the EOR980 samples
aged at 60°C, 91°C and 120°C

Acrylamide/Sodium Acrylamide Tertiary Butyl Sulfonate Copolymer

To investigate the potential of predicting the acrylamide hydrolysis rate of acrylamide and sodium
acrylamide tertiary butyl sulfonate (NaATBS) copolymers, a study similar to that conducted with HPAM
was initiated using EOR909. The polymer was aged at temperatures of 60°C, 91°C and 120°C and
monitored by 13C NMR. Acrylamide/NaATBS copolymers have been shown to have greater resistance to
acrylamide hydrolysis than do HPAM polymers (Parker and Lezzi 1993). And the data from this study
mirror the finding of Parker and Lezzi. Parker and Lezzi observed first-order kinetics for the hydrolysis
of acrylamide. Although under the conditions of this study, the hydrolysis of acrylamide proceeds at a
slower rate, the kinetics approximated first-order for the first 200 days. Likewise, Parker and Lezzi
observed significant hydrolysis of the ATBS only after a critical amount of acrylamide hydrolysis had
occurred. During this initial period of acrylamide hydrolysis, there is very little evidence of any hydrolysis
of the ATBS group. It is only after initial hydrolysis of acrylamide appears to slow is evidence detected
for the hydrolysis of ATBS to polymeric acrylate and 1-amino-2-methylpropane-2-sulfonate (Figure 13).
SPE-179558-MS 13

Figure 13—Mole percent acrylate and ATBS composition of EOR909 polymer measured by 13C NMR at temperatures of 60°C, 91°C and
120°C. The insert expands the data for the first fifteen days of hydrolysis

Physical Properties of Aged Acrylamide/Acrylate and Acrylamide/NaATBS Copolymers

The results of the NMR studies indicate that the mechanism of polyacrylamide hydrolysis under simulated
reservoir conditions is invariant of temperature up to at least 120°C for a level of hydrolysis up to at least
60 mole percent of the polyacrylamide copolymer. Therefore, a polyacrylamide mobility control polymer
can be expected to hydrolyze in exactly the same way in a polymer flood application regardless of the
reservoir temperature, and the polymer will experiences the same sequence of changes in its polymer
microstructure during its hydrolysis in the reservoir. The rheological and other properties of the polymer
should likewise display a common progression of changes during the hydrolysis timeframe, invariant of
the reservoir temperature. The polymer solubility, for example, is negatively impacted by the thermally-
induced hydrolysis of the polymer. The thermal stability of the polyacrylamide copolymer during its time
in the reservoir will therefore determine whether the polymer can be used without precipitating. Once the
injection fluid containing the polymer is away from the wellbore and deep in the anaerobic, low shear
environment of the reservoir, then the stability of the HPAM mobility control polymer is limited by this
thermally-induced polymer hydrolysis. If the residence time in the reservoir is long enough and the
reservoir temperature is high enough, then the hydrolysis will eventually reach an extent that the acrylate
content of the polymer becomes so high that the polymer will precipitate out of the injected fluid. The
polymer precipitation is caused by the ionic crosslinking of the acrylate carboxylates on the polymer with
multivalent metal ions in the reservoir water.
In order to confirm that the polymer solubility properties were changing over the course of the NMR
experiments as the hydrolysis proceeded in a way that is independent of the hydrolysis temperature, the
polymer was recovered from the NMR samples in order to measure the cloud points of the polymer
dissolved in brines containing varying levels of hardness. Poly(acrylamide/acrylate) copolymer solutions
display a characteristic cloud point in hard brines, which indicates the lowest temperature at which the
14 SPE-179558-MS

polymer is no longer soluble in the brine. The cloud point is most affected by the level of hardness in the
water and the level of hydrolysis on the polymer. Higher levels of water hardness and higher levels of
polymer hydrolysis reduce the cloud point of the polymer in the brine. The cloud points were measured
here using test solutions containing 1000 ppm polymer in brine. The brine was comprised of 5 weight
percent sodium chloride and a total level of hardness ranging from 1,000 to 10,000 ppm. The total
hardness was comprised of equal weights of calcium and magnesium ions (i.e., 1,000 ppm hardness is
comprised of 500 ppm calcium ions and 500 ppm magnesium ions). Under these conditions, HPAM
polymers possessing degrees of hydrolysis ranging from about 40 to 60 mole percent acrylate should
display cloud points ranging from about 0°C to about 150°C (Moradi-Araghi and Doe 1987).
In Figure 14, the cloud point results are shown for hydrolyzed and unhydrolyzed EOR980. The
unhydrolyzed polymer, possessing an acrylate level of 30 mole percent, displays a cloud point of about
100 °C at 10,000 ppm hardness. After EOR980 was aged at 60°C for 675 days in the NMR tube, it attained
a level of hydrolysis of 55 mole percent. At this point the cloud point exceeded 100°C at 1000 ppm
hardness but falls below 50°C at 10,000 ppm hardness. For comparison, also shown are the cloud points
of EOR980 aged as in the NMR experiment but at a higher temperature of 120°C for only six days, where
it attained a level of hydrolysis of 63 mole percent. Regardless of the hydrolysis temperature used to
produce the samples, the clouds point observed are consistent with those previously reported at similar
levels of polymer hydrolysis (Moradi-Araghi and Doe 1987). The cloud point results also show the
expected trend. As the acrylate content of the copolymer increases from 55 to 63 mole percent, the effect
is to lower the cloud points, i.e., the solubility, of the polymer in hard brines. Short screening tests of this
nature conducted at high temperatures might be used to enable the screening of the thermal stability and
rheological changes of a prospective mobility control polymer product considered for use under the
conditions of a specific target application (reservoir temperature, water chemistry, and polymer residence
time). Since the actual residence time of a mobility control polymer used in a polymer flood may exceed
two years, a short term test would be quite useful.

Figure 14 —Cloud points of EOR980 and thermally hydrolyzed EOR980 measured at 1000 ppm polymer concentration in 5 wt.% NaCl
over varying hardness levels. The EOR 980 (30/70 mole percent acrylate/acrylamide, squares) was thermally hydrolyzed at 60°Cover
575 days to a level of 55/45 mole percent acrylate/acrylamide (diamonds), and at 120°Cover 6 days to a level of 63/27 mole percent
acrylate/acrylamide (triangles)
SPE-179558-MS 15

Another important property that is impacted by the thermally-induced hydrolysis of the polymer is the
bulk viscosity that the polymer imparts to the injected fluid. It is this viscosity that provides the improved
mobility control as the plug of injected polymer solution sweeps across the formation during the polymer
flood. For a poly(acrylate/acrylamide) copolymer like EOR980, the bulk viscosity of a solution of the
polymer in hard brine will decrease once the thermally-induced hydrolysis of the poly(acrylate/
acrylamide) polymer causes the acrylate content in the copolymer to exceed about 40 mole percent
(Sheng). This is because the acrylate content of the polymer becomes high enough to induce ionic
crosslinking of the polymer via the multivalent ions in the water. For the more thermally stable ATBS
copolymers like EOR909 the situation is more complex, however, because the ATBS comonomer
improves the in-reservoir polymer solubility in two ways. First, the ATBS forms a mer unit on the polymer
that is relatively stable and will hydrolyze at a much slower rate than the acrylamide mer unit, so it makes
the mobility control polymer more inherently thermally stable. Secondly, the ATBS mer unit bears an
anion of a strong acid (a sulfonate anion) that, unlike the carboxylate of the anionic acrylate mer unit,
resists the ionic crosslinking that will lower the viscosity and ultimately cause the polymer to precipitate
if the hydrolysis is carried out long enough. The NMR studies can help determine the stabilizing
mechanisms as the polymer hydrolyzes during the course of its use in a particular application. For
example, the thermally-induced compositional changes in the EOR909 polymer are shown by NMR in
Figure 13. The ATBS mer unit is stable at 90°C for at least one year and at 120°C for at least 8 days.
Unlike the acrylamide mer unit, the ATBS mer unit displays only a minor drop in its level in the polymer
over these timeframes. Within this aging timeframe, any changes in the viscosity of the polymer cannot
be due to changing levels of the NaATBS mer unit. The viscosity changes actually observed within this
timeframe are shown in Figure 15, where the EOR909 75/25 acrylamide/NaATBS copolymer was aged
at 90°C and 120°C over 25 days while the bulk viscosities of deoxygenated 1500 ppm solutions in sea
water prepared from the polymer over the course of the aging period were measured. At 90°C, the
viscosity of the EOR909 remains relatively constant, and inspection of Figure 13 reveals that the acrylate
level, produce by the hydrolysis of acrylamide on the polymer, should remain below about 10 mole
percent during this timeframe. At 120°C, however, the bulk viscosity of the polymer falls rapidly over this
two week timeframe (Figure 15), while the acrylate level increase above 25 mole percent (Figure 13).
Apparently, the performance of the more thermally stable NaATBS copolymer first fails at high
temperatures not because of ATBS hydrolysis, but because the rapid acrylamide hydrolysis causes the
acrylate level to reach a point where the polymer solubility is severely compromised. At a 90°C reservoir
temperature, one would expect that when the viscosity performance eventually fails, it will fail due to
acrylamide hydrolysis as well, except at a much longer point in time. In this case, a polymer possessing
a higher level of NaATBS should be stable for a longer period at 90°C.
16 SPE-179558-MS

Figure 15—The bulk viscosity of deoxygenated 1500 ppm solutions of EOR909 (75/25 poly(acrylamide/ATBS) copolymer prepared from
polymer aged at 90°C (blue triangles), and at 120°C (red triangles)

Since these and previous NMR analyses show that the polyacrylamide copolymer itself will change in
composition over the course of its use in a polymer flood conducted even at moderate reservoir
temperatures, one needs to be careful in selecting a mobility control polymer based on the initial
rheological properties of the polymer that is introduced into the injection fluid. Minor differences in
rheology between similar polymers produced by different manufacturing methods, for example, would be
expected to even out as the polymer undergoes hydrolysis in the reservoir. On the other hand, in order to
accurately compare polymers of significantly different compositions, a view of the entire compositional
profile under simulated reservoir conditions is sometimes necessary to gain a full understanding of relative
polymer performance.

Conclusions
● The mechanism of polyacrylamide hydrolysis under simulated reservoir conditions is invariant of
temperature up to 120°C and up to at least 55 mole percent hydrolysis of the polyacrylamide
copolymer. Within this range of conditions, the polymer sequence distribution and rate of
hydrolysis can be predicted for an HPAM polymer.
● The deviation in the predicted rate of HPAM hydrolysis once 55 mole percent hydrolysis is
achieved suggests a change in the mechanism of hydrolysis with polymer structure.
● The progression of changes to the polymer structure during the hydrolysis process can be
correlated to changes in relevant physical properties of the polymer, and these property changes
should be considered when choosing a mobility control polymer for a polymer flood application.

Acknowledgement
The authors would like to recognize Aaron Majerczak’s contribution in the preparation of the NMR
samples and collection of the NMR data.
SPE-179558-MS 17

References
Al-Muntasheri, G. A., Nasr-El-Din, H. A., Peters, J. A. and Zitha, P. L. J. 2008. Thermal decomposition and hydrolysis
of polyacrylamide-co-tert-butyl acrylate. Eur. Poly. J. 44, 1225–1237.
Halverson, F., Lancaster, J. E. and O’Connor M. N. 1985. Sequence Distribution of Carbonyl Groups in Hydrolyzed
Polyacrylamide. Macromolecules 18, 1139 –1144.
Levitt, D. B. and Pope G. A. 2008. Selection and Screening of Polymers for Enhanced-Oil Recovery. Paper SPE 113845
presented at the Improved Oil Recovery Symposium, Tulsa, OK 19-23 April
Moradi-Arighi, A. and Doe, P.H. 1987. Hydrolysis and Precipitation of Polyacrylamide in Hard Brines at Elevated
Temperatures. SPERE 189 –198.
Parker, W. O. and Lezzi, A. 1993. Hydrolysis of sodium-1acryamido-2-methylpropansulfonate copolymers at elevated
temperature in aqueous solution via 13C n.m.r. spectroscopy. Polymer 34 (23), 4913–4918.
Spildo, K., Sæ, E.I.Ø. 2015. Effect of charge distribution on the viscosity and viscoelastic properties of partially
hydrolyzed polyacrylamide. Energy & Fuels 29 5609 –5617.
Sheng, James. Modern Chemical Enhanced Oil Recovery: Theory and Practice. Gulf Professional Publishing, 2010, page
103.
Truong, N. D., Galin, J. C., Francois, J. and Pham, Q. T. 1986. Microstructure of acrylamide-acrylic acid copolymers: 1.
As obtained by alkaline hydrolysis. Polymer 27, 459 –466.
Truong, N. D., Galin, J. C., Francois, J. and Pham, Q. T. 1986. Microstructure of acrylamide-acrylic acid copolymers: 2.
As obtained by direct copolymerization. Polymer 27, 467–475.
Yasuda, K., Okajima, K. and Kamide, K. 1988. Study of Alkaline Hydrolysis of Polyacrylamide by 13C NMR. Polymer
Journal 20 (12), 1101–1107.