You are on page 1of 9

SPE 72366

Multi Year Gas Storage Deliverability Results Using a Proppant Surface

Modification Agent
Gregory A. Kozera, Stanley F. Willis, Halliburton Energy Services, Inc., J. Stanley Shaw, Columbia Gas Transmission Corp.

Copyright 2001, Society of Petroleum Engineers, Inc. information can easily get buried in a sea of data. Wells using the
This paper was prepared for presentation at the SPE Eastern Regional Meeting held in once-new product or process are forgotten unless a serious
Canton, Ohio, 17-19 October 2001. problem occurs. The reason the product or process was first used
This paper was selected for presentation by an SPE Program Committee following review of may even be forgotten if it solved the problem.
information contained in an abstract submitted by the author(s). Contents of the paper, as
presented, have not been reviewed by the Society of Petroleum Engineers and are subject to While results may vary from formation to formation, this
correction by the author(s). The material, as presented, does not necessarily reflect any longer-term look at the results of SMA, injected with the proppant,
position of the Society of Petroleum Engineers, its officers, or members. Papers presented at
SPE meetings are subject to publication review by Editorial Committees of the Society of should help fracture treatment designers convince the skeptics
Petroleum Engineers. Electronic reproduction, distribution, or storage of any part of this paper
for commercial purposes without the written consent of the Society of Petroleum Engineers is that proppant-flowback control can achieve its immediate objec-
prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 tive and match or exceed long-term performance expectations.
words; illustrations may not be copied. The abstract should contain conspicuous
acknowledgment of where and by whom the paper was presented. Write Librarian, SPE, P.O.
Box 833836, Richardson, TX 75083-3836, U.S.A, fax 01-972-952-9435.
The gas-storage industry has recognized that storage-well and
storage-field performance declines over time. In a 1993 study,
This paper presents post-fracture deliverability results for 56 gas
the Gas Research Institute (GRI) determined that, on average,
-storage wells over a two- to five-year period. Of the wells
the storage industry loses over 5% of its deliverability annually.1
studied, 32 had a surface modification agent (SMA) applied to
Some fields or types of fields may naturally maintain deliverability
the proppant during the hydraulic fracturing treatment. The other
better than others. The operator performs various types of well
24 wells included in this study had no additives applied to the
work in annual programs to combat deliverability losses. The
proppant for sand flowback control.
most effective means of maintaining existing levels of
The deliverability results from actual flow tests are used to
deliverability in many fields is to treat wells with hydraulic
evaluate how absolute open flows (AOF) held up over multiple
fracture treatments (fracture jobs).
injection and withdrawal cycles involving significant pressure
differentials. The results are compared for the large data sets of
wells that did and did not include the SMA in the fracture
The operator owns and operates 42 underground gas-storage
treatment. In general, the data set for wells that did not include
fields. Many of the fields have been continuously operated for 50
an SMA have been through four or five complete pressure
years or more and are regulated by the Federal Energy Regula-
cycles. After some operational problems, the need for proppant
tory Commission (FERC). To expand deliverability, or perform
flowback control became apparent, and later frac treatments
many other types of well, pipe, or compression construction, the
usually included the SMA. The wells that included a SMA have
operator must first obtain approval from the FERC. Addition-
been through two or three complete pressure cycles.
ally, it should maintain existing levels of deliverability because
This study was completed to alleviate the operators con-
all deliverability is under contract to various customers. The
cern about the long-term impact of the SMA on the well
deliverability is aggregated over all 42 fields so that a customer
performance. While the SMA effectively achieved the short-
does not contract deliverability from any specific field.
term goal of proppant-flowback control, the operator was
The results presented in this study are from wells that were
concerned that the fracture might be more prone to a reduction
either part of two expansion projects between 1997 and 1999 or
in conductivity or simple clogging.
annual deliverability maintenance programs. With 42 storage
Typically, a new product or process gets a lot of attention
fields, conducting well work-overs in every field each year
and study. Many times the well results are tracked only for a year
makes little sense. Instead, work is grouped in programs to
or less and the product or process becomes common or dis-
reduce overall cost and environmental impact. A 1996
carded. Longer-term studies on a product or process are less
deliverability maintenance program resulted in subsequent op-
common, perhaps due to time, manpower, cost constraints,
erational complications from produced proppant and led to the
personnel changes, and a variety of other reasons. Over time,

References at the end of the paper.


decision to begin including a proppant-flowback control agent to with the expensive additional procedures, problems were still
each treatment. encountered with sand production in the field. The sand was
In 1996, hydraulic fracture treatments were pumped in eight most likely carried by water as the wells continued to clean up
wells in one field that had only 22 active injection/withdrawal through the first withdrawal season.
wells, equalling more than one third of the field in a one-year The actual amount of proppant produced by the wells was
period. Meanwhile, a pig launcher and pig receiver were added small, a fraction of the amount of proppant included in the
to the mainline from the field to allow internal inspection of the fracturing treatment. In fact, testing on a large group of wells
pipe and to improve liquid removal from the gas stream. A new drilled and fractured in the late 1970’s showed that most wells
slug catcher was also part of the new liquid-removal equipment. were still in nearly the same stimulated condition that treatments
Pipeline construction caused the field to remain shut-in almost were targeting twenty years later (for example, skin factors in the
two months past the start of the traditional withdrawal season. range of –3.5 to –4.5). The long-term effectiveness of twenty-
The field was brought online during the peak of the heating year-old treatments means that most of the frac sand is staying in
season, holding near-maximum volumes. The field was flowed place over the long term. The correlation of proppant production
into the field line at relatively high rates to return to its expected with a number of factors, such as the impact of permeability and
volume position, resulting in some proppant production. Proppant cased-hole vs. openhole completions, was also scrutinized. While
reached the slug catcher and eroded some fittings in the dump- some situations were obviously worse than others, the need for
flow control equipment. sand control was widespread across the study wells.
The new equipment was in service only a few weeks before
the slug catcher dump mechanism failed. The immediate failure of Characteristics of SMA
equipment designed for a long service life persuaded the operator As reported by P.D. Nguyen et al.,2 SMA was developed as a
to develop methods for mitigating erosion from frac sand in field new way to increase proppant surface friction by coating the
piping. Proppant problems at the dumps also prompted the moni- proppant particles, enabling higher flowback rates without
toring of pipe wall thickness on the first ells off each well that producing proppant. Laboratory testing has shown that SMA-
dropped the well line toward the ground. treated proppant can resist movement by up to four times the
While no erosion problems in the primary gas-stream fluid velocity when compared to untreated proppant. This
piping were observed, the dump mechanisms continued to be resistance permits aggressive flowback rates without the
problematic. Among other components, 1-in. valves and tees problems of proppant returns (Fig. 1).
on drip dump lines and the larger slug catcher 2-in. dump lines Testing has shown that adding SMA to the proppant at low
were cut. The lines were manned during use and were obvious closure stresses increases conductivity by up to 20% (Fig. 2). The
safety concerns. Proppant also cut the lines in water tanks and mechanism that prevents flowback of the proppant also reduces
accumulated in the pig receiver, and several thousand pounds the settling rate. This rate increase creates large openings in the
of proppant accumulated in several tanks. Erosion occurred proppant pack, which provide excellent channels for fluid and gas
only on dump lines because those were the only places where flow at closure pressures up to 4,000 psi (Figs. 3 and 4).
the proppant reached a high velocity. Hardened dump valves
and seats and dump line reconfigurations improved the erosion Subject Storage Fields
situations. However, because most of the piping was designed The 56 wells in this study are located in several similar storage
for a solid-free gas stream, reducing/eliminating the proppant fields in Kanawha, Jackson, Wood, and Wirt Counties, West
production seemed to be a better solution. Virginia, in the vicinity of the 1 Tcf Elk-Poca (Sissonville)3 Field.
Options to clean up the wells and eliminate further proppant The storage zone in each of the fields/wells is the Devonian-age
production after fracturing were evaluated and found to be Oriskany sandstone. The storage zone is bounded by up-dip
limited. A primary limitation involved environmental concerns. permeability barriers, down-dip saltwater, and Carbonate units
The operator flows wells to a tank after fracturing to eliminate above and below. The Oriskany is a fine- to medium-grained
the release of liquids. Flowback tanks are equipped with baffles calcareous sandstone cemented by quartz. Calcite is a minor
to remove liquids from the flow stream before venting the gas. In cement that reduces formation quality in some layers. Porosity is
many cases, this policy did not allow the wells to be flowed hard about 10% in the better wells. While some natural fracturing is
enough because the rate was limited by a ¾-in. choke. Even observed, possibly contributing to the performance of the best
though flow rates were often in the range of 15,000 to 20,000 wells, matrix permeability is dominant throughout the fields.
Mcf/D, subsequent higher rates into the pipeline occurred from Discovery pressures in these fields were approximately
several wells, especially in the storage field where some of the 1,900 psig. Average total depth (TD) for the wells is about 5,200
aforementioned problems occurred. ft. The average bottomhole temperature is 115°F. Maximum
In addition to the initial flowback procedure, the wells storage pressures on the fields range from 1,675 to 1,800 psig
described above underwent a procedure that included washing measured at the surface. AOF of over 100 MMcf/D have been
the wellbores clean with coiled tubing to remove any sand. Even achieved following stimulation treatments on the best wells.

Typical AOF’s are in the 30 to 60 MMcf/D range, and the better vertical distribution of proppant within a fracture. SMA
average AOF of the 56 study wells is just over 45,000 Mcf/D. allows for more aggressive well-cleanup procedures after frac-
Flow rates at some wells, during peak conditions, may reach 50% turing stimulation and does not require any shut-in time for
of AOF rates. activation and curing.
SMA met the following required criteria for the operator’s
Solution storage field:
The operator was initially reluctant to include additives in the • It allowed for immediate flowback after treatment.
fracture treatments to improve sand control at the wellbore. • It was easy to clean out of a wellbore in the event of a
The designed flow rates from the wells were high, so reduced screenout.
fracture conductivity, hence reduced deliverability, was unac-
ceptable. As the operator was seeking an optimal sand-control
• Based on laboratory results, it appeared to enhance fracture
solution, a new product (SMA) was offered that promised flow capacity and deliverability.
proppant flowback control and yet increased fracture conduc- • It was cost effective.
tivity.2 The SMA was applied to the proppant and a reduction The only uncertainty was whether it would control proppant
in proppant flowback during initial well cleanup procedures flowback under the pressure differentials in the operator’s storage
was immediately noticed. The reduction in proppant flowback fields. To answer this question, we ran the material on some
during the withdrawal seasons was noted as well. The long- fracturing jobs.
term impact of SMA to well performance was a concern to the
operator and an unknown factor when the first treatments were Job Design
pumped. The flow test results presented in a later section show Gas-storage wells typically contain high-permeability forma-
that the long-term impact is favorable. tions. Maximum deliverability requires a short, wide fracture,
A method for controlling proppant was necessary to reduce which can be obtained with the tip-screenout method (TSO). The
or eliminate damage to surface equipment from proppant in SMA additive further enhances proppant-pack conductivity.
recently stimulated storage wells. Several options were consid- Data from past fracturing treatments provided the input for
ered. Both curable and precured resin-coated sand would work, a fracturing design simulator. From the simulator, a model was
but could restrict deliverability. Thermoplastic ribbons/flake created for each storage field. Each fracture design had the pad
fibers were also considered, but again deliverability could be volume tailored to the formation leak-off characteristics. The
restricted. To be successful, the proppant flowback control pad volume should be depleted as the last sand stage enters the
method needed to be cost-effective, allow immediate flowback, formation, resulting in a TSO. Table 1 lists a typical job design.
work in varying well conditions, and not restrict deliverability. A low-gel, borate-crosslinked fluid was used as the fractur-
A major challenge with gas-storage wells is designing a ing fluid. It provided an instant crosslink and good sand transport
proppant-flowback control method or material that can with- at a much lower Guar loading than conventional borate-
stand peak reservoir pressure differential drawdown multiple crosslinked fluids.4 An enzyme breaker was used at a concentra-
times over several years. The proppant had to remain in place and tion that provided a break within two hours, enabling quick fluid
deliverability could not be lost. This need is what makes gas- turnaround. A foaming agent/surfactant helped recover fluid,
storage operations unique and more challenging compared to while a fluid-loss control agent controlled fluid leak-off in the
normal E&P operations. In normal E&P operations, the proppant pad. A bactericide was used to prevent contamination of the
flowback-control material would have to withstand the greatest formation. SMA was added to the proppant at 0.18 gal/sk.
pressure differential and the highest flowrates after fracturing Finally, nitrogen was used at 15 to 20% as a flowback aid and for
and during the first year of operation. After this time, reservoir reducing the amount of fluid entering the formation. In early
pressure is reduced and never returns to its initial pressure. In storage season treatments, a higher percentage of nitrogen is
gas-storage operations, the storage field is refilled and returns to used during the flush to prevent killing the well.
maximum reservoir pressure every year. High drawdown pres-
sure differentials occur at least once per year. Well Performance After Multiple Seasons
In 1997, SMA was developed and introduced to the industry A storage season at these fields is described as the traditional
for controlling proppant flowback in low-temperature wells. The “inject-gas-in-the-summer” and “withdraw-gas-in-the-winter”
material is applied as a liquid additive to water-based fluids on- model, although minor reversals in the greater trend do occur.
the-fly during hydraulic fracturing stimulation treatments. This Pressures at the wellbore may fluctuate by as much as 1,500 psig
agent instantly covers the proppant with a thin, tacky coating that from the season highs to lows.
dramatically increases intergrain friction and the fluid velocity Well-performance results in this study are based on com-
required to cause fluidization and mobility of the proppant pack. parisons of AOF vs. time. Other terms commonly used for AOF
This tackiness also significantly reduces the proppant settling may include open flow capacity or potential. The AOF is
rate and ultimately the density of the settled pack, resulting in a obtained by extrapolating tests performed at several moderate

rates, as opposed to opening the well wide to full flow. The up points to the initial 3-hr flows can be misleading and is therefore
pressure function is plotted vs. flow rate on a log-log plot and is noted.
typically referred to as a back-pressure or deliverability plot. The The 91% maintained performance would be approximately
methods are more fully discussed in Craft and Hawkins.5 Slopes equivalent to a 2% yearly decline in well performance. While
on deliverability curves are quantified by an n value and are this value is much better than the industry average, there is still
generally constant vs. flow time unless a major event (such as a room for improvement. Wells 331 and 353 were even excluded
fracturing job) occurs to the well. The lines from the flow points to obtain the more favorable results. Remediation on these
are extrapolated to AOF conditions by the difference in maxi- wells is now necessary because well performance is less than
mum well pressure to atmospheric pressure. The value of n takes pre-fracturing levels. Proppant depths will be checked in the
into account non-Darcy effects and tubulars (done at surface). wells to verify that much of the proppant is now in the wellbore.
The flow string ranges from 7 in. down to 3 ½-in. pipe. The AOF Much of the proppant that caused operational problems may
values are presented in Tables 2 through 4. have flowed out of these wells. The treatment for well 331
Table 2 lists 24 control wells where no SMA was used with included 22,000 lb of sand, and 353 included 38,000 lb of sand.
the proppant. Tables 3 and 4 both list study wells that had SMA Both wells were cased-hole completions and have permeabil-
added to the proppant. The column headers are consistent for the ity-thickness values (kh) of 2,200 and 3,200 md-ft respectively.
three tables. The first column lists the study well designation The permeability-thickness values and skin factor values were
used to track the results. The second column lists the year the obtained from the multi-point flow/buildup tests that serve as
hydraulic fracture treatment was pumped in the well. The next the basis for the initial post-frac AOF. Skin factors were
column lists the initial post-frac AOF rate which is extrapolated evaluated as a function of rate. Both the SMA and non SMA
from a multi-point flow/pressure buildup test and is usually wells reached designed levels of stimulation and both showed
based on 3-hr flow. The final four columns list the follow-up adequate fracture conductivity.
well-test results as a percentage of the initial post-frac AOF The average initial post-frac AOF for the 16 wells listed in
presented in the third column. Table 3 is 32,066 Mcf/D. All of these wells had an SMA included
The initial post-frac well tests each followed one complete with the proppant for sand flowback control. Table 3 wells are
storage cycle. Following the stimulation treatments, the wells in generally located in the same fields as Table 2, but well quality
some storage fields clean up and reach full potential after the first was slightly diminished, as shown by the average AOF of only
withdrawal season. Other fields take about two withdrawal 33,282 Mcf/D. Seven of the wells have been through three
seasons to reach full potential. This variation in clean-up time storage cycles and the other nine wells have been through two
has more to do with the basic operation of the field as a whole as storage cycles. The group as a whole has maintained an average
opposed to the well quality. However, some poorer deliverability AOF of 33,282 Mcf/D, or 104% of the initial or baseline value.
wells may require even more time. The longer clean up occurs at The deliverability of wells listed in Table 3 likely decreased
fields that are relatively more pipeline-constrained and have about 3% from the initial AOF to the AOF after two or three
lesser pressure differential pulls. Table 3 lists 16 study wells withdrawal seasons when the 7% time factor is applied. Those
with SMA that are in fields which typically clean up after one results are still showing evidence of only half the decline rate of
withdrawal season. Table 4 lists 16 study wells with SMA that wells that did not include the SMA with the proppant. Further-
are in fields that typically clean up after two withdrawal seasons. more, no subsequent fracturing failures (such as 331 and 353
The average initial post-frac AOF for the 24 wells listed in above) were noted in this group.
Table 2 is 57,546 Mcf/D. None of these wells had an SMA or any The average initial post-frac AOF for the 16 wells listed in
other additive included with the proppant for proppant-flowback Table 4 is 38,936 Mcf/D. All of these wells had an SMA included
control. One third of the wells have been through five storage with the proppant for sand flowback control. The wells included
cycles, and the other wells have been through four storage cycles. in Table 4 are located in different fields than the study wells in
The group as a whole has maintained an average AOF of 52,016 Tables 2 and 3, and the deliverability is back to top-quality wells
Mcf/D, or 90% of the initial or baseline value. Two of the wells with an average AOF of 47,637 Mcf/D. Six of the wells have
suffered exceptionally poor performance, wells 331 and 353. been through three storage cycles and the other ten wells have
Excluding wells 331 and 353, the group maintained 98% of the been through two storage cycles. The group as a whole has
baseline performance. maintained an average AOF of 47,637 Mcf/D, or 122% of the
An earlier internal study, based on multi-point flow tests initial or baseline value. Obviously, these wells were still clean-
with one extended flow period, indicated that deliverability ing up after the first withdrawal season when the initial post-frac
decreases 7% from one hour to three hours on stimulated wells. multi-point test was completed. Only one well showed any
The deliverability of wells listed in Table 1 decreased only 2% deliverability decrease, and no subsequent fracturing failures
from the initial AOF to the AOF after four or five withdrawal were noted in this group. The Table 4 results are difficult to
seasons. However, owing to the time-sensitive nature of the C compare directly to Table 2 results, but qualitatively indicate
value, perhaps the 2% decrease is in addition to the 7% time that SMA works at better wells.
factor to yield 91% of maintained AOF. Comparing the 1-hr follow-

The authors intend to continue monitoring the longer-term Acknowledgements

performance of these study wells. Not all of the wells will be tested The authors thank Halliburton Energy Services, Inc. and Colum-
each year, as was the case with the data in Tables 2 through 4. The bia Gas Transmission Corp. for their support and permission to
tests will be sporadic, but most wells will be checked eventually. publish this paper.

Conclusions References
• SMA was effective in the immediate objective of control- 1. “State-of-Technology Assessment and Evaluation of Gas Storage
ling proppant flowback after stimulation treatments in low- Well Productivity Enhancement Techniques,” Topical Report,
Contract No: GRI-93/0001, Gas Research Institute, Chicago,
temperature wells.
Illinois, December 1993.
• SMA clearly demonstrated a long-term positive impact on 2. Nguyen, P.D., et al.: “Enhancing Fracture Conductivity Through
deliverability. Surface Modification of Proppant,” paper SPE 39428 presented
• Wells treated without SMA can have significant degrada- at the 1998 Formation Damage Control Conference, Lafayette,
tion of deliverability. Note wells 331 and 353. 18-19 February.
3. The Atlas of Major Appalachin Gas Plays, West Virginia Geo-
• SMA was used successfully on wells with low and high logical and Economic Survey, Publication V-25, 1996, funded by
flowrates with no negative impact on deliverability. The U.S. Department of Energy, pg. 120. Edited by J.B. Roen and B.J.
highest-volume well in this study had a flowrate of approxi- Walker, contributions from D.G. Patchen and J.A. Harper.
mately 90 MMcf/D AOF. 4. Terracina, J.M., et al.: “Low Gel Loading Increases Well Produc-
• Most importantly, the positive deliverability results have tion in Eastern U.S.,” paper SPE 51070 presented at the 1998
Eastern Regional Conference, Pittsburgh, 9-11 November.
continued over several injection and withdrawal cycles.
5. Craft and Hawkins, Applied Petroleum Reservoir Engineering,
• Proppant flowback under peak deliverability continues to (Englewood Cliffs, N.J., Prentice-Hall, Inc., 1959) pp. 326-328.
be controlled even after several injection and withdrawal 6. Cullender, M.H.: “The Isochronal Performance Method of Deter-
cycles. Damage to surface equipment, such as pipe, valves, mining the Flow Characteristics of Gas Wells,” Trans. AIME
and tanks caused by fracturing sand has been virtually (1955) 204, 137.

Table 1—Typical Job Design

Stage Clean Volume Proppant Concentration Slurry Rate Nitrogen Rate
(gal) (ppg) (bpm) (scf/min)
Pad 6,000 0.00 12.8 4,400
1-lb sand 2,800 1.25 13.5 4,400
2-lb sand 2,800 2.50 14.2 4,400
3-lb sand 1,600 3.75 15.0 4,400
4-lb sand 1,600 5.00 15.7 4,400
Flush 2,500 0.00 12.8 4,400

Table 2—Post-Frac AOF vs. Time Study for Non-SMA Wells

Study Well Frac Year Initial 1998 Test 1999 Test 2000 Test 2001 Test
Designation Post-Frac AOF Followup Followup Followup Followup
(Mcf/D) (% AOF) (%AOF) (% AOF) (% AOF)
49 1997 46302 100 — 89 94
71 1997 54783 100 — — 82
90 1997 62063 100 — 97 99
92 1997 26289 100 — — 103
93 1997 39317 100 — — 91
109 1997 81556 100 — 74 83
110 1997 30615 100 — 101 103
111 1999 38995 — — 100 105
199 1997 68043 100 — 88 94
222 1996 95509 122 — 114 99
224 1997 48128 100 — — 94
225 1997 110948 100 — — 79
226 1999 34164 — — 100 119
324 1996 68094 100 — 87 93
327 1996 84365 107 — 117 113
331 1996 65699 108 — 20 39
335 1995 38651 104 — — 110
336 1996 36459 83 — — 81
353 1996 70356 — — — 9
354 1996 55684 101 — — 84
355 1996 45765 98 — — 94
375 1997 73371 100 — 100 127
379 1997 86535 100 — — 86
383 1997 19401 100 — — 228

Table 3—Post-Frac AOF vs. Time Study for SMA Wellsa

Study Well Frac Year Initial 1998 Test 1999 Test 2000 Test 2001 Test
Designation Post-Frac AOF Followup Followup Followup Followup
(Mcf/D) (% AOF) (%AOF) (% AOF) (% AOF)
43 1998 65847 — 100 96 86
54 1998 40060 — 100 102 98
56 1999 11085 — — 100 95
58 1999 22313 — — 100 110
72 1998 34556 — 100 72
79 1999 24892 — — 100 148
94 1998 20286 — — 100 99
96 1998 42851 — 100 86 87
102 1999 48067 — — 100 100
104 1998 32985 — 100 102
108 1999 30453 — — 100 123
216 1998 44762 — 100 84 87
229 1999 29593 — — 100 119
422 1999 43338 — — 100 136
423 1999 2693 — — 100 247
424 1999 19222 — — 100 119
Fields achieved cleanup in 1 year.

Table 4—Post-Frac AOF vs. Time Study for SMA Wellsa

Study Well Frac Year Initial 1998 Test 1999 Test 2000 Test 2001 Test
Designation Post-Frac AOF Followup Followup Followup Followup
(Mcf/D) (% AOF) (%AOF) (% AOF) (% AOF)
287 1999 53176 — — 100 115
238 1998 37821 — 100 n/a 103
260 1998 28673 — 100 n/a 78
294 1998 25345 — 100 91 95
297 1999 50940 — — 100 112
313 1999 56235 — — 100 141
316 1998 43239 — 100 92 102
317 1999 23832 — — 100 122
339 1998 67883 — 100 121 122
342 1999 71669 — — 100 111
344 1998 59573 — 100 117 101
347 1999 16623 — — 100 122
411 1999 14478 — — 100 142
413 1999 6474 — — 100 218
414 1999 27136 — — 100 117
416 1999 39875 — — 100 245
Fields achieved cleanup in 2 years.

20/40, 1%
0.009 20/40, 2%
Proppant Rate (ft3/D)

0.008 16/30, 0%
0.007 16/300, .5%
0.006 16/30, 2%
0.005 16/30, 1%
0.004 20/40, 0%
0.003 12/20, 0%
0 1 2 3 4 5 6 7 8
Fluid Rate (B/D)

Fig. 1—Effect of fluid flow rate on flowback of SMA-treated and untreated proppants, as seen in SPE 39428.

Ottawa sand
2 lb/ft2 water
Conductivity, md-ft 6,000 140°F






0 SMA-Treated
2,000 Untreated
6,000 Fiber-Based
Closure Stress, psi

Fig. 2—Effect of closure stress on the conductivity of sand packs, as seen in SPE 39428.


No Treatment

% Settled

SMA-Treated Sand,
0.12 gal/sk

40 Low-Guar Borate-Crosslinked Fluid

(25lb/Mgal), 70°F
20 20/40-Mesh Ottawa Sand

0 5 10 15 20 25 30 35
Time (min)

Fig. 3—The coating on SMA sand grains allows them to form a looser pack than an untreated sand pack, as seen in SPE 39428.

Fig. 4—Closeup of vugular regions in the vertical slot packed with SMA-treated sand, as seen in SPE 39428.