SPE

Society of Petrolelm Engineers of AIME

SPE 13557

Alumina Gel Drilling Fluid for High~Temperature Use

by J. Block, W.R. Grace & Co.
SPE Member

Copyright 1985, Society of Petroleum Engineers

This paper was prf!sented at the International Symposium on Oilfield and Geothermal Chemistry held in Phoenix, ~rlzona, April9~11, 1985. T~e material
is subject to correction by the author. Permission to copy is restricted tp an abstract of not more than 300 words. Wnte SPE, P.O. Box 833836, Richardson,
Texas 75083-3836. Telex: 730989 SPE DAL.

ABSTRACT desirable for drilling fluids because the fluid

will be thin at the drill bit (high shear rates)
A new clay-free drilling fluid has been and sweep cuttings away, and thick in the annulus
developed for use at temperatures up to at least (low shear rates) and keep cuttings.~uspended.
177°C (350°F). The fluid consists of an alumina The increase in drilling rates obtained with a
gel (pseudoboehmite) viscosifier stabilized with shear-thinning mud syst~m has been described by
various bulky anions. These bulky anions include Walker ( 4, 5) •
'inorganics, such as sulfate, and organics, such as The rheological properties of the system
tartrate and citrate. Filtration control is are determined by the alumina gel component. This
provided by a compatible cross-linked poly(vinyl material is the pseudoboehmite form of alumina and
alcohol). The fluid is highly shear-thinning (the can be represented chemically by the formula
n value in the power law model is less than 0.3), AlO(OH)•xH 2 0. However, pseudoboehmite does not
has shale stabilizing properties (based on provide sufficient filtration control by itself.
laboratory tests with sodium bentonite and various The addition of standard filtration controllers
shales), is divalent ion tolerant, and can also such as starches and cellulose derivatives do not
tolerate moderate NaCl levels. . provide both filtration control and shear-thinning
~
rheology. It was therefore necessary to develop
The system has been successfully field our own compatible water-loss controller,
tested in a deep well (4,823 m (15,824 ft) total cross-linked poly (vinyl alcohol).
depth) in western Oklahoma. Good properties were
maintained through both the unweighted and Poly (vinyl alcohol) is a nonionic water
weighted portions of the well. Penetration rates soluble polymer which does not significantly
and rig utilization rates were compared with an change the rheological properties of the alumina
offset well drilled with a dispersed bentonite gel. However, the molecular weight of these
mud and the trends were favorable. If further polymers is not very high, and therefore
fieid tests confirm laboratory data, this new mud cross-linking was necessary to increase the
system should be able to de+ay (or eliminate in molecular weight and produce a more effective
some cases) the conversion to oil based muds. system.
INTRODUCTION It is believed that the alumina gel and
the cross-linked poly (vinyl alcohol) interact
A new clay-free drilling f~uid has been with each other via hydrogen bonding to form one
recently developed by Block et al. (1, 2). The species which is shear-thinning and provides
fluid is made up of two components, an alumina gel filtration control.
viscosifier, and a cross-linked poly (vinyl
alcohol) (PVA) filtration controller. The system, as initially formulated was
stable in laboratory tests up to 121°C. At higher
The fluid is shear-thinning temperatures, e.g., 150°C and above, the system
(pseudoplastic) and follows the power-law (3) thinned (decrease inK value) and lost much of its
rheological model between approximately 10 and shear thinning properties (increase inn value).
1,000 sec- 1 • The power-lawn value is about 0.1 - The filtration control properties also declined.
0.3 which indicates a high degree of X-ray diffraction of the heated product showed a
pseudoplastic i.ty. This type of rheology is highly ·substantial increase in crystallinity, and it was
theorized that this conversion from a non
crystalline gel to a crystalline material was the
References and illustrations at end of paper. cause of the high temperature failure.

85
2 ALUMINA GEL DRILLING FLUID FOR HIGH TEMPERATURE USE SPE 13557

Recent work on aluminum chemistry has All high temperature aging was carried
shown that the anions present during the gel out in a roller oven at the temperatures indicated
formation, or added after gel formation, have a . for 16 hours. After the aging period, the samples
significant effect on the gel properties. For were cooled and mixed for 5 minutes on a Hamilton
example, Nail, White and Hem (6) showed that the Beach mixer.
chloride ion retards crystal formation and
.inhibits particle growth. Hare and Vickerman (7) The laboratory pH measurements were made
·round similar results with sulfate ion, and with an Altex model 3500 digital pH meter. All
Kerkhof et al (8) with carbonate ion. Kwong and samples were adjusted to pH 9.5 with dilute NaOH
Huang found that citrate (9) and tannate (10) before and after the aging tests.
inhibited crystallization of aluminum hydroxide
and distorted the crystal latice. Violante and Field pH measurements ~ere made with an
Violante (11) showed that both tartrate and Altex Chem-Mate portable pH meter.
citrate were strongly adsorbed by alumina gels.
A Phillips model 12045 B/3 X-ray
Based on these findings, it was felt that Diffractometer was used to obtain x-ray
the alumina gel might be stabilized with various diffraction patterns.
anions, especially bulky anions, and increase the
upper temperature limit of the system. RESULTS AND DISCUSSION
POWER-LAW MODEL 121 °C Lab Tests
The power-law describes shear-thinning Typical lab test results for the AlO(OH)
liquids and is expressed mathematically as: cross-linked poly (vinyl alcohol) system are shown
in Table I. A typical rheological plot showing
"[ =
the shear thinning properties of the system is
shown in Figure 1. The data show that the system
where 1 the shear stress, is stable at 121°C (250 9 F) and is not appreciably
Y the shear rate, affected by moderate levels of NaCl, calcium and
K a constant often referred to magnesium. The tolerance to divalent ions is not
as the consistency surprising because poly (vinyl alcohol) is of
coefficient, and course, nonionic. At high NaCl levels the
n a constant, the power law cross-linked PVA begins to be salted out,
index. resulting in higher filtration values.
If n is unity, then the liquid is The clay stabilizing properties of the
Newtonian; if n is less than unity, then the system are demonstrated by the results shown in
liquid is shear-thinning. The lower the n value, Table II. It is felt that the cross-linked poly
the more shear-thinning the liquid. (vi0Yl alcohol), possibly in conjunction with the
alumina gel, forms a relatively impervious film
K is a viscosity-like term, and is equal around the shale, thus retarding water
to the viscosity when the shear rate is unity. penetration.
The constants K and n are readily FIELD TEST
obtained by plotting the log of the shear stress
versus the log of the shear rate. The slope is n
and the intercept at a shear rate of unity is K. A field test of the AlO(OH)-cross-linked
Because the system obeys the power-law model poly (vinyl alcohol) system was run in 1982 in
(between shear rates of approximately 10 and 1,000 Beckham County, OK. The mud was added at 1 ,372m
sec- 1 ) , all dat~ will be reported using the (4,500 ft) and was used to a total depth of 4,823
constants n and K. The Bingham plastic constants, m (15,824 ft), No significant problems were
PV and YP, were not obtained because the Bingham encountered and the mud properties are summarized
model assumes a Newtonian fluid with a yield in Table III.
point, and does not apply to a shear-thinning
fluid. The results show that the shear-thinning
character of the mud was maintained throughout the
EXPERIMENTAL well, unweighted and weighted. In addition, good
filtration control was also maintained.
All laboratory rheological measurements Penetration rate comparisons were made with an
were made with a Haake RV3 rheometer using an MV-I offset well drilled with a dispersed bentonite
rotor. All field measurements were made with a mud, and indications were that a 20% improvement
Baroid multi-speed rheometer. was obtained with the alumina gel system. Further
discussion of this field test is presented
The filtration measurements were made elsewhere (2).
using the API procedures described in API Bulletin
13B.

86
SPE 13557 JACOB BLOCK 3

177°C Test Results ACKN OWL EDGEMENT

Table IV summarizes lab data obtained The author gratefully acknowledges the
after hot rolling the AlO(OH)-poly (vinyl alcohol) work of L.J. Yeager, R.J. Davenport, K.M. O'Connor
system at 177°C (350°F) for 16 hours. It can be and D.D. MacLean who obtained most of the
seen that the rheological and filtration control laboratory and field data.
properties have deteriorated. The AlO(OH)
crystallite size, obtained by x-ray diffraction, REFERENCES
has increased appreciably indicating that the
alumina is now much more crystalline. It is felt 1. Block, J. "New Mud System Gets Field Trial,"
that this increase in crystallinity accounts for Oil & Gas J. p. 79 (January 11, 1982).
the loss of shear-thinning properties and quite
possibly the increase in filtration, due to a 2. Block, J., and Jewell, J.E. "Field Test of
break-down in the alumina gel-cross-linked poly Alumina Gel - Cross-Linked Poly (Vinyl
(vinyl alcohol) interaction product. Alcohol) Drilling Fluid," Transactions of the
1984 Drilling Technology Conference p. 95
177°C Test Results With Bulky Anions (1984).
The 177°C roller oven tests were repeated 3. Metzner, A.B. "Non-Newtonian Technology:
with the addition of some of the bulky anions Fluid Mechanics and Transfers", Advances in
known to stabilize alumina gels. The results Chemical Engineering , Academic Press, New
(Table V) show that sulfate, tartrate and citrate . York p. 87 (1956).
were effective in maintaining the rheological
properties of the alumina gel, whereas chlorides 4. Walker, R.E. "Mud Hydraulics - 3", Oil & Gas
were ineffective. Apparently, these bulky anions J., p. 59 (August 30, 1976).
attach (adsorbtion or ion exchange) to the surface
of the gel and sterically hinder crystallization. 5. Walker, R.E. "Mud Hydraulics - 4", Oil & Gas
This was confirmed by x-ray diffraction which J., p. 63 (September 13, 1976).
showed that the crystallite size of the
tartrate-containing samples was only 23 A (after 6. Nail, S.L., White, J.L., and Hem, S.L., J.
aging at 350°F for 16 hours) compared to 45 A for Pharm. Sci. (65) p. 1255 ( 1976).
the NaCl-containing samples.
7, Hare, A.S., and Vickerman, J.C., J, Chern.
Sulfate (at the higher level) and Soc., Faraday Trans. 1, (77) p. 1113 (1981).
tartrate also surprisingly maintained good
filtration control. It is believed that the good 8. Kirkhof, N.J., White, J.L., and Hem, S.L., J.
filtration control obtained with sulfate and Pharm. Sci. (66) p. 1533 (1977).
tartrate is a result of alumina gel stabilization,
i.e., if the alumina is maintained as a gel, then 9. Kwong, K.F., and Huang, P.M., Soil Sci. Am.
it will function better with the cross-linked poly J. (43) p. 1107 (1979).
(vinyl alcohol). Because the alumina was still
gelatanous, the lack of filtration control with 10. Kwong, K.F., and Huang, P.M., Geoderma (26)
citrate is surprising, However, since the K p.179(1981).
values were so high, it may be that less alumina
gel is required, and this may lead to improved 11. Violante, A., and Violante, P., Clays and
filtration. Clay Minerals (28) p. 425 (1980).
Additional tests were run with a weighted
system, with and without the addition of sodium
potassium tartrate (Rochelle salt). The results
(Table VI) again show that the tartrate stabilizes
the alumina gel-cross-linked PVA system.

CONCLUSIONS

Bulky anions such as tartrate and sulfate

are effective in stabilizing the alumina
gel-cross-linked poly (vinyl alcohol) system up to
at least 177°C (350°F). It is believed that these
anions function by sterically retarding the
crystallization of the alumina gel. If additional
field tests in high temperature wells confirm
laboratory data, then this new mud system should
be able to delay or eliminate the conversion to
oil-based muds.

87
 Table I Properties of the Alumina Gel- Cross-linked PVA System (121°C) Tabl<? II Shnle Stabilizing Properties o[ the Alumina-Gel -

Before Heating (d) After Heating (d) Cros~-1 i~~~-P~stem (a)

Ambient Ambient
Mud Addit~ve (%) (a) n ~ Filt.(c) _n_ ~ Filt. (c)
Final Pellet
None 0.21 5.? 7.8 0.18 15.0 11 .8 Pellet Composition Mud System Tested Volume (ml) (b)

CaCl 2 (1) 0.29 2.7 7. 0 0.19 8.2 7. 7 Sodium BcmtonltP H,O >80

CaCl 2 (?) 0.27 2.8 6.9 0.21 ~ .5 7.6 Sodium Bentonite Sodium bentonite {2.8%)
Polyanionic cellulose (0. 28%) lJO
CaC1 2 (3) 0_28 2. ~ 8.6 0.2ll 3-2 7.0
Sodium Bentonite AlO(OH) (1.2%), CLPVA (1 .61) 20
C:tC I, (',) o.~fi 3.2 ll.8 0.:.'3 11.8 li.6
Sodium Bentonite Sodium bentonite (?.8%), lJO
Polyanionic cellulose (0.28%)
r1gCl 2 ( 1) 0.21 ll.O 7.9 0.17 15.7 7.9 KCl (17.5%)

MgCl, (;>) o.;';> 3.8 8.9 0. Jli 15.:; 9.5 Sodium Bentonite AlO(DH) (1.2%), CLPVA (1.6%), 25
KCl (17.5%)
MgC1 2 ( 3) 0.21 ll.ll 9.2 0.16 8.6 10.2
Anahuac Shale H2 0 35

NaCl (3.5) 0.26 lj. ~ 6.ll 0.20 9:9 6.6 Anahuac Shale AlO(OH) (1.2%), CLPVA (1.6%) 12.5

NaCl (7) 0.22 ll.8 6.ll 0.23 ·r .6 7.2

(a) Pellets made by pressing 10 g of mater·ial at llO,OOO pounds
NaCl (10) 0.29 2.9 2ll .8 0.25 3.7 ll3 .8
pressure for 1 min. Pellet is 3.1 em in diameter and 0.6 em

(a) Initial mud contained 1 .~% AlO(OH), 1.5% CLPVA and 5.ll% high. Pellet volume is ll.5 cm 3 •

kaolin. All samples adjusted to pH 9.5, before and after heating. (b) Final pellet volumP determined af'ter static aging ror 16

Compositions i1.re weight percent based on final mud. hours at 250°F in test mud. Values are averages of at least

(b) lb-sec/100ft 2 two results.

(e) ml/30 min

(d) l?JOC for 16 hours in roller oven

:r·able £II Summary of Hud Properties - Field Test - Beckham County, OK Table IV Performance Test Results After Heating at 177°C (350°F}

lb-~ec
Before Heating After Heating
Mud wt Gels Ambient API AlO(OH) Ambient AlO( OH)
(
Depth <ill (lb/gal) 100 ft 2
{lb/1 OOft 2 ) (b) Filt. (ml/30min) Ambient Crystal. Filt. Crystal.
Mud Composition(a) n K Filt. (c) Size (A) _n_ ~ __ (c_)__ Size (A)
6,080 9.2 0.06 9.0 12/lll 1ll .3
(I)
AlO(OH) (2-~%),
~ 10' 3~8 9.0 0.13 9.0 9/12 1ll .8 CLPVA (1 .6%)
pH 9.5 0.17 9. 3 6.4 22 O.ll1 0.52 ll1 .6 li5
13,067 9.5 0.16 lj .9 5/10 7.0

~ 1ll ,886 1lJ.li 0.17 11 .8 11/18 4.5

(a) Weight percent based on final mud
\.J.J 15,82ll (a) 1ll.8 0.17 16.0 14/23 lj .8
\J'\ (b) lb-sec/100ft 2

\.11 (c) ml/30 min

(;t) Tot:1l depth (l>ot.tom hole t"rnperatur<' approximately lOS -
~ (d) 177°C for 16 hours in roller oven.
]lQDC).

(b) 10 second/ 10 minute

 Table V :erformance Test Re~ults after Heating at 177°C (350°F) -
Addi~~Dulky Anions (a)

After Heating

Ambient
Additive (%) n K (b) Filtration (c)

Sodium Chloride ( 1. 0) 0.23 1 .9 69.5

Sodium Chloride (2. 0) 0.40 0.6 31.2
Sodium Sulfate ( 1 '0) 0.19 8.3 67.0
Sodium Sulfate (;2.0) 0.28 14.8 6,0

Sodium Potassium Tartrate (0.5) o. 11 18.0 12.0

Sodium Potassi~m Tartrate ( 1. 0) 0.14 15.5 7.5

Sodium Citrate (0.5) 0.15 28.Q 32.5
Soqium Citrate ( 1 . 0) 0.?0 25.0 27.4

(a) Initial mud contains AlQ(OH) (2.4%), CLP.VA (1 .6%) at pH 9.5;

n = 0.17, K = 9.3, Ambient Filtration= 6.4,

(b) lb,...sec/lOOft 2

(c) ml/30 min

Table VI Effect of Tartrate in a Weighted System (a)

Before Heat in~ After Heatin~

I
Rochelle
Al.O(OH) CLPVA Salt Ambient Ambient
(%)(Q) (%)(b) ~ n --L Filt. (d) n K Filt. (d)

0.5 0.75 0.00 0.30 1?.0 2.q 0.66 0.27 3.8

0.18 1.3 0.23 0.13 7.4 3.0 0.20 3.6 3.2
0. 27 0.85 0.35 o. 16 11,0 12.2 0.08 ~3.0 13.8

(a) System weighted to 12.3 lb/gal with barite

(b) Al.l concentrations are weight percent based on final system.

(c) lb-sec/100ft 2

(d) ml/~0 min

SPE J 3 55.7
 ('.....
t.ri
L.l'"'t
1'<\
r--1

~
Cf)

100
~
AlO(OH) :- 1.5%
CLPVA : 1.6%
Kaolin :. 5.7%
pH- : 9.5

Shear
Stress 2
{lb/100ft ) .
1(}

n = 0 .. 24

K = 3.2

1 I I I f I f I I I I I ( I J- I I I I I I I I· I I I I 1- I

I 10 lOQ- 1,.000
. -1
Shear Raee (sec }

Fig. 1-Typical rheological plot for the arumina ge~/crosslinked PVA system.