Ll

SPE 30498
Society of Petroleum Engineers

Formate Brines for Drilling and Completion: State of the Art

Siv K. Howard, Shell Research BV
SPE Member

Copyright 1995, Society of Petroleum Engineers, Inc.

agreement exists among the various environmental institu-
This paper was prepared for presentation at the SPE Annual Technical Conference & Ex· tions and governments about what is the most acceptable
hlbltion held In Dallas. U.S.A., 22·25 October, 1995.
solution - if any at all.
This paper was selected for presentation by an SPE Program Committee following review of The safest solution to this problem is to avoid it in the first
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 place. If cuttings are still to be disposed on the seabed, the
to correction by the author(s). The material, as presented, does not necessarily reflect any use of non-toxic water based drilling fluids, disappearing
position of the Society of Petroleum Engineers, Its officers. or members. Papers presented
at SPE meetings are subject to pUblication review by Editorial Comities of the Society of
from the cuttings pile during settling, is a safe way forward.
Petroleum Engineers. Permission to copy Is restricted to an abstract of not more than 300 However, not only environmental requirements have to be
words. Illustrations may not be copied. The abstract should contain conspicuous acknow·
ledgements of where and by whom the paper is presented. Write librarian, SPE, P.O. Box
considered when selecting a drilling fluid. Also technical re-
833836, Richardson, TX 75083·3836, U.S.A., fax 01·214·952·9435. quirements have to be fulfilled, such as temperature stability,
good hydraulics, shale stability, tolerance to contaminations,
material compatibility, reservoir compatibility, and recycling
possibilities. Conventional water based drilling fluids can
Abstract often not compete with the oil based or pseudo oil based sys-
Low solids drilling fluids based on formate brines (sodium, tems in most of these areas. Drilling fluids based on formate
potassium and caesium salts of formic acid) were originally brines (sodium, potassium, and caesium salts of formic acid),
designed to minimise frictional pressure losses in slim hole however, have been found to fulfil all of the above men-
drilling applications. In addition, their unique capability of tioned requirements. These fluids were first designed for use
stabilising polymers to high temperatures made them more as deep slim hole drilling fluids because of their temperature
temperature resistant than any other polymer based drilling stabilising effect on polymers and their high densities.
fluids. Subsequent work has shown that these brines, because Neither bentonite nor solid weight material are needed and
of their high densities and low corrosivity, are also ideal therefore these low solids drilling fluids have very good
completion and packer fluids. Formate brines have excellent rheological properties. Later, they were also shown to be
HSE profiles and they are compatible with reservoir fluids, environmentally more acceptable than other commonly used
good shale stabilisers, gas hydrate inhibitors, and scale dis- brine systems, shale stabilising, and compatible with
solvers. Also, a technique has been found for cost effective reservoir fluids and common drilling equipment materials. A
clean-up and recycling of formate based drilling fluids. technique has also been developed for recycling of these
The commercialisation and introduction of these fluids into brines.
the field (especially caesium formate) has taken a long time, The formate brines also have great potential for use as
due to high prices and few manufacturers. This situation is completion and packer fluids, as densities up to 2.3 sa can
now changing, as the number of manufacturers is increasing, be achieved without any solid weight material. The heaviest
and buy-back arrangements have been made available. Also, formate brine, caesium formate, is for the time being intended
a number of successful drilling and completion trials have as a replacement for the highly toxic and corrosive zinc
been carried out. bromide brine.

General Introduction Introduction to the Formate Brines

Recent changes in environmental legislations have driven the
industry away from oil based drilling fluids. The most Properties of Formate Brines. The formate salts of alkali
popular solution by far is the use of pseudo oil based muds, metals are very soluble in water and form brines of very high
which are based on synthetic hydrocarbons. However, these densities. The three salts that have been found useful for
systems are still not proven to be environmentally fully ac- drilling and completion fluids are sodium formate
ceptable, and their future is not certain. The main concern (NaCOOH), potassium formate (KCOOH), and caesium
about these fluids is biodegradation, both aerobic and an- formate monohydrate (CsCOOH-H20). Sodium formate is
aerobic. A whole range of synthetic base fluids now exists, the least soluble of the three and can reach a density of about
from those that degrade slowly and only aerobically to those 1.33 SO. Potassium formate is more soluble, with a
that are broken down very rapidly anaerobically. Common to maximum brine density of about 1.59 SO, and caesium
all is that they do affect the environment, either by staying on formate can reach as far as 2.3 SO [1]. The densities of the
the seabed for many years or by degrading rapidly with short three formate brines as a function of concentration (% weight
but more drastic effects on the environment. Major dis- and molar concentration) are shown in Fig 1. Although cae-
483
2 FORMATE BRINES FOR DRILLING AND COMPLETION: STATE OF THE ART 1995 SPE 30498

sium formate is the most soluble of the three on a weight ba- In the PARCOM recommended tests, all formates fall in the
si:;;, it is worthwhile noticing that the highest molar concen- categories "non-toxic" or "practically "no,n-toxic", whereas
tration can' be obtained with potassium formate. For certain zinc bromide is "highly toxic" or "moderately toxic".
applications, for example, temperature stabilisation of poly- Biodegradability. Sodium, potassium, and caesium formate
mers, the molar concentration of formate ions should be as have all been tested according to GECD 301 D (Closed
high as possible, and a optimised mixture of potassium and Bottle Test). Also, OECD 301 E (Modified OECD Screening
caesium formate is therefore recommended for densities Test) has been carried out on sodium and potassium formate.
between 1.59 and 2.3 SG. The results are listed in Table 4. It can be concluded that the
The viscosities of the three formate brines are shown in formates are readily biodegradable, and they all pass the
Fig. 2 as a function of brine density. As can be seen, apart "Time-window criterion"(substances that achieve 60% bio-
from highly concentrated potassium formate, the formate degradation must do so within 10 days of attaining the 10%
brines have relatively low viscosities,. which will not contrib- level).
ute much to the bulk viscosity of the drilling fluid.
All formate brines exhibit very low freezing points and Polymer Compatibility. The temperature stability of three
crystallisation temperatures [1]. True crystallisation tem- different polymers in' the formates and several other' com-
peratures have ,been measured for sodium, potassium,and monly used brine systems have been measured and com-
caesium formate, and for potassium acetate (according to pared:' The polymers that have, been investigated are the vis-
APIl3J, TCT), and are shown in Fig. 3 together with those cosifier xanthan and the two fluid loss polymers PAC (poly
for sodium and potassium chloride. anionic cellulose) and a cross-linked derivatised starch.
Temperature behaviour of polymers in formate brines.
HSEProfile. A large amount of environmental testing has Polymers behave differently when they are exposed to high
been carried ,out on the formate brines. This ,has shown that temperatures. Some polymers, typically biopolymers, have a
all formate brines have favourable toxicological profiles and so-called transition temperature (Fig. 4a). The transition
are biodegradable [2], Caesium formate is somewhat more temperature of a polymer is defined as the temperature at
toxic than the other formates,particularly to freshwater spe- which the polymer undergoes an order-disorder conforma-
cies,' This heavy brine is,hciwever, intended for use as a re- tional change. This conformational change is accompanied by
placement for zinc bromide that is significantly more toxic a massive loss in viscosity and an increase in the rate of hy-
than caesium formate. drolytic degradation by two orders of magnitude [3]. It has
Mammalian Toxicity. Due to the lack of literature data, been shown [1,3,4] that the transition temperature of these
caesium formate has been tested extensively for acute oral types of polymers can be modified by the influence of various
and dermal toxicity, eye irritation, skin sensitisation, mut- brines. Some ions, the. so-called water structure breakers
agenicity, and clastogenteity (Table 1). Table 2 lists the ("salting-in ions"), will lower the transition temperatures,
acute orat toxicity of the three formate salts· in comparison while others are water structure ,makers ("salting-out ions")
with several other salts (taken from the open literature). and will raise the transition temperature. The formate ion has
Mammalian toxicity testing on caesium formate shows that, been shown' to raise the transition temperature [1,3]. The
like sodium and potassium formate, it has a favourable toxi- 'commonly used viscosifier, '~anthan gum (a microbial poly-
cological profile. In particular, the lack of eye irritation and saccharide), is a typical example of a polymer with a transi-
skin sensitisation confers Significant advantages over the tion temperature.
alternative high-density brine system,zinc bromide. The transition, temperatures of xanthan gum in several
Ecotoxicity. The formates have been tested for aquatic commonly used brine systems have been measured and plot-
toxicity on several freshwater and marine species. Zinc ted' as function of' brine density (Fig.S). These· transition
bromide was also included in the testing, to evaluate caesium temperatures were found by measuring viscosity at increasing
formate as a replacement for this commonly used highly toxic temperatures in solutions of 5 gIL xanthan in various
brine system. Testing has also been carried out on potassium concentrations of brines. A Haake RV20 ,viscometer
chloride and potassium acetate for comparison. All test re- (D 100/300 sensor) was used at constant shear rate (500 s-I).
sults are listed in Table 3. The transition temperatures were determined as shown in Fig.
Aquatic toxicity testing on the marine and on the fresh- 4a. With the exception of calcium chloride, calcium bromide,
water species shows that sodium formate and in most cases and zinc bromide, the brines were tested up to concentrations
potassium formate oan be categorised as "non-toxic". Parallel close to saturation, Calcium chloride started gelling' at con-
testing on potassium chloride and potassium acetate shows centrations above about 1.4 SO. Neither calcium bromide nOr
that the potassium ion contributes more to the toxicity than zinc bromide could be measured to saturation because of
the· formate ion. Caesium formate can be categorised mainly quick polymer degradation at higher concentrations. As can
as, "non-toxic" and ,"practically non·toxic". Exceptions exist, be seen from the results, only sodium and potassium chloride
however, for freshwater algae where it is !'moderately toxic". have the ability to inorease the transition temperature in a
All tests carried out on zinc bromide confirm that it is "highly similar way as the formates. At higher densities, however, the
toxic" or "moderately toxic"'. formates are the only known brines that can increase the
transition temperature of xanthan.
484
SPE 30498 SIV K. HOWARD 3

Other biopolymers such as PAC and starch do not have a molecular weight polymer was used for the testing, and that a
transition temperature like xanthan. It has, however, been lower performance can be expected for a higher molecular
found that, in spite of this, their behaviour is strongly influ- weight polymer.
enced by the type and concentration of the brine. The viscos-
ity behaviour of these polymers typically decrease with in- Material Compatibility. The compatibility of the base fl~id
creasing temperature, as illustrated in Fig. 4b. This viscosity with common materials is of great importance both for
decrease is found to be reversible up to a specific tempera- drilling, completion, and packer fluids. Some testing has
ture. If this temperature is exceeded, a permanent viscosity therefore been carried out to assess the effect of formate both
loss is experienced, probably caused by polymer degradation. on metals and elastomers.
Viscosities of fresh water and an almost saturated potassium Corrosion. Existing brine systems such as sodium, potas-
formate brine, both with 15 gIL of a high molecular weight sium, calcium, and zinc chlorides and bromides are known to
PAC, have been measured. These measurements were done cause serious corrosion problems. For example, all halide
by measuring viscosity (at 500 s-l) while heating to various salts, especially the chlorides, are very corrosive to stainless
temperatures and cooling it down again. In the potassium steel as they cause both pitting and stress corrosion cracking
formate brine, most of the viscosity could be regained after (SCC). As pitting is also promoted in acidic solutions, cor-
heating to 200 a C, while in fresh water the viscosity was rosion is very difficult to avoid in zinc bromide because of
permanently lost already at 150 a C. A clear correlation the inevitably low natural pH of this brine. Formates, how-
between the viscosity and the abilities of the corresponding ever, are not halides, they have an easily adjustable pH, and
systems to control fluid loss was seen. Good fluid loss control are therefore likely to cause fewer corrosion problems.
could be obtained with PAC in both fresh water and Nevertheless some corrosion testing has been conducted in
potassium formate at temperatures up to the point where a order to confirm the low corrosivity of the formate brines.
permanent viscosity loss was experienced. Above this tem- One study was carried out to compare the corrosion on 4140
perature the systems could not control fluid loss. steel and Inconel 718 in caesium formate and zinc bromide
Long term stability of polymers in formate brines. In the brines. Seven day coupon tests were carried out at 180a C and
design of temperature stable drilling and completion fluids, 1450 psia. Nitrogen was used as pressurisation gas. Zinc
the transition temperature or the temperature at which the bromide was tested as delivered from the supplier without
polymer loses its viscosity is not as important as the tempera- addition of corrosion inhibitor. Both brines were tested with-
ture at which the polymer can survive over a longer period of out any pH adjustments (i.e. zinc bromide at pH - 1-2; cae-
time. Neither the measurement of the transition temperature sium formate at pH - 12). The results are shown in Table 5.
of xanthan, nor of the temperature of PAC or starch where a Another study was conducted on AISI 4145 steel in a typi~
permanent viscosity loss is experienced are suited for de- cal sodium formate drilling fluid at two pH values and in a
termining the temperature stability of drilling and completion concentrated potassium formate brine. Two tests were done.
fluids. These measurements have all been carried out by a The first was an electrochemical corrosion test. The results of
temperature scan of less than one hour, while in a typical this test are given in Table 6. As can be seen, the corrosion
drilling fluid, the polymers could be exposed to these high rates measured under deaerated conditions at 80 a C are
temperatures for a much longer period of time. significantly higher than the rest of the results (including
For testing ofdrilling fluids, 16 hours temperature stability aerated at 80°C and deaerated at 120°C). This was assumed
is normally required. Tests have therefore been carried out in to be associated with changes in the nature of the surface
order to determine the temperature that these three polymers scale on the steel, and should be further investigated. The
can withstand for 16 hours without any major loss in viscos- other test was a Stress Corrosion Cracking (SCC) test. As
ity (and/or ability to control fluid loss). Concentrations of 5 shown in Table 7, no SCC was observed under the conditions
gIL of xanthan, 10 gIL of PAC (ultra low molecular weight), examined. Earlier studies on a magnesium chloride based
and 20 gIL of starch were tested in several brine systems. The mud in the same lab confirmed that the formate brines are
solutions were aged for 16 hours at various temperatures, superior to the chloride brines in terms of preventing local-
cooled down again, and the viscosity (and/or fluid loss) was ised corrosion and SCC in AISI 4145 steel.
measured. The temperature at which the polymers, during Elastomers. Elastomer testing was carried out on a potas-
ageing, experienced a loss of viscosity or fluid loss of 50%, sium formate brinein an independent laboratory. A saturated
has been defined as the 16 hours stability temperature. The potassium formate brine (75% w/w) was selected for the
16 hours stability temperatures for xanthan, PAC, and starch testing as this is thought to be the worst case (highest
in various brines and fresh water are plotted in Fig. 6. The concentration of formate ions, (Fig. Ib». Thirteen different
brine systems were all close to saturation, apart from the cal- elastomers were tested for time periods of 7 days and 8
cium chloride brine, which experienced gelling at higher con- weeks. All elastomers were tested at 120°C and the most
centrations. As can be seen, the effect of the salts on the temperature stable ones also at 175°C. The results are sum-
temperature stability is very similar for xanthan, PAC, and marised in Table 8. The elastomers were received from the
starch, and the 16 hours stability temperature has been found supplier in the form of a-rings, something that made one of
to be about 25 a C lower than the transition temperature of the standard test procedures (hardness) difficult. An alter-
xanthan for the investigated brine systems. However, in the native test method was therefore used, which caused a large
case of PAC, it is important to be aware that an ultra low uncertainty in the results.
485
4 FORMATE BRINES FOR DRILLINO AND COMPLETION: STATE OF THE ART 1995 SPE 30498

Four FKM's (PIuoro-carbon ,"Viton") were also included Formate Based Drilling Fluids
in the tests. These elastomers are not recommended to use at
a pH above 10. As the pH in the test fluid was 10.5, this can Formulation of Drilling Fluids. Low solids drilling fluids
explain the rather poor performance of these four elastomers. can be formulated based on a sodium formate brine, a potas-
On the other hand, HNBR and NBRare both not rec- sium formate brine, or a mixture of the two, covering the
ommended for use in ZnBr2 or CaBr2 as cross-linking will range from 1.0 to 1.6 SO. As sodium formate is cheaper than
occur and the elastomers will become extremely hard. Al- potassium formate, to save cost a mixture· should always be
though no comparative study has been carried out on these made with as much sodium formate as possible. For the same
brines, it seems from the results that this phenomenon will reason, in the density range from 1.6 to 2.3 SO, a mixture of
not occur (or to a much lesser extent) in the formate brines. potassium and caesium formate is required with as much
potassium formate as possible. Alternatively, to save costs, a
Microbial Growth. One of the advantages of the formate drilling fluid in this range can be formulated on a pure
brines over other brines that are used for drilling and com- potassium formate brine weighted to required density with
pletion is that they are biodegradable (when diluted) and, solid weight material.
therefore, environmentally acceptable. However, when for- Dependent on the temperature, density, and rheology re-
mate brines are used as completion/packer fluids in low con- quirements, and the need to minimise impairment, various
centrations, it is important that the formate concentration is formulations can be recommended. For example, when a po-
high enough to inhibit bacterial growth. The formate levels tassium formate brine is being used at the standard concen-
that will inhibit bacterial growth have therefore been deter- tration, i.e., 1.57 SO, it is important to be aware that the
mined both for aerobic bacteria and for anaerobic sulphate- brine viscosity itself is significantly higher (10 cP) than it is
reducing bacteria in a diluted sodium formate brine, as this is at a concentration corresponding to, for example, 1.50 SO (5
the brine that will normally be used in low density applica- cP) (Fig. 2). When low rheology is a requirement, it might
tions. With both bacteria, the critical concentration was ap- therefore be advantageous to choose a slightly lower brine
proximately 1 M (6.8%w), which corresponds to a brine concentration, and use some additional weight material.
density of about 1.04 SO. As both these tests used "ideal Similarly, if the system will be exposed to low temperatures,
feeding" conditions, which in the field will rarely occur, these problems with crystallisation will occur at a much higher
tests have to be seen as reflecting· the worst case scenario. temperature (-15°C) in a 1.57 SO brine than for example in a
Therefore, in any formate brine system above 1.04 SO, 1.50 SO brine (-35°C) (Fig.3). On the other hand, if the fluid
problems with microbial growth are very unlikely to take will be exposed to high temperatures, the highest possible
place. concentration might be required to protect the polymers
(Figs. 5 and 6)
Formation Impairment. A large number of formation dam- Viscosifiers. All work that has been carried out so far
age mechanisms exist. These include interactions between shows that xanthan gum is superior in creating the desired
solids and fluids in the drilling/completion fluid with solids shear thinning rheology in the formate brines. As shown ear-
and fluids in the reservoir. No full study on the effect of the lier (Figs. 5 and 6), xanthan is also stable to high tempera-
formate brines/drilling fluids on reservoir rocks or fluids has tures in formate brines. The xanthan concerttration will vary
yet been carried out. depending on the required viscosity of the drilling fluid and
However, two common damage mechanisms can be the viscosity of the base brine. In a concentrated potassium
avoided/reduced by the use of formates. The first is the dam- formate brine, typically 0.5 ppb will be used, and in a more
age caused by solids invasion into the formation. It has been diluted brine, a concentration Closer to about 0.75 ppb might
shown [5] that significant damage is caused by improper be found more suitable.
removal of the mud cake solids deposited on the surface of Fluid loss polymers. As with xanthan, fluid loss polymers
the formation, By using a properly designed drilling fluid will be stable to higher temperatures in formate brines (Fig.
based on a high density brine, unnecessary solids required as 6). Most formulations so far have been formulated with PAC,
weight material can be completely avoided. Very thin, firm, but starch has also successfully been used in formate
and easily dissolvable filter cakes can be formed by the use of formulations. Several synthetic fluid loss polymers have been
only small amounts of calcium carbonate (or, if preferred, tested in the formate systems, but, especially at higher brine
with sized salt particles). concentrations, difficulties might occur in hydrating the
The second damage mechanism that can be omitted by the polymers in the base brine. Three types of PAC have been
use of formate brines (only monovalent ions) is the formation used to formulate formate drilling fluids: low molecular
of damaging precipitates that can occur when brines based on weight, extremely low molecular weight, and ultra low mo-
divalent cations come in contact with formation water con- lecular weight. Typically, a better temperature stability and a
taining sulphate or carbonate ions. Even the divalent alkaline- lower high shear rheology (PV) can be obtained with a lower
earth-metal salts of formic acid are highly soluble, and as a molecular weight PAC than with a higher molecular weight
result it is thought unlikely that any damaging precipitates PAC. However, for good filtration control, a higher molecu-
would be formed through contact between formate brines and lar weight PAC might be advantageous. Therefore, the ideal
normal formation· water components. PAC is often a combination of the higher and lower molecu-
lar weights. Typically, a combination of extremely low and
486
SPE 30498 SlV K. HOWARD 5
ultra low molecular weight will give a good compromise concentrations of PAC have been varied in order to obtain a
between rheology and fluid loss properties. A combination of good combination of rheology and fluid loss control.
two molecular weight PACs has also been shown to give a Another type of weight/filtercake material that has been
synergistic effect, in that fluid loss can be better controlled by used in the formate muds is sized salt. Drilling fluids have
the mixture than by either of the polymers alone. been formulated and used with sized sodium chloride as filter
Weightlfiltercake material. Although the formate drilling cake material. Closer analysis of this mud system, however,
fluids are often referred to as "solids free", this is not entirely showed that this salt dissolved and potassium chloride and/or
true, as a minimum amount of solids is required for the for- sodium formate were formed. A more suitable salt to be used
mation of a filtercake. The ideal filtercake material in a for- with the potassium formate mud is therefore potassium chlo-
mate drilling fluid is CaC03 (chalk), which can easily be re· ride. However, little testing has been carried out on this
moved with acid. An amount of about 10 to 20 ppb is suffi- system yet.
cient to create a thin and very efficient filtercake. Typical
formulations for drilling fluids based on three concentrations Hydraulics. The formate drilling fluids were originally de-
(1.50, 1.57, and 1.60 SO) of potassium formate brines are veloped as deep slim hole drilling fluids [6] with one of the
given in Table 9. The properties of these fluids at four differ- most important purposes being the minimisation of frictional
ent temperatures are shown in Table 10. It can be seen here pressure losses.
how the temperature stability increases with the brine con- Traditional drilling fluids utilise solids both for viscosity
centration. On the other hand, the plastic viscosity of the control (bentonite) and as a weight material (barite, hematite,
fluids becomes higher with the increased brine concentration etc.) High solids loading in a drilling fluid develops high
due to the increased brine viscosity. Also, it is important to frictional pressure losses during circulation in deep slim
keep in mind that the crystallisation temperature is increasing holes, resulting in low hydraulic power transmission to the
rapidly with the brine concentration, and in the 1.60 SO fluid, downhole mud motors, and the occurrence of high equivalent
the crystallisation temperature is close to room temperature. circulating densities (ECD's) and high swab/surge pressures
Ideally, in formate based drilling fluids, no weight material in the narrow annulus.
is required as all densities can be reached with the brines A typical formate based drilling fluid, on the other hand,
themselves. However, as caesium formate is still expensive, does not require any solid weight material, and as it is com-
high density drilling fluids based on a weighted potassium patible with polymeric viscosifiers (xanthan gum) up to high
formate brine are often required. Best suited as weight temperatures, the use of bentonite is also superfluous. The
material is calcium carbonate (chalk). In addition to being an combined effect of the highly shear thinning behaviour of the
excellent filtercake material, this acid soluble solid is also xanthan gum and the minimum solids loading, make formate
suited for weighting up the mud to a density of about 1.7 SO. based drilling fluids the best choice when low frictional pres-
Above this density, the plastic viscosity of the mud is becom- sure losses are required.
ing very high, and the fluid loss properties are diminishing. Drag reduction is also a phenomenon that should be Con-
An altemative weight material at this density is siderite sidered when slim hole drilling fluids are designed. Drag re-
(FeC03), which also has the advantage of being removable duction (or friction reduction) is the ability of certain addi-
with acid. Potassium formate drilling fluids can be formu- tives to reduce frictional pressure losses in turbulent flow.
lated with FeC03 up to densities of about 2.0 SO. One way of describing this phenomenon is as an extended
For densities above 2.0 SO, some heavier weight material laminarisation, that is, a delay in transition from laminar to
is required. Barite cannot be used in combination with po- turbulent flow [7]. Drag reduction cannot be predicted from
tassium formate because it dissolves (Table 12). Two types conventional API rheology measurements, and the phenome-
of weight material have been found to work at these condi- non is unrelated to shear thinning behaviour. Typical exam-
tions. The first is manganese tetraoxide (Mn304), which has ples of drag reducers are high molecular weight polymers. It
a density of 4.8 and an average particle size of about 0.5 mi- has been shown in a flow line with a diameter of 2", that vis-
cron. Although the properties of this mud system might be cosifying water with a polymer lowers the frictional pressure
quite good at 2.0 SO, the plastic viscosity becomes very high losses in the pipe with a factor of about three from what it
at 2.3 SO, and the fluid loss is not easily controlled. Due to would have been in pure water [8]. Similar viscosification
the fine particles, this weight material also has the with bentonite showed the opposite effect, the frictional pres-
disadvantage of causing dust problems on the rig and there sure losses increased with a factor of three. This effect was
are some indications that it might cause formation impair- shown to increase significantly with decreasing pipe diame-
ment. ter. Both these investigations [7,8] confirm the fact that low-
Hematite (Fe203) can be used for formulating drilling solids, polymer based drilling fluids, such as the formates, are
fluids of 2.3 SO, with good rheology and fluid loss proper- expected to drastically decrease the frictional pressure losses
ties. However, also this weight material has disadvantages in the turbulent flow regime. Some simple testing should,
such as abrasiveness and a higher tendency of sag. however, be carried out in order to obtain more exact data on
Table 11 shows typical properties of muds that have been these particular systems.
made up with the above mentioned weight materials. All
these muds are based on a standard 1.57 SO potassium for- Shale Stability. Drilling t1uids based on concentrated for-
mate brine viscosified with 0.5 ppb xanthan. The types and mate brines have also demonstrated a great potential as shale
487
6 FORMATE BRINES FOR DRILLING AND COMPLETION: STATE OF THE ART 1995 SPE 30498

drilling fluids [9,10). The shale stabilising properties of the 60°Cat the surface will dissolve 620 gIL magnesium chloride
formate brines are related toa combination of two different salt, The use of a saturated potassium chloride brine will not
mechanisms. improve the performance under these temperature conditions.
The first mechanism is the reduction of the hydraulic flow However, saturated sodium chloride and potassium formate
of fluid from the mud to the shale, caused by the relatively brines will both caLIse a drastic reduction of the solubility. In
high filtrate viscosities of the formate drilling fluids (Fig. 2). the sodiuffi'chloride brine, the solubility is reduced to 78 gIL,
In addition, in non-fractured low permeability (k$ 10 nD) and in the potassium formate brine, it is as low as 38 gIL.
shales, a second stabilising mechanism takes place as these Saturated sodium formate brine, was also shown to reduce
shales have been proven to act as selective "membrane" sys" the· solubility of magnesium chloride significantly, but re-
tems. The low water activity ofa concentrated brine system, precipitation problems made the measurements difficult. Of
generates an osmotic pressure that may stimulate backflow of the tested brines, potassium formate is therefore concluded to
shale pore water. This backflow will reduce the net flow of be the best suited for drilling of the Zechstein bischofite
fluid from the mud to the shale, resulting in slower hydration formation, with a performance twice as high as the second
and a ,slower rate of pore-pressure elevatiOn. If the osmotic best, which is. sodium chloride.
backflow is larger than the hydraulic flow into the shale and The effect of contamination of magnesium chloride on a
the far-field response is slow compared. to water transport potassium formate mud was also studied. 10% (wlw) of
from the shale to the mud, dt~hydra:tion and lowering of the MgCl2 salt was added to a potassium formate mud system.
pore+pressure in thenear.;wellbore zone will oocur. This will The sampl~ was hot rolled for 16 hours at 175°C, and the
increase, the formation strength and . 'the •near-wellbore standard mud properties were measured. The mud had' only
effective stresses, which both benefit wellbore stability. slightly increased its viscosity and the fluid loss properties
had improved, probably due to the salt crystals in the mud
Drilling througll salt formations. When drilling' through a (MgCI2 and Mg(COOH)z).
salt formation, for example, the magnesiu'mchloridecontain- Due to this low degree of dissolution of magnesium chlo-
ing Zechstein bischofite formation, three possible problems ride in the formates, the potential formation damage that can
can arise. The first is the actual dissolution· of the formation, be .caused by the divalent magnesium ions is also pro-
causing big wash-outs and thereby hole cleaning and cement- portionally reduced.
ing 'problems. The second problem is related .to the' drilling
fluid itself. As the salt dissolves in the drilling fluid, the Solids Contamination. A potassium I formate based drilling
properties of the drilling fluid might change. In the case of fluid, weighted to 2.3 SG with hematite has been tested for its
divalent ions such,as for example magnesium, this could for tolerance to solids contamination. To compare, a pseudo oil
example lead to problems related to cross~lirtking of poly- based mud and an oil based mud with the same densities were
mers. The third problem will arise if the same fluid is also requested from a mud company. All three fluids were
used for drilling the reservoir. If the drilling fluid has pioked weighted with hematite.' The three. drilling fluids were con-
up. magnesium ions from the' salt sectiort, formation impair- taminated with up to 80 ppb of solids. As solids, a mixture of
ment caused by precipitation of divalent ions can also be 25% bentonite and 75% OCMAHymodPrima clay was used.
ex:pected. The'high contamination level wa~ chosen to demonstrate the
The best solution to avoid these 'problems is to minimise extremely good solids tolerance of the formate drilling fluids.
the amount of salt that can dissolve in the drilling fluid by Also, previous experience has shown that a higher
drilling with a fluid based on a concentrated brine. Typically contamination level is required in the lab than in the field in
a pre-heated, oYer saturated magnesium chloride based mud order to achieve the same mudprbperties. This is probably
has been used to drill the Zechstein bischofite·formation. This dueto the difficulty of obtaining the very fine particle sizes
mu?; is heated at surface to keep the amount of additional . which are accumulated in the field muds.
magnesium chloride saltthat can be dissolved in it as low as ' Fromthe·results shown in Fig. 9, it can be seen that the
possible. This system has disadvantages such as high formate based system is only slightly affected by the con-
viscosity (PV) and it requires expensive surface equipment. .' tamination, and is far more resistant to' solids contamination
Other alternatives are to drill with other types of concentrated than both the oil based and the pseudo oil ,based drilling fluid.
brines, in which the concentration of magnesium chloride' that
can be dissolved is as lowaspossible. Clean-up and Recycling. A simple and cost effective' tech-'
In order to assess the abilities of potassium formate to drill nique has been developed to recover up to 80% of the for-
through a magnesium chloride formation, a small study was mate brine from a formate drilling fluid. The technique con-
conduoted on the solubility of magnesium chloride in various sists of a simple pH treatment. The method has'.been opti-
saturated brine systems: The brines tested were sodium misedin the 1abol'atOfY, and, tested on used drilling. fluids
chloride, potassium' chloride, and potassium formate, which from the field both in the laboratory and in a large scale test.
were all saturated at room temperature. For comparison, The method consists of the following 3 steps:
heated magnesium chloride, saturated at 60°C, was inclUded. l.Lowering the pH to dissolve part of the chalk that is al-
The results are shown in ifig. 7. From this, it can be Seen that ready in the system.
at a circLllating bottom hole temperature. of, for example, 2. Raising the pH by adding a base. This causes precipitation
100°C, a magnesium chloride drilling fluid that is saturated at of chalk and enmeshment of the drilled solids, in the pre-
488
SPE 30498 SIV K. HOWARD 7

cipitates. In addition a drastic viscosity reduction is ob- mate muds, but in most cases, the technique performs better
tained in the system due to precipitation of polymers. than the standard coagulationlflocculation technique.
3. Separating and readjusting the pH. Best separation will be
obtained by using a filter press (up to 80%), but evidence Other use of formate brines
from the laboratory is that a centrifuge or shale shaker are
also suitable. Scale dissolution. Concentrated alkali metal formate brines
Note that for certain muds, step 1 can be omitted. By using have been shown to have exceptional properties for dissolv-
the conjugated acid and base of the formate salt in the drilling ing very high levels of several alkaline-earth metal sulphate
fluid, no dilution will occur, and all added components are scales. The solubilities of these sulphates in formate brines
actually contributing to the make-up of a more concentrated are shown in Table. 12. As can be seen, concentrated potas-
formate brine. The remaining filtrate/centrate is normally sium formate is by far the most effective system, but also
clear, but might contain a small amount of solids, mainly sodium and caesium formate dissolve significantly higher
polymers and chalk. As these are natural components of the levels of sulphates than for example the halide brines (e.g.,
drilling fluid, the centrate can easily be used for make-up of chlorides, bromides).
new mud. For a complete clean-up, these solids can easily be This high solubility of sulphates in the concentrated for-
filtered out. mate brines has to be taken into consideration when design-
Evidence suggests that three independent mechanisms are ing drilling or completion fluids. For example use of barite
involved in the separation process: (BaS04) as a weight material should be avoided due to the
1. Precipitation of chalk. The technique of forming larger relatively high toxicity of the free barium ions (Ba2 +). On the
particles by precipitation, also referred to as "Sweep floc co- other hand, these brines, especially potassium formate brine,
agulation"[ll] is based on the principle that colloids may have great potential for use in scale removal treatments.
serve as condensation nuclei for precipitates or may become
enmeshed as the precipitates settle. Fig. 8a shows the particle
size distribution in the mud before and after treatment. Hydrate Inhibition. Brines are also known for being very
2. "Salting out" (precipitation) of polymers. By increasing the effective inhibitors for gas hydrates. They are so-called
pH, polymers are precipitated out of the solution. Polymers "inert" inhibitors, i.e., inhibitors that do not enter the gas hy-
are known to be salted out more easily at high concentrations drate phase, but influence the thermodynamic equilibrium
of water structure making ions [3], reducing the amount of through their effect on the activity of water. The ions in the
free water available for hydration. The effect of the pH on the brines interact with the dipoles of the water molecules with a
viscosity of xanthan in water and in a concentrated potassium much stronger bond than the van der Waals forces which
formate brine is shown in Fig. 8b. cause clustering around the apolar solute molecule [12]. This
3. Flocculation. Flocculation may occur depending on the clustering also causes a decrease in the solubility of potential
concentration and nature of polymers present in the drilling hydrate molecules in the water (a phenomenon known as
fluid and can therefore not easily be controlled. A flocculated "salting-out") as a secondary effect. Both of these effects
system has the advantage of producing a cleaner centrate but combine to require substantially more subcooling to over-
gives a lower recovery due to the high water content of the come the structural changes and cause hydrates to form. The
floes. effect of salts on the hydrate formation temperature can be
The amount of acid and base to be added should be opti- calculated, using the freezing point depression data for the
mised for each mud system. The precipitated particles in- brines. As can be seen from the crystallisation temperature
crease in size with the amount of acid added. If the formate curves in Figure 3, the formates, especially potassium for-
concentration is high enough, a separation can also be ob- mate, have a substantial effect on the freezing point. A po-
tained without the use of acid, but the recovered brine is nor- tassium formate brine of a density of 1.4 SO or higher lowers
mally clearer when acid is being used. Overdosing with acid the freezing point to about -60°C, while a sodium formate
results in the formation of carbon dioxide. A minimum brine of a density above 1.2 SO lowers it to about -25°C.
amount of base is required for good separation. Overdosing Based on the theory described by Sloan [12], there is a
with base has no consequences, except for the increased very simple and useful "rule of thumb" to predict the effect
amount of acid required to re-adjust the pH. The influence of on the hydrate formation temperature for any natural gas:
factors such as addition- and mixing time and temperature Hydrate equilibrium temp. (in brine) = Hydrate equilibrium
have been shown to be of minor importance. temp. (in water) - 0.8 X freezing point depression (brine).
The technique works on drilling fluids based on sodium, Applying this rule to the formates gives a gas hydrate for-
potassium, and caesium formate brines. Drilling fluids con- mation temperature in a potassium formate brine (> 1.4 SO)
taining chalk separate better, but a successful separation has which is 48°C lower than in fresh water. In sodium formate
been conducted on a typical used field mud that was based on (> 1.2 SO), the hydrate formation temperature is about 20°C
a concentrated potassium formate brine and using manganese lower than in fresh water.
tetraoxide as filtercake/weight material.
The technique has also been tested on other types of mud Field Experience
systems. The recovery is normally not as high as on the for- Within Shell, three successful field trials have been carried
out with sodium formate drilling fluids.
489
8 FORMATE BRINES FOR DRILLING AND COMPLETION: STATE OF THE ART 1995 SPE 30498

A -1.1 SG mud was used by NAM to drill a 4 1/8" hori- • stabilise polymers to higher temperatures than any other
zontal sidetrack in Berkel-S with a Coil Tubing Unit (CTU). known brine system
A brine/polymer based system was selected for this job to • stabilise shales
minimise frictional pressure losses in this small diameter • have a low corrosion potential
hole. The required density was very low, and could easily be • be compatible with most elastomers
reached by use of other brine systems (NaCl, KCl). However, • inhibit bacterial growth (at densities above 1.04 SG)
sodium formate was selected based on environmental • have a high tolerance for solids contamination
considerations. No drilling fluid related problems occurred • be compatible with salt containing formations
during the trial. A lost circulation situation made it desirable • be compatible with formation fluids
to change out the fluid to a less expensive KClIgypsum mud. • inhibit gas hydrate formation
This made it possible to perform a direct comparison between • be easily re-cyclable
the frictional pressure losses that were experienced with the • dissolve several sulphate scales
two fluids. The circulation pressure losses were found to be This makes the formate brines ideal completion or packer
noticeably lower in the sodium formate mud than in the fluids, good scale dissolvers, and an excellent base for drill-
KCI/gypsum mud. Lubricity was rather difficult to assess. ing fluids. Drilling fluids based on formate brines have ap-
Qualitative indications, however, such as easier control with plications in the following areas:
WOB and smoother tripping in-and-out in formate mud than • Deep slim hole drilling. The formates allow the combina-
in KCI/gypsum mud were observed. Production from this tion of the very low frictional pressure losses that can be
well has exceeded expectation. achieved in a brine/polymer based system with high tem-
Norske Shell has drilled two horizontal sidet.rack with perature stability and high hydrostatic pressures.
sodium formate drilling fluids of about 1.20 SG. • Shale drilling. The formates enhance stability through a
Brine/polymer/chalk systems were selected to avoid impair- combination of reduced hydraulic flow by filtrate viscosity
ment, as the filtercake is easily removed with acid. Sodium enhancement and stimulation of osmotic backflow of pore
formate was preferred over other brines (NaCI, KCI) for en- water.
vironmental reasons. No drilling fluid related problems were • Reservoir drilling. The formates require no weight mate-
encountered during drilling, and both wells have a high rial, giving the opportunity to design drilling fluids con-
productivity. taining only the desired type and amount of particles. In
Statoil has used a drilling fluid based on a 1.5 SG potas- addition, no damage due to precipitation of divalent ions
sium formate brine to drill the reservoir section of Gullfaks occurs.
C-30 [13]. The fluid was chosen as low ECD's, tight control • Salt drilling. Major wash-outs can be avoided due to the
of filtration, and low solids content were required. The fluid low solubility of most salts in the formate brines. Also, the
that was used was weighted to about 1.6-1.7 SG with man- formate muds have a great tolerance for salt contamination.
ganese tetraoxide. Manganese tetraoxide was selected as • "Gas hydrate drilling". Depending on type and concen-
weight material because it was assessed to be compatible tration of formate brine, the hydrate formation temperature
with an open hole completion with pre-packed screens. Be- Can be brought down by up to 48°C.
cause of the very small particle size of the manganese Difficulties related to too few manufacturers and high prices
tetraoxide (-0.5 micron), these solids were expected to be have so far hampered the introduction of these fluids into the
small enough to flow back through the screens. However, it is field. This situation is now in the process of changing, as the
also important to keep· in mind that such small solids will number of manufacturers is increasing and buy-back arrange-
invade much deeper into the formation than other commonly ments have been made available. The use of formates is
used solids, and the potential for formation impairment increasing steadily, and in addition to a number of comple-
should not be excluded. Also in this application, no fluid tions, some successful drilling applications have been com-
related drilling problems were encountered. During drilling pleted.
or when tripping no indications of hole-instability were ex-
perienced, and the cuttings (containing up to 40% clay) were AcknOWledgements
of better quality (more discrete) than had previously been I wish to thank John Downs (Forbrico, Ltd.) for the initial
seen with other water-based drilling fluids. Based on drilling work on formate and continued support.
performance, rig-time savings and well productivity, the field I would also like to thank Rob Houben for excellent ex-
trial was concluded to be successful. perimental assistance and Eric van Oort and Arie Baak for
their contributions on shale stability and scale dissolution.
Conclusions Thanks are also given to Gerard Bol for stimulating dis-
Formate brines have been extensively investigated for their cussions and reviewing and to NAM and NSEP for carrying
use as drilling and completion fluids. The formate brines out the field trials.
have been found to:
• cover the full density range between 1.0 to 2.:3 SG References
• have very low crystallisation temperatures 1. Downs, J.D.: "Formate brines: Novel drilling and completion
• have exceptionally good HSE profile fluids for demanding environments," paper SPE 25177 pre-
sented at the 1993 SPE International Symposium on Oilfield
Chemistry, New Orleans, March 2-5.
490
SPE 30498 SIV K. HOWARD 9

2. Downs, J.D., Killie, S., Whale, G. and Inglesfield, C.:

"Development of Environmentally Benign Formate"Based
Drilling and Completion Fluids," paper SPE 25143 presented at
the 1994 2nd International Conference on Health, Safety &
Environment, Indonesia, January 25-27.
3. Clarke-Sturman, A.J., Pedley, J.B., and Sturla, PL.: "InflUence
of anions on the properties of microbial polysaccharides in so-
lution," Int. J. Biol. Macromol. (December 1986) 8, 355.
4. Downs, J.D.: "High temperature stabilisation of xanthan in
drilling fluids by use of formate salts," Physical chemistry of
colloids and interfaces in oil production, Paris (1991) 197-202.
5. Francis, P.A., Eigner M.R.P., Patey, I.T.M., Spark, I.S.C.:
"Visualisation of Drilling"Induced Formation Damage Mecha-
nisms using Reservoir Conditions Core Flood Testing," paper
SPE 30088 presented at the 1995 European Formation Damage
Conference, The Hague, May 15-16.
6. Downs, J.D.,: "Formate Brines: New Solutions to Deep Slim-
Hole Drilling Fluid Design Problems," paper SPE 24973 pre-
sented at the 1992 European Petroleum Conference, Cannes,
Novemberl6-18.
7. Savins, J.O.: "Drag Reducing Additives Improve Drilling Fluid
Hydraulics," Oil & Gas Journal (March 1995),79.
8. Darley, H.C.H. and Hartfiel, A,: "Tests Show that Polymer
Fluids Cut Drill·Pipe Pressure Losses," The Oil and Gas Jour-
nal (June 1974),70.
9, Oort, E. van, Hale, AH., and Mody, F.K.,: "Critical Parameters
in Modelling the Chemical Aspects of Borehole Stability in
Shales and in Designing Improved Water"Based Shale Drilling
Fluids," paper SPE 28309 presented at the 1994 69th Annual
Technical Conference and Exhibition, New Orleans, September
25-28.
1O.0ort, E. van, Hale, A.H., and Mody, F.K.: "Manipulation of
Coupled Osmotic Flows for Stabilisation of Shales Exposed to
Water Based Drilling FlUids," paper SPE 30499 to be presented
at the 1995 SPE Annual Technical Conference & Exhibition,
Dallas, October 22-25.
I 1. Benefield, L.D., Judkins, J.F. and Weand B.L.: Process Chem-
istry for Water And Wastewater Treatment, Prentice-Hall Inc.,
Englewood Cliffs, NJ (1982) 216.
12. Sloan, RD. Jr,: Clathrate hydrates of natural gases, Marcel
Dekker Inc., New York 10016 (1990) 171,499.
13,Svendsen, 0., Toften, J,K., Marshall, D.S" and Hermansson,
C.L.: "Use of a Novel Drill-In/Completion Fluid Based on Po-
tassium Formate brine on the First Open Hole Completion in
the Gullfaks Field," paper SPE/IADC 29409 presented at the
1995 SPE/IADC Drilling Conference, Amsterdam, Feb. 28 -
March 2.

$1 Metric Conversion Factors

ppb x 2.853 E + 00::; kg/m 3
SO x 1.0 E + 03 ::; kg/m 3
cP x 1.0 E - 03 ::; Pa·s
Lb/lOO sq ft x 4.788 E" 01 ::; Pa
D x 9.869 E - 13 = m 2

491
 10 FORMATE BRINES FOR DRILLING AND COMPLETION: STATE OF THE ART 1995 SPE 30498

TABLE 1-MAMMALIAN TOXICITY OF CAESIUM FORMATE TABLE 4-AEROBIC BIODEGRADATION OF FORMATE BRINES
MONOHYDRATE (SCREENING TESTS\
Test Result Test NaCOOH KCOOH CSCOOH.H2O
Acute oral toxicity rat LDl;() 200 - 2,000 mg/kg
OECD 3010 102% 92% 83%
Acute dermal toxiCity rat LD"n >2,000 mg/kg (28 days) (16 mg/L) (18 mg/L) (45 mg/L)
Eye Irritation Non-Irritant .(grade 1) OECD 301C 90% (11.8 mg/L) 80% (11.7 mg/L)
Skin sensitisatlon No overt reaction (28 days) 88% (31.2 mg/L) 89% .(3D.4 mg/L)
Ames test (mutagenicity) Negative
Clastogenlclty Negative TABLE 5--COUP.ON TESTS ON 4140 STEeL AND INOCEL718
Brine Metal COrrosion rate
TABLE 2-ACUTE ORAL TOXICITY OF FORMATE SALTS AND [mm/yearl
RELATE.D SALTS Caesium formate 4140 0.033
Rat Mouse Unknown 2.27 SG Inconel718 0.Q33
Salt LDsO LDsO LD so
Zinc bromide 4140 0.263
[mg/kg] [mg/kg] [mg/kg]
2.27 SG Inconel718 0.033
SODIUM FORMATE 11 200
Sodium chloride 3000 4000
TABLE 6-ELECTROCHEMICAL CORROSION RATe TeSTS ON
Sodium bromide 3500 7000
4145 STEEL
Sodium carbonate 4090 6600
Test. sample .Condltlon Corrosion Comments
POTASSIUM FORMATE 5500 rate
Potassium acetate 3250 [mm/year]
Potassium chloride 2600 1500 Sodium formate 80°C aerated 0,1 - 0.5 good to fair
PotasSium carbonate 1870 2570 drilling fluid 80°C de-aerated negligible excellent
pH = 10.2
CAESIUM FORMATE "200 - 2 000
Sodium formate 20°C aerated 0.1 - 0.6 . good to fair
Caesium chloride 2600 2306 drilling fluid 80°0 aerated 0;6 - 1 fair
Caesium nitrate 2390 2300 pH =7.5 80°C de.aerated "1 - 3.5 poor
Caesium carbonate 2333 2170 120°C de-aerated negligible excellent
Caesium sUlphate 2830 3180 Potassium formate 80"C aerated 0.2~0.5 good to fair
brine 80°C de-aerated "1 - 10 poor to unaccept.
CALSIUM FORMATE 2650 1920
pH = 10 120°C aerated 0.5 - 1.5 fair to poor
Calcium chloride 1000 1940 120"C de-aerated negligible excellent
Calcium bromide 4068 " The relatively poor performance· at 80·C In 2 of lhedeallrated systems Is unexplainable.
Calcium carbonate 6450 7340 and possibly rela·ted to changlls In the surfacll scale of, the steel,

Zinc bromide 1000

TABLE ~ACUTE AQUATIC TOXICITY AND ALGISTATIC ACTIVITY

Species Test type Type species NaCOOH HCOOH KCI KCH 2 COOH CsCOOH·H?O ZnBr?
Brown Shrimp (Crangon crangon) LC"n 96 h s,w.shrlmp 6200 1300 1000 1800 91
Juvenile turbot (Scophthalmus max/mus) LO"n96 h s,w..f1sh 6100 1700 1800 1900 260
Larval turbot (Scophthalmus max/mus) LCl;() 48 h s.W. larval lish 1400 7.6
Rainbow trout (OncorhynchUs myklss) LC"n 96 h l,w,lish >10 000 3500 4100 5100 2100
Acartia tonsa (Dana) EC 50 48 h s.w, calanoid 3900 300 340 1.7
copepod
Daphnia magn<J (Straus) EC"n48 h l.w. crustacean >1000 540 35 6.4
Oyster embryo Crassostrea gigas EC"n 24 h s,w. oyster embryos 1200 0.33
"Skeletonema costatum Eh C",n72 h s,w. algae 1600 3400 1000 0.32
Scenedesmus sUbspicatus EbC SO 72 h l,w. algae >1000 >1000 1.6

Raphidoce/ls suboapitata EbC SO 72 h l.w. algae 2.0 0.41

• PARCOM recommended species s.w, = seawater I,w, = freshwater

492
 SPE 30498 SIV K. HOWARD 11

TABLE 7-STRESS CORROSION CRACKING (SCC) TESTS ON

4145 STEEL
Test sample Condition Comments
Sodium formate 150°C aerated No SCC, No pitting,
drilling fluid Negligible general corrosion
pH = 10.2
Sodium formate 80°C aerated No SCC, No pitting,
drilling fluid Negligible general corrosion
pH = 7.5 150°C de-aerated No SCC, No pitting,
Negligible general corrosion
Potassium for- 80°C aerated No SCC, No pilling,
mate brine Negligible general corrosion
pH = 10

TABLE 8--RESULTS OF ELASTOMER TESTING IN A 75% POTASSIUM FORMATE BRINE. TESTING HAS BEEN DONE AFTER
EXPOSURE TO TWO TEMPERATURES (120 AND 175°C) AT TWO EXPOSURE TIMES (7 DAYS AND 8 WEEKS)

120°C 175°C
"'Temp Thickness Mass **Hardness Thickness Mass **Hardness
limit change change change change change change
rOC] [%1 [%1 [%1 [%1 [%1 [%1
Elastomer 7 days 8wks. 7 days 8wks. 7 days 8wks 7 days 8wks. 7 days 8wks. 7 days 8wks.
Atlas 790 200 -1.2 +0.2 0 +0.7 +1.1 0 +0.4 +0.3 +1.2 +2.5 -1.1 -2.1
Atlas 7182B 200 -1.1 +0.1 0 -0.6 0 +1.1 -0.5 -0.7 +6.3 -0.1 +1.1 +1.1
EPDM 7204 270 -2.1 +0.1 0 -0.1 -10.6 -1.2 -0.4 -0.5 +0.3 +0.1 +5.7 +6.9 b)
Fluore171481 1) 230 +2.7 -0.1 0 +0.7 0 +1.1 +0.2 -6.2 -26.6 -12.0 -1.6 +2.6w)
Neoprene 7065 120 +0.1 +1.2 0 +0.8 -3.6 +2.5
HNBR 2269 2) 160 c2.2 +0.3 0 +1.4 -1.1 +1.1
Viton TC 1220-11 1) 175 +2.3 +0.8 0 0 -2.1 +1.1 +1.0 -2.9 -0.3 -7.0 -1.6 +2.1 w)
Vlton TC 1220-12 1) 175 -0.2 +0.8 0 +2.7 -3.1 +3.3 +4.0 +29.0 +15.9 +28.7 -1.1 b) -13.7
Nitrile 4058-902) 120 -2.1 +0.3 0 +2,3 +4.5 +10.5
Viton 9062-95 1) 175 +0.2 +0.3 0 +2.5 -2.1 +2.2 b) +0.6 +1.7 -1.4 -20.9 Ow) +1.1 w)
Carboxyl. HNBR 2311-90 2) 180 -1.4 +0.3 0 +3.7 +11.5 +4.4 -1.3 -7.2 +15.5 +11.6 0 +6.5
Carboxyl. NBR 2067-90 2) 120 -2.2 +0.4 0 +2.2 0 +3.5
EPDM 5778-90 160 -1.8 0 0 +1.0 +10.7 0
• Temperature limit as specified by supplier.
•• The hardness measurements were difficult to perform due to the shape of the elastomers, and the results should be taken as Indications rather than absolute values.
h) hard b) brittle w) weak
') FKM's (Fluorooarbon's) are very often attacked by alkali solutions (pH> 10), which oan explain the bad performance In the concentrated potassium formate brine (pH - 10.5).
2) HNBR and NBR cannot be used In CaBr o or ZnBr 0 as It hardens due to crosslinking. Based on these test results, such crosslinking seems not to occur In the formates (only monovalent ions).

TABLE 9--TYPICAL FORMULATION OF A POTASSIUM

FORMATE DRILLING FLUID
Compound Amount
TABLE 1Q-HPHT FLUID LOSS [mL] OF THREE UNWEIGHTED
K-formate brine (1.50, 1.57, or 1.60 SG) 350ml POTASSIUM FORMATE DRILLING FLUIDS.
'Xanthan 0.5 g Brine HPHT HPHT HPHT HPHT HPHT
'PAC (extra low molecular weight) 2g density (150°C) (150°C) (170°C) (175°C) (180°C)
'PAC: (low molecular weight) 1g [SG] BHR 'AHR 'AHR 'AHR 'AHR
Chalk (CaC0 3) 20 g 1.50 15 18 31 -- _.
KHC0 3 or K2C0 3 0.5 g 1.57 14 14 18 39 --
formic acid pH=10 1.60 13 14 18 20 33
'Xanthan and PAC polymers are mixed In with a high shear mixer for 30 minutes. 'Formulatlons as in Table 9. Hot rolling temperatures = Test temperatures.

493
12 FORMATE BRINES FOR DRILLING AND COMPLETION: STATE OF THE ART 1995 SPE 30498

TABLE 11-COMPOSITION AND PROPERTIES OF SEVERAL WEIGHTED POTASSIUM FORMATE DRILLING FLUIDS.
AI-L SAMPLES ARE HOT ROLLED AT 150'C.
Density Weight! Filter- Fluid Loss polymer (PAC) PV (aHR/AHR) YP (BHR/AHR) HPHT (150'C) Comments
ISG] oake [ppb] loP] Ilbs/100 tt 2] (BHR/AHR) [mL]
Ultra low Extr.low Low mol.
mol weight. mol. weight weight
1.60 CaCO" (fine) 2 1 26/26 17/15 1:l/13 Unwelghted
1.70 CaCOf.\ (line) - 2 54/41 34/16 24/80
1,70 CaCOf.\ (ooarse) 4 40/28 1917 19/23 Larger chalk particles reqUired
1,70 FQCO" 2 2 33/32 15/14 16/15
2.0 FeCOf.\ 1 1.5 53/63 17/17 38/33
2,3 Mn"O" - 1 1,5 63/89 75/25 80/>100 Bad flUid loss and rheology
2.3 Fe?O" . . 1 60/55 25/20 32/83 Formulated lor low rheology
2,3 Fe?O" - . 2 115/95 54/41 14/16 Formulated lor low fluid loss

TAt,lLE 12-S0LUBILITY OF ALKALINE MRTH SULP"lATES AFTER 16 HOURS AT 85°C IN SATURATED BRINES
Solutions Water Sodium salts Potassium salts Caesium salts

NaCOOH NaCI NaBr KCOOH KCI KBr CsCOOH, H2O CS2S04

(40%w) (26%w) (46%w) (75%w) (24%w) (34%w) (82%w) (60%w)
Dissolved !lOlids [mg/L]
6aS04 2 80 8 3 5800 16 14 600 2
Bartte 1 160 50 20 2600 80 50 700 8
SrS04 26 2700 300 250 110000 700 500 180000 350
CaS04·2aq 1200 2800 9000 9000 90000 15000 10000 80000 600
Scale (BaS04) nd 80 50 14 7000 50 30 800 1
Scale (SrSO ) nd nd nd nd 600 nd nd nd nd

~--- - -_ _ - - - - e,eOOH.H20
2.2 C$000H.H20 2,2
....... KeOOlI
2 ....... KCOOH --NaeOOH

e:~ 1.8- . - - . NaCOOH

] 1.6 1 - - - - - - - "
......
j ....
1.4 .... ....
1.2

I~:;';'---+-------"""'--......j 1#OO=;c..........._ _+__-+--_........_ - - t -_ _>------4

o 20 40 60 80 o 10 12 14
a) Format<: concentration ['lOw) b) Formate concentration (moI/L]

Fig. 1-Denslty of sodium, potassium, and oaeslum formste brines at 20'C.

a) Density as f\IOctlon of % (weight) salt. b) Density as function of molar concentration.

I
IZ I - Sodium formate 2S'C
I
I - Sodium formate 60'C
10 I
I
• • • • Potassium formate 2S·C
I
S; 8 I PotaSsium formate 60'C
~
I
i';> Caesium formate 25·C
'Iii I
0
u 6
.
I
~ I
- - - - Caesium formate 60·C

4
I
I
I

.
..
,"
2
. "~I
. ........".:.--.
-
a
1.2 1.4 1.6 1.8 2 2.2
Density [SOl

Fig. 2-Vlscoslty of sodium, pots$slum, and caesium formate brines as function of brine density.

494
SPE 30498 SlV K. HOWARD 13

30 . . . - - - - - - - - - - - - - - - - - - - / ,
20
- - - - Sodium chloride
10
Potassium chloride
U
o
0
~
I-<
-10 - .....- - Sodium formate
E-20
<I.l , , , 0' " Potassium acetate
lr -30
~ -40 • Potassium formate
-50 • Caesium formate
-60
-70 +----.---t---+----.---t---+--J
1.2 1.4 1.6 1.8 2 2.2
Brine density [SG]

Fig. 3--Crystallisation temperatures of the formate brines and potassium acetate brine as function of density. The crystallisation
temperatures of sodium and potassium chloride brines are plotted for comparison.

\ " .Transition temperature

-q- ... - . . -
I ~
Increasing temperature

Viscosity
Temperature thinning
Viscosity
/ Polymer
break-down

a) Temperature b) Temperature
Fig. 4-Viscosity of
a) a polymer with a typical transition temperature (for example xanthan) as function of temperature
b) a polymer without a transition temperature (for example PAC or starch) as function of temperature.

220 -
• KCOOH
200 -
-. - 0-·' KCl
180
• NaCOOH
160
U0
•• - 0- - - NaCl
'-' 140 , 1)-'-
S • CsCOOH
E-< ~ ~"'........
,"" •• " , - " 6 " " " " " "
-"-"-
Gelling" --<>
-"t:r.
120
100 -
- - ---............
;::- -x- _...... ...........
'6...
-- • -x, - • -<> - - NaBr

" • " c," "" CaCl2

80 '--- "6 , ..........

D~radation
,,
' 'x"" d'
l.>I~,gra ahon
- - 'X' " - ZnBr2

60 -" ~,- CaBr2

1 1.1 1.2 1.3 1.4 1.5 1.6 1.7
Density [SG]
• Water

Fig. 5-Transltlon temperature of xanthan as function of brine density for some commonly used salts.
495
14 FORMATE BRINES FOR DRILLING AND COMPLETION: STATE OF THE ART 1995 SPE 30498

Fresh water
(1.0 SO)
~ • Xanthan
800 - r - - - - - - - - - - - - - - , - - - - - ,
Calcium chloride
(1,39 SO) ~PAC 700
Sodium bromide
~ 600
(1,53 SO)
Sodium chloride
o Starch .~ 500
(1.19 SO)
Potassium chloride
(1.16 SO)
8 400
~ 300
Sodium formate
(1,32 SO) ~ 200
Pota."ium formate '
(1,59 SO) 100
o L.. . ~~_...=!~fE:t:~~±:~rf;N~a~C~I~J KCOOH

Sli lOll 120 1411 1611 ISO 2110 20 40 60 80 100 120

Temperature 1°C]

Fig. 6-16 hours stability temperatures for xanthan, PAC, and Fig.7-$Orubility of MgCI~. 6H 2 0 In various saturated brine
starch. For xanthan and PAC these temperatures are determined as systems. For a bottom hole circulating temperature of 100'C,
the ,temperature Where the fluid loses half of its viscosity. the solubility In a heated (60'C at surface) MgCI 2 brine is
For the starch, where the viscosity is too low to measure, 620g/L, while in a saturated potassium formate brine this is
API fluid loss measurements have been used for determination. reduced to about 38 giL.

---Water
• • • • Before treatment
12 ,.-~~~----- _ _ _ Sodium formate
411 , . - -__- - - - - - 1
- - Potassium formate
10 35

30
S
25
Viscosity [cP) 20
0/0 Particles 6

15
4
10 . .- - - - - - - _ - -...... -.....;..-_-1
2

0+---+---I----I----!---+---1
o 10 15 20 25 30
10 100 1000
b) Amount 10M KOH added [mUL)
a) Particle size I !lm]

Fig. 8-Recycllng of drilling fluids by use of the pH treatment technique.

a) Particle size distribution In a sodium formate drilling fluid before and after treatment.
b) Viscosity of 2.5 giL xanthan In fresh water, 44% (w/w) sodium formate, and In a 75% (w/w) pqtassium formate brine, as function of
amount 10M KOH added. Both solutions start at neutral pH before addition.

140 70,---------------,

120 - - - Oil Based Mud 60 • Oil Based Mud

- - -..... Pseudo Oil :E' _ •••••• Pseudo Oil

ii;'100 Sf 511
~ based Mud Based Mud
8
.~ SO ::: 40
l'J - .. -e-- , Potassium
.~
- •• - Potassium
Fo,nnte Based
d Formate Based
60

•• __ """"'" .-
.":a
0
Mud l30
'0
Mud
0:: 40 l"._~"_ ] 20
><
20 10

1I 0+---1----1----+---<
1I 211 411 611 SO o 20 40 60 SO
Solids contamnation [ppb] b) Solids contamination [ppb]
a)

Fig. 9-- Effect of solids contamination on three different drilling fluids. Measurements are done at 50'C after 16 hrs hot rolling at 175'C
a) Effect on plastic viscosity PV b)Effect on yield point VP.

496