SPE 132237

Scale Prediction for Oil and Gas Production

Amy T. Kan, Mason B. Tomson, Rice University

Copyright 2010, Society of Petroleum Engineers

This paper was prepared for presentation at the CPS/SPE International Oil & Gas Conference and Exhibition in China held in Beijing, China, 8–10 June 2010.

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

Abstract
Scale prevention is important to ensure continuous production from existing reserves that produce brine. Wells could be
abandoned prematurely due to poor management of scale and corrosion. The objective of this paper is to present an overview of
scale prediction and control and the current research at Rice University to solve these problems. In this paper, the challenges of
scale prediction at high temperature, pressure, and TDS are reviewed. An accurate model to predict pH, scale indices, density, and
inhibitor needs at these conditions are discussed. First, the various scale types found in oil and gas production and the condition
under which they form are discussed. Secondly, the relationship of pH, alkalinity, organic acids, carbonates and CO2 distribution
is discussed. Thirdly, the temperature, pressure, TDS dependence of the thermodynamic equilibrium constants and activity
coefficients is discussed. Lastly, the accuracy of a scale prediction program and its application is discussed. Based on a simple
propagation of error estimation, the overall estimated error for calcite SI is ± 0.1. The program has been validated with literature
solubility data for 6 minerals, pH data at 25 and 60 C, and density of high TDS solutions and up to 12.7 lb/gal weighting fluid.

Introduction
By all accounts, oil and natural gas will be the major components of global energy production for decades to come. Although
proved oil and gas reserves may be declining, oil and gas production can increase with the development of new technologies to
extract energy from unproved reserves and improved economics of production from both known reservoirs and marginal fields.
For example, oil and gas have gone to deeper and tighter formation with the advance of new production technologies. These new
developments in oil and gas production also bring challenges in scale prediction and inhibition, e.g., at high temperature (150-
200ºC), pressure (1,000-1,500 bars) and TDS (>300,000 mg/L) commonly experienced at these depths. Brines from different
zones are often mixed in the production tubing, subsea tieback or a common facility. Often the oil, gas, and brine composition in
each zone is quite different and consequently will cause significant scaling. Steam, seawater or alkaline surfactant flooding are
employed to mobilize residual or heavy oil. Many of these cost saving measures can result in significant scaling problems. Scale
prediction, control and treatment are vital to the success of these processes.
A major component of scale and corrosion management is the ability to accurately predict the brine chemistry, pH, and scaling
tendency of a production system. The typical variables in oil and gas fluid components are shown in Table 1. Due to the
complexity of the system, scale prediction is not as straight forward as one might imagine. In this paper, we attempted to address
the following: (1) the theoretical background of scale prediction and reliability; (2) the impact of brine salinities, composition, and
temperatures on pH, scale and corrosion; (3) the impact of organic acids; (4) the reliability of scale prediction at high temperature,
pressure, and in complex brine; (5) the impact of hydrate inhibitors on scale formation and prediction; and (6) the prediction of
scale inhibitor need. Experimental data to validate the model prediction of pH, saturation index (SI), and density calculations are
also examined and an estimate of error due to analysis is determined. ScaleSoftPitzerTM (SSP)1, a scale prediction software
developed by the authors, is used to illustrate the interrelationship of these parameters.

Common scale types found in oil and gas production.

Common oilfield scales can be classified into “pH independent” and “pH sensitive” scales. The scaling tendency of sulfates
(calcium sulfate, barite and celestite) and halite scales are not a strong function of brine pH. The carbonates (calcite, dolomite, and
siderite) and sulfide scales are acid soluble and their scaling tendencies are strongly influenced by the brine pH. For pH sensitive
scales, the scale prediction is more complicated since issues that control the brine pH also affect their scaling tendencies.
Barite. Barite is the mineral name for barium sulfate (BaSO4). When barite is formed rapidly (relative to geologic times) from
produced water it often contains up to one strontium ion (Sr2+) per approximately seven barium ions (i.e., approximately
2 SPE 132237

Ba0.88Sr0.12SO4). Unfortunately, radioactive, radium ions (226Ra2+) also substitute into the barite crystal lattice and are the major
source of what is called "naturally occurring radioactive materials," or NORM’s.
Barite scale formation can occur due to a pressure drop during production from a reservoir wherein the brine is saturated with
BaSO4, but barite scale formation is generally a consequence of mixing high sulfate waters (such as sea water) with formation
water during water flooding operations or a result of mixing brine from a high barium zone with brine from a high sulfate zone.
Essentially, all minerals in a reservoir are at equilibrium with the undisturbed connate, or natural, water. USGS did a survey of
log(barium) vs. log(sulfate) for a large group of wells2. The results showed that wells produce brine that is saturated with respect
to barite at downhole conditions, within experimental error. Therefore, it is suggested that when production takes place from a
single zone and the well is not flooded, the brine is probably at saturation with barite down hole. The saturation index of barite
increases as both the temperature and pressure decrease, but it can be demonstrated that the typical T and P drop, from a well,
producing from a single zone, would rarely be sufficient to cause barite scale to nucleate and grow in the tubing.
Generally though barite scale occurs when either brines from multiple zones or brines from different wells are mixed, or when
there is flooding of high sulfate water, as occurs with sea water flooding. When multiple zones are produced and it turns out that
one of the brines is high in sulfate and the other is high in barium and they mix at the well bottom, barite scale forms at the bottom
of the well, but generally delayed for a few feet till mixing is complete. In both of these cases the SI value for barite can be
exceptionally high and large amounts of scale can form quickly. This can even produce a supersaturation level that is not
controllable with inhibitors, regardless of the concentration of inhibitor. These are probably the most dangerous conditions, but
unfortunately they appear to occur rather commonly.
Flooding produces a very special and rather peculiar scenario of events for barite scale formation. Since the injected brine
flows over the formation solids for considerable distances (excess of a few feet is sufficient), the injected flow lines will establish
equilibrium with barite in the flow field. Therefore, the special problems of barite scale in flooding are just a transient problem
that occurs as the produced brine goes from mostly connate brine to mostly injected brine. Once the wells has experience injection
brine breakthrough the scale problem will likely disappear, or simply be the same as if the brine were saturated at downhole
conditions. The important issues of concern are generally related to the few feet around the injection well and the production well
where the conditions can change rapidly3. In subsea completions often methanol or MEG will be injected to prevent hydrates, and
these hydrate inhibitors can very easily cause the barite scale problem to be very severe.
Three precipitation regimes can be identified in a well, depending upon the saturation index (SI) of the produced brine (Figure
1). The exact critical SI values that divide these three regions are highly dependent upon the temperature. At lower temperatures,
the critical SI is much higher as indicated semiquantitatively in Figure 1.
A. If barite SI in the perforation region at the bottom of the well is greater than about ~1.5 (120 C) then there will be virtually
instantaneous precipitation in the mixing brines. This will reduce the SI value, but with a large mass of seed crystals produced
in the process. Then seeded growth precipitation will commence in the solution and on the tubing walls at the bottom on the
well. This condition should be rare, but it can be difficult to inhibit with chemical intervention.
B. If barite SI in the perforation region at the bottom of the well is greater than about 0.3 to 1.5 then, in the absence of inhibitors,
there will be nucleation and precipitation on the tubing walls at the bottom of the well. The kinetics of this scale growth on
the tubing walls will be quite rapid and scale is expected to form and build-up at the bottom of the tubing as often occurs. Due
to low barite solubility, the mass of barite scale formed per unit brine is typically small and may not be detected for a long
time. For this reason the pattern of mineral buildup is often such that by the time the brine reaches the surface, it is calculated
to be at equilibrium, and one is tempted to suspect that there is no problem. In these cases, the buildup of barite can
sometimes be calculated by taking the reservoir brine component values minus the surface values and summing for the barrels
of brine produced. In this case, either Ba2+ or SO42- concentration will be very low (the limiting reagent effect). Low
concentrations of Ba2+ (<1.0 mg/L) or SO42- (<15 mg/L) are difficult to measure. This probably accounts for scatter in field
data.
C. If barite SI in the perforation region at the bottom of the well is less than about 0.3 then nucleation will be delayed up the well.
This will also apply to wells wherein the brine is at equilibrium with barite in the formation, as can occur in the absence of
seawater flooding. As the brine flows, the pressure decreases and the temperature decreases, both of which increase the SI
value for barite. All else being equal, as the SI value increases the rate of nucleation and scale growth increase, but as the
temperature decreases the rates decrease quite rapidly. This can cause barite to nucleate and grow further up the tubing
toward the surface or this reduced temperature can prevent nucleation and scale buildup altogether. Either scale forms toward
the top of the tubing or not at all.
Calcium sulfates. Three calcium sulfate phases commonly occur as scales. They differ by the number of water molecules in
the crystal formula. Gypsum (CaSO4.2H2O) is the stable calcium sulfate phase at low temperatures (from room temperature up to
about 40° to 90°C). Above about 120°C, anhydrite (CaSO4) is the commonly reported phase. At intermediate temperatures
hemihydrate (CaSO4.1/2H2O), also called "plaster of Paris," is often reported in brines with high total dissolved solids (TDS)
content. The transition temperatures are strongly dependent upon the salt content of the brine and other specific chemical species.
SPE 132237 3

At any specific temperature, a higher concentration of salt will tend to favor the formation of a solid calcium sulfate phase with
fewer waters of hydration in the crystal formula. The details of how and which solution conditions affect the phase relationships
for the calcium sulfates are not known well enough to be predictive, but is certainly related to the water activity. In the presence of
methanol or MEG, as hydrate inhibitors, anhydrite was observed at much lower temperatures than expected due to the lowering
water activity in the presence of methanol or MEG4, 5.
The precipitation of calcium sulfate scale phases is often a consequence of the mixing of incompatible waters during water-
flooding. Also, calcium sulfate precipitation can result from a pressure drop when production is from a reservoir where the brine is
saturated with calcium sulfate, or from an increase in temperature during the processing of the brine on the surface (e.g., heater),
membrane filtration, steam flood, or when large quantities of thermodynamic hydrate inhibitors are used for hydrate control4-6.
Celestite. Celestite is the mineral name for SrSO4. The solubility of celestite salt is less studied than calcium or barium
sulfates7. There is a wide variability in literature reported celestite solubility8-11. Two issues appear to contribute to the
discrepancy: the presence of smallest sized particles that control the solubility, and the susceptibility of this mineral to surface
poisoning. Monnin and Galinier reviewed the solubility of celestite in complex solution matrices. They concluded that the
celestite solubility in NaCl is well behaved according to Pitzer model, but its solubility in sulfate brine is not. The unexpected
solubility is attributed to the effect of strong Sr2+-SO42- ion-pair interaction.
Halite. Halite is the mineral name for NaCl. The Pitzer theory was derived from the NaCl thermodynamic data measured by
Pitzer. Therefore, the scale prediction of halite should be accurate up to 300 C. Halite solubility is 6.03 molal or 311.3 g/L at 0
C. Therefore, halite does not typically form in less than about 350,000 mg/L TDS brine. When the produced brine is near
saturation with halite downhole, a small drop in temperature can cause halite to precipitate. Therefore, halite scale can be
problematic for a gas well with high TDS brine and coling as the pressure drops. Due to the high halite solubility, if precipitation
occurs, the mass of NaCl formed is large. The well tends to be plugged up fast when halite SI exceeds 0.0. The solubility of halite
in pure water and in electrolytes have been thoroughly studied 12. Note that halite solubility is reduced in the presence of
methanol. Halite scale can become a problem at much lower TDS values, circa and above 180,000 mg/L in the presence of
methanol4.
Silica. Silica and silicate chemistry is quite complicated and highly varied in rocks and reservoir brines. Silicate scale is rarely
a problem in “lower” temperature productions. “How low is low?” is not completely clear. As long as the temperature is below
300 to 400 F, it is unusual to have silicate scale formation. Since silicates typically have a rather simple monotonic increase in
solubility with temperature, the reverse is also true, i.e., nearly all produced waters are supersaturated with silicates as the T
decreases. Yet, we rarely see silicate precipitates. The silica monomer species in solution tend to slowly polymerized to larger
and larger polymers with time, but this reaction is much slower at lower temperatures. The temperature of transition to “silica
problems” seems to be near 400 to 500 F (204 to 260 C).
Sulfides. The sulfide scales include FeS, FeS2, PbS, CaS, and ZnS, to mention a few. "Sour" fields, which produce a
substantial amount of hydrogen sulfide (H2S) gas, tend to precipitate sulfide minerals. Many of the deeper and hotter gas fields
contain 25 percent by volume, or more, H2S. H2S is very corrosive to most steels and tends to form metal sulfide precipitates.
Several investigators have reported the formation of zinc sulfide (ZnS, or sphalerite) at depth, even in "sweet," or low sulfide
brines. Even though discrete formulas, such as ZnS and FeS, are used in discussing these sulfides, it should be emphasized that
these are only idealizations and that in most cases the actual solid compositions are not well known. Further, little is known about
what controls the conversion of acid soluble FeS to acid insoluble FeS2 (Pyritization). Considering the importance of these
materials to production, surprisingly little is known about the prediction and control of metal sulfide deposits. The research group
of the authors is developing the fundamental knowledge to better understand issues related to sulfide scale control.
Siderite. A special relationship has been reported to exist between iron carbonate (FeCO3) and iron oxide formation and
corrosion control as a function of temperature. At low temperatures FeCO3 is slow to form, but so is corrosion. As the temperature
increases to about 100°C, the rate of FeCO3 formation accelerates. Somewhere between 90° and 110°C, FeCO3 will begin to
decompose and form magnetite, Fe3O4, which is a better corrosion-prevention film. At higher temperatures the corrosion rates tend
to decrease to some lower limiting value. The relationship of calcite scale formation to corrosion control has already been
mentioned, but Sontheimer 13 has suggested that in general the important first phase to precipitate is FeCO3, not CaCO3 14, 15. It is
generally observed that in the absence of corrosion the ratio of Ca2+ to Fe2+ is about 100:1, due to the corresponding Ksp ratios, but
essentially pure FeCO3 has been reported in several fields around the world.
Many reports on the relationship of scale deposits to corrosion control have been published, including the formation of metal
oxides, phosphates, phosphonates, chromates and molybdates, to mention a few. There are numerous opportunities in the gas and
oil business for using such precipitates in corrosion control programs, especially in fields that produce substantial volumes (about
500 Bpd, or more) of brine. It is expected that more relationships between scale and corrosion will be tested as the need for high
efficiency corrosion control increases along with salt water production.
Dolomite To our knowledge there is no way to precipitate dolomite in the laboratory. It forms over thousands to millions of
years in the field. To date, no one has been able to precipitate dolomite in the lab at any SI. The SI is more related to dissolution
4 SPE 132237

of dolomite to equilibrium and to tell whether a formation is probably saturated with dolomite. We have looked at the dissolution
of dolomite in the Kuwait cores and find that dissolution is rapid and apparently to thermodynamic equilibrium with dolomite; yet,
when we try to reverse the process it appears that the brine remains supersaturated with respect to dolomite and probably what
happens is that calcite, or a magnesium containing calcite (called, magnesium calcite) forms and is essentially at equilibrium with
calcite SI ~ 0.0, but much more work is needed on this important problem.
Calcite. Calcite, essentially CaCO3, is the most common type of scale found. The crust of the earth contains about 12 percent
by weight of calcite and closely related carbonates. If the pressure drop from bottom hole to the surface is a factor of 5 to 10, or
greater, it is reasonable to expect that calcite scale will form; this is particularly true for sweet wells that are not being
waterflooded. Calcite scale formation is generally a consequence of the pressure drop that accompanies production. Simply put,
this pressure drop removes carbon dioxide (CO2) from solution and increases the solution pH and causes calcite to precipitate.
Also, there is a secondary consequence of the pressure drop; the inherent solubility of calcite in salt water decreases as the pressure
decreases. Both of these effects tend to cause calcite to precipitate during production. Calcite crystals are composed mostly of
calcium carbonate (CaCO3), but often contain up to 20 percent iron or magnesium carbonate, so that the average formula for
calcite might be represented as:

Ca0.8 to 1.0(Fe,Mg)0.2 to 0.0CO3

Although naturally occurring calcite such as Iceland spar is often essentially pure CaCO3, the scale formed from flowing brine
generally contains several mole-percent iron. This coprecipitated iron is often from corrosion products deeper in the well, but can
also be a result of naturally occurring siderite, FeCO3, or other materials in the reservoir. Most divalent metal ions, such as
manganese (Mn2+), lead (Pb2+) and strontium (Sr2+), can substitute into the calcite crystal lattice 16. Alsairi et al.17 studied the
coprecipitation of iron and calcium carbonate. Calcium ions have a strong influence in increasing the solubility of iron carbonate.
On the other hand, ferrous iron did not significantly affect the solubility of calcium carbonate. Although calcium carbonate has
higher solubility and also the saturation index of calcium carbonate was lower than that of ferrous iron carbonate, calcium
carbonate is the preferred phase to precipitate and the precipitate has a higher molar ratio of Ca/Fe than that in solution.
pH definition and the interpretation of pH value of brine pH determines the scaling tendency of the pH sensitive scales and
is of value to interpreting other brine chemistry, such as in scale, corrosion or emulsion formation. Yet, the procedures used to
calculate or to measure pH are often of questionable reliability in oil field applications. The pH is defined to be equal to the
negative of the logarithm (base ten) of the hydrogen ion activity on a molality scale (moles of H+ per 1.000 kg of water) with the
reference state activity coefficient equal to 1.000 at infinite dilution in water, i.e., pH   log 10 (a  )   log 10 ( m    ) .
H H H
There are at least three fundamental problems with the concept of pH and they can only be approximately addressed
thermodynamically 18: (1) It is impossible to separate the required activity coefficient product, (      ) ; (2) There will always
H Cl
be a junction potential, Ej(V) between the solution and the reference electrode and (3) the infinite dilution transfer potential from
one solution to another (e.g. pure water to methanol/water) cannot be determined. With that realization, a practical approach is
adapted to standardize the pH measurement by comparing the electromotive force (emf) of a sample to the emf of a standard
reference buffer measured with a glass electrode under well defined conditions. We use the operational definition of the pH as is
used by NIST which is pH  pH (S)  ( E X  E S ) ( 2.303RT F) , where pH is the value read on a pH-meter, pH(S) is the value
assigned to as standard reference buffer (e.g., pH(S) = 4.008 for 0.05 m potassium hydrogen phthalate buffer at 25 C,), ES is the
voltage in the standard reference buffer, EX is voltage in the unknown solution. The term, ( 2.303RT F) is the Nernst factor, where
F is the Faraday constant (= 96,484.6 coulombs/equivalent), R is the gas constant (= 8.31441 J(mol oK)-1) and T = Kelvin
temperature. Then, (2.303RT)/F = 0.05916 V at 298.15 oK. In short, pH is the value read on a pH-meter, when calibrated as
above. How consistent are the NBS standards within themselves and how closely do they correspond to the thermodynamic
concept of pH? The uncertainty of pH(S) appears to be about ±0.006 at 25 C and somewhat larger at other temperatures. The
uncertainty relative to the thermodynamic standard appears to be ±0.003 pH units.
Since pH is determined on a relative scale to a standard reference buffer, the pH measurement must be done under well defined
conditions in the field. Furthermore, any oxidation and precipitation reactions of the brine after it is collected can alter both the
total alkalinity of the brine and reduce the pH. We have proposed a sampling procedure by first lowering the temperature of the
brine with ice water while brine is still pressurized and then dropping the pressure of the brine under a blanket of CO2 gas. If a
sample is stored refrigerated, this will yield a chemically unaltered brine at full strength 19. Given the difficulties in obtaining
meaningful pH values for field brine samples, we have developed rigorously accurate equations to eliminate the need to use
measured pH in SI calculations. Essentially, we mathematically substitute the measured PCO2 for pH, as will be discussed later. To
be exact, if pH is the chosen parameter for scale prediction, the conditions we recommend are pH measured at STP and at
equilibrium with gas of composition exactly the same as the produced gas. This is approximately what would be measured at a
low pressure separator on the surface after the gas has cooled to room temperature and the pressure is reduced. Next, the measured
SPE 132237 5

pH should be corrected for both the salinity and cosolvent effects on the electrode junction potential in order to estimate the true
pH.
For any real solution and pH measurement there will always be a junction potential, Ej, due to the difference in the salt content
of the solution and the reference electrode filling solution, i.e., (E(V)  E o  (RT F) ln(m H  m Cl  H   Cl )  E j ) . The junction
potential, Ej, is a result of two processes. First, there is generally a concentration gradient from the solution to the reference
electrode filling solution, at the liquid junction and this concentration gradient causes a voltage difference. Secondly, the cations
and anions typically diffuse at different rates, given by their transference numbers or essentially their mobility (o/F). KCl is used
as a filling solution because the mobility of K+ and Cl- are very similar {73.52(K+) vs. 76.34(Cl-) S·cm2/eq.} Values of Ej can be
estimated by use of the Henderson equation18, 20. Also, Ej, can be measured using HCl and the values generally vary from 0.02 to
0.5 pH units, depending upon ionic strength. For dilute aqueous solutions, Ej is generally approximately constant, ± 1 mV, or less,
relative to the standard buffer solutions. The authors established a correlation equation to estimate the change in junction potential
as a function of salt concentration 5, 6, pHmeter reading = pHtrue - pHj = pHtrue – 0.13·(I)1/2, where pH j is the change in junction
potential in unit of pH and I is the ionic strength in molality. If the brine contains hydrate inhibitors, e.g., methanol, monoethylene
glycol, etc., a further complication arises related to the precise meaning of the reference state for pH. A detail discussion on the
cosolvent effects has been discussed previously 5, 6.

Effect of temperature, pressure, gas phase CO2 and H2S, and TDS on brine pH. The pH of water can be affected
by temperature, pressure, CO2 and H2S contents and TDS. The following calculation, based on SSP, is used to illustrate the effect
of T, P, CO2 in the gas phase and TDS on brine pH. The pH of pure water is equal to the negative logarithm of the square root of
the ionization constant of water, i.e., pH = - log10(Kw1/2) and at 25 C Kw = 10-14 and pH = 7.00. Figure 2 is the calculated pH of
pure water vs. T with no added salt, which is the predominant part of the T dependence of the pH. The salt effect on pH can be
derived from solving the equilibrium ( K w  a  a OH ) and the charge balance ([H+] = [OH-] equations simultaneously,
H
1/ 2
pH  - log 10 (a H  )   log 10 ( K w  (  H   OH  ) 1 2 ) That is, the pH differs from the pure water value, above, only by the square
root of the ratio of the activity coefficients of the hydrogen and the hydroxide ions. The activity coefficient of the hydroxide ion is
almost always less than that of the hydrogen ion. At 1.0 M NaCl, or 23,000 mg/L Na+ and 35,450 mg/L Cl-, and 25 C (from
SSP): H+ = 0.876 and OH- = 0.610 and pH = -½ log10(10-14·0.876/0.610) = 6.92. Due to the salt effect on pH electrode junction
potential, the pH reading will be that corrected, i.e., pHmeter reading = 6.92 – 0.13·(1.0 M)1/2 = 6.92 – 0.13 = 6.79. In the presence of
carbon dioxide, the pH will decrease and the effects of T and P can become large and cause the steel to corrode, i.e. the classical
“CO2 corrosion.” 21. In the presence of 1% CO2 in the gas and at room temperature and one atm pressure, the pHmeter reading = 4.68.
Increasing the temperature to 200 F, but still at 1 atm pressure, raises the pH from 4.68 to 4.96. This is due to the fact that at the
same PCO2 = 0.01 atm, the amount of CO2 that is pushed into solution is less than that at 77 F (25 C). There is a secondary
impact of the change in the ionization constants of carbonic acid and of water, but the primary effect is due to the decreased
solubility of CO2 in water at the higher temperature. Next, raising the P to 10,000 psi at the same T of 200 F would lower the pH
to 3.76. This is because the increased pressure greatly increases the PCO2 and the amount of CO2 that is pushed into the salt water.
Overall, these calculations illustrate the effect of temperature, pressure, TDS, and CO2 on pH. Water at neutral pH can become
corrosive downhole or scale can form at the surface from nonscaling calcite saturated brine.

Relationship of alkalinity, bicarbonate, sulfide, and organic acids

Bicarbonate is a crucial variable in carbonate scale prediction and it buffers the brine pH. Since bicarbonate concentration can
change with pressure down a well, it is not a conservative parameter to be used in scale prediction. The change with pressure is
strongly dependent upon the specific PCO2, the relative volume of water, gas and oil produced and the total carboxylic acids
present. A useful way to address this problem is by using the concept of alkalinity. The advantage of this concept derives from the
fact that although the actual concentration of the weak-acid anions and pH change with pressure and PCO2, the alkalinity is
constant, as long as no net acids or bases are added. Once the total alkalinity and organic acids concentrations are determined by
the procedure discussed below, the theoretically corrected bicarbonate concentration and/or pH at any given operation temperature
and pressure can be calculated.
The origin of alkalinity (Alk) comes from reactions of acids from digenesis and metabolism with minerals in the formation,
such as CaCO3, Fe(OH)2, MgSiO3, Na2SiO3, etc., to form weak acid buffer solutions, as shown by the following reaction scheme:

(H 2CO3  HAc  H 2S  HCl)aq min

erals

 H 2CO3  HAc  H 2S  H 2O  NaCl
HCO3- Ac- HS- OH - SiO 2
2- 2- 
CO 3 S H
6 SPE 132237

It is essentially a statement of the charge balance for a solution [Alk =Na+ + K+ + 2Ca2+ + 2Mg2+… - Cl- - 2SO42- …], i.e., sum
of strong base cations minus the sum of strong acid anions. For a system containing only carbonate, sulfide, and carboxylic acid
species, alkalinity is defined rigorously by the following expression:

Alk  [HCO3-] + 2[CO32-] + [HS-] + 2[S2-] + [Ac-] + [OH-] – [H+] 1

The twos in front of [CO32-] and [S2-] express the fact that one mole of each ion is capable of neutralizing two moles of acid, H+.
The term [Ac-], acetate ion concentration, is the sum of all the concentrations of ionized carboxylic acids in the brine, e.g.,
{[Acetate-] + [Propionate-] + [Butyrate-] + etc.} If the brine pH is above about 10 to 11 pH (which is very unusual, but possible),
then it is necessary to add contributions to alkalinity from boric acid (H3BO3) and silicic acid (H4SiO4) and ammonia (NH3). For
the effect of phosphates and polymers to be significant the concentrations would have to be more than 10 to 100 mg/L, active.
There is considerable debate about the numerical value of the second ionization constant of H2S; values range from K2,H2S = 10-13
to 10-18. In either case, the concentration of S2- is essentially never of importance in the calculation of alkalinity and will be
omitted in this and all further calculations, for illustration purposes.
The predominant Alk term is often [HCO3-]; this is especially true in the absence of substantial H2S, organic acids and when
the pH is less than about 8 to 8.5 pH (the pH range wherein CO32- becomes significant). However, oil field brines often contain
aliphatic carboxylic acids, up to 5000 mg/L. The highest concentrations of carboxylates tend to be in waters from reservoirs at
temperatures of 80 to 100 ºC22-24. When brine contains a substantial amount of organic acids, the conventional assumption that
bicarbonate concentration is equal to alkalinity is no longer valid. This is illustrated by the following example. At 200 F, 1,000 psi,
and 1 M TDS  60,000 mg/L; the conditional equilibrium constants25 are K 1H 2CO 3'  10 5.94 , K CO
H
2'
= 10-2.04 atm/M, K 'HAc = 10-4.48. In
this specific example, the alkalinity is equal to the sum of bicarbonate and acetate,
Alkalinity  [HCO 3 ]  [AC ]  (10
- - 7.98
 PCO ) a H  THAc  10

-4.48
 (a H  10 )
-4.48
.
2

Since the Alkalinity and the THAc are both constants and the PCO2 changes due to the flow of the well, the pH varies with the
PCO2 When the PCO2 is low, toward the surface of the well, the pH will be higher or a will be lower and the second term on the
H
-
right hand side will become THAc1.00 ( = [AC ]), i.e., all the acetic acid will be ionized and [HCO 3- ] will be less than it otherwise
would be if the Ac- concentration had been ignored. Then, as the gas is compressed down the well the CO2 in gas phase will be
forced back into the brine which will lower the pH making a   K HAc at pH = 4.48. At this pH only ½ of the THAc ( = [AC-])
H

will be ionized and the actual amount of [ HCO 3- ] will increase which will increase the SI(calcite) quadratically (see calcite SI eq
below).
In Figure 3 are plotted the HCO3- concentration and brine pH as a function of PCO2 and acetic acid concentration in the brine.
As shown in Figure 3b (top line), the bicarbonate concentration is identical between 0.01 to 100 psi PCO2 in the absence of organic
acids. However, the bicarbonate concentration changed considerably between 0.01 to 100 psi PCO2 when the organic acids are >
25% of total alkalinity. For example, with Alkalinity = 610 mg/L = 0.01 equiv./l and THAc = 0.009 M with 5 % CO2 (STP), the
pHSTP  6.00. At this point, [HCO 3- ] 0.00112 M and [ Ac - ]  0.00888 M. In the well at one hundred times this pressure, PCO2, gas =
5 atm: [HCO 3- ] 0.00370 M and [Ac - ]  0.00630 M at 4.85 pH. This shows that the [HCO 3- ] has increased by 330 %. This will
change the SI by log(3.3)2 = 1.04 SI units vs. what would have been calculated without correcting for the acetic acid effect.

Simultaneous alkalinity and weak acids determination. Reliable methods to accurately measure true alkalinity are
scarce, especially when organic weak acids are present that smear the titration endpoint of bicarbonate. Oxidation and precipitation
reaction of the brine can also alter the true total alkalinity of the brine. Similarly, measuring the organic acid compositions
required either a GC or LC procedure and the method is tedious and expensive. We have developed a rigorous theoretical solution
of this problem and proposed a corresponding experimental method to accurately measure both bicarbonate and weak acids
simultaneously with minimal experimental difficulty and no costly instrumentation 26 27. The new titration method is based upon
simultaneous analysis of the titration curve determined at fixed PCO2 and emphasizes the titration shape (profile) instead of the
endpoint inflection, as is done presently using a recommended procedures and apparatus (Figure 4). When the brine is purged with
CO2 gas, each mole of HS- will be replaced by one mole of CO2 gas to form one mole of bicarbonate with no net change in total
alkalinity, i.e., HS- + CO2, gas  → HCO3- + H2S gas .
Typically, [CO 32 ] and [OH  ] concentrations are negligibly low. The charge balance equation of alkalinity titration can be
expressed in the following form after substituting the equilibrium constants and activity coefficients:
SPE 132237 7

K1H 2 CO3  K CO
H
2
 PCO 2 ,gas   g , CO 2 TAc
Alk0  HCl( M)   10pH 
 HCO   10 pH   Ac  
3   1 2
K  
 HAc HAc,aq 
TAc
 a  10pH 
b  10 pH  1

Where Alk0 is the initial brine alkalinity (equiv.-Liter-1) and each mole of added HCl exactly neutralizes one equivalent per liter of
alkalinity. Note that the activity coefficients and equilibrium constants in Eq. 3 are constant for a specific brine and therefore, the
product of these constants in the first and second terms of Eq. 3 can be regarded as two adjustable constants, a and b. The titration
data (HCl added versus pH) can be fitted to Eq. 3 by non-linear least squares procedure by minimizing the sum of the squares of
the difference between the calculated and the experimental pH, subject to the constraints of the function, Smin : SUM(pHcalc, i -
pHexp, i) 2 being fitted to determine the values (Alk0, TAc, a and b). A software (“BCC Total Alkalinity, Carboxylic Acid
Simultaneous Titration at Constant PCO2©”), using Excel Solver program, has been developed for the calculation. This method
has been tested on three oilfield brines by four independent laboratories of oil and service companies. Excellent agreement was
observed from the blind round robin test results (See Table 2).
Typical reactions that may alter the initial brine alkalinity during sample storage include oxidation and precipitation, as
illustrated in the following reactions.
Fe2  0.25O  2.5H 2O  Fe(OH)3,solid  2H  (Oxidation)
2

Ca 2  2HCO3  CaCO3, solid  H 2 CO 3, aq (Precipitation)

Fe 2  2HCO 3  FeCO3, solid  H 2 CO 3, aq (Precipitation)
Fe2   2HS  FeS  H 2Saq (Precipitation)
In each of the reactions, two equivalents of bicarbonate alkalinity will be consumed.
The following correction procedures are recommended for brine that may be altered during brine storage.
1. Filter brine with a 0.45 mm filter.
2. Analyze total alkalinity and organic acid concentration according to recommended procedure.
3. Acidify and measure Fe and Ca content on the filter and calculate the equivalent Fe and Ca concentration in the original
brine that form precipitates.
4. Correct total alkalinity by the following equation.
Alk 0 (mg/L)  Alk titration (mg/L)  2.184  [ Fe ppt ] (mg/L)  3.050  [Ca ppt ](mg / L)
A series of experiments were done to test the feasibility of this equation for correcting the effect of precipitation and Fe(II)
oxidation on alkalinity measurements. In Figure 5 is plotted the decline in measured alkalinity where a synthetic brine containing 1
M NaCl, Fe(NH4)2(SO4)2) (154 mg/L as Fe2+ and NaHCO3 ( 471 mg/L as HCO3) were stored in two different sampling bottles.
PVC bottles are significantly better than HDPE bottles in terms of an oxygen barrier. The oxygen permeabilities are 4 and 185 cc.-
mil/100m2-24 hrs-atm for PVC and HDPE, respectively. Over a ten days period, the alkalinity declined by only about 8% when the
brine was stored in the PVC bottle with no headspace. The alkalinity declined about 35% when the same brine was stored in the
HDPE bottle. The measured alkalinity in the HDPE bottle declined steadily over time due to gradual Fe(II) oxidation and
precipitation. However, the loss of alkalinity due to oxidation and precipitation can be quantitatively corrected. There is no
statistical difference among the corrected alkalinities and the total alkalinity initially present in the brine, Alk0.
Lastly, the various terminologies used in typical brine analysis can also add to confusion. In SSP, the total alkalinity is
expressed as bicarbonate equivalents in unit of mg/L HCO3-. Below is a list relating the various terminologies.
1. If alkalinity or bicarbonate concentration is given in unit of mg/L of CaCO3, then total alkalinity (mg/L HCO3-) =
1.22·(mg/L CaCO3)
2. If both mg/L HCO3- and mg/L CO32- are given, then total alkalinity (mg/L HCO3-) =2.303·(mg/L CO32-) + (mg/L HCO3)
3. If you are given a value for phenolphthalein alkalinity as mg/L CaCO3 in addition to bicarbonate alkalinity as mg/L
CaCO3, then total alkalinity (mg/L HCO3-) = 1.22·( phenolphthalein alkalinity + bicarbonate alkalinity, mg/L CaCO3).
Relationship of pH, PCO2, Alkalinity, and SI.
In SSP, the total moles of each gas component present in the gas, oil, and brine are used at all subsequent temperature and
pressure combinations to calculate pH as well as all other solution species at the new temperatures and pressures. Only two out of
the three key parameters, i.e., PCO2, alkalinity, or pH, are needed to calculate the total mass of CO2. Once the pH and PCO2 are
measured or calculated, the total moles of CO2 at specific volumes of gas, oil, and water, temperature and pressure can be
calculated from Eqs. 3 and 4.
8 SPE 132237

- 2-
nTCO 2  n CO
g
2
 n CO
o
2
 n CO
w
2
 n HCO
w
3
 n CO
w
3
 PCO 2  (Vg / ZRT  a 2 ) 3
 1 K 1H 2 CO 3 K 1H 2CO 3 K HCO 3 
a 2   CO 2,g (K CO 2
 m /   K CO 2
 m 
w   2  4
o/g o CO 2 ,o H
 aH  aH 2
 
 CO 2,aq HCO 3 CO 3 
Where Vg is the gas volume, Z is gas compressibility, and ’s are the activity coefficients. Similarly, the total moles of H2S and
hydrocarbon can be calculated. Using the principle of mass balance, the pH and PCO2, PH2S can be calculated from nTCO2 and
nTH2S at other temperatures and pressures.
The following illustration is to show how changes in volume of water or of gas will affect the calculated downhole pH by
changing the relative fraction of gas phase carbon dioxide with pressure. At STP condition with 0.01 atm PCO2, and the g/o/w
ratio of 2000 scf/1 STB/10 STB, the solution pH is 7.25 and dissolved CO2 concentration ([CO2,aq]) is 2.8·10-4 m. When the
system is pressurized to 400 atm and 25 C, PCO2 become 0.44 atm, more CO2 dissolved into the aqueous phase ([CO2,aq]=0.0120
m) and the solution become more acidic (pH 5.76). At the bubble point (513 atm) and 25 C, the gas phase disappears and all CO2
is dissolved in the oil and brine phases, the aqueous phase CO2 concentration increases to 0.0123 m and the pH drops to 5.62.
Scale Prediction Scale formation tendency can be represented by the thermodynamic-based saturation index, SI, and is defined
in the following for calcite:
CaCO 3,solid  Ca 2  CO 32
 Ion Activity Product   a Ca 2  a CO 32  
SI(Calcite)  Log10    log10  Calcite  5
 K sp (T, P)  
 K sp (T, P) 

By repeated substitution it is possible to express Eq.5 as follows:

CaCO 3,solid  H 2 CO 3, aq  Ca 2   2HCO 3
 a Ca 2  a 2HCO   a a2 
  log  Ca 2  HCO 3   log  K 2 
K HCO 3 HCO 3 6
SI  log 10  3 2
(T, P ) 
 a CO , aq K K H 2 CO 3  10
 a CO ,aq  10
 K K H 2 CO 3 
 2 sp 1   2   sp 1 
 a Ca 2  a 2HCO   m m 2
 ( ) 
2

where log 10  3   log  Ca 2  HCO 3  Ca 2  HCO 3 

10  
 a CO ,aq  m CO 2 , aq  CO 2, aq
 2   

The SI values for all scale types, e.g., BaSO4 or FeCO3, are calculated via an equation analogous to Eq 5. When SI values at both
the bottom hole and surface conditions are calculated, SI (= SIsurface - SIbottom hole) can be determined. This SI approach represents
the change in SI vs. (T,P) up the well. If the mineral(s) are at equilibrium in the formation, SIbottom hole = 0. Any calculated deviation
from SIbottom hole = 0.00 is a result of uncertainties in the various measured parameters, the equilibrium constants or calculation
procedure itself. Thereby, most analytical and theoretical uncertainties might be minimized by using the SI approach. This is
often a reliable way to interpret scale phenomena. Note that the carbonic acid activity in Eq. 6 can be related to CO2 partial
pressure by Henry’s law. Therefore, CO2 gas phase concentration can substitute pH in calcite SI calculations.
CaCO 3,solid  H 2 O  PCO 2  Ca 2   2HCO 3
 a Ca 2  a 2HCO- K HCO 3   a Ca 2  a 2HCO- K HCO 3 
SI Calcite  Log10   2
 Log  2 7
HCO 3  10  HCO 3 CO2 
3 3
CO 2
 a CO2,aq K sp K1   PCO2 γ gas K sp K1 K H 

where K1HCO 3 and K HCO 2
3
represent the first and second ionization constants of carbonic acid;  CO
2
gas
represents the fugacity
coefficient of CO2 in the gas; and, K H represents Henry’s law constant for the partitioning of CO2 between gas and water. This
CO2

is the recommended procedure, when possible, because generally the partial pressure of CO2 is better known and more easily
measured than the surface brine pH although SSP software will accommodate all combinations.

Temperature and Pressure dependence of equilibrium constants. In saturation index calculations, the temperature and pressure
dependence of the thermodynamic equilibrium constants and the activity coefficients are needed. Temperature and pressure
dependence of equilibrium constants has been extensively reported by Helgeson group28-30. The Helgeson theory is derived from
the temperature and pressure dependence of the partial molar volume and appropriate integration and differentiation with
Maxwell’s equations to yield the free energy of formation and the equilibrium constants at any temperature and pressure. These
equations contain 10 to 88 adjustable parameters. However, at the typical oil and gas production temperatures and pressures, the
SPE 132237 9

equations can be greatly simplified because it is below the critical temperature and pressure of water (373.946 C and 22,060 kPa).
Most equilibrium constants can be represented by about five adjustable parameters, as has been done by USGS2, which was used
in SSP. Figure 6 is a comparison of logarithmic barite solubility product (log(Ksp, barite)) calculated by Helgeson Supcrit92 versus
that calculated by SSP. Excellent agreements are observed between these two predictions in the temperature range of oil and gas
production (0-250 C).

Activity coefficients. The Pitzer model is the most accepted ion interaction model for high ionic strengths. It has been shown to
accurately model the behavior of electrolyte solution up to 6 mol/Kg H2O 31. At high TDS, the high ion density of solution can lead
to binary interaction of species of equal and opposite charge and ternary interactions between three or more ions. The Pitzer model
takes into account both the short range interactions in concentrated solution and the long range electrostatic effects of Dehye-
Hükel-type model31, 32. The excess free energy is assumed to be a virial (power series) expansion of binary and ternary interactions
with a leading term of a Debye-Hückel type. Essentially:

Pitzer  RT  {w w f  w w   ij ( I) n i n j  w w   ijk n i n j n k } with w w  kg of water

G Excess  1 2 8
i, j ijk

Where I = ionic strength in molal (m), n is moles,  ij and  ijk are binary and ternary constants and
 I1 / 2 2 
f   A    ln(1  1.2  I1 / 2 )  with A   0.3915 at 25 C and 1 atm for water..
 1  1.2  I
1/ 2
1.2 
Then
 G Excess / RT  9
ln( i )   Pitzer   Pitzer activity coefficients after considerable rearrangement.
 n i T , P , n
j

For example for a metal (M) and cations ions (c), anions (a), and neutral molecules (n):
 G Excess
Pitzer / RT

ln( M )   

 n M  T ,P,n j
10
 z 2m F(I)   m a (2B Ma  ZC Ma )   m c (2 Mc   m a Mca ) 
a c a

 m a m a ' Maa '  z M  m c m a C ca  2 m n  nM

a  a' c a n

The Pitzer coefficients, BMa , C Ma ,  Mc , Mca , Maa ' , C ca , and  nM , are functions of temperature, pressure ionic strength, and
composition. For NaCl, a total of 36 constants were included in Pitzer’s original derivation, but, for other ions, many parameters
have not been determined. No standard reference set of Pitzer coefficients has been established in literature. A consistent set of
Pitzer coefficients are included in SSP program after a thorough review of available Pitzer coefficients and experimental
observations7, 32-41. Note that the ion pair formation of 2-2 electrolytes, e.g., CaCO 30 , has been included in the BMa implicitly 31.
Several researchers proposed that an additional explicit ion pair term should be added if the association constant is greater than 500
for 2-2 type ion pairs or 20 for 2-l type ion pairs32, 34. However, there seems to be lack of sufficient evidence to support this
deviation from the original Pitzer code.
Scale Prediction by ScaleSoftPitzerTM. ScaleSoftPitzer (SSP)TM1 is a software program developed by the authors to
predict the scaling tendency of mineral scales common in oil and gas production. It is designed to calculate the pH, scaling index,
and inhibitor needs. The program also calculates density and heat capacity of fluids based on Pitzer theory. Minerals included: 1.
calcite (CaCO3); 2. barite (BaSO4); 3. gypsum (CaSO4·2H2O); 4. hemihydrate (CaSO4·½H2O); 5. anhydrite (CaSO4); 6. celestite
(SrSO4); 7. fluorite (CaF2); 8. sphalerite (ZnS); 9. iron-sulfide (FeS); 10. siderite (FeCO3); 11. halite (NaCl); 12. silica and
silicates; 13. lead sulfide (PbS); 14. brucite (Mg(OH)2) and 15 dolomite (CaMg(CO3)2). SSP uses a carefully selected set of
temperature and pressure dependence of thermodynamic equilibrium constants and Pitzer activity coefficients to calculate the
scaling index (Eq 5). Fugacity coefficients for the gas components are calculated from a Peng Robinson equation of state. The
effect of oil and gas composition on gas partitioning is estimated from the Vasquez and Beggs correlation of gas in solution to gas
specific gravity and API oil gravity42. The SSP program has four different ways to co-mingle brines, gas, oil, and seawater and
numerous “what-if” and GoalSeek calculations to simulate the various scenarios encountered in oil and gas production.
SSP also predicts the effect of hydrate inhibitors on mineral scaling tendency. For the nonelectrolyte effects (methanol, MEG
or TEG), the saturation index in the presence of hydrate inhibitors is defined to be the sum of the value in the absence of an
hydrate inhibitor (i.e., the value calculated by most computer codes) plus a term that is only a function of the added hydrate
10 SPE 132237

inhibitor, SI  SI( brine only)  SI(due to added MeOH, MEG, or TEG) . In this equation, SI(brine, only) refers to the saturation
index calculated for the produced brine before any hydrate inhibitor has been added and the value of
SI (due to added MeOH, MEG, or TEG) represents the effect of added hydrate inhibitor on scale formation at temperature and
ionic strength. This functional form of temperature and ionic strength dependence of SI(due to added MeOH, MEG, or TEG) was
suggested by Pitzer and others and has been tested by the present authors and found to represent experimental data on solubility
very well4-6, 43.
In Table 3 are listed typical sources of error for calcite SI value using a brine data from a particular gas well in South Texas as
example. For the purpose of illustration, the brine composition has been slightly modified to include the minor components. Most
cations can be measured accurately with ICP to within ±10% errors. However, some ions, such as Mg2+, K+, Fe2+, and Zn2+ are
often not measured in field brine. Ca/Fe: ratio is often about 100, if there is no corrosion taking place (calcite/siderite). Measured
pH can easily be off by ±0.5 pH units and it must be measured at the well site. The alkalinity is often off by ± (30 to 100) %,
unless is done by the method recommended above which produces both “true” alkalinity and total carboxylic acids simultaneously.
The overall estimated error in SI can be calculated from the various sources of error, sum the squares and the square root.
Estimated error in each parameter is listed on Column 9 (parameter). The contribution of each term is listed under Column 11. Based
upon a simple propagation of error formula, the estimated error in SI = 0.0827 (=(SI/Parameter)22parameter}1/2), i.e., SI(calcite)
= 0.664  0.083. This neglects systematic errors from equilibrium constants and activity coefficients, etc.
The program has been validated with large numbers of experimental data8, 12, 44-50 at a wide range of temperature, pressure and
electrolyte compositions (Figure 7 and Table 4). In Figure 8 is plotted the effects of methanol on salt solubilities and SSP SI
predictions. Hydrate inhibitors, especially methanol, have a strong impact on mineral salt solubilities and SSP is able to model the
effect. The range of applicability is listed in Table 5. Since the solution is assumed to be at equilibrium with a mineral phase, the
theoretical SI for all experiments should be 0.00, while SI > 0.0 represents a supersaturated solution and SI < 0.0 represents an
undersaturated solution. Excellent agreement between laboratory measured solubilities and the calculated SI values are observed.
Table 4 summarized the calculated SI for six minerals and 123 solubility data points. The mean SI is -0.013 and standard deviation
of 0.14, an excellent agreement to the theoretical SI of 0.00, validating the applicability of ScaleSoftPitzer for predicting scale
formation under these diverse oil field conditions. However, it should be noted that the relevant solubility data in complex brines
are very limited to sufficiently test the model in complex, high TDS brines. Similarly, there are few high pressure data to validate
the model prediction at HTHP conditions. The author’s research group is currently aiming to resolve such knowledge gaps.
Validation of pH calculation. For a known solution containing a fixed concentration of sodium bicarbonate and acetic acid
at equilibrium with the CO2 partial pressure, the solution pH can be calculated from the following charge balance equation.
[ Na  ]  [H  ]  [HCO3 ]  2[CO 32 ]  [AC ]  [OH  ]  0 or
10 pH K H  PCO2   g ,CO2  K1  PCO2   g ,CO2 K1H 2 CO 3  K HCO
CO 2 H 2 CO 3
K CO 2 3
TAc Kw
[ Na  ]   (10 pH ) 2 
H 2
  10 pH  2   0
 H  HCO  CO  10 pH   Ac  10 pH  OH
3 3   1
K  
 HAc HAc,aq 

11
Kassa7 reported the pHs of two solutions containing 16.53 mM NaHCO3 and 8.53 and 35.18 mM acetic acid at equilibrium with 0
– 600 psi CO2 partial pressure at 60 C. In Figure 9a is the comparison of the pH values calculated by SSP versus that measured by
Kassa7. The agreement between calculated and measured pH values is excellent, indicating that the equilibrium constants and
activity coefficients used in the above equation agree with measurements (Figure 9a). Furthermore, the pH of several calcite
saturated brine solutions containing various weight% of either methanol or MEG were measured in the laboratory. These
experiments were done with a background electrolyte of 1 molal NaCl, 0 - 0.1 m Ca2+ and a CO2 partial pressure of 0 – 1 atm
PCO2. Excellent agreement of the measured and predicted pHs was also observed (Figure 9b).
Validation of density calculation. The excess molar volume can be estimated from the pressure derivative of Pitzer activity
coefficients, i.e., Viex  log( i P)RT . Therefore, the density of the salt solution can be estimated from Pitzer theory by Eq. 12.
1000   m i  MWi 12

1000 /  H 2O   ( Vi0  Viex )  m i
Excellent agreement has been observed between the calculated density of 16 concentrated salt solutions and that reported in the
CRC handbook of chemistry and physics (Table 6) as well as the density of commercial weighting fluid up to 12.7 lb/gal density
(Figure 10).
Validation of minimum inhibitor concentration and critical SI prediction. As shown in Figure 1, scale may not form
at low SI. For calcite, experience has shown that the transition from "controllable" to "uncontrollable” scale formation occurs over
SPE 132237 11

a narrow SI range. The SI transition value from controllable to uncontrollable scale formation can vary from about 2.0 to a
maximum of 2.5 for calcite and may depend upon flow conditions, temperature, specific brine compositions, inhibitor type, and
other parameters. Even though the precise transition SI v a lu e is n o t w e l l d e f in e d , th e r e i s a l mo s t universal agreement
that such a region exists and care should be taken to avoid any production conditions that will push SI values into this range of
massive and rapid uncontrollable scale formation. SSP includes a model to predict if scale inhibitor treatment is needed and if so,
the minimum inhibitor concentration to inhibit scale (MIC). The model is derived from a large number of nucleation kinetic data
of barite, calcium sulfates, and calcite and the inhibition efficiencies of 8 scale inhibitor25, 51-54 (eq. 13-15).
t 0 sec  10α0 α1/SIα2/T( K)α3/(SIT( K)) 
 
13
β + β SI + β /T(  K) + β3 pH +β4logRinh 
b inh  L/mg = 10 0 1 2 14
1  t (sec) 
Cinh mg/L  = log  inh  15
b inh (L/mg)  t 0 (sec) 
Where, for example, t0 is the time required for scale to form a precipitate and tinh is the time required for scale inhibition. Based on
the scale inhibition model, the calcite SI and the minimum inhibitor concentration at different temperatures are calculated for a
brine composition of a gas well in Texas and plotted in Figure 11. This graph shows an exponential increase in inhibitor
concentration at around SI = 2.3 for calcite, which is an excellent agreement with the field observed transition SI for uncontrollable
scale formation.
McElhiney et al.55 used this program to calculate the concentration of Ba2+ that can be mixed with sulfate reduced seawater
without barite precipitation. Due to the high sulfate concentration in seawater, barite tolerance is less than 1 mg/L when an oilfield
brine is mixed with seawater. Using sulfate removal technology alone, Ba2+ tolerance can be increased by a factor of 40. Adding
inhibitor treatment (20 mg/L DTPMP), Ba2+ tolerance can be increased by a factor of 100 (See Fig. 12). Such an effects can be
predicted by SSPand is validated experimentally (Figure 12, round solid dots are experimental data).

Conclusions
Scale prediction and control are complex in nature. This paper summarized the state of the art in scale prediction and control for
oil and gas application. The following conclusions can be made:
1. Brine pH is difficult to measure with needed accuracy due to both theoretical and practical limitations. If gas phase CO2
concentration is known, it should always be used in place of measured pH.
2. Alkalinity measurements carry a large margin of error, unless the simultaneous alkalinity and organic acids titration
method discussed in this paper is used.
3. SSP is simple and versatile to simulate a wide range of conditions. The scale predictions agree well with literature
solubility data for a temperature range of 0-200 C, a pressure range of 0-15,000 psi, and a TDS range of 0-350,000
mg/L with a standard error of 0.1 SI unit.
4. There is not enough literature data at > 200 C temperature, > 15,000 psia pressure, in complex brine of high TDS to
validate scale predictions and more research is needed in these HTHP ranges.
5. SSP predicts the mineral scaling tendency in the presence of methanol and MEG.
6. SSP can calculate the brine pH, in the presence and absence of hydrate inhibitors, with excellent precision.
7. SSP can calculate the density of fluids up to 12.7 lb/gal.

Acknowledgements:
This work was financially supported by a consortium of companies including Baker-Petrolite, BJ Chemical Services, BP,
Champion Technologies, Inc., Chevron, Inc., ConocoPhillips, Inc., Halliburton, Hess, Kemira, Marathon Oil, MI SWACO, Multi-
Chem, Naclo, Petrobras, Saudia Aramco, Shell, Inc., StatOil, and Total, and National Science Foundation through the Center for
biological and Environmental Nanotechnology [EEC-0118007], US EPA ORD/NCER/STAR nanotechnology program, China-
U.S. Center for Environmental Remediation and Sustainable Development, and Advanced Energy Consortium.

Nomenclature:
pH j = the change in junction potential in unit of pH.
I = the ionic strength, molal.
Alk = Alkalinity (molal)
Alk0 = Initial alkalinity in brine (molal)
K CO
H
2
=Henry’s law constant for CO2 partition in gas/water phase (molal atm-1)
12 SPE 132237

K1H 2 CO 3 = the first ionization constant of carbonic acid

K HCO
2
3
= the second ionization constant of carbonic acid
K HAc = The dissociation constant of acetic acid
TAc = Total concentration of all acetate species (molal) = HAc+Ac-
 CO2, g =CO2 fugacity coefficient in gas
CO,o = CO2 activity coefficient in oil
n CO 2
total = CO2 in all phases (moles)

n CO
g
2
= CO2 in gas phase (moles)
n CO
w
2
= CO2 in aqueous phase (moles)

n HCO
w
3
= bicarbonate in aqueous phase (moles)
2
n CO
w
3
= carbonate in aqueous phase (moles)
PCO2 = partial pressure of CO2 (atm)
Vg = gas volume (l)
Z = gas compressibility
R = the gas constant
T = temperature (K)
ai = ion activity
mo = mass of oil (Kg)
mw = Ww= mass of water (Kg)
i = Pitzer activity coefficient
SI = saturation index
Ksp = solubility product
G Excess
Pitzer = excess freee energy
ij = Pitzer binary constants
ijk = Pitzer ternary constant
A =Debye Hückel constant for osmotic pressure
Ni = moles of components
BMa , C Ma ,  Mc , Mca , Maa ' , Cca , and  nM = Pitzer second and third virial coefficients
I= parameters to nucleation time
I= parameters to predict inhibition efficiency
Cinh = minimum inhibitor concentration (mg/L)
binh =inhibition efficiency coefficient
Rinh = lattice cation to anion molar ratio
t0 = nucleation time (sec)
tinh = inhibition time (sec)

References

1. Tomson, M.B. and A.T. Kan, SSP2009. 2009, Rice University, Brine Chemistry Consortium: Houston, TX.
2. Kharaka, Y.K., W.D. Gunter, and P.K. Aggarwal. "Solmineq 88: A computer program for geochemical modeling of
water-rock interactions", U. S. Geological Survey, 88-4227.
3. Shen, D., G. Fu, H. Al-Saiari, A.T. Kan, and M.B. Tomson, "Barite dissolution/precipitation kinetics in porous media and
in the presence and absence of a common scale inhibitor" SPE J. (2009) (In press).
4. Kan, A.T., G. Fu, and M.B. Tomson, "Effect of methanol and ethylene glycol on sulfates and halite scale formation" Ind.
Eng. Chem. Res. (2003) 42, 2399.
5. Lu, H., A.T. Kan, and M.B. Tomson, "Effects of Monoethylene Glycol on Carbonate Equilibrium and Calcite solubility in
Gas/Monoethylene Glycol/NaCl/water Mixed Systems" SPE J. (2010) (In Press).
6. Kan, A.T., G. Fu, and M.B. Tomson, "Effect of methanol on carbonate equilibrium and calcite solubility in a
SPE 132237 13

gas/methanol/water/ salt mixed system" Langmuir (2002) 18, 25, 9713.

7. Kaasa, B.: "Prediction of pH, mineral precipitation and multiphase equilibria during oil recovery", Institutt For Uorganisk
Kjemi, Norge Teknisk-Naturvitenskapelige Universitet, Trondhein (1998).
8. Howell, R.D., K. Raju, and G. Atkinson, "Thermodynamics of "Scale" Mineral Solubilities. 4. Experimental
Measurements of SrSO4(s) in H2O and Aqueous NaC1 from 25 to 250 o C and from 1 to 500 bar" J. Chem. Eng. (1992)
37, 464.
9. Reardon, E.J. and D.K. Armstrong, "Celestite (SrSO4(S)) Solubility in Water, Seawater and NaC1 Solution" Geochimica et
Cosmochimica Acta (1986) 51, 63.
10. Felmy, A.R., D. Rai, and J.E. Amonette, "The Solubility of Barite and Celestite in Sodium Sulfate: Evaluation of
Thermodynamic Data" J. SOlution Chemistry (1990) 19, 2, 175.
11. Davis, J.W. and A.G. Collins, "Solubility of Barium and Strontium Sulfates in Strong Electrolyte Solutions"
Environmental Science and Technology (1971) 5, 1039.
12. Pitzer, K.S., J.C. Peiper, and R.H. Busey, "Thermodynamic properties of aqueous sodium chloride solutions." J. Phys.
Chem. Ref. Data (1984) 13, 1.
13. Sontheimer, H., W. Kolle, and V. Snoeyink, "The siderite model fo the formation of corrosion resistant scales" J. Am.
Wat. Works Assoc. (1981), 572.
14. Greenberg, J. and M.B. Tomson, "The precipitation and dissolution kinetics and equilibria of aqueous ferrous carbonate
vs temperature" App. Geochem. (1992) 7, 185.
15. Tomson, M.B. and M.L. Johnson, "How ferrous carbonate kinetics impacts oilfield corrosion." SPE Production
Engineering (1991).
16. Deer, W.A., R.A. Howie, and J. Zussman, An Introduction to the Rock Forming Minerals, Halsted Press, 1983) 528.
17. Alsaiari, H.A., A. Kan, and M.B. Tomson, "Effect of Calcium and Iron (II) Ions on The Precipoitation of Calcium and
Ferrous Carbonate" SPE J. (In press) (2010).
18. Bates, R.G., Determination of pH - Theory and Practice, Vol. Second Edition, A Wiley-Interscience Publication, Canada
(1973).
19. Tomson, M.B. and J.E. Oddo, Handbook of Calcite Scale measurment, prediction and control., Hart Publications, Inc.,
1996).
20. Bard, A.J. and L.R. Raulkner, Electrochemical Methods: Fundamental and Applications, John Wiley & Sons, New York,
1980) 718.
21. deWaard, C. and D.E. Williams. "Prediction of Carbonic Acid Corrosion in Natural Gas Pipelines", Proc., First
International Conference on the Internal and External Protection of Pipes, Univ. of Durham, UK (1975).
22. Lundegard, P.D. and L.S. Land, "Carboxylic acid anions in formation waters, San Joaquin Basin and Louisiana Gulf
Coast, U.S.A.--Implications for clastic diagenesis. Discussion" Applied Geochemistry (1993) 8, 3, 297.
23. Lundegard, P.D. and Y.K. Kharaka, "Geochemistry of organic acids in subsurface waters. Field data, experimental data,
and models" ACS Symposium Series (Chem. Model. Aqueous Syst. 2) (1990) 416, 169.
24. Carothers, W.W. and Y.K. Kharaka, "Aliphatic acid anions in oil-field waters - implications for origin of natural gas. "
American Association of Petroleum Geologists Bulletin (1978) 62, 12, 2441.
25. He, S.L., A.T. Kan, and M.B. Tomson. "A new interactive software for scale prediction, control and management", Proc.,
SPE Annual Technical Conference and exhibition, San Antonio, TX (1997).
26. Tomson, M.B., A.T. Kan, G. Fu, and L. Cong, "Measurement of total alkalinity and carboxylic acid and their relation to
scaling and corrosion" SPE J. (2006) 11, 1, 103.
27. Kan, A.T., G. Fu, and M.B. Tomson. "SPE 93265 Prediction of scale inhibitor squeeze and return in calcite-bearing
formation", Proc., 2005 SPE International Symposium on Oilfield Chemistry, Houston, TX (2005).
28. Helgeson, H.C., "Thermodynamics of Complex Dissociation in Aqueous Solution at Elevated Temperatures" J. of Phys.
Chem. (1967) 71, no. 10, pp. 3121.
29. Johnson, J.W., E.H. Oelkers, and H.C. Helgeson, "SUPCRT92: A software package for calculating the standard molal
thermodynamic properties of minerals, gases, aqueous species, and reactions from 1 to 5000 bar and 0 to 1000°C"
Computers & Geosciences (1992) 18, 7, 899.
30. Tanger, J.C. and H.C. Helgeson, "Calculation of the thermodynamic and transport properties of aqueous species at high
pressures and temperatures; revised equations of state for the standard partial molal properties of ions and electrolytes"
Am J Sci (1988) 288, 1, 19.
31. Pitzer, K.S., Thermodynamics, 3rd Ed ed, McGraw-Hill, New York (1995).
32. Harvie, C.E., N. Møller, and J.H. Weare, "The prediction of mineral solubilities in natural waters: The Na-K-Mg-Ca-H-
Cl-SO4-OH-HCO3-CO3-CO2-H2O system to high ionic strengths at 25°C" Geochimica et Cosmochimica Acta (1984)
48, 4, 723.
14 SPE 132237

33. He, S. and J.W. Morse, "Prediction of gypsum, anhydrite and halite solubility in brines under subsurface conditions."
Comput. Geosci. (1993) 19, 1.
34. He, S. and J.W. Morse, "The carbonic acid system and calcite solubility in aqueous Na-K-Ca-Mg-Cl-SO4 solutions from
0 to 90°C" Geochimica et Cosmochimica Acta (1993) 57, 15, 3533.
35. Harvie, C.E. and J.H. Weare, "The prediction of mineral solubilities in natural waters: the Na---K---Mg---Ca---Cl---SO4--
-H2O system from zero to high concentration at 25° C" Geochimica et Cosmochimica Acta (1980) 44, 7, 981.
36. Greenberg, J.P. and N. Møller, "The prediction of mineral solubilities in natural waters: A chemical equilibrium model for
the Na-K-Ca-Cl-SO4-H2O system to high concentration from 0 to 250°C" Geochimica et Cosmochimica Acta (1989) 53,
10, 2503.
37. Felmy, A.R. and J.H. Weare, "The prediction of borate mineral equilibria in natural waters: Application to Searles Lake,
California" Geochimica et Cosmochimica Acta (1986) 50, 12, 2771.
38. Christov, C. and N. Moller, "Chemical equilibrium model of solution behavior and solubility in the H-Na-K-OH-Cl-
HSO4-SO4-H2O system to high concentration and temperature" Geochimica et Cosmochimica Acta (2004) 68, 6, 1309.
39. Millero, F., F. Huang, T. Graham, and D. Pierrot, "The dissociation of carbonic acid in NaCl solutions as a function of
concentration and temperature" Geochimica et Cosmochimica Acta (2007) 71, 1, 46.
40. Duan, Z. and D. Li, "Coupled phase and aqueous species equilibrium of the H2O-CO2-NaCl-CaCO3 system from 0 to
250 °C, 1 to 1000 bar with NaCl concentrations up to saturation of halite" Geochimica et Cosmochimica Acta (2008) 72,
20, 5128.
41. Monnin, C., "A thermodynamic model for the solubility of barite and celestite in electrolyte solutions and seawater to
200°C and to 1 kbar" Chemical Geology (1999) 153, 1-4, 187.
42. Bradley, H.B., Petroleum Engineering Handbook, 1987).
43. Kan , A.T., H. Lu, and M.B. Tomson, "Effects of Monoethylene Glycol on Carbon dioxide partitioning in
Gas/Monoethylene Glycol/water/Salt Mixed Systems" Industrial & Engineering Chemistry Research (2010), (In press).
44. Ellis, A.J., "The Solubility of Calcite in Carbon Dioxide Solutions" Amer. J. of Sci. (1959) 257, May, pp. 354.
45. Ellis, A.J., "The Solubility of Calcite in Sodium Chloride Solutions at High Temperatures" Amer. J. of Sci. (1963) 261,
March, pp. 259.
46. Wolf, M., O. Greitkopf, and R. Puk, "Solubility of calcite in different electrolytes at temperatures between 1 0 ° and 60
°C and at C02 partial pressures of about 1 kPa" Chemical Geology (1989) 76, 291.
47. Templeton, C.C., "Solubility of Barium Sulfate in Sodium Chloride Solutions from 25 C. to 95 C." J. of Chem. and Eng.
Data (1960) 5, Oct., pp. 514.
48. Blount, C.W., "Barite Solubilities and Thermodynamic Quantities up to 300 C. and 1400 Bars" Amer. Min. (1977) 62, pp.
942.
49. Bock, E., "On The Solubility of Anhydrous Calcum Sulfate and of Gypsum in Concentrated Solutions of Sodium
Chloride at 25, 30, 40, and 50 C." Can. J. of Chem. (1961) 39, pp. 1746.
50. Marshall, W.L., R. Slusher, and E.V. Jones, "Solubility and Thermodynamic Relationships for CaSO4 in NaCl-H2O
Solutions from 40 to 200 C, 0 to 4 Molal NaCl" J. Chem. and Eng. Data (1964) 9, 2, pp. 187.
51. Fan, C., A.T. Kan, G. Fu, M.B. Tomson, and D. Shen, "Quantitative evaluation of calcium sulfate precipitation kinetics in
the presence and absence of scale inhibitors" SPE J. (2010) (In press).
52. He, S.L., A.T. Kan, and M.B. Tomson, "Mathematical Inhibitor Model for Barium Sulfate Scale Control" Langmuir
(1996) 12, 7, 1901.
53. He, S.L., A.T. Kan, and M.B. Tomson, "Inhibition of calcium carbonate precipitation in NaCl brines from 25 to 90 C."
Applied Geochemistry (1999) 14, 17.
54. Xiao, J.A., A.T. Kan, and M.B. Tomson, "Prediction of BaSO4 precipitation in the presence and absence of a polymeric
inhibitor: Phosphino-polycarboxylic acid" Langmuir (2001) 17, 15, 4668.
55. McElhiney, J.E., M.B. Tomson, and A.T. Kan, SPE 100480 Design of low sulfate seawater injection based upon kinetic
limits, in 2006 SPE International Oilfield Scale Symposium 2006, SPE: Aberdeen, Scotland, U.K.
SPE 132237 15

Table 1. Range of variables to be considered in oilfield scale prediction and control.

Temperature Pressure Oil Brine Processes
25 to 500 F 14 to 30,000 Psi Aromatics Na+, K+, Ca2+, Equil, rate, & mass
or or Paraffins Mg2+, Ba2+, Sr2+, transfer processes:
-4 to 260 C 1 to 2,000 atm Waxes Fe2+, Zn2+, Pb2+, Cl-, Gas, solids,
Asphaltenes SO42-, HCO3-, Br-, redox, corrosion,
Naphthenic acids HS-, HAc (acids), bio., minerals,
SiO2 NORMs
MeOH,
MEGs, Inhibition kinetics &
0.00 to 400,000 mg/L concentrations for eight
TDS different inhibitors and
blends.
Table 2. Round Robin Test Results of the Rice University Total Alkalinity, Carboxylic Acid Simultaneous Titration at Constant PCO2©
Analysis.
Brine # Lab. Alkalinity Std Organic Std. Brine Lab. Alkalinity Std Organic Std. Brine # Lab. Alkalinity Std Organic Std.
# (mg/L) Dev Acid Dev. # # (mg/L) Dev Acid Dev. # (mg/L) Dev Acid Dev.
(mg/L) (mg/L) (mg/L)
1 1 1082 3 1234 14 2 1 401 2 557 47 3 1 514 10 15 12
1 1 1123 3 1254 13 2 1 386 4 418 50 3 1 484 32 49 33
1 1 1024 2 1138 10 2 1 377 15 229 64 3 2 529 91 60 73
1 2 1126 3 1175 10 2 2 386 6 494 60 3 2 532 107 41 89
1 2 1119 3 1172 12 2 2 377 3 483 31 3 3 525 1 35 7
1 3 1122 5 1041 46 2 3 329 3 486 134 3 3 543 4 0 0
1 3 1111 2 1220 24 2 3 337 1 381 34 3 4 516 4 0 0
1 3 1120 7 1270 34 2 4 353 3 432 39 3 4 516 4 0 0
1 4 1125 1 1248 5 2 4 354 4 431 60 3
1 4 1106 2 1235 9
GC analysis 1186 297 0
1133 222 0
Mean 1106 3 1199 18 359 5 435 58 520 32 25 27
Standard 10 1 22 4 8 1 31 10 6 15 9 13
Error
Standard 32 2 69 13 24 4 93 31 17 43 24 35
Deviation
Table 3. Estimated error in calcite SI prediction due to errors in brine analysis.
1 2 3 4    8   11
Estimated
Parameter Parameter SI / param.  2 Error of Estimated
 SI    SI
2
 2 
Initial Final SI with SI)/   parameters    param. Error from
Parameter value value final value SI (Final-Initial)  parameter  parameter parameter)2  (Final  Initial)  each term
TF 340 350 0.73139 0.06718 -6.72E-03 4.51E-05 5 2.50E+01 1.13E-03 3.36E-02
P psia 7000 7700 0.56316 -0.10105 1.44E-04 2.08E-08 100 1.00E+04 2.08E-04 1.44E-02
Na mg/L 19872 20872 0.63995 -0.02426 2.43E-05 5.89E-10 1000 1.00E+06 5.89E-04 2.43E-02
K mg/L 500 1500 0.64977 -0.01444 1.44E-05 2.09E-10 25 6.25E+02 1.30E-07 3.61E-04
Mg mg/L 54 154 0.65621 -0.008 8.00E-05 6.40E-09 5 2.50E+01 1.60E-07 4.00E-04
Ca mg/L 6500 6600 0.67016 0.00595 -5.95E-05 3.54E-09 325 1.06E+05 3.75E-04 1.94E-02
Sr mg/L 700 800 0.66265 -0.00156 1.56E-05 2.43E-10 35 1.23E+03 2.99E-07 5.47E-04
Ba mg/L 550 650 0.66308 -0.00113 1.13E-05 1.28E-10 27.5 7.56E+02 9.65E-08 3.11E-04
Fe mg/L 12 20 0.66487 0.00066 -8.25E-05 6.81E-09 2 4.00E+00 2.72E-08 1.65E-04
Zn mg/L 10 20 0.66403 -0.00018 1.80E-05 3.24E-10 2 4.00E+00 1.30E-09 3.60E-05
Pb mg/L 1 0 0.66421 0 0.00E+00 0.00E+00 2 4.00E+00 0.00E+00 0.00E+00
Cl mg/L 43000 44000 0.67331 0.0091 -9.10E-06 8.28E-11 2150 4.62E+06 3.83E-04 1.96E-02
SO4 mg/L 5 15 0.66431 1E-04 -1.00E-05 1.00E-10 5 2.50E+01 2.50E-09 5.00E-05
F mg/L 1 11 0.66407 -0.00014 1.40E-05 1.96E-10 1 1.00E+00 1.96E-10 1.40E-05
Br mg/L 10 20 0.66425 4E-05 -4.00E-06 1.60E-11 1 2.00E+00 3.20E-11 5.66E-06
SiO2 mg/L 10 20 0.66421 0 0.00E+00 0.00E+00 1 3.00E+00 0.00E+00 0.00E+00
Alk mg/L 281 291 0.7019 0.03769 -3.77E-03 1.42E-05 14.05 1.97E+02 2.80E-03 5.29E-02
Ac mg/L 0 10 0.62817 -0.03604 3.60E-03 1.30E-05 10 1.00E+02 1.30E-03 3.60E-02
CO2 % 1.04 1.14 0.63055 -0.03366 3.37E-01 1.13E-01 0.01 1.82E-04 2.06E-05 4.54E-03
pH=7.1627 pH=7.1240 -0.66421
H2S % 0.0283 0 0.97266 0.30845 1.09E+01 1.19E+02 0.0005 2.50E-07 2.97E-05 5.45E-03
pH=7.1627 pH=7.1722 -0.66421
Gas 103SCF 8500 8600 0.66386 -0.00035 3.50E-06 1.22E-11 425 1.81E+05 2.22E-06
Oil B 1000 1100 0.67015 0.00594 -5.94E-05 3.53E-09 50 2.50E+03 8.82E-06 2.97E-03
Brine B 100 200 0.66984 0.00563 -5.63E-05 3.17E-09 5 2.50E+01 7.92E-08 2.82E-04

        SI­initial =  0.66421 SUM Sqrs = 6.84E-03

Estimated error in Calcite SI value, (SUM Sqrs)1/2 = 0.08270
Table 4. Overall quality of SSP SI prediction.
Scale Type I (m) T (C) P (psi) SI  S.D. n
Calcite 0 - 6.24 0 - 200 14.7 - 174 0.047  0.121 30
Barite 0 - 5.0 22 - 250 14.7 - 14,417 -0.013  0.147 36
Halite 6.097 - 10.4 0 - 300 14.7 -1,244 -0.003  0.041 13
Gypsum 0 - 6.08 0 - 70 14.7 - 50 0.036  0.144 18
Anhydrite 0 - 6.30 150 14.7 -0.087  0.081 7
Celestite 0-4 20 - 150 14.7 - 7,409 -0.137  0.156 19
Summary Estimated Range of SSP Applicability
All Scales 0 - 10 0 - 200 14.7 - 15,000 -0.013  0.14 123

Table. 5. Applicable range of SSP SI prediction in the presence of hydrate inhibitors.

Co-Solvent Conc.
Mineral Co-Solvent I (mol/Kg H2O) T (ºC)
(wt%)
Barite MeOH 0 – 75 0–3 0 – 200
Barite MEG 0 – 75 1–3 0 – 200
Barite TEG 0 – 90 1–3 0 – 25
Halite MeOH 0 – 90 2.8 – 6.2 0 – 200
Halite MEG 0 – 80 6.2 – 8.6 0 – 200
Celestite MeOH 0 – 50 0–3 0 – 25
Gypsum MeOH 0 – 50 0–3 25
Calcite MeOH 0 – 75 0–3 0 – 100
Calcite MEG 0 – 75 0–3 0 – 100
SPE 132237 19

Table 6. SSP Predicted Density vs. density of concentrated salt solution reported in CRC handbook of
chemistry and physics (20 °C).

20
Density ( D 4 , g/mL)
reported in CRC Density (g/mL)
handbook of Deviation
Salt solution Conc. (M) calculated by
Chemistry and (g/ml)
SSP
Physics

Acetic acid 6.255 1.043 1.049 0.006

Na Acetate 4.243 1.160 1.170 0.010
KCl 3.742 1.168 1.162 0.006
NaCl 5.326 1.197 1.185 -0.012
MgCl2 6H2O 4.021 1.276 1.287 0.011
BaCl2 1.597 1.279 1.282 0.003
MgSO4 7H2O 2.799 1.296 1.262 -0.034
CaCl2 2H2O 5.03 1.395 1.420 0.026
HCl 6.022 1.098 1.097 -0.001
NaBr 5.495 1.414 1.421 0.007
SrCl2 2.293 1.298 1.303 0.005
H2SO4 5.313 1.303 1.288 -0.015
KHCO3 2.801 1.169 1.181 0.012
K2CO3 4.093 1.414 1.473 0.059
NaHCO3 0.743 1.043 1.042 -0.001
Na2CO3 1.638 1.157 1.170 0.013
Average 0.006±0.019
deviation
20 SPE 132237

Above some critical SI value nucleation and

SI > +1.5 (120 C) crystal growth will be spontaneous and can not be
SI > +3 (25C) prevented by added scale inhibitors. This critical
SI value for barite is not well known and depends
upon T, pH, and the inhibitor selected.

SI ≈ + 1.5 (120 C)

SI ≈ + 3.0 (25C) Solutions in this region can be inhibited
with barite scale inhibitors. Nucleation and
crystal growth can be delayed for long
periods of time.

SI ≈ + 0.3 (120 C)

SI ≈ + 1.3 (25C) Precipitation and
growth occurs only on
seed crystals of barite.
SI = 0.00 Equilibrium

If the SI < 0, only dissolution can occur.

SI = negative

Time (Not to scale)

Figure 1. Schematic diagram to illustrate the relationship of SI and kinetics. The lines are time to precipitate or dissolve to equilibrium for
barite. A similar diagram could be constructed for any type of scale formation, except the specific T and SI values would change.
Typically, the more soluble a mineral, the lower the corresponding SI value.
 7.5
pH of pure water, no salt

6.5

5.5
0 25 50 75 100 125 150 175 200 225
T(C)

Figure 2. Plot of pure water pH versus temperature.

TAc (M)
TAc (M) 0.012
0
10.00
0 0.01 0.001
9.00 0.001

HCO3 conc., M
0.002
0.008
8.00 0.002 0.003
0.003 0.006 0.004
pH

7.00 0.004 0.005

6.00 0.005 0.004
0.006
0.006 0.002 0.007
5.00
0.007 0.008
4.00 0.008 0
0.009
0.009
01

05

01

05
0

0
2

2
20

20
1

5
0.01
10

10
0.

0.
0.

0.

0.

0.

0.01 0.011
P-CO2, psia 0.011 P-CO2, psia

Figure 3. Plots of the variation of pH and HCO3 concentration versus PCO2, gas and TAc concentrations, all at a final Alk = 0.01 eq/L  610
-

mg/L as HCO3 . The equilibrium constants used are at 212 F and 15 psia and 1 M I. Note that with any specific well, the change in
-

pressure will typically be a factor of 100x, or less.

22 SPE 132237

a.

2 3 4
9 pH
7
1 5 6

8
6
Gas Titration
Saturation apparatus
Bottle

b.

8.0

7.0

6.0
pH

5.0

4.0

3.0
0.00 0.01 0.01 0.02 0.02
HC l adde d (M, Exp)

"Exp" "Calc"

Figure 4a. Schematic diagram of the titration apparatus. The parts are (1) CO2/N2 gas cylinder, (2)gas regulator, (3)regulating valve, (4)
1/8" OD Teflon tubing, (5) glass bottle with a GL-45 2 valve cap, (5) sparging filter, (6) 8 oz jar, with a 1.5” stirring bar; (7) magnetic stirrer;
(8) pH electrode and (9) pH meter. Figure 4b. Typical titration results curve fitted by the BCC Total Alkalinity and Carboxylic Acid
Simultaneous Titration at Constant PCO2 program.
SPE 132237 23

a. b.

Alkalinity by Titration Corrected Alkalinity

PVC HDPE PVS HDPE
Alkalinity (mg/L)

550

Alkalinity (mg/L)
550
500 500
450 450
400 400
350 350
300
300
0 2 4 6 8 10 0 2 4 6 8 10
Time (days)
Time (days)
2+ -
Figure 5. (a) Measured alkalinity of a Fe (154 mg/L) containing HCO3 solution (471 mg/L) in 1 M NaCl versus brine storage time in PVC vs.
HDPE bottles and (b) Corrected alkalinity of the data in Figure 5a by the procedure.
24 SPE 132237

SSP2009 SUPCRT92

-9
0 50 100 150 200 250
-9.5

Barite log(Ksp)
-10

-10.5

-11

-11.5

-12
Temperature (C)

SSP2009 SUPCRT92

-9
0 50 100 150 200 250
-9.5
Barite log(Ksp)

-10

-10.5

-11

-11.5

-12
Temperature (C)
1
Figure 6. Comparison of based 10 logarithm of barite solubility product calculated by ScaleSoftPitzer (SSP2009) versus Helgeson’s
29
SUPCRT92 .
SPE 132237 25

a. b.

Calcite, 140 F Barite, 203 F Barite

Halite, 32-536 F Gypsum, 122 F
1.0
Anhydrite, 122 F Celestite, 169 F
2 0.5

Barite SI
1.5
0.0
Calculated SI

1
0.5 -0.5
0 1 m NaCl
-0.5 -1.0
-1 0 100 200 300 400
-1.5 Temperature (ºF)
-2
0 1 2 3 4 5 6 7 8 9
Ionic Strengh

c. d.

Barite Calcite
205 F 372 F 478 F Ellis, 1 m NaCl, 12 atm CO2
Wolf, 1 m NaCl, 0.01 atm CO2
1 Ellis, 0 m NaCl, 12 atm CO2
1.0
0.5
Barite SI

0.5
Calcite SI

0 0.0

-0.5 -0.5

-1.0
-1
0 200 400 600
0 5000 10000 15000 20000 Temperature (F)
Pressure (psia)

Figure 7. Plot of Pitzer theory based SI of selected literature reported mineral salt solubility data, where ScaleSoftPitzer was used. (a) the
calculated SI of selected solubility data for calcite, barite, halite, gypsum, anhydrite, and celestite at different temperature and ionic
; (2) Barite SI of solubility data at 77-400 F ; (c)
8, 12, 44-50 27
strength where the pressure is typically close to the vapor pressure of water
Barite SI of solubility data at 200 -400 F and 0 – 15,000 psia pressure ; and (d) Calcite SI of solubility data from 77-500 F and vapor
48
44-46
pressure of water .
26 SPE 132237

a. b.

1.E-01 0% MeOH 25% MeOH 49% MeOH 75% MeOH

Gypsum
Ba, Sr, or Ca Conc.
(moles/Kg H2O)

1.E-02 2

1.E-03 1

Barite SI
Celestite
1.E-04 0
0 100 200 300 400
1.E-05 -1
Barite
1.E-06 -2

0.0 0.1 0.2 0.3 0.4 Temp (F)

Cosolvent Conc. (mole fraction)

c. d.

Obs. Literature 0% MeOH 25 % MeOH 50% MeOH 75% MeOH

NaCl Solubility (mol/Kg

1.0
7
6 25 C
5
H2O)

4 Halite SI
3 0.0
2 0 100 200 300 400

0 50 100
MeOH (wt%)
-1.0
Temperature (F)

Figure. 8. (a) Sulfates salts solubility in methanol/NaCl/Water solutions; (b) Calculated barite SI by SSP using barite solubility data in 0-
75% Methanol/NaCl/water solution at 77-400 F; (c) Literature reported and observed halite solubility in methanol/NaCl/Water solution at
25 C; and (d) Calculated halite SI by SSP using halite solubility data in 0-75% Methanol/NaCl/water solution at 77-400 F.
a. b.
0-50% MeOH, 0-0.5 atm PCO2, 0-0.1 m Ca
Calc pH (I) Obs pH (I) 0-75% MeOH, 1 atm PCO2
Calc pH (II) Obs pH (II)
25-75% MEG, 1 atm PCO2
6.5
6.30
NaHCO3=16.53 mM, 60C y = 0.998x
6.10

SSP calc. pH
6.0 I. HAC = 8.53 mM R = 0.98
II. HAC= 35.18 mM
5.90
5.5

5.70
pH

5.0
I 5.50

4.5
II
5.30
5.3 5.8 6.3
4.0
0 100 200 300 400 500 600 Measured pH
PCO2 (psi)

Figure 9. SSP calculated versus measured pHs. (a) The pH of two solutions at equilibrium with different CO2 partial pressure in the gas
7
phase. The experimental data were from Kaasa . The two solutions containing 16.53 mM sodium bicarbonate and either 8.53 or 35.18 mM
acetic acid. The pHs were measured at 60 C and the solution is at equilibrium with 0 – 500 psi CO2 partial pressure (0-40 bar). (b) The pH
of calcite saturated solutions containing various weight% of methanol or MEG at 0-1 atm of CO2 partial pressure.

13.0
12.5 y = 0.9981x
R2 = 0.9998
SSP calc density (lb/gal)

12.0
11.5
11.0
10.5
10.0
9.5
9.0
8.5
8.0
8.0 8.5 9.0 9.5 10.0 10.5 11.0 11.5 12.0 12.5 13.0
Measured density (lb/gal)

Figure 10 Calculated versus measured density of a commercial weighting fluid at 25 C. Measured density is provided by a Brine
Chemistry Consortium member.
28 SPE 132237

Calcite SI NTMP

3 70

60
2.5

50
2

NTMP (mg/L)
Calcite SI

40
1.5
30

1
20

0.5
10

0 0
0 50 100 150
Temperature (C)

Figure 11. Calculated calcite SI and inhibitor need versus time. The brine composition of a gas well in Texas is used for illustration.
SPE 132237 29

No Inh 1 mg/L DTPMP 5 mg/L DTPMP

10 mg/L DTPMP 20 mg/L DTPMP Exp Obs
10000

(20) (20)
1000

Ba in Formation Water (mg/L)

100 (10)

10

0.1
0 500 1000 1500 2000 2500 3000
SO4 in Sulfate Reduced Seawater (mg/L)

Figure 12. Plot of barium versus sulfate concentrations that can be tolerated for 120 minutes without precipitation in the presence 0-20
mg/L of a scale inhibitor (diethylenetriaminepenta(phosphonic acid, DTPMP). . The results were calculated by SSP GoalSeek function. The
calculated results were compared to the laboratory observed results where 10 or 20 mg/L DTPMP were used and the concentrations were
labeled in parentheses in the plot.