SPE-3188-PA

An Evaluation of Scale Inhibitors0.J. Vetter, SPE-AIME, Union Oil Co. of California

Scale Inhibition Is Still an ArtScale inhibition is still closer to being an art than ascience. The literature deals only to a very smallextent with the problems of scale inhibition. Most ofthe published data do not consider the basic facts ofcrystallization and inhibition, such asreactionmecha-~i~w,~~nd kinetics. Because these fundamental con-siderations are often neglected, there are many re-corded failures of scale treating proceduresin oil andgas fields.

This neglect of fundamentals and lack of data issurprising because the formation of scale is a rathercostly problem for some industries, above all thepetroleum industry. We lose many millions of dol-lars in revenue every year because scale creates flowrestrictions in our fields.

Accurate evaluation of scale inhibitors is the firstand most important step toward effective scalecontrol.

How Are Scale InhibitorsPresently Tested?Test Machines

Several investigators’-’ have been concerned with thedevelopment of test machines designed to simulatethe conditions of scale deposition in the field. Thedegree of scale buildup in these test devices is usedto screen and rank commercial products and experi-mental chemicals.

~~~lo !fl!l!pQns“~”..

The use of scale coupons in the field is a test some-what related to the use of test machines in the labo-

ratory. Steel coupons are inserted in flow lines, andthe amount of scale they accumulate is used to evalu-ate an inhibitor that has been injected into the systemupstream of the coupon.

Precipitation Test

In the “chelation test,” ‘ -11.-./l*hawhicn s’houki be CaIIGu .Ileprecipitation test, two chemically incompatible solu-tions are combined. The precipitation of scale-forming solids — for example, calcium carbonate(CaCO,,), calcium sulfate (CaSO,) or barium sulfate(BaSO,) is measured in the presence of an inhibitor ofvarying concentrations by means of chemical analy-ses. The amounts of precipitate obtained are thencompared with the amount of precipitate formed bymixing solutions containing no scale inhibitor. Themore solids that are kept in solution by a givenamount of inhibitor, or the less inhibitor that isneeded to keep a given amount of solids in solution,the more effective and desirable the inhibitor isthought to be.

This test is widely used for laboratory evaluationof scale inhibitors. The test is very quick and con-venient to run, gives fast results, does not requirecostly equipment, and can be run by relatively un-trained personnel. Unfortunately, the reproducibilityof the result is sometimes very poor, and data re-ported by different testers is often contradictory.

The “Common” Method

Then, of course, there is the “common” mefhod oftesting: an inhibitor is tried in the field without pre-

A rating o,f inhibitors in the order of their effectiveness is a relative matter and dependsupon test conditions. The most important conditions are the degree of supersaturationand the temperature; the higher the supersaturation and the higher the temperature theless eflective a scale inhibitor becomes.

AUGUST, 1972 TPT 997

vious testing. The success or failure of the job de-cides the “quality” of the inhibitor.

Precipitation Is More Significant ThanAdherence of CrystalsIf aninhibitor prevents the crystallization of a scale-forming compound (CaCO:,, CaSQ ~, etc.), w% ~~~-not see a need for another inhibitor to prevent theadherence of crystals to themselves or to hardwarein the field. On the other hand, if an inhibitor does..-. -- ..-. +-r=+911+75tifln f@ lcnnwle&Te of its effectsnut plev Cnl ~1Y3La,,.Lu.lui.,. .- ..... .. . . ~on adherence could be very beneficial, Therefore, weshould measure the effects of the inhibitor on bothcrystallization and adherence; i.e., we should keep thetwo effects apart and measure them separately.

i4n inhibitor that prevents the adherence of scale-forming crystals but not the formation of tliein maynot be very useful in oil fields. Such an inhibitor maywell be effective for preventing scale in pipe. But ifthe scale is formed inside the porous medium’ (e.g.,an oil-producing zone), a lack of adherence may notprevent formation damage in the well, since the non-adherent crystals would simply be filtered out, causinga decrease of permeability by “scale”. Therefore, inour tests, we do not at the present attempt to evalu-ate adherence effects.

Test Machines Ignore Oilwell ConditionsTwo main arguments are generally offered for usingtest machines:

1. They measure simultaneously the effects of theinhibitor on the precipitation and the adherence ofcrystals.

2. They simulate actual well conditions.The first argument is valid. However, we still do

not know how much of the scale buildup can beattributed to the effects of the inhibitor on precipita-tion and how much on adherence. This means thatwe still have to run tests to measure either the pre-cipitation effect or the adherence effect alone, thuscreating the need for an additional test.

The second argument is not valid. The adherenceof individual crystals to a metal surface or to othercrystals will depend largely on both flow and scalingconditions. It is quite likely that both sets of condi-tions in the test are entirely different from the cor-responding conditions in an actual field. Therefore,an inhibitor that is rated excellent after evaluation inthe machine could be ineffective under conditionsfWIId in m actual oil well. This is one of the mainreasons why we do not recommend evaluating in-hibitors in test machines.

Results Obtained with Scale CouponsAre Almost MeaninglessThe particular location in the field where we deposita coupon may not happen to offer serious scalingproblems. Thus, the fact that we find no scale in agiven location does not necessarily mean that theinhibitor effectively prevents scale formation.

For example, we may have found scale on a cou-pon before we injected an inhibitor. It may be thatafter the inhibitor is injected there is no scale on thecoupon; however, heavy scaling may still take place

at the bottom of the well, within the perforation, orat any other point away from the coupon. In such a~a~e, the inhihitor ~as not So]ved our problem, butLI,w ...... . . . .the coupon test rates it “excellent,” nevertheless.

Our objections to the use of test machines appiylargely also to the use of coupons.

Objectives of This StudyThe studies discussed below have a threefold objec-tive:

1. To improve the precipitation test by taking intoaccount the effects of temperature, supersatura-tion, and chemical composition.

2. To determine the compatibility of inhibitorswith oilfield effluents.

3. To rank inhibitors for specific applications.

Precipitation Test Measures Effectiveiiess ofInhibitors on CrystallizationWe have evaluated a large number of inhibitors withthis test. All of these inhibitors belong to one of thefollowing chemical families:

1. Polymers,2. Esters of phosphoric acid,3. Phosphonates.

Inhibitors belonging to all of these three classes arecommercially available.

We have not tested inorganic phosphates becauseof their well known disadvantages (i.e., they convertrapidly to ineffective and damaging o-phosphates)even though they may find a limited application underconditions of low temperature and high PH. How-ever, many species from the other three classes havebeen tested. The following paragraphs describe ourexperience and give some explanations as to whythe data reported in the literature seem to showcontradictions.

We included into our precipitation tests only thethree major oilfield scales; namely, CaCO,,, CaSO.,,and BaS04. Only one of these compounds was pres-ent in each precipitation test. Mixed scales, particu-larly the combination CaCO,,-CaSO,, are frequentlyencountered in the field. These mixed scales containindividual crystals of each species; e.g., a CaCO~-CaSO, scale consists of individual CaCO,, and CaSO~crystals. Therefore, the data obtained from our pre-cipitation tests can also be applied if these mixedscales are found and if we treat the species in thesemixed scales separately.

Sodium chloride was the only impurity presentduring the tests. Other impurities, frequeiitly eil-countered in oilfield brines, may have an effect onour results. These possible effects will be studied infuture work.

The Precipitation Experiments

Two solutions were prepared for each percipitationexperiment as follows:

1. One solution contained the predetermined con-centration of the inhibitor plus twice the desiredconcentration of Ca+ + ions as CaCl,.

. . -l ...:_- _ -+.:m-~ +h. c m- nmnllnt~. 1 ne dk S.UIULIU1l LO1lLCUIIUI L!J& o&I- . .... ....

of inhibitor plus equivalent concentrations of SO.,--ions in the form of NalSO~.

998 JOURNAL OF PETROLEUM TECHNOLOGY

The second solution also contained a calculatedamount of NaCl to bring the total NaCl concentra-tion after combining the two solutions to 0,5 mol or2.92 percent NaC1.

The tests were run at three different temperatures:75”F, 21 O°F, and 350”F. The two solutions (100 mleach) were brought to the same temperature and thencombined. A simple oil bath was used for the experi-ments up to 2 10°F. A pressure bomb was used forthe experiments at 350”F.

Fig. 1 shows a schematic cross-section of one ofthe bombs. One solution was poured into the Teflon-lined bomb; the other was poured into the glass. Thenthe glass container was carefully inserted into thebomb, which was closed and placed in the oil bath.After the temperature of the bomb had stabilized at350”F, the bomb was shaken thoroughly a few times.This broke the glass insert, and the two solutions werecombined. The solutions were not agitated. All thetests were run for 24 hours, at which time a sampleof the clear liquid was taken and analyzed for theCa+’ concentration. An atomic absorption spectrom-eter was used for the analytical determination. Theamount of precipitation was then calculated. Similarprocedures were used with the BaSO, runs, (The cal-culated data derived from the analytical data aregiven in Tables 1 through 6.)

Some of the crystals were used for optical examina-tion under a polarization microscope. The rest of thesample was filtered, mixed with acetone, and driedin an air stream. The dry crystals were then examinedin an X-ray spectrometer.

A very common mistake is to run a precipitationtest without adjusting the impurities; for example,the NaC[ concentration, If we combine a solutioncontaining a given concentration of BaCl, with a solu-tion containing a given concentration of Na,SO,, ourfinal solution will contain a certain concentration ofNaC!, We can write the fo!!owing equation:

For example, if we combine Y ml of a solution con-taining X nicd BaClu per liter %th Y m,!Of a sO!’JtiQncontaining X mo] Na,SO, per liter, we obtain 2 Y mlof an X/2 molar BaSO, solution. which is also an Xmolar NaCl solution. If the concentration, X, of theBaCl, or Na.SO, solution is changed, we automatical-ly change also the X (NaCl concentration) of the finalsolution. However. we know that the solubilities ofBaSO, (see Templeton’ ‘1)and CaSO, (see Vetter andPhillips’) are changed by the NaCl concentration.This means that the degree of supersaturation (S) andthe concentration of saturation will change in differ-ent ways because of the definition of supersaturationi:

~=cm–c.a, . . . . . . . . (2)L“

where Cm is the initial concentration of the final solu-tion, and C,, is the equilibrium concentration atsaturation.

Since C“ changes with (he N aul concentration, S..-.

in Eq. 2 is ~iven as a function of two variables (C,,,and Co). It N more convenient, and our conclusions

about the effect of supersaturation are correct, if weadjust the NaCl concentration in the final solution toa constant value by adding a calculated amount ofNaCl to the Na,SO, solution. Then Eq. 2 becomesa simple relation of one variable (C,,,) which we variedin our experiments. In other words, S and C,,, are nowlinearly proportional. This is not true when we do notadjust the NaCl concentration but vary Cm. For ex-ample, if we were to increase the C,,, values withoutadjusting the NaCl concentration, the C,, values forgypsum would not then be constant.’ In this case, theCo would also increase, but at a different rate; i.e., S

as a function of C., would not be a linear relation,and we could not directly compare the inhibition data(Tables 4 through 6) for different Cm values. How-ever, this is done commonly by comparing the effec-tiveness of the same inhibitors, as determined bydifferent investigators using different initial concen-trations of scaling materials (Cm). This is one of themain reasons for the contradictory results reportedby different investigators.

We have found 0.5 N NaCl in the final solution aconvenient concentration. If equal volumes of CaCl,and Na,SO, solutions having equal polarities inBaCl, and Na,SO, are combined, the molar concen-tration of additional NaCl in the Na,SO, solution can

0.5 – xeasily be calculated: z molar NaCl concen-

tration in the Na,SO, solution will give us a 0.5 molarNaCl concentration in the final solution as long asthe magnitude of X stays below 0.5 mol.

The Kinetics of Crystallization

The literature” points out that the effect of inhibitorson the prevention of precipitation cannot be ex-plained by stoichiometrical considerations. The ef-fective concentrations of an inhibitor are so smallthat hindrances in crystal nucleation and growth mustplay the main role. Therefore. we must know some

0 .,,. 1 .__.:-. -~ ,.-.+,,]l;~ot;fimim tm l,n A-rctand01 Ine DaSIC KI1lCLIO UL LAywllluul”ll w w..._......-

the results and value of the precipitation test.

StoinlessSteel

o

HeavyGlrIssWell

1

-/KnockOut-Bottom

Fig, l—Pressure bomb for precipitation test atelevated temperatures.

AUGUST, 1972 999

The very complex rules of crystal nucleation andgrowth are of particular interest. Some recent publi-cations’-” describe the effect of supersaturation, het-erogeneities, temperature, and time on the nuclea-tion and growth of crystals in aqueous systems. Thenumber of nuclei formed in any phase of precipita-tion, and the pattern and velocity of crystal growth,depend on these conditions. The main driving forcefor both the nucleation and the growth is the super-saturation level, which of course changes more or lessrapidly during crystallization. For example, the effectof supematuration is so predominant that even themechanism of growth changes with the degree ofsupersaturation.’ The morphology of the precipitatedcrystals is quite different at high and low supersatu-ration because the order of the growth law changesfrom 3 to 2 when the supersaturation decreases. Thismeans that the crystal size distributions or the shapeof the crystals, is a complex function of supersatura-tion, which in itself is a function of temperature,concentration, time, and chemical composition.

Worthlessness of a single Test

The complex and only partially known crystal nuclea-tion and growth process are further complicated whenwe introduce scale inhibitors into the crystallizationsystem. To obtain an impression of the effect of in-hibitors on crystallization, we must vary the condi-tions that are controlling the crystallization process.Commonly, this is not done; only the concentrationof the inhibitor is vaned, whereas the other condi-tions — above all, the supersaturation — are notchanged. These conditions however, will change in.. mL-.-c,.... hove $9 V~Aryactual wells and fieids. ~IKICLUIG, we .,-*-

them in our tests to be able to predict how the in-hibitors will work under varying field conditions.

Effect of Supematuration

Because of the important role of supersaturation inall phases of the crystallization process, we can alsoexpect that supersaturation will have a large in.tluenceon the effectiveness of scale inhibitors. If supersatu-

InhibitorConcentration

(ppm)

0248

122050

250

InhibitorConcentration

(ppm)

0248

122050

250

InhibitorConcentration

(ppm)

0248

122050

250

TABLE I—EFFECT OF Batj04 SUPERSATURATION ON INHIBITOR EFFECTIVENESS

(Temperature: 73° F; NaCl: 0.5 mol = 2.92 percent)

Percent of Potential Precipitate Kept in Solution

Polyelect rolyte

C,)t =5:.. 10-amol C,,,= 1\ 10-3 mol

o 0100 6100 40100 70100 90100 100100 100100 100

Ester

Cn, =5x 10-~mol C,,, = 1x 10-:1mol

o100100100100100100100

0“ 60

100100100100100100

Phosphonate

Cn, =5x 10-~mol C., =lx 10-3mol

o 0100 5100 60100 100100 100100 100100 100100 100

TABLE 2—EFFECT OF CaSOd SUPERSATURATION ON INHIBITOR EFFECTIVENESS(Temperature: 210” F; NaCl: 0.5 mot = 2.92 percent)

Percent of Potential Precipitate Kept in Solution

Polyelectrolyte Ester Phosphonate

C,,, =4.5 , 10-: mol C~=8x 10-: mol C,,, =4.5X 10-~mol C., =8x 10-: mol C,n=4.5x 10-~mol C. =8x 10-~ mol

o 0 0 0 0 00 0 85 0 70 0

50 0 90 2 80 270 0 90 4 85 480 0 95 10 90 690 0 100 iO 100 g

100 0 100 20 100 15100 0 100 30 100 80

TABLE 3—EFFECT OF CaC03 SUPERSATURATION ON INHIBITOR EFFECTIVENESS(Temperature: 73°~ NaCl: 0.5 mol = 2.92 percent)

Percent of Potential Precipitate Kept in Solution

Polyelectrolyte Ester Phosphonate

C~=2 .10-3 mol

o47

100100100100100100

C., = 10-1 mol

oQ

o02

101222

C~=2x 10-3 mol Cm= 10-1mol C., =2x 1O-*mol Cm=lO-l mol

o40

100100100100100100

00012

103045

0100100100100100100100

000012

1550

1000 JOURNAL OF PETROLEUM TECHNOLOGY

ration has a large effect on the rate of nucleation, andif the inhibitor prevents crystallization by suppressingnucleation, we should find a vastly decreasing effi-ciency of the inhibitor with increasing supersatura-tion. Results of our experiments (Tables 1 through 3)show precisely this behavior. The initial supersatu-ration (S) is changed in these experiments, W-biletemperature and other conditions are kept constant.The numbers given in Tables 1 through 3 are the per-centages of scale-forming compound kept in solutionby different inhibitor concentrations. For example,-nJu w~e~~ w~=~~~~ w=k~tate kept in solution b

4 ppm inhibitor means that the inhibitor prevents 30percent of the precipitation that would otherwise haveoccurred.

Effect of Temperature

Because the effect of supersaturation is a function of~mon u other conditions, and be-the Rrnpei-ature, -... -..= . .

cause the effect of inhibition seems to be a functionof the supersaturation, we can expect a large effect

of the temperature on the inhibition. Indeed, Tables4 through 6 show this predictable temperature effect.The data given in these tables are the percents ofscale-forming compound kept in solution by inhibi-tors of varying concentrations.

We can see from Tables 4 through 6 that there is~ ~rid,ca! temapera~ure for every inhibitor. The inhi-bitor’s effectiveness decreases -with increasing tem-perature. This “temperature effectiveness” relation isdifferent for every inhibitor, and changes also withthe chemical composition of the scale. However, everyinhibitor tends to exhibit a critical temperature above. . .. . .. -4 -11 ~k Ifiw-ct criticalwhich no mhlbmon occurs iiLw x ,Ie w ..~.. ---

temperatures are shown by ester inhibitors. Underour test conditions, ester-type inhibitors frequentlyfailed to work above 250°F, whereas the other twotypes of inhibitors were still active at 350°F.

At critical temperature, the supersaturation ap-proaches zero. As the supersaturation increases, the

. --l-:-l. : L:h:+: “maximum temperature at ‘WILIW liilllULL1&. =“-..mI rc &-

creases. The supersaturation reaches a maximum

TABLE *EFFECT OF TEMPERATURE ON INHIBITOR EFFECTIVENESS (BaSO.t PREClplTATlON)

InhibitorConcentration

(ppm)

o248

122050

250

(Cm: 5 x 10-’ mol; NaCl: 0.5 mol = 2.92 percent) -

Percent of Potential Precipitate Kept in Solution

Polyelectrolyte Ester Phosphonate

73° F 210” F 350” F 73° F 210” F 350” F 73°F 210” F 350” F—— . — .

0 0 0 0 0 0 0 0 0100 80 0 100 100 0 100 100 0100 90 20 100 100 0 100 100 5100 100 40 100 100 0 100 100 9100 100 60 100 100 0 100 100 30100 100 100 100 100 0 100 100 55100 100 100 100 100 0 100 100 100100 100 100 100 100 0 100 100 100

TABLE 5-EFFECT OF TEMPERATURE ON INHIBITOR EFFECTIVENESS (CaSOd PRECIPITATION)(Cm: 0.08 mol; NaCl: 0.5 mol = 2.92 percent)

Percent of Potential Precipitate Kept in %iutiorl

InhibitorConcentration

Polyelectrolyte Ester Phosphonate

(ppm) 73° F 210” F 350” F 73° F 210” F 350” F 73° F 210” F 350” F—— . — — .

0 0— .

G G o 0 0 0 0 02 0 0 0 2 0 0 2 0 04 15 0 0 60 2 0 4 2 08 70 0 0 80 4 0 20 4 2

12 80 0 0 90 10 0 40 6 820 80 0 0 90 10 0 40 8 1050 80 0 0 90 20 0 70 15 20

250 90 0 0 90 30 0 90 80 20

TABLE 6-EFFECT OF TEMPERATURE ON INHIBITOR EFFECTIVENESS (CaC03 PRECIPITATION)(Cm: 2 x 10-’ mol; NaCl: 0.5 mol = 2.92 percent)

Percent of Potential Precipitate Kept in Solution

InhibitorConcentration

Polyelectrolyte Ester Phosphonate

(ppm) 73° F 210° F 350” F 73° F 210” F 350” F 73° F 210” F 350” F— — — .

0 0 0— —

o 0 0 0 0 0 02 47 0 0 40 7 0 100 10 74 100 0 0 100 10 0 100 20 108 100 20 0 100 15 0 100 50 35

12 100 32 10 100 25 0 100 100 3820 100 35 20 100 55 0 100 100 4250 100 40 40 100 65 0 100 100 42

250 100 100 50 100 65 0 100 100 45

AUGUST, 1972 1001

value (critical supersaturation) above which no in-hibition occurs.

Effect of Inhibitor Concentration

The expression “threshold effect” is used” to explainthe large effect of comparatively small concentrationsof inhibitors on scale formation. Quite often, we hearthe statement that certain inhibitors should be used ina certain concentration (threshold concentration) toassure “good” or “sufficient” scale inhibition. Thesestatements are not correct, as we can see in Tables 1through 6. The magnitude of inhibitor concentrationrequired to keep a given amount of scale in solutionis nor a ~oiisiaiit mid does not show a “threshold ef-fect”. This magnitude vanes and depends mainly onthree factors:

1. Supersaturation level,2. Temperature,3. Chemical composition of the scale.

The higher the supersaturation and the higher thetemperature, the larger must be the inhibitor concen-tration to prevent the precipitation of a given amountof scale per unit volume of solution. In addition, agradual increase of the inhibitor concentration grad-ually increases the amount of potential scale kept insolution per unit volume of water. Thus a 10 ppm in-hibitor, desi=~ated by the vendor to be the “requiredthreshold concentration” can result in

1. A correct treatment (well does not scale up), or2. An undertreatment (well scales up), or3, An overtreatment (waste of inhibitor).The right concentration in an actual well will de-

pend on the scaling conditions of this well. Some-times, high concentrations of 50 ppm or more arerequired to keep 100 percent of the potential scale insolution. On the other hand, we have found condi-tions (low temperature and low level of supersatura-tion) where 1 ppm of inhibitor was sui%cient to keep100 percent of the potential scale in solution.

Efiect oi inhiiiitor Coinpositkm

The results of our precipitation tests (see Tables 1through 6) suggest the following rules for practicalapplications of the inhibitors:

1.

7-.

3,

EstersBaSO, inhibition: good to excellent up to approxi-

mately 250°F at fairly small concentrations (10to 20 ppm).

CaCO,, inhibition: some fair, some totally insuffi-cient. Not advisable for temperatures higherthan 250”F.

CaSO, inhibition: good to excellent up to approxi-mately 2500F. No effect above this temperature.

PhosphonatesBaSO, inhibition: good to excellent up to 350°F

at fairly small concentrations (10 to 20 ppm).CaCO:, inhibition: good; effective up to 350”F.CaSO, inhibition: good; effective up to 350”F.

PolvelectrolvtesBaSO. inhibition: excellent; effective up to 350°F.CaCO,, inhibition: small effectiveness above 250°F.

. ,.,.-1Uw

CaSO, inhibition: not very effective above about250°F.

Questionable Value of Optical and X-ray Inspection

Up to this point, we have analyzed only the liquidphase in the precipitation test. Examining the solidphase (crystals) also allows us to draw some import-ant conclusions.

Precipitated CaCO:+ crystals are too small to ex-amine with an optical microscope; an electron micro-scope must be used to examine them. The optical mi-croscope is generally worthless for examining the pre-cipitation of BaS04 crystals, also, and for the same~~~~Q~.

In addition, all considerations regarding the relationbetween nucleation rate and crystal size (measured bymicroscopic examination) are highly questionable inthe BaSO, crystallization. This is because of the pos-sible coagulation of small crystals to larger speci-mens. r

A combination of microscope and X-ray diffrac-tometer seems to be a good tool for examining thechange of crystallization characteristics of CaSO~compounds caused by different concentrations of in-hibitors. However, interpreting the results of such anexamination can be very tricky, as shown in Fig. 2.The microscopic examination seems to indicate thatthe material shown in the center picture is ditlerentfrom that in the left-hand picture, but the same asthat in the right-hand picture. Actually the oppositeis true. The left-hand and the center pictures bothshow anhydrite (RX). And even though the crystalswere obtained in the same way, the crystal habits areentirely clifferent. On the other hand, the habits ofthe crystals in the center and the right are identical,but the CaSO, modification has been changed. Thepicture to the right represents a mixture of anhydrite(approximately 80 percent) and gypsum (approxi-mately 20 percent). Quite often we found behaviorsuch as that demonstrated in Fig. 3. The crystal size

... . ..__ :_ .L:.-.. ~ “aa” +,.”+;~“increases wltn mcreasmg mhiulwl LOMLGILLLCLLJUA. ‘UP

to a concentration of about 50 ppm, and then quitesuddenly the crystal size becomes very small again.The X-ray diffraction spectrum showed gypsum in allsix pictures. From the decrease in the precipitation(see Table 5) and the prevailing crystal size (Fig. 3),we can draw the conclusion that the crystal popula-tion (number of crystals per volume unit) decreasedconsiderably at inhibitor concentrations between Oand 50 ppm. Between 50 and 250 ppm, the popula-tion increased drastically, whereas the linear size de-creased by a factor of at least 100. This suggests anunexpected mechanism of the inhibitor effect.

.4t concentrations up to 50 ppm, the inhibitor inter-feres with the nucleation of the crystals. Above 50ppm, the inhibitor does not interfere much with thenucleation, but prevents the growth of the crystals.

Unfortunately, the reproduciiiiiity of these picturestaken through the microscope is poor. Sometimes thecrystal size in the 250-ppm experiment is larger thanshown in Fig. 2. In a few cases, a new crystal modifi-cation can be observed: bassanite (CaSO, ● 1%H, O),anhydrite, or a mixture of any two of the three CaSO,modifications. Our work on this subject continues.

JQIURNAL OF PETROLEUM TECHNOLOGY

w) INHIBITOR

Fig. 3—Preci pitation of calcium sulfate (temperature:210” F; C,,,: 0.08 mol CaSO,, 0.5 mol NaCl).

1003. ..n. Tc.-r in7-1

Compatibility of Inhibitors withOilfield EfTluents Must Not Be NeglectedUp to the present time, conventional fie”ld practicehas been to test only the positive effect of inhibitors:their action in preventing scale formation. This is notenough. Because scale inhibitors produce negative ef-fectsalso, they can cause damage inthefield. There-fore, in evaluating inhibitors, we must test thenega-tive characteristics of these compounds. There aretwo particularly costly sources of interference withnormal oilfield operations:

1. Reaction of the inhibitor with ions normallydissolved in oilfield brines.

2. Formation of emulsions if the aqueous inhibitorsolution is contacted with crude oil.

Compatibility with Water Systems

Field brines contain varying concentrations of diva-lent cations, such as Ca++,Mg’+, and Ba++.Some scaleinhibitors react with these cations to form sparinglysoluble salts. If these salts that are voluminous anddifficult to dissolve are formed inside a producing for-mation, or are constantly fed into a water-line system,the salt buildup can form a thick layer of scale: a Ca,Mg, or Ba salt of the inhibitor that looks very muchlike a true scale. For example, in one water-source.. . . .... . $-> -nn $omfl.,thrn~gh fh~Wdi, a Scak? inhibitur w~>lGu *U1lS.U.1..J ..- v

annulus to prevent the formation of a CaCO, scale inthe tubing. After some months, a heavy scale hadformed. Chemical analysis revealed, however, that itwas not the expected CaCO,, but the reaction productbetween the inhibitor and the Ca’+ ion in the water.

The formation of a precipitating Ca-sait is given bythe volubility product of this Ca-salt, which in turn isa function of temperature, inhibitor concentration,Ca-+ ion concentration, pH, and, to a small extent, theNaCl concentration,

Test Procedure. To determine the ranges in whichprecipitation of the Ca-salt occurs, we titrated manysolutions containing given concentrations of Ca++ions

t 1<~I

11ptl INHIBITOR(ON(EN7MION 100.000ppm

IOU [a++ [ON[[NIRAJION 100,000ppm I10,000ppm ‘1

--- Clm? !,000 ppm 39 - mm. [Ioudy $00 ppm 4

— pret,p,fOte )00 ppm IS10 ppm 6

8 - Oppm 7

li~j~~~j 89 I

ml 10n NoOH —

Fig. 4-Precipitation of Ca-salts in 50 ml ofpolyelectrolyte inhibitor.

1004

and inhibitors with a 10 N NaOH. A Beckman pH-meter Model 76 and Beckman pH and reference elec-trodes were used for these titrations.

Results. Some of the titration curves obtained areshown in Figs. 4 and 5. All the titration results of atypical polymer-type inhibitor (polyelectrolyte) areplotted in Fig. 6 (Ca’+ concentration vs pH) and inFig. 7 (inhibitor concentration vs pH). The curves inFigs. 6 and 7 must be interpreted so that no precipita-tion occurs under any condition to the left (or below)any given curve, Whereas precipitation must be ex-pected at conditions to the right of (or above) anycurve,

Our results indicate that the polyelectrolytes andmany of the phosphonates form sparingly soluble Caor Ba ~a]~~:These inhibitors must be used with ex-treme caution. One phosphonate inhibitor and allthe ester-type inhibitors are very compatible with thedivalent ions in the water. These latter inhibitorswould not cause any problem in this connection.

Compatibility with Oil Systems

Test Procedure. Some of the aqueous inhibitor solu-tions form stable emulsions if contacted with crudeoil. To test this tendency of inhibitors, we shake 10riii d k~rosefie with 10 ~i of the inhibitor solution.Each inhibitor is tested in six different concentrations(100, 10, 1,0.1,0.01, and 0.001 percent = 10 ppm).After shaking the two phases for approximately 2minutes, the time required for a complete separationof the two phases is observed. With a “good” inhibi-tor, it takes approximately 30 seconds for this separa-tion to occur. Some of the “bad” inhibitors requiremore than 12 days.

Results. Our results show, in general, that the poly-electrolytes and most of the phosphonates form solu-tions that do not emulsify with kerosene, whereasmany of the esters are bad offenders. This means thatmost esters can offer serious problems in the field,whereas polyelectro~ytes and most phosphonates do

14r

INHIBITORCONCINIRAIION: 100,000 ppm [11I0,000ppm (8)1,000 ppm (9)

cosccux hazy

12500 ppm (10)

--- clear 100 ppm (11)— precipimta 50 Qpm (11

10 ppm (13)1 ppm ([4)O ppm (151

Hl;;3~;;; ;9 I

ml 10n No OH —

Fig. 5—Precipitation of Ca-salts in 50 ml ofpolyelectrolyte inhibitor.

JOURNAL OF PETROLEUM TECHNOLOGY

not cause Droblems.Actuall~, the test should be run with crude oil in-

stead of kerosene. We know from experience thatsometimes emulsions are formed when cm”de oil isused whereas none were observed when kerosene wasthe organic phase in the test. Economically, there isno way to test all scale inhibitors with all oils. There-fore, we place two restrictions on the test:

1. If an inhibitor shows emulsification with kero-sene, we can expect it to form emulsions with the oil.This inhibitor is not used in the field if emulsificationis a problem.

2. If an inhibitor does not form an emulsion withkerosene, it still could offer a problem with the crudeoil. Therefore, the suggested inhibitor must be testedwith the oil of the well in which the inhibitor is used(on-site test).

Ranking of Inhibitors Is Valid for OnlyA Given Set of Conditions

: L:L:._ h., tPQtin O t~.e~. with the de-We evaluate Inl,lultu.a “Y .W. ....=~rib~d procedures. They are then ranked on the basisof a cost-to-performance ratio.

From the test results, we could precisely predict theperformance of the inhibitors in the field if we couldreproduce field conditions in our laboratory tests.This, of course, we cannot do. Field conditions in awell may change unpredictably; changes may evenoccur without our knowledge. Since inhibitor per-formance depends on these conditions, we try to bringour test conditions as close as possible to the field con-ditions. Inhibitors are then ranked for this set of testconditions. The inhibitor that looks most promising inthe laboratory test is then tried in a field test. If the in-hibitor fails in the field or shows any unexpected be-havior, we can only conclude that our field informa-tion was inaccurate. Unfortunately, a change of theconditions can change our ranking of inhibitors. Twodifferent wells in the same field can have scaling con-ditions that are different enough to require two differ-ent inhibitors. For these reasons, we cannot make ageneral ranking of inhibitors.

100,000—— 1000ppmINHIBIToR

10,000–

tppm [a’+ 4

,/

1,000.–-. -. !OOOO?Q~N!!!lO!-_

\

Precipitation-. \

\—— NOPrecipitation

\I I ( 1 I2 4 6 6 10 12 14

pH w

Fig. &Precipitation of Ca-salts of polyelectrolyteinhibitor as function of pH and Ca concentrations.

After testing a great number of inhibitors, we cantell in many instances from the behavior of the in-hibitor in the field which conditions were misjudgedbefore a specific inhibitor was applied. Sometimes thescaling conditions of a well are so severe that none ofthe commercially available inhibitors can fully handlethe problem. In this case, even the inhibitor rankedfirst will fail in the field.

What Remains To Be Done?The precipitation and compatibility tests describedabove are the main parts of our inhibitor evaluation.There are, of course, more tests required to predictthe complicated behavior of an inhibitor under allfield conditions. Our evaluation procedure involvesalso more elaborate tests to allow a careful examina-tion of the following inhibitor properties.

1. Chemical composition (qualitative and quanti-tative),

2. Adsorption-desorption characteristics on differ-ent rock materials,

3. Thermal stability,C-. . . ..a”mcca”mcc4. Lullu>lvtillb..,

5. Interaction with other oilfield chemicals.Our work on these tests is still in progress. It wouldlead beyond the scope of this paper to describe thetests here.

ConclusionsThe following paragraphs summarize our conclusionsabout test procedures and the effectiveness of pre-sently available inhibitors:

1. As to the chemical classes of inhibitors, poly-elecirolytes (polymers) are highly effective in BaSOA,less so in CaCO,, and even less in CaSO, inhibition.Esters are highly effective in CaSO, and somewhatless in BaS04 inhibition; they are below average inCaCO, inhibition. The esters are ineffective at highertemperatures. Phosphorzates are the best inhibitors inCaCO, inhibition. They are also good in high-temperature inhibition of BaSO, and CaSO,.

2. Precipitation tests are the most suitable method

100,000/

t 10,000=~ So,ooo~ 100= 1,000E=

100Precipitation —

- No Precipilolion

10 I , ,2 4 6 8 10 12 1

pH —

Fig. 7—Precipitation of Ca-salts of polyelectrolyteinhibitor as function of pH and inhibitor concentrations.

I

AUGUST, 1972 1005

,.

for measuring the effectiveness of inhibitors for fielduse.

3. A rating of inhibitors in fhe order .of their ef-fectiveness is only relative and depends upon test con-ditions. The most important conditions are the degreeof supersaturation and the temperature; the higher thesupersaturation and the higher the temperature, theless effective a scale inhibitor becomes.

4. Precipitation tests show limits of conditionswithin which the inhibitor will work. There is a cri-tical temperature (at which the supersaturation ap-proaches zero) and a critical supersaturation. No in-hibitor is effective at values greater than these twocritical values.

5. Critical temperatures make scale inhibition ex-tremely dificult in high-temperature wells. Betweenapproximately 250°F and 350°F, no compiete itihi-

.-. -+- ~. J a crime u-alinu can ~akebition can & giiitl dlld%> L.u., w...- .--..-..=

place despite the presence of an inhibitor. No inhi-bition can be expected at temperatures above ap-proximately 350°F.

6, There is a maximum temperature (below crit-ical temperature) associated with any supersatura-tion. No inhibition can be achieved above tinese maxi-mum temperatures. As the supersaturation increasesfrom zero to critical supersaturation the maximumtemperature decreases from critical temperature to aminimum value.

7. The concept of “threshold” concentrations ofinhibitor in produced water, as used in current fieldpractice, is not valid. Application of threshold con-centrations can mean overtreatment or undertreat-ment, depending on the physical or chemical condi-tions at any given location within the system (field or

well).8. Microscopic investigation of crystal properties

can result in misleading conclusions. Determinationof crystal habits in the presence of scale inhibitorsshould include at least a combination of microscopeand X-ray spectrometer. Even then, many of the dataare not very reproducible. Much more work must bedone in this field.

9. For practical inhibitor evaluation, it is import-ant to test the emulsifying properties and compatibil-ity with formation fluids. Severe fortnation damage

., -IL., :m+~---+~mnf inhihitnr ~nll]fion withcan De miiseu UY UILGLa&.LoL.” . . . . .. U..=. --- ._

brine or oil.

Original manuscript received in Society of Petroleum Engineersoffice SePt. 17, 1970. Revised manuscript received March 7, 1972.Paper (SPE 3188) was first presented at SPE 41st Annual Califor-nia Regional Fall Meeting, held in Santa Barbara, Nov. 28-30, 1970.~ Coti, rj@ 197z America” Institute of Mining, Metallurgical, and

Petroleum Engineers, Inc.

10, There is no ideal inhibitor. All the inhibitorshave shortcomings that make them applicable in spe-~jfic ~ipda~on~ ~niy. The ~eiect~~n of an inhibitor is

almost always a compromise between its positive andnegative characteristics.

11, The wide variety of conditions in field appli-cations dictates a variety of scale inhibitors.

12. To apply laboratory results properly in thefield, accurate field information is vital. Field infor-mation is often to difficult to obtain and is thus in-accessible. In these instances, inhibitor selection bylaboratory test may prove to be “wrong” because thetest has been based on “wrong” conditions.

AcknowledgmentsI wish to acknowledge with many thanks the work oflx 1 c....+t ~.d G T H~nk, who performed theIvl. ~. Val LWLL L4’. -. -.

laboratory tests; R. A. Rogers, who provided valuablehelp with the manuscript; R. C. Phillips and C. R.Pearson, who supervised most of the field work; andthe Union Oil Co. of California for permission topublish this paper.

References

1. Bsharah, L.: “Test Unit Evaluates Scale Inhibition” Oiland Gas J. ( 1969) 166-170.

2.Jones, L. W.: “Development of a Mineral Scale Inhibi-tor,” Corrosion ( 1961) 17, 110-114.

3. Weintritt, D. J. and Cowan, C. J.: ‘~unique characterk-

tics of Barium Sulfate Scale Deposltlon,” J. Pet. Tech.(Oct. 1967) 1381-1394.

4.Bradley, B. W.: “Environment Simulator for Oil WellScale Inhibition,” U. S. Patent 3,470,735 ( 1969).

5. Vetter, O. J. G. and Phillips, R. C., Jr.: “Prediction ofthe Deposition of Calcium Sulfate Scale Under Down-hole Conditions,” J. Pet, Tech. (Oct., 1970) 1299-1308.

6. Ralston, P. H.: “Scale Control with Aminomethylene-phosphonates,” J. Pet. Tech. (Aug., 1969) 1029-1036.

7. Doremusj R. H.: “crystallization of Slightly SolubleSalts from Solution,” J. Phys. Chem. ( 1970) 74, 1405-1408.

8. Randolph, A. D.: “How to Approach the Problems ofCrystallization,” Chem. ,??ng. ( 1970) 80-96.

9. Doremus, R. H.: “Precipitation Kinetics of Ionic Saltsfrom Solution,” 1. F%ys. Chem. (1958) 62, 1068-1075.

10. Cruft, E. F. and Chaeo, P. C.: “Nucleation Kinetics ofthe Gypsum-Anhydrite System,” Third Symposmm onSalt, The Northern Ohio Geol. Sot., Inc., Cleveland(1970) 1.109-117.

11. Nancollas, G. H. and Purdie, N.: “Crystallization of Ba-a-------- :- AJII,-m,c ‘&ml,,tion “

rltirii CMplliLLG 111 A.IWWUU. u.-.. -.., Trans., FaradaySOC. ( 1963) 59, 735-740.

12. Turnbull, D.: “The Kinetics of Precipitation of BariumSulfate from Aqueous Solution,” Ada Metalhmgica,( 1963) 1.684-691.

13. Templeton, C. C.: “Volubility of Barium Sulfate in So-~:m Chloride Solutions From 2500 to 9S 0C,” J. Chem.

n .- t~ocn~ c <1A-51AJHIg.Data {,7””, 4, J,-TJ.”. J_PT

1006 JOURNAL OF PETROLEUM TECHNOLOGY