Behavior of Calcium salts at Boiler Temperatures - American Chemical

much stress has been laid by Hall (Í), Partridge (£), and others, upon theratio of carbonate to sulfate for the preven- tion of scale. Calcium sulfa...
0 downloads 0 Views 754KB Size
LAROEMODERN CENTRAL POWER PLANT

Behavior of Calcium Salts at Boiler Temperatures FREDERICK G. STRAUR, Engineering Experiment Station, University of Illinois, Urbana, Ill.

T

HE trend of the desigii

bonate in a laboratory boiler. From these results they ttf modern steam plants is toward higher pressures and higher rating. Since arrived a t the conclusion that a t 185O C. the ratio of carhoncapital investments are also increasing, the question ate to sulfate was not greater than 0.01. Thus, it appears evident that, in order t.o draw definite conof "boiler outage" or the amount of t.ime a boiler is out of service becomes very important. In attempting to explain clusions as to the cause of scale fonnation and its prevention, the cause of scale fonnation and methods of preventing it, more data are desirable as to what are the stable solid and much stress has been laid by Hall (f), Partridge (Z), and liquid phases of the various salts encountered in boiler waten others, upon the ratio of carbonate to sulfate for the preven- a t boiler temperatures. The author liss run a series of tests tion of scale. Calcium sulfate forms a dense adherent scale involving the calcium salts such as carlionate, sulfate, phoson the heating &faces, while calcium carbonate does not phate, silicate, and hydroxide (separately and combined), in usually form a n adherent scale in boiler waters having an ap- order to determine the several equilibria itwolved. preciable hydroxide alkalinity. The earlier explanation TBs1-s ON CALCIGM SALTS offered by Hall was that calcium sulfate, having a retrograde AwAaa.rus. The equipment used in making these tests solubility (decreasing with increase in temperature), became less soluble at the heating surface and came out of solution as has been reported in detail (4,6). A photograph and diagram scale. Calcium carbonate, according to the small amount of of the steel bomb used for the tests have already been reprodata available, was assumed to be more soluhle with increase duced (6). The solid and solution were put into the lower in temperature. Thus, it would remain in solution at the bomb. The completely assembled unit was heated in a conheating surface and be thrown out of solution in the main stant-temperature air bath. When the lower bomb had been body of the boiler and form sludge. Working on this assump- held a t this temperature for the desired period of time, the tion, Hall determined the solubility of calcium sulfate in water valve between the bombs was opened and part of the liquid at temperatures up to 200" C. and calculated the soln~ility in the lower bomb was forced through the filter and the conproduct at these temperatures. IIe made a rough calculation necting capillary steel tube into the top bomb. The valve of the solubility product of calcium carbonate from data avail- was then closed, the unit removed from the heating chamable a t lower temperatures. The relation between the two ber, cooled, and the volume of the sample collected in the solubility products was taken as the amount of carbonate ion top bomb measured. necessary to cause preferential precipitation of calcium carMEASUKEMENT OF YOLUNEOF SAMPLE. The volnnie of bonate. Ball gave this figure as 0.0883 for the ratio of car- each top bomb was determined by using a standard top havbonate to sulfate at 185" C., and stated that at higher tem- ing a glass tube in it. The amount of water necessary to fill the bomb to the mark on the tube was determined for each peratures this figure would increase. Partridge (Z) pointed out that calcium carbonate undoubt- bomb. When the bomb was removed after sampling, the edly became less soluble with increase in temperature. The standard top wa8 put on and redistilled water, free from author (4,6) reported the results of tests run on the solu- carbon dioxide, was added through the glass tube from a bility of calcium carbonate at temperatures between 182" arid buret. The amount of water added, deducted from the total 316" C. and showed that the solubility decreases with in- volume of the bomb, gave the volume of the sample. Thii crease in temperature. Partridge, Schroeder, and Adams (.i)volume could be determined within 0.2 cc. with a total volume calculated the ratio of carbonate to sulfate, employing a rela- of about 250 cc., thus giving a possible error of about 0.1 per tion between the stoichiometric solubility product determined cent in measuring the volume of the liquid. directly and correlated by means of the ionic strength as a The bomb was then rinsed out three times with redistilled measure of the total ionic strength. They also determined water, free from carbon dioxide, into a 500-cc. Erlenmeyer the ratio roughly by preferential precipitation of calcium car- flask. If the hydroxide content was to be determined, the 113 4

October, 1932

INDUSTRIAL AND ENGINEERING CHEMISTRY

1175

solution was transferred to a 500-cc. volumetric flask and temperature baths. Thus, a set of bombs could be sampled ; diluted to volume with the purified water. One portion was removed from the bath a t 8 A. M.; cleaned, charged, and rethen pipetted out to be run for hydroxide and sulfate, and the turned by 3 p. M.; and be up to the desired temperature by remaining portion run for carbonate and calcium. The size 8 P. M . In this way every fourth day was found to be a conof the samples to be used depended on the approximate venient sampling time. amount of the material present. Thus, with a low sulfate HYDROXIDEDETERMINATION. The hydroxide was decontent, 250-cc. samples were used for hydroxide and sulfate termined by pipetting the solution into a 500-cc. Erlenmeyer and with a high sulfate content only 100 cc. were used. If flask, adding 25 cc. of a standard solution of 0.02 N sodium the hydroxide content was not to be determined, all of the hydroxide containing about 10 per cent barium chloride; sample was run for carbonate and sulfate. after allowing the mixture to stand stoppered for about 2 PROCEDURE. Since six bombs were run a t one time in the hours, the solution was titrated with 0.02 N hydrochloric constantrtemperature boxes, it was found advantageous to acid to the phenolphthalein end point (colorless). A blank run the six sanides through the analytical procedure to- was run using the same volume of carbon dioxide-free water gether. When the solutions ;ere as that contained in the sample prepared before adding to the being tested. The difference b e bombs, a blank using the same tween the quantity of acid used amount of the carbon dioxideThe reaction CaSO, NazCOa CaCOa for the blank and that used for free redistilled water as was used the sample gave a direct measure Na2S04 has been investigated at temperafor the bombs was put in a 500-cc. of the hydroxide content. Tests tures between 182" and 282" C. The effect Erlenmeyer flask, and the same conducted by the author to check of the preseme of sodium hydroxide and sodium solution added as was added to the accuracy of this method (6) one of the bombs. The flask was for the determination of hydroxchloride combined in the liquid phase has also stoppered and set aside to be run ide showed that the possible error been studied. Owing to the change in equilibthrough the analytical procedure was about 3 p. p. m. as sodium rium between the ions in the liquid phase at the with the samples obt,ained from hydroxide. elevated temperatures and room temperature the bombs after they had been CARBONATED ET E R M I N A where the solutions are analyzed, it has not been held a t the d e s i r e d t e m p e r a TION. The sample to be tested ture for the proper period of was transferred to a 1000-cc. possible to delermine the exact equilibrium. time. This b l a n k s e r v e d to E r l e n m e y e r flask, previously However, data are presented which indicate the check the concentration of the swept clean with carbon dioxideliquid-phase concentrations measured at room s o l u t i o n in the b o m b a t the f r e e a i r . The flask was contemperature which will cause the precipitation start, as well as the analytical nected with a reflux condenser. of either calcium sutfate or calcium carbonate at procedure. A stream of c a r b o n dioxideThe solids added to the bombs free air was passed through the the elevated temperatures. were either c a l c i u m s u l f a t e flask, through the c o n d e n s e r , or calcium c a r b o n a t e . The through granulated zinc, through calcium sulfate was added in c a l c i u m c h l o r i d e , and finally previous tests both as CaSOl and CaS04.2H20,and the result- through anhydrone. The dry air was then passed through a ing solubilities were found to be identical, apparently irre- weighed Wesson bulb containing ascarite and anhydrone. spective of which form was used. For the tests being re- The entire train was swept free from carbon dioxide before ported, CaSO, was added as the solid. When calcium car- adding the flask containing the sample. After the flask was bonate was added, ground crystals of Iceland spar were used; added, dilute hydrochloric acid (1: 1) was added by means of the latter was of standardization quality. an adding funnel put through the rubber stopper; the solution The chemicals to be added in solution were made up in was heated to boiling and boiled slowly, with the air being about 0.1 molal solutions by adding the chemicals to carbon passed through slowly. After about 30 minutes the Wesson dioxide-free redistilled water. The sodium hydroxide solu- bulb was removed and weighed, the difference in weight being tion was stored in a paraffin-lined bottle. a measure of the carbon dioxide present in the solution. The water was obtained by distillation after adding a small Weighed amounts of pure sodium carbonate were put in soluamount of sulfuric acid and potassium permanganate to dis- tion in carbon dioxide-free water, and the carbon dioxide tilled water and collecting the middle 60 per cent of the dis- content tested by this method. It was found that the method tillate. The redistilled water was washed free from carbon was accurate to within one p. p. m. of carbonate (0.02 millidioxide by bubbling carbon dioxjde-free air through it for 24 mole per liter) when a 250-cc. sample was used. hours. Tests run on liter samples of this water showed the SULFATEDETERMINATION. The solution left after the carbon dioxide to be not more than 2 parts per million. hydroxide determination was made distinctly acid to methyl The desired amounts of the concentrated chemicals were orange by adding a few drops of concentrated hydrochloric calculated, and the total volume deducted from the volume acid. It was then boiled down to about 75 cc. volume, alto be added to the bombs (usually 400 cc.). This amount of lowed to stand for 12 hours, filtered, washed, ignited, and redistilled carbon dioxidefree water was added to a 500-cc. weighed in tared crucibles as barium sulfate. It was thought Erlenmeyer flask and heated until boiling just started. The that, since the barium sulfate was precipitated by the addition necessary volumes of solutions of the chemicals were added, of an alkaline barium chloride solution in the cold, the preand the hot solution was poured into the clean dry bomb con- cipitate might occlude other materials. However, tests run taining about 50 millimoles of the desired solid. The bomb by this method, using standard solutions of sodium sulfate, was immediately closed. gave results for the sulfate content within 2 p. p. m. of sulfate The bombs were then put in the constant-temperature air (0.02 millimole per liter) of those obtained in a volume of 250 baths, brought up to temperature within 5 hours, and held a t cc. in which the barium sulfate was precipitated from a hot, constant temperature for 84 hours. Tests run a t 182" C. slightly acid solution. gave the same results after 36 hours a t this temperature as CALCIUM DETERMINATION. The calcium was determined when held there 84 hours. However, the &-hour period was after the carbonate content determination. The acid soluused since it fitted in with the operation of four constant- tion left after the carbon dioxide had been removed was boiled

+.

+

A

INDUSTRIAL AND ENGINEERING CHEMISTKY

1176

down to about 75 cc. volume, made slightly alkaline with ammonium hydroxide, and then just acid to methyl orange indicator with hydrochloric acid. About 20 cc. of a 4 per cent solution of ammonium oxalate were added to the boiling solution; after a few minutes of boiling, the solution was made slightly alkaline with ammonium hydroxide and boiled for 15 minutes. After cooling, the solution was made distinctly alkaline with ammonium hydroxide, allowed to stand 12 hours, filtered, and washed with distilled water, slightly alkaline with ammonium hydroxide, until free from ammonium oxalate. The precipitate was dissolved in hot sulfuric acid solution (about 25 cc. concentrated sulfuric acid to 1 liter of water), heated to 80" C., and titrated with 0.05 N potassium permanganate solution. The potassium permanganate solution was standardized a t regular intervals against sodium oxalate obtained from the Bureau of Standards. If much iron was present, it was removed as ferric hydroxide prior to running the calcium. If the volume of the ferric hydroxide was appreciable, it was dissolved in hydrochloric acid and reprecipitated, and the filtrate added to the first filtrate and run for calcium. The calcium was determined with an error of *0.2 milligram. Thus, on a 250-cc. sample the error would be * 0.02 millimole per liter.

DISCUSSION OF RESULTS The first system studied, the results of which are given in Table I, involves a study of the reaction: CaCOa Na2S04 CaS04 Na2C03

+

TEMP.

c.

182

207

244

282

ADDED

--Millimoles

0.48 1.20 1.20 3.60 12.00 18.00 0.24 0.48 1.20 3.60 12.00 18.00 0.48 1.20 3.60 12.00 18.00 3.60 12.00 18.00

So4 ATEND OF TEST

TABLE 11. SULFATEAND CARBONATE CONTENT OF SOLUTIONS OF

SODIUM SULFATE HEATEDIN CONTACT WITH SOLIDCALCIUH CARBONATE so4 cos

TEMP.

c.

182

207

p e r liter-

1.07 1.63 1.56 3.28 11.50 16.80 0.43 0.50 1.20 3.25 11.90 17.05 0.23 0.71 2.78 10.40 15.40 2.62 10.75 16.00

244

Cos

ATEND OF TEST

0.39 0.51 0.55 0.73 1.12 1.21 0.52 0.67 0.75 0.90 1.05 1.40 0.34 0.90 1.63 2.33 3.73 1.61 3.83 3.60

RATIO

cos:so4

Millimoles P. p . m. 0.36 0.22 0.31 0.19 0.35 0.22 0.22 0.14 0.09 0.06 0.07 0.05 121 0.79 0.84 1.34 0.39 0.62 0.17 0.27 0.09 0.06 0.08 0.05 1.47 0.92 1.27 0.79 0.59 0.37 0.22 0.14 0.24 0.15 0.62 0.39 0.22 0.35 0.14 0.22

SUMMARY OF TESTS. Of the various equilibria studied, the following appears to be of particular interest when considering the cause of sulfate scale formation:

If the solubility of calcium sulfate and calcium carbonate is known, one should be able to calculate the ratio of carbonate to sulfate which is necessary to cause precipitation of calcium carbonate in preference to calcium sulfate. However, such data depend not only upon the solubility of calcium sulfate and calcium carbonate alone, but in solutions of ionic strength comparable to that in boiler waters, and also with a mixed solid. Tests have been run under the following conditions a t the beginning: Solid, CaSO4; in solution Na&O3 Solid, CaCOa; in solution NsSO4 Solid, CaS04; in solution, NaCl, NaOH, Na&O, Solid, CaCOa; in solution NaCI, NaOH, Naps04 Complete analyses were made of the solutions removed when equilibrium had apparently been established. The results obtained from a study of these two systems, amroached from both sides of the equilibrium, are given in Tibles I to V.

+

starting from the left and proceeding to the right. The results show sulfate and carbonate in millimoles per liter, both a t the start and when apparent equilibrium had been reached. The ratio of carbonate to sulfate is also calculated both on the basis of millimoles per liter and parts per million. This ratio does not reach a constant but is continually.decreasing as the concentration of sulfate increases.

TABLE I. SULFATE. AND CARBONATE CONTENT OF SOLUTIONS OF SODIUMCARBONATE HEATED IN CONTACT WITH SOLIDCALCIUM SULFATE NaaCOa

VOl. 2-1, No. 10

282

NazSOI ADDED

E N D AT E N D RATIO TEST O F T E B T COX:Sod IMzllimoles per literMillimoles P . p . m.

0.25 0.50 1.25 3.75 12.50 18.75 0.25 0.50 1.25 3.75 12.50 18.75 0.50 1.25 3.75 12.50 18.75 3.75 12.80 18.75

AT OF

0.19 0.48 1.40 3.67 12.2 17.4 0.16 0.39 1.06 3.74 11.10 18.36 0.46 0.98 2.85 11.38 17.23 2.96 10.46 17.02

0.55 0.56 0.73 1.04 1.30 1.22 0.55 0.65 0.74 0.82 1.31 1.44 0.66 0.78 1.74 1.52 1.60 1.47 1.56 3.94

2.90 1.17 0.52 0.28 0.10 0.07 3.43 1.67 0.70 0.22 0.12 0.08 1.43 0.80 0.61 0.13 0.09 0.49 0.15 0.23

1.81 0.73 0.32 0.17 0.06 0.04 2.14 1.05 0.44 0.14 0.08 0.05 0.89 0.50 0.38 0.08 0.06 0.30 0.09 0.14

Table I1 gives the result of equilibrium tests of the reaction: CaC03

+ NapS04tr Cas04 + Na2CO8

Starting with calcium carbonate and adding sodium sulfate, the sulfate is all retained in the solution at the lower temperatures, and over 90 per cent remains even a t the higher temperatures. This would indicate that the solids contained both calcium carbonate and calcium sulfate. I n these tests the ratio of carbonate to sulfate was not constant but decreased with increase in the sulfate content. At the higher sulfate concentrations, however, the ratios agree fairly well with those reported in Table I. The results of the tests run using solid calcium sulfate and a solution containing sodium chloride, sodium hydroxide, and sodium carbonate a t the start are given in Table 111. The sodium chloride was 2.5 millimoles per liter, and the sodium hydroxide 11.5 millimoles per liter a t the start of all tests. These results are very interesting in that they show the carbonate to remain almost constant a t all concentrations and at all temperatures, with the'exception of 282" C. where the carbonate increases in the last two tests reported. This may be due to the coating of the calcium sulfate with calcium carbonate, where an excess of sodium carbonate was obtained. The remaining hydroxide varies to some extent and may be influenced by the variation in the amount of solid a t the start since this was not constant. The almost constant value of the reported carbonate is undoubtedly due to the fact that in the presence of sodium hydroxide another equilibrium is established which is apparently independent of the one under consideration. Thus, if sodium carbonate is present in solution a t these temperatures and corresponding steam pressures, the following reaction may take place: NaHC08 3NaOH COZ 2Na&03 2H20

+

+

+

When a samole of the solution is analvzed at room temDerature, the carbonAe is reported, whereas it undoubtedly ixisted as

IN DUSTR IAL AN D EN GINEER IN G CHEMISTK Y

October, 1932

1177

acid carbonate a t the elevated temperature. The acid carbonate present a t these higher temperatures would depend upon the partial pressure of the carbon dioxide and the hydroxide concentration. Being present as acid carbonate, no reaction would take place between it and the calcium sulfate. Under these conditions the reported carbonate, having been acid carbonate a t the higher temperatures, would remain constant and be almost independent of the sulfate ooncentration.

carbonate and calcium sulfate in the presence of the various ions and in solutions of ionic strengths similar to those encountered. Thus, in the presence of sodium hydroxide the solubility of calcium carbonate becomes less than 0.0 millimole, while the solubility of calcium sulfate is 0.90 millimole a t 182' C. and 0.23 a t 316' C. (5). Under these conditions the presence of a ratio of carbonate to sulfate a t the temperatures under consideration of greater than approximately 0.01 (3) should cause precipitation of the carbonate in preference to the sulfate. TABLE111. SULFATE, CARBONATE, AND HYDROXIDE CONTENT If a solution of sodium hydroxide is added to solid calcium OF SOLUTIONS CONTAINING SODIUM CARBONATE, HYDROXIDE,carbonate and heated to the temperatures under consideraAND CHLORIDE, HEATEDIN CONTACTWITH SOLIDCALCIUM tion, the ratio of carbonate to sulfate will be infinite since the SCLFATF sulfate equals zero. There will be some acid carbonate in (NaC1 added, 2.5 millimoles per liter in,all tests: NaOH added, 11.5 millimoles per liter in all tests) solution, and the equilibrium will be: sn. fyn. nu A T E N D A T E X D A; END RATIO TEST OF TEST OF TEST Cos: SO4 .Millimoles wer liter Millimoles P. w. m. --/

TEXF.

c.

182

207

244

282

NazCOs ADDED r

0.44 0.44 0.68 0.68 1.40 3.80 3.80 12.00 12.00 18.00 0.44 0.68 1.40 3.80 12.00 12.00 18.00 0.44 0.68 1.40 1.40 3.60 12.00 12.00 18.00 0.44 0.44 0.68 1.40 1.40 3.80 12.00 18.00 18.00

.I-o

CaCOs

OF

'__7

1.93 1.71 1.77 2.25 2.82 4.88 4.68 11.70 12.78 22.40 1.45 1.92 2.80 4.60 12.30 12.62 18.05 1.35 1.71 2.50 2.61 4.44 12.72 12.82 19.85 0.85 0.88 1.47 2.20 2.61 4.83 12.75 19.30 18.15

0.29 0.43 0.41 0.44 0.37 0.33 0.27 0.40 0.46 0.78 0.42 0.28 0.59 0.54 0.41 0.73 0.54 0.31 0.41 0.94 0.40 1.05 0.47 0.66 0.85 0.22 0.46 0.37 0.27 0.61 0.57 0.66 3.78 2.33

4.7 5.9 5.2 4.5 4.0 3.2 3.1 3.9 3.6 2.6 5.3 6.0 5.5 4.60 2.88 5.33 2.9 3.3 3.5 4.0 3.0 2.15 2.8 2.7 3.76 3.2 4.0 2.8 2.6 3.5 2.88 2.6 4.0 3.5

+ 2NaOH

Na&Os

+ Ca(OH)2

If sodium sulfate is added in sufficient amount to make the ratio of carbonate to sulfate less than 0.01, calcium sulfate should precipitate. The carbonate content is high because of the reaction between the sodium hydroxide and the calcium carbonate. Before the ratio of carbonate to sulfate will be less than 0.01, it is necessary to have the sulfate at least 30 millimoles per liter (or 5200 parts per million sodium sulfate), which is much higher than the sulfate content of the tests re ported and explains why no calcium sulfate was formed in the tests reported in Table V. TABLEIV. CALCIUM, CAFLBONATE, AND HYDROXIDE CONTENT OF SOLUTIONS CONTAINXNQ SODIUMHYDROXIDE HEATEDIN CONTACT WITH SOLIDCALCIUM CARBONATE (COS in solution without any solid phase, 0.20 millimole per liter) OH Ca AT END COS AT END OH A T END

TEMP.

0.26 0.52 0.25 0.12 0.24 0.12 0.05 0.19 0.12

0.16 0.32 0.16 0.08 0.15 0.08 0.03 0.12 0.08

When a solid calcium carbonate is in contact with a sohtion of sodium hydroxide the solubility of calcium is less than or 0.0 millimole per liter as shown in Table IV. 0.0 p. p. The sodium hydroxide might react with the calcium carbonate bicarbonate, ~~~i~ to form calcium hydroxide and the sodium bicarbonate would be controlled by the partial pressure of the carbon dioxide. The results obtained in Table IV indicate that this is possible, The calcium solubility is 0.0 millimole; the hydroxide concentration has been reduced; and the carbonate varies to a large extent. The addition of sodium sulfate to this system should have no influence on the carbonate content if no calcium sulfate existed or was formed as a solid phase. However, if calcium sulfate was formed, then the carbonate should drop to the lower value reported in Table 111. The fact that the sulfate remaining a t the end is the same as that added, except in the very low concentrations where occlusion of the sodium sulfate by the calcium carbonate or reaction with the iron container would readily explain the small losses, would indicate that no calcium sulfate existed as a solid. Under these conditions the results reported in Table V should not check those in Table 111. The solids in the tests reported in Table V were undoubtedly basic calcium carbonate as well as calcium hydroxide. If calcium hydroxide alone had been formed, there would have been a much larger increase in carbonate content. This indicates that a basic calcium carbonate was undoubtedly formed. I n order to predict the carbonate ion necessary to cause preferential precipitation of calcium carbonate, it would be necessary to determine the relative solubility of the calcium

ADDED

TEST OF TEST Millimoles pe7 litw

OF

c.

OF

TEST

TABLEV. SULFATE, CARBONATE, AND HYDROXIDE CONTENT OF SOLUTIONS CONTAINING SODIUMSULFATE,HYDROXIDE, AND CHLORIDE HEATEDIN CONTACT WITH SOLID CALCIUM CARBONATE (NaCI, 2.5 millimoles per liter in all tests. NaOH, 11.5 millimoles per liter in a U t e s t b 504 cos OH

so4

TEMP.

C. IS2

207

244

282

ADDED

A T E N D AT END

OF

-MMzllimoles 0.25 0.50 0.50 1.25 1.25 3.75 3.75 12.50 18.76 0.26 0.26 0.50 1.25 1.25 3.75 12.50 18.75 0.25 0.25 0.50 1.25 3.75 12.5 18.76 0.25 0.25 0.50 1.25 3.75 12.50 18.75 18.75

0.32 0.48 0.48 1.32 1.28 3.96 3.86 8.95 18.8 0.12 0.13 0.16 0.18 1.40 .3.70 13.0 18.65 0.12 0.15 0.18 1.86 4.75 13.5 19.4 0.09 0.11 0.13 1.96 2.81 13.8 21.4 19.6

OF

A T ~ N D OF TEeT

per lite7-T

0.62 0.66 0.96 1.08 1.09 1.19 1.00 1.27 1.28 0.20 0.26 0.18 0.61 1.09 0.70 0.70 1.21 0.55 0.41 0.32 1.58 0.94 0.38 2.24 0.86 0.82 0.70 1.93 1.12 2.30 2.78 1.33

6.6 5.1 6.0 5.5 6.5 4.8 6.5 7.0 5.8 6.9 6.6 4.9 6.2 6.8 6.0 5.0 6.5 6.7 6.5 7.2 4.8 5.5 6.3 6.0

4.6 4.0 3.7 1.1 2.5 2.8 3.2 2.5

RATIO

cos:804 Millimolea P. p . m. 1.94 1.21 1.37 0.85 2.00 1.25 0.82 0.51 0.85 0.53 0.30 0.19 0.26 0.16 0.14 0.09 0.068 0.04 1.67 1.04 2.00 1.25 1.12 0.70 3.38 2.10 0.78 0.49 0.19 0.12 0.054 0.03 0.065 0.04 4.58 2.86 2.74 1.71 1.78 1.11 0.85 0.53 0.20 0.12 0.028 0.02 0.12 0.08 9.55 5.95 7.45 4.65 5.40 3.36 0.9s 0.61 0.40 0.25 0.16 0.10 0.13 0.08 0.068 0.04

1178

INDUSTRIAL AND ENGINEERING CHEMISTRY

The results in Table I11 show that the reported carbonate is independent of the sulfate concentration up to a t least 22 millimoles per liter of sulfate. Thus, if solid calcium sulfate is in contact with a solution containing sodium hydroxide, the introduction of sodium carbonate into the solution forms carbonate ions which, because of the low sulfate content from the calcium sulfate solubility, .give a carbonate concentration sufficient to precipitate calcium carbonate, which reduces the carbonate. As long as solid calcium sulfate exists, all of the sodium carbonate added will react to precipitate calcium carbonate and form soluble sodium sulfate until the sulfate ccxentration becomes high enough to require an appreciable carbonate increase. However, this concentration is higher than 22 millimoles of sulfate (about 3000 p. p. m. sodium sulfate) Since the carbonate reported is the total of the acid carbonate and carbonate existing a t the higher temperature, and since there is no marked increase in the reported carbonate even with the sulfonate equal to 22 millimoles per liter, the carbonate existing a t the higher temperatures must be low in respect to the acid carbonate. The data reported in Tables 111tfo V indicate that, in order to maintain a stable phase of calcium carbonate, only a small amount of sodium carbonate is necessary. Thus, with the sulfate as high as 22.40 millimoles per liter (2150 p. p. m.), a carbonate content as low as 0.78 millimole per liter (33 p. p. m.) should prevent the formation of calcium sulfate. The solubility of calcium carbonate, being zero for all practical cases, gives further insight into the question of scale formation and prevention in steam boilers. Carbonate scale forms readily in feed water heaters and hot water lines, owing to the breakdown of the calcium bicarbonate to form the more insoluble calcium carbonate. Such precipitation takes place rapidly and the precipitate forms on the portions being heated, as well as on the cooler portions. However, with the normal carbonate entering the boiler, which contains free hydroxide in the majority of instances, the solubility is reduced to zero, and the solid phase forms evenly throughout the boiler irrespective of the temperatures. If a t any time the hydroxide and carbonate alkalinity disappear and the sulfate increases, then sulfate scale is formed. The solubility of the calcium sulfate is appreciable. Consequently, it is deposited at the heating surface. Small amounts of sodium carbonate in the presence of hydroxide are sufficient to prevent formation of the solid-phase calcium sulfate, even at a temperature as high as 282" C. A feed water which contains a small amount of sodium carbonate in addition to the theoretical amount required to react with the calcium sulfate present will form sodium hydroxide in the boiler. However, in the average boiler not over 90 per cent of the sodium carbonate decomposes to form sodium hydroxide, and 10 per cent remains in the boiler. Thus, if the total alkalinity (expressed as sodium carbonate) of a boiler water becomes 500 p. p. m., there will be 50 p. p. m. of sodium

CRUDEBARITEINDUSTRY. In 1931 the crude barite industry in the United States was characterized b declines in total mine production, shipm'ents, and prices, a n i an increase in total stocks at the mines. The total quantity of crude barite mined and sold in the United States, which had been annually more than 200,000 tons for the past six years (1925-30, inclusive), fell to about 174,500 tons in 1931, or 26 per cent below the corresponding figure for 1930. On the other hand, to.tal *ports of crude or unmanufactured barite for consumption in the United States in 1931 increased 40.2 per cent in quantity as compared with 1930. With the exception of 1929, im orts in 1931 were higher than for any year for which imports ofunmanufactured barite have been recorded-namely, 1884-1931, inclusive. The total value of crude barite mined and sold in the United States in 1931, which had exceeded $1,000,000 annually for the nine-year period 1922-30, inclusive, dropped in 1931 to approxi-

Vol. 24, No. 10

carbonate remaining in the boiler (when analyzed a t room temperature), or approximately 0.5 millimole of carbonate, and calcium sulfate scale should be prevented even in a boiler operating a t 1000 pounds steam pressure. The direct interpretation of these solubility data in terms of boiler operation should be made conservatively. These are equilibrium data, and, until they are correlated with data from actual boiler operation, no specific conclusions should be drawn. There is a possibility that the rate of steam generation a t the heating surface may have more to do with the rate of scale formation than the question of solubility, and, furthermore, it is possible that the equilibrium conditions may not be maintained in the various portions of an operating boiler. CONCLUSIONS 1. At temperatures between 182' and 282" C. in the reaction, CaSO, NalCOI +3 CaCOa NanSOc

+

+

the presence of as low a concentration as 1.5 millimoles per liter (90 p. p. m.) of carbonate will force the reaction to the right, even when the sulfate concentration is around 18.0 millimoles per liter (1728 p. p. m.). 2. I n the following reaction a t temperatures between 182" and 282' C., CaS04

+ XatCOa

CaCOa

+ Na2S04

in the presence of sodium hydroxide (5.0 millimoles per liter) and sodium chloride (2.5 millimoles per liter), the presence of as low a concentration as 0.5 millimole per liter (30 p. p. m.) of carbonate will force the reaction to the right, even when the sulfate concentration is around 18.0 millimoles per liter (1728 p. p. m.). 3. The ratio of carbonate to sulfate is not a constant for the reactions studied owing t b the fact that the carbonate measured a t room temperatures undoubtedly exists to a large extent as acid carbonate a t the higher temperatures.

LITERATURECITED (1) Hall, R. E., Carnede Inst. Tech., Bull. 24 (1927). (2) Partridge, E. P., Univ. Mich., Eng. Research Bull. 15 (1930). (3) Partridge, E. P., Schroeder, W. C., and Adams, R . C., Jr., Am.SOC.Mech. Eng.,Reprint 37, Dec., 1931. (4) Straub, F.G., Ibid., 47, Dec., 1931. (5) Straub, F.G., IND.ENO.CHEM.,24, 914 (1932). (6) Straub, F.G., Ibid., Anal. Ed., 4, 290 (1932). RECEIVED April 18, 1932. Presented before the Division of Water, Sewage, and Sanitation Chemistry at the S3rd Meeting of the American Chemical Society, New Orleans, La., March 28 to April 1, 1932. Releaeed by permission of the Director of the Engineering Experiment Station, University of Illinois. Part of the research conducted in the Chemical Engineering Division of the Engineering Experiment Station, and financed by the Utilities Research Commission of Chicago.

mately $994,600, or about 35 per cent below the total value reported for 1930. The total average value per ton was $5.70, or 85 cents a ton below that of 1930. With one exception (1929), stocks of crude barite at the mines have steadily increased since 1923, and reached a maximum in 1931,

Missouri held first place, as usual, in order of output; Georgia ranked second, California third, and Tennessee fourth. Washington County, Mo., produced about 70 per cent of the total shipments of crude barite from Missouri in 1931, and the combined output of Georgia and Missouri constituted more than 75 per cent of the total out.put of crude barite in the United States &ring the year. The commodity is largely utilized in this country by the manufacturers of barium Drodricts and chemicals and is exported by them in manufactbed form.