The Solubility of Citric and Tartaric Acids in Water

5/26/2018 The Solubility of Citric and Tartaric Acids in Water

1/3

Dec., 1937 THESOLUBILITYF CITIUCAND TARTARICCIDSN WATER 2547[CONTRIBUTIONF RO M THE CHEMICAL LABORATORYF THE BROOKLYNOLLEGE O F PHAXMACY, LONGSLAND UNIVERSITY]

The Solubility of Citric and Tartaric Acids in WaterBY LAWRENCE. DALMAN

In a recent investigation1 a few measurementswere reported for the solubilities of citric and tar-taric acids in water. A search of the literaturereveals practically no data with which to comparethe results obtained for citric acid. While Sei-dell2 has reported an isolated solubility at 25and Kremann and Eite13 have obtained somemeasurements at lower temperatures from freez-ing point data, no systematic investigation ap-pears to have been made of the solubility of citricacid over the usual range of temperature from 0to 100.The figures obtained for the solubility of tar-taric acid were found to be about 1% lower thanthose given in International Critical table^')^and Seidells Tables5 which are both taken fromthe results originally reported a t 5 intervals from0 to 100 by Leidie.6 It is not stated how thesemeasurements were made except that they wereobtained with the aid of two equations. Whenthese values are plotted, the solubility curve ap-pears to consist of two sections which intersectbetween 40 and 45, although the saturating solidphase is reported to be anhydrous tartaric acidthroughout the entire temperature range. Theauthor does not offer an interpretation of thisirregularity and it was therefore decided to repeatthese measurements. The present investigationpresents the results obtained for the solubilitymeasurements of citric and tar taric acids at tendegree intervals from 0 to 100.

MaterialsCitric and Tartaric Acids.-The citric and tartaric acids

were supplied by Charles Pfizer and Company. Theseacids were thrice recrystallized from water, dried andstored over sulfuric acid. Titration against sodiumhydroxide gave results which agreed within 0.1 of theirtheoretical values.

Standard Sodium Hydroxide.-This solution was pre-pared and used according to accepted standards. It waspreserved in a container protected against dust, carbondioxide, etc., and standardized at regular intervals.Procedure.-Solubilities were measured by analysis ofthe saturated solutions obtained by agitation of the acids

(1) Dslman, TEIS JOURNAL, 69,775 1937).2)Seidell, Bull. N o . 67, H y g . Lab. 254, 1910).3) Kremann and Eitel, Rec. trao. chim. 44,539 1923).4) I. C T. ol. IV, . 251.5) Seidell, Solubilities, p. 710.6) LckIie, Comfi t rend. 96 87 1882).

in water enclosed in glass-stoppered tubes and rotated inan electrically controlled thermostat for at least s i x hours.This was found to be sufficient time to bring about equi-librium although the samples below 50 were rotatedfrom eight to ten hours. The measurement a t 0 wasobtained in a well-stirred mixture of ice and water, whilethat at 100 was taken in boiling distilled water. Thetemperature control was *0.02, read directly from athermometer standardized by the Bureau of SGndards.

For analysis samples of the saturated solution weretaken up in a pipet (warmed when necessary) and weighedin a small Vial. To prevent ingress of solid material, theend of the pipet was protected by means of a piece ofcloth held securely by thread. The acid content of thesample was determined by t itration with standard sodiumhydroxide using phenolphthalein as indicator. ll deter-minations were made in triplicate and averaged. Thedata thus obtained have been expressed in weight per cent.and plotted in Fig. 1 It is believed that the measure-ments are precise within O.l , although it was found thatthey could be checked on the average within 0.06%.

8580

75

2 70886 5 -3B

60

55

0 20 40 60 80 100Temperature, C.

Fig. 1.-The solubilities of citric and tar-taric acids in water: 0 Leidies curve fortartaric acid.

ResultsCitric Acid.-It will be seen from Fig 1 that

the solubility curve for citric acid consists of twostraight lines which intersect a t approximately


2/3

254s LAWREXCE. D a m m VOI. 5936O, thus indicating the existence of two solidphases. To determine the composition of thesesolid phases, the crystals from the sample tubeswere well centrifuged, weighed and dried formoisture content after which their acid contentwas measured by titrat ion with alkali. The re-sults of these analyses showed that the solidscorresponding to the curve above the break werepractically pure citric acid, while those corre-sponding to the one below contained 91.1 to 91.4%acid which is approximately that required by theformula H~C ~~O,-HZO.t is evident that thetwo solid phases existing in this system are citricacid monohydrate which is stable below the transi-tion point and anhydrous citrio acid which is stableabove.

Attempts to determine this transition tempera-ture accurately by means of cooling curves ob-tained from mixtures stirred in a Beckmann freez-ing-point tube repeatedly proved unsuccessful.The most probable explanations appear to be thateither the transition occurs exceedingly slowly orthat the heat change is too small to produce meas-urable results as read on a tenth degree ther-mometer. In view of this failure it was decidedto determine the solubility curves to their point ofintersection and to at tempt by analysis of thesolid phases to locate the point a t which thechange of phase occurs. The results show thatthis change occurs between 35.7 and 36.9'. Afinal sample determined at 35.8' gave crystalswhich contained 94.7% acid. This figure lies be-tween that of t he anhydrous and the hydratedacid which is 91.43 , thus indicating a mixtureof the two solids. Since the composition of thesolution showed no appreciable change after fur-ther rotation in the bath, it is evident that equilib-rium had been attained and the transition tem-perature is therefore reported as 35.8'. The satu-rated solution at this temperature contains67.61% acid. This figure agrees well with thecomputed value which is found to be 67.63y0(average of values from the two equations).

The following equations for the solubilities ofthe hydrated and anhydrous citric acids expressedin percentages by weight have been obtained fromthe data by the method of least squares

Monohydrated citric acid solubility = 48.95594 0.52311Anhydrous citric acid solubility = 57.8564 0.26161

The solubilities computed by these equations agreevery well wt the observed values, the averagedeviation being 0,027 nd 0,02570while the maxi-

mum deviation is 0.08 and 0.05% for the anhy-drous and hydrated acids, respectively.

The solubility of citric acid a t 25' was foundto be 62.070/0. This is 0.36% higher than thatreported by SeidelL2 It is hardly feasible tomake a comparison of these results with thoseobtained by Kremann and Eite13 a t 0, 10 and15' since their values appear to be much too highand do not show a consistent curve.Tartaric Acid,-The results obtained for thesolubility measurements of tartaric acid are rep-resented graphically in Fig. 1. In contradis-tinction to the irregular curve obtained by Leidie,it is apparent tha t the solubility curve for tartaricacid closely approximates a straight line. Thereis no eviderice of a break near 45' and the saturat-ing phase was found to be anhydrous tartaricacid throughout the entire range of temperaturefrom 0 to 100'. It is of interest to note thatwhile Leidie's curve (shown by the shaded circlesin Fig. 1) lies about 1.5% above at 0 it crossesthe author's curve a t 90 and lies 0.2% below at100'.

The solubility of tartaric acid a t 25' was foundto be 58.48%. Seidell's2 figure for the corre-sponding solubility is given as 57.90%, while thatreported by Leidiee is 59.790/0. Perhaps the bestcheck on the accuracy of these measurements isthe agreement to be found between the observedvalues and those derived from the equation

Tartaric acid solubility = 51.8573+ 0.26431the constants having been derived by the methodof least squares. The average deviation betweenthe experimentally determined solubilities andthose computed by this equation is 0.058%, whilethe maximum deviation is 0.1%.It was suspected that the measurements at 90and 100' might fall away somewhat from thestraight line solubility curves, since these acidshad been found to decompose in an oven at thesame temperature. While no serious loss inweight is noted when these acids are heated at70 for a week, or at 100' for a few hours, there

are distinct signs of decomposition (browning)when they are heated for several days a t 100'.It is quite probable that decomposition likewiseoccurs when solutions of these acids are subjectedto the same conditions. However, since the 90and 100' samples were rotated in the bath for atleast six hours, i t appears that solutions of citricand tartaric acids can be heated safely at 100' fora few hours without appreciable decomposition.


3/3

Jlec., 3 )3 i PARTIALOLAL OLUME F POTASSIUMALTS 2549Summary below this transition temperature citric acid

The solubilities of citric and tartaric acids in monohydrate exists, while above anhydrousThe solubility curve for tartaric acid is a

straight line representing solutions in equilibriumwith the anhydrous form of the acid.BROOKLYN,. Y. RECEIVEDUNE 23, 1937

water have been determined a t loo intervals citric acid is the stable phase.(including 250) Over the range of temperature0-looo.

The solubility curve for citric acid consists oftwo straight line curves which intersect at 35.8'

[CONTRIBUTIONFROM THE DEPARTMENTF CHEMISTRY, NORTH AKOTAGRICULTURALCOLLEGE]The Partial Molal Volumes of Potassium Chloride, Potassium Bromide andPotassium Sulfate in Sodium Chloride Solutions

BY HENRY . WIRTHSince the discovery by Masson' that the ap-

parent molal volume of a dissolved salt is a linearfunction of the square root of the volume concen-tration, several investigator^ ^ have made ex-haustive tests of the applicability of this relation-ship. It was found valid for a number of saltsover a surprisingly large coiicentration range.Root4 derived from Masson's rule a simple equa-tion relding density and volume concentrationwhich has found wide application.

Redlich and Rosenfeld5 derived from the De-bye-Hiickel theory a linear relationship betweenthe square root of the volume concentration andthe partial molal volume of a dissolved salt. Thisgave a partial theoretical basis to Masson's ruleas the apparent molal volume is closely related tothe partial molal volume. The expression ob-tained wasI.; is the partial molal volume of the salt in asolution of concentration c; 't;, is the partial molalvolume at infinite dilution; q is a complex factorinvolving the temperature, compressibility, thetype of electrolyte, the dielectric constant and itschange with pressure; and s one-half the sum-mation of the number of ions times the square ofthe valence of the ion. When considering a solu-tioii containing two or more electrolytes the lastterm in Eq. 1becomes p ( Z . ~ c ) ~ ~ ~ .

The factor ZWCorresponds to the ionic strengthx1a volume basis, so that the partial molal volumeof a salt in solution should be a linear function of

the square root of the volume ionic strength. This

/I = v + qw% C','? (1)

11 Mas.-on, Phil. Ma: : [ 7 ] 8 2 8 1929).12 Scott, J P h y s . C h e m . 35, 2316 (1931).3) Geffcken, Z . p h y si k . C he m . A166, 1 1931).

(4) Root, THIS OURNAL, 55, 850 (1933).(6) Redlich and Rosenfeld, Z . p h y s i k . C h e m . A l S S , 65 1931).

relationship was tested by determining the partialmolal volumes of potassium chloride, potassiumbromide and potassium sulfate in sodium chloridesolutions of different concentrations.

MethodsPreparation of Solutions.-The salts used were either

Baker Analyzed or Mallinckrodt Reagent quality andwere not further purified. Salts to be weighed were driedat 350-400 . Solutions for the density determinationswere prepared by adding a weighed amount of the driedsalt (potassium chloride, potassium bromide or potassiumsulfate) to a weighed amount of water or sodiuni chloridesolution. The volume concentration of the added saltcould then be calculated using the observed density.Enough sodium chloride solution for each series was pre-pared and its concentration determined from the density.This concentration was corrected fot: the change caused bythe addition of another salt. The concentrations reportedare expressed as moles per liter of solution.

The sodium chloride solutions used were approximately0.04, 0.16, 0.36, 0.64 and 1.0 normal. To each of thesesolutions was added sufficient potassium chloride, potas-sium bromide or potassium sulfate to make the volumeionic strength of the added salt approximately 0.04, 0.16,0.36 and 0.64. Solutions in water of the latter salts ofvolume ionic strength 0.04, 0.16, 0.36, 0.64 and 1.0 werealso prepared.

Density Determination.-The density of each of theabove solutions was determined by means of the sinkermethod. The solutions whose densities were to bemeasured were placed in heavily silver plated coppercans immersed in a thermostat a t 25'. Metallic con-tainers were used to decrease the time required to attaintemperature equilibrium. The temperature was main-tained constant to O . O 0 l o . The sinker was suspendedin the solution by means of a fine platinum wire whichwas coated with platinum black where it passed throughthe liquid surface. This wire could be fastened to a panof balance which was supported over the thermostat.

DiEerences in density were calculated directly usingthe equation

d, , = (m, J / v 2)

Documents

The Solubility of Citric and Tartaric Acids in Water