Compound Formation in Binary Fatty-Acid Mixtures

Compound Formation in Binary Fatty-Acid MixturesAuthor(s): M. Fieldes and L. HartmanSource: Proceedings of the Royal Society of London. Series A, Mathematical and PhysicalSciences, Vol. 233, No. 1193 (Dec. 20, 1955), pp. 195-202Published by: The Royal SocietyStable URL: http://www.jstor.org/stable/99877 .

Accessed: 04/05/2014 03:22

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].

.

The Royal Society is collaborating with JSTOR to digitize, preserve and extend access to Proceedings of theRoyal Society of London. Series A, Mathematical and Physical Sciences.

http://www.jstor.org

This content downloaded from 130.132.123.28 on Sun, 4 May 2014 03:22:00 AMAll use subject to JSTOR Terms and Conditions

http://www.jstor.org/action/showPublisher?publisherCode=rsl

http://www.jstor.org/stable/99877?origin=JSTOR-pdf

http://www.jstor.org/page/info/about/policies/terms.jsp


Compound formation in binary fatty-acid mixtures

By M. FIELDES AND L. HARTMAN

Department of Scientific and Industrial Research, Wellington, New Zealand

(Communicated by E. Marsden, F.R.S.-Received 26 April 1955)

Long crystal spacings d1 and short spacings d2 and d3 of C12-C14, ClA-C16 and C16-C18 fatty- acid mixtures have been determined using a high resolution X-ray diffractometer in an attempt to elucidate the question of compound formation in these mixtures. Both d1 and d3 spacings showed significant changes in relation to the composition of the mixtures whereas d2 spacings remained practically unchanged. The changes in d1 and d3 spacings are interpreted as denoting limited intersolubility of the component fatty acids and partial compound formation.

Numerical correlation found between the above changes seems to indicate a shortening of the distance between alternate carbon atoms of the higher melting component in mixtures containing 30 to 70 mol. percent of this component.

INTRODUCTION

Despite numerous investigations the peculiar shape of the solidification point curves of binary systems formed by adjacent even-numbered n-fatty acids remains still an unsolved problem. The existence of an inflexion in these curves at the

equimolar composition, beside a eutectic point at about 73 mol. % of the lower

melting acid, had been tentatively explained by Roozeboom (i899) as due to

compound formation. The curve could be then split into halves at the 50 mol. % point to yield two systems of type III according to Roozeboom's classification of solid solutions. In view of the tendency of fatty acids towards association (Robertson 1902, 1903) the compound suggested by Roozeboom is now usually envisaged as consisting of the two components of the system held together by hydrogen bonding. This concept is supported by the observation (Francis, Piper & Malkin I930) that equimolar mixtures of fatty acids show crystal spacings approaching the arithmetical mean of those of the pure components. However, Slagle & Ott (i933) advanced arguments against this theory which, as far as we are aware, have not been effectively refuted. On determining long crystal spacings of C10-Cl, acids they found only one spacing for each particular composition. When these spacings were plotted against composition it was found that the

points were scattered closely around the straight line connecting the pure acids. The occurrence of the mean spacing of the two components in the equimolar mixture was therefore explained by Slagle & Ott as being a special case of a general rule. This led them to the assumption of a continuous series of solid solutions, and the concept was extended to mixtures of even-numbered acids. The much wider scattering of spacings in such mixtures as well as the inflexion at the

equimolar point was attributed to changes in modification (polymorphism). The last assumption is difficult to reconcile with the fact that inflexion in the

equimolar range is shown by a number of other long-chain particles, but only by those which crystallize in double molecules. Thus anilides of fatty acids and alkyl

[ 195 ]



M. Fieldes and L. Hartman

iodides which form double molecules show an inflexion, whereas the mono- molecular ethyl esters of fatty acids do not (Ann. Rep. Chem. Soc. I939). This and certain discrepancies between the X-ray data of Slagle & Ott (I933) and those of Francis, Collins & Piper (1937) who favour the compound theory, prompted us to re-examine the crystal spacings of some carefully purified fatty-acid mixtures. In addition to long crystal spacings d, recorded by the above mentioned

authors, short spacings d2 and d3 which denote lateral distances between the

fatty-acid chains were determined. Results obtained in this investigation have led us to conclusions differing from those of Slagle & Ott.

EXPERIMENTAL

Three binary fatty-acid mixtures Ci2-C14, C14-C16 and Cl6-C,l were examined. Lauric, palmitic and stearic acids were purified by fractional distillation of their

methyl esters followed by crystallization of the free acids as described by Francis et al. (I937). Their characteristics are summarized in table 1. The solidification

point curve of the palmitic-stearic acid pair agreed closely with that of de Visser

(I898). TABLE 1. CHARACTERISTICS OF FATTY ACIDS

saponification equivalent r ~- - melting-point setting-point

acid calc. found ( C) ( C) lauric 200-31 200.1 44-15 43.7 myristic 228-37 228-3 54.3 53-72 palmitic 256-4 256.6 62-8 62.4 stearic 284.5 284.3 69-9 69-1

For the determination of crystal spacings, a Philips Geiger X-ray diffractometer was used which was expected to allow a higher degree of resolution than the X-ray diffraction equipment employed in previous investigations. Samples were examined using FeKac radiation and the apparatus was standardized at

angles corresponding to crystal spacings 1-818, 3.340 and 4-255 A of a specimen of a-quartz supplied by the manufacturer.

By melting the samples on glass plates in the usual manner, and cooling rapidly, a good reproducibility of + 001 A was obtained for short spacings, but the

reproducibility for long spacings was + 0-5 A. These unsatisfactory results were attributed to the poor orientation of the crystals, and to improve the accuracy of long spacings, the following procedure of preparing samples was adopted:

Two overlapping 12 mm diameter holes were drilled in a microscope slide

providing an aperture of approximately 20 x 12 mm. The perforated slide was fastened to another slide with two clamps and approximately 0-2 g of powdered fatty-acid material was poured into the hollow space and melted. The melt was cooled rapidly and, with some practice, it was possible to separate the two slides without damaging the smooth surface of the crystalline cake contained in the

perforated slide. The smooth surface was exposed to the X-ray beam. Increased

preferred orientation of the crystals and greater accuracy in locating the specimen by this method resulted in an overall reproducibility of long spacings of + 0 1 A.

196



Compound formation in binary fatty-acid mixtures 197

Figure 1 shows the X-ray diffraction patterns of stearic acid specimens obtained

by solidifying a melt and crystallizing from acetone respectively. In both specimens two long crystal spacings of 39.40 and 43-71 A corresponding to C and B forms (Francis & Piper 1939) were observed. In the specimen obtained by the

1st. order of d, 39;40A

43-71A1

f'

a diffraction pattern of stearic acid crystallized

from acetone background fro lass 110.7Ao M9222A

\2nd.orderofd, 3rd.orderof,d

1970A 1457A 1313A

21-85A

4371A 13-13A 43'71A

(i ( 39-40A

i 1

IF;~~ ~ 19-70A .i1107A b diffraction pattern of

stearic acid cooled after

~\ (~I |~I | melting

FIGURE 1. X-ray diffraction putterns of stearic acid.




solidification of the melt the amount of the B form was only small and did not affect the peak value of third order diffraction line of the C form, whereas the

peaks of the first order diffraction lines were mutually affected. Since the same

applied to other acids and to their mixtures the values of d1 spacings were calculated from the third order diffraction lines which were also less affected by the

a

b

""C C? IZO dF (D

F4

mol. %, higher melting acid

FIGaURE 2. Long crystal spacings d, of binary fatty-acid mixtures. a, palmitic-stearic; b, myristic-palmitic; c, lauric-myristic.

absorption of the X-ray beam by the specimen. The amount of B form being small the discussion will be limited to C form, but it might be mentioned that the

shape of the curves obtained by plotting dl spacings against composition was similar for both B and C forms. Peaks in the region from 80 to 120 A appearing in the diffraction pattern (figure 1) which were at first attributed to periodicity within the specimen or to internal reflexionwere later found to be due to low angle reflexions from the glass mounts.

198



Compound formation in binary fatty-acid mixtures

Variations of the long crystal spacings d,, and the short spacings d3 are shown

graphically in figures 2 and 3, in which the spacings in angstrom units are plotted against the composition of fatty-acid mixtures expressed in mol. % of the

higher melting acid. Changes observed in these spacings were considered significant. On the other hand, d2 spacings remained practically constant, showing for all mixtures values of about 4-14 A with a deviation of + 0*02 A.

O ? - a

3'82- \

b

5 , Oo c _ / O

.9 3 78-f - - b ? s

3.74 /

8/

3 70 I 1 I I . - I 1--- 0 20 40 60 80 100

mol. %, higher melting acid

FIGURE 3. Short crystal spacings d3 of binary fatty-acid mixtures. a, palmitic-stearic; b, myristic-palmitic; c, lauric-myristic.

DISCUSSION

Before we examine the X-ray data recorded in this work it might be mentioned that Slagle & Ott's argument in favour of a continuous series of solid solutions does not exclude the possibility of an equimolecular compound. A compound forming a series of solid solutions with either component of the fatty-acid system would also give rise to one definite spacing characteristic for each mixture. The difference would be only in the interpretation of the inflexion at the equimolar point. However, it will be shown below that X-ray data do not seem to support the concept of complete intersolubility of the fatty acids examined.

The diagrams of long crystal spacings d, (figure 2) can be divided into three sections. On the left-hand side of each diagram (up to 30 mol. % of the higher melting acid) the slope of the curve is steep. In the middle portion of the curve, which extends from 30 to approximately 70 mol. % of the higher melting acid, there is flattening and a minimum at about 40 mol. % followed by a marked increase of d1 spacings. The last section of the curve is flat and the spacings are

very close to that of the pure higher melting acid, which confirms the results obtained by Francis et al. (I937) for the palmitic-stearic acid pair. The shape of the curve obtained for C,6--C18 acids by Slagle & Ott is somewhat different. Apart from the previously mentioned occurrence of small amounts of B modification

only one long crystal spacing was observed by us for each particular mixture on

199




rapid cooling of the melt, whereas Francis et al. recorded several sets of spacings depending on the method of preparing the specimens.

The changes of d1 spacings acquire significance when compared with those of the short crystal spacings d3 (cf. figure 3). The smallest values for d3 spacings are shown by pure acids but whereas the addition of 5 % of the higher melting acid results in a marked increase of these spacings (left-hand side of the diagram) the addition of up to 20 % of the lower melting acid produces no perceptible increase

(right-hand side of the diagram). It will be shown below that there are reasons for assuming a connexion between the mode of changes of the long and short

spacings and the intersolubility of the component fatty acids. The steep rise of dI and d3 spacings as the proportion of the higher melting acid

increases could be associated with the increasing difficulty of forming composite structures. In fact, beyond a limiting concentration of the higher melting acid there is a breakdown in the structure as shown by the appearance of a eutectic

point at about 27 mol. % of this acid.

TABLE 2. CORRELATION BETWEEN CHANGES IN d3

SPACINGS AND STAGGER DISTANCES, d

differences

crystal spacings (A) d1 d3 Ad3 Ad

mol. % higher melting acid 100 40 100 40 lauric-myristic 31-3 29.3 3.73 3.84 0.11 0.105 myristic-palmitic 35.2 33.4 3.71 3.79 0.08 0.086

palmitic-stearic 39.4 37.8 3.70 3.76 0-06 0.069

The middle portion of the d, diagrams shows, as already stated, a minimum and at this point the values of d1 spacings approach the arithmetic mean of the

spacings for the pure acids. If this were accompanied by the appearance of a d3

spacing similar to that of a pure acid it would signify the presence of a new uniform phase corresponding to an equimolecular compound. Actually there is no decrease of d3 spacings, which would indicate that even in this range the

compound formation-if any-is by no means complete. However, this is not

unexpected. As there is no apparent reason why the affinity between two different

fatty-acid molecules should be greater than between two molecules of the same

kind, one would expect at the equimolar point no more than 50 % compound formation.

The fact that d1 spacings in this region approach, nevertheless, the mean spacing of the two acids could be explained by assuming a shortening of the distance between alternate carbon atoms of the higher melting acid which makes possible its accommodation in the composite structure.

The nearest approach to the mean spacing is at about 40 mol. % of the higher melting acid and at this point there is a correlation between the shortening of the above acid and the increase of d3 spacing, which was established as follows:

The distance between adjacent carbon atoms in a fatty-acid chain is 1-54A and between alternate carbons 2-52A corresponding to the included angle of

200



Compound formation in binary fatty-acid mixtures 201

109? 28' between carbon to carbon bonds (Miiller I927, 1928; Vand, Morley & Lomer I95I). The chain having the form of flat zigzags (Miiller I927, 1928) the

staggered distance, d, between adjacent carbon atoms measured perpendicularly to the long axis of the molecule in the zigzag plane is

d = V{(1.54)2- (1-26)2}A = 0-885A.

If, as in the case of stearic acid, d, spacing changes from the value of 39-4A for the pure acid to 378 A for the mixture containing 40 mol. % of this acid, the distance d changes by the increment Ad and

d+Ad= /(.154)2_(1 26x 37 8) A 0 A 39.4 A =0 .954A,

hence Ad = 0-069A.

Table 2 shows that the calculated values for Ad at 40 mol. % of the higher melting acid are in good agreement with the changes of d3 spacing determined

experimentally. The direction of d3 spacings is possibly at an angle with the plane of the zigzags, but differences arising therefrom would be within the experimental error for angles not exceeding 45?. That changes in the length of fatty-acid chains are easily realized follows from the work of Vand et al. (I951) who observed a

shortening of similar magnitude as shown in table 2 in lauric acid compared with strontium laurate.

As the amount of the higher melting acid increases beyond the equimolar proportion, d1 spacings show a sharp increase until they become equal to those of the pure acid. This almost coincides with a sudden decrease of d3 spacings which also assume the value similar to that of the pure acid. Thus in the last portion of both d1 and d3 diagrams the curves become flat. Since d3 spacings, which deter- mine to a certain extent the packing density of molecules, have reached in this

region their lowest value, it can be assumed that a sufficiently high concentration of the higher melting acid produces a structure in which the molecules of the lower melting acid are freely accommodated. The different slope of the curves on

opposite sides of the diagrams shown in figures 2 and 3 would therefore imply that the lower melting acid is more easily accommodated in the structure of the

higher melting acid than the higher in that of the lower, which appears reasonable. Thus X-ray data do not seem to support a complete intersolubility of fatty acids,

but are compatible with some degree of compound formation which could explain the inflexion in the solidification point curves. This suggests a compromise between the opposite opinions discussed in the present paper, and the view is put forward that the formation of both compound and solid solutions is only partial, and that one state represents the limiting case of the other.

In an earlier communication (Fieldes & Hartman I951) we have suggested a possible correlation between the changes of d3 spacings and the solidus-liquidus diagram for lauric-myristic acids determined dilatometrically by Jantzen (I93I).

According to this diagram as reproduced and interpreted by Bailey (I950), the area of maximum intersolubility would extend to perhaps 35 % lauric and 65 % myristic acids on one side, and to 20 % myristic and 80 % lauric on the other,

Vol. 233. A. I4




these percentages being probably too high. Several areas are outlined in the diagram, including an equimolar compound with some dissolved lauric acid, but since there are no numerical data in Jantzen's communication and the diagram itself is only available from second-hand sources, all conclusions based on it need to be treated with caution.

Recently, however, Jantzen's work received corroboration from Kofier (I954) who published a phase diagram of the palmitic-stearic acid pair containing both solidus and liquidus curves. This diagram determined by thermo-microscopic methods shows three phases, the middle one extending from 30 to 50 mol. % of stearic acid and corresponding to the compound area. In addition, Kofler's observations seem to confirm the presence of a true eutectic and peritectic point in the system. This provides further evidence against the existence of a continuous series of solid solutions, although the concept of 'a stabilized intermediate

phase' is advanced by Kofier in preference to compound formation.

The authors wish to express their gratitude to Dr E. Marsden, F.R.S., for valuable suggestions.

REFERENCES

Annual Reports Chem. Soc. I939 35, 262. Bailey, A. E. 1950 Melting and solidification of fats, p. 199. New York:

Interscience Publishers, Inc. de Visser, L. E. 0. I898 Rec. Trav. Chim. Pays-Bas, 17, 182. Fieldes, M. & Hartman, L. I95I Nature, London, 168, 74. Francis, F., Collins, F. J. E. & Piper, S. H. I937 Proc. Roy. Soc. A, 158, 691. Francis, F. & Piper, S. H. I939 J. Amer. Chem. Soc. 61, 577. Francis, F., Piper, S. H. & Malkin, T. 1930 Proc. Roy. Soc. A, 128, 214. Jantzen, E. I931 Z. angew. Chem. 44, 482. Kofler, A. I954 Microchim. Acta, p. 444. Muller, A. 1927 Proc. Roy. Soc. A, 114, 542. Muiller, A. I928 Proc. Roy. Soc. A, 120, 437. Robertson, P. W. 1902 Trans. N. Z. Inst. 35, 452. Robertson, P. W. 1903 J. Chem. Soc. 83, 1425. Roozeboom, H. B. W. I899 Z. Phys. Chem. 30, 396. Slagle, F. B. & Ott, E. I933 J. Amer. Chem. Soc. 55, 4404. Vand, V., Morley, W. M. & Lomer, T. R. I95i Acta Cryst. 4, 324.

202



Documents

Compound Formation in Binary Fatty-Acid Mixtures