870
F. F. RAWLINGS, JR.,
AND
was determined experimentally. Within the limits of error of measurement, the heat of dilution was zero, i.e., m a O 3 . 1 6 = 0.0 f 0.6 cal./mole, corresponding to a dilution from 0.0543 to 0.0528 molal. Combination of the two measured heats of solution yields AH303.16 = 735 f 15 cal./mole for the reaction Na2SOd (V) = NalSO, (111)
(3)
This result, in conjunction with the heat content values of Table 11, leads to A H z g s . 1 ~= 743 f 20 cal./mole and A H 4 6 0 = 740 f 120 cal./mole. From
E. C. LINGAFELTER
VOl. 77
the work of Kracek and Gibson,' it is estimated that AFo = 0 for this reaction at 450 f 5°K. Hence, A s 0 4 5 0 = 1.64 0.30, from which ASoz98 = 1.70 may be obtained by means of entropy increments in Table 11. The latter result and Pitzer and Coulter's value for the entropy of NazSO4 (V) (s0Z98,16 = 35.73 f 0.10) yield SO298.16 = 37.4 f 0.5 cal./deg. mole for Na2S04 (111). Columns 6 and 7 of Table I1 were obtained by combining the values of AHm and AS450 with the heat content and entropy increments of Na2S04 (111, I, 1) above 450°K.
*
BERKELEY, CALIFORNIA
[CONTRIBUTION FROM THE DEPARTMENT OF CHEMISTRY O F
THE
UNIVERSITY O F WASHINGTON]
X-Ray Crystallography of the Sodium n-Alkyl Sulfates1 B Y F. F. RAWLINGS, J R . , ~AND E.
c. LIKG.4FELTER
RECEIVED AUGUST23, 1954 An X-ray crystallographic study has been made of the sodium n-alkyl sulfates, for comparison with previous studies of the sodium 1-alkanesulfonates. Six distinct phases are reported, one of which is very similar to the a-phase of the sulfonates. Unit cells and space groups are given for all six phases.
Introduction As an extension of our previous studies of the sodium l-alkane sulfonates, R S O S N ~ , ~ -we ? have thought it of interest to compare the sodium n-alkyl sulfates, RSOdNa, which may be considered as related to the sulfonates by the replacement of one methylene group by an oxygen atom. The alkyl groups which have been included in the present study are C6, C?JCS, c9, c10, c11,c12, c14~c16, ClS and Czo. In the work on the sulfonates, six distinct hydrated phases and one anhydrous phase have been recognized and designated CY, p, y, 6, E, {, g in the order of their discovery. Because of the extreme similarity in molecular structure between the sulfonates and the sulfates, and therefore the possibility of appearance of identical or similar crystal structures, i t was decided to use the same series for the sulfate phases. Thus we shall discuss in this paper the a-phase of the sulfates, which is almost identical with the a-phase of the sulfonates* and the 1, K , 1,p and v phases, which are different from those observed for the sulfonates. Experimental With the exception of the n-eicosyl alcohol, the normal alcohols used in this work were the purest alcohols obtainable commercially, and no further attempt was made to increase the purity except for the n-hexyl and n-heptyl alcohols. The n-hexyl and n-heptyl alcohols were fractionally dis(1) This work was supported in part under contract DA-04-200ORD-236 with the O 5 c e of Ordnance Research. (2) Taken from a thesis submitted by F. F. Rawlings, Jr., in partial fulfillment of the requirements for the Ph.D. degree. Procter and Gamble Fellow, 1949-1990. (3) L. H. Jensen and E. C . Lingafelter, THISJOURNAL, 66, 1946 (larl). (1) 1.. H. Jensen and E. C. Lingafelter, ibid., 66, 1729 (1946). (5) E. C . Lingafelter and I,. H. Jensen, Acta C y y s f . . 3, 257 (1950). (6) J. E. Minor and E. C . Lingafelter, ibid.. 4, 183 (1951). 76, 5761 (7) L. A. Wilcox and E. C. Lingafelter, THISJOURNAL,
(1053). (8) F. F. Rawlings, J r . , and E. C . Liogafelter, i b i d . , 74, 1852 (1950).
tilled under vacuum using a three-foot column having an outside diameter of 18 mm. and packed with small glass helices. The alcohols were obtained from several companies, including the Eastman Kodak Co., Columbia Organic Chemicals Co., Halogen Chemicals, Inc., and the Matheson Co. The n-eicosyl alcohol was made in this Laboratory from n-octadecyl bromide (Halogen Chemicals, Inc., Columbia, S . C.) and liquid ethylene oxide (Eastman Kodak Co. white label), using the Grignard reaction. The purity of the n-eicosyl alcohol (m.p. 6545.5') was checked by comparing the powder pattern with those of known alcohols, namely, the n-hexadecyl and n-octadecyl alcohols. All sodium n-alkyl sulfates were prepared by the method of Lottermoser and Stalls with the exception of the sodium n-hexyl sulfate and sodium n-eicosyl sulfate which were prepared following the procedure outlined by Dreger .Io The crystals were grown from mixed solvents, primarily ethanol and water, by slow evaporation of the solvent fxom The saturated solutions a t constant temperature (&lo). best crystals were grown from solutions which contained no solvents other than ethanol and water. With the sulfonates, improved crystal formation can often be attained by the addition of such solvents as carbon tetrachloride, 1,4dioxane and glycerol. In the case of the sodium n-alkyl sulfates these additions were usually more detrimental than beneficial. The data in Table I show the most favorable conditions for growing each of the phases. I t should be mentioned that the crystals of the iota and lambda phases are very unstable a t room temperature and effloresce to form the more stable alpha phase almost immediately. The r-phase is also unstable a t room temperature and changes t o the CYphase by absorption of water vapor. The compounds having a chain of less than eleven carbon atoms form only gels if the percentage of water is raised above 5% in ethanolwater mixtures. The crystals, with the exception of the aphase in the region of C,Oto CI2and the K-phase for CM and C,,,, were very small and poorly formed. All the crystals are extremely thin and show a considerable tendency to bend and warp. An anhvdrous phase, which we have designated the phase, is known i o exist from powder patterns but single crystal growth has not been attempted up t o the present time due t o the elaborate precautions which must be taken to grow anhydrous crystals. The hydrate determinations were made in the apparatus (9) A. Lottermoser and F . Stoll, Kolloid-Z., 63, 49 (1933). (10) E. E. Dreger, G . I. Keim, G. D. Miles, Leo Shedlovsky and John Ross, I n d . Eng. Chcm., 36, 610 (1944).
X-RAY CRYSTALLOGRAPHY OF SODIUM %-ALKYL SULFATES
Feb. 20, 1955
TABLE I CRYSTAL GROWTH Substance
Phase
Solvent
Temp., OC.
a A
Ethanol Ethanol Ethanol Ethanol 95% Ethanol, CC14 Ethanol 95% Ethanol Ethanol 9.570 Ethanol 95% Ethanol H20, 1,4-dioxane 9570 Ethanol 95% Ethanol, H20 95% Ethanol 95% Ethanol, H20 95% Ethanol 95% Ethanol 95% Ethanol, H20 95% Ethanol, H20 95% Ethanol 95% Ethanol 95% Ethanol, HzO 95% Ethanol 95% Ethanol, H20
25-26 25-26 40-41 25-26 25-26 40-41 25-30 45-50 25-30 25-30 25-26 25-30 25-26 25-30 25-26 35-36 40-4 1 30-3 1 25-26 57-58 35-36 25-26 40-41 25-26
P a
X P
or P
a
a 1
a 1
a 1
a
a K V
or K V
K Y
Approx. % Ha0
1 1 1 1 5 1 5 1 5 5 5 5 25
scope with a graduated stage are found to be very convenient for rapid identification of the phases. A comparison of the observed and calculated (from and bo values) end angles are given in Table 11. TABLE I1 COMPARISON OF OBSERVED AND CALCULATED ENDANGLES Unit cell parameters in Angstrom units. Chain Phase length
or
c12
i
CIS
K
h ~.r Y
cla Ce C, Cla
5 50
5 5 15 20
5 5 5
length
Discussion The crystals of all phases are monoclinic, thin tabular on (OOl), usually elongated in the a direction and terminated by (111) and (111). Occasionally crystals were found terminated by (211) and (271) or by (121) and (121). The end angles (the angle between the zone axes defined by the terminating faces and (001)) as measured with a micro(11) E. C. Lingafelter, L. H. Jensen and A. E. Markharn, J . Phys. C h e m . . 57, 428 (1953). ( I ? ) I-. A . Wilcox. P h . D . Thesis, University of Washington, Seattle, Wash., 1951.
bo
Terminating planes
16.40 10.30 ( i i i ) ( i i i ) 9.46 14.02 (111) (1%) 9.44 9.15 ( i i i ) ( i i i ) (211) (2i1) 9.46 14.10 (111) (131) 8.50 6.11 (111)(lfl) 9.86 5.25 (111) (171) (121) (1%)
End angle Calcd. Obsd.
115044' 68'2' 91'48' 54034' 67'44' 108'36' 123'56' 150'12'
TABLE I11 CRYSTALLOGRAPHIC DATAFOR SODIUM n-ALKYL Chain
described by Lingafelter, Jerisen and.Markham," although, with the exception of the a-phase, it was not possible to make equilibrium measurements a t 15 mm. pressure due to unstability of hydrates and the tendency for the alkyl sulfate to hydrolyse a t elevated temperatures. Above about 50°, the Toepler pump was used and only the total amount of hydration was measured. Density measurements were made in most cases by the flotation method using 1,4dioxane and carbon tetrachloride. In the case of the A- and 1-phases, which are both monohydrate phases and decompose rapidly in air a t room temperature, the problem of determining the densities was very difficult. It has been shown by Wilcoxl* for the sulfonates that the y-phase not only decomposes in air a t room temperature but also decomposes immediately in mixtures of 1,4-dioxane and carbon tetrachloride. Both the y-phase sulfonate and the r- and X-phase sulfates change over to the a-phase in such solvents. The densities of the y-phase sulfonates were measured in a water- and sulfonate-saturated mixture of benzene and carbon tetrachloride. This modification also gave fairly constant values for the 1- and A-sulfate crystals. As a check, the Ce and C, A-phase and the C1, 1-phase were r u n using their mother liquor as one solvent and carbon tetrachloride as the other. The C ~ r-phase Z density was measured in a mixture of pure water and water saturated with potassium iodide. Since checks were obtained it is felt that these values are fairly accurate. X-Ray crystallographic data were obtained from rotation, equi-inclination Weissenberg, and precession photographs using Cu K a radiation (1.5418 A.).
a0
115~ 66" 92" 54" 68' 108" 124' 146'
In Table I11 are listed the X-ray crystallographic results (cell elements, space groups, number of molecules) and the hydration values and densities of all of the crystals investigated.
20 20
87 1
B
bo
(2)
(A,)
(2)
SULFATES Density g./cm.:'
Obsd.
(")
Calcd.
or-phase, RSO+Na.1/sH20(see text), z =I 32-eighth hydrate, Aa or A2/a for n even; Ia or I2/a for n odd Cs
16.46
10.41
16.49 16.48 16.45 16.42 16.40 16.38 16.38 16.37
10.37 10.38 10.38 10.36 10.30 10.29 10.30 10.37
c 7
Cs Cs Cio CII Cir Cu Cis Cis
57.79 62.69
68.06
95'54' 94'54' 96'48'
73.08 78.28 88.52 98.88 106.48
98'18' 100°30' 102'36' 104'36'
96"OO'
47.30 52.54 57.48 62.46 67.60 72.68 77.46 87.04 96.50 103.04
1.27 1.23 1.19 1.18 1.17 1.13 1.12 1.11
1.352 1.302 1.266 1.235 1.207 1.187 1.179 1.153 1.130 1.137
r-phase, RSOdNa.Ht0, Pa, P2/a or P21/a, z = 10-monohydrate CII Cir
9.51 8-46
13.99 14.02
32.74 35.74
K-phase, RS04Na.'/rHzO, CIS Cis Cm
9.49 9.44 9.50
9.21 9.15 9.16
46.98 49.05 53.48
98'6' 92'12''
32.41 35.71
1.12 1.08
1.124 1.073
P21/a, z = 8-quarterhydrate 91'33'' 94'00' 93°48'a
46.96 48.92 53.36
1.14 1.16 1.15
1.119 1.169 1.157
A-phase, RSO,Na.H20, Aa or A2/a, z = 16-monohydrate CS C7
9.46 9.38
14.10 14.05
38.03 44.91
103'30' 97°36"'
36.98 44.52
1.16 1.05
p-phase, RSOlNa, Aa or A2/n, z = 8-indefinite Cs
C, Cs
8.52 8.50 8.48
6.14 6.11 6.10
40.38 43.61 47.36
9Z036" 9l012' 93°39'n
40.34 43.60 47.26
1.31 1.29 1.26
1.200 1.068
hydrate 1.298 1.295 1.274
v-phase, RSO,Na.HzO; CZ,Cm or C2/m, z = 4-mOnOhydrate Cir CIS Ca
9.86 9.78 9.82
5.25 5.27 5.25
40.13 41.48 45.35
91'31'' 92'11' 95°24'a
40.12 41.45 45.15
1.16 1.20 1.18
1.159 1.212 1.193
Probably the correct values for @ as compared with other members of series but poor upper level pictures make it impossible t o be certain.
The a-phase of the sodium n-alkyl sulfates is quite similar to that of the sodium l-alkanesulfonate^.^,' Not only do they have the same space group and degree of hydration and very similar cell dimensions, but the general distribution of intensity in corresponding diffraction patterns is very similar, particularly for the longer chains
F. F. RAWLINGS, JR.,
-
AND
E. C. LINGAFELTER
VOl. 77
3
are 4 chains along the c-axis.) The constant value of doO1-dleast 8quares for the sulfonates thus 18 3 indicates a constant tilt of r = cos-' (4.7391 a k a 0 a . . a . - . .f di 5.100) = 21'40'. The long spacings of the sulc - a- a e a a a a .-a 14 5 fates, however, fall into three groups. For the 0 % C 3 shorter chains, 8 to 11 carbon atoms, the devia- a i m a a ' a a a a e l2 tion plot shows a linear variation with a slope of 3. - a - a e :a -a a -4 v C a - a ' a a a .C. a 10 8 3 0.35. The slope of do01 os. N would therefore have ca a * @ a e a a a a a constant slope of 4.739 0.33 = 5.09, indicat3 3 aa a % ing a tilt of T = cos-' (5.09/5.10) = 0'. For the I o intermediate chains. 12 to 16 carbon atoms. the _----.---.I A constant value of dool - dleast squares indicates 30" 50' 70' 90' 110' 130' a tilt of 22O, the same as for the sulfonates. 40" 60' 80" 100' 120" 140' Finally, the ClaH37SOaNa has a long spacing P. w h i r h i q a h n i i t .? ,& qhnrter t h a n t h e nrerlirtd FIX. - 1.-possible values ot B for sodium I-alkanesulfonates, value, wfiich indicates a tilt still larger tban that 0, and sodium n-alkyl sulfates, 0 . of the sulfonates. The data on the other phases are scanty. lengths. There are, however, some slight but There can be no question as to the existence of the definite differences. several phases, but in each case, only one chain The a. and bo for the sulfates are about 16.4 length has afforded good enough crystals to unand 10.3 k., respectively, while the corresponding equivocally determine the space group and select values for the sulfonates are 16.8 and 10.1 8. the appropriate value for p. It therefore seems A comparison of the possible values for p for the advisable to defer any discussion of these phases sulfates and sulfonates is shown in Fig. 1. The set until additional data are available. a t the extreme right for the sulfates gives ANo. of carbon atoms in sodium alkanesulfonate. centering for even numbers of carbon atoms and I 6 8 10 12 14 16 18 for odd; the extreme right hand set for the sulfonates gives I for even and A for odd. I n the case of the sulfonates, the values chosen were those most nearly corresponding to the probable direction of the chains. In the case of the sulfates, as shown below, the chain direction is probably not constant. We have therefore chosen the indicated values, which form the set nearest to 90' which gives Aa (or ,42/a) as the space group for -1.0 those having even numbers of carbon atoms and I a (or I2/a) for those having odd numbers. A comparison of the long spacings also shows a E * definite difference between the sulfates and the : : sulfonates. This is best illustrated by Fig. 2, in which we have plotted the deviations of the long +I -2.0 spacings from a linear function of chain length. . The linear function used is dleastsqusres = 16.618 4- d 4.739N1 where N is the number of carbon atoms in the chain for the sulfonates and N - 1 (to account for the extra oxygen atom) is the number of carbon -3.0 atoms in the chain for the sulfates. This function was derived from a least squares treatment of the long spacings of the sulfonates alone. If the long spacings of a homologous series are found to fit a linear function of the chain length, i t is reasonable 6 8 10 12 14 16 18 to assume that the chains have a constant tilt No. of carbon atoms in sodium alkylsulfate. from the normal to (0011, the magnitude of the tilt, T , being given by the relation cos r = (slope of Fig. 2.-Long spacings, & I , of sodium 1-alkanesulfonate, 0, and sodium alkyl sulfates, 0 . do01os. N)/3.100 (5.100 8.is the expected increase in the direction of the chains, 4 X 1.275 since there SEATTLE, WASHIXGTOX a
e
e
a
a
a m . .
~1
25
O
r
I
+
-
5
D
r;i-
