Aqueous Solutions of Azoniaspiroalkane Halides (27) R . G. Deissler and J. S. Boegli. ASME Trans., 80, 1417 (1958). (28) E. G. Linder and A. P. Davis, quoted by G. Glocker and S. C. Lind, “The Electrochemistry of Gases and Other Dielectrics”, Wiley, New York, N.Y., 1939, p 200. (29) G. Poma and G. Bassi, Gazz. Chim. /tal.. 51, 71 (1921). (30) G. Poma and A. Nesti, Gazz.Chim. /tal., 51, 80 (1921). (31) J. H. Baxendaie and R. D. Sedgwick, Trans. Faraday SOC., 57, 2157 (1961). (32) C. E. Bricker and H. R . Johnson, lnd. Eng. Chem. Anal. Ed., 17, 400
1527 ( 1945). (33) S. DePaoii and 0. P. Strausz, Can. J. Chem., 48, 3756 (1970). (34) M. J. Buerger in “Phase Transformations in Solids”, R . Smoiuchowski, J. E. Mayer, and W. A. Weyl, Ed., Wiley, New York, N.Y., 1951, p 183. (35) F. R. N. Nabano, “Theory of Crystal Dislocations”, Clarendon Press, Oxford, 1967, p 595. (36) P. J. Sullivan and W. S. Kosk, J. Am. Chem. SOC., 85,384 (1963). (37) R. H. Johnsen, J. Phys. Chem.. 85, 2144 (1961). (38) R . P. Porter and W. A. Noyes. J. Am. Chem. SOC., 81, 2307 (1959).
Aqueous Solutions of Azoniasprioalkane Halides. 1. Enthalpies of Solution, Dilution, and Transfer David P. Wilson and Wen-Yang Wen* Jeppson Laboratory, Chemistry Depsrrnmnt, Clark Universky, Worcester, Massachusetts 0 16 10 (Received February 5, 1975) Publication costs assisted by the National Science Foundation
Enthalpies of solution of azoniaspiroalkane halides, (CH2)nN+(CH2)nX- ( n = 4, 5, or 6 and X = Br; for n = 5, X = C1 and I also), in HzO have been measured calorimetrically in dilute solution a t 25’. For the three bromide salts the enthalpies of solution were also measured in DzO a t 25’. Apparent molal heat contents and enthalpies of transfer obtained are compared with those of the corresponding tetraalkylammonium halides, and the results are discussed in terms of the cation structure (particularly with respect to the presence or absence of flexible alkyl chains). Forming closed loops from the alkyl chains in tetraalkylammonium ions seem to alter their thermodynamic properties greatly and weaken their hydrophobic interactions with water considerably.
Introduction Azoniaspiroalkane halides have so far been of interest to synthetic organic chemists and workers in the pharmacology field. As a group of curariform-type muscle relaxants, they have been found to inhibit certain enzyme activity, and attempts have been made to correlate the antienzymatic activity of the ions with their ~ t r u c t u r e . l - ~ There seem to be rather few reports on the physical properties of solutions of these compounds in the literature. The surface tensions of aqueous solutions of some of the halides have been reported by Thomas and CloughU4 The apparent molal volumes a t infinite dilution for a few iodides were determined by Barlow et aL6 with a pycnometric method on dilute aqueous solutions a t 25’. In their NMR line width study of 79Br nuclei in aqueous solutions of various ammonium, phosphonium, and sulfonium salts, Forsen and his coworkers have included azoniaspiro[4.4]nonane bromide.6 Lowe and Rendal17 reported viscosity and density, as well as conductance measurements of dilute solutions of some “monocyclic” quaternary ammonium iodides. We are interested in solutions of this type of salt primarily in connection with the studies on solutions of tetraalkylammonium salts. Aqueous solutions of symmetrical tetraalkylammonium salts show many interesting properties which are now being actively investigated in various laboratories.a10 As pointed out by Friedrnanll and others,12 we have to consider the following factors for a better understanding of properties of solutions containing large tetraalkylammonium ions: (i) flexibility of alkyl chains, (ii)
nonspherical shape of the cations, and (iii) penetration of the anion into the space between the alkyl chains. The flexibility of alkyl chains in a large tetraalkylammonium ion would be greatly reduced if these alkyl chains are linked to form loops. Since these azoniaspiro ions have no terminal methyl group, comparison of the solution properties of these salts with tetraalkylammonium salts of corresponding molecular size is expected to reveal differences between flexible alkyl chains and relatively inflexible methylene groups forming loops. In this communication we report our experimental results on the enthalpies of solution and dilution of some of the halides in H20 and D20 a t 25O. Experimental Section 1. Materials. The general methods for preparing some of the azoniaspiroalkane halides ( C H Z ) ~ N + ( C H ~ ) were ~Xintroduced by von Braun and coworkers13 in 1906 and expanded in his later a r t i ~ 1 e s . lWe ~ have modified some of the recently improved methods for synthesizing these compounds by Blicke and H ~ t e l l i n g , ’and ~ by Thomas and his coworkers.16J7 For convenience we shall use abbreviated names for these compounds, denoting only the halide and the number of carbon atoms in each ring. For example, a spiro quaternary ammonium halide, C12Hz4NBr, will be called “6.6 bromide” in place of its IUPAC name of 7-azoniaspiro[6.6]tridecane bromide or its common name, 1,l-spirobi(hexamethy1eniminium bromide). Our method of synthesis may be summarized as follows. Reagent grade a,w-dihaloalkane (0.1 mol) was allowed to The Journal of Physical Chemistiy, Vol. 79, No. 15, 1975
David P. Wilson and Wen-Yang Wen
1528
react with equimolar cyclic amine (having the same number of methylene groups as the dihaloalkane) in water or 2-propanol containing 0.1 mol of NaOH. After evaporation to dryness, the crude product was extracted and precipitated with chloroform and purified by Soxhlet extractions with methyl ethyl ketone followed by several recrystallizations from 2-propanol and absolute ethanol. (Some variations are necessary for synthesis of different compounds. For details, see ref 18.) Yields of the reaction, melting point, results of analysis, and solid densities obtained for the five compounds are given. (a) (CH&N+(CHz)dBr-, “4.4 bromide’’ (5-azoniaspiro[4.4]nonane bromide): final yield, 8.0 g (39%), mp 269-270’ (dec), lit. 25O-252’,l5-l7 261-262’,16 261.5-262°.5 Anal. C ~ H ~ G61.23; N , Br, 38.70; d = 1.40 g/ml. (b) (CH2)5Nf(CH2)5Br-, “5.5 bromide” (6-azoniaspir0[5.5]undecane bromide): final yield, 15.0 g (64%), mp 331-332’, lit. 300°,5 3040,17 3O9O,l6 311-3120,15 335-336°.4 Anal. C ~ O H Z O65.67; N , Br, 33.96; d = 1.34 g/ml. (c) (CH&,N+(CHz)&l-, “5.5 chloride” (6-azoniaspir0[5.5]undecane chloride): final yield, 5.4 g (28%), mp 350351’ (dec), lit. 310-311°.15 Anal. CIOHZON, 81.27; C1, 18.65; d = 1.15 g/ml. (d) (CH2)5Nf(CHz)51-, “5.5 iodide” (6-azoniaspir0[5.5]undecane iodide): final yield, 8.0 g (28%), mp 253255’, no literature value available. Anal. CloHzoN, 53.17;39 I, 44.91; d = 1.54 g/ml. (e) (cH2)6N+(CHz)&r-, “6.6 bromide” (7-azoniaspiro[6.6]tridecane bromide): final yield, 8.1 g (31%), mp 284.5~ N , Br, 30.47; d = 285.5’, lit. 281-282’.15 Anal. C ~ Z H ~69.32; 1.37 g/ml. The salts were analyzed gravimetrically for the total amount of cations and anions. The cation analysis followed the procedure of Flaschka and Barnard,lg where the quaternary ammonium ion was precipitated in neutral solution as the tetraphenylboron salt by addition of a twofold excess of purified 3% sodium tetraphenylboron reagent. The anion analysis followed the standard silver halide gravimetric procedure. Triplicate determinations on the cation and anion agreed to within 0.2%. Proton magnetic resonance spectra were taken to check the purity of the compounds and to see if any side reaction such as polymerization occurred in the ring closure process. Densities of the solid compounds were determined by a flotation method similar to that described by Wulff and Heigl.20 The solvents used to form mixtures were benzene and carbon tetrachloride. Deuterium oxide was obtained from the Stohler Isotope Chemicals, Inc. and used without further purification. Its proton content after experiment was checked with a highresolution NMR spectrometer indicating that the DzO content was 99.7% or higher. 2. Calorimeter. A constant-temperature environment or isoperibol calorimeter used has been described previous1y.21J2The calorimeter with some recent modificationslB was standardized by measuring the heat of solution and tris(hydroxymethy1)aminomethane neutralization of (THAM) in excess 0.103 m HCl. The average heat of solution a t 25O for seven runs was -7118.0 cal/mol with 95% confidence level of f 7 . 3 cal/mol (1 cal = 4.1840 J). This result compares favorably with other values reported, particularly the most recent ones.23-27The main sources of error in the calorimetric experiment were the heats of breaking the glass bulbs containing samples add the error in the measurements of the temperature rise, AT, depending on The Journal of Physical Chemisfry, Vol. 79, No. 15, 1975
TABLE I: Enthalpies of Solution of Azoniaspiroalkane Halides in H20 at 25” a C8H1,NBr (4.4 B r )
Ci,H2,NBr (5.5 B r )
10‘ rnSi/? (01
5.43 8.20 10.44 13.46 17.69 21.85 26.16 31.97 35.18 41.57 47.42 55.02
-
AH, 1097 1121 1133 1132 1126 1112 1091 1064 1022 998 945 895 824
i
* *
t
i
* *
k i
* * -i-
*
102
10 6 7 9 9 9 10 10 10 10 10 10 10
(0 )
5.12 8.39 11.40 14.69 17.69 20.60 23.51 27.03 30.08 34.15 36.69 41.33 43.47 48.67 54.80
CI2Hz4NBr(6.6 B r )
10‘ (0 )
102
3813 14 3830 i 14 3828 i 14 3843 * 18 3837 i 16 3840 f 20 3821 i t 1 3803 i 2 1 3776 * 22 3762 i 22
CloHzoNCl (5.5 C1) l o 2 rnSil2
AH, 2150 i 10 2171 x 7 2175 * 9 2170 * 10 2158 * 11 2143 i. 11 2125 i- 12 2108 + 12 2085 * 12 2061 i 13 2029 i 13 2008 i 13 1967 i 14 1948 i 14 1898 + 14 1840 ~t14
C,*H,,NBr (6.6 B r )
AFTS
3.49 3.56 5.46 6.11 7.53 10.67 14.55 18.95 22.50
TnSi/2
iil,‘/?
28.12 31.91 32.57 37.13 41.50 41 -71 49.86 54.08 59.32 62.62
AR, 3738 + 23 3720 i 23 3713 i 23 3683 * 23 3665 + 24 3647 * 24 3610 * 24 3573 i: 24 3531 i 25 3511 i 25
CiOH2,NI (5.5 I)
-
-
AH,
102 Yns’/?
AH,
4960 * 20 5007 i. 19 * 5.13 4976 i 23 7.57 4990 i 24 i 10.30 4974 i: 26 i * 13.17 4962 + 26 17.71 4938 * 27 i 21.45 4916 * 27 i i 27.22 4873 i 28 30.19 4851 i 29 f i 34.17 4818 k 29 37.80 4786 * 29 a m, = moles of salt per 1000 g of HzO: A H s is the enthalpy of solution in calories per mole. (01
3.90 6.76 10.13 13.57 16.56 20.73 24.13 28.20 33.74 37.84
-1268 -1256 -1234 -1236 -1234 -1240 -1253 -1267 -1286 -1318 -1340
i i
15 10 11 12 12 13 13 13 13 13 13
(0 1
3.04
the tolerances of the resistance elements of the Wheatstone bridge. The heats of bulb breaking ranged from 0 to 0.012 cal for bulb sizes from 0.25 to 1.5 ml. After many runs, the blank heats of bulb opening could be estimated to a precision of f0.003 cal. The error in the estimation of AT is about hO.l% which is the primary source of random error for large samples.
Aqueous Solutions of Azoniaspiroalkane Halides
1529
Results Enthalpies of solution and apparent molal heat contents of azoniaspiroalkane halides in H2O at 25’ and some of the bromides in D2O a t 25O are tabulated in Tables 1-111. For additive enthalpies of solution, where consecutive samples are introduced into solution and the individual enthalpy values recorded, the uncertainty in the enthalpy of solution for the ith consecutive sample bulb is
where Qj is the experimental heat change on dissolution and M , ( m j ) is the molar enthalpy of solution. The above uncertainty determines the errors of the data reported in Tables I-IV. The low-concentration enthalpy data have been extrapolated to infinite dilution using the Debye-Huckel limiting law. The apparent molal heat contents, & , ( m ) ,a t concentration m is given by’
4 d m ) = AR,(rn) - AR,(l/m) TABLE 11: Apparent Molal Heat Contents of Azoniaspiroalkane Halides in HzO at 25” bL, cal/mol
4.4 Bra
m1I2
5.5 Bra
6.6 Brb
5.5 Clb
5.5 IC
0 0 0 0 0 0 0.05 23 21 23 22 22 0.10 36 23 17 34 18 0.15 26 7 -1 5 31 -8 0 -20 4 -2 1 -42 18 -3 7 0.25 -2 5 -52 -67 -3 -7 1 0.30 -61 -87 -89 -2 8 -109 0.35 -99 -128 -115 -56 -151 0.40 -140 -172 -143 -198 0.45 -182 -218 -171 0 -50 -227 -264 -205 0.55 -273 -243 0.60 -283 a Average errors of data for the salt are about 15 cal/mol. Average errors of data for the salt are about 20 cal/mol. Average errors of data for the salt are about 25 cal/mol.
where AI?,(l/m) is the molar enthalpy of solution of a salt at infinite dilution.
Discussion 1. Apparent Molal Heat Contents i n Water. Our results for the azoniaspiroalkane halides are plotted in Figures 1 and 2 together with those of selected alkali halides28 and tetraalkylammonium halide^.^^,^^ According to Levine and Wood30 the effect of pair-wise interactions on the heat content per mole of salt may be summarized as follows. At low concentrations the deviation from the Debye-Huckel limiting law stems from both the like and oppositely charged interactions. As the concentration increases, the effect of oppositely charged interaction is reduced and the effect of the like-charged interaction is increased. Comparison of our results with those of the tetraalkylammonium halides seems to reveal the following trends. (i) In the concentration range studied, 4~ for 4.4 Br and 6.6 Br decrease with the concentration increase much faster than those of corresponding Et4NBr and PrdNBr, respectively. (ii) Below 0.09 m, the decreasing order of @L for the spiro bromides is 4.4 > 5.5 > 6.6, but above 0.09 m the order is 6.6 > 4.4 > 5.5. For tetraalkylammonium bromides, the decreasing order of $L is Me4NBr > Pr4NBr > Et4NBr for concentrations below 0.04 m, but it changes to Pr4NBr > MedNBr > Et4NBr for concentrations above 0.04 m and below 1.4 rn.30 (iii) The higher concentration (above 1.2 m) behavior of & for R4N+X- is “regular” depending on the size of the cation (Bu4N+ > Pr4N+ >- Et4N+ > Me4N+). The @L curve crossover of 6.6 Br with 5.5 Br a t about 0.1 m suggests that the trend for I#JL a t higher concentrations may similarly become “regular”. It is suspected that when 4.4 Br and 5.5 Br curves crossover as eventually do their tetraalkylammonium counterparts a t higher concentrations, the crossover may not take place until much higher concentrations. Aside from these details, the 4~ values of our three spiro bromides are close to each other and considerably lower (more negative) than those of the corresponding tetraalkylammonium bromides, presumably due to the “wrong shape” of these spiro ions as far as the surrounding water structure is concerned. In addition to the shape, the smaller size of the spiro ions should also contribute to their solu-
TABLE 111: Enthalpies of Solution and Apparent Molal Heat Contents of Azoniaspiroalkane Bromides in DzO at 25”
1550 * 10 1572 + 6 1577 f 8 1569 i 8 1556 i 9 1545 f 9 1510 i 9 1472 i 9 1428 * 10 1400 i 10 1355 * 10
0 4225 + 15 0 3.74 4240 i 13 15 5.90 4245 i 16 20 8.52 4245 + 18 20 -4 11.46 4240 i 19 15 -2 7 14.64 4214 i 20 -1 1 -40 * -62 21.21 4179 i 20 -46 29.18 -7 8 -9 i 23.68 4164 i 2 1 -61 34.08 -122 -156 28.87 4130 i 22 -9 5 37.22 -150 32.60 4107 i 22 -118 42.02 -195 36.64 4077 i 23 -148 a m‘ = moles of salt per 55.51 mol of DzO.AHs is the enthalpy of solution in calories per mole. is the apparent molal heat content in calories per mole. 0
5.14 8.55 12.53 16.36 20.36 24.54
0
0
22 27 19 6 -5
3.64 5.78 9.06 12.52 17.26 23 -06 27.03 34.79
2576 2596 2592 2590 2572 2549 2514 2485 2420
15 12 15 16 i 16 i 17 17 f 17 i 17 i
0
i i i
20 16 14
The Journal of Physical Chemistry, Vol. 79, No. 15, 1975
1530
David P. Wilson and Wen-Yang Wen
TABLE IV: Ionic Enthalpies of Transfer from HzO to D2O at 25"
AH,,,
AH,,,
Ion
kcal/mol
(cH,),N+(cH,), (cH,),N+(cH,)~ (cH,),N+(cH,), (cH,),N+
0.36 0.02 0.34 i 0.02 0.32 i 0.02 0.43'
a
kcal/mol
Ion
*
( c ~ H , ) ~ N + 0.21" (c~H,),N+ -0.05" (c,H,),N+ -0.230 Br0 .09'
From ref 33.
I
I 2001
IO0
1
- LiSr
I
-- _- _ -NaBr --_ KBi
- -Fr4NBt
\
',Me4NBr -300 0.1
0
0.2
0.3
0.4
0.5
0.6
0.7
0.8
m Figure 1. Apparent molal heat contents of the azoniaspiroalkane bromides in H20 at 25' compared with those of some alkali bromides and tetraalkylammonium bromides.
-
-
L O
-E
0-'
0
4" -100-200Et4NI
-300 0
0.1
0.2
m
0.3
0.4
Figure 2. Apparent molal heat contents of 5.5 halides In H20 at 25' compared with those of some tetraalkylammonium halides. tion properties. Our recent experimental results31 show that the apparent molal volumes at infinite dilution of 4.4 cation and 6.6 cation in water are about 85% of those of Et4N+ ion and Pr4N+ ion, respectively. We also found that the difference in apparent molal volumes a t infinite dilution between Pr4NBr and 6.6 Br is about 10.5 ml/mol greater than the corresponding difference between Et4NBr and 4.4 Br. It means that the apparent molal volume per methylene or methyl group is larger when the group is farther away from the charged nitrogen center. This seems to provide some indication of the flexibility of the alkyl chains in the larger tetraalkylammonium ion. Since the terminal groups should be most flexible, they are expected to make a substantial contribution to the excess in apparent molal volumes.31 The Journal of Physical Chemistry, Vol. 79, No. 15, 1975
In a related way, the apparent molal heat content of a larger tetraalkylammonium ion is expected to be influenced by the swaying terminal methyl groups, since the flexible alkyl chains can fit into the surrounding water structure more easily than the relatively inflexible methylene groups connected to form a tight ring. 2. Enthalpy of Transfer. The thermodynamic functions for the transfer of a salt a t infinite dilution from one solvent to another reflect differences in solvation, work of cavity formation, dipole interaction, etc. One term which contributes significantly to the enthalpy of solvation in a highly structured solvent is that due to changes in the solvent structure induced by the presence of the solute species. For a transfer process between two aqueous media differing primarily by degree of structuredness, such as DzO H20,32*33 this effect may be the primary contribution to the enthalpy of transfer. Krishnan and Friedma113~have studied the enthalpies of transfer of sodium n- alkylsulfonates from water to heavy water and water-alcohol mixtures as a function of the length of the alkyl chains. Their results illustrated clearly the strong interaction between the hydration regions near the chains and the terminal methyl groups. The molal enthalpies of transfer for the azoniasprioalkane bromides can be obtained from the enthalpies of solution in HzO and D20 given in Tables I and 111. If we assume a value of 0.09 kcal/mol for the enthalpy of transfer of Br- ion from H2O to D20 as reported by Krishnan and Friedman,32 then we can obtain the ionic enthalpies of transfer a t infinite dilution for the cations. The values obtained are given in Table IV together with those of tetraalkylammonium ions.32 Based on the ratio of Walden product in DzO to that in HzO, Kay and Evans35 concluded that Et4N+ ion has very little effect on the structure of water, Me4N+ ion has structure breaking effect, while Pr4N+ ion and Bu4N+ ions have structure enhancing effect. These interpretations are consistent with the results on the enthalpy of transfer reported by Krishnan and Friedma11.3~If we compare our enthalpy of transfer data with those of tetraalkylammonium ions, the hiit, values of our spiro ions are all greater than that of Et4N+ ion and less than that of Me4N+ ion. Levine and Wood30 have plotted b ~ ( D 2 0 )- +L(H~O) against aquamolality m' for the R4NBr series. Their plots are compared with our plots for the spiro compounds in Figure 3. Three curves for the spiro bromides are seen to cluster below the R4NBr curves. The curve for our 6.6 Br (which is the highest among the three spiro compounds) turns lower than that of Me4NBr, leading to a dubious implication that it is a water-structure breaker, if we are to follow the frequently used argument based on the sign and trend of heats of transfer from H2O to DzO. We are puzzled by this "structure breaking" implication for the (CHz),jN+(CHz)e ion, because our recent studies on apparent molal heat capacities and viscosities of the salt indicate that 6.6 ion is definitely a structure maker.31 Caution should be exercised in any structural interpretation which relies solely on the enthalpy of transfer data a t one temperature. In their recent studies of bolaform electrolytes in HzO and DzO, Burns and Verral136 found that AB,, (D20 HzO) values are small positive for all of the solutes investigated. This would classify the solutes as H20) structure breakers. However, their AC,(tr) (DzO values are relatively large positive, which would classify the solutes as structure makers.
-
+-
-
Aqueous Solutions of Azoniaspiroalkane Halides
1531
tion. Acknowledgment is also made to the donors of the Petroleum Research Fund, administered by the American Chemical Society. References a n d Notes I. B. Wilson, J. Bid. Chem., 197, 215 (1952). F. Bergmann and A. Shimoni, Biochem. Biophys. Acta, 10, 49 (1953) 8. D. Roufogalis and J. Thomas, J. Pharm. Pharmacol.,22, 649 (1970). J. Thomas and D. Clough. J. Pharm. Pharmacol,, 15, 167 (1963). R . B. Barlow, B. M. Lowe, J. D. Pearson, H. M. Rendall, and G. M. Thompson, Mol. Pharmacol., 7, 357 (1971). (6) H. Wennerstrom, B. Lindman, and S. Forsen, J. Phys. Chem., 75, 2936 11971). (7) E. M.'Lowe and H. M. Rendall, Trans. Faraday Soc., 67, 2318 (1971); 68., 2191 11972). - -, (8) F. Franks, "Effects of Solutes on the Hydrogen Bonding in Water" in "Hydrogen-Bonded Solvent Systems", A. K. Covington and P. Jones, Ed., Taylor & Francis, London, 1968, p 41. (9) W.-Y. Wen, "Aqueous Solutions of Symmetrical Tetraalkylammonium Salts" in "Water and Aqueous Solutions", R. A. Horne, Ed., Wiley, New York, N.Y., 1972, Chapter 15, p 613. (IO) W.-Y. Wen, J. Solution Chem., 2 , 253 (1973). (11) P. S. Ramanathan, C. V. Krishnan, and H. L. Friedman, J. Solution Chem., 1, 237 (1972). (12) R. W. Kreis and R. H. Wood, J. Phys. Chem., 75, 2319 (1971). (13) J. von Braun, C. Muller, and E. Beschke, Berichte, 39, 4337 (1906). (14) J . von Braun, Berichte, 49, 966 (1916); 56, 1994 (1923). (15) F. F. Blicke and E. B. Hoteliing, J. Am. Chem. Soc., 76, 5099 (1954). (16) J. Thomas, J. Med. Pharm. Chem., 3, 45 (1961). (17) 8. D. Roufogalis and J. Thomas, J. fharm. Pharmacol., 20, 153 (1968). (18) D. P. Wilson, Ph.D. Thesis, Clark University, 1974. (19) H. Flaschka and A. J. Barnard, Jr., Adv. Anal. Chem. Instrum., 1, 24 (1960). (20) P. Wulff and A. Heigl, 2.Phys. Chem. A, 153, 187 (1931). (21) R. B. Cassel and W.-Y. Wen, J. Phys. Chem., 76, 1369 (1972). (22) R. B. Cassel, Ph.D. Thesis, Clark University, 1971. (23) R. J. Irving and I. Wadso, Acta Chem. Scand., 16, 195 (1964). (24) S. R . Gunn, J. fhys. Chem., 69, 2902 (1965). (25) I. Petersson, Acta /M€KO, Proc. lnt. Meas. Conf., 4th, 2, 337 (1968). (26) M. V. Kllday and E. J. Prosen, Abstracts, 26th Annual Calorimetry Conference, Orono. Maine, 1971. (27) J. D. Navratil and F. L. Oetting, Abstracts, 27th Annual Calorimetry Conference, Park City, Utah, 1972. (28) V. B. Parker, Natl. Stand. Ref. Data. Ser., Natl. Bur. Stand., No. 2 (1965). (29) S. Lindenbaum, J. fhys. Chem., 70, 814 (1966). (30) A. S.Levine and R. H. Wood, J. fhys. Chem., 77, 2390 (1973). (31) Unpublished resuits of W.-Y. Wen, A. LoSurdo, C. Jolicoeur, and J. Boileau. Details will be published shortly. (32) C. V. Krishnan and H. L. Friedman, J. Phys. Chem., 74, 2356 (1970). (33) C. Jolicoeur and G. Lacroix, Can. J. Chem., 51, 3051 (1973). (34) C. V. Krishnan and H. L. Friedman, J. Solution Chem., 2, 37 (1973). (35) R. L. Kay and D. F. Evans, J. fhys. Chem., 69, 4216 (1965); 70, 366, 2325 (1966). (36) J. A. Burns and R. E. Verrali, J. Solution Chem., 2, 489 (1973). (37) P. R. Philip and C. Jolicoeur, J. Solution Chem., 4, 105 (1975). (38) H. S. Frank, "Structural Models", in "Water, A Comprehensive Treatise", Vol. 1, F. Franks, Ed., Plenum Press, New York, N.Y., 1972, p 527, footnote. (39) The low percentage of cation analysis reported here was due to the difficulties encountered In filtering the fine collodial precipitates formed with the sodium tetraphenylboron reagent. The 5.5 iodide prepared is believed to be better than 99 % pure.
(1) (2) (3) (4) (5)
-40-
\4:4
Br
~
In an interesting study, Philip and J o l i ~ o e u rused ~ ~ the scaled-particle theory to investigate the solvent isotope effects. Their conclusion should be taken as a timely warning for our simple interpretations. "In the DzO HzO transfer functions of inert solutes, departure from hard-sphere behavior can be attributed to two possible sources: a solute geometrical effect and a local solvent molecular reorientational (structural) effect; the latter does not seem dominant in any of the cases examined here." For this reason and others,38 it is important to study various properties of the solutes by different techniques, and we are currently engaging in the measurements of apparent molal volumes and heat capacities, viscosities, free energies, as well as some nuclear magnetic properties of these azoniaspiroalkane halides in solutions. In conclusion, our experiments show that forming closed loops from the alkyl chains in tetraalkylammonium ions has a great effect upon the thermodynamic properties, particularly those which have been regarded as characteristic of aqueous tetraalkylammonium salts. The hydrophobic interaction with the solvent water is weakened considerably for the azoniaspiroalkane ions presumably due to their "'wrong shape", lack of swaying terminal methyl groups, and flexible alkyl chains.
-
Acknowledgments. I t is our great pleasure to acknowledge the financial support of the National Science Founda-
.
The Journal of Physical Chemistry, Voi. 79,No. 15, 1975
