3688
THOMAS C. WEHMANAND ALEXANDER I. POPOV
Charge-Transfer Complexes of Mono- and Disubstituted Tetrazoles with n-Electron Acceptors
by Thomas C. Wehman and Alexander I. Popov Department of Chemistry, Michigan State University, East Lansing, Michigan
(Received A p r i l 11, 1966)
The complexing ability of the tetrazole ring was investigated in order to determine the a-donor ability of tetrazoles. It is shown that 5-substituted and 1,j-substituted tetrazoles readily form charge-transfer complexes with x acids. Formation constants of these complexes were measured spectrophotometrically in dichloromethane at 25'. There is a reasonable agreement between the magnitudes of the formation constants and the inductive effects of the substituent groups on the tetrazole ring. The formation constant of the pentamethylenetetrazole complex with 1,3,5-trinitrobenzene has also been measured by nuclear magnetic resonance technique.
Introduction indicate the extent of x-electron donor ability of the tetrazole ring. In this respect, possible complexes Substituted tetrazoles have been noted for their with tetracyanoethylene (hereafter abbreviated as wide spectrum of neurological activity which, depending TCNE) were investigated since it has been shown by on the substituent group, ranges from strong stimulanumerous investigators that TCKE is a strong x tion of the nervous system to a depressant action.' acid.6 Also, complexing tendencies of the tetrazoles Since it is entirely conceivable that the neurological were studied with other x acids such as tetracyanoactivity is related to the physicochemical properties , 1,3,5-triquinodimethane (TCNQ), chloranil (CA) of the tetrazoles, a thorough investigation of tetrazole (TSB), and 2,4,7-trinitrofluorenone nitrobenzene derivatives was initiated some time ago. Among other (TNF). properties, the ability of the tetrazole ring to form complex compounds was investigated and it was soon Experimental Part obvious that the tetrazoles formed moderately strong With the exception of pentamethylenetetrazole, all complexes with a number of Lewis acids such as the halogens2or the transition metal ions.3 Although the complexes formed by tetrazoles were, (1) F. W. Schueler, S. C. Wang, R. N. Featherstone, and E. G . in general, well characterized, the manner in which Gross, J . Pharmaeol. E z p t l . Therap., 97, 266 (1949), and references the tetrazole ligand was bonded to the respective Lewis listed therein. acid was not clear. Brubaker and D ~ u g h e r t y ~ ~(2) (a) A . I. Popov, C. C. Bisi, and M. Craft, J . Am. Chem. SOC.,80, 6513 (1958); (b) A. I. Popov, R. E. Humphrey, and W. B. Person, have suggested that in the case of metal-tetrazole comibid., 8 2 , 1850 (1960); (c) J. W. Vaughn, T. C. Wehman, and A . I. Popov, J. Inorg. Nucl. Chem., 2 6 , 2027 (1964). plexes, the bonding may occur either through one or two (3) (a) A. I. Popov and R. D. Holm, J . Am. Chem. SOC., 81, 3250 nitrogens of the tetrazole ring or that the central (1959); (b) C. H. Brubaker and N. A. Daugherty, ibid., 83, 3779 metal ion may coordinate to the n-electron system of (1961); (c) C. H. Brubaker and G. L. Gilbert, Inorg. Chem., 2, 1216 (1963). the tetrazole ring. The last alternative was originally (4) (a) A . D. Harris, H. H. Herber, H. B. Jonassen, and G. K . favored by Jonassen and his c o - ~ o r k e r s ,but ~ in a Werthein, J . Am. Chem. SOC.,85, 2927 (1963); (b) H. B. Jonassen, recent paper on the tetrazole-iron(I1) c ~ m p l e x , ~ T. 0. Terry, and A. D. Harris, J . Inorg. Nucl. Cllem., 2 5 , 1239 (1963). they cite stronger evidence for a u bond from the nitro(5) A. D. Harris, H. B. Jonassen, and R. D. Archer, Inorg. Chem., 4 , gen to the metal ion. 147 (1965). It seemed to us that a study of possible complex (6) R. E. Merrifield and W. D. Phillips, J . Am. Chem. SOC.,8 0 , 2778 formations between tetrazoles and Lewis x acids may (1958). ~
T h e .Journal of Physical Chemistry
~~
~~
~
_ _ _ _ _ _ _ _ _ ~
CHARGE-TRANSFER COMPLEXES OF MONOAND DISUBSTITUTED TETRAZOLES
other tetrazoles used in this investigation were synthesized by the following methods. a. &Substituted tetrazoles were obtained by treating sodium amide with the corresponding nitrile' following the general reaction R-C-N-H
\/ N
Using the above method, the following, previously reported tetrazole derivatives were obtained: 5methyltetrazole, mp 149O, lit. 148'; 5-ethyltetrazole, mp 96', lit. 98'; 5-propyltetrazole1 mp 64', lit. 64'; 5-c~-phenylpropyltetrazole,mp go', lit. 93'; and 5benzyltetrazole, mp 125', lit. 125'. I n addition, the following two new compounds were synthesized : 5p-chlorobenzyltetrazole (Anal. Calcd: C, 49.34; HI 3.62; N, 28.80. Found: HI 3.61; C, 49.08; N, 28.70; mp 164') and 5-p-methylbenzyltetrazole (Anal. Calcd: C, 62.07; HI 5.78; N, 32.15. Found: C, 62.04; HI 5.78; N, 32.10; mp 161'). b. l15-Substituted tetrazoles were prepared by the method of Herbst, et ala8
done at room temperature of approximately 25'. All nmr measurements were made on carbon tetrachloride solutions with a Varian Associates A-60 spectrometer operating at 60 Mc/sec. The probe temperature was approximately 35'. Tetramethylsilane was used as the reference for chemical shift measurements. Methods of solution preparation for mole ratio investigations have been described previously.2 In A i solutions general, work was done with to of TCNE. A rather large excess of the tetrazoles had to be added to produce measurable spectral shifts. In all cases, the tetrazole complexes were too weak to be obtained in the solid form.
Results Formation constants of the TCSE-tetrazole complexes were obtained from spectrophotometric data by using the Ketelaar modification of the Benesi-Hildebrand m e t h ~ d . A ~ series of solutions was prepared in which the concentration of a tetrazole (donor) was variable and much higher than the concentration of TCNE (acceptor). Absorption spectra were obtained in the 300-450-mp region. Since tetrazoles do not absorb in this region, Ketelaar's equation could be used. 1
R1-C--N-R2
The following compounds were prepared by the above method: l-ethyl-5-phenyltetrazole, mp 69O, lit. 69'; 1-phenyl-bethyltetrazole, mp 46', lit. 49'; and 1,5-diphenyltetrazole, mp 146O, lit. 145'. Pentamethylenetetrazole (PAIT) was obtained from the Knoll Pharmaceutical Co. and purified as described previously ;*a p-chloranil, 1,3,5-trini trobenzene, and 2,4,7-trinitrofluorenonewere obtained from Eastman. They were purified by several recrystallizations. The melting points of purified products agreed well with the literature values. Tetracyanoethylene was originally obtained from the Du Pont Co. and subsequently from Eastman. It was recrystallized twice from chlorobenzene and then sublimed twice in an inert atmosphere, mp 199', lit. 200'. Tetracyanoquinodimethane was obtained as pure material from the Du Pont Co., mp 294'. Absorption measurements in the visible and ultraviolet regions were made on a Cary recording spectrophotometer >lode1 14, in silica cells of 1.00, 5.00, and 10.00 f 0.01 cm path lengths. Measurements were
3689
et
-
1
Ea
e,
-
ea
1 X-+KrCd
1 eo
-
€8
where Et is the apparent molar absorptivity of the solution (ie., absorbance divided by the total concentration of TCNE); ea is the molar absorptivity of T C S E ; is the molar absorptivity of the complex; C d is the total concentration of the donor = equilibrium concentration of tetrazole; and K f is the formation constant of the complex. A plot of l / ( t t - ea) us. 1 / C d gives a straight line. The intercept then yields the value of e,, and the slope, that of Kf. The results are given in Table I. The slope and the intercept were obtained by using a standard least-squares Fortran computer program on a CDC 3600 computer. It has been pointed out in a recent paperlo that the linearity of a Ketelaar plot alone is not a valid proof for the existence of only 1:l complex in solution but that the value of the formation constant should be independent of the wavelength at which the measurements were made. Table I indicates that in most cases (7) W. G. Finnegan, K. A. Henry, and R. Lofquist, J . Am. Chem. SOC.,8 0 , 3908 (1958).
(8) E.K.Harvill, R. M.Herbst, E. C. Schreiner, and C. W. Roberts, J. Org. Chem., 15, 662 (1950). (9) J. A. A. Ketelaar, C. van der Stolpe, A. Goudsmit, and W.
Dscubas, Ree. Trat. Chim., 71, 1104 (1952). (10) G. D.Johnson and R. E. Bowen, J . Am. Chem. Soc., 8 7 , 1655 (1965).
Volume 70. Number 11 ,vovember 1966
3690
THOMAS
c. WEHMAN AND ALEXANDERI. POPOV
Table I : Tetracyanoethylene Complexes with Tetrazoles Wavelength, rnr
Tetraaole
PMT
300
305
310
315
320
1.33 2260
1.32 1630
1.32 1120
1.30 760
1.28 490
1.31 f 0 . 0 2
1.90 250
1.94 850
1.80 600
1.88 f 0.07
1-Cyclohexyl-5-E tTzb
325
330
335
340
Kava
1-Cyclohexyl-5-MeTa
1.30 2030
1.50 1250
1.38 900
1.39 f 0 . 1 0
1-Me-5-cyclohexyl-Te
1.94 1430
2.20 900
1.93 400
2.02 f 0.14
5-PrTz
2.00 3.60
1.59 2.60
1.85 1.80
1.81 f 0.20
0.84 1250
1.08 1690
1.5-DiPhTz
0.89 1190
l-Et-5-PhTz 1.21 1000
1-Ph-5-EtTz a
Formation constants are given in liters per mole.
1.22
850
Table I1 : Charge-Transfer Complex of TCNE with 5-Benzyltetrazole 370 mr
390
410
mr
mp
mr
Kav
KC
2.26
2.27
2.31
2.28
2.28 f 0.02
E
#77
89
98
89
The Journal of Physical Chemistry
1.44 470
1.39 440
1.20 630
1.18 540
0.90 f 0.15 1.48 390
1.53 350
1.46~t0.07 1.21 f 0.03
* M e = methyl; E t = ethyl; P r = propyl; P h = phenyl; Te = tetrazole.
this additional condition is reasonably well fulfilled within the limitations of the measurements. The only exception seems to be the complex of the 1,5-diphenyltetrazole, but here again the trend is not outside the expected experimental error. It is interesting to note that in all cases the absorption maxima of new charge-transfer bands were not observed, but rather the absorption band of the TCNE broadened considerably and extended to lower frequencies (Figure 1). Absorption measurements were, consequently, carried out on the rather steep side of the new absorption band with consequent loss of accuracy. An absorption maximum, however, was obtained in the case of 5-benzyltetrazole (Figure 2 ) . The experimental data are given in Table 11. While there seems to be little doubt that the new band is due to the complex, in view of relatively low molar absorptivity, it is doubtful that this is a charge-transfer band. Since TCNE is very reactive chemically, it was im-
350
1.24 720
0.78 1090
portant to determine whether the spectral change observed in the TCNE-tetrazole systems was not due to a side reaction. Tetracyanoethylene readily forms the radical anion TCNE - or it can undergo a reduction to H2TCNE. Both of these substances were synthesized and their spectra were determined. These spectra showed good agreement with the data in the literature." The comparison of these spectra with the results obtained in the TCNE-tetrazole systems made it evident that the spectral changes observed in these systems above could not be due to the formation of either the radical anion or of the reduction product. It is also known that TCNE can react iryeversibly with strong bases. Solutions of TCKE were prepared in methylene chloride containing piperidine, triethylamine, or pyridine. The spectra of these solutions were found to be time dependent and slow, and irreversible formations of colored adducts were observed. The spectral changes were entirely different from those observed with the tetrazoles. Spectral shifts similar to those obtained in TCNEtetrazole systems were also observed when PAIT was added to solutions of other ir acids, namely trinitrobenzene (TNB), chloranil (CA) , trinitrofluorenone (TNF), and tetracyanoquinodimethane (TCNQ). (11) (a) 0. W. Webster, W. Mahler, and R. E. Benson, J . Am. Chem. Soc., 84, 3678 (1962); (b) J. Prochorow and A. Tramer, Bull. Acad. Polon. Sci., Ser. Sci. Math., Astr. Phys., 12, 589 (1964).
CHARGE-TRANSFER COMPLEXES OF MONO-AND DISUBSTITUTED TETRAZOLES
3691
Table 111: Comparison of Complex Strengths Donors
Formation constants TNB
TNF
I¶
0.10
0.06
7.5*
0.33
0.23
(very small)
0.15
...
0.55
(very small)
1.52
TCNE
TCNQ
CA
1.31
0.22
0.16
PMT ( K , in CHZCIZ) Benzene’ ( K , in CC4) Hexamethylbenzene“ ( K , in CCh)
1.03
2.54
1.40
a Values from L. J. Andrews and R. M. Keefer, “Molecular Complexes in Organic Chemistry,” Holden-Day, Inc., San Francisco, Calif., 1964, p 97. The values of formation constants given in the reference are converted to liter mole-’ units to compare with our values. Reference 2a.
’
0.8
0.7
0.E
0 .e
e5 c
d
O.d
0.1
0 .I
0.
weve1engtIl in mlr
400
350
450
500
HavelcnKtb i n mp
Figure 1. Curve A, 5.00 X lo-’ M TCNE in CHZCh; curve B, 6.0 X 10-4 M TCNE large excess of benzene (benzene-TCNE c-t absorption band); other curves, 5.00 X 10-8 M TCNE excess of 5benzyltetrazole; figures on the curves indicate mole ratios Tz:TCNE.
+
+
Formation constants of the resulting complexes were calculated by the same technique and the results are given in Table HI. It is obvious from Table I11 that the r acids listed in the table are much weaker r-electron acceptors than TCKE. The absolute values of formation constants of the PRlT complexes with CA, TNB, and T N F cannot be very accurate and probably should not be taken too seriously. Since the formation constants are so small, measurable spectral changes can only be obtained with very large excesses of one of the reagents,
Figure 2. Pentamethylenetetrazole-TCNE complex in CHpClz: curve A, 1.00 X M TCNE; other curves, PMT 1.00 X loe3 M TCNE at various mole ratios PMT: TCNE.
+
and this greatly magnifies the experimental error.12 It seems that the data may at least indicate the relative strengths of the P M T complexes. As seen from Table 111, the strength of the PMT complexes follows closely the trends of the other two donors, benzene and hexa‘methylbenzene. This is another indication that tetrazoles form T complexes with T acids. Recently, a new method for the determination of complex formation constants in solutions has been proposed by Hanna and Ashbaugh.13 They have shown that in a donor-acceptor interaction to form a 1:1 charge-transfer complex, the chemical shift of the ac(12) W. B. Person, J. Am. Chem. SOC.,87, 167 (1965). (13) M. W. Hanns and A. L. Ashbaugh, J . Phys. Chem., 68, 811 (1964).
Volume 70, Number 11 Nmember 1966
THOMAS C. WEHMAN AND ALEXANDER I. POPOV
3692
where A is the difference between the observed shift of the acceptor protons in the presence of the donor and the chemical shift of the uncomplexed acceptor, A, is the difference between the shift of the acceptor protons in pure complex and the shift of the uncomplexed acceptor, K , is the formation constant of the complex, and Cd is the total concentration of the donor. Just as in the case of optical spectra, one measures the chemical shift of the acceptor protons in a series of solutions containing varying concentrations of the donor and plots l / A 21s. 1/Cd. A straight-line plot is obtained in the case of 1:l complexes and the values of A, and K, are obtained from the slope and the intercept. This nmr technique was used very successfully by Foster and Fyfe in the study of the dinitrobenzene and trinitrobenzene complexes with aromatic compounds. l5 The method was used in this investigation for the determination of the PRIT-TIVB complex formation constant in carbon tetrachloride solutions. Plot of l / A us. l/(PMT) gave a satisfactory straight line. (See Figure 3.) All chemical shifts were measured from the standard tetramethylsilane with an estimated accuracy of 0.2 cps. The formation constant of the complex was found to be K, = 1.3, which is higher by an order of magnitude than the constant determined spectrophotometrically. This is not surprising, however, since the latter was determined in dichloromethane which is a polar solvent, as compared to carbon tetrachloride, and can itself participate in the complexation reaction.16 I n order to verify this point, the formation constant of the complex was also determined spectrophotometrically in carbon tetrachloride solutions. Once again, no e-t band was observed and the measurements had to be made at the broad absorption tail as in the case of TCXE-tetrazole complexes. The results are summarized in Table IV. Good agreement of the two values obtained by quite different techniques seems to confirm the validity of our assumption that the increase in absorption such as shown in Figure 1 is due to complex formation.
Discussion It seems reasonable to conclude from the above data that the tetrazole ring does indeed possess some Tdonor ability and is capable of forming charge-transfer The Journal qf Physical Chemistry
r
0.60
ceptor protons is related to the strength of the donoracceptor interaction. They derived a relationship analogous to that of Benesi and Hildebrand used in spectrophotometric studies.14 The equation is
O.'O
t'
I
0.00
I
I
I
1 conc.
I
2 .o
1.0
I
3.0
rn
Figure 3. Plot of the TNB proton shift with increasing concentration of PMT.
Table IV: Spectrophotometric Determination of K2 for the Trinitrobenzene-Pentamethylene Tetrazole Complex in Carbon Tetrachloride
K2 E
267.5
272.5
mr
mr
277.5
w
1.67 7550
1.52 4300
1.73
K . " 1.64 =t0.10
2160
complexes with T acids. As seen from Table I, 11, and 111, there is a reasonable agreement between the electron inductive effect of the substituent groups and the stability of the complexes. It is also int,eresting to compare the formation constants of the 1-cyclohexyl-5-methyltetrazole-TCNE complex with that of 1-methyl-5-cyclohexyltetrazoleTCNE complex as well as the complexes formed by 1phenyl-5ethyltetrazole and l-ethyl-5-phenyltetrazole. It is seen that when the larger group is in the 1-position of the tetrazole ring, the stabilities of the complexes are less than when the groups are reversed. This can be explained by the fact that the carbon-carbon bond with the substituent in the 5-position is copolanar with the tetrazole ring. On the other hand, the nitrogen-carbon (14) H. A. Benesi and J. H. Hildebrand, J . Am. Chem. SOC.,7 1 , 2703 (1949). (15) R. Foster and C. A. Fyfe, Trans. Faraday Soc., 61, 1626 (1965). (16) For example, for the N.N-dimethylaniline-TNB complex, K O is 9.5 in cyclohexane, 3.4 in carbon tetrachloride, and 0.2 in sectetrachloroethane; R. Foster and D. L. Hammick, J. Chem. Soc., 2685 (1954).
LOCATION OF ADSORBED ETHYLENE IN A ZEOLITE
bond, with the substituent in the 1-position, is at an angle to the tetrazole ring. A large substituent group in the 1-position would, therefore, extend above and below the tetrazole ring and partially shield it from TCNE or other T acids with corresponding lowering of the stability of the complex. It is interesting to note that the formation constants of the benzene-TCNE and PMT-TCNE complexes are comparable (Table 111). On the other hand, the formation constant of iodine monochloride complexes with P M T is larger by three orders of magnitude than that of benzene-IC1 complex.2a This seems to indicate that in the former case,
3693
the complexation occurs through one of the nitrogen atoms on the P M T ring. A study of the crystal structure of the PICIT-IC1 complex is presently in progress and should answer this point unambiguously.
Acknowledgment. The authors gratefully acknowledge the support of this work by Research Grant MH07825 from the Institute of Mental Health, U. S. Public Health Service. They also wish to thank Professor W. B. Person of the University of Florida for many helpful discussions.
On the Location of Adsorbed Ethylene in a Zeolite
by D. J. C. Yates Esso Research and Engineering Co., Process Research Divwwn, Linden, New Jersey
(Received May 2, 1966)
During a previous spectroscopic investigation of the adsorption of ethylene on a series of ion-exchanged near-faujasite zeolites, it was found that Cd-X and Ag-X adsorbed ethylene quite strongly. I n the case of Ag-X, there were two distinct CH deformation bands in the adsorbed phase, which suggested two discrete sites for the adsorbed ethylene. The location of adsorbed ethylene in Ag-X has been studied in detail by adsorption studies, both gravimetric and volumetric, of ethylene and argon. It has been found that the most strongly held ethylene is adsorbed in the supercages, probably on those silver ions on the walls of the supercages. The more weakly held ethylene is probably interacting with the silver ions situated in the hexagonal windows between the supercages and the sodalite cages.
I n some recent work on the adsorption of ethylene on a series of near-faujasite zeolites,l it was determined that the ethylene adsorbed directly on the cations by T bonding. With Ag-X, the spectroscopic data showed the presence of two types of adsorbed ethylene. Further jnformat,ion on the location of adsorbed ethylene in Ag-X has now been obtained from adsorption isotherms of ethylene and argon. The most strongly held ethylene is adsorbed in the supercages, probably on those silver ions on the walls. The more weakly held ethylene is probably interacting with those
ions situated in the windows between the supercages and sodalite cages.
Experimental Section Gravimetric isotherms were measured2 with a Cahn microbalance. The volumetric argon and ethylene isotherms were measured in a conventional Pyrex (1) J. L. Carter, D. J. C. Yatetes, P. J. Lucchesi, J. J. Elliott, and V. Kevorkian, J . Phus. C h a . , 70, 1126 (1966). (2) D. J. C. Yates, ibid.; 69, 1676 (1965).
Volume 70, Number 1 1
Sovemliei l&,%
