CIIAKLES S.KRAIHAXZEL A N D ROBERT WEST
3G70
Vol.
s-1.
TABLE I1 SPIN-SPIN COUPLING COKSTANTS IS PERFLCOROALKYL DERIVATIVES OF SULFUR HEXAFLUORIDE* Compound
JCFI-SF~
JCFz-sFi
JCF~C-OFI
-'CFzC-CFz
JCF~CC-CFZ
JCFZ-BFR~ JCF2-SFab
JsF.-F~'
JC.F~-CP,
15.70 9.33 5 24,00 15.10 9.40 23.50 13.60 < 14.36 8.56 4.82 152.19 CF~CF~CF~CF~SF~~ 17.0 10.80 2,42 4.93 2.47 145.96 6 CF3SFaCF2SFse 22.92 21 . 4 2 5.28 1 A1 .87 a Coupling constants are given in units of cycles per second. SF, indicates the apex fluorine atom of the SFb g r o u p J W - Fis ~the coupling constant between the apex and base fluorine atoms in the SFs group. JCF~-C-JF, must be about 8-9 c.p.s. in this compound t o account for t h e SFa multiplet observed. Some assumptions concerning the relative magnitudes to be expected for certain of these coupling constants have been made in assigning them t o particular pairs of groups. e JSF~-C-SF~ must be about 10-12 c.p.s. in this compound to account for the SFamultiplet at -103.9 p.p.m. in Fig. 4. ( CFaCFz)SF4 CF3SF4CFZCF3 CF3SF.iCFzCOOCH3 CF3CF:SFj
three bond:;, one would also expect the throughbond coupling in the F-C-C-F fragment to be appreciable. It has indeed been foundL3 t h a t J trans = 18.6 c.p.s. and J gauche = 16.2 c.p.s. for CF2Br-CFBr2 and, a t least for the trans isomer, the through-space coupling should be small here. The hypothesis of Petrakis and SederholmL2t h a t through-bond coupling is essentially zero for interaction f h m i g h three single bonds therefore seems untenable and some other explanation of the low values for JCF~-CCF~ and JcF~--cF~ niust be sought. If the trans and gaiicize coupling constants were of opposite sign, then rotational averaging might indeed lead t o the observed results. Unfortunately (13) S. L . hfanatt a n d D. D. Ellernan, J. A l i i . Cheiiz. S O L .84, , 1305 ( 1962).
[ C o x m m c n u s FROM
Tiiti
the only study of relative signs of F-F coupling constants which has been madeL3indicates t h a t both trans and g u i d e coupling constants in CF2BrCFBr2have the same sign. Accidental mutual cancellation of through-bond and through-space interaction terms might also result in low J values but the problem remains a puzzling one. Acknowledgments.--We are indebted to Professor J. A. Young and R. D. Dresdner of the University of Florida for the gift of these compounds and for much helpful discussion and to Dr. P. T. Narasimhan and Dr. J. C. 11-oodbrey for assistance in the preliminary stages of this work. The continued support of this research b y the Atomic Energy Commission, through a grant, is gratefully acknowledged.
I~EPARTMEST OF CHEMISTRY, USIVERSITSOF WISCOSSIS, MADISON, WISCONSIN]
Hot Bands in the Infrared Spectra of Acetylenes in Solution BY CHARLESS.KRXIIIANZEL AND ROBERT \VEST RECEIVED APRIL 16, 196% The side hand accompanying the acetylenic C-13 stretcliing infrared band in solutions of acetylenes is sliown to I)c a "hot tjarid" absorption resulting from molecules in the first excited state with respect to the C-H bending mode. This appears to be the first reported observation of infrared hot bauds in solution.
Splitting of the acetylenic carbon-hydrogen stretching band has been noted in recent infrared spectral studies of A weak side band usually appears about 20 c m - l lower in frequency than the main band. Previous workers have suggested t h a t this vibrational doubling is due to Fermi resonance between the acetylenic C-H fundamental and a summation tone.'J T o investigate the nature of the side band more fully, we have investigated the spectral absorption of a number of acetylenes as a function of temperature, in the region of the C-H fundamental and first overtone. Experimental T h e m e t l i d s of purification and preparation of the acetylenes and their physical constants have been described elsewhere." Met1iyl:rcetyle:ie (Matheson Compressed
Gases; 95yo miti. purity) was used without purification . Reagent grade carbon tetrachloride and 2,2,4-trimethylpentane from freshly opened bottles viere used as solvents. Infrared spectra in the C-H stretching region x e r e determined with a Perkin-Elmer Model 112 spectrophotometer employing a lithium fluoride prism. T h e temperature of the 1 mm. cell eniployed was maintained to within f0.5'. Slit widtiis were 0.08&0.10 mm. T h e same instrument, with a cesium bromide prism, was used for studies in t h e 400-700 c x i ~ . - region. ~ Spectra in the 5000-7000 c m - l region were obtained with a Cary Model 14M Spectrophotoineter equipped with thermostatted cell blocks. T h e temperature was constant t o i 0.1". T h e path length used x i s 10 cm. and the slit widths were 0.2-0.32 mm. for slit heights of 20 ~nin. T h e scanning speed was 5.k./sec. The values of n / n a used in the calculatioiis are optical density raticis for the side band compared t o the inairi band; The optical density o f the side band was taken t o be the rnasitnuni difference between the actual ahsorption curve and the approsimated l0n. frequency wing of the rnain b a n d ; the latter u'as obtained hl- reflection of the main band about its center. Ratios of n;n, for the spectra in the fundamental regicin arc 11igI 'r,jvalues.
Discussion weak band is present approxiinately 20 c t n - ' lower in frequency than the main acetylenic carbon-hydrogen stretching fundamental, iii each of
HOTBAMX
IN THE INFR.lREI1
X i1
SPECTRA OF ~CETYLENES
TABLE I POSITIOSS O F FUNDAMEXTALS A S D HOT BAXDS IS A4CETPLEXESIS Chf.-' Compound
uCH
WE+ vi-vi
2vc11+ 2vCH
ui-vi
Concn., XI at 25'
Temp. range, " C . (no. of meas.)
Avg. v i calcd.
3316 3298 6530 6429 cu. 0 . 1 19.9-55.1(5) 665 C H a C r CH 3314 3296 6525 6486 ,056 20.8-49. T ( 5 ) 650 n-C*HgC=CH 3314 3295 6525 6486 .052 19.8-68.3 (5) 655 n-CjHIIC=CH d d 6537 6497 ,041 19.8-57.4 ( 4 ) 650 n-C5H11C=CHu 3314 3295 6527 6488 ,040 19.5-60.0 ( 7 ) mn n-C6H13C=CH 3314 3295 6527 6488 .06 2 6 , s and 66.0' 663' n-CsHi,C=CH 2599 2585 5158 5126 ,124 20.2-60.8 ( 5 ) 535 Z-C~HLJCECD 5126 ca. . 1 5 2 8 . 5 and 66.5' 560" TZ.-C~,H~~C=CD 2599 5158 2585 6445 ,090 - 1.4-39.6 ( 5 ) 690 6482 3293 3238 Me3SiCZCH 6480 6344 ,055 20.1-67.6( 5 ) 695 3293 3278 Ph3SiCECH 6480 2 8 . 0 and 70.0' 675" 3293 3278 6444 .U6 Ph3SiC~CH 6473 ca. .03 20.0-59.0 (5) 605 EtSCrCH 6512 3308 3289 b From measurements in the first overtone region except where noted. a In 2,2,4-trimethylpentane. t h e fundamental region. S o t measurable because of interferences.
6CH obsd.
630 627 (627) (627) (627) 630 492 492 672 or 651 d
d
669 Measurements in
the monosubstituted acetylenes which were studied may be taken to be equal to the ratio of the ab(Table I ) . The splitting is about 40 cm.-' in the sorbancies of the fundamental and the associated first overtone region. Nyquist and Potts consider hot band if the probabilities of the two transitions the possibility t h a t the side band represents a are assumed to be nearly equal. By choosing summation of the first overtone of the acetylenic proper degeneracy values, the energy level E or the C-H deformation mode (1230 cm.-l) with the frequency vi of the corresponding normal mode of C=C stretching mode (2120 cm.-l); the observed the molecule may be determined. intensity would then be the result of Fermi resonance with the carbon-hydrogen stretching mode.Z We have previously noted t h a t the C-D stretching absorption in deuterated acetylenes has a similar split at tern.^ This observation makes Fermi resonance seem an unlikely explanation for the side band, since an entirely different summation would have t o be involved in the deuterated compounds. T h e side band cannot be attributed to rotationvibration interaction because molecular rotation should be effectively prevented in solution. Intermolecular hydrogen bonding cannot be responsible for the side band, because the band persists a t low concentrations where self-association of acetylenes b y hydrogen bonding is known t o be ~iegligible.~ Furthermore, the intensity of the side band incyeases with increasing temperature (Fig. l ) , whereas the opposite would be true of hydrogenbonded C-H bands. T h e effect of temperature on the intensity of the side band leads to what we believe to be the correct explanation for this absorption. Similar teniperature-dependent side bands, called "hot bands, '' 15,300 15P2 15,500 have often been observed in the infrared spectra of WAVELENGTH A. gases.5 The hot band represents the same specFig. 1.-The first overtone of t h e -C-H stretching tral transition as the fundamental with which i t is absorption of 1-heptyne, 0.040 JI in CCI,; dashed line, associated except t h a t the hot band transition 19.5'; solid line, 69.0'. occurs only for molecules which are in a low-lying energy level above the ground state. T h e distriThe hot bands observed in the spectra of acetbution of molecules between the ground state and ylenes in solution will be discussed with the aid this particular excited state is given b y the Boltz- of the partial energy level diagram in Fig. 2 , for mann equation, where n is the relative number of molecules of Cavsymmetry. T h e ground state is molecules in an energy state E of degeneracy g non-degenerate. The energy level in which the above the ground state, a t temperature T excited molecules reside will correspond to a low energy deformation band (vi). For acetylenes ? = g e - E / k T = g e-hG,c/kT of Cav symmetry, the E-C=C RsE-C=C-H, no go Eo and C e C H deformation modes, which are both no and go refer to the ground state. The ratio of the numbers of molecules in the two states, %/no, doubly degenerate, are of a suitable energy to give rise t o hot bands. Absorption of radiation corre( 5 ) G . Herzberg, "Infrared a n d R a m a n Spectra of Polyatomic sponding to the acetylenic carbon-hydrogen stretch Molecules," D. Van Nostrand Co , I n c , Princeton, N J , 1945, pp. 266-269, 389-392. (Yk) gives rise to molecules in the levels V k and V k
CHARLES S.KR.ZIII.INZEL AND ROBERT~TE'EST
3672
EA
2Yk +y
AL
O
+ +
V i ; the transitions are V k and vk Vi - Vi. The level vk vi must also be doubly degenerate. Vi - vi Similarly, the transitions ? V k and L)vk may occurr and will give rise to hot bands associated with the first overtone of V k . The identical degeneracies for the two low frequency bending modes of monosubstituted acetylenes of symmetry C;jx.permit a clear distinction between the two possibilities for v,. In Table I are given some observed values for vi obtained from near infrared spectra for methylacetylene, trimethylsilylacetylene and triphenylsilylacetylene. The agreement of calculated values over the temperature range studied for each acetylene is excellent. These results indicate t h a t the hot band is arising from molecules in the energy level vi corresponding to the acetylenic carbon-hydrogen deformation band, which occurs near 650 cm.-l. T h e alternate (E-C=C) deformation mode occurs near 350 cm.-l in 1-propyne and a t even lower frequencies in silylacetylenes .6 *Although I-hexyne, 1-heptyne and 1-octyne are of lower symmetry, the near degeneracy of the carbon-hydrogen bending mode is attested to by the fact t h a t only one band is observed in the 630 cm.-' region. Calculations on these 1-alkynes employing a degeneracy ratio of two to one also lead to a frequency of about 650 cm.-] for v i , supporting the argument given above. Still further support is obtained from studies with 1-octyne-dl. The temperature variation in this deuterated acetylene leads to a frequency for vi of about 550 cni.-'. The C-D bending mode occurs near 300 cm.-1.7 Similar studies of temperature dependence were carried out for a few of the compounds in the fundamental region (Table I ) . dlthough the results are less accurate, they lead to approximately the same frequencies for vi. All of the evidence is thus consistent with the assignment of the side bands as hot bands resulting from molecules in the first excited state with respect
+
(6) H . Biichert and \V, Zeil, Z . p h y s i k . Chein. ( F r a n k f u r t ) , 29, 3 1 7 ( 1546 1) .
(7) R. J Grisenthwaite and H . I\'. Thompson, Tiairs & . , 5 0 , 212 (19543.
to BCH, the acetylenic C-H deformation mode. If the fundamental and first overtone are designated as VCH and 2 V C H , their associated hot bands should be designated as VCH ti^^ - tie^ and L ) V C H ti^^ - ~ C H , respectively. T h e results, both in the fundamental and first overtone region, lead to a value of about -20 cm.-' for the anharmonicity constant x relating the two modes. Such constants are normally much smaller for large molecules. The unusually large anharmonicity indicates t h a t VCH and t i interact ~ ~ strongly and probably are largely decoupled from the rest of the vibrations in the molecule. A series of hot bands corresponding to these have been observed in the spectra of gaseous acetylene. The relevant anharmonicity constant in acetylene, x14, is - 16.46 cm.-1.3 Quite different hot bands associated with V C H have also been observed previously in the gas spectra of methylacetylene and its deuterated derivatives.'-'O These hot bands result from excitation with respect to the methyl deformation; the anharmonicity is much smaller (1-2 cm.-l). Similar excitation undoubtedly takes place in solution, b u t the resulting hot bands are not resolved from the broad C-H fundamental. The previous workers did not notice the widely-split hot band due to the 630 em.-' energy level in the spectrum of gaseous methylacetylene, perhaps because its absorption is thinly spread over the complicated rotation-vibration fine structure. So far as the authors are aware, the bands described above are the first reported examples of hot bands observed in the infrared spectra of liquids or solutions. However, i t does not seem that the large anharmonicity observed in substituted acetylenes should be unique, and with increasing use of infrared spectrometers with moderately high resolving power i t seems likely t h a t other examples of hot bands observable in solution will be discovered." n'hen possible i t may prove quite convenient to study such hot bands in solution because interference due to rotational levels will thereby be eliminated.
+
2yk
Fig. 9.--Siiriplified energy level diagram for acetylenes nf Cav symmetry.
+
Yol. S f
Fainday
+
.-IDDED IN PROOF.--.% recent communication reports independent observation of the v(x1 8CH - 8ca not band in 1heptyne [Phan Tan Huongand Jean Lascombe, Comp. Rend , 254, 2543 (1962)].
+
Acknowledgments.-Acknowledgment is made to the donors of the Petroleum Research Fund, administered by the American Chemical Society, for partial support of this research. The authors are indebted to Dr. Monroe Evans for assistance with both theoretical and experimental aspects of the work and to D r . Xrthur Maki for helpful discussions. C. S.K . wishes to thank the Eastman Kodak Co. for a fellowship for 1960-1961. ( 8 ) H . C . Allen, J r , E. D. Tidwell and E. K . Plyler, J . R e s e n r r h S a i l BIW.S t a n d a r d s , 57, 213 (1956) ( 9 ) G. Herzherg, F. P a t a t and H . Verleger, J . P h y s Chrm., 41, 123 t1937). (10) D. R . J. Boyd and H. W'. Thompson, T i a n s . F n r a d a y .Poi., 48, 493 (1952); ibid., 49, 141 (1953); LZ. I. Christensen and H . IV Thompson, ibid., 52, 1439 (1956). (11) An example has already come t o t h e authors' attention through private communicatiun with Prof. T. I*, B r o w n , who reports ohservation of probable hot hands in t h e infrared spectrum of p-benzoquinone ( T . L. Brown, S ~ r c l i o c h i ~ nAcla . 18, in press.)
