3034
H. C. ALLEN,JR., E, D. TIDWELL AND E. K. PLYLER [CONTRIBUTION FROM
THE
Vol. 78
NATIONAL BUREAUO F STANDARDS]
The Infrared Spectrum of Acetylene-dll BY HARRY C. ALLEN,JR., EUGENE D. TIDWELL AND EARLEK. PLYLER RECEIVED FEBRUARY 20, 1956 The vibrational-rotational spectrum of acetylene-& has been observed and measured from 1800 t o 9200 cm? Rotational structure sufficient t o enable a n analysis was found for 2 8 bands. From the average combination-difference values of the 14 best bands the rotational constants were found t o be, BO = 0.99169 rt 0.00015 cm.-l and DO = 1.08 It 0.26 X em.-‘ The vibrational energy system is fully discussed. It is pointed out t h a t the vibrational energy levels cannot be fit by the usual quadratic expansion of the energy in terms of the vibrational quantum numbers. Evidence is presented t o show t h a t at least cubic terms must be included in t h e expansion.
Introduction Although the infrared spectra of acetylene and acetylene-& have been extensively ingestigated under high resolution,2-6 no high resolution work has been reported on the partially deuterated species in the non-photographic infrared. The vibrational energy systems of CzHz and C2D2 are still not well understood in spite of the extensive work. Since C2HD is less symmetric, there are no vibrational transitions which are forbidden by symmetry and hence there is a better chance to observe enough bands to enable an understanding of its energy system and perhaps furnish some clues which will help unravel the other two molecules. The spectrum of C2HD has been observed from 1800 to 9200 cm.-’. In this region 28 bands were observed for which rotational analyses could be made. In addition several Q branches were measured for which the P and R branch absorption was too weak to be observed. Experimental In the spectral region from 1 to 5.5 p , there have been observed 28 bands of CzHD for which the centers have been determined. These bands varied considerably in intensity __ 58’60
cn-’
--
5940
5900
0. v , + v , + v ; - v :
/
and the measurements were made on the gas in two cells of 1 m . and 10 m. in length and at pressures from a few millimeters to 76 cm. (Hg). Two instruments were used in obtaining the data. One instrument employs a grating of 15,000 lines/in. and a PbS detector. Other details of this spectrometer have been previously published.’ The second instrument contained a 7500 lines/in. grating and a cooled PbTe cell and it was used for the observations in the region from 3 to 5.5 p . A detailed description of this instrument has been reported.8 The rotational lines nere measured by comparison w ~ t h the fringe system from a Fabry-Perot interferometer. T h e absorption spectrum and the fringe system were recorded simultaneously on a two-pen recorder. T h e fringe system was calibrated by superimposing well known spectral lines on the chart at intervals throughout the band. This method of measurement is well suited for high precision measurements. When bands are not seriously overlapped by other bands (hot bands and isotopic bands) the band centers can be determined to a few hundredths of a wave number. T h e CzHD gas was made by Dr. Hellman of this Bureau. The gas was analyzed by a mass spectrometer and showed 16.8% CZHZ,47.87, CzHD and 35.4T0 CzD2. T h e spectra of the three components caused serious overlapping only in the region of 6600 em.-’. T h e results obtained in two regions of the spectrum are shown in Figs. 1 and 2 . TVhite recorder paper was used SO that photographs could be made of the actual tracing from the instrument. Figure 1 shows the bands u1 U S and U I -u8 U: - u: in the region of 5900 em.-’. Figure 2 is the observed trace for the region 2550 and shows the three bands u d , ug u: - vi, and va u: - u:.
+
+
+
~
+
+
Rotational Analysis.-The bands were analyzed by first assigning J values to the observed absorption peaks. An equation for each J value in a given band was then obtained of the form Y
=
vg
+ (B‘ + B ” ) m + (B’ - B”)m*+ 4Dma (1)
+
where m = J 1 for the R-branch and m = - J for the P-branch. The set of equations for each band was solved by the method of least squares c P R 4 to obtain values for yo, B’, B” and Fig. 1.-The U I + ug and the u1 + ug 4 u: - u: bands of C2HD in the D. The least squares reductions were region of 5900 em.-’. The rotational quantum numbers are given below carried out on SQAC, the NBS electhe trace for the PI + ~1 + V: band. A ten meter cell with a pressure of 16 tronic cornnuter. The observed ab.~ cm. (Hg) of CzHD was used for this region. sorption frequencies and rotational assignments for several of the stronger bands are ( 1 ) T h e work reported herein was supported by t h e U. S. Atomic given in Table I. Energy Commission. (2) E. E. Bell and H. H . Nielsen, J. Chem. Phys., 18, 1382 (1950). The value of Bo varied somewhat from band to (3) R.M.Talley and A. H. Nielsen, ibid., 22, 2030 (1954). band. This is believed to be due to the influence of (4) B. D.Saksena, ibid., 20, 95 (1952). overlapping hot bands. In order to get a best value (5) G.Herzberg, “Infrared and R a m a n Spectra,” D. Van h’ostrand ~~
Co., Inc., New York, N. Y., 1945, pp, 290-293. (6) H. C. Allen. L. R . Blaine and E. K. Plyler, J . Reseavch Null. Bur. b t a n d a r d s , in press.
(7) N. Gailar a n d E. K. Plyler, ibid., 4 6 , 393 (1952). ( 8 ) E. K. Plyler, L. R. Blaine and M. S. Connor, J . O p f SOC.A m . , 45, 102 (1955).
July 5 , 1956
INFRARED SPECTRUM OF ACETYLENE-&
3035
H. C. ALLEN,JR., E. D. TIDWELL AND E. K. PLYLER
3036
n c;
Vol. 78
July 5 , 1966
INFRARED SPECTRUM OF ACETYLENE-&
L
L
3037
H. C. ALLEN,JR., E. D. TIDWELL AND E. K. PLYLER
3038 2 5 2 0 cm-'
2560
30
35
25
15
20
I
I
I
I I I I
0
5
I
20
10
5
2 +
I
,
25
!
I
I
I
l I , l
I
~
l
l
20
l
/
I
!
I
l
l Il
5
10
2 +
5
O
15
I0
R-
I
I I I I I I I I I I I I I
'
I
2640
2600
+ P I
Vol. 78
,
20
25
30
I I I I I I I I I I I I I I 1/ I 1 I I I I 1I I I 1 I I
0
20
25
P Q R +
1 1 l l / 1 / 1 1 1 1 1 1 / 1 1 l / l I 10
0
P Q R +
+
+
Y : - Y : bands of C2HD in the region of 2580 cm.-'. The rotational quantum Fig. 2.-The pa, Y~ U: - Y: and numbers for v8 are marked below the spectrum. A one meter cell with a pressure df 1.5 cm. (Hg) was used for observing this band.
TABLE I1 INFRARED SPECTRUM OF GHD YO
+ B'J'* + B"1"Z 1200.63 1850.177 1852.46 1853.75 2065.24 2568.29 2580.81 2583.638 3086.61 3335.58 3995.777 4362.11 4385.76 4397.47 4415.882 4513.57 4643.244 4915.719 5123.114 5610.03 5795.014 5894.15 5913.312 6564.24 6569.36 6932.20 7620.34 8409.18 9139.02
vi
va
?:U
vdl
1'
1 1 1 1 1 1 1 1
VI
Vkl
VL
1' 1' 11
1'
1'
1'
1' 1' 1' 20
2 1 1
1' 1 1
1 1
1'
1'
1' 1' 2
1
1 2 2
1'
1 1
1'
1'
1
11
1' 1'
1
2 2
01
b;--B"'
(B
Lower VI
I'
1 1
1 1 1 2 2
0s'
1'
2 3
1 1
0.0050 .00432 ,0038 .00422 .00699 ,00645 .00677 .00670 .00614 .00480 .00508 .00128 ,0003 ,0108 .01091 .00092 .00198 .0094 1 .01338 .01390 .01372 ,0113 .01143 .0119 .01221 .01755 .02016 .01386 .01617
-
(B' B') calcd.
0.0042 0.00422 0.00670 0.00670 0.00670 0.00480
0.0109 0.01092
0.01340
0.0115 .01150 .0960 .0960 .01762 .02010 .01382 ,01630
Bo = 0.99169 f 0.00015 cm.-' Do = 1 . 0 8 f 0.26 X lod ~ m . - '
for Bo, use was made of the combination difference relation AF2' = R(J - 1) - P(J 4- 1) = 4Brr(J '/2) -I- 8D"(J '/2)' (2) The AF2 values were obtained by averaging the AF, values of the 14 best bands. These equations were then solved by the method of least squares. The values of the rotational constants obtained in this manner are, BO = 0.99169 f 0.00015 cm.-l and DO= 1.08 f 0.26 X cm.-'. The values of YO
and B' - B" obtained from the solutions of eq. 1 are given in Table 11. From the analyses of VI, v p and v g one obtains directly values of cyl, cy2 and cy3 in the expansion of the inertial constants in terms of the vibrational constants
+ ic ai( vi + w z ) =l 5
B" = Be
(3)
The values of the constants are a1 = 0.00480 cm.-l, a2 = 0.00422 and cy3 = 0.00670. These values can be checked by calculating B, for the states in which
July 5 , 1956
3039
INFRARED SPECTRUM O F ACETYLENE-dl
only VI,vz and Va are greater than zero. This has series nvl except for n = 2 which i t predicts about been done and the results are included in the last 1 cm.-l higher than i t is observed. Further, if column of Table 11. It is readqy seen that the cubic uoss terms were not important, then the agreement is excellent in all cases except 2vl a t A2v values for the two series should be equal, a The reason for this is not immedi- condition which, again, is not fulfilled. 6569.36 cm.-'. ately apparent. As we shall see later this band is TABLE IV also a misfit in the vibrational analysis. The Be value cannot be determined until adequate values of TABLE SHOWINGBANDSERIESREGULARITIES a4 and a5 are obtained. These values cannot be 1),vmv,'ud u, crn.-l AP Aav A2V obtained from the present work since no Q-branches 0 00000 were resolved and no 1-type doubling was observed. 3335.58 However Bo values have been determined in this lOOd0 3335.58 101.80 Laboratory for C2Hz and CZDZas well as C2HD. 5.06 3233.78 The values were determined from the average AFz 6569,36 20000 96.74 values of many bands and a least squares reduction 0.10 3137.04 of the data. The results are tabulated in Table I11 9706.4 30000 96.64 along with the bond distances which one can calcu3040.40 late from these inertial constants. 12746.8 40000 TABLE I11 MOLECULAR CONSTANTS OF ACETYLENE^ Bo, cm.-1 CzHz 1.17627 f 0 . 0 0 0 1 6 CzHD 0.99169 zkO.00015 CzDz 0.84788 f 0 . 0 0 0 0 9 Bo's used YCH, Arc-c, A. C2Hz and CzHD 1.054 1.210 CZHZand CZDZ 1.056 1.209 CZHDand CZDZ 1.055 1.210 Av. 1,055 1.210 The atomic constants used are taken from the recent work of J. A. Bearden and John S. Thomsen "A Survey of Atomic Constants," a report prepared for Bureau of Ordnance, U. S. Navy, 1955.
Vibrational Analysis.-The vibrational energy levels of a molecule are usually expanded in terms of the vibrational quantum numbers in an equation of the form E,
~ i Vi
- EO = i
XikVCvk i , k 2;
-
c YifkV,ViVk + . . .
(4)
ijk
I n most spectra studied so far, it has been sufficient to neglect terms above those quadratic in the vibrational quantum numbers. For CzHD this approximation seems to be inadequate. Consider v1 and its overtones as is done in Table IV. The first, second and third differences are tabulated. If this series could be represented adequately by a quadratic expression , the second differences would be equal. If this series could be adequately represented by a cubic expression, the third differences would be equal. Clearly neither of these conditions is fulfilled. However if 2vz were about 1 cm.-l higher, then the third differences could be equal and a cubic expression would suffice. The CzHD band at 6569 is badly overlapped by both CzHz and C2D2 as well as hot bands of CzHD. It is conceivable that the analysis here is in error, which would help straighten things out. One can also consider the vs as is done in the second group of series n v l Table IV. Again we clearly see that a quadratic term is not adequate. However if one uses the Azv to calculate ~111,a set of constants uy,Xll and En can be obtained which faithfully calculates the
+
00100
2583.64
10100
5913.31
3329.67 103.96 2.23
3225.71 20100
9139.02
101.73 3123.98
30100 00000
12263.0
0 2583.64
00100
2583.64
44.21 2.07
2539.43 00200
5123.07
00300 01000
7620.34 1853.75
01100
4415.88
42.14 2497.29
2562.13 45.81 2.08
2516.32 01200
6932.20
01300
9404.8
43.72 2472.60
As further evidence the data and differences for the two series nva and v2 nv3 are also given in Table 11. These series cannot be followed to sufficiently high values of v3 to determine whether or not the third differences are constant. Certainly the third differences are the same for each series, hence an unambiguous value of Y333 can be determined. Again the respective values of Azv in the two series are not equal indicating a sizable contribution from such terms as YIIZ and Y1zZ. This evidence points quite conclusively to the fact that a t least terms cubic in the vibrational quantum numbers are necessary to represent the vibrational energy levels. Several constants have been determined and are given in Table V.
+
TABLE V VIBRATIONAL CONSTANTS OF CaHD wl0 = 3386.92 cm.-l XI1 = 51.77 YIII = -0.43 08' 2606.45 X33 = 23.16 Y333 = - 0 . 3 5 x z 3 Yzza = 20.71 Yz,, = 0 . 8 0 XI3 Y133 = 4 . 3 5 Y113 = 1.56
+ +
Constants from levels in which all
X2, = 21.51 an.-' X24 = 3 . 1 6 vl( =
518.26
a n - 1
vi
1, the resulting constants are probably combinations of Xij’s and Yijk’s. Constants which can be evaluated in this way are given in Table IV where the designa-
[COXTRIBUTION FROM
THE
Vol. 78
tion used is that appropriate to a quadratic equation. A final piece of information which can be deduced is the band center of vi. This can be determined three different ways: v3 and v3 - v:, vz v3 v: and vz 4- v3 vi - vi, and yg vi and v3 v i - vq. The values obtained for v,‘ are 518.40, 518.25 and 518.32 cm.-‘. Since there is no reason to prefer any one of these values, the average 518.31 cm.-’ should represent the band center satisfactorily. The poor agreement can be explained by the fact that these values were deduced from hot bands which were badly overlapped by a strong CzHD band. Hence the band centers may be in error as much as 0.1 cm.-‘. Acknowledgments.-The authors wish to thank Dr. M. Hellman for preparing the sample and Mr. Joseph M. Cameron for performing the least squares reduction on SEAC.
+
+-
+ + +
WASHINGTOS25, D. C.
FRICK CHEMICALLABORATORY, PRIXCETOX VNIVERSITY,AND EASTMAN KODAKCOMPANY ROCHESTER, N. Y.]
THE
RESEARCH LABORATORIES,
Dipole Moments and Structures of Two Heteropolar Cyanine Dye Molecules’~2 BY
ANTHONY J. PETRO, CHARLES P. SNYTH
AND
LESLIEG. s. BROOKER
RECEIVEDDECEMBER 29, 1955 The molecular dipole moments of two cyanine dyes have been measured in very dilute solutions in dioxane. Comparison of the observed moment values 9.8 and 13.3 X lo-’’ with those calculated for the possible resonating structures indicates t h a t a full, or nearly full, electronic charge is carried by the acidic radical of the molecule, while a n equal positive charge is equally divided between the nitrogens of the two basic nuclei.
Dipole moment measurements carried out in the Carbon Chemicals Corporation was refluxed overnight with sodium and then fractionally distilled as needed. The Frick Laboratory have been used, 3*4 in conjunction fraction used boiled a t 101.2-101.3° at 758.0 mm. and had a with absorption data in the visible region, to help dielectric constant of 2.2005 =k 0.0002 a t 30”. Dioxane was clarify the resonance conditions in various non- chosen as solvent because of the greater solubility of the ionized dyes. 4 long series of papers by Brooker dyes in this medium than in benzene. of the very slow rate of solution of the dyes, heat and colleagues dealing with these molecules is listed wasBecause used to dissolve the solids. In the more concentrated in reference 4. Two new dyes are of particular solutions, which were evidently supersaturated, the solids interest because of their possibly large polarities. recrystallized after two days, so t h a t measurements were The two compounds are 1,3-bis-(2-methoxyethyl)- carried out as soon as possible after the solutions were pre5- [bis-1,3-(3-ethyl-2 -benzothiazolinylidene)-2-pro- pared. The dielectric constants of the solutions were measured pylidenel-barbituric acid and 1,3-diethyl-5-[bis- with a heterodyne-beat apparatus as described e l s e ~ h e r e . ~ * ~ 1,3- ( 3-ethyl - 2 - benzothiazolinylidene) - 2 - propyli- The dielectric cell was similar t o t h a t previously described’ denel-a-thiobarbituric acid, both belonging to the with one modification. Filaments of lead glass (Corning No. 8871) instead of mica strips were used as spacing group of dyes known as the holopolar cyanines.6 material between the concentric condenser plates, the glass They will hereafter be referred to as I and 11,respec- being fused t o the platinum in a high temperature oven.8 tively. This construction gave added stability t o the cell and thus an invariable cell constant. Experimental The dipole moments were measured in dioxane solutions a t 30’. “Practical” 1,4-dioxane from the Carbide and (1) This research was supported by t h e United States Air Force through t h e Office of Scientific Research of t h e Air Research a n d Development Command. Reproduction, translation. publication, use or disposal in whole or in part b y or for t h e United States Government is permitted. (2) This paper represents a part of t h e work t o be submitted b y Rlr A. J. Petro t o t h e Graduate School of Princeton University in partial fulfillment of the requirements for t h e degree of Doctor of Philosophy. (3) L. XI. Kushner and C. P. Smyth, THIS J O U R N A L , 71, 1401 (1949). (4) C. P . Smyth, “Dielectric Behavior a n d Structure.” McGrawHill Book Co., New York, N. Y., 1955, pp. 347-354. ( 5 ) L. G. S. Brooker and F. L. White, U. S. Patent 2,730,004 (1056)
Because of the low solubilities of the dyes (range of mole fractions used: I, (0.5-1.5) X 11, (1.2-4.4) X it was considered unprofitable t o measure the densities of the more dilute solutions. Consequently, the detisity of only the most concentrated solution was measured in each case, with an Ostwald-Sprengel pycnometer, and a linear dependence of specific volume v 1 on ~ the mole fraction cz was assumed. The values of the slopes p‘ in the equation VlZ
=
v1
+ P’cz
were then calculated from the specific volumes of the pure solvent and the most concentrated solution. The effect of (6) G. L. Lewis and C. P. Smyth, J. Chcm. P h y s . , 7 , 1085 (1939) (7) C. P. S m y t h and S . 0. Morgan, THISJOURNAL, 50, 1547 (1928). (8) T h e cell was constructed b y Mr B. B . Howard of t h e Frick Laboratory.
