- Email: [email protected]
Centrifugal distortion and internal rotation analysis of the rotational spectra of N-methylmethanimine d0 and d5
JOURNAL
OF MOLECULAR
SPECTROSCOPY
107,250-260 (1984)
Centrifugal Distortion and Internal Rotation Analysis of the Rotational Spectra of N-Methylmethanimine do and d5 J.
DEMAISON
AND J. BURIE
Luboratoire de SpectroscopicHertzienne. L.A. 249, Universitf de Lille I, F 59655-Villeneuve d*AscqCedex, France
J. M.
DENIS’
Laboratoire de Chimie Organique Physique, E.R.A. 827, Universitkde Lille I, F 59655-Villeneuve djlscq Cedex, France AND
B. P.
VAN
EIJCK
Department of Structural Chemistry, Universityof Utrecht,Padualaan 8, 3584 CH Utrecht. The Netherlands The groundstatemillimeter-wave spectraof CH3N=CH2 and CD3N = CD2 havebeenmeasured.
The rotational constants, centrifugal distortion constants, and barrierhindering internal rotation of the methyl group have been determined for both species. For the parent species I, and <(i, a) werealso obtained, and for the perdeuteriated species the quadrupoie coupling constants of “N were determined. INTRODUCTION
The imines are compounds of growing usefulness and interest to chemists and biochemists, but their physical properties have been little studied because they are very reactive. The microwave spectrum of N-methylmethanimine (CHsN=CH2) was first studied by Sastry and Curl (1). They prepared CH3N=CH2 by dripping 1,3,5trimethylhexahydrosymtriazine onto a bed of fused alumina and SiOz maintained at 45O’C. They have determined the rotational constants, the dipole moment, the quadrupole coupling constants of nitrogen, and the barrier to internal rotation of the methyl group. Later, by measuring the microwave spectra of the isotopic species CH2DN=CHz, Curl et al. (2) showed that one methyl proton eclipses the C=N double bond. Recently Bak and Svanholt (3), by measuring some new lines of CHjN=CHz and the microwave spectrum of CD3N=CDz, have determined approximate centrifugal distortion constants for these two species, which have been produced by pyrolysis of dimethylamine d,, or d7 on a quartz surface at 950°C. I Presentaddress: Groupe de Physico-Chimie Structurale, E.R.A. 389, Campus de Beaulieu, 35042Rennes Cedex. France
0022-2852184 $3.00 Copyrisht 0 1984by Academic
250 F’rers, Inc.
All rights of reproduction in any form reserved.
ROTATIONAL
SPECTRUM
251
OF N-METHYLMETHANIMINE
However, these studies have been limited up to 50 GHz. As CH3N=CH2 is relatively light (A - 52.5 GHz), the greatest part of the strong lines are above 50 GHz. Therefore, to predict the frequencies of the strong lines which are of great potential interest to radioastronomers, it is necessary to measure first the millimeter-wave spectrum. This was the aim of the present work, for which a very simple and efficient procedure has been used to prepare the N-methylmethanimine. This procedure is quite general and could be used for other molecules of the same type. EXPERIMENTAL
DETAILS
Reactive molecules have been recently synthesized by vacuum dynamic gas/solid phase reaction (4, 5). CH3N=CH2 and CD3N=CD2 have been prepared here by “one-tube” multistep sequences carried out in a single vacuum line which was pumped continuously through the absorption cell (Fig. 1). N-Chlorination of the dimethylamine ((CH3)*NH or (CD3)2ND) on solid N-chlorosuccinimide (NCS) at room temperature is followed by an a-elimination reaction of the resulting iV-chloramine over solid potassium tert-butoxide (KO’Bu) at 90°C. This process can be schematically represented as
(CH3NH s
(CH3)2NCl T
CH,N=CH;!.
The only gaseous by-product, t-BuOH, was eliminated by a cold trap at -82°C. A cell pressure of about 5 mTorr was maintained by cooling the dimethylamine at -55°C. At the outset of this investigation the purity of the compound was checked by its low-temperature ir, and ‘H and 13C NMR spectra (5). The lifetime of this species in the closed cell was more than 1 min. The measurements were performed with the Lille millimeter-wave spectrometer. Phase-stabilized klystrons (Varian, 68-80 GHz) supply a harmonic generator (Custom Microwave) with fundamental power. To obtain a good sensitivity, superheterodyne detection was used; the local oscillator was kept at a constant frequency difference of 600 MHz from the source oscillator. After the detection, the signal was digitally averaged and then processed by a microcomputer (ITT 2020), which calculated the line frequencies and which allowed-if necessary-digital filtering (line smoothing, baseline subtraction). For details see (6).
i, __ -_ _ _ --__ _ _ -@ a
-b
L ‘-=-7y
‘
_-
Cell
-ntlH -. _~ = -_ ---_ *
II &Il=Wz;__I,, -- -
FIG. 1. Preparation of N-methylmethanimine: a, dimethylamine; b, methanol bath at -55”C, c, solid N-chlorosuccinimide; d, solid potassium tert-butoxide; e, furnace at +9O”C, f, trap at -82°C.
10
9
I6
I8
1
3
3
2
3
3
2
2
I
2
14
I4
15
I6
16
17
17
18
I9
19
10
17
16
I5
I4
15
13
12
I3
11
3
12
3
0
12
9
8
12
3
II
I3
3
IO
I
2
2
I
3
16
16
17
I7
17
1
2
I5
0
2
I4
19
2
I3
I9
2
2
I3
I
II
12
2
2
II
I8
I9
I5
17
I6
I5
16
I4
13
I2
12
II
II
10
9
18.52
41.03
-42.71
54.91
44.66
46.80
54.26
48.91
51.00
144
220
212
229
238
233
218
229
225
211
219 222
55.27 53.10 -30.55
219
217
216
213
215
214
213
209
211
-7.90
57.69
60.75
13.69
IO
65.31
0
928
9
937
I
836
10
73.11
827
~35
IO
85.93
725
726
734 36.
23.18
624
6~3
213
100.40
625
634
211 212
15.39 109.53
2
4
2
5
3
5
2
3
3
5
5
212 212
15.28 110.74
422
423
431
212
212
043.84
844.88
564.39
347.00
093.15
260.89
197.13
061.77
464.76
164.38
432.20
921.48
019.19
885.45
273.47
612.23
032.82
109.71
450.75
915.82
009.93
I3
-0.16
0.07
0.09
-0.07
-0.15
-0.12
-0.08
-0.04
-0.10
-0.12
-0.02
0.07
-0.17
0. I I
0.16
-0.
0.16
0.15
0.20
0.23
0.27
0.23
0.24
004.46
0.23
722.39
-0.10
0.24
0.27
0.24
vA(calc.-ohs.)
713.76
560.74
124.62
333.78
479.07
vA(obs.)
432
3
v*-VK(Calc.)
26.18
K;
100.74
K;
322
J”
321
1(; +
331
K;
330
J’
Transition Frequencies (MHz) in CH,N=CH2
TABLE I
144
220
212
229
238
233
218
229
225
211
222
219
219
217
216
213
214
214
213
209
210
212
212
211
212
212
212
212
825.31
803.57
607.18
291.75
048.43
214.04
142.57
012.85
413.75
-0.
I5
0.35
0.02
0.27
-0.08
-0.07
0.22
-0.03
-0.09
-0.18
0.02 194.99
0.10 379.06
0.83
0.17
0.16
0.02
0.24
0.23
0.24
0.25
0.31
0.29
0.28
0.32
0.43
0.31
0.28
0.30
vE(calc.-ohs.)
866.18
026.09
827.70
212.72
596.39
967.43
036.52
364.78
879.69
986.71
904.00
612.82
698.26
449.46
109.27
307.59
379.07
vK(obs.)
2
4
5
29
29
30
4
5
5
4
5
5
4
4
4
33
33
33
33
34
34
35
36
31
32
4
3
31
4
4
28
31
4
28
3
4
27
4
4
26
30
4
26
30
4
2
25
3
4
25
25
2
24
26
I
23
2
4
23
24
3
20
24
I
2
20
32
31
30
30
28
29
20
29
28
28
28
27
26
27
25
25
27
24
24
23
23
22
24
22
23
21
22
23
23
20
21
la
19
36
35
34
33
33
33
32
32
32
31
30
30
30
30
29
29
29
28
27
27
26
26
26
25
25
24
24
24
24
23
22
20
20
3
3
3
6
4
3
6
6
3
2
5
5
3
2
6
3
I
3
5
3
3
3
2
3
I
5
I
I
0
3
4
I
0
33
32
31
27
29
30
27
26
29
29
25
26
27
28
24
26
28
25
23
24
24
23
25
23
24
20
23
24
24
21
18
19
20
9.48
229 237
-13.42
222
231
303
217
230
210
214
239
236
315
214
223
I56
216
307
220
225
225
-13.06
-12.10
14.12
-15.46
-10.50
11.80
13.12
-8.31
-0.89
7.78
3.73
-2.30
-0.24
8.29
1.36
18.68
5.33
-5.04
318
230
13.68 II.74
311
313
233
144
215
319
306
306
I50
156
238
26.40
14.77
15.54
-10.69
15.10
62.37
57.20
20.77
-21.20
17.05
44.38
0.02
803.32
087.87
373.40
174.88
445.12
637.18
971.61
735.62
821.09
140.32
456.30
976.63
551.29
847.72
435.22
819.13
275.88
450.48
467.95
228.80
061.27
913.87
650.79
620.01
256.41
901.05
792.69
238.22
418.83
0.38
0.35
0.27
1.63
0.38
0.25
-I .30
-1.37
0. I2
-0.18
0.24
0.38
0.06
-0.35
-0.49
0.02
-0.04
-0.18
0.41
-0.16
-0.19
-0.59
-0.25
-0.09
-0.11
0.38
-0.09
-0.02
0.25
0.55 -0.18
595.57
-0.18
088.63
308.62
231.30
186.62
722.17
0.32
I7
163.83 385.49 100.92 816.69
231 222 220 237
0.24 460.65
303
0.42
0.36
0.28
-1.43
0.32
-0.91
959.42 647.70
-1.04
0.15
-0.38
0.52
0.51
0.10
-0.21
0.07
-0.16
0.70
-0.
-0.17
-0.58
-0.28
-0.12
0.67
-0.11
-0.27
0.69
-0.11
217
230
829.37
214
141.42
440.25
972.77
553.56
426.65
210
239
236
315
214
I56
817.72
445.13
220
216
472.70
219.33
049.51
900.18
605.43
240.88
911.46
777.61
574.89
109.69
291.50
225
225
318
230
313
233
I44
215
306
150
156
238
254
DEMAISON
ET AL.
CH3N=CH2
The rotational constants are rather large and p = & = 1.53 D (I), so the lines are rather strong. First, an approximate spectrum was predicted with the constants of Bak (3). Some lines showed a rather well-resolved quadrupole hyperfine structure due to 14N which greatly facilitated their assignment. Then higher J transitions were assigned using the “bootstrap” method as described by Kirchkoff (7). The predicted internal rotation splitting was also of great help for the assignment. A list of measured frequencies is given in Table I. As the quadrupole hypertine structure could be well reproduced by the constants of Sastry (Z), no new determination of the quadrupole coupling constants was attempted. The hypothetical unperturbed frequencies of the split lines were calculated using the intensity-weighted mean of the multiplets (8). To obtain molecular parameters from the observed frequencies, a least-squares program based on the Internal Axis Method was used (9). The elements off-diagonal in u were removed by a Van Vleck transformation. Then the eigenvalues of the Hamiltonian were directly calculated by matrix diagonalization. Basis functions expicu(3k + u) with k ranging from -9 to +9 and a Van Vleck transformation up to u = 2 were sufficient to obtain stable results within a few kilohertz. As in Ref. (9), quartic centrifugal distortion terms were defined in the Principal Axis System (representation I’), transformed to the Internal Axis System, and added to the Hamiltonian. All data of Refs. (1) and (3) and Table I were used. The determinable quantities derived from the fit to the spectrum are listed in Table II, together with their standard deviations. All the parameters are well determined. The correlation of ICYand BKis the highest with 0.886; the standard deviation is- 0.38 MHz for 87 A lines and 0.43 MHz for 82 E lines; the mean splitting being Au = 33.76 MHz, and the greatest splitting 111.28 MHz for the 432 - 4~~ line. CD3N=CD2
The spectrum of CD3N=CD2 was identified using the same method as for CH3N=CH2. A list of measured frequencies is given in Tables III and V. The hyperfine structure due to i4Nquadrupole coupling to overall rotation could be at least partially resolved for many lines. The hyperfine structure was analyzed with first-order perturbation theory. The diagonal elements xgg of the quadrupole coupling tensor determined from the splittings of Table III are listed in Table IV, together with those of the parent species. The fitting parameters were x+ and x-. The correlation coefficient of these two parameters is 0.13 1. Since the molecule has C, symmetry the only nonzero off-diagonal tensor element is x&. By rotating the principal inertial axis SySteII’I with isotopic substitution and combining the diagonal elements xgg of the two isotopic species, the off-diagonal element xab may be, in principle, calculated. Unfortunately, in the present case the angle of rotation is practically negligible (-0.1”). It explains why the quadrupole constants are nearly identical for the two molecules. The value found for the constant xcc = 3.59 (18) MHz is in good agreement with that found for the similar molecule CHz=NH, xcc = 3.591 (8) MHz (10). Some lines were found split by internal rotation (Table V), but as these splittings are small and few, only V, could be determined. The moment of inertia of the methyl top was taken to be Z, = 6.30 u - AZ and the angle 0 (methyl top to a axis) = 29”.
ROTATIONAL
SPECTRUM
255
OF N-METHYLMETHANIMINE
TABLE II Molecular Parameters for N-Methylmethanimine in Representation I’
CD3N = CD2
CH3N = CH2
d
c
I(31)
34 577.460(18)
52
B/MHZ
10 670.2820(96)
8 165,012(23)
8 161.1468(28)
C/MHZ
9 371.4939(99)
7 185.594(34)
7 191.2333(27)
3.1976(23)
Ia/“.; 01” v3/cm
514.71
34 579.433(26)
A/MHZ
a -I
AJ/kHz
25.660(45) 713.71(32) 7.5584(43)
6.30 29
b
b
715(8)
A&kHz
-34.189(86)
-5.83(26)
+/kHz
639.18(100)
148.75(63)
GJ/kHz $/kHz
1.39557(83) 19.844(140)
4.1357(23)
4.1348(36)
0.691
16.772(28) 122.65(33) 0.68403(32)
lO(37)
-48.464(26)
5.55(61)
-3.22(55)
HKJ’Hz
8 denotes
the angle between the a-axis and assumed value simultaneous fit the A and E lines fit of the A lines only (neglecting internal
the
internal
rotation
axis.
rotation).
This latter quantity was obtained from an assumed structure (I). The rotational constants, centrifugal distortion constants, and the barrier hindering internal rotation of the CD3 group are determined using the same least-squares program as for CH3N=CH2 and the experimental frequencies of Table V. When the splitting was too small to be observed, the input data is VA* = (Ye + YE)/~ (instead of VAand Ye). The derived constants are listed in Table II. The standard deviation of the fit is 0.21 MHz for the A (or A*) lines and 0.23 MHz for the E lines. Five correlation coefficients are greater than 0.98: (VJ, C) = 0.983; (V,, AJ~) = 0.999; (C, AJ~) = 0.985 (C, 6,) = 0.986; and (A,,, AK) = 0.998. These high correlations show that the fit is not satisfactory. Although the deuteriation of the methyl group does not seem to significantly change the barrier, this good agreement is certainly accidental: for CD,N=CD2, I, and t9are assumed parameters, and a small change of either Z, or 0 changes all the other parameters by many times their standard deviation. The most sensible parameters are V3 and AJK. To obviate these difficulties we have simply fitted the A lines of CD3N=CD2 to the centrifugal distorted Hamiltonian of Watson using a type Z’representation. The derived constants
DEMAISON
256
ET AL.
TABLE III Nuclear Quadrupole Hypertine Structure of CD3N=CD2 b
a J’
K’ + J” c
K’ a
K” a
F’
+ F”
”
%
AvHFS
obs. (MHz)
I
I
0
I
I
0
c
0 2
I 2
27 27
383.90 385.95
obs.
A?iFS
calc.obs.
(MHz)
(MHz)
-2.05
-0.05
2
I
1
2
0
2c
3 2
3 2
28 28
382.40 383.60
1.20
0.003
3
I
2
3
0
3c
4 3
4 3
29 29
925.50 926.90
1.40
0.014
4
0
4
3
I
3c
5 4
4 3
37 37
040.90 042.20
1.30
0.07
4
I
3
4
0
4C
5 4
5 4
32 32
072.60 074.10
1.50
0.07
6 5
6 5
34 34
896.70 898.50
17 I8 16
16
17 IS I6
17
17 18 I6
17 IS I6
I8 19 17
I8 I9 I7
19 20 I8
I9 I8 20
20 21 I9
I9 20 I8
21 22 20
21 22
5
I
4
5
0
17
I
16
16
2
I7
I7
I8
I9
20
21
22
25
25
2
3
3
I
2
4
4
I
2
I6
I7
15
17
I6
I8
I0
19
I8
I9
I7
21
18
22
25
24
I
2
2
0
3
3
3
0
25
24
I
gc
I5
17
16
I7
I9
I7
It3
19
25
A-species
unless
b)
Splitting
with
cl
Taken
d)
E-species
from
the
>
I6 18
22
25 26 24
25
25 26 24
25 26 24
thesis
234
220.738
157
082.883
157
081.318
157
392.688
157
391.900
lb0
951.641
160
950.826
159
331.448
159
329.462
237
405.349
237
484.217
158
939.837
158
940.586
154
197.882
I54
198.582
228
208.183
228
206.913
1
-0.034
-1.565
-0.007
-0.788
0.07
-0.815
-0.018
-1.986
-0.028
d
-1.132
0.002
>
0.749
-0.007
>
1
24 26
1
232
077.380
232
076.228
stated
of
-0.09
t
21 23
to
1.80 -0.663
)
otherwise respect
221.401
1
20
22 23 21
25
a)
15 17
234
first H.
component
SVANHOLT
(15)
0.700
0.009
-I .270
-0.149
-1.152
-0.139
ROTATIONAL
SPECTRUM
257
OF N-METHYLMETHANIMINE
TABLE III-Continued
J’
26
28
30
30
34
35
K'a I
3
2
4
4
3
K’
f
J”
“;
“1
F’
c
v
p
c
25
26
28
27
30
32
26
0
28
2
I
30
30
3
34
3
35
2
26
27
29
28
31
33
obs. (MHZ)
26 27 25
26 27 25
239 123.221
28
2.3
232 651.499
27 29
27 29 }
232 650.359
30 31 29
30 31 29 1
238 382.522
30 31 29
30 31 29
f
231 783.230
34 35 33
34 35 33 I
I58 477.900
35
35
230 537.583
34 36
34 36 t
238 536.255
239 121.903
238 381.196
AvHFSobs.
talc.-obs. *"HrS (MHZ)
(MHz)
-1.318
-0.032
-1.140
0.098
-1.326
0.006
-0.728
0.045
-0.808
0.019
-1.328
0.048
231 783.958
I58 477.092
are shown in the third column of Table II. The standard deviation of the fit is then only 186 kHz and the highest correlation (A,, aK) = 0.964. CONCLUSION
The results of the present analysis are sufficient for the prediction of all strong transitions of CH3N=CH2 throughout the microwave and millimeter-wave ranges. TABLE IV Nitrogen Quadrupole Coupling Constants of N-Methylmethanimine
(MHz) a
CD3N = CD2
x+ = Xbb x
+
x,,
= Xbb - xc,
-I
CH3N = CH2
.28(37)
-8.460(79)
derived constants
Xaa
‘bb
X cc
a) from ref. (I)
1.28(37)
-4.87(19)
3.59(18)
l.9(3)
-5.1(2)
3.2(2)
DEMAISON ET AL.
258
TABLE V Transition Frequencies (MHz) in CD3N=CD2 J’
K;
“;
1
J”
“;
“,
“A”E(ealc.)
VA(&.)
vl\kalc.-obs.)
vE(Obs.)
vg(calc.-ohs.)
21,
202
0.34
28 382.64
0.20
28 382.24
312
303
0.34
29 925.90
0. I I
29 925.50
0.17
404
313
-0.29
37 041.20
0.10
37 041.60
-0.02
413
404
0.35
32 073.00
0.26
32 072.70
0.21
514
505
0.36
34 897.30
0.08
34 896.80
0.22
615
606
0.37
38 481.70
0.16
38 481.30
0.20
0.31
241 521.62
-0.02
241 307.93
-0.00
II
5
I, 12
0.26
6
,I
4
,
-0.42
241 520.87
5
7
1,
4
8
-0.09
24, 560.10
5
8
12
4
9
-0.11
241
385.19
12
5
7
I*
4
8
-0.31
241
307.36
15
5
10
15
4
II
-0.21
24"
182.41
16
I
16
15
0
15
0.04
237
756.53
16
0
16
15
1
15
0.01
232
782.44
16
5
,I
16
4
12
-0.20
239
569.34
II
I
16
16
2
15
-0.01
234 220.96
17
2
16
17
I
17
0.77
157 081.84
0.22
157 080.85
0.45
17
3
15
17
2
16
cl.22
156 392.50
0.16
156 392.17
0.27
17
5
12
17
4
13
-0.19
238 784.77
159 329.17
0.34
17
5
13
17
4
14
-0.19
239 937.37
18
3
16
18
2
17
0.16
160 95l.lO
18
5
14
18
4
15
-0.19
239 562.10
19
I
18
19
0
19
0.82
19
j
15
19
4
I6
-0.19
*o
2
18
19
3
17
20
5
16
20
4
17
-0.19
21
4
17
2,
3
18
-0.23
158 940.34
21
5
17
*I
4
I8
-0.19
238
0.1 I
159 330.12 239
185.12
23, 484.59 238 827.14
514.13
* * * * * * *
t f * i
* * * t *
0.26 0.21 0.26 -0.02 0.05 0.08 -0.01 -0.08
-0.01 -0.02 0.23 -0.07 0.21 -0.1, -0.12 -O.lZ 0.07 -0.20
22
4
18
22
3
19
-0.30
154 198.35
*
-0.02
22
5
18
ZL
4
19
-".1!l
238 276.78
* *
-0.16
*
-0.28
23
5
19
23
4
20
-0.21
238
150.85
24
5
20
24
4
21
-0.22
238
175.81
25
I
24
25
0
25
0.95
228 207.26
0.03
228 206.06
0.28
25
2
24
25
I
25
0.85
232 076.54
-0.01
232 075.47
0.22
25
5
21
25
4
22
-0.24
238
394.48
26
1
25
26
0
26
0.95
239
122.34
239
121.09
0.12
26
5
22
26
4
23
-0.15
238 851.84
27
5
23
27
4
24
-0.27
239 593.77
28
3
26
28
2
27
-0.25
232
28
5
24
28
4
25
-0.29
240
29
2
27
29
I
28
0.02
29
3
27
29
2
27
-0.19
29
4
26
29
3
27
-0.51
225 506.02
30
2
28
30
I
29
-0.03
238
30
4
27
30
3
28
-0.52
31
4
28
31
3
29
-0.53
238 555.20
32
5
27
71
6
26
-0.06
228
33
4
29
33
3
30
-0.37
150 887.70
34
4
30
34
3
31
-0.31
158 477.01
35
3
32
35
2
33
-0.13
238 536.70
37
6
31
37
5
32
-0.07
238
38
6
32
78
5
33
-0.m
231 433.81
40
4
36
40
3
3)
0.02
650.74 665.96
225 996.85
381.64
266.94
501.77
229 890.91
*
* * * * *
f
*
-0.23
-0.30 -0.17 -0.32 -0.29 -0.25 -0.28 -0.19
-0.I)
158 583.69
-0.06
225 506.74
-0.35
231
-0.36
-0.26
783.47
-0.31
235 555.82
-0.35
150 888.15
-0.1,
158 477.37
-0.04
0.01 -0.03 * * * *
0.01 -0.34 0.25 0.44 0.45
ROTATIONAL
SPECTRUM
OF lb’-METHYLMETHANIMINE
259
TABLE VI Predicted Frequencies for CH3NCH2
Z.3”‘) 5,975 6.312 6.518 1.991 1.495 3.012 6.604 6.606 2.001 2.412 8.872 9.899 7.804 10.844 3.778 6.736 6.535 11.670 5.703 4.732 12.350 2.548 3.836 3.019 2.271 6.425 I.570
a) Calculated b) Line strength
frequency for the
in MHz
and
statistical
uncertainty (one
standard
deviation),
A
=
A
species,
E = E species
A species
But, the differences between calculated and observed frequencies (Table I) are greater than the experimental accuracy (about 30 kHz), especially for the high-Z lines. This result seems quite general and probably shows that the rigid frame-rigid top model with one torsional degree of freedom is failing for the high-J transitions (II, 12). The experimental value of the potential barrier V, = 2040.65 (91) cal/mol is in rather good agreement with the value found previously by Bak, I’, = 2007 cal/mol (3), assuming +(i, a) = 27.52”. The moment of inertia, ZO,of the methyl group could be accurately determined from the internal rotation analysis, Z? = 3.1976 us A’. Z, may be determined quite independently using the inertial defects of the normal and deuteriated species (13): Z, = -[A(CD3N=CD2)
- A(CH3N=CH,)]
= 3.1 19u*A2. We note that Z, < Zz. This behavior is rather general, as may be seen from an inspection of Table V of Ref. (14).
DEMAISON
260
ET AL.
A list of predicted transitions is given in Table VI. The frequencies in this table include all transitions with sufficient intensity over the range 60- 130 GHz. The line strength of each transition is given only for the A species because it is not significantly different for the E species. Further predictions may be obtained upon request from the authors. ACKNOWLEDGMENT This investigationhas been supported in part by the Centre National de la Recherche Scientifique (LA 249, ATP no. 9.82.10). RECEIVED:
January
23, 1984 REFERENCES
1. K. V. L. N. SASTRYAND R. F. CURL, J. Chem. Phys. 41, 77-80 (1964). 2. J. T. YARDLEY,J. HINZE, AND R. F. CURL, J. Chem. Phys. 41, 2562-2563 (1964). 3. B. BAK AND H. SVANHOLT,Acta Chem. Stand. A 31, 755-758 (1977). 4. J. C. GUILLEMINAND J. M. DENIS,Angew. Chem. Suppl. 1982; 15 15-1524. J. C. GUILLEMINAND J. M. DENIS,Angew. Chem. Int. Ed. Engi. 21,690-690 (1982). 5. J. M. DENIS,unpublishedresults. 6. J. BURIE,D. BOUCHER,J. DEMAISON,AND A. DUBRULLE,J. Physique 43, 1319-1325 (1982). 7. W. H. KIRCHHOFF, J. Mol. Spectrosc. 41, 333-380 (1972). 8. H. D. RUDOLPH,Z. Naturjbrsch. A 23, 540-543 (1968). 9. B. P. VAN EIJCK,J. VAN OPHEUSDEN, M. M. M. VAN SCHAIK,AND E. VAN ZOEREN,J. Mol. Spectrosc. 86,465-479 (198 1). 10. R. D. BROWN,P. D. GODFREY,AND D. A. WINKLER,Aust. J. Chem. 35,667-672 (1982). II. J. DEMAISON,A. DUBRULLE,D. BOUCHER,J. BURIE,AND B. P. VAN ELXK, J. Mol. Spectrosc. 94, 211-214 (1982). I2. J. SHERIDAN,W. B~SSERT,AND A. BAUDER,J. Mol. Spectrosc. 80, 1-I 1 (1980). 13. D. R. HERSCHBACH AND V. W. LAURIE,J. Chem. Phys. 40, 3142-3153 (1964). 14. J. DEMAISON,D. S~HWOCH,B. T. TAN, AND H. D. RUDOLPH,J. Mol. Spectrosc. 83,391~400 15. B. BAK, privatecommunication.
(1980).
OF MOLECULAR
SPECTROSCOPY
107,250-260 (1984)
Centrifugal Distortion and Internal Rotation Analysis of the Rotational Spectra of N-Methylmethanimine do and d5 J.
DEMAISON
AND J. BURIE
Luboratoire de SpectroscopicHertzienne. L.A. 249, Universitf de Lille I, F 59655-Villeneuve d*AscqCedex, France
J. M.
DENIS’
Laboratoire de Chimie Organique Physique, E.R.A. 827, Universitkde Lille I, F 59655-Villeneuve djlscq Cedex, France AND
B. P.
VAN
EIJCK
Department of Structural Chemistry, Universityof Utrecht,Padualaan 8, 3584 CH Utrecht. The Netherlands The groundstatemillimeter-wave spectraof CH3N=CH2 and CD3N = CD2 havebeenmeasured.
The rotational constants, centrifugal distortion constants, and barrierhindering internal rotation of the methyl group have been determined for both species. For the parent species I, and <(i, a) werealso obtained, and for the perdeuteriated species the quadrupoie coupling constants of “N were determined. INTRODUCTION
The imines are compounds of growing usefulness and interest to chemists and biochemists, but their physical properties have been little studied because they are very reactive. The microwave spectrum of N-methylmethanimine (CHsN=CH2) was first studied by Sastry and Curl (1). They prepared CH3N=CH2 by dripping 1,3,5trimethylhexahydrosymtriazine onto a bed of fused alumina and SiOz maintained at 45O’C. They have determined the rotational constants, the dipole moment, the quadrupole coupling constants of nitrogen, and the barrier to internal rotation of the methyl group. Later, by measuring the microwave spectra of the isotopic species CH2DN=CHz, Curl et al. (2) showed that one methyl proton eclipses the C=N double bond. Recently Bak and Svanholt (3), by measuring some new lines of CHjN=CHz and the microwave spectrum of CD3N=CDz, have determined approximate centrifugal distortion constants for these two species, which have been produced by pyrolysis of dimethylamine d,, or d7 on a quartz surface at 950°C. I Presentaddress: Groupe de Physico-Chimie Structurale, E.R.A. 389, Campus de Beaulieu, 35042Rennes Cedex. France
0022-2852184 $3.00 Copyrisht 0 1984by Academic
250 F’rers, Inc.
All rights of reproduction in any form reserved.
ROTATIONAL
SPECTRUM
251
OF N-METHYLMETHANIMINE
However, these studies have been limited up to 50 GHz. As CH3N=CH2 is relatively light (A - 52.5 GHz), the greatest part of the strong lines are above 50 GHz. Therefore, to predict the frequencies of the strong lines which are of great potential interest to radioastronomers, it is necessary to measure first the millimeter-wave spectrum. This was the aim of the present work, for which a very simple and efficient procedure has been used to prepare the N-methylmethanimine. This procedure is quite general and could be used for other molecules of the same type. EXPERIMENTAL
DETAILS
Reactive molecules have been recently synthesized by vacuum dynamic gas/solid phase reaction (4, 5). CH3N=CH2 and CD3N=CD2 have been prepared here by “one-tube” multistep sequences carried out in a single vacuum line which was pumped continuously through the absorption cell (Fig. 1). N-Chlorination of the dimethylamine ((CH3)*NH or (CD3)2ND) on solid N-chlorosuccinimide (NCS) at room temperature is followed by an a-elimination reaction of the resulting iV-chloramine over solid potassium tert-butoxide (KO’Bu) at 90°C. This process can be schematically represented as
(CH3NH s
(CH3)2NCl T
CH,N=CH;!.
The only gaseous by-product, t-BuOH, was eliminated by a cold trap at -82°C. A cell pressure of about 5 mTorr was maintained by cooling the dimethylamine at -55°C. At the outset of this investigation the purity of the compound was checked by its low-temperature ir, and ‘H and 13C NMR spectra (5). The lifetime of this species in the closed cell was more than 1 min. The measurements were performed with the Lille millimeter-wave spectrometer. Phase-stabilized klystrons (Varian, 68-80 GHz) supply a harmonic generator (Custom Microwave) with fundamental power. To obtain a good sensitivity, superheterodyne detection was used; the local oscillator was kept at a constant frequency difference of 600 MHz from the source oscillator. After the detection, the signal was digitally averaged and then processed by a microcomputer (ITT 2020), which calculated the line frequencies and which allowed-if necessary-digital filtering (line smoothing, baseline subtraction). For details see (6).
i, __ -_ _ _ --__ _ _ -@ a
-b
L ‘-=-7y
‘
_-
Cell
-ntlH -. _~ = -_ ---_ *
II &Il=Wz;__I,, -- -
FIG. 1. Preparation of N-methylmethanimine: a, dimethylamine; b, methanol bath at -55”C, c, solid N-chlorosuccinimide; d, solid potassium tert-butoxide; e, furnace at +9O”C, f, trap at -82°C.
10
9
I6
I8
1
3
3
2
3
3
2
2
I
2
14
I4
15
I6
16
17
17
18
I9
19
10
17
16
I5
I4
15
13
12
I3
11
3
12
3
0
12
9
8
12
3
II
I3
3
IO
I
2
2
I
3
16
16
17
I7
17
1
2
I5
0
2
I4
19
2
I3
I9
2
2
I3
I
II
12
2
2
II
I8
I9
I5
17
I6
I5
16
I4
13
I2
12
II
II
10
9
18.52
41.03
-42.71
54.91
44.66
46.80
54.26
48.91
51.00
144
220
212
229
238
233
218
229
225
211
219 222
55.27 53.10 -30.55
219
217
216
213
215
214
213
209
211
-7.90
57.69
60.75
13.69
IO
65.31
0
928
9
937
I
836
10
73.11
827
~35
IO
85.93
725
726
734 36.
23.18
624
6~3
213
100.40
625
634
211 212
15.39 109.53
2
4
2
5
3
5
2
3
3
5
5
212 212
15.28 110.74
422
423
431
212
212
043.84
844.88
564.39
347.00
093.15
260.89
197.13
061.77
464.76
164.38
432.20
921.48
019.19
885.45
273.47
612.23
032.82
109.71
450.75
915.82
009.93
I3
-0.16
0.07
0.09
-0.07
-0.15
-0.12
-0.08
-0.04
-0.10
-0.12
-0.02
0.07
-0.17
0. I I
0.16
-0.
0.16
0.15
0.20
0.23
0.27
0.23
0.24
004.46
0.23
722.39
-0.10
0.24
0.27
0.24
vA(calc.-ohs.)
713.76
560.74
124.62
333.78
479.07
vA(obs.)
432
3
v*-VK(Calc.)
26.18
K;
100.74
K;
322
J”
321
1(; +
331
K;
330
J’
Transition Frequencies (MHz) in CH,N=CH2
TABLE I
144
220
212
229
238
233
218
229
225
211
222
219
219
217
216
213
214
214
213
209
210
212
212
211
212
212
212
212
825.31
803.57
607.18
291.75
048.43
214.04
142.57
012.85
413.75
-0.
I5
0.35
0.02
0.27
-0.08
-0.07
0.22
-0.03
-0.09
-0.18
0.02 194.99
0.10 379.06
0.83
0.17
0.16
0.02
0.24
0.23
0.24
0.25
0.31
0.29
0.28
0.32
0.43
0.31
0.28
0.30
vE(calc.-ohs.)
866.18
026.09
827.70
212.72
596.39
967.43
036.52
364.78
879.69
986.71
904.00
612.82
698.26
449.46
109.27
307.59
379.07
vK(obs.)
2
4
5
29
29
30
4
5
5
4
5
5
4
4
4
33
33
33
33
34
34
35
36
31
32
4
3
31
4
4
28
31
4
28
3
4
27
4
4
26
30
4
26
30
4
2
25
3
4
25
25
2
24
26
I
23
2
4
23
24
3
20
24
I
2
20
32
31
30
30
28
29
20
29
28
28
28
27
26
27
25
25
27
24
24
23
23
22
24
22
23
21
22
23
23
20
21
la
19
36
35
34
33
33
33
32
32
32
31
30
30
30
30
29
29
29
28
27
27
26
26
26
25
25
24
24
24
24
23
22
20
20
3
3
3
6
4
3
6
6
3
2
5
5
3
2
6
3
I
3
5
3
3
3
2
3
I
5
I
I
0
3
4
I
0
33
32
31
27
29
30
27
26
29
29
25
26
27
28
24
26
28
25
23
24
24
23
25
23
24
20
23
24
24
21
18
19
20
9.48
229 237
-13.42
222
231
303
217
230
210
214
239
236
315
214
223
I56
216
307
220
225
225
-13.06
-12.10
14.12
-15.46
-10.50
11.80
13.12
-8.31
-0.89
7.78
3.73
-2.30
-0.24
8.29
1.36
18.68
5.33
-5.04
318
230
13.68 II.74
311
313
233
144
215
319
306
306
I50
156
238
26.40
14.77
15.54
-10.69
15.10
62.37
57.20
20.77
-21.20
17.05
44.38
0.02
803.32
087.87
373.40
174.88
445.12
637.18
971.61
735.62
821.09
140.32
456.30
976.63
551.29
847.72
435.22
819.13
275.88
450.48
467.95
228.80
061.27
913.87
650.79
620.01
256.41
901.05
792.69
238.22
418.83
0.38
0.35
0.27
1.63
0.38
0.25
-I .30
-1.37
0. I2
-0.18
0.24
0.38
0.06
-0.35
-0.49
0.02
-0.04
-0.18
0.41
-0.16
-0.19
-0.59
-0.25
-0.09
-0.11
0.38
-0.09
-0.02
0.25
0.55 -0.18
595.57
-0.18
088.63
308.62
231.30
186.62
722.17
0.32
I7
163.83 385.49 100.92 816.69
231 222 220 237
0.24 460.65
303
0.42
0.36
0.28
-1.43
0.32
-0.91
959.42 647.70
-1.04
0.15
-0.38
0.52
0.51
0.10
-0.21
0.07
-0.16
0.70
-0.
-0.17
-0.58
-0.28
-0.12
0.67
-0.11
-0.27
0.69
-0.11
217
230
829.37
214
141.42
440.25
972.77
553.56
426.65
210
239
236
315
214
I56
817.72
445.13
220
216
472.70
219.33
049.51
900.18
605.43
240.88
911.46
777.61
574.89
109.69
291.50
225
225
318
230
313
233
I44
215
306
150
156
238
254
DEMAISON
ET AL.
CH3N=CH2
The rotational constants are rather large and p = & = 1.53 D (I), so the lines are rather strong. First, an approximate spectrum was predicted with the constants of Bak (3). Some lines showed a rather well-resolved quadrupole hyperfine structure due to 14N which greatly facilitated their assignment. Then higher J transitions were assigned using the “bootstrap” method as described by Kirchkoff (7). The predicted internal rotation splitting was also of great help for the assignment. A list of measured frequencies is given in Table I. As the quadrupole hypertine structure could be well reproduced by the constants of Sastry (Z), no new determination of the quadrupole coupling constants was attempted. The hypothetical unperturbed frequencies of the split lines were calculated using the intensity-weighted mean of the multiplets (8). To obtain molecular parameters from the observed frequencies, a least-squares program based on the Internal Axis Method was used (9). The elements off-diagonal in u were removed by a Van Vleck transformation. Then the eigenvalues of the Hamiltonian were directly calculated by matrix diagonalization. Basis functions expicu(3k + u) with k ranging from -9 to +9 and a Van Vleck transformation up to u = 2 were sufficient to obtain stable results within a few kilohertz. As in Ref. (9), quartic centrifugal distortion terms were defined in the Principal Axis System (representation I’), transformed to the Internal Axis System, and added to the Hamiltonian. All data of Refs. (1) and (3) and Table I were used. The determinable quantities derived from the fit to the spectrum are listed in Table II, together with their standard deviations. All the parameters are well determined. The correlation of ICYand BKis the highest with 0.886; the standard deviation is- 0.38 MHz for 87 A lines and 0.43 MHz for 82 E lines; the mean splitting being Au = 33.76 MHz, and the greatest splitting 111.28 MHz for the 432 - 4~~ line. CD3N=CD2
The spectrum of CD3N=CD2 was identified using the same method as for CH3N=CH2. A list of measured frequencies is given in Tables III and V. The hyperfine structure due to i4Nquadrupole coupling to overall rotation could be at least partially resolved for many lines. The hyperfine structure was analyzed with first-order perturbation theory. The diagonal elements xgg of the quadrupole coupling tensor determined from the splittings of Table III are listed in Table IV, together with those of the parent species. The fitting parameters were x+ and x-. The correlation coefficient of these two parameters is 0.13 1. Since the molecule has C, symmetry the only nonzero off-diagonal tensor element is x&. By rotating the principal inertial axis SySteII’I with isotopic substitution and combining the diagonal elements xgg of the two isotopic species, the off-diagonal element xab may be, in principle, calculated. Unfortunately, in the present case the angle of rotation is practically negligible (-0.1”). It explains why the quadrupole constants are nearly identical for the two molecules. The value found for the constant xcc = 3.59 (18) MHz is in good agreement with that found for the similar molecule CHz=NH, xcc = 3.591 (8) MHz (10). Some lines were found split by internal rotation (Table V), but as these splittings are small and few, only V, could be determined. The moment of inertia of the methyl top was taken to be Z, = 6.30 u - AZ and the angle 0 (methyl top to a axis) = 29”.
ROTATIONAL
SPECTRUM
255
OF N-METHYLMETHANIMINE
TABLE II Molecular Parameters for N-Methylmethanimine in Representation I’
CD3N = CD2
CH3N = CH2
d
c
I(31)
34 577.460(18)
52
B/MHZ
10 670.2820(96)
8 165,012(23)
8 161.1468(28)
C/MHZ
9 371.4939(99)
7 185.594(34)
7 191.2333(27)
3.1976(23)
Ia/“.; 01” v3/cm
514.71
34 579.433(26)
A/MHZ
a -I
AJ/kHz
25.660(45) 713.71(32) 7.5584(43)
6.30 29
b
b
715(8)
A&kHz
-34.189(86)
-5.83(26)
+/kHz
639.18(100)
148.75(63)
GJ/kHz $/kHz
1.39557(83) 19.844(140)
4.1357(23)
4.1348(36)
0.691
16.772(28) 122.65(33) 0.68403(32)
lO(37)
-48.464(26)
5.55(61)
-3.22(55)
HKJ’Hz
8 denotes
the angle between the a-axis and assumed value simultaneous fit the A and E lines fit of the A lines only (neglecting internal
the
internal
rotation
axis.
rotation).
This latter quantity was obtained from an assumed structure (I). The rotational constants, centrifugal distortion constants, and the barrier hindering internal rotation of the CD3 group are determined using the same least-squares program as for CH3N=CH2 and the experimental frequencies of Table V. When the splitting was too small to be observed, the input data is VA* = (Ye + YE)/~ (instead of VAand Ye). The derived constants are listed in Table II. The standard deviation of the fit is 0.21 MHz for the A (or A*) lines and 0.23 MHz for the E lines. Five correlation coefficients are greater than 0.98: (VJ, C) = 0.983; (V,, AJ~) = 0.999; (C, AJ~) = 0.985 (C, 6,) = 0.986; and (A,,, AK) = 0.998. These high correlations show that the fit is not satisfactory. Although the deuteriation of the methyl group does not seem to significantly change the barrier, this good agreement is certainly accidental: for CD,N=CD2, I, and t9are assumed parameters, and a small change of either Z, or 0 changes all the other parameters by many times their standard deviation. The most sensible parameters are V3 and AJK. To obviate these difficulties we have simply fitted the A lines of CD3N=CD2 to the centrifugal distorted Hamiltonian of Watson using a type Z’representation. The derived constants
DEMAISON
256
ET AL.
TABLE III Nuclear Quadrupole Hypertine Structure of CD3N=CD2 b
a J’
K’ + J” c
K’ a
K” a
F’
+ F”
”
%
AvHFS
obs. (MHz)
I
I
0
I
I
0
c
0 2
I 2
27 27
383.90 385.95
obs.
A?iFS
calc.obs.
(MHz)
(MHz)
-2.05
-0.05
2
I
1
2
0
2c
3 2
3 2
28 28
382.40 383.60
1.20
0.003
3
I
2
3
0
3c
4 3
4 3
29 29
925.50 926.90
1.40
0.014
4
0
4
3
I
3c
5 4
4 3
37 37
040.90 042.20
1.30
0.07
4
I
3
4
0
4C
5 4
5 4
32 32
072.60 074.10
1.50
0.07
6 5
6 5
34 34
896.70 898.50
17 I8 16
16
17 IS I6
17
17 18 I6
17 IS I6
I8 19 17
I8 I9 I7
19 20 I8
I9 I8 20
20 21 I9
I9 20 I8
21 22 20
21 22
5
I
4
5
0
17
I
16
16
2
I7
I7
I8
I9
20
21
22
25
25
2
3
3
I
2
4
4
I
2
I6
I7
15
17
I6
I8
I0
19
I8
I9
I7
21
18
22
25
24
I
2
2
0
3
3
3
0
25
24
I
gc
I5
17
16
I7
I9
I7
It3
19
25
A-species
unless
b)
Splitting
with
cl
Taken
d)
E-species
from
the
>
I6 18
22
25 26 24
25
25 26 24
25 26 24
thesis
234
220.738
157
082.883
157
081.318
157
392.688
157
391.900
lb0
951.641
160
950.826
159
331.448
159
329.462
237
405.349
237
484.217
158
939.837
158
940.586
154
197.882
I54
198.582
228
208.183
228
206.913
1
-0.034
-1.565
-0.007
-0.788
0.07
-0.815
-0.018
-1.986
-0.028
d
-1.132
0.002
>
0.749
-0.007
>
1
24 26
1
232
077.380
232
076.228
stated
of
-0.09
t
21 23
to
1.80 -0.663
)
otherwise respect
221.401
1
20
22 23 21
25
a)
15 17
234
first H.
component
SVANHOLT
(15)
0.700
0.009
-I .270
-0.149
-1.152
-0.139
ROTATIONAL
SPECTRUM
257
OF N-METHYLMETHANIMINE
TABLE III-Continued
J’
26
28
30
30
34
35
K'a I
3
2
4
4
3
K’
f
J”
“;
“1
F’
c
v
p
c
25
26
28
27
30
32
26
0
28
2
I
30
30
3
34
3
35
2
26
27
29
28
31
33
obs. (MHZ)
26 27 25
26 27 25
239 123.221
28
2.3
232 651.499
27 29
27 29 }
232 650.359
30 31 29
30 31 29 1
238 382.522
30 31 29
30 31 29
f
231 783.230
34 35 33
34 35 33 I
I58 477.900
35
35
230 537.583
34 36
34 36 t
238 536.255
239 121.903
238 381.196
AvHFSobs.
talc.-obs. *"HrS (MHZ)
(MHz)
-1.318
-0.032
-1.140
0.098
-1.326
0.006
-0.728
0.045
-0.808
0.019
-1.328
0.048
231 783.958
I58 477.092
are shown in the third column of Table II. The standard deviation of the fit is then only 186 kHz and the highest correlation (A,, aK) = 0.964. CONCLUSION
The results of the present analysis are sufficient for the prediction of all strong transitions of CH3N=CH2 throughout the microwave and millimeter-wave ranges. TABLE IV Nitrogen Quadrupole Coupling Constants of N-Methylmethanimine
(MHz) a
CD3N = CD2
x+ = Xbb x
+
x,,
= Xbb - xc,
-I
CH3N = CH2
.28(37)
-8.460(79)
derived constants
Xaa
‘bb
X cc
a) from ref. (I)
1.28(37)
-4.87(19)
3.59(18)
l.9(3)
-5.1(2)
3.2(2)
DEMAISON ET AL.
258
TABLE V Transition Frequencies (MHz) in CD3N=CD2 J’
K;
“;
1
J”
“;
“,
“A”E(ealc.)
VA(&.)
vl\kalc.-obs.)
vE(Obs.)
vg(calc.-ohs.)
21,
202
0.34
28 382.64
0.20
28 382.24
312
303
0.34
29 925.90
0. I I
29 925.50
0.17
404
313
-0.29
37 041.20
0.10
37 041.60
-0.02
413
404
0.35
32 073.00
0.26
32 072.70
0.21
514
505
0.36
34 897.30
0.08
34 896.80
0.22
615
606
0.37
38 481.70
0.16
38 481.30
0.20
0.31
241 521.62
-0.02
241 307.93
-0.00
II
5
I, 12
0.26
6
,I
4
,
-0.42
241 520.87
5
7
1,
4
8
-0.09
24, 560.10
5
8
12
4
9
-0.11
241
385.19
12
5
7
I*
4
8
-0.31
241
307.36
15
5
10
15
4
II
-0.21
24"
182.41
16
I
16
15
0
15
0.04
237
756.53
16
0
16
15
1
15
0.01
232
782.44
16
5
,I
16
4
12
-0.20
239
569.34
II
I
16
16
2
15
-0.01
234 220.96
17
2
16
17
I
17
0.77
157 081.84
0.22
157 080.85
0.45
17
3
15
17
2
16
cl.22
156 392.50
0.16
156 392.17
0.27
17
5
12
17
4
13
-0.19
238 784.77
159 329.17
0.34
17
5
13
17
4
14
-0.19
239 937.37
18
3
16
18
2
17
0.16
160 95l.lO
18
5
14
18
4
15
-0.19
239 562.10
19
I
18
19
0
19
0.82
19
j
15
19
4
I6
-0.19
*o
2
18
19
3
17
20
5
16
20
4
17
-0.19
21
4
17
2,
3
18
-0.23
158 940.34
21
5
17
*I
4
I8
-0.19
238
0.1 I
159 330.12 239
185.12
23, 484.59 238 827.14
514.13
* * * * * * *
t f * i
* * * t *
0.26 0.21 0.26 -0.02 0.05 0.08 -0.01 -0.08
-0.01 -0.02 0.23 -0.07 0.21 -0.1, -0.12 -O.lZ 0.07 -0.20
22
4
18
22
3
19
-0.30
154 198.35
*
-0.02
22
5
18
ZL
4
19
-".1!l
238 276.78
* *
-0.16
*
-0.28
23
5
19
23
4
20
-0.21
238
150.85
24
5
20
24
4
21
-0.22
238
175.81
25
I
24
25
0
25
0.95
228 207.26
0.03
228 206.06
0.28
25
2
24
25
I
25
0.85
232 076.54
-0.01
232 075.47
0.22
25
5
21
25
4
22
-0.24
238
394.48
26
1
25
26
0
26
0.95
239
122.34
239
121.09
0.12
26
5
22
26
4
23
-0.15
238 851.84
27
5
23
27
4
24
-0.27
239 593.77
28
3
26
28
2
27
-0.25
232
28
5
24
28
4
25
-0.29
240
29
2
27
29
I
28
0.02
29
3
27
29
2
27
-0.19
29
4
26
29
3
27
-0.51
225 506.02
30
2
28
30
I
29
-0.03
238
30
4
27
30
3
28
-0.52
31
4
28
31
3
29
-0.53
238 555.20
32
5
27
71
6
26
-0.06
228
33
4
29
33
3
30
-0.37
150 887.70
34
4
30
34
3
31
-0.31
158 477.01
35
3
32
35
2
33
-0.13
238 536.70
37
6
31
37
5
32
-0.07
238
38
6
32
78
5
33
-0.m
231 433.81
40
4
36
40
3
3)
0.02
650.74 665.96
225 996.85
381.64
266.94
501.77
229 890.91
*
* * * * *
f
*
-0.23
-0.30 -0.17 -0.32 -0.29 -0.25 -0.28 -0.19
-0.I)
158 583.69
-0.06
225 506.74
-0.35
231
-0.36
-0.26
783.47
-0.31
235 555.82
-0.35
150 888.15
-0.1,
158 477.37
-0.04
0.01 -0.03 * * * *
0.01 -0.34 0.25 0.44 0.45
ROTATIONAL
SPECTRUM
OF lb’-METHYLMETHANIMINE
259
TABLE VI Predicted Frequencies for CH3NCH2
Z.3”‘) 5,975 6.312 6.518 1.991 1.495 3.012 6.604 6.606 2.001 2.412 8.872 9.899 7.804 10.844 3.778 6.736 6.535 11.670 5.703 4.732 12.350 2.548 3.836 3.019 2.271 6.425 I.570
a) Calculated b) Line strength
frequency for the
in MHz
and
statistical
uncertainty (one
standard
deviation),
A
=
A
species,
E = E species
A species
But, the differences between calculated and observed frequencies (Table I) are greater than the experimental accuracy (about 30 kHz), especially for the high-Z lines. This result seems quite general and probably shows that the rigid frame-rigid top model with one torsional degree of freedom is failing for the high-J transitions (II, 12). The experimental value of the potential barrier V, = 2040.65 (91) cal/mol is in rather good agreement with the value found previously by Bak, I’, = 2007 cal/mol (3), assuming +(i, a) = 27.52”. The moment of inertia, ZO,of the methyl group could be accurately determined from the internal rotation analysis, Z? = 3.1976 us A’. Z, may be determined quite independently using the inertial defects of the normal and deuteriated species (13): Z, = -[A(CD3N=CD2)
- A(CH3N=CH,)]
= 3.1 19u*A2. We note that Z, < Zz. This behavior is rather general, as may be seen from an inspection of Table V of Ref. (14).
DEMAISON
260
ET AL.
A list of predicted transitions is given in Table VI. The frequencies in this table include all transitions with sufficient intensity over the range 60- 130 GHz. The line strength of each transition is given only for the A species because it is not significantly different for the E species. Further predictions may be obtained upon request from the authors. ACKNOWLEDGMENT This investigationhas been supported in part by the Centre National de la Recherche Scientifique (LA 249, ATP no. 9.82.10). RECEIVED:
January
23, 1984 REFERENCES
1. K. V. L. N. SASTRYAND R. F. CURL, J. Chem. Phys. 41, 77-80 (1964). 2. J. T. YARDLEY,J. HINZE, AND R. F. CURL, J. Chem. Phys. 41, 2562-2563 (1964). 3. B. BAK AND H. SVANHOLT,Acta Chem. Stand. A 31, 755-758 (1977). 4. J. C. GUILLEMINAND J. M. DENIS,Angew. Chem. Suppl. 1982; 15 15-1524. J. C. GUILLEMINAND J. M. DENIS,Angew. Chem. Int. Ed. Engi. 21,690-690 (1982). 5. J. M. DENIS,unpublishedresults. 6. J. BURIE,D. BOUCHER,J. DEMAISON,AND A. DUBRULLE,J. Physique 43, 1319-1325 (1982). 7. W. H. KIRCHHOFF, J. Mol. Spectrosc. 41, 333-380 (1972). 8. H. D. RUDOLPH,Z. Naturjbrsch. A 23, 540-543 (1968). 9. B. P. VAN EIJCK,J. VAN OPHEUSDEN, M. M. M. VAN SCHAIK,AND E. VAN ZOEREN,J. Mol. Spectrosc. 86,465-479 (198 1). 10. R. D. BROWN,P. D. GODFREY,AND D. A. WINKLER,Aust. J. Chem. 35,667-672 (1982). II. J. DEMAISON,A. DUBRULLE,D. BOUCHER,J. BURIE,AND B. P. VAN ELXK, J. Mol. Spectrosc. 94, 211-214 (1982). I2. J. SHERIDAN,W. B~SSERT,AND A. BAUDER,J. Mol. Spectrosc. 80, 1-I 1 (1980). 13. D. R. HERSCHBACH AND V. W. LAURIE,J. Chem. Phys. 40, 3142-3153 (1964). 14. J. DEMAISON,D. S~HWOCH,B. T. TAN, AND H. D. RUDOLPH,J. Mol. Spectrosc. 83,391~400 15. B. BAK, privatecommunication.
(1980).