8 (DEG)

10

5 UF

₆

E ₀ = 50 eV

7.86eV

<> <> ~

0 ^{0 0}

<> <>

0 0

a

^{d 0} ₀

-

.5

₀ ^9.26eV

>

_{Q) . I} ^{4.2eV A A A}

CX>

.

J' A A

LO

H

.05

"

^H

₊ ⁺

⁺

.01

-0

0

.005

.OOIL--~---~--~--~----~__...__~...__~--~--~--

0 20

40 60 80

>

_Q)

an .

co

H

"

^H

5

₀

-

D

.5

0

WF6

E

₀

= 50eV

0 _ r i

~

0 0 ^v=

12.6 eV{x I 0)

ll.75eV(X 10)

D

+

20 40 60 80

8 (DEG)

Figure 7.

3.3. Paper III: Doublet+ Quartet Transitions and Doublet+ Doublet Electronic Transitions in N02 by Electron Impact

Doublet +Quartet and Doublet + Doublet Electronic Transitions tn N02 by Electron Impactt

' §

R. Rianda, R. P. Frueholz and A. Kuppennann · Arthur Amos Noyes Laboratory of Chemical Physics

California Institute of Technology Pasadena, California 91125 USA

The electron-impact energy-loss spectrum of nitrogen dioxide (N02) has been measured at impact energies of 25, 50 and 75 eV, and scattering angles varying from 5° to 80°. A previously unreported spin-forbidden doublet+ quartet transition was observed at 4.49 eV,_ in excellent agreement with theorettcal calculations. Doublet +doublet transitions were observed at 2.95, 5.81, 7.48, 8.64; 9.69, 10.52, 10.68, 10.94 and 11.20 ev in agreement wiith previous experimental and theoretical work.

In addition, numerous doublet+ doublet transitions to super...;excited states were also observed.

t This work was supported by a contract

=!=work perfonned in partial fulfillment of the requirements for the PhD degree i"n Chemistry at the California Institute of Technology.

§ ·Present address: The Aerospace Corporation,

P.O.

Bc.>x 92957, Los Angeles, CA 90009.

Introduction

The electronic structure of N02 has been the subject of extensive experimental 1 - l 4 and theoretical investigations. 15 - 26 Agreement be- tween calculated and experimental doublet + doublet excitation energies has been quite good; however, prior to the present investigation, no experimental confinnation of calculated doublet ⁺ quartet transition energies was available. ln order to further elucidate the electronic structure of N02 with particular emphasis on the low lying quartet states, we have applied the technique of low-energy variable-angle electron-im- pact spectroscopy to this molecule.

Electron impact spectroscopy h.as been shown to be a very powerful tool for the investigation of both spin and syrrmetry forbidden transi- tions. 27 • JO The shape of the differential cross section (.DCS) versus scattering angle curve i's a sensitive indi'cator of the forbi ddeness or allowe.dness of a transi'tion. Optically allowed processes have very sharply forward peaked DCS's for impact energies 15 eV or more above the excitation threshold. In contrast, spin forbidden transitions (ones in which the spin quanturn number changes by unHyl have a nearly isotropic DCS over the angular range 10° to 80°. Such transitions occur by the mechani.sm of electron exchange. 31 Consequently, the ratio of the inten-

sity of a spin forbi·dden transition to that of a fully allowed transition i.ncreases by one to two orders of magnitude as the scattering angle

i.'ncreases from 10° to 80°.

The configuration of the X2A ground state of N02 is lla 1)

2 (lb2l²

2 2 2 2 2 2 2 2 2 1 26

(.2a1) (3a

1) (2b2) (4a

1} (Sa

1) (3b

2) (lb1) (4b2) (la2) (6a1) •

Transitions may occur to either valence or Rydberg-like orbitals.

Single valence or Rydberg excitations from the 6a1 orbital may only result in doublet excited states. In addition, valence transitions from lower lying orbi'tals to the 6a

1 may also only result in doublet excited states. All other transi'tions will result in both doublet and quartet excited states. The excitations to valence excited states are of par- tkular i-nterest in this study since, because of orbi'tal overlap consider- ati:ons, the doublet-quartet splitttngs are expected to be larger than those for Rydberg exci'tati.ons, leading to easier resolution of quartet states from doublet states.

Experimental

The e 1 ectron spectrometer use.d i.TI thi.'s study was that described by Kuppennann et ai.²⁷ Spectra of N0

2 i'n the energy loss ran9e Q-+ 20 eV

were.

obtai:ned at i.mpact energi:es of 2~, 50 and 75 eV and scattering angles from

s

⁰ to 80°. ·sample pressures ill the scatteri.ng cha.mber were typi'cally 4 mTorr as tndicated by an uncalibrated Schulz-Phelps tonization gauge whi'le the incident electron beam current was approximately 80. nArnps, The energy resolution was electron opti:call,y set to 120 meV (fWHM). The

N02 sample was obtained from Matheson Gas Products with a stated puri.ty of 99.5% and was subjected to several freeze-pump-thaw cycles before use.

Di's<:uss i'on

fi:gure 1 s.hows. the electron i.mpact spectrum of NQ2 from ^l.6 to 9,6 eV ener9y 1 oss at an i,mpact energy of 50 eV and scatteri.n9 angles of 10°

and 80°. Relatively intense features appear in both spectra at 2.95, 5.81, 7.48 and 8.64 eV. No inelastic electronic features are observed below 1.6 eV. Two electric dipole allowed transitions, the

x

^{2Al-+ A281}

and the

X

²A1 ^-+ 1²8₂, are predicted to lie near 3 eV energy loss. 18 - 24126 Krauss et al.~³ on the basts of~ long vibrational progression observed in their 100 ev,o⁰ electron impact spectrurn, have assigned the feature with an tntenstty maximum at about 2.96 eV in their spectrurn to the

x

²A

1 ^-+ 128 2 valence transi'ti'on. Our observed transition energy of 2.95 eV is in

excellent agreement with theirs and we assign this feature to that same transtti'on.

A transition with an i·ntensity maximum at about 2167A (5.72 eV) has 0

Ileen observed in the optical spectrum of Nakayama et al. 5 In our spectra, a feature is observed with an intensity maximum at 5.81 eV in very good agreement wi_th the results of Nakayama et al. 5 Thi"s transition has been tentati_vely identified as the

x

^2A₁ ^-+

^C

transition by Herzberg9 (no

synrnetry designatton has" been given l; however, it has been suggested that

0

the progression observed between 2000 and 2350A (.S.62 -+ 5.28 eV} may be

~2 -2 9 ' 23

part of the X A1 ^-+ B B2 system. The CI calculations of Shih et al.

i'ndicate only one transttton in the 5.5 to 6 eV range, the

X

²A1 ^-+

B

²

82

at 5.51 eV. Addtttonal optical studies by Coon et al. 11 have i_ndicated a transition with an i'ntensity maximum at 5.22 eV which has been assigned to the '

x

^2A '1 ^-+

s

²

^a·

2· Shi_·h et al. 23 have suggested that the true vert'ical excitation energy may be higher than that of the most probable vibrational t ransl' i-on. "t. I n a ltron, prev10us e ectron lrnpac s u les dd. . . 1 . t t d. lO h ave o -b served only one transition in this region, at about 5.7 eV. It ts

therefore not clear from our results whether the 5.81 eV feature may be

. -2 -2 - - .

assigned as the X A1 ⁺ B B2 or whether the X ⁺ B transition, which is quite weak (the absorption coefficient is roughly 20% of that of the X ⁺ C transition5

), is simply not resolved from an

x

^2A₁ ^{+ C} transition of unknown synrnetry.

The transition which we observe with an intensity maxi"mum at 7.48 eV

0 0

has been reported previously as a broad maximum between 1750.A and 1600A (7.1 and 7.75 eV) in the optical work of Nakayama et al. 5 and as an

tntense peak at 7.5 eV in the electron impact results of Edqvist et a1. 10 This feature has been assigned as the

X

2A1 ⁺

o

^2A₁ transition corresponding to the Rydberg excitation 6a₁ ⁺ 3s. Shih et al. 23 have calculated an exci:tation energy of 7 .71 eV for this transition i.n good agreement with experi.mental results.

The transition which we observe with a maximum at 8,64 eY h.as been reported previously by Price and Simpson¹ at 1450.A {_8,55 eV) and by Edqvist et a1. 10 at

~.6 ~V.

This transition has been assigned as due to excitation from the 6a1 orbital to the 3px,y,z orbi.tal. The calculated energtes for excitation to the 3px, 3py and 3pz orbitals are 8,84, 8.62 and 8.73 eV, respectively 23 and are tn excellent agreement with the experimental work.

In addition to these previously reported doublet + doublet trans- itions, we observe a feature at 4. 49 eV i.n the 80° spectrum which is not observed in the 10° spectrum and has not been reported previously. The DCS' s for thi.s feature and sever(il other features are shown i.n fi.gure 2.

The method used for obtai.ning DCS curves h.as been descri'bed previously. 29

The DCS for the 4.49 eV feature is constant to within a factor of about 2 over the angular range 10° to 80°. This behavior is exactly as is ex- pected for a spin forbidden transition, so the feature may be definitively assigned as doublet -+ quartet in nature. Two doublet-to-quartet trans- itions have been theoretically predicted to lie in this region, the

X

²A -+ 4B (la -+ 2b } and the 1 2 2 1

X

2A -+ 4A (4b -+ 2b ) 18 - 20 ,22 - 26 Due 1 2-2 1·

to resolution limitations, we are unable to detennine which of these transitions i's the primary contributor to the 4.49 eV feature we observe.

figure 3 shows the spectrum of N02 at an impact energy of 75 eV and a scattering angle of 5° in the energy loss region 9.5 to 19.5 eV. The excitation energi-es .of the numerous transitions which appear in this

spectrum are listed in Table 1. Edqvist et al.,10 also using the electron tmpact technqi ue, h.ave assi'9ned the features i"n thi.s energy loss re9i on to various Rydberg series. Our results are in excellent agreement with.·

those of Edqvist et a1. 10 ln addition to the features which may be correlated wi·th previous·1y identified transitions, we observe several weak transitions between 12.5 and 14 eV and between 15.5 and 16 eY energy loss, including one fairly intense feature at 13.86 eV which appears at excitation energies which exactly correspond to intense transitions of nitric oxide. 30 We thereforebeli.eve that these features are due to a small nitric oxide impurity.

In sumnary, we have used the technique of low-energy, variable- angle, electron impact spectroscopy to investi'gate the excited states of N02. We have observed a previously unreported doublet-+ quartet trans- ition at 4.49 eV. I:n addition, doublet-+ doublet transitions were ·

observed at excttation energies which were in good agreement with pre- vious experimental and theoretical work.

State Excitation

x

^{2 A} 1 Ground State

A

^{2 B} 1 6a 1 + 2b 1 2B 2 4b 2 + 6a 1 2A 2 la 2 + 6a 1 48 2 la 2 +·2b 1 4A 2 4b2 + 2bl

Table 1 Transition Energies for N0 2 Previous Theoretical (eV) Experimental (eV) 0 0 2.13~ 2.58~ 1.75~ 2.45e 2.81 -2.84m 2.8{ 2.46q 2;79~ 2.93i 2.977< .. 3.15~ 3.33~ 3.03e ~3.cl'; 2.96~ 3.oP 3.4( 2.43~ 3.22h • k 3.4o; 3.59 b . c d 2 . 46 ' 1. 68' 4 .13 3.09~ 3.4( 2.159' h i· k 3.25, 3.61; 3.63 3.38~ 3.80~ 4.6! 2.88~ 4.04~ 4.85i • k 4.47~ 4.53 3.43~ 3.81~ 4.7! h . 3.1~ 4.09, 4.7tz. . k 4. 3(i, 4 .37

This Worik CeV) 0 2.95 4.49

Table 1 (continued) State Excitation 2A 2 4b2 + 2bl

8

2

s

2 1a 2 + 2b 1 2A 2 4b2 + 2bl

o

^{2 A} 1 6a 1 + 3s 28 1 6a 1 + 3py 2A 1 6a 1 + 3pz 28 2 6a 1 + 3px 28 2 la 2 + 2b 1 48 2 4b 2 + 3s 28 2 6a 1 ,4b 2 + 2bf 4A 2 la 2 + 3s 28 2 4b 2 + 3s 2A 2 la 2 + 3s 2A 1 4b2 + 3p C1

Theoretical (eV)

4.02~

5.o/ a j;_· ~ h 3.95, 3.7 ' 5.1 ' .5.51 6.5~ 6.76h

dZ

b h 7. ' 9 .87' 7. 71 8.62h 8.73h 8.84h 9.07h 9.20h 9.33h 9.62h 9.70 9.86

Previous ExPerimental (eV) 5.22P 5. 72~ 5. 7P 7.1-7.75~ 7.sP 0. 55

,Z·a

.65~ 8.

ff

9.6f? 10.'4f?

This :Work (eV) 5.81 7.48 8.64 9.69 10.52

Table 1 (continued) Previous State Excitation Theoretical {eV) Experimental (eV) This Work (eV) 2A 2 la 2-+ 3s 10.9 28 1 la 2-+ -2pcr 10.85 10.94 28 2 4b 2-+ 3s 11.12 11.20 28 2 4b 2-+ 4s 11.55 11.61 2 B2 4b 2-+ 5s 12.11 12.20 28 4b 2-+ 6s 12.40 12.46 2 14.7 14.62 26 3b 2 -+ 3s 15.38 15.36 2 2 2 3b -+ 3p. 16.35 16.37 A2' ~2 2 1T 2 2 Al' Bl 3b2 -+ 3d'IT 17.22 17.23 28 2 3b 2 -+ 4s 17.34 17 .35 2 2 A2, B2 3b2 -+ 4p'IT 17.63 17.69 26 . ' 3b 2 -+ 5s 18.01 18.03 2 28 2 3b 2 -+ 6s 18.32 18.33 28 2 3b 2 -+ 7s 18.49 18.5 28 2 4~1 -+ 3p0' 18.63 18.65

.

- Ref. 15, 0 ~ef. 16, aRef~ 17, aRef. 18, eRef. 19, JRef. 20, 9'Ref. 22, hRef. 23, t·Ref. 24, .iRef. 25, kRef. 26, iRef •· 1, mRef. 2, nRef. 5, 0 Ref. 9, PRef. 11, qRef. 13.

1. W. C. Price and D. M. Simpson, Trans. Far. Soc., 37, 106 {_1941).

2. T. C. Hall, Jr. and F. E. Blacet, J. Chem. Phys.,20, 1745 (1952).

3. K. Mori, Sci. of Light, 3, 62;{1954).

4. K. Mori, Sci. of Light, 4, 130 {1955).

5. T. Nakayama, M. Y. Kitamura and K. Watanabe, J. Chem. Phys., 30, 1180 {_19.59).

6. R. K, Ritchie and A. D. Walsh, Proc.· Roy. Soc., 267A, 395 (1962).

7. Y. Tanaka and A. S. Jursa, J. Chem •. Phys., 36, 2493 (1962).

8. A. E. Douglas and K. P. Huber, Can. J. Phys., 43, 74 {_1965).

9. G. Herzberg, Electroni.c Spectra of Polyatomic Molecules, (Yan Nostrand, Princeotn, NJ, 19.66)_, p. 507-50.9.

10. O. Edqvist, E. Li"ndho1m, L. E. Se1i.n, L. Asbri'nk, C. E. Kuyatt, S. R. Mielczarek, J. A. Simpson and -1. fischer-Hja1mars, Physica Seri pta, 1, 172 (1970.).

11.

J.-B.

^Coon,

F. A.

Cesani and

F.

P. Huberman, J. Chern. phys., 52, 1647 (1970)_.

12. J. L. Hardwick and J. C. D. Brand, Chem. Ph.ys. Lett., 21, 458 (J973).

13.

M.

Krauss, R.

J.

Celotta, S.

R.

Mielczarek and C. E. Kuyatt, Chem.

14.

15.

16.

17.

18.

Phys. Lett., 27, 285 (19741.

R.

J. Pirkle and

V.

T. Jones,

J.

Mel. Spect,, 54, 375

w.

^{H. Fink,}

J.

Chem. Phys., 49, 5054 (1968 }.

w.

^{H. fink,}

J.

Chern. Phys., 54, 2911 ()9?1l.

J.

^E. Del Bene , J . Chem. Phys. , 54, 3487 (l 971 )_.

R.

A. Gangi and L. Burnel1e, J. Chem. Phys., 55, 851

l1975)..

(J97.l).

19. P. J. Hay, J. Chem. Phys., 58, 4706 {_1973).

20. G. D. Gillespie, A. U. Khan, A. C. Wahl, R. P. Hosteny and M. Krauss,

J. Chem. Phys., 63, 3425 (1975).

21. G.D. Gtllespie and A. U. Khan, J. Chem. Phys., 65, 1624 (1976}.

22. C. F. Jackels and E. R. Davidson, J. Chem. Phys., 65, 2941 (1976).

23. S.-K. Shi·h, S. D. Peyerimhoff and R •. J. Buenker, Chem. Phys. Lett., 46, 201 (19771.

24. P. A. Benfoff, J. Chem. Phys., 68, 340.5 (l978L 25. C. f. Jackels, J. Chem. Phys., 70, 4664 (1979)_.

26. N. C. Handy, J. D. Goddard and H.F. Schaefer Ill, J. Chem. Phys., 71, 426 (1979}.

27. (_a)_ A. Kuppemann, J. K. Rice ands. Trajmar, J. Phys, Chem., 72, 3894 (,1968).

(b) S. Trajmar, J. K. Rice and

A.

Kuppennann, Advan. Chem. Phys., 18, 15 (19701.

(c} A. Kuppennann,

W.

M. Flicker and O. A. Mosher, Chem.

Rev.,

^79,

77 {_1979).

28. O. A. Mosher, W. M. Fli.cker and A. Kuppennann, J. Chem. Ph.ys., 59, 6502 (J973)..

29. O. A. Mosher, W. M. Fli.cker and A. Kuppermann, J. Chem. Phys., 62, 2600 (1975}.

30. R.

P.

^Frueholz,

R.

Rianda and A. Kuppennann, Chern. Ph.ys., 31, 315 (1978)..

31. J. ~- Oppenheimer, Phys. Rev., 29, 433 (1927).

Fig. 1: Electron energy-loss spectrum of N02 at a scattering angle of (a) 10°, and (b) 80°; 50 eV impact energy; 8 x 10-8 A incident beam current; 5 mTorr sample pressure reading from an uncali- brated Shulz-Phelps gauge; resolution approximately 0.12 eV

(FWHM).

fig. 2: Dtfferential cross-s-ecti:ons of N02 as a function of scattering angle at an incident electron energy .of 50 eV; for elastic scattering

C+l

and for excited states:

n

_{2_95}

(Ol,

^Q₄^•₄₉^(p),

D5•81 ^(~)., and 07.48 ^(AL The elastiC peak DCS was Jnultiplted by 0.1 before plotting. The letters D and Q stand for doublet and quartetupperstates, respectively, and the index represents the corresponding transitton energy.

fi:g. 3: Electron energy-loss spectrmn of N02 at a scattering angle of

s

^0, 75 eY tnci.dent electron energy; 8 x 10-8 A incident beam current; 3 mTorr sample pressure meas.ured by an uncal i·brated Schulz Phelps gauge.

6000

4000

-

^"'C c: ₀ 2000

(.) Q)

~

_(/)

-

^c _::J

0 0

-

(.)

>- r-

CJ)

z

¹⁰

w r-

z

5

0 2

N02

E₀

=

^{50 eV}

8= 10°

N02

E₀= 50 eV

e

^=80°

3 4 5 6 7

~E

(eV) Figure I

8

(a)

(b)

9

. 5

- . I

§ .05

>.

L.

0 L.

.._

.0 L.

- _en

0

u .01

0

.005

20 40 60 80

e ^(deg)

Figure 2

>

_Q)

l{)

I'- 0 C\I II l{)

Q 0 ^II

z w

^Q:)

0 0

0 0

0 0

(.0 ~

(P uo:las/siu no:l)

co

.,...._

>

CL>

vw -<]

N

0 0

0 0

C\J

Al.ISN3..LN I

CL>

~

::J

·-

Cl

LL

3.4 Paper IV: Singlet ⁺ Triplet Transitions in C=N Containing Molecules ^by Electron Impact.

Singlet ⁺ Triplet Transitions in C=N Containing Molecules by Electron Impact

Ronald Rianda, Robert P. Frueholz, and Aron Kupperrnann Arthur Amos Noyes Laboratory of Chemical Physics

California Institute of Technology Pasadena, California 91125

The electron-impact excitation spectra of hydrogen cyanide (HCN), acetonitrile (CH3CN), malononitrile (CH2(CN)2), propionitrile (C2H5CN), and butyronitrile (C3H7CN) have been studied experimentally at impact energies of 25, 50 and 75 eV and at scattering angles from 5° to 80°.

Results for hydrogen cyanide are in excellent agreement with previous work. Previously unobserved singlet ⁺ triplet transitions of acetoni- trile, propionitrile and butyronitrile are reported. Also, the first study of the electronic spectrum of malononitrile is reported. Tentative assignments for transitions observed are reported.

Although much has been published on the electronic spectroscopy of C=N containing molecular systems, especially hydrogen cyanide, the

assignments of transitions observed are still very much in question.¹-16

In order to resolve some of these questions, we have applied the tech- nique of low-energy, variable-angle, electron-impact spectroscopy to hydrogen cyanide, acetonitrile, malononitrile, propionitrile and butyro- nitrile.

Electron impact spectroscopy has proven to be a very powerful tech- nique for studying the electronic structure of molecules.¹⁷-20 The shape of the differential cross-section (DCS) versus scattering angle curve is a great aid in the identification of the nature of a transition.

Excitations which are fully allowed with respect to dipole selection rules exhibit very strongly forward peaked DCS's. The DCS's of such transitions are most intense at 0° scattering angle and decrease by about two orders of magnitude as the scattering angle (e} is increased from 10° to 80°. Spin forbidden transitions exhibit DCS's which are nearly constant, to within a factor of two or three, over a similar angular range. Spin-allowed but symmetry-forbidden transitions display an inteMDediate behavior. The DCS is forward peaked but not so much as that of a fully allowed transition. Thus, measuring the DCS of a transitfon may add valuable information as to the type of transition observed. The lower energy region of the spectra of CN containing

*

*

nJolecules i"s expected to be dominated by TI-+n and n~7T transitions

within the C~N group. Therefore, hydrogen cyanide, the simplest and most thoroughly studied member of this group, will serve as a model to assist in assigning transitions in the remainder of the group. Hydrogen cyanide is linear in its ground state and thus belongs to the Coov point group

The electronic configuration in the ground state is 1₀2 2₀2 3₀2 402 502 _1TI4 yielding the ground state symmetry of lL+. 5, 6 The higher energy regions of the spectra are expected to be dominated by Rydberg transitions.

Hydrogen cyanide is bent in many of its excited states and in these cases and possesses Cs symmetry.³ This results in a splitting of the~ or- bitals into a' and a" components. In Cs symmetry, the ground state is la¹² 2a' 2 3a 12 4a 12 5a 12 6a 12 la 112 (1A^{11) .}5

' ⁶ Assignments for transitions of the alkyl cyanides will be listed in accordance with the C₀₀v symmetry of the C =N group only.

Experimental

The electron spectrometer used in this study was that described by Kuppermann et al~¹⁷ Spectra of hydrogen cyanide, acetonitrile, pro- pionitrile, butyronitrile and malononitrile in the energy loss range 0 ⁺ 17 eV were obtained at impact energies of 25, 50 and 75 eV and scattering angles from 5° to 80°. Sample pressures in the scattering chamber were typically 2 - 4 mTorr, as indicated by an uncalibrated Schulz-Phelps ionization gauge while the incident electron beam current varied from 10 to 100 nAmps. The samples of hydrogen cyanide (99.5%),

acetonitrile (99+%), propionitrile (98+%), butyronitrile, and malononitrile {99+%) were obtained from Fumico Inc., Aldrich Chemical Co.~ Inc. ,!·Eastman Kodak Co., Matheson, Coleman and Bell, andAldrichChemical Co., Inc.,

respectively. The minimum stated purity. when available, is giveri in parentheses. All samples were subjected to several trap-to-trap dis- tillations· (HCN) or freeze-pump thaw cycles before u~e. No indication of impurity was observed in any spectrum.

Results and Discussion

Figure 1 shows the electron impact spectra of hydrogen cyanide (HCN), acetonitrile {CH3CN), rnalononitrile (CH2(CN)2), propionitrile (C2H5CN) and butyronitrile (C3H7CN) from 4.6 to 11.6 eV energy loss at an impact energy of 25 eV and s~attering angles of 10° and 80° {70° for HCN). The spectra of hydrogen cyanide at 10° and 70° scattering angles are repeated in figure 2 so that the vibrational structure we report may be more

readily observed. Differential cross-section versus scattering angle curves for severa1 of the features appearing in this energy loss region in hydrogen cyanide, acetonitrile and malononitrile are shown in figures 3, 4 and 5, respectively~ The features in the spectra of propibnitrile and butyronitrile are so heavily overlapped that differential cross- sections could not be obtained. Fi·gure 6 shows the spectra of hydrogen cyanide, acetonitrile, malononitrile propionitrile and butyronitrile from 10 to 17 eV energy loss at an impact energy of 75 eV and a scat- tering ang1e of 5°.

Hydrogen Cyanide

Intense features are observed with intensity maxima at 8.75, 10.17, 10.41, 10.70, 10.86, 11.03, 11.30 and 11.50 eV in the 10° spectrum of hydrogen cyanide. A considerable amount of vibrational structure is evident on the 8.75 eV feature. In addition to these intense features much weaker features with apparent intensity maxima at about 6.4 eV and 7.8 eV are observed. The feature at 6.4 eV has been previously observed in both electron impact¹⁰ and optical spectra³ and has been attributed to the

X

^lL+-+

A1Au

^{(n-+ n*)} excitation. The theoretical calculations of Schwenzer et al.⁶ indicate that this transition is the lowest lying singlet -+ singlet excitation. They have adjusted their calculated transition energies to match the observed value of 6.48 eV. The X' 1L:+-+

A

^1A11

transition is forbidden in the linear configuration (the excited state bond angle is 125.0°) so the intensity is expected to be low, The second feature in the low angle spectrum of hydrogen cyanide occurs between 7.5 and 8.0 eV. This trarisition has also been previously observed at 7.5 eV in the electron impact spectra of Chutjian et a1.¹⁰ Schwenzer et al.⁶ predict the

X

^1L:+-+

B

^1A11 transition, which is also forbidden in the linear configuration, to occur at 7.52 eV. However, Asbrink et al. ,¹¹ on the basis of HAM/3 calculations have suggested that the transitions observed at 6.48 eY and 7.5 eV in both the optical and electron impact spectra result from transitions to triplet excited states and that due to low intensities the

i.

^{1L:+ + 1 1A11 and}

X

^{1L:+ + 2 1A11} transitions are not observed. In our 25 eV, 70° spectrum (figure 2b), we observe a broad feature between 5~5 and 7.25 eV and another feature at J.86 eV

assigned as singlet +triplet transitions. Vibrational structure is evident on both features. Chutjian et a1.¹⁰ also observed transitions at 6.0, 6.8, 7.1-7.9 and 7.9 eV which they have assigned on the basis of the calculations of Schwenzer et al.⁶ to the transitions

X ¹r+ + 1 3A^1, 1 3A^{11, 2 3}A^1, X lL:+ + 2 3A^11, X ¹

t

+ 3 3A¹ and

X ¹r+ + ^{3 3}A^11, respectively. No vibrational structure was observed in their spectra at their stated resolution of 80 to 100 meV. Close

examination of our 25 eV, 10° spectrum reveals the presence of vibrational structure which also appears with much greater intensity in the 25 eV,

80° spectrum and may definitely be assigned to singlet + triplet transitions.

Unfortunately, it is only possible at this time to state that our results are consistent with assignment of the transitions observed between 5.5 and 7,25 and between 7.5 and 8.0 eV in both optical and low-angle electron impact spectra to singlet+ triplet excitations.

The first intense feature in the 10° spectrum of hydrogen cyanide appears between 8.0 and 9.63 eV. Previous investigations have identified two transitions in this region, the

X

l~+ +

C

^{1A1 and the}

X

^{l~+ + D 1A1 •3 .. 11}

The 0-0 bands of these transitions have been reported to be at 8.139 and 8.881 eV, respectively. We observe a progression in the bending mode, v2, of about .10 eV beginning at 8,03 eV. This is in excellent

~ -1

agreement with the v2 frequency for the C state of 869 cm (.108 eV) obtained by Herzberg.³ No vibrational structure was resolved in this energy region in the previous electron impact studies of Chutjian et al.¹⁰ and Tam and Brion.⁷ Fridh and Asbrink8 have observed and assigned the