Regioselectivity of the photochemical addition of ammonia, phosphine, and silane to olefinic and acetylenic nitriles

Article abstract of DOI:10.1002/(SICI)1521-3765(19980615)4:6<1074::AID-CHEM1074>3.0.CO;2-B

An investigation of the regioselectivity and mechanisms of the photochemical addition of NH₃, PH₃, and SiH₄ to olefinic and acetylenic nitriles is described. The photolysis of NH₃ in the presence of acrylonitrile led to the α-addition product 2-aminopropanenitrile (2), propanenitrile, and 2,3-dimethylbutanedinitrile (3). When NH₃ was photolyzed in the presence of substituted derivatives (crotononitrile, methacrylonitrile, or 1-cyclohexenecarbonitrile), the α-addition products were still obtained. However, under similar reaction conditions, only the β-addition products, 7 and 8, were obtained from acrylonitrile and PH₃, or acrylonitrile and SiH₄, respectively. On the other hand, the photolysis of 2-butynenitrile and NH₃ gave the β-addition products, (Z)- and (E)-3-aminocrotononitrile (10). The photolysis of these acetylenic nitriles with PH₃ or SiH₄ also gave the β-adducts (12) and (13). The α-addition of NH₃ proceeds by the stepwise addition of H· and ·NH₂, respectively, to the α,β-unsaturated nitriles. The β-addition products are formed by a radical chain mechanism initiated by photochemically generated radicals. The radical chain pathway provides an explanation for a number of previously described photochemical additions to olefins and acetylenes. Photochemical processes similar to the addition of ammonia and phosphine to unsaturated organic compounds may have played a role in the evolution of the atmosphere of the primitive Earth, and may even be currently occurring in the atmospheres of other planets.

Full text of DOI:10.1002/(SICI)1521-3765(19980615)4:6<1074::AID-CHEM1074>3.0.CO;2-B

FULL PAPER
Regioselectivity of the Photochemical Addition of Ammonia, Phosphine, and
Silane to Olefinic and Acetylenic Nitriles
Jean-Claude Guillemin,* Curt M. Breneman, Jeffrey C. Joseph, and James P. Ferris
Abstract: An investigation of the regio-
selectivity and mechanisms of the pho-
tochemical addition of NH3, PH3, and
SiH4 to olefinic and acetylenic nitriles is
described. The photolysis of NH3 in the
presence of acrylonitrile led to the a-
addition product 2-aminopropanenitrile
(2), propanenitrile, and 2,3-dimethylbu-
tanedinitrile (3). When NH3 was photo-
lyzed in the presence of substituted
derivatives (crotononitrile, methacrylo-
nitrile, or 1-cyclohexenecarbonitrile),
the a-addition products were still ob-
tained. However, under similar reaction
conditions, only the b-addition products,
7 and 8, were obtained from acrylonitrile
and PH3, or acrylonitrile and SiH4,
respectively. On the other hand, the
photolysis of 2-butynenitrile and NH3
gave the b-addition products, (Z)- and
(E)-3-aminocrotononitrile (10). The
photolysis of these acetylenic nitriles
with PH3 or SiH4 also gave the b-adducts
(12) and (13). The a-addition of NH3
are formed by a radical chain mecha-
nism initiated by photochemically gen-
erated radicals. The radical chain path-
way provides an explanation for a num-
ber of previously described photo-
chemical additions to olefins and acety-
lenes. Photochemical processes similar
to the addition of ammonia and phos-
phine to unsaturated organic com-
pounds may have played a role in the
evolution of the atmosphere of the
primitive Earth, and may even be cur-
rently occurring in the atmospheres of
other planets.
.
proceeds by the stepwise addition of H
.
and NH2, respectively, to the a,b-unsa-
turated nitriles. The b-addition products
Keywords: ab initio calculations ´
alkenes ´ alkynes ´ radical reactions
´ photochemistry
imidazole;[2] however, there have been very few studies of the
photochemical addition reactions of unsaturated nitriles. The
subsequent transformation of these compounds occurs mainly
in photochemically driven reactions. It is not possible to infer
the regioselectivity of the photochemical addition reactions of
unsaturated nitriles, esters, and acids because of the conflict-
ing nature of the limited number of reports available in the
literature. Photolysis of ethyl propynoate in the presence of
isopropanol resulted in the b-addition of the isopropanol
group to the triple bond [Eq. (1)],[10] while a-addition is the
principal reaction pathway in the photochemical addition of
indene to acrylonitrile [Eq. (2)],[11] in which the ground state
of acrylonitrile adds to the excited state of indene. Both a- and
b-additions were observed in the photochemical reaction of
ammonia to ammonium acrylate [Eq. (3)].[12] b-Addition was
the exclusive product at a high photon flux or with UV
Introduction
Unsaturated nitriles play a key role in many of the pathways
proposed for the prebiotic synthesis of biological molecules.
These nitriles are formed in electric discharge reactions
designed to simulate events on the primitive Earth.[1, 2] They
are found on Titan[3, 4] and in the interstellar medium.[5, 6]
Michael addition (b-addition) of nucleophiles to acetylenic
nitriles is an important step in the prebiotic synthesis of the
pyrimidine ring system,[2, 7] while the addition of HCN and
NH3 to propynenitrile results in the formation of the dinitrile
of aspartic acid.[1] Aminonitriles are utilized in almost every
step in research designed to elucidate the low-energy routes to
prebiotic molecules.[8]
Photochemical reactions of unsaturated aminonitriles play
an important role in the prebiotic synthesis of purines[9] and
OH
hν
(1)
(2)
Me
C C
CHCO2Et
(CH3)2CHOH
+
HC CCO2Et
[*] Dr. J.-C. Guillemin
MeH
Department of Chemistry, Rensselaer Polytechnic Institute
Troy, NY 12180-3590 (USA)
1
Current address:
hν
+
H2C CHCN
+
Á
Â
Laboratoire de Synthese et Activation de Biomolecules
ESA CNRS No.6052, ENSCR, 35700 Rennes (France)
Fax : (‡ 33)299-87-13-84
CH3CHCN
CH3CHCN
E-mail: jean-claude.guillemin@ensc-rennes.fr
H3C
H2N
Prof. Dr. C. M. Breneman, Dr. J. C. Joseph, Prof. Dr. J. P. Ferris
Department of Chemistry, Rensselaer Polytechnic Institute
Troy, NY 12180-3590 (USA)
hν
H2C CHCO2- NH4+ + NH3
CH CO2- NH4 + NH2CH2CH2CO2- NH4
+
+
(3)
1074
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
0947-6539/98/0406-1074 $ 17.50+.50/0
Chem. Eur. J. 1998, 4, No. 6

1074±1082
Table 1. Photolysis of acrylonitrile and ammonia.[a]
Irradiation
l [nm]
Pressure [Torr]
NH3 Acrylonitrile
Light absorbed
by NH3 [%]
Products
3[b] [%]
2[b] [%]
Propanenitrile [%]
185, 254
185, 254
185, 254
185, 254
206
206
206
206
10
50
300
560
10
50
300
560
50
50
10
3
50
50
10
3
9[c]
33[c]
94[c]
99[c]
30[d]
68[d]
98[d]
100[d]
0
0
0.5
4.5
0
0
0.7
4.5
0
0
0
0
0.4
3.0
0
0
0.5
3
1.3
8.5
0
traces
3.2
10
[a] Irradiation time ˆ 3 h. [b] 2: 2-Aminopropanenitrile; 3: diastereomers of 2,3-dimethylbutanedinitrile. [c] UVabsorbance of acrylonitrile was estimated to
be 178 cm 1 atm 1 at l ˆ 185 nm and 22 cm 1 atm 1 at l ˆ 206 nm.[15]
propanenitrile (alaninenitrile, 2) was observed when ammo-
radiation (l ˆ 140 ± 160 nm) from a hydrogen lamp. Both a-
nia was photolyzed in the presence of acrylonitrile (Scheme 1,
Table 1). Two other products, propanenitrile and the diaster-
eomers of 2,3-dimethylbutanedinitrile (3), were also identi-
fied. These three products were only observed if NH3 was
alanine and b-alanine were observed when a low-intensity
laser source (l ˆ 266 nm) was used. It should be noted that the
identity of the photoproducts rests entirely on their retention
times from an ion-exchange column, and therefore requires
confirmation.
An investigation of the regioselectivity and possible
mechanisms of the photochemical addition of ammonia
(NH3), phosphine (PH3), and silane (SiH4) to acetylenic and
olefinic nitriles is presented in this report. A comparison of
the photoproducts provided insight into the reaction pathway.
hν
NH3
.
.
NH2
+
H
.
.
.
+
H2C CHCN
H
CH3CHCN
M
.
+
NH2
CH3CHCN
CH3CHCN
NH2
Results and Discussion
Scheme 1. Photolysis of ammonia in the presence of acrylonitrile.
Photoaddition to a,b-unsaturated nitriles: The photolysis of
NH3 in the presence of olefinic nitriles resulted in the a-
.
addition of NH2 [Eq. (4)]. The formation of 2-amino-
present in excess and was the light-absorbing species. No
addition products were observed when acrylonitrile was the
principal light-absorbing species. Trace amounts of propyne-
nitrile and HCN were detected in a control reaction when
acrylonitrile was irradiated by itself in the gas phase but the
isomers of 3 were not observed.[13]
MeH
+
NC C C CN
H3C
H2N
hν
(4)
CH3CH2CN
+
CH CN
H C CHCN
+
NH3
2
H Me
2
3
Comparable reaction products were obtained by the
photolysis of NH3 in the presence of crotononitrile, meth-
acrylonitrile, and 1-cyclohexenecarbonitrile (Tables 2 and 3).
The low vapor pressure (0.7 Torr at 258C) of 1-cyclohexene-
carbonitrile precluded photochemical studies where it was
present in excess, but the formation of 1-aminocyclohexane-
carbonitrile (4) was observed in the presence of excess
ammonia. The corresponding dicyano compound, the diaster-
eomers of 2,3-dicyclohexylbutanedinitrile (5), and the satu-
rated nitrile, cyclohexanecarbonitrile (6), were also obtained
[Eq. (5)].
Â
Â
Â
Â
Abstract in French: La regioselectivite et le mecanisme de
lꢀaddition photochimique de NH3, PH3 et SiH4 sur des nitriles
ethyleniques ou acetyleniques sont etudies. La photolyse de
lꢀammoniac en presence dꢀun acrylonitrile conduit a un a-
aminonitrile, produit dꢀa-addition, a un dinitrile et a un
alkylnitrile. Dans les memes conditions experimentales, PH3
et SiH4 en presence dꢀun nitrile ethylenique ou acetylenique
conduisent aux produits de b-addition de meme que NH3 en
Â
Â
Â
Â
Â
Â
Â
Á
Á
Á
Ã
Â
Â
Â
Â
Â
Â
Ã
Â
Â
Â
presence dꢀun nitrile acetylenique. Un processus mettant en jeu
Â
une reaction radicalaire en chaîne explique la formation des
produits de b-addition. Avec NH3, la formation dꢀadduits a
CN
Á
Â
provient de lꢀaddition, en premiere etape, dꢀun radical hydro-
CN
hν
+
+ NH3
CN
+
(5)
CN
Á
Â
gene sur lꢀinsaturation; les radicaux formes nꢀont pas une
NH2
CN
Â
Á
Á
energie suffisante pour arracher un hydrogene a lꢀammoniac et
4
6
5
Â
Â
debuter une reaction en chaîne. Sous irradiation solaire, des
hν
Á
processus photochimiques similaires a ceux de lꢀaddition de
(6)
(7)
H2PCH2CH2CN + CH3CH2CN
H2C CHCN
H2C CHCN
PH3
+
Â
lꢀammoniac ou de la phosphine sur des composes organiques
insatures peuvent avoir jouer un role important dans lꢀevolu-
ution de lꢀatmosphere primitive terrestre et peuvent actuelle-
ment sꢀeffectuer dans les atmospheres dꢀautres planetes.
7
Â
Ã
Â
Á
hν
SiH4
+
H3SiCH2CH2CN + CH3CH2CN
Á
Á
8
Chem. Eur. J. 1998, 4, No. 6
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
0947-6539/98/0406-1075 $ 17.50+.25/0
1075

J.-C. Guillemin et al.
FULL PAPER
Table 2. Photochemical addition of NH3 to crotononitrile.[a]
Irradiation
l [nm]
Pressure [Torr]
NH3 Crotononitrile
Light absorbed
by NH3 [%]
Products
15[b] [%]
16[b] [%]
n-Butanenitrile [%]
185, 254
185, 254
185, 254
185, 254
206
206
206
206
5
25
300
560
5
25
300
560
25
25
10
3
25
25
10
3
6[c]
24[c]
94[c]
98[c]
2[d]
0
0
0.5
4.5
0
0
0.7
4.5
0
0
1
2
0
0
1.5
6
0
0
1.5
2
0
0
1.5
6
8[d]
72[d]
94[d]
[a] Irradiation time ˆ 3 h. [b] 16: 2-Aminobutanenitrile; 15: diastereomers of 2,3-diethylbutanedinitrile. [c] UV absorbance of crotononitrile was estimated
to be 271 cm 1 atm 1 at l ˆ 185 nm. [d] UV absorbance of crotononitrile was estimated to be 140 cm 1 atm 1 at l ˆ 206 nm.
Table 3. Photochemical addition of NH3 to methacrylonitrile.[a]
Irradiation
l [nm]
Pressure [Torr]
NH3 Acrylonitrile
Light absorbed
by NH3 [%]
Products
18[b] [%]
2-Butanenitrile [%]
17[b] [%]
185, 254
185, 254
185, 254
185, 254
206
206
206
206
206
10
50
300
560
10
50
50
10
3
50
50
10
3
7[c]
27[c]
92[c]
99[c]
2[d]
11[d]
80[d]
96[d]
99[d]
0
‡
0
‡
‡
‡
0.2
4.9
‡
0.8
6.9
‡
0.2
3.0
‡
50
‡
‡
‡
300
560
560
0.5
5.3
8.0
2.0
0.7
3.0
3.0
10
14
1
[a] Irradiation time ˆ 3 h; ‡ : detected but <0.1% yield. [b] 17: 2-Amino-2-methylpropanenitrile; 18: 2,2,3,3-tetramethylbutanedinitrile. [c] UVabsorbance
1
1
of methacrylonitrile was estimated to be 240 cm 1 atm at l ˆ 185 nm. [d] UV absorbance of methacrylonitrile was estimated to be 57 cm 1 atm at l ˆ
206 nm.
Table 5. Photochemical reaction of SiH4 and acrylonitrile.[a]
On the other hand, when PH3 was photolyzed in the
presence of acrylonitrile, the b-addition product, 3-phosphi-
nopropanenitrile (7), was observed [Table 4, Eq. (6)]. Pro-
panenitrile is also formed but 3 was not detected. Photolysis of
SiH4 and acrylonitrile gave 3-silylpropanenitrile (8) and
propanenitrile [Table 5, Eq. (7)]. However, SiH4 does not
absorb light of l ˆ 185 nm;[14] only acrylonitrile is activated by
radiation of this wavelength. Products 7 and 8 were only
observed when PH3 or SiH4 were present in excess, even if
they were not the light-absorbing species.
Irradiation
l [nm]
Pressure [Torr]
SiH4
Products[c, d]
Propanenitrile [%]
H3C3N[b]
8 [%]
185, 254
185, 254
185, 254
185, 254
185, 254
70
75
60
40
8
2
8
8
40
75
3.0
1.6
0.6
traces
0
traces
2.7
1.0
0
0
[a] Irradiation time ˆ 1 h. [b] 100% light absorbed at l ˆ 185 nm by
acrylonitrile. [c] 8: 3-Silylpropanenitrile. [d] Traces of unidentified prod-
ucts have been observed but the a-adduct MeCH(CN)SiH3 was never
detected.
It was initially surprising that a-aminonitriles were ob-
tained by the photoaddition of NH3 to acrylonitrile while b-
addition products were obtained from the photoaddition of
PH3 and SiH4. The observation of 3 as a photoproduct clearly
.
acetylene proceeds 700 times more rapidly than NH2 addi-
.
indicated the formation of the radical CH3C (H)CN as a
tion.[17] Hydrogen atom addition to acrylonitrile proceeds at
the b-carbon because the radical formed is stabilized by
delocalization of its p orbital by the p-bond of the nitrile
group. Ab initio calculations at several levels of theory show
that hydrogen atom addition to the b-carbon of acrylonitrile
gives a more stable radical than if the addition was at the a-
carbon or the nitrile group (Table 6). The same addition
reaction intermediate in the NH3 ± acrylonitrile system. This
observation suggests that the addition reaction proceeds by a
stepwise process in the pathway given in Scheme 1. Here the
.
.
photochemically generated H and NH2 add to acrylonitrile,
with the hydrogen atom adding first. The more rapid addition
.
.
of H is consistent with the calculation that H addition to
Table 4. Photochemical addition of PH3 to acrylonitrile.[a]
Irradiation
l [nm]
Pressure [Torr]
Acrylonitrile
Light absorbed
at 185 nm by PH3[b] [%]
Products[c]
Propanenitrile [%]
PH3
7[d] [%]
P2H4 [%]
185, 254
185, 254
185, 254
185, 254
185, 254
70
70
60
40
6
2
4
6
40
60
95
91
85
36
5
2.4
2.4
1.9
0.3
0
1.4
7.0
6.3
2.1
0.2
0.4
traces
traces
0
traces
[a] Irradiation time ˆ 1 h. [b] UVabsorbance of acrylonitrile was estimated to be 178 cm 1 atm 1 at l ˆ 185 nm,[15] UVabsorbance of PH3 was estimated to be
100 cm 1 atm 1 at l ˆ 185 nm.[16] [c] The a-adduct MeCH(CN)PH2 was never detected. [d] 7: 3-Phosphinopropanenitrile.
1076
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
0947-6539/98/0406-1076 $ 17.50+.50/0
Chem. Eur. J. 1998, 4, No. 6

Addition to Nitriles
1074±1082
.
Table 6. Calculated energies of H addition products of acrylonitrile.
The formation of b-substituted products in the photo-
addition of PH3 and SiH4 to acrylonitrile indicated that
.
CH3C (H)CN was not a reaction intermediate. The major
difference between NH3 addition and the addition of PH3 and
SiH4 is the much higher NH bond energy of NH3
(107.4 kcal)[18] versus that of PH in PH3 (78 kcal),[19] and SiH
in SiH4 (90.3 kcal).[18] Spin-projected MP2/6-31G‡ ab initio
.
calculations (Figure 1) suggest that the abstraction of H from
.
NH3 by CH3C (H)CN is not energetically favored, while the
much lower bond energies of PH3 and SiH4 suggest that
.
.
abstraction occurs to generate PH2 and SiH3, respectively.
.
The b-adduct of NH3 with acrylonitrile is observed because H
addition generates a stabilized a-cyano radical (Scheme 2).
Hydrogen abstraction from PH3 and SiH4 is fast because of
the high partial pressures of PH3 or SiH4 in the reaction cell.
Consequently, dimerization of the radical intermediates is not
observed with PH3 and SiH4, while dimer 3 is observed in the
NH3 addition reaction.
The mechanism of the addition of PH3 and SiH4 differ from
that of the addition of NH3 in that PH3 and SiH4 add by a
radical chain process initiated by any radical species that can
abstract hydrogen from these gases. The resulting radical
.
.
chain process then generates the PH2 or SiH3 radicals which
mechanism accounts for the regioselec-
tivity of the photochemical addition of
NH3 to crotononitrile, methacrylo-
nitrile, and 1-cyclohexenecarbonitrile.
It should be noted that ab initio
calculations involving molecules with
unpaired electrons are often more dif-
ficult to interpret than those of closed-
shell molecules. The reason for this
difficulty stems from the potential for
spin contamination of the final wave-
function resulting from the admixture
of states of higher multiplicity with the
Figure 1. Energetics of hydrogen abstraction from NH3.
desired state. This can lead to distorted
geometries if optimization is attempted,
or to artificially low energies in single-
hν
point calculations. Several options are available to minimize
these effects. First, it is possible to specify that the spatial part
of the wavefunctions for a- and b-electrons be identical. This
type of calculation uses an ROHF (restricted open-shell
Hartree ± Fock) Hamiltonian which cannot result in spin
contamination, but it can also incorrectly represent the spin
density and the underlying electron density of the system.
UHF (unrestricted Hartree ± Fock) calculations use separate
spatial descriptions for each of the electrons, but can lead to
the spin-contamination problems described above. For com-
parative purposes, all of the ab initio calculations in this paper
were performed with several levels of theory for both
geometry optimization of the radical species as well as for
the calculation of the final energies of each species. The most
reliable method used in this work to remove spin contam-
ination and to account for electron correlation is PMP2/6-
31G‡, a method in which the Mùller± Plesset second-order
perturbation method is used to give a spin-projected, corre-
lated wavefunction.
.
.
PH2
+
PH3
H
.
.
+ H2C CHCN
H
CH3CHCN
.
+
PH3
CH3CHCN
.
PH2
CH3CH2CN
+
.
.
PH2
H2C CHCN
+
H2PCH2CHCN
.
.
+ PH3
H2PCH2CHCN
PH2
H2PCH2CH2CN +
Scheme 2. Photolysis of phosphine in the presence of acrylonitrile.
perpetuate the chain. The equivalent of the radical chain
addition of PH3 or silylphosphines[21] to olefins has been
proposed. In the case of NH3 the radicals which add are
generated by NH3 photolysis. In the case of PH3 the radicals
which initiate the chain reaction are formed by dissociation of
PH3 to PH2 and H ,[22] but in the case of SiH4 the radicals are
[20]
.
.
Chem. Eur. J. 1998, 4, No. 6
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
0947-6539/98/0406-1077 $ 17.50+.25/0
1077

J.-C. Guillemin et al.
FULL PAPER
generated by photolysis of the acrylonitrile since SiH4 does
not absorb light at wavelengths equal to or greater than
185 nm.[14]
An investigation of the photolysis of a mixture of NH3
(400 Torr) and 2-butynenitrile (100 Torr) at l ˆ 254 nm for
24 h at 48C was not successful. 2-Butynenitrile has a weak
absorption at l ˆ 254 nm,[25] while NH3 does not absorb at
all.[26] The only observed products were the E and Z isomers of
3-aminocrotononitrile (10), however, comparable yields of
this isomer mixture were formed in a nonirradiated control so
it could not be determined if these were photoproducts.
It was not possible to study the gas-phase photochemical
addition of NH3 to butynedinitrile because the thermal
process proceeded almost immediately at room temperature
upon mixing the gases. A mixture of butynedinitrile (14 Torr)
and NH3 (59 Torr) resulted in a 57% yield of 2-amino-2-
butenedinitrile (11), which consists of 92% Z and 8% E
isomers [Eq. (9)].[27] Compounds (E)-11 and (Z)-11 were
The greater ease of hydrogen abstraction from PH3 and
SiH4 versus NH3 also explains the formation of propanenitrile
if PH3 or SiH4 are irradiated in the presence of acrylonitrile.
The radical formed by the b-addition of a hydrogen atom to
acrylonitrile can abstract a hydrogen atom from PH3 or SiH4
to form propanenitrile (Scheme 2). The same radical, formed
by the photolysis of NH3 in the presence of acrylonitrile
(Scheme 1), cannot abstract a hydrogen from NH3 so no
propanenitrile is formed by this route. The relatively low
proportion of propanenitrile, as compared with the reaction of
acrylonitrile with PH3 and SiH4, must be formed by the
.
combination of a hydrogen atom with CH3C (H)CN.
H
H
NC
H2N
NC
hν
Photoaddition to a,b-unsaturated acetylenic nitriles: Photo-
lysis of mixtures of NH3 and propynenitrile resulted in b-
addition to the triple bond with the formation of the E and Z
isomers of 3-amino-2-propenenitrile (9). The structure of this
compound was determined by 1H and 13C NMR spectro-
scopy.[23] Control reactions established that the thermal b-
addition of NH3 to propynenitrile proceeds almost as rapidly
as the photoinitiated reaction. Consequently, it is not certain
that the observed b-addition products are actually the
consequence of photochemical initiation. Subsequent studies
were performed with 2-butynenitrile: the thermal addition
reaction proceeds much more slowly than the photochemi-
cally initiated process.
+
C
C
C
C
(9)
NH3 + NCC CCN
H2N
CN
CN
(E)-11
(Z)-11
identified by 1H and 13C NMR spectroscopy and high-
resolution mass spectrometry (HRMS). No hydrazine
(N2H4) was detected as a photoproduct when NH3 was
photolyzed. This is because N2H4 is rapidly decomposed to
N2 and H2 by the hydrogen atoms produced during NH3
photolysis.[28]
The photolysis of a mixture of propynenitrile and PH3 or
SiH4 yielded 3-phosphinoacrylonitrile [12, Eq. (10)] and 3-
silylacrylonitrile [13, Eq. (11)] respectively, along with acryl-
Photolysis of 40:2.5 Torr mixtures of NH3 and 2-butyne-
nitrile, respectively, in which NH3 absorbed >95% of the
light, resulted in the b-addition of NH2 to the triple bond and
the formation of the E and Z isomers of 3-aminocrotononi-
trile (10) [Table 7, Eq. (8)]. Conversions of 1 ± 3% were deter-
hν
HC CCN
HC CCN
H2PCH CHCN
(10)
(11)
PH3
+
.
12
hν
H3SiCH CHCN
SiH4
+
13
H
H
H2N
H3C
H3C
hν
+
C
C
C C
(8)
NH3 + H3CC CCN
CN H2N
CN
onitrile (Tables 8 and 9). The (Z) and (E) phosphines 12 were
obtained in 85:15 ratio and characterized by 1H and 31P NMR
spectroscopy and high-resolution mass spectrometry
(HRMS). For each isomer, three signals were observed in
(E)-10
(Z)-10
1
mined by H NMR after irradiation times of 0.25 ± 1 h. The
yield did not increase if the irradiation time was extended
because a polymer formed on the inside of the cell which
absorbed the UV light. The yield was not changed by the
addition of ethane (150 Torr) or N2 (600 Torr) to the reaction
mixture.[24] 2-Aminocrotononitrile[8] was not detected in the
NMR spectra of the product mixture. Traces of crotononitrile
were observed after a short irradiation time, but disappeared
on longer irradiation.
1
the H NMR spectrum. For the Z isomer, the signal at d ˆ
3.85, with a coupling constant J ˆ 215 Hz, is typical of the
hydrogens attached to a phosphorus atom. The signals of the
vinylic protons were observed at d ˆ 5.87 and 7.04 with a
coupling constant 3J(H,H)cis ˆ 12.6 Hz and the expected
multiplicity. The observation of the molecular ion by HRMS
(calcd for C3H4NP: 85.0081, found: 85.0079) confirms the
structure. The (Z)- and (E)-3-silylacrylonitriles (13) were
Table 7. Photochemical addition of ammonia to 2-butynenitrile.
Irradiation
l [nm]
Pressure [Torr]
2-Butynenitrile N2
Light absorbed
by NH3 [%]
Irradiation time [min]
(Temperature [8C])
(Z)-10[c]
[%]
(E)-10[c]
[%]
[a]
[b]
NH3
Ethane
185, 254
185, 254
185, 254
185, 254
40.0
40.0
40.0
40.0
2.5
2.5
2.5
2.5
0.0
600.0
0.0
0.0
0.0
150.0
0.0
> 95
> 95
> 95
> 95
15 (25)
5 (25)
10 (25)
30 ( 8)
0.47
0.61
0.70
0.37
1.02
1.07
1.34
0.76
0.0
[a] The extinction coefficient of NH3 at l ˆ 185 nm was taken to be 87.0 cm 1 atm 1. [b] The extinction coefficient of 2-butynenitrile at l ˆ 185 nm was taken
1 [25]
to be 13.5 cm 1 atm
.
[c] 10: (Z)- and (E)-aminocrotononitrile.
1078
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
0947-6539/98/0406-1078 $ 17.50+.50/0
Chem. Eur. J. 1998, 4, No. 6

Addition to Nitriles
1074±1082
.
Table 8. Photochemical reaction of PH3 and propynenitrile.[a]
initiated by the a-cyano radicals formed by H addition to the
b-carbon of 2-butynenitrile, to generate a cyanovinyl radical
(14), the most stable of all the possibilities (Tables 10 and 11).
Irradiation[a] Pressure [Torr] Light
Products[c]
12[d] [%] Acrylo-
nitrile [%]
l [nm]
PH3 HC3N
absorbed by
PH3 [%]
P2H4 [%]
[b]
.
This abstracts hydrogen from NH3 to generate NH2. The
185, 254
185, 254
185, 254
254
254
254
75
40 40
8
75
8
98
87
42
0
0
0
0.4
1.3
0.45
0.1
2.9
0.63
1.3
0.06
0.36
0.03
2.0
0.45
ꢀ 0.3
0
0
0
0
0
.
Table 10. Propynenitrile H addition series.
74
8
40 40
75
8
[a] Irradiation time ˆ 1 h at l ˆ 185 and 254 nm or 3 h irradiation at l ˆ 254 nm. A
medium-pressure mercury lamp Philips TUV6W was used for these experiments.
[b] Percent light absorbed at l ˆ 185 nm; UV absorbance of PH3 was estimated to
1
be 100 cm 1 atm at l ˆ 185 nm.[16] [c] Products formed from the photolysis of
HC3N alone[32] were never observed. [d] 12: 3-Phosphinoacrylonitrile.
Table 9. Photochemical reaction of SiH4 and propynenitrile.[a]
Irradiation
l [nm]
Pressure [Torr]
SiH4
Products[c]
[e]
HC3N[b]
13[d] [%] Acrylonitrile[d] Ph(CN)3
COT[e]
[%]
[%]
[%]
185, 254
185, 254
185, 254
254
254
254
140
40
7
100
40
8
40
40
100
30
40
75
0.4
0.26
nd
0.58
0.12
0
0.25
0.09
nd
0.34
0.10
0
0.2
0
0
0
0.41
0.55
traces
0.3
traces
0
0
0.5
[a] Irradiation time ˆ 1 h at l ˆ 185 and 254 nm or 3 h at l ˆ 254 nm. A medium-
pressure mercury lamp Philips TUV6W was used for these experiments. [b] 100%
light absorbed by propynenitrile. [c] Traces of unidentified silanes have also been
observed. [d] 13: 3-Silylacrylonitrile. [e] Products formed in the photolysis of HC3N
by itself.[32]
characterized by 1H NMR spectroscopy. The Z:E ratio (87:13)
was determined by integration of the SiH3 signals at d ˆ 4.01
and 3.92, respectively. The signals of the vinylic protons of the
Z isomer were observed at d ˆ 6.25 and 6.85. Control
reactions established that neither PH3 nor SiH4 added to
propynenitrile under these conditions in the absence of UV
irradiation. Diphosphine (P2H4), a product of PH3 photoly-
sis,[29] was also detected as a product when PH3 was irradiated
in the presence of propynenitrile. Attempts to detect disilane
(Si2H6)[30] when propynenitrile was irradiated in the presence
of SiH4 were unsuccessful.
.
Table 11. 2-Butynenitrile H addition series.
The photochemical b-addition of NH3 to 2-butynenitrile
stands in marked contrast to the a-addition of ammonia to
acrylonitrile, but is consistent with the mechanistic principles
derived from the above studies of the photoaddition reactions
of a,b- unsaturated nitriles (Scheme 3). The reaction is
hν
.
.
NH2
NH3
CN
+
H
.
.
formation of NH2 initiates the chain reaction process leading
.
+
H
H3C C C
H3CCH CCN
to the b-substitution products. Spin-projected open-shell ab
initio calculations performed at a reliable level of theory
which includes the effects of electron correlation (PMP2/6-
31G‡) suggests that the abstraction of a hydrogen atom from
14
.
.
+
NH2
H3CCH CCN
H3CCH CHCN
+ NH3
C CN
.
.
NH2 + H3C
C
1
H3C
C
CCN
NH3 by this radical is nearly thermoneutral (0.3 kcalmol ,
NH2
Figure 1), a value so close to zero that, in the absence of an
unusually large DS8 of reaction, the only conclusion that can
be drawn is that the reaction is highly probable. The same
calculation gives an enthalpy of abstraction of hydrogen from
.
.
+ NH3
+
NH2
H3C
C
H3C
C
CCN
CHCN
NH2
NH2
Scheme 3. Photolysis of ammonia in the presence of 2-butynenitrile.
Chem. Eur. J. 1998, 4, No. 6
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
0947-6539/98/0406-1079 $ 17.50+.25/0
1079

J.-C. Guillemin et al.
FULL PAPER
1
NH3 by the a-cyanoethyl radical of ‡12.2 kcalmol , which
clearly shows that the reaction will probably not occur in the
absence of the same special entropy considerations (Figure 1).
Observation of the b-substituted product of 2-butynenitrile
demonstrates that hydrogen abstraction does indeed take
the reaction proceeds by a radical chain pathway which can be
explained by the initial addition of the radical X to the
.
unsaturated bond to give the most stable radical. The organic
radical formed then abstracts a hydrogen atom from HX to
.
give the addition product and another X to propagate the
.
place with the formation of NH2, which then undergoes b-
reaction. In the addition of NH3 to olefins, the radical formed
was not sufficiently energetic to abstract a hydrogen atom
from NH3. Thus, it was necessary to generate the radicals
required for the addition reaction by photolysis of NH3. If the
inorganic component does not absorb all the light, then the
product mixture will be complicated by the addition of HX to
hydrocarbons formed by photolysis of the starting olefin. If
the inorganic component used does not absorb the UV light
used, the radical chain pathway will still occur if the radicals
formed by irradiation of the hydrocarbon are sufficiently
energetic to abstract a hydrogen from the inorganic gas.
Solar UV was the most abundant energy source on the
primitive Earth[37] and this radiation is still the most prevalent
energy source in our solar system. Consequently, the disequi-
librium species present in the atmospheres of the planets and
their moons in our solar system are formed by solar UV light.
Unsaturated hydrocarbons and unsaturated nitriles are pro-
duced in processes initiated by these photons and the
compounds formed undergo the light-initiated addition
reactions described in this report. Of particular interest is
the a-addition of NH3 to acrylonitrile to form an a-amino-
nitrile, which in turn can be hydrolyzed to the amino acid
alanine. A wide array of aminonitriles, and hence amino acids,
may have been formed by the addition of NH3 to substituted
acrylonitriles in a reducing atmosphere on the primitive
Earth. These same NH3 addition reactions could have
occurred on Titan before all its atmospheric ammonia was
photolyzed to molecular nitrogen and hydrogen. The a-
aminonitriles formed would have precipitated on the cold
surface of this moon. Photochemically generated adducts of
NH3 with the unsaturated hydrocarbons present on Jupiter
are a likely source of the nitrogen-containing organics found
there.[38] The P-containing adducts are probably formed in
much smaller amounts because the partial pressures of PH3
are much lower than those of NH3 on Jupiter.[39] The
structures of these adducts and of those formed with SiH4
follow directly from the regiochemistries of the addition
processes delineated in these investigations.
addition to the ground state 2-butynenitrile (Scheme 3). This
reaction proceeds by the radical chain pathway proposed
previously for the addition of PH3 and SiH4 to acrylonitrile.
SiH4 and PH3 also undergo radical chain addition to
propynenitrile to give the b-substituted acrylonitrile products.
The photochemical addition of silane must be initiated by
the radicals formed by irradiation of propynenitrile,[31]
because SiH4 does not absorb light at l ˆ 185 nm.[14] The
observation of benzene-1,3,5-tricarbonitrile, a known photo-
product of propynenitrile, establishes that propynenitrile is
absorbing the UV light.[32]
The photoaddition mechanisms derived from the present
study provide mechanistic interpretations for previously
reported photochemical addition reactions. It is now clear
that the photochemical b-addition of isopropanol to methyl
propynoate[10] [Eq. (1)] proceeds by the radical chain process
observed in the photochemical addition of ammonia to 2-
butynenitrile. The photochemical addition of PH3 to fluoro or
simple olefins[33] is now understandable on the basis of the
radical chain mechanistic pathway developed in the present
study. When PH3 is principal absorbing species, then the main
reaction products can be explained by the radical chain
mechanism. Addition to propene yields n-propanephosphine
and, in
a lower yield, iso-propanephosphine [Eq. (12),
Scheme 4]. The n-propanephosphine is the product expected
.
.
CH3CHCH2PH2
PH2
H3CCH CH2
+
.
CH3CHCH2PH2 + PH3
.
PH2
CH3CH2CH2PH2
+
Scheme 4. Radical chain mechanism for the photoaddition of phosphine to
propene.
from the radical chain process. The iso-propanephosphine is
probably also formed by this pathway, but the process is less
regioselective for a simple olefin. Irradiation of PH3 in the
presence of 1-propyne yields the (Z)- and (E)-1-propene-
phosphines[34] expected by the radical chain mechanism
[36]
[Eq. (13)]. Similarly, the addition of H2S[35, 36] and GeH4
to propyne yields the corresponding (Z)- and (E)-propene-
thiols and -germanes.
Experimental Section
Caution: Phosphines and silanes are pyrophoric compounds and must be
used with great care in oxygen-free atmospheres.
hν
CH3CH2CH2PH2
+
(CH3)2CH(PH2)
(12)
(13)
PH3 + H3CCH CH2
General: UV spectra were determined on a Cary 219 spectrophotometer;
1H, 31P, and 13C NMR spectra on a Varian XL-200, a Varian Unity 500, and a
Bruker ARX400; GC/MS was performed on a Hewlett Packard 5987A
spectrometer; GC analyses were performed on a Hewlett Packard 5890
with a J and W DB-5 capillary column (30 m  0.2 mm id). Propynenitrile,
2-butynenitrile, and butynedinitrile were prepared by the procedure of
Moureu and Bongrand[40] as modified by Miller and Lemmon.[41] 1-
Cyclohexenecarbonitrile,[42] 2,3-dicyclohexylbutanedinitrile (5),[43] and
2,2,3,3-tetramethylbutanedinitrile (18)[44] were prepared by literature
procedures. The diastereomers of 2,3-dimethylbutanedinitrile (3) and 2,3-
hν
+
H3CCH CHPH2
(H3C)H2PC CH2
H3CC CH
+
PH3
Conclusions
These investigations have determined the reaction pathway
for the photochemical addition of inorganic compounds [HX]
to simple and activated olefins and acetylenes. In most cases,
diethylbutanedinitrile (15) were prepared by the procedure of Whiteley
[45]
and Marianelli;
2-aminopropanenitrile (2), 2-aminobutanenitrile (16),
1080
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
0947-6539/98/0406-1080 $ 17.50+.50/0 Chem. Eur. J. 1998, 4, No. 6

Addition to Nitriles
1074±1082
and 2-amino-2-methylpropanenitrile (17) by the procedure of Bold et al.[8]
Cyclohexanecarbonitrile (19)[46] was prepared by dehydration of the
corresponding amide with P4O10 in sea sand. 3-Phosphinopropanenitrile
(7)[47] and 3-silylpropanenitrile (8)[48] were prepared as previously reported.
Commercial samples of acrylonitrile, crotononitrile, methacrylonitrile, 3-
aminocrotononitrile ((E)-10 and (Z)-10), butanenitrile, and isobutaneni-
trile were purified by distillation or sublimation before use.
(s, NH), 3398 (s, NH), 3020 (vs, NH), 2240 (w, CN), 2205 (m, CN), 1624 (s,
1
ˆ
C C) cm
.
Photochemical addition of PH3 or SiH4 to a,b-unsaturated nitriles: In a
typical experiment, phosphine (40 Torr) was added to propynenitrile
(40 Torr) in a quartz photolysis cell, and the mixture was immediately
irradiated. A pale yellow solid, presumably oligomeric products, formed in
the cell within 5 min. The cell was irradiated for 1 h, then it was evacuated
on a vacuum line and the volatile products were condensed onto a cold
finger (77 K). A solvent (CDCl3) was added at this step. The cold finger was
disconnected from the vacuum line and dry nitrogen was added until the
pressure had risen by 1 atm. The liquid nitrogen was removed and the
products and solvent were collected in a NMR tube fitted to the bottom of
the finger.
Photochemistry: Photolyses were performed in cylindrical quartz cells
(10 cm  2.8 cm). The gas mixtures were prepared on a Hg-free vacuum
line equipped with a Baratron Type 270B signal conditioner and a 370HA-
1000 vacuum gauge. The light sources were a low-pressure mercury lamp
with principal emissions at l ˆ 184.9 and 253.7 nm, a iodine discharge lamp
with principal emission at l ˆ 206.2 nm,[49] and a Philips lamp TUV6W with
principal emission at l ˆ 253.7 nm. The light from the iodine lamp was
filtered through 1 cm of distilled water to remove shorter wavelengths.[50]
Controls were always performed to establish that there were indeed
photochemical reactions: the gas mixtures were allowed to stand at room
temperature in the absence of light for the same period of time that the
corresponding samples were irradiated. The products of photochemical
reactions and controls were analyzed by NMR. In the calculation of the
percent light absorbed by NH3 the absorption coefficients of NH3 were
estimated to be 87 and 47 cm 1 atm 1 at l ˆ 185 and 206 nm, respectively.[51]
3-Phosphinopropanenitrile (7):[47] 31P NMR (CDCl3, 258C): d ˆ 137.4
(ttt, 1J(P,H) ˆ 195.5 Hz, 2J(P,H) ˆ 10.5 Hz, 3J(P,H) ˆ 5.7 Hz); 1H NMR
(CDCl3, 258C, TMS): d ˆ 1.85 (dtm, 2H, 2J(P,H) ˆ 10.5 Hz, 3J(H,H) ˆ
7.2 Hz), 2.56 (tm, 2H, 3J(H,H) ˆ 7.2 Hz, 3J(P,H) ˆ 5.7 Hz), 2.83 (dm, 2H,
1J(P,H) ˆ 195.5 Hz).
3-Silylpropanenitrile (8):[48] 1H NMR (CDCl3, 258C, TMS): d ˆ 1.14 (tq,
2H, 3J(H,H) ˆ 8.0 Hz, 3J(H,H) ˆ 3.8 Hz), 2.45 (t, 2H, 3J(H,H) ˆ 8.0 Hz),
3
3.58 (t, 3H, J(H,H) ˆ 3.8 Hz).
The absorption coefficient of PH3 at l ˆ 185 nm were estimated to be
3-Phosphinoacrylonitrile (12) (Z:E ˆ 85:15): Z isomer: 31P NMR (CDCl3,
258C): d ˆ 136.0 (tdd, 1J(P,H) ˆ 215 Hz, 2J(P,H) ˆ 19.2 Hz, 3J(P,H) ˆ
10.6 Hz); 1H NMR (CDCl3, 258C, TMS): d ˆ 3.85 (ddd, 2H, 1J(P,H) ˆ
215 Hz, 3J(H,H) ˆ 7.2 Hz, 4J(H,H) ˆ 1.7 Hz; PH2), 5.87 (ddt, 1H,
3J(H,H)cis ˆ 12.6 Hz, 3J(P,H) ˆ 10.6 Hz, 4J(H,H) ˆ 1.7 Hz; CHCN), 7.04
1 [16]
100 cm 1 atm
.
The absorption coefficient of propynenitrile at l ˆ 185
and 206 nm were estimated to be 15 and 16,[32] respectively.
Photochemical addition of NH3 to a,b-unsaturated nitriles: In a typical
experiment, ammonia (50 Torr) was added to propynenitrile (50 Torr) in a
quartz photolysis cell and the mixture was immediately irradiated. A brown
solid, presumably oligomeric products, formed in the cell within 5 min.
After irradiation for 1 h, the cell was quickly rinsed with either CH2Cl2 or
CDCl3 and immediately analyzed. The polymeric product was insoluble in
these solvents. The spectral analysis of the extract proved the presence of 3-
amino-2-propenenitrile (9).[23] UV (CH2Cl2): lmax ˆ 247 nm; Z isomer: 1H
NMR (CDCl3, 258C, TMS): d ˆ 3.97 (1H, dt, 3J(H,H)cis ˆ 8.3 Hz,
2
3
3
(ddt, 1H, J(P,H) ˆ 19.2 Hz, J(H,H)cis ˆ 12.6 Hz, J(H,H) ˆ 7.2 Hz; CHP).
E isomer: 31P NMR (CDCl3, 258C): d ˆ 132.8 (tdd, 1J(P,H) ˆ 209 Hz,
2J(P,H) ˆ 17.9 Hz, 3J(P,H) ˆ 5.6 Hz); 1H NMR (CDCl3, 258C, TMS): d ˆ
3.70 (ddd, 2H, 1J(P,H) ˆ 209 Hz, 3J(H,H) ˆ 6.7 Hz, 4J(H,H) ˆ 1.9 Hz; PH2),
5.81 (ddt, 1H, 3J(H,H)Htrans ˆ 17.9 Hz, 3J(P,H) ˆ 5.6 Hz, 4J(H,H) ˆ 1.9 Hz;
CHCN), 7.18 (ddt, 1H, 2J(P,H) ˆ 3J(H,H)trans ˆ 17.9 Hz, 3J(H,H) ˆ 6.7 Hz;
CHP); HRMS: calcd. for C3H4NP: 85.00814, found: 85.0079; MS: m/z
4J(H,H) ˆ 0.8 Hz, H2), 4.80 (2H, broad, NH2), 6.77 (1H, dt, 3J(H,H)cis
ˆ
‡
‡
‡
(%) ˆ 86 (3.7) [M ], 85 (86.7) [M
H], 83 (22.8) [M
3H], 82 (24.2)
3
8.3 Hz, J(H,H) ˆ 10.6 Hz, H3); 13C NMR (CDCl3, 258C, TMS): d ˆ 62.5
‡
‡
‡
‡
[M
4H], 59 (19.6) [M
CN], 58 (61.9) [M
HCN], 57 (100) [M
2
(ddt, 1J(CH) ˆ 178.9 Hz, 2J(CH) ꢀ J(CH) ˆ 6.2 Hz), 118.8 (d, 2J(CH) ˆ
H2CN].
13.8 Hz), 150.3 (d, 1J(CH) ˆ 169.4 Hz); E isomer: 1H NMR (CDCl3,
258C, TMS) d ˆ 4.29 (1H, dt, 3J(H,H)trans ˆ 13.9 Hz, 4J(H,H) <0.4 Hz, H2),
4.55 (2H, broad, NH2), 7.00 (1H, dt, 3J(H,H)trans ˆ 13.9 Hz, 3J(H,H) ˆ
10.7 Hz, H3); 13C NMR (CDCl3, 258C, TMS): d ˆ 64.9 (dt, 1J(CH) ˆ
168.6 Hz, 2J(CH) ˆ 6.8 Hz), 121.8 (d, 2J(CH) ˆ 5.4 Hz), 151.6 (ddt,
1
3-Silylacrylonitrile (13) (Z:E ˆ 83:17): Z isomer: H NMR (CDCl3, 258C,
TMS): d ˆ 4.01 (d, 3H, 3J(H,H) ˆ 3.5 Hz; SiH3), 6.25 (d, 1H, 3J(H,H)cis
ˆ
15.0 Hz; CHCN), 6.85 (dq, 1H, 3J(H,H)cis ˆ 15.0 Hz, 3J(H,H) ˆ 3.5 Hz;
CHSi). E isomer: 1H NMR (CDCl3, 258C, TMS): d ˆ 3.92 (dd, 3H,
2
3
1J(CH) ˆ 166.7 Hz, J(CH) ꢀ J(CH) ˆ 4.2 Hz); MS (70 eV): m/z (%) ˆ 69
3J(H,H) ˆ 3.3 Hz, 4J(H,H) ˆ 0.6 Hz; SiH3), 6.03 (dd, 1H, 3J(H,H)trans
ˆ
‡
‡
‡
‡
19.5 Hz, 4J(H,H) ˆ 0.6 Hz; CHCN), 7.00 (dq, 1H, 3J(H,H)Htrans ˆ 19.5 Hz,
3J(H,H) ˆ 3.3 Hz, CHSi). This data is in good agreement with that reported
for the trimethylsilyl derivatives.[53]
(3) [M ], 68 (100) [M
[M
H], 67 (47) [M
2H], 66 (23) [M
3H], 52 (17)
‡
‡
NH2 H], 41 (83) [M
HCN]. The product yield was determined
by a comparison of the area of the 1H NMR spectrum with that of an
internal standard of CHCl3. It was not possible to detect the presence of 2-
amino-2-propenenitrile or 2-aziridinecarbonitrile.[8]
Computational methods: All ab initio calculations were performed on a
Silicon Graphics Indigo II workstation and Gaussian 94, RevB. The
computational data reported in Tables 6, 10, and 11 and Figure 1 are
presented in the accepted format of the Theoretical Model/Basis Set. In
cases where different levels of theory were used in the geometry
optimization and energy determination portions of the calculation, the
Model/Basis reported first refers to the energy calculation and the second
Model/Basis refers to the theoretical model used in the geometry
optimization. When only one Model/Basis data designation is given, both
the geometry optimization and energy calculations were performed with
the same method. In most cases, multiple theoretical models were used and
their results reported to facilitate comparisons between the methods. Bold
results indicate the lowest energy isomers of a given series. Since the
calculations were performed on open-shell doublet species, several
methods of handling the unpaired electron were tested. ROHF optimized
energies were compared to UHF energies with the ROHF geometries. The
full UHF geometry-optimized results are presented with S2 values before
and after spin annihilation (S2, S2A). Where electron correlation was
expected to contribute significantly to an energy comparison result,
projected MP2 open-shell calculations were used at the 6-31G‡ level of
theory.
The slow thermal addition of NH3 to propynenitrile is complete within 12 h
to form a mixture of (E,Z)-3-amino-2-propenenitrile (9). The rate of this
addition appears to increase with time and a yield of <0.5% was observed
after 1 h.
The same procedure was used in the investigation of the photochemical
addition of ammonia to the other unsaturated nitriles used in this study,
with the exception that a photolysis time of 3 h was used for the olefinic
nitriles. The photolyses of acrylonitrile alone yielded a polymeric material
and <0.5% of HCN and propynenitrile. The vapor pressure of 1-
cyclohexenenitrile is 0.7 Torr at room temperature, therefore, it was only
possible to study the reaction of 0.7 Torr of this unsaturated nitrile with
excess (440 Torr) NH3. The addition of ammonia to 2-butynenitrile gave
the E and Z isomers of 3-aminocrotononitrile (10).[52]
2-Amino-2-butenedinitrile (11):[27] Butynedinitrile (50 Torr) and ammonia
(50 Torr) were introduced into a photolysis cell; a brown solid formed in ꢀ
1 min. The walls of the cell were then washed with either CH2Cl2 or CDCl3.
Evaporation of the solvent in vacuo led to the crude product (Yield: 21%).
The two isomers were obtained in a 8:1 ratio. Major isomer: 1H NMR
(CDCl3, 258C, TMS): d ˆ 4.50 (s, 1H), 6.70 (brd, 2H); 13C NMR (CDCl3,
258C, TMS): d ˆ 74.3 (d, 1J(CH) ˆ 184.1 Hz), 114.1 (d, 2J(CH) ˆ 6.3 Hz),
115.7, 133.4; Minor isomer: 1H NMR (CDCl3, 258C, TMS):
d ˆ 4.75 (s, 1H), 6.70 (brd, 2H); 13C NMR (CDCl3, 258C, TMS): d ˆ 77.6
(d, 1J(CH) ˆ 180 Hz), 112.8, 116.8, 133.8; IR (CHCl3) (E ‡ Z) : nÄ ˆ 3501
Acknowledgments: We thank Dr. Yoji Ishikawa and Dr. R. R. Jacobson
for performing some of the initial photolyses of ammonia and propyneni-
trile. Financial support was provided by NASA grant NAG5-4557. J.-C.G.
Chem. Eur. J. 1998, 4, No. 6
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
0947-6539/98/0406-1081 $ 17.50+.25/0
1081

J.-C. Guillemin et al.
FULL PAPER
Â
thanks the Programme National de Planetologie (INSU-CNRS, France) for
financial support.
[25] P. Bruston, H. Poncet, F. Raulin, C. Cossart-Magos, R. Courtin, Icarus
1989, 78, 38 ± 53.
[26] B. A. Thompson, P. Harteck, R. R. Reeves, Jr., J. Geophys. Res. 1963,
68, 6431 ± 6436.
Received: September 24, 1997 [F833]
[27] H. Mommaerts, Bull. Acad. Sci. R. Belg. 1941, 27, 579 ± 597.
[28] W. A. Noyes, P. A. Leighton, The Photochemistry of Gases, Reinhold,
New York, 1941, 372 ± 373.
[29] J. P. Ferris, R. Benson, J. Am. Chem. Soc. 1981, 103, 1922 ± 1927; J. P.
Ferris, A. Bossard, H. Khwaga, ibid. 1984, 106, 318 ± 324.
[30] G. G. A. Perkins, E. R. Austin, F. W. Lampe, J. Am. Chem. Soc. 1979,
101, 1109 ± 1115.
[1] R. A. Sanchez, J. P. Ferris, L. E. Orgel, Science 1966, 154, 784 ± 785.
[2] J. P. Ferris, R. A. Sanchez, L. E. Orgel, J. Mol. Biol. 1968, 33, 693 ± 704.
[3] V. G. Kunde, A. C. Aikin, R. A. Hanel, D. E. Jennings, W. C. Maguire,
R. E. Samuelson, Nature 1981, 292, 686 ± 688.
[31] D. W. Clarke, J. P. Ferris, Icarus 1995, 115, 119 ± 125.
[32] J. P. Ferris, J.-C. Guillemin, J. Org. Chem. 1990, 55, 5601 ± 5606.
[33] A. R. Stiles, F. F. Rust, W. E. Vaughan, J. Am. Chem. Soc. 1952, 74,
3282 ± 3284; R. L. Whistler, C.-C. Wang, S. Inokawa, J. Org. Chem.
1968, 33, 2495 ± 2497; R. Fields, R. N. Haszeldine, N. F. Wood, J. Chem.
Soc. (C) 1970, 1370 ± 1375; J.-C. Guillemin, L. Lassalle, S. Le Serre,
Adv. Space Res. 1997, 19, 1093 ± 1102.
[34] J.-C. Guillemin, T. Janati, L. Lassalle, Adv. Space Res. 1995, 16, 85 ± 92.
[35] O. P. Strausz, T. Hikida, H. E. Gunning, Can. J. Chem. 1965, 43, 717 ±
721.
Â
[4] A. Coustenis, T. Encrenaz, B. Bezard, B. Bjoraker, G. Graner, G.
Â
Dang-Nhu, E. Arie, Icarus 1993, 102, 240 ± 260.
[5] A. P. C. Mann, D. A. Williams, Nature 1980, 283, 721 ± 725.
[6] B. E. Turner, J. Mol. Evol. 1980, 15, 79 ± 101.
[7] J. P. Ferris, O. S. Zamek, A. M. Altbuch, H. J. Freiman, J. Mol. Evol.
1974, 3, 301 ± 309.
[8] G. Ksander, G. Bold, R. Lattmann, C. Lehmann, T. Früh, Y.-B. Xiang,
K. Inomata, H.-P. Buser, J. Schreiber, E. Zass, A. Eschenmoser, Helv.
Chim. Acta 1987, 70, 1115 ± 1172; G. H. Bold, Ph.D. Thesis, ETH
Zürich, 1984.
[9] J. P. Ferris, L. E. Orgel, J. Am. Chem. Soc. 1966, 88, 1074.
[10] G. Büchi, S. H. Feairheller, J. Org. Chem. 1969, 34, 609 ± 612.
[11] J. J. McCullough, C. W. Huang, Can. J. Chem. 1969, 47, 757 ± 766.
[12] E. V. Khoroshilova, N. P. Kuzꢁmino, Y. A. Matveets, Laser Chem.
1988, 8, 13 ± 18.
[36] J.-C. Guillemin, L. Lassalle, T. Janati, Planet. Space Sci. 1995, 43, 75 ±
82.
[37] S. L.Miller, Cold Spring Harbor Symposium on Quantitative Biology
17 ± 27, Cold Spring Harbor Press, Cold Spring Harbor, NY, 1987.
[38] T. C. Keane, F. Yuan, J. P. Ferris, Icarus 1996, 112, 205 ± 207.
[39] S. K. Atreya, Atmospheres and Ionospheres of the Outer Planets and
Their Satellites, Springer, Berlin, 1986.
[40] C. Moureu, J. C. Bongrand, Ann. Chim. (Paris) 1920, 14, 47 ± 58.
[41] F. A. Miller, D. C. Lemmon, Spectrochim. Acta A 1967, 23, 1415 ± 1423.
[42] M. Oda, A. Yamamuro, T. Watabe, Chem. Lett. 1979, 1427 ± 1430.
[43] C. G. Overberger, M. B. Berenbaum, Org. Synth. Collective Vol. IV
1966, 273 ± 274.
[44] R. Scheffold, J. Löliger, H. Blaser, P. Geisser, Helv. Chim. Acta 1975,
58, 49 ± 64.
[45] R. V. Whiteley, Jr., R. S. Marianelli, Synthesis 1978, 5, 392 ± 394.
[46] K. Mai, G. Patil, Tetrahedron Lett. 1986, 27, 2203 ± 2206.
[47] M. M. Rauhut, I. Hechenbleikner, H. A. Currier, F. C. Schaefer, V. P.
Wystrach, J. Am. Chem. Soc. 1959, 81, 1103 ± 1107.
[48] O. W. Steward, O. R. Pierce, J. Org. Chem. 1961, 26, 2943 ± 2947.
[49] P. Harteck, R. R. Reeves, Jr., B. A. Thompson, Z. Naturforsch. 1964,
19A, 2 ± 6.
[50] W. E. Groth, U. Schurath, R. N. Schindler, J. Phys. Chem. 1968, 72,
3914 ± 3920.
[51] J. P. Ferris, Y. Ishikawa, J. Am. Chem. Soc. 1988, 110, 4306 ± 4312.
[52] E. Bullock, B. Gregory, Can. J. Chem. 1965, 43, 332 ± 336.
[53] E. Lukevics, V. Dirnens, A. Kemme, J. Popelis, J. Organomet. Chem.
1996, 521, 235 ± 244.
[13] D. M. Gale, J. Org. Chem. 1970, 35, 970 ± 972.
[14] Y. Harada, J. N. Murrell, H. H. Sheena, Chem. Phys. Lett. 1968, 1,
595 ± 596; U. Itoh, Y. Toyoshima, H. Onuki, N. Washida, T. Ibuki, J.
Chem. Phys. 1986, 85, 4867 ± 4872
[15] P. A. Mullen, M. K. Orloff, Theoret. Chim. Acta 1971, 23, 278 ± 284.
[16] D. Kley, K. H. Welge, Z. Naturforsch. 1965, 20A, 124 ± 131; G.
Di Stephano, M. Lenzi, A. Margani, A. Mele, C. Nguyen Xuan,
Photochem. 1977, 7, 335 ± 344.
[17] W. A. Payne, L. J. Stief, J. Chem. Phys. 1976, 64, 1150 ± 1155; S. R.
Bosco, D. F. Nava, W. F. Brobst, L. J. Stief, ibid. 1984, 81, 3505 ± 3511.
[18] R. Walsh, Acc. Chem. Res. 1981, 14, 246 ± 252; D. F. McMillen, D. M.
Golden, Ann. Rev. Phys. Chem. 1982, 33, 493 ± 532.
[19] A. S. Manocha, D. W. Setzer, M. A. Wickramaaratchi, Chem. Phys.
1983, 76, 129 ± 146.
[20] M. M. Rauhut, H. A. Currier, A. M. Semsel, V. P. Wystrach, J. Org.
Chem. 1961, 26, 5138 ± 5145.
[21] D. M. Schubert, P. F. Brandt, A. D. Norman, Inorg. Chem. 1996, 35,
6204 ± 6209.
[22] J. P. Ferris, R. Benson, J. Am. Chem. Soc. 1981, 103, 1922 ± 1927.
[23] H. U. Sieveking, W. Lüttke, Angew. Chem. 1969, 81, 432 ± 433; Angew.
Chem. Int. Ed. Engl. 1969, 8, 458 ± 459.
[24] D. W. Clarke, J. P. Ferris, J. Geophys. Res. 1996, E3, 101, 7575 ± 7584.
1082
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
0947-6539/98/0406-1082 $ 17.50+.50/0
Chem. Eur. J. 1998, 4, No. 6