C-C bond cleavage of acetonitrile by a carbonyl iron complex with a silyl ligand

Article abstract of DOI:10.1021/om0208319

The photoreaction of a silyl iron complex Cp(CO)₂Fe(SiMe₃) (1) in acetonitrile in the presence of P(NMeCH₂)₂(OMe) (L) yielded Cp(CO)LFeMe (2), CpL₂FeMe (3), and CpL₂Fe(CN) (4), showing that carbon-carbon bond cleavage of acetonitrile was achieved. These C-C bond cleavage products were also obtained in the photoreaction of 1 with 1 equiv of MeCN in THF in the presence of L. The reaction with CD₃CN showed that the methyl group on the iron in the products is derived from acetonitrile. The corresponding reaction of Cp(CO)₂Fe(ER₃) (ER₃ = CH₃, GeMe₃, SnMe₃) generated a CO/L exchange complex, Cp(CO)LFe(ER₃), showing that a silyl ligand on the iron is indispensable for the C-C bond cleavage of acetonitrile. Theoretical studies on the C-C bond cleavage were performed using the hybrid DFT-B3LYP method. The direct C-C bond oxidative addition of acetonitrile to the 16e species Cp(CO)Fe(SiMe₃) expected to readily form from 1 in the photoreaction conditions has a very high activation barrier of 52.7 kcal/mol, suggesting that the oxidative addition is not an appropriate reaction pathway. A more feasible pathway was proposed. The end-on coordination of acetonitrile nitrogen to Cp(CO)Fe(SiMe₃), followed by the rearrangement to a CN side-on complex, with the activation energy of 14.8 kcal/mol occurs, and then the insertion of the CN bond into the Fe-Si bond with a small activation energy of 4.0 kcal/mol and the successive C-C bond cleavage of acetonitrile on the Fe coordination sphere with the activation energy of 15.0 kcal/mol take place to give Cp(CO)MeFe(CNSiMe₃). The isolation of an iron complex with a methyl group derived from acetonitrile and a silylisocyanide ligand was attained in the photoreaction of Cp(CO)₂Fe(SiPh₃) in MeCN in the presence of PPh₃. The product Cp (PPh₃)MeFe(CNSiPh₃) was confirmed by the X-ray structure analysis. The reaction mechanism leading to the iron cyanide complex has also been discussed.

Full text of DOI:10.1021/om0208319

Organometallics 2004, 23, 117-126
117
C-C Bon d Clea va ge of Aceton itr ile by a Ca r bon yl Ir on
Com p lex w ith a Silyl Liga n d
Hiroshi Nakazawa,*,† Takafumi Kawasaki,‡ Katsuhiko Miyoshi,‡
Cherumuttathu H. Suresh,§ and Nobuaki Koga§
Department of Chemistry, Graduate School of Science, Osaka City University, Sugimoto,
Sumiyoshi-ku, Osaka 558-8585, J apan, Department of Chemistry, Graduate School of Science,
Hiroshima University, Higashi-Hiroshima 739-8526, J apan, and Graduate School of
Information Science and Venture Business Laboratory, Nagoya University, Chikusa-ku,
Nagoya 464-8601, J apan
Received October 4, 2002
The photoreaction of a silyl iron complex Cp(CO)2Fe(SiMe3) (1) in acetonitrile in the
presence of P(NMeCH2)2(OMe) (L) yielded Cp(CO)LFeMe (2), CpL2FeMe (3), and CpL2Fe-
(CN) (4), showing that carbon-carbon bond cleavage of acetonitrile was achieved. These
C-C bond cleavage products were also obtained in the photoreaction of 1 with 1 equiv of
MeCN in THF in the presence of L. The reaction with CD3CN showed that the methyl group
on the iron in the products is derived from acetonitrile. The corresponding reaction of
Cp(CO)2Fe(ER3) (ER3 ) CH3, GeMe3, SnMe3) generated a CO/L exchange complex, Cp(CO)-
LFe(ER3), showing that a silyl ligand on the iron is indispensable for the C-C bond cleavage
of acetonitrile. Theoretical studies on the C-C bond cleavage were performed using the hybrid
DFT-B3LYP method. The direct C-C bond oxidative addition of acetonitrile to the 16e species
Cp(CO)Fe(SiMe3) expected to readily form from 1 in the photoreaction conditions has a very
high activation barrier of 52.7 kcal/mol, suggesting that the oxidative addition is not an
appropriate reaction pathway. A more feasible pathway was proposed. The end-on coordina-
tion of acetonitrile nitrogen to Cp(CO)Fe(SiMe3), followed by the rearrangement to a CN
side-on complex, with the activation energy of 14.8 kcal/mol occurs, and then the insertion
of the CN bond into the Fe-Si bond with a small activation energy of 4.0 kcal/mol and the
successive C-C bond cleavage of acetonitrile on the Fe coordination sphere with the activation
energy of 15.0 kcal/mol take place to give Cp(CO)MeFe(CNSiMe3). The isolation of an iron
complex with a methyl group derived from acetonitrile and a silylisocyanide ligand was
attained in the photoreaction of Cp(CO)2Fe(SiPh3) in MeCN in the presence of PPh3. The
product Cp (PPh3)MeFe(CNSiPh3) was confirmed by the X-ray structure analysis. The
reaction mechanism leading to the iron cyanide complex has also been discussed.
In tr od u ction
U11 complex. The nitrile activation has been attained
by the direct oxidative addition of the C-C bond toward
a transition metal center, though in many cases the
mechanism has not been elucidated.
Since nitriles (R-CtN) have relatively strong R-C
and CtN bonds and have lone pair electrons on the
nitrogen atom, they have been widely used as ligands
in transition metal complexes, from which they can be
readily released to afford reactive and coordinatively
unsaturated transition metal species. In other words,
nitriles are generally resistant to the C-C bond cleav-
age by transition metal complexes. There are some
reports exhibiting C-C bond cleavage of nitriles by
transition metal complexes. They involve mainly group
10 transition metal triads1-8 and one Mo,9 Cu,10 and
During the course of our investigation on iron phos-
phenium complexes,12 we encountered C-C bond cleav-
age of acetonitrile promoted by an iron complex bearing
a silyl ligand. Our further investigation concentrating
on this C-C bond activation reaction revealed that a
silyl ligand on an iron is indispensable for nitrile C-C
bond cleavage, and the bond cleavage is induced by silyl
migration from iron to the nitrogen in the coordinating
nitrile. We here report our experimental results and
* Corresponding author. E-mail: nakazawa@sci.osaka-cu.ac.jp.
† Osaka City University.
(7) (a) Garcia, J . J .; J ones, W. D. Organometallics 2000, 19, 5544.
(b) Garcia, J . J .; Brunkan, N. M.; J ones, W. D. J . Am. Chem. Soc. 2002,
124, 9547.
(8) Miller, J . A. Tetrahedron Lett. 2001, 42, 6991.
(9) Churchill, D.; Shin, J . H.; Hascall, T.; Hahn, J . M.; Bridgewater,
B. M.; Parkin, G. Organometallics 1999, 18, 2403.
(10) Marlin, D. S.; Olmstead, M. M.; Mascharak, P. K. Angew.
Chem., Int. Ed. 2001, 42, 4752.
(11) Adam, R.; Villiers, C.; Ephritikhine, M.; Lance, M.; Nierlich,
M.; Vigner, J . J . Organomet. Chem. 1993, 445, 99.
(12) Nakazawa, H. J . Organomet. Chem. 2000, 611, 349, and
references therein.
‡ Hiroshima University.
§ Nagoya University.
(1) Burmeister, J . L.; Edwards, L. M. J . Chem. Soc. (A) 1971, 1663.
(2) Gerlach, D. H.; Kane, A. R.; Parshall, G. W.; J esson, J . P.;
Muetterties, E. L. J . Am. Chem. Soc. 1971, 93, 3543.
(3) Parshall, G. W. J . Am. Chem. Soc. 1974, 96, 2360.
(4) Morvillo, A.; Truco, A. J . Organomet. Chem. 1981, 208, 103.
(5) Favero, G.; Morvillo, A.; Turco, A. J . Organomet. Chem. 1983,
241, 251.
(6) Abla, M.; Yamamoto, T. J . Organomet. Chem. 1997, 532, 267.
10.1021/om0208319 CCC: $27.50 © 2004 American Chemical Society
Publication on Web 12/02/2003

118 Organometallics, Vol. 23, No. 1, 2004
Nakazawa et al.
theoretical considerations as to acetonitrile C-C bond
cleavage by an iron complex. During the preparation of
this article, a quite similar paper was published by
Bergman, Brookhart, and their co-workers for a rho-
dium complex.13
derived from 1. To collect more information on the
unexpected reaction, we examined the following reac-
tions. (i) The same reaction was performed by using
commonly used phosphine, that is PMe3 and PPh3, in
place of the diamino-substituted phosphite. In both
cases, the corresponding products Cp(CO)(phosphine)-
FeMe and Cp(phosphine)2Fe(CN) were formed, whereas
Cp(phosphine)2FeMe was not obtained. (ii) The reaction
shown in eq 1 but in the absence of a phosphorus
compound was examined, and it was found that the
reaction was messy and intractable oily products were
formed. The phosphorus compound added to the reaction
system seems to help the isolation of methyl- and cyano-
iron complexes derived from acetonitrile C-C bond
cleavage (see theoretical considerations described be-
low). (iii) To examine whether acetonitrile has to be used
as a solvent, reaction of 1 with an equimolar amount of
MeCN in the presence of L was conducted in THF. The
results are shown in eq 2. The three iron complexes (2,
3, and 4) were formed as in eq 1, which shows that the
C-C bond cleavage reaction occurred. In this case,
Cp(CO)LFe(SiMe3),14 corresponding to a CO/L exchange
product of 1, was also formed. Although the respective
yields in eq 2 were slightly reduced, 65% of complexes
derived from 1 were identified. (iv) When CD3CN was
Resu lts a n d Discu ssion
(a ) Exp er im en ta l Resu lts. Iron silyl complex 1 was
dissolved in acetonitrile and subjected to a photoreaction
for 4 h in the presence of 2 equiv of diamino-substituted
phosphite L (L stands for P(NMeCH2)2(OMe) in this
paper). The reaction yielded mainly three kinds of
complexes (eq 1). The 31P NMR spectrum of the reaction
mixture showed three dominant singlets around 180
ppm in addition to a singlet at 124.51 ppm, which is
attributed to free L.
Two of the three iron complexes (2 and 4) could be
isolated by column chromatography. The reaction mix-
ture was loaded on an alumina column and eluted with
CH2Cl2 to obtain a yellowish orange complex (2) and
then eluted with acetone to obtain a yellow complex (4).
Complex 2 was identified with Cp(CO)LFeMe, which
had already been reported.14 Complex 4 exhibits an
intense IR absorption at 2063 cm-1 in CH2Cl2, indicative
of a terminal CN ligand. The 13C NMR spectrum, which
contains a triplet at 153.6 ppm (J CP ) 42 Hz), also
supports the existence of a CN ligand. The spectroscopic
data and the single-crystal X-ray diffraction study (see
below) clearly show that 4 is CpFe(CN)L2. Considerable
difficulty was encountered in our attempts to isolate 3
as a pure solid due to the high reactivity. Formation of
1
3 was evidenced by the H, 13C, and 31P NMR measure-
ments of the reaction mixture. The yields for the three
complexes based on 1 were 15% (isolation) and 26%
(from NMR) for 2, 35% (from NMR) for 3, and 25%
(isolation) and 32% (from NMR) for 4. Therefore, on the
basis of the NMR data, 93% of complexes derived from
1 were identified after the reaction. Similar results were
obtained when Cp(CO)2Fe(SiEt3) and Cp(CO)2Fe(SiPh3)
in place of Cp(CO)2Fe(SiMe3) were used as a starting
iron complex.
These three iron-containing complexes have no silyl
groups. To elucidate the fate of the SiMe3 group, we
examined the organic products in the reaction mixture
and found the formation of Me3SiCN and Me4Si in 66%
and 27% yields, respectively, based on 1 in eq 1.
The reaction in eq 1 obviously shows that the C-C
bond of acetonitrile is cleaved by some iron complex
used in the reaction mentioned in (iii), Cp(CO)LFe(CD3)
was obtained. This confirms that the methyl group in 2
is derived from acetonitrile but not from, for example,
a SiMe3 group. It is likely that CpL2Fe(CD3) was also
formed, though clear spectroscopic data could not be
obtained. (v) To elucidate the role of the silyl ligand on
the iron, the reactions of the corresponding alkyl,
germyl, and stannyl complexes were examined (eq 3).
In all cases, CO/L exchange reaction takes place to give
Cp(CO)LFe(ER3).15 Since the 31P NMR spectra of the
reaction mixtures showed that the yields of 8, 9, and
10 were more than 80%, the CO/L exchange reaction
proceeds dominantly in all cases. The isolation yields
were, however, 39% for 8, 36% for 9, and 38% for 10
presumably due to the loss of the product during the
purification processes. From these results, it can be said
(13) Taw, F. L.; White, P. S.; Bergman, R. G.; Brookhart, M. J . Am.
Chem. Soc. 2002, 124, 4192.
(14) Nakazawa, H.; Yamaguchi, Y.; Mizuta, T.; Ichimura, S.;
Miyoshi, K. Organometallics 1995, 14, 4635.
(15) (a) Nakazawa, H.; Yamaguchi, Y.; Miyoshi, K. Organometallics
1996, 15, 1337. (b) Nakazawa, H.; Yamaguchi, Y.; Kawamura, K.;
Miyoshi, K. Organometallics 1997, 16, 4626.

C-C Bond Cleavage of Acetonitrile
Organometallics, Vol. 23, No. 1, 2004 119
Ta ble 1. Su m m a r y of Cr ysta l Da ta for 4 a n d 30
4
30
formula
fw
C
16H31Fe N5O2P2
C
43H38NPSiFe
443.25
683.69
color, habit
cryst dimens, mm
cryst syst
unit cell dimens
a, Å
brown, cubic
0.25 × 0.25 × 0.15
monoclinic
red, cubic
0.20 × 0.18 × 0.07
monoclinic
9.1740(2)
17.7330(6)
12.5210(4)
92.412(2)
2035.1 (4)
Cc(#9)
4
17.643(4)
8.135(1)
25.519(5)
106.006(7)
3520(1)
Cc(#9)
4
b, Å
c, Å
â, deg
V, Å3
space group
Z
D
calcd, g cm-3
1.447
1.290
F igu r e 1. ORTEP drawing of 4 showing the atom-
numbering scheme. The thermal ellipsoids are drawn at
the 50% probability level. Hydrogen atoms are omitted for
clarity.
F(000)
936.00
9.18
55.0
1432.00
5.39
55.0
µ, cm-1
2θmax, deg
no. of reflns
measd
obsd(I > 3σ(I))
obsd(I > 2σ(I))
struct soln
2471
2354
13631
5567
direct method
(SIR92)
360
0.027
0.043
direct method
(SIR92)
424
0.067
0.074
no. of params
Ra
Rw
b
a
b
2
2
R ) ∑||Fo| - |Fo||/|∑|Fo|. Rw ) [∑w||Fo| - |Fc|| /∑w|Fo| ]1/2 and
2
2 -1
w ) 1/σ2(Fo) ) [σc (Fo) + (p2/4)Fσ
] .
Ta ble 2. Selected Bon d Dista n ces (Å) a n d An gles
(d eg) for 4 a n d 30
that a silyl ligand on the iron is indispensable for the
C-C bond cleavage.
4
30
Fe1-P1
Fe1-P2
Fe1-C1
Fe1-C2
Fe1-C3
Fe1-C4
Fe1-C5
Fe1-C6
P1-O1
P1-N2
P1-N3
P2-O2
P2-N4
P2-N5
C6-N1
2.1848(7)
2.1558(7)
2.100(3)
2.087(3)
2,089(3)
2.086(3)
2.086(3)
1.907(3)
1.635(2)
1.678(2)
1.682(2)
1.623(2)
1.686(3)
1.675(2)
1.166(4)
Fe1-P1
Fe1-C1
Fe1-C2
Fe1-C3
Fe1-C4
Fe1-C5
Fe1-C6
Fe1-C7
P1-C8
P1-C14
P1-C20
C1-N1
N1-Si1
Si1-C26
Si1-C32
Si1-C38
2.182(2)
1.760(5)
2.045(5)
2.110(6)
2.084(7)
2.087(6)
2.094(6)
2.129(6)
1.847(6)
1.835(5)
1.846(5)
1.181(7)
1.701(6)
1.857(6)
1.862(6)
1.862(6)
The crystal structure of 4 was determined by X-ray
analysis. An ORTEP drawing is displayed in Figure 1.
Crystallographic data and selected bond distances and
angles are summarized in Tables 1 and 2.
Complex 4 has a piano-stool configuration: the iron
has a cyclopentadienyl ligand bonded in an η5 fashion,
a terminal CN ligand, and two diamino-substituted
phosphite ligands. The Fe1-C6 and C6-N1 bond
distances (1.907 and 1.166 Å) and the Fe1-C6-N1
angle (172.7°) are similar to those for cis-[Fe(phen)2-
(CN)2]‚3H2O, which is also an Fe(II) complex with
terminal cyanide ligands.16 The Fe-P bond distances
(2.1848 and 2.1558 Å) are also similar to those for Cp-
(CO){P(NMeCH2)2(OEt)}Fe{CH2PPh3} (2.164 Å)14 and
Cp(CO){P(NMeCH2}2(OEt)}Fe{SiMe2PPh3} (2.1548 Å),17
which are Fe(II) complexes with a diamino-substituted
phosphite ligand.
(b) Th eor etica l Resu lts. It is expected that at first
the irradiation of the coordinately saturated 1 will lead
to the dissociation of one of the carbonyl groups and
thereby the formation of a vacant site on the Fe atom
of the resulting compound Cp(CO)Fe(SiMe3), 11. We
investigated several reaction paths from 11 for the C-C
bond cleavage of acetonitrile with the DFT calculations
(see the Experimental Section), as shown below, to find
a plausible reaction path.
P1-Fe1-P2
P1-Fe1-C6
P2-Fe1-C6
Fe1-P1-O1
Fe1-P1-N2
Fe1-P1-N3
Fe1-P2-O2
Fe1-P2-N4
Fe1-P2-N5
Fe1-C6-N1
95.64(3)
91.41(8)
91.28(8)
107.34(8)
126.91(8)
118.34(9)
111.31(8)
116.6(1)
122.49(9)
172.7(3)
P1-Fe1-C1
P1-Fe1-C2
C1-Fe1-C2
Fe1-P1-C8
Fe1-P1-C14
Fe1-P1-C20
Fe1-C1-N1
C1-N1-Si1
N1-Si1-C26
N1-Si1-C32
N1-Si1-C38
94.5(2)
92.8(2)
90.9(3)
113.1(2)
116.1(2)
119.3(2)
173.5(5)
156.7(5)
108.1(3)
109.4(3)
107.4(3)
the interacting C-H bond. The C-Si bond of 1.938 Å is
also longer, suggesting the agostic interaction of the
C-Si bond with the iron atom.18e
(i) P a th I: Oxid a tive Ad d ition of C-C Bon d . The
formation of the products 2, 3, and 4 in eq 1 could be
The optimized structure of 11 is shown in Figure 2.
One of the methyl groups of the trimethylsilyl ligand of
11 is inclined toward the vacant site on the Fe atom.
This is caused by an agostic interaction18 between the
Fe atom and the methyl C-H bond, which elongates
(18) (a) Brookhart, M.; Green, M. L. H. J . Organomet. Chem. 1983,
250, 395. (b) Koga, N.; Obara, S.; Kitaura, K.; Morokuma, K. J . Am.
Chem. Soc. 1985, 107, 7109. (c) Koga, N.; Obara, S.; Morokuma, K. J .
Am. Chem. Soc. 1984, 106, 4625. (d) Obara, S.; Koga, N.; Morokuma,
K. J . Organomet. Chem. 1984, 270, C33. (e) Koga, N.; Morokuma, K.
J . Am. Chem. Soc. 1988, 110, 108. (f) Suresh, C. H.; Koga, N.
Organometallics 2001, 20, 4333.
(16) Zhan. S.; Meng. Q.; You. X.; Wang, G.; Zheng, P.-J . Polyhedron
1996, 15, 2655.
(17) Kawamura, K.; Nakazawa, H.; Miyoshi, K. Organometallics
1999, 18, 1517.

120 Organometallics, Vol. 23, No. 1, 2004
Nakazawa et al.
The main electronic change when going from 14 to
15 is the cleavage of the Fe-Si bond and the formation
of the N-Si bond. The superior strength of the N-Si
bond over the Fe-Si bond may account for the higher
stability of 15 as compared to 14. Note that 15 is also
an Fe(II) complex with an 18-electron configuration.
If the methyl group of the imino ligand migrates to
the iron atom, the methyl iron complex could be formed.
We can expect such a reaction because it is similar to a
decarbonylation reaction (RCO-[M] f R-[M]-CO).20
It has been found that a distortion in 15 that breaks
the FerN bond leads to the transition state TS4. This
process requires an activation energy of 15.0 kcal/mol.
The Fe-N bond breaking as we go from 15 to TS4
accounts for a decrease in the formal electron count on
the iron atom by two electrons, meaning that the iron
atom will formally have a vacant d orbital in TS4. As a
result the important orbital interaction realizing the
three-center transition state TS4 arises. This is depicted
in Figure 5. The methyl anion migrates to the vacant d
orbital on iron, and the orbital of the CN group that
was responsible for the CC bond is left as the π* orbital.
The orbital interaction depicted in Figure 5 is of great
importance, as it is usually encountered in the carbonyl
insertion reactions of d8 square-planar palladium and
platinum complexes.20 The product of this step, Cp(CO)-
MeFe(CNSiMe3), 16, is 5.9 kcal/mol more stable than
15, which is presumably due to the presence of a strong
Fe-C bond with double-bond character, viz., the Fe-C
bond of length 1.796 Å (cf. Figure 4).
The reaction profile in Figure 4 well explains the
rupture of the acetonitrile C-C bond by the Fe(II)
complex. The highest activation barrier of 15.0 kcal/mol
is well within the reach of a feasible chemical reaction.
Once 16 is formed, the formation of the product 2 in eq
1 can be easily achieved by a silylisocyanide/phosphite
exchange reaction. Further CO/phosphite exchange
reaction on 2 would give complex 3.
(iii) P a th III: F or m a tion of Cp (CO)MeF e(NC-
SiMe3) a n d Su bsequ en t Rea ction s. While we found
the reaction path for the Fe-Me bond formation, we
have to look for the reaction path for the Fe-CN bond
formation. At this point we come back to the Fe(II)
system 16, the Cp(CO)MeFe(CNSiMe3) complex (cf.
Figure 3). It is expected that the dissociation of the
ligand trimethylsilyl isocyanide from 16 can lead to the
reaction paths for the Fe-CN bond formation. One
possibility is that the dissociated trimethylsilyl isocya-
nide isomerizes to trimethylsilyl cyanide, of which the
Si-C bond is broken by Cp(CO)MeFe, 17. In Figures 6
and 7, the results for this reaction path are illustrated.
F igu r e 2. Path I: The oxidative addition of the CC bond
of acetonitrile to Fe(II) complex. The depicted values are
ZPE-corrected potential energy and the Gibbs free energy
(in square brackets) in kcal/mol relative to 11 + acetoni-
trile.
explained by the reaction mechanism including the
direct C-C bond oxidative addition of acetonitrile.
However, the oxidative addition is less likely to happen.
As shown in Figure 2, this reaction starts from 12 with
the end-on coordination of acetonitrile to 11 and passes
through the three-centered transition state TS1 to lead
to Cp(CO)MeFe(CN)(SiMe3), 13. Since this reaction
requires a very high activation barrier of 52.7 kcal/mol,
we rule out its possibility.
(ii) P a th II: C-C Bon d Clea va ge via th e CN
Bon d In ser tion in to F e-Si Bon d . Next, let us
consider the reaction from 12 described in Figure 3. The
energy profile for this reaction is depicted in Figure 4
along with some bond length parameters. Complex 12
is coordinatively saturated, and without further per-
turbations on the electronic structure, none of the
products in eq 1 (2, 3, and 4) could be formed from 12.
So we consider the transformation of 12 to the CN
π-coordinated species 14. This process takes place
through TS2 with the activation energy of 14.8 kcal/
mol.
In 14, the insertion of the CN bond into the Fe-Si
bond is possible. The calculations showed that this
requires only a small activation energy of 4.0 kcal/mol
(cf. TS3 in Figure 3). The Mulliken charge on silicon
and nitrogen atoms of 14 was found to be 0.929 and
-0.268, respectively, and therefore the silicon atom may
strongly attract the negatively charged nitrogen atom
toward its coordination region. The geometry of the
silicon center in TS3 is very typical of a hypervalent
silicon atom19 because the Si-C bond anti to the newly
formed N-Si bond is longer compared to the other Si-C
bonds by 0.03 Å (cf. Figure 4 for bond lengths). From
TS3, a somewhat stable imino complex with the coor-
dination of a nitrogen lone pair to the Fe atom, Cp(CO)-
Fe(C(Me)dNSiMe3), 15, is formed.
(19) (a) Muguruma, C.; Koga, N.; Hatanaka, Y.; El-Sayed, I.;
Mikami, M.; Tanaka, M. J . Phys. Chem. A 2000, 104, 4928. (b) El-
Sayed, I.; Hatanaka, Y.; Muguruma, C.; Shimada, S.; Tanaka, M.;
Koga, N.; Mikami, M. J . Am. Chem. Soc. 1999, 121, 5095. (c)
Weinmann, M.; Huttner, G.; Lang, H. J . Organomet. Chem. 1998, 561,
131. (d) Bearpark, M. J .; McGrady, G. S.; Prince, P. D.; Steed, J . W. J .
Am. Chem. Soc. 2001, 123, 7736. (e) Perrin, L.; Maron, L.; Eisenstein,
O. Inorg. Chem. 2002, 41, 4355.
(20) (a) Koga, N.; Morokuma, K. Chem. Rev. 1991, 91, 823. (b) Koga,
N.; Morokuma, K. In Quantum Chemistry: The Challenge of Transition
Metals and Coordination Chemistry; Veillard, A., Ed.; NATO ASI Series
176; : Reidel: Dordrecht, 1986; p 351. (c) Koga, N.; Morokuma, K. J .
Am. Chem. Soc. 1985, 107, 7230. (d) Koga, N.; Morokuma, K. J . Am.
Chem. Soc. 1986, 108, 6536. (e) Nakamura, S.; Dedieu, A. Chem. Phys.
Lett. 1984, 111, 243. (f) Fukuoka, A.; Fukagawa, S.; Hirano, M.; Koga,
N.; Komiya, S. Organometallics 2001, 20, 2065.

C-C Bond Cleavage of Acetonitrile
Organometallics, Vol. 23, No. 1, 2004 121
F igu r e 3. Path II: The cleavage of the CC bond of acetonitrile through insertion of acetonitrile into the Fe-Si bond. The
depicted values are ZPE-corrected potential energy and the Gibbs free energy (in square brackets) in kcal/mol relative to
11 + acetonitrile.
F igu r e 4. ZPE-corrected potential energy profile (in kcal/mol) for path II. The partial structures of the geometries (the
Cp ring is excluded for clarity) are also depicted along with some bond length (in Å) parameters (the view is from the top
of the Cp ring).
exchange is 21.0 kcal/mol endothermic and energetically
more feasible than the following isomerization of tri-
methylsilyl isocyanide to trimethylsilyl cyanide with the
activation energy of 23.9 kcal/mol. Trimethylsilyl cya-
nide is more stable than trimethylsilyl isocyanide. This
is in accord with the experimental fact that trimethyl-
silyl cyanide is thermodynamically more favorable.21
Since the binding energy of trimethylsilyl cyanide (23.7
kcal/mol) is larger than that of acetonitrile (21.3 kcal/
mol), the former would replace the latter to give tri-
methylsilyl cyanide complex Cp(CO)MeFe(NCSiMe3),
F igu r e 5. Three-center orbital interaction in the transition
state TS4.
18. In fact, in the present experiments trimethylsilyl
As we can see in Figure 6, the dissociation of trimeth-
cyanide was identified in the reaction mixture (described
ylsilyl isocyanide from 16 requires a large dissociation
in section (a )), and therefore the formation of 18 is quite
energy of 42.3 kcal/mol. However, such a reaction is not
likely to occur.
difficult to imagine because the unstable intermediate
Complex 18 is an 18-electron Fe(II) complex in which
17 formed in that reaction can be easily stabilized by
the nitrogen of the trimethylsilyl cyanide has an end-
the coordination of a solvent acetonitrile molecule. At
on coordination to the iron atom, and many attempts
the present level of theory, the energy of 21.3 kcal/mol
is released when acetonitrile coordinates to 17. In other
(21) Thayer, J . S. Inorg. Chem. 1968, 7, 2599. (b) Seckar, J . A.;
words, the acetonitrile/trimethylsilyl isocyanide ligand
Thayer, J . S. Inorg. Chem. 1976, 15, 501.

122 Organometallics, Vol. 23, No. 1, 2004
Nakazawa et al.
the migration takes place in a single step without any
intermediate.22 The activation barrier is only 7.6 kcal/
mol, as this transition state is quite similar to the
structure of 21. The methyl migration product 22 can
be best described as the Fe(II) system with Cp(CN)Fe
bound to tetramethylsilane via two Fe‚‚‚H-C interac-
tions. At this point, the formation of product 4 in eq 1
is quite easy to explain, as it corresponds to the
coordination of two phosphite ligands to 22 that replaces
the Fe‚‚‚H-C interactions. Actually, Me4Si, which is
expected to be released from 22, was experimentally
detected (vide supra).
(iv) P a th IV: Oxid a tive Ad d ition of C-H Bon d
of Aceton itr ile a n d Su bsequ en t Rea ction s. There
is yet another possibility that we have explored for the
C-C bond cleavage of acetonitrile as well as the
formation of the cyanide complex (cf. Figure 9). In this
mechanism, acetonitrile approaches 11 so that one of
its methyl C-H bonds interacts with the iron atom. This
gives a weak complex 23. Note that 23 is less stable
than 11+acetonitrile by 3.5 kcal/mol, indicating that the
agostic interaction in 11 is stronger than the interaction
between 11 and acetonitrile in 23. The C-H‚‚‚Fe
interaction in 23 leads to the C-H bond oxidative
addition product 24 through the transition state TS9.
The activation energy for this step is only 6.3 kcal/mol.
Note that complex 24 can be best described as a Cp(CO)-
(CH2CN)Fe(HSiMe3) complex. If the carbonyl group
dissociates from 24, the vacant d orbital thus produced
on iron will show a good interaction with the CN group
of the CH2CN ligand as seen in 25. The complex 25 is
also characterized by the Fe‚‚‚H-Si interaction.
In the next step through the transition state TS10,
the C-C bond of the acetonitrile fragment breaks. This
step requires a moderately high activation energy of
25.0 kcal/mol. This value is not too large for the reaction
to take place, because in 25 the coordination of the CN
bond to Fe imposes strain on the CC bond in the FeCC
triangle to activate it. The product of this step, Cp(CN)-
Fe(CH2)(HSiMe3), 26, is a carbene complex with Fe‚‚‚
H-Si interaction.
In the next step from 26, the SiMe3 group of HSiMe3
migrates to the carbene ligand. This reaction passes
through the transition state TS11 to lead to Cp(CN)-
Fe(H)(CH2SiMe3), 27.23 The CH2SiMe3 group of 27 then
undergoes a minor change in its orientation with respect
to the cyano group to lead to TS12. Then the hydrogen
atom migrates to the carbon atom of the CH2SiMe3
fragment, so that finally Cp(CN)Fe(H3CSiMe3), 28, is
given. The structure of complex 28 has two Fe‚‚‚CH
interactions similar to that of complex 22 in path III.
The main difference between 28 and 22 is that in the
former the Fe‚‚‚CH interactions are from the same
methyl group, while in the latter such interactions are
from different methyl groups. The formation of the
product 4 in eq 1 is achieved when H3CSiMe3 in 28 is
replaced with the phosphites.
F igu r e 6. ZPE-corrected potential energy profile for the
formation of Cp(CO)MeFe(NCSiMe3), 18, from Cp(CO)-
MeFe(CNSiMe3), 16. Potential energy and the Gibbs free
energy (in square brackets) relative to 11 + acetonitrile
are also depicted along with the optimized geometries. All
values are in kcal/mol.
to locate a transition state for the oxidative addition of
the Si-CN bond of trimethylsilyl cyanide failed. Dis-
sociation of the carbonyl group giving a coordinatively
unsaturated reactive intermediate, CpMeFe(NCSiMe3),
19, is required to realize the Si-CN bond cleavage.
The breaking of the Si-CN bond starts from the side-
on trimethylsilyl cyanide complex 20. The isomerization
of the end-on complex 19 to 20 passes through the
transition state TS6 with the activation energy of 15.4
kcal/mol (cf. Figures 7 and 8). Transformation of 19 to
20 weakens the Fe-N bond as it is elongated by 0.259
Å.
The Si-CN bond breaking first passes through the
transition state TS7 with the activation energy of 11.7
kcal/mol, in which the interaction of the Si-CN bond
with the Fe atom makes the N atom dissociate. In the
product of this step, 21, the largely deformed trimeth-
ylsilyl cyanide seems to coordinate to iron with the help
of the hypervalent character of the silicon atom.18f
Though the Si-CN distance of 2.057 Å is only 0.165 Å
longer than that of 20, the large Fe-C-N angle of
166.0° and the short Fe-Si distance of 2.363 Å suggest
that 21 is CpMeFe(CN)(SiMe3) with the interaction
between the Fe atom and the C-H bond as indicated
by the long CH bond of 1.116 Å and the short Fe‚‚‚H
distance of 1.948 Å. This agostic interaction tilts the
SiMe3 ligand to make the silyl orbital interact with the
CN π orbitals. As a matter of fact, the structure of the
SiMe3 ligand of 21 is similar to that of 11.
Once the Si-CN bond is weakened in 21, the pos-
sibility of the migration of the methyl group from iron
to silicon exists.18f This possibility is explored in the next
step, and the transition state TS8 is located for that.
Although the isomerization of 21 to CpMeFe(CN)(SiMe3)
with the Me and SiMe3 ligands being cis to each other
and their succeeding reductive elimination are expected
as a possible reaction mechanism, the results show that
(22) In the model reaction with SiH3 in place of SiMe3, there is
CpMeFe(CN)(SiH3) with the Me and SiH3 groups being cis to each other
as an intermediate.
(23) Instead of the SiMe3 group, the hydrogen atom of the newly
formed Fe-H bond of 26 can also migrate back to the carbene ligand.
The activation energy needed for this was the same as that for the 26
to 27 conversion. However, the reaction profile consisting of the
hydrogen migration product that leads to the agostic complex 28 was
placed at higher energy level than that described in the reaction profile
depicted in Figure 9. So we do not describe this path in detail.

C-C Bond Cleavage of Acetonitrile
Organometallics, Vol. 23, No. 1, 2004 123
F igu r e 7. Path III: The cleavage of the Si-CN bond of trimethylsilyl cyanide activated by Fe(II) complex. ZPE-corrected
potential energy and the Gibbs free energy (in square brackets) in kcal/mol relative to 11 + acetonitrile are depicted.
F igu r e 8. ZPE-corrected potential energy profile (in kcal/mol) for path III. The partial structures of the geometries (Cp
ring is excluded) are also depicted along with some bond length (in Å) parameters (the view is from the top of the Cp ring).
As we can see, the reaction path IV described above
and path III (cf. Figures 6, 7, and 8) can give the cyanide
complex 4 in eq 1. However, path III is the most
preferred one, as it needs the highest activation energy
of 23.9 kcal/mol as compared to that of 25.0 kcal/mol
required for the CC bond cleavage step of path IV (cf.
Figure 9). Further, path IV is thermodynamically less
favorable than path III, as it contains intermediates and
transition states with very high energies. Moreover,
during the reaction path IV, HSiMe3 dissociation is
possible. For instance, that from 25 is endothermic by
only 8.0 kcal/mol, indicating that HSiMe3 dissociation
is easier than CC bond cleavage and that the cyano-
methyl complex should be detected.
(v) P a t h V: Me3Si Migr a t ion in t h e Me3SiNC
Com p lex. Another possibility for the Fe-CN bond
formation is via the Me3Si migration to the iron atom
in Cp(CO)MeFe(CNSiMe3), 16. However, this reaction
can be neglected because the DFT calculations showed
that this process required a large activation energy of
51.3 kcal/mol. The CO dissociation from 16 reduces the
activation energy to 37.2 kcal/mol, but it is still too large.
(c) F u r th er Exp er im en ts th a t Va lid a te Th eor eti-
ca l Resu lts. In the theoretical considerations, it is
assumed that at first a carbonyl ligand is liberated from
1 by photolysis to give the 16e species Cp(CO)Fe(SiMe3).
This was supported by the experimental fact that
Cp(CO)LFe(SiMe3) was detected in the reaction of 1
with an equimolar amount of MeCN in the presence of
L in THF. We have also seen that the theoretical
consideration as to the C-C bond cleavage by the iron
silyl complex proposed that silyl migration from iron to
the nitrogen in coordination nitrile takes place to give
an iron complex having methyl and silylisocyanide
ligands (complex 16 in Figure 3). In fact, the detection
of the compound Me3SiCN as described in section (a )
strongly supported the formation of 16 as well as the
whole mechanism given in Figure 4 (path II). It is felt
that the detection of complex 16 itself will further
confirm the mechanism.
Therefore, to isolate complex 16, several attempts
were made, and it was finally found that a photoreaction
of Cp(CO)2Fe(SiPh3) (29) in the presence of PPh3 in
acetonitrile yielded Cp(PPh3)MeFe(CNSiPh3) (30) (eq 4).

124 Organometallics, Vol. 23, No. 1, 2004
Nakazawa et al.
F igu r e 9. Path IV: Oxidative addition of the C-H bond of acetonitrile and subsequent reactions. The depicted values
are ZPE-corrected potential energy and the Gibbs free energy (in square brackets) in kcal/mol relative to 11 + acetonitrile.
Complex 30 was isolated as red crystals. The 31P NMR
spectrum showed a signal at 87.60 ppm, which was
assigned to coordinating PPh3. In the 13C NMR spec-
trum, there were two characteristic resonances:
a
doublet at -21.07 ppm with J ) 23.5 Hz due to a methyl
ligand and a doublet at 221.23 ppm with J ) 35.5 Hz
due to a silylisocyanide ligand. The complex exhibited
an intense IR absorption at 1926 cm-1 attributable to
CN stretching. These spectroscopic data suggest that
the isolated complex 30 is formulated as Cp(PPh3)MeFe-
(CNSiPh3).
F igu r e 10. ORTEP drawing of 30 showing the atom-
numbering scheme. The thermal ellipsoids are drawn at
the 50% probability level. Hydrogen atoms are omitted for
clarity.
The structure was confirmed by X-ray analysis, which
is depicted in Figure 10. It clearly shows that C-CN
bond cleavage took place. The C-N bond distance (1.181
Å) is longer than that for [Cp*(PMe3)(iPr)Rh(CNSi-
Ph3)]+,13 indicating that π-back-donation from the iron
fragment to the π* orbital of the CN multiple bond is
stronger than that from the Rh fragment. This is
reasonably understood from the cationic Rh fragment.
It is noteworthy that the C-N bond distance for 30 is
longer than that for 4 (1.166 Å). Since both are elec-
tronically neutral Fe(II) complexes, it shows that the
CNSiPh3 ligand is a better π-acceptor than the CN
ligand. The C-N-Si angle for 30 is 156.7°, whereas that
for [Cp*(PMe3)(iPr)Rh(CNSiPh3)]+ is 175.9° (cf. Tables
1 and 2). The angle for 30 seems to be due to steric
repulsion between the bulky PPh3 and SiPh3 groups.
F igu r e 11. Reaction sequences of acetonitrile CC bond
cleavage depicted with line drawings.
(11) proposed that the mechanism of the C-C bond
cleavage corresponds to the insertion of CN of acetoni-
trile to the Fe-Si bond, followed by methyl migration
to iron to give complex 16. The reaction sequences are
depicted with line drawings in Figure 11. The CN
insertion and the methyl migration required activation
energies of 14.8 and 15.0 kcal/mol, respectively. The low
Con clu sion s
The reactions mentioned above are quite rare ex-
amples of the activation of a C-C bond connected to a
cyanide group promoted by a transition metal complex
with a silyl ligand.13 The experimental results and the
DFT level calculations starting from [Cp(CO)Fe(SiMe3)]

C-C Bond Cleavage of Acetonitrile
Organometallics, Vol. 23, No. 1, 2004 125
in the photoreaction of 1 with 1 equiv of CD3CN in the presence
activation barrier for the CN insertion reaction is
attributed to the hypervalent character of the silicon
atom and the charge distribution of the nitrogen and
silicon atoms. Both experiment and theory supported
the formation of Me3SiCN in the reaction mixture. The
formation of Fe(II) cyanide complex (4) is due to the
reaction that led to the cleavage of the Si-CN bond of
Me3SiCN in the coordination sphere of the CpMe(CO)-
Fe complex. In the formation of complexes 2, 3, and 4,
the hypervalent character of the silicon atom played a
major role.
of L in THF.
2
2-CD3: 1H NMR (δ, in CDCl3): 4.37 (s, 5H, C5H5). D NMR
(δ, in CHCl3): -0.23 (s, CD3). 13C NMR (δ, in CDCl3): -26.33
2
1
(d of sept, J PC ) 30.4 Hz, J DC ) 20.3 Hz, FeCD3), 82.56 (d,
2J PC ) 1.2 Hz, C5H5), 222.48 (d, 2J PC ) 46.6 Hz, CO). 31P NMR
(δ, in CDCl3): 177.68 (s).
P h otor ea ction of 5, 6, or 7 in MeCN in th e P r esen ce
of P (NMeCH2)2(OMe). Cp(CO)2FeMe (3.4 mmol), MeCN (70
mL), and P(NMeCH2)2(OMe) (6.8 mmol) were put in a Pyrex
Schlenk tube, and the solution was irradiated with a 400 W
medium-pressure mercury arc lamp at 0 °C for 4 h. After the
solvent was removed, the residue was loaded on an alumina
column. A yellow complex eluted with hexane was collected
and dried in vacuo to give yellow powders of 8.14 The isolation
yield based on Cp(CO)2FeMe was 39%. The reactions of 6 and
7 were undertaken similarly to give 9 (isolation yield, 36%)
and 10 (isolation yield, 38%), respectively.
Exp er im en ta l Section
Gen er a l Rem a r k s. All reactions were carried out under
an atmosphere of dry nitrogen using standard Schlenk tech-
niques. Column chromatography was done quickly in the air.
MeCN, CH2Cl2, and THF were distilled from CaH2, P2O5, and
sodium metal, respectively, and then they were stored under
a dry nitrogen atmosphere. P{NMeCH2}2(OMe),24 Cp(CO)2Fe-
(SiMe3),25 and Cp(CO)2Fe(SiPh3)26 were prepared according to
the literature method.
P r ep a r a tion of 30. Cp(CO)2Fe(SiPh3) (0.77 g, 1.76 mmol),
PPh3 (0.93 g, 3.55 mmol), and MeCN (20 mL) were put in a
Pyrex Schlenk tube, and the solution was irradiated with a
400 W medium-pressure mercury arc lamp at 0 °C for 4 h. An
orange powder thus formed was washed with MeCN (5 mL ×
4) and dried in vacuo. The powder was dissolved in toluene (3
mL), and the solution was charged on a Celite column. An
orange band eluted with toluene was collected, and the solvent
was removed under reduced pressure to give 30 as an orange
powder (0.39 g, 0.57 mmol, yield 33%). IR (νCN, in CH2Cl2):
IR spectra were recorded on a Shimadzu FTIR-8100A or a
Perkin-Elmer Spectrum One spectrometer. A J EOL LA-300
1
multinuclear spectrometer was used to obtain H, 2D, 13C, and
31P NMR spectra. 1H and 13C NMR data were referenced to
Me4Si, and 31P NMR data were referenced to 85% H3PO4. The
chemical shift in 2D NMR was referred to the resonance of
CDCl3 at 7.24 ppm. Elemental analysis data were obtained
on a Perkin-Elmer 2400 CHN elemental analyzer.
3
1926. 1H NMR (δ, in CDCl3): -0.22 (d, J PH ) 6.6 Hz, 3H,
FeCH3), 4.08 (s, 5H, C5H5), 7.06-7.49 (m, 30H, Ph). 13C NMR
2
(δ, in CDCl3): -21.07 (d, J PC ) 23.5 Hz, FeCH3), 83.94 (s,
P h ot or ea ct ion of 1 in MeCN in t h e P r esen ce of
P (NMeCH2)2(OMe). Cp(CO)2Fe(SiMe3) (3.4 mmol), MeCN (70
mL), and P(NMeCH2)2(OMe) (6.8 mmol) were put in a Pyrex
Schlenk tube, and the solution was irradiated with a 400 W
medium-pressure mercury arc lamp at 0 °C for 4 h. After the
solvent was removed, the residue was loaded on an alumina
column. An orange complex eluted with CH2Cl2 and a yellow
complex eluted with acetone were collected respectively and
dried in vacuo to give yellow powders of 2 and 4. The isolation
yields were 15% for 2 and 25% for 4 based on 1. The
spectroscopic data for 2 were identical to those for Cp(CO)-
{P(NMeCH2)2(OMe)}FeMe, which has already been reported.14
4: IR (νCN, in CH2Cl2): 2063. 1H NMR (δ, in CDCl3): 2.70
C5H5), 127.54 (d, J CP ) 8.7 Hz, PPh3), 127.92 (s, SiPh3), 128.69
(d, J CP ) 1.9 Hz, PPh3), 130.08 (s, SiPh3), 133.32 (d, J CP ) 10.0
Hz, PPh3), 133.34 (s, SiPh3), 135.04 (s, SiPh3), 137.96 (d,
2
J CP ) 36.6 Hz, PPh3), 221.23 (d, J CP ) 35.5 Hz, FeCN). 31P
NMR (δ, in CDCl3): 87.60. Anal. Calcd for C43H38FeNPSi: C,
75.54; H, 5.60; N, 2.05. Found: C, 75.75; H, 5.62; N, 1.55.
X-r a y Str u ctu r e Deter m in a tion for 4 a n d 30. Suitable
crystals of 4 and 30 were obtained through crystallization from
toluene/hexane and from CH2Cl2/hexane, respectively, and
were mounted individually on a glass fiber. All measurements
for 4 and 30 were made on a Mac Science DIP2030 diffracto-
meter and on a Rigaku/MSC Mercury CCD diffractometer,
respectively, with graphite-monochromated Mo KR radiation
(λ ) 0.710 73 Å) at 200 K. Crystal data, data collection
parameters, and results of the analysis are summarized in
Table 1.
The structures were solved by direct methods with the
program SIR9227 and expanded using Fourier techniques.28
Positions of hydrogen atoms were determined from subsequent
difference Fourier maps. The non-hydrogen atoms were refined
anisotropically. Hydrogen atoms were refined isotropically. An
extinction correction was applied. All calculations were per-
formed using the program package teXsan.29
3
3
(d, J PH ) 4.8 Hz, 3H, NCH3), 2.72 (d, J PH ) 5.1 Hz, 3H,
3
3
NCH3), 2.82 (d, J PH ) 4.8 Hz, 3H, NCH3), 2.84 (d, J PH ) 5.1
Hz, 3H, NCH3), 3.00-3.05 (m, 4H, NCH2), 3.14-3.30 (m, 4H,
3
3
NCH2), 3.20 (d, J PH ) 5.3 Hz, 3H, OCH3), 3.21 (d, J PH ) 5.5
Hz, 3H, OCH3), 4.28 (s, 5H, C5H5). 13C NMR (δ, in CDCl3):
2
33.73 (m, NCH3), 51.22 (d, J PC ) 6.2 Hz, OCH3), 51.30 (d,
2J PC ) 6.2 Hz, OCH3), 51.47 (s, NCH2), 51.73 (s, NCH2), 78.74
(s, C5H5), 153.58 (t, J CP ) 41.9 Hz, CN). 31P NMR (δ, in CH2-
2
Cl2): 174.72. Anal. Calcd for C16H31FeN5O2P2: C, 43.36; H,
7.05; N, 15.80. Found: C, 43.54; H, 7.12; N, 15.80.
1
Formation of 3 was evidenced by the H, 13C, and 31P NMR
Deta ils of Com p u ta tion a l Meth od . All the molecular
geometries were optimized at the DFT level of theory using
the B3LYP hybrid functional30 with the Gaussian 98 suite of
programs.31 For Fe, the basis set LanL2DZ was used.32 In this
basis set the 10 innermost electrons of Fe have been replaced
by an effective core potential (ECP) of Hay and Wadt.32 For
H, C, N, O, and Si, 6-31G(d) basis functions were selected.33
measurements of the reaction mixture mentioned above,
together with those of the reaction mixture containing mainly
2 and 3 in the photoreaction of [Cp(CO)2FeMe] with L in
acetonitrile.
3
3: 1H NMR (δ, in CDCl3): -0.59 (t, J PH ) 6.0 Hz, 3H,
FeCH3), 4.11 (s, 5H, C5H5). 13C NMR (δ, in CDCl3): -27.77 (t,
2J PC ) 34.4 Hz, FeCH3), 79.13 (s, C5H5). 31P NMR (δ, in
CDCl3): 180.25 (s).
(27) Aotomare, A.; Burla, M. C.; Camalli, M.; Cascarano, M.;
Giacovazzo, C.; Gaugliardi, A.; Polidori, G. J . Appl. Crystallogr. 1994,
27, 435.
(28) Beurskens, P. T.; Admiraal, G.; Beurskens, G.; Bosman, W. P.;
de Gelder, R.; Israel, R.; Smits, J . M. M. The DIRDIF-94 Program
System; Technical report of the crystallography laboratory; Univeristy
of Nijmegen; Nijmegen, The Netherlands, 1994.
Formation of Cp(CO)LFe(CD3) (2-CD3) was evidenced by the
1H, 2D, 13C, and 31P NMR measurements of the reaction
mixture, together with those of the reaction mixture produced
(24) Ramirez, F.; Paatwardham, A. V.; Kugler, H. J .; Smith, C. P.
J . Am. Chem. Soc. 1967, 89, 6272.
(29) teXsan,
A Crystal Structure Analysis Package; Molecular
(25) King, R. B.; Pannel, K. H. Inorg. Chem. 1968, 7, 1500.
(26) Cerveau, G.; Colomer, E.; Corriu, R.; Douglas, W. E. J . Orga-
nomet. Chem. 1977, 135, 373.
Structure Corporation: The Woodlands, TX, 1985 and 1992.
(30) (a) Becke, A. D. J . Chem. Phys. 1993, 98, 1372. (b) Becke, A. D.
J . Chem. Phys. 1993, 96, 5648.

126 Organometallics, Vol. 23, No. 1, 2004
Nakazawa et al.
This method would give reliable geometries.34 Normal coor-
dinate analysis has been performed for all stationary points
to characterize the transition states (TSs) and equilibrium
structures. Therefore, the energy minimum structures reported
in this paper show positive eigenvalues of the Hessian matrix,
whereas TSs have one negative eigenvalue. For most TSs, the
analysis including the visualization of the negative frequency
was sufficient to specify the corresponding reaction path. In
some cases, intrinsic reaction coordinate (IRC) calculations35
near the TS region followed by geometry optimization of both
reactants and products were performed to confirm the con-
nectivity of the TSs. Unscaled vibrational frequencies were
used to calculate zero-point energy (ZPE) corrections to total
energy. The Gibbs free energies were also calculated employing
the usual approximations of statistical thermodynamics (ideal
gas, harmonic oscillator, and rigid rotor) at the temperature
of 298.15 K and the pressure of 1.00 atm. Unless otherwise
noted, the ZPE-corrected energies are used in the discussion.
(31) Frisch, M. J .; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J . R.; Zakrzewski, V. G.; Montgomery, J . A.,
J r.; Stratmann, R. E.; Burant, J . C.; Dapprich, S.; Millam, J . M.;
Daniels, A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J .;
Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo,
C.; Clifford, S.; Ochterski, J .; Petersson, G. A.; Ayala, P. Y.; Cui, Q.;
Morokuma, K.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.;
Foresman, J . B.; Cioslowski, J .; Ortiz, J . V.; Stefanov, B. B.; Liu, G.;
Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin, R. L.;
Fox, D. J .; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.;
Gonzalez, C.; Challacombe, M.; Gill, P. M. W.; J ohnson, B.; Chen, W.;
Wong, M. W.; Andres, J . L.; Gonzalez, C.; Head-Gordon, M.; Replogle,
E. S.; Pople, J . A. Gaussian 98, Revision A.11; Gaussian, Inc.:
Pittsburgh, PA, 1998.
(32) Hay, P. J .; Wadt, W. R. J . Chem. Phys. 1985, 82, 299.
(33) Hehre, W. J .; Ditchfield, R.; Pople, J . A. J . Chem. Phys. 1972,
56, 2257. (b) Hariharan, P. C.; Pople, J . A. Theor. Chim. Acta 1973,
28, 213. (c) Francl, M. M.; Pietro, W. J .; Hehre, W. J .; Binkley, J . S.;
DeFrees, D. J .; Pople, J . A.; Gordon, M. S. J . Chem. Phys. 1982, 77,
3654.
Ack n ow led gm en t. Part of the calculations were
carried out at the Research Center for Computational
Science of Okazaki National Research Institutes, J apan.
We acknowledge financial support (14654123 and
14044071) from Ministry of Education, Science, Sports,
and Culture, J apan.
Su p p or tin g In for m a tion Ava ila ble: X-ray crystallo-
graphic data (PDF) and optimized geometries of all the
systems and their energy values. This material is available
free of charge via the Internet at http://pubs.acs.org.
OM0208319
(34) Foresman, J . B.; Frisch, A. E. Exploring Chemistry with
Electronic Structure Methods, 2nd ed.; Gaussian Inc.: Pittsburgh, PA,
1995; Chapter 7, p 158.
(35) (a) Fukui, K. Acc. Chem. Res. 1981, 14, 363. (b) Gonza´lez, C.;
Schlegel, H. B. J . Chem. Phys. 1991, 95, 5853.