CNR and CO insertion reactions of 2,6-xylyl isocyanide with p-chlorobenzylpentacarbonylmanganese

Article abstract of DOI:10.1021/om9903644

Stirring a solution of (CO)₅MnCH₂C₆H₄-p-Cl, 1, with 2 equiv of 2,6-xylyl isocyanide in toluene in the presence of a catalytic amount of PdO produced isocyanide insertion products including the bis(iminoacyl) (CO)₃(CN-xylyl)Mn[C(=N-xylyl)C(=N-xylyl)CH₂C ₆H₄-p-Cl], 3, and the first octahedrally based tris(iminoacyl) (CO)₄Mn[C(=N-xylyl)C(=N-xylyl)C(=N-xylyl)-CH₂C ₆H₄-p-Cl], 6. The same reaction in THF afforded the carbonyl insertion product (CO)₃-(xylyl-NC)₂MnC(O)CH₂C₆H ₄-p-Cl, 9. Reaction of 1 with 1 equiv of xylyl isocyanide in THF without PdO produced the expected acyl (CO)₄(CN-xylyl)MnC(O)CH₂C₆H₄-p-Cl, 11, along with a minor quantity of 3. Exposure of the reaction mixture to CO₂ afforded the unique dimanganese species (CO)₄Mn[C{=CH(C₆H ₄-p-Cl)}N-(xylyl)C(O))OMn(CO)₄[C(=NH-xylyl)-CH ₂C₆H₄-P-Cl]], 12, in which the manganese centers are linked through a carbamato bridge. X-ray structures of 3, 6, and 12 are reported.

Full text of DOI:10.1021/om9903644

5594
Organometallics 1999, 18, 5594-5605
CNR a n d CO In ser tion Rea ction s of 2,6-Xylyl Isocya n id e
w ith p-Ch lor oben zylp en ta ca r bon ylm a n ga n ese
Thomas M. Becker,1a J ohn J . Alexander,*,1a J eanette A. Krause Bauer,1a
J effrey L. Nauss,1b and Fred C. Wireko1c
Department of Chemistry, University of Cincinnati, P.O. Box 210172,
Cincinnati, Ohio 45221-0172, Molecular Simulations Inc., 9685 Scranton Road,
San Diego, California 92121, and The Procter and Gamble Co., Miami Valley Laboratories,
11810 E. Miami River Road, P.O. Box 398707, Cincinnati, Ohio 45239-8707
Received May 17, 1999
Stirring a solution of (CO)5MnCH2C6H4-p-Cl, 1, with 2 equiv of 2,6-xylyl isocyanide in
toluene in the presence of a catalytic amount of PdO produced isocyanide insertion products
including the bis(iminoacyl) (CO)3(CN-xylyl)Mn[C(dN-xylyl)C(dN-xylyl)CH2C6H4-p-Cl], 3,
and the first octahedrally based tris(iminoacyl) (CO)4Mn[C(dN-xylyl)C(dN-xylyl)C(dN-xylyl)-
CH2C6H4-p-Cl], 6. The same reaction in THF afforded the carbonyl insertion product (CO)3-
(xylyl-NC)2MnC(O)CH2C6H4-p-Cl, 9. Reaction of 1 with 1 equiv of xylyl isocyanide in THF
without PdO produced the expected acyl (CO)4(CN-xylyl)MnC(O)CH2C6H4-p-Cl, 11, along
with a minor quantity of 3. Exposure of the reaction mixture to CO2 afforded the unique
dimanganese species (CO)4Mn[C{dCH(C6H4-p-Cl)}N-(xylyl)C(O)OMn(CO)4[C(dNH-xylyl)-
CH2C6H4-p-Cl]], 12, in which the manganese centers are linked through a carbamato bridge.
X-ray structures of 3, 6, and 12 are reported.
1. In tr od u ction
shown to vary in the following order: aromatic isocya-
nide > CO > aliphatic isocyanide.7 However, predicting
whether acyls or iminoacyls will be formed is not always
straightforward. Extended Hu¨ckel MO studies suggest
that iminoacyls are more thermodynamically stable
than their acyl counterparts. Also, the kinetic activation
energy barrier is larger for isocyanide insertions.8 These
findings suggest that the transformation of an acyl to
the more thermodynamically stable iminoacyl product
should be possible. Pizzano and co-workers have now
observed the first example of thermal conversion from
a molybdenum acyl to a molybdenum iminoacyl com-
plex.9
Workers in our laboratory have previously investi-
gated the reaction of alkylpentacarbonylmanganese
compounds with various isocyanides.10-12 Stirring any
of several aromatic or aliphatic isocyanides with various
benzylpentacarbonylmanganese complexes in THF pro-
duces primarily manganese acyls, p-XC6H4CH2C(O)Mn-
(CNR)(CO)4 (X ) H, Cl, or OMe), which are the result
of carbonylation and isocyanide addition. When the
manganese acyls that contain a terminal aliphatic
isocyanide are heated, the decarbonylation product,
p-XC6H4CH2Mn(CNR)(CO)4, results.12 However, if the
isocyanide is aromatic p-tolyl isocyanide, thermolysis
Organometallic insertion reactions can create new
C-C bonds and hence have attracted widespread inter-
est.2,3 Carbon monoxide insertion has been extensively
studied,4 while insertion reactions of other small mol-
ecules are now receiving increasing attention. One class
of molecules that has shown insertion chemistry is the
isocyanides, CNR.5 Carbon monoxide and isocyanides
are isolobal; therefore, similar types of reactivity are
expected for each. Isocyanides are generally regarded
as better σ donors and weaker π acceptors than carbon
monoxide.6 Unlike carbon monoxide, isocyanides contain
an R group, which can be varied to change the electronic
or steric properties.
Molecules containing both CO and CNR ligands often
display a competition between CO insertion (giving acyl
products) and isocyanide insertion (giving iminoacyl
products). The facility of insertion reactions in an Fe
system containing both carbonyls and isocyanides was
* To whom correspondence should be addressed.
(1) (a) University of Cincinnati, Cincinnati, OH. (b) Molecular
Simulations Inc., San Diego, CA. (c) The Procter and Gamble Co.,
Miami Valley Laboratories, Cincinnati, OH.
(2) Collman, J . P.; Hegedus, L. S.; Norton, J . R.; Finke, R. G.
Principles and Applications of Organotransition Metal Chemistry;
University Science Books: Mill Valley, CA, 1987.
(3) Alexander, J . J . In The Chemistry of the Metal-Carbon Bond,
vol.2; Hartley, F. R., Patai, S., Eds.; J . F. Wiley and Sons: Ltd., London,
1985; p 339.
(4) (a) Wojcicki, A. A. Adv. Organomet. Chem. 1973, 11, 87. (b)
Kuhlmann, E. J .; Alexander, J . J . Coord. Chem. Rev. 1980, 33, 195.
(5) (a) Singleton, E.; Oosthuizen, H. E. Adv. Organomet. Chem. 1983,
22, 209. (b) Yamamoto, Y. Coord. Chem. Rev. 1980, 32, 193. (c) Treichel,
P. M. Adv. Organomet. Chem. 1973, 11, 21.
(7) de Lange, P. P. M.; Fruhauf, H. W.; Kraakman, M. J . A.; van
Wijnkoop, M.; Kranenburg, M.; Groot, A. H. J . P.; Vrieze, K.; Fraanje,
J .; Wang, Y.; Numan, M. Organometallics 1993, 12, 417.
(8) Berke, H.; Hoffmann, R. J . Am. Chem. Soc. 1978, 100, 7724.
(9) Pizzano, A.; Sanchez, L.; Altmann, M.; Monge, A.; Ruiz, C.;
Carmona, E. J . Am. Chem. Soc. 1995, 117, 1759.
(10) Motz, P. L.; Alexander, J . J .; Ho, D. M. Organometallics 1989,
8, 2589.
(6) Cotton, F. A.; Wilkinson, G.; Murillo, C. A.; Bochmann, M.
Advanced Inorganic Chemistry, 6th ed.; J ohn Wiley & Sons: New York,
1999; p 246.
(11) Motz, P. L.; Williams, J . P.; Alexander, J . J .; Ho, D. M.
Organometallics 1989, 8, 1523.
(12) Kuty, D. W.; Alexander, J . J . Inorg. Chem. 1978, 17, 1489.
10.1021/om9903644 CCC: $18.00 © 1999 American Chemical Society
Publication on Web 12/02/1999

CNR and CO Insertion Reactions
Organometallics, Vol. 18, No. 26, 1999 5595
yields a diazabutadiene dimer, A.11 The proposed mech-
anism suggests formation of a highly reactive (alkyl)
(isocyanide) intermediate resulting from decarbonyla-
tion followed by rapid and preferential insertion of
aromatic isocyanide to give a Mn iminoacyl intermedi-
ate, which dimerizes.
to attempt to isolate intermediates in these reactions
in order to gain more information on the mechanism of
their production. Previous work in this laboratory
suggests that insertion of p-tolyl isocyanide in manga-
nese alkyl species is rapid.10,11 However, a study of
rhenium complexes in acetonitrile showed that the
insertion of the bulky aromatic 2,6-dimethylphenyl
isocyanide (xylyl isocyanide) is significantly slower than
that of p-tolyl isocyanide. The rate of insertion of xylyl
isocyanide into rhenium-carbon bonds was found to be
slower than p-tolyl isocyanide by approximately a factor
of 2.17 Yamamoto and co-workers18 and de Lange and
co-workers7 have estimated the relative sizes of isocya-
nide ligands. de Lange developed cone angles in a
manner similar to the development for phosphines.19
The slower insertion of the larger xylyl group (remote
cone angle ) 69°) in the Re system17 compared with
p-tolyl (remote cone angle measured for phenyl isocya-
nide ) 57°) indicates that the size of the aromatic
isocyanide is important in its insertion chemistry on Re.
These results prompted us to investigate the reactions
of alkylpentacarbonylmanganese complexes with xylyl
isocyanide and to compare the results to those obtained
with aliphatic isocyanides and the smaller aromatic
isocyanide, p-tolyl isocyanide.
Reactions of benzyl manganese carbonyls with p-tolyl
isocyanide in the presence of PdO were studied in order
to isolate simple isocyanide substitution products rather
than carbonylation products.10 PdO has been shown to
catalyze the substitution of CO by isocyanides.13-15 The
anticipated products were (CNR)(CO)4MnCH2C6H4-p-
X, containing a terminal p-tolyl isocyanide. However,
when the reaction was run in toluene with PdO, only
isocyanide insertion products were isolated.
2. Resu lts a n d Discu ssion
2.1. Rea ction in Tolu en e in th e P r esen ce of P d O.
Reactions of (CO)5MnCH2C6H4-p-Cl (1) with 2 equiv of
xylyl isocyanide in the presence of PdO in toluene at
room temperature were investigated under conditions
similar to those of the p-tolyl isocyanide investigation.11
Multiple products were isolated from the reaction
mixture. One minor product, Mn2(CO)9(xylyl-NC) (2),
is isolated in approximately a 7% yield. It is the result
of the reduction of 1. Reduction of manganese(I) carbo-
nyl complexes by an aromatic isocyanide to afford
substitution products of dimanganese decacarbonyls has
been reported by J oshi and co-workers.20 Also, small
amounts of Mn2(CO)8(CN-p-tolyl)2 and Mn2(CO)9(CN-
p-tolyl) were isolated by Motz21 from the reaction of 1
with p-tolyl isocyanide in THF at room temperature.
Complex 2 has previously been reported by Albers and
Coville, and the spectroscopic data for 2 match those in
the literature.13
All other products (3, 4, 5, and 6) isolated from the
reaction are depicted in Scheme 1 and are the result of
multiple isocyanide insertions. The product isolated in
the highest yield from the PdO-catalyzed reaction is a
manganese bis(iminoacyl) metallacycle, 3. The manga-
nese complex, 3, is a tricarbonyl species containing a
terminal xylyl isocyanide and a manganacycle that is
the result of insertion of 2 equiv of xylyl isocyanide into
the metal-alkyl bond. It is similar to the p-tolyl
manganacycle B with the following exceptions: xylyl
isocyanides replace p-tolyl isocyanides, and a terminal
Complexes B and C, which contain 2 equiv of isocyanide,
were isolated even if the reagents were mixed in a 1:1
ratio. It has been shown that C is the kinetic product
that converts to the thermodynamic product, B, over
time.11,16
The reaction of manganese alkyls with p-tolyl isocya-
nide, with and without PdO, results in products con-
taining inserted isocyanides (A, B, C). It is of interest
(17) Padolik, L. L.; Alexander, J . J .; Ho, D. M. J . Organomet. Chem.
1992, 440, 153.
(18) Yamamoto, Y.; Aoki, K.; Yamazaki, H. Inorg. Chem. 1979, 18,
1681.
(13) Albers, M. O.; Coville, N. J . S. Afr. J . Chem. 1982, 35, 139.
(14) Harris, G. W.; Coville, N. J . Organometallics 1985, 4, 908.
(15) Coville, N. J .; Stolzenberg, A. M.; Muetterties, E. L. J . Am.
Chem. Soc. 1983, 105, 2499.
(19) Tolman, C. A. J . Am. Chem. Soc. 1970, 92, 2956.
(20) J oshi, K. K.; Pauson, P. L.; Stubbs, W. H. J . Organomet. Chem.
1963, 1, 51.
(21) Motz, P. L. Ph.D. Dissertation, University of Cincinnati, 1988,
p 107.
(16) Laughlin, S. K.; Alexander, J . J . Unpublished results.

5596 Organometallics, Vol. 18, No. 26, 1999
Becker et al.
Sch em e 1
Mn is a distorted octahedron. The bite of the bis-
(iminoacyl) is 64.4 (4)°. The three carbonyls are es-
sentially linear, and the terminal isocyanide angle,
Mn-C(6)-N(2), is 173.6 (11)°. The C(5)-N(1)-C(15)
angle and the C(4)-N(3)-C(30) angle around the sp2
nitrogens are 126.1(10)° and 118.8(10)°, respectively.
The N(1)-C(5) and C(4)-N(3) bond distances of
1.300(13) and 1.275(13) Å, respectively, are as expected
for C-N double bonds. The Mn-C(4) distance of 2.039-
(12) Å is rather short for a manganese-acyl bond22 but
only slightly shorter than the Mn-iminoacyl bond
distance of 2.067 Å in B.10 The Mn-N(1) distance of
2.115(9) Å is slightly longer that the Mn-imine N
distance of 2.079 Å in B10 and in a similar bis-
(iminoacyl), four-membered manganacycle reported by
Cooper and co-workers.23 The terminal CO distances are
typical of those in alkyl manganese carbonyl com-
plexes.22 The Mn-C(6) bond length of the terminal
isocyanide is 1.945(13) Å and the C-N triple bond
distance is 1.137(13) Å, which are usual for manganese
isocyanide complexes.24
F igu r e 1. ORTEP diagram of 3.
xylyl isocyanide replaces a terminal CO, giving a fac-
tricarbonyl product. Interestingly, the xylyl isocyanide
tetracarbonyl product analogous to B was not isolated.
The structure of 3 has been confirmed by single-
crystal X-ray analysis. An ORTEP diagram of 3 is shown
in Figure 1. Crystal and diffraction data, including bond
angles and distances, are listed in Tables 1 and 2.
Complex 3 contains a four-membered bis(iminoacyl) ring
composed of Mn, N1, C5, and C4. The geometry around
The IR, 13C NMR, and mass spectra are consistent
with the structure shown and are unremarkable. The
(22) Treichel, P. M. In Comprehensive Organometallic Chemistry,
vol. 4; Wilkinson, G., Stone, F. G. A., Abel, E. W., Eds.; Pergamon
Press: New York, 1982; pp 1-159.
(23) Utz, T. L.; Leach, P. A.; Gieb, S. J .; Cooper, N. J . Organome-
tallics 1997, 16, 4109.

CNR and CO Insertion Reactions
Organometallics, Vol. 18, No. 26, 1999 5597
Ta ble 1. Su m m a r y of Cr ysta l Da ta for 3, 6, a n d 12
Ta ble 2. Selected Bon d Dista n ces [Å] a n d An gles
[d eg] for 3
Crystal Data and Structure Refinement for 3
empirical formula
fw
temp
wavelength
cryst syst
space group
unit cell dimens
C37H33ClMnN3O3
658.05
294(2) K
1.54178 Å
orthorhombic
Pbca
Mn-C(2)
Mn-C(1)
Mn-C(4)
Mn-C(5)
O(2)-C(2)
N(1)-C(5)
N(2)-C(7)
N(3)-C(30)
C(5)-C(23)
1.78(2)
Mn-C(3)
Mn-C(6)
Mn-N(1)
O(1)-C(1)
O(3)-C(3)
N(2)-C(6)
N(3)-C(4)
C(4)-C(5)
1.808(11)
1.945(12)
2.120(9)
1.146(13)
1.152(11)
1.137(13)
1.275(13)
1.47(2)
1.825(14)
2.039(11)
2.587(12)
1.142(13)
1.300(13)
1.416(14)
1.419(14)
1.523(14)
a ) 18.239(6) Å; R ) 90°
b ) 15.971(5) Å; â ) 90°
c ) 23.062(7) Å; γ ) 90°
6718(4) Å3, 8
volume, Z
density (cald)
abs coeff
F(000)
cryst size
θ range for data collection
limiting indices
no. of reflns collected
no. of indep reflns
refinement method
1.301 Mg/m3
C(2)-Mn-C(3)
C(3)-Mn-C(1)
C(3)-Mn-C(6)
C(2)-Mn-C(4)
C(1)-Mn-C(4)
C(2)-Mn-N(1)
C(1)-Mn-N(1)
C(4)-Mn-N(1)
C(3)-Mn-C(5)
C(6)-Mn-C(5)
N(1)-Mn-C(5)
C(5)-N(1)-Mn
C(6)-N(2)-C(7)
O(1)-C(1)-Mn
O(3)-C(3)-Mn
N(3)-C(4)-Mn
N(1)-C(5)-C(4)
C(4)-C(5)-C(23)
C(4)-C(5)-Mn
N(2)-C(6)-Mn
88.5(6)
91.5(5)
C(2)-Mn-C(1)
C(2)-Mn-C(6)
C(1)-Mn-C(6)
C(3)-Mn-C(4)
C(6)-Mn-C(4)
C(3)-Mn-N(1)
C(6)-Mn-N(1)
C(2)-Mn-C(5)
C(1)-Mn-C(5)
C(4)-Mn-C(5)
C(5)-N(1)-C(15)
C(15)-N(1)-Mn
C(4)-N(3)-C(30)
O(2)-C(2)-Mn
N(3)-C(4)-C(5)
C(5)-C(4)-Mn
N(1)-C(5)-C(23)
N(1)-C(5)-Mn
C(23)-C(5)-Mn
93.8(6)
86.1(5)
95.7(5)
89.6(5)
85.0(5)
4.242 mm-1
2736
171.3(5)
102.7(6)
163.5(6)
167.1(5)
99.1(5)
64.4(4)
90.9(5)
88.4(4)
30.0(3)
0.20 × 0.20 × 0.10 mm
3.83-60.02°
0 < h < 20, 0 < k < 17, 0 < l < 25
5183
91.0(5)
92.8(4)
4988
137.1(5)
129.1(5)
34.4(4)
126.1(10)
137.8(7)
118.8(10)
179.1(13)
119.4(10)
93.7(8)
full-matrix least-squares on F2
no. of data/restraints/params 4976/0/407
goodness-of-fit on F2
final R indices [I > 2σ(I)]
R indices (all data)
1.028
R1 ) 0.0882, wR2 ) 0.1925
R1 ) 0.2871, wR2 ) 0.2975
0.00042(7)
95.3(7)
171.2(13)
176.2(12)
177.2(13)
146.8(10)
106.5(10)
127.3(11)
51.9(6)
extinction coeff
largest diff peak and hole
0.350 and -0.616 e Å-3
Crystal Data and Structure Refinement for 6
empirical formula
fw
temp
wavelength
cryst syst
space group
unit cell dimens
C38H33ClMnN3O4
686.06
296(2) K
0.71073 Å
monoclinic
126.2(11)
54.7(6)
178.6(8)
173.6(11)
room-temperature 1H NMR spectrum of 3 contains a
multiplet from 7.21 to 6.92 ppm due to the phenyl
protons. An AB quartet is observed at 3.59 ppm (J )
12.25 Hz) due to the diastereotopic methylene protons.
The methyl region contains three sharp signals at 2.25,
2.18, and 2.07 ppm. Very broad peaks are also evident
between 2.40 and 2.10 ppm. The broadening suggests
that a dynamic process is occurring involving some of
the xylyl methyl groups at room temperature, vide infra.
Complex 3 contains three xylyl isocyanides. One is
in a terminal position on the metal, contains C(6) (from
the crystal structure solution), and is labeled I in Figure
1. The other two xylyl groups are labeled as II and III.
Xylyl II is part of the iminoacyl group in which the
carbon is labeled as C(4) and the nitrogen is not directly
included in the metallacycle. Xylyl III is part of the
iminoacyl in which both the carbon C(5) and the
nitrogen N(1) are incorporated into the manganacycle.
Variable-temperature 1H NMR gives some insight
into the dynamic process that these xylyl groups un-
dergo and the assignments of their methyl groups.
P21/n
a ) 10.317(1) Å; R ) 90°
b ) 19.602(1) Å; â ) 99.89(1)°
c ) 17.113(1) Å; γ ) 90°
3409.4(3) Å3, 4
1.337 Mg/m3
volume, Z
density (calcd)
abs coeff
F(000)
cryst size
θ range for data collection
limiting indices
0.510 mm-1
1424
0.50 × 0.25 × 0.10 mm
1.59-28.31°
-13 < h < 11, -21 < k < 25,
-12 < l < 22
no. of reflns collected
no. of indep reflns
refinement method
16370
7947 (Rint ) 0.0505)
full-matrix least-squares on F2
no. of data/restraints/params 6869/0/424
goodness-of-fit on F2
1.035
final R indices [I > 2σ(I)]
R indices (all data)
largest diff peak and hole
R1 ) 0.0543, wR2 ) 0.0930
R1 ) 0.1463, wR2 ) 0.1269
0.254 and -0.233 e Å-3
Crystal Data and Structure Refinement 12
empirical formula
fw
temp
wavelength
cryst syst
C41H30Cl2Mn2N2O10
891.45
296(2) K
1.54184 Å
triclinic
P1h
1
space group
unit cell dimens
Figure 2 shows the H NMR methyl region at various
a ) 11.841(2) Å; R ) 97.18(3)°
b ) 12.160(2) Å; â ) 105.32(3)°
c ) 15.940(3) Å; γ ) 98.40(3)°
2157.4(7) Å3, 2
1.372 Mg/m3
temperatures. At various temperatures overlap of sig-
nals occurs; however the -30 °C spectrum begins to
show all five resonances that are resolved at lower
temperatures. The -30 °C spectrum contains signals
at 2.37, 2.27, 2.14, 2.13, and 2.06 ppm. There is some
overlap of the 2.14 and 2.13 peaks. The integration of
these peaks is 1:1:2:1:1, respectively, which is appropri-
ate for six methyl groups. As the temperature increases,
two peaks (formerly located at 2.37 and 2.13 ppm)
broaden until coalescence occurs at approximately 30
°C. The exact coalescence temperature is difficult to
volume, Z
density (calcd)
abs coeff
F(000)
cryst size
θ range for data collection
limiting indices
6.375 mm-1
908
0.25 × 0.20 × 0.15 mm
2.0-58.0°
0 < h < 12, -13 < k < 13,
-17 < l < 16
no. of reflns collected
no. of indep reflns
refinement method
6421
5881 (Rint ) 0.0165)
full-matrix least-squares on F2
no. of data/restraints/params 5881/0/518
goodness-of-fit on F2
final R indices [I > 2σ(I)]
R indices (all data)
0.961
(24) (a) Bright, D.; Mills, O. S. J . Chem. Soc. 1974, 219. (b) Sarapu,
A. C.; Fenske, R. F. Inorg. Chem. 1972, 11, 3021. (c) Ericsson, M.-S.;
J agner, S.; Ljungstro¨m, E. Acta Chem. Scand., Ser. A 1980, A34, 535.
(d) Ericsson, M.-S.; J agner, S.; Ljungstro¨m, E. Acta Chem. Scand., Ser.
A 1979, A33, 371.
R1 ) 0.0656, wR2 ) 0.1376
R1 ) 0.1226, wR2 ) 0.1619
0.00042(11)
extinction coeff
largest diff peak and hole
0.314 and -0.250 e Å-3

5598 Organometallics, Vol. 18, No. 26, 1999
Becker et al.
sidered as relative but not absolute. The total time
interval studied in each simulation was 50 ps. The
simulation found the torsion angle of xylyl I (defined
as C4-Mn-C7-C12) varies at all temperatures, indi-
cating rapid rotation. The torsion angle of xylyl II (C4-
N3-C30-C35) is essentially fixed at low temperatures,
and rotation frequency increases as the temperature
increases. The simulation at various temperatures
suggests the torsion angle of xylyl III (Mn-N1-C15-
C16) over time does not change, suggesting that xylyl
III does not rotate even at higher temperatures.
The dynamics simulation agrees with the observed
1
variable-temperature H NMR signals from the methyl
groups on the xylyl isocyanide and assists in their
assignments. The freely rotating group, xylyl I, is on
the terminal isocyanide and corresponds to the room-
temperature 1H NMR signal at 2.18 ppm. The other two
xylyl ligands are incorporated into the manganacycle.
Xylyl II is the group in which the iminoacyl carbon is
found in the four-membered metallacycle and the ni-
trogen is not located in the ring. The frequency of
rotation of xylyl II increases as the temperature in-
creases and therefore fits the observed room-tempera-
1
ture H NMR broad peak at 2.24 ppm. Both the carbon
and the nitrogen of xylyl III are located in the ring of
the metallacycle. Xylyl III has essentially no rotation
at tested temperatures and corresponds to the room-
temperature 1H NMR signals at 2.25 and 2.07 ppm.
These results indicate that the manganacycle restricts
the rotation of the included xylyl groups.
1
F igu r e 2. Variable-temperature H NMR for 3.
determine due to the overlap of other methyl signals
complicating the spectrum. A new peak begins to form
at 2.24 ppm in the 55 °C spectrum. A ∆Gq of 61.5 kJ /
mol for the dynamic process leading to the methyl
exchange was calculated on the basis of the Eyring
equation.25 The rate constant used in the calculation of
∆Gq was approximated from the change in the line
widths of an NMR peak in the slow exchange region
(-50 to 20 °C). The other three peaks remain essentially
unchanged. There are slight shifts in some of the peaks
of the nonexchanging methyls in the 1H NMR due to
the overlap with the dynamic signals.
These results indicate that there are three distinct
xylyl groups. One exhibits fast exchange on the NMR
time scale at all temperatures. This gives rise to the
signal at 2.14 ppm at -30 °C. In another xylyl group,
methyl exchange does not occur to any significant
extent; this group gives rise to two distinct methyl
signals at all temperatures (2.27 and 2.06 ppm at -30
°C). The third xylyl group exhibits slow exchange at
lower temperatures and increasingly rapid exchange by
rotation at higher temperatures (2.37 and 2.13 ppm at
-30 °C and 2.24 ppm at 55 °C). The exchange mecha-
nism is presumably rotation around the N-C(ipso)
bond.
Another bis(iminoacyl) complex, 4, is isolated in trace
amounts from the PdO-catalyzed reaction in toluene.
The spectroscopic data are consistent with the proposed
structure, which contains a “dangling” bis(iminoacyl)
group. A complex similar to 4 was isolated by Motz10
and co-workers. They isolated a manganese tetracar-
bonyl with a bulky terminal tert-butyl isocyanide and a
“dangling” bis(iminoacyl) ligand formed from p-tolyl
isocyanides. The spectroscopic data for each “dangling”
bis(iminoacyl) are similar.
1
The H NMR spectrum of 4 contains a multiplet due
to aromatic protons, a singlet at 3.51 ppm from the
benzylic protons, and singlets at 2.31, 2.11, and 2.01
ppm from the xylyl methyl groups. Unlike 3, all methyl
signals of 4 are sharp at room temperature, suggesting
free rotation around the N-C bonds. Also unlike the
1
1H NMR of 3, the benzylic protons in the H NMR of 4
give rise to a singlet. This is consistent with the
existence of a plane of symmetry in 4 containing the
metal, two carbonyls, the terminal isocyanide, and the
dangling ligand coupled with free rotation around the
C-C and C-N bonds; thus, the methylene protons are
equivalent by NMR.
Another product, isolated as a colorless solid, is the
cyclic carbene 5, the xylyl isocyanide analogue of C that
was isolated by Motz and co-workers.10 The spectral
data of 5 compare favorably with data previously
reported for C. Some features include an N-H signal
in the IR spectrum at 3296 cm-1, a broad peak at 7.04
To determine which xylyl groups in 3 exhibit fast
rotation, no rotation, and rotation of varying frequency,
molecular dynamics simulations were performed at
several temperatures (see Experimental Section for
simulation details). Solvent considerations were not
taken into account in the molecular simulations. The
temperatures for the dynamics studies should be con-
1
ppm in the H NMR spectrum arising from N-H, and
a peak at 6.40 ppm from the olefinic proton.
Also isolated from the reaction is the green solid 6.
The green color is in stark contrast to the other Mn(I)
species containing isocyanide and/or iminoacyl ligands,
(25) Gu¨nther, H. NMR Spectroscopy, 2nd ed.; J ohn Wiley and
Sons: Chichester, England, 1995; p 342.

CNR and CO Insertion Reactions
Organometallics, Vol. 18, No. 26, 1999 5599
Ta ble 3. Selected Bon d Dista n ces [Å] a n d An gles
[d eg] for 6
Mn-C(1)
Mn-C(3)
Mn-C(5)
O(1)-C(1)
O(3)-C(3)
N(1)-C(7)
N(2)-C(5)
C(5)-C(6)
C(7)-C(24)
1.819(4)
1.844(4)
2.067(3)
1.144(3)
1.144(3)
1.293(4)
1.274(4)
1.519(4)
1.505(4)
Mn-C(4)
Mn-C(2)
Mn-N(1)
O(2)-C(2)
O(4)-C(4)
N(1)-C(31)
N(3)-C(6)
C(6)-C(7)
1.831(4)
1.860(4)
2.082(2)
1.134(4)
1.145(4)
1.461(4)
1.273(4)
1.491(4)
C(1)-Mn-C(4)
C(4)-Mn-C(3)
C(4)-Mn-C(2)
C(1)-Mn-C(5)
C(3)-Mn-C(5)
C(1)-Mn-N(1)
C(3)-Mn-N(1)
C(5)-Mn-N(1)
C(7)-N(1)-Mn
O(1)-C(1)-Mn
O(3)-C(3)-Mn
N(2)-C(5)-C(6)
C(6)-C(5)-Mn
N(3)-C(6)-C(5)
N(1)-C(7)-C(6)
89.2(2)
C(1)-Mn-C(3)
89.58(14)
86.6(2)
94.1(2)
84.34(13)
87.84(13)
90.88(12)
93.21(12)
94.41(14) C(1)-Mn-C(2)
170.43(14) C(3)-Mn-C(2)
98.68(13) C(4)-Mn-C(5)
171.62(13) C(2)-Mn-C(5)
178.86(12) C(4)-Mn-N(1)
91.55(12) C(2)-Mn-N(1)
80.19(11) C(7)-N(1)-C(31) 120.3(3)
119.0(2)
175.5(3)
177.2(3)
113.5(3)
109.8(2)
129.9(3)
114.3(3)
C(31)-N(1)-Mn
O(2)-C(2)-Mn
O(4)-C(4)-Mn
N(2)-C(5)-Mn
N(3)-C(6)-C(7)
C(7)-C(6)-C(5)
120.7(2)
175.8(3)
177.3(3)
136.6(2)
115.7(3)
114.2(3)
F igu r e 3. ORTEP diagram of 6.
2 is produced from the reduction of the starting material
by added isocyanide, as discussed earlier. The proposed
mechanism for the formation of the isocyanide insertion
products 3-6 is shown in Scheme 1.
N(1)-C(7)-C(24) 125.2(3)
C(6)-C(7)-C(24) 120.6(3)
which are generally yellow, red, or colorless. The green
color is likely a result of increased conjugation within
the complex. The spectral data are consistent with the
tris(iminoacyl) complex pictured in Scheme 1. This is,
to our knowledge, the first isolated octahedrally based
tris(iminoacyl) complex. The IR has four strong bands
between 2068 and 1954 cm-1, consistent with a tetra-
carbonyl complex. The 1H NMR indicates aromatic,
methylene, and methyl protons in the expected regions
with proper ratios.
From the spectroscopic data it is unclear whether 6
is a five-membered metallacycle (a possible resultant
from ring closure by the imino nitrogen of the first
inserted isocyanide) or a four-membered metallacycle
(a possible resultant from ring closure by the imino
nitrogen of the second inserted isocyanide). Previous
work suggested that the probable structure was 6. First,
the square-planar Pd and Ni tris(iminoacyl) complexes
exhibited a structure that contains five atoms in the
metallacycle.26-28 Second, manganese five-membered
rings are less constrained than four-membered rings in
pseudo-octahedral complexes, allowing for an N-Mn-C
angle closer to the 90°.
A single-crystal X-ray structure solution was per-
formed to confirm the structure of 6. Crystallographic
parameters can be found in Table 1, and selected bond
angles and distances are recorded in Table 3. An ORTEP
diagram of 6 is shown as Figure 3. This pseudo-
octahedral complex has an angle of 84.34(13)° for C4-
Mn-C5, which is slightly larger than those found in
five-membered bis(iminoacyl) metallacycles of palla-
dium and nickel (80.3-81.3°).26-28 All other angles and
distances are unremarkable.
The initial step is likely the PdO-catalyzed substitu-
tion of a carbonyl by an isocyanide to create intermedi-
ate 7, which is neither isolable nor spectroscopically
observable. Rapid migration of the benzyl group to the
isocyanide carbon and addition of a second equivalent
of xylyl isocyanide leads to the mono(iminoacyl) inter-
mediate 8 (which also is neither observed nor isolated).
From this intermediate, two pathways are possible.
Product 5 is formed by attack on the polarized terminal
isocyanide carbon by the imino nitrogen lone pair
followed by migration of a proton from the benzyl carbon
to the isocyanide nitrogen. This pathway to a cyclic
carbene is also proposed in the p-tolyl isocyanide
chemistry previously reported.10 If intermediate 8 in-
serts the second added isocyanide and the vacant
coordination site is taken by a third isocyanide, the
“dangling” bis(iminoacyl) 4 is produced. Then, a third
isocyanide insertion followed by ring closure by the
imino nitrogen lone pair leads to the green solid 6. If a
terminal carbonyl is displaced in 4 by the ring closure
from the imino nitrogen coordination to the manganese,
manganacycle 3 is created. Other studies within our
group indicate that 3 converts to 6 stirring under a CO
atmosphere.29
The observed reactivity and proposed mechanism
with xylyl isocyanide are similar in some aspects to the
chemistry found with p-tolyl isocyanide (earlier de-
scribed in the Introduction), although different in
significant other aspects.10 Both aromatic isocyanides
form cyclic carbenes (5 and C) from their reactions with
1. The p-tolyl cyclic carbene C rearranges to the tetra-
carbonyl bis(iminoacyl)metallacycle B. However the
xylyl cyclic carbene 5 does not undergo this transforma-
tion and is stable over 7 days at 50 °C in CDCl3. The
reason for this is unclear.
The mixture produced from the reaction depicted in
Scheme 1 in toluene contains products 2-6. The dimer
The major product from the p-tolyl reaction is the
tetracarbonyl bis(iminoacyl) B. The xylyl isocyanide
analogue of B is not observed. Instead, three other
(26) Yamamoto, Y.; Tanase, T.; Yanai, T.; Asano, T.; Kobayashi, K.
J . Organomet. Chem. 1993, 456, 287.
(27) Carmona, E.; Mar´ın, J . M.; Palma, P.; Poreda, M. L. J .
Organomet. Chem. 1989, 377, 157.
(28) Tanase, T.; Ohizumi, T.; Kobayashi, K.; Yamamoto, Y. Orga-
nometallics 1996, 15, 3404.
(29) Homrighausen, C. L.; Alexander J . J . Unpublished results.

5600 Organometallics, Vol. 18, No. 26, 1999
Becker et al.
Sch em e 2
products are isolated: 3, 4, and 6. The proposed mech-
anisms of the reaction of 1 with p-tolyl and xylyl
isocyanide both include an intermediate, which contains
one inserted isocyanide and one terminal isocyanide (8
in Scheme 1). p-Tolyl isocyanide undergoes a second
insertion followed by rapid coordination of the imino
nitrogen to the manganese to form the tetracarbonyl
manganacycle B. Xylyl isocyanide also undergoes an
insertion of a second isocyanide; however, instead of
coordination of the imino nitrogen to the vacant coor-
dination site on the metal, a third equivalent of xylyl
isocyanide fills the vacant manganese site to produce
4. Apparently, the bulk of the xylyl isocyanide slows the
attack by the imino nitrogen on the metal and, instead,
the coordination of a free isocyanide is favored. If a
carbonyl in 4 is displaced by ring closure, then the
metallacycle 3 is formed. Alternatively, 4 can insert a
third isocyanide followed by ring closure to form the tris-
(iminoacyl) manganacycle 6. Surprisingly, this third
insertion is found only with the larger xylyl isocyanide.
A third insertion of isocyanide is favored for xylyl
isocyanide presumably because coordination of a third
equivalent of isocyanide is faster than ring closure
which forms bis(iminoacyl) metallacycles.
The IR spectrum of 9 displays two bands at 2161 and
2124 cm-1 from the symmetric and antisymmetric Ct
N stretches, typical facial tricarbonyl absorptions, and
a broad stretch at 1628 cm-1 characteristic of a metal
acyl. The spectroscopic results are similar to those for
the tert-butyl isocyanide analogue of 9.16
The proposed mechanism of this reaction is depicted
in Scheme 2. The key step that differentiates this
reaction from the one in toluene is the equilibrium
between 1 and 10, where S is the coordinating solvent
THF. This equilibrium lies much further to the left in
toluene. This solvent-dependent equilibrium has been
proposed for other carbonylations on manganese and
iron.3 The labile solvent ligand in the acyl 10 is rapidly
replaced with xylyl isocyanide, producing 11. This
complex is not isolated because, in the presence of PdO,
a carbonyl is rapidly substituted by a second equivalent
of isocyanide, producing the stable fac-tricarbonyl 9. It
is unlikely that substitution occurs before carbonylation.
It has been observed (see Scheme 1) that, if terminal
carbonyls and a terminal aromatic isocyanide are
present, isocyanide insertion is favored over carbony-
lation. Hence, if substitution were the first step, imi-
noacyls (not an acyl) would form. Iminoacyl products
might be expected to result from the decarbonylation
of 11. However, no products of this type were observed
in this room-temperature reaction.
2.2. Rea ction in THF in th e P r esen ce of P d O. The
same reaction performed in THF (Scheme 2) affords
different products. The only product identified and
isolated from the THF reaction is the disubstituted acyl
(CO)3(xylyl-NC)2MnC(O)CH2C6H4-p-Cl, 9, in 29% yield.
2.3. Rea ction w ith ou t P d O in THF . Reaction of 1
with xylyl isocyanide in THF (without PdO) was also
investigated. As earlier indicated, a number of other

CNR and CO Insertion Reactions
Organometallics, Vol. 18, No. 26, 1999 5601
isocyanides react with 1 under these conditions to
produce products of carbonylation with the addition of
a terminal isocyanide p-X-C6H4CH2C(O)Mn(CO)4(CNR).
A 1:1 ratio of xylyl isocyanide and 1 was stirred at room
1
temperature. The IR spectrum and the H NMR spec-
trum of the reaction mixture both indicate that three
complexes are present. One complex isolated from the
reaction mixture is unreacted starting material 1.
Another product is the expected manganese tetracar-
bonyl acyl with a terminal xylyl isocyanide 11. An
additional isolated complex is the bis(iminoacyl) met-
allacycle 3, which contains a terminal xylyl isocyanide.
By employing a 1:1 ratio of 1:xylyl isocyanide, one would
expect leftover 1 if the metallacycle 3 (which requires 3
equiv of xylyl isocyanide) is produced. The formation of
a bis(iminoacyl) manganacycle in this reaction has
precedent. Reaction of 1 with p-tolyl isocyanide also
produces a bis(iminoacyl) minor product.21 However, 3
is a bis(iminoacyl) tricarbonyl complex with a terminal
isocyanide, while the p-tolyl bis(iminoacyl) is the tet-
racarbonyl complex, C.
The ratio of species in the reaction mixture was
determined by 1H NMR. The methylene proton peak
areas for each complex were determined and their ratios
calculated. This integration determined that the product
mixture consists of 44% 1, 34% 11, and 22% 3. This ratio
of species and the IR spectrum indicate that 100% of
the isocyanide has been consumed in the formation of
11 and 3 (1 mol of xylyl-NC in each mole of 11 ) 0.34
equiv and 3 mol of xylyl-NC in each mole of 3 ) 3 ×
0.22 ) 0.66 equiv). Although significant amounts of both
11 and 3 are present in the reaction mixture, purifica-
tion by column chromatography and recrystallization
resulted in low isolated yields of pure products 11 and
3 (2% and 11%, respectively). The spectroscopic data are
appropriate for complex 11 and are consistent with
those of other manganese acyls previously reported.12
Other xylyl isocyanide insertion products (such as 6)
apparently are not present in the reaction in THF
without PdO (or are present in trace amounts) as they
are not resolved from the 1H NMR baseline of the
reaction mixture. In THF, a dissociative CO substitution
(via ring closure) is apparently favored to produce the
bis(iminoacyl) 3 over the insertion of a third isocyanide
to produce the tris(iminoacyl) 6, which was obtained in
toluene. A dissociative mechanism for carbonyl substi-
tution has been previously shown for several Mn(I)
complexes.30 It is reasonable to suggest that in polar
THF, the intermediate or transition state for dissocia-
tive substitution of CO may be more stable than in
toluene, thus affording only 3 in THF. The bulk of the
isocyanide may also contribute to a smaller rate for the
third insertion than that for ring closure. The barrier
seems not to be thermodynamic since 6 is stable in THF
solution.
F igu r e 4. ORTEP diagram of 12.
nese centers are linked through a carbamato bridge. The
dimer contains two inserted isocyanides, one on each
metal center. A five-membered manganacycle is formed
on one metal center. Hydrogen transfer (formally from
the benzylic group attached to the metallacycle) to an
imino nitrogen has also occurred.
The structure of 12 has been established by single-
crystal X-ray analysis. An ORTEP diagram of 12 is
shown in Figure 4. Crystal and diffraction data are
listed in Table 1, and bond angles and distances are
listed in Table 4.
Complex 12 contains a five-member metallacycle
consisting of Mn(1)-C(9)-N(10)-C(11)-O(12). The mean
deviation from this plane is 0.0416 Å; the dihedral angle
between this plane and the plane defined by Mn(2)-
O(13)-C(8)-C(6)-C(29) is 17°. The coordination around
Mn(1) is distorted octahedral. The C(9)-Mn(1)-O(12)
is 80.2°, which is slightly larger than the average angle
of 77.8° found in other reported five-membered manga-
nacycles.10 The coordination around Mn(2) is also dis-
torted octahedral. The carbonyl ligands coordinated to
both metals are linear and are unremarkable in their
bond angles and distances. The Mn(1)-Mn(2) distance
is 5.32 Å, indicating that no metal-metal bond is
present.
The C(9)-C(14) bond length is 1.346(8) Å, indicative
of a C-C double bond. The C(29)-N(30) bond length is
1.285(7) Å, the C(11)-N(10) bond length is 1.354(7) Å,
and the C(9)-N(10) bond length is 1.443(7) Å, indicating
a double bond, a bond that is between single bond and
double bond in character, and a single bond, respec-
tively. The C(11)-N(10) bond length in the carbamate
is similar to that observed in tungsten and tantalum
complexes.31 The hydrogen atom, H(30), was located
directly from the difference electron maps, and the
N(30)-H(30) distance is 0.942 Å. The C(29)-C(39) bond
length of 1.518(8) Å is as expected for a C-C single
bond. The Mn(1)-O(12) bond distance is 2.031(4) Å, and
the Mn(2)-O(13) bond length is 2.043(4) Å. These
Mn-O lengths are within the range of bond lengths
reported for Mn-O(carbonate)32 and Mn-O(carboxy-
late)33 bonds (2.020-2.042 Å) found trans to carbonyl
ligands. The bridging carbamato C-O bond lengths are
During the workup of the reaction mixture, an
unexpected yellow crystalline solid, 12, was isolated in
6% yield. This product was not in the original reaction
mixture and so was formed during the workup, which
was not performed under an argon atmosphere. Com-
plex 12 is a manganese dimer with each manganese
center containing four terminal carbonyls. The manga-
(31) (a) Chisholm, M. H.; Extine, M. W. J . Am, Chem. Soc. 1974,
96, 6214. (b) Chisholm, M. H.; Extine, M. W. J . Am. Chem. Soc. 1977,
99, 792.
(30) (a) Atwood, J . D.; Brown, T. L. J . Am. Chem. Soc. 1975, 97,
3380. (b) Atwood, J . D.; Brown, T. L. J . Am. Chem. Soc. 1976, 98, 3155.
(32) Mandal, S. K.; Ho, D. M.; Orchin, M. Organometallics 1993,
12, 1714.

5602 Organometallics, Vol. 18, No. 26, 1999
Becker et al.
Ta ble 5. Rea ction Con d ition s Em p loyed to
Ta ble 4. Selected Bon d Dista n ces [Å] a n d An gles
[d eg] for 12
Elu cid a te th e Mech a n ism of F or m a tion of 12
Mn(1)-C(3)
Mn(1)-C(4)
Mn(1)-O(12)
Mn(2)-C(6)
Mn(2)-C(8)
Mn(2)-O(13)
O(1)-C(1)
1.787(8)
1.839(10)
2.031(4)
1.778(7)
1.849(8)
2.043(4)
1.124(8)
1.149(8)
1.125(8)
1.126(8)
1.264(7)
1.354(7)
1.444(7)
1.446(8)
Mn(1)-C(2)
Mn(1)-C(1)
Mn(1)-C(9)
Mn(2)-C(5)
Mn(2)-C(7)
Mn(2)-C(29)
O(2)-C(2)
1.821(8)
1.871(9)
2.074(6)
1.842(8)
1.874(9)
2.046(6)
1.137(8)
1.131(9)
1.156(7)
1.155(8)
1.278(7)
1.443(7)
1.285(7)
1.346(8)
O(3)-C(3)
O(4)-C(4)
O(5)-C(5)
O(7)-C(7)
O(12)-C(11)
N(10)-C(11)
N(10)-C(21)
N(30)-C(31)
O(6)-C(6)
O(8)-C(8)
O(13)-C(11)
N(10)-C(9)
N(30)-C(29)
C(9)-C(14)
C(3)-Mn(1)-C(2)
C(2)-Mn(1)-C(4)
C(2)-Mn(1)-C(1)
C(3)-Mn(1)-O(12)
C(4)-Mn(1)-O(12)
C(3)-Mn(1)-C(9)
C(4)-Mn(1)-C(9)
O(12)-Mn(1)-C(9)
C(6)-Mn(2)-C(8)
C(6)-Mn(2)-C(7)
C(8)-Mn(2)-C(7)
C(5)-Mn(2)-O(13)
C(7)-Mn(2)-O(13)
C(5)-Mn(2)-C(29)
C(7)-Mn(2)-C(29)
89.6(3) C(3)-Mn(1)-C(4)
93.0(4) C(3)-Mn(1)-C(1)
94.0(4) C(4)-Mn(1)-C(1)
173.6(3) C(2)-Mn(1)-O(12)
91.8(3) C(1)-Mn(1)-O(12)
99.5(3) C(2)-Mn(1)-C(9)
84.1(3) C(1)-Mn(1)-C(9)
80.7(2) C(6)-Mn(2)-C(5)
88.5(3) C(5)-Mn(2)-C(8)
92.5(4) C(5)-Mn(2)-C(7)
91.6(3) C(6)-Mn(2)-O(13)
87.9(3) C(8)-Mn(2)-O(13)
86.9(3) C(6)-Mn(2)-C(29)
84.6(3) C(8)-Mn(2)-C(29)
90.9(3) O(13)-Mn(2)-C(29)
94.5(4)
88.3(3)
172.4(3)
90.5(3)
85.3(3)
170.7(3)
88.5(3)
92.5(4)
92.8(3)
173.5(3)
176.1(3)
95.4(2)
93.1(3)
177.0(3)
83.1(2)
The product mixtures from H2O, CO2, and air added
to the reactants had the same composition as the
reaction mixture obtained under an argon atmosphere
as determined by IR spectroscopy. This indicates that
none of these species react with the starting materials
to give 12. However, if the product mixture is placed
under a carbon dioxide atmosphere over days, the yield
of 12 increases significantly, from 6% to 33%. This
indicates that the dimer is formed slowly and that CO2
is required. The IR spectrum after the CO2 exposure
indicates that all of the acyl 11 has been consumed,
suggesting that 11 is involved in the production of 12.
Also, the yield of 12 is similar to the yield of 11
calculated from the 1H NMR of the initial reaction
mixture prior to workup. From these results the mech-
anism for the formation of 12 is proposed in Scheme 3.
C(11)-O(12)-Mn(1) 114.5(4) C(11)-O(13)-Mn(2) 129.1(4)
C(11)-N(10)-C(9)
C(9)-N(10)-C(21)
O(1)-C(1)-Mn(1)
O(3)-C(3)-Mn(1)
O(5)-C(5)-Mn(2)
O(7)-C(7)-Mn(2)
C(14)-C(9)-N(10)
N(10)-C(9)-Mn(1)
118.3(5) C(11)-N(10)-C(21) 119.5(5)
122.2(5) C(29)-N(30)-C(31) 129.6(6)
177.2(8) O(2)-C(2)-Mn(1)
174.5(7) O(4)-C(4)-Mn(1)
175.9(7) O(6)-C(6)-Mn(2)
175.4(8) O(8)-C(8)-Mn(2)
116.3(5) C(14)-C(9)-Mn(1)
175.8(8)
178.0(8)
178.1(8)
174.5(6)
136.9(5)
106.7(4) O(12)-C(11)-O(13) 122.9(6)
O(12)-C(11)-N(10) 119.1(6) O(13)-C(11)-N(10) 118.0(6)
C(9)-C(14)-C(15)
128.8(6) N(30)-C(29)-C(39) 116.2(6)
N(30)-C(29)-Mn(2) 117.2(5) C(39)-C(29)-Mn(2) 126.4(5)
C(40)-C(39)-C(29) 115.1(6)
1.264(7) and 1.278(7) Å for C(11)-O(12) and C(11)-
O(13), respectively. These C-O bond lengths are shorter
than expected for a C-O single bond (∼1.34 Å) and
longer than expected for a C-O double bond (∼1.20 Å).
The bond distances in the bridging carbamate suggest
delocalization throughout the carbamate. The spectro-
scopic data are consistent with the structure of 12,
including the IR spectrum, which shows multiple ter-
minal carbonyl bands from 2095 to 1936 cm-1 and a
band at 1554 cm-1 from the CdN stretch.
Since 12 (containing a bridging carbamate) is a very
unexpected product based on the known Mn chemistry,
it is of interest to determine the mechanism of its
formation. Complex 12 was not present in the initial
reaction mixture. It was formed when the reaction
mixture in a dichloromethane/hexane solution was
cooled in dry ice over several days. A number of
experiments were carried out to explore the mechanism
of formation of 12. The question is the source of the
“extra” O in the µ-carbamate. The original reaction was
performed under an inert atmosphere, while the workup
was not. The reaction conditions investigated are sum-
marized in Table 5.
The first step is the formation of 11. Over time, 11
decarbonylates and isocyanide insertion occurs, forming
a coordinatively unsaturated mono(iminoacyl) interme-
diate 13. Manganese mono(iminoacyl) complexes have
been proposed as highly reactive intermediates in other
systems.11 This mono(iminoacyl) intermediate, 13, re-
acts via attack on electrophilic CO2 by the imino
nitrogen lone pair giving 14, which is not spectroscopi-
cally observable. Intermolecular attack by 14 on the
metal center of another unsaturated mono(iminoacyl)
intermediate, 13, occurs followed by an intramolecular
attack of the manganese center by the oxygen of CO2,
thus closing the ring. Intermolecular attack by the CO2
oxygen may occur before intramolecular attack; other-
wise one would expect the isolation of a monomeric
kinetic product, a carbamate metallacycle, 15. It may
be that intramolecular attack does occur reversibly
giving 15, but that only proton transfer leads to a
thermodynamically stable product, the dimer 12. Pre-
sumably, 12 is not formed in the reaction conducted in
the presence of CO2 because 13 reacts faster with
isocyanide rather than with CO2. Currently, efforts in
our lab are ongoing using manganese acyl complexes
with various aromatic isocyanides to investigate their
reactions with CO2, CS2, and other cumulenes.
(33) (a) Herrmann, W. A.; Schweizer, I.; Skell, P. S.; Ziegler, M. L.;
Wiedenhammer, K.; Nuber, B. Chem. Ber. 1979, 112, 2423. (b) Cotton,
F. A.; Darensbourg, D. J .; Kolthammer, B. W. S. Inorg. Chem. 1981,
20, 1287.

CNR and CO Insertion Reactions
Organometallics, Vol. 18, No. 26, 1999 5603
Sch em e 3
downfield from tetramethylsilane. Elemental analyses were
performed by Chemisar Laboratories, Guelph, Canada. Mass
spectra were recorded on a VG 30-250 quadrupole mass
spectrometer with an Ion Tech saddle field ion (FAB) gun. The
gun was operated employing argon as the bombarding gas with
an ion current of 1 mA and an ion gun potential of 8 kV. The
typical matrix was 3-nitrobenzyl alcohol. The instrument was
scanned for positive ions from 25 to 1200 Da with a scan time
of 4.9 s/scan. Melting points were determined on a Mel-Temp
melting point apparatus and are uncorrected.
All solvents were dried prior to use employing standard
techniques. PdO was purchased from Aldrich and used as
received. Dimanganese decacarbonyl was purchased from
Strem Chemical Co. and used as received. Xylyl isocyanide
was purchased from Fluka and used without further purifica-
tion. (CO)5MnCH2C6H4-p-Cl, 1, was prepared according to the
published method.12
3.1. Gen er a l Con d ition s for Rea ction s in th e P r esen ce
of P d O. Complex 1 and xylyl isocyanide (1:2 mol ratio) were
placed in a flask. A catalytic amount of PdO (approximately
0.02 g) was added. Toluene was injected into the flask, and
the solution was stirred at least overnight. The color of the
mixture changed from yellow to red. The solvent was removed
in vacuo, and complexes 2-6 were isolated as described
below.
2.4. Su m m a r y. The complex product mixtures, found
in the described reactions with xylyl isocyanide, provide
difficult circumstances in which to isolate the various
products. Alkylmanganesepentacarbonyls reacted with
xylyl isocyanide in the presence of PdO and produced
several products distinct from those isolated from reac-
tion with the smaller p-tolyl isocyanide, including the
first tris(iminoacyl) manganacycle. It was also found
that when a manganese acyl complex containing a
terminal xylyl isocyanide was exposed to CO2, a unique
dimanganese complex, which is linked by a carbamato
bridge, was formed.
3. Exp er im en ta l Section
All reactions were performed under argon using standard
Schlenk techniques34 unless otherwise noted. Infrared spectra
were recorded on a Perkin-Elmer 1600 Series FTIR spectro-
photometer. IR cells were 1.0 mm solution cells with NaCl
windows. 1H and 13C NMR data were recorded using a Bruker
AM-250 spectrometer, and the chemical shifts are given in ppm
(34) Shriver, D. F.; Drezdzon, M. A. The Manipulation of Air-
Sensitive Compounds, 2nd ed.; Wiley: New York, 1986.

5604 Organometallics, Vol. 18, No. 26, 1999
Becker et al.
Isola tion of Mn 2(CO)9(CN-xylyl) (2). Complex 1 (0.822
g, 2.57 mmol), xylyl isocyanide (0.680 g, 5.18 mmol), and PdO
(approximately 0.02 g) were placed in a flask and stirred in
30 mL of toluene for 4 days. A 15 mL sample of the solution
was removed from the flask, and the solvent was removed in
vacuo. A silica gel column was prepared, and the residue was
eluted from 15% THF in hexane. The first band was yellow.
Solvent was removed via rotary evaporation. Hexane was
added and the solution cooled to 0 °C. Complex 2 (0.0441 g,
8.7% yield) was isolated as a yellow solid. The spectroscopic
data compared favorably to that previously reported.20
spectrum: 686 (M+ + 1), 575, (M+ - 4CO). Anal. Found: C,
65.10; H, 5.32. Calcd for C38H33: C, 66.52; H, 4.85.
Isola tion of (CO)3(CN-xylyl)2Mn C(O)CH2C6H4-p-Cl] (9).
The reaction of 1 (0.785 g, 2.46 mmol) with xylyl isocyanide
(0.646 g, 4.92 mmol) was run in 20 mL of THF with PdO (0.02
g). The mixture was stirred for 22 h, and the solvent was
removed in vacuo. Two chromatographies were performed. The
first was on silica gel with 5% THF in hexane. An orange-
yellow fraction was rechromatographed on silica gel with 25%
hexane in THF, affording an orange band. The solvent was
removed, hexane/CH2Cl2 was added, and the mixture was
cooled to 0 °C. Yellow solids of 9 (0.3067 g, 29% yield) were
isolated.
Sp ectr a l d a ta for 9. IR (THF), cm-1: CtN 2161 (m), 2124
(m), CtO 2010 (vs), 1966 (vs), 1944 (vs), CdO 1628 (m, br).
1H NMR (CDCl3), ppm: 7.30-7.00 (m, Ph, 10H), 4.20 (s, CH2,
2H), 2.41 (s, CH3, 12H). 13C NMR (CDCl3), ppm: 137-117
(singlets, Ph), 70.2 (s, CH2), 18.6 (s, CH3), 18.0 (s, CH3). Mass
spectrum: 555 (M+), 471 (M+ - 3CO). Anal. Found: C, 62.47;
H, 4.27. Calcd for C29H24: C, 62.77; H, 4.36.
Isolation of (CO)3(CN-xylyl)Mn [C(dN-xylyl)C(dN-xylyl)-
CH2C6H4-p-Cl] (3). Complex 3 was isolated from reaction
mixtures with and without PdO. 1 (0.822 g, 2.57 mmol) was
stirred in the presence of PdO (approximately 0.02 g) and xylyl
isocyanide (0.680 g, 5.18 mmol) in 30 mL of toluene for 4 days.
The solvent was removed, and the residue was chromato-
graphed on two successive silica gel columns and eluted with
15% THF/hexane. The first red band was isolated from each
column, and the solvent was removed. THF was added, and
the mixture was cooled to 0 °C, affording 3 (0.148 g, 9% yield)
as a red solid. Suitable crystals of 3 for X-ray analysis were
grown from CH2Cl2/hexane in an NMR tube at 0 °C.
Sp ectr a l Da ta for 3. IR (CHCl3): cm-1 CtN 2121 (s), Ct
O 2008 (vs), 1950 (vs), 1925 (vs). 1H NMR (CDCl3), ppm: 7.21-
6.92 (m, Ph, 13H), 3.59 (AB quartet, J ) 12.25 Hz, CH2, 2H),
2.25 (s, Mn-N-xylyl-CH3, 3H), 2.24 (br, Mn-C-N-xylyl-CH3,
6H), 2.18 (s, terminal-xylyl-CH3, 6H), 2.07 (s, Mn-N-xylyl-
CH3, 3H). 13C NMR (CDCl3), ppm: 221.1 (CO), 219.7 (CO),
215.0 (CO), 206.1 (Mn-terminal-CNR), 187.9 (Mn-C), 179.1
(Mn-N-C), 153.1-122.9 (singlets, Ph), 33.5 (CH2), 22.7-18.4
(m, CH3). Mass spectrum: 658 (M+ + 1), 574 (M+ - 3CO).
Anal. Found: C, 67.2; H, 5.18. Calcd for C37H33: C, 67.5; H,
5.05.
3.2. Gen er a l Rea ction Con d ition s w ith ou t P d O. Com-
plex 1 and xylyl isocyanide (1:1 mol ratio) were placed in a
reaction flask. THF was added, and the mixture was allowed
to stir overnight. The color of the solution changed from yellow
to red, and the solvent was removed in vacuo. Isolation of the
greatest yields of complexes 3, 11, and 12 from this reaction
is described below.
Isola tion of fa c-(CO)3(CN-xylyl)Mn [C(dN-xylyl)C(dN-
xylyl)CH2C6H4-p-Cl] (3). Xylyl isocyanide (0.400 g, 3.05
mmol) was stirred with 1 (0.972 g, 3.04 mmol) in 30 mL of
THF for 18 h. After solvent removal, the residue was chro-
matographed on a silica gel column with a 1:9 Et2O/hexane
solution as the eluent. An orange band was isolated and
rechromatographed employing the same conditions. Again, an
orange band was isolated. The solvent was removed, and the
red solid 3 was recrystallized from hexane (0.0773 g, 11.6%
yield). Spectroscopic data were given earlier in this report.
Isola tion of (CO)4(CN-xylyl)Mn C(O)CH2C6H4-p-Cl (11).
Complex 1 (1.51 g, 4.71 mmol) was stirred in 20 mL of THF
with xylyl isocyanide (0.619 g, 4.72 mmol) for 19 h. After
solvent removal, the residue was recrystallized from CH2Cl2/
hexane at 0 °C. Pale yellow needles of 11 (0.037 g, 2.0% yield)
were isolated.
Isola tion of (CO)4(CN-xylyl)Mn [C(dN-xylyl)C(dN-xy-
lyl)-CH2C6H4-p-Cl] (4). The experimental details can be found
above in the preparation of 2. Complex 4 (0.0041 g, 0.5% yield)
was isolated from an orange chromatography band in which
the solvent was removed and orange solids were isolated.
Sp ectr a l Da ta for 4. IR (CDCl3), cm-1: CtN 2112 (s), Ct
O 2016 (s), 1996 (s), 1973 (s), 1949 (vs, br), CdN 1620 (m),
1601 (m). 1H NMR (CDCl3), ppm: 7.16-6.14 (m, Ph, 13H), 3.51
(s, CH2, 2H), 2.31 (s, CH3, 6H), 2.11 (s, CH3, 6H), 2.07 (s, CH3,
6H). No analysis could be obtained due to the extremely small
amount of material isolated.
Sp ectr a l Da ta for 11. IR (THF), cm-1: CtN 2164 (m), Ct
1
O 2065 (s), 2004 (sh), 1985 (vs), 1972 (sh), CdO 1637 (m). H
NMR (CDCl3), ppm: 7.24-7.01 (m, Ph, 7H), 4.15 (s, CH2, 2H),
2.42 (s, CH3, 6H). Anal. Found: C, 56.27; H, 3.40. Calcd for
C21H15: C, 55.83; H, 3.35.
Isola t ion of (CO)4Mn [C{dCH (C6H 4-p -Cl)}N-xylyl)C-
(NH-xylyl)] (5). The experimental details are described above
in the preparation of 2. Complex 5 (0.0252 g, 3.5% yield) was
isolated from the fifth chromatography band. The solvent was
removed and washed with hexane. The residue was recrystal-
lized from CH2Cl2 affording 5 as colorless crystals.
Isola tion of (CO)4Mn [C{dCH(C6H4-p-Cl)}N-(xylyl)C-
(O)OMn (CO)4[C(dNH-xylyl)CH 2C6H4-p-Cl]] (12). A THF
solution of 1 (0.590 g, 1.85 mmol) and xylyl isocyanide (0.242
g, 1.85 mmol) was stirred for 22 h. The solvent was removed
in vacuo. The flask was filled with CO2, and CH2Cl2/hexane
was added via syringe. The sealed flask was cooled to 0 °C
and afforded the yellow solid 12 (0.273 g, 33.1% yield). Crystals
of 12 suitable for X-ray analysis were grown from CH2Cl2/
hexane at 0 °C.
Sp ectr a l Da ta for 12. IR (CHCl3), cm-1: CtO 2095 (m),
2077 (m), 2025 (sh), 2007 (vs), 1988 (s), 1936 (s), CdN 1554
(m). 1H NMR (CDCl3), ppm: 7.2-6.4 (m, Ph, N-H, 15 H), 6.04
(s, CdCH, 1H), 4.01 (s, CH2, 2H), 2.08 (s, CH3, 6H), 1.75 (s,
CH3, 6H). 13C NMR (CDCl3), ppm: 142-116 (singlets, Ph), 50.9
(s, CH2), 17.8 (s, CH3), 17.5 (s, CH3). Mass spectrum: 891 (M+),
778 (M+ - 4CO), 750 (M+ - 5CO), 722 (M+ - 6CO), 666 (M+
- 8CO). Anal. Found: C, 55.5; H, 3.9. Calcd for C41H30: C,
55.2; H, 3.4.
Sp ectr a l Da ta for 5. IR (CHCl3), cm-1: N-H 3296 (w), Ct
1
O 2066 (m), 1983 (vs), 1973 (vs), 1945 (s). H NMR (CDCl3),
ppm: 7.30-7.14 (m, Ph, 10H), 7.04 (s, br, NH, 1H), 6.40 (s,
CdCH, 1H), 2.36 (s, CH3, 6H), 2.34 (s, CH3, 6H). 13C NMR
(CDCl3), ppm: 140-117 (m, Ph), 18.6 (s, CH3), 18.0 (s, CH3).
Mass spectrum: 554 (M+), 443 (M+ - 4CO). Anal. Found: C,
62.21; H, 4.61. Calcd for C29H24: C, 62.77; H, 4.36.
Isolation of (CO)4Mn [C(dN-xylyl)C(dN-xylyl)-CH2C6H4-
p-Cl] (6). A green fraction of the chromatography column
described above in the preparation of 2 was taken, and the
solvent was removed giving 6 (0.062 g, 2.7% yield)) as a green
solid.
Sp ectr a l Da ta for 6. IR (CDCl3), cm-1: CtO 2068 (s), 1991,
1
(vs), 1977 (vs), 1955 (vs). H NMR (CDCl3), ppm: 7.24-6.76
(m, Ph, 13H), 3.94 (s, CH2, 2H), 2.15 (s, CH3, 6H), 2.00 (s, CH3,
6H), 1.88 (s, CH3, 6H). 13C NMR (CDCl3), ppm: 132-122
(singlets, Ph), 18.7 (s, CH3), 18.4 (s, CH3), 17.8 (s, CH3). Mass
3.3. X-r a y Diffr a ction Stu d ies of 3, 6, a n d 12. Crystals
of 3, 6, and 12 suitable for X-ray diffraction were obtained from
CH2Cl2/hexane. For room-temperature X-ray examination and

CNR and CO Insertion Reactions
Organometallics, Vol. 18, No. 26, 1999 5605
data collection, each crystal was coated with a light film of
epoxy resin and mounted on a glass fiber. Intensity data for 3
and 12 were collected using graphite-monochromated Cu KR
radiation, while 6 was analyzed using graphite-monochro-
mated Mo KR radiation.
) 0.5). The solvent contribution from the reflection data was
subtracted out using SQUEEZE,37 and the modified reflection
data were subsequently used in the final stages of refinement.
The three structural refinements converged with crystal-
lographic agreement factors summarized in Table 1.
3.4. Molecu la r Mod elin g of 3. Modeling was carried out
using Insight II (version 95.2) and Discover (version 4.0), both
from Molecular Simulations Inc. The ESFF force field was
used, as it had suitable parameters for the organometallic
complex.38,39
Energy minimization of the complex began with coordinates
obtained from X-ray crystallography. The minimization pro-
tocol was a cascade of increasingly more accurate methods
starting with the steepest descents method and continuing
through conjugate gradients to the quasi-Newton-Raphson
method using the Broyten, Fletcher, Goldfarb, and Shanno
(BFGS) approach. At each step the maximum number of steps
used for each method was 250 to a final convergence of 0.001.
No cutoff was used for calculating nonbonded interactions. A
constant dielectric constant of 1.0 was used.
Molecular dynamics simulations were carried out at a
constant temperature and volume with no nonbonded cutoff
and a dielectric constant of 1.0. The system was heated in 100
steps from an initial temperature of 298 K to the target
temperature. Temperature was maintained by velocity scaling.
After heating, data were collected over 50 000 steps (50 ps) at
10 fs intervals. The system reached equilibrium very quickly,
as evidenced by the stability of temperature and potential
energy through the course of the simulation (data not shown).
Lattice parameters for 3 and 12 were obtained by least-
squares refinement of the angular settings from 25 reflections
lying in a 2θ range of 10-60°. Intensity data for 3 was collected
using θ-2θ scans in the range 4° < 2θ < 120°. For 12, intensity
data were collected using θ-2θ scans in the range 4° < 2θ <
115°. Lattice parameters for 6 were obtained by least-squares
refinement of the angular information from 201 reflections
measured at 0.3° increments of ω with three different 2θ and
φ values and a detector-to-crystal distance of 4.959 cm.
Intensity data for 6 were collected as 10 s data frames
measured at 0.3° intervals of ω covering nearly one hemisphere
of intensity data with a maximum 2θ value of 56.62°. The data
for all three compounds were corrected for decay, Lorentz, and
polarization effects. Data for 3 and 6 were also corrected for
absorption effects (3: based on measured ψ scans, correction:
min transmission 0.7645, max transmission 1.0000; 6: semiem-
pirical absorption correction using SADABS, correction: min
transmission 0.7663, max transmission 0.9280).35
Structures for 3, 6, and 12 were solved by a combination of
direct methods using SHELXTL v5.0336 and the difference
Fourier technique and were refined by full-matrix least-
squares on F2 (data for 6 were used out to 0.8 Å resolution).
Non-hydrogen atoms were refined with anisotropic displace-
ment parameters. Weights were assigned as w-1 ) σ2(Fo2) +
(aP)2 + bP where P ) 0.33333Fo2 + 0.66667Fc2 and a ) 0.0930,
b ) 23.9669 for 3, a ) 0.0310, b ) 1.1469 for 6, and a ) 0.1101,
b ) 1.2962 for 12. An extinction correction of the form
k[1+0.001(øFc2λ3)]/sin(2θ)]-1/4 was applied (3, k ) 0.28507 and
ø ) 0.00042(7); 12, k ) 0.4638 and ø ) 0.00042(11)). Aromatic
and methylene H atom positions for 6 were calculated on the
basis of a geometric criterion, while methyl H atom positions
for 3 and 6 were calculated after the location of one H atom
from the electron density map. All H atom positions for 3 and
6 were allowed to ride on their respective atoms. All methyl
groups in 6 had disordered hydrogen positions, and occupan-
cies were fixed at 0.5. In 12, the H atom positions were
calculated on the basis of a geometric criterion and allowed to
ride on their respective atoms with the exception of H30. H30
was located directly from the difference electron density map,
and its position was allowed to refine. H atom isotropic
temperature factors for all three structures were defined as
U(C)*a ) U(H) where a ) 1.5 for methyl hydrogens and a )
1.2 for the remaining hydrogens. Compound 12 packs in the
crystalline lattice with a badly disordered CH2Cl2 (occupancy
Ack n ow led gm en t. Mr. J im Carlson obtained the
mass spectra. We thank E.M. Science for a gift of silica
gel. T.M.B. is grateful for financial support in the form
of a Quantum Chemical (now Millennium Petrochemi-
cals) fellowship and from a scholarship from the Re-
search Scholars program in the Department of Chem-
istry at the University of Cincinnati. J .A.K.B. sincerely
thanks Dr. Charles Campana (Bruker AXS Inc.) and Dr.
Ewa Skrzypczak-J ankun (University of Toledo) for
assistance in using the CCD diffractometer system for
6. Data were collected through the Ohio Crystal-
lographic Consortium, funded by the Ohio Board of
Regents 1995 Investment Fund (CAP-075), located at
the University of Toledo, Instrumentation Center in
A&S, Toledo, OH 43606. The authors thank Prof. J ohn
S. Ricci and Mr. J effrey B. Fortin of the University of
Southern Maine, Portland, ME, for help in collecting
crystallographic data.
Su p p or tin g In for m a tion Ava ila ble: Lists of H atom
coordinates and isotropic thermal parameters, anisotropic
thermal parameters for other atoms, bond distances, bond
angles, and torsion angles are available for 3, 6, and 12 as
well as molecular modeling graphs depicting torsion angles
versus time for all three xylyl isocyanides in 3.
(35) Data for 3 were collected on a Rigaku AFC5R rotating anode
diffractometer. The TEXSAN suite of programs were used for data
collection and data processing, Molecular Structure Corp.: Woodlands,
TX. Data for
6 were collected on a Siemens SMART 1K CCD
diffractometer. The SMART v4.05 and SAINT v4.05 programs were
used for data collection and data processing, Siemens Analytical X-ray
Instruments, Inc.: Madison, WI. SADABS was used for the application
of a semiempirical absorption correction, G. M. Sheldrick, University
of Goettingen, Germany. Data for 12 were collected on a Siemens P4
diffractometer. The XSCANS suite of programs were used for data
collection and data processing, Siemens Analytical X-ray Instruments,
Inc.: Madison, WI.
(36) SHEXLXTL v5.03 was used for the structure solution and
generation of figures and tables, Siemens Analytical X-ray Instru-
ments, Inc.: Madison, WI. Neutral-atom scattering factors were used
as stored in SHELXTL v5.03.
OM9903644
(37) SQUEEZE is a routine implemented in PLATON-92, which
allows solvent contributions to be eliminated from the reflection data.
A. L. Spek, University of Utrecht, Utrecht, The Netherlands.
(38) Forcefield-Based Simulations; Molecular Simulations Inc.: San
Diego, April 1997.
(39) Shi, S.; Yan, L.; Shaulski, J .; Thacher, T. Manuscript in
preparation.