Reaction of [fac-(CO)3(dppe)Mn(OH2)]BF4 with 13CO and conversion of the resulting tetracarbonyl to labeled isocyanato complexes

Article abstract of DOI:10.1016/S0022-328X(00)00503-9

When ¹³CO is bubbled into a CH₂Cl₂ solution of [fac-(CO)₃(dppe)Mn(OH₂)]BF₄ (dppe = Ph₂PCH₂CH₂PPh₂), the aqua ligand is replaced by the CO to give the tetracarbonyl, [(CO)₃(¹³CO)(dppe)Mn]BF₄ (1) in which the labeled CO appears exclusively at the site originally occupied by the aqua ligand. Treatment of this aqua complex with NaN₃ leads to an approximate 50:50 mixture of isocyanates corresponding to attack, with equal probability, at either the labeled CO or its axial partner. Neither of the two CO groups trans to the P atoms in the tetracarbonyl are affected.

Full text of DOI:10.1016/S0022-328X(00)00503-9

www.elsevier.nl/locate/jorganchem
Journal of Organometallic Chemistry 610 (2000) 118–121
Note
1
3
Reaction of [ fac-(CO) (dppe)Mn(OH )]BF with CO and
3
2
4
conversion of the resulting tetracarbonyl to labeled isocyanato
complexes
Thomas M. Becker, Koka Jayasimhulu, Milton Orchin *
Department of Chemistry, Uni6ersity of Cincinnati, Cincinnati, OH 45221-0172, USA
Received 17 May 2000; received in revised form 2 August 2000
Abstract
1
3
When CO is bubbled into a CH Cl solution of [ fac-(CO) (dppe)Mn(OH )]BF (dppe=Ph PCH CH PPh ), the aqua ligand
2
2
3
2
4
2
2
2
2
13
is replaced by the CO to give the tetracarbonyl, [(CO) ( CO)(dppe)Mn]BF (1) in which the labeled CO appears exclusively at
3
4
the site originally occupied by the aqua ligand. Treatment of this aqua complex with NaN leads to an approximate 50:50 mixture
3
of isocyanates corresponding to attack, with equal probability, at either the labeled CO or its axial partner. Neither of the two
CO groups trans to the P atoms in the tetracarbonyl are affected. © 2000 Published by Elsevier Science S.A. All rights reserved.
Keywords: Tetracarbonyl complexes; Labeled isocyanato complexes; 13CO
1
. Introduction
plexes with functional groups s-bonded to the Mn [2].
The chemistry of some such functional groups has
proven to be substantially different from that of the
corresponding functional groups in traditional organic
chemistry, obviously reflecting the influence of the tran-
sition metal [3,4]. The aqua ligand in [ fac-
The octahedral Mn(I) complex [ fac-(CO) (dppe)-
3
Mn(OH )]BF is readily prepared by a simple, high-
yield two step procedure [1].
2
4
Mn (CO) +dppe+1-pentanol
2
10
(
CO) (dppe)Mn(OH )]BF is readily exchanged with
3 2 4
“
fac-(CO) (dppe)MnH
(1)
(2)
D O with the retention of facial geometry so that even
3
2
were the reaction to proceed by dissociation of the
ligand, the new ligand occupies the position of the
departing ligand. However, when the aqua ligand is
replaced by CO, the resulting tetracarbonyl complex,
fac-(CO) (dppe)MnH+HBF4
3
“
[ fac-(CO) (dppe)Mn(OH )]BF
3 2 4
Owing to the facial geometry and the presence of the
chelating phosphorus atoms, the aqua ligand must oc-
cupy an axial position trans to CO. We have shown
that the aqua ligand in this complex is readily replaced
at room temperature by a wide variety of neutral as
well as anionic ligands to give the corresponding com-
[
(CO) (dppe)Mn]BF , contains two different geometric
4 4
sets of COs, one set of two, each trans to a P atom, and
the other set of two axial COs trans to each other. It
might be assumed that when an azido group attacks a
carbonyl of the tetracarbonyl complex, either of the
two now essentially symmetry equivalent axial COs
would be attacked to give the identical isocyanato
complex provided scrambling does not occur. In order
*
Corresponding author. Tel.: +1-513-556-9200; fax: +1-513-556-
239.
E-mail address: orchinm@ucbeh.san.uc.edu (M. Orchin).
1
3
to test this assumption, several experiments with CO
were conducted and are reported herein.
9
0
022-328X/00/$ - see front matter © 2000 Published by Elsevier Science S.A. All rights reserved.
PII: S0022-328X(00)00503-9

T.M. Becker et al. / Journal of Organometallic Chemistry 610 (2000) 118–121
119
13
2. Experimental
2.2. Preparation of [(CO) ( CO)(dppe)Mn]BF (1)
3
4
[
fac-(CO) (dppe)Mn(OH )]BF was prepared from
Compound
1
was
prepared
from
[ fac-
3
2
4
fac-(CO) (dppe)MnH as described in the literature [1].
(CO) (dppe)Mn(OH )]BF according to the directions
3
3 2 4
fac-(CO) (dppe)Mn(NCO) was also prepared in accor-
dance with the literature [5]. CO (99%) was purchased
from Aldrich. Solvents were used as received.
given for [(CO) (dppe)Mn]BF (see Section 2.1) with
4 4
3
13
13
the following exceptions: CO was bubbled through a
solution of the aqua complex (0.152 g, 0.237 mmol) for
ca. 1 min and the solution was stirred for ca. 24 h.
13
2
.1. Preparation of [(CO) (dppe)Mn]BF4
Crystalline [(CO) ( CO)(dppe)Mn]BF (1) (0.123 g,
4
3
4
0
.189 mmol), was obtained from a CH Cl –hexane
2 2
This tetracarbonyl was previously prepared from fac-
mixture.
(
CO) (dppe)MnH without, however, isolating the aqua
3
intermediate [6]. In this procedure, the pure aqua com-
13
2
.3. Stirring [(CO) (dppe)Mn]BF with CO
4 4
plex, [ fac-(CO) (dppe)Mn(OH )]BF (0.138 g, 0.215
3
2
4
mmol), was placed in a round bottomed flask and 20 ml
of CH Cl was added and the solution stirred. A very
quick evacuation was performed and a needle was then
inserted and CO was bubbled through the solution for
The tetracarbonyl, [(CO) (dppe)Mn]BF4 (0.107 g),
4
2
2
was placed in a round bottomed flask and approxi-
mately 20 ml of CH Cl was added and the solution
2
2
was stirred. A very quick evacuation was performed
approximately
3 min. The solution was stirred
1
3
and a needle was then inserted and CO was bubbled
through the solution for ca. 1 min. The solution was
overnight (ca. 18 h) to give [(CO) (dppe)Mn]BF . Isola-
4
4
tion techniques were the same as those previously re-
ported and the spectral data match those previously
reported [5,6].
13
stirred under the CO and monitored by IR for a total
of 4 days. There was no evidence of a reaction.
1
3
2
.4. Reaction of [(CO) ( CO)(dppe)Mn]BF with NaN
3 4 3
to gi6e labeled isocyanates
The reaction conditions were similar to the literature
preparation for unlabeled fac-(CO) (dppe)Mn(NCO)
3
1
3
[
5]. [(CO) ( CO)(dppe)Mn]BF (0.103 g, 0.158 mmol)
3 4
was dissolved in 20 ml of methanol and NaN was
3
Table 1
added (0. 11 g, 0. 17 mmol). The mixture was stirred
for 1 day. The solvent was removed by rotary evapora-
tion and the resulting solids were extracted with ben-
zene. The benzene solution was concentrated and
hexane was added. On cooling yellow crystals (0.058 g)
resulted.
a
Mass spectroscopy results (m/z)
13C
Parent compound
Fragment
incorporation
+
C29, [(CO) (dppe)Mn]
,
Ratio 538:537
3
537
[
[
(CO) (dppe)Mn]BF4
Theory
Observed
Observed
0.32
1.2
1.5
4
2.5. Mass spectra
(CO) (13CO)(dppe)-
3
Mn]BF4 (1)
Mass spectra were recorded on a Kratos MS-25
high-performance mass spectrometer. Spectra of the
+
C27, [(dppe)Mn(NCO)]
,
Ratio 496:495
0.30
495
unlabeled complexes, [(CO) (dppe)Mn]BF4 and fac-
4
fac-(CO) (dppe)-
Theory
3
(
CO) (dppe)Mn(NCO), were used for comparison with
Mn(NCO)
3
Observed
0.33
the mass spectra of the products isolated in Sections 2.2
and 2.4. The relevant data are found in Table 1.
Mixture of
fac-(CO) (dppe)-
Theory:
ꢀ (no 495)
1.30
3
13
100% (N13CO)
Mn(N CO) (2),
13
13
(
CO)2( CO)(dppe)-
50% (N CO),
2.6. IR spectra
Mn(NCO) (3)
50% (NCO)
1
3
2
7
5% (N CO),
5% (NCO)
0.63
IR spectra were recorded on a Perkin–Elmer Spec-
trum One FTIR instrument employing 1.0 mm NaCl
liquid cells. IR data for all reactions are found in Table
2. The Perkin–Elmer instrument was utilized to deter-
mined peak areas after a deconvolution program was
employed to separate overlapping bands.
Observed
1.28
a
All ratios include appropriate contributions from the natural
abundance of 13C in the fragments. Theoretical ratios were calculated
based on Table 2.2 of Ref. [7].

1
20
T.M. Becker et al. / Journal of Organometallic Chemistry 610 (2000) 118–121
Table 2
Infrared spectra, w (cm ) in CH Cl
−
1
2
2
[
(CO) (dppe)Mn]BF4 a
4
Carbonyl groups
2093 vs 2024 sh
2081 vs 2023 sh
2014 vs 1998 vs
1996 vs 1985 vs
[
(CO) (13CO)(dppe)Mn]BF4 (1)
3
Carbonyl groups
b
fac-(CO) (dppe)Mn(NCO)
3
NCO
2240s
Carbonyl groups
2028 vs 1957 vs
1919 vs
1
3
13
Mixture of fac-(CO) (dppe)Mn(N CO) (2) and (CO)2( CO)(dppe)Mn(NCO) (3)
3
Area (%T×cm−1)
−2156
−2165
1881 s
N13CO
NCO
Carbonyl groups
2182 s
2240 s
2026 vs 1956 vs
1919 s
a
The IR data for [(CO) (dppe)Mn]BF have been reported previously [5].
4
4
b
Complete IR data in C H for fac-(CO) (dppe)Mn(NCO) [5] and in CH Cl for the carbonyl groups have been reported previously [9].
6
6
3
2
2
3
. Results and discussion
an isocyanato group in the same geometric position. If
the attack were to occur exclusively on the labeled CO,
at the position originally occupied by the aqua ligand,
then 100% of the isocyanato group should be the
labeled isocyanate 2 (Scheme 1), and no labeled COs
should remain. However, if the attack of the azide were
to occur equally on either of the axial COs then half of
the isocyanato groups would contain the label and the
other half of the label would still be on the axial CO;
When unlabelled tetracarbonyl is prepared by our
13
standard procedure and then treated with CO, no
incorporation of the label is observed indicating that
once formed the monolabeled tetracarbonyl will not
undergo further exchange with ligated COs, i.e. no
complex with multi-labeled COs need be considered.
When the labeled tetracarbonyl 1 (Scheme 1) is pre-
−
pared and treated with NaN to produce the corre-
50% of 2 and 50% of 3 (Scheme 1). If attack of N3
3
sponding isocyanato complex, the question arises as to
which of the two sets of coordinated COs is attacked.
The reaction of the azide with the CO proceeds via a
pathway analogous to the Curtius rearrangement,
where the CO originally attacked by the azide results in
were on the equatorial COs of 1 (Scheme 1), the
resulting isocyanato complex would have a mer-
configuration, but this possibility need not be consid-
ered because only the fac-isomer is obtained. Another
possibility should be mentioned. If during a dissociative
Scheme 1.

T.M. Becker et al. / Journal of Organometallic Chemistry 610 (2000) 118–121
121
displacement of the aqua ligand, the three remaining
COs somehow became equivalent before the fourth and
labeled CO became ligated, then the isocyanato group
formed subsequently would contain only one quarter of
the label; 25% of 2. Although the chelating phosphine
would make such a pentacoordinate intermediate un-
likely, it is a conceivable intermediate. The question
two axial COs in the resulting tetracarbonyl are indis-
tinguishable in their reaction with the azide.
3.2. IR spectra
Comparison of the IR data (Table 2) of the unla-
belled tetracarbonyl with the monolabeled tetracar-
bonyl 1 show that as expected, all of the CO stretching
frequencies in 1 are shifted to lower frequencies by
isotopic labeling.
13
then is what percentage of the NCO group is N CO?
00% if only the labeled CO is attacked; 50% if the
1
labeled CO and its axial partner CO are equally at-
tacked; or 25% if all four COs become equivalent
before attack. In order to answer this question, infrared
and mass spectra of the reaction products were ob-
tained and the results are discussed below.
The IR of the products from the reaction of the
labeled tetracarbonyl 1 with NaN3 clearly show
−1
stretches for both the NCO in 3 (2240 cm ) and
13
−1
N CO in 2 (2182 cm ). If one assumes as a first-order
approximation that the isotope effect is associated ex-
1
3
13
3
.1. Mass spectra
clusively with CꢀO in Nꢀ CꢀO a shift to lower
−1
frequency of approximately 51 cm is predicted. Simi-
larly, if one assumes that the isotope effect is associated
The relevant mass spectra data are summarized in
13
13
Table 1. None of the complexes exhibit a molecular ion
exclusively with Nꢀ C in Nꢀ CꢀO a shift to lower
−1
+
’
(
M
) peak. All peak numbers are m/z values.
The spectrum of the non-labeled tetracarbonyl con-
tains the peak 537, corresponding to the fragment
frequency of approximately 48 cm
is predicted [8].
is, therefore, not unrea-
−
1
The observed shift of 58 cm
sonable. In order to quantify the relative amounts of 2
and 3, a deconvolution was performed to eliminate the
slight overlap of the peaks. The areas of the isocyanato
bands were calculated and found to be very similar
indicating a mixture of approximately 50% 2 and 50%
3, a conclusion consistent with the mass spectral analy-
sis.
+
+
’
(CO) (dppe)Mn] =(M −CO). The 538:537 peak
3
[
intensity ratio is 1.2 while the expected value for a C29
fragment is 0.32 [7]. This higher than expected ratio
observed is due to contributions from the incorporation
of hydrogen, i.e. {[(CO) (dppe)Mn] +H}. The labeled
tetracarbonyl 1 has a 538:537 peak intensity ratio of
+
3
1
.5. This increase in the ratio from 1.2 to 1.5 indicates
1
3
incorporation of CO but a quantitative comparison is
not possible owing to the {[(CO) (dppe)Mn] +H]}
+
3
Acknowledgements
interference and the fact that we cannot determine
which of the four COs is lost. Hence, attention was
directed to the isocyanato spectra.
We thank Professor Bruce S. Ault for valuable dis-
cussions on the IR data. A paper in preparation is
devoted to a detailed interpretation of the IR data of
these and other similar complexes.
The
pure,
unlabelled
isocyanate,
fac-
(
[
CO) (dppe)Mn(NCO), has the fragment peak 537,
3
+
+
’
(CO) (dppe)Mn] =(M −NCO), and again its
3
5
{
38:537 peak intensity ratio shows interference from
+
[(CO) (dppe)Mn] +H]} and accordingly this ratio is
3
References
not useful. However, the fragment peak 495,
+
+
’
[
(NCO)(dppe)Mn] =(M −3CO), is also observed
[
1] T.M. Becker, J.A. Krause Bauer, C.L. Homrighausen, M. Orchin,
Polyhedron 18 (1999) 2563.
for the unlabeled isocyanato complex and the 496:495
peak intensity ratio is consistent with the 496:495 peak
intensity ratio for a C27 fragment. This ratio was used
to obtain the quantitative information required to an-
swer the question of what percentage of the NCO
group is N CO in the mixture of isocyanato com-
plexes. The observed and theoretical ratios are shown
in Table 1 and clearly indicate that an approximate
[2] T.M. Becker, J.A. Krause Bauer, C.L. Homrighausen, M. Orchin,
J. Organomet. Chem. 602 (2000) 97.
[
3] G.Q. Li, R.M. Bums, S.K. Mandal, J.A. Krause Bauer, M.
Orchin, J. Organomet. Chem. 549 (1997) 89.
[
[
4] J. Freudenberg, M. Orchin, Organometallics 1 (1982) 1408.
5] S.K. Mandal, D.M. Ho, M. Orchin, Polyhedron 11 (1992) 2055.
1
3
[6] M. Orchin, S.K. Mandal, J. Feldman, Inorg. Synth. 32 (1998)
98.
2
[
7] F.W. McLafferty, F. Turecek, Interpretation of Mass Spectra,
fourth ed., University Science Books, Sausalito, CA, 1993.
8] M. Orchin, H.H. Jaffe, Symmetry, Orbitals, and Spectra, Wiley-
Interscience, New York, 1971, p. 233.
5
0:50 mixture of 2 and 3 are found in the reaction
mixture. Therefore, we conclude that no ligand rear-
rangement occurs in the substitution of the aqua ligand
in [ fac-(CO) (dppe)Mn(OH )]BF by CO and that the
[
[9] Q.G. Li, M. Orchin, J. Organomet. Chem. 535 (1997) 43.
3
2
4
.