Reactions of "in-Situ" generated Cr(CO)5CN-C(Cl)- PR3 (R = Ph, NMe2) with Potential Dipolarophiles. New Examples of [3+2] Cycloadditions with Concomitant Interligand CC Bond Formation [1]

Article abstract of DOI:10.1002/zaac.200900310

The highly functional chromio nitrile ylides Cr(CO)₅C=N-C(Cl)- PR₃ (R = Ph, NMe₂) (2a, b) form readily in mixtures of Cr(CO)₅CNCCl₃ (1) with the respective phosphanes. Their reactions without isolation with the acylating agent Cr(CO)₅CN(C=O)Ar (Ar = C₆H₄-P-NO₂), MeO₂CC≡ CCO₂Me, ketones (R₂C=O, R = Me, Et; 2R = 9-fluorenediyl) and phenyl isocyanate to give the Cl/COAr-substitution- (6a), alkyne insertion- (10) and [3+2] cycloaddition products (13, 14) are described. Formation of the latter is accompanied by an interligand CC coupling with loss of P(NMe ₂)₃ and OP(NMe₂)₃ (after hydrolysis). Their reactions are compared with, those of the parent compounds L_nMCN-C(H)-PPh₃.

Full text of DOI:10.1002/zaac.200900310

ARTICLE
DOI: 10.1002/zaac.200900310
Reactions of “in-Situ” Generated Cr(CO)5CN–C(Cl)–PR3 (R = Ph, NMe2)
with Potential Dipolarophiles. New Examples of [3+2] Cycloadditions with
Concomitant Interligand CC Bond Formation [1]
Samyoung Ahn,[a] Volker Langenhahn,[a] Wilfried Sperber,[a] Wolfgang Beck,[b] and
Wolf Peter Fehlhammer*[a,b]
Keywords: Cycloaddition; Heterocycles; Carbenes; Bridging ligands; Isocyanide ligands;
Chromium and dichromium complexes
Abstract. The highly functional chromio nitrile ylides Cr(CO)5C≡N– tion- (6a), alkyne insertion- (10) and [3+2] cycloaddition products (13,
C(Cl)–PR3 (R = Ph, NMe2) (2a, b) form readily in mixtures of 14) are described. Formation of the latter is accompanied by an interli-
Cr(CO)5CNCCl3 (1) with the respective phosphanes. Their reactions gand CC coupling with loss of P(NMe2)3 and OP(NMe2)3 (after hy-
without isolation with the acylating agent Cr(CO)5CN(C=O)Ar (Ar = drolysis). Their reactions are compared with those of the parent com-
C6H4-p-NO2), MeO2CC≡CCO2Me, ketones (R2C=O, R = Me, Et; 2R = pounds LnMCN–C(H)–PPh3.
9-fluorenediyl) and phenyl isocyanate to give the Cl/COAr-substitu-
[9]. This as well as the nature of the species that subsequently
Introduction
form can easily be deduced from what is known to occur in
Trichloromethyl isocyanide is one of the exciting small mol-
Appel’s two-component-system CCl4/PR3 [10]. A comparison
ecules with an inexhaustible reactivity potential on top. To date
of the organoelement with the organometallic series is shown
it only exists as a ligand in Cr(CO)5CNCCl3 (1) whose synthe-
in Scheme 1.
sis from [Cr(CN)(CO)5]– and aryldiazonium ions in CHCl3 by
what we call “radical alkylation of cyanido complexes” obvi-
ously makes special demands on the redox properties not met
by other organometallics [2, 3]. Reactions of 1 with nucleo-
philes (amines, thiols) generally set in at the isocyanide carbon
atom (C1) followed by successive Cl-substitution at C3 [4, 5].
Occasionally only the C1 atom remains of the original isocya-
nide forming new isocyanide or diaminocarbene complexes
with the amine reactant [6, 7]. Imidazole, in contrast, interacts
with C3 exclusively giving rise to pentacarbonyl[tris(imida-
zolyl)methyl isocyanide]chromium [8].
From reactions with phosphanes, correspondingly, P-func-
tionalized methylisocyanide complexes result, though the site
of primary attack is certainly not C3 but the chloro substituents
Scheme 1. Species generated in mixtures of CCl4/PR3 (left column) or
1/phosphanes, respectively (right column) (see text).
* Prof. Dr. W. P. Fehlhammer
E-Mail: wpfehlhammer@gmx.de
[a] Institut für Anorganische und Analytische Chemie
Freie Universität Berlin
Depending on the stoichiometry and the reaction condi-
tions – such as the presence of traces of water – compounds
A–D were identified, all of which have organometallic coun-
terparts 2–5. Only compound 2 has not been isolated as such,
but the trapping of 2a in Wittig-type reactions with carbonyl
Fabeckstr. 34–36
14195 Berlin, Germany
[b] Department Chemie und Biochemie
Universität München,
Butenandtstr. 5–13
81377 München, Germany
492
© 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Z. Anorg. Allg. Chem. 2010, 636, 492–498

Reactions of In-Situ Generated Cr(CO)5CN–C(Cl)–PR3 (R = Ph, NMe2)
compounds to give the corresponding α-chloroalkenyl isocya-
nide complexes proves its intermediary existence [1]. As met-
allo-nitrile ylides, capable of undergoing [3+2] cycloadditions,
compounds 2 and 5 (after deprotonation) were of particular
interest, the more as our synthetic access to 5 [and
Cr(CO)5CNCHPPh3] so far required several tricky steps
through free functional isocyanides lacking metal stabilization
[11, 12]. “In situ”-generated 2b has already been reported to
react with ketones in the presence of secondary amines afford-
ing that sort of [3+2] cycloaddition products (vide infra and
lit. [1]).
2. Reaction with MeO2CC≡CCO2Me
This paper presents further reactions of the chloro-substi-
tuted chromio-nitrile ylides 2, which were carried out in order
to confirm their intermediary existence, to investigate the role
of the chlorine atom and to demonstrate the dependence of the
reaction course on the nature of the phosphane.
Again we compare the reaction with acetylene dicarboni-
cacid diester of the “in situ” generated chloro derivative 2a
with that of the parent compound M(CO)5CN–CH–PPh3
(where M = W): whereas the latter gave rise to the betain-like
4-phosphoniopyrrol-2-ato complex species 9 [13], in the case
of 2a all evidence points to the fact that 1,3-migration of the
PPh3 group or, respectively, insertion of the acetylene into the
P–C bond had occurred. The main driving force for the cycli-
zation to 9 probably is the ring aromatization, which requires
the C3–H bond to be cleaved, a process not feasible for 2a.
Here, the primary addition product with MeO2C–C≡C–CO2Me
stabilizes by rearrangement into the well-known α-chloroal-
kenyl isocyanide structural unit [1] (cf. Section 3) and a new
phosphorus ylide, which is much less reactive because of ex-
tensive conjugation. This also explains why the final product
10 does not show tendency towards ring closure at all.
Results and Discussion
1. Reaction with Cr(CO)5CNC(=O)C6H4-p-NO2
Acylisocyanide complexes such as Cr(CO)5CNC(=O)C6H4-
p-NO2 are known to act as mild acylating agents towards vari-
ous kinds of nucleophiles. Its reaction with 2a, prepared “in
situ” from 1 + 2 PPh3, proceeds with elimination of the chlo-
rine atom (formally as Cl+) at C3 to give complex 6a. Obvi-
ously, this reduction (dechlorination) is the preferred way of
stabilization for the presumed primary addition product. A
third equivalent of PPh3 is needed, which eventually reappears
in the salt-like (Ph3PCl+) yet partially hydrolyzed precipitate
together with the leaving group [Cr(CN)(CO)5]–.
An unequivocal proof for both structural units mainly arises
from the characteristic 13C NMR spectroscopic data of the iso-
cyano, the chlorine carrying and the ylidic carbon atoms; due
to its bonding to phosphorus the latter appears as a doublet
1
with the expected order of magnitude for J(P,C) (Experimen-
tal Section).
The nature of 6a is unequivocally settled by its spectroscopic
data (Experimental Section), which perfectly agree with those
of 6b, the product of the reaction of Cr(CO)5CN–CH–PPh3
with trifluoroacetic acid anhydride [13]. Note that the forma-
tion of 6b required the abstraction of H+ by a base, which
usually was the ylide itself (“Umylidierung”). The CN stretch-
ing vibrations [6a/b: 2129 m/2127 m cm–1] as well as the 13C
NMR spectroscopic shifts of the isocyano (6a/b: δ = 166.2/
168.7), the ylidic (6a/b: δ = 77.2/70.8) and the acyl carbon
atoms (6a/b: δ = 183.2/172.1) are almost identical with the
latter two existing as doublets (or even as pseudo-dd’s) be-
cause of coupling with 31P (and 19F). A surprising difference,
however, turns up in the ν(C=O)acyl values [6a/b: 1695 m/
Very indicative of the P-ylidic nature of 10 also is a δ(31P)
1596 s, 1584 s, 1571 m (?) cm–1]; still there is no doubt that of 20.3, particularly in comparison with a shift of δ = 9.5 for
the strong absorption at 1526 cm–1 must be assigned to the the phosphorus resonance in 9. The rather low IR spectro-
antisymmetric stretching of the NO2 substituent [14].
scopic frequency of the CN stretching vibration is typical for
In neither case the conceivable cyclization to a 4-phospho- isocyano groups linked to a sp2 carbon atom. The assignment
nio-oxazolatochromium species (7) took place, which is cer- of ν(C=C), on the other hand, is questionable; however, there
tainly a consequence of the interplay of the stabilization effect is indirect evidence in the form of a doubling of the 13C signals
of chromium(0) on the isocyano group and the highly electron- for (CO)trans and (CO)cis which results from the presence of
deficient substitution on C3 and Cacyl, which reduces the nu- E and Z isomers. Similar observations including doubled CN
cleophilicity of the acyl oxygen atom. That acylium ions are resonances have been made on α-chloroalkenyl isocyanide
able to act as dipolarophiles towards metallo-nitrile ylides was complexes Cr(CO)5CNC(Cl)=CRR' [1].
shown
earlier
by
the
synthesis
of
8
from
To the best of our knowledge, this is the first example of an
alkyne insertion into a C–Pylide bond.
[Pt(Cl)(CNCH2CO2Et)(PPh3)2]BF4, NEt3 and PhCOCl [15].
Z. Anorg. Allg. Chem. 2010, 492–498
© 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.zaac.wiley-vch.de
493

S. Ahn, V. Langenhahn, W. Sperber, W. Beck, W. P. Fehlhammer
ARTICLE
3. [3+2] Cycloaddition and C3=C3 Bond Formation in the
Three-Component System [1/P(NMe2)3/Dipolarophile]: Di-
chromium Complexes with Carbene-Isocyanide Bridging Li-
gands
(In parentheses it should perhaps be pointed out that in prin-
ciple there is a second factor strongly determining the reaction
course i or ii, viz. the electrophilicity of the isocyano group,
which can be adjusted by proper choice of the metal; in the
platinum(II) complex [Pt(Cl)(CN-CH-PPh3)(PPh3)2]+ the par-
ent 1,3-dipole is sufficiently activated to undergo [3+2] cy-
cloadditions with aldehydes in spite of the presence of PPh3
[1]).
It stands to reason that 11 and 12 are formed via the same
intermediate H. In the decisive step, the phosphonio substitu-
ent in the 4-position of the latter is replaced either by the sec-
ondary amine to give – after HCl-elimination – 11, or by a
second molecule of the isocyano(chloro) ylide to give – pre-
sumably via 12 – the “mixed” C=C-linked μ-oxazolin-2-ylid-
ene/α-chloroalkenyl isocyanide bis(pentacarbonylchromium)
complexes 13 [Equation (1)].
For both types of products striking parallels can be drawn to
earlier findings in our laboratories. Thus, a reaction of the par-
ent metallo-nitrile ylides M(CO)5CNCHPPh3 (M = Cr, W) with
triphenylketeneimine resulted in an exceptional lability to-
wards substitution of the 4-PPh3 group of the heterocyclic pri-
mary product I, which led to the C–C coupled dinuclear inter-
mediate J and further to K [Equation (2)] [16]. As for 11, a
host of such 4-amino-Δ3-oxazolin-2-ylidene complexes that
exhibit an extremely high tendency of formation had already
been synthesized by very versatile and efficient three-compo-
3.1 Ketones as Dipolarophiles
The mentioned α-chloroalkenyl isocyanide complexes were
synthesized from 2a and ketones [1]. Modification of the phos-
phane component led to an extension of this chemistry with
widely differing results, which are summarized below.
In the first step, the “in situ” generated compound
Cr(CO)5CN–CCl–PR3 (R = Ph, NMe2) attacks the carbonyl
carbon atom. Which of the two possible consecutive processes
then occurs – i to give α-chloroalkenyl isocyanide (and phos-
phane oxide) or ii leading to oxazolin-2-ylidene(ato) com-
plexes – depends on the relative oxygenophilicity of the phos-
phorus atom. As this is less pronounced in the R = NMe2 case,
the [3+2] cycloaddition is preferred, ending up with the hetero-
cyclic products 11 (with a secondary amine as fourth compo-
nent) [1] and 12 or 13, respectively (in the absence of amine)
[Equation (1)].
(1)
(2)
494
www.zaac.wiley-vch.de
© 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Z. Anorg. Allg. Chem. 2010, 492–498

Reactions of In-Situ Generated Cr(CO)5CN–C(Cl)–PR3 (R = Ph, NMe2)
nent additions (3CC) between cyanido complexes, isocyanides peated chromatography and fractional recrystallization. For this
and carbonyl compounds [17].
reason we refrained from a separation of the isomers of 13c.
The IR spectra of the E and Z isomers of 13a and 13b exhibit
slightly different ν(NH), ν(CN) and ν(C=C) absorption bands;
it is not possible, however, to tell to which configuration the
individual bands belong. All ν(CN) bands are of low intensity
and fall in the narrow range of 2115–2125 cm–1, which is char-
acteristic for α-chloroalkenyl isocyanide complexes of Cr0 [1].
The poor solubility of the new compounds prevented us from
carrying out more extensive NMR spectroscopic measurements
(Experimental Section).
In order to avoid hydrolysis of the presumed intermediate
12, a work-up procedure without chromatography was chosen
for the reaction with fluorenone. As is distinct from the needle-
shaped crystals of complex 13a, we now obtained cubic ones
whose IR spectrum clearly lacked a ν(NH) band. Though the
compound was analyzed correctly as 12 (R2 = fluorenediyl),
the X-ray structure determination only showed complex 13a
(see below). Obviously slow hydrolysis also occurs in the solid
during crystal manipulation.
Complexes 13 form E and Z isomers, which exist in a ratio of
≈ 1:1 according to the IR spectra. Because of their similar Rf
values, their separation aroused severe problems requiring re-
As mentioned above, X-ray structure analysis of a supposed
intermediate 12 came up with the structure of 13a instead,
which coexists with the byproduct P(NMe2)3=O of its hydroly-
sis in the crystal. Figure 1 shows an ORTEP drawing of the
molecular structure of 13a with pertinent bonding parameters,
whereas Figure 2 depicts an asymmetric unit containing the
hydrogen-bonded pair of 13a + O=P(NMe2)3. The distance be-
tween O2 of the phosphane oxide and N1 of the heterocycle
amounts only 2.64(3) Å, this means we deal with a rather
strong NH···O bond according to the range of bond lengths
(2.50–2.7 Å) allocated to this category in the literature [18].
Relative to an average N,O-carbene complex the Cr1–C11 dis-
tance of 1.98(3) Å is unusually short as is the C11–O1 bond
length [1.30(3) Å], which is even shorter than the Ccarbene–N
bond length [C11–N1: 1.39(3) Å] [19]. An N1–C1 bond of
similar length, an exocyclic isolated CC double bond and two
single bonds (C1–C2, C2–O1) to and from the ketone carbon
atom (C2) complete the stereochemical picture, which can be
understood as a consequence of the presence of the electron-
withdrawing chloro(isocyano)vinylidene substituent in the 4-
position of the heterocycle though the poor quality of the struc-
ture prevents a more detailed interpretation. For comparison,
however, the formally carbanionic nature of the N-analogue
11 (and related 3CC products) and the extensive conjugative
interaction along the O–Ccarbene–N–C–N(exo-ammonium)
atomic sequence, which results in practically identical bond
lengths, should be noted [1, 17]. The isocyanide complex part
is inconspicuous with the exception of the C21–N2–C3 angle
of 160(3)° deviating considerably from linearity. Non-linear
[M]C≡N–C arrangements of terminally coordinated isocya-
nides were found in complexes both with particularly electron-
deficient ligands such as (OC)5Cr–CN–C(=NCF3)–NC–
Cr(CO)5 [C–N–C 167.8(5), 167.2(7)°] and with particularly
Figure 1. Molecular structure of compound 13a with the crystallo-
graphic numbering scheme (ORTEP plot). The thermal ellipsoids have
been drawn to include 50 % probability. Selected bond lengths /Å and
angles /°: Cr1–C11 1.98(3), Cr1–C14 1.78(3), (Cr1–Ccis av
C21 1.90(3), Cr2–C22 1.79(3), (Cr2–Ccis av
C11–N1 1.39(3), N1–C1 1.41(3), C1–C2 1.50(4), C2–O1 1.45(4),
C21–N2 1.16(4), N2–C3 1.39(3), C3–Cl 1.69(3), C3–C1 1.32(5),
N1···O2 2.64(3); Cr1–C11–O1 124(2), Cr1–C11–N1 130(2), N1–C11–
O1 107(2), C11–N1–C1 113(2), N1–C1–C2 103(2), C1–C2–O1
103(2), C2–O1–C11 114(2), Cr2–C21–N2 178(3), C21–N2–C3 160(3).
)
1.86, Cr2–
)
1.86, C11–O1 1.30(3),
electron-rich
metal
components,
e.g.,
trans-
[Mo(CNMe)2(dppe)2] [C–N–C 156(1)°] [20]. Interesting devi-
ations from the expected linearity of the structural element
[M]–C≡N–[M'] also occur in supercomplexes of the type
(NEt4)2[Si{N≡C–Cr(CO)5}6] [C–N–C 172.8(3), 168.5(3)°] or
(NEt4)3[Ni{N≡C–W(CO)5}5] [C–N–C 164(3), 156(3),
150(4)°] [21]. In this case the bending might be a consequence
of the steric repulsion between the fluorene group and cis-CO
ligands on Cr2. At the Cr1 site it is the largely staggered orien-
tation of the carbenoid heterocycle with respect to the cis-CO
ligands (interplanar angle: 39°), which ensures steric relief.
Figure 2. Asymmetric unit of 13a + O=P(NMe2)3 with N1–H···O2 hy-
drogen bridge.
Z. Anorg. Allg. Chem. 2010, 492–498
© 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.zaac.wiley-vch.de
495

S. Ahn, V. Langenhahn, W. Sperber, W. Beck, W. P. Fehlhammer
ARTICLE
(excitation energy 80 eV), (pos-FAB) Varian MAT CH 5 DF (neutral
xenon source at 3 keV). Microanalyses (C, H, N): Heraeus CHN-Rapid
elemental analyzer. Melting points (uncorrected): Gallenkamp MFB-
595 apparatus.
3.2 Reaction with PhN=C=O
The reaction of phenyl isocyanate with the “in situ”-gener-
ated chromium coordinated isocyano(chloro)methylenephos-
phorane (OC)5Cr–CN–CCl–P(NMe2)3 follows exactly the stoi-
chiometry and mechanism of the reaction with ketones (cf.
3.1), i.e., the final product must be formulated as 14. No at-
tempts have been made to isolate the dinuclear analogue of 12
prior to hydrolysis.
Pentacarbonyl[isocyano(p-nitrobenzoyl)methylenetriphenylphos-
phorane]chromium (6a): A solution of triphenylphosphane (524 mg,
2.0 mmol) in diethyl ether (10 mL) was added dropwise at room temp.
to a mixture of 1 (336 mg, 1.0 mmol) and pentacarbonyl(p-nitroben-
zoylisocyanide)chromium (368 mg, 1.0 mmol) in diethyl ether
(20 mL). The orange-red solution lightened, and
a precipitate
([PPh3Cl]Cl) formed. After stirring for another 4 h, the precipitate was
filtered off (G4 frit) and the solvent was removed in high vacuum. The
oily residue was dissolved in a small amount of diethyl ether and puri-
fied by column chromatography on neutral Al2O3 using diethyl ether/
petroleum ether (1:1) as eluent. The orange-red second fraction was
collected and the solvents were evaporated to dryness under high vac-
uum. Recrystallization of the orange-red powder from toluene at
–20 °C gave orange needles in 63 % yield. M. p. 117 °C.
C32H19CrN2O8P (642.33): calcd. C 59.83, H 2.95, N 4.36; found C
59.31, H 3.56, N 4.57. IR (KBr): ν = 3064 m [ν(CH)aryl], 2129 m
˜
[ν(CN)], 2059 s, 1927 vs, br. [ν(CO)], 1695 m [ν(CO)acyl], 1526 s
[νas(NO2)], 1346 s [νs(NO2)] cm–1. H NMR (CDCl3, 250 MHz): δ =
1
As in the case of 10, the 13C NMR spectroscopic data were
extremely useful for the detection of special structural ele-
ments such as the α-chloroalkenyl isocyanide part with its
characteristic narrow δ ranges for the isocyano (180–187) and
C–Cl carbon atoms (113 3). Two sets of CO signals – the
one at higher field (δ = 215.1, 214.1) presumably arising from
the isocyanide-coordinating Cr(CO)5 group – in combination
with two ν(CO)-A1 IR features prove the dinuclear nature of
14, yet unlike in 10 there is no further doubling of lines, which
could be associated with cis/trans isomerism. As for the car-
bene complex part, neither the Ccarbene signal nor a ν(NH) band
has been found; the NH group might be involved in strong
hydrogen bonding as established in the crystal structure of 13a,
a possible counterpart could be the exocyclic amido oxygen
atom of a neighboring molecule.
8.26 (m, 2 H, C6H4), 7.99 (m, 2 H, C6H4), 7.65 (m, 15 H, Ph). 13C{1H}
NMR (CDCl3, 250 MHz): δ = 217.6 (COtrans), 214.7 (COcis), 183.2
2
(d, COacyl), J(P,C) = 18 Hz, 166.2 (CNR), 148.5–121.4 (Caryl), 77.2
1
(d, CNCP), J(P,C) = 135 Hz. 31P{1H} NMR (CDCl3, 400 MHz): δ =
20.6. MS (+FAB, Xe, DMSO, 3-nitrobenzylalcohol), m/z (int. %): 643
(13) [M]+, 559 (18) [M – 3CO]+, 530 (21) [M – 4CO]+, 502 (18) [M –
5CO]+.
Synthesis of 10:
A solution of triphenylphosphane (524 mg,
2.0 mmol) in diethyl ether (10 mL) was added dropwise at room temp.
to a mixture of 1 (336 mg, 1.0 mmol) and acetylenedimethylcarboxyl-
ate (124 mg, 1.0 mmol) in diethyl ether (20 mL). The yellow solution
turned dark red, and a precipitate ([PPh3Cl]Cl) formed immediately.
After stirring for another 4 h, the precipitate was filtered off (G4 frit)
and the solvent was removed in high vacuum. The dark red resinous
residue was dissolved in the least possible amount of diethyl ether and
purified by column chromatography on neutral Al2O3 using diethyl
ether/petroleum ether (1:2) as eluent. The orange second fraction was
collected and the solvents were evaporated to dryness under high vac-
uum resulting in an orange-yellow microcrystalline solid, which al-
ready was analytically pure. M.p. 113 °C, yield: 73 %.
C31H21ClCrNO9P (669.76): calcd. C 55.59, H 3.13, N 2.09; found C
In comparison, no C3C3 coupling through phosphane substi-
tution with a second molecule of the ylide occurs in [3+2]
cycloadditions between the parent metallo-nitrile ylids LnM-
CNCHPPh3 (MLn = Pt(Cl)(PPh3)2, Cr(CO)5, W(CO)5, BPh3)
and isocyanates; as in the case of LnM-CNCHR (R = Tos,
CO2Et) the primary product gains stability by a formal C4 to
N3 -shift of the proton affording aromatic imidazol-5-olat-2-
ylidene species [11, 22–25]. Even with these apparently more
simple reagents, however, one has to grapple with diverse
complications such as the formation of site-isomeric oxazole
derivatives or 1:2 (ylide/isocyanate) addition products [22, 26].
56.21, H 3.66, N 2.18. IR (KBr): ν = 3059 w [ν(CH)aryl], 2950 m
˜
[ν(CH)alkyl], 2119 m [ν(CN)], 2041 s, 1998 s, 1944 vs, br. [ν(CO)],
1719 s [ν(CO2Me)], 1626 m [ν(C=C)] cm–1. 1H NMR (CDCl3,
250 MHz): δ = 7.72 (m, 15 H, Ph), 3.60 (s, 3 H, Me), 3.45 (s, 3 H,
Me). 13C{1H} NMR (CDCl3, 250 MHz): δ = 215.7, 215.3 (COtrans),
213.5, 213.2 (COcis), 186.9 (=CCO2Me), 180.6 (CrCN), 168.1 (d,
PCCO2Me), 2J(P,C) = 12 Hz, 133.7–127.3 (Caryl), 113.1 (C=CCl),
52.2, 50.1 (OCH3), 46.1 (d, PCCO2Me), 1J(P,C) = 128 Hz. 31P{1H}
NMR (CDCl3, 400 MHz): δ = 20.3. MS (+FAB, Xe, DMSO, 3-nitrob-
enzylalcohol), m/z (int. %): 670 (3) [M]+, 586 (3) [M – 3CO]+, 558
(7) [M – 4CO]+, 530 (41) [M – 5CO]+, 451 (37) [M – 5CO – Cr]+.
Experimental Section
All manipulations were performed under pure argon using Schlenk-
tubes and vacuum techniques. The solvents were dried and distilled General Procedure for the Complexes 13a-c: A stoichiometric
prior to use. Complex 1 was prepared according to a published proce- amount of the respective ketone was added to a solution of 1 (1.0 g,
dure [2]. Silica gel (100–200 μm) from ICN and Al2O3 90 active (neu- 2.98 mmol) in CH2Cl2 (40 mL) and the mixture was cooled to –60 °C.
tral, 0.063–0.200 mm) from Merck were used for column chromatog- To this suspension a solution of P(NMe2)3 (1.08 mL, 5.94 mmol) in
raphy. IR: Beckman IR 12 double beam infrared spectrometer and CH2Cl2 (10 mL) was added dropwise whereupon the mixture quickly
Perkin–Elmer 983 IR spectrometer. NMR: Bruker AM 250, JEOL FX turned dark violet. The cooling device was removed and stirring was
90 and JEOL Lambda 400 spectrometers. MS: (EI) Varian Mat 711 continued for 2 h at room temp. The solvent was removed in vacuo;
496
www.zaac.wiley-vch.de
© 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Z. Anorg. Allg. Chem. 2010, 492–498

Reactions of In-Situ Generated Cr(CO)5CN–C(Cl)–PR3 (R = Ph, NMe2)
the remaining brown oil was extracted with diethyl ether and subject (COtrans), 214.1 (COcis), 182.2 (CNR), 166.3 (>C=O), 132.5–125.8
to chromatography on silica gel (2 × 15 mL column). Elution with (Ph), 115.6 (CCl).
petroleum ether/diethyl ether (30:1) gave a first phase, which contained
X-ray Structure Determination of Complex 13a: In order to avoid
hydrolysis of complex 12, the brown oil obtained from 1, fluorenone
and P(NMe2)3, according to the general procedure for 13a–c, was di-
rectly recrystallized from dichloromethane layered with diethyl ether
at –25 °C, which means that chromatography was skipped. A piece
measuring 0.42 × 0.38 × 0.20 mm was cut out from a cube-shaped
crystal and mounted on fiberglass. All geometry and intensity data
were collected on a STOE four circle diffractometer at –50 °C in the
ω-2θ scan mode using Mo-Kα-radiation (λ = 0.71073 Å) and a graph-
ite-monochromator. Neither absorption nor extinction corrections were
carried out. The structure was solved by direct methods and developed
using alternative cycles of full-matrix least-squares refinement and dif-
ference-Fourier synthesis (programs SHELXS-86, SHELXL-93 [27,
28]). With the exception of chromium and chlorine, some of the light
atoms of the carbene-isocyanide bridging ligand (O1, N1, N2, C1, C2,
C3), P and O2 of the side product, and Cl8 and Cl9 of the built-in
CH2Cl2 solvent molecule, only an isotropic model was calculated. The
molecular plot was produced with the ORTEP program [29].
a small amount of Cr(CO)5CNCH2Cl. An orange fraction appeared
thereafter, which could not be identified. More polar eluents were em-
ployed to elute the main products 13a–c (13a, 13c: petroleum ether/
diethyl ether, 10:1–10:2; 13b: petroleum ether/diethyl ether, 20:1–
10:1) as mixtures of isomers. Thin layer chromatography showed the
presence of practically equal amounts of E and Z isomers in each case.
Separation of the isomers of 13a and 13b was finally successful
through (four times) repeated chromatography followed by fractional
recrystallization from CH2Cl2 at –25 °C. The separation of the isomers
of 13c was not attempted.
13a (mixture): Yellow crystals, m.p. 128 °C (dec.), yield 11 %.
C27H9ClCr2N2O11 (676.81): calcd. C 47.91, H 1.34, N 4.14; found C
48.02, H 1.58, N 4.44. IR (KBr) isomer I: ν = 3433 w [ν(NH)], 2114
˜
m [ν(CN)], 2067 m, 2033 s, 1980 s, 1932 vs, br. [ν(CO)], 1660 m
[ν(C=C)], 1452 s [ν(N–C–O)]; isomer II: ν = 3400 m [ν(NH)], 2113
˜
m [ν(CN)], 2069 m, 2028 s, 2008 w, 1931 vs, br., 1873 s [ν(CO)],
1
1673 m [ν(C=C)], 1452 s [ν(N–C–O)]. H NMR (CDCl3) (mixture):
δ = 9.2 (NH), 7.1–7.8 (Har). MS (mixture): m/z (%) = 675 (11) [M]+;
592 (2), 564 (14), 536 (11), 508 (18), 480 (3), 452 (3), 424 (13), 396
(22) [M – nCO]+ (n = 3–10).
C27H9ClCr2N2O11 (676.82) + O=P[N(CH3)2]3 (179.20) + CH2Cl2
(84.93), monoclinic, C2/c (Nr. 15), a = 39.854(18), b = 9.406(6), c =
23.858(12) Å, β = 108.02(4)°, V = 8507.48 Å3, Z = 8, ρcalcd.
=
1.469 Mg·m–3, temperature = 223 K, 2θ range for data collection =
4.0–37.0°, reflections collected = 3426, observed reflections (≥3σ|F|) =
1541, parameters (refined) = 294, R = 10.1 %, Rw = 11.5 %, weighting
scheme w = 1/σ(F2).
13b (isomer I): Yellow crystals, m.p. 159 °C (dec.), yield 22 %; (iso-
mer II): yellow crystals, m.p. 140 °C (dec.), yield 19 %.
C17H7ClCr2N2O11 (554.68): calcd. C 36.81, H 1.27, N 5.05; found (iso-
mer I) C 36.65, H 1.45, N 4.39; (isomer II) C 35.92, H 1.57, N 4.96.
IR (KBr) isomer I: ν = 3427 w [ν(NH)], 2110 m [ν(CN)], 2065 m,
˜
CCDC737354 contains supplementary crystallographic data for this pa-
per. These data can be obtained free of charge from The Cambridge
Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
1985 w, 1926 vs, br. [ν(CO)], 1670 w [ν(C=C)], 1426 s [ν(N–C–O)];
isomer II: ν = 3385 w [ν(NH)], 2113 m [ν(CN)], 2067 m, 2027 w,
˜
1
1936 vs, br. [ν(CO)], 1675 w [ν(C=C)], 1415 s [ν(N–C-O)]. H NMR
(CD2Cl2) (mixture): δ = 9.95 (NH), 1.75 (CH3). MS (isomer I): m/z
(%) = 554 (29) [M]+, 498 (2), 470 (4), 442 (18), 414 (100), 386 (38),
358 (5), 330 (5), 302 (20), 274 (20) [M – nCO]+ (n = 2–10); 165 (26) Acknowledgement
[CrL – COR2 + H]+; (isomer II) m/z (int.): 554 (20) [M]+; 470 (3),
Financial support for this work by the Fonds der Chemischen Industrie
442 (9), 414 (55), 386 (48), 358 (18), 330 (4), 302 (17), 274 (17) [M –
(FCI), the Deutsche Forschungsgemeinschaft (DFG), the Bundesminis-
terien für Bildung und Wissenschaft (BMBW) (Graduiertenkolleg
“Synthesis and Structure of Low Molecular Compounds”) and the
BAYER AG is gratefully acknowledged.
nCO]+ (n = 3–10); 165 (29) [CrL – COMe2 + H]+.
13c: Yellow crystals. IR (KBr) (mixture): ν = 3400 w, 3386 w [ν(NH)],
˜
2115 w [ν(CN)], 2065 w, 2033 s, 1961 s, 1921 vs, br. [ν(CO)], 1666
w [ν(C=C)], 1433 s [ν(N–C–O)] MS (mixture): m/z (%) = 582 (25)
[M]+, 470 (17), 442 (100), 414 (34), 386 (6), 330 (15), 302 (17) [M –
nCO]+ (n = 4–7, 9,10); 165 (21) [CrL – COEt2 + H]+.
Synthesis of 14: Compound 1 (336 mg 1.0 mmol) and PhNCO
(0.12 mL, 1.0 mmol) were dissolved in diethyl ether (20 mL). To this
References
[1] Reactions at the coordinated trichloromethyl isocyanide, Part
VIII. Part VII: V. Langenhahn, G. Beck, G. Zinner, D. Lentz, B.
Herrschaft, W. P. Fehlhammer, J. Organomet. Chem. 2007, 692,
2936.
solution,
a mixture of tris(dimethylamino)phosphane (0.37 mL,
2.0 mmol) in diethyl ether (5 mL) was added dropwise at room temp.
The yellow solution turned dark red, and an oily precipitate formed
mainly consisting of [PPh3Cl]Cl. After stirring for another 2 h, the
solution was siphoned off and the solvents were evaporated to dryness
under high vacuum. The red resinous residue was redissolved in a
[2] W. P. Fehlhammer, F. Degel, Angew. Chem. 1979, 91, 80: Angew.
Chem. Int. Ed. Engl. 1979, 18, 75.
[3] W. P. Fehlhammer, F. Degel, G. Beck, Chem. Ber. 1987, 120, 461.
small amount of ether and purified by column chromatography on neu- [4] G. Beck, W. P. Fehlhammer, Angew. Chem. 1988, 100, 1391;
Angew. Chem. Int. Ed. Engl. 1988, 27, 1344.
tral Al2O3 using diethyl ether/petroleum ether as eluent. The second
fraction contained the product and was collected. Removal of the sol-
vent left a light red powdery solid, which already was sufficiently pure.
Yield 61 %, m.p. 93 °C. C21H5ClCr2N3O11 (614.69): calcd. C 41.03,
H 0.81, N 6.83; found C 42.15, H 1.27, N 7.01. IR (KBr): ν = 2125
s [ν(CN)], 2060 s, 2043 s, 1936 vs, br. [ν(CO)], 1682 m [ν(amide)],
[5] W. P. Fehlhammer, G. Beck, Chem. Ber. 1989, 122, 1907.
[6] W. P. Fehlhammer, G. Beck, J. Organomet. Chem. 1989, 369,
105.
[7] W. P. Fehlhammer, S. Ahn, G. Beck, J. Organomet. Chem. 1991,
˜
411, 181.
[8] S. Ahn, W. Sperber, W. P. Fehlhammer, J. Organomet. Chem.
1
1609 m [ν(C=C)], 1498 [νas(N–C–N)]. H NMR (CDCl3, 400 MHz):
1995, 489, 27.
δ = 7.42 (m, 2 H, Ph), 7.24 (m, 2 H, Ph), 7.03 (m, 1 H, Ph). 13C{1H}
NMR (CDCl3, 400 MHz): δ = 218.0 (COtrans), 215.3 (COcis), 215.1
[9] S. Ahn, G. Beck, W. P. Fehlhammer, J. Organomet. Chem. 1991,
418, 365.
Z. Anorg. Allg. Chem. 2010, 492–498
© 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.zaac.wiley-vch.de
497

S. Ahn, V. Langenhahn, W. Sperber, W. Beck, W. P. Fehlhammer
ARTICLE
[10] R. Appel, F. Knoll, W. Michel, W. Morbach, H. D. Wihler, H.
1999, 182, 175; F. E. Hahn, M. C. Jahnke, Angew. Chem. Int. Ed.
2008, 47, 3122.
Veltmann, Chem. Ber. 1976, 109, 58.
[11] G. Zinner, W. P. Fehlhammer, Angew. Chem. 1985, 97, 990; [20] a) G. Beck, Ph. D. Thesis, Free University of Berlin, 1987; b) J.
Angew. Chem. Int. Ed. Engl. 1985, 24, 979.
Chatt, A. J. L. Pombeiro, R. L. Richards, G. H. D. Royston,
K. W. Muir, R. Walker, J. Chem. Soc., Chem. Commun. 1975,
708.
[12] G. Zinner, M. Bakola-Christianopoulou, W. P. Fehlhammer in Or-
ganometallic Syntheses (Eds.: R. B. King, J. J. Eisch), Elsevier,
Amsterdam 1988, p. 82.
[21] M. Fritz, D. Rieger, E. Bär, G. Beck, J. Fuchs, G. Holzmann,
W. P. Fehlhammer, Inorg. Chim. Acta 1992, 198–200, 513.
[13] G. Zinner, Ph. D. Thesis, Free University of Berlin 1988.
[14] L. J. Bellamy, The Infrared Spectra of Complex Molecules, Me- [22] W. P. Fehlhammer, K. Bartel, A. Völkl, D. Achatz, Z. Natur-
thuen & Co, London, 1958.
[15] W. P. Fehlhammer, K. Bartel, W. Petri, J. Organomet. Chem. [23] W. P. Fehlhammer, G. Zinner, M. Bakola-Christianopoulou, J.
1975, 87, C34. Organomet. Chem. 1987, 331, 193.
[16] W. P. Fehlhammer, G. Zinner, G. Beck, J. Fuchs, J. Organomet. [24] W. P. Fehlhammer, A. Völkl, U. Plaia, G. Beck, Chem. Ber. 1987,
Chem. 1989, 379, 277. 120, 2031.
[17] a) D. Rieger, S. D. Lotz, U. Kernbach, J. Fuchs, W. P. Fehlham- [25] W. P. Fehlhammer, H. Hoffmeister, U. Eckert, Z. Naturforsch.,
mer, Chem. Ber. 1993, 126, 2243; b) D. Rieger, S. D. Lotz, U. Teil B 1993, 48, 1448.
forsch., Teil B 1982, 37, 1044.
Kernbach, C. André, J. Bertran-Nadal, W. P. Fehlhammer, J. Or- [26] a) A. Völkl, Ph. D. Thesis, University of Erlangen, 1983; b) A.
ganomet. Chem. 1995, 491, 135.
Völkl, G. Beck, W. Beck, W. P. Fehlhammer, publication in prep-
aration.
[18] a) Cf. e.g.:G. C. Pimentel, A. L. McClellan, The Hydrogen Bond,
W. H. Freeman, San Francisco, 1960; b) It should be added, that [27] G. M. Sheldrick, SHELX-86, Univ. of Cambridge, UK, 1986.
the sum of the van-der-Waals radii of N and O is 2.90 Å:N. N. [28] G. M. Sheldrick, SHELX-93, A Program for Crystal Structure De-
Greenwood, A. Earnshaw, Chemie der Elemente, VCH, Weinheim
termination, Göttingen, Germany, 1993.
1988, p. 72; c) The N···O separation compares well with the corre- [29] C. K. Johnson, ORTEP II, Rep. ORNL-5138, Oak Ridge National
sponding distance in a dicarbenerhenium(V)-OPPh3 adduct: F. E.
Hahn, L. Imhof, Organometallics 1997, 16, 763.
Laboratory, Oak Ridge, TN, USA, 1970.
[19] a) F. E. Hahn, M. Tamm, T. Lügger, Angew. Chem. Int. Ed. Engl.
1994, 33, 1356; b) M. Tamm, F. E. Hahn, Coord. Chem. Rev.
Received: June 23, 2009
Published Online: January 14, 2010
498
www.zaac.wiley-vch.de
© 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Z. Anorg. Allg. Chem. 2010, 492–498