Remarkably stable Copper(I) α-carbonyl carbenes: synthesis, structure, and mechanistic studies of alkene cyclopropanation reactions

Article abstract of DOI:10.1021/om8007376

The reaction of [tBu₂P(NSiMe₃)₂-k ²Cu(C₂H₄) (1) with various aryl diazo esters [p-X- C₆H₄]C(N₂)[C(O)R] allowed us to synthesize the corresponding a-carbonyl copper(I) carbene complexes [tBu ₂P(NSiMe3)2-/f2Ar|Cu=C[p-X-C6H₄][C(O)R] (8), where X = OMe, NO,. The rotation barriers around the Cu-C_carbene and C _carbene -C_Ar bonds or their low limits were determined for some of these compounds by ¹H-VT-NMR spectroscopy. Carbene 8g with X = OMe and R = OCH(p-Cl-CI-C₆H₄)₂ was isolated in analytically pure, crystalline form as the first stable representative of this important class of compounds. Its solid-state molecular structure revealed an orthogonal position of the carbene fragment relative to both the ligand plane and the ester C=0 group and a remarkably short Cu-C_carbene distance of 1.822(4) A. Compound 8g reacted with styrene stereoselectively to give the corresponding trans-cyclopropane derivative and [tBu₂P(NSiMe ₃)₂-K²N]Cu(η²-CH ²=CHPh). The stoichiometric cyclopropanation of styrene with 8g and the previously described diarylcarbene [tBu²P(NSiMe ³)2-k²N]Cu=C[p-NO²-C⁶H ⁴]² (6) in toluene-d₈ revealed that the reactions are first order in both the copper carbenes and the alkene. The activation parameters for 8g (?H# = 51.5(9) kJ/mol and ?S# = -127.1(28) J/(mol K)] and for 6 (?H# = 53.4(8) kJ/mol and ?S# = -152.1(23) J/(mol K)) were derived from the kinetics of the cyclopropanation processes. Thermal decomposition of carbene 8g in toluene-d₈ displayed first-order kinetics until 20-25% conversion with activation parameters ?H# = 85.5(24) kJ/mol and ?S# = -49.0(76) J/(mol K). Solutions of 6 in toluene-d₈ also decompose in a first-order fashion with ?H# = 66.1(20) kJ/mol and ?S# =- 125.5(56) J/(mol K). A Hammett study employing 8g and para-substituted styrenes afforded p =- 1.06(19), demonstrating the electrophilic nature of a-carbonyl copper(I) carbene (Fischer-type) complexes. The electronic structure of 8 with X = R = OMe was investigated by DFT methods.

Full text of DOI:10.1021/om8007376

Organometallics 2009, 28, 1049–1059
1049
Remarkably Stable Copper(I) r-Carbonyl Carbenes: Synthesis,
Structure, and Mechanistic Studies of Alkene Cyclopropanation
Reactions†
Igor V. Shishkov, Frank Rominger, and Peter Hofmann*
Organisch-Chemisches Institut der UniVersita¨t Heidelberg, Im Neuenheimer Feld 270,
D-69120 Heidelberg, Germany
ReceiVed July 31, 2008
The reaction of [tBu2P(NSiMe3)2-κ2N]Cu(η2-C2H4) (1) with various aryl diazo esters [p-X-
C6H4]C(N2)[C(O)R] allowed us to synthesize the corresponding R-carbonyl copper(I) carbene complexes
[tBu2P(NSiMe3)2-κ2N]CudC[p-X-C6H4][C(O)R] (8), where X ) OMe, NO2. The rotation barriers around
the Cu-Ccarbene and Ccarbene-CAr bonds or their low limits were determined for some of these compounds
1
by H-VT-NMR spectroscopy. Carbene 8g with X ) OMe and R ) OCH(p-Cl-C6H4)2 was isolated in
analytically pure, crystalline form as the first stable representative of this important class of compounds.
Its solid-state molecular structure revealed an orthogonal position of the carbene fragment relative to
both the ligand plane and the ester CdO group and a remarkably short Cu-Ccarbene distance of 1.822(4)
Å. Compound 8g reacted with styrene stereoselectively to give the corresponding trans-cyclopropane
derivative and [tBu2P(NSiMe3)2-κ2N]Cu(η2-CH2dCHPh). The stoichiometric cyclopropanation of styrene
with 8g and the previously described diarylcarbene [tBu2P(NSiMe3)2-κ2N]CudC[p-NO2-C6H4]2 (6) in
toluene-d8 revealed that the reactions are first order in both the copper carbenes and the alkene. The
activation parameters for 8g (∆Hq ) 51.5(9) kJ/mol and ∆Sq ) -127.1(28) J/(mol K)] and for 6 (∆Hq
) 53.4(8) kJ/mol and ∆Sq ) -152.1(23) J/(mol K)) were derived from the kinetics of the cyclopropanation
processes. Thermal decomposition of carbene 8g in toluene-d8 displayed first-order kinetics until 20-25%
conversion with activation parameters ∆Hq ) 85.5(24) kJ/mol and ∆Sq ) -49.0(76) J/(mol K). Solutions
of 6 in toluene-d8 also decompose in a first-order fashion with ∆Hq ) 66.1(20) kJ/mol and ∆Sq )
-125.5(56) J/(mol K). A Hammett study employing 8g and para-substituted styrenes afforded F )
-1.06(19), demonstrating the electrophilic nature of R-carbonyl copper(I) carbene (Fischer-type)
complexes. The electronic structure of 8 with X ) R ) OMe was investigated by DFT methods.
inactive and had to be reduced in situ either by addition of
hydrazine or by heating with the diazoalkane used.4,5 The role
of cupracyclobutanes remained unclear. According to Aratani,
the results obtained with chiral salicylaldimine ligands were
consistent with intermediate cupracyclobutanes formed from
copper carbene complexes and olefins.4 However, although the
predominant cyclopropane configuration was predicted correctly,
the hypothesis did not adequately explain the preferred trans
selectivity observed in most cases.6 An alternative suggestion
based on copper(I) carbenes as the active cyclopropanating
intermediates was made in 1988 by Pfaltz.5 Utilizing C2-
symmetric semicorrin ligands, the expected enantioselectivities
of cyclopropanes formed by styrene attack upon the electrophilic
carbene center were in agreement with experimental results. The
results could be explained nicely in terms of different steric
repulsion of the R-acyl substituent of the carbene with bulky
groups of the spectator ligand in the diastereomeric transition
states.5 The proposed Pfaltz model (Figure 1) also explained
well the gross influence of the olefin and diazoalkane structures
on the cis/trans selectivity, whereas the effect of the semicorrin
ligand was predicted to be negligible, since the olefinic
substituents were too remote to have any significant interaction
Introduction
Copper catalysis for reactions of diazo compounds with
olefins has been known for about a century,1 but homogeneous
catalysts for this process started to appear only 60 years later.
In 1966, Nozaki, Noyori, et al. used chiral copper(II) salicyl-
aldimine complexes in cyclopropanation reactions of styrene
with ethyl diazoacetate (EDA) as the first application of
enantioselective homogeneous transition-metal catalysis.2,3 Al-
though the reported enantioselectivities were low, the principle
of chiral catalyst development based upon generating and
distinguishing between diastereomeric transition states involving
transition metals was established. Copper(II) carbenes of the
d9-ML5 type were initially proposed to be the active cyclopro-
panating species. However, the importance of copper(I) in the
catalytic cycle was recognized and proven by Aratani and Pfaltz
in the 1980s. Copper(II) complexes as such were shown to be
* To whom correspondence should be addressed. Tel: ++49-6221-
548502. Fax: ++49-6221-544885. E-mail: ph@oci.uni-heidelberg.de.
† Dedicated to Prof. Dietmar Seyferth on the occasion of his 80th
birthday.
(1) For a more recent review of copper-catalyzed cyclopropanation see:
Kirmse, W. Angew. Chem. 2003, 115, 1120; Angew. Chem., Int. Ed. 2003,
42, 1088. DelMonte, A. J.; Dowdy, E. D.; Watson, D. J. Top. Organomet.
Chem. 2004, 6, 97.
(4) Aratani, T. Pure Appl. Chem. 1985, 57, 1839.
(5) Fritschi, H.; Leutenegger, U.; Pfaltz, A. HelV. Chim. Acta 1988, 71,
1553.
(2) Nozaki, H.; Moriuti, S.; Takaya, H.; Noyori, R. Tetrahedron Lett.
1966, 5239.
(3) Nozaki, H.; Takaya, H.; Moriuti, S.; Noyori, R. Tetrahedron 1968,
24, 3655.
(6) Doyle, M. P. In Catalytic Asymmetric Synthesis, 2nd ed.; Ojima, I.,
Ed.; Wiley-VCH: Weinheim, Germany, 2000; p 191.
10.1021/om8007376 CCC: $40.75
2009 American Chemical Society
Publication on Web 01/16/2009

1050 Organometallics, Vol. 28, No. 4, 2009
Hofmann et al.
reactions21swe report here the synthesis and spectroscopy of a
number of R-carbonyl copper(I) carbenes. The isolation of a
remarkably stable representative of this new class of Cu(I)
organometallic compounds in analytically pure form, its struc-
tural characterization, and detailed kinetic studies may help to
unravel the true nature of such a species and to deepen the
understanding of the observed stereo- and enantioselectivities
in cyclopropanation reactions, if instead of our achiral imino-
phosphanamide a chiral spectator ligand is used.5,6,22 Kinetic
studies with the diaryl copper carbene 620 are also presented in
this paper.
Figure 1. Pfaltz’s model for explaining the observed enantiose-
lectivities in copper-catalyzed cyclopropanation.
withthespectatorligand.Severaltheoretical7-11andmechanistic12-14
studies utilizing neutral and cationic copper complexes as
catalytically active species have been performed. Kinetic
investigations by Kochi12 and Perez13,14 ascertained the presence
of a preequilibrium before the rate-determining N2 loss of the
diazoalkane. Copper(I) alkene complexes were found to be the
resting states of the catalytic cycle, and they were the only
isolated or detected intermediates until 2000.
Results and Discussion
The reaction of various R-carbonyl diazo compounds with
the copper olefin complex 1 (Scheme 1) afforded the corre-
sponding deeply colored carbene complexes 8 in steady-state
concentrations significantly higher than that found previously
for complex 3 (16%).17 Although various 2-(p-nitrophenyl)-2-
diazoacetates reacted with 1 at 25 °C to form carbenes 8a,b in
a rather low (<20%) steady-state concentrations, performing
the reaction at 70 °C for 5-10 min raised the Cu(I) carbene
concentrations to 35% and 60% for 8a,b, respectively. Interest-
ingly, ethyl 2-(o-nitrophenyl)-2-diazoacetate, the ortho stereo-
isomer of 8a, reacted with 1 only slowly to form brown and
insoluble, so far unidentified products in most solvents. Amide
7c was more reactive than the esters and gave carbene 8c in a
38.5% steady-state concentration already at room temperature.
Subsequent heating of the mixture to 50 °C only led to an
increase of the concentration to 46%. Note that replacement of
the NMe2 group of 7c by the sterically more demanding NiPr2
group of 7d completely suppresses the formation of the
corresponding NiPr2-substituted carbene 8d even at elevated
temperatures. This clearly demonstrates that large substituents
at the carbonyl carbon may completely block carbene formation
and, consequently, the following cyclopropanation of an olefin.
Carbenes 8b,c reacted with styrene to form exclusively trans
cyclopropanes.
Then, however, the synthesis of remarkably stable, neutral
iminophosphanamide copper(I) olefin complexes such as 17,15
allowed the isolation of the stable copper diazocarbonyl complex
216 and the detection in solution of the first, quite reactive
Fischer-type copper(I) carbene 3, which was active in the
cyclopropanation of styrene (Figure 2).17 Compound 3 is
stabilized by the tailor-made anionic ligand and could be
completely characterized by spectroscopic methods. The in-
tensely colored copper(I) carbene 3 was observed in a steady-
state concentration that precluded its isolation in analytically
pure form and thus hampered a detailed study of its properties
and reactivity. A report of a second (cationic) copper(I) carbene
4 by Barluenga et al. followed soon after,18 although its
reactivity in cyclopropanation was not reported. Recently,
Warren and Dai disclosed the solid-state structure of the
diphenylcopper(I) carbene 5 supported by an anionic ꢀ-diketim-
inate ligand.19a,b They have also demonstrated the possibility
of formation of binuclear µ-CPh2 copper(I) carbenes such as
5a. The closely related, also neutral Cu carbene 6 was prepared
in our group and used in reaction with diazo compounds to
investigate the commonly accepted mechanism of carbene dimer
formation.20 Just last year, Peters has reported the spectroscopic
characterization in solution of another diphenylcopper(I) car-
bene, utilizing a bulky bis(phosphino)borate spectator ligand
for Cu(I).19c
The highly nucleophilic p-methoxyphenyl diazo esters 7e-g
turned out to be very reactive toward complex 1 already at room
temperature. In case of compound 7g, the somewhat complicated
alcohol HOCH(p-ClC6H4)2 has been employed for ester forma-
tion. It is responsible for the high crystallization tendency of
the corresponding Cu carbene 8g from saturated hydrocarbons.
The isolation of 8g is possible in acceptable yield and in
analytically pure form.
Despite the great progress toward unraveling the mechanism
of copper-catalyzed cyclopropanations achieved over the past
decade, the isolation of analytically pure R-carbonyl copper
carbenes for mechanistic studies and their structural character-
ization remained a challenging task.17 Employing the copper
olefin complex 1 and various diazocarbonyl compoundssthe
most commonly used carbene sources in cyclopropanation
1
In H NMR spectra of compounds 8a-g, the two diaste-
reotopic tBu groups at P give rise to two separated doublets
Figure 2

Copper(I) R-Carbonyl Carbenes
Scheme 1a
Organometallics, Vol. 28, No. 4, 2009 1051
a
Steady-state concentrations in solution indicated in parentheses.
Table 1. Selected 13C{1H} and 31P{1H} NMR Spectroscopic Data for
Compounds 3 and 8a-ga
Figure 3. 1H-VT-NMR (500.13 MHz, toluene-d8) spectrum of 8f,
showing coalescence of the ortho and meta aryl protons.
CudC
CdO
δ(31P)
1JPC
2JPSi
12.6
11.2
11.7
11.8
no data
14.4
3
229.9
219.0
221.1
231.7
232.8
235.8
230.5
177.9
178.3
177.9
177.8
177.9
177.0
175.9
64.3
67.7
66.7
67.0
60.7
59.7
59.5
63
Table 2. Experimental Rotation Barriers ∆Gq of the Aryl Group
and the Carbene Unit for Some Copper(I) Carbenes (kJ/mol)
8a
8b
8c
8e
8f
8g
61.7
61.9
62.4
63.7
64.4
64.4
carbene
Ccarbene-CAr bond
36.8 ( 3.0
not measured
49.8 ( 0.9
50.1 ( 1.2
fast on NMR time scale
CudC bond
317
8e
8f
8g
8b
>60
62.2 ( 1.1
68.8 ( 1.9
>67
14.3
a Chemical shifts are given in ppm and coupling constants in Hz.
>81
(3JPH ) ca. 14 Hz), indicating a perpendicular position of the
carbene unit relative to the spectator ligand plane. The carbene
carbon in the 13C{1H} NMR spectra appears in the range of
219-235.8 ppm, verifying the presence of an electrophilic
carbene center. As is obvious from Table 1, an increase of
electron density in the aromatic ring attached to the carbene
carbon leads to a deshielding of the carbene signals (except
amide 8c) and a shielding of the phosphorus resonance in the
31P{1H} NMR spectra. Such a correlation cannot be found for
the carbon signals of the carbonyl groups. The higher electron
donating capacity of the aryl ring also leads to an increase of
1
The low-temperature H NMR spectra of compounds 8e-g
reveal that four protons of the aromatic ring give rise to four
doublets, indicating slow rotation on NMR time scale around
the Ccarbene-CAr bond. Warming the sample in case of 8f leads
to coalescence of the aromatic signals (Figure 3). The activation
barriers ∆Gq for 8f,g derived for this process by means of H-
1
VT-NMR spectroscopy are listed in Table 2. The hindered
rotation observed is a consequence of the donation of electron
density from the electron-rich π-system of the aromatic ring
toward the electrophilic carbene center. In the perpendicular
transition state 8fq, the carbene carbon cannot share electron
density with the aromatic ring. This is in accordance with the
electrophilic character of the carbene carbon centers in 8e-g
and clearly requires a qualitative description of these complexes
as electrophilic (Fischer-type) carbenes.
From Table 2 it is easily seen that an introduction of the
π-donor p-OMe into the aromatic ring of 3 dramatically raises
the aryl group rotation barrier (by 13 kJ/mol), whereas the
π-acceptor NO2 has the opposite effect. This is exactly to be
expected, since the π-system of the aromatic ring in 8f is more
electron-rich than in case of 3, leading to an improved interaction
with the empty p orbital of the carbene carbon and to a
strengthening of the Ccarbene-CAr bond. In complex 8b, the
presence of the NO2 group in the aryl ring lowers its electron
density and the carbene-aryl interaction becomes less pro-
nounced. For this reason, the rotation of the aromatic ring in
compound 8b was found to be fast on the NMR time scale even
at -80 °C.
1
2
the JPC and JPSi coupling constants.
(7) Straub, B. F.; Gruber, I.; Rominger, F.; Hofmann, P. J. Organomet.
Chem. 2003, 684, 124.
(8) Rasmussen, T.; Jensen, J. F.; Østergaard, N.; Tanner, D.; Ziegler,
T.; Norrby, P.-O. Chem. Eur. J. 2002, 8, 177.
(9) Fraile, J. M.; Garc´ıa, J. I.; Mart´ınez-Merino, V.; Mayoral, J. A.;
Salvatella, L. J. Am. Chem. Soc. 2001, 123, 7616.
(10) Bu¨hl, M.; Terstegen, F.; Lo¨ffler, F.; Meynhardt, B.; Kierse, S.;
Mu¨ller, M.; Na¨ther, C.; Lu¨ning, U. Eur. J. Org. Chem. 2001, 11, 2151.
(11) Fraile, J. M.; Garc´ıa, J. I.; Gil, M. J.; Mart´ınez-Merino, V.; Mayoral,
J. A.; Salvatella, L. Chem. Eur. J. 2004, 10, 758.
(12) Salomon, R. G.; Kochi, J. K. J. Am. Chem. Soc. 1973, 95, 3300.
(13) D´ıaz-Requejo, M. M.; Nicasio, M. C.; Pe´rez, P. J. Organometallics
1998, 17, 3051.
(14) D´ıaz-Requejo, M. M.; Belderrain, T. R.; Nicasio, M. C.; Prieto,
F.; Pe´rez, P. J. Organometallics 1999, 18, 2601.
(15) Straub, B. F.; Eisentra¨ger, F.; Hofmann, P. Chem. Commun. 1999,
2507.
(16) Straub, B. F.; Rominger, F.; Hofmann, P. Organometallics 2000,
19, 4305.
(17) Straub, B. F.; Hofmann, P. Angew. Chem. 2001, 113, 1328; Angew.
Chem., Int. Ed. 2001, 40, 1288.
(18) Barluenga, J.; Lo´pez, L. A.; Lo¨ber, O.; Toma´s, M.; Garc´ıa-Granda,
S.; Alvarez-Ru´a, C.; Borge, J. Angew. Chem. 2001, 113, 3495; Angew.
Chem., Int. Ed. 2001, 40, 3392.
The relatively high stability of the copper carbene 8f allowed
the determination of the rotation barrier of the carbene unit
relative to the [PN2Cu] plane. Coalescence of the signals for
(19) (a) Dai, X.; Warren, T. H. J. Am. Chem. Soc. 2004, 126, 10085.
(b) Badiei, Y. M.; Warren, T. H. J. Organomet. Chem. 2005, 690, 5989.
(c) Mankad, N. P.; Peters, J. C. Chem. Commun. 2008, 1061.
(20) (a) Hofmann, P.; Shishkov, I. V.; Rominger, F. Inorg. Chem. 2008,
47, 11755. (b) Shishkov, I. V. Ph.D. Thesis, Universita¨t Heidelberg, 2006.
(21) Doyle, M. P.; McKervey, M. A.; Ye, T. Modern Catalytic Methods
for Organic Synthesis with Diazo Compounds,; Wiley: New York, 1998.
(22) Pfaltz, A. In Transition Metals for Organic Synthesis; Beller, M.,
Bolm, C., Eds.; Wiley: Weinheim, Germany, 1998; Vol. 1, p 100.

1052 Organometallics, Vol. 28, No. 4, 2009
Hofmann et al.
Figure 4. 1H-VT-NMR (500.13 MHz, toluene-d8) spectrum of 8f,
showing the coalescence of the diastereotopic tBu groups.
Figure 5. ORTEP plot of 8g. Selected bond lengths (Å) and angles
(deg): Cu-C(1) ) 1.822(4), Cu-N(1,2) ) 2.003(3), 1.993(3),
P-N(1,2) ) 1.604(3), 1.607(3), C(1)-C(2) ) 1.484(6), C(1)-C(31)
) 1.434(5), C(2)-O(2) ) 1.202(5), C(2)-O(3) ) 1.353(5),
C(31)-C(32) ) 1.418(5), C(32)-C(33) ) 1.370(6), C(33)-C(34)
) 1.402(6), C(34)-O(1) ) 1.353(5); N(2)-Cu-N(1) ) 77.8(1),
P-Cu-C(1))177.9(1),C(1)-Cu-N(2))142.2(1),C(1)-Cu-N(1)
) 140.0(1), C(31)-C(1)-C(2) ) 114.4(3), O(2)-C(2)-C(1) )
125.0(4). Only one hydrogen is shown for clarity. The given
geometric data are mean values of two independent molecules.
the two diastereotopic tBu groups of 8f can be observed by
1H-VT-NMR spectroscopy upon warming of the carbene solu-
tion in toluene-d8 (Figure 4). A Gibbs free activation enthalpy
of ∆Gq ) 68.8 ( 1.9 kJ/mol was found for this process. This
value is significantly higher than the olefin rotation barriers of
37-53.7 kJ/mol found in copper(I) olefin complexes [tBu2P-
(NSiMe3)2-κ2N]Cu(η2-olefin) determined earlier, where olefin
) styrene, norbornene, maleic anhydride.7,15 This indicates that
the carbene moiety is a better π-acceptor than norbornene or
even an η2-coordinated maleic anhydride. The Cu-carbene
rotation barrier in 8e can be determined in a manner similar to
(1.434(5) Å), which thus has partial double-bond character and
leads to the observed slow rotation around the C(1)-C(31) bond
(Table 1) on the NMR time scale. The p-methoxy group lies in
the plane of the aromatic ring, maximizing the conjugation
(C(34)-O(1) ) 1.353(5) Å).
1
that for 8f by means of H-VT-NMR spectroscopy. Here an
activation barrier of ∆Gq ) 62.2 ( 1.1 kJ/mol was found for
this dynamic process. This value is ca. 6 kJ/mol lower than in
the case of 8f, most probably because of the steric demand of
the tBu group in the latter.
Interestingly, the carbonyl group of 8g is oriented perpen-
dicular to the carbene moiety (angle between the [Cu(1),C(1),
C(2),C(31)] and [C(1),C(2),O(2),O(3)] planes: 86.8°). This
should be the case for electronic reasons, because the CdO
group tries to adopt an orientation with minimized interaction
of its acceptor π-system with the “empty” p orbital of the
carbene carbon. In fact, such an orientation of the carbonyl
function has been predicted in theoretical calculations7-9 and
is consistent with a “carbocationic” character of the carbene
carbon. However, this effect only persists in the solid-state
structure, since in solution there is facile rotation around
the C(1)-C(2) bond even at -60 °C on the NMR time scale.
The remarkably short CudC distance (1.822(4) Å) in 8g is
indicative of a double bond and is the shortest copper-carbon
distance that has been reported so far. It is slightly shorter than
the one reported for Warren’s carbene 5 (1.834(3) Å)19 and
significantly shorter than found in Barluenga’s cationic copper
carbene (1.882(3) Å)18 or for many copper(I) carbenes with
Wanzlick-Arduengo carbene ligands23,24 with their minimal
π-acceptor character. The structural features of carbene 8g are
nicely consistent with the mechanistic scenario proposed by
Pfaltz5 for explaining and predicting product distributions in
enantioselective Cu-catalyzed cyclopropanations.
The magnitude of the carbene fragment rotation barrier should
reflect the significance of the copper to carbene back-bonding.
This, in turn, should also be a function of the variable electronic
properties of the aromatic ring at the carbene carbon. This
assumption is very well demonstrated by comparison of the
experimental ∆Gq values obtained for the sterically equivalent
compounds 8b,f. Replacing the strongly π-donating p-OMe
group in 8f by the strongly π-accepting one (p-NO2) results in
an increase of the rotation barrier of the carbene moiety by more
than 12 kJ/mol (lower limit) because of the strengthened Cu-C
back-bonding. The exact ∆Gq value for compound 8b cannot
be determined, as it decomposes too rapidly above 80 °C.
Nevertheless, the low limit of the rotation barrier of the carbene
fragment in 8b was established to be at least 81 kJ/mol.
Compound 8g turned out to be the first stable Cu(I)
R-carbonyl carbene, which was isolated in analytically pure form
and as crystalline material. The compound crystallizes with two
independent molecules in the unit cell, and Figure 5 shows only
one of them. In the solid state, the copper atom is in a trigonal-
planar environment as expected, and the atoms Cu, C(1), C(2),
and C(31) share the same plane. The angle P-Cu-C(1) is
177.9°, and the carbene unit is nearly orthogonal to the [PN2Cu]
plane, maximizing back-bonding (the angle between planes
[PN2Cu] and [Cu,C(1),C(2),C31] is 97.7°). Notably, the aromatic
ring is twisted by only 5.9° relative to the [C(31),C(1),C(2)]
plane. This fact could play a significant role in the stabilization
of 8g due to the interaction of the π system of the electron-rich
aryl ring with the empty p orbital of the carbene carbon. This
interaction is also reflected in a rather short C(1)-C(31) bond
It is remarkable that, according to NMR data and elemental
analysis, compound 8g crystallized together with pentane in a
(23) Raubenheimer, H. G.; Cronje, S.; Olivier, P. J. J. Chem. Soc., Dalton
Trans. 1995, 313.
(24) Raubenheimer, H. G.; Cronje, S.; van Rooyen, P. H.; Olivier, P. J.;
Toerien, J. G. Angew. Chem. 1994, 106, 687; Angew. Chem., Int. Ed. Engl.
1994, 33, 672.

Copper(I) R-Carbonyl Carbenes
Organometallics, Vol. 28, No. 4, 2009 1053
1:0.46 ratio, and the cocrystallized solvent could not be removed
even at 10-3 mbar. The ratio was found to be 1:0.48 in a second
synthesis of 8g under the same reaction conditions.
by Woo and co-workers for explaining the high stereoselectivity
observed in cyclopropanations of styrene with mesityldiazo-
methane catalyzed by the iron porphyrin complex Fe(TTP) (TTP
) meso-tetra-p-tolylporphyrin).26 The reaction of pent-1-yne
with 8g yielded, as expected from the acidity of the alkyne, the
free iminophosphanamide ligand and yellow copper acetylide.
Dimethyl acetylenedicarboxylate was unreactive toward copper
carbene 8g. Complex 8g is the first stable, isolated, and
analytically pure copper(I) R-carbonyl carbene which could be
shown to be active in cyclopropanations of olefins.
Although complex 8g is rather stable, its solutions in toluene-
d8 slowly decompose at room temperature to give the free
iminophosphanamide ligand tBu2P(NSiMe3)(NHSiMe3) (δ(31P)
30.1), a solvated copper iminophosphanamide fragment
[tBu2P(NSiMe3)2-κ2N]Cu(toluene-d8) (δ(31P) 63.3, which is very
close to that of the corresponding [tBu2P(NSiMe3)2-κ2N]-
Cu(C6D6) with δ(31P) 63.2) and several other phosphorus-
containing products, which are difficult to identify. The 1H NMR
spectrum of 8g after thermal decomposition displayed a large
number of different products and was not informative. Possibly,
the decomposition proceeds via reactive copper ketenes (after
a Wolff rearrangement), although such a hypothesis has not been
confirmed. Interestingly, our diaryl carbene 6 also decomposed
in toluene-d8 solutions to yield only tetrakis(p-nitrophenyl)eth-
ylene, the “carbene dimer” [tBu2P(NSiMe3)2-κ2N]Cu(toluene-
d8), and some free ligand as the main products. Solutions of 8g
are stable at low temperature; in the solid state, compound 8g
possesses even greater stability and can be stored without
decomposition at -25 °C in a glovebox for several months.
The kinetic stabilization by the steric demand and the high
basicity of the anionic iminophosphanamide σ- and π-donor
ligand, the enforced small N-Cu-N bite angle of 77.8° of the
four-membered neutral (chelate)Cu ring system, the strong
π-donor substituent OMe of the carbene phenyl ring, and the
sterically demanding special ester group are responsible for
the unprecedented stability of 8g. The high-lying HOMO of
the small-bite-angle d10-ML2 fragment25 [tBu2P(NSiMe3)2-
κ2N]CuI results in pronounced π back-bonding from Cu(I)
toward the carbene which, together with a strong Cu(I)-Ccarbene
σ-bond resulting from a low-lying LUMO at the copper
fragment, leads to a strong and short CudC bond. It should be
noted here that the characterization of the tailor-made Cu(I)
fragment [tBu2P(NSiMe3)2-κ2N]CuI as a “good” back-bonding
fragment is not meant on an absolute but on a relative scale,
specific for copper(I) units of the type d10-L2Cu. In comparison
to many other back-bonding isolobal and isoelectronic metal
fragments its π-donor capability toward π-acceptor ligands is
weak. Nevertheless, the anionic iminophosphanamide spectator
ligand causes the [tBu2P(NSiMe3)2-κ2N]Cu(I) building blocks
isolobal also with a singlet methylenesto function as a strong
π-donor relative to other cationic or neutral d10-L2Cu fragments.
[tBu2P(NSiMe3)2-κ2N]CuI therefore is particularly well suited
for stabilizing metal-ligand coordination in terms of the
Dewar-Chatt-Duncanson picture, sufficiently stabilizing the
first stable R-carbonyl copper(I) carbenes 8.
Kinetic studies with R-carbonyl carbene 8g and diaryl carbene
27
1
6 were performed by means of H NMR spectroscopy. For
product analysis, we assume here that the decrease of the
copper(I) carbene concentration in the presence of an olefin is
a result of two independent processes: carbene decomposition
and cyclopropanation of the olefin by the copper carbene, i.e.
-d[carbene] ⁄ dt ) kcyclop[carbene][olefin] + kdec[carbene]n
If the decomposition proceeds in a first-order fashion (n )
1) and the olefin concentration is kept constant (pseudo-first-
order conditions), one may write
-d[carbene] ⁄ dt ) [carbene]{kpseudo+ kdec} ) [carbene]kobs
where
kpseudo ) kobs - kdec ) kcyclop[olefin]
(kcyclop ) actual second-order cyclopropanation rate constant,
kdec ) carbene decomposition rate constant, kobs ) observed
first-order cyclopropanation rate constant, kpseudo ) pseudo-first-
order constant corrected with respect to decomposition).
Solutions of carbene 8g in toluene-d8 decompose in a first-
order manner until 21-25% conversion is reached. Upon further
conversion, a strong deviation from first-order kinetics was
observed: i.e., the rate of decomposition became considerably
faster than that after ca. 3000 s. Presumably, such a deviation
may result from a reaction of 8g with its decomposition
products. Decomposition of diaryl carbene 6 could be ap-
proximated with the first-order kinetics if data points until ca.
1.5 half-lives were collected. The first-order constants for
decomposition obtained over a temperature range of 303-333
K (for 8g) or 343-373 K (for 6) allowed the construction of
Eyring plots28 which afforded the activation parameters listed
in Table 3. These values may be compared to those found for
the decomposition of Warren’s carbene 5.19 Note that ∆Hq
a,b
for the carbene loss in 6 is much lower than ∆Hq in the
R-carbonyl carbene case 8g. The large negative entropy of
activation in all cases can be attributed to solvent effects. It
implies an associative mechanism with a possible coordination
of the solvent in the transition state. Indeed, our studies of
Carbene 8g reacted stereoselectively with an excess of styrene
to form the trans-cyclopropane derivative (i.e., ester and the
phenyl substituents are trans to each other) and the copper
styrene complex [tBu2P(NSiMe3)2-κ2N]Cu(η2-CH2dCHPh) in
a nearly 1:1 ratio, indicating almost quantitative conversion of
8g to cyclopropane. It is reasonable to assume that the large
substitutes at the carbene center are responsible for the high
trans selectivity observed. In addition, another factor favoring
the formation of the trans product could be arene π-π
interactions in the transition state between the p-methoxyphenyl
substituent of the carbene carbon and the phenyl ring of styrene.
The importance of such π-π interactions was suggested earlier
(26) Hamaker, C. G.; Mirafzal, G. A.; Woo, K. L. Organometallics 2001,
20, 5171.
(27) The concentration of 8g was monitored by the decrease of the
integral intensity of the singlet at δ 3.07 ppm assigned to the methoxy group
of 8g. The kinetic studies with diarylcarbene 6 were performed in a similar
manner, except that the integral intensity of the doublet at δ 8.03 ppm
assigned to the four aromatic ortho protons was followed to control the
carbene concentration. The triplet at δ 1.00 ppm and the quintet at 2.21
ppm were assigned to pentane and C6D5CD2H (toluene), respectively, and
were used as convenient internal standards. The target peaks and the
reference signals for the kinetic measurements did not overlap with any
decomposition or cyclopropanation products. Only in the case of p-methoxy-
substituted styrene did the reference signal at δ 2.21 (toluene) overlap with
the signal of the corresponding cyclopropanation product. In this case a
methyl resonance of trace pentane was used to follow the reaction.
(28) See the Supporting Information.
(25) Hofmann, P.; Heiss, H.; Mu¨ller, G. Z. Naturforsch., B 1987, 42,
395. The Walsh diagram described in this paper for the frontier orbital
energy changes of d10-L2M metal fragments as a function of their L-M-
L angle has been reported multiple times in the literature. .

1054 Organometallics, Vol. 28, No. 4, 2009
Hofmann et al.
Table 3. Activation Parameters for Thermal Decomposition and
Cyclopropanation of Styrene with Carbenes 6 and 8g and with
Warren’s Carbene 5
Hammett-type. The results presented in Tables 6 and 7 clearly
demonstrate the electrophilic nature of the copper(I) carbenes.
Styrenes with electron-donating para substituents accelerate the
cyclopropanation rate, whereas the electron-deficient p-(trifluo-
romethyl)styrene has the opposite effect. This is in accordance
with theoretical model calculations which predicted a substan-
tially higher activation barrier for cyclopropanation of ethylene
compared to the more electron-rich vinyl alcohol.7 The reaction
of the diarylcarbene 6 with p-CF3-C6H4CHdCH2 proceeded so
slowly that only small amounts of the corresponding cyclopro-
pane were detected by 1H NMR spectroscopy, and the decom-
position products dominated in the reaction mixture. For that
reason, it was not possible to obtain reliable first-order cyclo-
propanation constants for this olefin. Nevertheless, the data
indicate qualitatiVely that electron-poor olefins are also less
reactive in cyclopropanation with diarylcarbenes such as 6.
Interestingly, Warren and Dai have reported that p-(trifluorom-
ethyl)styrene had little influence on the reaction rate compared
to styrene in cyclopropanations employing diphenyl copper
carbene 5.19a We cannot rationalize this observation at this point.
A reasonable linear correlation in a log(kx/kH) vs standard
σ+ substituent constants Hammett plot (Figure 6) was found
with F ) -1.06(19). This value may be compared to that
reported for cyclopropanation of para-substituted styrenes with
EDA and neutral copper(I) complexes such as TpMeCuC2H4 (F
) -0.85(7), TpMe ) hydrotris(3,5-dimethyl-1-pyrazolyl)bo-
rate)30 and with -1.23 to -0.99 (with an average of -1.11)
for a BpCuL series [Bp ) dihydrobis(1-pyrazolyl)borate].13 The
Hammett reaction constant F determined for 8g is substantially
larger (more negative) than the one derived for the cyclopro-
panation with a cationic bis-oxazoline copper(I) complex (F )
-0.51),8 and for the stoichiometric reaction of para-substituted
styrenes with the isolated highly electrophilic carbene
{Fe(TPFPP)[C(Ph)(CO2Et)]} (F ) -0.41(5), TPFPP ) meso-
tetrakis(pentafluorophenyl)porphyrin).31
∆Hq, kJ/mol
∆S , J/(mol K)
∆Gq, kJ/mola
q
8g
6
dec
cyclop
dec
85.5(24)
51.5(9)
66.1(20)
53.4(8)
87.8(42)
43.5(13)
-49.0(76)
-127.1(28)
-125.5(56)
-152.1(23)
-33.4(125)
-135.0(38)
100.1(48)
89.4(18)
103.5(37)
98.7(15)
97.8(80)
83.7(24)
cyclop
5
decb
cyclopb
a At 298 K. b See ref 19a,b.
solvent effects show that the decomposition of 8g in coordinating
solvents such as THF-d8 proceeds appreciably faster than in
toluene-d8.
Carbenes 8g and 6 slowly react with an excess of styrene
and exhibit clean pseudo-first-order kinetics in toluene-d8. The
observed rate constants are doubled upon doubling the concen-
tration of styrene in the reaction mixtures, indicating that the
reaction is also first order in styrene and second order overall.
The actual rate constants measured at various temperatures
allowed the construction of the corresponding Eyring plots, from
which the activation parameters of the carbene decomposition
and the cyclopropanation were derived (Table 3). These values
are in full agreement with an associative mechanism and are
somewhat different from those reported for the stoichiometric
cyclopropanation of styrene with Warren’s terminal carbene 5
and from the cyclopropanation of vinyl acetate with (CO)5Wd
CH(p-MeOC6H4) (∆Hq ) 42.0(13) kJ/mol and ∆Sq ) -143.6(29)
J/(mol K)).29
Comparison of the activation parameters for styrene cyclo-
propanation and carbene decomposition reveals a significantly
larger ∆Hq for carbene loss relative to carbene transfer to
styrene. This can be expected, since the breaking of the strong
CudC bond (calculated De of 257.9-312.4 kJ/mol7) is not yet
compensated by the C-C bond formation in the transition state.
The experimentally derived activation barriers ∆Gq for
cyclopropanation of styrene are 40-50 kJ/mol higher than those
found in theoretical calculations on small neutral model sys-
tems.7 However, this difference is understandable, as the large
substituents of the spectator ligand as well as at the carbene
carbon should definitely destabilize the corresponding associa-
tive transition states.
The actual second-order rate constants determined for the
cyclopropanation of styrene with 8g over the temperature range
of 303-333 K exceed the decomposition rates obtained at these
temperatures by a factor of 17-62 (Table 4). This rate differen-
ce rationalizes the efficient conversion of the carbene to the
corresponding cyclopropane. On the other hand, the second-
order rate constants determined for cyclopropanation of styrene
with 6 over a temperature range of 323-373 K (Table 5) exceed
the decomposition rates obtained at the same temperatures by
only a factor of 2-5. This is consistent with the appearance of
significant amounts of the carbene dimer tetrakis(p-nitrophe-
nyl)ethylene in the reaction of 6 with styrene20 and underscores
the need to use an excess of an alkene under typical catalytic
cyclopropanation conditions with a low concentration of the
copper catalyst.
For both carbenes 6 and 8g, the data points fit better to σ+
than to σ. This is to be expected, since reactions which proceed
along with the concomitant development of an electron defi-
ciency are better correlated with substituent constants based upon
e.g. an SN1 bond breaking reaction than the ionization of benzoic
acids.32 In conjunction with a relatively small negative F value
of -1.06 obtained for carbene 8g, these data suggest an early
associative transition state with only a moderate positive charge
being developed at the benzylic carbon (Figure 7). Therefore,
the picture of the transition state derived for copper carbenes
from our experimental results is also in good agreement with
the Pfaltz model5 and with the results obtained from previous
theoretical calculations.7-9
DFT calculations28 were employed for copper carbene 8e.
This system is very similar to the X-ray characterized carbene
8g, where only the larger ester group -COOCH(p-ClC6H4)2
replaces -COOCH3. BP86/SV(P) geometry optimization of 8e
yielded a minimum energy structure with an orthogonal position
of the carbene moiety relative to the PN2Cu planes, as was found
earlier for high-level DFT calculations utilizing smaller imino-
phosphanamide model systems.7
The geometrical features of the optimized copper carbene 8e
are in reasonable agreement with the solid-state structure of 8g,
Complexes 8g and 6 were also utilized in cyclopropanation
reactions of para-substituted styrenes p-R-C6H4CHdCH2 (R )
OCH3, CH3, H, CF3) in order to test the electronic influence of
olefin substrate structures via potential LFE correlations of the
(30) D´ıaz-Requejo, M. M.; Pe´rez, P. J.; Brookhart, M.; Templeton, J. L.
Organometallics 1997, 16, 4399.
(31) Li, Y.; Huang, J.-S.; Zhou, Z.-Y.; Che, C.-M.; You, X.-Z. J. Am.
Chem. Soc. 2002, 124, 13185.
(32) Connors, K. A. Chemical Kinetics; VCH: Weinheim, Germany,
1990.
(29) Fischer, H.; Mauz, E.; Jaeger, M. J. Organomet. Chem. 1992, 427,
63.

Copper(I) R-Carbonyl Carbenes
Organometallics, Vol. 28, No. 4, 2009 1055
Table 4. Observed First-Order and Actual Second-Order Rate Constants for Styrene Cyclopropanation with 8g over a Temperature Range of
303-333 K at a 20:1 Olefin/Carbene Ratio under Pseudo-First-Order Conditions ([styrene] ) [2.75(1) × 10-1 M)
T, K
kobs, s-1
kdec, s-1
kpseudo, s-1
kcyclop, L/(mol s)
303
313
320
327
333
[5.94(6)] × 10-4
[1.160(4)] × 10-3
[1.89(1)] × 10-3
[3.09(4)] × 10-3
[4.64(4)] × 10-3
[3.27(9)] × 10-5
[9.71(21)] × 10-5
[2.02(6)] × 10-4
[3.83(24)] × 10-4
[8.09(29)] × 10-4
[5.61(6)] × 10-4
[1.063(5)] × 10-3
[1.69(1)] × 10-3
[2.71(5)] × 10-3
[3.83(5)] × 10-3
[2.04(2)] × 10-3
[3.87(2)] × 10-3
[6.14(5)] × 10-3
[9.85(17)] × 10-3
[1.39(2)] × 10-2
Table 5. Observed First-Order and Actual Second-Order Rate Constants for Styrene Cyclopropanation with 6 over a Temperature Range of
303-333 K at a 40:1 Olefin/Carbene Ratio under Pseudo-First-Order Conditions ([styrene] ) [6.26(2)] × 10-1 M)
T, K
kobs, s-1
kdec, s-1
kpseudo, s-1
kcyclop, mol-1 s-1
323a
343
353
363
373
[1.51(6)] × 10-4
[5.37(8)] × 10-4
[9.995(75)] × 10-4
[1.78(4)] × 10-3
[3.08(5)] × 10-3
[3.93(16)] × 10-5
[1.80(7)] × 10-4
[3.40(10)] × 10-4
[6.44(26)] × 10-4
[1.27(5)] × 10-3
[1.11(6)] × 10-4
[3.57(11)] × 10-4
[6.60(12)] × 10-4
[1.13(5)] × 10-3
[1.81(8)] × 10-3
[1.78(10)] × 10-4
[5.71(17)] × 10-4
[1.06(2)] × 10-3
[1.81(8)] × 10-3
[2.89(12)] × 10-3
a kdec at 323 K was obtained by extrapolation of the Eyring plot for decomposition of 6.
1.888(6) Å)23,24,33 and falls within the lower range for the Cu-C
bonds calculated for Wanzlick-Arduengo carbene copper
complexes in MP234 or BP8635 calculations (1.84-1.89 Å).
However, the CudC separation calculated for 8e is significantly
larger compared to the values computed for a small iminophos-
phanamide model system such as [H2P(NH)2-κ2N]Cud
CH(HCdO) (1.794 Å)7 and for a cationic copper carbene model
complex (1.782 Å).9 Such a discrepancy can be first of all
attributed to steric reasons.
The calculated HOMO-1 of 8e (Figure 8) shows π-back-
bonding from the copper atom to the carbene carbon. A
population analysis reveals that the copper-carbon π-bond is
localized predominantly on copper (76.9%) mainly as a d-orbital
contribution (85.1%). Mixing with symmetry-equivalent s and
p orbitals was found to be <1% and 14.2%, respectively. The
carbon part of this bond (23.1%) is represented by a pure p
orbital (100%). A population analysis for 8e gave a Mulliken
charge of the carbene carbon of -0.0163. This bonding pattern
clearly identifies 8e as an electrophilic Fischer-type carbene.
The LUMO of 8e has its major coefficient at the carbene
fragment’s carbon atom as an “empty” p orbital. These results
are in agreement with earlier calculations on neutral ꢀ-diketimi-
nate-supported diphenylcarbene copper complexes.19b The
intense color of the d10 complexes 8 can also be explained on
the basis of these features.17
Table 6. Rate Constants for Cyclopropanation of Para-Substituted
Styrenes p-R-C6H4CHdCH2 by Copper Carbene 8g at 303 K and a
1:20 Carbene/Olefin Ratio
R
kobs, s-1
kdec, s-1
kpseudo, s-1
kX/kH
H
CF3
CH3
[6.63(9)] × 10-4 [3.27(9)] × 10-5 [6.30(9)] × 10-4 1.000
[3.20(3)] × 10-4 [3.27(9)] × 10-5 [2.87(3)] × 10-4 0.4559
[1.75(2)] × 10-3 [3.27(9)] × 10-5 [1.72(2)] × 10-3 2.725
OCH3 [8.64(56)] × 10-3 [3.27(9)] × 10-5 [8.61(56)] × 10-3 13.661
Table 7. Rate Constants for Cyclopropanation of Para-Substituted
Styrenes p-R-C6H4CHdCH2 by Copper Carbene 6 at 343 K and a
1:40 Carbene/Olefin Ratio
R
kobs, s-1
kdec, s-1
kpseudo, s-1
kX/kH
H
CH3
[5.37(8)] × 10-4 [1.80(7)] × 10-4 [3.57(11)] × 10-4 1.00
[1.18(1)] × 10-3 [1.80(7)] × 10-4 [1.00(2)] × 10-3 2.808
OCH3 [4.47(14)] × 10-3 [1.80(7)] × 10-4 [4.29(14)] × 10-3 12.02
Figure 6. Superimposed Hammett plots from cyclopropanation
reactions of p-R-substituted styrenes using compounds 6 and 8g.
Figure 8. Graphical representation of HOMO-1 (left) and LUMO
(right) of copper carbene 8e.
Conclusions
Employment of a tailor-made, sterically demanding, electron-
rich iminophosphanamide ligand leads to a sufficient stabiliza-
tion of a number of R-carbonyl copper(I) carbenes 8 to allow
Figure 7. Qualitative transition state representation for styrene
cyclopropanations with copper carbenes 8.
although the CudC, Cu-N, and P-N bonds are slightly too
long on the BP86/SV(P) level. The calculated copper-carbene
bond in 8e (1.850 Å) is 0.02-0.04 Å shorter than in some
copper(I) N-heterocyclic carbene complexes (1.868(6) and
(33) Arduengo, A. J.; Dias, H. V. R.; Calabrese, J. C.; Davidson, F.
Organometallics 1993, 12, 3405.
(34) Boehme, C.; Frenking, G. Organometallics 1998, 17, 5801.
(35) Nemcsok, D.; Wichmann, K.; Frenking, G. Organometallics 2004,
23, 3640.

1056 Organometallics, Vol. 28, No. 4, 2009
Hofmann et al.
equivalent portions, and kept at -78 °C. All kinetic studies were
performed by 1H NMR spectroscopy on the Bruker 500 MHz NMR
spectrometer equipped with an internal thermocouple sensor. The
sample was allowed to reach thermal equilibrium for ca. 4 min
prior to measurements. The errors of the rate constants and the
activation parameters were calculated as described in ref 32.
Ethyl Diazo(2-nitrophenyl)acetate.39 To a well-stirred solution
of ethyl (2-nitrophenyl)acetate (300 mg, 1.434 mmol), Bu4NBr (0.28
eq., 130 mg, 0.403 mmol), 18-crown-6 (13 mg), and TsN3 (230
mg) in 15 mL of C6H6 was added dropwise a 20% aqueous solution
of KOH (15 mL). The reaction mixture was stirred at 38 °C for
1 h, and the benzene phase slowly acquired a yellow color. The
temperature was maintained at 40 °C for 1.5 h, and the organic
layer changed its color to red. TLC control showed almost complete
conversion of the starting material. An additional 40 mg of TsN3
was added, and stirring was continued at 38 °C for 40 min. Et2O
was added, and the organic phase was separated. The water
phase was diluted with more H2O (20 mL) and extracted with Et2O
(15 mL). The combined organic extracts were washed with H2O
(20 mL) and brine (10 mL), dried over Na2SO4, and evaporated to
give a yellow oil which was purified by column chromatography
(SiO2, EtOAc-hexane 1:5) to yield 0.26 g (77%) of a yellow-orange
oil. Crystallization from EtOH gave an analytically pure sample.
Mp: 48 °C. 1H NMR (250 MHz, CDCl3, 298 K): δ 7.82-7.85 (m,
1H, arom), 7.42-7.49 (m, 1H arom), 7.24-7.36 (m, 2H, arom),
their characterization in solution and in two cases even their
isolation in analytically pure form. Their relative stability
enabled us to determine the carbene fragment vs Cu(I) metal
fragment rotation barriers, which indicated a significant back-
bonding character of the CudC bond. The electron-rich aryl
ring at the carbene carbon decreases the metal to carbene back-
bonding but stabilizes the carbene center by means of a donor
interaction of the aryl ring π-system with the carbene’s empty
p orbital. The solid-state structure of carbene 8g is in agreement
with the Pfaltz model of cyclopropanation stereochemistry and
with predictions from quantum chemistry. To our knowledge,
compound 8g is the first stable structurally characterized
R-carbonyl copper carbene. Reaction of this carbene with styrene
proceeds stereoselectively and affords trans-cyclopropane and
a copper styrene complex in ca. 1:1 ratio. Kinetic experiments
with pure 8g and previously characterized diaryl carbene 6 are
fully consistent with an associative mechanism. The relative
small negative F values derived from the Hammett plot for 8g
suggest an early transition state where only a partial positive
charge is building up at the benzylic carbon in the rather early
transition state. DFT calculations performed on the full system
8e are consistent with the description of this compound as an
electrophilic Fischer-type d10-ML2(carbene) complex.
Finally we note that copper ethylene complex 1 also catalyzes
the aziridination of styrene with PhIdNTs as a nitrene source,
and synthetic as well as mechanistic and theoretical studies
devoted to the characterization of the presumably highly
electrophilic nitrenes bound to the copper-iminophosphanamide
fragment will be a subject of further studies in our laboratory.
3
3
4.08 (q, JHH ) 7.1 Hz, OCH2), 1.09 (t, JHH ) 7.1 Hz, CH3).
tert-Butyl (4-Nitrophenyl)acetate.40 A 100 mL flask equipped
with a calcium chloride drying tube and a magnetic stirring bar
was charged with (4-nitrophenyl)acetic acid (5 g, 27.6 mmol), dry
CH2Cl2 (50 mL), tBuOH (8 mL, 82.20 mmol), and DMAP (2.70
g, 22.08 mmol). The solution was stirred and cooled in an ice bath
to 0 °C while carboxydiimide (8.20 g, 39.75 mmol) was added in
three portions. The reaction mixture was stirred at 0 °C for 15 min,
and then the cooling bath was removed and stirring was continued
at room temperature for 34 h. During this time, a white precipitate
was formed. The reaction mixture was filtered through a layer of
Celite. The Celite was washed with CH2Cl2, and the combined
organic fractions were washed with cold concentrated citric acid
(2 × 30 mL), cold saturated NaHCO3 (2 × 30 mL), and brine (30
mL), dried over Na2SO4, and evaporated. The product was purified
by column chromatography (SiO2, EtOAc-hexane 1:7) to yield
6.05 g (75%) of a colorless liquid, which crystallized on storage in
the refrigerator. An analytical sample was obtained by recrystal-
lization of the product from hexane. Yield: 4.24 g (65%). Mp: 55
Experimental Section
Unless otherwise stated, all experiments were performed under
an Ar atmosphere using a Braun glovebox or by Schlenk techniques.
Tetrahydrofuran, pentane, hexane, toluene, and diethyl ether were
distilled over benzophenone sodium-ketyl and kept under Ar.
Toluene-d8 was stirred over CaH2 and then degassed three times
prior to use. Tetrahydrofuran-d8 was distilled over sodium and
degassed three times. Pyridine, piperidine, iPr2NH, and NEt3 were
distilled under Ar over KOH. DBU was purified by vacuum
distillation. Commercially available EDA was distilled under
vacuum prior to use and kept under Ar. Styrene and p-methylstyrene
were distilled under vacuum prior to use and degassed three times.
p-Methoxystyrene and p-(trifluoromethyl)styrene were passed through
a short neutral Al2O3 column and degassed three times. All other
commercially available reagents, including ABSA (p-acetamido-
benzenesulfonyl azide), were used as received. Solutions of LDA
were freshly prepared from iPr2NH and nBuLi and used im-
mediately. Ethyl diazo(4-nitrophenyl)acetate,36 TsN3 (tosyl azide),37
TryN3 (2,4,6-triisopropylbenzenesulfonyl azide),38 and complex 115
were prepared according to known procedures.
1
3
°C. H NMR (250 MHz, CDCl3, 298 K): δ 8.17 (d, JHH ) 8.75
3
Hz, 2H, arom), 7.43 (d, JHH ) 8.75 Hz, 2H, arom), 3.63 (s, 2H,
ArCH2), 1.43 (s, 9H, C(CH3)3). Anal. Calcd for C12H15NO4: C,
60.75; H, 6.37; N, 5.90. Found: C, 60.72; H, 6.38; N, 5.90.
tert-Butyl Diazo(4-nitrophenyl)acetate (7b). To a mixture of
tert-butyl (4-nitrophenyl)acetate (1.5 g, 6.325 mmol) and ABSA
(1.5 equiv, 2.28 g, 9.488 mmol) in dry THF (10 mL) was added
Et3N (1.5 equiv, 1.32 mL, 9.488 mmol). The reaction mixture
became yellow-brown. After 10 min of stirring at room temperature
TLC showed the reaction to be not yet complete. DBU (1.5 equiv,
1.42 mL, 9.488 mmol) was added slowly (exothermic reaction),
and the reaction mixture was stirred at room temperature for another
10 min, until TLC showed the reaction to have gone to completion.
H2O was added, and the product was extracted two times with Et2O.
The combined ether fractions were washed with concentrated citric
acid (2 × 30 mL) solution, water (2 × 30 mL), and brine, dried
(Na2SO4), filtered, and evaporated to give a yellow oil. Column
chromatography (EtOAc-hexane 1:7) afforded a yellow solid (1.62
g, 97.6%) which can be recrystallized from EtOAc-hexane 1:7
The NMR spectra were recorded on Bruker 500, 300, and 250
1
MHz spectrometers at temperatures specified below. H and 13C
NMR chemical shifts are reported in parts per million and are
referenced to the deuterated solvent used. 31P NMR spectra are
calibrated either with external 85% H3PO4 or, where possible, by
traces of free iminophosphanamide ligand (δ 30.1). Mass spectra
were recorded on a JMS 700 instrument (JEOL). Elemental analyses
were performed in the “Mikroanalytisches Laboratorium der
Chemischen Institute der Universita¨t Heidelberg”.
The solutions of copper carbenes 6 and 8g and olefins for kinetic
measurements were prepared using a volumetric flask, divided in
(39) Prepared by the method of: Lombardo, L.; Mander, L. N. Synthesis
1980, 368, which has been slightly modified.
(40) Prepared by the method of: Neises, B.; Steglich, W. Organic
Syntheses; Wiley: New York, 1992; Collect. Vol. 70, p 93, which was
slightly modified.
(36) Regitz, M. Chem. Ber. 1965, 1210.
(37) Curphey, T. G. Org. Prep. Proced. Int. 1981, 13 (2), 112.
(38) Harmon, R. E.; Wellman, G.; Gupta, S. K. J. Org. Chem. 1973,
38, 11.

Copper(I) R-Carbonyl Carbenes
Organometallics, Vol. 28, No. 4, 2009 1057
with addition of small portions of CHCl3. Yield after recrystalli-
zation: 1.46 g (88%), light-yellow solid. Mp: 118-119 °C. 1H NMR
(500.13 MHz, CDCl3, 298 K): δ 8.19 (d, 3JHH ) 8.95 Hz, 2H, arom),
7.63 (d, 3JHH ) 8.95 Hz, 2H, arom), 1.55 (s, 9H, C(CH3)3). 13C{1H}
NMR (125.77 MHz, CDCl3, 298 K): δ 162.9 (CdO), 144.8, 134.6,
124.2, 123.1 (all arom H), 83.2 (C(CH3)3), 28.3 (C(CH3)3). Anal.
Calcd for C12H13N3O4: C, 54.75; H, 4.98; N, 15.96. Found: C, 54.72;
H, 4.91; N, 16.00.
solution of TsN3 (1.15 equiv, 420 mg, 2.14 mmol) in THF (2 mL).
The reaction mixture was stirred at -75 °C for 10 min, the cooling
bath was removed, and the mixture was warmed to ambient
temperature. Upon warming, a white precipitate was formed, and the
viscosity increased. The reaction mixture was stirred at room
temperature for 3 h. After the reaction had been completed (TLC),
THF was removed in vacuo below 40 °C, the residue was quenched
with Et2O (40 mL) and H2O (30 mL), and the layers were separated.
The water phase was washed with ether (20 mL), and the combined
organic extracts were washed with water (30 mL) and brine (20
mL), dried (Na2SO4), and evaporated to give an orange oil which
waspurifiedbycolumnchromatography(50mLSiO2,EtOAc-hexane
1:7). Yield: 0.34 g (88.7%), orange solid. An analytical sample
and a sample for carbene synthesis were prepared by crystallization
N,N-Diisopropyl-2-(4-nitrophenyl)acetamide. This compound
was prepared from iPr2NH and (4-nitrophenyl)acetic acid chloride
followed by purification of the product by column chromatography
(EtOAc-petroleum ether 1:2.5) and recrystallization from EtOAc-
hexane 1:5. The crude product (mp 70-71 °C) was used further
1
without additional purification. H NMR (250 MHz, CD2Cl2, 298
1
3
3
of the product from EtOAc-hexane 1:7 (mp 50 °C). H NMR
K): δ 8.14 (d, JHH ) 8.8 Hz, 2H, arom), 7.39 (d, JHH ) 8.9 Hz,
(500.13 MHz, CDCl3, 298 K): δ 7.37 (d, 3JHH ) 8.55 Hz, 2H, arom),
3
2H, arom), 3.91 (sep, JHH ) 6.6 Hz, 1H, NCH(CH3)2), 3.75 (s,
3
3
6.93 (d, JHH ) 8.55 Hz, 2H, arom), 3.84 (s, 3H, OCH3), 3.80 (s,
2H, CH2), 3.43 (br. s, 1H, NCH(CH3)3), 1.37 (d, JHH ) 6.6 Hz,
3
3H, OCH3). Anal. Calcd for C10H10N2O3: C, 58.25; H, 4.89; N,
13.59. Found: C, 58.17; H, 4.84; N, 13.63.
6H, NCH(CH3)3), 1.09 (d, JHH ) 6.6 Hz, 6H, NCH(CH3)3).
2-Diazo-N,N-diisopropyl-2-(4-nitrophenyl)acetamide (7d). To
a stirred solution of N,N-diisopropyl-2-(4-nitrophenyl)acetamide (0.5
g, 1.89 mmol) and ABSA (0.5 g, 2.083 mmol) in absolute THF (8
mL) was added dropwise DBU (0.31 mL, 2.083 mmol) at 0 °C.
The cooling bath was removed, and the reaction mixture was stirred
at room temperature for ca. 1.5 h (TLC control). An additional 80
mg of AzN3 and 0.07 mL of DBU were added, and the dark solution
was stirred for a further 30 min. The solvent was removed on a
rotary evaporator, and the residue was purified by column chro-
matography on deactivated SiO2 (EtOAc-hexane 1:2.5) to give
an orange oil which quickly crystallized on standing. The product
was recrystallized from warm EtOAc-hexane 1:5. The mother
liquor after crystallization was concentrated by evaporation followed
by crystallization of more product from light petroleum. Yield: 50%.
Method B.39 To a well-stirred mixture of methyl (4-methox-
yphenyl)acetate (0.5 g, 2.78 mmol), Bu4NBr (250 mg, 0.778 mmol),
TryN3 (1.03 g, 3.336 mmol), 18-crown-6 (30 mg), and C6H6 (10
mL) was added dropwise a 66% aqueous solution of KOH (15 mL).
The reaction mixture was stirred at 39 °C for 25 min, an additional
100 mg (0.4 equiv) of TryN3 was added, and the mixture was stirred
at 38 °C for another 15 min. Ether was added, and the layers were
separated. The organic phase was washed with water and brine,
dried (Na2SO4), and evaporated. Yield of the product after chro-
matography (EtOAc-hexane 1:7): 86 mg (15%). Spectroscopic data
were identical with those of a sample prepared by method A.
Bis(4-chlorophenyl)methyl (4-Methoxyphenyl)acetate. This
compound was prepared from bis(4-chlorophenyl)methanol (3.06
g, 12.10 mmol), pyridine (2.57 mL, 31.85 mmol), and (4-
methoxyphenyl)acetyl chloride (2 mL, 12.74 mmol) in toluene (150
mL). Extractive workup followed by recrystallization from
EtOAc-petroleum ether 1:5, column chromatography (SiO2,
EtOAc-petroleum ether 1:5), and one more recrystallization from
the same solvent mixture afforded 2.68 g (54.5%) of analytically
1
Rf ) 0.95 (EtOAc-hexane 1:2.5). Mp: 87-88 °C. H NMR (250
3
MHz, CDCl3, 298 K): δ 8.18 (d, JHH ) 9.2 Hz, 2H, arom), 7.28
3
3
(d, JHH ) 9.2 Hz, 2H, arom), 3.74 (sep, JHH ) 6.7 Hz, 2H,
3
NCH(CH3)2), 1.35 (d, JHH ) 6.7 Hz, 12H, NCH(CH3)3). Anal.
Calcd for C14H18N4O3: C, 57.92; H, 6.25; N, 19.30. Found: C, 57.77;
H, 6.20; N, 19.18.
1
pure material. Mp: 95-96 °C. H NMR (250 MHz, CDCl3, 293
N,N-Dimethyl-2-(4-nitrophenyl)acetamide. (4-Nitrophenyl)ace-
tic acid chloride (3 g, 15.04 mmol) was dissolved in dry C6H6 (100
mL), and the solution was cooled to 0 °C. Me2NH gas was bubbled
through the solution, and a precipitate was formed. The reaction
mixture was poured on water, and the organic layer was separated,
washed with water, dried (Na2SO4), and evaporated to give a solid
which was recrystallized from EtOAc-hexane 1:5. Yield: 2.58 g
3
3
K): δ 7.31 (d, JHH ) 8.6 Hz, 4H, arom), 7.21 (d, JHH ) 8.8 Hz,
2H arom), 7.19 (d, 3JHH ) 8.6 Hz, 4H, arom), 6.89 (d, 3JHH ) 8.8
Hz, 2H, arom), 6.80 (s, 1H, OCH), 3.84 (s, 3H, OCH3), 3.69 (s,
2H, ArCH2). Anal. Calcd for C22H18Cl2O3: C, 65.85; H, 4.52; Cl,
17.67. Found: C, 65.64; H, 4.54; Cl, 17.53.
Bis(4-chlorophenyl)methyl Diazo(4-methoxyphenyl)acetate
(7g). Bis(4-chlorophenyl)methyl (4-methoxyphenyl)acetate (1 g,
2.46 mmol) and ABSA (0.89 g, 3.69 mmol) were suspended in
absolute THF, and then DBU (0.55 mL, 0.56 g, 3.69 mmol) was
added in one portion. The reaction mixture was stirred for 1 h at
room temperatue and then allowed to stand at ambient temperature
in a dark place for 7 days. TLC (EtOAc-hexane 1:7) showed
complete conversion of the starting material. The solvent was
removed in vacuo, and the residue was purified by column
chromatography (SiO2, EtOAc-hexane 1:5) to give a red-orange
oil that was dried at 10-3 mbar. Yield: 1.05 g of an orange oil,
purity >90%. The product is light sensitive and slowly decomposes
at room temperature; therefore, it must be stored under Ar in a
dark, cold (-35 °C) place. 1H NMR (250 MHz, CDCl3, 298 K): δ
7.27-7.41 (m, 10H, arom), 7.00 (s, OCH), 6.97 (d, 3JHH ) 8.9 Hz,
2H, arom), 3.84 (s, 3H, OCH3). MS-EI: m/z 398.0 (2%) [(M -
N2)+].
1
(82.4%). Mp: 89-90 °C. H NMR (250 MHz, CDCl3, 298 K): δ
3
3
8.15 (d, JHH ) 8.8 Hz, 2H, arom), 7.40 (d, JHH ) 8.8 Hz, 2H,
arom), 3.78 (s, 2H, CH2), 3.03 (s, 3H, NCH3), 2.96 (s, 3H, NCH3).
Anal. Calcd for C10H12N2O3: C, 57.69; H, 5.81; N, 13.45. Found:
C, 57.59; H, 5.78, 13.40.
2-Diazo-N,N-dimethyl-2-(4-nitrophenyl)acetamide (7c). This
compound was prepared in a manner similar to that described for
7d, from N,N-dimethyl-2-(4-nitrophenyl)acetamide and ABSA in
THF. Extractive workup followed by column chromatography
(SiO2, EtOAc-hexane 2:1) gave a yellow-orange product. Crystal-
lization from EtOAc-hexane 1:5 afforded red-orange plates. Yield:
1
45%. Mp: 118-119 °C. H NMR (500 MHz, toluene-d8, 298 K):
3
3
δ 7.91 (d, JHH ) 8.9 Hz, 2H, arom), 6.98 (d, JHH ) 8.9 Hz, 2H,
arom), 2.48 (s, 6H, N(CH3)2). 13C NMR (125.77 MHz, toluene-d8,
298 K): δ 162.9 (s, CdO), 145.2, 135.7, 124.2, 123.4 (all s, arom
C), 62.6 (s, C ) N2), 36.8 (s, N(CH3)2). Anal. Calcd for C10H10N4O3:
C, 51.28; H, 4.30; N, 23.92. Found: C, 51.26; H, 4.29; N, 23.92.
Methyl Diazo(4-methoxyphenyl)acetate (7e). Method A. To
a stirred solution of LDA (1.1 equiv, 2.05 mmol) in dry THF (7
mL) was added dropwise methyl(4-methoxyphenyl)acetate (0.3 mL,
0.335 g, 1.86 mmol) via a syringe at -75 °C. The reaction mixture
was stirred at -75 °C for 1 h followed by dropwise addition of a
[tBu2P(NSiMe3)2-K2N]CudC[(p-NO2-C6H4)(COOEt)] (8a). Cop-
per complex 1 (40 mg, 0.097 mmol) and diazo compound 7a (0.85
eq, 20 mg, 0.083 mmol) were mixed in 0.5 mL of toluene-d8 and
placed in an NMR tube after complete dissolution. The solution in
the NMR tube was heated to 70 °C for 5 min, and the steady-state
concentration of carbene complex 8a reached 35%. 1H NMR

1058 Organometallics, Vol. 28, No. 4, 2009
Hofmann et al.
(500.13 MHz, toluene-d8, 235 K): δ 8.62 (d, 3JHH ) 8.5 Hz, 2 arom
H), 7.32 (d, 3JHH ) 8.5 Hz, 2 arom H), 4.20 (q, 3JHH ) 7.2 Hz, 2H,
CH2), 1.37 (d, 3JHP ) 14.3 Hz, 9H, C(CH3)3), 1.30 (d, 3JHP ) 14.6
Hz, 9H, C(CH3)3), 0.33 (s, 18H, Si(CH3)3). The signal of the CH3
group overlaps with tBu group signals of the carbene complex and
of the starting ethylene complex. 31P{1H} NMR (202.47 MHz,
toluene-d8, and the resulting dark violet solution was placed in an
NMR tube. After 10 min of shaking at room temperature, the NMR
measurement was performed at 235 K. Carbene 8f was formed in
a steady-state concentration of 59.2%. 1H NMR (500.13 MHz, 235
K, toluene-d8): δ 9.47 (br s, 1H, o-C6H4), 7.85 (br s, 1H, o-C6H4),
6.46 (br s, 1H, m-C6H4), 6.17 (br s, 1H, m-C6H4), 2.98 (s, 3H,
OCH3), 1.81 (s, 9H, OC(CH3)3), 1.47 (d, 3JHP ) 13.8, 9H, C(CH3)3),
2
1
toluene-d8, 235 K): δ 67.7 (s + sat., JPSi ) 11.2 Hz, JPC ) 60.8
Hz). 13C{1H} NMR (125.77 MHz, toluene-d8, 235 K): δ 219.0
3
1.45 (d, JHP ) 14.0 Hz, 9H, C(CH3)3), 0.48 (s, 18H, Si(CH3)3).
1
31P{1H} NMR (202.47 MHz, 235 K, toluene-d8): δ 59.7 (s + sat,
(CudC), 178.3 (CdO), 37.7 (d, JPC ) 61.7 Hz, C(CH3)3), 37.6
(d, 1JPC ) 61.7 Hz, C(CH3)3), 27.7 (s, C(CH3)3), 27.6 (s, C(CH3)3),
5.5 (s, Si(CH3)3); aromatic signals omitted.
1
2JPSi ) 14.4 Hz, JPC ) 64.1 Hz). 13C{1H} NMR (125.77 MHz,
235 K, toluene-d8): δ 235.8 (CudC), 177.0 (C dO), 55.6 (s,
1
1
[tBu2P(NSiMe3)2-K2N]CudC[(p-NO2-C6H4)(COOtBu)] (8b).
The copper complex 1 (40 mg, 0.097 mmol) and the diazo
compound 7b (0.9 equiv, 23 mg, 0.0876 mmol) were mixed in 0.5
mL of toluene-d8, and when dissolution was complete, the solution
was transferred to an NMR tube. The color slowly changed to dark
violet, but even after 40 min at room temperature the carbene
concentration remained very low. The solution in the NMR tube
was heated to 70 °C for 10 min, and the steady-state concentration
of the carbene complex 8b reached 60.2%. 1H NMR (500.13 MHz,
OC(CH3)3), 37.5 (d, JPC ) 64.4 Hz, C(CH3)3), 37.4 (d, JPC
)
64.4 Hz, C(CH3)3), 29.1 (s, OC(CH3)3), 28.1 (s, C(CH3)3), 28.0 (s,
C(CH3)3), 5.75 (s, Si(CH3)3). 1H-VT-NMR (500.13 MHz, toluene-
d8,): o-H, Tc ) 270 ( 6 K, kc ) 1807 Hz; m-H, Tc ) 254 ( 4 K,
kc ) 305 Hz, tBu, Tc ) 314.1 ( 4 K, kc ) 23.5 Hz.
[tBu2P(NSiMe3)2-K2N]CudC{(p-MeO-C6H4)[COOCH(p-
ClC6H4)2]} (8g). Copper complex 1 (100 mg, 0.243 mmol) and
diazo compound 7g (100 mg, 0.231 mmol) were stirred for 20 min
in absolute hexane (14 mL) under a slight dynamic vacuum at
36-37 °C. The solvent was carefully removed at -3 °C and 10-3
mbar, and the dark residue was treated with absolute hexane (8
mL), vigorously stirred at room temperature for 2 min, and filtered
via a filter cannula into another Schlenk vessel cooled to -3 °C.
The clear, deep violet solution was slowly cooled to -60 °C and
kept at this temperature for 2 days. Dark crystals precipitated, and
the supernatant solvent was removed via cannula. The remaining
solid was washed with cold (-78 °C) absolute pentane (8 mL) and
dried under high vacuum (10-3 mbar) at -20 °C and finally for 2
min at room temperature to yield 126 mg of dark crystalline
material. According to elemental analysis, it was contaminated with
the starting diazo compound 7g and cocrystallized hexane (both
ca. 17 mol %). Recrystallization from absolute pentane gave a
sample completely free of the starting diazo ester (8g · 0.46C5H12,
100 mg, 53%). According to the 1H NMR and elemental analysis,
the product crystallized with 0.46 equiv of pentane. 1H NMR
3
toluene-d8, 235 K): δ 8.59 (d, JHH ) 8.7 Hz, 2 arom H), 7.25 (d,
3JHH ) 8.7 Hz, 2 arom H), 1.74 (s, 9H, OC(CH3)3), 1.36 (d, 3JHP
)
3
14.1 Hz, 9H, C(CH3)3), 1.34 (d, JHP ) 13.9 Hz, 9H, C(CH3)3),
0.33 (s, 18H, Si(CH3)3). 31P{1H} NMR (202.47 MHz, toluene-d8,
235 K): δ 66.7 (s + sat., 2JPSi ) 11.7 Hz, 1JPC ) 62.1 Hz). 13C{1H}
NMR (125.77 MHz, toluene-d8, 235 K): δ 221.1 (CudC), 177.9
(CdO), 81.7 (s, OC(CH3)3), 37.62 (d, JPC ) 61.9 Hz, C(CH3)3),
37.59 (d, JPC ) 61.9 Hz, C(CH3)3), 28.4 (s, OC(CH3)3), 27.7 (s,
C(CH3)3), 27.6 (s, C(CH3)3), 5.53 (s, Si(CH3)3); signals correspond-
ing to the aromatic ring omitted.
[tBu2P(NSiMe3)2-K2N]CudC[(p-NO2-C6H4)(CONMe2)] (8c).
To a mixture of copper complex 1 (40 mg, 0.0973 mmol) and diazo
compound 7c (20.5 mg, 0.0876 mmol) was added toluene-d8, and
after the reaction mixture was shaken at ambient temperature for 5
min, an intense violet solution was obtained. This solution was
1
placed in an NMR tube. H NMR (500.13 MHz, toluene-d8, 263
3
3
3
K): δ 8.85 (d, JHH ) 8.8 Hz, 2 arom H), 7.45 (d, JHH ) 8.8 Hz,
(500.13 MHz, THF-d8, 238 K): δ 9.30 (d, JHH ) 8.2 Hz, 1 arom
3
3
2 arom H), 2.78 (s, 3H, N(CH3)2), 2.40 (s, 3H, N(CH3)2), 1.44 (d,
H), 7.50 (d, JHH ) 8.5 Hz, 4 arom H), 7.47 (d, JHH ) 8.2 Hz, 1
3JHP ) 14.4 Hz, 9H, C(CH3)3), 1.43 (d, JHP ) 14.4 Hz, 9H,
3
arom H), 7.39 (d, 3JHH ) 8.5 Hz, 4 arom H), 7.21 (s, CH), 7.16 (d,
C(CH3)3), 0.29 (s, 18H, Si(CH3)3). 31P{1H} NMR (202.47 MHz,
3JHH ) 7.0 Hz, 1 arom H), 6.81 (d, JHH ) 7.60 Hz, 1 arom H),
3
1
2
3
toluene-d8, 263 K): δ 67.0 (s + sat, JPC ) 62 Hz, JPSi ) 11.8
Hz). 13C{1H} NMR (125.77 MHz, toluene-d8, 263 K): δ 231.7 (s,
CudC), 177.8 (s, CdO), 155.1, 147.7, 132.4, 126.1 (all s, arom
3.88 (s, OCH3), 1.36 (d, JHP ) 14.2 Hz, 9H, C(CH3)3), 1.28 (d,
3JHP ) 14.5 Hz, 9H, C(CH3)3), -0.05 (s, Si(CH3)3). 31P{1H} NMR
1
(202.47 MHz, THF-d8, 238 K): δ 59.5 (s + sat, JPC ) 63.9 Hz,
1
1
2JPSi ) 14.3 Hz). 13C{1H} NMR (125.77 MHz, THF-d8, 238 K): δ
230.5 (s, CudC), 175.9 (s, CdO), 166.9, 143.9, 143.7, 140.3, 134.1,
134.0, 129.7, 129.3, 117.9, 117.1 (all arom H), 75.0 (s, CH), 56.6
C), 37.4 (d, JPC ) 62.4 Hz, C(CH3)3), 37.2 (d, JPC ) 62.4 Hz,
C(CH3)3), 35.1 (s, NCH3), 33.0 (s, NCH3), 27.7 (d, 2JCP ) 2.2 Hz,
2
3
C(CH3)3), 27.3 (d, JCP ) 2.2 Hz, C(CH3)3), 5.26 (d, JCP ) 2.2
2
2
Hz, Si(CH3)3).
(s, OCH3), 37.43 (d, JCP ) 64.4 Hz, C(CH3)3), 37.39 (d, JCP
64.4 Hz, C(CH3)3), 27.8 (s, C(CH3)3), 27.7 (s, C(CH3)3), 5.1 (s,
Si(CH3)3). H-VT-NMR (500.13 MHz, THF-d8,): o-H, Tc ) 277
)
[tBu2P(NSiMe3)2-K2N]CudC[(p-MeO-C6H4)(COOMe)] (8e).
Copper complex 1 (40 mg, 0.097 mmol) and diazo compound 7e
(18 mg, 0.087 mmol) were dissolved in toluene-d8 (0.5 mL). Gas
evolution was observed, and the reaction mixture became intensively
violet. The reaction mixture was agitated at room temperature for
1 min, transferred to an NMR tube, and kept at room temperature
for a further 4 min. 1H NMR (500.13 MHz, toluene-d8, 237 K): δ
1
( 5 K, kc ) 917.5 Hz; m-H, Tc ) 257 ( 4 K, kc ) 174 Hz. MS-
FAB (NPOE): 782.5 (20%), [MH+], 725.4 (14%), [MH+ - tBu],
correct isotopic pattern. Anal. Calcd for C36H52Cl2CuN2O3PSi2 ·
0.46C5H12: C, 56.35; H, 7.05; N, 3.43; Cl, 8.71; P, 3.80. Found: C,
56.47; H, 6.89; N, 3.72; Cl, 8.73; P, 3.81. UV-vis (in THF): λmax
527 nm (ꢀ ) (1.1-1.7) × 104).
3
9.61 (br s, 1H arom (ortho)), 6.79 (d, JHH ) 7.35 Hz, 2H, arom),
[tBu2P(NSiMe3)2-K2N]Cu(C6D6). The free iminophosphanamide
ligand (0.32 g, 1 mmol) was dissolved in hexane (5 mL), treated
with 1.6 M of n-BuLi (0.75 mL, 1.2 mmol), and then added to a
cooled (-70 °C) suspension of CuBr · Me2S (0.246 g, 1.2 mmol)
in hexane (6 mL). After it was stirred at room temperature for 3 h,
the reaction mixture was allowed to stand for 2 days, and the liquid
above the black solid precipitate was decanted with a filter cannula.
The clear solution obtained was evaporated at 10-3 mbar to give
ca. 0.20 g of an oil, which was analyzed by NMR. 1H NMR (C6D6,
500.13 MHz): 1.43 (d + sat, 3JHP ) 14 Hz, 1JCH ) 126.5 Hz, 18H,
tBu), 0.65 (s + sat, 1JCH ) 117.8 Hz, 18H, SiMe3). 31P{1H} NMR
(C6D6, 202.47 MHz): δ 63.2 (s).
6.52 (br s., 1H arom (ortho)), 3.73 (s, 3H, OCH3), 2.96 (s, 3H,
3
3
OCH3), 1.48 (d, JHP ) 13.9 Hz, 9H, C(CH3)3), 1.39 (d, JHP
)
14.3 Hz, 9H, C(CH3)3), 0.47 (s, 18H, Si(CH3)3). 31P{1H} NMR
(202.48 MHz, toluene-d8, 237 K): δ 60.74 (s). 13C{1H} NMR
(125.77 MHz, toluene-d8, 237 K): δ 232.8 (s, CudC), 177.9 (s,
CdO), 37.5 (d, 2JCP ) 63.7 Hz, C(CH3)3), 37.4 (d, 2JCP ) 63.7 Hz,
C(CH3)3), 28.2 (s, C(CH3)3), 28.0 (s, C(CH3)3), 5.65 (s, Si(CH3)3).
1H-VT-NMR (500 MHz, toluene-d8): tBu, Tc ) 301 ( 4 K, kc )
101.4 Hz.
[tBu2P(NSiMe3)2-K2N]CudC[(p-MeO-C6H4)(COOtBu)] (8f).
Copper complex 1(40 mg, 0.097 mmol) and diazo compound 7f
(0.9 equiv, 22 mg, 0.088 mmol) were dissolved in 0.5 mL of

Copper(I) R-Carbonyl Carbenes
Organometallics, Vol. 28, No. 4, 2009 1059
Table 8. Crystal Data and Structure Refinement Details for 8g
based on the Laue symmetry of the reciprocal space. The structure
was solved by direct methods and refined against F2 with a full-
matrix least-squares algorithm using the SHELXTL-PLUS (5.10)
software package. A total of 924 parameters were refined, hydrogen
atoms were treated using appropriate riding models, and disordered
solvent electron density was assigned using several carbon atoms
with different occupancies. Crystal data and structure refinement
details for 8g are summarized in Table 8.
empirical formula
formula wt
temp (K)
wavelength (Å)
cryst syst
space group
Z
C27H44CuN4O4PSi2
782.39
200(2)
0.710 73
triclinic
j
P1
4
a (Å)
b (Å)
c (Å)
12.5036(2)
18.8771(3)
20.9013(3)
91.8980(10)
94.7360(10)
92.4700(10)
4908.62(13)
1.059
Acknowledgment. Support of this work (part of the Ph.D.
Thesis of I.V.S., Universita¨t Heidelberg, 2006) by the
Deutsche Forschungsgemeinschaft (SFB 623 “Molecular
Catalysts: Structure and Functional Design”) and by the
Fonds der Chemischen Industrie is gratefully acknowledged.
R (deg)
ꢀ (deg)
γ (deg)
V (Å3)
Dcalcd (g/cm3)
abs coeff (mm-1
cryst shape
)
0.66
polyhedron
0.38 × 0.20 × 0.14
1.0-24.7
Supporting Information Available: An ORTEP diagram of 8g
with a complete list of bond lengths and angles, ORTEP plots of
diazo compounds 7b,d,f and ethyl diazo(2-nitrophenyl)acetate,
computational details and Cartesian coordinates of the optimized
cryst size (mm3)
θ range for data collecn (deg)
no. of collected rflns
41 715
no. of indep rflns
16 670
structure of 8e, 1H, 31P, and 13C NMR spectra of compound 8g, 31
P
no. of obsd rflns
10 070
1.00
goodness of fit on F2
NMR spectra of 6 and 8g after complete decomposition, and time
vs concentration and Eyring plots for carbenes 6 and 8. This material
is available free of charge via the Internet at http://pubs.acs.org.
final R indices (I > 2σ(I))
R1 ) 0.053, wR2 ) 0.138
0.65 and -0.37
largest diff peak and hole (e Å-3
)
Crystal Structure Analysis of 8g. Crystals of 8g suitable for
X-ray analysis were obtained by crystallization from a concentrated
solution of crude 8g in pentane at 0 °C. Data were collected on a
Bruker SMART diffractometer, and the structure solution was
performed using the SHELXTL-PLUS (5.10) software package.41
Intensities were corrected for Lorentz and polarization effects, and
an empirical absorption correction was applied using SADABS42
OM8007376
(41) Sheldrick, G. M. SHELXTL-PLUS, Bruker Analytical X-ray
Division, Madison, WI, 1997.
(42) Sheldrick, G. M. SADABS, Bruker Analytical X-ray Division,
Madison, WI, 2001.