Complexes of Gilman reagents with C-S and C-N double bonds: δ or π bonding?

Article abstract of DOI:10.1021/ja103068h

Upon rapid injection, a variety of thiocarbonyl compounds react with the Gilman reagent Me₂CuLi at -100 °C inside the probe of an NMR spectrometer to give high yields of complexes. Typical examples of substrates include carbon disulfide, methyl dithioacetate, methyl dithiobenzoate, thiobenzophenone, ethylene trithiocarbonate, and phenyl isothiocyanate. Evidence suggesting the formal oxidation state of copper in these complexes to be Cu^III is presented. The last example was particularly interesting, since it involved a transient intermediate that was identified as a complex with a C-N double bond. Methyl isothiocyanate gave a stable C-N double-bond complex.

Full text of DOI:10.1021/ja103068h

Published on Web 06/11/2010
Complexes of Gilman Reagents with C-S and C-N Double Bonds:
σ or π Bonding?
Steven H. Bertz,* Yasamin Moazami, Michael D. Murphy, Craig A. Ogle,* Joshua D. Richter, and
Andy A. Thomas
Department of Chemistry, UniVersity of North Carolina-Charlotte, Charlotte, North Carolina 28223
Received April 12, 2010; E-mail: sbertz1@uncc.edu; cogle@uncc.edu
Scheme 2. Structures of Substrates 1, Complexes 3, and
Carbophilic Addition Products 4 (2 ) Me2CuLi; n/d ) not detected)
Abstract: Upon rapid injection, a variety of thiocarbonyl compounds
react with the Gilman reagent Me2CuLi at -100 °C inside the probe
of an NMR spectrometer to give high yields of complexes. Typical
examples of substrates include carbon disulfide, methyl dithio-
acetate, methyl dithiobenzoate, thiobenzophenone, ethylene trithio-
carbonate, and phenyl isothiocyanate. Evidence suggesting the
formal oxidation state of copper in these complexes to be CuIII is
presented. The last example was particularly interesting, since it
involved a transient intermediate that was identified as a complex
with a C-N double bond. Methyl isothiocyanate gave a stable C-N
double-bond complex.
In 1941, Kharasch and Tawney discovered that catalytic amounts
of copper salts change the regioselectivity of the reaction between
Grignard reagents and R-enones from 1,2- to 1,4-addition.1 In 1966,
House, Respess, and Whitesides demonstrated that stoichiometric
organocuprate(I) reagents (Gilman reagents) also give 1,4-addition to
R-enones.2 In 1985, it was shown that CuI reagents, catalytic or
stoichiometric, change the regioselectivity of addition to thiocarbonyl
compounds from thiophilic to carbophilic.3 Is this analogy between
R,ꢀ-unsaturated carbonyl compounds and thiocarbonyl compounds a
superficial one, or is it a reflection of a deeper connection?
We investigated the reactions of a variety of substrates 1 containing
C-S double bonds with the Gilman reagent Me2CuLi (2) using rapid-
injection NMR (RI-NMR) spectroscopy,4,5 and we can now report
that the analogy between the thiocarbonyl group and the R-enone
system is a deep one, as both functional groups form complexes with
Gilman reagents on the reaction pathway, as illustrated in Scheme 1.
Upon injection of carbon disulfide (1a; 32 µmol) into a solution of
2·LiCl (30 µmol) in THF-d8 that was spinning in the probe of an
NMR spectrometer at -100 °C, a new species rapidly formed. In the
13C NMR spectrum, peaks for 1a (193.90 ppm) and 2·LiCl (-9.07
ppm) disappeared, and new peaks at 279.19, 20.10, and -10.54 ppm
1
and new peaks at 0.37 and -0.76 ppm appeared. The HMBC spectrum
had cross-peaks between the methyl groups and also between the
methyls and the thiocarbonyl carbon. Complex 3a formed equally well
with 2·LiX (X ) Cl, Br, I, CN).
appeared. The H NMR peak for 2·LiCl (-1.48 ppm) disappeared,
Scheme 1. Analogous Reactions of Thiocarbonyl Compounds and
R,ꢀ-Unsaturated Carbonyl Compounds with Me2CuLi (2)
The NMR data are consistent with the pseudo-square-planar
structure 3a (Scheme 2), which was confirmed by two-bond 13C-13C
coupling across Cu. In all previous examples, trans coupling was
significantly larger in magnitude than cis.6 Using 13CS2, we were able
to measure 2Jtrans ) 28.8 Hz for coupling to the upfield methyl (MeR)
and 2Jcis ) 4.5 Hz for coupling to the downfield methyl (Meꢀ). When
13CS2 was added to 3a at -100 °C, no incorporation of the labeled
ligand was observed.
When injected into solutions of 2·LiI in THF-d8 at -100 °C, methyl
dithioacetate (1b) and methyl dithiobenzoates 1c-e formed 3b-e,
respectively. In all cases, the Gilman reagent was completely converted
to complex with a small excess of substrate, although warming to -90
9
10.1021/ja103068h  2010 American Chemical Society
J. AM. CHEM. SOC. 2010, 132, 9549–9551 9549

C O M M U N I C A T I O N S
Table 1. 1H and 13C NMR Shifts for Methyl Groups MeR and Meꢀ
Phenyl isothiocyanate (1j) reacted rapidly with 2·LiI at -100 °C
to afford a fleeting intermediate 3j′ (t1/2 ) 40 s) and a final, stable
complex 3j. The methyl 13C shifts for 3j were very similar to those
for 3a-h, and therefore, we assign a structure that also has a
complexed C-S double bond. The methyl 13C shifts (obtained by using
13C-labeled 2·LiI) for intermediate 3j′ were far outside the range for
3a-h. The difference, ∆(13C) ) 4.91 ppm, is close to the ∆(13C) value
of 4.44 ppm for C-C double-bond complex 6. It is interesting to note
that the difference ∆(1H) ) 0.44 ppm for 3j′ is in the range ∆(1H) )
0.22-0.45 ppm for complexes 3c-f, which have at least one phenyl
group attached to a C bonded to Cu.
Are the double-bond complexes discussed above best described as
trigonal d10 CuI π complexes, as shown in Scheme 1, or are they
pseudo-square-planar d8 CuIII σ bonded structures, as shown in Scheme
2? Calculations support the view that the charges on the ligands in
CuIII complexes are significantly smaller than they are in CuI
complexes.9 This difference is reflected in the 13C NMR shifts of
lithium dimethylcuprate(I) (ca. -10 ppm) and lithium tetramethyl-
cuprate(III) (ca. 15 ppm).5 Similarly, the 13C NMR data summarized
in Table 1 show a clear difference between the complexes involving
C-C and C-N double bonds on one hand and those involving C-S
double bonds on the other.
1H shifts (ppm)a
13C shifts (ppm)b
complex (bond)
MeR
Meꢀ
∆(1H)
MeR
Meꢀ
∆(13C)
3a (CdS)
3b (CdS)
3c (CdS)
3d (CdS)
3e (CdS)
3f (CdS)
3g (CdS)
3h (CdS)
3i′ (CdN)
3j′ (CdN)
3j (CdS)
6 (CdC)
-0.76
-0.96
-0.68
-0.74
-0.52
-0.55
-0.76
-1.00
-0.48
-0.47
-0.79
-1.12
0.37
-0.16
-0.29
-0.29
-0.27
-0.33
0.12
-0.03
0.52
-0.03
0.43
-0.10
1.13
0.80
0.39
0.45
0.25
0.22
0.88
0.97
1.00
0.44
1.22
1.02
-10.54
-8.58
-9.31
-9.46
-9.18
-9.02
-9.51
-9.34
-1.48
-2.76
-9.09
-5.04
20.10
7.17
9.35
9.19
9.18
9.13
8.84
4.06
-1.25
2.15
10.81
-0.60
30.64
15.75
18.66
18.65
18.36
18.15
18.35
13.40
0.23
4.91
19.90
4.44
b
a ∆(1H) ) δH(Meꢀ) - δH(MeR). ∆(13C) ) δC(Meꢀ) - δC(MeR).
°C was necessary to complete the reaction with 1b in a reasonable
time (<1 h). The colors of the solutions of these complexes were shades
of yellow to red.
There were two salient features in the NMR spectra of these complexes:
(i) a pair of 1H peaks and the corresponding 13C peaks (Table 1) for the
methyl groups MeR (upfield) and Meꢀ (downfield) attached to Cu and (ii)
a characteristic 13C peak at ca. 80 ppm for the C atom attached to Cu and
S (Scheme 2). For the methyl groups on Cu, the upfield 13C peak
The difference in charge is also responsible for the dramatic
difference in the reactivity of these complexes toward methanol (4
equiv), as observed using RI-NMR spectroscopy. At -100 °C, lithium
dimethylcuprate(I) reacted completely before the first spectrum was
obtained (<1 s). In contrast, lithium tetramethylcuprate(III) reacted very
slowly at -100 °C but gave a quantitative yield of methane upon
warming to -60 °C.10
1
corresponded to the upfield H peak (HMQC/HSQC) and likewise for
the downfield peaks in these and the rest of the complexes reported herein.
NOESY cross-peaks between the hydrogens of Meꢀ and those of the
substituents were particularly prominent.
At -100 °C, trifluoromethyl derivative 1e completely displaced 1c
from complex 3c to give complex 3e, but it did not react with methoxy
complex 3d. At -60 °C, 1e converted complex 3d to 3e (plus free
1d). Likewise, 1c converted complex 3d to 3c (plus free 1d) at -60
°C. Consequently, the order of stability is 3e > 3c > 3d, i.e., electron-
poor ligands form more stable complexes than electron-rich ones. This
observation suggests a “push-pull” interaction between the methyl
and thiocarbonyl ligands.
Injection of a deep-blue solution of thiobenzophenone (1f) into a
colorless solution of 2·LiCl (both in THF-d8) at -100 °C gave an
emerald-green solution of 3f. The 13C NMR shift of the C bonded to
Cu and S was 81.89 ppm. A number of transition-metal complexes of
thiobenzophenone are known,7 and the 13C shifts for the CdS groups
are in the range 150-180 ppm. We attribute the large upfield
displacement in our case to significantly higher sp3 character. (For
comparison, it should be noted that the 13C shift of the methine in
diphenylmethanethiol is 48 ppm.8)
A mixture of R-enone complexes 6 and 6·LiI4 reacted rapidly at
-100 °C. In fact, the rate of methanolysis was much higher than the
rate of dissociation of the complexes (k-1; see Scheme 1).4 Complex
3i′ reacted slowly at -100 °C. Complex 3c gave no reaction at -100
°C; however, methane was generated upon warming to -60 °C. The
results, C-C > C-N . C-S, are indicative of their positions on the
π-σ continuum, with C-C double bonds close to π and C-S double
bonds close to σ.
Facile ligand exchange is typical of square-planar d8 complexes,
and it usually proceeds via an associative mechanism, either pseudo-
rotation or an SN2-like reaction. For ligands such as ours with low-
lying LUMOs, we posit initial coordination of the new ligand by a
filled d orbital followed by the pseudorotation pathway. In contrast,
on the basis of theoretical calculations, Ga¨rtner et al.11 have proposed
an SN2 mechanism for the displacement of chloride by methyl in d8
CuIII complexes.
Ethylene trithiocarbonate (1g) reacted with 2·LiI at -100 °C to
afford complex 3g quantitatively. NOESY confirmed the methyl group
nearest the ethylene bridge to be Meꢀ.
η2-Complexes of Gilman reagents with C-C double bonds are well-
known, and we have now prepared and identified the first such
complexes of these valuable synthetic reagents with C-N and C-S
double bonds, which also have considerable potential.
Alkoxy derivative 1h reacted with 2·LiI at -100 °C to yield
complex 3h, which was significantly less stable than the previous ones
(t1/2 ≈ 1 h); nevertheless, it was fully characterized over several runs.
While oxygen substitution was tolerated, sp3 nitrogen was not:
methyl pyrrolidine-1-carbodithioate, N,N-dimethylthioformamide, and
N,N,N′,N′-tetramethylthiourea were unreactive toward 2·LiI.
In contrast, injection of methyl isothiocyanate (1i), which contains
an sp2 nitrogen, into 2·LiI (both in THF-d8 at -100 °C) gave a
relatively stable complex with methyl 13C shifts that were far removed
from those of 3a-h. The 13C peak at 207.85 ppm was 83 ppm
downfield from the corresponding peak in the substrate, essentially
the same effect that was observed upon complexation of carbon
disulfide (85 ppm). Consequently, we assign structure 3i′ to this new
product, which also has an uncomplexed C-S double bond. Finally,
a strong NOE between the NMe group and Meꢀ was observed, which
is consistent with the assigned structure.
Acknowledgment. We thank the NSF for Grants 0353061 and
0321056 and Tom DuBois for helpful discussions.
Note Added after ASAP Publication. Due to a production error,
the uncorrected proof version was published ASAP on June 11, 2010.
Typographical corrections have now been made and ref 9b has been
added. Corrections were also made to Scheme 2 and Table 1. The
corrected version was published on June 28, 2010.
Supporting Information Available: Typical NMR spectra. This
material is available free of charge via the Internet at http://pubs.acs.org.
References
(1) Kharasch, M. S.; Tawney, P. O. J. Am. Chem. Soc. 1941, 63, 2308.
(2) House, H. O.; Respess, W. L.; Whitesides, G. M. J. Org. Chem. 1966, 31,
3128.
9
9550 J. AM. CHEM. SOC. VOL. 132, NO. 28, 2010

C O M M U N I C A T I O N S
(3) Bertz, S. H.; Dabbagh, G.; Williams, L. M. J. Org. Chem. 1985, 50, 4415.
(4) Background: Bertz, S. H.; Carlin, C. M.; Deadwyler, D. A.; Murphy, M. D.;
Ogle, C. A.; Seagle, P. H. J. Am. Chem. Soc. 2002, 124, 13650.
(5) Procedure: Bartholomew, E. R.; Bertz, S. H.; Cope, S. K.; Murphy, M. D.;
Ogle, C. A.; Thomas, A. A. Chem. Commun. 2010, 46, 1253.
(6) (a) Bertz, S. H.; Cope, S.; Murphy, M.; Ogle, C. A.; Taylor, B. J. J. Am.
Chem. Soc. 2007, 129, 7208. (b) Bertz, S. H.; Cope, S.; Dorton, D.;
Murphy, M.; Ogle, C. A. Angew. Chem., Int. Ed. 2007, 46, 7082. (c)
Bartholomew, E. R.; Bertz, S. H.; Cope, S.; Dorton, D. C.; Murphy,
M.; Ogle, C. A. Chem. Commun. 2008, 1176. (d) Bartholomew, E. R.;
Bertz, S. H.; Cope, S.; Murphy, M.; Ogle, C. A. J. Am. Chem. Soc.
2008, 130, 11244.
(7) Beck, R.; Sun, H.; Li, X.; Camadanli, S.; Klein, H.-F. Eur. J. Inorg. Chem.
2008, 3253. and refs 7 and 8 cited therein.
(8) Makioka, Y.; Uebori, S.; Tsuno, M.; Taniguchi, Y.; Takaki, K.; Fujiwara,
Y. J. Org. Chem. 1996, 61, 372.
(9) (a) Hu, H.; Snyder, J. P. J. Am. Chem. Soc. 2007, 129, 7210. (b) Snyder,
J. P. J. Am. Chem. Soc. 1995, 117, 11025.
(10) Bertz, S. H.; Murphy, M. D.; Ogle, C. A.; Thomas, A. A. Chem. Commun.
2010, 46, 1255.
(11) Ga¨rtner, T.; Yoshikai, N.; Neumeier, M.; Nakamura, E.; Gschwind, R. M.
Chem. Commun. 2010, 46, 4625.
JA103068H
9
J. AM. CHEM. SOC. VOL. 132, NO. 28, 2010 9551