Organocuprate(III) chemistry: Synthesis and reactivity of amido, cyano, phosphido and thiolato ate complexes of copper(III)

Article abstract of DOI:10.1039/b921412g

Lithium tetramethylcuprate(iii) 1 reacts readily at -100 °C with appropriate sources of H⁺ or X⁺ (X = Br, I) to remove a methyl and in some cases incorporate the counterion (e.g., arylthio or cyano) to give stable complexes. These derivatives can in turn serve as starting materials for other Cu^III complexes.

Full text of DOI:10.1039/b921412g

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Organocuprate(III) chemistry: synthesis and reactivity of amido,
cyano, phosphido and thiolato ate complexes of copper(III)
Steven H. Bertz,*a Michael D. Murphy,b Craig A. Ogle*b and Andy A. Thomasb
Received (in College Park, MD, USA) 13th October 2009, Accepted 10th December 2009
First published as an Advance Article on the web 25th January 2010
DOI: 10.1039/b921412g
Lithium tetramethylcuprate(III) 1 reacts readily at ꢀ100 1C with
appropriate sources of H+ or X+ (X = Br, I) to remove a
methyl and in some cases incorporate the counterion (e.g.,
arylthio or cyano) to give stable complexes. These derivatives
can in turn serve as starting materials for other CuIII complexes.
The transient intermediacy of organocopper(III) species in
reactions of organocopper(I) reagents has been demonstrated
for conjugate addition, SN2 and SN20 reactions by using rapid
injection nuclear magnetic resonance (RI-NMR) spectro-
scopy.1–3 By running the reactions of EtI with Gilman reagents
Me2CuLiꢁLiX 2a–d (X = a, I; b, Br; c, Cl; d, CN) in the
presence of stabilizing ligands (e.g., DMAP, Bu3P, t-BuNC,
etc.), we were able to prepare and characterize a number of
neutral organocopper(III) complexes.4 This was the first family
of ‘‘engineered’’ CuIII complexes.
We recently reported the preparation of lithium tetra-
methylcuprate(III) 1 by treatment of 2,3-dichloropropene with
the mixture, Me2CuLiꢁLiX–2MeLi, prepared from CuX
Fig. 1 1H–31P HMBC plot for product 4. The 1H NMR peak at
(X = I, Br, Cl) and 4 equiv. of MeLi.5 We can now report
that by treating 1 with selected reagents capable of removing a
methyl group, a variety of CuIII ate complexes have been
prepared and characterized by NMR spectroscopy. Since most
of the intermediates observed to date in reactions of organo-
copper reagents have been CuIII ate complexes,1–3 understanding
the reactivity of this class of compounds is a high priority.
Injection of 60 mL of a 0.5 M solution of BrCN in freshly
distilled (Na/K) THF-d8 into a solution of 30 mmol of 1 in
ca. 0.5 mL of THF-d8 in an unused NMR tube, spinning in
the probe of an NMR spectrometer at ꢀ100 1C, gave a
virtually instantaneous and quantitative conversion to lithium
cyano(trimethyl)cuprate(III) 3 and methyl bromide. Complex 3
has been well-characterized by Gschwind and co-workers, in
spite of the fact that it was formed in low yield in their reaction
mixture.6 The conversion of 1 to 3, a known compound,
supports the structure assigned to 1 by spectroscopic studies.5
ꢀ0.18 ppm (ꢀ20 1C) is from 1, and the one at 0.20 ppm is from
methane.
to 1. The MeBr side-product from the preparation of 3 did not
react with the extra equivalent of MeLi.
When excess methanol was injected into the NMR tube, 1 was
slowly converted back to 3, presumably via capture of an inter-
mediate by LiCN, along with the co-production of methane. In a
control experiment, injection of methanol (2 equiv.) into a freshly
prepared solution of 1 at ꢀ100 1C resulted in a slow evolution of
methane, which accelerated appreciably at ꢀ60 1C. In stark
contrast, injection of methanol into a solution of the classic Gilman
reagent 2a at ꢀ100 1C gave an instantaneous burst of methane.
When 60 mL of 1 M lithium diphenylphosphide solution in
THF-d8 was injected into a solution of 3 and MeBr (from 1 and
BrCN, vide supra) at ꢀ100 1C, both reacted rapidly to afford a
mixture of Me3Cu(PPh2)Li 4 and Ph2PMe,w characterized by
1H, 13C and 31P NMR spectroscopy. HMBC (1H–13C
and 1H–31P) and HSQC (1H–13C) were particularly useful
(e.g., Fig. 1) for characterizing the products herein.
Injection of 2 equiv. of MeLi into the mixture of 3 and
MeBr at ꢀ100 1C resulted in the smooth conversion of 3 back
a Complexity Study Center, Mendham, NJ 07945, USA.
E-mail: sbertz@complexitystudycenter.org
b Department of Chemistry, University of North Carolina–Charlotte,
Charlotte, NC 28223, USA. E-mail: cogle@uncc.edu;
Fax: +1 704-687-3151; Tel: +1 704-687-2524
At ꢀ100 1C, the P-atom in 4 (23.81 ppm, with the P-atom in
Ph2PMe set at ꢀ27.00 ppm) was coupled across Cu to the
ꢂc
This journal is The Royal Society of Chemistry 2010
Chem. Commun., 2010, 46, 1255–1256 | 1255

View Article Online
C-atoms (18.93 and 15.55 ppm) with two-bond coupling
2
2
constants, Jtrans(13C–31P) = 74.3 Hz and Jcis(13C–31P) o 1 Hz,
1
respectively. The three-bond H–31P coupling constants were
measured at ꢀ20 1C, where they were better resolved than at
3
ꢀ100 1C: Jtrans(1H–31P) = 3.6 Hz (d, 0.48 ppm, 3H) and
3Jcis(1H–31P) = 9.5 Hz (d, ꢀ0.21 ppm, 6H, see Fig. 1).
Complex 4 was remarkably stable up to 0 1C; at 20 1C it
decomposed with a half-life of ca. 2 min. We note that
diphenylphosphine was not acidic enough to protonate and
demethylate 1. Furthermore, Ph2PLi does not undergo
exchange with 1 to give 4 (cf. Fig. 1).w Even BuLi does not
displace methyl from 1 in the temperature range ꢀ100 to 0 1C.
In contrast to its facile reactions with MeLi and Ph2PLi, 3
did not react with EtMgCl at a detectable rate at ꢀ100 1C or
upon warming to ꢀ40 1C. Similarly, EtMgCl did not displace
cyanide from EtCu(CN)MgCl in this temperature range.
N-Bromosuccinimide (NBS) reacted slowly with 1 at ꢀ100 1C,
and the only product observed was ethane. N-Iodosuccinimide
(NIS) reacted rapidly under these conditions, but again the
only product was ethane. Apparently, any methyl halide that
may have formed reacted further. Since control reactions
established that MeBr and MeI do not react with 1 at ꢀ100 1C,
they presumably reacted with the organocopper(I) species that
resulted from reductive elimination. Lithium succinimido-
(trimethyl)cuprate(III) was not observed in these reactions.
Injection of 60 mL of a 0.5 M solution of p-(tert-butyl)thio-
phenol in THF-d8 into the usual solution of 1 (30 mmol)
at ꢀ100 1C gave Me3Cu(SC10H13)Li 5. Small amounts
(ca. 10%) of Me2Cu(SC10H13)2Li were also present. When 5
was treated with BrCN, the arylthio group was rapidly
replaced by CN to give 3.
Phosgene did not react with 1 at ꢀ100 or ꢀ80 1C, but it did
at ꢀ60 1C to give ethane and acetyl chloride, which did not
react further. Halo(trialkyl)cuprate(III) complexes had
a
fleeting existence at ꢀ100 1C,2 so it is not surprising that 8
was not observed.
It appears that there is a robust analogy between organo-
cuprate(III) and organocuprate(I) reactivity. For example,
cyanide is readily displaced from Me3CuIII(CN)Li 3 by MeLi
to give Me4CuIIILi 1, just as it is from MeCuI(CN)Li to afford
the cyano-Gilman reagent, Me2CuILiꢁLiCN 2d.7 Moreover,
both CuIII and CuI cyano-complexes do not react with
EtMgCl under our conditions. Just as 1 is the CuIII analog
of the Gilman reagent, Me3CuIII(SAr)Li 5 is the CuIII analog
of MeCuI(SAr)Li, a S-heterocuprate(I) (Posner reagent),8
and both of these react with EtMgCl at ꢀ80 1C. Both
Me3CuIII(PPh2)Li 4 and MeCuI(PPh2)Li have noteworthy
thermal stability.9 In contrast, 1 and 2 differ dramatically in
their reactivities towards C-centered electrophiles, such as acid
chlorides, where only 2 reacts.
We thank the NSF for grants 0353061 and 0321056.
Notes and references
w Residual 1 was present (cf. Fig. 1) from the preparation of 3, which
allows us to say that 1 does not react with Ph2PLi at ꢀ100 1C. We used
a deficit of BrCN to make sure there was no excess to react with
Ph2PLi.
Injection of 5 with EtMgCl in THF-d8 at ꢀ100 1C gave
no reaction; however, at ꢀ80 1C they afforded the desired
product, EtMe3CuLi 6, which we observed previously as an
intermediate in the reaction of 2 with EtI.2
1 S. H. Bertz, S. Cope, M. Murphy, C. A. Ogle and B. J. Taylor,
J. Am. Chem. Soc., 2007, 129, 7208–7209.
2 S. H. Bertz, S. Cope, D. Dorton, M. Murphy and C. A. Ogle,
Angew. Chem., Int. Ed., 2007, 46, 7082–7085.
3 E. R. Bartholomew, S. H. Bertz, S. Cope, M. Murphy and
C. A. Ogle, J. Am. Chem. Soc., 2008, 130, 11244–11245.
4 E. R. Bartholomew, S. H. Bertz, S. Cope, M. Murphy, D. C. Dorton
and C. A. Ogle, Chem. Commun., 2008, 1176–1177.
5 E. R. Bartholomew, S. H. Bertz, S. K. Cope, M. D. Murphy,
C. A. Ogle and A. A. Thomas, Chem. Commun., 2010, DOI:
10.1039/b924000d.
6 T. Gartner, W. Henze and R. Gschwind, J. Am. Chem. Soc., 2007,
¨
129, 11362–11363.
7 (a) S. H. Bertz, J. Am. Chem. Soc., 1990, 112, 4031–4032;
(b) S. H. Bertz, J. Am. Chem. Soc., 1991, 113, 5470–5471.
8 G. H. Posner, An Introduction to Synthesis Using
Organocopper Reagents, Wiley, New York, 1980.
9 (a) S. H. Bertz, G. Dabbagh and G. M. Villacorta, J. Am. Chem.
Soc., 1982, 104, 5824–5826; (b) S. H. Bertz and G. Dabbagh, J. Org.
Chem., 1984, 49, 1119–1122.
Injection of 60 mL of a 0.5 M solution of imidazole (ImH) in
THF-d8 into the usual solution of 1 (30 mmol) at ꢀ100 1C
afforded Me3Cu(Im)Li 7. ImH also converted 3 to 7 under
these conditions, but at a much slower rate. Complex 7
was stable at low temperatures (ꢀ100 to ꢀ80 1C), but
decomposed at ꢀ60 1C. In contrast to its facile reactions with
1 and 5, BrCN did not react at a discernable rate with 7
at ꢀ100 1C.
ꢂc
This journal is The Royal Society of Chemistry 2010
1256 | Chem. Commun., 2010, 46, 1255–1256