Synthesis and structure of intermediates in copper-catalyzed alkylation of diphenylphosphine

Article abstract of DOI:10.1021/ic100816u

Cu(I) catalysts for alkylation of diphenylphosphine were developed. Treatment of [Cu(NCMe)₄][PF₆] (1) with chelating ligands gave [CuL(NCMe)][PF₆] (2; L = MeC(CH₂PPh₂) ₃ (triphos), 3; L = 9,9-dimethyl-4,5-bis(diphenylphosphino)xanthene (XantPhos)). These complexes catalyzed the alkylation of PHPh₂ with PhCH₂Br in the presence of the base NaOSiMe₃ to yield PPh₂CH₂Ph (4). The precursors Cu(dtbp)(X) (dtbp =2,9-di-t-butylphenanthroline, X = Cl (5) or OTf (6)), CuCl, and 1 also catalyzed this reaction, but dtbp dissociated from 5 and 6 during catalysis. Both 2 and 3 also catalyzed alkylation of PHPh₂ with PhCH ₂Cl/NaOSiMe₃, but XantPhos dissociation was observed when 3 was used. When CH₂Cl₂ was used as the solvent for alkylation of PhCH₂Cl with precursors 2 or 3, or of PhCH(Me)Br with 2, it was competitively alkylated to yield PPh₂CH₂Cl (7), which was formed exclusively using 2 in the absence of a benzyl halide. Cu(triphos)-catalyzed alkylation of PhCH(Me)Br gave mostly PPh₂CHMePh (8), along with some Ph₂P-PPh₂ (9), which was also formed in attempted alkylation of dibromoethane with this catalyst. The phosphine complexes [Cu(triphos)(L′)][PF₆] (L′ = PH₂Ph (10), PH₂CH₂Fc (Fc = C₅H₄FeC ₅H₅, 11), PHPh₂ (12), PHEt₂ (13), PHCy₂ (Cy = cyclo-C₆H₁₁, 14), PHMe(Is) (Is = 2,4,6-(i-Pr)₃C₆H₂, 15), PPh₂CH ₂Ph (16), PPh₂CH₂Cl (17)), and [Cu(XantPhos)(L′)][PF₆] (L′ = PHPh₂ (18), PPh₂CH₂Ph (19)) were prepared by treatment of 2 and 3 with appropriate ligands. Similarly, treatment of dtbp complexes 5 or 6 with PHPh₂ gave [Cu(dtbp)(PHPh₂)(X)] (X = OTf (20a) or Cl (20b)), and reaction of PPh₂CH₂Ph (4) with 1 formed [Cu(PPh₂CH₂Ph)₃][PF₆] (21). Complexes 2, 3, 11-14, 16, 17, 19, and 21 were structurally characterized by X-ray crystallography. Deprotonation of diphenylphosphine complex 12 in the presence of benzyl bromide gave diphenylbenzylphosphine complex 16, while deprotonation of 12 in CD₂Cl₂ gave 17 containing a PPh₂CD₂Cl ligand. Low-temperature deprotonation of the soluble salt 12-[B(Ar_F)₄] (Ar_F = 3,5-(CF ₃)₂C₆H₃) in THF-d₈ gave the phosphido complex Cu(triphos)(PPh₂) (22). Thermally unstable 22 was characterized by NMR spectroscopy and, in comparison to 12, by density functional theory (DFT) calculations, which showed it contained a polarized Cu-P bond. The ligand substitution step required for catalytic turnover was observed on treatment of 16 or 17 with PHPh₂ to yield equilibrium mixtures containing 12 and the tertiary phosphines 4 or 7; equilibrium constants for these reactions were 8(2) and 7(2), favoring complexation of the smaller secondary phosphine in both cases. These observations are consistent with a proposed mechanism for catalytic P-C bond formation involving deprotonation of the cationic diphenylphosphine complex [Cu(triphos)(PHPh₂)][PF ₆] (12) by NaOSiMe₃ to yield the phosphido complex Cu(triphos)(PPh₂) (22). Nucleophilic attack on the substrate (benzyl halide or CH₂Cl₂) then yields the tertiary phosphine complex [Cu(triphos)(PPh₂CH₂X)][PF₆] (X = Ph (16) or Cl (17)), and ligand substitution with PHPh₂ regenerates 12.

Full text of DOI:10.1021/ic100816u

7650 Inorg. Chem. 2010, 49, 7650–7662
DOI: 10.1021/ic100816u
Synthesis and Structure of Intermediates in Copper-Catalyzed
Alkylation of Diphenylphosphine
Matthew F. Cain,† Russell P. Hughes,† David S. Glueck,*,† James A. Golen,‡ Curtis E. Moore,‡ and
Arnold L. Rheingold‡
†6128 Burke Laboratory, Department of Chemistry, Dartmouth College, Hanover, New Hampshire 03755, and
‡Department of Chemistry, University of California, San Diego, 9500 Gilman Drive, La Jolla, California 92093
Received April 26, 2010
Cu(I) catalysts for alkylation of diphenylphosphine were developed. Treatment of [Cu(NCMe)4][PF6] (1) with chelating
ligands gave [CuL(NCMe)][PF6] (2; L = MeC(CH2PPh2)3 (triphos), 3; L = 9,9-dimethyl-4,5-bis(diphenylphosphino)-
xanthene (XantPhos)). These complexes catalyzed the alkylation of PHPh2 with PhCH2Br in the presence of the base
NaOSiMe3 to yield PPh2CH2Ph (4). The precursors Cu(dtbp)(X) (dtbp =2,9-di-t-butylphenanthroline, X = Cl (5) or OTf
(6)), CuCl, and 1 also catalyzed this reaction, but dtbp dissociated from 5 and 6 during catalysis. Both 2 and 3 also
catalyzed alkylation of PHPh2 with PhCH2Cl/NaOSiMe3, but XantPhos dissociation was observed when 3 was used.
When CH2Cl2 was used as the solvent for alkylation of PhCH2Cl with precursors 2 or 3, or of PhCH(Me)Br with 2, it was
competitively alkylated to yield PPh2CH2Cl (7), which was formed exclusively using 2 in the absence of a benzyl halide.
Cu(triphos)-catalyzed alkylation of PhCH(Me)Br gave mostly PPh2CHMePh (8), along with some Ph2P-PPh2 (9),
which was also formed in attempted alkylation of dibromoethane with this catalyst. The phosphine complexes
[Cu(triphos)(L0)][PF6] (L0 = PH2Ph (10), PH2CH2Fc (Fc = C5H4FeC5H5, 11), PHPh2 (12), PHEt2 (13), PHCy2 (Cy =
cyclo-C6H11, 14), PHMe(Is) (Is = 2,4,6-(i-Pr)3C6H2, 15), PPh2CH2Ph (16), PPh2CH2Cl (17)), and [Cu(XantPhos)-
(L0)][PF6] (L0 = PHPh2 (18), PPh2CH2Ph (19)) were prepared by treatment of 2 and 3 with appropriate ligands.
Similarly, treatment of dtbp complexes 5 or 6 with PHPh2 gave [Cu(dtbp)(PHPh2)(X)] (X = OTf (20a) or Cl (20b)), and
reaction of PPh2CH2Ph (4) with 1 formed [Cu(PPh2CH2Ph)3][PF6] (21). Complexes 2, 3, 11-14, 16, 17, 19, and 21
were structurally characterized by X-ray crystallography. Deprotonation of diphenylphosphine complex 12 in the
presence of benzyl bromide gave diphenylbenzylphosphine complex 16, while deprotonation of 12 in CD2Cl2 gave 17
containing a PPh2CD2Cl ligand. Low-temperature deprotonation of the soluble salt 12-[B(ArF)4] (ArF = 3,5-
(CF3)2C6H3) in THF-d8 gave the phosphido complex Cu(triphos)(PPh2) (22). Thermally unstable 22 was charac-
terized by NMR spectroscopy and, in comparison to 12, by density functional theory (DFT) calculations, which showed
it contained a polarized Cu-P bond. The ligand substitution step required for catalytic turnover was observed on
treatment of 16 or 17 with PHPh2 to yield equilibrium mixtures containing 12 and the tertiary phosphines 4 or 7; equi-
librium constants for these reactions were 8(2) and 7(2), favoring complexation of the smaller secondary phosphine in
both cases. These observations are consistent with a proposed mechanism for catalytic P-C bond formation involving
deprotonation of the cationic diphenylphosphine complex [Cu(triphos)(PHPh2)][PF6] (12) by NaOSiMe3 to yield the
phosphido complex Cu(triphos)(PPh2) (22). Nucleophilic attack on the substrate (benzyl halide or CH2Cl2) then yields
the tertiary phosphine complex [Cu(triphos)(PPh2CH2X)][PF6] (X = Ph (16) or Cl (17)), and ligand substitution with
PHPh2 regenerates 12.
Introduction
an enantiomerically pure DiPAMP analogue,2 enantioselective
tandem alkylation/arylation of primary phosphines to yield
1-phosphaacenaphthenes,3 and in studies of substrate- and
catalyst-controlled selectivity in alkylation of bis(secondary
We recently developed Pt catalysts for asymmetric alkyla-
tion of secondary phosphines1 and used them for synthesis of
*To whom correspondence should be addressed. E-mail: glueck@
dartmouth.edu.
(2) Anderson, B. J.; Glueck, D. S.; DiPasquale, A. G.; Rheingold, A. L.
Organometallics 2008, 27, 4992–5001. (Addition/Correction: Anderson, B. J.;
Glueck, D. S.; DiPasquale, A. G.; Rheingold, A. L. Organometallics 2009, 28,
386).
(3) Anderson, B. J.; Guino-o, M. A.; Glueck, D. S.; Golen, J. A.;
DiPasquale, A. G.; Liable-Sands, L. M.; Rheingold, A. L. Org. Lett. 2008,
10, 4425–4428.
(1) (a) Scriban, C.; Glueck, D. S. J. Am. Chem. Soc. 2006, 128, 2788–2789.
(b) Scriban, C.; Glueck, D. S.; Golen, J. A.; Rheingold, A. L. Organometallics
2007, 26, 1788–1800. (Addition/Correction: Scriban, C.; Glueck, D. S.; Golen,
J. A.; Rheingold, A. L. Organometallics 2007, 26, 5124). (c) Glueck, D. S.
Synlett 2007, 2627–2634. (d) Glueck, D. S. Coord. Chem. Rev. 2008, 252,
2171–2179.
r
pubs.acs.org/IC
Published on Web 07/09/2010
2010 American Chemical Society

Article
Inorganic Chemistry, Vol. 49, No. 17, 2010 7651
Scheme 1. Synthesis of Catalyst Precursors
phosphines).4 Independently, Bergman, Toste, and co-work-
ers developed Ru catalysts for similar selective alkylations.5
Mechanistic studies in both systems suggested that these re-
actions proceeded via phosphido intermediates, and P-C
bond formation occurred by nucleophilic attack of the
M-PR2 groups at alkyl halide electrophiles. Ligand substitu-
tion and proton transfer to a base then yielded the product
and regenerated the catalyst. The role of the metal appeared
to be two-fold: (a) it activated the phosphine substrate by
making it more nucleophilic,6 and (b) it induced rapid pyr-
amidal inversion at the P-stereogenic phosphido ligand,7
M-PR(R0), which was the origin of stereoselectivity in these
reactions.8 Since both roles appear to be fulfilled generally
across the periodic table, it is not clear that platinum group
metals are required. Instead, cheaper first-row metals might
make these catalysts more practical.
Figure 1. ORTEP diagram of the cation in one of the two independent
molecules of [Cu(triphos)(NCMe)][PF6] (2). Selected average bond
lengths (A) and angles (deg): Cu-N 1.939(5), Cu-P1 2.2688(17), Cu-P2
2.2540(18), Cu-P3 2.2742(17); N-Cu-P1 127.41(16), N-Cu-P2
117.91(16), N-Cu-P3 118.67(15), P1-Cu-P2 94.91(6), P2-Cu-P3
97.98(6), P1-Cu-P3 93.38(6).
˚
Here, we report development of Cu(I) catalysts, which
were inspired by stoichiometric alkylation of copper(I) phos-
phido complexes with benzyl bromides.9 The weaker Cu-P
bonds in these labile complexes were expected to speed up
ligand substitution processes, especially in comparison to the
known catalysts, which feature “inert” square planar Pt(II)
and octahedral Ru(II).10 However, especially with a view to
future development of asymmetric catalysts, this lability must
not extend to the ancillary ligand, which must be tightly
bound to avoid displacement by the phosphine substrate and
products, or formation of inactive μ-phosphido dimers.11
Therefore, we investigated chelate ligands such as tridentate
triphos (MeC(CH2PPh2)3) and bidentate XantPhos (9,9-
dimethyl-4,5-bis(diphenylphosphino)xanthene) and dtbp
(2,9-di-t-butyl-phenanthroline).
Figure 2. ORTEP diagram of the cation in [Cu(XantPhos)(NCMe)]-
[PF6] 2CH2Cl2 (3 2CH2Cl2). The anion and solvent are not shown.
Selected bond lengths (A) and angles (deg): Cu-P 2.2692(5), Cu-N
1.941(2), P-Cu-P 116.46(2), P-Cu-N 121.717(12).
3
3
˚
The crystal structures of these complexes showed the
expected pseudo-tetrahedral and trigonal planar geomet-
ries (Figures 1 and 2; see Table 1 and the Supporting
Information for crystallographic details).12
Results and Discussion
Cu-Catalyzed Alkylation of Diphenylphosphine. Com-
plexes 2 and 3 catalyzed the alkylation of PHPh2 with
PhCH2Br in the presence of the base13 NaOSiMe3 to yield
PPh2CH2Ph (4); so did the precursors Cu(dtbp)(X) (X=
Cl (5) or OTf (6)),14 CuCl, and 1 (Scheme 2, Table 2,
entries 1-9). Although the copper-free background reac-
tion also occurred (Table 2, entries 10-11), the catalytic
reactions (with 10 mol % loading) were faster, occurring
in a few minutes. The phosphine complexes were robust,
but dissociation of dtbp was observed in the reactions of 5
and 6.
Synthesis of Catalyst Precursors. Treatment of [Cu-
(NCMe)4][PF6] (1) with chelating ligands gave [Cu(triphos)-
(NCMe)][PF6] (2) and [Cu(XantPhos)(NCMe)][PF6] (3)
in high yields (Scheme 1).
(4) (a) Chapp, T. W.; Glueck, D. S.; Golen, J. A.; Moore, C. E.;
Rheingold, A. L. Organometallics 2010, 29, 378–388. (b) Chapp, T. W.;
Schoenfeld, A. J.; Glueck, D. S. Organometallics 2010, 29, 2465–2473.
(5) (a) Chan, V. S.; Stewart, I. C.; Bergman, R. G.; Toste, F. D. J. Am.
Chem. Soc. 2006, 128, 2786–2787. (b) Chan, V. S.; Chiu, M.; Bergman, R. G.;
Toste, F. D. J. Am. Chem. Soc. 2009, 131, 6021–6032.
(6) Glueck, D. S. Dalton Trans. 2008, 5276–5286.
(7) Rogers, J. R.; Wagner, T. P. S.; Marynick, D. S. Inorg. Chem. 1994, 33,
3104–3110.
(12) For the structures of related Cu(XantPhos) complexes, see:
(a) Kaltzoglou, A.; Fassler, T. F.; Aslanidis, P. J. Coord. Chem. 2008, 61,
1774–1781. (b) Huang, J.; Chan, J.; Chen, Y.; Borths, C. J.; Baucom, K. D.;
Larsen, R. D.; Faul, M. M. J. Am. Chem. Soc. 2010, 132, 3674–3675.
(13) NaOSiMe3 was used, as in related catalyses (reference 1), to reduce
the rate of the background reaction by minimizing deprotonation of
uncoordinated PHPh2.
(8) Glueck, D. S. Chem.;Eur. J. 2008, 14, 7108–7117.
€
(9) (a) Meyer, C.; Grutzmacher, H.; Pritzkow, H. Angew. Chem., Int. Ed.
€
€
Engl. 1997, 36, 2471–2473. (b) Meyer, C.; Scherer, M.; Schonberg, H.; Ruegger,
H.; Loss, S.; Gramlich, V.; Gr€utzmacher, H. Dalton Trans. 2006, 137–148.
(10) Langford, C. H.; Gray, H. B. Ligand Substitution Processes; W. A.
Benjamin: New York, 1966.
(11) Scriban, C.; Wicht, D. K.; Glueck, D. S.; Zakharov, L. N.; Golen,
J. A.; Rheingold, A. L. Organometallics 2006, 25, 3370–3378.
(14) Gandhi, B. A.; Green, O.; Burstyn, J. N. Inorg. Chem. 2007, 46,
3816–3825.

7652 Inorganic Chemistry, Vol. 49, No. 17, 2010
Cain et al.
Table 1. Crystallographic Data for the Cu Complexes 2, 3 2CH2Cl2, 11-14 THF, 16 1.5CH2Cl2, 17 EtOH, and 19 2CH2Cl2
3
3
3
3
3
compound
formula
2
3 2CH2Cl2
3
11
12
13
14 THF
3
16 1.5CH2Cl2
3
17 EtOH
3
19 2CH2Cl2
3
C43H42CuF6 C43H39Cl4
C52H52Cu C54.5H53Cl3 C45H52CuF6 C57H70CuF6 C61.5H59Cl3
C56H57ClCuF6 C60H53Cl4
NP4
874.20
P2(1)/c
CuF6NOP3
998.00
P2(1)/m
F6FeP5
1065.18
P2(1)/n
11.866(3) 10.3838(6)
32.018(9) 12.8816(7)
CuF6P5
1146.71
P1
OP5
941.26
P2(1)/n
13.0803(6)
16.9559(8)
OP5
1103.52
P2(1)
12.1730(3)
17.3557(4)
CuF6P5
1236.83
P2(1)/n
16.0842(7)
16.9120(7)
21.1911(9)
90
OP5
1113.86
P2(1)/c
10.627(2)
25.660(5)
19.383(4)
90
100.073(4)
90
5204.0(19)
4
1.422
CuF6OP4
1233.24
P1
formula wt
space group
˚
a, A
29.1600(11) 8.6254(12)
13.449(2)
13.521(2)
17.421(3)
74.159(2)
70.067(2)
73.684(2)
2802.5(8)
2
˚
b, A
14.0993(6)
20.7714(8)
90
110.820(3)
90
17.945(3)
14.093(2)
90
96.776(4)
90
2166.0(5)
2
1.530
˚
c, A
12.795(3) 21.8230(12) 21.4791(10) 12.9747(3)
90 86.5950(10) 90 90
R, deg
β, deg
γ, deg
98.794(6) 77.5010(10) 107.3490(10) 103.3610(10) 92.301(3)
69.2370(10) 90
90
4804(2)
4
90 90
2666.98(11) 5759.7(4)
2
1.374
2.500a
3
˚
V, A
Z
7982.2(5)
8
2664.3(3)
2
1.429
0.768
4547.1(4)
4
1.375
0.714
4
D(calcd), g/cm3 1.455
1.473
0.971
1.426
3.621a
1.461
0.756
μ(MoKR),
2.811a
0.924
0.686
mm-1
temp, K
100(2)
4.99
11.97
100(2)
3.09
8.04
150(2)
4.47
9.68
100(2)
5.93
14.43
100(2)
4.06
10.95
100(2)
3.65
9.58
120(2)
5.58
14.89
100(2)
6.43
13.38
100(2)
8.06
21.66
R(F), %b
Rw(F2), %b
P
P
P
P
a Cu KR radiation was used. b Quantity minimized: Rw(F2) = [ w(Fo2 - Fc2)2/ w(Fo2)2]1/2; R = Δ/ (Fo), Δ = |(Fo - Fc)|, w = 1/[σ2(Fo2) þ (aP)2
2
þ bP], P = [2Fc þ Max(Fo2,0)]/3. A Bruker CCD diffractometer was used in all cases.
Scheme 2. Cu-Catalyzed Alkylation of PHPh2 with PhCH2Br
Alkylation by the CH2Cl2 solvent to yield 7 also
occurred in the Cu(triphos)-catalyzed reaction of PHPh2
with the more hindered benzyl bromide PhCH(Me)Br,
which gave PPh2CHMePh (8) as the major product17
along with PPh2CH2Cl (7) and the biphosphine Ph2P-
PPh2 (9; Table 2, entry 21). Only 8 and 9 were formed
in THF, much faster than in the background reaction
(Scheme 4; Table 2, entries 20 and 22). Cu(triphos)-
mediated alkylation of PhCH(Me)Cl was much slower,
yielding mostly 8 along with a small amount of 9; tri-
phos dissociation was also observed (Table 2, entries
23-24).
Biphosphine 9 was formed cleanly in Cu-catalyzed
reaction of PHPh2 with dibromoethane in THF or
CH2Cl2 (Scheme 5; Table 2, entries 25-27). Several
related reactions of phosphido anions with dibromo-
ethane and related alkyl or aryl halides are known;
they have been proposed to occur via nucleophilic attack
on halogen instead of the “normal” SN2 attack at
carbon.18
We have not investigated the ligand effects on the
reactions in Table 2 in detail. However, the success of
the presumably tridentate triphos ligand, and the disso-
ciation observed for bidentate XantPhos and dtbp sug-
gested that strong chelation by a polydentate ligand was
important to stabilize catalytic intermediates.
Triphos complex 2 also catalyzed alkylation of PHPh2
with benzyl chloride in tetrahydrofuran (THF) or CH2Cl2
to yield 4 (Scheme 3, Table 2, entries 12-13). This reac-
tion was slower than alkylation with PhCH2Br, but the
even slower background process was now insignificant
(Table 2, entries 16-17). When XantPhos precursor 3 was
used, 4 was also formed, but XantPhos dissociation was
observed (Table 2, entries 14-15).
When methylene chloride was used as the solvent for
reactions of benzyl chloride, it was competitively alkyla-
ted to yield mixtures of 4 and the known chloromethyl-
phosphine PPh2CH2Cl (7; Table 2, entries 12 and 14).15 In
contrast, alkylation of diphenylphosphine with benzyl
bromide proceeded smoothly in CH2Cl2 with 2, 3, or other
catalyst precursors (Table 2, entries 1-3, 5, and 7-9).
When no benzyl halide was added, Cu(triphos)-catalyzed
alkylation of methylene chloride solvent gave 7 in high
yield (Scheme 3, Table 2, entry 18; no background reac-
tion occurred (entry 19)). Although such a stoichiometric
reaction was reported for a Re phosphido complex, this is
the first observation of catalysis.16
Mechanism of Cu-Catalyzed Alkylation of Diphenylpho-
sphine. We hypothesized that these reactions proceeded by
the catalytic cycle shown in Scheme 6, in which benzyl bro-
mide is shown as a sample electrophile.
(17) (a) Heydenreich, F.; Mollbach, A.; Wilke, G.; Dreeskamp, H.;
Hoffmann, E. G.; Schroth, G.; Seevogel, K.; Stempfle, W. Isr. J. Chem.
1972, 10, 293–319. (b) Millauer, H.; Brungs, P. U.S. 5808163, 1998. (c) Berens,
U. EP 1582527A1, 2005.
(18) (a) Seyferth, D.; Wood, T. G.; Henderson, R. S. J. Organomet. Chem.
1987, 336, 163–182. (b) Gillespie, D. G.; Walker, B. J. Tetrahedron Lett. 1975,
16, 4709–4710. (c) Gillespie, D. G.; Walker, B. J.; Stevens, D.; McAuliffe, C. A.
J. Chem. Soc., Perkin Trans. 1 1983, 1697–1703. (d) Gillespie, D. G.; Walker,
B. J. J. Chem. Soc., Perkin Trans. 1 1983, 1689–1695. (e) Hewertson, W.;
Taylor, I. C.; Trippett, S. J. Chem. Soc. C 1970, 1835–1839. (f) Hewertson, W.;
Taylor, I. C. J. Chem. Soc. D 1970, 119–120. (g) Issleib, K.; Muller, D.-W.
(15) (a) Braussaud, N.; Ruther, T.; Cavell, K. J.; Skelton, B. W.; White,
A. H. Synthesis 2001, 626–632. (b) Burck, S.; van Assema, S. G. A.; Lastdrager,
B.; Slootweg, J. C.; Ehlers, A. W.; Otero, J. M.; Dacunha-Marinho, B.; Llamas-
Saiz, A. L.; Mar, O.; van Raaij, M. J.; Lammertsma, K. Chem.;Eur. J. 2009, 15,
8134–8145.
(16) (a) Buhro, W. E.; Georgiou, S.; Hutchinson, J. P.; Gladysz, J. A.
J. Am. Chem. Soc. 1985, 107, 3346–3348. (b) Buhro, W. E.; Zwick, B. D.;
Georgiou, S.; Hutchinson, J. P.; Gladysz, J. A. J. Am. Chem. Soc. 1988, 110,
2427–2439.
€
Chem. Ber. 1959, 92, 3175–3182. (h) Issleib, K.; Doll, G. Chem. Ber. 1961, 94,
€
2664–2669. (i) Issleib, K.; Doll, G. Chem. Ber. 1963, 96, 1544–1550.

Article
Inorganic Chemistry, Vol. 49, No. 17, 2010 7653
Table 2. Cu-Catalyzed Alkylation of Diphenylphosphinea
entry
catalyst precursor
substrate
product
solvent
time
yield (%)
1
2
3
4
5
6
7
8
[Cu(NCMe)4][PF6] (1)
PhCH2Br
PhCH2Br
PhCH2Br
PhCH2Br
PPh2CH2Ph (4)
PPh2CH2Ph (4)
PPh2CH2Ph (4)
PPh2CH2Ph (4)
PPh2CH2Ph (4)
PPh2CH2Ph (4)
PPh2CH2Ph (4)
PPh2CH2Ph (4)
PPh2CH2Ph (4) Ph2P-PPh2 (9)
PPh2CH2Ph (4) [PPh2(CH2Ph)2][Br]
PPh2CH2Ph (4) [PPh2(CH2Ph)2][Br]
PPh2CH2Ph (4) PPh2CH2Cl (7)
PPh2CH2Ph (4)
CH2Cl2 15 min
CH2Cl2 <10 min
CH2Cl2 2 h
73
82
87
89
81
98
d
[Cu(triphos)(NCMe)][PF6] (2)
[Cu(triphos)(NCMe)][PF6] (2)b
[Cu(triphos)(NCMe)][PF6] (2)
THF
CD2Cl2 <15 min
THF <15 min
<15 min
[Cu(XantPhos)(NCMe)][PF6] (3) PhCH2Br
[Cu(XantPhos)(NCMe)][PF6] (3) PhCH2Br
Cu(dtbp)(Cl) (5)c
PhCH2Br
PhCH2Br
PhCH2Br
PhCH2Br
PhCH2Br
PhCH2Cl
PhCH2Cl
CD2Cl2 <15 min
CD2Cl2 <15 min
CH2Cl2 15 min
CD2Cl2 15 min
THF 15 min
CH2Cl2 1 h
Cu(dtbp)(OTf) (6)e
CuCl
d
9
84f
g
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
none
none
h
[Cu(triphos)(NCMe)][PF6] (2)
[Cu(triphos)(NCMe)][PF6] (2)
[Cu(XantPhos)(NCMe)][PF6] (3) PhCH2Cl
[Cu(XantPhos)(NCMe)][PF6] (3) PhCH2Cl
none
none
[Cu(triphos)(NCMe)][PF6] (2)
none
[Cu(triphos)(NCMe)][PF6] (2)
[Cu(triphos)(NCMe)][PF6] (2)
none
[Cu(triphos)(NCMe)][PF6] (2)
none
[Cu(triphos)(NCMe)][PF6] (2)
[Cu(triphos)(NCMe)][PF6] (2)
none
75i
93
75 j,k
100k
l
THF 15 min
CH2Cl2 15 min
PPh2CH2Ph (4) PPh2CH2Cl (7)
PPh2CH2Ph (4)
PPh2CH2Ph (4)
PPh2CH2Ph (4)
PPh2CH2Cl (7)
THF
CD2Cl2 72 h
THF 3 h
27 h
PhCH2Cl
PhCH2Cl
CH2Cl2
m
CH2Cl2 25 min
CH2Cl2 48 h
86
0n
100o
p
CH2Cl2 PPh2CH2Cl (7)
PhCH(Me)Br PPh2CH(Me)Ph (8) Ph2P-PPh2 (9)
THF
PhCH(Me)Br PPh2CH(Me)Ph (8) PPh2CH2Cl (7) Ph2P-PPh2 (9) CH2Cl2 20 min
15 min
PhCH(Me)Br PPh2CH(Me)Ph (8) Ph2P-PPh2 (9)
THF
THF
THF
THF
15 min
72 h
75 h
q
PhCH(Me)Cl
PhCH(Me)Cl
Br(CH2)2Br
Br(CH2)2Br
Br(CH2)2Br
PPh2CH(Me)Ph (8)
PPh2CH(Me)Ph (8)
Ph2P-PPh2 (9)
Ph2P-PPh2 (9)
Ph2P-PPh2 (9)
74r
s
15 min
77t
71u
v
CH2Cl2 15 min
THF 15 min
a Standard conditions for catalytic alkylation: 50 mg of PHPh2 (0.27 mmol), 1 equiv of electrophile, 10 mol % Cu catalyst precursor, 2 mL of solvent, 1
equiv of NaOSiMe3. Reactions were monitored by 31P NMR spectroscopy, then worked up as described in the Experimental Section and Supporting
Information. Isolated yields are reported. b 1 mol % catalyst precursor. c 0.51 mmol of PHPh2. d Phosphine 4 was the only product observed by 31P NMR,
but the 1H NMR spectrum showed free dtbp. e 0.47 mmol of PHPh2. f Mostly 4, with a trace of 9. g 22% conversion; 2:1 ratio of 4 to the salt
[PPh2(CH2Ph)2][Br]. h 61% conversion to a mixture of 4 and the salt [PPh2(CH2Ph)2][Br] (25:1). i 93:7 mixture of 4 and 7 (88:12 before workup). j 97:3
mixture of 4 and 7. k XantPhos dissociation was observed. l <5% conversion. m 5% conversion; 14% conversion after 16.5 h. n No reaction occurred.
o 97:3 ratio of 8 to 9 before and after workup e. p The initial ratio of 8:7:9 was 68:28:4. q After 15 min, about 16% conversion to a 2.5:1 mixture of 8 and 9.
r triphos dissociation was observed. Phosphine 8 was obtained in 94% purity, containing 9, triphos, and another unidentified byproduct (δ -21.4) s 3%
conversion. t Contained traces of Ph2P-PPh2(O), PHPh2, and another unidentified byproduct. u Contained a trace of Ph2P-PPh2(O). v ca. 1% conversion;
even after 48 h, only 2% conversion of PHPh2.
Scheme 3. Cu(triphos)-Catalyzed Alkylation of PHPh2 with Benzyl
Halides and CH2Cl2
Scheme 5. Cu(triphos)-Catalyzed Reaction of PHPh2 with Dibro-
moethane
Scheme 6. Hypothetical Mechanism of Cu-Catalyzed Alkylation of
PHPh2 with Benzyl Bromide
Scheme 4. Cu(triphos)-Catalyzed Reactions of PHPh2 with Second-
ary Benzyl Halides
such as a benzyl halide, would yield a cationic tertiary
phosphine complex (C). Finally, ligand substitution on C
would regenerate A. To investigate these steps, we pre-
pared examples of the proposed intermediates.
Intermediates A and C (Cationic Phosphine Complexes).
The phosphine complexes [Cu(triphos)(L0)][PF6] (L0 =
PH2Ph (10), PH2CH2Fc (Fc = C5H4FeC5H5, 11), PHPh2
Reaction of a catalyst precursor, like 2 or 3, with the
phosphine substrate was expected to yield a cationic
secondary phosphine complex A, which on deprotona-
tion by NaOSiMe3 would yield a nucleophilic phosphido
intermediate (B). Attack on the electrophilic substrate,

7654 Inorganic Chemistry, Vol. 49, No. 17, 2010
Cain et al.
Scheme 7. Synthesis of Copper Phosphine Complexes 10-21a
Figure 3. ORTEP diagram of the cation in [Cu(triphos)(PH2CH2Fc)]-
[PF6] (11). Only the PH hydrogen atoms, which were located and refined,
are shown.
a Ligands L: PH2Ph (10), PH2CH2Fc (11), PHPh2 (12 and 18), PHEt2
(13), PHCy2 (14), PHMe(Is) (15), PPh2CH2Ph (16 and 19), PPh2CH2Cl
(17). For 20a, X = OTf; for 20b, X = Cl.
(12), PHEt2 (13), PHCy2 (Cy = cyclo-C6H11, 14), PHMe-
(Is) (Is = 2,4,6-(i-Pr)3C6H2, 15), PPh2CH2Ph (16), PPh2-
CH2Cl (17)), and [Cu(XantPhos)(L0)][PF6] (L0 = PHPh2
(18), PPh2CH2Ph (19)) were prepared by treatment of 2
and 3 with appropriate ligands. Similarly, reaction of dtbp
complexes 5 or 6 with PHPh2 gave [Cu(dtbp)(PHPh2)X]
(X = OTf (20a), Cl (20b)), and reaction of PPh2CH2Ph (4)
with 1 formed [Cu(PPh2CH2Ph)3][PF6] (21) (Scheme 7).
Spectroscopic characterization of these complexes was,
in most cases, straightforward. The PH groups in the pri-
mary and secondary phosphine complexes were useful
spectroscopic probes, with JPH coupling constants of
about 300-400 Hz and IR PH stretching vibrations in
Figure 4. ORTEP diagram of the cation in [Cu(triphos)(PHPh2)][PF6]
(12). Only the PH hydrogen atom, which was located and refined, is
shown.
the range 2300-2400 cm .
-1 19 Phosphine binding to copper
resulted in large 31P NMR coordination chemical shifts,
with broad 31P NMR signals due to the quadrupolar 65Cu
and 63Cu nuclei.20
1
However, the 31P and H NMR spectra of 20a (X =
OTf) and 20b (X = Cl) in CD2Cl2 were markedly dif-
ferent, suggesting that the anion X was bound to copper
in one or both complexes. The conductivities of ∼1 ꢀ
10-3 M solutions of 20a, 20b, and [Cu(dtbp)(NCMe)]-
[PF6]14 in nitromethane were 139, 54, and 149 Ω-1 cm2
mol-1, respectively, consistent with an ionic three-coordi-
nate structure for triflate complex 20a and with the pre-
sence of both three- and four-coordinate forms of chlor-
ide 20b in this polar solvent.21 For simplicity, Scheme 7
shows 20 as four-coordinate, but this likely depends on
the anion and the solvent.
Complexes 11-14, 16-17, 19, and 21 were structurally
characterized by X-ray crystallography (Figures 3-9; see
Tables 1 and 3 and the Supporting Information for more
details and the structure of 21).
Although the steric and electronic properties of the mono-
dentate ligands in the complexes [Cu(triphos)(L)][PF6]
However, the structures of the diphenylphosphine
complexes of XantPhos (18) and dtbp (20) were less easy
to identify. Elemental analysis and mass spectrometry
were consistent with the formula of 18. The three-coordi-
nate structure of its PPh2CH2Ph analogue 19 was
established by its 31P NMR spectrum, which included a
doublet and a triplet (JPP =98 Hz), and by X-ray crystal-
lography (see below). However, room-temperature NMR
spectra of 18 contained very broad signals, and cooling
the sample did not simplify them. This behavior might
result from rapid exchange with a small amount of
PHPh2. For dtbp complex 20, the bulky ligand was ex-
pected to promote formation of three-coordinate cations
[Cu(dtbp)(PHPh2)][X], as observed in the structures of
the precursors Cu(dtbp)(X) (X = Cl (5), OTf (6)).14
(19) Kourkine, I. V.; Maslennikov, S. V.; Ditchfield, R.; Glueck, D. S.;
Yap, G. P. A.; Liable-Sands, L. M.; Rheingold, A. L. Inorg. Chem. 1996, 35,
6708–6716.
(20) Black, J. R.; Levason, W.; Spicer, M. D.; Webster, M. J. Chem. Soc.,
Dalton Trans. 1993, 3129–3136.
(21) (a) Geary, W. J. Coord. Chem. Rev. 1971, 7, 81–122. (b) Tye, J. W.;
Weng, Z.; Giri, R.; Hartwig, J. F. Angew. Chem., Int. Ed. 2010, 49, 2185–2189.
(c) Tye, J. W.; Weng, Z.; Johns, A. M.; Incarvito, C. D.; Hartwig, J. F. J. Am.
Chem. Soc. 2008, 130, 9971–9983.

Article
Inorganic Chemistry, Vol. 49, No. 17, 2010 7655
Figure 5. ORTEP diagram of the cation in [Cu(triphos)(PHEt2)][PF6]
(13). Only the PH hydrogen atom, which was located and refined, is
shown.
Figure 8. ORTEP diagram of the cation in [Cu(triphos)(PPh2CH2Cl)]-
[PF6] EtOH (17 EtOH). The solvent is not shown.
3
3
Figure 6. ORTEP diagram of the cation in [Cu(triphos)(PHCy2)]-
[PF6] THF (14 THF). The solvent and disorder in a cyclohexyl group
are not shown.
3
3
Figure 9. ORTEP diagram of the cation in [Cu(XantPhos)(PPh2CH2Ph)]-
[PF6] 2CH2Cl2 (19 2CH2Cl2). The solvent is not shown.
3
3
phosphine complexes, as also seen in the recently reported
structure of [Cu(triphos)(PPh3)][BF4].22
The structure of 12 was also investigated by density
functional theory (DFT) calculations. As shown in Table 3,
the crystallographic structure of 12 was in excellent agree-
ment with the computed one. Although the DFT results
˚
overestimated the Cu-P bond lengths by about 0.1 A,
they reproduced the crystallographically established trend
that the Cu-P(PHPh2) bond was slightly shorter than the
Cu-P(triphos) bonds, and they also matched the angles at
copper well. An overlay of the crystallographic and DFT
structures of the cation of 12 is included in the Supporting
Information.
Intermediate B (Terminal Cu-Phosphido Complexes).
Formation by Deprotonation of Cationic Phosphine Com-
plexes and Reaction with Electrophiles. No isolable neu-
tral terminal Cu-phosphido complexes have been repor-
ted,23-25 so we expected proposed intermediates B to be
Figure 7. ORTEP diagram of the cation in [Cu(triphos)(PPh2CH2Ph)]-
[PF6] 1.5CH2Cl2 (16 1.5CH2Cl2). The solvent is not shown.
3
3
varied, the data in Table 3 showed that the crystal
structures were similar, with the expected pseudo-tetra-
hedral geometry at copper. On replacing the P-H sub-
stituent in 12 with a P-CH2Ph or P-CH2Cl group in 16
or 17, all the Cu-P bond lengths increased slightly,
consistent with increased steric hindrance in the tertiary
(22) Yin, Q.; Gan, X. Acta Crystallogr. 2010, E66, m369.
(23) For anionic terminal phosphido copper complexes, see: (a) Cowley,
A. H.; Giolando, D. M.; Jones, R. A.; Nunn, C. M.; Power, J. M. J. Chem.
Soc., Chem. Commun. 1988, 208–209. (b) Martin, S. F.; Fishpaugh, J. R.; Power,
J. M.; Giolando, D. M.; Jones, R. A.; Nunn, C. M.; Cowley, A. H. J. Am. Chem.
Soc. 1988, 110, 7226–7228.

7656 Inorganic Chemistry, Vol. 49, No. 17, 2010
Cain et al.
˚
Table 3. Selected Bond Lengths (A) and Angles (deg) in the Cationic Complexes [Cu(triphos)(L)][PF6]
L
Cu-P(L)
Cu-P (triphos)
P-Cu-P(L)
P-Cu-P (triphos)
PH2CH2Fc (11)
PH2Mesa
PHPh2 (12)
PHPh2 (12-DFT)
PHEt2 (13)
PHCy2 (14)
2.2278(12)
2.241(3)
2.2262(9)
2.3298
2.2466(7)
2.2531(6)
2.2648(11)
2.2642(13)
2.2852(11)
2.2883(11) 2.2817(11) 2.2781(11)
2.279(3) 2.289(3) 2.262(3)
2.2692(9) 2.2776(9) 2.2804(9)
2.3913 (ave)
2.2817(6) 2.2866(6) 2.2922(6)
2.2914(6) 2.3040(6) 2.3367(6)
2.2971(11) 2.3041(10) 2.3296(10)
2.3085(14) 2.3130(14) 2.3093(12)
2.2983(12) 2.3177(12) 2.3314(12)
118.36(4) 128.01(4) 119.15(4)
125.81(11) 119.50(10) 118.55(10)
124.83(3) 120.47(3) 120.24(3)
122.25 (ave)
121.61(2) 127.32(2) 118.75(3)
125.51(3) 122.13(2) 120.73(2)
121.90(4) 127.92(4) 119.08(4)
125.49(5) 120.10(5) 121.54(5)
122.79(4) 124.27(4) 117.67(4)
94.50(4) 92.00(4) 96.77(4)
94.6(1) 97.2(1) 94.13(9)
96.58(3) 94.95(3) 92.28(3)
94.16 (ave)
93.92(2) 95.55(2) 91.66(2)
95.35(2) 92.93(2) 91.72(2)
93.25(4) 94.25(4) 91.92(4)
91.78(4) 90.84(5) 99.53(5)
91.17(4) 96.58(4) 97.65(4)
PPh2CH2Ph (16)
PPh2CH2Cl (17)
PPh3
b
a Kourkine, I. V.; Maslennikov, S. V.; Ditchfield, R.; Glueck, D. S.; Yap, G. P. A.; Liable-Sands, L. M.; Rheingold, A. L. Inorg. Chem. 1996, 35,
6708-6716. b Yin, Q.; Gan, X. Acta Crystallogr. 2010, E66, m369.
Scheme 8. Deprotonation of 12: Formation of Cu-Phosphido Com-
highly reactive. Attempts to observe such intermediates
with dtbp or XantPhos ligands were unsuccessful. Treat-
plex 22 and Its Reactions with Electrophilesa
ment of [Cu(dtbp)(PHPh2)(OTf)] (20a) with NaOSiMe3
at -78 °C in THF-d8 gave diphenylphosphine and dtbp,
1
observed by 31P and H NMR spectroscopy already at
-70 °C. On warming to room temperature, the products
were dtbp and a brick-red, insoluble precipitate, which is
presumably the known [Cu(PPh2)]n.25 Consistent with this
assignment, independently prepared [Cu(PPh2)]n did not
react with dtbp under conditions in which the less sterically
hindered N∼N ligands bipy, phen, or neocuproine (2,9-
Me2phen) readily solubilized this oligomer to give dimeric
or trimeric [Cu(N∼N)(PPh2)]n.9 Similarly, low-temperature
reaction of NaOSiMe3 with [Cu(XantPhos)(PHPh2)][PF6]
(18) gave XantPhos and several other unidentified P-con-
taining compounds.
In contrast, we obtained evidence for the intermediate
Cu(triphos)(PPh2) (22) by trapping it with electrophiles
and by direct NMR observation. Deprotonation of cat-
ionic PHPh2 complex 12 in the presence of benzyl bro-
mide gave diphenylbenzylphosphine complex 16, while
deprotonation of 12 in CD2Cl2 gave 17-D2 containing a
PPh2CD2Cl ligand (Scheme 8). These observations were
consistent with the expected intermediacy of 22, which
was not observed under these conditions. However, gen-
eration of 22 in the absence of an electrophile was hind-
ered by its limited solubility in nonpolar solvents. There-
fore, soluble 12-[B(ArF)4] (ArF = 3,5-(CF3)2C6H3) was
prepared by anion exchange. Its deprotonation at low
temperature in THF-d8 gave 22 (Scheme 8), which was
readily identified by its 31P NMR spectrum (-50 °C, δ
-23.4 (d, J = 27 Hz, triphos); -30.3 (q, J = 27 Hz, PPh2));
in Cu(PPh3)(PPh2) the phosphido group gave rise to a
31P NMR signal at δ -32.3 (d, J = 80 Hz; 30 °C, C6D6).25
Intermediate 22 was also characterized by 1H NMR
spectroscopy.
a [Cu] = Cu(triphos). The anion for cations 12, 16, and 17-D2 was PF6;
22 was generated by deprotonation of 12-B(ArF)4.
comparison to precursor cation 12, by DFT calculations
(Table 4).
On deprotonation, both the Cu-P(PHPh2) bond and
the Cu-P(triphos) bonds lengthened slightly (for triphos,
˚
the average Cu-P bond length in cation 12 was 2.3913 A;
compare 2.4290 A in 22). This may reflect reduced
˚
electrostatic Cu-P attraction on conversion of cationic
12 to neutral 22, the increased steric demands of the
phosphido lone pair in comparison to the P-H group,
or changes in hybridization at phosphorus.26 While the
P-Cu-P(L) angles showed greater variation in 22 than in
12, their average value (124.37° (22)) did not change much
from that in 12 (122.25°). In contrast, the P-Cu-P-
(triphos) angles decreased slightly. The sum of angles at
P in the phosphido complex was 322.2°; compare to
328.5° for an idealized tetrahedral geometry. This pyra-
midal structure is consistent with earlier observations on
Pt and Re terminal phosphido complexes.27
Phosphido complex 22 underwent noticeable decom-
position on warming to -30 °C and did not survive at
room temperature; identifiable products included triphos
and PHPh2. To explore its high reactivity and nucleophi-
lic behavior, intermediate 22 was characterized further, in
The nucleophilic nature of the M-X groups (X = OH,
OR, SR, NR2, PR2) in complexes with late transition
metal-heteroatom bonds has been ascribed either to
(24) An equilibrium between dimeric [Cu(dppe)(PPh2)]2 and monomeric
Cu(dppe)(PPh2) was proposed on the basis of ebullioscopic molecular weight
measurements: (a) Van Koten, G.; Noltes, J. G.; Spek, A. L. J. Organomet.
Chem. 1978, 159, 441–463. (b) Greiser, T.; Weiss, E. Chem. Ber. 1978, 111,
516–522.
(26) (a) Deeming, A. J.; Doherty, S.; Marshall, J. E.; Powell, J. L.; Senior,
A. M. J. Chem. Soc., Dalton Trans. 1993, 1093–1100. (b) Zhuravel, M. A.;
Glueck, D. S.; Zakharov, L. N.; Rheingold, A. L. Organometallics 2002, 21,
3208–3214. (c) Bonnet, G.; Kubicki, M. M.; Moise, C.; Lazzaroni, R.; Salvadori,
P.; Vitulli, G. Organometallics 1992, 11, 964–967. (d) Reference 16a.
(27) (a) Mastrorilli, P. Eur. J. Inorg. Chem. 2008, 4835–4850. (b) Eichen-
seher, S.; Delacroix, O.; Kromm, K.; Gladysz, J. A. Organometallics 2005, 24,
245–255.
(25) Two-coordinate Cu(PPh3)(PPh2) was observed by 31P NMR spec-
troscopy: Lemmen, T. H.; Goeden, G. V.; Huffman, J. C.; Geerst, R. L.;
Caulton, K. G. Inorg. Chem. 1990, 29, 3680–3685.

Article
Inorganic Chemistry, Vol. 49, No. 17, 2010 7657
˚
Table 4. Selected Bond Lengths (A) and Angles (deg) in the Complexes
[Cu(triphos)(L)]nþ (L = PHPh2, n = 1, 12; L = PPh2, n = 0, 22), from DFT
Calculations
Cu-P
(triphos)
P-Cu-P
(triphos)
L
Cu-P(L)
P-Cu-P(L)
PHPh2 (12)
2.3298
2.3911
2.3909
2.3918
2.4478
2.4411
2.3981
122.21
120.06
124.48
132.90
114.86
125.35
94.00
94.01
94.47
91.12
91.44
89.68
PPh2 (22)
2.3374
Table 5. Calculated Orbital Energies (eV) for Metal Phosphido Complexes and
Phosphines
P(p) HOMO
energy
P(hybrid) NHOMO
energy
compound
Li(triphosH6)-PH2
Na(triphosH6)-PH2
K(triphosH6)-PH2
Cu(triphosH6)-PH2
Cu(triphos)-PPh2
PH2 anion
PPh2 anion
PPh2-Li
PPh2-Na
PPh2-K
PPh2-Cu
PH2-Li
PH2-Na
;0.1360
;0.1335
;0.1166
;0.1589
;0.1419
þ0.0347
;0.0039
;0.1668
;0.1543
;0.1371
;0.2007
;0.1808
;0.1741
;0.1445
;0.2346
;0.2152
;0.2135
;0.2017
;0.2108
;0.1983
;0.0598
;0.0938
;0.2433
;0.2322
;0.2220
;0.2476
;0.2538
;0.2497
;0.2281
;0.2674
PH2-K
PH2-Cu
P(hybrid) HOMO energy (eV)
PH2-H
PPh2-H
PPh2-CH2Ph
;0.2801
;0.2354
;0.2257
repulsive π-symmetry interactions between filled metal
orbitals and X lone pair orbitals, and/or to the ionic
nature of the polarized Mδþ-Xδ- bonds.28,6 We used
DFT calculations on a series of phosphines and phosphi-
do complexes to investigate the observed nucleophilicity of
22. Alkali metal phosphido complexes were included to assess
the effect of ionic bonding in cases where repulsive orbital
interactions were not possible. Table 5 lists the calculated
highest occupied molecular orbital (HOMO) and next high-
est occupied molecular orbital (NHOMO) energies.6
In the free PH2; anion the HOMO is the b2 p-orbital
lying perpendicular to the molecular plane (Figure 10A),
as expected, and the NHOMO is the a1 MO lying in the
molecular plane (Figure 10B); the NHOMO contains
considerable s-character, making it lower in energy than
the HOMO. When PH2; is protonated to give a covalent
P-H bond, the p-orbital is used to give a low lying σ-
bonding MO, leaving the original PH2; NHOMO as the
new HOMO of the tertiary phosphine (Figure 10C).
However, when a more electropositive element is used
to make a metal-phosphido compound, the new bond
Figure 10. DFT Kohn-Sham HOMOs and NHOMOs for phosphido
anions and metal phosphido complexes.
HOMO of the phosphido complex, and as such is pre-
sumably responsible for the nucleophilic character of the
metal-phosphido entity. As shown in Table 5, the energy
of this HOMO is raised on changing the metal from Li to
Na to K, in agreement with electronegativity ideas. For
CuPH2 the HOMO and NHOMO remain unchanged in
nature (Figures 10F and 10G); compared to the alkali
metal phosphides there is more covalency in the HOMO,
as expected for the less electropositive Cu, but the HOMO
is still higher in energy than that in PH3. Adding tri-
phosH6 (triphosH6 = MeC(CH2PH2)3) ligands to the
MPH2 compounds raises the energy of the HOMO in
each case, while not changing its overall character. Pic-
tures of all the valence MOs of CuPH2 are provided in the
Supporting Information.
;
formed using the PH2 p-orbital has considerable ionic
character, as shown for LiPH2 in Figure 10D; while the
energy of this molecular orbital (MO) is lowered relative
to free PH2 it is still high enough in energy to be the
;
(28) (a) Caulton, K. G. New J. Chem. 1994, 18, 25–41. (b) Holland, P. L.;
Andersen, R. A.; Bergman, R. G. Comments Inorg. Chem. 1999, 21, 115–129.

7658 Inorganic Chemistry, Vol. 49, No. 17, 2010
Cain et al.
These observations carry through to CuPPh2 (Figure 10H)
and importantly also for Cu(triphos)(PPh2) 22 (Figure 10I),
in which the HOMO is still mostly phosphorus p in
character, and considerably higher in energy than
either PHPh2 or PPh2CH2Ph, consistent with their
relative nucleophilicities.
Scheme 9. Ligand Substitution in Cationic Cu(triphos) Complexes
Thus, the observed nucleophilicity of 22 could be
explained in terms of a polarized Cu-P bond; the domi-
nant contribution of the phosphorus p-orbital to the
HOMO composition results from the relative electrone-
gativities of these elements. In the alternative model, the
HOMO of 22 is a σ*-antibonding MO, whose increased
energy results from filled-filled repulsive Cu-P orbital
interactions. These computational results are similar to
those for related Cu(I) anilido complexes, where the
Kohn-Sham HOMOs were calculated to be primarily
NHPh p-π orbitals, with minor π*-character from mixing
with a Cu d-orbital.29
Scheme 10. Proposed Mechanism for Cu(triphos)-Catalyzed Alkyla-
tion of Diphenylphosphine
Ligand Substitution. The ligand substitution step re-
quired for catalytic turnover was observed on treatment
of 16 or 17 with PHPh2 to yield equilibrium mixtures also
containing 12 and the tertiary phosphines 4 or 7; equilib-
rium constants for these reactions were about 8(2) and
7(2) respectively, favoring complexation of the smaller
secondary phosphine (Scheme 9). These reactions oc-
curred on mixing and reached equilibrium within min-
utes; identical results were obtained starting from 12.30
observed dissociation of bidentate XantPhos and dtbp lig-
ands confirmed our fears that this lability could be a problem.
However, Cu catalysts containing tridentate triphos were
more robust and enabled a new reaction, catalytic phosphi-
nation of methylene chloride. In future work, we plan to
investigate the scope and limitations of such P-C bond-
forming reactions, and to develop related asymmetric cata-
lysis using chiral analogues of triphos.34
Conclusions
Theseobservationsareconsistent with a proposedmechan-
ism for catalytic P-C bond formation, involving deprotona-
tion of the cationic diphenylphosphine complex [Cu(triphos)-
(PHPh2)][PF6] (12) by NaOSiMe3 to yield the phosphido
complex Cu(triphos)(PPh2) (22). Nucleophilic attack on the
substrate (benzyl halide or CH2Cl2) then yields the tertiary
phosphine complex [Cu(triphos)(PPh2CH2X)][PF6] (X = Ph
(16) or Cl (17)), and ligand substitution with PHPh2 regen-
erates 12 (Scheme 10).
Experimental Section
General Experimental Details. Unless otherwise noted, all
reactions and manipulations were performed in dry glassware
under a nitrogen atmosphere at ambient temperature in a dry-
box or using standard Schlenk techniques. Petroleum ether (bp
38-53 °C), CH2Cl2, ether, THF, and toluene were dried over
alumina columns similar to those described by Grubbs.35 NMR
spectra were recorded by using a Varian 300 or 500 MHz
spectrometer. 1H or 13C NMR chemical shifts are reported
versus Me4Si and were determined by reference to the residual
1H or 13C solvent peaks. 31P NMR chemical shifts are reported
versus H3PO4 (85%) used as an external reference. Coupling
constants are reported in hertz (Hz), as absolute values unless
noted otherwise. Unless indicated, peaks in NMR spectra are
singlets. IR spectra were recorded on KBr disks and are reported
in cm-1. Quantitative Technologies Incorporated provided
elemental analyses. Mass spectrometry was performed at the
University of Illinois. Unless otherwise noted, reagents were
from commercial suppliers; these compounds were prepared by
the literature methods: [Cu(NCMe)4][PF6] (1),36 PH2CH2Fc,37
PHMe(Is),38 Cu(dtbp)(Cl) (5) and Cu(dtbp)(OTf) (6),14 and
[K][B(ArF)4].39
This proposed mechanism is related to those suggested for
catalytic C-X (X = N, O, S) bond formation mediated by
nucleophilic Cu(I) amide, alkoxide, and thiolate complexes.31,6
In contrast, Cu-catalyzed aryl amination and etherification
may proceed via Cu(I)/Cu(III) cycles or other processes invol-
ving electron transfer,32 while little is known about the mechan-
ism of Cu-catalyzed phosphination of aryl halides.33
Development of the copper-catalyzed alkylations of di-
phenylphosphine reported here exploited the anticipated
nucleophilicity of a Cu(I) terminal phosphido intermediate
and rapid ligand substitution at the labile copper center. The
(29) Goj, L. A.; Blue, E. D.; Delp, S. A.; Gunnoe, T. B.; Cundari, T. R.;
Pierpont, A. W.; Petersen, J. L.; Boyle, P. D. Inorg. Chem. 2006, 45, 9032–
9045.
(30) On treatment of [Cu(XantPhos)(PPh2CH2Ph)][PF6] (19) with
PHPh2, the 31P NMR spectrum (CD2Cl2, 21 °C) was consistent with rapid
ligand substitution on the NMR time scale. Three broad peaks (δ -4.7,
-10.2, -24.7), assigned to PPh2CH2Ph, XantPhos, and PHPh2, respectively,
were observed.
(31) Gunnoe, T. B. Eur. J. Inorg. Chem. 2007, 1185–1203.
(32) (a) Jones, G. O.; Liu, P.; Houk, K. N.; Buchwald, S. L. J. Am. Chem.
Soc. 2010, 132, 6205–6213. (b) References 21b, 21c.
(33) (a) Gelman, D.; Jiang, L.; Buchwald, S. L. Org. Lett. 2003, 5, 2315–
2318. (b) Van Allen, D.; Venkataraman, D. J. Org. Chem. 2003, 68, 4590–4593.
(c) Tani, K.; Behenna, D. C.; McFadden, R. M.; Stoltz, B. M. Org. Lett. 2007, 9,
2529–2531. (d) Huang, C.; Tang, X.; Fu, H.; Jiang, Y.; Zhao, Y. J. Org. Chem.
2006, 71, 5020–5022.
(34) (a) Burk, M. J.; Harlow, R. L. Angew. Chem., Int. Ed. 1990, 29, 1462–
1464. (b) Burk, M. J.; Feaster, J. E.; Harlow, R. L. Tetrahedron: Asymmetry
1991, 2, 569–592. (c) Ward, T. R.; Venanzi, L. M.; Albinati, A.; Lianza, F.; Gerfin,
T.; Gramlich, V.; Ramos Tombo, G. M. Helv. Chim. Acta 1991, 74, 983–988.
(35) Pangborn, A. B.; Giardello, M. A.; Grubbs, R. H.; Rosen, R. K.;
Timmers, F. J. Organometallics 1996, 15, 1518–1520.
(36) Kubas, G. J. Inorg. Synth. 1990, 28, 68–70.

Article
[Cu(triphos)(NCMe)][PF6] (2). A white slurry of [Cu(NCMe)4]-
Inorganic Chemistry, Vol. 49, No. 17, 2010 7659
under vacuum. The residue was dissolved in 10 mL of a 10%
THF/petroleum ether solution. The solution was passed
through a silica plug, and the filtrate was pumped down under
vacuum to yield a white solid (61 mg, 0.22 mmol, 82%). In the
experiment with 1 mol % catalyst loading, the reaction was
complete after 2 h; 20% THF/petroleum ether was used for
workup, and the yield was 87%. A similar procedure was used
for the other entries in Table 2 involving benzyl bromide and
chloride (entries 1 and 4-17); see the Supporting Information
for details.
[PF6] (1; 250 mg, 0.671 mmol) in 5 mL of THF was treated with a
solution of triphos (419 mg, 0.671 mmol) in 2 mL of THF and
rapidly stirred overnight. The reaction mixture became somewhat
more clear and homogeneous, but still contained an off-white tint.
It was pumped down under vacuum. The residue was washed with
petroleum ether and then recrystallized from methylene chloride
layered with petroleum ether at -30 °C to give a white solid
(585 mg, 0.670 mmol, 99.8%). Crystals suitable for elemental
analysis and X-ray crystallography were obtained by dissolving
the white solid in a minimal amount of methylene chloride followed
by diffusion of petroleum ether vapors at -30 °C. Scale-up: Start-
ing with 1.789 g of [Cu(NCMe)4][PF6] gave 3.947 g of [Cu(triphos)-
(NCMe)][PF6] (94% yield).
Anal. Calcd. for C43H42P4NF6Cu: C, 59.08; H, 4.84; N, 1.60.
Found: C, 58.83; H, 4.89; N, 1.56. HRMS m/z calcd for C41H39-
P3Cu (M-NCMe)þ: 687.1561. Found: 687.1561. 31P{1H} NMR
(CD2Cl2): δ -20.6 (broad), -143.4 (septet, J = 711). 1H NMR
(CD2Cl2): δ 7.36 (br, 12H, Ar), 7.24 (t, J = 8, 6H, Ar), 7.15 (t, J =
8, 12H, Ar), 2.69 (br, 3H, CH3), 2.49 (br, 6H, CH2), 1.62 (br, 3H,
CH3). 13C{1H} NMR (CD2Cl2): δ 134.7 (m, Ar), 131.9 (m, Ar),
130.2 (Ar), 129.0 (m, Ar), 122.6 (NC), 39.1 (m, CH3), 36.4 (br,
MeC), 35.7 (br, CH2), 3.7 (NCCH3). 19F{1H} NMR (CD2Cl2):
δ -73.5 (d, J = 711).
[Cu(XantPhos)(NCMe)][PF6] (3). A slurry of [Cu(NCMe)4]-
[PF6] (1; 150 mg, 0.403 mmol) in 5 mL of THF was treated with a
slurry of XantPhos (233 mg, 0.403 mmol) in 5 mL of THF and
stirred for 2 h. The reaction mixture became homogeneous and
was pumped down under vacuum. The residue was washed with
petroleum ether (2 ꢀ 10 mL) and dried under vacuum. The crude
product was then dissolved in CH2Cl2, filtered through Celite,
and the resulting solution was layered with petroleum ether and
cooled to -30 °C yielding a white crystalline solid (280 mg, 0.338
mmol, 84%). Crystals suitable for elemental analysis and X-ray
crystallography were obtained by slow diffusion of petroleum
ether vapors into a methylene chloride solution at -30 °C.
Scale-up: 300 mg of [Cu(NCMe)4][PF6] gave 629 mg of [Cu-
(XantPhos)(NCMe)][PF6] (94% yield).
Anal. Calcd. for C41H35CuNOP3F6: C, 59.46; H, 4.26; N,
1.69. Found: C, 59.05; H, 4.28; N, 1.62. HRMS m/z calcd for
C39H32CuOP2 (M-NCMe)þ: 641.1224. Found: 641.1237. 31P-
{1H} NMR (CD2Cl2): δ -13.0 (XantPhos), -143.5 (septet, J =
711, PF6). 1H NMR (CD2Cl2): δ 7.68 (dd, J = 2, 8, 2H, Ph), 7.46
(t, J = 7, 4H, Ph), 7.36 (t, J = 8, 8H, Ph), 7.31-7.27 (m, 8H, Ph),
7.23 (t, J = 8, 2H, Ph), 6.72 (br m, 2H, Ph), 2.23 (3H, NCMe),
1.71 (6H, Me). 13C{1H} NMR (CD2Cl2): δ 154.5 (t, J = 6, Ph),
133.7 (t, J = 2, Ph), 133.5 (t, J = 8, Ph), 131.8 (Ph), 130.8 (Ph),
130.2 (t, J = 19, Ph), 129.3 (t, J = 5, Ph), 128.1 (Ph), 125.4 (t,
J = 2, Ph), 120.7 (Ph), 118.7 (t, J = 16, NCMe), 36.0 (CMe2),
28.5 (Me), 2.5 (NCMe).
Cu-Catalyzed Alkylation of CH2Cl2; Synthesis of PPh2CH2Cl
(7; Table 2, Entry 18). A solution of [Cu(triphos)(NCMe)][PF6]
(2; 23 mg, 0.027 mmol, 10 mol %) and PHPh2 (50 mg, 0.27 mmol)
in 2 mL of CH2Cl2 was added to a stirring solution of NaOSiMe3
(30 mg, 0.27 mmol) in 1 mL of CH2Cl2, resulting in a flash of
yellow color, which dissipated within a minute. After 25 min, all
of the PHPh2 had been consumed, according to 31P NMR moni-
toring. The solution was pumped down under vacuum, the
residue was dissolved in 10 mL of 10% THF/petroleum ether,
and this solution was passed through a silica plug. The filtrate
was pumped down, giving a clear oil (54 mg, 0.23 mmol, 86%).
NMR data for 7 matched literature reports.15
31P{1H} NMR (CDCl3): δ -7.9. H NMR (CDCl3): 7.52-
1
7.49 (m, 4H, Ph), 7.42-7.40 (m, 6H, Ph), 4.08 (d, J = 5, 2H,
CH2). 13C{1H} NMR (CDCl3): 135.6 (d, J = 12, Ph), 133.2 (d,
J = 19, Ph), 129.6 (Ph), 128.9 (d, J = 7, Ph), 41.0 (d, J = 27,
CH2).
Cu-Catalyzed Alkylation of PHPh2 with PhCHMe(Br);
Synthesis of PPh2CHMe(Ph) (8; Table 2, Entry 20). A solution
of (1-bromoethyl)benzene (37 μL, 0.27 mmol) in 1 mL of THF
was added to a stirred slurry of [Cu(triphos)(NCMe)][PF6] (2;
23 mg, 0.027 mmol) and PHPh2 (50 mg, 0.27 mmol) in 2 mL of
THF. A solution of NaOSiMe3 (30 mg, 0.27 mmol) in 2 mL of
THF was added with stirring, resulting in a bright yellow solu-
tion, which became milky white as precipitation occurred within
minutes. After 15 min, 31P NMR spectroscopy revealed full con-
version to a 33:1 mixture of PPh2CH(Me)Ph (8) and Ph2P-PPh2
(9). The reaction mixture was pumped down under vacuum and
the residue was extracted with a 10% THF/petroleum ether
solution. The extract was passed through a silica plug, and the
filtrate was pumped down, giving oily, white solid 8 (79 mg,
0.27 mmol, 100%) in 97% purity by 31P NMR spectroscopy.
The identity of the impurity (9) was confirmed by spiking the
product with Ph2P-PPh2 (Aldrich).
NMR spectra for 8 matched literature data.17 31P{1H} NMR
(CDCl3): δ 3.6 (97%, PhCHMePPh2), -13.9 (3%, Ph2P-PPh2).
1H NMR (CDCl3): δ 7.68-7.65 (m, 2H, Ph), 7.45-7.42 (m, 3H,
Ph), 7.24-7.10 (m, 10H, Ph), 3.57 (apparent quintet, J = 7.5,
1H, CH), 1.45 (m, 3H, CH3).
Cu-Catalyzed Reaction of PHPh2 with Dibromoethane; Synthe-
sis of Ph2P-PPh2 (9; Table 2, Entry 26). An NMR tube was
charged with a solution of [Cu(triphos)(NCMe)][PF6] (2; 23 mg,
0.027 mmol) in a few drops of CH2Cl2, a solution of PHPh2
(50 mg, 0.27 mmol) in 1 mL of CH2Cl2, and 1,2-dibromoethane
(12 μL, 0.13 mmol). This mixture was added to a stirred slurry of
NaOSiMe3 (30 mg, 0.27 mmol) in 1 mL of CH2Cl2, and the cata-
lysis was monitored by 31P NMR spectroscopy. After 15 min,
the reaction was complete, yielding Ph2P-PPh2 (9) and a small
amount of its oxide, Ph2P-PPh2(O);40 no Ph2PCH2Cl was
formed. The reaction mixture was pumped down under vacuum.
The residue was dissolved in 10 mL of a 10% THF/petroleum
ether solution and passed through a silica plug. The resulting
solution was pumped down under vacuum affording a white
solid (35 mg, 0.095 mmol, 71%), which contained Ph2P-PPh2
and a trace of Ph2P-P(O)Ph2, as observed by 31P NMR spec-
troscopy. The formation of 9 was confirmed by 31P and 1H
Cu-Catalyzed Alkylation of PHPh2 with PhCH2X (X =Br,
Cl); Synthesis of PPh2CH2Ph (4; Entries 2 and 3, Table 2).
[Cu(triphos)(NCMe)][PF6] (2; 23 mg, 0.027 mmol (10 mol %)
or 2.3 mg, 0.0027 mmol (1 mol %)) was dissolved in a few drops
of CH2Cl2 and treated with a solution of PHPh2 (50 mg, 0.27 mmol)
in 1 mL of CH2Cl2, then with dry, degassed benzyl bromide (32 μL,
0.27 mmol). This mixture was added to a rapidly stirring slurry of
NaOSiMe3 (30 mg, 0.27 mmol) in 1 mL of CH2Cl2. The reaction,
monitored by 31P NMR spectroscopy, was complete in less than
10 min, but was likely faster; a large amount of precipitate
formed within 1 min. The reaction mixture was pumped down
(37) (a) Pet, M. A.; Cain, M. F.; Hughes, R. P.; Glueck, D. S.; Golen,
J. A.; Rheingold, A. L. J. Organomet. Chem. 2009, 694, 2279–2289.
(b) Goodwin, N. J.; Henderson, W.; Nicholson, B. K. Chem. Commun. 1997,
31–32.
(38) Brauer, D. J.; Bitterer, F.; Dorrenbach, F.; Hessler, G.; Stelzer, O.;
Kruger, C.; Lutz, F. Z. Naturforsch. B 1996, 51, 1183–1196.
(39) Brookhart, M.; Grant, B.; Volpe, A. F., Jr. Organometallics 1992, 11,
3920–3922.
(40) Irvine, D. J.; Glidewell, C.; Cole-Hamilton, D. J.; Barnes, J. C.;
Howie, A. J. Chem. Soc., Dalton Trans. 1991, 1765–1772.

7660 Inorganic Chemistry, Vol. 49, No. 17, 2010
Cain et al.
NMR after spiking this material with an authentic sample of
Ph2P-PPh2 (Aldrich).
Anal. Calcd. for C53H50CuF6P5 1.2CH2Cl2: C, 58.06; H,
3
4.71. Found: C, 57.51; H, 4.75. One equivalent of CH2Cl2 per
copper complex was observed by X-ray crystallography. Inte-
gration of this sample’s H NMR spectrum indicated the pre-
[Cu(triphos)(PH2Ph)][PF6] (10). A stirring white slurry of
[Cu(triphos)(NCMe)][PF6] (2; 50 mg, 0.057 mmol) in 5 mL of
THF was treated with a solution of PH2Ph (6.3 mg, 0.057 mmol)
in 1 mL of THF resulting in a homogeneous solution with a pink
tint. The mixture was stirred for 3 h; it became significantly
darker. The reaction mixture was pumped down, and the residue
was washed with petroleum ether. The crude product was
recrystallized from methylene chloride layered with petroleum
ether at -30 °C to give a flaky gray-pink solid (52 mg, 0.055
mmol, 96%). White crystals suitable for elemental analysis were
obtained by dissolving the solid in THF followed by diffusion of
petroleum ether vapors at -30 °C. Scaleup: 200 mg of 2 gave the
product in quantitative yield.
1
sence of 1.2 molecules of CH2Cl2. HRMS m/z calcd for
C53H50CuP4 (Mþ): 873.2160. Found: 873.2160. 31P{1H} NMR
(CD2Cl2): δ -16.2 (broad), -143.5 (septet, J = 711). 31P{1H}
NMR (CD2Cl2, -70 °C): -16.0 (broad, triphos), -19.4 (broad,
PHPh2), -143.7 (septet, J = 711, PF6). 1H NMR (CD2Cl2): δ
7.71 (t, J = 7, 4H, Ar), 7.56 (t, J = 7, 2H, Ar), 7.41 (t, J = 7, 4H,
Ar), 7.25 (m, 6H, Ar), 7.04 (br, 24H, Ar), overlapping 6.99 (dq,
J = 381, 7, 1H, PH), 2.61 (br, 6H, CH2), 1.70 (br, 3H, CH3).
13C{1H} NMR (CD2Cl2): δ 134.1 (m, Ar), 133.5 (d, J = 13,
P-Ar), 131.9 (q, J = 5, Ar), 131.1 (d, J = 1, P-Ar), 130.6 (m,
P-Ar), 130.3 (Ar), 129.7 (d, J = 9, P-Ar), 129.1 (q, J = 3, Ar),
39.1 (q, J = 11, CH3), 36.2 (broad, overlapping CH3C-CH2).
IR (Nujol): 2917, 2726, 1963, 1483, 1480, 1465, 1434, 1378, 1095,
Anal. Calcd. for C47H46CuF6P5: C, 59.85; H, 4.92. Found: C,
59.76; H, 4.97. HRMS m/z calcd for C47H47CuP4 (MHþ):
798.1924. Found: 798.1965. 31P{1H} NMR (CD2Cl2): δ -16.4
(broad), -91.0 (very broad), -143.5 (septet, J = 711). 1H NMR
(CD2Cl2): δ 7.76 (m, 2H, Ar), 7.58 (t, J = 8, 1H, Ar), 7.47 (m,
2H, Ar), 7.25 (t, J = 7, 6H, Ar), 7.13 (broad, 12H, Ar), 7.08 (t,
J = 8, 12H, Ar), 5.98 (dq, J = 9, 313, 2H, PH2), 2.57 (broad, 6H,
CH2), 1.69 (q, J = 3, 3H, CH3). 13C{1H} NMR (CD2Cl2): δ
133.9 (m, Ar), 133.6 (d, J = 13, P-Ar), 131.8 (q, J = 5, Ar),
130.8 (d, J = 1, P-Ar), 130.4 (Ar), 129.8 (d, J = 10, P-Ar),
129.1 (q, J = 3, Ar), 123.9 (dq, J = 4, 38, P-Ar), 39.0 (q, J = 11,
CH3), 36.2 (m, CH3C), 35.9 (m, CH2). IR (Nujol): 2919, 2324
(P-H stretch), 1964, 1585, 1573, 1418, 1378, 1332, 1306, 1095,
876, 835, 761, 736, 700, 557 cm-1
.
[Cu(triphos)(PHEt2)][PF6] (13). A white slurry of [Cu(triphos)-
(NCMe)][PF6] (2; 50 mg, 0.057 mmol) was stirred in 5 mL of THF
and treated with a solution of PHEt2 (5.2 mg, 0.057 mmol) in 1 mL
of THF to give a clear solution. The mixture was stirred for 4 h,
pumped down under vacuum, and the residue was washed with
petroleum ether. The crude product was recrystallized from
methylene chloride layered with petroleum ether at -30 °C to
give an off-white/beige solid (53 mg, 0.057 mmol, 100%). Crystals
suitable for X-ray and elemental analyses were obtained by
dissolving this material in a minimal amount of methylene chlo-
ride followed by diffusion of petroleum ether vapors at -30 °C.
Scaleup: 200 mg of 2 gave the product in quantitative yield.
Anal. Calcd. for C45H50CuF6P5: C, 58.54; H, 5.46. Found: C,
58.13; H, 5.70. HRMS m/z calcd for C45H51CuP4 (MHþ):
778.2237. Found: 778.2274. 31P{1H} NMR (CD2Cl2): δ -16.7
(broad), -33.4 (broad), -143.5 (septet, J = 711). 1H NMR
(CD2Cl2): δ 7.27 (t, J = 7, 6H, Ar), 7.15-7.11 (br, 24H, Ar), 4.85
(dm, J = 300, 1H, PH), 2.55 (6H, CH2), 2.17 (m, 4H, PHCH2),
1.67 (q, J = 3, 3H, CH3), 1.36 (dt, J = 16, 8, 6H, CH3). 13C{1H}
NMR (CD2Cl2): δ 134.6 (m, Ar), 131.8 (q, J = 5, Ar), 130.3 (Ar),
129.1 (q, J = 3, Ar), 39.0 (m, CH3), 36.3 (m, CH3C), 36.0 (m,
CH2), 15.3 (dq, J = 4, 17, P-CH2), 11.6 (d, J = 2, PCH2CH3). IR
(Nujol): 3059, 2728, 2674, 2301 (P-H stretch), 1572, 1482, 1416,
1403, 1400, 1378, 1333, 1310, 1276, 1267, 1184, 1166, 1161, 1157,
1072, 1025, 999 cm-1
.
[Cu(triphos)(PH2CH2Fc)][PF6] (11). A white slurry of [Cu-
(triphos)(NCMe)][PF6] (2; 50 mg, 0.057 mmol) was stirred in
5 mL of THF and treated with a yellow solution of FcCH2PH2
(13.3 mg, 0.057 mmol) in less than 1 mL of THF; the mixture
became yellow-orange and homogeneous. After stirring over-
night, the reaction mixture was pumped down, and the residue
was washed with petroleum ether. The crude product was then
recrystallized from methylene chloride layered with petro-
leum ether at -30 °C to give a golden brown solid (61 mg,
0.057 mmol, 100%). X-ray quality crystals were grown by dis-
solving the solid in warm methanol and allowing it to slowly cool
at -30 °C. Scaleup: 200 mg of 2 gave the product in quantitative
yield.
1147, 1096, 1073, 1041, 1027, 1001, 970, 958, 922 cm-1
.
Anal. Calcd. for C52H52CuF6FeP5: C, 58.63; H, 4.92. Ana-
lyses were consistently low in carbon, for example: Found: C,
57.97; H, 4.79. HRMS m/z calcd for C52H53P4CuFe (MHþ):
920.1743. Found: 920.1766. 31P{1H} NMR (CD2Cl2): δ -16.6
(broad), -94.3 (very broad), -143.5 (septet, J = 711). 1H NMR
(CD2Cl2): δ 7.26 (t, J = 7, 6H, Ar), 7.17 (br, 12H, Ar), 7.12 (t,
J = 7, 12H, Ar), 4.92 (dq, J = 8, 305, 2H, PH2), 4.32 (br, 2H,
Fc), 4.22 (br, 7H, Fc), 3.33 (br, 2H, PCH2Fc), 2.52 (br, 6H,
CH2), 1.66 (br, 3H, CH3). 13C{1H} NMR (CD2Cl2): δ 134.1 (m,
Ar), 131.8 (m, Ar), 130.4 (Ar), 129.2 (m, Ar), 85.0 (Fc), 69.4 (Fc),
68.71-68.67 (m, Fc), 39.0 (apparent quartet, J = 11, CH3), 36.1
(m, C-CH3), 35.7 (m, MeC-CH2), 19.9 (dq, J = 5, 18, P-CH2Fc).
IR (Nujol): 2902, 2728, 2329 (P-H stretch), 1481, 1435, 1378,
[Cu(triphos)(PHCy2)][PF6] (14). A white slurry of [Cu(triphos)-
(NCMe)][PF6] (2; 50 mg, 0.057 mmol) in 5 mL of THF was treated
with a solution of PHCy2 (11.3 mg, 0.057 mmol) in 2 mL of THF
to give a clear solution. The mixture was stirred for 5 h and
pumped down under vacuum. The residue was washed with
petroleum ether (2 ꢀ 10 mL) and dried under vacuum. The crude
product was recrystallized from CH2Cl2 layered with petroleum
ether at -30 °C to give a white solid (56 mg, 0.054 mmol, 95%).
Crystals suitable for X-ray and elemental analysis were obtained
by dissolving the solid in THF followed by diffusion of petroleum
ether vapors at -30 °C. Scaleup: 200 mg of 2 gave a quantitative
yield of the product.
1104, 1095, 1025, 999, 923, 875, 736 cm-1
.
Anal. Calcd. for C53H62CuF6P5 0.5THF: C, 61.98; H, 6.37.
3
[Cu(triphos)(PHPh2)][PF6] (12). A white slurry of [Cu(triphos)-
(NCMe)][PF6] (2; 50 mg, 0.057 mmol) was stirred in 5 mL of THF
and treated with a solution of PHPh2 (10.7 mg, 0.057 mmol) in
1 mL of THF to give a clear solution. The mixture was stirred
overnight and was cloudy by morning. The solvent was pumped
down, and the residue was washed with petroleum ether. The
crude product was recrystallized from methylene chloride layered
with petroleum ether at -30 °C to give a white precipitate (56 mg,
0.055 mmol, 97%). X-ray quality crystals were obtained by
dissolving the white solid in a minimal amount of methylene chlo-
ride followed by diffusion of petroleum ether vapors at -30 °C.
Scale-up: 1.0 g of [Cu(triphos)(NCMe)][PF6] gave [Cu(triphos)-
(PHPh2)][PF6] in quantitative yield.
Found: C, 62.35; H, 6.32. One equivalent of THF per copper
complex was observed by X-ray crystallography. Before ele-
mental analysis, a sample was placed under vacuum; 1H NMR
integration then showed that 0.5 equiv of THF remained.
HRMS m/z calcd for C53H63CuP4 (MHþ): 886.3176. Found:
886.3135. 31P{1H} NMR (CD2Cl2): δ -7.7 (broad), -18.2
1
(broad), -143.5 (septet, J = 711). H NMR (CD2Cl2): δ 7.31
(t, J = 7, 6H, Ar), 7.15-7.10 (broad, 24H, Ar), 4.53 (dm, J =
293, 1H, PH), 2.59 (br, 6H, CH2), 2.20 (m, 2H, Cy), 1.98 (m, 4H,
Cy), 1.73-1.61 (overlapping m, 3H Me þ 6H Cy), 1.37 (m, 5H,
Cy), 1.25 (m, 5H, Cy). 13C{1H} NMR (CD2Cl2): δ 134.6 (m, Ar),
132.1 (q, J = 5, Ar), 130.3 (Ar), 129.1 (q, J = 3, Ar), 39.1 (q, J =
10, C-CH3), 36.7 (br d, J = 4, C-CH3), 35.9 (m, CH3C-CH2),

Article
Inorganic Chemistry, Vol. 49, No. 17, 2010 7661
32.6 (d, J = 4, Cy), 31.9 (d, J = 3, Cy), 31.8 (Cy), 27.0 (d, J = 10,
Cy), 26.8 (d, J = 12, Cy), 25.6 (Cy). IR (Nujol): 3052, 2728,
2670, 2314 (P-H stretch), 1585, 1573, 1483, 1462, 1450, 1418,
1403, 1378, 1334, 1310, 1269, 1229, 1183, 1173, 1160, 1096, 1073,
Anal. Calcd. for C54H51ClCuF6P5: C, 60.74; H, 4.81. Found:
C, 60.92; H, 4.40. HRMS m/z calcd. for C54H51P4ClCu (Mþ):
921.1926. Found: 921.1923. 31P{1H} NMR (CD2Cl2): δ 11.6
(Ph2PCH2Cl), -16.7 (triphos), -143.5 (septet, J = 711, PF6).
1H NMR (CD2Cl2): δ 7.60-7.56 (m, 6H, Ph), 7.37-7.33 (m, 4H,
Ph), 7.29 (t, J = 7.5, 6H, Ph), 7.05 (t, J = 7, 12H, Ph), 6.96 (br,
12H, Ph), 4.46 (2H, Ph2PCH2Cl), 2.64 (6H, CH2-triphos), 1.68
(3H, CH3). 13C{1H} NMR (CD2Cl2): δ 134.2 (m, Ph), 133.4 (d,
J = 13, Ph), 132.2 (dd, J = 5, 10, Ph), 131.6 (d, J = 1, Ph),
130.6-130.4 (m, Ph), 130.4 (Ph), 129.5 (d, J = 9, Ph), 129.2 (dd,
J = 3, 6, Ph), 40.6 (m, Ph2PCH2Cl), 39.1 (q, J = 11, CH3), 37.2
(m, CH3C), 36.0 (m, CH2-triphos).
1043, 1026, 1000, 964, 916, 875, 734 cm-1
.
[Cu(triphos)(PHMe(Is))][PF6] (15). A white slurry of [Cu-
(triphos)(NCMe)][PF6] (2; 200 mg, 0.23 mmol) in 5 mL of THF
was treated with a solution of PHMe(Is) (58 mg, 0.23 mmol) in
2 mL of THF, resulting in a clear solution. The mixture was stirred
for 3 h, pumped down under vacuum, and washed with petroleum
ether (2 ꢀ 10 mL). The crude product was recrystallized from met-
hylene chloride layered with petroleum ether at -30 °C to give a
white solid (243 mg, 0.22 mmol, 98%). Crystals suitable for ele-
mental analysis were obtained from a second recrystallization from
CH2Cl2/petroleum ether at -30 °C.
[Cu(XantPhos)(PHPh2)][PF6] (18). A slurry of [Cu(XantPhos)-
(NCMe)][PF6] (3; 200 mg, 0.24 mmol) in 2 mL of THF was treated
witha solution of PHPh2 (45 mg, 0.24 mmol) in 2 mL of THF. The
resulting clear solution was stirred for 1 h and then pumped down
under vacuum. The residue was washed with petroleum ether (2 ꢀ
10 mL) and then recrystallized from methylene chloride layered
with petroleum ether at -30 °C affording a white solid (220 mg,
0.226 mmol, 93%). 31P and1H NMR spectra at room temperature
were very broad, consistent with a dynamic process; on cooling
to-70 °C, the peaks remained broad, and more of them appeared.
Anal. Calcd. for C51H43CuF6OP4: C, 62.93; H, 4.45. Found:
C, 62.99; H, 4.44. HRMS m/z calcd for C51H43CuOP3 (Mþ):
827.1823. Found: 827.1816. 31P{1H} NMR (CD2Cl2): δ -9.4
(broad with a significant shoulder), -28.2 (broad), -143.5 (septet,
J = 711, PF6). 1H NMR (CD2Cl2): δ 7.71 (d, J = 7), 7.41 (broad),
7.24 (v. broad triplet with broad shoulder, J= 8), 6.68 (broad), 5.94
(v. broad d, J = 327, PH), 1.95, 1.72 (broad). The broadness of
these peaks precluded meaningful integration.
Anal. Calcd. for C57H66CuP5F6: C, 63.18; H, 6.14. Found: C,
62.95; H, 5.73. HRMS m/z calcd for C57H66CuP4 (Mþ):
937.3412. Found: 937.3411. 31P{1H} NMR (CD2Cl2): δ -17.8
(broad, triphos), -81.6 (broad, PHMe(Is)), -143.5 (septet, J =
711, PF6). 1H NMR (CDCl3): δ 7.22-7.20 (m, 6H, Ph), 7.17 (br,
2H, Ph), 7.12 (br, 12H, Ph), 7.04-7.00 (br m, 12H, Ph), 6.38
(dm, J = 313, 1H, PH), 3.49 (br, 2H, CH), 2.97 (septet, J = 7,
1H, CH), 2.49 (br, 6H, CH2), 2.07 (m, 3H, PHMe), 1.62 (br, 3H,
CH3), 1.33-1.28 (br m, 18H, CH3(Is)). 13C{1H} NMR (CD2Cl2):
δ 151.9 (Ar), 134.5 (m, Ar), 132.0 (m, Ar), 130.4 (Ar), 129.2 (m,
Ar), 124.8 (dm, J = 32, Ar), 122.9 (br, Ar), 39.1 (m, CH3), 36.4
(br, MeC), 35.9(m, CH2), 34.6 (i-Pr), 33.4(br, i-Pr), 25.0 (br, i-Pr),
23.9 (d, J = 3, i-Pr), 12.5 (dm, J = 16, P-Me). IR (Nujol): 2953,
2923, 2854, 2397 (P-H stretch), 1463, 836, 694 cm-1
.
[Cu(triphos)(PPh2CH2Ph)][PF6] (16). A slurry of [Cu(triphos)-
(NCMe)][PF6] (2; 100 mg, 0.114 mmol) in 5 mL of THF was
treated with a solution of benzyldiphenylphosphine (4; 32 mg,
0.11 mmol) in 1 mL of THF and rapidly stirred. Within 1 min, the
solution became clear and homogeneous, but it was stirred for 1 h
toensure completeconversion. The reactionmixture was pumped
down and washed with petroleum ether (2 ꢀ 10 mL). The residue
was recrystallized from methylene chloride layered with petro-
leum ether at -30 °C to give a white, fluffy solid (130 mg, 0.114
mmol, 100%). Crystals suitable for X-ray analysis were obtained
by diffusion of petroleum ether vapors into a solution of 16 in
CH2Cl2 over the course of several days at -30 °C, leading to the
formation ofa large white crystal. Whenthe solution was pipetted
away from the crystal, it desolvated, but when the solution was
transferred to a separate vial, it suddenly formed several white
snowflake-like crystals of X-ray quality.
Anal. Calcd. for C60H56CuP5F6: C, 64.95; H, 5.09. Found: C,
64.49; H, 4.84. HRMS m/z calcd for C60H56CuP4 (Mþ): 963.2629.
Found: 963.2604. 31P{1H} NMR (CD2Cl2): δ 9.2 (PPh2CH2Ph),
-17.6 (triphos), -143.5 (septet, J= 712, PF6). 1HNMR(CD2Cl2):
δ 7.54-7.49 (m, 6H, Ph), 7.31-7.24 (m, 10H, Ph), 7.07 (t, J = 7.5,
13H, Ph), 6.95 (br, 14H, Ph), 6.64 (d, J = 8.5, 2H, Ph), 3.99 (d, J =
3.5, 2H, CH2), 2.62 (6H, CH2), 1.67 (d, J = 3.5, 3H, CH3). 13C{1H}
NMR (CD2Cl2): δ 134.5 (m, Ph), 134.0 (d, J = 14, Ph), 133.5 (d,
J = 7, Ph), 132.2 (q, J = 6, Ph), 131.9 (br, Ph), 131.0 (d, J = 2, Ph),
130.3 (Ph), 130.2 (d, J = 4, Ph), 129.2 (m, Ph), 129.1 (Ph), 128.5 (d,
J = 2, Ph), 127.3 (d, J = 2, Ph), 39.1 (m, CH3), 37.9 (m, CH2-P),
37.2 (br, MeC), 35.9 (m, CH2-triphos).
[Cu(XantPhos)(PPh2CH2Ph)][PF6] (19). A solution of [Cu-
(XantPhos)(NCMe)][PF6] (3; 50 mg, 0.060 mmol) in 1 mL of
CH2Cl2 was treated with PPh2CH2Ph (4;17 mg, 0.060 mmol) in
1 mL of CH2Cl2 and stirred for 1 h. The reaction mixture was
pumped down under vacuum. The residue was washed with
petroleum ether (2 ꢀ 10 mL), redissolved in 3 mL of methylene
chloride and passed through Celite. The transparent CH2Cl2
solution was then layered with petroleum ether and cooled
to -30 °C resulting in the precipitation of a white crystalline
solid (56 mg, 0.053 mmol, 88%). X-ray quality crystals were
obtained by slow diffusion of petroleum vapors into a methylene
chloride solution at -30 °C.
Anal. Calcd. for C58H49CuF6OP4: C, 65.51; H, 4.64. Found:
C, 64.97; H, 4.51. HRMS m/z calcd for C58H49CuOP3 (Mþ):
917.2292. Found: 917.2322. 31P{1H} NMR (CD2Cl2): δ -0.6 (t,
J = 98, PPh2CH2Ph), -12.2 (d, J = 98, XantPhos), -143.6
(septet, J = 711, PF6). 1H NMR (CD2Cl2): δ 7.76 (dd, J = 1.5,
7.5, 2H, Ph), 7.38 (t, J = 7.5, 4H, Ph), 7.33 (m, 2H, Ph), 7.26 (t,
J = 7.5, 2H, Ph), 7.17 (t, J = 7.5, Ph), 7.14-7.12 (m, Ph),
7.10-7.06 (m, 17H total for 7.17-7.06, Ph), 6.97-6.96 (m, 4H,
Ph), 6.94-6.90 (m, 8H, Ph), 6.71-6.69 (m, 2H, Ph), 3.47 (d, J =
7.5, 2H, Ph), 1.75 (Me, 6H). 13C{1H} NMR (CD2Cl2): δ 154.2 (t,
J = 6, Ph), 135.0 (Ph), 133.5 (t, J = 2, Ph), 133.1 (t, J = 8, Ph),
132.3 (d, J = 4, Ph), 132.2 (Ph), 130.9 (Ph), 129.6-129.2
(overlapping peaks, Ph), 129.2 (Ph), 129.1 (d, J = 6, Ph),
128.9 (Ph), 127.6 (d, J = 3, Ph), 125.6 (m, Ph), 118.5 (t, J =
16, Ph), 36.0 (CMe2), 33.0 (d, J = 16, CH2), 29.0 (CMe2).
[Cu(dtbp)(PHPh2)(OTf)] (20a). A solution of Cu(dtbp)(OTf)
(6; 200 mg, 0.40 mmol) in 2 mL of CH2Cl2 was treated with a
solution of PHPh2 (74 mg, 0.40 mmol) in 2 mL of CH2Cl2. The
yellow-orange homogeneous solution was stirred for 1 h, then
pumped down under vacuum. The resulting residue was washed
with petroleum ether (2 ꢀ 10 mL). The crude product was re-
crystallizedfrommethylenechloride layeredwithpetroleum ether
at -30 °C, giving a yellow solid (252 mg, 0.365 mmol, 92%).
Anal. Calcd. for C33H35CuF3N2O3PS: C, 57.34; H, 5.10; N,
4.05. Found: C, 56.94; H, 5.00; N, 3.99. HRMS m/z calcd for
C32H35N2PCu (Mþ): 541.1834. Found: 541.1833. 31P{1H}
NMR (CD2Cl2): δ -24.9. 1H NMR (CD2Cl2): δ 8.65 (d, J = 8.5,
[Cu(triphos)(PPh2CH2Cl)][PF6] (17). A slurry of [Cu(triphos)-
(NCMe)][PF6] (2; 155 mg, 0.178 mmol) in 1 mL of THF was
treated with a solution of PPh2CH2Cl (50 mg, 0.21 mmol) in 2 mL
of THF resulting in a clear, homogeneous solution, which was stir-
red for 1 h, then pumped down under vacuum. The residue was
washed with petroleum ether (2 ꢀ 10 mL) to remove the excess
phosphine and redissolved in 3 mL of CH2Cl2. The solution was
then passed through Celite and recrystallized by layering the filt-
ratewithpetroleum ether at -30 °C yielding a whitesolid (187 mg,
0.175 mmol, 99%). Crystals suitable for X-ray diffraction were
obtained by dissolving the sample in warm ethanol and allowing it
to slowly cool at -30 °C.

7662 Inorganic Chemistry, Vol. 49, No. 17, 2010
Cain et al.
2H, dtbp), 8.12 (d, J = 8.5, 2H, dtbp), 8.10 (2H, dtbp), 7.53-7.50
(br m, 2H, P-Ph), 7.42-7.36 (br m, 8H, P-Ph), 6.40 (d, J = 341, 1H,
PH), 1.60(18H, t-Bu). 13C{1H} NMR (CD2Cl2):δ170.3 (Ar), 143.0
(Ar), 140.0 (Ar), 133.7 (d, J = 12.5, Ar), 131.5 (Ar), 129.7 (d, J =
10, Ar), 128.0 (Ar), 126.7 (Ar), 123.2 (Ar), 38.5 (C(CH3)3), 30.6
(C(CH3)3). 19F{1H} NMR (CD2Cl2): δ -79.2.
[Cu(dtbp)(PHPh2)(Cl)] (20b). A solution of Cu(dtbp)(Cl) (5;
100 mg, 0.26 mmol) in 2 mL of CH2Cl2 was treated with a solution
of PHPh2 (48 mg, 0.26 mmol) in 1 mL of CH2Cl2. The reaction
mixture remained orange and was stirred for 1 h. The solution was
pumped down under vacuum, washed with petroleum ether (2 ꢀ
10 mL), and recrystallized from CH2Cl2 layered with petroleum
ether at -30 °C to afford an orange solid (94 mg, 0.16 mmol, 62%).
31P{1H} NMR (CDCl3): δ -33.2. 1H NMR (CDCl3): δ 8.30
(d, J = 8.5, 2H, dtbp), 7.94 (d, J = 8, 2H, dtbp), 7.80 (2H, dtbp),
led to some decomposition, with widespread decomposition
occurring at -30 °C. Some of the decomposition products iden-
tified by 31P NMR spectroscopy at -30 °C were triphos (δ -26.1)
and PHPh2 (δ -40.3), along with several other unidentified
peaks at δ -15.8, -18.6, -20.6, and -34.4. At room tempera-
ture, the 31P NMR spectrum showed no signals corresponding
to 22; however, triphos (δ -25.0) and PHPh2 (δ -39.9) were
observed along with several unidentified peaks at -20.9, -23.9,
-26.8, and -34.0 ppm.
NMR data for Cu(triphos)(PPh2) (22):31P{1H} NMR (d8-THF,
-50 °C): δ -23.4 (d, J = 27, triphos), -30.3 (q, J = 27, PPh2). 1H
NMR (d8-THF, -50 °C): δ 7.86 (br, 4H, BArF), 7.69 (br, 8H,
BArF), 7.30 (br, 12H, Ph), 7.07 (t, J = 7, 8H, Ph), 6.91 (t, J = 7,
12H, Ph), 6.81 (br, 8H, Ph), 2.43 (br, 6H, CH2), 1.53 (br, 3H, CH3).
Ligand Substitution: Equilibrium Between [Cu(triphos)-
(PPh2CH2Ph)][PF6] (16), PHPh2, [Cu(triphos)(PHPh2)][PF6] (12)
and PPh2CH2Ph (4). A solution of [Cu(triphos)(PHPh2)][PF6]
(12; 50 mg, 0.049 mmol) in less than 1 mL of CD2Cl2 was treated
with a solution of PPh2CH2Ph (4; 14 mg, 0.049 mmol) in 1 mL of
CD2Cl2 and transferred to an NMR tube. The 31P NMR
spectrum was monitored over the course of several hours, but
did not change from the initial spectrum taken less than 15 min
after the addition of PPh2CH2Ph. An identical spectrum re-
sulted from treatment of [Cu(triphos)(PPh2CH2Ph)][PF6] (16;
50 mg, 0.045 mmol) with 1 equiv of PHPh2 in CD2Cl2. From
integration of the broad 31P NMR signals observed in both
experiments, Keq = 8(2), favoring coordination of PHPh2.
Ligand Substitution: Equilibrium Between [Cu(triphos)-
(PPh2CH2Cl)][PF6] (17), PHPh2, [Cu(triphos)(PHPh2)][PF6] (12)
and PPh2CH2Cl (7). A solution of [Cu(triphos)(PHPh2)][PF6]
(12; 52 mg, 0.051 mmol) in less than 1 mL of CD2Cl2 was treated
with a solution of PPh2CH2Cl (7; 12 mg, 0.051 mol) in 1 mL of
CD2Cl2. The resulting homogeneous solution was transferred to
an NMR tube, then analyzed by 31P NMR spectroscopy after
30 min. An identical spectrum resulted when the mixture was
generated by treatment of [Cu(triphos)(PPh2CH2Cl)][PF6] (17;
57 mg, 0.054 mmol) with PHPh2 (10 mg, 0.054 mmol) in CD2Cl2.
From integration of the broad 31P NMR signals observed in
both experiments, Keq = 7(2), favoring coordination of PHPh2.
DFT Computations. Gas phase structures were optimized using
the hybrid B3LYP functional41 and the triple-ζ LACV3P**þþ
basis set,42 which uses extended core potentials on heavy atoms
and a 6-311G**þþ basis for other atoms, as implemented in the
7.56 (m, 4H, Ph), 7.34 (m, 2H, Ph), 7.25 (m, 4H, Ph), 5.85 (d, JPH
313, 1H, PH), 1.83 (18H, t-Bu).
=
Stoichiometric Deprotonation/Alkylation of [Cu(triphos)-
(PHPh2)][PF6] (12): PhCH2Br. An NMR tube was charged with
a solution of [Cu(triphos)(PHPh2)][PF6] (12; 50 mg, 0.049 mmol)
in 1 mL of CD2Cl2. Neat PhCH2Br (6 μL, 0.05 mmol) was added
via syringe, followed by a solution of NaOSiMe3 (6 mg, 0.05 mmol),
and the reaction was monitored by 1H and 31P NMR spectroscopy.
After 15 min, complete conversion to [Cu(triphos)(PPh2CH2Ph)]-
[PF6] (16) was observed.
Stoichiometric Deprotonation/Alkylation of [Cu(triphos)-
(PHPh2)][PF6] (12): CD2Cl2. A solution of [Cu(triphos)(PHPh2)]-
[PF6] (12; 50 mg, 0.049 mmol) in less than 1 mL of CD2Cl2 was
placed in an NMR tube, which was capped with a rubber sep-
tum and sealed with Parafilm. The NMR tube was then cooled
to -78 °C in a dry ice/acetone bath and injected via syringe with a
solution of NaOSiMe3 (55 mg, 0.49 mmol) in less than 1 mL of
CD2Cl2. The reaction mixture was kept at -78 °C and analyzed by
low temperature NMR spectroscopy at 10 °C intervals beginning
at -70 °C. Formation of [Cu(triphos)(PPh2CD2Cl)][PF6] (17-D2)
was first observed starting at -40 °C. On warming to room temp-
erature, this complex was formed as the major product and identi-
1
fied by comparison of its 31P and H NMR spectra to indepen-
dently prepared 17 (as expected, for 17-D2, the CH2 1H NMR
signal was not observed). Some decomposition products were also
observed by 31P NMR (δ -8.8, -19.8, -21.3).
Ion-Exchange; Preparation of 12-B(ArF)4. A solution of [Cu-
(triphos)(PHPh2)][PF6] (12; 100 mg, 0.098 mmol) in CH2Cl2 was
treated with a slurry of K[B(ArF)4] (89 mg, 0.098 mmol) in
approximately 25 mL of water. The mixture was transferred to a
separatory funnel and vigorously shaken. The layers were dis-
tinct, but the interface was very cloudy even after long periods
of equilibration. The organic layer was separated, washed (3 ꢀ
25 mL) with water, and dried over MgSO4. The solution was
passed over a frit, and the filtrate was pumped down yielding an
oily solid (22 mg, 0.012 mmol, 13%). The low yield appeared to
arise from poor separation of 12-B(ArF)4 from the aqueous
layer. The 31P{1H} NMR (CDCl3) spectrum (δ -16.5) confirmed
that no PF6; was present.
€
Jaguar suite of programs (Jaguar, versions 7.0-7.5, Schrodinger,
LLC, New York, NY: 2007-2009). All computed structures were
confirmed as energy minima by calculating the vibrational fre-
quencies by second derivative analytic methods, and confirming
the absence of imaginary frequencies. Thermodynamic quantities
were calculated assuming an ideal gas, and are zero point energy
corrected.
Acknowledgment. D.S.G. thanks the National Science
Foundation for support. R.P.H. also thanks the National
Science Foundation.
1H NMR (CDCl3): δ 7.73 (br, 8H, BArF), 7.64-7.60 (over-
lapping doublets, J = 11, 4H, Ar), 7.52 (br, 4H, BArF), 7.49
(overlapping triplet, J = 7, 2H, Ar), 7.34 (m, 4H, Ar), 7.19 (t,
J = 7, 6H, Ar), 6.98-6.92 (br m, 24H, Ar), 6.52 (dq, J = 7, 316,
1H, PH), 2.52 (br, 6H, CH2), 1.26 (3H, CH3).
Supporting Information Available: Additional experimental
details, details of the X-ray crystallographic studies, including
crystallographic information files (CIF), additional NMR spec-
tra, and details of the DFT calculations. This material is
available free of charge via the Internet at http://pubs.acs.org.
Deprotonation of 12-B(ArF)4; Generation of Cu(triphos)-
(PPh2) (22). A solution of [Cu(triphos)(PHPh2)][B(ArF)4] (12-
B(ArF)4; 21 mg, 0.12 mmol) in less than 1 mL of d8-THF was
placed in an NMR tube, which was capped with a rubber
septum, and sealed with Parafilm. The NMR tube was cooled
to -78 °C in a dry ice/acetone bath and injected via syringe with
a solution of NaOSiMe3 (14 mg, 0.12 mmol) in 1 mL of d8-THF.
The reaction mixture was kept at -78 °C and analyzed by low
temperature NMR spectroscopy. Complex 22 was observed at
-70 and -50 °C by 31P NMR spectroscopy. Warming to -40 °C
(41) (a) Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785–789.
(b) Stephens, P. J.; Devlin, F. J.; Chabalowski, C. F.; Frisch, M. J. J. Phys. Chem.
1994, 98, 11623–11627. (c) Becke, A. D. J. Chem. Phys. 1993, 98, 5648–5652.
(d) Becke, A. D. J. Chem. Phys. 1993, 98, 1372–1377.
(42) (a) Dunning, T. H.; Hay, P. J. In Modern Theoretical Chemistry, Vol.
4: Applications of Electronic Structure Theory.; Schaefer, H. F., III, Ed.;
Plenum: New York, 1977. (b) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985,
82, 270–283. (c) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 299–310.
(d) Wadt, W. R.; Hay, P. J. J. Chem. Phys. 1985, 82, 284–298.