Copper-Phosphido Intermediates in Cu(IPr)-Catalyzed Synthesis of 1-Phosphapyracenes via Tandem Alkylation/Arylation of Primary Phosphines

Article abstract of DOI:10.1021/acs.organomet.8b00198

Tandem alkylation/arylation of primary phosphines PH₂R (R = Ph, Cy, Fc, FcCH₂; Fc = ferrocenyl) with 5-bromo-6-chloromethylacenaphthene (1) and 2 equiv of NaOSiMe₃ using the catalyst precursor Cu(IPr)(Cl) gave a series of 1-phosphapyracenes (R-PyraPhos, 2a-d), which were isolated as borane adducts 3a-d. Similar reactions of the chiral air-stable primary phosphines PH₂Ar? (Ar? = (S)-binaphthyl (4), (R)-MeO-binaphthyl (5)) to yield 2e,f and 3e,f were diastereoselective (dr = 2:1 and 1.2:1, respectively), and chromatography gave a highly enriched sample of one diastereomer of 3f. The mechanism of catalysis was investigated by NMR monitoring and independent syntheses of potential intermediates. The phosphido complexes Cu(IPr)(PHAr′) (Ar′ = Ph, (R)-MeO-binaphthyl) were generated in equilibrium mixtures, along with Me₃SiOH, from Cu(IPr)(OSiMe₃) and PH₂Ar′. They reacted with benzyl chloride 1 to yield Cu(IPr)(Cl) and the secondary phosphines PHAr′(CH₂Ar) (Ar = Br-acenaphthyl); addition of NaOSiMe₃ yielded PyraPhos derivatives 2a,f. Deprotonation of the cations [Cu(IPr)(PHAr′CH₂Ar)][OTf] (Ar′ = Ph, (R)-MeO-binaphthyl) was investigated as a route to the secondary phosphido complexes Cu(IPr)(PAr′CH₂Ar) (13). We propose that C-Br oxidative addition in the Cu(I)-phosphido intermediates 13 followed by P-C reductive elimination from Cu(III)-phosphido complexes forms the PyraPhos ring, with diastereoselection arising from rapid pyramidal inversion of Cu-phosphido groups.

Full text of DOI:10.1021/acs.organomet.8b00198

Article
Cite This: Organometallics XXXX, XXX, XXX−XXX
Copper−Phosphido Intermediates in Cu(IPr)-Catalyzed Synthesis of
1‑Phosphapyracenes via Tandem Alkylation/Arylation of Primary
Phosphines
Ge Wang,† Sarah K. Gibbons,† David S. Glueck,*,† Connor Sibbald,‡ James T. Fleming,‡ Lee J. Higham,‡
and Arnold L. Rheingold§
†Department of Chemistry, Dartmouth College, 6128 Burke Laboratory, Hanover, New Hampshire 03755, United States
‡School of Chemistry, Bedson Building, Newcastle University, Newcastle upon Tyne NE1 7RU, U.K.
§Department of Chemistry, University of California, San Diego, 9500 Gilman Drive, La Jolla, California 92093, United States
S
* Supporting Information
ABSTRACT: Tandem alkylation/arylation of primary phosphines
PH2R (R = Ph, Cy, Fc, FcCH2; Fc = ferrocenyl) with 5-bromo-6-
chloromethylacenaphthene (1) and 2 equiv of NaOSiMe3 using
the catalyst precursor Cu(IPr)(Cl) gave a series of 1-phosphapyr-
acenes (R-PyraPhos, 2a−d), which were isolated as borane adducts
3a−d. Similar reactions of the chiral air-stable primary phos-
phines PH2Ar* (Ar* = (S)-binaphthyl (4), (R)-MeO-binaphthyl
(5)) to yield 2e,f and 3e,f were diastereoselective (dr = 2:1 and
1.2:1, respectively), and chromatography gave a highly enriched
sample of one diastereomer of 3f. The mechanism of catalysis was
investigated by NMR monitoring and independent syntheses of
potential intermediates. The phosphido complexes Cu(IPr)(PHAr′)
(Ar′ = Ph, (R)-MeO-binaphthyl) were generated in equilibrium mixtures, along with Me3SiOH, from Cu(IPr)(OSiMe3) and PH2Ar′.
They reacted with benzyl chloride 1 to yield Cu(IPr)(Cl) and the secondary phosphines PHAr′(CH2Ar) (Ar = Br-acenaphthyl);
addition of NaOSiMe3 yielded PyraPhos derivatives 2a,f. Deprotonation of the cations [Cu(IPr)(PHAr′CH2Ar)][OTf] (Ar′ =
Ph, (R)-MeO-binaphthyl) was investigated as a route to the secondary phosphido complexes Cu(IPr)(PAr′CH2Ar) (13).
We propose that C−Br oxidative addition in the Cu(I)−phosphido intermediates 13 followed by P−C reductive elimination
from Cu(III)−phosphido complexes forms the PyraPhos ring, with diastereoselection arising from rapid pyramidal inversion of
Cu−phosphido groups.
phospholanes are privileged ligands in asymmetric catalysis.10
We report here increased scope for this reaction with a variety of
primary phosphine substrates, including chiral ones, which led to
isolation of a highly diastereomerically enriched P-stereogenic
INTRODUCTION
■
Metal-catalyzed formation of phosphorus−carbon bonds is
frequently used to prepare organophosphorus compounds.1
Many of these reactions use precious metals such as palladium
and rhodium, but catalysis by base metals would be cheaper and
PyraPhos derivative. Using a copper catalyst precursor bearing
an N-heterocyclic carbene (NHC) ligand, as in B−D, enabled
identification of potential intermediates in the catalytic cycle by
NMR monitoring and independent synthesis. These observa-
tions provided information on the properties of the rare ter-
minal copper−phosphido group, including the first P-stereogenic
chiral example, and its role in catalysis.
more sustainable.2 For example, several methods for copper-
catalyzed P−C bond formation by coupling of secondary
phosphines with electrophiles have been reported.3 However,
little is known about the key proposed intermediates, mono-
meric terminal copper phosphido complexes.4 Chart 1 shows
the only directly observed catalytically relevant terminal Cu−
phosphido intermediates, A and B,3m,5 and the only isolated
complexes, C and D.6−8 The P-stereogenic derivatives
LnCu(PRR′), potential intermediates in catalytic asymmetric
synthesis of chiral phosphines, have not yet been reported.
We recently described tandem alkylation/arylation of phenyl-
phosphine with 5-bromo-6-chloromethylacenaphthene (1), the
base NaOSiMe3, and the catalyst precursor CuCl to yield a
P-stereogenic 1-phosphapyracene (Ph-PyraPhos (2a); Scheme 1).9
This type of reaction might be useful if the P configuration of
the chiral product could be controlled by the catalyst, because
RESULTS AND DISCUSSION
■
Copper-Catalyzed Alkylation/Arylation of Primary
Phosphines. Synthesis and Structure of Phosphapyr-
acene−Boranes 3a−f. Several Cu(I) catalyst precursors, includ-
ing CuCl and [Cu(NCMe)4][PF6], were screened for the title
reaction in THF with 2 equiv of the base NaOSiMe3.
Received: April 5, 2018
© XXXX American Chemical Society
A
DOI: 10.1021/acs.organomet.8b00198
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Chart 1. Terminal Copper−Phosphido Complexes
Scheme 2. Copper-Catalyzed Synthesis of PyraPhos−BH3
Derivatives 3a−f (Fc = Ferrocenyl)
The new PyraPhos-boranes 3b−f were characterized by
spectroscopy, by elemental analyses, and by X-ray crystallog-
raphy (Figures 1−5; more details are included in the
Scheme 1. Copper-Catalyzed Synthesis of
1-Phosphapyracene 2a
Commercially available Cu(IPr)(Cl) was active for the syn-
thesis of 1-phosphapyracenes 2a−f from a variety of primary
phosphines, including alkyl- and aryl-substituted derivatives, as
well as the unusual air-stable chiral primary phosphines 4 and 5
(Scheme 2).11 The intermediate secondary phosphines 6,
observed in the reaction mixtures and identified by indepen-
dent synthesis of 6a,f, were converted to phosphapyracenes.9
These reactions were slow, typically proceeding in 2−3 days at
50 °C with 5 mol % catalyst loading. After the catalytic reac-
tions were complete, the reaction mixtures were treated with
BH3·THF, and the resulting air-stable PyraPhos−boranes 3a−f
were isolated and purified by column chromatography and
recrystallization. In most cases (except for 3e), the reactions
were chemoselective, with few byproducts formed and isolated
yields ranging from 34% to 95%. The reactions with 4 and 5
were diastereoselective (2:1 and 1.2:1 diastereomer ratios,
respectively), and further separation by chromatography gave a
highly enriched sample of one diastereomer of the (R)-MeO-
binaphthyl derivative 3f (∼99:1 dr). In contrast, when
Au(IPr)(Cl) was used as a catalyst precursor with PH2Ph,
the secondary phosphine 6a was observed but, instead of cata-
lytic turnover to yield phosphapyracene 2a, a mixture of prod-
ucts formed.
Figure 1. ORTEP diagram of Cy-PyraPhos(BH3) (3b) with 30%
probability ellipsoids. Hydrogen atoms are omitted for clarity.
Figure 2. ORTEP diagram of Fc-PyraPhos(BH3) (3c) with 30%
probability ellipsoids. Hydrogen atoms are omitted for clarity.
Supporting Information). Changing the P substituent had little
effect on the structures (Table 1). The PyraPhos rings were
B
DOI: 10.1021/acs.organomet.8b00198
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Figure 3. ORTEP diagram of FcCH2-PyraPhos(BH3) (3d) with 30%
probability ellipsoids. Hydrogen atoms are omitted for clarity.
Figure 5. ORTEP diagram of (R)-MeO-Binaphthyl-PyraPhos(BH3)
(3f) with 30% probability ellipsoids. Hydrogen atoms are omitted for
clarity. Two of the four independent molecules (A and D) in the unit
cell are shown. The crystal contained four independent molecules, a
mixture of two diastereomers having the same binaphthyl axial chirality
but differing in the P configuration. Two of them (A and B) were SP,
and the other two (C and D) were RP. The last two molecules,
especially C, were disordered.
Ph-PyraPhos complex 9 before catalytic turnover. With gold,
the catalytic formation18 of secondary phosphine 6a from PH2Ph
and benzyl chloride 1 suggested that problems with the cycli-
zation step prevented gold-catalyzed synthesis of Ph-PyraPhos.
Deprotonation of 10, which might result in formation of 11,
was intended to probe this process and enable comparison of
analogous gold and copper phosphido intermediates.
In experiments more directly related to catalysis, treatment of
the precursor Cu(IPr)(Cl) with NaOSiMe3 in THF gave the
known Cu(IPr)(OSiMe3) (Scheme 4).19 Upon addition of PH2Ph
or the chiral primary phosphine 5, the reaction mixtures turned
yellow. 31P{1H} and 1H NMR monitoring showed that the pri-
mary phosphido complexes Cu(IPr)(PHAr′) (Ar′ = Ph (12a),
(R)-MeO-binaphthyl (12f)) were formed, in apparent equilibrium
mixtures with the starting materials and Me3SiOH (Scheme 4
and Table 2 (entries 3 and 12); Keq was about 1 in both cases).
As expected, treatment of cation 7 with NaOSiMe3 gave a simi-
lar mixture.
Figure 4. ORTEP diagram of (S)-binaphthyl-(RP)-PyraPhos(BH3)
(3e) with 30% probability ellipsoids. Hydrogen atoms are omitted for
clarity.
almost planar, with small deviations of the peri substituents
(P and C) from the planes of the acenaphthene rings. The
geometry at phosphorus was distorted tetrahedral, constrained
by the PyraPhos ring, in which the C−P−C bond angles ranged
from 92.9(3)° (R = Fc) to 95.3° (R = (S)-binaphthyl). All other
bond angles at phosphorus were larger, up to 120.61(14)°
(C(Ar)−P−B for R = Cy).
Mechanism of Copper-Catalyzed Phosphapyracene
Synthesis. The mechanism of catalysis was investigated by stoi-
chiometric reactions and by independent synthesis or generation
of proposed intermediates, with a focus on comparison of the
achiral and chiral substrates PH2Ph (a) and PH2(R)-MeO-
Binaphthyl (f), whose steric and electronic differences make
PH2Ph a pyrophoric liquid, while the chiral phosphine is an air-
stable solid.12
Cationic complexes 7−9 containing primary, secondary, or
tertiary arylphosphines were prepared by reaction of Cu(IPr)-
(OTf)13 with PH2Ph, the secondary phosphines 6a,f, or
Ph-Pyraphos (2a),14 and the analogous gold complexes 10
and 11 were synthesized from Au(IPr)(OTf) and 6a or 2a
(Scheme 3).15 31P NMR data for these cations showed the
expected coordination chemical shift16 and increase in JPH upon
phosphine binding (Table 2, entries 1, 2, 4−6, 8−10, 13, and 14).17
We planned to deprotonate copper complexes 7 and 8 to
generate phosphido intermediates, which could eventually yield
Although two diastereomers of 12f are possible, only one
1
31P{1H} NMR signal and one set of H NMR peaks were
observed at room temperature (Table 2, entry 12). When the
sample was cooled to −85 °C in the NMR probe, the chemical
1
shifts of selected H NMR signals of 12f changed significantly,
but there was no evidence in either 1H or 31P NMR spectra for
two isomers. These observations are consistent either with the
selective formation of one diastereomer or with interconversion
between the diastereomers which remains fast on the NMR
time scale even at low temperature. This P epimerization could
occur by pyramidal inversion, as observed previously for metal
phosphido complexes.17b,20
Addition of the benzyl chloride 1 to the Cu(IPr)(OSiMe3)/
PH2Ph mixture gave secondary phosphine 6a and Cu(IPr)(Cl),
plus a small amount of Ph-PyraPhos (2a) (route A, Scheme 4).
When 12a was generated, along with NaOTf, from cation 7,
reaction with 1 instead gave the secondary phosphine complex
C
DOI: 10.1021/acs.organomet.8b00198
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Table 1. Selected Bond Lengths (Å) and Angles (deg) for the PyraPhos−Boranes 3a−f
R
a
b
Ph (3a)
Cy (3b)
Fc (3c)
CH2Fc (3d)
H-binaphthyl (3e)
MeO-binaphthyl (3f)
P−CH2
P−C(Ar)
P−C(R)
P−B
CH2−P−B
C(Ar)−P−CH2
C(Ar)−P−C(R)
C(Ar)−P−B
C(R)−P−CH2
C(R)−P−B
1.851(8)
1.815(7)
1.845(3)
1.90(4)
1.869(3)
1.914(6)
1.799(6)
1.791(6)
1.905(6)
115.2(3)
92.9(3)
1.8542(16)
1.8066(17)
1.8325(16)
1.9157(19)
114.66(8)
94.85(7)
1.874(5)
1.786(4)
1.839(4)
1.907(4)
109.8(3)
95.3(2)
1.889(8)
1.786(7)
1.814(7)
1.908(8)
111.0(3)
94.7(4)
1.811(3)
1.835(3)
1.920(3)
115.7(14)
94.9(3)
112.95(14)
94.34(13)
106.80(12)
120.61(14)
107.62(13)
112.60(14)
105.3(2)
115.9(14)
110.2(3)
113.0(12)
106.4(3)
120.2(3)
103.4(3)
115.4(3)
109.04(8)
113.43(8)
105.38(7)
117.03(8)
108.11(17)
117.87(19)
110.35(19)
113.9(2)
113.8(3)
114.5(4)
107.3(3)
113.5(4)
a
b
Average values for the two enantiomeric forms in the disordered crystal; data from ref 9. Average values for three of the four independent
molecules (data for the disordered fourth molecule was omitted).
Scheme 3. Synthesis of Cationic M(IPr) Primary, Secondary,
and Tertiary Phosphine Complexes (M = Cu, Au; [M] =
M(IPr))
Table 2. 31P NMR Data for Phosphines and Related IPr−
Copper and −Gold Phosphine and Phosphido Complexes
a
(Ar = Br-acenaphthyl, Ar* = (R)-MeO-binaphthyl)
entry
compound
δ(31P)
JPH (Hz)
1
2
3
4
5
6
7
8
9
PH2Ph
−125.2
198
314
188
216
323
392
[Cu(IPr)(PH2Ph)][OTf] (7)
Cu(IPr)(PHPh) (12a)
−103.5
−124 (br)
−37.8
−32.4
2.9
PHPh(CH2Ar) (6a)
[Cu(IPr)(PHPh(CH2Ar))][OTf] (8a)
[Au(IPr)(PHPh(CH2Ar))][OTf] (10)
Cu(IPr)(PPh(CH2Ar)) (13a)
Ph-Pyraphos (2a)
−59
−6.4
−0.9
[Cu(IPr)(Ph-Pyraphos)][OTf] (9)
10 [Au(IPr)(Ph-Pyraphos)][OTf] (11)
11 PH2Ar* (5)
12 Cu(IPr)(PHAr*) (12f)
13 PHAr*(CH2Ar) (6f)
36.7
−129.5
−124.7
−39.3
−42.4
−35.5
−38.6
200
209
211
208
289
325
14 [Cu(IPr)(PHAr*(CH2Ar))][OTf] (8f)
8a-OTf (route B, Scheme 4). When the mixture from route A
was treated with a second equivalent of NaOSiMe3, 6a was
slowly converted to 2a over 2 days. In a similar experiment with
12f and 1, the secondary phosphine 6f was immediately formed
as a 7:1 mixture of diastereomers, along with Cu(IPr)(Cl)
(Scheme 4). This ratio changed to 1:1.4 over several days, pre-
sumably via proton-transfer processes common in epimeriza-
tion of phosphines containing P−H bonds.21 If 12f had been
present as a single diastereomer, its reaction with 1 should have
given a single diastereomer of 6f. Although it is possible that
rapid epimerization of 6f could have given the observed 7:1 ratio,
that is not consistent with the continued slow epimerization.
Therefore, it is likely that P epimerization in copper−phosphido
complex 12f is rapid on the NMR time scale (for additional
discussion, see also Scheme 8 and accompanying text below).22
The potential intermediate secondary phosphido complexes
Cu(IPr)(PAr′(CH2Ar)) (Ar′ = Ph (13a), (R)-MeO-binaphthyl
(13f)) were not observed in these experiments. Attempts to gen-
erate 13a,f independently via deprotonation of the coordinated
15 intermediates from NaN(SiMe3)2 + [Cu(IPr)
−9.1
(PHAr*(CH2Ar))][OTf] (8f)
−15.6
a
In THF or THF-d8, chemical shift reference 85% H3PO4.
secondary phosphines in cations 8a,f led to different outcomes,
depending on the base and the phosphine (Scheme 5 and Table 2,
entries 7 and 15). With 8a, excess NaOSiMe3 (4 equiv) gave the
secondary phosphine 6a immediately, plus some Ph-PyraPhos
(2a), followed by complete conversion to Ph-PyraPhos. The initial
results were similar with 8f and NaOSiMe3, but conversion of
6f to Ar*-PyraPhos 2f was much slower.
In contrast, addition of NaN(SiMe3)2 (1.3 equiv) to 8a gave
rise to a mixture containing 8a, cationic Ph-PyraPhos complex 9a,
and several unidentified intermediates. One of these (31P NMR
δ −59, with no P−H coupling), which disappeared as more 9a
formed over 1 h,23 was also formed in a cleaner reaction of 8a
with CsOH·H2O (ca. 3 equiv), which led to 9a over 22 h.
On the basis of its generation from 8a with two different bases
D
DOI: 10.1021/acs.organomet.8b00198
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
slowing P epimerization at low temperature, we treated 8f with
NaN(SiMe3)2 at −78 °C to give a mixture containing mostly
the secondary phosphine 6f (1.3:1 mixture of diastereomers)
and small amounts of Ar*-PyraPhos 2f (1:1) and two uniden-
tified intermediates (1:1, Scheme 5). When the mixture was
warmed to room temperature, the intermediates disappeared
and additional PyraPhos 2f was formed; more slowly, 6f was
converted to 2f. The 31P NMR chemical shifts of the interme-
diates (δ −9.1 and −15.6, Table 2, entry 15) were quite dif-
ferent from that assigned to 13a (δ −59, Table 2, entry 7) in
the analogous reaction of P-Ph cation 8a with NaN(SiMe3)2,
suggesting that this material was not the target phosphido
complex 13f.25 We presume that the steric and electronic prop-
erties of the P-aryl substituents control the differences in reac-
tivity of P-Ph and P-Ar* cations 8a,f.12
Scheme 4. Stoichiometric Reactions of Cu(IPr)(Cl) with
NaOSiMe3, PH2Ar′, and the Benzyl Chloride 1 ([Cu] =
Cu(IPr); Ar′ = Ph (a), (R)-MeO-binaphthyl (f); Ar =
Br-acenaphthyl)
These observations are consistent with equilibria favoring
Cu(IPr)(OSiMe3) and secondary phosphines 6a,f over sec-
ondary phosphido complexes 13a,f and Me3SiOH, while 13a
was observable with a different base. The formation in these
experiments of Ph-PyraPhos 2a or its cationic copper complex
9a, which may undergo exchange, also seems to depend on the
base and conditions; treatment of 9a with [NOct4][Br] gave 2a
and Cu(IPr)(Br).26 Replacing bromide with Ph-PyraPhos was
also possible, as treatment of a mixture of Cu(IPr)(Br) and the
phosphine with NH4PF6 gave 9-PF6 (Scheme 5).
Consistent with the lack of activity of Au(IPr)(Cl) in
catalysis, treatment of the gold secondary phosphine cation 10
with NaOSiMe3 gave secondary phosphine 6a, along with two
other unidentified products which lacked P−H bonds, but
neither Ph-PyraPhos (2a) nor its cationic gold complex 11
were formed (Scheme 6).
Scheme 5. Reactions of Cationic Secondary Phosphine
Complexes 8a,f with Bases (Ar′ = Ph (a) or Ar* = (R)-MeO-
binaphthyl (f))
Scheme 6. Reaction of Gold Cation 10 with NaOSiMe3 To
Give Secondary Phosphine 6a and Not Ph-PyraPhos 2a or
Its Gold Complex 11
On the basis of these observations, we propose the mecha-
nism shown in Scheme 7 for the copper-catalyzed tandem
alkylation/arylation. Treatment of the catalyst precursor Cu(IPr)-
(Cl) with NaOSiMe3 yields the silanolate complex Cu(IPr)-
(OSiMe3), which reacts with a primary phosphine to give an
equilibrium mixture containing primary phosphido complex 12.
Nucleophilic attack of 12 on the benzyl chloride 1 forms sec-
ondary phosphine 6 and Cu(IPr)(Cl). Reaction of this mixture
with NaOSiMe3 generates phosphido intermediate 13, pre-
sent in an unfavorable equilibrium with Cu(IPr)(OSiMe3), 6,
and Me3SiOH. Oxidative addition of the C−Br bond in the
and its decomposition to yield 9a, we propose that this inter-
mediate is phosphido complex 13a (Scheme 5).24
In hopes of observing both diastereomers of the analogous
secondary phosphido complex Cu(IPr)(PAr*CH2Ar) (13f) by
E
DOI: 10.1021/acs.organomet.8b00198
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Scheme 7. Proposed Mechanism of Copper-Catalyzed Formation of PyraPhos Derivatives from Primary Phosphines ([Cu] =
Cu(IPr), Ar = Br-acenaphthyl)
pendant aryl group forms the Cu(III) intermediate 14,27 whose
reductive elimination yields Cu(IPr)(Br) and PyraPhos 2.9
Neither of the proposed Cu−phosphido complexes 13 and 14
could be observed under catalytic conditions, which is con-
sistent either with rate-determining formation of 13 followed
by fast oxidative addition/reductive elimination or with rate-
determining oxidative addition or reductive elimination, where
the concentration of 13 is low in the disfavored equilibrium
with Cu(IPr)(OSiMe3). Finally, bromide/NaOSiMe3 meta-
thesis regenerates the copper−silanolate intermediate, which
can react with a primary or secondary phosphine to continue
the cycle.28
Scheme 8. Asymmetric Induction via Pyramidal Inversion
and Cyclization in Diastereomeric Copper−Phosphido
Complexes (Ar* = (S)-binaphthyl or (R)-MeO-binaphthyl)
Proposed Origin of Asymmetric Induction in Catalysis
with Chiral Primary Phosphines 4 and 5. The proposed
mechanism is also consistent with formation of diastereomeri-
cally enriched PyraPhos derivatives 3e,f from chiral substrates 4
and 5. The chiral copper−phosphido complexes analogous to
12−14 (Scheme 7) would all be expected to exist as mixtures
of two diastereomers, differing in P configuration, while the
binaphthyl configuration is fixed. Interconversion between these
diastereomers should be fast because of the rapid P-inversion in
metal phosphido complexes, consistent with the observations on
12f.20,22 As in related systems, if this step is faster than P−C bond
formation, the product ratio is controlled by Curtin−Hammett
kinetics, depending on the equilibrium and rate constants
shown in Scheme 8 for intermediate 14f, as an example.29
favoring bonding of hard anionic donors (silanolate, halide)
over soft neutral phosphines. This should limit potential
product inhibition but may slow the rate of catalysis by limiting
the concentration of key copper−phosphido complexes in unfa-
vorable equilibria. The use of earth-abundant copper instead of
the precious metal platinum previously used for such catalytic
processes is promising for future applications.30 In particular,
the diastereoselective synthesis of binaphthyl-PyraPhos deriv-
atives 2e,f and NMR evidence for rapid pyramidal inversion in
copper phosphido complex 12f suggest that using a chiral NHC
ligand on copper to develop asymmetric catalysis for the synthe-
sis of enantiomerically enriched PyraPhos derivatives may be
possible.
CONCLUSIONS
■
We have shown that a well-defined Cu(I)−NHC catalyst pre-
cursor mediates the tandem alkylation/arylation of primary
phosphines to yield the PyraPhos ring and obtained mecha-
nistic information on the role of copper−phosphido complexes,
including the first P-stereogenic one, in this transformation.
Our observations suggest that the [Cu(IPr)]+ fragment is hard,
F
DOI: 10.1021/acs.organomet.8b00198
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
overlapped, ace-CH2 + benzyl-CH2), 1.93−1.88 (m, 1H), 1.88−1.76
(m, 3H), 1.76−1.71 (m, 1H), 1.69−1.65 (m, 1H), 1.41−1.32 (m, 1H),
1.27−1.05 (m, 4H), 1.05−0.50 (m, broad, 3H, BH3). 13C{1H} NMR
(CDCl3): δ 148.2 (d, J = 2, Ar, quat), 143.3 (Ar, quat), 138.5 (d, J =
13, Ar, quat), 138.2 (d, J = 8, Ar, quat), 132.6 (Ar, quat), 129.7 (d, J =
9, Ar), 126.1 (d, J = 52, Ar, quat), 124.7 (d, J = 7, Ar), 121.0 (d, J =
10, Ar), 120.8 (Ar), 36.5 (d, J = 29, Cy CH), 31.6 (d, J = 1, ace-CH2),
30.9 (ace-CH2), 28.6 (d, J = 37, benzyl-CH2), 26.44 (d, J = 13,
Cy CH2), 26.39 (Cy CH2), 26.34 (d, J = 11, Cy CH2), 26.1 (d, J = 2,
Cy CH2), 25.7 (d, J = 1, Cy CH2).
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 glovebox or using
standard Schlenk techniques. Petroleum ether (bp 38−53 °C) or
pentane, CH2Cl2, ether, THF, and toluene were dried over alumina
columns similar to those described by Grubbs.31 NMR spectra were
recorded with 500 or 600 MHz spectrometers. 1H or 13C NMR chemi-
cal shifts are reported vs Me4Si and were determined by reference to
1
the residual H or 13C solvent peaks. 31P NMR chemical shifts are
Fc-PyraPhos(BH3) (3c). An orange solution of ferrocenylphos-
phine (65 mg, 0.30 mmol) in 2 mL of THF was added to a solution of
Cu(IPr)(Cl) (7 mg, 0.014 mmol, 5 mol %) in 2 mL of THF. A solu-
tion of NaOSiMe3 (74 mg, 0.66 mmol, 2.2 equiv) in 2 mL of THF was
then added, followed by a solution of the benzyl chloride 1 (84 mg,
0.30 mmol, 1 equiv) in 2 mL of THF. Part of the orange reaction mix-
ture (about 1/10) was transferred to an NMR tube and sealed under
N2. The rest of the solution was transferred to a 50 mL Schlenk flask
and put under positive N2 pressure. The NMR tube and the Schlenk
flask were both heated at 50 °C. After 19 h, the 31P NMR spectrum of
the reaction mixture in the NMR tube showed the formation of the
intermediate secondary phosphine (δ −57.6 (81%)) and a small amount
(14%) of the desired tertiary phosphine product (−17.0 ppm). The
reaction was complete in 42 h. Under positive N2 pressure, 0.1 mL of a
BH3·THF solution (1 M in THF, 0.1 mmol) was added dropwise to the
NMR tube and another 0.5 mL of the BH3·THF solution (0.5 mmol)
was added to the reaction mixture in the Schlenk flask. The 31P NMR
spectrum confirmed the formation of the desired pyraphos−borane
adduct.
The reaction mixtures in the NMR tube and Schlenk flask were
combined, and 9 mL of 1 M aqueous NH4Cl was added. The mixture
was then extracted with CH2Cl2 (10 × 3 mL). The CH2Cl2 layer was
washed with 15 mL of water and then dried with MgSO4. Removing
the solvent under reduced pressure gave 120 mg of an orange solid.
Purification of this crude product with silica gel chromatography using
15/1 pentane/ethyl acetate was not successful, as the product seemed
to be quite insoluble in the eluent and blocked the column. Another
trial with 1/1 pentane/CH2Cl2 as eluent gave several portions of the
product with different purities. Each portion was recrystallized by
vapor diffusion of pentane into a CH2Cl2 solution to give pure orange
solid (63 mg total (53% yield)), which was also suitable for X-ray
crystallography.
Anal. Calcd for C23H22BFeP: C, 69.75; H, 5.60. Found: C, 69.52;
H, 5.29. HRMS: m/z calcd for C23H21BFeP (M − H)+, 395.0823,
found, 395.0835; calcd for C23H19FeP (M − BH3)+, 382.0574, found,
382.0569. 31P{1H} NMR (CDCl3): δ 32.3 (apparent d, J = 63).
1H NMR (CDCl3): δ 7.77 (dd, J = 7, 7, 1H), 7.41 (dm, J = 7, 1H), 7.39
(br, apparent AB pattern, JAB = 7, 1H), 7.32 (br, apparent AB pattern,
JAB = 7, 1H), 4.40−4.38 (m, 1H, Fc), 4.36 (5H, Fc), 4.35−4.33
(m, 2H, Fc), 4.17−4.14 (m, 1H, Fc), 3.82 (AB, JAB = 17, 1H, benzyl-
CH2), 3.68 (ABX, JAB = 17, JAX = 10, 1H, benzyl-CH2), 3.51−3.44
(m, 4H, ace-CH2), 1.50−0.80 (m, broad, 3H, BH3). 13C{1H} NMR
(CDCl3): δ 148.3 (d, J = 2, Ar, quat), 143.3 (Ar, quat), 138.4 (d, J = 8, Ar,
quat), 137.9 (d, J = 15, Ar, quat), 132.2 (Ar, quat), 129.0 (d, J = 10, Ar),
128.1 (d, J = 57, Ar, quat), 124.8 (d, J = 7, Ar), 121.1 (d, J = 10, Ar),
120.8 (Ar), 72.1 (d, J = 11, Fc), 71.8 (d, J = 7, Fc), 71.5 (Fc, quat),
71.2 (d, J = 8, Fc), 70.6 (d, J = 10, Fc), 69.7 (5C, Fc C5H5), 34.6
(d, J = 42, benzyl-CH2), 31.6 (ace-CH2), 31.0 (ace-CH2).
FcCH2-PyraPhos(BH3) (3d). An orange solution of the primary
phosphine (79 mg, 0.34 mmol) in 2 mL of THF was added to a solu-
tion of Cu(IPr)(Cl) (8 mg, 0.016 mmol, 5 mol %) in 2 mL of THF.
A solution of NaOSiMe3 (84 mg, 0.75 mmol, 2.2 equiv) in 2 mL of
THF was then added, followed by a solution of the benzyl chloride 1
(96 mg, 0.34 mmol, 1 equiv) in 2 mL of THF. Part of the orange
reaction mixture (about 1/10) was transferred to an NMR tube and
sealed under N2. The rest of the solution was transferred to a 50 mL
Schlenk flask and put under positive N2 pressure. The NMR tube and
the Schlenk flask were both heated at 50 °C. After 19 h, the 31P NMR
spectrum of the reaction mixture in the NMR tube showed the for-
mation of the intermediate secondary phosphine (δ −50.0 (81%))
reported vs H3PO4 (85%) used as an external reference. Coupling
constants are reported in Hz, as absolute values. Unless indicated,
peaks in NMR spectra are singlets. Instead of the expected quartet
resonances due to P−B coupling in the 31P{1H} NMR spectra of the
phosphine−boranes 3, poorly resolved multiplets or apparent doublets
were observed.32 Quantitative Technologies Incorporated/Intertek
Pharmaceutical Services (Whitehouse, NJ) provided elemental ana-
lyses. Mass spectrometry was performed at the University of Illinois.
Compound purity was established by a combination of elemental
analyses and NMR spectroscopy (see the Supporting Information for
NMR spectra). Reagents were obtained from commercial suppliers,
except for [Cu(NCMe)4][PF6],33 FcCH2PH2,17a,34 the chiral binaph-
thylphosphines,11 secondary phosphine 6a and Ph-PyraPhos 2a,9
Cu(IPr)(OTf),13 Au(IPr)(OTf),15 and Pt(dppe)(Me)(Cl),35 which
were prepared by literature procedures.
Ph-PyraPhos(BH3) (3a). We previously reported the complete
characterization of 3a, which was prepared using a CuCl catalyst pre-
cursor.9 Cu(IPr)(Cl) was also active. A solution of PH2Ph (6 mg,
0.05 mmol) in 0.2 mL of THF was added to a solution of Cu(IPr)(Cl)
(2 mg, 0.004 mmol, 8 mol %) in 0.2 mL of THF. A solution of the
benzyl chloride 1 (15 mg, 0.05 mmol) in 0.2 mL of THF was added,
followed by a solution of NaOSiMe3 (14 mg, 0.12 mmol, 2.2 equiv) in
0.2 mL of THF. The mixture turned light yellow immediately and was
transferred to a NMR tube. After 2 h, the 31P NMR spectrum showed
that the intermediate secondary phosphine 6a (80%) was the major
species, together with trace amounts of Ph-PyraPhos (2a; 3%)
and unreacted PH2Ph (15%). The reaction to form 2a was complete
after 3 days at room temperature, according to 31P NMR spectroscopy.
Cy-PyraPhos(BH3) (3b). A solution of the primary phosphine
(19 mg, 0.16 mmol) in 0.3 mL of THF was added to a solution of
Cu(IPr)(Cl) (4 mg, 0.008 mmol, 5 mol %) in 0.2 mL of THF. A solu-
tion of NaOSiMe3 (40 mg, 0.36 mmol, 2.2 equiv) in 0.4 mL of
THF was then added, followed by a solution of the benzyl chloride 1
(45 mg, 0.16 mmol, 1 equiv) in 0.4 mL of THF. The reaction mixture
was sealed in an NMR tube under N2 and heated at 50 °C. After 21 h,
the 31P NMR spectrum of the reaction mixture showed the formation
of the intermediate secondary phosphine (δ −36.3, 30%) and the
tertiary phosphine product (δ 1.0, 55%). The reaction was complete in
40 h. Under positive N2 pressure, 0.32 mL of a BH3·THF solution
(1 M in THF, 0.32 mmol, 2 equiv) was added dropwise to the NMR
tube. The 31P NMR spectrum confirmed the formation of the
PyraPhos−borane adduct (δ 46.4, apparent d, J = 62).
A 5 mL portion of 1 M aqueous NH4Cl and 5 mL of ether were
added to the reaction mixture. The aqueous layer was separated and
extracted again with 5 mL of ether. The organic layers were combined
and dried with MgSO4. Removing the solvent under reduced pressure
gave a sticky light brown solid. Purification of this crude product
with silica gel chromatography using 15/1 pentane/ethyl acetate as
eluent gave 23 mg of a white solid with approximately 7% impurities.
Slow evaporation of the column eluent gave X-ray-quality crystals.
Recrystallization by vapor diffusion of pentane into a CH2Cl2 solution
at room temperature gave a white solid (18 mg, 38% yield) with approxi-
mately 1% impurities (31P NMR δ 65.2), which was presumably the
phosphine oxide Cy-Pyraphos(O).
HRMS: m/z calcd for C19H21BP (M − 3H)+, 291.1474; found,
291.1469. 31P{1H} NMR (CDCl3): δ 44.8 (apparent d, J = 63). 1H NMR
(CDCl3): δ 7.71 (dd, J = 7, 7, 1H), 7.40 (dm, J = 7, 1H), 7.37 (apparent
AB pattern, JAB = 7, 1H), 7.30 (apparent AB pattern, JAB = 7, 1H), 3.54
(ABX, JAB = 17, JAX = 7, 1H, benzyl-CH2), 3.50−3.42 (m, 5H,
G
DOI: 10.1021/acs.organomet.8b00198
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
and a small amount (10%) of the desired tertiary phosphine product
(−7.2 ppm). The reaction was complete in 3 days. Under positive N2
pressure, 0.1 mL of a BH3·THF solution (1 M in THF, 0.1 mmol) was
added dropwise to the NMR tube and another 0.6 mL of the BH3·
THF solution (0.6 mmol) was added to the reaction mixture in the
Schlenk flask. The 31P NMR spectrum confirmed the formation of the
desired PyraPhos−borane adduct (δ 40.3).
The reaction mixtures in the NMR tube and Schlenk flask were
combined, and 10 mL of 1 M aqueous NH4Cl was added. The mixture
was extracted with ether (10 × 3 mL), and the organic layer was dried
with MgSO4. Removing the solvent under reduced pressure gave an
orange solid. Purification of this crude product with silica chroma-
tography using 1/1 pentane/CH2Cl2 as eluent was not successful, giving
73 mg of solid with impurities. Recrystallization by vapor diffusion of
pentane into a CH2Cl2 solution at room temperature gave a pure
orange solid (48 mg, 34% yield), which was also suitable for X-ray
crystallography.
Anal. Calcd for C24H24BFeP: C, 70.29; H, 5.90. Found: C, 69.86;
H, 5.76. Although these results are outside the range viewed as estab-
lishing analytical purity, they are provided to illustrate the best values
obtained to date; please see the Supporting Information for NMR
spectra. HRMS: m/z calcd for C24H24BFeP (M)+, 410.1058, found,
410.1063. 31P{1H} NMR (CDCl3): δ 38.3 (apparent d, J = 57). 1H NMR
(CDCl3): δ 7.57 (dd, J = 7, 7, 1H), 7.36 (dm, J = 7, 1H), 7.28 (apparent
AB pattern, JAB = 7, 1H), 7.25 (apparent AB pattern, JAB = 7, 1H), 4.04
(5H, Fc), 4.01−3.99 (m, 2H, Fc), 3.91 (q, J = 2, 1H, Fc), 3.79−3.78
(m, 1H, Fc), 3.48−3.41 (m, 6H, overlapped, ace-CH2 + benzyl-CH2),
3.08 (ABX, JAB = 14, JAX = 9, 1H, Fc-CH2), 3.03 (ABX, JAB = 14, JBX = 9,
1H, Fc-CH2), 1.21−0.50 (m, broad, 3H, BH3). 13C{1H} NMR
(CDCl3): δ 148.4 (d, J = 2, Ar, quat), 143.2 (Ar, quat), 138.2 (d, J =
14, Ar, quat), 138.1 (d, J = 8, Ar, quat), 132.0 (Ar, quat), 129.4 (d, J =
9, Ar), 126.7 (d, J = 53, Ar, quat), 124.7 (d, J = 7, Ar), 120.9 (d, J =
10, Ar), 120.7 (Ar), 78.7 (d, J = 2, Fc, quat), 69.5 (d, J = 1, Fc), 69.0
(d, J = 2, Fc), 68.8 (5C, C5H5), 68.1 (Fc), 67.8 (Fc), 31.6 (ace-CH2),
31.1 (d, J = 26, Fc-CH2), 30.9 (ace-CH2), 29.7 (d, J = 37, benzyl-CH2).
binaphthyl-PyraPhos(BH3) (3e). A yellow solution of (S)-
binaphthylphosphine (30 mg, 0.10 mmol) in 0.5 mL of THF was
added to a suspension of [Cu(NCMe)4][PF6] (4 mg, 0.011 mmol,
10 mol %) in 0.5 mL of THF. A solution of NaOSiMe3 (26 mg,
0.23 mmol, 2.2 equiv) in 0.5 mL of THF was added; the mixture
turned from yellow to red immediately. A solution of the benzyl
chloride 1 (30 mg, 0.11 mmol, 1 equiv) in 0.5 mL of THF was then
added. The red reaction mixture was stirred at room temperature and
monitored by 31P NMR spectroscopy. After 1 day, the 31P NMR spec-
trum showed the formation of the intermediate secondary phosphine
(81%, two diastereomers in a 1:1 ratio: δ −43.6, −44.2) and a small
amount (4%) of the desired tertiary phosphine product (δ −9.7,
−10.6). The reaction was complete in 3 days and gave the tertiary
phosphine as two diastereomers in 2:1 ratio: δ −14.8, −14.0. Under
positive N2 pressure, 0.2 mL of a BH3·THF solution (1 M in THF,
0.2 mmol, 2 equiv) was added dropwise with a syringe to the reaction
mixture. The 31P NMR spectrum confirmed the formation of the
PyraPhos−borane adduct (δ 37.0).
crystals used for X-ray analysis; see the Supporting Information for NMR
spectra.
A reaction with Cu(IPr)(Cl) as catalyst precursor was also carried
out on a small scale and monitored by 31P NMR spectroscopy.
A yellow solution of the chiral primary phosphine (15 mg, 0.05 mmol)
in 0.2 mL of THF was added to a solution of Cu(IPr)(Cl) (2 mg,
0.004 mmol, 8 mol %) in 0.2 mL of THF. A solution of NaOSiMe3
(13 mg, 0.12 mmol, 2.2 equiv) in 0.2 mL of THF was added, followed
by a solution of the benzyl chloride 1 (15 mg, 0.05 mmol) in 0.2 mL of
THF. The mixture was transferred to a NMR tube and heated to
50 °C. The reaction, monitored by 31P NMR spectroscopy, was done
in 2 days, with a dr of 1.9:1 (δ −14.8, −13.9, respectively). Several
other byproducts were formed, as the 31P NMR spectrum showed
extra peaks at 12.2 ppm (39%) and −56.6 ppm (8%). These byproducts
were also observed in approximately the same percentage when
[Cu(NCMe)4][PF6] was used as catalyst precursor. Addition of BH3·
THF gave the PyraPhos−borane adduct (31P NMR: δ 37.1 ppm). The
reactivity and selectivity of catalyst precursors Cu(IPr)(Cl) and
[Cu(NCMe)4][PF6] was about the same, but reaction of the cheaper
nitrile complex precursor occurred at room temperature instead of
50 °C for Cu(IPr)(Cl), which was more convenient.
Anal. Calcd for C33H26BP: C, 85.36; H, 5.64. Found: C, 83.16;
H, 5.33. HRMS: m/z calcd for C33H25BP (M − H)+, 463.1787, found,
463.1783. The NMR data below were obtained from the sample with
A:B diastereomer ratio 2:1. When signals from individual diaster-
eomers could be distinguished, those are noted: for example, as 1H, B.
When signals overlapped, they are reported as 1H (A+B), meaning
1HA + 1HB. 31P{1H} NMR (CDCl3): δ 35.8 (broad). 1H NMR
(CDCl3): δ 8.59 (dd, J = 13, 9, 1H, A), 8.56 (dd, J = 13, 9, 1H, B),
8.10 (apparent d, J = 9, 1H, A+B), 7.94 (d, J = 8, 1H, A+B), 7.58
(d, J = 8, 1H, B), 7.54 (dd, J = 7, 7, 1H, B), 7.52−7.48 (m, 1H, A+B),
7.46 (d, J = 8, 1H, A), 7.37 (apparent t, J = 7, 1H, B), 7.28 (dm, J = 7,
1H, B), 7.25−7.19 (m, 1H, A+B), 7.18−7.13 (m, 2H, 2A+2B), 7.12
(dd, J = 7, 1, 2H, A), 7.08 (dd, J = 7, 7, 1H, A), 7.06 (d, J = 7, 1H, A),
7.01 (d, J = 8, 1H, B), 6.98−6.94 (m, 1H, A+B), 6.88 (d, J = 9, 1H, B),
6.87 (d, J = 9, 1H, A), 6.72 (dd, J = 8, 7, 1H, A), 6.68 (d, J = 8, 1H, A),
6.48 (dm, J = 7, 1H, A), 6.40−6.37 (m, 2H, B), 6.30 (d, J = 7, 1H, B),
3.70 (ABX, JAB = 17, JAX = 9, 1H, benzyl, A), 3.60 (AB, JAB = 17, 1H,
benzyl, A), 3.52−3.14 (m, overlapped, 8 ace-H (A+B) + 2 benzyl-H
(B)), 1.50−0.70 (m, broad, 3H, BH3 (A+B)). 13C{1H} NMR
(CDCl3): δ 147.8 (d, J = 2, Ar, quat, B), 146.9 (d, J = 2, Ar, quat, A),
144.8 (Ar, quat, B), 144.5 (Ar, quat, A), 142.5 (Ar, quat, A), 142.4 (Ar,
quat, B), 137.81 (d, J = 9, Ar, quat, B), 137.76 (d, J = 16, Ar, quat, B),
137.6 (d, J = 16, Ar, quat, A), 137.2 (d, J = 9, Ar, quat, A), 134.64
(d, J = 2, Ar, quat, B), 134.58 (d, J = 2, Ar, quat, A), 134.1 (d, J = 4, Ar,
quat, A), 133.6 (apparent d, J = 8, Ar, quat, A+B), 133.1 (d, J = 3, Ar,
quat, B), 132.8 (Ar, quat, B), 132.5 (Ar, quat, B), 132.3 (Ar, quat, A),
132.1 (Ar, quat, A), 131.9 (Ar, quat, A), 131.8 (Ar, quat, B), 130.4
(d, J = 21, Ar, B), 130.2 (d, J = 21, Ar, A), 129.0 (Ar, B), 128.7 (d, J =
10, Ar, A), 128.6 (d, J = 10, Ar, B), 128.2 (Ar, A), 128.1 (d, J = 14, Ar,
A), 127.99 (d, J = 14, Ar, B), 127.97 (Ar, B), 127.86 (Ar, overlapped,
A+B), 127.82 (Ar, B), 127.78 (Ar, A), 127.5 (Ar, B), 127.4 (Ar, A),
127.25 (Ar, A), 127.22 (Ar, A), 127.1 (Ar, B), 127.0 (d, J = 43, Ar,
quat, A), 126.72 (d, J = 42, Ar, quat, B), 126.66 (Ar, overlapped, A+B),
126.1 (Ar, B), 125.9 (Ar, 2 overlapping peaks, A), 125.7 (Ar, B), 125.5
(Ar, B), 125.19 (Ar, A), 125.17 (d, J = 60, Ar, quat, A+B), 124.2 (d, J = 8,
Ar, A), 124.1 (Ar, A), 123.9 (d, J = 8, Ar, B), 123.0 (Ar, B), 120.8
(d, J = 11, Ar, B), 120.4 (d, J = 11, Ar, A), 120.2 (Ar, B), 120.1 (Ar, A),
35.1 (d, J = 41, benzyl-CH2, A), 33.0 (d, J = 41, benzyl-CH2, B), 31.5
(ace-CH2, B), 31.4 (ace-CH2, A), 30.8 (ace-CH2, B), 30.7 (ace-CH2, A).
MeO-binaphthyl-PyraPhos(BH3) (3f). A solution of primary
phosphine 5 (25 mg, 0.079 mmol) in 0.3 mL of THF was added to a
solution of Cu(IPr)(Cl) (2 mg, 0.004 mmol, 5 mol %) in 0.3 mL of
THF. A solution of NaOSiMe3 (20 mg, 0.18 mmol, 2.2 equiv) in
0.3 mL of THF was added; the solution turned from colorless to
yellow immediately. A solution of the benzyl chloride 1 (23 mg,
0.082 mmol, 1 equiv) in 0.3 mL of THF was added. The yellow
solution was transferred to a NMR tube and sealed under N2. The
tube was heated at 50 °C, and the reaction was monitored by
31P NMR spectroscopy. After 19 h, the 31P NMR spectrum showed the
A 5 mL portion of 1 M aqueous NH4Cl was added slowly to the
reaction mixture together with 10 mL of ether. The water layer was
carefully removed with a pipet. The organic layer was further washed
with 2 × 5 mL of water and dried with MgSO4. Removing the solvent
under reduced pressure gave a yellow solid, which was purified by
chromatography on silica (Rf = 0.2 in 5/1 pentane/ethyl acetate) to
1
give 17 mg (35% yield) of a white solid, whose H NMR spectrum
showed a 2:1 ratio of diastereomers (A:B).
The reaction was also carried out on a larger scale (starting with
372 mg of primary phosphine) and was complete in 2 days. The ratio
of the two diastereomers was 1.8:1. After addition of borane−THF
and workup with chromatography as above, several eluted fractions were
collected, from which a small amount of X-ray-quality crystals was
obtained after 4 days as the solvent slowly evaporated at room tem-
perature. The 1H NMR spectrum of the crystals showed a more
diastereoenriched sample (5:1 A:B). However, satisfactory elemental
analysis data could not be obtained, even with the same batch of
H
DOI: 10.1021/acs.organomet.8b00198
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
formation of the intermediate secondary phosphine (two diaster-
eomers: δ −39.0 (51%), −42.2 (38%)) and a small amount (11%) of
the tertiary phosphine product (two diastereomers, around −14.0 ppm).
The reaction was complete in 42 h, with dr ≈ 1.2:1. Under positive N2
pressure, 0.16 mL of a BH3·THF solution (1 M in THF, 0.16 mmol,
2 equiv) was added dropwise with a syringe. The 31P NMR spectrum
confirmed the formation of the desired PyraPhos−borane adduct
(δ 36.1 ppm, broad).
A 3 mL portion of 1 M aqueous NH4Cl was added slowly to the
reaction mixture together with 5 mL of ether. The water layer was
carefully removed with a pipet and extracted again with 5 mL of ether.
The combined organic layer was dried with MgSO4. Removing the
solvent under reduced pressure gave 40 mg of white solid. This crude
product was purified by chromatography on silica (Rf = 0.16 in 3/1
pentane/ethyl acetate). Two portions of pure product were collected.
The first 13 mg was enriched in diastereomer A (2.3:1 A:B). The next
24 mg of product was slightly enriched in diastereomer B (0.8:1 A:B).
Thus, a total of 37 mg (95% yield) of white solid was collected, which
was pure as indicated by NMR spectra. Vapor diffusion of pentane into
an ether solution of the 24 mg portion at room temperature gave small
X-ray-quality crystals.
In a smaller scale experiment using 11 mg of the primary phosphine
and 10 mol % of the Cu(IPr)(Cl) catalyst precursor, the reaction was
first conducted at room temperature. Within 1 day, most of the
primary phosphine converted to the secondary phosphine. However,
in the following 3 days at room temperature, the reaction seemed to
have stopped and the secondary phosphine was still the major com-
pound present. Then the mixture was heated at 50 °C and the reaction
was complete in 2 days. After addition of borane and workup, 11 mg
(63% yield) of product was collected after silica gel chromatography as
a mixture of two diastereomers A and B (1.2:1 ratio).
A large-scale experiment starting with 307 mg of primary phosphine
was also conducted with the same procedure as the first one. After 22 h,
the primary phosphine was all converted to the intermediate secondary
phosphine (two diastereomers: 31P{1H} NMR δ −39.0, −42.1). How-
ever, it took 1 week for complete conversion to PyraPhos (δ −13.95,
−14.00). After addition of BH3·THF and workup, a pentane/ethyl
acetate mixture (10/1) was used as eluent for silica gel chroma-
tography. Twelve fractions were collected. The first fraction (7 mg)
was highly diastereoenriched (A:B = 99:1). The second fraction
(11 mg) had a 20:1 ratio (A:B), while the third fraction (20 mg) was
8:1. The last fraction (31 mg) was approximately 1:3 A:B. The total
mass of the 12 fractions was 338 mg (70% yield). Repeated chroma-
tography of the middle fractions gave more diastereoenriched sam-
ples, and we were able to obtain around 30 mg total of highly
diastereoenriched sample (A:B ≈ 20:1). X-ray-quality crystals (mixture
of two diastereomers) were also obtained from the chromatography
eluent, after slow evaporation of the solvent at room temperature.
Although a sample from the same batch of X-ray crystals was used for
elemental analysis, the results were low in carbon; see the Supporting
Information for NMR spectra.
123.7 (d, J = 8, Ar), 122.9 (Ar), 120.4 (d, J = 11, Ar), 119.7 (Ar), 119.0
(d, J = 4, Ar, quat), 111.2 (Ar), 55.3 (OCH3), 32.9 (d, J = 41, benzyl-
CH2), 31.4 (ace-CH2), 30.7 (ace-CH2).
From comparison of the NMR spectra of A and of A:B mixtures, the
spectra of diastereomer B could be assigned. 31P{1H} NMR (CDCl3):
1
δ 34.8 (broad). H NMR (CDCl3): δ 8.54 (dd, J = 13, 8, 1H), 8.08
(dd, J = 8, 2, 1H), 7.95 (d, J = 8, 1H), 7.55 (dd, J = 7, 7, 1H), 7.50
(ddd, J = 8, 7, 1, 1H), 7.42 (d, J = 8, 1H), 7.28−7.26 (m, 1H), 7.20
(ddd, J = 8, 7, 1, 1H), 7.19−7.14 (m, 3H), 6.96−6.92 (m, 2H), 6.82
(d, J = 8, 1H), 6.39 (d, J = 7, 1H), 6.29 (d, J = 9, 1H), 3.50−3.28
(m, overlapped, 4 ace-H + 1 benzyl-H), 3.29 (3H, OCH3), 3.24 (d, J =
18, 1H, benzyl), 1.35−0.80 (m, broad, 3H, BH3). 13C{1H} NMR
(CDCl3): δ 153.8 (Ar, quat), 147.3 (d, J = 2, Ar, quat), 142.1 (Ar,
quat), 141.5 (Ar, quat), 137.52 (d, J = 16, Ar, quat), 137.48 (d, J = 9,
Ar, quat), 134.9 (Ar, quat), 133.3 (Ar, quat), 133.2 (d, J = 8, Ar, quat),
131.7 (Ar, quat), 130.7 (d, J = 22, Ar), 129.4 (Ar), 129.0 (d, J = 10,
Ar), 128.0 (Ar), 127.9 (d, J = 13, Ar), 127.74 (Ar), 127.71 (Ar), 127.5
(d, J = 33, Ar, quat), 127.2 (d, J = 18, Ar, quat), 126.63 (Ar), 126.58
(Ar), 126.4 (Ar), 125.0 (d, J = 60, Ar, quat), 124.5 (Ar), 123.8 (d, J = 8,
Ar), 123.0 (Ar), 119.98 (d, J = 11, Ar), 119.96 (Ar), 118.9 (d, J = 4, Ar,
quat), 110.2 (Ar), 54.7 (OCH3), 33.2 (d, J = 41, benzyl-CH2), 31.5
(ace-CH2), 30.8 (ace-CH2).
Stoichiometric Reaction of Cu(IPr)(Cl) with PH2Ph, NaO-
SiMe3, and the Benzyl Chloride 1 To Form Ph-PyraPhos (2a).
A solution of PH2Ph (5 mg, 0.04 mmol) in 0.3 mL of THF was added
to a solution of Cu(IPr)(Cl) (22 mg, 0.045 mmol, 1 equiv) in 0.3 mL
of THF. A solution of NaOSiMe3 (5 mg, 0.04 mmol, 1 equiv) in
0.3 mL of THF was added. The solution turned from colorless to
bright yellow immediately. The 31P{1H} NMR spectrum of the mix-
ture showed two peaks, which were assigned to PH2Ph (δ −125.2) and
the copper−phosphido complex Cu(IPr)(PHPh) (12a; broad, δ −124).
Then a solution of the benzyl chloride 1 (13 mg, 0.045 mmol, 1 equiv)
in 0.3 mL of THF was added. The mixture turned light yellow-orange.
The 31P{1H} NMR spectrum showed the formation of secondary phos-
phine 6a (δ −37.8) and a small amount of Ph-PyraPhos 2a (δ −6.4).
This mixture was treated with a second equivalent of NaOSiMe3 (5 mg,
0.04 mmol, in 0.3 mL of THF). The 31P NMR signal of 6a slowly
disappeared, while more Ph-PyraPhos (2a) was formed; complete con-
version at room temperature took 2 days.
Observation of [Cu(IPr)(PH2Ph)][OTf] (7) and [Cu(IPr)(PHPh-
(CH2Ar))][OTf] (8a-OTf) via Stoichiometric Reaction of Cu(IPr)-
(OTf) with PH2Ph, NaOSiMe3, and the Benzyl Chloride 1. A solu-
tion of PH2Ph (29 mg, 0.26 mmol) in 1 mL of THF was added to a
solution of Cu(IPr)(OTf) (156 mg, 0.26 mmol, 1 equiv) in 1 mL of
THF. The 31P NMR spectrum showed a sharp triplet at −103.5 ppm
(JP−H = 314 Hz), perhaps due to exchange between PH2Ph and
[Cu(IPr)(PH2Ph)][OTf] (7). (The 31P NMR signals of copper-
coordinated phosphines are usually broad, in contrast to the sharp
signal observed here, and no signal due to free PH2Ph was observed.)
1H NMR (C6D6): δ 7.24 (t, J = 8, 2H, IPr Ar), 7.13−7.10 (m, 2H, Ar),
7.08 (d, J = 8, 4H, IPr Ar), 6.94−6.92 (m, 1H, Ph), 6.90−6.86 (m, 2H,
Ph), 6.47 (2H, IPr CHCH), 4.22 (d, J = 312, 2H, PH2), 2.60
(m, 4H, i-Pr CH), 1.24 (d, J = 7, 12H, i-Pr Me), 1.03 (d, J = 7, 12H,
i-Pr Me).
Anal. Calcd for C34H28BOP: C, 82.60; H, 5.71. Found: C, 80.99;
H, 5.64. HRMS: m/z calcd for C34H27BOP (M − H)+, 493.1893; found,
493.1890. The NMR data of diastereomer A were obtained from the
highly diastereoenriched (A:B = 99:1) sample. 31P{1H} NMR
1
A solution of NaOSiMe3 (29 mg, 0.26 mmol, 1 equiv) in 1 mL of
THF was added; the mixture turned bright yellow immediately. The
31P{1H} NMR spectrum showed two peaks, assigned to PH2Ph
(δ −124.8) and the copper−phosphido complex Cu(IPr)(PHPh)
(12a; δ −123, broad). A solution of the benzyl chloride 1 (73 mg,
0.26 mmol, 1 equiv) in 1 mL of THF was added. The 31P{1H} NMR
spectrum showed partial conversion of 12a to give a broad peak
at −33 ppm due to [Cu(IPr)(PHPh(CH2Ar))][OTf] (8a-OTf).
[Cu(IPr)(PHPh(CH2Ar))][OTf] (8a-OTf). A solution of Cu(IPr)-
(OTf) (179 mg, 0.30 mmol) in 2 mL of THF was added to a solution
of the secondary phosphine 6a (106 mg, 0.30 mmol, 1 equiv) in 2 mL
of THF. The mixture was stirred for 2 min, and then solvent was
removed under vacuum to give 282 mg (99% yield) of white solid.
NMR spectra were obtained in CD2Cl2, although slow decomposition
was observed in solution.
(CDCl3): δ 34.9 (apparent d, J = 49). H NMR (CDCl3): δ 8.63
(dd, J = 13, 8, 1H), 8.11 (dd, J = 8, 2, 1H), 7.95 (d, J = 8, 1H), 7.49
(ddd, J = 8, 7, 1, 1H), 7.35 (d, J = 8, 1H), 7.14 (ddd, J = 8, 7, 1, 1H),
7.12−7.10 (m, 2H), 7.08 (ddd, J = 8, 7, 1, 1H), 7.06 (d, J = 8, 1H),
6.98 (dd, J = 7, 7, 1H), 6.91 (ddd, J = 8, 7, 1, 1H), 6.86 (d, J = 9, 1H),
6.464 (d, J = 9, 1H), 6.460 (d, J = 8, 1H), 6.42 (dm, J = 7, 1H), 3.97
(dd, J = 17, 10, 1H, benzyl), 3.61 (3H, OCH3), 3.53 (d, J = 17, 1H,
benzyl), 3.38−3.15 (m, 4 ace-H), 1.35−0.80 (m, broad, 3H, BH3).
13C{1H} NMR (CDCl3): δ 153.9 (Ar, quat), 146.6 (d, J = 2, Ar, quat),
142.4 (Ar, quat), 141.2 (Ar, quat), 137.3 (d, J = 16, Ar, quat), 137.1
(d, J = 9, Ar, quat), 134.9 (d, J = 2, Ar, quat), 133.7 (Ar, quat), 133.3
(d, J = 8, Ar, quat), 132.9 (Ar, quat), 130.7 (d, J = 22, Ar), 129.3 (Ar),
128.6 (d, J = 10, Ar), 128.0 (Ar), 127.95 (d, J = 13, Ar), 127.7 (Ar),
127.4 (d, J = 28, Ar, quat), 127.2 (d, J = 15, Ar, quat), 126.8 (Ar),
126.6 (Ar, 2C), 126.3 (Ar), 125.0 (d, J = 60, Ar, quat), 124.7 (Ar),
I
DOI: 10.1021/acs.organomet.8b00198
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
HRMS (m/z): calcd for C46H52N2PCuBr (M+), 805.2347; found,
805.2347. 31P{1H} NMR (CD2Cl2): δ −29.2. 31P{1H} NMR (THF):
δ −33.1. 19F{1H} NMR (CD2Cl2): δ −78.8. 1H NMR (CD2Cl2):
δ 7.53 (apparent t, J = 8, 2H), 7.51 (d, J = 8, 1H), 7.42 (broad, 1H),
7.30−7.28 (m, 6H), 7.21 (apparent t, J = 7, 2H), 7.11 (broad, d, J = 6,
1H), 6.98 (broad, d, J = 6, 1H), 6.79 (broad, 2H), 6.73 (broad, 1H),
4.88 (broad, d, J = 344, P−H), 4.11 (broad, d, J = 11, 1H, benzyl-
CH2), 3.66 (broad, 1H, benzyl-CH2), 3.31 (4H, ace-CH2), 2.52 (sept,
J = 7, 4H, CH), 1.21 (d, J = 7, 12H, CH3), 1.08 (broad, 12H, CH3).
13C{1H} NMR (CD2Cl2): δ 179.2 (broad, NHC-carbene), 148.5 (Ar,
quat), 148.0 (broad, Ar, quat), 146.1 (Ar, quat), 142.4 (Ar, quat),
135.3 (Ar), 134.7 (Ar, quat), 134.4 (broad, Ar), 133.6 (Ar), 131.6
(broad, Ar), 131.3 (Ar), 129.6 (Ar), 128.0 (Ar, quat), 124.9 (Ar),
124.6 (Ar), 121.4 (Ar), 120.6 (Ar), 113.1 (Ar, quat), 30.6 (ace-CH2),
30.5 (ace-CH2), 29.9 (very broad, benzyl-CH2), 29.2 (CH), 25.1
(CH3), 24.2 (CH3). The triflate CF3 signal was not observed.
[Cu(IPr)(Ph-PyraPhos)][OTf] (9-OTf). A clear solution of Cu(IPr)-
(OTf) (99 mg, 0.16 mmol) in 2 mL of THF was added to a solution of
Ph-PyraPhos (45 mg, 0.16 mmol, 1 equiv) in 2 mL of THF. The mixture
was stirred for 20 min, and then solvent was removed under vacuum to
give 142 mg (98% yield) of white powder. The complex was relatively
stable in solution and could be characterized by NMR and mass
spectrometry. However, we were not able to obtain it in pure form,
since it decomposed slowly in the solid state at room temperature, and
attempted recrystallization by diffusion of pentane vapor into a THF
solution also led to decomposition.
conversion of 8a-OTf had occurred after 45 min, and the major spe-
cies in the reaction mixture was 9-OTf.
Reaction with CsOH·H2O. Thirty minutes after addition, some
[Cu(IPr)(Ph-PyraPhos)][OTf] (9-OTf; δ −6.3) had formed, together
with the sharp peak at −59 ppm. After 22 h, there was no more
starting material left, and the −59 ppm peak also disappeared.
[Cu(IPr)(Ph-PyraPhos)][OTf] (9-OTf) was the major component in
the mixture after the reaction was done.
[Au(IPr)(PHPh(CH2Ar))][OTf] (10). A solution of Au(IPr)(OTf)
(277 mg, 0.38 mmol, light purple due to slight decomposition) in
2 mL of THF was added to a solution of the secondary phosphine 6a
(134 mg, 0.38 mmol, 1 equiv) in 2 mL of THF. The 31P NMR
spectrum showed a broad peak at 2.9 ppm for the product. The mix-
ture was filtered through a Celite plug, and a light brown filtrate was
collected, which was further concentrated under vacuum to around
0.5 mL. While a small portion of the filtrate was taken out and pumped
down under vacuum to give a solid for NMR characterization, the
major part was used in recrystallization via vapor diffusion of pentane
into this concentrated THF solution. NMR spectra obtained in CD2Cl2
confirmed the right structure. However, the sample decomposed
slowly during recrystallization, and we could not obtain the product in
pure form.
HRMS (m/z): calcd for C46H52N2PAuBr (M+), 939.2717; found,
939.2695. 31P{1H} NMR (CD2Cl2): δ 6.1 (broad). 31P{1H} NMR
1
(THF): δ 2.9 (broad). H NMR (CD2Cl2): δ 7.65 (apparent t, J = 8,
2H), 7.64 (d, J = 7, 1H), 7.50 (apparent t, J = 7, 1H), 7.48 (2H), 7.40
(d, J = 8, 4H), 7.26 (broad apparent t, J = 7, 2H), 7.14 (d, J = 7, 1H),
6.99 (d, J = 7, 1H), 6.82 (dd, J = 13, 8, 2H), 6.70 (dd, J = 7, 3, 1H),
6.02 (dm, JP−H = 392, 1H, PH), 4.37 (ddd, J = 15, 5, 5, 1H, benzyl-
CH2), 3.73 (ddd, J = 15, 10, 8, 1H, benzyl-CH2), 3.33 (4H, ace-CH2),
2.54 (sept, J = 7, 4H, CH), 1.27 (d, J = 7, 12H, CH3), 1.23 (d, J = 7,
12H, CH3). 13C{1H} NMR (CD2Cl2): δ 189.8 (NHC-carbene), 148.7
(d, J = 4, Ar, quat), 148.6 (Ar, quat), 146.4 (Ar, quat), 142.5 (d, J = 2,
Ar, quat), 135.5 (Ar), 134.8 (d, J = 12, Ar), 133.73 (d, J = 7, Ar),
133.70 (Ar, quat, overlapped), 132.7 (d, J = 2, Ar), 131.8 (Ar), 129.6
(d, J = 11, Ar), 127.8 (Ar, quat), 125.4 (Ar), 125.0 (Ar), 121.5 (Ar),
120.5 (d, J = 4, Ar), 113.4 (Ar, quat), 31.8 (d, J = 30, benzyl-CH2),
30.7 (ace-CH2), 30.5 (ace-CH2), 29.4 (CH), 25.2 (CH3), 24.3 (CH3).
The triflate CF3 signal was not observed.
HRMS (m/z): calcd for C46H51N2PCu (M+), 725.3086; found,
725.3094. 31P{1H} NMR (CD2Cl2): δ 2.9 (broad). 19F{1H} NMR
1
(CH2Cl2): δ −78.9. H NMR (CD2Cl2): δ 7.57 (apparent t, J = 8,
2H), 7.38−7.27 (m, overlapped, 10H), 7.16 (apparent t, J = 7, 2H),
7.15−7.11 (overlapped, 1H), 6.69 (dd, J = 12, 7, 2H), 3.51−3.45
(m, 4H, ace-CH2), 3.46 (ABX, JAB = 18, m, overlapping, 1H, benzyl-
CH2), 3.39 (ABX, JAB = 18, JBX = 8, 1H, benzyl-CH2), 2.59 (sept, J = 7,
4H, CH), 1.26 (d, J = 7, 12H, CH3), 1.18 (d, J = 7, 12H, CH3).
13C{1H} NMR (CD2Cl2): δ 177.7 (broad, NHC-carbene), 149.4 (Ar,
quat), 145.9 (Ar, quat), 144.2 (Ar, quat), 138.8 (d, J = 7, Ar, quat),
138.4 (d, J = 14, Ar, quat), 134.1 (Ar, quat), 132.3 (Ar, quat), 131.8
(Ar), 131.7 (d, J = 15, Ar), 131.2 (Ar), 130.1 (d, J = 13, Ar), 129.5
(d, J = 11, Ar), 125.6 (d, J = 7, Ar), 124.7 (Ar), 124.6 (Ar), 121.53
(d, J = 11, Ar), 121.49 (Ar), 34.7 (d, J = 30, benzyl-CH2), 31.8 (ace-
CH2), 31.1 (ace-CH2), 28.9 (CH), 25.2 (CH3), 23.6 (CH3). The
triflate CF3 signal was not observed.
Reaction of [Cu(IPr)(PHPh(CH2Ar))][OTf] (8a-OTf) with Excess
NaOSiMe3. To a solution of [Cu(IPr)(PHPh(CH2Ar))][OTf]
(8a-OTf; 10 mg, 0.01 mmol) in 0.4 mL of THF was added a solu-
tion of NaOSiMe3 (4 mg, 0.04 mmol, 4 equiv) in 0.4 mL of THF. The
mixture was transferred to a NMR tube and monitored by 31P NMR
spectroscopy. Secondary phosphine 6a (80%) formed immediately,
together with some Ph-PyraPhos (2a; 20%). The reaction was com-
plete in 8 h with full conversion of 8a-OTf to 2a.
Reaction of [Cu(IPr)(Ph-PyraPhos)][OTf] (9-OTf) with Excess
[NOct4][Br]. A solution of [Cu(IPr)(Ph-PyraPhos)][OTf] (9-OTf; 21
mg, 0.024 mmol, 1 equiv) in 0.5 mL of THF was added to a solution
of [NOct4][Br] (16 mg, 0.029 mmol, 1.2 equiv) in 0.5 mL of THF.
The 31P NMR spectrum showed that 9-OTf was consumed and
Ph-PyraPhos (−6.5 ppm) was formed.
Reactions of [Cu(IPr)(PHPh(CH2Ar))][OTf] (8a-OTf) with the
Bases NaN(SiMe3)2 and CsOH•H2O. The two reactions were set up
together for comparison. A solution of [Cu(IPr)(PHPh(CH2Ar))]-
[OTf] (8a-OTf; 40 mg, 0.042 mmol) in 1 mL of THF was divided
evenly into two parts. One half was added to 5 mg of NaN(SiMe3)2
(0.027 mmol, 1.3 equiv), and the other half was added to approxi-
mately 10 mg of CsOH·H2O (0.06 mmol, 2.9 equiv). The mixtures
were transferred to two NMR tubes and monitored by 31P NMR
spectroscopy.
[Au(IPr)(Ph-PyraPhos)][OTf] (11). A solution of Au(IPr)(OTf)
(97 mg, 0.13 mmol) in 2 mL of THF was added to a solution of
Ph-PyraPhos (2a; 36 mg, 0.13 mmol, 1 equiv) in 2 mL of THF. The
mixture was stirred at room temperature for 20 min, and then the sol-
vent was removed under vacuum to give 129 mg (97% yield) of white
solid, which was relatively stable in CD2Cl2.
31P{1H} NMR (CD2Cl2): δ 36.6 (broad). 19F{1H} NMR (CD2Cl2):
1
δ −78.9. H NMR (CD2Cl2): δ 7.60 (apparent t, J = 8, 2H), 7.48
(2H), 7.43−7.35 (m, overlapped, 8H), 7.27 (dd, J = 8, 8, 1H), 7.20
(apparent t, J = 8, 2H), 6.80 (dd, J = 13, 8, 2H), 3.74 (AB, JAB = 18,
1H, benzyl-CH2), 3.59 (ABX, JAB = 18, JBX = 8, 1H, benzyl-CH2),
3.52−3.45 (m, 4H, ace-CH2), 2.58 (sept, J = 7, 4H, CH), 1.28 (t, J = 7,
24H, CH3). 13C{1H} NMR (CD2Cl2): δ 189.8 (broad, NHC-carbene),
150.3 (Ar, quat), 146.4 (Ar, quat), 144.7 (Ar, quat), 139.1 (d, J = 8, Ar,
quat), 138.8 (d, J = 16, Ar, quat), 133.7 (Ar, quat), 132.7 (d, J = 15,
Ar), 132.6 (Ar), 131.7 (Ar), 130.3 (d, J = 12, Ar), 129.8 (d, J = 11, Ar),
126.2 (d, J = 8, Ar), 125.4 (Ar), 125.1 (Ar), 122.0 (d, J = 11, Ar), 121.9
(Ar), 36.5 (d, J = 35, benzyl-CH2), 32.3 (ace-CH2), 31.5 (ace-CH2),
29.5 (CH), 25.3 (CH3), 24.2 (CH3). The triflate CF3 signal was not
observed.
Generation and Reactivity of Cu(IPr)(PHAr*) (12f). A solution
of NaOSiMe3 (5 mg, 0.045 mmol) in 0.5 mL of THF-d8 was added to
solid Cu(IPr)(Cl) (22 mg, 0.045 mmol), and a solution of (R)-MeO-
BinaphthylPH2 (14.3 mg, 0.045 mmol) in 0.5 mL of THF-d8 was
added to the resulting mixture to give a yellow solution. The 31P{1H}
1
NMR and H NMR spectra showed the presence of a mixture (about
1:1:1:1) of Cu(IPr)(OSiMe3), Cu(IPr)(PHAr*), Me3SiOH, and PH2Ar*
Reaction with NaN(SiMe3)2. Ten minutes after addition, a signifi-
cant amount of [Cu(IPr)(Ph-PyraPhos)][OTf] (9-OTf; δ −6.3) had
formed. In addition to the starting material at −37.6 ppm, a small peak
at −59 ppm, perhaps Cu(IPr)(PPhCH2Ar) (13a; Scheme 5), was also
observed, which almost disappeared after another 15 min. Complete
(Ar* = (R)-MeO-binaphthyl).
Cu(IPr)(PHAr*) (12f). 31P{1H} NMR (THF-d8, 21 °C): δ −124.7.
1
31P NMR (THF-d8, 21 °C): δ −124.7 (dd, J = 206, 9). H NMR
(THF-d8, 21 °C; signals overlapping those due to PH2Ar* and
J
DOI: 10.1021/acs.organomet.8b00198
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
[Cu]-OSiMe3 are noted): δ 7.74−7.72 (m, overlap, 1H), 7.69 (d, J = 8,
1H), 7.56−7.52 (m, 3H, overlap, 1H Ar + 2H IPr CHCH), 7.44
(2H, IPr CHCH, overlapping signals for [Cu]-OSiMe3 and [Cu]-
PHAr* (2H each)), 7.34 (d, J = 8, 4H, IPr Ar, overlapping signals for
[Cu]-OSiMe3 and [Cu]-PHAr* (4H each)), 7.28−7.26 (m, overlap,
1H), 7.13−7.09 (m, overlap, 2H), 7.01 (apparent t, J = 7, 1H), 6.91−
6.89 (m, overlap, 2H), 6.87 (apparent t, J = 7, 1H), 6.68 (d, J = 8, 1H),
6.63 (apparent t, JHH = JPH = 9, 1H), 3.48 (3H, OMe), 2.68−2.59 (m,
8H, IPr CH, overlapping signals for [Cu]-OSiMe3 and [Cu]-PHAr*),
1.96 (d, J = 206, 1H, Cu-PHAr*), 1.22−1.20 (m, 24H, IPr Me).
NMR data for the other components of the mixture were not
reported previously in this solvent, so they are included here for
convenience.
7.10−7.05 (m, 3H, Ar CH), 6.99−6.96 (m, 3H, Ar CH), 6.93 (d, J = 7,
1H, Ar CH A), 6.86 (d, J = 9, 2H, Ar CH), 6.82 (d, J = 7, 1H, Ar CH
A), 4.09−4.00 (br m, 1H, P-CH2), 3.76 (3H, OMe A), 3.70 (3H, OMe
B), 3.68−3.62 (m, 1H A, P-CH2), 3.55−3.45 (m, 1H B, P-CH2),
3.26−3.19 (m, 4H, ace-CH2). The P−H signals were not observed, but
in C6D6, they were a very broad multiplet at δ 4.2. 13C{1H} NMR
(THF-d8, 25 °C): δ 156.1 (quat A), 155.4 (quat B), 147.6 (quat A),
147.5 (quat B), 145.8 (d, J = 3, quat B), 145.6 (d, J = 2, quat A), 142.5
(quat A), 142.4 (quat B), 140.1 (d, J = 20, quat B), 140.0 (d, J = 20,
quat A), 135.9 (d, J = 17, quat B), 135.6 (d, J = 16, quat A), 135.2
(quat), 134.7 (d, J = 4, CH B), 134.5 (quat A), 134.4 (quat B), 134.1
(d, J = 4, quat A), 134.0 (quat B), 133.3 (quat), 133.0 (quat), 132.9
(d, J = 5, CH), 132.8 (d, J = 6, CH), 132.7 (d, J = 5, CH), 132.5 (d, J =
5, CH), 130.6 (CH A), 130.5 (CH B), 130.0 (quat A), 129.95 (quat
B), 128.7 (CH), 128.5 (CH), 128.4 (CH), 127.6 (d, J = 2, CH B),
127.5 (d, J = 2, CH A), 127.0 (d, J = 15, CH), 126.8 (CH B), 126.6
(d, J = 2, CH A), 126.53 (CH B), 126.49 (CH A), 126.4 (CH B),
126.3 (CH A), 125.7 (CH B), 125.68 (CH A), 123.97 (CH A), 123.95
(CH B), 120.5 (m, CH), 114.1 (CH B), 113.6 (CH A), 56.1 (OMe B),
56.0 (OMe A), 30.4 (CH2), 30.3 (d, J = 2, CH2), 29.9 (d, J = 18, CH2 B),
29.5 (d, J = 19, CH2 A).
MeO-binaphthyl-PH2 (5). 31P{1H} NMR (THF-d8, 21 °C): δ −129.5.
31P NMR (THF-d8, 21 °C): δ −129.5 (apparent td, J = 200, 4). 1H NMR
(THF-d8, 21 °C): δ 8.05 (d, J = 9, 1H), 7.88 (apparent dd, J = 8, 3, 2H),
7.85 (d, J = 8, 1H), 7.73 (dd, J = 8, 6, 1H), 7.54 (d, J = 9, 1H), 7.39
(t, J = 7, 1H), 7.28 (apparent t, J = 7, 1H), 7.18 (apparent q, J = 7,
2H), 7.09 (d, J = 8, 1H), 6.90 (d, J = 8, 1H), 3.75 (3H, OMe), 3.51
(ABX pattern, J = 200, 12, 2H, PH2).
1
Cu(IPr)(OSiMe3). H NMR (THF-d8, 21 °C): δ 7.47 (t, J = 8, 2H,
Generation of [Cu(IPr)(PHAr*(CH2Ar))][OTf] (8f-OTf) and Its
Low-Temperature Reaction with NaN(SiMe3)2. To Cu(IPr)(OTf)
(11 mg, 0.02 mmol) was added a solution of PH((R)-MeO-binaphthyl)-
(CH2Ar) (6f; 10 mg, 0.02 mmol) in 0.6 mL of THF-d8. The resulting pale
yellow solution was characterized by variable-temperature NMR spec-
troscopy. It contained a minor impurity (∼5% by 31P{1H} NMR
integration): δ −20.5, perhaps PAr*(CH2Ar)2, which was present in
the phosphine.
IPr Ar), 7.44 (2H, IPr CHCH), 7.34 (d, J = 8, 4H, IPr Ar), 2.68−
2.59 (m, 4H, IPr CH), 1.33 (d, J = 7, 12H, IPr Me), 1.22 (d, J = 7,
12H), −0.50 (9H, OSiMe3).
1
Me3SiOH. H NMR (THF-d8, 21 °C): δ 0.0 (br, 9H, SiMe3); the
OH signal could not be confidently assigned. In the mixture, both this
Me3Si peak and the −0.50 ppm peak due to Cu(IPr)(OSiMe3) were
broad at room temperature, but they became sharp on cooling to −35 °C.
The benzyl chloride 1 (13 mg, 0.046 mmol) was added to the
solution; the color changed to orange-yellow and a 7:1 mixture of the
two diastereomers of the secondary phosphine Ar*PH(CH2Ar) (6f)
was observed by 31P{1H} and 31P NMR spectroscopy (δ −39.4
(minor, dd, J = 210, 11), −42.4 (major, dd, J = 208, 5)), along with a
small amount of the primary phosphine and the Cu-PHAr*
HRMS: m/z calcd for C61H62BrCuN2OP (MH+), 1011.3079; found,
m/z 1011.3084. 31P{1H} NMR (THF-d8, 25 °C): δ −35.5 (br, B),
−38.6 (br, A). The ratio of diastereomers A and B was 1.5:1. 31P NMR
(THF-d8, 25 °C): δ −35.4 (br d, J = 289, B), −38.6 (br d, J = 325, A).
31P NMR (THF-d8, −65 °C; the signals were sharper at this tempera-
ture): δ −34.7 (d, J = 346, B), −38.7 (d, J = 340, A). 1H NMR signals
overlapped in most cases for the mixture of diastereomers. Some peaks
could be assigned to the major and minor isomers A and B. 1H NMR
(THF-d8, 25 °C): δ 8.10 (d, J = 9, 1H, Ar CH A), 7.99 (t, J = 9, 1H, Ar
CH), 7.92 (t, J = 6, 1H, Ar CH), 7.80 (d, J = 8, 1H, Ar CH B), 7.75
(t, J = 7, 1H, Ar CH), 7.58 (2H, IPr CHCH), 7.56 (d, J = 9, 1H, Ar
CH A), 7.49 (t, J = 8, 2H, IPr p-Ar), 7.42 (d, J = 7, 1H, Ar CH), 7.31
(d, J = 8, 4H A, IPr m-Ar), 7.29 (d, J = 8, 4H B, IPr m-Ar), 7.24−7.05
(m, overlapping, 4H, Ar CH), 6.97−6.82 (m, 3H, Ar CH), 6.72−6.66
(m, 2H, Ar CH), 6.49 (br m, 1H, Ar CH B), 4.62 (br d, J = 322, 1H,
PH), 3.90 (ABX, J = 14, 4, P-CH2), 3.80 (ABX, J = 14, P-CH2), 3.90−
3.75 (br m, P-CH2; 2H total for P-CH2 signals), 3.63 (3H A, OMe),
3.57 (overlapping, 3H B, OMe), 3.25−3.16 (br m, 4H, ace-CH2), 2.57
(m, 4H, IPr CH), 1.17 (apparent t, J = 7, 24H, IPr CH3).
To a THF-d8 solution of [Cu(IPr)(PH((R)-MeO-binaphthyl)-
(CH2Ar))][OTf] at −78 °C was added a solution of NaN(SiMe3)2
(4 mg, 0.03 mmol, 1.5 equiv) in 0.4 mL of THF-d8; the solution
became slightly more yellow. The reaction was monitored via variable-
temperature NMR spectroscopy in a precooled NMR probe. The
31P{1H} NMR spectrum at −65 °C showed that none of the starting
cation was present. The major 31P-containing product was the sec-
ondary phosphine 6f (two diastereomers A:B, 1:1.3 ratio). Other signals
were assigned to the impurity from the starting material (δ −25.3, 4%),
Ar*-PyraPhos (2f; two diastereomers C:D with overlapping signals at
δ −17.1), and two additional minor signals E:F, for which no P−H
coupling was observed (δ −9.1, −15.6, 1:1 ratio). When the tem-
perature was raised to room temperature, these signals disappeared
and more Ar*-PyraPhos was formed. Note: the 31P NMR chemical
shifts were temperature dependent.
1
intermediate. The H NMR spectrum showed that Cu(IPr)(Cl) was
the major species, along with some Cu(IPr)(OSiMe3), presumably
because of inexact stoichiometry on this small scale. After 4 days, the
ratio of secondary phosphines had changed to 1:1.4, but some decom-
position was also observed.
1
Cu(IPr)(Cl). H NMR (THF-d8, 21 °C): δ 7.51 (2H, CHCH),
7.49 (t, J = 8, 2H, Ar), 7.35 (d, J = 8, 4H, Ar), 2.68−2.60 (m, 4H, CH),
1.31 (d, J = 7, 12H, Me), 1.23 (d, J = 7, 12H, Me).
PHAr*(CH2Ar) (6f). To Ar*PH2 (5, 79 mg, 0.25 mmol) was added
a solution of Pt(dppe)(Me)(Cl) (8 mg, 0.05 mmol, 20 mol %) and
NaOSiMe3 (55 mg, 0.5 mmol, 2 equiv) in 2 mL of THF, resulting in a
light yellow solution, which was added to the benzyl chloride 1
(71 mg, 0.25 mmol) to give an orange solution. The reaction mixture
was stirred overnight. Solvent was removed in vacuo, the orange solid
residue was extracted with 5 mL of ether, and the solution was filtered
through Celite to give a yellow solution. The solution was cooled to
−22 °C, yielding a tan solid. The solid was dissolved in 9/1 pentane/
THF and filtered through a silica plug, yielding a white solid (40 mg,
29%). Alternatively, 10 mL of pentane was added to the original
orange solution and the reaction mixture was filtered through a silica
plug, yielding a colorless solution. The solvent was removed in vacuo,
resulting in a white solid (91 mg, 65%).
We could not get satisfactory elemental analyses; results were
consistent with phosphine oxidation. Anal. Calcd for C34H26BrOP: C,
72.73; H, 4.67. Calcd for C34H26BrO2P: C, 70.72; H, 4.54. Found: C,
71.32; H, 4.35. HRMS: m/z calcd for C34H27BrOP[MH]+, 561.0983;
found, 561.0977.
NMR spectra are reported for a mixture of two diastereomers (A:B,
dr = 1.3:1), with clear differences between diastereomers labeled
accordingly. 31P{1H} NMR (THF, 25 °C): δ −39.6 (A), −42.7 (B).
31P NMR (THF-d8, 25 °C): δ −39.3 (dd, J = 14, 211, A), −42.4 (dd,
31P{1H} NMR (THF-d8, −65 °C): δ −9.1 (E), −15.6 (F), −17.1
(overlapping, PyraPhos), −31.0 (A), −36.2 (B). 31P NMR (THF-d8,
−65 °C): δ −9.1 (E), −15.5 (F), −17.0 (apparent d, J = 21, PyraPhos),
−30.6 (dd, J = 15, 211), −36.7 (dd, J = 6, 209). 31P{1H} NMR (THF-d8,
25 °C): δ −14.2 (PyraPhos C), −14.3 (PyraPhos D), −39.3 (A), −42.4
(B). Over the next 6 days, secondary phosphine 6f was converted to
PyraPhos 2f (1.7:1 ratio), according to 31P{1H} NMR spectroscopy.
1
J = 6, 209, B). H NMR (THF-d8, 25 °C): δ 8.00 (d, J = 10, 1H,
Ar CH B), 7.99 (d, J = 9, 1H, Ar CH A), 7.83−7.80 (overlapping m,
4H, Ar CH), 7.78 (d, J = 8, 1H, Ar CH B), 7.72 (d, J = 8, 1H, Ar CH A),
7.56−7.50 (overlapping m, 5H, Ar CH), 7.38−7.34 (overlapping m, 3H,
Ar CH), 7.24−7.19 (m, 2H, Ar CH), 7.15−7.13 (m, 3H, Ar CH),
K
DOI: 10.1021/acs.organomet.8b00198
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Reaction of [Cu(IPr)(PHAr*(CH2Ar))][OTf] (8f) with NaO-
SiMe3. Cation 8f was prepared as above from Cu(IPr)(OTf) (11 mg,
0.02 mmol) and 6f (10 mg, 0.02 mmol) in 1 mL of THF-d8. This mix-
ture was added to NaOSiMe3 (3 mg, 0.03 mmol, 1.5 equiv), resulting in
a slightly darker yellow color, and the reaction was monitored by
31P{1H} NMR spectroscopy. As observed with NaN(SiMe3)2 above, the
major P-containing products were the diastereomers of 6f, plus a small
amount of PyraPhos 2f; more 2f formed over the course of 1 day.
Reaction of [Au(IPr)(PHPh(CH2Ar))][OTf] (10) with Excess
NaOSiMe3. A light purple solution of [Au(IPr)(PHPh(CH2Ar))]-
[OTf] (65 mg, 0.07 mmol) in approximately 0.7 mL of THF was
added to a solution of NaOSiMe3 (13 mg, 0.12 mmol, 1.6 equiv) in
0.3 mL of THF. The mixture was transferred to a NMR tube and
monitored by 31P NMR spectroscopy, which showed the formation of
PHPh(CH2Ar) (6a; 31P{1H} NMR δ −37.8) and two other uniden-
tified peaks at 14.7 and −13.2 ppm. [Au(IPr)(Ph-PyraPhos)][OTf]
(11, δ 36.7) was not observed.
Attempted Synthesis of Ph-PyraPhos with Au(IPr)(Cl) as
Catalyst Precursor. A solution of PH2Ph (8 mg, 0.07 mmol) in 0.3 mL
of THF was added to a solution of Au(IPr)(Cl)(2 mg, 0.003 mmol,
5 mol %) in 0.3 mL of THF. A solution of NaOSiMe3 (18 mg, 0.16 mmol,
2.2 equiv) in 0.3 mL of THF was added, followed by a solution of the
benzyl chloride 1 (20 mg, 0.07 mmol, 1 equiv) in 0.3 mL of THF. The
mixture was transferred to a NMR tube and monitored by 31P NMR spec-
troscopy. Secondary phosphine 6a (δ −37.8) formed as PH2Ph was
consumed. After 12 h, along with 8 (47%) and PH2Ph (31%), two
other unidentified peaks at −11 ppm (11%) and −100.5 ppm (11%)
were observed, but Ph-PyraPhos 2a (δ −6.4) was not.
ACKNOWLEDGMENTS
■
Research at Dartmouth College was supported by the U.S.
National Science Foundation (CHE-126578 and CHE-
1562037). The Newcastle group thanks the Engineering and
Physical Sciences Research Council (EPSRC) for a Career
Acceleration Fellowship (LJH EP/G005206/1) and its
National Mass Spectrometry Service Centre, Swansea, UK, and
Newcastle University for two studentships (C.S. and J.T.F.).
REFERENCES
■
(1) (a) Glueck, D. S. Top. Organomet. Chem. 2010, 31, 65−100.
(b) Tappe, F. M. J.; Trepohl, V. T.; Oestreich, M. Synthesis 2010,
2010, 3037−3062. (c) Wauters, I.; Debrouwer, W.; Stevens, C. V.
Beilstein J. Org. Chem. 2014, 10, 1064−1096. (d) Koshti, V.; Gaikwad,
S.; Chikkali, S. H. Coord. Chem. Rev. 2014, 265, 52−73. (e) Rosenberg,
L. ACS Catal. 2013, 3, 2845−2855. (f) Bange, C. A.; Waterman, R.
Chem. - Eur. J. 2016, 22, 12598−12605.
(2) Catalysis Without Precious Metals; Bullock, R. M., Ed.; Wiley-
VCH: Weinheim, Germany, 2010.
(3) For a review, see: (a) Zhang, H.; Zhang, X.-Y.; Dong, D.-Q.;
Wang, Z.-L. RSC Adv. 2015, 5, 52824−52831. See also: (b) Zhang, S.-
L.; Bie, W.-F.; Huang, L. Organometallics 2014, 33, 5263−5271.
(c) Gelman, D.; Jiang, L.; Buchwald, S. L. Org. Lett. 2003, 5, 2315−
2318. (d) Tani, K.; Behenna, D. C.; McFadden, R. M.; Stoltz, B. M.
Org. Lett. 2007, 9, 2529−2531. (e) Krout, M. R.; Mohr, J. T.; Stoltz, B.
M. Org. Synth. 2009, 86, 181−193. (f) Bel
́
́
anger, E.; Pouliot, M.-F.;
Generation of [Cu(IPr)(Ph-PyraPhos)][PF6] (9-PF6) from
Stoichiometric Alkylation/Arylation of PH2Ph. A solution of
PH2Ph (25 mg, 0.23 mmol) in 1 mL of THF was added to a solution
of Cu(IPr)(Cl) (110 mg, 0.23 mmol, 1 equiv) in 1 mL of THF. To the
mixture was added a solution of NaOSiMe3 (56 mg, 0.5 mmol,
2.2 equiv) in 1 mL of THF, followed by a solution of the benzyl
chloride 1 (64 mg, 0.23 mmol, 1 equiv) in 1 mL of THF. The mixture
was stirred at room temperature. After 10 min, almost all PH2Ph had
reacted to give secondary phosphine 6a. The full conversion from
intermediate 6a to PyraPhos 2a took 2 days. After the reaction was
complete, 48 mg of NH4PF6 (1.3 equiv) was added to the reaction
mixture, forming a light pink suspension. The 31P{1H} NMR spectrum
showed a broad peak at −3.7 ppm due to [Cu(IPr)(Ph-PyraPhos)]-
[PF6] (9-PF6), which decomposed in the mixture over 1 day.
Paquin, J.-F. Org. Lett. 2009, 11, 2201−2204. (g) Van Allen, D.;
Venkataraman, D. J. Org. Chem. 2003, 68, 4590−4593. (h) Kondoh,
A.; Yorimitsu, H.; Oshima, K. J. Am. Chem. Soc. 2007, 129, 4099−
4104. (i) Isley, N. A.; Linstadt, R. T. H.; Slack, E. D.; Lipshutz, B. H.
Dalton Trans. 2014, 43, 13196−13200. (j) Leyva-Perez, A.; Vidal-
Moya, J. A.; Cabrero-Antonino, J. R.; Al-Deyab, S. S.; Al-Resayes, S. I.;
Corma, A. J. Organomet. Chem. 2011, 696, 362−367. (k) Chen, Y.-R.;
Feng, J.-J.; Duan, W.-L. Tetrahedron Lett. 2014, 55, 595−597. (l) Cain,
M. F.; Reynolds, S. C.; Anderson, B. J.; Glueck, D. S.; Golen, J. A.;
Moore, C. E.; Rheingold, A. L. Inorg. Chim. Acta 2011, 369, 55−61.
(m) Cain, M. F.; Hughes, R. P.; Glueck, D. S.; Golen, J. A.; Moore, C.
E.; Rheingold, A. L. Inorg. Chem. 2010, 49, 7650−7662.
(4) (a) The crystal structure of Cu(PPh3)3(PH2)·(acetone) was
reported, but without other characterization data: Hill, T. N. Z.
Kristallogr. - New Cryst. Struct. 2013, 228, 120−122. (b)
[CuBr(P(IPr)2)][OTf] was described as a P(I) compound containing
an unusual cationic phosphido ligand supported by one normal and
one abnormal NHC substituent: Schwedtmann, K.; Holthausen, M.
H.; Feldmann, K.-O.; Weigand, J. J. Angew. Chem., Int. Ed. 2013, 52,
14204−14208. (c) Terminal copper−phosphido complexes observed
by NMR spectroscopy include Cu(triphos)(PPh2)3m and Cu(PPh3)
(PPh2): Lemmen, T. H.; Goeden, G. V.; Huffman, J. C.; Geerst, R. L.;
Caulton, K. G. Inorg. Chem. 1990, 29, 3680−3685. [Li][Cu-
Me3(PPh2)] may have a Li−P−Cu interaction: Bertz, S. H.;
Murphy, M. D.; Ogle, C. A.; Thomas, A. A. Chem. Commun. 2010,
46, 1255−1256. (d) For similar lithium cuprates, see: 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. Cowley,
A. H.; Giolando, D. M.; Jones, R. A.; Nunn, C. M.; Power, J. M. J.
Chem. Soc., Chem. Commun. 1988, 208−209. (e) An equilibrium
between dimeric [Cu(dppe)(PPh2)]2 and monomeric Cu(dppe)
(PPh2) was proposed on the basis of ebullioscopic molecular weight
measurements: van Koten, G.; Noltes, J. G.; Spek, A. L. J. Organomet.
Chem. 1978, 159, 441−463. (f) Terminal Cu−phosphido complexes
may be involved in P-alkylation of [Cu(diimine)(PPh2)]n oligomers:
Meyer, C.; Grutzmacher, H.; Pritzkow, H. Angew. Chem., Int. Ed. Engl.
ASSOCIATED CONTENT
■
S
* Supporting Information
The Supporting Information is available free of charge on the ACS
Publications website at DOI: 10.1021/acs.organomet.8b00198.
Additional experimental information, NMR spectra, and
details of the X-ray crystallographic studies (PDF)
Accession Codes
CCDC 1829752−1829757 contain the supplementary crystal-
lographic data for this paper. These data can be obtained free of
charge via www.ccdc.cam.ac.uk/data_request/cif, or by email-
ing data_request@ccdc.cam.ac.uk, or by contacting The
Cambridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, UK; fax: +44 1223 336033.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail for D.S.G.: Glueck@Dartmouth.Edu.
1997, 36, 2471−2473. Meyer, C.; Scherer, M.; Schonberg, H.;
̈
ORCID
Ruegger, H.; Loss, S.; Gramlich, V.; Grutzmacher, H. Dalton Trans.
̈ ̈
2006, 137−148. (g) Terminal Cu(NHC)−phosphido complexes,
formed from trimeric phosphido-bridged precursors, were proposed
intermediates in catalytic hydrophosphination of alkynes: Yuan, J.;
Zhu, L.; Zhang, J.; Li, J.; Cui, C. Organometallics 2017, 36, 455−459.
David S. Glueck: 0000-0002-8438-8166
Notes
The authors declare no competing financial interest.
L
DOI: 10.1021/acs.organomet.8b00198
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
(h) Cu-phosphido complexes were proposed as intermediates in Cu-
mediated synthesis of P-macrocycles via deprotonation/alkylation of
coordinated secondary phosphines: Nell, B. P.; Swor, C. D.; Henle, E.
A.; Zakharov, L. N.; Rinehart, N. I.; Nathan, A.; Tyler, D. R. Dalton
Trans. 2016, 45, 8253−8264. Kang, Y. B.; Pabel, M.; Pathak, D. D.;
Willis, A. C.; Wild, S. B. Main Group Chem. 1995, 1, 89−98.
(5) Okugawa, Y.; Hirano, K.; Miura, M. Angew. Chem., Int. Ed. 2016,
55, 13558−13561.
primary or secondary phosphine to a Cu(IPr)−phosphido complex
does not follow consistent trends. For PHPh2 (δ −41.4) and Cu(IPr)
(PPh2) (C; Chart 1, δ −27.8), the PPh2 signal was downfield from that
of the phosphine (ref 6). However, Cu(IPr)(PHPh) (δ −124) and
PH2Ph (δ −125.2) or Cu(IPr)(PHAr*) (δ −124.7) and PH2Ar* (δ
−129.5) had similar 31P{1H} NMR chemical shifts (see Table 2), and
the shift for Cu(IPr)(P(SiMe3)2) (C; Chart 1, δ −268, ref 7), was
upfield from that of the secondary phosphine (δ −237.3): Uhlig, W.;
Tzschach, A. Z. Anorg. Allg. Chem. 1989, 576, 281−283. Therefore,
the data for PHPh(CH2Ar) (6a, δ −37.8) and 13a (δ −59) are
consistent with the assignment of 13a as Cu(IPr)(PPh(CH2Ar)), but
not definitive.
(25) Because these temperature-sensitive intermediates were formed
as minor components of the mixture, they could be characterized only
by 31P{1H} and 31P NMR spectroscopy, which showed the absence of
one-bond P−H coupling. Their 31P NMR shifts are not consistent
with the secondary phosphido complex 13f, but they might also be
Cu(III) phosphido complexes 14f (Scheme 7) or cationic PyraPhos
complexes 9f. Similarly, the proposed structure of P-Ph intermediate
13a is speculative, and we cannot rule out the possibility that it is 14a,
but it is clearly not 9a, which was prepared independently.
(26) 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.
(6) Fortman, G. C.; Slawin, A. M. Z.; Nolan, S. P. Organometallics
2010, 29, 3966−3972.
(7) Khalili Najafabadi, B.; Corrigan, J. F. Dalton Trans. 2015, 44,
14235−14241.
(8) Laitar, D. S. Synthetic and catalytic studies of Group 11 N-
heterocyclic carbene complexes. Ph.D. thesis, MIT, 2006; http://hdl.
handle.net/1721.1/36268.
(9) Wang, G.; Guino-o, M. A.; Glueck, D. S.; Golen, J. A.; Daley, C. J.
A.; Rheingold, A. L. Dalton Trans. 2015, 44, 9943−9954.
(10) (a) Phosphorus Ligands in Asymmetric Catalysis. Synthesis and
Applications; Borner, A., Ed.; Wiley-VCH: Weinheim, Germany, 2008.
̈
(b) Phosphorus(III) Ligands in Homogeneous Catalysis: Design and
Synthesis; Kamer, P. C. J., van Leeuwen, P. W. N. M., Eds.; Wiley:
Chichester, U.K., 2012.
(11) Ficks, A.; Sibbald, C.; Ojo, S.; Harrington, R. W.; Clegg, W.;
Higham, L. J. Synthesis 2013, 45, 265−271.
(27) (a) Font, M.; Parella, T.; Costas, M.; Ribas, X. Organometallics
2012, 31, 7976−7982. (b) Casitas, A.; Ribas, X. Chem. Sci. 2013, 4,
2301−2318.
(12) Stewart, B.; Harriman, A.; Higham, L. J. Organometallics 2011,
30, 5338−5343.
(13) Munro-Leighton, C.; Blue, E. D.; Gunnoe, T. B. J. Am. Chem.
Soc. 2006, 128, 1446−1447.
(28) The Ar−Br bond could also be activated in processes which
avoid Cu(III), as in: Van Hoveln, R.; Hudson, B. M.; Wedler, H. B.;
Bates, D. M.; Le Gros, G.; Tantillo, D. J.; Schomaker, J. M. J. Am.
Chem. Soc. 2015, 137, 5346−5354.
(14) For the related cationic phosphine complexes [Cu(IPr)(PPh3)]
[PF6] and [Cu(IPr)(P(t-Bu)3][BF4], see: (a) Guo, S.; Lim, M. H.;
Huynh, H. V. Organometallics 2013, 32, 7225−7233. (b) Lazreg, F.;
Slawin, A. M. Z.; Cazin, C. S. J. Organometallics 2012, 31, 7969−7975.
(15) (a) Tsui, E. Y.; Muller, P.; Sadighi, J. P. Angew. Chem., Int. Ed.
2008, 47, 8937−8940. (b) Cavarzan, A.; Scarso, A.; Sgarbossa, P.;
Strukul, G.; Reek, J. N. H. J. Am. Chem. Soc. 2011, 133, 2848−2851.
(c) See the Supporting Information for the crystal structure of
Au(IPr)(OTf).
(29) (a) Glueck, D. S. Synlett 2007, 2007, 2627−2634. (b) Glueck,
D. S. Chem. - Eur. J. 2008, 14, 7108−7117.
(30) 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.
(31) Pangborn, A. B.; Giardello, M. A.; Grubbs, R. H.; Rosen, R. K.;
Timmers, F. J. Organometallics 1996, 15, 1518−1520.
(16) The large chemical shift differences between analogous copper
and gold complexes are consistent with previous observations; see
Table S1 in the Supporting Information.
(17) (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) Glueck, D. S. Coord. Chem. Rev. 2008, 252, 2171−2179; Coord.
Chem. Rev. 2011, 255, 356 (erratum).
(32) Brunel, J. M.; Faure, B.; Maffei, M. Coord. Chem. Rev. 1998,
178−180, 665−698.
(33) Kubas, G. J. Inorg. Synth. 2007, 28, 68−70.
(34) Goodwin, N. J.; Henderson, W.; Nicholson, B. K.; Fawcett, J.;
Russell, D. R. J. Chem. Soc., Dalton Trans. 1999, 1785−1793.
(35) Clark, H. C.; Jablonski, C. R. Inorg. Chem. 1975, 14, 1518−1526.
(18) Johnson, M. W.; Shevick, S. L.; Toste, F. D.; Bergman, R. G.
Chem. Sci. 2013, 4, 1023−1027.
(19) (a) Wyss, C. M.; Tate, B. K.; Bacsa, J.; Gray, T. G.; Sadighi, J. P.
Angew. Chem., Int. Ed. 2013, 52, 12920−12923. (b) This complex was
also prepared in ref 6, by the reaction of Cu(IPr)(OH) with
Me3SiNMe2, but mistakenly formulated as Cu(IPr)(SiMe3): Fortman,
G. C.; Nolan, S. P., personal communication, 2010.
(20) Rogers, J. R.; Wagner, T. P. S.; Marynick, D. S. Inorg. Chem.
1994, 33, 3104−3110.
(21) (a) Albert, J.; Magali Cadena, J.; Granell, J.; Muller, G.; Panyella,
D.; Sanudo, C. Eur. J. Inorg. Chem. 2000, 2000, 1283−1286. (b) Bader,
̃
A.; Nullmeyers, T.; Pabel, M.; Salem, G.; Willis, A. C.; Wild, S. B.
Inorg. Chem. 1995, 34, 384−389. (c) Huang, Y.; Li, Y.; Leung, P.-H.;
Hayashi, T. J. Am. Chem. Soc. 2014, 136, 4865−4868.
(22) Fast P-epimerization in 12f can be explained most simply by
pyramidal inversion of the phosphido group, as observed for other
metal complexes (see refs 17b and 20). A reviewer suggested another
possible pathway, via reversible P protonation by Me3SiOH or
Ar*PH2, as in the related processes for secondary phosphines (ref 21).
Unfortunately, we were not able to isolate 12f in order to test this idea.
(23) We rejected the possibility that the −59 ppm 31P{1H} NMR
signal was due to deprotonated 6a, because treatment of 6a with n-
BuLi in THF gave rise to a signal at −20.5 ppm.
(24) In principle, the 31P{1H} NMR chemical shift could provide
structural information for 13a. However, such data for conversion of a
M
DOI: 10.1021/acs.organomet.8b00198
Organometallics XXXX, XXX, XXX−XXX