P-P Menschutkin preparation of prototypical phosphinophosphonium salts

Article abstract of DOI:10.1039/c2cc33082b

Reactions of Me₃P with alkyl- or arylchlorophosphines yield phosphinophosphonium salts in quantitative yields, demonstrating a Menschutkin P-P methodology that has potentially broad application for element-element bond formation.

Full text of DOI:10.1039/c2cc33082b

View Online / Journal Homepage / Table of Contents for this issue
This article is part of the
Frontiers in Molecular
Main Group Chemistry
web themed issue
Guest editors: Manfred Scheer and Richard
Layfield
All articles in this issue will be gathered together
online at
www.rsc.org/cc/maingroup

Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2012, 48, 7359–7361
www.rsc.org/chemcomm
COMMUNICATION
P–P Menschutkin preparation of prototypical phosphinophosphonium
saltswz
Saurabh S. Chitnis,a Elizabeth MacDonald,b Neil Burford,*ab Ulrike Werner-Zwanzigerb and
Robert McDonaldc
Received 29th April 2012, Accepted 31st May 2012
DOI: 10.1039/c2cc33082b
Reactions of Me3P with alkyl- or arylchlorophosphines yield
phosphinophosphonium salts in quantitative yields, demonstrating
a Menschutkin P–P methodology that has potentially broad
application for element–element bond formation.
Table 1 lists the products identified by P-31 NMR spectro-
scopy from mixtures of R3P (R = Me, Ph) with R0nPCl3ꢀn
(R0 = Me, Ph; n = 0, 1, 2) or dmpe (dimethylphosphinoethane,
Me2PCH2CH2PMe2) as illustrated in Schemes 1 and 2. Me3P
reacts with Me2PCl or MePCl2 on contact (white powders
formed from the gas phase) to give compounds 3 and 4,
respectively, and dmpe reacts with Me2PCl or MePCl2 on
contact to give compounds 5 and 6, respectively. All reactions
proceed quantitatively according on the stoichiometry of the
mixture. The reactions also occur in acetonitrile with immediate
precipitation of the products. Reactions of Me3P or dmpe with
Ph2PCl and PhPCl2 in dichloromethane give the analogous
mono- and dicationic products 7–10. Product 8 is partially
converted to Me3PCl2 and cyclo-P4Ph4 over several weeks at
room temperature. Reaction of Me3P with PCl3 yielded a
mixture of insoluble polymeric material, Me3PCl2 and 11,
identified by comparison of P-31 chemical shifts and coupling
constants with those of known triphosphenium cations.10 The
composition of products 3–10 was established by elemental
analysis and P-31 NMR spectroscopy and their ionic structure
was confirmed by comparison of chemical shifts with those of the
previously reported [O3SCF3]ꢀ salts of phosphinophosphonium
cations (Table 1).
The classical Menschutkin reaction1 yields an alkylammonium
salt from an alkyl halide and an amine, representing a Lewis
basic nitrogen center interacting with a Lewis acidic carbon
center to eliminate a halide anion. The reaction offers a facile
approach to N–C bond formation, but should be generally
applicable between atomic centers that offer sufficient basicity and
centers that participate in a bond that involves a sufficiently effective
leaving group. For example, specific imines displace a chloride ion
from a chloroaminophosphine to give an oniophosphine,
represented by 1.2 Moreover, NMR spectroscopic data and
elemental analytical data indicate Pn–Pn0 bond formation
(Pn = P, As, Sb) in reactions of pnictines with halopnictines.3–7
We now provide definitive confirmation of a homoatomic
P–P Menschutkin reaction and demonstrate that Me2PCl and
Ph2PCl are both sufficiently Lewis acidic to engage methylated
phosphines as donors to form P–P bonds in phosphinophos-
phonium 2 frameworks. We have used the reaction to prepare
the chloride salt of the prototypical phosphinophosphonium
cation [Me3PPMe2]+, 3, which has been comprehensively charac-
terized and compared with salts formed from chlorophosphines
that have been activated to phosphenium cations by halide
abstraction.8,9
a Department of Chemistry, University of Victoria, Victoria,
BC V8W3V6, Canada. E-mail: nburford@uvic.ca;
Fax: + 1 250 721 7147; Tel: + 1 250 721 7150
Although the solubility of compounds 3–6, 8, and 10 is too
low to obtain solution P-31 NMR data, P-31 CP/MAS NMR
spectra for the solids provides definitive data, as shown for
example in Fig. 1 for the precipitate from the reaction between
Me3P and Me2PCl. The distinctive pattern of two doublets
(one in the phosphine region and the other in the phosphonium
region) is consistent with that observed for [Me3PPMe2][O3SCF3].11
In contrast to reactions of PMe3 or dmpe, P-31 NMR spectra of
mixtures involving PPh3 with Me2PCl, MePCl2, Ph2PCl, PhPCl2,
b Department of Chemistry, Dalhousie University, Halifax,
NS B3H4J3, Canada
c X-ray Crystallography Laboratory, Department of Chemistry,
University of Alberta, Edmonton, Alberta T6G2G2, Canada
w This article is part of the ChemComm ‘Frontiers in Molecular Main
Group Chemistry’ web themed issue.
z Electronic supplementary information (ESI) available: CCDC
879397–879400. For ESI and crystallographic data in CIF or other
electronic format see DOI: 10.1039/c2cc33082b
c
This journal is The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 7359–7361 7359

Table 1 P-31 NMR chemical shifts, 1JP–P coupling constants and spin systems for phosphinophosphonium cations as chloride and triflate salts
(NMR data obtained for solutions, unless denoted by CP/MAS)
1
Dd (ppm); JP–P (Hz, ꢁ15); spin system
Cation
Counterion = [Cl]ꢀ
Counterion = [O3SCF3]ꢀ
11
[Me3PPMe2]+ (3)
[(Me3P)2PMe]2+ (4)
[dmpe(PMe2)2]2+ (5)
[dmpePMe]2+ (6)
[Me3PPPh2]+ (7)
[(Me3P)2PPh]2+ (8)
[dmpe(PPh2)2]2+ (9)
[dmpePPh]2+ (10)
[(Me3P)2P]+ (11)
20, ꢀ66 (CP/MAS); 249; AX
20, ꢀ63 (CP/MAS); broad; AXX0
30, ꢀ75 (CP/MAS); 272; AX
50, ꢀ86 (CP/MAS); 296, 271; AXX0
15, ꢀ24; broad; AX
18, ꢀ59; 274; AX
19
25, ꢀ63; 285, 295; AXX0
24, ꢀ58; broad; AX; (This work)
11
53, ꢀ88; 278, 283; AXX0; (This work)
15, ꢀ23; 289; AX
b
19
a97, ꢀ51, 31,2; broad; A, A, AX
24, ꢀ46; 286, 298; AXX0
18
19, ꢀ15; broad; AX
21, ꢀ21; 300; AX
48, ꢀ77 (CP/MAS); broad; AX2
18, ꢀ146 (CP/MAS); 434; AX2
53, ꢀ68; 277, 275; AXX0; (This work)
15, ꢀ156; 439, 438; AXX0; (This work)
a
b
Me3PCl2, ref. 20. cyclo-P4Ph4, ref. 21.
Scheme 1
Scheme 2
Fig. 2 Thermal ellipsoid plot (50% probability) of [Me3PPMe2](3)[Cl]
in the solid state. H atoms are omitted for clarity. Selected bond lengths
(A) and angles (1): X-ray: P1–P2 = 2.1767(6), P1–C1 = 1.790(2), P1–C2/
C20 = 1.7855(15), P2–C3/C30 = 1.8329(15), C1–P1–C2/C20 = 108.9,
C2–P1–C20 = 106.3, C3–P2–C30 = 100.7, P1–P2–C3/C30 = 98.2,
P1–P2–Cl2 = 178.4, C1–P1–P2–C3/C30 = 51.1; cf. Calc. gas phase
(MP2/cc-pVTZ): P1–P2 = 2.190, P1–C1 = 1.808, P1–C2/C20 = 1.805,
P2–C3/C30 = 1.843, C1–P1–C2/C20 = 108.2, C2–P1–C20 = 108.4,
C3–P2–C30 = 100.9, P1–P2–C3/C30 = 98.2. C1–P1–P2–C3/C30 = 51.2.
calculated (MP2/cc-pVTZ) gas-phase structure of the cation
(Fig. 2 caption).
While the P–P bond length in 3 [2.1767(6) A, 2.190 A
(MP2)] is predictably shorter than that in the neutral
Me2PPMe2 [2.212(1) A,12 2.213 A (MP2)], it is surprisingly shorter
than that in the dication [Me3PPMe3][OTf]2 [2.198(2) A,11 2.227 A
(MP2)], although the differences are small. On the other
hand, comparisons of the P–C distances can be understood
in terms of the increasing cationic charge affecting a small but
significant shortening of the P–C bonds at the phosphonium
centers. The similarity of the P–C bonds in P2Me4 with those
at P2 in 3, and the fact that P–P distance is independent of
charge suggests that the cationic charge is accommodated by
the P–C bonds at the phosphonium centers. The effect is
replicated in the calculated structures of Me2PPMe2,
[Me3PPMe2]+ and [Me3PPMe3]2+ (see Supporting Information).
The nearly linear P1–P2–Cl2 angle (178.41) in the solid state
structure of 3[Cl] models the finale of the SN2 mechanism
envisaged for a Menschutkin process, featuring a disphenoidal
acceptor site (P2) with the incoming phosphine and the outgoing
halide on the same axis, analogous to the trigonal bipyramidal
acceptor carbon centre in the C–N Menschutkin reaction.13
The Born-Fajans-Haber cycle for the formation of [Me3PPMe2]-
[Cl] from Me2PCl and Me3P shown in Fig. 3 has been constructed
using experimentally measured14 enthalpies of vaporization (DHvap),
a calculated value for the chloride-ion-affinity of [Me2P]+
Fig. 1 31P CP/MAS solid state NMR spectrum of [Me3PPMe2][Cl],
* = spinning sidebands, w = impurity.
and PCl3 show only signals corresponding to starting materials,
even after prolonged heating.
Crystals of [Me3PPMe2][Cl] (3[Cl]) have been obtained from
a dilute CH3CN solution, and characterized by X-ray crystallo-
graphy as an ionic compound, as shown in Fig. 2. The crystals
and the powder do not melt, but both sublime in the same
temperature range. In the solid state, 3 adopts a staggered
conformation about the P–P bond. The closest interion P–Cl
distance (3.614 A) occurs at P2 and trans to P1 on the P1–P2
axis, but is significantly larger than the sum of the van der
Waals’ radii of P and Cl (3.55 A). Each chloride ion is stabilized
by five Cl–H contacts (2.760–2.860 A) involving the methyl
groups from three adjacent cations. The observed conformation,
bond lengths, and angles of 3 are in good agreement with the
c
7360 Chem. Commun., 2012, 48, 7359–7361
This journal is The Royal Society of Chemistry 2012

interpret the reaction as a bimolecular Menschutkin process,
driven by the DHP–P and DHlat of the phosphinophosphonium
salt and facilitated by the occupation of the P–Cl s*-MO in
the acceptor. We envisage that appropriate consideration of
these parameters will lead to broader application of the
Menschutkin approach to element–element bond formation.
In summary, formation of the prototypical phosphinophos-
phonium chloride salt by direct combination of a phosphine
and a chlorophosphine demonstrates the P–P Menschutkin
reaction, which has potentially broad applicability.
Fig. 3 Born-Haber-Fajans cycle for the formation of [Me3PP-
Me2](3)[Cl] from Me2PCl and Me3P (all values in kJ molꢀ1).
We thank the Natural Sciences and Engineering Research
Council of Canada, the Canada Research Chairs Program, the
Canada Foundation for Innovation and the Nova Scotia Research
and Innovation Trust Fund for funding and NMR-3 for use of
instrumentation.
Notes and references
1 N. Menschutkin, Z. Physik. Chem., 1890, 5, 589.
2 R. Reed, R. Reau, F. Dahan and G. Bertrand, Angew. Chem., Int.
Ed. Engl., 1993, 32, 399–401.
3 W. Seidel, Z. Anorg. Allg. Chem., 1964, 8, 141–150.
4 S. F. Spangenberg and H. H. Sisler, Inorg. Chem., 1969, 8,
1006–1010.
5 J. C. Summers and H. H. Sisler, Inorg. Chem., 1970, 9, 862–869.
6 J. A. Boon, H. L. Byers, K. B. Dillon, A. E. Goeta and
D. A. Longbottom, Heteroat. Chem., 3000, 11, 226–231.
7 K. B. Dillon and P. K. Monks, Dalton Trans., 2007, 1420–1424.
8 N. Burford and P. J. Ragogna, J. Chem. Soc., Dalton Trans., 2002,
4307–4315.
9 C. A. Dyker and N. Burford, Chem.–Asian J., 2008, 3, 28–36.
10 A. Schmidpeter, S. Loschmidt and W. S. Sheldrick, Angew. Chem.,
Int. Ed. Engl., 2003, 24, 226–227.
11 J. J. Weigand, S. D. Riegel, N. Burford and A. Decken, J. Am.
Chem. Soc., 2007, 129, 7969–7976.
Fig. 4 Pictorial representation of the LUMOs in (i) Me2PCl,
(ii) Ph2PCl at the MP2/cc-pVTZ level. View perpendicular to the sv
plane of symmetry. H-atoms omitted for clarity. See also Supporting
Information.
(DHCIA),15 a calculated value for the P–P bond strength
in [Me3PPMe2]+ (DHP–P), and a Volume-Based-Thermo-
dynamics16 calculation for the value of the lattice enthalpy
(DHlat), to give the overall reaction an exothermicity (DHrxn
)
of 14.2 kJ molꢀ1. The DHP–P value (356 kJ molꢀ1) for
[Me3PPMe2]+ is dramatically larger than that calculated for
P2Me4 (256 kJ molꢀ1, G3 level).17 The small magnitude of
DHrxn renders the overall thermodynamics sensitive to small
changes in DHP–P and DHlat and we speculate that these
variations are the reason that previous reports of reaction
mixtures containing longer-chained trialkylphosphines
(triethyl-, tri-n-butyl-, and tri-n-octyl-) and alkyl- or arylhalo-
phosphines described thermally unstable adducts and redox
decomposition products.3–5 In particular, the inverse relationship
between DHlat and the volume of the interacting ions makes
DHrxn less negative for larger trialkylphosphines The inclusion of
a strong Lewis acid into the mixture of Me2PCl and Me3P as a
halide abstractor will render DHrxn more negative as the positive
DHCIA of [Me2P]+ will be compensated by the negative DHCIA
of the abstractor (e.g. ꢀ321 kJ molꢀ1 for AlCl3).22
12 O. Mundt, H. Riffel, G. Becker and A. Simon, Z. Naturforsch. Teil
B., 1988, 43, 952.
13 S. Shaik, A. Ioffe, A. C. Reddy and A. Pross, J. Am. Chem. Soc.,
1994, 116, 262–273.
14 Me2PCl: A. B. Burg and P. J. Slota, J. Am. Chem. Soc., 1958, 80,
1107–1109Me3P: L. H. Long and J. H. Sackman, Trans. Faraday
Soc., 1957, 53, 1606–1611.
15 DHCIA for [Me2P]+ was calculated using the experimentally known
DHCIA for AlCl3 as
Information.
a reference enthalpy. See Supporting
16 H. D. B. Jenkins and L. Glasser, J. Chem. Eng. Data, 2011, 568,
874–880.
17 K. B. Borisenko and D. W. H. Rankin, J. Chem. Soc., Dalton
Trans., 2002, 3135–3141.
18 N. Burford, P. J. Ragogna, R. MacDonald and M. J. Ferguson,
Chem. Commun., 2003, 2066–2067.
19 D. Knackstedt, Investigations of Phosphenium insertion into
Phosphorus-Phosphorus Bonds, M.Sc. Dissertation, Dalhousie
University, Halifax, NS, 2011.
20 K. George, A. L. Hector, W. Levason, G. Reid, G. Sanderson,
M. Webster and W. Zhang, Dalton Trans., 2011, 40, 1584–1593.
21 M. Baudler, B. Carlsohn, W. Bohm and D. Reuschbach, Z. Naturforsch.
Teil B, 1976, 31, 558.
22 Y. U. Pervova, M. V. Korobov and L. N. Sidorov, Russ. J. Phys.
Chem., 1992, 66, 635.
The large value of DHCIA for [Me2P]+ (770 kJ molꢀ1
)
precludes heterolytic dissociation of Me2PCl as a plausible
step prior to P–P bond formation. Ab initio calculations
indicate that the LUMO for both Me2PCl and Ph2PCl is
the P–Cl s*-antibonding orbital (Fig. 4). On this basis, we
c
This journal is The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 7359–7361 7361