Base-Induced 1,3-Sigmatropic rearrangement of mesitylphosphonium salts

Article abstract of DOI:10.1002/ejic.201301203

Attempted synthesis of the ylide dianion [2,4,6-Me₃C ₆H₂P(CHR)₃]^2- (2,4,6-Me ₃C₆H₂ = mesityl, R = H or Me) by the reaction of mesitylphosphonium iodides [2,4,6-Me₃C₆H ₂PR₃]⁺I^- (R = Me, 1; R = Et, 2) with tBuLi at reflux does not result in the anticipated deprotonation of the phosphorus-bonded R groups. Instead, quantitative 1,3-sigmatropic rearrangement occurs to give new benzylic phosphonium salts [(3,5-Me₂C ₆H₃)CH₂PR₃]⁺I ^- (R = Me, 6; R = Et, 7), in which the phosphonium centre, the R ₃P group, is transferred to an ortho-CH₃ group. In situ ³¹P NMR spectroscopic studies show that the reaction is base-activated and stoichiometric with respect to tBuLi. DFT calculations support the conclusion that the rearrangement is thermodynamically favourable in the gas phase and in THF and show that the rearrangement is enthalpically driven. Copyright

Full text of DOI:10.1002/ejic.201301203

SHORT COMMUNICATION
DOI:10.1002/ejic.201301203
Base-Induced 1,3-Sigmatropic Rearrangement of
Mesitylphosphonium Salts
CLUSTER
ISSUE
Sophia A. Solomon,*[a] Lucy K. Allen,[a] Sarah B. J. Dane,[a] and
Dominic S. Wright*[a]
Keywords: Sigmatropic rearrangement / Phosphonium salts / Reaction mechanisms / Ylides
Attempted synthesis of the ylide dianion [2,4,6-Me3C6H2P-
(CHR)3]2– (2,4,6-Me3C6H2 = mesityl, R = H or Me) by the re-
action of mesitylphosphonium iodides [2,4,6-Me3C6H2PR3]+I–
(R = Me, 1; R = Et, 2) with tBuLi at reflux does not result in
the anticipated deprotonation of the phosphorus-bonded R
groups. Instead, quantitative 1,3-sigmatropic rearrangement
occurs to give new benzylic phosphonium salts [(3,5-
Me2C6H3)CH2PR3]+I– (R = Me, 6; R = Et, 7), in which the
phosphonium centre, the R3P group, is transferred to an
ortho-CH3 group. In situ 31P NMR spectroscopic studies show
that the reaction is base-activated and stoichiometric with re-
spect to tBuLi. DFT calculations support the conclusion that
the rearrangement is thermodynamically favourable in the
gas phase and in THF and show that the rearrangement is
enthalpically driven.
head (RЈ) and pendant (R) groups, aiming to improve the
solubility and the donor/acceptor character of the tripodal
ylide frameworks obtained on deprotonation. We note that,
apart from our own studies, almost nothing is known about
the reactivity of these key precursors. Relevant to the cur-
rent work, the synthesis of phosphonium salt [2,4,6-
Me3C6H2PMe3]+I– (1) was first reported around 80 years
ago but,[5] to the best of our knowledge, it has never been
explored further. In this paper we show that the bridgehead
(RЈ) group is not merely a spectator but that its choice is
highly important to the formation of stable tripodal ylide
ligands. Specifically, the presence of the mesityl function-
ality gives rise to a rearrangement, previously unknown in
phosphonium chemistry, in which the phosphonium centre
of the ligand is transferred quantitatively onto the ortho-
CH position in the presence of tBuLi.
Introduction
Applications of phosphorus ylides as ligands in inorganic
chemistry were first introduced by Schmidbaur, opening up
a rich area of transition and main group metal coordination
chemistry.[1,2] Our interest in this area has recently focused
on the triple deprotonation of the phosphonium salts
[RЈP(CH2R)3]+X– (X = a halide ion) to give ylides of the
type [RЈP(CHR)3]2– (Scheme 1).[3] Previously, we were able
to structurally authenticate the first (and currently the only)
example of this type of organometallic ligand, present in
the lithiate [{PhP(CH2)3}2(Li·THF)4].[3] We also showed
that the [PhP(CH2)3]2– dianion of the latter can be transfer-
red smoothly to other metal centres.[4] The metal-exchange
reaction with FeBr2 afforded the unusual hydride complex
[{PhP(CH2)3Fe}4(μ4-H)]–, composed of a tetrahedron of
FeII atoms with a μ4-H at the centre of the cluster.
Results and Discussion
A particular issue with the previously investigated
[PhP(CH2)3]2– ligand is the low solubility of the lithiate
[{PhP(CH2)3}2(Li·THF)] and the metal complexes formed
in metal–ligand exchange reactions.[6] This drawback has
thwarted further exploration of the coordination chemistry
of the [PhP(CH2)3]2– ligand by us to date. The starting point
in the current studies was therefore to provide related lithi-
ates containing a broader range of [RЈP(CHR)3]2– ligands
that have greater solubility, making them more useful pre-
cursors for metal-exchange reactions (particularly with re-
gard to obtaining crystalline products). This can be
achieved by modifying the bridgehead (RЈ) or pendant (R)
groups. In order not to crowd the ligands sterically in the
Scheme 1. Deprotonation of phosphonium cations [RЈP(CH2R)3]+
to give tripodal ylide ligands of the type [RЈP(CHR)3]2–.
Encouraged by these results, we have begun to synthesise
a range of phosphonium salts containing various bridge-
[a] Chemistry Department, Cambridge University,
Lensfield Rd. Cambridge, CB2 1EW, U.K
E-mail: dsw1000@cam.ac.uk
http://www.ch.cam.ac.uk/person/dsw1000
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejic.201301203.
Eur. J. Inorg. Chem. 2014, 1615–1619
1615
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
SHORT COMMUNICATION
domain of the three anionic C atoms, we elected to focus
Phosphonium salts 1 and 2 were fully characterised by
on the bridgehead position. Mesityl derivative [2,4,6- 1H, 13C and 31P NMR spectroscopy, mass spectrometry and
Me3C6H2P(CH2)3]2– was one obvious target since the mes- C, H elemental analysis. The H-coupled 31P NMR spec-
ityl group has long been employed to increase solubility and trum of 1 is particularly diagnostic, showing a 10-line bino-
aid crystallisation in ligand chemistry.
mial multiplet at δ = 20.1 (2JH–P = 13.7 Hz) ppm, while the
Phosphonium salt [2,4,6-Me3C6H2P(CH3)3]+I– (1) was spectrum of 2 consists of a broad multiplet at δ = 39.0 ppm.
easily obtained by a modification of the procedure outlined In addition, the single-crystal X-ray analyses of both 1 and
in the literature[5] in which 2,4,6-Me3C6H2PCl2 is reacted 2 were obtained in order to provide further, unequivocal
with MeLi (2 equiv.) in Et2O, the LiCl produced is filtered support for their structures prior to studies of their re-
off, and the filtrate is reacted directly with MeI (1 equiv.) arrangement with tBuLi. The structure of 1 shows no unex-
(Scheme 2a). It was obtained in 59% yield by crystallisation pected features and is presented in Figure 1.[9] The structure
of the residue from anhydrous methanol (see Exp. Sect.). of 2 can be seen in the Supporting Information.
This procedure is also readily applicable to the modifica-
tions of the pendant (R) groups, as shown by the synthesis
of [2,4,6-Me3C6H2PEt3]+I– (2) (80% crystalline yield) by
using EtLi and EtI in place of MeLi and MeI (see Support-
ing Information). The 2,4,6-Me3C6H2PCl2 starting material
is easily prepared by the reaction of 2,4,6-Me3C6H2Li with
PCl3 according to the method of Oshikawa et al.[7] How-
ever, the monosubstitution reaction cannot easily be con-
trolled where less sterically bulky groups are involved. In
these cases we have found that the reactions of RЈZnCl
(generated in situ from ZnCl2 and RЈLi at –78 °C) with
PCl3 offers a general route to the precursors RЈPCl2, which
can be used in an identical way, as previously mentioned,
to give phosphonium salts [RЈPR3]+I–. (Scheme 2b).[8]
Phosphonium salts [(4-tBuC6H4)PMe3]+I– (3), [(4-MeC6H4)-
PMe3]+I– (4) and [(2-MeC6H4)PMe3]+I– (5) were prepared
in this way (see Supporting Information).
Figure 1. Structure of the [2,4,6-Me3C6H2PMe3]+ cation of 1. Se-
lected bond lengths [Å] and angles [°]: P(1)–C(1) 1.806(3), P(1)–
C(10,11) range 1.786(3)–1.796(3); C–P–C range 103.3(1)–118.6(2).
Surprisingly, in preliminary small-scale studies, the lithi-
ation of 1 with tBuLi in THF (1:3 equiv., respectively) led
to the isolation of crystals of benzylic phosphonium salt
[(3,5-Me2C6H3)CH2PMe3]+I– (6). This was fully character-
ised by analytical and spectroscopic techniques. While (like
1
1) the H-coupled 31P NMR spectrum of 6 reveals an ap-
parent 10-line multiplet (in theory 12-line), this is found at
significantly more positive chemical shift (δ = 26.1 ppm)
than that in 1 (δ = 20.1 ppm). The presence of a benzylic
CH2 group is also obvious from the 13C and 1H NMR spec-
1
tra, which show doublets for the C (δ = 30.7, JC-P
=
2
50.3 Hz) and H atoms (δ = 4.10, JH-P = 16.00 Hz). The
X-ray structure of 6 is entirely consistent with these data
(Figure 2).[9]
The probable mechanism for this reaction is intramolecu-
lar on the basis of entropy and kinetics, and a direct indica-
tion of this is provided by the fact that a large reduction of
the concentration of the starting phosphonium salts has no
apparent effect on the reaction rate. However, we cannot
exclude a contribution from an intermolecular pathway at
this stage. The proposed intramolecular mechanism is out-
lined in Scheme 3. In Step 1, the deprotonation of the ortho-
Scheme 2. (a) Synthesis of 1 and 2, (b) general approach to phos-
phonium salts of the type [RЈP(R)3]+I–, starting from PCl3 (full
experimental details on the syntheses and characterisation of 1, 2,
3, 4 and 5 are found in the Supporting Information) and (c) all
phosphonium salts synthesised and the rearranged products.
Eur. J. Inorg. Chem. 2014, 1615–1619
1616
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
SHORT COMMUNICATION
Scheme 3. The proposed mechanism of the rearrangement reaction
producing 6 (R = Me) and 7 (R = Et).
Figure 2. Structure of the [(3,5-Me2C6H3)CH2PMe3]+ cation of 6.
Selected bond lengths [Å] and angles [°]: P(1)–C(7) 1.817(6), P(1)–
C(10,11,12) range 1.779(6)–1.793(6); C–P–C range 107.6(3)–
110.9(3), P(1)–C(7)–C(1) 109.6(4).
reactions of 3 and 4 with 3 equiv. of tBuLi in THF strongly
suggest that the ligands are deprotonated solely at the P–
Me groups, giving complexes of the type [{4-RЈЈC6H4-
P(CH2)3}(Li·THF)n]m {RЈЈ = tBu (8), RЈЈ = Me (9)}. Par-
methyl group occurs, followed by intramolecular attack of ticularly diagnostic are the upfield shifts of the doublet res-
the anionic C centre onto the phosphonium centre (Step 2). onances of the P(CH3)3 protons of 3 and 4 on addition of
The overall reaction is potentially base-catalysed, because tBuLi (moving from ca. δ = 2.60 to 1.52 ppm), which is
protonation of the final intermediate could be ac- associated with the build-up of negative charge on the CH2
complished by deprotonation of the initial phosphonium groups of the [RЈP(CH2)3]2– anions (see Supporting Infor-
salt. In an attempt to understand the mechanism further, a mation).[3] The fact that no rearrangement of the 2-Me
series of in situ 31P NMR spectroscopic studies were under- phosphonium salt 5 was also observed under these condi-
taken in THF as the solvent. These studies reveal that: (1) tions suggests that the rearrangement is highly substrate-
conversion of 1 into 6 does not occur (even with prolonged dependent.
reflux) without the addition of tBuLi; (2) complete conver-
Although this phosphonium rearrangement has, to the
sion occurs within 18 h at reflux, in THF; and (3) quantita- best of our knowledge, never been observed before, it
tive conversion of 1 into 6 requires a 1:1 stoichiometric ratio clearly bears a close relationship to a few previously re-
of tBuLi/1, and substoichiometric amounts of tBuLi result ported reactions involving ortho-C–H deprotonation of 2-
in an incomplete reaction (i.e., addition of 0.5 equiv. of methyl-substituted phenol, aniline and aryl Grignard rea-
tBuLi results in the rearrangement of only half of the 1 into gents.[11] Perhaps the closest of these is the rearrangement
6). These features strongly indicate that the reaction is not of 2-methylphenylmagnesium bromide into benzylmagne-
catalytic and that Steps 1 and 3 (Scheme 3) are not coupled. sium bromide.[11b]
In the absence of an added acid to complete the proton-
Hybrid DFT calculations were performed with the
ation in Step 3 of the reaction, the only possible proton B3LYP[12] functional and the LanL2DZ basis set[13] in or-
source available is the solvent (THF). Previous precedents der to examine the thermodynamics of the rearrangement
for the role of THF as a proton source in reactions with of the phosphonium cations of 1 and 2 into benzylic phos-
alkali metal organometallics are well established.[10] On the phonium cations 6 and 7, respectively. The results for the
basis of the above NMR spectroscopic studies, a prepara- gas-phase and solution-phase (THF) reactions are shown
tive-scale synthesis of 6 was developed (giving a 48% crys- in Table 1.
talline yield) by the 1:1 reaction of tBuLi/1 at reflux in THF
These calculations show (as expected) that the gas-phase
for 18 h (see Exp. Sect.). The crude solid residue from the and solution-phase reactions are thermodynamically
reaction is crystallised from EtOH (acting jointly as a sol- favourable, ΔG° becoming more favourable in THF. This
vent and proton source).
dependence on solvent polarity is perhaps unsurprising,
We were also able to show that the same rearrangement since the stability of the phosphonium cation reactants and
occurs in the case of phosphonium salt 2, which gives benz- products will be very sensitive to solvent–cation interac-
ylic phosphonium salt [(3,5-Me2C6H3)CH2PEt3]+I– (7) (in tions. Clearly the rearrangement is strongly enthalpically
55% crystalline yield after workup, see Supporting Infor- driven, ΔH° accounts for most of the thermodynamic fav-
mation). In situ 31P NMR spectroscopic studies show ex- ourability and ΔS° is small and positive in both the gas-
actly the same features as described previously for the re- phase and in THF. Interestingly, DFT calculations of the
1
arrangement of 1 into 6. In contrast, in situ H and 31P rearrangement of the 2-Me phosphonium derivative 5 show
NMR spectroscopic studies of the crude products of the a significant reduction in favourability, in line with the ex-
Eur. J. Inorg. Chem. 2014, 1615–1619
1617
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
SHORT COMMUNICATION
ringe and the crystals were dried and isolated. Yield (first batch)
5.07 g (59%). 1H NMR (400.14 MHz, CDCl3, +25 °C): δ = 2.29
[br. s, 3 H, p-CH3(mes)], 2.61 [br. s, 6 H, 2ϫ o-CH3(mes)], 2.65 [d,
Table 1. DFT calculations on the reactions 1 Ǟ 6 and 2 Ǟ 7 in the
gas phase and in THF.
Reaction[a] ΔG° [kJmol–1] ΔH° [kJmol–1] ΔS° [Jmol–1 K–1]
4
2JP-H = 12.00 Hz, 9 H, P(CH3)3], 6.98 (d, JP-H = 4.00 Hz, 2 H, Ar
CH) ppm. 13C{1H} NMR (100.62 MHz, CDCl3, +25 °C): δ = 15.9
Gas-phase
1
[d, JP-C = 54 Hz, P(CH3)3], 21.1 [br. s, p-CH3(mes)], 24.7 [s, o-
1 Ǟ 6
2 Ǟ 7
–48.6
–38.2
–41.7
–31.3
19.0
17.4
1
3
CH3(mes)], 115.5 [d, JP-C = 82 Hz, ipso-PC(mes)], 132.5 (d, JP-C
2
= 11 Hz, Ar CH), 142.4 [d, JP-C = 11 Hz, o-C(CH3)(mes)], 144.8
THF
[s, p-C(CH3)(mes)] ppm. 31P NMR (161.98 MHz, CDCl3, +25 °C,
rel. 85% H3PO4 in D2O): δ = 20.1 (binomial decet, 2JP-H = 13.7 Hz)
[31P(1H), δ = 20.1 (s)] ppm. Electrospray HR-MS (positive ion):
calcd. for [2,4,6-Me3C6H2P(CH3)3]+ 195.1303; found 195.1299.
[2,4,6-Me3C6H2P(CH3)3]+I– (1): calcd. C 44.7, H 6.3, found C 44.7,
H 6.3.
1 Ǟ 6
2 Ǟ 7
–62.1
–57.5
–53.0
–49.2
26.8
25.3
[a] The data in the gas phase concern the isolated cations, whereas
those in THF relate to the salts (including I–), all data is at
298.15 K.
6: To a solution of 1 (1.0 mmol, 0.32 g) in anhydrous THF (20 mL)
was added dropwise tert-butyllithium (1.0 mmol, 0.59 mL, 1.7 m
solution, pentanes) at –78 °C, and the mixture was stirred (30 min).
The reaction mixture was warmed to room temperature to give a
cloudy yellow solution. A reflux condenser was attached and the
reaction was heated to reflux (18 h), which resulted in a pale yellow
cloudy solution. The reaction was cooled to room temperature to
give a crop of fine colourless crystals. The solution was removed
by syringe, and the solid was dried under vacuum. Yield 78 mg
perimentally observed substrate dependence.[14] Further
theoretical studies are currently underway to probe the
mechanism of the reaction, particularly the transition state
and the intermediate involved in the second step of the reac-
tion (Scheme 3) and will be reported in a future full paper
on this work.
1
2
(48%). H NMR (400.14 MHz, CDCl3, +25 °C): δ = 2.17 [d, JP-
H = 16.00 Hz, 9 H, P(CH3)3], 2.31 [s, 6 H, m-CH3(Ar); 4.10, d, 2JP-
H = 16.00 Hz, 2 H, Ar-CH2P(CH3)3], 6.96 (br. s, 2 H, Ar CH), 6.99
(br. s, 1 H, Ar CH) ppm. 13C{1H} NMR (100.62 MHz, CDCl3,
+25 °C): δ = 8.92 [d, 1JP-C = 54.3 Hz, P(CH3)3], 21.2 [s, m-CH3(Ar)]
Conclusions
The primary conclusion of the current study is that mes-
itylphosphonium salts rearrange intramolecularly in the
presence of organometallic bases to give benzylic phos-
phonium salts. Although this new rearrangement may have
some synthetic uses, its main influence on the design of new
phosphonium precursors to the desired [RЈP(CHR)3]2– li-
gands is that the mesityl substituent should be avoided on
aryl bridgeheads (RЈ). Instead, using substituents (like
those present on the phosphonium salts 3–5) is a better
strategy for promoting greater solubility (without the dan-
ger of rearrangement).
1
2
30.7 [d, JP-C = 50.3 Hz, CH2P(CH3)3], 127.2 [d, JP-C = 9.0 Hz,
3
5
ipso-PC(Ar)], 127.6 (d, JP-C = 5.0 Hz, Ar CH), 130.4 (d, JP-C
=
4.0 Hz, Ar CH), 139.4 [d, JP-C = 4.0 Hz, m-C(CH3)(Ar)] ppm. 31
P
4
NMR (161.98 MHz, CDCl3, +25 °C, rel. 85% H3PO4 in D2O): δ =
2
26.1 (apparent 10-line multiplet, JP-H = 14.5, 12.9 Hz) [31P(1H), δ
= 26.1 (s)] ppm. Electrospray HR-MS (positive ion): calcd. for
[(3,5-Me2C6H3)CH2PMe3]+ 195.1381; found 195.1373. [(3,5-
Me2C6H3)CH2PMe3]+I– (6): calcd. C 44.7, H 6.3, found C 44.3, H
6.0.
CCDC-962443 (for 1), -962444 (for 2), -962446 (for 3), -962445 (for
4), -973078 (for 5), -962447 (for 6) and -962448 (for 7) contain the
supplementary crystallographic files for this paper. These data can
be obtained free of charge from The Cambridge Crystallographic
Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
Experimental Section
All syntheses were undertaken under dry, O2-free N2, on a vacuum
line. A glove-box was used to handle and store the products. The
phosphonium salts are not air-sensitive but are hygroscopic to vary-
ing degrees. The syntheses of 1 and 6 are given here, and the synthe-
ses and characterisation of 2, 3, 4, 5, 7, 8 and 9 are provided in the
Supporting Information. It can be noted that the literature pro-
cedure for the synthesis of 1 (ref.[5]) only gives a brief outline of the
method and no detailed characterisation.
Supporting Information (see footnote on the first page of this arti-
cle): 1H, 13C, 31P, 31P{1H} NMR spectra, electrospray HR-MS
(positive ion) of 1, 2, 3, 4, 5, 6 and 7. Syntheses of 2, 3, 4, 5, 7, 8
and 9.
Acknowledgments
1: To a solution of 2,4,6-Me3C6H2PCl2 (26.5 mmol, 5.87 g) in an-
hydrous Et2O (60 mL) was added dropwise methyllithium
(55.0 mmol, 35 mL, 1.6 m solution, diethyl ether) at –78 °C and the
mixture was stirred (15 min). The reaction was warmed to room
temperature and left to stir (18 h). The LiCl precipitate was filtered
(P4, Celite) to give a colourless, cloudy solution (the cloudiness is
LiCl, which is impossible to remove completely). The filtrate was
treated with iodomethane (26.5 mmol, 3.76 g, 1.65 mL) dropwise
at room temperature, with stirring, to give a precipitate, and the
reaction mixture was stirred (18 h). The diethyl ether was removed
under vacuum to give a white solid, replaced with anhydrous meth-
anol (50 mL), and any remaining undissolved solid was heated into
solution. The solution was stored at –30 °C (18 h) to give large
colourless, needle-like crystals. The solution was removed by sy-
We gratefully acknowledge the European Union [S. A. S., L. K. A.,
D. S. W., Advanced Investigator European Research Council
(ERC) award], Procter and Gamble, PLC and the Engineering and
Physical Sciences Research Council (EPSRC) (S. B. J. D., Industrial
Case Award) for financial support. We also thank Dr. J. E. Davies
for collecting X-ray data on 1, 2, 3, 4, 5, 6 and 7 and Dr. A. Sim-
perler and the National Service for Computational Chemistry Soft-
ware (NSCCS) for help with DFT calculations.
[1] H. Schmidbaur, Angew. Chem. 1983, 95, 980; Angew. Chem.
Int. Ed. Engl. 1983, 22, 907; H. Schmidbaur, Pure Appl. Chem.
1980, 52, 1057; H. Schmidbaur, Pure Appl. Chem. 1978, 50, 19;
H. Schmidbaur, Acc. Chem. Res. 1974, 7, 62.
Eur. J. Inorg. Chem. 2014, 1615–1619
1618
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
SHORT COMMUNICATION
[2] W. C. Kaska, Coord. Chem. Rev. 1983, 48, 1.
=
1445.03(9) (Å3), μ(Mo-Kα)
=
2.295 mm–1, ρcalc
=
[3] R. J. Less, V. Naseri, D. S. Wright, Organometallics 2009, 28,
1.481 Mgm–3, T = 180(2) K. Total reflections 10144, unique
3355 (Rint = 0.041). R1 = 0.076 [IϾ2σ(I)] and wR2 = 0.207
(all data).
3594.
[4] C. Berges Serrano, R. J. Less, M. McPartlin, R. E. Mulvey, V.
Naseri, D. S. Wright, Organometallics 2010, 29, 5754.
[5] W. C. Davies, J. Chem. Soc. 1935, 462.
[6] V. Naseri, S. B. J. Dane, S. A. Solomon, D. S. Wright, unpub-
lished results.
[10] P. Stanetty, H. Koller, M. Mihovilovic, J. Org. Chem. 1992, 57,
6833; J. Clayden, S. A. Yasin, New J. Chem. 2002, 26, 191;
A. R. Kennedy, J. Klett, R. E. Mulvey, D. S. Wright, Science
2009, 326, 707, and references therein.
[7] T. Oshikawa, M. Yamashita, Chem. Ind. (london) 1985, 126.
[8] The method was adapted from that described in E. Vedegs,
O. Daugulis, J. Am. Chem. Soc. 2003, 125, 4166 (Supporting
Information, Section SI-3).
[11] a) S. Meyerson, P. N. Rylander, E. L. Eliel, J. D. McCollum, J.
Am. Chem. Soc. 1959, 81, 2606; b) L. Zhang, S. Su, W. Hongp-
ing, S. Wang, Tetrahedron 2009, 65, 10022; c) P. Malik, D.
Chakraborty, V. Ramkumar, Polyhedron 2010, 29, 2142; d) P.
Malik, D. Chakraborty, V. Ramkumar, Polyhedron 2013, 32,
836.
[12] A. D. Becke, J. Chem. Phys. 1993, 98, 5648; S. H. Vosko, L.
Wilk, M. Nusair, Can. J. Phys. 1980, 58, 1200; C. Lee, W. Yang,
R. Parr, Phys. Rev. B 1988, 37, 785; P. J. Stephens, F. J. Devlin,
C. F. Chabalowski, M. J. Frisch, J. Phys. Chem. 1994, 98,
11623.
[9] Crystal data: Data were collected with a Nonius KappaCCD
diffractometer equipped with an Oxford cryostream low-tem-
perature device. Crystals were mounted directly from solution
using a perfluorohydrocarbon oil that freezes at low tempera-
ture (T. Kottke, D. Stalke, J. Appl. Crystallogr. 1993, 26, 615).
Data were solved by direct methods and refined by full-matrix
least-squares on F2 (G. M. Sheldrick, SHELX-97, Göttingen,
1997). 1: C12H20IP, M = 322.15, monoclinic, space group P21/
m, Z = 2, a = 8.2103(3) Å, b = 7.2465(3) Å, c = 11.9460(5) Å,
α = γ = 90°, β = 95.76(3)°, V = 707.15(5) Å3, μ(Mo-Kα) =
2.345 mm–1, ρcalc = 1.513 Mgm–3, T = 180(2) K. Total reflec-
tions 4670, unique 1732 (Rint = 0.037). R1 = 0.026 [IϾ2σ(I)]
and wR2 = 0.059 (all data). 6: C12H20IP, M = 322.15, mono-
clinic, space group P21/c, Z = 4, a = 15.5484(4) Å, b =
6.0821(2) Å, c = 15.7758(7) Å, α = γ = 90°, β = 104.395(2)°, V
[13] W. R. Wadt, P. J. Hay, J. Chem. Phys. 1985, 82, 284–298; P. J.
Hay, W. R. Wadt, J. Chem. Phys. 1985, 82, 270; P. J. Hay, W. R.
Wadt, J. Chem. Phys. 1985, 82, 299.
[14] Data for 5: gas-phase ΔG°
= =
–14.6 kJmol–1, ΔH°
–11.7 kJmol–1, ΔS° = 10.1 Jmol–1 K–1 and in solution ΔG° =
–26.7 kJmol–1, ΔH° = –21.4 kJmol–1, ΔS° = 15.4 Jmol–1 K–1.
Received: September 16, 2013
Published Online: December 9, 2013
Eur. J. Inorg. Chem. 2014, 1615–1619
1619
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim