Light-induced rearrangement of thioether-substituted phosphanide ligands: Scope and limitations of a remarkable isomerization

FULL PAPER

DOI: 10.1002/chem.201203918

Light-Induced Rearrangement of Thioether-Substituted Phosphanide

Ligands: Scope and Limitations of a Remarkable Isomerization

Keith Izod,* Ewan R. Clark, Pamela Foster, Rebecca J. Percival, Ian M. Riddlestone,

William Clegg, and Ross W. Harrington^[a]

Abstract: Treatment of the thioether-

substituted secondary phosphanes

of 1_PH with either PhCH₂Na or

PhCH₂K yields the heavier alkali metal

THF (1_Se)] as the sole products. The

compounds 3_Sa and 4_Sa may be desol-

vated to give the cyclic oligomers

R²PH(C₆H₄-2-SR¹) [R²=(Me₃Si)₂CH,

R¹=Me (1_PH), iPr (2_PH), Ph (3_PH);

R²=tBu, R¹=Me (4_PH); R²=Ph, R¹=

Me (5_PH)] with nBuLi yields the corre-

sponding lithium phosphanides, which

were isolated as their THF (1–5_Pa) and

tmeda (1–5_Pb) adducts. Solid-state

structures were obtained for the ad-

ducts [R²P(C₆H₄-2-SR¹)]Li(L)_n[R²=

(Me₃Si)₂CH, R¹=nPr, (L)_n=tmeda

(2_Pb); R²=(Me₃Si)₂CH, R¹=Ph, (L)_n=

tmeda (3_Pb); R²=Ph, R¹=Me, (L)_n=

(THF)_1.33(5_Pa); R²=Ph, R¹=Me,

(L)_n=([12]crown-4)₂(5_Pc)]. Treatment

complexes

[{(Me₃Si)₂CH}P(C₆H₄-2-

SMe)]M(THF)_n[M=Na (1_Pd),

N

K

[[{(Me₃Si)₂CH}P(Ph)

((3_S)₆) and [[tBuP(Me)

((4_S)₈), respectively.

ACHTUNGTRENNUNG

(1_Pe)]. With the exception of 2_Pa and

2_Pb, photolysis of these complexes with

white light proceeds rapidly to give the

ACHTUNGTRENNUNG

A

study reveals that the phosphanide–thi-

olate rearrangement proceeds by intra-

molecular nucleophilic attack of the

phosphanide center at the carbon atom

of the substituent at sulfur. For 2_Pa/2_Pb,

competing intramolecular b-deprotona-

tion of the n-propyl substituent results

in the elimination of propene and the

formation of the phosphanide–thiolate

thiolate

species

[R²P(R¹)

ACTHNGUTERN(NUG C₆H₄-2-

S)]M(L)_n[M=Li, L=THF (1_Sa, 3_Sa–

5_Sa); M=Li, L=tmeda (1_Sb, 3_Sb–5_Sb);

M=Na, L=THF (1_Sd); M=K, L=

Keywords: alkali metals · photoly-

sis · P ligands · rearrangement ·

sulfur

dianion [{(Me₃Si)₂CH}PACHTNUTGRNEUNG

(C₆H₄-2-S)]^2ꢀ.

Introduction

mediated polymerization of a-olefins;^[4]2) the palladium-

mediated copolymerization of ethylene with polar mono-

mers, such as acrylonitrile and vinyl acetate;^[5]3) the ruthe-

nium-mediated allylation of indoles;^[6]and 4) the rhodium-

catalyzed hydrogenation and hydrosilylation of unsaturated

amino esters and ketones.^[7]

In the large majority of these cases, the supporting ligands

possess an ortho-phenylene linker group between the phos-

phorus and sulfur functionalities. Usually, complexes con-

taining ligands of this type are thermally and photolytically

stable; however, in a small number of cases it has been re-

ported that certain sulfur-functionalized phosphanes under-

Over the last five decades, tertiary phosphanes have become

preeminent as ligands for the support of transition-metal-

based catalysts.^[1]This is due, at least in part, to the ease

with which their steric, electronic, and stereochemical prop-

erties may be modified. More recently, sterically hindered

phosphanes have been key to the discovery of an entirely

new field of chemistry associated with the formation of frus-

trated Lewis pairs.^[2]Such is the importance of phosphane

ligACHTUNGTRENNUNGands that there continues to be substantial interest both

in the development of new methods for their synthesis and

in their stabilities and reactions.

A relatively recent development in this area has been the

use of potentially hemilabile, sulfur-functionalized phos-

phane ligands for the support of catalytically-active transi-

tion-metal complexes.^[3]Several recent reports have shown

that tertiary phosphane–thiolate and tertiary phosphane–sul-

fonate ligands are useful supporting scaffolds for catalytic

transformations, including: 1) the nickel- or ruthenium-

ꢀ

go C S cleavage reactions when ligated to transition-metal

centers. For example, Lindoy and others have shown that

2-(diphenylphosphino)thioanisole complexes of Pd^IIor Pt^II

ꢀ

undergo S Me cleavage on heating under reflux for several

hours in DMF, to give the corresponding phosphane–thio-

late complexes (Scheme 1a).^[8]Similar demethylation reac-

tions have been observed on treatment of late-transition-

metal (Ni^II, Pd^II, or Pt^II) complexes of thioether-functional-

ized tertiary phosphanes with nucleophiles (for example,

Scheme 1b).^[9]A small number of C S cleavage reactions in

ꢀ

[a] Dr. K. Izod, Dr. E. R. Clark, P. Foster, R. J. Percival,

I. M. Riddlestone, Prof. W. Clegg, Dr. R. W. Harrington

Main Group Chemistry Laboratories, School of Chemistry, Bedson

Building, University of Newcastle, Newcastle upon Tyne, NE1 7RU

(UK)

related heteroatom-substituted thioether ligands have also

been reported.^[10]These reactions appear to be thermally

driven and typically do not result in a change of oxidation

state at the metal centers. In contrast, Darensbourg and co-

workers have shown that Ni⁰complexes of 2-(diphenylphos-

phino)thioanisole yield Ni^IIcomplexes of the corresponding

E-mail: k.j.izod@ncl.ac.uk

Supporting information for this article is available on the WWW

under http://dx.doi.org/10.1002/chem.201203918.

Chem. Eur. J. 2013, 00, 0 – 0

ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

&

1

&

ÞÞ

These are not the final page numbers!

significantly from the foregoing transition-metal-mediated

ꢀ

C S cleavage reactions in that 1) the reaction does not re-

quire the presence of external nucleophiles; 2) although

light-induced, the reaction does not involve oxidation of the

metal center; 3) the reaction does not appear to proceed via

a radical-based mechanism; and 4) the methyl group ap-

pears to migrate between the S and P atoms within the

same ligand.

To understand this reaction better, we have embarked on

a systematic investigation of the light-induced isomerization

of thioether-substituted phosphanide ligands; in this contri-

bution we demonstrate the generality of this reaction and il-

lustrate its scope and limitations. We also present evidence

to support our thesis that this reaction involves an intramo-

lecular, concerted migration of a substituent from sulfur to

phosphorus.

Scheme 1. Examples of transition-metal mediated phosphanide–thiolate

rearrangements.

thiolate ligand on exposure to light (Scheme 1c).^[11]This re-

action yields ethane as a side-product and is inhibited by the

radical scavenger TEMPO (2,2,6,6-tetramethylpiperidine-N-

oxide) and is therefore believed to involve the generation of

Results and Discussion

ꢀ

methyl radicals by a homolytic S Me cleavage reaction.

More recently, Pramanik and co-workers have reported the

Synthesis of secondary phosphane precursors: The thio-

ethers 2-BrC₆H₄SR¹(R¹=nPr (I), Ph (II)) were prepared by

ACHTUNGTRENNUNG

ꢀ

thermally induced C S cleavage of alkyl-2-(phenylazo)-

the reaction of 1,2-dibromobenzene with the corresponding

sodium thiolate in the polar solvent 1,3-dimethyl-2-imidazo-

lidinone; extended reaction times (typically 24 h) and high

temperatures were required to force these reactions to com-

pletion, but both I and II were isolated in good yield and

purity. Lithium–halogen exchange reactions between either

I or II and nBuLi at 08C generated the corresponding aryl-

lithium compounds, which were reacted in situ with

thioether ligands complexed to Rh^III.^[12]These rhodium-

mediated cleavage reactions generate a variety of organic

side-products, consistent with a radical-based mechanism;

this was attributed to PPh₃-induced reduction of the Rh^III

center to Rh^II, followed by elimination of an alkyl radical

and re-oxidation of the metal center to Rh^III(Scheme 2).

ꢀ

The cleavage of C S bonds is a key step in the hydro-

desulfurization of petroleum feedstocks, and so there is con-

[(Me₃Si)₂CH]PCl₂

to

give

the

chlorophosphanes

tinuing interest in the development of new methods for the

cleavage of C S bonds and in the mechanisms of these reac-

[(Me₃Si)₂CH]P(Cl)(C₆H₄-2-SR¹) (R¹=nPr (2_PCl), Ph

(3_PCl)); these were shown by H and ³¹P{¹H} NMR spectro-

1

ꢀ

tions.^[13]

scopy to be sufficiently clean for use without further purifi-

cation. Treatment of either 2_PCl or 3_PCl with LiAlH₄gave

the secondary phosphanes [(Me₃Si)₂CH]P(H)(C₆H₄-2-SR¹)

(R¹=nPr (2_PH), Ph (3_PH)) as colorless oils in good yield

(Scheme 4). Similarly, the reaction between commercially

available 2-BrC₆H₄SMe and nBuLi gave 2-LiC₆H₄SMe,

which was reacted in situ with tBuPCl₂, followed by LiAlH₄,

to give the secondary phosphane tBuP(H)(C₆H₄-2-SMe)

(4_PH) as a colorless oil.

ꢀ

Scheme 2. PPh₃-mediated, photoinduced C S bond cleavage.

We recently presented some preliminary results showing

that the thioether-functionalized lithium phosphanide

[{(Me₃Si)₂CH}P(C₆H₄-2-SMe)]Li

isomerization to the

(C₆H₄-2-S)]LiACTHUNGTERN(UNG tmeda) (1_Sb) on exposure

(tmeda) (1_Pb)^[14]undergoes

ACHTUNGTRENNUNG

corresponding

thiolate

[{(Me₃Si)₂CH}P(Me)

N

to ambient light (Scheme 3).^[15]This isomerization differs

Scheme 4. Synthesis of thioether-substituted phosphanes 2_PH–5_PH. Re-

agents and conditions: i) R¹SNa, 1,3-dimethyl-2-imidazolidinone, reflux,

24 h; ii) nBuLi, ꢀ788C, Et₂O, 2 h; iii) R²PCl₂, Et₂O, ꢀ788C, 12 h; iv)

LiAlH₄, Et₂O, reflux, 2 h; v) PhP(Cl){N

Et₂O, 1 h.

ACHTUGNRTEN(NUGN iPr)₂}, Et₂O, 12 h; vi) 2HCl,

Scheme 3. Photoinduced phosphanide–thiolate isomerization.

&

2

&

www.chemeurj.org

ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Chem. Eur. J. 0000, 00, 0 – 0

ÝÝ

These are not the final page numbers!

Thioether-Substituted Phosphanide Ligands

FULL PAPER

In contrast to the relatively straightforward syntheses of

THF cleanly give the corresponding alkali metal salts

2_PH–4_PH, reactions between PhPCl₂and one equivalent of

[{(Me₃Si)₂CH}P(C₆H₄-2-SMe)]MCAHTUNGRTNEN(GU THF) (M=Na (1_Pd), K

in situ-generated 2-LiC₆H₄SMe gave mixtures of the sec

G

(1_Pe)).

A

G

It has been suggested previously that the dealkylation of

transition-metal-coordinated thioethers proceeds by nucleo-

philic attack at the S-alkyl carbon atom and that the electro-

philicity of this center is enhanced by donation of one of the

sulfur lone pairs to the metal center, leading to a decrease

in electron density at the sulfur and adjacent alkyl carbon

centers.^[9]We attribute the demethylation observed for 1_PH

to a similar nucleophilic attack by nBuLi at the S-methyl

carbon of a P,S-chelating phosphanide ligand bound to a

lithium cation. The lack of any demethylation chemistry for

4_PH and 5_PH appears to be a consequence of the somewhat

reduced steric congestion at the phosphorus center in these

ligands. For 1_PH the large alkyl substituent at phosphorus in-

hibits deprotonation at this site, leading to competitive de-

methylation at sulfur, whereas deprotonation at phosphorus

is extremely rapid for 4_PH and 5_PH and so this is favored

over demethylation at sulfur.

phosphane PhP(C₆H₄-2-SMe)₂, along with unreacted

PhPCl₂, owing to the small size of the phenyl substituent.

This difficulty was overcome by protection of one of the

chloride substituents as its diisopropylamide derivative: The

reaction between PhP(Cl)

2-LiC₆H₄SMe proceeded cleanly to give PhP(C₆H₄-2-SMe)-

(NiPr₂). Deprotection with two equivalents of ethereal HCl

ACHTUNGTRENNUNG

and reduction with LiAlH₄gave the desired secondary phos-

phane PhP(H)(C₆H₄-2-SMe) (5_PH) in good yield as a color-

less oil.

Metalation reactions: We previously reported that the reac-

tion between [(Me₃Si)₂CH]P(H)(C₆H₄-2-SMe) (1_PH) and

one equivalent of nBuLi in diethyl ether results in concur-

rent deprotonation and demethylation to give the phospha-

nide–thiolate

(tmeda)}₂(1_PSb) after crystallization in the presence of

tmeda, leaving half an equivalent of 1_PH unreacted

(Scheme 5) (tmeda=N,N,N’,N’-tetramethylethylene-

ACHTUNGTRENNUNG

diamine).^[15]

complex

G

The effect of sulfur coordination is likely to be significant-

ly diminished in 2_PH, owing to the inductive effect of the

ethyl group adjacent to the SCH₂fragment, and in 3_PH,

where the carbon atom adjacent to sulfur is part of an aro-

matic ring. For these compounds, the electron-withdrawing

effect of the lithium cation in the phosphanide complexes

[{(Me₃Si)₂CH}P(C₆H₄-2-SR¹)]Li

ACHTNUTRGNE(UNG OEt₂)_nis likely to be some-

ꢀ

what decreased and so C S cleavage is disfavored with re-

spect to deprotonation at phosphorus, in spite of the large

substituent at the latter site. In accord with this, treatment of

either 2_PH or 3_PH with one equivalent of nBuLi in di

ACHTUNGTRENNUNGethyl

Scheme 5. Double lithiation of 1_PH to give 0.5 equivalents of 1_PSb.

ether proceeds smoothly to give the corresponding phos-

The present study reveals that this concurrent deprotona-

tion/demethylation reaction is not a general phenomenon,

but appears to be limited to 1_PH. For example, treatment of

either 4_PH or 5_PH with one equivalent of nBuLi in diethyl

ether gives pale yellow-orange solutions, the ³¹P{¹H} NMR

spectra of which contain signals that are due solely to the

phanide complexes [{(Me₃Si)₂CH}P(C₆H₄-2-SR¹)]Li

ACHTUNGTRENNUNG(OEt₂)_n

(R¹=nPr, Ph) as the sole phosphorus-containing products.

The ¹H, ¹³C{¹H}, ³¹P{¹H}, and ⁷Li NMR spectra of the

phosphanide complexes 1_Pa–5_Pa, 1_Pb–5_Pb, 1_Pd, and 1_Pe are

7

consistent with the above formulations. The ³¹P{¹H} and Li

NMR spectra of both 2_Pb and 3_Pb in C₆D₆exhibit a 1:1:1:1

quartet and a doublet, respectively, due to coupling between

these two nuclei (J_PLi=63.3 (2_Pb), 70.2 Hz (3_Pb)), suggesting

that P–Li exchange is slow on the NMR timescale for these

phosphanide complexes [R²P(C₆H₄-2-SMe)]Li

each case; signals due to the corresponding phosphanide–

thiolate dianions [R²P(C₆H₄-2-S)]^2ꢀwere not observed in

these spectra.

ACHTUNGTRNE(NUNG OEt₂)_nin

AHCTUNGTRENNUNG

1

compounds. The H, ¹³C{¹H}, and ³¹P{¹H} NMR spectra of

However, we now find that the course of the reaction of

1_PH with nBuLi is solvent-dependent: reactions between

any of the thioanisole-substituted phosphanes 1_PH, 4_PH, or

5_PH and one equivalent of nBuLi in the strongly coordinat-

ing solvent THF proceed cleanly to give the corresponding

trinuclear 5_Pa (see below) in [D₈]toluene possess a single set

of ligand signals, while a single signal is observed in the Li

7

NMR spectrum of this complex. This suggests either that

5_Pa exists as a single species containing one lithium environ-

ment in solution, or that this complex is subject to rapid dy-

namic equilibria such that a time-averaged set of signals is

observed for the two unique lithium centers and two ligand

environments. Crystals of 5_Pa, the related separated ion pair

[PhP(C₆H₄-2-SMe)][Li(12-crown-4)] (5_Pc), the tmeda ad-

ducts 2_Pb and 3_Pb, and the potassium salt 1_Pe were studied

by X-ray crystallography; details of their solid-state struc-

tures are given below.

ACTHNUTRGNEUNG

phosphanide complexes [R²P(C₆H₄-2-SMe)]Li(THF)_n(R²=

(Me₃Si)₂CH, n=2 (1_Pa); R²=tBu, n=2 (4_Pa); R²=Ph, n=

1¹/₃(5_Pa)) as the sole phosphorus-containing species. Addi-

tionally, treatment of 1_PH, 4_PH, or 5_PH with preformed

nBuLi

the tertiary amine-coordinated phosphanides [R²P(C₆H₄-2-

SMe)]Li

(tmeda) (R²=(Me₃Si)₂CH (1_Pb), tBu (4_Pb), Ph

ACHTUNGTRENNUNG(tmeda) in either diethyl ether or THF cleanly gives

ACHTUNGTRENNUNG

(5_Pb)). Phosphanide–thiolate dianions are not observed

under any of these reaction conditions. We also find that re-

actions between 1_PH and either PhCH₂Na or PhCH₂K in

Photolysis reactions: In our preliminary report, we showed

that irradiation of solutions of 1_Pb in toluene in borosilicate

Chem. Eur. J. 2013, 00, 0 – 0

ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemeurj.org

&

3

&

ÞÞ

These are not the final page numbers!

K. Izod et al.

ꢀ

glass vessels using a standard 20 W fluorescent lamp led to

age of S CACTHNGUTERNNUG

(sp²) bonds. Notable exceptions include the Ni^II-

mediated desulfurization of dibenzothiophenes reported by

clean isomerization to the corresponding thiolate 1_Sb

(Scheme 3).^[15]We now find that irradiation of the thioan

A

Jones and Vicic,^[16]the Cu^I-mediated cleavage of the

ꢀ

G

S C(Ar) bond of

a

polydentate pyrimidine-2-thiolate

ligand,^[17]and the Ru^II-mediated ring-opening of a pyridyl-

substituted dibenzothiophene,^[18]all of which reactions are

thermally induced.

G

ACHTUNGTRENNUNG

Scheme 6), suggesting that the photolytic rearrangement of

We find that irradiation of the S-phenyl-substituted com-

plex 3_Pa with a standard 11 W low-energy lamp results in

ꢀ

clean S CACHTUNGTRNEUNG

(sp²) cleavage and isomerization to the corre-

sponding thiolate [{(Me₃Si)₂CH}P(Ph)ACHTUNGTRENNUG(C₆H₄-2-S)]LiACHTUNGTRENNUNG(THF)_n

(3_Sa) (Scheme 6). Once again, after irradiation with white

light for 30 minutes, flame-sealed samples of 3_Pa in

[D₈]toluene exhibit no evidence in their ³¹P{¹H} and

¹H NMR spectra for species other than 3_Sa. We were unable

to obtain crystals of 3_Sa; however, this complex may be de-

Scheme 6. Phosphanide–thiolate rearrangement.

AHCTUNGTRENNUNG

thioanisole-substituted phosphanides is largely unaffected by

the nature of the remaining substituent at phosphorus, R².

However, the rate of isomerization appears to be a function

of the steric bulk of R², increasing with decreasing size of

this substituent in the order: 1_Pa<4_Pa<5_Pa; while isomeriza-

tion of 1_Pa takes ca. 60 h on exposure to natural daylight,

the isomerizations of 4_Pa and 5_Pa take approximately 2 h

and 20 min, respectively, under the same conditions.

crystals of this compound were readily obtained for analysis

by X-ray crystallography. Indeed, the photolysis and desol-

vation steps may be combined: simultaneous irradiation of

soACHTNURTGNENlUG uCAHUTGTNRENNUGtions of 3_Pa in toluene heated under reflux for 20 h leads

to essentially quantitative conversion to (3_S)₆, which precipi-

tates from the cooled solution as a colorless powder. It

should be noted, however, that isomerization of 3_Pa to 3_Sa

does not proceed in the absence of light, even when toluene

solutions are heated to 908C for an extended period.

We were similarly able to desolvate 4_Sa by heating solu-

tions of this complex under reflux in methylcyclohexane to

give the octameric cluster [{tBuP(Me)ACHTUNRTGNE(UNG C₆H₄-2-S)}Li]₈((4_S)₈).

Convenient access to (4_S)₈may be obtained by irradiation of

a solution of 4_Pa in methylcyclohexane while heating under

reflux for 12 h; details of the solid-state structures of (3_S)₆

and (4_S)₈are given below.

While seemingly dependent on the nature of the substitu-

ent at sulfur (see below), the phosphanide–thiolate re-

ꢀ

The majority of previously reported C S cleavage reac-

tions observed for phosphane–thioether ligands are thermal-

ly induced, although even these are relatively small in

number.^[8–10,12]However, the photolytic dealkylation of tran-

sition-metal-coordinated tertiary phosphane–thioether lig-

ACHTUNGTRENNUNGands has been reported by Darensbourg and co-workers

(see above).^[11]Such transition-metal-mediated reactions

differ markedly from the rearrangements described here:

the light-induced demethylation of (Ph₂PC₆H₄-2-SMe)₂Ni

(III) yields the bisACHTUNRGTNE(UNG thiolate) complex (Ph₂PC₆H₄-2-S)₂Ni;

that is, elimination of the S-methyl groups is accompanied

by oxidation from Ni⁰to Ni^II. Further investigation showed

that this reaction proceeds via radical intermediates: photol-

ysis of III in the presence of the radical scavenger TEMPO

yields 2,2,6,6-tetramethyl-1-piperidinylmethoxide, while pho-

tolysis of the ethyl-substituted analogue (Ph₂PC₆H₄-2-

SEt)₂Ni (IV) yields ethane and higher hydrocarbons as

ACHTUNGTRENaNUGN rrangement is relatively tolerant of the metal center, al-

though efficient rearrangement appears to require signifi-

ꢀ

cant ionic character in the M P bond. In this regard, we re-

cently reported that photolysis of the related diphosphager-

mylene and -stannylene compounds [{(Me₃Si)₂CH}P(C₆H₄-2-

SMe)]₂E (E=Ge, Sn),^[19]in which the E P bond is essential-

ꢀ

ly covalent, yields the diphosphane [{(Me₃Si)₂CH}P(C₆H₄-2-

SMe)]₂as the dominant product in each case, along with

either elemental germanium or tin (that is, reductive elimi-

nation is more rapid than phosphanide–thiolate rearrange-

ment for these compounds). In accordance with the forego-

ing, we find that irradiation of either of the sodium or potas-

major side-products.^[11]In contrast, the S Me cleavage ob-

ꢀ

served on irradiation of 1_Pa, 4_Pa, and 5_Pa is not accompanied

by metal redox chemistry, but may be envisaged as proceed-

ing via intramolecular attack of the phosphanide center at

ꢀ

the S Me carbon atom (see below for a more detailed in-

ꢀ

vestigation of the mechanism of this reaction). These reac-

tions do not generate the multiple side-products expected

for a radical-mediated reaction, but yield the corresponding

tertiary phosphane–thiolate as the sole product. In support

of this, EPR spectra of samples of 1_Pb irradiated in situ

within the cavity of the spectrometer provide no evidence

for the presence of radical intermediates.

sium salts 1_Pd or 1_Pe, in which the Na P and K P bonds

have predominantly ionic character, in THF with an 11 W

low-energy lamp leads to complete conversion to the corre-

sponding thiolate complexes [{(Me₃Si)₂CH}P(Me)

ACHTUNGTRENNUNG(C₆H₄-2-

S)]MACHTNGUTRNE(NUG THF)_n[M=Na (1_Sd), K (1_Se)] within 2 minutes.

1

The H, ¹³C{¹H}, and ³¹P{¹H} NMR spectra of the phos-

phane–thiolate complexes 3_Sa–5_Sa, 1_Sd, and 1_Se are largely

as expected. However, for (3_S)₆, 1_Sd, and 1_Se, the methine

proton of the (Me₃Si)₂CH substituent gives rise to a rather

ꢀ

While the cleavage of S C

limited, precedent, there are very few reports of the cleav-

(sp³) bonds has some, albeit

ACHTUNGTRENNUNG

&

4

&

www.chemeurj.org

ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Chem. Eur. J. 0000, 00, 0 – 0

ÝÝ

These are not the final page numbers!

Thioether-Substituted Phosphanide Ligands

FULL PAPER

low-field signal (d 1.91 ((3_S)₆), 2.61 (1_Sd), 2.65 ppm (1_Se);

compare with 0.71 (3_Pb), 0.24 (1_Pd), 0.25 ppm (1_Pe)). Exami-

nation of the crystal structure of (3_S)₆(see below) reveals

that the methine proton in this compound lies close to the

plane of the aromatic ring of the thiolate group and so

should experience a downfield shift due to a substantial ring

current effect, implying that a similar oligomeric structure is

adopted in solution. We therefore tentatively attribute the

unusual shifts of the methine protons in (3_S)₆, 1_Sd, and 1_Se to

chemical shift anisotropy associated with their proximity to

the aromatic ring of the thiolate ligand in each case. For the

heavier alkali metal compounds 1_Sd and 1_Se, this suggests

that an oligomeric structure is adopted, in which the meth-

phane–thiol and generating more secondary phosphanide,

which in turn isomerizes to the thiolate; thus, over time all

of the free secondary phosphane is consumed.

Unexpectedly, while 1_Pa, 3_Pa, 4_Pa, and 5_Pa undergo clean

isomerization to the corresponding thiolates, photolysis of

the n-propylthioether-substituted phosphanide complexes

[{(Me₃Si)₂CH}P(C₆H₄-2-S-nPr)]Li(L) (L=(THF)₂(2_Pa),

tmeda (2_Pb)) is not so straightforward. Irradiation of a deep

red solution of 2_Pb in [D₈]THF for 3 h with an 11 W low-

energy lamp yields a pale yellow solution, the ³¹P{¹H} NMR

spectrum of which exhibits major signals at ꢀ87.0 (A),

ꢀ28.7 (B), ꢀ20.9 (C), and ꢀ18.4 (D) ppm in the approxi-

mate ratio 4:2:6:1. Similar behavior is observed when C₆D₆

solutions of 2_Pb or 2_Pa are irradiated under these conditions.

By comparison with our previous report, we attribute

ACHTUNGTRENNUNGine protons are constrained to lie near the aromatic rings of

the ligands, and that such oligomers persist even in the

strong donor solvent THF. Unfortunately, we have so far

been unable to grow single crystals of 1_Sd and 1_Se suitable

for X-ray diffraction and so cannot confirm this hypothesis

for these two compounds.

peak A to the phosphanide–thiolate [{(Me₃Si)₂CH}P

ACHTUNGTRENNUNG(C₆H₄-2-

S)]{Li(tmeda)}₂(1_PSb); thus, photolysis of 2_Pb leads, in part,

AHCTUNGTRENNUNG

to dealkylation at sulfur. Although it is not possible to

assign the remaining signals unambiguously, their chemical

shifts indicate that these species are either secondary phos-

phanide–thiols or tertiary phosphane–thiolates, suggesting

that migration of the n-propyl group from sulfur to phos-

phorus competes with the n-propyl elimination reaction. Sig-

nal C is observed with very low intensity in the ³¹P{¹H}

NMR spectrum of the pre-photolysis sample of 2_Pb, which is

due to unavoidable exposure to light during sample loading

into the NMR spectrometer. As the appearance of this

signal is not accompanied by the appearance of signal A, we

tentatively attribute peak C to the expected alkyl migration

In the above photolysis experiments, we were somewhat

1

surprised to observe that signals in the H and ³¹P{¹H} NMR

spectra owing to traces of the free secondary phosphanes

R²P(H)(C₆H₄-2-SR¹), which are present due to a small

amount of hydrolysis during sample preparation, disap-

peared upon irradiation to be replaced by signals due to the

corresponding tertiary phosphane–thiol R²P(R¹)(C₆H₄-2-

SH). To confirm this observation, a 1:1 mixture of 1_PH and

1_Pa in toluene solution was subjected to photolysis and this

reaction was monitored by ³¹P NMR spectroscopy. After ir-

radiation of the solution with an 11 W low-energy lamp for

10 minutes, the initial signal in the ³¹P NMR spectrum owing

to 1_Pa (ꢀ64.2 ppm) was completely replaced by a new signal

at ꢀ37.1 ppm, due to the corresponding thiolate 1_Sa. At this

point the signal at ꢀ66.7 ppm due to 1_PH remained largely

product [{(Me₃Si)₂CH}PACHTUGNTERN(NNUG nPr)ACHTNURTGEG(NUNN C₆H₄-2-S)]LiACHTNUGTREN(GUNN tmeda) (2_Sb).

1

The H NMR spectrum of the post-photolysis sample exhib-

its a multitude of peaks due to n-propyl and aromatic

groups, along with a very low-intensity set of multiplets

lying between 4.89 and 5.83 ppm. The major signals in the

¹H spectrum again suggest that the n-propyl group remains

intact in one or more of the products, while the low-intensity

signals between 4.89 and 5.83 ppm correspond exactly to the

signals expected for propene; the low intensity of these pro-

pene signals is consistent with evolution of the majority of

the propene formed into the headspace of the sample, which

was sealed under vacuum.

unaffected and only

a small peak was observed at

ꢀ18.6 ppm, which is due to the tertiary phosphane–thiol

{(Me₃Si)₂CH}P(C₆H₄-2-SH) (1_SH). After irradiation for a

further hour approximately half of the phosphane 1_PH had

been converted into 1_SH, while after a further 5 h ir-

ACHTUNGTRENNUNGradiation, almost all of the initial quantity of 1_PH had been

converted. Irradiation of pure samples of 1_PH–5_PH, under

the same conditions used for the photolysis of the phospha-

nide derivatives 1_Pa, 3_Pa, 4_Pa, and 5_Pa, had no effect even

after several days exposure, clearly demonstrating that these

Based on the foregoing, we suggest the following mecha-

nism for the formation of 1_PSb from 2_Pb (Scheme 7). Attack

of the phosphido phosphorus center at the b-CH₂group of

the propyl substituent leads to elimination of propene and

secACHTUNGTRENNUNGondACHTUNGTRENNUNGary phosphanes are stable towards light.

We attribute the conversion

of the secondary phosphanes

into the corresponding tertiary

phosphane–thiols to a thiolate-

mediated reaction: photolysis

of

the

phosphanide

R²P(Li)(C₆H₄-2-SR¹) leads to

the formation of the thiolate

R²P(R¹)(C₆H₄-2-SLi). This is

sufficiently basic that it depro-

ACHTUNGTRENNUNGtonates the free secondary

phosphane, yielding free phos- Scheme 7. Mechanism for the formation of 1_PSb from 2_Pb. R²=CH

ACHTNUGTRENUNG(SiMe₃)₂, L=tmeda.

Chem. Eur. J. 2013, 00, 0 – 0

ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemeurj.org

&

5

&

ÞÞ

These are not the final page numbers!

K. Izod et al.

formation of the secondary phosphane–thiolate intermediate

[{(Me₃Si)₂CH}P(H)(C₆H₄-2-S)]Li(tmeda) (2_Sb’), which is im-

We showed above that the rate of the phosphanide–thio-

late rearrangement is diminished by increasing steric bulk of

the substituent at phosphorus. This competition experiment

also demonstrates that the rate of rearrangement is reduced

by increased steric bulk at the sulfur substituent: after ir-

N

ACHTUNGTRENNUNG

mediately consumed and so is not observed. We have shown

in our previous preliminary report, that secondary phos-

phanes of this type rapidly undergo deprotonation in the

presence of thiolates and that the resulting phosphanide–thi-

olate dianions are stable towards protonation even in the

presence of excess secondary phosphane (1_PH undergoes

concurrent demethylation and deprotonation, leaving half of

the starting phosphane intact). We therefore suggest that

2_Sb’ is rapidly deprotonated by 2_Sb (formed simultaneously

in the reaction) to give 1_PSb, yielding the tertiary phos-

AHCTUNTGRENNrGNU adiation of the mixed sample of 3_Pa and 4_Pa for 1 h we ob-

served complete conversion of 4_Pa into 4_Sa, while a signifi-

cant proportion of 3_Pa remained; after 2 h the rearrange-

ment was complete for both compounds.

We noted above that the rate of phosphanide–thiolate re-

arrangement increases with increasing ionic character of the

ꢀ

M P bond (that is, with increasing charge density at the

phane–thiol [(Me₃Si)₂CH]P

N

phosphorus center). Consistent with this, we observe that

the rearrangement of the potassium derivative 1_Pe proceeds

more rapidly than that of the corresponding lithium com-

plex. To investigate this phenomenon further, we studied

the rearrangement of the isolated anion [R²P(C₆H₄-2-

product; we attribute one of the two peaks B and D in the

³¹P{¹H} NMR spectrum to this latter species.

Mechanism of the phosphanide–thiolate rearrangement: On

the basis that the phosphanide–thiolate rearrangement reac-

tions of 1_Pa, 3_Pa, 4_Pa, and 5_Pa give single products and that

these reactions are EPR-silent, we may reasonably rule out

a radical-based mechanism. There are thus two feasible

mechanisms by which the reaction may proceed: 1) intramo-

lecular migration of a methyl group between the sulfur and

phosphorus centers within a ligand; and 2) intermolecular

migration of a methyl group from the sulfur center of one

ligand to the phosphorus center of a second ligand.

A kinetic experiment in which the concentration of 1_Pb

was monitored by ³¹P{¹H} NMR spectroscopy while irradiat-

ing with white light for fixed periods reveals the reaction to

be first-order in 1_Pb (Figure 1), clearly implying that the re-

action proceeds via an intramolecular mechanism. To con-

SR¹)]^ꢀ, prepared using cryptand

um counterion. Thus, addition of one equiv of cryptand-

[2.1.1] to a solution of 1_Pb in diethyl ether results in immedi-

ate precipitation of the putative separated ion pair

[{(Me₃Si)₂CH}P(C₆H₄-2-SMe)][Li(cryptand[2.1.1])] (1_Pf).

ACHTUNGTNER[NUNG 2.1.1] to sequester the lithi-

AHCTUNGTRENNUNG

A

ACHTUNGTRENNUNG

This complex was dissolved in [D₈]THF and its rearrange-

ment was monitored at 208C over time by ³¹P{¹H} NMR

spectroscopy, irradiating the sample for fixed periods when

outside the NMR spectrometer. At the same time we moni-

tored the rearrangement of the analogous tmeda complex

1_Pb under the same conditions. The resulting data (Figure 1)

clearly show that rearrangement is significantly faster for

the separated ion pair complex 1_Pf than for the contact ion

pair complex 1_Pb. This is consistent with the reaction pro-

ceeding by intramolecular nucleophilic attack of the phos-

phorus center at the carbon atom of the sulfur substituent

R², since charge will be localized on phosphorus to a greater

extent in 1_Pf than in the contact ion pair 1_Pb, and therefore

the phosphorus center will be more nucleophilic in the

former compound.

The rearrangement of 1_Pa, 3_Pa, 4_Pa, and 5_Pa under the con-

ditions described above necessarily yields racemic mixtures

of the corresponding chiral tertiary phosphane-functional-

ized thiolates. Attempts to prepare these compounds stereo-

selectively as a single enantiomer met with limited success;

however, these experiments provide further information re-

garding the mechanism of the rearrangement reaction. Ad-

dition of the naturally occurring chiral diamine (ꢀ)-spar-

teine to either 4_Pa or 5_Pa gives the corresponding complexes

[R²P(C₆H₄-2-SMe)]Li[(ꢀ)-sparteine] (R²=tBu (4_Pg), Ph

(5_Pg)), which were not isolated but which were generated

and used in situ. Irradiation of solutions of these adducts in

[D₈]toluene with an 11 W low-energy lamp for 12 h resulted

in quantitative conversion to the corresponding thiolates

&

^

Figure 1. Percentage conversion of 1_Pb to 1_Sb ( ) and 1_Pf to 1_Sf ( ) in

[D₈]THF on exposure to ambient fluorescent light at 208C.

firm this we irradiated a mixture of 3_Pa and 4_Pa in THF with

an 11 W low-energy lamp and monitored the reaction at set

intervals by ³¹P{¹H} NMR spectroscopy. After 2 h the spec-

trum of the irradiated sample exhibited signals at ꢀ17.9 and

ꢀ23.2 ppm, which may be assigned to the intramolecular mi-

gration products 3_Sa and 4_Sa, respectively, but no evidence

for the potential products of migration between phospha-

nide species, that is, no signals were observed for either

[(ꢁ)-R²P(Me)

ACHTUNGTRENNUNG

Ph (5_Sg)). These reactions are rather sluggish, possibly due

to the steric bulk of the sparteine co-ligand, and so extended

irradiation times were required for complete conversion.

The ³¹P{¹H} NMR spectrum of diastereomeric 4_Sg exhibits

signals at ꢀ17.1 and ꢀ21.8 ppm, owing to the two potential

[{(Me₃Si)₂CH}P(Me)

G

(THF)_n

(1_Sa)

or

[tBuP(Ph)(C₆H₄-2-S)]LiACHTGNUTRENNUNG

A

&

6

&

www.chemeurj.org

ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Chem. Eur. J. 0000, 00, 0 – 0

ÝÝ

These are not the final page numbers!

Thioether-Substituted Phosphanide Ligands

FULL PAPER

epimers at phosphorus, in an approximately 2:1 ratio. Simi-

larly, the ³¹P{¹H} NMR spectrum of 5_Sg exhibits signals at

ꢀ33.7 and ꢀ34.4 ppm in an approximately 2:1 ratio. These

sponding

tertiary

phosphane–thiols

{(Me₃Si)₂CH}P(Ph)(C₆H₄-2-SH) (3_SH) and tBuP(Me)(C₆H₄-

2-SH) (4_SH) as a colorless solid and a colorless oil, respec-

1

results clearly suggest that the presence of

G

tively. The H and ³¹P NMR spectra of 3_SH and 4_SH show

a real, if somewhat limited, effect on the stereo

A

these compounds are formed exclusively; in the ¹H NMR

spectra of 3_SH and 4_SH the SH proton is observed as a dou-

blet (J_PH=11.0 Hz) at 4.70 ppm and a broad singlet at

4.95 ppm, respectively.

come of the rearrangement reaction. This implies that the

lithium cation is associated with the phosphorus atom of the

phosphanide ligand during the rearrangement reaction and

is consistent with the slow rate of P–Li exchange suggested

by the ³¹P{¹H} and ⁷Li NMR spectra of 2_Pb and 3_Pb (see

above).

The photolysis reactions were carried out in standard

boroACHTUNGTRENNUNGsilicate glass vessels, which filter out the majority of

mid-to-high energy UV radiation, and occur at essentially

the same rate when the white light source is substituted by a

blacklight (Phillips 30 W fluorescent blacklight emitting at

420 nm; line width at half maximum height ca. 20 nm). The

In contrast, treatment of 1_Sb with two equivalents of ethe-

real HCl gives the corresponding protonated compound

1_SH’ as a colorless solid, which may be crystallized from hot

methylcyclohexane as colorless blocks in moderate yield.

The ³¹P NMR spectrum of 1_SH’ exhibits a sharp doublet at

1

ꢀ20.4 ppm (J_PH=406.1 Hz), while its H NMR spectrum ex-

hibits a very broad doublet centered at 6.90 ppm and with

the same coupling constant. This is consistent with protona-

tion at phosphorus rather than sulfur, implying formation of

the unusual zwitterionic tertiary phosphonium thiolate

phosphanide compounds 1–5_PACTHNUTRGENUGN(a–g) range from yellow to red

in color and this color dissipates on conversion to the corre-

sponding thiolate compounds. Consistent with this, the UV/

Vis spectrum of a freshly prepared sample of 1_Pb exhibits a

significant absorption at 257 nm, with a weak, broad should-

er at approximately 370 nm; after irradiation with an 11 W

low-energy lamp for 30 minutes, the major peak remains un-

changed, while the shoulder at 370 nm disappears. This

clearly suggests that the phosphanide–thiolate isomerization

is associated with an electronic transition occurring in the

violet to near-UV region of the spectrum. However, the

exact nature of this transition is, as yet, unclear.

[(Me₃Si)₂CH]P(Me)(H)ACTHUNGETRNNU(G C₆H₄-2-S) (1_SH’); in accord with

this, no peak corresponding to an SH proton was observed

in the ¹H NMR spectrum of this compound. The constitution

of 1_SH’ was confirmed by X-ray crystallography (see below).

It is clear that there is a delicate balance between proto-

nation at the sulfur or phosphorus sites of these phosphane–

thiolate anions; we note that protonation of 1_Sb with a

strong acid gives the zwitterion 1_SH’, whereas the alternative

protonation product 1_SH is formed during the photolysis of

1_Pa in the presence of 1_PH (see above).

We therefore conclude that the phosphanide–thiolate re-

arrangement outlined above is activated by absorption of

light in the violet to near-UV region and proceeds via a con-

certed intramolecular migration of the S-alkyl or S-aryl

Solid-state structures: i) phosphanide complexes: The struc-

tures of 2_Pb and 3_Pb (Figure 2) closely resemble that of the

lithium phosphanide 1_Pb. The asymmetric unit of 2_Pb con-

tains two independent molecules that differ in stereochemis-

try at the sulfur atoms: molecule 1 has P_SS_Rstereochemistry

whereas molecule 2 has P_SS_Sstereochemistry. As a conse-

quence, the bond lengths and angles of the two molecules of

2_Pb are rather dissimilar. Both diastereomers of each mole-

cule 1 and 2 occur in the centrosymmetric crystal structure.

In contrast, 3_Pb crystallizes with a single molecule in the

asymmetric unit, as a racemic mixture of P_RS_Rand P_SS_S

ꢀ

group from sulfur to phosphorus. Where P Li contacts are

possible (that is, with the exception of 1_Pf and 5_Pc), this

process occurs within the intact contact ion pair.

Protonolysis: As expected, treatment of 5_Sb with two equiv-

alents of ethereal HCl precipitates one equivalent of the salt

tmedaH⁺Cl^ꢀand yields the tertiary phosphane–thiol

PhP(Me)(C₆H₄-2-SH) (5_SH) as a colorless oil (Scheme 8).

stereoACTHNUTRGENUGiN somers.

In both compounds, each lithium ion is coordinated by

the P and S atoms of a chelating phosphanide–thioether

ligand, generating five-membered chelate rings (bite angles:

81.08(19) and 71.77(17)8 (2_Pb), 72.19(14)8 (3_Pb)). The coor-

dination about lithium is completed in each case by two ni-

trogen atoms of a molecule of tmeda. Furthermore, mole-

cule 1 of 2_Pb exhibits two short contacts between the lithium

atom and the methylene carbon atoms of the tmeda co-

ligand (Li1···C19 2.735(8), Li1···C20 2.692(9) ꢁ), while mole-

cule 2 exhibits a single short contact to one of the methylene

carbons of the tmeda co-ligand (Li2···C42 2.731(8) ꢁ); the

structure of 3_Pb exhibits two short Li···C(methylene) con-

tacts to the tmeda co-ligand (Li···C22 2.749(7), Li···C23

Scheme 8. Protonolysis of phosphane–thiolates 1_Sb, 3_Sb–5_Sb..

1

The H and ³¹P NMR spectra of this oil are consistent with

protonation at the sulfur atom; although we were unable to

observe the SH proton directly, owing to rapid exchange,

the proton-coupled ³¹P NMR spectrum consists of a sharp

singlet at ꢀ36.0 ppm, consistent with a tertiary phosphane

center. Similarly, hydrolysis of (3_S)₆and (4_S)₈gives the corre-

ꢀ

2.786(7) ꢁ). The Li P distances in 2_Pb (2.487(7) and

2.550(6) ꢁ) are somewhat dissimilar, but lie in the typical

range of such distances in lithium phosphanides,^[20]as does

Chem. Eur. J. 2013, 00, 0 – 0

ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemeurj.org

&

7

&

ÞÞ

These are not the final page numbers!

K. Izod et al.

ꢀ

um complexes of thioethers; the Li S distance in 3_Pb is

ꢀ

2.665(3) ꢁ. For comparison, the Li S distance in the related

phosphidoborane complex [{[(Me₃Si)₂CH]P(C₆H₄-2-SMe)-

ꢀ

(BH₃)}Li

in the cubane cluster [PhLi

2.635(4) ꢁ.^[24]

(THF)]₂is 2.608(2) ꢁ,^[23]while the Li S distances

(SMe₂)]₄range from 2.576(4) to

In contrast, 5_Pa crystallizes with an unusual trinuclear

structure containing two different lithium environments

(Figure 3). Li1 is coordinated by two P atoms from the adja-

cent phosphanide ligands and by the oxygen atoms of two

Figure 3. Molecular structure of 5_Pa with ellipsoids set at 40% probability

and hydrogen atoms and one THF disorder component omitted for clari-

ty. Selected bond lengths [ꢁ] and angles [8]: Li1–P1 2.576(5), Li1–P3

2.607(5), Li2–P1 2.514(5), Li2–P2 2.529(5), Li3–P2, 2.544(6), Li3–P3

2.516(6), Li2–S2 3.113(6), Li2–S1 2.626(5), Li3–S2 3.017(6), Li3–S3

2.620(6), Li1–O1 1.937(5), Li1–O2 1.982(5), Li2–O3, 1.929(6), Li3–O4

1.904(6); P1-Li1-P3 110.86(18), P1-Li2-P2 117.7(2), P3-Li3-P2 125.8(2),

Li1-P1-Li2 124.64(17), Li2-P2-Li3 111.89(18), Li1-P3-Li3 115.90(17), P1-

Li2-S1 75.34(14), P2-Li2-S2 63.78(13), P2-Li3-S2 65.19(14), P3-Li3-S3

79.40(16).

molecules of THF, affording a distorted tetrahedral geome-

try about this center. In contrast, Li2 and Li3 are each coor-

dinated by the P and S atoms of two chelating phosphanide

ligands (bite angles at Li2 63.78(13) and 75.34(14)8; bite

angles at Li3 65.19(14) and 79.40(16)8), and by the O atom

of a molecule of THF, affording a distorted trigonal bipyra-

midal geometry, in which the two S atoms occupy the axial

positions, at these centers. Thus, two of the phosphanide lig-

Figure 2. Molecular structures of a) the two independent molecules of

2_Pb, and b) 3_Pb, with ellipsoids set at 40% probability and hydrogen

atoms omitted for clarity. Short Li···CACTHNUGTRNEUNG(tmeda) contacts are not shown. Se-

lected bond lengths [ꢁ] and angles [8]: 2_Pb: Li1–P1 2.487(7), Li1–S1

2.527(7), Li1–N1 2.070(8), Li1–N2 2.061(8), C1–P1 1.885(4), C8–P1

1.803(4), C13–S1 1.783(4), Li2–P2 2.550(6), Li2–S2 2.625(7), Li2–N3

2.092(8), Li2–N4 2.068(7), C23–P2 1.883(4), C30–P2 1.812(4), C35–S2

1.763(5); P1-Li1-S1 81.08(19), P2-Li2-S2 71.77(17), N1-Li1-N2 90.5(3),

N3-Li2-N4 88.9(3). 3_Pb: Li1–P1 2.475(5), Li1–S1 2.665(5), Li1–N1

2.063(6), Li1–N2 2.067(6), C1–P1 1.887(3), C8–P1 1.805(3), C13–S1

1.784(3); P1-Li1-S1 72.19(14), N1-Li1-N2 89.0(2).

AHCTUNGTRENNUNG

Li2 and Li3 through both its P and S atoms, generating a

puckered six-membered (PLi)₃ring.

ꢀ

The P Li distances span the range 2.514(5)–2.607(5) ꢁ

ꢀ

the P Li distance in 3_Pb (2.475(5) ꢁ). For example, the

and are typical of Li

G

ꢀ

Li P distances in [Li

G

distances in 5_Pa span a much larger range: the Li2 S1 and

Li3 S3 distances to the simple chelating ligands are 2.626(5)

and 2.620(6) ꢁ and these distances are similar to the Li S

distances in 2_Pb and 3_Pb and in other thioether-coordinated

2.573(5) ꢁ,^[21]while the Li P distances in [Li(P{CH-

ꢀ

(Me₃Si)₂}₂)]₂are 2.481(10) and 2.473(9) ꢁ.^[22]Similarly, the

ꢀ

Li S distances in 2_Pb (2.527(7) and 2.625(7) ꢁ) differ by

0.1 ꢁ, but lie in the typical range for Li S distances in lithi-

ꢀ

lithium complexes. However, the Li2 S2 and Li3 S2 distan-

&

8

&

www.chemeurj.org

ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Chem. Eur. J. 0000, 00, 0 – 0

ÝÝ

These are not the final page numbers!

Thioether-Substituted Phosphanide Ligands

FULL PAPER

ces between the Li cations and the m₂-bridging sulfur atom

are 3.116(3) and 3.017(6) ꢁ, respectively.

Addition of two equivalents of [12]crown-4 to diethyl

ether solutions of 5_Pa yields the salt [PhP(C₆H₄-2-SMe)]

[Li([12]crown-4)₂] as a red solid, which precipitates from

solution. Recrystallization of this material from diethyl

ether/THF gave red blocks of the THF solvate [PhP(C₆H₄-2-

SMe)][Li([12]crown-4)₂]·THF (5_pc) suitable for X-ray crys-

tallography (Figure 4).

Figure 4. Structure of 5_Pc with ellipsoids set at 40% probability and hy-

drogen atoms and THF of crystallization omitted for clarity. Selected

bond lengths [ꢁ] and angle [8]: Li-O1 2.272(8), Li-O2 2.449(7), Li-O3

2.278(7), Li-O4 2.379(7), Li-O5 2.365(7), Li-O6 2.418(7), Li-O7 2.386(7),

Li-O8 2.382(7), P-C1 1.813(3), P-C13 1.829(3), S-C6 1.760(3); C1-P-C13

103.28(15).

Compound 5_Pc crystallizes as a separated ion pair. The

lithium cation is coordinated by the eight O atoms of two

molecules of [12]crown-4 in the typical manner. The two ar-

omatic rings of the isolated anion are not coplanar; the di-

hedral angle between the mean planes of these rings is

Figure 5. a) Asymmetric unit and b) polymeric structure of 1_Pe with ellip-

soids set at 40% probability and hydrogen atoms omitted for clarity. Se-

lected bond lengths [ꢁ] and angles [8]: K–P 3.3308(15), K–P’ 3.3014(15),

K–S 3.2575(16), K–S’ 3.3094(17), K–O 2.678(4), K–C6’’ 3.225(5), P–C1

1.890(4), P–C8 1.808(5), S–C13 1.780(5); P-K-S 56.09(3), P’-K-S’ 55.89(3),

P-K-P’ 125.45(4), S-K-S’ 88.20(4). The primed and double-primed atoms

are in adjacent units generated by symmetry.

ꢀ

52.48. The P C1 and P C13 distances are 1.813(3) and

1.829(3) ꢁ, respectively, and these distances are similar to

ꢀ

the corresponding P C distances in 5_Pa (P C(Ph) 1.819(3)–

ꢀ

1.845(3) ꢁ; P C(Ar) 1.805(3)–1.822(3) ꢁ), suggesting that

ii) Thiolate complexes: Compound (3_S)₆crystallizes with a

coordination of the anion to lithium has little impact on the

degree of delocalization of charge from the phosphorus

center to the aromatic rings of the phosphanide ligand.

The potassium salt 1_Pe crystallizes as a chain polymer in

which adjacent potassium ions are bridged by the phospho-

rus and sulfur atoms of a phosphanide–thioether ligand

(Figure 5). The coordination about each potassium ion is

completed by the oxygen atom of a molecule of THF and by

hexagonal prism structure with crystallographic three-fold

¯

inversion (3=S₆) symmetry, in which each lithium ion is co-

ordinated by three m₃-bridging thiolate sulfur atoms and by

one phosphorus atom (Figure 6). Each phosphane–thiolate

ligand thus chelates a lithium ion (bite angle 80.43(19) ꢁ)

such that each lithium ion adopts a distorted tetrahedral

geometry.

The hexagonal prism motif adopted by this complex is

rather unusual for a lithium thiolate, although this motif has

been observed in several lithium alkoxides. To the best of

our knowledge, the only other lithium complex to crystallize

with such a Li₆S₆hexagonal prismatic core is the thioamidi-

ꢀ

a short K···Me Si contact (K···C6’’ 3.225(5) ꢁ).

The K P distances (3.3014(15) and 3.3308(15) ꢁ) are sim-

ilar to previously reported K P contacts; for example, the

ꢀ

[20]

ꢀ

K P distances in [{(mes*)PH}K]₁are 3.271(2), 3.181(2),

and 3.357(2) ꢁ,^[25]while the K P distances in

(nBu)}]₆(8),^[28]which adopts a paddle-

nate [Li{CSACHTNGUTRENNU(G NtBu)ACHTUNGTRENNUNG

ꢀ

[{(tBuPhP)K}₂(py)]₁

range

from

3.2324(14)

to

wheel structure in which the thioamidinate ligands bridge

the six edges which connect the Li₃S₃rings. In contrast, the

phosphane–thiolate ligands in (3_S)₆bridge alternate edges

around each of the Li₃S₃rings.

3.3727(14) ꢁ.^[26]The K S distances (3.3094(17) and

ꢀ

3.2575(16) ꢁ) are also similar to previously reported

ꢀ

K S(thioether)

distances;

(2-KO)

for

example,

the

ꢀ

K S distance in [1,1’-S

G

E

The Li S distances in (3_S)₆(2.483(7), 2.429(7), and

(Et₂O)] is 3.3145(7) ꢁ.^[27]

2.530(6) ꢁ) are similar to the corresponding distances in 8

Chem. Eur. J. 2013, 00, 0 – 0

ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemeurj.org

&

9

&

ÞÞ

These are not the final page numbers!

K. Izod et al.

Figure 7. a) Molecular structure of (4_S)₈with ellipsoids set at 40% proba-

bility and hydrogen atoms and minor disorder components omitted for

clarity. b) The Li₈S₈P₈core of (4_S)₈. The molecule has a crystallographic

C₂axis, which runs horizontally in (a).

Figure 6. a) Molecular structure of (3_S)₆with ellipsoids set at 40% proba-

bility and hydrogen atoms omitted for clarity. b) The Li₆S₆P₆core of

(3_S)₆. Selected bond lengths [ꢁ] and angles [8]: Li–P 2.653(7), Li–S

2.429(7), Li_A–S 2.483(7), Li_B–S 2.530(6), P–C1 1.855(4), P–C8 1.848(5),

S–C19 1.766(4); Li-S-Li_A124.6(3), S-Li-S_D95.5(2), Li-S-Li_B85.4(2), Li_A-

S-Li_B84.2(2), S-Li-S_C112.7(3), S-Li-P 80.43(19), S_D-Li-P 137.1(3), S_C-Li-

P 127.0(3). Suffix letters A–D denote symmetry-related atoms.

in which the phosphanide ligands bridge alternate edges of

each Li₄S₄octagon. To the best of our knowledge, this is the

first time that this structural motif has been observed in

alkali metal chalcogenolate chemistry. Indeed, there is little

precedent for this structural motif across the periodic table:

Kawamoto and co-workers have reported the structures of

four modifications of related Cu^Icomplexes [Cu{SC₆H₄-2-

ꢀ

(Li S 2.380–2.660(7) ꢁ) and lie in the typical range for

(thiolate) bonds.^[29]For example, the Li S distances in

ꢀ

Li S

ꢀ

the two independent molecules of (2,4,6-tBu₃C₆H₂S)Li-

(THF)₃are 2.449(11) and 2.459(12) ꢁ.^[30]The Li P distance

ꢀ

(2.653(7) ꢁ) falls in the usual range of distances for bonds

N=CHACTHGNUTERNNUG

(C₄H₄-4-X)}]₈(X=OMe, NMe₂, CF₃),^[32]although in

between lithium and a tertiary phosphane center;^[20]for ex-

these complexes the ligands bridge the edges of the prism

between the two octagonal faces in a paddle-wheel fashion,

and Mulvey and co-workers have reported the structure of

the lithium primary amide [(tBuNH)Li]₈.^[33]

ꢀ

ample, the Li P distance in [Li{OC

ACHTNUGTNER(UNNG tBu)₂CH₂PPh₂}]₂is

2.651(6) ꢁ.^[31]

Compound (4_S)₈crystallizes with an unusual octagonal

prismatic structure (Figure 7). Unfortunately, despite several

attempts to grow crystals of suitable quality and despite the

collection of several data sets, the structure obtained for

(4_S)₈was of relatively poor quality. Nevertheless, the overall

structure and the connectivity of the atoms are unambigu-

ous. Each lithium ion is coordinated by the phosphorus and

sulfur atoms of a chelating phosphanide ligand and by the

sulfur atoms of two adjacent phosphanide ligands. Thus, the

sulfur atom of each phosphanide ligand acts as a m₃bridge

between the lithium cations, generating a Li₈S₈Ferris wheel,

Crystals of the zwitterionic compound 1_SH’ were obtained

by protonolysis of 1_Sb with ethereal HCl. This compound

crystallizes as a discrete molecular species (Figure 8); the

hydrogen atoms were located in a difference map and the H

atom bound to phosphorus was freely refined. There is no

short contact between this H atom and the sulfur center: the

aromatic ring is tilted such that the sulfur atom is directed

away from the phosphorus-bound hydrogen atom and the

S···H distance (2.96 ꢁ) is significantly greater than the sum

of the van der Waals radii of S and H (1.89 ꢁ).^[34]The S C-

ꢀ

&

10

&

www.chemeurj.org

ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Chem. Eur. J. 0000, 00, 0 – 0

ÝÝ

These are not the final page numbers!

Thioether-Substituted Phosphanide Ligands

FULL PAPER

Experimental Section

All manipulations were carried out using standard Schlenk techniques

under an atmosphere of dry nitrogen. Diethyl ether, THF, light petro-

AHCTUNGTRENNUNG

all other solvents were stored over a potassium film. Deuterated benzene,

toluene and THF were distilled from potassium, CDCl₃was distilled

from CaH₂; all NMR solvents were deoxygenated by three freeze–pump–

thaw cycles and were stored over activated 4 ꢁ molecular sieves. NaS-

nPr was prepared by the reaction of nPrSH with NaH in diethyl ether,

[{(Me₃Si)₂CH}P(C₆H₄-2-SMe)]LiACTHNUTRGENNUG CAHTUNGTRENNUGN

(tmeda)^[15]and PhPCl{N(iPr)₂}^[35]were

prepared by previously published procedures; n-butyllithium was pur-

chased from Aldrich as a 2.5m solution in hexanes while anhydrous HCl

was purchased from Aldrich as a 2.0m solution in diethyl ether. All other

compounds were used as supplied by the manufacturer.

Figure 8. Molecular structure of 1_SH’ with ellipsoids set at 40% probabili-

ty and C-bound hydrogen atoms omitted for clarity. Selected bond

lengths [ꢁ]: P1–H1 1.299(16), P1–C1 1.7901(14), P1–C8 1.7805(15), P1–

C14 1.7921(15), S1–C13 1.7293(15).

Photolysis experiments were carried out by irradiating a solution of phos-

phanide complex in toluene in a standard borosilicate glass Schlenk with

a domestic Phillips Genie 11 W low-energy fluorescent lamp (color tem-

perature 2700 K, luminous flux 610 Lm) at a horizontal distance of ap-

proximately 10 cm.

¹H and ¹³C{¹H} NMR spectra were recorded on a JEOL ECS500 spec-

trometer operating at 500.16 and 125.65 MHz, respectively, or a Bruker

Avance300 spectrometer operating at 300.15 and 75.47 MHz, respective-

ly; chemical shifts are quoted in ppm relative to tetramethylsilane.

³¹P{¹H} and ⁷Li{¹H} NMR spectra were recorded on a JEOL ECS500

spectrometer operating at 202.35 and 194.38 MHz, respectively; chemical

shifts are quoted in ppm relative to external 85% H₃PO₄and 1.0m LiCl,

respectively. Elemental analyses were obtained by the Elemental Analy-

sis Service of London Metropolitan University.

ACHTUNGTRENNUNG(aryl) distance (1.7293(15) ꢁ) is somewhat shorter than the

corresponding distances in 1_Sc and (3_S)₆(1.778(4) and

1.766(4) ꢁ, respectively), consistent with greater multiple

bond character in the former.

Conclusion

The present study demonstrates that the light-induced isom-

ACHTUNGTRENNUNGerization of ortho-phenylene-bridged phosphanide–thioether

ligands to the corresponding tertiary-phosphane–thiolates is

a general phenomenon. This highly unusual isomerization is

Representative syntheses (full details of all synthetic procedures and

characterization data may be found in the Supporting Information):

ꢀ

accelerated with increasing ionic character of the P M bond

Synthesis of a 2-bromophenylthioether. 2-(nPrS)C₆H₄Br (I): 1,2-dibromo-

benzene (3.86 mL, 7.55 g, 32.01 mmol) was added to a solution of NaS-

nPr (3.00 g, 30.56 mmol) in 1,3-dimethyl-2-imidazolidinone (40 mL). This

mixture was heated to 1008C for 24 h and then allowed to cool to room

of the phosphanide starting material and is inhibited by in-

creased steric bulk at either the phosphorus or sulfur cen-

ters. Mechanistic studies reveal that the isomerization is

first-order with respect to the phosphanide and indicate that

this process occurs through the intramolecular migration of

a substituent from sulfur to phosphorus in the intact ion

pair. The isomerization proceeds cleanly for both alkyl and

aryl substituents at sulfur, although a competing alkene

elimination pathway is possible for sulfur substituents pos-

sessing a b-hydrogen atom, leading to the formation of

phosphanide—thiolate dianion side-products.

temperature. The supernatant was deACTHNUTRGNEUGNcanted and the solid residue was ex-

tracted into light petroleum (3ꢂ20 mL). The combined organic extracts

were washed with water, the aqueous phase was extracted with diethyl

ether (2ꢂ30 mL), and the combined organic extracts were dried over

MgSO₄. The solvent was removed in vacuo to yield a yellow oil, which

was purified by distillation (1008C/10^ꢀ2mmHg). Yield 4.18 g, 59.2%.

¹H NMR (CDCl₃): d 1.06 (t, J_HH=7.8 Hz, 3H, CH₂Me), 1.72 (m, 2H,

CH₂Me), 2.90 (t, J_HH=7.1 Hz, 2H, CH₂S), 6.99 (m, 1H, ArH), 7.23 (m,

2H, ArH), 7.52 (m, 1H, ArH). ¹³C{¹H} NMR (CDCl₃): d 13.59 (CH₂Me),

21.88 (CH₂Me), 34.78 (CH₂S), 123.16, 126.13, 127.50, 127.61, 132.87,

138.48 (Ar).

The light-induced phosphanide–thiolate rearrangement of

alkali metal complexes of ortho-phenylene-bridged phospha-

nide–thioether complexes described above differs from pre-

vious reports of related isomerizations in transition metal

complexes of this ligand type in three important respects:

1) migration of substituents occurs within the ligand, rather

than between ligands; 2) the isomerization is light-induced,

but does not involve the intermediacy of radicals; and 3) the

isomerization does not involve reduction or oxidation of the

metal center.

Synthesis

of

a

thioether-substituted

secondary

phosphane.

{(Me₃Si)₂CH}PH(C₆H₄-2-S-nPr) (2_PH): nBuLi (4.8 mL, 12.00 mmol) was

added to a cold (08C) solution of 2-(nPrS)C₆H₄Br (2.75 g, 11.90 mmol) in

diethyl ether (20 mL) and this mixture was stirred at 08C for 3 h. The

pale yellow solution was added dropwise to a cold (ꢀ788C) solution of

{(Me₃Si)₂CH}PCl₂(3.43 g, 13.13 mmol) in diethyl ether (20 mL). This

mixture was allowed to attain room temperature and was stirred for 16 h.

The pale precipitate was removed by filtration and solid LiAlH₄(0.50 g,

13.17 mmol) was carefully added to the filtrate. This mixture was heated

under reflux for 2 h and then allowed to cool to room temperature. De-

gassed brine (40 mL) was carefully added and the organic phase was de-

X-ray crystallography reveals that both the phosphanide

starting materials and their thiolate rearrangement products

adopt a wide range of structural motifs in the solid state, in-

cluding monomers, trimers, hexamers, and octamers.

AHCTUNGTREGcNNUN anted. The aqueous phase was extracted into light petroleum (3ꢂ

20 mL) and the combined organic components were dried over activated

4 ꢁ molecular sieves. The solution was filtered and solvent was removed

in vacuo from the filtrate to give 2_PH as a colorless oil. Yield 2.52 g,

62.0%. ¹H NMR (CDCl₃): d 0.05 (s, 9H, SiMe₃), 0.10 (d, J_PH=1.0 Hz,

Chem. Eur. J. 2013, 00, 0 – 0

ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemeurj.org

&

11

&

ÞÞ

These are not the final page numbers!

K. Izod et al.

9H, SiMe₃), 0.76 (dd, J_HH=6.5 Hz, J_PH=1.8 Hz, 1H, CHP), 1.04 (t, J_HH

=

1.06 mmol) in THF (10 mL). This mixture was stirred at room tempera-

ture in the absence of light for 15 min. Solvent was removed in vacuo

and the sticky red solid was dissolved in toluene (15 mL). This solution

was heated under reflux whilst illuminating with an 11 W low-energy

lamp for 20 h. The pale solids were isolated by filtration and the solid

was washed with a little light petroleum. Residual solvent was removed

in vacuo to give (3_S)₆as a colorless powder. Single crystals of (3_S)₆suit-

7.3 Hz, 3H, CH₂Me), 1.70 (m, 2H, CH₂Me), 2.92 (m, 2H, CH₂S), 4.49

(dd, J_PH=216.6 Hz, J_HH=6.5 Hz, 1H, PH), 7.11 (m, 1H, ArH), 7.22 (m,

1H, ArH), 7.27 (m, 1H, ArH), 7.45 (m, 1H, ArH). ¹³C{¹H} NMR

(CDCl₃): d 0.84 (SiMe₃), 5.43 (d, J_PC=0.5 Hz, CHP), 13.69 (CH₂Me),

22.43 (CH₂Me), 36.20 (CH₂S), 125.31, 128.27, 128.46 (Ar), 133.61 (d,

J

_PC=12.3 Hz, Ar). ³¹P NMR (CDCl₃): d ꢀ66.3 (d, J_PH=216.6 Hz).

AHCTUNGTERGaNNUN ble for X-ray crystallography were grown by carefully layering a saturat-

Synthesis

[{(Me₃Si)₂CH}P(C₆H₄-2-S-nPr)]LiAHCUTNGTRENNUNG

of

a

thioether-substituted

(tmeda)

lithium

phosphanide.

(0.5 mL,

ed solution in diethyl ether with n-hexane; colorless blocks deposited

within 1 week. Yield 0.31 g, 76%. ¹H NMR ([D₈]THF): d ꢀ0.15 (s, 9H,

SiMe₃), 0.01 (s, 9H, SiMe₃), 1.91 (s, 1H, CHP), 6.56 (m, 1H, ArH), 6.66

(m, 1H, ArH), 6.99 (m, 1H, ArH), 7.13 (m, 3H, ArH), 7.23 (m, 1H,

(2_Pb): nBuLi

1.25 mmol) was added to a solution of 2_PH (0.44 g, 1.28 mmol) and

tmeda (0.20 mL, 0.15 g, 1.33 mmol) in diethyl ether (15 mL), excluding

light as far as possible. The deep red solution was stirred for 1 h and then

solvent was removed in vacuo. The sticky red solid was crystallized from

cold (ꢀ308C) n-hexane as dark orange blocks suitable for X-ray crystal-

lography. Yield 0.45 g, 75.6%. ¹H NMR (C₆D₆): d 0.42 (s, 18H, SiMe₃),

0.63 (s, 1H, CHP), 0.91 (t, J_HH=7.3 Hz, 3H, CH₂Me), 1.59 (m, 2H,

CH₂Me), 1.63 (br s, 4H, CH₂N), 1.82 (br s, 12H, NMe₂), 3.22 (m, 2H,

CH₂S), 6.53 (m, 1H, ArH), 7.06 (m, 1H, ArH), 7.28 (m, 1H, ArH), 7.32

(m, 1H, ArH). ¹³C{¹H} NMR (C₆D₆): d 1.57 (d, J_PC=5.6 Hz, SiMe₃), 4.05

ArH), 7.53 (m, 2H, ArH). ¹³C{¹H} NMR ([D₈]THF): d 1.60 (d, J_PC

=

5.7 Hz, SiMe₃), 2.94 (d,

J_PC=2.9 Hz, SiMe₃), 10.39 (d, J_PC=46.2 Hz,

CHP), 118.94, 126.66, 127.80, 127.88, 131.37, 135.70 (Ar), 135.89 (d, J_PC

=

20.1 Hz, Ar), 141.67, 143.19 (d,

J_PC=20.1 Hz, Ar), 160.35 (d, J_PC

=

32.4 Hz, Ar). ⁷Li{¹H} NMR ([D₈]THF): d 3.4. ³¹P{¹H} NMR ([D₈]THF):

d ꢀ18.5. Elemental analysis calcd (%) for C₁₉H₂₈LiPSSi₂: C 59.65, H 7.38;

found: C 59.74, H 7.47.

(d, J_PC=65.5 Hz, CHP), 13.32 (CH₂Me), 22.87 (CH₂Me), 32.31 (d, J_PC

=

Protonation of

a lithium thiolate. {(Me₃Si)₂CH}P(Me)(H)ACHTUNGTRENNUNG(C₆H₄-2-S)

18.2 Hz, CH₂S), 45.54 (NMe₂), 56.31 (CH₂N), 114.61 (d, J_PC=3.2 Hz, Ar),

121.94 (d, J_PC=35.7 Hz, Ar), 126.60, 133.99 (Ar), 167.35 (d, J_PC=53.5 Hz,

Ar). ⁷Li{¹H} NMR (C₆D₆): d 4.7 (d, J_LiP=63.3 Hz). ³¹P{¹H} NMR (C₆D₆):

(1_SH’): Anhydrous HCl (1.01 mL of a 2.0m solution in diethyl ether,

2.02 mmol) was added to a solution of 1_Sb (0.45 g, 1.01 mmol) in light pe-

troleum (10 mL) and this mixture was stirred for 15 min. The pale precip-

itate was removed by filtration and washed with light petroleum (2ꢂ

20 mL). Solvent was removed in vacuo from the combined organic wash-

ings and the resulting pale yellow solid was crystallized from hot methyl-

cyclohexane to give pale yellow blocks of 1_SH’ suitable for X-ray crystal-

lography. Yield 0.15 g, 47.2%. ¹H NMR (CDCl₃): d 0.04 (s, 9H, SiMe₃),

d

ꢀ63.5 (q,

J_LiP=63.3 Hz). Elemental analysis calcd (%) for

C₂₂H₄₆N₂Si₂SPLi: C 56.85, H 9.98, N 6.03; found: C 57.04, H 9.89, N 5.93.

Synthesis of a tertiary phosphane-substituted lithium thiolate by photoly-

sis of a lithium phosphanide. [[{(Me₃Si)₂CH}P(Ph)ACHTUNGRTNEN(UG C₆H₄-2-S)]Li]₆((3_S)₆):

nBuLi (0.42 mL, 1.05 mmol) was added to a solution of 3_PH (0.40 g,

Table 1. Crystallographic data for 2_Pb, 3_Pb, 5 _Pa, 5_Pc, 1_Pe, (3_S)₆, (4_S)₈, and 1_SH’.

Compound

2_Pb

3_Pb

5_Pa

5_Pc

1_Pe

(3_S)₆

(4_S)₈

1_SH’

CCDC no.

formula

M_r

cryst size

[mm³]

crystal

system

space

group

a [ꢁ]

908626

908627

908628

908629

908630

908631

908632

908633

C₁₄H₂₇PSSi

314.6

C₂₂H₄₆LiN₂PSSi₂C₂₅H₄₄LiN₂PSSi₂C₅₅H₆₈Li₃O₄P₃S₃C₃₃H₅₂LiO₉PS

464.8 498.8 1003.0 662.7

0.18ꢂ0.15ꢂ0.15 0.06ꢂ0.03ꢂ0.005 0.35ꢂ0.30ꢂ0.30 0.30ꢂ0.15ꢂ0.15 0.35ꢂ0.10ꢂ0.10 0.10ꢂ0.10ꢂ0.10

C₁₈H₃₄KOPSSi₂C₁₁₄H₁₆₈Li₆P₆S₆Si₁₂C₈₈H₁₂₈Li₈P₈S₈

424.8 2295.4 1754.7

0.50ꢂ0.40ꢂ0.40 0.40ꢂ0.20ꢂ0.20

triclinic

orthorhombic

monoclinic

rhombohedral

monoclinic

¯

P1

P2₁2₁2₁

P2₁

P2₁/c

R3

P2/c

P2₁/c

11.1585(7)

14.8236(8)

18.2149(10)

87.235(4)

88.661(5)

74.724(5)

2902.9(3)

4

9.363(3)

11.105(3)

16.551(5)

90.707(3)

103.756(3)

114.217(3)

1512.7(8)

2

12.7379(5)

15.4215(7)

27.7054(9)

8.9343(2)

17.1740(3)

11.8147(3)

11.6050(9)

24.0993(16)

9.0386(7)

25.5140(17)

17.3071(12)

16.9419(7)

17.2008(8)

21.2577(11)

7.5455(3)

15.8994(5)

15.7440(6)

b [ꢁ]

c [ꢁ]

a [8]

b [8]

108.017(2)

111.142(9)

104.932(5)

102.615(4)

g [8]

V [ꢁ³]

Z

5442.4(4)

4

0.267

1723.93(7)

2

0.191

2357.7(3)

4

0.487

9756.9(11)

3

3.048

5985.6(5)

2

0.289

1843.19(12)

4

0.378

m [mm^ꢀ1

]

0.260

0.254

trans coeff

range

0.955–0.962

0.985–0.999

0.912–0.924

0.945–0.972

0.848–0.953

0.650–0.750

0.869–0.893

0.864–0.928

reflns measd

unique reflns

R_int

27151

10215

0.041

6738

12058

5197

0.051

3626

32262

11700

0.049

8900

9299

6082

0.021

4964

17320

4143

0.059

3606

15210

3823

0.087

2554

28317

7809

0.038

6842

10797

3916

0.019

3479

reflns with

F²>2s

refined pa-

rameters

R (on F,

545

299

654

407

224

223

661

174

0.067

0.051

0.050

0.055

0.063

0.062

0.108

0.030

F²>2s)^[a]

R_w(on F², all 0.192

0.133

0.098

0.138

0.179

0.250

0.078

data)^a

goodness of

1.111

1.028

1.014

1.183

1.051

1.119

1.067

fit^a

max, min el,

density [eꢁ^ꢀ3

1.34, ꢀ1.36

0.86, ꢀ0.34

0.30, ꢀ0.25

0.99, ꢀ0.22

0.73, ꢀ0.38

1.15, ꢀ0.44

0.64, ꢀ0.47

0.37, ꢀ0.27

]

2

[a] Conventional R=Sj jF_oj-jF_cj j/SjF_oj ; R_w=[Sw

(F_o-F_c²)²/Sw(F_o)²]^1/2; S=[Sw(F_o-F_c²)²/(number of data - number of params)]^1/2for all data.

ACHTUNGTRNEUNGN ACHTNUGTRENNUNG ACTHNUGTRENNGUN

&

12

&

www.chemeurj.org

ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Chem. Eur. J. 0000, 00, 0 – 0

ÝÝ

These are not the final page numbers!

Thioether-Substituted Phosphanide Ligands

FULL PAPER

0.28 (s, 9H, SiMe₃), 1.82 (d, J_PH=9.7 Hz, 3H, PMe), 2.80 (d, J_PH

=

[5] a) T. Kochi, S. Noda, K. Yoshimura, K. Nozaki, J. Am. Chem. Soc.

2007 , 129 , 8948; b) S. Ito, K. Munakata, A. Nakamura, K. Nozaki, J.

Am. Chem. Soc. 2009 , 131 , 14606; c) E. Drent, R. van Dijk, R. van

Ginkel, B. van Oort, R. I. Pugh, Chem. Commun. 2002 , 744; d) W.

Weng, Z. Shen, R. F. Jordan, J. Am. Chem. Soc. 2007 , 129 , 15450;

e) K. M. Skupov, L. Piche, J. P. Claverie, Macromolecules 2008 , 41 ,

2309.

[6] B. Sundararaju, M. Achard, B. Demersemann, L. Toupet, G. V. M.

Sharma, C. Bruneau, Angew. Chem. 2010 , 122 , 2842; Angew. Chem.

Int. Ed. 2010 , 49 , 2782.

22.6 Hz, 1H, CHP), 6.87 (m, 1H, ArH), 6.90 (br d, J_PH=406.1 Hz, 1H,

PH), 7.07 (m, 1H, ArH), 7.14 (m, 1H, ArH), 7.53 (m, 1H, ArH). ¹³C{¹H}

NMR (CDCl₃): d 1.34 (d, J_PC=4.9 Hz, SiMe₃), 2.11 (d, J_PC=3.8 Hz,

SiMe₃), 4.92 (d, J_PC=12.6 Hz, CHP), 7.88 (d, J_PC=37.9 Hz, PMe), 120.24

(d, J_PC=10.1 Hz, Ar), 122.31 (Ar), 130.46 (d, J_PC=12.1 Hz, Ar), 131.75

(Ar), 135.04 (d, J_PC=6.6 Hz, Ar), 161.93 (Ar). ³¹P NMR (CDCl₃): d

ꢀ20.4 (d, J_PH=406.1 Hz). Elemental analysis calcd (%) for C₁₄H₂₇PSSi₂:

C 53.45, H 8.65; found C 53.55, H 8.56.

Crystal structure determinations of 1_Pe, 2_Pb, 3_Pb, 5_Pa, 5_Pc, (3_S)₆, (4_S)₈, and

1_SH’: Except for 3_Pb, measurements were made at 150 K on an Oxford

Diffraction Gemini A Ultra diffractometer using CuKa radiation (l=

1.54184 ꢁ) for (3_S)₆and MoKa radiation (l=0.71073 ꢁ) otherwise; data

for 3_Pb were obtained using synchrotron radiation (l=0.6889 ꢁ), a Crys-

tal Logic kappa diffractometer and Rigaku Saturn 724+ CCD detector

at beamline I19 of Diamond Light Source. Cell parameters were refined

from the observed positions of all strong reflections. Intensities were cor-

rected semi-empirically for absorption, based on symmetry-equivalent

and repeated reflections. The structures were solved by direct methods

and refined on F²values for all unique data. Table 1 gives further details.

All non-hydrogen atoms were refined anisotropically, and hydrogen

atoms were constrained with a riding model; U(H) was set at 1.2 (1.5 for

methyl groups) times U_eqfor the parent atom. Twofold disorder of one

THF ligand was resolved for 5_Pa with the help of restraints on the geom-

etry and displacement parameters of these disorder components. Disor-

der was also resolved for two of the four crystallographically independent

[7] a) D. A. Evans, F. E. Michael, J. S. Tedrow, K. R. Campos, J. Am.

Chem. Soc. 2003 , 125 , 3534; b) X. Caldentey, X. C. Cambeiro, M. A.

Pericas, Tetrahedron 2011 , 67 , 4161.

[8] a) L. F. Lindoy, S. B. Livingstone, T. N. Lockyer, Inorg. Chem. 1967 ,

6 , 652; b) T. N. Lockyer, Aust. J. Chem. 1974 , 27 , 259.

[9] a) D. M. Roundhill, W. B. Beaulieu, U. Bagchi, J. Am. Chem. Soc.

1979 , 101 , 5428; b) A. Benefiel, D. M. Roundhill, W. C. Fultz, A. L.

Rheingold, Inorg. Chem. 1984 , 23 , 3316; c) A. Benefiel, D. M.

Roundhill, Inorg. Chem. 1986 , 25 , 4027.

[10] For examples see: a) S. Pal, S. N. Poddar, G. Mukherjee, Transition

Metal Chem. 1994, 19, 449; b) C. A. McAuliffe, Inorg. Chem. 1973 ,

12 , 2477; c) S. Mandal, N. Paul, P. Banerjee, T. P. Mondal, S. Goswa-

mi, Dalton Trans. 2010 , 39 , 2717; d) S. Ohkoshi, Y. Ohba, M. Iwaizu-

mi, S. Yamauchi, M. Ohkoshi-Ohtani, K. Tokuhisa, M. Kajitani, T.

Akiyama, A. Sugimori, Inorg. Chem. 1996 , 35 , 4569; e) D. Shimizu,

N. Takeda, N. Tokitoh, Chem. Commun. 2006 , 177; f) F. Guyon, M.

Knorr, A. Garillon, C. Strohmann, Eur. J. Inorg. Chem. 2012 , 282.

[11] a) J. S. Kim, J. H. Reibenspies, M. Y. Darensbourg, J. Am. Chem.

Soc. 1996 , 118 , 4115; b) J. S. Kim, J. H. Reibenspies, M. Y. Dare-

nsbourg, Inorg. Chim. Acta 1996 , 250 , 283.

[12] K. Pramanik, U. Das, B. Adhikari, D. Chopra, H. Stoeckli-Evans,

Inorg. Chem. 2008 , 47 , 429.

[13] a) R. J. Angelici, Acc. Chem. Res. 1988 , 21 , 387; b) R. J. Angelici,

Coord. Chem. Rev. 1990 , 105 , 61; c) C. Bianchini, A. Meli, J. Chem.

Soc. Dalton Trans. 1996 , 801; d) C. Bianchini, A. Meli, Acc. Chem.

Res. 1998 , 31 , 109.

ligACHTUNGTRENNUNGands of (4_S)₈; restraints were applied to the geometry and displacement

parameters of all except Li atoms. The satisfactory refinement of this

structure was hindered by highly disordered solvent (presumably tol-

uene), which appears to be easily lost from the crystals on isolation from

the mother liquor; the contribution of this disordered solvent to the dif-

fraction was treated by the SQUEEZE procedure of PLATON.^[36]The

structure is of lower precision than the others reported here. Other pro-

grams were standard Oxford Diffraction CrysAlisPro, Rigaku Crystal-

Clear, and Bruker APEX2 for data collection and processing, and

SHELXTL for structure solution, refinement, and molecular graphics.^[37]

CCDC 908626–908633 contain the supplementary crystallographic data

for this paper. These data can be obtained free of charge from The Cam-

bridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data re-

quest/cif.

[14] The new compounds described herein have been labeled in the fol-

lowing manner: The nature of the substituents R¹and R²is denoted

by an Arabic numeral, according to Scheme 4, this is followed by a

subscript, which indicates the location of the chloride, proton, or

negative charge in the (pro)ligand (that is, P, S, or both P and S),

and finally by H, Cl, or a lower-case letter, which indicates either

the protio- or chloro- form or the nature of the metal and coligands

(a Li

e K(THF)_n, f Li

a letter indicates no coligands). Thus 1_Pb indicates ligand 1 deproto-

nated at phosphorus as its Li(tmeda)complex, whereas 1_PH indicates

the P-protio form of this ligand and 1_Sb indicates the thiolate prod-

uct obtained on photolysis of 1_Pb.

N

b

Li

N

c Li([12]crown-4)₂, d NaACHTUNGTRENNUNG(THF)_n,

Acknowledgements

A

E

ACHTUNGTRENNUNG

The authors are grateful to the EPSRC and Newcastle University for

support, and to Diamond Light Source for access to synchrotron facili-

ties.

AHCTUNGTRENNUNG

[15] S. Blair, K. Izod, R. W. Harrington, W. Clegg, Organometallics 2003 ,

22 , 302.

[1] a) PhosphorusACHTUNGTRENNUNG(III) Ligands in Homogeneous Catalysis: Design and

[16] D. A. Vicic, W. D. Jones, J. Am. Chem. Soc. 1999 , 121 , 7606.

[17] C. Huang, S. Gou, H. Zhu, W. Huang, Inorg. Chem. 2007 , 46 , 5537.

[18] M. Shibue, M. Hirotsu, T. Nishioka, I. Kinoshita, Organometallics

2008 , 27 , 4475.

Synthesis (P. C. J. Kamer, P. W. N. M. van Leeuven, Eds.), John

Wiley & Sons; Chichester (UK), 2012; b) Phosphorus Compounds

in Asymmetric Catalysis, Synthesis and Applications (A. Bçrner,

Ed.), Wiley-VCH; Weinheim, 2008.

[19] K. Izod, E. R. Clark, R. W. Harrington, W. Clegg, Organometallics

2012 , 31 , 246.

[20] K. Izod, Adv. Inorg. Chem. 2000 , 50 , 33.

[21] G. Stieglitz, B. Neumꢃller, K. Dehnicke, Z. Naturfosch. B 1993, 48,

156.

[22] P. B. Hitchcock, M. F. Lappert, P. P. Power, S. J. Smith, J. Chem. Soc.

Chem. Commun. 1984 , 1669.

[23] K. Izod, J. M. Watson, R. W. Harrington, W. Clegg, Dalton Trans.

2011 , 40 , 11712.

[24] M. M. Olmstead, P. P. Power, J. Am. Chem. Soc. 1990 , 112 , 8008.

[25] G. W. Rabe, G. P. A. Yap, A. L. Rheingold, Inorg. Chem. 1997 , 36 ,

1990.

[2] For reviews, see: a) D. W. Stephan, G. Erker, Angew. Chem. 2010 ,

122 , 50; Angew. Chem. Int. Ed. 2010 , 49 , 46; b) D. W. Stephan, Org.

Biomol. Chem. 2008 , 6 , 1535; c) D. W. Stephan, Dalton Trans. 2009 ,

3129; d) D. W. Stephan, Chem. Commun. 2010 , 46 , 8526.

[3] For recent reviews of the chemistry of P,S and related hybrid ligands

and their applications in catalysis, see: a) D. K. Dutta, B. Deb,

Coord. Chem. Rev. 2011 , 255 , 1686; b) J. R. Dilworth, N. Wheatley,

Coord. Chem. Rev. 2000 , 199 , 89; c) W.-H. Zhang, S. W. Chien,

T. S. A. Hor, Coord. Chem. Rev. 2011 , 255 , 1991; d) F. L. Lam, F. Y.

Kwong, A. S. C. Chan, Chem. Commun. 2010 , 46 , 4649.

[4] a) S. Noda, T. Kochi, K. Nozaki, Organometallics 2009 , 28 , 656;

b) X. Zhou, S. Bontemps, R. F. Jordan, Organometallics 2008, 27,

8421; c) L. Piche, J.-C. Daigle, J. P. Claverie, Chem. Commun. 2011 ,

47 , 7836.

[26] G. W. Rabe, S. Kheradmandan, H. Heise, I. A. Guzei, L. M. Liable-

Sands, A. L. Rheingold, Main Group Chem. 1998, 25, 6541.

Chem. Eur. J. 2013, 00, 0 – 0

ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemeurj.org

&

13

&

ÞÞ

These are not the final page numbers!

K. Izod et al.

[27] P. L. Arnold, J. J. Hall, L. S. Natrajan, A. J. Blake, C. Wilson, Dalton

Trans. 2003 , 1053.

[28] T. Chivers, A. Downard, G. P. A. Yap, Inorg. Chem. 1998 , 37 , 5708.

[29] M. D. Janssen, D. M. Grove, G. van Koten, Prog. Inorg. Chem. 1997 ,

46 , 97.

[30] G. A. Sigel, P. P. Power, Inorg. Chem. 1987 , 26 , 2819.

[31] L. M. Engelhardt, J. M. Harrowfield, M. F. Lappert, I. A. Mackin-

non, B. H. Newton, C. L. Raston, B. W. Skelton, A. H. White, J.

Chem. Soc. Chem. Commun. 1986 , 846.

[32] a) T. Kawamoto, M. Nishiwaki, M. Nishimija, K. Nozaki, A. Iga-

shira-Kamiyama, T. Konno, Chem. Eur. J. 2008 , 14 , 9842; b) T. Ka-

wamoto, N. Ohkashi, I. Nagasawa, H. Kuma, Y. Kushi, Chem. Lett.

1997 , 553.

[33] N. D. R. Barnett, W. Clegg, L. Horsburgh, D. M. Lindsay, Q.-Y. Liu,

F. M. Mackenzie, R. E. Mulvey, P. G. Williard, Chem. Commun.

1996 , 2321.

[34] A. Bondi, J. Phys. Chem. 1964 , 68 , 441.

[35] M. Mag, J. Muth, K. Jahn, A. Peyman, G. Kretzschmar, J. W. Engels,

E. Uhlmann, Bioorg. Med. Chem. 1997 , 5 , 2213.

[36] A. L. Spek, Acta Crystallogr. Sect. D 2009 , 65 , 148.

[37] a) CrysAlisPro, Oxford Diffraction Ltd (now part of Agilent Tech-

nologies Ltd), Oxford, UK, 2008; b) APEX2, Bruker AXS Inc.,

Madison, WI, USA, 2010; c) G. M. Sheldrick, Acta Crystallogr. Sect.

A 2008 , 64 , 112.

Received: November 1, 2012

Published online: && &&, 0000

&

14

&

www.chemeurj.org

ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Chem. Eur. J. 0000, 00, 0 – 0

ÝÝ

These are not the final page numbers!

Thioether-Substituted Phosphanide Ligands

FULL PAPER

Phosphanide Ligands

K. Izod,* E. R. Clark, P. Foster,

R. J. Percival, I. M. Riddlestone,

W. Clegg, R. W. Harrington &&&& — &&&&

Light-Induced Rearrangement of Thio-

ether-Substituted Phosphanide

Ligands: Scope and Limitations of a

Remarkable Isomerization

Phosphanides under the spotlight:

Exposure of ortho-phenylene-bridged

phosphanide–thioether complexes to

white light leads to rapid isomerization

to the corresponding tertiary phos-

phane–thiolate (see scheme). This

process is general for a range of sub-

stituents at phosphorus or sulfur and

appears to proceed by a concerted

migration of the alkyl/aryl group from

sulfur to phosphorus within the ligand.

Chem. Eur. J. 2013, 00, 0 – 0

ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemeurj.org

&

15

&

ÞÞ

These are not the final page numbers!

Article Doi

Light-induced rearrangement of thioether-substituted phosphanide ligands: Scope and limitations of a remarkable isomerization

DOI: 10.1002/chem.201203918

Source and publish data:

Authors:

Article abstract of DOI:10.1002/chem.201203918

Full text of DOI:10.1002/chem.201203918

Products guided by the article

R&D Labs maybe for 1441422-41-8

Relevant to this article

Hot Product