Deoxygenation of carbon dioxide by electrophilic terminal phosphinidene complexes

Dynamic Article LinksC<Chemical Science

Cite this: Chem. Sci., 2012, 3, 3526

www.rsc.org/chemicalscience EDGE ARTICLE

Deoxygenation of carbon dioxide by electrophilic terminal phosphinidenecomplexes†

Christian Schulten,a Gerd von Frantzius,a Gregor Schnakenburg,a Arturo Espinosab and Rainer Streubel*a

Received 25th July 2012, Accepted 3rd September 2012

DOI: 10.1039/c2sc21081a

Deoxygenation of carbon dioxide was achieved using transient terminal phosphinidene chromium and

tungsten complexes 2a,b. The overall reaction is exothermic according to DFT calculations on the

model terminal P-methyl phosphinidene complex Me-2b; this was also supported by the calculated

thermodynamic oxygen-transfer potential. The oxaphosphiran-3-one complex intermediates 3a,b

possess an unprecedented bonding situation as some characteristics of a side-on bound carbon dioxide

to the (formally) low-coordinated phosphorus centre come to the fore. This is expressed by equidistant

P–C and P–O bonds and unusual bond strength relationship, i.e. P–C > P–O, as revealed by the relaxed

force constants and other related parameters. The decomposition of 3a,b via CO extrusion yields

terminal phosphinidene oxide complexes 4a,b which dimerise to the final products, the 1,3-dioxa-2,4-

diphosphetane complexes 5a,b–8a,b. Additional experimental evidence for the transient formation of

phosphinidene oxide complexes 4a,b was obtained by a cross dimerisation experiment using transient

chromium and tungsten complexes 2a,b. First comparative investigations on the reaction of Li–Cl

phosphinidenoid complex 10 and CO2 at low temperature revealed the formation of the carbamoyl–

phosphane complex 11.

Introduction

The challenge to use carbon dioxide as a synthetic building block

has recently received increasing interest in the area of green and

sustainable chemistry and catalysis, and the direct conversion of

carbon dioxide into urea as an industrially important process

provides an extra stimulus.1,2 In this regard, various addition

reactions have been studied involving electrophiles and nucleo-

philes3 and, for example, nucleophilic carbenes add to carbon

dioxide to form betaines4,5 but not to give stable a-lactones

(oxiranones) I (Scheme 1, E¼C). The latter are obtained only via

epoxidation of ketenes having perfluorinated large alkyl

substituents; in this case gentle heating also led to ring frag-

mentation and thus to carbon monoxide and the corresponding

ketone, which represents a net deoxygenation reaction of carbon

dioxide.6 In comparison, transition metal complexes have been

widely used to bind and thus activate carbon dioxide.7 In some

cases, deoxygenation of carbon dioxide was initiated by its side-

aInstitut f€ur Anorganische Chemie, RheinischeFriedrich-Wilhelms-Universit€at Bonn, Gerhard-Domagk-Str. 1, 53121Bonn, Germany. E-mail: [email protected]; Fax: +49-228-73-9616;Tel: +49-228-73-5345bDepartamento de Qu�ımica Org�anica, Universidad de Murcia, Campus deEspinardo, 30100 Murcia, Spain. E-mail: [email protected]; Fax: +34 868-88-4149; Tel: +34 868-88-7489

† Electronic supplementary information (ESI) available: Homodesmoticreactions used for computing RSE. CCDC reference number 806428. ForESI and crystallographic data in CIF or other electronic format see DOI:10.1039/c2sc21081a

3526 | Chem. Sci., 2012, 3, 3526–3533

on coordination as in complexes II8,9 (Scheme 1) and some Mo,10

Fe,11 and Ni12 h2–CO2 complexes were structurally charac-

terised. The deoxygenation mechanism based on transient or

stable complexes II to give the metal oxo complex and carbon

monoxide was corroborated recently by a DFT study13 for the

case of a Ni0 N-heterocyclic carbene (NHC) complex.14

Computationally and experimentally described are oxasilir-

anone (E ¼ SiH2)15 and oxaziridinone (E ¼ NH),16 whereas

oxagermiranone (E ¼ GeH2)17 is known in silico only; oxa-

phosphiranones (E ¼ PR) and complexes thereof are unknown,

so far. However, with the discovery of metal-free organophos-

phorus catalysed fixation of carbon dioxide,16,18–21 evidence is

growing that main group elements may very well enter the clas-

sical transition metal domain of small molecule activation22–24 as

recently put forward by Power.25

Other recent examples are the deoxygenation of CO2 initiated

by coordination to an NHC,26 and the organocatalytic fixation of

CO227 by a frustrated Lewis pair (FLP) system,28 or by addition

Scheme 1 Known deoxygenation processes of CO2 via transient

a-lactones I (oxiranones, E ¼ C) or via side-on carbon dioxide transition

metal complexes II.

This journal is ª The Royal Society of Chemistry 2012

to an organic substrate activated by an NHC.29 More recently,

carbon dioxide insertion into a strained P–N bond of a cyclic

amidophosphorane was used for CO2 ‘‘sequestration’’, taking

advantage of ‘‘masked’’ FLP reactivity.30 Regarding this back-

ground, the work of Issleib and others may seem as modern

classics but still is fundamental for metal phosphanides MPHR

(M ¼ Na, K; R ¼ Ph, cyclohexyl) which yield metal phosphi-

nocarboxylates M[HRPCO2] with CO2.31

Here, deoxygenation of carbon dioxide by thermally generated

electrophilic terminal phosphinidene complexes is reported. This

is compared to the reaction of a Li–Cl phosphinidenoid complex

with carbon dioxide that yields a carbamoylphosphane complex,

instead. DFT calculations on the deoxygenation process point to

an oxaphosphiranone complex as the reactive intermediate

which decomposes to yield diastereomeric 1,3-dioxa-2,4-

diphosphetane complexes.

Results and discussion

Transient electrophilic terminal phosphinidene complexes 2a,b,

generated by thermolysis of 2H-azaphosphirene complexes

1a,b,32,33 react at 75 �C with carbon dioxide (50 bar) to yield

selectively cis and trans 1,3-dioxa-2,4-diphosphetane complexes

5a,b and 6a,b which were isolated and unambiguously charac-

terised (Scheme 2).34

The deoxygenation of CO2 by terminal phosphinidene

complexes35 2a,b is assumed to proceed via intermediate oxa-

phosphiranone complexes 3a,b (i), which subsequently extrude

CO to form phosphinidene oxide complexes 4a,b (ii). Subse-

quently, 4a,b dimerise to yield 1,3-dioxa-2,4-diphosphetane

complexes 5a,b and 6a,b36 as final products (iii). Further evidence

for the deoxygenation process and the intermediacy of 4a,b was

obtained by a cross dimerization experiment that furnished the

mixed metal 1,3-dioxa-2,4-diphosphetane complexes cis 7a,b and

trans 8a,b appearing each as pair of diastereomers (Scheme 2). A

related situation was observed recently in thermal reactions of

complexes 1 with isocyanates in which the dinuclear 1,3-dioxa-

2,4-diphosphetane complexes were also obtained as final prod-

ucts (in lower yields).36

A first attempt was made to gain selective access to the

assumed oxaphosphiranone complex 3b using the Li–Cl

phosphinidenoid complex 10,37 which has been successfully

used in the synthesis of oxaphosphirane complexes.37,38

Deprotonation of P-chloro phosphane complex 9 39 by lithium

Scheme 2 Proposed mechanism for the formation of complexes

3a,b–6a,b.

This journal is ª The Royal Society of Chemistry 2012

diisopropylamide in the presence of 12-crown-4 and subsequent

reaction with CO2 at �60 �C yielded complex 11 (Scheme 3) as

the only product to be detected by 31P NMR spectroscopy at

ambient temperature. 31P NMR reaction monitoring (�60

to �30�C) revealed, besides complex 11, exclusively a broad

signal of a transient compound at 112.6 ppm, the 1J(W,P)

coupling of which could not be observed. Two mechanistic

options for the formation of complex 11 seem to be reasonable:

(i) the nucleophilic attack of complex 10 at CO2 and ring

closure to yield 3b with subsequent ring opening by diisopro-

pylamine to give 11 or (ii) reaction of 10 with an initially

formed carbamoylamine to yield 11 (Scheme 3).

The 1,3-dioxa-2,4-diphosphetane complexes 5–8 resonate at

low field (>240 ppm), whereby the 31P NMR shift of the trans

isomers appear downfield of the cis isomers for both metal

complexes. The Dd(31P) value of the tungsten and chromium

complexes is about 65 ppm. The constitution of carbamoyl-

phosphane complex 11 was confirmed by single-crystal X-ray

diffraction analysis (Fig. 1).

DFT calculations

To get further insights into the carbon dioxide deoxygenation

process, the model phosphinidene tungsten complex [(CO)5W-

(PMe)] (Me-3b, Scheme 4) and the phosphinidene oxide model

complex [(CO)5W(OPMe)] (Me-4b) were used (Fig. 2, steps i–ii)

for the DFT calculations; oxaphosphiran-3-one Ia and

complexes thereof (Ib,c,Me-3b, 3bsyn, 3banti) were also calculated

for the first time.

Oxaphosphiranone Ia shows significant differences of the ring

geometry compared with the corresponding parent oxaphos-

phirane (P–C 1.823, P–O 1.724, C–O 1.417 [�A]):40 the phos-

phorus–oxygen bond is shorter in the oxaphosphirane, while the

opposite holds for the endocyclic C–O bond. Upon complexation

with Cr(CO)5 (Ib) the endocyclic P–C, P–O as well as the

exocyclic C–O bond shorten slightly; the characteristic CO

stretch vibration of �1955 cm�1 shifts towards shorter wave-

numbers (Table 1).

The related bond strengths were inspected in order to obtain

insight into the consequences of the geometrical bond changes.

Among several possible bond-strength related descriptors, the

widespread used41 Wiberg’s bond index (WBI)42 was computed.

Additionally, the electron density at bond critical points, r(rc),

Scheme 3 Proposed reaction courses for the formation of complex 11.

Chem. Sci., 2012, 3, 3526–3533 | 3527

Fig. 1 Molecular structure of complex 11 in the crystal (50% probability

level; hydrogen atoms are omitted for clarity, except that bound to P).

Selected bond lengths [�A] and angles [�]:W–P 2.4755(11), P–C(1) 1.807(4),

P–O(1) 1.672(3), O(2)–C(2) 1.219(5), O(1)–C(2) 1.368(5), N(1)–C(2)

1.344(5);W–P–C(1) 119.85(16), O(1)–P–C(1) 100.11(17), C1–P–H 105(2),

P–O(1)–C(2) 117.5(3), O(1)–C(2)–N(1) 113.4(4), O(2)–C(2)–N(1)

125.9(5), O(1)–C(2)–O(2) 120.7(4).

Scheme 4 Scope of calculations, B3LYP/6-311G(d,p), LanL2DZ (W).

successfully used in quantifying many other different bonding

situations41i–k,43 and derived from Bader’s atoms-in-molecules

(AIM) theory44 was also included (Table 1). Both parameters

Fig. 2 BSSE corrected reaction Gibbs free energies and ZPE-corrected energi

complex Me-2b to yield phosphinidene oxide Me-4b. On path A: intermedi

(nucleophilic attack at carbon) and TS-II (extrusion of CO). Path B: dissociat

the (P)O–CO bond).

3528 | Chem. Sci., 2012, 3, 3526–3533

point to the following strength order for the endocyclic bonds:

P–O < P–C � C–O. The same trend was observed for the bond

stiffness quantified by the relaxed force constants, k0 (Table 1), a

robust and safely transferable parameter obtained as the recip-

rocal value of the compliance constant.41i,45 Worth mentioning is

that differences in strength stiffness are enlarged upon

complexation within the group of endocyclic bonds PO/PC/CO

in Ib, and the P–O bond becomes remarkably weakened.

An important feature of small rings is that their reactivity is

often triggered by enhanced ring strain.46 Therefore, ring strain

energies (RSE) of Ia,b were quantitatively evaluated by means of

appropriate homodesmotic reactions41i,j,47 (see the ESI†), where

the number and type of bonds and the valences of all atoms are

kept.48 In relation to the reported RSE for the 1,3,3-trimethyl

substituted oxaphosphirane (94.2 kJ mol�1) and its P-Cr(CO)5complex (92.0 kJ mol�1),47 the parent oxaphosphiranone Ia was

found to be much more strained (141.3 kJ mol�1, at LPNO-

NCEPA1/def2-TZVPP//B3LYP-D/def2-TZVP). Following the

same behaviour as the aforementioned fully saturated ring, the

chromium complex Ib displays a slightly higher value (144.7 kJ

mol�1) compared to unligated Ia.

Thermodynamic oxygen-transfer potentials (TOP’s)49 of the

redox couples [(CO)5M(PR)]/[(CO)5M(OPR)] and CO/CO2 are

shown in Fig. 3. By definition a species Z is oxidised by YO, if

TOPZ/ZO < TOPY/YO. Thus, for a phosphinidene complex

[(CO)5M(PR)] to be oxidised by CO2 a TOP < �386 kJ mol�1 is

required, which is the case for R¼Me (Me-2b), CH(SiMe3)2 (2b)

and for Niecke’s complex (Cr, R ¼ NiPr2),50 but not for R ¼ H

and CF3.

The first step in the oxidation of terminal phosphinidene

model complex Me-2b by carbon dioxide (Fig. 2, path A) is the

nucleophilic attack of the phosphorus centre at the carbon of CO2

(TS-I, Fig. 4).

Selected bond lengths of transition states TS-I and TS-II are

collected in Table 2.

Addition of Me-2b to CO2 proceeds with a barrier of 62 kJ

mol�1 to yield oxaphosphiran-3-one model complex Me-3b

(Fig. 2, Table 2). For this new class of compounds the fully

es for the deoxygenation of CO2 bymethyl-model tungsten phosphinidene

ate formation of oxaphosphiran-3-one Me-3b via transition states TS-I

ion of the van der Waals complex via transition state TS-III (breaking of

This journal is ª The Royal Society of Chemistry 2012

Table 1 Computed characteristic bond features of oxaphosphiranones Ia

P–C P–O P–M endoC–O exoC–On(C]O)

Ia 1.846 (0.870) [15.07] {2.32} 1.842 (0.736) [12.30] {1.95} — 1.310 (1.146) [32.96] {5.41} 1.183 (1.836) [44.77] {14.19} 1957Ib 1.836 (0.828) [15.66] {2.29} 1.839 (0.665) [12.94] {1.82} 2.287 (0.548) [7.35] {1.36} 1.317 (1.120) [31.96] {4.67} 1.181 (1.849) [44.96] {14.19} 1952

a Bond length [�A], WBI in parentheses, electron density at bond critical points (r(rc)� 102 [e a0�3]) in square brackets, relaxed force constants (k0 [mdyn

per �A]) in curly brackets and exocyclic CO stretch vibration [cm�1].

Table 2 Bond lengthsa of transition states TS-I,II,b and of oxaphos-phiran-3-one tungsten complexes 3bc and Me-3b

P–C P–O P–M C–O1 C–O2

TS-I 2.178 2.312 2.430 1.219 1.173TS-II 1.810 1.603 2.457 1.834 1.148Me-3b 1.824 1.832 2.458 1.326 1.1833bsyn 1.812 1.862 2.488 1.326 1.1893banti 1.814 1.854 2.501 1.326 1.188

a [�A]. b Fig. 4. c Fig. 5.

Fig. 3 Thermodynamic oxygen-transfer potentials52 (DG, kJ mol�1) of

redox couples [(CO)5M(PR)]/[(CO)5M(OPR)] and Et-NC/Et-NCO,

CO/CO2 at B3LYP/6-311g(d,p), LanL2DZ(W); the arrow indicates the

direction of a more favoured oxygen transfer.

Fig. 4 Computed structures for the transition state model tungsten

complexes TS-I (nucleophilic attack at the CO2 carbon centre) and TS-II

(extrusion of CO).

Fig. 5 Calculated structures of oxaphosphiran-3-one complexes 3banti

and 3bsyn.

substituted complexes 3bsyn and 3banti were additionally calcu-

lated (the syn- and anti-prefixes refer to the position of

hydrogen at the a-carbon at phosphorus relative to the W(CO)5moiety, Fig. 5). For the methyl model oxaphosphiranone

complex Me-3b the barrier to the back-reaction amounts to

This journal is ª The Royal Society of Chemistry 2012

only 23 kJ mol�1, which essentially means that complex Me-3b

is located in a high energy local minimum with a small barrier

to the back-reaction and a barrier to further reaction compa-

rable in size for its formation. While intra- and exocyclic CO

distances are almost constant for the three oxaphosphiran-3-

one complexes, the large CH(SiMe3)2 substituent in 3bsyn/anti

effects a longer P–W and P–O bond relative to Me-3b, whereas

the endocyclic P–C bond is longer in the methyl model

complex. Especially noteworthy are the structures of Me-3b

and the two conformers 3bsyn and 3banti which, by their almost

identical P–C and P–O bonds, reveal the characteristics of a

side-on bonded complex of a CO2 ligand to the phosphorus

centre rather than a covalent three-membered ring structure.

Characteristic IR vibrational frequencies of oxaphosphiran-3-

one tungsten complexes are the exocyclic CO stretchings at 1903–

1935 cm�1 (Table 3), that vibrate at lower frequency than the

uncomplexed and P-chromium ligated models (Table 1), and

which are well separated in the spectrum before the CO

stretchings of the metal fragment (2000–2200 cm�1). Computed

vibrations agree well with the experimentally reported range of

1945-1990 cm�1 for perfluorinated oxiranone.6 Phosphorus

NMR chemical shifts of this new class of compounds (Table 3)

are at low field from 107 ppm (Me-3b) to 120 (3banti) and 136

ppm (3bsyn).

With a barrier of 63 kJ mol�1 Me-3b reacts further under

extrusion of CO (steps iia and b, Fig. 2) via TS-II (Fig. 4). The

phosphorus–oxygen bond in this transition state possesses a

value (1.6 �A) oriented more at the 1.5 �A of resulting phosphini-

dene oxide complexMe-4b35 than the 1.8�A ofMe-3b. The overall

reaction energy for the deoxygenation of carbon dioxide by

terminal phosphinidene model complex Me-2b amounts

to �26 kJ mol�1.

One-step dissociation of carbon monoxide from the initially

formed van der Waals complex between the phosphinidene

model Me-2b and CO2 (path B) can be ruled out on the basis of

the required high-energy TS (184 kJ mol�1) involved (Fig. 2).

Chem. Sci., 2012, 3, 3526–3533 | 3529

Table 3 Characteristic CO stretchingsa and phosphorus NMR chemicalshiftsb of oxaphosphiran-3-one complexes of tungsten

CO (sym) cisCO

trCO(sym) endoC–O exoC–O d(31P)

Me-3b 2161 2069 2064 1105 1935 1073bsyn 2155 2061 2042 1112 1903 1363banti 2155 2059 2042 1108 1908 120

a n [cm�1]. b d(31P) [ppm].

Phosphinidene oxide model complex [(CO)5W(OPMe)]

(Me-4b) finally dimerises with a barrier of 94 kJ mol�1 to yield

trans-1,3-dioxa-2,4-diphosphetane methyl model complexMe-6b

in an exergonic reaction (Fig. 6).

Fig. 6 Energy profile for the dimerization of phosphinidene oxide model

complex [(CO)5W(OPMe)] (Me-4b) to yield 1,3-dioxa-2,4-diphosphetane

model complex Me-6b.

Conclusions

Deoxygenation of carbon dioxide using transient terminal

phosphinidene complexes of chromium and tungsten was ach-

ieved. According to DFT calculations using terminal P-methyl

phosphinidene model complex Me-2b the overall reaction is

exothermic. This was also supported by the calculated thermo-

dynamic oxygen-transfer potential and, hence, a series of

powerful reductants was established. The barriers on two

concurrent mechanistic pathways involve intermediate oxa-

phosphiran-3-one complexes 3a,b rather than dissociation of CO

from the initially formed van der Waals complex. Remarkably,

the structures of complexes 3a,b reveal an unprecedented

bonding situation that has characteristics of a side-on bound

carbon dioxide to the (formally) low-coordinated phosphorus

centre (¼ non-classical complex); this comes together with an

extremely unusual bond strength relationship, i.e. P–C > P–O, as

revealed by the relaxed force constants and bond strength related

parameters. The decomposition of 3a,b yields terminal phos-

phinidene oxide complexes (and CO) which dimerise under

thermal conditions to the final products, the 1,3-dioxa-2,4-

diphosphetane complexes 5a,b–8a,b. Experimental evidence for

the transient formation of phosphinidene oxide complexes 4a,b

was obtained by a cross dimerization experiment using 2a,b. First

comparative investigations on the reaction of Li–Cl phosphini-

denoid complex 10 and CO2 at low temperature revealed the

formation of the carbamoylphosphane complex 11. Here, further

studies are required to unveil the unknown mechanism.

3530 | Chem. Sci., 2012, 3, 3526–3533

Experimental section

General procedures

All operations were performed in an atmosphere of deoxygen-

ated and dried argon using standard Schlenk techniques with

conventional glassware. Solvents were distilled from sodium

wire/benzophenone in an argon atmosphere. NMR data were

recorded on a Bruker Avance 300 spectrometer (1H: 300.13

MHz; 13C: 75.5 MHz; 31P: 121.5 MHz) at 30 �C (if not otherwise

noted) using C6D6, CDCl3 or CD2Cl2 as solvent and internal

standard; shifts are referenced to tetramethylsilane (1H; 13C, X ¼19.867187 MHz) and 85% H3PO4 (

31P, X¼ 40.480742 MHz); ssatand dsat denote signals with satellites. Mass spectra were recor-

ded on a MAT 95 XL Finnigan (EI, 70 eV, 184W) spectrometer

(selected data given). Elemental analyses were performed using

an ElementarVarioEL instrument.

Procedure for the synthesis of the complexes 5a,b and 6a,b: a

solution of 1.0 g (2.14 mmol) of 2H-azaphosphirene complex 1a,

dissolved in 10 ml of toluene, were transferred into an autoclave.

Carbon dioxide was then pressured into the autoclave (50 bar)

and the mixture was heated to 75 �C for 5 hours. After 5 hours

the mixture was cooled to room temperature and transferred into

a Schlenk vessel. After addition of 2 spatulas of silica gel all

volatile components were removed in vacuo (�1 � 10�2 mbar).

The residue was subjected to column chromatography at low

temperature (neutral SiO2, column dimensions 2� 8 cm,�30 �C,eluent: petroleum ether (50/70)/Et2O 95/5). From the (optically)

well separated bands the first fraction contained 6a (or 6b) which

was obtained and the second fraction contained 5a (or 5b); all

compounds were obtained as pale yellow solids after evaporation

of the solvents.

Complex 5a: yield: 113 mg (0.15 mmol, 14%): 31P{1H} NMR:

d¼ 308.1 (sSat); these and all other analytical data are as reported

before.33

Complex 6a: yield: 442 mg (0.53 mmol, 50%): 31P{1H} NMR:

d¼ 321.8 (sSat); these and all other analytical data are as reported

before.33

Complex 5b: yield: 94 mg (0,10 mmol, 12%); 31P{1H} NMR:

d ¼ 243.2 (sSat, J(W,P) ¼ 375.9, 328.1 Hz); these and all other

analytical data are as reported before.33

Complex 6b: yield: 385 mg (0.37 mmol, 45%); 31P{1H} NMR:

d ¼ 256.9 (sSat, J(W,P) ¼ 354.8, 325.0 Hz,); these and all other

analytical data are as reported before.33

Procedure for the synthesis of the complexes 7a,b and 8a,b: a

solution of 307mg (0.50mmol) of 2H-azaphosphirene complex 1a

and 212 mg (0.50 mmol) of 1b, dissolved in 10 ml of toluene, were

transferred into an autoclave. Carbon dioxide was then pressured

into the autoclave (50 bar) and themixturewas heated to 75 �C for

5 hours. After 5 hours the mixture was cooled to room tempera-

ture and transferred into a Schlenk vessel. After addition of 2

spatulas of silica gel all volatile components were removed in

vacuo (�1 � 10�2 mbar). The residue was subjected to column

chromatography at low temperature (neutral SiO2, column

dimensions 2 � 8 cm, �30 �C, eluent: petroleum ether (50/70)/

Et2O 95/5). Complexes 8a,b were obtained together with 7a,b, as

pale yellow compounds from the first fraction (easy visible band)

after evaporation. The mixture was then dissolved in Et2O and

stored at�30 �C over night. Complexes 8a,b crystallised together

This journal is ª The Royal Society of Chemistry 2012

with 7a,b. Therefore, the following analytical data were obtained

from NMR spectra from solutions of this mixture.

Complex 7a/b: 31P{1H} NMR: d ¼ 306.1 (dSat,2J(P,P) ¼

16.7 Hz), 244.1 (dSat,2J(P,P) ¼ 24.3 Hz; 1J(W,P) could not be

determined); these and all other analytical data are as reported

before.33Complex 8a/b: 31P{1H}NMR: d¼ 319.0 (dSat,2J(P,P)¼

16.7 Hz), 259.3 (dSat,2J(P,P) ¼ 16.7 Hz, 1J(W,P) ¼ 339.8 Hz);

these and all other analytical data are as reported before.33

Procedure for the synthesis of complex 11: a solution of

832 mg (1.45 mmol) of the [bis(trimethylsilyl)methyl]chlor-

ophosphane complex 9 and 256 mg (1.45 mmol) of 12-crown-4,

dissolved in 5 ml of Et2O, was added dropwise to a cooled

(�78 �C) solution of LDA (170 mg (1.59 mmol) in 5 ml Et2O).

The reaction mixture was stirred for 10 min at �78 �C before

warming to�60 �C. CO2 was bubbled through the yellow-orange

solution for about 1 min. An immediate colour change to col-

ourless and precipitation of LiCl was observed. The mixture was

warmed to room temperature (3 hours), filtered and dried in

vacuo (�1� 10�2 mbar). After extraction with n-pentane (10 ml),

the solution was concentrated to 3 ml and stored over night

at �25 �C. Complex 11 crystallised as a pale yellow product.

Complex 11: yield: 750 mg (1.13 mmol 78%); 1H NMR: d ¼0.05 (s, 9H, Si(CH3)3), 0.23 (s, 9H, Si(CH3)3), 0.87 (dd, 6H,3J(H,H) ¼ 6.7 Hz, iPr), 1.15 (d, 1H, 3J(H,H) ¼ 2.3 Hz, PCH),

1.29 (dd, 6H, 3J(H,H) ¼ 6.6 Hz, CHiPr), 2.99 (q, 1H, 3J(H,H) ¼6.6 Hz, CHiPr), 4.29 (q, 1H, 3J(H,H) ¼ 6.6 Hz, CHiPr), 8.40 (dd,

1H, 1J(P,H) ¼ 365.3 Hz, 1J(W,H) ¼ 9.5 Hz, 3J(H,H) ¼ 2.3 Hz,

PH); 31P{1H} NMR: d ¼ 83.59 (1J(W,P) ¼ 277.6); 13C{1H}

NMR: d ¼ 198.7 (d, 2J(P,C) ¼ 27.8 Hz; trans-CO), 196.9 (d,1J(W,C)¼ 125.8 Hz, 2J(P,C)¼ 7.5 Hz, cis-CO), 48.1 (s, iPr), 45.6

(s, iPr), 22.1 (d, 1J(P,C) ¼ 5.8 Hz PCH), 20.3 (d, 1J(P,C) ¼ 5.8

Hz, PCH), 20.2 (s, NCMe2), 1.87 (d, 3J(P,C) ¼ 3.3 Hz, SiMe3),

0.21 (d, 3J(P,C)¼ 3.3 Hz, SiMe3); MS (EI, 70 eV, 184W):m/z (%):

631 ([M]+, 40), 603 ([M � 1CO]+, 100), 575 ([M � 2CO]+, 45),

547 ([M � 3CO]+, 45), 517 ([M � 4CO]+, 63), 502 ([M � 5CO]+,

40); elemental analysis (%) calculated for C19H34NO7PSi2W: C

34.60, H 5.20, N 2.12; found: C 34.64, H 5.20, N 2.00%.

Computational details

Geometry optimizations, frequency and energetic calculations

were performed with the GAUSSIAN03 program suite51 at

B3LYP/6-311g(d,p), with the LanL2DZ basis set at W. Basis set

superposition errors (BSSE) were corrected with the counterpoise

method.All reported energies include correction for the zero point

energy (ZPE) term. Nuclear magnetic shieldings were calculated

using the GIAO method52 with ADF(2009.01)53 at VWNBP86/

TZ2P SO ZORA (LDA: VWN, gradient correction: Becke88/

Perdew86, SO ¼ spin orbit, ZORA ¼ zeroth order regular

approximation).54Minima and transition states of all compounds

were characterised by a number of zero or one imaginary

frequencies. RSEwere computed by single point (SP) calculations

at the LPNO-NCEPA1 level,55 using the def2-TZVPP basis set

and the geometries andZPE correction obtained at theB3LYP-D/

def2-TZVP level of theory,56 using the ORCA electronic structure

program package.57 Bond properties were computed using the

more polarised def2-TZVPP basis set. Electron densities at BCP

were computed with the AIM2000 program58 and WBI were

obtained from the Natural Bond Orbital (NBO) analysis.

This journal is ª The Royal Society of Chemistry 2012

X-Ray crystallographic analysis

Suitable yellow single crystals of 11 were obtained from concen-

trated n-pentane solutions upon decreasing the temperature from

ambient temperature to �25 �C. Data were collected on Nonius

Kappa CCD diffractometer equipped with a low-temperature

device (Cryostream,OxfordCryosystems) at 123Kusing graphite

monochromated Mo-Ka radiation (l ¼ 0.71073 �A). The struc-

tures were solved by Patterson methods (SHELXS-97) and

refined by full-matrix least squares on F2 (SHELXL-97).59

Crystal structure data of complex 11 (C19H34NO7PSi2W):

crystal size 0.24 � 0.24 � 0.22 mm, triclinic, P-1, a ¼10.7620(3), b ¼ 11.0051(3), c ¼ 25.8757(9) �A, a ¼ 85.0791(15)�,b ¼ 82.5425(16)�, g ¼ 64.6070(15)�, V ¼ 2743.59(14) �A3, Z ¼ 4,

rcalc ¼ 1.599 Mg m�3, 2qmax ¼ 29�, collected (independent)

reflections ¼ 30036 (13638), Rint ¼ 0.0391, m ¼ 4.391 mm�1, 589

refined parameters, 1 restraints, R1 (for I > 2s(I)) ¼ 0.0644,

wR2 (for all data) ¼ 0.0727, max./min. residual electron

density ¼ 1.996/�1.525 e �A�3.

The CCDC number 806428 (11) contains the supplementary

crystallographic data 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 or in the ESI.†

Acknowledgements

Financial support by the Deutsche Forschungsgemeinschaft

(STR 411/25-2 and 411/26-1) and the COST action CM0802

‘‘PhoSciNet’’ is gratefully acknowledged. We thank the J€ulich

Supercomputing Centre (JuRoPa@JSC; HBN12) and the

computing centre of the RWTH-Aachen for computing time.

G. S. thanks Prof. A. C. Filippou for support.

Notes and references

1 (a) M. Aresta, Carbon Dioxide as Chemical Feedstock, Wiley-VCH2010; (b) T. Sakakura, J.-C. Choi and H. Yasuda, Chem. Rev.,2007, 107, 2365–2387.

2 C. A. Tsipis and P. A. Karipidis, J. Phys. Chem. A, 2005, 109, 8560–8567.

3 D. J. Darensbourg, Inorg. Chem., 2010, 49, 10765–10780.4 D. E. Milligan and M. E. Jacox, J. Chem. Phys., 1962, 36, 2911–2917.

5 D. Kovacs and J. E. Jackson, J. Phys. Chem. A, 2005, 105, 7579–7587.6 P. L. Coe, A. Sellars, J. Colin Tatlow, G. Whittaker andH. C. Fielding, J. Chem. Soc., Chem. Commun., 1982, 362–363.

7 (a) G. Fachinetti, C. Floriani, A. Chiesi-Villa and C. Guastini, J. Am.Chem. Soc., 1979, 101, 1767–1775; (b) I. Castro-Rodriguez andK. Meyer, Chem. Commun., 2006, 1353–1368, and references citedtherein.

8 U. Jayarathne, P. Chandrasekaran, H. Jacobsen, J. T. Mague andJ. P. Donahue, Dalton Trans., 2010, 39, 9662–9671.

9 S. Gambarotta, C. Floriani, A. Chiesi-Villa and C. Guastini, J. Am.Chem. Soc., 1985, 107, 2985–2986.

10 S. Komiya, M. Akita, N. Kasuga, M. Hirano and A. Fukuoka,J. Chem. Soc., Chem. Commun., 1994, 1115–1116.

11 M. Aresta, C. F. Nobile, V. G. Albano, E. Forni and M. Manassero,J. Chem. Soc., Chem. Commun., 1975, 636–637.

12 J. Li and Z. Lin, Organometallics, 2009, 28, 4231–4234.13 C. H. Lee, D. S. Laitar, P. Mueller and J. P. Sadighi, J. Am. Chem.

Soc., 2007, 129, 13802–13803.14 R. Becerra, J. P. Cannady and R. Walsh, J. Phys. Chem. A, 2002, 106,

4922–4927.15 A. Herize, J. R. Mora, J. Lezama, E. Marquez, T. Cordova and

G. Chuchani, J. Phys. Org. Chem., 2009, 22, 170–179.16 S. P. So and W.-K. Li, J. Phys. Chem. A, 2004, 108, 4002–4007.

Chem. Sci., 2012, 3, 3526–3533 | 3531

17 V. Griffiths and J. C. Tebby, J. Chem. Soc., Chem. Commun., 1981,607–608.

18 J. Fournier, C. Bruneau and P. H. Dixneuf, Tetrahedron Lett., 1989,30, 3981–3982.

19 C. Bruneau and P. H. Dixneuf, J. Mol. Catal., 1994, 74, 97–107.20 Y. Kayaki, M. Yamamoto and T. Ikariya, J. Org. Chem., 2007, 72,

647–649.21 G. D. Frey, V. Lavallo, B. Donnadieu, W. W. Schoeller and

G. Bertrand, Science, 2007, 316, 439–449.22 (a) C. M. Fr€omming, E. Otten, G. Kehr, R. Fr€ohlich, S. Grimme,

D. W. Stephan and G. Erker, Angew. Chem., 2009, 121, 6770–6773;C. M. Fr€omming, E. Otten, G. Kehr, R. Fr€ohlich, S. Grimme,D. W. Stephan and G. Erker, Angew. Chem., Int. Ed., 2009, 48,6643–6646; (b) D. W. Stephan, Dalton Trans., 2009, 3129–3136.

23 R. C. Neu, E. Y. Ouyang, S. J. Geier, X. Zhao, A. Ramos andD. W. Stephan, Dalton Trans., 2010, 39, 4285–4294.

24 (a) D. W. Stephan and G. Erker, Angew. Chem., 2010, 122, 50–81; (b)X. Zhao and D. W. Stephan, Chem. Sci., 2012, 3, 2123–2132.

25 P. P. Power, Nature, 2010, 463, 171–177.26 L. Gu and Y. Zhang, J. Am. Chem. Soc., 2010, 132, 914–915.27 Y. Kayaki, M. Yamamoto and T. Ikariya, Angew. Chem., 2009, 121,

4258–4261; Y. Kayaki, M. Yamamoto and T. Ikariya, Angew. Chem.,Int. Ed., 2009, 48, 4194–4197.

28 (a) A. Berkefeld, W. E. Piers andM. Parvez, J. Am. Chem. Soc., 2010,132, 10660–10661; (b) G. M�enard and D. W. Stephan, Angew. Chem.,2011, 123, 8546–8549; G. M�enard and D. W. Stephan, Angew. Chem.,Int. Ed., 2011, 50, 8396–8399; (c) G. M�enard and D. W. Stephan,J. Am. Chem. Soc., 2010, 132, 1796–1797; (d) A. E. Ashley,A. L. Thompson and D. O’Hare, Angew. Chem., 2009, 121, 10023–10027; A. E. Ashley, A. L. Thompson and D. O’Hare, Angew.Chem., Int. Ed., 2009, 48, 9839–9843.

29 V. Nair, V. Varghese, R. R. Paul, A. Jose, C. R. Sinu andR. S. Menon, Org. Lett., 2010, 12, 2653–2655.

30 (a) L. J. Hounjet, C. B. Caputo and D. W. Stephan, Angew. Chem.,2012, 124, 4792–4795; (b) For reactions of electron-richtrisaminophosphanes with carbon dioxide, see: R. W. Light,L. D. Hutchins, R. T. Paine and C. F. Campana, Inorg. Chem.,1980, 19, 3597–3604; (c) For reactions of acyclicamidophosphoranes with carbon dioxide, see: R. G. Cavell,K. I. The and L. V. Griend, Inorg. Chem., 1981, 20, 3813–3818.

31 (a) K. Issleib and H. Weichmann, Chem. Ber., 1964, 97, 721–724; (b)K. Diemert, T. Hahn and W. Kuchen, J. Organomet. Chem., 1994,476, 173–181; (c) K. Diemert, T. Hahn, W. Kuchen, D. Mootz,W. Poll and P. Tommes, Z. Naturforsch., B: J. Chem. Sci., 1995,50, 209; (d) the parent phosphanyl carboxylic acid was characterisedonly as [(CO)5Cr{H2PCO2H}] (ref. 2 and 3).

32 R. Streubel, F. Ruthe and P. G. Jones, Eur. J. Inorg. Chem., 1998,571–574.

33 R. Streubel, A. Ostrowski, S. Priemer, U. Rohde, J. Jeske andP. G. Jones, Eur. J. Inorg. Chem., 1998, 257–261.

34 Reactions of 1a, b in toluene at 75 �C and atmospheric pressure ofCO2 were unselective; among the products formed are those knownto be formed under argon atmosphere (otherwise unchangedconditions).

35 (a) F. Mathey, N. H. Tran Huy and A. Marinetti, Helv. Chim. Acta,2001, 84, 2938–2957; (b) F. Mathey, Angew. Chem., 2003, 115, 1616–1643; F. Mathey, Angew. Chem., Int. Ed., 2003, 42, 1578–1604; (c)K. Lammertsma, Top. Curr. Chem., 2003, 229, 95–119; (d)H. Aktas, C. J. Slootweg and K. Lammertsma, Angew. Chem.,2010, 122, 2148–2159; H. Aktas, C. J. Slootweg andK. Lammertsma, Angew. Chem., Int. Ed., 2010, 49, 2102–2113.

36 (a) C. Schulten, G. von Frantzius, G. Schnakenburg and R. Streubel,Heteroat. Chem., 2011, 22, 275–286; (b) for 1,3-dioxo-2s4l5,4s4l5-diphosphetanes, see: H. M. Schiffner, PhD thesis, University ofBonn, 1991.

37 A. €Ozbolat, G. von Frantzius, J. M. Perez, M. Nieger and R. Streubel,Angew. Chem., 2007, 119, 9488–9491; A. €Ozbolat, G. von Frantzius,J. M. Perez, M. Nieger and R. Streubel, Angew. Chem., Int. Ed.,2007, 46, 9327–9330.

38 R. Streubel, M. Bode, J. M. P�erez, G. Schnakenburg, J. Daniels andP. G. Jones, Z. Anorg. Allg. Chem., 2009, 635, 1163–1171.

39 R. Streubel, S. Priemer, F. Ruthe and P. G. Jones, Eur. J. Inorg.Chem., 2000, 1253–1259.

40 T. P.M.Goumans,A.W.Ehlers,K.LammertsmaandE.U.W€urthwein,Eur. J. Org. Chem., 2003, 2941–2946, here, the relative stability of

3532 | Chem. Sci., 2012, 3, 3526–3533

phosphiranone was compared to its methylene–oxaphosphirane isomer,but their RSE values were not reported.

41 For recent examples see: (a) C. Bleiholder, F. Rominger andR. Gleiter, Organometallics, 2009, 28, 1014–1017; (b) A. Noor,G. Glatz, R. M€uller, M. Kaupp, S. Demeshko and R. Kempe, Nat.Chem., 2009, 1, 322–325; (c) P. R. Schreiner, H. P. Reisenauer,J. Romanski and G. Mloston, Angew. Chem., Int. Ed., 2009, 48,8133–8136; (d) H. Helten, S. Fankel, O. Feier-Iova, M. Nieger,A. Espinosa Ferao and R. Streubel, Eur. J. Inorg. Chem., 2009,3226–3237; (e) F. Ot�on, I. Ratera, A. Espinosa, K. Wurst,T. Parella, A. T�arraga, J. Veciana and P. Molina, Chem.–Eur. J.,2010, 16, 1532–1542; (f) M. Feller, E. Ben-Ari, M. A. Iron,Y. Diskin-Posner, G. Leitus, L. J. W. Shimon, L. Konstantinovskiand D. Milstein, Inorg. Chem., 2010, 49, 1615–1625; (g)H. Braunschweig, K. Radacki and K. Schwab, Chem. Commun.,2010, 46, 913–915; (h) J. D. Epping, S. Yao, M. Karni, Y. Apeloigand M. Driess, J. Am. Chem. Soc., 2010, 132, 5443–5455; (i)A. Espinosa and R. Streubel, Chem.–Eur. J., 2011, 17, 3166–3178;(j) A. Espinosa, C. G�omez and R. Streubel, Inorg. Chem., 2012, 51,7250–7256; (k) A. Espinosa and R. Streubel, Chem.–Eur. J., 2012,DOI: 10.1002/chem.201201057.

42 K. B. Wiberg, Tetrahedron, 1968, 24, 1083–1096.43 For recent examples applied to intramolecular bonding see: (a) J. Ran

and M. W. Wong, Aust. J. Chem., 2009, 62, 1062–1067; (b) H. Wang,I. G. Csizmadia, I. Marsi, G. A. Chasse, D. Fang and B. Viskolcz,J. Chem. Phys., 2009, 131, 035105/1–035105/8; (c) F. Ot�on,A. Espinosa, A. T�arraga, I. Ratera, K. Wurst, J. Veciana andP. Molina, Inorg. Chem., 2009, 48, 1566–1576; (d) H. Raissi,A. F. Jalbout, B. Abbasi, F. Fazli, F. Farzad, E. Nadim and A. deLeon, Int. J. Quantum Chem., 2010, 110, 893–901.

44 R. F. W. Bader, in Atoms in Molecules: A Quantum Theory, OxfordUniversity Press, Oxford, 1990.

45 (a) W. T. Taylor andK. S. Pitzer, J. Res. Natl. Bur. Stand., 1947, 38, 1;(b) P. G. Maslov, Dokl. Akad. Nauk SSSR, 1949, 67, 819; (c)J. C. Decius, J. Chem. Phys., 1963, 38, 241–248; (d) K. Brandhorstand J. Grunenberg, ChemPhysChem, 2007, 8, 1151–1156; (e)K. Brandhorst and J. Grunenberg, Chem. Soc. Rev., 2008, 37,1558–1567.

46 C. J. M. Stirling, Pure Appl. Chem., 1984, 56, 1781–1796.47 O. Krahe, F. Neese and R. Streubel, Chem.–Eur. J., 2009, 15, 2594–

2601. RSE computed at SCS-MP2/def2-TZVPP//BP86/def2-TZVP.48 P. George, M. Trachtman, C. W. Bock and A. M. Brett, Tetrahedron,

1976, 32, 317–323.49 D. V. Deubel, J. Am. Chem. Soc., 2004, 126, 996–997.50 E. Niecke, M. Engelmann, H. Zorn, B. Krebs and G. Henkel, Angew.

Chem., 1980, 92, 738–739; E. Niecke, M. Engelmann, H. Zorn,B. Krebs and G. Henkel, Angew. Chem., Int. Ed. Engl., 1980, 19,710–712.

51 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone,B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li,H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng,J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda,J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao,H. Nakai, T. Vreven, J. A. Montgomery, Jr, J. E. Peralta,F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin,V. N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari,A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi,N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B. Cross,V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann,O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski,R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth,P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels,€O. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski, andD. J. Fox, Gaussian 03, Revision E.01, Gaussian, Inc., WallingfordCT, 2009.

52 (a) G. Schreckenbach and T. Ziegler, J. Phys. Chem., 1995, 99, 606; (b)G. Schreckenbach and T. Ziegler, Int. J. Quantum Chem., 1997, 61,899; (c) S. K. Wolff and T. Ziegler, J. Chem. Phys., 1998, 109, 895.

53 ADF2009.01, SCM, Theoretical Chemistry, Vrije Universiteit,Amsterdam, The Netherlands, http://www.scm.com.

54 (a) E. van Lenthe, E. J. Baerends and J. G. Snijders, J. Chem. Phys.,1993, 99, 4597; (b) E. van Lenthe, E. J. Baerends and J. G. Snijders,J. Chem. Phys., 1994, 101, 9783; (c) E. van Lenthe, J. G. Snijdersand E. J. Baerends, J. Chem. Phys., 1994, 105, 6505–6516; (d)

This journal is ª The Royal Society of Chemistry 2012

E. van Lenthe, R. van Leeuwen, E. J. Baerends and J. G. Snijders, Int.J. Quantum Chem., 1996, 57, 281; (e) C. Fonseca Guerra,J. G. Snijders, G. te Velde and E. J. Baerends, Theor. Chem. Acc.,1998, 99, 391; (f) E. van Lenthe, A. E. Ehlers and E. J. Baerends,J. Chem. Phys., 1999, 110, 8943; (g) G. te Velde, F. M. Bickelhaupt,S. J. A. van Gisbergen, C. Fonseca Guerra, E. J. Baerends,J. G. Snijders and T. Ziegler, J. Comput. Chem., 2001, 22, 931.

55 F. Wennmohs and F. Neese, Chem. Phys., 2008, 343, 217–230.56 Including the Grimme’s damped semiempirical correction accounting

for the major part of the contribution of dispersion forces to theenergy: (a) S. Grimme, J. Comput. Chem., 2004, 25, 1463–1476; (b)S. Grimme, J. Comput. Chem., 2006, 27, 1787–1799.

This journal is ª The Royal Society of Chemistry 2012

57 ORCA – An ab initio, DFT and semiempirical SCF-MO package.Written by F. Neese, Max Planck Institute for BioinorganicChemistry, D-45470 M€ulheim/Ruhr, 2012, Version 2.9.0, http://www.mpibac.mpg.de/bac/logins/neese/description.php; for a recentupdate, see: F. Neese, WIREs Comput. Mol. Sci., 2012, 2, 73–78.

58 AIM2000 v. 2.0, designed by F. Biegler-K€onig, and J. Sch€onbohm,2002, http://www.aim2000.de/(a) F. Biegler-K€onig, J. Sch€onbohmand D. Bayles, J. Comput. Chem., 2001, 22, 545–559; (b) F. Biegler-K€onig and J. Sch€onbohm, J. Comput. Chem., 2002, 23, 1489–1494.

59 (a) SHELXS-97,Acta Crystallogr., Sect. A: Found. Crystallogr., 1990,46, 467; (b) G. M. Sheldrick, SHELXL-97; University of G€ottingen,1997.

Chem. Sci., 2012, 3, 3526–3533 | 3533