Synthetic and computational study of geminally bis(supermesityl) substituted phosphorus compounds

Dalton

Transactions

View Article Online

View Journal | View Issue

PAPER

Synthetic and computational study of geminally bis-

(supermesityl) substituted phosphorus compounds†‡

Cite this: Dalton Trans., 2013, 42, 1437

Conor G. E. Fleming, Alexandra M. Z. Slawin, Kasun S. Athukorala Arachchige,

Rebecca Randall, Michael Bühl and Petr Kilian*

Reaction chemistry of an extremely sterically encumbered phosphinic chloride (Mes*)₂P(vO)Cl (Mes* =

2,4,6-tri-t-butylphenyl, supermesityl) was investigated. This compound, as well as other compounds

bearing two supermesityl groups placed geminally at the central phosphorus atom, shows extremely low

reactivity at the phosphorus centre. Nevertheless, some synthetically signiﬁcant transformations were

possible. Reduction with hydridic reagents under forcing conditions yielded the phosphine oxide

(Mes*)₂P(vO)H and a secondary phosphine Mes*(2,4-tBu₂C₆H₃)PH. Deprotonation of (Mes*)₂P(vO)H

gave the corresponding phosphinite, which aﬀorded very crowded tertiary phosphine oxides (Mes*)₂P-

(vO)R (R = Me and Et) on reactions with electrophiles. While the reaction of the phosphine Mes*(2,4-

tBu₂C₆H₃)PH with sulfur was surprisingly facile (although under forcing conditions), we have been

unable to chlorinate or deprotonate this phosphine. All new compounds were fully characterised with

multinuclear NMR, IR, Raman, MS, microanalyses and single crystal X-ray diﬀraction. Our computations

(B3LYP and M06-2X level) show that strain energies of (synthetically accessible) geminally substituted

compounds are extremely high (180 to 250 kJ mol⁻¹), the majority of the strain is stored as boat distor-

tions to the phenyl rings in Mes* substituents.

Received 28th August 2012,

Accepted 2nd November 2012

DOI: 10.1039/c2dt31971c

www.rsc.org/dalton

as a principal precursor for introduction of Mes* groups into a

wide range of species.^4,5The second reason is the specific geo-

Introduction

The preparation of the first stable diphosphene Mes*PvPMes* metry of Mes* groups, with flanking ortho-positioned tBu

by Yoshifuji in 1981¹led to a flourishing of activity in the groups which impart good “encapsulating” properties. To

research of low coordinate and multiple bonded main group date, the Cambridge Structural Database holds ca. 1500 struc-

species, resulting in a wide recognition of the key role of extre- tures containing the supermesityl group; interestingly almost

mely bulky substituents (initially mainly Mes*, 2,4,6-tri-t-butyl- half of these contain phosphorus atoms directly connected to

phenyl, supermesityl) in the stabilisation of peculiar bonding Mes* groups (i.e. Mes*–P motif).

and generation of unusual reactivity. Sterically protective

The geometry of Mes* substituents is particularly favour-

groups continue to be used extensively in the low coordination able for preventing oligomerisation of low-valent species,

main group and organometallic chemistry, as well as in the however the clash of ortho-positioned t-Bu groups from two

stabilisation of open shell species.²In addition to supermesityl Mes* groups makes geminal disubstitution a challenging

many other extremely bulky groups have been developed over (however not impossible) task. We were interested to find out

the past few decades, such as terphenyls or yet bulkier deriva- how the extreme steric shielding in geminally disubstituted

tives of supermesityl.³

species changes the reactivity patterns of the central phos-

Nevertheless, supermesityl remains a popular choice of phorus atom. Furthermore, we wanted to probe whether the

sterically protective groups. There are two major reasons for excessive shielding could confer stability to normally fleeting

this. The first is the ease of synthesis of Mes*Br, which serves two coordinated phosphorus species, such as the phosphe-

nium (Mes*₂P)⁺or the phosphinyl radical (Mes*₂P). Our survey

showed that the first literature example of a phosphorus

species bearing two geminal Mes* groups was the phosphinic

chloride Mes*₂P(vO)Cl 1.^6,7Compound 1 was made as early

as 1979 by Yoshifuji via the reaction of Mes*Li with POCl₃.

This is in contrast with the related reaction of two molar

equivalents of Mes*Li with PCl₃which did not give geminally

disubstituted Mes*₂PCl. Instead Mes*PCl₂and Mes*PvPMes*

EaStChem School of Chemistry, University of St Andrews, Fife KY16 9ST, St Andrews,

UK. E-mail: pk7@st-andrews.ac.uk; http://chemistry.st-andrews.ac.uk/staﬀ/pk/group/

†Dedicated to Professor David Cole-Hamilton on the occasion of his retirement

and for his outstanding contribution to transition metal catalysis.

‡Electronic supplementary information (ESI) available. CCDC 898340–898344

and 898346. For ESI and crystallographic data in CIF or other electronic format

see DOI: 10.1039/c2dt31971c

This journal is © The Royal Society of Chemistry 2013

Dalton Trans., 2013, 42, 1437–1450 | 1437

View Article Online

Paper

Dalton Transactions

were obtained.⁸In addition, geminally disubstituted thiopho- yields) by reactions of arylcopper(^I) reagents with bulky diaryl-

sphinic chloride Mes*₂P(vS)Cl was not detected when Mes*Li phosphine chlorides.¹⁹

was reacted with PSCl₃.⁶There is only one other successful syn-

In summary, there seems to be only a narrow window of

thesis mentioned in the literature of a compound with a phos- synthetic opportunity for attaching two Mes* groups on the

phorus centre geminally substituted by two Mes* groups, phosphorus centre geminally, which is essentially represented

which is the phosphaalkene Mes*–PvCH–P(Mes*)₂documen- by the reaction of Mes*Li with POCl₃(giving phosphinic chlor-

ted by Karsch. Unfortunately, no synthetic and limited spectro- ide 1). Therefore in our eﬀorts to disclose the chemistry of

scopic data (³¹P NMR only) of the phosphaalkene were these extremely sterically encumbered phosphorus species we

reported.⁹Additionally, single crystal diﬀraction data for employed 1 as our principal synthon.

Mes*₂PH were deposited with CCDC as a private communi-

cation but neither physical data nor synthetic methods had

been oﬀered to date.¹⁰

The reactivity of 1 was studied in some detail previously.

Unexpected reactivity was reported in early work by Majoral,

Results and discussion

Reactivity of Mes*₂P(O)Cl

the azide Mes*₂P(vO)(N₃) is not formed in the reaction of 1

with NaN₃. Instead intramolecular dehydrohalogenation

occurs with formation of phosphindole 6 as a major product.¹¹

Mes*₂P(O)Cl 1 was synthesised by a literature method, which

involves the reaction of two equivalents of Mes*Li with

POCl₃.^6,7In our hands, yields of 1 up to 65% were achieved

after purification by column chromatography on silica. The lit-

erature NMR data were collected at low working frequencies

(¹H at 90 MHz, ¹³C at 25.2 MHz);⁶we provide higher quality

data with detailed assignments in the Experimental section.

Both ¹H and ¹³C NMR spectra are complex due to a number of

prochiral groups present in the molecule; for example all four

aromatic hydrogen atoms are anisochronous (δ_H7.47, 7.36,

1

was used as a source of the phosphonyl radical

Mes*₂P˙(vO) in several ESR studies. The phosphonyl radical

was formed by UV irradiation of 1 in the presence of electron

rich olefins,^12–14or by X-ray irradiation of single crystals of 1.¹⁵

Tordo has used also the phosphine oxide Mes*₂P(vO)H 2 as a

precursor for generation of the same (phosphonyl) radical,

however no synthetic data of 2 were provided and the reported

characterisation is limited to (incomplete) ³¹P NMR data.^14a

The transient formation of the related phosphinyl radical

Mes*₂P˙ was observed (by ESR) in photolysis of 1 and also in

the reaction of Mes*PCl₂with Mes*Li.¹²Compound 1 was

subject to several theoretical studies owing to its peculiar

structure.^16,17

An interesting illustration of the profound eﬀect of rela-

tively small change in the bulk of the protective group is seen

in the contrasting chemistry of Mes* groups and 2,4,6-triiso-

propylphenyl (Trip) groups. Thus the reaction of 2,4,6-triiso-

propylphenylcopper(^I) with pnictogen trihalides aﬀorded the

respective tris(2,4,6-triisopropylphenyl) phosphine, arsine,

stibine and bismuthine in moderate to low yields,¹⁸whilst

only monosubstitution on the P atom was observed in the reac-

tion of Mes*Li with PCl₃(see above). Moreover, a range of

extremely crowded tertiary phosphines with at least two ter-

phenyl or Trip substituents were synthesised (albeit in low

1

7.24 and 7.15 ppm) in H NMR of 1. Since there is no formal

stereogenic centre in 1, the anisochronicity of the prochiral

groups and atoms is likely to be a consequence of a hindered

or blocked rotation around both P–C bonds.

Our initial reactivity studies of 1 focused on reduction. A

range of hydridic reducing reagents was screened (Table 1)

with a view to obtain phosphorus(^III) species as desired syn-

thons for further transformations. Compound 1 turned out to

be very inert; no reaction was observed after heating 1 with

Ph₃SiH, NaBH₄or NaH in THF for several hours under reflux.

Notably, heating 1 with HSiCl₃/Et₃N (a widely employed

reagent in reduction of phosphine oxides)²⁰under reflux in

toluene also resulted in a recovery of the starting material

exclusively. On the same note, no reaction was observed when

Mes*₂P(O)Cl 1 was treated with LiAlH₄, [iBu₂AlH]₂(DIBAL) or

LiBEt₃H (superhydride) in THF at r.t. for several hours

(Table 1). Nevertheless, in boiling THF 1 reacts slowly with

Table 1 Reaction conditions for reduction of Mes*₂P(O)Cl 1

Reducing agent

Solvent

Temperature (°C)

Time (hours)

Major product

LiAlH₄

[iBu₂AlH]₂

LiBEt₃H

THF

RT

10

No reaction

NaBH₄

NaH

Ph₃SiH

HSiCl₃/Et₃N

LiAlH₄

LiAlH₄+ Me₃SiCl

LiAlH₄

LiAlH₄+ AlCl₃

LiAlH₄+ Me₃SiCl

LiAlH₄+ CeCl₃

THF

Toluene

THF

65 (reflux)

110 (reflux)

65 (reflux)

101 (reflux)

6

4

No reaction

Mes*₂P(O)H 2

Mes*(2,4-tBu₂C₆H₃)PH 3

Mes*PH₂

6 to 9

5

6

7 to 9

5

6

THF

1,4-Dioxane

1438 | Dalton Trans., 2013, 42, 1437–1450

This journal is © The Royal Society of Chemistry 2013

View Article Online

Dalton Transactions

Paper

be a result of hindered rotation around the P–C bonds. All

signals in both the aliphatic and aromatic regions sharpen sig-

nificantly at −60 °C (Fig. 1). At this temperature the four o-tBu

groups become distinctly anisochronic, while three distinct

singlets (1 : 2 : 1 intensity ratio) are observed for the four aro-

matic protons. In addition to ³¹P, ¹H and ¹³C NMR, Mes*₂-

P(O)H 2 was fully characterised by IR, Raman and ES MS, and

its purity was verified by microanalyses. It has also been

characterised by a single crystal X-ray diﬀraction (see below).

When Mes*₂P(O)Cl 1 was reacted with mixed reducing

reagents LiAlH₄/Me₃SiCl or LiAlH₄/AlCl₃in a high boiling ethe-

real solvent (1,4-dioxane, b.p. 101 °C) for a prolonged period

the reduction to a trivalent phosphorus species was achieved.

Somewhat unexpectedly, the major product of these reactions

was Mes*(2,4-tBu₂C₆H₃)PH 3 rather than Mes*₂PH (Scheme 1,

Table 1). Thus in this reaction loss of one of the flanking t-Bu

groups on the aromate occurs in addition to the reduction to a

trivalent phosphorus species. Such a C–C bond splitting is for-

mally a reverse Friedel–Crafts reaction, which however nor-

mally requires a source of protons in addition to the Lewis

acid. The loss of the t-Bu group results in a significant relax-

ation of the steric crowding within the molecule of 3 (see

Table 3 and X-ray discussion), which is likely to be the driving

force for the reaction. Purification of the crude product from

the reaction by column chromatography aﬀorded 3 in yields of

up to 72% (from the LiAlH₄/Me₃SiCl reaction). Despite the loss

of the t-butyl group, the phosphorus atom in 3 remains

strongly shielded (δ_P−67.1) vs. that in (non-crowded) Ph₂PH

(δ_P−40.2). Mes*(2,4-tBu₂C₆H₃)PH 3 was fully characterised by

³¹P, ¹H and ¹³C NMR, IR, Raman and ES MS, its purity was ver-

ified by microanalyses. It has also been characterised by a

single crystal X-ray diﬀraction (see below).

Scheme 1 Reduction reactions of Mes*₂P(O)Cl 1.

LiAlH₄to give the corresponding secondary phosphine oxide

Mes*₂P(O)H 2 (Scheme 1). The reaction is slightly accelerated

by an addition of stoichiometric amounts of Me₃SiCl, or by

replacing the THF solvent with (higher boiling) 1,4-dioxane.

Using the THF solution of LiAlH₄(rather than the solid form)

did not result in improved yields or shorter reaction times.

The reaction was monitored by ³¹P{¹H} NMR, and was deemed

complete when all the starting material was consumed. ³¹P

and ¹H NMR spectra of the mixture after the reaction indicated

that whilst 2 was a major product in these reductions, signifi-

cant amounts of P–C bond cleavage products Mes*H, Mes*PH₂

and Mes*P(O)H₂were also co-formed. Purification by column

chromatography on silica aﬀorded 2 in good purity with yields

of up to 56%.

Mes*₂P(O)H 2 has been mentioned in the literature pre-

viously, it has been used as a precursor for phosphonyl radical

Mes*₂P(O).^14aHowever, no details were given on its prep-

1

aration and apart from the magnitude of the J(PH) coupling

(478 Hz)^14ano other characterisation data were reported. The

³¹P NMR data of Mes*₂P(O)H 2 acquired by us (δ_P12.9, J_HP

=

1

483.0 Hz) indicate that the phosphorus atom in Mes*₂P(O)H is

only slightly more shielded vs. that in the related non-crowded

1

phosphine oxide Ph₂P(O)H (δ_P21.5, J_HP= 480.6 Hz),²¹whilst

We note that the reaction conditions used in the prep-

aration of both 2 and 3 are rather harsh when compared to

those used in analogous reductions of the less sterically hin-

dered secondary phosphine oxides. For example, reduction of

the phosphine oxide Ph₂P(O)H to the respective phosphine

Ph₂PH using DIBAL proceeded within 10 min at r.t. with a

yield of 86%.²²There are many additional examples of clean

phosphine oxide reductions using the reducing reagents listed

in Table 1 under mild conditions.²³

the magnitude of ¹J(PH) is very similar in both these com-

pounds. All lines in ¹H and ¹³C NMR spectra of Mes*₂P(O)H

are significantly broadened at r.t. (see Fig. 1), this is likely to

Sato reported that the reduction of secondary phosphine

oxides R₂P(vO)H to secondary phosphine–borane adducts

R₂PH·BH₃is achieved using LiAlH₄/NaBH₄/CeCl₃reducing

reagent at r.t. with good yields.²⁴We have attempted to reduce

1 using a LiAlH₄/CeCl₃mixed reducing reagent (in 1,4-

dioxane), however no reaction was observed at r.t., whilst

heating in a boiling solvent resulted in extensive P–C bond

cleavage and formation of Mes*PH₂as the major product.

In order to further test the synthetic utility of 1, we have

attempted its reaction with alkyllithiums. Phosphinic chlor-

ides R₂P(O)Cl react with organometallics (such as R′MgX or

R′Li) to give the corresponding phosphine oxides R₂R′P(O),

albeit often only with moderate yields.²⁵Interestingly, no reac-

tion has taken place between Mes*₂P(O)Cl 1 and MeLi after

heating under reflux in hexanes for 90 minutes as judged by

Fig. 1 ¹H NMR spectra of 2 at 298 K (top) and 213 K (bottom) showing

ﬂuxional character of aromatic (left) and aliphatic (right) signals. The doublet

with a large coupling constant on the very left is due to the P–H hydrogen

atom. Intensity of the left part of spectra was increased for clarity. CD₂Cl₂was

used as a solvent, working frequency was 499.9 MHz.

This journal is © The Royal Society of Chemistry 2013

Dalton Trans., 2013, 42, 1437–1450 | 1439

View Article Online

Paper

Dalton Transactions

Scheme 2 The reactions of the phosphine oxide Mes*₂P(O)H 2.

³¹P NMR spectroscopy, further illustrating inertness of the iodide Mes*₂P(O)I respectively. However, the latter reaction

crowded phosphinic chloride 1.

proceeded cleanly (indicated by ³¹P NMR) to a phosphindole 6,

which has been isolated as a colourless solid in good yields

(70%). We have also observed that phosphindole 6 is formed

via intramolecular quenching when the THF solution of the

phosphinite 2a is left at room temperature for several hours

(without the addition of the electrophile). Related cyclisation

reactions leading to inclusion of the proximate tBu group in

Mes* into a new ring have been reported in the literature,^8,26

namely phosphindole 6 was obtained by an intramolecular

dehydrohalogenation of Mes*₂P(O)Cl 1 in pyridine/DMF.¹¹The

Reactivity of Mes*₂P(O)H 2

Treating THF solution of Mes*₂P(O)H 2 with nBuLi at 0 °C

gives the corresponding phosphinite Mes*₂P(O)⁻2a. The

deprotonation is quantitative as indicated by ³¹P NMR spectra,

which show that at 0 °C the signal of the starting material is

replaced with

a new high frequency signal (δ_P106.4)

10 minutes after addition of nBuLi. Visually the formation of

the anion 2a results in a colour change from colourless to

rusty-red.

1

identity of 6 was established by ³¹P (δ_P54.9), H, ¹³C{¹H} NMR

and MS (ES+), our NMR data were in good agreement with the

literature.¹¹In addition, we also acquired single crystal X-ray

diﬀraction data of 6 (see below). We have only been able to

grow relatively small crystals of 6, hence our structural data are

rather poor. However they are adequate to demonstrate con-

nectivity and support assignment of the structure.

The reduction of the tertiary phosphine oxide 4 with hydri-

dic reducing reagents was investigated using similar con-

ditions to those used in reduction of Mes*₂P(O)Cl 1. The

reduction proceeded slowly with LiAlH₄/Me₃SiCl in boiling 1,4-

dioxane, however rather than the expected tertiary phosphine

Mes*₂PMe or possibly Mes*(2,4-tBu₂C₆H₃)PMe, a secondary

phosphine Mes*MePH was obtained as a major product (iso-

lated yield after column chromatography purification was 42 to

67%). In addition, a significant amount of another secondary

phosphine Mes*(2,4-tBu₂C₆H₃)PH 3 has also been formed in

these reactions. The identities of both Mes*MePH and 3 were

confirmed by ³¹P, ¹H and ¹³C NMR, and in the case of

In the next step, the reactivity of the phosphinite 2a towards

electrophiles was investigated. Thus 2a in THF reacted with

methyl iodide to give tertiary phosphine oxide Mes*₂P(O)Me 4

(Scheme 2). This reaction proceeds at room temperature, whilst

the analogous reaction with ethyl iodide to give Mes*₂P(O)Et 5

requires slight heating (50 °C) to be driven to completion. Pro-

gress of both reactions is easily monitored by observing a colour

change from rusty red to colourless/pale yellow.

While the conversion to 4 and 5 is almost quantitative as

judged by ³¹P NMR spectroscopy, analytically pure materials

were obtained after recrystallisation only in moderate to poor

yields (50 and 7%, respectively). Both 4 and 5 were fully charac-

1

terised by ³¹P, H and ¹³C NMR, IR, Raman and ES MS, their

purity was verified by microanalyses. Both 4 and 5 have also

been characterised by a single crystal X-ray diﬀraction.

Reactions of the in situ generated phosphinite Mes*₂P(O)⁻

2a with sterically more demanding alkyl halides benzylbro-

mide, iPrBr and tBuCl were performed under conditions

similar to those above. The ³¹P NMR spectra of the mixtures

after the reaction revealed that complex mixtures of phos-

phorus containing products were formed in each case.

Mes*MePH also by

a single crystal X-ray diﬀraction.

Mes*MePH was synthesised and well characterised earlier

(including X-ray diﬀraction),^27,28our ³¹P NMR data [δ_P(¹H

1

2

coupled, CDCl₃) −91.3 (d of q, J_HP= 225.8, J_HP= 5.0)] are in

Also the reactions of phosphinite 2a with bromine or iodine

did not give the expected phosphinic bromide Mes*₂P(O)Br or

good agreement with the literature.

1440 | Dalton Trans., 2013, 42, 1437–1450

This journal is © The Royal Society of Chemistry 2013

View Article Online

Dalton Transactions

Paper

Reactivity of Mes*(2,4-tBu₂C₆H₃)PH 3

Table 2 Reaction conditions used in attempted chlorinations of 3

Despite the loss of the tBu group from one of the Mes* groups,

the phosphorus atom in the secondary phosphine 3 remains

strongly sterically shielded. Therefore we decided to investigate

the reactivity of this formally P(^III) compound in some detail.

Phosphido anions obtained by deprotonation of secondary/

primary phosphines are important intermediates in a range of

transformations; the deprotonation reaction of 3 was therefore

of significant importance to us. However, we failed to deproto-

nate 3 by reacting it with nBuLi in hexane (with heating); only

starting material was observed in the ³¹P NMR spectrum of the

mixture after the reaction. Similarly, no deprotonation was

observed on treatment of ethereal solutions of 3 with LDA or

sBuLi at room temperature. A complex mixture of phosphorus

containing products was obtained when 3 was treated with

tBuLi at room temperature. Our observations are in line with

those made by Stelzer,²⁷who reported that the deprotonations

of Mes*RPH with R = Me, iPr, Ph proceeded smoothly at

−78 °C, whilst the sterically encumbered derivative Mes*-

MesPH did not undergo deprotonation with nBuLi even at

room temperature.

In addition to low acidity of the P-bonded hydrogen atom, 3

shows also low aﬃnity towards borane. Thus no reaction was

observed when 3 was treated with H₃B·SMe₂at room tempera-

ture or even on heating in toluene under reflux.

Extremely crowded chlorophosphines (R₂PCl) were shown

to be useful precursors for phosphenium radicals R₂P˙ via one

electron reductions.¹²In an attempt to gain access to crowded

chlorophosphine derived from 3 (i.e. Mes*(2,4-tBu₂C₆H₃)PCl

12, see Scheme 3), we have performed an extensive reactivity

scan of 3 with a variety of chlorinating reagents and conditions

(Table 2). Peculiar reactivity was observed. No reaction

occurred on exposure of solutions of 3 in chloroform, tetra-

chloromethane or 1,1,2,2-tetrachloroethane to sunlight for

Chlorination

agent

Temperature

Solvent (°C)

Major

product

Time

CHCl₃

CCl₄

Cl₂HC–CHCl₂

PCl₃

Cl₂CvO

Neat

RT

Weeks

No reaction

Toluene 110 (reflux)

Toluene RT

Toluene 110 (reflux)

5 hours No reaction

5 hours (2,4-tBu₂C₆H₃)

PCl₂7

Cl₂CvO +

NEt₃

Cl₂SvO

Toluene RT

1 hour

Complex mixture

1 hour

Complex mixture

several weeks. On the same note, no reaction was observed on

heating 3 with PCl₃in toluene, or on treating 3 with phosgene

(as toluene solution) at room temperature. On the other hand,

when 3 was treated with phosgene in boiling toluene, dichloro-

phosphine (2,4-tBu₂C₆H₃)PCl₂7 was obtained as the major

phosphorus containing product as indicated by ³¹P NMR

(δ_P166.9 ppm in CDCl₃, literature value δ_P167.2 in C₆D₆).²⁹

We have also attempted to perform the chlorination of 3 with

phosgene in the presence of triethylamine or using thionyl

chloride. Both these reactions aﬀorded complex mixtures of

products as revealed by ³¹P NMR.

In light of the extremely limited reactivity of 3 described

above, it appeared rather surprising to observe a clean reaction

of 3 with elemental sulfur, albeit under rather forcing con-

ditions (heating under reflux in Et₃N/toluene, Scheme 3). The

sulfide 8 was obtained in 45% yield after purification by

column chromatography. Interestingly, the phosphorus chemi-

cal shift observed in sulfide 8 (δ_P17.3 in CDCl₃) is very similar

to that in much less crowded Ph₂P(S)H (δ_P23.4 in CDCl₃).³⁰

8 was fully characterised by ³¹P, H and ¹³C NMR, IR, Raman

and ES MS (including exact mass determination); its purity

1

Scheme 3 The chlorination and thionation reactions of the secondary phosphine 3.

This journal is © The Royal Society of Chemistry 2013

Dalton Trans., 2013, 42, 1437–1450 | 1441

View Article Online

Paper

Dalton Transactions

Table 3 Bond distances (Å), angles (°) and distances from mean phenyl planes (Å) in compounds 2, 4, 5, 6, 3 and 8

2

4

5

6

3

8

Bond distances/Å

P(1)–E(1)^a

P(1)–C(1)

1.518(4)

1.809(4)

1.858(4)

1.525(4)

1.861(5)

1.854(5)

1.4888(13)

1.892(3)

1.864(3)

1.411(7)

1.854(9)

1.824(11)

—

1.853(2)

1.868(3)

1.9576(15)

1.842(3)

1.846(3)

P(1)–C(19)

Bond angles/°

C(1)–P(1)–C(19)

E(1)–P(1)–C(1)^a

E(1)–P(1)–C(19)^a

114.57(17)

110.89(18)

111.49(18)

111.5(2)

109.37(19)

112.12(19)

109.54(10)

111.55(9)

112.20(8)

108.6(5)

111.7(5)

111.5(5)

104.00(10)

—

110.23(13)

114.75(10)

119.59(9)

Distances from the mean phenyl plane/Å

C(1)^b

P(1)^b

0.106

0.658

0.137

1.076

0.367

0.069

0.517

0.532

0.132

0.976

0.146

1.001

0.491

0.564

0.603

0.684

0.101

0.745

0.164

1.076

0.452

0.290

0.715

0.625

0.145

1.205

0.017

0.117

0.473

0.527

0.178

0.187

0.053

0.351

0.001

0.043

0.221

0.175

0.078

—

0.011

0.249

0.038

0.196

0.012

0.065

0.142

—

C(19)^c

P(1)^c

C(7)^b

C(15)^b

C(25)^c

C(33)^c

^a2, 4, 5, 6 E = O, 8 E = S. ^bDistance from the phenyl mean plane fitted through atoms C(1) to C(6). ^cDistance from the phenyl mean plane fitted

through atoms C(19) to C(24).

was verified by microanalyses. The P–H bands in both IR and

Raman of 8 were extremely weak. Sulfide 8 has been also

characterised by a single crystal X-ray diﬀraction (see below).

Selenation of P(^III) species using elemental selenium gener-

ally requires harsher conditions than corresponding thiona-

tion. The reaction of 3 with grey selenium was attempted in

boiling toluene (both with and without the addition of Et₃N).

Unsurprisingly, starting materials were recovered exclusively.

X-ray discussion

The X-ray structures of 2, 4 and 5 are shown in Tables 3 and 5,

Fig. 2–5, respectively. Molecule 5 crystallises as solvate with

one molecule of dichloromethane, no co-crystallised solvent is

present in the crystals of 2 and 4. Analysis of the bond angles

around the central phosphorus atoms reveals only slight devi-

ations from tetrahedral geometry in 2, 4 and 5, the largest of

these deviations being C(1)–P(1)–C(19) angle in 2 (114.57(17)°).

The two Mes* groups in 2, 4 and 5 are positioned so that the

angles between the two mean phenyl planes are 33.0°, 21.3°

and 29.0° respectively. The phenyl rings in 2, 4 and 5 adopt a

boat form, with ipso-atoms C(1) and C(19) placed significantly

(0.10 to 0.16 Å, respectively) above their respective phenyl ring

mean planes. Moreover, the C(1)–P(1) and C(19)–P(1) bonds

are significantly bent out of these mean phenyl ring planes,

resulting in a P atom lying 0.658 to 1.076 Å above the mean

phenyl plane (see Fig. 5). The o-tBu groups are bent out slightly

to the other side of their respective phenyl mean planes, and

more significantly, they are also bent away from the phos-

phorus centre. The latter distortion is well observable as

expansion/compression of angles C(1)–C(6)–C(15) (127.47°)

and C(5)–C(6)–C(15) (115.16°) in 2; similar deviations are

observed for all four o-tBu groups in each 2, 4 and 5. As

expected, the p-tBu groups adopt normal geometry with their

quaternary carbon atoms lying close to their respective phenyl

ring planes.

Fig. 2 Molecular structure of 2. Thermal ellipsoids are set at 40% probability,

hydrogen atoms but H(1) are omitted for clarity. tBu groups are drawn as a wire-

frame for clarity.

Fig. 3 Molecular structure of 4. Thermal ellipsoids are set at 40% probability,

hydrogen atoms are omitted for clarity. tBu groups are drawn as a wireframe for

clarity.

As mentioned in the introduction, a crystal structure of

Mes*₂PH has been deposited with CSDS,¹⁰however no syn-

thetic or other details were given. The overall geometry of

1442 | Dalton Trans., 2013, 42, 1437–1450

This journal is © The Royal Society of Chemistry 2013

View Article Online

Dalton Transactions

Paper

Fig. 6 Molecular structure of phosphindole 6. Thermal ellipsoids are set at

40% probability, hydrogen atoms are omitted for clarity. tBu groups are drawn

as a wireframe for clarity.

Fig. 4 Molecular structure of 5. Thermal ellipsoids are set at 40% probability.

Hydrogen atoms and solvated molecules of CH₂Cl₂are omitted for clarity. tBu

groups are drawn as a wireframe for clarity.

Fig. 5 Partial view (Mes*P fragment) of the molecular structure of 5 to illus-

trate the extent of out-of-plane distortion within one of the Mes* groups. The

para-tBu group is drawn as a wireframe for clarity.

Fig. 7 Molecular structure of the sulﬁde 8. Thermal ellipsoids are set at 40%

probability, hydrogen atoms but H(1) are omitted for clarity.

Mes*₂PH is similar to that of 2, 4 and 5. Thus Mes*₂PH dis-

plays C–P–C angles 109.5° and 111.6° in the two molecules

within the unit cell, [109.54(10) in 5]. The maximum devi-

ations of the ipso-carbon and the phosphorus atom from the

respective phenyl mean plane are 0.126 and 0.806 Å, while

phenyl–phenyl interplanar angles are 41.6° and 44.8° in the

two molecules within the unit cell of Mes*₂PH.

The X-ray structure of phosphindole 6 is shown in Tables 3

and 5 and Fig. 6. The geometry around the P(1) atom is dis-

torted tetrahedral, with the angle C(19)–P(1)–C(36) being

slightly acute (96.3(5)°). The P(1)vO(1) bond length in 6 (1.411

(7) Å) is slightly shortened vs. those observed in 2, 4 and 5

(1.489(1)–1.525(4) Å). Notably, it is also shorter than the PvO

bond in the related non-crowded phosphine oxide Ph₂P(O)Me

(1.494(2) Å).³¹The phosphindole moiety within 6 is essentially

planar, with atom C(33) displaced by 0.18 Å from the mean

plane being the largest perturbation. The molecule adopts

similar overall geometry to that observed in 2, 4 and 5, with a

phenyl–phenyl interplanar angle of 44.2°. The Mes* group is

significantly distorted, it adopts (once again) a boat form with

an ipso-atom C(1) placed 0.145 Å above the mean phenyl

plane. The C(1)–P(1) bond is significantly bent out of its adja-

cent phenyl ring plane, more so than in 2, 4 and 5. This

results in P(1) lying 1.205 Å above this mean plane. As noted

earlier for 2, 4 and 5, all ortho-tBu groups in 6 are also bent

away from the central phosphorus atom.

The X-ray structure of the sulfide 8 is shown in Tables 3

and 5 and Fig. 7. The molecule possesses one less ortho-tBu

group vs. 2, 4 and 5, which allows it to adopt relatively relaxed

geometry with almost perpendicular orientation of the phenyl

rings with respect to each other (71.9° between the phenyl

mean planes). The phenyl ring within the Mes* group is

planar, whilst the phenyl ring within the 2,4-tBu₂C₆H₃group

displays only very moderate skewing towards a boat confor-

mation. Little or no out-of-plane displacement is observed for

the tBu groups, while the in-plane bending of ortho-tBu groups

is significant as seen from the diﬀerence between the two

(example) complementary angles C(1)–C(6)–C(15) (129.2(2)°)

and C(5)–C(6)–C(15) (113.2(3)°). The geometry around the

central P(1) atom is approximately tetrahedral. The P(1)–S(1)

bond length (1.958(2) Å) is identical to that found in the steri-

cally congested bis(alkyl) phosphine sulfide (tBu)₂P(H)vS

(1.9666(6) Å).³²

The X-ray structure of 3 is shown in Tables 3 and 5 and

Fig. 8. The molecule 3 adopts an overall relaxed geometry

This journal is © The Royal Society of Chemistry 2013

Dalton Trans., 2013, 42, 1437–1450 | 1443

View Article Online

Paper

Dalton Transactions

ortho-tBu groups away from the phosphorus atom is less pro-

nounced in comparison to those in 2, 4 and 5.

Computational section

In order to assess the strain in our geminally disubstituted

compounds we performed DFT calculations for 2, 4 and

selected model compounds. The B3LYP and M06-2X optimised

structures of 2 closely resemble the structure in the solid state,

which was used as a starting point. In particular the distor-

tions of the benzene rings are largely preserved in the gas-

phase equilibrium geometries.

Except in a molecular mechanics description, the “strain”

associated with distortions of bond distances, angles, torsions,

and nonbonded distances from optimum values³³is not

uniquely defined. We used the equations shown in Scheme 4

for strain evaluation; the main results of our computations are

shown in Table 4. The homodesmotic³⁴eqn (1) describes the

transfer of the tBu groups from two (presumably unstrained)

tri-t-butylbenzene molecules onto a mononuclear organophos-

phorus framework. We will use the energy associated with this

reaction, ΔE₁, as the most direct way to quantify steric strain. A

more indirect way, motivated by chemical reactivity, is pro-

vided by eqn (2a) and (2b), which represent the P–C bond

Fig. 8 Molecular structure of 3. Thermal ellipsoids are set at 40% probability,

hydrogen atoms but H(1) are omitted for clarity.

similar to that observed for 8. The phenyl rings are almost per-

pendicular (88.6°) to each other, thus largely avoiding the

mutual steric clashes of the ortho-tBu groups. The C(1)–P(1)–

C(19) angle in 3 is slightly compressed (104.0(1)°) vs. that in

the sulfide 8 (110.23(13)°). No significant distortions of the

phenyl rings are observed and the in-plane bending of the

Scheme 4 Isodesmic eqn (1) and bond dissociation eqn (2a) and (2b) used in computations.

Table 4 Strain energies in kJ mol⁻¹for compounds 2, 4 and 9 evaluated at the B3LYP-D3 level (in parentheses: M06-2X), both including zero-point energies

a

Compound

ΔE₁

ΔE_2a

ΔE_2b

ΔE₂= ΔE_2b− ΔE_2a

2

4

9

183.1 (186.4)

249.7

180.9

230.9 (240.2)

170.2

196.3

358.8 (379.6)

352.3

307.5

127.9 (139.4)

182.1

111.2

^aFor definition see eqn (1), (2a) and (2b) in Scheme 4.

1444 | Dalton Trans., 2013, 42, 1437–1450

This journal is © The Royal Society of Chemistry 2013

View Article Online

Dalton Transactions

Paper

to 266.2 kJ mol⁻¹at B3LYP+ZPE) showing that these weak

interactions play an important role in our compounds with a

large number of sub-van der Waals contacts.

X-ray diﬀraction data show that on going from 2 to 4 (i.e.

replacing the hydrogen atom with a methyl group), the two

supermesityl rings are pushed closer to each other. This is

seen from the closure of the C(1)–P(1)–C(19) bond angle from

114.6° (B3LYP 114.0°) to 111.5° (B3LYP 111.9°). As a conse-

quence the distortions in the P(Mes*)₂moiety are enhanced,

and the total strain evaluated through ΔE₁is increased to ca.

250 kJ mol⁻¹(Table 4).

Scheme 5 NMR numbering scheme for compounds 3 (E = none) and 8 (E = S).

To probe the diﬀerences in the strain between phosphine

oxides and phosphines, we performed computations also for

(Mes*)₂PH (9), i.e. the P(III) analogue of 2. Because the C–P–C

and C–P–H angles are smaller in phosphines than in phos-

phine oxides, increased repulsion between the Mes* groups

might be expected in 9 vs. 2. Indeed, the B3LYP optimized

C–P–C angle closes from 114.0° in 2 to 111.1° in 9. However,

little diﬀerence in strain is computed between 2 and 9 (see

ΔE₁values in Table 4). The diﬃculty to prepare the bulky

phosphine 9 (encountered in our experimental work) may

rather be related to its lower P–C_ArBDE compared to the phos-

phine oxide 2 (cf. ΔE_2avalues in Table 4).³⁵On the other hand,

an even lower P–C_ArBDE is predicted for 4, which was isolated

with little diﬃculties. Apparently other factors aﬀect kinetic

stability of 9, such as the spatial accessibility of the P centre.

In addition we also optimised (Mes*)₂P(vO)tBu, which is

unknown and could not be made through our synthetic meth-

odology (see above). The bulky tBu group pushes the

Mes* groups even closer, cf. the C_Ar–P–C_Arbond angle of

109.7°, and thus imparts even larger distortions on the

P(Mes*)₂moiety than H or Me substituents. Notably, the

optimised P–C_Arand P–C_Alkdistances in (Mes*)₂P(vO)tBu

exceed 1.9 Å and 2.0 Å, respectively [1.892(3) and 1.840(3) in 5].

Unsurprisingly, an excessively high amount of strain is pre-

dicted for this molecule, 359 kJ mol⁻¹according to eqn (1)

(B3LYP-D3 level).

dissociation reactions in the sterically encumbered title com-

pound (eqn (2a)) and in the related unstrained system

(eqn (2b)). The diﬀerence in the associated bond dissociation

energies (BDEs, D₀values) ΔE_2b− ΔE_2ashould to a large

extent be due to the strain in the bis(supermesityl) com-

pounds. However, we are aware that electronic eﬀects contrib-

ute as well, e.g. from the diﬀerent stabilisation of the P- and C-

centred radicals through the electron-rich supermesityl

groups, and will discuss only selected ΔE₂values for interpret-

ative purposes.

According to our calculations of ΔE₁values, the total strain

in the prototypical compound 2 is substantial, more than

180 kJ mol⁻¹(both at B3LYP-D3 and M06-2X levels). The

largest proportion of this energy is stored in the distortion of

the aryl rings; when the tBu groups in the B3LYP optimised

structure of 2 are replaced by H atoms (placed at the distances

found in the other aromatic C–H bonds), the resulting dis-

torted parent 2′ is higher in energy than the fully optimised 2′

by 170.6 kJ mol⁻¹(B3LYP level). Corresponding distortions in

compound 1 had been studied at a lower theoretical level (HF)

before, however dispersion interactions were not accounted for

in this early study.¹⁷It is noteworthy that the strain in 2,

evaluated through ΔE₁, increases significantly if dispersion

interactions are not taken into account (from 183.1 kJ mol⁻¹

Table 5 Crystal and structure reﬁnement data for compounds reported in this paper

Compound

2

3

4

5·CH₂Cl₂

6

8

Empirical formula

M

Crystal system

Space group

a/Å

b/Å

c/Å

α/°

C₃₆H₅₉OP

538.84

Orthorhombic

Pbca

10.7468(19)

20.780(5)

29.975(6)

90

6694(3)

8

C₃₂H₅₁

466.73

P

C₃₇H₆₁OP

552.86

Monoclinic

P2₁

11.039(4)

10.854(5)

14.610(6)

90

100.892(10)

90

1719.0(12)

2

C₃₈H₆₃OP CH₂Cl₂

651.82

Monoclinic

P2₁/c

20.653(4)

11.1027(18)

17.718(3)

90

111.573(4)

90

C₃₆H₅₇OP

536.82

Orthorhombic

P2₁2₁2₁

9.005(3)

11.305(4)

32.513(11)

90

C₃₂H₅₁PS

499.79

Monoclinic

P2₁/c

17.188(6)

10.133(4)

17.471(5)

90

92.519(10)

90

3039.7(17)

4

Triclinic

ˉ

P1

9.398(5)

12.868(7)

14.381(7)

108.860(7)

100.484(9)

105.905(8)

1511.2(13)

2

β/°

90

γ/°

V/Å³

3778.0(12)

4

3309.8(19)

4

Z

Total reflections

Independent reflections

R_int

R(obs)

wR(all data)

6057

5365

0.0470

0.1060

0.2903

5488

4124

0.0480

0.0562

0.1730

5639

5048

0.0570

0.0765

0.2115

6894

5803

0.0322

0.0468

5739

2735

0.6117

0.1657

0.4747

5281

3448

0.1206

0.1342

0.1505

0.2935

This journal is © The Royal Society of Chemistry 2013

Dalton Trans., 2013, 42, 1437–1450 | 1445

View Article Online

Paper

Dalton Transactions

over 30 minutes and the solution was then stirred for a further

30 minutes at −78 °C. Phosphoryl chloride (5.0 ml,

Conclusion

Compounds with two supermesityl groups placed geminally at 51.9 mmol) was added over 20 minutes. The resulting solution

the central phosphorus atom show extremely low reactivity at was maintained at −78 °C for 2 hours and was then allowed to

the phosphorus centre. Thus no reaction is generally observed rise to ambient temperature overnight. The solvent and excess

under conditions used for reactions of their unhindered phosphoryl chloride were removed in vacuo. The yellow solid

counterparts. Use of harsher conditions may promote the residue was taken up in dichloromethane and washed with 3 ×

desired reactivity, however can also result in C–C bond clea- 100 ml of water. The organic layer was separated and dried

vage (formation of 3), or the formation of complex mixtures. over anhydrous magnesium sulfate. The volatiles were

Rather common reactivity is also insertion into a proximate C– removed in vacuo and the resultant yellow solid was chromato-

H bond (formation of 6).

graphed on silica gel. Elution with hexane gave the by-product

A notable exception to the observed lousy reactivity (of 1,3,5-tri-t-butylbenzene as the first band. The eluent was then

species described in this paper) is the deprotonation of the changed to 97 : 3 hexane : ethyl acetate which furnished 1 as a

secondary phosphine oxide 2, which proceeds smoothly with white solid (yield range 7.26–8.74 g, 54–65%); mp 206–210 °C

n-BuLi. The resulting phosphinate anion is a useful synthon (found: C, 75.24; H, 10.15. Calc. for C₃₆H₅₈ClOP: C, 75.42; H,

for the formation of extremely bulky phosphine oxides 4 and 10.20); IR (KBr disk) ν_max/cm⁻¹2962vs (ν_CH), 2908s (ν_CH),

5. By the way of contrast, we have not been able to deprotonate 1583m, 1236s (ν_PvO), 614s, 205s; Raman (powder in glass

the secondary phosphine 3. Attempts to prepare bulky chloro- capillary) ν_max/cm⁻¹2967vs (ν_CH), 2911vs (ν_CH), 1584m, 1238m

4

phosphine Mes*(2,4-tBu₂C₆H₃)PCl by chlorination of 3 failed, (ν_PvO), 821s, 267s; δ_H(300.1 MHz; CDCl₃) 7.47 (1H, dd, J_PH

=

4

however the reaction of

amenable.

3 with sulfur was surprisingly 4.8, J_HH= 1.8, ArH), 7.36 (1H, dd, J_PH= 4.6, J_HH= 1.9, ArH),

4

7.24 (1H, dd, J_PH= 5.8, J_HH= 2.0, ArH), 7.15 (1H, dd, J_PH

=

4

Computations show that strain energies of synthetically 6.0, J_HH= 2.1, ArH), 1.55 (9H, s, o-tBu), 1.50 (9H, s, o-tBu),

accessible geminally substituted compounds are extremely 1.23 (18H, s, 2 × p-tBu), 0.69 (9H, s, o-tBu), 0.66 (9H, s, o-tBu);

2

high and their P–C_Arbond dissociation energies are signifi- δ_C(75.5 MHz, CDCl₃) 161.7 (d, J_PC= 9.9, o-C), 159.3 (d, J_PC

cantly lower than those of their unstrained relatives. Surpris- 7.1, o-C), 155.1 (d, J_PC= 12.7, o-C), 153.3 (d, J_PC= 11.9, o-C),

=

2

4

ingly a similar amount of strain (evaluated through ΔE₁, see 151.1 (d, J_PC= 4.5, p-C), 150.9 (d, J_PC= 5.0, p-C), 131.5 (d,

Table 4) was found in the secondary phosphine Mes*₂PH and ¹J_PC= 126.0, i-C), 127.7 (d, J_PC= 134.5, i-C), 126.1 (d, J_PC

=

1

3

its oxide Mes*₂P(vO)H (both ca. 180 kJ mol⁻¹), whilst the ter- 15.7, m-C), 125.6 (d, J_PC= 15.7, m-C), 124.2 (d, J_PC= 16.1, m-

3

tiary phosphine oxide Mes*₂P(vO)Me is strained significantly C), 123.3 (d, J_PC= 17.1, m-C), 42.1 (s, o-C(CH₃)₃), 42.0 (s,

more (ca. 250 kJ mol⁻¹).

o-C(CH₃)₃), 41.7 (s, o-C(CH₃)₃), 41.6 (s, o-C(CH₃)₃), 41.1 (d,

⁵J_PC= 3.1, p-C(CH₃)₃), 41.0 (d, J_PC= 4.0, p-C(CH₃)₃), 33.4 (s,

5

o-C(CH₃)₃), 33.0 (s, o-C(CH₃)₃), 32.5 (s, o-C(CH₃)₃), 31.7 (s,

o-C(CH₃)₃), 29.9 (broad s, 2 × p-C(CH₃)₃); δ_P(121.5 MHz,

CDCl₃) 45.3 (s); MS (ES+) m/z 573.36 (M + H⁺).

Experimental section

General notes

Mes*2P(O)H 2. A suspension of lithium aluminium hydride

Unless stated otherwise, experiments were carried out in stan- (2.00 g, 52.8 mmol) in THF (100 ml) was cooled in ice. To this

dard Schlenk glassware or in a glove box with strict exclusion powdered 1 (3.56 g, 6.22 mmol) was added in small batches

of air and moisture, using a nitrogen or argon atmosphere. Sol- with stirring. The suspension was heated under reflux;

vents were dried on an MBraun solvent purification system. samples of the reaction mixture were taken periodically and

Column chromatography was performed in air with non- investigated by ³¹P NMR. The reaction was deemed complete

degassed solvents. The 2-bromo-1,3,5-tri-t-butylbenzene when the starting material vanished from the spectrum to be

(Mes*Br) was made according to a literature procedure in two replaced by a transient (presumably aluminium) complex of

steps from benzene.^4,5Unless stated otherwise, the new com- compound 2 (s, δ_P102.4). The time needed to drive the reac-

pounds were fully characterized by ³¹P{¹H}, ¹H and ¹³C{¹H} tion to completion varied from 6 to 9 hours. The mixture was

NMR, including measurement of ¹H{³¹P}, ³¹P (¹H coupled), cooled in ice and then water (5 mL) was added very slowly to

¹H–¹H DQF COSY, ¹H–³¹P HSQC and ¹H–¹³C HSQC exper- hydrolyse the remaining hydride with evolution of H₂gas. The

iments. Measurements were performed at 25 °C unless said resulting suspension was filtered using glass sinter (porosity

otherwise; 85% H₃PO₄was used as an external standard in ³¹P S3) and the solid collected on the filter was washed through

1

NMR. H and ¹³C NMR shifts are relative to TMS, which was with dichloromethane (3 × 30 ml). The volatiles were evapor-

used as an internal standard or residual solvent peaks were ated in vacuo. The yellow residue was then taken up with more

used for calibration (CHCl₃δ_H7.26, δ_C77.2 ppm). J values are dichloromethane and washed with water (3 × 100 ml). The

given in Hz. NMR numbering scheme shown in Scheme 5 was organic layer was separated and dried with anhydrous mag-

used for compounds 3 and 8.

nesium sulfate. The volatiles were then removed in vacuo. The

Mes*₂P(O)Cl 1. A solution of Mes*Br (15.27 g, 46.9 mmol) yellow solid obtained was chromatographed on silica gel.

in THF (200 ml) was cooled to −78 °C. A hexane solution of Elution with hexane gave the by-products 1,3,5-tri-t-butylben-

1

n-butyllithium (20 ml of 2.5 M solution, 50 mmol) was added zene and 2,4,6-tri-t-butylphenylphosphine [δ_P−129.9 (t, J_PH

=

1446 | Dalton Trans., 2013, 42, 1437–1450

View Article Online

Dalton Transactions

Paper

210.1 Hz)].³⁶The eluent was then altered to 97 : 3 hexane : ethyl uncontaminated fractions were combined and after evapor-

acetate. The by-product 2,4,6-tri-t-butylphenylphosphine oxide ation of volatiles in vacuo 3 was obtained as a white powder

1

[δ_P−11.3 (t, J_PH= 482.4 Hz)]³⁷emerged slightly ahead of the (yield range 1.67–2.19 g, 55–72%).

1

desired product 2 (δ_P12.9 (d, J_HP= 483.0 Hz)), so the first few

Method 2: A stirred suspension of lithium aluminium

fractions were tested by ³¹P NMR until the first uncontami- hydride (0.52 g, 13.7 mmol) and aluminium trichloride

nated fraction egresses. After combining the uncontaminated (2.01 g, 15.2 mmol) in 1,4-dioxane (40 ml) was cooled in ice.

fractions and evaporation of volatiles in vacuo 2 was obtained Powdered 1 (0.93 g, 1.62 mmol) was added in small batches.

as a white powder (yield range 1.27–1.87 g, 38–56%); mp The suspension was heated under reflux, the progress of the

179–181 °C (found: C, 79.61; H, 11.36. Calc. for C₃₆H₅₉OP: C, reaction was monitored by ³¹P NMR. When all the starting

80.25; H, 11.04); IR (KBr disk) ν_max/cm⁻¹2965vs (ν_CH), 2909vs material was consumed (typically 7–9 hours) the reaction

(ν_CH), 1586m, 1191s (ν_PvO), 982s, 614s; Raman (powder in mixture was cooled in ice and water (∼5 ml) was added slowly

glass capillary) ν_max/cm⁻¹2969vs (ν_CH), 2909vs (ν_CH), 1595m, to hydrolyse the remaining hydride with evolution of H₂gas.

1192s (ν_PvO), 822s, 567s; δ_H(300.1 MHz; CDCl₃, 298 K) 8.91 The same workup procedure including chromatography purifi-

1

(1H, d, J_PH= 482.8, P–H), 7.70–6.70 (4H, br s, 4 × ArH), cation was used as in Method 1 above. After combining the

1.80–1.40 (18H, br m, 2 × o-tBu), 1.21 (18H, s, 2 × p-tBu), uncontaminated fractions and evaporation of volatiles in vacuo

1.00–0.40 (18H, br s, 2 × o-tBu); δ_C(100.6 MHz, CDCl₃, 298 K) 3 was obtained as a white powder (yield range 0.19–0.23 g,

155.0–156.6 (br m, C_q), 150.4–152.2 (br m, C_q), 121.9–127.8 (br 26–30%); mp 158–162 °C (found: C, 82.13; H, 11.39. Calc. for

m, 4 × CH), 37.6–43.2 (br m, 4 × o-C(CH₃)₃), 34.5 (s, 2 × C₃₂H₅₁P: C, 82.35; H, 11.01); IR (KBr disk) ν_max/cm⁻¹2963vs

p-C(CH₃)₃), 32.1–34.0 (br m, 4 × o-C(CH₃)₃), 30.4–31.7 (br s, 2 × (ν_CH), 2867vs (ν_CH), 2427w (ν_PH), 1597m, 1473m, 1390m,

p-C(CH₃)₃); δ_P{₁_H}(121.5 MHz, CDCl₃) 12.9 (s); δ_P(121.5 MHz, 1393m, 1361m, 1237m, 879s; Raman (powder in glass capil-

CDCl₃) 12.9 (d, ¹J_HP= 483); δ_H(499.9 MHz; CD₂Cl₂; 213 K) 8.99 lary) ν_max/cm⁻¹2967vs (ν_CH), 2906vs (ν_CH), 2429w (ν_PH), 1597s,

1

(1H, d, J_PH= 483.3, P–H), 7.53 (1H, br s, ArH), 7.42 (2H, br s, 1460m, 1202s, 1048m, 925s, 823s, 685s; δ_H(300.1 MHz; CDCl₃)

4

2 × ArH), 7.13 (1H, br s, ArH), 1.72 (9H, s, o-tBu), 1.63 (9H, s, 7.47 (1H, br s, H3 or H5), 7.41 (1H, br d, J_PH= 2.6, H3 or H5),

4

3

o-tBu), 1.32 (18H, s, 2 × p-tBu), 0.86 (9H, s, o-tBu), 0.72 (9H, 7.28 (1H, dd, J_PH= 4.4, J_HH= 2.1, H21), 6.69 (1H, m, J_HH

=

4

1

s, o-tBu); δ_C(125.7 MHz, CD₂Cl₂; 213 K) 161.4 (s, o-C), 155.8 (d, 8.4, J_HH= 2.1, H23), 6.05 (1H, d, J_PH= 236.5, P–H), 5.77 (1H,

²J_PC= 3.4, o-C), 155.6 (d, J_PC= 11.1, o-C), 154.8 (d, J_PC= 6.1, dd, J_HH= 8.1, J_PH= 5.1, H24), 1.52 (9H, s, 25-Me₃), 1.40 (9H,

2

3

1

o-C), 151.8 (s, p-C), 150.8 (s, p-C), 129.9 (d, J_PC= 106.2, i-C), s, 7-Me₃or 15-Me₃), 1.37 (9H, s, 7-Me₃or 15-Me₃), 1.29 (9H, s,

128.1 (¹J_PC= 96.6, i-C), 126.6 (d, J_PC= 11.7, m-C), 124.6 (d, 29-Me₃), 1.14 (9H, s, 11-Me₃); δ_C(75.5 MHz, CDCl₃) 155.6 (d,

3

³J_PC= 11.3, m-C), 124.5 (d, J_PC= 13.3, m-C), 123.3 (d, J_PC

=

²J_PC= 21.6, C2 or C6), 154.5 (s, C2 or C6), 149.7 (d, J_PC= 17.4,

3

2

3

11.1, m-C), 40.8 (s, o-C(CH₃)₃), 40.1 (d, J_PC= 3.1, o-C(CH₃)₃), C20), 149.2 (s, C4 or C22), 148.3 (s, C4 or C22), 133.2 (s, C24),

1

39.9 (s, o-C(CH₃)₃), 39.3 (s, o-C(CH₃)₃), 34.8 (s, o-C(CH₃)₃), 34.5 132.9 (d, J_PC= 28.3, C19), 128.4 (d, J_PC= 29.9, C1), 121.6 (s,

(s, 2 × p-C(CH₃)₃), 33.1 (s, o-C(CH₃)₃), 32.9 (s, o-C(CH₃)₃), 31.2 C3 or C5), 121.5 (s, C3 or C5), 121.4 (s, C21), 121.2 (s, C23),

(s, o-C(CH₃)₃), 30.9 (s, 2 × p-C(CH₃)₃); MS (ES+) m/z 539 37.4 (s, C7 or C15), 37.2 (s, C7 or C15), 36.3 (s, C25), 34.1 (s,

(M + H⁺), 561 (M + Na⁺).

C11), 33.5 (s, C29), 32.5 (s, 25-Me₃), 32.3 (s, 7-Me₃or 15-Me₃),

(2,4-tBu₂C₆H₃)(Mes*)PH 3. Method 1:

A

suspension of 32.1 (s, 7-Me₃or 15-Me₃), 30.3 (s, 29-Me₃), 30.1 (s, 11-Me₃); δ_P

lithium aluminium hydride (2.50 g, 65.8 mmol) in 1,4-dioxane _{₁_H}(121.5 MHz, CDCl₃) −67.1 (s); δ_P(121.5 MHz, CDCl₃) −67.1

(100 ml) was cooled in ice. To this trimethylsilyl chloride (d, ¹J_HP= 237.3); MS (ES+) m/z 467 (M + H⁺), 505 (M + O + Na⁺),

(9.2 ml, 72.0 mmol) was added slowly, followed by powdered 1 988 [(M + O)₂+ Na⁺].

(3.73 g, 6.51 mmol). The flask was fitted with a reflux conden-

Mes*₂P(O)Me 4. To a solution of 2 (1.30 g, 2.41 mmol) in

ser and heated to 100 °C. Samples of the reaction mixture were THF (40 ml) cooled in ice, n-butyllithium (1.0 ml of 2.5 M solu-

taken periodically and investigated by ³¹P NMR, the heating tion in hexanes, 2.50 mmol) was added dropwise. The mixture

was continued until all the starting material was consumed, was stirred for 10 minutes, during which time the solution

typically 5 hours. The mixture was cooled in ice and then water turned red-brown. Methyl iodide (1.25 ml, 20 mmol) was

(5 ml) was added very slowly to hydrolyse the remaining added slowly at 0 °C, resulting in the dark solution giving way

hydride with evolution of H₂gas. The resulting suspension to a pale yellow one. The volatiles were removed in vacuo and

was filtered using a glass sinter (porosity S3) and the solid col- the creamy residue was taken up in dichloromethane and

lected on the filter was washed through with dichloromethane washed with water (3 × 10 ml). The organic layer was separated

(3 × 30 ml). The volatiles were evaporated in vacuo. The yellow and dried over anhydrous magnesium sulfate. The volatiles

residue was then taken up with fresh dichloromethane and were removed in vacuo and the resulting solid was recrystal-

washed with water (3 × 100 ml). The organic layer was separ- lized from acetonitrile to give 4 as colourless crystals (0.69 g,

ated and dried with anhydrous magnesium sulfate. The vola- 50%); mp 164–168 °C (found: C, 80.32; H, 11.01; calc. for

tiles were then removed in vacuo. The yellow solid obtained C₃₇H₆₁PO: C, 80.38; H, 11.12); IR (KBr disk) ν_max/cm⁻¹2961vs

was chromatographed on silica gel using hexane as an eluent. (ν_CH), 1587m, 1467m, 1188s (ν_PvO), 1023s, 876s; Raman

The by-product 2,4,6-tri-t-butylphenylphosphine [(δ_P−129.9 (t, (powder in glass capillary) ν_max/cm⁻¹2963vs (ν_CH), 2907vs

¹J_PH= 210.1 Hz)]³⁶emerged only slightly ahead of 3, so the (ν_CH), 1591s, 1465s, 1448s, 1203s (ν_PvO), 812s, 573s; δ_H

4

first few fractions were tested by ³¹P NMR. The (300.1 MHz; CDCl₃) 7.41 (1H, dd, J_PH= 3.3, J_HH= 2.3, Ar-H),

Dalton Trans., 2013, 42, 1437–1450 | 1447

View Article Online

Paper

7.31 (1H, dd, J_PH= 3.9, J_HH= 2.2, Ar-H), 7.14 (1H, dd, J_PH

Dalton Transactions

4

=

589.4 (M + Na⁺); MS (exact mass, ES+) 567.4700 (M + H⁺),

4.2, ⁴J_HH= 2.2, Ar-H), 7.13 (1H, dd, ⁴J_PH= 3.5, ⁴J_HH= 2.2, Ar-H), C₃₈H₆₄PO requires 567.4695.

2.60 (3H, d, ²J_PH= 11.4, P-CH₃), 1.56 (9H, s, o-tBu), 1.45 (9H, s,

Phosphindole 6. To a solution of 2 (0.14 g, 0.26 mmol) in

o-tBu), 1.24 (9H, s, p-tBu), 1.23 (9H, s, p-tBu), 0.67 (9H, s, THF (8 ml) cooled in ice n-butyllithium (0.15 ml of 2.5 M solu-

o-tBu), 0.59 (9H, s, o-tBu); δ_C(75.5 MHz, CDCl₃) 160.0 (d, ²J_PC

6.7, o-C), 157.1 (d, J_PC= 9.3, o-C), 156.1 (d, J_PC= 5.8, o-C), was stirred for 15 minutes, during which time a red-brown

151.7 (d, ²J_PC= 7.4, o-C), 149.0 (d, ⁴J_PC= 4.1, p-C), 148.6 (d, ⁴J_PCcolour developed. A solution of iodine (0.5 g, 1.97 mmol) in

= 3.6, p-C), 132.7 (d, ¹J_PC= 97.0, i-C), 130.1 (d, ¹J_PC= 114.3, i-C), THF (8 ml) was added dropwise. The solution was allowed to

=

tion in hexanes, 0.37 mmol) was added dropwise. The mixture

2

3

125.3 (d, J_PC= 11.5, m-C), 124.5 (d, J_PC= 13.8, m-C), 123.2 (d, rise to room temperature and then slowly heated to 50 °C. It

³J_PC= 11.7, m-C), 122.9 (d, J_PC= 13.1, m-C), 41.1 (d, J_PC= 3.3, was maintained at this temperature for an hour during which

o-C(CH₃)₃), 41.0 (d, J_PC= 3.8, o-C(CH₃)₃), 40.8 (d, J_PC= 2.8, the purple colour of the solution persisted. The volatiles were

3

o-C(CH₃)₃), 39.5 (d, J_PC= 2.0, o-C(CH₃)₃), 33.6 (s, o-C(CH₃)₃), removed in vacuo. The dark residue was dissolved in dichloro-

33.3 (s, 2 × p-C(CH₃)₃), 33.2 (s, o-C(CH₃)₃), 32.4 (s, o-C(CH₃)₃), methane (10 ml) and the resulting solution was shaken with a

1

32.4 (d, J_PC= 69.5, P-CH₃), 31.7 (s, o-C(CH₃)₃), 30.0 (s, 2 × p-C saturated aqueous solution of sodium thiosulfate (10 ml). The

(CH₃)₃); δ_P{₁_H}(121.5 MHz, CDCl₃) 34.1 (s); δ_P(121.5 MHz, pale yellow organic layer was washed with water (3 × 20 ml),

CDCl₃) 34.1 (q, ²J_HP= 11.4); MS (ES+) m/z 553.4 (M + H⁺), 575.4 and dried over anhydrous magnesium sulfate. The volatiles

(M + Na⁺); MS (exact mass, ES+): 553.4533 (M + H⁺), C₃₇H₆₂PO were then removed in vacuo to give 6 as a colourless solid

requires 553.4538.

(0.10 g, 71.8%); δ_P(109.4 MHz, CDCl₃) 54.9; MS (ES+) m/z

Mes*₂P(O)Et 5. To a solution of 2 (0.72 g, 1.30 mmol) in 559.4 (M + Na⁺).

THF (25 ml) cooled in ice, n-butyllithium (0.6 ml of 2.5 M sol-

(2,4-tBu₂C₆H₃)(Mes*)PH(vS) 8. Triethyl amine (3.0 ml,

ution in hexanes, 1.50 mmol) was added dropwise. The 21.5 mmol) was added to a stirred suspension of 3 (0.26 g,

mixture was stirred for 15 minutes, during which time the 0.56 mmol) and sulfur (0.03 g, 0.77 mmol) in toluene (20 ml)

solution turned red-brown. Ethyl iodide (1.0 ml, 12.5 mmol) at room temperature. The sulfur dissolved over 10 minutes to

was added slowly. The stirred solution was allowed to warm up leave a yellow solution, which was subsequently heated under

to room temperature and then slowly heated to 50 °C at which reflux for 90 minutes. During this time the solution turned

point the solution became pale yellow. The volatiles were dark red. The volatiles were then removed in vacuo. The result-

removed in vacuo, the resulting creamy solid was dissolved in ing dark red/brown solid was chromatographed on silica gel

dichloromethane and washed with water (3 × 10 ml). The with 97 : 3 hexane : ethyl acetate as an eluent. 8 was obtained

organic layer was separated and dried over anhydrous mag- as a pale yellow powder (0.125 g, 45%); mp 144–148 °C (found:

nesium sulfate. The volatiles were removed in vacuo to give C, 76.89; H, 10.17. Calc. for C₃₂H₅₁PS: C, 77.10; H, 10.31); IR

crude 5 as a creamy yellow solid (0.71 g, 96%). This solid was (KBr disk) ν_max/cm⁻¹2961vs (ν_CH), 2868vs (ν_CH), 1597s, 1475m,

recrystallized from acetonitrile to give analytically pure 5 as 1394s, 1362s, 1106m, 663s (ν_PvS), 609m; Raman (powder in

colourless crystals (0.05 g, 7%); mp 175–180 °C (found: C, glass capillary) ν_max/cm⁻¹2968vs (ν_CH), 2908vs (ν_CH), 1597s,

80.46; H, 11.14; calc. for C₃₈H₆₃PO: C, 80.51; H, 11.20); IR 1467m, 1449m, 1202s, 926s, 828s, 685s, 664s (ν_PvS), 564m; δ_H

1

(KBr disk) ν_max/cm⁻¹2963vs (ν_CH), 1597m, 1476m, 1390m, (300.1 MHz; CDCl₃) 8.95 (1H, d, J_PH= 485.5, P–H), 7.66 (1H,

1363m, 1184s (ν_PvO), 1026m, 604s; Raman (powder in dd, ³J_PH= 18.5, ³J_HH= 8.4, H24), 7.51 (1H, dd, ⁴J_PH= 6.1, ⁴J_HH

=

4

glass capillary) ν_max/cm⁻¹2963vs (ν_CH), 2907vs (ν_CH), 1598s, 1.9, H21), 7.49 (2H, d, J_PH= 4.4, H3 and H5), 7.01 (1H, m,

4

1581s, 1469s, 1446s, 1185m (ν_PvO), 821s, 573s, 267s; ³J_HH= 8.4, J_HH= 1.9, H23), 1.57 (9H, s, 25-Me₃), 1.42 (18H, s,

4

δ_H(300.1 MHz; CDCl₃) 7.49 (1H, dd, J_PH= 2.7, J_HH= 1.6, 7-Me₃and 15-Me₃), 1.26 (9H, s, 11-Me₃), 1.22 (9H, s, 29-Me₃);

4

Ar-H), 7.28 (1H, dd, J_PH= 3.6, J_HH= 1.8, Ar-H), 7.23 (1H, dd, δ_C(75.5 MHz, CDCl₃) 153.2 (d, ²J_PC= 3.0, C2 and C6), 153.1 (d,

⁴J_PH= 3.8, ⁴J_HH= 1.7, Ar-H), 7.08 (1H, dd, ⁴J_PH= 2.7, ⁴J_HH= 1.8, ²J_PC= 7.6, C20), 151.1 (s, C4 or C22), 151.0 (s, C4 or C22), 137.3

2

1

Ar-H), 2.65–2.95 (2H, m, P-CH₂CH₃), 1.58 (9H, s, o-tBu), (d, J_PC= 13.6, C24), 127.2 (d, J_PC= 66.0, C1), 126.5 (d, J_PC

=

3

1.43 (9H, s, o-tBu), δ1.40 (3H, t, J_PH= 7.1, PCH₂CH₃), 76.8, C19), 124.3 (m, C3, C5 and C21), 120.5 (d, J_PC= 13.4,

1.23 (18H, s, 2 × p-tBu), 0.69 (9H, s, o-tBu), 0.60 (9H, s, o-tBu); C23), 38.9 (s, C7 and C15), 37.4 (s, C25), 34.0 (s, C11), 33.9 (s,

C29), 33.6 (s, 7-Me₃and 15-Me₃), 32.2 (s, 25-Me₃), 31.0 (s, 29-

9.0, o-C), 157.8 (d, J_PC= 5.7, o-C), 153.7 (d, J_PC= 8.1, o-C), Me₃), 30.1 (s, 11-Me₃); δ_P{₁_H}(121.5 MHz, CDCl₃) 17.3 (s);

2

δ_C(75.5 MHz, CDCl₃) 161.4 (d, J_PC= 7.4, o-C), 158.1 (d, J_PC

=

2

4

1

3

150.2 (d, J_PC= 3.4, p-C), 150.0 (d, J_PC= 4.1, p-C), 133.2 (d, δ_P(121.5 MHz, CDCl₃) 17.3 (dd, J_HP= 486.0, J_HP= 17.6); MS

¹J_PC= 109.2, i-C), 132.5 (d, J_PC= 90.5, i-C), 127.2 (d, J_PC

=

(ES+) m/z 521.3 (M + Na⁺); MS (exact mass, ES+): 521.3337

1

3

11.2, m-C), 125.9 (d, J_PC= 9.4, m-C), 125.8 (d, J_PC= 7.3, m-C), (M + Na⁺), C₃₂H₅₁PSNa requires 521.3347.

3

124.4 (d, J_PC= 12.8, m-C), 42.6 (d, J_PC= 3.3, o-C(CH₃)₃),

Computational details. Geometries were fully optimized at

42.2 (d, J_PC= 2.4, o-C(CH₃)₃), 42.1 (d, J_PC= 3.3, o-C(CH₃)₃), the B3LYP/6-31G*⁽*⁾level³⁸of density functional theory (DFT),

3

41.1 (d, J_PC= 2.1, o-C(CH₃)₃), 35.0 (s, o-C(CH₃)₃), 34.8 (s, i.e. including polarisation functions on all non-hydrogen

o-C(CH₃)₃), 34.7 (s, 2 × p-C(CH₃)₃), 33.7 (s, o-C(CH₃)₃), 33.6 (s, atoms and the H atoms bonded to P, together with a fine inte-

1

o-C(CH₃)₃), 33.3 (d, J_PC= 63.7, PCH₂CH₃), 31.4 (s, p-C(CH₃)₃), gration grid (75 radial shells with 302 angular points per

2

31.5 (s, p-C(CH₃)₃), 13.0 (d. J_PC

= 4.2, PCH₂CH₃); δ_Pshell). For 2, the solid state structure was used as starting

(121.5 MHz, CDCl₃) 47.5 (br s); MS (ES+) m/z 567.4 (M + H⁺), points for the optimisation. The nature of the minima was

1448 | Dalton Trans., 2013, 42, 1437–1450

View Article Online

Dalton Transactions

Paper

verified by computations of the harmonic frequencies at the

same level of theory. Energies are reported at the B3LYP-D3/6-

31G*⁽*⁾+ZPE level, i.e. including the empirical dispersion cor-

rection by Grimme³⁹and zero-point energies. Selected species

were reoptimised at the M06-2X level.⁴⁰All computations were

performed using the Gaussian09 suite of programs.⁴¹

1408; Y. Wang and G. H. Robinson, Chem. Commun., 2009,

5201–5213.

4 S. R. Ditto, R. J. Card, P. D. Davis and D. C. Neckers, J. Org.

Chem., 1979, 44, 894–896.

5 R. Knorr, C. Pires, C. Behringer, T. Menke, J. Freudenreich,

E. C. Rossmann and P. Böhrer, J. Am. Chem. Soc., 2006, 128,

14845–14853; A. H. Cowley, N. C. Norman and M. Pakulski,

Inorg. Synth., 1990, 27, 235–237.

X-ray crystallography

6 M. Yoshifuji, I. Shima and N. Inamoto, Tetrahedron Lett.,

1979, 20, 3963–3964.

7 M. Yoshifuji, I. Shima, N. Inamoto, K. Hirotsu and

T. Higuchi, Angew. Chem., Int. Ed. Engl., 1980, 19,

399–400.

8 B. Cetinkaya, P. B. Hitchcock, M. F. Lappert, A. J. Thorne

and H. Goldwhite, Chem. Commun., 1982, 12, 691–693.

9 H. H. Karsch, F. H. Kohler and H.-U. Reisacher, Tetrahedron

Lett., 1984, 25, 3687–3690.

10 M. Nieger, E. Niecke and U. Fischer, CCDC deposition No

720427, 2009, deposited as private communication.

11 A. Baceiredo, G. Bertrand, P. Mazerolles and J.-P. Majoral,

Chem. Commun., 1981, 22, 1197–1198.

12 B. Cetinkaya, A. Hudson, M. F. Lappert and H. Goldwhite,

Chem. Commun., 1982, 11, 609–610.

13 M. M. Wienk, R. A. J. Janssen and E. W. Meijer, J. Phys.

Chem., 1995, 99, 9331–9336.

14 (a) N. J. Winter, J. Fossey, B. Beccard, Y. Berchadsky, F. Vila,

L. Werbelow and P. Tordo, J. Phys. Chem., 1986, 90, 6749–

6750; (b) Y. Ayant, A. Thevand, L. Werbelow and P. Tordo,

J. Magn. Reson., 1987, 72, 251–256.

15 B. F. M. de Waal, O. M. Aagaard and R. A. J. Janssen, J. Am.

Chem. Soc., 1991, 113, 9471–9479.

X-ray quality crystals of 2, 3, 4, 5, 6 and 8 were obtained by

slow evaporation of solutions in dichloromethane. Table 5 lists

details of data collections and refinements. Data for com-

pounds 2, 3, 4, 5 and 6 were collected at 93(2) K using graphite

monochromated MoKα radiation (λ = 0.71073 Å) from a high

brilliance Rigaku MM007 generator and a Rigaku Mercury ccd

detector with ω and φ scans. X-ray crystal data for compound 8

were collected using the St Andrews Robotic Diﬀractometer

(Saturn724 CCD) at 125 K with graphite monochromated

MoKα radiation (λ = 0.71073 Å). Intensity data were collected

using ω steps accumulating area detector images spanning at

least a hemisphere of reciprocal space. Intensities were cor-

rected for Lorentz-polarisation and for absorption. The struc-

tures were solved by direct methods. The positions of the

hydrogen atoms were idealized except for structures 2, 3 and 8,

which were refined isotropically subject to distance constraint.

The Flack parameter for compound 4 was 0.57(17). Refine-

ments were done by full-matrix least squares based on F²

using the program SHELXTL.⁴²

CCDC 898340–898344 and 898346 contain the supplemen-

tary crystallographic data for this paper.

16 C. W. Bird, Tetrahedron, 1992, 48, 1675–1682.

17 HF/STO-6G level: M. Yoshifuji, I. Shima, N. Inamoto and

T. Aoyama, Tetrahedron Lett., 1981, 22, 3057–3060.

18 S. Sasaki, K. Sutoh, F. Murakami and M. Yoshifuji, J. Am.

Chem. Soc., 2002, 124, 14830–14831.

19 S. Sasaki, R. Chowdhury and M. Yoshifuji, Tetrahedron

Lett., 2004, 45, 9193–9196.

20 H. Mahdavi and J. Amani, Tetrahedron Lett., 2009, 50,

5923–5926.

Acknowledgements

We thank EPSRC, EaStCHEM and European Phosphorus

Sciences Network (PhoSciNet, COST action CM0802) for finan-

cial support. Calculations were performed using the EaSt-

CHEM Research Computing Facility maintained by H. Fruchtl.

We thank Mrs M. Smith for measurement of low temperature

NMR spectra and Mrs C. Horsburgh for MS measurements.

21 T. H. Woste and M. Oestreich, Chem.–Eur. J., 2011, 17,

11914–11918.

22 C. A. Busacca, J. C. Lorenz, N. Grinberg, N. Haddad,

M. Hrapchak, B. Latli, H. Lee, P. Sabila, A. Saha,

M. Sarvestani, S. Shen, R. Varsolona, X. Wei and

C. H. Senanayake, Org. Lett., 2005, 7, 4277–4280.

References

1 M. Yoshifuji, I. Shima and N. Inamoto, J. Am. Chem. Soc.,

1981, 103, 4587–4589.

2 For recent reviews see: M. Yoshifuji, Pure Appl. Chem., 2005, 23 (a) C. A. Busacca, R. Raju, N. Grinberg, N. Haddad,

77, 2011–2020; P. P. Power, Chem. Rev., 1999, 99,

3463–3503; P. Power, Chem. Rev., 2003, 103, 789–809.

3 P. P. Power, J. Organomet. Chem., 2004, 689, 3904–3919;

R. C. Smith, E. Urnezius, K. Lam, A. L. Rheingold and

J. D. Protasiewicz, Inorg. Chem., 2002, 41, 5296–5299;

P. James-Jones, H. Lee, J. C. Lorenz, A. Saha and

C. H. Senanayake, J. Org. Chem., 2008, 73, 1524–1531;

(b) P. N. Kapoor and L. M. Venanzi, Helv. Chim. Acta, 1977,

60, 2824–2829; (c) L. Horner, H. Hoﬀmann and P. Beck,

Chem. Ber., 1958, 91, 1583–1588.

T. Sasamori, E. Mieda, N. Nagahora, K. Sato, D. Shiomi, 24 T. Imamoto, T. Kusumoto, N. Suzuki and K. Sato, J. Am.

T. Takui, Y. Hosoi, Y. Furukawa, N. Takagi, S. Nagase and

N. Tokitoh, J. Am. Chem. Soc., 2006, 128, 12582–12588;

T. Sasamori and N. Tokitoh, Dalton Trans., 2008, 1395–

Chem. Soc., 1985, 107, 5301–5303; T. Imamoto, T. Oshiki,

T. Onozawa, T. Kusumoto and K. Sato, J. Am. Chem. Soc.,

1990, 112, 5244–5252.

Dalton Trans., 2013, 42, 1437–1450 | 1449

View Article Online

Paper

Dalton Transactions

25 L. D. Quin,

A

Guide to Organophosphorus Chemistry, 37 M. Yoshifuji, K. Shibayama, K. Toyota and N. Inamoto,

J. Wiley, New York, 1st edn, 2000, p. 98. Tetrahedron Lett., 1983, 24, 4227–4228.

26 (a) M. Yoshifuji, I. Shima, K. Ando and N. Inamoto, Tetra- 38 (a) A. D. Becke, J. Chem. Phys., 1993, 98, 5648–5642;

hedron Lett., 1983, 24, 933–936; (b) A. H. Cowley and

M. Pakulski, Tetrahedron Lett., 1984, 25, 2125–2126.

(b) C. Lee, W. Yang and R. G. Parr, Phys. Rev. B: Condens.

Matter, 1988, 37, 785–789.

27 D. J. Brauer, F. Bitterer, F. Dörrenbach, G. Hessler, 39 S. Grimme, J. Antony, S. Ehrlich and H. Krieg, J. Chem.

O. Stelzer, C. Krüger and F. Lutz, Z. Naturforsch., 1996, 51b,

1183–1196.

28 D. K. Wicht, I. Kovacik, D. S. Glueck, L. M. Liable-Sands,

Phys., 2010, 132, 154104.

40 Y. Zhao, N. E. Schultz and D. G. Truhlar, J. Chem. Theory

Comput., 2006, 2, 364–382.

C. D. Incarvito and A. L. Rheingold, Organometallics, 1999, 41 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,

18, 5141–5151.

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, N. J. 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, Ö. Farkas,

J. B. Foresman, J. V. Ortiz, J. Cioslowski and D. J. Fox,

GAUSSIAN 09 (Rev. A.02), Wallingford CT, 2009.

29 C. M. D. Komen and F. Bickelhaupt, Synth. Commun., 1996,

26, 1693–1697.

30 M. Stankiewicz, J. Nycz and J. Rachon, Heteroat. Chem.,

2002, 13, 330–339.

31 F. Dornhaus, M. Bolte, H. Lerner and M. Wagner,

Eur. J. Inorg. Chem., 2006, 24, 5138–5147.

32 G. Y. Li and W. J. Marshall, Organometallics, 2002, 21, 590–591.

33 K. Wiberg, Angew. Chem., Int. Ed. Engl., 1986, 25, 312–322.

34 Homodesmotic equations are a subclass of isodesmic

equations, where not only the number of CC and CH

bonds are preserved on each side, but also the hybridiz-

ation of the C atoms involved, see e.g.: S. E. Wheeler,

K. N. Houk, P. v. R. Schleyer and W. D. Allen, J. Am. Chem.

Soc., 2009, 131, 2547–2560.

35 K. Hemelsoet, F. Van Durme, V. Van Speybroek,

M.-F. Reyniers and M. Waroquier, J. Phys. Chem. A, 2010,

114, 2864–2873.

36 M. Yoshifuji, K. Toyota, K. Shibayama and N. Inamoto, 42 G. M. Sheldrick, Acta Crystallogr., Sect. A: Fundam. Crystal-

Chem. Lett., 1983, 10, 1653–1656.

logr., 2008, 64, 112–122.

1450 | Dalton Trans., 2013, 42, 1437–1450

Article Doi

Synthetic and computational study of geminally bis(supermesityl) substituted phosphorus compounds

DOI: 10.1039/c2dt31971c

Source and publish data:

Authors:

Article abstract of DOI:10.1039/c2dt31971c

Full text of DOI:10.1039/c2dt31971c

Products guided by the article

R&D Labs maybe for 1417613-53-6

Relevant to this article

Hot Product