Preparation of the chiral amine salts [(η5-C5H5)Ru(PPh3) (CNtBu)(NH2R)]PF6, RNH2 = α-methylbenzylamine, 1-cyclohexylethylamine, and 1-(1-naphthyl)ethylamine: X-ray single crystal structures of the diastereomeric salts (SRuRC)-

Polyhedron 25 (2006) 2303–2317

www.elsevier.com/locate/poly

Preparation of the chiral amine salts

[(g⁵-C₅H₅)Ru(PPh₃)(CN^tBu)(NH₂R)]PF₆, RNH₂=

a-methylbenzylamine, 1-cyclohexylethylamine, and

1-(1-naphthyl)ethylamine: X-ray single crystal structures of

the diastereomeric salts (S_RuR_C)- and (R_RuR_C)-

[(g⁵-C₅H₅)Ru(PPh₃)(CN^tBu){NH₂CH(Me)(Ph)}]PF₆

*

Dawn M. Wong, Stephen J. Simpson

Department of Chemistry and Applied Chemistry, Cockcroft Building, University of Salford, Salford M5 4WT, United Kingdom

Received 6 January 2006; accepted 24 January 2006

Available online 3 March 2006

Abstract

Diastereomeric salts [(g⁵-C₅H₅)Ru(PPh₃)(CN^tBu)(NH₂R)]PF₆containing both a chiral metal centre and an a-chiral amine ligand

1

have been prepared. The use of diamagnetic anisotropy in their H NMR spectra to distinguish the diastereomers has been supported

by single crystal X-ray crystallography. The structures of the salts containing the ligands cyclohexylamine and a-methylbenzylamine

(both diastereomers) have been determined at low temperature. Decomplexation of the amine ligand from puriﬁed diastereomeric salts

to aﬀect resolution of the metal centred chirality has been attempted. The chiral isonitrile ligand containing compounds

[(g⁵-C₅H₅)Ru(PPh₃)(CNCH(Me)Ph)(Cl)] and [(g⁵-C₅H₅)Ru(PPh₃)(CNCH(Me)Ph)(NH₃)]PF₆have been prepared from (S)-a-methyl-

benzylisonitrile and the single crystal X-ray structure of both diastereomeric salts of the ammine compound has been determined.

Keywords: Metal; Chirality; Diastereomers; Ruthenium; Amines; Isonitriles; X-ray

1. Introduction

the ﬁrst optically active organotransition metal compounds

containing a chiral metal centre were only reported in 1969

[2,3]. Brunner [4,5] has developed and reviewed this ﬁeld

and cautioned against common pitfalls of data interpreta-

tion when resolution of metal centres and particularly their

conﬁgurational stability is reported [6–8].

We were interested in using readily available homochiral

amines to eﬀect resolution of a racemic mixture of ruthe-

nium compounds of the general type [(g⁵-C₅H₅)Ru-

(PPh₃)(L)(Cl)] by diastereomer formation followed by

removal of the chiral auxiliary to eﬀect resolution of the

metal centre. Low-spin d⁶Ru(II) was anticipated to be

reasonably conﬁgurationally stable in both ﬁve- and six-

coordination but to possibly present some problems in

the decomplexation step related to its kinetic inertness.

The exploration and utilization of metal centred chiral-

ity in organometallic chemistry is less developed than that

of carbon centres in modern organic chemistry. A major

reason for this diﬀerence is the kinetic lability of most

metal centres especially when monodentate ligands are

present; thus studies of octahedrally coordinated Cr(III)

and low-spin Co(III) complexes with ligand and metal cen-

tered chirality dominate the literature. Werner [1] reported

the ﬁrst resolution of a coordination compound in 1911 but

*

Corresponding author. Tel./fax: +44 1392 263427.

E-mail address: s.j.simpson@exeter.ac.uk (S.J. Simpson).

doi:10.1016/j.poly.2006.01.016

2304

D.M. Wong, S.J. Simpson / Polyhedron 25 (2006) 2303–2317

The ammine ligand in [(g⁵-C₅H₅)Ru(PPh₃)(CN^tBu)-

(NH₃)]PF₆had proved to be resistant to replacement by

carbonylation and to deprotonation [9].

gave the salts [(g⁵-C₅H₅)Ru(PPh₃)(CN^tBu)(NH₂CH{Me}-

Ph)]PF₆(3), [(g⁵-C₅H₅)Ru(PPh₃)(CN^tBu)(NH₂CH{Me}-

Cy)]PF₆(4), and [(g⁵-C₅H₅)Ru(PPh₃)(CN^tBu)(NH₂CH-

{Me}Np)]PF₆(5), respectively. Initial crystallization of 3

from dichloromethane–diethyl ether (1:3) at À78 ꢁC gave

ﬁne yellow crystals, which proved to be predominantly

the R_RuS_Cdiasteromer (3RS) in 74% diastereomeric excess.

Concentration of the ﬁltrate and crystallization at À25 ꢁC

gave a second set of yellow crystals, which proved to be

predominantly the S_RuS_Cdiastereomer (3SS) in 98% d.e.

The infrared spectrum of 3 contains four bands due to

m(NH) at 3325, 3308, 3284, and 3265 cm^À1conﬁrming the

presence of two diastereomers with the bands at 3325

and 3284 cm^À1being due to the more soluble diastereomer

3SS. While the deﬁnitive assignment of absolute structure

for the two diastereomeric salts was made by single crystal

X-ray crystallography our initial assignment was made by

There are very comprehensive nomenclature systems for

the unique identiﬁcation of isomeric transition metal com-

pounds but they have proved very cumbersome for routine

application [10,11]. We have opted for the simple approach

related to that used by organic practitioners. The stereo-

chemical descriptors used for organometallic compounds

are a modiﬁcation of the Cahn–Ingold–Prelog rules and

treat the cyclopentadienyl group as a single atom of atomic

weight 60 for prioritization [12,13]. The structural dia-

grams in this paper are drawn to normally show the metal

chirality denoted as S_Rufor convenience (Cp > PPh₃>

NH₂R > CN^tBu) but unless explicitly stated the starting

materials and products represented are racemic. One of

the spectroscopic probes used throughout this work is that

of diamagnetic anisotropy; protons held over an aromatic

1

consideration of the H NMR spectra of the enriched dia-

1

ring system will normally be shielded (d decreases) in H

stereomers. The spectrum of the least soluble compound

3RS has a cyclopentadienyl singlet resonance at d 4.30

and a doublet at d 1.11 for the a-methyl group whereas

the more soluble compound 3SS has these resonances at

d 4.72 and d 1.29, respectively. The diastereomers 3RR

and 3SR were also prepared, from R(+)-a-methylbenzyl-

amine to ensure complementarity and reciprocity and this

pair were used for the X-ray single crystallographic studies

(Scheme 1).

NMR spectra while protons held over a triple bond will

be deshielded (d increases). The eﬀect is usually observable

˚

at distances up to 4 A and drops oﬀ rapidly although mea-

˚

surable eﬀects up to 14 A have been observed in polyaro-

matic systems [14,15].

2. Results and discussion

Treatment of [(g⁵-C₅H₅)Ru(PPh₃)(CN^tBu)(Cl)] (1) with

a halide abstractor in the presence of cyclohexylamine

produced yellow crystalline [(g⁵-C₅H₅)Ru(PPh₃)(CN^tBu)-

(NH₂C₆H₁₁)]PF₆(2). The halide abstraction system could

be either thallium hexaﬂuorophosphate in dichlorometh-

ane or the methanolic potassium hexaﬂuorophosphate–

thallium (I) carbonate mixture used previously to prepare

ammine complexes [9]. The ﬁrst system gave a 97% yield

compared with 38% for the latter indicating some degree

of competition between methanol and the amine for the

solvento cationic intermediate in that synthetic route. The

main features of spectroscopic interest are the diﬀerentia-

tion of the diastereotopic NH protons by 1.21 ppm in the

¹H NMR spectrum and weak bands at 3311, 3270, and

1587 cm^À1in the infrared spectrum due to the amine

ligand. Encouraged by the degree of chemical shift diﬀer-

ence which suggested that one of the amine protons is

oriented into the region near the phenyl rings of the triphe-

nylphosphine ligand and hence shielded, implying strongly

preferred conformations and a useful spectroscopic probe,

we decided to prepare a family of cations containing C-

chiral amines in order to investigate diastereomeric resolu-

tion. The chiral amines used were of at least 98% e.e. and

the diastereomeric excesses reported below have not been

recalculated; given the integration errors for routine ¹H

NMR spectroscopy the diastereomeric excesses reported

probably underestimate the true selectivities.

Davies has demonstrated the use of a-methyl group

shielding by the aromatic ring current of triphenylphos-

phine to distinguish stereoisomers of the metal acyls

[(g⁵-C₅H₅)Fe(PPh₃)(CO)(COCH(Me)R)]; in S_FeS_C/R_FeR_C

diastereomers the group resonates between d 0.00 and d

0.50 whereas in the S_FeR_C/R_FeS_Cdiastereomers the reso-

nance is found between d 0.8 and d 1.30 [16]. Fig. 1 illus-

trates the relative relationships between the ruthenium

and iron compounds; note that the priority rules reverse

the metal centre descriptors on changing from an amine

group to an acyl group.

The single crystal X-ray structures of 3SR and 3RR

(vide infra) conﬁrm the basic conformations shown in

Fig. 2 and support the additional assumption that the phe-

nyl ring of the amine ligand shields the cyclopentadienyl

protons in 3SR and 3RS. Our criterion is that the diaste-

1

reomer with the highest H NMR chemical shifts for both

the cyclopentadienyl and a-methyl groups is the RR or SS

diastereomer while that with the lowest values for both

chemical shifts is the RS or SR isomer in the isonitrile ser-

ies. Davies has published extensively on the conformational

preferences of the alkyl groups in the iron acyl complexes

[17]; we have not performed similar modeling studies but

the close spectroscopic similarities suggest that qualita-

tively the structures indicated are correct, in particular that

the methine hydrogen points in towards the isonitrile or

carbonyl group.

Reaction of 1 with thallium hexaﬂuorophosphate in

dichloromethane and S(À)-a-methylbenzylamine, R(À)-

1-cyclohexylethylamine, or ( )-1-(1-naphthyl)ethylamine

While the described preparation of 3 yields an overall

racemic product in that equal amounts of 3RS and 3SS

are produced from 2 and S(À)-a-methylbenzylamine

D.M. Wong, S.J. Simpson / Polyhedron 25 (2006) 2303–2317

2305

t

CN Bu

(i) or (ii)

Ru

N

Cl

H

PPh

3

(2)

(1)

(ii)

t

Ph

CN Bu

(iii)

Ru

H

CH

N

3

H

PPh

3

(3)

o

(i) KPF , Tl CO , MeOH, amine, 60 C (ii) TlPF , amine, CH Cl , RT (iii) amine, CH Cl

6

2

3

6

2

Scheme 1. Synthetic routes to 2 and 3.

molecular hydrogen cation [(g⁵-C₅H₅)Ru(PPh₃)(CN^tBu)-

(g²-H₂)]PF₆[18] with S(À)-a-methylbenzylamine in deute-

rodichloromethane by ¹H NMR spectroscopy at room

temperature. Addition of 0.3 equivalents of the amine

immediately generated only 3SS and the further addition

of 0.7 equivalents of amine subsequently generated 3RS

within 10 min; the ﬁnal ratio of diastereomers was 50:50

and remained unchanged after 1 h at room temperature.

Under the ambient conditions the Lewis acid fragment

[(g⁵-C₅H₅)Ru(PPh₃)(CN^tBu)]⁺formed initially does not

racemise and despite a kinetic preference for 3SS the lack

of racemization prevents diastereomeric induction. Under

these conditions there is no interconversion of 3SR and

3SS as evidenced both by this experiment and from moni-

toring enriched samples over a period of days. Further, the

single crystals used for the X-ray structural determinations

of 3RR and 3SR, respectively were each dissolved in

t

CN^tBu

N

CN Bu

Ph

Me

Ru

H

CH₃

H

Ph

N

H

Ph₂P

3SR

3SS

R

Me

CO

C

CO

C

Fe

H

CH₃

Fe

H

R

O

Ph₂P

RR

RS

Fig. 1. Relationship of chirality descriptors to structure.

1

CD₂Cl₂at 22 ꢁC and H NMR spectra of the solutions

were immediately obtained; in each case the appropriate

single diastereomer was the only observed species and the

other diastereomers, 3SR and 3RR, respectively, were not

observed after several hours.

CH₃

H

CN^tBu

N

H

Ru

H

CH₃

H

There has been a considerable history of investigating

the ease of racemization of chiral at metal compounds

[4–8] with the general conclusion that such centres are often

not stereochemically rigid; we conclude that in this case the

rate of racemization is much less than the rate of complex-

ation and that none of the ligands attached to ruthenium in

the ﬁnal products reversibly dissociate (Scheme 2).

The salt [(g⁵-C₅H₅)Ru(PPh₃)(CN^tBu)(NH₂CH{Me}-

Cy)]PF₆(4) was also enriched on initial crystallization to

produce 4SR as the least soluble diastereomeric salt in

61% d.e. and the more soluble isomer 4RR in 79% d.e.

H

CN^tBu

H

Ph₂P

S_RuR_C

R_RuR_C

Fig. 2. Idealized solution conformation of 3SR and 3RR.

thermodynamically, this was not necessarily anticipated. A

kinetic experiment was therefore carried out by monitoring

the formation of 3SS and 3RS from the reaction of the

2306

D.M. Wong, S.J. Simpson / Polyhedron 25 (2006) 2303–2317

H

CN^tBu

H

CN^tBu

Ru

H

PPh₃

k₁

PPh₃

k₂

k₁

CN^tBu

k₃

k₄

Ru

CN^tBu

PPh₃

k₅

k₆

NH₂R^*

CN^tBu

NH₂R^*

Ru

PPh₃

k₅> k₆>> k₃,k₄

Scheme 2. Curtin–Hammett scheme for the interconversion of 3.

Similarly a single crystallization of [(g⁵-C₅H₅)Ru(PPh₃)-

(CN^tBu)(NH₂CH{Me}Np)]PF₆(5) gave the RS/SR prod-

uct in 73% d.e. and the more soluble RR/SS product

in 56% d.e. A further recrystallization of the RS/SR prod-

uct increased the diastereomeric excess up to 91%; both

products in this case being racemic because racemic ( )-

1-(1-naphthyl)ethylamine was used. In contrast to the large

were again partially separable with the R_RuS_Cproduct

(7RS) being the least soluble. The assignment of R_Ruste-

reochemistry in this diastereomeric salt was made by com-

1

paring the cyclopentadienyl resonances in the H NMR

spectra at d 5.13 and 4.57 with the lower chemical shift

being assigned to 7RS and the higher to 7SS; this reﬂects

the pattern found for the diastereomers of 3 and 5 induced

by the orientation of the aromatic substituent on the amine

ligand. The small quantity of 7SS obtained after the initial

crystallization which gave a sample of 7RS in 42% d.e.

combined with the low chemical yield, prevented the full

1

diﬀerence in H NMR chemical shift, (0.43 ppm), for the

cyclopentadienyl resonances of 3RS and 3SS these reso-

nances diﬀer by 0.001 ppm for the diastereomers of 4

suggesting that the aromatic ring of the amine in 3 can

shield these protons. In the case of 5 which also contains

an aromatic group in the amine ligand the separation is

0.54 ppm for the diastereomers.

Further diastereomeric enrichment of 3 and 5 was car-

ried out to establish the principle and usually a maximum

of three crystallizations was suﬃcient to give products of

ca. 98.5% d.e. for 3 and 5. Enrichment of 4 was not

explored fully because the spectroscopic parameters of 4

made accurate quantitation problematic beyond the 90%

d.e. level.

1

assignment of the H NMR spectrum and means that we

have used a single parameter in this case.

Decomplexation of the bound amine from diastereo-

merically enriched samples of 3, 4, or 7 should produce

enantiomerically enriched samples of the metal moieties.

A variety of procedures was explored using [(g⁵-C₅H₅)-

Ru(PPh₃)(CN^tBu)(NH₂C₆H₁₁)]PF₆(2) as a model com-

pound. Reaction of 2 with carbon monoxide (5 atm.,

294 K, 120 h.) in dichloromethane, with iodomethane or

tetraethylammonium ﬂuoride or tetraethylammonium

cyanide in dichloromethane, with methanolic sodium

methoxide, with sodium iodide or sodium azide in ace-

tone, with lithium triethylborohydride in THF (all at

294 K, 1–48 h), or with 2-methyl-1-butanethiol in reﬂux-

ing methanol gave in each case high recovery (>80%) of

The reaction of [(g⁵-C₅H₅)Ru(PPh₃)(CO)Cl] (6) with

thallium hexaﬂuorophosphate andS(À)-a-methylbenzyl-

amine in dichloromethane produced a very low yield

(3%) of yellow crystals of [(g⁵-C₅H₅)Ru(PPh₃)(CO)-

(NH₂CH{Me}Ph)]PF₆(7). The two diastereomeric salts

D.M. Wong, S.J. Simpson / Polyhedron 25 (2006) 2303–2317

2307

2 and no other tractable products. Treatment of 2 with

excess sodium iodide and excess trimethylamine-N-oxide

in methanol under reﬂux generated [(g⁵-C₅H₅)-

Ru(PPh₃)(CN^tBu)(I)] (8) in moderate yield together with

recovered 2. Since this reaction probably proceeds by

the generation of iodine and its subsequent reaction with

2 we proceeded to react 2 with iodomethane in reﬂuxing

1,2-dichloroethane or by photolysis of a solution in

dichloromethane containing iodine. Careful optimization

of these reactions suggested initial conditions for the

attempted decomplexation of 3, the chiral metal centre

target. Samples of 3RS (90% d.e.) and 3SS (96% d.e.)

were converted via the iodomethane route into 8 in 98%

and 82% isolated yield, respectively. A sample of 3RS

of the same purity was also converted via the photolysis

route in 54% isolated yield.

the diastereomers (Table 1). In particular the cyclopentadie-

nyl group and a-methyl resonances reﬂect the neighbouring

phenyl ring induced shieldings. The N-methyl group shiel-

dings also reﬂect the existence of two diastereomers with

these groups pointing towards the isonitrile unsaturation

(deshielded) and two with these groups pointing away

(shielded).

Calculation of the total quantities of each diastereomer

isolated reveals the overall ratio is 12.3%:41.4%:

4.9%:41.4%; fortuitously the quantity of 10RSS and

10SRS material is similar and demonstrates that the metal

centre prefers to induce opposite chirality at the nitrogen

centre. While the isolated material does not represent

100% of the theoretical yield it is clear that a preference

of at least 4:1 is present. Fig. 3 below shows all four

diastereomers.

The samples of 8 obtained were assayed for enantio-

meric purity by two methods; reaction with S(À)-a-methyl-

benzylamine to regenerate 3RS and 3SS and by a ¹H

NMR spectroscopic method. Addition of R(À)-2,2,2-tri-

ﬂuoro-1-(9-anthryl)ethanol (R(À)-TFAE) to racemic 8

causes doubling of the cyclopentadienyl and tert-butyl

resonances with the optimum separation (300 MHz: 5.7

and 3.6 Hz, respectively) occurring on addition of 1.2

equivalents of TFAE. Both methods conﬁrmed that the

samples of 8 obtained on decomplexation from 3RS and

3SS were racemic ( 2%). The high chemical stability of

3 and the nature of the forcing conditions required to

produce 8 demonstrates that our resolution strategy was

unsuccessful.

The reaction of a secondary amine ligand could poten-

tially generate an additional chiral centre and [(g⁵-C₅H₅)-

Ru(PPh₃)(CN^tBu)(Cl)] (1) was reacted with diethylamine

and TlPF₆in dichloromethane to conﬁrm that a secondary

amine could ligate to the ruthenium centre. The orange-

yellow crystals of [(g⁵-C₅H₅)Ru(PPh₃)(CN^tBu)(NHEt₂)]PF₆

(9) were moderately air-sensitive and showed a single band

at 3269 cm^À1(m_NH) in the infrared spectrum in contrast to

the two bands seen for a primary amine containing cation.

The ¹H and ¹³C{¹H} NMR spectra of 9 demonstrated

that the ethyl groups of the diethylamine ligand were dia-

stereotopic as expected. S(À)-N-methyl-1-phenylethylamine

reacted similarly to yield yellow crystalline [(g⁵-C₅H₅)-

Ru(PPh₃)(CN^tBu)(NH(Me)CH{Me}Ph)]PF₆(10) as a mix-

ture of four diastereomers in the ratio 9.6%:38.5%:3.9%:

48.0% while concentration of the initial mother-liquor

and crystallization produced the same diastereomers in

the ratio 21.7%:52.2%:8.7%:17.4%. While these precise

ratios reﬂect the degree of crystallization achieved they also

indicate the relative stability of the nitrogen chiral centres,

these being freely selectable under the reaction conditions

unlike the ruthenium and carbon chiralities, one of which

is not racemisable under the conditions and the other of

which is set. Analysis of the ¹H NMR spectra of these

mixtures using the criteria established above for the relative

disposition of the metal and carbon chiralities suggests

the assignments S_RuS_NS_C:S_RuR_NS_C:R_RuR_NS_C:R_RuS_NS_Cfor

The conformations shown are those that can be reason-

ably predicted based upon the structures of the diastereo-

meric salts of 3RR and 3SR analyzed earlier by H NMR

spectroscopy and conﬁrmed by X-ray single crystal struc-

ture determination. It is clear that the N-methyl group

would prefer to be external (SRS,RSS) rather than internal

(RRS,SSS) on steric grounds.

1

Table 1

Comparison of chemical shifts for the diastereomers of 3 and 10

Assignments

3RS

3SS

10RSS

10RRS

10SRS

10SSS

CPh

7.25

4.29

3.35

3.69,

1.87

6.97

4.72

3.34

3.49,

2.13

7.05

4.64

3.63

2.05

6.65

4.74

3.73

obscured

6.80

4.81

3.88

2.51

6.15

4.83

3.78

1.85

C₅H₅

CHMe

NH

NHMe

CHMe

CMe₃

2.29

1.34

1.25

2.53

1.40

1.18

2.23

1.55

1.30

2.46

1.50

1.26

1.11

1.30

1.25

Ph

Me

H

CN^tBu

N

Me

Ru

H

Ph

N

CN^tBu

H

Me

PPh₃

S_RuR_NS_C

R_RuS_NS_C

Ph

Me

H

CN^tBu

N

H

Me

Ru

Me

CN^tBu

H

Ph

N

Me

H

PPh₃

R_RuR_NS_C

S_RuS_NS_C

Fig. 3. The diastereomers of 10 derived from S(À)-N-methyl-1-

phenylethylamine.

2308

D.M. Wong, S.J. Simpson / Polyhedron 25 (2006) 2303–2317

An alternative method to produce a ruthenium complex

3. X-ray crystal structures

resolved at the metal centre which could be taken on

further in asymmetric chemistry would involve the

preparation of [(g⁵-C₅H₅)Ru(PPh₃)(CNZ)(Cl)], where Z

contains a chiral element. Combining our experience of

the amine chemistry and the utility of the ring shielding

eﬀect as a spectroscopic probe we decided to use S(À)-a-

methylbenzylisonitrile (S(À)-Ph(Me)CHNC) as a ligand.

A sample of S(À)-a-methylbenzylamine was determined

The X-ray single crystal structure of 2 was determined at

223 K; many crystals of 2 were examined under a polariz-

ing microscope and found to be twinned but a few single

crystals were located and examined. The crystal used for

the data collection was of an acceptable quality for the

intended purpose. The crystal specimens of 3RR and 3SR

were of very high optical quality and highly faceted; it

would be possible to resolve a mixed sample by triage,

picking the diastereomers apart under a microscope. Both

structures were determined at 233 K (Table 2).

1

to be >99.6% e.e by H NMR spectroscopy using R(À)-

TFAE as a chiral auxiliary [19]. Conversion of this amine

to the isonitrile was achieved in 50% yield by a phase trans-

fer Hofmann carbylamine procedure; polarimetry and sub-

sequent reactions indicate that the optical purity was

retained in the product.

The crystal of 2 contained molecules with the R-conﬁg-

uration at ruthenium and the absolute structure parameter

of 0.31(8) reﬂects the correct choice for the single chiral

centre. Analysis of a second crystal revealed that it only

contained molecules with an S-conﬁguration at ruthenium

(Rogers g 1.11(12)), thus the bulk sample is probably best

described as a racemic conglomerate. The unit cell also

contained one molecule of dichloromethane per cation.

Racemic twin reﬁnement was attempted on the data

obtained for the ﬁrst crystal and did not signiﬁcantly

improve the reﬁnement; the nature of the sample did not

suggest that further exploration was necessary or required.

The overall geometry of the cation is pseudo octahedral

as expected with the cyclohexyl group in a chair conforma-

tion with the proton on C(41) pointing in towards the iso-

nitrile ligand (Fig. 4). One of the amine protons points

towards the centre of a phenyl ring while the other is ori-

ented away as anticipated providing the origin of the

chemical shift diﬀerentiation in the ¹H NMR spectrum.

Reaction of [(g⁵-C₅H₅)Ru(PPh₃)₂(Cl)] with S(À)-a-

methylbenzylisonitrile in hot toluene gave a bright orange

powder after chromatography in 45% yield. Characteriza-

tion of this material showed it to be a racemic mixture of

the diastereomeric salts [(g⁵-C₅H₅)Ru(PPh₃)(CNCH-

(Me)Ph)(Cl)] (11SS,11RS). The solid and solution infrared

spectra of the product contained bands at 2102 and

2104 cm^À1assignable to m(CN), while the mass spectrum

contained the parent ion and then subsequent losses of

the chloro group and the isonitrile group. It proved possi-

ble with considerable eﬀort to enrich a sample to 30% d.e.

by crystallization from toluene–hexane mixtures in order to

1

assign the individual H NMR parameters to each diaste-

1

reomer. The H NMR spectroscopic signals of the CHMe

(d 4.87 and 4.70) and CHMe (d 1.29 and 1.39) groups were

baseline resolved in the diastereomer mixture and could be

used quantitatively. The remoteness of the chiral centre in

11 in comparison to the amine complexes such as 3 did not

permit complete conﬁdence in an absolute assignment

because the linear isonitrile functionality removes the a-

methyl group away from the ring current of the triphenyl-

phosphine aromatic rings, but after looking at the solid

state structures of 12, we assign the signals at d 4.87 and

1.29 to 11SS.

˚

The ruthenium–nitrogen bond length of 2.172(8) A

˚

compares with 2.190(5), 2.216(2) and 2.174(8) A for

The mixture (11SS,11RS) was converted into the

ammine salts [(g⁵-C₅H₅)Ru(PPh₃)(CNCH(Me)Ph)(NH₃)]-

PF₆(12SS,12RS) by means of ammonium hexaﬂuorophos-

phate and thallium carbonate in hot methanol with a view

to improving the separation of the diastereomeric centres

by crystallization. It should be noted that the ruthenium

chirality descriptors reverse when the chloro ligand is

replaced by the ammine ligand. Thus, assuming retention

at the metal centre, 11SS produces 12RS and 11RS pro-

duces 12SS. The yellow crystalline product obtained in

47% yield proved to an equimolar mixture of the diastereo-

meric salts. All attempts to eﬀect partial resolution failed

and indeed single crystals of 12 proved to contain both dia-

stereomeric centres. In particular the single crystal used for

the X-ray single crystal structure determination was later

dissolved in CD₂Cl₂and conﬁrmed this point and that

the equimolar mixture was perfectly conﬁgurationally sta-

ble in solution.

Fig. 4. The cation present in 2 showing 25% probability ellipsoids.

Table 2

Crystallographic data collection parameters for 2, 3SR, 3RR, and (12RS,12SS)

Identiﬁcation code

Empirical formula

Formula weight

Temperature (K)

2 Æ CH₂Cl₂

₃₅H₄₄Cl₂F₆N₂P₂Ru

840.63

223(2)

3SR

C₃₆H₄₀F₆N₂P₂Ru

777.71

3RR

C₃₆H₄₀F₆N₂P₂Ru

777.71

(12RS,12SS)

C₃₂H₃₂F₆N₂P₂Ru

721.61

C

233(2)

˚

Wavelength (A)

0.71073

monoclinic

P2₁

0.71073

monoclinic

P2₁

0.71073

monoclinic

P2₁

0.71073

monoclinic

P2₁

Crystal system

Space group

Unit cell dimensions

˚

a (A)

11.048(3)

16.983(3)

11.204(2)

10.083(2)

18.628(3)

10.704(2)

9.444(2)

18.932(3)

11.022(2)

9.902(4)

17.934(8)

18.283(9)

˚

b (A)

˚

c (A)

a (ꢁ)

b (ꢁ)

c (ꢁ)

90

114.04(2)

90

1919.9(8)

90

117.68(2)

90

1780.4(6)

90

115.13(1)

90

1784.1(6)

90

92.63(4)

90

3243(3)

3

˚

Volume (A )

Z

2

4

Density (calculated) (Mg/m³)

Absorption coeﬃcient (mm^À1

F(000)

1.454

0.687

860

0.35 · 0.25 · 0.25

1.99–27.50

1.451

0.589

796

0.40 · 0.35 · 0.30

2.28–31.07

1.448

0.588

796

0.55 · 0.40 · 0.25

2.04–31.07

1.478

0.641

1464

0.45 · 0.20 · 0.14

2.27–25.09

)

Crystal size (mm³)

H range for data collection (ꢁ)

Index ranges

À1 6 h 6 14, À22 6 k 6 1,

À14 6 l 6 13

5598

À2 6 h 6 13, 0 6 k 6 27,

À15 6 l 6 14

7316

À2 6 h 6 13, 0 6 k 6 27,

À16 6 l 6 15

7368

0 6 h 6 11, À1 6 k 6 21,

À21 6 l 6 21

6726

Reﬂections collected

Independent reﬂections [R_int

Completeness to h_max(%)

Reﬁnement method

Data/restraints/parameters

Goodness-of-ﬁt on F²

Final R indices [I > 2r(I)]

R indices (all data)

]

4865 [0.0356]

99.2

5859 [0.0209]

99.8

5835 [0.0276]

99.4

6342 [0.0499]

99.3

full-matrix least-squares on F²

4864/1/433

full-matrix least-squares on F²

5859/18/424

full-matrix least-squares on F²

5835/34/460

full-matrix least-squares on F²

6342/1/776

1.096

1.024

1.085

1.022

R₁= 0.0712, wR₂= 0.1755

R₁= 0.0988, wR₂= 0.1905

0.31(8)

1.703 and À0.640

R₁= 0.0330, wR₂= 0.0755

R₁= 0.0425, wR₂= 0.0808

0.00(3)

0.554 and À0.453

R₁= 0.0333, wR₂= 0.0867

R₁= 0.0398, wR₂= 0.0918

À0.01(3)

R₁= 0.0544, wR₂= 0.1111

R₁= 0.0945, wR₂= 0.1290

0.03(6)

0.529 and À0.377

Absolute structure parameter

Largest diﬀerential peak and hole (e A

À3

˚

)

0.561 and À0.614

2310

D.M. Wong, S.J. Simpson / Polyhedron 25 (2006) 2303–2317

[(g⁵-C₅H₅)Ru(PPh₃)₂(NH₂CH₂Ph)]BF₄[20], [(g⁵-C₅H₅)Ru-

(PPh₃)(P{OMe}₃)(NH₂CMe₃)]SO₃CF₃[21], and [(g⁵-

C₅H₅)Ru(Cy₂PCH₂CH₂PCy₂)(NH₂C₈H₁₇)]SO₃CF₃[22],

respectively.

The structures of 3SR (Fig. 5) and 3RR (Fig. 6) have the

same gross geometry as that found in 2, and are both abso-

lute structure determinations with parameter values of

À0.01(3) and 0.00(3), respectively. The ruthenium–nitrogen

˚

bond lengths are 2.168(3) and 2.184(3) A, respectively, and

there are other small metric diﬀerences at the ruthenium

centre (Table 3). These diﬀerences for the diastereomeric

cations are likely to be due to steric factors such as the dis-

position of the benzyl group. The most striking structural

diﬀerence between 3RR and 3SR is the position of the a-

methyl group caused by the preference of the benzyl hydro-

gen atom to point inwards towards the isonitrile group in

both diastereomers. The proximity of the a-methyl group

in 3SR to the aromatic rings of the triphenylphosphine

ligand and of the a-phenyl group to the cyclopentadienyl

ring hydrogens supports the earlier ¹H NMR spectroscopic

analysis where both these interactions produce shielding

relative to 3RR. The diﬀerentiation of the two NH reso-

Fig. 6. The cation present in 3RR showing 25% probability ellipsoids.

Table 3

˚

Selected bond lengths (A) and bond angles (ꢁ) for 3RR and 3SR

1

3SR

3RR

nances in the H NMR spectrum of each diastereomeric

Ru(1)–P(1)

Ru(1)–N(1)

Ru(1)–C(6)

2.2948(10)

2.184(3)

1.936(3)

2.2964(10)

2.168(3)

1.946(4)

salt can also be seen to be a consequence of shielding from

aromatic rings. It is implicit in this analysis that the solid

state structures and solution structures for each diastereo-

meric cation are similar; given that rotation around the

Ru(1)–N(1) bond and about the N(1)–C(8) bond is likely

to be restricted in both diastereomers this is a reasonable

outcome.

P(1)–Ru(1)–N(1)

P(1)–Ru(1)–C(6)

N(1)–Ru(1)–C(6)

Ru(1)–N(1)–C(8)

Ru(1)–C(6)–N(2)

C(6)–N(2)–C(7)

89.71(9)

89.16(10)

95.65(13)

121.6(2)

171.9(3)

169.0(4)

90.53(9)

88.79(10)

90.10(15)

118.1(2)

175.2(4)

176.5(2)

The X-ray single crystal structure of 12 was carried out

at 233 K and the crystal contained both diastereomers of

the cation in the unit cell, permitting both a relative and

an absolute structure determination. The outcome is that

the isonitrile ligand was conﬁrmed to be of S-conﬁgura-

tion. Visual comparison of the two diastereomeric cations

suggests why the 12RS (Fig. 7) and 12SS (Fig. 8) com-

pounds are diﬃcult to fractionally separate and do co-crys-

tallize; the volumes and molecular shapes are very similar

in that only the disposition of the a-methyl group perturbs

the near mirror symmetry in the unit cell (see Table 4).

4. Conclusions

Diastereomers containing only two chiral centres can be

designated as p-(RR or SS) or n-(RS or SR) [23,24]. The

most soluble diastereomeric salts in dichloromethane are

the p-forms of 3, 4, 5, and 7; possibly this relationship will

also hold if small modiﬁcations are made to these generic

molecules (e.g., change of cyclopentadienyl group, use of

PR₂Ph in place of PPh₃, ring para-substitution at phenyl

rings of phosphine or amine). The use of ring shielding

1

eﬀects in H NMR spectroscopy has proved a useful tool

Fig. 5. The cation present in 3SR showing 25% probability ellipsoids.

in determining diastereomer ratio and relative assignment

D.M. Wong, S.J. Simpson / Polyhedron 25 (2006) 2303–2317

2311

Fig. 7. The cation present in 12RS showing 25% probability ellipsoids.

Fig. 8. The cation present in 12SS showing 25% probability ellipsoids.

Table 4

Selected bond lengths (A) and bond angles (ꢁ) for 12RS and 12RR

˚

when one or more chiral centres are present. For example

2 crystallized as a racemic conglomerate where each single

crystal is itself homochiral but the bulk sample is racemic, 3

can be crystallized to yield single crystals of each of the

four diastereomers, while 12 generated by using a homochi-

ral ligand produces single crystals containing both possible

diastereomers.

12RS 12RR

Ru(1)–P(1)

Ru(1)–N(1)

Ru(1)–C(6)

C(6)–N(2)

2.301(3) 2.298(3)

Ru(2)–P(3)

Ru(2)–N(3)

Ru(2)–C(56)

C(56)–N(4)

2.153(9)

1.917(14)

1.156(15)

2.182(9)

1.913(15)

1.175(17)

P(1)–Ru(1)–N(1)

P(1)–Ru(1)–C(6)

N(1)–Ru(1)–C(6)

Ru(1)–C(6)–N(2)

C(6)–N(2)–C(7)

91.5(3)

88.6(3)

91.4(4)

177.7(11)

174.6(12)

P(3)–Ru(1)–N(3)

P(3)–Ru(2)–C(56)

N(3)–Ru(2)–C(56)

Ru(2)–C(56)–N(4)

C(56)–N(4)–C(57)

91.5(3)

88.4(4)

89.7(5)

176.6(12)

165.5(15)

All of the cationic compounds in this work are conﬁgu-

rationally stable in solution at room temperature for peri-

ods in excess of one week. In particular the single crystal

actually used for each structural measurement was re-

1

dissolved and monitored by H NMR spectroscopy within

for this family of compounds and extends the work of

Davies on cyclopentadienyliron compounds. The crystallo-

graphic work illustrates some of the variations possible

minutes of solution and then periodically. The solution

spectra exactly mirrored the composition determined in

the crystal. The nature of the ligands used in this study

2312

D.M. Wong, S.J. Simpson / Polyhedron 25 (2006) 2303–2317

and the strong metal–ligand bonding undoubtedly prevents

epimerization which has been found to be a major problem

elsewhere [6–8]. Indeed the bond strengths in compounds

such as 2 and 3 prevented our ultimate aim of producing

resolved at ruthenium materials for subsequent asymmetric

synthesis.

5.2. Preparation of S(À)-a-methylbenzylisonitrile

{S(À)-CNCHMePh}

The published procedure for the synthesis of tert-butyl-

isonitrile was adapted [29,30].

A two-neck round-

bottomed ﬂask (500 ml) equipped with a magnetic stirrer

bar, a reﬂux condenser, and a pressure equalising drop-

ping funnel was charged with sodium hydroxide (24 g,

0.6 mol). Stirring was commenced and deionized water

(3 · 30 ml) was added in portions to maintain smooth

stirring. The funnel was charged with a mixture of

S(À)-a-methylbenzylamine (20 ml, 0.158 mol), chloroform

(15 ml, 0.188 mol) and benzyltriethylammonium chloride

(0.4 g, 1.75 mmol) in dichloromethane (30 ml). The mix-

ture was added dropwise to the stirred, warm solution

(at ca. 45 ꢁC) over a 30-min period. The reaction mixture

began to reﬂux slowly 10 min after initiation of the addi-

tion and subsided within 2 h; stirring was continued for

an additional hour. The reaction mixture was diluted with

ice and deionized water (1:1, 150 ml), and the organic

layer (pale yellow) was separated and retained. The col-

ourless aqueous layer was extracted with dichloromethane

(2 · 30 ml), and the combined extracts were successively

washed with deionized water (20 ml), aqueous sodium

chloride solution (5% w/w, 20 ml), and dried over excess

anhydrous magnesium sulphate for 1 h.

5. Experimental

5.1. General comments

All reactions and preparations were carried out using con-

ventional Schlenk tube techniques and all solvents were

degassed prior to use. Chromatographic work-ups were per-

formed with columns made up of Grade IV neutral alumina

prepared with petroleum ether (40–60 ꢁC). Cyclohexylamine

and diethylamine were dried over barium oxide and distilled

before use. R(+)- and S(À)-a-methylbenzylamine (99.6%

e.e.) were supplied by Seal Sands Chemicals Ltd. The

amines R(À)-a-cyclohexylethylamine (98% e.e.) and ( )-1-

(1-naphthyl)ethylamine from Aldrich were used as supplied.

S(À)-N-methyl-1-phenylethylamine (98% e.e.) was supplied

by Fluka Chemika. R(À)-2,2,2-triﬂuoro-1-(9-anthryl)etha-

nol (99.5% e.e., Aldrich) was dried under reduced pressure

at 70 ꢁC overnight. Photolysis reactions were performed

using a P.W. Allen A409 medium pressure mercury lamp

(313, 366 nm). Infrared spectra were recorded on a Perkin–

Elmer 1710 FT-IR instrument calibrated against polystyrene

ﬁlm and only signiﬁcant absorption bands are reported.

Optical rotation measurements were made on an Optical

Activity Ltd. AA–100 polarimeter. Nuclear magnetic reso-

nance spectra were recorded on Bruker AC300

(300.13 MHz, ¹H NMR; 75.47 MHz, ¹³C; 121.49 MHz,

³¹P) and Avance 400 (400.13 MHz, ¹H NMR;

The drying agent was removed by ﬁltration, washed

with dry dichloromethane (10 ml), and the pale yellow

combined ﬁltrate and washings were distilled at ca. 65 ꢁC

using a Vigreux column (1.5 cm · 10 cm) to remove the sol-

vent (dichloromethane).

The remaining yellow solution was placed under nitro-

gen, and distilled under reduced pressure to remove ﬁnal

traces of dichloromethane, unreacted chloroform (at

25 ꢁC, 13 mmHg), and any excess S(À)-a-methylbenzyl-

amine (34–39 ꢁC, 3 mmHg). Finally, the fraction which dis-

tilled at 29–30 ꢁC (0.01 mmHg) was collected as a

colourless oil. Yield 11.0 ml (50%).

1

100.61 MHz, ¹³C; 161.98 MHz, ³¹P) spectrometers. All H

and ¹³C NMR chemical shifts are expressed in ppm relative

to SiMe₄(0.0 ppm) and ³¹P shifts relative to P(OPh)₃

(126.5 ppm). All spectra were measured at room tempera-

ture and homonuclear decoupling and DEPT spectra were

obtained when appropriate. Elemental analyses were

obtained by Butterworth Laboratories, London. Mass spec-

tral analyses were performed by Fast Atom Bombardment

(FAB) on a Kratos Concept S1 Spectrometer. Crystal struc-

ture data were collected on a Siemens R3m/V Diﬀractometer

at 233 K for [(g⁵-C₅H₅)Ru (PPh₃)(CN^tBu)(NH₂Cy)]PF₆(2),

(S_RuR_C)- and (R_RuR_C)-[(g⁵-C₅H₅)Ru(PPh₃)(CN^tBu)-

(NH₂CHMePh)]PF₆(3SR) and (3RR), and at 293 K for

[(g⁵-C₅H₅)Ru(PPh₃)(CNCH(Me)Ph)(NH₃)]PF₆(12). The

starting compounds [(g⁵-C₅H₅)Ru(PPh₃)(CN^tBu)Cl] (1)

[25,26], [(g⁵-C₅H₅)Ru(PPh₃)(CO)Cl] (6) [27], and [(g⁵-

C₅H₅)Ru(PPh₃)(CN^tBu)I] (8) [28] were prepared by the pub-

lished literature procedures.

IR (neat): m_max2140vs cm^À1(CN). ¹H NMR (CDCl₃): d

3

7.29–7.42 (m, 5H, Ph), 4.81 (q{1:1:1}t, 1H, J(HH) 7 Hz

2

3

and J(NH) 2 Hz, CHMeN), 1.67 (d{1:1:1}t, 3H, J(HH)

7 Hz and ³J(NH) 2 Hz, CHMeN) ppm. ¹³C{¹H} NMR

(CDCl₃): d 156.6 (t, J(CN) 4.5 Hz, CN), 138.6 (s, C_ipso

,

Ph), 128.9 (s, C_ortho, Ph), 128.3 (s, C_para, Ph), 125.4 (s,

1

C

_meta, Ph), 53.8 ({1:1:1}t, J(NC) 6 Hz, NCHPh), 25.1 (s,

CHMe). Optical rotation [a]₅₈₉À30.5ꢁ (21ꢁC, neat d

0.97 g cm^À1).

{Lit. [31] IR: m_max2141 (CN), 759, 698 cm^À1; ¹H NMR:

d 7.20 (m, 5H, Ph), 4.70 (m, 1H, CHMeN), 1.60 (d, 3H,

CHMeN) ppm. ¹³C{¹H} NMR (proton coupled): d 157.0

(t, J(CN) 4.3 Hz, CN), 53.8 (dt, NCHPh), 25.0 (q, CHMe)

ppm. Lit. [32,33] IR: m_max3000, 2950, 2160 (CN), 1600,

1

The optical purity of the R(+)- and S(À)-a-methylben-

1500, 770, 700 cm^À1; H NMR (CCl₄): d 7.35 (s, 5H, Ph),

1

zylamines was checked by H NMR spectroscopy using

4.83 (q{1:1:1}t, 1H, ³J(HH) 7 Hz and ²J(NH) 2 Hz,

3

R(À)-2,2,2-triﬂuoro-1-(9-anthryl)ethanol [19]. The optical

purity of the R(À)-a-cyclohexylethylamine and S(À)-N-

methyl-1-phenylethylamine was not veriﬁed.

CHMeN, 1.68 (d{1:1:1}t, 3H, J(HH) 7 Hz and J(NH)

2 Hz, CHMeN; Optical rotation [33,34] [a]₅₈₉À30ꢁ (20 ꢁC,

neat)}.

D.M. Wong, S.J. Simpson / Polyhedron 25 (2006) 2303–2317

2313

5.2.1. [(g⁵-C₅H₅)Ru(PPh₃)(CN^tBu)(NH₂Cy)]PF₆(2)

5.2.3. (R_RuR_C)-[(g⁵-C₅H₅)Ru(PPh₃)(CN^tBu)-

(NH₂CHMePh)]PF₆(3RR)

The mother-liquor from the crystallization procedure

above was reduced in volume (to ca. 10 ml) under reduced

pressure, and was left to crystallize out at À80 ꢁC for 2

days, producing yellow crystals which were isolated by

ﬁltration. Yield 0.07 g (21%) 98% d.e.

A Schlenk tube was charged with a solution of 1

(1.00 g, 1.83 mmol) in dichloromethane (40 ml), thallium

hexaﬂuorophosphate (1.60 g, 4.58 mmol), and a magnetic

stirrer bar. The orange solution was stirred in the dark

under nitrogen at room temperature for 1 h to give a

tomato-orange suspension. Excess cyclohexylamine

(2 ml, 17.48 mmol) was added and the solution was stir-

red in the dark at room temperature overnight. Filtra-

tion, evaporation of the ﬁltrate under reduced pressure,

and crystallization of the yellow residue from dichloro-

methane and diethylether (1:3, 40 ml) at À30 ꢁC for 2

days aﬀorded bright yellow crystals of 2. Yield 1.34 g

(97%). (Calc. for C₃₄H₄₂F₆N₂P₂Ru Æ CH₂Cl₂: C, 50.00;

H, 5.28; N, 3.33. Found: C, 49.93; H, 5.33; N, 3.30%).

IR (Nujol): m_max3311w, 3270w, 1587vw (NH), 2121m

IR (Nujol): m_max3325w, 3284w, 1591w (NH), 2126s

(CN), 1712w (Ph overtone) and 840s cm^À1(PF₆). ¹H

NMR (CD₂Cl₂): d 7.23–7.45 (m, 15H, PPh₃), 6.96–6.98

(m, 5H, Ph), 4.72 (s, 5H, C₅H₅), 3.49 (br.t, 1H, ^2,3J(HH)

10 Hz, NH), 3.35 (m, 1H, NCHMe), 2.12 (br.d, 1H,

3

²J(HH) 10 Hz, NH), 1.29 (d, 3H, J(HH) 7 Hz, Me), 1.25

(s, 9H, CMe₃) ppm. ¹³C{¹H} NMR (CD₂Cl₂): d 134.11

2

(d, C_ipso, PPh₃), 133.63 (d, J(PC) 11 Hz, C_ortho), 130.98

3

(s, C_para), 129.17 (d, J(PC) 8 Hz, C_meta), 128.23 (s, C_para

,

1

(CN), and 842s cm^À1(PF₆). H NMR (CD₂Cl₂): d 7.2–

Ph), 126.21 (s, C_meta, Ph), 81.81 (s, C₅H₅), 61.65 (s,

NCHPh), 30.65 (s, CMe₃), 22.63 (s, CHMe) ppm.

³¹P{¹H} NMR (CD₂Cl₂): d 59.08 (s, PPh₃) ppm.

7.5 (m, 15H, PPh₃), 4.74 (s, 5H, C₅H₅), 3.20 (br.t, 1H,

^2,3J(HH) 11 Hz, NH), 1.99 (br.d, 1H, ²J(HH) 12 Hz,

NH), 0.7–1.90 (m, 11H, Cy), 1.25 (s, 9H, CMe₃) ppm.

¹³C{¹H} NMR (CD₂Cl₂): d 134.20 (d, C_ipso, PPh₃),

133.77 (d, ²J(PC) 11 Hz, C_ortho), 131.05 (s, C_para),

129.14 (d, ³J(PC) 10 Hz, C_meta), 81.94 (s, C₅H₅), 59.29

(s, C₁, Cy), 58.00 (s, CMe₃), 35.82 (s, C₂, Cy), 33.13

(s, C₃, Cy), 30.73 (s, CMe₃), 25.49 (s, C₄and C₅, Cy),

25.29 (s, C₆, Cy) ppm. ³¹P{¹H} NMR (CD₂Cl₂): d

59.91 (s, PPh₃), À145.40 (septet, ¹J(PF) 712 Hz, PF₆)

ppm. MS (FAB): 611 [M À PF₆]⁺(50%), 512(100),

429(41).

5.2.4. (R_RuS_C)-[(g⁵-C₅H₅)Ru(PPh₃)(CN^tBu)-

(NH₂CHMePh)]PF₆(3RS)

To a solution of 1 (0.16 g, 0.29 mmol) in dichlorometh-

ane (30 ml) was added thallium hexaﬂuorophosphate

(0.4 g, 1.15 mmol) and excess S(À)-a-methylbenzylamine

(0.8 ml, 6.21 mmol). This yellow/orange solution was stir-

red overnight in the dark, at room temperature. After sol-

vent removal under reduced pressure, ﬁne yellow crystals

were obtained by crystallising the residue at À80 ꢁC from

dichloromethane and diethylether (1:3, 20 ml) overnight.

Yield 0.13 g (57%) 74% d.e.

5.2.2. (S_RuR_C)-[(g⁵-C₅H₅)Ru(PPh₃)(CN^tBu)-

(NH₂CHMePh)]PF₆(3SR)

A solution of 1 (0.23 g, 0.42 mmol) in dichloromethane

(30 ml) was treated with thallium hexaﬂuorophosphate

(0.42 g, 1.20 mmol) and excess R(+)-a-methylbenzylamine

(0.8 ml, 6.21 mmol). This yellow-orange solution was stir-

red in the dark at room temperature for 24 h. After ﬁltra-

tion and solvent removal from the ﬁltrate under reduced

pressure, ﬁne yellow crystals were obtained by crystalliz-

ing the residue at À30 ꢁC from dichloromethane and

diethylether (1:3, 20 ml) overnight. Yield 0.19 g (58%)

80% d.e. (Calc. for C₃₆H₄₀F₆N₂P₂Ru.CH₂Cl₂: C, 51.51;

H, 4.91; N, 3.25. Found: C, 50.93; H, 4.82; N, 3.39%.)

IR (Nujol): m_max3307w, 3264w, 1580vw (NH), 2115m

(CN), 1712w (Ph overtone), and 840s cm^À1(PF₆). ¹H

NMR (CD₂Cl₂): d 7.25–7.51 (m, 15H, PPh₃), 7.21–7.25

(m, 5H, Ph), 4.30 (s, 5H, C₅H₅), 3.71 (br.t, 1H,

^2,3J(HH) 11 Hz, NH), 3.35 (m, 1H, NCHMe), 1.88

5.2.5. (S_RuS_C)-[(g⁵-C₅H₅)Ru(PPh₃)(CN^tBu)-

(NH₂CHMePh)]PF₆(3SS)

The mother-liquor from above was concentrated to low

volume (to ca. 10 ml) under reduced pressure, and was left

to crystallize out in the dark at À30 ꢁC for 7 h, from which

yellow crystals were isolated. Yield 0.05 g (22%) 98% d.e.

5.2.6. (S_RuR_C)-[(g⁵-C₅H₅)Ru(PPh₃)(CN^tBu)-

(NH₂CHMeCy)]PF₆(4SR)

To a solution of 1 (0.21 g, 0.384 mmol) in dichlorometh-

ane (30 ml) was added thallium (I) hexaﬂuorophosphate

(0.40 g, 1.15 mmol) and excess R(À)-1-cyclohexylethyl-

amine (1.0 ml, 6.81 mmol). This orange solution was stirred

at room temperature in the dark under nitrogen for 67 h.

After solvent removal under reduced pressure, ﬁne yellow

crystals were obtained by crystallising the yellow residue

at À30 ꢁC from dichloromethane and diethylether (1:3,

40 ml) overnight. Yield 0.13 g (43%) 61% d.e. (Calc. for

C₃₆H₄₆F₆N₂P₂Ru Æ CH₂Cl₂: C, 51.16; H, 5.57; N, 3.22.

Found: C, 52.98; H, 5.57; N, 3.45%.) IR (Nujol): m_max

3312m, 3275w, 1584w (NH), 2116s (CN), 1721br.w (Ph

2

(br.d, 1H, J(HH) 11 Hz, NH), 1.30 (s, 9H, CMe₃), 1.11

3

(d, 3H, J(HH) 7 Hz, Me) ppm. ¹³C{¹H} NMR (CD₂Cl₂):

2

d 134.09 (d, C_ipso, PPh₃), 133.78 (d, J(PC) 11 Hz, C_ortho),

3

131.11 (s, C_para), 129.26 (d, J(PC) 10 Hz, C_meta), 128.86

(s, C_para, Ph), 127.42 (s, C_meta, Ph), 81.71 (s, C₅H₅),

1

61.76 (s, NCHPh), 30.82 (s, CMe₃), 25.34 (s, CHMe)

overtone) and 843vs cm^À1(PF₆). H NMR (CD₂Cl₂): d

ppm. ³¹P{¹H} NMR (CD₂Cl₂):

d

60.04 (s, PPh₃),

7.28–7.50 (m, 15H, PPh₃), 4.744 (s, 5H, C₅H₅), 3.15 (br.dd,

À145.30 (septet, ¹J(PF) 712 Hz, PF₆) ppm. MS (FAB):

1H, ^2,3J(HH) 12, 6 Hz, NH), 2.09 (qd, 1H, ^3,3J(HH) 6,

633 [M À PF₆]⁺(21%), 512(100), 429(54).

4 Hz, NCHMe), 1.24 (s, 9H, CMe₃), 0.70 (d, 3H, J(HH)

3

2314

D.M. Wong, S.J. Simpson / Polyhedron 25 (2006) 2303–2317

1

7 Hz, Me), 0.7–0.98, 1.05–1.45 and 1.65–1.74 (m, 11H,

2· C_bridging, Np), 133.66 (d, J(PC) 45 Hz, C_ipso, PPh₃),

133.71 (d, ²J(PC) 11 Hz, C_ortho, PPh₃), 131.19 (s, C_para

PPh₃), 129.67, 128.96, 126.81, 126.34, 126.09, 123.19, and

^2,3,3J(HH) 12, 12, 4 Hz, Cy) ppm. ¹³C{¹H} NMR

,

1

(CD₂Cl₂): d 133.77 (d, J(PC) 44 Hz, C_ipso, PPh₃), 133.71

(d, ²J(PC) 11 Hz, C_ortho, PPh₃), 131.10 (s, C_para, PPh₃),

129.24 (d, ³J(PC) 9 Hz, C_meta, PPh₃), 81.88 (s, C₅H₅),

60.99 (s, NCHCy), 42.59 (s, C₁, Cy), 30.66 (s, CMe₃),

26.84 (s, C₂, Cy), 26.71 (s, C₃, Cy), 26.34 (s, C₄and C₅,

Cy), 25.49 (s, C₆, Cy), 17.02 (s, CHMe) ppm. ³¹P{¹H}

NMR (CD₂Cl₂): d 59.57 (s, PPh₃) ppm. MS (FAB): 639

[M À PF₆]⁺(20%), 512(100), 429(31).

122.57 (s, 7· C_aryl, Np), 129.32 (d, J(PC) 10 Hz, C_meta,

3

PPh₃), 81.50 (s, C₅H₅), 58.34 (s, NCHNp), 30.68 (s,

CMe₃), 25.53 (s, CHMe) ppm. ³¹P{¹H} NMR (CD₂Cl₂):

d 59.46 (s, PPh₃) ppm.

5.2.9. (R_RuR_C,S_RuS_C)-[(g⁵-C₅H₅)Ru(PPh₃)(CN^tBu)-

(NH₂CHMe{1-C₁₀H₇})]PF₆(5RR,5SS)

The mother-liquor from above was reduced in volume

(ca. 10 ml) under reduced pressure, and was then left to

crystallize out in the dark at À30 ꢁC for 46 h, from which

yellow crystals were isolated. Yield 0.07 g (23%) 56% d.e.

of the least soluble (5RS,5SR). (Calc. for C₄₀H₄₂F₆-

N₂P₂Ru Æ 0.5CH₂Cl₂: C, 55.90; H, 4.98; N, 3.22. Found:

C, 56.29; H, 4.75; N, 3.40%.) IR (Nujol): m_max3314w,

3257w, 1587w (NH), 2108m (CN) and 841s cm^À1(PF₆).

¹H NMR (CD₂Cl₂): d 7.70 (m, 2H, Np), 7.20–7.59 (m,

5.2.7. (R_RuR_C)-[(g⁵-C₅H₅)Ru(PPh₃)(CN^tBu)-

(NH₂CHMeCy)]PF₆(4RR)

The mother-liquor from above was concentrated to low

volume (ca. 10 ml) under reduced pressure, and was then

left to crystallize out in the dark at À30 ꢁC overnight, from

which yellow crystals were isolated. Yield 0.11 g (37%) 79%

d.e.

IR (Nujol): m_max3315w, 3274w, 1585w (NH), 2121m

(CN), 1725br.w (Ph overtone) and 843s cm^À1(PF₆). H

19H, Np and PPh₃), 7.40 (t, 1H, J(HH) 7 Hz, Np), 4.35

1

3

NMR (CD₂Cl₂): d 7.28–7.51 (m, 15H, PPh₃), 4.745 (s,

5H, C₅H₅), 2.95 (br.t, 1H, ^2,3J(HH) 11 Hz, NH), 2.00 (m,

1H, NCHMe), 1.26 (s, 9H, CMe₃), 0.99 (d, 3H, J(HH)

(m, 1H, NCHMe), 4.77 (s, 5H, C₅H₅), 3.84 (br.t, 1H,

^2,3J(HH) 11 Hz, NH), 3.05 (br.d, 1H, ²J(HH) 11 Hz,

3

NH), 1.42 (d, 3H, J(HH) 7 Hz, Me), 1.21 (s, 9H, CMe₃)

7 Hz, Me), 0.52–0.70, 1.0–1.3 and 1.60–1.83 (m, 11H,

ppm. ¹³C{¹H} NMR (CD₂Cl₂): d ꢀ133.58 (d, ²J(PC)

^2,3,3J(HH) 11, 11, 4 Hz, Cy) ppm. ¹³C{¹H} NMR

11 Hz, C_ortho, PPh₃), ꢀ133.37 (d, ¹J(PC) 45 Hz, C_ipso

,

1

(CD₂Cl₂): d 134.02 (d, J(PC) 44 Hz, C_ipso, PPh₃), 133.83

PPh₃), ꢀ129.13 (d, ³J(PC) 10 Hz, C_meta, PPh₃), 128.30,

126.65, 126.25, 122.25, and 121.90 (s, C_aryl, Np), 81.81 (s,

C₅H₅), 56.93 (s, NCHNp), 30.78 (s, CMe₃), 22.95 (s,

CHMe) ppm. ³¹P{¹H} NMR (CD₂Cl₂): d 59.57 (s, PPh₃)

ppm. MS (FAB): 682 [M À PF₆]⁺(34%), 511(100), 428(44).

(d, ²J(PC) 11 Hz, C_ortho, PPh₃), 131.12 (s, C_para, PPh₃),

129.22 (d, ³J(PC) 10 Hz, C_meta, PPh₃), 81.95 (s, C₅H₅),

61.35 (s, NCHCy), 45.22 (s, C₁, Cy), 30.74 (s, CMe₃),

29.61 (s, C₂, Cy), 27.04 (s, C₃, Cy), 26.50 (s, C₄and C₅,

Cy), 25.91 (s, C₆, Cy), 16.24 (s, CHMe) ppm. ³¹P{¹H}

NMR (CD₂Cl₂): d 59.68 (s, PPh₃) ppm.

5.2.10. (R_RuS_C)-[(g⁵-C₅H₅)Ru(PPh₃)(CO)-

(NH₂CHMePh)]PF₆(7RS)

5.2.8. (R_RuS_C,S_RuR_C)-[(g⁵-C₅H₅)Ru(PPh₃)(CN^tBu)-

(NH₂CHMe{1-C₁₀H₇})]PF₆(5RS,5SR)

To a yellow-brown solution of 6 (0.50 g, 1.02 mmol) in

dichloromethane (30 ml) was added thallium hexaﬂuoro-

phosphate (0.81 g, 2.33 mmol) and S(À)-a-methylbenzyl-

amine (1 ml, 7.76 mmol). The brown mixture was stirred

under nitrogen at 45–50 ꢁC for 17.5 h (overnight) to give

a cream brown suspension. Filtration, washing the oﬀ-

white residue with dichloromethane (3 · 5 ml) gave a

brown ﬁltrate. After solvent removal under reduced pres-

sure, the brown residue was extracted with diethylether

(5 · 10 ml) and ﬁltered to give a yellow solution. This

was placed onto an alumina column (2 · 15 cm), the

excess amine was eluted with light petroleum ether

(60 ml) and the pale yellow product band was eluted with

dichloromethane (60 ml) to give a pale yellow solution.

After solvent removal under reduced pressure, a mustard

yellow powder was obtained by crystallising the residual

oil from diethylether (10 ml) at À80 ꢁC for 54 h. Yield

0.04 g (5%) 42% d.e. (Calc. for C₃₂H₃₁F₆NOP₂Ru: C,

53.19; H, 4.32; N, 1.94. Found: C, 53.30; H, 4.60; N,

1.75%.) IR (Nujol): m_max3306w, 3267w, 1587vw (NH),

An orange solution of 1 (0.20 g, 0.366 mmol) and

thallium (I) hexaﬂuorophosphate (0.40 g, 1.15 mmol) in

dichloromethane (30 ml) was stirred for 90 min to give a

tomato-orange solution. Excess ( )-1-(1-naphthyl)ethyla-

mine (1.0 ml, 6.21 mmol) was added. The mixture was stir-

red at room temperature in the dark under nitrogen for

52 h to give a bright yellow solution. After solvent removal

under reduced pressure, the yellow treacly solid was

washed with diethylether (2 · 15 ml). Extraction with

dichloromethane (2 · 15 ml), ﬁltration, and concentration

under reduced pressure (ca. 10 ml) gave a yellow solution.

Addition of diethylether (30 ml) aﬀorded ﬁne yellow crys-

tals on storing at À30 ꢁC under nitrogen overnight. Yield

0.05 g (17%) 73% d.e.

IR (Nujol): m_max3303w, 3271vw, 1587vw (NH), 2108m

1

(CN) and 843vs cm^À1(PF₆). H NMR (CD₂Cl₂): d 7.93–

3

7.97 (m, 2H, Np), 7.88 (d, 1H, J(HH) 8 Hz, Np), 7.24–

7.63 (m, 19H, Np and PPh₃), 4.35 (m, 1H, NCHMe),

4.23 (s, 5H, C₅H₅), 3.68 (br.t, 1H, ^2,3J(HH) 11 Hz, NH),

1

1950sh.w, 1941m (CO) and 843s cm^À1(PF₆). H NMR

2

3

2.15 (br.d, 1H, J(HH) 11 Hz, NH), 1.29 (d, 3H, J(HH)

7 Hz, Me), 1.23 (s, 9H, CMe₃) ppm. ¹³C{¹H} NMR

(CD₂Cl₂): d 138.97 (s, C_ipso, Np), 134.18 and 130.82 (s,

(CD₂Cl₂/CDCl₃): d 7.55–7.60 and 7.34–7.40 (m, 15H,

PPh₃), 7.26 (m, 5H, Ph), 4.57 (s, 5H, C₅H₅), 4.26 (br.t,

1H, ^2,3J(HH) 11 Hz, NH), 3.26 (br.sextet {br.m}, 1H,

D.M. Wong, S.J. Simpson / Polyhedron 25 (2006) 2303–2317

2315

³J(HH) 7 Hz, NCHMe), 1.95 (br.d, 1H, ²J(HH) 11 Hz,

(0.20 g, 0.57 mmol) and excess S(À)-N-methyl-1-phenylethyl-

amine (0.25 ml, 1.72 mmol). This yellow-orange solution

was stirred in the dark at room temperature for 24 h. After

ﬁltration and solvent removal from the ﬁltrate under

reduced pressure, ﬁne yellow crystals were obtained by

crystallizing the residue at À30 ꢁC from dichloromethane

and diethylether (1:3, 15 ml) overnight. Yield 0.18 g (62%).

IR (Nujol): m_max3263w, 1585vw (NH), 2122s (CN), and

841br.s cm^À1(PF₆). MS (FAB): 647 [M À PF₆]⁺(11%),

511(100), 454(11), 429(38).

3

NH), 1.09 (d, 3H, J(HH) 7 Hz, CHMe) ppm. ¹³C{¹H}

2

NMR (CD₂Cl₂/CDCl₃): d 203.92 (d, J(PC) 19 Hz, CO),

134.34 (d, C_ipso, PPh₃), 133.95 (d, ²J(PC) 11 Hz, C_ortho

PPh₃), 132.11 (s, C_para, PPh₃), 129.92 (d, J(PC) 10 Hz,

,

3

C

_meta, PPh₃), 129.54 (s, C_ortho, Ph), 129.12 (s, C_para, Ph),

128.12 (s, C_meta, Ph), 86.39 (s, C₅H₅), 64.19 (s, NCHPh),

25.59 (s, CHMe) ppm. ³¹P{¹H} NMR (CD₂Cl₂/CDCl₃):

d 55.58 (s, PPh₃) ppm. MS (FAB): 578 [M À PF₆]⁺(37%),

457(63), 429(100).

The ¹H NMR spectrum of the crystals revealed four dia-

stereomers were present; integration of the cyclopentadie-

nyl ligand singlets produced a ratio of 5:20:2:25 for

SSS:SRS:RRS:RSS. Concentration of the mother-liquor

above to ca. 7 ml and cooling to À25 ꢁC gave yellow crys-

tals, yield 0.05 g. (17%), whose spectrum showed the diaste-

reomer ratio to be 5:12:2:4.

5.2.11. (S_RuS_C)-[(g⁵-C₅H₅)Ru(PPh₃)(CN^tBu)-

(NH₂CHMePh)]PF₆(7SS)

The partial data was obtained from the spectra obtained

from the sample of (7RS) above.

¹H NMR (CD₂Cl₂/CDCl₃): d 7.40–7.52 and 7.27–7.34

(m, 15H, PPh₃), 6.95 (m, 5H, Ph), 5.13 (s, 5H, C₅H₅),

3

1.39 (d, 3H, J(HH) 7 Hz, CHMe) ppm. ¹³C{¹H} NMR

10RSS: ¹H NMR (CD₂Cl₂): d 7.33–7.54 (m, 15H, PPh₃),

7.05 (m, 5H, Ph), 4.64 (s, 5H, C₅H₅), 3.63 (quintet, 1H,

³J(HH) 7 Hz, NHCH), 2.29 (d, 3H, ³J(HH) 6 Hz,

(CD₂Cl₂/CDCl₃): d 142.67 (s, C_ipso, PPh₃), 133.85 (d,

²J(PC) 11 Hz, C_ortho, PPh₃), 131.56 (s, C_para, PPh₃),

3

129.38 (d, ³J(PC) 10 Hz, C_meta, PPh₃), 127.03 (s, C_meta

,

NHCH₃), 2.05 (br.q, 1H, J(HH) 6 Hz, NH), 1.34 (d, 3H,

Ph), 86.71 (s, C₅H₅), 64.06 (s, NCHPh), 23.17 (s, CHMe)

ppm. ³¹P{¹H} NMR (CD₂Cl₂/CDCl₃): d 54.25 (s, PPh₃)

ppm.

³J(HH) 6 Hz, CHCH₃), 1.25 (s, 9H, Bu) ppm. ¹³C{¹H}

NMR (CD₂Cl₂): d 139.68 (s, C_ipso, Ph), 133.40 (d, J(PC)

44 Hz, C_ipso, PPh₃), 133.39 (d, ²J(PC) 11 Hz, C_ortho

PPh₃), 131.39 (s, C_para, PPh₃), 129.61 (d, J(PC) 10 Hz,

,

3

5.2.12. [(g⁵-C₅H₅)Ru(PPh₃)(CN^tBu)(NHEt₂)]PF₆(9)

A Schlenk tube was charged with a solution of 1 (0.20 g,

0.37 mmol) in dichloromethane (30 ml), thallium hexaﬂu-

orophosphate (0.19 g, 0.54 mmol), and a magnetic stirrer

bar. The orange solution was stirred in the dark under

nitrogen at room temperature for 1 h to give a tomato-

orange suspension. Excess diethylamine (0.4 ml,

3.87 mmol) was added and the solution was stirred in the

dark at room temperature overnight. Filtration, evapora-

tion of the ﬁltrate under reduced pressure, and crystalliza-

tion of the yellow residue from dichloromethane and

diethylether (1:3, 20 ml) at À30 ꢁC aﬀorded orange-yellow

crystals of 9. Yield 0.16 g (60%).

C_meta, PPh₃), 129.08 (s, C_ortho, Ph), 128.97 (s, C_para, Ph),

127.85 (s, C_meta, Ph), 82.18 (s, C₅H₅), 68.08 (s, NCHPh),

58.57 (s, CMe₃), 45.70 (s, NHCH₃), 30.48 (s, CMe₃),

23.99 (s, CHMe) ppm. ³¹P{¹H} NMR (CD₂Cl₂): d 58.3

(s, PPh₃) ppm.

10SRS: ¹H NMR (CD₂Cl₂): d 7.32–7.51 (m, 15H,

PPh₃), 6.80 (m, 5H, Ph), 4.81 (s, 5H, C₅H₅), 3.88 (qd,

1H, ³J(HH) 7 Hz, ⁴J(PH) 1.8 Hz, NHCH), 2.51 (br.s,

3

1H, NHCH₃), 2.23 (d, 3H, J(HH) 6 Hz, NHCH₃), 1.55

(d, 3H, ³J(HH) 7 Hz, CHCH₃), 1.30 (s, 9H, Bu) ppm.

¹³C{¹H} NMR (CD₂Cl₂): d 139.41 (s, C_ipso, Ph), 133.52

(d, J(PC) 44 Hz, C_ipso, PPh₃), 133.56 (d, ²J(PC) 10 Hz,

C

_ortho, PPh₃), 131.39 (s, C_para, PPh₃), 129.62 (d, ³J(PC)

IR (Nujol): m_max3269w, 1613vw (NH), 2129m, 2062sh.w

9 Hz, C_meta, PPh₃), 129.64 (s, C_ortho, Ph), 127.82 (s, C_para

+

1

(CN), and 841s cm^À1(PF₆). H NMR (CD₂Cl₂): d 7.26–

C_meta, Ph), 82.44 (s, C₅H₅), 65.73 (s, NCHPh), 58.62 (s,

7.49 (m, 15H, PPh₃), 4.78 (s, 5H, C₅H₅), 2.78 (m, 2H,

^2,3J(HH) 7 Hz, CH₂CH₃), 2.67 (m, 2H, ^2,3J(HH) 7 Hz,

CH₂CH₃), 1.68 (br.s, 1H, NH), 1.20 (s, 9H, CMe₃), 1.08

(t, 3H, ^2,3J(HH) 7 Hz, CH₂CH₃), 0.44 (t, 3H, ^2,3J(HH)

7 Hz, CH₂CH₃) ppm. ¹³C{¹H} NMR (CD₂Cl₂): d 133.88

CMe₃), 41.35 (s, NHCH₃), 30.75 (s, CMe₃), 24.80 (s,

CHMe) ppm. ³¹P{¹H} NMR (CD₂Cl₂): d 58.1 (s, PPh₃)

ppm.

1

10SSS: H NMR (CD₂Cl₂): d 6.15 (m, 5H, Ph), 4.83 (s,

3

5H, C₅H₅), 3.78 (q, 1H, J(HH) 7 Hz, NHCH), 2.51 (br.s,

2

3

(d, C_ipso, PPh₃), 133.37 (d, J(PC) 11 Hz, C_ortho), 131.32

1H, NHCH₃), 2.46 (d, 3H, J(HH) 6 Hz, NHCH₃), 1.85

(s, C_para), 129.52 (d, ³J(PC) 10 Hz, C_meta), 82.08 (s,

C₅H₅), 58.45 (s, CMe₃), 56.13, 51.54 (2s, CH₂), 30.45 (s,

CMe₃), 14.67, 14.08 (2s, CH₃) ppm. ³¹P{¹H} NMR

(CD₂Cl₂): d 58.3 (s, PPh₃), À145.40 (septet, ¹J(PF)

712 Hz, PF₆) ppm. MS (FAB): 585 [M À PF₆]⁺(15%),

512(100), 454(9), 429(37).

(br.s, 1H, NHCH₃), 1.50 (d, 3H, J(HH) 7 Hz, CHCH₃),

3

1.26 (s, 9H, Bu) ppm. ¹³C{¹H} NMR (CD₂Cl₂): d 82.30

(s, C₅H₅), 60.79 (s, CMe₃), 47.18 (s, NHCH₃), 30.64 (s,

CMe₃), 14.75 (s, CHMe) ppm. ³¹P{¹H} NMR (CD₂Cl₂):

d 55.7 (s, PPh₃) ppm.

10RRS: ¹H NMR (CD₂Cl₂): d 6.65 (m, 5H, Ph), 4.74 (s,

5H, C₅H₅), 3.73 (qd, 1H, ³J(HH) 7 Hz, ⁴J(PH) 1.8 Hz,

NHCH), 2.53 (d, 3H, ³J(HH) 6 Hz, NHCH₃), 1.40 (d,

3H, ³J(HH) 7 Hz, CHCH₃), 1.18 (s, 9H, Bu) ppm.

¹³C{¹H} NMR (CD₂Cl₂): d 81.65 (s, C₅H₅), 41.05 (s,

NHCH₃), 30.36 (s, CMe₃), 20.80 (s, CHMe) ppm.

5.2.13. [(g⁵-C₅H₅)Ru(PPh₃)(CN^tBu)(NH(Me)-

CHMePh)]PF₆(10)

A solution of 1 (0.20 g, 0.37 mmol) in dichloromethane

(30 ml) was treated with thallium hexaﬂuorophosphate

2316

D.M. Wong, S.J. Simpson / Polyhedron 25 (2006) 2303–2317

5.2.14. (R_RuS_C,R_RuS_C)-[(g⁵-C₅H₅)Ru(PPh₃)-

(CNCHMePh)Cl] (11RS,11SS)

CHMe), 1.43 (d, 3H, ³J(HH) 7 Hz, CHMe) ppm.

¹³C{¹H} NMR ((CD₃)₂CO): d 135.41 (d, ¹J(PC) 45 Hz,

A suspension of 1 (0.70 g, 0.964 mmol) and excess (S)-a-

methylbenzylisonitrile (0.5 ml, 3.58 mmol) in toluene

(30 ml) was stirred and heated under reﬂux at 100 ꢁC under

nitrogen for 24 h. The cooled orange solution was reduced

in volume (to ca. 10 ml) under reduced pressure and put

onto an alumina column (1.5 · 20 cm.). PPh₃was eluted

with toluene (20 ml) and the orange product band eluted

with dichloromethane (30 ml) and ultimately acetone

(10 ml) to give an orange solution. Solvent removal under

reduced pressure gave a frothy orange solid. Wash with

petroleum ether (b.p. 40–60 ꢁC, 2 · 10 ml) followed by dry-

ing under reduced pressure aﬀorded a bright orange pow-

der. Yield 0.26 g (45%) [11SS:11RS 49:51].

C

_ipso, PPh₃), 134.29 (d, ²J(PC) 11 Hz, C_ortho, PPh₃),

131.29 (s, C_para, PPh₃), 129.58 (d, ³J(PC) 10 Hz, C_meta

,

PPh₃), 128.88 (s, C_ortho, PPh₃), 126.48 (s, C_para, Ph),

126.36 (s, C_meta, Ph), 82.52 (s, C₅H₅), 57.38 (s, NCHPh),

25.22 (s, CHMe), 25.09 (s, CHMe) ppm. ³¹P{¹H} NMR

(CD₂Cl₂): d 58.47 (s, PPh₃) and 58.17 (s, PPh₃) ppm. MS

(FAB): 577 [M À PF₆]⁺(24%), 560(100), 429(53).

5.3. Decomplexation of amine ligands

5.3.1. [(g⁵-C₅H₅)Ru(PPh₃)(CN^tBu)(NH₂Cy)]PF₆(2)

Method A. Complex 2 (0.10 g, 0.13 mmol), trimethyla-

mine N-oxide dehydrate (0.05 g, 0.45 mmol), and sodium

iodide (0.08 g, 0.53 mmol) were placed in a Fischer–Porter

bottle and suspended in methanol (30 ml). After heating

and stirring at 85 ꢁC for 48 h the solvent was removed

under reduced pressure. Extraction with dichloromethane

(30 ml) gave an orange-red solution which was washed with

water (3 · 10 ml). The dichloromethane layer was dried

with anhydrous magnesium sulphate, ﬁltered and the sol-

vent removed. The orange foam (0.09 g) was identiﬁed as

Recrystallization of a small sample (0.15 g) from toluene

and hexane (1:3, 20 ml) at À30 ꢁC for 16 h aﬀorded an oily

orange solid after ﬁltration and solvent removal under

reduced pressure. Yield 0.02 g [11SS:11RS 65:35].

IR (Nujol): m_max2104, 2102s cm^À1(CN). MS (FAB): 595

[M]⁺(38%), 560(85), 429(93).

11SS: ¹H NMR (CDCl₃): d 7.41–7.54 and 7.22–7.33 (m,

15H, PPh₃), 7.10–7.13 and 7.03–7.06 (m, 5H, Ph), 4.87 (q,

1H, ³J(HH) 7 Hz, CHMe), 4.55 (s, 5H, C₅H₅), 1.39 (d, 3H,

³J(HH) 7 Hz, CHMe) ppm. ¹³C{¹H} NMR (CDCl₃): d

1

a mixture (70:30) of 8 and unreacted 2 by H NMR and

infrared spectroscopies.

1

2

136.65 (d, J(PC) 44 Hz, C_ipso, PPh₃), 133.71 (d, J(PC)

Method B. A Fischer–Porter bottle was charged with 2

(0.20 g, 0.27 mmol), iodomethane (2 ml, 32 mmol), and

1,2-dichloroethane (30 ml). Heating and stirring for 22 h

at 120 ꢁC gave a dark red solution. Cooling, removal of

solvent, and washing the residue with petroleum ether

(2 · 5 ml) then water (3 · 5 ml) gave a red solid which

was dried under reduced pressure and identiﬁed as 8

11 Hz, C_ortho, PPh₃), 129.41 (s, C_para, PPh₃), 128.65 (s,

3

C

_para, Ph), 127.82 (d, J(PC) 11 Hz, C_meta, PPh₃), 125.37

(s, C_meta, Ph), 81.66 (s, C₅H₅), 56.20 (s, NCHPh), 25.01

(s, CHMe) ppm.

11RS: ¹H NMR (CDCl₃): d 7.41–7.54 and 7.22–7.33 (m,

15H, PPh₃), 7.10–7.13 and 7.03–7.06 (m, 5H, Ph), 4.70 (q,

1H, ³J(HH) 7 Hz, CHMe), 4.53 (s, 5H, C₅H₅), 1.29 (d, 3H,

³J(HH) 7 Hz, CHMe) ppm. ¹³C{¹H} NMR (CDCl₃): d

1

(0.16 g, 95%) by H NMR and infrared spectroscopies.

Method C. A stirred solution of 2 (0.10 g, 0.13 mmol)

and iodine (0.10 g, 0.39 mmol) in dichloromethane was

photolyzed for 24 h at room temperature. The product

solution was washed with sodium thiosulfate solution

(10%, 3 · 20 ml), the separated organic layer was washed

with water (3 · 20 ml), then dried with granular calcium

chloride. Removal of the solvent under reduced pressure

gave 8 (0.08 g, 95%).

1

2

136.47 (d, J(PC) 44 Hz, C_ipso, PPh₃), 133.71 (d, J(PC)

11 Hz, C_ortho, PPh₃), 129.41 (s, C_para, PPh₃), 128.65 (s,

3

C

_para, Ph), 127.82 (d, J(PC) 11 Hz, C_meta, PPh₃), 125.37

(s, C_meta, Ph), 81.66 (s, C₅H₅), 56.09 (s, NCHPh), 25.01

(s, CHMe) ppm.

5.2.15. (R_RuS_C,R_RuS_C)-[(g⁵-C₅H₅)Ru(PPh₃)-

(CNCHMePh)(NH₃)]PF₆(12RS,12SS)

A mixture of (11RS,11SS) (0.21 g, 0.353 mmol), ammo-

nium hexaﬂuorophosphate (0.25 g, 1.54 mmol), and thal-

lium (I) carbonate (0.50 g, 1.07 mmol) in methanol

(40 ml) was stirred at 60 ꢁC for 22 h. Removal of the

solvent under reduced pressure and crystallization of the

residue from dichloromethane–diethylether (1:3, 40 ml) at

À30 ꢁC aﬀorded yellow crystals of (12RS,12SS). Total

yield 0.12 g (47%) [racemate]. (Calc for C₃₂H₃₂F₆N₂P₂Ru Æ

3CH₂Cl₂: C, 43.06; H, 3.92; N, 2.87. Found: C, 43.56; H,

3.63; N, 3.13%.) IR (Nujol): m_max3370m, 3296w, 1622w

5.3.2. (S_RuS_C)-[(g⁵-C₅H₅)Ru(PPh₃)(CN^tBu)-

(NH₂CHMePh)]PF₆(3SS)

A sample of 3RS (96% d.e.) was treated under Method B

above. The isolated 8 was recrystallized from dichloro-

1

methane and petroleum ether. Analysis of the H NMR

spectra obtained in the absence and presence of R(À)-

TFAE (1.05 equiv) showed the product to be a racemic

mixture of 8R and 8S.

5.3.3. (R_RuS_C)-[(g⁵-C₅H₅)Ru(PPh₃)(CN^tBu)-

(NH₂CHMePh)]PF₆(3RS)

(NH), 2119br.s (CN) and 838br.vs cm^À1(PF₆). H NMR

1

((CD₃)₂CO): d 7.36–7.58 (m, 30H, PPh₃), 7.25–7.36 and

7.11–7.14 (m, 10H, Ph), 5.21 (q, 1H, ³J(HH) 7 Hz, CHMe)

and 5.19 (q, 1H, ³J(HH) 7 Hz, CHMe), 4.88 (s, 10H,

C₅H₅), 2.60 (br, 6H, NH₃), 1.52 (d, 3H, ³J(HH) 7 Hz,

A sample of 3RS (90% d.e.) was treated under Method B

above. The isolated 8 was treated with thallium hexaﬂuoro-

phosphate and excess S(À)-N-methyl-1-phenylethylamine

using the procedures described earlier. The crude product

D.M. Wong, S.J. Simpson / Polyhedron 25 (2006) 2303–2317

2317

before crystallization was shown by ¹H NMR spectroscopy

to be a racemic mixture of 3RS and 3SS.

tary data associated with this article can be found, in the

online version, at doi:10.1016/j.poly.2006.01.016.

A sample of 3RS (90% d.e.) was treated under Method

1

C above. The product 8 was analyzed by the H NMR–

TFAE method and by conversion back to 3. Both methods

revealed the original sample of 8 to be racemic.

References

[1] A. Werner, Chem. Ber. 44 (1911) 1887.

[2] H. Brunner, Angew. Chem., Int. Ed. Engl. 8 (1969) 382.

[3] R. Dabard, A. Meyer, G. Jaouen, C.R. Seances, Acad. Sci., Ser. C

C286 (1969) 201.

5.4. X-ray crystallography

[4] H. Brunner, Angew. Chem., Int. Ed. Engl 38 (1999) 1194.

[5] H. Brunner, Eur. J. Inorg. Chem. (2001) 905.

[6] J.W. Faller, J. Parr, A.R. Lavoie, New J. Chem. 27 (2003) 899.

[7] H. Brunner, T. Zwack, Organometallics 19 (2000) 2423.

[8] M. Pfeﬀer, Organometallics 19 (2000) 2427.

[9] F.M. Conroy-Lewis, S.J. Simpson, J. Organomet. Chem. 396 (1990)

83.

[10] A. von Zelewsky, Stereochemistry of Coordination Compounds,

Wiley, Chichester, 1995.

[11] G.J. Leigh, Nomenclature of Inorganic Chemistry, Blackwell,

Oxford, 1990.

[12] K. Stanley, M.C. Baird, J. Am. Chem. Soc. 97 (1975) 6598.

[13] T.E. Sloan, Top. Stereochem. 12 (1981) 1.

[14] N.H. Martin, N.W. Allen III, K.D. Moore, L. Vo, J. Mol. Struct.

(Theochem) 454 (1998) 161.

[15] Y.H. Lai, Z.L. Zhou, J. Org. Chem. 62 (1997) 925.

[16] S.G. Davies, I.M. Dordor, J.C. Walker, P. Warner, Tetrahedron

Lett. 25 (1984) 2709.

[17] B.K. Blackburn, S.G. Davies, M. Whittaker, in: I. Bernal (Ed.),

Stereochemistry of Organometallic and Inorganic Compounds, vol. 3,

Elsevier, Amsterdam, 1989, and references therein.

[18] F.M. Conroy-Lewis, S.J. Simpson, J. Chem. Soc., Chem. Commun.

(1986) 506.

[19] E. Francotte, R.W. Lang, T. Winkler, Chirality 3 (1991) 177.

[20] J.A. Cabeza, I. del Rio, M. Elena-Navarro, S. Garcia-Granda, Acta

Crystallogr., Sect. E: Struct. Rep. Online 57 (2001) m162.

[21] F.L. Joslin, M.P. Johnson, J.T. Mague, D.M. Roundhill, Organo-

metallics 10 (1991) 41.

The crystals of 2, 3SR, 3RR, and 12 were grown by slow

layer diﬀusion of diethyl ether into dichloromethane solu-

tions. Suitable crystal specimens were mounted on a glass

ﬁbre with epoxy resin and cooled on the goniometer head

by means of nitrogen gas. Precession photographs and

intensity data were collected on a Siemens R3m/V diﬀrac-

tometer using graphite monochromatized Mo Ka X-rays.

Cell dimensions were obtained from 25 to 50 centred reﬂec-

tions with 2h values from 20ꢁ to 32ꢁ. Intensity data were

collected using a 2h–h scan technique. The intensities of

three reﬂections measured periodically showed a decrease

of less than 1% over the data collection for all four data-

sets. An empirical absorption correction was applied using

azimuthal scan data for twelve selected reﬂections. The

structures were solved by standard heavy atom routines

and reﬁned by full-matrix least-squares methods. All non-

hydrogen atoms were given anisotropic temperature factors.

Hydrogen atoms were placed in the models at calculated

positions and allowed to ride on their respective carbon

atoms. The hexaﬂuorophosphate anions were rotationally

disordered in both 3SR and 3RR and were modeled by sim-

ple partial occupancy methods. Table 2 contains the collec-

tion and solution parameters for all four determinations.

[22] F.L. Joslin, M.P. Johnson, J.T. Mague, D.M. Roundhill, Organo-

metallics 10 (1991) 2781.

[23] J. Jacques, A. Collet, S.H. Wilen, Enantiomers, Racemates, and

Resolutions, Krieger, Malabar, 1981, pp. 251–252.

[24] I. Ugi, Z. Naturforsch. 20B (1965) 405.

[25] M.I. Bruce, R.C. Wallis, Aust. J. Chem. 34 (1981) 209.

[26] F.M. Conroy-Lewis, Ph.D. Thesis, University of Salford, 1987.

[27] S.G. Davies, S.J. Simpson, J. Chem. Soc., Dalton Trans. (1984)

993.

[28] F.M. Conroy-Lewis, A.D. Redhouse, S.J. Simpson, J. Organomet.

Chem. 366 (1989) 357.

Acknowledgements

We thank the E.P.S.R.C. for a studentship (D.M.W.)

and Johnson-Matthey Chemicals Ltd. for a generous loan

of ruthenium trichloride. Dr. N.M. Boag (Salford) is

thanked for stimulating discussions and we acknowledge

the use of the E.P.S.R.C. Chemical Database Service at

Daresbury [35].

[29] G.W. Gokel, R.P. Widera, W.P. Weber, Org. Synth. 55 (1976) 96.

[30] G.W. Gokel, R.P. Widera, W.P. Weber, Org. Synth. Coll. Vol. 6

(1988) 232.

Appendix A. Supplementary data

[31] K. Gollnick, S. Koegler, D. Maurer, J. Org. Chem. 57 (1992) 229.

[32] S. Terashima, K. Takashima, T. Sato, S. Yamada, Chem. Pharm.

Bull 21 (1973) 1135.

[33] S. Yamada, K. Takashima, T. Sato, S. Terashima, J. Chem. Soc. D

(1969) 811.

[34] D. Seebach, G. Adam, T. Gees, M. Schiess, W. Weigand, Chem. Ber.

121 (1988) 507.

[35] The United Kingdom Chemical Database Service, D.A. Fletcher,

R.F. McMeeking, D. Parkin, J. Chem. Inf. Comput. Sci. 36 (1996)

746.

The supplementary crystallographic data for this paper

are contained in the deposition ﬁles CCDC-277650 (2),

CCDC-277651 (3SR), CCDC-277652 (3RR), and CCDC-

277653 (12). These data can be obtained free of charge

via ccdc.cam.ac.uk/data_request/cif, by emailing data_

request@ccdc.cam.ac.uk, or by contacting The Cambridge

Crystallographic Data Centre, 12 Union Road, Cam-

bridge, CB2 1EZ, UK: fax +44 1223 336033. Supplemen-

Article Doi

Preparation of the chiral amine salts [(η5-C5H5)Ru(PPh3) (CNtBu)(NH2R)]PF6, RNH2 = α-methylbenzylamine, 1-cyclohexylethylamine, and 1-(1-naphthyl)ethylamine: X-ray single crystal structures of the diastereomeric salts (SRuRC)-

DOI: 10.1016/j.poly.2006.01.016

Source and publish data:

Authors:

Article abstract of DOI:10.1016/j.poly.2006.01.016

Full text of DOI:10.1016/j.poly.2006.01.016

Products guided by the article

R&D Labs maybe for 908140-64-7

Relevant to this article

Hot Product