Experimental and theoretical studies on the mechanism of the C-S bond activation of PdII thiolate/thioether complexes

Article

pubs.acs.org/Organometallics

Experimental and Theoretical Studies on the Mechanism of the C−S

Bond Activation of Pd^IIThiolate/Thioether Complexes

Sushil Kumar,^†Fabrice Guyon,*^,†Michael Knorr,^†Stephane Labat,^‡Karinne Miqueu,*^,‡

́

Christopher Golz,^§and Carsten Strohmann^§

^†Institut UTINAM, UMR CNRS 6213, Universite

Gray, 25030 Besancon, France

^‡CNRS/UNIV PAU & PAYS ADOUR, Institut des Sciences Analytiques et de Physico-Chimie pour l’environnement et les

́ ́ ́

Bourgogne Franche-Comte, Faculte des Sciences et des Techniques, 16 Route de

̧

́

Materiaux, UMR 5254, 64000 Pau, France

^§Technische Universitat Dortmund, Anorganische Chemie, Otto-Hahn Strasse 6, 44227 Dortmund, Germany

̈

S

* Supporting Information

ABSTRACT: Two equivalents of L (L = 4-methylthio-2-

thioxo-1,3-dithiole-5-thiolate or Medmit) react with cis-Pd-

(PR₃)₂Cl₂(R = Ph and Et) to aﬀord Pd(PR₃)(η¹-L)(η²-L) (R

= Et: 1 ; R = Ph: 2) complexes, which have been characterized

by X-ray crystallography. These compounds are dynamic in

solution due to an exchange of the thiomethyl groups on

palladium. Variable-temperature ¹H NMR spectroscopy reveals

a low coalescence temperature (173 K). Treatment of Pd(diphos)Cl₂(diphos = dppe or dppm) with 2 equiv of L aﬀords thiolato

complexes Pd(dppe)(η¹-L)₂(3) and Pd(dppm)(η¹-L)₂(4). Whereas 3 is rigid in solution with ﬁrm η²-coordination of dppe and

η¹-coordination of the thiolate, two linkage isomers Pd(η²-dppm)(η¹-L)₂and Pd(η¹-dppm)(η¹-L)(η²-L) coexist in a solution of

4. L coordinated on Pd^IIundergoes a S-demethylation reaction leading to dithiolene complexes and MeL. This transformation

requires high temperature, and its eﬃciency depends on the nature of the phosphines as well as the nature of the metal (Pd vs

Pt). DFT calculations reveal that the most likely mechanism depends on the lability of phosphines. Starting from M(PR₃)₂(η¹-

L)₂(M= Pd and Pt; R = Ph and Et), the favored sequence implies decoordination of one triethyl phosphine (M(PEt₃)(η¹-L)(η²-

L)₂as intermediate) or two triphenylphosphines (Pd(η²-L)₂as intermediate) followed by oxidative addition and reductive

elimination (OA/RE) reactions. In the case of PEt₃, this OA/RE sequence can also compete with an intramolecular nucleophilic

+

addition (A_N), which can be described as an attack of a thiolate sulfur atom on a CH₃carbocation. An intramolecular S_N2

process was found to be the most feasible, starting from M(dppe)(η¹-L)₂(M= Pd and Pt), with the nucleophile approaching the

thiomethyl group at an angle of 180° with respect to the C_CH−S bond. The inﬂuence of the coligand has also been studied

3

experimentally. Structurally characterized disulﬁde L−L dimer has been isolated upon reaction of 2 equiv of L with MCl₂(M =

Pd and Pt).

by complexation of the thioether group to the metal center,

imparting a diminution of electron density at the sulfur atom

and consequently an increase of electrophilicity of the C atom

bonded to this sulfur. Moreover, this transformation is driven

thermodynamically by the conversion of a dative S_thioether→ M

interaction in a strong covalent M−S_thiolatebond. The nature of

ancillary ligand also impacts the S−C bond cleavage. Goswami

et al. have studied the reaction of 2-methylthioaniline with

Pt(pap)Cl₂(pap = 2-(phenylazo)pyridine) and Pt(bpy)Cl₂(bpy

= 2,2′-bipyridine).⁷A rupture of the H₃C−S bond was

observed with the strong π-acidic coligand pap to aﬀord

Pt(pap) (HN∩S) (Scheme 1b), whereas the 2,2′-bipyridine

ancillary ligand failed to promote the demethylation of Pt(bpy)

(HN∩SCH₃). Recently, the activation of the C−S bond of the

PS₃-type tripodal tetradentate ligands, tris(2-tert-

INTRODUCTION

■

Since the ﬁrst report on a C−S bond activation, namely, the S-

dealkylkation of dimethylsulﬁde by PtCl₂dating from 1883,¹

this topic is still of interest. Driven primarily by interests in the

hydrodesulfurization of fossil fuels, various aspects of the C−S

activation process have been since extensively studied allowing

the development of new synthetic methods including

stoichiometric as well as catalytic reactions.²An interesting

class of compounds which undergo S-dealkylation reactions are

the group 10 E∩S_thioetherchelate complexes (E = N, P, As, S,

and Se).³In the latter, the C−S_thioethercleavage may be

homolytic and photochemically induced as in Ni⁰complexes of

2-(diphenylphosphino)thioanisole⁴and in bis(S-benzyl-1,2-

diphenyl-1,2-ethylenedithiolato)M^II(Ni, Pd, and Pt) complexes

(Scheme 1a).⁵Dealkylation reactions occur also heterolytically

by reaction with nucleophiles such as amines or chloride. Such

an alkyl transfers to nucleophiles have been characterized as

Menschutkin-type S_N2 reactions.⁶This substitution is favored

Received: January 16, 2017

A

DOI: 10.1021/acs.organomet.7b00039

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

Scheme 1. Examples of C−S Bond Activation of E∩S_thioetherLigands (E = N, P, S, and Se)

butylthiophenyl)phosphine and tris(2-isopropylthiophenyl-

phosphine), by group VIII metal compounds has been studied

in depth.⁸Reaction of tris(2-tert-butylthiophenyl)phosphine

with PdCl₂(PhCN)₂and PtCl₂(cod) at room temperature

resulted in the elimination of t-BuCl and formation of

MCl{P(2-SC₆H₄)(2-t-BuSC₆H₄)₂}(Scheme 1c). In contrast,

the C−S activation in the isopropyl derivatives required more

forcing conditions (333 K). The more facile elimination of t-

BuCl compared to that of i-PrCl was attributed to the higher

stability of the tert-butyl cation vs an isopropyl cation. In the

case of bis(methyl)selenosalen and bis(methyl)thiasalen

podands (Scheme 1d), the mechanistic pathway of the S_N2

substitution at the methyl group by chloride (stemming from

the inorganic starting metallic salts) has been theoretically

compared to that of a process implying an oxidative addition of

the chalcogen−C_alkylbond on Pd^IIor Pt^IIfollowed by reductive

elimination of alkyl chloride. DFT calculations indicated that

the experimentally observed chalcogen−C_(alkyl)bond activation

occurred more likely via S_N2 mechanism instead of a sequence

implying changes in oxidation state of the metal.⁹However, the

oxidative addition of a C−S bond on a Pt^IIhas been recently

observed in a PtMe₂(RN∩SCH₃(RN∩SCH₃= 1,2-

C₆H₄(CHNR)(SCH₃)), and the resulting dinuclear Pt^IV

thiolate complex has been characterized (Scheme 1e). Note

that in this case the activation involves an aryl−sulfur bond

instead of an alkyl−sulfur bond.¹⁰We reported previously that

thiolates trans-Pt^II(PEt₃)₂(η¹-L)₂and Pt^II(dppm)(η¹-L)₂(L =

4-methylthio-2-thioxo-1,3-dithiole-5-thiolate or Medmit; dppm

=1,1-bis(diphenylphosphanyl)methane) undergo S-demethyla-

tion by heating in acetonitrile or toluene solutions, yielding

dithiolene complexes Pt^II(PEt₃)₂(dmit) and Pt^II(dppm)(dmit)

along with MeL (Scheme 1f).¹¹A sequence implying S−CH₃

oxidative addition followed by reductive elimination driven by

the formation of a stable dithiolene chelate complex was

proposed, although no Pt^IVintermediate could be evidenced.

The intramolecular C−S oxidative addition on a low-valent

transition metal requires the proximity of the thioether

fragment to the latter.¹²With hemilabile L ligand (η¹-thiolate

or η²-thiolate/thioether coordination mode), the nature of the

metal and that of the coligands may aﬀect the CH₃S → M

dative interaction and consequently the reactivity of coordi-

nated L. Our experimental investigation has been now extended

to the reactivity of L coordinated on Pd^IIP₂(P₂= (PPh₃)₂,

(PEt₃)₂, dppm, and dppe). The experimental conditions of the

demethylation of L on palladium are discussed and compared

to those observed with the platinum analogues. To provide a

detailed picture of the S−CH₃bond activation mechanism in

these thiolate complexes, various possible mechanistic pathways

B

DOI: 10.1021/acs.organomet.7b00039

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

Scheme 2. Synthesis of Complexes 1−4

1

Table 1. Comparison of the M−P and M−S Bond Lengths (Å) as Well as the Thiomethyl H NMR Chemical Shift of

M(PR₃)(η¹-L)(η²-L) Complexes (M = Pd and Pt)

a

compound

M−P

M−S_thiolate

M−S_thioether

δ_CH(ppm)

3

2.2746(14)

2.3179(13)

2.3290(13)

2.3052(10)

2.3278(10)

2.2978(5)

2.3449(5)

2.3040(5)

2.3441(5)

2.2892(9)

2.3379(9)

2.2937(9)

2.3359(9)

2.3602(13)

2.66

2.68

2.61

Pd(PEt₃)(η¹-L)(η²-L) (1)

2.2593(11)

2.2605(9)

2.2663(5)

2.2459(10)

2.2504(10)

2.3418(10)

2.3489(5)

2.3355(5)

2.3303(9)

2.3217(9)

Pt(PEt₃)(η¹-L)(η²-L)¹¹

b

Pd(PPh₃)(η¹-L)(η²-L) (2)

b

Pt(PPh₃)(η¹-L)(η²-L)¹¹

2.60

a

b

At 295 K. Two independent molecules in the unit cell.

have been explored by density functional theory (DFT)

(B3LYP), and the inﬂuence of phosphines (labile or chelating)

has been assessed in order to examine the impact onto the

privileged pathway of this S−CH₃bond activation (nucleophilic

substitution/addition mechanisms or oxidative addition/reduc-

tive elimination sequence). The inﬂuence of the coligand has

also been evidenced experimentally with the formation of L−L

disulﬁde dimer obtained by reaction of L with MCl₂(M = Pt

and Pd). Note that the reactivity of the dmit dithiolene skeleton

has been extensively studied¹³due to the applications of these

systems in conducting materials.¹⁴

and consequently the substitution of one phosphine by a

thiomethyl group. In all these complexes, the hydrogen atoms

of the thiomethyl groups appear at room temperature as a

singlet. The resonances displayed by 1 (2.66 ppm) and 2 (2.61

ppm) are downﬁeld-shifted with respect to the signals observed

in 3 (2.25 ppm) and Pd(η²-dppm)(η¹-L)₂(2.40 ppm). Thus,

the dative S(thioether) → Pd interaction in 1 and 2 exchanges

swiftly in solution between the thiomethyl groups of the two

sulfur ligands present in the coordination sphere of the metal.

This behavior is identical to that observed in the analogues

platinum complexes which exhibit quite the same chemical

shifts for the hydrogen atoms of L (Table 1).¹⁶However,

whereas the nonequivalence of the thiomethyl groups was

RESULTS AND DISCUSSION

■

1

established below 240 K in the H NMR spectra of the Pt(II)

Reactivity of L toward Pd(PR₃)₂Cl₂and Pd(P∩P)Cl₂

Precursors. The reaction of cis-Pd(PR₃)₂Cl₂(PR₃= PEt₃

and PPh₃) with 2 equiv of CsL in reﬂuxing MeOH aﬀorded as

main products complexes Pd(PR₃)(η¹-L)(η²-L) (R = Et: 1; R =

Ph: 2) in moderate yields (∼60%). Using the procedure

reported by Olk and co-workers for the synthesis of

Pd(dppe)(η¹-L)₂(3) (dppe = 1,2-bis(diphenylphosphanyl)-

ethane),¹⁵chelate compound Pd(dppm)(η¹-L)₂(4) has been

synthesized in 75% yield as a red solid (Scheme 2).

complexes, the freezing of the dynamic exchange requires much

more lower temperature with the Pd(II) complexes. Using

CD₂Cl₂as solvent, the coalescence temperature was estimated

near the freezing point of CD₂Cl₂(173 K). This result is in

accordance with the well-established eﬀect of the central metal

ion on the rate of square planar substitutions.¹⁷The molecular

structures of Pd(Et₃)(η¹-L)(η²-L) and Pd(PPh₃)(η¹-L)(η²-L)

were determined by single crystal X-ray diﬀraction studies.

Pd(PEt₃)(η¹-L)(η²-L) crystallizes in the monoclinic system and

space group P2₁/c (Figure 1). Pd(PPh₃)(η¹-L)(η²-L) crystal-

These complexes are highly soluble in chlorinated solvents

but poorly soluble in MeOH. The ¹H NMR spectra of 1 and 2

reveal the presence of two sulfur-rich ligands for one phosphine

lizes in the triclinic system and space group P1, with two

̅

C

DOI: 10.1021/acs.organomet.7b00039

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

with the values of covalent radii (1.39 Å for Pd and 1.36 Å for

Pt).¹⁸

Lability of the Phosphine Group. The reactivity of L on

cis-[MP₂Cl₂] (M = Pd and Pt) is sensitive to the nature of the

metal in function of the lability of the phosphine. This latter is

higher in the case of palladium complexes. Whereas the

bis(thiolate) trans-Pt(PEt₃)₂(η¹-L)₂is the main product

obtained by reaction at room temperature of 2 equiv of L

with cis-Pt(PEt₃)₂Cl₂,¹¹the presence of Pd(PEt₃)₂(η¹-L)₂is

identiﬁed only as a minor product in the NMR spectra (δ =

16.6 ppm in the ³¹P{¹H} NMR and δ = 2.48 ppm for the SMe

1

groups in the H NMR) of the crude product obtained by

reaction of 2 equiv of L with cis-Pd(PEt₃)₂Cl₂. As previously

described, the major species is [Pd(PEt₃)(η¹-L)(η²-L)] (1).

The evolution of M(PEt₃)₂(η¹-L)₂toward M(PEt₃)(η¹-L)(η²-

L) seems not driven by the strength of the M−S_thiomethyl

Figure 1. ORTEP diagram (50% probability level) of 1. H atoms are

omitted for clarity. Selected bond lengths [Å] and angles [deg]: Pd−P

2.2746(14), Pd−S(5) 2.3179(13), Pd−S(1) 2.3602(13), Pd−S(10)

2.3290(13), P−Pd−S(5) 91.99(5), P−Pd−S(1) 178.20(5), P−Pd−

S(10) 85.71(4), S(5)−Pd−S(1) 89.60(4), S(5)−Pd−S(10) 177.61(5),

S(1)−Pd−S(10) 92.71(4).

1

interaction since the thiomethyl H NMR chemical shifts are

very close in the respective Pd and Pt complexes (Table 1).

With Pd(PPh₃)₂Cl₂as starting material, Pd(PPh₃)(η¹-L)(η²-L)

(2) is the single complex identiﬁed in the NMR spectra when

performing the reaction in reﬂuxing MeOH. Addition of 1

equiv of PPh₃to a toluene-d₈solution of 2 leads to broad peaks

centered at 32.6 ppm (2) and −5 ppm (PPh₃) in the ³¹P{¹H}

NMR spectrum, whereas no change is observed in the ¹H NMR

spectrum (except the increase of the intensity of the signals in

the aromatic range). At 363 K, the exchange between

coordinated and free PPh₃(according to the equilibrium

Pd(PPh₃)(η¹-L)(η²-L) + *PPh₃⇆ Pd(*PPh₃)(η¹-L)(η²-L) +

PPh₃) is suﬃciently fast so that only one “averaged” entity is

observed as a very broad hump. Thus, the dissociation of the

triphenylphosphine ligand of complex 2 is also facile. The

absence of formation of Pd(PPh₃)₂(η¹-L)₂by addition of PPh₃

on 2 is also shown by UV−visible experiments since 2 shows

transitions at 311 and 416 nm which are insensitive to the

addition of an excess of triphenylphosphine (Figure S1). At 363

K, a toluene-d₈solution of 2 exhibits very broad peaks in the ¹H

and ³¹P{¹H} NMR spectra (Figure 3). These observations may

indicate that the equilibrium Pd(PPh₃)(η¹-L)(η²-L) ⇆ Pd(η²-

L)₂+ PPh₃exists in solution at high temperature.

The higher phosphine lability in palladium complexes was

also noticed with the diphosphine dppm ligand. Whereas Pt(η²-

dppm)(η¹-L)₂is rigid in solution, an equilibrium between

Pd(η²-dppm)(η¹-L)₂and the coordination isomer Pd(η¹-

dppm)(η¹-L)(η²-L) is observed in solution (Scheme 2). In

the ³¹P{¹H} NMR spectrum at 295 K (Figure 4), the chelating

dppm ligand gives rise to a broad singlet at −38.3 ppm,^17a,19

and the dangling dppm is characterized by two broad signals

centered at 27.5 ppm (−Ph₂P−Pd) and −23.2 ppm (−CH₂−

PPh₂).^17a,20Note that the thiomethyl protons of Pd(η¹-

dppm)(η¹-L)(η²-L) appear at room temperature as a singlet

at 2.58 ppm suggesting that a dynamic η¹-L/η²-L exchange

occurs at room temperature. From the NMR spectra, the

Pd(η²-dppm)(η¹-L)₂/Pd(η¹-dppm)(η¹-L)(η²-L) ratio can be

estimated to be 9:1 at 293 K. In contrast, the ﬁve-membered

metallacycle Pd(η²-dppe) seems not to be hemilabile in

solution since only one sharp singlet at 59.6 ppm is observed

in the ³¹P{¹H} NMR spectrum corresponding to compound 3.

No trace of Pd(η¹-dppe)(η¹-L)(η²-L) is detected by NMR. It is

noteworthy that dissociation of one M−P bond in complexes 1,

2, and 4 is induced by an intramolecular interaction between a

thiomethyl group and the metal. For MP₂(dithiolene)

complexes, such phosphine dissociation is sometimes observed

but intermolecular M···S interactions generating dimetallic

independent complexes and one CHCl₃molecule in a general

position in the unit cell (Figure 2). In both complexes, the Pd

Figure 2. ORTEP diagram (50% probability level) of 2. H atoms and

solvent molecules are omitted for clarity. Only one of the two

independent molecules in the unit cell is depicted. Selected bond

lengths [Å] and angles [deg] {equivalent values measured for the

second independent molecule in the unit cell}: Pd(1)−P(1) 2.2605(5)

{2.2663(5)}, Pd(1)−S(1) 2.2978(5) {2.3040(5)}, Pd(1)−S(2)

2.3489(5) {2.3355(5)}, Pd(1)−S(6) 2.3449(5) {2.3441(5)}, P−

Pd(1)−S(1) 92.453(18) {91.348(18)}, P−Pd(1)−S(2) 175.945(19)

{175.033(19)}, P−Pd(1)−S(6) 86.615(19) {89.83(2)}, S(1)−Pd-

(1)−S(2) 90.821(18) {90.453(18)}, S(1)−Pd(1)−S(6) 178.486(19)

{177.891(19)}, S(2)−Pd(1)−S(6) 90.161(19) {88.230(19)}.

atom displays a slightly distorted square-planar coordination

with cis angles in the range of 85.65−92.77° and 86.62−92.45°

for 1 and 2, respectively. The Pd−S_thiolatedistances (2.317 and

2.329 Å for 1; 2.297, 2.304, 2.344, and 2.344 Å for 2) are

similar to those reported for other Pd−S_thiolatebonds,^8a,17but of

note is that they are shorter than those measured in the

Pd(dppe)(η¹-L)₂complex (2.385 and 2.375 Å).¹⁵This

observation is in agreement with the greater thermodynamic

trans inﬂuence of phosphine compared to that of thiolate

ligand. In the Pd(η²-L)-metallacycle, the Pd−S_thioetherbond

length is, as expected, signiﬁcantly longer than the Pd−S_thiolate

one given the diﬀerent charges of the donor atoms. Note

however that in compound 2 the Pd−S_thioetherdistance is

comparable to the Pd−S_thiolatebond length of the Pd(η¹-L)

fragment. The metric parameters of these palladium complexes

are comparable to those of the platinum analogues (Table 1)

with the M−L distances being slightly longer, in agreement

D

DOI: 10.1021/acs.organomet.7b00039

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

Figure 3. ³¹P{¹H} of Pd(PPh₃)(η¹-L)(η²-L) (2) in toluene-d₈at 298 K (top) and 363 K (bottom).

Figure 4. ³¹P{¹H} spectrum at 295 K of a CDCl₃solution of Pd(dppm)(L)₂(4).

Scheme 3. Demethylation Reactions Aﬀording Dithiolene Complexes

species [MP(μ-dithiolene)]₂appear to be the driving force in

this case.²¹

Me₂dmit however takes place when a solution of 4 in DMF is

heated at 403 K for 6 h. In addition, for comparison we reacted

Pd(PR₃)₂Cl₂and Pt(PR₃)₂Cl₂(R = PEt₃and PPh₃) with 2

equiv of CsL in acetonitrile at 345 K for 17 h. Analysis of the

reaction mixtures revealed a M(PR₃)₂(dmit)/M(PR₃)(η¹-L)(η²-

L) ratio of 100/0 for M = Pt and R = Ph, 80/20 for M = Pt and

R = Et, 40/60 for M= Pd and R = Et, and 5/95 for M = Pd and

R = PPh₃. It should be stressed that in both of these Pd and Pt

complexes the thermally induced demethylation occurs without

the presence of an external nucleophile, contrasting with the

reactivity pattern depicted in Scheme 1b,d.

Demethylation Reaction. The L ligand coordinated on Pd

is also prone to demethylation (Scheme 3). Indeed, NMR

spectroscopy revealed a partial evolution (13% yield) of

bis(thiolate) compound 3 to the dithiolene complex Pd(η²-

dppe)(dmit) (δ = 56.7 ppm in the ³¹P{¹H} NMR) and

1

concomitant formation of Me₂dmit (δ = 2.49 ppm in the H

NMR) after heating in acetonitrile at 345 K for 6 h. Under the

same conditions, a demethylation rate of 82% has been

determined for the Pt derivative, whereas dppm-ligated Pd

complex 4 was reluctant to demethylate. The formation of

Pd(η²-dppm)(dmit) (δ = −39.0 ppm in the ³¹P NMR) and

Reactivity toward MCl₂(M = Pd and Pt). Two

equivalents of CsL were also reacted with MCl₂(M = Pd

E

DOI: 10.1021/acs.organomet.7b00039

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

and Pt) in reﬂuxing ethanol. A priori, this reaction might lead to

M(η²-L)₂species identiﬁed as thermodynamically stable

molecule in the theoretical study (see below). After separation

of a black powder, the workup of the ﬁltrate aﬀorded the L−L

dimer as an orange microcrystalline solid in low yield (Scheme

4). This disulﬁde, soluble in halogenated solvents, is stable at

Due to the cost and the stoichiometric need of PtCl₂or

PdCl₂in the reaction described above, we investigated an

alternative method to synthesize dimer L−L. The oxidative

coupling of thiols to disulﬁde has been extensively studied, and

besides protocols using iodine or air as oxidizing agents, a wide

range of methods have been developed.²⁶An eﬃcient and mild

procedure involving 1 equiv of hydrogen peroxide and a

catalytic amount of iodide has been recently reported.²⁷Using

this method, dimer L−L was synthesized in 40% yield starting

with the CsL thiolate dissolved in water. L−L reacts in CDCl₃

by oxidative addition across 1 equiv of the zero-valent

Pt(PPh₃)₂(PhCCPh) complex to aﬀording as product

[Pt(PPh₃)(η¹-L)(η²-L)] identiﬁed on the basis of its NMR

spectroscopic data.¹¹This reactivity is similar to that observed

between tetrathiocines and zero-valent group 10 transition

metal complexes yielding dithiolate complexes.²⁸

Scheme 4. Synthesis of L−L

ambient temperature. Spectroscopic characterizations reveal

1

only one singlet at 2.55 ppm in the H NMR spectrum and a

strong absorption at 1061 cm⁻¹in the IR spectrum, associated

with the presence of a CS bond. The Raman spectrum

recorded in CH₂Cl₂solution shows two bands in the S−S

stretching region situated at 460 and 508 cm⁻¹.²²The presence

of two bands may result from the coexistence of two rotamers

in solution diﬀering from the relative conformation, cis or trans,

of the C_αatoms linked by a disulﬁde bridge.

We also examined the reactivity of 2 equiv of CsL toward

M^II(cod)Cl₂(M = Pd and Pt, cod =1,5-cyclooctadiene) in

reﬂuxing ethanol. In both cases, only insoluble products have

been isolated. The presence of the dmit core was evidenced in

the IR spectra by the signature of the CS bond near 1050

cm⁻¹. No cod could be detected in the IR spectra of the solids.

Elemental analyses indicated a S/C ratio close to 3 in

accordance with the presence of L in this insoluble material.

A similar result was obtained after reaction of CsL with

Pd(SEt)₂Cl₂. Due to the very low solubility of the isolated

products, it was not possible to obtain information from NMR

spectroscopy, so at this stage, it would be speculative to

conclude on the structure of these compounds. Nevertheless,

these experiences point out the inﬂuence of the ancillary ligand

on the reactivity of L toward Pd^IIand Pt^II.

Theoretical Study. To gain more insight into the

mechanism of the C−S bond activation and the inﬂuence of

the phosphine ligands on the nature of the privileged pathway,

DFT calculations were carried out at the B3LYP/SMD-

(CH₃CN)/SDD+f(M),6-31G**(other atoms)//B3LYP/SDD

+f(M),6-31G** (other atoms) level of theory (see the

Computational Details section). The solvent eﬀect (CH₃CN)

was taken into account by the SMD solvation model. We

examined two speciﬁc cases involving either a labile phosphine,

PEt₃or PPh₃, for which the intermediates M(PR₃)(η¹-L)(η²-L)

have been experimentally isolated and characterized, or a

chelate phosphine (dppe), where no occurrence of M(η¹-

dppe)(η¹-L)(η²-L) complex was observed by NMR spectros-

copy.

The molecular structure of L−L was determined by X-ray

diﬀraction. The molecule is symmetric with a C₂axis

perpendicular to the S−S bond which inverts the two L units

(Figure 5). The S−S bond distance of 2.0929(11) Å is slightly

2−

shorter than those reported for the dianions C₆S₁₀featuring

two dmit units linked by one S−S bond (2.135(4) Å) and

2−

C₁₂S₁₆in which two 1,3-dithiole-2 thione rings are also

connected by one disulﬁde bond (2.157(12) Å).²³The S−S

separation in L−L is similar to the value of 2.0757(9) Å

observed in the rigid molecular C₆S₁₀compound whose

structure consists in two dmit units linked by two disulﬁde

bonds.²⁴The torsion angle of 101.3° measured by the dihedral

angle of the C−S−S−C linkage is usual. The stabilization

arising from dihedral angles near 90° has been explained by the

overlap of one σ*_S−Corbital and the 3p lone pair located on the

other sulfur atom which is maximized in such a conformation.²⁵

In contrast to the compound described here, the dianion

2−

C₆S₁₀does not follow this general rule since an unusually

acute dihedral angle of 51.9° was reported.²³In this case, the

twist was not favorable for hyperconjugation but allows

intramolecular π−π interactions between the 1,3-dithiole-2-

thione cores. For L−L, the dihedral angle of 101.3° is not

consistent with intramolecular π stacking, but the packing

diagram in the solid state exhibits intermolecular π−π

interactions with separation of the 1,3-dithiole-2-thione rings

of 3.49 Å (Figure 5).

For labile phosphines PEt₃and PPh₃, the starting point of the

theoretical study was complex M(PR₃)₂(η¹-L)₂, identiﬁed as

minor product when Pd(PEt₃)₂Cl₂was reacted with two equiv

of L and major product when using Pt(PEt₃)₂Cl₂. The latter

Figure 5. ORTEP diagram (50% probability level) of L−L (left) and view of the intermolecular π−π interactions in the solid state (right). Symmetry

transformations to generate equivalent atoms: #, −x + 1, y, −z + 1/2.

F

DOI: 10.1021/acs.organomet.7b00039

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

a

Scheme 5. Plausible Mechanisms for C−S Bond Activation in trans-M^II(PEt₃)₂(η¹-L)₂Complex

a

Including oxidative addition/reductive elimination sequence, nucleophilic substitution S_N2, and nucleophilic addition A_N. Pathways A correspond

to direct pathways, and pathways B to pathways involve phosphine predissociation.

can undergo direct demethylation or demethylation after

predissociation of one phosphine ligand through complex

M(PR₃)(η¹-L)(η²-L). From model complex Pd^II(PEt₃)₂(η¹-L)₂

and its analogue with aryl groups Pd^II(PPh₃)₂(η¹-L)₂, two

isomers have been localized on the potential energy surface

(PES), according to the position of the phosphine groups (cis

or trans). trans-Pd(PR₃)₂(η¹-L)₂was found to be around 5.6

kcal/mol (PEt₃)/9.2 kcal/mol (PPh₃) more stable than cis-

Pd(PR₃)₂(η¹-L)₂. Similar results have been obtained for the

platinum complexes, with an energy diﬀerence of around 4.1

kcal/mol (PEt₃)/9.3 kcal/mol (PPh₃). The privileged trans

conﬁguration is in agreement with experimental observations

since the unique formation of trans-[Pt^II(PR₃)₂(η¹-L)₂] has

been previously ascertained by the magnitude of NMR J_PtP

coupling constants and X-ray structures.¹¹Moreover, a survey

of structurally characterized PtP₂S₂complexes has shown that

the trans conﬁguration is the more common arrangement with

monodentate ligands.²⁹These theoretical data suggest a poor

probability of having the cis isomer in the mixture, according to

the Maxwell−Boltzmann distribution. Consequently, we

focused calculations on the reactivity of trans-M^II(PR₃)₂(η¹-

L)₂complexes and analyzed the diﬀerent plausible mechanisms

for the C−S bond activation.

Case of Labile Phosphine PEt₃. To rationalize the formation

of square planar dithiolene complex M^II(PEt₃)₂(dmit), we

supposed several mechanistic scenarios (Scheme 5): (i)

G

DOI: 10.1021/acs.organomet.7b00039

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

Figure 6. Energy proﬁles (ΔG^{CH CN}in kcal/mol) for direct mechanisms (pathways A) and phosphine-predissociation mechanisms (pathways B) for

C−S bond activation in trans-Pd^II(PEt₃)₂(η¹-L)₂complex, calculated at the B3LYP/SMD(CH₃CN)/SDD+f(M),6-31G** (other atoms)//B3LYP/

SDD+f(M),6-31G**(other atoms) level of theory. The values in parentheses correspond to the results for the corresponding trans-Pt^II(PEt₃)₂(η¹-

L)₂complex.³⁰

3

Pathways A: Direct mechanisms from trans-M(PEt₃)₂(η¹-L)₂

lone pair (n_S2) and the σ*_S3−CHorbital. At the transition state,

3

complex with the two alkylphosphines coordinated to the metal

the 5-partially bonded carbon adopts a trigonal bipyramidal

center throughout the entire process. Three diﬀerent pathways

geometry with S2 and S3 atoms in axial position. (path A-A_N)

have been considered: (path A-OA/RE) This is a two-step

In this pathway, the C−S bond activation proceeds through a

transition state involving a CH₃⁺carbocation equidistant to the

S_thiolateand S_thioetheratoms. The sulfur lone pair (p character) of

the thiolate S₂atom and the 2pπ(C) orbital of the carbocation

are correctly oriented to interact, favoring a nucleophilic attack.

For simplicity, this mechanism will be named hereafter

nucleophilic addition (A_N). (ii) Pathways B: Mechanisms

involving one phosphine predissociation from trans-M-

(PEt₃)₂(η¹-L)₂complex. The intramolecular substitution of

one PEt₃ligand by the thiomethyl group SMe aﬀords the

square planar species M^II(PEt₃)(η¹-L)(η²-L) (1^B). As pre-

viously described, several pathways can take place from this

intermediate: (path B-OA/RE) This is a multistep process

implying an oxidative addition/reductive elimination sequence

followed by phosphine recoordination. The penta-coordinated

process with oxidative addition of the S−C_Mebond to M(II)

leading to M^IV(PEt₃)₂(η¹-L)(dmit)Me intermediate (1a^A),

which can convert by reductive elimination of Me₂dmit to

M^II(PEt₃)₂(dmit). This intermediate adopts an octahedral

geometry with the two phosphine ligands at the axial position

of the square plane formed by methyl group and the three

sulfur atoms and may be related to the octahedral

Pt^IV(PMe₃)₂(dithiolene)₂complexes, recently described by

Donahue et al.^21a(path A-S_N2) This pathway corresponds to

an intramolecular concerted S_N2 reaction with a thiolate sulfur

atom acting as the nucleophile and M^II(PEt₃)₂(dmit) complex

being the leaving group. The nucleophile approaches the

thiomethyl group at an angle of 180° with respect to the C_CH

−

3

S bond, thus providing the optimal overlap between the sulfur

H

DOI: 10.1021/acs.organomet.7b00039

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

Figure 7. Main minima and TS involved in the phosphine-predissociation mechanisms: oxidative addition/reductive elimination sequence (path B-

OA/RE), nucleophilic addition (path-B-A_N), and S_N2 reaction (path B-S_N2) from trans-Pd^II(PEt₃)₂(η¹-L)₂complex.

oxidative addition product M^IV(PEt₃)(η¹-L)(dmit)Me (1a^B)

can be described as a square pyramidal with the methyl group

perpendicular to the square plane occupied by the P and the

three S atoms. This latter can then undergo reductive

elimination of the C−S_thioetherbond and ﬁnally form

M^II(PEt₃)₂(dmit) after recoordination of PEt₃. (paths B-S_N2

and B-A_N) These correspond to the S_N2 and nucleophilic

addition (A_N) mechanisms from M^II(PEt₃)(η¹-L)(η²-L) spe-

cies.

For the sake of clarity, only the most plausible mechanisms

will be described in detail hereafter. According to DFT

calculations, all energy proﬁles involving direct pathways

(pathways A, Scheme 5) can be ruled out because the

transition states (TS’s) lie 47.6 to 64.9 above the reactant

trans-Pd^II(PEt₃)₂(η¹-L)₂(Figure 6).³⁰In agreement with the

experimental observations, theoretical results (Figures 6 and 7)

point out that the formation of Pd^II(PEt₃)(η¹-L)(η²-L) (1^B) is

kinetically and thermodynamically favored [ΔG^#(TS^disso1^B) =

I

DOI: 10.1021/acs.organomet.7b00039

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

Figure 8. Energy proﬁles (ΔG^{CH CN}in kcal/mol) for C−S bond activation from trans-Pd^II(PPh₃)₂(η¹-L)₂calculated at the B3LYP/SMD(CH₃CN)/

SDD+f(M),6-31G**(other atoms)//B3LYP/SDD+f(M),6-31G**(other atoms) level of theory and involving the phosphine-predissociation

pathways B and C: (a) oxidative addition/reductive elimination sequence through 2^B, in red (B-OA/RE), (b) nucleophilic addition through 2^B, in

black (B-A^N), (c) S_N2 pathway (B-S_N2) through 2^B, in blue, and (d) oxidative addition/reductive elimination sequence through 2^C, in green (C-

OA/RE). The values in parentheses correspond to the results for the analogue trans-Pt^II(PPh₃)₂(η¹-L)₂complex.³⁴

3

15.2 kcal/mol and ΔG(trans-Pd^II(PEt₃)₂(η¹-L)₂→1^B) = −10.5

kcal/mol].³¹This step proceeds through a concerted

dissociative mechanism with a late TS featuring dissymmetric

P−Pd distances (2.316 and 3.198 Å), short Pd−S1/Pd−S2

bond lengths (2.385/2.384 Å), and a quasi-formed Pd−S3

bond (2.583 Å), only elongated around by 0.2 Å compared to

Pd−S1/Pd−S2 ones.

In light of these theoretical data, we have explored more

speciﬁcally the three dissociative mechanisms (B-OA/RE, B-

S_N2, and B-A_N) involving tetra-coordinated square planar

intermediate 1^B, which is stabilized by a strong coordination of

the thiomethyl group to the metal center [Pd−S3(Me): 2.425

Å, close to Pd−S1/PdS2: 2.369−2.390 Å, NBO analysis:

LP1(S3) → Pd: 8.6 kcal/mol and LP2(S3) → Pd: 96.2 kcal/

mol, Table S1]. DFT calculations (Figure 6) suggested that the

intramolecular S_N2-like reaction (B-S_N2, Scheme 5) is the less

feasible process. The TS lies at 43.3 kcal/mol (TS^SN21^B) above

the initial reactant Pd^II(PEt₃)₂(η¹-L)₂. The nucleophilic

addition (B-A_N) takes place through an activation barrier

(TS^AN1^B) lower in energy by around 6.2 kcal/mol and is

therefore the most privileged pathway. It can also compete with

oxidative addition/reductive elimination sequence (B-OA/RE

mechanism) since the rate-determining reductive elimination

step appears at almost the same energy as that of A_Nprocess

(TS^RE1^B, 2.5 kcal/mol higher in energy than TS^AN1^B, Figure 6).

However, in the OA/RE sequence, the oxidative addition step,

whose activation barrier is lower in energy than reductive

elimination by around 10 kcal/mol, can be reversible (penta-

coordinate intermediate 1a^Bis thermodynamically less stable

than 1^Bby around 27.3 kcal/mol). Consequently, the feasibility

of demethylation process through this sequence may be

impacted by promoting a competition between reductive

elimination of Me₂dmit at palladium (thermodynamic control)

leading to the dithiolene complex Pd^II(PEt₃)₂(dmit) or a

backward reaction giving rise to the regeneration of 1^B(kinetic

control). The activation barriers of the reductive elimination

step and the nucleophilic addition (A_N) mechanism are

signiﬁcant but accessible under the speciﬁc experimental

conditions (heating at 348 K during 17 h) and may explain

why the demethylation process is not eﬀective.

Analysis of the associated transition states (Figure 7)

revealed that in the nucleophilic addition process B-A_N, the

J

DOI: 10.1021/acs.organomet.7b00039

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

a

Scheme 6. Other Mechanism for C−S Bond Activation (Pathway C) in M^II(PPh₃)₂(η¹-L)₂Complex

a

Mechanism involves oxidative addition/reductive elimination sequence and going through M^II(η²-L)₂(complex 2^C).

methyl group is quasi-equidistant to S2 and S3 atoms

(C_CH...S2: 2.682 Å versus C_CH...S3: 2.794 Å), with a quasi-

bulky phosphine groups (cone angle of 132° for PEt₃and

145° for PPh₃).³²We turn our attention to predissociation

mechanisms since the intermediate Pd^II(PPh₃)(η¹-L)(η²-L) has

also been evidenced by NMR and X-rays. As previously noticed

for Pd^II(PEt₃)(η¹-L)(η²-L), NBO analysis allowed us to identify

a strong n_SMe→ Pd interaction in the PPh₃complex [Pd−S3:

2.417 Å, NBO analysis: LP1(S3) → Pd: 9.2 kcal/mol and

LP2(S3) → Pd: 97.0 kcal/mol, Table S1]. This interaction,

which is slightly more important than that with PEt₃as revealed

by NBO analysis and the shortening of the Pd−S3 distance

(theoretically and experimentally), is consistent with the

electronic eﬀect of the PPh₃ligand (weaker σ-donor eﬀect

than PEt₃which reduces charge on metal). The PPh₃-

predissociation step is even more kinetically and thermody-

namically favored than that with PEt₃phosphine

[ΔG^#(TS^disso2^B): 8.9 kcal/mol and ΔG(trans-Pd^II(PPh₃)₂(η¹-

L)₂→2^B): −19.4 kcal/mol].³³This highly exergonic step is in

line with the nondetection of [Pd^II(PPh₃)₂(η¹-L)₂] upon

reaction of Pd^II(PPh₃)₂Cl₂with CsL and the unique

identiﬁcation of complex Pd^II(PPh₃)(η¹-L)(η²-L) in the NMR

spectrum at room temperature. Comparison of the energy

proﬁles depicted in Figures 6 and 8 has shown that all the

diﬀerent intermediates (2^Band 2a^B) and transition states are

3

planar environment of the carbon center (ΣC: 351.4°). The

oxidative addition proceeds through a late 3-center transition

state involving Pd, S3, and C_CHatoms. The main geometric

3

features associated with TS^OA1^Bare a partially formed C_CH−Pd

3

bond (2.367 Å versus 2.115 Å in 1a^B), an elongated C_CH−S3

3

bond (2.498 Å versus 1.842 A in 1^B), and a classical Pd−S3

distance (2.367 Å). Likewise, reductive elimination transition

state TS^RE1^Bdisplays a 3-center structure (Pd, S2, and C_CH

3

atoms) with quasi-similar Pd−S distance (Pd−S2: 2.455 Å), a

shorter Me−S₂bond (2.316 Å) than the Me−S₃one in

TS^OA1^B, and a more slightly elongated C_CH−Pd bond (2.457

3

Å). Intrinsic reaction coordinate (IRC) calculations conﬁrmed

that these two TS are both connected to the same penta-

coordinated intermediate (1a^B), where the methyl group is in

apical position.

Case of the Labile Phosphine PPh₃. To evaluate the

inﬂuence of monodentate labile phosphines, we have explored

the activation of S−CH₃bond from the hypothetical

Pd^II(PPh₃)₂(η¹-L)₂complex, which involves slightly more

K

DOI: 10.1021/acs.organomet.7b00039

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

Figure 9. Main minima and TS involved in the oxidative addition/reductive elimination sequence for mechanism C, going through Pd^II(η²-L)₂

complex (2^C).

just downhill in energy by around 10 kcal/mol compared with

those of the PEt₃ligand. Thus, the activation barrier of the

oxidative addition step is slightly decreased compared to that of

the alkylphosphine ligand (∼4 kcal/mol). The reductive

elimination and nucleophilic substitution (B-A_N) steps are

calculated to be similar to that with PEt₃, suggesting that the

steric and electronic properties of the phosphine do not really

aﬀect this demethylation process. Likewise, the detailed

examination of the optimized geometries of TS^OA2^B, TS^RE2^B,

and TS^AN2^Bevidences no noticeable diﬀerence with their PEt₃

analogues (see Figure S3). These preliminary results emphasize

that a competition between oxidative addition/reductive

elimination sequence (B-OA/RE) and nucleophilic addition

(B-A_N) may also happen (diﬀerence around 0.5 kcal/mol

between the two TS involved in RE and A_Nreactions), as

previously observed with PEt₃ligand.

imental conditions (363 K, Figure 3) and can aﬀord easily

Pd^II(η²-L)₂complex (ΔG = −1.0 kcal/mol; ΔG^#: 18.7 kcal/

mol, Figure S4),³⁵in line with the experimental NMR

observations. The ability of demethylation from this

intermediate (2^C) has been studied computationally (Figure

8) to compare with the mechanism (B-OA/RE) in Scheme 5.

The activation barrier of the oxidative addition step was

calculated relatively close with or without dissociation of the

phosphine (ΔG^#= 34.4 kcal/mol vs 34.1 kcal/mol from 2^C). In

contrast, a higher diﬀerence was found for the reductive

elimination step, where a decrease of the activation barrier

around 7 kcal/mol was observed in the case of the mechanism

involving Pd^II(η²-L).³⁶Examination of the structures of the two

TS for pathway C indicated (i) the concerted formation of Pd−

C_CHbond (2.407 Å) and cleavage of S3−C_CHbond (2.471 Å)

3

in TS^OA2^Cas well as (ii) cleavage of Pd−C_CHbond (2.465 Å)

3

In light of the experimental data for Pd^II(PPh₃)(η¹-L)(η²-L)

complex (dissociation of PPh₃at high temperature and

exchange between coordinated and free PPh₃), we have

considered another pathway involving decoordination of the

second phosphine group from Pd^II(PPh₃)(η¹-L)(η²-L) and

going through Pd^II(η²-L)₂intermediate (2^C, Scheme 6). The

Pd^II(η²-L)₂complex (2^C) has been found as minimum on the

potential energy surface (PES). The dissociation of PPh₃from

Pd^II(PPh₃)(η¹-L)(η²-L) is straightforward under the exper-

and formation of S2−C_CHbond (2.365 Å) in TS^RE2^C.

3

Comparison of the geometrical features of these transition

states with those associated with pathway B-OA/RE revealed

weak modiﬁcations in the oxidative addition step (See Figures

S3 and 9, similar distances and ΣPd: 359.1° in TS^OA2^Cversus

ΣPd: 361.6° in TS^OA2^B), in agreement with similar activation

barriers. In contrast, the changes are more important in the

reductive elimination step. More speciﬁcally, the Pd center is

more distorted in TS^RE2^Bthan in TS^RE2^C(ΣPd: 367.4° in

L

DOI: 10.1021/acs.organomet.7b00039

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

Scheme 7. Privileged Pathways for C−S Bond Activation from M(PR₃)₂(η¹-L)₂

a

Scheme 8. Plausible Mechanisms from C−S bond activation in Pd^II(η²-dppe)(η¹-L)₂complex

a

Including oxidative addition/reductive elimination sequence, nucleophilic substitution S_N2, and nucleophilic addition. Pathways A correspond to

direct pathways, and pathways B are those involving phosphine predissociation.

TS^RE2^Bversus 359.6° for TS^RE2^C) because of steric constraint,

which can be one of the reasons of the energy diﬀerence around

9 kcal/mol.

Intriguingly, the demethylation reaction was predicted to

preferentially proceed through Pd^II(η²-L)₂intermediate (path-

way C) rather than directly going through Pd^II(PPh₃)(η¹-L)(η²-

L) complex (mechanisms B). This is pointed out by the

energetic span δE, which has been calculated around 39.3 kcal/

mol for mechanism C versus for 47.6 for mechanism B-OA/RE

and 47.1 for mechanism B-A_N.³⁷As previously described for the

OA/RE sequence in mechanism B with PEt₃ligand, DFT

calculations indicated that the oxidative step assumes an

important role on the feasibility of the demethylation process.

Indeed, penta-coordinated intermediate 2a^C(methyl group in

apical position and the Pd atom surrounded by four sulfur

atoms in a square planar environment) is not thermodynami-

M

DOI: 10.1021/acs.organomet.7b00039

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

Figure 10. Energy proﬁles (ΔG^{CH CN}in kcal/mol) for the direct pathways (nucleophilic substitution S_N2, nucleophilic addition A_N, and oxidative

addition/reductive elimination) for C−S bond activation from Pd^II(η²-dppe)(η¹-L)₂calculated at the B3LYP/SMD(CH₃CN)/SDD+f(M),6-

31G**(other atoms)//B3LYP/SDD+f(M), 6-31G**(other atoms) level of theory. The values in parentheses are those obtained for the analogue Pt

species.

3

cally favored. This results in a competition between two

reductive elimination pathways. The ﬁrst, corresponding to the

reverse oxidative addition step, gives Pd^II(η²-L)₂, which can

then be converted into Pd^II(PPh₃)(η¹-L)(η²-L) complex after

phosphine coordination, and the second one leads to

demethylation product Pd^II(PPh₃)₂(dmit). The weak thermo-

dynamic preference for this dithiolene complex (ΔG ≈ −4

kcal/mol from 2^Cor −5.1 kcal/mol from 2^B) and the readily

accessible activation barrier to once again regenerate the kinetic

product Pd^II(PPh₃)(η¹-L)(η²-L) from 2a^Cperfectly agree with

an ineﬃcient demethylation reaction and are consistent with

the experimental observations, i.e., a ratio of Pd^II(PPh₃)(η¹-

L)(η²-L)/Pd^II(PPh₃)₂(dmit) close to 20 upon heating the

reactants Pd(PPh₃)₂Cl₂and CsL at 348 K during 17 h.

Similarly, the high endergonicity of the oxidative addition

step for the PEt₃ligand tends to show that Pd^IIPEt₃(η¹-L)(η²-

L) species is the kinetically controlled product of the

demethylation process (ΔG^#= 11.7 kcal/mol for the reverse

oxidative addition process versus ΔG^#= 22.8 kcal/mol for the

formation of Pd^II(PEt₃)₂(dmit) by RE from 1a^B). On the

contrary, Pd^II(PEt₃)₂(dmit) seems to be the thermodynamically

controlled derivative and can originate from nucleophilic

addition reaction from 1^Band/or reductive elimination from

1a^B. Finally, the improvement of the demethylation process

with alkyl group on phosphorus (40/60 for PEt₃versus 5/95

for PPh₃) can be correlated to higher thermodynamic

stabilization of the dithiolene complex bearing a PEt₃ligand

compared to PPh₃(ΔG^dithiolene= −9.9 kcal/mol from 1^Bversus

ΔG^dithiolene= −4.7 kcal/mol from 2^B).

Thus, all these results highlight that in heteroleptic

monophosphine/L palladium complexes the C−S bond

activation can occur via two competitive pathways: a

nucleophilic addition reaction and/or an oxidative/reductive

elimination sequence (Scheme 7). However, DFT calculations

make a distinction between PEt₃and PPh₃derivatives. With

PEt₃, the two reactions take place simultaneously and directly

from Pd^II(PEt₃)(η¹-L)(η²-L), whereas the dissociation of the

triphenylphosphine from Pd^II(PPh₃)(η¹-L)(η²-L) implies that

oxidative addition/reductive elimination mechanism is favored

and occurs preferentially from Pd^II(η²-L)₂complex. The

thermodynamic stabilization of the ﬁnal products

[Pd^II(PR₃)₂(dmit) + Me₂dmit] compared to intermediate

N

DOI: 10.1021/acs.organomet.7b00039

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

Figure 11. Main geometrical parameters of TS^SN23^A(distances in Å and bond angles in deg) and main molecular orbitals (cutoﬀ: 0.04) involved in

direct S_N2 mechanism from Pd^II(η²-dppe)(η¹-L)₂.

Pd^II(PR₃)(η¹-L)(η²-L) seems to be the driving force for the

eﬃciency of the process and suggests that the latter is under

thermodynamic control (better yield with PEt₃). However, the

relatively high activation barriers of RE and A_Nsteps are

responsible for the poor yield of demethylation (<50%).

Case of the Chelate Phosphine dppe. To extend this work,

the demethylation process has also been investigated with the

strongly chelating diphosphine dppe, for which a pathway

implying a predissociation of a phosphine arm appears a priori

quite unlikely especially in the case of dppe. Nevertheless, the

opening of a dppe chelate ring has been already proposed in

mechanisms involving dihydrogen complex trans-[FeH(H₂)-

(dppe)₂]⁺.³⁸Therefore, we have theoretically considered both

direct (pathways A) and indirect pathways (pathways B) from

[Pd(η²-dppe)(η¹-L)₂] (Scheme 8).

The substitution of one phosphine arm by the thiomethyl

group requires a slightly less accessible barrier than that for the

monodentate phosphines (Figure S2) [ΔG^#(TS^disso3^B) = 19.7

kcal/mol versus 9−15 kcal/mol]. As expected, greater diﬀer-

ence appears in the thermodynamic feasibility of this

phosphine/thiomethyl exchange. While it was exergonic for

the labile phosphines [PEt: ΔG(TS^disso1^B) = −10.5 kcal/mol;

PPh₃: ΔG(TS^disso2^B) = −19.4 kcal/mol], it is endergonic for

the chelate dppe phosphine [ΔG(TS^disso3^B) = 8.3 kcal/mol],

showing that the feasibility of this largely depends on the nature

of the ligand and can be excluded with dppe.³⁹Looking at the

bonding situation in intermediate 3^B, no signiﬁcant diﬀerence is

observed (see Table S1) compared to those of the

corresponding PEt₃and PPh₃analogues, indicating that the

endergonicity of this step seems rather to be correlated to the

higher stability of the chelate Pd^II(η²-dppe)(η¹-L)₂complex. As

might be expected, the oxidative addition step from 3^Bis uphill

by around 10−20 kcal/mol compared to those calculated for

Pd^II(PR₃)₂(η¹-L)₂complexes (ΔG^#(TS^OA3^B) = 44.7 kcal/mol,

Figure S5), and the reductive elimination could not take place

because the activation barrier is unreachable (ΔG^#(TS^RE3^B) =

53.5 kcal/mol). The nucleophilic addition (A_N) and sub-

stitution (S_N2) are also energetically disfavored (ΔG^#> 55

kcal/mol), ruling out any mechanisms involving dissociation of

one phosphine arm (Figures S6 and S7).

Activation barriers for the direct oxidative addition/reductive

elimination sequence and A_Npathway were also found very

high in energy (Figure 10) [ΔG^#= 52−62 kcal/mol]. By

contrast, the transition state lies around 38.4 kcal/mol over

Pd^II(η²-dppe)(η¹-L)₂for the intramolecular S_N2 mechanism.

The barrier presents the same magnitude as that for the

oxidative addition step in the predissociative pathway involving

labile phosphines and can be therefore reached under the

experimental conditions (∼348 K in acetonitrile). It is

noteworthy that S_N2 substitution at methyl group by chloride

has also been proposed to account for the cleavage of Se−

C(alkyl) bond in bis(alkyl)selenosalen Pd(II) and Pt(II)

complexes (Scheme 1d),⁹but in this latter example, the

nucleophile (Cl⁻) is an external species.

The main geometric features of transition state TS^SN23^A

reveal simultaneous formation of a C_CH···S2_thiotherbond (2.236

3

Å) and concomitant displacement of the Me group from S3

sulfur (2.533 Å), with a S2···Me···S3 bond angle being around

160.09°. Plots of the MOs involved in this reaction reinforces

the intramolecular S_N2-like process (Figure 11), with HOMO−

1 and LUMO+4 of TS corresponding to the S2 lone pair of the

ligand L in interaction with the σ*_S2Corbital of the other ligand.

Examination of the geometric features of the initial reactants

Pd^II(η²-dppe)(η¹-L)₂, trans-Pd^II(PEt₃)₂(η¹-L)₂, and their corre-

sponding direct S_N2 transition states (TS^SN23^Aand TS^SN21^A)

reveals strong distortion of the geometry of the TS with PEt₃.

In TS^SN21^A, the metal presents a distorted square planar

geometry compared to trans-Pd^II(PEt₃)₂(η¹-L)₂(ΣPd: 374.1°

versus ΣPd: 360.4°, see Figure S7) with a strong decrease of the

S1PdS2 bond angle around 30−40° (173.6° in initial reactant

and 137.21° in TS) and a decrease of P1MP2 around 13°

(172.6° in initial reactant and 159.7° in TS). On the contrary,

with dppe ligand, these main geometrical parameters remain

quasi-similar in the two species (initial reactant ΣPd: 359.9°,

S1MS2: 94.4°, and P1MP2: 86.6° versus TS, ΣPd: 360.0°,

S1MS2: 92.1°, and P1MP2: 86.4°). These data conﬁrm that the

large diﬀerence between the direct S_N2 activation barriers of the

O

DOI: 10.1021/acs.organomet.7b00039

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

two systems originates from the bent PPdP (chelate

phosphine) versus linear (trans-PEt₃phosphines) geometry,

pointing out that the chelate eﬀect preorganizes the system for

the S_N2 pathway, becoming thus the most favorable. This is

conﬁrmed by the localization of the S_N2 transition states from

cis-Pd^II(PR₃)₂(η¹-L)₂complexes (R= Et and Ph; see Figures

S10 and S11 for the energy proﬁles and Figures S8 and S9 for

the geometrical features), which present activation barriers

(ΔG^#≈ 30 kcal/mol) close to that computed from Pd^II(η²-

dppe)(η¹-L), where no strong structural deformation was

evidenced.

the nature of the ancillary ligands. With the chelating dppe

phosphine, an intramolecular S_N2 process in which a S_thiolate

acts as nucleophile appears as most feasible pathway due to the

cis position of η¹-L ligands imposed by the stable Pd(η²-dppe)

metallacycle. With PPh₃ligand, the C−S bond was found to be

activated via an oxidative addition process leading to a penta-

coordinated Pd^IVintermediate from the thermodynamically

stable square planar Pd^II(η²-L) species, which after reductive

elimination aﬀords dithiolene complexes Pd^II(PPh₃)₂(dmit).

With the alkyl analogue (PEt₃), both nucleophilic addition A_N

and oxidative addition/reductive elimination sequence may

simultaneously occur through Pd^II(PEt₃)(η¹-L)(η²-L) complex,

leading to the demethylation product. This C−S bond

activation is under thermodynamic control, with the eﬃciency

correlated with the thermodynamic stabilization of the

demethylation product compared with key intermediate

Pd(PR₃)(η¹-L)(η²-L). Theoretical calculations also point out

that the nature of the metal Pt versus Pd does not impact the

privileged pathway. The greater eﬃciency of the demethylation

reaction generally observed with platinum complexes may

result from the greater stability of the Pt^IIP₂(dithiolene)

products versus their Pd^IIanalogues. It should be also

emphasized that the coordination sphere of the Pd^II/Pt^II

impacts not only the mechanism of the S−CH₃bond activation

but also the reactivity of L. The coupling of two L ligands by

reaction with PdCl₂or PtCl₂points out the inﬂuence of the

ancillary ligands in the reactivity of L coordinated on Pd^IIor Pt^II

center.

Inﬂuence of the Metal Center: Pd versus Pt. Finally,

inﬂuence of the metal center (Pt versus Pd) was also examined,

especially in the case of labile phosphines where the

[M(PR₃)₂(dmit)]/[M(PR₃)(η¹-L)(η²-L)] ratios have been

experimentally determined. In line with the experimental

observation, a lower lability of the phosphine ligand is

computed in the case of the platinum complexes (Figure S2).

The energy proﬁles of all the plausible mechanisms (Figures 6

and 8) were found to be very similar to those calculated for the

palladium complexes, showing that the metal does not really

impact the nature of the privileged pathway. The only

noticeable diﬀerences between the two metal centers rest on

two key points. The ﬁrst one corresponds to the relative

thermodynamic instability of oxidative addition product (1a^B/

2a^Band 2a^C), which makes this step more or less reversible and

leads to reformation of M(PR₃)(η¹-L)(η²-L). Penta-coordi-

nated intermediates 1a^B/2a^Band 2a^Cappear to be more stable

by around 8−10 kcal/mol with platinum than with palladium,

consistent with a strengthening of the M−C bond.⁴⁰

Consequently, the oxidative addition step is facilitated with

platinum center (lower reversibility), which can indirectly

further privilege the formation of the demethylation product, as

experimentally conﬁrmed. The second point relates to the

energy of the demethylation process from key intermediate

[M(PR₃)(η¹-L)(η²-L)]. This C−S bond activation is more

thermodynamically favored for the platinum complexes than for

the palladium ones (ΔG_rxaround −10 to −16 kcal/mol for Pt

compared to −5 to −10 kcal/mol for Pd), indicating a better

eﬃciency of the overall process for the platinum complexes

(thermodynamic control).

EXPERIMENTAL SECTION

■

General. All manipulations were performed using Schlenk

techniques in an atmosphere of dry oxygen-free argon. Elemental C,

H, and S analyses were performed on a Leco Elemental Analyzer CHN

900. The ³¹P{¹H} NMR spectra were recorded at 121.50 MHz on a

Bruker DRX 300 spectrometer. CsL,¹⁵Pd(PPh₃)₂Cl₂,⁴¹Pd-

(PEt₃)₂Cl₂,⁴²Pd(dppm)Cl₂,⁴³and Pd(dppe)L₂(3)¹⁵(³¹P{¹H}

NMR-CDCl₃: δ 59.6 ppm) were prepared according to the literature.

Samples of Pd(η²-dppe)(dmit), Pd(η²-dppm)(dmit), Pd(PPh₃₂)₋₂(dmit),

and Pd(PEt₃)₂(dmit) were prepared by reaction of the C₃S₅ligand

(dmit) with the appropriate [PdCl₂P₂] precursor⁴⁴and their ³¹P{¹H}

NMR spectra were recorded (a singlet was observed at 56.7, −39,0,

26.9, and 18.5 ppm, respectively). The bulk purity of new compounds

was established by NMR spectroscopy (¹H, ¹³C, and ³¹P) and

elemental analyses.

CONCLUSION

■

Synthesis of Pd(PEt₃)(η¹-L)(η²-L) (1). To a methanol (25 mL)

solution of cis-Pd(PEt₃)₂Cl₂(0.041 g, 0.1 mmol) was added cesium

1,3-dithiole-2-thioxo-4-methylthio-5-thiolate (0.076 g, 0.22 mmol).

The solution was reﬂuxed for 5 h. After cooling to room temperature,

the reaction mixture was ﬁltered and the dark brown precipitate was

extracted into dichloromethane. The extract was concentrated in vacuo

to aﬀord 1 (0.037 g, 57%). Single crystals of complex 1 were grown by

keeping a solution of 1 in chloroform/hexane mixture for 2 days at 4

°C. Anal. Calcd for C₁₄H₂₁PPdS₁₀(M = 647.4) C, 25.97 H, 3.27 S,

In this contribution, we studied the coordination of mixed

thiolate/thioether ligand L on Pd^IIP₂core and evidenced

competition between formation of PdP₂(η¹-L)₂and PdP(η¹-

L)(η²-L) complexes depending on the nature of the phosphine.

The S_thioether−Pd interaction substituted a P−Pd interaction

with the monophosphines PEt₃and PPh₃, whereas Pd(dppe)-

(η¹-L)₂was found to be stable at room temperature. With

dppm, the coexistence in solution of both species Pd(η²-

dppm)(η¹-L)₂and Pd(η¹-dppm)(η¹-L)(η²-L) is observed.

Experimental investigations revealed that the H₃C−S bond

activation occurred at high temperature leading to heteroleptic

phosphine/dithiolene complexes, PdP₂(dmit) and MeL. In

contrast to the other examples of C(sp³)−S bond cleavage

reported in the literature, which are mainly described as

resulting from substitution involving external nucleophile or are

more scarcely photochemically induced, the thermal C−S bond

activation reported in this work did not require the presence of

an external reactant. Theoretical study by DFT calculations

allowed us to establish the mechanism of the intramolecular

C−S bond activation in these complexes. The latter depends on

1

49.53; found: C, 26.17 H, 3.39 S, 49.23. H NMR (295 K, CDCl₃, δ)

2.66 (s, 6H, CH₃), 2.02 (m, 6H, P−CH₂−), 1.23 (m, 9H, CH₂−CH₃)

ppm. ¹³C{¹H} NMR (CDCl₃, δ) 8.3 (²J_C−P= 2.9 Hz), 15.9 (¹J_C−P

=

31.1 Hz), 22.9(−CH₃), 121.9 (C4), 152.2 (C5, ³J_C−P= 4.4 Hz), 214.6

(CS). ³¹P NMR (295 K, CDCl₃, δ) 35.7 ppm. IR (KBr disk, cm⁻¹):

1058 (ν_CS), 743, 628, 511 (ν_PEt3). UV/vis (CH₂Cl₂, λ_max, nm (ε, M⁻¹

cm⁻¹): 306 (32857), 414 (20952).

Pd(PPh₃)(η¹-L)(η²-L) (2). Complex 2 was prepared by using

precursor Pd(PPh₃)₂Cl₂following the same procedure as that for

complex 1 (70% yield). Anal. Calcd for C₂₆H₂₁PPdS₁₀(M = 791.5)

1

calcd C, 39.46 H, 2.67 S, 40.51; found: C, 39.68 H, 2.67 S, 40.17. H

NMR (295 K, CDCl₃, δ) 7.69−7.48 (m, 15 H, Ph), 2.61 (s, 6 H, CH₃)

ppm. ¹³C{¹H} NMR (CDCl₃, δ) 23.0 (−CH₃), 121.5 (C4), 127.7

P

DOI: 10.1021/acs.organomet.7b00039

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

(¹J_C−P= 54.0 Hz), 128.6 (²J_C−P= 11.2 Hz), 131.8, 134.4 (³J_C−P= 10.9

Hz), 153.1 (C5), 214.6 (CS). ³¹P NMR (295 K, CDCl₃, δ) 32.6

ppm. IR (KBr disk, cm⁻¹): 1057 (ν_CS), 741, 692, 517 (ν_PPh3). UV/vis

(CH₂Cl₂) λ_maxnm (ε, M⁻¹cm⁻¹): 311 (39762), 415 (21714).

Synthesis of Pd(dppm)L₂(4). Pd(dppm)Cl₂(0.056 g, 0.1 mmol)

and cesium 1,3-dithiole-2-thioxo-4-methylthio-5-thiolate (0.069 g, 0.2

mmol) were stirred in acetone (10 mL) at room temperature for 6 h.

After evaporation of the solvent, the orange residue was extracted with

CH₂Cl₂(2 × 7 mL). Ecaporation under reduced pressure aﬀorded 3 as

a red solid (0.068 g, 75%). Anal. Calcd for C₃₃H₂₈P₂PdS₁₀(M = 913.6)

The connectivity of the transition states and their adjacent minima was

conﬁrmed by intrinsic reaction coordinate (IRC) calculations.⁵²In

addition, single-point calculations were carried out on the geometries

obtained in gas phase, including the solvent (acetonitrile, CH₃CN) by

means of the SMD implicit solvation model.⁵³The electronic energies

were then ZPE and temperature corrected based on full gas-phase-

optimized geometry in order to obtain ΔH^{CH CN}and ΔG^{CH CN}values.

Natural bond orbital⁵⁴analyses (NBO, 5.9 version)⁵⁵have been used

to describe the bonding situation in the intermediates 1^B, 2^B, and 3^B.

Molecular orbitals were plotted with Molekel 5.4.⁵⁶The energetic span

(δE) has been calculated in the case of the reactivity of trans-

Pd^II(PPh₃)₂(η¹-L)₂in order to discriminate between mechanisms C

and B-OA/RE. It corresponds to the energetic diﬀerence between

TOF-determining intermediate (TDI) [Pd^II(PPh₃)(η¹-L)(η²-L) for

mechanism B-OA/RE and Pd^II(η²-L)₂for mechanism C] and the

TOF-determining transition state (TDTS, reductive elimination

TS).²⁹

3

1

calcd C, 43.38 H, 3.09 S, 35.10; found: C, 43.62 H, 3.17 S, 34.78. H

NMR (295 K, CDCl₃, δ) 7.66−7.45 (m, 20 H, Ph), 4.15 (br s, 2 H,

PCH₂P), 2.58(s, CH₃(η¹-L)₂), 2,40 (s, CH₃(η¹-L)(η²-L)) ppm. ³¹P

NMR (295 K, CDCl₃, δ) −38.3 ([Pd(η²-dppm)(η¹-L)₂]), 27.5

(−Ph₂P−Pd in [Pd(η¹-dppm)(η¹-L)(η²-L)], −23.25 (Ph₂P−CH₂in

[Pd(η¹-dppm)(η¹-L)(η²-L)]) ppm. IR (KBr disk, cm⁻¹): 1061 (ν_CS).

UV/vis (CH₂Cl₂) λ_max, nm: 308, 412. Note that due to an equilibrium

between Pd(η²-dppm)(η¹-L)₂and the coordination isomer Pd(η¹-

dppm)(η¹-L)(η²-L) identiﬁcation of all C signals was not possible in

the ¹³C{¹H} NMR (see Supporting Information).

ASSOCIATED CONTENT

* Supporting Information

■

S

Synthesis of L−L. Method A. To a solution of CsL (34 mg, 0.1

mmol) in ethanol (15 mL) was added 0.5 equiv of MCl₂(M = Pd or

Pt) under inert atmosphere. The resulting suspension was reﬂuxed for

5 h. Then, the mixture was cooled at room temperature and was

ﬁltered oﬀ. The dark residue was further dissolved in dichloromethane

to obtain a clear yellow solution of the disulﬁde compound.

Evaporation of the solvent aﬀorded an orange colored powder

which was recrystallized in dichloromethane-hexane mixture to get

single crystals for X-ray diﬀraction (4.6 mg, 22% with M = Pd).

Method B. A batch of CsL (0.34 g, 1.0 mmol) was dissolved in 15

mL of distilled water to obtain a red solution in a Schlenk tube. Then,

0.01 mmol of sodium iodide (1.5 mg, 1 mol %) and 1.0 mmol of 30%

H₂O₂(0.11 mL) were added directly, and the mixture was stirred at

room temperature for 24 h. Next day, a saturated aqueous solution (20

mL) of Na₂S₂O₃was added to get an orange precipitate of the disulﬁde

compound. The mixture was ﬁltered oﬀ and washed thoroughly with

water and cold methanol. The precipitate was dried under vacuum

(0.085 g, 40%). Anal. Calcd for C₈H₆S₁₀(M = 422.7) calcd C, 22.73

The Supporting Information is available free of charge on the

ACS Publications website at DOI: 10.1021/acs.organo-

met.7b00039.

Cartesian coordinates (XYZ)

Crystallographic information for 1, 2, and L−L (CIF)

UV−visible spectrum of 2, energy proﬁles for MP₂(η¹-

L)₂complexes, NBO analysis of the intermediate 1^B, 2^B

and 3^B, main minima and TS, energy proﬁles, free energy

proﬁles, energy for the phosphine-predissociation path-

way, main bond angles in the initial reactants and

corresponding TS, NMR spectra of 1, 2, 4 and L−L

(PDF)

AUTHOR INFORMATION

Corresponding Authors

*E-mail: fabrice.guyon@univ-fcomte.fr.

*E-mail. karinne.miqueu@univ-pau.fr.

■

1

H, 1.43 S, 75.85; found: C, 22.85 H, 1.47 S, 75.48. H NMR (CDCl₃,

300 MHz, δ) 2.55 (s) ppm. ¹³C{¹H} NMR (CDCl₃): 19.3 (−CH₃),

128.8, 147.6, 209.8 (CS) ppm. IR (KBr disk, cm⁻¹): 1061 (ν_CS).

UV/vis (CH₂Cl₂) λ_max, nm (ε, M⁻¹cm⁻¹): 275 (13786), 384 (15714).

Mp: 123 °C.

ORCID

Fabrice Guyon: 0000-0001-8905-2904

Notes

The authors declare no competing ﬁnancial interest.

Crystal Structure Determinations. The crystal structure

determinations were eﬀected at 173.15(2) K on a Xcalibur, Sapphire3

diﬀractometer. The crystal was kept at 173.15 K during data collection.

Using Olex2⁴⁵the structure was solved with the ShelXT⁴⁶structure

solution program using Direct Methods and reﬁned with the XL⁴⁷

reﬁnement package using Least Squares minimization. Crystallo-

graphic parameters are listed in Table S2. CCDC-1489553 (1),

1489549 (2), and 1489552 (L−L) contain the supplementary

crystallographic data for this paper. These data can be obtained free

of charge from the Cambridge Crystallographic Data Centre via www.

ccdc.cam.ac.uk/data_request/cif.

Computational Details. All calculations were performed using the

Gaussian 09 package⁴⁸and the B3LYP hybrid functional. B3LYP⁴⁹is a

three-parameter functional developed by Becke which combines the

Becke gradient-corrected exchange functional and the Lee−Yang−Parr

and Vosko−Wilk−Nusair correlation functionals with part of exact HF

exchange energy. Palladium and platinum atoms were described with

the relativistic electron core potential SDD and associated basis set,⁵⁰

augmented by a set of f-orbital polarization functions⁵¹and 6-31G**

basis set were employed for other atoms.

ACKNOWLEDGMENTS

Financial support from the Centre National de la Recherche

■

Scientiﬁque, the Universite

grant to S.K.) and Universite

́

de Franche-Comte

de Pau et des Pays de l′Adour

(UPPA) are gratefully acknowledged. UPPA, MCIA (Meso-

́

(post-doctoral

́

centre de Calcul Intensif Aquitain), and IDRIS under

Allocation 2016 (i2016080045) made by Grand Equipement

National de Calcul Intensif (GENCI) are also acknowledged

for computational facilities.

REFERENCES

(1) Blomstrand, C. W. J. Prakt. Chem. 1883, 27, 161−198.

■

(2) (a) Pan, F.; Shi, Z.-J. ACS Catal. 2014, 4, 280−288. (b) Wang, L.;

He, W.; Yu, Z. Chem. Soc. Rev. 2013, 42, 599−621. (c) Modha, S. G.;

Mehta, V. P.; Van der Eycken, E. Chem. Soc. Rev. 2013, 42, 5042−

5055. (d) Desnoyer, A. N.; Love, J. A. Chem. Soc. Rev. 2017, 46, 197−

238. (e) Ouch, K.; Mashuta, M. S.; Grapperhaus, C. A. Inorg. Chem.

2011, 50, 9904−9914. (f) Sargent, A. L.; Titus, E. P. Organometallics

1998, 17, 65−77. (g) Angelici, R. J. Organometallics 2001, 20 (7),

1259−1275. (h) Kuniyasu, H.; Ohtaka, A.; Nakazono, T.; Kinomoto,

M.; Kurosawa, H. J. Am. Chem. Soc. 2000, 122, 2375−2376.

Optimization of the diﬀerent minima and TS were carried out

without any symmetry restrictions in gas phase. All total energies and

Gibbs free energies have been zero-point energy (ZPE) and

temperature corrected using unscaled density functional frequencies.

Frequency calculations were undertaken to conﬁrm the nature of the

stationary points, yielding one imaginary frequency for transition

states, corresponding to the expected process, and zero for minima.

Q

DOI: 10.1021/acs.organomet.7b00039

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

(3) (a) Lindoy, L. F. Coord. Chem. Rev. 1969, 4, 41−71. (b) Lindoy,

L. F.; Livingstone, S. B.; Lockyer, T. N. Inorg. Chem. 1967, 6, 652−

656. (c) Lockyer, T. N. Aust. J. Chem. 1974, 27, 259−267. (d) Sacconi,

L.; Speroni, G. P. Inorg. Chem. 1968, 7, 295−299. (e) Roundhill, D.

M.; Beaulieu, W. B.; Bagchi, U. J. Am. Chem. Soc. 1979, 101, 5428−

5430.

(4) Kim, J. S.; Reibenspies, J. H.; Darensbourg, M. T. J. Am. Chem.

Soc. 1996, 118, 4115−4123.

(5) Schrauzer, G. N.; Rabinowitz, H. N. J. J. Am. Chem. Soc. 1968, 90,

4297−4302. (b) Ohkoshi, S.; Ohba, Y.; Iwaizumi, M.; Yamauchi, S.;

Ohkoshi-Ohtani, M.; Tokuhisa, K.; Kajitani, M.; Akiyama, T.;

Sugimori, A. Inorg. Chem. 1996, 35, 4569−4574.

(6) (a) Benefiel, A.; Roundhill, D. M. Inorg. Chem. 1986, 25, 4027−

4031. (b) Benefiel, A.; Roundhill, D. M.; Fultz, W. C.; Rheingold, A. L.

Inorg. Chem. 1984, 23, 3316−3324.

(7) Mandal, S.; Paul, N.; Banerjee, P.; Mondal, T. K.; Goswami, S.

Dalton Trans. 2010, 39, 2717−2726.

(8) (a) Takeda, N.; Tanaka, Y.; Oma, R.; Sakakibara, F.; Unno, M.

Bull. Chem. Soc. Jpn. 2016, 89, 922−930. (b) Takeda, N.; Tanaka, Y.;

Sakakibara, F.; Unno, M. Bull. Chem. Soc. Jpn. 2010, 83, 157−164.

(9) (a) Panda, S.; Rama Krishna, G.; Reddy, C. M.; Zade, S. S. Dalton

Trans. 2011, 40, 6684−6690. (b) Dutta, P. Kr.; Panda, S.; Rama

Krishna, G.; Reddy, C. M.; Zade, S. S. Dalton Trans. 2013, 42, 476−

483.

(10) Kim, S.; Boyle, P. D.; McCready, M. S.; Pellarin, K. R.;

Puddephatt, R. J. Chem. Commun. 2013, 49, 6421−6423.

(11) Guyon, F.; Knorr, M.; Garillon, A.; Strohmann, C. Eur. J. Inorg.

Chem. 2012, 2012, 282−291.

(12) Sgro, M. J.; Stephan, D. W. Organometallics 2012, 31, 1584−

1587.

(13) (a) Svenstrup, N.; Becher, J. Synthesis 1995, 1995, 215−235.

(b) Rauchfuss, T. B.; Stiefel, E. I. Prog. Inorg. Chem. 2003, 52, 1−54.

(14) (a) Garreau-de Bonneval, B.; Moineau-Chane Ching, K. I.;

Alary, F.; Bui, T.-T.; Valade, L. Coord. Chem. Rev. 2010, 254, 1457−

1467. (b) Kosaka, Y.; Yamamoto, H. M.; Nakao, A.; Tamura, M.; Kato,

R. J. Am. Chem. Soc. 2007, 129, 3054−3055. (c) Cassoux, P. Coord.

Chem. Rev. 1999, 185−186, 213−232.

2006, 71, 8268−8271. (c) Antoniow, S.; Witt, D. Synthesis 2007, 2007,

́

363−366. (d) Sanz, R.; Aguado, R.; Pedrosa, M. R.; Arnaiz, F. Synthesis

2002, 2002, 856−858. (e) Khazaei, A.; Zolfigol, M. A.; Rostami, A.

Synthesis 2004, 2959−2961. (f) Bao, M.; Shimizu, M. Tetrahedron

2003, 59, 9655−9659. (g) Uemura, S. In Comprehensive Organic

Synthesis; Trosty, B. M., Fleming, I., Eds.; Pergamon Press: Oxford,

1991; Vol. 7, p 757. (h) Tam, J. P.; Wu, C.-R.; Liu, W.; Zhang, J.-W. J.

Am. Chem. Soc. 1991, 113, 6657−6662.

(27) Kirihara, M.; Asai, Y.; Ogawa, S.; Noguchi, T.; Hatano, A.; Hirai,

Y. Synthesis 2007, 2007 (21), 3286−3289.

(28) Wrixon, J. D.; Hayward, J. J.; Raza, O.; Rawson, J. Dalton Trans.

2014, 43, 2134−2139.

(29) The structural parameters of PtP₂S₂complexes in which P-

donor ligands are only organomonophosphines have been recently

analyzed. With monodentate thiolates, 21 of the 31 examples exhibit a

trans conﬁguration: Melnik, M.; Mikus, P. Rev. Inorg. Chem. 2016, 36,

43−51.

(30) Recoordination of the phosphine (1b^B+ PEt₃→M^II(PEt₃)₂dmit

+ Me₂dmit) requires a barrier around ΔG^#= 18.5 kcal/mol (ΔH^#: 6.7

kcal/mol) for Pd complex and ΔG^#= 30.9 kcal/mol (ΔH^#= 18.5

kcal/mol) for the Pt one.

(31) Even if there is a possible overestimation by our calculations of

the entropic eﬀect, the formation of 1^Bremains favorable (ΔH = 2.3

kcal/mol, Figure S2).

(32) Tolman, C. A. Chem. Rev. 1977, 77, 313−348.

(33) ΔH^{CH CN}values are also consistent with this experimental

3

observation: ΔH^{CH CN}

= −3.8 kcal/mol < ΔH^{CH CN}= 2.3 kcal/

3

PPh₃

PEt₃

mol (Figure S2). Consequently, even if entropic eﬀects are

overestimated, they do not alter our conclusion (easier dissociation

of the phosphine with PPh₃ligand).

(34) For the palladium complex, recoordination of the ﬁrst

phosphine (2b^C+ PPh₃→2b^B, ΔG^#= 22.3 kcal/mol; ΔH^#= 9.5

kcal/mol) requires an activation barrier similar to the second one (2b^B

+ PPh₃→Pd^II(PPh₃)₂dmit + Me₂dmit, ΔG^#= 18.7 kcal/mol; ΔH^#=

10.4 kcal/mol).

(35) The TS associated to the phosphine decoordination from

Pd^II(PPh₃)(η¹-L)(η²-L) is a late transition state with a Pd−P distance

around 2.918 Å and a Pd−S(Me) of 2.477 Å quasi-formed (other Pd−

S distances: 2.37−2.41 Å).

(36) The TS corresponding to a A_Nmechanism is high in energy

(ΔG^#= 52.9 kcal/mol) and is consequently unlikely.

(15) Zeltner, S.; Olk, R.-M.; Joerchel, P.; Sieler, J. Z. Anorg. Allg.

Chem. 1999, 625 (2), 368−373.

(16) (a) Basolo, F.; Chatt, J.; Gray, H. B.; Pearson, R. G.; Shaw, B. L.

J. Chem. Soc. 1961, 2207−2215. (b) Wilkins, R. G. Kinetics and

Mechanism of Reactions of Transition Metal Complexes; Wiley-VCH:

New York, 2002.

(37) (a) Kozuch, S. WIREs Comput. Mol. Sci. 2012, 2, 795−815.

(b) Kozuch, S.; Shaik, S. Acc. Chem. Res. 2011, 44, 101−110. (c) Uhe,

A.; Kozuch, S.; Shaik, S. J. Comput. Chem. 2011, 32, 978−985.

(17) (a) Miranda, S.; Cerrada, E.; Luquin, A.; Mendia, A.; Laguna, M.

Polyhedron 2012, 43, 15−21. (b) Vinas, C.; Cirera, M. R.; Teixidor, F.;

Sillanpaa, R.; Kivekas, R. J. Organomet. Chem. 1997, 530, 89−94.

(18) Cordero, B.; Gomez, V.; Platero-Prats, A. E.; Reves, M.;

Echeverria, J.; Cremades, E.; Barragan, F.; Alvarez, S. Dalton Trans.

2008, 2832−2838.

́ ́

(38) (a) Basallote, M. G.; Duran, J.; Fernandez-Trujillo, M. J.; Manez,

M. A. J. Organomet. Chem. 2000, 609 (1−2), 29−35. (b) Basallote, M.

G.; Duran

́

, J.; Fernan

́

dez-Trujillo, M. J.; Gonzal

́

ez, G.; Man

́

ez, M. A.;

̃

Martínez, M. Inorg. Chem. 1998, 37, 1623−1628.

(39) The same conclusion can be drawn on the basis of the ΔH^{CH CN}

3

(19) Wrixon, J. D.; Hayward, J. J.; Rawson, J. M. Inorg. Chem. 2015,

54, 9384−9386.

values: ΔH^{CH CN}

= 12.0 kcal/mol > ΔH^{CH CN}= 2.3 kcal/mol >

3

dppe

PEt₃

(20) Martinez, J.; Adrio, L. A.; Antelo, J. M.; Pereira, M. T.;

Fernandez, J. J.; Vila, J. M. Polyhedron 2006, 25, 2848−2858.

(21) (a) Chandrasekaran, P.; Greene, A. F.; Lillich, K.; Capone, S.;

Mague, J. T.; DeBeer, S.; Donahue, J. P. Inorg. Chem. 2014, 53, 9192−

9205. (b) Wrixon, J. D.; Ahmed, Z. S.; Anwar, M. U.; Beldjoudi, Y.;

Hamidouche, N.; Hayward, J. J.; Rawson, J. Polyhedron 2016, 108,

115−121.

ΔE^{CH CN}

= −3.8 kcal/mol (Figure S2).

3

PPh₃

(40) (a) Luo, Y. R. Comprehensive Handbook of Chemical Bond

Energies; CRC Press, Boca Raton, FL, 2007. (b) Citir, M.; Metz, R. B.;

Belau, L.; Ahmed, M. J. Phys. Chem. A 2008, 112, 9584−9590.

(c) Skinner, H. A.; Connor, J. A. Pure Appl. Chem. 1985, 57, 5779−

5788.

(41) Miyaura, N.; Suzuki, A. Org. Synth. 1990, 68, 130−136.

(42) Doyle, J. R.; Slade, P. E.; Jonassen, H. B.; Rhoda, R. N. Inorg.

Synth. 1960, 6, 216−219. (b) Hartley, F. R. Organomet. Chem. Rev. A

1970, 6, 119−137.

(43) Steffen, W. L.; Palenik, G. J. Inorg. Chem. 1976, 15, 2432−2439.

(44) Vicente, R.; Ribas, J.; Cassoux, P. Nouv. J. Chim. 1984, 8, 653−

658.

(22) (a) Van Wart, H. E.; Lewis, A.; Scheraga, H. A.; Saeva, F. D.

Proc. Natl. Acad. Sci. U. S. A. 1973, 70, 2619−2673. (b) David, C.

Ph.D. Dissertation. Universite

2009.

́ ́

de Franche-Comte, Besanco̧ n, France,

(23) Breitzer, J. G.; Smirnov, A. I.; Szczepura, l. F.; Wilson, S. R.;

Rauchfuss, T. B. Inorg. Chem. 2001, 40, 1421−1429.

(24) Galloway, C. P.; Doxsee, D. D.; Fenske, D.; Rauchfuss, T. B.;

Wilson, S. R.; Yang, X. Inorg. Chem. 1994, 33, 4537−4544.

(25) Aida, M.; Nagata, C. Theor. Chim. Acta 1986, 70, 73−80.

(26) (a) Shimizu, D.; Takeda, N.; Tokitoh, N. Chem. Commun. 2006,

177−179. (b) Hunter, R.; Caira, M.; Stellenboom, N. J. Org. Chem.

(45) Dolomanov, O. V.; Bourhis, L. J.; Gildea, R. J.; Howard, J. A. K.;

Puschmann, H. J. Appl. Crystallogr. 2009, 42, 339−341.

(46) Sheldrick, G. M. Acta Crystallogr., Sect. A: Found. Adv. 2015, 71,

3−8.

R

DOI: 10.1021/acs.organomet.7b00039

Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

(47) Sheldrick, G. M. Acta Crystallogr., Sect. A: Found. Crystallogr.

2008, 64, 112−122.

(48) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;

Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci,

B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H.

P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.;

Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima,

T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.;

Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin,

K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.;

Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega,

N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.;

Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.;

Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.;

Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.;

Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O.;

Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09,

revision C.01; Gaussian, Inc.: Wallingford, CT, 2009.

(49) (a) Becke, A. D. J. Chem. Phys. 1993, 98, 5648−5652. (b) Becke,

A. D. Phys. Rev. A: At., Mol., Opt. Phys. 1988, 38, 3098−3100. (c) Lee,

C.; Yang, W.; Parr, R. G. Phys. Rev. B: Condens. Matter Mater. Phys.

1988, 37, 785−789.

(50) Andrae, D.; Haussermann, U.; Dolg, M.; Stoll, H.; Preub, H.

̈

Theor. Chim. Acta 1990, 77, 123−141.

(51) Ehlers, A. W.; Bohme, M.; Dapprich, S.; Gobbi, A.; Hollwarth,

̈

A.; Jonas, V.; Kohler, K. F.; Stegmann, R.; Veldkamp, A.; Frenking, G.

̈

Chem. Phys. Lett. 1993, 208, 111−114.

(52) (a) Gonzalez, C.; Schlegel, H. B. J. Chem. Phys. 1989, 90, 2154−

2161. (b) Gonzalez, C.; Schlegel, H. B. J. Phys. Chem. 1990, 94, 5523−

5527.

(53) Marenich, A. V.; Cramer, C. J.; Truhlar, D. G. J. Phys. Chem. B

2009, 113, 6378−6396.

(54) (a) Reed, E.; Curtiss, L. A.; Weinhold, F. Chem. Rev. 1988, 88,

899−926. (b) Foster, J. P.; Weinhold, F. J. Am. Chem. Soc. 1980, 102,

7211−7218. (c) Reed, A. E.; Weinhold, F. J. Chem. Phys. 1985, 83,

1736−1740.

(55) Glendening, E. D.; Badenhoop, J. K.; Reed, A. E.; Carpenter, J.

E.; Bohmann, J. A.; Morales, C. M.; Weinhold, F. NBO 5.0;

Theoretical Chemistry Institute, University of Wisconsin: Madison,

2001.

(56) Varetto, U. MOLEKEL 5.4; Swiss National Supercomputing

Centre: Lugano, Switzerland, 2012.

S

DOI: 10.1021/acs.organomet.7b00039

Organometallics XXXX, XXX, XXX−XXX

Article Doi

Experimental and theoretical studies on the mechanism of the C-S bond activation of PdII thiolate/thioether complexes

DOI: 10.1021/acs.organomet.7b00039

Source and publish data:

Authors:

Article abstract of DOI:10.1021/acs.organomet.7b00039

Full text of DOI:10.1021/acs.organomet.7b00039

Products guided by the article

R&D Labs maybe for 96042-53-4

Relevant to this article

Hot Product