Bis- N -heterocyclic carbene palladium(IV) tetrachloride complexes: Synthesis, reactivity, and mechanisms of direct chlorinations and oxidations of organic substrates

ARTICLE

pubs.acs.org/JACS

Bis-N-heterocyclic Carbene Palladium(IV) Tetrachloride Complexes:

Synthesis, Reactivity, and Mechanisms of Direct Chlorinations

and Oxidations of Organic Substrates

A. Scott McCall, Hongwang Wang, John M. Desper, and Stefan Kraft*

Department of Chemistry, Kansas State University, 213 CBC Building, Manhattan, Kansas 66506-0401, United States

S Supporting Information

b

ABSTRACT: This Article describes the preparation and

isolation of novel octahedral CH₂-bridged bis-(N-hetero-

cyclic carbene)palladium(IV) tetrachlorides of the gen-

eral formula LPd^IVCl₄[L = (NHC)CH₂(NHC)] from

LPd^IICl₂and Cl₂. In intermolecular, nonchelation-con-

trolled transformations LPd^IVCl₄reacted with alkenes and

alkynes to 1,2-dichlorination adducts. Aromatic, benzylic,

and aliphatic C-H bonds were converted into C-Cl

bonds. Detailed mechanistic investigations in the dichlor-

inations of alkenes were conducted on the 18VE Pd^IV

complex. Positive solvent eﬀects as well as kinetic measurements probing the impact of cyclohexene and chloride concentrations on

the rate of alkene chlorination support a Pd^IV-Cl ionization in the ﬁrst step. Product stereochemistry and product distributions

from various alkenes also support Cl^þ-transfer from the pentacoordinated Pd^IV-intermediate LPd^IVCl₃to oleﬁns. 1-Hexene/

þ

3-hexene competition experiments rule out both the formation of π-complexes along the reaction coordinate as well as in situ

generated Cl₂from a reductive elimination process. Instead, a ligand-mediated direct Cl^þ-transfer from LPd^IVCl₃^þto the π-system

is likely to occur. Similarly, C-H bond chlorinations proceed via an electrophilic process with in situ formed LPd^IVCl₃^þ. The

presence of a large excess of added Cl^-slows cyclohexene chlorination while the presence of stoichiometric amounts of chloride

accelerates both Pd^IV-Cl ionization and Cl^þ-transfer from LPd^IVCl₃^þ. H NMR titrations, T1 relaxation time measurements,

1

binding isotherms, and Job plot analysis point to the formation of a trifurcated Cl^-H-C bond in the NHC-ligand periphery as a

3 3 3

supramolecular cause for the accelerated chemical events involving the metal center.

’ INTRODUCTION

carbon-ligands on Pd^IVfrequently do not constitute “spectator

ligands” and become involved in reductive elimination events.

Unlike square-planar Pd^II-analogues, octahedral d⁶complexes

of Pd^IVare not prone to unwanted side reactions such as β-

eliminations,^2a,band their reactivity is almost exclusively domi-

nated by reductive eliminations;^2d,3a,5simple ligand substitu-

tions⁶as well as Pd^IVto M^IItransmetalations (M^II= Pd^II,^5g,7

Pt^II8) represent the only exceptions. This narrow scope of Pd^IV-

based transformations has eﬀectively precluded the study of

intermolecular Pd^IVchemistry with common organic functional

groups. Another practical limitation in catalytic Pd^II/Pd^IVcycles

is the common reliance on “chelation-controlled” or “ligand-

directed” activations^2e,9by Pd^IIto impart regioselectivity and to

facilitate otherwise sluggish reactions.¹⁰

Palladium-mediated transformations have revolutionized or-

ganic chemistry over the past 30 years.¹The main mode of

operation in catalytic reactions involves Pd⁰/Pd^IIcycles. How-

ever, in recent years, an alternative Pd^II/Pd^IVcouple has been

aggressively pursued in high-oxidation state routes that provide

complementary reactivities unattainable to Pd⁰/Pd^IIcouples.²

Two ﬁelds of particular interest in this arena focus on C-H bond

activations/functionalizations and oxidative difunctionalizations

of alkenes; under catalytic conditions, both of these reaction

classes proceed through similar cycles involving a substrate

activation step, an oxidation step, and a functionalization step

(Scheme 1). A general paradigm has emerged that prescribes

diﬀerent roles for the metal in its oxidation states þ2 and þ4;

speciﬁcally, activation steps that establish initial Pd-C bonds are

executed by Pd^II, while functionalizations that establish C-X

bonds are performed by Pd^IV.

Novel direct Pd^IV-activations/functionalizations may provide

an alternative to Pd^II/Pd^IVcycles that commonly require ele-

vated reaction temperatures¹¹to overcome barriers of rate-

determining^2b,cPd^IIactivation steps. Driven by an agenda to

widen the scope of Pd^IVchemistry, we had to take on two tasks:

The formation of the þ4 oxidation state on palladium is often

facilitated by the presence of strongly σ-donating carbon-

ligands,³and many isolable or spectroscopically characterized

Pd^IVcompounds are therefore organometallic in nature.⁴However,

Received: August 26, 2010

Published: January 19, 2011

r

2011 American Chemical Society

1832

dx.doi.org/10.1021/ja107342b J. Am. Chem. Soc. 2011, 133, 1832–1848

|

Journal of the American Chemical Society

ARTICLE

Scheme 1

Scheme 2

(i) provide a suitable nonreactive spectator ligand scaﬀold that

can stabilize the metal center at its oxidation state þ4 and

(ii) identify intermolecular reaction types in which suitable

groups are transferred from Pd^IVto substrates such as alkenes,

alkynes, or aromatic platforms. In this work, we focused on

studying stoichiometric Pd^IVreactions.

To address task (i), N-heterocyclic carbene ligands (NHCs)¹²

appeared to be a suitable choice as ancillary support for Pd^IV

due to their known propensity to stabilize highly oxidized

metal centers without ligand dissociation.¹³To this end, Arnold

and Sanford recently demonstrated that a Pd^IV-NHC-

dichloro-alkoxy-benzo[h]quinoline complex undergoes (benzo[h]-

quinoline)C-Cl reductive elimination with no C-Cl function-

alization involving the NHC portion.¹⁴It should be noted

though that C-C reductive eliminations involving NHCs have

occasionally been observed from NHC-Pd^II-methyl com-

plexes.¹⁵Given their ability to operate as catalysts for methane

activation in strongly oxidizing milieus [K₂S₂O₈/(CF₃CO)₂O/

CF₃COOH], we chose Strassner’s bis-chelating NHC complexes

LPd^IIX₂1-3¹⁶as a starting point for our explorations.

donors toward added chloride ions to form supramolecular

adducts of the type Cl^-(NHC)₂Pd^IVCl₄. We discovered that

3 3 3

these noncovalent interactions accelerate Pd^IV-Cl heterolysis to

form a zwitterionic intermediate Cl^-(NHC)₂Pd^IVCl₃^þ; we

3 3 3

also found that Cl^þ-transfers from these intermediates to alkenes

are accelerated in the presence of such C-H Cl^-hydrogen bonds.

3 3 3

’ RESULTS

Synthesis of Pd^IVTetrachlorides. The orange-yellow bis-

chelating NHC-Pd^IVcomplex 4 was generated from a suspen-

sion of LPd^IICl₂complex 1^16bin DMF and Cl₂in 86% yield

(Scheme 2) and in 76% yield from LPd^IIBr₂complex 2. While

poorly soluble in common organic solvents, we were able to obtain

¹H NMR spectroscopic data of 4 in DMF-d₇with aromatic

signals at δ 7.97, 7.80, two doublets for the bridging CH₂group at

δ 7.13, 6.85 (¹J = 13.1 Hz), and a CH₃singlet δ 4.44. Similarly,

the ¹³C NMR spectrum showed three ring signals at δ 139.08,

126.47, and 123.86 and a single peak for the bridging carbon at δ

63.42. ¹H and ¹³C NMR data are consistent with a C_s-symmetrical

tub-shaped structure analogous to Pd^II-precursor 1.^16b

In addressing task (ii), we directed our initial eﬀorts to the

exploration of intermolecular reactions of Pd^IV-tetrahalides, in

particular tetrachlorides. This choice was based on two criteria:

17

(a) Tetrachloride complexes such as (bipy)PdCl₄and (en)-

PdCl₄and (py)₂PdCl₄and [(NH₃)₄PdCl₂]Cl₂¹⁸have been isolated

previously and show limited thermal stability in the solid state.

(b) The principal viability of C-H halogenations of hydrocarbons

has been demonstrated in chelation-controlled Pd^II/Pd^IVcycles.¹⁹

Here, we report the synthesis and isolation of novel LPd^IVCl₄

(L = NHC-CH₂-NHC) complexes and their stoichiometric

intermolecular reactions with alkenes/alkynes as well as C-H

bonds; neither alkenes nor alkynes have been previously reacted

with isolated Pd^IVspecies of any kind. All substrates tested

here are devoid of potential directing groups. Mechanistic

studies on alkene dichlorinations revealed a stepwise process

including initial Cl^--loss from the metal center and direct, ligand-

Vapor diﬀusion (DMF/CH₂Cl₂) produced crystals of 4

suitable for X-ray crystal structure analysis. In the solid state,

the tetrachloride adopted an octahedral geometry around palla-

dium (Figure 1). Pd-C^carbenebonds [Pd1-C32, 2.015(4) Å;

Pd1-C22, 2.019(4) Å] experienced a bond lengthening of 0.046

and 0.049 Å as compared to 1,^16bhinting at diminished π-back-

bonding from a less π-basic d⁶-metal center.²⁰Equatorial “in-

plane” Pd-Cl^eqbond distances in 4 [Pd1-Cl1, 2.3769(10) Å;

Pd1-Cl2, 2.3813(9) Å] remained almost unchanged relative to

the Pd^II-precursor (Δ < 0.004 Å relative to 1) but were longer

relative to their perpendicular Pd-Cl^axcounterparts [Pd1-Cl3,

2.3189(9) Å; Pd1-Cl4, 2.3081(9) Å] by an average of 0.066 Å,

underlining a signiﬁcant trans-inﬂuence of NHC ligands.

þ

mediated Cl^þ-transfer from LPd^IVCl₃intermediates to

π-bonds without the involvement of [LPd^IVCl₃(η²-alkene)]^þ

complexes. Serendipitously, we discovered that C-H bonds of

the NHC-ligand backbone in LPd^IVCl₄can act as hydrogen-bond

In an eﬀort to bolster solubility in common organic solvents,

tetradecyl-substituted bis-imidazolium dichloride 5 and Pd-

(OAc)₂were converted to Pd(II) complex 6 as a white powder

1833

dx.doi.org/10.1021/ja107342b |J. Am. Chem. Soc. 2011, 133, 1832–1848

Journal of the American Chemical Society

ARTICLE

Figure 1. X-ray crystal structure of 4.

in 67% isolated yield (Scheme 2). Subsequent chlorination with

Cl₂/CH₂Cl₂yielded 7 as an orange-yellow powder quantita-

tively. Unfortunately, once 7 had precipitated from solution on

the time scale of a few minutes, it remained insoluble in common

1

organic solvents. We were able to obtain a H NMR spectrum

in CDCl₃prior to the onset of precipitation via in situ chlorination

of 6 and rapid spectral acquisition within 2 min. Two aromatic

doublets at δ 8.32 and δ 7.08 (broad singlets) and two doublets

2

(δ 7.48 and δ 7.07; J = 13.4 Hz) for the CH₂-bridge were

consistent with a tub-shaped C_s-symmetrical structure analogous

to that of 4. Serendipitously, we discovered that the addition of

the electrolytes Bu₄NCl or Bu₄NBF₄dramatically increased the

solubility of 7 in CDCl₃, and up to 0.1 M solutions could be

Figure 2. ¹H and ¹³C NMR chemical shifts for 7 in CDCl₃in the

presence of two diﬀerent electrolytes (Bu₄NBF₄and Bu₄NCl).

generated, which allowed for more time-consuming ¹H and ¹³

C

1

NMR characterizations.²¹We found that H NMR chemical

shifts of H⁵and H^exoexperienced notable downﬁeld shifts in

CDCl₃/Bu₄NCl by 0.37 and 1.34 ppm, respectively, relative to

CDCl₃/Bu₄NBF₄solutions, while their connected carbons C⁵

and C^bridgewere shifted upﬁeld in the ¹³C NMR spectrum by

0.51 and 0.61 ppm in CDCl₃/Bu₄NCl (Figure 2).

(Scheme 3) thermodynamically favors 7 Cl^-, and from a

3

Benesi-Hildebrand plot we obtained K_eq= 139.2 M^-1(ΔG⁰_Cl-

-2.9 kcal/mol).

=

For a structural localization of chloride binding, T1-relaxation

studies were conducted for both 7 and 7 Cl^-(Figure 5). We

3

A smaller upﬁeld shift by 0.08 ppm was observed for H⁴, while

C⁴was shifted downﬁeld by 0.44 ppm. Electrolyte-dependent

chemical shift diﬀerences for other protons and carbons were

much less pronounced, and variations along the side chain faded

out rapidly. Unambiguous assignments of all protons were

accomplished via a ¹H-¹H ROESY experiment. Using H^R,βas

a reference point, H⁴, H⁵, H^exo, and H^endowere identiﬁed through a

chain of neighborhood relationships, giving rise to ROE cross-

peaks of (H⁴, H^R,β) and (H⁵, H^exo) as well as TOCSY-type cross-

peaks for (H⁵, H⁴) and (H^exo, H^endo) (Appendix 1, Supporting

Information). Addition of other oxidants to bis-NHC-Pd^IIcom-

plexes failed to produce stable Pd^IVproducts.^22,23

found site-speciﬁc reductions in T1 values of 7 Cl^-relative to 7

3

by 25% and 17% for H⁵and H^exo, respectively, and an increase by

7% for H⁴, while other protons were virtually unaﬀected. Faster

relaxation for H⁵and H^exois consistent with the inﬂuence of

enhanced ³⁵Cl/³⁷Cl-quadrupolar relaxations upon Cl^--binding

via a Cl^-H-C hydrogen bond.²⁵

3 3 3

The combined data of chemical shift eﬀects, NMR titration

and Job plot, Benesi-Hildebrand analysis, and T1-relaxation

times are in support of a trifurcated hydrogen bond in 7 Cl^-

3

0

involving contacts of chloride to H^exoas well as H⁵/H⁵in the

ligand periphery (Scheme 3).

Reactivity. Small excesses (1.05 equiv) of LPd^IVCl₄(7 in

Chloride Binding Analysis. The remarkable impact of salts

on the solubility of 7 in CDCl₃prompted us to study their

Bu₄NBF₄/CDCl₃or 7 Cl^-in Bu₄NCl/CDCl₃) were reacted at

3

25 °C with a variety of substrates including alkenes/alkynes as

well as aromatic, benzylic, and aliphatic C-H bonds (Table 1).

LPd^IICl₂(6) was found to be the only organometallic product in

all cases. Reaction conditions were not optimized, and yields

were primarily limited by incomplete conversions (Table 1)

rather than by the occurrence of uncharacterized side reactions.

Alkenes/alkynes were converted into their 1,2-dichlorination

1

potential interactions with 7 in more detail. When H NMR

samples of 7 in CDCl₃/Bu₄NBF₄were sequentially titrated with

0.1 equiv aliquots of Bu₄NCl, we were able to follow progressive

downfield shifts for H^exoand H⁵and a progressive upfield shift

for H⁴(Figure 3a), which was expected on the basis of the

electrolyte-dependent chemical shifts of 7 (Figure 2). Notably,

chemical shift changes leveled out with increasing concentrations

of Bu₄NCl, which discredits the possibility of a bulk dielectric

adducts by 7 Cl^-(Table 1, entries 1-5) with side products

3

from allylic and vinylic chlorinations (entries 3-5). Aromatic

and benzylic C-H bonds of electron-rich substrates were

converted into C(sp²)-Cl and C(sp³)-Cl bonds (entries

1

effect (Figure 3b). The Job plot²⁴obtained from the H NMR

titration revealed a 1:1 binding ratio for Pd^IV:Cl^-(Figure 4) for

the new complex 7 Cl^-. The equilibrium 7 þ Cl^-h 7 Cl^-

6-8, 10-11), while p-xylene was inert toward 7 Cl^-even upon

3

1834

dx.doi.org/10.1021/ja107342b |J. Am. Chem. Soc. 2011, 133, 1832–1848

Journal of the American Chemical Society

ARTICLE

Figure 3. (a) ¹H NMR titration of 7 in CDCl₃/Bu₄NBF₄with aliquots of Bu₄NCl displaying changes in chemical shifts of aromatic and bridge protons.

(b) Saturation of chemical shift changes of H⁵at elevated Bu₄NCl concentration.

Figure 4. Job plot for mixtures of Bu₄NCl and LPd^IVCl₄(7 and 7 Cl^-

3

combined) in CDCl₃/Bu₄NBF₄solutions. Plotted is χ(Bu₄NCl) Δδ-

(H⁵) as a function of χ(Bu₄NCl) with χ(Bu₄NCl) = mol fraction

[Bu₄NCl]/([Bu₄NCl]þ[LPd^IVCl₄]) and Δδ(H⁵) = δ(H⁵)_max- δ(H⁵).

A maximum at χ(Bu₄NCl) = 0.48 is consistent with a 1:1 binding stoichio-

metry in complex 7 Cl^-.

3

Scheme 3

Figure 5. T1-relaxation studies of 7 and 7 Cl^-.

3

respective Cl₂-chlorination (Table 2). Furthermore, the abun-

dance of side products from allylic chlorination (Table 1, entry 3)

and vinylic chlorinations (Table 1, entries 4-and 5) was very

similar to respective abundances in Cl₂-chlorinations of cyclo-

hexene, styrene, and stilbene (Table 2). This suggests the

intermediacy of identical chloronium intermediates that deter-

mine stereochemistry and product distributions in chlorinations

with Cl₂and in chlorinations with 7 Cl^-.

3

heating to 60 °C. Methane was converted to methyl chloride,

Mechanism of Alkene Chlorinations. General Considera-

tions. We proceeded to elucidate the circumstances of Cl^þ-

albeit in extremely low yield (entry 12). Methyl Grignard (entry

transfer from LPd^IVCl₄(7 and 7 Cl^-) to alkenes and alkynes.

13) was oxidized to CH₃Cl and CH₃CH₃, and bromide reacted

with 7 to Br₂(entry 14).

3

Three plausible reaction manifolds were considered (Scheme 4):

In pathway 1, a direct, inner-sphere transfer of Cl^þwould occur

directly between LPd^IVCl₄and an alkene or alkyne. Neutral

species such as TlCl₃,²⁶PbCl₄,²⁷PCl₅,^26b,28SbCl₅,^26b,29

Alkene/alkyne dichlorinations in entries 2 and 3 showed

predominant trans-selectivities in their 1,2-addition products,

which matched selectivities obtained from electrophilic chlorina-

tions with Cl₂of identical substrates (Table 2). Equally, the

threo/erythro ratio (1.53/1) for 1,2-dichloro-1,2-dipheny-

32

MoOCl₄,³⁰CrO₂Cl₂,³¹or PhICl₂have previously been in-

voked in ionic Cl^þ-transfers to alkenes as reagents. Alternatively,

LPd^IVCl₄may undergo concerted Cl-Cl reductive elimination

(pathway 2) followed by Cl₂-chlorination of the alkene. Elimination

lethanes formed in the reaction of 7 Cl^-and stilbene (Table 1,

3

entry 5) was nearly identical to the threo/erythro ratio from the

1835

dx.doi.org/10.1021/ja107342b |J. Am. Chem. Soc. 2011, 133, 1832–1848

Journal of the American Chemical Society

ARTICLE

Table 1. Reactions of LPd^IVCl₄with Organic Substrates

^aReaction conditions: 10-20 mM substrate; 150 mM Bu₄NCl/CDCl₃; 1.05 equiv of (freshly prepared) 7 unless otherwise noted. ^bIdentiﬁed via ¹H

NMR spectroscopy chemical shift analysis and compared to either authentic samples or published spectra (for references, see the Supporting

Information). ^cDetermined via ¹H NMR integration using 1,1,2,2-tetrachloroethane as an internal standard. ^dHeadspace of a resealable J. Young NMR

tube was purged with 1 atm of CH₄prior to the reaction. ^eVolatiles were collected via ﬂask-to-ﬂask vacuum transfer using two J. Young NMR-tubes; ¹H

NMR spectra were recorded and compared to the initial crude product mixture. ^fIn THF-d₈with a suspension of 7 under sonication in the absence of

Bu₄NCl or Bu₄NBF₄. ^g5 equiv of Bu₄NBr was reacted with 7. ^hIdentiﬁed and quantiﬁed after complete consumption of 7 by trapping with cyclohexene

and quantifying the formed product 1,2-dibromocyclohexane and 3-bromocyclohex-1-ene via ¹H NMR spectroscopy.

of Cl₂is a common feature in inorganic complexes of Pd(IV)³³

such as (bipy)PdCl₄,¹⁷Pd(NH₃)₂Cl₄,³⁴Pd(py)₂Cl₄,³⁴[Pd(NH₃)₄-

Cl₂]Cl₂,³⁵and Et₄N[PdCl₅(PPr₃)]³⁶even though the precise

mechanism of chlorine loss in these complexes is unknown.³⁷In

the ionic pathway 3, an initial loss of Cl^-from LPd^IVCl₄would

form ion pair A (from 7) or zwitterionic analogue B (from

7 Cl^-; with intact Cl^-H-C hydrogen bonds on the ligand

agent with the alkene (pathway 3a). Literature precedence for

anion dissociation from octahedral Pd^IVcomplexes followed by

reductive C-O and C-F eliminations was reported by Sanford^6a

and Ritter.³⁸Goldberg has established sequential ligand dissocia-

tions followed by reductive eliminations in octahedral platinum

complexes.³⁹Alternatively, we considered the possibility of a

nucleophilic attack of the alkene to an electron-deficient Cl^δþ

-

3

3 3 3

backbone). A reductive Cl₂elimination from A or B may then

form Pd^II-cation C while Cl₂would act as the actual chlorinating

ligand in intermediates A or B (pathway 3b). This would be

reminiscent of reported S_N2-attacks of external nucleophiles on

1836

dx.doi.org/10.1021/ja107342b |J. Am. Chem. Soc. 2011, 133, 1832–1848

Journal of the American Chemical Society

ARTICLE

Table 2. Stereo- and Regioselectivity of Chlorination of

Alkenes from Reactions with 7 as well as with Cl₂

Figure 6. Relative rates (k_rel) for chlorination of 1-hexene and 3-hexene

with 7 Cl^-, Cl₂, and Cl₂/Bu₄NCl.

3

so far, calculations suggested that such species are energetically

feasible.⁴²It should be noted that chloropalladations of alkenes

or alkynes are known with Pd^II-chloride complexes.⁴³

Competition Experiments. In probing for the possibility of

in situ formation of Cl₂as the actual chlorinating agent in

pathways 2 and 3a, (Scheme 4), we conducted a competition

^aSolvent and salt additive identical to those described in Table 1. ^b0.8

equiv of Cl₂(saturated solution in CDCl₃) was rapidly added via syringe

to a premixed solution of substrate in CDCl₃/Bu₄NCl. ^cData from

Poutsma, M. L. J. Am. Chem. Soc. 1965, 87, 2161-2171. ^dReaction

performed neat at 25 °C with the exclusion of light and in the presence of

oxygen to eliminate the radical pathway. ^eData from Iskra, J.; Stavber, S.;

Zupan, M. Chem. Commun. 2003, 2496-2497. ^fData from Britsun, V.

N.; Serguchev, Y. A.; Ganushchak, N. I. Russ. J. Gen. Chem. 2001, 71,

261-264. ^g(Z)-product was below detection limit by ¹H NMR

spectroscopy. ^hData from Heasley, G. E.; Codding, C.; Sheehy, J.;

Gering, K.; Heasley, V. L.; Shellhamer, D. F.; Rempel, T. J. Org. Chem.

1985, 50, 1773-1776. ⁱReaction performed in CH₂Cl₂.

experiment in which 7 Cl^-was reacted with a 10 equiv/10 equiv

3

mixture of 1-hexene and 3-trans-hexene in CDCl₃/Bu₄NCl; the

resulting product ratio of 3,4-trans-dichlorohexane (8) and 1,2-

dichlorohexane (9) was 6.0:1 (Figure 6). In contrast, we

obtained a 40.1:1 ratio of 8 and 9 with Cl₂as the chlorinating

agent under otherwise identical conditions, which is close to

Poutsma’s ratio of 52:1 for Cl₂-chlorination of a neat mixture of

1-butene and 2-trans-butene.⁴⁴The apparent mismatch between

7 Cl^-and Cl₂in their internal versus terminal selectivities rules

3

out the possibility of in situ formed Cl₂from 7 Cl^-(pathways 2

3

and 3a). Furthermore, the 6.0:1 bias displayed by 7 Cl^-in the

3

Scheme 4

functionalization of the internal over the terminal hexane is at

odds with the formation of η²-alkene complex C from B prior to

chlorination in pathway 3c; metal complexes with terminal

alkenes are generally more stable than complexes with internal

alkenes,⁴⁵and functionalizations of alkenes by Pd^IIshow higher

rates for less substituted olefins.^1,46

Kinetic Experiments. To further elucidate the mechanism of

Pd^IV-chlorinations, we conducted a series of individual kinetic

measurements for the stoichiometric chlorinations of cyclo-

hexene (cy) (Table 3, entries 1-24), styrene (styr) (entries

25-27), and hexamethylbenzene (HMB) (entry 28) as sub-

strates in the reaction with 7 or 7 Cl^-. The relative abundances

3

of these two LPd^IVCl₄species were controlled in the equilibrium

7 þ Cl^-_freeh 7 Cl^-by adjusting [Cl^-]_free. Experimentally, we

3

could only follow the disappearance of [7] and [7 Cl^-] as a

3

1

weighed average by H NMR spectroscopy, and therefore we

deliberately adjusted [Cl^-]_freeto levels that ensured a predomi-

nant abundance of either 7 or 7 Cl^-. With relatively high

3

chloride concentrations of [Cl^-]_freeg 139 mM (entries 1-20

and 25-28), the major Pd^IVspecies in solution was 7 Cl^-

3

(χ₇₃_Cl-g 0.953); with relatively low chloride concentrations of

[Cl^-]_freee 0.64 mM, compound 7 was predominant (χ₇g

0.920, entries 21-24). Upon inspection of entries 1-20

(reactions of cy with 7 Cl^-) and entries 21-24 (reactions of

3

cy with 7), it is apparent that measured values for k_obsare overall

higher for 7 Cl^-than for 7 over a wide range of [cy]. At constant

3

free chloride concentration ([Cl^-]_free= 143 mM), higher con-

centrations of cyclohexene produce larger values for k_obs(entries

1-10). However, this relationship is not linear, and a plateau just

above k_obs= 20 ꢀ 10^-5s^-1is found for [cy] g 300 mM (entries

1 and 2). A [cy] versus k_obsplot for entries 2-10 represents this

saturation kinetics behavior graphically (Figure 3, top).⁴⁷Qua-

litatively, styrene displays similar plateau characteristics in its

sp³-carbons of Pd^IV-C^δþbonds.^2dMoreover, the transfer of

Cl^þto alkenes has been reported for cationic species such

þ 40

þ 28

as SbCl₄

,

PCl₄

,

and Cu^IICl^þ.⁴¹Last but not least, we

considered the formation of Pd^IV(η²-alkene) complex D followed

by chloropalladation (pathway 3c). Although spectroscopically

identified Pd^IVπ-complexes have not been reported in the literature

1837

dx.doi.org/10.1021/ja107342b |J. Am. Chem. Soc. 2011, 133, 1832–1848

Journal of the American Chemical Society

ARTICLE

Table 3. Observed Rates for the Chlorination of Cyclohex-

ene, Styrene, and Hexamethylbenzene with 7 and 7 Cl^-at

3

Various Substrate and Chloride Concentrations

[substrate] [Cl^-]_freemol fraction 7 Cl^-k_obsꢀ 10^-5

3

c

d

entry substrate^a

(mM)

(mM)^b

(χ₇₃_Cl-

)

(s^-1

)

1

2

cy

608

143

151

313

688

0.47

0.64

0.34

0.43

166

139

0.954

0.957

0.981

0.993

0.060

0.080

0.045

0.055

0.962

0.953

20.2 ( 0.3

20.2 ( 0.2

18.8 ( 0.2

17.9 ( 0.3

16.8 ( 0.3

16.4 ( 0.3

15.8 ( 0.5

14.9 ( 0.5

11.9 ( 0.6

10.6 ( 0.9

16.6 ( 0.1

19.5 ( 0.7

11.8 ( 0.3

8.8 ( 0.2

7.6 ( 0.3

6.9 ( 0.5

5.2 ( 0.8

5.9 ( 0.6

4.5 ( 0.4

3.1 ( 0.2

5.1 ( 0.2

4.6 ( 0.1

3.1 ( 0.1

2.8 ( 0.1

22.5 ( 0.4

23.3 ( 0.5

12.7 ( 0.2

22.2 ( 0.8

cy

300

3

cy

15.0

11.3

7.20

3.60

1.50

0.58

0.29

0.14

3.43

120

4

cy

5

cy

6

cy

Figure 7. Saturation kinetics plots for the chlorination of cyclohexene

with 7 (bottom line, dashed) and 7 Cl^-(top line, contiguous).

7

cy

3

8

cy

9

cy

follows a saturation kinetics trend similar to 7 Cl^-; however, the

3

k_obsplateau value of approximately 5 ꢀ 10^-5s^-1is about 5 times

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

cy

lower for 7 (Figure 7, bottom line) than for 7 Cl^-.⁴⁸

cy

3

While the addition of chloride ions can increase k_obsthrough

cy

shifting the equilibrium 7 þ Cl^-_freeh 7 Cl^-toward the side of

3

cy

3.46

2.77

0.44

0.37

0.32

0.28

0.25

0.23

190

the more reactive 7 Cl^-, a large excess of [Cl^-]_freehas a

3

cy

detrimental outcome on k_obs. This eﬀect reveals itself in entries 6

and 13 with virtually identical cyclohexene concentration ([cy] =

3.60 mM and 3.46 mM); a higher free chloride concentration in

entry 13 decreases the observed rate constant ([Cl^-]_free= 313

mM, k_obs= 11.8 ꢀ 10^-5s^-1) as compared to an environment

cy

with a lower free chloride concentration in entry 6 ([Cl^-]_free

=

cy

143 mM, k_obs= 16.4 ꢀ 10^-5s^-1). The rate suppression in entry

13 is observed despite a slight increase in the relative abundance

of the more reactive species 7 Cl^-in entry 13 (χ_{7 Cl-}g 0.981)

cy

16.7

9.07

5.36

300

3

as compared to entry 6 (χ₇₃_Cl-g 0.954). Increasing [Cl^-]_freeto

688 mM depressed k_obseven further (entries 15-20), and the

emergence of an inverse relationship between [Cl^-]_freeand k_obs

becomes apparent when comparing entries 15 and 20.

cy

styr

HMB

210

Quantitative Kinetic Analysis. Taken together, the saturation

kinetics in [cy] versus k_obs, the chloride-dependent depressions of

k_obsat high [Cl^-]_freeand low [cy], as well as the chemoselective

chlorination of internal versus terminal alkenes are at odds with

pathways 1, 2, 3a, and 3c (Scheme 4), but consistent with pathway 3b

14.4

38.9

^aSubstrates in the table are listed as cy = cyclohexene, styr = styrene, and

HMB = hexamethylbenzene. ^bFor entries 1-20 and 25-28, the

concentration of free chloride ([Cl^-]_free) was determined by subtracting

for both Pd^IVcomplexes 7 and 7 Cl^-. Next, we took on the task of

the amount of hydrogen-bonded Cl^-(within 7 Cl^-) from the total

3

extracting quantitative data pertaining to the elementary steps in

pathway 3b. First, we describe the overall disappearance of the total

chloride concentration [Cl^-]_totalin the system (from added Bu₄NCl).

In entries 15-18, small amounts of chloride were present as a 1:1

byproduct from allylic chlorination of cyclohexene. [Cl^-]_totalwas then

derived from the ¹H NMR integral of RRC(Cl)H of 3-chloro-cyclohex-

amount of Pd^IV([Pd^IV] = [7] þ [7 Cl^-]) as:

3

ꢀ

ꢁ

d Pd^IV

ꢀ

ꢁ

¼ - k_obsPd^IV

ð1Þ

ene. The hydrogen-bonded chloride portion within 7 Cl^-was sub-

3

dt

tracted from [Cl^-]_totalto determine [Cl^-]_free

.

^cDerived from 7 þ

Cl^-_freeh 7 Cl^-(K_eq= 139.2 M^-1); see Supporting Information eq 12

Our analysis is based on a rapid pre-equilibrium between 7 and

3

7 Cl^-in the presence of free chloride ions (7 þ Cl^-

h

and the Supporting Information chapter “Kinetic Analysis and Rate Law

free

7 Cl^-).⁴⁹Furthermore, we deﬁne k₇and k_{7 Cl-}as ﬁrst-order

3

Derivation” for details. ^dRates for disappearances of 7 and 7 Cl^-were

3

1

3

observed rate constants for the disappearance of 7 and 7 Cl^-(eq

determined by H NMR spectroscopy. Errors were calculated using

3

ANOVA regression analysis to a 95% conﬁdence interval and given as

the standard error for each measurement

2a,b). Essentially, k₇and k₇₃_Cl-mimic the role of observable rate

constants, even though our experimental technique of data

collection (¹H NMR spectroscopy) only provides for average

reaction with 7 Cl^-(entries 25-27), and k_obslevels out at

3

signals of 7 and 7 Cl^-. Combining eq 1 and eq 2a,b yields eq 3.

approximately 23 ꢀ 10^-5s^-1

.

3

Adding extra Pd^IIcomplex 6 (LPd^IICl₂) to the solutions of

cyclohexene or styrene did not aﬀect their k_obsvalues, which

excludes that Pd^II(alkene) complexes may play a role as a reactive

intermediate during the reaction. We generated only one data

point for the chlorination of hexamethylbenzene (HMB, entry

d½7ꢁ

dt

d½7 Cl^-ꢁ

-

3

-

¼ - k₇½7ꢁ and

¼ - k_{7 Cl}½7 Cl ꢁ ð2a,bÞ

3

dt

ꢀ

ꢁ

d Pd^IV

ꢂ

ꢃ

-

¼ - k₇½7ꢁ þ k_{7 Cl}½7 Cl ꢁ

ð3Þ

3

dt

28) with 7 Cl^-, and it matches the k_obsplateau values of

3

cyclohexene and styrene [k_obs(HMB) = 22.2 ꢀ 10^-5s^-1)].

Qualitatively, the reaction of 7 (entries 21-24) with cyclohexene

Combining eqs 1 and 3 and expressing the abundance of [7]

and [7 Cl^-] within [Pd^IV] through individual mole fractions χ₇

3

1838

dx.doi.org/10.1021/ja107342b |J. Am. Chem. Soc. 2011, 133, 1832–1848

Journal of the American Chemical Society

ARTICLE

Scheme 5

“steady-state” terms for the interplay of 7 and A (ﬁrst addend)

as well as 7 Cl^-and B (second addend) (for detailed deriv-

3

ations, see the Supporting Information).

A

k₁½cyꢁ

Cl_f^-_reeþ ½cyꢁ

!

k_obs¼ χ₇

A

ꢀ

ꢁ

k _{- 1}

A

k₂

B

k₁½cyꢁ

!

-

þ χ₇₃_Cl

ð5Þ

B

ꢀ

ꢁ

k _{- 1}

Cl_f^-_reeþ ½cyꢁ

B

k₂

Four limiting scenarios can be discussed in the context of eq 5:

(i) [Cl^-]_free> [Pd^IV], k_-1^B/k₂[Cl^-]_free, [cy]. In this case,

B

one obtains χ₇₃_Cl-≈ 1, χ₇≈ 0, and k_obsis reduced to

the rate constant k₁^Bfor chloride dissociation from 7 Cl^-

3

(eq 6). This scenario reﬂects the plateau level for the satu-

ration kinetics of 7 Cl^-(Figure 3, top), in which case the

3

initial ionization of 7 Cl^-is rate determining.

3

B

k_obs¼ k₁

ð6Þ

(ii) [Cl^-]_free, [Pd^IV], k_-1^A/k₂[Cl^-]_free, [cy]. In this

case, one obtains χ_{7 Cl-}≈ 0, χ₇≈ 1, and k_obsis reduced

A

to the rate constant³k₁for chloride dissociation from 7

A

(eq 7). This scenario reﬂects the plateau level for the

saturation kinetics of 7 (Figure 3, bottom) in which case

the initial ionization of 7 is rate determining.

A

k_obs¼ k₁

ð7Þ

(iii) [Cl^-]_free. [Pd^IV], k_-1^B/k₂[Cl^-]_free. [cy]. In this

case, one obtains χ₇₃_Cl-≈ 1, χ₇≈ 0, and k_obscan be

expressed according to eq 8. In this case, k_obsis propor-

tional to [cy] and inversely proportional to [Cl^-]_free. In

B

and χ₇._Cl-([7] = χ₇[Pd^IV] and [7 Cl^-] = χ₇._Cl-[Pd^IV])

provides eq 4.

chemical terms, this refers to a scenario in which 7 Cl^-

3

and B are eﬀectively part of a pre-equilibrium and the rate-

determining step is the transfer of Cl^þto cyclohexene.

-

7 Cl^-

k_obs¼ χ₇k₇þ χ

k_{7 Cl}

ð4Þ

3

B

k₁^Bk₂

½cyꢁ

Equation 4 manifests the additive approach to k_obs, which is

ꢀ

ꢁ

k_obs

¼

ð8Þ

separated into individual reactivities of 7 and 7 Cl^-as well as

B

k _{- 1}Cl^-_free

3

their respective abundances. This simple concept necessitates

that once intermediate A is formed from 7, it cannot directly

equilibrate with B on the time scale of the Cl^þ-transfer from A to

cyclohexene.⁵⁰Experimental evidence will be presented (vide

infra) in support of such slow ion pair dynamics.

Experimentally determined rates of disappearance of

7 Cl^-in the presence of very low concentrations of

3

cyclohexene ([cy] = 0.14-0.44 mM, [Cl^-]_free= 143

mM, 688 mM) (entries 10, 15-20, Table 3) produced

a linear relationship when plotting ln(k_obs) versus

ln([cy]/[Cl^-]_free) (Figure 8). A slope of 0.97 conﬁrmed

Scheme 5 lays out the conceptual framework for multistep

reactions including absolute rate constants for the elementary

steps. It is based on the equilibrium between 7 and 7 Cl^-.

the anticipated ﬁrst-order dependence on [cy]/[Cl^-]_free

.

3

Chloride dissociation from the metal centers in 7 and 7 Cl^-is

3

A

B

A

B

The inverse rate dependence on [Cl^-]_freecan be ex-

plained in chemical terms through a rate-determining

transfer of Cl^þfrom B to cyclohexene; the pre-equilib-

described through k₁and k₁, while k_-1and k_-1are rate

constants for the microscopic reverse chloride addition to

intermediates A and B. Furthermore, bimolecular Cl^þ-trans-

fers from A and B to cyclohexene to form E⁰are expressed

rium 7 Cl^-h B þ Cl^-

is shifted to the left by

free

3

A

B

increasing [Cl^-]_free. The switch in rate-determining step

as compared to scenario (i) is brought about by a

decrease of the [cy]/[Cl^-]_freeratio.

through k₂and k₂. No intermediates were detected in the

reactions of 7 and 7 Cl^-with cyclohexene (nor with any other

3

substrate), and we apply a steady-state model for A and B. On

the basis of this approach, eq 4 turns into a “bimodal” ex-

pression (eq 5) whose two components constitute individual

(iv) [Cl^-]_free, [Pd^IV], k_-1^A/k₂[Cl^-]_free. [cy]. We were

A

not able to explore this domain as practical concentrations of

1839

dx.doi.org/10.1021/ja107342b |J. Am. Chem. Soc. 2011, 133, 1832–1848

Journal of the American Chemical Society

ARTICLE

10^-5s^-1(ΔG_A^q= þ23.4 kcal mol^-1) (Figure 9), k₁^B= 21.0 ꢀ

A

q

10^-5s^-1(ΔG_B^q= þ22.5 kcal mol^-1), k₂/k_-1^A= 0.10 (δΔG_A

=

þ1.4 kcal mol^-1), and k₂/k_-1= 130 (δΔG_B= -2.9 kcal

B

q

0

mol^-1).⁵³Together with the already established value ΔG_Cl-

=

-2.9 kcal mol^-1for the equilibrium 7 þ Cl^-_freeh 7 Cl^-, these

3

values are the basis for an energy diagram, featuring a lower

barrier for Pd^IV-Cl heterolysis in 7 Cl^-than in 7 (ΔG_A

-

q

3

ΔG_B^q= þ0.9 kcal/mol) (Figure 9). It also underscores the rather

small energetic differences between TS₁and TS_Aas well as TS₂

and TS_B, allowing for the observation of [cy]-dependent satura-

tion kinetics. The interconversion of 7 and 7 Cl^-is too rapid on

3

the NMR time scale to spectroscopically distinguish between

them as a result of a low energetic barrier between them

[ΔG(exchange) e þ14.0 kcal mol^-1for exchange of hydro-

gen-bonded chloride in 7 Cl^-with external Cl^-via forma-

free

3

tion of intermediate 7].⁵⁴

A

B

A

B

Values for k₁, k₁, k₂/k_-1^A, and k₂/k_-1^Bfrom cks-simula-

tions were then entered into eq 5 to generate simulation plots for

k_obsat particular chloride concentrations (Figure 7). For

[Cl^-]_free= 143 mM, the k_obssimulation plot is represented with

a contiguous line (Figure 7, top); in this case, the major Pd^IV

Figure 8. Rates of chlorination of cyclohexene with 7 Cl^-(from 7 and

3

Bu₄NCl) at extremely low [cy]:[Cl^-]_freeratios (1:1000-1:3000) with a

slope ∼1 in the ln/ln plot.

component in solution was 7 Cl^-. For [Cl^-]_free= 0.43, the k_obs

3

simulation plot is represented with a dashed line (Figure 7,

bottom); in this case, the major Pd^IVcomponent in solution was

7. Both simulation plots are in good agreement with experi-

mental k_obsdata in Figure 7.

Solvent Effect. Solvent effects were investigated in the reac-

tions of cyclohexene with 7 in CDCl₃/Bu₄NBF₄and 4 in DMF-

d₇.⁵⁵While not mapping out a complete kinetic profile for 4 in

DMF-d₇, we were able to obtain a pseudofirst-order rate constant

[k_obs(DMF-d₇) = (214 ( 13) ꢀ 10^-5s^-1] with respect to 4

(23.6 mM), while the [cy] = 472 mM. At such high cyclohexene

concentration, it is reasonable to assume that k_obs(DMF-d₇)

solely reflects the chloride dissociation step (k₁) from the metal

and k_obs(DMF-d₇) = k₁(DMF-d₇). In comparing this value to k₁^A

=

4.2 ꢀ 10^-5s^-1(measured in CDCl₃/Bu₄NBF₄in the absence

Cl^-), we obtain the relative rate k_rel= k₁(DMF-d₇):k₁^A= 51:1.

Reaction with Aromatic Substrates. Various electron-rich

aromatic substrates reacted with 7 Cl^-, causing ring and

3

benzylic chlorinations in good yields (Table 1, entries 6-8, 10,

11), while p-xylene was inert even upon heating to 60 °C for 10 h

(entry 9). 4-Methylanisole was chlorinated in its 2- and 4-posi-

tions yet not in the 3-position (entry 11), 1,4-dimethoxybenzene

was smoothly converted to 2-chloro-1,4-dimethoxybenzene,

hexamethylbenzene underwent benzylic chlorination (entry 6),

while mesitylene (entry 8) and durene (entry 7) were function-

alized in both ring and benzylic positions. We followed the reac-

Figure 9. Energy diagram for the reactions of LPd^IVCl₄(7 and 7 Cl^-)

3

with cyclohexene to form chloronium intermediate E via LPd^IVCl₃_q

þ

intermediates A and B. Represented are enthalpies of ionization (ΔG₁

q

and ΔG₂) as well as the enthalpic diﬀerences between Cl^þtransfer from

q

LPd^IVCl₃^þto cyclohexene and Cl^-return to LPd^IVCl₃^þ(δΔG_A^qand δΔG_B).

tion of hexamethylbenzene and 7 Cl^-by ¹H NMR spectroscopy

3

and found a substrate-independent rate law d[7 Cl^-]/dt =

cyclohexene would have to be signiﬁcantly lower than in

scenario (iii), in which case [cy] had been only slightly

above our ¹H NMR detection limit.

3

k_obs[7 Cl^-] with k_obs=22.2ꢀ10^-5s^-1(CDCl₃, 25 °C, Table 2),

3

which matched the rate constant for chloride dissociation from

7 Cl^-(k₁= 21.0 ꢀ 10^-5s^-1) reasonably well. In contrast,

B

3

A

B

bimolecular rate laws d[7 Cl^-]/dt = k_bi[7 Cl^-] [substrate])

Simulations. To quantitatively determine values of k₁, k₁,

3

A

B

as well as (k₂/k_-1^A) and (k₂/k_-1^B), we simulated experimental

results using cks-software,⁵¹which uses a stochastic algorithm⁵²

and therefore serves for an analytical treatment of Scheme 5 that

is independent of the validity of eq 5 and eqs 6-8 from scenarios

(i), (ii), and (iii). We iteratively modeled the disappearances

were observed for 4-methylanisole (k_bi= 2.1ꢀ10^-3M^-1s^-1

)

and dimethoxybenzene (k_bi= 6.1 ꢀ 10^-3M^-1s^-1).

Reaction of 7 Cl^-with Methane and Side-Chain Chlo-

3

rination. We were able react CH₄with 7 Cl^-at elevated

3

temperatures to produce CH₃Cl in small quantities (entry 12,

Table 1). While a relatively low CH₄pressure (p_CH4= 1 atm)

may have prevented higher yields, we found predominant

of starting materials 7 and 7 Cl^-with the concentrations and

3

conditions provided in Table 3 (entries 1, 3-10, 15, 16, 17, and

18). We were able to obtain a reasonable and comprehensive fit

for all entries with the adjustable parameters set to: k₁^A= 4.2 ꢀ

intramolecular chlorination of the C₁₄H₂₉-chain of 7 Cl^-.

3

Reacting 7 Cl^-in the absence of organic substrates with excess

3

1840

dx.doi.org/10.1021/ja107342b |J. Am. Chem. Soc. 2011, 133, 1832–1848

Journal of the American Chemical Society

ARTICLE

Figure 10. Mass spectrum of chlorination products (LPd^IVCl₃^þ) from the reaction of 7 with excess Cl₂. Up to six H/Cl substitution events within the

ligand could be identiﬁed.

Cl₂produced multiple chlorination events in the side chain, when

exposed to 1 atm Cl₂at 25 °C for 16 h with the exclusion of light,

presumably via a catalytic Pd^II/Pd^IVcycle. The occurrence of up to six

substitutions of C-H bonds with C-Cl bonds in LPd^IVCl₄was

confirmed by MS(ESI) spectroscopy (Figure 10). A control experi-

ment using hexane in CDCl₃/Bu₄NCl solution and Cl₂showed no

aliphatic chlorination products over the same period.

mirrored by the absence of any isolable tetraoxygenated Pd^IV

complexes in the literature. The X-ray crystal structure of 4

reveals an isomorphous relationship to the tub-shaped bis-

NHC chelate in square planar Pd^IIprecursor 1.^16bInterest-

ingly, the Pd-C^carbenebonds were lengthened by 0.046 and

0.049 Å relative to the same bonds in 1, while Pd-Cl^eqbonds

on the opposite site were virtually unaltered in their bond

lengths (Δ < 0.004 Å). We attribute the Pd-C^carbenebond

lengthening to a decrease in π-backbonding.⁵⁶The existence

of d-π* interactions is in line with a growing literature

consensus on the general viability of π-backbonding in

metal-NHC complexes.^20,57Furthermore, NHC ligands

are known to exert a strong trans-influence on metal-ligand

bonds in the trans position;⁵⁸inspection of Pd-Cl bond

lengths in 4 corroborates this trend with longer equatorial

in-plane Pd-Cl^eqbonds as compared to shorter, unaffec-

ted axial Pd-Cl^axbonds (Δ_average= 0.066 Å). The longer

Pd-C^carbenebonds in 4 relative to 1 (diminished back-

bonding) paired with virtually no bond length differences

for Pd-Cl^eqin both 4 and 1 (unaltered trans-influence) under-

scores that π-backbonding and trans-influence are independent

effects that have no significant synergistic relationship. This conclu-

sion is consistent with the premise that trans-influences are trans-

mitted primarily through the σ-framework of a coordination

compound.⁵⁹

Reaction with Bromide. In an attempt to initiate Cl/Br

ligand substitution in LPd^IVCl₄, we reacted a suspension of 7 (1

equiv) in CDCl₃with Bu₄NBr (5 equiv). Within seconds, the yellow

1

solid decolorized and dissolved. An H NMR analysis showed the

quantitative formation of LPd^IIBr₂complex 6. The oxidation of Br^-

gave rise to Br₂, which was confirmed by adding cyclohexene (1

equiv) and identifying the adducts 1,2-dibromocyclohexane and

3-bromocyclohexene by ¹H NMR and ESI-MS spectroscopy.

’ DISCUSSION

Synthesis and Properties of LPd^IVCl₄Complexes. The

facile and clean conversion of LPd^IICl₂to LPd^IVCl₄was

accomplished with Cl₂or PhICl₂as oxidants. This method

has been used in the past to synthesize tetrachlorides (bipy)-

PdCl₄¹⁷and (en)PdCl₄and (py)₂PdCl₄and [(NH₃)₄PdCl₂]Cl₂.¹⁸

In contrast, the bromination of LPd^IIBr₂complex 2 pro-

duced a new material whose formation was apparently driven

by precipitation rather than by intrinsic driving forces

because we observed facile Br₂loss to regain 2. The reaction

of LPd^IVCl₄complex 7 with Br^-produced Br₂and LPd^IICl₂

complex 6 rather than engaging in Br^-/Cl^-ligand substitu-

tions (entry 14, Table 1). Given that the formation of bis-

NHC-Pd^IVbromides seems to be thermodynamically un-

favorable in solution, we were able to exploit these properties

by Cl₂oxidation of Pd^IIdibromide 2 (which is synthetically

more easily accessible than Pd^IIdichloride 2) to LPd^IVCl₄

product 4 with complete loss of Br^-ligands. Our attempts

to synthesize oxygenated complexes of the type LPd^IV-

(OR)_nCl_m(m þ n = 4) met with failure when a variety of

peroxides failed to convert starting materials. This failure is

A unique structural feature of CH₂-bridged bis-NHC complexes is

the nearly collinear alignment of the three C-H⁵and C-H^exobonds

(Figure 1; see C25-H25, C35-H35, C11-H11A). Through a

series of experiments including ¹H NMR titrations of Cl^-and T1

relaxation measurements, we established that these three C-Hbonds

could act as hydrogen-bond donors toward chloride ions to form

supramolecular adduct 7 Cl^-.

3

Halide binding to purely organic receptors has been reported. For

example, imidazolium cations are prone to form hydrogen bonds with

halides;⁶⁰Flood found that neutral C-H hydrogen-bond donors

from macrocyclic triazolophanes were able to hydrogen bond Cl^-to a

total of eight C-H bonds of the host with binding enthalpies in

CD₂Cl₂of up to -9.6kcalmol^-1with an average of -1.2kcalmol^-1

1841

dx.doi.org/10.1021/ja107342b |J. Am. Chem. Soc. 2011, 133, 1832–1848

Journal of the American Chemical Society

ARTICLE

per Cl^-H-C bond.⁶¹In comparison, the trifurcated hydrogen-

Mechanism of Alkene Chlorinations. We considered three

fundamental pathways that included direct bimolecular Cl^þ-transfer

from LPd^IVCl₄to the olefin (pathway 1, Scheme 4), loss of Cl₂

from LPd^IVCl₄(for example, via concerted reductive elim-

ination) and subsequent alkene chlorination by Cl₂(pathway

2), and, last, Cl^--dissociation from LPd^IVCl₄to form cation

LPd^IVCl₃^þ(pathway 3). For the latter, we considered reductive

elimination of Cl₂from LPd^IVCl₃^þ(pathway 3a); alternatively,

3 3 3

bonding interaction in 7 Cl^-(ΔG⁰_Cl-=-2.9 kcal mol^-1) amounts

3

to -0.97 kcal mol^-1per hydrogen bond. However, it should be

pointed out that the binding interaction of 7 and Cl^-could only be

determined in the presence of Bu₄NBF₄to dissolve 7in the ﬁrst place.

It is therefore conceivable that 7 and BF₄^-anions engage in weak

hydrogen-bonding contacts by themselves and addition of Cl^-simply

causes a binding competition. Therefore, ΔG⁰_Cl-may constitute a

relative binding aﬃnity, and its value of -2.9 kcal mol^-1would

therefore have to be regarded as a conservative lower limit. Interest-

þ

a direct Cl^þ-transfer from LPd^IVCl₃to the alkene was con-

templated (pathway 3b). A third plausible variation was seen

in the involvement of π-coordinated intermediates LPd^IVCl₃-

(η²-alkene)^þ(D) prior to intra- or intermolecular chloropalla-

dation (pathway 3c).

ingly, this hydrogen-bonding contact in the periphery of 7 Cl^-aﬀects

3

its chemical reactivity in comparison to the parent compound 7.

Related transition metal complexes capable of hydrogen

bonding anions to the periphery of the ligand scaﬀold have been

previously published by Beauchamp,⁶²Walter,⁶³Romeo,⁶⁴as

well as Crabtree and Eisenstein⁶⁵in compounds 10-13; these

Saturation Kinetics. In determining the relationship be-

tween alkene concentration and k_obs, we investigated the

disappearance of LPd^IVCl₄during its reaction with cyclohex-

ene (Table 3). In controlling the concentration of free

chloride⁷⁰(Cl^-_free), we were able to set the concentrations

interactions include Cl^-H-N hydrogen bonds (10-12)

3 3 3

and “carbon-based” F₃BF H-C hydrogen bonds in iridium

3 3 3

of 7 and 7 Cl^-and arrive at predominant abundances of

complex 13. However, complexes 10-13 are all based on

cationic metal centers, and hydrogen bonds merely supplement

underlying electrostatic interactions within each ion pair. To the

3

either species 7 (χ g 0.92) at low [Cl^-]_freeor 7 Cl^-(χ g

3

0.954) at high [Cl^-]_free. Saturation kinetics behavior was

independently found for both species at different plateau

levels (Figure 7). Such plateau domains for [cy]-independent

k_obsvalues rule out a strictly bimolecular process according

to pathway 1 (Scheme 4) but would be consistent with step-

wise pathways 2 and 3. At sufficiently high [cy]/[Cl^-]_free

best of our knowledge, 7 Cl^-constitutes the ﬁrst reported Cl^--

3

adduct to a neutral transition metal complex, with possible

electrostatic interactions only to the countercation Bu₄N^þ.

ratios g 2, we observe a unimolecular rate law d[7 Cl^-]/dt =

3

k_obs[7 Cl^-], and the rate is independent of [Cl^-]_free. How-

3

ever, at very low ratios [cy]/[Cl^-]_free< 1/1500, we were able

to manifest a domain with d[7 Cl^-]/dt µ 1/[Cl^-]_free. Rate

3

suppressions by [Cl^-]_freewould not be expected for pathway

2 but are supportive of pathway 3 with a rate-determining

loss of metal-bound chloride in the initial step followed by

fast Cl^þ-transfer to cyclohexene. At low [cy]/[Cl^-]_freeratios,

we are promoting the addition of Cl^-_freeto the metal center

in B at the expense of the forward reaction of B and

cyclohexene. In other words, the initial dissociation of 7 Cl^-

3

into B and Cl^-becomes reversible on merits of relative

free

abundances of Cl^-

over cy. Comparing unimolecular

free

plateau rates of compound 7 in CDCl₃/Bu₄NBF₄(d[7]/

dt = k_obs[7]) with unimolecular conversions of 4 in DMF-

d₇,⁷¹we found a 51-fold rate acceleration in the more polar

solvent.⁷²Again, such an effect would not be expected for

Reactivity. The reactivity of 7 toward a variety of functional

groups was studied (Table 1). Reactions were run in the presence

of an excess of chloride (150 mM Bu₄NCl in CDCl₃), and

pathway 2 with a rate-determining Cl₂loss from 7 Cl^-,

3

while rate-determining ionization (pathway 3) is fully con-

therefore the predominant Pd^IVspecies in solution was 7 Cl^-

sistent with a positive solvent effect.

3

rather than 7. Alkenes and alkynes were chlorinated to 1,2-

dichlorinated products; E/Z product ratios as well as side

products from allylic and vinylic chlorinations with cyclohexene,

styrene, and stilbene mirror results from chlorinations of iden-

tical alkenes/alkynes with Cl₂under electrophilic conditions

(Table 2). These matching outcomes suggest a stepwise me-

chanism involving the formation of organic chloronium inter-

mediates that subsequently react via trans-addition of Cl^-or via

proton loss to produce vinylic and allylic chlorination side

products. To understand the transfer of Cl^þfrom LPd^IVCl₄to

Competition Experiment. Further insight the nature of the

conversion of an alkene into a chloronium intermediate was

sought. Even under conditions that render the initial step (Cl₂

loss in pathway 2 or Cl^--dissociation in pathway 3) rate

determining, a competition experiment between two alkenes

with distinct degrees of substitutions would provide meaningful

relative rates for the alkene functionalization step. In reacting

7 Cl^-with a 10 equiv/10 equiv mixture of 1-hexene/3-hexene,

3

we found a 6:1 bias for chlorination of the internal over the

terminal alkene, which did not match with the bias that free Cl₂

displayed under identical conditions (40.1:1). This finding

excludes pathways that rely on in situ formed Cl₂as the active

chlorinating agent (pathways 2 and 3a). By the same token, we

π-systems, we studied the reaction of 7 and 7 Cl^-with alkenes

3

in more detail. Interactions of isolated Pd^IVcompounds with

alkenes or alkynes have never been studied, even though the

existence of (η²-alkene/alkyne)Pd^IVcomplexes was proposed in

Heck reactions^42,66and other carbopalladations,⁶⁷allylic C-H

acetoxylations,⁶⁸and cyclotrimerizations.⁶⁹

interpret the 6:1 preference of 7 Cl^-in favor of the more

3

substituted alkene as an indication against a process involving

π-complexes LPd^IVCl₃(η²-alkene)^þ. Although no mechanistic

1842

dx.doi.org/10.1021/ja107342b |J. Am. Chem. Soc. 2011, 133, 1832–1848

Journal of the American Chemical Society

ARTICLE

studies on potentially existing Pd^IVπ-complexes have been

reported so far, a plethora of reactions in the literature involving

migratory insertions^46a,bof alkenes into Pd^II-R bonds corrobo-

rate our chemoselectivity argument; for example, rates of Heck

reactions slow with an increasing degree of alkene substitu-

tion^1,46c,dand Pd^II-based carbopalladations,^46e-ghydro-

palladations,^46h,ias well as aminopalladations show identical

trends.^46j-lMost important, the Pd^II-Cl addition across iso-

prene showed kinetic chloropalladation of the less substituted

CdC bond.^46m,nThe generally higher reactivity of Pd^IIcom-

plexes toward terminal alkenes in metal-mediated reactions is at

odds with the intermediacy of LPd^IVCl₃(η²-alkene)^þ(C) in

pathway 3c. In contrast, in ligand-mediated alkene transforma-

of chloride and 7, the higher reactivity of 7 Cl^-over 7, and the

nonlinearity of this rate acceleration expose our system as a rare

case of a specific salt effect on a neutral molecule.

3

Effect of Chloride on the Distribution Factor k₂/k_-1. An

interesting facet of the kinetic analysis is the 1300-fold reversal of

A

B

distribution factors (k₂/k_-1^A= 1:10 and k₂/k_-1^B= 130:1) that

reflect the inherent preference of LPd^IVCl₃^þ(A and B) for Cl^þ-

transfer to cyclohexene as compared to Cl^--addition to the metal

center. Because short-lived intermediates A and B could not be

directly detected by ¹H NMR spectroscopy, structural proposals

can only be putative. As far as intermediate B is concerned, we

propose a zwitterionic supramolecular structure based on the fact

that hydrogen-bonded chloride lowers the barrier for loss of

metal bound chloride, and it is therefore likely that Cl^-exper_þi-

ences an even stronger association with the cation LPd^IVCl₃

73

tions, such as dichlorinations by CrO₂Cl₂as well as OsO₄-

catalyzed dihydroxylations,⁷⁴functional groups are directly in-

stalled on the CdC bond without prior π-coordination to the

metal. Ligand-mediated transformations prefer electron-rich

internal alkenes and are less sensitive toward steric factors.^73b

Equally, electrophilic alkene functionalizations with three-mem-

bered transition states, such as epoxidations or halogenations,

than with LPd^IVCl₄. Furthermore, the conversion of B to 7 Cl^-

3

is kinetically dependent on [Cl^-]_free. The need for an attack of

external chloride ions to B is consistent with a structure of B

containing an immobilized, hydrogen-bonded Cl^-that is un-

available for an internal return reaction.⁸⁵Structurally related ion

pairs that bind Cl^-via hydrogen bonds in their ligand periphery

have been isolated by others (compounds 10-12). The in situ

formation of a cationic Pd^IVcenters in chloroform has been

recently described by Sanford;^6awe propose that factors con-

tributing to the relative ease of formation of B include: (a) a

relatively tight zwitterionic constellation involving the counter-

anion; (b) the presence of two strongly σ-donating carbene

ligands;¹³and (c) the presence of π-donating chloride ligand.⁸⁶

Our understanding of the structure of A is more vague.

Heterolysis of a Pd^IV-Cl bond in 7 is likely to initially produce

an intimate ion pair LPd^IVCl₃^þCl^-, which may subsequently

also display a strong preference for internal alkenes.^73b,75

A

picture emerges in which chloronium intermediate is formed

via a ligand-mediated transfer of Cl^þfrom an intermediate

LPd^IVCl₃(A or B, depending on the precursor 7 or 7 Cl^-)

þ

3

to the alkene; from the viewpoint of the alkene, this reaction may

be considered an “S_N2-type” attack to a Cl-ligand in A or B.

While our reactions have not been extended to catalytic applica-

tions yet, it is apparent that these direct Pd^IV-based alkene

functionalizations diverge from the paradigm in Scheme 1 in

which Pd^IIand a O⁷⁶-, N^46j,77-, or C⁷⁸-nucleophile are added

across the olefin in the activation⁷⁹step followed by Pd^II/

Pd^IVoxidation^6b,19a,80and the C-X bond-forming event

(“functionalization”) on the oxidation state þ4; LPd^IVCl₄

has the ability to bypass otherwise necessary interactions between

the olefin and Pd^II.

react with external BF₄^-via an exchange pathway based on (-/

-

þ/-) triple ions⁸⁷to establish BF₄

H-C hydrogen bonds

3 3 3

on the periphery of the ligand similar to structure 13⁶⁵and

analogous to the proposed structure of B.⁸⁸Altering the anion

-

environment from BF₄for A to Cl^-for B has a noticeable

Supramolecular Catalysis in the Pd^IV-Cl Dissociation

Step. In Scheme 5, we summarize our conclusion of a two-step

chlorination mechanism; the opening event involves Cl^--dis-

sociation from LPd^IVCl₄via polarized transition states TS₁and

TS₂. The resulting intermediates A and B subsequently transfer

Cl^þto the alkene in a ligand-mediated process (TS_Aand TS_B).

Dynamic simulations of kinetic data from Table 3 allowed us to

impact on their relative reactivities k₂/k_-1that diverge 1300 fold

for k₂/k_-1and k₂/k_-1^B. This is in contrast to Romeo’s

observation that the reactivities toward DMSO exchange of

12a,b (as well as analogues with X = OTf, BF₄) are virtually

unaﬀected by the nature of the counterion.⁶⁴The explanation for

diﬀerent reactivities of A and B may lie in diﬀerent proximities

A

B

A

extract rate constants for Pd^IV-Cl heterolyses in 7 (k₁) and in

between cation LPd^IVCl₃^þto the counteranion X within X^-

3 3 3

7 Cl^-(k₁), and we found that k₁^Bis about 5 times larger than

(NHC)₂Pd^IVCl₃given the diﬀerences in ionic radii of Cl^-

B

þ

3

k₁^Aand the barrier height difference ΔG_B^q- ΔG_A^qamounts to

(1.81 Å)⁸⁹and BF₄(2.20 Å).⁹⁰Consequently, the eﬀective

-

-0.9 kcal/mol. We propose that the formation of a positive

charge on the metal center in transition state TS₁energetically

benefits from the proximity of an adjacent negative charge that is

held in place by a supramolecular contact. Effectively, the role of

Cl^-is that of a catalyst; attachment of chloride to the ligand

periphery facilitates the detachment of chloride from the metal.

This is noteworthy because electrostatic supramolecular catalysis

in which nearby ionic guests affect the reactivity of neutral hosts

through space is rarely seen with organic molecules⁸¹and is

unprecedented for transition metal complexes. The higher

positive charge on the metal center may be more pronounced in

A and may enhance charge-controlled reactivity (Cl^--capture)

over orbital control with the neutral substrate cyclohexene.

Variations in the degree of charge separation have been fre-

quently invoked in the literature to explain reactivity diﬀerences

in inorganic and organometallic ion pairs.⁹¹

Irrespective of the chemical preferences of A and B, their

reluctance to interconvert into each other on the time scale of

their reactions with cyclohexene (“forward reaction”) and chlo-

ride (“return reaction”) is remarkable in light of the often very

rapid intra- and intermolecular ion pair dynamics on the time

scale of 100-400 ps.⁹²Mechanistically, fast intermolecular ion

exchange channels, even in nonpolar solvents, involve the

formation of triple ions⁸⁷or quadrupoles.⁹³However, Romeo

found that X = PF₆^-/Cl^-exchange between 12b and 12a

proceeds comparatively slowly (k₂= 96 M^-1s^-1; ΔG^q= 14.9

kcal mol^-1). Based on a bimolecular rate law, this suggested a

reactivity of 7 Cl^-over 7 can also be explained by a special salt

3

effect⁸²in which externally added anions or cations affect the

reactivity of a system through specific interactions.⁸³In com-

pound 7 Cl^-, the chloride ion from the electrolyte hetero-

3

conjugates⁸⁴to the neutral receptor 7 via hydrogen bonding

and addition of chloride ions to 7 creates a nonlinear rate increase

as 7 Cl^-is formed until 7 is consumed. The specific 1:1 binding

3

1843

dx.doi.org/10.1021/ja107342b |J. Am. Chem. Soc. 2011, 133, 1832–1848

Journal of the American Chemical Society

ARTICLE

“direct bimolecular attack by the chloride on the acidic proton to

remove the less tightly bound hexaﬂuorophosphate ion” as

opposed to a conventional triple ion or quadrupole pathway.

We suggest that the reluctance of A to convert into B, even in the

presence of free chloride ions, has kinetic impediments similar to

those found in 12a,b.

Scheme 6

Reaction of B with Aromatic π-Systems. Expanding our

scope from isolated alkenes to aromatic π-systems, we discov-

ered that LPd^IVCl₄complex 7 Cl^-produced aromatic and

3

benzylic chlorination products, and we found evidence for the

þ

interaction of LPd^IVCl₃intermediate B with aromatic sub-

strates reminiscent of pathway 3b for the chlorination of alkenes

(Scheme 4). Chlorinations of aromatic substrates by 7 Cl^-both

3

on the ring as well in benzylic positions (if available) were

possible with electron-rich substrates (Table 1, entries 7, 8, 10,

11), while xylene (entry 9) was inert even at 60 °C. This is

consistent with an electrophilic delivery of Cl^þby B. In line with

this is the observed regioselectivity in the chlorination of

4-methylanisole in its 2- and 4-positions (entry 11) consistent

with the reported reactivity of this substrate in electrophilic

chlorinations.^94,95Equally, we propose that benzylic chlorina-

tions in entries 6, 7, and 8 are a consequence of initial Cl^þ-

delivery to the aromatic ring by B similar to the benzylic

chlorination mechanism proposed Baciocchi and Illuminati for

highly methylated arenes.⁹⁶For example, in the reaction of

organometallic Pd^IV-C_quartintermediates from 4-methylanisole

and 7 Cl^-(entry 11, Table 1) to explain the preferred formation

3

of 4-chloro-4-methylcyclohexa-2,5-dienone over 2-chloroanisole

as rather implausible.

7 Cl^-with durene, the initially formed intermediate B would

C-H bond activations by Pd^IVspecies were considered for the

ﬁrst time in 1971 in Pd(OAc)₂-catalyzed aromatic acetoxyla-

tions,^11dand recent sporadic reports have provided indirect

experimental evidence that in situ generated Pd^IVspecies may

indeed be capable of activating C-H bonds and form Pd^IV-Ar

bonds with¹⁰²or without chelation-control.¹⁰³In contrast to

3

electrophilically deliver Cl^-to the R- and the β-positions of

the π-system to form F and G (Scheme 6). While F then

produces the ring chlorination product, G undergoes a Bacioc-

chi-Illuminati sequence including deprotonation and 1,5-Cl-

shift from H to form the benzylic chlorination product. The

observed biases for aromatic over benzylic chlorination products

in entry 7 (aromatic/benzylic = 3.5:1) and entry 8 (aromatic/

benzylic = 12:1) compare reasonably well with those found

in previously reported electrophilic chlorinations of durene

(aromatic/benzylic = 5.9:1)⁹⁷and mesitylene with quantitative

aromatic substitution.⁹⁸

these reports, our ﬁndings suggest that LPd^IVCl₄complex 7 Cl^-

3

functionalizes aromatic and benzylic C-H bonds without the

intermediate formation of Pd^IV-C bonds. Product selectivities

and kinetic evidence lend to the view that intermediate B is the

reactive species in directly transferring Cl^þto the aromatic π-

system akin to Cl^þ-transfers from B to alkenes/alkynes.

Thermal Decomposition of 7 Cl^-via Side-Chain Chlo-

Kinetic support for the intermediacy of B in the reaction of

3

7 Cl^-with hexamethylbenzene (HMB) is next discussed. At

rination and Chlorination of Methane. In the absence of

3

[HMB] = 38.9 mM (CDCl₃/150 mM Bu₄NCl), the observed

rate constant for benzylic chlorination was k_obs= 22.2 ꢀ 10^-5s^-1

(Table 3); this value is essentially identical to the previously

determined k₁^B= 20.1 ꢀ 10^-5s^-1and is consistent with the rate-

determining formation of B in this reaction.

suitable substrates, Pd^IVcomplex 7 Cl^-gradually decomposes

3

in solution to Pd^II. Following this process by ¹H NMR spectros-

copy revealed that the reduction of the metal center is accom-

panied by C-H chlorinations of the C₁₄H₂₉side chain.¹⁰⁴This

process can be deliberately repeated in a catalytic fashion by

reoxidizing the newly formed L⁰Pd^IICl₂complex in the presence

of excess Cl₂to L⁰Pd^IVCl₄(L⁰= bis-NHC ligand with a

monochlorinated side chain).¹⁰⁵In the presence of 1 atm Cl₂

(25 °C/16 h/CDCl₃/Bu₄NCl), up to six side-chain chlorination

events were recorded by ESI-MS.

With less reactive aromatic substrates, the slow step in our

model may not be the unimolecular formation of B anymore but

instead the subsequent bimolecular Cl^þ-transfer from B to the π-

system (for example, in the formation of F and G in Scheme 6).

Indeed, the reactions of 7 Cl^-with the substrates 4-methylani-

3

sole (k_bi= 2.1 ꢀ 10^-3M^-1s^-1) and 1,4-dimethoxybenzene

The relevance of Pd^IVcomplexes in this chlorination process

(rather than Cl₂acting as the active aliphatic chlorinating agent)

was established in a negative control experiment in which hexane

(as a metal-free side-chain model) failed to produce chlorohex-

anes in the presence of 1 atm Cl₂under otherwise identical

reaction conditions. In an attempt to utilize this ability of

LPd^IVCl₄to functionalize aliphatic C-H bonds, we probed for

the complex’s intermolecular ability to functionalize methane.

Indeed, we identiﬁed the formation of CH₃Cl when a sample of

(k_bi= 6.1 ꢀ 10^-3M^-1s^-1) follow a bimolecular rate law:

d[7 Cl^-]/dt = k₁[7 Cl^-][substrate].

B

3

Even though our data on aromatic/benzylic chlorination are

consistent with a ligand-mediated Cl^þtransfer from in situ

generated B, we cannot deﬁnitively rule out metal-mediated

processes including formation of Pd^IV-Ar or Pd^IV-CH₂Ar

bonds. Such organometallic intermediates are commonly formed

in Pd^II-based aromatic C-H bond activations via electrophilic

aromatic substitutions,⁹⁹concerted metalation-deprotonation,¹⁰⁰

or palladation/β-elimination (Heck-like).¹⁰¹Moreover, Sanford

was able to form Pd^IV-Ar bonds from proposed [Pd^IV-O₂CR]

intermediates.¹⁰²However, we consider a facile formation of

CH₄(1 atm in CDCl₃/Bu₄NCl) and 7 Cl^-were heated to 100 °C

3

for 2 h in a sealed NMR tube. The extremely low yield in methyl

chloride of 2.9% was a consequence of competing side-

chain chlorination to (LPd^IVCl₄f L⁰Pd^IICl₂), while yields on

1844

dx.doi.org/10.1021/ja107342b |J. Am. Chem. Soc. 2011, 133, 1832–1848

Journal of the American Chemical Society

ARTICLE

Scheme 7

proceed via chloroarenium ion intermediates. Methane was

converted into methyl chloride in small yields. Through the

establishment of C-H Cl^-hydrogen bonds in the ligand

3 3 3

periphery, we were able to modulate chemical reactivities of

LPd^IVCl₄in alkene functionalizations. This supramolecular con-

tact accelerated both Cl^--dissociation from LPd^IVCl₄and Cl^þ-

transfer from LPd^IVCl₃^þ. We propose that our results pave the

way for direct activations of organic compounds via Pd^IV

chemistry in future catalytic applications.

’ ASSOCIATED CONTENT

S

Supporting Information. Detailed experimental proce-

b

dures, spectroscopic characterizations, crystallographic data of

compound 4, derivations of kinetic equations, and modeling of

kinetic data. This material is available free of charge via the

Internet at http://pubs.acs.org.

’ AUTHOR INFORMATION

the intermolecular functionalization may have also suﬀered from

the relatively low pressure of methane in the experimental setup.

Recalling the modus operandi of chlorinations of π-

Corresponding Author

skraft@ksu.edu

systems by 7 Cl^-, the chlorination of aliphatic C-H bonds

3

is likely to involve the in situ generation of B (pathway a,

’ ACKNOWLEDGMENT

Scheme 7). The latter may then insert Cl^þinto a C-H bond

of methane or a side-chain C-H bond in 7 Cl^-via an

Financial support from NIH/K-INBRE (P 20 RR16475)-

(SM), the Terry C. Johnson Cancer Center, the Kansas Lipido-

mics Research Center Analytical Laboratory (EPS 0236913, DBI

0521587), and Kansas State University is gratefully acknowl-

edged. We thank G .W. Conrad for his support and guidance and

D. R. M. Townsend, M. K. Higgins, A. L. May, and Y. Y. Yip for

helpful discussions.

3

electrophilic aliphatic substitution.¹⁰⁶Such a reaction is

reminiscent of electrophilic aliphatic methane chlorinations

with Cl₂/SbF₅,¹⁰⁷Cl₂/sulfated zirconia,¹⁰⁸and of electro-

philic alkane ﬂuorination by F₂.¹⁰⁹Alternatively, we also

considered an aliphatic C-H bond activation/functionaliza-

tion involving Pd^IV-C bonds in intermediate J (pathway b,

Scheme 7) because Pd^II-, Pt^II-,^16b,110or rare Pt^IV-¹¹¹me-

diated aliphatic C-H bond activations generally involve the

intermediacy of organometallic complexes; aliphatic C-H

bond activations by Pd^IVare unknown. Some indirect support

for the general viability of complexes containing Pd^IV-C bonds

was found in the reaction of 7 with CH₃MgCl, which produced

a 1:1 mixture of CH₃Cl and CH₃CH₃and LPd^IICl₂(6)

(entry 13, Table 1). The formation of ethane is indicative of

a Pd^IV-dimethyl intermediate LPd^IVCl₂(CH₃)₂that subse-

quently reductively eliminates CH₃CH₃; unlike their Pd^II-analogues,

dialkyl Pd^IVspecies are susceptible to undergo C-C reductive

elimination.^2dHowever, while CH₃Cl may have therefore been

formed from LPd^IVCl₃(CH₃) in entry 13, this does not necessi-

tate the formation of intermediate J in Scheme 7.

’ REFERENCES

(1) Tsuji, J. Palladium Reagents and Catalysts: New Perspectives for the

21st Century, 2nd ed.; John Wiley & Sons Ltd.: Chichester, 2004.

(2) (a) Xu, L. M.; Li, B. J.; Yang, Z.; Shi, Z. J. Chem. Soc. Rev. 2010,

39, 712–733. (b) Sehnal, P.; Taylor, R. J. K.; Fairlamb, I. J. S. Chem. Rev.

2010, 110, 824–889. (c) Sanford, M.; Lyons, T. W. Chem. Rev. 2010,

110, 1147–1169. (d) Canty, A. J. J. Chem. Soc., Dalton Trans. 2009,

10409–10417. (e) Chen, X.; Engle, K. M.; Wang, D. H.; Yu, J. Q. Angew.

Chem., Int. Ed. 2009, 48, 5094–5115.

(3) (a) Canty, A. J. Acc. Chem. Res. 1992, 25, 83–90. (b) Alsters, P. L.;

Engel, P. F.; Hogerheide, M. P.; Copijn, M.; Spek, A. L.; van Koten, G.

Organometallics 1993, 12, 1831–1844. (c) Dick, A. R.; Kampf, J. W.;

Sanford, M. S. J. Am. Chem. Soc. 2005, 127, 12790–12791. (d) Furuya,

T.; Ritter, T. J. Am. Chem. Soc. 2008, 130, 10060–10061.

(4) The ﬁrst isolable organometallic Pd^IVcompound was synthe-

sized in 1975: Uson, R.; Fornies, J.; Navarro, R. J. Organomet. Chem.

1975, 96, 307–312.

’ SUMMARY

Bis-NHC-Pd^IV-tetrachlorides (LPd^IVCl₄) were isolated for

the ﬁrst time and employed as a chlorinating agent and oxidant

for a variety of organic substrates, thus manifesting the feasibility

of direct Pd^IV-functionalizations. Although stoichiometric in

nature, this chemistry departs from the Pd^II/Pd^IVparadigm

(Scheme 1) that requires an initial activation of a substrate by

Pd^II. Furthermore, under ambient conditions, LPd^IVCl₄was able

to intermolecularly functionalize a variety of simple organic

substrates without the need for a chelation process. Mechanistic

studies on cyclohexene chlorinations elucidated a stepwise process

involving cationic intermediates LPd^IVCl₃^þas active chlorinating

agents; Cl^þ-transfers occurred via ligand-mediated routes.

We were able to chlorinate aromatic and benzylic C-H bonds

of electron-rich aromatic systems with LPd^IVCl₄. These reactions

(5) (a) Byers, P. K.; Canty, A. J.; Crespo, M.; Puddephatt, R. J.; Scott,

J. D. Organometallics 1988, 7, 1363–1367. (b) Canty, A. J.; Denney,

M. C.; Skelton, B. W.; White, A. H. Organometallics 2004, 23, 1122–

1131. (c) Canty, A. J.; Done, M. C.; Skelton, B. W.; White, A. H. Inorg.

Chem. Commun. 2001, 4, 648–650. (d) Canty, A. J.; Hoare, J. L.; Davies,

N. W.; Traill, P. R. Organometallics 1998, 17, 2046–2051. (e) Vanasselt,

R.; Rijnberg, E.; Elsevier, C. J. Organometallics 1994, 13, 706–720.

(f) Markies, B. A.; Canty, A. J.; Boersma, J.; van Koten, G. Organometallics

1994, 13, 2053–2058. (g) Markies, B. A.; Canty, A. J.; Janssen, M. D.;

Spek, A. L.; Boersma, J.; van Koten, G. Recl. Trav. Chim. Pays-Bas 1991, 110,

477–479. (h) Byers, P. K.; Canty, A. J.; Traill, P. R.; Watson, A. A.

J. Organomet. Chem. 1990, 390, 399–407. (i) Brown, D. G.; Byers, P. K.;

Canty, A. J. Organometallics 1990, 9, 1231–1235. (j) Canty, A. J.;

Watson, A. A.; Skelton, B. W.; White, A. H. J. Organomet. Chem. 1989,

367, C25–C28. (k) Byers, P. K.; Canty, A. J. Chem. Commun. 1988,

1845

dx.doi.org/10.1021/ja107342b |J. Am. Chem. Soc. 2011, 133, 1832–1848

Journal of the American Chemical Society

ARTICLE

639–641. (l) Byers, P. K.; Canty, A. J.; Skelton, B. W.; White, A. H. Chem.

Commun. 1986, 1722–1724.

(6) (a) Racowski, J. M.; Dick, A. R.; Sanford, M. S. J. Am. Chem. Soc.

2009, 131, 10974–10983. (b) Dick, A. R.; Hull, K. L.; Sanford, M. S.

J. Am. Chem. Soc. 2004, 126, 2300–2301.

(22) Bromination of LPd^IIBr₂(2) with Br₂in neat TFA produced a

yellow precipitate of putative LPd^IVBr₄. After the solvent was evaporated

and dried for 1 h under high vacuum, the sample turned colorless again,

and we regenerated 2 with an apparent loss of Br₂.

(23) Attempts to produce related LPd^IVX₄species from 1, 2, or 6

with O- or F-based oxidants [PhI(OAc)₂, PhI(OTFA)₂, t-BuOO-t-Bu,

(PhCO₂)O (all in CDCl₃), K₂S₂O₈, KHSO₅, (both in TFA-d), and

selectﬂuor (in CD₃CN)] caused no noticeable changes in the ¹H NMR

spectrum of the respective Pd^IIstarting material. Reaction of 6 with XeF₂

produced complex mixtures. Treatment of 6 with AgF to form LPdF₂in

situ followed by addition of XeF₂led to unidentiﬁable product mixtures

as well.

(7) Kruis, D.; Markies, B. A.; Canty, A. J.; Boersma, J.; van Koten, G.

J. Organomet. Chem. 1997, 532, 235–242.

(8) Aye, K. T.; Canty, A. J.; Crespo, M.; Puddephatt, R. J.; Scott,

J. D.; Watson, A. A. Organometallics 1989, 8, 1518–1522.

(9) (a) Daugulis, O.; Zaitsev, V. G. Angew. Chem., Int. Ed. 2005, 44,

4046–4048. (b) McCallum, J. S.; Gasdaska, J. R.; Liebeskind, L. S.;

Tremont, S. J. Tetrahedron Lett. 1989, 30, 4085–4088. (c) Tremont, S. J.;

Rahman, H. U. J. Am. Chem. Soc. 1984, 106, 5759–5760. (d) Zaitsev, V. G.;

Shabashov, D.; Daugulis, O. J. Am. Chem. Soc. 2005, 127, 13154–13155.

(10) As stated in ref 2b: “One ultimate challenge facing the Pd^IVﬁeld

is to selectively activate sp³C-H and C-C bonds, without the need for

proximal directing groups (e.g., imino or amino substituents).”

(11) (a) Yoneyama, T.; Crabtree, R. H. J. Mol. Catal. A: Chem. 1996,

108, 35–40. (b) Jintoku, T.; Taniguchi, H.; Fujiwara, Y. Chem. Lett.

1987, 1865–1868. (c) Stock, L. M.; Tse, K.; Vorvick, L. J.; Walstrum,

S. A. J. Org. Chem. 1981, 46, 1757–1759. (d) Henry, P. M. J. Org. Chem.

1971, 36, 1886–1890.

(24) (a) Hirose, K. J. Inclusion Phenom. Macrocyclic Chem. 2001, 39,

193–209. (b) Job, P. Ann. Chim. 1928, 9, 113. (c) Job, P. Comptes Rendus

1925, 180, 928–930.

(25) Remsing, R. C.; Swatloski, R. P.; Rogers, R. D.; Moyna, G.

Chem. Commun. 2006, 1271–1273.

(26) (a) Uemura, S.; Sasaki, O.; Okano, M. Bull. Chem. Soc. Jpn.

1972, 45, 1482–1484. (b) Onoe, A.; Uemura, S.; Okano, M. Bull. Chem.

Soc. Jpn. 1976, 49, 345–346.

(27) Henniger, P. W.; Dukker, L. J.; Havinga, E. Recl. Trav. Chim.

Pays-Bas 1966, 85, 1177–1187.

(12) (a) Diez-Gonzalez, S.; Marion, N.; Nolan, S. P. Chem. Rev.

2009, 109, 3612–3676. (b) de Fremont, P.; Marion, N.; Nolan, S. P.

Coord. Chem. Rev. 2009, 253, 862–892. (c) Normand, A. T.; Cavell, K. J.

Eur. J. Inorg. Chem. 2008, 2781–2800. (d) Diez-Gonzalez, S.; Nolan, S. P.

Coord. Chem. Rev. 2007, 251, 874–883. (e) Herrmann, W. A.; Schutz,

J.; Frey, G. D.; Herdtweck, E. Organometallics 2006, 25, 2437–2448.

(f) Crudden, C. M.; Allen, D. P. Coord. Chem. Rev. 2004, 248, 2247–

2273. (g) Herrmann, W. A. Angew. Chem., Int. Ed. 2002, 41, 1290–1309.

(h) Herrmann, W. A.; Weskamp, T.; Bohm, V. P. W. Adv. Organomet.

Chem. 2001, 48, 1–69. (i) Herrmann, W. A.; Kocher, C. Angew. Chem.,

Int. Ed. Engl. 1997, 36, 2162–2187.

(28) Timokhin, B. V. Usp. Khim. 1985, 54, 2027–2043.

(29) Heasley, V. L.; Rold, K. D.; Titterington, D. R.; Leach, C. T.;

Gipe, B. T.; McKee, D. B.; Heasley, G. E. J. Org. Chem. 1976, 41, 3997–

4001.

(30) Nugent, W. A. J. Org. Chem. 1980, 45, 4533–4534.

(31) Chung, S. K. Tetrahedron Lett. 1978, 3211–3214.

(32) Tanner, D. D.; Gidley, G. C. J. Org. Chem. 1968, 33, 38–43.

(33) Gulliver, D. J.; Levason, W. Coord. Chem. Rev. 1982, 46, 1–127.

(34) Kukushkin, Y. N.; Sedova, G. N.; Vlasova, R. A. Zh. Neorg. Khim.

1978, 23, 1877–83.

(35) Babaeva, A. V.; Khananova, E. Y. Zh. Neorg. Khim. 1965, 10,

(13) Jain, K. R.; Herrmann, W. A.; Kuhn, F. E. Curr. Org. Chem.

2008, 12, 1468–1478.

(14) Arnold, P. L.; Sanford, M. S.; Pearson, S. M. J. Am. Chem. Soc.

2009, 131, 13912–13913.

2579–81.

(36) Gulliver, D. J.; Levason, W. J. Chem. Soc., Dalton Trans. 1982,

1895–1898.

(37) In a concerted loss of Cl₂, two Pd-Cl bonds would break at the

same time concomitant with a change of the oxidation state of the metal

from þ4 to þ2. In contrast, a stepwise route would entail an initial

dissociation of Cl^-followed by a Cl^þtransfer from LnPd^IVCl₃^þto Cl^-

to form Cl₂. The metal is reduced in the second step. This ionic sequence

is reminiscent of pathway 3b (Scheme 4).

(38) Furuya, T.; Benitez, D.; Tkatchouk, E.; Strom, A. E.; Tang, P.;

Goddard, W. A.; Ritter, T. J. Am. Chem. Soc. 2010, 132, 3793–3807.

(39) (a) Jensen, M. P.; Wick, D. D.; Reinartz, S.; White, P. S.;

Templeton, J. L.; Goldberg, K. I. J. Am. Chem. Soc. 2003, 125, 8614–

8624. (b) Crumpton-Bregel, D. M.; Goldberg, K. I. J. Am. Chem. Soc.

2003, 125, 9442–9456. (c) Williams, B. S.; Goldberg, K. I. J. Am. Chem.

Soc. 2001, 123, 2576–2587.

(15) (a) Kuhl, O. Coord. Chem. Rev. 2009, 253, 2481–2492.

(b) Cavell, K. J. Chem. Soc., Dalton Trans. 2008, 6676–6685. (c) Graham,

D. C.; Cavell, K. J.; Yates, B. F. J. Chem. Soc., Dalton Trans. 2005, 1093–

1100. (d) Clement, N. D.; Cavell, K. J. Angew. Chem., Int. Ed. 2004, 43,

3845–3847. (e) Cavell, K. J.; McGuinness, D. S. Coord. Chem. Rev. 2004,

248, 671–681. (f) McGuinness, D. S.; Saendig, N.; Yates, B. F.; Cavell,

K. J. J. Am. Chem. Soc. 2001, 123, 4029–4040.

(16) (a) Muehlhofer, M.; Strassner, T.; Herrmann, W. A. Angew.

Chem., Int. Ed. 2002, 41, 1745–1747. (b) Strassner, T.; Muehlhofer, M.;

Zeller, A.; Herdtweck, E.; Herrmann, W. A. J. Organomet. Chem. 2004,

689, 1418–1424.

(17) Gray, L. R.; Gulliver, D. J.; Levason, W.; Webster, M. J. Chem.

Soc., Dalton Trans. 1983, 133–141.

(40) Uemura, S.; Onoe, A.; Okano, M. Bull. Chem. Soc. Jpn. 1974, 47,

(18) Babaeva, A. V.; Khananova, E. Y. Dokl. Akad. Nauk SSSR 1964,

159, 586–7.

692–697.

(41) Gamlen, P. H.; Henty, M. S.; Roberts, H. L. J. Chem. Soc., Dalton

Trans. 1983, 1373–1376.

(42) Sundermann, A.; Uzan, O.; Martin, J. M. L. Chem.-Eur. J. 2001,

(19) (a) Kalyani, D.; Dick, A. R.; Anani, W. Q.; Sanford, M. S. Org.

Lett. 2006, 8, 2523–2526. (b) Kalyani, D.; Dick, A. R.; Anani, W. Q.;

Sanford, M. S. Tetrahedron 2006, 62, 11483–11498. (c) Whitﬁeld, S. R.;

Sanford, M. S. J. Am. Chem. Soc. 2007, 129, 15142–15143.

(20) Arnold, P. L.; Zlatogorsky, S.; Jones, N. A.; Carmichael, C. D.;

Liddle, S. T.; Blake, A. J.; Wilson, C. Inorg. Chem. 2008, 47, 9042–9049.

(21) Less coordinating salts such as LiClO₄, LiOTf, or NaBPh₄did

not assist dissolving 7 in CDCl₃, CD₂Cl₂, or CDCl₂CDCl₂in any

appreciable fashion. This suggests that the establishment of a trifurcated

7, 1703–1711.

(43) (a) B€ackvall, J. E.; Nilsson, Y. I. M.; Gatti, R. G. P. Organome-

tallics 1995, 14, 4242–4246. (b) B€ackvall, J. E.; Nilsson, Y. I. M.;

Andersson, P. G.; Gatti, R. G. P.; Wu, J. C. Tetrahedron Lett. 1994, 35,

5713–5716. (c) Wipke, W. T.; Goeke, G. L. J. Am. Chem. Soc. 1974, 96,

4244–4249.

(44) Poutsma, M. L. J. Am. Chem. Soc. 1965, 87, 4285–4293.

(45) Collman, J. P.; Hegedus, L. S.; Norton, J. R.; Finke, R. G.

Principles and Applications of Organotransition Metal Chemistry; Univer-

sity Science Books: Mill Valley, CA, 1987; p 149.

(46) (a) Cavell, K. J. Coord. Chem. Rev. 1996, 155, 209–243. (b) Tye,

J. W.; Hartwig, J. F. J. Am. Chem. Soc. 2009, 131, 14703–14712.

(c) G€urtler, C.; Buchwald, S. L. Chem.-Eur. J. 1999, 5, 3107–3112.

(d) Heck, R. F. Acc. Chem. Res. 1979, 12, 146–151. (e) Larock, R. C.;

hydrogen bond in 7 Cl^-is necessary to break up packing forces in the

3

solid state of 7. Similarly, BF₄^-may be able to interact with 7 in a similar

fashion, yet to a weaker extent than Cl^-. The observation that at least 5

equiv of Bu₄NBF₄is needed to dissolve 1 equiv of 7 while only 1 equiv of

Bu₄NCl is required qualitatively agrees with this hypothesis. Anions with

even poorer hydrogen-bonding capabilities such as ClO₄^-, TfO^-, or

BPh₄^-are apparently not able to disrupt the solid state network of 7.

1846

dx.doi.org/10.1021/ja107342b |J. Am. Chem. Soc. 2011, 133, 1832–1848

Journal of the American Chemical Society

ARTICLE

Takagi, K. J. Org. Chem. 1984, 49, 2701–2705. (f) Larock, R. C.; Fried,

C. A. J. Am. Chem. Soc. 1990, 112, 5882–5884. (g) Albeniz, A. C.;

Espinet, P.; Lin, Y. S. Organometallics 1995, 14, 2977–2986. (h) Stakem,

F. G.; Heck, R. F. J. Org. Chem. 1980, 45, 3584–3593. (i) Mabbott, D. J.;

Maitlis, P. M. J. Chem. Soc., Dalton Trans. 1976, 2156–2160. (j) Liu,

G. S.; Stahl, S. S. J. Am. Chem. Soc. 2006, 128, 7179–7181. (k) Brice, J. L.;

Harang, J. E.; Timokhin, V. I.; Anastasi, N. R.; Stahl, S. S. J. Am. Chem.

Soc. 2005, 127, 2868–2869. (l) Paiaro, G.; Derenzi, A.; Palumbo, R.

Chem. Commun. 1967, 1150–1151. (m) B€ackvall, J. E. Acc. Chem. Res.

1983, 16, 335–342. (n) B€ackvall, J. E.; Nordberg, R. E.; Nystrom, J. E.

Tetrahedron Lett. 1982, 23, 1617–1620.

(56) We gratefully acknowledge an alternative explanation by a

reviewer who pointed out the “possibility that the relatively long

NHC-Pd^IVdistance might reﬂect the poor overlap between the soft

carbene carbon and the much hardened Pd^IVcenter.”

(57) (a) Khramov, D. M.; Lynch, V. M.; Bielawski, C. W. Organo-

metallics 2007, 26, 6042–6049. (b) Fantasia, S.; Petersen, J. L.; Jacobsen,

H.; Cavallo, L.; Nolan, S. P. Organometallics 2007, 26, 5880–5889.

(c) Jacobsen, H.; Correa, A.; Costabile, C.; Cavallo, L. J. Organomet.

Chem. 2006, 691, 4350–4358. (d) Scott, N. M.; Nolan, S. P. Eur. J. Inorg.

Chem. 2005, 1815–1828.

(58) Baba, E.; Cundari, T. R.; Firkin, I. Inorg. Chim. Acta 2005, 358,

(47) We omitted data from entry 1 (Table 3) in Figure 3 for visual

clarity regarding early data points. However, k_obsvalues are identical in

entries 1 and 2, which underlines the validity of a [cy]-independent k_obs

plateau.

2867–2875.

(59) (a) Appleton, T. G.; Clark, H. C.; Manzer, L. E. Coord. Chem.

Rev. 1973, 10, 335–422. (b) Pidcock, A.; Richards, R. E.; Venanzi, L. M. J.

Chem. Soc. A 1966, 1707–1710.

(60) Baker, M. V.; Brown, D. H.; Heath, C. H.; Skelton, B. W.;

White, A. H.; Williams, C. C. J. Org. Chem. 2008, 73, 9340–9352.

(61) Li, Y. L.; Flood, A. H. J. Am. Chem. Soc. 2008, 130, 12111–

12122.

(62) (a) Fortin, S.; Beauchamp, A. L. Inorg. Chem. 2001, 40, 105–

112. (b) Fortin, S.; Fabre, P. L.; Dartiguenave, M.; Beauchamp, A. L.

J. Chem. Soc., Dalton Trans. 2001, 3520–3527.

(63) (a) B€ottcher, L.; Scholz, A.; Walther, D.; Weisbach, N.; G€orls,

H. Z. Anorg. Allg. Chem. 2003, 629, 2103–2112. (b) Rau, S.; Bottcher, L.;

Schebesta, S.; Stollenz, M.; Gorls, H.; Walther, D. Eur. J. Inorg. Chem.

2002, 2800–2809.

(64) Romeo, R.; Nastasi, N.; Scolaro, L. M.; Plutino, M. R.; Albinati,

S.; Macchioni, A. Inorg. Chem. 1998, 37, 5460–5466.

(48) We corrected k_obsdata from entries 21-24 by subtracting rate

contributions of small amounts of 7 Cl^-(χ_{7 Cl-}= 0.080-0.045) in the

3

system. These corrections amounted to less than 1.6 ꢀ 10^-5s^-1and did

not aﬀect the conclusion that 7 Cl^-is noticably more reactive than 7.

3

(49) In the presence of chloride ions, chemical shifts of 7 and 7 Cl^-

3

are indistinguishable on the NMR time scale, and their equilibration

(7 þ Cl^-_freeS 7 Cl^-) is therefore faster than substrate chlorinations

3

(t_1/2of 1 h).

(50) An alternative explanation for the bias against the formation of

B in entries 21-24 (Table 3) may be sought by assuming that the

structure of intermediate A is that of the free cation LPd^IVCl₃^þdevoid of

any localized interactions within a tight io_þn pair with Cl^-or BF₄^-. A

hypothetical equililibrium B S LPd^IVCl₃þ Cl^-with K_eq(B) =

free

[LPd^IVCl_þ₃^þ][Cl^-_free]/[B] would then favor the formation of

LPd^IVCl₃at the expense of B at low [Cl^-_free]. A reasonable lower

estimate for the ion pair association constant K_eq(B) in the non-polar

solvent CDCl₃(that only includes electrostatic attractions but no

speciﬁc hydrogen-bonding eﬀects) would be g100.000 M^-1based on

Bu₄N^þClO₄^-association constants in solvents with ε e 10 ( Janz, C. G.;

Tomkins, R. P. T. Non-Aqueous Electrolytes Handbook; Academic Press:

New York, 1972). Using the lowest free chloride concentration in our

experiments [Cl^-_free] = 0.00034 M (entry 23, Table 3) as well as K_eq(B)

g 100.000 M^-1in [B]/[ LPd^IVCl₃^þ] = 1/(K_eq(B)[Cl^-_free]) (deriv_þed

(65) Appelhans, L. N.; Zuccaccia, D.; Kovacevic, A.; Chianese, A. R.;

Miecznikowski, J. R.; Macchioni, A.; Clot, E.; Eisenstein, O.; Crabtree,

R. H. J. Am. Chem. Soc. 2005, 127, 16299–16311.

€

(66) Herrmann, W. A.; Ofele, K.; Von Preysing, D.; Schneider, S. K.

J. Organomet. Chem. 2003, 687, 229–248.

(67) Wong, P. K.; Stille, J. K. J. Organomet. Chem. 1974, 70, 121–132.

(68) Pilarski, L. T.; Selander, N.; B€ose, D.; Szabo, K. J. Org. Lett.

2009, 11, 5518–5521.

(69) Trost, B. M.; Lautens, M. Tetrahedron Lett. 1985, 26, 4887–

4890.

(70) In nonpolar solvents, quarternary ammonium salts are not

dissociated to any appreciable degree with dissociation enthalpies of 24.3

kcal mol^-1for tetra-i-amylammonium chloride in benzene [ Fuoss,

R. M.; Kraus, C. A. J. Am. Chem. Soc. 1933, 55, 3614–3620] or 6.3 kcal

mol^-1for tetrabutylammonium chloride in CH₂Cl₂. (a) Balt, S.;

Duchattel, G.; Dekieviet, W.; Tieleman, A. Z. Naturforsch., B: J. Chem.

Sci. 1978, 33, 745–749. (b) Svorstol, I.; Hoiland, H.; Songstad, J. Acta

Chem. Scand., Ser. B 1984, 38, 885–893].

(71) A direct solvent comparison for one exclusive species was not

possible, as 7 was insoluble in DMF-d₇and in other solvents with high

dielectric constants, while compound 4 was only soluble in DMF-d₇yet

insoluble in other organic solvents. The addition of electrolytes such as

from the equilibrium equation above) yields the ratio [B]/[ LPd^IVCl₃

]

g 34. On the basis of the conservative estimate for the value of K_eq(B),

even our experimentally lowest value of [Cl^-_free] would therefore not be

low enough to produce signiﬁcant quantities of free [LPd^IVCl₃^þ] at the

expense of B. Therefore, this thermodynamic argument is an unlikely

explanation for the bias against the formation of B in entries 21-24.

We therefore favor a kinetic interpretation of an inhibited formation

of B on the timescale of k_-1and k₂from an intermediate _þA that is best

-

described as hydrogen-bonded adduct BF₄

LPd^IVCl₃

.

3 3 3

(51) cks was developed by IBM-Almaden Lab and can be down-

loaded free of charge from http://www.almaden.ibm.com/st/computa-

tional_science/ck/.

(52) (a) Bunker, D. L.; Garrett, B.; Kleindienst, T.; Long, G. S.

Combust. Flame 1974, 23, 373–379. (b) Gillespie, D. T. J. Comput. Phys.

1976, 22, 403–434.

Bu₄NCl or Bu₄NBF₄did not rectify these solubility problems.

P

(72) The dielectric constants of DMF-d₇( = 36.7) and CDCl₃

P

(

= 4.8) diverge greatly. However, the addition of electrolytes

P

(53) Without the ability to measure in situ concentrations of A and

B, absolute values of k_-1^A, k_-1^B, k₂^A, and k₂^Bcould not be determined.

(54) The upper value of ΔG(exchange) = 14.0 kcal mol^-1for the

exchange process 7 Cl^-S 7 þ Cl^-_freeS 7 Cl^-was determined from

(10^-2-10^-1M range) to unpolar solvents ( < 12) can increase the

dielectric constant by 100% [ (a) Cachet, H.; Cyrot, A.; Fekir, M.;

Lestrade, J. C. J. Phys. Chem. 1979, 83, 2419–2429. (b) Cavell, E. A. S.;

Knight, P. C. Z. Phys. Chem. 1968, 57, 331–334. (c) Svorstol, I.; Hoiland,

H.; Songstad, J. Acta Chem. Scand., Ser. B 1984, 38, 885–893].

(73) (a) Nelson, D. J.; Li, R. B.; Brammer, C. J. Phys. Org. Chem.

2004, 17, 1033–1038. (b) Awasthy, A. K.; Rocek, J. J. Am. Chem. Soc.

1969, 91, 991–996.

3

Δν = 148 Hz from frequencies associated with proton H⁵in discreet

samples of pure 7 as well as 7 Cl^-[operating frequency: 400 MHz;

k(exchange) g 2.22 (Δν)].

3

(55) The choice of recruiting diﬀerent Pd^IV-tetrachlorides in the two

solvents was due to limited solubility characteristics of LPd^IVCl₄species;

ditetradecyl-substituted complex 7 could only be dissolved in chlori-

nated solvents in the presence of electrolytes; 7 was virtually insoluble in

other common organic solvents (with or without adding Bu₄NCl) such

as hexane, diethylether, benzene, THF, ethyl acetate, acetone, acetonitrile,

DMF, and methanol. Similar limitations were encountered with dimethyl-

substituted 4 that showed measurable solubility exclusively in DMF.

(74) Sharpless, K. B.; Williams, D. R. Tetrahedron Lett. 1975, 3045–

3046.

(75) Nelson, D. J.; Li, R. B.; Brammer, C. J. Org. Chem. 2001, 66,

2422–2428.

(76) (a) Wang, A. Z.; Jiang, H. F.; Chen, H. J. J. Am. Chem. Soc. 2009,

131, 3846–3847. (b) Li, Y.; Song, D.; Dong, V. M. J. Am. Chem. Soc.

2008, 130, 2962–2964.

1847

dx.doi.org/10.1021/ja107342b |J. Am. Chem. Soc. 2011, 133, 1832–1848

Journal of the American Chemical Society

ARTICLE

(77) (a) Alexanian, E. J.; Lee, C.; Sorensen, E. J. J. Am. Chem. Soc.

2005, 127, 7690–7691. (b) Streuﬀ, J.; H€ovelmann, C. H.; Nieger, M.;

Muniz, K. J. Am. Chem. Soc. 2005, 127, 14586–14587. (c) Desai, L. V.;

Sanford, M. S. Angew. Chem., Int. Ed. 2007, 46, 5737–5740. (d) Muniz,

K.; H€ovelmann, C. H.; Streuﬀ, J. J. Am. Chem. Soc. 2008, 130, 763–773.

(78) Kalyani, D.; Sanford, M. S. J. Am. Chem. Soc. 2008, 130, 2150–

2151.

Patara, A. Tetrahedron Lett. 1972, 4161–4164. (d) Baciocchi, E.;

Illuminati, G. Tetrahedron Lett. 1975, 2265–2267. (e) Baciocchi, E.;

Mandolini, L. Tetrahedron 1987, 43, 4035–4041. (f) Baciocchi, E.; Galli,

C. J. Phys. Org. Chem. 1995, 8, 563–565.

(97) Andrews, L. J.; Keefer, R. M. J. Am. Chem. Soc. 1960, 82, 5823–

5826.

(98) Fittig, R.; Hoogewerﬀ, S. Liebigs Ann. Chem. Pharm. 1869, 150,

323–338.

(79) We deﬁne the term “activation” as the step that establishes

Pd-C σ-bonds from either an alkene or a C-H bond.

(99) (a) Catellani, M.; Chiusoli, G. P. J. Organomet. Chem. 1992, 425,

151–154. (b) Martin-Matute, B.; Mateo, C.; Cardenas, D. J.; Echavarren,

A. M. Chem.-Eur. J. 2001, 7, 2341–2348. (c) Park, C. H.; Ryabova, V.;

Seregin, I. V.; Sromek, A. W.; Gevorgyan, V. Org. Lett. 2004, 6, 1159–

1162. (d) Pivsa-Art, S.; Satoh, T.; Kawamura, Y.; Miura, M.; Nomura, M.

Bull. Chem. Soc. Jpn. 1998, 71, 467–473. (e) Lane, B. S.; Brown, M. A.;

Sames, D. J. Am. Chem. Soc. 2005, 127, 8050–8057. (f) Li, J. J.; Giri, R.;

Yu, J. Q. Tetrahedron 2008, 64, 6979–6987.

(100) (a) Gorelsky, S. I.; Lapointe, D.; Fagnou, K. J. Am. Chem. Soc.

2008, 130, 10848–10849. (b) Lafrance, M.; Fagnou, K. J. Am. Chem. Soc.

2006, 128, 16496–16497. (c) Lafrance, M.; Rowley, C. N.; Woo, T. K.;

Fagnou, K. J. Am. Chem. Soc. 2006, 128, 8754–8756. (d) Garcia-

Cuadrado, D.; de Mendoza, P.; Braga, A. A. C.; Maseras, F.; Echavarren,

A. M. J. Am. Chem. Soc. 2007, 129, 6880–6886. (e) Garcia-Cuadrado, D.;

Braga, A. A. C.; Maseras, F.; Echavarren, A. M. J. Am. Chem. Soc. 2006,

128, 1066–1067.

(80) (a) Stang, P.; Zhdankin, V. Chem. Rev. 1996, 96, 1123–1178.

(b) Zhdankin, V. V.; Stang, P. J. Chem. Rev. 2002, 102, 2523–2584.

(c) Zhdankin, V. V.; Stang, P. J. Chem. Rev. 2008, 108, 5299–5358.

(d) Deprez, N. R.; Sanford, M. S. Inorg. Chem. 2007, 46, 1924–1935.

(e) Desai, L. V.; Hull, K. L.; Sanford, M. S. J. Am. Chem. Soc. 2004, 126,

9542–9543. (f) Hull, K. L.; Anani, W. Q.; Sanford, M. S. J. Am. Chem. Soc.

2006, 128, 7134–7135. (g) Furuya, T.; Kaiser, H. M.; Ritter, T. Angew.

Chem., Int. Ed. 2008, 47, 5993–5996. (h) Wang, X. S.; Mei, T. S.; Yu, J. Q.

J. Am. Chem. Soc. 2009, 131, 7520–7521.

(81) (a) Smith, P. J.; Kim, E. I.; Wilcox, C. S. Angew. Chem., Int. Ed.

Engl. 1993, 32, 1648–1650. (b) Smith, P. J.; Reddington, M. V.; Wilcox,

C. S. Tetrahedron Lett. 1992, 33, 6085–6088. (c) Heuze, K.; Meziere, C.;

Fourmigue, M.; Batail, P.; Coulon, C.; Canadell, E.; Auban-Senzier, P.;

Jerome, D. Chem. Mater. 2000, 12, 1898–1904.

(82) Winstein, S.; Clippinger, E.; Fainberg, A. H.; Robinson, G. C.

J. Am. Chem. Soc. 1954, 76, 2597–2598.

(101) Glover, B.; Harvey, K. A.; Liu, B.; Sharp, M. J.; Tymoschenko,

M. F. Org. Lett. 2003, 5, 301–304.

(102) Hull, K.; Lanni, E.; Sanford, M. J. Am. Chem. Soc. 2006, 128,

14047–14049.

(83) Typically, special salt eﬀects play out in intercepting in situ

formed ion pairs (intimate or solvent separated) through cation or anion

exchange, thus altering the reactivity of the system, for example, by

slowing internal return reactions. Rate accelerations are nonlinear, and

(103) Rosewall, C.; Sibbald, P.; Liskin, D.; Michael, F. J. Am. Chem.

Soc. 2009, 131, 9488–9489.

k

_obsreaches a plateau or a linear domain that itself is only aﬀected only by

the common salt eﬀect (rate changes through increase in ion strength).

See: Dvorko, G. F.; Ponomareva, E. A.; Ponomarev, N. E. Russ. J. Gen.

Chem. 2006, 76, 1368–1385.

(84) The deﬁnition of heteroconjugation is “the conjugation of an

anion A^-with a hydrogen-bond donor”. See: Kolthoﬀ, I. M.; Chantooni,

M. K. J. Am. Chem. Soc. 1963, 85, 2195–2201.

(104) When 7 Cl^-was heated at 100 °C in CDCl₃for 30 min, ¹H

3

NMR signals associated with Pd^IVcomplex 7 Cl^-completely disap-

3

peared. New multiplets consistent with side-chain chlorinations in the

β-, γ-, and δ-site were found at δ 4.6-4.4, 4.2-4.0, and 3.75-3.65

integrating for 0.52 H, 0.35 H, and 0.10 H relative to the aromatic signals

that are indicative of a pseudosymmetrical Pd^IIscaﬀold (δ 7.85 (2H),

6.80 (2H)).

(85) An internal return reaction of B would require a rearrangement

(105) In contrast to Strassner’s proposed Pd^II/Pd^IVcatalysis [ref

16b] for methane activation with 1 and 2, we suggest an “inverted” Pd^IV/

Pd^IIsequence of events, in which the role of Pd^IIis limited to the reaction

with Cl₂to regenerate Pd^IV, which may serve to expand the horizon of

Strassner’s platform in the future.

of peripherally bonded Cl^-to the metal center and would initially form

7; subsequently, 7 Cl^-is rapidly formed in the presence of Bu₄NCl.

3

(86) (a) Poulton, J. T.; Sigalas, M. P.; Folting, K.; Streib, W. E.;

Eisenstein, O.; Caulton, K. G. Inorg. Chem. 1994, 33, 1476–1485.

(b) Poulton, J. T.; Folting, K.; Streib, W. E.; Caulton, K. G. Inorg. Chem.

1992, 31, 3190–3191.

(106) Olah, G. A. Acc. Chem. Res. 1987, 20, 422–428.

(107) Olah, G. A. Pure Appl. Chem. 1981, 53, 201–207.

(108) Batamack, P.; Bucsi, I.; Molnar, A.; Olah, G. A. Catal. Lett.

1994, 25, 11–19.

(109) (a) Rozen, S. Acc. Chem. Res. 1988, 21, 307–312.

(b) Chambers, R. D.; Parsons, M.; Sandford, G.; Thomas, E.; Trmcic,

J.; Moilliet, J. S. Tetrahedron 2006, 62, 7162–7167.

(110) (a) Shilov, A. E.; Shul’pin, G. B. Chem. Rev. 1997, 97, 2879–

2932. (b) Heyduk, A. F.; Zhong, H. A.; Labinger, J. A.; Bercaw, J. E.

Activation and Functionalization of CH Bonds; ACS Symposium Series

885; American Chemical Society: Washington, DC, 2004. (c) Lersch,

M.; Tilset, M. Chem. Rev. 2005, 105, 2471–2526.

(111) Nizova, G. V.; Krevor, J. V. Z.; Kitaigorodskii, A. N.; Shulpin,

G. B. Bull. Acad. Sci. USSR, Div. Chem. Sci. (Engl. Transl.) 1982, 31,

2480–2482.

(87) (a) Fuoss, R. M.; Kraus, C. A. J. Am. Chem. Soc. 1933, 55, 2387–

2399. (b) Fuoss, R. M.; Kraus, C. A. J. Am. Chem. Soc. 1933, 55, 2387–

2399. (c) Marcus, Y.; Hefter, G. Chem. Rev. 2006, 106, 4585–4621.

(88) As our information on A is limited to kinetic data, we cannot

exclude tight or solvent-separated ion pairs LPd^IVCl₃^þBF₄and

-

LPd^IVCl₃^þ(solvent)BF₄^-from consideration for the structure of A.

(89) Shannon, R. D. Acta Crystallogr., Sect. A: Found. Crystallogr.

1976, 32, 751–767.

(90) Hofstetter, C.; Pochapsky, T. C. Magn. Reson. Chem. 2000, 38,

90–94.

(91) Macchioni, A. Chem. Rev. 2005, 105, 2039–2073.

(92) (a) Yabe, T.; Kochi, J. K. J. Am. Chem. Soc. 1992, 114, 4491–

4500. (b) Simon, J. D.; Peters, K. S. J. Am. Chem. Soc. 1982, 104, 6542–

6547.

(93) (a) Fuoss, R. M.; Kraus, C. A. J. Am. Chem. Soc. 1935, 57, 1–4.

(b) Winstein, S.; Klinedinst, P. E.; Robinson, G. C. J. Am. Chem. Soc.

1961, 83, 885–895. (c) Petrucci, S.; Masiker, M. C.; Eyring, E. M.

J. Solution Chem. 2008, 37, 1031–1035.

(94) Smith, J. R. L.; McKeer, L. C.; Taylor, J. M. J. Chem. Soc., Perkin

Trans. 2 1989, 1529–1536.

(95) Smith, J. R. L.; McKeer, L. C.; Taylor, J. M. J. Chem. Soc., Perkin

Trans. 2 1988, 385–391.

(96) (a) Baciocchi, E.; Illuminati, G. Tetrahedron Lett. 1962, 637–

641. (b) Baciocchi, E.; Ciana, A.; Illuminati, G.; Pasini, C. J. Am. Chem. Soc.

1965, 87, 3953–3957. (c) Illuminati, G.; Mandolini, L.; Baciocchi, E.;

1848

dx.doi.org/10.1021/ja107342b |J. Am. Chem. Soc. 2011, 133, 1832–1848

Article Doi

Bis- N -heterocyclic carbene palladium(IV) tetrachloride complexes: Synthesis, reactivity, and mechanisms of direct chlorinations and oxidations of organic substrates

DOI: 10.1021/ja107342b

Source and publish data:

Authors:

Article abstract of DOI:10.1021/ja107342b

Full text of DOI:10.1021/ja107342b

Products guided by the article

R&D Labs maybe for 71518-99-5

Relevant to this article

Hot Product