Solvent-free aromatic C-H functionalisation/halogenation reactions

Article abstract of DOI:10.1039/c0dt00385a

The solvent-free, palladium-catalysed reaction of anilides with CuCl ₂ in the presence or absence of copper acetate yields ortho-chlorinated anilides in good to excellent yields, even on a large scale (100 mmol). By contrast, the equivalent reactions with copper bromide, either solvent free or in 1,2-dichloroethane, in the presence or absence of palladium, under air or inert conditions, gave the products of simple electrophilic bromination. Mechanistic studies highlighted the involvement of palladacyclic intermediates, one of which was characterised crystallographically, which undergo subsequent reaction with copper(ii) chloride to yield the chlorinated anilide products.

Full text of DOI:10.1039/c0dt00385a

View Online / Journal Homepage / Table of Contents for this issue
This article is published as part of the Dalton Transactions themed issue entitled:
Dalton Discussion 12: Catalytic C-H and C-X Bond
Activation
Published in issue 43, 2010 of Dalton Transactions
Read about the Dalton Discussion 12 meeting on the Dalton Transactions Blog
Image reproduced with the permission of Jennifer Love
Articles in the issue include:
PERSPECTIVES:
Arylation of unactivated arenes
Aiwen Lei, Wei Liu, Chao Liu and Mao Chen
Dalton Trans., 2010, DOI: 10.1039/C0DT00486C
The mechanism of the modified Ullmann reaction
Elena Sperotto, Gerard P. M. van Klink, Gerard van Koten and Johannes G. de Vries
Dalton Trans., 2010, DOI: 10.1039/C0DT00674B
Cross coupling reactions of polyfluoroarenes via C–F activation
Alex D. Sun and Jennifer A. Love
Dalton Trans., 2010, DOI: 10.1039/C0DT00540A
HOT ARTICLE:
The unexpected role of pyridine-2-carboxylic acid in manganese based oxidation catalysis with
pyridin-2-yl based ligands
Dirk Pijper, Pattama Saisaha, Johannes W. de Boer, Rob Hoen, Christian Smit, Auke Meetsma,
Ronald Hage, Ruben P. van Summeren, Paul L. Alsters, Ben L. Feringa and Wesley R. Browne
Dalton Trans., 2010, DOI: 10.1039/C0DT00452A
Visit the Dalton Transactions website for more cutting-edge inorganic and organometallic research
www.rsc.org/dalton

PAPER
www.rsc.org/dalton | Dalton Transactions
Solvent-free aromatic C–H functionalisation/halogenation reactions†‡
Robin B. Bedford,*a Jens U. Engelhart,a Mairi F. Haddow,a Charlotte J. Mitchellb and Ruth L. Webstera
Received 28th April 2010, Accepted 18th June 2010
DOI: 10.1039/c0dt00385a
The solvent-free, palladium-catalysed reaction of anilides with CuCl2 in the presence or absence of
copper acetate yields ortho-chlorinated anilides in good to excellent yields, even on a large scale
(100 mmol). By contrast, the equivalent reactions with copper bromide, either solvent free or in
1,2-dichloroethane, in the presence or absence of palladium, under air or inert conditions, gave the
products of simple electrophilic bromination. Mechanistic studies highlighted the involvement of
palladacyclic intermediates, one of which was characterised crystallographically, which undergo
subsequent reaction with copper(II) chloride to yield the chlorinated anilide products.
alternative halogen sources in palladium catalysed directed C–H
functionalisation/halogenation reactions.8 They have even been
Introduction
shown to function as catalysts under appropriate conditions.9
A disadvantage with many C–H functionalisation processes, in
particular halogenation reactions, is that they often require acidic
(acetic acid or trifluoroacetic acid) or industrially disfavoured
solvents such as 1,2-dichloroethane (DCE). We recently communi-
cated our preliminary findings on the use of solvent free conditions
for direct arylations and halogenations via C–H activation,10 and
now report in detail our results for a range of direct halogenation
reactions along with initial mechanistic findings.
Transition metal-catalysed cross-coupling reactions form the
bedrock of many contemporary organic syntheses.1 Whilst such
reactions are extremely useful and versatile, they are rapidly being
complemented and in some instances supplanted by catalytic C–
H functionalisation processes,2,3 such as direct functionalisation
of aromatics (Scheme 1), not least because these methodologies
allow greater step-economy and atom efficiency than comparable
cross-coupling reactions.
Results and discussion
Scheme 1 Catalytic aromatic C–H functionalisation.
Palladium-catalysed chlorination reactions
One class of reaction that has proved amenable to this protocol
is selective aromatic halogenation, where product distribution can
be controlled largely by the site of metal-mediated C–H activation
(Scheme 2).4 Forty years ago Fahey reported the ortho-selective
chlorination of azobenzene with chlorine.5 More recently Sanford
and co-workers have designed elegant catalytic halogenation
protocols that proceed via palladium-catalysed directed C–H
functionalisation using N-halosuccinimides as milder halogen
sources.6 Similarly Yu and co-workers used in situ prepared XOAc
reagents.7
A particularly useful C–H functionalisation/halogenation reac-
tion is Shi’s palladium-catalysed ortho-chlorination and bromina-
tion of anilides (Scheme 3) using a mixture of two equivalents each
of Cu(OAc)2 and CuX2 in 1,2-dichloroethane.8a In an attempt to
circumvent the use of DCE we reinvestigated this reaction, but
under air without the use of solvent.
Scheme 3 C–H functionalisation/halogenation of anilides.
Scheme 2 Directed C–H functionalisation/halogenation.
Copper salts have also been exploited either as oxidative
halogenating reagents or as oxidising agents in the presence of
We were delighted to find that under these conditions good
to excellent activity was seen in the chlorination of a range of
anilides using CuCl2/Cu(OAc)2 and 5 mol% palladium acetate11
as catalyst with significantly reduced reaction time (2 h) compared
to the 48 h required in 1,2-dichloroethane solution (Table 1). The
reactions were very easily performed: the components were ground
for ~ 30 s with a pestle and mortar and then the resultant powders
heated in an open tube under air. Work-up was equally simple—the
product mixture powder was poured directly onto a silica column
and then subjected to flash chromatography. Using this method
aSchool of Chemistry, University of Bristol, Cantock’s close, Bristol, UK
BS8 1TS. E-mail: r.bedford@bristol.ac.uk
bGlaxoSmithKline, Stevenage, UK SG1 2NY
† Based on the presentation given at Dalton Discussion No. 12, 13–15th
September 2010, Durham University, UK.
‡ Electronic supplementary information (ESI) available: Experimental
details, spectroscopic data, X-ray structural data. CCDC reference number
780622. For ESI and crystallographic data in CIF or other electronic
format see DOI: 10.1039/c0dt00385a
10464 | Dalton Trans., 2010, 39, 10464–10472
This journal is
The Royal Society of Chemistry 2010
©

both acetyl and pivalyl p-toluides were ortho-chlorinated in high
yields (entries 1 and 2). Unsubstituted acetanilide, 1d, was cleanly
chlorinated with complete ortho-selectivity in reasonable isolated
yield (entry 4). By contrast, no reaction was observed with the
2-methyl, 2-bromo or 4-cyano-substituted acetanilides 1e, f and
g (entries 5–7), however when these acetanilides were replaced by
their equivalent pivanilides 1h–j then activity was restored (entries
8–10), although the yield obtained with 1j was low.
The introduction of meta-fluoro, -methyl, or -ethoxy sub-
stituents onto the acetanilide (1k–m) led to some para- chlorina-
tion (entries 11–13, 15), presumably by a competitive electrophilic
chlorination mechanism (entries 11–13). Indeed when the reaction
with the electron-rich substrate 1m was repeated in the absence
of palladium then a very similar product mixture was obtained
indicating that only electrophilic chlorination is operative in this
case (entry 14). By contrast re-running the reactions in entries 1,
4 and 11 without palladium yielded no products.
We next briefly examined the ortho-chlorinations of benzamides
and aryl carbamates and these results are summarised in Table 2.
In contrast to the palladium-catalysed ortho-C–H arylation of
benzamides, which requires an N–H for success,12 some—albeit
very modest—activity was seen with N,N-dimethylbenzamide
(entry 1). However the introduction of an N–H gave far superior
performance, possibly indicating that the ability to form an anionic
substrate is important. Good activity was seen with para- and
meta-substituted anilides (entries 3–5), but again some competitive
para-chlorination was seen with the 3-fluorobenzamide 3e (entry
5). By contrast the ortho-substituted benzamides 3f and g did not
react. Certain aryl carbamates, 5, were also found to react, however
the success of the reactions was dependent on the aromatic
substituents (entries 8–10).
We next examined the potential for reaction scale-up using
toluide 1a as the test substrate and the results from this study
are summarised in Scheme 4. We found that with reactions
performed on the 1 mmol scale, it was not always necessary to use
copper acetate (see mechanistic section below for more details)
and therefore we examined 10 mmol scale reactions in both the
presence and absence of this salt. It can be seen that in a small
diameter tube, optimal activity was obtained when using copper
acetate, however, in a larger diameter beaker with a greater surface
area in contact with the air, this extra oxidant proved unnecessary.
Under these conditions we were able to get excellent conversion to
product 2a using 100 mmol (14.9 g) of 1a. It should be noted that
when a beaker was used it was necessary to cover it with an inverted
filter funnel to capture any product that sublimed. Product work
up consisted simply of extraction, washing and recrystallisation.
Solvent-free brominations
The results from an optimisation study on the ortho-bromination
of toluide 1a with CuBr2 and Cu(OAc)2 (Table 3) showed that
reducing the palladium loading had no effect on the extent of
bromination, even when palladium was omitted altogether (entries
1–5). This strongly suggests that the reactions proceeded via simple
electrophilic substitution rather than C–H activation. Indeed
under optimised, palladium-free conditions (Table 3, entry 5) a
representative range of anilides all gave products consistent with
electrophilic bromination, rather than ortho-C-functionalisation
(Table 4).
The disparity between the chlorination reactions, which proceed
predominantly through directed C–H activation, and the bromi-
nation reactions, which proceed through simple electrophilic pro-
cesses, is in line with the results obtained recently by Stahl and co-
workers who demonstrated different mechanisms of bromination
and chlorination of electron-rich aromatics catalysed by copper
salts in acetic acid solution.13
Specifically they showed that the copper-catalysed aerobic
oxidation of bromide to bromine occurs under mild, acidic
conditions and that it is the resultant bromine that is responsible
for electrophilic bromination.
In Shi and co-workers study on the halogenation of anilides
using copper(II) bromide,8a it was concluded that bromination
proceeds via directed C–H functionalisation. This is in con-
trast with both the results of bromination under solvent-free
conditions obtained above and Stahl’s studies on electron-rich
aromatic substrates.13 We therefore conducted a comparison of
bromination reactions in both DCE under Shi’s conditions and
in the absence of solvent and these results are summarised in
Table 5.
With 1d as substrate under solvent-free conditions in the absence
of palladium we obtained p-brominated 7c as the sole product
(entry 1). When the reaction was repeated in the presence of
palladium then substantial, competitive ortho-bromination was
observed (products 7g and h, entry 2) indicating that a C–H
functionalisation/bromination is viable. In general, the reactions
generate substantial amounts of acetic acid and when this reaction
was repeated in a closed vial then much less ortho-bromination was
observed (entry 3). This is consistent with Stahl’s observation that
acid is required for the copper-catalysed formation of bromine
from bromide under mild conditions.13
In Shi’s original report, all the brominations were conducted
on para-substituted anilides, except for the reaction of 1l. We
found that under solvent- and palladium-free conditions only the
para-brominated product 7b was produced to any great extent
(entry 5). The same product was furnished when the reaction
was repeated in the presence of palladium (entry 6), although in
this case we also obtained 7i, again presumably via a competitive
ortho-C–H functionalisation. Repeating the reaction under Shi’s
conditions in DCE solution either in the presence or absence
of palladium (entries 7 and 8) in our hands yielded only 7b via
electrophilic bromination. This was confirmed unequivocally by a
single crystal X-ray structure (Fig. 1) of the isolated product from
the reaction performed under Shi’s conditions (entry 7). A re-
examination of the 1H and 13C NMR spectroscopic data reported
for the claimed ortho-brominated species 7j8a showed them to be
very close to those for 7b14,15 (with the exception of the NH peak in
a
Scheme 4 Large scale ortho-chlorination. Spectroscopic yield (1H
NMR). b Covered with inverted filter funnel. c Isolated yield.
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 10464–10472 | 10465
©

Table 1 Solvent-free palladium-catalysed ortho-chlorination of anilidesa
Entry
1
Substrate
Product/s
Yieldb (spec.)c
93 (20)d (0)e
2
3
4
5
84 (>99)
32
57 (85)
(82)f
(0)e
No reaction
No reaction
No reaction
6
7
8
98
66
9
10
11
26 g
2h: 64
2i: (14)
2h: (53)f
2h: (0)e
2i: (0)e
12
13
2j: 63
2k: (11)
2j: (66)f
2l: 55
2m: 25
2n: 14
14
1m
2l: (53)e
2m: (34)e
2n: (13)e
a Conditions: Anilide (1.0 mmol), Cu(OAc)2 (2.0 mmol), CuCl2 (2.0 mmol), Pd(OAc)2 (5 mol%), ground then 120 ◦C, 2 h. b Isolated yield. c Determined
by 1H NMR, 1,3,5-C6H3(OMe)3 internal standard. d 2 mol% Pd(OAc)2. e No Pd(OAc)2. f No Cu(OAc)2. g 26 h.
10466 | Dalton Trans., 2010, 39, 10464–10472
This journal is
The Royal Society of Chemistry 2010
©

Table 2 Solvent-free palladium-catalysed chlorination of benzamides and aryl carbamatesa
Entry
1
Substrate
Product/s
Yield (spec. yield)b
(17)
2
3
4
69
(47)c
57
(52%)c
76
5
4e: 10d (46)
4f: 27
4g: (14)
6
7
No reaction
No reaction
8
9
60 (70) (6)c
48
10
No reaction
a Conditions: benzamide or carbamate (1.0 mmol), Cu(OAc)2 (2.0 mmol), CuCl2 (2.0 mmol), Pd(OAc)2 (5 mol%), ground then 120 ◦C, 18 h. b Determined
by 1H NMR, 1,3,5-C6H3(OMe)3 internal standard. c No Cu(OAc)2. d 16 h.
the 1H NMR spectrum) and quite distinct from those of a genuine
sample of 7j.14,15 We therefore suggest that in both solution or
solvent free, in the presence or absence of palladium, under inert
or aerobic atmosphere, the predominant bromination products
obtained from Cu(OAc)2/CuBr2 mixtures are those resulting from
simple electrophilic bromination. This can be mitigated to some
extent, allowing partial ortho-C–H bromination, by ensuring that
the reactions are performed solvent free, in an open reaction
vessel.
Mechanistic considerations
(a) Intermediation of palladacyclic species. Palladacycles
are postulated to be intermediates in the ortho-halogenation of
anilides with CuX2/Cu(OAc)2 in DCE solution8a and it seemed
likely that such species could also play a role in the solvent-free
chlorination reactions described herein. Indeed the reactions of
1a or b with palladium acetate under solvent-free conditions
yielded a mixture of starting anilide and the palladacycles 8a and
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 10464–10472 | 10467
©

Table 3 Optimisation of the bromination of anilide 1aa
Table 5 Comparison of bromination in DCE and under solvent-free
conditionsa
Entry Mol% Pd MCl
m
n
Temp., ◦ Time, h Spec. Yield, %b
C
1
2
3
4
5
6
7
8
5
1
0.1
0.01
0
0
0
0
0
0
0
0
0
CuBr2
CuBr2
CuBr2
CuBr2
CuBr2
CuBr2
LiBr
NaBr
KBr
CuBr2 0.1
CuBr2
CuBr2
CuBr2
CuBr2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
4
4
4
2
2
1
2
1
120
120
120
120
120
100
120
120
120
120
120
120
120
120
1
1
1
1
>99 c
>99
93
>99
>99
87
46
0
0
65
Entry
Conditions
Spec. yield a
1
R = H (1d)
15
22
1
1
1
1
1
1
1
7c
>99
40
7g
0
20
7h
0
40
1
2
Solvent free, Pd free
Solvent free, 5 mol%
Pd(OAc)2
9
10
11
12
13
14
3
4
Solvent free, 5 mol%
Pd(OAc)2, closed tube
Solvent free, 5 mol%
Pd(OAc)2, closed tube,
80 ◦C, 18 h
85
81
15
12
0
5
0
2
1
1
50
50
64
48
0
a Conditions: 1a (1.0 mmol), Cu(OAc)2, CuBr2, Pd(OAc)2, ground then
120 ◦C, time specified. b Determined by 1H NMR, 1,3,5-C6H3(OMe)3
internal standard. c Isolated yield.
R = Me (1l)
7b
82
78b
7i
7j
0
0
5
6
Solvent free, Pd free
Solvent free, 5 mol%
Pd(OAc)2
trace
21b
Table 4 Bromination of anilidesa
7
Pd(OAc)2 (10 mol%),
DCE, 90 ◦C, 48 h,
argon
80b
trace
trace
0
Entry
1
Substrate
Product/s
Yield (%)
82b
8
Pd free, DCE, 90 ◦C,
48 h, argon
80
0
a Determined by 1H NMR, 1,3,5-C6H3(OMe)3 internal standard. b Isolated
yields.
2
3
97
84
4
5
92
75
a Conditions: Anilide (1.0 mmol), Cu(OAc)2 (2.0 mmol), CuBr2 (2.0 mmol),
ground then 120 ◦C, 1 h. b 0.5 mmol anilide scale.
Scheme 5 Formation and reactivity of palladacycles. a Spectroscopic yield
(determined by 1H NMR, 1,3,5-C6H3(OMe)3), remainder unreacted 1.
b16 respectively after 2 h at 80 ◦C, as determined by 1H
NMR spectroscopy (Scheme 5). It is apparent that under these
conditions the bulkier substrate 1b was more reactive.
Both complexes could be prepared in better yield by the reaction
of palladium acetate with 1a or b in acetic acid. The structure of
complex 8a was determined by a single crystal X-ray diffraction
study, the results of which are summarised in Fig. 2.
In order to determine whether palladacycle 8a is indeed
significant in catalysis we next compared the rates of ortho-
chlorination of 1b with either palladium acetate or 8a (Fig. 3).
Fig. 1 X-ray structure of 7b formed from reaction in DCE solution,
thermal ellipsoids set at the 30% probability level.
10468 | Dalton Trans., 2010, 39, 10464–10472
This journal is
The Royal Society of Chemistry 2010
©

10-4 mmol s-1 respectively).§ At both catalyst loadings 8a showed
shorter induction times than palladium acetate, consistent with
the formation of the palladacycle in situ in these latter cases. The
induction periods observed with 8a were probably due to a lag in
heat transfer through the stationary powder.
It is apparent from Scheme 5 that the formation of the
palladacyclic intermediate from the smaller substrate 1a was less
efficient than with bulkier 1b. Indeed a competition reaction
revealed that the rate of formation of the palladacycle from 1b
is nearly twice that from 1a (Scheme 5). The same size effect was
noted above in the catalytic reactions, with pivanilides showing
higher yields than equivalent acetanilides (compare Table 2 entries
1, 3–7 with entries 2, 8–10). This was further reflected in the rates
of chlorination of 1a and 1b (Fig. 4), with the latter reaction
over ~ 3 times slower (rate of starting material consumption:
-0.55) ¥ 10-4 mmol s-1 v. -2.4 ( 0.22) ¥ 10-4 mmol s-1 respectively).
Doubling the amount of pivanilide 1b had no appreciable effect
on the rate.
Fig. 2 X-ray structure of complex 8a (one of two similar molecules in
asymmetric unit), thermal ellipsoids set at the 30% probability level.
Fig. 4 Varying anilide substrate. Solid lines indicate best fit over linear
regions used to determine maximum rates. Conditions: 1b or b (1.0 or
2.0 mmol), Cu(OAc)2 (2.0 mmol), CuCl2 (2.0 mmol), Pd(OAc)2 (5 mol%),
ground then 120 ◦C. Substrate (trend line R2): 1b, 1.0 mmol (0.993);
1a, 1.0 mmol (0.986); 1b, 2.0 mmol (0.995).
Fig. 3 Varying palladium source and loading. Solid lines indicate best
fit over linear regions used to determine maximum rates. Conditions: 1b
(1.0 mmol), Cu(OAc)2 (2.0 ◦mmol), CuCl2 (2.0 mmol), Pd catalyst (2.5 or
5 mol%), ground then 120 C. Catalysts and loadings (trend line R2):
8a, 5 mol% Pd (0.995); Pd(OAc)2, 5 mol% Pd (0.994); 8a, 2.5 mol% Pd
(0.989); Pd(OAc)2, 2.5 mol% Pd (0.993). b Conversion to 2b, determined
by 1H NMR.
(b) Roles of copper salts. Having established that palladium
catalysis is essential for the directed chlorination reactions (al-
though it plays little role in bromination reactions) and that the
reaction proceeds via the formation of a palladacycle, we decided
to investigate the roles of the copper salts in the chlorinations in
more detail, in the first instance using 1a as the substrate (Table 6).
Under aerobic conditions reducing copper acetate loadings had
no affect on activity, even when the salt was omitted altogether
(entries 1–3). By contrast when the reactions were repeated under
an inert atmosphere (entries 4 and 5) then it is apparent that
copper acetate can function as a reasonably efficient stoichiometric
oxidant as can copper chloride alone, although to a more limited
extent.
There was essentially no difference observed between the rates
of formation of 2b using either 8a or palladium acetate at
5 mol% Pd loading (7.2 ( 0.4) ¥ 10-4 mmol s-1 v. 7.1 ( 0.5) ¥
10-4 mmol s-1 respectively).§ Reducing the palladium loading to
2.5 mol% led to an approximate halving of the rate in both cases
indicating that the reactions are first order in palladium. Again
there was no significant rate difference between the palladacycle
and palladium acetate (3.3 ( 0.25) ¥ 10-4 mmol s-1 v. 3.0 ( 0.2) ¥
When the equivalent reactions where run in DCE solution
under Shi’s conditions but in the absence of copper acetate, either
under air or argon, then the same modest performance resulted
(entries 9 and 10) indicating that in solution Cu(OAc)2 is required
§ Note maximum rates cannot be expressed as [M]/s in the powder phase.
Maximum rates in all cases are for reactions based on varying one
component from ‘standard’ conditions of 1 mmol substrate, 2 mmol each
of the copper salts and 5 mol% catalyst.
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 10464–10472 | 10469
©

Table 6 Effect of varying copper salts loadings
Entry MCl
m
n
Temp., ◦
C
Time, h
Spec. Yield, %a
1
2
3
4
5
6
7
8
CuCl2
CuCl2 0.5
CuCl2
CuCl2
CuCl2
LiCl
LiCl
LiCl
CuCl2
CuCl2
2
2
2
2
2
2
4
4
4
2
2
120
120
120
120
120
120
120
120
120
120
2
2
2
2
97b
96
0
2
0
2
0.5
0
0
98
79c
28c
70
2
20
20
20
2
23
0
9
10
49d,e
47d,c
0
2
a Determined by 1H NMR, 1,3,5-C6H3(OMe)3 internal standard. b Isolated
yield. c Under argon. d DCE, 90 ◦C, 48 h. e Under air.
Fig. 5 Varying copper salt amounts. Solid lines indicate best fit over
linear regions used to maximum determine rates. Conditions: 1b (1 mmol),
Pd(OAc)2 (0.05 mmol), CuCl2 (2 or 4 mmol), Cu(OAc)2 (2 or 4 mmol),
ground then 120 ◦C, 2h. Copper salt loadings (trend line R2): CuCl2 and
Cu(OAc)2, 2 mmol (0.994); CuCl2, 4 mmol, Cu(OAc)2, 2 mmol (0.972);
CuCl2, 2 mmol, Cu(OAc)2, 4 mmol (0.993).
for optimal performance. Comparing entries 3 and 9 it can be
concluded that the oxidation step is more facile in the solid state.
CuCl2 could be replaced by LiCl under solvent-free conditions
with reasonable results, provided that copper acetate was present.
Reducing the amount of copper acetate in this case proved to
be deleterious and no activity was seen in the absence of copper,
demonstrating that copper must be present for the oxidative step
to occur efficiently under air.
We next investigated the effect of removing the copper acetate
from selected catalytic reactions described above. The reaction
with acetanilide, 1c, was largely unaffected (Table 1, entry 4) whilst
the reactions of the 3-substituted substrates 1k and 1l showed no
competitive electrophilic chlorination in the absence of copper
acetate (Table 1, entries 11 and 12). The reactions of acetamides
3b and c showed only a slight reduction in yield (Table 2, entries 2
and 3), by contrast leaving out copper acetate proved to be highly
deleterious with the carbamate substrate 5a (Table 2, entry 8).
Fig. 5 shows the effect of changing copper salt loadings on
the rate of ortho-chlorination of 1b. Doubling the amounts of
either of the copper salts led to a significant reduction in rate,
interestingly the new rates were essentially identical, irrespective
of which copper salt was increased (4 equiv. CuCl2 = 4.9 (
0.55) mmol s-1; 4 equiv. Cu(OAc)2 = 4.9 ( 0.3) mmol s-1).§ A
likely explanation for this, given that copper acetate appears to
play no role under aerobic conditions in the powder phase, is that
the rate retardation was due to a lower rate of diffusion in the
reaction mixture in each case.
that undergoes the elimination step. Reductive elimination of aryl
chlorides from Pd(II) is rare, typically facilitated by very bulky
phosphine ligands.17 Consequently alternative mechanisms have
been invoked for ortho-halogenation that proceed via reductive
elimination from palladacyclic intermediates in oxidation state IV
or III.18 At present we do not have sufficient evidence to determine
the likely oxidation state of intermediates in our case but are
currently investigating this further.
Conclusions
In summary we have demonstrated that a range of anilides, benza-
mides and carbamates can be ortho-chlorinated under solvent free
conditions using palladium catalysis and copper chloride under
air and that the process can be readily scaled to 100 mmol of
substrate. In contrast to equivalent reactions performed in DCE
solution, the use of copper acetate as a co-oxidant is not required
in most instances. Bromination reactions in either the solid state
or DCE solution proceed via electrophilic bromination rather
than sequential C–H activation/bromination. Mechanistic studies
reveal that the chlorination reactions proceed via the formation of
Pd(II) metallacycles followed by reductive elimination from as yet
uncharacterised intermediates.
The fact that copper acetate plays no role in the chlorination
of 1b in the absence of solvent under air enabled us to probe
the intimate role of CuCl2 further. Heating palladacycle 8a with
4 equivalents of CuCl2 under solvent-free conditions gave no
reaction after 4 h, but when this was repeated with added acetic
acid then quantitative liberation of 2b was observed (Scheme 5).
Presumably whilst the reaction mixture is still a powder, the
addition of acetic acid allows sufficient diffusion to allow reaction
to occur. In all the reactions reported here some acetic acid is
produced, either from palladium or copper acetate.
We thank the EPSRC and GSK for funding under the collabo-
rative EPSRC Programme for Synthetic Organic Chemistry with
AZ-GSK-Pfizer.
Experimental
General procedure for the preparation of 2-chloroarenes using
CuCl2
To a mortar was added the substrate (1.0 mmol), copper acetate
(363 mg, 2.0 mmol), copper chloride (266 mg, 2.0 mmol) and
Pd(OAc)2 (0.05 mmol). The reagents were ground with a pestle to
form a homogenous powder. The reaction mixture was transferred
While this data provides evidence for the formation of the
product 2b from the palladacycle 8a by reductive elimination, it
yields no information on the nature of the palladium intermediate
10470 | Dalton Trans., 2010, 39, 10464–10472
This journal is
The Royal Society of Chemistry 2010
©

to a thick walled glass vial. The vessel was then heated at 120 ◦C for
the appropriate length of time, under an atmosphere of air without
stirring. The vial was then allowed to cool to room temperature,
mixed with a spatula to form a free-flowing homogenous powder,
and added directly to the top of a chromatography column.
The product was purified by flash chromatography, eluting with
ethylacetate/hexanes.
Table 7 Crystallographic data for 7b complex 8a
7b
8a
Size/mm
0.48 ¥ 0.22 ¥ 0.12 0.36 ¥ 0.18 ¥ 0.18
Empirical Formula (excluding C9H10BrNO
unknown solvent molecules)
C189H270Cl6N12O37Pd12
M
228.09
Monoclinic
4791.67
Crystal system
Trigonal
¯
Space group
P21/c
R3
˚
a/A
4.0401(10)
23.7100(5)
9.7444(2)
90.00
44.9022(6)
44.9022(6)
25.3871(7)
90.00
˚
b/A
General procedure for the preparation of bromoanilides using
CuBr2
˚
c/A
a (◦)
b (◦)
g (◦)
100.007(1)
90.00
919.22(4)
4
4.422
100
90.00
120.00
44328.1(15)
6
0.814
100
0.91,27.49
To a mortar was added anilide (1 mmol), copper acetate (363 mg,
2 mmol) and copper bromide (447 mg, 2 mmol). The reagents
were ground with a pestle to form a homogenous powder. The
reaction mixture was transferred to a thick walled glass vial. The
vessel was then heated at 120 ◦C for the appropriate length of
time, under an atmosphere of air without stirring. The vial was
then allowed to cool to room temperature, mixed with a spatula
to form a free-flowing homogenous powder, and added directly to
the top of a chromatography column. The product was purified
by flash chromatography, eluting with ethylacetate/hexanes.
3
˚
V/A
Z
m/mm-1
T/K
qmin,max
Completeness
1.72,32.94
0.998 to q = 27.50◦ 0.998 to q = 27.49◦
Reflections: total/independent 10973/3192
22570/167944
0.0439
0.0465, 0.1254
1.097, -0.875
Rint
0.0320
0.0291, 0.0694
0.938, -0.738
Final R1 and wR2
-3
˚
Largest peak, hole/e A
Procedure for the preparation of palladacycle 8a
from a single crystal mounted on a glass fibre. Intensities were
integrated19 from several series of exposures measuring 0.5◦ in w
or j. An absorption correction was applied based on equivalent
reflections using SADABS.20 The structure was solved using
ShelXS and refined against all Fo2 data with hydrogen atoms riding
in calculated positions using SHELXL.21 Crystal structure and
refinement data are given in Table 7.
The appropriate anilide (4 mmol) was added to a round-bottom
flask along with Pd(OAc)2 (4 mmol) and acetic acid (3.5 ml). The
reaction mixture was heated to 80 ◦C and stirred under air at
this temperature for 2 h. The solvent was removed in vacuo and
chloroform (15 ml) was added. The vessel was cooled to 0 ◦C
to allowed complete precipitation of the product. The mother
liquor was removed by cannula filtration through glass paper. The
precipitate was washed a further 3 times with cold chloroform.
The precipitate was dried in vacuo. 8a: Green solid, 1.27 g (89%);
1H NMR (400 MHz; CD2Cl2) d 7.71 (br. s, 1H), 7.04 (s, 1H), 6.81
(dd, J 7.9, 1.3 Hz, 1H), 6.48 (d, J 7.9 Hz, 1H), 2.24 (s, 3H), 2.09
(s, 3H), 0.87 (s, 9H); 13C NMR (100 MHz; CD2Cl2) d 180.7, 173.3,
134.7, 132.3, 128.9, 125.3, 118.1, 114.5, 38.2, 26.6, 24.0, 20.7; IR
neat, n/cm-1 3436, 2971, 1611, 1571, 1532, 1471, 1410, 746, 688,
666; m.p. (decomp.) 197.0 ◦C; Anal. calcd. for C28H38N2O6Pd2: C,
47.3; H, 5.4; N, 3.9. Found: C, 47.3; H, 5.0; N, 3.5.
X-Ray structure determination of complex 8a
Crystals of 8a suitable for X-ray analysis were grown by slow
diffusion of a dichloromethane solution layered with cyclohexane.
The X-ray diffraction was carried out at 100 K on a Bruker
˚
APEX II diffractometer using Mo-Ka radiation (l = 0.71073 A).
The data collection was performed using a CCD area detector
from a single crystal mounted on a glass fibre. Intensities were
integrated19 from several series of exposures measuring 0.5◦ in w
or j. An absorption correction was applied based on equivalent
reflections using SADABS,20 and structures were refined against
2
Procedure for kinetic studies
all Fo data with hydrogen atoms riding in calculated positions
using SHELXL.22 Crystal structure and refinement data are given
in Table 7. The structural model was found to contain residual
electron density which could not be satisfactorily resolved, and
so the data were modelled using the SQUEEZE algorithm, as
implemented in PLATON.21 The estimate of the number of extra
electrons in the unit cell is 6606, which approximately equates to
an extra 184 electrons per molecule of complex in the crystal.
Since there is also ordered cyclohexane and dichloromethane in
the structure, it is difficult to say with certainty to what the residual
electron density relates.
The appropriate quantity of anilide, CuCl2, Cu(OAc)2 and palla-
dium source were mixed to form a homogeneous powder using
a pestle and mortar. The mixture was transferred to a thick-
walled vial and placed in an oil bath, pre-heated to 120 ◦C. A
small quantity of the mixture was removed using a spatula at the
appropriate time interval and dissolved in EtOAc. The solution
was passed through a plug of glass paper, the solvent removed
1
under reduced pressure and the sample analysed by H NMR
(CDCl3).
X-Ray structure determination of complex 7b
References
Crystals of 7b suitable for X-ray analysis were grown by slow
diffusion of a dichloromethane solution layered with cyclohexane.
The X-ray diffraction was carried out at 100 K on a Bruker
APEX II diffractometer using Mo-Ka radiation (l = 0.71073 A).
Data collections were performed using a CCD area detector
1 A. de Meijere and F. Diederich, Metal-Catalyzed Cross-Coupling
Reactions, 2nd ed., Wiley-VCH, Weinheim, 2004.
2 For recent overviews, see: (a) Selective Functionalization of C–H
Bonds (thematic issue), Chem. Rev., 2010, 110, 575–1211; (b) Catalytic
Aromatic C–H Activation (Symposium in Print), Tetrahedron, 2008,
˚
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 10464–10472 | 10471
©

64, 5963–6138; (c) Handbook of C–H Transformations, ed. G. Dyker,
Wiley-VCH, Weinheim, 2005.
9 X. Chen, X.-S. Hao, C. E. Goodhue and J.-Q. Yu, J. Am. Chem. Soc.,
2006, 128, 6790.
10 R. B. Bedford, C. J. Mitchell and R. L. Webster, Chem. Commun., 2010,
46, 3095.
11 Reducing the palladium loading to 2 mol% proved deleterious (entry 1),
presumably due to mass-transport limitation in the reaction mixture.
12 Y. Kametani, T. Satoh, M. Miura and M. Nomura, Tetrahedron Lett.,
2000, 41, 2655.
13 L. Yang, Z. Lub and S. S. Stahl, Chem. Commun., 2009, 6460.
14 Genuine samples of 7a and 7j were prepared by acetylation of the
commercially available bromotoluides.
3 (a) X. Chen, K. M. Engle, D.-H. Wang and J.-Q. Yu, Angew. Chem.,
Int. Ed., 2009, 48, 5094; (b) D. Alberico, M. E. Scott and M. Lautens,
Chem. Rev., 2007, 107, 174; (c) I. V. Seregin and V. Gevorgyan, Chem.
Soc. Rev., 2007, 36, 1173; (d) K. Godula and D. Sames, Science, 2006,
312, 67.
4 (a) T. W. Lyons and M. S. Sanford, Chem. Rev., 2010, 110,
1147.
5 (a) D. R. Fahey, J. Chem. Soc. D, 1970, 417; (b) D. R. Fahey,
J. Organomet. Chem., 1971, 27, 283.
6 (a) A. R. Dick, K. L. Hull and M. S. Sanford, J. Am. Chem. Soc.,
2004, 126, 2300; (b) K. L. Hull, W. Q. Anani and M. S. Sanford, J. Am.
Chem. Soc., 2006, 128, 7134; (c) D. Kalyani, A. R. Dick, W. Q. Anani
and M. S. Sanford, Tetrahedron, 2006, 62, 11483.
15 See ESI§ for full spectroscopic analysis.
16 Complex 8b has been reported previously, see: (a) H. Horino and N.
Inoue, 1979, 26, p. 2403; (b) H. Horino and N. Inoue, J. Org. Chem.,
1981, 46, 4416.
7 For related directed aromatic C–H iodination with I2(PhI(OAc)2 see:
(a) R. Giri, X. Chen and J.-Q. Yu, Angew. Chem., Int. Ed., 2005, 44,
2112; (b) R. Giri, X. Chen, X-S. Hao, J.-J. Li, J. Liang, Z.-P. Fan and
J.-Q. Yu, Tetrahedron: Asymmetry, 2005, 16, 3502; (c) J.-J. Li, R. Giri
and J.-Q. Yu, Tetrahedron, 2008, 64, 6979; (d) T.-S. Mei, R. Giri, N.
Maugel and J.-Q. Yu, Angew. Chem., Int. Ed., 2008, 47, 5215.
8 (a) X. Wan, Z. Ma, B. Li, K. Zhang, S. Cao, S. Zhang and Z. Shi, J. Am.
Chem. Soc., 2006, 128, 7416; (b) X. Zhao, E. Dimitrijevic´ and V. M.
Dong, J. Am. Chem. Soc., 2009, 131, 3466; (c) B. Song, X. Zheng, J.
Mo and B. Xu, Adv. Synth. Catal., 2010, 352, 329.
17 A. H. Roy and J. F. Hartwig, J. Am. Chem. Soc., 2001, 123, 1232.
18 (a) S. R. Whitfield and M. S. Sanford, J. Am. Chem. Soc., 2007, 129,
15142; (b) D. C. Powers and T. Ritter, Nat. Chem., 2009, 1, 302; (c) N. R.
Deprez and M. S. Sanford, J. Am. Chem. Soc., 2009, 131, 11234.
19 Bruker (2008) ApexII, Bruker AXS Inc. Madison, Wisconsin, USA.
20 G. M. Sheldrick, SADABS V2008, University of Go¨ttingen, Germany.
21 G. M. Sheldrick, Acta Crystallogr., Sect. A: Found. Crystallogr., 2008,
64, 112–122.
22 SHELXTL program system version V6.14, Bruker Analytical X-ray
instruments Inc. Madison, WI, 1998.
10472 | Dalton Trans., 2010, 39, 10464–10472
This journal is
The Royal Society of Chemistry 2010
©