Copper-catalysed aromatic-Finkelstein reactions with amine-based ligand systems

Article abstract of DOI:10.1039/c7cy00538e

A new efficient and low-cost ligand, diethylenetriamine, has been utilised to promote the iodination of 16 different bromo-substrates via the copper catalysed Finkelstein halogen exchange reaction under mild conditions. In contrast to earlier methods, the use of inert atmosphere conditions was not required to obtain high yields and purity. Studies on the speciation of the catalyst in solution indicate rapid disproportionation of copper(i) in the presence of diethylenetriamine to give copper(0) and a bis-ligated copper(ii) complex which is characterised by X-ray diffraction. This copper(ii) complex was also shown to be catalytically active in the halogen exchange reaction. In contrast, no significant disproportionation was observed using dimethylethylenediamine as the ligand, and the solid-state structures of a copper(i) dimeric complex and a 2D polymeric network of copper(i) iodide tetramers are reported. The catalytic activity of diethylenetriamine and dimethylethylenediamine with both copper(i) and copper(ii) salts are compared, and possible mechanistic implications discussed.

Full text of DOI:10.1039/c7cy00538e

Catalysis
Science &
Technology
View Article Online
View Journal
PAPER
Copper-catalysed aromatic-Finkelstein reactions
with amine-based ligand systems†
Cite this: DOI: 10.1039/c7cy00538e
*
Xiaodong Jin and Robert P. Davies
A new efficient and low-cost ligand, diethylenetriamine, has been utilised to promote the iodination of 16
different bromo-substrates via the copper catalysed Finkelstein halogen exchange reaction under mild
conditions. In contrast to earlier methods, the use of inert atmosphere conditions was not required to ob-
tain high yields and purity. Studies on the speciation of the catalyst in solution indicate rapid disproportion-
ation of copperIJI) in the presence of diethylenetriamine to give copper(0) and a bis-ligated copperIJII) com-
plex which is characterised by X-ray diffraction. This copperIJII) complex was also shown to be catalytically
active in the halogen exchange reaction. In contrast, no significant disproportionation was observed using
dimethylethylenediamine as the ligand, and the solid-state structures of a copperIJI) dimeric complex and a
2D polymeric network of copperIJI) iodide tetramers are reported. The catalytic activity of diethyl-
enetriamine and dimethylethylenediamine with both copperIJI) and copperIJII) salts are compared, and possi-
ble mechanistic implications discussed.
Received 20th March 2017,
Accepted 20th April 2017
DOI: 10.1039/c7cy00538e
rsc.li/catalysis
Klapars and Buchwald in 2002 which successfully overcame
some of these drawbacks.16 More recently, diamine ligated
Introduction
Aryl halides are much studied and valued chemicals, being
both targets and building blocks in many synthesis endeav-
ours. They occur widely in natural products, agrochemicals,
pharmaceuticals and materials.1 In addition the carbon–halo-
gen bond is an important synthetic precursor for a wide range
of transition-metal catalysed coupling reactions, most notably
for the formation of new carbon–carbon and carbon–hetero-
atom bonds.2,3 Although aryl chlorides and bromides are gen-
erally cheaper with wider commercial availability than the cor-
responding iodides, they suffer from much lower reactivity
due to the stronger aryl-halide bond.4 This can limit or pro-
hibit their utility in many bond-forming reactions.
CuIJI) systems have been further adapted using microwave as-
sistance to reduce the reaction time17 or incorporated into
continuous-flow processes using a sodium iodide packed-bed
reactor.18
Despite these recent advances, the mechanism of the
copper-catalysed Finkelstein reaction remains opaque.
CopperIJI) salts are by far the most widely used pre-catalysts,
and the general consensus in the literature is for an oxidative
addition/reductive elimination mechanism in which the cata-
lyst cycles between
a CuIJI) and CuIJIII) oxidation state
(Scheme 1).4,19,20 In order to protect the CuIJI) catalyst from
The aromatic Finkelstein reaction is a halide exchange re-
action which can be used to convert the less reactive aryl-
chlorides or bromides into the corresponding aryl-iodide
using either a nickel or copper based catalyst. In early re-
ports, both the copper and nickel mediated processes re-
quired harsh reaction conditions (temperatures in excess of
150 °C) and stoichiometric catalyst loadings. In addition,
reported yields were often poor for the replacement of lighter
halogen atoms with heavier ones (aryl-Cl/aryl-Br to aryl-I
conversion).5–15 An efficient copper-catalysed coupling of aryl
bromides promoted by diamine ligands was reported by
Department of Chemistry, Imperial College London, South Kensington, London,
SW7 2AZ, UK. E-mail: r.davies@imperial.ac.uk
† Electronic supplementary information (ESI) available: See ESI for full details of
all experimental methods. CCDC 1538590–1538592. For ESI and crystallographic
data in CIF or other electronic format see DOI: 10.1039/c7cy00538e
Scheme
1 Proposed halide exchange reaction mechanism via
oxidative addition.19
This journal is © The Royal Society of Chemistry 2017
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
being deactivated by oxidation these reactions are usually car-
ried out under vigorous air and moisture free conditions.
There are some obvious parallels between the proposed
mechanism for the aromatic Finkelstein reaction (Scheme 1)
and that of copper-catalysed carbon–nitrogen and carbon–ox-
ygen bond formation via Ullmann catalysis. In both cases the
oxidative addition of the aryl halide to a CuIJI) centre to give a
CuIJIII) intermediate is a key (often turnover limiting) step.21
Although there have been many recent studies on the mecha-
nism and scope of the Ullmann reaction, both by us22,23 and
others,24,25 this has not translated to date to the related
copper-catalysed Finkelstein reaction.
In order to begin to address this, we report in this paper
on a ligand screening program to optimise and improve the
copper-catalysed Finkelstein halide exchange reaction. Al-
though effective for a range of aryl-bromide and -iodide sub-
strates, the best currently employed ligand systems, DMEDA
and N,N-dimethyl-1,2-cyclohexanediamine,16,26 still require
high temperatures and vigorous air-free conditions, show
poor or no reactivity with aryl-chlorides, and are relatively ex-
pensive (Scheme 2). In this paper a new and much less ex-
pensive ligand system, diethylenetriamine, has been shown
to offer comparable reaction yields for a wide range of
bromo-substrates and without the requirement of inert-
atmosphere conditions. Further studies have also been un-
dertaken in order to better understand the mechanism of the
reaction and the potential role played by the ligand.
Fig. 1 Ligands studied and corresponding yields. Reaction conditions:
1a (1 mmol) was reacted with NaI (2 mmol) in dry dioxane at 110 °C in
a sealed vial for 24 hours using CuI (5 mol%) and L1–L8 (10 mol%) as
catalyst. Yields were determined by GC and averaged on two runs. All
reactions were carried out under N2.
Results & discussion
Screening of ‘Ullmann reaction’ ligands
were chosen since although NaI is soluble in both of these
solvents, NaBr is only partially soluble. This therefore favours
the desired bromo- to iodo-exchange reaction over the com-
peting reverse iodo- to bromo-exchange.16,28
Given the similarities between the proposed mechanisms for
the copper-catalysed Finkelstein and Ullmann reactions (see
Introduction), a range of ligands that have recently been con-
firmed to be effective in the Ullmann reaction were initially
selected for screening (Fig. 1).27 The choice of solvent was
thought to be of high importance: dioxane and acetonitrile
The results from the initial ligand screening in dioxane
(Fig. 1) showed L1 to give almost quantitative conversion
(consistent with the literature results for this ligand16), but
all other ligands (L2–L8) gave surprisingly zero or close to
zero conversions. These reactions were repeated using aceto-
nitrile as the solvent but with no improvement in the ob-
served yields (see ESI†). The poor solubility of CuI in the sol-
vent/ligand mixture for all ligands except L1, may at least in
part account for these results. Moreover, the reverse reaction
carried out in DMSO (in which NaI, NaBr and CuI all show
good solubility), for the conversion of ArI to ArBr using NaBr
as the bromide source also gave low yields (less than 2%) for
L2–L8. Thus, the poor yields in Fig. 1 cannot be attributed to
the dominance of the reverse reaction over the forward reac-
tion. L1 on the other hand successfully promoted the reverse
reaction under these conditions giving the aryl-iodide in 74%
yield (see ESI†). It is therefore apparent that effective ligands
reported for the Ullmann reaction are not readily transpos-
able to the Finkelstein reaction, despite the purported similar
mechanisms of these two reactions. Focus was therefore
switched to the study of amine-based ligands related to or de-
rived from L1 in order to seek further improvements.
Scheme 2 Comparison between previous studies16 and this work.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2017

View Article Online
Catalysis Science & Technology
Paper
Screening of amine-based ligands
Based on the good performance of dimethylethylenediamine
(L1) in the initial screening reactions, a range of tetramine,
triamine, diamine and monoamine ligands were selected for
further study (Fig. 2). The results from these studies reveal
many of the ligands to be inactive for the Finkelstein reac-
tion. This is despite the good solubility of CuI in these li-
gand/solvent mixtures. The best performing ligands were ob-
served to be those originally reported by Buchwald, L1 and
L14,16 and the new triamine ligands L19, L21 and L22. Based
on these results we are able to speculate on the requirements
for a good ligand candidate: 1. two or more amine groups; 2.
at least one NH group; 3. NHMe is better that NH2 which is
better than NH(R) where R is bulky. The reasons behind
these selectivity criteria are not yet clear, however it is likely
that good coordination (chelation) of the metal centre is re-
quired with too much steric bulk (including tertiary amine
centres) inhibiting the assumed oxidative addition step with
ArBr. The preference of NHMe over NH2 could possibly be
rationalised by the increased electron density on the nitrogen
atom which can help to stabilise the putative copperIJIII) inter-
mediate. The opposite is true in L15 where the electron den-
sity is lower on the nitrogen atoms due to delocalisation onto
the aryl ring, thus leading to an absence of catalytic behav-
iour. The tendency of L23 and L24 to coordinatively saturate
the copper centre29 may explain their lower performance
when compared to L1 and L19.
Optimisation of the reaction with diethylenetriamine (L19)
Although the original Buchwald ligands (L1 and L14) gave
the highest yields, the performance of L19 in acetonitrile was
also promising. Given that L19 is significantly less expensive
than either L1 or L14 (L19 £5.40 per mol, L1 £465.40 per mol,
L14 £2625.60 per mol at time of writing)30 further attempts
were made to optimise and improve the reaction using this li-
gand (Table 1). At slightly higher temperatures (110 °C) close
to quantitative conversions were obtained with L19 (entry 2).
Halving the catalyst and ligand loading to 2.5 and 5 mol%
respectively, only slightly diminished the yield (entry 3). Low-
ering the amount of NaI from 2 to 1.2 equivalents had no ef-
fect, still giving 99% yield (entry 4). Interestingly the CuI/L19
catalytic system shows excellent air- and water-tolerance (en-
tries 5 & 6). In comparison, the introduction of any air or
moisture to CuI with L1 or L14 either under Buchwald's origi-
nal conditions (dioxane, 110 °C) or under these new condi-
tions (acetonitrile, 110 °C)† led to a significant drop in cata-
lytic activity (Fig. 3).
Given that employing L19 gave excellent yields even with-
out the exclusion of air, the reaction scope was extended to a
wider range of bromo substrates (Scheme 3). All reactions
were carried out under both N2 and air (yields in parentheses
in Scheme 3 refer to reactions without any inert atmosphere
protection). It is noteworthy that the reaction could be ex-
tended beyond the preparation of aryl iodides, with hetero-
aryl iodide 2-iodo pyridine (2b) and vinyl iodide
Fig. 2 Screening of amine ligands. Reaction conditions: 1a (1 mmol)
was reacted with NaI (2 mmol) in dry dioxane (110 °C) or acetonitrile
(82 °C) in a sealed vial under N2 for 24 hours using CuI (5 mol%) and
L1, L9–L24 (10 mol%) as catalyst. Monoamine ligands were used with
20 mol% loading. All yields were determined by GC and averaged on
two runs.
β-iodostyrene (2c) both successfully synthesised from their
bromo precursors. In terms of ortho-substituents,
This journal is © The Royal Society of Chemistry 2017
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
Table 1 Optimisation of reaction conditions with L19a
Entry
Catalyst
L19
Temp.
Yield (%)
1
2
5% CuI
5% CuI
2.5% CuI
5% CuI
5% CuI
5% CuI
10%
10%
5%
10%
10%
10%
82 °C
83
99
98
99
99
96
110 °C
110 °C
110 °C
110 °C
110 °C
3
4b
5c
6d
a
Reaction conditions: 1a (1 mmol) was reacted with NaI (2 mmol) in dry acetonitrile at 82 or 110 °C in a sealed vial under N2 for 24 hours
b
using CuI (2.5–5 mol%) and ligand (5–10 mol%) as catalyst. All yields were determined by GC and averaged on two runs. 1.2 eq. NaI were
used. The reaction mixture was stirred in air for 10 min before heating to reaction temperature. 11.6 μL distilled water (1% of total volume)
c
d
was added and the reaction mixture stirred in air for 10 min before heating to reaction temperature.
2-bromoaniline (1h) and 2-bromoanisole (1j) both gave good
yields of the aryl-iodide products. Further comparison be-
tween the ortho, meta and para substituted isomers, indi-
cated that the functional group positions did not significantly
affect the final yields (2f, 2g, 2j). However the protocol did
not perform quite so well with substrates containing
electron-withdrawing groups (1k, 1l, 1m and 1p), with the ex-
ception of the nitrile group (1i) which gave over 90% yield of
2i. In contrast, substrates containing electron-donating
groups gave much higher yields (88% to >99%). Observed
yields under N2 and air are broadly comparable (≤5% differ-
ence) indicating that the exclusion of air is not required for
this protocol. However, an exception to this is the reaction
using substrates with moderate electron-withdrawing groups
(iodine 2k and ester 2l) which gave much lower conversions
and yields under air. For these lower yielding reactions,
higher catalyst and ligand loading (10 mol% CuI and 20
mol% L19) or longer reaction duration (48 h) all failed to give
any improvement. Aryl-chloride activations were also
attempted by reacting chlorobenzene with NaI under these
Scheme
conditions:
3
Reaction scope with different substrates. aReaction
(1 mmol) was reacted with NaI (2 mmol) in dry
1
Fig. 3 Air and moisture tolerance of the halogen exchange reaction
acetonitrile at 110 °C for 24 hours using CuI (5 mol%) and L19 (10
mol%) as catalyst. All yields were determined by GC and averaged on
two runs (yields in parentheses are the reactions exposed to air).
b1-Bromo-4-iodobenzene was used as substrate.
shown in Table
1 with L1 and L14 (using Buchwald's published
conditions in dioxane at 110 °C),16 and L19 (using the newly optimised
conditions of acetonitrile at 110 °C).
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2017

View Article Online
Catalysis Science & Technology
Paper
optimised conditions, but no product was observed. However,
by increasing the reaction temperature to 140 °C (in penta-
nol) low conversions and yields (both 8%) of 4-chlorotoluene
were achievable (see ESI†).
In summary, L19 is shown to promote the copper
catalysed halide exchange reaction in acetonitrile with good
reaction scope, especially for substrates without electron
withdrawing groups. Yields reported are comparable to ear-
lier studies by Buchwald using L1 and L14, with the benefit
of a significantly cheaper ligand system and no requirement
for inert-atmosphere protection.16,20,31
Fig. 4 Molecular structure of Cu2I2L12.
L1 (both at 5% copper loading) revealed almost identical re-
action profiles with 87% yield at 24 h (Fig. 6). However, crys-
tals of the polymeric network [Cu4I4L12]∞ (again at 5% load-
ing with respect to copper) showed lower reactivity and only
73% yield at 24 h. The lower reaction rate may relate to the
loading of L1 which at 2.5 mol% is half that of the other two
experiments. Further studies on the dependence of the rate
on [L1] supported this view, with a positive order in [L1] ob-
served (Fig. 6). From these experiments it can be seen that
the application of the crystalline pre-assembled complex as a
catalyst offers no significant rate benefits over using the
equivalent stoichiometric mixture of copperIJI) iodide and
ligand.
Catalyst and mechanism studies on the L1 and L19
promoted copper-mediated aromatic Finkelstein reaction
In order to better understand the differences between L1 and
L19 in the catalytic reaction, the solid-state structures of the
copper complexes of these ligands were investigated and re-
action kinetic profiling carried out.
Treatment of CuI with L1 in dioxane/hexane gave crystals
of the dimeric complex, Cu2I2L12 which was characterised by
single crystal X-ray diffraction studies‡ (Fig. 4). The dimeric
motif is similar to that observed in other published diamine
complexed copper halides,32–35 comprising of a central Cu2I2
ring. Cu–I distances are in the range 2.5797 (11) Å to 2.6358
(12) Å and the Cu⋯Cu distance is 2.5710 (19) Å. Each copper
is chelated by a diamine ligand with Cu–N bond distances of
2.138 (6) Å (Cu1–N1) and 2.155 (7) Å (Cu1–N2).
When a similar solution of CuI and L1 in acetone was
treated with 2 equivalents NaI (a reagent in the Finkelstein
reaction) it was possible to obtain crystals of a different struc-
ture, namely [Cu4I4L12]∞. X-ray diffraction studies‡ revealed
an unusual polymeric network built around rare36 tetrameric
[Cu4I4] clusters (Fig. 5). Each copper within the tetrameric
unit coordinates monodentately to a different L1 molecule
with a Cu–N bond distance of 2.064 (4) Å. This gives rise to a
2D network structure (Fig. 5, bottom). Cu–I distances are in
the range 2.6701 (7) to 2.7477 (7) Å. The Cu⋯Cu distances, at
2.627 (10) and 2.6734 (12) Å, are slightly longer than those
observed in Cu2I2L12.
Analogous to the studies with L1 detailed above, attempts
were also made to prepare crystals of L19 with copperIJI) io-
dide. However, on addition of L19 to CuI the solution was
discerned to rapidly turn from colourless to blue, and after
stirring for one hour the formation of a copper mirror was
The 1 : 2 L1 : Cu ratio in [Cu4I4L12]∞ is lower than that in
the reaction solution (1 : 1) and also lower than that in
crystalline Cu2I2L12 (1 : 1). The unaccounted for L1 likely re-
mains in solution coordinated to the Na+ cations, as further
evidenced by NMR spectroscopic experiments (see ESI†). Note
that it was not possible to directly grow crystals of [Cu4I4L12]∞
from a 2 : 1 mixture of CuI and L1 despite several attempts,
mainly due to the poor solubility of CuI when sub-
stoichiometric ligand loading was used.
Monitoring of the reaction kinetics for the Finkelstein re-
action with crystals of Cu2I2L12 and a 1 : 1 mixture of CuI and
‡ Full details concerning the data sets and crystal resolutions can be found in
the respective CIF files deposited at the Cambridge Crystallographic Data Centre
under the allocated deposition numbers CCDC 1538590–1538592.
Fig. 5 Crystal structure of [Cu4I4L12]∞ showing two Cu4I4 units (top);
view of the extended structure of [Cu4I4L12]∞ (bottom).
This journal is © The Royal Society of Chemistry 2017
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
Fig. 6 Comparison of the reaction profiles of CuI/L1 and crystals of
[Cu2I2L12] and [Cu4I4L12]∞. Reaction conditions: 1a (3 mmol) was
reacted with NaI (6 mmol) in dry acetonitrile (3 mL) at 82 °C.
Fig. 8 Comparison of the reaction profiles using CuI/L19 and crystals
of [CuL192]I2. Reaction conditions: 1a (3 mmol) was reacted with NaI (6
mmol) in dry acetonitrile (3 mL) at 110 °C.
also observed. This behaviour was repeated even when using
pre-dried reagents with the vigorous exclusion of air and
moisture. In contrast, an analogous solution with L1
remained colourless and no solid deposits were observed
even after prolonged stirring under N2. From the blue solu-
tion of L19 with CuI crystals of the CuIJII) complex [CuL192]I2
were obtained. The presence of CuIJII) along with the forma-
tion of a copper mirror is strongly suggestive of dispropor-
tionation of the CuIJI). X-ray crystallographic characterisation‡
of [CuL192]I2 (Fig. 7) reveals the copperIJII) centre to be coordi-
nated to six amine groups of two L19 ligands with Cu–N
bond distances in the range 2.009 (5) Å to 2.454 (5) Å. These
distances are longer than the comparable Cu–N bond dis-
tances in the previously reported mono-ligated [CuIJL19)I2]
complex (1.986 Å to 2.034 Å).37
that improved conversions were obtained with a copperIJII)
catalyst/pre-catalyst is surprising given that copperIJI) is
thought to be central to the catalytic cycle, undergoing oxida-
tive addition with the aryl halide reagent (Scheme 1).
It is not clear if copperIJII) is reduced to copperIJI) (poten-
tially promoted by L1938,39) which then undergoes the cata-
lytic cycle shown in Scheme 1, or if an alternative mechanism
based on copperIJII) is operational. In order to explore the pos-
sibility of a single electron transfer mechanism occurring,
radical trap experiments using 2-(allyloxy) bromobenzene as
the radical probe were carried out. If the reaction goes
through radical activation, the cyclised product would be
expected to form more rapidly than the halogen exchange
product.40,41 However, from the 1H-NMR studies only the
iodo-product was present with no cyclised products observed
The reaction profile of the halogen exchange reaction
using crystals of [CuL192]I2 as catalyst was obtained (Fig. 8).
Excellent yields were achieved, with conversion rates higher
than those achieved with a mixture of CuI and L19. The fact
Scheme 4 Reaction with 2-(allyloxy) bromobenzene (1q) as a radical
probe. Reaction conditions: 1q (1 mmol) was reacted with NaI (2
mmol) in dry acetonitrile at 110 °C.
Fig. 7 Molecular structure of [CuL192]I2.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2017

View Article Online
Catalysis Science & Technology
Paper
Table 2 Catalytic activity of copperIJII) salts with L19a
Entry
Copper salt
Yield (%)
1
2
3
4
5
6
CuI
CuIJacetate)2
CuCl2
CuBr2
CuIJCF3SO3)2
CuIJNO3)2
99
76
70
62
58
43
a
Reaction conditions: 1a (1 mmol) was reacted with NaI (2 mmol) in dry acetonitrile at 110 °C under N2 using crystals or CuI as catalyst for
24 h.
(Scheme 4). This suggests that the reaction pathway is not
containing electron-withdrawing groups were however less
tolerant of air/moisture than the other substrates.
following a radical mechanism (or alternatively the iodin-
ation step is extremely fast). It is worth noting that several ex-
amples of related catalytic copperIJII)-promoted cross cou-
plings have also been reported in the literature, but the
mechanisms of these reactions also remains unclear.42–46
The successful application of the copperIJII) complex
[CuL192]I2 in the Finkelstein reaction led us to explore the
catalytic performance of other CuIJII) salts in the presence of
L19. These studies (Table 2) showed that despite lower yields
when compared with copperIJI) iodide (entry 1), all the
copperIJII) salts were active giving moderate to good yield of
the iodo product (entries 2–6). Moreover, each of these
copperIJII) salts failed to catalyse the halide exchange reaction
in the absence of L19 or when using L1 in place of L19. Thus
repetition of entries 2–6 under either 10 mol% L1 loading or
under ligand free conditions gave yields and conversions of
less than 3%.
In order to probe the differences in catalyst speciation
when L1 and L19 were employed as ligands, we studied the
coordination chemistry of both of these compounds with
copperIJI) iodide. Solid-state structures of [Cu2I2L12],
[Cu4I4L12]∞ and [CuL192]I2 are reported. In the presence of
L19, copperIJI) underwent rapid disproportionation to give the
mononuclear copperIJII) complex [CuL192]I2. In contrast, no
evidence of disproportion was observed with L1, and the
copperIJI) complexes [Cu2I2L12] or [Cu4I4L12]∞ were obtained.
Studies on the catalytic activity of these crystalline complexes
reveal [CuL192]I2 to be a good catalyst for the halide exchange
reaction despite containing copperIJII) rather than copperIJI).
Further studies using copperIJII) salts in conjunction with L19
also showed remarkably good catalytic activity, whereas no
catalytic activity was observed using the same copperIJII) salts
with L1. Based on these results we speculate that copper un-
dergoes fast redox chemistry much more readily in the pres-
ence of L19 than with L1, and it is this that is responsible for
its catalytic activity with copperIJII) as well as the improved air
and moisture tolerance. However, the mechanism of these re-
actions is still not clear and further studies to better under-
stand this are ongoing in our group.
Conclusions
We have screened a number of ligands for application in the
copper-catalysed aromatic Finkelstein reaction. Ligands
which have recently been reported to show superior activity
in the related copper-catalysed Ullmann amination reaction,
such as 1,10-phenanthroline, 2-acetylcyclohexanone and di-
methyl glycine, were all shown to be non-active in the
Finkelstein reaction. This is despite the supposed similarity
in reaction mechanism for these two processes. Further li-
gand screening, this time of amine based ligands, identified
diethylenetriamine (L19) to be a promising candidate. Good
yields were obtained using L19 under mild conditions, di-
rectly comparable to those previously reported for L1 and
L14. However, unlike L1 and L14, the use of L19 was shown
not to require the employment of air and moisture free con-
ditions. In addition, L19 is at least 100× less expensive
per mole than either L1 or L14. The scope of the reaction was
extended to 16 examples (both under N2 and in air) with
good conversions reported in all cases. Aryl-bromides
In summary, we have reported a new cheap and effective
ligand – diethylenetriamine (L19) – for both copperIJI) and
copperIJII) mediated aromatic Finkelstein reactions, which
shows not only satisfactory yields over a wide range of sub-
strates but also good tolerance to air and moisture.
Acknowledgements
We thank Andrew J. P. White for assistance with obtaining
and solving the X-ray data.
Notes and references
1 D. A. Petrone, J. Ye and M. Lautens, Chem. Rev., 2016, 116,
8003–8104.
This journal is © The Royal Society of Chemistry 2017
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
2 G. C. Fu, Acc. Chem. Res., 2008, 41, 1555–1564.
3 D. Ma and Q. Cai, Acc. Chem. Res., 2008, 41, 1450–1460.
4 T. D. Sheppard, Org. Biomol. Chem., 2009, 7, 1043.
5 H. Finkelstein, Ber. Dtsch. Chem. Ges., 1910, 43, 1528–1532.
6 R. G. R. Bacon and H. A. O. Hill, J. Chem. Soc.,
1964, 1097–1107.
7 R. G. R. Bacon and H. A. O. Hill, J. Chem. Soc.,
1964, 1108–1112.
8 R. Cramer and D. R. Coulson, J. Org. Chem., 1975, 40,
2267–2273.
9 K. Takagi, N. Hayama and T. Okamoto, Chem. Lett., 1978, 7,
191–192.
10 K. Takagi, N. Hayama and S. Inokawa, Bull. Chem. Soc. Jpn.,
1980, 53, 3691–3695.
11 T. Tsou and J. Kochi, J. Org. Chem., 1980, 45, 1930–1937.
12 G. Axelrad, S. Laosooksathit and R. Engel, J. Org. Chem.,
1981, 46, 5200–5204.
25 I. P. Beletskaya and A. V. Cheprakov, Coord. Chem. Rev.,
2004, 248, 2337–2364.
26 D. S. Surry and S. L. Buchwald, Chem. Sci., 2010, 1, 13–31.
27 F. Monnier and M. Taillefer, Angew. Chem., Int. Ed.,
2009, 48, 6954–6971.
28 K. A. Cannon, M. E. Geuther, C. K. Kelly, S. Lin and A. H. R.
Macarthur, Organometallics, 2011, 30, 4067–4073.
29 V. Manríquez, M. Campos-Vallette, N. Lara, N. González-
Tejeda, O. Wittke, G. Díaz, S. Diez, R. Muñoz and L.
Kriskovic, J. Chem. Crystallogr., 1996, 26, 15–22.
30 L1, £528, 100 g; L14, £1665, 100 mL; L19, £50, 1 L. Price
obtained from United Kingdom|Sigma-Aldrich, 04 DEC
2016.
31 L. Li, W. Liu, H. Zeng, X. Mu, G. Cosa, Z. Mi and C. J. Li,
J. Am. Chem. Soc., 2015, 137, 8328–8331.
32 E. R. Strieter, B. Bhayana and S. L. Buchwald, J. Am. Chem.
Soc., 2009, 131, 78–88.
13 H. Suzuki, A. Kondo, M. Inouye and T. Ogawa, Synthesis,
1986, 1986, 121–122.
14 S. H. Yang, C. S. Li and C. H. Cheng, J. Org. Chem., 1987, 52,
691–694.
15 A. Casitas and X. Ribas, Copper-Mediated Cross-Coupling
Reactions, John Wiley & Sons, Inc., Hoboken, NJ, USA, 2013,
pp. 239–251.
33 D. A. Haitko, J. Coord. Chem., 1984, 13, 119–122.
34 M. F. Garbauskas, D. A. Haitko and J. S. Kasper,
J. Crystallogr. Spectrosc. Res., 1986, 16, 729–738.
35 L. M. Engelhardt, R. I. Papasergio and A. H. White, Aust. J.
Chem., 1984, 37, 2207–2213.
36 S. Naik, J. T. Mague and M. S. Balakrishna, Inorg. Chem.,
2014, 53, 3864–3873.
16 A. Klapars and S. L. Buchwald, J. Am. Chem. Soc., 2002, 124,
14844–14845.
37 D. J. Hodgson, D. K. Towle and W. E. Hatfield, Inorg. Chim.
Acta, 1991, 179, 275–280.
17 K. A. Cannon, M. E. Geuther, C. K. Kelly, S. Lin and A. H. R.
MacArthur, Organometallics, 2011, 30, 4067–4073.
18 M. Chen, S. Ichikawa and S. L. Buchwald, Angew. Chem., Int.
Ed., 2015, 54, 263–266.
19 A. Casitas, M. Canta, M. Solà, M. Costas and X. Ribas, J. Am.
Chem. Soc., 2011, 133, 19386–19392.
38 G. Zhang, H. Yi, G. Zhang, Y. Deng, R. Bai, H. Zhang, J. T.
Miller, A. J. Kropf, E. E. Bunel and A. Lei, J. Am. Chem. Soc.,
2014, 136, 924–926.
39 H. Yi, Z. Liao, G. Zhang, G. Zhang, C. Fan, X. Zhang, E. E.
Bunel, C.-W. Pao, J.-F. Lee and A. Lei, Chem. – Eur. J.,
2015, 21, 18925–18929.
20 X. Feng, L. Li, X. Yu, Y. Yamamoto and M. Bao, Catal.
Today, 2016, 274, 129–132.
40 A. Annunziata, C. Galli, M. Marinelli and T. Pau, Eur. J. Org.
Chem., 2001, 2001, 1323–1329.
21 A. Casitas, X. Ribas, A. Casitas, M. Costas, X. Ribas, S. S.
Stahl, T. Parella, J. Benet-Buchholz, A. Schweiger, G.
Mitrikas, M. Solà, A. Llobet and T. D. P. Stack, Chem. Sci.,
2013, 4, 2301.
22 S. Sung, D. Sale, D. C. Braddock, A. Armstrong, C. Brennan
and R. P. Davies, ACS Catal., 2016, 6, 3965–3974.
23 S. Sung, D. C. Braddock, A. Armstrong, C. Brennan, D. Sale,
A. J. P. White and R. P. Davies, Chem. – Eur. J., 2015, 21,
7179–7192.
41 S. E. Creutz, K. J. Lotito, G. C. Fu and J. C. Peters, Science,
2012, 338, 647–651.
42 G. Franc and A. Jutand, Dalton Trans., 2010, 39, 7873–7875.
43 A. Ouali, J. F. Spindler, H. J. Cristau and M. Taillefer, Adv.
Synth. Catal., 2006, 348, 499–505.
44 A. Ouali, J.-F. F. Spindler, A. Jutand and M. Taillefer, Adv.
Synth. Catal., 2007, 349, 1906–1916.
45 H. J. Cristau, P. P. Cellier, S. Hamada, J. F. Spindler and M.
Taillefer, Org. Lett., 2004, 6, 913–916.
24 E. Sperotto, G. P. M. van Klink, G. van Koten and J. G. de
Vries, Dalton Trans., 2010, 39, 10338–10351.
46 P. F. Larsson, A. Correa, M. Carril, P. O. Norrby and C.
Bolm, Angew. Chem., Int. Ed., 2009, 48, 5691–5693.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2017