Iron-Catalyzed Isopropylation of Electron-Deficient Aryl and Heteroaryl Chlorides

Article abstract of DOI:10.1002/adsc.201601097

Traditional methods for the preparation of secondary alkyl-substituted aryl and heteroaryl chlorides challenge both selectivity and functional group tolerance. This contribution describes the use of statistical design of experiments to develop an effective procedure for the preparation of isopropyl-substituted (hetero)arenes with minimal isopropyl to n-propyl isomerization. The reaction tolerates electronically diverse aryl chloride coupling partners, with excellent conversion observed for strongly electron-deficient aromatic rings, such as esters and amides. Electron-rich systems, including methyl- and methoxy-substituted aryl chlorides, were found to be less reactive. Furthermore, the reaction was found to be most successful when heteroaryl chlorides were submitted to the cross-coupling protocol. By mapping substituent effects on reaction selectivity, we were able to show that electron-deficient aryl chlorides are essential for efficient coupling, and use electronic structure calculations to predict the likelihood of successful coupling through the estimation of the electron affinity of each aryl chloride. Moderate isolated yields were achieved with selected aryl chlorides, and moderate to good isolated yields were obtained for all the heteroaryl chlorides coupled. Excellent selectivity was observed when a 2,6-dichloroquinoline was used, allowing mono-substitution on a challenging substrate. (Figure presented.).

Full text of DOI:10.1002/adsc.201601097

FULL PAPERS
DOI: 10.1002/adsc.201601097
Iron-Catalyzed Isopropylation of Electron-Deficient Aryl and
Heteroaryl Chlorides
James N. Sanderson,a,b,* Andrew P. Dominey,a and Jonathan M. Percyb,*
a
GlaxoSmithKline, Medicines Research Centre, Gunnels Wood Road, Stevenage, SG1 2NY, U.K.
Pure and Applied Chemistry, University of Strathclyde, Thomas Graham Building, 295 Cathedral Street, Glasgow, G1
b
1XL, U.K.
E-mail: james.sanderson@strath.ac.uk or jonathan.percy@strath.ac.uk
Received: October 6, 2016; Revised: December 13, 2016; Published online: && &&, 0000
Supporting information for this article can be found under: http://dx.doi.org/10.1002/adsc.201601097.
Abstract: Traditional methods for the preparation of ping substituent effects on reaction selectivity, we
secondary alkyl-substituted aryl and heteroaryl chlor- were able to show that electron-deficient aryl chlor-
ides challenge both selectivity and functional group ides are essential for efficient coupling, and use elec-
tolerance. This contribution describes the use of stat- tronic structure calculations to predict the likelihood
istical design of experiments to develop an effective of successful coupling through the estimation of the
procedure for the preparation of isopropyl-substitut- electron affinity of each aryl chloride. Moderate iso-
ed (hetero)arenes with minimal isopropyl to n- lated yields were achieved with selected aryl chlor-
propyl isomerization. The reaction tolerates electron- ides, and moderate to good isolated yields were ob-
ically diverse aryl chloride coupling partners, with tained for all the heteroaryl chlorides coupled. Ex-
excellent conversion observed for strongly electron- cellent selectivity was observed when a 2,6-dichloro-
deficient aromatic rings, such as esters and amides. quinoline was used, allowing mono-substitution on
Electron-rich systems, including methyl- and me- a challenging substrate.
thoxy-substituted aryl chlorides, were found to be
less reactive. Furthermore, the reaction was found to Keywords: density functional theory (DFT); design
be most successful when heteroaryl chlorides were of experiments; iron catalysis; isopropyl group;
submitted to the cross-coupling protocol. By map- Kumada cross-coupling
Introduction
to problems of both selectivity and functional group
tolerance. To avoid these difficulties, it is often intro-
The isopropyl group is an important structural motif duced early in a synthetic process, as demonstrated in
in both active pharmaceutical ingredients (APIs) such the preparation of 1 and analogues of 2.[1a,3] The addi-
as Rosuvastatin 1 and natural products including (À)- tion of an isopropyl group via late-stage functionaliza-
Standishnal 2 (Figure 1).[1]
tion is less practical; introducing an isopropyl group
The introduction of this group can require harsh to a scaffold in a medicinal chemistry programme, for
and forcing conditions including those employed in example, to explore the effects of size and hydropho-
the Friedel–Crafts alkylation (Scheme 1);[2] this leads bicity within a binding pocket, can therefore be chal-
lenging.
Over the last 40 years, the use of transition metal
(TM)-catalyzed cross-coupling reactions as accurate,
mild and reliable reaction chemistries has burgeoned.
The rapid expansion of this area of research has pro-
vided the synthetic chemist with a palette of new
techniques to create bonds with superb levels of con-
trol and selectivity. The importance of palladium-cata-
lyzed processes triggered the award of the Nobel
Prize for Chemistry to Heck, Negishi and Suzuki in
2010.[4] Despite the ubiquity and flexibility of palladi-
um-catalyzed C–C bond forming reactions, certain
Figure 1. Active pharmaceutical ingredient Rosuvastatin and
natural product (À)-Standishnal both contain the isopropyl
moiety.
Adv. Synth. Catal. 0000, 000, 0 – 0
1
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÞÞ
These are not the final page numbers!

FULL PAPERS
asc.wiley-vch.de
Scheme 2. Fꢁrstnerꢂs example of the isopropylation of an ac-
tivated pyridine substrate.
ride
3 was demonstrated by Fꢁrstner in 2002
(Scheme 2).[18]
Since Fꢁrstnerꢂs publication, very few examples of
the iron-catalyzed coupling of secondary alkyl
Grignard reagents have been reported. Cook et al.
demonstrated the isopropylation of aryl sulfamates
and sulfonamides in 2012, although significant isopro-
pyl to n-propyl isomerization was observed [i),
Scheme 3].[19] The use of an iron fluoride pre-catalyst
was demonstrated to give a higher isopropyl to n-
propyl ratio than iron chloride. Perry et al. also inves-
tigated the use of isopropylmagnesium chloride as
part of a wider study into the application of NHC li-
Scheme 1. Some methods for the introduction of an isopro-
pyl group. A: Friedel–Crafts alkylation.[10] B: Alkyl organo-
metallic in a cross-coupling process.[5b,6] C: Catalytic petro-
chemical preparation.[11] D: Alkenyl organometallic in
a cross-coupling process followed by hydrogenation.[12]
.
substrates remain challenging. The installation of sec-
ondary alkyl groups via cross-coupling methodology
often leads to undesired by-products resulting from b-
hydride elimination pathways.[5] Significant efforts
have been made in recent years to address these
problems through the application of palladium cataly-
sis with sterically hindered phosphine and N-heterocy-
clic carbene (NHC) ligands;[5b,6] most of the progress
in this area has been made with Suzuki–Miyaura
transformations. Secondary alkyl organoboron species
are competent cross-coupling partners in certain ap-
plications, but they are generally prepared in a waste-
ful and step-inefficient manner, often via transmetal-
lation from a Grignard or organolithium reagent.[7]
A
Kumada (RMgX)[8] or Negishi (RZnX)[9] type cross-
coupling reaction of secondary organometallics, di-
rectly using the more reactive reagent, would there-
fore offer significant advantages in relation to effi-
ciency and atom economy (Scheme 1).
The use of precious metal catalysts is often chal-
lenged in a pharmaceutical process research and de-
velopment environment, due not only to cost of
goods and questionable sustainability of supply, but
also because of the toxicity arising from the presence
of trace residues in APIs.[13] In recent years, there has
been a significant resurgence of interest in the appli-
cation of earth-abundant metals in cross-coupling re-
actions;[14] iron has been pre-eminent due to its abun-
dance, low toxicity, low cost and high versatility.[15]
Pioneering research by Vavon and Mottez in 1944[16]
was advanced by Kochi in his seminal publication in
1971.[17] Early reports focused on the reaction of pri-
mary alkyl Grignard reagents with aryl or vinyl hal-
ides; a single example of the iron-catalyzed coupling
Scheme 3. Iron-NHC Kumada-type isopropylation of carbo-
cyclic electrophiles as demonstrated by i) Cook,[19] ii)
of isopropylmagnesium chloride with heteroaryl chlo- Perry,[20] and iii) Nakamura.[21]
Adv. Synth. Catal. 0000, 000, 0 – 0
2
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÞÞ
These are not the final page numbers!

FULL PAPERS
asc.wiley-vch.de
gands in iron-catalyzed reactions.[20] It was noted that Through the strategic use of statistical design of ex-
a higher isopropyl to n-propyl ratio could be achieved periments (DoE), we hoped to develop an effective,
through the use of a large excess (6 equivalents) of more general and straightforward preparation of iso-
Grignard reagent, albeit at the expense of yield [ii), propylarenes and -heteroarenes from the correspond-
Scheme 3]. More recently, Nakamura et al. applied an ing chlorides via an iron-catalyzed reaction.
iron/NHC system to couple electron-rich aryl chlor-
ides with both primary and secondary alkyl Grignard
reagents.[21] However, attempted couplings with iso- Results and Discussion
propylmagnesium chloride led to the extensive forma-
tion of isomerized by-products [iii), Scheme 3].[19] The In 2013 Fox et al. demonstrated an operationally
use of an iron fluoride pre-catalyst was shown to be simple reaction between primary alkyl Grignard re-
necessary for a productive reaction in this case. In agents, such as n-butylmagnesium chloride, and elec-
general, the iron/NHC catalyst systems gave high tron-deficient aryl chlorides.[25] However, the reaction
levels of isopropyl to n-propyl isomerization, probably was less successful when more bulky organometallics
caused by the elevated temperatures required for ef- were used; for example, only a 48% conversion (25%
fective cross-coupling.[19–21]
isolated yield) of 1-chloro-4-(trifluoromethyl)benzene
Most recently, Wang et al. applied a simple FeCl3/ 13 was reported from the reaction with phenethyl-
N-methyl-2-pyrrolidinone (NMP) system to couple magnesium chloride (16, Scheme 6). No examples of
a small selection of pyrimidin-2-yl tosylates with al-
kylmagnesium bromides (Scheme 4).[22] The good
yields reported for this process suggested that the sec-
ondary:primary selectivity was high.
Scheme 6. Foxꢂs alkylation of electron-deficient aryl chlor-
ides with isolated yields.
cross-coupling processes of secondary alkyl Grignard
reagents were reported. Our preliminary study also
demonstrated
that
tetramethylethylenediamine
Scheme 4. Wangꢂs isopropylation of heteroaryl tosylates.[22]
.
(TMEDA) was an ineffective additive for the desired
transformation (see the Supporting Information). Al-
though NMP raises concerns due to its reprotoxic ef-
In most cases, the iron-catalyzed cross-coupling of fects, it has proven to be a successful reaction compo-
secondary alkyl Grignard reagents used less reactive nent in numerous iron-catalyzed processes.[15b,26] NMP
chloride coupling partners, instead of the more expen- was therefore chosen as the additive in a primary in-
sive and less readily available bromides or iodides. vestigation into the Kumada-type cross-coupling of
Although aryl nucleophile/alkyl halide coupling is isopropylmagnesium chloride and electron-deficient
well-known (Scheme 5),[23] the use of alkyl organome- aryl chlorides, with a view to applying the methodolo-
tallics also circumvents the handling of volatile alkyl gy to other secondary alkyl Grignard reagents in the
chlorides which are potentially toxic alkylating agents. future.
Based on the precedent of Fox et al., we chose the
isopropylation of 1-chloro-4-(trifluoromethyl)benzene
13a as a model reaction for a DoE investigation
(Scheme 7). DoE can offer many advantages over the
classical one-factor-at-a-time (OFAT) approach.[27]
Due to its multivariate nature it is possible to opti-
mize several components at once, leading to a signifi-
cant reduction in the overall amount of experimenta-
tion. The statistical analysis used in an experimental
design also means that it is possible to reveal interac-
tions between components, a distinct benefit over
OFAT studies.
Scheme 5. Iron-catalyzed cross-coupling of aryl Grignard re-
agent with an alkyl halide.[24]
.
Adv. Synth. Catal. 0000, 000, 0 – 0
3
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÞÞ
These are not the final page numbers!

FULL PAPERS
asc.wiley-vch.de
sented by a linear model. A further 10 experiments
were therefore carried out to expand the previous in-
vestigation and allow greater definition of non-linear
effects. Following the addition of this data, statistical
analysis suggested that a quadratic response surface
would offer the best statistical representation (see the
Supporting Information). The most significant factor
was an interaction observed between the loadings of
catalyst and NMP, as suggested by the first investiga-
tion. The improved model was used to generate a 3-
dimensional plot of this interaction (Figure 2); the op-
timal region for product formation is highlighted in
red.
Scheme 7. Model reaction chosen for the DoE investigation.
Stat-Easeꢂs Design-Expert 7 (DX7) was used to
prepare a small DoE with 4 factors (Table 1), requir-
ing an initial set of only 12 experiments. Reaction
mixtures were analyzed using GC, with naphthalene
as an internal standard. The results were then import-
ed directly into DX7 and the response to changes in
the chosen factors was analyzed. Duplicate reactions
were performed throughout the investigation to dem-
onstrate good reproducibility and give confidence in
the subsequent statistical analysis (see the Supporting
Information).
Table 1. Chosen factors and responses for the preliminary
experimental design.
Factor
Range
A
B
C
D
temperature
Grignard addition time
ligand loading
À20 to 408C
0 to 30 minutes
10–300ꢃcatalystmol%
0.1 to 10 mol%
Figure 2. Three-dimensional response surface for catalyst
and ligand loading.
catalyst loading
We then tried to use our DoE results to study the
The results from this first series of experiments formation of isomer 13c, hydrodehalogenated[28] com-
showed that efficient conversion was achieved with pound 13d and biaryl 13e (Scheme 6). Unfortunately,
the higher catalyst loadings. A number of two-factor we were unable to generate a statistically significant
interactions were observed: the most significant of model for the formation of the undesired by-products,
these was between catalyst loading and NMP. By in- which may be related to the very low levels observed
creasing the level of NMP a higher level of the de- throughout the DoE (see the Supporting Informa-
sired product was observed; this effect was more pro- tion). However, it was clear that higher catalyst load-
nounced when a catalyst loading of 10 mol% was ing gave more of both the isomerized product, and
used (see the Supporting Information). A similar (but the biaryl 13e formed from reductive homocoupling
inverse) relationship was observed with temperature; of two aryl chloride species.[29] Any optimal conditions
the lower temperatures used gave the best reaction must therefore balance the productive and non-pro-
profiles. As with NMP, the response to temperature ductive pathways. Upon completion of the experimen-
was increased when more catalyst was present (see tal design we used DX7 to predict a set of optimal
the Supporting Information).
conditions to achieve the highest level of isopropylat-
During data analysis it became apparent from reac- ed product (Table 2), together with a 95% confidence
tions carried out at the centre of the design-space that interval. Pleasingly, when implementing these condi-
the chosen experimental design (24–1) was not optimal. tions a conversion of 90% was observed; only 4% of
The experiments at the centre of the design space the n-propyl isomer was formed in this reaction, a sec-
[5.05 mol% Fe(acac)3, 783 mol% NMP] gave the high- ondary:primary ratio of almost 23:1. This is a higher
est levels of product; this behaviour cannot be repre- ratio than those reported under iron/NHC-catalyzed
Adv. Synth. Catal. 0000, 000, 0 – 0
4
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÞÞ
These are not the final page numbers!

FULL PAPERS
asc.wiley-vch.de
Table 2. Predicted optimal conditions (P) and experimental
results (E) with a 95% confidence interval of 91–96%.
7.5 mol%. Control experiments showed no reaction
with Cu(acac)2 or Pd(acac)2, and minimal reactivity
with Ni(acac)2; Fe(acac)2 displayed a similar product
distribution to Fe(acac)3, analogous to earlier reac-
tions with iron chlorides.[18] A series of experiments
were also conducted at this point to test both experi-
mental reproducibility and confirm the selection of
both temperature and addition time for the Grignard
reagent (see the Supporting Information).
Temp.
[8C]
Ligand
[ꢃcat.mol%]
Catalyst
[mol%]
Product [GC
%]
P À20
E À20
199
200
7.8
7.8
95
90
conditions in the literature.[19–21] The remaining mate-
The reaction gave the highest level of the desired
rial was found to be the reductive homocoupling coupling product 13b, 17b–38b (Table 3) when strong-
product 13e, at 6%. ly inductively electron-withdrawing substituents were
With the optimized reaction conditions in hand, present on the aryl chloride. The best substrates were
a series of aryl chlorides were tested in the iron-cata- the 3- and 4-substituted trifluoromethylbenzenes 13a
lyzed cross-coupling with isopropylmagnesium chlo- and 17a (entries 1 and 2); conversion to the desired
ride (Table 3). The experiments throughout the DoE product was higher for the meta-species, consistent
used 97% purity Fe(acac)3; recent reports have shown with the inductively electron-withdrawing (ÀI) effect
marked catalytic activity related to trace impurities in of the trifluoromethyl group. The alkoxycarbonyl
iron salts.[30] It was possible to use 99.9% trace metals groups of 19a and 20a (entries 4 and 5) were tolerated
basis Fe(acac)3 and reduce the catalyst loading to with good conversion to product in both meta- and
Table 3. Aryl chlorides tested in the iron-catalyzed cross-coupling reaction. GC conversion (isolated yields in parentheses).
Entry
Substrate
X
Ar-Cl
(a)
Ar-i-Pr
(b)
Ar-n-Pr
(c)
Ar-H
(d)
Ar-Ar
(e)
Branched:Linear
(b:c)
1
2
3
4
5
6
7
8
13
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
4-CF3
3-CF3
ND
ND
ND
5
ND
<1
ND
ND
5
ND
ND
ND
23
5
2
19
2
17
83
54
54
90 (45)
97
84 (46)
80 (23)
88
76 (48)
71
64
70
69
59
69
54
62
60
57
53
40
13
7
4
ND
ND
16
15
10
24
12
14
11
1
32
31
22
22
28
18
35
37
4
6
3
23:1
97:1
NA
80:1
44:1
NA
71:1
32:2
10:1
69:1
59:1
NA
NA
31:1
60:1
57:1
27:1
NA
NA
N/A
NA
NA
22:1
<1
ND
<1
2
ND
1
2
7
1
1
ND
ND
2
1
1
4-COMOR[a]
4-CO2Me
3-CO2Me
4-pyridyl[b]
4-Cl
ND
ND
ND
ND
16[c]
20[c]
7
30
9
ND
2
3-Cl
4-SO2NEt2
4-F
3-F
4-CN
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
3-CN
4-CHF2
4-OCHF2
4-SMe
4-OCF3
3-OCF3
4-CH2F
4-Me
3-Me
4-OMe
3-OMe
10
9
7i[d]
9
2
ND
ND
ND
ND
ND
1
6
ND
2
2
ND
8
38
38
31
42
6
3
22
66
27
[a]
MOR=morpholine.
2-(4-Chlorophenyl)pyridine.
[b]
[c]
[d]
Terphenyl also observed (see the Supporting Information).
Biaryl also observed (see the Supporting Information). ND=not detected, NA=not applicable, Volumes=1 LkgÀ1 of
limiting reagent.
Adv. Synth. Catal. 0000, 000, 0 – 0
5
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÞÞ
These are not the final page numbers!

FULL PAPERS
asc.wiley-vch.de
para-cases. Moreover, no reaction was observed be-
tween the ester and isopropylmagnesium chloride. Di-
chlorobenzenes 22a and 23a (entries 7 and 8) gave
moderate conversion to the isopropylated product,
but were found to form high levels of both hydrode-
halogenated and homocoupled species. Interestingly,
both biphenyl and terphenyl were observed by GC-
MS, suggesting that oxidative addition was still com-
petitive after the first substitution (see the Supporting
Information). Aryl chlorides containing fluorine
atoms 25a and 26a (entries 10 and 11) and cyano
groups 27a and 28a (entries 12 and 13) were also reac-
tive, although they produced higher amounts of unde-
sired by-products. Difluoromethyl- and difluorome-
thoxybenzenes 29a and 30a (entries 14 and 15) also
gave moderate levels of isopropyl addition. Although
methyl-substituted starting materials 35a and 36a (en-
tries 35 and 36) failed to give the desired products,
possibly because they were too electron-rich, they
have been included for completeness (vide infra). The
meta-methoxy-substituted starting material 38a
(entry 23) gave a higher conversion than the para-an-
alogue 37a (entry 22) – suggesting that the reaction
was aided by a ÀI effect.
Although necessary to form the desired isopropy-
lated product, the presence of relatively large
amounts of NMP made the isolation of the simple
substituted cumenes 13b and 19b quite difficult.
Quenching reactions with either sulfuric or citric acid
enabled efficient removal of NMP from the organic
phase in most cases.
Figure 3. Heteroaryl chlorides prepared using the iron-cata-
lyzed Kumada cross-coupling. Isolated yields (conversion in
parentheses).
explain the reduction in isomerization for pyridyl sub-
A pre-packed reverse phase (C18) column was strates.[31]
used in an attempt to trap the organic product due to
Chloropyridines gave good to excellent conversion
its low polarity (Rf of 0.69 in neat hexane), allowing to the desired isopropyl analogues (4, 39 to 43,
the polar solvents to be washed away. The process Figure 3, Table 4). Purification of these substrates
was successful, removing all traces of both THF and using either normal phase or reverse phase column
NMP prior to further purification. Subsequent distilla- chromatography resulted in moderate to good isolat-
tion of the reaction mixture from 13b gave a 45% iso- ed yields. Methoxy-substituted pyridine 4 was isolated
lated yield of the desired compound. It was possible in 79% yield, a significant improvement over the pre-
to apply this procedure to 4-methoxycarbonyl product vious report.[18] Alkoxycarbonyl-substituted pyridines
19b, giving an isolated yield of 23% from a challenging 41 and 42 gave very good conversion; the tert-butyl
product mixture. These very small and volatile mole- ester 41 displayed lower conversion than the ethyl an-
cules represent worst case scenarios for product isola- alogue 42, possibly due to steric effects. Cyanopyri-
tion; similar retention times and boiling points led to dine 43 demonstrated reduced selectivity, much like
difficulty in separating hydrodehalogenated material the cyanobenzenes 27b and 28b (entries 10 and 11,
from the desired product with the aryl chlorides.
Table 3). Normal phase chromatography, used in the
To expand the substrate scope of the iron-catalyzed purification of 41 and 42 gave a reduced yield due to
Kumada cross-coupling further, we chose to investi- close running impurities; reverse phase chromatogra-
gate a number of heteroaryl chlorides. Excellent con- phy used in the purification of 39 and 40 also led to
version was observed in all cases, with some sub- challenges related to volatility and aqueous solubility.
strates demonstrating quantitative formation of the
We have already commented on an apparent rela-
isopropylated material (Figure 3). Interestingly, the tionship between the inductive effects of substituents
secondary:primary selectivity observed with many of and conversion; in an attempt to quantify this rela-
them was much higher than those for the benzenoid tionship, we tested a series of substituted 4-tolyl chlor-
analogues, even with reactants leading to lower levels ides (Table 5). Increasing the number of fluorine
of the desired product. An alternative mechanism has atoms, and therefore the ÀI effect, gave a higher con-
previously been proposed by Jutand et al., which may version to the isopropylated product. A modest trend
Adv. Synth. Catal. 0000, 000, 0 – 0
6
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÞÞ
These are not the final page numbers!

FULL PAPERS
asc.wiley-vch.de
Table 4. Product distributions for the reaction between 3-chloropyridine and a selection of other Grignard reagents. Conver-
sion by GC.[a]
Entry
Product
RMgCl
Ar-Cl
(a)
Ar-R
(b)
Ar-R’
(c)
Ar-H
(d)
Ar-Ar
(e)
Desired Product
1
2
3
39
47
48
isopropyl
methyl
ND
93
82
ND
NA
ND
18
ND
7
ND
67
ND
ND
sec-butyl
33
ND
4
5
49
50
cyclopentyl
isobutyl
ND
77
52
23
NA
ND
30.0
ND
18
ND
6
51
cyclohexyl
16
82
NA
ND
2
[a] R’=isomerized; ND=not detected; NA=not applicable.
Table 5. (Fluoro)alkyl and (fluoro)alkoxy substituted aryl
chlorides.
[32]
Entry Substrate
R
i-Pr (b) sp
log10(kX/kH)
1
2
3
4
5
6
7
11
27
32
33
30
28
35
CF3
89.8
61.5
12.6
6.7
0.54
0.32
0.11
0.62
0.46
À0.23
CF2H
CH2F
Me
OCF3
OCF2H 59.9
OMe 2.7
À0.17 À0.50
52.5
0.35
0.18
0.39
0.45
À0.27 À0.90
Figure 4. Hammett plot for 4-fluoroalkyl analogues.
A step like this could be accommodated within
(R2 =0.93, 4 points) was observed between log10(% a cycle carried by Fe(I),[29] Fe(0) or Fe(ÀII)[18] spe-
conversion) and sP; however, the results for the three cies.[33] The elucidation of a reaction mechanism lies
methoxy substituents did not fit this trend.
far outside the remit of this manuscript; instead, we
The Hammett plot (Figure 4) shows a small positive sought a triage method which could be used to predict
reaction constant (1=1.72) which suggests a very lim- the likely degree of success for more complex sub-
ited build up of negative charge in the transition state strates. Activation of the aryl chloride in the OA step
of the rate-determining step; this would be consistent presumably involves electron transfer to it, so the
with electron donation from the catalytic metal centre LUMO energy might be related to the rate of entry
to the arene in an oxidative addition (OA) reaction. to the cycle and therefore of consumption of starting
Adv. Synth. Catal. 0000, 000, 0 – 0
7
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÞÞ
These are not the final page numbers!

FULL PAPERS
asc.wiley-vch.de
material. We calculated the LUMO energies of 13a
and 17a–38a using the method of Dixon et al.
(B3LYP/6-31+G*);[34] FMOs were examined to
ensure that an unoccupied orbital associated with the
C–Cl bond was selected. Unfortunately, no significant
correlation was observed between conversion/yield
and the calculated energies of the chosen orbital (see
the Supporting Information).
We also evaluated the electron affinity of the aryl
chlorides, following work published by Gillmore
et al.[35] This involved the calculation of the energies
of the aryl chloride ground states and radical anions
(Scheme 8). The resulting value has been demonstrat-
Scheme 8. Representation of the computational method
used to determine aryl chloride electron affinity.
Figure 5. Plot of conversion to desired product versus calcu-
lated electron affinity. Black circles – data from Table 3.
Red crosses – new substrates to test predictivity.
ed to be associated with the reduction potential of the
molecule through an inverse linear correlation. Many
calculations failed due to dissociation of chloride chlorides were coupled with primary alkyl Grignard
anion from the aryl radical at the reported level of reagents;[25] substrates 13a, 18a and 19a (Entries 1, 3
theory (B3LYP/6-31G*). This pathway could be and 4 – Table 3) gave successful cross-coupling in
avoided in some cases by the use of a larger basis set both Foxꢂs primary alkylation and our isopropylation.
(B3LYP/6-31+ +G**) although C–Cl bond scission
To test these hypotheses we calculated DE for
small series of trisubstituted aryl chlorides
still occurred.[36] With a different functional and
a
slightly larger basis set (M06-2X/6-311+ +G**), it (Figure 6) along with a number of Foxꢂs substrates.
was possible to successfully estimate the electron af- The new aryl chlorides are shown as red crosses in
finity of the aryl chloride in 21 out of 23 cases.
Figure 5. At high values of DE we see a good fit
The observed conversion to isopropylated product within the limits of the earlier reactions, with the
was plotted against the calculated energy gap lower DE we observe much lower predictivity. The re-
(Figure 5). The data fell into three regions, defined by actants investigated from Foxꢂs primary alkylation
the electron affinity in eV. Below 1.01 eV we saw the
poorest conversion to desired product; these sub-
strates are the methyl- and methoxy-substituted ben-
zenes. A lower electron affinity seems to give rise to
a less productive reaction. Between 1.17 and 1.28 eV,
we see a small grouping where 40–60% conversion is
observed in most cases. Above 1.49 eV, greater than
60% conversion is observed. This suggests that
a higher electron affinity leads to a more productive
reaction. These data could be fitted by an exponential
3 parameter curve (see the Supporting Information),
with an R2 value of 0.78.
The data obtained from the DFT investigation sug-
gest that it may be possible to determine whether or
not a substrate will couple successfully prior to carry-
ing out an experiment. This could save valuable time
and precious resources, especially if the substrate is
a late-stage intermediate in a synthetic pathway. The
modelling also appears to be applicable to the work
carried out by Fox et al., where electron-deficient aryl dictivity of our computational method.
Figure 6. Trisubstituted aryl chlorides used to test the pre-
Adv. Synth. Catal. 0000, 000, 0 – 0
8
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÞÞ
These are not the final page numbers!

FULL PAPERS
asc.wiley-vch.de
À208C, 7.5 mol% Fe(acac)3, 1500 mol% NMP and a concen-
were all shown to give greater than 90% conversion
and found to have DE values between 1.6 and 2.2 eV
(see the Supporting Information). This strongly sug-
gests that reactants which have a calculated DE of
greater than 1.5 eV will give productive alkylation
under either our conditions, for isopropylation, or
Foxꢂs conditions, for primary alkylation.
tration of 40 volumes (LkgÀ1 of material).
General Procedure following the DoE Optimization
(Conversion Experiments)
Isopropylmagnesium chloride (2M in THF, 1.7 mL,
3.4 mmol) was added to a solution of (hetero)aryl chloride
(2 mmol) and Fe(acac)3 (53 mg, 0.15 mmol, 7.5 mol%) in
THF (7.2 mL) and NMP (2.9 mL) at À208C. The mixture
was stirred for 30 minutes then quenched with 1M HCl
(5 mL). The pH of heteroaryl halide solutions was adjusted
using Na2CO3. The reaction mixture was sampled for GC/
GCMS or HPLC/LCMS analysis.
Conclusions
Following a simple DoE optimization it was possible
to couple isopropylmagnesium chloride with electron-
deficient aryl chlorides, in very high conversion.
These reactions gave excellent selectivity for the
branched isomer, with minimal isomerization to the
n-propyl by-product. It was easy to apply the newly
developed reaction conditions to heteroaryl chlorides,
with excellent conversions and good isolated yields in
many cases. Isomerization, known to present a signifi-
cant problem for secondary alkyl species, was mini-
mal. The conditions developed were mild and highly
selective, offering significant advantages over classical
Friedel–Crafts alkylation, and other iron-catalyzed
processes. Furthermore, the coupling of (hetero)aryl
chlorides allows the use of the cheapest and most
General Procedure following the DoE Optimization
(Isolation Experiments)
Method A [preparation of 1-isopropyl-4-(trifluoromethyl)-
benzene 13b]: Isopropylmagnesium chloride (2M in THF,
28.2 mL, 56.5 mmol) was added to a solution of 1-chloro-4-
(trifluoromethyl)-benzene (6 g, 33.2 mmol) and Fe(acac)3
(0.88 g, 2.49 mmol, 7.5 mol%) in THF (159 mL) and NMP
(48 mL) at À208C. The mixture was stirred for 30 minutes
then quenched with a pH 2 buffer solution (240 mL, 0.125M
H2SO4 +0.375M Na2SO4). This mixture was further diluted
with water (400 mL) and loaded onto a 120 g Biotage KP-
C18-HS reverse phase column. The mixture was first eluted
widely available halide or phenol-derived starting ma- with water (1.2 L) then diethyl ether (300 mL). The diethyl
ether fraction was concentrated under vacuum to afford the
crude material as a yellow oil. This was further purified by
distillation (66–688C, 45 mbar) to give 1-isopropyl-4-(tri-
fluoromethyl)benzene as a clear colourless oil; yield: 2.8 g
terials. Attempts were made to predict the effective-
ness of coupling reactions using electronic structure
calculations. A predictive trend was observed between
conversion to the desired product and in silico elec-
tron affinity, calculated from energy differences be-
tween radical anions and ground states.
It was also possible to deploy cyclohexylmagnesium
chloride in the reaction due to its similarity to isopro-
pylmagnesium chloride although other Grignard re-
1
(45%). H NMR (400 MHz, CDCl3): d=7.55 (d, J=8.1 Hz,
2H), 7.34 (d, J=8.1 Hz, 2H), 2.97 (hept, J=6.8 Hz, 1H),
1.27 (d, J=6.9 Hz, 6H); 13C NMR (101 MHz, CDCl3): d=
153.0, 128.3 (q, J=32.2 Hz), 125.4 (q, J=3.7 Hz), 124.6 (q,
J=271.6 Hz), 34.3, 23.9; 19F NMR (377 MHz, CDCl3): d=
À67.6. IR: n=2966, 1619, 1323, 1118, 1069, 606 cmÀ1; MS
agents were less effective. However, through the use (EI): m/z=188.0; HR-MS (ASAP-TOF): m/z=180.0800,
calculated: 180.0813.
of a simple DoE optimization it may be possible to
expand the scope and selectivity with a wider set of
Grignard reagents and aryl chlorides, particularly if
the desired application is the introduction of a primary
or secondary alkyl group on a large-scale.
Preparation of methyl 4-isopropylbenzoate 19b: Clear col-
1
ourless oil; yield: 1.41 g (25%). H NMR (400 MHz, chloro-
form-d): d=7.96 (d, J=8.3 Hz, 1H), 7.29 (d, J=8.1 Hz,
1H), 3.90 (s, 2H), 2.96 (hept, J=6.9 Hz, 1H), 1.27 (d, J=
7.0 Hz, 3H); 13C NMR (101 MHz, CDCl3): d=167.29,
154.44, 129.87, 127.93, 126.60, 126.58, 52.07, 34.40, 23.84; IR:
n=2961, 1719, 1273, 1109, 773, 706 cmÀ1; MS (EI): m/z=
179.0; HR-MS (ESI+): m/z=179.1068, calculated: 179.1067.
Method B (preparation of tert-butyl 2-isopropylnicotinate
42): Isopropylmagnesium chloride (2M in THF, 1.9 mL,
3.77 mmol) was added to a solution of tert-butyl 2-chloroni-
cotinate (473 mg, 2.2 mmol) and Fe(acac)3 (58.7 mg,
0.17 mmol, 7.5 mol%) in THF (7.6 mL) and NMP (3.2 mL)
at À208C. The mixture was stirred for 30 minutes then
quenched with 1M HCl (5 mL) and neutralized with 5M
sodium hydroxide (2 mL). The resulting mixture was diluted
with water (10 mL) and ethyl acetate (15 mL). The phases
were separated and the organic phase was washed with
brine (5ꢃ10 mL). The organic phase was dried over magne-
sium sulfate and concentrated under vacuum. The resulting
residue was dry loaded onto a pre-packed normal phase
Experimental Section
Design of Experiments Optimization
The experimental design(s) were carried out on a 2 mmol
scale, with a concentration determined by the statistical soft-
ware (Design Expert 7 ꢄ). The reactions were carried out
using an Electrothermal Integrity 10 Reaction Station
(STEM block) to allow accurate temperature control on 8–
10 reactions in a single run. This series of experiments sug-
gested that the optimal conditions were: À208C, 7.8 mol%
Fe(acac)3, 1560 mol% NMP and a concentration of 40 vol-
umes (LkgÀ1 of material). The conditions were adjusted to:
Adv. Synth. Catal. 0000, 000, 0 – 0
9
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÞÞ
These are not the final page numbers!

FULL PAPERS
asc.wiley-vch.de
column and purified by flash chromatography (25% diethyl
ether in hexane). This gave tert-butyl 2-isopropylnicotinate
as a yellow oil; yield: 0.34 g (68%). 1H NMR (400 MHz,
CDCl3): d=8.64 (dd, J=4.7, 1.8 Hz, 1H), 7.92 (d, J=7.8 Hz,
1H), 7.13 (d, J=7.8 Hz, 1H), 3.75 (hept, J=6.8 Hz, 1H),
1.60 (s, 9H), 1.30 (d, J=6.8 Hz, 7H); 13C NMR (101 MHz,
CDCl3): d=166.4, 165.8, 150.6, 136.9, 127.0, 119.9, 81.2, 31.8,
27.7, 21.8; IR: n=2966, 1721, 1255, 1132, 1069, 783cmÀ1; MS
(EI): m/z=222.0; HR-MS (ESI+): m/z=222.1491, calculat-
ed: 222.1489.
1H), 3.96 (hept, J=6.8 Hz, 1H), 1.45 (d, J=6.8 Hz, 6H).
13C NMR (101 MHz, CDCl3): d=166.4, 142.1, 136.5, 129.7,
127.67, 127.0, 126.4, 124.9, 119.1, 31.1, 22.4. IR: n=3050,
2963, 1561, 820, 744, 680 cmÀ1; MS (ESI+): m/z=172.2; HR-
MS (ESI+): m/z=172.1115, calculated: 172.1121.
Preparation of 6-isopropylpicolinonitrile 44: Yellow oil;
1
yield: 0.64 g (61%); H NMR (400 MHz, chloroform-d): d=
7.73 (t, J=7.8 Hz, 1H), 7.51 (dd, J=7.6, 1.0 Hz, 1H), 7.39
(dd, J=8.0, 1.0 Hz, 1H), 3.10 (hept, J=6.9 Hz, 1H), 1.30 (d,
J=7.0 Hz, 6H); 13C NMR (101 MHz, CDCl3): d=169.6,
137.3, 133.3, 126.0, 124.5, 117.8, 36.5, 22.4; IR: n=2968,
2236, 1588, 1445, 811, 733 cmÀ1; MS (EI): m/z=146.0; HR-
MS (ESI+): m/z=147.0913, calculated: 147.0917.
Preparation of ethyl 2-isopropylnicotinate 43: Yellow oil;
yield: 0.29 g (69%); 1H NMR (400 MHz, CDCl3): d=8.64
(dd, J=4.7, 1.8 Hz, 1H), 7.92 (dd, J=7.8, 1.9 Hz, 1H), 7.13
(dd, J=7.8, 4.8 Hz, 1H), 3.75 (hept, J=6.8 Hz, 1H), 1.60 (s,
9H), 1.30 (d, J=6.8 Hz, 7H); 13C NMR (101 MHz, CDCl3):
d=166.4, 165.8, 150.6, 136.9, 127.0, 119.9, 81.6, 31.8, 27.7,
21.8; IR: n=2977, 1716, 1132, 1071 cmÀ1; MS (EI): m/z=
194.1; HR-MS (ESI+): m/z=194.1175, calculated: 194.1176.
Method C (preparation of 3-isopropyl-6-phenylpyridazine
46): Isopropylmagnesium chloride (2M in THF, 17.8 mL,
35.7 mmol) was added to a solution of 3-chloro-6-phenylpyr-
idazine (4 g, 21.0 mmol) and Fe(acac)3 (0.56 g, 1.57 mmol,
7.5 mol%) in THF (75 mL) and NMP (18 mL) at À208C.
The mixture was stirred for 30 minutes then quenched with
0.5M citric acid (100 mL) and neutralized with 0.5M
sodium carbonate (100 mL). The organic phase was separat-
ed, and the aqueous phase was extracted using diethyl ether
(2ꢃ100 mL). The combined organics were dried over
sodium sulfate and concentrated under vacuum. The result-
ing residue was dry loaded onto a pre-packed normal phase
column and purified by flash chromatography (0–20%
EtOAc in heptane). This gave 3-isopropyl-6-phenylpyrida-
Method D (preparation of 2-isopropylpyridine 39): Iso-
propylmagnesium chloride (2M in THF, 17.8 mL,
35.7 mmol) was added to a solution of 2-chloropyridine (2 g,
1.67 mL, 17.6 mmol) and Fe(acac)3 (0.47 g, 1.32 mmol,
7.5 mol%) in THF (63 mL) and NMP (25 mL) at À208C.
The mixture was stirred for 30 minutes then quenched with
0.5M citric acid (25 mL) and neutralized with 0.5M sodium
carbonate (25 mL). The organic phase was separated, and
the aqueous phase was extracted using tert-butyl methyl
ether (3ꢃ100 mL). The combined organics were dried over
sodium sulfate and concentrated under vacuum. The result-
ing residue was dry loaded onto a pre-packed C18 (reverse
phase) column and purified by flash chromatography [5–
70% (acetonitrile+ammonia) in (water+sodium bicarbon-
ate)]. This gave 2-isopropylpyridine as a clear colourless oil;
1
yield: 1.2 g (56%). H NMR (400 MHz, chloroform-d): d=
8.53 (d, J=4.2 Hz, 1H), 7.60 (t, J=7.6 Hz, 1H), 7.16 (d, J=
7.9 Hz, 1H), 7.09 (dd, J=7.6, 4.8 Hz, 1H), 3.06 (hept, J=
6.9 Hz, 1H), 1.30 (d, J=6.9 Hz, 6H); IR: n=2964, 1591,
1475, 1433, 731 cmÀ1; MS (EI): m/z=122.1; HR-MS (ESI+):
m/z=122.0968, calculated: 122.0964.
zine as
a
yellow solid; yield: 3.1 g (75%). 1H NMR
(400 MHz, CDCl3): d=8.09 (d, J=8.1 Hz, 2H), 7.79 (d, J=
8.8 Hz, 1H), 7.56–7.44 (m, 3H), 7.41 (d, J=8.8 Hz, 1H),
3.36 (hept, J=7.0 Hz, 1H), 1.42 (d, J=7.0 Hz, 6H);
13C NMR (101 MHz, CDCl3): d=166.9, 157.5, 136.7, 129.8,
129.1, 127.0, 125.0, 124.2, 34.8, 22.5; IR: n=3048, 2957, 1422,
750, 693; MS (EI): m/z=197.2; HR-MS (ESI+): m/z=
199.1221, calculated: 199.1230.
Preparation of 3-isopropylpyridine 40: Orange oil; yield:
1
0.5 g (47%); H NMR (400 MHz, chloroform-d): d=8.49 (s,
1H), 8.43 (d, J=4.9 Hz, 1H), 7.54 (d, J=7.9 Hz, 2H), 7.22
(dd, J=7.7, 4.9 Hz, 1H), 2.94 (hept, J=7.0 Hz, 2H), 1.27
(dd, J=6.9, 1.0 Hz, 9H); 13C NMR (101 MHz, CDCl3): d=
148.77, 147.44, 143.88, 133.92, 123.50, 31.92, 23.80; IR: n=
2962, 1422, 713 cmÀ1; MS (EI): mz=?????; HR-MS (ESI+):
m/z=122.0968, calculated: 122.0964.
Preparation of (4-isopropylphenyl)(morpholino)metha-
1
none 18b: Clear colourless oil; yield: 0.94 g (46%). H NMR
(400 MHz, CDCl3): d=7.34 (d, J=8.2 Hz, 2H), 7.26 (d, J=
8.1 Hz, 2H), 3.73 (br. s, 6H), 3.50 (br. s, 2H), 2.93 (hept, J=
6.9 Hz, 1H), 1.25 (d, J=6.9 Hz, 6H); 13C NMR (101 MHz,
CDCl3): d=170.6, 150.9, 132.7, 127.3, 126.6, 66.9, 48.2, 42.6,
34.0, 23.8. IR: n=2960, 2856, 1629, 1422, 1277, 1010,
837 cmÀ1; MS (EI): m/z=232.2; HR-MS (ESI+): m/z=
234.1489, calculated: 234.1478.
Preparation of 2-methoxy-6-isopropylpyridine 41: Clear
colourless oil; yield: 1.7 g (79%). 1H NMR (400 MHz,
chloroform-d): d=7.47 (dd, J=8.2, 7.3 Hz, 1H), 6.71 (d, J=
7.3 Hz, 1H), 6.53 (d, J=8.1 Hz, 1H), 3.92 (s, 3H), 2.94
(hept, J=6.9 Hz, 1H), 1.27 (d, J=6.9 Hz, 6H); 13C NMR
(101 MHz, CDCl3): d=165.43, 163.63, 138.85, 113.04, 107.35,
53.18, 36.04, 22.52; IR: n=2964, 1578, 1286, 1029, 799,
732 cmÀ1; MS (ESI+): m/z=152.1; HR-MS (ESI+): m/z=
152.1065, calculated: 152.1070.
Method E (preparation of 6-chloro-2-isopropylquinoline
47): Isopropylmagnesium chloride (2M in THF, 7.6 mL,
15.2 mmol) was added to a solution of 2,6-dichloroquinoline
(2.5 g, 12.6 mmol) and Fe(acac)3 (0.33 g, 0.95 mmol,
7.5 mol%) in THF (48 mL) and NMP (18 mL) at À208C.
The mixture was stirred for 30 minutes then quenched with
0.5M citric acid (25 mL) and neutralized with 0.5M sodium
carbonate (25 mL). The organic phase was separated, and
the aqueous phase was extracted using tert-butyl methyl
ether (3ꢃ100 mL). The combined organics were dried over
sodium sulfate and concentrated under vacuum. The result-
Preparation of 2-(4-isopropylphenyl)pyridine 21b: Clear
colourless oil; yield: 1.0 g (48%); 1H NMR (400 MHz,
CDCl3): d=8.67 (d, J=4.7 Hz, 1H), 7.92 (d, J=8.3 Hz, 2H),
7.77–7.66 (m, 2H), 7.33 (d, J=8.2 Hz, 2H), 7.19 (dd, J=6.2,
4.9 Hz, 1H), 2.97 (hept, J=6.8 Hz, 1H), 1.29 (d, J=6.9 Hz,
6H); 13C NMR (101 MHz, CDCl3): d=158.0, 150.0, 149.7,
137.2, 136.7, 127.0, 127.0, 121.9, 120.4, 34.1, 24.1. IR: n=
2959, 2869, 1587, 1466, 1435, 779 cmÀ1; MS (EI): m/z=197.1;
HR-MS (ESI+): m/z=198.1273, calculated: 198.1277.
Preparation of 1-isopropylisoquinoline 45: Red oil; yield:
1
2.96 g (71%); H NMR (400 MHz, CDCl3): d=8.50 (s, 1H),
8.22 (d, J=8.4 Hz, 1H), 7.81 (d, J=8.1 Hz, 1H), 7.65 (t, J=
7.5 Hz, 1H), 7.58 (t, J=7.6 Hz, 1H), 7.48 (d, J=5.7 Hz,
Adv. Synth. Catal. 0000, 000, 0 – 0
10
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÞÞ
These are not the final page numbers!

FULL PAPERS
asc.wiley-vch.de
ing residue was dry loaded onto a pre-packed C18 (reverse
phase) column and purified by flash chromatography [5–
70% (acetonitrile+ammonia) in (water+sodium bicarbon-
ate)]. This gave 6-chloro-2-isopropylquinoline as a yellow
[13] C. E. Garrett, K. Prasad, Adv. Synth. Catal. 2004, 346,
889.
[14] P. Chirik, R. Morris, Acc. Chem. Res. 2015, 48, 2495.
[15] a) I. Bauer, H.-J. Knçlker, Chem. Rev. 2015, 115, 3170;
b) B. D. Sherry, A. Fꢁrstner, Acc. Chem. Res. 2008, 41,
1500; c) C. Bolm, J. Legros, J. Le Paih, L. Zani, Chem.
Rev. 2004, 104, 6217.
[16] V. G. P. Mottez, C. R. Acad. Sci. 1944, 218, 557.
[17] M. Tamura, J. K. Kochi, J. Am. Chem. Soc. 1971, 93,
1487.
1
oil; yield: 1.9 g (73%). H NMR (400 MHz, chloroform-d):
d=7.98 (dd, J=8.8, 4.3 Hz, 2H), 7.74 (d, J=2.3 Hz, 1H),
7.60 (dd, J=9.0, 2.4 Hz, 1H), 7.35 (d, J=8.6 Hz, 1H), 3.24
(hept, J=6.9 Hz, 1H), 1.39 (d, J=6.9 Hz, 6H); 13C NMR
(101 MHz, CDCl3): d=168.10, 146.26, 135.56, 131.37, 130.79,
130.20, 127.63, 126.22, 120.30, 37.37, 22.55; IR: n=2964,
1597, 1490, 1071, 830 cmÀ1; MS (ESI+): m/z=206.2; HR-MS
(ESI+): m/z=206.0728, calculated: 206.0731.
[18] A. Fꢁrstner, A. Leitner, M. Mendez, H. Krause, J. Am.
Chem. Soc. 2002, 124, 13856.
[19] T. Agrawal, S. P. Cook, Org. Lett. 2013, 15, 96.
[20] M. C. Perry, A. N. Gillett, T. C. Law, Tetrahedron Lett.
2012, 53, 4436.
[21] R. Agata, T. Iwamoto, N. Nakagawa, K. Isozaki, T. Ha-
takeyama, H. Takaya, M. Nakamura, Synthesis 2015,
47, 1733.
Acknowledgements
We thank Dr. Colin Edge, Dr. Martin Saunders and Dr. Greg
M. Anderson for their support with Gaussian 09 calculations
and HPC hardware at GSK in Stevenage. We also thank
GSK and the University of Strathclyde for funding.
[22] X. Chen, Z.-J. Quan, X.-C. Wang, Appl. Organomet.
Chem. 2015, 29, 296.
[23] A. Rudolph, M. Lautens, Angew. Chem. 2009, 121,
2694; Angew. Chem. Int. Ed. 2009, 48, 2656.
[24] a) R. Martin, A. Fꢁrstner, Angew. Chem. 2004, 116,
4045; Angew. Chem. Int. Ed. 2004, 43, 3955; b) T.
Nagano, T. Hayashi, Org. Lett. 2004, 6, 1297; c) M. Na-
kamura, K. Matsuo, S. Ito, E. Nakamura, J. Am. Chem.
Soc. 2004, 126, 3686; d) R. B. Bedford, D. W. Bruce,
R. M. Frost, M. Hird, Chem. Commun. 2005, 4161;
e) G. Cahiez, V. Habiak, C. Duplais, A. Moyeux,
Angew. Chem. 2007, 119, 4442; Angew. Chem. Int. Ed.
2007, 46, 4364.
References
[1] a) M. Baumann, I. R. Baxendale, Beilstein J. Org.
Chem. 2013, 9, 2265; b) T. Minami, M. Iwamoto, H.
Ohtsu, H. Ohishi, R. Tanaka, A. Yoshitake, Planta
Med. 2002, 68, 742.
[2] M. Rueping, B. J. Nachtsheim, Beilstein J. Org. Chem.
2010, 6, 6.
[3] a) L.-Q. Li, M.-M. Li, D. Chen, H.-M. Liu, H.-c. Geng,
J. Lin, H.-B. Qin, Tetrahedron Lett. 2014, 55, 5960;
b) B. N. Kakde, P. Kumari, A. Bisai, J. Org. Chem.
2015, 80, 9889; c) B. N. Kakde, A. Parida, P. Kumari, A.
Bisai, Tetrahedron Lett. 2016, 57, 3179.
[4] a) D. G. Brown, J. Bostrçm, J. Med. Chem. 2016, 59,
4443; b) S. D. Roughley, A. M. Jordan, J. Med. Chem.
2011, 54, 3451.
[25] P. J. Rushworth, D. G. Hulcoop, D. J. Fox, J. Org.
Chem. 2013, 78, 9517.
[26] a) T. L. Mako, J. A. Byers, Inorg. Chem. Front. 2016, 3,
766; b) G. Cahiez, H. Avedissian, Synthesis 1998, 1199.
[27] a) R. Leardi, Anal. Chim. Acta 2009, 652, 161; b) H.
Tye, Drug Discovery Today 2004, 9, 485.
[28] W. M. Czaplik, S. Grupe, M. Mayer, A. Jacobi von
Wangelin, Chem. Commun. 2010, 46, 6350.
[5] a) D. L. Reger, E. C. Culbertson, Inorg. Chem. 1977,
16, 3104; b) R. Jana, T. P. Pathak, M. S. Sigman, Chem.
Rev. 2011, 111, 1417.
[6] C. Valente, S. C¸ alimsiz, K. H. Hoi, D. Mallik, M. Sayah,
M. G. Organ, Angew. Chem. 2012, 124, 3370; Angew.
Chem. Int. Ed. 2012, 51, 3314.
[7] N. Miyaura, A. Suzuki, Chem. Rev. 1995, 95, 2457.
[8] T. Kohei, S. Koji, K. Yoshihisa, Z. Michio, F. Akira, K.
Shun-ichi, N. Isao, M. Akio, K. Makoto, Bull. Chem.
Soc. Jpn. 1976, 49, 1958.
[9] a) A. O. King, N. Okukado, E.-i. Negishi, J. Chem. Soc.
Chem. Commun. 1977, 683; b) A. Joshi-Pangu, M.
Ganesh, M. R. Biscoe, Org. Lett. 2011, 13, 1218.
[10] G. K. Surya Prakash, P. Yan, B. Tçrçk, I. Bucsi, M.
Tanaka, G. A. Olah, Catal. Lett. 2003, 85, 1.
[29] J. Kleimark, A. Hedstrçm, P.-F. Larsson, C. Johansson,
P.-O. Norrby, ChemCatChem 2009, 1, 152.
[30] a) R. B. Bedford, M. Nakamura, N. J. Gower, M. F.
Haddow, M. A. Hall, M. Huwe, T. Hashimoto, R. A.
Okopie, Tetrahedron Lett. 2009, 50, 6110; b) I. Thome,
A. Nijs, C. Bolm, Chem. Soc. Rev. 2012, 41, 979; c) S. L.
Buchwald, C. Bolm, Angew. Chem. 2009, 121, 5694;
Angew. Chem. Int. Ed. 2009, 48, 5586.
[31] G. Lefevre, A. Jutand, Chem. Eur. J. 2014, 20, 4796.
[32] C. Hansch, A. Leo, R. W. Taft, Chem. Rev. 1991, 91,
165.
[33] J. Kleimark, P.-F. Larsson, P. Emamy, A. Hedstrçm, P.-
O. Norrby, Adv. Synth. Catal. 2012, 354, 448.
[34] C.-G. Zhan, J. A. Nichols, D. A. Dixon, J. Phys. Chem.
A 2003, 107, 4184.
[11] O. V. Shutkina, O. A. Ponomareva, I. I. Ivanova, Pet.
[35] a) A. L. Speelman, J. G. Gillmore, J. Phys. Chem. A
2008, 112, 5684; b) E. J. Lynch, A. L. Speelman, B. A.
Curry, C. S. Murillo, J. G. Gillmore, J. Org. Chem. 2012,
77, 6423; c) D. D. Mꢅndez-Hernꢆndez, J. G. Gillmore,
L. A. Montano, D. Gust, T. A. Moore, A. L. Moore, V.
Mujica, J. Phys. Org. Chem. 2015, 28, 320.
Chem. 2013, 53, 20.
[12] a) E. Peyroux, F. Berthiol, H. Doucet, M. Santelli, Eur.
J. Org. Chem. 2004, 2004, 1075; b) T. Bach, S. Heuser,
Angew. Chem. 2001, 113, 3283; Angew. Chem. Int. Ed.
2001, 40, 3184; c) C. Dai, G. C. Fu, J. Am. Chem. Soc.
2001, 123, 2719; d) G. Bastug, S. P. Nolan, Organome-
tallics 2014, 33, 1253.
[36] C. Costentin, M. Robert, J.-M. Savꢅant, J. Am. Chem.
Soc. 2004, 126, 16051.
Adv. Synth. Catal. 0000, 000, 0 – 0
11
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÞÞ
These are not the final page numbers!

FULL PAPERS
12 Iron-Catalyzed Isopropylation of Electron-Deficient Aryl
and Heteroaryl Chlorides
Adv. Synth. Catal. 2017, 359, 1 – 12
James N. Sanderson,* Andrew P. Dominey,
Jonathan M. Percy*
Adv. Synth. Catal. 0000, 000, 0 – 0
12
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÞÞ
These are not the final page numbers!