Trifluoroacetic acid in 2,2,2-trifluoroethanol facilitates SNAr reactions of heterocycles with arylamines

Article abstract of DOI:10.1002/chem.201304336

Small-molecule drug discovery requires reliable synthetic methods for attaching amino compounds to heterocyclic scaffolds. Trifluoroacetic acid-2,2,2-trifluoroethanol (TFA-TFE) is as an effective combination for achieving S_NAr reactions between

Full text of DOI:10.1002/chem.201304336

DOI: 10.1002/chem.201304336
Full Paper
&
Synthetic Methods
Trifluoroacetic Acid in 2,2,2-Trifluoroethanol Facilitates SNAr
Reactions of Heterocycles with Arylamines**
Benoit Carbain, Christopher R. Coxon, Honorine Lebraud, Kristopher J. Elliott,
Christopher J. Matheson, Elisa Meschini, Amy R. Roberts, David M. Turner,
Christopher Wong, Celine Cano, Roger J. Griffin, Ian R. Hardcastle, and Bernard T. Golding*[a]
Abstract: Small-molecule drug discovery requires reliable
synthetic methods for attaching amino compounds to heter-
ocyclic scaffolds. Trifluoroacetic acid-2,2,2-trifluoroethanol
(TFA-TFE) is as an effective combination for achieving SNAr
reactions between anilines and heterocycles (e.g., purines
and pyrimidines) substituted with a leaving group (fluoro-,
chloro-, bromo- or alkylsulfonyl). This method provides a vari-
ety of compounds containing a “kinase-privileged fragment”
associated with potent inhibition of kinases. TFE is an advan-
tageous solvent because of its low nucleophilicity, ease of
removal and ability to solubilise polar substrates. Further-
more, TFE may assist the breakdown of the Meisenheimer–
Jackson intermediate by solvating the leaving group. TFA is
a necessary and effective acidic catalyst, which activates the
heterocycle by N-protonation without deactivating the ani-
line by conversion into an anilinium species. The TFA-TFE
methodology is compatible with a variety of functional
groups and complements organometallic alternatives, which
are often disadvantageous because of the expense of re-
agents, the frequent need to explore diverse sets of reaction
conditions, and problems with product purification. In con-
trast, product isolation from TFA-TFE reactions is straightfor-
ward: evaporation of the reaction mixture, basification and
chromatography affords analytically pure material. A total of
45 examples are described with seven discrete heterocyclic
scaffolds and 2-, 3- and 4-substituted anilines giving product
yields that are normally in the range 50–90%. Reactions can
be performed with either conventional heating or micro-
wave irradiation, with the latter often giving improved
yields.
Introduction
The bis-arylaniline unit has been termed a “kinase-privileged
fragment” owing to its frequent appearance as a structural
motif in kinase inhibitors, as exemplified by imatinib
(Figure 1).[1] Given the importance of kinases in current drug
discovery programs,[1] it is vital that efficient synthetic methods
are available to access bis-arylanilines. In this context, one of
the most important reaction classes aiding the synthesis of bis-
arylanilines is the attachment of an aniline to a heterocyclic
scaffold.[2] Numerous examples of this class have been de-
scribed whereby an amine is coupled to a heteroaryl halide
catalysed by a metallo-phosphine species (e.g., Buchwald–
Figure 1. Example of a kinase-privileged fragment (e.g., X=N, Y=H) and
a drug (imatinib) containing this fragment.
Hartwig methodology).[3] An alternative approach achieves
such couplings by direct reaction (SNAr) of a suitably activated
heteroaryl halide with an amine (e.g.,[4]). In projects seeking to
identify potent lead inhibitors of kinases, for example, cyclin-
dependent kinases,[5,6] we required series of arylamino-hetero-
cycles for the development of structure–activity relationships
(SAR). Initially, we employed classical SNAr conditions whereby
a 2-fluoropurine was heated with an aniline in 2-butanol.[7,8]
Whereas this procedure was satisfactory with relatively basic/
nucleophilic anilines, it failed with weakly nucleophilic anilines,
such as 4-nitroaniline. Although the desired products might
have been accessible by treating the aniline with a 2-iodopur-
ine in the presence of a palladium/phosphine catalyst, we ex-
plored alternative SNAr conditions, resulting in the discovery of
the versatile and reliable protocol described herein. In earlier
studies we found that the treatment of 2-fluoro-6-cyclohexyl-
[a] Dr. B. Carbain, Dr. C. R. Coxon, H. Lebraud, Dr. K. J. Elliott,
Dr. C. J. Matheson, Dr. E. Meschini, A. R. Roberts, Dr. D. M. Turner,
Dr. C. Wong, Dr. C. Cano, Prof. R. J. Griffin, Dr. I. R. Hardcastle,
Prof. B. T. Golding
Newcastle Cancer Centre, Northern Institute for Cancer Research
Bedson Building, Newcastle University
Newcastle Upon Tyne, NE1 7RU (UK)
E-mail: bernard.golding@newcastle.ac.uk
[**] B.T.G. conceived the studies, which were co-supervised with C.C., R.J.G. and
I.R.H.; B.T.G. wrote the manuscript aided by B.C., C.C., H.L. and R.J.G.; exper-
imental work was conducted by B.C., H.L., C.R.C., K.J.E., C.J.M., E.M., A.R.R.,
D.M.T. and C.W.
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201304336.
Chem. Eur. J. 2014, 20, 2311 – 2317
2311
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
methoxypurine with aniline appeared to accelerate at later re-
action times.[7] This was ascribed to the formation of hydro-
fluoric acid, which autocatalysed the reaction, presumably by
activation of the purine by N-protonation. An investigation, in
which various acid catalysts and solvents were explored, led to
the identification of trifluoroacetic acid-2,2,2-trifluoroethanol
(TFA-TFE) as a convenient and efficacious combination for ef-
fecting SNAr reactions with purines and pyrimidines.[5,7,9] The
benefit of acidic catalysis for SNAr reactions of appropriate het-
erocycles had already been recognised long ago by Banks[10,11]
and Chapman,[12,13] and subsequent to our initial studies, fur-
ther quantified by Liu and Robins.[14] The use of acidic catalysis
in SNAr reactions employed for the scale-up synthesis of drugs
is increasingly commonplace.[4,15]
With 2-substituted anilines the reaction is only effective when
the 2-substituent is electron-releasing (e.g., OMe). Although
most of the entries in the tables are for heterocycles, we have
also studied 2,4-dinitrofluorobenzene. In this case, and of
mechanistic significance (see below), is the observation that
whereas TFE is a beneficial solvent with respect to reaction
rate, addition of TFA has no accelerating effect because sub-
strate reactivity cannot be enhanced by protonation.
Mechanistic studies
Reactions of anilines with 4,7-dichloroquinoline at 258C were
studied by varying the concentration of TFA in TFE from 0.5 up
to 10 equivalents (Table 3). The optimal consumption of 4,7-di-
chloroquinoline with aniline occurred with approximately
1 equiv TFA. This gave good conversion of substrate into prod-
uct with the reaction being essentially complete after 512 min.
Further experiments were conducted with 4-substituted ani-
lines (4-methoxy, 4-chloro, 4-trifluoromethyl and 4-cyano) of
different basicity/nucleophilicity (Table 3; see also Supporting
Information). Optimal conversions were in the range 0.5–
2.5 equiv TFA, although only the reactions of 4-methoxy and 4-
chloro-aniline were complete after 512 min (4-trifluoromethyl,
88% complete; 4-cyano, 36% complete with 1 equiv TFA; see
Tables S5 and S6 in the Supporting Information). For the heter-
oaryl substrates in Tables 1 and 2, the preferred amount of
added TFA was found to be in the range 2.5–5 equivalents,
compared to 2-amino-6-chloropurine with aniline (Table 3).
Comparative SNAr reactions of 1-fluoro-2,4-dinitrobenzene
(Sanger’s reagent) were performed with 4-methoxyaniline in
TFE alone and in TFA-TFE at 228C. In TFE alone, the reaction
was complete within 5 min, whereas in the presence of TFA
(4 equivalents) the reaction was only complete after at least
6 h (Table S7 in the Supporting Information).
In this study, we describe the application of TFA-TFE to facili-
tate SNAr reactions of arylamines with a variety of heterocyclic
templates and leaving groups, which can be selected from
fluoro-, chloro-, bromo- and alkylsulfonyl. The roles of both
TFA and TFE in facilitating SNAr reactions are discussed and
partly validated experimentally. The methodology is generally
applicable to diverse heterocyclic substrates. It complements
organometallic methodology and is even preferable in many
cases because of the seldom need to explore varied sets of
conditions. Furthermore, the TFA-TFE methodology is compati-
ble with a variety of functional groups, some of which (e.g., al-
kynyl, thioamido, 1,3-dithian-2-yl) may be unsuitable for metal-
assisted couplings. In addition, the use of certain protected
heteroaryl substrates and anilines (e.g., N9-tetrahydropyranyl-
purines, N-butyloxycarbonyl-anilines) can reduce the number
of steps in a synthetic sequence by in situ deprotection during
the TFA-TFE reaction.
Results
Synthesis of substrates
The syntheses of heterocyclic substrates used in this study,
when not commercially available, have either been de- Discussion
scribed[5,7–9] or utilised standard methods (see the Experimental
General comments on the methodology
Section and the Supporting Information).
The SNAr reaction is a classical method for the introduction of
functional groups into aromatic and heteroaromatic systems,[16]
and can be traced back at least to 1854 when Pisani[17] treated
2,4,6-trinitrochlorobenzene with ammonia to give picramide.
The epithet SNAr came much later and is attributed to A. J.
Parker (1961)[18] after Bunnett and Zahler[19–21] had defined the
mechanism of such reactions as an addition-elimination pro-
cess via a Meisenheimer–Jackson[22,23] intermediate. Today the
SNAr reaction is still much used, but is frequently replaced by
metal-mediated cross-coupling between an aryl halide or tri-
flate and an amine (e.g., Buchwald–Hartwig reactions[24–28]).
Although many powerful protocols have been described for
such reactions, especially utilising palladium or copper cata-
lysts, the cocktail of reagents required and myriad of published
catalysts employed often demands time-consuming optimisa-
tion with new targets. For this reason and also considering re-
agent costs, the SNAr reaction is still a mainstay whenever an
Scope of the reaction
Table 1 and Table 2 summarise the panel of heterocycles and
anilines that have been examined (further examples are given
in the Supporting Information, Table S1). The methodology
therefore embraces a wide selection (45 examples) of halogen-
ated (hetero)aryl substrates (purine, pyrimidine, pyrazolo[3,4-
d]pyrimidine, [1,2,3]triazolo [3,4-d]pyrimidine, imidazo[1,2-
a]pyrimidine, pyrazolo[1,5-a]pyrimidine, quinoline, dinitroben-
zene) and a range of substituted anilines. Instead of a halide,
the leaving group can also be an alkylsulfonyl group. Yields are
generally good to excellent, although modest yields may be
obtained with less reactive systems, for which microwave heat-
ing is invariably beneficial. The most reactive anilines are those
with substituents at the 3- and/or 4-positions, including elec-
tron-withdrawing groups (NO2, SO2NH2, CO2H, CF3 and CN).
Chem. Eur. J. 2014, 20, 2311 – 2317
www.chemeurj.org
2312
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
arylamino group requires attach-
Table 1. Purine derivatives for SNAr reactions with anilines facilitated by TFA-TFE.
ment to
plate.[2,8,29]
a heterocyclic tem-
In earlier studies we focused
on the use of TFA-TFE with con-
ventional heating.[7] We have
now found that the more diffi-
cult substitutions can be facilitat-
ed by microwave heating.[15,30]
Furthermore, it is sometimes
convenient to use an aniline hy-
drochloride as reactant, with the
hydrogen chloride liberated pro-
viding acidic catalysis. Most reac-
tions proceed with minimal by-
product formation, although tri-
fluoroethanol may occasionally
compete with anilines of low nu-
cleophilicity and yield a trifluor-
oethoxy-substituted by-product.
Product isolation is straightfor-
ward: evaporation of the reac-
tion mixture, basification to neu-
tralise trifluoroacetate salts and
finally chromatography gives an-
alytically pure material.
Substrate
Product
TFA [equiv] Heat at reflux
mW
t [h] Yield [%] t [h] Yield [%]
1
2
5
5
–
–
1.5[a] 82
48
74
–
–
–
–
3
4
5
5
24
24
71
70
0.5[a] 89
Mechanisms of SNAr reactions
facilitated by TFA-TFE
5
6
2.5
2.5
–
–
–
–
2[a]
43
90
The rates of SNAr reactions of ar-
omatic compounds are gov-
erned by numerous factors, but
amongst these, the benefit of
electron-withdrawing substitu-
ents is well-known.[20,31] For het-
eroaromatic systems, the loca-
tion of a nitrogen atom in an
1[a]
a
or g position to a leaving
7
8
2.5
5
–
–
–
–
1[a]
90
71
group is advantageous.[20] Fur-
thermore, protonation of a nitro-
gen atom of a heteroaromatic
ring can enhance the rate of
a SNAr reaction.[10–13] However,
addition of an acid may deacti-
vate the nucleophilic component
of the reaction. The aniline
needs to be in its non-protonat-
ed form, and thus increasing the
amount of TFA will augment the
population of unreactive proton-
ated aniline. On the other hand,
reducing the concentration of
TFA will compromise the activa-
tion of a nitrogenous heterocy-
cle and the rate of reaction will
be reduced (Table 3). The pro-
1[b]
9
2.5
5
–
–
–
–
1[b]
68
83
10
1[b]
11
5
–
–
0.5[a] 66
Chem. Eur. J. 2014, 20, 2311 – 2317
www.chemeurj.org
2313
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
For the heterocycles described
herein there are no reliable pKa
data available, and it is therefore
difficult to quantify the degree
of protonation because there are
several basic sites. By analogy
with adenine derivatives, the pri-
mary sites of protonation of
the purines are expected to be
N-1 and N-7, and possibly
Table 1. (Continued)
Substrate
Product
TFA [equiv] Heat at reflux
mW
t [h] Yield [%] t [h] Yield [%]
12
5
24
61
–
–
[a] mW (microwave irradiation) at 1408C; [b] mW at 1608C.
N-3.[14,37,38]
Protonation
of
a purine at N-1 (or N-3) followed
by attack of an aniline leads to
posed effect of changing acid concentration on reaction rate
with one reactant protonated (e.g., purine in the present
study) and the other not (aniline) is well established for
a number of cases, for example, oxime formation.[32] The results
in Table 3 imply that although the pKa determines the concen-
tration of free aniline, it is the pKa of the nitrogenous heteroar-
yl substrate that is the key parameter dictating the optimal
quantity of TFA. On account of the similarity of TFE and water
the Meisenheimer–Jackson intermediate. The basic mechanistic
scheme is illustrated in Scheme 1 for the SNAr reaction of ani-
lines with 6-cyclohexylmethoxy-2-fluoropurine. The beneficial
role of TFE may reside in its ability to solvate the leaving
group by hydrogen bonding during the breakdown of the Mei-
senheimer–Jackson intermediate. This suggestion requires that
decomposition of the intermediate is rate limiting[20,23,39] and is
most likely when fluoride is the leaving group. A study of TFA-
catalysed reactions of 6-halopurines (halo=F, Cl, Br and I) with
aniline in acetonitrile concluded that loss of halide from the
Meisenheimer–Jackson intermediate was indeed rate limit-
ing.[14] The importance of fluoride solvation in solvolysis of
alkyl fluorides (e.g., 4-methoxybenzyl fluoride in water) is well-
known.[33] A marked effect of TFE relative to ethanol in solvoly-
sis of glycosyl fluorides was ascribed to better solvation of
fluoride by TFE.[40] In another recent study, it was shown that
for reaction of 2,5,6-trifluoronicotinenitrile with the amino
group of 3-isopropoxy-1H-pyrazol-5-amine in tetrahydrofuran,
monoprotonated 1,4-diaza [2,2,2]-bicyclooctane assisted the
loss of the 6-fluoro group from the intermediate.[41] In this
case, formation of the Meisenheimer–Jackson intermediate
was rate limiting, with fluoride loss being a fast process.
with respect to polarity (EN for TFE: 0.90; for water: 1.00),[33]
T
pKa values of anilines in TFE are likely to be similar to those
measured in water (compare pKa values of anilines in methanol
and ethanol, which correlate closely with the polarities of
these solvents relative to water).[34] It should be noted that at
258C, with or without TFA, no reaction was observed for 4,7-di-
chloroquinoline and benzylamine (pKa 9.34) during 8.5 h in
TFE. In this case, benzylamine is deactivated by TFA protona-
tion, whereas in TFE alone the lack of activation of the quino-
line by protonation also precludes any reaction.
The finding that 1-fluoro-2,4-dinitrobenzene reacts much
faster with 4-methoxyaniline in TFE compared to TFA-TFE, is
consistent with the premise that in this case there is no activa-
tion of the fluoro substrate by protonation. The presence of
TFA merely serves to reduce the concentration of free 4-me-
thoxyaniline. It has been proposed that fluoride loss from the
reversibly generated Meisenheimer–Jackson intermediate is
rate limiting when weakly nucleophilic anilines react with 1-
fluoro-2,4-dinitrobenzene.[35,36]
Compared to, for example, ethanol, TFE is markedly more
acidic (pKa 12.4[30] vs. 16). This property of TFE may enhance
the reactivity of a monoprotonated heterocycle by hydrogen
bonding to a non-protonated heteroatom. Besides playing
a key mechanistic role (see below) TFE is also an advantageous
Scheme 1. Proposed intermediates in the TFA-facilitated SNAr reactions of anilines (X=H, MeO, NO2, Br, SO2NH2, CO2H, Cl, CF3 and CN) exemplified with 6-cy-
clohexylmethoxy-2-fluoropurine.
Chem. Eur. J. 2014, 20, 2311 – 2317
www.chemeurj.org
2314
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Conclusion
Table 2. Further heterocyclic derivatives for SNAr reactions with anilines facilitated by TFA-TFE.
The use of trifluoroacetic acid in
2,2,2,-trifluoroethanol is an effi-
cient, practical and reliable
methodology for the acidic cat-
alysis of SNAr reactions of ani-
lines with nitrogen-containing
heteroaromatic substrates, ex-
emplified by 45 distinct exam-
ples (Tables 1–3, and Table S1 in
the Supporting Information). We
anticipate the use of TFA in TFE
to facilitate SNAr reactions of nu-
merous other heterocycles, with
optimisation of reaction condi-
tions following the principles
outlined above. We have found
TFA-TFE to be immensely useful
for the synthesis of numerous
heterocyclic derivatives required
for SAR studies in medicinal
chemistry research programs.
For example, the methodology
proved invaluable for the discov-
ery of NU6102, one of the most
potent and selective inhibitors
of the cyclin-dependent kinase
CDK2.[6] We have recently dem-
onstrated the value of TFA-TFE
in the synthesis of N7-methyl-
and N7-ethyl-purines,[46] and also
for facilitating the Paal–Knorr
synthesis of pyrroles under mi-
crowave heating.[47] TFE alone
has many benefits as a solvent[48]
including facilitation of: imino-
Diels–Alder reactions,[49] reduc-
tive aminations of aldehydes
and ketones,[50] and nucleophilic
ring openings of epoxides.[51]
The “magical” solvent properties
of TFE and other fluorinated
solvents (e.g., 1,1,1,3,3,3-hexa-
fluoro-2-propanol) have been
denoted as a “booster effect” rel-
ative to non-fluorinated ana-
logues and quantified by com-
putational studies.[52]
Substrate
Product
TFA [equiv]
Heat at reflux
t [h] Yield [%]
13
14
5
24
24
73
82
5
15
2.5
18
73
16
17
5
5
18
24
35
34
18
17.6
48
49
19[a]
1
0
9
1
95
80
20[a]
[a] At 208C.
solvent because of its low nucleophilicity, ease of removal (b.p.
788C) and hence suitability for recycling, as well as its ability to
solubilise numerous polar substrates.
Experimental Section
General information
For details concerning equipment and chemicals see the Support-
ing Information.
Chem. Eur. J. 2014, 20, 2311 – 2317
www.chemeurj.org
2315
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
to give the desired product as a pale yellow oil (660mg,
Table 3. Product [%] at 32 min for reactions of 4,7-dichloro-quinoline, 2-amino-6-
chloropurine and 1-fluoro-2,4-dinitrobenzene with anilines, varying the concentration
of TFA (0, 0.5, 1, 2.5, 5 and 10 equiv) in TFE at 258C.
89%). Rf =0.53 (DCM/MeOH 9:1); UV lmax (EtOH):
274 nm; IR (KBr): v˜ =2940, 2860, 2550, 1600, 1570 cmÀ1
;
1H NMR (500 MHz, CDCl3): d=1.14–1.18 (21H, m,
CH(CH3)2), 7.09 (1H, t, J=7.3 Hz, Ar-H), 7.21 (1H, brs,
NH), 7.32–7.35 (2H, m, Ar-H), 7.37 (1H, s, H-8), 7.53–7.57
(2H, m, Ar-H), 11.75 ppm (1H, brs, N9-H); 13C NMR
(125 MHz, CDCl3): d=11.3, 18.7, 100.8, 101.8, 120.7,
123.7, 129.3, 139.3, 140.9, 142.5, 153.2, 156.5, 184.8 ppm;
HRMS calcd for C22H30N5Si [M+H]+ 392.2270, found
392.2263.
Heteroaryl
substrate
Aniline
TFA equivalents
0
0
0.5
37
1
2.5
5
10
0
51
53
0
6
3
6-Cyclohexylmethoxy-7-methyl-2-(4’-sulfamoylanilino)-
7H-purine (12): The title compound was synthesised ac-
cording to the general procedure A using 6-cyclohexyl-
methoxy-2-fluoro-7-methyl-7H-purine (50mg, 0.19mmol),
sulfanilamide (65mg, 0.38mmol) and TFA (0.07mL,
0.95mmol) in TFE (4mL). The crude product was purified
by MPLC on silica (EtOAc/MeOH 9:1) to give a pale
brown powder (48mg, 61%). Rf =0.48 (EtOAc/MeOH
9:1); m.p. 252–2538C; UV lmax (EtOH): 292 nm; IR (neat
0
3
48
35
60
53
60
51
0
0
0
0
17
22
19
0
0
0
0
powder):
v ;
max =3373, 3313, 2922, 1569, 1504 cmÀ1
1H NMR (300 MHz, [D6]DMSO): d=0.95–1.80 (11H, m, H-
cyclo), 3.91 (3H, s, CH3), 4.33 (2H, d, J=5.4 Hz, OCH2),
7.11 (2H, s, NH2), 7.70 (2H, d, J=8.4 Hz, Ar-H), 7.97 (2H,
d, J=8.4 Hz, Ar-H), 8.19 (1H, s, H-8), 9.65 ppm (1H, s,
NH); 13C NMR (75 MHz, [D6]DMSO); d=25.6, 26.3, 29.6,
33.9, 37.1, 71.7, 108.7, 117.5, 126.8, 135.8, 144.8, 147.0,
155.1, 157.4, 162.9 ppm; MS (ES+) m/z 417 [M+H]+;
HRMS calcd for C19H25N6O3S [M+H]+ 417.1703, found
417.1704.
8[a]
11[a]
11[a]
0
0
9[a]
26[a]
73[a]
68[a]
4-(6-Amino-4-cyclohexylmethoxy-5-pyrimidin-2-ylami-
no)benzoic acid (14): The title compound was prepared
following general procedure A using 4-cyclohexyl-
[a] Yield [%] at 128 min.
methoxy-2-n-butylsulfonylpyrimidin-6-amine
(0.30g,
Caution! TFA and TFE should be handled with care. TFA causes
severe burns and TFE is irritating to the respiratory system and
skin and can seriously damage the eyes.
0.92mmol), 4-aminobenzoic acid (0.25g, 1.83mmol) and TFA
(0.36mL, 4.59mmol) in TFE (3mL). The system was cooled to RT
and concentrated under reduced pressure to yield an orange solid.
An aqueous 2m NaOH solution (20mL) was added to the residue
while stirring. After extraction with EtOAc (3ꢁ30mL), the aqueous
layer was acidified until pH<2 with a 4m HCl solution. A second
extraction with EtOAc (4ꢁ30mL) was performed on the aqueous
acidic phase. The organic fractions were combined, washed with
water (3ꢁ30mL) and dried over Na2SO4. After the solvent was
evaporated and the crude compound was recrystallized from
MeOH, the title compound was recovered as an off-white solid
(0.26g, 82%). Rf =0.27 (DCM/MeOH 95:5); m.p. 272–2758C; UV lmax
(EtOH): 302 nm; IR (neat powder): vmax =3514, 3404, 3105, 2932,
General procedure A
Typically, to a stirred suspension of the appropriate heteroaryl sub-
strate (1 equiv) and the aniline (2 equiv) in TFE ([heteroaryl sub-
strate]=0.1m) was added TFA (2.5 equiv) dropwise. The resulting
solution was heated at reflux for 24h under a nitrogen atmos-
phere. The solvent was removed in vacuo and the residue was re-
dissolved in EtOAc (10mL). The solution was washed with saturat-
ed sodium bicarbonate solution (3ꢁ10mL), and the aqueous ex-
tracts were combined and washed with EtOAc (10mL). The com-
bined organic layers were dried (MgSO4 or Na2SO4) and the solvent
was removed to give a residue that was purified as indicated for
each product. Four representative examples of this procedure are
given below with others to be found in Supporting Information.
2847, 1673, 1567, 1510 cmÀ1 1H NMR (300 MHz, [D6]DMSO): d=
;
0.95–1.29 (6H, m, H-cyclo), 1.68–1.79 (5H, m, H-cyclo), 4.01 (2H, d,
J=5.9 Hz, OCH2), 5.28 (1H, s, H-5), 6.43 (2H, brs, NH2), 7.78 (2H, d,
J=8.7 Hz, Ar-H), 7.88 (2H, d, J=8.7 Hz, Ar-H), 9.35 (1H, brs, NH),
12.45 ppm (1H, brs, OH); 13C NMR (75 MHz, [D6]DMSO): d=24.8,
25.7, 28.9, 36.6, 70.0, 78.4, 117.0, 129.5, 139.1, 145.2, 155.0, 158.6,
166.7, 169.6 ppm; MS (ES+) m/z 343.34 [M+H]+; HRMS calcd for
C18H23N4O3 [M+H]+ 343.1765, found 343.1770.
6-(2-(Triisopropylsilyl)ethynyl)-N-phenyl-9H-purin-2-amine
(4):
The title compound was synthesised according to general proce-
dure A using 2-fluoro-6-((triisopropylsilyl)ethynyl)-9H-purine (0.46g,
1.43mmol), aniline (260 mL, 2.86mmol) and TFA (551 mL, 7.15mmol)
in TFE (7mL). The crude material was purified by MPLC on silica
(DCM/MeOH 19:1) to give the desired product as a yellow oil/gum
(390mg, 1.00mmol, 70%). The title compound was also synthes-
ised following general procedure B for 30min using 2-fluoro-6-((trii-
sopropylsilyl) ethynyl)-9H-purine (600mg, 1.90mmol), aniline
(0.50mL, 3.80mmol) and TFA (0.7mL) in TFE (19mL). The crude
product was purified by MPLC on amine silica (DCM/MeOH 95:5)
8-Methyl-N-(4-phenoxyphenyl)-9H-purin-2-amine (28): The title
compound was synthesised following general procedure A using
2-fluoropurine intermediate (70mg, 0.46mmol) and 4-phenoxyani-
line (0.170g, 0.92mmol) were treated with TFA (177 mL, 2.30mmol)
in TFE (6mL). The crude product was purified by MPLC on silica
(DCM/MeOH 19:1) to afford a pale pink solid (0.101g, 75%). Rf =
0.39 (DCM/MeOH 19:1); m.p. 228–2318C; UV lmax (EtOH): 277 nm;
IR (neat powder): vmax =3454, 3253, 3194, 3044, 2924, 2161,
Chem. Eur. J. 2014, 20, 2311 – 2317
www.chemeurj.org
2316
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
1
1617 cmÀ1; H NMR (500 MHz, [D6]DMSO): d=2.46 (3H, s, CH3), 6.95
[9] C. Wong, R. J. Griffin, I. R. Hardcastle, J. S. Northen, L.-Z. Wang, B. T.
Golding, Org. Biomol. Chem. 2010, 8, 2457.
(2H, dd, J=1.1, 7.7 Hz, Ar-H), 7.00 (2H, d, J=9.0 Hz, Ar-H), 7.07
(1H, dddd, J=1.1, 1.1, 7.4, 7.4 Hz, Ar-H), 7.36 (2H, ddd, J=1.1, 7.4,
7.7 Hz, Ar-H), 7.83 (2H, d, J=9.0 Hz, Ar-H), 8.63 (1H, s, H-6), 9.45
(1H, s, NH), 12.68 ppm (1H, s, N9-H); 13C NMR (125 MHz,
[D6]DMSO): d=14.9, 117.1, 119.7, 119.8, 122.4, 128.3, 129.8, 137.5,
146.4, 149.3, 151.4, 154.3, 155.9, 158.1 ppm; HRMS calcd for
C18H16N5O [M+H]+ 318.1349, found 318.1352.
[10] C. K. Banks, J. Am. Chem. Soc. 1944, 66, 1127.
[11] C. K. Banks, J. Am. Chem. Soc. 1944, 66, 1131.
[12] N. B. Chapman, C. W. Rees, J. Chem. Soc. 1954, 1190.
[13] N. B. Chapman, D. Q. Russell-Hill, J. Chem. Soc. 1956, 1563.
[14] J. Liu, M. J. Robins, J. Am. Chem. Soc. 2007, 129, 5962.
[15] J. R. Schmink, C. M. Kormos, W. G. Devine, N. E. Leadbeater, Org. Process
Res. Dev. 2010, 14, 205.
[16] C. K. Ingold, Structure and Mechanism in Organic Chemistry, G. Bell,
London, 1953.
[17] M. F. Pisani, C. R. Hebd. Seances Acad. Sci. 1854, 34, 852.
[18] J. Miller, A. J. Parker, J. Am. Chem. Soc. 1961, 83, 117.
[19] J. F. Bunnet, Bull. Hist. Chem. 1996, 21, 33.
General procedure B
As described in general procedure A, except that reagents were
placed in a sealed microwave vial and the reactions were per-
formed under microwave irradiation (1408C for 90min). A repre-
sentative example of this procedure is given above with others to
be found in Supporting Information.
[20] J. F. Bunnett, Q. Rev. Chem. Soc. 1958, 12, 1.
[21] J. F. Burnett, R. E. Zahler, Chem. Rev. 1951, 49, 273.
[22] J. Meisenheimer, Liebigs Ann. Chem. 1902, 323, 205.
[23] F. Terrier, Chem. Rev. 1982, 82, 77.
[24] J. F. Hartwig, Angew. Chem. 1998, 110, 2154–2177; Angew. Chem. Int. Ed.
1998, 37, 2046–2067.
[25] J. P. Wolfe, H. Tomori, J. P. Sadighi, J. Yin, S. L. Buchwald, J. Org. Chem.
2000, 65, 1158.
[26] D. S. Surry, S. L. Buchwald, Angew. Chem. 2008, 120, 6438–6461; Angew.
Chem. Int. Ed. 2008, 47, 6338–6361.
[27] I. P. Beletskaya, A. V. Cheprakov, Organometallics 2012, 31, 7753.
[28] J. F. Hartwig, Acc. Chem. Res. 2008, 41, 1534.
[29] O. R. Barbeau, C. Cano-Soumillac, R. J. Griffin, I. R. Hardcastle, G. C. M.
Smith, C. Richardson, W. Clegg, R. W. Harrington, B. T. Golding, Org.
Biomol. Chem. 2007, 5, 2670.
[30] C. O. Kappe, Angew. Chem. 2004, 116, 6408–6443; Angew. Chem. Int. Ed.
2004, 43, 6250–6284.
[31] N. A. Senger, B. Bo, Q. Cheng, J. R. Keeffe, S. Gronert, W. Wu, J. Org.
Chem. 2012, 77, 9535.
Product formation as a function of TFA concentration
In a sealed tube equilibrated at 258C, containing a 0.25m solution
of the heteroaryl substrate (1 equiv) in TFE, was added the aniline
(2 equiv). Timing started upon introduction of TFA and aliquots
were taken at defined intervals (0.5, 1, 2, 4, 8, 16, 32, 64, 128, 256,
512 min) to follow the formation of the product and the disappear-
ance of the heteroaryl substrate. All aliquots were quenched with
a 1:1 (v/v) solution of saturated sodium bicarbonate and ethyl ace-
tate. The organic extracts were analysed by LC-MS to monitor the
percentage of product formed (equivalent to the percentage of re-
acted starting material) by integration of their respective peaks.
[32] W. P. Jencks, J. Am. Chem. Soc. 1959, 81, 475.
[33] M. M. Toteva, J. P. Richard, J. Am. Chem. Soc. 2002, 124, 9798.
[34] B. G. Cox, Acids and Bases: Solvent Effects on Acid-Base Strength, Oxford
University Press, Oxford, 2013.
[35] L. Forlani, V. Tortelli, J. Chem. Res. Synop. 1982, 62.
[36] L. Forlani, J. Chem. Res. Synop. 1984, 260.
[37] V. Markowski, G. R. Sullivan, J. D. Roberts, J. Am. Chem. Soc. 1977, 99,
714.
[38] J. Liu, J. R. Barrio, N. Satyamurthy, J. Fluorine Chem. 2006, 127, 1175.
[39] N. S. Isaacs, Physical Organic Chemistry, 2nd ed., Longman Scientific &
Technical, 1995.
Acknowledgements
The synthetic studies were part of medicinal chemistry re-
search within the Northern Institute for Cancer Research and
were financially supported by Cancer Research UK. The authors
thank the EPSRC National Mass Spectrometry Service at the
University of Wales (Swansea) for mass spectrometry analyses,
and Dr. K. Haggerty for HPLC and LC-MS analyses.
[40] M. L. Sinnott, W. P. Jencks, J. Am. Chem. Soc. 1980, 102, 2026.
[41] L. C. Chan, B. G. Cox, I. C. Jones, S. Tomasi, J. Phys. Org. Chem. 2011, 24,
751.
[42] E. A. Castro, M. Gazitffla, J. G. Santos, J. Org. Chem. 2005, 70, 8088.
[43] R. Kaliszan, P. Wiczling, M. J. Markuszewski, Anal. Chem. 2004, 76, 749.
[44] J. M. Vandenbelt, C. Henrich, S. G. Vanden Berg, Anal. Chem. 1954, 26,
726.
Keywords: catalysis · heterocyclic drugs · SNAr · synthetic
methods · trifluoroethanol
[1] A. M. Aronov, B. McClain, C. S. Moody, M. A. Murcko, J. Med. Chem.
2008, 51, 1214.
[2] S. D. Roughley, A. M. Jordan, J. Med. Chem. 2011, 54, 3451.
[3] D. S. Surry, S. L. Buchwald, Chem. Sci. 2011, 2, 27.
[45] G. N. Okafo, R. Brown, P. Camilleri, J. Chem. Soc. Chem. Commun. 1991,
864.
[46] H. Lebraud, C. Cano, B. Carbain, I. R. Hardcastle, R. W. Harrington, R. J.
Griffin, B. T. Golding, Org. Biomol. Chem. 2013, 11, 1874.
[47] T. J. Blackburn, S. Ahmed, C. R. Coxon, J. Liu, X. Lu, B. T. Golding, R. J.
Griffin, C. Hutton, D. R. Newell, S. Ojo, A. F. Watson, A. Zaytzev, Y. Zhao,
J. Lunec, I. R. Hardcastle, Med. Chem. Commun. 2013, 4, 1297.
[48] J.-P. BØguØ, D. L. Bonnet-Delpon, B. Crousse, Synlett 2004, 18.
[49] P. Buonora, J.-C. Olsen, T. Oh, Tetrahedron 2001, 57, 6099.
[50] M. Tajbakhsh, R. Hosseinzadeh, H. Alinezhad, S. Ghahari, A. Heydari, S.
Khaksar, Synthesis 2011, 490.
´
[4] L. Ciszewski, L. Waykole, M. Prashad, O. Repic, Org. Process Res. Dev.
2006, 10, 799.
[5] F. Marchetti, C. Cano, N. J. Curtin, B. T. Golding, R. J. Griffin, K. Haggerty,
D. R. Newell, R. J. Parsons, S. L. Payne, L. Z. Wang, I. R. Hardcastle, Org.
Biomol. Chem. 2010, 8, 2397.
[6] T. G. Davies, J. Bentley, C. E. Arris, F. T. Boyle, N. J. Curtin, J. A. Endicott,
A. E. Gibson, B. T. Golding, R. J. Griffin, I. R. Hardcastle, P. Jewsbury, L. N.
Johnson, V. Mesguiche, D. R. Newell, M. E. M. Noble, J. A. Tucker, L.
Wang, H. J. Whitfield, Nat. Struct. Biol. 2002, 9, 745.
[7] H. J. Whitfield, R. J. Griffin, I. R. Hardcastle, A. Henderson, J. Meneyrol, V.
Mesguiche, K. L. Sayle, B. T. Golding, Chem. Commun. 2003, 0, 2802.
[8] F. Marchetti, K. L. Sayle, J. Bentley, W. Clegg, N. J. Curtin, J. A. Endicott,
B. T. Golding, R. J. Griffin, K. Haggerty, R. W. Harrington, V. Mesguiche,
D. R. Newell, M. E. M. Noble, R. J. Parsons, D. J. Pratt, L. Z. Wang, I. R.
Hardcastle, Org. Biomol. Chem. 2007, 5, 1577.
[51] E. N. Jacobsen, Acc. Chem. Res. 2000, 33, 421.
[52] A. Berkessel, J. A. Adrio, D. Hüttenhain, J. M. Neudçrfl, J. Am. Chem. Soc.
2006, 128, 8421.
Received: November 5, 2013
Published online on January 23, 2014
Chem. Eur. J. 2014, 20, 2311 – 2317
www.chemeurj.org
2317
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim