Applicability aspects of transition metal-catalyzed aromatic amination protocols in medicinal chemistry

Article abstract of DOI:10.1002/adsc.200700133

The application of palladium- and coppercatalyzed reactions for the aromatic amination of pharmacologically relevant scaffolds is investigated. The focus is set on the scope of several protocols for the introduction of amines of broad structural diversity, allowing for the synthesis of numerous derivatives of one biological hit structure for screening in biological assay systems. Thus, attaining optimized yields and TONs had not a major priority, most important were practical aspects, that is no further purification and drying of reagents and solvents had to be envisaged, ideally only a few transition metalbased protocols had to be applied for synthesizing structurally diverse compounds in sufficient amounts (several milligrams) for screening without any finetuning of conditions and catalytic systems.

Full text of DOI:10.1002/adsc.200700133

FULL PAPERS
DOI: 10.1002/adsc.200700133
Applicability Aspects of Transition Metal-Catalyzed Aromatic
Amination Protocols in Medicinal Chemistry
Stefan Tasler,a,* Jan Mies,a and Martin Langa
a
4SC AG, Am Klopferspitz 19a, 82152 Planegg-Martinsried, Germany
Fax : (+49)-89-700-76329; stefan.tasler@4sc.com
Received: March 13, 2007; Revised: July 11, 2007
Supporting information for this article is available on the WWW under http://asc.wiley-vch.de/home/.
Abstract: The application of palladium- and copper- rification and drying of reagents and solvents had to
catalyzed reactions for the aromatic amination of be envisaged, ideally only a few transition metal-
pharmacologically relevant scaffolds is investigated. based protocols had to be applied for synthesizing
The focus is set on the scope of several protocols for structurally diverse compounds in sufficient amounts
the introduction of amines of broad structural diver- (several milligrams) for screening without any fine-
sity, allowing for the synthesis of numerous deriva- tuning of conditions and catalytic systems.
tives of one biological hit structure for screening in
biological assay systems. Thus, attaining optimized
yields and TONs had not a major priority, most im- Keywords: amination; copper; cross-coupling; homo-
portant were practical aspects, that is no further pu- geneous catalysis; medicinal chemistry; palladium
Introduction
As amine-substituted aromatics and heteroaromat-
ics represent desirable frameworks in medicinal
Starting with almost simultaneous landmark publica- chemistry, aromatic amination protocols had to sur-
tions from Hartwigꢀs and Buchwaldꢀs laboratories,[1] vive scrutiny under “non-optimal” reaction conditions
numerous articles on transition metal-mediated aro- for transition metal catalysis as usually encountered
matic amination reactions of aryl halides were pub- in R&D laboratories to prove their general applicabil-
lished during the past decade by several working ity. Generally that is, solvents and reagents are used
groups on almost every aspect of this reaction type.[2,3] as bought without further drying or purification, and
Major focus was laid primarily on Pd-catalyzed reac- weighing procedures are performed in normal atmos-
tions. The use of catalytic copper(I) species, however, phere. Furthermore, substrates are usually substituted
accrued relevance due to systematic substrate evalua- in a more complex way than those carefully chosen as
tion with a 4-year-onset,[4] again triggered by work of test substrates for methodical evaluations of a tech-
Buchwaldꢀs group.[5] In the meantime, the synthetic nique. As time is an important factor in pharmaceuti-
chemist has the choice of several palladium- and cal research, extensive investigations into subtle
copper-based amination protocols which can display changes of conditions (different ligand derivatives, in-
complementary advantages for different conversions organic bases, Pd sources within one given method,
(i.e., switching chemoselectivity),[6–8] and even a few concentration, etc.) for obtaining optimized yields are
procedures are now available using nickel species.[9] usually not feasible. Thus, a most general and simple
Most of the procedures elaborated had in common protocol was required, with a broad application for
that research was performed under ideal conditions the quick synthesis of small libraries with a defined
for transition metal catalysis: standard set-ups for structural diversity. For reasons of simplicity and time
drying of solvents were used, substrates and agents saving, certain protocols including pre-mixing of re-
were assembled in a dry-box or at least the bulk of agents or a certain order of assembling of all ingredi-
the base was stored under an inert atmosphere, start- ents were not followed. In most cases, slightly more
ing amines were distilled or purified by passing over a transition metal catalyst was applied than published
column of alumina, or – in several cases – ligands or within the original protocols, since with no further pu-
catalysts were freshly prepared to guarantee a certain rification of solvents and amines, turnover numbers
quality.
(TONs) were expected to be lower as compared to
2286
ꢁ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2007, 349, 2286 – 2300

Transition Metal-Catalyzed Aromatic Amination Protocols in Medicinal Chemistry
FULL PAPERS
the literature. Furthermore, it was not the major focus
to decrease Pd quantities to establish the high catalyt-
ic potential of a method, but to obtain significant
amounts of product for biological screening without
the necessity of optimizing yields.[10,11]
The advantages of transition metal-catalyzed aro-
matic amination reactions over other chemical ways
of amine introduction are supposed to be manifold:
a) the broad methodical work published offers a
procedure for the coupling of almost any amine, that
is, anilines, aliphatic amines, amides, N-heterocycles,
carbamates, etc.;
b) by relying on differently active leaving groups
(e.g., Cl vs. Br) at the aromatic reaction partner, a
stepwise introduction of two different substituents can
be achieved with clearly defined regioselectivity;
c) several different electronic conditions are toler-
ated within the aromatic ring, which is favorable over
amine introduction by SNAr-type reactions, which usu-
ally call for nitro substitution or at least substitution
with strongly electron-withdrawing groups located in
ortho or para position to the leaving group, or a nitro-
gen within a heterocyclic framework.
Figure 1. Selection of efficient ligands for aromatic amina-
tion reactions.
upon amination of all of the bromine sites, with gen-
erally an excess of amine being present to avoid
double arylation of primary amines. Thus, catalytic
systems were chosen which were known to leave
chloroarenes unaltered: transformations were con-
ducted using DPPF (1, method A) and BINAP (2,
method B) as chelating ligands, according to Hart-
wigꢀs and Buchwaldꢀs early procedures for the amina-
Results and Discussion
Two-Fold Aromatic Amination of 2,5-Dihalogenated
Arenes
Starting with the synthesis of a new series of seroto- tion of bromoarenes.[13,16]
nin receptor ligands,[12] a 2,4- and 2,5-diamination pat-
Despite the fact that nitroarenes are usually classi-
tern of nitroarenes was envisaged to be installed by fied as base-sensitive substrates, rarely tolerating
two-fold aromatic amination. Usually, 2,4-diaminated strong bases like sodium alkoxides in amination reac-
nitroarenes have been realized by subsequent SNAr tions,[2,3,13,17] the DPPF-based procedure resulted in
reactions on the 2,4-dihalogenated analogues (for a 25% isolated yield. On the other hand, with BINAP
discussion, see Supporting Information). However, (2) as ligand, Cs2CO3 had better been chosen accord-
with a focus on scaffold variations including the re- ing to the recommendation in the literature,[13] and –
placement of the nitro functionality of these sub- as cited – only decomposition of starting material was
strates by other polar “head groups”, an SNAr ap- observed with NaO-t-Pent.[18] In general, sodium tert-
proach was not feasible anymore – especially not for pentoxide was used throughout our work rather than
a 2,5-substitution pattern, with the second leaving sodium tert-butoxide, as tert-pentoxides were men-
group being located in an unfavorable position for tioned by Nolan and co-workers to be somewhat su-
SNAr chemistry.
perior to the corresponding tert-butoxides under cer-
A first method evaluation was performed on 1- tain amination conditions, most likely due to their en-
bromo-4-chloro-2-nitrobenzene (Table 1), amination hanced solubility in organic solvents.[19] Furthermore,
of which with benzylamine (entry 1) should occur on sodium tert-pentoxide was expected to be less hygro-
the bromine site with excellent selectivity over the scopic than the corresponding butoxide,[20] which, of
chlorine site.[13] Even though protocols utilizing di- course, would result in a significant advantage over
alkylphosphino-biphenyl ligands (like 3, Figure 1) time as all bases were stored on the shelf without fur-
seemed to be among the most active Pd/ligand sys- ther precautions.
tems for aminations of bromo- and chloroarenes,[14]
For additional evaluation of the most suitable pro-
with good chemoselectivities having been report- cedures, protocols were applied for the cross-coupling
ed,[10,15] they were not applied here for substrates of benzylamine with 1-bromo-4-chloro-2-nitrobenzene
bearing both leaving groups simultaneously. Given (entry 1), in which water was employed as (co-)sol-
the excellent TONs of this catalytic system as were vent (methods C1 and D1), thus eliminating any pos-
reported, amination reactions might not have stopped sible disadvantages due to residual moisture in sol-
Adv. Synth. Catal. 2007, 349, 2286 – 2300
ꢁ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2287

Stefan Tasler et al.
FULL PAPERS
Table 1. Two-fold amination of 2,5-dihalogenated nitroarenes.
[a]
A: Ar-Br/amine 1.0:1.1; 0.05equivs. (DPPF)PdCl
,
0.15equivs. DPPF
(
1), 1.25equivs. NaO- t-Pent, THF
2
(2 mLmmolÀ1 Ar-Br), 1008C, 4 h; B1: Ar-Br/amine 1.0:1.1; 0.01 equivs. Pd2dba3, 0.03 equivs. BINAP (2), 1.4 equivs.
NaO-t-Pent, toluene (2 mLmmolÀ1 Ar-Br), 808C, 20 h; C1: Ar-Br/amine 1.0:1.1; 0.01 equivs. Pd2dba3, 0.04 equivs.
(biph)P
(OAc)2, 0.20 equivs. (biph)P
1.0:1.1; 0.03 equivs. Pd2dba3, 0.12 equivs. (biph)P
20 h; C4: Ar-Cl/amine 1.0:1.3; 0.05equivs. Pd 2dba3, 0.20 equivs. (biph)P
(2 mLmmolÀ1 Ar-Cl), 1108C, 20 h; D1: Ar-Br/amine 1.0:1.05; 0.01 equivs. Pd2dba3, 0.04 equivs. P
equivs. (cetyl)NMe3Br, 1.5equivs. KOH in H O (45%), toluene (1.2 mLmmolÀ1 Ar-Br), 908C, 20 h; D2: Ar-Br/amine
ACHTREUNG
2
A
AHCTREUNG
AHCTREUNG
AHCTREUNG
2
1.0:1.05; 0.01 equiv. Pd2dba3, 0.016 equivs. P
ACHTREUNG
AHCTREUNG
(1 mLmmolÀ1 Ar-Br), 1108C, 20 h; F1: Ar-Br/amine 1.0:1.1; 0.03 equivs. Pd
ACHTRE(UGN OAc)2, 0.045equivs. Xantphos ( 4), 1.4
equivs. Cs2CO3, dioxane (1 mLmmolÀ1 Ar-Br), 1108C, 20 h.
Starting bromoarene detected (LC-MS), some decomposition occurred.
Traces of product and starting bromoarene detected (LC-MS), decomposition occurred.
Starting bromoarene detected/recovered.
Only formation of side product 6 observed (LC-MS).
Traces of product found, starting bromoarene recovered, some decomposition occurred, side product 4-chloro-2-nitro-1-
[b]
[c]
[d]
[e]
[f]
phenoxybenzene isolated.
[g]
[h]
Starting bromoarene recovered, biaryl 7a isolated, halogen exchange (I for Br) observed.[24]
Isolated yield for methyl ester; tert-pentyl ester formation (transesterification) detected by LC-MS (Me vs. t-Pent ca. 3:1,
not standardized).
Isolated yields, yields are not optimized; DME=1,2-dimethoxyethane, monoglyme.
vents used as bought. In this case, the advantages conducted as a modification of the original Buchwald
gained by the choice of solvent justified risking side protocol,[6] in which a specialized, highly sterically en-
reactions by over-amination as the Pd/ligand systems cumbered dicyclohexylphosphino-biphenyl ligand 3c
utilized usually qualify for effective oxidative addition had been used (as compared to 3a in our case), which
to the chloro-arene bond as well. Interestingly, the usually necessitated a certain fine-tuning of conditions
best result was achieved applying method C1 (43%), for different substrate subtypes. For our purposes
2288
asc.wiley-vch.de
ꢁ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2007, 349, 2286 – 2300

Transition Metal-Catalyzed Aromatic Amination Protocols in Medicinal Chemistry
FULL PAPERS
(biph)PACHTREUNG(t-Bu)2 3a was chosen as standard representa-
tive of the class of monodentate biphenyl P-ligands
through all but one of our respective reactions. This
ligand had been introduced by Buchwald within a
protocol claimed to be among the most benign amina-
tion procedures with respect to generality and scope
for amination reactions of chloro- and bromoar-
enes.[14]
For an application of procedures based on PCAHTRE(UNG t-Bu)3,
a highly active but air-sensitive (pyrophoric) ligand,
the corresponding air-insensitive HBF4 salt was
chosen, enabling handling under normal atmos-
phere.[21] Utilization of KOH in a solvent mixture tol-
uene/water under phase-transfer conditions (method
D1) was supposed to be more compatible with base-
sensitive functional groups as compared to the appli-
cation of sodium alkoxides according to the literature
(cf. also yield of method A vs. method C1).[22] For an
attempted room temperature amination based on the
same catalytic system, NaO-t-Pent had to be used in
toluene (method D2), as weaker inorganic bases
(K3PO4 or Cs2CO3) displaying a better compatibility
with nitroarenes would have called for elevated reac-
tion temperatures.[17,23] However, in both cases (meth-
ods D1 and D2) problems were reported using pri-
Figure 2. Isolated side products from amination reactions of
nitroarenes.
mary amines other than anilines, and the reactions non-catalytic Ullmann coupling is indicative of a reac-
(entry 1) proceeded accordingly. tion rate being too slow for the amination process, so
Despite the best coupling result of 43% for the that biaryl formation became competitive and used
amination entry 1 using method C1, the DPPF-based up all the catalytic copper species. Racemic trans-
protocol A was chosen for further aminations of bro- N,N’-dimethyl-1,2-cyclohexanediamine
(rac-5,
moarenes although the yield was inferior in this case. Figure 1) was chosen as standard ligand for copper-
The latter method proved to be a more general proce- based aminations throughout this work according to
dure, however, accompanied by acceptable yields for its advantageous reactivity profile as compared to nu-
a more diverse subset of aryl bromide substrates (not merous other diamine ligands elaborated by Buch-
shown).[18] Furthermore, this protocol employs the waldꢀs group.[8,25]
least cost expensive P-ligand among those depicted in
Figure 1, resulting in highest reaction rates (max. con- of primary and secondary tert-butyl carbamates to
version reached usually upon 3–4 h reaction time) aryl bromides, P(t-Bu)3 and sodium phenolate were
with complete chemoselectivity.
chosen as reagents,[17,26] the ligand again being used as
The second amination step on the benzylamino sub- its HBF4 salt (method D3). The use of (biph)P(t-Bu)2
Following a Hartwig protocol for the introduction
AHCTREUNG
ACHTREUNG
strate attained (entry 1), that is an exchange of chlor- 3a in this conversion (method C3) was in accordance
ine with N-methylpiperazine, was attempted with the to an effective amination of several chloroarenes
ligand (biph)PACHTREUNG(t-Bu)2 3a (method C2) – but only to (also nitro-substituted) with oxazolidinones, although
find side product formation, according to LC-MS it was stated that reactions proceeded quite sluggishly
maybe a self-dechloro-dimerization to give product 6 if substrates were equipped with an ortho substitu-
(Figure 2), with the benzylic NH of the substrate out- ent.[27] Eventually, a protocol based on Xantphos (4,
performing piperazine as nucleophile upon deproto- Figure 1), a ligand highly effective for amination reac-
nation. For a completion of this sequence, a Boc pro- tions of bromoarenes with weak N-nucleophiles,[28–30]
tection of the benzylamine unit was scheduled, which proved to be most effective (method F1, 66% isolated
was not readily achieved on substrate of entry 1 upon yield). Standard amination at the chlorine site in a
first amination due to the electron-withdrawing prop- second amination step now proceeded smoothly with
erties of the ortho nitro group. Therefore, a direct in- K3PO4 as base (entry 2, method C2, 82%).
corporation of N-Boc-benzylamine was attempted
Another way to circumvent the side product forma-
using Pd chemistry as well as a copper-based protocol tion as encountered in entry 1, second step, was the
optimized for the introduction of mainly amides to exchange of positions of bromine and chlorine within
bromoarenes (method E1).[25] Within the latter the aromatic substrate and hence an amination with a
method, the detection of biaryl 7a resulting from a secondary amine prior to the introduction of a pri-
Adv. Synth. Catal. 2007, 349, 2286 – 2300
ꢁ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2289

Stefan Tasler et al.
FULL PAPERS
mary amine. This approach was realized for a methyl group at the chlorine site – however, was shown later
benzoate (entry 3), and was also necessitated for an to give superior results using NaO-t-Pent as compared
anisole substrate (entry 4) since the corresponding 2- to K3PO4 (cf. Table 4, entries 18 and 19) in spite of
bromo-5-chloro-derivative was not commercially the potential for decomposition and/or transesterifica-
available for this head group. Conditions chosen for tion, with a stronger base being required to allow for
the two-fold amination sequence seemed to be superi- an active catalytic cycle.
or for the electron-rich substrate (entry 4 vs. entry 3),
which might simply be explained by a poor substrate
compatibility of the ester with NaO-t-Pent, though. Two-Fold Amination of 2,4-Dihalogenated
For the first amination of methyl benzoate (entry 3) Nitroarenes
at the bromine site, this argument seemed to be a
valid explanation for the moderate yield obtained, as Within the 2,4-dihalogenated substrate series, the 4-
a similar conversion of a structural isomer of this substituent was usually chosen to become N-methylpi-
ester (Table 4, entry 18) could be improved signifi- perazine, so that a most convergent synthetic strategy
cantly when switching to a BINAP (2)/Cs2CO3 system would start with installing this substituent first, thus
(vide infra). The second amination using (biph)PACHTRE(UGN t- requiring the bromine in said position. Several struc-
Bu)2 3a – that is, in an ortho position to the head turally diverse products obtained by such a two-fold
Table 2. Two-fold amination of 2,4-dihalogenated nitroarenes.
[a]
A: Ar-Br/amine 1.0:1.1; 0.05equivs. (DPPF)PdCl
(2 mLmmolÀ1 Ar-Br), 1008C, 4 h; C2: Ar-Cl/amine 1.0:1.3; 0.10 equivs. Pd
equivs. K3PO4, DME (2 mLmmolÀ1 Ar-Cl), 1008C, 20 h; C4: Ar-Cl/amine 1.0:1.3; 0.05equivs. Pd 2dba3, 0.20 equivs.
(biph)P
(t-Bu)2 3a, 1.4 equivs. NaO-t-Pent, toluene (2 mLmmolÀ1 Ar-Cl), 1108C, 20 h.
Formation of aryl (tert-pentyl) ether observed (<3%), cf. also ref.[14]
,
0.15equivs. DPPF
(
1), 1.25equivs. NaO- t-Pent, THF
2
A
ACHTREUNG
ACHTREUNG
[b]
[c]
[d]
[e]
[f]
Ar-Cl/amine 1.0:1.1; 0.03 equivs. Pd
ACHRTUNEG(OAc)2, 0.06 equivs. (biph)PACHRTE(UGN t-Bu)2 3a.
Major amounts of starting chloroarene unaltered (LC-MS).
35% isolated side product of Heck reaction, starting chloroarene recovered.
15% isolated side product 9.
Isolated yields, yields are not optimized.
2290
asc.wiley-vch.de
ꢁ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2007, 349, 2286 – 2300

Transition Metal-Catalyzed Aromatic Amination Protocols in Medicinal Chemistry
FULL PAPERS
amination sequence are depicted in Table 2, following Therefore, the latter sequence was conducted likewise
the final procedure selection made for the 2-bromo-5- with the 2,4-disubstituted nitroarenes to give quite
chloro series (Table 1, entry 1). The first amination good yields (entries 11 and 12). Some structural varie-
step was accomplished with different piperazines in ty was attained with the introduction of N-heterocy-
mediocre yields (24–38%, DPPF-based protocol, cles, even though yields were quite poor. In these
NaO-t-Pent as base, method A). (Biph)P
procedures (C2 and C4) allowed for the introduction (method E2)[8] was superior to palladium chemistry
of benzylamines (Table 2, entries 5, 9 and 10), N-ami- using P
(t-Bu)3 as ligand,[17] again as its HBF4 salt
ACHTRE(UNG t-Bu)2-based cases (entries 13 and 14), a Cu-based protocol
AHCTREUNG
nopyrroles (entry 6) and 2-substituted ethylamines (method D4). With pyrroles as N-nucleophiles and Pd
(entries 7 and 8) in the 2-position. In general, amina- catalysis, Heck-type arylation at a carbon atom of the
tion reactions of 2-chloronitroarenes proceeded quite pyrrole became dominant (entry 13), which is in ac-
sluggishly, either due to only very low TONs achieved cordance – but still with surprising preference in our
with K3PO4 as base with the major amount of starting case – with literature data for corresponding aryla-
chloroarene still being present unaltered (entries 5 tions of indoles with ortho substituted aryl bro-
and 8), or due to decomposition encountered with the mides.[8,17] Such a side reaction[33] could be excluded
stronger base NaO-t-Pent[13,17] (entries 7–10) – which when 3,5-dimethylpyrazole was used (entry 14), but
is in good agreement with literature data.[14]
yields from the Pd-catalyzed conversion were still
Amination with N-aminopyrrole as an aniline ana- poor (7%). Somewhat enhanced yields were obtained
logue succeeded with quite an excellent result for the N-arylation of pyrroles and pyrazoles using
(entry 6, 69%), not even encountering Heck-type side copper chemistry, standard protocols of which were
reactions as with the substrate of entry 7, which could originally established for aryl iodides as substrates,[7]
be subdued in the latter case by switching from and have been adapted here according to further
K3PO4 (method C2) to the stronger base NaO-t-Pent progress achieved for the N-arylation of indoles.[8,34]
(C4). With N-(2-aminoethyl)pyrrole (entry 7), utiliza- For Cu-based reactions of entries 13 and 14, desired
tion of a stronger base most likely enhanced or accel- conversions were hampered by significant side prod-
erated the formation of a palladium-amido complex uct formation (biaryl formation, halogen exchange).
À
within the catalytic cycle, thus rendering C N bond
A second amination step with N-methylpiperazine
À
formation competitive to Heck-type C C coupling. A using our standard Pd protocols C2 or C4, was only
vice versa phenomenon was observed with 2-phenoxy- successful with the latter procedure to give 7% and
ethylamine as substrate (entry 8), with which a major 30% for the pyrrole and pyrazole derivatives, respec-
side reaction occurred when NaO-t-Pent was tively (entries 13 and 14). For benzamides (entries 15
utilized, giving rise to diaryl ether 9. For the forma- and 16) and pyridazinone (entry 17), Pd in combina-
tion of the latter, a b-elimination in species 8 was tion with Xantphos (4) proved to be the method of
postulated (Figure 2), resulting in an intermediary choice (method F1).[28] Reaction of the latter resulted
(aryl)Pd(phenoxide) species. This side reaction indeed in an incomplete conversion and a poor yield, though,
became less competitive when the weaker base K3PO4 with K3PO4 being slightly favorable over Cs2CO3 as
was applied, and a moderate yield of 26% was ach- base (entry 17).[35] Final amination with N-methylpi-
ieved without any formation of diaryl ether 9.
perazine proceeded smoothly in all three cases with
As there are only a few examples of amination re- decent yields ranging from 48 to 76% (entries 15–17
actions on chloroarenes with weak N-nucleophiles and also 11 and 12).
like amides, carbamates or N-heterocycles using Pd or
Despite the fact that, in some instances, a copper-
Cu catalysis,[17,25,27,32] a less favorable amination se- based procedure represented the only way to attain
quence had to be chosen for the introduction of these significant amounts of product (entries 13 and 14),
amines in which divergence was already introduced yields were generally unsatisfactory. In all Cu-cata-
within the first amination step at the 2-bromine site lyzed amination attempts on 2-bromonitroarenes
followed by introduction of N-methylpiperazine at (Table 1, entry 2; Table 3, entries 13–15), a bromine-
the chlorine site (Table 3). For pyridazinone, the ne- iodine exchange[24] and a formation of biaryl 7 was
cessity of such a change in strategy was confirmed by observed, the latter resulting from a competing non-
its failed introduction into a 2-chloro-4-(N-methylpi- catalytic Ullmann coupling, which would result in loss
perazin-1-yl)-nitrobenzene using ligand 3a (method of catalytically active copper in the appropiate oxida-
C4).
Earlier results had shown a final introduction of
tion state.
Given the occurring side reactions with 2-bromoni-
benzylamines in the 2-position with chlorine as leav- troarenes in these conversions and the fact that steri-
ing group to give low yields (<10%, Table 2, en- cally demanding pyrazoles, like 3,5-dimethylpyrazole,
tries 5, 9 and 10), but – in contrast – an attachment of were stated to be only viable substrates for N-aryla-
Boc-protected benzylamine at the 2-bromine position tion with unhindered aryl iodides used neat as sol-
to be an excellent alternative (Table 1, entry 2). vent[34] – which, however, would not have been a fea-
Adv. Synth. Catal. 2007, 349, 2286 – 2300
ꢁ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2291

Stefan Tasler et al.
FULL PAPERS
Table 3. Two-fold amination of 2,4-dihalogenated nitroarenes.
[a]
(t-Bu)2 3a, 1.4 equivs. K3PO4, DME (2 mLmmolÀ1
C2: Ar-Cl/amine 1.0:1.3; 0.10 equivs. PdACHTERNGU(OAc)2, 0.20 equivs. (biph)PACHTREUGN
Ar-Cl), 1008C, 20 h; C4: Ar-Cl/amine 1.0:1.3; 0.05equivs. Pd dba3, 0.20 equivs. (biph)P
G
2
toluene (2 mLmmolÀ1 Ar-Cl), 1108C, 20 h; D4: Ar-Br/amine 1.0:1.2; 0.04 equivs. Pd2dba3, 0.08 equivs. P
equivs. Cs2CO3, toluene (3 mLmmolÀ1 Ar-Br), 1008C, 20 h; E1: Ar-Br/amine 1.0:1.1; 0.07 equivs. CuI, 0.14 equivs. rac-5,
2.0 equivs. K2CO3, toluene (1 mLmmolÀ1 Ar-Br), 1108C, 20 h; E2: Ar-Br/amine 1.0:1.2; 0.10 equiv. CuI, 0.30 equivs. rac-
5, 2.1 equivs. K3PO4, toluene (1 mLmmolÀ1 Ar-Br), 1108C, 20 h; F1: Ar-Br/amine 1.0:1.1; 0.03 equivs. Pd
ACHTRE(UGN OAc)2, 0.045
equivs. Xantphos (4), 1.4 equivs. Cs2CO3, dioxane (1 mLmmolÀ1 Ar-Br), 1108C, 20 h; F2: Ar-Br/amine 1.0:1.1; 0.02
equivs. Pd2dba3, 0.06 equivs. Xantphos (4), 1.4 equivs. K3PO4, dioxane (1 mLmmolÀ1 Ar-Br), 1008C, 20 h.
Side products of Heck reaction isolated, starting bromoarene recovered.
No conversion, only starting chloroarene detectable.
[b]
[c]
[d]
[e]
[f]
Starting bromoarene recovered, biaryl 7b isolated, halogen exchange (I for Br) observed.[24]
No starting chloroarene detectable.
Starting bromoarene recovered, biaryl 7b and biaryl of reductive coupling on Cl site of nitroarene isolated, minor quanti-
ties of 5 N-arylated (detected by LC-MS), halogen exchange (I for Br) observed.[24]
Reaction conducted at 1008C.
LC-MS indicated complete conversion for both reactions, differences in yields due to work-up.
Starting bromoarene detected (LC-MS).
[g]
[h]
[i]
[k]
NMR indicated side product of O-arylation of pyridazinone.
Isolated yields, yields are not optimized.
sible alternative with 2-bromo-4-chloro-1-nitroben- Two-Fold Amination of 2,4-Dihalogenated Non-
zene – an 18% yield in entry 14 seems to be quite ex- Nitroarenes
cellent. In this case, the observed halogen exchange
might have been an important factor favoring product Upon exchange of the nitro head group for other
formation.[36]
electron-withdrawing functionalities, like carboxylic
2292
asc.wiley-vch.de
ꢁ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2007, 349, 2286 – 2300

Transition Metal-Catalyzed Aromatic Amination Protocols in Medicinal Chemistry
FULL PAPERS
Table 4. Two-fold amination of 2,4-dihalogenated benzoic esters and benzonitriles.
[a]
A: Ar-Br/amine 1.0:1.1; 0.05equivs. (DPPF)PdCl
,
2
0.15equivs. DPPF
(
1), 1.25equivs. NaO- t-Pent, THF
(2 mLmmolÀ1 Ar-Br), 1008C, 4 h; B2: Ar-Br/amine 1.0:1.1; 0.03 equivs. Pd
AHCTREUNG
AHCTREUNG
ACHTREUNG
2
(t-Bu)2 3a, 1.4 equivs. NaO-t-Pent, toluene (2 mLmmolÀ1 Ar-Cl), 1108C, 20 h.
0.20 equivs. (biph)PACHTREUNG
[b]
Side product formation detected by LC-MS: methyl ester vs. tert-pentyl ester vs. aryl (tert-pentyl) ether 5:2:1 (not standar-
dized).
[c]
[d]
[e]
Major amounts of starting chloroarene unaltered.
Value given in parentheses accounts for additionally isolated product as tert-pentyl ester.
Isolated yield for methyl ester; tert-pentyl ester formation detected by LC-MS (Me vs. t-Pent ca. 3:1, not standardized).
Isolated yields, yields are not optimized.
ester or nitrile groups (Table 4), or electron-donating with benzyl derivatives increased from 21–38%
alkoxy groups (Table 5), amination reactions proceed- within the methyl benzoate subset (entries 18 and 19)
ed generally in good yields for both the first and the to around 55% within the benzonitrile series (en-
second amination step. For the introduction of N- tries 21 and 22). In accordance with the experiences
methylpiperazine into methyl benzoate (entry 18), a made for an attachment of 2-phenoxyethylamine to
milder variant using BINAP (2) and Cs2CO3 (method nitrobenzenes (Table 2, entry 8), said amine was intro-
B2) proved to be superior over DPPF (1)/NaO-t- duced smoothly to a 2-chlorobenzonitrile using K3PO4
Pent, giving rise to more than twice the product out- as base in 43% (entry 23).
come, which is in good accordance with literature
For electron-rich substrates (Table 5), the first ami-
statements.[13,14,16,17] In the second amination reaction nation with N-methylpiperazine succeeded similarly
with benzyl- and furfurylamine (entries 18 and 19), to the respective step on benzonitriles with around
however, K3PO4 was not able to produce satisfactory 65% yield (method A, cf. Table 4, entry 21). Subse-
conversions of chloroarenes. Thus, the stronger but quent introduction of benzyl-, furfuryl- and phene-
more nucleophilic base NaO-t-Pent was required for thylamine was achieved in excellent yields of 50 to
obtaining optimized yields (method C4 vs. C2), al- 75% (entries 24–26). As is deducible from the litera-
though significant transesterification occurred.[37] Phen- ture,[28] a Xantphos-based protocol for the implemen-
ethylamine was coupled accordingly with 30% yield tation of carbamates was not suitable for electron-rich
(entry 20).
bromoarenes (no coupling of N-Boc-benzylamine
Without a possibility for this transesterification side with 1-benzyloxy-2-bromo-4-chlorobenzene detecta-
reaction to take place, yields for the second amination ble, method F1).
Adv. Synth. Catal. 2007, 349, 2286 – 2300
ꢁ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2293

Stefan Tasler et al.
FULL PAPERS
Table 5. Two-fold amination of 2,4-dihalogenated phenol ethers.
[a]
A: Ar-Br/amine 1.0:1.1; 0.05equivs. (DPPF)PdCl
,
2
0.15equivs. DPPF
(
1), 1.25equivs. NaO- t-Pent, THF
(2 mLmmolÀ1 Ar-Br), 1008C, 4 h; C2: Ar-Cl/amine 1.0:1.3; 0.10 equivs. Pd
A
ACHTRE(UGN tBu)2 3a, 1.4
ACHTREUNG
ACHTREUNG
[b]
[c]
Incomplete conversion, starting chloroarene detectable.
LC-MS indicated complete conversion, differences in yields due to work-up.
Isolated yields, yields are not optimized; n.a.=not applied.
For transformations which had to be classified as Scheme 1). The diminution in efficacy when changing
failures, including conversions of certain acetoxy, to 2-(3-aminophenyl)-benzothiazole (13, 36% yield)
acetamide and sulfonamide haloarenes, cf. Supporting seemed to be simply a result of increased nucleophi-
Information.
licity of the amine due to the shift of the electron-
withdrawing thiazole unit from the para to the meta
position, thus rendering Xantphos (4) a sub-optimal
ligand.
With more complex molecules for which a direct
nucleophilic amination failed due to the low nucleo-
Amination of Chloropyrimidines and -quinazolines:
Xantphos-Based Catalysis
Among kinase inhibitors, aminopyrimidines and -qui- philicity of the amine substrates, a Xantphos-based
nazolines belong to the most potent scaffolds.[38] protocol for palladium-catalyzed amination reactions
Direct nucleophilic substitution of chlorine in 2-posi- (method F2) yielded the desired products. Yields as
tion of the former with anilines – accelerated by HCl depicted in Table 6, however, should not be used for
formed intermediary – is usually the synthetic mode comparison of efficacy as a fast and simple purifica-
of choice,[39] but failed for the conversion of 2-chloro- tion procedure was applied (a rinsing sequence), so
4-(N-methylpiperazino)pyrimidine (10). Weak nucleo- that some product loss occurred depending on their
philes in Pd-catalyzed aromatic aminations are gener- solubility. Generally, when 4-chloroquinazolines 15
ally converted smoothly using Xantphos (4) as ligand were aminated using a further functionalized amino-
(vide supra),[28–30] which was likewise suitable for ami- benzothiazole 16, conversions were quite excellent,
nation of chloropyrimidines and -pyridines as sub- isolated yields for 17 ranged from 40 to 81%. A free
strates,[29] whereas (biph)P
ACHTRE(UNG t-Bu)2 3a or BINAP (2) phenolic hydroxy group did not interfere with the
can effectively couple chloropyridines with good N- amination process (81% yield, entry 4) and even the
nucleophiles.[3,14,40] This general notion was confirmed corresponding 4-chloro-7-acetoxyquinazoline (R1 =
with 2-(4-aminophenyl)-benzothiazole (11), a phen- Me, R2 =Ac; not shown) was converted successfully
ylogous 2-aminothiazole and thus weak nucleophile, using this procedure.[41] A 2-aminobenzooxazole 18
which was attached to chloropyrimidine 10 in an ex- was likewise introduced with excellent 75% yield.
cellent yield of 73% (method F2, Xantphos), as com-
A good example of complete chemoselectivity is
pared to only 12% of product 12 being isolated when shown in Scheme 2, where aminobenzothiazoles 20
(biph)PACHTREUNG(t-Bu)2 3a was used as ligand (method C2, were exclusively attached to the 4-chloroquinazoline
2294
asc.wiley-vch.de
ꢁ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2007, 349, 2286 – 2300

Transition Metal-Catalyzed Aromatic Amination Protocols in Medicinal Chemistry
FULL PAPERS
Scheme 1. C2: Ar-Cl/amine 1.0:1.2; 0.04 equivs. Pd2dba3, 0.16 equivs. (biph)PCATHER(UNG t-Bu)2 3a, 1.4 equivs. K3PO4, DME
(2 mLmmolÀ1 Ar-Cl), 1008C, 20 h; F2: Ar-Cl/amine 1.0:1.2; 0.04 equivs. Pd2dba3, 0.12 equivs. Xantphos (4), 1.4 equivs.
[a]
K3PO4, dioxane (4 mLmmolÀ1 Ar-Cl), 1008C, 20 h. Incomplete conversion, starting chloropyrimidine detectable. Isolated
yields, yields are not optimized.
Scheme 2. C2: Ar-Br/amine 1.0:1.1; 0.05equivs. Pd 2dba3, 0.15equivs. (biph)P
ACHTREUNG
(10 mLmmolÀ1 Ar-Br), 1008C, 20 h; C4: Ar-Hal/amine 1.0:1.1; 0.05equivs. Pd 2dba3, 0.15equivs. (biph)P
AHCTREUNG
equivs. NaO-t-Pent, toluene (10 mLmmolÀ1 Ar-Hal), 1108C, 20 h; C6: Ar-Cl/amine 1.0:1.1; 0.1 equivs. Pd
ACHTREUNG
equivs. (biph)PCy2 3b, 1.4 equivs. NaO-t-Pent, toluene (10 mLmmolÀ1 Ar-Cl), 1008C, 20 h; F2: Ar-Cl/amine 1.0:1.0; 0.02
equivs. Pd2dba3, 0.06 equivs. Xantphos (4), 1.4 equivs. K3PO4, dioxane (5mLmmol À1 Ar-Cl), 1008C, 6 h; F3: Ar-Cl/amine
1.0:1.0; 0.02 equivs. Pd2dba3, 0.06 equivs. Xantphos (4), 1.4 equivs. K3PO4, 0.16 equivs. H2O, dioxane (3 mLmmolÀ1 Ar-Cl),
[a]
[b]
1008C, 20 h. Reaction time 20 h. No conversion, only starting haloarene detectable. Isolated yields, yields are not opti-
mized.
15 despite a bromine or chlorine substituent at the ar- sions, as was suggested in the literature for certain
omatic ring (Xantphos-based protocols F2 and F3). transformations, most likely caused by higher solubili-
For conversion of 20, X=Br, addition of water was ty of the base.[25,29]
advantageous (method F3) for achieving high conver-
Adv. Synth. Catal. 2007, 349, 2286 – 2300
ꢁ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2295

Stefan Tasler et al.
FULL PAPERS
Table 6. Amination of 4-chloroquinazoline with 2-aminobenzothiazoles and -oxazoles.[a]
[a]
F2: Ar-Cl/amine 1.0:1.0; 0.02 equivs. Pd2dba3, 0.06 equivs. Xantphos (4), 1.4 equivs. K3PO4, dioxane (10 mLmmolÀ1 Ar-
Cl), 1008C, 20 h.
Conducted with 5mL dioxane/mmol Ar-Cl.
Conducted with 1.1 equivs. K3PO4, 5mL dioxane/mmol Ar-Cl, 100 8C, 6 h.
Conducted with 3 mL dioxane/mmol Ar-Cl.
[b]
[c]
[d]
Isolated yields, yields are not optimized.
Unfortunately, a subsequent replacement of the Parallel Amination Chemistry
halogen in 21 by N-methylpiperazine failed under var-
ious conditions [X=Cl, methods C4 and C6, the For another series of compounds, displaying inhibito-
latter comprising (biph)PCy2 3b, Figure 1; X=Br, ry effects on certain protease activities, a parallel syn-
methods C2 and C4]. This problem seemed to be in- thetic approach using Pd-catalyzed amination reac-
trinsic to the aminobenzothiazole unit and was most tions was established. Using a GreenHouse Parallel
likely not caused by other parts of the complex mole- SynthesiserTM from Radleys Discovery Technology
cule 21 as was shown in a likewise unsuccessful ami- (RDT), vials were not sealed separately but were
nation of 20 (X=Br) directly with N-methylpipera- heated in a joined argon atmosphere with nickel con-
zine (methods C2 and C4), which did not yield com- densing fingers positioned within every individual
pound 22.[42] With N-methylpiperazine being installed glass reaction tube. Some adaptations of standard
differently by a multistep synthesis not involving pal- method C4 had to be performed to allow for an easy
ladium chemistry, the complete unit 20 (X=N-meth- charging of all vials with catalyst. For the arylation of
ylpiperazinyl) was finally attached to chloroquinazo- a certain piperidine derivative (here 24) with differ-
line 15 to give 23 with 56% yield (method F2).
ently substituted bromoarenes 25 (Table 7), stock sol-
utions of 24 and of Pd2dba3 and (biph)P(t-Bu)2 3a
AHCTREUNG
were prepared, aliquots of which were pipetted into
each reaction vial (method C7). Products 26 were iso-
lated with 10 to 40% yield upon filtration and HPLC
2296
asc.wiley-vch.de
ꢁ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2007, 349, 2286 – 2300

Transition Metal-Catalyzed Aromatic Amination Protocols in Medicinal Chemistry
FULL PAPERS
Table 7. Parallel amination reactions using a GreenHouse
Parallel SynthesiserTM [a]
purification, with only 8-bromoquinoline resulting in
no significant conversion (entry 7).
.
Conclusions
Generally, Pd-catalyzed amination reactions were
shown to be successful for a greatly diversified set of
substrates, even though a profound fine-tuning of the
catalytic system seems unavoidable when mediocre
yields are not tolerable. In our case, the major goal
was to gain sufficient quantities of product for biolog-
ical testing and yields obtained were thus acceptable.
As a most versatile and generally applicable proce-
dure, Buchwaldꢀs system based on (biph)PCAHTRE(UNG t-Bu)2 3a
as ligand with either NaO-t-Pent or K3PO4 as base
was identified: this protocol yielded good results for
chloroarenes in combination with secondary alicyclic
amines, benzylamines (except with 2-chloronitroar-
enes[14]), aminopyrroles and 2-substituted ethylamines,
and was likewise usable for bromoarene substrates.
For a chemoselective first amination of bromochloro-
arenes, less reactive catalytic systems were chosen
which usually were only amenable to aryl bromides.
Diversity was best introduced on aryl bromides (thus
in a first amination step), which still allow for amina-
tions with a greater variety of amines as compared to
aryl chlorides. The appropriate choice of ligand in
these cases was crucial for certain subgroups of N-nu-
cleophiles, hence amination reactions could not be
performed using a general catalytic system. The most
robust system for the arylation of bromoarenes with
benzylamines and piperazines was according to Hart-
wigꢀs DPPF-based procedure. Aminations using
BINAP/Cs2CO3 proved to be an excellent alternative
for substrates bearing base sensitive functional
groups, N-heterocycles were best arylated by copper
catalysis. For substituted carbamates, primary and sec-
ondary amides and other weak N-nucleophiles like 2-
aminobenzothiazoles and -oxazoles, Xantphos-based
protocols were identified to be very versatile and reli-
able, enabling amination reactions of aryl bromides,
chloropyrimidines and chloroquinazolines. However,
electron-rich aryl bromides, like ortho-alkoxybro-
moarenes, were found to represent challenging sub-
strates for Xantphos-mediated aminations, only very
rare examples of which have been published so
far.[28,29,31] In all cases of amination reactions of bro-
mochloroarenes, the chlorine site was never aminated
prior to the bromine site.
[a]
C7: stock solution I – amine 24 in toluene/DMA (4:1,
10 mLmmolÀ1 amine); stock solution II
Pd2dba3,
(biph)P
(t-Bu)2 3a (Pd/L=1:2) in toluene (200 mLmmolÀ1
Pd2dba3); reaction vial was charged with 1.4 equivs. NaO-
t-Pent, 1 mL of each stock solution I and II, and
1.0 equivs. of aryl bromide 25 dissolved in 1 mL toluene;
final conditions: Ar-Br/amine 1.0:1.0; 0.05equivs. Pd 2dba3,
–
AHCTREUNG
Arylation of 5-membered N-heterocycles proved to
be tricky for substrate combinations chosen here,
which were not suitable to assess the general applica-
bility of these procedures. Proceeding still with poor
yields but at least yielding some product, copper-
based protocols showed to be superior over Pd-cata-
lyzed reactions, though. Generally, 2-bromonitroar-
0.2 equivs. (biph)PACHTRE(UNG t-Bu)2 3a, 1.4 equivs. NaO-t-Pent, tolu-
ene/DMA (14:1, 30 mLmmolÀ1 Ar-Br), 1008C, 20 h.
Product traces detected by LC-MS, along with only start-
ing bromoarene.
[b]
Isolated yields, yields are not optimized.
Adv. Synth. Catal. 2007, 349, 2286 – 2300
ꢁ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2297

Stefan Tasler et al.
FULL PAPERS
benzonitrile (Table 4, entries 21–23), 2-bromo-4-chloroani-
sole (Table S1, entry 1; Supporting Information), 4-bromo-2-
chlorobenzenesulfonamide (Table S1, entry 4). All sub-
strates, reagents and solvents were used without further pu-
rification. Yields were not further optimized, no efforts were
made to isolate product from all HPLC fractions or mother
liquors.
enes represented problematic substrates for Cu-cata-
lyzed amination processes, as side reactions were
quite competitive, using up the catalytic species. Ad-
ditionally, combinations of sterically demanding sub-
strates (i.e., ortho substituted) were reportedly unsuit-
able for such Cu-mediated conversions.
For a few substrates like certain acetoxy and sulfon-
amide substituted bromoarenes, none of the general
protocols yielded any product at all. According to the
literature, some challenging substrates might be con-
verted successfully by utilizing specialized ligands like
3c (Figure 1), calling for an elaborated fine-tuning of
the catalytic system, which, however, was not aimed
at in this work.
General Procedure for Transition Metal-Catalyzed
Amination Reactions
An oven-dried vial with screw top was charged with all solid
or oily reactants: catalyst, ligand (except for amine ligand
rac-5), base, aryl halide, and – if applicable – the respective
amine. The tube was evacuated and purged with argon, and
solvent and – if applicable – liquid reactants (e.g., N-methyl-
piperazine, amine ligand rac-5) were added with a syringe or
Eppendorf pipette. The vial was sealed (screw cap with
septum) and heated for a certain period. Ratios of reactants
and substrates as well as temp. and reaction time can be de-
Finally, aminations of nitroarenes and methyl ben-
zoates confirmed their potential for decomposition in
coupling
reactions
applying
NaO-t-Pent
as
base.[2,3,13,14,16,17] Interestingly, this effect was strongly
pronounced for aminations of arylchlorides with duced from the Table and Scheme captions within the main
text.
Reaction scale ranged from 0.1 to 8.0 mmol aryl halide,
(biph)PACHTREUNG(t-Bu)2 3a as ligand and of aryl bromides using
BINAP (2) – with DPPF (1), however, quite decent
yields of 25–38% were attained (Tables 1, 2 and 4).
Despite this incompatibility with base-sensitive head
groups, NaO-t-Pent became base of choice for certain
amination reactions of such ortho substituted aryl
chlorides, in which the use of K3PO4 in combination
with ligand 3a resulted in almost no conversion, or
side reactions (e.g., Heck-type coupling) dominated,
with the stronger base accelerating the catalytic cycle
for amination reactions and thus subduing unwanted
reaction channels (Tables 2-4, entries 7, 13, 18, 19).
As with SNAr amination reactions,[43] Pd-catalyzed
ones can also be performed as parallel syntheses
using standard reactor blocks which enable a reaction
performance in an inert atmosphere.
most standard reactions of Tables 1 to 5were conducted at
a 0.4 or 0.8 mmol scale, reactions of Scheme 1 and Scheme 2
and Table 6 were started with 0.1–0.3 mmol aryl chloride,
parallel reactions of Table 7 were conducted with 0.1 mmol
starting material. For upscale reactions, 1.0 to 8.0 mmol aryl
halide were used.
Supporting Information
Work-up and spectral data for all compounds, including side
products, are given in the Supporting Information.
Acknowledgements
For a detailed comparison of the scope of transition
metal catalyzed aromatic amination and SNAr proto-
cols, see Supporting Information.
Synthetic support by Dr. A. Wuzik and a skilled preparative
HPLC/LC-MS performance by O. Müller were highly appre-
ciated.
References
Experimental Section
[1] a) J. Louie, J. F. Hartwig, Tetrahedron Lett. 1995, 36,
3609–3612; b) A. S. Guram, R. A. Rennels, S. L. Buch-
wald, Angew. Chem. 1995, 107, 1456–1459; Angew.
Chem. Int. Ed. Engl. 1995, 34, 1348–1350.
[2] For review articles, see: a) J. F. Hartwig, Angew. Chem.
1998, 110, 2154–2177; Angew. Chem. Int. Ed. Engl.
1998, 37, 2046–2067; b) D. Prim, J.-M. Campagne, D.
Joseph, B. Andrioletti, Tetrahedron 2002, 58, 2041–
2075; c) ref.[3]
[3] A. R. Muci, S. L. Buchwald, Top. Curr. Chem. 2002,
219, 131–209.
[4] For a review, see: S. V. Ley, A. W. Thomas, Angew.
Chem. 2003, 115, 5558–5607; Angew. Chem. Int. Ed.
Engl. 2003, 42, 5400–5449.
General Remarks
Pd2dba3, all phosphine and amine ligands were purchased
from STREM Chemicals except for rac-BINAP (2) and
(biph)PCy2 (3b), which were obtained from Sigma Aldrich
and Acros, respectively. PdACTHRE(UNG OAc)2 was purchased from
Acros, CuI and (DPPF)PdCl2 from Sigma Aldrich. Sodium
tert-pentylate, K3PO4, sodium phenolate trihydrate and
KOH were ordered from Sigma Aldrich, KOH was finely
ground and dried in high vacuum over night prior to use.
Cs2CO3 and K2CO3 were obtained from Acros. All organic
solvents were purchased from Sigma Aldrich as Sure/SealTM
,
H2O was de-ionized and degassed prior to use. The follow-
ing halogenated arenes were commercially available: 1-
bromo-4-chloro-2-nitrobenzene (Table 1, entries 1 and 2), 5-
bromo-2-chloroanisole (Table 1, entry 4), 4-bromo-2-chloro-
[5] A. Kiyomori, J.-F. Marcoux, S. L. Buchwald, Tetrahe-
dron Lett. 1999, 40, 2657–2660.
2298
asc.wiley-vch.de
ꢁ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2007, 349, 2286 – 2300

Transition Metal-Catalyzed Aromatic Amination Protocols in Medicinal Chemistry
FULL PAPERS
[6] X. Huang, K. W. Anderson, D. Zim, L. Jiang, A. Kla-
pars, S. L. Buchwald, J. Am. Chem. Soc. 2003, 125,
6653–6655.
chemistry, see ref.[6] and J. E. Ney, J. P. Wolfe, J. Am.
Chem. Soc. 2005, 127, 8644–8651.
[22] R. Kuwano, M. Utsunomiya, J. F. Hartwig, J. Org.
[7] A. Klapars, J. C. Antilla, X. Huang, S. L. Buchwald, J.
Am. Chem. Soc. 2001, 123, 7727–7729.
[8] J. C. Antilla, A. Klapars, S. L. Buchwald, J. Am. Chem.
Chem. 2002, 67, 6479–6486.
[23] Complete chemoselectivity for amination of the bro-
mine site over chlorine site in 1,3-bromochlorobenzene
under these conditions at room temperature has been
reported: X. Z. Yan, J. Pawlas, T. Goodson III, J. F.
Hartwig, J. Am. Chem. Soc. 2005, 127, 9105–9116.
[24] A. Klapars, S. L. Buchwald, J. Am. Chem. Soc. 2002,
124, 14844–14845.
Soc. 2002, 124, 11684–11688.
[9] For Ni catalysis, see: a) J. P. Wolfe, S. L. Buchwald, J.
Am. Chem. Soc. 1997, 119, 6054–6058; b) C. Desmar-
ets, R. Schneider, Y. Fort, J. Org. Chem. 2002, 67,
3029–3036; c) E. Brenner, R. Schneider, Y. Fort, Tetra-
hedron 1999, 55, 12829–12842; d) S. Tasler, B. H. Lip-
shutz, J. Org. Chem. 2003, 68, 1190–1199.
[25] A. Klapars, X. Huang, S. L. Buchwald, J. Am. Chem.
Soc. 2002, 124, 7421–7428.
[26] In the original protocol (ref.[17]), NaOPh was prepared
as an anhydrous base in contrast to a trihydrate as
commercially available used within this work.
[10] These demands are in clear contrast to those for indus-
trial scale-up, examples of which using Pd-catalyzed
amination reactions have been published most recently:
elaborate fine-tuning of reaction conditions has to be
performed to optimize yields, reduce side reactions, fa-
cilitate workup or decrease hazardous potential due to
exothermic conversions and to decrease amount of cat-
alyst to get dirt-cheap and thus economically attractive
processes; see S. L. Buchwald, C. Mauger, G. Mignani,
U. Scholz, Adv. Synth. Catal. 2006, 348, 23–39.
[11] Several applications of catalyzed aromatic aminations
in industrial research have been reported in the recent
past, these – however – rather focused on the introduc-
tion of a defined amine into a special target molecule,
including a detailed ligand screen and condition optimi-
zation, and did not aim at the generation of diverse li-
braries around one structural core unit. For a review on
such applications with a focus on practical issues of the
methodology, see B. Schlummer, U. Scholz, Adv. Synth.
Catal. 2004, 346, 1599–1626.
[27] A. Ghosh, J. E. Sieser, M. Riou, W. Cai, L. Rivera-
Ruiz, Org. Lett. 2003, 5, 2207–2210.
[28] J. Yin, S. L. Buchwald, J. Am. Chem. Soc. 2002, 124,
6043–6048.
[29] J. Yin, M. M. Zhao, M. A. Huffman, J. M. McNamara,
Org. Lett. 2002, 4, 3481–3484.
[30] For other Xantphos applications in couplings of bro-
moarenes and amides, carbamates, oxazolidinones and
ureas, see: a) B. H. Yang, S. L. Buchwald, Org. Lett.
1999, 1, 35–37; b) S. Cacchi, G. Fabrizi, A. Goggiama-
ni, G. Zappia, Org. Lett. 2001, 3, 2539–2541; c) G. A.
Artamkina, A. G. Sergeev, I. P. Beletskaya, Tetrahedron
Lett. 2001, 42, 4381–4384; d) ref.[31]; for arylation of
anilines using Xantphos, see: e) B. C. Hamann, J. F.
Hartwig, J. Am. Chem. Soc. 1998, 120, 3694–3703
(Xantphos=DPPX); f) Y. Guari, D. S. van Es, J. N. H.
Reek, P. C. J. Kamer, P. W. N. M. van Leeuwen, Tetra-
hedron Lett. 1999, 40, 3789–3790; g) M. C. Harris, O.
Geis, S. L. Buchwald, J. Org. Chem. 1999, 64, 6019–
6022.
[12] S. Tasler, J. Kraus, A. Wuzik, O. Müller, A. Aschen-
brenner, E. Cubero, R. Pascual, J.-R. Quintana-Ruiz,
A. Dordal, R. Mercꢂ, X. Codony, Bioorg. Med. Chem.
Lett., submitted.
[13] J. P. Wolfe, S. L. Buchwald, J. Org. Chem. 2000, 65,
1144–1157.
[31] a) R. G. Browning, H. Mahmud, V. Badarinarayana,
C. J. Lovely, Tetrahedron Lett. 2001, 42, 7155–7157;
b) A. G. Sergeev, G. A. Artamkina, I. P. Beletskaya,
Tetrahedron Lett. 2003, 44, 4719–4723, for electron-rich
[14] J. P. Wolfe, H. Tomori, J. P. Sadighi, J. Yin, S. L. Buch-
wald, J. Org. Chem. 2000, 65, 1158–1174.
bromoarenes,
a 3,5-(CF3)2-Xantphos derivative has
[15] In ref.[14], a competition experiment is presented in
which 4-bromotoluene was aminated with morpholine
in the presence of 4-chlorotoluene to proceed in favor
of the aryl bromide in a 97:3 ratio.
been utilized.
[32] For a succesful amination of an aryl chloride with
indole using the sterically more congested representa-
tive of the monodentate biphenyl-phosphine ligands 3c,
see ref.[6]
[16] M. S. Driver, J. F. Hartwig, J. Am. Chem. Soc. 1996,
118, 7217–7218.
[33] A DPPF based protocol did not qualify as an alterna-
tive, as it was only established for aminations of steri-
cally unhindered aryl bromides (that is, no ortho substi-
tution): G. Mann, J. F. Hartwig, M. S. Driver, C. Fer-
nandez-Rivas, J. Am. Chem. Soc. 1998, 120, 827–828.
[34] Shortly after finishing these experiments, generalized
procedures for the N-arylation of pyrroles and pyra-
zoles were also published, with conditions for the
former being identical to those described herein
(method E2) and for the latter preferring K2CO3 as
base over K3PO4; see: J. C. Antilla, J. M. Baskin, T. E.
Barder, S. L. Buchwald, J. Org. Chem. 2004, 69, 5578–
5587.
[17] J. F. Hartwig, M. Kawatsura, S. I. Hauck, K. H. Shaugh-
nessy, L. M. Alcazar-Roman, J. Org. Chem. 1999, 64,
5575–5580.
[18] For a test reaction using for example, benzylamine and
4-bromobenzonitrile using conditions A and B1, similar
conversions were observed using either BINAP or
DPPF (quantitative), but only traces of product forma-
tion were observed applying method C1.
[19] M. S. Viciu, R. M. Kissling, E. D. Stevens, S. P. Nolan,
Org. Lett. 2002, 4, 2229–2231.
[20] Lancaster catalogue 2004/2005, sodium tert-pentoxide
(OrderCode X15112) vs. sodium tert-butoxide (Order-
Code X15051).
[35] A further possibility for such a conversion might be a
copper catalyzed reaction, even though only estab-
[21] M. R. Netherton, G. C. Fu, Org. Lett. 2001, 3, 4295–
4298; for an application of this method in amination
Adv. Synth. Catal. 2007, 349, 2286 – 2300
ꢁ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2299

Stefan Tasler et al.
FULL PAPERS
lished for iodoarenes in combination with similar
amine substrates; see ref.[7]
2003, 13, 2961–2966; c) I. R. Hardcastle, C. E. Arris, J.
Bentley, F. T. Boyle, Y. Chen, N. J. Curtin, J. A. Endi-
cott, A. E. Gibson, B. T. Golding, R. J. Griffin, P. Jews-
bury, J. Menyerol, V. Mesguiche, D. R. Newell,
M. E. M. Noble, D. J. Pratt, L.-Z. Wang, H. J. Whitfield,
J. Med. Chem. 2004, 47, 3710–3722; d) S. Verma, D.
Nagarathnam, J. Shao, L. Zhang, J. Zhao, Y. Wang, T.
Li, E. Mull, I. Enyedy, C. Wang, Q. Zhu, M. Altieri, J.
Jordan, T.-T.-A. Dang, S. Reddy, Bioorg. Med. Chem.
Lett. 2005, 15, 1973–1977.
[36] Ref.[34] suggested a prior in situ Cu-catalyzed halogen
exchange for bromoarenes to enhance their reactivity
towards Cu-mediated aminations of N-heterocycles in
otherwise problematic substrate combinations. This ex-
change, however, is hampered for electron-rich bro-
moarenes with ortho substitution according to ref.[24]
[37] Similar side product formation was described in, e.g.,
refs.[13,16]
[38] a) O. Prien, ChemBioChem. 2005, 6, 500–505; b) T. L.
Underiner, B. Ruggeri, D. E. Gingrich, Curr. Med.
Chem. 2004, 11, 731–745; c) M. E. M. Noble, J. A. En-
dicott, L. N. Johnson, Science 2004, 303, 1800–1805;
d) A. Huwe, R. Mazitschek, A. Giannis, A. Angew.
Chem. 2003, 115, 2170–2187; Angew. Chem. Int. Ed.
2003, 42, 2122–2138; e) A. J. Bridges, Chem. Rev. 2001,
101, 2541–2571.
[39] For selected examples, see: a) T. Kumagai, T. Okubo,
H. Kataoka-Okubo, S. Chaki, S. Okuyama, A. Nakaza-
to, Bioorg. Med. Chem. 2001, 9, 1349–1355; b) G. A.
Breault, R. P. A. Ellston, S. Green, S. R. James, P. J.
Jewsbury, C. J. Midgley, R. A. Pauptit, C. A. Minshull,
J. A. Tucker, J. E. Pease, Bioorg. Med. Chem. Lett.
[40] S. Wagaw, S. L. Buchwald, J. Org. Chem. 1996, 61,
7240–7241.
[41] Only small quantities of starting 4-chloroquinazoline
were detected by LC-MS to have lost the acetyl group.
[42] Recently, similar conversions were published to be ac-
complished successfully using bulky electron-rich biaryl
phosphine ligands: M. D. Charles, P. Schultz, S. L.
Buchwald, Org. Lett. 2005, 7, 3965–3968.
[43] Y. Dong, J. Y. Roberge, Z. Wang, X. Wang, J. Tamasi,
V. Dell, R. Golla, J. R. Corte, Y. Liu, T. Fang, M. N.
Anthony, D. M. Schnur, M. L. Agler, J. K. Dickson Jr. ,
R. M. Lawrence, M. M. Prack, R. Seethala, J. H. M.
Feyen, Steroids 2004, 69, 201–217.
2300
asc.wiley-vch.de
ꢁ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2007, 349, 2286 – 2300