Efficient palladium catalysts for the amination of aryl chlorides: A comparative study on the use of phosphium salts as precursors to bulky, electron-rich phosphines

Article abstract of DOI:10.1016/j.tet.2005.06.067

Alkyl-di-(1-adamantyl)phosphonium salts are practical ligand precursors for the palladium-catalyzed amination of aryl chlorides. In the presence of typically 0.5 mol% Pd(OAc)₂ and 1 mol% of ligand precursor a variety of activated and deactivated aryl chlorides can be aminated in good to excellent yield (73-99%). Applying optimized conditions catalyst turnover numbers up to 10,000 have been achieved.

Full text of DOI:10.1016/j.tet.2005.06.067

Tetrahedron 61 (2005) 9705–9709
Efficient palladium catalysts for the amination of aryl chlorides:
a comparative study on the use of phosphium salts as precursors to
bulky, electron-rich phosphines
b
b,
Amit Tewari,a Martin Hein,a, Alexander Zapf and Matthias Beller
*
*
a
¨ ¨
Institut fur Chemie, Universitat Rostock, Albert-Einstein-Straße 3a, D-18059 Rostock, Germany
¨ ¨
Leibniz-Institut fur Organische Katalyse an der Universitat Rostock e.V. (IfOK), Albert-Einstein-Str. 29a, D-18059 Rostock, Germany
b
Received 24 March 2005; revised 3 May 2005; accepted 17 June 2005
Available online 18 July 2005
Abstract—Alkyl-di-(1-adamantyl)phosphonium salts are practical ligand precursors for the palladium-catalyzed amination of aryl
chlorides. In the presence of typically 0.5 mol% Pd(OAc)2 and 1 mol% of ligand precursor a variety of activated and deactivated aryl
chlorides can be aminated in good to excellent yield (73–99%). Applying optimized conditions catalyst turnover numbers up to 10,000 have
been achieved.
q 2005 Elsevier Ltd. All rights reserved.
1. Introduction
phosphonium salts, that meets the above mentioned criteria.
Noteworthy are the high catalyst productivity, and the air
stability of the ligand precursors, which makes them easy to
operate.
The palladium-catalyzed C–N bond formation of aryl
halides (Buchwald–Hartwig reaction) is a rapidly develop-
ing field of interest due to the importance of anilines and
amino-substituted heteroarenes as natural products, drugs,
agrochemicals, and fine chemicals.1 In order to apply such
reactions in the fine chemical industry significant cost
reduction of a typical lab-scale synthesis is an important
requirement. Efforts to substitute costly starting materials
such as aryl iodides or triflates by economically more
attractive chloro- and bromoarenes, reduction of catalyst
concentration, etc. are to be seen in this respect. Despite
numerous advances in C–N cross coupling processes,2 an
important factor for industrial applications of palladium-
catalyzed amination reactions is the development of more
efficient and economically attractive catalysts. Important
criteria for such improved catalysts include (1) no special
handling of metal complexes and ligands, (2) broad
substrate scope, as well as (3) ability to operate under
mild reaction conditions and at low catalyst concentration.
In this respect there exists still a need for easy-to-use
ligands, which lead to highly active catalyst systems, are
easily tunable and allow for scale-up.
Previously we demonstrated, that basic, sterically demand-
ing phosphines with 1-adamantyl substituents3 (Fig. 1) are
suitable ligands for palladium-catalyzed C–C, C–N, and
C–O bond forming reactions. Among the various aryl-4 and
alkyl-di-(1-adamantyl)phosphines5 (cataCXiumw
A
ligands)6 prepared especially di-(1-adamantyl)-n-butyl-
phosphine (1) and di-(1-adamantyl)benzylphosphine (7)
showed good to excellent catalyst performance for different
functionalization reactions of aryl chlorides.7 Very recently,
we also reported on the preparation of the respective
phosphonium salts via alkylation of di-(1-adamantyl)-
phosphine with alkyl or benzyl halides.8 In this regard it is
noteworthy that phosphonium salts of sterically hindered
alkylphosphines became interesting as ligand precursors for
palladium-catalyzed coupling reactions due to their
Herein, we report a novel catalytic system based on
Keywords: Amination; Anilines; Aryl chlorides; Palladium; Phosphines.
*
Corresponding authors. Tel.: C49 381 4986428; fax: C49 381 4986412
(M.H.); tel.: C49 381 12810; fax: C49 381 12815000 (M.B.); e-mail
addresses: martin.hein@uni-rostock.de; matthias.beller@ifok-rostock.de
Figure 1. Selected examples of alkyl-di-(1-adamantyl)phosphines
(cataCXiumw A ligands).
0040–4020/$ - see front matter q 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2005.06.067

9706
A. Tewari et al. / Tetrahedron 61 (2005) 9705–9709
increased stability against air and moisture.9 However, to
the best of our knowledge phosphonium salts have not been
tested as ligands in the amination of aryl chlorides.
Scheme 1. Coupling reaction of chlorobenzene and tert-butylamine.
2. Results and discussion
As a starting point of our investigations we examined the
amination of chlorobenzene as an example of a non-
activated aryl chloride, with a slight excess of the sterically
hindered tert-butylamine (Scheme 1, Table 1).
Table 1. Variation of ligands for the coupling reaction of chlorobenzene
and tert-butylamine
Entry
Ligand
precursor
Conv. (%)a
Yield (%)a
At first the influence of the ligand structure was studied in
the presence of 0.5 mol% Pd(OAc)2 and 1 mol% of
phosphonium salt with 1.2 equiv NaOtBu as the base in
toluene at 120 8C in a sealed tube. These are typical reaction
conditions, which have been often used for C–N coupling
reactions. Selected results are summarized in Table 1.
Interestingly, the variation of the alkyl group has a
significant effect on the coupling reaction. Thus, the
modular synthesis of cataCXiumw A ligands should allow
for an easy fine tuning of the ligand properties for other
substrates, too.
1
2
3
4
5
6
7
8
1
100
100
88
100
26
100
100
63
89
94
69
90
26
84
88
59
H1CIK
H2CIK
H3CIK
H4CIK
H5CBrK
H6CBrK
H7CBrK
a Average of two runs, determined by GC using hexadecane as internal
standard.
Table 2. Palladium-catalyzed coupling of aryl chlorides and amines in the presence of H1C K
I
and H5CBrKa
Entry
Aryl chloride
Amine
H1C K
I
H5CBrK
Conv. (%)b
100
Yield (%)b
95
Conv. (%)b
Yield (%)b
1
100
99
2c
100
99
100
90
3
4
100
100
93
73
100
100
93
87
5
6
7
100
100
100
75
93
99
100
100
100
62
93
79
8
9
100
100
95
99
100
100
97
99
10
100
99
100
99
a Conditions: 5 mmol aryl chloride, 6 mmol amine, 6 mmol NaOtBu, 0.5 mol% Pd(OAc)2, 1 mol% ligand, 5 mL toluene, 120 8C, 20 h.
b Average of two runs, determined by GC using hexadecane or diethyleneglycol di-n-butyl ether as internal standard.
c Reaction was carried out on 2 mmol scale.

A. Tewari et al. / Tetrahedron 61 (2005) 9705–9709
9707
Control experiments with the free n-BuPAd2 ligand (1,
Table 1, entry 1) indicated that the corresponding
phosphonium salt H1CIK can be used without any problem
(Table 1, entry 2). Good conversion for our model reaction
is also observed for di-(1-adamantyl)-iso-propyl- (H3CIK),
di-(1-adamantyl)-allyl- (H5CBrK), and di-(1-adamantyl)-
(2-methoxyethyl)phosphonium salts (H6CBrK, Table 1,
entries 4, 6, 7).
only little attention has been paid to the efficiency
(productivity) of the respective palladium catalyst in
amination reactions.11 In general, TONs in the range of
100 are obtained. However, Hartwig et al. have demon-
strated very recently that bidentate electron-rich phosphines
with a ferrocene backbone lead to highly active and
productive catalysts for the amination of aryl chlorides
with primary amines.12
Next, the general usefulness of our ligands was examined
and is shown in Table 2. Due to the ease of synthesis and
the catalytic performance in the model reaction we selected
di-(1-adamantyl)-n-butyl- and di-(1-adamantyl)-allyl-
phosphonium salts (H1CIK, H5CBrK) for more detailed
studies.
The results of our study with lower catalyst concentrations
are summarized in Table 3. Reactions of N-methylaniline
with chlorobenzene, 3-chlorobenzotrifluoride and
4-chloroanisole proceed with very good yield (89–94%;
TONs 8900–9400) in the presence of only 0.01 mol%
Pd(OAc)2 and 0.02 mol% of di-(1-adamantyl)-n-butyl-
phosphonium iodide at 120 8C.
Various secondary amines and tert-butylamine can be
coupled with different aryl chlorides to give the desired
products in high yields. Reactions of chlorobenzene with
secondary amines (acyclic, cyclic and aromatic amines)
occurred in yields of more than 87% (Table 2, entries 1, 3,
4). In general, the amination of acyclic secondary amines
with aryl halides is more challenging than similar reactions
with cyclic secondary amines. Nevertheless the di-(1-
adamantyl)-allylphosphonium salt is quite effective in the
coupling of di-n-butylamine and chlorobenzene (Table 2,
entry 4). In most of our examples we could not observe a
strong dependence of the product yield from donor or
acceptor substitution of the chloroarene. For example,
electron-rich chloroanisole as well as the electron-poor
trifluoromethyl-substituted chlorobenzenes react cleanly
with N-methylaniline (Table 2, entries 6, 9 and 10). On
the other hand, despite favourable electronic conditions,
reaction of 4-chlorobenzonitrile with morpholine gave a
lower yield of 75% due to side reactions of the nitrile group
(Table 2, entry 5).
Table 3. Reaction of aryl chlorides with N-methylaniline at low catalyst
concentrationa
Entry
1
Aryl chloride
Conv. (%)b
100
Yield (%)b
94
2
100
89
3
100
93
a Reaction conditions: 5 mmol aryl chloride, 6 mmol N-methylaniline,
6 mmol NaOtBu, 0.01 mol% Pd(OAc)2, 0.02 mol% H1CIK, 5 mL
toluene, 120 8C, 20 h.
b Average two runs, determined by GC using diethyleneglycol di-n-butyl
ether as internal standard.
On the other hand only low conversion and yield is detected
under these conditions for the reaction of morpholine with
2-chloropyridine (5–10% yield; TON 500–1000).
In addition to simple aryl chlorides, also heterocyclic
and
chloroarenes
such
as
2-chloropyridine
Finally, we investigated aminations of various aryl chlorides
at lower temperatures (60–80 8C). In order to achieve faster
conversion a catalyst concentration of 1 mol% Pd(OAc)2
has been used. As shown in Table 4 reactions of
chlorobenzene and 4-chlorobenzotrifluoride with N-methyl-
aniline gave 71 and 73% yield, respectively, (Table 4,
entries 1–2), which indicates that the amination proceeds
2-chloroquinoline react well with different amines and
demonstrate the scope of our catalyst system 0(Table 2,
entries 2, 8). Furthermore, sterically hindered 2,2 -dimethyl
substituted anilines can be obtained in quantitative yield
(99%) in the presence of 0.5 mol% palladium catalyst
(Table 2, entry 7).
By comparing the performance of the n-butyl-substituted
phosphonium salt H1CIK with the allyl-substituted deriva-
tive H5CBrK it is obvious that the former one is the
(slightly) better ligand precursor in most reactions.
However, in case of the arylation of di-n-butylamine
H5CBrK gave reproducibly better results. The reasons for
this behaviour are so far unclear, but it suggests that
optimized yields can be obtained by further variation of the
alkyl group.
Table 4. Amination of aryl chlorides at lower temperaturea
Entry
Aryl chloride
Amine
Temp. Yield
(%)b
(8C)
1
60
71
2
3
60
80
73
98
Normally, we carried out our catalytic experiments using a
standard procedure (see Section 3) with 0.5 mol% Pd(OAc)2
and 1 mol% of phosphonium salt. Clearly, this is a sufficient
small amount for lab-scale syntheses of substituted anilines.
However, it is well-known that catalyst turnover numbers
(TONs) of around 1000–10,000 are required to consider
larger scale applications.10 Therefore, it is surprising that
a Reaction conditions: 5 mmol aryl chloride, 6 mmol amine, 6 mmol
NaOtBu, 1 mol% Pd(OAc)2, 2 mol% H1C K
, 5 mL toluene, 120 8C,
20 h.
I
b Average two runs, determined by GC using hexadecane or diethylene-
glycol di-n-butyl ether as internal standard.

9708
A. Tewari et al. / Tetrahedron 61 (2005) 9705–9709
significantly slower at this temperature. Nevertheless, in
some cases the use of lower reaction temperatures can be
reasonable if one or both of the substrates contain sensitive
groups. As an example the amination of 2-chloropyridine
with N-(tert-butoxycarbonyl)piperazine is presented, which
gave an excellent yield (98%) at 80 8C (Table 4, entry 3).
3.1.3. N,N-Di-n-butylaniline. MS (EI, 70 eV): m/z (%): 205
[MC], 162, 120, 105, 77.
3.1.4. N-tert-Butyl-2,6-dimethylaniline. MS (EI, 70 eV):
m/z (%): 177, 162, 121.
3.1.5. N-(4-Cyanophenyl)morpholine. MS (EI, 70 eV):
m/z (%): 188 [MC], 130, 102.
In summary, we have shown that phosphonium salts of
alkyl-di-(1-adamantyl)phosphines allow for an efficient
synthesis of a variety of substituted anilines from aryl or
heteroaryl chlorides and amines. Good to excellent yields
(75–99%) are obtained at comparatively low catalyst
concentration (0.5 mol% Pd(OAc)2; 120 8C). By simply
reducing the metal and ligand amount optimized catalyst
turnover numbers up to ca. 10,000 have been observed. In
addition, the coupling reactions proceed under milder
conditions (60–80 8C), albeit with higher catalyst loading.
3.1.6. N-(4-Methoxyphenyl)-N-methylaniline. MS (EI,
70 eV): m/z (%): 213 [MC], 198, 77.
3.1.7. N-Methyl-N-[4-(trifluoromethyl)phenyl]aniline.
MS (EI, 70 eV): m/z (%): 251 [MC], 77.
3.1.8. N-Methyl-N-[3-(trifluoromethyl)phenyl]aniline.
MS (EI, 70 eV): m/z (%): 251 [MC], 145, 77.
An important advantage of the presented method is the easy
handling of catalyst and ligand precursors. Hence, it is not
necessary to exclude strictly air or moisture. Due to the
modular synthesis of cataCXiumw A ligands a fine tuning of
the ligand properties for other substrates is easily possible
and should lead to further improved catalyst performance.
3.1.9. N-(2-Pyridyl)morpholine. MS (EI, 70 eV): m/z (%):
164 [MC], 133, 107, 79.
3.1.10. N-Benzyl-N0-(2-quinolyl)piperazine. Yellow solid;
1H NMR (400 MHz, CDCl3): dZ7.78 (d, 3J(H,H)Z9.1 Hz,
3
3
1H), 7.61 (d, J(H,H)Z8.5 Hz, 1H), 7.50 (d, J(H,H)Z8.
3 Hz, 1H), 7.45 (m, 1H), 7.21 (m, 6H), 6.87 (d, 3J(H,H)Z9.
3 Hz, 1H), 3.68 (t, 3J(H,H)Z5.1 Hz, 4H), 3.49 (s, 2H), 2.51
3
(t, J(H,H)Z5.1 Hz, 4H); 13C NMR (101 MHz, CDCl3):
3. Experimental
dZ157.9, 148.3, 137.8, 129.9, 129.7, 128.7, 127.6, 127.0,
123.5, 122.7, 109.9, 63.6, 53.5, 44.5; MS (EI, 70 eV): m/z
(%): 303 [MC], 157, 128, 91.
Chemicals were obtained from Aldrich, Fluka and Merck
KGaA and used without further purification. Solvents were
1
dried according to standard procedures. H and 13C NMR
3.1.11. N-tert-Butoxycarbonyl-N0-(2-pyridyl)piperazine.
chemical shifts refer to tetramethylsilane (0 ppm) and
CDCl3 (77.0 ppm), respectively. Column chromatography
was carried out using silica gel 60 (0.063–0.2 mm Fluka).
1
Yellow solid; H NMR (250 MHz, CDCl3): dZ8.18 (m,
1H), 7.48 (m, 1H), 6.63 (m, 2H), 3.52 (s (br), 8H), 1.47 (s,
9H); 13C NMR (63 MHz, CDCl3): dZ159.3, 154.8, 148.0,
137.6, 113.6, 107.2, 79.9, 45.1, 43.3 (br), 28.4; MS (EI,
70 eV): m/z (%): 263 [MC], 190, 120, 107, 78.
3.1. General procedure (Buchwald–Hartwig amination)
A 30 mL pressure tube was loaded with Pd(OAc)2 (5.6 mg,
0.025 mmol), the ligand precursor (0.050 mmol), and
NaOtBu (577 mg, 6.0 mmol) and was purged with argon.
Then, toluene (5 mL), the aryl chloride (5.0 mmol), and the
amine (6.0 mmol) were added successively. The mixture
was stirred for 20 h at 120 8C. After cooling to room
temperature the mixture was diluted with diethyl ether
(5 mL) and washed with water (10 mL). The organic phase
was dried over MgSO4, concentrated under vacuum and the
product was isolated by column chromatography (ethyl
acetate/n-hexane or acetone/n-hexane). Alternatively,
diethyleneglycol di-n-butyl ether or hexadecane was
added as internal standard and quantitative analysis was
done by gas chromatography. The commercially available
products were identified by comparison of their GC/MS data
with the data of authentic samples, known products were
characterized by NMR and mass spectroscopy (for more
data see Ref. 4c and cited literature there).
Acknowledgements
This work has been financed by the State of Mecklenburg-
Vorpommern and the Bundesministerium fu¨r Bildung und
Forschung (BMBF).
References and notes
1. (a) Schlummer, B.; Scholz, U. Adv. Synth. Catal. 2004, 346,
1599–1626. (b) Hartwig, J. F. In Negishi, E.-I., Ed.; Handbook
of Organopalladium Chemistry for Organic Synthesis;
Wiley-Interscience: New York, 2002; Vol. 1; p 1051. (c)
Yang, B. H.; Buchwald, S. L. J. Organomet. Chem. 1999, 576,
125–146.
2. (a) Christmann, U.; Vilar, R. Angew. Chem. 2005, 117,
370–378. Angew. Chem., Int. Ed. 2005, 44, 366–374. (b)
Bedford, R. B.; Cazin, C. S. J.; Holder, D. Coord. Chem. Rev.
2004, 248, 2283–2321. (c) Urgaonkar, S.; Verkade, J. G. J.
Org. Chem. 2004, 69, 9135–9142. (d) Bedford, R. B.; Blake,
M. E. Adv. Synth. Catal. 2003, 345, 1107–1110. (e) Stauffer,
S. R.; Hartwig, J. F. J. Am. Chem. Soc. 2003, 125, 6977–6985.
3.1.1. N-Phenylmorpholine. MS (EI, 70 eV): m/z (%): 163
[MC], 105, 77.
3.1.2. Methyldiphenylamine. MS (EI, 70 eV): m/z (%):
183 [MC], 167, 104, 77.

A. Tewari et al. / Tetrahedron 61 (2005) 9705–9709
9709
(f) Zim, D.; Buchwald, S. L. Org. Lett. 2003, 5, 2413–2415. (g)
Hopper, M. W.; Utsunomiya, M.; Hartwig, J. F. J. Org. Chem.
2003, 68, 2861–2873. (h) Kuwano, R.; Utsunomiya, M.;
Hartwig, J. F. J. Org. Chem. 2002, 67, 6479–6486.
Ehrentraut, A.; Zapf, A.; Beller, M. Adv. Synth. Catal. 2002,
344, 209–217. (f) Stambuli, J. P.; Bu¨hl, M.; Hartwig, J. F. J.
Am. Chem. Soc. 2002, 124, 9346–9347. (g) Zapf, A.;
Ehrentraut, A.; Beller, M. Angew. Chem. 2000, 112,
4315–4317. Angew. Chem., Int. Ed. 2000, 39, 4153–4155.
(h) Ehrentraut, A.; Zapf, A.; Beller, M. Synlett 2000,
1589–1592.
¨
3. (a) Gorlich, J. R.; Schmutzler, R. Phosphorus, Sulfur Silicon
¨
Relat. Elem. 1995, 102, 211–215. (b) Gorlich, J. R.;
Schmutzler, R. Phosphorus, Sulfur, Silicon Relat. Elem.
1993, 81, 141–148.
6. These ligands are commercially available from Strem and
Degussa Homogeneous Catalysts (DHC).
4. For catalytic applications see: (a) Harkal, S.; Kumar, K.;
Michalik, D.; Zapf, A.; Jackstell, R.; Rataboul, F.; Riermeier,
T.; Monsees, A.; Beller, M. Tetrahedron Lett. 2005, 46,
3237–3240. (b) Zapf, A.; Jackstell, R.; Rataboul, F.;
Riermeier, T.; Monsees, A.; Fuhrmann, C.; Shaikh, N.;
Dingerdissen, U.; Beller, M. Chem. Commun. 2004, 38–39.
(c) Rataboul, F.; Zapf, A.; Jackstell, R.; Harkal, S.; Riermeier,
T.; Monsees, A.; Dingerdissen, U.; Beller, M. Chem. Eur. J.
2004, 10, 2983–2990. (d) Harkal, S.; Rataboul, F.; Zapf, A.;
Fuhrmann, C.; Riermeier, T.; Monsees, A.; Beller, M. Adv.
Synth. Catal. 2004, 346, 1742–1748. (e) Aranyos, A.; Old, D.
W.; Kiyomori, A.; Wolfe, J. P.; Sadighi, J. P.; Buchwald, S. L.
J. Am. Chem. Soc. 1999, 121, 4369–4378.
7. For an excellent review on palladium-catalyzed functionaliza-
tions of aryl chlorides see: Littke, A. F.; Fu, G. C. Angew.
Chem. 2002, 114, 4350–4386. Angew. Chem., Int. Ed. 2002,
41, 4176–4211.
8. Tewari, A.; Hein, M.; Zapf, A.; Beller, M. Synthesis 2004,
935–941.
9. (a) Netherton, M. R.; Fu, G. C. Org. Lett. 2001, 3, 4295–4298.
¨
(b) Kollhofer, A.; Pullmann, T.; Plenio, H. Angew. Chem.
2003, 115, 1086–1088. Angew. Chem., Int. Ed. 2003, 42,
1056–1058.
10. (a) Zapf, A.; Beller, M. Top. Catal. 2002, 19, 101–109. (b) de
Vries, J. G. Can. J. Chem. 2001, 79, 1086–1092. (c)
Blaser, H.-U.; Studer, M. Appl. Catal., A: Gen. 1999, 189,
191–204.
5. For catalytic applications see: (a) Michalik, D.; Kumar, K.;
Zapf, A.; Tillack, A.; Arlt, M.; Heinrich, T.; Beller, M.
Tetrahedron Lett. 2004, 45, 2057–2061. (b) Datta, A.; Ebert,
K.; Plenio, H. Organometallics 2003, 22, 4685–4691. (c)
Datta, A.; Plenio, H. Chem. Commun. 2003, 1504–1505. (d)
Benedi, C.; Bravo, F.; Uriz, P.; Fernandez, E.; Claver, C.;
Castillon, S. Tetrahedron Lett. 2003, 44, 6073–6077. (e)
11. For an earlier example from our side see: Ehrentraut, A.; Zapf,
A.; Beller, M. J. Mol. Catal. 2002, 182–183, 515–523.
12. Shen, Q.; Shekhar, S.; Stambuli, J. P.; Hartwig, J. F. Angew.
Chem. 2005, 117, 1395–1399. Angew. Chem., Int. Ed. 2005,
44, 1371–1375.