Ruthenium porphyrins-catalyzed atom-efficient amination of C-H bonds by arylazides

Article abstract of DOI:10.1142/S1088424610002501

Benzylic amines are synthesized in yield up to 90% by the Ru(TPP)CO-catalyzed amination of both exocyclic and endocyclic benzylic C-H bonds. The choice of arylazides as nitrogen sources confers to the methodology a good sustainability due to the formation of molecular nitrogen as the only stoichiometric by-product. A preliminary mechanistic investigation evidenced a critical role of the hydrocarbon concentration to drive the chemoselectivity of the reaction.

Full text of DOI:10.1142/S1088424610002501

Journal of Porphyrins and Phthalocyanines
J. Porphyrins Phthalocyanines 2010; 14: 732–740
DOI: 10.1142/S1088424610002501
Published at http://www.worldscinet.com/jpp/
Ruthenium porphyrins-catalyzed atom-efficient amination
of C-H bonds by arylazides
Daniela Intrieri, Alessandro Caselli, Fabio Ragaini, Sergio Cenini and
Emma Gallo*
Dipartimento di Chimica Inorganica, Metallorganica e Analitica “Lamberto Malatesta”, Università di Milano,
and ISTM-CNR, Via Venezian 21, 20133 Milano, Italy
Received 18 June 2010
Accepted 28 July 2010
ABSTRACT: Benzylic amines are synthesized in yield up to 90% by the Ru(TPP)CO-catalyzed
amination of both exocyclic and endocyclic benzylic C-H bonds. The choice of arylazides as nitrogen
sources confers to the methodology a good sustainability due to the formation of molecular nitrogen as
the only stoichiometric by-product. A preliminary mechanistic investigation evidenced a critical role of
the hydrocarbon concentration to drive the chemoselectivity of the reaction.
KEYWORDS: ruthenium, homogenous catalysis, C-H amination.
with unsaturated or saturated hydrocarbons, affording,
INTRODUCTION
with high chemo- and stereo-selectivities, aziridines and
The development of new strategies for the synthesis
amines, respectively [9]. The reaction of saturated hydro-
of aza-derivativies in an economical way is of scientific
carbons with organic azides is responsible for the direct
interest due to the importance such species have as pre-
transformation of a C-H into a C-N bond, a research
cursors for biological and pharmaceutical compounds
area of high impact and extreme synthetic utility. Some
[1, 2]. Nitrogen-containing molecules can be efficiently
examples of intramolecular C-H aminations by organic
obtained in one pot by a direct insertion of a nitrene func-
azides, giving nitrogen-based heterocycles [10] such as
tionality [RN] into an organic framework [3–5]. Among
carbazoles [11, 12], pyrroles [11], indoles [11, 13, 14],
nitrogen sources available to perform amination reac-
indolines [15] and benzimidazoles [16], have already
tions, organic azides (RN3) represent a versatile class
been reported.
of starting materials [6]; they are very reactive and, in
Among the transition metal complexes employed as
several cases, are easily formed from the corresponding
catalysts in these reactions, metalloporphyrins exhib-
amines. The peculiarity of the insertion of a nitrene moi-
ited very good catalytic activity for the intramolecular
ety into an organic skeleton by employing organic azides
C-H amination of organic sulfonyl or phosphoryl azides,
is the generation of the eco-friendly molecular nitrogen as
forming benzosultams [17] and cyclophosphoramidates
the only stoichiometric side-product. Therefore, organic
[18], respectively. On the other hand, to the best of our
azides can be considered as atom-efficient nitrene trans-
knowledge, there are only two examples of intermo-
fer reagents.
lecular C-H aminations by organic azides catalyzed by
Organic azides can be thermally or photochemically
metal porphyrins [19–21]. Among them the amination
activated [7, 8]. But to achieve a satisfactory reaction
of benzylic substrates by arylazides catalyzed by cobalt
chemoselectivity, the presence of a transition metal is
porphyrin complexes was reported by us some years
required. Under these conditions, organic azides react
ago [19, 20]. Besides, we have used aryl azides as nitro-
gen sources in aziridination reactions [22–26] because
of their convenient reactivity/stability relationship and
SPP full member in good standing
their easy synthesis starting from commercially available
anilines.
*Correspondence to: Emma Gallo, email: emma.gallo@unimi.
it, tel: +39 (0)2-50314372, fax: +39 (0)2-50314405
Copyright © 2010 World Scientific Publishing Company

RuthenIum PORPhyRInS-CatalyzeD atOm-effICIent amInatIOn Of C-h bOnDS by aRylazIDeS
733
if R' = H
received. 4(NO2)C6H4N3 [31], 3,5(CF3)2C6H3N3 [31],
CoII(P)
R
CoII(P)
R
CH
4(tBu)C6H4N3 [23], 2,6(NO2)2C6H3N3 [23], 4BrC6H4N3
[32], tetraphenylporphyrin (TPPH2) [33] and Ru(TPP)CO
[34] were synthesized by methods reported in the
literature or using minor modifications thereof. The
purity of hydrocarbons and arylazides employed was
R
R'
C
NAr
C
Ar'
+ArN3, -N2
R'
+ ArN3,
-N2, -ArNH2
NHAr
Ar'
Ar'
P = porphyrin
Scheme 1. Benzylic amination by arylazides catalyzed by
Co(porphyrin) complexes
1
checked by GC-MS or H NMR analysis. NMR spectra
were recorded at room temperature on a Bruker AC-300,
on a Bruker Avance DRX-300, operating at 300 MHz
The general path for the synthesis of benzylic amines
and imines employing arylazides (ArN3) as aminating
agents in the presence of cobalt porphyrins is shown in
Scheme 1.
1
for H, at 75 MHz for 13C and at 282 MHz for 19F, or
on a Bruker Avance DRX-400 spectrometer operating at
400 MHz for 1H and at 100 MHz for 13C. Chemical shifts
(ppm) are reported relative to TMS. The 1H NMR signals
of the compounds described in the following have been
attributed by COSY and NOESY techniques. Assign-
ments of the resonance in 13C NMR were made using
the APT pulse sequence and HSQC and HMBC tech-
niques. GC-MS analyses were performed on a Shimadzu
QP5050A instrument. Infrared spectra were recorded on
a Varian Scimitar FTS 1000 spectrophotometer. Elemen-
tal analyses and mass spectra were recorded in the ana-
lytical laboratories of Milan University.
The collected analytical data for N-benzyl-4-
nitrobenzenamine, 1b [19]; N-benzylidene-4-nitro-
benzenamine [19], 3,5-bis(trifluoromethyl)-N-(1-phenyl-
ethyl)benzenamine, 2a [19]; 4-nitro-N-(1-phenylethyl)
benzenamine, 2b [35]; 3,5-bis(trifluoromethyl)-N-(2-
methyl-1-phenylpropyl)benzenamine, 3a [30]; N-benz-
hydryl-4-nitrobenzenamine, 4b [19]; 3,5-bis(trifluoro-
methyl)-N-(2-phenylpropan-2-yl)benzenamine, 5a [30];
4-nitro-N-(2-phenylpropan-2-yl)benzenamine, 5b [19]
are in agreement with those reported in the literature.
The reaction of ArN3 with hydrocarbons containing a
benzylic group, Ar′RR1CH, initially produces the corre-
sponding amines Ar′RR1C-NHAr. When either R or R′ is
hydrogen, the catalytic reaction always proceeds further,
giving the corresponding imine Ar′RC=NAr by reaction
with another arylazide molecule. The corresponding ani-
line is the by-product of the imine formation. A mecha-
nistic study of the cobalt porphyrins-catalyzed reaction
evidenced a reversible coordination of the arylazide to
the metal, affording an unstable Co(II) adduct that reacts
with the hydrocarbon forming the aminated product [19].
Kinetic and spectroscopic data excluded the formation of
a “formal” cobalt imido complex, whereas imido inter-
mediates are usually proposed for ruthenium porphyrins-
catalyzed amination reactions [27–29].
Recently, we reported a complete characterization of
Ru(TPP)(NAr)2 (TPP = dianion of tetraphenylporphyrin,
Ar = 3,5-(CF3)2C6H3) and showed that it inserts a nitrene
group into benzylic C-H bonds under both stoichiometric
and catalytic conditions, strongly suggesting that it is the
key intermediate in the ruthenium porphyrin-catalyzed
amination of hydrocarbons by aryl azides [30]. However,
employing such a sensitive and difficult to isolate com-
plex as catalyst is not practical. In this paper, we report
on the use of the much more easily synthesized and even
commercially available Ru(TPP)CO as catalyst for the
amination of benzylic substrates. Results acquired up to
now indicate a different chemoselectivity with respect
to the cobalt-catalyzed reactions. In fact, in the presence
of ruthenium porphyrins, benzylic amines have been
obtained starting from benzylic hydrocarbons that nor-
mally afforded the corresponding imines in the presence
of cobalt porphyrins. The results of a preliminary mecha-
nistic study are also reported.
Synthesis
General procedure for catalytic reactions. Method A.
In a typical run, Ru(TPP)CO (36.0 mg, 4.85 × 10-2 mmol)
and the azide (0.606 mmol) were dissolved into the hydro-
carbon (30 mL). The reaction solution was then heated at
100 °C using a preheated oil bath. The consumption of
the azide was monitored by TLC up to the point where its
spot was no longer observable, and then by IR spectros-
copy measuring the characteristic azide absorbance in the
region 2095–2130 cm-1. The reaction was considered to
be finished when the absorbance of the latter peak was
below 0.03 (using a 0.5 mm thick cell). The solution was
then concentrated to dryness and the residue was purified
by flash chromatography using n-hexane/ethyl acetate =
9:1 as eluent mixture or analyzed by 1H NMR with 2,4-
dinitrotoluene as an internal standard. All reaction times
and products yields are reported in Table 1. Method B. The
amination reaction was repeated refluxing Ru(TPP)CO
(10.0 mg, 1.35 × 10-2 mmol) and the azide (0.674 mmol)
in the hydrocarbon (30 mL). The consumption of the
azide was monitored by TLC up to the point where its
spot was no longer observable, and then by IR spectros-
copy measuring the characteristic azide absorbance in the
EXPERIMENTAL
General
Unless otherwise specified, all the reactions were car-
ried out in a nitrogen atmosphere employing standard
Schlenk techniques and magnetic stirring. Benzene, tolu-
ene and isopropyl benzene were distilled over sodium-
benzophenone and kept under nitrogen. All other starting
materials were commercial products and were used as
Copyright © 2010 World Scientific Publishing Company
J. Porphyrins Phthalocyanines 2010; 14: 733–740

734
D. IntRIeRI et al.
Table 1. Ru(TPP)CO-catalyzed benzylic amination at 100 °Ca
Entry
1
Product
NHAr
Ar
Time, minb
Yield, %
3,5(CF3)2C6H3, 1a
4(NO2)C6H4, 1b
50
60
14c
25d
2
3
3,5(CF3)2C6H3, 2a
4(NO2)C6H4, 2b
60
31c
80d
180
NHAr
67c
57c
10d
31d
48d
NHAr
3,5(CF3)2C6H3, 3a
4(NO2)C6H4, 3b
4(tBu)C6H4, 3c
2,6(NO2)2C6H3, 3d
4-BrC6H4, 3e
30
90
70
25
70
65c
55d
Ph
4
3,5(CF3)2C6H3, 4a
4(NO2)C6H4, 4b
25
60
NHAr
5
6
3,5(CF3)2C6H3, 5a
4(NO2)C6H4, 5b
25
50
31c
41d
NHAr
NHAr
3,5(CF3)2C6H3, 6a
4(NO2)C6H4, 6b
30
90
53c
54c
60c
50c
NHAr
7
3,5(CF3)2C6H3, 7a
4(NO2)C6H4, 7b
25
120
a General procedure for catalytic reactions: Ru(TPP)CO (36 mg, 4.85 × 10-2 mmol) in the hydrocar-
bon as the reaction solvent (30 mL) at 100 °C; mol ratio Ru(TPP)CO/ArN3 = 4:50. b Time required
c
d
1
to reach complete conversion of the starting ArN3. Isolated yield. Determined by H NMR
employing 2,4-dinitrotoluene as internal standard.
region 2095–2130 cm-1. The reaction was considered to
(d, 2H, J = 9.20 Hz, HAr), 7.25–7.37 (m, 5H, HAr), 6.48
(d, 2H, J = 9.20 Hz, HAr), 4.89 (broad, 1H, NH), 4.23 (d,
1H, J = 6.24 Hz, CHNH), 2.05–2.19 (m, 1H, CH(CH3)2),
1.04 (d, 3H, J = 6.78 Hz, CH3), 0.95 (d, 3H, J = 6.78
Hz, CH3). 13C NMR (CDCl3, 75 MHz): δ, ppm 153.1 (C),
141.0 (C), 138.5 (C), 129.0 (CH, 2C overlapping), 128.0
(CH), 127.3 (CH, 2C overlapping), 126.6 (CH), 112.2
(CH), 64.0 (CH), 35.0 (CH), 20.0 (CH3), 19.1 (CH3). MS
(EI): m/z 270 [M]+.
be finished when the absorbance of the azide measured
was below 0.03 (using a 0.5 mm thick cell). The solu-
tion was then concentrated to dryness and the residue was
analyzed by 1H NMR with 2,4-dinitrotoluene as an inter-
nal standard. All the reaction times and products yields
are reported in Table 2.
Analytical data for N-benzyl-3,5-bis(trifluoro-
methyl)benzenamine, 1a. Anal. calcd. for C15H11F6N
(319.08): C, 56.43; H, 3.47; N, 4.39. Found: C, 56.70;
Analytical data for N-(2-methyl-1-phenylpropyl)-
4-tert-butylbenzenamine, 3c. Anal. calcd. for C20H27N
(281.44) C, 85.35; H, 9.67; N, 4.98. Found: C, 85.21;
1
H, 3.50; N, 4.27. H NMR (CDCl3, 300 MHz): δ, ppm
7.31–7.43 (m, 5H, HAr), 7.19 (s, 1H, HAr), 6.99 (s, 2H,
HAr), 4.56 (broad, 1H, NH), 4.39 (s, 1H, CH2). 13C NMR
(CDCl3, 75 MHz): δ, ppm 149.0 (C), 138.0 (C), 132.9 (q,
J = 33.0 Hz, C-CF3, 2C overlapping), 129.3 (CH), 128.3
(CH), 127.9 (CH), 123.9 (q, J = 271.5 Hz, CF3, 2C over-
lapping), 112.3 (CH), 110.8 (broad, CH), 48.4 (CH2). 19F
NMR (CDCl3, 282 MHz): δ, ppm -63.53. MS (EI): m/z
319 [M]+.
1
H, 9.74; N, 5.23. H NMR (CDCl3, 300 MHz): δ, ppm
7.28–7.33 (m, 5H, HAr), 7.11 (d, 2H, J = 7.12 Hz, HAr),
6.51 (d, 2H, J = 7.12 Hz, HAr), 4.09 (broad, 1H, CHNH),
4.07 (broad, 1H, NH), 2.01 (m, 1H, CH(CH3)2), 1.24 (s,
t
9H, Bu), 1.00 (d, 3H, J = 6.78 Hz, CH3), 0.92 (d, 3H,
J = 6.78 Hz, CH3). 13C NMR (CDCl3, 75 MHz): δ, ppm
128.1 (CH, 2C overlapping), 127.3 (CH, 2C overlap-
ping), 126.8 (CH), 126.0 (CH, 2C overlapping), 110.0
Analytical data for N-(2-methyl-1-phenylpropyl)-
4-nitrobenzenamine, 3b. Anal. calcd. for C16H18N2O2
(270.33) C, 71.09; H, 6.71; N, 10.36. Found: C, 71.32; H,
6.95; N, 10.54. 1H NMR (CDCl3, 300 MHz): δ, ppm 8.00
t
(CH, 2C overlapping), 31.5 (CH3, 3C overlapping, Bu),
19.7 (CH3), 18.7 (CH3), quaternary carbons and two CH
were not detected.
Copyright © 2010 World Scientific Publishing Company
J. Porphyrins Phthalocyanines 2010; 14: 734–740

RuthenIum PORPhyRInS-CatalyzeD atOm-effICIent amInatIOn Of C-h bOnDS by aRylazIDeS
Table 2. Ru(TPP)CO-catalyzed benzylic amination in refluxing hydrocarbona
735
Entry
1
Product
Ar
Time, minb
Yield, %c
3,5(CF3)2C6H3, 1a
4(NO2)C6H4, 1b
60
90
66
25
NHAr
2
3
3,5(CF3)2C6H3, 2a
4(NO2)C6H4, 2b
240
300
85
20
NHAr
NHAr
3,5(CF3)2C6H3, 3a
4(NO2)C6H4, 3b
60
45
75
40
4
5
3,5(CF3)2C6H3, 5a
4(NO2)C6H4, 5b
69
75
90
30
NHAr
NHAr
3,5(CF3)2C6H3, 6a
4(NO2)C6H4, 6b
25
60
80
30
a General procedure for catalytic reactions: Ru(TPP)CO (10 mg, 1.35 × 10-2 mmol) in refluxing
hydrocarbon (30 mL); mol ratios Ru(TPP)CO/ArN3 = 1:50. b Time required to reach complete
c
1
conversion of the starting ArN3. Determined by H NMR employing 2,4-dinitrotoluene as
internal standard.
Analytical data for 2,6-bis(nitro)-N-(2-methyl-
1-phenylpropyl)benzenamine, 3d. Anal. calcd. for
C16H17N3O4 (315.32) C, 60.94; H, 5.43; N, 13.33. Found:
C, 61.26; H, 5.69; N, 13.55. 1H NMR (CDCl3, 300 MHz):
δ, ppm 8.56 (d, 2H, J = 8.3 Hz, HAr), 7.71–7.74 (m, 2H,
HAr), 7.53–7.57 (m, 3H, HAr), 6.80 (t, 1H, J = 8.3 Hz, HAr),
4.24 (m, 2H, NH e CHNH), 1.69 (m, 1H, CH(CH3)2),
0.93 (m, 6H, (CH3)2). 13C NMR (CDCl3, 75 MHz): δ,
ppm 168.1 (C), 134.5 (CH, 2C overlapping), 131.2 (CH,
3C overlapping), 129.2 (CH, 2C overlapping), 114.3
(CH), 68.5 (CH), 39.1 (CH), 14.4 (CH3), 11.3 (CH3). Two
quaternary carbons were not detected.
overlapping), 129.1 (CH), 128.0 (CH), 127.3 (CH), 123.5
(q, J = 271.1 Hz, CF3, 2C overlapping), 112.7 (CH), 110.5
(CH), 62.8 (CH). 19F NMR (CDCl3, 282 MHz): δ, ppm
-63.53. MS (EI): m/z 395 [M]+.
Analytical data for N-(3,5-bis(trifluoromethyl)
phenyl)-2,3-dihydro-1H-inden-1-amine, 6a. Anal. calcd.
for C17H13F6N (345.10): C, 59.13; H, 3.79; N, 4.06.
1
Found: C, 59.02; H, 3.90; N, 4.17. H NMR (CDCl3,
300 MHz): δ, ppm 7.28–7.38 (m, 4H, HAr), 7.21 (s,
1H, HAr), 7.07 (s, 2H, HAr), 5.08 (m, 1H, CHNH), 4.39
(broad, 1H, NH), 2.95–3.09 (m, 2H, CH2), 2.62–2.70 (m,
1H, NHCHH), 1.92–2.00 (m, 1H, NHCHH). 13C NMR
(CDCl3, 75 MHz): δ, ppm 144.0 (C), 133.0 (q, J = 32.4
Hz, C-CF3, 2C overlapping), 128.9 (CH), 127.4 (CH),
125.5 (CH), 124.5 (CH), 123.9 (q, J = 271.1 Hz, CF3),
112.6 (CH), 110.6 (CH), 58.9 (CH), 33.8 (CH2), 30.7
(CH2). 19F NMR (CDCl3, 282 MHz): δ, ppm -63.50. MS
(EI): m/z 345 [M]+.
Analytical data for N-(2-methyl-1-phenylpropyl)-
4-bromobenzenamine, 3e. Anal. calcd. for C16H18BrN
(304.22) C, 63.17; H, 5.96; N, 4.60. Found: C, 63.51;
1
H, 6.06; N, 4.82. H NMR (CDCl3, 300 MHz): δ, ppm
7.26–7.33 (m, 5H, HAr), 7.14 (d, 2H, J = 8.8 Hz, HAr),
6.40 (d, 2H, J = 8.8 Hz, HAr), 4.10 (broad, 1H, NH), 4.08
(m, 1H, CHNH), 2.06 (m, 1H, CH(CH3)2), 1.00 (d, 3H,
J = 6.8 Hz, CH3), 0.93 (d, 3H, J = 6.8 Hz, CH3). 13C NMR
(CDCl3, 75 MHz): δ, ppm 131.7 (CH, 2C overlapping),
128.3 (CH, 3C overlapping), 127.5 (CH, 2C overlapping),
115.0 (CH, 2C overlapping), 55.3 (CH), 34.7 (CH), 19.6
(CH3), 18.7 (CH3). Three quaternary carbons were not
detected. MS (EI): m/z 304 [M]+.
Analytical data for 2,3-dihydro-N-(4-nitrophenyl)-
1H-inden-1-amine, 6b. Anal. calcd. for C15H14N2O2
(245.11): C, 70.85; H, 5.55; N, 11.02. Found: C, 70.52;
1
H, 5.89; N, 11.45. H NMR (CDCl3, 300 MHz): δ, ppm
8.15 (d, 2H, J = 9.16 Hz, HAr), 7.27–7.38 (m, 4H, HAr),
6.67 (d, 2H, J = 9.16 Hz, HAr), 5.10–5.15 (m, 1H, CHNH),
4.74 (m, 1H, NH), 3.06–3.11 (m, 1H, CHH), 2.96–3.02
(m, 1H, CHH), 2.63–2.68 (m, 1H, NHCHH), 1.96–2.02
(m, 1H, NHCHH). 13C NMR (CDCl3, 75 MHz): δ, ppm
152.7 (C), 143.6 (C), 142.7 (C), 138.2 (C), 128.6 (CH),
127.0 (CH), 126.5 (CH, 2C overlapping), 125.2 (CH),
124.1 (CH), 111.5 (CH, 2C overlapping), 58.3 (CH),
33.5 (CH2), 30.2 (CH2). MS (EI): m/z 254 [M]+.
Analytical data for N-benzhydryl-3,5-bis(trifluoro-
methyl)benzenamine, 4a. Anal. calcd. for C21H15F6N
(395.11): C, 63.80; H, 3.82; N, 3.54. Found: C, 64.02;
1
H, 3.90; N, 3.40. H NMR (CDCl3, 400 MHz): δ, ppm
7.31–7.37 (m, 6H, HAr), 7.22–7.26 (m, 4H, HAr), 7.19 (s,
1H, HAr), 6.93 (s, 2H, HAr), 5.60 (m, 1H, CHNH), 4.72
(broad, 1H, NH). 13C NMR (CDCl3, 100 MHz): δ, ppm
147.7 (C), 141.3 (C), 132.1 (q, J = 32.7 Hz, C-CF3, 2C
Analytical data for N-(3,5-bis(trifluoromethyl)
phenyl)-1,2,3,4-tetrahydronaphthalen-1-amine, 7a. Anal.
Copyright © 2010 World Scientific Publishing Company
J. Porphyrins Phthalocyanines 2010; 14: 735–740

736
D. IntRIeRI et al.
R'
calcd. for C18H15F6N (359.11): C, 60.17; H, 4.21; N, 3.90.
Found: C, 59.92; H, 3.90; N, 4.17. H NMR (CDCl3,
R
1
H
R'
NHAr
R
300 MHz): δ, ppm 7.32–7.36 (m, 1H, HAr), 7.21–7.26 (m,
2H, HAr), 7.16–7.20 (m, 1H, HAr), 7.02 (s, 2H, HAr), 4.68–
4.72 (m, 1H, CHNH), 4.35 (broad, 1H, NH), 2.80–2.90
(m, 2H, CHNHCH2CH2CH2), 1.98–2.06 (m, 1H, CHN-
HCH2CH2CH2), 1.86–1.95 (m, 1H, CHNHCH2CH2CH2).
13C NMR (CDCl3, 75 MHz): δ, ppm 148.2 (C), 138.1 (C),
137.0 (C), 132.0 (q, J = 32.4 Hz, C-CF3, 2C overlapping),
129.8 (CH), 129.4 (CH), 128.1 (CH), 126.8 (CH), 124.0
(q, J = 271.0 Hz, CF3, 2C overlapping), 112.2 (CH, 2C
overlapping), 110.4 (CH), 51.5 (CH), 29.5 (CH2), 28.8
(CH2), 19.7 (CH2). 19F NMR (CDCl3, 282 MHz): δ, ppm
-63.50. MS (EI): m/z 359 [M]+.
ArN3
[Ru]
H
1, R = R' = H
2, R = H; R' = CH3
3, R = H; R' = iPr
4, R = H; R' = Ph
5, R = R' = CH3
[Ru]
NHAr
n
n
6, n = 1
7, n = 2
[Ru] = Ru(TPP)CO
Scheme 2. Amination reaction of benzylic hydrocarbons by
Analyticaldatafor1,2,3,4-tetrahydro-N-(4-nitrophe-
nyl)naphthalen-1-amine, 7b. Anal. calcd. for C16H16N2O2
(268.31): C, 71.62; H, 6.01; N, 10.44. Found: C, 71.44; H,
5.89; N, 10.65. 1H NMR (CDCl3, 300 MHz): δ, ppm 8.13
(d, 2H, J = 9.00 Hz, HAr), 7.00–7.18 (m, 4H, HAr), 5.91 (d,
2H, J = 9.00 Hz, HAr), 4.18 (broad, 1H, CHNH), 3.77 (d,
1H, J = 7.90 Hz, NH), 2.48–2.63 (m, 2H, CH2), 1.42–1.49
(m, 4H, CH2). 13C NMR (CDCl3, 75 MHz): δ, ppm 162.7
(C), 152.7 (C), 138.1 (C), 136.5 (C), 129.8 (CH), 129.3
(CH), 128.2 (CH), 127.0 (CH), 126.8 (CH, 2C overlap-
ping), 111.5 (CH, 2C overlapping), 51.4 (CH), 29.5 (CH2),
29.1 (CH2), 19.8 (CH2). MS (EI): m/z 268 [M]+.
ArN3 catalyzed by Ru(TPP)CO
Benzylic substrates containing an exocyclic benzylic
carbon (Table 1, entries 1–5) were reacted with at least
two different arylazides, 3,5(CF3)2C6H3N3 and 4(NO2)
C6H4N3. Data reported in Table 1 reveals similar or even
better yields but always longer reaction times for the reac-
tions employing 4(NO2)C6H4N3 as the aminating reagent.
It is worth noting that only in the case of the amination of
toluene (Table 1, entry 1) was the formation of the imine,
N-benzylidene-4-nitrobenzenamine or N-benzylidene-3,
5-bis-trifluoromethylbenzenamine, deriving from the
reaction of 1a,b with another molecule of azide observed
by GC-MS. In all other cases, the aniline deriving from
the starting arylazide was the only by-product of the reac-
tions. On the other hand, when a substituent was present
on the benzylic position (Table 1, entries 2, 3, 4) any fur-
ther reaction of the benzylic amine with arylazide was
inhibited by the increase in the steric hindrance on the
benzylic hydrogen atom (Scheme 3).
The formation of imine starting from isopropyl ben-
zene is impossible due to the double substitution on the
C-H benzylic bond, but the yield of amines 5a and 5b are
not satisfactory because of the contemporary formation
of the anilines deriving from the corresponding azides.
Isobutylbenzene was also aminated employing 4(tBu)
C6H4N3, 2,6(NO2)2C6H3N3 and 4(Br)C6H4N3 (Table 1,
entry 3).
Kinetic measurements
To a Schlenk flask under a dinitrogen atmosphere were
added the catalyst, 4(NO2)C6H4N3 and toluene in this
order. When required, benzene was added at this point.
The flask was capped with a rubber septum and immedi-
ately placed in an oil bath preheated at 75 °C. The solu-
tion was stirred for one minute to dissolve all reagents
and 0.2 mL solution was withdrawn for IR analysis.
The consumption of the azide was followed by IR
spectroscopy (νN3 = 2120 cm-1) by withdrawing samples
of the solution at regular times. Since all reactions are
run in the presence of a large excess of toluene, the
apparent first-order constants were fitted to the equation
-d[ArN3]/dt = kapp[Ru(TTP)CO][ArN3]. The concentra-
tion of Ru(TPP)CO was calculated on the exact amount
of catalyst weighed in each run and was considered to
remain constant during the reaction.
At the moment we do not have enough data to give
a definite explanation for the observed trends in the
R
R
R'
RESULTS AND DISCUSSION
+ ArN3, -N2
Ru(TPP)CO
H
N
H
Synthetic results
+ ArN3
- ArNH2, -N2
In order to compare the chemoselectivity of the ben-
zylic amination catalyzed by Ru(TPP)CO with that
catalyzed by Co(porphyrin) complexes, several ben-
zylic substrates were reacted at 100 °C with arylazides
employing the hydrocarbon as reaction solvent and the
catalytic ratio Ru(TPP)CO/azide = 4:50 (Scheme 2).
Collected data is reported in Table 1.
Ru(TPP)CO
R
N
R''
Scheme 3. Ru(TPP)CO-catalyzed benzylic imine formations
Copyright © 2010 World Scientific Publishing Company
J. Porphyrins Phthalocyanines 2010; 14: 736–740

RuthenIum PORPhyRInS-CatalyzeD atOm-effICIent amInatIOn Of C-h bOnDS by aRylazIDeS
737
selectivity towards the benzylic amine, imine, or aniline
In order to improve the catalytic efficiency, the hydro-
carbon aminations were repeated lowering the catalytic
ratio Ru(TPP)CO/azide from 4:50 to 1:50 and running the
reactions at higher temperature. The reactions described
in Table 1 were performed with the refluxing hydrocar-
bon as the reaction solvent. The amination of diphenyl
methane and 1,2,3,4-tetrahydronaphthalene were not car-
ried out as the boiling points of those hydrocarbons are
very high (264 °C and 207 °C, respectively). Data col-
lected is reported in Table 2.
Comparing these data with those in Table 1, an
opposite behavior was observed when employing
3,5(CF3)2C6H3N3 or 4(NO2)C6H4N3 as nitrene sources.
The amination by 3,5(CF3)2C6H3N3 occurred with a
general improvement of the yields. However, the reac-
tions performed with 4(NO2)C6H4N3 formed the corre-
sponding benzylic amine in lower yields with respect to
those reported in Table 1. This result can be explained by
the different thermal stability of these two aryl azides.
Indeed, with the exception of toluene, the boiling points
of all tested hydrocarbons is always higher than 130 °C.
The presence of two CF3 groups on the aryl moiety of
3,5(CF3)2C6H3N3 conferred it a better stability than
that of 4(NO2)C6H4N3 and a thermal decomposition of
3,5(CF3)2C6H3N3 at T > 130 °C was avoided. Consider-
ing the different catalyst loading for the benzylic amina-
tions of Tables 1 and 2, a comparison between reaction
times is impossible.
product. However, the following arguments can account
for the observed results: 1) Our previous studies on the
cobalt-porphyrin catalyzed benzylic amination of hydro-
carbons by arylazides [19] unequivocally showed that the
activation of the C-H bond proceeds by an intermediate
having a marked radical character on the benzylic carbon.
Formation of such a radical is strongly accelerated by the
presence of an amino group in the α position. The cobalt
catalyst is not very active and thus the reactive intermedi-
ate should be able to select the most reactive substrate in
solution (the just-formed amine) with respect to the less
reactive one (the starting hydrocarbon) despite the latter
being present in a much larger concentration, being the
reaction solvent. Even a preliminary examination of the
reaction rates shows the ruthenium-tetraphenylporphyrin
catalyst to be much more active. This implies that the
reactive intermediate will react with the first suitable sub-
strate it will meet, most likely the hydrocarbon, allowing
the accumulation of the amine in solution. 2) As for the
cobalt-catalyzed reaction, the reaction is favored by elec-
tron-withdrawing substituents on the azide aryl ring. In
fact, the best yield was achieved performing the reaction
with 3,5(CF3)2C6H3N3 and only 10% of benzylic amine
was obtained when a tert-butyl group was present in the
para position of the aryl group of the azide. Conversely,
steric hindrance disfavors the amination reaction and
the reaction of isobutylbenzene with 2,6(NO2)2C6H3N3
afforded the corresponding amine only in low yield. The
rate of the amination reaction should not be confused with
the rate of azide consumption, which is very high for the
dinitro-substituted azide. Indeed, formation of the active
intermediate by reaction of the azide with the ruthenium
porphyrin complex or hydrogen abstraction of the latter
to generate aniline should be much less sensitive to steric
effects than a coupling reaction to generate a new C-N
bond. Thus, the azide consumption rate remains high, but
the main product is dinitroaniline. 3) Steric hindrance
surely also affects the reaction of the initially formed
benzylic amine with another azide molecule. Thus, it is
not surprising that the only case in which imine forma-
tion was detected with the ruthenium catalyst involves
toluene as the substrate. The corresponding benzylic
amines are the only ones to have a secondary benzylic
carbon atom, whereas a tertiary benzylic carbon atom is
formed in all other cases.
Imine compounds were never formed and, in all the
cases investigated, the reaction by-product was the ani-
line deriving from the aryl azide decomposition.
For each amination by 3,5(CF3)2C6H3N3 shown in
Table 2 the yield was higher than 66%, showing that the
positive effect of raising the temperature prevails on the
lower catalyst loading. In fact, the best yield of 90% has
been obtained using isopropylbenzene, which has the
highest boiling point among the tested hydrocarbons.
Preliminary mechanistic study
As already pointed out in the introduction, we have
recently reported that the species active in the benzylic
and allylic aminations catalyzed by Ru(TPP)CO is
the bis-imido complex Ru(TPP)(NAr)2 formed by the
reaction of the pre-catalyst Ru(TPP)CO with two mol-
ecules of 3,5(CF3)2C6H3N3 [30] (Scheme 4). Complex
Ru(TPP)(NAr)2 displayed a good activity for catalytic
TheRu(TPP)COcatalystisalsoactivefortheamination
of substrates containing an endocyclic benzylic
C-H bond (Table 1, entries 6 and 7) and the cor-
responding amines have been isolated in good
yields. It must be underlined that compound 7b,
deriving from the reaction of tetrahydronaphtha-
lene with 4(NO2)C6H4N3, has a pharmaceutical
interest, being a drug that, antagonizing nico-
tinic acetylcholine receptors (nAChRs), is used
to reduce the addiction of nicotine self-adminis-
tration by negative allosteric modulation [36].
NAr
Ph
Ph
N
Ph
Ph
N
N
+ ArN3
Ru
Ru
N
N
N
-N2, -CO
N
N
Ph
Ph
70%
Ph
Ph
CO
NAr
Ar = 3,5(CF3)2C6H3
Scheme 4. Synthesis of Ru(TPP)(NAr)2
Copyright © 2010 World Scientific Publishing Company
J. Porphyrins Phthalocyanines 2010; 14: 737–740

738
D. IntRIeRI et al.
nitrene transfer reactions to hydrocarbons and aminated
product yields up to 99% were obtained.
is approximately first-order in toluene concentration up
to a 7 M concentration, but an inhibiting effect of higher
toluene concentrations is evident.
To better investigate the inhibition effect observed in
the benzylic amination at high toluene concentrations, we
have repeated the reaction with benzene/toluene ratios of
75:25, 50:50 and neat toluene, analyzing the composition
of the reaction mixture by GC-MS. As shown in Fig. 2,
the increase in toluene concentration caused an increase
in the imine/amine ratio.
In the case of the allylic amination of cyclohexene, we
have already reported that the reaction of Ru(TPP)(NAr)2
with cyclohexene yielded an amido derivative that is
still catalytically active. The X-ray molecular structure
of the bis-amido complex Ru(TPP)(ArNC6H9)2 (Ar =
3,5(CF3)2C6H3) was solved. A kinetic study of the allylic
amination by arylazides catalyzed by ruthenium porphy-
rin complexes is under investigation.
To investigate the mechanism of the benzylic ami-
nation catalyzed by Ru(TPP)CO, a kinetic study was
undertaken employing 4-nitrophenylazide and toluene
as substrates. The reaction was followed by IR spectros-
copy, monitoring the intensity of the 2120 cm-1 absorp-
tion of the arylazide. All kinetic experiments in this paper
were run at 75 °C to avoid boiling of benzene in those
reaction in which it was present. The dependence of the
reaction rate with respect to the toluene concentration has
been investigated employing a catalytic ratio Ru(TPP)
CO/arylazide = 4:50.
The kinetic and selectivity data together indicate that
the reaction pathway is complex and not exactly the
same as for the corresponding allylic amination reaction.
A more in-depth study will be necessary to clarify all
aspects of this interesting reaction. The presently avail-
able data is consistent with an active role of a ruthenium-
imido intermediate, which is formed in a slow step by
reaction of a ruthenium complex with the azide. However,
the complex kinetic behavior with respect to toluene con-
centration clearly indicates that more than one reaction
pathway is available to this intermediate. The formation
of the benzylic amine obviously requires the interaction
of the intermediate with toluene and this process can
account for the increasing branch of the rate vs. [toluene]
plot (Fig. 1c). On the other hand, we must assume that
The reaction rate showed a first-order dependence on
azide and [Ru(TPP)CO] concentrations. However, an
irregular dependence on toluene concentration (the other
solvent being benzene) was observed (Fig. 1). The rate
Fig. 1. Kinetic dependence of the reaction rate with respect to (a) 4-nitro-phenylazide, (b) Ru(TPP)CO and (c) toluene
concentrations
Copyright © 2010 World Scientific Publishing Company
J. Porphyrins Phthalocyanines 2010; 14: 738–740

RuthenIum PORPhyRInS-CatalyzeD atOm-effICIent amInatIOn Of C-h bOnDS by aRylazIDeS
739
Fig. 2. GC-MS spectra of reactions run with benzene/toluene ratios of (a) 75:25, (b) 50:50 and (c) neat toluene. A = 4(NO2)C6H4NH2,
B = 4(NO2)C6H4N=CHC6H5, C = 4(NO2)C6H4NH-CH2C6H5
a different type of interaction between the intermediate
and toluene which leads to a less catalytically active spe-
cies can also occur. The formation of a less active species
explains both the slower rate at high toluene concentra-
tion and the increased amount of imine formed under
these conditions. Indeed, the same situation encountered
with the less active cobalt-porphyrin catalysts discussed
above would result. A less reactive intermediate is able
to select the most reactive partner among those present
in solution, even if the concentration of the benzylic
amine is relatively low with respect to that of a less reac-
tive partner such as the starting hydrocarbon. It should be
noted that formation of a completely inactive complex
would explain the rate drop, but not the change in selec-
tivity. The available data are insufficient to identify this
less active intermediate and speculating on its identity at
this stage is premature. Attempts are in progress to isolate
and characterise this complex, so that its reactivity and
kinetic behavior can be investigated.
Moreover, the preliminary mechanistic study per-
formed indicates an important role of the hydrocar-
bon concentration to drive the chemoselectivity of the
reaction.
Acknowledgements
We thank the MiUR (Programmi di Ricerca
Scientifica di Rilevante Interesse Nazionale PRIN
2007HMTJWP_004) for financial support and Metalor
Technologies SA for a generous supply of ruthenium
trichloride.
REFERENCES
1. Dux E. Chem. Rev. (Deddington, U. K.) 2009; 19:
11–15.
2. Lewis JC, Bergman RG and Ellman JA. Acc. Chem.
Res. 2008; 41: 1013–1025.
3. Davies HML and Manning JR. Nature 2008; 451:
417–424.
4. Diaz-Requejo MM and Perez PJ. Chem. Rev. 2008;
108: 3379–3394.
CONCLUSION
In conclusion, we have reported on the catalytic
activity of Ru(TPP)CO in the amination of benzylic
substrates, demonstrating that the treatment of the
same benzylic substrate with an aromatic azide, in the
presence of ruthenium or cobalt porphyrin complexes
produces different products. Therefore, these two meth-
odologies are complementary because they can be
alternately used, depending on the desired aminated
product.
5. Mueller P and Fruit C. Chem. Rev. 2003; 103:
2905–2919.
6. Cenini S, Gallo E, Caselli A, Ragaini F, Fantauzzi
S and Piangiolino C. Coord. Chem. Rev. 2006; 250:
1234–1253.
7. Henery-Logan KR and Clark RA. Tetrahedron Lett.
1968; 801–806.
8. Broeckx W, Overbergh N, Samyn C, Smets G and
L’Abbe G. Tetrahedron 1971; 27: 3527–3534.
Copyright © 2010 World Scientific Publishing Company
J. Porphyrins Phthalocyanines 2010; 14: 739–740

740
D. IntRIeRI et al.
9. Braese S, Gil C, Knepper K and Zimmer-
mann V. Angew. Chem., Int. Ed. Engl. 2005; 44:
5188–5240.
24. Gallo E, Buonomenna MG, Vigano L, Ragaini F,
Caselli A, Fantauzzi S, Cenini S and Drioli E. J.
Mol. Catal. A 2008; 282: 85–91.
10. Minozzi M, Nanni D and Spagnolo P. Chem.–Eur. J.
2009; 15: 7830–7840.
11. Shou WG, Li J, Guo T, Lin Z and Jia G. Organome-
tallics 2009; 28: 6847–6854.
12. Stokes BJ, Jovanovic B, Dong HJ, Richert KJ,
Riell RD and Driver TG. J. Org. Chem. 2009; 74:
3225–3228.
13. Shen M, Leslie Brooke E and Driver TG. Angew.
Chem., Int. Ed. Engl. 2008; 47: 5056–5059.
14. Stokes BJ, Dong H, Leslie BE, Pumphrey AL and
Driver TG. J. Am. Chem. Soc. 2007; 129: 7500–7501.
15. Sun K, Sachwani R, Richert KJ and Driver TG. Org.
Lett. 2009; 11: 3598–3601.
16. Shen M and Driver TG. Org. Lett. 2008; 10:
3367–3370.
17. Ruppel JV, Kamble RM and Zhang XP. Org. Lett.
2007; 9: 4889–4892.
18. Lu H, Tao J, Jones JE, Wojtas L and Zhang XP. Org.
Lett. 12: 1248–1251.
19. Ragaini F, Penoni A, Gallo E, Tollari S, Gotti CL,
Lapadula M, Mangioni E and Cenini S. Chem.–Eur.
J. 2003; 9: 249–259.
25. Caselli A, Gallo E, Fantauzzi S, Morlacchi S,
Ragaini F and Cenini S. Eur. J. Inorg. Chem. 2008;
3009–3019.
26. Fantauzzi S, Gallo E, Caselli A, Piangiolino C,
Ragaini F, Re N and Cenini S. Chem.–Eur. J. 2009;
15: 1241–1251.
27. Leung SK-Y, Tsui W-M, Huang J-S, Che C-M,
Liang J-L and Zhu N. J. Am. Chem. Soc. 2005; 127:
16629–16640.
28. Liang J-L, Yuan S-X, Huang J-S and Che C-M. J.
Org. Chem. 2004; 69: 3610–3619.
29. Au S-M, Fung W-H, Cheng M-C, Che C-M and
Peng S-M. Chem. Commun. 1997; 1655–1656.
30. Fantauzzi S, Gallo E, Caselli A, Ragaini F, Casati
N, Macchi P and Cenini S. Chem. Commun. 2009;
3952–3954.
31. Tanno M, Sueyoshi S and Kamiya S. Chem. Pharm.
Bull. 1982; 30: 3125–3132.
32. Zhu W and Ma D. Chem. Commun. (Cambridge,
U. K.) 2004; 888–889.
33. Adler AD, Longo FR, Finarelli JD, Goldmacher J,
Assour J and Korsakoff L. J. Org. Chem. 1967; 32:
476.
34. Gallo E, CaselliA, Ragaini F, Fantauzzi S, Mascioc-
chi N, Sironi A and Cenini S. Inorg. Chem. 2005;
44: 2039–2049.
20. Cenini S, Gallo E, Penoni A, Ragaini F and Tollari
S. Chem. Commun. 2000; 2265–2266.
21. Lu H-J, Subbarayan V, Tao J-R and Zhang XP.
Organometallics 2010; 29: 389–393.
22. Piangiolino C, Gallo E, Caselli A, Fantauzzi S,
Ragaini F and Cenini S. Eur. J. Org. Chem. 2007;
743–750.
23. Fantauzzi S, Gallo E, Caselli A, Piangiolino C,
Ragaini F and Cenini S. Eur. J. Org. Chem. 2007;
6053–6059.
35. Imamoto T, Iwadate N and Yoshida K. Org. Lett.
2006; 8: 2289–2292.
36. Yoshimura RF, Hogenkamp DJ, Li WY, Tran
MB, Belluzzi JD, Whittemore ER, Leslie FM and
Gee KW. J. Pharmacol. Exp. Ther. 2007; 323:
907–915.
Copyright © 2010 World Scientific Publishing Company
J. Porphyrins Phthalocyanines 2010; 14: 740–740