Thiosemicarbazone copper complexes as competent catalysts for olefin cyclopropanations

Article abstract of DOI:10.1016/j.jorganchem.2012.03.018

New copper complexes of several thiosemicarbazones have been prepared and characterized. All complexes have been prepared by employing Cu (II) acetate hydrate, but analytical and spectroscopical data for the isolated complexes revealed that in most cases

Full text of DOI:10.1016/j.jorganchem.2012.03.018

Journal of Organometallic Chemistry 714 (2012) 94e103
Contents lists available at SciVerse ScienceDirect
Journal of Organometallic Chemistry
journal homepage: www.elsevier.com/locate/jorganchem
Thiosemicarbazone copper complexes as competent catalysts for olefin
cyclopropanations
Nabil S. Youssef a, Ahmed M.A. El-Seidy a, Marco Schiavoni b, Brunilde Castano b, Fabio Ragaini b,
,
Emma Gallo b, Alessandro Caselli b
*
a Inorganic Chemistry Department, National Research Centre, P.O. 12622, Dokki, Giza, Egypt
b Dipartimento di Chimica, Università degli Studi and ISTM-CNR, Via Venezian 21, 20133 Milano, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
New copper complexes of several thiosemicarbazones have been prepared and characterized. All
complexes have been prepared by employing Cu (II) acetate hydrate, but analytical and spectroscopical
data for the isolated complexes revealed that in most cases a reduction to copper (I) occurred. Cyclo-
propanation reactions of several olefins by ethyldiazoacetate (EDA) in the presence of catalytic amounts
of the complexes were examined. The reported results showed that all complexes are competent cata-
lysts for the cyclopropanation reaction of unactivated olefins. Cyclopropanes were obtained in high yields
(up to 97%, TON up to 18,400) with moderate to excellent diastereoselectivities (up to >99%).
Ó 2012 Elsevier B.V. All rights reserved.
Received 13 February 2012
Received in revised form
6 March 2012
Accepted 7 March 2012
Keywords:
Schiff bases
Copper complexes
Cyclopropanation reactions
Homogeneous catalysis
Thiosemicarbazone
1. Introduction
(I) rather than copper (II) was established as the active catalyst,
where diazo compounds were found to reduce Cu (II) salts to Cu (I)
Cyclopropane derivatives are an important family of chemical
compounds that plays a prominent role in organic chemistry [1].
Naturally occurring and synthetic cyclopropanes are endowed with
a rich spectrum of biological properties [2] and a large number of
compounds carrying a cyclopropane ring have been synthesized
and described in the literature. As a result, great efforts have been
made to develop efficient stereoselective methods for the synthesis
of cyclopropanes [3]. A particularly versatile method is the metal-
catalysed cyclopropanation of olefins with diazo compounds, for
which several efficient homogeneous catalysts have been devel-
oped [4]. Nevertheless, the synthesis of these compounds remains
a considerable challenge, especially due to their interesting bio-
logical properties as well as their use as starting materials and
intermediates in organic synthesis. Among transition metal cata-
lysts for this reaction, copper complexes have attracted increasing
interest in the last years [5e15], especially due to their high effi-
ciency and to their lower cost when compared to other metal
derivatives, such as catalysts based on rhodium [16e18] or ruthe-
nium [19e21]. A variety of copper (II) and copper (I) sources are
known to catalyse the cyclopropanation reaction although copper
[22]. Nevertheless Cu (I) complexes are challenging to synthesize
and isolate due to the intrinsic instability of cuprous compounds
[11]: under many conditions disproportionation of Cu (I) to Cu (0)
and Cu (II) is thermodynamically favoured. We have recently
reported that Schiff bases derived from the condensation of
hydrazinecarbothioamide or phenyl thiosemicarbazone with 3-
acetyl-2H-chromen-2-one are suitable ligands for the synthesis of
copper (II) complexes very active as cyclopropanation catalysts
[23,24]. The structure of the thiosemicarbazide moiety confers
a good chelating capacity and the latter can be increased by
employing a suitable aldehyde or ketone for the formation of the
Schiff base possessing a further donor atom to render the ligand
tridentate [25].
We report here that Schiff bases derived from the condensation
reaction of hydrazinecarbothioamide with substituted salicylalde-
hydes are suitable ligands for copper and that the derived
complexes are competent catalysts for the cyclopropanation of
olefins with ethyldiazoacetate (EDA). Rarely metal complexes can
give high selectivities in cyclopropanation reactions together with
high turnover number (TON) [26]. In this work, different ligands
have been synthesized and characterized, by changing the steric
and electronic properties of the starting aldehyde employed in the
condensation reaction.
* Corresponding author. Tel.: þ39 02 50314372; fax: þ39 02 50314405.
E-mail address: alessandro.caselli@unimi.it (A. Caselli).
0022-328X/$ e see front matter Ó 2012 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2012.03.018

N.S. Youssef et al. / Journal of Organometallic Chemistry 714 (2012) 94e103
95
2. Experimental
(195.24g/mol):C, 49.21; H, 4.65; N, 21.52. Found:C, 49.64;H, 4.55;N,
21.51. IR (KBr,
/cmꢀ1): 3441m (nOH), 3318se2987s (nas and nsNH2),
3172s (nOH), 1616s (dNH2), 1603s ( C]N), 1539s, 1265s (nCO), 1061m
C]S), 829m (nsCS), 751s. UV/vis (10ꢀ5 M, DMSO (nm)):
¼ 276, 340.
2 (H2L2): in this case, in order to separate any trace of starting
n
2.1. Materials and methods
n
(n
l
All the reagents employed for the preparation of the ligands and
their complexes were of the highest grade available and used without
further purification. Copper (II) acetate monohydrate was purchased
from Aldrich and used as received. All aldehydes and thio-
semicarbazide were of reagent grade and used as purchased. Thio-
semicarbazones,1e8 and10e11, werepreparedaccordingtoaslightly
modified procedure with respect to what reported in the literature
[27]. 2-Hydroxybenzaldehyde N-ethylthiosemicarbazone (9) was
purchased from SigmaeAldrich. Unless otherwise stated, all catalytic
tests were carried out under an atmosphere of purified dinitrogen
using modified Schlenk techniques. Used solvents were dried and
distilled before use by standard methods. Benzene, cyclohexene and
thiosemicarbazide, a further purification by column chromatog-
raphy was needed (eluant n-hexane:ethyl acetate ¼ 7/3) (yield 80%).
1H NMR (300 MHz, DMSO-d6):
d
¼ 11.33 (s,1H, NH), 9.84 (s,1H, OH),
8.26 (s, 1H, H(7)), 8.02 (br s, 2H, NH2), 7.29 (d, 4J ¼ 2.3 Hz, 1H, H(4)),
7.16 (d, 4J ¼ 2.3 Hz, 1H, H(6)), 1.40 (s, 9H, CH3), 1.27 (s, 9H, CH3). 1H
NMR (300 MHz, CDCl3):
d
¼ 10.17 (s,1H, NH), 9.93 (s,1H, OH), 8.13 (s,
1H, H(7)), 7.44 (s,1H, H(4)), 7.08 (s,1H, H(6)), 6.57 (br s, 2H, NH2),1.45
(s, 9H, CH3),1.32 (s, 9H, CH3).13C NMR (75 MHz, CDCl3):
d
¼ 177.9 (C8),
155.1 (C2),150.4 (C7),142.4 (Cq),137.2 (Cq),128.3 (C4),127.0 (C6),116.3
(Cq), 35.5 (Cq), 34.5 (Cq), 31.8 (CH3), 29.8(CH3). MS (EI): m/z ¼ 307
(Mþ). Anal. Calcd. for C16H25N3OS (307.45 g/mol): C, 62.50; H, 8.20;
1-octene were distilled over sodium, styrene and
a-methylstyrene
N, 13.67. Found: C, 62.38; H, 8.06; N, 13.60. IR (KBr, n
/cmꢀ1): 3452m
were distilled overcalcium hydrideandstored underdinitrogen. NMR
spectra were recorded on an Avance 300-DRX Bruker instrument,
operating at 300 MHz for 1H and at 75 MHz for 13C. Chemical shifts
(ppm) are reported relative toTMS. The 1H NMRsignals of compounds
described in the following have been attributed by COSY and NOESY
techniques. Assignments of the resonances in 13C NMR were made
usingtheAPTpulsesequence,HSQCandHMQCtechniques.Elemental
analysesandmassspectrawererecordedintheanalyticallaboratories
of Milan University. GC analyses were performed on a Shimadzu 2010
FAST-GC equipped with an automatic sampler AOI-20i.
Some of the ligands and theirmetal complexes wereanalysedforC,
H, N and M contents at the Microanalytical Laboratory, Faculty of
Science, Cairo University, Egypt. IR spectra of the ligands and their
metal complexes were measured using KBr discs with a Jasco FT/IR
300E Fourier transform infrared spectrophotometer and/or on a Var-
ian Scimitar FTS 1000 spectrophotometer covering the range
400e4000 cmꢀ1 and in the 500e100 cmꢀ1 regionusingpolyethylene-
sandwiched Nujol mulls on a Perkin Elmer FT-IR 1650 spectropho-
tometer. The electronic spectra of the ligands and their complexes
were obtained in DMSO solutions using an Agilent 8453 UVevisible
recording spectrophotometer. Molar conductivities of the metal
complexes in DMSO (10ꢀ3 M) were measured using a dip cell and
a Bibby conductimeter MC1 at room temperature. The resistance
measured in ohms and the molar conductivities were calculated
(
nOH), 3274e2961s (nas and nsNH2), 3166s (nNH), 1715s (dNH2), 1606s
C]N), 1579m, 1538s, 1290m, 1265m (nCO), 1108m, 1067w ( C]S),
¼ 274, 340.
¼ 11.36 (s,1H, NH),
(n
n
826m (nsCS), 769m. UV/vis (10ꢀ5 M, DMSO (nm)):
l
3 (H2L3): (89%) 1HNMR (300 MHz, DMSO-d6):
d
9.19 (s,1H, OH), 8.39 (s,1H, H7), 8.05 (s,1H, NH2), 7.85 (s,1H, NH2), 7.51
(d, 1H, 3J ¼ 7.5 Hz, H6), 6.96 (dd, 1H, 3J ¼ 7.5 Hz, 4J ¼ 1.3 Hz, H4), 6.77
(pst,1H, 3J ¼ 7.5 Hz, H5), 3.80 (s, 3H, OCH3).13C NMR (75 MHz, DMSO-
d6):
d
¼ 178.6 (C8), 148.7 (C2), 140.7 (C3), 140.4 (C7), 121.6 (C1), 119.9
(C6), 119.0 (C5), 113.7 (C4), 56.8 (CH3). MS (EI): m/z ¼ 225 (Mþ). Anal.
Calcd. for C9H11N3O2S (225.27 g/mol): C, 47.99; H, 4.92; N, 18.65.
Found: C, 47.93; H, 5.02; N, 18.40. IR (KBr,
3343e2975m (nas and nsNH2), 3167m (nNH), 1621m (dNH2), 1598s
C]N),1588m,1536s,1281m,1263s (nCO),1059m ( C]S), 822m (nsCS),
778m. UV/vis (10ꢀ5 M, DMSO (nm)):
¼ 323, 390.
4 (H2L4): (85%) 1H NMR (300 MHz, DMSO-d6):
n
/cmꢀ1): 3461s (nOH),
(n
n
l
d
¼ 11.04 (s, 1H,
NH), 9.52 (s, 1H, OH), 8.18 (s, 1H, H7), 7.81 (s, 1H, NH2), 7.64 (s, 1H,
NH2), 7,50 (d,1H, 3J ¼ 8.8 Hz, H6), 6.20 (dd,1H, 3J ¼ 8.8 Hz, 4J ¼ 2.2 Hz,
H5), 6.08 (d,1H, 4J ¼ 2.2 Hz, H3), 3.30 (q, 4H, 3J ¼ 6.9 Hz, N(CH2CH3)2),
1.09 (t, 6H, 3J ¼ 6.9 Hz, N(CH2CH3)2). 13C NMR (75 MHz, DMSO-d6):
d
¼ 177.3 (C8), 159.0 (C2), 151.0 (C4), 143.3 (C7), 129.9 (C6), 108.2 (C1),
104.8 (C5), 98.2 (C3), 44.7 (CH2), 13.4 (CH3). MS (EI): m/z ¼ 266 (Mþ).
Anal. Calcd. for C12H18N4OS (266.36 g/mol): C, 54.11; H, 6.81; N,
21.03. Found: C, 54.52; H, 6.79; N, 21.20. IR (KBr, n
/cmꢀ1): 3409s
(
nOH), 3302e2960m (nas and nsNH2), 3175m (nNH), 1630s (dNH2), 1610s
C]N),1590s,1553s,1286m,1245s (nCO),1058m ( C]S), 821m (nsCS),
¼ 273, 364.
5 (H2L5): (90%) 1H NMR (300 MHz, DMSO-d6):
according to the equation: L ¼ V ꢁ K ꢁ Mw/g ꢁ
U
, where L, molar
(n
n
conductivity (Uꢀ1 cm2 molꢀ1); V, volumeofthecomplexsolution(ml);
787w. UV/vis (10ꢀ5 M, DMSO (nm)):
l
K, cell constant 0.92 cmꢀ1; Mw, molecular weight of the complex; g,
d
¼ 11.52 (s, 1H,
weight of the complex; and
U, resistance measured in ohms. Magnetic
NH), 10.06 (s, 1H, OH), 8.33 (s, 1H, H(7)), 8.22 (br s, 2H, NH2), 8.06 (d,
moments at 298 K were determined using the Gouy method with Hg
[Co(SCN)4] as calibrant[28]. XPSanalyses werecarried outusinganM-
probe apparatus (Surface Science Instruments).
4J ¼ 2.6 Hz, 1H, H(4)), 7.52 (d, 4J ¼ 2.6 Hz, 1H, H(6)). 13C NMR
(75 MHz, DMSO-d6):
d
¼ 179.0 (C8), 151.4 (C2), 138.5 (C7), 130.5 (C4),
126.0 (Cq), 125.1 (C6), 125.0 (Cq), 123.5 (Cq). MS (EI): m/z ¼ 264 (Mþ).
Anal. Calcd. for C8H7Cl2N3OS (263.13 g/mol): C, 36.38; H, 2.67; N,
2.2. Synthesis of the ligands
15.91. Found: C, 36.54; H, 2.66; N, 15.75. IR (KBr, n
/cmꢀ1): 3464s
(
nOH), 3347s (nasNH2), 3154m (nNH), 1612s (
C]S), 817m (nsCS). UV/vis (10ꢀ5 M, DMSO (nm)):
6 (H2L6): (50%) 1H NMR (300 MHz, DMSO-d6):
nC]N), 1153s, 1099m
2.2.1. Typical procedure for the synthesis of the ligands
(n
l
¼ 275, 346.
Thiosemicarbazide (202.2 mg, 2.22 mmol) was added to a hot
(75 ꢂC) solution of salicylaldehyde (271.3 mg, 2.22 mmol) in ethanol
(20 ml). The reaction mixture was refluxed for 2.5 h. The reaction
mixture was then concentrated to ca. 10 ml. The precipitated white
product was filtered off, washed with water, methanol, recrystal-
lized from ethanol and dried under vacuum (412.7 mg, 95%).
Analytical and spectroscopical data for this compound are in
agreement with those reported in the literature [29,30]. For the IR
spectrum see Lobana et al. [31].1 (H2L1): 1H NMR (300 MHz, DMSO-
d
¼ 11.50 (s, 1H,
NH), 10.04 (br s, 1H, OH), 8.29 (s, 1H, H(7)), 8.20 (br s, 2H, NH2), 8.11
(br s, 1H, H(4)), 7.74 (d, 4J ¼ 2.3 Hz, 1H, H(6)). 13C NMR (75 MHz,
DMSO-d6):
d
¼ 178.9 (C8), 152.8 (C2), 139.3 (C7), 136.0 (C4), 129.9
(C6), 126.0 (Cq), 113.8 (Cq), 112.9 (Cq). MS (EI): m/z ¼ 353 (Mþ). Anal.
Calcd. for C8H7Br2N3OS (353.03 g/mol): C, 27.22; H, 2.00; N, 11.90.
Found: C, 23.34; H, 2.01; N, 11.91. IR (KBr,
3357se3015m (nas and nsNH2), 3155m (nNH), 1611s (
1535s, 1288s, 1265m (nCO), 1138s, 1094m ( C]S), 859w, 815w, 714w.
UV/vis (10ꢀ5 M, DMSO (nm)):
¼ 321, 346, 413.
n
/cmꢀ1): 3468s (nOH),
n
C]N), 1542s,
n
d6):
d
¼ 11.33 (s,1H, NH), 9.84 (s,1H, OH), 8.34 (s,1H, H(7)), 8.07 (br s,
l
1H, NH2), 7.90 (br s, 1H, NH2), 7.87 (d, J ¼ 7.5 Hz, 1H, H(3)), 7.18 (pst,
J ¼ 7.5 Hz,1H, H(4)), 6.83 (d, J ¼ 7.5 Hz,1H, H(6)), 6.79 (pst, J ¼ 7.5 Hz,
1H, H(5)). MS (EI): m/z ¼ 195 (Mþ). Anal. Calcd. for C8H9N3OS
7 (H2L7): (84%) 1H NMR (300 MHz, DMSO-d6):
d
¼ 11.40 (s, 1H,
NH), 10.21 (s, 1H, OH), 8.29 (s, 1H, H(7)), 8.20 (d, 4J ¼ 2.6 Hz, 1H,
H(6)), 8.14 (br s, 2H, NH2), 7.33 (dd, 3J ¼ 8.7 Hz, 4J ¼ 2.6 Hz,1H, H(4)),

96
N.S. Youssef et al. / Journal of Organometallic Chemistry 714 (2012) 94e103
6.82 (d, 3J ¼ 8.7 Hz, 1H, H(3)). 13C NMR (75 MHz, DMSO-d6):
C
16H24CuN3OS (369.99 g/mol): C, 51.94; H, 6.54; N, 11.36. Found: C,
d
¼ 179.0 (C8), 156.4 (C2), 138.1 (C7), 134.0 (C4), 129.2 (C6), 123.7 (C1),
51.90; H, 6.31; N, 11.10. IR (KBr,
n
/cmꢀ1): 3472s, 3294s, 2960s, 1600s,
119.0 (C3), 112.0 (C5). MS (EI): m/z ¼ 273 (Mþ). Anal. Calcd. for
1596m, 1528m, 1432s, 1332s, 1301m, 1172s, 1026w, 839w, 784m.
C8H8BrN3OS (274.14 g/mol): C, 35.05; H, 2.94; N, 15.33. Found: C,
UV/vis (10ꢀ5 M, DMSO (nm)):
l
¼ 275, 335, 407. UV/vis (10ꢀ3 M,
35.06; H, 2.59; N, 15.01. IR (KBr,
3250se2997m (nas and nsNH2), 3162s (nNH), 1655s (dNH2), 1610s
C]N), 1601s, 1545s, 1294m, 1264m (nCO), 1064 ( C]S), 819m (nsCS).
UV/vis (10ꢀ5 M, DMSO (nm)):
¼ 270, 347, 400.
8 (H2L8): (31%). 1H NMR (300 MHz, DMSO-d6):
n
/cmꢀ1): 3456s
(nOH),
DMSO (nm)):
8.8
l
¼
583. Conductance
L
(DMSO 10ꢀ3 M):
m
U
ꢀ1 cm2 molꢀ1
.
meff. (25 ꢂC) ¼ diamagnetic.
(n
n
[CuHL3]n (14) was collected as a green powder (58%). MS (FABþ):
l
m/z ¼ 288 (Mþ þ 1), 513 [2M ꢀ Cu þ 2]þ. Anal. Calcd. for
d
¼ 11.51 (s, 1H,
C9H10CuN3O2S (287.81 g/mol): C, 37.56; H, 3.50; N, 14.60. Found: C,
NH), 10.05 (s, 1H, OH), 8.34 (s, 1H, H(7)), 8.19 (br s, 2H, NH2), 7.78 (d,
37.42; H, 3.39; N, 14.20. IR (KBr,
3200m, 1618m, 1602s, 1336m, 1307m, 1246s, 1218s, 1085w, 968w,
857w, 775m, 734m. UV/vis (10ꢀ5 M, DMSO (nm)):
¼ 260, 329,
n
/cmꢀ1): 3449s, 3347m, 3303s,
3J ¼ 9.5 Hz, 1H, H(4)), 7.22 (dt, 3J ¼ 9.5 Hz, 4J ¼ 2.9 Hz, 1H, H(6)). 19
F
NMR (282 MHz, DMSO-d6):
d
¼ ꢀ122.0 (pst, 3J ¼ 9.5 Hz, 1F,
l
F(5)), ꢀ131.1 (d, 3J ¼ 9.6 Hz, 1F, F(3)). MS (EI): m/z ¼ 231 (Mþ). Anal.
398. UV/vis (10ꢀ3 M, DMSO (nm)):
l
¼ 577. Conductance Lm (DMSO
Calcd. for C8H7F2N3OS (231.22 g/mol): C, 41.56; H, 3.05; N, 18.17.
10ꢀ3 M): 3.5
U
ꢀ1 cm2 molꢀ1
.
meff. (25 ꢂC) ¼ diamagnetic.
Found: C, 41.54; H, 3.14; N, 18.02. IR (KBr,
n
/cmꢀ1): 3430s, 3324m,
[CuHL4]n (15) was collected as a light brown powder (90%). MS
(FABþ): m/z ¼ 329 (Mþ þ 1), 595 [2M-Cu þ 2]þ. Anal. Calcd. for
C12H17CuN4OS (328.9 g/mol): C, 43.82; H, 5.21; N, 17.03. Found: C,
3289s, 3176s, 1618s, 1588m, 1541s, 1485s, 1459s, 1378s, 1298m,
1226m, 1104m, 1011m, 987m, 832m.
10 (H2L10): (66%). 1H NMR (400 MHz, DMSO-d6):
d
¼ 11.75 (s, 1H,
44,24;H,5.07;N,16.75.IR(KBr, n
/cmꢀ1):3409s,3291m, 3103s,2967m,
NH), 10.04 (s, 1H, NHPh), 9.97 (br s, 1H, OH), 8.50 (s, 1H, H(7)), 8.08
(d, J ¼ 7.6 Hz, 1H, H(6)), 7.58 (d, J ¼ 7.8 Hz, 2H, PhHortho), 7.37 (pst,
J ¼ 7.8 Hz, 2H, PhHmeta), 7.24 (pst, J ¼ 7.6 Hz, 1H, H(4)), 7.20 (pst,
J ¼ 7.8 Hz, 1H, PhHpara), 6.89 (d, J ¼ 7.6 Hz, 1H, H(3)), 6.84 (pst,
1637m,1610s,1587s,1354s,1297m,1244s,1134s,1051w, 829w, 785m.
UV/vis (10ꢀ5 M, DMSO (nm)):
l
¼ 270, 388. UV/vis (10ꢀ3 M, DMSO
(nm)):
l
¼ 573. Conductance Lm (DMSO 10ꢀ3 M): 1.2
U .
ꢀ1 cm2 molꢀ1
[CuHL5]n (16a) was collected as a brown powder (73.2%). MS
(FABþ): m/z ¼ 326 (Mþ þ 1). Anal. Calcd. for C8H6Cl2CuN3OS
(326.67 g/mol): C, 29.41; H,1.85; N,12.86. Found: C, 29.32; H,1.80; N,
J ¼ 7.6 Hz, 1H, H(5)). 13C NMR (75 MHz, DMSO-d6):
d
¼ 176.6 (C8),
157.5 (C2), 141.0 (C7), 140.0 (CPh), 132.2 (C4), 128.9 (CPh-meta), 127.9
(C6),126.5 (CPh-ortho),126.0 (CPh-para), 121.1 (C1),120.1 (C3),116.9 (C5).
MS (EI): m/z ¼ 271 (Mþ). Anal. Calcd. for C14H13N3OS (271.34 g/
mol): C, 61.97; H, 4.83; N, 15.49. Found: C, 61.40; H, 5.15; N, 15.29. IR
12.82. IR (KBr, n
/cmꢀ1): 3439s, 3247m, 3154s, 1617s, 1613s, 1477s,
1441s, 1342m, 1319m, 1213m, 1182s, 866w, 772m, 760m. UV/vis
(10ꢀ5 M, DMSO (nm)):
l
¼ 274, 331, 418. UV/vis (10ꢀ3 M, DMSO
(KBr,
1509s, 1271s, 1260m, 1208s, 1151m, 1080m, 1033m, 794w, 755m,
742m, 692m. UV/vis (10ꢀ5 M, DMSO (nm)):
¼ 283, 344.
11 (H2L11): (79%). 1H NMR (400 MHz, DMSO-d6):
¼ 11.66 (s, 1H,
n
/cmꢀ1): 3380m, 3150s, 2992m, 1621m, 1604m, 1593m, 1539s,
(nm)):
l
¼ 591. Conductance Lm (DMSO 10ꢀ3 M): 8.0
U .
ꢀ1 cm2 molꢀ1
[CuHL6]n (17) was collected as a brown powder (51%). MS
(FABþ): m/z ¼ 414 (Mþ þ 1), 766 [2M-Cu þ 2]þ, 829 [2M þ 2]þ. Anal.
Calcd. for C8H6Br2CuN3OS (415.57 g/mol): C, 23.12; H, 1.46; N, 10.11.
l
d
NH), 10.15 (s, 1H, NHPh), 9.98 (br s, 1H, OH), 8.38 (s, 1H, H(7)), 7.50
(d, J ¼ 7.5 Hz, 2H, PhH), 7.38 (pst, J ¼ 7.5 Hz, 2H, PhH), 7.32 (d,
J ¼ 2.1 Hz, 1H, H(4)), 7.20 (m, 2H, H(6) and PhH), 1.41 (s, 9H, CH3),
Found: C, 23.18; H, 1.50; N, 9.98. IR (KBr, n
/cmꢀ1): 3468s, 3232m,
3135m, 3066m, 1618s, 1608m, 1596m, 1474s, 1435s, 1410m, 1340m,
1308m, 1218m, 1167m, 867w, 730m. UV/vis (10ꢀ5 M, DMSO (nm)):
1.29 (s, 9H, CH3). 13C NMR (75 MHz, DMSO-d6):
d
¼ 177.1 (C8), 154.1
l
¼ 276, 341, 410. UV/vis (10ꢀ3 M, DMSO (nm)):
l
¼ 585. Conduc-
(C2), 149.1 (C7), 141.8 (CPh), 140.2 (C5), 136.9(C3), 129.1 (CPh-meta),
126.7 (CPh-para), 126.6 (C4), 126.4 (CPh-ortho)126.1 (C6), 118.7 (C1), 35.5
(CCMe3), 34.8 (CCMe3), 32.1 (CCH3), 30.2 (CCH3). MS (EI): m/z ¼ 383
(Mþ). Anal. Calcd. for C22H29N3OS (383.55 g/mol): C, 68.89; H, 7.62;
tance L (DMSO 10ꢀ3 M): 4.2 Uꢀ1 cm2 molꢀ1
.
m
[CuHL7]n (18) was collected as a brown powder (67%). MS
(FABþ): m/z ¼ 336 (Mþ þ 1). Anal. Calcd. for C8H7BrCuN3OS
(336.68 g/mol): C, 28.54; H, 2.10; N, 12.48. Found: C, 28.77; H, 2.09;
N, 10.96. Found: C, 68.58; H, 8.00; N, 10.91. IR (KBr,
3142s, 2956s, 1560s, 1545s, 1272s, 1210s, 1087m, 803w, 747m. UV/
vis (10ꢀ5 M, DMSO (nm)):
¼ 283, 345.
n
/cmꢀ1): 3325s,
N, 11.99. IR (KBr, n
/cmꢀ1): 3418s, 3268m, 2999s, 1637s, 1600s, 1496s,
1464s, 1341m, 1294m, 1187s, 1038m, 869w, 809m, 751w. UV/vis
l
(10ꢀ5 M, DMSO (nm)):
(nm)):
5.2
ꢀ1 cm2 molꢀ1
l
¼ 271, 335, 406. UV/vis (10ꢀ3 M, DMSO
l
¼
578. Conductance
L
m
(DMSO 10ꢀ3 M):
2.3. Synthesis of the complexes
U
.
meff. (25 ꢂC) ¼ 1.18 B.M.
[CuHL8]n (19) was collected as a dark green powder (33%). MS
2.3.1. Typical procedure for the synthesis of the copper complexes
(FABþ): m/z ¼ 294 (Mþ þ 1). Anal. Calcd. for C8H6CuF2N3OS
A
solution of copper (II) acetate monohydrate (222 mg,
(293.76 g/mol): C, 32.71; H, 2.06; N, 14.30. Found: C, 33.11; H, 1.74;
1.11 mmol) in the minimum required amount of methanol was
added dropwise to a hot (75 ꢂC) solution of H2L1, 1 (214 mg,
1.10 mmol) in ethanol (50 ml) (75 ꢂC). The reaction mixture was then
refluxed for 2.5 h. The brownprecipitate so formed was recovered by
filtration and was recrystallized from refluxing ethanol. The
collected solid was suspended in cold ethanol (3 ꢁ 10 ml), separated
by centrifugation, then suspended in diethyl ether (3 ꢁ 10 ml),
separated by centrifugation and dried under vacuum.
N, 14.40. IR (KBr,
1561m, 1465s, 1300m, 1267s, 1125s, 1069w, 997m, 849w, 832s,
n
/cmꢀ1): 3430m, 3301s, 3154m, 2874m, 1587s,
745w. Conductance L (DMSO 10ꢀ3 M): 1.0 Uꢀ1 cm2 molꢀ1
.
m
[CuHL9]n (20) was collected as a light brown powder (85%). 1H
NMR (300 MHz, CDCl3):
d
¼ 11.75 (br s, 1H, CH), 11,00 (br s, 1H, OH
or NH), 8.00e7.80 (br s, 2H, ArH), 5.98 (br, 2H, ArH), 4.11 (br s, 1H,
NH), 1.75 (br, 2H, CH2), 1.58 (br, 2H, CH3). MS (FABþ): m/z ¼ 286
(Mþ þ 1). Anal. Calcd. for C10H12CuN3OS (285.83 g/mol): C, 42.02; H,
[CuHL1]n (12) was collected as a light brown powder (233 mg,
83%). MS (FABþ): m/z ¼ 258 (Mþ þ 1). Anal. Calcd. for C8H8CuN3OS
(257.78 g/mol): C, 37.27; H, 3.13; N, 16.30. Found: C, 37.66; H, 3.00;
4.23; N, 14.70. Found: C, 41.67; H, 4.13; N, 14.92. IR (KBr, n
/cmꢀ1):
3369s, 2968m, 1596s, 1555m, 1506s, 1474s, 1438m, 1320m, 1280m,
1204s, 1156w, 1073w, 835w, 745m. UV/vis (10ꢀ5 M, DMSO (nm)):
N,16.14. IR (KBr,
n
/cmꢀ1): 3467s, 3367s, 3352s, 3267s,1599s,1596m,
l
¼ 275, 332, 397. UV/vis (10ꢀ3 M, DMSO (nm)):
l 577. Conductance
1511s, 1492s, 1317m, 1278m, 1210m, 1152w, 809m, 751s. UV/vis
L
m (DMSO 10ꢀ3 M): 3.1 Uꢀ1 cm2 molꢀ1
.
(10ꢀ5 M, DMSO (nm)):
l
¼ 279, 329, 396. UV/vis (10ꢀ3 M, DMSO
[CuHL10]n (21) was collected as a light brown powder (62%). MS
(nm)):
l
578. Conductance Lm (DMSO 10ꢀ3 M): 2.5
U
ꢀ1 cm2 molꢀ1
.
(FABþ): m/z ¼ 334 (Mþ þ 1), 605 [2M-Cu þ 2]þ. Anal. Calcd. for
meff. (25 ꢂC) ¼ diamagnetic.
C14H12CuN3OS (333.88 g/mol): C, 50.36; H, 3.62; N, 12.59. Found: C,
[CuHL2]n (13) was collected as a brown powder (88%). MS
(FABþ): m/z ¼ 370 (Mþ þ 1), 677 [2M-Cu þ 2]þ. Anal. Calcd. for
50.25; H, 3.72; N, 12.28. IR (KBr,
2937m, 1595m, 1508m, 1498m, 1458s, 1431s, 1314m, 1251w, 1210m,
n
/cmꢀ1): 3404w, 3235m, 3024m,

N.S. Youssef et al. / Journal of Organometallic Chemistry 714 (2012) 94e103
97
1178m, 1088w, 866w, 769m, 747m. UV/vis (10ꢀ5 M, DMSO (nm)):
The collected analytical data for cis and trans ethyl-2-
phenylcyclopropanecarboxylate [33], cis and trans ethyl-2-p-tolylcy-
clopropanecarboxylate [33], cis and trans ethyl-2-methyl-2-
phenylcyclopropanecarboxylate [34], ethyl-2,2-dphenylcyclopropane
carboxylate [33], cis and trans ethyl-2-(4-chlorophenyl)-2-methyl
cyclopropanecarboxylate [35], cis and trans ethyl-2-naphthylcy
clopropanecarboxylate [36], cis and trans ethyl-2-hexylcyclopropane
carboxylate [37], cis and trans bicyclo[4.1.0]heptane-7-carboxylic
acid ethyl ester [37], and cis and trans ethyl chrysantemate [37] are
in agreement to those reported in the literature.
l
¼ 287, 323, 408. UV/vis (10ꢀ3 M, DMSO (nm)): no ded band was
detected. Conductance Lm (DMSO 10ꢀ3 M): 3.5 Uꢀ1 cm2 molꢀ1
.
[CuHL11
] (22) was collected as a dark green powder (77%). MS
n
(FABþ): m/z ¼ 446 (Mþ þ 1), 829 [2M-Cu þ 2]þ. Anal. Calcd. for
C22H28CuN3OS (446.09 g/mol): C, 59.23; H, 6.33; N, 9.42. Found: C,
58.92; H, 6.46; N, 9.33. IR (KBr, n
/cmꢀ1): 3421m, 3221m, 2955s,
2904m, 2868m, 1592s, 1541s, 1498s, 1548s, 1438s, 1314m, 1299m,
1256s, 1167s, 1065w, 832w, 745m. UV/vis (10ꢀ5 M, DMSO (nm)):
l
¼ 324, 417. UV/vis (10ꢀ3 M, DMSO (nm)):
l 592. Conductance L
m
(DMSO 10ꢀ3 M): 2.5 Uꢀ1 cm2 molꢀ1
.
3. Results and discussion
2.3.2. Alternative synthesis of the copper complex 16b
A solution of copper (II) acetate monohydrate (200.0 mg,
1.00 mmol) in the minimum required amount of methanol was
added dropwise to a hot (60 ꢂC) solution of H2L5, 5 (191.0 mg,
0.723 mmol) in methanol (50 ml). The reaction mixturewas refluxed
for 3 h and then concentrated to half volume. The dark green
precipitate so formed was recovered by filtration, it was washed
with ethanol, then with diethyl ether and dried under vacuum over
anhydrous CaCl2. [CuL5$2H2O] (16b) was collected as a dark green
powder (204.9 mg, 78.4%). Anal. Calcd. for C8H9Cl2CuN3O3S
(361.69 g/mol): C, 26.57; H, 2.51; N, 11.91; Cu 17.57. Found: C, 26.42;
3.1. Synthesis
Ligands were synthesized by reacting hydrazinecarbothioamide
with substituted salicylaldehydes in refluxing ethanol (Scheme 1).
Pure ligands could be obtained by simple recrystallization from
ethanol. Only in the case of ligand 2 it was necessary to perform
a column chromatography over silica in order to obtain a compound
free from traces of hydrazinecarbothioamide. The copper complexes
were synthesized by treating the ligands in refluxing ethanol with
copper (II) acetate monohydrate dissolved in the minimum required
amount of methanol (Scheme 2). In all cases the formation of a dark
greenor brownprecipitatewas observedat the very beginningof the
addition of the copper acetate. The reaction mixture was then
refluxed and the product collected upon filtration after cooling at
room temperature. A recrystallization step from hot ethanol was
needed to obtain a crystalline powder. Traces of water could be
removed only after repeated washing with ethanol and diethyl
ether. Both ligands and complexes were obtained in good yields.
H, 2.63; N, 11.74; Cu 17.44. IR (KBr, n
/cmꢀ1): 3480s, 3464s, 3277s,
3182s,1748w,1620s,1610m,1601s,1473s,1439s,1314s,1212s,1164w,
950m, 867m, 630w, 520w, 431w, 364w. meff. (25 ꢂC) ¼ 1.79 B.M.
2.3.3. Detection of acetic aldehyde in the synthesis of complex 20
A
solution of copper (II) acetate monohydrate (14.9 mg,
0.076 mmol) in the minimum required amount of water was added
to a hot (70 ꢂC) solution of H2L9, 9 (15.2 mg, 0.068 mmol) in ethanol
(5 ml) in a pressure tube. The reaction mixture was heated to 80 ꢂC
for 2.5 h and then cooled to 0 ꢂC. After centrifugation, a sample of
the supernatant solution was withdrawn and analysed by GC.
Acetic aldehyde was detected [32].
3.2. Spectroscopy
The mass spectra (EI) of the Schiff base ligands 1e8 and 10e11
revealed molecular ion peaks which are coincident with the formula
weights for these ligands and support the identity of their structures.
Although numerous reports in the literature on the use of thio-
semicarbazide derived Schiff bases have been published in the last
years [29e31,38e41], reported spectroscopicaland analytical data are
often scant; 13C NMR spectra are seldom described and in some cases,
especiallyconcerning IRspectra, reported dataarenot ingood relative
agreement (see later). The 1H and 13CNMRspectraof theligandsarein
agreement with the proposed structure. Enol and keto tautomers are
very close in energy for Schiff bases and may compete for stability. As
expected for phenol Schiff bases derivatives, the enol form prevails
and we never observed any keto-enol tautomerism in DMSO-d6 (nor
in CDCl3) solutions [42] (see Experimental section). The IR spectra of
powders show the characteristic intense bands in the range
2.4. Typical procedure for the catalytic cyclopropanation of olefins
CuHL1 (12) (1.3 mg, 5.1$10ꢀ3 mmol) and
a-methylstyrene
(0.660 ml, 5.1 mmol) were suspended in distilled dichloroethane
(9 ml) and the reaction mixture was heated under stirring at 70 ꢂC.
Thendichloroethanesolution (1 ml) of EDA (0.266ml,2.53mmol)was
slowly added by a syringe pump during 100 min. The end of the
reaction was checked by IR, following the disappearance of the band
due to the stretching of N2 moiety at 2114 cmꢀ1. After the addition of
EDA, a complete conversion of EDA (IR absorbance <0.025) was
observed; the solution was analysed by GC after the addition of 2,4-
dinitrotoluene (230 mg, 1.27 mmol) as internal standard. Yields
were confirmed by 1H NMRanalysisofthecrudeand,inselectedcases,
products were isolated and purified by column chromatography.
1618e1590 cmꢀ1 associated with the n(C
]
N)
frequencies. The n(OH)
Scheme 1. . Formation of the thiosemicarbazone ligands H2L (1e11) and numbering scheme adopted.

98
N.S. Youssef et al. / Journal of Organometallic Chemistry 714 (2012) 94e103
Scheme 2. Formation of the copper complexes [CuHL]n (12e22).
band of the phenolic oxygen is found almost at the same frequencies
for all ligands, in the range 3468e3409 cmꢀ1, whilst the medium to
intense band in the region 3300e2900 cmꢀ1 have been attributed to
n(NH2) andn(NH). Thefreeligandsexhibitmediumintensitybandsinthe
region1099e1011 cmꢀ1 and859e793cmꢀ1, attributedrespectivelyto
1e8 U
ꢀ1 cm2 molꢀ1. These low values indicate that these complexes
are non-electrolytes due to the absence of any counter ion in their
structures [45].
The absence of water, as well as of coordinate acetate ions, in the
metalcomplexesisalsoindicatedbytheelementalanalyses,whichare
in perfect agreement with a 1:1 ratio metal/ligand of general formula
[CuHL]n (with the exception of complex 16b, which is in agreement
with the formula CuL5$2H2O, see below). This general formula is
supported also by the result of the FAB mass analysis of the metal
complexes. In general the fast atom bombardment (FAB) gives rise to
molecular ions of the type [Mþ þ 1] and, to the best of our knowledge,
very seldom [Mþ þ 2] peaks are observed. For all copper complexes
(with the exception of 16b) we observed a clean base peak corre-
sponding to a general formula [CuHLþ þ 1], which seems to indicate
that the ligand has just lost one hydrogen atom in the complexation
reaction and thus cannot behave as a dianionic ligand, as would be
expected. The mass spectra of complex 13 is reported in Fig. 1 and
compared with the spectrum simulated for the cluster [CuHL2]þ.
It can be seen that, even if with low intensity, an ion peak cor-
responding to the molecular formula [CuLþ þ 1] (369 m/e) is
present. The same pattern is found in all the mass spectra of the
metal complexes (see experimental section).
If we assume, as all analytical and spectroscopical data support,
that the metal to ligand ratio is 1:1 and that we do not have any
other ancillary ligand present in the metal complexes, we must
conclude that the copper complexes are not monomeric in the solid
state and that dimeric (or oligomeric) structures are formed. If the
ligand behaves as monoanionic (A, Fig. 2), then, to balance charges,
the copper atom must have been reduced to a formal I oxidation
state. On the other hand, if the ligand behave as dianionic (B, Fig. 2),
the metal ion should be formally described as copper (II).
The two structures A and B differ only for one proton, and for the
copper oxidation state. In the absence of coordinating solvents, we
must assume that these molecules are not monomers, but must
aggregate. Actually, all these metal complexes are almost insoluble
in most organic solvents, and a good solubility is observed only in
DMSO and DMF. Unfortunately, any attempt to grow crystals suit-
able for X-ray structural determination from these solvents met
with failure. When we tried to record NMR spectra in deuterated
DMSO, we observed broad signals that prevented any attribution of
the chemical shifts. The only complex that showed a good solubility
even in chlorinated solvents was 20. In this case, it was possible to
record a 1H NMR spectrum in CDCl3, that is in agreement with
a diamagnetic complex (see experimental section).
n(C]S) andtons(CS). Thereisnogeneralagreementintheliteraturedata
in the identification of ns(CS), that for some authors is at lower
frequencies (see for example Ref. [30]); however, reported data in the
experimental section are in accordance with those reported by
Campbell in his review [43]. This band is particularly diagnostic for
transition metal complexes of thiosemicarbazide and thio-
semicarbazones, since upon coordination could be shifted almost
100 cmꢀ1 to lower frequencies. A shift of this order would indicate
aconsiderablechange inthebondorder, suchaswouldresultfromthe
formation of a strong metal-sulfur bond. As expected, for all the
synthesizedcoppercomplexes,thefrequenciescorrespondingtons(CS)
were found in the range 809e730 cmꢀ1, therefore indicating the
coordination of the copper ion to the S atom. This fact is supported
even by the disappearance of the n(C
The complexation of the metal ion is accompanied also by a negative
] .
S) band at around 1050 cmꢀ1
shift of the frequencies corresponding to n(C
]
(from the spectral
N)
region 1618e1590 cmꢀ1 to lower wave numbers) and to namideII (from
1550 to 1537 cmꢀ1 to 1527e1465 cmꢀ1) and a positive shiftof the n(CO)
(from 1271e1245 cmꢀ1 to 1318e1278 cmꢀ1) [40]. All these data
support that the ligand donor atoms, O, N, S, chelate the copper atom
in a tridentate manner. Noteworthy, we observed for all metal
complexes the persistence of
a sharp band in the region
3484e3404 cmꢀ1. Normally absorption bands assigned to the n(OH)
modes oflatticewaterappearasbroadand rather intense bandsinthe
spectral region of 3550e3200 cmꢀ1 (antisymmetric and symmetric
OH stretchings) and are accompanied by d(OH) in the region
1630e1600 cmꢀ1, which are not observed in the present case. It
should be pointed out, however, that also the OH group in hydroxo
complexes lacks the HOH bending mode near 1600 cmꢀ1 [44]. The
absence of coordinated water molecules within the coordination
sphere of these complexes (with the exception of complex 16b, see
experimental section) is further supported by the absence of bands in
the 950e930 cmꢀ1 and 635e615 cmꢀ1 regions, due to H2O rocking
and wagging, respectively. On the other hand, the existence of coor-
dinated water molecules within the coordination sphere of complex
[CuL5$2H2O] (16b) is supported by the presence of bands at 3480,
1610, 950 and 630 cmꢀ1, due to OH stretching, HOH deformation, H2O
rocking and H2O wagging respectively. The complexes under study
did not show any band which may be attributed to acetate ligand
coordinated to the central metal atom [44].
The solid copper complexes 12e14 are diamagnetic at room
temperature. The typical magnetic momentum value for
mononuclear copper (II) compounds with a S ¼ ½ spin state are
The molar conductivities L of the metal complexes 12e22
m
dissolved in DMSO at 10ꢀ3 M were found to be in the range

N.S. Youssef et al. / Journal of Organometallic Chemistry 714 (2012) 94e103
99
Fig. 1. FAB mass spectrum of complex 13 (left) compared with the simulated spectra for the cluster [CuHL2]þ
.
expected in the range 1.79e1.95 B.M [46]. Only in the case of
complex 18 we measured a room temperature magnetic moment of
1.18 B.M. which seems to indicate a partial antiferromagnetic
coupling of spins at this temperature [47]. Further studies on the
magnetism of these molecules at low temperature will be the
subject of a study in the next future.
a spinespin interaction between the paramagnetic copper ions.
This polymeric chain would break upon dissolution in strongly
coordinating solvents such as DMSO or DMF. It should be pointed
out that Cu (I) complexes are not stable towards oxidation and that
the experimental conditions used in the present work and the high
dilutions needed to record an UVevis spectrum in DMSO hampers
the characterization of copper (I) in solution. On the other hand,
only the presence of copper in the oxidation state (I) would explain
the ion peaks of general formula [CuHLþ þ 1] observed in the mass
spectra. To gain a better insight to the oxidation state of the central
metal atom, we decided to carry out an X-ray photoelectron spec-
troscopy (XPS) analysis of selected complexes. XPS analysis, in fact,
is helpful in identifying the oxidation states of elements in various
compounds because the binding energy (BE) measured for the core
electrons undergoes a “chemical shift” as a result of changes in the
chemical environment of the atoms [49]. The Cu2p spin-orbit
doublets (2p1/2 and 2p3/2) obtained from complex 12 are illus-
trated in Fig. 4. A characteristic feature of the Cu2þ (3d9) is a shift in
the Cu2p photoelectron peak doublet to higher BE side compared to
Cuþ (3d10). This can be appreciated in the Cu2p3/2 peak that is
actually split in two components: one major peak at ꢀ932 eV (A,
Fig. 4) that can be ascribed to copper (I), and a smaller peak
at ꢀ935 eV (B, Fig. 4), in the typical range for a copper (II) atom. A
small shake-up satellite (C Fig. 4) can be seen on the lower kinetic
energy (KE) (higher BE), which is normally associated with copper
(II). Similar spectra were obtained for complexes 13 and 20.
The UVevis spectra of the thiosemicarbazone ligands in DMSO
(10ꢀ5 M) showed broad bands in the range 270e275 nm assignable
to the phenyl ring
p / p* transitions. This band is not affected to
a major extent upon coordination of the copper atom. The n /
p
*
transitions associated with the C]N group [48] around 350 nm in
the spectra of the free ligands are generally shifted to higher energy
on complexation. In the spectra of some complexes this latter band
shifts to higher energies. The shifts of these bands with respect to
those of the free ligand indicate coordination of phenolic oxygen
and azomethine thioenol moieties to the metal ions. There are also
charge-transfer transitions partially responsible for the intense
colours of some of the complexes. The absorption bands with high
intensity observed at 420e430 nm are assigned to charge-transfer
transitions. The UVevis spectra of the complexes (10ꢀ3 M in DMSO)
showed broad single band at 570e590 nm (ε in the range
200e300 Mꢀ1 cmꢀ1, Fig. 3) which is consistent with a distorted
square pyramidal geometry at a Cu (II) metal atom [41].
These experimental data provide grounds for suggesting
a polynuclear structure for the metal complexes 12e22 with
These data suggest that copper is present as copper (I) in the
complexes. The fact that small peaks due to the presence of some Cu
H
N
NH2
N
NH2
N
N
S
S
CuII
CuI
O
O
N
NH2
H
N
N
NH
N
S
CuI
O
S
CuII
H
O
B
H
N
m = 256
NH
N
S
A
CuI
O
H
m = 257
Fig. 3. UVevis spectra for complexes 13e15, 18 and 20 (see legend) in the region
500e700 nm; 10ꢀ3 M solutions in DMSO.
Fig. 2. Possible molecular formulae (sketched as monomers) for complex 12:
(monoanionic ligand); B (dianionic ligand).
A

100
N.S. Youssef et al. / Journal of Organometallic Chemistry 714 (2012) 94e103
Table 1
Catalytic cyclopropanation of a
-methylstyrene by EDA.a
Entry
Catalyst
Conversion (%)b
Yield (%)c
trans/cisc
1
2
3
4
5
6
7
8
12
13
14
15
16a
16b
17
18
19
20
21
22
>99%
>99%
>99%
>99%
>99%
>99%
>99%
>99%
>99%
>99%
>99%
>99%
79
78
58
62
56:44
56:44
55:44
56:44
57:43
57:43
55:45
56:44
54:46
56:44
57:43
55:45
97 (68)d
97
93
92
54
81
87
88
9
10
11
12
a
Experimental conditions: EDA (2.52 mmol) dissolved in C2H4Cl2 (1 ml) was
Fig. 4. The Cu2p spin-orbit doublets obtained for complex 12.
slowly added (100 min) to a hot (70 ꢂC) solution of cat (5.1 ꢁ 10ꢀ3 mmol) and
a-
methylstyrene (5.1 mmol) in dichloroethane (9 ml).
(II) are present suggests that the surface of the particles was easily
oxidized. We must underline that XPS analysis is a surface technique
and that a partial oxidation of the surface may always occur [49].
To investigate if the employed solvent (ethanol) in the reaction
may be responsible for the reduction of copper during the
complexation, complex 16b was synthesized in methanol instead.
In this case we isolated a paramagnetic complex of copper (II) (meff.
(25 ꢂC) ¼ 1.79 B.M.) corresponding to the molecular formula
[CuL5$2H2O]. When the reaction was repeated in isopropanol (a
better reducing agent than ethanol) a less pure product but with
a chemical composition similar to 16a was obtained instead (see
experimental). The fact that copper (II) is reduced to copper (I) in
reaction with thioylcarbamoyl ligand [50] and with thio-
semicarbazones [51] is not unprecedented. However, in reported
cases, reduction took place in the absence of alcohols. To clarify the
role played by the alcohol in the reduction from Cu (II) to Cu (I), we
have conducted a GC analysis of the reaction mixture to detect if an
oxidation product from ethanol was observed. In order to do this,
we had to run the reaction in the absence of methanol, that under
our experimental conditions interfere with the detection of acetic
aldehyde, by dissolving the copper acetate in the minimum of
water and performing the reaction in a pressure tube to avoid any
possible evaporation of the formed volatile products. Under these
conditions, we were able to detect a reasonable amount of acetic
aldehyde after the reaction took place (see experimental).
b
Conversion of the starting EDA.
Determined by GC (yield based on EDA).
Isolated yield.
c
d
Table 2
Cyclopropanation of
a
-methylstyrene with EDA catalysed by complex 16b, CuL5
Conversion(%)b Yield (%)c trans/cisc
(H2O)2.a
Entry Cat/EDA/olefin Time (min)
1
1/500/1000
1/1000/1500
1/10000/20000 380
1/20000/40000 420
7/5/5d
> 99
87
95
77
74
57:43
57:43
56:44
55:45
2e
3f
4g
100/100/100d > 99
92
92
a
Experimental conditions: EDA was added to
a
solution of 16b (2.73 mg,
7.5 ꢁ 10ꢀ3 mmol) and
a
-methylstyrene in dichloroethane (10 ml) at 70 ꢂC.
b
Conversion of the starting EDA.
Determined by GC (yield based on EDA).
After complete consumption of the starting EDA, the catalytic cycle was restored
c
d
twice by addition of EDA and a-methylstyrene; global yield is reported.
e
16b (1.35 mg, 3.7 ꢁ 10ꢀ3 mmol) in 9 ml of dichloroethane was used; EDA dis-
solved in C2H4Cl2 (1 ml) was slowly added.
f
This solution was prepared by dissolving 16b (1.35 mg, 3.7 ꢁ 10ꢀ3 mmol) in
dichloroethane (10 ml); 1 ml of this solution was added to the solution of
a-
methylstyrene in 9 ml of dichloroethane.
g
This solution was prepared by dissolving 16b (1.35 mg, 3.7 ꢁ 10ꢀ3 mmol) in
dichloroethane (10 ml); 0.5 ml of this solution were added to the solution of
methylstyrene in 9.5 ml of dichloroethane.
a-
subsequent transfer of the carbene moiety to the C]C double bond.
In Table 1 are collected all the obtained results in the cyclo-
propanation reaction of a-methylstyrene with the different cata-
3.3. Catalytic activity
lysts. In all cases cyclopropanes were obtained in yields from good
to excellent. Fumarate and maleate, the homo-coupling product of
EDA, were the only other product detected and accounted for the
missing mass balance of the reactions. The diastereoselective
outcome of the reaction is rather poor in all cases (almost equi-
molar amount of trans and cis cyclopropanes) and does not seem to
be influenced by the steric and/or electronic requirement of the
thiosemicarbazone ligand.
The catalytic activity of complexes 12e22 in cyclopropanation
reactions has been investigated. As a model reaction we choose the
cyclopropanation of
a
-methylstyrene by EDA (EDA ¼ ethyldia-
zoacetate) (Scheme 3). Catalytic reactions were run by slow addition
of EDA by syringe pump (100 min) to a stirred solution containing the
olefin and the metalcomplex indichloroethane underdinitrogen (Cu/
EDA/a
-methylstyrene ratio 1:500:1000) at 70 ꢂC. In all cases a quan-
Remarkably, almost no difference in the chemical yield was
observed employing a catalyst where copper is present as Cu (I) or
complex 16b, the only monomeric copper (II) catalyst tested. This
finding, however, is not surprising, since it is commonly accepted
that EDA reduces Cu (II) to Cu (I). We next optimised the reaction
titative conversion of the starting EDA was observed at the end of the
addition, as judged by IR spectroscopy, checking the disappearance of
the band due to the stretching of the N2 moiety (
n
¼ 2114 cmꢀ1).
All tested copper complexes exhibited a remarkable catalytic
activity towards the decomposition of ethyldiazoacetate, and the
Scheme 3.

N.S. Youssef et al. / Journal of Organometallic Chemistry 714 (2012) 94e103
101
conditions and catalyst loadings, employing complex 16b. The
results are summarised in Table 2.
When EDA was added in one portion to the reaction mixture
(entries 1, 3 and 4, Table 2), an induction period was observed,
that, in the case of complex 16b, can be necessary to reduce
copper (II) to copper (I). In fact, this induction period is not
observed upon the second and third additions (entry 1, Table 2).
Noteworthy, the copper complex does not lose its catalytic
Table 3
Cyclopropanation of olefins with EDA.a
Entry
1
Olefin
Cat
Time (min)
12
Conversion(%)b
Yield(%)c
68
trans/cisc
16b
>99
72:28
2d
16b
16b
100d
100d
>99
>99
99 (90)
82 (68)
92:8
3d
57:43
4
16b
16b
100d
>99
>99
90 (63)
e
5
6
31
76
71:29
12
100d
19
>99
>99
>99
>99
>99
63
65
75
76
78
68:32
69:31
80:20
73:27
66:44
16b
16a
21
100d
100d
100d
22
7
8
16b
16a
11
>99
>99
67
92
91:9
91:9
100d
16b
16a
32
>99
>99
82
98
64:36
50:50
100d
9e
16b
16b
100d
100d
>99
>99
85 (51)
90 (84)
58:41:1:0
10f
>99
11
21
100d
>99
15
>99
a
Experimental conditions: EDA was added to a solution of cat (5.1 ꢁ 10ꢀ3 mmol) and olefin in dichloroethane (10 ml) at 70 ꢂC.
Conversion of the starting EDA.
Determined by GC; isolated yield in parenthesis (yield based on EDA).
EDA dissolved in C2H4Cl2 (1 ml) was slowly added to the solution containing the catalyst and the olefin in dichloroethane (9 ml).
The two major products were determined to be (1R,2R,20R) and (1R,2R,20S) by comparison with literature data [52]. The fourth possible diastereoisomer was not detected.
Only one diasteroeisomer (1R,2R, 20R) was isolated [52].
b
c
d
e
f

102
N.S. Youssef et al. / Journal of Organometallic Chemistry 714 (2012) 94e103
activity after 3 consecutive runs. In contrast to the prolonged
EDA addition time generally required to reduce the formation of
homo-coupling products in cyclopropanation, we found complex
16b to be rather insensitive to this: if we compare entry 1,
Table 2 with entry 6, Table 1, a decrease only from 97% to 87% in
4. Conclusions
The straightforward synthesis of several thiosemicarbazide
derived Schiff base copper complexes has been reported. The
presence of N, S, O donor atoms of thiosemicarbazones renders the
study their coordination to transition metals very interesting.
Spectral and analytical collected data in this study suggested that,
at least in part, reduction to copper (I) has occurred. Further studies
in determining the crystal structure of these complexes in the
absence of strong coordinating solvents are on-going in our lab.
However, all synthesized complexes showed excellent catalytic
activities in cyclopropanation reactions and TON up to 18,400 could
be obtained. In contrast to the prolonged EDA addition time
generally required to reduce the formation of homo-coupling
products in cyclopropanation, we found those complexes very
selective. Furthermore, a single addition of EDA is required to yield
the desired cyclopropanes in excellent yields. Moreover, the cata-
lysts are very robust and no decrease in yield was observed even
after three catalytic runs. Several cyclopropanes have been
obtained in good to excellent yields even from non-activated
cyclopropane yield is observed. At a 16b/EDA/a-methylstyrene
ratio of 1/1000/1500 with slow addition of EDA products are
obtained with excellent yields (entry 2, Table 2). A 1.5 fold
excess of olefin with respect to EDA is however necessary to
reduce the formation of homo-coupling products. We were able
to further reduce the catalyst loading and TON up to 18,400 have
been obtained (entry 4, Table 2). Even if
a poor diaster-
eoselection was observed, to the best of our knowledge, this is
the higher TON reported for a single site copper catalyst in
homogeneous catalytic cyclopropanation [23]. A TON of 100,000
has been recently reported by us with a dinuclear copper pol-
yoxometalate catalyst [26].
To determine the general applicability of copper thio-
semicarbazone complexes, cyclopropanation reactions of a series of
styrene derivatives with varied electronic and steric properties
were carried out, using EDA as carbene source (Table 3). At a cat/
EDA/olefin ratio of 1/500/1000 at 70 ꢂC, the complexes catalysed
the cyclopropanation of a range of substrates with a quantitative
conversion and with selectivities ranging from good to excellent.
The results are summarised in Table 3.
olefins. In the case of the cyclopropanation of (ꢀ)
a-pinene, out of
four possible diastereoisomers, only one product was formed that
could be isolated pure in 84% yield.
Acknowledgements
When styrene was employed as substrate, we observed
a decrease in the reaction rate and in cyclopropane products with
respect to the case of a-methylstyrene (compare entry 1, Table 3
We thank MiUR (PRIN projects) for financial support and Prof.
Claudia Bianchi for XPS analyses.
and entry 6, Table 1). However, in this case the diaster-
eoselectivity is improved (trans/cis ¼ 2.6). Better yields in cyclo-
propane products and higher diastereoselectivities (trans/
cis ¼ 11.5) were obtained when electron donating substituents are
present in the para position of the aromatic ring (entry 2, Table 3). If
References
[1] M. Rubin, M. Rubina, V. Gevorgyan, Chem. Rev. 107 (2007) 3117.
[2] J. Salaun, Top. Curr. Chem. 207 (2000) 1.
[3] H. Lebel, J.-F. Marcoux, C. Molinaro, A.B. Charette, Chem. Rev. 103 (2003) 977.
[4] D. Intrieri, A. Caselli, E. Gallo, Eur. J. Inorg. Chem. (2011) 5071.
[5] A.J. Burke, E.d.P. Carreiro, S. Chercheja, N.M.M. Moura, J.P. Prates Ramalho,
A.I. Rodrigues, C.I.M. dos Santos, J. Organomet. Chem. 692 (2007) 4863.
[6] J.M. Fraile, J.I. García, A. Gissibl, J.A. Mayoral, E. Pires, O. Reiser, M. Roldán,
I. Villalba, Chem. Eur. J. 13 (2007) 8830.
[7] T. Portada, M. Roje, Z. Raza, V. Caplar, M. Zinic, V. Sunjic, Eur. J. Org. Chem.
(2007) 838.
[8] C. Ricardo, T. Pintauer, J. Organomet. Chem. 692 (2007) 5165.
[9] Z.-H. Xu, S.-N. Zhu, X.-L. Sun, Y. Tang, L.-X. Dai, Chem. Commun. (2007) 1960.
[10] C.-T. Yeung, H.-L. Yeung, C.-S. Tsang, W.-Y. Wong, H.-L. Kwong, Chem. Com-
mun. (2007) 5203.
[11] A. Caselli, F. Cesana, E. Gallo, N. Casati, P. Macchi, M. Sisti, G. Celentano,
S. Cenini, Dalton Trans. (2008) 4202.
[12] R. Safiullin, V. Dokichev, L. Yakupova, R. Sultanova, S. Khursan, R. Zaripov,
Y. Tomilov, Kinet. Catal. 49 (2008) 43.
[13] R. Ahuja, A.G. Samuelson, J. Organomet. Chem. 694 (2009) 1153.
[14] J. Perez, D. Morales, L.A. Garcia-Escudero, H. Martinez-Garcia, D. Miguel,
P. Bernad, Dalton Trans. (2009) 375.
[15] I.V. Shishkov, F. Rominger, P. Hofmann, Organometallics 28 (2009) 1049.
[16] M.P. Doyle, J. Org. Chem. 71 (2006) 9253.
[17] A.G.H. Wee, Curr. Org. Synth. 3 (2006) 499.
[18] J.A.S. Howell, Dalton Trans. (2007) 1104.
[19] G. Maas, Chem. Soc. Rev. 33 (2004) 183.
[20] G.-Y. Li, J. Zhang, P.W.H. Chan, Z.-J. Xu, N. Zhu, C.-M. Che, Organometallics 25
(2006) 1676.
[21] V. Dragutan, I. Dragutan, L. Delaude, A. Demonceau, Coord. Chem. Rev. 251
(2007) 765.
[22] R.G. Salomon, J.K. Kochi, J. Am. Chem. Soc. 95 (1973) 3300.
[23] N.S. Youssef, E. El-Zahany, A.M.A. El-Seidy, A. Caselli, S. Cenini, J. Mol. Catal. A:
Chem. 308 (2009) 159.
4-chloro-
decrease in the yield was observed instead (entry 3, Table 3). It can
be seen that in the absence of an -substituent on the styrene
derivative the formation of the trans cyclopropane is always fav-
oured (entries 1, 2 and 5, Table 3). Steric hindrance at the position
a-methylstyrene was employed as substrate, a slight
a
a
does not hamper the reaction and good yields were obtained with
1,1-diphenyl ethylene (entry 4, Table 3).
Even aliphatic alkenes that are generally less reactive in
cyclopropanation reactions, gave excellent results. Though the
time of the cyclopropanation reaction for these substrates
increased slightly if compared to what observed for styrene
derivatives (compare entries 6, 7 and 8, Table 3, with entry 6,
Table 1) yields ranging from good to excellent were always
obtained. Even in these cases trans cyclopropane compounds were
obtained as major products and a remarkable diastereoselectivity
(trans/cis ¼ 10.1) was observed for cyclohexene (entry 7, Table 3).
With 2,5-dimethyl-2,4-hexadiene, an important precursor to
chrysanthemic acid [53], the catalytic reaction yielded the desired
cyclopropanes (cyclopropanation of only one double bond was
observed, [37]) in very good yields (98%) (entry 8, Table 3),
without the need for a large excess of the olefin. Out of the four
possible diastereoisomers that could be obtained in the cyclo-
propanation of (ꢀ)
b-pinene, two major isomers were obtained.
The two major isomers are those deriving from a formal attack of
the carbene moiety on the less sterically hindered side of the
[24] N.S. Youssef, E. El-Zahany, A.M.A. El-Seidy, A. Caselli, S. Fantauzzi, S. Cenini,
Inorg. Chim. Acta 362 (2009) 2006.
[25] M. Belicchi-Ferrari, F. Bisceglie, G. Pelosi, S. Pinelli, P. Tarasconi, Polyhedron 26
(2007) 5150.
[26] I. Boldini, G. Guillemot, A. Caselli, A. Proust, E. Gallo, Adv. Synth. Catal. 352
(2010) 2365.
[27] S. Purohit, A.P. Koley, L.S. Prasad, P.T. Manoharan, S. Ghosh, Inorg. Chem. 28
(1989) 3735.
[28] K.S. Bose, B.C. Sharma, C.C. Patel, Inorg. Chem. 12 (1973) 120.
[29] W. Yi, R.-H. Cao, Z.-Y. Chen, L. Yu, L. Ma, H.-C. Song, Chem. Pharm. Bull. 57
(2009) 1273.
olefin [52]. On the other hand, when (ꢀ)
a-pinene was used as the
substrate, only one diastereoisomer (1R,2R,20R, Ref. [52]) was
isolated as a clean product in good yield (entry 10, Table 3). The
only substrate that failed to give good yields in the present study
was methylfuroate, (entry 11, Table 3). However, in this case, even
if in low yield, only the trans isomer (attack only at the non-
substituted double bond) was detected [54].

N.S. Youssef et al. / Journal of Organometallic Chemistry 714 (2012) 94e103
103
[30] I. Ðilovic, M. Rubcic, V. Vrdoljak, S.K. Pavelic, M. Kralj, I. Piantanida, M. Cindric,
Bioorg. Med. Chem. 16 (2008) 5189.
[41] A.D. Naik, P.A.N. Reddy, M. Nethaji, A.R. Chakravarty, Inorg. Chim. Acta 349
(2003) 149.
[31] T.S. Lobana, A. Sanchez, J.S. Casas, A. Castineiras, J. Sordo, M.S. Garcia-Tasende,
E.M. Vazquez-Lopez, J. Chem. Soc. Dalton Trans. (1997).
[42] M. Flores-Leonar, N. Esturau-Escofet, J.M. Mendez-Stivalet, A. Marin-Becerra,
C. Amador-Bedolla, J. Mol. Struct. 1006 (2011) 600.
[32] The GC analysis was performed on a HP 5890 series II CG equipped with
[43] M.J.M. Campbell, Coord. Chem. Rev. 15 (1975) 279.
[44] K. Nakamoto, Infrared and Raman Spectra of Inorganic and Coordination
Compounds Part B, , In: Applications in Coordination, Organometallic, and
Bioinorganic Chemistry Wiley-Interscience, 1998, p 384.
[45] E. Tas, M. Aslanoglu, A. Kilic, O. Kaplana, H. Temel, J. Chem. Res. 2006 (2006) 242.
[46] V.T. Kasumov, Spectrochim. Acta, Part A 57 (2001) 1649.
[47] A.P. Gulya, V.I. Prisakar, V.I. Tsapkov, S.A. Buracheva, S.N. Spynu,
N.P. Bezhenar, Pharm. Chem. J. 42 (2008) 326.
[48] E. Tas, M. Aslanoglu, M. Ulusoy, H. Temel, J. Coord. Chem. 57 (2004) 677.
[49] S.K. Chawla, N. Sankarraman, J.H. Payer, J. Electron, Spectrosc. Relat. Phenom.
61 (1992) 1.
[50] N. Gunasekaran, P. Ramesh, M.N.G. Ponnuswamy, R. Karvembu, Dalton Trans.
40 (2011) 12519.
a
TCD detector and with a HP-plotQ (divinylbenzene/styrene polymer)
capillary column under He flux. Acetic aldehyde (compared with a pure
sample) was eluted in 2.7 min; ethanol in 3.4 min; acetic acid in 8.1 min. A
blank test with the same solvent used for the reaction was performed: in this
case, a small amount of acetic aldehyde (or methanol, that has the same
retention time under these conditions) was also observed in traces, but the
intensity of this peak, with respect to that of ethanol, was 1/4 than in the
analysis of the reaction mixture.
[33] L. Huang, Y. Chen, G.-Y. Gao, X.P. Zhang, J. Org. Chem. 68 (2003) 8179.
[34] A. Berkessel, P. Kaiser, J. Lex, Chem. Eur. J. 9 (2003) 4746.
[35] S. Fantauzzi, E. Gallo, E. Rose, N. Raoul, A. Caselli, S. Issa, F. Ragaini, S. Cenini,
Organometallics 27 (2008) 6143.
[36] Y. Chen, X.P. Zhang, J. Org. Chem. 72 (2007) 5931.
[51] T.S. Lobana, P. Kumari, R. Sharma, A. Castineiras, R.J. Butcher, T. Akitsu,
Y. Aritake, Dalton Trans. 40 (2011) 3219.
[52] M. Hagar, F. Ragaini, E. Monticelli, A. Caselli, P. Macchi, N. Casati, Chem.
Commun. 46 (2010) 6153.
[53] M. Itagaki, K. Suenobu, Org. Process Res. Dev. 11 (2007) 509.
[54] C. Bohm, O. Reiser, Org. Lett. 3 (2001) 1315.
[37] A.G.M. Barrett, D.C. Braddock, I. Lenoir, H. Tone, J. Org. Chem. 66 (2001) 8260.
[38] Y. Tupolova, V. Lukov, V. Kogan, L. Popov, Russ. J. Coord. Chem. 33 (2007) 301.
[39] A. Cukurovali, I. Yilmaz, S. Kirbag, Transit. Met. Chem. 31 (2006) 207.
[40] S. Floquet, M.-L. Boillot, E. Riviere, F. Varret, K. Boukheddaden, D. Morineau,
P. Negrier, New J. Chem. 27 (2003) 341.