Dibutyltin perfluoroalkanecarboxylates: Synthesis, NMR characterization and in vitro antitumour activity

Article abstract of DOI:10.1016/S0022-328X(00)00367-3

Three dibutyltin perfluoroalkanecarboxylates have been synthesized, characterized by ¹H-, ¹³C-, ¹⁹F- and ¹¹⁷Sn-NMR, M?ssbauer, IR and mass spectroscopy. The structure of tetra-n-butylbis(trifluoroacetato)distann

Full text of DOI:10.1016/S0022-328X(00)00367-3

www.elsevier.nl/locate/jorganchem
Journal of Organometallic Chemistry 608 (2000) 63–70
Dibutyltin perfluoroalkanecarboxylates: synthesis, NMR
characterization and in vitro antitumour activity
Martine Kemmer a,b, Hassan Dalil b, Monique Biesemans b,c, Jose´ C. Martins c,
Bernard Mahieu d, Ernst Horn e, Dick de Vos f, Edward R.T. Tiekink g,
Rudolph Willem b,c, Marcel Gielen a,b,
*
a Uni6ersite´ Libre de Bruxelles, Faculte´ des Sciences, A6enue F.D. Roose6elt 50, B-1050 Brussels, Belgium
b Department of General and Organic Chemistry (AOSC), Faculty of Applied Sciences, Vrije Uni6ersiteit Brussel, Pleinlaan 2,
B-1050 Brussels, Belgium
c High Resolution NMR Centre, Vrije Uni6ersiteit Brussel, Pleinlaan 2, B-1050 Brussels, Belgium
d Catholic Uni6ersity of Lou6ain, CPMC, B-1348 Lou6ain-la-Neu6e, Belgium
e Department of Chemistry, Uni6ersity of Tsukuba, Ibaraki 305, Japan
f Medical Department, PCH Nederland, Pharmachemie B.V., NL-2003 RN Haarlem, The Netherlands
g Department of Chemistry, The Uni6ersity of Adelaide, Adelaide 5005, Australia
Received 27 March 2000
Abstract
Three dibutyltin perfluoroalkanecarboxylates have been synthesized, characterized by 1H-, 13C-, 19F- and 117Sn-NMR,
Mo¨ssbauer, IR and mass spectroscopy. The structure of tetra-n-butylbis(trifluoroacetato)distannoxane has been elucidated by
X-ray crystallography. The in vitro antitumour activity of the three compounds against seven human tumour cell lines was found
to be as high as or even higher than that for reference compounds used clinically. © 2000 Published by Elsevier Science S.A.
Keywords: Dibutyltin perfluoroalkanecarboxylates; NMR; Antitumour activity
compare their in vitro antitumour activities with those
of di-n-butyltin fluorinated aromatic carboxylates. For
this purpose, bis(trifluoroacetato-, -pentafluoropropi-
onato- and -heptafluorobutyrato)tetra-n-butyldistan-
noxanes were synthesized and characterized by 1H-,
13C-, 19F- and 117Sn-NMR as well as Mo¨ssbauer and
electrospray mass spectrometry. Crystallographic data
are also presented for the trifluoroacetate compound.
The trifluoroacetate is mentioned in an earlier study on
a comparative evaluation of its toxic and mutagenic
properties [9].
1. Introduction
Several organotin fluoro-substituted aromatic car-
boxylates are very active in vitro against human cancer
cell lines [1–6]; Tables 1 and 2 summarize their
activities.
Table 1 reveals high antitumour activities [1–4] for a
range of di- and triorganotin fluoro-substituted ben-
zoates against two cell lines and Table 2 shows that
organotin perfluorophenyl-substituted carboxylates are
in general more active than their non-fluorinated
analogs against a broader panel of cell lines [5–7]. Data
available for organometals with fluorinated organic
groups indicated enhanced activity [8]. Hence, we ex-
plored the antitumour activity of some aliphatic di-n-
butyltin perfluoroalkanecarboxylates in order to
2. Results and discussion
2.1. Synthesis
The different reactions that can in principle be used
for the synthesis of the di-n-butyltin perfluoroalkane-
carboxylates are generally characterized by excellent
yields [10–12].
* Corresponding author. Tel.: +32-2-6293279; fax: +32-2-
6293281.
E-mail address: mgielen@vub.ac.be (M. Gielen).
0022-328X/00/$ - see front matter © 2000 Published by Elsevier Science S.A.
PII: S0022-328X(00)00367-3

64
M. Kemmer et al. / Journal of Organometallic Chemistry 608 (2000) 63–70
Table 1
In vitro ID50 inhibition doses (in ng ml−1) against the mammary
tumour MCF-7 and the colon carcinoma WiDr of di- and tri-organ-
otin fluoro-substituted benzoates [1–4]
Compound
MCF-7
WiDr
Fig. 1. Structure of the 1/2 compounds of type a.
(2-FC6H4CO2)2SnBu2
(3-FC6H4CO2)2SnBu2
(4-FC6H4CO2)2SnBu2
(3-CF3C6H4CO2)2SnBu2
(2,3-F2C6H3CO2)2SnBu2
(2,6-F2C6H3CO2)2SnBu2
(3,5-F2C6H3CO2)2SnBu2
(3-FC6H4CO2)2SnEt2
74
39
90
102
38
98
52
1235
11
242
271
309
434
277
326
416
3705
14
3-FC6H4CO2SnPh3
4-FC6H4CO2SnPh3
15
14
2,3-F2C6H3CO2SnPh3
3,5-F2C6H3CO2SnPh3
2-FC6H4CO2SnBu3
17
18
35
38
22
91
81
13
14
17
B1
26
2
330
360
134
353
227
277
2,5-F2C6H3CO2SnBu3
2,6-F2C6H3CO2SnBu3
[(2-FC6H4CO2SnBu2)2O]2
[(4-FC6H4CO2SnBu2)2O]2
[(2,3-F2C6H3CO2SnBu2)2O]2
[(2,5-F2C6H3CO2SnBu2)2O]2
[(2,6-F2C6H3CO2SnBu2)2O]2
[(3,5-F2C6H3CO2SnBu2)2O]2
Fig. 2. Structure of the 1/1 compounds of type b.
32
25
43
in a tin/acid molar ratio 1/2 or 1/1 providing respec-
tively a di-n-butyltin dicarboxylate of type a (Fig. 1):
(Bu2SnOPr)2O+4RCOOH
“2Bu2Sn(OOCR)2+2PrOH+H2O
The first step of the one-pot synthesis [10] is the
preparation of tetra-n-butyldipropoxydistannoxane
or a dimeric tetra-n-butyldicarboxylatodistannoxane
of type b (Fig. 2):
2Bu2SnO+2PrOH“(Bu2SnOPr)2O+H2O
2(Bu2SnOPr)2O+4RCOOH
Usually, the second step is the reaction at room
temperature of this distannoxane with a carboxylic acid
“[(Bu2SnOOCR)2O]2+4PrOH
Table 2
In vitro ID50 inhibition doses (ng ml−1) against six human tumour cell lines, two mammary cancers (MCF-7, EVSA-T), a colon carcinoma
(WiDr), an ovarian cancer (IGROV), a melanoma (M19 MEL) and a renal cancer (A 498) of di- and triorganotin pentafluorophenyl-substituted
carboxylates and their non-fluorinated analogues [5–7], and of some clinically used reference compounds
Compound
MCF-7
EVSA-T
WiDr
IGROV
M19 MEL
A 498
{[Bu2Sn(O2CC6F5)]2O}2
{[Bu2Sn(O2CCH2C6F5)]2O}2
{[Bu2Sn(O2CCHꢀCHC6F5)]2O}2
Bu2Sn(O2CCH2C6F5)2
Bu3SnO2CC6F5
Bu3SnO2CCH2C6F5
Bu3SnO2CCHꢀCHC6H5
Bu3SnO2CC6H5
Bu3SnO2CCH2C6H5
Bu3SnO2CCHꢀCHC6H5
Ph3SnO2CC6F5
Ph3SnO2CCH2C6F5
Ph3SnO2CCHꢀCHC6F5
Ph3SnO2CC6H5
44
55
32
10
15
14
13
130
42
23
45
14
20
650
4
39
43
37
214
275
234
145
20
12
14
70
50
45
90
17
30
500
5
22
1500
967
225
11
53
60
41
20
25
12
11
100
50
38
44
20
34
470
5
86
114
66
36
60
52
45
110
85
47
130
18
40
800
13
30
5300
558
442
16
76
105
135
51
50
51
54
190
75
63
120
30
19
B3
B3
B3
15
11
B3
17
6
9
30
950
21
110
B3
3
1100
422
475
8
Ph3SnO2CCH2C6H5
Ph3SnO2CCHꢀCHC6H5
Carboplatin
Cis-platin
5-Fluorouracil
12
16
52
5500
699
750
10
780
169
297
60
3500
2253
143
90
Doxorubicine

M. Kemmer et al. / Journal of Organometallic Chemistry 608 (2000) 63–70
65
The compounds reported, tetra-n-butylbis(trifluoro-
acetato)distannoxane (1), tetra-n-butylbis(pentafluoro-
propionato)distannoxane (2) and tetra-n-butylbis-
(heptafluorobutyrato) distannoxane (3) are of type b.
Surprisingly, they could not be obtained when using a
1/1 molar ratio, but only from 1/2 reaction conditions.
The compounds of type a could not be obtained in
pure form utilizing this method. Di-n-butyltin bis(hep-
tafluorobutyrate) (3a) was the only compound of type a
that could be obtained, however, mixed with its hydrol-
ysis product 3. Compound 3a appears indeed very
sensitive to moisture, since it is hydrolysed in situ into
the corresponding type b compound 3. This is ac-
counted for by the reaction:
Table 3
1H-NMR data in CDCl3 for compounds 1–3; chemical shift in ppm,
3J(1H–1H) homonuclear coupling constants in Hz in parentheses
(t=triplet, q=quartet; m=complex pattern)
1a
2
3
H (a), H (b)
H (g)
H (d)
1.56–1.75 m
1.39 tq (7.7)
0.93 t (7)
1.55–1.75 m
1.38 tq (7.7)
0.92 t (7)
1.52–1.74 m
1.35 tq (7.7)
0.89 t (7)
0.92 t (7)
0.91 t (7)
0.88 t (7)
4R2Sn(OOCR%)2+2H2O
a In agreement with literature data [12].
“[(R2SnOOCR%)2O]2+4R%COOH
Table 4
All compounds decompose rapidly if they are not
stored in dry media. The literature describes the decom-
position of 1 into (CF3COO)SnBu2OSnBu2(OH) [11].
13C-NMR data in CDCl3 for the compounds 1–3; nJ(13C–117/119Sn)
values are given in parentheses; a single value is given when the
nJ(13C–117Sn) and nJ(13C–119Sn) couplings are unresolved
1
2
3
2.2. Characterization
CH2 (a)
CH2 (a)
CH2 (b)
CH2 (b)
CH2 (g)
CH2 (g)
CH3 (d)
30.1 (627/658)
28.6 (596/622)
27.8
30.6 (646/672)
29.1 (604/631)
28.2
30.0 (644/677)
28.6 (602/630)
27.6
2.2.1. NMR spectroscopy
Compounds 1–3 were characterized by 1H-, 13C-,
27.4
27.8
27.3
27.1 (116/121)
27.1 (136/141)
13.9
27.1 (115/121)
27.1 (145)
13.8
26.6 (116)
26.5 (142)
13.2
19F- and 117Sn-NMR in CDCl3 solution. 2D H–13C
1
HMBC (heteronuclear multiple bond correlation)
[13,14] and HMQC (heteronuclear multiple quantum
coherence) NMR spectra [15] were recorded for com-
pound 3 in order to assign the 1H and 13C resonances of
Table 5
19F chemical shifts with respect to CFCl3 and 13C chemical shifts for
the n-butyl groups. The basic H and 13C-NMR data
1
the perfluoroalkanecarboxylate groups of compounds 1–3 in CDCl3
for the compounds 1 to 3 are given in Tables 3 and 4.
2D 13C–19F HMQC spectroscopy was needed for
compound 2 in order to achieve simultaneous assign-
ment of the 19F and 13C resonances of the pe-
rfluoroalkanecarboxylate moiety. This was even more
1
2
3
CF3
l
l
13C
19F
−77.8
114.4 (q)
289
−85.4
118.7 (qt)
286
−83.5
117.6 (qt)
288
1J(13C–19F)
2J(13C–19F)
CF2 (b) l 19F
necessary because in the H decoupled 13C spectrum,
1
35
34
the fluorinated carbons give rise to noisy overlapping
multiplets which are difficult to unravel. The resulting
19F and 13C assignments are based on the assumption
−112.0
108.5 (ttq)
266
34; 34
−129.1
108.2 (tt)
266
34
162.4 (t)
28
l
13C
1J(13C–19F)
2J(13C–19F)
CF2 (a) l 19F
2
that the J(13C–19F) coupling constants are larger than
−122.7
107.0 (tq)
265
39
163.2 (t)
27
3
the J(13C–19F) ones, as supported by literature data
l
13C
n
[16]. Subsequently, the J(13C–19F) coupling constants
1J(13C–19F)
2J(13C–19F)
COO l 13C
2J(13C–19F)
as well as 13C chemical shifts of the fluorinated carbons
1
160.3 (t)
37
were assigned from non H-decoupled 13C spectra, in
order to optimize the dynamic range. Table 5 reports
13C and 19F chemical shifts as well as nJ(13C–19F)
coupling constants for these resonances.
Table 6
Compound 3a exhibits a single 117Sn resonance at
167.0 ppm, characteristic for type a compounds. For
compounds 1–3, two 117Sn resonances are observed
(Table 6) and assigned to the endocyclic and exocyclic
tin atoms (Fig. 2). These data are in agreement with
literature data on type b compounds described previ-
ously [17–20]. Noticeable, however, is a significant
deviation of the 117Sn chemical shifts of compound
117Sn chemical shifts in ppm of compounds 1–3 (CDCl3), with
unresolved 2J(117SnO119Sn) and 2J(117SnO117Sn) coupling constants
1
2
3
l(117Sn)
−169.0
−175.2
113
−175.0
−175.9
122
−174.3
−175.0
122
2J(117SnO119/117Sn)

66
M. Kemmer et al. / Journal of Organometallic Chemistry 608 (2000) 63–70
(1694–1701 cm−1) and bidentate perfluorocarboxy-
lates (1656–1663 cm−1), respectively, in agreement
with literature data [21].
Table 7
Mo¨ssbauer data of compounds 1–3 (quadrupole splitting QS, isomer
shift IS, and line widths Y1 and Y2) in mm s−1 relative to calcium
stannate
QS
IS
Y1
Y2
2.2.5. X-ray crystal structure of compound 1
The
molecular
structure
of
1,
{[Bu2Sn-
1
2
3
4.14
3.37
4.20
3.29
3.72
1.43
1.45
1.45
1.42
1.45
0.92
0.92
1.03
1.03
0.95
0.92
0.92
1.03
1.03
0.99
(O2CCF3)2]2O}2 is shown in Fig. 3 and selected inter-
atomic parameters are collected in Table 10. The
structure determination is less than optimal owing to
significant thermal motion associated with the light
atom positions, however, the molecular connectivity is
unambiguous. The centrosymmetric structure features
a central Bu4Sn2O2 core to which are linked two
Bu2Sn entities with the result that the O(1) atom is
three coordinate. Links between the endo- and exo-
cyclic tin centres are provided by bridging carboxylate
ligands that form slightly asymmetric bond distances:
Table 8
Fragment-ions observed in electrospray mass spectra for compounds
1–3
Fragment
1
2
3
RCOO(SnBu2O)2Bu2Sn.H2O+
HOSnBu2OSnBu2OSnBu2+
(RCOO)Bu2SnOSnBu2+
(HO)Bu2SnOSnBu+2
897
751
647
501
751
697
501
751
501
,
,
Sn(1)–O(2) 2.373(5) A and Sn(2)–O(3) 2.233(6) A.
The exocyclic tin atom is also coordinated by the O(4)
atom so that Sn(2)–O(4) is 2.200(3) A. The
,
,
Sn(2)···O(5) separation of 3.071(4) A represents a very
weak bonding interaction as the close contact ap-
proach by the O(5) atom has a stereochemical effect,
distorting the coordination geometry about the Sn(2)
atom so that the C–Sn(2)–C angle is expanded to
142.1(3)°. A similar intramolecular Sn(1)···O(4) contact
Table 9
IR w(COO)asym frequencies observed for compounds 1–3 (KBr pel-
lets)
1
2
3
1660
1694
1656
1695
1663
1701
,
of 2.889(3) A results in a comparable expansion in the
C–Sn(1)–C angle, i.e. 143.1(2)°. Each of the indepen-
dent tin atoms in the structure exists in a distorted
trigonal bipyramidal geometry with a C2O trigonal
plane and axial positions occupied by oxygen atoms.
The structure reported here for compound 1 resembles
closely the predominant motif found for compounds
of the general formula {[R2Sn(O2CR%)2]2O}2 [22,23].
1 with respect to literature data [12] (−178.6 and
−177.5 ppm). These data probably correspond to a
decomposition product because, following our experi-
ence on the stability of such compounds, the experi-
mental procedures used by the authors [10,12] could
lead to such a decomposition.
2.2.6. In 6itro antitumour screening
The compounds 1–3 were screened against seven
human cancer cell lines, the six lines mentioned in
Table 2 and H226, a lung cancer. The screening was
performed with aqueous solutions containing 1% of
DMSO or ethanol. The ID50 values observed are given
in Table 11 and compared to those of some clinically
used reference compounds [24,25], i.e. doxorubicine
(DOX), cis-platin (CPT), 5-fluorouracil (5-FU),
methotrexate (MTX) and etoposide (ETO).
2.2.2. Mo¨ssbauer spectroscopy
The Mo¨ssbauer parameters (Table 7) of compounds
1 and 2 reveal two types of tin sites, one with a
QS=3.4 mm s−1 and another one with QS=4.2 mm
s
−1, characteristic for larger (alkyl)2Sn bond angles
than in an ideal trigonal bipyramid, as effectively ob-
served for 1 in the X-ray structure (see Section 2.2.5).
The screening results indicate that the compounds
1–3 are more active than cis-platin, 5-fluorouracil and
etoposide. They are, however, globally less active than
doxorubicine and methotrexate, except for the H226
tumour against which they are strikingly active.
Against MCF-7 and WiDr, their in vitro activity is
comparable to that of di-n-butyltin fluorinated aro-
matic carboxylates, but lower than those of the tri-n-
butyl- and triphenyltin ones (see Tables 1 and 2).
2.2.3. Electrospray mass spectrometry
The most important fragment ions in the monoiso-
1
topic mass spectra (12C, H, 19F, 16O, 120Sn) recorded
in the cationic mode from water–acetonitrile solutions
of compounds 1–3 are described in Table 8.
2.2.4. IR spectroscopy
The two w(COO) absorptions observed in the IR
spectrum (Table 9) are assigned to monodentate

M. Kemmer et al. / Journal of Organometallic Chemistry 608 (2000) 63–70
67
Fig. 3. Molecular structure and crystallographic numbering scheme for compound 1, {[Bu2Sn(O2CCF3)2]2O}2.
3. Experimental
Table 10
,
Selected geometric parameters (A, °) for compound 1,
{[Bu2Sn(O2CCF3)2]2O}2
3.1. Syntheses
Sn(1)–O(1)
Sn(1)–O(2)
Sn(1)–C(21)
Sn(2)–O(3)
Sn(2)–C(31)
2.058(3)
2.373(5)
2.104(4)
2.233(6)
2.130(5)
Sn(1)–O(1)ia1
Sn(1)–C(11)
Sn(2)–O(1)
Sn(2)–O(4)
Sn(2)–C(41)
2.170(3)
2.121(5)
2.023(3)
2.200(3)
2.101(6)
Di-n-butyltin oxide is reacted with an excess (4 ml) of
propanol in 150 ml of benzene. After 3 h of reflux with
elimination (Dean–Stark) of the ternary azeotrope ben-
zene/water/propanol, the reaction mixture is allowed to
cool down to room temperature; the appropriate
amount of acid is subsequently added slowly and the
mixture further stirred for one night. The solvent is
evaporated. Purification is achieved by crystallization
from a suitable solvent (Table 12) under a dry atmo-
sphere. Compound 3 required successive recrystalliza-
tions in undried heptane in order to hydrolyze 3a
completely into 3. Heating must be avoided in order to
prevent decomposition. Anal. Found (calc.): 1, C 34.12
(33.93); H 4.92 (5.13); F 15.46 (16.10); Sn, 33.56
(33.93); 2, C 32.82 (32.71); H 4.04 (4.49); F 22.77
(23.51); Sn 29.47 (29.38); 3, C 31.99 (31.75); H 3.62
(4.00); F 28.36 (29.29); Sn 26.08 (26.15).
O(1)–Sn(1)–O(1)i
O(1)–Sn(1)–C(11)
O(1)I–Sn(1)–O(2)
O(1)I–Sn(1)–C(21)
O(2)–Sn(1)–C(21)
O(1)–Sn(2)–O(3)
O(1)–Sn(2)–C(31)
O(3)–Sn(2)–O(4)
O(3)–Sn(2)–C(41)
O(4)–Sn(2)–C(41)
Sn(1)–O(1)–Sn(1)i
Sn(1)–O(1)–Sn(2)i
Sn(1)–O(3)–C(1)
77.8(1)
110.2(1)
167.7(1)
97.7(2)
88.5(2)
91.9(2)
107.9(2)
172.7(2)
89.0(2)
O(1)–Sn(1)–O(2)
O(1)–Sn(1)–C(21)
O(1)I–Sn(1)–C(11)
O(2)–Sn(1)–C(11)
90.3(1)
105.7(2)
98.0(2)
83.0(2)
C(11)–Sn(1)–C(21) 143.1(2)
O(1)–Sn(2)–O(4)
O(1)–Sn(2)–C(41)
O(3)–Sn(2)–C(31)
O(4)–Sn(2)–C(31)
80.8(1)
109.6(2)
84.7(2)
98.3(2)
92.5(2)
C(31)–Sn(2)–C(41) 142.1(3)
102.2(1)
123.2(1)
139.4(5)
Sn(1)–O(1)–Sn(2)
Sn(1)–O(2)–C(1)
Sn(2)–O(4)–C(3)
134.5(2)
135.8(4)
114.3(3)
a Symmetry operation i: −x, −y, 1−z.

68
M. Kemmer et al. / Journal of Organometallic Chemistry 608 (2000) 63–70
Table 11
Inhibition doses ID50 in vitro (ng ml−1) of compounds 1, 2 and 3 against seven human tumour cell lines, two mammary cancers (MCF-7,
EVSA-T), a colon carcinoma (WiDr), an ovarian cancer (IGROV), a melanoma (M19 MEL), a renal cancer (A 498) and a non small cell lung
cancer (H226)
MCF-7
EVSA-T
WiDr
IGROV
M19 MEL
A 498
H226
1
2
3
57
64
191
10
699
750
18
79
93
135
8
422
475
5
354
339
340
11
967
225
B3
150
113
121
123
60
169
297
7
70
101
141
16
558
442
23
151
131
177
90
2253
143
37
147
120
213
199
3269
340
DOX
CPT
5-FU
MTX
ETO
2287
3934
2594
317
580
505
1314
Table 12
Reactant amounts, yields, recrystallization solvents and melting points of the diorganotin carboxylates 1–3
Compound
synthesized
Amount of acid (g (mmol))
Amount of dibutyltin oxide (g (mmol))
Yield (%)
Recrystallization solvent
1
2
3
1.37 (12.0)
1.99 (12.1)
2.80 (12.1)
1.51 (6.07)
1.51 (6.07)
1.62 (6.51)
95
90
72
hexane
hexane
heptane
3.2. NMR measurements
tively). The 119Sn, 117Sn and 19F reference frequencies
were calculated from the absolute references J(119Sn)=
37.290 665 MHz, J(117Sn)=35.632 295 MHz [32,33]
and J(19F)=94.094 003 MHz [32].
The 19F-NMR spectra as well as most other 1D
spectra were acquired on Bruker AC250 and DRX250
instruments equipped with a Quattro probe tuned to
250.13, 62.93, 89.15, and 235.19 MHz for 1H, 13C,
117Sn, and 19F nuclei, respectively. The 19F detected 2D
19F–13C HMQC correlation spectra were acquired on
the AC250 spectrometer, as previously [6], without 13C
decoupling using the resources of two NMR spectrome-
ters, in a ‘master–slave’ relationship, as described by
Berger [26,27]. 19F acquisition was performed on the
3.3. Mass spectrometry
The electrospray mass spectra were recorded in the
cationic mode on a Micromass Quattro II instrument
coupled with a Masslynx system (ionization in an elec-
tric field of 3.5 kV; source temperature 80°C; source
pressure 1 atm; analyser pressure 10−5 mbar) [34,35].
1
AC250 instrument with the H coil of the QNP probe
being tuned to the 19F resonance frequency, while 13C
pulses were generated on the AMX500 instrument un-
der direct control of the AC250 one and sent to the
X-coil of the QNP probe. The 19F–13C HMQC spectra
were acquired in the phase-sensitive mode using TPPI
[28], and processed as described elsewhere [6]. All 2D-
NMR spectra and some of the 1D spectra were
recorded on a Bruker AMX500 spectrometer operating
3.4. Mo¨ssbauer spectroscopy
The Mo¨ssbauer spectra were recorded as described
previously [36].
3.5. X-ray crystallography
Intensity data for
a colourless crystal of 1,
1
at 500.13, 125.77, and 186.50 MHz for the H, 13C, and
{[Bu2Sn(O2CCF3)2]2O}2, were collected at room tem-
perature on a Rigaku AFC7R diffractometer employ-
ing Mo–Ka radiation (u=0.71073 A) and the ꢀ:2q
scan technique to 2qmax of 60.0°. The data set was
corrected for Lorentz and polarization effects [37] as
well as for absorption based on psi-scans. Data that
satisfied the I]3.0|(I) criterion of observability were
used in the subsequent analysis. Crystal data and refin-
ement details are given in Table 13.
119Sn nuclei, respectively. 13C and 119Sn BB proton-de-
coupled 1D spectra, 13C DEPT spectra as well as the
,
1
proton-detected 1D and 2D H–13C HMQC [15] and
HMBC [13,14] spectra were acquired using the pulse
sequences of the Bruker program library, adapted to
include gradient pulses [29,30] as described recently
1
[31]. H and 13C chemical shifts were referenced to the
1
standard Me4Si scale from respectively residual H and
13C–2H solvent resonances of chloroform (CHCl3, 7.23
The structure was solved by direct methods [38] and
refined by a full-matrix least-squares procedure based
1
and CDCl3, 77.0 ppm for H and 13C nuclei, respec-

M. Kemmer et al. / Journal of Organometallic Chemistry 608 (2000) 63–70
69
Table 13
Crystallographic data for compound 1, {[Bu2Sn(O2CCF3)2]2O}2
4. Supplementary material
Crystallographic data for compound 1 have been
deposited with the Cambridge Crystallographic Data
Centre, CCDC No. 141898. Copies of this information
may be obtained free of charge from The Director,
CCDC, 12 Union Road, Cambridge CB2 1EZ, UK
Formula
C40H72F12O10Sn4
1415.7
Triclinic
Formula weight
Crystal system
Space group
(
P1
Unit cell dimensions
,
a (A)
12.041(2)
14.423(4)
10.182(2)
106.98(2)
101.87(1)
111.84(2)
1468.1(8)
1
0.2×0.4×0.7
1.601
700
7140
6791
(fax:
+44-1223-336033;
e-mail:
deposit@ccdc.
,
b (A)
cam.ac.uk or www: http://www.ccdc.cam.ac.uk).
,
c (A)
h (°)
i (°)
k (°)
Acknowledgements
3
,
V (A )
Z
Crystal size (mm3)
We thank Eng I. Verbruggen for recording the NMR
spectra and Dr G. Laus and M. Desmet for the mass
spectra. We are grateful to H.J. Kolker, Dr J. Verweij,
Professor Dr G. Stoter, Dr J.H.M. Schellens, Labora-
tory of Experimental Chemotherapy and Pharmacol-
ogy, Department of Medical Oncology, Rotterdam
Cancer Institute, NL-3008 AE, Rotterdam, The
Netherlands, for performing the in vitro tests. This
research was supported by the Belgian Fonds voor
Kollektief Fundamenteel Onderzoek (F.K.F.O. grant
no. 2.0094.94, R.W., M.B.), the Belgian Nationale Lo-
terij (grant no. 9.0006.93, R.W., M.B.), and the Fund
for Scientific Research Flanders (Belgium, grants nos.
G.0192.98, R.W., M.B. and G.0074.00, M.G.), the
Ministe`re de l’Education Nationale du Luxembourg
(grants nos. BFR93/051, BPU96/130, BPU97/138,
BPU98/071 (M.K.) and the Action Vaincre le Cancer
du Luxembourg (M.K.) and the Australian Research
Council (E.R.T.T.). J.C.M. is a Postdoctoral Fellow of
the Fund for Scientific Research Flanders (Belgium)
(FWO).
Dcalc (g cm−3
F(000)
)
No. of data collected
No. of unique data
No. of reflections with I53.0|(I)
R
4136
0.067
0.069
0.59
Rw
−3
,
Residual zmax (e A
)
on F [37]. The Sn and O atoms were refined with
anisotropic displacement parameters, C and F atoms
were refined isotropically, and hydrogen atoms were
included in the model at their calculated positions
,
(C–H 0.95 A). The atoms comprising the molecule are
subject to high thermal motion. Attempts to collect
data at reduced temperatures were unsuccessful as the
crystals fragmented, a frequent experience encountered
for fluorinated samples of the type reported here. Two
positions were resolved for the F atoms comprising
each CF3 group. These were refined with site occupancy
factors of 0.6 and 0.4 (primed atoms) determined from
the refinement. The carbon atoms were refined so that
References
[1] M. Gielen, A. El Khloufi, M. Biesemans, F. Kayser, R. Willem,
Appl. Organomet. Chem. 7 (1993) 201.
[2] M. Gielen, M. Biesemans, A. El Khloufi, J. Meunier-Piret, R.
Willem, J. Fluorine Chem. 64 (1993) 279.
,
C–C bond distances were constrained to 1.50 A. The
geometric parameters determined for the light atom
positions are thus unreliable, however, the molecular
connectivity has been established by this study. The
refinement was continued until convergence employing
a weighting scheme of the form 1/[|2(F)+0.00002ꢀFꢀ2].
Selected interatomic parameters are collected in Table
13 and the crystallographic numbering scheme em-
ployed is shown in Fig. 3 which was drawn with ORTEP
[39] at 25% probability ellipsoids.
[3] M. Gielen, A. El Khloufi, D. de Vos, H.J. Kolker, J.H.M.
Schellens, R. Willem, Bull. Soc. Chim. Belg. 102 (1993) 761.
[4] M. Gielen, E.R.T. Tiekink, A. Bouhdid, D. de Vos, M. Biese-
mans, I. Verbruggen, R. Willem, Appl. Organomet. Chem. 9
(1995) 639.
[5] M. Gielen, A. El Khloufi, M. Biesemans, R. Willem, Appl.
Organomet. Chem. 7 (1993) 119.
[6] R. Willem, A. Bouhdid, M. Biesemans, J.C. Martins, D. de Vos,
E.R.T. Tiekink, M. Gielen, J. Organomet. Chem. 514 (1996) 203.
[7] R. Willem, A. Bouhdid, B. Mahieu, L. Ghys, M. Biesemans,
E.R.T. Tiekink, D. de Vos, M. Gielen, J. Organomet. Chem. 531
(1997) 151.
3.6. Antitumour screening
[8] H. Plenio, Chem. Rev. 97 (1997) 3363.
[9] N.N. Belyaeva, T.A. Bystrova, Y.A. Rezvazova, V.I.
Arkhangl’skii, Gig. Sanit. 5 (1976) 10; Chem. Abstr. 85 (1976)
104765z.
The protocol followed for the antitumour screenings
has already been reported [24,25].

70
M. Kemmer et al. / Journal of Organometallic Chemistry 608 (2000) 63–70
[10] A.G. Davies, D.C. Kleinschmidt, P.R. Palan, S.C. Vasishtha, J.
Chem. Soc. (1971) 3972.
[11] V. Peruzzo, G. Tagliavini, J. Organomet. Chem. 66 (1974) 437.
[12] R.P. Houghton, A.W. Mulvaney, J. Organomet. Chem. 517
(1996) 107.
[13] A. Bax, M.F. Summers, J. Am. Chem. Soc. 108 (1986) 2093.
[14] A. Bax, M.F. Summers, J. Magn. Reson. 67 (1986) 565.
[15] A. Bax, R.H. Griffey, B.H. Hawkins, J. Magn. Reson. 55 (1983)
301.
[16] H.-O. Kalinowski, S. Berger, S. Braun, Carbon 13 NMR Spec-
troscopy, Wiley, 1988, pp. 576–586.
[26] S. Berger, J. Fluorine Chem. 72 (1995) 117.
[27] S. Berger, N. Wetzel, TAMU-NMR Newslett. (1994) 12.
[28] D. Marion, K. Wu¨thrich, Biochem. Biophys. Res. Commun. 113
(1983) 967.
[29] J. Keeler, R.T. Clowes, A.L. Davis, E.D. Laue, Methods Enzy-
mol. 239 (1994) 145.
[30] G.W. Vuister, R. Boelens, R. Kaptein, R.E. Hurd, B.K. John,
P.C.M. Van Zijl, J. Am. Chem. Soc. 113 (1991) 9688.
[31] R. Willem, A. Bouhdid, F. Kayser, A. Delmotte, M. Gielen, J.C.
Martins, M. Biesemans, B. Mahieu, E.R.T. Tiekink,
Organometallics 15 (1996) 1920.
[17] M. Gielen, E.R.T. Tiekink, A. Bouhdid, D. de Vos, M. Biese-
mans, I. Verbruggen, R. Willem, Appl. Organomet. Chem. 9
(1995) 639.
[18] M. Gielen, A. Bouhdid, F. Kayser, M. Biesemans, D. de Vos, B.
Mahieu, R. Willem, Appl. Organomet. Chem. 9 (1995) 251.
[19] M. Gielen, J. Meunier-Piret, M. Biesemans, R. Willem, A. El
Khloufi, Appl. Organomet. Chem. 6 (1992) 59.
[20] A. Meriem, R. Willem, M. Biesemans, B. Mahieu, D. de Vos, P.
Lelieveld, M. Gielen, Appl. Organomet. Chem. 5 (1991) 195.
[21] R. Graziani, G. Bombieri, E. Forselini, P. Furlan, V. Peruzzo,
G. Tagliavini, J. Organomet. Chem. 125 (1977) 43.
[22] E.R.T. Tiekink, Appl. Organomet. Chem. 5 (1991) 1.
[23] E.R.T. Tiekink, Trends Organomet. Chem. 1 (1994) 71.
[24] V.T. DeVita Jr., S. Hellman, S.A. Rosenberg (Eds.), Cancer:
Principles and Practice of Oncology, Lippincott-Raven, Philadel-
phia, PA, 5th ed., 1997.
[32] J. Mason, Multinuclear NMR, Plenum Press, New York, 1987,
pp. 625–629.
[33] A.G. Davies, P.G. Harrison, J.D. Kennedy, R.J. Puddephatt,
T.N. Mitchell, W.J. McFarlane, J. Chem. Soc. A (1969) 1136.
[34] R.B. Costy, J. Tanura, B. Brusselman, Anal. Chem. 64 (1992)
1561.
[35] G. Lawson, R.H. Dahm, N. Ostah, E.D. Woodland, Appl.
Organomet. Chem. 10 (1996) 125.
[36] M. Bouaˆlam, R. Willem, M. Biesemans, B. Mahieu, J. Meunier-
Piret, M. Gielen, Main Group Met. Chem. 14 (1991) 41.
[37] TEXSAN, Structure analysis package, Molecular Structure Corpo-
ration, Woodlands, TX, 1992.
[38] P.T. Beurskens, G. Admiraal, G. Beurskens, W.P. Bosman, S.
Garc´ıa-Granda, J.M.M. Smits, C. Smykalla, The DIRDIF pro-
gram system, Technical Report of the Crystallography Labora-
tory, University of Nijmegen, The Netherlands, 1992.
[39] C.K. Johnson, ORTEP-II, Report ORNL-5138, Oak Ridge Na-
tional Laboratory, Oak Ridge, TN, 1976.
[25] Y.P. Kepers, G.J. Peters, J. Van Ark-Otte, B. Winograd, H.M.
Pinedo, Eur. J. Cancer 27 (1991) 897.
.