Mononuclear ferrocenophane structural motifs with two thiourea arms acting as a dual binding site for anions and cations

Article abstract of DOI:10.1021/ic801879x

The synthesis of a new type of mononuclear ferrocenophane-based thiourea, in which the ferrocene moiety is simultaneously attached to two thiourea groups directly from 1,1′-bis(isothiocyanato)ferrocene, is reported. These nitrogen-rich structural motifs s

Full text of DOI:10.1021/ic801879x

Inorg. Chem. 2009, 48, 1566-1576
Mononuclear Ferrocenophane Structural Motifs with Two Thiourea Arms
Acting as a Dual Binding Site for Anions and Cations
Francisco Oto´n,† Arturo Espinosa,† Alberto Ta´rraga,*,† Imma Ratera,‡ Klaus Wurst,§ Jaime Veciana,*,‡
and Pedro Molina*,†
Departamento de Qu´ımica Orga´nica, Facultad de Qu´ımica, Campus de Espinardo, UniVersidad de
Murcia, E-30100 Murcia, Spain, Institut de Ciencia de Materials de Barcelona (CSIC)/
CIBER-BBN, Bellaterra, E-08193 Barcelona, Spain, and Institut fu¨r Allgemeine Anorganische and
Teoretische, UniVersita¨t Innsbru¨ck, Innrain 52a,
A-6020 Innsbru¨ck, Austria
Received October 2, 2008
The synthesis of a new type of mononuclear ferrocenophane-based thiourea, in which the ferrocene moiety is
simultaneously attached to two thiourea groups directly from 1,1′-bis(isothiocyanato)ferrocene, is reported. These
nitrogen-rich structural motifs show remarkable ion-sensing properties because of the presence of the redox active
ferrocene unit and the thiourea bridges, which unexpectedly act as a dual binding site for anions and metal ions.
They display a selective downfield shift of the thiourea protons and a remarkable cathodic shift of the ferrocene/
ferrocenium redox couple with F-, AcO-, H2PO4-, and HP2O73- anions, whereas the selective recognition of Hg2+
metal cations is achieved either by electrochemical or by spectral measurements. The preferred binding modes are
proposed for the most representative complexes by means of density functional theory based theoretical calculations
showing the Janus-like faces of the receptor.
Introduction
In particular, acyclic and macrocyclic amide- and urea-
functionalized ferrocene derivatives have all been shown to
undergo substantial cathodic perturbations of the respective
metallocene redox couple in the presence of a variety of
anions of biological and environmental importance.1b,2,3
However, there is only one example of thiourea/ferrocene
redox-active cationophore, although it could not be exploited
in the selective electrochemical sensing of anions.4
Artificial neutral receptors for anions became attractive
targets for study because of their analogy to natural systems
and because their selectivities were higher than in the case
of charged ligands. A number of anion receptors have been
reported which have various functional groups acting as
anion binding sites. In neutral hosts, anion binding is usually
accomplished through hydrogen bonds, and in this context,
thiourea subunits are currently used in the design of neutral
receptors for anions, owing to their ability to act as H-bond
donors.1
Regarding the cation-sensing properties of the thiourea
functionality and on the basis of the strong thiophilic affinity
of Hg(II), some selective chemodosimeters for Hg(II) utiliz-
The redox behavior of a ferrocene moiety in the selective
electrochemical sensing of anions has also been exploited.2
(2) (a) Beer, P. D.; Cadman, J. Coord. Chem. ReV. 2000, 205, 131–155.
(b) Beer, P. D.; Hayes, E. J. Coord. Chem. ReV. 2003, 240, 167–189.
(c) Beer, P. D.; Bayly, S. R. Top. Curr. Chem. 2005, 255, 125–162.
(3) (a) Oto´n, F.; Ta´rraga, A.; Velasco, M. D.; Espinosa, A.; Molina, P.
Chem. Commun. 2004, 1658–1659. (b) Oto´n, F.; Ta´rraga, A.; Velasco,
M. D.; Molina, P. Dalton Trans. 2005, 1159–1161. (c) Oto´n, F.;
Ta´rraga, A.; Espinosa, A.; Velasco, M. D.; Bautista, D.; Molina, P. J.
Org. Chem. 2005, 70, 6603–6608. (d) Oto´n, F.; Ta´rraga, A.; Espinosa,
A.; Velasco, M. D.; Molina, P. J. Org. Chem. 2006, 71, 4590–4598.
(e) Oto´n, F.; Ta´rraga, A.; Espinosa, A.; Velasco, M. D.; Molina, P.
Dalton Trans. 2006, 3685–3692.
* To whom correspondence should be addressed. E-mail: pmolina@
um.es (P.M.), atarraga@um.es (A.T.), vecianaj@icmab.es (J.V.).
†
Universidad de Murcia.
Institut de Ciéncia de Materials de Barcelona (CSIC).
Universita¨t Innsbru¨ck.
‡
§
(1) (a) Gale, P. A. Coord. Chem. ReV. 2003, 240, 1–226. (b) Beer, P. D.;
Gale, P. A. Angew. Chem., Int. Ed. 2001, 40, 486–516. (c) Schmidtch-
en, F. P.; Berger, M. Chem. ReV. 1997, 97, 1609–1646. (d) Sessler,
J. L.; Gale, P. A.; Cho, W.-S. Anion Receptor Chemistry; RSC
Publishing: Cambridge, 2006, pp 193-202. (e) Gunlaugsson, T.;
Glynn, M.; Tocci, G. M.; Kruger, P. E.; Pfeffer, F. M. Coord. Chem.
ReV. 2006, 250, 3094–3117.
(4) Beer, P. D.; Cavis, J. J.; Drillma-Milgrom, A.; Szemes, F. Chem.
Commun. 2002, 1716–1717.
1566 Inorganic Chemistry, Vol. 48, No. 4, 2009
10.1021/ic801879x CCC: $40.75  2009 American Chemical Society
Published on Web 01/22/2009

Mononuclear Ferrocenophane Structural Motifs
ing desulfurization reaction of thioureas have been reported.5
However, no examples dealing with the use of thiourea
derivatives as chemosensor molecules for metal cations have
been reported, to the best of our knowledge.
carried out using the same salt solution but with pure solvent, as
opposed to the solution containing the receptor. UV-vis titrations
were performed as follows: a solution of the receptor in DMSO (c
) 1 × 10-4 M) was prepared, and it was titrated with the
appropriate cation salts at room temperature. The titrations were
monitored using a VARIAN 5000 spectrophotometer. Crystal-
lographic data were measured with a Nonius Kappa CCD, and the
structure was solved by direct methods (SHELXS-97)6 and refined
by full-matrix least-squares methods on F2 (SHELXL-97).7
Computational Details. Calculated geometries were fully
optimized in the gas phase with tight convergence criteria at the
density functional theory (DFT) level with the Gaussian 03
package,8 using initially the B3LYP9 functional. Thereafter, the
resulting geometries were refined with the hybrid meta functional
mPW1B9510 that has been recommended for general purpose
applications and was developed to produce a better performance
where weak interactions are involved,10b such as those between
ligands and heavy metals.11 Comparison of calculated with
experimental (X-ray diffraction crystallography) geometries was
performed by computing the root-mean-square deviation (rmsd)
extended over all heavy (non-H) atoms excluding solvent dimeth-
ylformamide (DMF) molecules. Unless otherwise stated, the
6-311G** basis set was used for all atoms, adding diffuse functions
on donor atoms (F, O, N, and S; denoted as aug6-311G**) as well
as the SDD basis set, with effective core potential, for Hg, Cd, and
Pb. All reported data were obtained from these gas-phase optimized
geometries by means of single-point calculations. Energy values
were computed at the same level, considering solvent DMSO effects
by using the Cossi and Barone’s conductor-like polarizable
continuum model (CPCM) modification12 of the Tomasi’s PCM
formalism13 and correcting the basis set superposition error by
means of the counterpoise approach,14 except for metal complexes
because of convergence problems. All energies are uncorrected for
the zero-point vibrational energy. The reduced basis set 6-311G**
Generally, anion recognition motifs are often structurally
complicated and require an elaborate and sophisticated
synthetic process. Therefore, the development of chemosen-
sors capable of recognizing and sensing anions and cations
at the same time is one of the most challenging fields from
the viewpoint of organic and supramolecular chemistry. From
this perspective, we decided to combine in a highly preor-
ganized system the redox activity of the ferrocene moiety
with the strong hydrogen-bonding ability of the thiourea
groups. Despite its homoditopic nature, the unprecedented
capability for binding anions and metal cations is one
interesting characteristic of these new structural motifs.
Experimental Section
General Methods. All reactions were carried out under N2 and
using solvents which were dried by routine procedures. Column
chromatography was performed with the use of a column of
dimensions 60 × 4.5 cm and of silica gel (60 A C.C. 70-200 µm,
sds) as the stationary phase. NMR spectra were recorded at 200,
300, and 400 MHz. The following abbreviations for stating the
multiplicity of the signals in the NMR spectra were used: s (singlet),
bs (broad singlet), d (doublet), t (triplet), bt (broad triplet), st
(pseudotriplet), dt (double triplet), m (multiplet), and q (quaternary
carbon). The electrospray ionization (ESI) mass spectra were
recorded on an AGILENT V spectrometer. Elemental analyses were
carried out on a Carlo-Erba EA-1108 analyzer. Cyclic voltammo-
grams (CV) and Osteryoung square wave voltammograms (OSWV)
were performed with a conventional three-electrode configuration
consisting of platinum working and auxiliary electrodes and a
saturated calomel electrode (SCE) reference electrode. The experi-
ments were carried out with a 10-3 M solution of sample in dimethyl
sulfoxide (DMSO) containing 0.1 M (n-C4H9)4NPF6 (TBAHP) as
a supporting electrolyte. All of the potential values reported are
relative to the Fc+/Fc couple at room temperature. Deoxygenation
of the solutions was achieved by bubbling nitrogen for at least 10
min, and the working electrode was cleaned after each run. The
CVs were recorded with a scan rate increasing from 0.05 to 1.00
V s-1, while the OSWVs were recorded at a scan rate of 0.1 V s-1
with a pulse hight of 25 mV and a step time of 50 ms. Typically,
the receptor (1 × 10-3 mol) was dissolved in solvent (5 mL), and
TBAHP (base electrolyte; 0.193 g) was added. The guest under
investigation was then added as a 0.1 M solution in appropriate
solvent using a microsyringe, while the CV properties of the solution
were monitored. Ferrocene was used either as an external or internal
reference both for potential calibration and for reversibility criteria.
Under similar conditions, the ferrocene has E° ) 0.390 V vs SCE,
and the anodic peak-cathodic peak separation is 67 mV. Micro-
calorimetric titrations were carried out using an isothermal titration
calorimeter, and they were performed as follows: a solution of the
receptor in DMSO (c ) 1 × 10-3 M) was prepared, and it was
titrated with the appropriate alkylammonium salt at 25 °C. The
original heat pulses were normalized using reference titrations
(6) Sheldrick, G. M. SHELXS-97, program for crystal structure solutions;
University of Go¨ttingen: Go¨ttingen, 1997.
(7) Sheldrick, G. M. SHELXL-97, program for refinement of crystal
structures; University of Go¨ttingen: Go¨ttingen, 1997.
(8) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Montgomery J. A., Jr.; Vreven, T.; Kudin,
K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone,
V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.;
Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa,
J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene,
M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken, V.;
Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev,
O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala,
P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.;
Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas,
O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.;
Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.;
Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.;
Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.;
Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen,
W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian 03, Revision
B.03; Gaussian, Inc.: Wallingford, CT, 2004.
(9) Bartolottiand, L. J.; Fluchick, K. In ReViews in Computational
Chemistry; Lipkowitz, K. B., Boyd, B. D., Eds.; VCH: New York,
1996; Vol. 7, pp 187-216.
(10) (a) Zhao, Y.; Truhlar, D. G. J. Phys. Chem. A 2004, 108, 6908–6918.
(b) Zhao, Y.; Truhlar, D. G. J. Phys. Chem. A 2005, 109, 5656–5667.
(11) For instance, see Mu˜niz, J.; Sansores, L. E.; Mart´ınez, A.; Salcedo,
R. J. Mol. Struct. 2007, 820, 141–147.
(12) (a) Barone, V.; Cossi, M. J. Phys. Chem. A 1998, 102, 1995–2001.
(b) Cossi, M.; Rega, N.; Scalmani, G.; Barone, V. J. Comput. Chem.
2003, 24, 669–681.
(13) (a) Miertus, S.; Scrocco, E.; Tomasi, J. J. Chem. Phys. 1981, 55, 117–
129. (b) Cammi, R.; Mennucci, B.; Tomasi, J. J. Phys. Chem. A 2000,
104, 5631–5637.
(5) (a) Yang, X.-F.; Li, Y.; Bai, Q. Anal. Chim. Acta 2007, 584, 95–100.
(b) Lee, M. H.; Cho, B. K.; Yoon, J.; Kim, J. S. Org. Lett. 2007, 9,
4515–4518. (c) Wu, F.-Y.; Zhao, Y.-Q.; Ji, Z.-J. J. Fluoresc. 2007,
17, 460–465. (d) Wu, J.-S.; Hwang, I.-C.; Kim, K. S.; Kim, J. S. Org.
Lett. 2007, 9, 907–910.
(14) Boys, S. F.; Bernardi, F. Mol. Phys. 1970, 19, 553–566.
Inorganic Chemistry, Vol. 48, No. 4, 2009 1567

Oto´n et al.
Scheme 1
(without diffuse functions), together with SDD-ecp (for Hg, Cd,
and Pb), was used to perform the natural bond orbital (NBO)
population analysis. Bond orders were characterized by the Wiberg’s
bond index (WBI)15 and calculated with the NBO method as the
sum of squares of the off-diagonal density matrix elements between
atoms. The topological analysis of the electronic charge density
was conducted by means of the Bader’s atoms-in-molecules
(AIM)16 methodology using the AIM2000 software.17
Preparation of 1,1′-Bis(isothiocyanato)ferrocene 2. A well-
stirred solution of 1,1′-bis(N-triphenylfosforanilideneamino)fer-
rocene 1 (0.1 g, 0.18 mmol) in CS2 (8 mL) was heated at 45 °C
under nitrogen for 10 h. Then, the solvent was removed under
reduced pressure, and the remaining solid was slurried with diethyl
ether and filtered. The filtrate was concentrated to dryness and
recrystallized from hexane at -40 °C to give 2 (0.02 g, 25%) as a
yellow solid. Mp 88-92 °C. IR (Nujol): ν ) 2954, 2918, 2854,
2120, 2071, 1463, 1306, 1256, 1208, 1105, 1025, 817, 721 cm-1
.
1H NMR (200 MHz, CDCl3, at T ) 60 °C): δ ) 4.53 (st, 4H);
4.22 (st, 4H). 13C NMR (50 MHz, CDCl3, T ) 60 °C): δ ) 132.1
(q), 67.9 (CH), 67.1 (CH). ESI MS: m/z (%): 300 (M+, 43), 277
(48), 262 (59). Elem anal. Calcd (%) for C12H8FeN2S2: C, 48.01;
H, 2.69; N, 9.33. Found: C, 48.22; H, 2.51; N, 9.20.
General Procedure for the Preparation of 1,1′-Bis(thiou-
reido)[11]ferrocenophanes, 3 and 4. To a solution of 1,1′-
bis(isothiocyanato)ferrocene 2 (0.1 g, 0.33 mmol) in freshly distilled
dry THF (80 mL) was added dropwise a solution of the appropriate
amine (1.0 mmol), in the same solvent (80 mL), under nitrogen.
The solution was stirred for 3 h at room temperature, and the
resulting orange solid was filtered, washed with diethyl ether, and
recrystallized from CH2Cl2/diethyl ether (1:1) to give the appropriate
1,1′-disubstituted ferrocene derivative.
1H), 7.16-7.26 (m, 7H), 7.39 (d, 4H, 3J ) 7.8 Hz), 7.75 (bs, 2H).
13C NMR (50 MHz, CDCl3): δ ) 23.8 (CH3), 28.8 (CH), 47.2
(CH2), 123.0 (CH), 125.6 (CH), 127.6 (CH), 128.9 (CH), 130.4
(CH), 131.0 (q), 135.7 (q), 147.1 (q), 180.4 (q). ESI MS: m/z (%):
489 (M+ - 1, 100). Elem anal. Calcd (%) for C28H34N4S2: C, 68.53;
H, 6.98; N, 11.42. Found: C, 68.71; H, 7.26; N, 11.64.
X-ray Data for Compound 3. C26H32FeN4S2 × 2 DMF, M )
j
666.72, triclinic, space group P1 (No. 2), a ) 11.2153(3) Å, b )
12.7201(4) Å, c ) 13.2468(4) Å, R ) 109.900(2)°, ꢀ ) 98.020(2)°,
γ ) 99.179(2)°, V ) 1715.52(9) Å3, T ) 233(2) K, F(000) ) 708,
crystal size ) 0.20 × 0.15 × 0.02 mm3, Z ) 2, Mo KR (λ )
0.71073 Å), 9690 reflections collected, 5211 [R(int) ) 0.0329]
independent reflections, wR2 (all data) ) 0.1299, goodness-of-fit
is 1.056.
Crystallographic data for compound 3 reported in this paper have
been deposited with the Cambridge Crystallographic Data Centre,
CCDC 693271. Copies of this information may be obtained free
of charge from The Director, CCDC, 12 Union Road, Cambridge
CB21EZ,U.K.(Fax,+44-1223-336033;e-mail,deposit@ccdc.cam.ac.uk;
or on-line, http://www.ccdc.cam.ac.uk).
3. 0.103 g, 60%; mp 221-224 °C (dec). IR (Nujol): ν ) 3321,
3082, 3047, 1690, 1592, 1557, 1339, 1294, 1196, 1166, 1098, 1054,
1038, 1020, 976, 938, 874, 794 cm-1 1H NMR (300 MHz,
.
[D6]DMSO, at T ) 60 °C): δ ) 1.23 (d, 12H, J ) 6.3 Hz), 3.08
(m, 2H), 4.02 (bs, 4H), 4.73 (bs, 4H), 4.82 (bs, 4H), 6.99 (s, 1H),
7.14 (s, 1H), 7.71 (s, 2H), 8.95 (s, 2H). 13C NMR (75.3 MHz,
[D6]DMSO, T ) 60 °C): δ ) 22.9 (CH3), 27.4 (CH), 42.9 (CH2),
63.1 (CH), 63.9 (CH), 97.9 (q), 120.4 (CH), 121.8 (CH), 132.6
(q), 142.3 (q), 180.7 (q). ESI MS: m/z (%): 519 (M+ - 1, 100).
Elem anal. Calcd (%) for C26H32FeN4S2: C, 59.99; H, 6.20; N, 10.76.
Found: C, 60.23; H, 6.03; N, 10.51.
4. 0.096 g, 67%; mp 230-233 °C (dec). IR (Nujol): ν ) 3332,
3236, 3196, 1688, 1591, 1537, 1333, 1291, 1195, 1105, 1023, 970,
Results and Discussion
1
935, 874, 802 cm-1. H NMR (400 MHz, [D6]DMSO at T ) 80
Synthesis. The aza-Wittig reaction18 of 1,1′-bis(N-triph-
enylphosphoranylidenamino)ferrocene 13a with CS2 allowed
the isolation of 1,1′-bis(isothiocyanato)ferrocene 2 in 25%
yield after recrystallization (Scheme 1). The reaction of the
bisheterocumulene 2 with 1,3-bis(aminomethyl)-4,6-diiso-
propylbenzeneandm-xylylendiamineaffordedthebis(thioureido)-
[11]ferrocenophanes 3 and 4 in 60% and 67% yield,
respectively. Likewise, and only for comparison purposes,
the model compound 5 was prepared, in 91% yield, by
reaction of 1,3-bis(aminomethyl)-4,6-diisopropylbenzene with
phenylisothiocyanate. The 1H and 13C NMR spectra (which
were recorded at 60 and 80 °C, respectively, because of the
better resolution of peaks), elemental analyses, and mass
spectra were consistent with the proposed structures (see the
Experimental Section).
3
°C): δ ) 4.04 (bs, 4H), 4.68 (bs, 4H), 4.82 (d, 4H, J ) 5.6 Hz),
3
7.06 (d, 2H, J ) 7.2 Hz), 7.23-7.25 (m, 2H), 7.79 (s, 2H), 8.88
(s, 2H). 13C NMR (100.4 MHz, [D6]DMSO at T ) 80 °C): δ )
45.7 (CH2), 63.0 (CH), 63.7 (CH), 97.9 (q), 122.0 (CH), 123.9 (CH),
127.2 (CH), 139.2 (q), 180.9 (q). ESI MS: m/z (%): 435 (M+ - 1,
100). Elem anal. Calcd (%) for C20H20FeN4S2: C, 55.05; H, 4.62;
N, 12.84. Found: C, 55.30; H, 4.89; N, 12.71.
5. 0.147 g, 91%; mp 193-196 °C. IR (Nujol): ν ) 3394, 3153,
1595, 1529, 1315, 1298, 1252, 1186, 1100, 1068, 982, 911, 736
1
3
cm-1. H NMR (200 MHz, CDCl3): δ ) 1.16 (d, 12H, J ) 6.8
Hz), 3.05 (m, 2H), 4.83 (d, 4H, J ) 5.0 Hz), 6.08 (t, 2H), 7.02 (s,
(15) Wiberg, K. Tetrahedron 1968, 24, 1083–1096.
(16) Bader, R. F. W. Atoms in Molecules: A Quantum Theory; Oxford
University Press: Oxford, 1990.
(17) (a) AIM 2000 v. 2.0, designed by Biegler-Ko¨nig, F. W., and
Scho¨nbohm, J., 2002. Home page: http://www.aim2000.de/. Biegler-
Ko¨nig, F.; Scho¨nbohm, J.; Bayles, D. J. Comput. Chem. 2001, 22,
545–559. (b) Biegler-Ko¨nig, F.; Scho¨nbohm, J. J. Comput. Chem.
2002, 23, 1489–1494.
(18) (a) Molina, P.; Alajarin, M.; Arques, A. Synthesis 1982, 596–597. (b)
Arques, A.; Molina, P. Curr. Org. Chem. 2004, 8, 827–843.
1568 Inorganic Chemistry, Vol. 48, No. 4, 2009

Mononuclear Ferrocenophane Structural Motifs
Figure 2. Calculated (mPW1B95/aug6-311G**) structures for (a) 4C2 and
(b) 4qCs
.
4 · (DMF)2 by means of quantum chemical calculations at
the DFT level of theory, to check the performance of the
selected computational level and its applicability in the study
of its binding ability toward several anions and metal salts.
The quality of several basis sets was checked by comparison
of the calculated geometries of 4·(DMF)2 using two different
functionals, with that determined by single-crystal X-ray
diffraction for 3·(DMF)2 after replacing the iso-propyl groups
by H atoms. Even with the standard B3LYP functional, the
agreement with the experimental data (see the Computational
Details) is remarkable19 and systematically enhanced when
using the hybrid meta functional mPW1B95.10 Quantum
chemical calculations performed on 4 also revealed the
existence of a minimum energy conformation 4qCs (Figure
2a) with both thiourea groups pointing to the same side of
the molecule (quasi-Cs symmetry). Either in solution or in
the gas phase, this absolute minimum stationary point can
be interconverted into a relative minimum structure with C2-
symmetry, 4C2 (Figure 2b), by means of a transition state
representing an energetic barrier of 12.70 kcal· mol-1, low
enough to allow moderately fast interconversion at room
temperature, and featuring one of the thiourea groups almost
parallel to the phane ring and with the NH groups pointing
inward. The C2-symmetric minimum is less stable by only
1.68 kcal · mol-1 but results in the preferred geometry in the
crystalline state, probably because it enables a more compact
packing.
Figure 1. (a) Crystal structure of compound 3 (ORTEP molecular
representation) and (b) crystal packing of compound 3 along the a axis,
showing the intermolecular hydrogen bonds.
Additionally, single crystals of compound 3 were grown
by slow evaporation of a dichloromethane solution containing
5% DMF for solubility, used for its X-ray crystal determi-
nation. An ORTEP view of compound 3 is shown in Figure
j
1. Compound 3 crystallizes in the triclinic space group P1
with two molecules in the unit cell. The structure reveals
almost eclipsed cyclopentadienyl rings with a dihedral angle
of 7° between them because of the structure of the bis(thio-
ureido) ring which shows a trans configuration of the two
thiourea units. The 4,6-diisopropyl-m-xylylene unit is almost
perpendicular to the cyclopentadienyl rings. The solid-state
packing of compound 3 (Figure 1b) is best described as
centrosymmetrically related pairs of molecules. Stacking of
these related pairs occurs with a head-to-tail pairing. The
relative arrangement and the large distances between neigh-
boring molecules exclude the presence of hydrogen bonds
among neighboring thiourea units, which instead form
hydrogen bonds with the oxygen atom of the DMF solvent
present in the crystal structure. Under these circumstances,
the largest driving forces for such molecular packing are van
der Waals and short π· · · ·H-C and C-H· · · ·N interactions,
which lead to an efficient space filling.
Anion-Sensing Properties. At first, the electrochemical
properties of receptors 3 and 4 on their own as well as in
the presence of variable concentrations of F-, Cl-, Br-,
-
-
-
AcO-, NO3 , HSO4 , H2PO4 , and HP2O73- as guest anionic
(19) Calculated RMSD in 4 0.0844 Å at the B3LYP/aug6-311G** level
of theory and 0.0793 and 0.0763 Å when using the mPW1B95
functional and aug6-31G* or aug6-311G** basis sets, respectively.
These deviations reflect the fact that calculations do not take into
account the crystal packing.
We have also obtained the structure of the simplest
ferrocenophane of this series, 4, as well as the complex
Inorganic Chemistry, Vol. 48, No. 4, 2009 1569

Oto´n et al.
Table 1. Voltammetric Data of the Free Receptors 3 and 4 and Their
Anionic Complexes Obtained in DMSO at 278 K
c
receptora
E1/2 (V)
∆E1/2 (V)
Eab (V)
Kred/Kox
3
-0.18
-0.33
-0.30
-0.33
-0.27
-0.22
-0.39
-0.37
-0.42
-0.34
3·AcO-
3·H2PO4
3·HP2O7
3·F-
-0.15
-0.12
-0.15
-0.09
0.09
0.09
0.11
0.12
343
107
343
33
-
3-
4
4·AcO-
4·H2PO4
4·HP2O7
4·F-
-0.17
-0.15
-0.20
-0.12
0.06
0.05
0.04
0.04
748
343
2405
107
-
3-
a Anions added as their tetrabutylammonium salts. b Anodic peak
potential of the irreversible wave. c See refs 23 and 24.
species were investigated using CV and OSWV.20 Each free
receptor exhibited a reversible one-electron redox wave21
typical of a ferrocene derivative, at the half-wave potential
values shown in Table 1, calculated versus the ferrocenium/
ferrocene (Fc+/Fc) redox couple, which are identical to those
obtained from the corresponding OSWV peaks (see the
Supporting Information). The acyclic thiourea 5, where the
ferrocene moiety has been replaced by two phenyl groups,
has been prepared and studied to determine the electrochemi-
cal behavior of the thiourea units present in this kind of
receptors. However, the OSWV of the model compound 5
does not show any electrochemical process within the range
in which all voltammetric experiments have been carried out
(E ) -0.4 to 0.4 V vs Fc+/Fc).
Figure 3. OSWV (solid lines) and linear sweep voltammetry (dashed lines)
3-
of (a) receptor 3 and (b) complex 3·HP2O7
.
corresponding to anodic processes, have confirmed that the
irreversible oxidation involves a one-electron process, as
evaluated from the comparison of the heights of the Fe(II)/
Fe(III) and irreversible oxidation waves (see the Supporting
Information). Additionally, to elucidate the origin of the wave
that appears at higher potential, studies of reversibility of
the complexation process have been carried out. Addition
of aqueous HBF4 (5 or 6 equiv) recovered the original
voltammogram, thus demonstrating the reversibility of the
complexation.
In general, for receptor-anion systems exhibiting two-
wave behavior, both half-wave potentials can be obtained
from the voltammetric data as both redox couples are
simultaneously detected, and then the binding enhancement
factor (BEF)22 could be calculated (Table 1). This factor,
which defines the quantity Kox/Kred, is deduced from the
equation ∆E1/2 ) (RT/nF) ln(Kred/Kox).23 Because Kred and
Kox correspond to the equilibrium constants for the com-
plexation processes by the reduced and oxidized form of the
ligands, this factor indicates the times that the complexation
in the reduced form of the receptors is more difficult than
that in the oxidized one.
Titration studies, with the addition of several anions as
their tetrabutylammonium salts (TBA+) to an electrochemical
solution of receptors 3 and 4 (c ) 1 × 10-3 M) in the
appropriate solvent, containing 0.1 M [n-Bu4N]PF6 as
supporting electrolyte, demonstrate that while addition of F-,
-
3-
AcO-, H2PO4 , and HP2O7 anions to these receptors
promotes remarkable responses, indicating the interaction of
thiourea binding sites with these anions (see below), addition
of Cl-, Br-, HSO4 , and NO3- anionic species had no effect
-
on their CV and OSWV, even when present in a large excess
(up to 1/10, receptor/anion, ratio; see the Supporting
Information). Nevertheless, the results obtained in OSWV
on the stepwise addition of substoichiometric amounts of
the appropriate guest anionic species revealed the appearance
of a second wave that shifted cathodically from -0.20 V
3-
for 4 · HP2O7 to -0.09 V for 3 · F- (Table 1; Figure 3).
Surprisingly, at the end of the titration, a new irreversible
oxidation wave at more anodic potential (Ea ) 0.09-0.12
V for receptor 3 and 0.04-0.06 V for receptor 4) also
appeared in the voltammogram.
Linear sweep voltammetry experiments carried out in the
complexes, which display the presence of two waves
On the other hand, the acyclic thiourea 5, which, as it was
mentioned before, did not show any electrochemical response
upon addition of 5 equiv of the above-mentioned anions,
gives rise to an oxidation process, with the appearance of a
wave at Ep ) -0.23 V. These results seem to confirm the
fact that the waves exhibited by receptors 3 and 4 are related
to the presence of the thiourea bridges in their structures
(see the Supporting Information). We propose that this
irreversible oxidation wave is due to the following: prior to
the recognition process, the oxidation potential of the thiourea
(20) OSWV technique has been employed to obtain well-resolved potential
information, while the individual redox processes are poorly resolved
in the CV experiments in which individual E1/2 potentials cannot be
easily or accurately extracted from these data. (a) Serr, B. R.; Andersen,
K. A.; Elliot, C. M.; Anderson, O. P. Inorg. Chem. 1988, 27, 4499–
4504. (b) Richardson, D. E.; Taube, H. Inorg. Chem. 1981, 20, 1278–
1285.
(21) The criteria applied for reversibility were a separation of 60 mV
between cathodic and anodic peaks, a ratio for the intensities of the
cathodic and anodic currents Ic/Ia of 1.0, and no shift of the half-
wave potentials with varying scan rates.
(22) Beer, P. D.; Gale, P. A.; Chen, Z. AdV. Phys. Org. Chem. 1998, 31,
1–90.
(23) Richardson, D. E.; Taube, H. Inorg. Chem. 1981, 20, 1278–1285.
(24) Lo, K. K.-W.; Lau, J. S.-Y.; Fong, V. W.-Y. Organometallics 2004,
23, 1098–1106.
1570 Inorganic Chemistry, Vol. 48, No. 4, 2009

Mononuclear Ferrocenophane Structural Motifs
-
Figure 4. ITC titration plots of receptors 3 (left) and 4 (right) with H2PO4 (as TBA+ salt). Top: raw data. Bottom: normalized integration data of the
evolved heat per injection in terms of kcal mol-1 of injectant (H2PO4-) plotted against the molar ratio (H2PO4-/ligand).
Table 2. ITC Dataa of Anionic Complexes Derived from Receptors 3
and 4
fragment is higher than 1.3 V;24 however, after the addition
oftheanionandtheconsequentformationoftheanion-receptor
hydrogen bonding complex, the oxidation potential of the
thiourea group is reduced, making the oxidation more
feasible.25
receptora
anion
AcO-
Kas
L:Ab
∆Hc
T∆Sc
∆Gc
3
3
6.4 × 103
4.19 × 105
3.7 × 103
4.1 × 106
9.3 × 104
2.4 × 103
2.1 × 104
1.6 × 106
1.0 × 105
1:1
1:1
1:2
1:1
1:2
1:1
1:1
1:1
1:2
-1.27
-2.25
1.87
-4.34
-1.78
-2.28
-1.96
-4.85
-1.15
3.92
5.42
6.75
4.69
5.00
2.32
3.90
3.64
5.66
-5.19
-7.67
-4.88
-9.03
-6.78
-4.60
-5.86
-8.49
-6.81
-
H2PO4
3-
3
HP2O7
Isothermal titration calorimetry (ITC) provides useful
insight into the nature of the binding interactions. It
constitutes a method well suited for determining the affinity
of a molecular guest toward an artificial host compound.
Advantages of this method such as the capacity to elucidate
the enthalpic and entropic components of the free energy
for studying anion recognition properties have been recently
reported.26 ITC experiments were carried out by adding
aliquots of the appropriate anion (c ) 1.4 × 10-2 M) to a
solution of the receptors 3 or 4 (c ) 1 × 10-3 M) at 298 K
in DMSO (see the Supporting Information). Typical ITC
4
4
4
AcO-
H2PO4
HP2O7
-
3-
a The solvent used was DMSO. b Ligand/anion stoichiometry. c kcal/mol-1
.
of monocharged anions, and the resulting thermodynamic
parameters and complexation data are collected in Table 2.
Apart from F- ion, which eludes the analysis as no significant
-
heat effect was recorded in the titration, AcO-, H2PO4 , and
3-
HP2O7 anions all showed well behaved responses.
-
plots from the titration of 3 and 4 with the H2PO4 anion
Large values for the complexation enthalpy and entropy
-
were obtained from ITC measurements for AcO-, H2PO4 ,
are depicted in Figure 4. The integration of the head pulses
obtained in each titration step (Figure 4a,c) gives titration
curves (Figure 4b,d) rendering the molar enthalpy ∆H° as
the step height and the free energy ∆G° from the slope in
the inflection point. The association constants, Kas, and the
stoichiometry of complexation are derived as independent
parameters from the fitting of curves, and finally, the molar
∆S° may be calculated from the Gibbs-Helmholtz equation.
Considerably high affinities were found in the studied series
and HP2O73- anions, with both magnitudes being responsible
for the exoergonic character of these complexations. The
large positive values found for ∆S can be ascribed to the
increase of disorder produced during the complexation by
the replacement of the DMSO molecules in the solvation
sphere of such anions. If one assumes that the solvation
changes of anion complexation by 3 or 4 are similar because
the core binding sites are identical, the much increased
entropy contribution seen with receptor 3 would indicate a
massive loosening of the host structure brought about by
anion binding.
(25) Gunnlaugsson, T.; Fruger, P. E.; Lee, T. C.; Parkesh, R.; Pfeffer, F. M.;
Hussey, G. M. Tetrahedron Lett. 2003, 44, 6575–6578.
(26) (a) Schmidtchen, F. P. Org. Lett. 2002, 4, 431–434. (b) Tobey, S. L.;
Anslyn, E. V. J. Am. Chem. Soc. 2003, 125, 14807–14815. (c) Lee,
C.-H.; Na, D.-D.; Yoon, D.-W.; Won, D.-H.; Cho, W. S. V.; Lynch,
M.; Shevchuk, S. V.; Sessler, J. L. J. Am. Chem. Soc. 2003, 125, 7301–
7306. (d) Sambrook, M. R.; Beer, P. D.; Wisner, R. L.; Paul, R. L.;
Cowly, A. R.; Szemes, F.; Drew, M. G. B. J. Am. Chem. Soc. 2005,
127, 2292–2302. (e) Sessler, J. L.; An, D.; Cho, W.-S.; Lynch, V. M.;
Marquez, M. Chem. Eur. J. 2005, 11, 2001–2011.
The experimental data points indicate the formation of a
1:1 complex for AcO- anion and 1:2 for HP2O73- anion. In
-
the case of H2PO4 anion, a 1:1 binding model was found
for receptor 4, whereas in the titration with receptor 3 an
exothermic primary 1:1 association step is followed by an
Inorganic Chemistry, Vol. 48, No. 4, 2009 1571

Oto´n et al.
endothermic event of lower affinity (Figure 4b). This lower
affinity binding step shows the ordinary pattern of ion-pairing
in a polar solvent,27 that is, a highly positive entropy in
connection with a weak positive enthalpy (Table 2).
Free energy contributions in the anion complexation of
hosts 3 and 4 must have two sources. One must come from
the binding of the anions with the DMSO and another from
the binding of the host with the anions. Insights about these
two contributions can be obtained by inspecting the selectiv-
ity of the bisthiourea hosts 3 and 4 toward the anions. In
view of the basicity of the anions, the selectivity should be
-
in the order of AcO- (pKa AcOH ) 4.74) > H2PO4 (pKa
H3PO4 ) 2.12) > HP2O73- (pKa ) 0.85 H4P2O7). However,
3-
the ferrocenophane hosts bind HP2O7 more strongly than
-
H2PO4 and AcO-. This tendency has also been observed
for cyclic thioureas.28 Accordingly, the observed selectivity
3-
-
HP2O7 > H2PO4 > AcO- must be due to the difference
between the solvation of DMSO to the anions. It has been
well documented that polar aprotic solvents such as DMSO
are capable of electron-pair donation (Lewis basic) but are
not very effective on the Lewis acidity scale. Thus for AcO-,
-
H2PO4 , and HP2O73- anions, the DMSO must stabilize the
positively polarized atoms of the anions (i.e., carbon and
phosphorus, respectively), rather than the negatively charged
oxygen atoms. However, because the positively polarized
1
Figure 5. Partial H NMR spectrum in DMSO-d6 of (a) free receptor 3,
(b) 3 and 2 equiv of AcO-, (c) 3 and 2 equiv of H2PO4-, (d) 3 and 2 equiv
3-
of F-, and (e) 3 and 2 equiv of HP2O7
.
-
3-
phosphorus atoms of H2PO4 and HP2O7 anions are
surrounded by four oxygen atoms, they cannot be effectively
solvated by DMSO. On the contrary, the corresponding
carbon atom of the AcO- anion must be more accessible to
the solvation by DMSO, as revealed by previous thermo-
dynamic data (∆Gsolv of AcO- in DMSO ) -62.7 kcal/
mol).29 Accordingly, the comparatively lower binding con-
the formation of such bonds rigidified the receptor, which is
responsible for the shifts also observed in the methylene,
ferrocene, and H-2 aromatic proton signals upon complex-
ation. It should be highlighted that the presence of the
thiourea peaks in the spectra of complexes is clear evidence
that the ligands are not deprotonated by the anions and that
the chemical and spectroscopical changes are caused by the
coordination of the anions. All of these results confirm the
formation of strong complexes between ligands 3 and 4 and
the anions, showing also a good selectivity toward F-,
-
stant observed for AcO- anion than those for H2PO4 and
HP2O73- anions can be attributed to the stronger interaction
of the former with the solvent than that of the phosphorus-
based anions.
1H NMR titration experiments were also used to monitor
the anion recognition process. Thus, thiourea receptors 3 and
4 were found to bind F-, AcO-, H2PO4-, and HP2O73- anions
in DMSO solution (c ) 5 × 10-3 M in DMSO-d6) as
HP2O73-, H2PO4 , and AcO- anions.
To confirm the nature of these complexes, electrospray
-
mass spectrometry experiments were registered on the
3-
complexes formed upon the addition of F-, HP2O7
,
1
H2PO42-, and AcO-. The detection of the peak corresponding
to the complex was possible only in the case of the complex
with the largest association constant, HP2O73-. In all other
cases, the free receptor peak was found. The mass spectrum
evidenced by significant changes to the H NMR spectrum
of each receptor upon addition of such set of guest anions.
As expected, large downfield shifts in the resonance for
the two N-H thiourea protons were observed upon the
-
3-
addition of aliquots of F-, AcO-, H2PO4 , and HP2O7
3-
of 3 with HP2O7 exhibits a peak (m/z ) 938, 100%)
assigned to the complex [3 · H2P2O7 · (C4H9)4N]-, and the
mass spectrum of 4 with HP2O73- shows a peak (m/z ) 854,
44%) assigned to the complex [4 · H2P2O7 · (C4H9)4N]- ac-
companied with the peak of the free receptor (m/z ) 436,
100%; see the Supporting Information). The observation of
anions to the receptors 3 and 4 (see Table 2 and Figures
14-21 in the Supporting Information). This fact clearly
demonstrates that the thiourea moiety is involved in the
receptor-anion binding event (see Figure 5 and Supporting
Information), suggesting the formation of complexes in
which both thiourea arms are involved. As a consequence,
3-
both complexes with the HP2O7 anion is in accordance
with the largest ∆G changes observed for such an anion from
ITC measurements.
(27) (a) Rekharsky, M.; Inoue, Y.; Tobey, S.; Metzger, A.; Anslyn, E. J. Am.
Chem. Soc. 2002, 124, 14959–14967. (b) Jadhav, V. D.; Schmidtchen,
F. P. Org, Lett. 2005, 7, 3311–3314. (c) Valik, M.; Kral, V.; Hertweck,
E.; Schmidtchen, F. P. New J. Chem. 2007, 31, 703–710.
(28) (a) Bu¨hlmann, P.; Nishizaya, S.; Xiao, K. P.; Umezawa, Y. Tetrahe-
dron 1997, 53, 1647–1654. (b) Sasaki, S.; Mizuno, M.; Naemura, K.;
Tobe, Y. J. Org. Chem. 2000, 65, 275–283.
Binding of the aforementioned anions by the action of the
simplest receptor 4 has also been modelized by means of
DFT-based calculations. Under the working conditions, that
is, using naked anions (no countercation was explicitly
included) in the modelization, unless otherwise being stated,
(29) Pliego, J. R., Jr.; Rivero, J. M. Chem. Phys. Lett. 2002, 355, 543–
546.
1572 Inorganic Chemistry, Vol. 48, No. 4, 2009

Mononuclear Ferrocenophane Structural Motifs
in comparison to hydrogen bonds in model small mol-
ecules.30 Also, the relatively large positive 32F(rc) values
found are usually a good diagnostic for the ionic character
of the bonds.31
In all four receptor-anion complexes, we have also found
three or four additional hydrogen bonds located within the
receptor part that connect the thiocarbonyl groups with the
H atoms belonging to either the Cp rings or the CH2 groups
adjacent to the m-phenylene moiety (Figure 6). As also
outlined for the experimental X-ray structure of 3, the
calculated uncomplexed receptor 4qCs already shows these
intramolecular hydrogen bonds,32 thus explaining its high
preorganization and therefore the low energetic cost to be
paid for its adaptation upon complexation (ligand strain) with
most of the anionic guests.
Cation-Sensing Properties. The presence of sulfur atoms
in the molecule drove us also to study the cation-binding
ability of these receptors. Therefore, the metal recognition
properties toward Li+, Na+, K+, Mg2+, Ca2+, Ni2+, Cd2+,
Zn2+, Pb2+, and Hg2+ metal ions were evaluated by electro-
chemical analysis. In DMSO, no perturbation of the OSWV
of 3 and 4 was observed upon addition of Li+, Na+, K+,
Mg2+, Ca2+, Ni2+, Cd2+, Zn2+, and Pb2+ metal ions. However,
on addition of Hg2+ ions to receptors 3 and 4, the original
wave decreases in intensity while a new wave appears at a
more anodic potential. In the case of receptor 3, the new
wave appears at E1/2 ) 0.002 V (∆E1/2 ) 184 mV; BEF )
1290) with the addition of 1.5 equiv of Hg2+, whereas for
receptor 4, it appears at E1/2 ) 0.038 V (∆E1/2 ) 272 mV;
BEF ) 39650) when 2 equiv of this cation were added
(Figure 7).
The interference in the selective response of ligands 3 and
4 in the presence of Hg2+ metal ions from the other metal
cations tested was also studied by using cross-selectivity
experiments. Thus, addition of 1 equiv of Li+, Na+, K+, Ca2+,
Mg2+, Cd2+, Ni2+, Zn2+, or Pb2+ cations in CH3CN to 1 equiv
of the receptors 3 or 4 in DMSO did not give any optical
response. However, further addition of 1 equiv of Hg2+ to
the above-mentioned solutions gave an identical optical
response to that observed upon addition of 1 equiv of Hg2+
to a solution containing only 1 equiv of the appropriate ligand
and which is free of the other metal cations (see the
Supporting Information). These results clearly demonstrate
Figure 6. Calculated (mPW1B95/aug6-311G**) structures for (a) [4·F]-,
(b) [4·OAc]-, (c) [4·H2PO4]-, and (d) [4·HP2O7.TMA2]- with TMA+
cations omitted for clarity. The ligand is displayed in capped sticks while
the guest is highlighted in a ball-and-stick representation.
optimizations were carried out in the gas phase, observing
in all cases that 1:1 ligand/anion complexes derived from
the most stable receptor 4qCs (Figure 6) resulted to be more
favored than those derived from the other isomer 4C2 (see
the Supporting Information).
It is worth mentioning that in no case ligand deprotonation
was observed in agreement with experimental (1H NMR)
evidence. Instead, the four thiourea NH groups of the ligand
are oriented in a convergent fashion toward the hydrogen-
acceptor atoms of the anionic guests that result partially
wrapped. Strong interaction with the anionic guests is
evidenced by the high value of the total WBI15 extended
over all ligand-anion contacts (Table 3), which favorably
compares to those calculated as a reference for the dimers
of hydrogen fluoride (0.024), water (0.025), or ammonia
(0.017) at the same level. We have also used the Bader’s
AIM methodology16 to perform a topological analysis of the
electronic charge density, F(rc), to search for significant bond
critical points (BCPs) around the spatial region where the
noncovalent host-guest interactions are taking place. Indeed,
the sets of four hydrogen bonds so far described for every
ligand-anion complex can be appropriately characterized
by their respective BCPs (Table 3) whose electron density,
F(rc), agrees with the proposed relatively high bond strength,
(30) Calculated hydrogen bonds X-H· · ·X in dimers of model small
molecules are as follows: for (HF)2, dH· · ·F ) 1.850 Å, WBI 0.024,
3
5
F(rc) ) 2.39 × 10-2 e/a0 , and 32F(rc) ) 9.98 × 10-2 e/a0 ; for (H2O)2,
dH· · ·O ) 1.943 Å, WBI 0.025, F(rc) ) 2.48 × 10-2 e/a0 , and 32F(rc)
3
) 8.86 × 10-2 e/a0 ; for (NH3)2, dH· · ·N ) 2.252 Å, WBI 0.017, F(rc)
5
3
5
) 11.65 × 10-2 e/a0 , and 32F(rc) ) 14.92 × 10-2 e/a0 . For
comparison, the calculated covalent bonds X-H· · ·X in dimers of
the same model small molecules: for (HF)2, dF-H ) 0.938 Å, WBI
3
5
0.673, F(rc) ) 0.350 e/a0 , and 32F(rc) ) -2.567 e/a0 ; for (H2O)2,
dO-H ) 0.979 Å, WBI 0.750, F(rc) ) 0.347 e/a0 , and 32F(rc) )
3
5
-2.318 e/a0 ; for (NH3)2, dN-H ) 1.028 Å, WBI 0.838, F(rc) ) 0.323
e/a0 , and 32F(rc) ) -1.425 e/a0 .
3
5
(31) Nakanishi, W.; Nakamoto, T.; Hayashi, S.; Sasamori, T.; Tokitoh, N.
Chem. Eur. J. 2007, 13, 255–268, and references cited therein.
(32) For 4qCs dCS· · ·HCp ) 2.909 Å, WBI ) 0.002, F(rc) ) 0.55 × 10-2
3
5
e/a0 , 32F(rc) ) 1.61 × 10-2 e/a0 ; dCS· · ·HCp ) 2.785 Å, WBI ) 0.003,
F(rc) ) 1.15 × 10-2 e/a0 , 32F(rc) ) 3.57 × 10-2 e/a0 ; dCS· · ·HCAr
)
3
5
2.560 Å, WBI ) 0.019, F(rc) ) 1.78 × 10-2 e/a0 , 32F(rc) ) 5.18 ×
3
10-2 e/a0 ; and dCS· · ·HCAr ) 2.530 Å, WBI ) 0.023, F(rc) ) 1.85 ×
5
10-2 e/a0 , 32F(rc) ) 5.30 × 10-2 e/a0 .
3
5
Inorganic Chemistry, Vol. 48, No. 4, 2009 1573

Oto´n et al.
Table 3. Relevant Calculated Parameters Related to the Complexing Behavior of 4 towards Anions (A) and Metal Salts (MAn)
a
a
[4qCs ·F]-
[4qCs ·OAc]-
[4qCs ·H2PO4]-
[4qCs ·HP2O7]3-
[4C2 ·(HP2O7)2]6-
4qCs ·Hg(OTf)2
b
dNH· · ·A
1.604
1.724
1.918
1.770
1.896
1.934
1.840
1.875
1.897
1.801
1.823
2.007
1.860
1.897
1.986
1.822e
2.069
2.057
2.015
1.945
2.151
2.005
∑ WBIL· · ·A
0.311
0.213
0.188
0.216
0.266
0.170
d
F(rc[NH· · ·A])c (32F(rc[NH· · ·A]
)
5.43 (16.36)
4.09 (13.29)
2.65 (8.98)
2.03 (6.71)
14.20
4.14 (11.71)
2.90 (9.88)
2.61 (9.33)
2.21 (7.89)
11.86
3.36 (11.14)
3.08 (10.58)
2.94 (10.24)
2.54 (9.16)
12.44
3.64 (11.38)
3.54 (10.84)
2.39 (7.48)
1.81 (5.57)
12.46
3.16 (10.52)
2.96 (10.02)
2.39 (8.26)
2.24 (8.17)
12.49
4.04 (8.89)e
2.05 (7.11)
14.90
c
∑ F(rc[L· · ·A]
)
d
∑ F2F(rc[L· · ·A]
)
45.34
38.81
42.73
38.87
41.12
42.68
b
dCS· · ·M (WBICs· · ·M
)
2.470 (0.451)
2.471 (0.448)
7.44 (13.10)
7.44 (13.20)
F(rc[Cs· · ·M])c (32F(rc[Cs· · ·M]))d
a
c
d
Every HP2O7 anion bears two TMA cations. b In Å. In e/a0 × 102. In e/a0 × 102. e Formal (covalent) N · · ·H bond resulting from deprotonation
3-
3
5
3
5
by the pyrophosphate anion (dO-H ) 1.012 Å, WBI ) 0.621, F(rc) ) 0.3056, e/a0 ; 32F(rc) ) -1.9136 e/a0 ).
Figure 8. (a) Variation of the UV-vis spectrum of 3 (c ) 1 × 10-4 M)
increasing amounts of Hg2+ (as triflate salt).
Figure 7. OSWV of 3 in DMSO/[(n-Bu)4N]PF6 (0.1 M) before (red) and
in DMSO and (b) UV-vis titration curve, obtained upon addition of
after (blue) the addition of Hg2+ (as triflate salt).
analyzed quantitatively.35 No changes were observed in the
UV-vis spectra upon addition of Li+, Na+, K+, Mg2+, Ca2+,
Ni2+, Cd2+, Zn2+, or Pb2+ even when they were added in a
large excess to the DMSO solutions of any of these ligands.
However, significant spectrophotometric changes in the
metal-ligand absorption bands were observed on addition
of increasing amounts of Hg2+ metal ions.
For both ligands 3 and 4, the most prominent features
observed during the complexation processes are the progres-
sive increase of the corresponding metal-ligand transition
band and the appearance of well-defined isosbestic points
(Figure 8a), indicative of the presence, in the solution, of
only two absorbing species in equilibrium: the free ligand
(L) and the complex (L·Hg2+). Thus, addition of increasing
amounts of a solution of Hg2+ in CH3CN (c ) 2.5 × 10-2
M) to a solution of 3 in DMSO (c ) 1 × 10-4 M) caused a
progressive increase of the band located at λ ) 439 nm (ε
) 490 M-1 cm-1) accompanied with a small bathochromic
that these receptors have an excellent affinity for Hg2+ over
those ions. The demand for sensing Hg2+ metal ions in
competitive media such as DMSO is growing rapidly.33
Previous studies on ferrocene-based ligands have shown
that bands ascribed to the lowest energy metal-ligand
transitions (LE bands) in the absorption spectra are perturbed
upon complexation.34 Therefore, the metal recognition
properties of the thiourea-metallocene dyads 3 and 4 were
evaluated, additionally, by using UV-vis spectrophotometry
with the set of cations previously mentioned. Titration
experiments for DMSO solutions of these ligands (c ) 1 ×
10-4 M) and the corresponding ions were performed and
(33) Among other things, see: (a) Shunmugan, R.; Gabriel, G. J.; Smith,
C. E.; Aamer, K. A.; Tew, G. N. Chem. Eur. J. 2008, 14, 3904–3907.
(b) Diez-Gil, C.; Caballero, A.; Ratera, I.; Ta´rraga, A.; Molina, P.;
Veciana, J. Sensors 2007, 3481–3488. (c) Diez-Gil, C.; Martinez, R.;
Ratera, I.; Ta´rraga, A.; Molina, P.; Veciana, J. J. Mater. Chem. 2008,
18, 1997–2002.
(34) (a) Marder, S. R.; Perry, J. W.; Tiemann, B. G. Organometallics 1991,
10, 1896–1901. (b) Coe, B. J.; Jones, C. J.; McCleverty, J. A.; Bloor,
D.; Cross, C. J. J. Organomet. Chem. 1994, 464, 225–232. (c) Mu¨ller,
T. J.; Netz, A.; Ansorge, M. Organometallics 1999, 18, 5066–5074.
(d) Carr, J. D.; Coles, S. J.; Hassan, M. B.; Hurthouse, M. B.; Malik,
K. M. A.; Tucker, J. H. R. J. Chem. Soc., Dalton Trans. 1999, 57.
(35) Specfit/32 Global Analysis System, 1999-2004 Spectrum Software
Associates (SpecSoft@compuserve.com). The Specfit program was
acquired from Bio-logic, SA (www.bio-logic.info) in January 2005.
The equation to be adjusted by non-linear regression, using the above
mentioned software, was ∆A/b ) {K11∆εHG[H]tot[G]}/{1 + K11[G]},
where H ) host, G ) guest, HG ) complex, ∆A ) variation in the
absorption, b ) cell width, K11 ) association constant for a 1:1 model,
and ∆εHG ) variation of molar absorptivity.
1574 Inorganic Chemistry, Vol. 48, No. 4, 2009

Mononuclear Ferrocenophane Structural Motifs
Figure 10. (a) UV-vis spectra recorded during the reversibility experiments
carried out in CH2Cl2 for 3 with Hg(CF3SO3)2 and (b) absorbance registered
after each cycle.
The 1:1 stoichiometry of the complex formed between
ligand 4 and Hg2+ cations was also confirmed by electrospray
ionization mass spectrometry, in which three main peaks
were observed: a strong peak because of the fragment
[(42-2H) · Hg]+ (44%) together with two peaks of higher
intensity because of [4·Hg]+ (100%) and [(4-1H)]+ (98%),
respectively.
Figure 9. ITC titration plots of receptor 3 with Hg2+ (as triflate salt). Top:
raw data. Bottom: normalized integration data of the evolved heat per
injection in terms of kcal mol-1 of injectant (Hg2+) plotted against the molar
ratio (Hg2+/ligand). To determine the values of the thermodynamic variables
(∆H, ∆G, and ∆S), the ITC data have been fitted to a 1:1 binding model.
For the reported constants to be taken with confidence,
we have proved the reversibility of the complexation process
by performing the following experimental test: 1 equiv of
Hg(CF3SO3)2 was added to a solution of the ligand 3 in
CH2Cl2 to obtain the complexed 3 · Hg2+ species whose
OSWV and UV-vis spectrum were recorded. The CH2Cl2
solutions of the complexes were washed several times with
water until the color of the solution changed to that of the
receptor. The organic layers were dried, the corresponding
shift (∆λ ) 6 nm and ∆ε ) 1010 M-1 cm-1). Two well-
defined isosbestic points were found (λ ) 292 and 313 nm),
indicating that a neat interconversion between the uncom-
plexed and complexed species occurs. The resulting titration
profile suggests a 1:1 (L · Hg2+) binding model (Figure 8b).
Similar results were obtained for ligand 4, the detection limits
being 2.3 × 10-5 M for both ligands.36
1
optical spectrum, OSWV, and H NMR spectrum were
recorded, and they were found to be the same as those of
the corresponding free receptor 3. Afterward, 1 equiv of
Hg(CF3SO3)2 was added to this solution, and the initial
UV-vis spectrum and OSWV of the complex 3 · Hg2+ were
fully recovered. These experiments were carried out over
several cycles, and the optical spectra were recorded after
each step and found to be fully recovered on completion of
the step, thus demonstrating the high degree of reversibility
of the complexation/decomplexation processes (Figure 10).
The lower solubility of 4 in CH2Cl2 prevented us from
carrying out this experiment on this receptor.
The molecular structure of the complex which is formed
upon addition of mercury(II) triflate to receptor 3 or 4 has
been established with the aid of quantum chemical calcula-
tions. As described previously, we have used compound 4
as the only receptor model and assumed the 1:1 ligand/metal
stoichiometry claimed as the most stable. The quasi-Cs
stereoisomeric form of the receptor seems to be well suited
for complexing oxoanionic salts of the soft highly thiophilic
Hg2+ cation because of its two Janus-like faces. Thus the
host simultaneously features an NH-rich surface for anion
binding as previously described, furthermore contributing to
increase the rigidity of the already preorganized ligand,
Furthermore, ITC studies have been carried out to calculate
the association constants and the thermodynamic data related
to the complexation process. The experiments were achieved
by adding aliquots of the appropriate cation (c ) 7 × 10-3
M) to a solution of the receptors 3 or 4 (c ) 1 × 10-3 M)
at 298 K in DMSO. The association constants (Kas) were
calculated, as previously described, by nonlinear least-squares
analysis of the data obtained from these ITC experiments,
which also show that the recognition processes take place
with a 1:1 stoichiometries (L · Hg2+) and are also in agree-
ment with the data previously obtained by the UV-vis
titration experiments. The magnitudes of the corresponding
association constants are 5.2 × 104 M-1 and 2.0 × 104 M-1
for ligands 3 and 4, respectively. The thermodynamic data
obtained show that for compound 3, the complexation
process is mainly enthalpy driven (∆H ) -5.45 kcal/mol,
T∆S ) 0.91 kcal/mol, and ∆G ) -6.36 kcal/mol; Figure
9), while in the case of compound 4, the driving forces are
both enthalpy and entropy (∆H ) -1.96 kcal/mol, T∆S )
3.65 kcal/mol, and ∆G ) -5.61 kcal/mol).
(36) Shortreed, M.; Kopelman, R.; Kuhn, M.; Hoyland, B. Anal. Chem.
1996, 68, 1414–1418.
Inorganic Chemistry, Vol. 48, No. 4, 2009 1575

Oto´n et al.
upon addition of the appropriate amounts of Hg2+ cation to
the complexed species 3 · anion.
Conclusion
We have designed the first members of a potential class
of easy-to-synthesize neutral ferrocenophane-based thiourea
receptors and examined their binding properties toward
various guest ions using electrochemical, spectral, and optical
techniques, as well as by DFT-based quantum chemical
calculations. Our synthetic methodology, which allows the
preparation of ferrocenophane-based cyclic architectures,
with the ferrocene unit directly linked to two thiourea groups,
is based on the reaction of the 1,1′-bis(isothiocyanato)fer-
rocene 2 with diamino compounds. The rigidity of this new
structural motif imposes two different binding sites (het-
eroionic-sensing Janus faces38): a sulfur-rich face as an ideal
candidate to sense selectively the thiophilic Hg(II) metal
cations and a face with four directional N-H bonds able to
Figure 11. Calculated (mPW1B95/aug6-311G**/SDDecp) structure for
complex 4·Hg2+. The ligand is represented in capped sticks while the guest
is highlighted in a ball-and-stick representation. Counteranions are omitted
for clarity.
whereas two thiocarbonyl groups on the other side lie almost
parallel to each other and close enough (dS· · · S ) 4.559 Å in
the free receptor 4qCs) to easily allow chelation of a Hg2+
cation. The calculated structure for [4 · Hg2+] (Figure 11)
shows an almost planar-symmetric central receptor grabbing
the Hg2+ ion with the tweezer-like thiocarbonyl groups. It
is worth mentioning that the Hg2+ cation is essentially
dicoordinated, as suggested by the almost linear S-Hg-S
angle (155.3°), and does not promote oxidation of the
ferrocene moiety as in related systems.37 The interaction of
the NH groups of the receptor with anions seems to increase
the electron density at the central 1,3-phenylene C2 atom
and consequently could enable it to behave as donor atom
toward the Hg2+ cation.
-
recognize anions. Selective binding to F-, AcO-, H2PO4 ,
3-
and HP2O7 anions, in a polar solvent (DMSO) where
hydrogen-bonding interaction between the thiourea functional
groups and the anions is usually weakened by competing
solvent molecules, was observed for these ferrocene-based
thioureas. Remarkably, these receptors exhibit not only a high
selectivity for Hg(II) cations over other cations with a strong
anodic shift of the ferrocene oxidation wave (∆E1/2 ) 272
mV) but also a significant change in the absorption spectrum.
These new structural motifs pave the way to the design of a
new generation of homoditopic receptors in which the
ferrocene/ferrocenium redox couple and the ability of the
thiourea moiety to act as binding site may operate coopera-
tively within the molecule. This synergistic relation may
create a rare class of molecular systems which can behave
as redox-switching homotopic receptors, with the dual
capability of selectively sensing anions and cations.
The observed selectivity of receptor 4 toward Hg2+ in
comparison to other metal cations has also been studied
theoretically. With this aim, the structures of complexes
[4qCs · M2+] for M2+ ) Cd2+ and Pb2+ were also obtained
(see the Supporting Information), their formation being 8.6
and 21.7 kcal · mol-1 less stable, respectively, than that of
Hg2+ at the working level of theory.
To study the mutual interference of anions and Hg2+ in
the recognition ability of studied receptors, we also made
competitive electrochemical experiments. Thus, the CV
obtained upon addition of HP2O73-, H2PO42-, and AcO-
anions to an electrochemical solution of the 3 · Hg2+ species
showed that the first two molar equivalents of these anions
have no affinity for the thiourea residues present in the
complex 3 · Hg2+. However, addition of three equivalents
promotes the appearance of the redox wave ascribed to the
free receptor 3. Moreover, further additions of these anions
(up to five equivalents) to such an electrochemical solution
induced a cathodic shift of the redox wave, the magnitude
of such a redox-potential being the same as that observed
when those anions were added to the free receptor 3. Thus,
these results indicate that the added anions are apparently
able to stoichiometrically bind the Hg2+ cation, avoiding their
binding by the thiourea groups present in 3 · Hg2+ host. It is
also worth mentioning that a similar behavior was observed
Acknowledgment. We gratefully acknowledge the finan-
cial support from Fundacio´n Se´neca (Agencia de Ciencia y
Tecnolog´ıa de la Regio´n de Murcia) Projects 02970/PI/05
and 04509/GERM/06 (Programa de Ayudas a Grupos de
Excelencia de la Regio´n de Murcia, Plan Regional de Ciencia
y Tecnolog´ıa 2007/2010), the DGI Spain Project CTQ2006-
06333/BQU, and the Instituto Carlos III, MIyC, through
“Acciones CIBER” Ciber BBN. This paper is dedicated to
Prof. Josep Font Cierco, on the occasion of his retirement.
Supporting Information Available: Figures, tables of crystal
data and structure refinement, and tables of the theoretical calcula-
tions. This material is available free of charge via the Internet at
http://pubs.acs.org.
IC801879X
(37) Caballero, A.; Espinosa, A.; Ta´rraga, A.; Molina, P. J. Org. Chem.
2008, 73, 5489–5497.
(38) From the mythological roman god Janus, who is represented with a
double-faced head, each looking in opposite directions.
1576 Inorganic Chemistry, Vol. 48, No. 4, 2009