Convergent synthesis of meso-ferrocenyl porphyrins for TiO2 sensitization

Article abstract of DOI:10.1016/j.tetlet.2012.06.093

Electron-donating ferrocenyl moieties have been incorporated into supramolecular carboxyporphyrin architectures Zn(II)-5-ferrocenyl-10,20- bistolyl-15-(4-methylbenzoate)porphyrin (trans-Fc-ZnP-CO₂Me), Zn(II)-5-ferrocenyl-10,15,20-tris(4-methylbenzoate)porphyrin [trans-Fc-ZnP-(CO₂Me)₃], and Zn(II)-5,15-bisferrocenyl-10, 20-bis(4-benzoate)porphyrin [trans-Fc₂-ZnP-(CO₂Me) ₂] for self-assembly on metal oxide nanoparticles. Efficient and economic synthesis has required a convergent strategy toward reduced symmetry trans-A₂B₂ and trans-AB₂C substitution patterns about the porphyrin macrocycle minimizing the production of porphyrin side products and increasing yields of the target ferrocenylporphyrins. Preliminary spectroscopic data in solution and in the solid state bound to mesoporous TiO₂ films are discussed.

Full text of DOI:10.1016/j.tetlet.2012.06.093

Tetrahedron Letters 53 (2012) 4700–4703
Contents lists available at SciVerse ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Convergent synthesis of meso-ferrocenyl porphyrins for TiO2 sensitization
⇑
Joseph P. Harney, Timothy Dransfield, Jonathan Rochford
Center for Green Chemistry, Department of Chemistry, University of Massachusetts, Boston, 100 Morrissey Boulevard, Boston, MA 02125, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
Electron-donating ferrocenyl moieties have been incorporated into supramolecular carboxyporphyrin
architectures Zn(II)-5-ferrocenyl-10,20-bistolyl-15-(4-methylbenzoate)porphyrin (trans-Fc-ZnP-CO2Me),
Zn(II)-5-ferrocenyl-10,15,20-tris(4-methylbenzoate)porphyrin [trans-Fc-ZnP-(CO2Me)3], and Zn(II)-5,15-
bisferrocenyl-10,20-bis(4-benzoate)porphyrin [trans-Fc2-ZnP-(CO2Me)2] for self-assembly on metal
oxide nanoparticles. Efficient and economic synthesis has required a convergent strategy toward reduced
symmetry trans-A2B2 and trans-AB2C substitution patterns about the porphyrin macrocycle minimizing
the production of porphyrin side products and increasing yields of the target ferrocenylporphyrins. Pre-
liminary spectroscopic data in solution and in the solid state bound to mesoporous TiO2 films are
discussed.
Received 11 June 2012
Accepted 19 June 2012
Available online 26 June 2012
Keywords:
Porphyrin
Ferrocene
Dipyrromethane
TiO2
Ó 2012 Elsevier Ltd. All rights reserved.
Introduction
Results and discussion
Among the many potential clean, renewable, alternative energy
technologies, the dye-sensitized solar cell (DSSC) represents a
highly promising candidate.1–3 The porphyrin macrocycle repre-
sents a promising class of organic dyes for DSSCs with record
power conversion efficiencies of 12% recently reported when elec-
tron-donating diphenylamine substituents have been introduced
to the meso-position of the ring.4 In fact, along with their large vis-
ible molar extinction coefficients, it is the ability to tune the elec-
tronic properties of the porphyrin ring through appropriate
substitution at the meso-positions which make them particularly
attractive for semiconductor photosensitization in DSSCs.5 Efficient
ferrocene-to-porphyrin photoinduced electron transfer has been
demonstrated previously for the Zn(II)-5-ferrocenyl-10,15,20-riph-
enylporphyrin complex and inspired by the latter report we are
currently investigating the impact of such photoinduced charge
separations on the photocurrent response in DSSC devices.6 How-
ever, this has required a synthetic approach adapted to incorporate
peripheral carboxylic acid groups on the porphyrin ring, in addi-
tion to the ferrocenyl substituent, designed specifically for covalent
attachment to metal oxide electrodes. A convergent synthesis of
mono- and bisferrocenyl porphyrin dyads is here presented where
the electron donating ferrocene substituent is incorporated at the
meso-position of the porphyrin macrocycle in addition to benzoic
acid substituents at, minimum, one of the remaining meso-posi-
tions (Fig. 1). Preliminary UV–Vis studies of each dye system are
presented in solution and adsorbed on mesoporous TiO2 nanopar-
ticle thin films.
Initial attempts to synthesize a trans-AB2C ferrocenylporphyrin
similar to trans-Fc-ZnP-CO2Me were made via a Lewis acid cata-
lyzed mixed aldehyde condensation of 5-mesityldipyrromethane,
methyl-4-formylbenzoate, and ferrocenecarboxaldehyde using
standard Lindsey conditions.7 This synthetic method was initially
performed due to the reported benefits of using 5-mesityldipyr-
romethane which is well known to significantly reduce the forma-
tion of other porphyrin side products caused by ‘scrambling’.8 To
our surprise we observed a total of six porphyrin products from
this synthesis as opposed to the anticipated three.
It is likely that the scrambling here observed is due to electronic
effects of the meso-ferrocenyl substituent which counteract the
benefit of the rigid mesityl substituent. Ring opening of the por-
phyrinogen intermediate (prior to aromatization via p-choranil
oxidation) and rearrangement of substituents at the 5,10,15,20-
positions, that is scrambling, are driven by stabilization of a posi-
tive charge induced by ring protonation—this positive charge likely
being stabilized by the electron rich ferrocene group. Attempts to
purify the resulting porphyrin mixture, although relatively suc-
cessful, were not deemed worthwhile in the long run due to the
very poor yield of desired product and difficulty of separation.
For this reason a stepwise scheme was devised adapted from re-
ported methods to yield a single product (Scheme 1).9
Ferrocene carboxaldehyde 1 was condensed with pyrrole in the
presence of the Lewis acid catalyst indium chloride by adaption of
a literature procedure producing the known 5-ferrocenyldipyrrom-
ethane 2 in 67% yield.10 Acylation was then carried out at reduced
temperature to introduce the p-carboxytolyl group at the 1 and 9
positions of the dipyrromethane producing 3 in 72% yield. It has
been reported that purification of 1,9-diacyldipyrromethanes is
⇑
Corresponding author.
E-mail address: jonathan.rochford@umb.edu (J. Rochford).
0040-4039/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.tetlet.2012.06.093

J. P. Harney et al. / Tetrahedron Letters 53 (2012) 4700–4703
4701
CO2H
Fe
N
N
N
N
HO2C
Zn
MeO2C
Zn
CO2H
N
N
N
N
HO2C
ZnTCPP
trans-Fc-ZnP-CO2Me
CO2Me
CO2Me
Fe
N
N
Fe
Fe
MeO2C
Zn
N
N
Zn
N
N
N
N
MeO2C
MeO2C
Fc-ZnP-(CO2Me)3
trans-Fc2-ZnP-(CO2Me)2
Figure 1. Structures of meso-ferrocenylporphyrin complexes here investigated alongside the reference compound ZnTCPP.
O
O
O
nBu
nBu
HN
HN
(i)
N
N
HN
HN
(iii)
(ii)
O
H
Fe
Fe
Fe
Fe
Sn
O
67%
< 10 %
72%
4
1
3
2
82%
(iv)
HO
H
HN
HN
NH
NH
(i)
54 %
(v)
Fe
Fe
N
N
MeO2C
Zn
MeO2C
MeO2C
+
HO
N
N
17%
O
5
6
7
8
Scheme 1. Reagents and conditions: (i) pyrrole, InCl3, rt (ii) 2,6-Me2PhMgBr, p-toluoyl chloride, 0 °C (iii) NEt3, (n-Bu)2SnCl2 (iv) NaBH4, THF, MeOH, 0 °C. (v) CH2Cl2, Sc(OTf)3,
2,6-ditertbutylpyridine, p-chloranil; Zn(OAc)2, CH2Cl2, MeOH (vi) THF, NaOH.
made easier through the complexation of the nitrogen and oxygen
atoms with a di-n-butyltin adduct in a tetrachelate fashion.9 Com-
plexation also improves the shelf life of the 1,9-diacyldipyrrome-
thane intermediates. Preparation of the di-n-butyltin adduct of 3
produced complex 4 in low yield (<10%) ascribed to the electron
donating properties of the ferrocenyl substituent which presum-
ably decreases the acidity of the N–H protons. Tin complexation
was therefore overlooked with the benefit of precluding a synthetic
step. Compound 3 was successfully isolated in very good yield
(82%) via silica-gel flash chromatography (dichloromethane/trieth-
ylamine; 99:1), albeit with streaking on the silica (a minor disad-
vantage of avoiding tin complexation). The reduction of 3 to the
1,9-dipyrromethane dicarbinol 7 using sodium borohydride pro-
ceeded quantitatively (>90%). 2+2 MacDonald type condensation
of 5-(4-methyl benzoate)dipyrromethane and dicarbinol 7 condi-
tions, followed by direct zinc insertion of the crude material pro-
duced the methyl ester 8 in relatively good yield.
The bisferrocenyl porphyrin trans-Fc2-ZnP-(CO2Me)2 was also
successfully isolated as a single porphyrin product. The trans-
A2B2 substitution pattern about the porphyrin ring allows a more
straightforward MacDonald 2+2 condensation from 5-ferrocenyldi-
pyrromethane and methyl-4-formylbenzoate.
Standard BF3 catalysis conditions at room temperature in
dichloromethane resulted in scrambling and multiple porphyrin
products for this reaction. However, a single product was formed
with minimal scrambling when acetonitrile was used as the sol-

4702
J. P. Harney et al. / Tetrahedron Letters 53 (2012) 4700–4703
CO2Me
H
CO2Me
O
H
Fe
10 R =
11 R =
NH
NH
HN
HN
(i - ii)
Fe
Fe
N
N
R
Zn
R
or
N
N
7 - 8 %
CO2Me
2
2 or 6
O
MeO2C
CO2Me
5
Scheme 2. Reagents and conditions: (i) CH3CN, BF3ÁOEt2, NH4Cl, 0 °C, p-chloranil (ii) Zn(OAc)2, CH2Cl2, MeOH (iii) THF, aq. NaOH.
(a)
(b)
ZnTCPP
blank TiO2
TiO2 / ZnTCPP
TiO2 / trans-Fc-ZnP-CO2H
TiO2 / trans-Fc2-ZnP-(CO2H)2
1.0
0.8
0.6
0.4
0.2
0.0
trans-Fc-ZnP-CO2H
trans-Fc2-ZnP-CO2H
1.0
0.8
0.6
0.4
0.2
0.0
1.0
0.8
0.6
0.4
0.2
0.0
500
600
700
800
400
500
600
700
800
400
500
600
700
800
Wavelength (nm)
Wavelength (nm)
Figure 2. UV–Vis spectra (a) in methanol (b) on TiO2. Analogous spectra were observed for Fc-ZnP-CO2H and Fc-ZnP-(CO2H)3, for clarity only the former is included.
vent in the presence of ammonium chloride as the water scaven-
ger. A relatively low yield (7%) is observed under these conditions
consistent with literature reports.8 However this is a small trade-
off for the benefit of isolating the single targeted porphyrin product
10. Zinc insertion was carried out in a similar manner to trans-Fc-
ZnP-CO2Me. Porphyrin Fc-ZnP-(CO2Me)3 was prepared under
analogous minimal scrambling conditions yielding free base fol-
lowed by zinc insertion yielding the methyl ester 11 in good yield.
Hydrolysis gave the target compound in relatively good yield
phyrin systems relative to ZnTCPP. The photoelectrochemical
properties of these systems are currently under investigation.
Acknowledgments
J. Rochford thanks the University of Massachusetts Boston for
financial support.
References and notes
11
(Scheme 2).
1. Liska, P.; Vlachopoulos, N.; Nazeeruddin, M. K.; Comte, P.; Gratzel, M. J. Am.
Chem. Soc. 1988, 110, 3686–3687.
2. Vlachopoulos, N.; Liska, P.; Augustynski, J.; Gratzel, M. J. Am. Chem. Soc. 1988,
110, 1216–1220.
The UV–Vis absorption spectra of trans-Fc-ZnP-CO2Me, Fc-ZnP-
(CO2Me)3, and trans-Fc2-ZnP-(CO2Me)2 in methanol are shown in
Figure 2(a) overlayed with the reference compound ZnTCPP. Sub-
stantial electronic coupling of the redox active ferrocene and the
porphyrin is evident. For example, the Soret absorption is signifi-
cantly broadened due to vibronic coupling of the porphyrin S2 state
and the ferrocene substituent resulting in large increases of the
peak full width half maxima (FWHM) from 11 nm for ZnTCPP up
to 32 nm for trans-Fc2-ZnP-(CO2Me)2. In addition, intensity bor-
rowing of both the ferrocenyl MLCT and the porphyrin Q band
transitions is evident. The three complexes trans-Fc-ZnP-CO2Me,
Fc-ZnP-(CO2Me)3, and trans-Fc2-ZnP-(CO2Me)2 were easily hydro-
lyzed under basic conditions and covalently attached to mesopor-
ous TiO2 nanoparticle films.12 UV–Vis absorption spectra of cover
glass/TiO2/porphyrin films are shown in Figure 2(b). Results indi-
cate an increased absorption cross section of the ferrocenyl por-
3. O’Regan, B.; Gratzel, M. Nature 1991, 353, 737–740.
4. Yella, A.; Lee, H.-W.; Tsao, H. N.; Yi, C.; Chandiran, A. K.; Nazeeruddin, M. K.;
Diau, E. W.-G.; Yeh, C.-Y.; Zakeeruddin, S. M.; Grätzel, M. Science 2011, 334,
629–634.
5. Boyle, N. M.; Rochford, J.; Pryce, M. T. Coord. Chem. Rev. 2010, 254, 77–102.
6. Rochford, J.; Rooney, A. D.; Pryce, M. T. Inorg. Chem. 2007, 46, 7247–7249.
7. Lee, C. H.; Lindsey, J. S. Tetrahedron 1994, 50, 11427–11440.
8. Littler, B. J.; Miller, M. A.; Hung, C. H.; Wagner, R. W.; O’Shea, D. F.; Boyle, P. D.;
Lindsey, J. S. J. Org. Chem. 1999, 64, 1391–1396.
9. Zaidi, S. H. H.; Loewe, R. S.; Clark, B. A.; Jacob, M. J.; Lindsey, J. S. Org. Proc. Res.
Dev. 2006, 10, 304–314.
10. Venkatraman, S.; Kumar, R.; Sankar, J.; Chandrashekar, T. K.; Sendhil, K.;
Vijayan, C.; Kelling, A.; Senge, M. O. Chem.-Eur. J. 2004, 10, 1423–1432.
11. 1H NMR spectroscopic data for selected compounds: (4) (CDCl3) d: 2.43 (6H, s,
Me), 4.11–4.26 (9H, m, Cp), 5.32 (1H, s, methine), 6.07 (2H, s, b), 6.73 (2H, s, b),
7.29 (4H, d, J = 8.10 Hz, ph), 7.78 (4H, d, J = 8.10 Hz, ph), 10.26 (2H, s, NH) ppm.
trans-Fc-ZnP-CO2Me (8) (CDCl3) d: 2.72 (6H, s, Me), 4.07 (3H, s, CO2Me), 4.24

J. P. Harney et al. / Tetrahedron Letters 53 (2012) 4700–4703
4703
(5H, s, Cp), 4.83 (2H, d, J = 1.80 Hz, Cp), 5.55 (2H, d, J = 1.80 Hz, Cp), 7.57 (4H, d,
J = 7.80 Hz, tolyl), 8.10 (4H, d, J = 7.80 Hz, tolyl), 8.28 (2H, d, J = 8.40 Hz, ph),
8.37 (2H, d, J = 8.40 Hz, ph), 8.80 (2H, d, J = 5.70 Hz, b), 8.88–8.93 (4H, m, b),
10.20 (2H, d, J = 4.80 Hz, b) trans-(Fc)2-ZnP-(CO2Me)2 (10) (CDCl3) d: 4.12 (6H,
s, OMe), 4.21 (10H, s, Cp), 4.83 (4H, s, Cp), 5.50 (4H, s, Cp), 8.29 (4H, d, J = 7.26
Hz, ph), 8.43 (4H, d, J = 7.26 Hz, ph), 8.74 (4H, d, J = 4.50 Hz, b), 10.11 (4H, d, J =
4.50 Hz, b) Fc-ZnP-(CO2Me)3 (11) (CDCl3) d: 4.08–4.14 (9H, m, OMe), 4.24 (5H,
s, Cp), 4.85 (2H, s, Cp), 5.55 (2H, s, Cp), 8.25–8.43 (12H, m, ph), 8.81– 8.84 (6H,
m, b), 10.22 (2H, d, J = 4.80 Hz, b).
12. Rochford, J.; Chu, D.; Hagfeldt, A.; Galoppini, E. J. Am. Chem. Soc. 2007, 129,
4655–4665.