Molecular engineering of photoinduced charge separation Citation for published version (APA): Marcos Ramos, A. (2003). Molecular engineering of photoinduced charge separation. Technische Universiteit Eindhoven. https://doi.org/10.6100/IR571230 DOI: 10.6100/IR571230 Document status and date: Published: 01/01/2003 Document Version: Publisher’s PDF, also known as Version of Record (includes final page, issue and volume numbers) Please check the document version of this publication: • A submitted manuscript is the version of the article upon submission and before peer-review. There can be important differences between the submitted version and the official published version of record. People interested in the research are advised to contact the author for the final version of the publication, or visit the DOI to the publisher's website. • The final author version and the galley proof are versions of the publication after peer review. • The final published version features the final layout of the paper including the volume, issue and page numbers. Link to publication General rights Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. • Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal. If the publication is distributed under the terms of Article 25fa of the Dutch Copyright Act, indicated by the “Taverne” license above, please follow below link for the End User Agreement: www.tue.nl/taverne Take down policy If you believe that this document breaches copyright please contact us at: [email protected]providing details and we will investigate your claim. Download date: 28. Sep. 2020
Transcript
Molecular engineering of photoinduced charge separation
Citation for published version (APA):Marcos Ramos, A. (2003). Molecular engineering of photoinduced charge separation. Technische UniversiteitEindhoven. https://doi.org/10.6100/IR571230
DOI:10.6100/IR571230
Document status and date:Published: 01/01/2003
Document Version:Publisher’s PDF, also known as Version of Record (includes final page, issue and volume numbers)
Please check the document version of this publication:
• A submitted manuscript is the version of the article upon submission and before peer-review. There can beimportant differences between the submitted version and the official published version of record. Peopleinterested in the research are advised to contact the author for the final version of the publication, or visit theDOI to the publisher's website.• The final author version and the galley proof are versions of the publication after peer review.• The final published version features the final layout of the paper including the volume, issue and pagenumbers.Link to publication
General rightsCopyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright ownersand it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.
• Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal.
If the publication is distributed under the terms of Article 25fa of the Dutch Copyright Act, indicated by the “Taverne” license above, pleasefollow below link for the End User Agreement:www.tue.nl/taverne
Take down policyIf you believe that this document breaches copyright please contact us at:[email protected] details and we will investigate your claim.
•–) are energetically feasible in the three polar solvents (Table 2.2).
The experimental and estimated energies of the various neutral and charge-separated states of
the OAn-OPV-C60 triad are depicted in Figure 2.4, assuming chlorobenzene as the medium. Figure 2.4
also shows the photophysical reactions that have been identified in the triad or its model compounds,
with the prevailing processes highlighted in black, vide infra.
1OAn*-OPV-C601OAn*-OPV-C60
OAn-1OPV*-C60
OAn-OPV-C60
OAn-1OPV*-C60OAn-1OPV*-C60
OAn-OPV-C60OAn-OPV-C60
OAn-OPV•+-C60• -
OAn•+-OPV-C60• -
OAn-OPV•+-C60• -OAn-OPV•+-C60• -
OAn•+-OPV-C60• -OAn•+-OPV-C60• -
OAn•+-OPV•--C60
OAn-OPV-1C60*OAn-OPV-1C60*
OAn-OPV-3C60*OAn-OPV-3C60*
kET
kCS
kCSH
kPLOPV
k´ET
kISCC60
OAn-3OPV*-C60
kISCOPV
kPLC60
Energy
k´CS
k´CR
kCR1
kCR2
Figure 2.4. Schematic energy levels (in chlorobenzene) and photoinduced processes in OAn-OPV-
C60. The rate constants are collected in Table 2.3. The fastest processes have been highlighted in
black (vide infra).
Photoinduced multistep energy and electron transfer in a molecular triad
27
2.4 Photophysical processes in solution
2.4.1 Photoluminescence spectroscopy
Photoluminescence (PL) experiments on the triad, dyads, and model compounds, were used to
study the energy and electron transfer reactions that occur in these molecules. These transfer
relaxation pathways are expected to quench the PL of the chromophores, especially when their rate
constants are higher than that of the intrinsic decay.
500 600 700 8000
3
6
300 400 500 600 7000.0
0.5
1.0b
a
PL
Inte
nsity
(a.
u.)
Wavelength (nm)
Nor
mal
ized
PL
Inte
nsity
Wavelength (nm)
Figure 2.5. (a) PL spectra of OPV-C60 in toluene (dashed line) and of OAn-OPV-C60 in toluene (solid
line) and o-dichlorobenzene (dotted line). (b) Excitation spectra of the 710 nm emission of OPV-C60
(dashed line) and OAn-OPV-C60 (solid line) compared to the absorption spectrum of OAn-OPV-C60
(dashed line), all in toluene.
The PL at 499 nm of the OPV moiety of the OAn-OPV-C60 triad, dissolved in toluene, is
highly quenched (quenching factor Q > 4000) compared to the PL of the corresponding OPV
molecule. Apart from a residual emission at 499 nm,25 photoexcitation of the OPV moiety of OAn-
OPV-C60 results in a weak PL signal at 715 nm (Figure 2.5a), characteristic of the fluorescence
emission band of fulleropyrrolidines.9d In toluene the quantum yield of this emission is nearly
identical to that of the reference compound MP-C60. The same result is observed for the OPV-C60
dyad, although in this case the PL quenching of the OPV is somewhat less (Q ≈ 1500).25 The strong
Chapter 2
28
quenching of the 1OPV* singlet-excited state (S1) in the OPV-C60 dyad has previously been studied in
detail and was found to involve an ultrafast photoinduced intramolecular singlet-energy transfer (ET)
towards the fullerene moiety (1OPV*-C60 → OPV-1C60*), which occurs with a time constant of less
than 190 fs.9e,26 By analogy, the same ET process must occur in the OAn-OPV-C60 triad. In
accordance with this proposition, the excitation spectrum of the fullerene fluorescence of OAn-OPV-
C60 coincides with the corresponding absorption spectrum of OAn-OPV-C60 (Figure 2.5b). It is
interesting to note that the fullerene excitation spectra of OPV-C60 and OAn-OPV-C60 differ
appreciably at lower wavelengths, where the OAn moiety absorbs. The excellent agreement between
the absorption spectrum of the OAn-OPV-C60 triad and the fullerene excitation spectrum implies that
not only excitation of OPV, but also excitation of OAn is responsible for the emission of the fullerene.
This points to an efficient sequential singlet-energy transfer in the triad (Figure 2.4), that starts with
the formation of the 1OAn*-OPV-C60 singlet-excited state and ends at the OAn-OPV-1C60* fullerene
singlet-excited state.
In a more polar solvent, like o-dichlorobenzene (ε = 9.93), the PL of the OPV unit of the
OAn-OPV-C60 triad is quenched to a similar extent as in toluene (Figure 2.5a). In contrast, the PL of
the fullerene moiety at 715 nm is significantly quenched in this more polar solvent as compared to
toluene (Q ≈ 400, Figure 2.5a). The same result has previously been observed for the OPV-C60 dyad9d,e
and gives evidence of a photoinduced charge separation (CS) that occurs from the fulleropyrrolidine
singlet-excited state and produces the OPV•+-C60•– charge-separated state. For the OAn-OPV-C60 triad,
a photoinduced CS will initially produce a similar state (OAn-OPV•+-C60•–), which may then go
through a charge shift (CSH) to generate the energetically more favorable OAn•+-OPV-C60•– state
(Figure 2.4).
For a further comparison, we have studied the PL of the OAn-OPV dyad in solvents of
different polarity. Figure 2.6a shows that the emission of 1(OAn-OPV)*, dissolved in toluene, stems
exclusively from the OPV moiety, irrespective of the excitation wavelength (330 nm (OAn) or 444
nm (OPV)). This implies that an efficient singlet-energy transfer occurs from the excited 1OAn* state
to OPV (1OAn*-OPV → OAn-1OPV*). In toluene, the fluorescence quantum yield of OAn-OPV is
slightly (~10%) higher than that of OPV, but the PL is progressively quenched with increasing solvent
polarity (Figure 2.6a), providing quenching factors of Q ≈ 2, 9, and 22 for chlorobenzene, o-
dichlorobenzene, and benzonitrile, respectively. The quenching of OAn-OPV PL after excitation at
440 nm in more polar solvents is attributed to an intramolecular photoinduced electron transfer
reaction in the excited state to produce the OAn•+-OPV•– state. The excitation spectrum of the residual
emission of OAn-OPV recorded in chlorobenzene (Figure 2.6b) closely corresponds to the absorption
spectrum of OAn-OPV. This indicates that the 1OAn*-OPV → OAn-1OPV* singlet-energy transfer is
significantly faster than an electron transfer from the same state (1OAn*-OPV → OAn•+-OPV•–).
Photoinduced multistep energy and electron transfer in a molecular triad
29
Hence, OAn•+-OPV•– is primarily formed via the OAn-1OPV* singlet state, irrespective of the
excitation wavelength.
400 500 600 700
0
20
40
60
80
100
120
300 400 500 600
0.0
0.5
1.0 b
a
Cou
nts
x 10
-4
Wavelength (nm)
Nor
mal
ized
inte
sity
Wavelength (nm)
Figure 2.6. (a) PL spectra of OAn-OPV in toluene (solid squares), chlorobenzene (open squares), o-
dichlorobenzene (closed circles), and benzonitrile (open circles) solutions with excitation at 444 and
330 nm (for toluene only, dashed line). (b) Excitation spectrum of the OPV emission of OAn-OPV in
chlorobenzene (dashed line) and compared to the absorption spectrum (solid line).
2.4.2 Near steady state photoinduced absorption (PIA) spectroscopy
Near steady state PIA spectroscopy in the microsecond and millisecond time domain is a very
sensitive technique (detection limit ∆T/T ~ 10-6) to probe small concentrations of long-lived
photoexcitations such as triplet states and intermolecular charge-separated states. The PIA spectrum
of OAn-OPV-C60 in toluene solution, recorded with excitation at 458 nm, exhibits a band at 1.78 eV
with a shoulder at 1.52 eV, characteristic of the long-lived (~40 µs) triplet state of the
fulleropyrrolidine moiety (OAn-OPV-3C60*) formed via intersystem crossing from the fullerene
singlet-excited state. 9d
The photoinduced electron transfer in OAn-OPV-C60 in o-dichlorobenzene and subsequent
charge shift or charge recombination will likely occur in the picosecond to nanosecond time regime
and cannot be resolved with the near-steady state PIA technique. However, absorptions of the charged
Chapter 2
30
states can be observed in mixtures of the different model compounds of the triad in o-dichlorobenzene
solution, by using the redox activity of the corresponding triplet states. Photoexcitation of OPV (at
458 nm) or MP-C60 (at 528 nm) will result in the formation of the corresponding excited triplet states.
These triplet states (3OPV* and MP-3C60*) can undergo an electron transfer to one of the other redox
active chromophores present in solution to produce an intermolecularly charge-separated state, which
is long-lived because the formed cation and anion radicals diffuse away in solution.
0.5 1.0 1.5 2.00.0
0.5
1.0
75 Hz 2500 Hz
Energy (eV)
0.0
2.5
c
-∆T
/T x
104
b
a
0.0
0.5
1.0
1.5
Figure 2.7. (a) Normalized photoinduced absorption spectra of the mixtures OPV/MP-C60 (1:1) (solid
line) and OAn/MP-C60 (1:1) (dashed-line) in o-dichlorobenzene (excitation at 528 nm with 25 mW
and modulation frequency of 275 Hz). (b) Photoinduced absorption spectra of the mixtures OAn-
OPV/MP-C60 (1:1) (solid line) or OAn/OPV/MP-C60 (1:1:1) (dashed line) in o-dichlorobenzene
(excitation at 528 nm with 25 mW and modulation frequency of 275 Hz). (c) Normalized photoinduced
absorption spectra of the mixture OAn/OPV/MP-C60 (1:4:4) in o-dichlorobenzene recorded at
modulation frequencies of 75 Hz (solid line) and 2500 Hz (dashed line) (excitation at 528 nm with 25
mW).
Photoinduced multistep energy and electron transfer in a molecular triad
31
Accordingly, selective photoexcitation of MP-C60 at 528 nm in mixtures with OPV or OAn in
o-dichlorobenzene solution produces the charge-separated states OPV•+/MP-C60•– and OAn•+/MP-C60
•–
respectively, which are characterized by the absorptions of OPV•+ (at 0.68 and 1.52 eV), OAn•+ (at
1.40 eV), and MP-C60•– (at 1.24 eV) (Figure 2.7a). Likewise, photoexcitation of MP-C60 in a mixture
with OAn-OPV in o-dichlorobenzene results in a band at 1.40 eV (Figure 2.7b) attributed to OAn•+-
OPV radical cation. Although in this mixture the OAn-OPV•+ radical cation can also be formed, it will
probably rearrange by an intramolecular redox reaction to the OAn•+-OPV state within the timescale
of the experiment. In an equimolar mixture of OAn, OPV, and MP-C60, charge transfer from the
positively charged OPV+ onto the OAn becomes intermolecular, i.e. diffusion limited and slower.
This allows for the detection of a weak residual absorption band characteristic of the OPV•+ radical
cation in the PIA spectrum at 0.68 eV (Figure 2.7b).
Even in a mixture where both OPV and MP-C60 are present in fourfold excess with respect to
OAn, the PIA spectrum is still dominated by the absorption of the OAn•+ radical cation (Figure 2.7c).
At high modulation frequency (2500 Hz) the relative intensity of the OPV•+ radical cation band at
0.68 eV increases compared to low modulation frequency (75 Hz) (when normalized at 1.24 eV),
indicating that in this mixture OPV•+ has a shorter lifetime than the OAn•+ radical cation, consistent
with the proposed OPV•+ + OAn → OPV + OAn•+ redox reaction.
2.4.3 Femtosecond pump-probe spectroscopy in polar solvents
The formation and decay of the transient charged species generated after photoexcitation of
the OPV moiety in OAn-OPV-C60 and the reference compounds OPV-C60 and OAn-OPV have been
investigated with sub-picosecond pump-probe spectroscopy in solvents of different polarity.
Upon excitation at 450 nm of OPV-C60 or OAn-OPV-C60, the transient absorption at 1450 nm
(0.85 eV) that is associated with the OPV•+ radical cation, i.e. the OPV•+-C60•– and OAn-OPV•+-C60
•–
states, exhibits a rise and a decay component (Figure 2.8). It is important to note that the charge
separation occurs subsequent to the ultrafast singlet-energy transfer: OAn-1OPV*-C60 → OAn-OPV-1C60* (kET ≥ 5.3 × 1012
s-1).9e Fitting of the temporal differential absorption data at 1450 nm to a
biexponential function27 provides the rate constants for forward and backward electron transfer
reactions (Table 2.3). As can be seen in Figure 2.8 (and Table 2.3), the rate constants for the charge
separation (kCS) are equal in the dyad and triad within experimental error. The rates for charge
separation in the dyad and triad increase significantly with solvent polarity from kCS = 4.3 × 1010 s
-1 in
chlorobenzene to kCS = 2.0 × 1011 s-1 in benzonitrile. A similar increase is found for the charge
recombination. This effect of the medium on the rate constant has often been observed in donor-
acceptor dyads and is in agreement with Marcus theory, vide infra.28
Chapter 2
32
-60
-40
-20
0
0 50 100 150 200 250
-60
-40
-20
0
-60
-40
-20
0
a
b
c
∆T/T
(a.
u.)
Time delay (ps)
Figure 2.8. Differential transmission dynamics of the OPV•+ transition at 1450 nm for OAn-OPV•+-
C60•–
(solid line) and OPV•+-C60•– (dashed line) in (a) chlorobenzene, (b) o-dichlorobenzene, and (c)
benzonitrile after excitation at 450 nm.
In comparison to the dyad, the charge shift (CSH) from OAn-OPV•+-C60•– to OAn•+-OPV-
C60•– provides an extra pathway in the triad for the decay of the OPV•+ radical other than charge
recombination (CR1, Figure 2.4) to the ground state. Hence, the OPV•+ radical cation absorption may
disappear faster in the triad than in the dyad. The experimental data in Figure 2.8 show, however, that
this difference is only discernable in chlorobenzene, i.e. the least polar solvent. This is a direct
consequence of the increased rate for charge recombination in more polar solvents (Figure 2.8), which
implies that the probability for the competing (slower) charge shift is strongly reduced. Non-linear
least squares analysis of the experimental OPV•+ traces (Figure 2.8) does not give a clear indication of
the magnitude of the rate of charge shift; it yields values of kCR1 and (kCR1 + kCSH) that are essentially
the same (see table 3). This may be explained in part by a difference in kCR1 for the dyad and the triad
and in part by the difficulties in the data analysis itself. Due to the fact that the rates for formation and
decay of the OPV•+ species are very similar, it is difficult to estimate the rate. Non-linear-least-
Photoinduced multistep energy and electron transfer in a molecular triad
33
squares-fit procedures invariably yield rate parameters for formation and decay that are strongly
correlated.
Table 2.3. Rate constants for intrinsic decay of the lowest-energy singlet-excited chromophore (k0OPV,
k0C60), singlet-energy transfer (kET), charge separation (kCS), charge recombination (kCR1), and charge
shift (kCSH) in the studied compounds in toluene (TOL), chlorobenzene (CB), o-dichlorobenzene
(ODCB), benzonitrile (BZN).
Compound TOL
k (ns-1)
CB
k (ns-1)
ODCB
k (ns-1)
BZN
k (ns-1) OPV k0
OPV 0.83 0.79 0.74 0.69
MP-C60 k0C60 0.68 0.72 0.75 0.68
OPV-C60 kET ≥ 5300 n.d. ≥ 5300 n.d.
kCS - 40±2 50±9 250±40
kCR1 - 12±1 19±3 37±8
OAn-OPV-C60 kET ≥ 5300 n.d. ≥ 5300 n.d.
kCS - 48±5 71±12 201±54
kCR1+ kCSH - 11±1 16±3 61±19
OAn-OPV k´CS - 16.4 15.6 40.8
k´CR - 1.0 1.4 4.1
Additional information on the CSH reaction can be obtained by measuring the photoinduced
absorption at 1030 nm (1.20 eV). At this probe wavelength the S1 states 1OPV* and 1MP-C60*, as well
as the radical ions MP-C60•–, and OAn•+ contribute to the induced absorption (Figure 2.7). The
differential transmission at 1030 nm of OPV-C60 and OAn-OPV-C60 in chlorobenzene undergoes a
strong rise and drop within 2 ps (Figure 2.9). This transient signal is associated with the formation and
decay of the 1OPV* singlet-excited state and involves the Sn←S1 transition of the OPV unit.9e The
short lifetime of the 1OPV* state is due to the ultrafast ET onto the fullerene moiety as described
above. After the ET, the 1030 nm signal for the OAn-OPV-C60 triad remains constant over the
timescale of the experiment (1 ns), while for the dyad the signal decays to zero within 200 ps (Figure
2.9). It is proposed that the remaining signal is due to absorption by C60•– and OAn•+ radical ions, and
hence characteristic of the OAn•+-OPV-C60•– charge separated state. Using the following extinction
coefficients for these species (7.7 × 104 M-1cm-1 (OPV(S1), estimate), 7 × 103 M-1cm-1 (MP-C60(S1) 29),
7 × 103 M-1cm-1 (MP-C60•– 29), 8.2 × 103 M-1cm-1 (OAn•+ 20)) the time profile for the absorption at 1030
nm (Figure 2.9) has been modeled for both the OPV-C60 dyad and the OAn-OPV-C60 triad. Rate
Chapter 2
34
constants were taken from the transient absorption measurements at 1450 nm probe wavelength
(Table 2.3) and the rate for charge shift (kCSH) is used as an adjustable parameter.30 The results show a
large induced absorption due to the OPV(S1) species. The duration of this transient is mainly
determined by the width of the laser pulses used. For the modeling a cross-correlation of 500 fs
(FWHM) for the pump and probe pulse was assumed. For the dyad, a smooth trace is observed after
the initial contribution of the excited OPV moiety. This is consistent with the fact that the C60(S1) and
the C60•– groups have almost equal extinction coefficients. For the triad, a long-lived signal is
observed which can be modeled taking kCSH = 5 × 109 s-1 and kCR1 = 6 × 109 s-1. Taking a higher
(lower) value for kCSH results in a curve that bends downward (upward) after 60 ps. Using kCSH = 5 ×
109 s-1 results in a calculated yield of formation for the OAn•+-OPV-C60•– species of about 0.4 per
absorbed photon. This result can be compared with the following crude estimate. At 30 ps mainly the
Oan-OPV•+-C60•– state is present of which only the C60
•– makes a major contribution to the absorption
at 1030 nm. At t= 200 ns only the OAn•+-OPV-C60•– state is present and OAn•+ and C60
•–contribute
almost equally to the transient absorption. The experimental observation that the induced absorption
at 1030 nm hardly changes between 30 and 200 ps thus indicates that the probability of formation of
OAn•+-OPV-C60•– out of OAn-OPV•+-C60
•– is roughly one half, in good agreement with the estimate
above. Figure 2.9 also shows the induced absorption signal at 1450 nm as observed for the triad. This
latter signal has been modeled using the same parameters as mentioned above.
0.1 1 10 100 1000
-100
-50
0
∆T/T
(a.
u.)
Time delay -1 (ps)
Figure 2.9. Differential transmission dynamics of OAn-OPV-C60 (open circles) and OPV-C60 (closed
squares) in chlorobenzene monitored at 1030 nm with excitation at 450 nm. The induced absorption
signal at 1450 nm as observed for the triad is shown with open squares. The lines correspond to a
numerical simulation based on the model described in the text. For the simulation the rate constants
from Table 3 are used and the rate for charge shift is used as an adjustable parameter. The time delay
has been shifted by 1 ps to show the signals on a logarithmic plot.
Photoinduced multistep energy and electron transfer in a molecular triad
35
We also studied the reference compound OAn-OPV with pump-probe spectroscopy in
solution. In particular, it is of interest to investigate whether an OAn-1OPV*-C60 → OAn•+-OPV•–-C60
electron transfer reaction can compete with the OAn-1OPV*-C60 → OAn-OPV-1C60* energy transfer.
As a result of the electron-hole symmetry in conjugated oligomers, the optical spectra of OPV
radical cations and OPV radical anions are expected to be very similar and, hence, we probe the
formation of OPV•– at 1450 nm.31 The transient differential transmission recorded for OAn-OPV,
dissolved in solvents with increasing polarity, at 1450 nm after excitation at 455 nm of the OPV
chromophore is shown in Figure 2.10. In full agreement with the PL quenching of the OAn-OPV
fluorescence (Figure 2.6), the OAn•+-OPV•– charge-separated state is only formed in the three polar
solvents (chlorobenzene, o-dichlorobenzene, and benzonitrile) but not in (less polar) toluene (Figure
2.10). The rate constants for charge separation and charge recombination (k´CS and k´CR, Table 2.3)
increase with increasing polarity. Table 2.3 reveals that the rate for the OAn-1OPV* → OAn•+-OPV•–
charge separation (k´CS = 1.6 - 4.1 × 1010 s
-1) is significantly less than the rate of the 1OPV*-C60 →
OPV-1C60* energy transfer (kET ≥ 5.3 × 1012 s
-1) and as a consequence the OAn•+-OPV•–-C60 state is
unlikely be formed in a significant yield from OAn-1OPV*-C60. Hence, excitation of OAn-OPV-C60 in
solution will always provide OAn-OPV-1C60* as an intermediate state, irrespective of the excitation
wavelength and solvent polarity.
0 200 400 600 800 1000-0.3
-0.2
-0.1
0.0
∆T/T
(a.
u.)
Time delay (ps)
Figure 2.10. Differential transmission dynamics of OAn•+-OPV•– in toluene (closed squares),
chlorobenzene (closed circles), o-dichlorobenzene (open circles), and benzonitrile (open squares)
monitored at 1450 nm with excitation at 455 nm.
Charge recombination in OAn•+-OPV•– is almost an order of magnitude slower than in
OPV•+-C60•– and occurs in the nanosecond regime. In this respect, a remarkable phenomenon was
observed when the fluorescence lifetimes of OAn-OPV were recorded. In the most polar solvents o-
Chapter 2
36
dichlorobenzene and benzonitrile the lifetime of the OAn-OPV emission is significantly reduced to
~140 and ~100 ps, compared to the OPV model compound (1.36 and 1.44 ns, respectively). This is in
accordance with the proposed OAn-1OPV* → OAn•+-OPV•– electron transfer reaction, which reduces
the lifetime of the 1OPV* state. The lifetimes of 1OPV* and OAn-1OPV* in toluene are very similar
(1.20 and 1.38 ns, Figure 2.9) as expected because electron transfer does not occur here. However,
and surprisingly, a significant increase of fluorescence lifetime was observed for OAn-1OPV* in
chlorobenzene (2.65 ns) compared to OPV (1.27 ns) (Figure 2.11). At first glance this increase in
fluorescence lifetime seems to contradict the observed OAn-1OPV* → OAn•+-OPV•– charge
separation reaction, which is approximately 20 times faster than the intrinsic decay (Figure 2.10,
Table 2.3). Instead of an increase in fluorescence lifetime, a decrease would be expected. Note that
except for a loss in intensity (only a factor of 2), the fluorescence spectra of OPV and OAn-OPV are
identical (Figure 2.6), and that the emission is thus from the OAn-1OPV* state. These experimental
observations lead to the conclusion that in chlorobenzene the singlet OAn-1OPV* state and the
OAn•+-OPV•– charge-separated state are nearly degenerate and that back electron transfer from
OAn•+-OPV•– reproduces the OAn-1OPV* singlet state. The estimated change in free energy for OAn-1OPV* → OAn•+-OPV•– of only ∆G0 = – 0.07 eV (Table 2.2) supports this suggestion.
0 5 10
10
100
1000
10000
OAn-OPV: TOL
OAn-OPV: CB
OPV: CB
OPV: TOL
Cou
nts
Time (ns)
Figure 2.11. Time-resolved fluorescence of OPV and OAn-OPV in toluene (TOL) and chlorobenzene
(CB) recorded with excitation at 400 nm.
2.4.4 Kinetic considerations
The final outcome of a photoexcitation not only depends on the energetics of the reaction, but
also on the kinetics. Marcus theory provides an estimate for the free energy barrier (∆G‡) for electron
transfer reactions based on the change in free energy (∆G0) and the reorganization energy (λ) via:
Photoinduced multistep energy and electron transfer in a molecular triad
37
( )
λλ
4
20‡ +∆=∆ G
G (2.2)
The reorganization energy consists of an internal contribution (λ i) and a solvent term (λs),
which can be approximated via the Born-Hush approach to give after summation:
−
−
++=+= −+
s2
0
2
isi11111
2
1
4 επελλλλ
nRrr
e
cc
(2.3)
The rate constants for the different processes, are not only a function of the energy barrier
∆G‡, but also of the reorganization energy (λ) and the electronic coupling (V) between donor and
acceptor in the excited state according to the equation:
( )
+∆−
=
Tk
GV
Tkhk
B
202
21
B2
2
4exp
4
λλ
λπ
(2.4)
The values of ∆G0, λ, and ∆G‡ calculated on the basis of Eqs. 2.1 to 2.4 collected in Table
2.2, show that the initial charge separation (kCS) is in the Marcus normal region (–∆G0 < λ). As the
polarity of the solvent increases, the OAn-OPV•+-C60•– charge-separated state is stabilized with a
concomitant increase of the reorganization energy. The combination of these trends results in
reduction of the barrier for charge separation in more polar solvents (Table 2.2). As a consequence,
the rate for charge separation (kCS) is expected to increase with polarity as has been found
experimentally (Figure 2.8, Table 2.3). Charge recombination in OAn-OPV•+-C60•– is in the Marcus
inverted region (–∆G0 > λ). The use of Eq. (2.4) in the inverted region often underestimates the true
rate constant because of nuclear tunneling,32 but will be used here for qualitative comparison. We find
that the barrier for charge recombination is strongly reduced in more polar solvents, consistent with
the higher recombination rate (kCR1, Table 2.3). Apart from recombination to the ground state, the
OAn-OPV•+-C60•– state may undergo a charge shift to form OAn•+-OPV-C60
•–. The charge shift occurs
in the normal region (–∆G0 < λ) and is energetically less favorable than the recombination (Table 2.3).
However, in contrast to the recombination, the barrier for charge shift is reduced with decreasing
polarity. Hence, the balance between charge recombination and the competing charge shift, will move
towards the latter with decreasing polarity. As a result, the barrier for the charge shift in OAn-OPV•+-
C60•– is lower than that for charge recombination in chlorobenzene, but not in o-dichlorobenzene and
benzonitrile. This is in full agreement with the experimental result that the charge shift was more
easily observed in chlorobenzene (Figure 2.8 and 2.9).
Chapter 2
38
The energy barriers for relaxation to the ground state in OAn•+-OPV-C60•– are small.
However, besides the energy barrier, the rate constant (Eq. 2.4) is also determined by the electronic
coupling V. This electronic coupling depends exponentially on the center-to-center distance between
the donor and acceptor via ))(exp()( 0cc02
02 RRRVV −−= β , with R0 the contact distance. Hence, V
is orders of magnitude less for OAn•+-OPV-C60•– (Rcc = 30.0 Å) than for OAn-OPV•+-C60
•– (Rcc = 15.4
Å).33 The reduction of the electronic coupling V, caused by the longer distance between the centers of
positive and negative charge density in OAn•+-OPV-C60•–, is the origin of the increase in lifetime for
OAn•+-OPV-C60•– compared to OAn-OPV•+-C60
•–.
The rate constants of the various photoinduced processes in solvents of different polarity have
been calculated using Eq. 2.4 relative to the corresponding rate constant in chlorobenzene by
assuming that V is independent on the solvent, and are collected in Table 2.4.
Table 2.4. Rate constants for the charge separation (CS), charge recombination (CR1), charge shift
(CSH), and charge recombination (CR2) in o-dichlorobenzene (ODCB) and benzonitrile (BZN) with
relative to the corresponding rate constant in chlorobenzene (CB), calculated using Eq. 2.4.
rate constant CS CR1 CSH CR2
k(CB) 1 1 1 1
k(ODCB) 2.00 89 0.58 6.25
k(BZN) 2.16 980 0.23 5.06
2.5 Photophysical processes in the solid state
2.5.1 Near steady state PIA spectroscopy
Near steady state PIA spectra of thin films of OPV-C60 and OAn-OPV-C60 were recorded at
80 K with excitation at 458 nm. The PIA spectrum of the OPV-C60 film (Figure 2.12) exhibits the
signals of OPV•+ at 0.68 and 1.52 eV and that of C60•– at 1.24 eV, characteristic of a charge-separated
state. The PIA spectrum of a film of OAn-OPV-C60 (Figure 2.12) lacks the bands at 0.68 and 1.52 eV
and exhibits only one broad band that peaks at 1.25 eV. This signal is attributed to the overlapping
transitions of the OAn•+ and C60•– radical ions and gives evidence for the formation of an
intramolecular (OAn-OPV•+-C60•–) or intermolecular (OAn•+-OPV-C60 / OAn-OPV-C60
•–) charge-
separated state. The charge-separated state in the OPV-C60 and OAn-OPV-C60 films, measured with
this PIA technique, extend into the millisecond time domain. The PIA band at 1.25 eV increases with
the pump intensity (I) following a square-root power law (–∆T/T ~ I0.5). This suggests a non-geminate
bimolecular recombination of the photoinduced charges. We propose that the long-lived (ms domain)
Photoinduced multistep energy and electron transfer in a molecular triad
39
charges observed with near steady state PIA in films of OPV-C60 and OAn-OPV-C60 at 80 K, are
associated with a small fraction of positive and negative charges that have escaped from geminate
recombination by charge migration to other sites in the film where they became trapped and are thus
associated with different molecules in the film (i.e. OAn•+-OPV-C60 and OAn-OPV-C60•–).
0.5 1.0 1.5 2.0
0.0
0.5
1.0
1.5
2.0
∆T
/T x
104
Energy (eV)
Figure 2.12. Photoinduced absorption spectra of OAn-OPV-C60 (solid line) and OPV-C60 (dotted
line) in thin films. Recorded at 80 K with excitation at 458 nm (25 mW) and a modulation frequency
of 275 Hz.
2.5.2 Femtosecond pump-probe spectroscopy
Femtosecond spectroscopy shows that in films of OPV-C60 and OAn-OPV-C60 at room
temperature a charge-separated state is formed within 0.5 ps, as evidenced by the instantaneous rise of
the 1450 nm differential transmission (Figure 2.13a) associated with OPV•+ radical cations. Although
a lack of higher time-resolution precludes an unambiguous conclusion, we have no evidence that a
singlet-energy transfer precedes the electron transfer reaction in the solid state. Possibly, the
photoinduced electron transfer in the solid state is predominantly intermolecular. The OPV•+-C60•–
state is much longer lived in the film than in solution (Table 2.3). This phenomenon has also been
observed in other donor-acceptor dyads and has been attributed to the migration of opposite charges to
different sites in the film.8h,9d,34 In contrast, the OPV•+ signal in the film of the triad rapidly decays,
with a time constant of 20 ps over the first 100 ps. This is interpreted to result from an intermolecular
or intramolecular charge shift (CSH) from OPV•+ to OAn•+. The time profile of the 1030 nm
differential transmission (Figure 2.13b) is consistent with the formation of an OAn•+-OPV-C60•– state
in the triad and explains the short lifetime of the OAn-OPV•+-C60•– state (Figure 2.13a). This time
profile cannot be fitted to monoexponential decay, and suggests several lifetimes. This fact can be
rationalized by a combination of a direct charge recombination and an indirect charge recombination
Chapter 2
40
after migration of the charges in the film.8h,9d It should be noticed however, that contrary to what is
observed for the OPV-C60 dyad, the OAn•+-OPV-C60•– state seems to be longer lived in solution than
the OAn•+-OPV-C60 / OAn-OPV-C60•– state in the film. As a tentative explanation for this difference
we propose that in solution the triads are isolated from each other such that the weak electronic
coupling between the OAn and the C60 decelerates the intramolecular charge recombination in OAn•+-
OPV-C60•–. In the film the triads are in intimate contact with each other, and intermolecular charge
recombination between OAn•+-OPV-C60 and OAn-OPV-C60•– can occur.
0 100 200 300 400 500 600-120
-80
-40
0
0 200 400 600 800 1000-100
-80
-60
-40
-20
0b
a
Time delay (ps)
∆T/T
(a.
u.)
∆T/T
(a.
u.)
Time delay (ps)
Figure 2.13. (a) Differential transmission dynamics at 1450 nm of thin films of OAn-OPV-C60 (open
circles) and OPV-C60 (closed circles) at 298 K with excitation at 450 nm. (b) Differential transmission
dynamics at 1030 nm of a thin film of OAn-OPV-C60 at 298 K with excitation at 450 nm.
2.6 Conclusions
A molecular triad, OAn-OPV-C60, with a redox gradient in a linear array has been
synthesized. The photophysical processes that may occur in this system are schematically depicted in
Figure 2.4 and have been investigated in solution and in thin films with photoluminescence and
transient absorption spectroscopy.
Photoinduced multistep energy and electron transfer in a molecular triad
41
In solution, photoexcitation of either chromophore of the OAn-OPV-C60 triad results in an
ultrafast (sequence of) singlet-energy transfer (ET, Figure 2.4) and provides a singlet state on the
fulleropyrrolidine unit (OAn-OPV-1C60*), irrespective of the polarity of the solvent. The competitive
process of charge separation from the primary 1OAn*-OPV-C60 or OAn-1OPV*-C60 states are more
than one order of magnitude slower and were not observed in the triad. In toluene, the singlet OAn-
OPV-1C60* state decays via intersystem crossing (ISC) to the OAn-OPV-3C60* triplet state and via PL
to the ground state. In more polar solvents, the singlet OAn-OPV-1C60* state gives rise to an
intramolecular charge separation reaction (CS) that generates the OAn-OPV•+-C60•– state. The rate for
this forward electron transfer reaction increases with the polarity of the solvent from kCS = 4.8 × 1010 s
-1
in chlorobenzene to kCS = 2.0 × 1011 s
-1 in benzonitrile, in qualitative agreement with Marcus theory.
Because the oxidation potential of the OAn segment is below that of the OPV unit, the primary OAn-
OPV•+-C60•– charge-separated state may undergo an intramolecular redox reaction, or charge shift
(CSH), to form OAn•+-OPV-C60•–. The charge recombination in OAn-OPV•+-C60
•– (CR1), however,
competes with the charge shift. Because the charge recombination is slowed down in less polar
solvents, the quantum yield for the charge shift is the highest (~0.4) in the least polar solvent in which
electron transfer occurs, i.e. chlorobenzene. The charge recombination in the secondary OAn•+-OPV-
C60•– charge-separated state (CR2) is significantly slower (kCR2 < 1 × 109
s-1) than that of the primary
OAn-OPV•+-C60•– state (kCR1 = 1.1 × 1010
s-1). The observed trends in the various rate constants (kCS,
kCSH, kCR1, kCR2, k´SC and k´CR) with changing solvent polarity are in qualitative agreement with Marcus
theory when the free energies of the charge-separated states are determined using a continuum model
(Eq. 2.1).
In thin films, charge generation on the OPV unit of OAn-OPV-C60 is much faster (kCS ≥ 3.0 ×
1012 s
-1) and likely predominantly intermolecular. In the films a subsequent charge-shift occurs from
the primary OPV•+ radical cation to an OAn•+ radical cation with a rate close to kCSH = 5.0 × 1010 s
-1.
Because we consider it likely that in the film the primary charge-separated state involves two
molecules, also the charge shift probably involves an intermolecular OAn-OPV•+-C60 → OAn•+-OPV-
C60 reaction, with the negative charge located on the fullerene unit of a third molecule (OAn-OPV-
C60•–). The lifetime of the charges formed in the film (OAn•+-OPV-C60 / OAn-OPV-C60
•–) is somewhat
less as compared to the solution.
The experiments on OAn-OPV-C60 demonstrate that in solution the OAn-OPV•+-C60•– →
OAn•+-OPV-C60•– charge shift and the resulting spatial extension of the charges increase the lifetime
of the charge-separated state compared to OAn-OPV•+-C60•–, because the electronic coupling between
the redox active groups is strongly reduced. This provides a rationale to explain the long-lifetime of
the charge-separated state in conjugated polymer:C60 blends in terms of charge migration. The major
differences in the kinetics of the electron transfer reactions observed after photoexcitation of OAn-
Chapter 2
42
OPV-C60 in solution or in thin films, further demonstrate that intermolecular interactions are of crucial
significance in this respect. Creating and investigating well-defined multichromophoric
supramolecular donor-acceptor assemblies, consisting of many judiciously positioned chromophores,
will enable a more detailed understanding of photoinduced charge-separation processes in natural and
artificial systems.
2.7 Experimental section
All reagents and solvents were used as received or purified using standard procedures. C60 was
purchased from BuckyUSA. NMR spectra were recorded on a Varian Unity Inova and a Varian Unity Plus at frequencies of 500 and 125 MHz for 1H and 13C nuclei; a Varian Mercury Vx at frequencies of 400 and 100 MHz for 1H and 13C nuclei or Varian Gemini 2000 at frequencies of 300 and 75 MHz for 1H and 13C nuclei, respectively. Tetramethylsilane (TMS) was used as an internal standard for 1H NMR and CDCl3, CD3COCD3 or CS2 for 13C NMR. Infrared (FT-IR) spectra were recorded on a Perkin-Elmer Spectrum One UATR FT-IR. Elemental analyses were preformed on a Perkin Elmer 2400 series II CHN Analyzer. Matrix assisted laser desorption ionization time-of-flight mass spectrometry (MALDI-TOF MS) was performed on a Perseptive DE PRO Voyager MALDI-TOF mass spectrometer using a dithranol matrix. All HPLC analyses were performed on a Hewlett Packard HP LC-Chemstation 3D (HP 1100 Series) with DAD detection using an Inertsil® 5 Si column (250x3 mm). A Shimadzu LC-10AT system combined with a Polymer Laboratories MIXED-D column (Particle size: 5µm; Length/I.D. (mm): 300 × 7.5) and UV detection was employed for size exclusion chromatography (SEC), using CHCl3 as an eluent (1 mL/min). (E,E)-4-{4-(4-methyl-2,5-bis[(S)-2-methylbutoxy]styryl)-2,5-bis[(S)-2-methylbutoxy]styryl}-2,5-bis[(S)-2-methylbutoxy}-5-benzaldehyde-dimethylacetal (2). Amberlite IR 120 (1.5 g), trimethyl orthoformate (20 mL) and 1 (1.2 g, 1.78 mmol) where added to 100 mL methanol. The suspension was stirred under an argon atmosphere at 70 °C for 2 h. The reaction mixture was cooled to room temperature and 1.5 g of Na2CO3 was added. The suspension was filtered and the solvent was removed in vacuo to yield 1.38 g (91%) of 2, which was used without further purification. 1H NMR (CDCl3, 400 MHz): δ (ppm) 7.50 (d, 1H),7.49 (s, 2H), 7.44 (d, 1H), 7.18 (s, 1H), 7.17 (s, 1H), 7.16 (s, 1H), 7.10 (s, 1H), 7.08 (s, 1H), 6.73 (s, 1H), 3.92-3.74 (m, 12H), 3.42 (s, 6H), 2.24 (s, 3H), 1.98-1.88 (m, 6H), 1.69-1.54 (m, 6H), 1.39-1.26 (m, 6H), 1.10-0.96 (m, 36 H); 13C NMR (CDCl3, 100 MHz): δ (ppm) 151.69, 151.10, 150.96, 150.67, 150.45, 128.17, 127.71, 127.57, 126.98, 126.56, 125.21, 123.19, 123.06, 122.54, 121.71, 116.32, 111.80, 110.14, 109.94, 109.26, 108.39, 99.75, 74.69, 74.35, 74.28, 74.23, 73.73, 73.38, 54.34, 35.11, 35.08, 35.05, 34.96, 34.89, 26.36, 26.26, 26.21, 16.79, 16.70, 16.40, 11.46, 11.37, 11.33. N-(4-Diphenylaminophenyl)-N-phenyl-3-aminobenzaldehyde (4). To a tube fitted with a magnetic stirrer was added 3 (0.8 g, 2.38 mmol), 3-bromobenzaldehyde (1.32 g, 7.14 mmol), Pd2(dba)3 (0.022g, 0.024 mmol), BINAP (0.044 g, 0.071 mmol) and Cs2CO3 (1.16 g, 3.57 mmol). After purging with argon, freshly distilled toluene (11.9 mL) was added. The reaction mixture was heated at 100 °C under Ar atmosphere. After 48 h Pd2(dba)3 (0.022g, 0.024 mmol), BINAP (0.044 g, 0.071 mmol) and Cs2CO3 (1.16 g, 3.57 mmol) were added and the mixture was heated for another 72 h. After cooling to room temperature, the reaction mixture was filtered over Celite 545 and concentrated in vacuo. Column chromatography (SiO2, heptane/CH2Cl2 1:1, Rf = 0.4) and evaporation to dryness from heptane yielded 0.643 g (62 %) of a yellow powder. IR (UATR) ν (cm-1) 3034, 2923, 2851, 1698, 1583, 1485, 1277, 1263, 749, 690, 626. 1H NMR (CDCl3, 400 MHz): δ (ppm) 9.92 (s, 1H), 7.56 (t, 1H), 7.43(dt, 1H), 7.37 (t, 1H), 7.34-7.23 (m, 7H), 7.13-7.11 (m, 6H), 7.08-6.96 (m, 7H); 13C NMR (CDCl3, 100 MHz): δ (ppm) 192.24, 149.14, 148.00, 147.45, 144.17, 142.11, 137.91, 129.98, 129.70, 129.45, 128.18, 126.37, 125.41, 124.62, 124.20, 123.67, 123.07, 122.91, 122.66. Anal. Cald for C31H24N2O: C, 84.5; H, 5.5; N, 6.4. Found: C, 84.1; H, 5.1; N, 6.1. (E)-N,N’-(Diphenyl)-N’-(4-diphenylaminophenyl)-3-aminobenzaldimine (5). To a suspension of 4 (0.55 g, 1.24 mmol) in ethanol (50 mL), was added aniline (0.14g, 1.50 mmol). The reaction mixture was heated at 85 °C for 4 h. After cooling to room temperature the product precipitated slowly
Photoinduced multistep energy and electron transfer in a molecular triad
43
from the ethanol. The product was obtained as 0.508 g (79%) of a yellow powder after washing with ethanol. IR (UATR) ν (cm-1) 3034, 1628, 1589, 1486, 1267, 751, 693. 1H NMR (CDCl3, 400 MHz): δ (ppm) 8.34 (s, 1H), 7.61 (t, 1H), 7.53 (dt, 1H), 7.38-7.31(m, 3H), 7.28-7.10 (m, 16H), 7.03-6.97 (m, 7H); 13C NMR (CDCl3, 100 MHz): δ (ppm) 160.31, 151.99, 148.47, 147.78, 147.56, 143.29, 142.34, 137.40, 129.62, 129.33, 129.19, 129.09, 126.32, 125.91, 125.66, 125.26, 123.95, 123.88, 123.53, 122.82, 122.52, 122.44, 120.85. Anal. Cald for C37H29N3: C, 86.2; H, 5.7; N, 8.1. Found: C, 85.8; H, 5.3; N, 8.0. MALDI-TOF MS (Mw = 515.65) m/z = 515.23 [M]+. (E,E,E)-4-[4-{4-[N-(4-Diphenylaminophenyl)-N-(phenyl)-3-aminostyryl]2,5-bis[(S)-2-methylbutoxy]styryl}-2,5-bis[(S)-2-methylbutoxy]styryl}-2,5-bis[(S)-2-methylbutoxy]benzaldehyde (6). Schiff base 5 (0.450 g, 0.873 mmol) and acetal 2 (0.775 g, 0.873 mmol) were dissolved in DMF (5 mL). The mixture was heated to 80 °C under an argon atmosphere and potassium tert-butoxide (0.352 g, 3.142 mmol) was added. The reaction mixture was stirred for 3 h. After cooling to room temperature, the reaction mixture was poured on ice, washed with HCl 3N and brine, dried over MgSO4 and concentrated in vacuo. Column chromatography (SiO2, toluene/cyclohexane 7:3 Rf = 0.4, and heptane/CH2Cl2 6:2, Rf = 0.2) and evaporation to dryness from heptane yielded 0.795 g (72%) of an orange powder. IR (UATR) ν (cm-1) 3062, 2960, 2917, 2873, 1675, 1589, 1504, 1490, 1422, 1262, 1200, 1039, 968, 744, 693. 1H NMR (CDCl3, 500 MHz): δ (ppm) 10.43(s, 1H), 7.64 (d, 1H), 7.54 (d, 1H), 7.51 (d, 1H), 7.50 (d, 1H), 7.40 (d, 1H), 7.33 (s, 1H), 7.28-6.98 (m, 29H), 3.79 3.98 (m, 12H), 1.98-1.91 (m, 6H), 1.70-1.57 (m, 6H), 1.39-1.31 (m, 6H), 1.13-0.94 (m, 36H); 13C NMR (CDCl3, 100 MHz): δ (ppm) 189.06, 156.43, 151.44, 151.26, 151.04, 150.64, 148.21, 147.84, 142.91, 142.70, 139.13, 135.22, 129.45, 129.19, 129.18, 128.50, 128.45, 127.45, 126.74, 126.48, 126.42, 125.46, 125.30, 123.97, 123.79, 123.74, 123.56, 123.19, 122.98, 122.56, 122.43, 122.40, 122.21, 121.90, 120.29, 110.64, 110.29, 110.17, 109.82, 109.74, 109.64, 74.39, 74.25, 74.11, 74.05, 73.91, 73.69, 35.13, 35.01, 34.93, 34.88, 34.84, 26.38, 26.34, 26.33, 26.18, 16.88, 16.86, 16.81, 16.76, 16.63, 11.52, 11.46, 11.38, 11.32. Analysis Calcd for C85H102N2O7 C, 80.8; H, 8.1; N, 2.2. Found: C, 80.7; H, 7.6; N, 2.2. MALDI-TOF MS (Mw = 1263.75) m/z = 1263.71 [M]+. N-Methyl-2<4-[4-{4-[N’-(4-Diphenylaminophenyl)-N’-(phenyl)]3-aminostyryl]-2,5-bis[(S)-2-methylbutoxy]styryl}-2,5-bis[(S)-2-methylbutoxy]styryl]-2,5-bis[(S)-2-methylbutoxy]>-3,4-fulleropyrrolidine (7). A solution of aldehyde 6 (0.2 g, 0.158 mmol), finely ground N-methylglycine (84.4 mg, 0.949 mmol) and C60 (227 mg, 0.316 mmol) in chlorobenzene (60 mL) was stirred and refluxed in the dark under an atmosphere of dry nitrogen for 18 h. After cooling to room temperature the solvent was removed in vacuo and the remaining residue was purified by column chromatography on silica gel (toluene/cyclohexane: 2/1, Rf = 0.4) to afford triad 7 as a ~1/1 mixture of diastereomers according to HPLC (eluent: toluene/cyclohexane 50/50 v/v; flow: 1 mL/min.; peaks at tr = 9.1 and 10.3 min.). Traces of impurities were effectively removed after a 2nd chromatographic purification on silica gel (CS2/toluene: 1/0 to 7/3 Rf = 0.3) to afford analytically pure material (assay >99.5%) according to HPLC and GPC analysis. The product was precipitated from a concentrated toluene solution with methanol (100 mL) and the resulting solid was washed with methanol (2 × 100 mL) and finally dried in vacuo at 55 ºC. The triad was obtained as a light brown powder (137 mg, 43%). IR (UATR) ν (cm-1) 2957, 2914, 2872, 1589, 1502, 1491, 1264, 1190, 965, 750, 694, 526. 1H NMR (CS2, 500 MHz): δ (ppm) 7.61-6.96 (m, 35H), 5.62 (s, 1H), 5.05 (d, 1H), 4.41 (d, 1H), 4.15-3.70 (m, 12H), 2.92 (s, 3H), 2.16-1.26 (m, 18H), 1.02-1.26 (m, 36H). 13C NMR (CS2, 125 MHz): δ (ppm) 156.50, 154.91, 154.84, 154.18, 154.15, 153.53, 151.61, 150.93, 150.76, 150.71, 150.67, 150.63,147.79, 147.40, 147.06, 146.60, 146.57, 146.52, 146.13, 146.07, 146.03, 146.00, 145.93, 145.89, 145.85, 145.75, 145.50, 145.47, 145.37, 145.26, 145.19, 145.07, 145.04, 145.01, 144.98, 144.91, 144.50, 144.37, 144.31, 144.15, 142.92, 142.86, 142.51, 142.46, 142.43, 142.40, 142.35, 142.12, 142.10, 142.06, 141.99, 141.98, 141.95, 141.92, 141.79, 141.63, 141.53, 141.51, 140.06, 139.98, 139.69, 139.51, 139.07, 136.35, 135.99, 135.88, 135.86, 134.49, 134.44, 129.42, 129.20, 129.17, 127.99, 127.68, 127.66, 127.40, 127.27, 126.88, 126.30, 125.24, 125.14, 124.85, 124.80, 123.79, 123.70, 123.61, 123.29, 122.69, 122.56, 122.52, 122.47, 122.29, 121.96, 120.39, 114.29, 114.16, 110.04, 109.63, 109.40, 109.22, 108.71, 108.61, 76.53, 75.43, 75.40, 73.97, 73.91, 73.69, 73.59, 73.52, 73.44, 72.95, 72.91, 69.70, 68.98, 40.02, 35.41, 35.31, 35.29, 35.25, 35.20, 35.18, 35.17, 26.86, 26.84, 26.80, 26.49, 17.14, 17.09, 17.06, 17.02, 17.00, 16.80, 12.04, 12.03, 12.00, 11.94, 11.89, 11.85. Analysis Calcd for C147H107N3O6 C, 87.7; H, 5.4; N, 2.1. Found: C, 87.8; H, 5.0; N, 2.1. MALDI-TOF MS (Mw = 2011.44) m/z = 2011.18 [M]+. (E,E)-2-{4-(4-methyl-2,5-bis[(S)-2-methylbutoxy]styryl)-2,5-bis[(S)-2-methylbutoxy]styryl}-1,4-bis[(S)-2-methylbutoxy]-5-deuteriobenzene (9). n-BuLi 1.6 M (0.13 mL, 0.21 mmol) was added dropwise to a solution of 8 (0.15 g, 0.16 mmol) in freshly distilled diethyl ether (3 mL) at –10° C. The reaction mixture was stirred for 5 minutes. After the addition of
Chapter 2
44
D2O (0.5 mL) the cooling bath was removed and the reaction mixture was stirred for 2 h at room temperature. The mixture was estracted with diethyl ether dried over MgSO4 and concentrated in vacuo. Column (SiO2; heptane/CH2Cl2: 1/1, Rf = 0.3) and recrystallization from ethanol yielded 54 mg (41%) of the product as yellow crystals. 1H NMR (CDCl3, 400 MHz): δ (ppm) 7.49 (d, 1H), 7.48 (s, 2H), 7.44 (d, 1H), 7.19(s, 1H), 7.18 (s, 1H), 7.17 (s, 1H), 7.10 (s, 1H), 6.82 (s, 1H), 6.72(s, 1H), 3.92-3.71 (m, 12H), 1.99-1.82 (m, 6H), 1.68-1.54 (m, 6H), 1.38-1.23 (m, 6H), 1.20-0.86 (m, 36H); 13C NMR (CDCl3, 100 MHz): δ (ppm) 151.68, 151.14, 150.94, 150.90, 150.44, 128.21, 127.61, 127.54, 127.10, 125.23, 123.24, 122.98, 122.82, 121.74, 116.31, 113.96, 111.58, 110.36, 109.85, 108.34, 74.68, 74.46, 74.30, 74.26, 73.59, 73.38, 35.11, 35.06, 35.04, 34.97, 34.80, 26.36, 26.34, 26.26, 26.19, 16.80, 16.70, 16.56, 16.40, 11.46, 11.40, 11.37, 11.31. Anal. Cald for C53H79DO6: C, 77.8; H, 10.2. Found: C, 77.5; H, 9.1. MALDI-TOF MS (Mw = 813.96) m/z = 813.53[M]+.
(E,E,E)-2-[4-{4-[N-(4-Diphenylaminophenyl)-N-(phenyl)-3-aminostyryl]-2,5-bis[(S)-2-methylbutoxy]styryl}-2,5-bis[(S)-2-methylbutoxy]styryl]-1,4-bis[(S)-2-methylbutoxy]-5-deuteriobenzene (10). Schiff base 5 (0.025 g, 0.049 mmol) and 9 (0.40 g, 0.049 mmol) were dissolved in DMF (5 mL). The mixture was heated at 80 °C under an argon atmosphere and potassium tert-butoxide (0.020 g, 0.176 mmol) was added. The reaction mixture was stirred for 5 h at 80 °C. After cooling to room temperature, the reaction mixture was poured onto ice and washed with HCl 3N and brine. The organic layer was dried over MgSO4 and concentrated in vacuo. Column chromatography (SiO2, toluene/cyclohexane 7:3, Rf = 0.6) and recyrstallization from hexane, heptane and a few drops of CH2Cl2 yielded 30 mg of 10 (50%) as a yellow powder. IR (UATR) ν (cm-1) 2960, 2917, 2873, 1588, 1504, 1489, 1255, 1205, 1042, 970, 744, 693. 1H NMR (CD3COCD3, 400 MHz): δ (ppm) 7.61 (s, 2H), 7.60 (d, 1H), 7.55 (d, 1H), 7.47(d, 1H), 7.35-6.95 (m, 30H), 4.01-3.76 (m, 12H), 2.09-1.83 (m, 6H), 1.74-1.55 (m, 6H), 1.45-1.26 (m, 6H), 0.95-1.15 (m, 36H); 13C NMR (100 MHz): δ (ppm) 155.25, 152.89, 152.84, 152.72, 150.10, 149.58, 149.54, 144.87, 144.58, 141.06, 131.32, 131.03, 130.98, 130.19, 129.47, 129.13, 128.96, 128.91, 128.15, 127.30, 127.04, 125.39, 125.35, 125.16, 124.93, 124.87, 124.73, 124.68, 124.37, 124.34, 124.30, 123.17, 122.16, 115.71, 113.14, 112.34, 111.95, 111.58, 111.39, 75.61, 75.56, 75.53, 75.36, 74.72, 36.76, 36.68, 36.63, 36.44, 27.85, 27.81, 27.60, 18.04, 17.98, 17.95, 17.92, 17.60, 12.63, 12.60, 12.58, 12.52, 12.39. Anal. Cald for C85H102N2O7: C,81.6; H, 8.2; N, 2.3. Found: C, 81.2; H, 7.7; N, 2.2. MALDI-TOF MS (Mw = 1236.74) m/z = 1235.62 [M]+.
Electrochemistry. Cyclic voltammograms were measured in 0.1 M tetrabutylammonium hexafluorophosphate (TBAPF6) as a supporting electrolyte in dichloromethane (or THF) using a Potentioscan Wenking POS73 potentiostat. The working electrode was a Pt disk (0.2 cm2), the counter electrode was a Pt plate (0.5 cm2), and a saturated calomel electrode (SCE) was used as reference electrode, calibrated against Fc/Fc+ (+0.43 V).
Absorption and Photoluminescence. UV/visible/near-IR absorption spectra were recorded on a Perkin Elmer Lambda 900 spectrophotometer. Fluorescence spectra were recorded on an Edinburgh Instruments FS920 double-monochromator spectrometer and a Peltier-cooled red-sensitive photomultiplier. Time-correlated single photon counting. Time-correlated single photon counting fluorescence studies were performed using an Edinburgh Instruments LifeSpec-PS spectrometer. The LifeSpec-PS comprises a 400 nm picosecond laser (PicoQuant PDL 800B) operated at 2.5 MHz and a Peltier-cooled Hamamatsu micro-channel plate photomultiplier (R3809U-50). Lifetimes were determined from the data using the Edinburgh Instruments software package.
Near steady state PIA. Solutions were prepared in a nitrogen-filled glove box in order to exclude interference of oxygen during measurements. The PIA spectra were recorded between 0.5 and 3.0 eV by exciting with a mechanically modulated cw Ar ion laser (λ = 458 or 528 nm, 275 Hz) pump beam and monitoring the resulting change in transmission of a tungsten-halogen probe light through the sample (∆T) with a phase-sensitive lock-in amplifier after dispersion by a grating monochromator and detection, using Si, InGaAs, and cooled InSb detectors. The pump power incident on the sample was typically 25 mW with a beam diameter of 2 mm. The PIA (-∆T/T ≈ ∆αd) was directly calculated from the change in transmission after correction for the PL, which was recorded in a separate experiment. PIA and PL spectra were recorded with the pump beam in a direction almost parallel to the direction of the probe beam. The solutions were studied in a 1 mm near-IR grade quartz cell at room temperature. Solvents for PIA measurements were distilled under nitrogen before use. The solid-state measurements were performed on films, drop cast from chloroform solution, on quartz substrate and held at 80 K in an Oxford continuous flow cryostat.
Photoinduced multistep energy and electron transfer in a molecular triad
45
Transient subpicosecond photoinduced absorption. The femtosecond laser system used for pump-probe experiments consists of an amplified Ti/sapphire laser (Spectra Physics Hurricane). The single pulses from a cw mode-locked Ti/sapphire laser were amplified by a Nd-YLF laser using chirped pulse amplification, providing 150 fs pulses at 800 nm with an energy of 750 µJ and a repetition rate of 1 kHz. The pump pulses at 450 nm were created via optical parametric amplification (OPA) of the 800 nm pulse by a BBO crystal into infrared pulses which were then two times frequency doubled via BBO crystals. The probe beam was generated in a separate optical parametric amplification set-up in which 1030 and 1450 nm pulses were created. The pump beam was focused to a spot size of about 1 mm2 with an excitation flux of 1 mJ cm-2 per pulse. For the 1030 and 1450 nm pulses a RG 850 nm cut-off filter was used to avoid contributions of residual probe light (800 nm) from the OPA. The probe beam was reduced in intensity compared to the pump beam by using neutral density filters of OD = 2. The pump beam was linearly polarized at the magic angle of 54.7° with respect to the probe, to cancel out orientation effects in the measured dynamics. The temporal evolution of the differential transmission was recorded using Si or an InGaAs detector by a standard lock-in technique at 500 Hz. Solutions in the order of 2-5 × 10-4 M were excited at 450 nm, i.e. providing primarily excitation of the OPV part within the molecules. 2.8 References and notes 1 (a) The Photosynthetic Reaction Center, Deisenhofer, J., Norris, J. R., Eds.; Academic Press: New York,
1993. (b) Anoxygenic Photosynthetic Bacteria, Blankenschip, R. E., Madigan, M. T., Bauer, C. E., Eds.;
17 (a) Siegrist, A. E. Helv. Chim. Acta 1967, 50, 906. (b) Siegrist, A. E.; Meyer, H. R.; Weber, K. Helv.
Chim, Acta 1969, 52, 2521.
18 The bromine has been replaced by a deuterium atom in order to simplify the 1H-NMR spectrum of the
molecule.
19 Murata, Y.; Shine, H. J. J. Org. Chem. 1969, 34, 3368.
20 van Meurs, P. J. High-Spin Molecules of p-Phenylenediamnine Radical Cations, PhD Thesis, Eindhoven
University of Technology, 2002, ISBN 90-386-2574-X.
21 van Hal, P. A.; Beckers, E. H. A.; Peeters, E.; Apperloo, J. J.; Janssen, R. A. J. Chem. Phys. Lett. 2000,
328, 403.
22 Weller, A. Z. Phys. Chem. Neue Folge 1982, 133, 93.
23 The assumption that charges are located at the centers, is of course a simplification of the actual situation
in which charges are delocalized. Especially for C60, the charge is not expected to be at the center, but
rather at the outer surface.
24 Williams, R. M.; Zwier, J. M.; Verhoeven, J. W. J. Am. Chem. Soc. 1995, 117, 4093.
Chapter 2
48
25 The PL lifetime of the residual emission at 510 nm contained a contribution of long lifetime (~ 1 ns) and
is in part attributed to residual PL emission by minor impurities in the sample.
26 In toluene the energy level of initial charge separated state [OAn-OPV•+-C60•–] is higher than that of
OAn-OPV-1C60* due to the destabilization in the nonpolar solvent. In such case one can also expect that
initial electron transfer from 1OPV* to C60 occurs to yield OAn-OPV•+-C60•–, which subsequently
recombines to generate OAn-OPV-1C60*. The stepwise pathway would compete with the direct energy
transfer pathway from 1OPV* to C60 to produce the same state. However, in view of the high rate of the
energy transfer (190 fs) it is very unlikely that this process occurs and the same rate of 190 fs has been
found in ODCB, where the OAn-OPV•+-C60•– state is below the OAn-OPV-1C60* state (Ref 9e).
27 The function f(t)= a0(exp(-a1(t-a3) - exp(-a2(t-a3)) + a4) was used to fit the data and all 5 parameters (a0 –
a4) were optimized.
28 (a) Wasielewski, M. R.; Niemczyk, M.P.; Svec, W. A.: Pewitt, E. B. J. Am. Chem. Soc. 1985, 107, 1080.
(b) Asahi, T.; Ohkohchi, M.; Matsusaka, R.; Mataga, N.; Zhang, R. P.; Osuka, A.; Maruyama, K. J. Am.
Chem. Soc. 1993, 115, 5665. (c) Macpherson, A. N.; Liddell, P. A.; Lin, S.; Noss, L.; Seely, G. R.;
DeGraziano, J. M.; Moore, A. L.; Moore, T. A.; Gust, D. J. Am. Chem. Soc. 1995, 117, 7202. (d) Kroon,
K.; Verhoeven, J. W.; Paddon-Row, M. N.; Oliver, A. M. Angew. Chem Int. Ed. Engl. 1991, 30, 1358. (e)
Imahori, H.; El-Khouly, M. E.; Fujitsuka, M.; Ito, O.; Sakata, Y.; Fukuzumi S. J. Phys. Chem. A 2001,
105, 325.
29 Guldi, D. M.; Prato, M. Acc. Chem. Res. 2000, 33, 695.
30 The following kinetic scheme has been used to model the photoinduced absorption at 1030 nm. In this
scheme, the intramolecular recombination of charges in OAn+-OPV-C60– and also charge generation from
electronic states other than the singlet-excited state of the C60 moiety have been neglected.
CRCOPV
kkk
kkk
CSHCSET
↓↓↓
−− →−−→→ −•+•−•+•
6000
60601601 COPVOAnCOPVOAn)S(C)OPV(S
For this model the following set of coupled differential equations can be written describing the change in
the concentration of each of the four species involved over time.
−−−−
+−+−+
+−
=
−−−−
−•+•
−•+•
−•+•
−•+•
]COPV[OAn
]COPV[OAn
)]S([C
)][OPV(S
000
0)(0
00)(
000)(
]COPV[OAn
]COPV[OAn
)]S([C
)][OPV(S
60
60
160
1
1
1
0
0
60
60
160
1
60
CR
CSHCRCS
CSC
ET
ETOPV
k
kkk
kkk
kk
dt
d
The singlet-excited state of the OPV is assumed to be generated by photon absorption at time t = 0 and
thus the relevant boundary value condition is:
Photoinduced multistep energy and electron transfer in a molecular triad
49
==
−−−−
−•+•
−•+•
0
0
0
)](0)[OPV(S
)0(
]COPV[OAn
]COPV[OAn
)]S([C
)][OPV(S 1
60
60
160
1
t
A solution to the above set of equations is:
))(exp()0]()OPV(S[))](OPV(S[ 011 tkkt ETOPV +−=
[ ]))(exp())(exp()0]()OPV(S[)]()(SC[ 60
6000
001160 tkktkk
kkkk
kt C
CSETOPV
CCSET
OPVET +−++−−
−−+=
[ ]
[ ]
[ ]
+−
−−+−−+
++−−−+−−+
−
+
+−
−−+−−+
=−− −•+•
))(exp())((
))(exp())((
))(exp())((
)0)](OPV(S[
)](COPVOAn[
1
1010
0
1000
0
1000
1
60
60
60
6060
60
tkkkkkkkkkk
kk
tkkkkkkkkkk
kk
tkkkkkkkkkk
kk
t
CSHCR
CSHCRETOPV
CSHCRC
CS
CSET
CCS
CSHCRC
CSC
CSETOPV
CSET
ETOPV
CSHCRETOPVC
CSETOPV
CSET
[ ]
[ ]
[ ]
+−−+−−+
++−−+−−+
−
+
+−−+−−+
++−
+−−+−−+−
++−+−−+−−+
+
+−
+−−+−−+−
×=−− −•+•
))()((
))()((
))()((
))(exp())()((
)(exp())()((
))(exp())()((
)0)](OPV(S[)](COPVOAn[
11000
01000
11000
1
11000
0
01000
0
1000
160
60
606060
60
60
60
606060
60
CSHCRCSHCRETOPVC
CSETOPV
CSHCSET
CCSCSHCR
CCS
CCSET
OPVCSHCSET
CSHCRCSHCRETOPVC
CSETOPV
CSHCSET
CSHCR
CSHCRCSHCRETOPVC
CSETOPV
CSHCSET
CCSC
CSCSHCRC
CSC
CSETOPV
CSHCSET
ETOPV
CSHCRCSHCRETOPVC
CSETOPV
CSHCSET
kkkkkkkkkk
kkk
kkkkkkkkkk
kkk
kkkkkkkkkk
kkk
tkkkkkkkkkkkk
kkk
tkkkkkkkkkkkk
kkk
tkkkkkkkkkkkk
kkk
t
In these equations it can be verified quite easily that they satisfy the boundary condition at t = 0.
Furthermore, setting kCS = 0, the yield of C60(S1) (abbrv. φ(C60(S1))) is given by kET/( kET + k0OPV).
Similarly, setting kCR1 = 0, the yield of OAn-OPV•+-C60•– is φ(C60(S1)) kCS /( kCS + k0
C60). Finally, the yield
of OAn• +-OPV-C60•– is: φ (C60(S1)) φ(OAn-OPV•+-C60
•–) kCSH /( kCSH + kCR1).
Chapter 2
50
31 Deussen, M; Bässler, H. Chem. Phys. 1992, 164, 247.
32 Jortner, J. J. Chem. Phys. 1976, 64, 4860.
33 Imahori et. al. reported that the electronic coupling in a triad (0.019 cm-1) is much less compared to that
in a dyad (3.9 cm-1), see ref. 15o.
34 (a) Thomas, K. G.; Biju, V.; Guldi, D. M.; Kamat, P. V. George, M. V. J. Phys. Chem. B 1999, 103,
8864. (b) Imahori, H.; Hasobe, T.; Yamada, H.; Kamat, P. V.; Barazzouk, S.; Fujitsuka, M.; Ito, O.;
Fukuzumi, S. Chem. Lett. 2001, 784.
Chapter 3
Photoinduced multistep electron transfer in
oligoaniline - oligo(p-phenylene vinylene) -
perylene arrays
Abstract
Multistep energy and electron transfer have been studied in solutions containing two different OAn-
OPV-PERY-OPV-OAn pentads, corresponding to a donor(1)-donor(2)-acceptor-donor(2)-donor(1)
symmetrical arrangement. The pentads have been synthesized by double coupling of an oligoaniline-
oligo(p-phenylene vinylene) (OAn-OPV) dyad to a central perylene diimide (PERY) segment, directly
connected (pentad 1) and via a saturated spacer (pentad 2). Photoexcitation of any of the
chromophores yields in first instance the OAn-OPV•+-PERY•—-OPV-OAn charge separated state. This
state can relax to the ground state via charge recombination or generate the OAn•+-OPV-PERY•–-
OPV-OAn state via a charge shift. In pentad 1, the intramolecular charge separation and charge
recombination are extremely fast. The same processes are slowed down in pentad 2, because the
saturated spacer reduces the electronic coupling between the OPV donor and PERY acceptor. For
pentad 1 in toluene, the OAn-OPV•+-PERY•–-OPV-OAn and the OAn•+-OPV-PERY•–-OPV-OAn states
are nearly isoenergetic and thus there is no driving force for the charge shift process in this apolar
solvent. In more polar media, the charge shift competes with the fast charge recombination to the
ground state with an estimated efficiency of 0.22. The slower charge recombination of OAn-OPV•+-
PERY•–-OPV-OAn to the ground state in pentad 2, with respect to that in pentad 1, is beneficial for
the charge shift process that occurs with an efficiency of 0.28 in polar media. The low electronic
coupling between the OAn and PERY chromophores accounts for the remarkably long-lived OAn•+-
OPV-PERY•–-OPV-OAn geminate ion pair (τ > 1 ns).
Chapter 3
52
3. 1 Introduction
Fuelled by intriguing applications in molecular electronics, light harvesting, photocatalysis,
and artificial photosynthesis, photoinduced energy and electron transfer in covalently linked donor-
acceptor molecules attract enormous attention in recent years.1 For practical applications it is often of
interest to combine a high rate for charge separation with a low rate for charge recombination. For this
purpose, multichromophoric arrays have been designed in which the initial photoinduced charge
separation reaction is followed by charge migration reactions along a well-defined redox gradient that
ultimately provides the spatial separation of the photogenerated charges, which is essential to lower
the rate for charge recombination. Various elegantly designed triads, tetrads, pentads, etc. have been
synthesized and investigated in detail.2,3
Multistep electron transfer and associated charge migration is possibly also the origin of the
longevity of charges in solid state blends of donor and acceptor materials used in organic and polymer
solar cells.4 In these cells, excitons, created by the absorption of light, dissociate at the donor-acceptor
interface forming an electron-hole pair. When this electron and hole escape from geminate
recombination and diffuse away from the interface, they can be collected at electrodes as a
photocurrent.
Recent studies on covalently linked oligo(p-phenylene vinylene) (OPV) donors and perylene
diimide (PERY) acceptors in liquid crystalline OPV-PERY-OPV triads,5,6 π-stacks of hydrogen-
bonded OPV-PERY-OPV trimers,7 and alternating (OPV-PERY)n copolymers,8 have revealed that
photoinduced electron transfer can be extremely fast in solution (< 1 ps) and that the lifetime is
generally rather short (<1 ns). Because of their high absorption coefficients in the visible region and
their charge transport properties, OPV9-14 and PERY15-19 chromophores have also attracted interest for
organic solar cells.
With the aim to extend the lifetime of the charge-separated state, two new multichromophoric
molecular arrays incorporating a central OPV-PERY-OPV triad augmented with two p-oligoaniline
(OAn) moieties as additional donor, have been designed (1 and 2 in Figure 3.1). In both donor(1)-
donor(2)-acceptor-donor(2)-donor(1) pentads, the OAn and OPV donors are electronically decoupled
in the ground state by an m-phenylene linkage and operate essentially as isolated redox active
segments. The two pentads differ in the connectivity between the central PERY segment and the OPV
donor. In pentad 1 the OPV and PERY units are directly connected providing a close proximity of the
chromophores and possibly some overlap of the π-electronic system, even though the two
chromophores are not coplanar. In 2 the OPV-PERY distances are increased and their π-overlap is
fully interrupted by saturated bonds. The redox potentials of the OAn, OPV, and PERY segments
favor the OAn•+-OPV-PERY•–-OPV-OAn state as the lowest-energy charge-separated state after
photoexcitation. To elucidate the mechanism of multistep photoinduced electron transfer in these
pentads and to assess the lifetime of the charge separated state, the photophysical properties of 1 and 2
Photoinduced multistep electron transfer in multichromophoric arrays
53
have been studied using photoluminescence and transient photoinduced absorption spectroscopy in
solvents of different polarity in comparison to those of the isolated chromophores and dyad
combinations. The rates for charge separation, migration, and recombination are rationalized using
Marcus theory and a continuum model for describing the free energy.
OO
O
OO
OOO
O
OO
ONNN N
NN
O
O
O
O
1
OO
O
OO
O
NN
OO
O
OO
ON N
NO
ONN
N O
O
O
O
O
O
2
N N NN
O
O
O
O
OO
O
OO
ON N
O
O
O
O
OO
O
OO
O
OAn OPV PERY
OAn-OPV
Figure 3.1 Chemical structures of pentads 1 and 2 and reference compounds.
3.2 Synthesis
The synthesis of pentads 1 and 2 and the reference compound OAn-OPV are depicted in
schemes 3.1 to 3.3. The synthesis of pentad 1 (Scheme 3.1) starts with the coupling of aldehyde 3,
whose synthesis has been described in chapter 2, to diethyl 4-nitrobenzyl phosphonate 4 via a Wittig-
Horner reaction, affording nitro compound 5. The reduction of the nitro group to the terminal amine 6
was achieved using stannous dichloride and ethanol in ethyl acetate. Condensation of 6 with 3,4,9,10-
perylenetetracarboxy-dianhydride in imidazole and catalytic amounts of Zn(OAc)2 afforded pentad
OAn-OPV-PERY-OPV-OAn 1. Purification of 1 was achieved via extensive column chromatography
and preparative size exclusion chromatography, which resulted in a low yield for this last step. For the
Chapter 3
54
synthesis of reference dyad OAn-OPV 8, aldehyde 3 was reacted with diethyl benzylphosphonate 7 in
a Wittig-Horner reaction (Scheme 3.2).
N NOR*
OR*OR*
R*OR*O
R*OO
N NOR*
OR*OR*
R*OR*O
R*OR
N
N NOR*
OR*OR*
R*OR*O
R*OO
O
NO2(EtO)2P
O
O
6. R = NH2
5. R = NO2
1
a
b
c
2
OR* =
43
+
Scheme 3.1. Synthesis of OAn-OPV-PERY-OPV-OAn (1). a. t-BuOK, DMF/THF (2/1), r.t., 2 h, 86%;
b. SnCl2·H2O, EtOH, EtOAc, 95° C, 5 h, 70%; c. Imidazole, Zn(OAc)2, 160 °C, 4 h, 15%.
N NOR*
OR*OR*
R*OR*O
R*OO
N NOR*
OR*OR*
R*OR*O
R*O
(OEt)2P
O
O
3
8
a
OR* =
7
+
Scheme 3.2. Synthesis of OAn-OPV (8). a. t-BuOK, DMF/THF (2/1), r.t., 2.5 h, 73%.
Photoinduced multistep electron transfer in multichromophoric arrays
55
In pentad 2 (Scheme 3.3) a saturated spacer has been intercalated between the OAn-OPV
segments and the perylene diimide central unit. This saturated spacer was first introduced on the
perylene unit by a double condensation of (S)-(+)-leucinol with 3,4,9,10-
tetracarboxyperyleneanhydride, affording the dihydroxy terminated perylene diimide 10. The amino
function in OAn-OPV 6 was reacted with phosgene in toluene to yield isocyanate 9. Pentad 2 was
synthesized by double reaction of isocyanate 9 with the hydroxyl functions of 10 using dibutyltin
dilaurate as catalyst.
N NOR*
OR*OR*
R*OR*O
R*ONCO
OHNN
OH
O
O
O
O
N NOR*
OR*OR*
R*OR*O
R*ONH2
O
N
O
O
O
N NOR*
OR*OR*
R*OR*O
R*ON
H
O
6
9 10
2
a
b
OR* =
+
2
Scheme 3.3. Synthesis of OAn-OPV-PERY-OPV-OAn (10). a. Phosgene, toluene, 95 °C, 16 h, 100%;
b. Dibutylin dilaurate, CH2Cl2, reflux, 20 h, 47%.
All final compounds were characterized using 1H-NMR spectroscopy, mass spectrometry, and
size exclusion chromatography.
In analogy with a previously studied analogous molecular triad based on OPV-PERY-OPV,6
the pentads and reference dyad were studied with 1H-NMR spectroscopy. Pentad 1 and reference dyad
8 feature the spectral characteristics of molecularly dissolved species when measured in deuterated
dichloromethane. However, pentad 2 features four clearly separated and well-resolved signals for the
perylene protons, whereas in a molecular dissolved state only two different signals are expected
(Figure 3.2a). This difference between the aromatic protons points to the formation of aggregates via
Chapter 3
56
intermolecular hydrogen bonding as has been established for the analogous triads.6 The spectrum of
pentad 2 recorded in THF (Figure 3.2b) shows the signals typical for molecular dissolved species and
indicates that intermolecular hydrogen bonding is much less present in THF.
7.88.28.69.09.49.8
N-H
N-H
a
b
1 2 3 4
1-4
7.88.28.69.09.49.8
N-H
N-H
a
b
1 2 3 4
1-4O RORNN
O
O
O
ONH O
NO H
2 1
13
3
4
4
2
R = OPV-OAn
Figure 3.2. 1H-NMR spectra of pentad 2 in dichloromethane (a) and THF (b).
The 1H-NMR results indicate that pentad 1 and reference dyad 8 are molecular dissolved in
most organic solvents of intermediate polarity. Pentad 2, however, will dimerize or aggregate in a
variety of solvents due to intermolecular hydrogen bonding of the urethane functionalities. This has
implications for the opto-electronic behavior of pentad 2 and therefore the solvent for 2 must be
carefully selected. Tetrahydrofuran seems to be a solvent that allows for the study of pentad 2 in its
molecularly dissolved state.
3.3 Electronic properties and energetic considerations
Absorption spectroscopy. The absorption spectrum of pentad 1 in toluene solution (Figure
3.3 left) exhibits two strong absorption bands, one centered at 327 nm and the other featuring vibronic
fine structure with maxima at 464, 494 and 532 nm. For comparison, the absorption spectra of the
corresponding individual chromophores OAn, OPV and PERY as well as that of dyad OAn-OPV are
also plotted in Figure 3.3. Whereas all chromophores contribute to the absorption of the UV region of
the pentad, the absorption in the visible region is dominated by the π-π* transitions of the OPV and
PERY chromophores. The vibronic fine structure typical for the PERY chromophore is still present in
the visible region.
Photoinduced multistep electron transfer in multichromophoric arrays
57
300 400 500 6000.0
0.5
1.0
1.5N
orm
aliz
ed A
bsor
banc
es (
O.D
.)
Wavelength (nm)300 400 500 600
0.0
0.3
0.6
Abs
orba
nce
(O. D
.)
Wavelength (nm)
Figure 3.3. Left: UV/Visible absorption spectra of the OAn-OPV-PERY-OPV-OAn pentad 1 (solid
line) and model compounds OAn-OPV (solid squares), OAn (open squares), OPV (open triangles),
and PERY (open circles) recorded in toluene solution. Right: UV-Visible absorption spectra of the
OAn-OPV-PERY-OPV-OAn pentad 2 in tetrahydrofuran (solid line), dichloromethane (dashed line)
and toluene (dotted line) solutions.
The UV-visible absorption spectrum of pentad 2 (Figure 3.3 right) recorded in
tetrahydrofuran, consists of the overlapping absorptions of the individual chromophores OAn, OPV
and PERY. However, the low energy absorption band of the PERY at 532 nm exhibits a lower
intensity in pentad 2 with respect to pentad 1, suggesting aggregation, most probably via π-π
interactions. In toluene and dichloromethane the absorption spectra show a further reduction of the
intensity of the 532 nm peak and a weak contribution above 550 nm, which indicates aggregation of
the pentads. In these solvents, a combination of intermolecular hydrogen bonding and π-π interactions
is probably responsible for the aggregation.6
Electrochemistry. The electrochemical properties of pentads 1 and 2 and of the reference
compounds OAn, OPV, and PERY were investigated using cyclic voltammetry. The redox potentials
are collected in Table 3.1. The cyclic voltammogram of pentad 1 shows two reversible reduction
waves at –0.53 and –0.76 V, corresponding to the reduction of the PERY acceptor, and four reversible
oxidation waves of the OAn (+0.53 V and +1.03 V) and the OPV (+0.76 and 0.96 V) donor moieties
(potentials are given vs. SCE, calibrated against Fc/Fc+, recorded in dichloromethane with 0.1 M
TBAPF6) (Table 3.1). In dichloromethane pentad 2 also exhibits four oxidation waves at +0.53 and
+1.07 V associated with the OAn segment and at +0.74 and +0.90 V corresponding to the OPV
segment. In this solvent no reduction waves could be observed for pentad 2. The measurement of the
reduction potentials in dichloromethane was most likely hampered by the aggregation via hydrogen
Chapter 3
58
bonding. In THF, however, the cyclic voltammogram of pentad 2 exhibits the two reduction waves of
the PERY moiety at -0.60 and –0.89 V.
Table 3.1 reveals that the redox potentials of the OAn and OPV donors are hardly affected by
their linkage to the other units. In contrast, the reduction potentials of the PERY acceptor depend on
the substitution of the imide functionalities. The reduction potentials of 1 and 2 are shifted by +0.12
and +0.05 V respectively with respect to that of the PERY reference compound (-0.65 V). For pentad
2 the shift is less, owing to the similar alkyl substitution as in the PERY reference. In pentad 1, the
overlap of the π-electronic clouds of the OPV and PERY chromophores is possible, although the first
phenyl ring of the OPV moiety will not be coplanar with the PERY unit.
Table 3.1. One-electron redox potentials (E0) of OAn, OPV, PERY, OAn-OPV, and OAn-OPV-PERY-
OPV-OAn (vs. SCE) calibrated with Fc/Fc+ (in dichloromethane with 0.1 M TBAPF6).
Compound E0red (V) E0
ox (V)
OAn 0.53 / 1.02
OPV 0.73/ 0.80
PERY -0.65/-0.85
Pentad 1 -0.53/-0.76 0.53 / 0.76/ 0.96/ 1.07
Pentad 2 -0.60/-0.89a 0.53 / 0.74/ 0.90/ 1.07
a Measured in THF.
Energetic considerations. The scheme in Figure 3.4 illustrates the various photophysical
processes that may occur in the pentads upon illumination. Excitation of one of the redox-active
chromophores generates locally excited states, e.g. the OAn-OPV-1PERY*-OPV-OAn singlet excited
state. Exergonic processes such as energy transfer (ET) and charge separation (CS) might compete
with intrinsic decay processes generating cor OAn•+-OPV•–-PERY-OPV-OAn charge-separated states.
These charge-separated states either decay to the ground state via charge recombination (CR) or
evolve via charge shift (CSH) to the more stable OAn•+-OPV-PERY•–-OPV-OAn state.
Photoinduced multistep electron transfer in multichromophoric arrays
59
1OAn*-OPV-PERY1OAn*-OPV-PERY
OAn-1OPV*-PERYOAn-1OPV*-PERY
OAn-OPV-PERYOAn-OPV-PERY
OAn-OPV•+-PERY• -
OAn•+-OPV-PERY• -
OAn•+-OPV•--PERYOAn-OPV-1PERY*OAn-OPV-1PERY*kET2
kCS1
kCR1
kCR2
kCSH1
kPLOPV
kET1
kISCOPV
kPLPERY
Energy
kCS3
kCR3
kCS2
OAn-3OPV*-PERY
kCSH2
kCS4
kPLOAn
Figure 3.4. Schematic energy levels (in THF) and photoinduced processes in OAn-OPV-PERY-OPV-
OAn pentads 1 and 2. (In the scheme the pentads are defined as OAn-OPV-PERY for simplification)
The rate constants are collected in Table 3.2. The fastest processes have been highlighted in black.
The energies of the singlet-excited states are at 3.40, 2.39 and 2.33 eV for the OAn, OPV and
PERY chromophores respectively, as extracted from absorption spectroscopy. In contrast to the
singlet- excited states, the energy of the charged separated states is strongly influenced by the polarity
of the medium. The Weller equation (Eq. 3.1) provides a means of calculating the energy of the
charged states in function of the polarity of the solvent and distance between chromophores. 20
( ) ( )( )
−
+−−−−=∆ −+
sref0
2
ccs0
2
00redox0 1111
84AD
εεπεεπε rr
e
R
eEEEeG (3.1)
In this equation, Eox(D) and Ered(A) are the oxidation and reduction potentials of the donor
and acceptor molecules or moieties measured in a solvent with relative permittivity εref, E00 is the
Chapter 3
60
energy of the excited state from which the electron transfer occurs, and Rcc is the center-to-center
distance of the positive and negative charges in the charge separated state. The radii of the positive
and negative ions are given by r+ and r– and εs is the relative permittivity of the solvent, -e is the
elemental charge, and ε0 is the vacuum permittivity.
Using equation 3.1 the change in free energy of the different processes of pentads 1 and 2 in
solutions of low to high polarity, i.e. toluene (ε = 2.38), chlorobenzene (ε = 5.72), THF (ε = 7.51), and
o-dichlorobenzene (ε = 9.93), have been calculated (Table 3.2). For the calculation the radius of the
negative perylene radical anion, PERY•–, was set to r– = 4.7 Å8 and that of the radical cations of OPV
and OAn to r+ = 5.4521 and 4.8 Å (Chapter 2) respectively. The Rcc distances were determined,
assuming that the charges are located at the centers of the OAn, OPV, and PERY moieties.22 The Rcc
distances were, in pentad 1, 21 and 41 Å for the OAn-OPV•+-PERY•–-OPV-OAn and OAn•+-OPV-
PERY•–-OPV-OAn charge-separated states respectively. In pentad 2 the relative position between the
OAn-OPV segment and the PERY central unit is not constant due to the relatively flexible character
of the spacer separating them. The upper limit for the Rcc distances in pentad 2 (27 and 44 Å for OAn-
OPV•+-PERY•–-OPV-OAn and OAn•+-OPV-PERY•–-OPV-OAn respectively) have been determined
in a conformation in which all chromophores are coplanar.
The experimental and estimated energies of the various neutral and charge-separated states of
the pentads are depicted in Figure 3.4, assuming THF as the medium. In the energetic scheme, the
dominant photophysical processes are depicted in black to highlight them from the slower and
therefore less probable processes, which are depicted in gray. The slow processes are the intrinsic
decay to the ground state of the individual chromophores (kPLOAn, kPL
OPV and kPLPERY), the intersystem
crossing to the triplet state for the OPV chromophore (kISCOPV), and the formation of the OAn•+-OPV•–
-PERY-OPV-OAn charge-separated state either from the 1OAn*-OPV-PERY-OPV-OAn or OAn-1OPV*-PERY-OPV-OAn singlet excited states.23 The change in free energy for the fast processes is
collected in Table 3.2. Here, the charge separation (CS1) process refers to the reaction taking place
from the lowest singlet excited state, i.e. OAn-OPV-1PERY*-OPV-OAn → OAn-OPV•+-PERY•–-
OPV-OAn. This separated state can also originate directly from the OPV singlet excited state (kCS2),
however, energy transfer to generate the PERY singlet excited (kET2) state is likely to compete with
this process.
The charge separation process, charge shift and charge recombination (CS1, CSH1, CR1, and
CR2) are exergonic in all solvents, with the exception that the driving force for charge shift is zero in
toluene. Similar energy values are obtained for pentads 1 and 2 because the only difference taken in
consideration was the Rcc parameter.
Photoinduced multistep electron transfer in multichromophoric arrays
61
Table 3.2. Change in free energy (∆G0) with reference to the lowest singlet excited state,
reorganization energy (λ), and barrier (∆G‡) for charge separation (CS1, OAn-OPV-1PERY*-OPV-
OAn → OAn-OPV•+-PERY•–-OPV-OAn), charge recombination (CR1, OAn-OPV•+-PERY•–-OPV-
To unravel the different photophysical processes occurring on the picosecond to nanosecond
time domain, subpicosecond transient pump-probe spectroscopy has been performed at room
temperature on solutions of pentad 1 in solvents of different polarity. For pentad 2 the study has been
restricted to THF, because in this solvent aggregation is less pronounced. The experiments have been
performed by selective excitation of the OPV or PERY chromophores at 455 or 520 nm, respectively,
and by monitoring the transient absorptions at 1450, 900, and 700 nm. The absorption at 1450 nm is
0.5 1.0 1.5 2.0 2.5-1
0
1
2
3
4
-∆T
/T x
104
Energy (eV)
Photoinduced multistep electron transfer in multichromophoric arrays
65
caused by the low-energy transition of the OPV•+ radical cation21 and, hence, this transient signal
gives direct information on the rates of formation (kCS = kCS1+kCS2) and decay (kdecay) of the OAn-
OPV•+-PERY•–-OPV-OAn charge separated state. The decay of this charge-separated state is the sum
of the rate constants for charge recombination to the ground state and charge shift to the OAn•+-OPV-
PERY•–-OPV-OAn state (kdecay = kCR1 + kCSH1). The radical ions of all chromophores absorb at 700
nm, with estimated molar absorption coefficients of 7 × 103, 15 × 103 and ~80 × 103 M-1cm-1 for
OAn•+, OPV•+,27 and PERY•–26 respectively. The 1OPV* and 1PERY* states dominate the transient
absorption at 900 nm. Also at this wavelength all radical ions absorb, though with much lower molar
absorption coefficients for OPV•+ and PERY•– than at 700 nm. On the basis of the molar absorption
coefficients of the radical ions a OAn-OPV•+-PERY•–-OPV-OAn to a OAn•+-OPV-PERY•–-OPV-OAn
charge shift can be identified by comparing the intensities of transient absorptions at 700 nm
(dominated by the PERY•– radical anion), 900 nm (dominated by the S1 states of the OPV or PERY
chromophores), and 1450 nm (dominated by the OPV•+ radical cation).
3.4.3.1 Pentad 1
Apolar medium (toluene solution). The 1450 nm transient absorption of pentad 1 in toluene
solution after excitation at 450 nm exhibits a fast rise and a slow decrease of the (negative) intensity
(Figure 3.9). The rise occurs within 1 ps, which corresponds to a rate constant for charge separation
(kCS) of > 1000 ns-1. This fast formation of the OAn-OPV•+-PERY•–-OPV-OAn charge-separated state
is in agreement with the complete quenching of the PERY emission (Figure 3.5, right). The slow
decay (3×109 s-1) points to a long-lived charge-separated state (~ 366 ps) in toluene.
The transient absorption recorded at 700 nm essentially superimposes with the absorption at
1450 nm (Figure 3.9). This strongly suggests that both absorptions originate from the same charged
species, i.e. the OAn-OPV•+-PERY•–-OPV-OAn and, hence, no charge shift is taking place in toluene
solution.
The 900 nm differential transmission, after photoexcitation at 450 nm (OPV segment),
exhibits in the first picosecond an abrupt rise and decay of the intensity corresponding to the
formation and decay of the OPV singlet excited state (Figure 3.9). A similar feature is observed if the
PERY chromophore is excited instead, although in this case the short-lived absorption corresponds to
the PERY singlet excited state. Both S1 states evolve within 1 ps into the OAn-OPV•+-PERY•–-OPV-
OAn charge-separated state, as evidenced by the 1450 and 700 nm transient absorptions. Apart from
this initial transient feature, the absorption at 900 nm exhibits the same temporal evolution as the 1450
and 700 nm differential transmissions, which is associated with the OAn-OPV•+-PERY•–-OPV-OAn
charge-separated state.
Chapter 3
66
0 100 200 300 400 500 600
-6
-5
-4
-3
-2
-1
0
-4 -2 0 2 4 6
-5
0
∆T (
a. u
.)
Time (ps)
∆T (
a. u
.)
Time (ps)
Figure 3.9. Differential transmission dynamics of pentad 1 in toluene monitored at 1450 (open
circles), 900 nm (closed squares) and 700 (open triangles) with excitation at 450 nm. The inset shows
the differential transmissions on shorter time scale.
Polar media. The 1450 nm transient absorption reveals that the formation of the OAn-OPV•+-
PERY•–-OPV-OAn charge-separated state in pentad 1 in THF solution is as fast as in toluene, but that
the lifetime of the charged state is dramatically shorter (~ 12 vs. 366 ps) (Figure 3.10). Again, the
same kinetics are observed regardless which chromophore is excited, OPV or PERY.
In the differential transmission at 700 nm of pentad 1 in THF two different regimes can be
distinguished (Figure 3.10). The first regime consists of a fast rise and decay of the signal occurring in
the initial 30 ps. This initial feature is associated with the formation and decay of the OAn-OPV•+-
PERY•–-OPV-OAn charge-separated and, thus, the OPV•+ and PERY•– radical ions account for the
absorption in this first regime. After 30 ps most of the absorption has decayed but a less intense long-
lived signal remains, which constitutes the second regime. Because the 1450 nm absorption of the
OAn-OPV•+-PERY•–-OPV-OAn state has disappeared after 30 ps, the remaining signal at 700 nm
corresponds to the OAn•+-OPV-PERY•–-OPV-OAn charge separated state. In the first picosecond
after photoexcitation, the differential transmission at 900 nm features the formation and decay of the
singlet excited state of the OPV and PERY chromophores (depending on which has been excited). As
in toluene, this initial singlet excited state is immediately quenched by the formation of the OAn-
OPV•+-PERY•–-OPV-OAn charge-separated state. After 20 ps an almost constant signal remains that
corresponds to the OAn•+-OPV-PERY•–-OPV-OAn charge separated state.
Photoinduced multistep electron transfer in multichromophoric arrays
67
0 200 400 600 800 1000
-1
0
∆T (
a. u
.)∆T
(a.
u.)
∆T (
a. u
.)
Time (ps)
0 20 40 60 80 100
-1
0
Time (ps)
0 2 4 6 8
-1
0
Time (ps)
Figure 3.10. Differential transmission dynamics of pentad 1 in THF monitored at 1450 (open circles),
900 (solid squares) and 700 nm (open triangles) with excitation at 455 nm, measured on different time
scales.
Even though the rate for charge shift cannot be extracted from the data, an efficiency of 22%
for the process in THF can be estimated from the 700 nm transient absorption, using the molar
absorption coefficients of the different radical ions at this wavelength and assuming that the
maximum intensity at 3 ps corresponds to the OAn-OPV•+-PERY•– charge separated state, while at 50
ps the remaining absorption corresponds to the OAn•+-OPV-PERY•– charged state only.
Chapter 3
68
In chlorobenzene and o-dichlorobenzene, solvents of lower and higher polarity than THF, the
differential transmissions at 1450, 900 and 700 nm of pentad 1 exhibit similar time profiles as
observed in THF. Also the efficiencies for charge shift are similar in all the polar solvents.
0 200 400 600 800 1000
-1
0
∆T (
a. u
.)
Time (ps)
0 100 200 300 400
-1
0
∆T (
a. u
.)
Time (ps)
0 2 4 6 8 10 12
-1
0
∆T (
a. u
.)
Time (ps)
Figure 3.11. Differential transmission dynamics of pentad 2 in THF monitored at 1450 (open circles),
900 (solid squares) and 700 nm (open triangles) with excitation at 455 nm, measured on different time
scales.
Photoinduced multistep electron transfer in multichromophoric arrays
69
3.4.3.2. Pentad 2 in THF
The differential transmission at 1450 nm of pentad 2 dissolved in THF (Figure 3.11) reveals
that the charge separation process to the OAn-OPV•+-PERY•–-OPV-OAn charge-separated state is
slowed down to kCS1 = 285 ns-1, compared to 1000 ns–1 for the same process in pentad 1. The reduced
rate of charge formation is consistent with the lower photoluminescence quenching of the PERY
chromophore in pentad 2 compared to 1. Moreover, the lifetime of the charge-separated state is
influenced by the longer and non-conjugated connection between donor and acceptor. As can be
expected, the rate for charge recombination has decreased resulting in a longer lifetime of the primary
charge-separated state (OAn-OPV•+-PERY•–-OPV-OAn) in pentad 2 (100 ps) than in pentad 1 (12 ps).
The efficiency of the charge shift process to generate the OAn•+-OPV-PERY•–-OPV-OAn charge-
separated state has improved to 28%, judging from the 700 nm transient absorption.
3.5. Kinetic considerations
Marcus theory provides an estimate for the free energy barrier (∆G‡) for electron transfer
reactions based on the change in free energy (∆G0) and the reorganization energy (λ) via:
( )
λλ
4
20‡ +∆=∆ G
G (3.2)
The reorganization energy consists of an internal contribution (λ i) and a solvent term (λs),
which can be approximated via the Born-Hush approach to give after summation, with n the index of
refraction:
−
−
++=+= −+
s2
0
2
isi11111
2
1
4 επελλλλ
nRrr
e
cc
(3.3)
The rate constants for the different processes, are not only a function of the energy barrier
∆G‡, but also of the reorganization energy (λ) and the electronic coupling (V) between donor and
acceptor in the excited state according to:
( )
+∆−
=
Tk
GV
Tkhk
B
202
21
B2
2
4exp
4
λλ
λπ
(3.4)
Chapter 3
70
The values of ∆G0, λ, and ∆G‡ calculated on the basis of Eqs. 3.1 to 3.4 are collected in Table
3.2. For the initial charge separation (kCS1) the values of free energy variation and reorganization
energies are comparable (–∆G0 ~ λ), close to the Marcus optimal region. In such cases, the reaction
rate is governed mainly by the electronic coupling V between the donor and acceptor and the polarity
of the solvent becomes less important. The electronic coupling depends on the nature of the spacer
and on the separation of donor and acceptor via ))(exp()( 0cc02
02 RRRVV −−= β , with R0 the
contact distance. The electronic coupling in pentad 1 is probably high due to the semi-conjugated
fashion by which the OPV and the PERY chromophores are linked. This explains the extraordinarily
high rate constants for the process observed for pentad 1 in all solvents (kCS > 1000 ns-1). Charge
recombination in OAn-OPV•+-PERY•–-OPV-OAn is in the Marcus inverted region (–∆G0 > λ) for all
solvents (Table 3.2). The use of Eq. 3.4 in the inverted region often underestimates the true rate
constant because of nuclear tunneling,28 but will be used here for qualitative comparison. In general
the barrier for charge recombination is reduced upon increasing the polarity of the medium. In
toluene, the most apolar solvent, the energy barrier is remarkably high, due to the large difference
between the free and reorganization energies. This is in agreement with the rather long-lived (366 ps)
OAn-OPV•+-PERY•–-OPV-OAn charge-separated stated observed for pentad 1 in toluene. In the more
polar solvents, the energy barrier is up to two orders of magnitude less than in toluene and, hence
charge recombination is expected to be much faster than in toluene. This explains, to some extend, the
12 ps lifetime of the OAn-OPV•+-PERY•–-OPV-OAn charge-separated state in THF. The charge shift
occurs in the normal region (–∆G0 < λ) and is energetically less favorable than the recombination
process (Table 3.2). In addition, the energy barrier of this process is higher than that of charge
recombination. These two facts imply that the charge shift is probably a slower process than the
charge recombination, although the influence of V is not accounted for. The study on the analogous
OAn-OPV-C60 molecular triad reported in chapter 2, has shown that increasing the lifetime of the
OAn-OPV•+-PERY•–-OPV-OAn charge-separated state is beneficial for the slower charge shift
process to compete. In principle this condition is nicely fulfilled for pentad 1, in toluene solution.
However, the driving force for the charge shift is close to zero in toluene and, hence, the charge shift
does not occur. However, in the polar solvents, where charge recombination is much faster, the charge
shift can be observed. The efficiency of the charge shift is similar in all polar solvents, consistent with
the small variation in energies reported in Table 3.2.
For pentad 2 the rate for charge separation (kCS) is less than for pentad 1, even though the
energy values depicted in Table 3.2 predict a similarly low barrier. Whereas in pentad 1, the electron
can be transferred from the OPV donor, through a ‘conjugated’ pathway to the PERY acceptor, the
electron transfer in pentad 2 occurs through space or through the saturated spacer by superexchange
mechanism. Moreover, the Rcc distance is higher in pentad 2. Both effects result in a lower electronic
Photoinduced multistep electron transfer in multichromophoric arrays
71
coupling in the excited state between the OPV donor and the PERY acceptor in pentad 2, and thus a
slower charge separation process. The same explanation holds for the charge recombination process
and explains the longer lifetime of the OAn-OPV•+-PERY•–-OPV-OAn state of pentad 2 in THF
solution (100 ps) compared to that of pentad 1 (12 ps). As a result of the slower charge recombination
to the ground state a slightly higher efficiency for charge shift is found for pentad 2 than for pentad 1.
The electronic coupling between the OAn and the PERY moieties is very weak.
Consequently, the charge recombination process is very slow and a long lifetime is observed for the
OAn•+-OPV-PERY•–-OPV-OAn charge-separated state.
3.6 Conclusions
Two multichromophoric arrays have been synthesized by covalently linking OAn, OPV and
PERY chromophores in an OAn-OPV-PERY-OPV-OAn symmetrical arrangement. In pentad 2 all
chromophores are decoupled in the ground state while in pentad 1 some π-overlap is possible between
the OPV and the PERY chromophores. Because of its urethane functionality, pentad 2 dimerizes or
aggregates in a variety of solvents (except THF) due to intermolecular hydrogen bonding.
Upon illumination of the pentad a number of sequential photophysical events occur. These
processes are mainly multistep energy and electron transfers. Even though no unambiguous evidence
has been established for an ultrafast multistep singlet-energy transfer process (ET1 & ET2, Figure
3.4) resulting in the OAn-OPV-1PERY*-OPV-OAn state prior to the charge transfer and irrespective
of the excitation wavelength, some facts seem to indicate that such sequential event indeed occurs.
First, the excitation spectrum of the residual PERY emission in pentad 2, superimposes with the
absorption spectrum of the pentad (Figure 3.6, left). Second, the rate for intramolecular charge
separation that generates the OAn-OPV•+-PERY•–-OPV-OAn state is independent of the chromophore
that has been excited, strongly suggesting that they originate from the same state.
The sequential electron transfer starts by the formation of OAn-OPV•+-PERY•–-OPV-OAn.
This state can relax to the ground state via charge recombination or generate via a charge shift the
long-lived OAn•+-OPV-PERY•–-OPV-OAn charge-separated state (CR1 and CSH1, Figure 3.4). In
pentad 1, intramolecular charge separation to generate the OAn-OPV•+-PERY• –-OPV-OAn state is
extremely fast (kCS > 1000 ns-1), regardless the polarity of the solvent, because it occurs under optimal
conditions (–∆G0 ~ λ). The charge recombination of OAn-OPV•+-PERY•–-OPV-OAn to the ground
state occurs in the Marcus inverted region (–∆G0 > λ) and is therefore very fast in the polar solvents
and remarkably slow in toluene. The charge separation and recombination are slowed down in pentad
2, because the saturated spacer reduces the electronic coupling between the OPV donor and PERY
acceptor. In pentad 1, the charge shift process that generates OAn•+-OPV-PERY•–-OPV-OAn
competes with the faster charge recombination with an efficiency of 0.22. The slower charge
Chapter 3
72
recombination of OAn-OPV•+-PERY•–-OPV-OAn to the ground state in pentad 2, with respect to that
in pentad 1, is beneficial for the charge shift process that occurs with an efficiency of 0.28 in polar
media. Thus placing the OPV and PERY chromophores at longer distances seems to be the key to
improve the quantum yield for charge shift. As a matter of fact, the steady state PIA experiment on
the OAn-OPV and PERY mixture is an extreme situation of such a large distance, resulting in a high
efficiency because for every OAn-OPV•+/PERY•– the OAn•+-OPV/PERY•– is formed.
Pentads 1 and 2 show a few differences with respect to the OAn-OPV-C60 molecular triad
presented in chapter 2. In contrast to the fullerene-based triad, charge separation is possible in apolar
media for the pentads owing to the lower reduction potential of the perylene diimid acceptor.
Furthermore, in the pentads the various photophysical processes are less sensitive to the solvent
polarity (within the polar solvents) than in the OAn-OPV-C60 triad. This means that even though
similar trends are followed (for example, faster charge recombination in polar solvents), these changes
are almost negligible in the pentads. The most striking difference between the two multichromophoric
arrays originates from the fact that the charge shift process to the OAn•+-OPV-PERY•–-OPV-OAn is
able to compete with the charge recombination to the ground state in the pentads, even though the
latter process seems to be even much faster than in the OAn-OPV-C60 triad (Chapter 2).
In this chapter, the photophysical phenomena following photoexcitation have been
investigated in pentads 1 and 2 in an isolated form (molecularly dissolved). These molecules have the
potential to form ordered molecular aggregates, because of the strong and directing π−π interactions
of the perylene unit and additionally for pentad 2 the possibility of hydrogen bonding via the urethane
linkage. Future studies on the aggregated and solid states of the pentads might reveal promising bulk
photophysical properties.
3.7 Experimental section
For general methods and materials the reader is referred to chapter 2 of this thesis. Diethyl(4-nitrobenzyl) phosphonate (4). Triethyl phosphite (2.30 g, 13.88 mmol) and 4-nitrobenzyl bromide (2g, 9.25 mmol) were heated to 160 °C and stirred for 2 h. Subsequently, the mixture was cooled to 70 °C and the formed ethyl bromide and the excess of triethyl phosphite were distilled under reduced pressure. The residue was dissolved in ethyl acetate and filtrated over silica gel. The solvent was removed in vacuo to yield 2.2 g of diethyl 4-nitrobenzyl phosphonate. 1H NMR (CDCl3, 300 MHz): δ 8.18 (d, 2H), 7.5 (d, 2H), 4.07 (m, 4H), 3.27 (d, 2H), 1.28 (t, 6H); 13C NMR (CDCl3, 75 MHz): δ 146.78, 139.62 (d), 130.45(d), 123.44 (d), 62.20 (d), 34.61, 32.79, 16.15 (d). (E,E,E,E)-4-<4-[4-{4-[N-(4-Diphenylaminophenyl)-N-(phenyl)-3-aminostyryl]-2,5-bis[(S)-2-methylbutoxy]styryl]-2,5-bis[(S)-2-methylbutoxy]styryl}-2,5-bis[(S)-2-methylbutoxy]styryl>nitrobenzene (5).
Photoinduced multistep electron transfer in multichromophoric arrays
73
Diethyl(4-nitrobenzyl) phosphonate (73 mg, 0.27 mmol), was dissolved in anhydrous DMF (6 mL) under an argon atmosphere and KtBuO (35 mg, 0.32 mmol) was added to the solution at room temperature. After 15 minutes a solution of aldehyde compound 3 (307 mg, 0.243 mmol) in 3 mL DMF/THF (2/1) was added dropwise to the reaction mixture. Twenty minutes after the addition was completed, the product starts precipitating from solution. The reaction mixture was stirred for 2 h more. The product was isolated by filtration and washed with ethanol and methanol as a red powder (0.290 g, 86%). 1H NMR (CDCl3, 300 MHz): δ 8.22 (d, 2H), 7.65 (d, 1H), 7.63(d, 2H), 7.57 (d, 1H), 7.52 (s, 2H), 7.51 (d, 1H), 7.40 (d, 1H), 7.31-6.96 (m, 31H), 4.05-3.72 (m, 12H), 2.10-1.85(m, 6H), 1.70-1.45 (m, 6H ), 1.45-1.2 (m, 6H), 1.2-0.85 (m, 36H); 13C NMR (CDCl3, 75 MHz): δ 151.81, 151.25, 151.01, 148.21, 147.84, 146.43, 144.73, 142.90, 142.70, 139.15, 129.45, 129.18, 128.42, 128.31, 127.75, 127.59, 127.07, 126.63, 126.04, 125.45, 125.30, 125.11, 124.18, 123.78, 122.96, 122.79, 122.68, 122.42, 122.24, 120.29, 111.15, 110.67, 110.01, 109.89, 109.70, 109.53, 74.47, 74.25, 74.16, 74.04, 73.94, 35.13, 35.05, 34.94, 26.37, 16.85, 11.52, 11.38. MALDI-TOF MS (Mw =1382.87) m/z = 1382.0 [M]+; Anal. Calc for C92H107N3O8: C, 79.9; H, 7.8; N, 3.0. Found: C, 79.7; H, 7.8; N, 2.7. (E,E,E,E)-4-<4-[4-{4-[N-(4-Diphenylaminophenyl)-N-(phenyl)-3-aminostyryl]-2,5-bis[(S)-2-methylbutoxy]styryl]-2,5-bis[(S)-2-methylbutoxy]styryl}-2,5-bis[(S)-2-methylbutoxy]styryl>aniline (6). Under an Ar atmosphere nitro compound 5 (0.28 g, 0.2 mmol) was suspended in EtOAc (15 mL). SnCl2.H2O (0.36 g, 1.62 mmol) and EtOH (2 mL) were added and the mixture was heated to 85 °C. The reaction mixture was stirred for 5 h at 95 °C and subsequently cooled to room temperature. After cooling to room temperature the reaction mixture was poured into crushed ice. The aqueous phase was slightly basified by the addition of NaOH 0.1 M and was subsequently extracted three times with diethyl ether. The collected organic fractions were dried over MgSO4, and the solvent removed in vacuo. Purification by column chromatography (silica gel, pentane/CH2Cl2 1:1, Rf = 0.4), yielded 6 (0.19 g, 70%) as an orange solid. 1H NMR (CDCl3, 400 MHz): δ 7.52 (s, 4H), 7.42 (d, 1H), 7.36 (d, 2H), 7.31 (d, 1H), 7.27-6.97 (m, 31H), 6.69 (d, 2H), 3.93-3.89 (m, 12H), 1.70-1.45 (m, 6H), 1.45-1.2 (m, 6H), 1.2-0.85 (m, 36H); 13C NMR (CDCl3, 75 MHz): δ 151.23, 151.08, 151.02, 150.95, 148.17, 147.81, 145.94, 142.86, 142.69, 139.15, 129.43, 129.17, 128.77, 128.67, 128.30, 127.72, 127.53, 127.36, 127.21, 126.79, 126.43, 125.44, 125.30, 123.77, 123.71, 123.59, 122.68, 122.41, 122.25, 120.26, 119.88, 115.22, 110.62, 110.41, 109.85, 109.65, 74.44, 74.28, 74.06, 73.99, 35.13, 35.04, 34.98, 34.92, 26.37, 26.29, 16.89, 16.87, 16.82, 16.79, 11.54, 11.48, 11.41, 11.37. MALDI-TOF MS (Mw =1352.89) m/z = 1352.92 [M]+.
N,N’-Bis[(E,E,E,E)-4-<4-[4-{4-[N-(4-Diphenylaminophenyl)-N-(phenyl)-3-aminostyryl]-2,5-bis[(S)-2-methylbutoxy]styryl]-2,5-bis[(S)-2-methylbutoxy]styryl}-2,5-bis[(S)-2-methylbutoxy]styryl>phenyl]-3,4,9,10-perylene bis(dicarboximide) (1). Amine compound 6 (55 mg, 0.04 mmol), 3,4,9,10-perylenetetracarboxylic dianhydride (8 mg, 0.02 mmol), imidazole (0.5g, 7.3 mmol), and a catalytic amount of Zn(AcO)2 were mixed and heated till 160 °C. After stirring for 4 h, the reaction mixture was cooled to room temperature. The residue was purified by column chromatography (silica gel, ethyl acetate/ CHCl3 1:0 to 1:4, Rf = 0 to 0.4) and repetitive preparative size exclusion chromatography (Bio Beads S-X1 and S-X3, CH2Cl2) to yield 9 mg (15 %) of 1 as a dark red solid. 1H NMR (CD2Cl2, 400 MHz): δ 8.61-8.52 (m, 8H), 7.61 (d, 4H), 7.51-6.89 (m, 77H), 3.90-3.70 (m, 24H), 1.94-1.82 (m, 12H), 1.65-1.52 (m, 12H), 1.36-1.25 (m, 12H), 1.07-0.90 (m, 64H). MALDI-TOF MS (Mw =3062.08) m/z = 3060.64 [M]+. Diethyl benzylphosphonate (7). Triethyl phosphite (1.45 g, 8.77 mmol) and benzyl bromide (1 g, 5.84 mmol) were heated to 160 °C and stirred for 3 h. Subsequently, the mixture was cooled to 90 °C and the formed ethyl bromide and the excess of triethyl phosphite were distilled under reduced pressure. The residue was dissolved in ethyl acetate and filtrated over silica gel. The solvent was removed in vacuo to yield 1.05 g (79 %)of diethyl benzylphosphonate. 1H NMR (CDCl3, 300 MHz): δ 7.34 (m, 5H), 4.06-3.95 (m, 4H), 3.15 (dd, 2H), 1.23 (td, 6H); 13C NMR (CDCl3, 75 MHz): δ 131.55 (d), 129.70 (d), 128.45 (d), 126.78 (d), 62.02 (d), 34.63, 32.80, 16.29 (d). (E,E,E,E)-4-<4-[4-{4-[N-(4-Diphenylaminophenyl)-N-(phenyl)-3-aminostyryl]-2,5-bis[(S)-2-methylbutoxy]styryl]-2,5-bis[(S)-2-methylbutoxy]styryl}-2,5-bis[(S)-2-methylbutoxy]styryl>benzene (8). Phosphonate 7 (17 mg, 0.074 mmol) was dissolved in anhydrous DMF (1 mL) under an argon atmosphere and KtBuO (18 mg, 0.016 mmol) was added to the solution at room temperature. After 15 minutes, a solution of aldehyde compound 3 (54 mg, 0.042 mmol) in 2.2 mL DMF/THF (2/1) was added dropwise to the reaction mixture. The reaction mixture was stirred for 2.5 h. Evaporation of the solvent and precipitation in methanol yielded 7 (42 mg, 73 %) as an orange solid. 1H NMR (CD2Cl2, 400 MHz): δ 7.56-6.98 (m, 42 H), 3.98-3.81 (m,
Chapter 3
74
12H), 2.04-1.92 (m, 6H), 1.73-1.59 (m, 6H), 1.43-1.32 (m, 6H), 1.14-0.96 (m, 36H); 13C NMR (CDCl3, 100 MHz): δ 151.65, 151.50, 151.41, 148.73, 148.27, 143.46, 143.21, 139.54, 138.39, 129.84, 129.55, 129.06, 128.93, 128.71, 127.86, 127.80, 127.65, 126.91, 126.77, 126.02, 125.81, 124.11, 123.78, 123.68, 123.07, 122.92, 122.83, 122.25, 120.61, 110.96, 110.78, 110.15, 109.93, 74.76, 74.56, 74.42, 35.56, 35.49, 35.43, 26.77, 17.00, 11.70, 11.56. MALDI-TOF MS (Mw =1337.90) m/z = 1337.74 [M]+. (E,E,E,E)-4-<4-[4-{4-[N-(4-Diphenylaminophenyl)-N-(phenyl)-3-aminostyryl]-2,5-bis[(S)-2-methylbutoxy]styryl]-2,5-bis[(S)-2-methylbutoxy]styryl}-2,5-bis[(S)-2-methylbutoxy]styryl>phenylisocyanate (9). Amine compound 6 (50 mg, 0.04 mmol) was suspended in 20% phosgene solution in toluene (2.4 mL) and stirred at 95 °C for 16 h under an inert atmosphere. The reaction mixture was cooled to room temperature and the solvent removed in vacuo. The complete conversion of the amine to the isocyanate was monitored by IR spectroscopy by observing the disappearance of the amine peak at 3364 cm-1 and the formation of the isocyanate peak at 2264 cm-1 Compound 9 was used without further purification. MALDI-TOF MS (Mw =1378.91) m/z = 1378.64 [M]+. N,N’-Di-((S)-1-isobutyl-2-hydroxyethyl)-3,4,9,10-perylene bis(dicarboximide) (10). (S)-(+)-Leucinol (0.455 g, 3.88 mmol), 3,4,9,10-perylenetetracarboxylic dianhydride (0.63 g, 1.61 mmol), imidazole (10 g), and catalytic amounts of Zn(OAc)2 were heated to 160 °C and stirred for 1.5 h. After cooling to room temperature the reaction mixture was dissolved in methylene chloride and washed 2 times with HCl 1M, brine, and dried over MgSO4. The solvent was removed in vacuo to yield 0.72g (75%) of a dark red solid. 1H NMR (CD2Cl2, 400 MHz): δ 8.40-7.52 (m, 8H), 5.54-5.44 (m, broad signal, 2H), 4.60-4.48 (m, broad signal, 2H), 3.98-3.94 (m, 2H), 2.10-2.01 (m, 2H), 1.80-1.46 (m, broad signal, 4H), 1.10-0.92 (m, 12). MALDI-TOF MS (Mw = 590.2) m/z = 590.1 [M]+. N,N’-Bis[(S)-1-Isobutyl-2-[(E,E,E,E)-4-<4-[4-{4-[N-(4-Diphenylaminophenyl)-N-(phenyl)-3-aminostyryl]-2,5-bis[(S)-2-methylbutoxy]styryl]-2,5-bis[(S)-2-methylbutoxy]styryl}-2,5-bis[(S)-2-methylbutoxy]styryl>phenyl carbamicacid]ethyl] 3,4,9,10-perylene bis(dicarboximide) (2). A solution of the isocyanate (38 mg, 0.027 mmol) in dry methylene chloride was added to a solution of 6 (7.4 mg, 0.012 mmol) in dry methylene chloride. Dibutylin dilaurate (3.22 mg) was added as a catalyst, and the reaction mixture was heated at reflux for 20 h under an argon atmosphere. The crude mixture was cooled to room temperature, the solvent was removed in vacuo. The residue was purified by silica gel chromatography (CH2Cl2/pentane; 4:1, Rf = 0.15) and preparative size exclusion chromatography (Bio Beads S-X3 , CH2Cl2) to yield 20 mg (47 %) of 7 as a red solid. 1H NMR (CD2Cl2, 400 MHz): δ 9.38 (br s, 2H), 8.57, 8.21, 8.12, 7.79 (4 x d, 8H), 7.58-7.06 (m, 82H), 5.94-5.78 (br m, 2H), 5.35-5.18 (br m, 2H), 4.76-4.57 (br. m, 2H), 3.98-3.83 (m, 24H), 2.37-2.22 (m, 2H), 2.02-1.89 (m, 14H), 1.75-1.60 (m, 16H), 1.52-1.32 (m, 12H), 1.16-0.99 (m, 84H); 13C NMR (THF, 100 MHz): δ 165.00 (broad signal), 154.83, 153.07, 152.95, 150.14, 149.76, 144.83, 144.73, 141.33, 140.59, 136.06, 134.32, 132.50, 131.21, 131.09, 130.82, 130.60, 129.89, 129.79, 129.36, 129.18, 129.04, 128.89, 128.65, 128.34, 128.26, 127.74, 127.09, 126.99, 126.79, 126.79, 125.45, 125.06, 124.49, 124.37, 124.22, 124.12, 123.97, 123.28, 123.09, 121.83, 119.90, 112.15, 112.02, 111.47, 111.20, 111.13, 80.28, 75.64, 75.46, 53.07, 40.05, 37.07, 36.97, 36.89, 28.11, 27.26, 24.44, 23.65, 18.08, 12.73, 12.61.MALDI-TOF MS (Mw =3348.49) m/z = 3347.85 [M]+. 3.8 References and notes 1 (a) Molecular Electronics, Jortner, J., Ratner, M. Eds.; Blackwell: London, 1997. (b) Lehn, J.-M.,
Supramolecular Chemistry; VCH, Weinheim, 1995. (c) Electron Transfer in Chemistry Vol. I-IV,
Balzani, V., Wiley-VCH, Weinheim, 2001.
2 For recent reviews see: (a) Gust, D.; Moore, T. A.; Moore, A. L. Acc. Chem. Res. 2001, 34, 40. (b) Guldi,
D. M. Chem. Soc. Rev. 2002, 31, 22.
Photoinduced multistep electron transfer in multichromophoric arrays
75
3 (a) Liddell, P. A.; Kuciauskas, D.; Sumida, J. P.; Nash, B.; Nguyen, D.; Moore, A. L.; Moore, T. A.;
Gust, D. J. Amer. Chem. Soc. 1997, 119, 1400. (b) Imahori, H.; Yamada, K.; Hasegawa, M.; Taniguchi,
S.; Okada, T.; Sakata, Y. Angew. Chem., Int. Ed. Engl. 1997, 36, 2626. (c) Carbonera, D.; Di Valentin,
M.; Corvaja, C.; Agostini, G.; Giacometti, G.; Liddell, P. A.; Kuciauskas, D.; Moore, A. L.; Moore, T.
A.; Gust, D. J. Amer. Chem. Soc. 1998, 120, 4398. (d) Kuciauskas, D.; Liddell, P. A.; Moore, A. L.;
Moore, T. A.; Gust, D. J. Amer. Chem. Soc. 1998, 120, 10880. (e) Tamaki, K.; Imahori, H.; Sakata, Y.;
squares) and absorption spectrum (solid line) of an ITO/PEDOT:PSS/3/Al device.
Photoinduced electron transfer of conjugated polymers with pendant fullerenes
117
5.3. Low-bandgap π-conjugated polymers with pendant fullerenes
With the solar spectrum peaking at 700 nm, a mismatch occurs between the absorption
spectrum of polymer 3 and the solar emission. In order to improve the absorption of photons to create
charges, a backbone with low bandgap properties would be beneficial. Low-bandgap polymers are
typically obtained by alternating electron rich and electron deficient conjugated units along the
polymer chain. Another aspect of polymer 3 that needs to be addressed is its solubility. A less rigid
backbone might aid in solubilizing the donor-acceptor polymer.
BO
O
BO
OS S
NS
N
RR
R R
OC6H13
OC6H13
OC6H13
O*
S S
NS
N* n
RR
R R
+ 1
16. R = OC8H17
17. R = OC8H17
Scheme 5.3. Synthesis of low-bandgap polymers with pendant fullerenes.
These issues are taken in consideration in polymer 17, prepared by copolymerization of
boronic ester 16 with diiodo functionalized fullerene 1 using the Suzuki polycondensation reaction
(Scheme 5.3). 15 The alternation between thiophenes and benzothiodiazoles in monomers 16 confers
the low-bandgap character to the corresponding polymer, together with a less rigid structure.
The optical bandgap of polymer 17 is 1.7 eV, lower than that of polymer 3. The polymer is
also highly soluble in most common solvents and, according to SEC analysis, of rather high molecular
weight (Mw = 25.2 kg/mol, PDI =2.3). IR and UV-Visible absorption spectroscopies exhibit the
typical features of the C60 fragments that are evidence of the incorporation of the fullerene molecules
in the polymer chains. However, several characterization techniques seem to indicate that the polymer
does not correspond with the proposed structure. The SEC-trace of polymers 17 shows a bimodal
distribution, which indicates that likely more than one reaction mechanism occurs in the
polymerization. The bimodality is not observed in a reference polymer without the fullerene moieties
that has been made by reaction of 16 with 4. The resolution of the 1H-NMR spectrum is extremely
low due to peak broadening and the characteristic multiplet at 3 ppm of the –CH2– closest to the
fullerene moiety is not distinguishable.
Chapter 5
118
Photoinduced absorption spectroscopy of 17 reveals that charge separation occurs after
photoexcitation of the backbone. The PIA spectrum of 17 shows two bands at 0.48 and 1.08 eV of the
backbone radical cation with a residual triplet signal (Figure 5.7, left). The signal of the C60 radical
anion at 1.24 eV cannot be distinguished because of the radical cation absorptions present in the same
spectral region. As expected, the PIA spectrum of the corresponding reference polymer exhibits a
single photoinduced absorption attributed to the formation of the triplet state.
0.5 1.0 1.5 2.0-10
-5
0
5
10
15
Nor
mal
ized
-∆T
/T
Energy (eV)
-2 -1 0 1 21E-5
1E-3
0.1
10
I (m
A/c
m2 )
Bias (V)
Figure 5.7. Left: Photoinduced absorption spectra of polymer 17 (solid line) and its corresponding
reference polymer (dashed line). Right: I/V curves of an ITO/PEDOT:PSS/17/Al device in the dark
(dashed line) and under white-light illumination (solid line).
Following a procedure similar as for polymer 3, a photovoltaic device of 17 has been prepared
spincoating from a toluene solution. The semilogaritmic plot of the I/V characteristics in the dark and
under white-light illumination of the device are shown in Figure 5.7 (right). The I/V characteristics of
the cell show a diode behavior with a rectification ratio (RR) of 54.7 at ± 2 V. This rather low value is
related to low film thickness (~ 40 nm). Under white-light illumination (100 mW/cm2, AM1.5) an
open-circuit voltage of Voc= 0.75 V and a short circuit current of Isc = 0.31 mA/cm2 are obtained. The
fill-factor (FF) is 0.23. This low fill-factor can be attributed to a small parallel resistance caused by
the low film thickness.
5.4. Conclusions
Two novel π-conjugated polymers with pendant methanofullerenes have been synthesized
using palladium catalyzed cross-coupling reactions.
The first one, a hybrid polymer between p-phenylene vinylenes and p-phenylene ethynylenes
bearing fullerenes, 3, has been synthesized using the Sonogashira coupling between dihalogenated
and bisethynyl terminated aryl compounds 1 and 2. SEC reveals that polymer 3 has a lower molecular
Photoinduced electron transfer of conjugated polymers with pendant fullerenes
119
weight than its reference polymer 5. Higher molecular weights result in insolubility of the polymer
chains. Photoluminescence and PIA spectroscopies reveal that photoexcitation of 3 results in a
electron transfer from the polymer backbone to the pendant fullerenes. A device with the structure
ITO/PEDOT:PSS/3/Al has been prepared by spincoating from a chloroform solution. The values of Isc
and Voc have improved compared to previously reported π-conjugated polymer/fullerene solar cells.16
Moreover, this performance is given by a polymer with a fullerene loading of 31 wt %, a value
considerably smaller than that used commonly in ‘bulk-heterojunction’ solar cells (about 75%). The
performance of the device of 3 could be improved by increasing the solubility of the polymer, which
would enable the preparation of thicker active layers. Also, there is a mismatch between the solar
spectrum and the absorption spectrum of 3, meaning that the solar spectrum is not used in an optimal
way to generate photoinduced charges.
In an attempt to increase the processability and the absorption of light of polymer 3, a
covalently linked donor-acceptor polymer with low-bangap characteristics, 17, has been synthesized
using the Suzuki polycondensation. Characterization of these polymers reveals that although the
polymers contain fullerene molecules, the polymer does not correspond to a perfect alternation of the
comonomers. This is attributed to a significant amount of side reactions taking place, apart from the
Suzuki cross-coupling reactions. Although this polymer shows some interesting photophysical and
photovoltaic properties, the Suzuki polycondensation is not an ideal alternative to improve this type of
polymers.
Overall, the results indicate that a bicontinuous network of donor and acceptors, confined to a
molecular scale, is an attractive approach to new materials for photovoltaic applications, although
much has to be improved to bring the performance of these novel polymers to that of the best bulk-
heterojunction cells.
5.5. Experimental section
For general methods and materials the reader is referred to chapter 2 of this thesis. 1,4-Bis(2-ethylhexyloxy)-2-methylbenzene (7). Under an argon atmosphere methylhydroquinone (10 g, 80.5 mmol), 2-ethylhexyl-p-toluenesulfonate (48 g, 169 mmol) and tetrabutylammonium chloride (2.66 g, 9.6 mmol) were added to a suspension of K2CO3 (66.3 g, 480 mmol) in dry 2-butanone (160 mL). The reaction mixture was stirred for 16 h at reflux temperature. After cooling, the suspension was filtered and the solvent was removed in vacuo. The resulting crude product was purified by column chromatography (silica gel, hexane/CHCl3 2:1). Evaporation of the solvent yielded 28.3 g (93%) of 7 as a pure colorless oil: 1H NMR (CDCl3) δ 6.79 (d, 1H), 6.77 (d, 1H), 6.71 (dd, 1H), 3.83 (m, 4H), 2.25 (s, 3H), 1.7 (m, 2H), 1.48 (m, 16H), 0.98 (m, 12H); 13C NMR (CDCl3) δ 152.95, 151.50, 127.99, 117.62, 111.66, 111.35, 70.94, 70.79, 39.62, 39.45, 30.68, 30.52, 29.11, 29.07, 24.05, 23.83, 23.08, 16.38, 14.09, 11.21, 11.09; GC-MS (Mw = 348.56) m/z = 348 [M]+. 1-Bromo-2,5-bis(2-ethylhexyloxy)-4-bromomethylbenzene (8). Under an argon atmosphere, NBS (6.12 g, 34.4 mmol) and AIBN (1.72 g, 10.3 mmol) were added to a solution of 7 (10 g, 28 mmol) in dry CCl4 (28 mL). After stirring for 1h under reflux, the reaction mixture was cooled to
Chapter 5
120
room temperature. The mixture was filtered and the solvent evaporated. To remove the last traces of AIBN and NBS, hexane was added to the residue, followed by filtration and evaporation of the solvent. Subsequently, dry THF (28 mL) and NBS (6.63 g, 37 mmol) were added and the reaction mixture was stirred at reflux temperature for 1h. After evaporation of the solvent, hexane was added. The solution was filtered and the solvent removed in vacuo. After column chromatography (silica gel, hexane) and evaporation of the solvents, 3.5 g of 8 (24%) was obtained as a colorless oil. 1H NMR (CDCl3) δ 7.06 (s, 1H), 6.80 (s, 1H), 4.48 (s, 2H), 3.84 (m, 4H), 1.76 (m, 2H), 1.4 (m, 16H), 0.92 (m, 12H); 13C NMR (CDCl3) δ 151.20, 149.55, 125.87, 116.98, 115.61, 113.15, 72.35, 71.04, 39.50, 30.57, 30.46, 29.05, 23.97, 23.86, 23.01, 14.05, 11.16. Diethyl[2,5-bis(2-ethylhexyloxy)-4-bromo-benzyl]phosphonate (9). Triethyl phosphite (1.55 g, 9.33 mmol) and 8 (3.15 g, 6.22 mmol) were stirred at 160 oC for 1.5 h. The reaction mixture was cooled to 75 oC and the ethyl bromide, formed during the reaction, and the excess of triethyl phosphite were distilled under reduced pressure. The product 9 was a light yellow oil. Yield 3.30 g (100%). 1H NMR (CDCl3) δ 7.02 (d, 1H), 6.86 (d, 1H), 4.03 (m, 4H), 3.79 (m, 4H), 3.16 (d, 2H), 1.70 (m, 2H), 1.36 (m, 22H), 0.90 (m, 12H); 13C NMR (CDCl3) δ 150.94 (d), 149.34 (d), 120.08 (d), 116.55 (d), 116.25 (d), 110.50 (d), 72.17, 71.27, 61.76 (d), 39.39 (d), 30.38 (d), 28.93 (d), 23.74 (d), 22.88, 16.22 (d), 13.91, 11.01. 1,4-Dibromo-2-(3,7-dimethyloctyloxy)-5-methoxybenzene (12). A solution of Br2 (11.78 g, 73.74 mmol) in glacial acetic acid (30 mL) was added dropwise to a solution of 1-(3,7-methyloctoxy)-4-methoxybenzene (11) (10 g, 37.82 mmol) in glacial acetic acid (45 mL) at 65 oC. After stirring during 45 min at 65 oC, the temperature was raised to reflux during 1h. The solution was then cooled to room temperature and subsequently poured on water (560 mL) and made alkaline with 2N NaOH (660 mL). The aqueous phase was extracted with CH2Cl2 (3 x 200 mL). The combined organic layers were washed with brine and dried over MgSO4. The resultant crude product was purified by column chromatography (silica gel, hexane/CHCl3 2:1) Evaporation of the solvent yielded 12 g (75%) of 12 as a colorless oil: 1H NMR (CDCl3) δ 7.09 (s, 1 2H), 1.52 (m, 1H), 1.32 (m, 2H), 1.68 (m, 4H), 0.94 (d, 3H), 0.87 (d, 6H); 13C NMR (CDCl3) δ 150.34, 150.09, 118.38, 116.91, 111.11, 110.13, 68.57, 56.90, 39.13, 37.15, 35.97, 29.70, 27.92, 24.61, 22.66, 22.56, 19.64; GC-MS (Mw = 422.20) m/z= 422 [M]+. 2-(3,7-Dimethyloctyloxy)-5-methoxybezene-1,4-dialdehyde (13). Dibromide 12 (7 g, 16.58 mmol) was dissolved in dry diethyl ether (135 mL). The solution was cooled to –10 oC and 1.6 M n-buthyllithium hexane solution (24.87 mL) was added slowly. The reaction mixture was stirred for 5 min., then, the cooling bath was removed and dry DMF (3.23 mL) was added dropwise. The mixture was stirred for another hour at room temperature. After addition of 6 M HCl (30 mL), the organic layer was washed with water (2 x 100 ml) a saturated NaHCO3 solution (100 mL) and again water (100 mL). The organic layer was dried over MgSO4 and the solvent was evaporated. The resultant crude product was purified by column chromatography (silica gel, hexane/toluene 2:1, Rf = 0.2). Evaporation of the solvent yielded 3.02 g (56%) of 13 as a light yellow solid. 1H NMR (CDCl3) δ10.54 (s, 1H), 10.49 (s, 1H), 7.44 (s, 2H), 4.13 (m, 2H), 3.94 (s, 3H), 1.88 (m, 1H), 1.66 (m, 2H), 1.53 (m, 1H), 1.25 (m, 6H), 0.96 (d, 3H), 0.87 (d, 6H); 13C NMR (CDCl3) δ189.20, 155.47, 155.28, 129.08, 111.70, 110.58, 67.53, 56.10, 39.11, 37.13, 35.89, 29.83, 27.88, 24.60, 22.61, 22.51, 19.57. (E,E)-1,4-Bis[4-bromo-2,5-bis(2-ethylhexyloxy)styryl]-2-(3,7-dimethyloctyloxy)-5-methoxybezene (14). Phosphonate 9 (1.82 g, 3.23 mmol) was dissolved in dry DMF (10 mL) under an argon atmosphere and 0.43 g (3.9 mmol) of KtBuO were added to the solution. After 15 min, a solution of dialdehyde 13 (0.5 g, 1.56 mmol) in dry DMF (12 mL) was added dropwise and the reaction mixture was stirred for 3 h. The solution was poured on crushed ice and of 6 M HCl (200 mL) was added. The aqueous phase was extracted twice with diethyl ether and the combined organic layers were subsequently washed with 3 M HCl, water, a saturated aqueous solution of Na2CO3 and dried over MgSO4. The solvent was removed in vauo. Column chromatography (silica gel, hexane/CHCl3 4:1, Rf = 0.6) and posterior evaporation of the solvent afforded 0.7g (35%) of 14 as a greenish oil: 1H NMR (CDCl3) δ 7.5 (d, 1H), 7.49 (d, 1H), 7.45(d, 1H), 7.44 (d, 1H), 7.18 (s, 2H), 7.17 (s, 2H), 7.10 (s, 1H), 7.10 (s, 1H), 4.10(t, 2H), 3.92 (m, 11H), 1.9-0.8 (m, 79H); 13C NMR (CDCl3) δ 151.37, 151.08, 150.98, 150.90, 149.92, 149.88, 127.10, 127.02, 126.90, 126.84, 123.41, 123.35, 123.22, 122.63, 117.55, 117.41, 111.59, 111.55, 111.08, 110.72, 110.35, 108.45, 72.34, 72.25, 71.79, 71.50, 67.17, 55.91, 39.61, 39.53, 39.21, 37.29, 36.46, 30.72, 30.73, 30.50, 30.21, 29.10, 29.06, 27.92, 24.70, 24.15, 23.89, 23.87, 23.05, 22.63, 22.54, 19.88, 14.11, 14.08, 14.06, 11.32, 11.26, 11.23, 11.20; MALDI-TOF MS (Mw = 1139.32) m/z = 1138.59 [M]+.
Photoinduced electron transfer of conjugated polymers with pendant fullerenes
121
(E,E)-1,4-Bis[4-trimethylsilylethynyl-2,5-bis(2-ethylhexyloxy)styryl]-2-(3,7-dimethyloctyloxy)-5-methoxybezene (15). (Trimethylsilyl)acetylene (0.114 g, 1.17 mmol) and dibromide 14 (0.5 g, 0.39 mmol) were dissolved in anhydrous triethylamine (8 mL). Argon was purged trough the solution for 15 min and the temperature was raised to 80 oC. Then PdCl2 (6.98 mg, 0.04 mmol), triphenylphosphine (31 mg, 0.11 mmol) and copper(II) acetate (7.8 mg, 0.04 mmol) were added to the solution. The reaction mixture was stirred for 16 h. The solvent was removed in vacuo, the crude solid was dissolved in diethyl ether. The organic layer was washed with a saturated aqueous solution of NH4Cl and brine. After the organic layer was dried over MgSO4, the solvent was removed in vacuo. The residue was purified by column chromatography (silica gel, hexane/CHCl3 85:15, Rf = 0.4) and recrystalization from ethanol yielded 230 mg (50%) of 15 as yellow crystals: 1H NMR (CDCl3) δ7.48 (s, 2H), 7.47 (s, 2H), 7.15 (s, 2H), 7.11 (s, 1H), 7.10 (s, 1H), 6.94 (s, 2H), 4.07 (t, 2H), 3.90 (m, 11H), 1.89 (m, 1H), 1.77 (m, 5H), 1.51 (m, 16), 1.34 (m, 20H), 1.14 (m, 4H), 0.92 (m, 33H), 0.26 (s, 18H); 13C NMR (CDCl3) δ 154.90, 151.49, 151.01,150.40, 150.29, 128.72, 128.58, 127.26, 127.06, 123.94, 123.84, 123.48, 122.83, 117.07, 116.93, 112.17, 110.47, 110.05, 109,62, 108.47, 101.84, 98,86, 71.81, 71.70, 71.64, 71.35, 67.80, 55.94, 39.73, 39.94, 39.63, 39.24, 37.30, 36.47, 30.79, 30.76, 30.54, 30.24, 29.19, 29.15, 27.93, 24.69, 24.21, 23.92, 23.91, 23.09, 23.07, 22.64, 22.55, 19.90, 14.12, 14.08, 14.06, 11.35, 11.32, 11.28, 0.02; MALDI-TOF MS (Mw =1173.94) m/z = 1173.84 [M]+; Anal. Cald for C75H120O6Si2: C, 76.7; H, 10.3. Found: C, 76.64; H, 10.33. (E,E)-1,4-bis[4-ethynyl-2,5-bis(2-ethylhexyloxy)styryl]-2-(3,7-dimethyloctyloxy)-5-methoxybezene (2). To a solution of 15 (27 mg, 0.02 mmol) in dry THF was added 1 M tetrabuthylammonium fluoride in THF (0.023 mL). The reaction mixture was stirred for 1 min and subsequently filtrated over silica gel using chloroform as eluent. The solvent was removed in vacuo yielding 22.6 mg (100%) of a yellow solid which was used without further purification: 1H NMR (CDCl3) δ7.49 (s, 1H), 7.49 (s, 1H), 7.48 (s, 2H), 7.15 (s, 2H), 7.14 (s, 1H), 7.13 (s, 1H), 6.97 (s, 1H), 6.97 (s, 1H), 4.08 (t, 2H), 3.90 (m, 11H), 3.29 (s, 2H), 0.91 (m, 33H); 13C NMR (CDCl3) δ 154.87, 151.46, 150.99, 150.39, 150.28, 128.97, 128.81, 127.18, 127.01, 124.11, 124.03, 123.33, 122.73, 117.43, 117.27, 111.13, 110.38, 110.28, 109.88, 108.47, 81.41, 80.53, 80.50, 72.15, 72.05, 71.62, 71.34, 67.73, 55.93, 39.63, 38.57, 39.43, 39.22, 37.29, 36.45, 30.93, 30.76, 30.74, 30.52, 30.22, 29.68, 29.10, 29.06, 27.92, 24.69, 24.18, 23.89, 23.87, 23.05, 22.64, 22.54, 19.88, 14.10, 14.06, 11.31, 11.26, 11.23, 11.20. MALDI-TOF MS (Mw =1029.58) m/z = 1029.45 [M]+. Reference polymer (5). To a sealed tube fitted with a magnetic stirrer was added diiodo monomer 417 (14.85 mg, 0.025 mmol), diethylnyl monomer 2 (25 mg, 0.024 mmol), Pd(PPh3)4 (1.16 mg, 0.001 mmol), CuI (0.19 mg, 0.001 mmol), dry Et3N (0.3 mL) and dry ortodichlorobenzene (0.7 mL). The reaction mixture was degassed using freeze-pump-thaw cycles and heated at 75 °C under Ar atmosphere for 24 h. After cooling it to room temperature, the reaction mixture was added dropwise to rapidly stirred EtOH (30 mL). After stirring for 2 h, the precipitate was collected and dried under vacuum overnight. Polymer 5 was obtained as 25 mg (75%) of an orange solid. SEC (chloroform): Mw =35.8 kg/mol, PDI = 2.82. Polymer 3. This polymer was prepared by a procedure identical polymer 5 using diethylnyl monomer 2 (22 mg, 0.02 mmol), diiodo monomer 1 (27mg, 0.018mmol), Pd(PPh3)4 (0.92 mg, 0.0008 mmol), CuI (0.15 mg, 0.0008mmol), dry Et3N (0.3 mL) and dry orthodichlorobenzene (0.7 mL ). Polymer 3 was obtained as 34 mg (76%) of a brown solid. SEC (chloroform): Mw =16.2 kg/mol, PDI = 2.32. Photoinduced absorption. PIA measurements were performed between 0.25 eV and 3.0 eV by exciting thin films on quartz with a mechanically modulated (275 Hz, 25 mW 2 mm diameter, 488 nm) beam from a continuous wave argon ion laser (Spectra physics 2025). The change in transmission of probe light (∆T) was monitored with a phase-sensitive lock-in amplifier using Si, InGaAs, and cooled InSb detectors after dispersion by a triple grating monochromator. The photoinduced absorption, -∆T/T ≅ ∆αd, is directly calculated from the change in transmission after correction for fluorescence, which is recorded in a separate experiment. The lifetime of the photoexcitations has been determined by recording the intensity of the PIA bands as a function of the modulation frequency (ω) in the range of 30-4000 Hz. Photovoltaic cells. For photovoltaic cells, polyethylenedioxythiophene polystyrenesulfonate (PEDOT:PSS, Bayer AG) (90 nm) was spin coated on UV-ozone cleaned glass substrates covered with indium tin oxide (ITO) (140 nm), followed by spin coating a solution of 3 in chloroform to form the active layer of 30 nm as determined with a Tencor P-10 surface profiler. Finally, an aluminum back electrode (100 nm) was deposited in
Chapter 5
122
vacuum to give an active area of 4 mm2. I/V characteristics were measured under 100 mW/cm2 AM1.5 white-light illumination from a Steuernagel Solarconstant 1200 solar simulator with a Keithley 2400 Source Meter in inert nitrogen atmosphere at room temperature. Surface roughness was determined by AFM. IPCE measurements. Spectrally resolved photocurrent measurements were performed by illuminating the device with 1 mW/cm2 monochromatized light with a FWHM of ~6 nm from a Xe arc lamp. No noticeable degradation of the devices was observed during the measurement cycles. Light intensities were measured by a calibrated Si photodiode. The incident photon to current efficiency (IPCE) or external quantum efficiency was calculated from the photocurrent and light intensity as:
IPCE [%] =[ A / cm
[nm] [W / m ]
2
2
1240 ⋅
⋅
I
I
sc µλ
]
5.6 References 1 (a) Tang, C. W. Appl. Phys. Lett. 1986, 48, 183. (b) Granström, M.; Petrisch, K.; Arias, A. C.; Lux, A.;
Lux, M.; Andersson. M. R.; Friend, R. H. Nature 1998, 395, 257.
2 (a) Halls J. J. M.; Walsh, C. A.; Greenham, N. C.; Marseglia, E. A.; Friend, R. H.; Moratti, S. C.;
Holmes, A. B. Nature 1995, 376, 498. (b) Yu, G.; Gao, Y.; Hummelen, J. C.; Wudl, F.; Heeger, A. J.
Science 1995, 270, 1789.
3 Shaheen, S. E.; Brabec, J. C.; Padinger, F.; Fromherz, T.; Hummelen, J. C.; Sariciftci, N. S. Appl. Phys.
Lett. 2001, 78, 841.
4 (a) Nierengarten, J.-F.; Eckert, J.-F.; Nicoud, J.-F.; Ouali, L.; Krasnikov, V. V.; Hadziioannou, G. Chem.
Commun. 1999, 617. (b) Stalmach, U.; de Boer, B.; Videlot, C.; van Hutten, P. F.; Hadziioannou, G. J.
The size exclusion chromatography (SEC) trace of the polymer in chloroform solution does
not correspond to a perfect Gaussian distribution, which suggests aggregation of the polymer in
chloroform solution. Addition of 1% methanol does not alter the shape of the SEC trace (Figure 6.3).
The molecular weight of the polymer is Mw= 14.6 kg/mol with PDI = 1.6. The molecular weight has
been determined using polystyrene standards, which gives a rough indication, but is not quantitative
due to the different shape of the polymer with regard to polystyrene. In contrast to polystyrene, which
is in a random coil conformation under these conditions, the polyacetylene is most probably in a rigid
rod conformation. The IR spectrum of polymer 1 does not feature the characteristic terminal acetylene
absorption, present in the starting monomers at around 3300 cm-1. This indicates that the
polymerization reaction has proceeded at the terminal ethynyl position and that no starting monomer
is left in the polymers. The 1H-NMR spectrum of polymer 1 shows very broad resonances. Generally,
Polyacetylenes with pendant donor-acceptor dyads
131
polyphenylacetylenes prepared using [Rh(nbd)Cl]2 exhibit sharp line widths, because of the high
stereoregularity of these polymers.7 However, polyacetylene 1 bears donor and acceptor
chromophores that are highly prone to aggregate. The mobility of the polymer backbone is most
probably restricted by this aggregation and peak broadening of the proton ressonances occurs.11
Consequently, the characteristic vinylic proton corresponding to the acetylene backbone could not be
distinguished and therefore, the regularity of the polymer backbone could not be determined. The
good solubility of copolymer 1 with respect to polymer 11 shows however that random incorporation
of non-perylene oligomers, disrupts the perylene aggregation by which the processability is greatly
enhanced.
300 400 500 600 700
0
1
Nor
mal
ized
res
pons
e
Time (sec)
Figure 6.3. SEC traces of polymers 1 (solid line), 2 (open squares) and 3 (open triangles) measured
in chloroform with 1% methanol. Inset: SEC traces of polymer 2 recorded in chloroform (solid
squares) and in chloroform with 1% methanol (open squares).
6.3.3 Polyacetylene with pendant OPVE-PERY dyads (polymer 2)
Polymer 2 has been synthesized by polymerization of the terminal acetylene OPVE-PERY
dyad 20 (Scheme 6.3). This dyad has been assembled by the non-selective reaction of diiodinated
OPV 17 with ethynyl perylene 10.
The synthesis of the OPV chromophore started by the radical bromination of 4-iodotoluene
giving 14. Treatment of 14 with triethylphosphite afforded phosphonate 15. Diiodinated OPV 17 was
obtained after a double Wittig-Horner reaction of phosphonate 15 with dialdehyde 16.19 The coupling
between the OPV and the PERY was done using the palladium catalyzed Sonogashira cross-coupling
of ethynyl compound 10 with diiodinated OPV 17 affording 19. An excess of 17 was used in order to
minimize the double coupling reaction, and the dyad was obtained as the main product of a statistical
Chapter 6
132
mixture. Due to the different nature of the two alkoxy side chains on the central ring, 17 is not
symmetrical and 19 therefore consists of a mixture of two regioisomers (only one of which has been
represented in Scheme 6.3). Coupling of 19 with TMS-acetylene and subsequent desilylation gave
functional monomer 20.
Monomer 20 was polymerized for 16 hours. Already after 20 minutes of polymerization a red
solid starts precipitating out of solution. The polymer was worked up by precipitation from the
reaction mixture with methanol. Removal of the cyclic trimers was achieved by reprecipitation from
THF. The polymer is slightly soluble in chloroform and o-dichlorobenzene, but the solubility strongly
increases with a small percentage of ethanol in the chloroform.
IR
OO O
O O
O
I
I
NNO
O
O
O
O
O
O
R
tmsH
NNO
O
O
O
O
O
O
*
*
n
16 17
+
2
18. R = I19. R = 20. R =
13. R = H14. R = Br15. R = P(OEt)2O
a
b
c
d
e
f
g
Scheme 6.3. a. NBS, benzoyl peroxide, CCl4, reflux, 3.5 h, 25%; b. Triethylphosphite, 160 °C, 1.5 h,
100 %; c. t-BuOK, DMF, r.t., 3 h, 64 %; d. 10, Pd(PPh3)2Cl2, CuI, Et3N, toluene, 50 °C, 16 h, 47%; e.
TMA, Pd(PPh3)2Cl2, CuI, Et3N, toluene, 50 °C, 16 h, 80%; f. TBAF/Silica, THF, r.t., 30 min, 97%; g.
[Rh(nbd)Cl]2, Et3N, toluene, r.t. 16h, 60%.
In contrast to polymer 1, the SEC trace of 2 is enormously influenced by the addition of a
small percentage of methanol (Figure 6.3, and inset). The SEC trace in chloroform points to very
Polyacetylenes with pendant donor-acceptor dyads
133
strong aggregation, compared to polymer 1, but this is largely overcome by the addition of methanol.
But even methanol does not allow for complete disruption of the aggregation, which makes it difficult
to estimate an accurate molecular weight against polystyrene standards (Mw= 45.8 kg/mol with PDI
=1.9, using 1% methanol in chloroform as eluent). As for polymer 1, the 1H-NMR spectrum of
polymer 2 shows extremely broad resonances.
6.3.4 Polyacetylene with pendant OPE-PERY dyads (polymer 3)
For polymer 3 an acetylene terminated OPE-PERY dyad was synthesized. In this case, the
OPE donor segment has first been end-capped with two different functionalities (a TMS-ethynylene
group and a bromine) to facilitate a selective monocoupling to the phenylethynyl terminated perylene
acceptor 10.
For the synthesis of the bromine-ethynyl bifunctional donor OPE (Scheme 6.4) a repetitive
strategy developed by Ziener et al. was followed.12 The synthesis started by bisbromination and
subsequent bisalkylation of hydroquinone that gave dibromide 23. The exchange of one bromine for
one iodine to yield 24 was achieved by reaction of 23 with n-BuLi followed by treatment with 1,2-
diiodoethane, while working at temperatures below –80 °C. Selective coupling of TMS-acetylene
with 24 at the iodo-phenyl position yielded 25. In contrast to literature,12 iodo-selectivity was only
achieved by keeping the temperature of the reaction at 0 °C and by using exact amounts of TMS-
acetylene. One part of 25 was treated with TBAF to give acetylene 26. The other part was converted
into the iodo compound 27. The yield of the halogen exchange was only 77%, and several attempts of
separating 27 from 25 using column chromatography were unsuccessful. Nevertheless, the synthetic
procedure was continued and removal of 25 was achieved in the following synthetic step. An iodo-
bromo-selective cross-coupling of 26 with 27 gave dimer 28. Tetramer 31 was derived from 28 after
another cycle of desilylation, iodine-bromine exchange and a palladium catalyzed cross-coupling.
Although tetramer 31 was obtained in reasonable overall yield, the synthesis of the donor fragment is
hampered by difficult and tedious purification of some of the intermediates.
Coupling of the donor tetramer 31 with ethynyl terminated perylene 10, followed by treatment
with TBAF gave acetylene 33. An attempt to polymerize 33, resulted selectively in cyclic trimers due
to the alkoxy substituents on the first phenyl ring. Therefore 33 was extended by coupling it to 4-
iodotrimethylsilylethynylbenzene to afford monomer 35 after deprotection of the ethynyl function.
Polymer 3 was prepared using the usual polymerization conditions for 16 hours. The reaction
mixture was precipitated into methanol and THF to yield a red solid, which is highly soluble in
solvents of intermediate polarity such as chloroform, dichloromethane and o-dichlorobenzene.
Chapter 6
134
OH
OH
OH
OH
BrBr
OR
RO
BrBr
OR
RO
IBr
OR
RO
Br tms
OR
RO
Br H
OR
RO
tmsI
OR
RO
Br tms
OR
RO
Br H
OR
RO
I tms
OR
RO
Br tms
ONN
O
O O
O
OR
RO
R'
ONN
O
O O
O
OR
RO
R'
OR = O
ONN
O
O O
O
OR
RO
H
*
*n
2
2
2
4
a b c
de
f
g
h
i
j
4
4
l
m
32. R' = tms33. R' = H
n
o
34. R' = tms35. R' = H
21 22 23 24
25
26
27
28
29
30
31
4
3
k
Scheme 6.4. a. HBr, Br2, 100 °C, 24 h, 49%; b. ROTs, KOH, EtOH, reflux, 24 h, 57%; `c. n-BuLi,
1,2-diodoethane, THF, -60 °C, 72%; d. TMA, Pd(PPh3)2Cl2, CuI, Et3N, 0 °C, 1 h, 82%; e. TBAF,
THF, r.t.,1 min, 100%; f. n-BuLi, 1,2-diodoethane, THF, -60 °C, 77%; g. Pd(PPh3)2Cl2, CuI, Et3N, 0
°C, 3.5 h, 67%; h. TBAF, THF, r.t., 1 min, 100%; i. n-BuLi, 1,2-diodoethane, THF, -60 °C, 60%; j.
Pd(PPh3)2Cl2, CuI, Et3N, 0 °C, 3.5 h, 56%; k. 10, Pd(PPh3)2Cl2, CuI, Et3N, toluene, 50 °C, 16 h, 34%
; l. TBAF/Silica, THF, r.t., 30 min, 100% ; m. 4-Iodotrimethylsylilphenylacetylene Pd(PPh3)2Cl2, CuI,
Et3N, toluene, 50 °C, 16 h, 40% ; n. BAF/Silica, THF, r.t., 30 min, 99% ; o. [Rh(nbd)Cl]2, Et3N,
toluene, r.t. 16h, 30%.
Polyacetylenes with pendant donor-acceptor dyads
135
The molecular weight of the polymer according to SEC is Mw = 53.5 kg/mol with PDI= 1.5
(Figure 6.3). The high molecular weight, especially with respect to polymer 1, is partly due to the high
molecular weight of the monomers, but most probably to the good solubility of polymer 3.
Additionally, the different hydrodynamic volume caused by the large pendant dyads might contribute
to the high molecular weight. In the 1H-NMR spectrum of polymer 3, peak broadening is somewhat
less prominent than for polymers 1 and 2. The OPE segment is heavily substituted with alkoxy chains,
which may hamper aggregation and allow for a better resolution in the NMR spectrum.
6.4 UV/Visible absorption and circular dicroism spectroscopies
The absorption spectra of polymers 1, 2 and 3 in chloroform solution exhibit the electronic
transitions associated with their donor and acceptor chromophores (Figure 6.4). All spectra feature the
well-resolved vibronic structure absorption characteristic of the PERY chromophore with maxima at
490 and 530 nm, approximately. The donor moieties dominate the rest of the absorption spectrum.
Monosubstituted polyacetylenes mainly absorb in the UV region and have very low extinction
coefficients. 13 Therefore the contribution of the backbone to the absorption spectra can be considered
negligible.
In Figure 6.4 the absorption spectra of all monomers in chloroform have been plotted together
with those of the respective polymers. The polymers exhibit a red-shifted onset with respect to their
corresponding monomers. Also the normalized spectra of the monomers and the related polymers do
not superimpose. These differences in the spectra suggest that at least the perylene chromophore is
aggregating when incorporated in the polymeric chain. In the solid state the onset is even more red-
shifted and the absorption bands even broader. By comparing the absorption spectra of the polymers
in chloroform solution with those of the monomeric species and with that in the solid state it can be
concluded that already in chloroform solution the pendant chromophores and dyads are interacting
with each other and are preorganizing the polymer into its solid state conformation. Of all, 2 is the
polymer that experiences the highest aggregation in solution as the absorption spectra in solution
highly resembles the one in solid state.
None of the polymers showed a Cotton effect in chloroform solution, even though they are
decorated with stereocenters. The absence of optical activity might stem either from aggregation
phenomena that disrupt the helical arrangement or from the presence of racemic side-chains or
mixtures of side-chains with different chiral centers, resulting in non-preferential twist senses of the
helices.
Chapter 6
136
300 400 500 6000
1
Abs
orba
nce
(O.D
.)
Wavelength (nm)
0
1
0
1
b
c
a
Figure 6.4. Absorption spectra of the different polyacetylenes and monomers.(a) Polymer 1 in
solution (solid line), and solid state (doted line) and monomer 10 (dashed-dotted line) and 12 (dashed
line). (b) Polymer 2 in solution (solid line) and thin film (dotted line) and monomer 20 (dashed line).
(c) Polymer 3 in solution (solid line) and in solid state (dotted line) and monomer 35 (dashed line).
6.5 Photophysical properties
6.5.1 Solid state
Near steady-state photoinduced absorption (PIA) spectroscopy of thin films of polymers 1, 2
and 3 recorded at 80 K revealed the occurrence of charge separation upon illumination (Figure 6.5 left
and right). All PIA spectra exhibit the absorptions characteristic of the perylene bisimide radical anion
at 1.28, 1.54 and 1.73 eV.14 For polymer 3 two additional absorptions centered at 0.5 and at 2.0 eV are
observed. A similar PIA spectrum is obtained from a blend of a p-phenylene ethynylene polymer
(PPE)15 with a soluble perylene dye PERY (Figure 6.5 right), meaning that the additional absorption
bands observed in the PIA spectrum of polymer 3 correspond to the OPE•+ radical cations. The
Polyacetylenes with pendant donor-acceptor dyads
137
absorptions of the positively charged donor moieties in polymers 1 and 2 are expected at similar
energies as observed for the OPE segments.16 However, the PIA spectrum of polymers 1 and 2 display
very weak ill-defined absorptions at low energies. Similar low absorption of the OPV•+ radical cation
have been previously observed in the PIA spectrum of a molecular triad based on comparable
chromophores as in polymers 1 and 2, OPV3-PERY-OPV3 (Figure 6.5 left).17
0.5 1.0 1.5 2.0 2.5
-2
0
2
4
6
∆T/T
x 1
04
Energy (eV)
0.5 1.0 1.5 2.0 2.5
0
2
4
∆T/T
x 1
04
Energy (eV)
Figure 6.5. PIA spectra of thin films of the polyacetylene polymers. Left: polymers 1(solid line), 2
(dashed line) and OPV3-PERY-OPV3 (solid squares). Right: polymer 3 (solid line) and PPE/PERY
solid state blend (solid squares).
NN
O
O
O
O
* *
O
O
n
OO
OO
OO
OONN
O
O
O
O
OPV3-PERY-OPV3
PERY
PPE
Figure 6.6. Chemical structure of reference compounds OPV3-PERY-OPV3, PPE and PERY.
The rates of formation and decay of the charge-separated states in thin films of 2 and 3 have
been inferred from sub-picosecond pump-probe spectroscopy (Figure 6.7 and Table 6.1). The time
profile of the 1450 nm absorption, corresponding to the donor radical cation, has been measured after
exciting the polymers at 450 nm. Fitting of the temporal transient absorption to exponential functions
reveals that the charges are formed within 1 ps and that their lifetimes range from 300 to 400 ps. For
polymer 1 only extremely weak signals could be measured.
Chapter 6
138
0 200 400 600 800 1000-70
-60
-50
-40
-30
-20
-10
0
Time (ps)
∆T (
a. u
.)
Time (ps)
-3 0 3-60
-30
0
Figure 6.7. Differential transmission dynamics of the 1450 nm absorption of polymers 2 (solid
squares) and 3 (open squares) in a thin film, after excitation at 450 nm. The inset shows the 1450 nm
signals on short timescales.
Table 6.1. Lifetimes and rates for the charge separation (CS) and charge recombination (CR) of
polymers 1, 2 and 3 in the solid state and in chloroform solution.
Solid state CHCl3 solution
CS CR CS CR
τ (ps) k (s-1) τ (ps) k (s-1) τ (ps) k (s-1) τ (ps) k (s-1)
Polymer 1 n.d. n.d. <1 >1012 13 8·1010
Polymer 2 <1 >1012 340 3·109 2.2 5·1011 634 2·109
Polymer 3 <1 >1012 425 2·109 6 2·1011 400 2·109
6.5.2.Chloroform solution
The photoluminescence spectra of polymers 1, 2 and 3, recorded in chloroform solution after
excitation of either donor or acceptor, exhibit very weak emission intensities. In particular, the
fluorescence emission of the OPV chromophore is 60 times lower than that of the starting monomer
12. The PL of the PERY chromophore is also quenched in polymer 1, by a factor of 20 with respect to
the PL of the PERY model compound (Figure 6.6). For polymers 2 and 3 the PL quenching factor of
the PERY chromophore scales to 450 times. The modest PL quenching in copolymer 1, with respect
to 2 and 3, can be attributed to a less efficient charge transfer in the copolymer caused by the presence
of some domains of only OPV or PERY chromophores within the polymer chains.
Polyacetylenes with pendant donor-acceptor dyads
139
500 600 7000
50
Inte
nsity
(a.
u.)
Wavelength (nm)
Figure 6.8. PL spectra of monomers 10 (solid squares) and 12 (solid circles) and polymer 1 in
chloroform solution, after selective excitation of the OPV (open circles) and of the PERY (open
squares) chromophores at 400 and 520 nm respectively.
According to sub-picosecond pump-probe spectroscopy, charge separation occurs
immediately in copolymer 1 (Figure 6.9 left and Table 6.1), which is consistent with a face-to-face
orientation of donor and acceptor, the most probable orientation of two neighboring pendant
chromophores in the copolymer. The face-to-face orientation of the chromophores is also responsible
for a very fast recombination of the charges. Additional to the fast decay of the majority of the
charges, a weak signal remains that only decays very slowly. This residual signal most probably
corresponds to trapped charges in small domains of donor or acceptor units. The presence of small
domains of chromophore units with equal signature has also been inferred from fluorescence
spectroscopy.
In polymers 2 and 3 the charges are generated more slowly than in polymer 1, but what is
more important, their lifetimes have been increased by more than one order of magnitude (Figure 6.9
and Table 6.1). This is the result of the different design of the last two polymers. The spacer placed
between the donor and the acceptor moieties slows down the charge separation and recombination
processes by intercalating a distance between the redox centers and, because of its semirigid nature,
by preventing the folding of the PERY donor on top of the donor segment to a face-to-face
orientation.9b
Both for polymer 2 and for polymer 3 the formation of the charges in the solid state is faster
than in solution (Table 6.1). However, for polymer 2 the formation of the charges is increased by only
a factor of 2 upon going from solution to the solid state, whereas for polymer 3 this increases by a
factor of 6. This observation is in line with the results obtained by absorption spectroscopy, which
show that the conformation and aggregational state of polymer 2 in solution closely resembles that in
Chapter 6
140
the solid state. For polymer 3 this relationship was not manifested as strongly as for polymer 2.
Polymer 3 was found to be much better dissolved in chloroform and hence features a much larger
conformational flexibility. The decrease in conformational flexibility for polymer 3 upon going to the
solid state accounts for the chromophores lining up in close proximity, thus very rapidly generating
the charges.
0 200 400 600 800 1000
-5
0
∆T (
a. u
.)
Time (ps)
∆T (
a. u
.)
Time (ps)
-3 0 3
-5
0
0 200 400 600 800 1000
-100
-50
0
Time (ps)
∆T (
a. u
.)
Time (ps)
-5 0 5 10 15-100
-50
0
Figure 6.9. Differential transmission dynamics of the 1450 nm absorption of polymers 1 (left, solid
line), 2 (right, solid squares) and 3 (right, hollow squares) in chloroform solution, after excitation at
450 nm. The insets show the 1450 nm signals on short timescales.
Similar as for the charge separation in polymer 2 also the lifetime of the charges is decreased
by a factor of 2 upon going from solution to the solid state. This reduction of the lifetime results
probably from the combination of denser packing of the chromophores in the solid state and increased
interchain interactions, both speeding up the charge recombination in polymer 2 in the solid state.
Surprisingly, for polymer 3 a different behavior of the lifetime of the charges is observed. The
lifetime of the charges is similar in solution and in the solid state. This phenomenon might be the
result of the alkoxy substitution of the OPE element. Although a denser packing might also occur for
polymer 3 upon going to the solid state, the alkoxy side chains might hamper lateral interaction of the
OPE chromophores with acceptors belonging to other polymer chains, preventing ‘interchain
recombination’, and thus allowing longer lifetimes. The alkoxy side chains might play the same role
as the spacer between donor and acceptor in a single dyad.
6.6 Conclusions and outlook
Several polyacetylenes substituted with a OPV or OPE donor and a PERY acceptor have been
synthesized for the first time. By designing different architectures, the strengths and weaknesses of
the placement of the functional groups on processability and photophysical properties have been
explored.
Polyacetylenes with pendant donor-acceptor dyads
141
The polymers were prepared either by the copolymerization of OPV and PERY terminal
acetylenes or by the polymerization of homologous OPVE-PERY and OPE-PERY dyads. SEC
analyses exhibit the typical distribution corresponding to polymeric systems. The molecular weights
of the polymers are in the range of 15-55 kDa giving especially polymers 2 and 3 good processing
properties, due to their higher molecular weight and good solubility in common organic solvents.
Nevertheless, the low solubility of especially polymers 1 and 2 in toluene, used in the polymeriztion,
limits the chain length. Although the polydispersity of the polymers is good, the degree of
polymerization might eventually be optimized by a different choice of the polymerization solvent.
Even though the copolymerization approach as applied for polymer 1 offers the fastest
synthetic access to donor-acceptor functionalized polyacetylenes, the alternating arrangement of
donors and acceptors results in very short-lived charged-separated states. It remains to be seen if this
can be overcome by another design of the two monomers. On the contrary, polymers 2 and 3 show
promising photophysical properties owing to their more refined architecture, avoiding donor-acceptor
stacking and favoring donor-donor and acceptor-acceptor interactions. In these polymers the lifetime
of the photoinduced charge separated state is increased by at least more than one order of magnitude
with respect to polymer 1.
These results, together with the optimization of the polymerization conditions, make these
polyacetylenes an attractive system to explore in future experiments.
6.7 Experimental Section
For general methods the reader is referred to chapter 2 of this thesis. [(S)-1-Isobutyl-2-(4-iodo-phenyloxy)ethyl]carbamic acid tert-butyl ester (5). Iodophenol (4.34g, 19 mmol), 49 (4.25 g, 19 mmol) and triphenylphosphine (7.77 g, 29 mmol) were dissolved in toluene (60 mL). Diethyl azodicarboxylate (4.66 ml, 29 mmol) in toluene (30 mL) was slowly added while keeping the temperature below 35 °C. The reaction mixture was stirred for 16 h at room temperature and subsequently washed with 1N aqueous HCl. The organic phase was dried over MgSO4 and evaporated in vacuo. Column chromatography (n-heptane/EtOAc, 4/1, Rf = 0.3) and crystallization from hexane afforded the product (3.2 g, 40%) as white crystals.1HNMR (CDCl3, 300 MHz): δ 7.54 (d, 2H), 6.67 (d, 2H), 4.75-4.65 (m, 1H), 4.10-3.8 (m, 3H), 1.80-1.65 (m, 1H), 1.60-1.40 (m, 11 H), 1.00-0.88 (m, 6H); 13C NMR (CDCl3, 75 MHz): δ .158.50, 155.18, 138.04, 116.80, 82.91, 79.36, 70.25, 48.15, 41.00, 28.48, 24.94, 23.12, 22.38; Anal. Calc for C17H26INO3: C, 48.7; H, 6.2; N, 3.3. Found: C, 48.9; H, 6.1; N, 3.28. (S)-1-Isobutyl-2-(4-iodo-phenyloxy)-ethylamine (6). TFA (12 mL) was added to a solution of 5 (3 g, 7.3 mmol) in methylene chloride (12 mL). After stirring the reaction mixture for 16 h at room temperature under an argon atmosphere, NaHCO3 was added. The reaction mixture was subsequently washed with water, NaOH 1 N, water, brine and dried over Na2SO4.The product (2 g, 86%) was obtained after evaporation of the solvent as a colorless oil.1H NMR (CDCl3, 300 MHz): δ 7.54 (d, 2H), 6.68 (d, 2H), 3.86 (dd, 1H), 3.64 (dd, 1H), 3.22 (s, 1 H), 1.81-1.74 (m, 1 H), 1.25 (t, 2H), 0.97-0.92 (m, 6H); 13C NMR (CDCl3, 75 MHz): δ .158.64, 137.99, 116.76, 82.66, 73.69, 48.27, 42.98, 24.46, 23.27, 21.91.
Chapter 6
142
N-[(S)-1-Isobutyl-2-(4-iodo-phenyloxy)-ethyl]-N’-(1-ethylpropyl)-3,4,9,10-perylenebis(dicarboximide) (8). N-(1-ethylpropyl)-3,4,9,10-perylenetetracarboxylic monoanhydride monoimide 518 (1.91 g, 4.17 mmol), amine 4 (2 g, 6.26 mmol), imidazol (30 g, mmol), and catalytic amounts of Zn(AcO)2 were mixed and stirred for 2.5 h at 160 °C. After cooling to room temperature the solid reaction mixture was dissolved in CH2Cl2 and washed with HCl 1N, water, brine and dried over MgSO4. Purification by column chromatography (flash silica, CH2Cl2, Rf = 0.36) yielded 1.97 g (67%) of as a dark red solid.1HNMR (CDCl3, 300 MHz): δ 8.76-8.59 (m, 8H), 7.46 (d, 2H), 6.65(d, 2H), 5.77-5.69 (m, 1H), 5.11-5.03 (m, 1H), 4.72 (dd, 1H), 4.33-4.30 (dd, 1H), 2.33-2.21 (m, 3H), 2.00-1.90 (m, 2H), 1.81- 1.75 (m, 1H), 1.67-1.59 (m, 1H) 1.02 (d, 3H), 098 (d, 3H), 0.94 (t, 6H); 13C NMR (CDCl3, 100 MHz): δ 163.16, 158.20, 137.87, 133.58, 133.22, 131.15, 130.61, 128.72, 125.35, 123.18, 122.32, 122.12, 117.02, 82.98, 68.78, 57.77, 54.41, 38.63, 25.64, 25.04, 23.14, 22.60, 11.62. Anal. Calc for C41H35IN2O5: C, 64.6; H, 4.6; N, 3.7. Found: C, 64.3; H, 4.4; N, 3.5. MALDI-TOF MS (Mw = 762.65) m/z= 762.08 [M]+. N-[(S)-1-Isobutyl-2-(4-trimethylsylylethynyl-phenyloxy)-ethyl]-N’-(1-ethylpropyl)-3,4,9,10-perylenebis(dicarboximide) (9). A mixture of 8 (1.42 g, 1.86 mmol), Pd(PPh3)2Cl2 (0.104 g, 0.148 mmol), CuI ( 0.021 g, 0.107 mmol) was dissolved in anhydrous toluene/triethylamine (320 mL, 5:1). After argon was purged trough the solution for 15 min trimethylsylylacetylene (0.32 mL, 2.23 mmol) was added. The reaction mixture was heated at 50 oC for 4 h. The crude reaction mixture was filtered through Celite, evaporated to dryness, and chromatographied (flash silica, CH2Cl2, Rf = 0.35) to afford a red solid (1.2 g, 84 %). 1H NMR (CDCl3, 300 MHz): δ 8.56-8.36 (m, 8H), 7.34 (d, 2H), 6.84 (d, 2H), 5.76-5.70 (m, 1H), 5.10-5.03 (m, 1H), 4.80 (dd, 1H), 4.39-4.34 (dd, 1H), 2.33-2.20 (m, 3H), 2.04-1.90 (m, 2H), 1.87-1.75 (m, 1H), 1.68-1.61 (m, 1H), 1.07-0.94 (m, 12H), 0.18 (s, 9H); 13C NMR (CDCl3, 100 MHz): δ 163.80 (broad signal), 158.78, 134.36, 134.01, 133.37, 131.20, 129.33, 126.08, 123.47, 122.88, 122.71, 115.41, 114.61, 105.10, 92.43, 68.64, 57.72, 53.40, 51.40, 38.53, 25.55, 24.98, 23.10, 22.42, 11.38, -0.03. Anal. Cald for C41H35IN2O5: C, 64.6; H, 4.6, N, 3.4. Found: C, 64.3; H, 4.4, N, 3.4. MALDI-TOF MS (Mw = 732.96) m/z= 732.30 [M]+. N-[(S)-1-Isobutyl-2-(4-ethynyl-phenyloxy)-ethyl]-N’-(1-ethylpropyl)-3,4,9,10-perylenebis(dicarboximide) (10). To a solution of 9 (0.4 g, 0.56 mmol) and tetrabutylammonium fluoride on silica (0.177 g, 0.68 mmol) in 70 mL THF was stirred for 0.5 h under argon. The reaction mixture was filtered through Celite and evaporated to dryness. The product was obtained as a red powder (0.3 g, 81%) and was used without further purification. 1H NMR (CDCl3, 300 MHz): δ 8.63-8.50 (m, 8H), 7.35 (d, 2H), 6.84 (d, 2H), 5.76-5.70 (m, 1H), 5.11-5.02 (m, 1H), 4.80 (dd, 1H), 4.39-4.34 (dd, 1H), 2.94 (s, 1H), 2.33-2.20 (m, 3H), 2.04-1.90 (m, 2H), 1.87-1.75 (m, 1H), 1.68-1.61 (m, 1H), 1.07-0.94 (m, 12H); 13C NMR (CD2Cl2, 100 MHz): δ.163.81, 159.55, 134.15, 133.89, 133.75, 131.56, 131.01, 129.25, 125.87, 123.72, 122.90, 122.70, 115.16, 114.65, 83.76, 76.14, 69.31, 58.40, 51.64, 38.99, 25.93, 25.36, 23.23, 22.63, 11.71. MALDI-TOF MS (Mw =660.78) m/z = 660.38 [M]+. 1-Bromomethyl-4-iodobenzene (14). A mixture of 4-iodotoluene (10 g, 458.8 mmol), N-bromosuccinimide (9.79 g, 55 mmol) and benzoyl peroxide (0.44 g, 1.8 mmol) in dry carbon tetrachloride (14 mL) was stirred and heated to reflux for 3.5 h and then cooled and filtered. The red filtrate was washed with saturated sodium thiosulfate solution (10 mL), dried and filtered. The solvent was removed in vacuo and the solid residue was further purified by recrystalization in hexane. The product was obtained as an off-white solid (3.5 g, 25%). 1H NMR (CDCl3, 300 MHz): δ 7.67 (d, 2H), 7.15(d, 2H), 4.41 (s, 1H); 13C NMR (CDCl3, 100 MHz): δ 137.94, 137.40, 130.82, 94.12, 32.45. Diethyl(4-iodo-benzyl)phosphonate (15). Triethyl phosphite (2.28 g, 13 mmol) and 14 (3.40 g, 11 mmol) were stirred at 160 oC for 1.5 h under an argon atmosphere. The reaction mixture was cooled to 75 oC and the ethyl bromide, formed during the reaction, and the excess of triethyl phosphite were distilled under reduced pressure. The product was obtained as a light yellow oil. Yield 3.89 g (100%). 1H NMR (CDCl3) δ 7.54(d, 2H), 6.97 (dd, 2H), 4.00- 3.84(m, 4H), 2.98 (d, 2H), 1.17(t, 6H); 13C NMR (CDCl3) δ 137.32 (d), 131.48 (d), 131.20 (d), 92.09 (d), 61.93 (d), 34.00, 32.17, 16.16 (d). (E,E)-1,4-Bis(4-iodo-styryl)-2-(3,7-dimethyloctyloxy)-5-methoxybenzene (17). Phosphonate 15 (0.8 g, 2.25 mmol) was dissolved in dry DMF (9 mL) under an argon atmosphere and 0.32 g (2.8 mmol) of KtBuO were added to the solution. After 15 min, a solution of dialdehyde 1619 (0.29 g, 0.9 mmol) in dry DMF (6 mL) was added dropwise and the reaction mixture was stirred for 3.5 h. The solution was poured on crushed ice and of 6 M HCl (200 mL) was added. The aqueous phase was extracted twice with diethyl ether
Polyacetylenes with pendant donor-acceptor dyads
143
and the combined organic layers were subsequently washed with 3 M HCl, water, a saturated aqueous solution of Na2CO3 and dried over MgSO4. The solvent was removed in vauo. Column chromatography (flush silica gel, CH2Cl2/pentane, 4:1, Rf = 0.3) and evaporation of the solvent afforded 0.42 g (64%) of the product as a yellow solid: 1H NMR (CDCl3; 300 MHz): δ 7.67 (d, 2H), 7.67 (d, 2H), 7.47(d, 1H), 7.46 (d, 1H), 7.27 (d, 2H), 7.26 (d, 2H), 7.11 (s, 1H), 7.09 (s, 1H), 7.06 (d, 1H), 7.02 (d, 1H), 4.09 (t, 2H), 3.91 (s, 3H), 1.97-1.86 (m, 1H), 1.82-1.45 (m, 4H), 1.45-1.05 (m, 6H), 0.99 (d, 3H), 0.87 (d, 3H). 13C NMR (CD2Cl3, 100 MHz) δ 151.54, 151.19, 137.76, 137.54, 137.48, 128.25, 128.20, 127.73, 127.69, 126.58, 126.35, 124.20, 123.98, 110.35, 109.09, 92.46, 92.42, 67.75, 56.19, 39.27, 37.41, 36.49, 30.14, 28.03, 24.85, 22.49, 22.39, 19.57. MALDI-TOF MS (Mw =720.48) m/z= 719.95[M]+. N-[(S)-1-Isobutyl-2-<4-{(E,E)-4-[4-(4-iodostyryl)-2-(3,7-dimethyloctyloxy)-5-methoxy)styryl]phenylethynyl}phenyloxyl>ethyl]-N’-(1-ethylpropyl)-3,4,9,10-perylenebis(dicarboximide) (18). A degassed solution of 17 (0.19 g, 0.26 mmol), 10 (0.12 g, 0.18 mmol), Pd(PPh3)2Cl2 (0.01 g, 0.014 mmol), CuI (0.002 g, 0.01 mmol) in anhydrous toluene/triethylamine (30 mL, 5:1) was heated at 50 oC for 16 h. The crude reaction mixture was filtered through Celite, evaporated to dryness, and chromatographied (flash silica, CHCl3/pentane, 9:1, Rf = 0.5) to afford a red solid (0.102 g, 47 %). 1H NMR (CDCl3, 400 MHz): δ 8.56-8.36 (m, 8H), 7.64 (d, 1H), 7.62 (d, 1H), 7.44-7.38 (m, 8H), 7.22 (d, 1H), 7.21 (d, 1H), 7.06-6.97 (m, 4H), 7.01 (d, 1H), 7.00(d, 1H), 6.99 (d, 1H), 6.93-6.90 (m, 2 H), 5.81-5.74 (m, 1H), 5.11-5.03 (m, 1H), 4.82 (dd, 1H), 4.41-4.38 (dd, 1H), 4.08-3.99 (m, 2H), 3.87 (s, 3H), 2.34-2.22 (m, 3H), 2.03-1.81 (m, 4H), 1.78-1.13 (m, 4 H),1.58-1.16 (m, 6H), 1.06 (d, 3H), 1.02 (d, 3H), 0.98 (d, 3H), 0.96 (t, 6H), 0.86 (d, 3H), 0.85 (d, 3H); 13C NMR (CDCl3, 100 MHz): δ 163.80 (broad signal), 158.66, 151.35, 151.37, 151.04, 137.65, 137.58, 137.36, 137.30, 134.39, 134.04, 132.93, 131.62, 131.56, 131.18, 129.34, 129.30, 128.24, 128.18, 128.11, 127.64, 127.61, 126.77, 126.52, 126.41, 126.34, 126.29, 126.14, 126.08, 124.17, 123.99, 123.91, 123.72, 122.87, 122.74, 122.31, 115.58, 114.85, 110.19, 108.93, 92.54, 92.49, 90.28, 88.40, 68.69, 67.63, 57.70, 56.13, 51.45, 39.19, 38.58, 37.36, 36.43, 30.06, 29.66, 27.93, 25.58, 24.98, 24.80, 24.78, 23.12, 22.67, 22.58, 22.48, 19.74, 11.41. MALDI-TOF MS (Mw =1253.34) m/z= 1252.23[M]+. N-[(S)-1-Isobutyl-2-<4-{(E,E)-4-[4-(4-trimethylsylylethynyl-styryl)-2-(3,7-dimethyloctyloxy)-5-methoxy)styryl]phenylethynyl}phenyloxy>ethyl]-N’-(1-ethylpropyl)-3,4,9,10-perylenebis(dicarboximide) (19). A mixture of 18 (0.073 g, 0.058 mmol), Pd(PPh3)2Cl2 (0.005 g, 0.007 mmol), CuI ( 0.001 g, 0.005 mmol) was dissolved in anhydrous toluene/triethylamine (18 mL, 5:1). After argon was purged trough the solution for 15 min trimethylsylylacetylene (0.1 mL, 0.707 mmol) was added. The reaction mixture was heated at 50 oC for 16 h. The crude reaction mixture was filtered through Celite, evaporated to dryness, and chromatographied (flash silica, CH2Cl2, Rf = 0.5) to afford a red solid (57 mg, 80%). 1H NMR (CDCl3, 300 MHz): δ 8.56-8.37 (m, 8H), 7.45-7.37 (m, 12H), 7.06 (d, 1H), 7.05, 7.04, 7.03, 7.03 (4xs, 2H), 7.02 (d, 1H), 6.92-6.89 (m, 2H), 5.80-5.73 (m, 1H), 5.09-5.03 (m, 1H), 4.82 (dd, 1H), 4.41-4.37 (dd, 1H), 4.05-4.01 (m, 2H), 3.82 (s, 3H2.34-2.22 (m, 3H), 2.03-1.81 (m, 4H), 1.78-1.13 (m, 4 H), 1.58-1.16 (m, 6H), 1.06 (d, 3H), 1.02 (d, 3H), 0.98 (d, 3H), 0.96 (t, 6H), 0.86 (d, 3H), 0.85 (d, 3H) ), 0.18 (s, 9H); 13C NMR (CDCl3, 100 MHz): δ 163.77 (broad signal), 158.66, 151.38, 151.07, 137.99, 137.41, 137.33, 134.39, 134.04, 132.91, 132.21, 131.59, 131.55, 131.13, 130.72, 129.30, 128.19, 128.06, 126.72, 126.56, 126.47, 126.33, 126.27, 126.18, 126.08, 124.33, 124.06, 123.75, 123.38, 122.87, 122.71, 122.28, 121.80, 115.58, 114.84, 110.21, 108.92, 105.26, 94.99, 90.25, 88.39, 68.68, 67.65, 57.69, 56.11, 53.39, 51.45, 39.19, 38.56, 37.34, 36.41, 30.05, 27.93, 25.57, 27.98, 24.77, 23.09, 22.65, 22.55, 22.45, 19.71, 11.38, -0.054. MALDI-TOF MS (Mw =1223.65) m/z = 1222.36 [M]+. N-[(S)-1-Isobutyl-2-<4-{(E,E)-4-[4-(4-Ethynyl-styryl)-2-(3,7-dimethyloctyloxy)-5-(methoxy)styryl]phenylethynyl}phenyloxy>ethyl]-N’-(1-ethylpropyl)-3,4,9,10-perylenebis(dicarboximide) (20). A solution of 19 (50 mg, 0.041 mmol) and tetrabutylammonium fluoride on silica (0.05 mL of a 1M solution, 0.05 mmols) in 5 mL THF was stirred for 0.5 h under argon. The reaction mixture was filtered through Celite and evaporated to dryness. The product was obtained as a red powder (46 mg, 97 %) and was used without further purification. The complete deprotection of the ethynyl bond was confirmed by 1H-NMR spectroscopy as the disappearance of the singlet at 0.18 ppm corresponding to the TMS end group and the appearance of the signal for the ethynylinic proton at 3.10 ppm. 2,5-Dibromo-hydroquinone (22). p-Benzoquinone (15 gr, 136 mmol) was dissolved in concentrated HBr (320 mL). Bromine (22.5 gr, 140 mmol) was added slowly to the solution while stirring. After 7 h the reaction mixture was heated to 100 °C and stirred
Chapter 6
144
for another 12 h. An additional amount of bromine (7 g, 43 mmol) was added and reacted at 100 °C for another 3 h. The reaction mixture was cooled to room temperature and the precipitated solid washed with water. The product was purified by repetitive recrystalization in water with addition of charcoal and obtained as white crystals (17.98 g, 49%). 1H NMR (CD3CO CD3, 400 MHz): δ 8.58 (s, 2H), 7.16(s, 2H); 13C NMR (CDCl3, 100 MHz): δ 147.98, 120.05, 108.90. 1,4-Dibromo-2,5-bis(2-ethylhexyloxy)benzene (23). Dibromohydroquinone 22 (26.27 g, 98 mmol) was dissolved in a solution of potassium hydroxide (21.99 g, 392 mmol) in ethanol (1250 mL) and heated to reflux, after what 2-ethylhexyl-p-toluenesulfonate (97.56 g, 343 mmol) was added. The reaction mixture was cooled to room temperature after 16 h. The precipitate formed was filtered off and washed with methanol. The solvent of the liquid phase was removed in vacuo. The resulting crude product was purified by column chromatography ( silica gel, hexane, Rf = 0.35). Evaporation of the solvent yielded 27.81 g (57 %) of a colorless oil. 1H NMR (CD3CO CD3, 300 MHz): δ 7.08 (s, 2H), 3.83 (d, 4H), 1.80-1.68 (m, 2H), 1.60-1.26 (m, 18 H), 0.96-0.90 (m, 12H); 13C NMR (CDCl3, 75 MHz): δ 150.28, 118.29, 111.15, 72.64, 39.52, 30.53, 29.11, 23.96, 23.10, 14.14, 11.23. 2,5-Bis(2-ethylhexyloxy)-4-bromo-iodobenzene (24). A solution of 23 (28 g, 56 mmol), in THF (225 mL) was cooled to –90 °C. n-BuLi in hexane (22.74 mL, 2.5 M) was added at such a rate that the internal temperature did not exceed –80 °C. This cold solution was added via a cannula to 1,2-diiodoethane (17.05 g, 60.5 mmol) in THF (113 mL) at –60 °C. After addition, the cooling bath was removed, and the reaction mixture was stirred for 30 min. The color changed from yellowish to brown. The reaction mixture was washed with saturated aqueous Na2S2O5 untill it lost the color. The water phase was extracted three times with diethyl ether. The combined organic phases were dried over MgSO4 and the solvent removed in vacuo. The crude product was purified by flash chromatography (silica gel, hexane, Rf = 0.5). Evaporation of the solvent yielded 22.10 g (72%) of a colorless oil. 1H NMR (CDCl3; 400 MHz): δ 7.27 (s, 1H), 6.98 (s, 1H), 3.85-3.81 (m, 4H), 1.80-1.70 (m, 2H), 1.60-1.48 (m, 16H), 0.96-0.90 (m, 12H); 13C NMR (CDCl3, 100 MHz): δ 152.04, 150.06, 123.52, 116.25, 112.01, 84.22, 72.24, 72.01, 39.22, 39.20, 30.30, 30.24, 28.84, 28.90, 23.75, 23.68, 22.84, 13.93, 13.90, 11.03, 11.01. 2,5-Bis(2-ethylhexyloxy)-4-trimethylsilylethynyl-bromobenzene (25). To a degassed solution of 24 (22 g, 40.77 mmol) in Et3N (123 mL), Pd(PPh3)2Cl2 (0.28 g, 0.4 mmol) and CuI (0.15 g, 0.8 mmol) were added, while cooling in an ice bath. (Trimethylsilyl)acetylene (5.76 mL, 40.97 mmol) was added and the mixture was stirred at 0 °C. After 50 min the reaction was quantitative according to GC-MS. The solvent was removed in vacuo and the residue was dissolved in water and diethyl ether. The water phase was extracted three times with diethyl ether. The combined organic phases were washed with saturated aqueous NH4Cl and dried over MgSO4. The solvent was removed and the residue was flash chromatographed (silica gel, hexane, Rf = 0.25). The product was obtained as a colorless oil (17.11 g, 82%). 1H NMR (CDCl3; 300 MHz): δ 7.03 (s, 1H), 6.93 (s, 1H), 3.88-3.77 (m, 4H), 1.77-1.69 (m, 2H), 1.61-1.25 (m, 16H), 0.96-0.90 (m, 12H), 0.24 (s, 9H); 13C NMR (CDCl3, 75 MHz): δ 154.89, 149.40, 117.62, 117.55, 113.52, 112.26, 100.68, 98.99, 72.35, 71.94, 39.61, 39.47, 30.47, 29.10, 23.89, 23.05, 14.06, 11.25, 11.14,-0.07. 2,5-Bis(2-ethylhexyloxy)-4-ethynyl-bromobenzene (26). To a solution of 25 (5.49 g, 10.7 mmol) in dry THF was added 1 M tetrabuthylammonium fluoride in THF (10.79 mL). The reaction mixture was stirred for 1 min and subsequently filtrated over silica gel using chloroform as eluent. The solvent was removed in vacuo yielding 4.68 g (100%) of a slightly colored oil which was used without further purification: 1H NMR (CDCl3; 300 MHz): δ 7.07 (s, 1H), 6.96 (s, 1H), 3.85-3.82 (m, 4H), 3.26 (s, 1H), 1.77-1.69 (m, 2H), 1.61-1.25 (m, 16H), 0.96-0.90 (m, 12H). 2,5-Bis(2-ethylhexyloxy)-4-trimethylsilylethynyl-iodobenzene (27). Following a similar procedure as for the synthesis of compound 24, the reaction of 25 (8.5 g, 16.6 mmol) in THF (100 mL) with 2.5 M BuLi in hexane (6.7 mL) and 1,2-diiodoethane (5.52 g, 19.6 mmol) in THF (50 mL) gave, after chromatography (flash silica gel, hexane, Rf = 0.23), 27 (7.2 g, 77%, purity~ 90%). 1H NMR (CDCl3; 400 MHz): δ 7.24 (s, 1H), 6.93 (s, 1H), 3.77-3.50 (m, 4H), 1.77-1.69 (m, 2H), 1.61-1.25 (m, 16H), 0.96-0.90 (m, 12H), 0.24 (s, 9H); 13C NMR (CDCl3, 75 MHz): δ 155.03, 151.70, 123.38, 115.88, 113.23, 100.85, 99.21, 87.65, 72.12, 71.95, 39.62, 39.46, 30.53, 3047, 29.10, 29.06, 23.96, 23.50, 14.09, 11.27, 11.18, -0.08.
Polyacetylenes with pendant donor-acceptor dyads
145
4-[4-Trimethylsilylethynyl-2,5-bis(2-ethylhexyloxy)phenylethynyl]-2,5-bis(2-ethylhexyloxy)bromobenzene (28). To a degassed solution of 26 (4.71 g, 10.78 mmol) and 27 (5.5 g, 10.78 mmol) in Et3N (48 mL), Pd(PPh3)2Cl2 (76 mg, 0.1 mmol) and CuI (41 mg, 0.21 mmol) were added, while cooling in an ice bath. The mixture was stirred at 0 °C for 3.5 h. The solvent was removed in vacuo and the residue was dissolved in water and diethyl ether. The water phase was extracted three times with diethyl ether. The combined organic phases were washed with saturated aqueous NH4Cl, dried over MgSO4 and the solvent evaporated in vacuo. The residue was extensively purified by flash column chromatography (silica gel, pentane/diethyl ether, 100:0 to 99:1 Rf = 0.1-0.5). The product was obtained as a light yellow oil (5.9 g, 67). 1H NMR (CDCl3; 300 MHz): δ 7.09 (s, 1H), 6.98 (s, 1H), 6.93 (s, 1H) 6.92 (s, 1H), 3.92-3.78 (m, 8H), 1.79-1.72 (m, 4H), 1.63-1.37 (m, 16H), 1.32-1.30 (m, 16H), 0.97-0.84 (m, 24H), 0.26 (s, 9H); 13C NMR (CDCl3, 100 MHz): δ 154.35, 154.20, 153.44, 149.57, 117.82, 117.20, 117.02, 116.33, 114.47, 113.40, 113.12, 112.90, 101.22, 99.84, 90.81, 90.47, 72.28, 72.20, 71.88, 71.64, 39.63, 39.56, 39.54, 39.46, 30.60, 30.56, 30.50, 30.47, 29.09, 29.08, 29.02, 24.07, 24.03, 23.87, 23.07, 23.05, 14.09, 14.06, 11.30, 11.25, 11.11, -0.05. MALDI-TOF MS (Mw = 866.20) m/z = 866.38 [M]+. 4-[4-Ethynyl-2,5-bis(2-ethylhexyloxy)phenylethynyl]-2,5-bis(2-ethylhexyloxy)bromobenzene (29). To a solution of 28 (1.76 g, 2 mmol) in dry THF was added 1 M tetrabuthylammonium fluoride in THF (2 mL). The reaction mixture was stirred for 1 min and subsequently filtrated over silica gel using chloroform as eluent. The solvent was removed in vacuo. The product was purified by column chromatogragphy (flush silica gel, hexane/toluene, 4:1, Rf = 0.2). The product (1.55 g, 95%) is light yellow oil: 1H NMR (CDCl3; 300 MHz): δ 7.09 (s, 1H), 6.98 (s, 1H), 6.96 (s, 2H), 3.90-3.82 (m, 8H), 1.79-1.73 (m, 4H), 1.60-1.26 (m, 32H), 0.98-0.84 (m, 24H); 13C NMR (CDCl3, 75 MHz): δ 154.44, 154.24, 153.50, 149.61, 117.86, 117.54, 117.26, 116.70, 114.90, 113.22, 112.85, 112.48, 90.87, 90.26, 82.14, 80.05, 72.32, 72.20, 72.07, 71.97, 39.54, 39.41, 30.58, 30.48, 29.08, 29.03, 24.04, 23.87, 23.03, 14.03, 11.28, 11.10. 4-[4-Trimethylsilylethynyl-2,5-bis(2-ethylhexyloxy)phenylethynyl]-2,5-bis(2-ethylhexyloxy)iodobenzene (30). Following a similar procedure as for the synthesis of compound 24, the reaction of 28 (3.14 g, 3.55 mmol) in THF (100 mL) with 2.5 M BuLi in hexane (1.56 mL) and 1,2-diiodoethane (1.10 g, 3.91 mmol) in THF (50 mL) gave, after chromatography (flash silica gel, toluene/pentane, 1:4, Rf = 0.42), 30 (2.1 g, 63%, purity~85%). 1H NMR (CDCl3; 400 MHz): δ 7.29 (s, 1H), 6.93 (s, 1H), 6.92 (s, 1H), 6.87 (s, 1H), 3.90-3.81 (m, 8H), 1.79-1.71 (m, 4H), 1.54-1.26 (m, 32H), 0.97-0.85 (m, 24H), 0.26 (s, 9H); 13C NMR (CDCl3, 75 MHz): δ 154.34, 153.44, 151.89, 123.68, 117.01, 116.34, 115.50, 114.48, 113.90, 113.45, 101.22, 99.81, 91.00, 90.67, 87.18, 72.22, 72.05, 71.85, 71.62, 39.63, 39.56, 39.47, 30.57, 30.51, 30.49, 29.08, 29.03, 24.03, 23.93, 23.88, 23.03, 14.06, 11.30, 11.24, 11.14, -0.07. MALDI-TOF MS (Mw = 913.20) m/z = 912.47 [M]+. 4-<4-{4-[4-Trimethylsilylethynyl-2,5-bis(2-ethylhexyloxy)phenylethynyl]-2,5-bis(2-ethylhexyloxy) phenylethynyl}-2,5-bis(2-ethylhexyloxy)phenylethynyl >-2,5- bis(2-ethylhexyloxy)bromobenzene (31). To a degassed solution of 29 (1.20 g, 1.5 mmol) and 30 (1.4 g, 1.5 mmol) in Et3N (15 mL), Pd(PPh3)2Cl2 (52 mg, 0.075 mmol) and CuI (14 mg, 0.075 mmol) were added, while cooling in an ice bath. The mixture was stirred at 0 °C for 2.5 h. The solvent was removed in vacuo and the residue was dissolved in water and diethyl ether. The water phase was extracted three times with diethyl ether. The combined organic phases were washed with saturated aqueous NH4Cl, dried over MgSO4and the solvent evaporated in vacuo. The residue was purified with flash column chromatography (silica gel, toluene/hexane, 3:7 Rf = 0.25). The product was obtained as a greenish fluorescent oil (1.4 g, 60%). 1H NMR (CDCl3; 400 MHz): δ 7.12 (s, 1H), 7.01 (s,3H), 7.01 (s, 2H) 6.96 (s, 1H), 3.94-3.83 (m, 16H), 1.88-1.72 (m, 8H), 1.70-1.40 (m, 32H), 1.40-1.25 (m, 32H), 1.01-0.88 (m, 48H), 0.27 (s, 9H); 13C NMR (CDCl3, 100 MHz): δ154.33, 154.18, 153.67, 153.64, 1253.45, 149.56, 117.79, 117.17, 117.00, 116.62, 116.33, 114.57, 114.22, 114.15, 114.10, 113.39, 113.07, 112.94, 101.23, 99.81, 91.52, 91.41, 90.83, 90.60, 72.22, 72.14, 71.87, 71.81, 71.59, 39.62, 39.57, 39.53, 39.44, 30.59, 30.48, 30.45, 29.07, 28.99, 24.04, 23.86, 23.05, 23.00, 14.06, 14.02, 11.27, 11.22, 11.08, -0.09. MALDI-TOF MS (Mw = 1579.31) m/z = 1579.12 [M]+. N-[(S)-1-Isobutyl-2-<4-[4-{4-[4-Trimethylsilylethynyl-2,5-bis(2-ethylhexyloxy)phenylethynyl]-2,5-bis(2-ethylhexyloxy)phenylethynyl}-2,5-bis(2-ethylhexyloxy)phenylethynyl>-2,5-bis(2-ethylhexyloxy)phenylethynyl]phenyloxy>-ethyl]-N’-(1-ethylpropyl)- 3,4,9,10-perylenebis(dicarboximide) (32). A degassed solution of 31 (0.3 g, 0.19 mmol), 10 (0.12 g, 0.19 mmol), Pd(PPh3)2Cl2 (0.01 g, 0.014 mmol), CuI (0.002 g, 0.01 mmol) in anhydrous toluene/triethylamine (30 mL, 5:1) was heated at 50 oC for 16 h. The crude
Chapter 6
146
reaction mixture was filtered through Celite, evaporated to dryness, and chromatographied (flash silica, CH2Cl2/pentane, 3:2 to 4:1, Rf = 0.23-0.4) to afford a red solid (0.144 g, 34 %). 1H NMR (CDCl3, 400 MHz): δ 8.64-8.45 (m, 8H), 7.41 (d, 2H), 6.99 (s, 4H), 6.96 (s, 1H), 6.94 (s, 3H), 6.91 (d, 2H), 5.81-5.73 (m, 1H), 5.10-5.03 (m, 1H), 4.82 (dd, 1H), 4.40 (dd, 1H), 3.94-3.80 (m, 16H), 2.32-2.45 (m, 3H), 2.01-1.94 (m, 2H), 1.86-1.25 (m, 66H), 1.06-0.82 (m, 60H), 0.26 (s, 9H); 13C NMR (CDCl3, 100 MHz): δ 163.51, 158.66, 153.65, 151.32, 153.45, 131.48, 131.11, 132.88, 131.21, 129.43, 126.18, 122.95, 116.99, 116.62, 115.90, 114.78, 114.57, 114.24, 114.10, 113.84, 113.34, 101.23, 99.82, 94.82, 91.54, 91.37, 84.49, 71.86, 71.59, 68.69, 57.71, 51.45, 39.55, 38.57, 30.74, 30.57, 30.47, 29.06, 25.57, 24.98, 24.03, 23.94, 2.85, 23.04, 22.45, 14.034, 11.36, 11.27, -0.069 MALDI-TOF MS (Mw = 2159.17) m/z = 2158.31[M]+. N-[(S)-1-Isobutyl-2-<4-[4-{4-[4-Ethynyl-2,5-bis(2-ethylhexyloxy)phenylethynyl]-2,5-bis(2-ethylhexyloxy)phenylethynyl}-2,5-bis(2-ethylhexyloxy)phenylethynyl>-2,5-bis(2-ethylhexyloxy)phenylethynyl]phenyloxy>-ethyl]-N’-(1-ethylpropyl)-3,4,9,10-perylenebis(dicarboximide) (33). To a solution of 32 (76 mg, 0.035 mmol) and tetrabutylammonium fluoride impregnated on silica (0.042 mL of a 1M solution, 0.042 mmol) in 10 mL THF was stirred for 0.5 h under argon. The reaction mixture was filtered through Celite and evaporated to dryness. The product was obtained as a red powder (70 mg, 96%) and was used without further purification. The complete deprotection of the ethynyl bond was confirmed by 1H-NMR spectroscopy as the disappearance of the singlet at 0.2 ppm and the appearance of a singlet for the ethynylinic proton at 3.2 ppm. N-[(S)-1-Isobutyl-2-<4-[4-{4-[4-(4-Trimethylsilylethynyl-phenylethynyl)-2,5-bis(2-ethylhexyloxy)phenylethynyl]-2,5-bis(2-ethylhexyloxy)phenylethynyl}-2,5-bis(2-ethylhexyloxy)phenylethynyl]-2,5-bis(2-ethylhexyloxy)phenylethynyl>phenyloxy>-ethyl]-N’-(1-ethylpropyl)-3,4,9,10-perylenebis(dicarboximide) (34). A degassed solution of 33 (70 mg, 0.035 mmol), 4-iodotrimethylsylilphenylacetylene (34 mg, 0.123 mmol), Pd(PPh3)2Cl2 (2 mg, 0.003 mmol), CuI ( 0.3 mg, 0.002 mmol) in anhydrous toluene/triethylamine (6 mL, 5:1) was heated at 50 oC for 16 h. The crude reaction mixture was filtered through Celite, evaporated to dryness, and chromatographied (flash silica, CH2Cl2/pentane, 1:1 to 2:1, Rf = 0.1-0.3) to afford a red solid (30 mg, 40 %). 1H NMR (CDCl3, 400 MHz):δ 8.72-8.58 (m, 8H), 7.44 (s, 4H), 7.39 (d, 2H), 6.99 (s, 4H), 6.98 (s, 2H), 6.96 (s, 1H), 6.95 (s, 1H), 6.87 (d, 2H), 5.90-5.70 (m, 1H), 5.11-5.01 (m, 1H), 4.78 (dd, 1H), 4.38 (dd, 1H), 3.96-3.82 (m, 16H), 2.32-2.23 (m, 3H), 1.99-1.92 (m, 2H), 1.86-1.73 (m, 8H), 1.68-1.25 (m, 66H), 1.05-0.84 (m, 60H), 0.26 (s, 9H); 13C NMR (CDCl3, 100 MHz): δ 163.51 (broad signal), 158.66, 153.87, 153.67, 153.60, 134.72, 134.34, 132.88, 132.04, 131.85, 131.24, 129.61, 129.55, 126.43, 126.38, 123.64, 123.12, 122.97, 122.74, 116.63, 116.59, 116.54, 116.51, 116.47, 115.88, 114.78, 114.65, 114.31, 114.26, 114.13, 114.09, 113.86, 113.37, 104.71, 96.22, 94.83, 94.38, 91.74, 91.62, 91.58, 91.52, 91.42, 91.37, 88.16, 84.84, 71.96, 71.88, 68.65, 57.71, 51.45, 39.58, 38.55, 30.65, 30.59, 29.08, 25.57, 24.99, 54.03, 23.97, 23.95, 23.14, 23.06, 22.44, 14.06, 11.34, 11.31, 11.28, 11.23, 11.21, -0.10. MALDI-TOF MS (Mw = 2259.29) m/z = 2258.69[M]+. N-[(S)-1-Isobutyl-2-<4-[4-{4-[4-(4-Ethynyl-phenylethynyl)-2,5-bis(2-ethylhexyloxy)phenylethynyl]-2,5-bis(2-ethylhexyloxy)phenylethynyl}-2,5-bis(2-ethylhexyloxy)phenylethynyl]-2,5-bis(2-ethylhexyloxy)phenylethynyl]phenyloxy>-ethyl]-N’-(1-ethylpropyl)-3,4,9,10-perylenebis(dicarboximide) (35). To a solution of 34 (28 mg, 0.012 mmol) and tetrabutylammonium fluoride impregnated on silica (0.013 mL of a 1M solution, 0.013 mmol) in 5 mL THF was stirred for 0.5 h under argon. The reaction mixture was filtered through Celite and evaporated to dryness. The product was obtained as a red powder (26 mg, 99%) and was used without further purification. The complete deprotection of the ethynyl bond was confirmed by 1H-NMR spectroscopy as the disappearance of the singlet at 0.26 ppm and the arising of the signal for the ethynylinic proton at 3.2 ppm. Synthesis of the polymers: Polyphenylacetylene with pendant perylenes (11). To a mixture of the ethynyl terminated compound 10 (37 mg, 0.06mmol) and triethylamine (0.8mL) in toluene (20 mL) at room temperature [Rh(nbd)Cl]2 (2.6 mg, 0.006 mmol) was added under an argon atmosphere. The reaction mixture was stirred for 24 h and the polymer was isolated by precipitation into methanol as a red polymer. The polymer was insoluble in any common solvent.
Polyacetylenes with pendant donor-acceptor dyads
147
Polyphenylacetylene with pendant OPVs and perylenes (1). To a mixture of monomers 10 (29 mg, 0.044 mmol) and 12 (30 mg, 0.045 mmol) in triethylamine (0.8 mL) and toluene (20 mL), [Rh(nbd)Cl]2 (2.6 mg, 0.006 mmol) was added under an argon atmosphere. The reaction was stirred for 24 h and subsequently poured into stirring methanol. The polymer was isolated from the cyclic trimers side products using preparative size exclusion chromatography (Biobeads, SXIII, CH2Cl2) and obtained as a red solid (21 mg, 36%). SEC (chloroform/methanol, 99/1): Mw =14.6 kg/mol, PDI = 1.6. Polyphenylacetylene with pendant OPV-PERY dyads (2). To a solution of ethynyl monomer 20 (44 mg, 0.038 mmol) dissolved in toluene (13 mL) and triethylamine (0.6 mL), [Rh(nbd)Cl]2 (2 mg, 0.004 mmol) was added. The reaction was stirred for 16 h and subsequently poured into stirring methanol. The cyclic trimers were removed by precipitation in THF. The polymer was obtained as a red solid (26 mg, 60 %). SEC (chloroform/methanol, 99/1): Mw =45.8 kg/mol, PDI = 1.9. Polyphenylacetylenes with pendant OPE-PERY dyads (3). To a solution of ethynyl monomer 35 (26 mg, 0.012 mmo) dissolved in toluene (4 mL) and triethylamine (0.15 mL), [Rh(nbd)Cl]2 (0.55 mg, 0.001 mmol) was added. The reaction was stirred for 16 h and subsequently poured into stirring methanol. The cyclic trimers were removed by precipitation in THF. The polymer was obtained as a red solid (7 mg, 30 %). SEC (chloroform/methanol, 99/1): Mw = 53.5 kg/mol, PDI = 1.5. Absorption and Photoluminescence. UV/visible/near-IR absorption spectra were recorded on a Perkin Elmer Lambda 900 spectrophotometer. Fluorescence spectra were recorded on a Perkin Elmer LS 50B spectrometer, using a 4 nm bandwidth and optical densities of the solutions of 0.1 at the excitation wavelength. Transient subpicosecond photoinduced absorption. Solutions in the order of 10-5 M were excited at 450 nm, i.e. providing primarily excitation of the donor part within the molecules. 6.8 References and notes 1 (a) Yu, G.; Gao, J.; Hummelen, J. C.; Wudl, F.; Heeger, A. J. Science 1995, 270, 1789; (b) Halls, J. J. M.;
Walsh, C. A.; Greenham, N. C.; Marseglia, E. A.; Friend, R. H., Moratti, S. C.; Holmes, A. B. Nature
1995, 376, 498; (c) Shaheen, S. E.; Brabec, C. J.; Padinger, F.; Fromherz, T.; Hummelen, J. C.; Sariciftci,
N. S. Appl. Phys. Lett. 2001, 78, 841; (d) Schmidt-Mende, L.; Fechtenkötter, A.; Müllen, K.; Moons, E.;
Friend, R. H.; MacKenzie, J. D. Science, 2001, 293, 1119.
2 (a) Shinohara, K.; Yasuda, S.; Kato, G.; Fujita, M.; Shigekawa, H. J. Am. Chem. Soc. 2001, 123, 3619.
(b) Yashima, E.; Katsuhiro, M.; Okamoto, Y. Nature 1999, 399, 449.
3 Schenning, A. P. H. J.; Fransen, M.; van Duren, J. K. L.; van Hal, P. A., Janssen, R. A. J.; Meijer, E. W.
