Decelerating Charge Recombination Using Fluorinated Porphyrins in N,N-Bis(3,4,5-trimethoxyphenyl)aniline - Aluminum(III) Porphyrin - Fullerene Reaction Center Models

Article abstract of DOI:10.1021/jacs.0c01574

In supramolecular reaction center models, the lifetime of the charge-separated state depends on many factors. However, little attention has been paid to the redox potential of the species that lie between the donor and acceptor in the final charge separat

Full text of DOI:10.1021/jacs.0c01574

Subscriber access provided by BIU Pharmacie | Faculté de Pharmacie, Université Paris V
Article
Decelerating Charge Recombination Using Fluorinated
Porphyrins in N,N-bis(3,4,5-trimethoxyphenyl)aniline –
Aluminum(III) Porphyrin – Fullerene Reaction Center Models
Niloofar Zarrabi, Sairaman Seetharaman, Subhajyoti Chaudhuri, Noah Holzer,
Victor S. Batista, Art van der Est, Francis D'Souza, and Prashanth K Poddutoori
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.0c01574 • Publication Date (Web): 28 Apr 2020
Downloaded from pubs.acs.org on April 28, 2020
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 33
Journal of the American Chemical Society
Decelerating Charge Recombination Using Fluorinated Porphyrins in N,N-bis(3,4,5-
1
2
3
trimethoxyphenyl)aniline – Aluminum(III) Porphyrin – Fullerene Reaction Center Models
4
5
6
7
8
9
Niloofar Zarrabi,a Sairaman Seetharaman,b Subhajyoti Chaudhuri,c Noah Holzer,a Victor S. Batista,*,c Art van
der Est,*,d Francis D’Souza,*,b Prashanth K. Poddutoori*, a
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
aDepartment of Chemistry & Biochemistry, University of Minnesota Duluth, 1038 University Drive, Duluth,
b
Minnesota 55812, United States. Department of Chemistry, University of North Texas, 1155 Union Circle, #
305070, Denton, Texas 76203-5017, United States. cDepartment of Chemistry, Yale University, P.O. Box 208107,
New Haven, Connecticut 06520-8107, United States. dDepartment of Chemistry, Brock University, 1812 Sir Isaac
Brock Way, St. Catharines, ON, L2S 3A1, Canada.
ABSTRACT. In supramolecular reaction center models, the lifetime of the charge-separated state depends on
many factors. However, little attention has been paid to the redox potential of the species that lie between the
donor and acceptor in the final charge separated state. Here, we report on a series of self-assembled aluminum
porphyrin-based triads that provide a unique opportunity to study the influence of the porphyrin redox potential
independently of other factors. The triads, BTMPA-ImAlPorFn-Ph-C60 (n = 0, 3, 5), were constructed by linking
the fullerene (C60) and bis(3,4,5-trimethoxyphenyl)aniline (BTMPA) to the aluminum(III) porphyrin. The
porphyrin (AlPor, AlPorF3 or AlPorF5) redox potentials are tuned by substitution of phenyl (Ph), 3,4,5-
trifluorophenyl (PhF3) or 2,3,4,5,6-pentafluorophenyl (PhF5) groups in its meso positions. The C60 and BTMPA
units are bound axially to opposite faces of the porphyrin plane via covalent and coordination bonds, respectively.
Excitation of all of the triads results in sequential electron transfer that generates the identical final charge
separated state, BTMPA+-ImAlPorFn-Ph-C60, which lies energetically 1.50 eV above the ground state.
Despite the fact that the radical pair is identical in all of the triads, remarkably, the lifetime of BTMPA+-
ImAlPorFn-Ph-C60 radical pair was found to be very different in each of them, that is, 1240, 740 and 56 ns for
BTMPA-ImAlPorF5-Ph-C60, BTMPA-ImAlPorF3-Ph-C60 and BTMPA-ImAlPor-Ph-C60, respectively.
These results clearly suggest that the charge recombination is an activated process that depends on the midpoint
potential of the central aluminum(III) porphyrin (AlPorFn).
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 33
INTRODUCTION
1
The generation of long-lived high-energy charge-separated states is an essential component of the conversion of
sunlight energy into chemical energy in natural photosynthesis.1–4 To achieve energy conversion with a high
quantum yield, the photosynthetic systems use a series of electron transfer reactions along a chain of redox active
cofactors. The sunlight energy collected by the light harvesting antenna systems is funneled to the reaction center
complex where an extremely fast and highly efficient initial charge separation occurs. The charge separation is
then stabilized by a series of secondary electron transfer steps that ultimately generate species that are long lived
enough to allow oxidation and reduction of soluble donors and acceptors to occur on opposite sides of the
2
3
4
5
6
7
8
9
10
11
12
13
14 photosynthetic membrane. In oxygenic photosynthesis, this process generates the energy-rich compounds ATP,
15
NADPH and O2 that are used to drive respiration, carbon fixation and a host of other biochemical processes.
Mimicking the initial steps of photosynthetic energy conversion by using synthetic models is important to further
our understanding of the intricate details of electron transfer process in natural photosynthesis. Most importantly,
research in this area holds promise for technological advances in solar energy conversion and storage for future
energy needs.5–13 In this regard, many reaction center models have been designed and studied to mimic the
photoinduced charge separation of natural systems.9,10,14–25 However, unlike the reaction centers, charge
recombination is an important factor because the number of donors and acceptors is smaller, and the complexes
are not embedded in a membrane protein. In these model systems, the rate of charge separation and recombination
has been studied as a function of the driving force,26–28 reorganization energy,29–32 and electronic coupling,33,34
which depend on the distance between donor and acceptor units,30,35,36 the bridge and medium connecting them,37–
39 molecular conformation,32,40 etc. By optimizing these factors through careful design of the complexes, sequential
charge transfer yielding the desired long-lived charge separated states are achieved.41–44
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In many of these synthetic models, substituted porphyrin compounds are used because they share similar
structural, physical and chemical properties with the chlorophyll chromophores of the natural photosystems.45,46
In the artificial complexes, porphyrins are covalently or non-covalently connected to variety of redox active units.
In most cases, the donor and acceptor centers are connected to the periphery of the porphyrin ring either through
the meso- or pyrrolic (-) positions. This strategy results in complexes in which the charge separation occurs in
the porphyrin plane and is sometimes referred to as horizontal electron transfer. A smaller number of model
compounds have been studied in which “vertical” or “axial” electron transfer occurs perpendicular to the porphyrin
ring.27,30,47–60 One of the main reasons for this limitation is the use of transition metal ions in the porphyrin cavity
that do not allow more than one axial unit.
The most important factor governing biological electron transfer is the electronic coupling between the
donor and acceptor, which leads to the linear dependence of the logarithm of the rate on the distance as described
by the Moser-Dutton relationship.61 In molecular systems the electronic coupling depends strongly on the nature
of the bonding and the arrangement of the orbitals involved in the electron transfer. Therefore, it is fundamentally
2
ACS Paragon Plus Environment

Page 3 of 33
Journal of the American Chemical Society
important to consider the axial arrangement as an alternative to the more common horizontal arrangement for the
artificial model compounds. Moreover, the axial placement of the donor and acceptor units on opposite faces of
the porphyrin ensures that they are spatially well separated and that unwanted interactions are minimized. With
this objective recently we and others have reported several axial-bonding type reaction center models by exploiting
the unique properties of the main group porphyrins.35,62–70
1
2
3
4
5
6
7
8
Among the main group porphyrins, aluminum(III) porphyrins are unique because they can form two
different types of axial bonds: i) covalent Al—O bonds formed through condensation with carboxylic acids and
alcohols to form ester and ether linkages, respectively,71–74 and ii) Al—X coordination bonds (X = N, O) through
9
10
11
12
13
14 Lewis acid-base interactions between Lewis bases such as pyridine and imidazole and the Al center.35,64 These
15
two types of bonding allow a wide variety of donors and acceptors to be attached on opposite faces of the porphyrin
ring. Additionally, AlPor is highly fluorescent indicating that the excited singlet state lifetime is relatively long,
which is advantageous if electron transfer is to take place. AlPor also has a rich redox chemistry with two reversible
oxidation and reduction processes, allowing it to easily act as both an electron donor and acceptor. Furthermore,
the optical and redox properties of AlPor can be tailored by introducing appropriate substituents in meso- or -
positions of the porphyrin ring for desired applications. The combination of these properties makes AlPor a unique
candidate for constructing ‘axial-bonding’ type multi-component Donor-AlPor-Acceptor systems.
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
An important factor in determining the lifetime of the charge-separated state is the nature of the charge
recombination reaction. In designing such systems, it is usually assumed that the direct recombination occurs
between the radical ions of the final charge separated state and the rate is therefore determined primarily by the
electronic coupling between them and to a lesser extent on their midpoint potentials and the reorganization energy.
However, stepwise electron transfer or a hopping mechanism is also possible in which case the recombination
depends on the energies of the intermediate states. To optimize the lifetime of the final charge separated state, it
is therefore important to be able to distinguish between these possible recombination mechanisms.
Here we report three novel aluminum(III) porphyrin centered triads that allow this question to be addressed
by taking advantage of the properties of aluminum(III) porphyrin. The structures of the three triads (BTMPA-
ImAlPor-Ph-C60, BTMPA-ImAlPorF3-Ph-C60 and BTMPA-ImAlPorF5-Ph-C60) are shown in Scheme 1.
The aluminum(III) porphyrin acts as the primary electron donor/acceptor, fullerene is the primary/secondary
electron acceptor, and the imidazole-appended bis(trimethoxyphenyl)aniline derivative (BTMPA) acts as the
primary/secondary electron donor. We have selected fullerene as an electron acceptor due to its small
reorganization energy for electron transfer.75–77 As the secondary donor, we have designed a new compound
consisting of: (i) electron-rich N,N-bis(3,4,5-trimethoxyphenyl)aniline (BTMPA) and (ii) the Lewis base
imidazole (Im) attached to one another through a covalent linkage. The redox properties of aluminum(III)
porphyrin are tuned by substitution of its meso positions with either phenyl (Ph), 3,4,5-trifluorophenyl (PhF3) or
2,3,4,5,6-pentafluorophenyl (PhF5) groups. The reported systems are designed in such a way that following
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 33
excitation of aluminum(III) porphyrin, stepwise electron transfer along the axial direction is energetically
1
2
3

favorable and produces an identical final radical pair, BTMPA+-aluminum(III) porphyrin-C60 in all cases. In
these systems, the factors involving the donor BTMPA and acceptor C60 such as distance between them, the nature
of the bridge, their contributions to the reorganization and Gibbs free energies are kept constant. Thus, this design
allows the role of the central porphyrin aluminum(III) porphyrin in the charge separation and recombination
dynamics to be studied independently of other factors. Such systems are rarely explored,26,78 and to the best of our
knowledge this first such system in which the charge separation and recombination in the axial direction is
investigated as a function of the redox potential of the sensitizer. Time-resolved spectroscopy reveals that the triads
depicted in Scheme 1 undergo light-induced multistep electron transfer. We will show that stabilization of the
initial charge separation by secondary electron transfer occurs, and remarkably the stabilization as measured by
the overall lifetime of the charge separation is more pronounced in those triads in which aluminum(III) porphyrin
is substituted with electron withdrawing PhF3 and PhF5 groups which make it a better electron acceptor.
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
EXPERIMENTAL SECTION.
Synthesis. All chemicals, solvents and chromatographic materials were obtained from Aldrich Chemicals, Fisher
Chemicals or Alfa Aesar and were used as received. The synthesis details of the precursor porphyrins 5,10,15,20-
tetraphenylporphyrinatoaluminum(III)
hydroxide
(AlPor-OH),35,65
5,10,15,20-tetrakis(3,4,5-
trifluorofluorophenyl)porphyrinatoaluminum(III) hydroxide (AlPorF3-OH)64,79 and their corresponding reference
compounds AlPor-Ph35,65 and AlPorF3-Ph64,79 are reported elsewhere. The precursor 5,10,15,20-
tetrakis(pentafluorophenyl)porphyrin (H2PorF5) and its aluminum(III) derivative (AlPorF5-OH) were synthesized
using modified literature procedures,79 for details see the Supporting Information and Scheme S1. The preparation
of the dyad AlPor-Ph-C60 and the reference fullerene (C60COOMe) has been reported previously.65 The electron
donor 4-(1H-imidazol-1-yl)-N,N-bis(3,4,5-trimethoxyphenyl)aniline (BTMPA-Im, Scheme S2), the dyads
40 (AlPorF3-Ph-C60 and AlPorF5-Ph-C60, Scheme 2) and one of the reference compounds (AlPorF5-Ph Scheme S1)
41
42
were prepared as described in the Supporting Information. The self-assembled dyads (BTMPAAlPorFn-Ph, n =
43
44
0, 3, 5) and triads (BTMPAAlPorFn-Ph-C60, n = 0, 3, 5) were constructed by coordinating the BTMPA-Im with
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
AlPorFn-Ph and AlPorFn-Ph-C60, respectively, see Schemes 1 and 2. Because these self-assembled dyads and triads
are formed in a solution equilibrium they cannot be isolated easily for elemental and mass analysis.
4
ACS Paragon Plus Environment

Page 5 of 33
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 1. Construction of the dyads AlPorFn-Ph-C60 and triads BTMPA-Im AlPorFn-Ph-C60, n = 0, 3, 5.
Reaction conditions: (i) C60-Ph-COOH, dry CH2Cl2 and CH3OH, stirring at room temparature under nitrogen for
12-18 h. (ii) BTMAP-Im, o-dichlorobenzene (o-DCB), under titration conditions.
Scheme 2. Construction of the dyads BTMPA-ImAlPorFn-Ph, n = 0, 3, 5. Reaction conditions: (i) Ph-COOH,
dry CH2Cl2 and CH3OH, stirring at room temparature under nitrogen for 12-18 h. (ii) BTMAP-Im, o-DCB, under
titration conditions.
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 33
RESULTS AND DISCUSSION
1
Synthesis. The newly designed secondary electron donor BTMPA-Im, consisting of the electron donor BTMPA
bound to the Lewis base imidazole was synthesized according to Scheme S2. The synthesis and self-assembly of
dyads and triads are shown in Schemes 1 and 2. In the dyads AlPorFn-Ph-C60, fullerene is bound covalently to the
porphyrin by condensation of AlPorFn-OH (n = 0, 3, 5) with C60-Ph-COOH in quantitative yield (see Scheme 1).
The Al center in the dyad, AlPorFn-Ph-C60, acts as Lewis acid, therefore, Lewis bases such as imidazole appended
N,N-bis(3,4,5-trimethoxyphenyl)aniline (BTMPA-Im) can be bound via coordination bonding through the free
imidazole nitrogen to construct the supramolecular triads BTMPAAlPorFn-Ph-C60. A similar strategy was
2
3
4
5
6
7
8
9
10
11
12
13
14 employed to construct the dyads BTMPAAlPorFn-Ph starting from the reference compound AlPorFn-Ph (see
15
Figure 2).
Preliminary characterization of the covalently bound dyads and reference porphyrins was carried out by
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
ESI High Resolution mass spectrometry. The mass spectra of the dyads and reference porphyrin showed peaks,
which corresponds to the mass (m/z) of either [M+H]+ or [M – (PhCOO/C60-PhCOO)+CH3CN]+, for details see
experimental section in the Supporting Information as well as Figures S1-S3. The EI mass spectra of BTMPA-Im
is shown in Figure S4, and the spectra revealed [M]+ peak at m/z = 491.27 (calculated 491.2056). The melting
point studies indicate that all the dyads, AlPorFn-Ph-C60, and the reference compounds, AlPorF3-Ph and AlPorF5-
Ph, are stable up to 300C. The reference compound, AlPor-Ph and the secondary electron donor, BTMPA-Im,
have melting points of 293.5-294.1°C and 166.6-168.5°C, respectively. The IR spectra of AlPor-OH, C60-Ph-
COOH, AlPor-Ph and AlPor-Ph-C60 are shown in Figure S5. The AlPor-Ph-C60 spectrum reveals the peaks from
both AlPor and C60 units. The absence of the carboxylic acid C=O stretching peak at 1695 cm–1 and presence of a
characteristic C60 vibrational mode at 527 cm–1 confirms formation of the dyad AlPor-Ph-C60, i.e. C60 is connected
to AlPor through a carboxylate linker. Similar results were observed from the other two dyads, AlPorF3-Ph-C60
(Figure S6) and AlPorF5-Ph-C60 (Figure S7). The 1H NMR spectra of BTMPA-Im, AlPorF3-Ph-C60 and AlPorF5-
40 Ph-C60 are shown in Figures S8, S9 and S10, respectively. The data obtained from the spectra are summarized in
41
42
the experimental section of the supporting information. The upfield shift of the bridging phenyl protons a and b,
43
44
from 8.07 and 7.87 ppm in free C60-Ph-COOH65 to ~5.12 and ~6.80 ppm, respectively in the dyads is due to the
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
shielding effect of the porphyrin ring current at the center of the ring and confirms axial-bonding via the
carboxylate spacer. The protons from pyrrolidine ring appear at c (4.97 ppm), d and e (4.98, 4.25 ppm) and f (2.77
ppm) in free C60-Ph-COOH.65 In the dyad compounds, AlPorF3-Ph-C60 and AlPorF5-Ph-C60, these peaks were
shifted upfield to ~4.77, ~4.48, ~4.00 and ~2.40 ppm, respectively. Figure S11 shows the 1H NMR spectrum of a
1:1 mixture of AlPor-Ph-C60 and BTMPA-Im (bottom) along with the individual spectra of AlPor-Ph-C60 (middle)
and BTMPA-Im (top). In the coordination complex, the shielding due to the porphyrin ring also causes an upfield
shift of the BTMPA-Im protons on the imidazole unit, bridging phenyl, terminal phenyl as well as methoxy groups.
The magnitude of the shift depends on the distance of the protons from the porphyrin ring and the imidazole
6
ACS Paragon Plus Environment

Page 7 of 33
Journal of the American Chemical Society
protons display the greatest shift indicating that coordination occurs via the imidazole group. On the benzoate
bridging group to the fullerene, the protons, a, closest to the porphyrin ring show an increased upfield shift upon
coordination, suggesting that the aluminum(III) center lies out of the porphyrin plane in AlPor-Ph-C60 and is pulled
1
2
3
4
5
6
1
into the plane when BTMPA-Im coordinates. Similar H NMR titrations for AlPorFn-Ph-C60 (n = 3, 5) and
7
8
9
AlPorFn-Ph (n = 0, 3, 5) with BTMPA-Im give analogous results (data not shown). Thus, the NMR titrations
confirm the formation of the triads (BTMPA-ImAlPorFn-Ph-C60) and the dyads (BTMPA-ImAlPorFn-Ph) in
solution.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
UV−visible absorption spectroscopy. The UV−visible spectra of AlPorF3-Ph-C60 and its reference compounds
(AlPorF3-Ph and C60-Ph-COOMe) were measured in dichloromethane (CH2Cl2), and the spectra are shown in
Figure 1a. The band positions (Q-band and B- or Soret band) and their molar extinction coefficients are
summarized in Table 1. As shown in Figure 1a, the absorption spectrum of the dyad is essentially a linear
combination of its reference compounds and the positions and molar extinction coefficients (ε) of the porphyrin
bands are essentially unchanged in the dyad compared to the corresponding monomer porphyrins. Similar spectral
features were also observed from the dyads AlPor-Ph-C60 (Figure S12a) and AlPorF5-Ph-C60 (Figure S12b).
Overall, the absorption studies suggest that any interaction between the porphyrin (AlPorF3) and the linked
fullerene (C60) is sufficiently weak that it causes no discernable change in their electronic structure. This can be
attributed to their perpendicular orientation which decreases the electronic coupling between their respective π-
systems. The imidazole-appended donor, BTMPA-Im, has relatively weak and very broad absorption bands in the
ultraviolet region, Figure S12c.
Furthermore, the UV−visible spectra of dyads AlPorFn-Ph-C60 and their reference porphyrins, AlPorFn-Ph,
were measured in o-dichlorobenzene (o-DCB), and the spectra are shown in Figure S13. The observed spectral
trends and absorption profiles are consistent with CH2Cl2 data, although the C60 band in the dyads was obscured
40 due to the solvent cutoff. The absorption bands of AlPor at 550 and 410 nm have been chosen to excite the AlPor
41
42
for steady-state fluorescence and transient-absorption studies, respectively.
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
7
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 33
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
Figure 1. (a) Absorption spectra of AlPorF3-Ph (red) and AlPorF3-Ph-C60 (green), C60-Ph-COOMe (blue) in
CH2Cl2. Absorption titrations of (b) AlPor-Ph-C60 vs BTMPA-Im: BTMPA-Im was added up to 1.83 × 10−4 M in
10 μl (1.33 × 10−3 M) increments to 1 ml (4 × 10−6 M) solution of AlPor-Ph-C60; (c) AlPorF3-Ph-C60 vs BTMPA-
43 Im: BTMPA-Im was added up to 2.47 × 10−5 M in 10 μl (1.37 × 10−4 M) increments to 1 ml (4 × 10−6 M) solution
44
45
of AlPorF3-Ph-C60; (d) AlPorF5-Ph-C60 vs BTMPA-Im: BTMPA-Im was added up to 7.69 × 10−6 M in 5 μl (1.36
46
47
× 10−4 M) increments to 1 ml (4 × 10−6 M) solution of AlPorF5-Ph-C60. All the titrations were performed in o-
48
DCB. The inset shows the Benesi–Hildebrand plot.
49
50
51
52
53
54
55
56
57
58
59
60
8
ACS Paragon Plus Environment

Page 9 of 33
Journal of the American Chemical Society
Table 1. Optical and redox data of investigated compounds.
1
2
3
4
Potential [V vs Fc]a
Absorptionb
Fc/Fc+ = 0.84 V
Fluorescencec
max, nm

max [nm] (log ( [M−1 cm−1]))
Sample
Q, B-bands & Axial Ligand
Bands

Oxidation
Reduction
5
C60-Ph-COOMe
AlPor-Ph
-
1.22
−0.33, −0.73, −1.28
−0.86, −1.23
256 (5.06), 309 (4.55)
547 (4.23), 416 (5.65)
-
6
7
8
9
596, 648
595, 648
594, 646
593, 646
592, 648
594, 650
-
AlPor-Ph-C60
AlPorF3-Ph
1.19
−0.32, −0.70, −0.88, −1.25
−0.71, −1.09
548 (4.43), 416 (5.76), 255 (5.12)
546 (4.20), 413 (5.56)
1.39, 1.69
1.41, 1.75
1.72
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
AlPorF3-Ph-C60
AlPorF5-Ph
−0.36, −0.73, −1.09, −1.31
−0.49, −0.89
546 (4.30), 414 (5.60), 256 (4.84)
551 (4.30), 418 (5.59)
AlPorF5-Ph-C60
BTMPA-Im
1.70
−0.34, −0.43, −0.75, −0.95, −1.31
-
551 (4.36), 418 (5.59), 256 (5.05)
308 (4.95)
1.18
aRedox potentials were reported against ferrocene, where E1/2 (Fc/Fc+) = 0.84 V with 0.1 M TBAP in o-DCB in our experimental
conditions. bIn CH2Cl2, cIn o-DCB.
Axial coordination of ligands to AlPors causes a significant shift of its absorption bands. These shifts can
be rationalized as arising from a change in the planarity of the porphyrin ring as the Al center, which lies above
the plane of the four pyrrole nitrogen atoms in the dyad, is drawn toward the coordinating ligand. These spectral
changes allow the formation of self-assembled dyads and triads (Schemes 1 and 2) to be monitored by absorption
titrations, from which the binding constants can be obtained as summarized in Table S1. Figure 1b shows
absorption titrations of AlPorF3-Ph-C60 with BTMPA-Im in o-DCB. Upon addition of BTMPA-Im, the Soret and
Q bands of the porphyrin 416, 546 and 588 nm are shifted to 422, 561 and 600 nm, which is typical of coordination
of nitrogen ligands to AlPor.66,71 Isosbestic points are observed at 426 and 555 nm, indicating the formation of the
triad BTMPA-ImAlPorF3-Ph-C60. Benesi−Hildebrand analysis (Figure 1b, inset) gives a linear plot indicating
that a 1:1 complex is formed, and the slope yields a binding constant K = 1.7  105 M−1. Similar spectral changes
were observed in titrations of AlPor-Ph-C60 vs BTMPA-Im (Figure 1c) and AlPorF5-Ph-C60 vs BTMPA-Im (Figure
1d) confirming the formation of the triads BTMPA-ImAlPor-Ph-C60 and BTMPA-ImAlPorF5-Ph-C60,
respectively. The binding constants differ for the triads and follow the order: K(AlPor-Ph-C60 vs BTMPA-Im) <
K(AlPorF3-Ph-C60 vs BTMPA-Im) < K(AlPorF5-Ph-C60 vs BTMPA-Im). The trend is consistent with the expected
increase in the Lewis acidity of the Al center as an increasing number of fluorine atoms are added to the phenyl
substituents. Hence, the Al center is the strongest Lewis acid when the porphyrin is substituted with PhF5 and
weakest with Ph. Therefore, it binds the Lewis base most strongly in the PhF5 substituted porphyrin and most
weakly with Ph substituents.
Additional titrations were carried out to establish the formation of the dyads BTMPA-ImAlPorFn-Ph (n
= 0, 3, 5): AlPor-Ph vs BTMPA-Im (Figure S14a), AlPorF3-Ph vs BTMPA-Im (Figure S14b) and AlPorF5-Ph vs
BTMPA-Im (Figure S14c). The observed spectral changes were similar to those of the corresponding triads; thus,
they establish the formation of dyads BTMPA-ImAlPorFn-Ph. Furthermore, titrations of AlPorFn-Ph-C60 (n =
0, 3, 5) with Me-Im (Figures S15) and AlPorFn-Ph (n = 0, 3, 5) with Me-Im (Figures S16), were performed in o-
9
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 33
DCB, (Me-Im = 1-methylimidazole). The spectral trends are consistent with thier corresponding dyads and triads
and were employed as controls in analyzing the excited state properties of the investigated systems. Together with
the NMR and absorption titrations, formation of the triads BTMPA-ImAlPorFn-Ph-C60 (n = 0, 3, 5) and dyads
BTMPA-ImAlPorFn-Ph (n = 0, 3, 5) are established in solution.
1
2
3
4
5
6
7
8
9
10
11
12
13
Computation. The electronic structure of the dyads and triads were studied using Density Functional Theory
(DFT). All the complexes (dyads and triads) were optimized at the B3LYP80/def2-SVP81 level of theory using an
ultrafine grid in the Gaussian 1682 software. The optimizations were performed in a CPCM83 dielectric continuum
14 solvent model of dichlorobenzene as implemented in Gaussian 16. Figures 2 and S17-S19 show optimized
15
geometries and frontier orbitals of the triads and dyads (plotted with isovalue 0.04). The DFT calculations revealed
that the location of the frontier orbitals is sensitive to the meso-substitutions of the porphyrin.
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
BTMPA-ImAlPor-Ph-C60: As shown in Figure 2, the HOMO, LUMO, and HOMO-1 are mainly localized on
the AlPor, C60 and BTMPA moieties. However, the HOMO and HOMO-1 are significantly delocalized on to the
BTMPA-Im and AlPor, respectively. Figure S19 shows the optimized geometries of one of the reference dyad
BTMPA-ImAlPor-Ph and it revealed similar delocalization of its frontier orbitals. Figure S17 shows another
reference dyad, AlPor-Ph-C60, and as projected the HOMO and LUMO are localized on the AlPor and C60 units,
respectively. Based on these results it clear that there exist strong electronic interactions between the AlPor and
BTMA-Im units in the triad BTMPA-ImAlPor-Ph-C60 but not between AlPor and C60 in the AlPor-Ph-C60 dyad.
The calculated radii are found to be 5.62, 8.85 and 3.50 Å for BTMPA, AlPor, and C60, respectively. The center-
to-center distance between BTMPA and C60, BTMPA and AlPor and between AlPor and C60 were measured to be
21.4, 9.6 and 12.4 Å, respectively. The angle between N(C60)-N(Im)-N(BTMPA) estimated to be 157.1.
BTMPA-ImAlPorF3-Ph-C60: As indicated in the Figure 2, the HOMO, LUMO and HOMO-1 are localized on
the BTMPA-Im, C60 and AlPorF3, respectively. Unlike the previous triad (BTMPA-ImAlPor-Ph-C60), the
40 frontier orbitals in this triad are exclusively localized on their respective moieties, and therefore direct electronic
41
42
interaction can be ruled out between redox centers. Moreover, the angle between N(C60)-N(Im)-N(BTMPA) is
43
44
found to be 132.9. The decrease in the angle represents the bent geometry between BTMPA and AlPorF3 units.
45
46
The calculated radius for AlPorF3 was found to be 9.06Å. The center-to-center distance between BTMPA and C60,
47 BTMPA and AlPorF3 and between AlPorF3 and C60 were estimated to be 21.6, 9.5 and 12.3 Å, respectively. In the
48
dyad BTMPA-ImAlPorF3-Ph (Figure S19), the HOMO and LUMO are restricted to the BTMPA and AlPorF3
units, whereas in the dyad AlPorF3-Ph-C60 (Figure S17), the HOMO-1 and LUMO are on the AlPorF3 and C60
units, respectively.
49
50
51
52
53
54
55
56
57
58
59
60
BTMPA-ImAlPorF5-Ph-C60: The HOMO and LUMO are localized on BTMPA-Im and C60, respectively, see
Figure 2. The electron withdrawing effect of the pentafluorinated phenyl substituents on the porphyrin lowers the
energy of the highest occupied -orbital of the porphyrin so that it lies below the HOMO-1. Again, the frontier
10
ACS Paragon Plus Environment

Page 11 of 33
Journal of the American Chemical Society
orbitals in this system are also exclusively localized on their respective moieties, and therefore no electronic
interaction can be anticipated between redox centers. Moreover, the angle between N(C60)-N(Im)-N(BTMPA)
found to be 132.9 which is same as for the BTMPA-ImAlPorF3-Ph-C60. The HOMO and LUMO in one of the
reference dyads, BTMPA-ImAlPorF5-Ph (Figure S19), are confined to the BTMPA and AlPorF5, respectively.
In the second reference dyad, AlPorF5-Ph-C60 (Figures S17 and S18), the HOMO-5, and LUMO are on the AlPorF5
and C60 units, respectively. The calculated radius for AlPorF5 and the center-to-center distance between BTMPA
and C60, BTMPA and AlPorF5 and between AlPorF5 and C60 were found to be 9.07, 21.6, 9.4 and 12.3 Å,
respectively.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
Figure 2. Frontier orbitals of BTMPA-ImAlPor-Ph-C60 (Triad 1), BTMPA-ImAlPorF3-Ph-C60 (Triad 2), and
39 BTMPA-ImAlPorFn-Ph-C60 (Triad 3).
40
41
42
43
Energetics of dyads and triads. Cyclic voltammetry measurements of the newly synthesized compounds were
44
45
performed in o-DCB with 0.1 M tetrabutylammonium perchlorate (TBA.ClO4) as supporting electrolyte and
ferrocene as an internal standard. The cyclic voltammogram of AlPor-Ph-C60 and its reference compounds AlPor-
Ph and fullerene derivative (C60-Ph-COOMe) are reported elsewhere.65 Representative voltammograms are shown
in Figure 3, and the oxidation and reduction midpoint potentials are summarized in Table 1. The redox processes
of all of the compounds are found to be one-electron reversible based on the peak-to-peak separation values, and
the cathodic-to-anodic peak current ratio. During the cathodic scan, all three dyads (AlPor-Ph-C60, AlPorF3-Ph-
C60 and AlPorF5-Ph-C60) showed four reduction processes. The observed voltammograms are assigned to the
combination of porphyrin and C60 reduction processes. For the dyad AlPor-Ph-C60, the first two processes are
assigned to the reduction of C60 and the third process to the first reduction of the porphyrin. The fourth process is
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
11
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 12 of 33
a combination of the third reduction of C60 and the second reduction of the porphyrin. In the case of AlPorF3-Ph-
C60, the first, third, and fourth processes are assigned to the first reduction of C60, second reduction of porphyrin,
and third reduction of C60, respectively. Whereas, the second process is assigned to the combination of porphyrin
first reduction and C60 second reduction. For AlPorF5-Ph-C60, the first reduction of porphyrin partially overlaps
with the first reduction of C60. The remaining distinct second, third and fourth processes are assigned to the second
reduction of C60, second reduction of porphyrin and third reduction of C60, respectively. Comparison of the
porphyrin reduction potentials in the dyads AlPorFn-Ph-C60 and their reference compounds AlPorFn-Ph shows that
with increasing numbers of fluorine substituents on the phenyl groups, the porphyrin reduction potentials undergo
1
2
3
4
5
6
7
8
9
10
11
12
13
14 a positive shift as a result of the increasing electron withdrawing ability of the phenyl groups.
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
Figure 3. Cyclic voltammograms of (a) C60-Ph-COOMe, (b) BTMPA-Im, (c) AlPorF3-Ph, (d) AlPorF3-Ph-C60,
49
50
51
52
53
54
55
56
57
58
59
60
(e) AlPorF5-Ph and (f) AlPorF5-Ph-C60 with 0.1 M TBA.ClO4 in o-DCB. Voltammograms were measured with
ferrocene (0.84 V vs Ag wire) as an internal standard. Scan rate 100 mV/s.
The anodic scan of the dyads reveals one or two oxidation processes, which are assigned to the first and
second oxidation of porphyrin unit. Again, the potentials follow the same trend as the reduction potentials with a
shift to more positive values as the electron withdrawing ability of the porphyrin substituents increases.
12
ACS Paragon Plus Environment

Page 13 of 33
Journal of the American Chemical Society
Comparison of dyads and their reference compounds show that the oxidation processes in dyads are not strongly
perturbed by the addition of the fullerene. The secondary electron donor BTMPA-Im shows a one-electron process
in the anodic scan corresponding to its first oxidation. Additionally, the redox potentials of the triads (physical
mixtures of the dyads and BTMPA-Im) were measured to evaluate possible effects of the electron donor on the
potentials of the porphyrin. The observed voltammogram (not shown) for each triad was found to be just the sum
of its corresponding dyad and BTMPA-Im, which indicates that the porphyrin potentials are not perturbed by
coordination of the electron donor.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
The redox potentials can be used in combination with optical data to construct the energy level diagram of
15 the states involved in possible electron-transfer processes. Figure 4 summarizes the energy levels of the
16
investigated compounds. The energies of the lowest excited singlet states (E0-0) of AlPor (2.14 eV), AlPorF3 (2.14
eV) and C60 (1.75 eV) and the lowest excited triplet states of AlPor (1.61 eV) and C60 (1.55 eV) have been taken
from the literature.35,79,84 The singlet and triplet state energies of AlPorF3 and AlPorF5 have been calculated from
their absorption and emission spectra, see Figure S20. The estimated energies are 2.14, 1.65 and 1.53 eV for
1AlPorF5*, 3AlPorF3* and 3AlPorF5*, respectively. The energy of the charge-separated states ECS (relative to the
ground state) and the free-energy changes for charge separation (ΔGCS), hole stabilization (ΔGHS), and electron
shift (ΔGES) are estimated using the following equations 1 and 2 (also see footnotes of Table 2) and are summarized
in Table 2.
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
퐸퐶푆 = 퐸표1푥 (퐷) ― 퐸1푟푒푑(퐴) + 퐺푆
(1)
2
2
Here, GS is the ion-pair stabilization and incorporates both the solvent-dependent Coulomb energy change upon
ion-pair formation or recombination and the free energy of solvation of the ions, Equation 2:
GS = e2/(40)[(1/(2R+) + 1/(2R–)  1/RD-A) 1/S  (1/(2R+) + 1/(2R–)) 1/R]
(2)
where R+, R– and RD-A are donor radius, acceptor radius and the center-to-center distance between donor and
acceptor, respectively. S is the dielectric constant of the solvent used for the photophysical studies (9.93 and 2.4
for o-DCB and toluene, respectively). R is the dielectric constant of the solvent used for measuring the redox
potentials, in this case o-DCB. Using the radii from the DFT calculations, GS values of −0.15 and 0.02 - 0.06 eV
•+
are obtained for BTMPA•+-AlPorFn•− in o-DCB and toluene, respectively. For the radical ion-pairs AlPorFn -Ph-
C60•− in the dyads and BTMPA•+-ImAlPorFn-Ph-C60•− in the triads, the calculated Coulomb stabilization due to
the attraction between the cation and anion of the radical pair is small because of the relatively large RD-A values.
However, the calculated destabilization in toluene compared to o-DCB is unreasonably large, because of the small
•+
•−
•−
radius estimated for C60•−. Hence for AlPorFn -Ph-C60 and BTMPA•+-ImAlPorFn-Ph-C60 we have not
included the GS term in their energies. Using the energies obtained from the optical and electrochemical data, the
13
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 14 of 33
energy-level diagram was constructed for the dyads and triads in toluene (Figure 4) and o-DCB (Figure S21). The
calculated free-energy level diagrams suggest the following electron transfer processes are energetically favorable
in these triads upon excitation of AlPorFn: (i) oxidative electron transfer from the excited singlet state of AlPorFn
(1AlPorFn*) to fullerene followed by a hole shift to BTMPA, or (ii) reductive electron transfer from BTMPA to
1AlPorF3* followed by a charge shift to C60. However, the rates of these processes are determined by the electronic
coupling and activation energy, which depends on the reorganization energy as well as the free energy change.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 2. Experimentally calculated free energy change (ΔG) values of charge separation (CS) and charge
recombination (CR) processes of the investigated dyads and triads in toluene.
Compound
ECS1
1.51
-
ECS2
-
ECS3
ΔGCS1
0.63
-
ΔGCS2
-
ΔGCR
1.51
2.10
1.50
1.77
1.93
1.52
2.04
1.63
1.63
-
AlPor-Ph-C60
-
BTMPA-ImAlPor-Ph
BTMPA-ImAlPor-Ph-C60
AlPorF3-Ph-C60
2.10
2.10
-
0.04
0.04
-
1.50
1.51
1.77
-
0.63
0.37
-
-
-
1.52
-
BTMPA-ImAlPorF3-Ph
BTMPA-ImAlPorF3-Ph-C60
AlPorF5-Ph-C60
1.93
1.93
-
0.21
0.21
-
1.77
2.04
-
0.37
0.10
-
-
BTMPA-ImAlPorF5-Ph
BTMPA-ImAlPorF5-Ph-C60
1.63
1.63
0.51
0.51
1.52
2.04
0.10

CS1: AlPorFn-C60

CS2: BTMPA-AlPorFn
GCS = ECS  E0-0, E0-0 = 2.14 eV
GCR =  ECS
CS3: BTMPA-AlPorFn-C60

Figure 4. Energy level diagram of investigated triads BTMPA-ImAlPorFn-Ph-C60 in toluene.
14
ACS Paragon Plus Environment

Page 15 of 33
1
Journal of the American Chemical Society
Steady-state fluorescence studies. The steady-state fluorescence spectra of the dyads and triads were measured
in o-DCB (Figure 5) and toluene (Figure S22). Interestingly, very similar results were observed in both solvents.
This is unusual because fluorescence quenching by electron transfer is typically less efficient in non-polar solvents
such as toluene. The experiments were carried out at same concentration of porphyrin in all samples. The spectra
were measured with an excitation wavelength of 555 nm, which excites the Q-band transition of the porphyrin.
2
3
4
5
6
7
8
9
10
11
12
13
14
AlPorFn-Ph-C60: Figure 5 shows the fluorescence spectra of the dyads AlPor-Ph-C60, AlPorF3-Ph-C60 and
AlPorF5-Ph-C60 and its reference compound AlPor-Ph, AlPorF3-Ph and AlPorF5-Ph, respectively in o-DCB, and
15 the emission maxima are summarized in Table 1. All of the compounds exhibit two bands due to porphyrin
16
emission. The band positions in the dyad and its reference porphyrin are very similar indicating that the electronic
structure of the porphyrin is not perturbed by axial-linkage of C60 unit. However, the fluorescence of the porphyrin
is quenched, 80%, 66% and 50%, in AlPor-Ph-C60, AlPorF3-Ph-C60 and AlPorF5-Ph-C60, respectively. Based on
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60


the energy level diagram, the AlPor-Ph-C60, AlPorF3 -Ph-C60 and AlPorF5 -Ph-C60 charge separated states
lie energetically below the excited singlet state of porphyrin (1AlPorFn*), therefore the observed quenching is
assigned to oxidative electron transfer from 1AlPorFn* to C60. The calculated ΔGCS for this electron transfer was
found to be 0.64, 0.27 and 0.10 eV for AlPor-Ph-C60, AlPorF3-Ph-C60 and AlPorF5-Ph-C60, respectively (see
Table 2). The fact that the quenching follows the same trend as the driving force for the electron transfer suggests
that electron transfer lies in the normal region of the Marcus curve as would be expected.
15
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 16 of 33
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 5. Fluorescence spectra of (a) AlPor-Ph (red) and AlPor-Ph-C60 (green), (b) AlPorF3-Ph (red) and AlPorF3-
Ph-C60 (green), (c) AlPorF5-Ph (red) and AlPorF5-Ph-C60 (green), (d) AlPor-Ph vs BTMPA-Im: BTMPA-Im was
added up to 1.52 × 10−4 M in 10 μl (6.60 × 10−4 M) increments to a 1 ml (4 × 10−6 M) solution of AlPor-Ph, (e)
AlPorF3-Ph vs BTMPA-Im: BTMPA-Im was added up to 3.0 × 10−5 M in 10 μl (1.37 × 10−4 M) increments to a 1
ml (4 × 10−6 M) solution of AlPorF3-Ph, and (f) AlPorF5-Ph vs BTMPA-Im: BTMPA-Im was added up to 5.54 ×
10−6 M in 10 μl (6.7 × 10−5 M) increments to a 1 ml (4 × 10−6 M) solution of AlPorF5-Ph. All the titrations were
performed in o-DCB.
BTMPA-ImAlPorFn-Ph: In order to establish the formation of the dyads (Scheme 2) and evaluate possible
photoinduced processes between BTMPA-Im and AlPorFn, steady-state fluorescence titrations were performed
between BTMPA-Im and AlPorFn-Ph in o-DCB. The excitation wavelength in each of these titrations was adjusted
to the isosbestic point, which was obtained from the corresponding absorption titrations, so that the extinction
coefficient is the same for each sample solution. As shown in Figure 5d, upon addition of BTMPA-Im to AlPor-
Ph, the AlPor fluorescence peaks shift to longer wavelengths from 594 and 646 nm to 613 and 665 nm, respectively
along with a change in their relative intensities. The observed spectral changes support the formation of the dyad
BTMPA-ImAlPor-Ph in solution. To investigate the possible reductive electron transfer between the BTMPA
and AlPor, a control titration of AlPorF-Ph vs Me-Im (without BTMPA, Figure S23a) was performed. Spectral
16
ACS Paragon Plus Environment

Page 17 of 33
Journal of the American Chemical Society
changes similar to those found in the titration of AlPor-Ph vs BTMPA-Im are observed suggesting that they are a
result of the binding of the imidazole moiety and not due to the presence of BTMPA as an electron donor.
Therefore, the rate of the anticipated reductive electron transfer from BTMPA to 1AlPor* is negligible compared
to the fluorescence rate. This is consistent with the small free energy change of –0.01 eV calculated for this process
(GCS2 in Table 2) since in this case, the activation energy for the electron transfer is determined mostly by the
reorganization energy and is expected to be fairly large.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In the case of AlPorF3-Ph vs BTMPA-Im (Figure 5e) titrations, the fluorescence bands are red shifted
which again confirms the formation of the dyad BTMPA-ImAlPorF3-Ph. However, the fluorescence intensity
decreases upon addition of BTMPA-Im. The corresponding control titrations AlPorF3-Ph vs Me-Im (Figure S23b)
revealed band shifts to longer wavelengths but without a change in fluorescence intensity suggesting that the
fluorescence quenching during the titrations of AlPorF3-Ph vs BTMPA-Im is indeed due to the presence of
BTMPA and reductive electron transfer from BTMPA to 1AlPorF3*. This is also consistent with the much larger
value of –0.22 eV for the GCS2. Figure 5c shows the titrations of AlPorF5-Ph vs BTMPA-Im, which establish the
formation of BTMPA-ImAlPorF5-Ph. A significant fluorescence quenching is exhibited without a spectral shift
in presence of BTMPA-Im. In contrast, the control titrations AlPorF5-Ph vs Me-Im (Figure S23c) show a shift of
the bands but no quenching. This implies that essentially complete quenching occurs when BTMPA-Im binds and
the fluorescence observed in the titration with BTMPA-Im arises only from the fraction of unbound porphyrin
molecules. The observed substantial quenching is assigned to the reductive electron transfer from BTMPA to
1AlPorF5* and again is consistent with the larger free energy change of –0.44 eV and hence lower activation energy
for this process. Overall, the observed fluorescence quenching pattern in the dyads is consistent with the expected
behavior based on the Gibbs free energy values and electron transfer in the normal region of the Marcus curve.
BTMPA-ImAlPorFn-Ph-C60: Figure S24 shows the fluorescence titrations of AlPorFn-Ph-C60 vs BTMPA-Im in
o-DCB. The spectral trends are very similar to their corresponding AlPorFn-Ph vs BTMPA-Im titrations and
confirm the formation of the triad BTMPA-ImAlPorFn-Ph-C60 in solution. The excitation of AlPorFn in the triad
results in an initial charge separation either as the oxidative electron transfer from 1AlPorFn* to C60 or the reductive
electron transfer from BTMPA to 1AlPorFn*. Qualitatively, the observed spectral trends during the titrations show
that degree of fluorination of the porphyrin substituents influences which of the two possible electron transfer
reactions occurs. Upon addition of BTMPA-Im to AlPor-Ph-C60 (Figure S24a) the AlPor fluorescence bands shift
to longer wavelengths 613 and 665 nm without any additional quenching. By comparing these trends with control
titrations, AlPor-Ph vs BTMPA-Im (Figure 5d) and AlPor-Ph-C60 vs Me-Im (Figure S23d), it is reasonable to
predict that oxidative electron transfer from 1AlPor* to C60 is the initial process in the triad BTMPA-ImAlPor-
Ph-C60. In contrast, in the titrations of AlPorF3-Ph-C60 with BTMPA-Im (Figure S24b) the fluorescence
bands are shifted to 609 and 657 nm with a moderate decrease in their intensities. In this case, comparison with
control titrations of AlPorF3-Ph vs BTMPA-Im (Figure 5e) and AlPorF3-Ph-C60 vs Me-Im (Figure S23e), as well
17
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 18 of 33
as the energy level diagram (Figure 4) suggest that the observed decrease in fluorescence intensity is due to
reductive electron transfer from BTMPA to 1AlPorF3*. This implies that the rate of reductive electron transfer rate
is comparable to or faster than that of the oxidative electron transfer rate from 1AlPorF3* to C60. Lastly, Figure
S24c shows the fluorescence spectra of the AlPorF5-Ph-C60 with increasing amounts of BTMPA-Im. In this case,
addition of BTMPA-Im, leads to strong quenching without any significant shift of the emission bands. Consistent
with the control experiments, AlPorF5-Ph vs BTMPA-Im (Figure 5f) and AlPorF5-Ph-C60 vs Me-Im (Figure S23f),
as well as the energy diagram (Figure 4), the observed strong quenching in BTMPA-ImAlPorF5-Ph-C60 implies
that reductive electron transfer from BTMPA to 1AlPorF5* is the fastest process in this triad.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
Spectroelectrochemistry. To help interpret the transient spectral data of the products formed during
the process of charge injection, spectroelectrochemical studies were performed on AlPorF3-Ph and AlPorF5-Ph,
and BTMPA-Im in o-DCB containing 0.2 M TBA.ClO4. As shown in Figure S25, the oxidation of BTMPA-Im
results in a decrease in absorbance at 308 nm which is accompanied by the appearance of a new peak with a
maximum around 333 nm. An isosbestic point at 340 nm is also observed. The oxidation of AlPorF3 revealed the
appearance of new bands at 595 and 690 nm, whereas the reduction produced bands at 568 and 608 nm, Figure
S26. In the case of AlPorF5, the oxidation resulted in new peaks at 644 and 686 nm and reduction developed peaks
at 450 and 627 nm, Figure S27. The spectroelectrochemical data of the AlPor have been reported previously63 and
show that peaks at 600 and 685 nm are formed when AlPor is oxidized and at 570 and 610 nm when it is reduced.
Femtosecond Transient Absorption Spectroscopy. Having proposed the different photochemical charge
separation events in the AlPorFn-Ph-C60 dyads and BTMPA-ImAlPorFn-Ph-C60 triads from the energy levels
diagrams shown in Figure 4 and fluorescence titrations shown in Figure 5, femtosecond and nanosecond transient
absorption spectral studies were performed to secure direct experimental evidence of charge separation and also
to evaluate the lifetimes of the charge separated states. In the present study, we have used toluene rather than o-
DCB as the solvent because of its better photochemical stability under laser irradiation
Figure 6 shows the femtosecond transient spectra at the indicated delay times for AlPor-Ph, AlPorF3-Ph
and AlPorF5-Ph in the absence and presence of axial coordinating BTMPA-Im. The transient spectral features of
45 AlPor-Ph, AlPorF3-Ph and AlPorF5-Ph have a similar pattern, however, with subtle differences. In the case of
46
47
AlPor-Ph, the instantaneously formed 1AlPor* (see the spectrum at 1 ps in Figure 6a) has positive peaks at 448,
48
49
578, 611 and 1240 nm due to transitions originating from the 1AlPor* state. Negative peaks at 550, 595 and 651 nm
50
51
52
53
54
55
56
57
58
59
60
are also observed. By comparison with the absorption and fluorescence spectral data described above, the 550 nm
peak is assigned to ground state bleaching and the latter two peaks are assigned to stimulated emission. The lifetime
of the decay of positive peaks and recovery of the negative peaks is longer than the 3 ns time window accessible
in the experiment, which is consistent with the 7.88 ns fluorescence lifetime of AlPor, determined from time
correlated single photon counting measurements. The decay and recovery of the positive and negative peaks are
18
ACS Paragon Plus Environment

Page 19 of 33
Journal of the American Chemical Society
accompanied by new peaks emerging in the 500 and 840 nm region due to the formation of 3AlPor* by intersystem
1
2
*
crossing. In the case of AlPorF3-Ph, transient positive peaks of 1AlPorF3 are observed at 453, 595 and 1236 nm
3
4
while the transient negative peak at 550 nm due to ground state bleaching and at 650 nm due to stimulated emission
5
6
*
are also observed (Figure 6d). Slow intersystem crossing populating 3AlPorF3 leads to peaks at 490 and 835 nm
7
8
9
on a timescale consistent with the fluorescence lifetime of 3.74 ns. Finally, for AlPorF5-Ph photoexcitation results
in positive peaks at 454, 573, 610 and 1230 nm from the excited singlet state accompanied by negative peaks at
10
11
12
13
3
*
546, 589 and 645 nm due to ground state bleaching and stimulated emission (Figure 6g). The AlPorF5 state
formed by intersystem crossing resulted growth of new peaks at 450 and 832 nm on a ns timescale. Again, this is
14 consistent with the fluorescence lifetimes of 2.32 ns. In summary, AlPor-Ph, AlPorF3-Ph and AlPorF5-Ph all show
15
similar behavior with instantaneous formation of the excited singlet state, which decays with a lifetime of several
nanoseconds to the ground state and excited triplet state.
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 6. Femtosecond transient absorption spectra at the indicated delay times (ex = 410 nm) of (a) AlPor-Ph,
(c) BTMPA-ImAlPor-Ph, (d) AlPorF3-Ph, (f) BTMPA-ImAlPorF3-Ph, (g) AlPorF5-Ph, and (i) BTMPA-
ImAlPorF5-Ph in toluene. The central panel (b), (e), and (h) shows the time profile of the near-IR peak at 1240
nm in the absence (blue trace) and presence (red trace) of coordinated BTMPA to a given AlPorFn-Ph.
19
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 20 of 33
1
2
Figure 6, panels (c), (f) and (i) show the femtosecond transient absorbance spectra of dyads BTMPA-
3
4
ImAlPor-Ph, BTMPA-ImAlPorF3-Ph, and BTMPA-ImAlPorF5-Ph. The central panels (b, e and h) show
5
6
1
the time profiles of the near-IR peak located at 1240 nm due to singlet-singlet absorption of AlPorFn* in the
7
8
9
presence (red) and absence (blue) of BTMPA. Upon coordination of BTMPA-Im with AlPor-Ph, positive transient
peaks originating from BTMPA-Im1AlPor*-Ph are located at 452, 587, 627, 710 and 1248 nm, and negative
red-shifted transient peaks at 556 nm due to ground state bleaching, and at 608 and 660 nm due to stimulated
emission are observed. The energy level diagram shown in Figure 4 predicts BTMPA-AlPor lies only 0.01 eV
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
lower than AlPor* thus hole transfer from the porphyrin to the donor is expected to be slow or negligent.
Consistent with this, the two kinetic traces overlap in Figure 6b revealing little or no effect of BTMPA, that is, the
absence of significant amounts of a hole shift leading to BTMPA-ImAlPor-Ph. In the cases of BTMPA-
Im1AlPorF3*-Ph and BTMPA-Im1AlPorF5*-Ph, the hole shift leading to BTMPAAlPorF3 and BTMPA-
AlPorF5 were found to be exothermic by 0.22 and 0.44 eV, respectively. Under such conditions, new transient
peaks corresponding to the charge-separated states is expected. Spectroelectrochemical studies show that

BTMPA has a band at 333 nm (Figure S25) while AlPorF3 has bands at 568 and 608 nm (Figure S26) and
AlPorF5 has bands at 450 and 627 nm (Figure S27). The band at 333 nm due to BTMPA is not accessible with
the femtosecond setup used. The bands due to the porphyrin radical anion overlap with those of the excited singlet
state or have weak intensity making it difficult to positively identify them in the femtosecond data. However, the
kinetic time profiles of the 1240 nm near-IR peak provide some evidence for the expected charge separation. As
1
shown in Figures 6e and 6h, there is an indication of a slight acceleration of the decay of both AlPorF3* and
1AlPorF5* in the presence of BTMPA, which might be due to the hole shift in these two dyads. From the fitted
decay lifetimes, the rate constants for hole shift, kHS were estimated to be 0.80 x 109 s-1 and 1.83 x 109 s-1,
respectively, for BTMPA-ImAlPorF3-Ph and BTMPA-ImAlPorF5-Ph.
Next, photoinduced charge separation in the covalently linked AlPor-Ph-C60, AlPorF3-Ph-C60, and
AlPorF5-Ph-C60 and their coordination complexes with BTMPA-Im were performed. In agreement with our earlier
studies,35,65 the femtosecond data reveal that photoexcitation of AlPor in the dyad AlPor-Ph-C60 results in charge
separation from 1AlPor* to C60 leading to the formation of AlPor-C60. The absorption peak at ∼450 nm, the
near-IR peak at 1240 nm, and the negative peaks due to S1 emission at 590 and 650 nm decayed more rapidly than
in AlPor-Ph (see Figure 6a). Additionally, these processes were accompanied by new signals in the range of 600-
650 nm corresponding to AlPor and near 1000 nm from C60. Since the 1000 nm peak of C60 and 1240 nm
1
peak of AlPor* were far from other transient bands they can be used to estimate the rate constant for charge
separation, kCS. The transient spectra of BTMPA-ImAlPor-Ph-C60 at different delay times (Figure 7c) show
peaks characteristic of the charge-separated state and provide evidence for the charge separation from 1AlPor* to
20
ACS Paragon Plus Environment

Page 21 of 33
Journal of the American Chemical Society
C60. However, the decay of the absorbance change due to C60 at 1000 nm (Figure 7b, red trace) is longer than the
3 ns time window of our instrumental setup. Thus, in this case the rate constant for charge recombination, kCR was
evaluated from nanosecond transient spectral studies that will be discussed in the next section. If hole transfer
from BTMPA to AlPor occurs then one would expect the decay of C60 to be slower when BTMPA is bound,
that is, the larger separation of the radical cation and radical anion species in BTMPA-AlPor-C60- compared to
AlPor-C60 should stabilize the charge separation. A comparison of time profiles of C60 for AlPor-Ph-C60 and
1
2
3
4
5
6
7
8
9
10
11 BTMPA-ImAlPor-Ph-C60 is shown in the Figure 7b. As can be seen, the two traces are virtually identical except
12
for the rise time which is slightly shorter for the triad and a slight difference beyond 2000 ps suggesting that the
decay of C60 in the triad (blue trace) may be slightly slower than in the dyad (red trace). The fact that the two
traces are essentially identical up to ~2000 ps suggests that if hole transfer involving BTMPA occurs in this triad
it is slow.
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 7. Femtosecond transient absorption spectra at the indicated delay times (ex = 410 nm) of (a) AlPor-Ph-
C60, (c) BTMPA-ImAlPor-Ph-C60, (d) AlPorF3-Ph-C60, (f) BTMPA-ImAlPorF3-Ph-C60, (g) AlPorF5-Ph-C60,
-
and (i) BTMPA-ImAlPorF5-Ph-C60 in toluene. The central panels (b), (e) and (h) shows time profile of the C60
peak in the absence (red trace) and presence (blue trace) of coordinated BTMPA to a given AlPorFn-Ph-C60.
21
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 22 of 33
For the triads BTMPA-ImAlPorF3-Ph-C60 and BTMPA-ImAlPorF5-Ph-C60, the energy level diagrams
shown in Figure 4 predict that the hole transfer process is exothermic by 0.37 eV and 0.54 eV, respectively, which
is considerably more favorable than in BTMPA-ImAlPor-Ph-C60. Consistent with this, the effect of hole transfer
process further stabilizing the final charge separated states is clearly visible in the femtosecond transient
absorbance data. Figures 7d and g show the transient absorption spectra at selected delay time and in both dyads,
1
2
3
4
5
6
7
8

evidence for charge separation from the excited singlet state of the porphyrin leading to the formation of AlPorF3 -
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55



C60 and AlPorF5 -C60 is seen. The values of kCS were determined by fitting the decay of the 1230 nm peak
corresponding to singlet-singlet absorption of 1AlPorF3 and 1AlPorF5* and are listed in Table 3. As shown in the
*
middle panels of Figure 7e and h, the decay of the C60 peak (red traces) lasted beyond 3 ns (red trace), similar to
that observed for AlPor-Ph-C60 dyad. For BTMPA-ImAlPorF3-Ph-C60 and BTMPA-ImAlPorF5-Ph-C60

(Figures 7f and i) similar kCS values are obtained (see Table 3). The comparisons of the time profiles of the C60
peak of the triads (blue traces) dyads (red traces) show that the decay is significantly slower in the triads and is
slowest in BTMPA-ImAlPorF5-Ph-C60. This suggests that the decay of C60 competes with hole transfer from
BTMPA and that rate of hole transfer is faster BTMPA-ImAlPorF5-Ph-C60 as expected from the larger driving
force and lower activation energy.
Table 3. Rate constants for charge separation, kCS and charge recombination, kCR and overall lifetime of the radical
ion-pair, RIP evaluated from femto- and nanosecond transient absorption spectroscopy for the investigated dyads
and triads in toluene.
kCS, x 109 (s-1)
--
kCR, x 106 (s-1)
Compound
TMPA-Im AlPor-Ph
TMPA-Im AlPorF3-Ph
TMPAIm AlPorF5-Ph
AlPor-Ph-C60
Ref
tw

RIP (s)
--
--
0.80
--
--
tw
1.83
--
--
tw
66
6.45
26.3
1.85
1.51
17.9
1.35
0.81
0.038
0.54
0.66
0.056
0.74
1.24
AlPorF3-Ph-C60
2.52
tw
tw
tw
tw
tw
AlPorF5-Ph-C60
4.27
TMPA-Im AlPor-Ph-C60
TMPA-Im AlPorF3-Ph-C60
TMPA-Im AlPorF5-Ph-C60
tw = this work
2.53
1.21
3.44
Nanosecond Transient Absorption Spectroscopy.
Lastly, to evaluate the lifetime of the final charge-separated states, nanosecond transient spectral measurements
were performed on all of the newly assembled systems and the data are shown in Figure 8. Although relatively
weak, in all of the dyads and triads, the C60 peak at ~1000 nm could be observed indicating persistence of the
56 radical ion-pair into the nanosecond to microsecond time window. In addition to peaks of C60, transient features
57
3
3
in the 460 and 850 nm range corresponding to AlPorFn* and in the 700 nm range corresponding to C60* were
58
59
60
22
ACS Paragon Plus Environment

Page 23 of 33
Journal of the American Chemical Society
also observed. These results suggest that the singlet excited AlPorFn and/or the charge-separated species could
competitively populate the triplet excited states to some extent. The lifetime of the absorbance changes at 1000
nm due to C60 was used to calculate kCR and lifetime of the radical ion-pairs.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 8. Nanosecond transient absorption spectra at the indicated delay times (ex = 410 nm) of (a) BTMPA-
ImAlPor-Ph-C60, (b) AlPorF3-Ph-C60, (c) BTMPA-ImAlPorF3-Ph-C60, (d) AlPorF5-Ph-C60, and (e) BTMPA-
ImAlPorF5-Ph-C60 in toluene.
1
The data listed in Table 3 reveal several features. (i) The rates of hole transfer from AlPorFn* to
coordinated BTMPA are lower than those of electron transfer from 1AlPorFn* to C60. Thus, we surmise that in the
triads, electron transfer from the excited singlet state of AlPorFn to C60 occurs first and subsequently a hole transfer
from AlPorFn to BTMPA takes place. (ii) Generally, the rate constant for charge separation kCS correlates with
the free-energy difference between the excited singlet state of the porphyrin and the charge-separated state AlPor-
C60–, that is, a higher rate constant is observed for AlPor-Ph-C60 compared to the dyads with the fluorinated
AlPorF3-Ph-C60 and AlPorF5-Ph-C60. (iii) Coordination of BTMPA causes a slight increase in kCS. A possible
explanation for this is that coordination of BTMPA-Im to AlPorFn draws the porphyrin ring slightly closer to the
C60 unit by pulling the Al center into the porphyrin plane as suggested by DFT studies. (iv) Finally, the kCR values
decrease (RIP values increase) with increasing fluorination of porphyrin for both the dyads and triads. This is
especially intriguing considering the fact that the final charge separated state in the triads is the same and has the
same energy (1.50 eV). These results suggest that the central electron deficient AlPorF3 or AlPorF5 plays a key
23
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 24 of 33
role in slowing down the charge recombination process.
1
2
3
4
5
6
7
8
9


Recombination mechanism and lifetimes. The lifetimes of the charge-separated states AlPorFn -C60 in the
dyads can be rationalized by using the Marcus equation (3)85–87 to relate the driving force (GET) to the rate
constants (kET) of the charge separation or charge recombination. Here,  is the reorganization energy, V is the
electronic coupling matrix element, kB is the Boltzmann constant, h is the Planck constant, and T is the absolute
temperature.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
1
2
4휋3
)
푘퐸푇 =
2푉2푒푥푝 ― (
Δ퐺퐸푇 + 휆
(3)
2
(
)
[
]
4휆푘퐵푇
ℎ 휆푘퐵푇
The values of 38, 540 and 660 ns for CR in the dyads AlPor-Ph-C60, AlPorF3-Ph-C60 and AlPorF5-Ph-C60,
respectively, clearly show that the radical pair is stabilized by substitution of fluorine atoms on the phenyl
substituents of the porphyrin ring. This is because the corresponding driving force for recombination, (GET = –
1.50, –1.87, –2.04 eV) places the recombination far into the Marcus inverted region if the reorganization energy,
, is on the order of 0.2 eV as suggested from studies of other porphyrin fullerene dyads.31 Therefore, the charge
recombination becomes slower as the driving force for recombination increases. It is important to compare these
results with some related systems in the literature. For example, the charge separation and recombination in “axial-
bonding” type high potential zinc(II) porphyrin – C60 dyads have been studied as a function of driving force.88
However, the recombination rates did not follow the expected trend and the highly fluorinated zinc(II) porphyrin
– C60 dyads showed faster recombination than the non-fluorinated zinc(II) porphyrin-C60 dyad. This suggests that
either the electronic coupling or the recombination pathway differed in the different dyads. Numerous “horizontal”
zinc(II) porphyrin – C60 dyad compounds have also been reported in the literature.26,28 In these systems, the charge
37 recombination sometimes occurs to the C60 triplet state28 depending on the nature of the bridging groups and energy
38
of the charge separated state. For those in which direct recombination to the ground state occurs, the rates are
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
typically slower than observed here, suggesting that the electronic coupling is larger in the vertical arrangement.26

Unlike the dyads, all of the investigated triads produce the same radical pair, BTMPA-AlPorFn-C60
,
with an identical energy of 1.50 eV but with a significant difference in their lifetimes. We note, however, that
value of the energy is based on the midpoint potential of unbound BTMPA and some small variation in the energies
of the final radical pairs is expected. Nonetheless, the observed lifetime differences are too large to be reasonably
explained using the free energy differences (G) or different reorganization energies () of the final radical pair.
Instead, factors such as the electronic coupling (V) and/or a hopping/superexchange mechanism must be invoked
to explain the variation in the backreaction rates. In the case of BTMPA-ImAlPor-Ph-C60, the final radical pair
lifetime was found to be 56 ns and is only slightly longer than the 38 ns observed for the corresponding dyad. The
DFT calculations revealed that strong electronic coupling exists between the AlPor and BTMPA units as the
HOMO and HOMO-1 are delocalized significantly onto the BTMPA and AlPor, respectively. Moreover, the
24
ACS Paragon Plus Environment

Page 25 of 33
Journal of the American Chemical Society
estimated energies of the AlPor-C60, and BTMPA-AlPor-C60 are identical. Under these conditions, it is

1
2
3
4
5
6

likely that if the radical pair BTMPA-AlPor-C60 is formed, its recombination to AlPor-C60 probably has a

lifetime that is similar to that of the decay of AlPor-C60 to the ground state. In an incoherent hopping

mechanism, this would result in a lifetime for the BTMPA-AlPor-C60 radical pair that is similar to that of
7
8

AlPor-C60 in the dyad. In contrast, for BTMPA-ImAlPorF3-Ph-C60, a hopping mechanism is unlikely
9
because the electronic coupling between AlPorF3 and BTMPA is weaker and the possible intermediates AlPor-
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
C60 and BTMPA-AlPorF3 radical pairs are much higher in energy than the final radical pair BTMPA-
AlPorF3-C60. Thus, the lifetime of 740 ns probably represents direct recombination to the ground state and is an
order of magnitude longer than that of BTMPA-AlPor-C60. Similarly, in the triad BTMPA-ImAlPorF5-Ph-
C60, the final radical pair also likely decays directly to the ground state because of the high energies of the possible
intermediate radical pairs and weak electronic coupling. The lifetime of 1240 ns of the final radical pair BTMPA-




AlPorF5-C60 is slightly longer than that of BTMPA-AlPorF3-C60 suggesting that either there is a small
difference in the electronic coupling or the electron withdrawing effect of the porphyrin substituents leads to a

small stabilization of BTMPA-AlPorF5-C60 in the triad. Thus, the fact that the recombination lifetime is
different for the triads BTMPA-ImAlPorF3-Ph-C60 and BTMPA-ImAlPorF5-Ph-C60 indicates that the central
porphyrin (AlPorF3 vs AlPorF5) is important in determining the lifetime of the charge separation because it has an
impact on the energies of possible recombination intermediates and therefore on the possible recombination
mechanisms.
CONCLUSION
The triads studied here are related to other axial-bonding triads we have reported previously 35,63–67 with a
general formula of “Donor-Py/ImAl(III)Porphyrin-Acceptor”. A variety of donor and acceptor units have been
employed and in all of these systems, sequential electron transfer takes place resulting in stable charge separation
between the donor and acceptor. By varying the components of triads we have been able to study the sequential
charge separation and recombination processes as a function of factors such as the donor-acceptor distance,35
reorganization energy,89,90 Gibbs free-energy,64 and solvent polarity.67 The observed charge recombination
lifetimes can be rationalized using the Marcus equation and both the superexchange mechanism and incoherent
electron transfer have been invoked to explain the recombination rates.
Previously, the focus has been on varying the terminal redox units or bridging groups to study their effect
on the charge separation and recombination rates. Although these studies provide insight into the relationship
between the components of the complexes and the rates, it is sometimes unclear which variable is responsible for
observed differences in the rates. Here, by keeping the terminal redox units constant and making as small a
structural change as possible to the central porphyrin to change its redox potential, it is possible to keep some of
25
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 26 of 33
the electron transfer factors fixed, such as the distance and bridge between redox/photoactive active units,
reorganization energy, free energy of the final charge separated state, etc. As a result, the observed changes in the
rates can be clearly ascribed to the changes in the redox potential of the porphyrin. Similar triads but with the
donor and acceptor attached in the plane of the porphyrin have also been reported Imahori and co-workers26,78 but
in these systems, the central metal of the porphyrin was altered to change its redox potential. In contrast, to the
systems reported here, in which the recombination reaction is far into the Marcus inverted region, the
recombination reaction was close to the top of the Marcus curve and the back reaction rate was governed primarily
by the electronic coupling which was much smaller due to the different bonding arrangement. From this
1
2
3
4
5
6
7
8
9
10
11
12
13
14 comparison, it is clear that these two factors must be balanced against each other to achieve long-lived charge
15
separation. In systems, like the ones presented here in which the electronic coupling is fairly strong, it is important
that the back reaction should lie far into the Marcus inverted region. This means that the midpoint potential of the
central porphyrin should be chosen so that possible intermediates in a hopping mechanism of charge recombination
are high enough in energy to make them inaccessible.
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ASSOCIATED CONTENT
Supporting Information: Synthesis schemes and details, spectroscopic experiments, mass, IR and NMR spectra,
absorption spectra, binding constants, spectroscopic titrations, phosphorescence spectra, DFT optimized structures
with frontier orbitals, energy level diagram in o-DCB, and spectroelectrochemical data.
AUTHOR INFORMATION
Corresponding authors.
Prashanth K. Poddutoori. Department of Chemistry & Biochemistry, University of Minnesota Duluth, 1038
University Drive, Duluth, Minnesota 55812, United States. Email: ppk@d.umn.edu
Francis D’Souza. Department of Chemistry, University of North Texas, 1155 Union Circle, # 305070, Denton,
Texas 76203-5017, United States. Email: francis.dsouza@unt.edu
Art van der Est. Department of Chemistry, Brock University, 1812 Sir Isaac Brock Way, St. Catharines, ON, L2S
3A1, Canada. Email: avde@brocku.ca
Victor S. Batista. Department of Chemistry, Yale University, P.O. Box 208107, New Haven, Connecticut 06520-
8107, United States. Email: victor.batista@yale.edu
Authors.
Niloofar Zarrabi. Department of Chemistry & Biochemistry, University of Minnesota Duluth, 1038 University
Drive, Duluth, Minnesota 55812, United States.
26
ACS Paragon Plus Environment

Page 27 of 33
Journal of the American Chemical Society
Sairaman Seetharaman. Department of Chemistry, University of North Texas, 1155 Union Circle, # 305070,
Denton, Texas 76203-5017, United States.
1
2
3
4
5
6
Subhajyoti Chaudhuri. Department of Chemistry, Yale University, P.O. Box 208107, New Haven, Connecticut
06520-8107, United States.
7
8
9
Noah Holzer. Department of Chemistry & Biochemistry, University of Minnesota Duluth, 1038 University Drive,
Duluth, Minnesota 55812, United States.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Notes. The authors declare no competing financial interest.
ACKNOWLEDGEMENTS.
This work was supported by startup grants from the University of Minnesota Duluth, to PPK, and by the National
Science Foundation (Grant No. 1401188) to FD and by an NSERC Discovery Grant (2015-04021) to AvdE.
REFERENCES.
(1)
(2)
(3)
(4)
(5)
(6)
(7)
Barber, J.; Andersson, B. Revealing the Blueprint of Photosynthesis. Nature 1994, 370 (6484), 31–34.
https://doi.org/10.1038/370031a0.
Blankenship, R. E. Molecular Mechanisms of Photosynthesis; 2008.
https://doi.org/10.1002/9780470758472.
Deisenhofer, J.; Norris, J. R. Photosynthetic Reaction Center, Two-Volume Set: Photosynthetic Reaction
Center, Volume 2; Academic Press: San Diego, CA, 1993.
Hou, H. J. M.; Najafpour, M. M.; Moore, G. F.; Allakhverdiev, S. I. Photosynthesis: Structures,
Mechanisms, and Applications; Springer, Cham, 2017. https://doi.org/10.1007/978-3-319-48873-8.
Barber, J. Photosynthetic Energy Conversion: Natural and Artificial. Chem. Soc. Rev. 2009, 38 (1), 185–
196. https://doi.org/10.1039/b802262n.
Fukuzumi, S. Development of Bioinspired Artificial Photosynthetic Systems. Physical Chemistry
Chemical Physics. 2008, pp 2283–2297. https://doi.org/10.1039/b801198m.
Hambourger, M.; Moore, G. F.; Kramer, D. M.; Gust, D.; Moore, A. L.; Moore, T. A. Biology and
Technology for Photochemical Fuel Production. Chem. Soc. Rev. 2009, 38 (1), 25–35.
https://doi.org/10.1039/b800582f.
(8)
(9)
Armaroli, N.; Balzani, V. Solar Electricity and Solar Fuels: Status and Perspectives in the Context of the
Energy Transition. Chemistry - A European Journal. 2016, pp 32–57.
https://doi.org/10.1002/chem.201503580.
Fukuzumi, S.; Ohkubo, K.; Suenobu, T. Long-Lived Charge Separation and Applications in Artificial
Photosynthesis. Acc. Chem. Res. 2014, 47 (5), 1455–1464. https://doi.org/10.1021/ar400200u.
(10) Imahori, H.; Mori, Y.; Matano, Y. Nanostructured Artificial Photosynthesis. Journal of Photochemistry
and Photobiology C: Photochemistry Reviews. 2003, pp 51–83. https://doi.org/10.1016/S1389-
5567(03)00004-2.
(11) El-Khouly, M. E.; El-Mohsnawy, E.; Fukuzumi, S. Solar Energy Conversion: From Natural to Artificial
Photosynthesis. Journal of Photochemistry and Photobiology C: Photochemistry Reviews. 2017, pp 36–
83. https://doi.org/10.1016/j.jphotochemrev.2017.02.001.
(12) Nocera, D. G. The Artificial Leaf. Acc. Chem. Res. 2012, 45 (5), 767–776.
https://doi.org/10.1021/ar2003013.
(13) Kurumisawa, Y.; Higashino, T.; Nimura, S.; Tsuji, Y.; Iiyama, H.; Imahori, H. Renaissance of Fused
Porphyrins: Substituted Methylene-Bridged Thiophene-Fused Strategy for High-Performance Dye-
27
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 28 of 33
Sensitized Solar Cells. J. Am. Chem. Soc. 2019, 141(25), 9910-9919.
https://doi.org/10.1021/jacs.9b03302.
(14) Yamamoto, M.; Föhlinger, J.; Petersson, J.; Hammarström, L.; Imahori, H. A Ruthenium Complex–
1
2
3
4
Porphyrin–Fullerene-Linked Molecular Pentad as an Integrative Photosynthetic Model. Angew. Chemie -
Int. Ed. 2017, 56 (12), 3329–3333. https://doi.org/10.1002/anie.201612456.
5
(15) Fukuzumi, S.; Lee, Y. M.; Nam, W. Bioinspired Artificial Photosynthesis Systems. Tetrahedron 2020, 76.
https://doi.org/10.1016/j.tet.2020.131024.
(16) Sampaio, R. N.; Troian-Gautier, L.; Meyer, G. J. A Charge-Separated State That Lives for Almost a
Second at a Conductive Metal Oxide Interface. Angew. Chemie - Int. Ed. 2018, 57 (47), 15390–15394.
https://doi.org/10.1002/anie.201807627.
6
7
8
9
10
11
12
13
(17) Whang, D. R.; Apaydin, D. H. Artificial Photosynthesis: Learning from Nature. ChemPhotoChem 2018, 2
(3), 148–160. https://doi.org/10.1002/cptc.201700163.
14 (18) Guldi, D. M. Fullerene-Porphyrin Architectures; Photosynthetic Antenna and Reaction Center Models.
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Chem. Soc. Rev. 2002, 31 (1), 22–36. https://doi.org/10.1039/b106962b.
(19) Gust, D.; Moore, T. A.; Moore, A. L. Mimicking Photosynthetic Solar Energy Transduction. Acc. Chem.
Res. 2001, 34 (1), 40–48. https://doi.org/10.1021/ar9801301.
(20) Kodis, G.; Terazono, Y.; Liddell, P. A.; Andréasson, J.; Garg, V.; Hambourger, M.; Moore, T. A.; Moore,
A. L.; Gust, D. Energy and Photoinduced Electron Transfer in a Wheel-Shaped Artificial Photosynthetic
Antenna-Reaction Center Complex. J. Am. Chem. Soc. 2006, 128 (6), 1818–1827.
https://doi.org/10.1021/ja055903c.
(21) Kuramochi, Y.; Sandanayaka, A. S. D.; Satake, A.; Araki, Y.; Ogawa, K.; Ito, O.; Kobuke, Y. Energy
Transfer Followed by Electron Transfer in a Porphyrin Macrocycle and Central Acceptor Ligand: A
Model for a Photosynthetic Composite of the Light-Harvesting Complex and Reaction Center. Chem. - A
Eur. J. 2009, 15 (10), 2317–2327. https://doi.org/10.1002/chem.200801796.
(22) Gust, D.; Moore, T. A.; Moore, A. L. Solar Fuels via Artificial Photosynthesis. Acc. Chem. Res. 2009, 42
(12), 1890–1898. https://doi.org/10.1021/ar900209b.
(23) Wasielewski, M. R. Self-Assembly Strategies for Integrating Light Harvesting and Charge Separation in
Artificial Photosynthetic Systems. Acc. Chem. Res. 2009, 42 (12), 1910–1921.
https://doi.org/10.1021/ar9001735.
(24) Fiechter, S. Artificial Photosynthesis - An Inorganic Approach. Adv. Bot. Res. 2016, 79, 99–128.
https://doi.org/10.1016/bs.abr.2016.04.002.
(25) Gust, D. An Illustrative History of Artificial Photosynthesis. Adv. Bot. Res. 2016, 79, 1–42.
https://doi.org/10.1016/bs.abr.2016.02.004.
(26) Imahori, H.; Tamaki, K.; Guldi, D. M.; Luo, C.; Fujitsuka, M.; Ito, O.; Sakata, Y.; Fukuzumi, S.
Modulating Charge Separation and Charge Recombination Dynamics in Porphyrin-Fullerene Linked
Dyads and Triads: Marcus-Normal versus Inverted Region. J. Am. Chem. Soc. 2001, 123 (11), 2607–
2617. https://doi.org/10.1021/ja003346i.
(27) Schuster, D. I.; Cheng, P.; Jarowski, P. D.; Guldi, D. M.; Luo, C.; Echegoyen, L.; Pyo, S.; Holzwarth, A.
R.; Braslavsky, S. E.; Williams, R. M.; Klihm, G. Design, Synthesis, and Photophysical Studies of a
Porphyrin-Fullerene Dyad with Parachute Topology; Charge Recombination in the Marcus Inverted
Region. J. Am. Chem. Soc. 2004, 126 (23), 7257–7270. https://doi.org/10.1021/ja038676s.
(28) Wolf, M.; Costa, J. I. T.; Minameyer, M. B.; Drewello, T.; Tomé, A. C.; Guldi, D. M. Efficient Low
Driving Force Charge Separation in an Electron Deficient Zn-Porphyrin-Fullerene Donor-Acceptor
Conjugate. J. Phys. Chem. C 2019, 123 (46), 28093–28099. https://doi.org/10.1021/acs.jpcc.9b09790.
(29) Imahori, H.; Yamada, H.; Guldi, D. M.; Endo, Y.; Shimomura, A.; Kundu, S.; Yamada, K.; Okada, T.;
Sakata, Y.; Fukuzumi, S. Comparison of Reorganization Energies for Intra- and Intermolecular Electron
Transfer. Angew. Chemie - Int. Ed. 2002, 41 (13), 2344–2347. https://doi.org/10.1002/1521-
3773(20020703)41:13<2344::AID-ANIE2344>3.0.CO;2-7.
(30) Stangel, C.; Schubert, C.; Kuhri, S.; Rotas, G.; Margraf, J. T.; Regulska, E.; Clark, T.; Torres, T.;
Tagmatarchis, N.; Coutsolelos, A. G.; Guldi, D. M. Tuning the Reorganization Energy of Electron
Transfer in Supramolecular Ensembles - Metalloporphyrin, Oligophenylenevinylenes, and Fullerene -
28
ACS Paragon Plus Environment

Page 29 of 33
Journal of the American Chemical Society
And the Impact on Electron Transfer Kinetics. Nanoscale 2015, 7 (6), 2597–2608.
https://doi.org/10.1039/c4nr05165c.
(31) Imahori, H.; Tkachenko, N. V.; Vehmahen, V.; Tamaki, K.; Lemmetyinen, H.; Sakata, Y.; Fukuzumi, S.
An Extremely Small Reorganization Energy of Electron Transfer in Porphyrin-Fullerene Dyad. J. Phys.
Chem. A 2001, 105 (10), 1750–1756. https://doi.org/10.1021/jp003207n.
1
2
3
4
5
(32) Guldi, D. M.; Hirsch, A.; Scheloske, M.; Dietel, E.; Troisi, A.; Zerbetto, F.; Prato, M. Modulating Charge-
Transfer Interactions in Topologically Different Porphyrin-C60 Dyads. Chem. - A Eur. J. 2003, 9 (20),
4968–4979. https://doi.org/10.1002/chem.200304995.
(33) Di Valentin, M.; Bisol, A.; Agostini, G.; Carbonera, D. Electronic Coupling Effects on Photoinduced
Electron Transfer in Carotene - Porphyrin-Fullerene Triads Detected by Time-Resolved EPR. In Journal
of Chemical Information and Modeling; 2005; Vol. 45, pp 1580–1588. https://doi.org/10.1021/ci050183e.
(34) Higashino, T.; Yamada, T.; Yamamoto, M.; Furube, A.; Tkachenko, N. V.; Miura, T.; Kobori, Y.; Jono,
R.; Yamashita, K.; Imahori, H. Remarkable Dependence of the Final Charge Separation Efficiency on the
Donor-Acceptor Interaction in Photoinduced Electron Transfer. Angew. Chemie - Int. Ed. 2016, 55 (2),
629–633. https://doi.org/10.1002/anie.201509067.
(35) Poddutoori, P. K.; Lim, G. N.; Sandanayaka, A. S. D.; Karr, P. A.; Ito, O.; D’Souza, F.; Pilkington, M.;
Van Der Est, A. Axially Assembled Photosynthetic Reaction Center Mimics Composed of
Tetrathiafulvalene, Aluminum(Iii) Porphyrin and Fullerene Entities. Nanoscale 2015, 7 (28), 12151–
12165. https://doi.org/10.1039/c5nr01675d.
(36) Weiss, E. A.; Ahrens, M. J.; Sinks, L. E.; Gusev, A. V.; Ratner, M. A.; Wasielewski, M. R. Making a
Molecular Wire: Charge and Spin Transport through Para-Phenylene Oligomers. J. Am. Chem. Soc. 2004,
126 (17), 5577–5584. https://doi.org/10.1021/ja0398215.
(37) Schubert, C.; Margraf, J. T.; Clark, T.; Guldi, D. M. Molecular Wires-Impact of π-Conjugation and
Implementation of Molecular Bottlenecks. Chem. Soc. Rev. 2015, 44 (4), 988–998.
https://doi.org/10.1039/c4cs00262h.
(38) Gilbert, M.; Albinsson, B. Photoinduced Charge and Energy Transfer in Molecular Wires. Chemical
Society Reviews. 2015, pp 845–862. https://doi.org/10.1039/c4cs00221k.
(39) Ito, O.; Yamanaka, K. I. Roles of Molecular Wires between Fullerenes and Electron Donors in
Photoinduced Electron Transfer. Bull. Chem. Soc. Jpn. 2009, 82 (3), 316–332.
https://doi.org/10.1246/bcsj.82.316.
(40) Benniston, A. C.; Harriman, A. Charge on the Move: How Electron-Transfer Dynamics Depend on
Molecular Conformation. Chemical Society Reviews. 2006, pp 169–179.
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
https://doi.org/10.1039/b503169a.
(41) Guldi, D. M.; Imahori, H.; Tamaki, K.; Kashiwagi, Y.; Yamada, H.; Sakata, Y.; Fukuzumi, S. A
Molecular Tetrad Allowing Efficient Energy Storage for 1.6 s at 163 K. J. Phys. Chem. A 2004, 108 (4),
541–548. https://doi.org/10.1021/jp036382n.
(42) Imahori, H.; Guldi, D. M.; Tamaki, K.; Yoshida, Y.; Luo, C.; Sakata, Y.; Fukuzumi, S. Charge Separation
in a Novel Artificial Photosynthetic Reaction Center Lives 380 Ms. J. Am. Chem. Soc. 2001, 123 (27),
6617–6628. https://doi.org/10.1021/ja004123v.
(43) Luo, C.; Guldi, D. M.; Imahori, H.; Tamaki, K.; Sakata, Y. Sequential Energy and Electron Transfer in an
Artificial Reaction Center: Formation of a Long-Lived Charge-Separated State. J. Am. Chem. Soc. 2000,
122 (28), 6535–6551. https://doi.org/10.1021/ja993959z.
(44) Lee, S. H.; Blake, I. M.; Larsen, A. G.; McDonald, J. A.; Ohkubo, K.; Fukuzumi, S.; Reimers, J. R.;
Crossley, M. J. Synthetically Tuneable Biomimetic Artificial Photosynthetic Reaction Centres That
Closely Resemble the Natural System in Purple Bacteria. Chem. Sci. 2016, 7 (10), 6534–6550.
https://doi.org/10.1039/c6sc01076h.
(45) Otsuki, J. Supramolecular Approach towards Light-Harvesting Materials Based on Porphyrins and
Chlorophylls. J. Mater. Chem. A 2018, 6 (16), 6710–6753. https://doi.org/10.1039/c7ta11274b.
(46) Kadish, K. M.; Smith, K. M.; Guilard, R. The Porphyrin Handbook; 2012; Vol. 11.
https://doi.org/10.1016/C2009-0-22714-0.
(47) Fukuzumi, S.; Kashiwagi, Y. Photoinduced Electron Transfer in a Supramolecular Triad System
29
ACS Paragon Plus Environment