Excited-State Triplet Equilibria in a Series of Re(I)-Naphthalimide Bichromophores

Article abstract of DOI:10.1021/acs.jpcb.9b05688

We present the synthesis, structural characterization, electronic structure calculations, and the ultrafast and supra-nanosecond photophysical properties of a series of five bichromophores of the general structural formula [Re(5-R-phen)(CO)₃(dm

Full text of DOI:10.1021/acs.jpcb.9b05688

Subscriber access provided by Nottingham Trent University
Article
Excited-State Triplet Equilibria in a Series
of Re(I)-Naphthalimide Bichromophores
James E. Yarnell, Kaylee A Wells, Jonathan R. Palmer, Josue M Breaux, and Felix N. Castellano
J. Phys. Chem. B, Just Accepted Manuscript • DOI: 10.1021/acs.jpcb.9b05688 • Publication Date (Web): 13 Aug 2019
Downloaded from pubs.acs.org on August 13, 2019
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 50
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
Excited-State Triplet Equilibria in a Series of Re(I)-
Naphthalimide Bichromophores
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
†‡
†
†
‡
James E. Yarnell , Kaylee A. Wells , Jonathan R. Palmer , Josué M. Breaux and Felix N.
Castellano†*
†
Department of Chemistry, North Carolina State University, Raleigh, North Carolina 27695-8204,
USA
‡
Department of Chemistry & Chemistry Research Center, United States Air Force Academy,
Colorado Springs, Colorado 80840-6230 USA
ABSTRACT
We present the synthesis, structural characterization, electronic structure calculations, and the
ultrafast and supra-nanosecond photophysical properties of a series of five bichromophores of the
general structural formula [Re(5-R-phen)(CO)
3
(dmap)](PF ), where R is a naphthalimide (NI),
6
phen = 1,10-phenanthroline, and dmap is 4-dimethylaminopyridine. The NI chromophore was
systematically modified at their 4-positions with -H (NI), -Br (BrNI), phenoxy (PONI),
thiobenzene (PSNI), and piperidine (PNI), rendering a series of metal-organic bichromophores
(Re1-Re5, respectively) featuring variability in the singlet and triplet energies in the pendant NI
subunit. Five closely related organic chromophores as well as [Re(phen)(CO)
3
(dmap)](PF ) (Re6)
6
were investigated in parallel to appropriately model the photophysical properties exhibited in the
bichromophores. The excited state processes of all molecules in this study were elucidated using
ACS Paragon Plus Environment
1

The Journal of Physical Chemistry
Page 2 of 50
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
a combination of transient absorption spectroscopy and time-resolved photoluminescence (PL)
spectroscopy, revealing the kinetics of the energy transfer processes occurring between the
appended chromophores. The spectroscopic analysis was further supported by electronic structure
calculations which identified the origin of many of the experimentally observed electronic
transitions.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
INTRODUCTION
Rhenium(I) carbonyl-diimine (Re-CDI) complexes have been studied for many decades because
of their desirable photophysical and redox properties, chemical robustness, and synthetic
1
2, 3
flexibility; making them useful in fields such as photochemical molecular devices, solar energy
conversion,4, 5 luminescence sensing,6-14 and biotechnology,15, 16 among others.17-19 More recently,
these types of molecules have garnered interest as photocatalysts in carbon dioxide reduction,19-24
as the increase of atmospheric CO
2
has steadily risen across the world with no tangible solutions
in sight.25, 26 In these systems, the Re-CDI is used in conjunction with a sacrificial electron donor
and light to catalytically drive the reduction of CO
2
to CO. The most efficient photocatalyst to
reduction quantum yield
mitigation.
reduction is their inability
to absorb strongly in the visible region. To counter this, highly absorbing ruthenium(II) complexes
have been covalently linked to the Re-CDI CO reduction catalyst, resulting in greatly improved
+
date, [Re(bpy)(CO)
3
(P(OEt)
3
)] , was found to have a photocatalytic CO
2
2
7
of 0.38, demonstrating the utility of these types of complexes in atmospheric CO
2
Unfortunately, the drawback to many Re-CDI complexes used in CO
2
2
photocatalytic activity attributed to intermolecular electron transfer between the Ru(II) and Re(I)
species.28-30
ACS Paragon Plus Environment
2

Page 3 of 50
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
In many Re-CDI complexes, the observed photoluminescence (PL) originates from the triplet
metal-to-ligand charge transfer (MLCT) excited state.31-33 Since these molecules have the general
+
molecular formulae Re(L-L)(CO)
3
X or [Re(L-L)(CO)
3
L] , the MLCT excited state can be
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
energetically tuned through modification of the ancillary ligands (X or L) which coarsely
3
4
modulates the Re(I) d-orbital energies, and thus the HOMO energy. Additionally, the π* energies
of the chromophoric ligand (L-L), can also be modified, resulting in predictable changes to the
3
5
LUMO energy. Other types of excited states, such as ligand-centered (LC) excited states, can
also be observed in the triplet manifold of Re-CDI complexes if conjugated aryl systems or organic
chromophores are appended to either the ancillary or chromophoric ligands.36-40 Many of the
synergistic benefits of polychromophoric systems (i.e. excited state lifetime extension, increase in
absorption cross section, etc.) have been documented by our research group,33, 41-48 in addition to
many others.49-60
The attachment of the organic chromophore 4-piperidinyl-naphthalenimide (PNI) to a Re-CDI
complex was previously investigated, revealing profound changes in the resultant photophysical
3
3
properties with respect to the two isolated molecules. The Re(I) and PNI chromophores were
purposefully selected in that study for their characteristic spectroscopic infrared handles that
enabled transient vibrational spectroscopic interrogation of the energy transfer processes occurring
between these moieties. The comprehensive utilization of time-resolved PL emission, transient
absorption, and time-resolved infrared spectroscopies from the femtosecond to microsecond time
regimes, revealed “ping-pong” energy transfer occurring between the PNI and Re-CDI complex.
1
The initial ligand-centered singlet excited state was localized on the PNI chromophore ( PNI) and
underwent Förster-type resonance energy transfer (FRET) to the Re-CDI chromophore (휏
=
퐹푅퐸푇
3
43.5 ps) forming the MLCT excited state. After approximately 20 ns, the excited state returned to
ACS Paragon Plus Environment
3

The Journal of Physical Chemistry
Page 4 of 50
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
the PNI chromophore through a Dexter-like triplet–triplet energy transfer (TTET) process, finally
3
populating the long-lived triplet excited state localized on the PNI chromophore ( PNI). Due to
3
3
-1
the energetic proximity between the MLCT and PNI excited states (ΔE = 1680 cm ) and rapid
forward and reverse triplet-triplet energy transfer (rTTET) reactions with respect to the other
deactivation pathways, a thermal excited state equilibrium became established between these two
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
triplets resulting in a 3000-fold increase in the observed MLCT lifetime; this radiative process is
termed thermally-activated delayed PL. At the conclusion of that study, two questions remained
unanswered. First, if weakly emissive organic chromophores were used instead of PNI whose
quantum yield for fluorescence approaches unity,61-64 would singlet energy transfer to the Re-CDI
3
3
moiety still take place? Second, what MLCT/ NI energy gap is sufficient for sustaining the
equilibrium between these two excited states at room temperature. To address these questions, a
series of five Re(I) bichromophores (Re1-Re5) were synthesized by substituting the 4-position of
the NI subunit pendant to the 5-position of the phenanthroline ligand with various substituents: -
H (NI, Re1), -Br (BrNI, Re2), phenoxy (PONI, Re3), thiobenzene (PSNI, Re4), or piperidine
3
(
PNI, Re5) (see Figure 1) while conserving the Re-CDI complex’s MLCT excited state energetics
across all molecules. The naphthalimide substitutions modulated many of the observed
photophysical properties, including the energies of the associated NI excited states and
fluorescence quantum yields, as observed in the appropriate model chromophores.
ACS Paragon Plus Environment
4

Page 5 of 50
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 1. Naphthalimide chromophores evaluated in this study.
One objective was to investigate how the substitution of the naphthalimide subunit modulated
the efficiency of the energy transfer processes that occurs in the bichromophore. Ultrafast transient
absorption spectroscopy was used to compare the bichromophore to the appropriate model systems
to assign the observed excited state features. Despite being weakly fluorescent, there was still
energy transfer observed to the Re-CDI chromophore in the NI and BrNI containing
bichromophores Re1 and Re2. Of the five bichromophores interrogated, four were shown to
exhibit “ping-pong” energy transfer characteristics, whereas the PSNI containing molecule (Re4)
had photophysical properties echoing those of its model compound.
Another objective was to determine the energy gap between the low-lying triplet states on the
3
3
Re-CDI complex ( MLCT) and the naphthalimide chromophore ( NI) where the excited state
equilibrium would no longer be established at room temperature. To increase the energy gap with
respect to that observed in the Re-PNI complex, the new molecules here replaced the chloride
3
ligand with dimethylaminopyridine (dmap), resulting in a higher energy MLCT excited state.
Additionally, this generated an ionic complex that was more soluble in spectroscopic friendly
solvents including acetonitrile. Structural modification of the naphthalimides resulted in a 0.25 eV
-1
3
(2000 cm ) variation of the NI excited state energy across the series, with the NI being the highest
ACS Paragon Plus Environment
5

The Journal of Physical Chemistry
Page 6 of 50
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
3
and the PNI being the lowest in energy. The change in NI excited state energies resulted in triplet
3
excited state equilibrium and characteristic MLCT delayed PL being observed in the three
3
naphthalimides with the highest energy NI excited states (NI, BrNI, and PONI), whereas PSNI
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
and PNI exclusively featured ligand-localized photophysics with no observed MLCT emission.
EXPERIMENTAL
Reagents and Chemicals. All synthetic manipulations were performed under an inert and dry
nitrogen atmosphere using standard techniques. All reagents were purchased from VWR and used
as received. The precursors NNI, NNI-phen, PNI-phen, and Re(PNI-phen)(CO) Cl, as well as PNI,
3
were synthesized according to their published procedures.33, 44, 46 The synthetic procedures used to
make all the chromophores are outlined in Scheme 1. The identity and purity of all compounds
1
studied were confirmed using H NMR, ATR-FTIR, high-resolution mass spectrometry, and
elemental analysis. Synthetic details for all precursors and products in this study are included in
the Supporting Information. Spectroscopic samples were prepared using spectroscopic grade
tetrahydrofuran or acetonitrile and were degassed using the freeze-pump-thaw technique.65
Characterization data for the studied complexes:
1
2
-(p-tolyl)-1H-benzo[de]isoquinoline-1,3(2H)-dione (NI). H NMR (400 MHz, CD
2
Cl ), δ 8.60
2
(
dd, J = 7.2, 1.1 Hz, 2H), 8.31 (dd, J = 8.1, 1.2 Hz, 2H), 7.81 (dd, J = 8.3, 7.3 Hz, 2H), 7.37 (d, J
=
8.0 Hz, 2H), 7.18 (d, J = 8.2 Hz, 2H), 2.46 (s, 3H). MS [HR-ESI (CH
2
Cl ) m/z] 288.1027
2
+
[M+H] , calcd (C19
H
14NO
2
) 288.1025. Anal. Calcd (found) for C19
H
13NO
2
: C, 79.43 (79.14); H,
4
1
.56 (4.81); N, 4.88 (4.87). ATR-FTIR: 490, 516, 538, 781, 812, 1185, 1234, 1352, 1373, 1514,
-1
587, 1656, 1673, 1699 cm .
ACS Paragon Plus Environment
6

Page 7 of 50
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
1
[
Re(NI-phen)(CO)
3
(dmap)](PF
6
) (Re1). H NMR (400 MHz, CD
2
Cl ), δ 9.58 (ddd, J = 12.8,
2
5
.2, 1.4 Hz, 2H), 8.82 (dd, J = 8.4, 1.4 Hz, 1H), 8.66 (ddd, J = 16.2, 7.3, 1.2 Hz, 2H), 8.51 (dd, J =
8.5, 1.4 Hz, 1H), 8.48 – 8.43 (m, 2H), 8.25 (s, 1H), 8.16 (dd, J = 8.3, 5.2 Hz, 1H), 8.04 (dd, J =
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
8
2
.5, 5.1 Hz, 1H), 7.90 (ddd, J = 8.3, 7.3, 4.6 Hz, 2H), 7.65 (d, J = 7.4 Hz, 2H), 6.27 (d, J = 7.4 Hz,
]+, calcd (C34 185Re)
H), 2.91 (s, 6H). MS [HR-ESI (CH
2
Cl
2
) m/z] 766.1224 [M−PF
6
H
23
N
5
O
5
766.1229. Anal. Calcd (found) for C34
H
23
F
6
N
5
O
5
PRe·0.5H
2
O: C, 44.30 (44.15); H, 2.62 (2.40);
N, 7.60 (7.46). ATR-FTIR: 557, 730, 779, 838, 1234, 1350, 1587, 1626, 1673, 1709, 1903, 2058
-
1
cm .
1
6
-bromo-2-(p-tolyl)-1H-benzo[de]isoquinoline-1,3(2H)-dione (BrNI). H NMR (400 MHz,
Cl ), δ 8.69 – 8.62 (m, 2H), 8.42 (d, J = 7.9 Hz, 1H), 8.10 (d, J = 7.8 Hz, 1H), 7.90 (dd, J =
CD
2
2
8
.5, 7.3 Hz, 1H), 7.37 (d, J = 7.9 Hz, 2H), 7.18 (d, J = 8.2 Hz, 2H), 2.46 (s, 3H). MS [HR-ESI
+
13NO
79Br) 366.0130. Anal. Calcd (found) for
(
CH
2
Cl
2
) m/z] 366.0134 [M+H] , calcd (C19
H
2
C
19
H
12BrNO : C, 62.32 (62.04); H, 3.30 (3.27); N, 3.82 (3.81). ATR-FTIR: 422, 500, 518, 730,
2
81, 814, 1193, 1238, 1344, 1365, 1589, 1656, 1707 cm-1.
7
1
[
Re(BrNI-phen)(CO)
3
(dmap)](PF
6
) (Re2). H NMR (400 MHz, CD
2
2
Cl ), δ 9.58 (ddd, J = 12.1,
5.1, 1.3 Hz, 2H), 8.81 (ddd, J = 9.3, 8.6, 1.2 Hz, 2H), 8.72 (ddd, J = 15.2, 7.3, 1.2 Hz, 1H), 8.52 –
8
.43 (m, 2H), 8.26 (s, 1H), 8.23 – 8.13 (m, 2H), 8.08 – 7.96 (m, 2H), 7.65 (d, J = 7.4 Hz, 2H), 6.27
+
(
d, J = 7.4 Hz, 2H), 2.91 (d, J = 0.5 Hz, 6H). MS [HR-ESI (CH
2
Cl
2
) m/z] 844.0316 [M−PF
PRe·0.5H
6
] ,
1
85
79
calcd (C34
H
22
N
5
O
5
Re Br) 844.0334. Anal. Calcd (found) for C34
H
22BrF
6
N
5
O
5
2
O: C,
40.81 (40.58); H, 2.32 (2.17); N, 7.00 (6.72). ATR-FTIR: 418, 477, 514, 543, 557, 726, 781, 838,
234, 1364, 1363, 1589, 1626, 1673, 1713, 1905, 2028 cm-1.
1
1
6
-phenoxy-2-(p-tolyl)-1H-benzo[de]isoquinoline-1,3(2H)-dione (PONI). H NMR (400 MHz,
CD
2
Cl
2
), δ 8.78 (dd, J = 8.4, 1.2 Hz, 1H), 8.66 (dd, J = 7.3, 1.2 Hz, 1H), 8.45 (d, J = 8.3 Hz, 1H),
ACS Paragon Plus Environment
7

The Journal of Physical Chemistry
Page 8 of 50
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
7
.84 (dd, J = 8.4, 7.3 Hz, 1H), 7.55 – 7.47 (m, 2H), 7.39 – 7.30 (m, 3H), 7.27 – 7.22 (m, 2H), 7.18
(
d, J = 8.2 Hz, 2H), 6.96 (d, J = 8.3 Hz, 1H), 2.46 (s, 3H). MS [HR-ESI (CH
2
Cl
·0.1H
78.84); H, 4.55 (4.73); N, 3.67 (3.67). ATR-FTIR: 485, 696, 753, 779, 804, 1136, 1177, 1201,
2
) m/z] 380.1289
+
[M+H] , calcd (C25
H
18NO
3
) 380.1287. Anal. Calcd (found) for C25
H
17NO
3
2
O: C, 78.77
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(
-1
1
236, 1356, 1575, 1660, 1699 cm .
1
[Re(PONI-phen)(CO) (dmap)](PF
3
6
) (Re3). H NMR (400 MHz, CD
2
Cl ), 9.58 (ddd, J = 11.6,
2
5
7
1
.2, 1.4 Hz, 2H), 8.94 (dd, J = 8.5, 1.2 Hz, 1H), 8.82 (dd, J = 8.4, 1.4 Hz, 1H), 8.71 (ddd, J = 15.9,
.3, 1.2 Hz, 1H), 8.56 – 8.44 (m, 2H), 8.24 (s, 1H), 8.16 (dd, J = 8.3, 5.2 Hz, 1H), 8.08 – 8.01 (m,
H), 7.93 (ddd, J = 8.4, 7.3, 4.5 Hz, 1H), 7.65 (dd, J = 7.4, 2.2 Hz, 2H), 7.59 – 7.50 (m, 2H), 7.38
(td, J = 7.4, 1.2 Hz, 1H), 7.29 (dd, J = 7.5, 1.2 Hz, 2H), 7.01 (dd, J = 8.3, 4.4 Hz, 1H), 6.26 (dd, J
+
=
7.3, 5.2 Hz, 2H), 2.91 (s, 3H), 2.90 (s, 3H). MS [HR-ESI (CH
2
2
Cl ) m/z] 858.1478 [M−PF
6
] ,
calcd (C40
H
27
N
5
O
6
185Re) 858.1491. Anal. Calcd (found) for C40
H
27
F
6
N
5
O
6
PRe·H
2
O: C, 46.97
(47.06); H, 2.86 (2.68); N, 6.85 (6.79). ATR-FTIR: 557, 781, 840, 1206, 1236, 1254, 1356, 1579,
626, 1671, 1707, 1907, 2028 cm-1.
1
1
6
-(phenylthio)-2-(p-tolyl)-1H-benzo[de]isoquinoline-1,3(2H)-dione (PSNI). H NMR (400
MHz, CD Cl ), δ 8.73 (dd, J = 8.5, 1.1 Hz, 1H), 8.64 (dd, J = 7.3, 1.1 Hz, 1H), 8.35 (d, J = 7.9 Hz,
H), 7.84 (dd, J = 8.5, 7.3 Hz, 1H), 7.61 – 7.53 (m, 2H), 7.51 – 7.45 (m, 3H), 7.39 – 7.32 (m, 3H),
2
2
1
7
+
.16 (d, J = 8.1 Hz, 2H), 2.45 (s, 3H). MS [HR-ESI (CH
S) 396.1058. Anal. Calcd (found) for C25
N, 3.54 (3.54). ATR-FTIR: 445, 516, 685, 747, 777, 810, 904, 1134, 1175, 1185, 1230, 1346,
2
Cl
2
) m/z] 396.1058 [M+H] , calcd
(
C
25
H
18NO
2
H
17NO
2
S: C, 75.93 (75.66); H, 4.33 (4.56);
361, 1507, 1562, 1581, 1656, 1697 cm-1.
1
1
[
Re(PSNI-phen)(CO)
3
(dmap)](PF
6
) (Re4). H NMR (400 MHz, CD
2
Cl ), δ 9.57 (ddd, J = 12.6,
2
5.1, 1.4 Hz, 2H), 8.84 (ddd, J = 24.3, 8.5, 1.3 Hz, 2H), 8.70 (ddd, J = 16.3, 7.3, 1.1 Hz, 1H), 8.48
ACS Paragon Plus Environment
8

Page 9 of 50
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
(
dd, J = 8.5, 1.3 Hz, 1H), 8.36 (dd, J = 15.7, 8.0 Hz, 1H), 8.23 (s, 1H), 8.16 (dd, J = 8.3, 5.2 Hz,
1
H), 8.04 (dd, J = 8.5, 5.1 Hz, 1H), 7.93 (ddd, J = 8.5, 7.4, 4.7 Hz, 1H), 7.67 – 7.62 (m, 4H), 7.56
– 7.52 (m, 3H), 7.31 (dd, J = 8.0, 5.0 Hz, 1H), 6.26 (td, J = 5.5, 1.6 Hz, 2H), 2.91 (s, 3H), 2.89 (s,
]+, calcd (C40 S185Re) 874.1262. Anal.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
H). MS [HR-ESI (CH
2
Cl
2
) m/z] 874.1242 [M−PF
6
H
27
N
5
O
5
Calcd (found) for C40
H
27
F
6
N
5
5
O PReS: C, 47.06 (46.98); H, 2.67 (2.68); N, 6.86 (6.78). ATR-
-1
FTIR: 557, 728, 751, 779, 838, 1234, 1363, 1585, 1626, 1671, 1709, 1905, 2028 cm .
1
[
Re(PNI-phen)(CO)
3
(dmap)](PF
6
) (Re5). H NMR (400 MHz, CD
2
Cl ), δ 9.56 (ddd, J = 13.1,
2
5
8
.1, 1.4 Hz, 2H), 8.80 (dd, J = 8.3, 1.4 Hz, 1H), 8.66 – 8.44 (m, 4H), 8.22 (s, 1H), 8.14 (dd, J =
.3, 5.2 Hz, 1H), 8.02 (dd, J = 8.5, 5.1 Hz, 1H), 7.79 (ddd, J = 8.5, 7.3, 5.3 Hz, 1H), 7.64 (d, J =
7.4 Hz, 2H), 7.27 (dd, J = 8.2, 6.1 Hz, 1H), 6.26 (d, J = 7.4 Hz, 2H), 3.36 (m, J = 5.4 Hz, 4H), 2.91
+
(s, 6H), 1.92 (m, J = 10.7 Hz, 4H), 1.78 (m, 2H). MS [HR-ESI (CH
2
Cl
2
) m/z] 849.1945 [M−PF
PRe·H
6
] ,
calcd (C39
H
32
N
6
O
5
185Re) 849.1964. Anal. Calcd (found) for C39
H
32
F
6
N
6
O
5
2
O: C, 46.20
(46.17); H, 3.38 (3.16); N, 8.29 (8.10). ATR-FTIR: 557, 781, 840, 1232, 1365, 1583, 1626, 1664,
701, 1909, 2028 cm-1.
1
1
[
Re(phen)(CO)
3
(dmap)](PF
6
) (Re6). H NMR (400 MHz, CD
2
Cl ), δ 9.51 (dd, J = 5.1, 1.4 Hz,
2
2H), 8.79 (dd, J = 8.3, 1.4 Hz, 2H), 8.17 (s, 2H), 8.10 (dd, J = 8.3, 5.1 Hz, 2H), 7.53 (d, J = 7.4 Hz,
]+, calcd
2
H), 6.19 (d, J = 7.4 Hz, 2H), 2.86 (s, 6H). MS [HR-ESI (CH
2
Cl
2
) m/z] 571.0909 [M−PF
6
(
(
C
22
H
18
N
4
O
3
185Re) 571.0908. Anal. Calcd (found) for C22
H
18
F
6
N
4
O
3
PRe·0.5H O: C, 36.37
2
-1
36.52); H, 2.64 (2.42); N, 7.71 (7.59). ATR-FTIR: 557, 724, 836, 1626, 1897, 1919, 2025 cm .
1
13
General Techniques. H and C NMR spectra were recorded on a Varian Inova 400 NMR
1
operating at a working frequency of 400 MHz for H. All chemical shifts were referenced to
residual solvent signals. High-resolution electron spray ionization mass spectrometry (ESI-MS)
were performed at Michigan State University Mass Spectrometry Facility. Elemental analysis data
ACS Paragon Plus Environment
9

The Journal of Physical Chemistry
Page 10 of 50
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
was measured by Atlantic Microlab, Inc., Norcross, GA. Attenuated total reflectance fourier-
transform infrared (ATR-FTIR) spectroscopy was performed using the Bruker Alpha Platinum
ATR. Optical absorption spectra were measured on a Shimadzu UV-3600 spectrometer. Steady-
state luminescence spectra were obtained from an Edinburgh FS 920. Quantum yield
measurements were performed on NI and BrNI using anthracene in aerated ethanol as the standard
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
6
6
(ΦPL = 0.28), PONI using 9,10-diphenylanthracene in aerated cyclohexane as the standard (ΦPL
=
=
0.97),66 PSNI and Re1-Re6 using [Ru(bpy)
3
](PF
6
) in aerated acetonitrile as the standard (ΦPL
2
0.018), and PNI in acetonitrile using PNI in aerated toluene as the standard (ΦPL = 0.91).64
66
Nanosecond Transient Absorption Spectroscopy. Transient absorption measurements were
collected using a LP920 laser flash photolysis system (Edinburgh Instruments) described
67
previously. Samples were prepared to have a 25 μM concentration (3.0 mJ/pulse) in a 1 cm path
length quartz optical cell and degassed using the freeze-pump-thaw technique. All transient
absorption experiments were performed at room temperature. The reported difference spectra and
kinetic data are the average of 50 laser shots. The ground state absorption spectra were taken before
and after each experiment to ensure there was no sample degradation. The transient kinetic data
was evaluated using the fitting routines available in OriginPro 2016.
Ultrafast Transient Absorption Spectroscopy. The transient absorption measurements were
performed at the NCSU Imaging and Kinetic Spectroscopy (IMAKS) Laboratory using a mode-
6
8
locked Ti:sapphire laser (Coherent Libra) as described previously. The pump beam was directed
into a parametric amplifier (Coherent OPerA Solo) to generate the 350 and 400 nm excitation. The
probe beam was focused onto a calcium fluoride crystal to generate a white light continuum
between 350 to 775 nm. The pump beam was focused and overlapped with the probe beam through
a 2 mm path length cuvette. The ground-state absorption spectra were taken before and after each
ACS Paragon Plus Environment
1
0

Page 11 of 50
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
experiment to ensure there was no sample photo-degradation during the experiment. The transient
kinetic data at specific wavelengths was evaluated using the fitting routines available in OriginPro
2016. The global fit analysis was performed using Surface Xplorer available from Ultrafast
Systems.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Time-Resolved Emission Intensity Decay Measurements. Single wavelength emission intensity
decays at 77 K and room temperature were acquired with an LP920 laser flash photolysis system
(Edinburgh Instruments) using the Vibrant 355 LD Nd:YAG/OPO system (OPOTEK) as the
excitation source (λex = 355/410 nm, ∼3.0 mJ/pulse). Spectroscopic samples were prepared to
have a 25 μM concentration and PL transients were collected at emission maximums for most
experiments, with all exceptions noted in text. Time-gated emission spectra were collected using
the same system. Emission spectra and PL transients were collected with an iStar ICCD camera
(Andor Technology), which was controlled by the L900 software program (Edinburgh
Instruments). Reported time-gated emission spectra are the average of 50 laser flashes.
Density Functional Theory (DFT) Calculations. The calculations utilized in this study were
performed using the Gaussian 09 software package (revision D.01)69 and the computation
resources of the North Carolina State University High Performance Computing Center. Ground
state and triplet state geometry optimizations were performed using the B3LYP,70, 71 M06,72 and
CAM-B3LYP functionals,73 along with the def2-SVP basis set of the Alrichs group as
7
4
75
implemented in Gaussian 09. The Stuttgart−Dresden effective core potentials (ECPs) were used
to replace the core electrons in rhenium for all calculations. The polarizable continuum model was
76
used to simulate the acetonitrile solvent environment for all calculations. Frequency calculations
were performed on all optimized structures and no imaginary frequencies were obtained. The TD-
DFT calculations were performed using the same conditions as described for the geometry
ACS Paragon Plus Environment
1
1

The Journal of Physical Chemistry
Page 12 of 50
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
optimizations.77-79 The energy and oscillator strengths were computed for each of the 50 lowest
singlet excitations. The natural transition orbitals of the low-lying singlet and triplet transitions
were generated using GaussView 5.0.80
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
RESULTS AND DISCUSSION
Syntheses. The schemes used to synthesize all the chromophores in this study are collected in
Scheme 1. Briefly, the naphthalimide chromophores were prepared by departing from the
corresponding naphthalene anhydride (either unsubstituted or substituted with Br or NO in the 4-
2
position) and reacting with a primary amine (p-toluidine or 5-amino-1,10-phenanthronline) using
published protocols to generate the NI and BrNI model chromophores and ligands (NI-phen and
4
6
BrNI-phen), along with the nitronaphthalimide intermediates (NNI and NNI-phen). The nitro
group was substituted by refluxing the nitronaphthalimide in DMF with a phenol or thiolbenzene
under basic conditions or piperidine to yield the other three naphthalimide chromophores and
ligands (PONI, PSNI, PNI, PONI-phen, PSNI-phen, and PNI-phen).33, 44, 81, 82 The 5-substituted
34
phenanthroline ligands were coordinated to Re(CO)
5
Cl using published protocols. The chloride
ligand was subsequently removed by refluxing fac-Re(L-L)(CO)
3
Cl with silver triflate in ethanol,
forming the solvato-complex. The silver chloride precipitate was removed by filtration over Celite.
The dmap ligand was finally added to the isolated solvato-complex and refluxed in ethanol to yield
the desired product (Re1-Re6). The final isolated products (facial isomers) were structurally
1
characterized using H NMR spectroscopy, high resolution mass spectroscopy, elemental analysis,
and FT-IR spectroscopy. Furthermore, the molecular structures of the bichromophores Re1 to Re5
1
1
were confirmed using H- H COSY NMR spectroscopy.
ACS Paragon Plus Environment
1
2

Page 13 of 50
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
Scheme 1. Synthetic procedures for the preparation of NI, BrNI, PONI, PSNI and Re1-Re6.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ACS Paragon Plus Environment
1
3

The Journal of Physical Chemistry
Page 14 of 50
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
UV-Vis Electronic Spectra. The electronic spectra for the model compounds and
bichromophores measured in acetonitrile are presented in Figure 2, with additional spectroscopic
data collected in Table 1. The NI and BrNI chromophores both have structured lowest energy
electronic transitions with a maximum absorbance corresponding to a π→π* transition at 332 and
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
40 nm, respectively (Figure 2a). In PONI, PSNI, and PNI, the low energy transition, which is
now broad and featureless, red shifts progressively (359, 384, and 407 nm, respectively) and is
consistent with possessing significant charge transfer character as demonstrated in prior work,
Figure 2a.33, 44, 46 All five naphthalimide model chromophores possess relatively intense low
-1
-1
energy transitions with molar extinction coefficients ranging from 11300 to 17100 M cm , Figure
2a. The model Re(I) compound (Re6) has a broad and featureless absorption band centered at 368
nm which is assigned to the d휋(Re)→휋*(phen) MLCT transitions and a higher energy band at 276
nm which corresponds to the π→π* transitions localized on the phen ligand, Figure 2b. As also
illustrated in Figure 2b, the low energy electronic transitions observed in each of the
bichromophores (Re1-Re5) emulates that expected for a simple mixture of the composing
molecules. This is illustrated in Figures S40-S41, where the absorption spectrum of Re6 is added
to that of the appropriate NI model compound’s absorption spectrum. In all cases, the additive and
measured spectra of the five bichromophores were found to be in good quantitative agreement,
illustrating that there is no significant electronic communication between the respective subunits.
ACS Paragon Plus Environment
1
4

Page 15 of 50
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
2
2
1
1
5000
0000
5000
0000
5000
0
NI
A
BrNI
PONI
PSNI
PNI
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
50
300
350
400
450
500
Re1 B
Re2
Re3
Re4
Re5
5
4
0000
0000
30000
Re6
2
1
0000
0000
0
2
50
300
350
400
450
500
Wavelength (nm)
Figure 2. (A) Electronic spectra of NI, BrNI, PONI, PSNI, and PNI recorded in acetonitrile. (B)
Electronic spectra of Re1-Re6 recorded in acetonitrile.
Electronic Structure Calculations. Time-dependent density functional theory (TD-DFT)
calculations were performed on all the molecules to glean insight into the nature of the lowest
energy electronic transitions observed experimentally. The B3LYP, M06, and CAM-B3LYP
functionals were used to perform the ground state geometric optimization and the corresponding
TD-DFT calculations, Table S2. Across the series of naphthalimide model chromophores, the M06
functional yielded the best quantitative modeling of the transition energies, whereas the B3LYP
functional was consistently lower in energy and the CAM-B3LYP functional was consistently
ACS Paragon Plus Environment
1
5

The Journal of Physical Chemistry
Page 16 of 50
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
higher in energy. Unfortunately, both the B3LYP and M06 functionals did not adequately model
the PSNI chromophore, as both predicted a low energy transition with significant oscillator
strength that was not observed experimentally. By using the long-range corrected functional,
CAM-B3LYP, the energies of the calculated low energy transitions were all overcorrected
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(
rendered too high in energy), however the trend in relative energies observed in the experimental
data were revealed in the low energy transitions using this functional. Natural transition orbital
analysis for the S →S and S →S transitions using the M06 and CAM-B3LYP functionals was
performed on the naphthalimide chromophores, Figures S82-S86. The S →S transition across the
series can be best described as being π→π* in character, with a small amount of charge transfer
observed in PONI and a significant amount observed in both PSNI and PNI. The S →S transition
varied a greater amount across the series and its value was functional dependent. This transition is
0
1
0
2
0
1
0
2
considerably weaker than the S
experimentally.
0
→S transition (PSNI is the only exception) and was not observed
1
The same TD-DFT analysis was performed on the rhenium chromophores (Re1-Re6) with the
calculated transitions available in the Supporting Information, Table S4. Using both the B3LYP
and M06 functionals, the calculated charge transfer transition for the Re(I) model (Re6) was
predicted as too low in energy when compared to the experimental data (0.54 and 0.31 eV lower
in energy for the B3LYP and M06 functionals, respectively). Moreover, many new charge transfer
transitions in the bichromophores were calculated that were simply not observed experimentally.
As with the naphthalimide model compounds, the CAM-B3LYP functional gave the best
quantitative match to experimental data based on the trends observed experimentally. Natural
transition orbital analysis for the S
0
→S
1
, S
0
→S
2
, and S
0
→S transitions using the M06 and CAM-
n
ACS Paragon Plus Environment
1
6

Page 17 of 50
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
B3LYP functionals were performed on Re1-Re6, see Figures S87-S92. In the model MLCT
chromophore Re6, the S
0
→S transition can best be described as charge transfer with electronic
1
density originating from the metal and dmap ligands and transitioning to the phenanthroline ligand.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
This type of transition was observed in all the bichromophores near 3.66 eV (using the CAM-
B3LYP functional) and represents the S
0
→S
1
transition in Re1-Re3 and the S
0
→S transition in
2
Re4-Re5. The ligand-centered transition localized on the naphthalimide chromophore was also
clearly observed in all instances (based on energy and oscillator strength). Using the CAM-B3LYP
functional, the NI-localized LC transition is the S
0
→S
4
transition in Re1, the S
0
→S transition in
3
Re2 and Re3, and the S →S transition in Re4 and Re5.
0
1
ACS Paragon Plus Environment
1
7

The Journal of Physical Chemistry
Page 18 of 50
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Table 1. Static photophysical properties of the NI, BrNI, PONI, PSNI, PNI model compounds
and Re1-Re6 measured in acetonitrile (black) and THF (blue).
λ
abs max/nm
λ
(RT)
em max/nm
λ
em max/nm
Φ
in air
em
Φ
em
Molecule
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
-1
-1 [a]
[a]
[b,c]
[a]
[a,d]
(ε/M cm )
(77 K)
deaerated
---
[
e]
e]
NI
332 (14200)
330 (12500)
340 (17100)
374, 506
375, 486
387, 537
379, 514
431
538
0.0005
0.0004
0.0005
0.0005
0.18[g]
[
[e]
e]
BrNI
PONI
PSNI
PNI
Re1
553
---
---
---
---
[
3
39 (16300)
359 (15601)
57 (14300)
[
f]
416, 547
0.038[g]
3
422
[
f]
[h]
384 (14700)
382 (13900)
407 (11300)
518
500
450, 579
0.004
[
h]
0.006
0.084
0.81[
0.0030[h]
0.0028[h]
0.0014[h]
0.0014[h]
0.0023[h]
0.0014[h]
0.0011[h]
0.0010[h]
[f]
[i]
539
516
630[d]
614[d]
626[d]
617[d]
492, 605
i]
3
98 (11300)
334 (20400)
34 (20200)
536, 512
555, 516
0.014[h]
0.020[h]
0.011[h]
0.015[h]
0.014[h]
0.018[h]
0.0035[h]
0.0082[h]
0.0031[h]
0.011[h]
0.024[h]
0.029[h]
3
Re2
343 (23200)
343 (23900)
364 (21600)
Re3
435, 628[d] 546, 514
3
63 (24000)
392 (20700)
93 (20700)
415 (15800)
10 (15400)
368 (3160)
72 (3600)
611[d]
613[d]
613, 645[d]
548[d]
523[d]
602[d]
601[d]
Re4
577, 516
595, 512
507
3
Re5
0.0032[h]
0.0087[h]
0.011[h]
0.016[h]
4
Re6
3
[a]
Room temperature measurements were made with ACN or THF as a solvent. [b] 77 K
measurements were made with 2-methyl-THF. Emission maximum determined from time-
gated PL spectra using the LP 920. Samples deaerated using freeze-pump-thaw method.
[c]
[
d]
[e]
Quantum yields were measured using anthracene in aerated ethanol as the standard (ΦPL
=
6
6 [f]
[g]
0
.28).
Emission maximum determined from steady-state PL spectra using the FS 920.
Quantum yields were measured using 9,10-diphenylanthracene in aerated cyclohexane as the
6
6 [h]
standard (ΦPL = 0.97).
Quantum yields were measured using [Ru(bpy)
3
](PF
6
)
2
in aerated
66 [i]
acetonitrile as the standard (ΦPL = 0.018).
aerated toluene as the standard (ΦPL = 0.91).
Quantum yields were measured using PNI in
64
Static and Dynamic PL Spectroscopy. The static PL spectra recorded for the model
chromophores and Re(I) bichromophores in acetonitrile are presented in Figure 3, with additional
ACS Paragon Plus Environment
1
8

Page 19 of 50
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
spectroscopic data collected in Table 1. The NI and BrNI molecules have two observable emission
bands in aerated acetonitrile at room temperature, Figure 3a. The high energy band in both
molecules was structured with emission maxima at 374 and 387 nm, respectively, consistent with
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
fluorescence from the S π→π* excited state. The low energy PL observed was broad, featureless,
1
and sensitive to solvent polarity, with emission maxima at 506 and 537 nm, respectively.
Additionally, excitation spectra were measured in acetonitrile at the high and low energy emission
maxima for both dual-emitting chromophores. Regardless of the monitored wavelength, the
recorded spectra mimic the low energy transition observed in the electronic spectra of NI and
BrNI, supporting the assignment as dual emission (Figures S43-44). The low-energy emission
bands likely originate from a conformational change of the molecule and the formation of planar-
intramolecular charge transfer (PICT) excited states. This dual emission phenomenon has been
documented by multiple research groups and demonstrated to occur when the N-aryl group of the
1,8-napthalimide is substituted with an electronically donating substituent.63, 83-85 First documented
by Bérces and coworkers, a series of 1,8-napthalimides were prepared using an unsubstituted
phenyl ring as the model and adding electron donating or withdrawing substituents to the para
position of the N-aryl group (among many other NI derivatives). In their model system, only the
short-wave fluorescence emission (386 nm in ACN) originating from the S π→π* excited state
1
was observed. By adding an electron donating group (p-MeO) to the N-aryl group, the two low-
lying excited states undergo pseudo-Jahn-Teller coupling which allows the excited state to cross
to the nearly isoenergetic S potential energy surface. Once on this surface, the N-aryl group rotates
2
to nearly a coplanar geometry with the naphthalimide and stabilizes the charge transfer excited
state. In the p-MeO substituted molecule, the long-wave fluorescence (603 nm in ACN) originating
8
5
from the S
2
CT excited state is observed almost exclusively. In the case of NI and BrNI, the p-
ACS Paragon Plus Environment
1
9

The Journal of Physical Chemistry
Page 20 of 50
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
methyl on the N-aryl group is electronically more donating than hydrogen but less so than a p-
MeO so it is probable that some degree of pseudo-Jahn-Teller coupling takes place, therefore
supporting the observation of both short-wave and long-wave fluorescence bands in NI and BrNI.
The PONI, PSNI, and PNI molecules all have broad and featureless fluorescence spectra, with
maxima at 431, 518, and 539 nm, respectively, Figure 3a. When PONI, PSNI, and PNI are
dissolved in lower polarity solvents such as THF or toluene, they experience positive
solvatochromism whereas the corresponding PL spectra shift to lower energy with increasing
medium polarity. The combination of the broad featureless emission spectra and solvatochromic
nature of this PL are both indicative of charge transfer character in the emissive state of these three
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
molecules. In acetonitrile, the emission quantum yield of PONI is substantial (훷 = 0.18),
퐹
followed by PNI and PSNI (훷 = 0.084 and 0.004, respectively). When dissolved in a lower
퐹
polarity solvent, such as THF, the PNI becomes the most emissive (훷 = 0.81), followed by PONI
퐹
and PSNI (훷 = 0.038 and 0.006, respectively). In the case of NI, BrNI, PSNI, the low
퐹
fluorescence quantum yields are likely due to efficient intersystem crossing (ISC) from the S
1
6
1
excited state to the triplet manifold. The PONI chromophore has a substantially larger
fluorescence quantum yield and likely has less efficient ISC. The emission intensity of PNI is
extremely solvent dependent where fluorescence quantum yields greater than 90% persist in
nonpolar solvents such as toluene. In this case, the fluorescence is likely quenched through another
mechanism, such as twisted intermolecular charge transfer (TICT) states as the solvent polarity
plays a large role in TICT excited state decay.86-88
ACS Paragon Plus Environment
2
0

Page 21 of 50
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
NI
A
1
.0
.8
BrNI
PONI
PSNI
PNI
0
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
0.6
0.4
0.2
0.0
4
00
500
600
700
800
Re1
B
1
.0
.8
Re2
Re3
Re4
Re5
Re6
0
0.6
0.4
0.2
0.0
400
500
600
700
800
Wavelength (nm)
Figure 3. Room temperature emission spectra of (A) NI (휆ex = 325 nm), BrNI (휆ex = 325 nm),
PONI (휆ex = 355 nm), PSNI (휆ex = 395), and PNI (휆ex = 395) recorded in aerated acetonitrile and
of (B) Re1-Re6 (휆ex = 385) recorded in deaerated acetonitrile.
The Re(I) model complex (Re6) displays a broad, featureless PL emission band centered at 602
3
nm, which originates from the MLCT excited state, Figure 3b. The light emission of Re6 is
quenched by molecular oxygen and possesses a lifetime of 176 ns in deaerated acetonitrile (Figure
3
S48), consistent with the lowest excited state being triplet in origin. Similarly, MLCT PL was
observed in Re1-Re4, however the spectral profiles are slightly red-shifted to 628 nm in these
ACS Paragon Plus Environment
2
1

The Journal of Physical Chemistry
Page 22 of 50
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
molecules, Figure 3b. Furthermore, the naphthalimide fluorescence is quantitatively quenched in
Re1 and Re2 (not shown), predominantly quenched in Re3, appears as a shoulder in Re4, and is
the PL spectrum primarily observed in Re5, Figure 3b. Since prior studies have demonstrated that
Förster-like energy transfer is the primary quenching pathway of the naphthalimide fluorescence
in related metal-organic chromophores,33, 44 this implies that the energy transfer efficiency is
decreasing as the fluorescence of the appended naphthalimide red-shifts. The fact that there is less
spectral overlap between the singlet fluorescence donor emission (naphthalimide) and the acceptor
absorbance (Re-CDI) as the naphthalimide red-shifts supports these experimental observations
being consistent with resonance energy transfer quenching across the series. The PL quantum
yields of the bichromophores are smaller than that of Re6 with Re1-Re3 having about half the
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
efficiency when compared to the model chromophore (훷 = 0.013, 0.009, and 0.011,
푃퐿
respectively). The other two bichromophores (Re4 and Re5) have substantially less MLCT-based
PL and in the case of Re5, the observed spectrum is composed of predominately unquenched
fluorescence from the PNI moiety. As demonstrated in our previous study, the MLCT
phosphorescence in the bichromophores is mostly derived from the triplet excited state equilibrium
3
3
3
formed between the MLCT and LC states, with the population of the higher energy MLCT
33
3
excited state based on a Boltzmann distribution. Therefore, the lower the energy of the NI
3
excited state (as in Re4 and Re5), the lower the relative population of the MLCT excited state
and therefore less PL emanating from that state.
PL intensity decays of the metal-organic bichromophores were collected at 600 nm to investigate
3
the population decay kinetics of the MLCT state whose data are collected in Table 2. In Re1-Re3,
the emission intensity decays were adequately fit to biexponential functions whose average
ACS Paragon Plus Environment
2
2

Page 23 of 50
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
lifetimes were much longer lived with respect to the Re-CDI model complex, suggesting that
3
3
33, 44
equilibrium is indeed operative between the MLCT and NI excited states.
The biexponential
nature of these decays have been previously observed and attributed to concentration-dependent
self-quenching.33, 44, 46, 61 To simplify comparisons between molecules, average lifetimes were used
and it was determined that Re1 possessed the shortest lifetime when compared to Re2 and Re3
measured at matched concentrations under identical experimental conditions. Furthermore, time-
gated PL spectra acquired at numerous delay times confirmed that the emission originated from
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
the MLCT excited state in all instances, Figure S49. There was no evidence of any delayed
3MLCT PL in Re4 and Re5, instead only prompt emission from a combination of the
naphthalimide fluorescence and strongly quenched MLCT PL was observed. Time-gated PL
spectra confirmed these findings as there was little delayed emission observed beyond those
3
prompt signals seen in Re4 and Re5. Since these molecules have the lowest NI energies, it is
3
likely that the energy gap is too large to enable thermal population of the higher lying MLCT
excited state at room temperature and therefore no delayed PL could be observed.
3
Since phosphorescence was not observed from the LC excited state in any of the metal-organic
bichromophores at room temperature, each molecule was dissolved in 2-methyltetrahydrofuran (2-
MeTHF) and frozen at 77 K by immersion in liquid N . At this temperature, the distinct
2
3
phosphorescence from the NI excited states in each instance (Re1-Re5) was experimentally
1
3
distinguished from the NI and MLCT states using time-gated PL, Figure 4. The PL spectra
measured at 1 µs was broad and featureless, with a maximum near 515 nm for all bichromophores
(
Re1-Re5), very closely matching that of the model MLCT complex Re6. We conclude that the
3
origin of the emission observed at this time delay emanates from the MLCT state resident on the
ACS Paragon Plus Environment
2
3

The Journal of Physical Chemistry
Page 24 of 50
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Re-CDI complex. Extending the observation onset to 25 ms after the laser pulse, the observed PL
3
was red-shifted with respect to the MLCT PL spectrum, and is vibronically structured, suggesting
3
that the emission now originates from the NI excited state. To confirm these assignments, the
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
naphthalimide models were also measured at 77 K under identical conditions, revealing emission
spectra that were very close in energy and spectral features to those of the bichromophores
measured at 25 ms delay, Figure S50. The PL decays at 77 K were well modeled using
biexponential kinetics for the bichromophores where the first component corresponded to the
3
decay of the MLCT excited state (τ ~ 11 µs) and the second component corresponded to the decay
3
3
3
of the extremely long-lived NI excited state (τ ≥ 10 ms). Since both the MLCT and NI excited
states decay independently at 77 K, the excited state equilibrium observed at room temperature in
3
Re1-Re3 no longer takes place, thereby eliminating any observation of delayed MLCT PL.
Table 2. Time-resolved PL and TA data recorded for Re1-Re6 in acetonitrile.
휏
휏
휏
휏
휏
푃퐿
푃퐿 푎푣푒
14.5 µs
56.6 µs
54.7 µs
---
푇퐴
푇퐴 푎푣푒
푃퐿
Molecule
[
a]
[b]
[c]
[d]
[e]
(298 K)
(298 K)
(298 K)
(298 K)
(77 K)
Re1
Re2
Re3
Re4
Re5
Re6
6.5 µs, 27.0 µs
9.5 µs, 60.7 µs
16 µs, 64.6 µs
< 15 ns
6.6 µs, 19.6 µs
11.9 µs
11.1 µs, 544 ms
10.6 µs, 8.6 ms
12.6 µs, 612 ms
10.6 µs, 102 ms
9.2 µs, 288 ms
13.6 µs
21.9 µs, 81.3 µs 53.7 µs
18.8 µs, 70.0 µs 42.2 µs
34.2 µs, 186 µs
43.9 µs, 177 µs
182 ns
116 µs
136 µs
---
< 15 ns
---
176 ns
---
[
a]
PL decay of Re1-Re6 (25 ± 2 μM) measured at 600 nm in deaerated acetonitrile at room
[
b]
temperature.
components measured in deaerated acetonitrile at room temperature.
intensity decay of Re1-Re6 (25 ± 2 μM) measured at 470 nm (Re1), 480 nm (Re2), 480 nm
Re3), 670 nm (Re4), 480 nm (Re5), 475 nm (Re6) in deaerated acetonitrile at room temperature.
Average time constant determined from a weighted average of biexponential
[c]
Transient absorption
(
[
d]
Average time constant determined from a weighted average of biexponential components
[
e]
measured in deaerated acetonitrile at room temperature. PL decay of Re1-Re6 measured at 535
nm (Re1), 555 nm (Re2), 550 nm (Re3), 575 nm (Re4), 600 nm (Re5), 500 nm (Re6) in 2-
MeTHF at 77 K.
ACS Paragon Plus Environment
2
4

Page 25 of 50
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 4. Time-resolved 77 K PL emission spectra of Re1-Re6 at the specified delay times
following a 7 ns fwhm, 350 nm (Re1-Re3) or 400 nm (Re4-Re6) laser pulse in 2-MeTHF.
ACS Paragon Plus Environment
2
5

The Journal of Physical Chemistry
Page 26 of 50
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Transient Absorption Spectroscopy. When attaching the NI or BrNI structural motif to the Re(I)
chromophore, as in the case for Re1 and Re2, the observed excited state processes are consistent
with energy transfer occurring between the chromophores. As presented in Figure 5, following
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
50 nm excitation, a transient absorption feature near 475 nm is promptly observed and
quantitatively matches that of the S excited state absorptions observed in the NI moiety, Figure
1
S51. Unlike the model chromophore, this transient feature in Re1 decays rapidly (τ = 400 fs) with
a new absorption appearing at 370 nm (τ = 315 fs) along with a broad feature near 525 nm that
spans into the NIR region. This excited state transient is completely formed by 5 ps and is assigned
3
as the MLCT state of the Re(I) chromophore as it matches the excited state features observed in
the Re(I) model (Re6) measured under the same experimental conditions, Figure S68. The
1
3
transition from the LC, π,π* state localized on the NI chromophore to the MLCT state was found
to have a time constant of 400 fs as this matches the decay of the 470 nm transient feature and the
subsequent growth of the 370 nm feature, Figure S54. Going from 5 ps through the end of the
delay stage on the nanosecond timescale, the broad excited state absorbance from 500 nm to the
NIR decreases in intensity whereas the transient feature at 470 nm grows in; the latter has been
3
previously assigned to the LC triplet excited state localized on the NI chromophore. The kinetic
3
3
transition from the MLCT excited state to the LC excited state was tracked by analyzing the
growth of the 470 nm transient feature, which corresponded to a time constant of 4.1 ± 0.2 ns.
Since NI has its major low energy absorption band positioned well into the UV, the MLCT state
could be selectively excited in Re1, Figure S53. Using 400 nm as the excitation source, the initially
3
observed features at 510 fs revealed the presence of the MLCT state with a concomitant growth
3
of the LC feature (470 nm) occurring with a similar time constant as measured with 350 nm laser
pulses, τ = 4.0 ns. Re2 displays similar excited state features and kinetics to that measured in Re1,
ACS Paragon Plus Environment
2
6

Page 27 of 50
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
demonstrating that the addition of the bromo substituent to the naphthalimide does not drastically
alter the cascade of excited states or their relative time constants, Figures S56-S58.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
0.030
2
50 fs A
0.025
0.020
0.015
0.010
0.005
0.000
540 fs
850 fs
1.5 ps
3 ps
5
ps
-0.005
-0.010
4
00 450 500 550 600 650 700 750
0.020
5
5
ps
0 ps
1500 ps
3000 ps
6100 ps
B
0.015
500 ps
0.010
0.005
0.000
0.005
-
400 450 500 550 600 650 700 750
Wavelength (nm)
Figure 5. Excited-state absorption difference spectra of Re1 in ACN during short delay times
(A) and long delay times (B) following 350 nm pulsed excitation (105 fs fwhm).
When the PONI chromophore is attached to the Re(I) subunit, as in Re3, the effects of energy
transfer between chromophores are also clearly delineated. When using 350 nm laser pulses, the
excited state difference spectrum of Re3 at 250 fs quantitatively matches that of the PONI model,
ACS Paragon Plus Environment
2
7

The Journal of Physical Chemistry
Page 28 of 50
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
suggesting that the initial excited state is indeed localized on the PONI chromophore, Figure 6.
The excited state absorbance at 450 nm undergoes a similar thermalization process observed with
the PONI model, except now when the blue-shifted excited state absorbance at 425 nm decays, a
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
new excited state feature at 360 nm grows in and matches that of the MLCT excited state of the
Re(I) model Re6. The change in these absorption features suggests vibrational relaxation from the
initially populated Franck-Condon state localized on the PONI moiety (τ = 170 fs) is followed by
resonance energy transfer to the Re(I) chromophore (τ = 1.5 ps) with subsequent ultrafast
3
intersystem crossing to the MLCT excited state. When examining the kinetics at 370 nm, the data
is best fit to a sum of three exponentials yielding one time constant (τ = 26 ps) matching that
observed in Re6 measured under the same conditions, Figure S69. Towards the end of the delay
stage, the excited state absorbance at 360 nm and at 725 nm decrease in intensity as the excited
state feature at 475 nm, assigned as the triplet localized on the PONI, gains intensity. The growth
of this 475 nm feature is best fit to a single exponential function with a time constant of 4.4 ns,
nearly matching what was observed in Re1 and Re2, Figure S60. The change in these spectral
features suggest a triplet-triplet energy transfer process returning the excited state density from the
3
3
MLCT state back to the LC state on the PONI ligand.
ACS Paragon Plus Environment
2
8

Page 29 of 50
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
0
0
0
.02
.01
.00
A
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
-
0.01
0.02
2
4
50 fs
50 fs
2 ps
3 ps
5 ps
15 ps
50 ps
-
740 fs
1 ps
1.5 ps
-
0.03
4
00 450 500 550 600 650 700 750
0
.020
5
1
5
0 ps
50 ps
00 ps
1500 ps
3000 ps
6100 ps
B
0.015
0
0
0
.010
.005
.000
4
00 450 500 550 600 650 700 750
Wavelength (nm)
Figure 6. Excited-state absorption difference spectra of Re3 measured in acetonitrile across
short (A) and long (B) delay times following 350 nm pulsed laser excitation (105 fs fwhm).
Unlike Re1-Re3, when coupling the PSNI subunit to the Re(I) complex to yield Re4, there were
no significant changes in terms of the transient absorption difference spectra and kinetics in
comparison to the PSNI model chromophore. As displayed in Figure 7, the excited state absorption
features for the short and intermediate time ranges in Re4 nearly match those observed in PSNI,
3
3
suggesting that the LC formation does not proceed through the MLCT excited state as was
observed in Re1-Re3. Following the first relaxation process at 450 nm, the kinetics in Re4 are best
ACS Paragon Plus Environment
2
9