Photophysical properties of a boron analogue of coumarin

Article abstract of DOI:10.1039/d1cp02386a

In this report, we present a study into the structure and electronic properties of difluoroboronsalicylaldoxime (DFBS), a boron-based structural analog of coumarin. The modification of the heterocyclic ring of coumarin with boron results in a compound with similar structural parameters and molecular orbitals to coumarin. DFT and TDDFT calculations reveal a significant stabilization of the LUMO in DFBS; this is supported by a ～40 nm red shift of the lowest electronic transition in the absorption spectrum. Interestingly, DFBS is emissive, while unmodified coumarin is effectively non-radiative. Comparisons between DFBS, emissive coumarin variants, and unmodified coumarin suggest that the charge transfer character of the transition contributes to the fluorescence. This journal is

Full text of DOI:10.1039/d1cp02386a

PCCP
View Article Online
View Journal
PAPER
Photophysical properties of a boron analogue
of coumarin†
Cite this: DOI: 10.1039/d1cp02386a
b
Huayi Wang,a Briana R. Schrage,a Kana Takematsu
and
a
Christopher J. Ziegler
*
In this report, we present a study into the structure and electronic properties of difluoroboronsalicylaldoxime
(DFBS), a boron-based structural analog of coumarin. The modification of the heterocyclic ring of coumarin
with boron results in a compound with similar structural parameters and molecular orbitals to coumarin. DFT
and TDDFT calculations reveal a significant stabilization of the LUMO in DFBS; this is supported by a B40 nm
red shift of the lowest electronic transition in the absorption spectrum. Interestingly, DFBS is emissive, while
unmodified coumarin is effectively non-radiative. Comparisons between DFBS, emissive coumarin variants, and
unmodified coumarin suggest that the charge transfer character of the transition contributes to the
fluorescence.
Received 28th May 2021,
Accepted 12th August 2021
DOI: 10.1039/d1cp02386a
rsc.li/pccp
in polycyclic aromatic compounds provide useful targets. This
approach has not been widely explored as a synthetic strategy,
Introduction
Fluorophores continue to be essential for sensors, dyes, and and among BF2 chromophores that have been produced, their
energy harvesting applications, and the development of new photophysics and electronic structures have not been probed
fluorophore structures is a highly active area of investigation.1–3 extensively. A good example of such a system is difluoro-boron-
In particular, work on the design and characterization of new triaza-anthracene, a BF2 analogue of anthracene.7 There have only
BF2 containing chromophores has attracted significant interest been a few studies on this compound, and only one publication on
over the past decade. The most well studied and utilized its photophysical behaviour has appeared in the literature to date.7
example of this family of chromophores is BODIPY (4,4-
difluoro-4-bora-3a,4a-diaza-s-indacene, Fig. 1).1 First discov-
ered in 1968,1 the fluorescent properties of BODIPY were not
pursued until two decades later.4 Since its introduction as a
commercial fluorophore by Molecular Probes, BODIPY has
become one of the most important photoactive compounds
used in biological imaging and photonic materials.2–4
Alternate BF2 chromophore development has moved in several
directions, from modification of the BODIPY structure, as seen in
the aza-BODIPY class of compounds,5 to the creation of new
structural motifs, such as in the BOPHY (bis(difluoroboron)-1,2-
bis((pyrrol-2-yl)methylene)hydrazine) system.6 In the latter case, the
BOPHY fluorophore can be considered as a structural analogue of
dicyclopenta[b,g]naphthalene, just as BODIPY can be considered a
BF2 modified version of indacene (Fig. 1). Thus, both BODIPY and
BOPHY have analogous structures that are polycyclic aromatic
organic compounds. As a resource for the development and inves-
tigation of new BF2 based chromophores, the architectures present
a Department of Chemistry, University of Akron, Akron, Ohio 44325-3601, USA.
E-mail: ziegler@uakron.edu
b Department of Chemistry, Bowdoin College, Brunswick, Maine 04011, USA
† Electronic supplementary information (ESI) available: Spectroscopic data, DFT
Fig. 1 Comparison of organic chromophores and their organoborane
analogues.
data, and X-ray parameters. CCDC 2051349. For ESI and crystallographic data in
CIF or other electronic format see DOI: 10.1039/d1cp02386a
This journal is © the Owner Societies 2021
Phys. Chem. Chem. Phys.

View Article Online
Paper
Coumarin, related phenylpropanoids, and their structural ana-
PCCP
X-Ray intensity data were obtained on a Bruker CCD-based
logues have long attracted interest for their basic properties and and PHOTON II CPAD-based diffractometer with dual Cu/Mo
industrial applications, in particular, due to their photophysical ImuS microfocus optics (Cu Ka radiation, l = 1.54178 Å, Mo Ka
properties. Structurally modified synthetic coumarins, such as radiation, l =0.71073 Å). Crystals were mounted on a cryoloop
coumarin 314 and coumarin 504, are used as common gain media using Paratone oil and placed under a steam of nitrogen at
in laser dyes due to their efficient light emission.8 The 2H-1- 100 K (Oxford Cryosystems). The detector was placed at a
benzopyran-2-one unit itself, also referred to as unsubstituted distance of 5.00 cm from the crystal. The data were corrected
coumarin, has been incorporated into light harvesting, energy for absorption with the SADABS program. The structures were
¨
transfer, and biologically active FRET (Forster resonance energy refined using the Bruker SHELXTL Software Package (Version
transfer) dyes.8 Accordingly, unsubstituted coumarin has been the 6.1)9 and were solved using direct methods until the final
subject of both fundamental photophysical studies as well as anisotropic full-matrix, least squares refinement of F2
theoretical investigations (Fig. 1).
converged.
Cyclic voltammograms were obtained using a CHI 820D
In this report, we present an investigation into the photo-
physical properties of a boron adduct of salicylaldoxime, potentiostat with a standard three-electrode configuration.
difluoroboronsalicylaldoxime (DFBS) (Fig. 1). This species can Platinum wire was used as the counter electrode, a 2 mm in
be considered as a structural analogue of the unsubstituted diameter gold disk was used as the working electrode, and
coumarin. We examined their absorptive and fluorescent prop- silver wire was used as the reference electrode. The nonaqueous
erties. DFBS and unsubstituted coumarin show remarkably Ag/Ag+ reference electrode was prepared by immersing silver
similar UV-vis absorption spectra but differ strikingly in their wire in a dried acetonitrile solution of 0.01 M AgNO3 and 0.1 M
emissive properties: DFBS is emissive but unsubstituted cou- tetrabutylammonium hexafluorophosphate (TBAPF6). TBAPF6
marin is not. A close inspection of their crystal structures, DFT was used as the supporting electrolyte for all measurements.
(density functional theory) and TDDFT (time dependent density Acetonitrile was used as the solvent for all measurements. All
functional theory) calculations, and literature examples of potentials were referenced to the ferrocene/ferrocenium couple.
modified coumarins with increased emission provides insight The concentration of analyte was 2.0 mM, and the concen-
into how the differences in backbone structure alter the tration of supporting electrolyte was 0.2 M.
observed photophysical properties.
Synthesis
Salicylaldoxime. The synthesis of this compound was based on
a previously published procedure.10 Salicylaldehyde (5.00 mL,
47.8 mmol) is combined with sodium hydroxide (1.89 g, 47.2 mmol)
and hydroxylamine hydrochloride (8.02 g, 115 mmol) in solvent
mixture of 1 : 1 deionized water and ethanol. The solution was
Experimental
Materials
General. All reagent grade chemicals and solvents were
received from commercial retailers and used without further
purification. Solvents for synthesis were dried under 4 Å
stirred overnight at 70 1C. Ethanol was then evaporated leaving
yellow crude 2-hydroxybenzaldehyde oxime to crystalize out of
molecular sieves but otherwise used as is. HPLC-grade dichlor-
solution. 2-Hydroxybenzaldehyde oxime can be extracted with
omethane (DCM) was used for all fluorescence measurements.
dichloromethane, which is then neutralized with a dilute base
Freshly prepared fluorescence samples were degassed with
solution and washed with brine. Salicylaldoxime can be further
argon gas for 1 minute before measurements. Solvents dried
dried with magnesium sulfate and crystalized out of dichloro-
under 3 Å molecular sieves were used for absorption measure-
ments, and degassed and dried acetonitrile were used for all
electrochemistry measurements.
methane in 69% yield (4.52 g) as white crystals.
Difluoroboronsalicylaldoxime (DFBS). DFBS was synthesized
using a modified procedure first reported by Kliegel and
Nanninga.11 Dry salicylaldoxime (0.517 g, 3.77 mmol) was
Instruments
added to a dried solvent mixture of 4 : 1 chlorobenzene and
NMR spectra were recorded on 300 MHz and 400 MHz spectro- 1,4-dioxane. Molecular sieves can be added to solution. Boron
meters and chemical shifts were given in ppm relative to trifluoride etherate (0.522 mL, 4.23 mmol) was added slowly to
residual solvent resonances (1H NMR and 13C NMR spectra). the oxime solution. Solution was refluxed for 30 minutes and
High-resolution mass spectrometry experiments were per- filtered while hot to remove any salt formed as byproduct. DFBS
formed on a Bruker MicroTOF-III and MicroTOF-qIII instru- crystalizes in solvent mixture as clear crystals in 74% yield that
1
ments. Infrared spectra were collected on a Thermo Scientific dries to produce white to off white solids. DFBS: H NMR (d1-
Nicolet iS5 that was equipped with an iD5 ATR. UV-visible CDCl3, 300 MHz): d 7.03 (t, 1H, J = 7.61 Hz), 7.11 (d, 1H, J = 8.49
spectra were recorded on a Shimadzu UV-2600 UV-visible Hz), 7.37 (m/Jab, 1H), 7.56 (m, 1H), 8.29 (s, 1H), 8.68 (s, 1H). 19
spectrometer with slit width of 1 nm for all measurements. NMR (d1-CDCl3, 300 MHz): d 142.93.
Fluorescence emission spectra were measured on a Horiba
F
Computational details
FluoroMax-4 fluorimeter using quinine sulfate in 0.1 M sulfuric
acid as the reference dye for all quantum yield calculations. All DFT calculations were conducted using the 2019 R1 P1
A slit width of 2 nm was used for all emission measurements. version of GAMESS software running under a Windows OS.12
Phys. Chem. Chem. Phys.
This journal is © the Owner Societies 2021

View Article Online
PCCP
Paper
The molecular geometry coordinate inputs for DFBS and
unsubstituted coumarin were obtained from single crystal X-
ray crystallography data.13,14 Gas phase, ground state geome-
tries for both compounds were optimized using the RHF DFT
method, B3LYP exchange functional, and 6-31G(d,p) basis set.
Frequency calculations were carried out to ensure optimized
geometries represented local minima. Ground state energies
were then obtained using the optimized geometries. For the
initial TDDFT calculations, ground state geometries and the
same calculation parameters were used, and the first 25 excited
states were calculated. The same version of the GAMESS soft-
ware was used. Geometry optimization and energy calculations
were repeated for solution phase chemistry in 1,4-dioxane and
dichloromethane using PCM as the solvent model. The same
method, exchange functional, and basis set used for gas phase
calculations were employed for solution phase calculations.
Geometry optimization in solution was carried out with explicit
solvent molecules: either one DCM or one dioxane molecule
was placed next to the hydroxyl group for DFBS. Optimization
was then allowed to run until a local minimum was reached.
The same method was applied to coumarin with a solvent
molecule initially positioned next to the carbonyl group. Fre-
quency, single point energy, and excited state calculations did
not employ explicit solvent molecules.
Scheme 1 Synthesis of salicylaldoxime and DFBS.
Fig. 2 Crystal Structure of DFBS (left) and coumarin (right)13 with 35%
thermal ellipsoids. Hydrogen atoms on carbon positions have been
omitted for clarity.
Further TDDFT calculations for DFBS were performed using
Gaussian version 16 B.0115 at the Bowdoin High Performance
Computing Grid, utilizing Linux OS clusters. Gas phase and DCM
solution phase calculations were performed using three different
exchange functionals: B3LYP, CAM-B3LYP, and WB97XD. The same
basis set, 6-31G(d,p), was used to match pervious calculations. The
crystal structure of DFBS was used as the starting point geometry
and then optimized for absorption calculations. Excited state
geometry optimizations were performed in gas phase and in DCM
using the PCM solvation model for all emission calculations. All
calculations yielded positive frequencies. All orbitals were visualized
with MacMolPlt.16
deposited to the CCDC, CSD-1542948, for structural compar-
ison to DFBS.13,14 The structure is shown next to that of DFBS in
Fig. 2. While the gross structures and connectivity are similar,
there are several important differences between the two mole-
cules. Table 1 lists several bond length differences between the
two compounds. In terms of similarities, both compounds have
identical benzene rings on the unmodified end. With regard to
differences, coumarin exhibits an exocyclic carbonyl (as con-
firmed by the double bond length of 1.2160(7) Å), while DFBS
has an exocyclic hydroxide as part of an aldoxime functional
group (with an N–O bond length of 1.3763(17) Å). The two
compounds also differ in the bond lengths seen in the
Results and discussion
Table 1 Comparison of measured and calculated bond lengths for DFBS
and unsubstituted coumarin. Errors shown in measured values. Calculated
values rounded to the nearest 0.001 angstrom
We synthesized the coumarin analogue DFBS by the reaction of
BF3 with salicylaldoxime in a solvent mixture of chlorobenzene
and dioxane (Scheme 1). This compound was previously synthe-
sized by Kleigel and Nanninga following prior work by Umland
and Schleyerbach with the diphenylboron variant.11,17 We
completed the characterization of DFBS using NMR spectro-
scopy and mass spectrometry, and were able to elucidate its
crystal structure, which is shown in Fig. 2. Overall, DFBS has
the same connectivity as unsubstituted coumarin and is com-
prised of two six membered rings with an exocyclic oxygen
atom. Previously, the diphenylboron variant was elucidated by
X-ray crystallography, and the overall structural parameters are
similar to those seen in DFBS.18
DFBS
measured
Coumarin
measured
DFBS
calculated
Coumarin
calculated
Bond
N–O/C2QO2
C–O/C5–C4
O–B/C4–C3
B–N/C3–C2
N–C/C2–O1
1.3763(17)
1.3504(18)
1.433(2)
1.603(2)
1.2838(19)
1.437(2)
1.2160(7)
1.4404(5)
1.3521(5)
1.4536(5)
1.3708(8)
1.3721(8)
1.386
1.326
1.468
1.629
1.294
1.427
1.208
1.442
1.352
1.461
1.398
1.365
The structure of unsubstituted coumarin has been deter-
mined several times by X-ray crystallography in several crystal
forms. For coumarin, we selected a recent crystal structure C–C/O1–C1
This journal is © the Owner Societies 2021
Phys. Chem. Chem. Phys.

View Article Online
Paper
PCCP
heteroatom-containing ring of the compound. The largest
difference between the two compounds is due to the C–N–B
connection in DFBS. From our crystal structure we can see the
C–N bond retains its double bond character (1.2838(19) Å, on
the short side of the double bond) and the N–B bond is a single
bond (1.603(2) Å). We carried out gas phase DFT calculations to
predict the geometries of coumarin and DFBS, and the calcu-
lated bond lengths are listed in Table 1 alongside the observed
lengths. The calculations are in good agreement overall with
both structures.
One notable difference between the two crystal structures
can be seen in planarity; the BF2 unit of DFBS lies outside of the
plane defined by the C6 ring by B211 or B0.3 angstrom. This
nonplanar conformation is also seen in the gas phase DFT
geometry optimization for DFBS. DFT calculations were carried
out on a planar structure of DFBS to determine what causes the
lack of planarity seen in the crystal structure of DFBS. We
calculated the difference in energies between planar and non-
planar conformations of DFBS in which, for the planar con-
formation, the boron atom was forced to be coplanar with the
heteroatom ring. Without undergoing geometry optimization, a
single point energy was obtained on a planar DFBS. Several
iterations of this calculation were performed with different
positioning of the fluorides and slight rotations of the bonds,
but it was found that the planar conformation was only higher
in energy compared to the non-planar conformation by a small
amount: between 3 and 16 kJ molÀ1 depending on iteration.
In the 19F NMR spectra, we did not observe any temperature
dependent dynamic behaviour in the fluorine resonance
(Fig. S5, ESI†). Therefore, we estimate the energy barrier of
inversion is on the lower end of the calculated range, and we
conclude that the lack of planarity might be due to crystal
packing forces and the relative flexibility of the BF2 unit, and
that the compound is freely inverting in solution. The single
observed fluorine resonance is in agreement with the small
energy barrier to inversion (Fig. S3, ESI†). The BF2 unit, in this
case, does not provide rigidity, which often has been invoked as a
factor in establishing fluorescence in BODIPY-like fluorophores.19–23
In contrast, coumarin is as rigid, but not emissive.
Fig. 3 UV-vis absorption spectra of DFBS, salicylaldoxime, and coumarin
in DCM.
were avoided for all tests due to degradation of DFBS after
exposure to water. This effect was observed from immediate
shifts in the DFBS absorption spectra to match that of salicy-
laldoxime when a few drops of methanol was added to an
aprotic solution, however; DFBS shows good stability in aprotic
solvents such as DCM (Fig. S15, ESI†). A comparison of the
absorption and emission spectra for DFBS can be seen in Fig. 4.
Excitation at either 274 nm or 339 nm resulted in the same
emission spectra (Fig. S8, ESI†). A Stokes shift of 101 nm is
observed for the compound along with a calculated quantum
yield of 7.7%.
Both DFT and TDDFT methods were used to calculate the
structures and energies of the frontier orbitals for both cou-
marin and DFBS. The calculated structures and energies of the
ground state HOMOÀ1, HOMO, LUMO, and LUMO+1 are
shown in Fig. 5. A comparison of calculated excited state
transitions and experimental absorption spectra can be seen
in Fig. 6. More excited state transitions are listed on Tables S2–
S9 in the ESI.†
The two compounds have similar overall electronic struc-
tures with few differences in the frontier orbitals between
coumarin and DFBS. With regard to the energy levels of the
orbitals, both compounds have similar energies for their
UV-vis absorption spectra were also obtained for DFBS and
coumarin, along with the free base salicylaldoxime. Our mea-
sured absorption spectra of coumarin in acetonitrile and
methanol match previously reported absorption spectra of
coumarin in polar protic and polar aprotic solvents.24,25
A
comparison of the three absorption spectra is shown in
Fig. 3. Similar spectroscopic features are seen for all three
compounds but the absorption of both DFBS and coumarin
have a noticeable red shift of their higher energy absorption.
The lower energy absorption of DFBS is also further red-shifted
compared to the coumarin by almost 40 nm. The emission
produced by DFBS comes from the electronic state corres-
ponding to this shifted peak.
Unsubstituted coumarin has been reported to exhibit little
emission,26 but the DFBS compound is appreciably emissive.
Emission spectra was obtained for DFBS in both polar and
nonpolar solvents. Alcohols and aqueous solutions of DFBS
Fig. 4 A side by side comparison of the lower energy absorption (a.u.)
shown in solid purple, and normalized emission, solid blue, for DFBS in
DCM. 274 nm excitation wavelength was used for the emission
measurement.
Phys. Chem. Chem. Phys.
This journal is © the Owner Societies 2021

View Article Online
PCCP
Paper
The two compounds have similar electronic structures for
their HOMO orbitals, with more extended p conjugation seen
for coumarin than DFBS across the heteroatom ring, as
expected. The LUMO of the two compounds also show more
difference with more p contribution for coumarin coming from
its heterocyclic ring. Our calculated data for coumarin match
that of Kumar et al.27 The calculated energy differences of the
ground state orbitals correlate with the red shift that is
observed in the absorbance spectra of DFBS compared to
unsubstituted coumarin.
Gas phase TDDFT calculated transition energies of DFBS
and coumarin are slightly blue-shifted compared to the
obtained experimental spectra, but this phenomenon has been
Fig. 5 Gas phase DFT calculated ground state energy differences
between frontier orbitals of DFBS and coumarin. HOMOÀ1, HOMO,
LUMO, and LUMO+1 obtained via DFT calculations shown in ascending previously noted for the B3LYP functional.28–33 The transition
order.
energies and intensities correlate relatively well to experimental
spectra: two intense transitions are seen in the range of interest
for both DFBS and coumarin, and all four transitions are
similarly blue-shifted. Additional solution phase TDDFT calcu-
lations in DCM and dioxane using the PCM solvation model
(Fig. 6 and Tables S4–S7, ESI†) were performed on DFBS and
unsubstituted coumarin. Fig. S16 (ESI†) shows the geometry
optimization results for DFBS using the PCM solvent model
and an explicit solvent molecule. Hydrogen bonding between
DFBS and dioxane is suggested in the results, and the opti-
mized configuration of DFBS in dioxane positions a solvent
molecule next to the DFBS hydroxyl group. For unsubstituted
coumarin, no hydrogen bond specific arrangement of solvent
molecules was shown, as expected. For both compounds of
interest, the calculated transitions in DCM largely corrected the
gas phase blue-shift and show better agreement in terms of
energy; this is seen especially with the higher energy absorp-
tion. For DFBS, the calculated results in DCM have good
agreement with experimental results in both energy and oscil-
lator strength values; but the calculated oscillator strengths in
DCM for unsubstituted coumarin are in less agreement with
the lowest energy absorption calculated to have higher oscilla-
tor strength than the higher energy absorption. This inversion
of oscillator strength is not seen in the gas phase calculations
for unsubstituted coumarin.
Further TDDFT calculations using Gaussian were carried out
Fig. 6 Top: Absorbance spectra of DFBS in DCM and select calculated
for DFBS using the CAM-B3LYP and WB97XD exchange func-
tionals. CAM-B3LYP and WB97XD are both popular range
separated functionals that have shown better accuracy in
calculating absorption and emission properties.34,35 WB97XD
has also shown superior accuracy in calculations involving
TDDFT energy transitions in DCM using the PCM solvent model. Bottom:
Absorbance spectra of unsubstituted coumarin in DCM and select calcu-
lated TDDFT energy transitions in DCM.
HOMOs, but we observed a relative stabilization of the LUMO BODIPY.36 However, contrary to expectations, B3LYP gave
in DFBS but not in the LUMO+1. Cyclic voltammograms were results that were in better alignment with experimental results
obtained for both coumarin and DFBS in an effort to further for DFBS. As seen in Tables S8 and S9 (ESI†), CAM-B3LYP and
support this calculated result (Fig. S12, ESI†) but the com- WB97XD are in good agreement with each other for both
pounds are electrochemically reactive and the voltammograms absorption and emission calculations but overestimated the
are not reversible. There is also a small stabilization in the transition energies for DFBS. This overestimation is most
HOMOÀ1 and a small destabilization in the LUMO+1 in DFBS pronounced in the emission calculations. Fig. 7 overlays the
versus coumarin, resulting in larger HOMOÀ1/HOMO and obtained emission spectra of DFBS in DCM with the solvent
LUMO/LUMO+1 energy gaps. Conversely, the HOMO–LUMO corrected calculated result using the B3LYP functional. The
gap in DFBS is smaller than in coumarin, calculated to be good agreement of B3LYP with experimental results, along with
4.26 eV versus 4.61 eV.
the lack of noticeable molecular orbital structural changes
This journal is © the Owner Societies 2021
Phys. Chem. Chem. Phys.

View Article Online
Paper
PCCP
different solvent environments. This lack of reordering for
DFBS may reduce non-radiative relaxation pathways for DFBS
but does not fully explain the difference in observed photo-
physical behaviour between DFBS and coumarin.
The question remains as to why a boron analogue of
unsubstituted coumarin would exhibit emission whereas the
unsubstituted core of coumarin is non-emissive. The coumarin
structure has been looked at intensively because peripheral
modifications to its core structure induce fluorescence.44–50
Since the 1950s, electron donating groups such as hydroxyl
and amino substituents have been added onto the coumarin
core. Such single functional group modifications potentiate
emissivity but only in the ionic form of the molecule.44,51 This
likely indicates only the ionic forms of the modified coumarins
increase D–A character enough to turn on emission. Coumarins
modified with two peripheral substituents also turn on emis-
Fig. 7 Overlay of emission spectra of DFBS in DCM, 339 nm excitation,
with calculated emission transition energy in DCM using B3LYP. Calculated
emission was solvent corrected for DCM.
across all three functionals, suggested our continued use of this sion as long as the D–A character of the frontier orbitals is
functional for our comparisons of DFBS and unsubstituted increased with the modifications. Therefore, we would like to
coumarin.
propose a similar cause, albeit within the ring, for DFBS: an
For unsubstituted coumarin, prior studies have assigned the electropositive boron in the ring along with the electron-
lowest energy band as a (p,p*) transition based on vibration and withdrawing fluorine atoms increase D–A character of the DFBS
electronic states calculated using Fariser-Parr-Pople Hartree– excited states resulting in a larger quantum yield of emission.
Fock method.24 Follow up calculations using CNDO and INCO Some support for this hypothesis can be seen in the consis-
methods proposed a (n,p*) transition instead for the lowest tently higher calculated dipole moment for DFBS compared to
singly excited state.37 Relaxation from this (n,p*) transition was coumarin across all solvent environments (see Fig. S14, ESI†).
attributed as one of the main non-radiative pathways for
coumarin.38 More recent work by Kumar et al. using DFT and
TDDFT methods correlates with prior H-F calculations that the
low energy transitions of coumarin are p–p* type.27 While there
has been continuous debate on the photophysical properties of
unsubstituted coumarin, our data support the idea of close
lying (n,p*) and (p,p*) states reported previously.38–43 Relaxa-
tion from the (n,p*) state provides one possible non-radiative
pathway for coumarin but experimental and calculated results
support an initially excited (p,p*) state. The molar extinction
coefficients for coumarin’s lowest two excitations are quite
high, (B103 MÀ1 cmÀ1 to 104 MÀ1 cmÀ1), and the calculated
oscillator strengths in all three environments are also high
enough to indicate (p,p*) character transitions for both. Further
support comes from the structures of the modelled molecular
orbitals: both the HOMO - LUMO and HOMOÀ1 - LUMO
transitions indicate transitions between p orbitals. The n-
HOMO - LUMO transition (Fig. S13 and Tables S4, S6, ESI†),
with a very low oscillator frequency, lies close in energy to the
HOMOÀ1 - LUMO transition, and calculations indicate a
reordering of these two transitions depending on solvent
environment. The lowest energy transition remains (p,p*) in
character.
Conclusions
We have examined the structure and photophysics of the boron
analogue of coumarin, DFBS. The modification of the hetero-
cyclic ring greatly impacts the absorption and emission proper-
ties while retaining similarities to the overall orbital structure.
We believe the incorporated BF2 moiety alters the electronic
structure in two main ways that turn on emission: lack of
reordering of the non-radiative (n,p*) excited state and
increased D–A character.
We are continuing our work on boron-based chromophores
and fluorophores as part of our investigations into how boron
modification alters their fundamental properties. Immediate
future studies to flesh out the photophysics of DFBS include
performing fluorescence lifetime studies. This kinetic data
along with calculations to map potential relaxation barriers
for DFBS will give additional information on its relaxation
pathway and hopefully elaborate more in its fluorescent proper-
ties. Additional spectral grade or HPLC grade solvents, such as
spectral grade dioxane, may also be acquired for more concrete
absorption and emission data. Other studies will aim to look at
how additions of exocyclic functional groups to the DFBS
structure will affect its photophysics.
Calculated results for DFBS also show the two lowest energy
transitions result from (p,p*) transitions with corresponding
high molar extinction coefficients and relatively large oscillator
strengths. Like coumarin, the lowest energy transition for DFBS
is primarily HOMO - LUMO in character and the second
lowest energy transition is HOMOÀ1 - LUMO for all calcula-
tions. The main difference seen is that the n-HOMO - LUMO
Conflicts of interest
transition for DFBS does not seem to reorder in energy in There are no conflicts to declare.
Phys. Chem. Chem. Phys.
This journal is © the Owner Societies 2021

View Article Online
PCCP
Paper
19 E. Dai and W. Pang, et al., Chem. – Asian J., 2015, 10,
Acknowledgements
1327–1334.
K. T. would like to acknowledge Bowdoin College and Research
Corporation for Science Advancement (Cottrell Scholar Award)
for support of this work.
ˇ
´
20 S. Budzak and D. Jacquemin, Phys. Chem. Chem. Phys., 2018,
20, 25031.
21 C. Cheng and N. Gao, et al., Org. Lett., 2015, 17, 278–281.
22 J. F. Araneda, W. E. Piers and B. Heyne, et al., Angew. Chem.,
Int. Ed., 2011, 50, 12214–12217.
23 Y. Gawale and N. Sekar, J. Photochem. Photobiol., B, 2018,
178, 472–480.
24 P.-S. Song and W. H. Gordon, J. Phys. Chem., 1970, 74,
4234–4240.
25 J. S. Seixas de Melo and R. S. Becker, et al., J. Phys. Chem.,
1994, 98, 6054–6058.
26 B. N. Mattoo, Trans. Faraday Soc., 1956, 52, 1184–1194.
27 A. Kumar, R. Baccoli and A. Fais, et al., Appl. Sci., 2020,
10, 144.
28 K. R. Brorsen, Y. Yang and S. Hammes-Schiffer, J. Phys.
Chem. Lett., 2017, 8, 3488–3493.
Notes and references
1 A. Loudet and K. Burgess, Chem. Rev., 2007, 107, 4891–4932.
2 S. Boodts and E. Fron, et al., Coord. Chem. Rev., 2018, 371, 1–10.
3 G. Wesela-Bauman and M. Urban, et al., Org. Biomol. Chem.,
2015, 13, 3268.
4 K. Tram and Y. Hongbin, et al., Dyes Pigm., 2009, 82, 392–395.
5 N. Adarsh and M. Shanmugasundaram, et al., Chem. – Eur.
J., 2012, 18, 12655–12662.
6 I.-S. Tamgho and A. Hasheminasab, J. Am. Chem. Soc., 2014,
136, 5623–5626.
˜
7 J. Banuelos and F. L. Arbeloa, et al., Phys. Chem. Chem. Phys.,
29 P. Sharma, D. G. Truhlar and L. Gagliardi, J. Chem. Phys.,
2018, 148, 124305.
30 J. Mato and M. S. Gordon, Phys. Chem. Chem. Phys., 2018, 20,
2615–2626.
2011, 13, 3437–3445.
8 X.-Y. Sun and T. Liu, et al., RSC Adv., 2020, 10, 10826.
9 G. M. Sheldrick, Acta Crystallogr., Sect. A: Found. Crystallogr.,
2008, 64, 112–122.
31 J. L. Bao, P. Verma and D. G. Truhlar, Phys. Chem. Chem.
Phys., 2018, 20, 23072–23078.
10 G. Stokker, J. Org. Chem., 1983, 48, 2613–2615.
¨
11 W. Kliegel and D. Nanninga, Monatshefte fur Chemie, 1983,
32 S. S. Leang, F. Zahariev and M. S. Gordon, J. Chem. Phys.,
2012, 136, 104101.
33 L. Goerigk and S. Grimme, Phys. Chem. Chem. Phys., 2011,
13, 6670–6688.
34 C. Suellen and R. G. Freitas, et al., J. Chem. Theory Comput.,
2019, 15, 4581–4590.
35 C. Halsey-Moore and P. Jena, et al., Comput. Theor. Chem.,
2019, 1162, 112506.
114, 465–484.
12 G. M. J. Barca and C. Bertoni, et al., J. Chem. Phys., 2020,
152, 154102.
13 A. G. Shtukenberg and Q. Zhu, et al., Chem. Sci., 2017, 8,
4926–4940.
14 C. R. Groom and I. J. Bruno, et al., Acta Crystallogr., 2016,
B72, 171–179.
15 Gaussian 16, Revision B.01, M. J. Frisch, G. W. Trucks,
H. B. Schlegel, G. E. Scuseria, M. A. Robb,
J. R. Cheeseman, G. Scalmani, V. Barone, G. A. Petersson,
H. Nakatsuji, X. Li, M. Caricato, A. V. Marenich, J. Bloino,
B. G. Janesko, R. Gomperts, B. Mennucci, H. P. Hratchian,
J. V. Ortiz, A. F. Izmaylov, J. L. Sonnenberg, D. Williams-
Young, F. Ding, F. Lipparini, F. Egidi, J. Goings, B. Peng,
A. Petrone, T. Henderson, D. Ranasinghe, V. G. Zakrzewski,
J. Gao, N. Rega, G. Zheng, W. Liang, M. Hada, M. Ehara,
K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima,
Y. Honda, O. Kitao, H. Nakai, T. Vreven, K. Throssell,
J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro,
M. J. Bearpark, J. J. Heyd, E. N. Brothers, K. N. Kudin,
V. N. Staroverov, T. A. Keith, R. Kobayashi, J. Normand,
K. Raghavachari, A. P. Rendell, J. C. Burant, S. S. Iyengar,
J. Tomasi, M. Cossi, J. M. Millam, M. Klene, C. Adamo,
R. Cammi, J. W. Ochterski, R. L. Martin, K. Morokuma,
O. Farkas, J. B. Foresman and D. J. Fox, Gaussian, Inc.,
Wallingford CT, 2016.
36 V. E. Matulis and E. G. Ragoyja, et al., Int. J. Quantum Chem.,
2020, 120, e26159.
37 P. K. McCarthy and G. J. Blanchard, J. Phys. Chem., 1993, 97,
12205–12209.
38 B. H. Jhun and K. Ohkubo, et al., J. Mater. Chem. C, 2016,
4, 4556.
39 T.-J. Hsieh and C.-C. Su, et al., J. Mol. Struct., 2005, 741,
193–199.
40 S. Sebastian and S. Sylvestre, et al., Spectrochim. Acta, Part A,
2013, 101, 370–381.
41 S. Anup Thomas, B. Shiddlingeshwar and S. B. Patil, Lumi-
nescence, 2017, 32, 1328–1338.
42 R. J. Cave and E. W. Castner, Jr., J. Phys. Chem. A, 2002, 106,
12117–12123.
¨
43 C. M. Krauter and J. Mohring, et al., Phys. Chem. Chem.
Phys., 2013, 15, 17846.
44 R. M. Christie and C.-H. Lui, Dyes Pigm., 1999, 42
85–93.
45 V. F. Traven and L. I. Vorobjeva., et al., Can. J. Chem., 1997,
75, 365–376.
16 B. M. Bode and M. S. Gordon, J. Mol. Graph. Modeling, 1999,
16, 133–138.
46 K. Azuma and S. Suzuki, et al., Photochem. Photobiol. Sci.,
2003, 2, 443–449.
17 F. Umland and C. Schleyerbach, Angew. Chem., 1965, 77,
426–427.
47 M. Adamczyk and M. Cornwell, et al., Bioorg. Med. Chem.
Lett., 1997, 7, 1985–1988.
18 W. Kliegel, S. J. Rettig and J. Trotter, Can. J. Chem., 1984,
62, 1363.
This journal is © the Owner Societies 2021
Phys. Chem. Chem. Phys.

View Article Online
Paper
PCCP
48 N. Nizomov and A. U. Kholov, et al., J. Appl.Spectrosc., 2007, 50 R. H. Abu-Eittah and B. A. H. El-Tawil, Can. J. Chem., 1985,
74, 626–634. 63, 1173.
49 H. Mizoguchi and K. Kubo, et al., Berichte der Bunsenge- 51 S. Rangaswami and T. R. Seshadri, Proc. Ind. Acad. Sci.,
¨
sellschaft fur physikalische Chemie, 1997, 12, 1914–1920.
1940, 12, 375–380.
Phys. Chem. Chem. Phys.
This journal is © the Owner Societies 2021