The computational and experimental investigations of photophysical and spectroscopic properties of BF2 dipyrromethene complexes

Article abstract of DOI:10.1016/j.saa.2013.08.042

The electronic excited states of BF₂ dipyrromethene (2BrDPM, DPMI, DPMII, PM567 and 4PhDPM) complexes were investigated using the extended multi-configuration quasi-degenerate at the second order of perturbation theory (XMCQDPT2) and the second-order approximate coupled-cluster (CC2) methods. The excitation energies calculated by CC2 are significantly overestimated by 0.42-0.59 eV because of the substantial contributions of double excitation levels to excited states (>10%). However, the calculated XMCQDPT2 excitation energies agree well with experimental ones within the accuracy 0.11-0.20 eV. The very low lasing efficiency (7.8-8.4%) of 4PhDPM compound was explained by the T₁ → T₄ and T₁ → T₅ reabsorptions at XMCQDPT2 level of theory. The molecular photonics of pyrromethenes are studied using a combination of the first-principle and semi-empirical calculations. The main mechanism for the deactivation of the energy of the first singlet excited electronic state is the radiative electronic transition for DPMI, DPMII, PM567 and 4PhDPM compounds. Also, the main mechanism for the quenching of fluorescence in considered complexes (except DPMII compound) is the internal conversion. The processes of the internal conversion and intersystem crossing compete with each other in DPMII compound. The measured and calculated fluorescence quantum yields agree well for all considered molecules.

Full text of DOI:10.1016/j.saa.2013.08.042

Accepted Manuscript
The computational and experimental investigations of photophysical and spec‐
troscopic properties of BF dipyrromethene complexes
2
R.R. Valiev, A.N. Sinelnikov, Y.V. Aksenova, R.T. Kuznetsova, M.B. Berezin,
A.S. Semeikin, V.N. Cherepanov
PII:
S1386-1425(13)00916-5
http://dx.doi.org/10.1016/j.saa.2013.08.042
SAA 10901
DOI:
Reference:
To appear in:
Spectrochimica Acta Part A: Molecular and Biomo‐
lecular Spectroscopy
Received Date:
Revised Date:
1 June 2013
5 August 2013
Please cite this article as: R.R. Valiev, A.N. Sinelnikov, Y.V. Aksenova, R.T. Kuznetsova, M.B. Berezin, A.S.
Semeikin, V.N. Cherepanov, The computational and experimental investigations of photophysical and spectroscopic
properties of BF dipyrromethene complexes, Spectrochimica Acta Part A: Molecular and Biomolecular
2
Spectroscopy (2013), doi: http://dx.doi.org/10.1016/j.saa.2013.08.042
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

The computational and experimental investigations of photophysical and spectroscopic
2
properties of BF dipyrromethene complexes
1
,a
b
b
b
c
R. R.Valiev , A. N. Sinelnikov , Y. V. Aksenova , R. T. Kuznetsova , M. B. Berezin , A. S.
c
b
Semeikin , V. N. Cherepanov .
Department of General and Inorganic Chemistry, National Research Tomsk Polytechnic University,
Tomsk 634050, 43a Lenin Avenue, Building 2
a
b
Tomsk State University, 36 Lenin Avenue, 634050, Tomsk, Russian Federation
c
ISCTU, prospekt F. Engelsa 7, Ivanovo 153000, Russian Federation
Abstract
The electronic excited states of BF
investigated using the extended multi-configuration quasi-degenerate at the second order of perturbation theory
XMCQDPT2) and the second-order approximate coupled-cluster (CC2) methods. The excitation energies calculated by
2
dipyrromethene (2BrDPM, DPMI, DPMII, PM567 and 4PhDPM) complexes were
(
CC2 are significantly overestimated by 0.42-0.59 eV because of the substantial contributions of double excitation levels to
excited states (>10%). However, the calculated XMCQDPT2 excitation energies agree well with experimental ones within
the accuracy 0.11-0.20 eV. The very low lasing efficiency (7.8%-8.4%) of 4PhDPM compound was explained by the
1 4 1 5
T →T and T →T reabsorptions at XMCQDPT2 level of theory. The molecular photonics of pyrromethenes are studied
using a combination of the first-principle and semi-empirical calculations. The main mechanism for the deactivation of the
energy of the first singlet excited electronic state is the radiative electronic transition for DPMI, DPMII, PM567 and
4
PhDPM compounds. Also, the main mechanism for the quenching of fluorescence in considered complexes (except
DPMII compound) is the internal conversion. The processes of the internal conversion and intersystem crossing compete
with each other in DPMII compound. The measured and calculated fluorescence quantum yields agree well for all
considered molecules.
Keywords: internal conversion, intersystem crossing, photophysics, quantum yelds
1
Corresponding author at Department of Optics and Spectroscopy, Tomsk State University, Tomsk, 634050, Russian Federation. Tel:
+
73822-529-640;fax:+73822-529-651. E-mail: valievrashid@mail.ru
1

1. Introduction
BF dipyrromethenes are the class of compounds which have wide application in laser
2
technology because they have specific spectroscopic and photophysical properties [1-6]. The high
photostability, termal and chemical stability and the fluorescence from green to red parts of spectrum
and high fluorescence quantum yield are the main characteristics of dypyrromethene compounds [7-
12]. It is obviously, they rival with rhodamine dyes, which are often used in dye lasers. Therefore, the
new derivatives of dipyrromethenes are intensively synthesized and investigated using both
experimental and theoretical methods.
Theoretical investigation of photophysical properties of molecules implies the calculation of
the energies of electronic excitation levels and the rate constants of electronic transitions in molecules.
The knowledge of the values of the rate constants of electronic transitions allows to estimate the
probabilities of deactivation processes of excited electronic energies through the different relaxation
channels and to calculate the fluorescence and phosphorescence quantum yields.
At present time, the theoretical investigations of spectroscopic properties of big molecules
(more than 38 atoms) usually use the time-dependent density functional theory (TDDFT)[13], multi-
configuration second-order perturbation theory (CASPT2)[14] and the second-order approximate
coupled-cluster (CC2)[15-18]. However, the calculated characteristics of absorption and excitation
spectra using TDDFT and CC2 methods agree not well with experimental data when the excited states
have substantial contributions from double excitations [15,19]. Besides, the calculation at TDDFT
level of theory leads to the underestimation of the energy of charge transfer (CT) transitions [20,21].
Therefore, the choice of the method for the calculation depends on the specific system which is
investigated. It should be noted that the most of the theoretical works, where the photophysical
properties are studied, are only devoted to the estimation of the values of the rate constants of the
intersystem crossing (IS) and the radiative rate constants of the phosphorescence [22-26]. However,
the rate constants of the internal conversion (IC) are also required for the fluorescence quantum yield
2

calculations. The process of internal conversion is a radiationless transition between different
electronic and vibrational states within the same spin multiplicity. Calculation of the IC rate constants
is very expensive in computational cost and is not trivial when ab initio methods are used. These
difficulties appear because the matrix elements of non-adiabatic operator (NACME) must be
calculated [27,28]. In this case the computation efforts are concerned with the ab initio calculations of
the derivatives of wave function on coordinates [27]. To avoid this problem, the use of semi-empirical
methods for calculation of NACME can be applied to reduce the cost of such calculations [29,30]. In
the recent work of Valiev et al. [19] the rate constants of IC and IS for porphyrins molecules were
calculated using both ab initio and semi-empirical methods. These calculations show that the
estimations of fluorescence and phosphorescence quantum yields obtained by this way agree well with
experimental ones. Therefore, the algorithm in Ref. [19] can be applied to other organic molecules.
The aim of the work is to investigate the spectroscopic and photophysical properties of the 4,4-
difluoro-1,3,5,7-tetramethyl-2,6-diethyl-4-bora-3a,4a-diaza-s-indacene (DPMI), 4,4-difluoro-1,3,5,7-
tetramethyl-2,6-diethyl-8-phenyl-4-bora-3a,4a-diaza-s-indacene
(DPMII),
4,4-difluoro-1,3,5,7,8-
pentamethyl-2,6-diethyl-4-bora-3a,4a-diaza-s-indacene (PM567), 4,4-difluoro-1,3,5,7-tetra phenyl-4-
bora-3a,4a-diaza-s-indacene (4PhDPM) and 4,4-difluoro-2,6-dibromo-1,3,5,7-tetramethyl-4-bora-
3
a,4a-diaza-s-indacene (2BrDPM) compounds using experimental and theoretical methods. These
molecules are of interest because they have the high fluorescence quantum yields (> 90%) under
excitation of the S →S transition (except 2BrDPM) and are widely applied for lasing [5,6,9,11,12].
Therefore, their spectroscopic and photophysical properties are very interest for scientific society.
0
1
2. Calculation and experimental details
2.1. Synthesis
In this work, the synthesis of the DPMI, DPMII, PM567 compounds was performed according
to methods which are described in Ref. [31]. The 2BrDPM compound was synthesized using the
algorithm described in Ref. [32]. The 4PhDPM compound was synthesized according the scheme
3

which is shown in fig. 1. The 3,5 diphenylpyrrol (0.44 g. and 2.02 mM) and 2-formyl-3,5-
diphenylpyrrol (0.5 g. and 2.02 mM) were dissolved in a mixture of acetic acid (10 ml.) and acetic
anhydride (2.5 ml.) and then boiled within an hour. As a result, the solution became crimson and then
turn to a precipitate. Thereafter, the precipitate was filtered and washed with methanol and then was
dried. The acetic acid solution was poured into 200 ml of water, the precipitate was filtered off, dried
and chromatographed with the use the silica gel. The yield of the product is 0.7 g (77%).
Fig. 1. The synthetic scheme of 4PhDPM compound.
2
.2. Electronic absorption and fluorescence spectra measurements
Absorption spectra and luminescent characteristics were measured on
the SM2203
spectrometer (SOLAR, Belarus). The cyclohexan was used solvent for all measurements. The
fluorescence quantum yield was measured in respect to the rhodamine 6G with an error of 10% and the
-5
-6
concentration of the investigated compounds was 10 -10 M. The long-lived (ms) emission in frozen
solutions (at 77 K) was investigated using the Cary Eclipse spectrofluorimeter with an Optistat DN
cryostat (Oxford).
2.3. Calculations
The equilibrium geometries of electronic ground state of investigated complexes were
optimized at DFT level of theory using the Becke's three-parameter functional in combination with the
Lee-Yang-Parr correlation functional (B3LYP) [33,34] and the def2-TZVP basis set [35] in the
framework of the Turbomole 6.3 package [36]. The excitation energies were calculated at extended
multi-configuration quasi-degenerate at the second order of perturbation theory (XMCQDPT2) [37] in
Firefly [38] and at the CC2 level using the resolution of the identity approximation (RI) in Turbomole
6
.3 package [36].
4

The rate constants of IC and IS were estimated using method which is described in details in
the works of Valiev et al. [19] and Plotnikov [30]. Note that the XMCQDPT2 excitation energies and
the received equilibrium geometries of the electronic ground states at DFT level of theory were used
-
1
because the Stokes shift of fluorescence for these molecules is small (<1000 cm ). The NACME and
matrix elements of spin-orbital operator were computed at the INDO level using a set of fitting
parameters, chosen with respect to the solvent effects, for considered molecules [39].
The fluorescence quantum yields (ϕ ) for all compounds were estimated as
fl
kr
ϕ =
fl
k + k + k
r
IC
S1Ti
where k is the radiative rate constant of fluorescence, k is the rate constant of IC and k is the
r
IC
S1Ti
rate constant of the IS between the first excited singlet state (S
1
i
) and the i-th excited triplet state (T ).
3
. Results and discussions
.1. Molecular structures
3
The equilibrium geometries of the electronic ground states of all considered complexes are
shown in Fig. 2. The cartesian coordinates are given as supplementary material. Note that the
equilibrium geometries of the electronic ground state agree well with X-ray data [40-43]. The
frequencies analysis has confirmed that the obtained geometries of the considered compounds are
really equilibrium.
Fig. 2. The equilibrium structures of the considered complexes in the ground electronic state.
5

3
.1. Vertical excitation energies
The calculated vertical excitation energies and oscillator strengths (f), the peak positions of
bands of experimental electronic absorption spectra and the experimental estimation of oscillator
strengths for all investigated compounds are given in Tables 1-5. Note that the experimental excitation
energies of the first triplet state for 2BrDPM, DPMII compounds were taken from the
phosphorescence spectra from Ref. [32].
3
.1.1. 2BrDPM
The complete active space self consistent field (CASSCF) calculation is required before to
compute XMCQDPT2 excitation energies. Therefore, Hartree-Fock (HF) molecular orbitals (MO)
were chosen for the calculations with the CASSCF method. The HOMO, HOMO-n (where n=1..8) and
LUMO, LUMO+1 were used in the CASSCF calculation. Note that these MO were chosen in
accordance with that the electronic transitions, formed by these MO, in the configuration interaction
(CI) expansion of the first four electronic excited singlet states at CIS level of theory for 2BrDPM
compound have the maximum weights (≥0.2).
The optimized MO at the CASSCF level of theory are given in supplementary material for
2BrDPM complex and other compounds. Contributions of the most relevant transitions of CASSCF
calculation, the XMCQDPT2 calculated and experimental electronic excitation energies, oscillator
strengths of 2BrDPM compound are given in Table 1. As it is visible from Table 1, the discrepancy
between the theoretical and experimental excitation energies is 0.16 eV for the first excited singlet
state. The second band of electronic absorption spectrum of 2BrDPM is located in the range from 410
to 340 nm (see supplementary material). The position of the peak of the second band is 3.24 eV.
However, in order to determine accurately the energies of electronic transitions which form this
absorption band the femtosecond fluorescence upconversion study is required [44]. One can consider
that the second band is formed only by the 2-4 electronic transitions. Note that the using XMCQDPT2
method leads to the good agreement of the calculated and observed energy of the first triplet state.
6

Table 1. Contributions of the most relevant configuration state functions (the weights are given in parentless) of CASSCF
calculation, the XMCQDPT2 calculated and experimental electronic excitation energies (E) and oscillator strengths (f) for
the first four electronic excited states of 2BrDPM compound.
CASSCF
XMCQDPT2
Experiment
State
single
double
E(f)
E(f)
H→L (0.61)
H-1→L (-0.21)
H→L (0.59)
S
1
S
2
S
3
S
4
H→L;H-2→L (-0.17)
2.14(0.19)
2.85(0.04)
2.90(0.02)
4.11(0.01)
2.3(0.2)
H→L;H-1→L (0.23)
H→L;H→L (-0.20)
3.24(0.03)
H-1→L (0.60)
H→L (-0.24)
H-3→L (-0.37)
H-4→L (-0.30)
H→L;H→L (-0.68)
a
T
1
2
1.70
2.27
1.56
T
a
Measured in ethanol [32].
The results of CASSCF calculation shows that the 1-th, 2-th and 4-th excited singlets have the
substantial contributions (≥0.2) from the double excitations. It is expected that the TDDFT and CIS
methods lead to an overestimation of these energies because the TDDFT and CIS methods do not
include double excitations. It should be noted also that there is a good agreement (within the order)
between the calculated and experimental oscillator strengths of S
.1.2. DPMI
The HOMO, HOMO-n (n=1..3) and LUMO, LUMO+1 were chosen for the CASSCF
0 1
→S transition.
3
calculation. The criterion of the selection of MO is the same as in the case of 2BrDPM compound.
This criterion will be also applied for other considered compounds. The results of the calculations are
given in Table 2. As seen from the Table 2, the using of the XMCQDPT2 method leads to small
underestimation of the S
1
energy (0.1 eV). Also, there are significant contribution (≥ 0.2) of double
7

1 2 2 1
excitations in S and S states. It should be noted the energetic gap between T and S energy levels is
-
1
0
.02 eV (161 cm ).
3
.1.3. DPMII
The molecular orbitals were used in the CASSCF calculations are given in supplementary material.
The experimental and theoretical results are given in Table 3. The discrepancy between the
XMCQDPT2 calculated and experimental excitation energies of S
calculation at CASSCF level of theory shows that there are substantial double excitations (≥0.2) in the
-S excited states. Therefore, the TDDFT method leads to the significant overestimation of the
excitation energy of S state by 1.1 eV (see Table 3). Also, the discrepancy between the calculated and
experimental excitation energies of T state is 0.1 eV.
1
level is 0.27 eV. Also, the
S
1
4
1
1
Table 2 Contributions of the most relevant configuration state functions (the weights are given in parentless) of CASSCF
calculation, the XMCQDPT2 calculated and experimental electronic excitation energies (E) and oscillator strengths (f) for
the first two electronic excited states of DPMI compound.
CASSCF
XMCQDPT2
Experiment
State
single
double
E(f)
E(f)
S
1
2
H→L (-0.63)
H-1→L (0.53)
H-2→L (0.33)
H-3→L;H-→L (-0.19)
H→L;H-1→L (0.19)
2.23(0.22)
2.33(0.30)
S
2.98(0.02)
T
1
2
1.52
2.25
T
3
.1.4. PM567
HOMO, HOMO-n (n=1..3) and LUMO, LUMO+m (m=1..5) have to be included in CASSCF
calculations for PM567 complex. The results of the calculations of excitation energies, oscillator
strengths at XMCQDPT2 level of theory together with experimental ones are given in Table 4. It is
seen from Table 4 that the calculated and experimental excitation energies of the S
1
state agree well
(the discrepancy is 0.22 eV). It should be noted that the using CASSCF method gives the substantial
8

contributions of double excitations (≥0.2) to all singlet excited states. Therefore, the calculations of
excitation energies using TDDFT method leads to the significant overestimation of them.
Table 3. Contributions Contributions of the most relevant configuration state functions (the weights are given in parentless)
of CASSCF calculation, the XMCQDPT2 calculated and experimental electronic excitation energies (E) and oscillator
strengths (f) for the first four electronic excited states of DPMII compound.
B3LYP/6-31G(d,p)
CASSCF
XMCQDPT2
Experiment
[
45]
State
single
double
E(f)
E
E(f)
H→L (-0.59)
H-1→L (-0.27)
H-1→L (-0.55)
H→L (0.29)
H-2→L (0.59)
2.03(0.17)
3.40
S
1
H→L;H-2→L (-0.16)
H→L;H-2→L (-0.24)
2.30(0.20)
3.26(0.02)
S
2
S
3
S
4
2.68(0.01)
2.85(0.01)
4.05(0.03)
H→L;H-2→L (0.31)
H→L;H→L (0.71)
H-5→L (0.39)
H-5→L;H-5→L (0.27)
a
T
1
2
1.71
2.29
1.70
T
a
Measured in ethanol [32].
3
.1.5. 4PhDPM
In this case, the HF HOMO, HOMO-n (n=1..5) and LUMO were chosen for CASSCF
calculations. The calculation results obtained in the framework of XMCQDPT2 level of theory
together with the observed results are given in Table 5. This table shows a good agreement of the
calculated and observed both excitation energies of S
transition (the discrepancy doesn't exceeded one order of magnitude). At that the calculated and
observed excitation energies of T state can not be compared because the phosphorescence of 4PhDPM
compound is absent. Also, there is the problem concerned with the very low lasing efficiency (η) of
PhDPM complex in polar and nonpolar solvents. In this case, η equals to 7.8%-8.4% when the
1 0 1
state (0.17 eV) and oscillator strength of S →S
1
4
-
1
fluorescence quantum yield is 0.9 and excitation energy is 2.33 eV (18800 cm ). The most likely the
-1
reason of the low value of η is the reabsorption of the laser emission at the wavenumber 16500 cm on
9

the S
energies and the oscillator strengths for S
results of the calculations are given in Table 6. It is visible from this table that T
1
→S
i
and/or T
1
→T
i
transitions. In order to confirm this conclusion the calculations of excitation
→S and T →T electronic transitions were carried out. The
→T and T →T
1
i
1
i
1
4
1
5
electronic transitions give the large contribution to the formation of T-T reabsorption at the wavelength
of laser emission because T-T energies are close to laser wavenumber and the oscillator strengths of
these transitions are large.
Table 4. Contributions of the most relevant configuration state functions (the weights are given in parentless) of CASSCF
calculation, the XMCQDPT2 calculated and experimental electronic excitation energies (E) and oscillator strengths (f) for
the first four electronic excited states of PM567 compound.
B3LYP/6-31G(d,p)
CASSCF
XMCQDPT2
Experiment
[
45]
State
single
double
E(f)
E
E(f)
H→L (0.6)
2.08(0.17)
3
.40
S
1
H→L;H-2→L (-0.19)
2.30(0.20)
3.30(0.03)
H-1→L (-0.22)
H-2→L (-0.45)
H→L;H→L (-0.44)
H→L;H-1→L (0.18)
H→L;H-3→L (0.24)
S
2
S
3
S
4
3.45(0.01)
3.55(0.01)
4.11(0.03)
H-1→L (0.53)
H-3→L (-0.29)
H-2→L (-0.28)
H→L;H→L (-0.48)
T
1
2
1.71
2.29
T
3
.2. The CC2 calculation
The calculations of excitation energies and oscillator strengths, the single and double
contributions of excitation levels are given in Table 7. Note, the D1 diagnostics is 0.07 for all
calculations. Therefore, the electronic ground states of considered molecules are not the
1
multiconfigurational character [36]. As seen from Table 7, the calculated excitation energies of S state
are overestimated by 0.42-0.59 eV for all considered compounds. The reason of the overestimation is
the substantial contributions of double excitation levels (> 10%) [19,46]. Therefore, the CC2
10

calculation explains the bad agreement of the TDDFT excitation energies with experimental ones.
Note that only one triplet state (T ) is lower than the S state according to the results of CC2
calculations for all considered complexes. Therefore, the S energy can be deactivated by the internal
conversion (S →S
1
1
1
1
0
)
Table 5. Contributions of the most relevant configuration state functions (the weights are given in parentless) of CASSCF
calculation, the XMCQDPT2 calculated and experimental electronic excitation energies (E) and oscillator strengths (f) for
the first four electronic excited states of 4PhDPM compound.
CASSCF
XMCQDPT2
Experiment
State
single
double
E(f)
E(f)
S
S
S
1
2
3
H→L (-0.64)
H-2→L (-0.62)
H-2→L (0.62)
H-3→L (-0.37)
H-5→L (0.20)
H→L;H-4→L (0.20)
H-2→L;H→L (-0.26)
H-2→L;H→L (0.30)
2.01(0.19)
2.55(0.01)
2.72(0.01)
2.18(0.20)
3.30
S
4
H→L;H→L (-0.75)
3.45(0.001)
T
1
2
1.47
2.20
T
Table 6. The excitation energies (cm-1) and oscillator strengths (f) for the S1 →Si and T1→Ti electronic transitions of
PhDPM compound calculated by XMCQDPT2 level of theory.
4
Type
E(f)
Type
E(f)
S
S
S
1
1
1
→S
→S
→S
2
3
4
4370(0.0)
T
T
T
1
1
1
→T
→T
→T
3
4
5
6950(0.09)
15000(0.6)
18500(0.1)
5700(0.012)
11600(0.001)
5700(0.04)
1 2
T →T
and the intersystem crossing (S
considered compounds.
1 1
→T ) or by the photon emission (the radiative electronic transition) for
11

3
.3. The rate constants of electronic transitions
The calculation of excitation energies of the first singlet and triplet states for considered
compounds using XMCQDPT2 method lead to the good agreement with the experimental ones (the
discrepancy is less than 0.25 eV). Therefore, the XMCQDPT2 excitation energies were used to
calculate the IC and IS rate constants and to estimate the quantum fluorescence yields (ϕ ) for
fl
considered complexes. Also, the equilibrium geometries of the ground electronic states were used for
the modeling of the geometries for the first excited states for all considered molecules. This modeling
-1
is possible because the Stokes shifts are < 800 cm . The results of the calculations of the rate constants
of the intersystem crossing (k ), the internal conversion (k ) and the radiation rate constants (k )
S T
IC
r
1
1
are given in Table 8 for the studied compounds. It should be noted that the measurements of the
fluorescence spectra and the fluorescence quantum yields were carried out at very small concentrations
-5
-6
(
10 -10 M) of the complexes. Therefore, the intermolecular mechanisms of the quenching of
fluorescence can be neglected.
The results show that the main mechanism of the deactivation of the excited electronic energy
is the radiative electron transition for DPMI, DPMII, PM567 and 4PhDPM compounds. The
probability of the radiationless electronic transitions is significantly less than the probability of the
radiative electron transitions for DPMI, DPMII, PM567 and 4PhDPM compounds whereby these
compounds have the high fluorescence quantum yields (>0.9). Note that the presence of the heavy
atom of bromine in 2BrDPM compound leads to the competition between the processes of the
intersystem crossing and the photon emission. In this case, the quenching of fluorescence is caused
mainly by the intersystem crossing process. Therefore, the fluorescence quantum yield of 2BrDPM
complex is substantially less than it for DPMI, DPMII, PM567 and 4PhDPM compounds. It should be
also noted that the main mechanism for the quenching of fluorescence in DPMI, PM567 and 4PhDPM
molecules is the internal conversion. As to DPMII compound, the processes of the internal conversion
and the intersystem crossing compete each with other for this compound.
12

Table 7. The excitation energies (in eV) and oscillator strengths (f), the single and double contributions of the excitation
levels to excited states are calculated with using CC2 level of theory.
2
BrDPM
Type
E(f)
Exp
Single %
Double %
S
1
2.72(0.6)
1.95
2.3(0.2)
1.7
88
90
89
12
10
11
T
1
2
T
2.95
DPMI
Type
E(f)
Exp
Single %
Double %
S
1
2.73(0.6)
1.95
2.33(0.3)
88
97
97
12
T
1
2
3.0
3.0
T
2.95
DPMII
Type
E(f)
Exp
Single %
Double %
S
1
2.75(0.6)
1.96
2.30(0.3)
1.61
88
91
89
12
8.0
11
T
1
2
T
2.95
PM567
Type
E(f)
Exp
Single %
Double %
S
1
2.77(0.6)
2.02
2.18(0.3)
88
91
89
12
8.0
11
T
1
2
T
3.02
4
PhDPM
Type
E(f)
Exp
2.18(0.2)
Single %
Double %
S
1
2.54(0.6)
1.98
86
91
87
14
8.0
13
T
1
2
T
2.95
The validity of the obtained results is confirmed by the very good agreement of the calculated
fluorescence quantum yields with the experimental ones.
13

-
1
Table 8. The calculated rate constants (in s ) for the photophysical processes of the 2BrDPM, DPMI, DPMII, PM567 and
PhDPM compounds are given. The obtained fluorescence quantum yields arecompared to experimental values.
4
2BrDPM
DPMI
DPMII
PM567
4PhDPM
7
7
6
4
7
7
7
kr
4.0·10
1.1·10
4.3·10
2.7·10
3.2·10
3.7·10
3.5·10
6
7
6
6
6
kIC
k
5.3·10
4.9·10
0.87
4.2·10
5.5·10
2.7·10
6
5
6
2.4·10
4.9·10
2.0·10
S T
1
1
ϕ fl
0.47
0.45
0.83
0.90
0.87
0.92
0.90
ϕ (exp)
0.90
0.90-1.00
fl
4. Conclusion
The equilibrium molecular structures of the electronic ground states of 2BrDPM, DPMII,
PM567 and 4PhDPM compounds were obtained at DFT level of theory using B3LYP functional and
TZVP basis set. The electronic absorption and fluorescence spectra of the considered complexes were
observed. The all compounds have almost the same position of the maximums of the first absorption
band and the close values of intensity. The electronic structures of the electronic excited states were
investigated using XMCQDPT2 and CC2 methods. The application of both methods have shown the
substantial contributions of double excitation levels to excited states (>10%) of all considered
compounds. Therefore, the calculation of the excitation energies of the considered compounds leads to
the overestimation of them by 0.42-0.59 eV when the CC2 level of theory is used. In contrast, the
calculated XMCQDPT2 excitation energies agree well with experimental values (the discrepancy is
0
.11-0.20 eV). Also, in this work the low efficiency of the laser emittion of 4PhDPM complex was
explained by the T →T and T →T reabsorption.
1
4
1
5
The rate constants of the internal conversion and the intersystem crossing, the radiative rate
constants were calculated. The measured and estimated fluorescence quantum yields of the considered
compounds largely agree. The main mechanism of the deactivation of the electronic energy of the S
1
14

state for DPMI, DPMII, PM567 and 4PhDPM complexes is the radiative electron transitions. Also, the
main mechanism of the quenching of fluorescence in the DPMI, PM567 and 4PhDPM complexes is
the internal conversion. The processes of internal conversion and intersystem crossing compete each
with other in DPMII compound. As to the 2BrDPM compound, the quenching of fluorescence is
caused mainly by the intersystem crossing process.
Acknowledgements
This work has been supported by the RFBR 11-02-98019r, RFBR 12-02-90008 and GC
N14.518.11.7038. The calculations were performed at the SKYF CSC (Tomsk, Tomsk State
University) and the Finnish IT Center for Science, which is thanked for the generous amount of
computer time.
Referenses
[1] A. Costela, I. Garcia-Moreno, C. Gomez, O. Garcia, R. Sastre, J. Appl. Phys. 90 (2001) 3159.
[2] F. Liang, H. Zeng, Z. Sun, Y. Yuan, Z. Yao, Z. Xu, J. Appl. Phys. 18 (2001), 1841–1845.
[3] T. Susdorf, D. del Agua, A. Tyagi, A. Penzkofer, O. Garcia, R. Sastre, A. Costela, I. Garcia-
Moreno, Appl. Phys. B. 86 (2007) 537–545.
'
'
[
4] M. Alvarez, F. Amat-Guerri, A. Costela, I. Garcia-Moreno, C. Gomez, M. Liras, R. Sastre, Appl.
Phys. B. 80 (2005) 993–1006.
'
'
[
5] M. E. Perez-Ojeda, V. Martin, A. Costela, I. Garcia-Moreno, I. A. Cordoba and E. Pe a-Cabrera,
Appl. Phys. B. 106 (2012) 911–914.
6] T. N. Kopylova, S. S. Anufrik, G. V. Mayer, T. A. Solodova, E. N. Tel’minov, K. M. Degtyarenko,
[
L. G. Samsonova, R. M. Gadirov, S. Y. Nikonov, E. N. Ponyavina, V. V. Tarkovskii, G. G. Sazonko,
Russ. Phys. J. 55 (2013) 1137–1142.
[
[
7] A. Loudet, K. Burgess, Chem. Rev. 107 (2007) 4891–4932.
8] G. Ulrich, R. Ziessel, A. Harriman, Angew. Chem. Int. Ed. 47 (2008) 1184–1201.
15

[
9] R. T. Kuznetsova, Y. V. Aksenova, E. N. Tel’minov, L. G. Samsonova, G. V. Maier, T. N.
Kopylova, S. L. Yutanova, E. V. Antina, M. B. Berezin, G. B. Guseva, Optics and Spectroscopy 112
(2012) 746–754.
[10] S. Mula, A. K. Ray, M. Banerjee, T. Chaudhuri, K. Dasgupta, S. Chattopadhyay, J. Org. Chem.
73 (2008) 2148–2154.
[
11] T. Chaudhuri, S. Mula, S. Chattopadhyay, M. Banerjee, Spectrochim. Acta A 75 (2010) 739–744.
12] F. L. Arbeloa, T. L. Arbeloa, I. L. Arbeloa, I. Garcia-Moreno, A. Costela, R. Sastre, F. Amat-
[
Guerri, Chem. Phys. 236 (1998) 331–341.
[
[
13] F. Furche, R. Ahlrichs, J. Chem. Phys. 117 (2002) 7433–7447.
14] O. Rubio-Pons, O. Loboda, B. Minaev, B. Schimmelpfennig, O. Vahtras, H. A˚gren, Mol. Phys.
101 (2003) 2103–2114.
[
[
[
[
[
(
[
[
(
[
[
[
[
[
15] C. H ttig, F. Weigend, J. Chem. Phys. 113 (2000) 5154–5161.
16] C. H ttig, Adv. Quantum Chem. 50 (2005) 37–60.
17] O. Christiansen, H. Koch, P. Jörgensen, Chem. Phys. Letters 243 (1995) 409–418.
18] F. Weigend, M. H ser, H. Patzelt, R. Ahlrichs, Chem. Phys. Letters 294 (1998) 143–152.
19] R. R. Valiev, V. N. Cherepanov, V. Y. Artyukhov, D. Sundholm, Phys.Chem. Chem. Phys. 14
2012) 1150811517.
20] T. Yanai, D. P. Tew, N. C. Handy, Chem. Phys. Letters 393 (2004) 51–57.
21] R. Valiev, E. Ermolina, Y. Kalugina, R. Kuznetsova and V. Cherepanov, Spectrochim. Acta A 87
2012) 40–45.
22] B. Minaev, Spectrochim. Acta A 60 (2002) 3213–3224.
23] O. Loboda, I. Tunell, B. Minaev, H. A˚gren, Chem. Phys. Letters 312 (2005) 299–309.
24] B. F. Minaev, A. B. Minaev, Opt. and Spectroscopy 98 (2005) 214–219.
25] B. Minaev, E. Jansson, H. A˚ gren, J. Chem. Phys. 125 (2006) 234704.
26] S. Perun, J. Tatchen, C. M. Marian, ChemPhysChem 9 (2008) 282–292.
16

[
[
[
[
[
27] C. Marian, WIREs Rev., Comp. Molec. Sci. 2 (2012) 187–203.
28] G. Hirsch, P. J. Bruna, R. J. Buenker, S. D. Peyerimhoff, Chem. Phys. Letters 45 (1980) 335–347.
29] V. G. Plotnikov, V. A. Dolgikh, V. M. Komarov, Opt. and Spectroscopy 43 (1977) 522–527.
30] V. G. Plotnikov, Intern. J. Quantum Chem., 16 (1979) 527–541.
31] E. V. Antina, M. B. Berezin, A. S. Semeikin, N. A. Dudina, S. L. Yutanova, Russ. J. General
Chem. 80 (2010) 1214–1215.
32] R. T. Kuznetsova, Y. V. Aksenova, O. Orlovskaya, T. N. Kopilova, E. N. Telmonov, G. V. Maier,
[
E. Antina, S. L. Jutanova, M. B. Berezin, G. B. Guseva, L. A. Antina, A. S. Semeikin, High Energy
Chemistry(in Russian) 46 (2012) 464–475.
[
[
[
[
[
[
33] A. D. Becke, J. Chem. Phys. 98 (1993) 5648–5652.
34] C. Lee, W. Yang, R. G. Parr, Phys. Rev. B, 37 (1988) 785–789.
35] F. Weigend, R. Ahlrichs, Phys. Chem. Chem. Phys. 7 (2005) 3297–3305.
36] R. Ahlrichs, M. B r, M. H ser, H. Horn, C. Kölmel, Chem. Phys. Letters 162 (1989) 165–169.
37] A. A. Granovsky, J. Chem. Phys. 134 (2011) 214113.
38] Firefly, Alex A. Granovsky, Firefly version 8.0.0,
www http://classic.chem.msu.su/gran/firefly/index.html.
39] V. Y. Artyukhov, T. N. Kopylova, L. G. Samsonova, N. I. Selivanov, V. G. Plotnikov, V. A.
Sazhnikov, A. A. Khlebunov, G. V. Mayer and M. V. Alfimov, Russ. Phys. J. 51 (2008) 1097–1111.
[
[
[
[
[
[
[
40] K. Tram, H. Yan, H. A. Jenkins, S. Vassiliev, D. Bruce, Dyes and Pigments 82 (2009) 392–395.
41] H.-J. Xu, Acta Crystallographica Section E. 64 (2008) 1672.
42] E. D. Stevens, M. Shah, J. H. Boyer, Acta Crystallographica Section C. 46 (1990) 1148–1150.
43] Y. Sun, Acta Crystallographica Section E. 68 (2012) 1302.
44] P. Toele, H. Zhang, C. Trieflinger, M. G. J. Daub, Chem. Phys. Letters. 368 (2003) 66–75.
45] S. Mula, A. K. Ray, M. Banerjee, T. Chaudhuri, K. Dasgupta and S. Chattopadhyay, Chem. Phys.
Letters 73 (2007) 2146–2154.
17

[46] N. O. Winter, N. K. Graf, S. Leutwyler, C. H ttig, Phys. Chem. Chem. Phys. 15 (2013) 6623–
6
630.
18

19

20

Table 1. Contributions of the most relevant configuration state functions (the weights are given in parentless) of CASSCF
calculation, the XMCQDPT2 calculated and experimental electronic excitation energies (E) and oscillator strengths (f) for
the first four electronic excited states of 2BrDPM compound.
CASSCF
XMCQDPT2
Experiment
State
single
double
E(f)
E(f)
H→L (0.61)
H-1→L (-0.21)
H→L (0.59)
S
S
S
S
1
2
3
4
H→L;H-2→L (-0.17)
2.14(0.19)
2.85(0.04)
2.90(0.02)
4.11(0.01)
2.3(0.2)
H→L;H-1→L (0.23)
H→L;H→L (-0.20)
3.24(0.03)
H-1→L (0.60)
H→L (-0.24)
H-3→L (-0.37)
H-4→L (-0.30)
H→L;H→L (-0.68)
a
T
T
1
1.70
2.27
1.56
2
a
Measured in ethanol [32].
21

Table 2 Contributions of the most relevant configuration state functions (the weights are given in parentless) of CASSCF
calculation, the XMCQDPT2 calculated and experimental electronic excitation energies (E) and oscillator strengths (f) for
the first two electronic excited states of DPMI compound.
CASSCF
XMCQDPT2
Experiment
State
single
double
E(f)
E(f)
S
S
1
H→L (-0.63)
H-1→L (0.53)
H-2→L (0.33)
H-3→L;H-→L (-0.19)
H→L;H-1→L (0.19)
2.23(0.22)
2.33(0.30)
2
2.98(0.02)
T
T
1
1.52
2.25
2
22

Table 3. Contributions Contributions of the most relevant configuration state functions (the weights are given in parentless)
of CASSCF calculation, the XMCQDPT2 calculated and experimental electronic excitation energies (E) and oscillator
strengths (f) for the first four electronic excited states of DPMII compound.
B3LYP/6-31G(d,p)
CASSCF
XMCQDPT2
Experiment
[
45]
State
single
double
E(f)
E
E(f)
H→L (-0.59)
H-1→L (-0.27)
H-1→L (-0.55)
H→L (0.29)
H-2→L (0.59)
2.03(0.17)
3.40
S
1
H→L;H-2→L (-0.16)
H→L;H-2→L (-0.24)
2.30(0.20)
S
S
S
2
3
4
2.68(0.01)
2.85(0.01)
4.05(0.03)
3.26(0.02)
H→L;H-2→L (0.31)
H→L;H→L (0.71)
H-5→L (0.39)
H-5→L;H-5→L (0.27)
a
T
T
1
1.71
2.29
1.70
2
a
Measured in ethanol [32].
23

Table 4. Contributions of the most relevant configuration state functions (the weights are given in parentless) of CASSCF
calculation, the XMCQDPT2 calculated and experimental electronic excitation energies (E) and oscillator strengths (f) for
the first four electronic excited states of PM567 compound.
B3LYP/6-31G(d,p)
CASSCF
XMCQDPT2
Experiment
[
45]
State
single
double
E(f)
E
E(f)
H→L (0.6)
H-1→L (-0.22)
H-2→L (-0.45)
2.08(0.17)
3.40
S
1
H→L;H-2→L (-0.19)
2.30(0.20)
H→L;H→L (-0.44)
H→L;H-1→L (0.18)
H→L;H-3→L (0.24)
S
S
S
2
3
4
3.45(0.01)
3.55(0.01)
4.11(0.03)
3.30(0.03)
H-1→L (0.53)
H-3→L (-0.29)
H-2→L (-0.28)
H→L;H→L (-0.48)
T
T
1
1.71
2.29
2
24

Table 5. Contributions of the most relevant configuration state functions (the weights are given in parentless) of CASSCF
calculation, the XMCQDPT2 calculated and experimental electronic excitation energies (E) and oscillator strengths (f) for
the first four electronic excited states of 4PhDPM compound.
CASSCF
XMCQDPT2
Experiment
State
single
double
E(f)
E(f)
S
S
S
1
2
3
H→L (-0.64)
H-2→L (-0.62)
H-2→L (0.62)
H-3→L (-0.37)
H-5→L (0.20)
H→L;H-4→L (0.20)
H-2→L;H→L (-0.26)
H-2→L;H→L (0.30)
2.01(0.19)
2.55(0.01)
2.72(0.01)
2.18(0.20)
3.30
S
4
H→L;H→L (-0.75)
3.45(0.001)
T
T
1
1.47
2.20
2
25

-
1
Table 6. The excitation energies (cm ) and oscillator strengths (f) for the S
1
→S
i
and T
1
→Ti electronic transitions of
4
PhDPM compound calculated by XMCQDPT2 level of theory.
Type
E(f)
Type
E(f)
S
S
S
1
1
1
→S
→S
→S
2
3
4
4370(0.0)
5700(0.012)
11600(0.001)
5700(0.04)
T
T
T
1
1
1
→T
→T
→T
3
4
5
6950(0.09)
15000(0.6)
18500(0.1)
1 2
T →T
26

Table 7. The excitation energies (in eV) and oscillator strengths (f), the single and double contributions of the excitation
levels to excited states are calculated with using CC2 level of theory.
2BrDPM
Type
E(f)
Exp
Single %
Double %
S
1
2.72(0.6)
1.95
2.3(0.2)
1.7
88
12
10
11
T
T
1
90
2
2.95
89
DPMI
Type
E(f)
Exp
Single %
Double %
S
1
2.73(0.6)
1.95
2.33(0.3)
88
12
T
T
1
97
3.0
2
2.95
97
3.0
DPMII
Type
E(f)
Exp
Single %
Double %
S
1
2.75(0.6)
1.96
2.30(0.3)
1.61
88
12
8.0
11
T
T
1
91
2
2.95
89
PM567
Type
E(f)
Exp
Single %
Double %
S
1
2.77(0.6)
2.02
2.18(0.3)
88
12
8.0
11
T
T
4
1
91
2
3.02
89
PhDPM
Type
E(f)
Exp
Single %
Double %
S
1
2.54(0.6)
1.98
2.18(0.2)
86
14
8.0
13
T
T
1
91
2
2.95
87
27

-
1
Table 8. The calculated rate constants (in s ) for the photophysical processes of the 2BrDPM, DPMI, DPMII, PM567 and
PhDPM compounds are given. The obtained fluorescence quantum yields arecompared to experimental values.
4
2BrDPM
DPMI
DPMII
PM567
4PhDPM
7
7
6
4
7
7
7
kr
4.0·10
1.1·10
4.3·10
2.7·10
3.2·10
3.7·10
3.5·10
6
7
6
6
6
kIC
k
5.3·10
4.9·10
0.87
4.2·10
5.5·10
2.7·10
6
5
6
2.4·10
4.9·10
2.0·10
S T
1
1
ϕ fl
0.47
0.45
0.83
0.90
0.87
0.92
0.90
ϕ (exp)
0.90
0.90-1.00
fl
28

Graphical abstract
29