Dicyano and pyridine derivatives of retinal: Synthesis and vibronic, electronic, and photophysical properties

Article abstract of DOI:10.1021/jp103749b

Two new derivatives of all-trans retinal, one containing a pyridine and a cyanide substituent, all-trans-15-cyano-15′-pyridylretinal (2), and the other containing two cyanide substituents, all-trans-15,15′-dicyanoretinal (3), have been synthesized from al

Full text of DOI:10.1021/jp103749b

6766
J. Phys. Chem. A 2010, 114, 6766–6775
Dicyano and Pyridine Derivatives of Retinal: Synthesis and Vibronic, Electronic, and
Photophysical Properties
A. J. Cruz,† K. Siam,‡ C. Moore,† M. Islam,† and D. P. Rillema*,†
Department of Chemistry, Wichita State UniVersity, Wichita, Kansas 67260, and Department of Chemistry,
Pittsburgh State UniVersity, Pittsburgh, Kansas 66762
ReceiVed: April 26, 2010; ReVised Manuscript ReceiVed: May 20, 2010
Two new derivatives of all-trans retinal, one containing a pyridine and a cyanide substituent, all-trans-15-
cyano-15′-pyridylretinal (2), and the other containing two cyanide substituents, all-trans-15,15′-dicyanoretinal
(3), have been synthesized from all-trans-retinal (1) by the Knoevenagel reaction. Compound 2 crystallizes
j
in the space group P1; compound 3 crystallizes in the space group P21/n. The polyene chains are bowed. The
compounds exhibit CtN vibrations in the infrared and Raman regions and undergo irreversible oxidations.
Compounds 2 and 3 display electronic absorptions with maxima located at 442 and 487 nm and emission
peaks located at 583 and 500 nm, respectively. Density Functional Theory and Time Dependent Density
Functional Theory calculations reveal that the electronic transition from the HOMO to the LUMO can be
associated with the Ag f Bu electronic transition and a peak corresponding to the HOMO f LUMO+1
transition can be associated with the “cis-peak”.
Introduction
SCHEME 1: Compounds 2 and 3 Derived from 1
ꢀ-Carotenes serve as antennas for absorbing light in spectral
regions where chlorophylls do not absorb, and thus initiate
photosynthesis.1-8 According to group theoretical treatments
based on the C2h point group, the electronic ground state is 1Ag
in character while the symmetry-allowed transition found in the
visible region responsible for pigment coloration is due to the
population of the 1Bu state.1,9-22 Low-temperature (77 K)
emission studies have shown that another 1Ag excited state lies
1
between the ground state and the Bu excited state.13-15,23-26
Triplet states have also been previously investigated.27,28
In this study, we have synthesized two cyanine derivatives
of all-trans-retinal, ꢀ-carotene analogues (Scheme 1), as anten-
nas for attachment to transition metal complexes using an acid-
catalyzed approach. Crystal structures were determined for both
compounds by using single crystal X-ray diffraction, and their
electronic, vibronic and photophysical properties have been
characterized. Density-functional theory (DFT) and Time De-
pendent Density Functional Theory (TDDFT)29 calculations have
also been employed to compare their spectral properties to those
determined experimentally.
Experimental Section
Materials. The compound all-trans-retinal (Vitamin-A al-
dehyde) was obtained from Sigma. Malononitrile was obtained
from Fluka while the compound 4-pyridylacetonitrile monohy-
drochloride was purchased from Acros Scientific. Butyronitrile
used for emission and lifetime studies was obtained from Acros
Scientific and was fractionally distilled prior to use. Dried
benzene, diethyl ether, HPLC grade methanol, chloroform,
methylene chloride, and optima grade acetonitrile for UV/visible
spectral determination were purchased from Fisher Scientific.
Propionic acid and ammonium carbonate were purchased from
* To whom correspondence should be addressed. E-mail: paul.rillema@
wichita.edu.
† Wichita State University.
‡ Pittsburgh State University.
Aldrich. IR grade potassium bromide used as a pellet for the
infrared spectra and deuterated dimethyl sulfoxide (DMSO) for
10.1021/jp103749b  2010 American Chemical Society
Published on Web 06/02/2010

Dicyano and Pyridine Derivatives of Retinal
J. Phys. Chem. A, Vol. 114, No. 25, 2010 6767
NMR analysis were obtained from Aldrich. Ferrocene standard
and dried acetonitrile (sure-sealed bottle) used as an electro-
chemical solvent were purchased from Aldrich. Sulfur powder
used as a standard to calibrate the FT-Raman spectrometer was
provided by Thermo Nicolet. Electrochemical grade tetrabuty-
lammonium perchlorate was purchased from Southwestern
Analytical.
Ag/AgNO3 reference electrode in dried CH3CN was freshly
made and used for this study. The supporting electrolyte was
0.1 M tetrabutylammonium perchlorate. Ferrocene was added
as a reference. The sample preparation for emission studies
involved dissolving a small amount of sample in dried buty-
ronitrile and then measuring the absorbance of the solution. The
concentration of the solution was altered in order to achieve an
absorbance of ∼0.10 at the λmax of the sample. The sample
solution was placed in a quartz tube and was degassed a
minimum of five times with the freeze-pump-thaw method
prior to actual measurements. Emission quantum yields were
calculated by using eq 1,
Preparation of all-trans-15-Cyano-15′-pyridylretinal (2).
The following preparation was performed under an argon
atmosphere. All-trans-retinal (100.0 mg, 3.49 × 10-4 mol) was
placed in a 10 mL round-bottomed flask. The compound was
then dissolved in benzene and allowed to stir for 5 min. After
all the solid dissolved, 4-pyridylacetonitrile monohydrochloride
(58.3 mg, 5.30 × 10-4 mol) and ammonium carbonate (98.6
mg, 1.03 mmol) were added to the solution and the reaction
was allowed to stir further for about 5 min. Then, 1.00 mL of
propionic acid was added dropwise into the solution. After all
the reagents were added, the reaction was heated overnight under
argon with the temperature regulated at approximately 70 °C.
The reaction was then allowed to cool in an ice water bath and
the solvent was removed by lyophilization. The red-orange solid
was redissolved in chloroform and subjected to column chro-
matography through an oven-dried neutral alumina (∼10.0 g)
stationary phase. The mobile phase used was 7:1 chloroform:
methylene chloride. The first and final fractions were discarded
while the middle fractions were saved and examined for purity
by TLC. After combining the pure fractions, the solid was
recovered after freeze-drying the solution to remove the solvent.
The orange compound was dried in a vacuum oven and stored
in a desiccator. Color: brick-red. Yield: 80%. Anal. Calcd for
C27H32N2: C, 84.31; H, 8.39; N, 7.29. Found: C, 84.05; H, 7.27;
N, 7.27. IR (KBr pellet, cm-1): (CN) 2208 w, (-CdC-) 1542
sh, 3745 m, 2924 m, 1097 m, 1023 m. Raman (solid, cm-1):
(CN) 2207 w, (-CdC-) 1534 sh, (CdCsH) 1211 sh. 13C
NMR (CDCl3): δ ppm 150.7 (pyr), 140.7, 140.2, 137.5, 135.6,
131.5, 130.3, 129.6, 127.1, 119.6 (CN), 39.9, 34.5, 33.4, 29.2
(ꢀ-ring), 22.0, 19.4, 13.9.
Preparation of all-trans-15,15′-Dicyanoretinal (3). Com-
pound 3 was prepared analogous to compound 2 except
malononitrile (58.3 mg, 5.30 × 10-4 mol) was substituted for
4-pyridylacetonitrile monohydrochloride (58.3 mg, 5.30 × 10-4
mol). The dark-red compound was dried in a vacuum oven and
stored in a desiccator. Color: burgundy. Yield: 85%. Anal. Calcd
for C23H28N2: C, 83.07; H, 8.50; N, 8.43. Found: C, 82.98; H,
7.76; N, 8.45. IR (KBr pellet, cm-1): (CN) 2219 m, (-CdC-)
1544 sh, 3746 w, 2926 w, 1244 w, 1188 w. Raman (solid, cm-1):
(CN) 2215, (-CdC-) 1528, (CdCsH) 1171 sh.
Measurements. Infrared spectra were obtained with Perkin-
Elmer Model 1600 FT-IR and Nicolet Avatar model FT-IR
spectrophotometers. All samples were prepared as potassium
bromide pellets. Raman spectra were obtained with a Thermo
Nicolet Nexus FT-Raman module spectrophotometer. The
Nicolet instruments were accompanied by Omni software
programs. Proton (1H) and 13C NMR spectra were obtained with
Varian Mercury 300 MHz and Varian Inova 400 MHz Fourier
Transform-NMR spectrometers (internal standard TMS). Ul-
traviolet spectra were obtained with a Hewlett-Packard Model
8452A diode array spectrophotometer interfaced with an OLIS
software program. Elemental (C, H, N) analysis was performed
by MHW Laboratories, Phoenix, AZ. An EG&G PAR model
263A potentiostat/galvanostat was used to obtain cyclic volta-
mmograms. All measurements were carried out in a typical
H-electrochemical cell, using a platinum disk working electrode
(polished every run) and a platinum wire counter electrode. A
If
log(ꢁf) ) 0.919 log
- 7.934
(1)
-A
[
]
1 - 10
where ꢁf is the emission quantum yield, If is the emission
intensity, and A is the absorbance of the sample at the excitation
frequency.30 Low-temperature (77 K) emissions were measured
with a Spex Fluorolog 212 spectrofluorometer equipped with a
homemade sample holder connected to an Ar gas tank for
continuous deaeration of the sample compartment.
Calculations. Calculations were affected with Gaussian 03
(Rev. B.03)29 for UNIX. The molecules were optimized by using
Becke’s three-parameter hybrid functional B3LYP31 with the
local term32 of Lee, Yang, and Parr and the nonlocal term33 of
Vosko, Wilk, and Nassiar. The basis set 6-311G** was chosen
for all atoms and the geometry optimizations were all ran in
the gas phase. TDDFT34-40 calculations were employed to
produce a number of singlet excited states41 in the gas phase
based on the optimized geometry. All oscillator values and
singlet and triplet excited state values are presented in the
Supporting Information.
X-ray Crystallography. all-trans-15-Cyano-15′-pyridylreti-
nal (2) and all-trans-15,15′-Dicyanoretinal (3). Crystals were
affixed to a nylon cryoloop with oil (Paratone-n, Exxon) and
mounted in the cold stream of a Bruker Kappa-Apex II area-
detector diffractometer. The temperature at the crystals was
maintained at 150 K by using a Cryostream 700EX Cooler
(Oxford Cryosystems). A scan with 0.3° and time of 10 s was
employed along with graphite monochromated Molybdenum KR
radiation (λ ) 0.71073 Å) that was collimated to a 0.6 mm
diameter. Data collection, reduction, structure solution, and refine-
ment were performed with the Bruker Apex2 suite (v2.0-2).
The unit cell of 2 was determined from the setting angles of
50 reflections collected in 36 frames of data. Data were
measured with a redundancy of 7 with a CCD detector at a
distance of 50 mm from the crystal with a combination of ꢁ
and ω scans. All available reflections to 2θmax ) 54° were
harvested (37745 reflections, 5276 unique) and corrected for
Lorentz and polarization factors with Bruker SAINT (v6.45).
Reflections were then corrected for absorption (numerical
correction, µ ) 0.064 mm-1), interframe scaling, and other
systematic errors with SADABS 2004/1 (combined transmission
and other correction factors min/max ) 0.9731/0.9911).
The unit cell of 3 was determined from the setting angles of
74 reflections collected in 36 frames of data. Data were
measured with a redundancy of 7.8, using a CCD detector at a
distance of 50 mm from the crystal with a combination of ꢁ
and ω scans. All available reflections to 2θmax ) 54° were
harvested (37913 reflections, 4782 unique) and corrected for
Lorentz and polarization factors with Bruker SAINT (v6.45).
Reflections were then corrected for absorption (numerical

6768 J. Phys. Chem. A, Vol. 114, No. 25, 2010
Cruz et al.
TABLE 1: Crystal Data and Structure Refinement for 2 and 3
empirical formula
formula weight
temperature
C27H32N2 (2)
384.55
150 K
C23H28N2 (3)
332.47
150 K
wavelength
crystal system
space group
0.71073 Å
triclinic
P1
0.71073 Å
monoclinic
P21/n
j
unit cell dimensions
a ) 6.8351(7) Å
b ) 7.4409(7) Å
c ) 24.397(2) Å
R ) 89.250(6)°
ꢀ ) 85.685(6)°
γ ) 69.736(5)°
1160.61(19) Å3
2
a ) 10.2960(7) Å
b ) 11.5493(7) Å
c ) 17.6364(12) Å
R ) 90°
ꢀ ) 100.126(3)°
γ ) 90°
volume
Z
calcd density
absorption coeff
F(000)
2064.5(2) Å3
4
1.100 g/cm3
1.070 g/cm3
0.062 mm-1
720
0.064 mm-1
416
crystal size
0.43 × 0.32 × 0.14 mm
0.39 × 0.31 × 0.22 mm
crystal habit
crystal color
θ range for data collection
limiting indices
block
block
dark red
dark red
1.67° to 25.00°
2.12° to 27.11°
-13e h e 13
-14 e k e 14
-22 e l e 22
-8e h e 7
-8 e k e 8
-28 e l e 28
33891/4069 [R(int) ) 0.0862]
99.9%
no. of reflections collected/unique
completeness to θ ) 25.00°
refinement method
no. of data/restraints/parameters
refinement threshold
36637/4549 [R(int) ) 0.1076]
99.6%
full-matrix least-squares on F2
4069/0/268
full-matrix least-squares on F2
4549/0/231
I > 2σ(I)
2417
1.022
I > 2σ(I)
2506
1.048
no. of data > threshold
goodness-of-fit on F2
final R indices [I > 2σ(I)]
R indices (all data)
largest diff peak and hole
R1 ) 0.0627, wR2 ) 0.1455
R1 ) 0.1182, wR2 ) 0.1774
0.471 and -0.257 e- Å-3
R1 ) 0.0681, wR2 ) 0.1570
R1 ) 0.1353, wR2 ) 0.1968
0.468 and -0.172 e- Å-3
correction, µ ) 0.062 mm-1), interframe scaling, and other
systematic errors with SADABS 2004/1 (combined transmission
and other correction factors min/max ) 0.9796/0.9915).
The structures were solved (direct methods) and refined (full-
matrix least-squares against F2) with the Bruker SHELXTL
package (v6.14-1). All non-hydrogen atoms were refined by
using anisotropic thermal parameters. All hydrogen atoms were
included at idealized positions; hydrogen atoms were not refined.
Pertinent crystal, data collection, and refinement parameters are
given in Table 1.
X-ray Results. The crystal structures of 2 and 3 were
determined. Selected bond lengths, bond angles, and torsion angles
were tabulated in Table 2 and ORTEP diagrams are shown in
Figure 3. The retinal-pyridine adduct, in its crystal form, retained
its all-trans configuration as can be seen clearly in Figure 3. The
experimental bond distance for the CtN group was 1.14 Å. For
the polyene chain, localized CsC single bonds in both cases
averaged 1.44 Å while double CdC bonds averaged 1.36 Å. The
new CdC double bond that resulted between the methylene
(-CH2-) and carbonyl (CO) carbon after condensation gave an
average bond distance of 1.36 Å but the bond angles at the
condensation point were not the same. The C(22)-C(16)-C(17)
angle was 117.9° for 2 while the C(17)-C(18)-C(22) angle of 3
was 121.1°. For 2 the C(14)-C(15)-C(16)-C(17) torsion angle
was 177.8° while the C(21)-C(17)-C(16)-C(22) angle was 11.9°.
For 3, the C(16)-C(17)-C(18)-C(22) torsion angle was 178.0°.
The optimized structures of 2 and 3 were also calculated by using
density functional theory (DFT) and the corresponding theoretical
bond lengths and angles were correlated with the experimental data.
Results showed good correlation with alternating single-double
bonds present in the polyene chain. Both the actual crystal and
DFT optimized structures showed gradual bending in the polyene
chain with the methyl groups pointed away from the curvature.
From Figure 3, one also notes the lowering in symmetry of
the polyene derivatives from C2h suggested for ꢀ-carotenes to
possibly Cs due to substituent effects and to twisting of the
polyene chain.
Results and Discussion
Preparations. The preparation of 2 and 3 followed the
Knoevenagel reaction,42,43 which involved the condensation of
the aldehyde, retinal, with substrates lacking R-hydrogen atoms
adjacent to the coupling site. This criterion is required to enhance
the acidity of the methylene (-CH2-) protons.
Traditionally, this condensation reaction is catalyzed by using
only a weak base such as piperidine or NaNH2.42 In our case,
we used simultaneous base- and acid-catalyzed condensation
reactions to enhance the yield. The base catalyst used was
NH4CO3 while the acid catalyst chosen was propionic acid
(CH3CH2COOH). The reaction scheme is shown in Figure 1.
The mechanism of formation of the pyridyl adduct is presented
in Figure 2, which can be extended to the dicyano derivative.
In this reaction, the base catalyst was added first to the reaction
mixture to neutralize hydrochloride and to deprotonate one of
the methylene (-CH2-) protons. Then, the weak acid catalyst
was added to protonate the carbonyl oxygen generating the
carbonyl carbocation necessary to initiate the condensation
reaction.
Calculations (DFT/TDDFT). The HOMO and LUMO orbit-
als determined by DFT calculations are shown in Figure 4; lower
energy HOMO and higher energy LUMO orbitals are shown
in Figure 5.

Dicyano and Pyridine Derivatives of Retinal
J. Phys. Chem. A, Vol. 114, No. 25, 2010 6769
Figure 1. Knoevenagel condensation of all-trans-retinal with (2) 4-pyridylacetonitrile monohydrate and (3) malononitrile.
Figure 2. Mechanism for the acid- and base-catalyzed Knoevenagel condensation of all-trans-retinal with 4-pyridylacetonitrile monohydrate.
Table 3 lists the population analysis data and the percentage
electronic distribution of some selected states in 2 and 3. The
HOMO is predominantly polyene in character with 87%
occupancy in 2 and 67% in 3. The most remarkable difference
is that the HOMO of 3 also showed some important contribu-
tions from the ꢀ-ring (19% occupancy) and the CN group (15%
occupancy). Although the LUMO character lies on the chain
of 2 and 3, the electron density of the LUMO in 2 is distributed
throughout the molecule with a 10% occupancy coming out from
the pyridyl group.
observed in H - 1. In the latter case, a rather different π-bonding
interaction was observed with a nodal plane arising every third
atom of the polyene chain. In H - 3 the electron density is
localized only on the pyridyl ring with no contributions coming
from the polyene chain.
The electronic distributions of the orbitals in 3 are much more
intriguing than the pyridyl derivative. The electron density in
H-1 state is almost divided equally into three segments
throughout the molecule. One segment is localized in the ꢀ-ring
region; another segment is located in the middle of the polyene
chain, while the last one is concentrated in the dicyano group.
H-2, on the other hand, contains “banana-shaped” electron
density lobes extended throughout the molecule. It is also
The pyridyl group in 2 plays an important role in hosting the
excited electron. In some cases, the methyl groups of the polyene
chain have minor contributions in one of the states such as that

6770 J. Phys. Chem. A, Vol. 114, No. 25, 2010
Cruz et al.
calcd
TABLE 2: Selected Bond Lengths (Å) and Angles (deg) for 2 and 3
atoms 2
exptl
calcd
atoms 3
exptl
C(22)-N(2) (nitrile)
C(16)-C(22)
1.142
1.442
1.329
1.324
1.155
1.430
1.334
1.335
C(22)-N(1) (nitrile)
C(22)-N(2) (nitrile)
C(5)-C(9)
1.143
1.142
1.469
1.343
1.451
1.364
1.423
1.353
1.428
1.386
1.395
1.370
1.425
1.430
1.154
1.154
1.471
1.349
1.449
1.365
1.428
1.360
1.437
1.372
1.416
1.371
1.426
1.427
C(19)-N(1) (pyridyl)
C(20)-N(1) (pyridyl)
C(9)-C(10)
C(10)-C(11)
C(11)-C(12)
C(12)-C(13)
C(13)-C(14)
C(14)-C(15)
C(15)-C(16)
C(16)-C(17)
C(17)-C(18)
C(18)-C(22)
C(18)-C(23)
C(6)-C(7)
1.479
1.328
1.458
1.351
1.440
1.346
1.443
1.357
1.424
1.355
1.481
1.549
1.342
1.457
1.361
1.432
1.357
1.442
1.368
1.427
1.366
1.481
C(7)-C(8)
C(8)-C(9)
C(9)-C(10)
C(10)-C(11)
C(11)-C(12)
C(12)-C(13)
C(13)-C(14)
C(14)-C(15)
C(15)-C(16)
C(16)-C(17)
C(22)-C(16)-C(17)
C(22)-C(16)-C(15)
117.9
116.1
116.0
117.2
C(17)-C(18)-C(22)
C(17)-C(18)-C(23)
N(1)-C(22)-C(18)
N(2)-C(23)-C(18)
121.1
120.7
122.0
179.2
178.7
122.9
176.3
178.4
C(14)-C(15)-C(16)-C(17)
C(21)-C(17)-C(16)-C(22)
177.8
11.9
173.7
C(16)-C(17)-C(18)-C(22)
178.0
178.6
TABLE 3: Electron Distribution (%) in 2 and 3
important to note that in all transitions, the dicyano groups are
involved, particularly in the L+1 state.
Figure 6 shows the molecular orbital energy diagrams of 2
and 3. The HOMO-LUMO gaps are consistent with the
observed transition energy as discussed in the electronic
absorption section (see Table 6).
Vibrational Properties. The infrared and Raman spectros-
copy data are given in Table 4 while the Raman spectra of 2
and 3 are shown in Figure 7. Although the intensity of the
Raman scattering in 2 is a lot weaker than that of 3, the polyene
mode of vibration in all cases remains IR and Raman active.
The main points of interest in structural elucidation are the nitrile
and polyene groups before and after condensation. As shown
in Table 4, the carbonyl stretching frequencies of all-trans-retinal
appeared at around 1655 cm-1 (IR and Raman). The IR
frequency of the CtN vibration for malononitrile was located
at 2273 cm-1 while its Raman frequencies were found at 2295
states
2
3
LUMO+1
60 (chain)
2 (ꢀ-ring)
1 (CN)
77 (chain)
12 (ꢀ-ring)
11 (CN)
37 (pyridyl)
82 (chain)
1 (ꢀ-ring)
7 (CN)
LUMO
72 (chain)
3 (ꢀ-ring)
25 (CN)
10 (pyridyl)
87 (chain)
3 (ꢀ-ring)
5 (CN)
HOMO
67 (chain)
19 (ꢀ-ring)
15 (CN)
5 (pyridyl)
77 (chain)
5 (ꢀ-ring)
6 (CN)
11 (pyridyl)
1 (chain)
0 (ꢀ-ring)
0 (CN)
HOMO-1
HOMO-2
35 (chain)
49 (ꢀ-ring)
15 (CN)
57 (chain)
24 (ꢀ-ring)
19 (CN)
99 (pyridyl)
(sym) and 2265 cm-1 (asym). After condensation, the carbonyl
peak of retinal disappeared and peaks for the CtN vibration
appeared. The IR CtN frequency for 2 was located at 2208
cm-1 in close agreement with its Raman frequency at 2207
cm-1. For 3, the related frequencies were 2219 cm-1 in the IR
and 2215 cm-1 in the Raman. It is notable that the CN stretching
frequency in the pyridyl adduct is weaker than the malononitrile
counterpart. This behavior may be due to the effect of pyridine
ring that allows the CN to vibrate freely in the conjugation.
Retention of the polyene groups after the condensation was best
monitored by using Raman spectroscopy. These polyene group
frequencies are also listed in Table 4. The frequency of
alternating carbon-carbon double bond backbone in 1 was
observed at 1574 cm-1 in the IR and at 1568 cm-1 in the Raman.
For 2, the frequency of the polyene backbone was observed at
1542 cm-1 in the IR and at 1534 cm-1 in the Raman, while for
3 it was found at 1544 cm-1 in the IR and at 1528 cm-1 in the
Raman.
Figure 3. ORTEP diagram of 2 and 3.

Dicyano and Pyridine Derivatives of Retinal
J. Phys. Chem. A, Vol. 114, No. 25, 2010 6771
Figure 4. HOMO/LUMO pictures in 2 and 3 determined by DFT.
Figure 5. Orbital pictures of selected states involved in optical transition in 2 and 3.

6772 J. Phys. Chem. A, Vol. 114, No. 25, 2010
Cruz et al.
TABLE 4: Vibrational Stretching Frequencies (cm-1) of CN
and Polyene Groups
Some naturally occurring carotenoids (i.e., ꢀ-carotene) un-
dergo cation radical formation. The irreversible oxidation
observed in the CV of 2 resembles that of ꢀ-carotene.45 It is
possible for the compound to undergo a two-electron oxidation
reaction at the electrode surface to generate a dication inter-
mediate. This EC mechanism is well-established in most
carotenoids. In the case of the dicyano adduct, another irrevers-
ible peak appears after the initial one. This behavior has been
observed in naturally occurring compounds such as ꢀ-apo-8′-
carotenal wherein two- or three-electron-transfer processes are
involved.46
compd
functional group
exptl (IR)
exptl (Raman)
1
CdO stretch
1656
1574
1654
1568
1268
-CdC- skeletal
dCsH in-plane bending
CH3 stretch
2934
2208
1542
2
3
CN stretch
2207
1534
1211
-CdC- skeletal
dCsH in-plane bending
CH3 stretch
2924
2219
1544
CN stretch
2215
1528
1171
Electronic Spectra. The UV/visible spectral data of the
compounds are given in Table 6 and shown in Figure 9. Relative
to 1, compounds 2 and 3 showed bathochromic shifts. For 2,
-CdC- skeletal
dCsH in-plane bending
CH3 stretch
2925
λmax was shifted by 64 to 446 nm; for 3 it was shifted by about
82 to 464 nm. This bathochromic shift was due to the
lengthening of the polyene chain (1-D box). After the condensa-
tion, a new double bond was generated between the carbonyl
carbon of the retinal and the methylene groups of pyridine-CN
and malononitrile. Simulated UV/visible absorption spectra of
1, 2, and 3 were determined by using TDDFT calculations and
are overlaid with the actual electronic spectra (Figure 9). Results
showed good correlation between experimental and calculated
spectra. In the case of 2, the calculated spectrum underwent a
hypsochromic shift from 446 to 441 nm in its lowest energy π
f π* transition while 3 underwent a bathochromic shift from
464 to 487 nm. Calculations revealed that the lowest energy
transition occurred from the HOMO to LUMO in both 2 and 3
with an energy gap of 441 nm for the 2 and 489 nm for 3.
Experimentally, two other major bands were observed in the
spectra of both 2 and 3. For 2, a weak absorption band was
also observed near 290 nm while a similar band was observed
in the spectrum of 3 centered at 285 nm. These bands (Figure
9) correlated with the calculated 299 and 284 nm bands,
respectively.
Alkene protons in conjugated systems are known to be Raman
active modes. These proton frequencies of the methyl groups
and the polyene backbone protons are also listed in Table 4.
Cyclic Voltammetry. Oxidations for 1, 2, and 3 were
examined and the data are summarized in Table 5. Figure 8
shows the cyclic voltammograms of 2 and 3. Electrochemical
results showed irreversible oxidation waves for all three, which
is typical in most retinoids and carotenoids44-46 supporting their
antioxidant properties; no reversible reduction was observed.
For 2, the irreversible peak was observed at 0.78 V while for 3
it was observed at a higher potential of 0.98 V. Both were shifted
to a lower electrode potential with respect to all-trans-retinal
(1.10 V). The much larger shift for the 2 derivative may be
related to additional π conjugation with the pyridine ring.
As expected, the polyene chain is the major contributor to
the lowest-energy transitions. The HOMO of 2 (Figure 4) shows
the electron density as located on the polyene chain while the
LUMO shows some electron density has shifted to the pyridyl
side increasing the conjugative (π f π*) behavior between the
polyene and the pyridyl group. The HOMO of 3 is spread out
over the polyene chain, the dicyano group, and the ꢀ-ring. Some
p-orbital contribution in the case of methylene carbon is also
observed. The LUMO, on the other hand, had a more distorted
conjugated π-bonding system with its electron density shifted
more toward the dicyano group.
As shown in Figure 9, the simulated UV/visible spectra
obtained with TD-DFT calculations gave rise to four distinct
theoretical bands in 2 while three bands were calculated for 3.
Table 7 lists selected transitions, their corresponding oscillator
strengths, and the major molecular orbitals that contribute in
each transition.
The HOMO to LUMO transition of 2 has high oscillator
strength in agreement with its high absorption coefficient and
is the predominant peak calculated in the UV-visible spectrum
with its maximum centered at 442 nm. The other calculated
peaks merge pairwise into the peaks shown in Figure 9. For
example, the calculated peaks at 305 and 299 nm when added
together appear as one peak. The calculated peak at 299 nm
gave oscillator strength of 0.23. Under this band lie the following
transitions: HOMO f L+1, HOMO f L+3, and H-1 f
LUMO.
The calculated UV/visible data of 3 are also listed in Table
7. The HOMO-LUMO transition dominates with high oscillator
Figure 6. Molecular orbital energy diagrams of 2 and 3.

Dicyano and Pyridine Derivatives of Retinal
J. Phys. Chem. A, Vol. 114, No. 25, 2010 6773
Figure 7. Raman spectra of 2 and 3.
TABLE 5: Oxidation Electrode Potentials of Retinoidsa
Previous studies also showed that this approach can be extended
to nonsymmetric polyene chains.1,9-22,47-52
compd
E (V)
all-trans-retinal
2
3
1.10
0.78
0.98
The electronic energy level diagram proposed for ꢀ-carotenes
based on C2h symmetry is shown in Figure 10. Direct absorption
1
1
occurs from the S0 (1 Ag) f S2 (1 Bu)2 state. Vibrational
relaxation and internal energy conversion occurs to the (2 1Ag)
a 0.1 M TBAClO4 in acetonitrile.
1
state from which emission takes place to the (1 Ag) state.
TABLE 6: UV/Visible data of S0 1Ag f S2 1Bu transition in
1, 2, and 3
Previous calculations (semiempirical) have shown that the 1Ag
f
1Bu transitions in polyenes have high molar extinction
compd
E(λmax) (exptl)
ε (M-1 cm-1
)
E(λmax) (calcd)
coefficients.2 The HOMO f LUMO transitions of 2 and 3
located at 442 (f ) 1.8) and 487 nm (f ) 1.5), respectively,
correspond to this criterion.
1
2
3
382
446
464
44000
37673
86800
408
442
487
The optical transition at ∼280-290 nm has been assigned
as 1 Ag f 3 Ag and labeled the “cis-peak” corresponding to
perceived photoinduced cis-trans isomerization. From DFT
calculations, this corresponds to the HOMO to LUMO+1
transition for 2 located at 299 nm and for 3 located at 284 nm.
Emission Studies. The emission properties of 2 and 3 were
collected at 77 K; room temperature emission for both was weak.
Population of this first excited (2 1Ag) state at 77 K by internal
conversion resulted in an emission band at lower energy (λem
for 2 ) 583 nm and λem for 3 ) 500 nm) as shown in Figure
11. One interesting feature valid for both compounds is that
more intense emission occurred upon excitation at higher
strength 1.5 times greater than the others. Two peaks also are
observed located near the other calculated bands of high
oscillator strength. The theoretical peak observed at around 284
nm has oscillator strength of about 0.14 and involves predomi-
nately the population of the L+1 state. This information can
be related to symmetry adapted orbital labels deduced by using
Group theory for naturally occurring retinoids and carotenoids.
In general, the method uses Ag and Bu symmetry orbital labels
deduced when symmetry operations in C2h point group were
applied to ꢀ-carotene rather than treating it with π (π*) labels.
Figure 8. Cyclic voltammogram of 2 and 3.

6774 J. Phys. Chem. A, Vol. 114, No. 25, 2010
Cruz et al.
Figure 9. Experimental and calculated UV/visible data for 2 and 3.
TABLE 7: Calculated UV/Visible Data for 2 and 3
2
3
λ
(nm)
f
major contributors
λ
(nm)
f
major contributors
442 1.8130 HOMO f LUMO 487 1.4752 HOMO f LUMO
337 0.2685 H-1 f LUMO
374 0.6686 H-1 f LUMO
HOMO f L+1
327 0.0014 H-2 f LUMO
330 0.0547 H-2 f LUMO
HOMO f L+1
305 0.0069 H-3 f LUMO
284 0.1392 HOMO f L+1
H-2 f LUMO
H-1 f L+1
299 0.2312 HOMO f L+1
HOMO f L+3
265 0.0058 H-3 f LUMO
H-1 f L+1
H-1 f LUMO
279 0.0029 HOMO f L+2
HOMO f L+3
259 0.0184 H-1 f L+1
HOMO f L+2
H-3 f LUMO
265 0.1569 H-4 f LUMO
H-6 f LUMO
255 0.0005 H-4 f LUMO
HOMO f L+2
HOMO f L+3
wavelengths (λex ) 290 nm in 2 and λex ) 280 nm in 3) rather
than at the highly intense visible band. These data can be
compared to the emission behavior of all-trans-retinal deter-
mined in our laboratories. Excitation at 380 nm resulted in
emission at 570 nm. The emission intensity increased 50-fold
at 77 K. The quantum yields of emission for all-trans-retinal
determined in the lab with use of eq 1 at room temperature and
77 K are 0.001096 and 0.04465, respectively. The quantum yield
compares favorably to that for 2 (0.005465) and 3 (0.005371)
at 77K.
Figure 10. Energy diagram for 2 and 3.
dicyanoretinal, 3), retained their trans geometry as determined
by X-ray diffraction for each compound, but the chains were
bowed. Molecular orbitals determined for the compounds by
Density Functional Theory revealed that the highest occupied
molecular orbital (HOMO) and lowest unoccupied molecular
orbital (LUMO) were located primarily on the polyene chain
and the energy gap between them was consistent with the
observed optical spectrum with maxima located at 442 nm of 2
and 487 nm of 3. Optical transitions deduced from Time
Dependent Density Functional Theory calculations correlated
with experimental results and were interpreted based on the
group theoretical approach as previously applied to ꢀ-carotene,
with the lowest energy absorption in the visible region occurring
from the ground state 1 1Ag (HOMO) f 1 1Bu (LUMO) excited
state. Low-temperature (77 K) emission observed at 583 nm of
2 and 500 nm of 3 were assigned to decay from the 2 1Ag state
The excitation spectra peaked at 355 nm for 1, 445 nm in 2,
and 410 nm for 3. However, according to the Franck-Condon
principle, excitation occurs vertically to higher vibronic states
1
of the Bu state followed by internal conversion to the 21Ag
excited state, which then leads to vibrational relaxation followed
1
by fluorescence emission back to the ground (1 Ag) state.
Emission lifetimes in all three cases were more rapid than 5 ns,
which is the pulse width of our nanosecond laser.
Summary
1
to its 1 Ag ground state. TDDFT calculations further reveal
Two new derivatives of all-trans-retinal synthesized via acid-
catalyzed Knoevenagel condensation reaction, one containing
a pyridine and a cyano substituents (15,15′-cyanopyridylretinal,
2) and the other containing two cyano substituents (15,15′-
that the electronic transitions in the 280-290 nm region are
HOMO to LUMO+1 (1 1Ag f 3 1Ag) transitions and most likely
are associated with the “cis” peak, which is associated with
photoinduced “cis-trans” isomerization in ꢀ-carotenes.

Dicyano and Pyridine Derivatives of Retinal
J. Phys. Chem. A, Vol. 114, No. 25, 2010 6775
Figure 11. (a) Low-temperature (77 K) emission spectra of 2 and (b) low-temperature (77 K) excitation (red line) and emission (black line)
spectrum in 3.
(26) Sakai, T.; Ikeda, K. H.; Kawasaki, K. Appl. Phys. Lett. 1990, 56,
411.
(27) Evans, D. F. J. Chem. Soc. 1960, 1735.
Acknowledgment. We thank the support from the Wichita
State University High Performance Computing Center, the
Wichita State University Office of Research Administration, and
the Department of Energy.
(28) Raubach, R. A.; Guzzo, A. V. J. Phys. Chem. 1973, 75, 983.
(29) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vrevem, T.; Kudin, K. N.;
Burant, J. C.; Millam, J. R.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci,
B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.;
Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.;
Nakajima, T.; Honda, Y.; Hitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox,
J. E.; Hratchian, H. P.; Cross, H. B.; Bakken, V.; Adamo, C.; Jaramillo, J.;
Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.;
Pommelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.;
Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels,
A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.;
Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.;
Clifford, S.; Cioslowki, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz,
P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.;
Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson,
B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A., Gaussian 03,
Revision B.04; Gaussian Inc., Wallingford, CT, 2004.
(30) Kirgan, R.; Moore, C.; Witek, P.; Rillema, D. P. Dalton Trans.
2008, 3189.
Supporting Information Available: Tables of Cartesian
coordinates and excited states for 1-3 and CIF files for 2 and
3. This material is available free of charge via the Internet at
http://pubs.acs.org.
References and Notes
(1) Cogdell, R. J.; Frank, H. A. Biochim. Biophys. Acta 1987, 895,
63.
(2) Young, A. J.; Britton, G. Carotenoids in Photosynthesis, 1st ed.;
Chapman and Hall: London, UK, 1993.
(3) Borland, C. F.; McGarvey, D. J.; Truscott, T. G.; Cogdell, R. J.;
Land, E. J. J. Photochem. Photobiol. B 1987, 93.
(4) Palacios, R. E.; Kodis, G.; Herrero, C.; Ochoa, E. M.; Gervaldo,
M.; Gould, S. L.; Kennis, J. T. M.; Gust, D.; Moore, T. A.; Moore, A. L.
J. Phys. Chem. B 2006, 110, 2541.
(5) Griffiths, M.; Sistrom, W. R.; Cohen-Bazire, G.; Stanier, R. Y.
Nature 1955, 176, 1211.
(31) Becke, A. D. J. Chem. Phys. 1993, 98, 5648.
(32) Lee, C.; Yang, W.; Parr, R. G. Phys. ReV. B 1988, 37, 785.
(33) Vosko, S. H.; Wilk, L.; Nusair, M. Can. J. Phys. 1980, 58, 1200.
(34) Andrae, D.; Hauessermann, U.; Dolg, M.; Stoll, H.; Preuss, H.
Theor. Chim. Acta 1990, 77, 123.
(6) Krinsky, N. I. Photophysiology; Academic Press: New York, 1968;
Vol. III.
(7) Foote, C. S.; Chang, Y. C.; Denny, R. W. J. Am. Chem. Soc. 1970a,
92, 5218.
(35) Stratmann, R. E.; Scuseria, G. E.; Frisch, M. J. J. Chem. Phys.
(8) Foote, C. S.; Chang, Y. C.; Denny, R. W. J. Am. Chem. Soc. 1970b,
92, 5216.
(9) Frank, H. A.; Young, A. J.; Britton, G.; Cogdell, R. J. The
Photochemistry of Carotenoids; Klewer Academic: Dordrecht, The Neth-
erlands, 1999.
(10) Zoos, Z. G.; Ramasesha, S. Phys. ReV. B 1984, 29, 5410.
(11) Christensen, R. L.; Kohler, B. E. Photochem. Photobiol. 1973, 18,
293.
(12) Hudson, B. S.; Kohler, B. E.; Shulten, K. Excited States; Academic
Press: New York, 1987; Vol. VI.
(13) Koyama, Y.; Kuki, M.; Andersson, P. O.; Gillbro, T. Photochem.
Photobiol. 1996, 63, 243.
(14) Orlandi, G.; Zerbetto, F.; Zgierski, M. Z. Chem. ReV. 1991, 91,
867.
(15) Kohler, B. E.; Samuel, I. D. W. J. Chem. Phys. 1995, 103, 6253.
(16) Labhart, H. J. Chem. Phys. 1957, 27, 957.
(17) Tavan, P.; Schulten, K. J. Chem. Phys. 1979, 70, 5407.
(18) Tavan, P.; Schulten, K. J. Chem. Phys. 1986, 85, 6602.
(19) Petek, H.; Bell, A. J.; Choi, Y. S.; Yoshihara, K.; Tounge, B. A.;
Christensen, R. L. J. Chem. Phys. 1995, 102, 4726.
(20) Snyder, R.; Arvidson, E.; Foote, C.; Harrigan, L.; Christensen, R. L.
J. Am. Chem. Soc. 1985, 107, 4117.
1998, 109, 8218.
(36) Bauernschmitt, R.; Ahlrichs, R. Chem. Phys. Lett. 1996, 256, 454.
(37) Casida, M. E.; Jamorski, C.; Casida, K. C.; Salahub, D. R. J. Chem.
Phys. 1998, 108, 4439.
(38) Cossi, M.; Barone, V. J. Chem. Phys. 2001, 115, 4708.
(39) Barone, B.; Cossi, M. J. Phys. Chem. A 1998, 102, 1995.
(40) Cossi, M.; Rega, N.; Scalmani, G.; Barone, V. J. Comput. Chem.
2003, 24, 669.
(41) Stoyanov, S. R.; Villegas, J. M.; Rillema, D. P. Inorg. Chem. 2002,
41, 2941.
(42) Knoevenagel, E. Chem. Ber. 1894, 27, 2345.
(43) Ren, Z.; Cao, W.; Tong, W. Synth. Commun. 2002, 32, 3475.
(44) Grant, J. L.; Kramer, V. J.; Ding, R.; Kispert, L. D. J. Am. Chem.
Soc. 1988, 110, 2151.
(45) Jeevarajan, A. S.; Khaled, M.; Kispert, L. D. J. Phys. Chem. 1994,
98, 7777.
(46) Khaled, M.; Hadjipetrou, A.; Kispert, L. D.; Allendoerfer, R. D. J.
Phys. Chem. 1991, 95, 2438.
(47) Ishii, K.; Takeuchi, S.; Tahara, T. Chem. Phys. Lett. 2006, 418,
307.
(48) Snyder, R.; Arvidson, E.; Foote, C.; Harrigan, L.; Christensen, R. L.
J. Am. Chem. Soc. 1985, 107, 4117.
(49) Young, A. K.; Phillip, D.; Ruban, A. V.; Horton, P.; Frank, H. A.
(21) Granville, M. F.; Holtom, G. R.; Kohler, B. E. J. Chem. Phys. 1980,
72, 4671.
(22) Gillbro, T.; Cogdell, R. J. Chem. Phys. Lett. 1989, 158, 312.
(23) Hand, E. S.; Belmore, K. A.; Kispert, L. D. J. Chem. Soc., Perkin
Trans. 2 1993, 659.
(24) Gilmour, S.; Marder, S. R.; Tiemann, B. G.; Chen, L. T. J. Chem.
Soc., Chem. Commun. 1993, 432.
(25) Hashimoto, H.; Nakashima, T.; Hattori, K.; Yamada, T.; Mizoguchi,
T.; Koyama, Y.; Kobayashi, T. Pure Appl. Chem. 1999, 71, 2225.
Pure Appl. Chem. 1997, 69, 2125.
(50) Andersson, P. O.; Gillbro, T. J. Lumin. 1992, 51, 11.
(51) Moroni, L.; Gellini, C.; Salvi, P. R.; Schettino, V. J. Phys. Chem.
A 2000, 104, 11063.
(52) O’Neil, M. P.; Wasielewski, M. J. Chem. Phys. 1991, 95, 7212.
JP103749B