Femtosecond Spectroscopic Study of
Carminic Acid-DNA Interactions
Den Naturwissenschaftlichen Fakultäten
der Friedrich-Alexander-Universität Erlangen-Nürnberg
zur
Erlangung des Doktorgrades
vorgelegt von
Dipl.-Phys. Radu Comanici
Aus
Piatra Neamt
2007
Als Dissertation genehmigt von den Naturwissenschaftlichen Fakultäten der
Universität Erlangen-Nürnberg
Tag der mündlichen Prüfung: 29.07.2007
Vorsitzender der Promotionskommission:
Prof. Dr. Bänsch Eberhard
Erstberichterstatter:
Prof. Dr. Carola Kryschi
Zweitberichterstatter:
Prof. Dr. Rainer Fink
Contents
1. Introduction........................................................................................................................ 10
2. Materials and Methods ...................................................................................................... 13
2.1 Materials....................................................................................................................... 13
2.1.1.1 DNA ................................................................................................................... 13
2.1.1.2 Binding Mode..................................................................................................... 17
2.1.1.3 Structural, Electronic and Spectroscopic Properties of Carminic Acid ............. 19
2.2.1 Stationary Optical Spectroscopy............................................................................... 21
2.2.1.1 UV/VIS Absorption Spectroscopy..................................................................... 22
2.2.1.2 Fluorescence Spectroscopy ................................................................................ 25
2.2.1.3 Measurement of Fluorescence Quantum Yield and Fluorescence Lifetime ...... 27
2.2.2 Time-Resolved Optical Spectroscopy....................................................................... 31
2.2.2.1 Femtosecond Transient Absorption Spectroscopy............................................. 31
2.2.2.2 Femtosecond Fluorescence Up Conversion ....................................................... 41
3. Experimental...................................................................................................................... 48
3.1 Materials: Chemicals and Sample Solutions ............................................................... 48
3.2 Methods........................................................................................................................ 49
3.2.1 UV/VIS Absorption and Fluorescence Spectroscopy........................................... 49
3.2.2 Determination of the Fluorescence Quantum Yield ............................................. 49
3.2.3 Fluorescence Titration Experiments ..................................................................... 50
3.2.4 Femtosecond Transient Absorption Spectroscopy................................................ 50
3.2.5 Femtosecond Fluorescence Up-Conversion Technique........................................ 51
3.2.6 Computations ........................................................................................................ 53
4. Results and Discussion ...................................................................................................... 54
4.1. Stationary Optical Spectroscopy................................................................................. 54
4.2. Femtosecond Spectroscopy......................................................................................... 67
4.2.1 Transient Absorption Spectroscopy ...................................................................... 67
4.2.2 Fluorescence Up-Conversion Spectroscopy ......................................................... 70
5. Conclusions........................................................................................................................ 80
6. References.......................................................................................................................... 82
List of Figures
Fig. 2.1.1 Heterocyclic bases A: pyrimidines 1: uracil, 2: thymine, 3: cytosine; ................ 14
Fig. 2.1.2 Structure components of the common nucleotides. .............................................. 14
Fig. 2.1.3 Tautomers of uracil. .............................................................................................. 15
Fig 2.1.4 Binding of guanine with cytosine .......................................................................... 16
Table 2.1.5 Redox potential of the DNA bases at a pH value of 7. ...................................... 16
Fig. 2.1.5 Structure of a section of DNA............................................................................... 17
Table 2.1.1.2 Thermodynamic binding parameters for the interaction of doxorubicin,
daunorubicin, hydroxyrubicin and the β anomer of doxorubicin with calf thymus DNA; Keq
is the binding constant and n is the number of base pairs per binding site. .......................... 18
Table 2.1.1.3 Free energy of anthracycline antibiotic binding to calf thymus DNA. ........... 19
Fig. 1.3.1: Structure formula of carminic acid. ..................................................................... 20
Fig. 2.2.1 Spectrum of electromagnetic radiation: the spectral range of optical spectroscopy
is depicted in enlarged form. ................................................................................................. 22
Fig. 2.2.1.1 Schematic representation of a UV/VIS absorption spectrometer. ..................... 24
Figure2.2.1.2.1 Jabłoński term scheme. ................................................................................ 25
Fig. 2.2.1.2.2 Setup of a fluorescence spectrometer.............................................................. 27
Fig 2.2.2.1.1 Energy scheme of the electronic states involved in a pump-probe.................. 33
experiment; excited states relaxation dynamics detected by absorption changes of the probe
............................................................................................................................................... 33
pulse are: 1) bleaching; 2) excited state absorption; 3) stimulated emission. ....................... 33
Fig. 2.2.2.1.1 Schematic representation of the fs transient absorption spectroscopy
experiment; BS: Beam Splitter; FM: Flip Mirror; DS: delay stage; VA: variable attenuator;
P: polarizer; L: lens; WLC plate: rotating fused silica plate for white light continuum
generation; PM: parabolic mirror; SHG: BBO crystal for second harmonic generation; BC:
Berek compensator; BD : beam dump; BG: BG38 filter; GG: GG420 filter; ND: neutral
density filter. .......................................................................................................................... 36
Fig. 2.2.2.1.4 a) Intensity as a function of time for a Gaussian laser pulse b) Time
dependence of the frequency for a positive nonlinear index of refraction, n2....................... 39
Fig. 2.2.2.1.5. Representation of the wlc probe pulse as a composition of the temporally
shifted, different spectral sub-pulses having the same pulse width as the pump pulse......... 39
Table 2.2.2.2 Relative quantum efficiencies ηq (normalized relative to KDP), damage
thresholds Ithr, and cut-off wavelengths for fluorescence up-conversion with 800 nm pump
pulses. .................................................................................................................................... 44
Fig. 2.2.2.2.1 Femtosecond up-conversion technique apparatus; DM: dichroic mirror; HW:
half wave plate; CCD video camera for the visual superposition of the beams in the BBO
crystal; Mono: monochromator; PM: photomultiplier. ......................................................... 46
Scheme 1: Dissociation reaction of carminic acid. ................................................................ 54
Fig. 4.1.1: pH dependence of the spectral features of carminic acid in water measured by
UV/VIS absorption spectroscopy. ......................................................................................... 55
Fig. 4.1.2: Dual fluorescence of carminic acid with a blue emission peak at 470 nm (22700
cm-1) and an orange emission peak at 570 nm (15100 cm-1)................................................. 56
Fig. 4.1.3: Orange fluorescence of the tautomer at 15100 cm-1. ........................................... 57
Fig 4.1.4: Normalized fluorescence spectra of 5µM carminic acid in BPES (dashed line) and
5µM carminic acid with 5µM DNA in BPES (solid line)..................................................... 58
Fig. 4.1.5: Absorption of carminic acid in BPES; the band structure (thin solid line) was
analyzed by fitting with a superposition of four Gaussian function (dashed line)................ 59
Fig. 4.1.6: Absorption of carminic acid in DMSO; the band structure (thin solid line) was
analyzed by fitting with a superposition of four Gaussian function (dashed line)................ 60
Scheme 2: Molecular structure of the normal form of carminic acid (CAH) and the three
tautomers (CAH T1, CAH T2, CAH T3) .............................................................................. 61
Table 4.1.1: Calculated values of the total energy (ET), the binding energy (EB), the absolute
∆(S0-Sm)) and S0-Sn transition (∆
∆(S0-Sm))
energy (Eabs), the energy of the S0-Sm transition (∆
transition and of the oscillator strength (f). .......................................................................... 62
Fig. 4.1.7: The spectra of the orange fluorescence of 5 µM carminic acid in BPES. The band
structure of the spectra (thin solid line) was analyzed by fitting with a superposition of
Gaussian functions (dashed line). .......................................................................................... 64
Fig. 4.1.8: The spectra of the orange fluorescence of 5 µM carminic acid in DMSO; the
band structure of the spectra (thin solid line) was analyzed by fitting with a superposition of
Gaussian functions (dashed line). .......................................................................................... 65
Fig. 4.1.9: Concentration dependence of bound carminic acid, cB, on the DNA
concentration; the experimental data (dots) were obtained from fluorescence titration of 6
µM carminic acid with DNA in BPES and were fitted (solid line) employing the
relationship cB= cT⋅⋅ cDNA/(KB-1+cDNA) with KB= 5.0×
×105 (M nucleotide)-1. .......................... 67
Fig. 4.2.1.1 3-D plot of the temporal evolution of the transient absorption spectra obtained
for carminic acid in BPES. .................................................................................................... 68
Fig. 4.2.1.2 Transient absorption spectrum (thin solid line) recorded at the delay time τ= 1
ps; the fit (thick short dashed line) arises from the superposition of 8 Gaussian functions
assigned to four different time constants (thick solid, dashed, dashed dotted and dotted
lines). ..................................................................................................................................... 69
Fig. 4.2.2.1 Fluorescence-up conversion spectrum (dots) fitted by a superposition (thin solid
line) of two Gaussians (dashed line)...................................................................................... 70
Fig. 4.2.2.2 3-D plot of the temporal evolution of the fluorescence up-conversion spectra
obtained for 0.6 mM carminic acid in BPES......................................................................... 71
Fig.4.2.2.3: 3-D plot of the temporal evolution of the fluorescence up-conversion spectra
obtained for 0.6 mM carminic acid and 3.9 mM DNA in BPES........................................... 72
Fig. 4.2.2.4: Fluorescence up-conversion decays of 0.6 mM carminic acid in BPES (thin
solid line) detected at 14910 cm-1 in the parallel-polarization geometry (I(t)par) and in the
perpendicular-polarization geometry (I(t)perp); both fits were obtained using the parameters:
a = 0.37, b= 0.91, τ1= 1.7 ps, τ2= 33 ps and r(0)= 0.137. ...................................................... 73
Fig. 4.2.2.5: Fluorescence up-conversion decays of 0.6 mM carminic acid in BPES (thin
solid line) detected at 16490 cm-1 in the parallel-polarization geometry (I(t)par) and in the
perpendicular-polarization geometry (I(t)perp); both fits were obtained using the parameter:
a= 0.50, b= 0.92, τ1= 1.5 ps, τ2= 47 ps and r(0)= 0.202. ...................................................... 74
Fig. 4.2.2.6: Fluorescence up-conversion decays of 0.6 mM carminic acid and 3.9 mM
DNA in BPES (thin solid line) detected at 14910 cm-1 in the parallel-polarization geometry
(I(t)par) and in the perpendicular-polarization geometry (I(t)perp); both fits were obtained with
the parameter: a= 0.50, 0.97, τ1= 1.9 ps, τ2= 48 ps and r(0)= 0.182. .................................... 75
Fig. 4.2.2.7: Fluorescence up-conversion decays of 0.6 mM carminic acid and 3.9 mM
DNA in BPES (thin solid line) detected at 16490 cm-1 in the parallel-polarization geometry
(I(t)par) and in the perpendicular-polarization geometry (I(t)perp); both fits were obtained with
the parameter: a= 0.50, b= 0.97, τ1= 1.1 ps, τ2= 61 ps and r(0)= 0.243. ............................... 76
Table 4.2.2.1: The amplitudes, a and b, the time constants, τ1 and τ2, and the anisotropy,
r(0), were obtained from the best fit of the fluorescence up-conversion decay curves that are
detected at 14910 cm-1 and 16490 cm-1. ................................................................................ 78
Introduction
1. Introduction
Drugs for treatment of cancer on basis of anthracyclines have been extensively
studied for decades in an effort to optimize their therapeutic function. These compounds are
believed to develop their cytotoxic effect by penetrating into the tumor cell nucleus and
interacting there with DNA [1-4]. The formation of drug-DNA complexes is determined by
the structural features of the anthracyclines composed of a dihydroxy-anthraquinone
chromophore with one or two glycosyl side chains. The formation of intercalation
complexes has been observed to inhibit the DNA replication and the RNA transcription that
blocks the gene expression [3]. Irradiation with light enhances the cytotoxicity of
anthracyclines (e.g. daunomycin) by several orders of magnitudes [5-8]. This photoactivation effect is understood to originate from an ultrafast electron transfer reaction from a
G base of the DNA to the intercalated chromophore, which is associated with the oxidation
of the G base and the reduction of the dihydroxyanthraquinone [7, 8]. This hypothesis is
based on femtosecond spectroscopy studies of daunomycin-DNA and adriamycin-DNA
complexes, yielding a decrease of the S1 state lifetime of the drug by three orders of
magnitude when DNA is present [7]. The lifetime of the daunomycin-DNA complex with
τ= 290 fs was ascribed to the occurrence of the photo-induced electron transfer from the G
base to the 1,4-dihydroxyanthraquinone chromophore. On the other hand, other
radiationless decay processes such as intersystem crossing, internal conversion and excitedstate intramolecular proton transfer (ESIPT) may also be enhanced by conformational
reorganization that daunomycin experiences in the hydrophobic environment of the DNA
base pair stacking. Despite extensive research activities on the examination of photoactivated anthracycline-DNA complexes [7-11] to date there exists no unambiguous
evidence for photo-induced oxidation of the DNA and moreover, the excited-state relaxation
dynamics as well as the structural mechanism at the molecular level are more hypothetical
than really understood.
10
Introduction
The proton transfer is a fundamental process and stay for instance at the basis of
various biochemical reactions such as occurring in enzymes [12-17]. In technical
applications proton transfer reactions are employed to build up optical memories or as
fundamental process for proton transfer LASER.
Excited state intramolecular proton transfer (ESIPT) is induced by optical excitation
that changes the electronic configuration of the molecule and thereupon, promotes the
tautomerization reaction in the electronically excited state. Proton transfer may occur
between two functional groups, for instance between (–OH, -NH2) and (-C=O, -N=),
resulting either in a tautomerization reaction or in an incomplete transfer of a hydrogen
atom. Both are associated with the rearrangement of the molecular structure. This structural
rearrangement results into a neutral tautomer having an energetically distinct molecular
configuration in the excited state which is responsible for the Stokes shift of the tautomer
fluorescence relatively to the fluorescence of the normal form.
Femtosecond (fs) resolved transient absorption spectroscopy experiments of ESIPT
processes may give information of the structural relaxation dynamics following the proton
transfer. The drawback of transition absorption spectroscopy is that this technique does not
allow for the specific detection of only one photophysical deactivation process: the
fluorescence, the internal conversion or the intersystem crossing. To obtain an indepth
insight in the scenario of structural, radiationless and radiative relaxation the ESIPT
reactions of the compounds under study have been studied using complementary fsresolving techniques, as the fs transient absorption spectroscopy and the fluorescence up
conversion spectroscopy.
Our research work presents stationary and time resolved optical spectroscopic studies
of carminic acid-DNA complexes using standard techniques of stationary spectroscopy as
well as femtosecond-resolved transient absorption spectroscopy and fluorescence upconversion technique. We have focused onto carminic acid (7-α-D-glycopyranosyl)9,10,dihydro-3,5,6,8-tetrahydroxy-1-methyl-9,10-dioxo-2-anthracenecarboxylic acid), since
this natural dye consists of a tetrahydoxy-anthraquinone chromophore, with a pendant
11
Introduction
glycosyl moiety and thus constitutes the essential structural features of the anthracyclines.
The main impetus for us is to understand the complex mechanism and dynamics of the
excited-state relaxation of carminic acid in the presence of DNA. Carminic acid as a
carboxylic acid is expected to dissociate in water, so that two forms should coexist at pH =
7, the fully protonated acid (CAH) and the deprotonated anion (CA-).
Although the UV/VIS absorption spectra and the fluorescence spectra of carminic
acid and their dependence on the pH value had been still investigated in detail [18-20], no
contribution by tautomerization reactions, that may occur in the ground state (S0) as well as
in the excited state (S1), to the spectroscopic properties were taken into account. This is
really surprising, since excited-state intramolecular proton transfer (ESIPT) has been shown
to accelerate the rate of internal conversion by reducing the energy gap between the S0 and
S1 states [20]. In order to get a comprehensive understanding of the ESIPT kinetics of
carminic acid and its sensitivity to the microenvironment of the DNA base stacks, there are
performed ZINDO/S calculations of the electronic spectra of the four carminic acid
tautomers in both its non-dissociated and dissociated forms. In addition, have been studied
the binding of carminic acid to DNA upon performing fluorescence titration experiments.
12
Materials and Methods
2. Materials and Methods
2.1 Materials
2.1.1.1 DNA
Deoxyribonucleic acid (DNA) is a nucleic acid in the form of a double helix that
contains the genetic instructions specifying the biological development of all cellular forms
of life and many viruses. DNA is a long polymer of nucleotides that encodes the sequence
of the amino acid residues in proteins using the genetic code, a triplet code of nucleotides.
DNA consists of two complementary strands linked by hydrogen bonds. Each strand of
DNA is a chain of nucleotides consisting each of a heterocyclic base, a sugar and one or
more phosphate groups. In the most common nucleotides the base is a derivative of purine
or pyrimidine as being adenine (A), cytosine (C), guanine (G) or thymine (T) (see Figure
2.1.1), and the sugar is the pentose deoxyribose or ribose (see Figure 2.1.2) [21, 22].
Between the two strands, each base can only mate with one distinct base: resulting into one
of the following combinations: A+T, T+A, C+G and G+C. For instance, A on one strand of
double-stranded DNA will match properly only with a T on the other complementary strand.
A pair of two nucleotides is a so-called base pair [21].
The oxidation of DNA occurs most likely at guanine residues due to the lowest
reduction potential of this base in comparison with cytosine, thymine, and adenine [23, 24]
(see Table 2.1.5).
13
Materials and Methods
A
1
2
3
B
1
2
Fig. 2.1.1 Heterocyclic bases A: pyrimidines 1: uracil, 2: thymine, 3: cytosine;
B: purines 1: adenine, 2: guanine.
OO
P
O
O
O
-
-
P
O
O
P
O
O
Base
O- O
Nucleotide
monophosphate
Nucleotide
diphosphate
Nucleotide
triphosphate
HO
OH -(Ribose)
H - (Deoxiribose)
Fig. 2.1.2 Structure components of the common nucleotides.
The uracil base (A1) is a pyrimidine and have been found in RNA making a pair with
the adenine. In DNA it is replaced by thymine. Theoretically the uracil base can be
14
Materials and Methods
combined with any base, although it only mates with adenosine via a hydrogen bonding. As
hydrogen bond acceptor the uracil makes up to three hydrogen bonds. Uracil occurs as two
tautomers that are at pH=7 the lactam (keto) and the lactim (enol).
Fig. 2.1.3 Tautomers of uracil.
Another pyrimidine base is thymine of DNA (A2). Thymine can be formed by a
methylation reaction of the uracil base.
The third base in the class of pyrimidines is cytosine (A3). It forms with guanine a
pair with three hydrogen bonds. The methylation reaction of cytosine to 5-methylcytosin is
conducted by an enzyme. However, uracil will be produced in a side reaction.
The adenine (B1) base belongs to the purines. Originally it was called vitamin B4.
Adenine forms several tautomers that can be rapidly interconverted and are often considered
as being equivalent. In DNA adenine binds through two hydrogen bonds with thymine.
The second purine base is guanine that forms pairs with adenine in DNA and with
cytosine in RNA. Guanine occurs as two tautomers, a keto form and an enol form. It binds
with cytosine via three hydrogen bonds. The carbon atom 6 in Figure 2.1.4 (left) of guanine
acts as the hydrogen acceptor, while the nitrogen at position 1 and the carbon at position 2
act as hydrogen donors. In comparison with guanine the hydrogen donor of cytosine is the
amino group and the hydrogen acceptors are the carbon atom 2 and the nitrogen 3 (right).
15
Materials and Methods
O
NH2
N3
1
1
1
1
4
HN1
1
1
1
1
5
1
1
1
1
6
N
1
1
1
1
1
1 1
9
4
1 1
1
O
1
1
1
2
1
1 1
1
1
1
1
1 1
6
1
1
1
1
2
2
1
1
1
1
1
1
3
N
H 2N
N
H
1
1
1
8
1
5
1
7
N
H
Guanine
Cytosine
Fig 2.1.4 Binding of guanine with cytosine
The one-electron oxidation potential can be further lowered when G is contained as GG or
GGG in DNA sequences.
Base
redox potential
Adenine
1.42 V
Guanine
1.29 V
Thymine
1.7 V
Cytosine
1.6 V
Table 2.1.5 Redox potential of the DNA bases at a pH value of 7.
Electron transfer measurements in solution indicate that the redox potential of
guanine depends also on the substituent at the nitrogen (9). For example the redox potential
of guanosine (Gs) is higher for 0.1 V than that of guanine (G).
The double-stranded structure of DNA provides a simple mechanism for DNA
replication: the DNA double strand is first separated down the middle, and the other half of
each new single strand is recreated by an enzyme by finding the correct base in the mixture
16
Materials and Methods
of the four bases and pairing it with the original strand. This replication mechanism is
sensitive to the nearest environment. For instance, any chemical compound attached at the
DNA double strand may perturb the separation into single strands.
Fig. 2.1.5 Structure of a section of DNA.
2.1.1.2 Binding Mode
Numerous compounds, with varying sizes and diverse structurally groups, bind to the
duplex DNA molecule. The cation-substituted quinone derivative binds to the DNA with
different affinities depending on the structure of the quinones and the sequence of the DNA.
In the case of anticancer antibiotics the binding mode of anthracycline to DNA is
crucial for their efficiency in the cancer therapy. Daunomycin is one of the most known
17
Materials and Methods
anthracycline antibiotics and exhibits a unique DNA binding sequence specificity by
preferential binding to the triplet sequences 5’TACG and 5’TAGC [25].
Compound
Keq/105 (M-1)
n(bp)
δ logK/δlog [M+]
Doxorubicin
29.0±2.0
3.4±0.1
-0.96 ± 0.1
Daunorubicin
6.9±0.2
3.4±0.1
-1.25 ± 0.1
Hydroxyrubicin
1.9±0.1
3.4±0.1
-0.18 ± 0.1
β anomer of doxorubicin
0.15
4.3±0.5
-0.91 ± 0.1
Table 2.1.1.2 Thermodynamic binding parameters for the interaction of doxorubicin,
daunorubicin, hydroxyrubicin and the β anomer of doxorubicin with calf thymus DNA; Keq is the
binding constant and n is the number of base pairs per binding site.
The polyelectrolyte theory quantitatively describes the thermodynamic linkage
between cation and charged ligand binding to DNA. The cations are condensed around the
DNA in order to neutralize the highly negatively charged backbone. The cations are
nonspecifically bound and are mobile. Binding of positively charged ligand to the DNA
results in the release of bound cations, since the positive charge on the ligand may serve to
neutralize the DNA in the place of cation.
The doxorubicin and hydroxorubicin binding data are summarized in Table 2.1.1.2.
Both of these compounds possess the protonated daunosamine group and are still positively
charged at a pH value of 7. The β anomer of doxorubicin is a stereoisomer of the parent
compound with an altered orientation of the C-1’ adriamycine. The table also contains the
values of the slope (δ log K/δ log [M+]) ranging from -0.91 to -1.25 for anthracycline
antibiotic binding to DNA. The exception is hydroxyrubicin with a value of 0.18. For the
binding of a monovalent intercalator to DNA the quantity (δ log K/δ log [M+]) is predicted
to be -0.88 by Record and coauthors [26, 27] and -1.24 by Friedman and Manning [28].
18
Materials and Methods
Friedman and Manning calculated a value of (δ log K/δ log [M+]) = -0.24 for the binding of
an uncharged intercalator to DNA.
The observed free energy of antibiotic binding is defined as ∆Gobs° = - RT ln Kobs.
Values calculated for the binding constant are shown in the Table 2.1.1.3. The observed free
energy may be partitioned into two contributions:
∆Gobs° = ∆G° + ∆Gel
II.1.1
where ∆G° is the nonelectrostatic contribution to ∆Gobs° and ∆Gel is the polyelectrolyte
contribution to ∆Gobs°.[11]
R1
R2
Doxorubicin
OH
Daunorubicin
H
Hydroxyrubicin
OH
β anomer of doxorubicin
OH
Compound
∆Gobs°
∆G°
∆Gel°
NH2
-8.8
-7.8
-1.0
0
NH2
-7.9
-6.9
-1.0
0.9
-7.2
-7.0
-0.2
0.8
-5.7
-4.7
-1.0
3.1
OH
NH2
δ ∆G
Table 2.1.1.3 Free energy of anthracycline antibiotic binding to calf thymus DNA.
2.1.1.3 Structural, Electronic and Spectroscopic Properties of Carminic Acid
Carminic acid (7-α-DGlucopyranosyl-9,10-dihydro-3,5,6,8-tetrahydroxy-1-methyl9,10-dioxo-2-anthracene-carboxylic acid) is an antraquinone dye with a wide application
spectrum in the alimentary and textile industry. For instance it is used as dye in food and
drinks. The molecular structure (Figure1.3.1) consists of an anthraquinone chromophore, a
19
Materials and Methods
sugar residue, and a carboxyl group. Thus the carminic acid has a good solubility in water.
As being a carboxylic acid it is a weak acid so that its dissociation equilibrium crucially
depends on the pH value of the aqueous solution. The differences in electronic structures of
the dissociated and undissociated carminic acid are reflected by their absorption spectra
measured at different pH values between 2 and 9. At a pH value lower than 5, the carminic
acid is fully protonated (undissociated) and exhibits maximum absorption at 490 nm. With
rising pH value the absorption maximum is gradually shifted to 550 nm. The isosbestic
point appears at 515 nm [20].
OH
O
OH
HO
OH
HO
HO
O
OH
O
CH3 O
OH
OH
Fig. 1.3.1: Structure formula of carminic acid in 2D and 3D.
20
Materials and Methods
Cyclovoltammetric studies of the interaction of carminic acid with human serum
albumin (HSA) show that carminic acid has a sensitive linear sweep voltammetric reductive
peak at -0,54 V (vs. SCE). After the addition of HSE in the carminic acid solution the peak
decreased markedly [29].
2.2 Methods
2.2.1 Stationary Optical Spectroscopy
The electronic structure and the optical properties of organic molecules are examined
using stationary and time-resolved techniques of optical spectroscopy. Interactions between
photons and molecular states may lead to an indepth insight into the molecular electron
configuration and excited-state relaxation processes.
The fundamental methods for the study of molecular optical properties are the
stationary techniques of UV/VIS absorption and fluorescence spectroscopy. The wavelength
range of these techniques conventionally reaches from the UV(180 nm) to the near infrared
at 900 nm.
Spectroscopy is the study of matter by investigating light, sound, or particles that is
emitted, absorbed or scattered by the matter under investigation. The energy of a light
quantum is hυ = k · 1,24 · 10-4eV cm. The spectrum of electromagnetic radiation can be
situated between the 50 Hz, the normally AC frequency, and secondary high radiation at
1020 Hz shows in the Fig. 2.2.1.
21
Materials and Methods
UV C
400
UV B
500
600
315 280
370
UV A
Violet
Blue
Green
700
455
492
Yellow
Orange
800
622 597 575
680
Red
2000
0
Crimson
Near
infrared
750
300
Gamma radiation
Longest wave
Microwave
Radio wave
Röntgen radiation
IR
Cosmic radiation
105 104 103 100 10 1 100 10 1 100 10 1 100 10 1 100 10 1 100 10 1 100 1 100 10
m
km
102
10-12
105
10-9
-6
10
10
mm
µm
nm
pm
108
1011
1014
1017
1020
1023
10-6
10-3
100
103
106
109
0
3
6
9
12
-3
10
10
10
λ
10
10
Hz
eV
cm-1
Fig. 2.2.1 Spectrum of electromagnetic radiation: the spectral range of optical spectroscopy
is depicted in enlarged form.
2.2.1.1 UV/VIS Absorption Spectroscopy
The total energy of a molecule in its electronic ground state with exception of the translation
and internal nuclear energy is the sum of the electronic energy (Ee), the vibrational energy
(Ev) and the rotational energy (Er):
Et = Ee + Ev + Er
II.2.2.1
The total energy for the excited state (Et’) is the sum of the electronic vibrational and
rotational energy from the excited state (Ee’, Ev’, Er’). The absorption transition can be
defined as:
22
Materials and Methods
∆Ex = Ex’- Ex ,
II.2.2.2
where x represents t, e, v, or r. The ∆Er is in the order of magnitude of 10 cm-1, typical
values of ∆Ev are about 103 cm-1 and those of ∆Ee are around 3 • 104 cm-1. While excited
electronic states may be investigated using the UV/VIS absorption spectroscopy, the
rotational absorption transitions and vibrational transitions occur in far infrared and near
infrared spectrum, respectively.
The absorption of visible and UV light by organic molecules with a closed shell is
associated with electronic transitions between the electronic singlet ground state (S0) and
higher excited singlet states (Sn, with n=1,2,3,…). Transitions of σ electrons occur via
excitation at photon energies equivalent with wavelengths between 120 and 200 nm. At
wavelengths longer than 200 nm electrons in p-, d-orbitals and π molecular orbitals may be
excited.
To quantitatively characterize absorption transitions of molecular electronic states in
solution samples Lambert-Beer’s law defines the molar extinction coefficient ε. If a
monochromatic light beam with the intensity I0 is incident on a rectangular cell (i.e. cuvette)
with the path length d (in cm), containing the concentration c of the compound under study,
the intensity I of the transmitted light beam decay mono-exponentially over the path length
of the cuvette:
I = I0 e - ε c d ,
II.2.2.3
The prior conditions of Lambert-Beer’s law are that, first, the fraction of the incident
light absorbed by the sample is independent of the total intensity of the light source, second,
the concentration of the absorbing molecules is infinitely small and third, the incident light
beam falls orthogonally of the plan-parallel quartz windows of the sample cuvette. The so-
23
Materials and Methods
called absorption or absorbance, d, of the cuvette, is defined as the decadic logarithm of the
ratio I0/I:
A= lg
I0
= ε ⋅d ⋅c
I
II.2.2.6
where I0 is the intensity of the incident light, I is the intensity of the transmitted light, ε the
molar decadic extinction coefficient (in l cm-1 mol-1), c is the concentration (in mol/l) and d
is the path length of the cuvette (in cm).
The experimental absorption spectrum is plotted absorption, A, versus the
wavelength, λ.
Beam Splitter
Grating
Reference
PMT 2
Detectors
UV VIS lamps
Sample
PMT 1
Fig. 2.2.1.1 Schematic representation of a UV/VIS absorption spectrometer.
In practice, a chromophore giving rise to light absorption by a totally (dipole) allowed
optical transition has values of ε larger than 10 000 l/(mol cm). For the excitation of absorption
transitions in the UV regime a deuterium high pressure arc is employed, whereas a tungstenhalogen lamp provides for the excitation in the visible. The excitation light is dispersed using a
grating monochromator. The intensity of the monochromatic excitation light, I(λ), after the sample
cuvette, and that of the excitation light after the reference cuvette, I0(λ), are detected using exactly
24
Materials and Methods
the same type of photomultiplier tube (PMT1 and PMT2). The absorption spectrum for the
wavelength λ is obtained by subtracting the intensities of the transmitted light beams (A(λ1) = lg
I0(λ1) – lg I(λ1)).
2.2.1.2 Fluorescence Spectroscopy
Fluorescence is a dipole-allowed spontaneous emission of light that occurs from an
electronically excited singlet state. In the first excited singlet state the electron in the lowest
unoccupied molecular orbital (LUMO) is of opposite spin orientation relatively to the other
electron in the highest occupied molecular orbital (HOMO). Consequently, the return to the
ground state is spin allowed and occurs rapidly by emission of a photon. The fluorescence
lifetime of organic compounds in solution ranges from ps to hundreds of ns. Photo-physical
processes following an optical absorption transition of molecular systems are described in
terms of the Jabłoński term scheme (Figure 2.2.1.2.1). The ground state, the first and
second singlet states, are denoted by S0, S1 and S2, respectively, whereas T1 is the first
excited triplet state and the sublevels of each electronic state arise from the fundamental and
the overtones of one representative vibration with the vibrational quantum number 0, 1 and
2. The electronic absorption transition between the ground states S0 and the S1 or S2 occur
within 10-15 s.
S2
Internal
conversion
hυA hυA
Fluorescence
S0
Absorption
S1
Intersystem
Crossing
Internal
conversion
hυF
T1
Phosphorescence
hυP
2
1
0
Fig. 2.2.1.2.1 Jabłoński term scheme.
25
Materials and Methods
In general, the electronic excitation of molecules involves the occupation of higher
vibrational levels of the S1 or S2 states. With a few rare exceptions, molecules in condensed
phases rapidly relax to the vibration less first excited electronic state (S1). This nonradiative
relaxation process is the so-called vibrational relaxation that occurs in 10-12 seconds or less.
Nonradiative deactivation of higher excited singlet states to the S1 state takes place by
internal conversion with time constants between ps and ns that are given by the energy gap
between the involved states (Fermi’s Golden rule). Since the lifetime of the excited S1 state
is about 10-8 s, vibrational relaxation has to occur prior to fluorescence, internal conversion
and intersystem crossing. The radiationless deactivation processes of the S1 state as being
internal conversion and intersystem crossing occur in competition with the fluorescence.
While the internal conversion promotes the electronically excited molecules to their ground
state by dissipation of vibrational energy, the intersystem crossing occurs via spin flip as an
isoenergetic transition to the vibrational manifold of the triplet system. Its probability is
ruled by spin-orbit coupling and thereupon, by the nuclear configuration of the molecule.
The time constant of the intersystem crossing may vary between ps and ms.
Fluorescence relaxation to the ground state occurs to a higher exited vibrational
ground state. Deactivation of the triplet state (T1) takes place either by phosphorescence or
by internal conversion. Since both processes, the radiative and the radiationless one, are spin
forbidden, their time constants may reach values up to several seconds. In general the
fluorescence spectrum is red shifted in comparison to the absorption spectrum. This
phenomenon is called Stokes shift and may be explained by the Strickler-Berg relationship
[30].
Figure 2.2.1.2.2 shows a setup for measurements of fluorescence spectra. A
fluorescence spectrometer consists of an excitation part and a detection part. The excitation
light may be either generated by a xenon high-pressure arc, or by a laser. For the xenon
high-pressure arc the excitation wavelength is selected with a grating monochromator (i.e.
excitation grating). A beam splitter reflects a part of the excitation light to a photon counter
that monitors light intensity instabilities and enables therefore their compensation for the
26
Materials and Methods
final spectrum. The emitted fluorescence light is isotropic and has to be collected by a
converging lens. An ideal detection configuration is the collection of fluorescence emission
under an angle smaller than 90° relatively to the excitation beam. For anisotropy
measurements two polarizers are used, one in front of the sample and the other behind the
sample.
Excitation
grating
Emission
grating
Slit 1
Source Counter
Slit 4
PMT 2
Slit 2
Slit 3
PMT 1
Fig. 2.2.1.2.2 Setup of a fluorescence spectrometer.
2.2.1.3 Measurement of Fluorescence Quantum Yield and fluorescence lifetime
A quantitative measure for the fluorescence probability of an excited molecule is
acquainted by the determination of the fluorescence lifetime and the fluorescence quantum
yield. The fluorescence quantum yield is given by the number of emitted photons relative to
the number of the absorbed photons. In the case of compounds with a large fluorescence
quantum yield, approaching unity, the displayed emission is very intensive.
27
Materials and Methods
The fluorescence quantum yield is the ratio of the number of photons emitted from
the compound under study and the number of photons absorbed by the substance. The
processes governed the Γ and knr with both depopulating the excited state. The fraction of
fluorophores which decays by emission, and hence the quantum yield is given by:
Q=
Γ
Γ + k nr
II.1.3
where Г is the radiative decay rate.
The quantum yield can be close to one, if the radiationless decay rate is much smaller
than the rate of Stock’s losses. For simplicity, all nonradiative decay processes of the first
excited singlet state (S1) are accounted for with the rate constant knr. The lifetime of the S1
state is defined by the mean time the molecule spends in the excited state prior to their
return into the electronic ground state. The fluorescence lifetime is given by:
τ=
1
Γ + k nr
II.1.4
The fluorescence emission is a random (spontaneous) process: a few molecules emit
their photons at precisely t = τ. More than 60% of the fluorescing molecules decay earlier
than t = τ, while the other 40% will emit at t > τ.
I (t ) = I 0 e − t /τ
II.1.5
The natural radiative lifetime or intrinsic lifetime and is given by:
τn =
28
1
Γ
II.1.6
Materials and Methods
The natural radiative lifetime can be calculated from the absorption spectrum (ε) and the
fluorescence spectrum (F) of the fluorophore. The radiative decay rate Γ can be calculated
using:
Γ = 2.88 ⋅ 10
−19
n2
2.88 ⋅ 10
−19
∫ F (ν )dν
∫ F (ν )dν /ν
n2 <ν
−3
> −1
3
∫
∫
ε (ν )
dν =
ν
ε (ν )dν
ν
II.1.7
F(ν ) is the fluorescence spectrum as function of the wavenumber (in cm-1), ε(ν ) is
the absorption spectrum, and n is the refractive index of the medium. The integrals are
calculated for the S0 – S1 absorption transition and by integrating over the total fluorescence
spectrum, respectively.
The natural lifetime can be obtained from the measured fluorescence lifetime (τ) and
quantum yield:
τn =
τ
Q
II.1.8
This equation results from Eq. II.1.3, II.1.4 and II.1.6.. Many biochemical
fluorophores do not behave as predicted for aromatic compounds. There is often poor
agreement between the lifetime calculated using Eq. II.1.7. and II.1.8. that may be explained
by protein residues acting as fluorescence quenchers.
Scintillators are generally used because of their high fluorescence quantum yield.
High fluorescence quantum yields are a result of large Γ values. Hence, their fluorescence
lifetimes are generally short, around 1 nanoseconds. The quantum yield of phosphorescence
in condensed matter is extremely small at room temperature. The triplet to singlet transition
is spin forbidden, and the rate of spontaneous emission is about 103 s-1 or smaller.
29
Materials and Methods
Fluorescence anisotropy measurements are commonly used for biochemical studies
of macromolecules (i.e. proteins, DNA). This type of measurement technique provides
information of the size and shape of the molecules or the rigidity of various molecular
environments.
Dipole-allowed optical transitions are associated with the absorption or emission of
light the electric field vector is aligned in parallel to the transition moment. The transition
moment has a definite orientation with regard to the molecular axes. The random orientation
of fluorescent molecules in solution generates an isotropic environment. Upon excitation
with polarized light, one selectively excites those fluorescent molecules whose absorption
transition dipole momentum is parallel to the electric field vector of the excitation light.
This photo-selective excitation results in a partially oriented population of fluorophores and
in a partially polarized fluorescence emission. Emission also occurs with the light polarized
along the fixed axis in the fluorophore. The relative angle between these moments
determines the maximum measured anisotropy. The fluorescence anisotropy (r) and
polarization (P) can be calculated with help of the equations:
r=
I II − I ⊥
I II + 2 I ⊥
P=
I II − I ⊥
I II + I ⊥
II.1.9
II.1.10
where I II and I ⊥ are the fluorescence intensities of the vertically (||) and horizontally (┴)
polarized emission, when the sample is excited with vertically polarized light.
Decrease of fluorescence anisotropy can be caused by several phenomena. The most
common case is rotational diffusion. This type of diffusion occurs during the lifetime of the
excited state and displaces the emission dipole momentum of the fluorophore. Measurement
30
Materials and Methods
of this parameter provides information about the relative angular displacement of the
fluorophore between the time of absorption and the time of emission. The rotational
relaxation time of most fluorophores in solution ranges between 50 and 100 ps. This implies
that the molecule may rotate several times during its lifetime of the excited state (1-10 ns).
2.2.2 Time-Resolved Optical Spectroscopy
2.2.2.1 Femtosecond Transient Absorption Spectroscopy
In order to investigate interaction between solute and solvent, high time resolution is
an essential aspect. One of the spectroscopic experimental techniques used in this thesis is
the femtosecond-resolved transient absorption pump-probe spectroscopy. In such an
experiment a strong femtosecond (fs) pump pulse is used to initiate a photoreaction, and the
reaction dynamics is followed by recording the absorbance of an week monitoring pulse, as
a function of the time delay between the pump pulse and the monitoring pulse. In order to
eliminate the possible noises introduced by possible intensity fluctuation of the laser the
monitoring pulse is divided in a probe and a reference pulse. The probe pulse in spatially
overlapped with the pump pulse in the sample, while the reference pulse passes through a
region of the sample which is unaffected by pump pulse. The change of the optical density
(∆OD) is obtained as a negative logarithm of the ratio of the intensity of the probe and
reference pulse as follows:
∆OD = − log
I probe
I ref
II.2.1
Liquid phase absorption and emission spectra generally contain broad bands, which
makes it essential to examine the temporal behavior of the photoinduced reaction over a
31
Materials and Methods
wide range of probe wavelengths. In the experiment the probe pulse has a white light
continuum (wlc) spectrum. The wlc is generated by the χ(3) process of self-phase modulation
of an intense femtosecond laser pulse propagating through a dense but transparent medium
[31, 32], here a fused silica disc. Under proper experimental condition a wlc can be made to
extend from the near ultraviolet to near infrared. After the sample, the wlc is dispersed by a
spectrograph and detected by a charged coupled device (CCD) camera which allows for
simultaneous measurement of the intensities of the probe and reference pulse for a wide
range of wavelengths present in the wlc. In this sense the pump-probe signal, ∆OD (λ, ∆t),
is obtained as a function of the probe wavelength and the delay time ∆t between the pump
and the probe pulses.
In a heterodyne detection scheme, the electric field of the strong pump pulse,
Epump(z,t), and the electric field of the weaker probe pulse, Eprobe(z,t), generate a non-linear
polarization, PNL, in the sample, where z is the sample thickness. The non-linear
polarization acts as a source term in the Maxwell’s equation to generate a signal field, ES.
The probe field is called the local oscillator field [33], and in these terms the total measured
intensity is proportional to the square of the total field. In order to obtain the change in
optical density, a reference field is needed, Eref , which is the probe field unaffected by the
pump, Eprobe(0,t). The total detected intensity can then be written as [33]:
I
probe
(z,t) ∝| E
I ref ( z , t ) ∝ | E
I
probe
probe
probe
( z , t ) |2 = | E
probe
( 0 , t ) + E S ( z , t ) |2
( 0 , t ) | 2 = | E ref ( 0 , t ) | 2
II.2.2
( z , t ) − I ref ∝ − 2 Im( E * probe ( 0 , t ) E S ( z , t ))
The signal field ES is usually weak so that the detected intensity is approximately
proportional to the sum of the cross term, and the reference term, |ES (z,t)|2 , in Eq. II.2.2 is
neglected. In order to normalize the measured intensity the probe is divided by the reference
32
Materials and Methods
intensity, and to achieve the change in the optical density, the negative logarithm is taken of
this ratio as:
 I probe
∆OD = − log
 I
 ref

I
− I ref
 ≈ − 1 ⋅ probe
 ln(10)
I ref

II.2.3
The approximation in Eq. II.2.3 is valid only under the condition, when (Iprobe – Iref)/ Iref <<1
[34].
SN+M
Probe
2)
SN
S1
Pump
Probe
1)
Probe
3)
S0
Fig 2.2.2.1.1 Energy scheme of the electronic states involved in a pump-probe
experiment; excited states relaxation dynamics detected by absorption changes of the probe
pulse are: 1) bleaching; 2) excited state absorption; 3) stimulated emission.
In a pump-probe experiment there will be three contributions to the transient
absorption pump-probe signal (see Figure 2.2.2.1.1). First, the strong pump pulse excites
the molecule from the ground state S0 to a higher lying state SN. The probe pulse may also
excite the molecules from the ground state to the same excited state as the pump pulse,
which will result in a bleaching of the ground state, 1) in Figure 2.2.2.1.1. This may be
observed as an increase in the detected probe intensity since less probe photons is absorbed
relative the reference. An increase in probe pulse intensity is the same as negative ∆OD. For
the situation of the probe pulse (see 2) in Figure 2.2.2.1.1) the molecules in the excited state
undergo so-called transient absorption transitions. This is seen as an increase in ∆OD.
33
Materials and Methods
Finally there may be stimulated emission as the case 3) in the same figure and this is will be
detected as a decrease in the ∆OD.
However most of the samples are transparent in the pump wavelength region, both to
one- and two-photon absorption. In this case there is no excited state in the sample, and
consequently neither bleaching, excited-state absorption nor stimulate emission when the
probe pulse is present in the signal. But even in those cases there will be detectable signals,
when the probe is dispersed after the sample, due to the change in non-linear polarization of
the sample.
Positive values of ∆OD may be ascribed to either bleaching of the ground-state (S0)
population density or stimulated emission, while negative values of ∆OD are generally
assigned to absorption transitions from the first excited singlet state, S1, to higher singlet
states (i.e. transient absorption).
Both phenomena, S0-state bleaching and transient
absorption from the S1 state are induced by the pump pulse. Both, recovery of the S0-state
population density and decay of the transient absorption exhibit mono-exponential behavior
with a time constant that is the lifetime of the S1 state, if the latter is determined by pure
monomolecular photo-physical processes.
The experimental setup for the fs transient absorption pump-probe spectroscopy is
depicted in Figure 2.2.2.1.1. The central wavelength of the pump pulse is adjusted to the
energy gap between the electronic ground state (S0) and the first excited singlet state (S1) of
the molecules under study, while a white-light continuum is used for the probe pulse in
order to enable a large variety of absorption transitions from the S1 state to higher lying Sn
states. A delay stage is used to vary the path length of the probe pulse in small steps in order
to obtain a variable time delay between the pump and the probe pulses. The delay stage may
vary the path of the probe pulse down to steps of 1µm, corresponding to 3.3 femtoseconds.
The maximum delay time what can be applied between pump and probe pulses is about 1
nanosecond. A part of 5% from the 800 nm laser beam (fundamental) is reflected by a beam
splitter and sent to the diagnostics part where the time and frequency profile of the laser
pulse can be measured by an autocorrelator and a spectrometer. The 95% transmitted laser
34
Materials and Methods
beam is raised to a convenient level by a mirror periscope and a λ/2 plate rotates the
polarization of the beam from horizontal to vertical. This gives s-polarized light with respect
to reflection in the horizontal plane which is what the dielectric mirrors, frequently used in
set up, are optimized for. The two time magnifier telescope reduces the beam diameter to a
half and consists of a concave mirror focusing the beam and a convex mirror restoring the
beam collimation. The chopper triggers the detector and synchronizes it with the chopped
light. A second beam splitter divides the beam in a pump beam (90%) and a probe beam
(10%). The pump pulses are frequency doubled in a non linear crystal (BBO). The probe
pulses can be shifted in time by the delay stage. The wlc probe beam is split in a reference
beam and a probe beam. The probe pulses pass the sample cell to spatially overlap there
with the pump pulses, whereas the reference beam crosses the sample unaffected by the
pump beam.
35
Materials and Methods
95% 800nm
Ti:Sapphire laser
DM
λ/2
5%
autocorrelator
X2
telescope
spectrograph
CCD
pulse characterization
spectrograph CCD
DS
VA
λ/2
L
WLC plate
λ/2
10%
pump
P
periscope
BS
90%
chopper
SHG
400nm
DM
BC
FS
PM
ND
GG
BG
┴
||
probe
reference
sample
BD
Fig. 2.2.2.1.1 Schematic representation of the fs transient absorption spectroscopy
experiment; BS: Beam Splitter; FM: Flip Mirror; DS: delay stage; VA: variable attenuator; P:
polarizer; L: lens; WLC plate: rotating fused silica plate for white light continuum generation; PM:
parabolic mirror; SHG: BBO crystal for second harmonic generation; BC: Berek compensator; BD :
beam dump; BG: BG38 filter; GG: GG420 filter; ND: neutral density filter.
The sample solution is through a flow cell with a thickness of 2 millimeters. A
spectrograph disperses the probe and the reference beams, which are detected by a CCD
camera. A computer controls the experiment and plots the ∆OD as a function of delay time
and frequency.
36
Materials and Methods
The transient absorption spectroscopy is based on a Ti: Sapphire laser system that
consists of a mode-locked oscillator (Coherent, Mira) pumped by an argon ion laser
(Coherent, Innova 310) and a Ti: Sapphire regenerative amplifier (Alpha 1000 US, BM
Industries). This femtosecond laser system produces 40 fs output pulses at 800 nm with a
repetition rate of 1 kHz and energies of about 700 µJ. The pump pulses at 400 nm are
obtained by frequency doubling the output pulses in a 0.2 mm BBO crystal, whereas the wlc
probe pulses between 400 and 800 nm are generated by focussing a small fraction of the
fundamental pulses (5%) into a thin rotating fused silica disc. By applying the pump pulses
at 400 nm to the sample solution, carminic acid is non-resonantly excited into the S1 state,
while the excited-state decay dynamics are probed with the temporally delayed white-light
continuum pulses.
Wlc probe pulses are generated by self phase modulation (SPM) occurring as a
parametric χ(3) process in a Kerr medium. SPM involves the modulation of the refractive
index of the medium in response to ultrashort laser pulses. The modulated refraction index
modulates the pulse phases and thereupon the frequencies of the electromagnetic waves
contained in the wave packet. Self phase modulation is based on the Kerr effect accounting
for the nonlinear dependence of the refractive index n(t) on the intensity I(t) of the pulse
propagating along the space coordinate:
n(t ) = n0 + n2 ⋅ | E (t ) | 2 ,
II.2.1
where n0 denotes the linear refractive index, n2 the nonlinear refractive index and E(t) is the
electric field of the laser beam.
Combining the expression for nonlinear refracting index (II.2.1) with a linear wave
equation [35]:
∇2E −
n2 ∂2 E
c 2 ∂t 2
II.2.2
derived from the Maxwell’s equation, where c is the velocity of light and n is the refractive
index, gives rise to the nonlinear term on the right side of the nonlinear wave equation:
37
Materials and Methods
∇2E −
n02 ∂ 2 E
∂ 2 PNL
=
µ
0
c 2 ∂t 2
∂t 2
,
II.2.3
with:
µ 0 PNL = µ 0 P ( 3) = µ 0 ε 0 χ (3) EEE ≈
2n 0 n 2
| E |2 E ,
2
c
II.2.4
where the n22|E| term has been neglected and n2 and χ(3) are assumed to be instantaneous.
Expressing E as a plane wave with the angular frequency ω0 and the wave vector k
propagating in the z- direction with E = A(t)ei(kz- ω0t) + A*(t)e-i(kz- ω0t) and inserting this into
Eq. II.2.3 give the following equation for the amplitude A(t)[36]:
ω
∂A 1 ∂A
+
= i 0 n2 | A |2 A ,
∂z v g ∂t
2c
II.2.5
where vg is the group velocity. Both, the group velocity dispersion and linear absorption,
are neglected.
Solving Eq. II.2.5 with respect to the phase reveals a time dependence of the phase
Φ(t). The instantaneous frequency is the time derivation of the phase and can be written as
[37]:
ω (t ) =
ω ∂n(t )
∂
φ (t ) = ω 0 − 0
z
∂t
c ∂t
II.2.6
The frequency shift is:
δω (t ) = ω − ω 0 =
ω 0 n 2 ∂I (t )
,
z
∂t
c 2 ε 0 n0
II.2.7
where I(t) is the intensity of the pulse and is written as:
I (t ) = 2ε 0 cn | A(t ) | 2
38
II.2.8
Materials and Methods
Fig. 2.2.2.1.4 a) Intensity as a function of time for a Gaussian laser pulse b) Time
dependence of the frequency for a positive nonlinear index of refraction, n2.
The wlc probe pulse in the propagation direction is a repetition of monotone pulses with
continually increasing central wavelength. These spectral pulse components of the wlc pulse
have the same temporal pulse width as the wlc generating pulse which is the pump pulse.
λ
Fig. 2.2.2.1.5. Representation of the wlc probe pulse as a composition of the temporally
shifted, different spectral sub-pulses having the same pulse width as the pump pulse.
39
Materials and Methods
The fundamental nonlinear optical phenomena arising from strong light-matter
interactions are parametric three-wave mixing and four-wave mixing that give rise to second
harmonic generation (SHG), sum frequency generation (SFG),
difference frequency
generation (DFG) and third harmonic generation (THG). While SHG, DFG and SFG are
parametric χ(2) processes and are derived from the second order term of macroscopic
polarization P(2)=ε0 χ(2) E2, the THG is a χ(3) and is due to P(3)=ε0 χ(3) E3.
SHG, SFG and DFG directly result from P(2)=ε0 χ(2) E2 for electromagnetic waves the
with angular frequencies ω1 and ω2:
E = E1 cos(ω1t − k1 z ) + E 2 cos(ω 2 t − k 2 z )
II.2.9
That are incident on a non-centrosymmetric medium with a non zero χ(2). E1 and E2 are the
amplitudes of the two beams and k1 and k2 are the wave vectors in the direction z with ki
=2πn(λi)/λi. The second order polarization becomes [39]:
P ( 2) = ε 0 χ ( 2) E 2 ( z = 0) = ε 0 χ ( 2) ( E12 cos 2 ω1t + E22 cos2 ω 2 t + 2 E1 E2 cosω1t cosω 2 t ) =
II.2.10
1
1
1
= ε 0 χ ( 2)  ( E12 + E22 ) + E12 cos2 2ω1t + E22 cos 2 2ω 2 t + E1 E2 [cos(ω1 + ω 2 )t + cos(ω1 − ω 2 )t ]
2
2
2
P(2) contains as components a static electric field with ½ε0χ(2)(E12+E22), the SHG of both
fundamental waves with ω1 and ω2 (i.e. 2ω1, 2ω2), the SFG with ω1+ω2 and the DFG with
ω1-ω2. For efficient SFG and DFG the incoming and generated waves must fulfil the phase
matching condition:
k (ω1 ± ω 2 ) = k1 ± k 2
II.2.11
Because of ki =2πn(λi)/λi the phase matching condition can be fulfilled by changing the
refractive index n(λ) of one wave with respect to the others. In practice, this can be provided
40
Materials and Methods
by uniaxial birefringent crystals that have two different refractive index n0 and ne(θ) for the
ordinary and the extraordinary beams, where θ is the angle between optical axis and the
wave vector. The ordinary beam is polarized perpendicular to and the extraordinary beam
parallel to the plane defined by the propagation direction and the optical axis.
2.2.2.2 Femtosecond Fluorescence Up-Conversion
The fluorescence spectrum, quantum yield and lifetime of a chemical compound
depend on its environment and of the solute-solvent interaction. In particular, the
fluorescence lifetime may reflect the kinetics of a chemical reaction. Time-resolved
fluorescence spectroscopy measurements give detailed information about intra- and
intermolecular photophysical and photochemical processes. In the last 30 years, the laser
technology and especially, the ultrafast laser techniques enable precise measurements of
molecular excited-state relaxation dynamics in the nano-, pico- and femtosecond time range.
When analyzing the rise and the decay, often multiexponential of their fluorescence
intensity as a function of time, the pre-exponential factor and the fluorescence time constant
have to be determined accurately after a careful deconvolution using a computer-based
algorithm. Several techniques are used: phase modulation spectroscopy with a modulated
laser pump source on a time scale comparable to the decay times of interest, time-correlated
single photon counting under picoseconds laser excitation, ultra-fast streak cameras
designed for the conversion of time dependent light emission into distance dependent
electron impact on a phosphor screen. Because a sufficient resolution is not possible with
this technique for measurements of fast decay in the range 50 fs – 100 ps, the fluorescence
up-conversion in nonlinear crystals is the best choice.
Although femtosecond mode-locked Ti: Sapphire lasers which offer tunability and
pulse duration similar to those of dye laser but higher stability and ease of use are also
available, ultrafast kinetics in the range of femtosecond cannot be resolved in the
41
Materials and Methods
nanosecond or picoseconds technique since the limiting factor of their response function is
the detector. The time-gated up-conversion technique is by far the best choice for the
femtosecond time range.
The phenomenon of fluorescence up-conversion is based on the parametric threewave mixing process of SFG. The fluorescence emitted after an fs laser pulse excitation is
mixed with the other half of the laser pulse in a nonlinear optical crystal to generate sum
frequency radiation. The crystal acts as an ultra-fast optical “light gate” that is “open” when
the pump pulse is “on”. This method is particularly suited for laser pulses of moderate peak
powers but very high repetition rate. Common nonlinear crystals are cut from β-barium
borate (BBO), potassium dihydrogen phosphate (KDP), lithium iodide (LIO) or urea. In
general BBO crystals are used because of their optical transparency in the UV range.
The fluorescence light with the energy EF and the pump laser beam with the
energy EP are focused into the BBO crystal under phase matching conditions so that the sum
frequency light with the energy ES is produced:
E S = hν S
II.3.1
EP = hν P
II.3.2
E F = hν F
II.3.3
ν S = ν F +ν P
II.3.4
r
r
r
kS = kF + kP ,
II.3.5
The phase matching conditions are:
r
r
r
where kS , k F and k P are the wave vectors of the sum frequency, fluorescence and pump
pulses, respectively. In the case of collinear configuration the phase matching Eq. (II.3.5)
becomes:
nF
λF
42
+
nP
λP
=
nS
λS
II.3.6
Materials and Methods
where n is the crystal refractive index at the wavelength λ.
In uniaxial crystals with nx = ny = n0 (ordinary index) and nz =ne (extraordinary
index), the refractive indices vary with the wavelength according to Sell Meier’s equation:
n2 = A
B
+ Dλ2
2
C −λ
II.3.7
For type I crystals (O + O → E), the fluorescence and the pump are polarized parallel to
each others. In type II crystals (E + O → E) or (O + E →E), the pump and fluorescence
beams are polarized perpendicularly. For type I crystals the phase matching angle can be
calculated using the following equation:
1
1
− 2 )
n ( θ m ) n0 ,S
sin 2 θ m =
1
1
− 2
2
ne ,S n0 ,S
(
2
S
II.3.8
where nS(θm) satisfies the following equation:
nS (θ m ) = nO ,F
λS
λ
+ nO ,P S
λF
λP
II.3.9
The phase mismatch is given by
1
∆k = k S − k P − k p = (nSϖ S − nFϖ F − nPϖ P ) ,
c
where c is the velocity of light and ωi is the angular frequency ( ϖ i =
II.3.10
2πc
). Since
2 nλ i
fluorescence is polychromatic, isotropic radiation, the fluorescence light has to be collected
and to be focused onto the crystal and the phase matching angle has to be adjusted in
dependence on the wavelength of interest. The acceptance angle increases inversely with the
crystal length.
43
Materials and Methods
Only a narrow band of the fluorescence spectrum is up-converted for an appropriate
phase matching angle θm. The full spectrum is obtained by varying the angular position of
the crystal.
The quantum efficiency for up-conversion varies as the ratio of the power PP of the
pump of the beam to its area A in the BBO crystal, the square of both the thickness L and
the effective nonlinear coefficient of the crystal deff according to:
nq =
2π 2 d eff2 L2 PP / A
ce03λF λF n0.F n0 P ( θ m )
,
II.3.11
where ε0 is the permittivity in vacuum. The quantum efficiency, ηq, the damage threshold,
Ithr(GW/cm2), and the cut-off for the up-conversion fluorescence calculated for a pump
pulse wavelength at 800 nm, λf,min(nm) are summarized in Table 2.2.2.2.
Crystal type
ηq
Ithr(GW/cm2)
λf,min(nm)
KDP (ooe)
1.0
30 (ps regime)
344
LiIO3 (ooe)
44
15 (ps regime)
480
BBO (eoe)
4.1
50 (ps regime)
264
BBO (ooe)
6.5
50 (ps regime)
249
urea (eeo)
6
5 (ps regime)
313
Table 2.2.2.2 Relative quantum efficiencies ηq (normalized relative to KDP), damage
thresholds Ithr, and cut-off wavelengths for fluorescence up-conversion with 800 nm pump pulses.
A typical fluorescence up-conversion experiment is schematically described in
Figure 2.2.2.2.1. The femtosecond laser source is a Ti: Sapphire oscillator (Coherent
MIRA) pumped by continuous wave Coherent VERDI V10 laser with a 10 W output power.
Typical performances of the Ti: Sapphire oscillator is 1.8 W output power in the modelocked regime at 800 nm and 76 MHz repetition rate. The autocorrelation trace shows pulse
duration of 125 femtoseconds. The second harmonic at 400 nm is generated in a 1 mm thick
44
Materials and Methods
BBO crystal and separated from the fundamental 800 nm beam by a dichroic beam splitter.
After passage though a delay line, the residual fundamental is focused by a 100 mm lens
onto a 0.2 mm-thick type II BBO crystal producing the SFG signal. The 400 nm pulse beam
used for excitation is focused by a 1.5-inch concave mirror into the 1 mm thick flow quartz
cuvette cell what contains the solution sample. The fluorescence is collected by another
concave mirror with a 4 inch off axis. After the mirror the fluorescence passes through a 1
mm optical filter (Schott GG420) and is focused into the up-conversion crystal by a similar
concave mirror (4 inch off axis). The up-converted light beam is collected by a 150 mm UV
transparent lens, passes through another filter (Schott UG11), and is focused onto the
entrance slit of a 0.25 µm monochromator. The slit width corresponds to a 10 nm band-pass
filter. The spectrally selected UV light is detected by a photomultiplier connected to a lock–
in photon counter. All the experiments are performed at the magic angle between the
polarization of the excitation and the detection beams. The cross correlation trace between
the fundamental laser beam and the second harmonic give a full width at half maximum
(FWHM) value of 210 femtoseconds, confirmed by recording the Raman line (2900 cm-1 for
the C-H and 3500 cm-1 for the O-H stretching mode, respectively) of pure methanol with the
fluorescence up-conversion apparatus [40].
45
Materials and Methods
Verdi 10 W
800 nm
Ti: Saphire 2W
BBO
1 mm
800 nm
Lock in
CCD
DM
400 nm
HW
PM
Mono.
GG420
θ
BBO 0.2 mm
1 mm flow cell
Fig. 2.2.2.2.1 Femtosecond up-conversion technique apparatus; DM: dichroic mirror; HW: half
wave plate; CCD video camera for the visual superposition of the beams in the BBO crystal; Mono:
monochromator; PM: photomultiplier.
Time resolved fluorescence spectra are usually obtained by recording kinetics at
different wavelengths and using the spectra reconstruction method. An alternative method is
to measure directly time resolved spectra to avoid the too long acquisition of the kinetic
data. In principle the acquisition of one complete spectrum at a particular delay time after
excitation does not suffer from long term variation of the experimental conditions. There are
two points of view to consider. First, different spectral components of the fluorescence
propagate with different group velocities in the various traversed media between the sample
cell and the non linear crystal. The second point is the limited half bandwidth of sum
frequency generation, typically 20 nm at 450 nm in a 0.2 mm thick BBO crystal. The
temporal and spectral response has thus to be corrected for. For the temporal correction,
while scanning the monochromator, the group velocity dispersion (GVD) is calculated as a
function of wavelength and the difference in propagation time is compensated by adjusting
46
Materials and Methods
the delay line. For the spectral correction, the optimal phase matching angle is calculated as
a function of wavelength and the crystal is rotated to the correct angular position using a
step motor and a PC computer. Moreover, the spectra are corrected for the spectral response
of the up-conversion apparatus. This correction is based on the comparison of the
fluorescence spectrum at a long time after excitation (typically 500 ps, when the molecular
system is spectrally relaxed) to the steady-state fluorescence spectrum corrected itself for
the response function of the conventional spectrofluorometer [40].
47
Experimental
3. Experimental
3.1 Materials: Chemicals and Sample Solutions
Carminic acid was purchased from Aldrich (99% purity) and was used without
further purification. Salmon sperm DNA (low molecular weight) was obtained from Fluka
Biochemika. DNA concentrations were determined upon performing UV/VIS absorption
spectroscopy measurements at 260 nm using the extinction coefficient ε(260nm) = 6600 (M
nucleotide)-1 cm-1. Ultrapure water was obtained from a Millipore system. The buffer
designated as BPES [11] consists of Millipore water with 6 mM Na2HPO4, 2 mM NaH2PO4,
1 mM Na2EDTA, and 185 mM NaCl, all from Fluka, and has a pH value of 7.0.
UV/VIS absorption spectroscopy and fluorescence spectroscopy measurements of
carminic acid in BPES, Millipore water or dimethylsulfoxide (DMSO) (Merck, pro
analysis) were performed at carminic acid concentrations between 5 µM and 10 µM,
Sample solutions (carminic acid in BPES) used for the fs resolved spectroscopic
investigation were prepared to attain an optical density of about 0.5 at 400 nm for an optical
path of 0.4 mm or 2 mm. For the determination of the pKa value of carminic acid we have
titrated carminic acid in Millipore water at a concentration of 5 µM or 1 mM with an
aqueous 5µM NaOH solution (e.g. 1 mM NaOH). The pH value was potentiometrically
measured using a pH meter (pH 522, WTW) equipped with a glass electrode (pH Electrode
Blue Line 23pH, Schott). The results are presented in the next chapter.
48
Experimental
3.2 Methods
3.2.1 UV/VIS absorption and fluorescence spectroscopy
UV/VIS absorption spectra of carminic acid in BPES, in Millipore water or in
DMSO were recorded on a Perkin-Elmer Lambda 2 spectrometer. The fluorescence spectra
were taken with a Jobin-Yvon FluoroMax-3 spectrofluorometer in the magic-angle
polarization configuration. All the stationary spectroscopy experiments were performed at
room temperature for carminic acid concentrations between 5 µm and 10 µM, employing a
quartz cell with an optical path length of 10 mm. For the determination of the pKa value of
carminic acid the different pH values were set by titration either with 1 M NaOH or with 1
M HCl. All experiments on DNA were performed using the BPES buffer solution (pH= 7).
3.2.2 Determination of the Fluorescence Quantum Yield
The fluorescence quantum yield Φ of carminic acid in Millipore water was
determined with reference to rhodamine 6G in ethanol (Φ0= 95 % [15]) using the
relationship:
S (1 − 10 − A0 ) ⋅ n 2
Φ = Φ0
2
S 0 (1 − 10 − A ) ⋅ n0
IV.2.1
where S and S0 are the integral intensities of the fluorescence spectra measured in the
magic-angle polarization configuration in the range of 500 - 800 nm, A and A0 are the
absorbances at the excitation wavelength of 480 nm of carminic acid and rhodamine 6G,
respectively, n denotes the refractive index of water and n0 that of ethanol.
49
Experimental
3.2.3 Fluorescence Titration Experiments
The
fluorescence
spectra
were
taken
with
a
Jobin-Yvon
FluoroMax-3
spectrofluorometer in the magic-angle polarization configuration. For the fluorescence
titration experiments an excitation wavelength of λexc = 320 nm with a slit width of 10 nm
was typically used, and the spectrum was recorded from 360 nm to 800 nm with a detection
slit width of 10 nm. The relative fluorescence intensities were determined by integrating
over the whole spectrum.
3.2.4 Femtosecond Transient Absorption Spectroscopy
The setup used for the femtosecond transient absorption spectroscopy [41] is based
on a femtosecond Ti: Sapphire laser system that consists of a mode-locked oscillator
(Coherent, Mira) pumped by an argon ion laser (Coherent, Innova 310) and a Ti: Sapphire
regenerative amplifier (Alpha 1000 US, BM Industries). This femtosecond laser system
produces 40 fs output pulses at 800 nm with a repetition rate of 1 kHz and energies of about
700 µJ. The pump pulses at 400 nm were obtained by frequency doubling the output pulses
in a 0.2 mm BBO crystal, whereas white-light continuum (wlc) pulses between 400 and 800
nm were generated by focussing a small fraction of the fundamental pulses into a thin
rotating fused silica disc. By applying the pump pulses at 400 nm to the sample solution,
carminic acid was non-resonantly excited into the S1 state, while the excited-state decay
dynamics were probed with the temporally delayed wlc pulses.
Transient absorption spectra were obtained by determining the differential absorption
in the sample versus the delay time τ between the pump and the white-light continuum
pulses. To compensate for eventual intensity fluctuations, the wlc was split into two parts,
one overlapping with the pump beam in the sample and thus serving as probe pulse and the
50
Experimental
other one, non-overlapping with the pump beam, serving as a reference. After passage
through the sample, both probe and reference beams were dispersed in a spectrograph (Spex
270 M) and detected using a 1024 x 512 pixels CCD camera (Princeton Instruments). The
readout rate of the camera was 30 Hz achieved by a chopper that modulates both beams.
The transmitted white light intensity of the probe (Tsample) and the reference (Tref)
beams were simultaneously measured on the CCD for delay times τ ≥ 0 ps. This delay time
was varied between 0 and 200 ps by 30 fs steps. The differential absorption is defined by the
equation:
 Tsample (τ ) 
 Tsample (τ < 0) 
absorption = log 
 − log 

 Tref (τ ) 
 Tref (τ < 0) 
IV.3.1
In this equation, Tref(τ) does not imply any direct dependence of the reference beam on the
time delay τ, only that it was the reference signal recorded simultaneously with the probe
signal that was used thus compensating for any intensity fluctuations. Thus defined, the
transient absorption is negative and the bleaching of the ground-state population density is
positive. The sample solutions were pumped through a flow cell with a thickness of 0.4 mm
or 2 mm. The 0.4 mm-thick flow cell was used for carminic acid concentrations between 1
and 6×10-4 M, whereas sample solutions at a concentration above 1×10-3 M were
investigated in a 2 mm cell. The FWHM of the instrument response function is about 60 fs
in the wlc spectral range of 400 – 800 nm.
3.2.5 Femtosecond Fluorescence Up-Conversion Technique
The femtoseconds resolved fluorescence up-conversion experiments were performed
using a setup that has already been described in detail in the chapter 2.2.2.2. A femtosecond
laser system, consisting of a Ti: Sapphire laser (Coherent MIRA 900), pumped by a 10 W
51
Experimental
CW solid state laser (Coherent VERDI V10), produces 125 fs pulses at 800 nm with a
repetition rate of 76 MHz and a 1.8 W average output power. The second harmonic (SH) is
generated in a 0.5 mm thick BBO crystal and is separated from the fundamental beam by a
dichroic beam splitter. The SH is used as pump pulse for excitation of the fluorescence of
the sample, whereas the fundamental pulse serves as gating pulse for the sum frequency
(“up-conversion”) generation. The fluorescence light is focused into the up-conversion
crystal, and the up-converted light is spectrally analyzed using a 220 mm focal length
double grating monochromator (SPEX 1680). The detection and amplification of the
dispersed up-conversion light is achieved by using a photomultiplier (Hamamatsu R1527P)
in combination with a photon counter (Stanford SR400). The fluorescence up-conversion
experiments were performed at 0° and 90° angle between the polarization axes of excitation
and detection. The cross correlation trace between the laser fundamental (800 nm) and the
SH (400 nm) gives a FWHM value of 175 fs for the apparatus function.
The fluorescence up-conversion spectra were corrected first by subtracting the
background from the fluorescence up-conversion spectra. The background was recorded by
positioning the delay at “negative” time, so that the gating pulse arrives to the upconversion crystal well before the fluorescence signal. The current spectral correction curve,
R(λ), was determined experimentally by comparing the normalized fluorescence upconversion spectrum at long times (100 ps: I100ps(λ)) with the normalized steady state
spectrum, ISS(λ) recorded for the same sample solution, so that the relationship
R(λ)=ISS(λ)/I100ps(λ)
2.5.1
is obtained. All fluorescence up-conversion spectra were subsequently multiplied with R(λ).
52
Experimental
3.2.6 Computations
Semi-empirical quantum chemical computations of the molecular geometries,
orbitals and electronic energy states of the four carminic acid tautomers in the fully
protonated state (CAH) as well as in the singly deprotonated state (CA-) were performed
using the HYPERCHEM (release 7.1) package program. The geometry optimizations of all
structures were achieved employing the ZINDO/S self-consistent fields molecular orbital
(SCF MO) method at the restricted Hartree-Fock (RHF) level [42]. For the optimizations we
have applied the steepest-descent method in combination with the consecutively performed
conjugate gradient methods, Fletcher-Reeves and Polak-Ribiere (convergence limit of
4.18×10-4 kJ/mol and RMS gradient of 4.18 ×107 kJ/m mol). RHF open-shell and closedshell calculations of the electronic energy states and of the electronic transitions were
performed for single configuration interaction (CI) involving 160 total orbitals.
53
Results and Discussion
4. Results and Discussion
4.1. Stationary Optical Spectroscopy
Carminic acid possesses a carboxylic acid function and is thus expected to dissociate
in water:
OH
O
HO
HO
HO
OH
OH
O
HO
OH
OH
O
OH
OH
OH
O
CH3 O
HO
HO
O
O
OH
OH
O
-
+
+
H
CH3 O
OH
Scheme 1: Dissociation reaction of carminic acid.
The protonated carminic acid (CAH) and its anion (CA-) differ in their electronic
density distribution and thereupon, exhibit distinct dipole momentum and solvation shell
structure. Their different electronic configurations are reflected in the pH dependence of the
UV/VIS absorption spectrum of carminic acid dissolved in water (see Figure 4.1.1). As it is
obvious from Figure 4.1.1 there are in main two different dissociation equilibria, one exists
at a pH value of 5 and is shifted to CAH, while the other occurs at pH values above 5.
54
Results and Discussion
pH= 3
pH= 4
pH= 5
pH= 6
pH= 7
pH= 8
pH= 9
absorption
0.4
0.3
0.2
0.1
0.0
30000
25000
20000
15000
-1
wavenumber / cm
Fig. 4.1.1: pH dependence of the spectral features of carminic acid in water measured by
UV/VIS absorption spectroscopy.
For carminic acid in Millipore water a pKa value of 5.55 was determined upon
titration with either an aqueous NaOH solution or an aqueous HCl solution. The pKa value
of 5.55 implies that at pH = 7 the undissociated carminic acid, CAH, and its anion, CA-,
coexist at nearly the same concentration. Photo-exciting carminic acid in BPES (at pH= 7)
with 340 nm light generates two fluorescence emissions, one peaks at 15100 cm-1 (570 nm:
orange emission) and the other at 22700 cm-1 (470 nm: blue emission) ( see Figure 4.1.2).
55
normalized intensity
Results and Discussion
1.0
0.8
0.6
0.4
0.2
0.0
12000
15000
18000
21000
24000
27000
-1
wavenumber / cm
Fig. 4.1.2: Dual fluorescence of carminic acid with a blue emission peak at 470 nm (22700
-1
cm ) and an orange emission peak at 570 nm (15100 cm-1)
The intensity of the orange fluorescence is half of the blue fluorescence intensity.
The observation of the dual fluorescence is taken as strong indication to an excited state
intramolecular proton transfer (ESIPT) [13, 43-46]. In general, the excited-state (S1)
tautomer accessed by ESIPT has an energy value comparable to or slightly less than that of
the normal form, while the ground state (S0) of the tautomer is of considerably higher
energy. As a consequence, the fluorescence of the tautomer appears spectrally red shifted
with respect of that of the normal form. That is why we ascribe the blue fluorescence to the
normal form, whereas we expect the tautomer to emit the orange fluorescence. In contrast,
carminic acid that was dissolved in the aprotic solvent DMSO and was analogously photoexcited at 340 nm, exhibited the orange fluorescence only (see Figure 4.1.3).
56
normalized intensity
Results and Discussion
1.0
0.8
0.6
0.4
0.2
0.0
12000
15000
18000
21000
24000
27000
-1
wavenumber / cm
Fig. 4.1.3: Orange fluorescence of the tautomer at 15100 cm-1.
The lack of the blue fluorescence implies that the ESIPT occurs in this aprotic
solvent with a rather high probability. This is explained by the fact that DMSO does not
form external H bonds which stabilize the normal form of carminic acid so that ESIPT
becomes improbably in water. Another stabilization effect may arise from efficient
interactions of DNA with carminic acid. Photo-excited BPES sample solutions containing
the same concentrations of carminic acid and DNA, each at 5 µM, were observed to emit
the blue fluorescence with a significantly higher intensity than that of the orange
fluorescence (see Figure 4.1.4). For carminic acid in BPES the ratio of orange fluorescence
intensity to blue fluorescence intensity amounts to 2:1, whereas that one in the presence of
DNA is 0.8:1. The increase of the blue fluorescence intensity at the expense of the orange
fluorescence intensity does not only indicate to strong interactions between DNA and
carminic acid, but also manifests that the intercalated carminic acids experience a further
structural stabilization due to its rigid environment in the base pair stacks of DNA.
57
Results and Discussion
normalized intensity
carminic acid
carminic acid + DNA
1.0
0.8
0.6
0.4
0.2
0.0
12000
15000
18000
21000
24000
27000
-1
wavenumber / cm
Fig 4.1.4: Normalized fluorescence spectra of 5µM carminic acid in BPES (dashed line) and
5µM carminic acid with 5µM DNA in BPES (solid line)
In order to obtain information of the dissociation equilibrium in the S0 state, we have
recorded the UV/VIS absorption spectra of carminic acid in BPES (see Figure 4.1.5) and
carminic acid in DMSO (see Figure 4.1.6). The band structure of both spectra (thin solid
line) was analyzed by calculating them as a superposition of four Gaussian functions
(dashed line). As expected the peak areas and the central positions of the four calculated
peaks in the BPES spectrum are different from those in the DMSO spectrum. The large red
shifts of the peak positions in the carminic acid-BPES spectrum in comparison with that of
DMSO are explained by the more efficient solvation of carminic acid by the water
molecules forming strong H bonds to the oxo and hydroxy groups of carminic acid. On the
other hand, the peak areas are determined by the concentration of the respective species and
their extinction coefficients (i.e. oscillator strength). However, the appearance of these four
58
Results and Discussion
peaks in the absorption spectrum indicates to the coexistence of four different geometric
states of carminic acid in the S0 state.
experiment
Gauss Fit
10000
-1
17890 cm
8000
-1
19060 cm
6000
-1
22230 cm
-1
ε / M cm
-1
-1
19760 cm
4000
2000
0
15000
17500
20000
22500
25000
-1
wavenumber / cm
Fig. 4.1.5: Absorption of carminic acid in BPES; the band structure (thin solid line) was
analyzed by fitting with a superposition of four Gaussian function (dashed line)
59
Results and Discussion
experiment
Gauss Fit
10000
-1
18560 cm
8000
-1
19710 cm
-1
6000
22544 cm
-1
ε / M cm
-1
-1
20340 cm
4000
2000
0
15000
17500
20000
22500
25000
-1
wavenumber / cm
Fig. 4.1.6: Absorption of carminic acid in DMSO; the band structure (thin solid line) was
analyzed by fitting with a superposition of four Gaussian function (dashed line)
Two of the gauss fitting peaks are ascribed to the undissociated carminic acid and its
deprotonated anion. Both, the undissociated acid as well as the deprotonated anion may
undergo a tautomerisation reaction in the ground state, and thus both may occur in an
equilibrium distribution as the normal form and as one tautomer. These four species may
also be formed in DMSO, as naturally containing water up to 2 % in vol. In terms of this
rationale the spectrum of carminic acid in BPES was analyzed as follows: the relatively
large peaks at 22230 cm-1 and 19760 cm-1 were assigned to the normal and tautomer form of
the undissociated carminic acid (CAH and CAH T3, see Scheme 2), respectively, whereas
the small peak at 19060 cm-1 and the very large one at 17890 cm-1 represent the lowestenergy absorption transitions of the deprotonated CA- and its tautomer CA- T3 (see Scheme
2). In the case of carminic acid in DMSO as containing water at ca. 2 % vol. the two small
peaks (18560 cm-1 and 19710 cm-1) at the low-energy edge of the absorption spectrum were
60
Results and Discussion
analogously ascribed to the absorption transitions of CA-T3 and CA-, respectively. The very
large peak at 20340 cm-1 displays the absorption of CAH T3 and the smaller one at 22544
cm-1 that of CAH.
CAH T1:
CAH:
OH
OH
O
HO
OH
O
HO
OH
OH
HO
HO
O
OH
O
CH3
OH
HO
O
HO
O
OH
O
OH
O
OH
CH3
O
OH
OH
OH
CAH T2:
CAH T3:
O
OH
HO
OH
HO
OH
OH
HO
HO
O
OH
O
CH3
O
OH
HO
HO
OH
OH
O
O
OH
CH3
O
OH
OH
Scheme 2: Molecular structure of the normal form of carminic acid (CAH) and the three tautomers
(CAH T1, CAH T2, CAH T3)
Complementary information of the different geometric states of carminic acid in
BPES was obtained upon performing semiempirical ZINDO/S computations on respectively
four tautomers of the undissociated carminic acid and its deprotonated anion in vacuum. We
have calculated the optimized geometries, electronic states and electronic transitions of all
the eight carminic acid structures. The results are summarized in Table 4.1.1 The
undissociated normal form of carminic acid (CAH) has the lowest absolute energy, Eabs= -
61
Results and Discussion
8041.5 eV and the largest S0-S1 difference (22716 cm-1) with a rather large oscillator
strength of f= 0.269, which supports our assignment of the highest-energy peaks in the
absorption spectra. The similarly low absolute energy of the tautomer CAH T3 (Eabs= 8041.3 eV), its largely red-shifted absorption band (18900 cm-1) and the even higher
oscillator strength of f= 0.393 also coincide nicely with the spectroscopy data and moreover,
indicates to a S0-state tautomerization equilibrium between CAH and CAH T3. Because of a
different solvation effect the calculated absolute energy values of CA- and CA- T3, with
Eabs= -8014.2 eV and Eabs= -8014.4 eV, respectively, cannot be correlated with those of the
undissociated carminic acid. Nevertheless the nearly equal absolute energy values
significantly reflect the coexistence of CA- and CA- T3 in the ground state. We have
calculated the optimized geometries, electronic states and electronic transitions of all these
eight carminic acid structures. The results are summarized in Table 4.1.1:
ET/kcal/mol EB/kcal/mol Eabs/eV ∆(S0-Sm)/cm-1
f
∆(S0-Sn)/cm-1
f
CAH
-185444
-25700
-8041.5
22716
0.269
23492
0.037
CA-
-184815
-25378
-8014.2
20563
0.071
23496
0.148
CAH T1
-185427
-25693
-8040.7
19351
0.606
21801.1
0.056
CA- T1
-184851
-25421.3
-8015.8
18281.1
0.632
23061.6
0.031
CAH T2
-185431
-25698.3
-8040.9
18919
0.555
25395
0.070
CA- T2
-184811
-25579
-8014
19087.7
0.683
25944.7
0.041
CAH T3
-185440
-25707
-8041.3
18900
0.393
28473.5
0.029
CA- T3
-184820
-25397.7
-8014.4
15491.4
0.414
21758.3
0.051
Table 4.1.1: Calculated values of the total energy (ET), the binding energy (EB), the absolute
energy (Eabs), the energy of the S0-Sm transition (∆(S0-Sm)) and S0-Sn transition (∆(S0-Sm)) transition
and of the oscillator strength (f).
62
Results and Discussion
The undissociated normal form of carminic acid (CAH) has the lowest absolute
energy, Eabs= -8041.5 eV and the largest S0-S1 difference (22716 cm-1) with a rather large
oscillator strength of f= 0.269, which supports our assignment of the highest-energy peaks
in the absorption spectra. The similarly low absolute energy of the tautomer CAH T3 (Eabs=
-8041.3 eV), its largely red-shifted absorption band (18900 cm-1) and the even higher
oscillator strength of f = 0.393 also coincide nicely with the spectroscopy data and
moreover, indicates to a S0 -state tautomerization equilibrium between CAH and CAH T3.
Because of a different solvent effect the calculated absolute energy values of CA- and CAT3, with Eabs= -8014.2 eV and Eabs= -8014.4 eV, respectively, cannot be correlated with
those of the undissociated carminic acid. Nevertheless the nearly equal absolute energy
values consistently reflect the coexistence of CA- and CA- T3 in the ground state. The
difference in their oscillator strengths matches perfectly to the observation of a relatively
small absorption peak of CA- in comparison with the four times larger one of CA- T3 (see
Figure 4.1.5).
The absorption spectra and semiempirical computations support the hypothesis that
the tautomers CAH T3 and CA- T3 may be formed in the S0 state. This has been confirmed
by detecting the fluorescence of the tautomers subsequently by selective excitation at 480
nm. The Figures 4.1.7 and 4.1.8 depict the fluorescence spectra recorded from a BPES
solution (see Figure 4.1.7) and from a DMSO solution (see Figure 4.1.8). The Gaussian fit
of the BPES spectrum yielded three peaks positioned at 16487 cm-1, 14906 cm-1 and 14277
cm-1. The 16487 cm-1 peak was assigned to the fluorescence of CAH T3, whereas the peaks
at 14906 cm-1 and 14277 cm-1 display the fluorescence spectrum (electronic and vibronic
transition) of CA- T3.
63
normalized intensity
Results and Discussion
experiment
Gauss Fit
1.0
-1
14277 cm
0.8
-1
14906 cm
-1
16487 cm
0.6
0.4
0.2
0.0
12000
14000
16000
18000
20000
-1
wavenumber / cm
Fig. 4.1.7: The spectra of the orange fluorescence of 5 µM carminic acid in BPES. The band
structure of the spectra (thin solid line) was analyzed by fitting with a superposition of Gaussian
functions (dashed line).
The analysis of the DMSO spectrum resulted into two peaks (at 17440 cm-1 and
16487 cm-1) only, which were ascribed to fluorescence transitions of CAH T3.
64
normalized intensity
Results and Discussion
experiment
Gauss Fit
1.0
-1
16487 cm
0.8
-1
17440 cm
0.6
0.4
0.2
0.0
12000
14000
16000
18000
20000
22000
-1
wavenumber / cm
Fig. 4.1.8: The spectra of the orange fluorescence of 5 µM carminic acid in DMSO; the
band structure of the spectra (thin solid line) was analyzed by fitting with a superposition of
Gaussian functions (dashed line).
The fluorescence quantum yield of the carminic acid tautomers dissolved in
Millipore water was determined as a value of 2.7 %. This rather low fluorescence quantum
yield suggests that the deactivation of the S1 state may occur either through an efficient
intersystem crossing to the T1 state or by an ESIPT followed by internal conversion to the
S0 state. An argument for the ESIPT is the observation of the dependence of the blue
fluorescence intensity on the nature of the solvent shell.
A relative measure for the interactions between carminic acid and DNA was
achieved by performing fluorescence titration measurements. Carminic acid in BPES at the
total concentration of cT = 6 µM was excited at 320 nm. The total fluorescence spectrum
between 360 and 800 nm was recorded in the magic-angle polarization configuration. The
relative fluorescence intensities (I) were determined by integrating the fluorescence spectra.
The fluorescence titrations were performed upon detecting the fluorescence spectrum of
carminic acid in dependence on the DNA concentration cDNA (M in nucleotide) over the
range of 1 nM to 1 mM. Under the simplified approximation that at maximum one carminic
65
Results and Discussion
acid molecule may bind to one DNA nucleotide, only, the fluorescence titration data may be
described by the equation:
cB =
cT ⋅ c DNA
K B−1 + c DNA
V.1.1
with KB representing the intrinsic binding constant. The fluorescence intensities of carminic
acid in the absence of DNA (I0), in the course of the titration with DNA (I) and at high DNA
concentrations, i.e. the plateau value (I∞), are used to calculate the concentration of free
carminic acid:
c
F
=
cT ( I − I ∞ )
I0 − I∞
V.1.2
This relationship allows for the determination of the concentration of bound carminic
acid with cB = cT – cf. Figure 4.1.9 shows the fluorescence titration curve depicted as the
concentration of bound carminic acid, cB, versus the DNA concentration, cDNA. An intrinsic
binding constant, KB= 5.0×105 (M nucleotide)-1 provided the best fit (thick solid line) to the
experimental data (dots). The uncertainty in this parameter is about 10 %. This large binding
constant agrees very well with that reported for the doxorubicin-calf thymus DNA complex
with KB= 1.5×105 (M bp)-1 [11] and that of the daunomycin-DNA complexes with KB=
7×105 (M bp)-1[47]. Binding constants of that order of magnitude are typical for strong
intercalative bindings of chromophores to the DNA base pairs [11].
66
bound carminic acid / M
Results and Discussion
-6
6.0x10
-6
5.0x10
-6
4.0x10
experimental data
-1
-6
cB= cT cDNA/(KB + cDNA)
3.0x10
-6
2.0x10
-6
1.0x10
0.0
0.0
-5
4.0x10
-5
8.0x10
-4
1.2x10
-1
DNA concentration / M nucleotide
Fig. 4.1.9: Concentration dependence of bound carminic acid, cB, on the DNA
concentration; the experimental data (dots) were obtained from fluorescence titration of 6 µM
carminic acid with DNA in BPES and were fitted (solid line) employing the relationship cB= cT⋅
cDNA/(KB-1+cDNA) with KB= 5.0×105 (M nucleotide)-1.
4.2. Femtosecond Spectroscopy
4.2.1 Transient Absorption spectroscopy
The excited-state decay dynamics of carminic acid in BPES were investigated utilizing
fs-resolved transient absorption spectroscopy. Figure 4.2.1.1 represents the temporal
evolution of the transient absorption spectra between 0 and 100 ps illustrated as a 3-D plot.
The rather complicated band structure of the spectra obviously originates from a
67
Results and Discussion
superposition of photo-bleaching of the ground-state population densities (absorption > 0)
and transient absorption of the excited states (absorption < 0). In order to analyze the time
evolution of the involved spectral components, each spectrum was decomposed into 8
Gaussian functions, where photo-bleaching was accounted for by positive amplitude and the
transient absorption by negative amplitude. The Gaussian fits of the transient absorption
spectra for delay times τ from 0 ps to 100 ps in steps of 5 ps were calculated by keeping
constant the peak positions and the widths of the Gaussian functions, whereas the
amplitudes of the Gaussians were varied until the best fit was obtained.
on
absorpti
0,04
0,02
0,00
-0,02
-0,04
-0,06
-0,08
100
tim
80
e/
60
40
ps
20
24000
22000
-1
20000
18000
cm
/
16000
er
b
14000
m
w
en
v
a
u
Fig. 4.2.1.1 3-D plot of the temporal evolution of the transient absorption spectra obtained
for carminic acid in BPES.
68
Results and Discussion
The temporal evolution of every amplitude, as being a measure for the relative absorption of
the respective spectral component, could be described by a mono-exponential function. The
exponential fits yielded four different time constants for the 8 spectral components, which
are τta= 15±0.5 ps, 20±0.8 ps, 33±1.2 ps and 46±1.4 ps. This result agrees very well with the
observation of four different species of carminic acid (i.e. CAH, CAH T3, CA-, CA- T3) in
the stationary absorption spectrum. Hence each set of components with equal time constants
constitutes a transient absorption spectrum of one species of carminic acid, respectively.
Figure 4.2.1.2 exemplarily displays the transient absorption spectrum (thin solid line)
recorded at the delay time τ= 1 ps. The Gaussian fit (thick short dashed line) results from
the superposition of 8 Gaussian functions that are assigned to the four different time
transient absorption
constants (thick solid, dashed, dashed dotted and dotted lines).
ta spectrum
46 ps Gaussian
20 ps Gaussian
15 ps Gaussian
33 ps Gaussian
Gaussian superposition
0.02
0.00
-0.02
-0.04
-0.06
14000
16000
18000
20000
22000
24000
-1
wavenumber / cm
Fig. 4.2.1.2 Transient absorption spectrum (thin solid line) recorded at the delay time τ= 1
ps; the fit (thick short dashed line) arises from the superposition of 8 Gaussian functions assigned to
four different time constants (thick solid, dashed, dashed dotted and dotted lines).
69
Results and Discussion
4.2.2 Fluorescence up-conversion spectroscopy
Complementary information of excited-state decay behaviour were obtained from
fluorescence up-conversion experiments. The rather small fluorescence quantum yield of
carminic acid limits the detection of the up-converted fluorescence emission to the
wavenumber range of 13800 – 17200 cm-1, and thereupon to the observation of two
fluorescence bands, only, one peaked at 14910 cm-1 and assigned CA- T3 and the other one
fluorescence intensity
at 16490 cm-1 ascribed to CAH T3 (see Figure 4.2.2.1).
spectrum
Gaussians
Gaussian superposion
1.0
0.8
-1
16490 cm
0.6
-1
14910 cm
0.4
14000
15000
16000
17000
-1
wavenumber / cm
Fig. 4.2.2.1 Fluorescence-up conversion spectrum (dots) fitted by a superposition (thin solid
line) of two Gaussians (dashed line).
70
Results and Discussion
Figure 4.2.2.2 shows the 3D-plot of carminic acid in BPES at a concentration of 0.6
mM. Addition of DNA at a surplus concentration of 3.9 mM (in nucleotide) does not change
the spectral features but distorts a little their time evolution (see Figure 4.2.2.3).
1,2
0,8
0,6
0,2
0,0
20
16000
40
60
/ ps
80
17000
100
wa
t im e
ve
nu
m
0
15000
/c
m-
14000
1
0,4
be
r
intensity
1,0
Fig. 4.2.2.2 3-D plot of the temporal evolution of the fluorescence up-conversion spectra obtained
for 0.6 mM carminic acid in BPES.
71
Results and Discussion
0,8
0,6
0,0
16000
40
/ ps
80
17000
100
w
ti m e
60
av
en
u
20
be
15000
r/
14000
0,2
cm
-
1
0,4
m
intensity
1,0
Fig. 4.2.2.3: 3-D plot of the temporal evolution of the fluorescence up-conversion spectra
obtained for 0.6 mM carminic acid and 3.9 mM DNA in BPES.
To achieve indepth insight into the fluorescence decay dynamics and in particular, to
explore the influence of the intercalative interaction with DNA, the up-conversion decay
curves detected at 14910 cm-1 and 16490 cm-1, respectively, both, in the parallelpolarization and perpendicular-polarization geometry, were recorded (see Figure 4.2.2.4 –
4.2.2.7). Figures 4.2.2.4 and 4.2.2.5 depict the up-conversion decays of carminic acid in
BPES, whereas those of carminic acid in BPES in the presence of DNA are shown in
Figures 4.2.2.6 and 4.2.2.7.
72
fluorescence intensity
Results and Discussion
1000
I(t)par
fit(t)par
I(t)perp
fit(t)perp
800
600
400
200
0
-20
0
20
40
60
80
100 120 140
time / ps
Fig. 4.2.2.4: Fluorescence up-conversion decays of 0.6 mM carminic acid in BPES (thin
solid line) detected at 14910 cm-1 in the parallel-polarization geometry (I(t)par) and in the
perpendicular-polarization geometry (I(t)perp); both fits were obtained using the parameters: a =
0.37, b= 0.91, τ1= 1.7 ps, τ2= 33 ps and r(0)= 0.137.
73
fluorescence intensity
Results and Discussion
I(t)par
fit(t)par
I(t)perp
fit(t)perp
600
400
200
0
-20
0
20
40
60
80
100 120 140
time / ps
Fig. 4.2.2.5: Fluorescence up-conversion decays of 0.6 mM carminic acid in BPES (thin
solid line) detected at 16490 cm-1 in the parallel-polarization geometry (I(t)par) and in the
perpendicular-polarization geometry (I(t)perp); both fits were obtained using the parameter: a= 0.50,
b= 0.92, τ1= 1.5 ps, τ2= 47 ps and r(0)= 0.202.
74
Results and Discussion
fluorescence intensity
500
I(t)par
fit(t)par
I(t)perp
fit(t)perp
400
300
200
100
0
-20
0
20
40
60
80
100
120
140
time / ps
Fig. 4.2.2.6: Fluorescence up-conversion decays of 0.6 mM carminic acid and 3.9 mM DNA
in BPES (thin solid line) detected at 14910 cm-1 in the parallel-polarization geometry (I(t)par) and in
the perpendicular-polarization geometry (I(t)perp); both fits were obtained with the parameter: a=
0.50, 0.97, τ1= 1.9 ps, τ2= 48 ps and r(0)= 0.182.
75
Results and Discussion
fluorescence intensity
700
600
I(t)par
fit(t)par
I(t)perp
fit(t)perp
500
400
300
200
100
0
-100
0
20
40
60
80
100
120
140
time / ps
Fig. 4.2.2.7: Fluorescence up-conversion decays of 0.6 mM carminic acid and 3.9 mM
DNA in BPES (thin solid line) detected at 16490 cm-1 in the parallel-polarization geometry (I(t)par)
and in the perpendicular-polarization geometry (I(t)perp); both fits were obtained with the parameter:
a= 0.50, b= 0.97, τ1= 1.1 ps, τ2= 61 ps and r(0)= 0.243.
76
Results and Discussion
The decay curves were evaluated by a merged nonlinear fitting and deconvolution
process directly on the parallel (Ipar) and perpendicular (Iperp) signals, using the model
functions
i par
i
perp
(t ) = (1
+ 2 r (t
(t ) = (1
− r (t
)) f (t )
))
f
(t )
These were convoluted by the Gaussian model function, I(t) ∝ i(t) ⊗ G(t). In the fitting we
used
f ( t ) = const . + a ⋅ exp( −t / τ 1 ) + ( 1 − a ) ⋅ b ⋅ exp( −t / τ 2 )
and
r( t ) = r 0 ⋅ exp( −t / τ ro t )
where τrot = 196 ps [19]. Ipar and Iperp represent the fluorescence intensity that was detected
in the parallel and perpendicular polarization geometry, respectively. Table 4.2.2.1 contains
the amplitudes, a and b, the time constants, τ1 and τ2, as well as the anisotropy parameter
r(0), which were obtained from the best fit. The short time constants τ1 were ascribed to a
fast structural relaxation process, whereas the long time constants τ2, with 33±0.7 ps and
47±1 ps, coincide with two excited-state lifetimes that resulted from the Gaussian analysis
of the transient absorption spectroscopic data. These lifetimes were assigned to the excitedstates of the CA- T3 and CAH T3, respectively. As obvious from Figure 4.2.2.6 and Figure
4.2.2.7, the presence of DNA is associated with a significant prolongation of these lifetimes:
τ2 changes from 33 ps to 48 ps for CA- T3, whereas CAH T3 exhibits an increase from 47
ps to 61 ps. In addition the anisotropy parameter r(0) becomes significantly larger under the
influence of DNA. The latter effect confirms for a second time the efficient couplings of
carminic acid to DNA. Moreover, the prolongation of the lifetimes arises from the aprotic
rigid environment of the carminic acid incorporated between the DNA base pairs which
77
Results and Discussion
presumably slows down the nonradiative processes that are the internal conversion and the
intersystem crossing of carminic acid. Reverse ESIPT processes to the excited state of the
corresponding normal form are very unlikely to take place because of their higher excitedstate energy. Since fluorescence up-conversion were generated using a pump pulse at 400
nm, there are two excitation pathways to populate the excited-states of the tautomers, CAH
T3 and CA- T3. One pathway is direct photoexcitation of the tautomers in the ground state
and the other occurs via ESIPT from the excited state of the respective normal form (CAH
and CA- ). The rise of all fluorescence up-conversion decay curves (see Figures 4.2.2.4–
4.2.2.7) takes place within the instrumental response time (i.e. 270 fs).
a
b
τ1 / ps
τ2 / ps
r(0)
CA-T3: 14910 cm-1
0.37
0.91
1.7
33
0.137
CAHT3: 16490 cm-1
0.50
0.92
1.5
47
0.202
CA-T3+DNA: 14910 cm-1
0.50
0.97
1.9
48
0.182
CAHT3+DNA: 16490 cm-1
0.50
0.97
1.1
61
0.243
Table 4.2.2.1: The amplitudes, a and b, the time constants, τ1 and τ2, and the anisotropy,
r(0), were obtained from the best fit of the fluorescence up-conversion decay curves that are
detected at 14910 cm-1 and 16490 cm-1.
This may be explained by an ultrafast ESIPT with a proton-transfer time constant
smaller than 270 fs. Just as probable is that no ESIPT takes place and the observed
fluorescence originates from the directly photoexcited tautomers. On the other hand, the
solvent dependence of blue and orange fluorescence intensities (see Figures 4.1.2, 4.1.3 and
4.1.4) unambiguously indicates to the occurrence of ESIPT. This is consistent with the
results obtained from a fs transient absorption spectroscopy study of 1,8-dihydroxy-
78
Results and Discussion
anthraquinone by Neuwahl and Coworkers [46]. The authors reported the observation of a
practically instantaneous ESIPT with a proton-transfer time constant much smaller 100 fs.
79
Conclusions
5. Conclusions
In a buffer solution at pH= 7 carminic acid was found to occur as undissociated acid
(CAH) and as its deprotonated anion (CA-), both, at nearly the same concentration. The
Gaussian analysis of the UV/VIS absorption spectrum (see Figures 4.1.5 and 4.1.6) gave
indication to four species of carminic acid coexisting in the ground state. On the basis of
semiempirical molecular geometry computations these four species were assigned to the
normal forms of CAH and CA- and to their tautomers, CAH T3 and CA- T3. The existence
of four ground-state species was confirmed by femtosecond transient absorption
spectroscopy experiments. The Gaussian analysis of the spectroscopic data yielded four
different excited-state lifetimes: 15 ps, 20 ps, 33 ps and 46 ps. The solvent dependence of
the blue and orange fluorescence intensity gave indication to ESIPT. Carminic acid
dissolved in the aprotic solvent DMSO does not exhibit blue fluorescence. On the other
hand, in protic environment (BPES) the orange to blue fluorescence intensity ratio is about
2:1, whereas the presence of DNA alters this ratio to 0.8:1. This implies that ESIPT takes
instantaneously place in an aprotic medium that is also partially achieved for carminic acid
by intercalating between the base pairs of DNA. Fluorescence up-conversion experiments
conducted on carminic acid in BPES led to two lifetimes, 33 ps and 47 ps that were
assigned to the tautomers CA- T3 and CAH T3 of no dissociated and deprotonated carminic
acid, respectively. Since the rise time of fluorescence up-conversion transients corresponds
to the instrumental response time constant of 270 fs, only ultrafast ESIPT with a protontransfer time constant smaller than 100 fs may be involved in the excitation of the
tautomers. In the presence of DNA the observed lifetimes increase from 33 ps to 48 ps for
CA- T3 and from 47 ps to 61 ps for CAH T3. Their prolongation in the presence of DNA
may be taken as an additional evidence for the formation of intercalation complexes
between DNA and carminic acid. The rigid aprotic environment between the DNA base
pairs are expected to slow down the nonradiative deactivation processes of
80
the
Conclusions
electronically excited carminic acid which are internal conversion and intersystem crossing
[13].
81
References
6. References
1.
J. B. Chaires, N.D., D. M. Crothers, Biochemistry, 1985. 24: p. 260.
2.
X. Qu, J.O.T., I. Fokt, W. Pribe, J. B. Chaires, Proc. Nat. Acad. Sci., 2000(12032).
3.
W. A. Remers, The Chemistry of Antitumor Antibiotics. Vol. 1. 1979, New York:
Wiley.
4.
A. H. -J. Wang, G.U., G. J. Quigley, A. Rich, Biochemistry, 1987. 26(1152).
5.
A. Andreoni, A.C., A. Kisslinger, M. Mastrocinque, G. Portella, P. Riccio, G.
Roberti, Photochem. Photobiol., 1993. 57(851).
6.
D. Zhong, S.K.P., Ch. Wan, A. H. Zewail, Proc. Nat. Acad. Sci., 2001. 98(11873).
7.
X. Qu, C.W., D. Zhong, H. -C. Becker, A. H. Zewail, Proc. Nat. Acad. Sci., 2001.
98(1421).
8.
C. Wan, T.X.H.-C.B., A. H. Zewail, Chem Phys. Lett, 2005. 412(158).
9.
C. A. Frederick, L.D.W., G. Ugetto, G. A. van der Marel, J. H. van Boom, A. Rich,
A. H.-J. Wang, Biochemistry, 1990. 29(2538).
10.
J. B. Chaires, N.D., D. M. Crothers, Biochemistry, 1982. 21(3933).
11.
J. B. Chaires, W.T., D. E. Graves, T. G. Burke, Dissection of Free Energy Of
Anthracycline Antibiotic Binding to DNA: Electrostatic Contribut. J. Am. Chem.
Soc., 1993. 115: p. 5360-5364.
12.
T. Bountis, Proton transfer in hydrogen- bonded systems, ed. N. ASISeries. 1991,
New York, London: PlenumPress.
13.
T.P. Smith, K.A.Z., K.Thakur, G.C. Walker, K.Tominaga, P.F. Barbara, Ultrafast
studies on proton transfer in photostabilizers. J. Phys. Chem. 95, 1991: p. 10465.
14.
R.P. Bell, The proton in chemistry. 1959, London: Methuen & Co.
15.
R. Steward, The proton: applications to organic chemistry. 1985, Orlando: Academic
Press.
16.
E.F. Caldin, V.G., Proton transfer reactions. 1975, Chapman & Hall: London.
References
17.
P. Mitchell, Coupling of phosphorylation to electron and hydrogen transfer by a
chemiosonic type of mechanism, Nature, Editor. 1961: London.
18.
J.P. Rasimas, G.J.B., J. Phys. Chem., 1994. 98: p. 12949.
19.
J.P. Rasimas, G.J.B., J. Phys. Chem., 1995. 99: p. 11333.
20.
J.P. Rasimas, K.A.B., G. J. Blanchard, A molecular Loch-and-Key approach to
detecting solution phase self-assembly. A fluorescence and absorption study of
carminic acid in aqueous glucose solution. J. Phys. Chem. 1996, 1996. 100: p. 72207229.
21.
J. D: Watson, F.H.C.C., Molecular structure of nucleic acids; a structure for
deoxyribose nucleic acid. Nature, 1953. 171: p. 737 – 738.
22.
T.L. Gilchrist, Heterocyclic Chemistry. 3rd Edition ed. 1963.
23.
S. Steenken, S.V.J.J., J. Am. Chem. Soc., 1997. 119: p. 617.
24.
J.E. Rogers, L.A.K., Nucleic acid oxidation mediated by naphthalene and
benzophenoneimide and diimide derivatives: Consequencesfor DNA redox chemistry.
J. Am. Chem. Soc., 1999. 121: p. 3854-3861.
25.
Chaires, J.B.F., K. R.;Herrera, J. E.; Britt, M.; Waring, M. J., Biochemistry, 1987.
26: p. 8227-8236.
26.
Record, M.T.A., C. F.; Lohman, T. M., Rev. Biophysical, 1978. 11: p. 103-178.
27.
Record M. T., J.S., R.S., Nonspecific Protein-DNA Interaction. Ed. CRC Press: Boca
Raton Fl, 1990: p. 33-69.
28.
Friedmann, R.A.G.M., G.;, Biopolymers, 1984. 23: p. 2671-2714.
29.
Wei Sun, Y.H., Kui Jiao, Voltammetric albumin quantification based on its
interaction with carminic acid. J. Serb. Chem. Soc., 2006. 71(4): p. 385-396.
30.
J. R. Lakowicz, Principle Of Fluorescence Spectroscopy. Second Edition ed, New
York, Boston, Dordrcht, London, Moscow.
31.
A. E. Johnson and A. B. Myers, J. Chem. Phys., 1995. 102(9): p. 3519.
32.
A. E. Johnson and A. B. Myers, J. Phys. Chem., 1996. 100: p. 7778.
33.
P. Atkins, Molecular Quantum Dynamics. 1983: Oxford University Press.
References
34.
P. Bernath, Spectra of Atoms and Molecules. 1995: Oxford University Press.
35.
Sutherland, R., Handbook of Nonlinear Optics. 1996: Marcel Dekker.
36.
R. Alfano, The Super Continuum Laser Source. 1989: Springer-Verlag.
37.
C. Rullière, Femtosecond Laser Pulses. 1998.
38.
C. H. Brito Cruz, Dispersion effects on femtoseconds laser pulses. Winter collegeon
ultrafast phenomena, 1991.
39.
C. Shank and E. Ippen, Appl. Phys. Lett., 1974. 24,: p. 373.
40.
Gustavsson T.; Cassara, L.G., V.; Gurzadyan, G.; Mialocq, J.-C.; Pommeret, S.;
Sorgius, M.; van der Meulen, P., Femtosecond Spectroscopic Study of Relaxation
Processes of Three Amino-Substituted Coumarin Dyes in Methanol and Dimethyl
Sulfoxide. J. Phys. Chem. A, 1998. 102(23): p. 4229-4245.
41.
S. Pommeret, R.N., P. van der Meulen, M. Menard, T. Gustavsson, G. Vigneron,
Chem Phys Lett, 1998. 228: p. 833.
42.
J.E. Ridley, M.Z., Theor. Chim Acta. 32. 1973. 111.
43.
G. Smulevich, P.F., J. Chem. Phys., 1987. 87: p. 5657.
44.
G. Smulevich, P.F., A. Feist, M. P. Marzocchi, J. Chem. Phys., 1987. 87: p. 5664.
45.
M. P. Marzocchi, A.R.M., M. Casu, G. Smulevich, J. Chem. Phys., 1998. 107.
46.
F.V.R. Neuwahl, L.B., R. Righini, G. Buntix, Chem Phys 3, 2001. 87: p. 1227.
47.
J.B. Chaires, N.D., D.M. Crothers, Biochemistry, 1985. 24: p. 260.
Summary
Photo-excited carminic acid and carminic acid-DNA complexes in a buffer solution
at pH= 7 have been examined using a variety of spectroscopy techniques, that are in
particular, the femtosecond resolved fluorescence up-conversion and transient absorption
spectroscopy. The observation of dual fluorescence emission, one peaks at 470 nm and the
other at 570 nm, indicates to an excited-state (S1) intramolecular proton transfer (ESIPT). A
detailed analysis of the transient absorption measurements of an aqueous carminic-acid
solution at pH= 7 yielded four lifetimes for the excited state (S1): 8 ps, 15 ps, 33 ps and 46
ps. On the other hand, only two lifetimes, 34 ps and 47 ps, were observed by fluorescence
up-conversion spectroscopy because of the detection limitation to the long wavelength edge
of the carminic-acid spectrum. The four S1 lifetimes were ascribed to the coexistence of
respectively two tautomer (normal and tautomer) forms of carminic acid, in the
nondissociated state (CAH) and in the deprotonated state (CA-). The fluorescence upconversion measurements of carminic acid-DNA complexes exhibited a prolongation of the
fluorescence lifetimes. This effect was taken as evidence for the formation of intercalation
complexes between the carminic acid and the DNA. The intercalative binding of the
carminic acid to DNA was confirmed by the fluorescence titration experiments leading to a
binding constant of KB= 5.0×105 (M nucleotide)-1 that is typical for anthracycline-DNA
complexes.
Zusammenfassung
Die statischen und dynamischen Wechselwirkungen zwischen optisch angeregter
Carminsäure und DNA wurden mittels verschiedener, komplementärer Techniken der
stationären und zeithochaufgelösten optischen Spektroskopie detailliert untersucht. Die
stationäre Fluoreszenz und UV/VIS-Absorptionsspektroskopie die Identifizierung von vier
Tautomeren der Carminsäure im elektronischen Grundzustand ermöglichte, konnte in durch
Messung der UV-Vis-Absorptionsspektren in Abhängigkeit vom pH-Wert nicht nur der
pKa-Wert von Carminsäure bei pH= 7 bestimmt werden, sondern auch noch gleichzeitig die
Koexistenz von nur einfach deprotonierter Carminsäure mit der nicht-dissoziierten Säure für
pH-Werte zwischen pH= 2 und pH= 9 nachgewiesen werden. Die Koexistenz der vier
Tautomeren von Carminsäure im Grundzustand der einfach deprotonierten und
undissozierten Carminsäure wurde mit Ergebnissen aus semiempirischen ZINDO/S
Rechnungen bestätigt. Die statischen Wechselwirkungen und damit die Bindungskonstante
und der Bindungsmodus von Carminsäure-DNA-Assoziaten wurde durch Messung von
Fluoreszenz-Titrationskurven für verschiedene Carminsäure-Konzentrationen untersucht.
Die Carminsäure-DNA-Assoziate haben eine Bindungskonstante von KB= 5.0×105 (M
nucleotide)-1, was auf einen interkalativen Bindungsmodus hinweist. Die duale Fluoreszenzemission von Carminsäure mit Maxima bei 470 nm und bei 570 nm zeigt, dass ein
intramolekularer Protonentransfer im elektronisch angeregten S1-Zustand (ESIPT)
stattfindet. Da die Zeitkonstante des ESIPT von der Molekülgeometrie im S1-Zustand
bestimmt wird, wurde der ESIPT als Sonde für die dynamischen Wechselwirkungen
zwischen Carminsäure und DNA genutzt. Die Relaxationsdynamiken von optisch
angeregter Carminsäure und Carminsäure-DNA-Assoziaten wurden mittels Femtosekundenauflösender transienter Absorptionsspektroskopie und Fluoreszenz-Aufwärtskonversion
untersucht. Die detaillierte Analyse der gemessenen transienten Absorptionsspektren von
Carminsäure in einer Lösung mit pH= 7 ergab vier Lebensdauern für den S1-angeregten
Zustand: 8 ps, 15 ps, 33 ps und 46 ps, wobei zwei der vier Lebensdauern, mit 34 ps und 47
ps, durch Messungen mittels der Fluoreszenz-Aufwärtskonversion bestätigt wurden. Die
vier Lebensdauern werden zwei Tautomeren von Carminsäure (CAH) und deren jeweils
deprotonierte Formen (CA-) im S1-Zustand zugeordnet. Für die Carminsäure-DNAAssoziate wurde eine Zunahme der Lebensdauer im Vergleich zur reinen Carminsäure
nachgewiesen, was als weiterer Hinweis auf eine Interkalation der Carminsäuremoleküle in
die DNA bewertet wird.
Appendix
Used Software
Computation: Hyperchem 7.1
Spectral Analyses: Origin 5.0
SigmaPlot 7.0
Peak Fitting Modules
Laser systems
Ti: Sapphire laser: Coherent MIRA 900
Pump laser: solid state laser Coherent VERDI V10
Spectrographs
Absorption: Perkin Elmer Lambda 2
Fluorescence: Jobin-Yvon FluoroMax-3
Lebenslauf
Vorname
Radu Constantin
Name
Comanici
Geburtsdatum
09.01.1977
Geburtsort
Piatra-Neamt, Rumänien
Staatsangehörigkeit
Rumänisch
Familienstand
Verheiratet, keine Kinder
2002 - 2006 wissenschaftlicher Angestellter im Rahmen der Promotion im Institut für
Physikalische Chemie I an der Friedrich-Alexander Universität ErlangenNürnberg
2001 Masterdiplom unter Betreuung von Prof. Dr. Viorica Simon
1999 - 2001 Master an der Babes-Bolyai Universität Klausenburg in Rumänien
1999 Diplomprüfüng mit Diplomarbeit unter Betreuung von Prof. Dr. Traian Iliescu
1995 - 1999 Student an der Physik-Fakultät der Babes-Bolyai Universität Klausenburg in
Rumänien
1995 Abitur am Informatik-Gymnasium Nr. 1
1983- 1995 Schulausbildung mit Berufsabschluss als Software-Analyst und
Service für Computertechnik am Informatik-Gymnasium
Acknowledgement
I would like to thank everyone who has contributed in any way to the successful completion
of my PhD thesis.
First of all, I would like to thank Prof. Dr. Carola Kryschi for accepting me in her work
group and giving me the opportunity make my PhD thesis in the Physical Chemistry I
Institute of the Friedrich Alexander University of Erlangen. Also I want to thank all my
colleagues in the work group of Prof. Dr. Kryschi for the good collaboration and for the
friendly atmosphere.
On this way I want to express my acknowledgement to all my colleagues from this institute.
I would like to thank to the entire employees of the mechanic workshop good coordinated
from Mr. Friedhold Wölfel, for the speedy and professional expertise, for the support and
help with electrical and electronics instrumentation to Mr. Dipl. ing. Dirk Harnisch, to Ms
Julianne Roth for her help in administrating the used chemicals.
My special thanks goes to the Dr. Thomas Gustavson and entire collaborators the
Laboratoire Francis Perrin CEA in Gif sur Yvette France, for helping and coordinate the
femtoseconds measurements to realize the time resolved spectroscopy.
I would like to thank my mother and my brother for all their love and support as well as for
the understanding they shows to me in every circumstance.
Last but not least I would like to thank my wife, Karmen, for the great help, support and
understanding she gave to me during my PhD studies.