Second harmonic optical activity of tryptophan derivatives adsorbed at the air/water interface

Article abstract of DOI:10.1021/jp036252e

Optical activity in second harmonic reflection from the air/water interface of solutions of simple tryptophan (trp) derivatives has been investigated. The derivatives include D- and L-optical isomers of Boc-trp and DD- and LL-isomers of the dipeptide Boc-trp-trp, where Boc is tert-butyloxycarbonyl. An experimental method based on polarization modulation of fundamental radiation by rotation of a quarter-wave retardation plate has been used to completely characterize the second-order nonlinear susceptibilities of the chiral surfaces, for the fundamental wavelength ?? = 564 nm. Boc-trp and Boc-trp-trp show similar chiral contributions to the nonlinear susceptibilities, which indicates that the chiral response does not depend on intramolecular interactions between trp residues. The efficiency of second harmonic reflection has been shown to be resonantly enhanced at the two-photon level, in the region of the near-UV bands (a??280 nm) of the indole chromophore of tryptophan. The origin of the chiral response is discussed in terms of simple microscopic models, and a comparison is made with optical activity of tryptophanyl compounds in linear spectroscopy. The peptide gramicidin, incorporating several trp residues, showed no evidence of optical activity in second harmonic reflection from the air/water interface. This indicates that second harmonic optical activity is sensitive to structural effects such as conformational and orientational distributions of trp residues at the interface.

Full text of DOI:10.1021/jp036252e

1020
J. Phys. Chem. B 2004, 108, 1020-1029
Second Harmonic Optical Activity of Tryptophan Derivatives Adsorbed at the Air/Water
Interface
S. A. Mitchell* and R. A. McAloney
Steacie Institute for Molecular Sciences, National Research Council of Canada, 100 Sussex DriVe,
Ottawa, Ontario K1A 0R6, Canada
ReceiVed: August 1, 2003; In Final Form: NoVember 6, 2003
Optical activity in second harmonic reflection from the air/water interface of solutions of simple tryptophan
(trp) derivatives has been investigated. The derivatives include D- and L-optical isomers of Boc-trp and DD-
and LL-isomers of the dipeptide Boc-trp-trp, where Boc is tert-butyloxycarbonyl. An experimental method
based on polarization modulation of fundamental radiation by rotation of a quarter-wave retardation plate has
been used to completely characterize the second-order nonlinear susceptibilities of the chiral surfaces, for the
fundamental wavelength λ ) 564 nm. Boc-trp and Boc-trp-trp show similar chiral contributions to the nonlinear
susceptibilities, which indicates that the chiral response does not depend on intramolecular interactions between
trp residues. The efficiency of second harmonic reflection has been shown to be resonantly enhanced at the
two-photon level, in the region of the near-UV bands (∼280 nm) of the indole chromophore of tryptophan.
The origin of the chiral response is discussed in terms of simple microscopic models, and a comparison is
made with optical activity of tryptophanyl compounds in linear spectroscopy. The peptide gramicidin,
incorporating several trp residues, showed no evidence of optical activity in second harmonic reflection from
the air/water interface. This indicates that second harmonic optical activity is sensitive to structural effects
such as conformational and orientational distributions of trp residues at the interface.
1. Introduction
of interfacial biomolecules have also been carried out using sum
frequency generation vibrational spectroscopy.16-18
Following the first reports of optical activity in second
harmonic reflection from chiral surfaces in 1993-94,1-7 there
has been growing interest in the application of second harmonic
generation optical activity (SHG-OA) for studies of bio-
molecules adsorbed at interfaces. SHG-OA measurements have
the potential for providing unique information on conformations
and conformational changes of surface-active biological mol-
ecules. Such a potential is anticipated by analogy with well-
developed linear chiroptical measurements on biomolecules in
bulk solution.8 Second harmonic generation (SHG) is inherently
sensitive to surfaces and interfaces.9 Furthermore, the sensitivity
of SHG-OA measurements is very high in comparison with
linear chiroptical methods, potentially in the submonolayer
range.1
The theory of SHG-OA of chiral surfaces is well established
at the phenomenological level,19,20 and aspects of the micro-
scopic origin of the nonlinear optical response have been
studied.21-27 At this time there is a need for experimental
evidence on relevant microscopic mechanisms, and to establish
correlations that can provide input for the development of useful
descriptive models. In this work we have reinvestigated the Boc-
trp-trp and Boc-trp systems originally studied by Frey and co-
workers7 and by Smiley and Vogel.10-12 We have used recently
developed measurement techniques20,28-30 to provide a thorough
characterization of SHG-OA of the trp systems. Our results
provide new information on the origin of SHG-OA of tryptophan
derivatives at the air/water interface.
There have been relatively few SHG studies of amino acid
derivatives adsorbed at surfaces. Smiley and Vogel10-12 studied
aromatic amino acid derivatives at the air/water interface and
used resonance enhanced SHG to probe interfacial tryptophan
(trp) residues. Frey and co-workers7 reported observations of
SHG-OA for the dipeptide Boc-trp-trp adsorbed at the air/water
interface (Boc represents a tert-butyloxycarbonyl protecting
group). SHG-OA was observed for Boc-trp-trp when the second
harmonic of the laser wavelength was resonant with the indole
chromophore of tryptophan (λ/2 ∼ 280 nm). SHG-OA has also
been reported for supported films of bacteriorhodopsin,5 for the
protein cytochrome c at air/water interface,13 and most recently
for the peptide melittin bound to the surface of a planar-
supported lipid bilayer.14 Girault and co-workers15 have used
SHG to study the enzyme glucose oxidase adsorbed at air/water
interface and also the heme proteins myoglobin and hemoglobin
adsorbed at a liquid/liquid interface. Nonlinear optical studies
2. Experimental Section
2.1. Laser System and Setup for SHG. A schematic diagram
of the experimental setup for measurement of optical activity
in second harmonic reflection from an air/water interface is
shown in Figure 1. Tunable laser pulses from a femtosecond
optical parametric oscillator (OPA) are incident on the interface
at 60° from the surface normal. A Berek’s compensator (BC)
and Glan polarizer (GP) are used to initially select p-polarization
of the laser radiation, which is incident on an achromatic quarter-
wave retardation plate (AWP). Rotation of the AWP through
360° in a plane normal to the laser beam produces a well-defined
modulation of the polarization state of the fundamental radiation,
including left- and right-hand circular polarization states. Second
harmonic radiation reflected from the air/water interface in the
specular direction is resolved in selected plane-polarized states
by using a Rochon prism polarizer (RP). The AWP and RP are
10.1021/jp036252e CCC: $27.50 Published 2004 by the American Chemical Society
Published on Web 12/18/2003

Optical Activity of Tryptophan Derivatives
J. Phys. Chem. B, Vol. 108, No. 3, 2004 1021
Figure 1. Schematic diagram of experimental setup for measurement of optical activity in second harmonic reflection from an air/water interface.
Key: OPA, optical parametric amplifier; BC, Berek’s polarization compensator; GP, Glan laser prism; BS, beam splitter; L, lens; M, mirror; AWP,
achromatic wave plate; F, filter; RP, Rochon prism polarizer; PB, Pellin-Broca prism; PMT, photomultiplier tube. The PMT, setup for reference
SHG signal (REF), and laser source for 800 nm pump radiation are behind the plane of incidence at the air/water interface.
mounted directly in the incident and reflected laser beams,
respectively, at appropriate angles to the horizontal. Second
harmonic and fundamental light reflected from the air/water
interface are spatially separated by using a Pellin-Broca prism
(PB). Colored glass filters (F) are used after the AWP to
eliminate second harmonic radiation from the beam incident
on the air/water interface and after the PB to selectively transmit
second harmonic from the interface for detection on the
photomultiplier tube (PMT).
and sodium citrate dihydrate (Aldrich). All solutions were
prepared using Milli-Q water (resistivity > 18 MΩ cm).
Gramicidin D (gD) from Sigma is a linear polypeptide
complex consisting of gramicidin types A-C. The major
component is gramicidin A, with the sequence formyl-L-Val-
Gly-L-Ala-D-Leu-L-Ala-D-Val-L-Val-D-Val-L-Trp-D-Leu-L-Trp-
D-Leu-L-Trp-D-Leu-L-Trp-ethanolamine. A solution of gD in
methanol (1 mg/mL) was mixed with a solution of the lipid
1,2-dipalmitoyl-sn-glycero-3-phosphocholine (DPPC) in CHCl3
(1 mg/mL). The DPPC/gD mixture (50% mole ratio) was used
to prepare a layer at the air/water interface by dropping a few
microliters on the surface. This preparation resulted in a
homogeneous film with a surface pressure ∼25 mN/m. Solutions
of DHN were saturated solutions in water, carefully decanted,
and then filtered with a 450 µm Millipore filter.
Synthesis of Boc-(L)Trp-(L)Trp and Boc-(D)Trp-(D)Trp. The
Boc-protected L-tryptophan amino acid was coupled to the
corresponding methyl ester protected L-tryptophan with 1,3-
diisopropylcarbodiimide in the presence of N,N-diisopropyl-
ethylamine and 1-hydroxybenzotriazole in freshly distilled
dichloromethane. After purification by column chromatography,
the methyl ester was hydrolyzed to the free acid by treating
with sodium hydroxide in a water/DMF mixture. The reaction
is shown in Scheme 1. Neutralization and extraction with
dichloromethane yielded the Boc-protected LL-tryptophan dipep-
tide (LL-trp) in a 95% overall yield. The Boc-protected DD-
tryptophan dipeptide (DD-trp) was synthesized in the same
manner.
In all experiments, a reference SHG signal from a quartz plate
was measured simultaneously with the SHG signal from the
sample. The reference signal was used to normalize the SHG
signal for fluctuations in the incident laser power. In addition,
the SHG measurements were normalized to a signal that was
recorded when a z-cut quartz wedge was substituted in the
sample position.
The OPA is a traveling-wave optical parametric amplifier of
superfluorescence (Topas, Light Conversion). The Topas was
pumped with the output of a 1 kHz Ti:sapphire regenerative
amplifier (Positive light) seeded with the output of a femto-
second Ti:sapphire oscillator (Spectra-physics). Visible laser
pulses in the range 500-600 nm and with a duration of ∼100
fs were produced by the OPA with a repetition rate of 1 kHz.
Laser pulses with energy ∼2.0 µJ/pulse were focused on the
air/water interface by using a 30 cm length focal lens. The beam
waist at the focus had a diameter ∼140 µm, measured by
translating a knife edge across the beam waist while monitoring
the transmitted energy on a photodiode. The Berek’s polarization
compensator was obtained from New Focus, and the calcite
Glan-laser polarizer, from Newport. The achromatic wave plate,
with approximately quarter-wave retardance in the range 500-
800 nm, was supplied by Halbo Optics. The magnesium fluoride
Rochon polarizer was supplied by Karl Lambrecht, and the
Pellin-Broca prism, by B. Halle Nachfl. (Berlin).
We shall refer to L-trp (D-trp) and LL-trp (DD-trp) as Boc-trp
and Boc-trp-trp, respectively, when the enantiomeric form is
not important.
Solutions of the tryptophan compounds were prepared in 22
mM citrate buffer with pH in the range 3.3-3.4. Boc-trp solid
was dissolved in a solution of sodium citrate dihydrate using a
minimal amount of water (∼10 mL). Dissolution occurred after
10-15 min of sonication. Once dissolved, the citrate/trp solution
was slowly combined with the citric acid solution and diluted
to 1.0 mM in Boc-trp. The dipeptide solutions were prepared
using the same procedure, to 0.1 mM Boc-trp-trp and 22 mM
citrate buffer. The sample solutions were characterized by using
high-performance liquid chromatography, which confirmed that
only single tryptophan species were present, i.e., Boc-trp or Boc-
trp-trp. Circular dichroism (CD) spectroscopy (Jasco J-600
spectropolarimeter) was used to confirm the enantiomeric purity
of the materials. Distinct CD spectra were observed for the Boc-
trp and Boc-trp-trp solutions in the region 240-320 nm.
The sample cell is an open dish fabricated from Teflon, with
a well diameter 7.2 cm and depth 1.7 cm. It was filled with 40
mL of sample solution. The large diameter of the cell provided
a flat fluid surface for undistorted reflection of the laser beam.
The second harmonic signal was detected with a solar-blind
photomultiplier tube (Hamamatsu) and averaged by using a
gated integrator (Stanford Research Systems). Typically, 1000-
3000 pulses were averaged for each point of a wave plate
rotation trace. It was confirmed that the second harmonic signal
varied quadratically with the pulse energy of the fundamental
radiation. Repeated measurements on the same samples showed
no indication of laser induced degradation of the samples during
data acquisition.
3. Results and Analysis
2.2. Materials and Sample Preparation. NR-(tert-butoxy-
carbonyl)-L-tryptophan (L-trp) from Aldrich, NR-(tert-butoxy-
carbonyl)-D-tryptophan (D-trp) from Flulka, and 2,7-dihydroxy-
naphthalene (DHN) from Aldrich were used as received. Citrate
buffer was prepared from citric acid monohydrate (EM Science)
3.1. Second Harmonic Generation. Smiley and Vogel10 have
studied adsorption of Boc-trp at the air/water interface of a bulk
solution of Boc-trp in a citrate buffer with pH 3.2. Surface
pressure and SHG measurements indicated a well-defined

1022 J. Phys. Chem. B, Vol. 108, No. 3, 2004
Mitchell and McAloney
SCHEME 1
adsorption behavior with increasing bulk concentration. A
monolayer film with a surface density of Boc-trp ∼1.7 nm-2
was formed for a bulk concentration of Boc-trp ∼1 mM. We
have repeated the surface pressure and SHG measurements of
Smiley and Vogel and found general agreement. Similar
measurements on Boc-trp-trp indicated significantly higher
surface activity compared with Boc-trp. This was seen in the
results of surface pressure measurements and also from observa-
tions of SHG signals from the air/solution interface.
for Rpp in Figure 2 is shown as the second harmonic wavelength,
i.e., λ/2. We also show in Figure 2 the SHG efficiency measured
for the aqueous buffer that was used to prepare the L-trp
solutions, for the fundamental wavelength λ ) 564 nm. The
efficiency of second harmonic reflection from the L-trp solution
exceeded that from the buffer with no L-trp by a factor ∼10.
3.2. Second Harmonic Optical Activity. A detailed phe-
nomenological theory of second harmonic optical activity (SHG-
OA) of chiral surfaces including electric-dipole and higher order
interactions has been presented by Maki et al.,19 and a general
approach for measurement and characterization of SHG-OA has
been developed.28-30 Here we outline some essential features
for the particular case where only electric-dipole interactions
are considered. Attention is restricted to isotropic surfaces with
full rotational symmetry about the surface normal. A general
expression for the intensity of reflected second harmonic
radiation is given by eq 2, where Ep(ω) and Es(ω) are complex
valued amplitudes of p- and s-polarized components of the
fundamental electric field (with angular frequency ω) incident
on the nonlinear surface and f, g, and h are complex valued
coefficients that are interpreted as effective nonlinear suscep-
tibilities of the surface.
The efficiency of second harmonic reflection Rpp for p-
polarized fundamental and second harmonic radiation is defined
in terms of the associated irradiances as shown in eq 1.
Ip(2ω)
[Ip(ω)]2
Rpp
)
(1)
We measured Rpp for L-trp and gD adsorbed at air/water interface
by calibrating the second harmonic response with the signal
observed when a z-cut quartz crystal was substituted in the
sample position. We used d11 ) 0.34 pm/V for quartz with the
fundamental wavelength λ ) 564 nm, based on literature data31
for λ ) 1064 nm and application of Miller’s rule to estimate
the wavelength dependence of d11. Measurements of Rpp for
L-trp and gD were repeated for several fundamental wavelengths
in the range λ ) 500-605 nm, and the results are shown in
Figure 2. Also shown in Figure 2 is an absorption spectrum of
a bulk solution of L-trp in the citrate buffer. To facilitate
comparison with the absorption spectrum, the wavelength scale
2
2
I
2ω ) |fEp (ω) + gEs2(ω) + hEp(ω)Es(ω)|
(2)
Maki et al.19 have shown how the effective susceptibilities
are related to more general nonlinear susceptibility tensors. In
eq 2, particular values of the coefficients f, g, and h are
associated with specific polarization states of the second
harmonic radiation. For p-polarized second harmonic we
designate the coefficients as fp, gp, and hp, while fs, gs, and hs
are used for s-polarized second harmonic radiation. Additional
subscripts 1 and 2 are used to indicate the real and imaginary
parts of the coefficients, respectively: fp ) fp1 + ifp2, and
similarly for gp, hp, fs, gs, and hs. With the restriction to electric-
dipole interactions and isotropic surfaces, the effective suscep-
(2)
tibilities have the following properties: gs ) 0; hp fs
ø
,
XYZ
where ø(X2Y)Z is the element of the nonlinear susceptibility tensor
that is nonzero only for chiral surfaces. The chiral susceptibilities
fs and hp are not independent; they are related by a multiplicative
factor that depends on geometrical factors and optical constants
of the interface. There are thus eight independent, real valued
constants that completely specify the nonlinear response: fp1,
fp2, gp1, gp2, hp1, hp2, hs1, and hs2. Considering that the overall
phase of the response is not significant, this number is reduced
to seven by setting hs2 ) 0 and hs1 > 0. Of these seven, only
hp1 and hp2 are specifically associated with the chiral response;
i.e., they are nonzero only for chiral surfaces. The remaining
five constants are in general nonzero for both chiral and achiral
isotropic surfaces.
Figure 2. Spectra of second harmonic efficiency Rpp of L-trp (filled
circles) and gD (open squares) at air/water interface, in comparison
with absorption spectrum of L-trp in aqueous buffer solution (broken
line). The wavelength scale for Rpp corresponds with the second
harmonic of the fundamental wavelength (λ/2). The solid triangle shows
R
pp for the buffer solution with no added L-trp, for λ ) 564 nm. Error
bars show representative uncertainties in relative values.

Optical Activity of Tryptophan Derivatives
J. Phys. Chem. B, Vol. 108, No. 3, 2004 1023
TABLE 1: Second-Order Nonlinear Susceptibilities for DHN and Several Tryptophan Derivatives Adsorbed at the Air/Water
Interface, from Global Fits of Quarter-Wave Plate Rotation Data for λ ) 564 nma
susceptibility
DHN
D-trp
L-trp
DD-trp
LL-trp
fp
gp
fs
hs
0.067 - 1.1i
-0.075 - 1.0i
0.0
-0.69 + 0.54i
-0.28 + 0.44i
-0.021 - 0.032i
0.97
-0.64 + 0.60i
-0.54 + 0.73i
-0.020 + 0.091i
1.0
-0.66 + 0.25i
-0.21 + 0.50i
0.010 - 0.052i
1.1
-0.70 + 0.36i
-0.75 + 0.81i
-0.048 + 0.062i
1.1
0.56
øZZZ
øZXX
øXXZ
øXYZ
2.6 - 2.3i
-0.59 - 8.1i
2.3
0.95 + 1.7i
-2.7 + 4.4i
5.1
2.7 + 0.84i
-5.4 + 7.3i
5.4
1.9 - 0.63i
-2.6 + 6.1i
7.3
4.0 - 1.2i
-8.3 + 8.9i
6.5
0.0
-0.14 - 0.21i
-0.13 + 0.59i
0.08 - 0.41i
-0.34 + 0.44i
a Effective susceptibilities fp, gp, fs, and hs are given in arbitrary units, with a common normalization factor. Elements of the nonlinear susceptibility
tensor øIJK are absolute values in units of 10-22 m2 V-1. The phases of the susceptibilities are relative values, with hs constrained as real and
positive. See the text for details.
Geiger et al.28 and Maki et al.30 have described the use of a
quarter-wave retardation plate for single-valued determination
of second-order nonlinear susceptibilities of surfaces with in-
plane isotropy. The method involves rotation of the quarter-
wave plate to modulate the polarization state of the fundamental
radiation incident on the nonlinear surface. Using the analysis
and conventions of Maki et al.,30 the intensity of reflected second
harmonic radiation can be expressed as shown in eq 3.
polarized states with the plane rotated 45 and 135°, respectively,
from the p-direction. The complex amplitudes of the electric
field of the second harmonic radiation are Eq(45) ) (Es + Ep)/
x
x
2 and Eq(135) ) (Es - Ep)/ 2. Likewise, the effective
x
susceptibilities are fq(45) ) (fs + fp)/ 2 and fq(135) ) (fs - fp)/
x
2 and similarly for gq(45), gq(135) and hq(45), hq(135). Traces of
2ω(θ) for q45-, q135-, s-, and p-polarization states each provide
I
unique information on the effective susceptibilities fp, gp, hp,
and hs. Therefore, it is useful to simultaneously fit all four traces
with a unique set of susceptibilities (global fit). As noted above,
seven parameters are needed to describe the second-order
nonlinear response of an isotropic chiral surface. The work of
Geiger et al.28 and Maki et al.30 has shown that a unique
determination of all seven parameters can be made by means
of quarter-wave plate rotation measurements.
2
2
I
2ω ) (f12 + f2 )[(A2 + B2)2] + (g12 + g2 )[(C2 + D2)2] +
(h12 + h2 )[(A2 + B2)(C2 + D2)] + 2(f1g1 + f2g2)[(A2 -
2
B2)(C2 - D2) + 4ABCD] + 2(f1h1 + f2h2)[(A2 + B2)(AC +
BD)] + 4(f1g2 - f2g1)[(C2 - D2)AB - (A2 - B2)CD] +
2(f1h2 - f2h1)[(A2 + B2)(BC - AD)] + 2(g1h1 +
g2h2)[(C2 + D2)(AC + BD)] + 2(g1h2 - g2h1)[(C2 +
D2)(AD - BC)] (3)
3.3. Nonlinear Susceptibilities Measured by Achromatic
Wave Plate Rotation. An achromatic quarter-wave plate was
used to measure the susceptibilities of DHN, D-trp, L-trp, DD-
trp, and LL-trp at air/water interface, with the fundamental
wavelength λ ) 564 nm. For each system, four I2ω(θ) traces
were recorded by selecting s-, p-, q45-, and q135-polarization
states of the second harmonic. Hereafter we shall refer to these
traces as s-out, p-out, etc. The four traces were simultaneously
fit with seven parameters as described in section 3.2 (global
fit). Optimization was by minimizing the sum of the squares of
the deviations between calculated and observed (I2ω, θ) pairs
according to eq 3. Suitable Fresnel factors were included on
the right-hand side of eq 3 to account for the polarization
dependence of the transmission efficiency of the Pellin-Broca
prism. In Figures 3-5 we show measured and fitted I2ω(θ) traces
for DHN, D-trp, and L-trp. Similar results were obtained for
DD-trp and LL-trp. The agreement between measured and fitted
The quantities A-D specify the p- and s-polarized components
of the fundamental electric field (Ep and Es) incident on the
nonlinear surface. In the case where the fundamental radiation
incident on the wave plate is p-polarized with complex
amplitude Ep0 (and with Es0 ) 0), then Ep ) Ep0 (A + iB) and
0
Es ) Ep (C + iD), where A-D are given in eqs 4.
A ) cos(δ); B ) -sin(δ) cos(2θ); C ) 0;
D ) -sin(δ) sin(2θ) (4)
Here δ is the linear retardation of the wave plate (δ ) π/4 for
a quarter-wave plate) and θ is the angle between the fast axis
of the wave plate and the unit vector for p-polarized light (with
respect to the nonlinear surface). Right- and left-hand circularly
polarized states of the fundamental radiation are produced for
θ ) π/4 and θ ) -π/4, respectively.
The effective susceptibilities in eq 3 are fp, gp, and hp for
p-polarized second harmonic and fs, gs, and hs for s-polarized
second harmonic radiation. These quantities can be measured
by making a nonlinear fit of I2ω vs quarter-wave plate rotation
angle θ (for the selected polarization state), according to eq 3.
Such I2ω(θ) traces show characteristic features that allow quick
identification of a possible chiral contribution to the nonlinear
response. A nonzero chiral susceptibility hp (or fs) is immediately
apparent in producing an asymmetry of the I2ω(θ) trace about
the midpoint θ ) 180° of the range θ ) 0-360° of wave plate
rotation, for both p- and s-polarized second harmonic detection.
As shown by Geiger et al.,28 it is useful to measure I2ω(θ)
traces for q45- and q135- as well as s- and p-polarization states
of second harmonic radiation. Here q45 and q135 refer to plane-
I2ω(θ) traces was in all cases satisfactory. The parameters
obtained from the global fits are summarized in Table 1.
DHN was included in this study as an example of an achiral
system. Our results are indeed consistent with a vanishing chiral
response for DHN. Thus, the susceptibility hp was constrained
as zero in the fit for DHN, and this produced a satisfactory result
(Figure 3). However, certain features of the data for DHN
seemed at first to indicate the presence of a small chiral
response. In particular, there is a small asymmetry about θ )
180° in the p-out I2ω(θ) trace, noticeable for example at θ )
135 and 225° in Figure 3. The origin of this apparent chiral
response was traced to an artifact of the achromatic wave plate.
This was proven by rotating the wave plate 180° about its fast
axis, so the laser beam was incident on the opposite face of the
plate. The effect was to reverse the sign of the apparent chiral
response, such that the asymmetry about θ ) 180° in the p-out
I2ω(θ) trace was reversed. This artifact can be attributed to a

1024 J. Phys. Chem. B, Vol. 108, No. 3, 2004
Mitchell and McAloney
Figure 3. Global fits of wave plate rotation data I2ω(θ) for DHN at the air/water interface, for the indicated polarization states of second harmonic
radiation. (q1 and q2 represent q45 and q135 polarization states, respectively.) The wavelength of fundamental radiation was λ ) 564 nm. The second
harmonic intensity is shown in the same arbitrary units in all traces.
misalignment of the two parallel retardation plates (quartz and
magnesium fluoride) that comprise the achromatic retarder.
Ideally, the fast axis of one plate should be aligned with the
slow axis of the second plate. A misalignment of these axes
about the normal to the plates introduces chirality in the setup,
since the mirror plane defined by the normal to the plates and
the optic axis is removed by such a misalignment.
We characterized the achromatic wave plate by recording
θ-rotation traces Iω(θ) of the intensity of fundamental radiation
transmitted by the wave plate in s-, p-, q45- and q135-polarization
states, with p-polarized radiation incident on the wave plate.
Here the polarization states are identical with those defined in
section 3.2. The data were fit with four parameters that specify
the details of the model of misaligned retardation plates: ∆θ is
the angle between the fast and slow axes of the two plates, δ′
(δ) is the linear retardation of the magnesium fluoride (quartz)
plate, and θ0 gives the offset between the experimental variable
and the quantities A-D are given in eqns 8-11.
A ) cos(δ′) cos(δ) + sin(δ′) sin(δ) cos(2∆θ)
(8)
B ) -cos(δ′) sin(δ) cos(2θ) - sin(δ′) cos(δ) cos(2θ′) (9)
C ) -sin(δ) sin(δ′) sin(2∆θ) (10)
D ) -sin(δ′) cos(δ) sin(2θ′) - sin(δ) cos(δ′) sin(2θ) (11)
The amplitudes of the electric fields for q45- and q135
polarization states are related to Ep and Es as described in section
3.2. Intensities of the transmitted radiation in the various
polarization states are proportional to the absolute squares of
the corresponding electric field amplitudes. Note that eqs 8-11
become equivalent to eqs 4 when the misalignment ∆θ goes to
zero.
Measured traces of Iω(θ) for s-, p-, q45-, and q135-polarization
states were simultaneously fit with four parameters (∆θ, δ′, ∆δ,
and θ0), in the manner described above for I2ω(θ) traces (global
fit). The fits to the data were excellent and yielded optimized
parameters (all in angular units of deg): ∆θ ) 5, δ′ ) 85.7,
∆δ ) -1, and θ0 ) 5, for λ ) 564 nm. Although this set of
parameters is not unique, it nevertheless allows a useful
correction to be applied in the analysis of the I2ω(θ) data. A
careful investigation showed that it was unlikely that the
nonunique nature of the correction introduced significant errors
in the analysis. The correction is intended to compensate for
the small chiral artifact introduced in the data by the misalign-
ment of the elements of the achromatic wave plate. We applied
this correction by using eqs 8-11 in place of eqs 4 to calculate
the quantities A-D in eq 3. The parameters in eqs 8-11 were
fixed as given above; i.e., no additional free parameters were
introduced. Including this correction in the fits produced
markedly better results, as was apparent by eye and also from
-
θobs corresponding to the angular setting of the wave plate and
the angle θ′ between the fast axis of the magnesium fluoride
retardation plate and the p-polarization direction. Additional
quantities that depend on these parameters are defined in eqs
5-7.
θobs ) θ′ - θ0
(5)
θ ) θ′ - π/2 + ∆θ
δ ) δ′ - π/4 + ∆δ
(6)
(7)
θ is the angle between the fast axis of the quartz retardation
plate and the p-polarization direction. With this model the s-
and p-polarized components of the transmitted electric field are
0
0
0
Ep ) Ep (A + iB) and Es ) Ep (C + iD), where Ep is the
amplitude of the p-polarized radiation incident on the wave plate

Optical Activity of Tryptophan Derivatives
J. Phys. Chem. B, Vol. 108, No. 3, 2004 1025
Figure 4. Global fits of wave plate rotation data I2ω(θ) for D-trp at the air/water interface, for the indicated polarization states of second harmonic
radiation. (q1 and q2 represent q45 and q135 polarization states, respectively.) The wavelength of fundamental radiation was λ ) 564 nm. The second
harmonic intensity is shown in the same arbitrary units in all traces.
the magnitude of the sum of the squares of the deviations
between measured and fitted traces. The correction was applied
to all of our results and is included in Figures 3-6 and also in
Table 1.
not to show the corresponding mirror-image asymmetry (Figure
6). Thus, by using eqs 3 and 4 to fit the individual p-out traces,
it was seen that the chiral parameters (hp1 and hp2) did not show
the expected sign change for enantiomers. However, by inclusion
of the correction for the artifact in the fit, the parameters were
brought closer to the expected relation. The correction had a
much smaller effect on the fitted parameters for individual s-out
traces, and thus, the chiral component of the p-out data is more
sensitive to the distorting effect of the artifact.
It is useful to consider circular difference (CD) and linear
difference (LD) effects in SHG from chiral surfaces.29 These
are second harmonic intensity differences for left-hand and right-
hand circularly polarized fundamental radiation (CD) and for
q45- and q135-polarized fundamental radiation (LD). The plane-
polarized states q45 and q135 have been defined in section 3.2.
Expressions for the normalized intensity differences in terms
of effective susceptibilities of the chiral surface are shown in
eqs 12 and 13.29
No attempt has been made to correct the measurements for
the presence of a small background SHG signal that was
observed for the buffer solution with no added tryptophan
derivative or DHN. The associated error should be small, since
the background signal was approximately 10-fold weaker than
the working signals (Figure 2). The effective susceptibilities in
Table 1 have been uniformly scaled by a normalization constant
that was obtained from measurement of Rpp for z-quartz (section
3.1). For completeness, we also show in Table 1 the elements
of the nonlinear susceptibility tensors, as absolute values in units
of 10-22 m2/V. The detailed expressions that relate the effective
susceptibilities to the elements of the nonlinear susceptibility
tensor have been given by Maki et al.19 We assumed that the
index of refraction of the monolayer film was the same as that
of air, for both the fundamental and second harmonic frequen-
cies. It is difficult to assess the uncertainties present in the results
in Table 1. From repeated measurements on several of the trp
systems, it was seen the results were subject to significant
variability, but the main features were reproducible in the range
(30%.
4[(f2 - g2)h1 - (f1 - g1)h2]
f12 + f22 + g12 + g22 + h12 + h22 - 2(f1g1 + f2g2)
∆ICD
)
(12)
-4[(f1 + g1)h1 + (f2 + g2)h2]
f12 + f22 + g12 + g22 + h12 + h22 + 2(f1g1 + f2g2)
∆ILD
)
In Figures 4 and 5, the s-out I2ω(θ) traces for D-trp and L-trp
show well-defined asymmetry about θ ) 180°. This asymmetry
shows a mirror-image relation for D-trp and L-trp enantiomers,
as expected for chiral isotropic surfaces. Similar results were
obtained for LL-trp and DD-trp, as shown in Figure 6. In contrast,
no asymmetry appears in the s-out trace for the achiral system
DHN (Figure 3), as expected. These observations indicate that
the assumption of in-plane isotropy of the surfaces is justified.
(13)
The intensity differences are normalized to the average second
harmonic intensity for the two polarization states of the
fundamental. We calculated these quantities from the effective
susceptibilities for the trp systems in Table 1, and the results
are given in Table 2. ∆ICD and ∆ILD are given for both s- and
p-polarized second harmonic radiation.
Chiral asymmetry is generally less apparent in the p-out I2ω
-
(θ) traces for the trp systems compared with the s-out traces.
An exception is LL-trp, but here the enantiomer DD-trp appears
One expects to find a sign change in the intensity differences
for the enantiomeric pairs (D-trp, L-trp) and (DD-trp, LL-trp), with

1026 J. Phys. Chem. B, Vol. 108, No. 3, 2004
Mitchell and McAloney
Figure 5. Global fits of wave plate rotation data I2ω(θ) for L-trp at the air/water interface, for the indicated polarization states of second harmonic
radiation. (q1 and q2 represent q45 and q135 polarization states, respectively.) The wavelength of fundamental radiation was λ ) 564 nm. The second
harmonic intensity is shown in the same arbitrary units in all traces.
Figure 6. Fits of single traces of wave plate rotation data I2ω(θ) for DD-trp and LL-trp at the air/water interface, for the indicated polarization states
of second harmonic radiation. The wavelength of fundamental radiation was λ ) 564 nm. The second harmonic intensity is shown in the same
arbitrary units in all traces.
the absolute magnitudes of the intensity differences preserved.
A sign change is observed for most cases in Table 2, with the
exception of ∆ILD (s-out) for (D-trp, L-trp). However, the
absolute magnitudes of the intensity differences are generally
not in close agreement within the enantiomeric pairs. Exceptions
are the p-out intensity differences ∆ICD and ∆ILD for (DD-trp,
LL-trp), which show approximately conserved magnitudes and
the expected sign changes. A further consistent feature of the
results was noted when ∆ICD (s-out) was evaluated from
effective susceptibilities obtained by fitting single I2ω(θ) traces.

Optical Activity of Tryptophan Derivatives
J. Phys. Chem. B, Vol. 108, No. 3, 2004 1027
TABLE 2: Circular and Linear Differences of Second
Harmonic Intensity for Several Tryptophan Derivatives
Adsorbed at the Air/Water Interface, from Global Fits of
Quarter-Wave Plate Rotation Data for λ ) 564 nma
D-trp
L-trp
DD-trp
LL-trp
s-out
∆ICD
∆ILD
p-out
∆ICD
∆ILD
-0.13
0.35
0.077
-0.18
0.22
0.17
0.086
-0.035
-0.33
0.023
1.3
-0.18
-0.38
0.14
0.36
-0.16
a Intensity differences for s- and p-polarized components of second
harmonic radiation. See the text for details.
TABLE 3: Circular Differences of Second Harmonic
Intensity for L-trp and LL-trp Adsorbed at the Air/Water
Interface, from Fits of Single Traces of Quarter-Wave Plate
Rotation Data for Several Fundamental Wavelengthsa
wavelength/nm
542
564
581
603
L-trp: ∆ICD (s-out)
LL-trp: ∆ICD (s-out)
0.45
0.38
0.41
0.41
0.49
0.33
0.09
0.01
a Intensity differences for s-polarized component of second harmonic
radiation. See the text for details.
Figure 7. (a) Tilt (θ) and twist (ψ) angles of indole (C8NH7) relative
to laboratory coordinate axes, with Z the normal to the air/water
interface. The dark circle represents the nitrogen atom of indole. (b)
Hypothetical structure of Boc-D-trp at the air/water interface, viewed
from the +Z direction. Note that the indicated bond torsion is correlated
with the twist angle ψ. (c) Chemical structure of Boc-D-trp. The
asymmetric carbon center is marked with an asterisk.
These effective susceptibilities are distinct from those obtained
from the global fits. Being associated with single s-out traces,
they are not subject to the distorting effect of the artifact that
affects mainly the p-out traces (see above). The ∆ICD (s-out)
evaluated from single traces are -0.33 (D-trp), 0.37 (L-trp),
-0.33 (DD-trp), and 0.32 (LL-trp), which show enantiomeric
effects as expected for isotropic chiral surfaces. Such measure-
ments of ∆ICD (s-out) are robust because the signal is relatively
large and slowly varying in the region of the circular polarization
states of the fundamental (θ ) 45, 135°, ...). The situation is
much less favorable for the same measurements on p-out traces,
since the p-out signal intensities are close to zero and quickly
varying in these regions, as shown in Figure 5, for example.
For similar reasons, it is difficult to make accurate measurements
of ∆ILD on single s-out or p-out I2ω(θ) traces. Except for ∆ICD
(s-out), our most reliable measurements of the intensity differ-
ences are from the global fits (Table 2). Our preferred
measurements of ∆ICD (s-out) are from the fits to single I2ω(θ)
traces.
4. Discussion
In the following we consider the implications of our results
for the interpretation of the SHG-OA effects in terms of a
microscopic model. Our central result is a close similarity in
SHG-OA for the like enantiomers L-trp and LL-trp on one hand
and D-trp and DD-trp on the other. The presence of two trp
residues in the dipeptide has only a small effect on the observed
SHG-OA.
As shown in Figure 2, there is a correspondence between
the absorption spectrum of Boc-trp in solution and the second
harmonic spectrum of the air/solution interface. This indicates
that the nonlinear response with the fundamental wavelength λ
∼ 560 nm is resonantly enhanced at the two-photon level. The
resonance is associated with the indole chromophore of tryp-
We measured ∆ICD s-out intensity differences for L-trp and
LL-trp for several fundamental wavelengths in the range 542-
603 nm, by using effective susceptibilities obtained from fits
of single s-out I2ω(θ) traces. The results are summarized in Table
3, which incorporates minor corrections for the artifact in the
achromatic wave plate. L-trp and LL-trp show comparable values
of ∆ICD (s-out) in this range. In both cases there is little variation
with fundamental wavelength except for an abrupt drop in ∆ICD
(s-out) at the longest wavelength (603 nm).
1
1
tophan, with well-known electronic transitions to La and Lb
excited states in the region near 280 nm.32 The structure of
indole is shown in Figure 7. The 1La and 1Lb excitations can be
described as transitions within a planar π-electron system and
are thus polarized in the plane of the indole molecule. This is
a significant point when considering the origin of SHG-OA of
the trp systems. A planar chromophore is of course achiral.
However, the proximity of the chromophore to the asymmetric
carbon of tryptophan means that the chromophore is in a chiral
environment.
In general, our results show a close similarity between Boc-
trp and Boc-trp-trp. Let us emphasize that the bulk solutions
that were used were well characterized with respect to chemical
composition. Despite the similarities, there are certain features
of the SHG-OA results that allow a clear distinction to be made
between Boc-trp and Boc-trp-trp. These features are apparent
in the I2ω(θ) traces (Figures 4-6) and also in the details of the
effective susceptibilities in Table 1. We found that the I2ω(θ)
traces for L-trp did not vary with the concentration of the bulk
solution, in a range where the surface density of L-trp changed
by a factor ∼2. This suggests that the I2ω(θ) traces were not
affected by interactions between the L-trp molecules.
Our successful analysis of all I2ω(θ) traces strictly in terms
of an electric-dipole model, with gs ) 0 and hp fs, suggests
that magnetic effects are not important for the trp systems.
Recently, attention has been drawn to possible structural origins
of SHG-OA,26,27 within the framework of electric-dipole
interactions. Note that an unambiguous analysis of SHG-OA
relies on azimuthal isotropy of the surface (i.e., the surface must
be rotationally invariant) or the rotational symmetry of the
surface must be explicitly considered.33 For an isotropic surface,
asymmetry in the twist-angle (ψ) distribution of an achiral

1028 J. Phys. Chem. B, Vol. 108, No. 3, 2004
Mitchell and McAloney
chromophore about its polar axis may impart chirality to the
surface and, thereby, give rise to SHG-OA. The polar axis
defines the tilt angle (θ) of the chromophore relative to the
surface normal. The tilt (θ) and twist (ψ) angles are illustrated
in Figure 7 for the case of an indole chromophore. Simpson
and co-workers26,27 have suggested that asymmetry in the twist-
angle distribution of an achiral chromophore could arise from
the presence of a chiral center in the molecule. The chromophore
may be sufficiently removed from the chiral center that no direct
electronic coupling occurs, and the chromophore remains
effectively achiral. The chiral center has the effect of directing
assembly of the chromophore at the surface. Such an effect could
be important for the trp systems studied in this work, but we
believe that this is unlikely. To illustrate how the effect could
operate, consider the model structure for Boc-trp at the air/water
interface shown in Figure 7b. Note that the twist angle of the
indole chromophore is related to the torsion angle indicated in
Figure 7b. In particular, asymmetry in the twist angle distribution
implies restricted rotation with respect to the torsion. Such a
restricted rotation is unlikely considering the probable magnitude
of the torsional barrier.34 One could argue that intermolecular
interactions in the surface layer might effectively lock the
conformation, but this seems unlikely in view of the fluid nature
of the air/water interface.
the dipeptide Boc-trp-trp can be understood by considering that
the indole groups may be too far apart to allow significant
coupling. We found no indication of SHG-OA for the peptide
gramicidin, which includes several trp residues. This shows that
the presence of trp is not a sufficient condition for observation
of SHG-OA. The conformation and orientation of the peptide
at the interface are additional important factors. In future work
it is planned to continue this investigation of the origin of SHG-
OA in these biomolecular/surface systems.
5. Conclusion
We have measured the second-order nonlinear susceptibility
tensors for monolayer films of two enantiomerically pure
tryptophan derivatives and for an achiral hydroxynaphthyl
derivative at the air/water interface, for the fundamental
wavelength λ ) 564 nm. A satisfactory analysis of data from
quarter-wave plate rotation measurements was obtained by using
the assumption that only electric-dipole effects contributed to
the nonlinear response. Optical isomers of mono- and ditryp-
tophan compounds show approximately equal magnitudes and
opposite signs of the chiral components of the nonlinear
susceptibilities, as expected for isotropic surface films. Spectra
of the second harmonic efficiency versus fundamental wave-
length for the tryptophan derivatives showed resonance en-
hancement at the two-photon level, associated with the indole
chromophore of tryptophan. An interpretation of our results in
terms of an asymmetric orientational distribution of the (achiral)
indole chromophore appears not to be applicable. Instead, it is
suggested that the nonlinear chiral response may have its origin
in dipole coupling between chromophores, as in the case of
circular dichroism in linear spectroscopy of tryptophan deriva-
tives. From a practical point of view, it is significant that the
nonlinear response is observed in a lower energy spectral region
(λ/2 ∼ 280 nm) compared with the strongest chiral effects in
linear spectroscopy (λ < 230 nm). From a comparison of results
for mono- and ditryptophan derivatives, no evidence was found
for excitonic coupling effects in the nonlinear chiral response.
The peptide gramicidin showed no detectable optical activity
in second harmonic reflection from the air/water interface. This
shows that the presence of one or more tryptophan residues in
a peptide or protein is not a sufficient condition for observation
of optical activity in second harmonic reflection.
The above considerations lead to the suggestion that the origin
of SHG-OA for the trp systems is an intrinsic molecular chiral
response. Of course, Boc-trp and Boc-trp-trp are chiral mol-
ecules, but the asymmetric carbon centers are separated from
the indole chromophores by a methylene (-CH2-) linking
group (see illustration in Figure 7). These compounds neverthe-
less display strong circular dichroism (CD) in linear spectros-
copy in the spectral region below 230 nm35 and generally much
1
1
weaker CD effects in the region of the La and Lb excitations
of indole near 280 nm.36 These CD effects in linear spectroscopy
of tryptophanyl compounds have their origin in dipole-dipole
coupling between similar or dissimilar chromophores including
indole and amide groups that are present in a chiral arrangement.
The excited electronic states and associated intramolecular
interactions that are responsible for strong circular dichroic
effects in linear spectroscopy of tryptophan residues should also
be significant for SHG-OA.
A coupled chromophore, electric-dipole mechanism of SHG-
OA is potentially important for the dipeptide Boc-trp-trp, due
to the presence of indole chromophores in spatial proximity. In
the work of Frey and co-workers,7 significant SHG-OA was
observed for Boc-trp-trp at air/water interface but not for Boc-
trp. These observations are suggestive of coupling between trp
residues. However, in this study we found closely similar SHG-
OA for Boc-trp and Boc-trp-trp, in disagreement with the report
of Frey and co-workers. The origin of this discrepancy is not
clear. Our observations indicate that excitonic coupling between
indole chromophores is not a dominant mechanism of SHG-
OA for Boc-trp-trp. Instead, the proximity of indole to an
asymmetric carbon center may be the significant factor, as in
the case of optical activity in linear spectroscopy. We believe
that coupling between transition dipoles on indole and amide
(NHCO) chromophores could give rise to the observed SHG-
OA effects. These dissimilar chromophores are present in a
chiral arrangement (see the chemical structure of Boc-trp in
Figure 7). Such a mechanism for SHG-OA, involving coupling
of dissimilar chromophores, is a generalization of the mechanism
discussed by Hache et al.24 and by Belkin et al.,25 involving
the coupling of two identical chromophores. The relative
unimportance of interactions between indole chromophores in
Acknowledgment. The authors thanks Prabhat Arya and
Michael Barnes for chemical synthesis of Boc-trp-trp and Linda
Johnston for assistance in the work on gramicidin D.
References and Notes
(1) Petralli-Mallow, T.; Wong, T. M.; Byers, J. D.; Yee, H. I.; Hicks,
J. M. J. Phys. Chem. 1993, 97, 1383.
(2) Byers, J. D.; Yee, H. I.; Hicks, J. M. J. Chem. Phys. 1994, 101,
6233.
(3) Byers, J. D.; Yee, H. I.; Petralli-Mallow, T.; Hicks, J. M. Phys.
ReV. B 1994, 49, 14643.
(4) Kauranen, M.; Verbiest, T.; Maki, J. J.; Persoons, A. J. Chem. Phys.
1994, 101, 8193.
(5) Verbiest, T.; Kauranen, M.; Persoons, A.; Ikonen, M.; Kurkela, J.;
Lemmetyinen, H. J. Am. Chem. Soc. 1994, 116, 9203.
(6) Stolle, R.; Loddoch, M.; Marowsky, G. Nonlinear Opt. 1994, 8,
79.
(7) Crawford, M. J.; Haslam, S.; Probert, J. M.; Gruzdkov, Y. A.; Frey,
J. G. Chem. Phys. Lett. 1994, 229, 260.
(8) Fasman, G. D., Ed.; Circular dichroism and the conformational
analysis of biomolecules; Plenum Press: New York, 1996.
(9) Shen, Y. R. Annu. ReV. Phys. Chem. 1989, 40, 327.
(10) Smiley, B. L.; Vogel, V. J. Chem. Phys. 1995, 103, 3140.
(11) Vogel, V.; Smiley, B. L. Proc. SPIE 1992, 1922, 86.
(12) Smiley, B. L.; Vogel, V. Proc. SPIE 1994, 2125, 59.

Optical Activity of Tryptophan Derivatives
J. Phys. Chem. B, Vol. 108, No. 3, 2004 1029
(13) Petralli-Mallow, T. P.; Plant, A. L.; Lewis, M. L.; Hicks, J. M.
Langmuir 2000, 16, 5960.
(14) Kriech, M. A.; Conboy, J. C. J. Am. Chem. Soc. 2003, 125, 1148.
(15) (a) Rinuy, J.; Brevet, P. F.; Girault, H. H. Biophys. J. 1999, 77,
3350. (b) Perrenoud-Rinuy, J.; Brevet, P. F.; Girault, H. H. Phys. Chem.
Chem. Phys. 2002, 4, 4774.
(16) Watry, M. R.; Richmond, G. L. J. Phys. Chem. B 2002, 106, 12517.
(17) Kim, G. K.; Gurau, M. C.; Lim, S. M.; Cremer, P. S. J. Phys. Chem.
B 2003, 107, 1403.
(18) (a) Wang, J.; Clarke, M. L.; Zhang, Y.; Chen, X.; Chen, Z.
Langmuir 2003, 19, 7862. (b) Wang, J.; Buck, S. M.; Even, M. A.; Chen,
Z. J. Am. Chem. Soc. 2002, 124, 13302. (c) Wang, J.; Buck, S. M.; Chen,
Z. J. Phys. Chem. B 2002, 106, 11666.
(19) Maki, J. J.; Kauranen, M.; Persoons, A. Phys. ReV. B 1995, 51,
1425.
(20) Kauranen, M.; Verbiest, T.; Persoons, A. J. Mod. Opt. 1998, 45,
403.
(21) Hache, F.; Mesnil, H.; Schanne-Klein, M. C. J. Chem. Phys. 2001,
115, 6707.
(24) Hache, F.; Boulesteix, T.; Schanne-Klein, M. C.; Alexandre, M.;
Lemercier, G.; Andraud, C. J. Phys. Chem. B 2003, 107, 5261.
(25) Belkin, M. A.; Shen. Y. R.; Flytzanis, C. Chem. Phys. Lett. 2002,
363, 479.
(26) Simpson, G. J. J. Chem. Phys. 2002, 117, 3398.
(27) Simpson, G. J.; Perry, J. M.; Ashmore-Good, C. L. Phys. ReV. B
2002, 66, 165437.
(28) Geiger, F.; Stolle, R.; Marowsky, G.; Palenberg, M.; Felderhof, B.
U. Appl. Phys. B 1995, 61, 135.
(29) Maki, J. J.; Verbiest, T.; Kauranen, M.; Van Elshocht, S.; Persoons,
A. J. Chem. Phys. 1996, 105, 767.
(30) Maki, J. J.; Kauranen, M.; Verbiest, T.; Persoons, A. Phys. ReV. B
1997, 55, 5021.
(31) Roberts, D. A. IEEE J. Quantum Electron. 1992, 28, 2057.
(32) Callis, P. R. J. Chem. Phys. 1991, 95, 4230.
(33) Verbiest, T.; Kauranen, M.; Van Rompaey, Y.; Persoons, A. Phys.
ReV. Lett. 1996, 77, 1456.
(34) Electronic structure calculations on Boc-trp using the semiempirical
AM1 Hamiltonian indicated that the torsional barrier was less that 2 kcal/
mol, for a range of model geometries.
(22) Schanne-Klein, M. C.; Hache, F.; Brotin, T.; Andraud, C.; Collet,
A. Chem. Phys. Lett. 2001, 338, 159.
(23) Schanne-Klein, M. C.; Boulesteix, T.; Hache, F.; Alexandre, M.;
Lemercier, G.; Andraud, C. Chem. Phys. Lett. 2002, 362, 103.
(35) Auer, H. E. J. Am. Chem. Soc. 1973, 95, 3003.
(36) Strickland. E. H. CRC Crit. ReV. Biochem. 1974, 2, 113.