Characterizing the Photoinduced Switching Process of a

Aug 31, 2012 - Australian Nuclear Science and Technology Organisation (ANSTO), Locked Bag 2001, Kirrawee .... information on the spiropyran photoinduc...
3 downloads 0 Views 2MB Size
Download PDF
Article pubs.acs.org/Langmuir

Characterizing the Photoinduced Switching Process of a Nitrospiropyran Self-Assembled Monolayer Using In Situ Sum Frequency Generation Spectroscopy Tamim A. Darwish,*,† Yujin Tong,§,∥ Michael James,†,‡ Tracey L. Hanley,† Qiling Peng,§ and Shen Ye*,§ †

Australian Nuclear Science and Technology Organisation (ANSTO), Locked Bag 2001, Kirrawee DC, NSW 2232, Australia School of Chemistry, The University of New South Wales, Kensington, NSW 2052, Australia § Catalysis Research Center, Hokkaido University, Sapporo 001-0021, Japan ∥ Fritz-Haber-Institut der Max-Planck-Gesellschaft, Faradayweg 4-6, 14195 Berlin, Germany ‡

S Supporting Information *

ABSTRACT: Sum frequency generation (SFG) vibrational spectroscopy is employed to investigate the reversible, photoinduced spiro→merocyanine isomerization of a self-assembled monolayer, the result of attachment of nitrospiropyran to a gold surface using a dithiolane anchoring group. The attachment of these molecular “alligator clips” to spiropyran molecules provide an easily accessible method to self-assemble a robust monolayer of spiropyran on a gold surface, which allows photoswitching of the spiropyran units. Probing the symmetric and antisymmetric stretching modes of the nitro group allows the determination of the structural orientation of the charged moiety with respect to the surface normal as well as the isomerization rates under photoinduced switching conditions. The photoisomerization of the spiropyran SAM on the gold surface is much faster than the rates of switching spiropyrans in a solid crystalline form, and the rate of thermal relaxation of the opened to closed form in this study is found to be on the same time scale as the relaxation of spiropyran when present in solutions with polar solvents.

■

INTRODUCTION Photochromic spiropyran units respond to light and undergo a reversible isomerization between colorless spiro (SP) and colored merocyanine (MC) forms. The SP-forms are threedimensional, inert, and colorless, while the MC-forms are planar, are zwitterionic, possess a large dipole moment, and are colored. The large and the reversible change in these molecular properties has led to great interest in their use, not only for ophthalmic lenses,1,2 but also in cell imaging,3 molecular sensing,4,5 programmable molecular logic devices,6 photoprogrammable organic light-emitting diode (OLEDs),7 and molecular switches.8 The response of photochromic molecules such as spiropyran to external stimuli such as heat, UV or visible light depends critically on the interactions with their environment, as well as factors such as molecular packing, charge, and orientation. To date, most of the experimental studies on these molecules have been performed in solution, where the open MC-form is thermally unstable at room temperature. After photocoloration of the parent spiropyran compound, the MC-form molecules undergo a thermal back-reaction, returning to the more thermodynamically stable closed SP-form. While the switching properties of free spiropyran molecules in solution are well understood, comprehending how these molecules and their © 2012 American Chemical Society

photoisomerizing adducts geometrically assemble and switch in self-assembled monolayers (SAMs) requires significant further study at a molecular level. Relatively few studies have been performed on spiropyran monolayers assembled on solid surfaces. In most instances the photoswitchability of these surfaces was indirectly observed and characterized by studying their wettability9−11 and their electrochemical properties.12−14 Despite the large difference in the polarity between the open and closed states, (with dipole moment increasing from 15 × 10−30 C m in the SP-form to 60 × 10−30 C m in the MCform),15 spiropyran monolayer surfaces only show small changes in water contact angle (ΔΘ = 9°−14°) between the SP-form and the charged MC-form.11,16 The reason for such a small difference is likely to be not only due to the isomerization yield but also due to the orientation of the charged hydrophilic moieties and how they are exposed at the air/surface interface in the SAMs. Despite the widespread use of water contact angle measurements to infer molecular orientation in self-assembled films and functionalized surfaces, it has recently been demonstrated that this technique and others such as X-ray Received: May 30, 2012 Revised: August 13, 2012 Published: August 31, 2012 13852

dx.doi.org/10.1021/la302204f | Langmuir 2012, 28, 13852−13860

Langmuir

Article

complicated by the need for multiple protection and deprotection synthetic steps to avoid the oxidation of the thiol group. In our current study, we describe an easily accessible method to assemble nitrospiropyran units on a gold surface using a dithiolane anchoring group which does not suffer the above limitations of thiols. Cyclic disulfides such as 1,2-dithiolane [S(CH2)3S] have been widely used as molecular “alligator clips” to form SAMs on gold and other metal surfaces,28−30 and our study below demonstrates the ease with which these molecules may be synthesized and used to produce robust, photoswitchable molecular surfaces.

reflectometry should be used with some caution due to the susceptibility of these systems to adsorb nanoscale water layers at the air interface.17,18 Vibrational spectroscopy is an appropriate technique that can directly probe the structure and conformation of the surfacebound molecules. Previous studies on probing the photoswitchability of spiropyran or related molecules on surfaces using traditional infrared (IR) based techniques have been limited to porous silicon surfaces11 or to multilayer films19 due to the low sensitivity of these linear vibrational techniques to monolayer concentrations. Despite this, no detailed structural information on the spiropyran photoinduced isomerization process on the surfaces has been obtained. In our current study sum-frequency generation (SFG) vibrational spectroscopy was employed to investigate the photoinduced SP→MC isomerization and resulting molecular orientations in SAMs on a gold surface (Scheme 1). SFG vibrational spectroscopy is a second-

■

EXPERIMENTAL SECTION

Synthesis of SP-LA and Sample Preparation. The molecule used to form the SAMs used in this SFG study, 2-(3′,3′-dimethyl-6nitro-3′H-spiro[chromene-2,2′-indol]-1′-yl)ethyl (1,2-dithiolane-3)pentanoate (SP-LA), was synthesized by a one-step reaction from commercially available reagents according to the procedure of Tomasulo et al.31 In brief, the generic spiropyran 1-(2-hydroxyethyl)-3,3-dimethylindolino-6′-nitrobenzopyrylospiran (TCI America) was treated with (±)-α-lipoic acid in the presence of N,N′dicyclohexylcarbodiimide and 4-dimethylaminopyridine, and the resulting ester was isolated. A thin gold film with a thickness of ∼200 nm was prepared by thermal evaporation onto a Ti-primed glass slide (25 mm × 25 mm × 1 mm) at an evaporation rate of 6 nm/min. The gold surface was annealed by a propane gas/oxygen flame before further use. The goldcoated glass slides were then immersed in a SP-LA dichloromethane solution (0.5 mM) for 12 h at ambient temperature to facilitate selfassembly of the spiropyran monolayer. The modified gold slides were rinsed with dichloromethane several times and dried with Ar gas before SFG measurement. Irradiation of SAM Surfaces. UV and visible light exposure was carried out by using a UV lamp (350 nm, ∼3 mW/cm2) and a visible light lamp (580 nm, ∼2 mW/cm2), respectively. The lamp powers were assessed and were not found to cause any unfavorable SAM degradation by monitoring SFG signals that remained intrinsically constant during the light exposure time. The photochemical isomerization reactions of the nitrospiropyran monolayer on the gold surface were monitored under in situ conditions by SFG spectroscopy. The SFG chamber was continuously purged with N2 gas to keep out moisture and any ozone that might be generated over the gold surface due to the UV irradiation. SFG and FTIR Characterization. The details of our broad-band SFG system are described elsewhere.32−34 Briefly, a regenerative amplifier (SpiteFire PRO), seeded by a titanium:sapphire oscillator (Mai Tai) and pumped by a Nd:YLF laser (EMPower), generates a 2.2 mJ laser pulse at 800 nm with a 120 fs duration and a repetition rate of 1 kHz; 1.0 mJ of the output was used to pump an optical parametric amplifier system (TOPAS, Light Conversion Ltd.) to generate IR pulses that are tunable between 2.5 and 8 μm with a full width at halfmaximum of approximately 200 cm−1. The remaining output from the amplifier was sent to a homemade spectral shaper to generate a narrow-band pulse (ca. 10 cm−1) at 800 nm to improve the spectral resolution. In this experiment, the SFG measurements were mainly carried out in the IR frequency regions of C−H stretching (2800 and 3150 cm−1) and fingerprint (1300 and 1650 cm−1). The incident angles of the visible and IR beams were 70° and 50°, respectively. The intensity of the visible and infrared lasers at the sample surface were adjusted to 4 and 2 μJ per pulse to eliminate possible sample damage. No significant effects were found by the laser surface irradiation on the switching properties of the samples, which is due to both the wavelengths of the visible (800 nm) and infrared beams being out of the range required to trigger the SP/MC isomerization. All SFG spectra were recorded with a polarization combination of p-SFG, pvisible, and p-IR; spectra were normalized by an SFG spectrum from bare gold to correct the effect of the line shapes of the infrared beam. Multiple IR center frequencies (every 100 cm−1 between 1300 and

Scheme 1. Photoswitching between SP- and MC-Forms for 2-(3′,3′-Dimethyl-6-nitro-3′H-spiro[chromene-2,2′-indol]1′-yl)ethyl (1,2-dithiolane-3)pentanoate (SP-LA) Tethered to Gold Surfaces Used in this Studya

Three dihedral angles (α, β, γ) in the MC-form can alter to form several structural isomers. The MC-form here is shown as the TTC isomer (trans−trans−cis conformation of the three bonds constituting the methine bridge). a

order nonlinear optical technique in which an infrared and a visible laser are overlapped spatially and temporally at the interface and the resulting emission with the frequency equal to the sum of the two incidents is recorded.20 When the infrared frequency is resonant with a vibrational mode of the surface molecules, the sum frequency field is resonantly enhanced. SFG has several advantages over conventional vibrational spectroscopic techniques, such as surface selectivity and submonolayer sensitivity. It has been widely employed to study the molecular structure and dynamics of gas/liquid, gas/solid, and liquid/solid interfaces.21−23 Recently, SFG has also been employed to study the behaviors of some cis−trans photoswitchable surfaces, such as an azobenzene-functionalized SAM24 and azobenzenefunctionalized lipids at the air−water interface.25 In previous studies of spiropyran-functionalized SAMs, the spiropyran molecules were usually covalently attached to a suitable substrate via an anchoring group and an additional spacer unit (linker).9,10,12,18 A number of studies have reported the attachment of spiropyran to gold surfaces;13,26,27 however, these molecular surfaces were also prepared in multiple synthetic steps in order to connect a thiol-anchored linker to the spiropyran molecule. The chemistry of thiols is often 13853

dx.doi.org/10.1021/la302204f | Langmuir 2012, 28, 13852−13860

Langmuir

Article

1800 cm−1) were used in order to cover the whole spectral range of interest for the static SFG measurement. However, for the kinetic measurement, the IR center frequency was fixed at 1475 cm−1, which can cover the frequency region for symmetric and antisymmetric stretching modes of NO2 group, in order to have higher time resolution to follow the structural evolution process. Generally, each SFG spectrum was acquired over a 2 min period, which allowed sufficient time resolution to follow the structural evolution process. To check the reproducibility of the results, three samples were prepared and measured sequentially. Both the spectral intensity and kinetic results were reproducible. Complementary FTIR measurements were carried out on the bulk form of SP-LA in transmission using a Bio-Rad FTIR-896 spectrometer equipped with a MCT detector. About 1 mg of SP-LA was mixed with ∼150 mg of KBr powder to prepare the pellet for characterization. All experiments were carried out at room temperature (24 °C).

can only be observed when there is a net orientation of the NO2 dipole from the ordered molecular structure. To discuss the conformation of SP-LA at the interface, we will focus on the symmetric (νsNO2, 1335 cm−1) and antisymmetric (νasNO2, 1512 cm−1) stretching modes, partly due to the fact that NO2 modes show the most intense peak in the SFG spectrum due to its strong dipole moment (Figure 1a). In addition, unlike phenyl, methyl, or methylene groups, the SP-LA molecule possesses only one NO2 group, making the analysis more straightforward. Since the dipole moments of the symmetric and antisymmetric modes of NO2 group are perpendicular to each other, any orientational change of the NO2 group will be reflected in the intensity change of the two vibrational modes in the SFG spectra. Moreover, since the NO2 group and the nitrophenyl ring are always in the same plane, the orientational change of the former can be regarded as a marker to characterize the orientational change of the whole nitrophenyl and pyran groups, which undergo dramatic conformational change during the photoinduced isomerization. Initial State of the Closed Form of SP-LA on the Surface. The SFG spectrum of a freshly prepared SP-LA SAM (Figure 1a) showed a strong NO2 symmetric peak and a small antisymmetric peak. Since p-polarized laser beams are used for the SFG measurement, this means that only the vibrational modes that contain components of the dipole moment perpendicular to the surface plane can be monitored, while components of the local electric field parallel to the surface will be weakly observed.36 This suggests that the nitrophenylpyran rings are initially assembled on the surface in a way that the direction of the antisymmetric vibrational dipole moment of NO2 is almost parallel to the substrate (see details in the following section for SFG simulation). The existence of a positive peak for the symmetric mode of NO2 suggests that the NO2 group is pointing away from the surface based on the relative phase of the SFG signal in the C−H stretching region.37,38 This information, together with the orientation restrictions dictated by the point of attachment of the SP-LA on the surface, suggests that the plane of the nitrophenylpyran ring is horizontal with respect to the surface with a minimum twist angle ψ (rotational angular change about the symmetry axis of the NO2 group relative to the surface parallel). Moreover, the plane of the nitrophenylpyran ring is situated at a tilt angle θ from the surface normal so that the NO2 group is pointing away from the surface (θ = the angle between the symmetric axis dissecting the NO2 group and the surface normal axis) (see Figure 2). UV Irradiation of the Closed Form of SP-LA on the Surface. In the following, the effect of UV light illumination on the structure of the SP-LA SAM was investigated. Figure 3a displays the SFG spectrum during illumination of the SAM with UV light (λ = 350 nm). UV exposure leads to a decrease in the amplitude of the symmetric vibrational mode and the appearance of the antisymmetric vibrational mode of NO2, reaching saturation after 10 min. An increase of the antisymmetric peak, with a corresponding decrease of the symmetric peak, suggests significant geometric change in the orientation of the nitrophenylpyran region of the molecule. We assign the change in the vibrational spectra to a photoinduced SP→MC isomerization of the spiropyran. In the MC-form, it is likely that the direction of the dipole moment of NO2 symmetric vibrational mode no longer has a strong perpendicular component to the surface, which implies an increase in the tilt angle θ, and therefore, the amplitude of the

■

RESULTS AND DISCUSSIONS Figure 1 shows the SFG spectrum (a) of a freshly prepared SPLA monolayer anchored on gold surface and the FTIR

Figure 1. (a) SFG spectrum in fingerprint frequency region (1300− 1650 cm−1) of originally prepared spiropyran monolayer on a gold substrate. (b) FTIR transmission of SP-LA powder in KBr disk. The assignments of the peaks are given beside the peaks.

transmission spectrum (b) of bulk SP-LA molecules dispersed in a KBr pellet, in the frequency region between 1300 and 1650 cm−1. The two spectra are different in both peak shapes and intensities. The FTIR spectrum shows only downward peaks (in transmission mode), while the SFG spectrum shows more complex features. This is because the SFG signals shown in Figure 1a arise from both the resonant contribution from SPLA and the nonresonant contribution from the gold substrate. The two signals interfere with each other and give constructive (upward peaks) or destructive (downward peaks) features.21 Such peak line shape information can further be employed to determine the orientation of the dipole moment at the interface and will be discussed later. The peak assignments of the SFG spectrum can be made on the basis of the FTIR spectrum (Figure 1b). The major peaks in Figure 1b can be assigned to the NO2 symmetric stretch (νsNO2, 1335 cm−1), antisymmetric deformation of CH3 (δasCH3, 1460 cm−1) and CH2 (δasCH2, 1483 cm−1), NO2 antisymmetric stretch (νasNO2, 1512 cm−1), and aromatic breathing modes (δAro, 1578 and 1607 cm−1).11,19,35 Appearance of the peaks in Figure 1a implies that the SP-LA formed a well-ordered monolayer at the surface. This is due to the fact that SFG peaks, for example, NO2 peaks, 13854

dx.doi.org/10.1021/la302204f | Langmuir 2012, 28, 13852−13860

Langmuir

Article

ring perpendicular or nearly perpendicular to the substrate, which implies an increase in the twist angle, which results in the increase of the antisymmetric vibrational mode of NO2 (see the following section on SFG simulation). Visible Light Irradiation of the Open Form of SP-LA on the Surface. When a UV-saturated surface was irradiated with visible light (λ = 580 nm), a gradual decrease of the NO2 antisymmetric peak was observed while the NO2 symmetric peak remained fairly constant with the exception of a slight increase of the latter mode at the beginning (Figure 3b). The weakening of the antisymmetric vibrational modes of NO2 upon visible light irradiation while having the symmetric vibrational modes almost constant suggests that this process is dominated by a decrease in the twist angle ψ while the tilt angle θ remains fairly constant. This implies that the plane of the nitrophenyl group is now placed again horizontal to the gold surface with only a few perpendicular components for both the symmetric and antisymmetric vibrational modes. This is attributed to the MC→SP reisomerization and recombination of the two parts of the molecule to form the SP-form again. The subsequent switching between the SP→MC forms using UV and visible light, alternatingly, was observed to be reversible. The increase and decrease of the NO2 peaks in Figure 3b can be reproduced using UV and visible light, consecutively, after the first UV irradiation without any noticeable degradation in the SAM, as determined by the reproducibility of the SFG signals (not shown here). However, it was evident that the orientation of the SP-form of SP-LA in the SAM, after the first open/close switching cycle, does not return to the original orientation found directly after SAM formation. This was concluded from the symmetric peak and the tilt angle, which remain almost constant after the first visible light irradiation (switching from open state back to closed state) and do not go back to their initial values (i.e., in SAM as initially prepared). The opening and closing of the pyran ring in the first cycle is believed to induce structural changes in the SAM to minimize the crowding of the molecules on the surface and to allow space for the switching, which was also confirmed by probing the rate of isomerization (see below). The SFG signals attributed to the CH2 and CH3 vibrational modes in the region of 2800−3000 cm−1, which correspond to the linker and the two methyl groups on the indole ring, remain almost the same, indicating that no damage occurs after UV and visible light irradiations and that the orientation of the linker system and the indole part of the spiropyran molecule do not significantly change upon UV irradiation (see Figure 1S in the Supporting Information). SFG Simulation of the Three States of SP-LA on the Gold Surface. Since both the twist angle (ψ) and the tilt angle (θ) have to be considered in the molecular system, the analysis for the SFG signals becomes complicated. In order to quantitatively analyze the molecular orientation of the three extreme states of SP-LA on the gold surface from the SFG observations, simulation of the SFG intensities of νsNO2 and νasNO2 has been conducted on the basis of a number of assumptions.39−41 Details about the simulation can be found in the Supporting Information. In brief, the observed ppppolarized SFG signal from the gold surface is dominated by the zzz-component,41 which is a function of the tilt (θ) and twist (ψ) angles (Figure 2). Parts a and b of Figure 4 show the dependency of the symmetric and antisymmetric intensity of the SFG signals of NO2, respectively, on the tilt and twist

Figure 2. Representative orientation of the nitophenylpyran group with respect to the surface, defining the tilt angle θ (angular change of the symmetry axis of the NO2 group relative to the surface normal) and the twist angle ψ (rotational angular change about the symmetry axis of the NO2 group relative to the surface parallel). The lab coordinates are defined as x, y, z, with the x,y-plane being parallel to the gold surface and the z-axis being perpendicular to the surface plane.

Figure 3. SFG spectra in fingerprint frequency region (1300−1550 cm−1) of SP-LA SAM (a) following UV irradiation and (b) visible light irradiation, showing the symmetric (νs) and antisymmetric (νas) peaks of NO2 group.

NO2 symmetric signal decreases. On the other hand, the ringopening of the pyran group puts the plane of the nitrophenyl 13855

dx.doi.org/10.1021/la302204f | Langmuir 2012, 28, 13852−13860

Langmuir

Article

Figure 4. Simulated intensity of (a) the symmetric NO2 peak intensity χ(2),sszzz (left panel) and (b) the antisymmetric NO2 peak intensity χ(2),aszzz (right panel) as a function of the twist and tilt angle of the NO2 group. See Supporting Information for details about the simulation.

Scheme 2. Possible Conformations of SP-LA during Photoirradiation and the Orientation of the Nitrophenyl Group and the SFG Response for Symmetric (νsym) and Antisymmetric (νasym) Modes of NO2a

a

Conformation (a) shows a strong NO2 symmetric peak in the SFG spectrum, while conformation (b) shows a strong NO2 antisymmetric peak and a weak signal for the symmetric mode. When the molecule adopts the NO2 group conformation in (c), both modes give a weak SFG signal.

suggests that simulation is crucial for a quantitative understanding of the observed SFG spectra. On the basis of the simulation results that are presented in Figure 4, we are able to semiquantitatively discuss these orientation changes from the changes of the SFG spectral intensity of the NO2 group. Therefore, with this we are able to minimize the uncertainty in the orientation of the different states of SP-LA on the gold surface by limiting the range of values for the tilt and twist angles, and this gives a clearer indication on the conformations of SP-LA molecules at the surface, as shown in Scheme 2. The simulation results confirm our assignments about the molecular conformations at the interface, where for the initially prepared SAM the NO2 group points away from the surface, which gives a strong symmetric

angles, based on simulation and considering the symmetry of NO2 group (see Supporting Information). It is clear from these figures that the NO2 symmetric peak intensity decreases monotonically with the tilt angle, but it is insensitive to the twist angle and reaches minimum at θ = 90° (Figure 4a). On the other hand, the NO2 antisymmetric peak intensity shows more complicated features, as it increases with ψ for all θ. A maximum intensity of the NO2 antisymmetric peak was observed at θ = 55° and ψ = 90°, but minimum intensity is shown at both sides of the θ range (0°−90°). The latter feature indicates that normal expectation based on the IR dipole moment does not apply in this case, in which one expects that the z-component for the dipole moment of the NO2 antisymmetric mode reaches a maximum at θ = 90°. This 13856

dx.doi.org/10.1021/la302204f | Langmuir 2012, 28, 13852−13860

Langmuir

Article

Figure 5. Reaction kinetics for the (a) UV and (b) visible light irradiation processes of the SP-LA SAM.

that TTC isomer, which is the most favorable isomer in solution, is also favored in this monolayer. Kinetics of the Switching Processes of SP-LA. To quantitatively evaluate the photoinduced isomerization processes, the peak intensity of the antisymmetric stretch mode of NO2 as a function of UV and visible irradiation time is plotted in parts a and b of Figure 5, respectively. As the intensity of the UV or visible light is constant, the reaction can be regarded as a pseudo-first-order process. The results can be fitted by the equation ISFG1/2(t) = (ISFG0)1/2ekt + b, where ISFG is the peak intensity, ISFG0 is the intensity at t = 0, k is the observed rate constant, t is time, and b is the asymptotic value.34,46 The rate constant of the initial UV irradiation process and the one induced by visible light were determined to be ki(UV) = 3.4 × 10−3 s−1 and k(vis) = 1.9 × 10−4 s−1. The reversible UV swing rate constant was determined to be kr(UV) = 4.7 × 10−3 s−1, which is a little faster than the initial UV irradiation process and agrees with the change in orientation of the molecules from its original assembly after the first switching cycle to give an easier switching ability of the molecules in the SAM. When a UVsaturated surface was left in the dark while probing the SFG signal of the antisymmetric band, switching back was also observed; however, predictably the kinetics followed a slower MC→SP thermal isomerization rate, k(relaxation) = 9.0 × 10−5 s−1 (see Figure 2S in the Supporting Information). It is interesting to note that the photoswitchability of SP-LA on a gold surface measured in this study using SFG is of a comparable time scale to that observed for a similar type of spiropyran molecule assembled on single-walled carbon nanotubes (SWCNTs). The latter was measured indirectly by following the drain current as a function of time on these semiconducting transistors [i.e., k(UV) = 1.9 × 10−3 s−1, k(vis) = 1.5 × 10−3 s−1].47 Moreover, the time scale for the photoisomerization of the SP-LA SAM on gold is one order of magnitude slower than that observed for similar spiropyrans when doped into polymer films, as determined by their UV−vis spectra [k(UV) = 2.6 × 10−2 s−1, k(vis) = 1.2 × 10−2 s−1],46 but much faster than the rates of switching spiropyrans (thermal relaxation) in the solid crystalline form (ca. 9 × 10−8 s−1).48 This implies that spiropyran assembled on surfaces or in SAMs lacks the mobility that polymer films can provide through flexibility or the existence of local solvent pockets for the

peak but a very weak antisymmetric peak. Although we are still unable to exactly determine the absolute orientation angles (θ, ψ) from the present measurement solely, one can expect that θ should be quite small (