Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://chem.msu.ru/eng/lab/chemkin/chemkin-people/articles-pdf/yankova2.pdf Äàòà èçìåíåíèÿ: Mon Mar 14 12:11:20 2011 Äàòà èíäåêñèðîâàíèÿ: Mon Apr 11 06:04:23 2016 Êîäèðîâêà: Ïîèñêîâûå ñëîâà: ori

ISSN 0036 0244, Russian Journal of Physical Chemistry A, 2011, Vol. 85, No. 4, pp. 695­701. © Pleiades Publishing, Ltd., 2011. Original Russian Text © T.S. Yankova, N.A. Chumakova, A.Kh. Vorob'ev, 2011, published in Zhurnal Fizicheskoi Khimii, 2011, Vol. 85, No. 4, pp. 777­783.

PHOTOCHEMISTRY AND MAGNETOCHEMISTRY

Photoinduced Orientational Order of Dichloride Anion Radicals
T. S. Yankova, N. A. Chumakova, and A. Kh. Vorob'ev
Department of Chemistry, Moscow State University, Moscow, 119991 Russia e mail: ya.tatiana@gmail.com
Received March 31, 2010

Abstract--The photo orientation of dichloride anion radicals (Cl 2 ) in a glassy solution of 5 M LiCl is dis cussed. The quantitative characteristics of orientation of paramagnetic molecules were determined using the anisotropy of optical absorption and the angular dependence of the EPR spectrum. The orientational distri bution function of ordered anion radicals was determined by joint computer modeling of the EPR spectrum recorded at different directions of the symmetry axis of a sample relative to the magnetic field of a spectrom eter. It was found that the value of the order parameter (­0.1 ± 0.01), calculated from the orientational dis tribution function coincides with the value obtained under the measurements of the linear dichroism in the range of the detection error (­0.12 ± 0.01). Keywords: EPR spectra, photo orientation, dichloride anion radical, paramagnetic molecules. DOI: 10.1134/S0036024411040303

­·

INTRODUCTION Photo orientation is a light induced orientational order of molecules [1­4]. This phenomenon is observed upon the fulfillment of two conditions: (1) the probability of light being absorbed by a mole cule depends on the latter's orientation relative to the vector electrical component of the light wave; (2) having absorbed a photon, the molecule changes its spatial ori entation. When these conditions are met during the irra diation of an initial isotropic medium, the light adsorb ing molecules are accumulated in an orientation in which the probability of light adsorption is minimal. At present, the phenomenon of photo orientation is under intense scrutiny using the examples of azobenzene and other dyes, and of polymers based on them [5­11], due to the prospects for developing information record ing and storage devices [12­14], and for practical use in the field of polarized holography [15]. The main characteristic of materials with a partial ordering of molecules is the orientation factor or order parameter. This characteristic is most often deter mined from the linear dichroism spectrum of an anisotropic sample [3, 16­21]. Earlier [22], we devel oped another method for determining the orienta tional order characteristics that is applicable only to paramagnetic molecules and is based on computer modeling of the angular dependence of the EPR spec trum. The orientational distribution function obtained is the most comprehensive and informative descrip tion of the particles' orientational order. From the ori entational distribution function, we can calculate in

particular the orientation factor of the studied mole cules. The most important stage in the testing and devel opment of the new method is comparing the results obtained with those of the standard optical method of molecular ordering investigation. It is to this task that the present work is devoted. Dichloride anion radicals (Cl 2 ) in a glassy matrix of 5 M LiCl solution were chosen as the object of investigation, since it is known that the photo orienta tion phenomenon is observed in this system [22­24], ­· Cl 2 anion radicals have an optical adsorption band that is easy to measure [23], and an EPR spectrum that is described in the literature [25, 26]. It is also highly ­· important that the simplicity of the Cl 2 molecule's geometry allows us to determine unambiguously the directions of a dipole moment of optical transition and those of the main axes of g and hyperfine interactions tensors relative to the molecular coordinate system. EXPERIMENTAL An aqueous solution of LiCl (5 M) containing dis solved molecular chlorine was obtained by applying the desired amount of gaseous chlorine with a gas syringe. The solution was immersed in a quartz ampoule for EPR spectroscopy. A glassy sample was ­· obtained by rapid cooling of the ampoule to 77 K. Cl 2 radicals were generated by irradiation the sample with light having a wavelength of 254 nm. A BUF 4 non
­·

695


696

YANKOVA et al.

electrode mercury vapor lamp of low pressure with high frequency excitation was used for this purpose. Upon irradiation, the ampoule was rotated with a speed of 2­3 rps around its axis. The sample was irra diated until the value of the optical absorption of Cl 2 radicals reached a value of D = 2 at the maximum of the adsorption band (345 nm). The orientational order of Cl 2 radicals was pro duced by irradiation of an isotopic sample with a par allel beam of plane polarized light having a wave length of 365 nm. To accomplish this, a high pressure mercury vapor lamp (500 W) equipped with a system of quartz lenses, a standard kit of glass light filters, and a Glan polarization prism was used. The spectra of optical absorption were registered in polarized light on a SPECORD M40 spectrophotom eter; EPR spectra were registered using a Varian E3 EPR spectrometer. The spectra were registered at 77 K in quartz Dewar vessels. The angular depen dences of the EPR spectra were registered by turning a sample in the spectrometer's resonator with an incre ment of 20° using a goniometer; the accuracy of the angle arrangement was ±2°C. Method for the Computational Modeling of EPR Spectra The EPR spectra were modeled in an approxima tion of a weak external field accurate to within the sec ond order terms with allowance for the energy of nuclear Zeeman interaction and the contributions of forbidden transitions. The program for calculating the spectra allows the use of the most general form of an individual line's shape, a convolution of the Gaussian and Lorentz functions. Upon modeling, however, it was found that the shape of the individual line of Cl 2 anion radicals is described by a Gaussian function. Modeling also showed that the width of an individual line in the studied spectra depends on the projection of the magnetic moment of the nuclei, mI, and on a rad ical's orientation relative to the magnetic field. It is known [27] that in the spectra of paramagnetic particles registered in the solid phase, the dependence of the width of a spectral line on the projection of a nucleus's magnetic moment can be expressed as:
2 k = ak + bkmI + ckmI ,
­· ­· ­·

x = y = = (a + bm1 + cm1 ) + (a + bm2 + cm2 )
2 2 1/ 2

(

22

z = = a + b m1 + c m1 )

(

)

,

22

2 + (a + b m2 + c m2 )2 , where m1 and m2 are projections of the nuclear spin of ­· two nuclei of Cl 2 anion radicals. To consider the natural isotopic composition of ­· chlorine atom nuclei, the EPR spectrum of Cl 2 anion radicals was calculated as a sum of the spectra of three particles, 35Cl35Cl, 35Cl37Cl, and 37Cl37Cl. The natural content of 35Cl and 37Cl atoms is approximately 75.78 and 24.22%, and the ratio of summation coefficients of the three spectra was therefore 9.75 : 6.32 : 1. The constants of the hyperfine interaction of an unpaired electron with nuclei of 35Cl and 37Cl are proportional to the magnetic moments of the nuclei, which are 0.8219 and 0.6841 units of nuclear magneton, respec tively [28]. In accordance with the geometry of dichlo ride anion radicals, all anisotropic characteristics such as g tensor, the tensor of hyperfine interaction, and the characteristics of the half width of a line are described using axial tensors. The computational modeling of experimental EPR spectra of an isotropic sample was performed in order to determine the magnetic resonance parameters of ­· the Cl 2 radicals. In the course of modeling, the deter mined parameters (the characteristics of the individ ual line width, g tensor and hyperfine interaction ten sor components) were varied in order to find the min imum of the sum of squares of deviations of the spectrum calculated from the experimental spectrum. Minimization was performed using the nonlinear least squares method [29, 30]. The experimental and calcu ­· lated spectra of Cl 2 anion radicals in an isotropic sample are shown in Fig. 1. The magnetic resonance parameters obtained by modeling are shown in the table. The main values determined by the authors of [25] for the g and hyperfine interaction tensors are also shown in the table. We can see that the values obtained as a result of spectrum modeling are in rea sonable agreement with the literature data.

)

1/ 2

where k = x, y, z. The width of a Gaussian line of Cl 2 binucleate particles is therefore described in the present work with a second order tensor whose main values were calculated as follows:

­·

Modeling of the Angular Dependence of the EPR Spectrum The method for modeling the angular dependence of EPR spectra is described in [22]. To determine the orientational distribution function of paramagnetic particles, six spectra of an anisotropic sample, regis tered at different angles between the axis of anisotropy of the sample and the direction of the spectrometer's magnetic field, were modeled jointly at fixed magnetic
Vol. 85 No. 4 2011

RUSSIAN JOURNAL OF PHYSICAL CHEMISTRY A


PHOTOINDUCED ORIENTATIONAL ORDER

697

10 mT

Fig. 1. EPR spectrum of statistically oriented Cl tional modeling (circles).

­· 2

anion radicals in 5 M LiCl at 77 K (line) and the results from their computa

resonance parameters upon variation of the Ak coeffi cients of the series expansion of the orientational dis tribution function by a Legendre polynomial:

RESULTS AND DISCUSSION Photo Orientation: The Value of a Sample's Optical Dichroism The photo orientation of Cl 2 radicals in a glassy media has been studied quite extensively [22, 24, 31]. Photo orientation is specified by the probability of the light absorbance by a molecule being proportional to the square of the cosine of an angle between the vector of a molecule's dipole transition moment and the elec ­· trical component of the light. A Cl 2 anion radical ori ented properly for light absorbance reacts with a ­· neighboring chlorine anion of a matrix (Cl 2 + Cl­ Cl­ + Cl 2 ), and a similar anion radical with other spatial orientation is formed as a result. Through the
­· ­·

( ) =

k =0





AkPk(cos ),

(1)

where is the angle between the sample's axis of sy m metry and the direction of the Cl­Cl bond of a paramagnetic molecule (the direction of the z axis of a g tensor). In the course of modeling, it was found that using coefficients of expansion of the fourth order and higher does not lead to an improvement in the quality of the EPR spectra's description. Modeling was there fore performed using coefficients of the second order.
-i

Magnetic resonance parameters and half width of the Cl 2 anion radicals spectra in 5 M LiCl Direction relative to the Cl­Cl bond Perpendicular a â 104 13.30 ± 0.26 ­b â 10
4

c â 10

4

2.03865 ± 0.00006 2.050 [25]

A

35 SFI( Cl)

T

â 104, A

37 SFI( Cl)

T

â 104,

2.83 ± 0.56

3.64 ± 0.33

14.00 ± 0.29

11.77 ± 0.24

(8.5 ± 1.5) [25] 100.9 ± 0.24 84.06 ± 0.20

Parallel

3.98 ± 0.36

10.03 ± 0.79

14.00 ± 0.30

2.00791 ± 0.00015 2.004 [25]

(103 ± 1) [25]

Note: ASFI is a constant of hyperfine interaction; a, b, and c are components of the Gaussian width of an individual line, T. RUSSIAN JOURNAL OF PHYSICAL CHEMISTRY A Vol. 85 No. 4 2011


698 D

YANKOVA et al.

1.6

0.8

0

300

400

500 , nm
­· 2

Fig. 2. Optical absorbance spectrum of Cl in 5 M LiCl at 77 K.

anion radicals

multiple repetition of this process, the sample is depleted of radicals capable of absorbing light. Anisot ropy of the sample results and appears in both the opti cal spectra and the EPR spectra. The spectrum of the optical absorbance of Cl 2 anion radicals in 5 M LiCl at 77 K is shown in Fig. 2. The maximum of the absorption band corresponds to a wavelength of 345 nm. In order to minimize the measurement errors in calculating the value of optical
­·

dichroism, a region of the spectrum was chosen in which the optical density did not exceed D = 2. The regions of the sample's absorption spectra used for the calculations after photo orientation are shown in Fig. 3. We can see that the intensity of the recorded spectrum upon the perpendicular polarization of the orienting and probing beams is higher than that upon ­· parallel polarization. This means that Cl 2 molecules are oriented so that their vectors of the dipole moment of transition are mostly directed perpendicularly to the direction of light polarization. The sample obtains an axis of anisotropy that is codirectional with the electri cal component of the light. It is evident that the vector of the dipole moment of the above optical * transition is directed along the Cl­Cl bond. During ­· the photo orientation process, Cl 2 radicals are thus oriented mainly by their Cl­Cl bond, perpendicular to the electric component of the illuminating light. The orientational order of the vectors of dipole transition moment of molecules is characterized by the value of the sample's optical dichroism:

d=

D - D , D + 2D

(2)

where D|| and D are values of the optical absorbance in spectra with mutually parallel and perpendicular polarizations of the oriented and probing beams. The dichroism of the sample containing ordered ­· Cl 2 radicals was averaged over the range of wave

D

1.6

0.8

||

0

350

400

450

500 , nm

Fig. 3. Optical absorbance spectrum of Cl 2 anion radicals in 5 M LiCl at 77 K after irradiation with linear polarized light with a wavelength of 365 nm upon parallel (line) and perpendicular (circles) mutual polarization of the probing and illuminating beams. RUSSIAN JOURNAL OF PHYSICAL CHEMISTRY A Vol. 85 No. 4 2011

­·


PHOTOINDUCED ORIENTATIONAL ORDER

699

0°

(a)

90° 10 mT

(b)

Fig. 4. Angular dependence of the EPR spectrum of Cl

­· 2

anion radicals (a) and the results from their computational modeling (b).

lengths from 349 to 390 nm. The value obtained is d = ­0.12 ± 0.01. Orientational Distributional Function of Dichloride Anion Radicals Figure 4 shows the experimental angular depen dence of the EPR spectrum of a sample containing oriented Cl 2 radicals and the results from their com puter modeling. We can see that the values of the angu lar dependence (the maximum difference between the spectra registered at different angles of sample rota tion) in the calculations have good reproducibility, despite the ratio of different spectrum components being reproduced inaccurately.
RUSSIAN JOURNAL OF PHYSICAL CHEMISTRY A
­·

As was stated above, it is sufficient to use the second order of the function expansion by a Legendre polyno mial (i.e., only two coefficients of expansion (1)) to describe the orientational distribution function of dichloride anion radicals. The A0 coefficient charac terizes the number of particles in a sample, while the A2 value describes their orientational order. It was found that A2 = ­0.25 ± 0.02 (at A0 = 1). The orientational distribution function obtained upon modeling is shown in Fig. 5. The function is pre sented in the sample's coordinate system. The z axis corresponds the direction of the electrical component of the orienting light (the sample's axis of anisotropy). The three dimensional figure shown in Fig. 5 is the envelope of vectors coming from the coordinate origin in different directions; the length of each vector corre sponds to the number of particles oriented in the
Vol. 85 No. 4 2011


700

YANKOVA et al.


D = N l ( ) ( ) sin d ,
0



(3)

z

where N is the concentration of the light absorbing substance, l is the length of the optical path, is the angle between the direction of the light wave's electric vector and the direction of the dipole moment of tran sition in a molecule, and () is the coefficient of a molecule's optical absorbance. We can then obtain the expression for the main values of the optical absorp tion of a dichroic sample for the axial tensor of absorp tion coefficient (x = y z):
- x si n D = N l + 2P2( c o s ) z 3 2



0

x y

â


k=0





A2k P2k (cos )d,

(4)

Fig. 5. Orientational distribution function of Cl radicals.

­· 2

anion

- x sin D = N l - P2(cos ) z 3 2


0

direction of this vector. The flattened form of the dis tribution function confirms that the dichloride anion radicals are mainly oriented perpendicular to the direction of light polarization as a result of irradiation by the polarized light. The orientational order of the particles is generally characterized by the averaged value referred to as the ori entation factor (the order parameter) f = (3cos2 ­ 1)/2, in which the angle brackets denote averaging over all particles of the sample. It is easy to demonstrate that when the orientational distribution function of the molecules is expressed by series (1), the orientation factor is determined as f = (2/5)A2. The orientation factor of Cl 2 radicals obtained using EPR is thus fEPR = ­0.10 ± 0.01. The negative value of the orienta tional factor means that particles are mainly oriented perpendicular to the sample's axis of symmetry. Comparison of the Value of Optical Dichroism ­· and the Orientation Factor of Cl 2 Anion Radicals Obtained from the Orientational Distribution Function The coincidence of the direction of vector of dipole moment of optical transition with the direction of the ­· Cl­Cl bond of the Cl 2 radical allows us to compare with no additional complications the value of the opti cal dichroism and the radical orientation factor obtained from the orientational distribution function. The value of optical adsorption can be calculated from the orientational distribution function as follows:
­·

â

where is the average coefficient of a molecule's opti cal absorbance, and P2k(cos ) are Legendre polyno mials. It follows from (4) that the value of optical dichro ism is associated with the second expansion coefficient of the orientational distribution function's expansion as follows: D - D ( z - x) = d= A2 . D + 2D 15 For an approximation of ideal single axis absorp tion (x = y = 0, z 0), we obtain d = (2/5)A2, which is similar to the expression obtained for fEPR. The optical dichroism obtained according to for mula (2) is the orientational factor of the dipole moment vectors of molecule transition and was found to have a value of fopt = ­0.12 ± 0.01. The orientational factor determined by EPR (fEPR = ­0.10 ± 0.01) thus coincides in the range of experimental error with the orientational factor deter mined by optical spectroscopy (fopt = ­0.12 ± 0.01). CONCLUSIONS The photoinduced orientational order of dichlo ride anion radicals specified by the photo orientation phenomenon was investigated using optical and EPR spectroscopy. The value of linear dichroism for the optical absorption of an anisotropic sample was calcu ­· lated. The orientational distribution function of Cl 2 radicals was determined from an analysis of the angu
Vol. 85 No. 4 2011

k=0





A2k P2k (cos )d,

RUSSIAN JOURNAL OF PHYSICAL CHEMISTRY A


PHOTOINDUCED ORIENTATIONAL ORDER

701

lar dependence of the same sample's EPR spectrum. The value of the orientatonal factor (order parameter) of the z axis of the g tensor of a radical that lies along the Cl­Cl bond and coincides with the direction of the dipole moment vector of the observed optical tran sition was obtained from the orientational distribution function. It was found that the values of the order parameter obtained from the EPR spectra coincide quantitatively. This confirms the reliability of the method developed earlier for investigating orienta tional order by means of EPR spectroscopy. ACKNOWLEDGMENTS This work was financially supported by the Russian Foundation for Basic Research, project no. 09 03 00900. REFERENCES
1. A. C. Albrecht, J. Chem. Phys. 27, 1413 (1957). 2. A. C. Albrecht, Prog. React. Kinetic 5, 301 (1970). 3. J. Michl and E. W. Thulstrup, Spectroscopy with Polar ized Light: Solute Alignment by Photoselection, in Liquid Crystals, Polymers, and Membranes (VCH Publ., Wein heim, 1986). 4. Experimental Methods of High Energy Chemistry: The Guide, Ed. by M. Ya. Mel'nikov (Mosk. Gos. Univ., Moscow, 2009) [in Russian]. 5. I. Zebger, M. Rutloh, U. Hoffmann, et al., J. Phys. Chem. A 106, 3454 (2002). 6. A. Bobrovsky, V. Shibaev, and N. Boiko, Prog. Polym. Sci. 23, 729 (2003). 7. C. Cojocariu and P. Rochon, Pure Appl. Chem. 76, 1479 (2004). 8. D. Rais, S. Nespurek, Y. Zakrevskyy, et al., J. Optoelec tron. Adv. Mater. 7, 1371 (2005). 9. C. C. Jung, M. Rutloh, and J. Stumpe, J. Phys. Chem. B 109, 7865 (2005). 10. A. Bobrovsky and V. Shibaev, Polymer 47, 4310 (2006). 11. A. Bobrovsky, A. Ryabchun, A. Medvedev, et al., J. Photochem. Photobiol., A 206, 46 (2009).

12. S. Kawata and Y. Kawata, Chem. Rev. 100, 1777 (2000). 13. C. C. Corredor, Z. L. Huang, and K. D. Belfield, Chem. Mater. 19, 5165 (2007). 14. A. Bobrovsky, V. Shibaev, G. Elyashevich, et al., Polym. Adv. Technol. 21, 100 (2010). 15. L. Nikolova and P. S. Ramanujam, Polarization Holog raphy (Cambridge Univ., Cambridge, 2009), p. 88. 16. E. W. Thulstrup and J. Michl, J. Am. Chem. Soc. 104, 5594 (1982). 17. J. Schellman and H. P. Jensen, Chem. Rev. 87, 1359 (1987). 18. P. M. Dolan, D. Miller, R. J. Cogdell, et al., J. Phys. Chem. B 105, 12 134 (2001). 19. A. Rodger, J. Rajendra, R. Marrington, et al., Phys. Chem. Chem. Phys. 4, 4051 (2002). 20. M. A. R. B. Castanho, S. Lopes, and M. Fernandes, Spectroscopy 17, 377 (2003). 21. A. Bobrovsky, V. Shibaev, V. Hamplova, et al., Monatsh. Chem. 140, 789 (2009). 22. A. Kh. Vorobiev and N. A. Chumakova, J. Magn. Res. 175, 146 (2005). 23. A. Kh. Vorobiev and V. S. Gurman, Khim. Fiz 6, 1220 (1987). 24. A. Kh. Vorobiev and V. S. Gurman, Chem. Phys. 167, 341 (1992). 25. E. B. Zvi, R. A. Beaudet, and W. K. Wilmarth, J. Chem. Phys. 51, 4166 (1969). 26. J. E. Bennet, B. Mile, and B. Ward, J. Chem. Phys. 49, 5556 (1968). 27. Yu. V. Rakitin, G. M. Larin, and V. V. Minin, Interpre tation of EPR Spectra of Coordination Compounds (Nauka, Moscow, 1993) [in Russian]. 28. P. Paghavan, At. Data Nucl. Data Tables 42, 189 (1989). 29. J. E. Dennis, D. M. Gay, and R. E. Welsch, Trans. Math. Software 7, 348 (1981). 30. J. E. Dennis, D. M. Gay, and R. E. Welsch, Trans. Math. Software 7, 369 (1981). 31. A. Kh. Vorobiev, A. V. Polenok, and V. S. Gurman, Khim. Fiz. 9, 840 (1990).

RUSSIAN JOURNAL OF PHYSICAL CHEMISTRY A

Vol. 85

No. 4

2011