ENDOR Study on the Dynamic Properties of the First Stable

Jun 28, 2010 - SAR1 center is presented on the basis of thermodynamic properties of the R-hyperfine ... Faculty of Veterinary Medicine, University of ...
0 downloads 0 Views 745KB Size
Download PDF
7500

J. Phys. Chem. A 2010, 114, 7500–7505

ENDOR Study on the Dynamic Properties of the First Stable Paramagnetic Center in γ-Irradiated L-Alanine Crystals Boris Rakvin,*,† Nadica Maltar-Strmecˇki,‡ Daniel Kattnig,§ and Gu¨nter Grampp§ Ru{er BosˇkoVic´ Institute, P.O. Box 180, 10002, Zagreb, Croatia, Faculty of Veterinary Medicine, UniVersity of Zagreb, P.O. Box 466, 10002, Zagreb, Croatia, and Institute of Physical and Theoretical Chemistry, Graz UniVersity of Technology, Technikerstrasse 4/I, A-8010 Graz, Austria ReceiVed: April 29, 2010; ReVised Manuscript ReceiVed: June 10, 2010

Dynamic properties of the first stable L-alanine radical, SAR1, induced by γ-irradiation of L-alanine crystals, have been investigated by the electron nuclear double resonance technique (ENDOR). The study focuses on the dynamic properties of the R-proton hyperfine splitting in the temperature range from 180 to 320 K. In this region the motion of the NH3+ and CH3 groups exhibits slow and fast motional dynamics in comparison to the nuclear and electron Larmor frequencies, respectively. Evidence for different conformations of the SAR1 center is presented on the basis of thermodynamic properties of the R-hyperfine splitting. The activation processes causing the broadening of the ENDOR lines are studied. At room temperature the motional dynamics of the SAR1 center are modulated by the dynamics of the charged, neighboring NH3+ group. 1. Introduction The properties of crystalline L-alanine, one of the simplest amino acids, have been the subject of extensive spectroscopic investigations. In the crystal alanine exists as zwitterion, NH3+CH(CH3)COO-, taking part in three networks of intermolecular hydrogen bonds of unequal strength.1-3 Investigations of the temperature-induced changes in crystalline L-alanine provide a better understanding of the dynamic properties of more complex biological structures such as proteins. These studies are also relevant for their technological application as nonlinear optical devices.4 In addition, L-alanine shows specific radiation dosimetric properties in the solid state and, in combination with electron paramagnetic resonance (EPR) spectroscopy, is widely use as a dosimeter material.5 Temperature-induced changes in crystalline L-alanine6 also influence the stable paramagnetic centers generated by X-ray or γ-irradiation. These changes can be utilized for characterizing the dynamics of the centers and its neighboring lattice fragments. The first stable L-alanine radical, •CH(CH3)COOH, henceforth denoted SAR1, has been studied for many years and remains the most frequently studied paramagnetic amino acid center in the solid state.7-29 It has been examined by various EPR techniques (continuous wave and pulsed EPR, electron-nuclear double resonance, ENDOR, ENDOR induced EPR, and ELDOR). Despite the huge number of investigations, the molecular environment of the radical, in particular the unusually large displacement of the central carbon atom, CR, which contains 75% of the spin density, as well as the dynamic processes associated with the radical motion, are still not fully understood. One possible conformer of the radical within the L-alanine lattice, which has been adopted from an earlier ENDOR study13 is shown in Figure 1. Theoretical modeling attempts of the SAR1 center30-32 have mostly been limited to * To whom correspondence should be addressed: fax, +385-1-4680245; e-mail, [email protected]. † Ru{er Bosˇkovic´ Institute. ‡ Faculty of Veterinary Medicine, University of Zagreb. § Institute of Physical and Theoretical Chemistry, Graz University of Technology.

Figure 1. Structure of the first stable alanine radical, SAR1 (large spheres), displaced from the original position (denoted with small spheres) in the L-alanine crystal lattice as obtained from an earlier ENDOR study.13 The positions of the two nearest hydrogen atoms (H3 and H6) to O1 and O2 of SAR1 center are shown with corresponding two neighbor molecules. The presented plane was constructed containing C2, O1 and O2 atoms and it also contains C1 and C2 atoms due to planar radical model.

the center itself without taking interactions with the surrounding atoms in the crystal matrix into account. Moreover, the internal rotations of the CH3 and NH3+ groups in combination with the molecular zwitterionic form in the crystalline state produce even more complex local environments of the SAR1 center at room temperature. In recent calculations30-32 of the hyperfine coupling constants, HFCC, of SAR1 density-functional theory (DFT) has been combined with different approximations of the environment surrounding the center. These calculations result in several possible model structures showing an agreement in the range of 5-10% with experimental data. Within this accuracy, it was not possible to distinguish subtle contributions, which have their

10.1021/jp103883x  2010 American Chemical Society Published on Web 06/28/2010

First Stable L-Alanine Radical source in thermal averaging or the environmental surrounding. Thus, for an improved description of the SAR1 center a more precise experimental and theoretical evaluation of dynamical properties is essential. The EPR spectrum of irradiated L-alanine consists of three stable paramagnetic centers.22 Besides SAR1, which contributes approximately 60% to the total room temperature spectrum, there are two additional paramagnetic species: the second stable alanine center, SAR2, which contributes 40%, and a third alanine radical, SAR3, which accounts for a minor contribution to the cumulative room temperature spectrum. It is important to note that SAR2 (NH3+C•(CH3)COO-) does not exhibit a displacement of the CR atom and its theoretical modeling within the crystal lattice was, thus, more successful than the modeling of the SAR1 center.30-33 In previous studies an additional anisotropy of the motional dynamics of the CH3 group has been suggested.23,25,29 This raises the question of whether there is any evidence for motional dynamics of the entire SAR1 center. In order to probe for this possibility, a study of the temperature dependence of the R-proton HFCC has been undertaken. In a previous CW-EPR study of the R-proton hyperfine coupling in the high temperature region34 (300-450 K) the supposition was raised that the observable coupling constant is the result of the fast averaging of two different conformations of SAR1. Due to the relative small changes of the R-proton HFCC in comparison to the EPR line width, the thermal averaging could not be detected in the vicinity of room temperature by employing conventional CW-EPR. However, by employing ENDOR spectroscopy, the detected line width is reduced by nearly 2 orders of magnitude and, hence, the R-splitting is obtained with enhanced accuracy. In the present ENDOR study the internal dynamics of the R-proton have been studied. The ENDOR line widths and line shifts obtained for different crystal orientations as a function of temperature have been used to extract different mechanisms of motion contributing to the thermal averaging of the HFCC of the R-proton. Evidence for local dynamics, which are correlated with lattice internal rocking vibrations of the COO- group and with internal torsions of the NH3+ group, has been obtained in the monitored temperature interval. Model calculations of the SAR1 center suggest that four conformers could exist simultaneously.30 The present experimental study provides an informative basis for the two conformers that contain the transferred proton30 regiospecifically attached to O1 from a direction perpendicular to the CO1O2- plane. 2. Experimental Methods Single crystals of L-alanine were prepared by slow evaporation of saturated aqueous solutions at room temperature. The crystals were irradiated by γ-rays from a 60Co source with a dose of 10 kGy yielding alanine-based organic radicals stable at room temperature. ENDOR spectra were obtained by using a Bruker Elexsys 500 spectrometer equipped with a ER4131VT temperature controller. The temperature was kept constant to within (0.1 K. All ENDOR spectra were collected by applying a 100 kHz frequency modulation of the radiofrequency field.

J. Phys. Chem. A, Vol. 114, No. 28, 2010 7501

Figure 2. ENDOR spectrum of a single crystal of L-alanine measured at 295 K. The external magnetic field is oriented parallel to the b axis. The insert shows the corresponding EPR spectrum with the position of the locked magnetic field indicated by an arrow. The field position corresponds to the low field singlet line of the SAR1 center. The arrow indicates the free proton frequency. The lines corresponding to the β-protons and the R-proton are labeled.

of the region where they exhibit significant changes. This study is undertaken in order to elucidate the leading relaxation mechanism of the proton or the proton groups of the corresponding paramagnetic center. The same approach was applied and discussed in detail for the investigation of the internal motional dynamics of CH3 of SAR115,29 and NH3+ and CH3 of SAR2.29 It should be noted that in the monitored temperature interval, 200-320 K, the ENDOR lines of the NH3+ protons22,29 appear in the slow motion regime as three separate lines. By approaching the fast motional dynamic of the NH3+ group (1/τ > ∆Aanis, where ∆Aanis denotes the anisotropic component of the hyperfine proton splitting for the monitored crystal orientation and τ represents an average proton correlation time) these lines exhibit broadening before they eventually collapse into a single line at an average frequency position. On the other hand, the ENDOR lines that were assigned to the CH3 groups in the fast motional regime of SAR1 and SAR2 appear as singlet lines in the whole temperature interval monitored. A typical ENDOR spectrum of the SAR1 radical is shown in Figure 2. The intense singlet of the R-proton (labeled in the Figure 2) is convenient for further temperature-dependent studies. A series of ENDOR spectra of the R-proton line are presented in Figure 3 as a function of temperature. The spectra clearly show complex, nonlinear changes of the line position (Figure 3) and the line width (Figure 4) for temperatures decreasing from 320 to 200 K. As was shown earlier,15,22 the ENDOR spectrum recorded at the low-field line of EPR spectrum leads to enhanced R-proton signals due to significant contribution of nonsecular (electron spin nuclear spin cross-relaxation relaxation rate, W1x) and pseudosecular terms (nuclear relaxation rate, W1n) of the spin Hamiltonian15

H ) νeSz + νn

∑ Ikz + ∑ STkIk k

(1)

k

3. Results and Discussion Temperature-Dependent ENDOR Spectra of SAR1 Center. In general, ENDOR line shift and ENDOR line width are complex functions of various relaxation mechanisms existing in the matrix.35 The present work is focused on the detection of ENDOR linewidths as a function of temperature in the vicinity

Here, Tk is the electron-nuclear hyperfine interaction of the kth proton, and the other symbols have their usual meanings. The assignment of the R-proton ENDOR line can be checked by employing the known R-proton hyperfine tensor22 to calculate the expected X-band ENDOR frequency.36 For the magnetic

7502

J. Phys. Chem. A, Vol. 114, No. 28, 2010

Figure 3. Temperature dependence of the HR ENDOR line position of the SAR1 center obtained for the magnetic field oriented along the crystal b axis.

Figure 4. HR ENDOR line width, Γ, as a function of temperature. Γ is the half-width at the half-height.

field oriented along the b axis, the calculated frequency of the line position (ν ) 41.55 MHz) is in good agreement with that observed in experiment (ν ) 41.58 MHz). This supports the assignment of the R-proton line (as well as crystal axis). It should be noted that the changes in line width are expected due to the dominant contributions of the W1x and the W1n relaxation rates over standard spin-lattice W1e relaxation (which is characteristic for the β-proton couplings of the CH3 and the NH3+ groups29). For the SAR1 center the R-proton line exhibits line shifting, while the corresponding β-proton line (Figure 3) does not show detectable line shifts in the monitored temperature interval. This finding indicates the presence of an additional, thermally activated motional mechanism beside the expected and previously analyzed rotational mechanisms related to the CH3 and the NH3+ groups.29 The temperature shift of the R-proton line was also examined for other characteristic orientations of the magnetic field, i.e., along the a and c axis of the crystal (Figure 5). In the monitored temperature interval, the R-proton hyperfine components along the a and b axes shifted in opposite directions while the component along the c axis undergoes only a marginal change in comparison to the other two components. Temperature Dependence of HFCC of r-Proton of SAR1. In order to describe the R-proton coupling, it is convenient to determine the HFCC from temperature-dependent measurements along the three axes (Figure 5) and to calculate the temperaturedependent isotropic component. The isotropic component

Rakvin et al.

Figure 5. Temperature dependence of the HR ENDOR line transition frequency of the SAR1 center obtained for the magnetic field oriented along the a axis (b), b axis (9), and c axis ([).

Figure 6. Temperature dependence of the isotropic components of the HR line position. The solid line represents the best fit to eq 2. The fitting parameters are collected in Table 1.

exhibits an increase with increasing temperature as shown in Figure 6. The temperature dependence of Aiso can be described in terms of a simple two-level model, which has also been utilized for the R-tensor components at higher temperatures34

Aiso(T) )

1 exp(-∆E/kT) A + A 1 + exp(-∆E/kT) L 1 + exp(-∆E/kT) H Aiso(T) ) A0 + A1 tanh(∆E/2kT) A0 )

AL + AH 2

A1 )

AL - AH 2

(2)

where AL and AH are the R-proton coupling for the “low” and “high” temperature stable structures and the exponential function accounts for their population ratio. The best fit of the temperature dependence of Aiso(T) to relation 2 is shown as a continuous line in Figure 6. The obtained parameters are presented in Table 1. The above model reveals an activation process relating the

First Stable L-Alanine Radical

J. Phys. Chem. A, Vol. 114, No. 28, 2010 7503

TABLE 1: Parameters Related to the Temperature Dependence of the ENDOR Line Position of Hr Obtained by Fitting Equation 2 to the Experimental Data Presented in Figures 6 and 7 HR tensor component

A0 (MHz)

A1 (MHz)

∆E (K)

Aiso

-54.43 ( 0.20

-2.16 ( 0.19

751 ( 41

AL and AH structures (∆E ) 751 K ≈ 6.2 kJ/mol). For a generic R-proton radical fragment, •CRHRR1R2, one expects that the modulation of the umbrella angle,37 φ (where three bonds to CR bent equally from the horizontal plane describes the planar and tetrahedral case when φ ) 0° and φ ) 19.5°, respectively), induces rehybridization of the CR-centered orbital system.38 As a consequence the two structures are assumed to differ with respect to their conformation at the CR plane. The difference of the isotropic HFCC splitting is proportional to the difference in the spin density at •CRHR(CH3)(COOH) of the planar and the bent conformation. Model calculation37 indicates that symmetrical bending of all bonds connecting CR from the planar to near tetrahedral geometry (φ ) 20°) leads to a change in R-proton isotropic HFCC of ∼67% for the considered •CRHR(CH3)(COOH) structure. In the present case the change in Aiso of ∼8% (Table 1) corresponds to a variation of the SAR1 bending angle37 of ∼5°.

Figure 7. Natural logarithm of the temperature-dependent part of the line width, (Γ-Γ0), as a function of the inverse temperature for various protons of the SAR1 and SAR2 centers: filled circles denote HR and empty circles denote CH3, both corresponding to the SAR1 center; empty diamonds denote H1 of NH3+, and filled diamonds denote the center, respectively. Dashed lines indicate possible activated processes at the low and high temperature region. The corresponding activation energies are collected in Table 2. The experimental data for HR proton of SAR1 center are taken from present study and the other data presented in the figure have been taken from ref 29.

TABLE 2: Activation Energies Obtained from the Temperature Dependence of the ENDOR Line Widths for Various Protons of the SAR1 and SAR2 Centers As Shown in Figure 7a activation process ENDOR, line width SAR1; SAR1; SAR2; SAR2; a

HR CH3β CH3β H1β of NH3+

∆E1 (K) 656 (narrowing) 2178 (narrowing) 2088 (narrowing) 2486 (broadening)

∆E2 (K) 5386 5386 5878 5668

(broadening) (broadening) (narrowing) (broadening)

The activation processes have been extrapolated for the low, ∆E1, and high temperature region, ∆E2, respectively.

In accord with this result one would expect that three bonds to C1 bent equally from the horizontal plane (Figure 1) of around 5° for the model shown in Figure 1. The possibility of a nonsymmetrical configuration of methyl groups with respect to the plane of the radical center has been discussed earlier.31 This possibility was based on the HFCCs of the CH3 group, detected in the absence of motional dynamics, at low temperature13 (77 K). However, according to the above model31 the bending will produce a change of the spin density on the methyl carbon position that is smaller by ∼50% than that on the CR atom. The three symmetry equivalent, rotationally averaged β-methyl protons should experience proportionally smaller shifts in the ENDOR line position in the monitored temperature interval. Indeed, as mentioned above, the experimental ENDOR results show insignificant shifts of the β-methyl protons line in comparison to the R-proton line. The theoretical model given above describes the paramagnetic center outside of the crystal lattice. Since the SAR1 center is located in the lattice containing alanine in zwitterionic form, it is important to consider lattice effects on SAR1 if a more realistic description of the center is to be obtained. In one study,30 density-function theory has been used to investigate the SAR1 center in the solid state. Four conformers of the SAR1 center (henceforce I, II, III, IV) containing intermolecular hydrogen bonds have been found as candidates for the molecular structure of the radical. Two of the conformers (III, IV) are characterized by the fact that the transferred proton binds to the oxygen in the plane of the CO1O2- group. For the other two conformers (I, II) it binds perpendicular to the CO1O2plane. Since none of the transferred structures could be excluded, an equilibrium distribution between the four conformers at room temperature was assumed. Because the calculated HFCC of all conformers are close to each other, it was suggested that the assignment of conformers could not be solved by a proton EPR study.30 However, by carefully analyzing the proton ENDOR splitting, accompanied with thermodynamic properties, new evidence supporting the presence of conformers (I, II) emerges. The calculated differences between the isotropic HFCCs of I and II and of III and IV, respectively, is equal to 2 MHz and is comparable with the experimentally detected temperature dependence of the isotropic HFCC (Table 1). The difference among the isotropic HFCC of the CH3β protons, on the other hand, is significantly smaller between I and II (0.4 MHz) than between III and IV (2.3 MHz). This finding implicates similar shifts of the R-proton line when thermally averaging within the two groups of conformers (I, II) and (III, IV), while for the CH3β proton lines different shifts are expected. Since the line shift of the CH3β protons originating from thermal averaging of conformers I and II is expected to be significantly smaller than that resulting from averaging III and IV a better correlation of the experimental results with (I, II) emerges. Moreover, the thermal averaging mechanism is related to an activation process with an activation energy of ∆E ) 751 K ≈ 522 cm-1. One notes that this process coincides with the rocking vibration39,40 of the CO1O2- group in the L-alanine lattice (520 cm-1). The fact that the proposed mechanism involves the CO1O2- rocking vibration additionally supports the hypothesis of thermal averaging between conformers I and II, i.e., species that are related by transferring the proton perpendicularly to the CO1O2- plane. The presented experimental data hence suggest correlated dynamics of the out-of-plane motion of the radical center and the motion of the CO1O2- group. Indeed, a similarly prominent vibration mode has recently been calculated for the glycine radical (+NH3-•CH-CO2-) by employing DFT calculations and

7504

J. Phys. Chem. A, Vol. 114, No. 28, 2010

Rakvin et al.

molecular dynamics simulations.41 The obtained mode (702 cm-1) corresponds to an out-of-plane motion of the paramagnetic center inversely coupled with a similar motion of the carboxyl group. In particular this simulation suggests explanation of the temperature dependence of the isotropic HFCC of 13C and HR. Moreover, as in the present case, ∆Aiso of HR was predicted to change by a similar amount (∼2 MHz) when increasing the temperature from 100 to 300 K. Temperature Dependence of ENDOR Line Width for Hr Proton of SAR1. The changes of the line width (Figure 4) of the HR ENDOR line have been employed to deduce additional information on the motional dynamics of the SAR1 center. The ENDOR transitions of the SAR centers have been assigned earlier.13,14,22 ENDOR spectra are particularly suitable for temperature-dependent studies since each ENDOR line corresponds to a known proton, located within the radical or on the neighboring molecule. The Hamiltonian (1) is usually employed to account for the CH3β proton relaxation in the fast motional regime brought about by classical proton jumps between three potential wells. It is assumed that the correlation time, τ, accounting for this motion follows a simple activation law (τ ) τo exp(∆E/kT)). As was discussed earlier15 the ENDOR intensity of SAR1 strongly depends on the electron proton crossrelaxation, W1x, as well as on the nuclear relaxation rate W1n. It is also convenient to describe contributions to the ENDOR line width in terms of transition probabilities

ΓENDOR ) Γ0 +

1 (W + W1x + W1n) 2π 1e

(3)

In order to get more information on possible relaxation mechanisms, we study the ENDOR line width, Γ, of the HR proton line. In Figure 7 the natural logarithm of Γ-Γ0, where Γ0 represents the narrowest Γ detected in the monitored temperature interval, is presented as function of 1/T. In this plot a net thermodynamic contribution illustrates possible activation processes. The slopes of the dashed lines added to Figure 7 give the activation energies of the pertinent activation processes (Table 2) in the limit of high and low temperatures. This approach only indicates the presence of several activation processes contributing to Γ (as broadening or as narrowing contribution); individual relaxation mechanisms could not be identified undoubtedly. Generally, one expects42 that on increasing temperature Γ(HR) decreases due to motional averaging of anisotropic HFCCs with other matrix protons and increases due to motions of the entire molecule. The activation process with ∆E ∼ 656 K ≈ 456 cm-1 indicates the possible presence of motional narrowing due to the torsional motion of the NH3+ matrix protons (475 cm-1) for the low temperature region.39 At higher temperature the narrowing increases further suggesting that an additional activation process related to other matrix protons (for example the CH3 groups) contribute to this process. However, in the higher temperature region a significant broadening contribution to Γ(HR) with ∆E ∼ 5386 K strongly indicates a new motional mechanism, which is probably related to entire radical motion. In order to check this supposition, it is instructive to compare the thermal behavior of Γ(CH3β) of SAR1, which was studied earlier.29 The temperature-dependent line widths are shown in the Figure 7 for the same temperature interval. Again there are two distinct temperature regions, one region of Γ(CH3β) narrowing at low temperatures and another region of broadening at high temperatures. These two regions approximately coincide with the corresponding regions of HR. At low temperatures, however, different activation processes

seem operative in both cases, i.e., the activation energy of Γ(CH3β) amounts to ∆E ∼ 2178 K as can be seen in Figure 7 and Table 2. This activation energy could be related to the activation energy of proton dynamics of the neighboring CH3 group, which, for the undamaged (nonirradiated) crystals, is accessible by NMR.42,43 Anyway, it is most important to note that Γ(HR) and Γ(CH3β) show the same broadening contributions, as well as the same activation process, in the high temperature region (Figure 7). This result supports the above assumption that the high-temperature broadening mechanism is related to motion of the entire radical, affecting both groups of protons in the same way. With the aim of identifying possible differences in the dynamic behavior between SAR1 and SAR2, we apply the same procedure to the proton ENDOR lines of the CH3β group and H1 of the NH3+ group of SAR2.29 The line width behavior has been obtained previously and is also shown on the graph in Figure 7. Γ(CH3β) of SAR2 exhibits only narrowing behavior in the examined temperature interval. In the low temperature region the activation process is nearly the same as for Γ(CH3β) of SAR1. This is expected due to the same mechanism of matrix proton motion, which induces the narrowing of the line width. On the other hand the broadening contribution is not detected in the monitored temperature region. The dynamics related to the motion of the entire radical found for the SAR1 center is absent for SAR2. The temperature dependence of the ENDOR line width of H1 of the NH3+ group only shows broadening contributions due to slow motional dynamics of the NH3+ group, as was expected and discussed earlier.29 However, the broadening is related to two activation processes (Figure 7 and Table 2). The transition from one activation process to the other is detected at temperatures around 220 K. The activation process at higher temperatures (T > 220 K) coincides with the activation process of the SAR1 center (Table 2). The dynamic behavior of the SAR1 center as a whole was noted and discussed earlier.25 The results obtained here for the first time show clear experimental evidence for the different dynamic behavior of the SAR1 and the SAR2 centers. The underlying structures of these centers in the crystal lattice can easily explain the difference. In the case of SAR2, all hydrogen bonds with the lattice are preserved, while in the case of SAR1, due to rupture of the C-NH3+ bond, only hydrogen bonds at one side of the radical (at the COO- group) are expected. Thus, unequal bond strengths along the intermolecular network are responsible for the increased mobility of the SAR1 center in comparison to the SAR2 center. Despite the fact that X-ray and neutron single-crystal diffraction measurements support the same crystal structure at low temperatures (20 K) and at room temperature,1-3 there is other experimental evidence for lattice instability at around 220 K. For example: birefringence and light depolarization measurements39 indicate some subtle symmetry breaking as does the change of the T1 relaxation time of the amine group detected by 1H NMR spectroscopy around the same temperature.44-46 A possible explanation for this phenomenon is the strong dynamic Jahn-Teller effect due to the NH3+ charge-lattice coupling.6 Indeed, Figure 7 shows a slight change in the broadening activation process of the amino protons of the SAR2 center (Table 2) at temperatures around 220 K. The activation process for T < 220 K (∆E ) 2486 K ) 1728 cm-1) is comparable with the expected deformation mode39,40 δ(NH3+) of the amine group in the crystal (∼1700 cm-1). The deformation does affect hydrogen bonds in the lattice and contributes a small variation of the dipole moment of each molecule allowed by the symmetry

First Stable L-Alanine Radical of the lattice. The additional activation process of the amine group detected at higher temperatures (T > 220 K, ∆E ) 5668 K ) 3939 cm-1), relates to the enhanced motional dynamic of the NH3+ group of SAR2. Its activation energy is comparable to the activation energy (∆E ) 4641 K) of the motional correlation time of the NH3+ group in the undamaged crystal, which has been detected by proton NMR.42,43 Thus, the amine group of SAR2 could be considered as a probe for lattice instability around 220 K. Moreover, the detection of nearly the same broadening activation process for the SAR1 center (HR and CH3β) and theNH3+ group of SAR2 (Table 2) in the vicinity of room temperature could be explained by a similar mechanism, by which the charged amine group modulates the motion. 4. Conclusion Earlier studies10,15,25,29 on the dynamical properties of the paramagnetic centers in γ-irradiated alanine indicate the possibility of dynamics of the entire SAR1 center. These indications were based on the presence of nonisotropic correlation times of the CH3 group at low temperatures (200-300 K) and led to the proposal of two conformational states of SAR1, which are present at higher temperatures35 (300-450 K). However, a detailed picture of the origin of these motional dynamics remained elusive. In the present study the temperature dependence of the HR HFCC as well as ENDOR line widths of the SAR1 center were studied in the low temperature region (200-320 K). The obtained results yield new insights on the complex dynamic behavior of the SAR1 center. As was indicated, the R-proton HFCC of SAR1 exhibits a temperature dependence, which is related to the thermodynamical average of possibly two conformations differing in the pyramidalization of the CR. Moreover, the activation process for this thermal averaging coincides with the expected rocking motions of the COOH group, which is modulated by hydrogen bonding perpendicular to the CO1O2- plane. It should be noted that these hydrogen bonds only bind the center to the lattice from one side (hydrogen bonds of CO1O2- group). This differs from the SAR2 center and the undamaged molecule, for which hydrogen bonds are present on both sides of the molecule. Finally, we have detected the same activation process for the NH3+ group of SAR2 and the HR and CH3β protons of SAR1 in the vicinity of room temperature (Figure 7 and Table 2). This strongly suggests that the ionic NH3+ groups modulate the motional dynamics of the entire SAR1 center. The differences detected for the dynamic behavior of SAR1 in comparison to SAR2 support a larger variety of dynamical modes related to the SAR1 center. The above findings explain the origin of the complex dynamic behavior of the SAR1 center in the L-alanine lattice and supply evidence related to the lattice instability due to the dynamic Jahn-Teller effect at around 220 K. Acknowledgment. This work was supported by grants from the Croatian Ministry of Science, Education and Sports, Grant No. 098-0982915-2939, and the Scientific Bilateral Collaboration Croatia-Austria. References and Notes (1) Simpson, H. J.; Marsh, R. E. Acta Crystallogr. 1966, 20, 550.

J. Phys. Chem. A, Vol. 114, No. 28, 2010 7505 (2) Destro, R.; Marsh, R. E.; Bianchi, R. J. Phys. Chem. 1988, 92, 966. (3) Lehmann, M. S.; Koetzle, T. F.; Hamilton, W. C. J. Am. Chem. Soc. 1972, 94, 2657. (4) Monaco, S. B.; Davis, L. E.; Velsko, S. P.; Wang, F. T.; Eimerl, D.; Zalkin, A. J. Cryst. Growth 1987, 85, 252. (5) Regulla, D. Appl. Radiat. Isot. 2000, 52, 1023. (6) Barthes, M.; Bordallo, H. N.; De´noyer, F.; Lorenzo, J.-E.; Zaccaro, J.; Robert, A.; Zontone, F. Eur. Phys. J. B 2004, 37, 375. (7) Miyagawa, I.; Gordy, W. J. Chem. Phys. 1960, 32, 255. (8) Morton, J. R.; Horsfield, A. J. Chem. Phys. 1961, 35, 1142. (9) Horsfield, A.; Morton, J. R.; Whiffen, D. H. Mol. Phys. 1962, 5, 115. (10) Miyagawa, I.; Itoh, K. J. Chem. Phys. 1962, 36, 2157. (11) Sinclair, J. W.; Hanna, M. W. J. Phys. Chem. 1967, 71, 84. (12) Dzuba, S. A.; Salikhov, K. M.; Tsvetkov, Yu. D. Chem. Phys. Lett. 1981, 79, 568. (13) Kuroda, S.-i.; Miyagawa, I. J. Chem. Phys. 1982, 76, 3933. (14) Matsuki, K.; Miyagawa, I. J. Chem. Phys. 1982, 76, 3945. (15) Brustolon, M.; Cassol, T.; Micheletti, L.; Segre, U. Mol. Phys. 1986, 57, 1005. (16) Nakagawa, K.; Eaton, S. S.; Eaton, G. R. Appl. Radiat. Isot. 1993, 44, 73. (17) Angelone, R.; Forte, C.; Pinzino, C. J. Magn. Reson., Ser. A 1993, 101, 16. (18) Brustolon, M.; Segre, U. Appl. Magn. Reson. 1994, 7, 405. (19) Ghim, B. T.; Du, J.-L.; Pfenninger, S.; Rinard, G. A.; Quine, R. W.; Eaton, S. S.; Eaton, G. R. Appl. Radiat. Isot. 1996, 47, 1235. (20) Rakvin, B. Appl. Radiat. Isot. 1996, 47, 1251. (21) Maltar-Strmecˇki, N.; Rakvin, B. Appl. Radiat. Isot. 2005, 63, 375. (22) Sagstuen, E.; Hole, E. O.; Haugedal, S. R.; Nelson, W. H. J. Phys. Chem. A 1997, 101, 9763. (23) Rakvin, B.; Maltar-Strmecˇki, N. Spectrochim. Acta, Part A 2000, 56, 399. (24) Rakvin, B.; Maltar-Strmecˇki, N.; Cevc, P.; Arcˇon, D. J. Magn. Reson. 2001, 152, 149. (25) Maltar-Strmecˇki, N.; Rakvin, B.; Cevc, P.; Arcˇon, D. Appl. Magn. Reson. 2002, 22, 551. (26) Harbridge, J. R.; Eaton, S. S.; Eaton, G. R. J. Phys. Chem. A 2003, 107, 598. (27) Rakvin, B.; Maltar-Strmecˇki, N.; Ramsey, C. M.; Dalal, N. S. J. Chem. Phys. 2004, 120, 6665. (28) Rakvin, B.; Maltar-Strmecˇki, N. Chem. Phys. Lett. 2005, 415, 161. (29) Rakvin, B.; Maltar-Strmecˇki, N.; Kattnig, D.; Grampp, G. Mol. Phys. 2007, 105, 2087. (30) Ban, F.; Wetmore, S. D.; Boyd, R. J. J. Phys. Chem. A 1999, 103, 4303. (31) Lahorte, P.; De Proft, F.; Vanhaelewyn, G.; Masschaele, B.; Cauwels, P.; Callens, F.; Geerlings, P.; Mondelaers, W. J. Phys. Chem. A 1999, 103, 6650. (32) Pauwels, E.; Van Speybroeck, V.; Lahorte, P.; Waroquier, M. J. Phys. Chem. A 2001, 105, 8794. (33) Declerck, R.; Pauwels, E.; Van Speybroeck, V.; Waroquier, M. Phys. ReV. B 2006, 74, 245103. (34) Itoh, K.; Miyagawa, I. J. Mol. Struct. 1988, 85, 190. (35) Freed, J. H. Multiple Electron Resonance Spectroscopy; Dorio, M. M., Freed, J. H., Eds.; Plenum Press: New York, 1979; p 73. (36) Vrielinck, H.; De Cooman, H.; Tarpan, M. A.; Sagstuen, E.; Waroquier, M.; Callens, F. J. Magn. Reson. 2008, 195, 196. (37) Erling, P. A.; Nelson, W. H. J. Phys. Chem. A 2004, 108, 7591. (38) Dobbs, A. J.; Gilbert, B. C.; Norman, R. O. C. J. Chem. Soc., Perkin Trans. 2 1972, 6, 786. (39) Barthes, M.; Vik, A. F.; Spire, A.; Bordallo, H. N.; Eckert, J. J. Phys. Chem. A 2002, 106, 5230. (40) Vik, A. F.; Yuzyuk, Y. I.; Barthes, M.; Sauvajol, J.-L. J. Raman Spectrosc. 2005, 36, 749. (41) Pauwels, E.; Vertraelen, T.; De Cooman, H.; Van Speybroeck, V.; Waroquier, M. J. Phys. Chem. B 2008, 112, 7618. (42) Hoffmann, S. K. Mol. Phys. Rep. 2005, 41, 9. (43) Andrew, E. R.; Hinshaw, W. S.; Hutchins, M. G.; Sjo¨blom, R. O. I.; Canepa, P. C. Mol. Phys. 1976, 32, 795. (44) Long, J. R.; Sun, B. Q.; Bowen, A.; Griffin, R. G. J. Am. Chem. Soc. 1994, 116, 11950. (45) Beshah, K.; Olejniczak, E. T.; Griffin, R. G. J. Chem. Phys. 1987, 86, 4730. (46) Jackson, P.; Harris, R. K. J. Chem. Soc., Faraday Trans. 1995, 91, 805.

JP103883X