Differentiating Two Nitrosylruthenium Isomeric Complexes by Steady

Oct 18, 2016 - The [Ru(II)–NO+] group affects the structure and chemical reactivity of nitrosylruthenium(II) complexes. A characteristic infrared ab...
0 downloads 0 Views 3MB Size
Article pubs.acs.org/JPCB

Differentiating Two Nitrosylruthenium Isomeric Complexes by Steady-State and Ultrafast Infrared Spectroscopies Pengyun Yu,†,‡,∥ Yan Zhao,†,§,∥,⊥ Fan Yang,†,∥ Huifen Pan,§ Jianru Wang,§ Juan Zhao,† Wenming Wang,§ Hongfei Wang,*,§ and Jianping Wang*,†,‡ †

Beijing National Laboratory for Molecular Sciences, Molecular Reaction Dynamics Laboratory, Institute of Chemistry, the Chinese Academy of Sciences, Beijing, 100190, P. R. China ‡ University of Chinese Academy of Sciences, Beijing 100049, P. R. China § Institute of Molecular Science, Shanxi University, Taiyuan, 030006, P. R. China ABSTRACT: The [Ru(II)−NO+] group affects the structure and chemical reactivity of nitrosylruthenium(II) complexes. A characteristic infrared absorption band due to the nitrosyl (NO) stretching motion is shown in the frequency region 1800−1900 cm−1. In this work, linear infrared (IR) and nonlinear IR methods, including pump−probe and two-dimensional (2D) IR, were utilized to study the structures and dynamics of two isomeric nitrosylruthenium complexes [Ru(OAc)(2mqn)2NO] (H2mqn = 2-methyl-8-quinolinol) in cis and trans isomeric configurations in a weak polar solvent (CDCl3). Using the NO stretching mode as a vibrational probe, information about local structural dynamics of the Ru complex as well as solvent fluctuation dynamics was obtained. In particular, a “structured” solvent environment is believed to form in the vicinity of the NO group in the case of the cis isomer with the aid of a neighboring OAc ligand, which is the reason for more efficient vibrational relaxation but more inhomogeneously distributed solvent and thus associated slower spectral diffusion. Our results also suggest a more anharmonic potential surface for the NO stretching mode in the less stable trans isomer.

1. INTRODUCTION Nitric oxide (NO) plays key regulatory roles in mammalian biology; it has been shown to be an important signaling molecule in a wide variety of physiological and pathological processes, including the modulation of the immune and endocrine response, regulation of blood pressure, neurotransmission, and induction of apoptosis.1−3 The coordination of NO with metal cation is critical to the stability and activity of NO. There is considerable interest in designing transitionmetal−NO complexes to regulate the reactivity of NO.4−7 For example, Sodium nitroprusside has been studied extensively for probing Fe−NO bonding in several heme−NO proteins including myoglobin and nitrophorins.8−10 In comparison with the Fe−NO complexes, the Ru−NO complexes are more photoactive, which raised interest in the study of nitrosylruthenium complexes for both chemists and biologists. In a nitrosylruthenium complex, a generally accepted picture is NO+ bound to the Ru(II) center, largely based on the observation of a high-frequency infrared (IR) active NO stretching mode (νNO = 1820−1950 cm−1),6 vs free NO (∼1750 cm−1) or bound NO• (1650−1750 cm−1) in the Ru− NO species.8 Previous experiments and ab initio computations also indicate that the peculiarity of the [Ru(II)−NO+] group affects the structure and chemical reactivity of the nitrosylruthenium complexes. In the meantime, the frequency of the © 2016 American Chemical Society

nitrosyl stretching vibration (νNO) in nitrosyl metal complexes is very sensitive to the isomeric conformations and their local chemical environment. Therefore, the NO stretching mode becomes a characteristic IR spectroscopic probe to study the structure and reactivity of these complexes. In recent years, time-resolved IR spectroscopy, in particular the two-dimensional infrared (2D IR) method, has been used to unravel the ultrafast structures and dynamics of condensedphase molecules, particularly on the transition metal complexes,11−19 on the femtosecond to picosecond time scales. In general, the absorption line shape of 2D IR spectra contains the homogeneous and inhomogeneous contributions. Vibrational spectral diffusion (SD) is the time evolution of the frequency of a given vibration probe extracted from the inhomogeneously broadened absorptive 2D IR spectrum caused by equilibrium structural fluctuations; thus, it is a measure of the equilibrium structural dynamics of solute or solvent or both.20,21 The changes of the diagonal 2D IR spectral line shape caused by spectral diffusion are a measure of the frequency timecorrelation function (FTCF),22−24 which can be obtained by the inverse of the center line slope (CLS)25 of the timeReceived: August 10, 2016 Revised: September 27, 2016 Published: October 18, 2016 11502

DOI: 10.1021/acs.jpcb.6b08060 J. Phys. Chem. B 2016, 120, 11502−11509

Article

The Journal of Physical Chemistry B

species. Thus, the neighborhood of the NO group is structurally different in the two isomers. The chemical reactivity of two isomers can be different in such a way that, for example, upon photoexcitation, the cis isomer quickly changes into the trans conformation, while the trans can only slowly change into the cis isomer. Moreover, the cis isomer is generally easy to be synthesized with reasonable yield while the trans isomer is not, suggesting that the cis conformation is thermodynamically more stable.31 In this work, weak polar solvent with adequate solubility, that is, CDCl3, is used. Both linear and nonlinear IR experiments on the two isomers were carried out. The NO vibrational frequency, spectral line width, vibrational relaxation dynamics, and spectral diffusion dynamics were measured and compared for the two isomeric species, in order to understand the differences in reactivities of these isomeric nitrosylruthenium(II) complexes on the structural basis.

dependent 2D IR spectra, for instance. Thus, the FTCF is not only a quantitative description of the 2D IR spectral change but also a connection between the 2D IR spectroscopy and molecular structures and structural dynamics. In recent years, the applications of linear IR and 2D IR on nitrosyl complexes have been reported, including the solvatochromic behavior of Fe−NO complexes,26 metal− nitrosyl linkage isomerism in sodium nitroprusside,27 solvent polarity effect on vibrational dephasing dynamics of nitrosyl,28 structural dynamics in the distal pocket of nitrophorin 4,29 and vibrational frequency fluctuations of nitrosyl in aqueous solution.30 In our recent work,18 the structural dynamics of two cis [Ru(OAc)(2cqn)2NO] (H2cqn = 2-chloro-8-quinolinol) isomers were studied, by focusing on the NO stretching mode. One question remains, however; that is, can we derive any additional IR signatures of structures and/or dynamics for these isomeric Ru−NO complexes from a nonlinear IR spectroscopy method such as 2D IR? In other words, can we differentiate two similarly structured isomers based on their 2D IR spectral features? To this end, two structurally different (cis and trans isomeric) [Ru(OAc)(2mqn)2NO] (H2mqn = 2methyl-8-quinolinol) compounds were chosen, and their molecular structures are shown in Figure 1. There are subtle

2. EXPERIMENTAL SECTION 2.1. Materials. Cis [Ru(OAc)(2mqn)2NO] was synthesized using a previously described procedure,32 and the trans [Ru(OAc)(2mqn)2NO] complex was prepared by photoisomerization from the cis isomer, followed by further HPLC purification. The structures of both isomers were confirmed by proton NMR experiment. All solvents were purchased from Sigma-Aldrich and/or from locally available sources and were used without further purifications. In particular, the purity of CDCl3 is 99.8% in D. In addition, our NMR studies also indicate that DMSO may act as a ligand and gradually change the structure and composition of the complex. Thus, in this work, DMSO was not considered as a solvent. Moreover, it should be noted that there is 9% or so of the trans isomer presented in the cis isomeric sample if a spectroscopic measurement is carried out under room light, as shown by FTIR spectra (see below). 2.2. Linear IR Spectroscopy. The cis or trans isomeric complex was dissolved in CDCl 3 at nearly identical concentration (12.6 mM for the trans and 12.5 mM for the cis). Then, the sample solutions were loaded into a homemade IR sample cell containing two CaF2 windows of 25 mm in diameter and 2 mm in thickness separated by a 50 μm Teflon spacer. The FTIR (linear IR) spectra of the cis and trans isomers in the mid-IR frequency region were collected using a Nicolet 6700 spectrometer (Thermo Electron) at 0.4 cm−1 resolution. All FTIR experiments were performed at room temperature (22 °C). 2.3. Nonlinear IR Spectroscopy. The nonlinear IR experiments, namely, IR pump−probe and 2D IR, were performed using two independent experimental apparatuses. The IR pump−probe experiment was carried out using a homebuilt nonlinear IR spectrometer as described earlier.24,33 Briefly, a commercial ultrafast Ti:sapphire laser amplifier was used to generate a 3 mJ, sub 35 fs, 800 nm pulse at a repetition rate of 1 kHz. The laser was used to pump an optical parametric amplification (OPA). The signal and idler pulses in the near IR region generated by the OPA were used to generate a 70 fs (full width at half-maximum, fwhm), 5 μJ mid-IR pulse by a difference frequency generation (DFG) device using a 0.5 mm thick type-II AgGaS2 nonlinear crystal. The obtained IR pulse was tuned to a center frequency at 1850 cm−1 with a spectral width (fwhm) of ca. 240 cm−1. The mid-IR pulse was then split into two excitation beams with k1 (with a pulse energy of ca. 400 nJ/pulse) as pump and k2 (attenuated to much weaker

Figure 1. Structures of the cis (left) and trans (right) isomers of [Ru(OAc)(2mqn)2NO] (H2mqn = 2-methyl-8-quinolinol). The two isomers are defined on the basis of the relative positions of N and O atoms of the quinolinol ring connecting the Ru center. To show more clearly the relative positive positions of the NO ligand with respect to other ligands, two different drawings are provided, with ligating O atoms labeled.

but important structural differences between these isomers. In the cis isomer, the two ligating O atoms (O1 and O2) in the two 2mqn groups are in cis positions, while the two ligating N atoms in the two 2mqn groups are in trans positions. This is seen in Figure 1 (more clearly in the lower-left corner). On the other hand, in the trans isomer, the two ligating O atoms (O1 and O2) in the two 2mqn groups are in trans positions, and their two ligating N atoms are also in trans positions. This is seen in Figure 1 (more clearly in the lower right corner). Moreover, having the ON−Ru−OAc (where O3 is the ligating atom) more or less form the C2 axis, the trans isomer appears to be more symmetric than the cis isomer, although both cis and trans belong to Cs molecular point symmetry. In such a simplified picture, the free O atom of the OAc group and the NO group appear to be very far away from each other in the trans isomer, while they are very close to each other in the cis 11503

DOI: 10.1021/acs.jpcb.6b08060 J. Phys. Chem. B 2016, 120, 11502−11509

Article

The Journal of Physical Chemistry B

transition intensity, and anharmonicity of the NO stretching mode of these Ru complexes were computed. The polarizable continuum model (PCM)37 for solvent (ε = 4.71 for CDCl3) was used for the geometry optimization and frequency calculations. The ab initio calculations were performed using Gaussian 09.38

pulse energy, which was estimated to be tens of nJ per pulse) as probe. The two pulses were focused on the sample solution by a parabolic mirror with 100 mm focal length, and the emitted pump−probe signal was detected along the k2 direction. An IR monochromator equipped with a 64-element liquid-nitrogencooled mercury−cadmium−telluride (MCT) array detector was used to collect the pump−probe signal. In the current work, magic angle (54.7°) polarization condition was used in order to obtain the vibrational lifetime (T1). The IR pulse shaper-based 2D IR spectrometer is known to be more efficient in 2D IR data acquisition. In this work, 2D IR spectra of the cis and trans Ru−NO complexes were collected using a commercial 2D IR spectrometer (2DQuick, PhaseTech).34 Briefly, a commercial ultrafast Ti:sapphire laser amplifier with 3 mJ, 30 fs, 800 nm pulses at a repetition rate of 1 kHz was used to pump a commercial optical parametric amplifier (TOPAS) to generate mid-IR pulses with a typical pulse energy of 12−15 μJ, frequency centered at 1850 cm−1, and a spectral width (fwhm) of ca. 270 cm−1. However, only part of the output (6 μJ) was used in the 2D IR experiment. The IR pulse was first split into a strong pump pulse and a weak probe pulse using a 90:10 ZnSe beam splitter, which is aligned using a He−Ne laser beam. The pump pulse enters an IR pulse shaper with ca. 40% overall output efficiency and generates a collinear pulse pair. The total pulse energy of the pump pair and that of the probe were measured to be typically ca. 0.4 μJ and less than 0.1 μJ, respectively, at the focus point and under parallel polarization. Detailed descriptions and methods of alignment, calibration, and optimization of the output pulses for the IR pulse shaper can be found elsewhere.35 The intensity cross-correlation measurement of the pump and probe pulses showed an overall pulse duration (FHWM) of 70−100 fs typically, depending on the status of the pulse shaper and the specification of the input IR pulse. Nearly Fourier transform limited pulses can be obtained with a careful tuning of the second- and third-order dispersion coefficients.36 2D IR signal was generated in the pump−probe geometry, and was collected through an IR monochromator equipped with a 64-by-2 MCT IR array detector. The top-row MCT array was used for data acquisition, and the bottom-row array could be turned on and off as an instantaneous reference. However, in the present work, no reference was used because the 2D IR signal-to-noise ratio appeared to be reasonably good. The waiting time (T, between the pump pair and probe) was controlled by a combination of a ZnSe wedge pair (in the pump beam path) and a translation stage (in the probe beam path). Using a 100 line/mm IR grating, the detection frequency resolution was found to be better than 2 cm−1 for the pump−probe experiment. All nonlinear IR measurements were performed at room temperature (22 °C). FTIR spectra of samples were measured before and after the 2D IR experiment, and no spectral changes were observed in the two samples. Also, this work only focuses on the NO stretching mode, rather than on the interaction between that and CO stretching modes of the OAc ligand. 2.4. Ab Initio Calculations. Initial atomic coordinates of the cis and trans [Ru(OAc)(2mqn)2NO] complexes were obtained from the crystal structures determined by X-ray diffraction.32 All of the structures were fully optimized using the density functional theory (DFT), at the level of the B3LYP functional with the 6-311++G (d,p) basis set for the ligand atoms and the lanl2dz pseudopotential for the core electrons of the Ru atom. Anharmonic vibrational transition frequency,

3. RESULTS AND DISCUSSION 3.1. Linear IR Spectra and Band Assignment. The linear IR absorption spectra of the cis and trans isomers in CDCl3 are shown in Figure 2. In this solvent, there was only one

Figure 2. Normalized FTIR spectra of the cis and trans isomers of [Ru(OAc)(2mqn)2NO] in the NO stretching region in CDCl3. Voigt function fittings are shown in dashed lines. There is ca. 9% trans component presented in the cis conformer sample.

absorption peak in the NO stretching region for both the cis and trans isomers. The NO stretching frequency of the cis isomer is ωNO,cis = 1859 cm−1, and that of the trans isomer is ωNO,trans = 1844 cm−1. Using the Voigt line-shape function, the spectrum can be fitted reasonably well for the trans conformer. However, a small trans component (ca. 9%) is needed to fit the IR spectrum of the cis conformer, which was determined by fitting the IR spectrum, and on the basis of the theoretically estimated transition intensities of the two isomers shown in Table 2 (see below). Fitting parameters are listed in Table 1. The fwhm of the cis isomer is found to be slightly broader than that of the trans isomer. The frequency difference and the line width difference are explained below. Table 1. Experimentally Determined Vibrational Transition Frequency (ω0) and Voigt Line Width (FHWM) of the NO Stretching Mode from the Linear IR Spectra of the cis and trans Isomers of [Ru(OAc)(2mqn)2NO] in CDCl3 isomer

ωNO (cm−1)

FHWMVoigt (cm−1)

cis trans

1859 1844

17 16

The bonding between a linear NO ligand and a metal atom is widely described in terms of a synergic picture which is made up of the following components: (a) donation of electron density from a σ-type orbital of NO to a vacant d-orbital of the metal center, in addition to (b) donation of electron density from the occupied metal d-orbitals to the π*-antibonding orbitals of the NO ligand.39−41 In the case of terminal linear nitrosyls, the electron donation from the nitrosyl group to the Ru metal center in the form of [RuII−NO+] causes an increase of the NO bond order and thus a higher NO stretching frequency (ωNO). The observed ωNO,trans is lower than ωNO,cis, suggesting that the trans configuration has a weaker NO bond. Such a picture is consistent with the well-known trans11504

DOI: 10.1021/acs.jpcb.6b08060 J. Phys. Chem. B 2016, 120, 11502−11509

Article

The Journal of Physical Chemistry B strengthening effect for linear nitrosylruthenium(II) complexes,42,43 where the NO group acts as a strong π-electron acceptor and the phenolato oxygen acts as a π-electron donor (see section 3.5 for more details). The band assignment is supported by ab initio calculations (Table 2). Computations show that the force constant (k) and Table 2. Vibrational Transition Frequency (ωNO), Transition Intensity (I), Transition Dipole Moment (Δμ), Anharmonicity (Δ), and Force Constant (k) of the NO Stretching Mode of cis and trans Isomeric [Ru(OAc)(2mqn)2NO] in CDCl3 Using the PCM Solvent Model isomer

ωNO (cm−1)

I (km·mol−1)

Δμ (D)

Δ (cm−1)

k (N·cm−1)

cis trans

1925.8 1908.3

1722.9 1617.0

0.597 0.581

26.2 26.2

31.8 31.5

Figure 3. Population relaxation dynamics of the NO stretching mode of the cis isomer (a) and trans isomer (b) from magic-angle pump− probe IR experiment, and their single exponential fittings. Negative, the 1 → 2 transitions; positive, the 0 → 1 transitions. Probing frequencies are listed in Table 3.

Table 3. Fitting Parameters for the Vibrational Population Relaxation Dynamics (T1) at Probing Frequency (ωprobe) and Spectral Diffusion Time (τSD) of the NO Stretching Mode of the cis and trans Isomers of [Ru(OAc)(2mqn)2NO] in CDCl3

the vibrational transition dipole moment (Δμ) of the NO stretching mode of the cis and trans isomers in CDCl3 exhibit some differences. First, the values of Δμ and k of the cis isomer are slightly larger than those of the trans isomer (Table 2). Associated with a marginally smaller k, the trans isomer shows a lowered vibrational transition frequency. Also associated with a smaller Δμ, the trans isomer shows a lowered transition intensity. In addition, a subtle line width difference between the cis and trans isomers is also observed (17 cm−1 vs 16 cm−1, Table 1), which can be explained simply by the local chemical environment effect on the NO ligand. This can be seen easily in Figure 1. First, as mentioned earlier, the geometric configuration of the cis isomer allows a structured CDCl3 solvent cluster to be formed in the vicinity of the NO group, with the aid of the free O atom in the OAc ligand. Additionally, in the cis isomer, one of the methyl groups of the two 2mqn ligands is pointing toward the NO group on the opposite side of the OAc group. This prevents more solvent molecules from approaching the NO group. While in the trans isomer, the two methyl groups are in opposite positions and sitting on the “equatorial” plane formed by the two 2mqn groups, in which case, the geometrical obstructive effect on solvent is not as severe as that in the cis isomer. As a consequence, the local structural environment of the NO group and its solvent configuration vary substantially in the two isomers. This explains the slightly broadened line width observed in the cis isomer of the FTIR spectra (Figure 2). 3.2. Magic-Angle Pump−Probe Spectra and Vibrational Relaxation Dynamics. The magic-angle pump−probe spectra of the cis and trans isomers in CDCl3 in the NO stretching frequency region were collected. The results are shown in Figure 3 for a few dynamic traces probed at selected frequencies. Here, in each case, the positive signal represents the ground-state bleaching (v = 0 → v = 1, or simply 0 → 1, where v is vibrational quantum number), while the negative signal represents the first exited-state absorption (v = 1 → v = 2, or simply 1 → 2). The two peaks appear at different frequency positions as a result of the vibrational anharmonicity; i.e., the energy gap of the 1 → 2 transition is smaller than that of the 0 → 1 transition, so the former appears at the lowfrequency side. Each curve of dynamics can be fitted reasonably using only one exponential function, and the fitting parameters for the two isomers are listed in Table 3.

spectral diffusion time

lifetime measurement ωprobe (cm−1)

τSD (ps)

T1 (ps)

isomer

ν0→1

ν1→2

ν0→1

ν1→2

ν0→1

cis trans

1862 1846

1835 1820

7.3 ± 0.2 9.0 ± 0.2

8.1 ± 0.2 11.9 ± 0.2

4.8 ± 0.6 3.5 ± 0.2

Because the delay time is the mismatch between the strong pump pulse and weak probe pulse, the vibrational relaxation dynamics for the NO stretching vibration under magic-angle conditions can be used to evaluate the vibrational population relaxation time constant (i.e., lifetime, T1) without the influence of orientation relaxation. In general, the vibrational energy relaxation can occur first through both intra- and/or intermolecular energy dissipation channels, then through solvent modes eventually, among which the more efficient one will dominate. In CDCl3, the vibrational lifetime of the cis isomer was found to be 7.7 ps on average of the values for the 0 → 1 and 1 → 2 transitions, which was faster than that of the trans isomer (10.5 ps on average). Their different T1 times can be explained as a structural or geometric effect: for the vibrational relaxation of the excited NO stretching mode, through CDCl3 solvent is believed to be a more effective vibrational energy dissipation pathway. This is because NO and CDCl3 may form a weak O···D deuterium bond, which establishes a direct and relatively efficient intermolecular vibrational relaxation route. As has been pointed out earlier, there is possibly a “structured” solvent environment formed in the vicinity of the NO group only in the case of the cis isomer. Such a structured CDCl3 cluster favors a more efficient vibrational relaxation of the vibrationally excited NO state. However, for the trans species, this is not the case; the solvent molecules are likely to be randomly distributed around the NO group except for one that forms a weak O···D bond. This seems to be one of the reasons that the T1 time for the cis isomer is shorter. In addition, such a structured CDCl3 cluster in the vicinity of the NO group in the cis species also leads to a more inhomogeneous broadening, which is in perfect agreement with a more broadened linear IR spectra width in the cis isomer (Figure 2, Table 1). 11505

DOI: 10.1021/acs.jpcb.6b08060 J. Phys. Chem. B 2016, 120, 11502−11509

Article

The Journal of Physical Chemistry B

Figure 4. Pure absorptive 2D IR spectra of the NO stretching mode of the cis isomer of [Ru(OAc)(2mqn)2NO] in CDCl3 at various waiting times. Spectra are normalized by the intensity of the red positive peak. The dashed line indicates diagonal.

Figure 5. Pure absorptive 2D IR spectra of the NO stretching mode of the trans isomer of [Ru(OAc)(2mqn)2NO] in CDCl3 at various waiting times. Spectra are normalized by the intensity of the red positive peak. The dashed line indicates diagonal.

insignificant cis component seen in the fitting results in Figure 2. The structural fluctuations in both solute and solvent molecules will cause a fluctuation in vibrational frequency of the NO stretching mode. The equilibrium distribution of the frequency fluctuations of the 0 → 1 and 1 → 2 transitions is the reason for the observed line-shape and orientational changes of the 2D IR diagonal signals as a function of the waiting time. As seen from Figures 4 and 5, at a very short waiting time, the 2D IR peaks are elongated along the diagonal of the 2D plot. Such an elongation indicates the presence of the inhomogeneous broadening (plus homogeneous broadening) in the 2D line shape, in which the inhomogeneous contribution is caused by the structural fluctuations. At similar waiting time, the elongation for the cis and trans isomers differs, indicating different a time dependence of the structural inhomogeneity, which can be roughly quantified by the initial value of the FTCF (see below). The diagonal peaks become less elongated with increasing waiting time and eventually become upright, manifesting the presence of vibrational SD dynamics. Such a process seems to be slower for the cis isomer than for the trans isomer because, as viewed from Figures 4 and 5, the 2D signal becomes nearly upright oriented at 10 ps for the latter but not yet so for the former. The SD process can be quantitatively characterized using the waiting-time-dependent 2D IR spectra. A convenient way is to measure the center-line slope (CLS)25 of either the 0 → 1 transition (red signals in Figures 4 and 5) or the 1 → 2 transition (blue signals in Figures 4 and 5). In this work, we use CLS of the 0 → 1 transitions. Here a CLS is the inverse of the slope of the line that connects the maxima of the peaks of a series of cuts through the red peak, which are parallel to the horizontal (ωt) axis. The extracted CLS data points for the NO stretching mode of the cis and trans isomers at various waiting times T are shown in Figure 6 and are fitted to a single-

Of course, one may also argue that a better vibrational energy match is needed between solvent modes (overtone and/or combinations) and the NO modes;44 however, this does not contradict with the proposed structured solvent model. In addition, there is no obvious strong interaction between nitrosyl and other ligands, except that between NO and the Ru center through back electron transfer. In order to relax the vibrational energy, the Ru-ligand vibration, which is quite low in transition energy, has to couple to the NO-ligand vibrations, with the latter coupled to solvent motions. Thus, this complex pathway does not seem to be an efficient vibrational energy relaxation pathway for the excited NO stretching mode. 3.3. Waiting-Time-Dependent 2D IR Spectra and Equilibrium Structural Dynamics. The pure absorptive 2D IR spectra of the cis and trans isomers in the NO stretching region are shown in Figure 4 and Figure 5, respectively, at several waiting times. Note that the 2D frequency window of the cis species is made to differ from that of the trans species in order to center each 2D IR signal. Each 2D IR spectrum contains a pair of peaks arising from the 0 → 1 transition (red) on the diagonal (ωτ = ωt) and the 1 → 2 transition (blue) shifted along the horizontal (ωt) axis due to the anharmonicity of the NO stretching mode. Here, ωτ is obtained by Fourier transform of the 2D IR time-domain signal along the coherent time τ, and ωt is obtained by Fourier transform of the 2D IR time-domain signal along the detection time t. These 0 → 1 signals are slightly off the diagonal because of the overlap with the corresponding 1 → 2 signals. Further, because there is only one anharmonic oscillator (i.e., the NO stretching), there is only one pair of diagonal signals appearing in each panel of Figures 4 and 5, and no off-diagonal signals are involved. However, in Figure 4, a small portion of 2D IR signal resulting from the trans isomer in the case of the cis isomer is still seen in the lower left corner of each panel, which agrees with the 11506

DOI: 10.1021/acs.jpcb.6b08060 J. Phys. Chem. B 2016, 120, 11502−11509

Article

The Journal of Physical Chemistry B exponential decay (solid curve). The obtained spectral diffusion time constants are listed in Table 3.

Figure 7. Anharmonicity fitting of the NO stretching mode using the magic-angle pump−probe slices at 1.0 ps delay time: (a) cis isomer; (b) trans isomer. Black circles, experimental data; red curve, overall fitting; positive, 0 → 1 transitions; negative, 1 → 2 transitions. A pair of weak signals (ca. 9% of the total) due to the trans isomer are shown in panel a. A factor of −1 is applied on the pump−probe signal so that its signal polarity can be directly compared with that of 2D IR.

Figure 6. Center-line slope dynamics of the NO stretching mode of the cis isomer (a) and trans isomer (b) extracted from the 0 → 1 transition of the waiting-time-dependent 2D IR signal of [Ru(OAc)(2mqn)2NO] in CDCl3. The solid lines are single exponential fits to experimental data (circles).

Experimental results suggest that the anharmonic surface of the NO stretching oscillator is slightly different for the two isomeric species, which must be due to local chemical structural reasons and/or solute−solvent interaction reasons. The computed anharmonicities of the two Ru−NO species are listed in Table 2, in CDCl3 PCM solvent. The predicted value is 26.2 cm−1 for both isomers. Thus, computations only roughly reproduce the observed anharmonicities. 3.5. Structural and Nonlinear IR Characteristics of the cis and trans Isomers. Taking the results together, detailed structural and vibrational characteristics can be obtained. First, for both cis and trans complexes, the bond distance of the Ru O trans to the NO group is shorter than those cis to the NO group, which is observed in the crystal structure:45 in the crystal structure of the cis isomer, the RuO1 bond length is 1.97 Å, which is shorter than those of the RuO bonds that are cis to the NO group (RuO2 is 1.99 Å and RuO3 of the OAc group is 2.05 Å). On the other hand, in the crystal structure of the trans isomer, the bond length of RuO3 of the OAc group that is in the trans position to the NO group is 1.98 Å, which is shorter than those of RuO1 (2.02 Å) and RuO2 (2.04 Å). Such a phenomenon is known as the transstrengthening effect (as mentioned earlier) for the linear nitrosylruthenium(II) complexes because the NO group acts as a strong π-electron acceptor and the phenolato oxygen acts as a π-electron donor.42,43 These structural differences are also predicted by the DFT calculation (data not shown). These structural differences cause a subtle change in the electron state of the atom coordinated to the Ru center, which changes the RuN and NO bond lengths (slightly longer of the latter in trans isomer44), and thus leads to a lowered NO stretching frequency for the trans isomer. Further, it is interesting to compare the spectral parameters and the inbuilt structural and dynamics information between the Ru complexes studied here and studied previously. In our recent study,18 the IR spectrum of the NO mode of one of the cis isomers of [Ru(OAc)(2cqn)2NO] (H2cqn = 2-chloro-8quinolinol) (referred to as cis-1) was studied. The NO stretching mode in the presence of chloride (1877 cm−1, cis-1 isomer, in our recent study18) was found to be higher in frequency than that in the presence of a methyl group (1859 cm−1, cis isomer, in the present study), suggesting that the chemical groups around the NO group also affect the charge distribution of the Ru complex, and thus the NO bond order in these complexes.

The initial value of CLS (T = 0) is related to the total magnitude of inhomogeneous and homogeneous contributions in the line shape of the NO stretching mode. Figure 6 shows almost indistinguishable CLS (T = 0) values for both cis and trans forms, suggesting a similar spectral broadening, which generally agrees with the observation in the linear IR spectra that the fwhm in the cis form is marginally broader than that in the trans form (Figure 2 and Table 1). This actually indicates the insensitivity of CLS (T = 0) in reflecting the inhomogeneous contribution out of the total broadening. The time constant of the CLS obtained from the 0 → 1 transition, however, reflects the relaxation time of the vibrational spectral diffusion, which is purely due to the inhomogeneous broadening. It can be seen that, indeed from Figures 4 and 5, the 2D IR spectral orientation evolves relatively slower for the cis species than for the trans species. Moreover, the SD process can be used as a measure of the FTCF.22−24 The results in Figure 6 suggest that, for the NO stretching mode, the frequency time correlation of the 0 → 1 transition, i.e., c(t) = ⟨δω0−1(t)δω0−1(0)⟩, where δω0−1(t) = ω0−1(t) − ⟨ω⟩ and ⟨ω⟩ is the average frequency, relaxes slower in the cis species (4.8 ps) than in the trans species (3.5 ps). In addition, there is also a very small portion of the CLS relaxation (ca. 10%) at up to 15 ps, which could be due to a relatively slow structural fluctuation due to the Ru complex itself. The observed slightly broader line width in the cis form than in the trans form (17 cm−1 vs 16 cm−1) and slower vibrational frequency diffusion in the cis form than in the trans form (4.8 ps vs 3.5 ps) both can be explained as a structured solvent effect that the local environment of NO is more inhomogeneous in the cis form than in the trans form. 3.4. Anharmonicities. For a relatively isolated anharmonic oscillator, as far as the fundamental 2D IR spectroscopy is concerned, the diagonal anharmonicity is simply defined as Δ = ω0→1 − ω1→2, which is normally a positive value because the energy gap of the v = 1 → v = 2 transition is usually smaller than that of the v = 0 → v = 1 transition for the anharmonic oscillator. The transient spectra at 1 ps delay time are shown in Figure 7. By fitting the magic-angle pump−probe spectra at 1, 2, 3, 5, 8, 12, and 18 ps delay times, the anharmonicity of the NO-stretching mode is determined for each case. The results are 26.5 ± 0.2 and 28.5 ± 0.5 cm−1 for the cis and trans species, respectively, from the fitting of these data. 11507

DOI: 10.1021/acs.jpcb.6b08060 J. Phys. Chem. B 2016, 120, 11502−11509

Article

The Journal of Physical Chemistry B Furthermore, pump−probe and “static” (at a given waiting time T) and time-resolved 2D IR data add further details to the anharmonic vibrational characteristics of the NO stretching mode of these two isomers. A comparison of low vibrational levels of the NO stretching modes between the cis and trans isomers is shown in Figure 8.

biomolecules as spectroscopic markers for probing biostructures and their dynamics.



AUTHOR INFORMATION

Corresponding Authors

*Phone: (+86)-351-7018623. Fax: (+86)-351-7018623. E-mail: [email protected]. *Phone: (+86)-010-62656806. Fax: (+86)-010-62563167. Email: [email protected]. Author Contributions ∥

P.Y., Y.Z., F.Y.: Equal contributors.

Notes

The authors declare no competing financial interest. ⊥ Y.Z. is a one-year visiting student at MRDLab, ICCAS.



ACKNOWLEDGMENTS This work was supported by the Hundred Talent Fund of the Chinese Academy of Sciences (to J.W.), by the National Natural Science Foundation of China (21473212 to F.Y., 21543003 and 21671125 to H.W., and 21573243 to J.W.), and also by the Talent Plans of Shanxi Province and Shanxi Scholarship Council of China (to H.W.). The authors thank Ms. J. Liu for technical help.

Figure 8. Illustrative anharmonic potential surface of the NO stretching modes and the vibrational quantum levels in the cis and trans isomeric [Ru(OAc)(2mqn)2NO].

Static 2D IR data suggest a smaller vibrational anharmonicity of the NO stretching mode for the cis isomer, which is consistent with a more stable cis structure with a deeper and more harmonic potential surface than that for the trans structure. Thus, a relatively shallower potential surface and thus a larger anharmonicity are expected for the NO stretching mode in the trans isomer (Figure 8). The dynamical 2D IR data add more structural details to the NO−solvent interaction, in such a way that more inhomogeneity is expected in the vicinity of the NO group of the cis isomer because of the OAc and methyl groups in the neighborhood. The structured CDCl3 around the NO group is believed to be the reason for the faster T1 relaxation for the vibrationally excited NO state in the cis isomer, as observed in the pump−probe experiment.



REFERENCES

(1) Fukumura, D.; Kashiwagi, S.; Jain, R. K. The Role of Nitric Oxide in Tumour Progression. Nat. Rev. Cancer 2006, 6, 521−534. (2) Lundberg, J. O.; Gladwin, M. T.; Ahluwalia, A.; Benjamin, N.; Bryan, N. S.; Butler, A.; Cabrales, P.; Fago, A.; Feelisch, M.; Ford, P. C.; et al. Nitrate and Nitrite in Biology, Nutrition and Therapeutics. Nat. Chem. Biol. 2009, 5, 865−869. (3) Tennyson, A. G.; Lippard, S. J. Generation, Translocation, and Action of Nitric Oxide in Living Systems. Chem. Biol. 2011, 18, 1211− 1220. (4) Pereira, A. d. C.; Ford, P. C.; da Silva, R. S.; Bendhack, L. M. Ruthenium-Nitrite Complex as Pro-Drug Releases NO in a Tissue and Enzyme-Dependent Way. Nitric Oxide 2011, 24, 192−198. (5) Merkle, A. C.; McQuarters, A. B.; Lehnert, N. Synthesis, Spectroscopic Analysis and Photolabilization of Water-Soluble Ruthenium(iii)-Nitrosyl Complexes. Dalton Trans. 2012, 41, 8047− 8059. (6) Sharma, N.; Bhattarai, J. P.; Hwang, P. H.; Han, S. K. Nitric Oxide Suppresses L-Type Calcium Currents in Basilar Artery Smooth Muscle Cells in Rabbits. Neurol. Res. 2013, 35, 424−428. (7) Hatcher, L. E.; Raithby, P. R. Dynamic Single-Crystal Diffraction Studies Using Synchrotron Radiation. Coord. Chem. Rev. 2014, 277− 278, 69−79. (8) Spiro, T. G.; Soldatova, A. V.; Balakrishnan, G. CO, NO and O(2) as Vibrational Probes of Heme Protein Interactions. Coord. Chem. Rev. 2013, 257, 511−527. (9) Soldatova, A. V.; Ibrahim, M.; Olson, J. S.; Czernuszewicz, R. S.; Spiro, T. G. New Light on NO Bonding in Fe(III) Heme Proteins from Resonance Raman Spectroscopy and DFT Modeling. J. Am. Chem. Soc. 2010, 132, 4614−4625. (10) Nienhaus, K.; Maes, E. M.; Weichsel, A.; Montfort, W. R.; Nienhaus, G. U. Structural Dynamics Controls Nitric Oxide Affinity in Nitrophorin 4. J. Biol. Chem. 2004, 279, 39401−39407. (11) Khalil, M.; Demirdö ven, N.; Tokmakoff, A. Vibrational Coherence Transfer Characterized with Fourier-Transform 2D IR Spectroscopy. J. Chem. Phys. 2004, 121, 362−373. (12) Cahoon, J. F.; Sawyer, K. R.; Schlegel, J. P.; Harris, C. B. Determining Transition-State Geometries in Liquids Using 2D-IR. Science 2008, 319, 1820−1823. (13) Anna, J. M.; King, J. T.; Kubarych, K. J. Multiple Structures and Dynamics of [CpRu(CO)2]2 and [CpFe(CO)2]2 in Solution

4. CONCLUSIONS In this work, the vibrational behaviors of two isomeric species (cis and trans) of a nitrosylruthenium coordination compound have been characterized using linear and nonlinear IR, in particular the 2D IR methods. Using the NO stretching mode as a vibrational probe, information about structures and structural dynamics of the two Ru complexes as well as local solvent fluctuations was examined in a weak polar solvent (CDCl3). Linear IR spectra exhibit quite different vibrational frequencies and slight spectral widths for the two isomers. Magic-angle pump−probe spectra show different vibrational lifetimes, and also different vibrational anharmonicities, while 2D IR spectra reveal different spectral diffusion dynamics. Our results suggest that the obtained vibrational parameters, including the spectral line width, lifetime, and spectral diffusion time constant, can be used to differentiate the two isomeric nitrosylruthenium complexes. The results also suggest that the anharmonic potential surface of the NO stretching mode is influenced by its local structural environment, and also by solute−solvent interactions. Our work demonstrates that the combination of linear and nonlinear 2D IR methods can be used to differentiate the cis and trans isomeric species of transitional metal nitrosylruthenium coordination compounds in their equilibrium and even nonequilibrium states. Our work also provides a benchmark measurement for these nitrosylruthenium(II) complexes to be incorporated into 11508

DOI: 10.1021/acs.jpcb.6b08060 J. Phys. Chem. B 2016, 120, 11502−11509

Article

The Journal of Physical Chemistry B Revealed with Two-Dimensional Infrared Spectroscopy. Inorg. Chem. 2011, 50, 9273−9283. (14) King, J. T.; Kubarych, K. J. Site-Specific Coupling of Hydration Water and Protein Flexibility Studied in Solution with Ultrafast 2D-IR Spectroscopy. J. Am. Chem. Soc. 2012, 134, 18705−18712. (15) Yang, F.; Yu, P.; Zhao, J.; Wang, J. Simultaneously Probing Two Ultrafast Condensed-Phase Molecular Symmetry Breaking Events by Two-Dimensional Infrared Spectroscopy. ChemPhysChem 2013, 14, 2497−2504. (16) Woys, A. M.; Mukherjee, S. S.; Skoff, D. R.; Moran, S. D.; Zanni, M. T. A Strongly Absorbing Class of Non-Natural Labels for Probing Protein Electrostatics and Solvation with FTIR and 2D IR Spectroscopies. J. Phys. Chem. B 2013, 117, 5009−5018. (17) Yu, P.; Yang, F.; Zhao, J.; Wang, J. Hydration Dynamics of Cyanoferrate Anions Examined by Ultrafast Infrared Spectroscopy. J. Phys. Chem. B 2014, 118, 3104−3114. (18) Wang, J.; Yang, F.; Zhao, Y.; Yu, P.; Qiao, X.; Wang, J.; Wang, H. Photoisomerization and Structural Dynamics of Two Nitrosylruthenium Complexes: a Joint Study by NMR and Nonlinear IR Spectroscopies. Phys. Chem. Chem. Phys. 2014, 16, 24045−24054. (19) Fedoseeva, M.; Delor, M.; Parker, S. C.; Sazanovich, I. V.; Towrie, M.; Parker, A. W.; Weinstein, J. A. Vibrational Energy Transfer Dynamics in Ruthenium Polypyridine Transition Metal Complexes. Phys. Chem. Chem. Phys. 2015, 17, 1688−1696. (20) Skinner, J. L. Vibrational Line Shapes and Spectral Diffusion in Fluids. Mol. Phys. 2008, 106, 2245−2253. (21) Roberts, S. T.; Loparo, J. J.; Tokmakoff, A. Characterization of Spectral Diffusion from Two-Dimensional Line Shapes. J. Chem. Phys. 2006, 125, 084502−8. (22) Kuroda, D. G.; Hochstrasser, R. M. Dynamic Structures of Aqueous Oxalate and the Effects of Counterions Seen by 2D IR. Phys. Chem. Chem. Phys. 2012, 14, 6219−6224. (23) Yan, C.; Yuan, R.; Pfalzgraff, W. C.; Nishida, J.; Wang, L.; Markland, T. E.; Fayer, M. D. Unraveling the Dynamics and Structure of Functionalized Self-Assembled Monolayers on Gold using 2D IR Spectroscopy and MD Simulations. Proc. Natl. Acad. Sci. U. S. A. 2016, 113, 4929−4934. (24) Li, D.; Yang, F.; Han, C.; Zhao, J.; Wang, J. Correlated HighFrequency Molecular Motions in Neat Liquid Probed with Ultrafast Overtone Two-Dimensional Infrared Spectroscopy. J. Phys. Chem. Lett. 2012, 3, 3665−3670. (25) Kwak, K.; Park, S.; Finkelstein, I. J.; Fayer, M. D. FrequencyFrequency Correlation Functions and Apodization in Two-Dimensional Infrared Vibrational Echo Spectroscopy: A New Approach. J. Chem. Phys. 2007, 127, 124503. (26) Sando, G. M.; Zhong, Q.; Owrutsky, J. C. Vibrational and Rotational Dynamics of Cyanoferrates in Solution. J. Chem. Phys. 2004, 121, 2158−2168. (27) Lynch, M. S.; Cheng, M.; Van Kuiken, B. E.; Khalil, M. Probing the Photoinduced Metal−Nitrosyl Linkage Isomerism of Sodium Nitroprusside in Solution Using Transient Infrared Spectroscopy. J. Am. Chem. Soc. 2011, 133, 5255−5262. (28) Brookes, J. F.; Slenkamp, K. M.; Lynch, M. S.; Khalil, M. Effect of Solvent Polarity on the Vibrational Dephasing Dynamics of the Nitrosyl Stretch in an FeII Complex Revealed by 2D IR Spectroscopy. J. Phys. Chem. A 2013, 117, 6234−6243. (29) Cheng, M.; Brookes, J. F.; Montfort, W. R.; Khalil, M. pHDependent Picosecond Structural Dynamics in the Distal Pocket of Nitrophorin 4 Investigated by 2D IR Spectroscopy. J. Phys. Chem. B 2013, 117, 15804−15811. (30) Ohta, K.; Tayama, J.; Saito, S.; Tominaga, K. Vibrational Frequency Fluctuation of Ions in Aqueous Solutions Studied by Three-Pulse Infrared Photon Echo Method. Acc. Chem. Res. 2012, 45, 1982−1991. (31) Wang, H. F.; Hagihara, T.; Ikezawa, H.; Tomizawa, H.; Miki, E. Electronic Effects of the Substituent Group in 8-Quinolinolato Ligand on Geometrical Isomerism for Nitrosylruthenium(II) Complexes. Inorg. Chim. Acta 2000, 299, 80−90.

(32) Miki, K.; Tomizawa, H.; Ikezawa, H.; Miki, E. Crystal Structures of Three Geometrical Isomers of [Ru(OAc)(2mqn)2NO] (OAc = acetate, H2mqn = 2-methyl-8-quinolinol) and Their Thermal Isomerization Reactions. Inorg. Chim. Acta 1997, 257, 3−10. (33) Han, C.; Zhao, J.; Yang, F.; Wang, J. Structural Dynamics of NPropionyl-d-glucosamine Probed by Infrared Spectroscopies and Ab Initio Computations. J. Phys. Chem. A 2013, 117, 6105−6115. (34) Shim, S.-H.; Zanni, M. T. How to Turn Your Pump-Probe Instrument into a Multidimensional Spectrometer: 2D IR and Vis Spectroscopies via Pulse Shaping. Phys. Chem. Chem. Phys. 2009, 11, 748−761. (35) Middleton, C. T.; Mukherjee, S. S.; Zanni, M. T. ResidueSpecific Structural Kinetics of Proteins Through the Union of Isotope Labeling, Mid-IR Pulse Shaping, and Coherent 2D IR Spectroscopy. Methods 2010, 52, 12−22. (36) Shim, S.-H.; Zanni, M. T. Generation and characterization of phase and amplitude shaped femtosecond mid-IR pulses. Opt. Express 2006, 14, 13120−13130. (37) Mennucci, B.; Tomasi, J. Continuum Solvation Models: A New Approach to the Problem of Solute’s Charge Distribution and Cavity Boundaries. J. Chem. Phys. 1997, 106, 5151−5158. (38) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; et al. Gaussian 09, revision A.02; Gaussian, Inc.: Pittsburgh, PA, 2009. (39) De La Cruz, C.; Sheppard, N. A Structure-Based Analysis of the Vibrational Spectra of Nitrosyl Ligands in Transition-Metal Coordination Complexes and Clusters. Spectrochim. Acta, Part A 2011, 78, 7− 28. (40) Enemark, J. H.; Feltham, R. D. Principles of Structure, Bonding, and Reactivity for Metal Nitrosyl Complexes. Coord. Chem. Rev. 1974, 13, 339−406. (41) Andrews, L.; Citra, A. Infrared Spectra and Density Functional Theory Calculations on Transition Metal Nitrosyls. Vibrational Frequencies of Unsaturated Transition Metal Nitrosyls. Chem. Rev. 2002, 102, 885−912. (42) Bottomley, F. Nitrosyl Complexes of Ruthenium. Coord. Chem. Rev. 1978, 26, 7−32. (43) Lahiri, G. K.; Kaim, W. Electronic Structure Alternatives in Nitrosylruthenium Complexes. Dalton Trans. 2010, 39, 4471−4478. (44) Nitzan, A.; Mukamel, S.; Jortner, J. Energy Gap Law for Vibrational Relaxation of a Molecule in a Dense Medium. J. Chem. Phys. 1975, 63, 200−207. (45) Wang, J.; Pan, H.; Liu, J.; Zhao, Y.; Wu, Y.; Wang, H. Structures and Spectroscopic Properties of Three [Ru(OAc)(2mqn)2NO] (H2mqn = 2-methyl-8-quinolinol) Isomers: An Experimental and Density Functional Theoretical Insight. J. Mol. Struct. 2015, 1086, 232−238.

11509

DOI: 10.1021/acs.jpcb.6b08060 J. Phys. Chem. B 2016, 120, 11502−11509