High-Resolution Coherent Three-Dimensional Spectroscopy of Br2

Feb 20, 2013 - Chemistry Department, Spelman College, Atlanta, Georgia 30314, ... and Chemical Biology, Georgia Institute of Technology, Atlanta, Geor...
0 downloads 0 Views 974KB Size
Download PDF
Article pubs.acs.org/JPCA

High-Resolution Coherent Three-Dimensional Spectroscopy of Br2 Peter C. Chen,*,† Thresa A. Wells,† and Benjamin R. Strangfeld‡ †

Chemistry Department, Spelman College, Atlanta, Georgia 30314, United States Department of Chemistry and Chemical Biology, Georgia Institute of Technology, Atlanta, Georgia 30332, United States

‡

ABSTRACT: In the past, high-resolution spectroscopy has been limited to small, simple molecules that yield relatively uncongested spectra. Larger and more complex molecules have a higher density of peaks and are susceptible to complications (e.g., effects from conical intersections) that can obscure the patterns needed to resolve and assign peaks. Recently, high-resolution coherent two-dimensional (2D) spectroscopy has been used to resolve and sort peaks into easily identifiable patterns for molecules where pattern-recognition has been difficult. For very highly congested spectra, however, the ability to resolve peaks using coherent 2D spectroscopy is limited by the bandwidth of instrumentation. In this article, we introduce and investigate highresolution coherent three-dimensional spectroscopy (HRC3D) as a method for dealing with heavily congested systems. The resulting patterns are unlike those in highresolution coherent 2D spectra. Analysis of HRC3D spectra could provide a means for exploring the spectroscopy of large and complex molecules that have previously been considered too difficult to study.

■

INTRODUCTION Interest in developing coherent two-dimensional (2D) spectroscopy as a new optical analogue to multidimensional nuclear magnetic resonance (NMR) spectroscopy has been driven by the advantages of multidimensional spectroscopy over conventional one-dimensional 1D spectroscopies. Like most 2D techniques, extension to the second dimension helps improve resolution and eliminate congestion problems. Two-dimensional spectroscopies can also provide new information by exposing the relationship between peaks in a spectrum, which leads to information about molecular structure and behavior that is not normally available when using 1D techniques. This relationship is revealed by the presence and properties of peaks that are located in the off-diagonal region of the 2D spectrum. Most of the past work in coherent 2D spectroscopy has been for vibrational or electronic spectroscopy in condensed phase systems.1−6 During the past few years, high-resolution versions of coherent 2D7,8 spectroscopy and incoherent 2D9,10 spectroscopy have been used to improve the ability to generate electronic spectra of gas phase molecules, which are well-known for being rich in information but sometimes difficult or impossible to interpret when the peaks are restricted to one dimension. The advantages of high-resolution coherent 2D spectroscopy over conventional high-resolution electronic spectroscopy include the ability to sort otherwise disorganized peaks by species, quantum number, and selection rules. This sorting ability helps to reveal patterns among peaks that would otherwise appear patternless in conventional 1D spectra. Recently, further expansion of coherent 2D spectroscopy into the third dimension has been explored by several research groups.11−18 The work carried out so far has been in the condensed phase, where the challenge has been to extract additional information (e.g., on anharmonicity12) from a small © 2013 American Chemical Society

number of observed peaks that contain contributions from components (e.g., overtone and combination bands) that appear overlapped and unresolved in 1D spectra. In the gas phase, however, the challenge for high-resolution coherent 2D spectroscopy is different. Spectral congestion in high-resolution gas phase electronic spectra can be caused by a very high number and density of narrow peaks rather than a few broad ones. Congestion in high-resolution coherent 2D spectra makes it difficult to accurately identify the correct peaks for a given pattern and to assign peaks with a high degree of certainty. Similar congestion problems in the 2D NMR spectra of large molecules have driven the development of 3D NMR spectroscopy. The purpose of this article is to demonstrate how highresolution coherent three-dimensional (HRC3D) spectroscopy can be used to deal with severe congestion problems in highresolution gas phase electronic spectroscopy. These congestion problems are often due to the high number of overlapping peaks from many different rotational and vibrational transitions that have similar transition frequencies. This congestion worsens for larger molecules and may be exacerbated by perturbations and other effects caused by conical intersections. Perhaps the most successful method for dealing with severe congestion has been the use of the molecular jet to reduce the temperature and thereby the number of rotational peaks. Highresolution coherent 2D spectroscopy provides an alternative method for dealing with congestion that preserves rotational Special Issue: Prof. John C. Wright Festschrift Received: November 30, 2012 Revised: February 11, 2013 Published: February 20, 2013 5981

dx.doi.org/10.1021/jp3118049 | J. Phys. Chem. A 2013, 117, 5981−5986

The Journal of Physical Chemistry A

Article

controlled using the signal beam wavelength (ωpump = ωsignal + ωidler). The firing of the three pump lasers was synchronized using a delay generator (Stanford Research System DG 535). The three beams were combined using collinear phasematching before being focused into the sample cell. The pump beams were then removed by KG3 and BG40 absorption filters and a Semrock 633 nm notch filter. The remaining FWM beam was focused into a 1.25 m monochromator (SPEX 1250m) equipped with a 2400 g/mm grating and CCD (2048 elements, 13 μm pixels). In order to produce a single 2D slice of a coherent 3D spectrum, we set the dye laser to a specific wavelength, scanned the narrowband near-infrared OPO (ω3) wavelength, and collected a series of spectra using the monochromator and CCD (ω4). During the scan, the dye laser wavelength was continuously monitored using a Bristol Instruments 821 wavemeter. The resulting 1D spectra were stacked in order to produce the 2D slice. The wavelength of the light from the dye laser (ω1) could then be changed in order to produce different 2D slices within the third dimension. Multichannel detection (needed to reduce the amount of time required to collect data) along the ω4 axis was facilitated by the use of a degenerate OPO for ω2, which was not tuned but generated broadband light covering a continuous range of wavelengths from 1150 to 1700 nm. The three axes of the 3D spectra therefore correspond to ω1, ω3, and ω4, with intensity as an additional (unplotted) dimension. For plotting purposes, the results are most clearly shown as a series of stackable 2D plots where the x and y axes correspond to ω4 and ω3 and peaks appear as dots. These peak positions were obtained using a modified version of a computer program (nmrDraw) that includes a two-dimensional peak picker. Each 2D slice consists of approximately 5000 stacked 1D spectra, each taken with a specific ω1 wavelength. Small step sizes (0.002 nm) that were a factor of 3 times smaller than the line width of the ω3 beam were used in order to help distinguish FWM peaks from spurious peaks caused by cosmic rays that periodically hit the CCD detector. The real FWM peaks persist for several steps (appearing as a group of peaks along the y axis) and are reproducible, while the cosmic ray spikes appear as single sporadic events that are irreproducible. The acquisition time required to produce each 2D slice was approximately 2 h. The spectra were repeated, and spurious cosmic ray spikes were removed using a Fortran computer program that eliminated many (but not all) of the irreproducible peaks. The sample consisted of 0.001 atm of bromine in an evacuated 0.5 m long glass cell. Bromine was selected because of its simplicity as a diatomic molecule and the availability of published spectroscopic constants for its B and X states.23−25 The beam pulse energies were 3 mJ for the broadband OPO, 3 mJ for the narrowband tunable OPO, and 0.3 mJ for the dye laser. In order to ensure that the observed signal depended upon all three beams, we periodically blocked each beam and confirmed that the signal disappeared. The wavelengths of the dye and OPO beams were also changed to confirm that the signal was triply resonant.

information; the resolution is improved by expansion to a second dimension, and information about the relationship between peaks can be found from patterns in the off-diagonal region. In this region, peaks are sorted and grouped into recognizable structures (e.g., parabolas and x-shaped clusters) so that all peaks within a cluster involve the same vibrational quantum numbers and that the peaks are ordered by their rotational quantum numbers. However, recent high-resolution coherent 2D spectra of small molecules (NO2 and Br2) reveal many regions where the congestion is too heavy to permit accurate peak resolution and assignment).19,20 This problem needs to be addressed to permit the spectral analysis of larger and more complex molecules.

■

EXPERIMENTAL SECTION Nonlinear spectroscopy serves as a well-suited method for generating coherent multidimensional spectra because the resulting signal depends upon multiple beams that can have independently controllable frequencies, each of which can constitute an orthogonal axis in a multidimensional spectrum. Four wave mixing (FWM) techniques can therefore be used to generate spectra with up to 4 orthogonal frequency dimensions. By using three independently controllable beams to drive the FWM signal, fully (triply) resonant FWM signals can be generated that provide greater selectivity and resonant enhancement over doubly or singly resonant processes that might be used to create 1D or 2D spectra. The fully resonant FWM approach used in this work was pioneered by Wright and co-workers21,22 and makes effective use of the additional dimensions that are available when creating coherent 3D spectra. The use of a broadband beam for one of the fields can facilitate multichannel detection with a charge-coupled device (CCD), which helps reduce the time required to acquire data. The FWM signals produced in this work were driven by three independent input beams (ω1 = narrowband tunable visible dye laser beam, ω2 = broadband near-infrared OPO beam, and ω3 = narrowband near-infrared tunable OPO beam). The use of three independent beams to drive fully (triply) resonant FWM processes allows us to explore and maximize the capabilities of coherent 3D spectroscopy. A simplified diagram of the setup is shown in Figure 1. Three independent injection-seeded Nd:YAG lasers (SpectraPhysics

Figure 1. Simplified experimental layout of the coherent 3D spectrometer.

■

Lab 150, GCR 230, and PRO 250) were used to pump a dye laser (Coherent Scanmate plus, line width = 0.075 cm−1), a broadband optical parametric oscillator (OPO) (custom-built, line width > 3000 cm−1)), and a narrowband tunable OPO (SpectraPhysics MOPO 730, line width = 0.2 cm−1, pumped by 355 nm). The idler beam from the MOPO 730 was used in this experiment, but the wavelength of the idler beam was

RESULTS AND DISCUSSION Figure 2 shows a 2D slice from the coherent 3D spectrum of bromine gas. Compared to previous coherent 2D spectra, the 3D result has a much lower peak density; for bromine, the HRC3D peak density is typically around one peak per square 5982

dx.doi.org/10.1021/jp3118049 | J. Phys. Chem. A 2013, 117, 5981−5986

The Journal of Physical Chemistry A

Article

Figure 3. FWM energy level diagram (left) and predicted 2D slice from the resulting 3D spectrum (right). Vertical and horizontal vibrational/vibronic resonance lines intersect to give clusters that are represented by the circles and that each contain three rotationally resolved peaks. The a and b levels are from the X state, while the c and d levels are from the B state.

Figure 2. Coherent 3D spectrum of bromine when the dye laser was set to 612.334 nm. The dots mark the positions for 3776 of the most intense repeated peaks that were identified using a 2D peak picking program. The peaks forming the diagonal line were caused by unwanted stray light from the MOPO signal beam during the scan. The peaks forming the vertical line were caused by unwanted stray light from nearby fluorescent lights (Hg emission at λ = 546.075 nm). The circled clusters have been assigned; those enclosed in a box are presumably due to a different energy set of energy levels or a different FWM process yet to be determined.

to the bottom vibrational level of the ground electronic state (level a). The spacings between the vertical lines are therefore equal to the spacing between corresponding vibrational levels in the excited electronic state. The diagonal resonance lines are due to Raman-type resonances where ω4−ω3 matches the frequencies of overtone vibrations in the ground electronic state (the b levels). The spacings between the diagonal lines therefore are equal to the spacings between the corresponding ground state vibrational overtones. The spacings between the b levels in Figure 3 are drawn so that they are much more even than the spacings for level d. In other words, the d level spacings are drawn to show greater anharmonicity than that for the b levels. As a result, the x axis spacings in the simulated multidimensional spectrum are less even (decrease with increasing ω4 value), while the y axis spacings and the spacings between diagonal lines are more even. This behavior is similar to that observed experimentally in Figure 2. This match between the behavior described in Figure 3 and that observed in Figure 2 suggests that the FWM process in Figure 3 is a possible candidate for producing the peaks shown in Figure 2. To investigate this possibility, we used literature spectroscopic constants to calculate and then identify suitable energy levels that fit the observed resonance wavelengths, spacings, and patterns shown in Figure 2. For the ground electronic X state, we used spectroscopic constants from Gerstenkorn and Luc23 for the low values of v″ and constants from Focsa and Bernath25 for the high values of v″. For the excited electronic B state, we used constants from Gerstenkorn and Luc.24 The value of the rotational quantum number J throughout the entire FWM process was restricted by the fact that the dye laser wavelength was fixed at 612.334 nm. Since this particular FWM process is parametric (starting level = final level), application of the ΔJ = ±1 selection rule dictated that J could never deviate by more than 2 quanta from its value at the starting level (see Figure 4). The limiting factors that were used to help narrow down the possible energy levels are summarized as follows: (1) J″ and J′ for levels a and c are limited to a few possible values by the wavelength of the dye laser (ω1). (2) J′ for level d should be within +1 or −1 quanta of J″ for level a. The calculated value should agree with the ω4 (x axis) wavelengths of the observed peaks for the same isotopomer. (3) The relationship between J″ in level b and J″ in level a should obey the ΔJ = 0,+2 Ramantype selection rule for the same isotopomer.

nanometer, while the coherent 2D peak density is hundreds of peaks per square nanometer.20 Changes in the dye laser wavelength by a few tens of picometers cause dramatic changes in the resulting spectrum; the observed peaks disappear and are replaced by new peaks that exhibit different patterns. The peaks in Figure 2 that are enclosed in circles show a clear pattern with a high degree of symmetry. These peaks are grouped into three-peak clusters, two of which share the same ω4 (x axis) wavelength. By contrast, none of the three peaks within a cluster have the same y axis values. For the x axis region from 534 to 546 nm, the width of these clusters (dimension along the x axis) grows from 0.2 to 0.3 nm. Along the y axis, from 530 to 550 nm, the height of these clusters (distance between highest and lowest points) remains relatively constant at about 0.5 nm. The clusters also exhibit a similar pattern; the x axis resonance wavelengths of the circled clusters appear to repeat, but their y axis values do not. In other words, the set of possible x axis values of the peaks appears to be limited to a small number of repeating values, while the y axis resonances are unique for every peak. This overall pattern of circled clusters appears to be produced by intersections between two types of resonance lines: a set of evenly spaced diagonal resonance lines intersecting with a set of vertical resonance lines that gradually increase in spacing with increasing wavelength. Prior work in the field of high-resolution coherent 2D spectroscopy has revealed the fact that different FWM processes can produce different cluster patterns that can be modeled by intersecting resonance lines. For example, the CDRES process shows a rectangular cluster pattern that is produced by intersections between horizontal and vertical resonance lines, while the RECARS process can exhibit a parallelogram-shaped cluster pattern caused by intersections between diagonal resonance lines and vertical resonance lines.26 Figure 3 illustrates how the final two photons in a proposed FWM process can yield clusters based upon intersection between vertical and diagonal resonance lines. The four vertical resonance lines are produced when ω4 connects vibronic levels from the excited electronic state (represented by the d levels) 5983

dx.doi.org/10.1021/jp3118049 | J. Phys. Chem. A 2013, 117, 5981−5986

The Journal of Physical Chemistry A

Article

wavelength causes the coherence to follow the lower half of the diamond-shaped diagram. The next resonance induced by the broadband OPO involves levels c and b, where the allowed quantum numbers for b are J″ = x and J″ = x − 2. Level d can either be J′ = x + 1 or x − 1. The diamond diagram shows three possible pathways from level b to level a, all of which have different ω3 frequencies and two of which have similar ω4 frequencies. The right side of Figure 4 shows the predicted shape of the resulting three point cluster (shown in wavelength space), which is in agreement with the clusters shown in Figure 2. This analysis indicates that the width of the cluster is equal to the difference in energies between P and R processes associated with ω4. Both of these processes involve the same rotational quantum number J″ in the ground state; therefore, the width depends upon both the rotational quantum number J″ and the rotational constants in the excited electronic state. The height of the cluster is equal to the difference in energies between the P and R processes associated with ω3. However, both of these processes involve the same rotational quantum number in the excited electronic state J′, which is equal to J″ − 1. Therefore, the height depends upon the rotational quantum number (J″ − 1) and the rotational constants for high overtones in the ground electronic states. If the rotational constants for the ground and excited states are known, the size of the cluster could therefore be used to estimate the rotational quantum number. In order to further verify the accuracy of the assigned levels and FWM process, we calculated the dye laser wavelength needed for the corresponding R-type process (ω1 resonance: v″ = 1, J″ = 62 to v′ = 6, J′ = 63) to be 611.821 nm for the mixed isotopomer. This process should yield a pattern that can be predicted by the R-type diamond in the lower half of Figure 4. The resulting prediction was that two of the three peaks (shown in green) should be identical for both P- and R-type processes, but the lowest peak (red) in the P-type process should be absent and replaced by a different peak (yellow) in the R-type process. The resulting R-type cluster should be easily recognizable because it has a shape that is the reverse of the P-type cluster, with two of the three points remaining unchanged in position. Figure 5 shows the experimental results

Figure 4. Pathways showing the evolution of the rotational quantum number after each interaction between the molecule and one of the fields (left). The top is for a P-type process, and the bottom is for an R-type process. The right side shows the corresponding predicted cluster shape if the spectrum is plotted in wavelength space.

The following procedure was used to assign the observed peaks. First, the dye laser wavelength was compared with calculated transition wavelengths. The dye laser was fitted first because its wavelength was constant, it had the narrowest line width, and its wavelength was monitored using a wavemeter. Attempts to find levels that would satisfy all of these conditions failed for all isotopomers when using a starting state with v″ = 0. Changing the starting state to v″ = 1 for the mixed isotopomer (Br279−81) provided a satisfactory fit when J″ = 62 (the difference was less than 0.003 nm, which was within the experimental uncertainty of 0.004 nm based upon the instrumental error and laser line width). We then used those ground state quantum numbers (v″ = 1 and J″ = 62) to calculate ω4 wavelengths and compared the results with those experimentally determined by manually adjusting the MOPO wavelength to find the maxima of several peaks. The discrepancies between eight experimental and calculated values were all less than 0.010 nm, which is within the estimated experimental uncertainty (based upon the line width of the MOPO, the line width of the monochromator, and the spacing between the CCD pixels). Finally, seven experimental values for ω4−ω3 were compared with calculated values. The discrepancies between the calculated and measured MOPO signal beam were all less than 0.012 nm (also within the estimated experimental uncertainty). Therefore, the fits for all three pairs of experimental and calculated values were within the experimental uncertainty, and the resulting assignments maintained consistency for the isotopomer and ground state quantum numbers. The shape of the clusters can be understood by examining how selection rules determine the allowed levels for this FWM process. Figure 4 contains a diagram showing all possible pathways that satisfy rotational selection rules (ΔJ = ±1) and requirements for the FWM process shown in Figure 3. The diamond-like shape of this diagram (initially expanding from left to right, followed by contraction) is due to the fact that the starting and final rotational quantum number must be identical for a parametric process. If J″ = x initially, then the fixed dye wavelength can be resonant with either a J′ = x + 1 or a J′ = x − 1 level in the B state. The dye wavelength of 612.334 nm is very close to the calculated value of 612.3375 nm for a P-type resonance from v″ = 1, J″ = 62 to v′ = 6, J′ = 61. This dye laser

Figure 5. Coherent 3D spectrum of bromine when the dye laser is set to 611.819 nm, showing 680 of the most intense repeated peaks produced by an initial R-type process. The diagonal line was caused by unwanted stray light from the MOPO signal beam during the scan. The vertical line was caused by unwanted stray light from nearby fluorescent lights. The clusters that were predicted are circled; other apparent clusters have been enclosed in boxes. 5984

dx.doi.org/10.1021/jp3118049 | J. Phys. Chem. A 2013, 117, 5981−5986

The Journal of Physical Chemistry A

Article

resolution coherent 2D spectra. The resulting patterns are unlike those observed for high-resolution coherent 2D spectroscopy. For bromine, expansion to the third dimension results in peaks that are more widely separated and better resolved than they were in the coherent 2D spectra by approximately 2 orders of magnitude. Pattern recognition methods have been used to help determine the FWM process responsible for generating the peaks, to accurately assign peaks, and to provide immediate information (e.g., distinguish between P-type and R-type processes). The FWM process found to dominate the HRC3D spectra for the experiments described in this work resembles a CARS-type process. However, other FWM processes should be possible. The ability to identify the FWM process responsible for generating observed peaks is necessary in order to determine spectroscopic constants for molecules where little information is available, and methods for identifying and distinguishing different FWM processes are currently being explored. If such methods can be developed, HRC3D spectroscopy, perhaps combined with its2D counterpart, could prove to be an effective means for carrying out rotationally resolved spectroscopic analyses of molecules that have been considered too large and complex for high-resolution spectroscopy.

after setting the dye laser wavelength to 611.819 nm. The results match that predicted; the circled clusters have the pattern predicted for an R-type process, and the locations of the peaks are also as predicted. In addition to showing clusters that were predicted, Figures 2 and 5 also shows several unanticipated clusters (enclosed in rectangles). These smaller unanticipated clusters are presumably due to molecules with a lower initial lower J″ value, while the larger unanticipated clusters are probably caused by molecules that originally have a higher initial J″ value. Both P-type and R-type clusters can be seen in Figure 5. In general, additional unanticipated clusters could be caused by (1) the same FWM process involving different energy levels or (2) a different FWM process yet to be determined. Additional work is therefore needed in order to establish a better understanding of the full set of FWM processes that can contribute to HRC3D spectra. The results described in this article suggest that highresolution coherent 3D spectroscopy may provide significant advantages when trying to accurately assign peaks. For 1D and 2D spectroscopy, the accuracy of peak assignments may be controversial when the density of peaks is very high and the spectra are highly congested. For HRC3D, the peak resolution is subtantially improved because the peak density is reduced by several (approximately 2 for this study) orders of magnitude. Second, the reliability of peak assignments is likely to be improved by the required match with three independent experimental wavelengths (ω1, ω3, and ω4) rather than just the one or two required for 1D or 2D spectroscopy. In fact, the increased spacing between peaks and the requirement for 3 wavelength matches can make it challenging to initially locate peaks. For many spectroscopic techniques, such as fluorescence, Raman, and even high-resolution coherent 2D spectroscopy, peaks can readily be found after varying the wavelength of just one laser and/or the detection system. For the high-resolution coherent 3D experiments, however, both narrowband beams need to be positioned correctly in order to observe peaks; the spectra produced using the detection system may remain void of peaks if either of the two narrowband tunable sources is off by just a few picometers. After using three independent wavelengths to assign the peaks, further confirmation of the assignment and the FWM process can be made by simulating and then running an experiment where the selection rule is switched (from P-type to R-type or vice versa). A third advantage of coherent 3D spectroscopy is that it provides new and immediate information, based upon the observed patterns in the spectra. The cluster shape can be used to quickly differentiate P-type processes from R-type processes. The height of each cluster depends upon the rotational quantum numbers and the rotational constants for the ground electronic state. The width of each cluster depends upon the rotational quantum numbers and the rotational constants for the excited state. Finally, the distances between vertical columns depend upon the spacings between vibrational levels in the excited electronic state (i.e., the d levels), while the distances between diagonal lines of clusters depend upon the spacings between vibrational levels in ground electronic state (i.e., the b levels).

■

AUTHOR INFORMATION

Corresponding Author

*Phone: 404-270-5742. E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported by National Science Foundation grant CHE-0910232. Additional support was provided by NSF grant EEC-0310717. We wish to thank Zuri House, Notorious Scott, Kamilah Mitchell, and Christa Fields for their assistance in software development and running coherent 3D experiments.

■

REFERENCES

(1) Wright, J. C. Multiresonant Coherent Multidimensional Spectroscopy. Annu. Rev. Phys. Chem. 2011, 62, 209−230. (2) Jonas, D. Two-Dimensional Femtosecond Spectroscopy. Annu. Rev. Phys. Chem. 2003, 54, 425−463. (3) Cho, M. Coherent Two-Dimensional Optical Spectroscopy. Chem. Rev. 2008, 108, 1331−1418. (4) Mukamel, S. Multidimensional Femtosecond Correlation Spectroscopies of Electronic and Vibrational Excitations. Annu. Rev. Phys. Chem. 2000, 51, 691−729. (5) Fournier, F.; Gardner, E. M.; Kedra, D. A.; Donaldson, P. M.; Guo, R.; Butcher, S. A.; Gould, I. R.; Willison, K. R.; Klug, D. R. Protein Identification and Quantification by Two-Dimensional Infrared Spectroscopy: Implications for an All-Optical Proteomic Platform. Proc. Natl. Acad. Sci. 2008, 105, 15352−15357. (6) Moran, S. D.; Woys, A. M.; Buchanan, L. E.; Bixby, E.; Decatur, S. M.; Zanni, M. T. Two-Dimensional IR Spectroscopy and Segmental 13C Labeling Reveals the Domain Structure of Human γD-Crystallin Amyloid Fibrils. Proc. Natl. Acad. Sci. 2012, 109, 3329−3334. (7) Chen, P. C. High Resolution Coherent 2D Spectroscopy. J. Phys. Chem. A 2010, 114, 11365−11375. (8) Chen, P. C.; Joyner, C. C. Peak Separation and Sorting by Coherent 2D Resonance Raman Spectroscopy. Anal. Chem. 2005, 77, 5467−5473. (9) Gascooke, J. R.; Alexander, U. N.; Lawrance, W. D. Two Dimensional Laser Induced Fluorescence Spectroscopy: A Powerful Technique for Elucidating Rovibronic Structure in Electronic

■

CONCLUSIONS We have successfully carried out an initial demonstration of the use of HRC3D spectroscopy as a way of reducing spectral congestion in systems that are too congested for high5985

dx.doi.org/10.1021/jp3118049 | J. Phys. Chem. A 2013, 117, 5981−5986

The Journal of Physical Chemistry A

Article

Transitions of Polyatomic Molecules. J. Chem. Phys. 2011, 134, 184301. (10) Reilly, N. J.; Schmidt, T. W.; Kable, S. H. Two-Dimensional Fluorescence (Excitation/Emission) Spectroscopy As a Probe of Complex Chemical Environments. J. Phys. Chem. A 2006, 110, 12355− 12359. (11) Hayes, D.; Engel, G. S. Extracting the Excitonic Hamiltonian of the Fenna-Matthews-Olson Complex Using Three-Dimensional Third-Order Electronic Spectroscopy. Biophys. J. 2011, 100, 2043− 2052. (12) Ding, F.; Zanni, M. T. Heterodyned 3D IR Spectroscopy. Chem. Phys. 2007, 341, 95−105. (13) Turner, D. B.; Stone, K. W.; Gundogdu, K.; Nelson, K. A. Three-Dimensional Electronic Spectroscopy of Excitons in GaAs Quantum Wells. J. Chem. Phys. 2009, 131, 144510. (14) Cho, M. J. Nonlinear Response Functions for the ThreeDimensional Spectroscopies. Chem. Phys. 2001, 115, 4424−4437. (15) Fidler, A. F.; Harel, E.; Engel, G. S. Dissecting Hidden Couplings Using Fifth-Order Three-Dimensional Electronic Spectroscopy. J. Phys. Chem. Lett. 2010, 1, 2876−2880. (16) Garrett-Roe, S.; Hamm, P. Purely Absorptive Three-Dimensional Infrared Spectroscopy. J. Chem. Phys. 2009, 130, 164510. (17) Garrett-Roe, S.; Hamm, P. What Can We Learn from ThreeDimensional Infrared Spectroscopy? Acc. Chem. Res. 2009, 42, 1412− 1422. (18) Davis, J. A.; Hall, C. R.; Dao, L. V.; Nugent, K. A.; Quiney, H. M.; Tan, H. H.; Jagadish, C. Three-Dimensional Electronic Spectroscopy of Excitons in Asymmetric Double Quantum Wells. J. Chem. Phys. 2011, 135, 044510. (19) Chen, P. C.; Mitchell, K. Analysis of Polyatomic Molecules Using High Resolution Coherent Two-Dimensional Spectroscopy: Application to Nitrogen Dioxide. J. Chem. Phys. 2008, 129, 194301. (20) Chen, P. C.; Gomes, M. Two-Dimensional Coherent Double Resonance Electronic Spectroscopy. J. Phys. Chem. A 2008, 112, 2999−3001. (21) Nguyen, D. C.; Wright, J. C. Synchronous Scanning Methods for Organic Spectroscopy with Fully Resonant Four-Wave Mixing. Appl. Spectrosc. 1985, 39, 230−243. (22) Wright, J. C.; Carlson, R. J.; Hurst, G. B.; Steehler, J. K.; Riebe, M.T.; Price, B. B.; Nguyen, D. C.; Lee, S. H. Molecular, Multiresonant Coherent Four-Wave Mixing Spectroscopy. Int. Rev. Phys. Chem. 1991, 10, 349−390. (23) Gerstenkorn, S.; Luc, P.; Raynal, A.; Sinzelle, J. Description of the Absorption Spectrum of Bromine Recorded by Means of Fourier Transform Spectroscopy: the (B3Π0+u ← X1Σg+) system. J. Phys. 1987, 48, 1685−1696. (24) Gerstenkorn, S.; Luc, P. Analysis of the Long-Range Potential of 79 Br2 in the B3Π0+u State and Molecular Constants of the Three Isotopic Bromine Species 79Br2, 79,81Br2, and 81Br2. J. Phys. 1989, 50, 1417−1432. (25) Focsa, C.; Li, H.; Bernath, P. F. Characterization of the Ground State of Br2 by Laser-Induced Fluorescence Fourier Transform Spectroscopy of the B3Π0+u−X1Σg+ System. J. Mol. Spectrosc. 2000, 200, 104−119. (26) Chen, P. C.; Joyner, C. C. Vibrational and Vibronic Processes in Coherent 2D Resonance Raman Spectroscopy. J. Phys. Chem. A 2006, 110, 7989−7993.

5986

dx.doi.org/10.1021/jp3118049 | J. Phys. Chem. A 2013, 117, 5981−5986