Open Access
Add to BibSonomy
Abstract
Two-dimensional electronic spectroscopy (2DES) is a powerful method to study coherent and incoherent interactions and dynamics in complex quantum systems by correlating excitation and detection energies in a nonlinear spectroscopy experiment. Such dynamics can be probed with a time resolution limited only by the duration of the employed laser pulses and in a spectral range defined by the pulse spectrum. In the blue spectral range (<500 nm), the generation of sufficiently broadband ultrashort pulses with pulse durations of 10 fs or less has been challenging so far. Here, we present a 2DES setup based on a hollow-core fiber supercontinuum covering the full visible range (400-700 nm). Pulse compression via custom-made chirped mirrors yields a time resolution of <10 fs. The broad spectral coverage, in particular the extension of the pulse spectra into the blue spectral range, unlocks new possibilities for coherent investigations of blue-light absorbing and multichromophoric compounds, as demonstrated by a 2DES measurement of chlorophyll a.
© 2023 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Two-dimensional electronic spectroscopy (2DES) [1–5] is becoming a widely applied method to study optical excitations in a variety of materials, including atoms [6, 7], molecular [8–12] and biological systems [13,14], aggregated thin films [15,16], and organic [16–19] and inorganic semiconductors [20–24]. Applications to quantum systems in the strong coupling regime are just emerging [25–28]. In contrast to conventional pump-probe (PP) spectroscopy, a phase-locked pulse pair is used in 2DES to obtain information about the excitation pathways while maintaining a high time resolution defined by the pump pulse duration. Importantly, interferometric phase-stability of the pump pulse pair allows for obtaining frequency resolution of the excitation energy and thus to record 2DES maps that correlate the excitation and detection energy as a function of the pump-probe delay (waiting time). For this reason, 2DES is exceptionally powerful for investigating coherent and incoherent coupling processes, charge and energy transfer [3,29,30], many-body interactions [21,31] and vibronic couplings [16,32,33]. For many of the applications, 2DES experiments with time resolution of less than 10 fs, shorter than any vibrational period in the system, are desirable.
2DES experiments with sufficiently broadband pulses, covering, ideally, all optical transitions of interest of the system under investigation, are of particular current interest. They open the door to probing interactions in widely detuned quantum systems. Even though 2DES experiments with narrowband excitation can have specific advantages in selecting quantum pathways [34], broadband 2DES offers, in principle, unique access to coupling and relaxation phenomena in multichromophoric systems. For this, the chosen optical pulses have to be intense enough to induce a sufficiently strong third order nonlinearity in the sample and stable enough to provide 2DES spectra with high signal-to-noise ratio in a short acquisition time. The generation of laser pulses that meet all these requirements can be challenging. This holds in particular in the “blue” part of the spectral region (400-500 nm) where dispersion is pronounced and chirp compensation is challenging. Hence, 2DES experiments with a spectral bandwidth covering the full visible range from 400-700 nm have so far not been reported.
Such experiments are highly desirable for the understanding of many biologically relevant systems [35], since e.g. chlorophyll [10] and carotenoid [10,36] pigments in light-harvesting complexes [8,9] or flavoproteins such as cryptochromes [37] have optical transitions below 500 nm. The first challenge in realizing them is to find suitable light sources.
Unamplified bulk supercontinua can, in principle, provide sufficient bandwidth, their typical pulse energies, however, are too low [38]. Noncollinear optical parametric amplifiers (NOPAs) are routinely used to generate ultrashort pulses in the visible and NIR by converting infrared laser pulses, generated, e.g., by Ti-sapphire or fiber amplifier systems, into tunable and broadband pulses [39–43]. They employ nonlinear crystals to amplify weak bulk supercontinua, usually generated in sapphire or YAG crystals. Noncollinear phase-matching in NOPAs yields short pulses of <10 fs duration for wavelengths >500 nm [42,43] but for shorter wavelengths material dispersion limits the spectral phase matching bandwidth to a few 10 nm [44]. Advanced NOPA-based schemes [45,46] that employ additional sum-frequency generation have also been demonstrated and can achieve pulses width durations <10 fs [45].
An alternative approach for creating broad supercontinua relies on sets of thin glass plates [47–49]. Since the broad supercontinuum is generated incrementally this allows for higher pulse energies. To the best of our knowledge, reported supercontinua do not yet cover the full spectral range of interest and only reach to ∼450 nm at best.
Another approach for generating broad pulses for 2DES is to utilize supercontinuum generation via filamentation in gas [50–52] or by using hollow-core fibers [53–55]. This generates stable broadband pulses with spectra that can cover the full visible range [51,53,56,57]. So far, 2DES with such sources has, however, been limited to wavelengths longer than 450 nm [52,53]. Also, 2DES with <10 fs time resolution based on high-repetition rate (10 - 100 kHz) supercontinua has not been reported. Also, more advanced generation schemes such as light-field synthesis [57] have not yet been applied to 2DES.
Hence, we use here supercontinuum generation in a gas-filled hollow-core fiber [53,58–60] for 2DES. The compression of such pulses to achieve pulse durations <10 fs is a challenging task. Advanced dispersion control in the spectral domain via pulse shaping [38,60,61] or deformable mirrors [62,63] have limited performance in the blue part of the spectrum. Also, the blue spectral components may lead to photo damage in liquid-crystal-based spatial light modulators.
Therefore, we chose here to use custom-designed double-chirped mirrors (DCMs) [64], optimized for chirp compensation below 450 nm, for dispersion control of hollow-core fiber supercontinua. We use this source to implement a 2DES setup in a partially collinear geometry that covers the spectral range from 400 nm to 700 nm. To generate the pump-pulse pair we employ an in-line interferometer based on birefringent wedges (“TWINS”) [65]. Transient-grating frequency-resolved optical gating (TG-FROG) measurements characterize the compressed pulses and confirm a pulse duration of <10 fs showing dispersion correction across the entire bandwidth. The high time resolution is further demonstrated by probing vibrational coherences in dichloromethane solvent and rhodamine 6 G dye dissolved in water. The ability to study 2DES maps of multichromophoric systems covering the entire visible range is demonstrated by a 2DES experiment on the photosynthetic pigment chlorophyll a in ethanol. We anticipate that the demonstrated setup will be of interest for uncovering energy and charge transfer processes in a variety of physically, chemically and biologically relevant systems, including, e.g., blue-light absorbing flavoproteins [37] or multichromophoric light harvesting complexes [8].
2. Experimental setup
2.1 Supercontinuum generation
To generate ultrashort supercontinuum pulses we employ a hollow-core fiber system (Savanna, Ultrafast Innovations) that is pumped by a regenerative Ti-sapphire laser (Legend Elite, Coherent) operating at 10 kHz repetition rate. The laser delivers 28 fs pulses with 1 mJ pulse energy, centered at 800 nm with pulse-to-pulse fluctuations of <0.5%. Spectral broadening is achieved by focusing these pulses into a 1-m-long quartz capillary with an inner diameter of 200 µm that is placed in Neon atmosphere at an absolute pressure of ∼2.4 bar (see Fig. 1(a)). To avoid drifts of the laser focus position we employ a beam pointing stabilization unit (Aligna, TEM Messtechnik). A typical supercontinuum spectrum, recorded at the output of the hollow core fiber system and optimized for high spectral density below 500 nm, is presented in Fig. 1(b). This spectrum spans more than one octave from 350–1000 nm. Output pulse energies are usually ∼0.5 mJ. The high quality of the output beam profile (Fig. 1(b), inset) is an advantage of the fiber-based approach that usually yields beams with small spatial chirp [53]. A spectral analysis of the beam profile (Fig. S5) shows that no significant spatial chirp is present in our supercontinuum, except for a weak radial color dependence, with blue components being stronger at the center of the beam. To remove the infrared (IR) contribution, which also contains residual 800-nm pump light, we use a dichroic filter (700 nm Low GDD Dichroic Shortpass Ultrafast Filter, Edmund Optics) that only transmits a spectral window spanning 400-700 nm. Spectra with similar bandwidth can also be generated in Argon gas instead of Neon with pressures up to 0.3 bar. While several other groups report successful generation of stable supercontinua in Argon-filled hollow-core fibers [53–55], we observed that for our system spectra generated with Argon exhibit larger spectral modulations and, importantly, inferior pulse-to-pulse stability compared to the pulses generated in Neon. Therefore, supercontinuum generation in Neon was used in the following.
Fig. 1. Experimental setup and laser spectra. a: Schematic of the experimental setup to generate broadband supercontinua and for two-dimensional electronic spectroscopy (2DES). b: Typical hollow-core fiber output when operating with Neon (2.4 bar absolute pressure), spanning from 350-1000 nm. A picture of the beam profile (inset) shows a clean Gaussian mode. c: Laser spectrum behind the first set of chirped mirrors (DCM 1) with the fundamental spectrum und IR part removed. The spectra in b and c are measured with an intensity-calibrated fiber spectrometer. d: Laser power stability measured over one hour. The power fluctuations, displayed as a histogram in the inset, show a standard deviation of 0.8%.
2.2 Chirp compensation
While the spectrum in Fig. 1(b) indeed provides sufficient spectral bandwidth, the temporal profile of these pulses is highly chirped and this chirp is further increased by dispersion introduced in the setup. Therefore, an appropriate dispersion compensation scheme is essential, although difficult for such broadband pulses, especially for the blue part of the spectra. Even though similar supercontinuum spectra have already been demonstrated previously [51,53], the spectral bandwidth used in a subsequent experiment has usually been reduced and most often limited to wavelengths longer than ∼450 nm. A major reason for not incorporating the blue part of the supercontinuum in the spectroscopy experiment is the lack of suitable pulse compression solutions.
To compensate the supercontinuum chirp and additional setup dispersion, we first employed a set of commercially available double-chirped mirrors (DCM12, Laser Quantum) specified for the spectral range from 400 to 700 nm. As confirmed by Transient-Grating Frequency-Resolved Optical Gating (TG-FROG) and pump-probe (PP) measurements (Figs. S1-2 of Supplement 1), these DCMs only compensate a spectral range from ∼440-700 nm. While this part of the spectrum can be nicely compressed down to ∼6 fs, the blue spectral region <440 nm remains uncompressed. This is clearly seen as a tilted tail in the TG-FROG and PP map (Figs. S1-2). These pulses are thus not suited for studying compounds containing resonances in the blue spectral region with high time resolution.
We therefore went on to use custom-designed matched pairs of DCMs (batch C222K040 and C222K041, Layertec). These DCMs were optimized for the dispersion of 1.5 mm fused silica per bounce pair for a broad spectral range of 380-700 nm (see Figs. S3-4). To accurately characterize the pulses a TG-FROG measurement of pulses compressed by 10 DCM bounce pairs is displayed in Fig. 2(a). The straight shape of the TG-FROG trace as a function of the wavelength shows that all spectral components arrive simultaneously and thus the full spectral range is chirp-corrected, in contrast to the case of the DCM12 mirrors (Fig. S1). The retrieved pulse intensity profile and phase (Fig. 2(b)) show a full-width at half maximum (FWHM) of 5.8 fs. This is less than a factor of two beyond the bandwidth-limited duration of ∼3.2 fs obtained from the laser spectrum depicted in Fig. 1(c). Pump-probe measurements (Fig. 3) in section 3.1 also demonstrate such a high time resolution. Nevertheless, the retrieved temporal phase (red line in Fig. 2(b)) shows several characteristic phase jumps that arise from the residual chirp introduced by the custom-designed DCMs. These phase jumps and weak pre- and post-pulses seen in the intensity profile in Fig. 2(b) (black line) currently limit the time duration of the compressed pulses.
Fig. 2. Characterization of laser pulse duration and stability. a: Transient grating frequency resolved optical gating (TG-FROG) measurement allowing to accurately characterize the pulse duration of the compressed broadband pulses. b: Retrieved temporal intensity and phase of the laser pulses with a full-width at half-maximum (FWHM) duration of 5.8 fs. c: Single-shot pulse stability measured over a train of 10000 consecutive pulses. The inset shows a histogram of the pulse energy with a standard deviation of 0.8%. d: Spectral single-shot stability of the probe behind the setup measured over 1000 consecutive laser pulses. The spectra are recorded using the fast line camera without spectral intensity correction.
2.3 Stability
For the desired spectroscopic applications, it is required to not only achieve an optimum spectral profile, i.e., high spectral density for wavelengths below 500 nm, but also a good spatial beam profile and shot-to-shot stability. This can be achieved by 1) second- and third-order dispersion compensation of the pump pulses (individually controllable via the grating compressor of the amplifier system), 2) manual hollow-core fiber alignment (set via two XY stages), and 3) gas pressure variation. While it is usually straightforward to find the optimal compressor position, the fine-adjustment of the fiber alignment is more difficult, since it is used to optimize the throughput, spatial beam profile and spectral shape but also influences the output stability. Tuning the gas pressure usually results in a tradeoff between spectral broadening, where a higher pressure results in stronger broadening, and shot-to-shot stability, which deteriorates with increasing gas pressure. Good results were achieved using pressures between 2 to 2.4 bar absolute pressure for neon, depending on laser performance and fiber condition.
The long-term stability, measured behind the first pair of DCMs with a thermal sensor (S302C, Thorlabs) over one hour, is presented in Fig. 1(d). The data were sampled every second. We were able to greatly enhance the stability by properly shielding all laser beams up to the 2DES setup from air flow using tubing and by enclosing the hollow-core fiber apparatus. This way we achieve a stability of 0.8% (standard deviation), mainly limited by residual air turbulence and periodic fluctuations in temperature (<0.2°C) and relative humidity (<2%) caused by the air conditioning system of the laboratory.
Shot-to-shot fluctuations in pulse energy were monitored using a photo diode (PD) with a bandwidth of 10 MHz that has been synchronized to the laser system. Figure 2(c) presents the pulse energies of 10000 consecutive laser pulses, showing a standard deviation of 0.8%. The shot-to-shot fluctuations are slightly worse than those reported by Mewes et al. (0.3-0.5%) [55] and slightly better than those reported by Seiler et al. (1%) [54], both recorded for differentially pumped Argon-filled hollow-core fibers. To also characterize and optimize the spectral dependence of the shot-to-shot fluctuations, we measure 1000 consecutive single-shot spectra using a fast line camera (Aviiva EM4, e2 v) mounted to a grating monochromator (Acton SP-2150, Princeton Instruments, 150 l/mm grating blazed for 300 nm). The recorded mean spectrum and normalized root-mean-square error (RMSE) are plotted in Fig. 2(d). The spectral RMSE shows a flat profile with values close to 1% for the full spectral range. The spectral fluctuations are sensitive to the hollow-core fiber alignment, amplifier compressor position and gas pressure and are therefore closely monitored when optimizing the pulse generation.
2.4 Two-dimensional electronic spectroscopy setup
The schematic of our 2DES setup is shown in Fig. 1(a). After initial chirp compensation the beam is split via a beam splitter (BS) into a pump and a probe arm. Both beams are sent through variable neutral density filters (NDF) to tune the pulse energy and a set of fused silica wedge pairs (WP) to fine-tune the total and relative dispersion of the two arms. Both beams then pass through a custom-made rotating slotted wheel mounted to an optical chopper system (MC2000B, Thorlabs) with a 2:1 duty cycle for shot-to-shot acquisition. This allows to generate pulse sequences of pump on/probe on; pump on/probe off and pump off/probe on, which enable real-time scattering correction of the recorded data, similar to the acquisition scheme reported by Son et al. [52]. To generate the phase-stable pump pulse pair that is required for a 2DES experiment, we utilize a common-path interferometer based on birefringent α-BBO wedges and a set of polarizers (Translating Wedge-based Identical pulse eNcoding System, TWINS [65]). This approach allows to generate a pump-pulse pair with attosecond phase stability and has previously been demonstrated to operate in a wide range of wavelengths [66,67]. By moving a set of wedges in and out of the beam using a motorized linear translation stage (M112.1DG1, Physik Instrumente) the pump pulse delay τ, also labeled as the coherence time, can be scanned with high precision and moderately high speed. A thin glass plate (GP) behind the second polarizer reflects a fraction of the pump beam to a PD to monitor the pulse energy and to record a field autocorrelation every time a τ-scan is performed [67]. To compensate the additional dispersion introduced by the optical components a second pair of DMCs is placed in the pump arm.
Before pump and probe beam are recombined at the sample position, a retroreflector mounted on a motorized linear translation stage (M126.DG1, Physik Instrumente) is used to control the delay between the pump pulse pair and the probe pulse. The delay between the second pump and the probe pulse is denoted the waiting time T. A piezoelectric vibrating mirror with a modulation frequency of 1.25 kHz is used in the probe arm to destroy any coherence between the pump and probe pulses [67]. This suppresses the formation of a stable spectral interference pattern in the recorded probe spectrum arising from pump light scattering into the probe direction by modulating the phase between pump and probe in first order by π. While for the reported pulses the effect on time resolution for this approach is small (half oscillation period ∼0.8 fs at 500 nm), it may have to be considered when further pushing the pulse duration toward the few-cycle regime. The probe polarization can also be tuned with a polarizer.
Both pump and probe are focused using an off-axis parabolic mirror (75 mm reflective focal length) to a spot size of ∼30 × 30 µm2 as characterized with a CMOS camera (DCC1545 M, Thorlabs). The pulse compression is optimized by placing a 10-µm β-BBO crystal at the sample position and performing a second-harmonic FROG measurement recording the cross-correlation between pump and probe arms. In case a cuvette is used in the experiment, a substrate with the same thickness as the cuvette window is placed in the beam during chirp optimization to pre-compensate for the additional dispersion.
In a 2DES or PP experiment a sample is placed at the intersection between pump and probe and the transmitted (or reflected) probe beam is sent to a grating monochromator with a mounted sensitive high-speed line camera, yielding the detection energy axis Edet. For every sequence of three pulses provided by the chopping scheme we can record spectra Spu,pr where pu (pump) or pr (probe) can be either on or off. This results in differential transmission spectra
In a PP measurement we set τ = 0 fs and only scan the waiting time T. The PD signal can be used to correct, to first order, slow pump pulse fluctuations and thus to enhance the signal-to-noise ratio of the experiment by reducing the effect of drifts and fluctuations on the recorded dynamics. The pump power available at the sample position is ∼1 µJ.
3. Results and discussion
3.1 Coherent vibrational spectroscopy
To demonstrate the high temporal resolution of the experiment we perform a series of PP measurements on bare solvent and dissolved dye molecules. In these time-resolved vibrational spectroscopy experiments [68,69], the sample is impulsively excited by the short pump, launching coherent vibrational wavepacket motion in either the ground or excited state potential energy landscape, which will modulate the measured differential spectra. By choosing an off-resonant or on-resonant excitation, we can thus probe molecular ground or ground and excited state dynamics, respectively.
As a first sample we select dichloromethane, measured in a 1-mm quartz cuvette (Hellma) at room temperature. Figure 3(a) displays the PP map for waiting times up to 3 ps, recorded with parallel pump and probe polarization using a pump pulse energy of 30 nJ. Close to zero waiting time, the nonlinear signal is dominated by a strong spectral modulation that arises from a pump-induced change in the sample refractive index via off-resonant cross-phase modulation (XPM) [70]. For T > 0 fs, we observe persistent coherent oscillations arising from coherent vibrational motion in the ground state of dichloromethane, launched by impulsive stimulated Raman scattering. To better visualize the weak oscillatory components of the PP signal, we show in Fig. 3(b) residuals of the PP data, obtained after subtracting any slowly temporally-varying PP background in the PP map. In Fig. 3(b), the first 150 fs have been discarded due to the strong XPM contribution. The coherent solvent oscillations, impulsively launched by the off-resonant pump pulse, probe Raman modes of the solvent molecules [68]. A Fourier transform (Fig. 3(c)), obtained after filtering the residuals with a Kaiser-Bessel window, clearly reveals the frequencies of the vibrational modes, in good agreement with Raman spectra of dichloromethane reported in the literature [71]. Specifically, the Fourier transform spectrum shows two dominant peaks at 285 cm-1 and 703 cm-1 which arise from C-Cl2 bending and stretching intramolecular vibrations of the dichloromethane molecule, respectively [72]. Weaker features at 1422 cm-1 und 2986 cm-1 are in good agreement with the C-H2 bending and stretching modes, respectively [72]. The fastest mode that is resolved has a frequency of 2986 cm-1 which corresponds to a time-domain oscillation period of ∼11 fs, highlighting the high time resolution of the setup.
Fig. 3. Time-resolved vibrational spectroscopy measurements demonstrating the high temporal resolution of the experimental setup in Fig. 1. a: Pump-probe measurements of liquid dichloromethane in a 1-mm cuvette for waiting times up to 3 ps. Around waiting time zero, cross-phase modulation (XPM) signals are visible, for later waiting times coherent solvent oscillations can be observed. b: A closer look at the oscillatory component of the measurement is shown in form of spectral residuals (bottom) and a crosscut for a selected detection energy Edet (top). c: Fourier transform of the residuals after applying a Kaiser-Bessel window (bottom). The integrated Fourier spectrum (top) shows the characteristic Raman modes of dichloromethane [75], including the 2986 cm-1 mode, corresponding to ∼11 fs oscillation period in the time domain. d-f: The same as in a-c but for an aqueous solution of rhodamine 6 G.
The second sample under investigation is an aqueous solution of rhodamine 6 G, excited by using a resonant pump pulse with an energy of 6 nJ. The same analysis as for the measurement of dichloromethane is presented in Fig. 3(d)-(f). In contrast to the first experiment, the resonant electronic nonlinearity of the rhodamine 6 G molecule dominates the signal while the faint XPM signal is much weaker. For T > 0 fs, positive ground state bleaching (GSB) and stimulated emission (SE) can be observed for Edet < 2.7 eV, while for higher energies a negative differential transmission band denotes an excited state absorption (ESA) band. These signatures are in good agreement with previously reported transient absorption spectra of rhodamine 6 G in aqueous solution [73]. To isolate the oscillatory component of the experimental data, Fig. 3(e) presents residuals that clearly show coherent oscillations of the GSB/SE band. Energy-independent solvent oscillations, as observed in dichloromethane, are absent when using water as the solvent. The observed oscillations now reflect coherent vibrational wavepacket motion in the harmonic ground state of rhodamine 6 G and on the displaced excited state potential energy surface of the dye molecules [69]. A Fourier transform reveals a dominant mode at 608 cm-1 and several weaker modes with frequencies up to 1653 cm-1, in good agreement with reported Raman spectra [74].
These measurements convincingly demonstrate that time resolution and sensitivity of the experimental setup to probe coherent vibrational dynamics of high-frequency vibrational modes for a wide spectral range of electronic transitions.
3.2 Two-dimensional electronic spectroscopy of chlorophyll a
We now want to finally demonstrate the key advantage of the experimental setup for recording 2DES maps covering the full visible spectrum with high time resolution. To this aim, we select chlorophyll a (Sigma Aldrich, > 95%) as a chromophore. Chlorophyll a takes a prime role in photosynthesis and also finds use as a dye with large oscillator strength in, e.g., strong coupling experiments [76]. The linear absorption spectrum of chlorophyll a dissolved in ethanol is shown in Fig. 4(c), right panel. It consists of two main features, known as the Soret and Q bands at ∼430 nm and ∼665 nm, respectively. Both bands show considerable substructure, especially well pronounced for the Q band, where it reflects a high-frequency vibronic progression of the vibronically coupled Qy and Qx states [77,78]. So far, 2DES of chlorophyll molecules was mostly limited to investigating the Q band [79–81].
Fig. 4. Two-dimensional electronic spectroscopy (2DES) of chlorophyll a in ethanol covering the full visible range. a: Field autocorrelation measurement between the two pump pulses measured by a photo diode (PD). b: Spectrum and spectral phase obtained by Fourier transform of the autocorrelation measurement showing the excitation pulse profile and the finite phase difference between the two pump pulses. c: 2DES map at T = 2 ps of chlorophyll a (center panel). The linear absorption (right panel) shows two bands, the Soret band in the blue and the Q band in the red spectral range. Spectral crosscuts through these bands along the excitation and detection energy axis are shown in the bottom and left panel, respectively.
A 2DES measurement of chlorophyll a in ethanol using a pulse energy of 10 nJ (τ = 0 fs) and parallel polarized pump and probe pulses is shown in Fig. 4(c) for a waiting time of T = 2 ps. On the diagonal, it shows GSB/SE bands of the 0-0 transition of the Qy band (1.86 eV) and of the broad, box-like Soret band (∼2.86 eV) [82]. While the Soret band does not show a pronounced substructure, the Q band reveals strong above-diagonal cross-peaks likely reflecting Qx → Qy and vibrational relaxation [78,83–85]. The spectral region between the Q and Soret band shows weak negative ESA signals, consistent with the literature [86,87]. In addition, strong cross peaks, both above and below the diagonal, appear between the two bands. While the above-diagonal Soret-Qy cross-peak may result from incoherent population relaxation by Soret and Q band [83,85], the origin of the below-diagonal Q-Soret cross peak(s) is surprising. In principle, these cross peaks can again point to an incoherent relaxation channel from the Q to the Soret band. Given the large splitting between both bands of ΔE ≈ 1 eV, such a process is energetically highly unfavorable and thus seems unlikely. Such cross-peaks may arise from a Herzberg-Teller coupling that leads to an intensity-borrowing between Soret and Q bands in porphyrins [88]. Alternatively, the cross peaks can be a sign of coherent coupling between Q and Soret band. The signature of such a coherent coupling would be an oscillatory population transfer between both bands. Due to the large detuning between both states, the corresponding oscillation period (Rabi period) TR = h /ΔE should be ∼4 fs. The high time resolution provided by the spectral bandwidth would, in principle, allow us to give direct evidence for such largely off-resonant coherent couplings by probing the resulting 2DES peak oscillation along the waiting time axis. Experiments in this direction are currently underway.
4. Conclusion and outlook
We have described and demonstrated a broadband and ultrafast 2DES and PP setup based on hollow-core fiber white light generation that combines spectral coverage of the full visible range (400-700 nm) and a time resolution of less than 10 fs. Using a TWINS interferometer and operating in a partially collinear 2DES geometry allows for straightforward implementation and data acquisition with high stability. White light generation in neon-filled hollow-core fibers provides high-fluence pulses with good pulse-to-pulse stability and a spectral bandwidth covering more than an optical octave (∼350-1000 nm). The achieved spectral coverage in the 2DES experiments is mainly limited by the optical components of the setup and in particular by the double chirped mirrors used for pulse compression. It can easily be adapted if other spectral ranges are needed and possibly be extended even further. The long-term and shot-to-shot stability of the supercontinuum (∼0.8% fluctuations), in combination with a fast chopping and spectral acquisition scheme at full laser repetition rate, provides an excellent signal-to-noise ratio of the 2DES spectra. In the present work, compression of the pulses down to <10 fs is achieved by using custom-made DCMs as verified via TG-FROG and pump-probe measurements. These custom chirped mirrors overcome the insufficient chirp compensation of currently commercially available broadband solutions that include the blue spectral region. Residual spectral phase modulations introduced by the broadband mirror design result in weak pulse backgrounds within ∼100 fs around the main pulse (Fig. 2(a),(b)). Currently this limits the time resolution of the compressed pulses. Possibly, further improvements in pulse cleaning may be achieved using further improvements of the mirror design or advanced light field synthesis concepts [57].
The high time resolution of our 2DES setup, reached with broadband light pulses covering a wavelength range from below 400 nm to above 700 nm, makes the setup ideally suited for probing charge and energy transfer in light-harvesting proteins or blue-light absorbing flavoproteins such as the European robin cryptochrome 4, possibly involved in ability of magnetic-field sensing of songbirds [37]. Its high time resolution and broad bandwidth makes our 2DES setup an interesting tool for uncovering previously hidden coherent couplings in multichromophoric molecules, such as the chlorophyll molecules studied here, organic and inorganic semiconductors or quantum emitters in the strong or ultrastrong coupling regime. Its combination with advanced microscopy techniques promises access to the dynamics of individual quantum emitters and to the spatiotemporal transport of optical excitations in hybrid nanosystems.
Funding
Volkswagen Foundation (SMART); Niedersächsisches Ministerium für Wissenschaft und Kultur (DyNano); Deutsche Forschungsgemeinschaft (INST 184/163-1, INST 184/164-1, SFB1372).
Acknowledgments
We want to thank Matthias Wollenhaupt for making the hollow-core fiber available. We want to thank Lars Englert and Marcel Behrens for their help in the hollow-core fiber handling.
Disclosures
Sebastian Riese is currently employed at Layertec GmbH, which manufactured the chirped mirrors used in the experiments.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
Supplemental document
See Supplement 1 for supporting content.
References
1. H. Li, B. Lomsadze, G. Moody, et al., Optical multidimensional coherent spectroscopy (Oxford University, Oxford, 2023).
2. F. D. Fuller and J. P. Ogilvie, “Experimental Implementations of Two-Dimensional Fourier Transform Electronic Spectroscopy,” Annu. Rev. Phys. Chem. 66(1), 667–690 (2015). [CrossRef]
3. S. Biswas, J. Kim, X. Zhang, et al., “Coherent Two-Dimensional and Broadband Electronic Spectroscopies,” Chem. Rev. 122(3), 4257–4321 (2022). [CrossRef]
4. J. D. Hybl, A. W. Albrecht, S. M. G. Faeder, et al., “Two-dimensional electronic spectroscopy,” Chem. Phys. Lett. 297(3-4), 307–313 (1998). [CrossRef]
5. E. Fresch, F. V. Camargo, Q. Shen, et al., “Two-dimensional electronic spectroscopy,” Nat. Rev. Methods Primers 3(1), 84 (2023). [CrossRef]
6. X. C. Dai, M. Richter, H. B. Li, et al., “Two-Dimensional Double-Quantum Spectra Reveal Collective Resonances in an Atomic Vapor,” Phys. Rev. Lett. 108(19), 193201 (2012). [CrossRef]
7. D. F. Liang, L. S. Rodriguez, H. T. Zhou, et al., “Optical two-dimensional coherent spectroscopy of cold atoms,” Opt. Lett. 47(24), 6452–6455 (2022). [CrossRef]
8. T. Brixner, J. Stenger, H. M. Vaswani, et al., “Two-dimensional spectroscopy of electronic couplings in photosynthesis,” Nature 434(7033), 625–628 (2005). [CrossRef]
9. M. Maiuri, J. Rehault, A. M. Carey, et al., “Ultra-broadband 2D electronic spectroscopy of carotenoid-bacteriochlorophyll interactions in the LH1 complex of a purple bacterium,” J. Chem. Phys. 142(21), 212433 (2015). [CrossRef]
10. E. Meneghin, A. Volpato, L. Cupellini, et al., “Coherence in carotenoid-to-chlorophyll energy transfer,” Nat. Commun. 9(1), 3160 (2018). [CrossRef]
11. U. Bangert, F. Stienkemeier, and L. Bruder, “High-resolution two-dimensional electronic spectroscopy reveals the homogeneous line profile of chromophores solvated in nanoclusters,” Nat. Commun. 13(1), 3350 (2022). [CrossRef]
12. D. Timmer, F. Zheng, M. Gittinger, et al., “Charge Delocalization and Vibronic Couplings in Quadrupolar Squaraine Dyes,” J. Am. Chem. Soc. 144(41), 19150–19162 (2022). [CrossRef]
13. G. S. Engel, T. R. Calhoun, E. L. Read, et al., “Evidence for wavelike energy transfer through quantum coherence in photosynthetic systems,” Nature 446(7137), 782–786 (2007). [CrossRef]
14. G. D. Scholes, G. R. Fleming, L. X. Chen, et al., “Using coherence to enhance function in chemical and biophysical systems,” Nature 543(7647), 647–656 (2017). [CrossRef]
15. N. S. Ginsberg, Y. C. Cheng, and G. R. Fleming, “Two-Dimensional Electronic Spectroscopy of Molecular Aggregates,” Acc. Chem. Res. 42(9), 1352–1363 (2009). [CrossRef]
16. A. De Sio, E. Sommer, X. T. Nguyen, et al., “Intermolecular conical intersections in molecular aggregates,” Nat. Nanotechnol. 16(1), 63–68 (2021). [CrossRef]
17. E. Collini and G. D. Scholes, “Coherent Intrachain Energy Migration in a Conjugated Polymer at Room Temperature,” Science 323(5912), 369–373 (2009). [CrossRef]
18. A. De Sio, F. Troiani, M. Maiuri, et al., “Tracking the coherent generation of polaron pairs in conjugated polymers,” Nat. Commun. 7(1), 13742 (2016). [CrossRef]
19. A. A. Bakulin, S. E. Morgan, T. B. Kehoe, et al., “Real-time observation of multiexcitonic states in ultrafast singlet fission using coherent 2D electronic spectroscopy,” Nat. Chem. 8(1), 16–23 (2016). [CrossRef]
20. C. L. Smallwood and S. T. Cundiff, “Multidimensional Coherent Spectroscopy of Semiconductors,” Laser Photonics Rev. 12(12), 1800171 (2018). [CrossRef]
21. X. Q. Li, T. H. Zhang, C. N. Borca, et al., “Many-body interactions in semiconductors probed by optical two-dimensional Fourier transform spectroscopy,” Phys. Rev. Lett. 96(5), 057406 (2006). [CrossRef]
22. V. R. Policht, M. Russo, F. Liu, et al., “Dissecting Interlayer Hole and Electron Transfer in Transition Metal Dichalcogenide Heterostructures via Two-Dimensional Electronic Spectroscopy,” Nano Lett. 21(11), 4738–4743 (2021). [CrossRef]
23. K. W. Stone, K. Gundogdu, D. B. Turner, et al., “Two-Quantum 2D FT Electronic Spectroscopy of Biexcitons in GaAs Quantum Wells,” Science 324(5931), 1169–1173 (2009). [CrossRef]
24. X. T. Nguyen, K. Winte, D. Timmer, et al., “Phonon-driven intra-exciton Rabi oscillations in CsPbBr3 halide perovskites,” Nat. Commun. 14(1), 1047 (2023). [CrossRef]
25. T. M. Autry, G. Nardin, C. L. Smallwood, et al., “Excitation Ladder of Cavity Polaritons,” Phys. Rev. Lett. 125(6), 067403 (2020). [CrossRef]
26. R. Duan, J. N. Mastron, Y. Song, et al., “Isolating Polaritonic 2D-IR Transmission Spectra,” J. Phys. Chem. Lett. 12(46), 11406–11414 (2021). [CrossRef]
27. N. Peruffo, F. Mancin, and E. Collini, “Coherent Dynamics in Solutions of Colloidal Plexcitonic Nanohybrids at Room Temperature,” Adv. Opt. Mater. 11(15), 2203010 (2023). [CrossRef]
28. D. Timmer, M. Gittinger, T. Quenzel, et al., “Plasmon mediated coherent population oscillations in molecular aggregates,” Nat. Commun. 14(1), 8035 (2023). [CrossRef]
29. J. C. Dean and G. D. Scholes, “Coherence Spectroscopy in the Condensed Phase: Insights into Molecular Structure, Environment, and Interactions,” Acc. Chem. Res. 50(11), 2746–2755 (2017). [CrossRef]
30. E. Collini, “2D Electronic Spectroscopic Techniques for Quantum Technology Applications,” J. Phys. Chem. C 125(24), 13096–13108 (2021). [CrossRef]
31. J. Kim, S. Mukamel, and G. D. Scholes, “Two-Dimensional Electronic Double-Quantum Coherence Spectroscopy,” Acc. Chem. Res. 42(9), 1375–1384 (2009). [CrossRef]
32. D. M. Jonas, “Vibrational and Nonadiabatic Coherence in 2D Electronic Spectroscopy, the Jahn-Teller Effect, and Energy Transfer,” Annu. Rev. Phys. Chem. 69(1), 327–352 (2018). [CrossRef]
33. A. De Sio and C. Lienau, “Vibronic coupling in organic semiconductors for photovoltaics,” Phys. Chem. Chem. Phys. 19(29), 18813–18830 (2017). [CrossRef]
34. J. O. Tollerud and J. A. Davis, “Separating pathways in double-quantum optical spectroscopy reveals excitonic interactions,” Laser Photonics Rev. 11(1), 1600249 (2017). [CrossRef]
35. R. Borrego-Varillas, A. Nenov, L. Ganzer, et al., “Two-dimensional UV spectroscopy: a new insight into the structure and dynamics of biomolecules,” Chem. Sci. 10(43), 9907–9921 (2019). [CrossRef]
36. G. Marcolin and E. Collini, “Solvent-Dependent Characterization of Fucoxanthin through 2D Electronic Spectroscopy Reveals New Details on the Intramolecular Charge-Transfer State Dynamics,” J. Phys. Chem. Lett. 12(20), 4833–4840 (2021). [CrossRef]
37. D. Timmer, A. Frederiksen, D. C. Lunemann, et al., “Tracking the Electron Transfer Cascade in European Robin Cryptochrome 4 Mutants,” J. Am. Chem. Soc. 145(21), 11566–11578 (2023). [CrossRef]
38. N. M. Kearns, R. D. Mehlenbacher, A. C. Jones, et al., “Broadband 2D electronic spectrometer using white light and pulse shaping: noise and signal evaluation at 1 and 100 kHz,” Opt. Express 25(7), 7869–7883 (2017). [CrossRef]
39. M. Bradler and E. Riedle, “Sub-20 fs µJ-energy pulses tunable down to the near-UV from a 1 MHz Yb-fiber laser system,” Opt. Lett. 39(9), 2588–2591 (2014). [CrossRef]
40. Y. Song, A. Konar, R. Sechrist, et al., “Multispectral multidimensional spectrometer spanning the ultraviolet to the mid-infrared,” Rev. Sci. Instrum. 90(1), 013108 (2019). [CrossRef]
41. G. Cerullo and S. De Silvestri, “Ultrafast optical parametric amplifiers,” Rev. Sci. Instrum. 74(1), 1–18 (2003). [CrossRef]
42. C. Manzoni, D. Polli, and G. Cerullo, “Two-color pump-probe system broadly tunable over the visible and the near infrared with sub-30 fs temporal resolution,” Rev. Sci. Instrum. 77(2), 023103 (2006). [CrossRef]
43. A. Grupp, A. Budweg, M. P. Fischer, et al., “Broadly tunable ultrafast pump-probe system operating at multi-kHz repetition rate,” J. Opt. 20(1), 014005 (2018). [CrossRef]
44. C. Homann, C. Schriever, P. Baum, et al., “Octave wide tunable UV-pumped NOPA: pulses down to 20 fs at 0.5 MHz repetition rate,” Opt. Express 16(8), 5746–5756 (2008). [CrossRef]
45. R. B. Varillas, A. Candeo, D. Viola, et al., “Microjoule-level, tunable sub-10 fs UV pulses by broadband sum-frequency generation,” Opt. Lett. 39(13), 3849–3852 (2014). [CrossRef]
46. R. Borrego-Varillas, A. Oriana, L. Ganzer, et al., “Two-dimensional electronic spectroscopy in the ultraviolet by a birefringent delay line,” Opt. Express 24(25), 28491–28499 (2016). [CrossRef]
47. C. H. Lu, Y. J. Tsou, H. Y. Chen, et al., “Generation of intense supercontinuum in condensed media,” Optica 1(6), 400–406 (2014). [CrossRef]
48. P. He, Y. Y. Liu, K. Zhao, et al., “High-efficiency supercontinuum generation in solid thin plates at 0.1 TW level,” Opt. Lett. 42(3), 474–477 (2017). [CrossRef]
49. C. H. Lu, W. H. Wu, S. H. Kuo, et al., “Greater than 50 times compression of 1030 nm Yb:KGW laser pulses to single-cycle duration,” Opt. Express 27(11), 15638–15648 (2019). [CrossRef]
50. B. Spokoyny, C. J. Koh, and E. Harel, “Stable and high-power few cycle supercontinuum for 2D ultrabroadband electronic spectroscopy,” Opt. Lett. 40(6), 1014–1017 (2015). [CrossRef]
51. H. B. Zheng, J. R. Caram, P. D. Dahlberg, et al., “Dispersion-free continuum two-dimensional electronic spectrometer,” Appl. Opt. 53(9), 1909–1917 (2014). [CrossRef]
52. M. Son, S. Mosquera-Vazquez, and G. S. Schlau-Cohen, “Ultrabroadband 2D electronic spectroscopy with high-speed, shot-to-shot detection,” Opt. Express 25(16), 18950–18962 (2017). [CrossRef]
53. X. N. Ma, J. Dostal, and T. Brixner, “Broadband 7-fs diffractive-optic-based 2D electronic spectroscopy using hollow-core fiber compression,” Opt. Express 24(18), 20781–20791 (2016). [CrossRef]
54. H. Seiler, S. Palato, B. E. Schmidt, et al., “Simple fiber-based solution for coherent multidimensional spectroscopy in the visible regime,” Opt. Lett. 42(3), 643–646 (2017). [CrossRef]
55. L. Mewes, R. A. Ingle, A. Al Haddad, et al., “Broadband visible two-dimensional spectroscopy of molecular dyes,” J. Chem. Phys. 155(3), 034201 (2021). [CrossRef]
56. M. Nisoli, G. Sansone, S. Stagira, et al., “Ultra-broadband continuum generation by hollow-fiber cascading,” Applied Physics B: Lasers and Optics 75(6-7), 601–604 (2002). [CrossRef]
57. A. Wirth, M. T. Hassan, I. Grguras, et al., “Synthesized Light Transients,” Science 334(6053), 195–200 (2011). [CrossRef]
58. M. Nisoli, S. DeSilvestri, and O. Svelto, “Generation of high energy 10 fs pulses by a new pulse compression technique,” Appl. Phys. Lett. 68(20), 2793–2795 (1996). [CrossRef]
59. M. Nisoli, S. DeSilvestri, O. Svelto, et al., “Compression of high-energy laser pulses below 5 fs,” Opt. Lett. 22(8), 522–524 (1997). [CrossRef]
60. S. Kerbstadt, D. Timmer, L. Englert, et al., “Ultrashort polarization-tailored bichromatic fields from a CEP-stable white light supercontinuum,” Opt. Express 25(11), 12518–12530 (2017). [CrossRef]
61. T. Brixner and G. Gerber, “Femtosecond polarization pulse shaping,” Opt. Lett. 26(8), 557–559 (2001). [CrossRef]
62. E. Zeek, K. Maginnis, S. Backus, et al., “Pulse compression by use of deformable mirrors,” Opt. Lett. 24(7), 493–495 (1999). [CrossRef]
63. S. Bonora, D. Brida, P. Villoresi, et al., “Ultrabroadband pulse shaping with a push-pull deformable mirror,” Opt. Express 18(22), 23147–23152 (2010). [CrossRef]
64. G. Steinmeyer, D. H. Sutter, L. Gallmann, et al., “Frontiers in ultrashort pulse generation: Pushing the limits in linear and nonlinear optics,” Science 286(5444), 1507–1512 (1999). [CrossRef]
65. D. Brida, C. Manzoni, and G. Cerullo, “Phase-locked pulses for two-dimensional spectroscopy by a birefringent delay line,” Opt. Lett. 37(15), 3027–3029 (2012). [CrossRef]
66. J. Rehault, M. Maiuri, C. Manzoni, et al., “2D IR spectroscopy with phase-locked pulse pairs from a birefringent delay line,” Opt. Express 22(8), 9063–9072 (2014). [CrossRef]
67. J. Rehault, M. Maiuri, A. Oriana, et al., “Two-dimensional electronic spectroscopy with birefringent wedges,” Rev. Sci. Instrum. 85(12), 123107 (2014). [CrossRef]
68. M. Liebel, C. Schnedermann, T. Wende, et al., “Principles and Applications of Broadband Impulsive Vibrational Spectroscopy,” J. Phys. Chem. A 119(36), 9506–9517 (2015). [CrossRef]
69. G. Lanzani, G. Cerullo, and S. De Silvestri, Coherent Vibrational Dynamics (CRC, 2007).
70. K. Ekvall, P. van der Meulen, C. Dhollande, et al., “Cross phase modulation artifact in liquid phase transient absorption spectroscopy,” J. Appl. Phys. 87(5), 2340–2352 (2000). [CrossRef]
71. T. E. Acosta-Maeda, A. K. Misra, J. N. Porter, et al., “Remote Raman Efficiencies and Cross-Sections of Organic and Inorganic Chemicals,” Appl. Spectrosc. 71(5), 1025–1038 (2017). [CrossRef]
72. T. Shimanouchi, Tables of molecular vibrational frequencies: consolidated volume I, NSRDS-NBS 39. (National Bureau of Standards, Washington, D.C, 1972), pp. iv, 160 p.
73. M. Fedoseeva, R. Letrun, and E. Vauthey, “Excited-State Dynamics of Rhodamine 6 G in Aqueous Solution and at the Dodecane/Water Interface,” J. Phys. Chem. B 118(19), 5184–5193 (2014). [CrossRef]
74. S. Shim, C. M. Stuart, and R. A. Mathies, “Resonance Raman cross-sections and vibronic analysis of rhodamine 6 G from broadband stimulated Raman spectroscopy,” Chemphyschem 9(5), 697–699 (2008). [CrossRef]
75. P. Bishui, D. Mukherjee, and S. Sirkar, “Raman Spectra of CH2Cl2. CHCl3 And CCl4 At 64° K,” Indian J. Phys. 41, 553–558 (1967).
76. A. Singh, G. Sharma, R. Ghosh, et al., “Strong Dipole Interaction between Chlorophyll-a Molecules and Surface Plasmon Polaritons,” J. Phys. Chem. C 123(27), 16965–16972 (2019). [CrossRef]
77. M. Ratsep, J. Linnanto, and A. Freiberg, “Mirror symmetry and vibrational structure in optical spectra of chlorophyll a,” J. Chem. Phys. 130(19), 194501 (2009). [CrossRef]
78. J. R. Reimers, Z. L. Cai, R. Kobayashi, et al., “Assignment of the Q-Bands of the Chlorophylls: Coherence Loss via Q(x) - Q(y) Mixing,” Sci. Rep. 3(1), 2761 (2013). [CrossRef]
79. R. Moca, S. R. Meech, and I. A. Heisler, “Two-Dimensional Electronic Spectroscopy of Chlorophyll a: Solvent Dependent Spectral Evolution,” J. Phys. Chem. B 119(27), 8623–8630 (2015). [CrossRef]
80. K. L. Wells, Z. Y. Zhang, J. R. Rouxel, et al., “Measuring the Spectral Diffusion of Chlorophyll a Using Two-Dimensional Electronic Spectroscopy,” J. Phys. Chem. B 117(8), 2294–2299 (2013). [CrossRef]
81. E. Meneghin, C. Leonardo, A. Volpato, et al., “Mechanistic insight into internal conversion process within Q-bands of chlorophyll α,” Sci. Rep. 7(1), 11389 (2017). [CrossRef]
82. J. Du, T. Teramoto, K. Nakata, et al., “Real-Time Vibrational Dynamics in Chlorophyll a Studied with a Few-Cycle Pulse Laser,” Biophys. J. 101(4), 995–1003 (2011). [CrossRef]
83. W. P. Bricker, P. M. Shenai, A. Ghosh, et al., “Non-radiative relaxation of photoexcited chlorophylls: theoretical and experimental study,” Sci. Rep. 5(1), 13625 (2015). [CrossRef]
84. T. N. Do, H. L. Nguyen, S. Caffarri, et al., “Two-dimensional electronic spectroscopy of the Q(x) to Q(y) relaxation of chlorophylls a in photosystem II core complex,” J. Chem. Phys. 156(14), 145102 (2022). [CrossRef]
85. Y. Shi, J. Y. Liu, and K. L. Han, “Investigation of the internal conversion time of the chlorophyll a from S-3, S-2 to S-1,” Chem. Phys. Lett. 410(4-6), 260–263 (2005). [CrossRef]
86. L. De Boni, D. S. Correa, F. J. Pavinatto, et al., “Excited state absorption spectrum of chlorophyll a obtained with white-light continuum,” J. Chem. Phys. 126(16), 165102 (2007). [CrossRef]
87. Y. Silori, S. Chawla, and A. K. De, “Unravelling the Role of Water in Ultrafast Excited-State Relaxation Dynamics within Nano-Architectures of Chlorophyll a,” Chemphyschem 21(17), 1908–1917 (2020). [CrossRef]
88. P. P. Roy, S. Kundu, N. Makri, et al., “Interference between Franck-Condon and Herzberg-Teller Terms in the Condensed-Phase Molecular Spectra of Metal-Based Tetrapyrrole Derivatives,” J. Phys. Chem. Lett. 13(32), 7413–7419 (2022). [CrossRef]
