Refine
Year of publication
- 2016 (258) (remove)
Document Type
- Article (203)
- Doctoral Thesis (21)
- Postprint (18)
- Other (10)
- Review (4)
- Habilitation Thesis (2)
Is part of the Bibliography
- yes (258)
Keywords
- Magellanic Clouds (8)
- Sun: magnetic fields (6)
- intergalactic medium (6)
- methods: data analysis (6)
- stars: Wolf-Rayet (6)
- techniques: spectroscopic (6)
- galaxies: formation (5)
- stars: atmospheres (5)
- turbulence (5)
- X-rays: binaries (4)
Institute
- Institut für Physik und Astronomie (258) (remove)
In this thesis we use integral-field spectroscopy to detect and understand of Lyman α (Lyα) emission from high-redshift galaxies.
Intrinsically the Lyα emission at λ = 1216 Å is the strongest recombination line from galaxies. It arises from the 2p → 1s transition in hydrogen. In star-forming galaxies the line is powered by ionisation of the interstellar gas by hot O- and B- stars. Galaxies with star-formation rates of 1 - 10 Msol/year are expected to have Lyα luminosities of 42 dex - 43 dex (erg/s), corresponding to fluxes ~ -17 dex - -18 dex (erg/s/cm²) at redshifts z~3, where Lyα is easily accessible with ground-based telescopes. However, star-forming galaxies do not show these expected Lyα fluxes. Primarily this is a consequence of the high-absorption cross-section of neutral hydrogen for Lyα photons σ ~ -14 dex (cm²). Therefore, in typical interstellar environments Lyα photons have to undergo a complex radiative transfer. The exact conditions under which Lyα photons can escape a galaxy are poorly understood.
Here we present results from three observational projects. In Chapter 2, we show integral field spectroscopic observations of 14 nearby star-forming galaxies in Balmer α radiation (Hα, λ = 6562.8 Å). These observations were obtained with the Potsdam Multi Aperture Spectrophotometer at the Calar-Alto 3.5m Telescope}. Hα directly traces the intrinsic Lyα radiation field. We present Hα velocity fields and velocity dispersion maps spatially registered onto Hubble Space Telescope Lyα and Hα images. From our observations, we conjecture a causal connection between spatially resolved Hα kinematics and Lyα photometry for individual galaxies. Statistically, we find that dispersion-dominated galaxies are more likely to emit Lyα photons than galaxies where ordered gas-motions dominate. This result indicates that turbulence in actively star-forming systems favours an escape of Lyα radiation.
Not only massive stars can power Lyα radiation, but also non-thermal emission from an accreting super-massive black hole in the galaxy centre. If a galaxy harbours such an active galactic nucleus, the rate of hydrogen-ionising photons can be more than 1000 times higher than that of a typical star-forming galaxy. This radiation can potentially ionise large regions well outside the main stellar body of galaxies. Therefore, it is expected that the neutral hydrogen from these circum-galactic regions shines fluorescently in Lyα. Circum-galactic gas plays a crucial role in galaxy formation. It may act as a reservoir for fuelling star formation, and it is also subject to feedback processes that expel galactic material. If Lyα emission from this circum-galactic medium (CGM) was detected, these important processes could be studied in-situ around high-z galaxies. In Chapter 3, we show observations of five radio-quiet quasars with PMAS to search for possible extended CGM emission in the Lyα line. However, in four of the five objects, we find no significant traces of this emission. In the fifth object, there is evidence for a weak and spatially quite compact Lyα excess at several kpc outside the nucleus. The faintness of these structures is consistent with the idea that radio-quiet quasars typically reside in dark matter haloes of modest masses. While we were not able to detect Lyα CGM emission, our upper limits provide constraints for the new generation of IFS instruments at 8--10m class telescopes.
The Multi Unit Spectroscopic Explorer (MUSE) at ESOs Very Large Telescopeis such an unique instrument. One of the main motivating drivers in its construction was the use as a survey instrument for Lyα emitting galaxies at high-z. Currently, we are conducting such a survey that will cover a total area of ~100 square arcminutes with 1 hour exposures for each 1 square arcminute MUSE pointing. As a first result from this survey we present in Chapter 5 a catalogue of 831 emission-line selected galaxies from a 22.2 square arcminute region in the Chandra Deep Field South. In order to construct the catalogue, we developed and implemented a novel source detection algorithm -- LSDCat -- based on matched filtering for line emission in 3D spectroscopic datasets (Chapter 4). Our catalogue contains 237 Lyα emitting galaxies in the redshift range 3 ≲ z ≲ 6. Only four of those previously had spectroscopic redshifts in the literature. We conclude this thesis with an outlook on the construction of a Lyα luminosity function based on this unique sample (Chapter 6).
Context. For the spectral analysis of high-resolution and high signal-to-noise (S/N) spectra of hot stars, state-of-the-art non-local thermodynamic equilibrium (NLTE) model atmospheres are mandatory. These are strongly dependent on the reliability of the atomic data that is used for their calculation. Aims. New Kr IV-VII oscillator strengths for a large number of lines enable us to construct more detailed model atoms for our NLTE model-atmosphere calculations. This enables us to search for additional Kr lines in observed spectra and to improve Kr abundance determinations. Methods. We calculated Kr IV-VII oscillator strengths to consider radiative and collisional bound-bound transitions in detail in our NLTE stellar-atmosphere models for the analysis of Kr lines that are exhibited in high-resolution and high S/N ultraviolet (UV) observations of the hot white dwarf RE 0503-289. Results. We reanalyzed the effective temperature and surface gravity and determined T-eff = 70 000 +/- 2000 K and log (g/cm s(-2)) = 7.5 +/- 0.1. We newly identified ten Kr V lines and one Kr vi line in the spectrum of RE 0503-289. We measured a Kr abundance of 3.3 +/- 0.3 (logarithmic mass fraction). We discovered that the interstellar absorption toward RE 0503-289 has a multi-velocity structure within a radial-velocity interval of -40 km s(-1) < upsilon(rad) < +18 km s(-1). Conclusions. Reliable measurements and calculations of atomic data are a prerequisite for state-of-the-art NLTE stellar-atmosphere modeling. Observed Kr V-VII line profiles in the UV spectrum of the white dwarf RE 0503-289 were simultaneously well reproduced with our newly calculated oscillator strengths.
In this thesis, the two prototype catalysts Fe(CO)₅ and Cr(CO)₆ are investigated with time-resolved photoelectron spectroscopy at a high harmonic setup. In both of these metal carbonyls, a UV photon can induce the dissociation of one or more ligands of the complex. The mechanism of the dissociation has been debated over the last decades. The electronic dynamics of the first dissociation occur on the femtosecond timescale.
For the experiment, an existing high harmonic setup was moved to a new location, was extended, and characterized. The modified setup can induce dynamics in gas phase samples with photon energies of 1.55eV, 3.10eV, and 4.65eV. The valence electronic structure of the samples can be probed with photon energies between 20eV and 40eV. The temporal resolution is 111fs to 262fs, depending on the combination of the two photon energies.
The electronically excited intermediates of the two complexes, as well as of the reaction product Fe(CO)₄, could be observed with photoelectron spectroscopy in the gas phase for the first time. However, photoelectron spectroscopy gives access only to the final ionic states. Corresponding calculations to simulate these spectra are still in development. The peak energies and their evolution in time with respect to the initiation pump pulse have been determined, these peaks have been assigned based on literature data. The spectra of the two complexes show clear differences. The dynamics have been interpreted with the assumption that the motion of peaks in the spectra relates to the movement of the wave packet in the multidimensional energy landscape. The results largely confirm existing models for the reaction pathways. In both metal carbonyls, this pathway involves a direct excitation of the wave packet to a metal-to-ligand charge transfer state and the subsequent crossing to a dissociative ligand field state. The coupling of the electronic dynamics to the nuclear dynamics could explain the slower dissociation in Fe(CO)₅ as compared to Cr(CO)₆.
The molecular ability to selectively and efficiently convert sunlight into other forms of energy like heat, bond change, or charge separation is truly remarkable. The decisive steps in these transformations often happen on a femtosecond timescale and require transitions among different electronic states that violate the Born-Oppenheimer approximation (BOA). Non-BOA transitions pose challenges to both theory and experiment. From a theoretical point of view, excited state dynamics and nonadiabatic transitions both are difficult problems (see Figure 1(a)). However, the theory on non-BOA dynamics has advanced significantly over the last two decades. Full dynamical simulations for molecules of the size of nucleobases have been possible for a couple of years and allow predictions of experimental observables like photoelectron energy or ion yield. The availability of these calculations for isolated molecules has spurred new experimental efforts to develop methods that are sufficiently different from all optical techniques. For determination of transient molecular structure, femtosecond X-ray diffraction and electron diffraction have been implemented on optically excited molecules.
Gravity dictates the structure of the whole Universe and, although it is triumphantly described by the theory of General Relativity, it is the force that we least understand in nature. One of the cardinal predictions of this theory are black holes. Massive, dark objects are found in the majority of galaxies. Our own galactic center very contains such an object with a mass of about four million solar masses. Are these objects supermassive black holes (SMBHs), or do we need alternatives? The answer lies in the event horizon, the characteristic that defines a black hole. The key to probe the horizon is to model the movement of stars around a SMBH, and the interactions between them, and look for deviations from real observations. Nuclear star clusters harboring a massive, dark object with a mass of up to ~ ten million solar masses are good testbeds to probe the event horizon of the potential SMBH with stars. The channel for interactions between stars and the central MBH are the fact that (a) compact stars and stellar-mass black holes can gradually inspiral into the SMBH due to the emission of gravitational radiation, which is known as an “Extreme Mass Ratio Inspiral” (EMRI), and (b) stars can produce gases which will be accreted by the SMBH through normal stellar evolution, or by collisions and disruptions brought about by the strong central tidal field. Such processes can contribute significantly to the mass of the SMBH. These two processes involve different disciplines, which combined will provide us with detailed information about the fabric of space and time. In this habilitation I present nine articles of my recent work directly related with these topics.
The cell interior is a highly packed environment in which biological macromolecules evolve and function. This crowded media has effects in many biological processes such as protein-protein binding, gene regulation, and protein folding. Thus, biochemical reactions that take place in such crowded conditions differ from diluted test tube conditions, and a considerable effort has been invested in order to understand such differences.
In this work, we combine different computationally tools to disentangle the effects of molecular crowding on biochemical processes. First, we propose a lattice model to study the implications of molecular crowding on enzymatic reactions. We provide a detailed picture of how crowding affects binding and unbinding events and how the separate effects of crowding on binding equilibrium act together. Then, we implement a lattice model to study the effects of molecular crowding on facilitated diffusion. We find that obstacles on the DNA impair facilitated diffusion. However, the extent of this effect depends on how dynamic obstacles are on the DNA. For the scenario in which crowders are only present in the bulk solution, we find that at some conditions presence of crowding agents can enhance specific-DNA binding. Finally, we make use of structure-based techniques to look at the impact of the presence of crowders on the folding a protein. We find that polymeric crowders have stronger effects on protein stability than spherical crowders. The strength of this effect increases as the polymeric crowders become longer. The methods we propose here are general and can also be applied to more complicated systems.
Changes in extratropical storm track activity and their implications for extreme weather events
(2016)
Can the statistical properties of single-electron transfer events be correctly predicted within a common equilibrium ensemble description? This fundamental in nanoworld question of ergodic behavior is scrutinized within a very basic semi-classical curve-crossing problem. It is shown that in the limit of non-adiabatic electron transfer (weak tunneling) well-described by the Marcus–Levich–Dogonadze(MLD) rate the answer is yes. However, in the limit of the so-called solvent-controlled adiabatic electron transfer, a profound breaking of ergodicity occurs. Namely, a common description based on the ensemble reduced density matrix with an initial equilibrium distribution of the reaction coordinate is not able to reproduce the statistics of single-trajectory events in this seemingly classical regime. For sufficiently large activation barriers, the ensemble survival probability in a state remains nearly exponential with the inverse rate given by the sum of the adiabatic curve crossing (Kramers) time and the inverse MLD rate. In contrast, near to the adiabatic regime, the single-electron survival probability is clearly non-exponential, even though it possesses an exponential tail which agrees well with the ensemble description. Initially, it is well described by a Mittag-Leffler distribution with a fractional rate. Paradoxically, the mean transfer time in this classical on the ensemble level regime is well described by the inverse of the nonadiabatic quantum tunneling rate on a single particle level. An analytical theory is developed which perfectly agrees with stochastic simulations and explains our findings.
In the first part of my work I have investigated the ageing properties of the first passage time distributions in a one-dimensional subdiffusive continuous time random walk with power law distributed waiting times of the form $\psi(\tau) \sim \tau^{-1-\alpha}$ with $0<\alpha<1$ and $1<\alpha<2$. The age or ageing time $t_a$ is the time span from the start of the stochastic process to the start of the observation of this process (at $t=0$). I have calculated the results for a single target and two targets, also including the biased case, where the walker is driven towards the boundary by a constant force. I have furthermore refined the previously derived results for the non-ageing case and investigated the changes that occur when the walk is performed in a discrete quenched energy landscape, where the waiting times are fixed for every site. The results include the exact Laplace space densities and infinite (converging) series as exact results in the time space. The main results are the dominating long time power law behavior regimes, which depend on the ageing time. For the case of unbiased subdiffusion ($\alpha < 1$) in the presence of one target, I find three different dominant terms for ranges of $t$ separated by $t_a$ and another crossover time $t^{\star}$, which depends on $t_a$ as well as on the anomalous exponent $\alpha$ and the anomalous diffusion coefficient $K_{\alpha}$. In all three regimes ($t \ll t_a$, $t_a \ll t \ll t^{\star}$, $t \gg t^{\star}$) one finds power law decay with exponents depending on $\alpha$. The middle regime only exists for $t_a \ll t^{\star}$. The dominant terms in the first two regimes (ageing regimes) come from the probability distribution of the forward waiting time, the time one has to wait for the stochastic process to make the first step during the observation. When the observation time is larger than the second crossover time $t^{\star}$, the first passage time density does not show ageing and the non-ageing first passage time dominates. The power law exponents in the respective regimes are $-\alpha$ for strong ageing, $-1-\alpha$ in the intermediate regime, and $-1-\alpha/2$ in the final non-ageing regime. A similar split into three regimes can be found for $1<\alpha<2$, only with a different second crossover time $t^*$. In this regime the diffusion is normal but also age-dependent. For the diffusion in quenched energy landscapes one cannot detect ageing. The first passage time density shows a quenched power law $^\sim t^{-(1+2\alpha)/(1+\alpha)}$. For diffusion between two target sites and the biased diffusion towards a target only two scaling regimes emerge, separated by the ageing time. In the ageing case $t \ll t_a$ the forward waiting time is again dominant with power law exponent $-\alpha$, while the non-ageing power law $-1-\alpha$ is found for all times $t \gg t_a$. An intermediate regime does not exist. The bias and the confinement have similar effects on the first passage time density. For quenched diffusion, the biased case is interesting, as the bias reduces correlations due to revisiting of the same waiting time. As a result, CTRW like behavior is observed, including ageing. Extensive computer simulations support my findings.
The second part of my research was done on the subject of ageing Scher-Montroll transport, which is in parts closely related to the first passage densities. It explains the electrical current in an amorphous material. I have investigated the effect of the width of a given initial distribution of charge carriers on the transport coefficients as well as the ageing effect on the emerging power law regimes and a constant initial regime. While a spread out initial distribution has only little impact on the Scher-Montroll current, ageing alters the behavior drastically. Instead of the two classical power laws one finds four current regimes, up to three of which can appear in a single experiment. The dominant power laws differ for $t \ll t_a, t_c$, $t_a \ll t \ll t_c$, $t_c \ll t \ll t_a$, and $t \gg t_a,t_c$. Here, $t_c$ is the crossover time of the non-aged Scher-Montroll current. For strongly aged systems one can observe a constant current in the first regime while the others are dominated by decaying power laws with exponents $\alpha -1$, $-\alpha$, and $-1-\alpha$. The ageing regimes are the 1st and 3rd one, while the classical regimes are the 2nd and the 4th. I have verified the theory using numerical integration of the exact integrals and applied the new results to experimental data.
In the third part I considered a single file of subdiffusing particles in an energy landscape. Every occupied site of the landscape acts as a boundary, from which a particle is immediately reflected to its previous site, if it tries to jump there. I have analysed the effects single-file diffusion a quenched landscape compared to an annealed landscape and I have related these results to the number of steps and related quantities. The diffusion changes from ultraslow logarithmic diffusion in the annealed or CTRW case to subdiffusion with an anomalous exponent $\alpha/(1+\alpha)$ in the quenched landscape. The behavior is caused by the forward waiting time, which changes drastically from the quenched to the annealed case. Single-file effects in the quenched landscape are even more complicated to consider in the ensemble average, since the diffusion in individual landscapes shows extremely diverse behavior. Extensive simulations support my theoretical arguments, which consider mainly the long time evolution of the mean square displacement of a bulk particle.
During the summer of 2013, a 4-month spectroscopic campaign took place to observe the variabilities in three Wolf-Rayet stars. The spectroscopic data have been analysed for WR 134 (WN6b), to better understand its behaviour and long-term periodicity, which we interpret as arising from corotating interaction regions (CIRs) in the wind. By analysing the variability of the He ii lambda 5411 emission line, the previously identified period was refined to P = 2.255 +/- 0.008 (s.d.) d. The coherency time of the variability, which we associate with the lifetime of the CIRs in the wind, was deduced to be 40 +/- 6 d, or similar to 18 cycles, by cross-correlating the variability patterns as a function of time. When comparing the phased observational grey-scale difference images with theoretical grey-scales previously calculated from models including CIRs in an optically thin stellar wind, we find that two CIRs were likely present. A separation in longitude of Delta I center dot a parts per thousand integral 90A degrees was determined between the two CIRs and we suggest that the different maximum velocities that they reach indicate that they emerge from different latitudes. We have also been able to detect observational signatures of the CIRs in other spectral lines (C iv lambda lambda 5802,5812 and He i lambda 5876). Furthermore, a DAC was found to be present simultaneously with the CIR signatures detected in the He i lambda 5876 emission line which is consistent with the proposed geometry of the large-scale structures in the wind. Small-scale structures also show a presence in the wind, simultaneously with the larger scale structures, showing that they do in fact co-exist.
We report the detection of extended Ly alpha emission around individual star-forming galaxies at redshifts z = 3-6 in an ultradeep exposure of the Hubble Deep Field South obtained with MUSE on the ESO-VLT. The data reach a limiting surface brightness (1 sigma) of similar to 1 x 10(-19) erg s(-1) cm(-2) arcsec(-2) in azimuthally averaged radial profiles, an order of magnitude improvement over previous narrowband imaging. Our sample consists of 26 spectroscopically confirmed Ly alpha-emitting, but mostly continuum-faint (m(AB) greater than or similar to 27) galaxies. In most objects the Ly alpha emission is considerably more extended than the UV continuum light. While five of the faintest galaxies in the sample show no significantly detected Ly alpha haloes, the derived upper limits suggest that this is due to insufficient S/N. Ly alpha haloes therefore appear to be ubiquitous even for low-mass (similar to 10(8)-10(9) M-circle dot) star-forming galaxies at z > 3. We decompose the Ly alpha emission of each object into a compact component tracing the UV continuum and an extended halo component, and infer sizes and luminosities of the haloes. The extended Ly alpha emission approximately follows an exponential surface brightness distribution with a scale length of a few kpc. While these haloes are thus quite modest in terms of their absolute sizes, they are larger by a factor of 5-15 than the corresponding rest-frame UV continuum sources as seen by HST. They are also much more extended, by a factor similar to 5, than Ly alpha haloes around low-redshift star-forming galaxies. Between similar to 40% and greater than or similar to 90% of the observed Ly alpha flux comes from the extended halo component, with no obvious correlation of this fraction with either the absolute or the relative size of the Ly alpha halo. Our observations provide direct insights into the spatial distribution of at least partly neutral gas residing in the circumgalactic medium of low to intermediate mass galaxies at z > 3.
The Milky Way is only one out of billions of galaxies in the universe. However, it is a special galaxy because it allows to explore the main mechanisms involved in its evolution and formation history by unpicking the system star-by-star. Especially, the chemical fingerprints of its stars provide clues and evidence of past events in the Galaxy’s lifetime. These information help not only to decipher the current structure and building blocks of the Milky Way, but to learn more about the general formation process of galaxies.
In the past decade a multitude of stellar spectroscopic Galactic surveys have scanned millions of stars far beyond the rim of the solar neighbourhood. The obtained spectroscopic information provide unprecedented insights to the chemo-dynamics of the Milky Way. In addition analytic models and numerical simulations of the Milky Way provide necessary descriptions and predictions suited for comparison with observations in order to decode the physical properties that underlie the complex system of the Galaxy.
In the thesis various approaches are taken to connect modern theoretical modelling of galaxy formation and evolution with observations from Galactic stellar surveys. With its focus on the chemo-kinematics of the Galactic disk this work aims to determine new observational constraints on the formation of the Milky Way providing also proper comparisons with two different models. These are the population synthesis model TRILEGAL based on analytical distribution functions, which aims to simulate the number and distribution of stars in the Milky Way and its different components, and a hybrid model (MCM) that combines an N-body simulation of a Milky Way like galaxy in the cosmological framework with a semi-analytic chemical evolution model for the Milky Way. The major observational data sets in use come from two surveys, namely the “Radial Velocity Experiment” (RAVE) and the “Sloan Extension for Galactic Understanding and Exploration” (SEGUE).
In the first approach the chemo-kinematic properties of the thin and thick disk of the Galaxy as traced by a selection of about 20000 SEGUE G-dwarf stars are directly compared to the predictions by the MCM model. As a necessary condition for this, SEGUE's selection function and its survey volume are evaluated in detail to correct the spectroscopic observations for their survey specific selection biases. Also, based on a Bayesian method spectro-photometric distances with uncertainties below 15% are computed for the selection of SEGUE G-dwarfs that are studied up to a distance of 3 kpc from the Sun.
For the second approach two synthetic versions of the SEGUE survey are generated based on the above models. The obtained synthetic stellar catalogues are then used to create mock samples best resembling the compiled sample of observed SEGUE G-dwarfs. Generally, mock samples are not only ideal to compare predictions from various models. They also allow validation of the models' quality and improvement as with this work could be especially achieved for TRILEGAL. While TRILEGAL reproduces the statistical properties of the thin and thick disk as seen in the observations, the MCM model has shown to be more suitable in reproducing many chemo-kinematic correlations as revealed by the SEGUE stars. However, evidence has been found that the MCM model may be missing a stellar component with the properties of the thick disk that the observations clearly show. While the SEGUE stars do indicate a thin-thick dichotomy of the stellar Galactic disk in agreement with other spectroscopic stellar studies, no sign for a distinct metal-poor disk is seen in the MCM model.
Usually stellar spectroscopic surveys are limited to a certain volume around the Sun covering different regions of the Galaxy’s disk. This often prevents to obtain a global view on the chemo-dynamics of the Galactic disk. Hence, a suitable combination of stellar samples from independent surveys is not only useful for the verification of results but it also helps to complete the picture of the Milky Way. Therefore, the thesis closes with a comparison of the SEGUE G-dwarfs and a sample of RAVE giants. The comparison reveals that the chemo-kinematic relations agree in disk regions where the samples of both surveys show a similar number of stars. For those parts of the survey volumes where one of the surveys lacks statistics they beautifully complement each other. This demonstrates that the comparison of theoretical models on the one side, and the combined observational data gathered by multiple surveys on the other side, are key ingredients to understand and disentangle the structure and formation history of the Milky Way.
Imaging changes in molecular geometries on their natural femtosecond timescale with sub-Angstrom spatial precision is one of the critical challenges in the chemical sciences, as the nuclear geometry changes determine the molecular reactivity. For photoexcited molecules, the nuclear dynamics determine the photoenergy conversion path and efficiency. Here we report a gas-phase electron diffraction experiment using megaelectronvolt (MeV) electrons, where we captured the rotational wavepacket dynamics of nonadiabatically laser-aligned nitrogen molecules. We achieved a combination of 100 fs root-mean-squared temporal resolution and sub-Angstrom (0.76 angstrom) spatial resolution that makes it possible to resolve the position of the nuclei within the molecule. In addition, the diffraction patterns reveal the angular distribution of the molecules, which changes from prolate (aligned) to oblate (anti-aligned) in 300 fs. Our results demonstrate a significant and promising step towards making atomically resolved movies of molecular reactions.
We present results on ultrafast gas electron diffraction (UGED) experiments with femtosecond resolution using the MeV electron gun at SLAC National Accelerator Laboratory. UGED is a promising method to investigate molecular dynamics in the gas phase because electron pulses can probe the structure with a high spatial resolution. Until recently, however, it was not possible for UGED to reach the relevant timescale for the motion of the nuclei during a molecular reaction. Using MeV electron pulses has allowed us to overcome the main challenges in reaching femtosecond resolution, namely delivering short electron pulses on a gas target, overcoming the effect of velocity mismatch between pump laser pulses and the probe electron pulses, and maintaining a low timing jitter. At electron kinetic energies above 3 MeV, the velocity mismatch between laser and electron pulses becomes negligible. The relativistic electrons are also less susceptible to temporal broadening due to the Coulomb force. One of the challenges of diffraction with relativistic electrons is that the small de Broglie wavelength results in very small diffraction angles. In this paper we describe the new setup and its characterization, including capturing static diffraction patterns of molecules in the gas phase, finding time-zero with sub-picosecond accuracy and first time-resolved diffraction experiments. The new device can achieve a temporal resolution of 100 fs root-mean-square, and sub-angstrom spatial resolution. The collimation of the beam is sufficient to measure the diffraction pattern, and the transverse coherence is on the order of 2 nm. Currently, the temporal resolution is limited both by the pulse duration of the electron pulse on target and by the timing jitter, while the spatial resolution is limited by the average electron beam current and the signal-to-noise ratio of the detection system. We also discuss plans for improving both the temporal resolution and the spatial resolution.
We present results on ultrafast gas electron diffraction (UGED) experiments with femtosecond resolution using the MeV electron gun at SLAC National Accelerator Laboratory. UGED is a promising method to investigate molecular dynamics in the gas phase because electron pulses can probe the structure with a high spatial resolution. Until recently, however, it was not possible for UGED to reach the relevant timescale for the motion of the nuclei during a molecular reaction. Using MeV electron pulses has allowed us to overcome the main challenges in reaching femtosecond resolution, namely delivering short electron pulses on a gas target, overcoming the effect of velocity mismatch between pump laser pulses and the probe electron pulses, and maintaining a low timing jitter. At electron kinetic energies above 3 MeV, the velocity mismatch between laser and electron pulses becomes negligible. The relativistic electrons are also less susceptible to temporal broadening due to the Coulomb force. One of the challenges of diffraction with relativistic electrons is that the small de Broglie wavelength results in very small diffraction angles. In this paper we describe the new setup and its characterization, including capturing static diffraction patterns of molecules in the gas phase, finding time-zero with sub-picosecond accuracy and first time-resolved diffraction experiments. The new device can achieve a temporal resolution of 100 fs root-mean-square, and sub-angstrom spatial resolution. The collimation of the beam is sufficient to measure the diffraction pattern, and the transverse coherence is on the order of 2 nm. Currently, the temporal resolution is limited both by the pulse duration of the electron pulse on target and by the timing jitter, while the spatial resolution is limited by the average electron beam current and the signal-to-noise ratio of the detection system. We also discuss plans for improving both the temporal resolution and the spatial resolution.
Observing the motion of the nuclear wave packets during a molecular reaction, in both space and time, is crucial for understanding and controlling the outcome of photoinduced chemical reactions. We have imaged the motion of a vibrational wave packet in isolated iodine molecules using ultrafast electron diffraction with relativistic electrons. The time-varying interatomic distance was measured with a precision 0.07 angstrom and temporal resolution of 230 fs full width at half maximum. The method is not only sensitive to the position but also the shape of the nuclear wave packet.