Fritz-Haber-Institut der Max-Planck-Gesellschaft  Department of Physical Chemistry
· Home · News · Research · Publications · Events · People · About us ·

Recent Developments in the
Department of Physical Chemistry (2009)

Director: Martin Wolf

click here to get this text as pdf (4.1 Mb)

1. General
1.1 Current activities at the FHI and FU Berlin
1.2 Planning of the Department of Physical Chemistry 
2. Projects
2.1 Ultrafast Dynamics in Solids and at Interfaces
2.1.1 Dynamics of Interfacial Electron Transfer
2.1.2 Photoinduced Dynamics in Correlated Materials
2.1.3   Ultrafast Optical Spectroscopy of Low-energy Excitations 
2.2 Molecular Processes at Surfaces
2.2.1 Nanoscience with Functional Molecules
2.2.2 Tip Enhanced Raman Spectroscopy
2.2.3   Photoinduced Surface Reactions and Vibrational Spectroscopys 
2.3 Complex Dynamics
2.3.1 Spatiotemporal Self-Organization
2.3.2   Complex Systems 




Activity reports

2011 · 2009

2007 · 2005 · 2003

1. General

Self-organization is a wide-spread phenomenon in nature and manifests itself in that an ordered state is formed that is far from the thermodynamic equilibrium and often shows complex spatio-temporal dynamics. The infinite diversity of patterns in physical, chemical and biological systems can be classified with modern concepts of nonlinear dynamics and a kind of soft-control of their spatio-temporal evolvement.

1.1 Current activities at the FHI and FU Berlin

Due to the renovation of the historical Building A the members of the department are currently located at several sites of the FHI campus and at the FU Berlin. The departments of Inorganic Chemistry, Molecular Physics and Chemical Physics have generously provided space to relocate existing equipment and to setup new experiments in the lab area dedicated to the future Free Electron Laser at the FHI. With the completion of Building A in 2010 the projects from the FU Berlin will be transferred to the FHI. Currently the planning for a new building with special infrastructure for laser experiments is performed; the construction of this building shall be finished until the end of 2011.

At the FHI a new research group is currently set up by Julia Stähler, which will focus on ultrafast electron dynamics and correlation effects at interfaces using time- and angle-resolved photoemission spectroscopy. Furthermore, three groups are active, which have started their research under the former director, Prof. G. Ertl. The groups of Markus Eiswirth and Alexander Mikhailov study complex dynamics in chemical systems, in particular, nonequilibrium pattern formation in electrochemical systems and problems of nanobiology and active soft matter. The technique of tip-enhanced Raman spectroscopy is developed by the group of Bruno Pettinger to study individual molecules and nanostructures under UHV conditions.

The research group of Martin Wolf is still located at the FU Berlin and focuses on electron dynamics in correlated materials and adsorbate overlayers, molecular switches at surfaces, THz time-domain and vibrational spectroscopy as well as photoinduced surface reactions. In the last two years several younger scientists received offers for university professorships: Luca Perfetti and Uwe Bovensiepen accepted offers from the Ecole Polytechnique Palaiseau and University of Essen-Dusiburg, respectively. Also Petra Tegeder will continue her research on molecular switches on a W2-position at the FU Berlin.

In this context only some of the current activities at the FU will be transferred to the FHI: The scanning tunneling microscopy group, which is headed by Leonhard Grill, will focus on functional molecules on surfaces and their controlled manipulation by tunneling electrons, electric fields, chemical forces and light. Also the growth of controlled architectures from weakly bound supramolecular and covalently bound macromolecular nanostructures will be studied. The activities on photoinduced surface reactions (headed by Christian Frischkorn) will be developed further aiming towards control of reactions by light. Tobias Kampfrath (currently at AMOLF) will join the department in March 2010 and will set up a new group to study the dynamics of solid state materials by time-resolved optical spectroscopy and intense THz fields. In addition, several other research groups may be created in the next years. The final organizational structure of the department will consist of approximately ten small research groups headed by a junior or staff scientist.

1.2 Planning of the Department of Physical Chemistry

The research of the Department of Physical Chemistry will be focussed on the dynamics of elementary processes at surfaces and interfaces and in solids. A key goal is to develop a microscopic understanding of electronic interactions and correlations as well as of photoinduced reactions and molecular processes at a single molecule level. On the one hand the dynamics of such processes will be studied by ultrafast spectroscopy directly on the relevant time scales. Hereby a broad spectrum of laser-based techniques allows analyzing the dynamics of electron transfer processes and vibrational degrees of freedom at adsorbate covered surfaces, the dynamics of electronic excitations and scattering processes in solids, and the mechanisms of optically induced phase changes in highly correlated materials. On the other hand molecular processes at surfaces will be studied on a single molecule level and employing various schemes of optical excitations. This includes also studies of photo-induced surface reactions. Scanning probe microscopy (in part combined with optical excitation) allows imaging, manipulation and spectroscopy as well as inducing and probing of chemical processes of individual molecules. For the preparation and characterization of these systems various surface science methods are employed. Further activities of the department in the field of complex dynamics of chemical systems focus on problems of nanobiology and electrochemistry. Here, theoretical studies of reactive soft matter, individual molecular machines and networks of interacting molecular machines are performed, complemented by the studies of nonequilibrium pattern formation in electrochemical systems and general aspects of chemical kinetics.

In the next few years research at the Department of Physical Chemistry will be structured into three areas:

   •   Ultrafast Dynamics in Solids and at Interfaces
   •   Molecular Processes at Surfaces
   •   Complex Dynamics

In the following, some key goals and general research directions of these areas are discussed. A more detailed account of the current projects and their future planning is given in part 2.1 - 2.3.

Ultrafast Dynamics in Solids and at Interfaces: Research in this area aims towards the understanding of electronic correlations, interactions and scattering processes in solids and at interfaces and will be pursued in several directions:
(i) Some of the most fascinating problems in solid state physics like superconductivity or metal-insulator transitions arise from electron correlation effects and coupling between electronic and other degrees of freedom of a material (e.g. lattice, orbital or spin excitations). Such electron correlation effects are particularly strong in low-dimensional systems like layered materials or quasi one-dimensional systems. Direct time-resolved studies of the ultrafast dynamics of such highly correlated materials open a new dimension, which yields information complementary to conventional frequency domain spectroscopy or transport measurements. For example, photoexcitation can drive a material into different regions of a phase diagram, which are not accessible under equilibrium conditions. Furthermore, the coupling strength of collective modes to different states of the electronic band structure as well as their damping can be studied directly in the time domain. In particular, the dynamics of the opening and closing of electronic gaps at the Fermi level provides fundamental information about the mechanisms of photoinduced phase transitions.
Studies of the ultrafast dynamics of low-dimensional correlated systems will therefore be a key topic of research of the department. As time-resolved photo¬emission spectroscopy provides the most direct access to electronic structure, several experimental setups will be developed including high-harmonic generation (HHG) of VUV laser pulses at high repetition rates (10-100 kHz). The latter approach opens also the possibility to extend such studies into the attosecond time domain. These time-resolved studies will be complemented by high-resolution photoemission experiments at the corresponding photon energies using laboratory or synchrotron light sources.

(ii) Another important aspect in such studies of non-equilibrium dynamics of solids is how selectively the system is excited and brought out of equilibrium. As many fundamental excitation energies in solids (e.g. phonons, magnons) are in the range of several meV to a few 100 meV, excitation and optical spectroscopy of solids in the (far to near) infrared regime can most directly address such low-energy transitions. For example, resonant pumping of IR-active phonon modes provides a selective way of excitation, which is fundamentally different from the response to a hot carrier distribution created by absorption of visible or UV photons. So far most time-resolved studies of solids have been restricted to the latter regime, i.e. to excitation energies much larger that the fundamentals gaps or collective modes. A key goal is therefore, to develop schemes to selectively excite low frequency modes using the electrical (or magnetic) field of intense THz or IR pulses and to analyze the response of the material with appropriate spectroscopic (THz, optical or photoemission) techniques. These studies will be complemented by THz absorption and emission spectroscopy (see 2.1.3).

(iii) Electron dynamics at surfaces and at interfaces will be another direction of research. Interfaces (like molecule-metal or molecule-oxide) allow studying fundamental processes like interfacial electron transfer, carrier localization or electron solvation processes, which are of relevance for surface chemistry, organic photovoltaic and electrochemistry. The well established technique of time- and angle-resolved two-photon-photoemission (2PPE) will be developed further to obtain state-of-the art (<20 fs) time resolution with tuneable laser pulses combined with parallel momentum detection. Furthermore fundamental studies of the photoemission process will be performed, for example, by using phase-stabilized laser pulses to study the role of the carrier-envelope phase in nonlinear photoemission at surfaces.

(iv) Further activities in this area will depend on the future hiring of research group leaders. For example, time-resolved nonlinear optics could be applied as a highly interface- and symmetry-sensitive spectroscopy to study carrier or spin dynamics at interfaces of materials with anisotropic electronic or antiferromagnetic properties. Adding spin-resolved detection to time-resolved photoemission spectroscopy would provide access to the dynamics of spin-orbit interactions and spin-dependent scattering processes at surfaces.

Molecular Processes at Surfaces: The second major research area of the department is devoted to studies of elementary processes of molecules at surfaces using scanning probe techniques as well as to studies of photoinduced surface reactions. One of the long-term goals is to combine these approaches to study light induced processes on individual nanostructures. A further goal is to induce and control surface reactions by light via appropriate excitation schemes.

(i) Scanning tunneling microscopy at low temperatures opens fascinating possibilities for the study of molecular processes as it allows on the one hand to induce chemical processes or intramolecular changes via manipulation (by tunneling electrons, electric fields, chemical forces or light) at the single molecule level in a controlled way. On the other hand, the induced changes (and thus the molecular function) can be analyzed in detail by imaging and spectroscopic characterization before and afterwards, even with respect to the atomic scale environment. Furthermore, the ccontrolled growth of supra- and macromolecular nanostructures from functional building blocks on different surfaces is a key topic in the research activities, because it allows to assemble molecules in a specific, pre-defined architecture. Depending on the chemical structure of the molecules, various intermolecular interactions — from weak van der Waals to strong covalent forces — are used for the molecular coupling. In the next years, the activities will be extended to the growth of heterogeneous, covalently bound, nanostructures that include molecular units with a specific function. By using molecular wires as connection between the functional building blocks, the possibility of charge transfer/transport through such an architecture will be explored. Such systems could open the way towards functional molecular circuits at the atomic scale, the central issue in the field of Molecular Electronics.

(ii) Combining scanning probe microscopy with optical spectroscopy allows both topographic imaging as well as spectroscopic investigation of individual nanostructures or molecules without limitations from inhomogeneous broadening. Tip-enhanced Raman spectroscopy (TERS) exhibits such spatial resolution on the nanometer scale and allows the identification and characterization of molecular adsorbates by vibrational (Raman) spectroscopy. The current TERS activities of the department will be further developed to study individual Raman active molecules on single crystal and nanostructured surfaces prepared under UHV conditions. This will be done in cooperation with the Department of Chemical Physics (Prof. Freund). A long term goal is the combination of TERS (or local optical excitation) with low-temperature STM, which would combine the potential of atom manipulation with several means of spatial resolved optical/tunneling spectroscopy. Application of TERS to liquid-solid interfaces including electro¬chemical potential control could also be an attractive route.

(iii) Controlling chemical reactions by light is a longstanding dream of laser- chemistry, whereby, bond-selectivity competes with rapid vibrational energy redistribution (IVR). However, using amplitude- and phase-shaped femtosecond laser pulses optimum control of chemical reactions can be achieved for molecules in the gas phase or in solution. Transfer of such concepts to femtochemisty at metal surfaces is hindered by rapid dephasing and energy dissipation processes. Two routes to achieve light induced control over surface reactions will be investigated: One approach will combine optimum control of gas phase reactions in a flow reactor (to generate e.g. radicals) with surface femtochemistry on a catalyst in direct contact with the gas phase. The target of control could be either optimization of branching ratios (selectivity) or the reaction yield. Another route is based on resonan IR pumping of vibrational modes which are strongly coupled to the reaction coordinate and will emply intense femtosecond IR pulses in conjunction with visible or UV excitation of hot substrate electrons. The goal is to explore the reactivity and dynamics of vibrationally ‘hot’ adsorbates, which are resonantly heated by IR excitation. Such studies will be first performed using laser based IR sources in the lab (exciting high frequency vibrational modes), but will be later extended to experiments at the infrared free electron laser of the FHI to address also excitation of low-frequency (external) modes.

Complex Dynamics: Studies of complex dynamics in chemical systems continue a tradition of research on nonlinear dynamics and spatiotemporal pattern formation established by its former director, Prof. G. Ertl. In the next years, two main research directions will be pursued:

(i) The first aims on the understanding of complex nanoscale dynamics in biological systems and, specifically, in single-molecule protein machines. Modeling of slow conformational relaxation processes in proteins and of ligand-induced mechano-chemical motions in such macromolecules will be performed. Furthermore, complex self-organized collective dynamics in biochemical networks will be studied. This theoretical research may have important practical applications, with singe-molecule machines providing the basis of a new nanotechnology generation. Using low-temperature STM, elementary molecular interactions und self-assembly of molecular architectures in two dimensions can be also studied experimentally as described above. These links, joining experimental and theoretical investigations within the department, shall be explored.

(ii) The second activity includes experimental and theoretical investigations of spatiotemporal self-organization in electrochemical processes. Additionally, general aspects of chemical kinetics are analyzed. In the future mechanistic studies of electrochemical processes, which are relevant for applications in fuel cells and for energy conversion, shall be extended and will be studied in cooperation with the Department of Inorganic Chemistry (Prof. Schlögl).

The proposed activities of the department will depend in part on the future hiring of staff scientists and can become only fully operational after the new building is available. This will strongly enhance the cross links between different projects and allow to operate the laser infrastructure under optimum conditions.

2.   Projects
2.1   Ultrafast Dynamics in Solids and at Interfaces

Electronic correlation and interactions are at the heart of solid state physics and chemical bonding. In dissipative systems with many degrees of freedom (like solids and surfaces) time-resolved spectroscopy can provide unique information about the interaction between different subsystems (e.g. electronic and nuclear degrees of freedom), which cannot be obtained from frequency domain spectroscopy. The study of the non-equilibrium, ultrafast dynamics in solids and at interfaces allows therefore to gain fundamental insights into elementary processes like electron transfer and localization, electron-phonon coupling and energy relaxation.

2.1.1  Dynamics of Interfacial Electron Transfer

Electron dynamics at molecule-metal interfaces is of key relevance for a variety of application e.g. organic light emitting devices, nanoscale molecular electronic devices or organic solar cells. The investigation of interfacial electron dynamics requires access to energy and momentum distribution of a non-equilibrium electron population and a temporal resolution of femtoseconds, i.e. on the timescale where electronic excitations occur and relax. Femtosecond time- and angle-resolved two-photon photoelectron (2PPE) spectroscopy grants insight into the occupied and unoccupied electronic band structure at surfaces. Thereby, a first laser pulse (pump) creates a non-equilibrium electron population and a second, time-delayed pulse (probe), monitors the subsequent electron dynamics by photoexcitation of the transient electron distribution above the vacuum level.

So far, the group of Julia Stähler (and coworkers at the FU Berlin) has focussed on electron transfer (ET), localization, and relaxation dynamics at polar molecule-metal interfaces. These systems are particularly interesting, as excess electrons in polar environments such as ice or ammonia localize and stabilize (solvate) by rearrangement of the surrounding molecules. Systematic studies of the influence of the substrate’s electronic band structure, the adsorbate’s structure and morphology, solvation site, temperature, coverage, and solvent unveil the fundamental electron dynamics occurring at such interfaces: Metal electrons are photoexcited to unoccupied metal states and are injected into the adlayer via the adsorbate’s conduction band. The excited electrons localize at favorable sites and are stabilized by reorientations of the adjacent molecules. Concurrently, they decay back to the metal substrate, as it offers a continuum of unoccupied states. The process of electron solvation leads to a decreasing electronic coupling between the excess electron and the substrate states. This competition of electron back transfer and solvation facilitates the study of heterogeneous electron transfer as a function of coupling strength. For amorphous ice-metal interfaces, the electron transfer is initially dominated by the substrate’s electronic surface band structure. With increasing solvation, a transient barrier evolves at the interface that increasingly screens the electrons from the substrate and that determines the excess electron residence time [1]. 

Fig. 1. Left: Electron population traces for amorphous NH3/Cu(111) for various layer thicknesses. The residence time increases with increasing coverage. Right: The excess electron residence time rises exponentially with layer thickness due to a decreasing wave function overlap with the substrate as illustrated by the inset. The inverse range parameters 1, 2, and 3 correspond to three “snapshots” of electron solvation at early (<3 ps), intermediate (10-20 ps), and late time delays (30-70 ps) [2].

For amorphous NH3/Cu(111) interfaces, two species of solvated electrons were found, one exhibiting electron dynamics on femtosecond, the other one on picosecond timescales [2]. A similar transition between ET regimes was observed as for water ice. Using Xenon overlayers to test the influence on the electron binding energy, the solvation site (bulk versus surface) could be determined [3]. For NH3/Cu(111) interfaces, the excess electron is bound at the ammonia/vacuum interface and its distance to the metal substrate can thus be varied in a systematic way. Key results are displayed in Fig 1. The analysis of the exponential dependence of the tunneling rate on the layer thickness indicted a transient evolution of tunneling barrier (see Fig. 1 right) [2]. For thicker layers the weak coupling limit is reached, where the electron transfer is mediated by thermally activated rearrangement of the solvent. Upon crystallization, the electron dynamics slow down significantly both for D2O and NH3, as the electrons reside for minutes in the adlayer. After the ultrafast formation their energetic stabilization spans up to 17 orders of magnitude in time. Their high degree of screening is achieved by localization at orientational defects at the adsorbate-vacuum interface [4].

Currently, the group develops a new setup for 2PPE spectroscopy at the FHI that will have a significantly improved time resolution due to shorter laser pulse durations, higher pulse energies for high density excitations, and parallel detection of the electron momentum due to a hemispherical analyser equipped with a 2D-CCD camera. The regeneratively amplified fs-laser system (35 fs, 250 kHz) will pump both, a collinear and a non-collinear optical parametric amplifier (OPA/NOPA) which will provide independently tuneable light from 460-760 nm. In particular the NOPA, which delivers <20 fs laser pulses, will enable the observation of electronic transitions that have not been accessible for the previous setup. Using this improved experimental design, several new directions will become accessible:

Building on the understanding of the electron dynamics at polar molecule-metal interfaces, chemical reactions with such excess electrons can be studied. In particular, the enormous lifetime of excess electrons in crystalline solvents can be used to initiate electron-driven processes as has been already shown for the dissociation of CFCl3 on D2O/Ru(001) [5]. In addition, we plan to make use of the surface-bound excess electrons on amorphous ammonia layers on Cu(111). The coverage-dependent residence times and the transient presence of the excess electrons at the ammonia-vacuum interface may thus be utilised to trigger electron-induced reactions and study the respective timescales and cross sections.

Another exciting direction is the study the dynamics of strong correlated electron systems at interfaces, which are often related to electron localization phenomena and which can be induced by photodoping. This will be discussed in the subsequent section (2.1.2).

[1]   J. Stähler, U. Bovensiepen, M. Meyer, M. Wolf. Chem. Soc. Rev. 37 2180 (2008)
[2]   J. Stähler, M. Meyer, D. O. Kusmierek, U. Bovensiepen, M. Wolf.
J. Chem. Soc. Rev. 130 8797 (2008)
[3]   J. Stähler, M. Mehlhorn, U. Bovensiepen, M. Meyer, D. O. Kusmierek, K. Morgenstern,
M. Wolf. Phys. Rev. Lett. 98 206105 (2007)
[4]   U. Bovensiepen, C. Gahl, J. Stähler, M. Bockstedte, M. Meyer, F. Baletto, S. Scandolo,
X.-Y. Zhu, A. Rubio, M. Wolf. J. Phys. Chem. C 113 919 (2009)
[5]   M. Bertin, M. Meyer, J. Stähler, C. Gahl, M. Wolf, U. Bovensiepen.
Faraday Discussions 141 293 (2009)

2.1.2  Photoinduced Dynamics in Correlated Materials

One of the basic questions in solid state physics is to understand why a material behaves like an insulator or a metal. Systems with a half-filled band are usually expected to be metallic, however, may undergo a metal-to-insulator transition at low temperatures due to Peierls instabilities or electron correlations. This interplay between electronic and phonon degrees of freedom is of general importance for the understanding of various classes of highly correlated materials like superconductors, charge density wave (CDW) compounds or Mott insulators. Excitation with ultrashort laser pulses may induce a collaps of the respective gaps in the electronic structure via photodoping as well as excitation of collective (vibrational) modes in the material. Time- and angle-resolved photoelectron spectroscopy (trARPES) provides direct access to the dynamics of the electronic structure of such photoexcited materials. In particular both single particle excitations as well as collective modes (e.g. coherent phonons) can be analyzed via the temporal evolution of the spectral function. In contrast to 2PPE spectroscopy typically a UV/VUV laser pulse is used to probe the transient changes of the electron distribution function around the Fermi level (i.e. both the occupied and normally unoccupied electronic states).

In the last few years the photoemission group at the FU Berlin has used femtosecond trARPES to optically excite and probe two model systems, namely the Mott insulator 1T-TaS2 and the CDW compound TbTe3, to investigate the dynamics of insulator-to-metal transitions directly in the time domain. In TaS2 photoexcitation by an intense laser pulse leads to an ultrafast transition towards a gapless phase which is accompanied by periodic oscillations of the electronic states (charge density breathing mode), which is lasting for 20 ps without perturbing the insulating phase [1]. The qualitative difference between the oscillatory dynamics of the collective CDW mode and the quasi-instantaneous collapse of the electronic gap followed by a monotonic recovery of the electronic gap prove that TaS2 is indeed a Mott insulator, which is also supported by DMFT calculations. Moreover it is in clear contrast with the retarded (>100fs) response which is observed for the transient melting of the CDW phase in TbTe3 [2]. Two coherently excited collective modes are observed in this system; one of them only in the CDW phase at low fluence. The latter is attributed to the CDW amplitude mode, which modulates the spectral function particular strongly at the Fermi surface (see Fig. 2). 

Fig. 2. Time-resolved photoemission spectroscopy of the CDW compound TbTe3.Top: Fermi surface of the CDW phase. The red line shows investigated region of the Brillouin zone (BZ). (b)–(d) Pump-probe spectra recorded for different momentum vectors k||. Bottom: Snapshots of the electronic structure ofTbTe3 for different time delays after photoexcitation [2]. Note the delayed collapse of the CDW gap after 100 fs and the pronounced oscillations in the spectral function at k = kF (panel d) arising from the coupling to the CDW amplitude mode.

Using trARPES allows thus to identify the role of collective vibrations in the transition and to document the highly anisotropic coupling to the electronic system in real time [2]

In the future these studies will be expanded in three directions:

(i) Studies of the photoinduced dynamics of conventional (BCS) and high Tc superconductors: So far trARPES has been applied to investigate the cuprate superconductor Bi2212, where an ultrafast (<50 fs) electron thermalization and cooling of the electronic temperature on two distinct timescales has been observed. This observation of a bottleneck in the energy flow from the electrons to the lattice suggests that only a minor subset (20%) of all phonon modes contribute to the e-ph coupling with an interaction strength which is strongly anisotropic but weak [3]. Currently these studies are extended to the novel iron pnictides superconductors. A major goal is to investigate the dynamics of superconductors under very weak excitation conditions (i.e. without completely breaking the superconducting state). This will require substantial improvements of the sensitivity of the trARPES experiment.

(ii) The group of Julia Stähler plans to drive insulator-to-metal transitions by photodoping across the interface between layered Mott insulators and a metallic substrate. Strong electron correlation effects, leading to e.g. insulator-to-metal transitions, are often related to electron localization phenomena. As the charge density is crucial for these phase transitions, they can be either statically induced by chemical doping or pressure variation or dynamically by mode-selective excitation of phonons, lattice heating, and photodoping. However, all studies known to date apply photodoping to bulk materials only. In a new approach photoexcited hot electrons in a metal substrate will be photoinjected from the metal into the upper Hubbard band and induce — if their density is sufficiently large — a collapse of the gap, i.e. the adlayer will become metallic. These experiments will unveil the role of “pure” doping with electrons (or holes) in contrast to the previous experiments with bulk materials where photodoping in fact leads to electron-hole pair (exciton) formation.

(iii) The work on anisotropic CDW compounds will be continued with a systema¬tic study of several RTe3 (R = lanthanide) compounds, some of which exhibit two CDW gaps, which open at different temperatures. Closely related is the problem of Peierls instabilities in quasi-one-dimensional metallic chains, which can be epitaxially grown on semiconductor surfaces. As the current trARPES setup with 6 eV photons (and corresponding low kinetic energies of the photoelectrons) provides only limited access to the full Brillouin zone, a key development would be a VUV source which operates at 9 eV photon energy. First steps in this direction using high-harmonic-generation (HHG) are currently under way.

[1]   L. Perfetti, P. A. Loukakos, M. Lisowski, U. Bovensiepen, M. Wolf, H. Berger, S. Biermann,
and A. Georges, New J. Phys. 10, 053019 (2008)
[2]   F. Schmitt, P. Kirchmann, U. Bovensiepen, R. G. Moore, L. Rettig, M. Krenz, J.-H. Chu,
N. Ru, L. Perfetti, D. H. Lu, M. Wolf, I. Fisher, Z.-X. Shen, Science 321, 1649 (2008)
[3]   L. Perfetti, P. A. Loukakos, M. Lisowski, U. Bovensiepen, H. Eisaki, and M. Wolf,
Phys. Rev. Lett. 99, 197001 (2007).

2.1.3  Ultrafast optical spectroscopy of low-energy excitations

Many elementary excitations in nature occur at transition energies of the order of few meV to several 10 meV, i.e. at frequencies in the terahertz (THz) regime. Examples are quasi-free electrons in plasmas of ionized gases, Cooper pairs in superconductors, excitons in semiconductors or low frequency phonons and external vibrational modes of adsorbates. Electromagnetic pulses with THz frequencies have been proven to be efficient probes of such excitations thanks to their low photon energy (4.1 meV at 1 THz), their large bandwidth (typically more than one octave), and the possibility to detect the transient electric field (rather than intensity) directly in the time domain. In addition, the short pulse duration (typically less than 1 ps) makes it possible to study ultrafast processes.

In the last few years the THz group at the FU Berlin (Tobias Kampfrath and Christian Frischkorn) has implemented THz transmission spectroscopy, whereby the carrier dynamics in a thin sample is triggered by the absorption of a femtosecond laser pulse and, after some delay, probed by a broadband THz pulse. Recently, the mechanism of the far-infrared absorption of carbon-nanotube films has been identified to originate from electronic transitions over the band gap of nanotubes with a small electronic gap of ~10 meV [1]. This finding is important for the understanding of the THz transmittance as a sensitive probe of the doping level of nanotubes. Based on the same pump-probe technique, a new scheme to measure the electronic heat capacity of the high-temperature superconductor BSCCO has been developed. This approach relies on the fact that the energy absorbed from the incident pump laser pulse is first deposited in the electronic system of the solid and allows highly selective heating of the electrons. Currently these studies are extended to various doping levels (critical temperature) in BSCCO. Furthermore, the temperature dependence of optical conductivity of the conducting polymer PEDOT:PSS used in organic photovoltaics has been investigated.

On the other hand THz radiation generated by a laser-excited sample can serve as a probe of the sample dynamics. In THz emission experiments with ferromagnetic Fe films the emitted THz radiation was found to be extremely sensitive to the cap material (interface properties) of the Fe film. A detailed analysis shows that the THz radiation reflects the motion of spin-polarized electrons parallel to the two interfaces of the Fe film. This information might be highly relevant for spintronic devices based on the interface-dominated giant magneto-resistance effect.

Future work at the FHI using THz spectroscopy will use this experimental and theoretical know-how to investigate important materials such as high-temperature superconductors, organic materials for photovoltaics, ferroelectrics, and magnetically ordered systems. The experimental setups will be improved to further increase the signal-to-noise ratio and to make reflection measurements possible.

Instead of using THz radiation as a probe of electrons, intense THz transients of order MV/cm have been used to coherently control the state of excitons in Cu2O [2]. Key results, which demonstrate coherent population transfer from the 1s to the 2p orbitals, are shown in Fig. 3. These results encourage the use of shaped THz pulses to create excitons in any coherent state on demand. From a general perspective this work demonstrates the possibility to employ THz not only as probe of the carrier dynamics, but to resonantly induce non-linear processes in materials (e.g. Rabi cycles) in the THz regime, which can be directly monitored in the time domain via ultra-broadband electro-optic sampling. 

Fig. 3. Left: Scheme for coherent control of a bound exciton. A cold gas of excitons in Cu2O is generated by a fs laser pulse and subsequently an intense THz electromagnetic wave (red arrow) resonantly drives the transition from the 1s exciton ground state to the 2p first excited state. Right: (a) Real-time profile of the THz pump pulse (red) and measured reemitted THz fields (blue) for different driving field (0.065–0.5MV/cm). The maximum of the reemitted field shifts to earlier times even develops an oscillatory structure. (b) First-principles microscopic theory, showing for trace (v) a coherent population transfer with an efficiency of up to 80%, followed by two Rabi cycles [2].

In the future this approach will therefore be extended to drive basically various low-energy excitation of a physical system into a nonlinear regime. Highly relevant and fascinating examples are low-energy molecular vibrations or spin waves. Several strategies to generate and optimize intense THz pulses will be investigated. Furthermore we will explore in how far light-management principles from the field of nano-photonics can be implemented into the field of non-linear THz spectroscopy.

[1]   T. Kampfrath, K. von Volkmann, C. M. Aguirre, P. Desjardins, R. Martel, M. Krenz,
C. Frischkorn, M. Wolf, and L. Perfetti, Phys. Rev. Lett. 101, p. 267403 (2008).
[2]   S. Leinß, T. Kampfrath, K. v.Volkmann, M. Wolf, J. T. Steiner, M. Kira, S. W. Koch,
A. Leitenstorfer, and R. Huber, Phys. Rev. Lett. 101, p. 246401 (2008).

2.2   Molecular processes at surfaces

Chemical reactions at surfaces occur via a sequence of elementary steps, which are controlled by interactions on a molecular level. The study of molecular processes at surfaces on a single molecule level can therefore yield important contributions for a microscopic understanding of surface reactions. Inducing surface reactions by light opens a different route compared to thermal activation providing insight into the role of electronically and vibrationally excited states and the dynamics of energy flow between the various degrees freedom of the system. Studies of molecular processes at surfaces are performed by several groups which employ complementary spectroscopic techniques with high spatial or temporal resolution. 

2.2.1  Nanoscience with functional molecules

The group of Leonhard Grill investigates functional molecules on metal surfaces or on thin insulating films, which allow to decouple the molecules electronically from the substrate. By using scanning tunneling microscopy (STM) at low temperatures below 10 K, single molecules are imaged with sub-molecular resolution and characterized spectroscopically. Furthermore, the STM tip is used for manipulation by using the interatomic forces between tip and molecule, the tunneling electrons or the strong electric field in the junction. Such manipulation experiments allow to induce intramolecular conformational changes or to dislocate molecules or atoms with atomic scale precision in a controlled way.

The research of the group is focussed on the understanding of chemical processes at the single molecule level, molecular “nanomachines” with specific mechanical or electrical functio¬nalities, for instance molecular wheels or switches, respectively, and growth processes of molecular structures and architectures on surfaces. The latter is divided into supramolecular structures with rather weak intermolecular interactions and the bottom-up construction of (covalently bound) molecular nanostructures, the so-called “on-surface synthesis”.

One important class of molecules are molecular switches, which are molecules that exhibit at least two stable states (conformations) with characteristic properties. While these molecules are well studied in solution, the knowledge about the switching processes on a surface is still very limited. Based on the successful switching of the so-called TBA molecules in our group, we have systematically changed the substrate and added side groups to these molecules. The use of carboxylic acid groups leads to intermolecular hydrogen bonds and characteristic rosette assemblies in which a discrete number of six molecules is connected in a well defined structure. Moreover, the attachment of a methoxy group causes the formation of different close-packed structures on the surface. Interestingly, molecules are only able to switch in some of these structures and only in particular adsorption sites, leading to periodic switching. These results revealed for the first time the importance of the atomic-scale environment, i.e. surrounding molecules and surface atoms, on the switching capability of the molecules. 

Fig. 4. STM images of a single molecule before (left) and of a molecular chain (right) after the formation of intermolecular covalent bonds by “on-surface-synthesis”. The chemical structures of the initial building block and the chain are indicated [1].

The central research topic of the group of Leonhard Grill is the polymerization of molecules directly on the surface. This so-called “on-surface-synthesis”, which has been developed in the group in the last years, allows the formation of covalent bonds between molecular building blocks on the surface (see Fig. 4). As the dimensions and shape of the resulting macromolecular networks directly reflect the chemical structure of the initial building blocks, we could form different topologies as dimers, chains (see Fig. 4) and networks on gold. In the last year, this method has been extended to another molecule, terfluorene, and the formation of homogeneous conjugated polymers with lengths of more than 100 nm could be demonstrated. Moreover, the electronic transport has been measured through single molecular wires by pulling them up from a Au(111) surface with the STM tip and thus continuously changing their length up to more than 20 nanometers [2]. This particular manipulation experiment allows the determination of the conductance as a function of the molecular wire length in the junction. The conductance curves show not only an exponential decay, but also characteristic oscillations, as one molecular unit after another is detached from the surface during stretching.

[1]   L. Grill, M. Dyer, L. Lafferentz, M. Persson, M. V. Peters, S. Hecht,
Nature Nanotechn. 2, 687 (2007)
[2]   L. Lafferentz, F. Ample, H. Yu, S. Hecht, C. Joachim, L. Grill, Science 323, 1193 (2009)

2.2.2  Tip enhanced Raman spectroscopy

There is a growing demand for techniques to topographically and chemically investigate inter¬faces on a nanometer scale. Tip-enhanced Raman spectroscopy (TERS) has proved to be a powerful tool in this respect as its scanning probe component exhibits spatial resolution on the nanometer scale and its spectroscopic component permits the identification and characteriza¬tion of adsorbates. This approach has a great potential for application in nanoscience as it provides vibrational spectroscopy with very high sensitivity and nanometer resolution.

The group of Bruno Pettinger has transferred the TERS approach to UHV by using a unique concept that places (i) a high numerical aperture parabolic mirror adjustable in between the STM scanner and the sample and (ii) all other necessary optics on a common platform with the STM. In 2007 first promising results have been obtained [1]. However, the first version of this instrument permitted only the proof of principle. Therefore, the current task is to improve the instrument along various lines in order to make UHV-TERS applicable for advanced UHV studies. For example, the optical alignment could not always be maintained, in particular after pumping down and baking out. Consequently, piezo-driven mirrors have been installed permitting a re-alignment of the optical path inside the UHV chamber. In connection with renovation of building A and the movement of the instrument to a laboratory in the Department of Chemical Physics some of the improvements had to be postponed. A more elaborated task, which is currently under progress, is to add a preparation chamber to the system which allows the preparation and characterization of sample and tip under UHV conditions and the transfer to the TERS/STM unit for vibrational and structural investigations. This includes the preparation of structured interfaces such as nanoclusters on oxide films on a support, a topic that is pursued in cooperation with the Department of Chemical Physics. Currently new experiments are in progress. These include investigations with two-analyte systems in order to determine the spectroscopic properties of individual adsorbates. The measurements will be extend also towards optically non-resonant molecules at single crystal surfaces as well as at to individual metal nanoclusters on oxide surfaces. 

Fig. 5. Scheme for tip-enhanced Raman spectroscopy (ambient conditions).
A: Side-illumination approach
B: illuminated STM tip at various distances z. Mini- mum distance: d ~ 1 nm.
TER and background intensities for varying distance parameter Rs+ z and different retraction speeds. Rs is the effective radius of the tip apex (30 nm), z is the gap width (distance of STM tip to the surface). Inset: SEM image of the STM tip.

The group has also developed different setups for TERS and performed mechanistic investigations. One of the crucial properties of TERS is its tip-sample distance (z) dependence (see Fig. 5). The intensities of the Raman lines and the broad TERS background, decay rapidly with increasing tip-sample distance z, which is nearly complete within 10 nm withdrawal of the STM tip in the z direction. A ((Rs + d)/(Rs + z))10 dependence has been derived from a simple near-field model, where Rs is the tip radius, d is the minimum distance (~1 nm) [2]. In addition, the maximum of the broad Lorentzian-shaped TER background is substantially blue shifted in energy with z. This effect is ascribed to a corresponding blue shift of the energies of localized plasmon modes upon tip retraction.

Another TER study, highlighting the analytical power of TERS, concerns cobalt tetraphenyl-porphyrin (CoTPP) adlayers formed on a Au(111) single crystal [3]. The Raman vibrational fingerprints collected from the nm-sized near-field region just below the STM tip can be correlated with the adsorbate structures seen in the STM images. The TER features of the disordered phase is assigned to CoTPP complexes with CO and/or NO axial ligands, whereas the ordered phase does not show any indication of additional axial complexation of CoTPP.

[1]   J. Steidner and B. Pettinger, Phys. Rev. Lett. 100, 236101 (2008).
[2]   B. Pettinger, K. F. Domke, D. Zhang, G. Piccardi and R. Schuster,
Surf. Sci. 603, 1335 (2009).
[3]   K. F. Domke and B. Pettinge,
ChemPhysChem 10, 1794 (2009).

2.2.3  Photoinduced Surface Reactions and Vibrational Spectroscopy

Surface femtochemistry is initiated by ultrafast laser excitation of an adsorbate-covered metal surface, whereby the non-adiabatic coupling between transiently excited metal electrons and adsorbate vibrational degrees of freedom can efficiently mediate chemical reactions. Due to the ultrafast response of the metal electrons this process can serve as a fast “trigger” of surface reactions. On the other hand, vibrational resonant sum frequency generation (SFG) spectroscopy provides a surface sensitive tool to probe chemical species, with the potential to time-resolve such ultrafast reactive processes.

The group of Christian Frischkorn has studied the mechanism of several association and desorption reactions (H+HH2; C+OCO; H2O desorption) on Ru(0001) induced by excitation with intense fs-laser pulses [1, 2]. It could be demonstrated that the dynamics of the reaction products are determined predominantly by the ground state potential energy surface although the energy is brought into the system via electronic transitions to excited states. For future experiments, the group has setup a pulse shaper and plans to use temporally shaped laser pulses to control branching ratios of femtosecond laserinduced surface reactions. Of particular interest is the coupling of such surface reactions with gas phase species (e.g. radicals), which are generated by photodissociation with appropriately shaped laser pulses. These experiments will be combined with mass spectrometry and SFG spectroscopy to obtain insights into transient reactions products.

In a collaboration with the department of Chemical Physics (within the cluster of excellence UniCat), the group is currently investigating the UV-induced photochemistry of N2O and CH4 on MgO surfaces to elucidate the role excitons and color centers in this photoreaction. In the future these studies will be complemented by the surface photochemistry for the CHx radicals in conjunction with femtosecond laser excitation of hot substrate electrons/phonons and vibrational IR-pumping of the C-H stretch bond. 

Fig. 6. SFG spectroscopy of CO stretch vibrations on a 13 ML ferromagnetic Ni(001) film for two magnetization directions (S+ and S), which exhibit pronounced changes upon magnetization reversal. The lower panel shows the magnetic contrast of the SFG intensities S+ and S.

The group has also used IR-broadband SFG spectroscopy as a sensitive probe of the coupling between spin and nuclear degrees of freedom at adsorbate-ferromagnetic interfaces, namely CO molecules adsorbed on a thin Ni film. The SFG spectra in Fig. 6 show a significant magnetic contrast at the resonances of the CO stretch vibration. In particular, the magnetic contrast at the resonance of on-top CO species exhibits a temperature dependence which very different from the temperature dependence of the bulk magnetization of the nickel film as observed by magneto-optical Kerr effect and the non-resonant SFG signal. Spin-resolved DFT calculations where performed in collaboration with Theory department (Prof. M. Scheffler). The observed temperature dependence may be attributed to atom and symmetry resolved magnetization densities and their changes under the influence of the coupling of the CO stretch vibration with thermally exited external modes like the frustrated translation and rotation of the CO molecules.

Currently, SFG spectroscopy is also employed to study the vibrational signature of photoinjected electrons in crystalline D2O ice layers on a Ru(001) surface. A ‘giant’ increase of the SFG signal is observed (enhancement factor 103–104) at a specific resonance energy, which may be attributed to highly polarizable electron-water complexes formed after relaxation of the photoexcited excess electrons.

[1]   C. Frischkorn, J. Phys.: Condens. Matter 20, 313002 (2008)
[2]   S. Wagner, New J. Phys. 10, 125031 (2008)

2.3   Complex Dynamics

Studies of complex dynamics in chemical systems have been performed in two groups which build on the tradition of research on nonlinear dynamics and spatiotemporal pattern formation established in this department by its former director, Prof. G. Ertl. 

2.3.1  Spatiotemporal Self-Organization

The group of Markus Eiswirth has combined theoretical studies with experimental investigations. The attention has been focused on problems of nonequilibrium pattern formation in electrochemical systems. Additionally, general aspects of chemical kinetics and statistical descriptions have been analyzed.

Theoretical investigations of pattern formation using reaction-migration equations were extended to partially insulated ring electrodes; these studies reproduced among other behaviors the saltatory conduction resulting from a pattern induced by the insulated part (so that the mechanism is quite different from classical saltatory conduction in nerve axons). Furthermore, mechanistic studies were carried out in cooperation with H. Varela (Sao Carlos, Brazil) on the oxidation of alcohols on Pt. A mechanism explaining kinetic instabilities in methanol oxidation was developed, it was analyzed using stoichiometric networks, algebraic geometry and numerical simulations.

Experimental studies of spatiotemporal self-organisation in electrochemical systems included a systematic survey of patterns at low conductivity in the electro-oxidation of formic acid on a Pt ring as well as first results with geometries exhibiting close working and counter electrode, which were carried with the same reaction using a Pt ribbon. An understanding of the latter geometry is crucial for modeling industrial applications (e.g. fuel cells).

These mechanistic studies of electrochemical processes will be extended in the future in cooperation with the Department of Inorganic Chemistry (Prof. Robert Schlögl) and also in cooperation with the newly founded Ertl Center for Electrochemistry and Catalysis (Gwangju Institute of Science and Technology, Korea), where Markus Eiswirth has been appointed as a vice director.

Concepts from algebraic geometry (polynomial rings) can be used to analytically determine the stationary states of a chemical reaction network and solve the stability problem. This approach can often lead to complicated expressions which are difficult to analyze. It has been shown that these expressions can be simplified by forming quotient rings from the original polynomial ring in such a way that only the remainders need to be taken into account for the stability analysis. Further studies are needed in order to determine which method would work best for a given mechanism.

The order structures (majorization) can be used to characterize dynamical processes without (or before) specifying any kinetics. Earlier studies were extended to include sinks and sources via weak sub- and supermajorization. This gives a better description, e.g., of bubble size distributions during foam decay. The latter were further characterized by using the Lorenz curves and Renyi dimensions. Future applications may include gas evolution systems and the separation of fluids, but no experiments in this direction have been carried out so far. 

2.3.2  Complex Systems

The group of Alexander Mikhailov has completed its transition to new research directions. Theoretical studies of nonequilibrium pattern formation in surface chemical reactions are terminated and attention is focused instead on a class of problems related to nanobiology. Generally, the research on reactive soft matter, individual molecular machines and networks of interacting molecular machines is performed.

Weakly condensed systems, representing soft matter, are characterized by low cohesion and high structural lability. At equilibrium, they show a great variety of structural phase transitions taking place as temperature or other external parameters are changed. Chemical reactions in such systems can interfere with the phase transitions, leading to a wealth of nonequilibrium, stationary or dynamical, structures. Their special property is that such structures can be very small, with the characteristic scales extending to the nanodomain. Nonequilibrium soft matter plays a fundamental role in biological cells and may emerge as the basis of a new nanotechnology generation. Within the last two years, theoretical investigations of nonequilibrium Langmuir monolayers, biomembranes with active protein inclusions and thin liquid layers with floating molecular machines have been performed in the group. First results of such investigations are summarized in a review article of A. Mikhailov and G. Ertl which has appeared in 2009 in the special issue of ChemPhysChem.

At the level of single macromolecules, protein machines provide a spectacular example of nonequilibrium soft matter. In response to the energy supplied to them in the chemical form with individual ATP molecules, such proteins perform ordered internal mechanical motions, cyclically changing their spatial conformations. In molecular motors, such internal motions are used to generate mechanical work. In enzymes, they allow to facilitate catalytic events by bringing together the reacting molecules attached to a protein and by implementing a conformation optimal for the catalytic conversion. These important conformational motions are slow and typically have characteristic timescales of the order of tens of milliseconds. Therefore, they remain well beyond the limit of the modern all-atom molecular dynamics simulations, where only microsecond processes could be so far traced. Thus, coarse-grained descriptions are needed. The use of elastic-network models of proteins allows to speed up computer simulations and to follow slow ligand-induced conformational motions in protein machines. 

Fig. 7. Two snapshots from a simulation of the molecular motor HCV helicase interacting with DNA. The upper protein domains (red, blue) actively translocate along a strand, while the lower domain separates the two DNA strands by repeatedly breaking the base pairs.

Analyzing slow conformational relaxation motions in classical motor proteins (Myosin, Kinesin, F1-ATPase), it was found that they proceed along well-defined trajectories, stable against noise and external perturbations, - which is essential for the robust machine operation. As the next step, dynamical elastic-network modeling of whole operation cycles of the molecular motor HCV helicase, splitting the double DNA, has been for the first time performed (see Fig. 7). These investigations are carried out in collaboration with the Osaka University (Prof. T. Yanagida) and the Toronto University (Prof. R. Kapral).

In a cell, protein machines operate in ensembles characterized by complex networks of molecular interactions. Their collective operation resembles that of a factory with many interwoven assembly lines. This factory is however self-organized and, moreover, it can robustly function despite the high noise level and frequent structural perturbations caused by biological mutations. Understanding principal mechanisms of self-organization, control and evolution in dynamical networks of interacting machines is important not only in cell biology, but also for the design of future artificial productions systems which may involve a large number of active nanodevices.

The group participates in an international project devoted to the studies of self-organizing machine networks and collaborates within this project with the groups from the universities of Arizona (Prof. D. Armbruster), Sapporo (Prof. Y. Nishiura), Kyoto (Prof. Y. Kuramoto), ETH Zurich (Prof. D. Helbing) and the Max Planck Institute for Molecular Genetics (Prof. M. Vingron). Investigations are focused on dynamical self-organization in networks of interacting cyclical automata, providing simplified descriptions of real machine networks. Moreover, evolutionary optimization methods are employed to design, in computer simulations, network architectures with the properties of self-correction against errors, allowing such network-based systems to operate at a high level of noise and structural damage.

The group of A. Mikhailov is also involved in an initiative to establish the Berlin Center for Studies of Complex Systems which aims to promote international scientific exchanges and collaborations in this research field.

GO TO    
 ·  Top of page  ·  Site map  ·