Watching Rust Dissolve:
Ultrafast Time-Resolved X-Ray Absorption Measurements of the Reductive Dissolution of Iron Oxide Nanoparticles
Watching Rust Dissolve:
Ultrafast Time-Resolved X-Ray Absorption Measurements of the Reductive Dissolution of Iron Oxide Nanoparticles
Overview
The focus of my current research is to understand interfacial electron-transfer reactions at metal oxide surfaces. To do this, I use a variety of analytical, electrochemical, and spectroscopic techniques. The goal of my current work is to develop new ultrafast time-resolved x-ray spectroscopy methods to better understand the reductive dissolution of iron oxides.
Following Fast Reactions with Pump-Probe X-ray Spectroscopy
The only way to achieve very high time resolution in optical or x-ray spectroscopy is to use a “pump-probe” setup. Here, an initial very short pulse of light initiates the reaction of interest (the “pump”), followed by a second pulse of light (the “probe”), which is used for spectroscopy. Changes in the intensity or energy of the probe beam can be used to follow the reaction progress. By using extremely short pulses of light (femtoseconds to picoseconds, or 10-15–10 -12 seconds), the dynamics of the reaction can be monitored in real time and short-lived intermediates can be directly observed. While this type of “ultrafast” spectroscopy is now relatively commonplace when used with UV, visible, or IR light, it has only begun to be exploited for use with x-ray spectroscopy or scattering. The second main emphasis of the project I am working on is to develop new tools and methods to study interfacial electron transfer dynamics on the ultrafast timescale using x-ray spectroscopy and scattering.
Experimental Approach
Using a Molecular Dye for Light-Induced Electron Transfer
To simulate an iron-reducing microbe, we use a light-sensitive small molecule bound to the surface of the particle. Ideally, such a molecule must meet several criteria:
1)It must absorb light very strongly, and at wavelengths where the particles themselves do not absorb (≥500 nm);
2)It should bind strongly and irreversibly to the surface, and have a small molecular footprint to maximize packing density;
3)The excited state of the dye must be thermodynamically capable of transferring an electron to unoccupied states in iron oxide;
4)There must be good electronic coupling between the dye and the iron oxide surface so that the electron transfer step is fast.
Time-Resolved X-Ray Spectroscopy
To study the reaction, a flowing stream of a suspension of dye-sensitized maghemite (gamma-Fe2O3) nanoparticles particles, 2-3 nm in diameter, is excited by a laser at approximately 530 nm. Then, a second beam, either visible, IR, x-ray can be used to follow the reaction progress. A schematic diagram of the experimental setup for time-resolved fluorescence x-ray spectroscopy is shown in Figure 1.
Upon photoexcitation with a laser, an excited state of the molecular dye is formed, and subsequently an electron is injected into the iron oxide particle, thereby reducing an Fe(III) center to Fe(II). During the interaction with the x-ray probe beam, Fe atoms in the nanoparticles emit x-rays, whose energy is indicative of the oxidation state of the iron. Although x-ray spectroscopy is an ideal way to monitor the oxidation state of the iron in the mineral, it probes all the Fe atoms in the particles, and so it is important to maximize ratio of the amount of dye molecules bound to the particles to the total amount of Fe. To do this, we maximize the surface area of the particles by using nanometer sized particles — thereby allowing a maximum amount of dye to bind.
![]() |
Results
Sensitization by 2,7-Dichlorofluorescein
Preliminary results using the dye 2,7-dichlorofluorescein, shown in Figure 2, are promising. The dye absorbs visible light strongly, with molecular absorptivity of ~5 x 104 M-1 cm-1. When bound to maghemite at pH 4, the absorption maximum shifts from 485 nm to 515 nm. Binding only occurs at slightly acidic pH, where the surface charge of the particle is positive but the dye is singly protonated. Compared to other dyes with similar structures, absorption isotherms show very strong absorption by 2,7-dichlorofluorescein with near monolayer coverage. Finally, in solution the dye fluoresces very strongly (thus its name), but when bound to maghemite, that emission is completely eliminated. This is a strong indicator that the excited state of the dye is efficiently quenched by an electron transfer reaction to the particles.
Figure 2: The dye 2,7-dichlorofluorescein, which is chemically bound to the iron oxide nanoparticle surface, acts as a light-activated reducing agent.
Steady-State Optical Spectroscopy
To determine if Fe atoms in the iron oxide nanoparticles are indeed being reduced to aqueous Fe(II), I designed a methodology to monitor the buildup of Fe(II) in solution during prolonged exposure to light. Cyan-colored high-power LEDs, with their low cost and high intensity at 500 nm, have proved to be an ideal illumination source. An O2-free, aqueous suspension of dye-sensitized maghemite nanoparticles is exposed for 24 hours to the LED light source, and is then tested for the presence of Fe(II) in solution using a common spectroscopic indicator. Ferrozine, in the presence of Fe(II) but not Fe(III), turns from pale yellow to deep purple. During light exposure, a significant concentration of Fe2+(aq) is evolved for dye-sensitized iron oxide nanoparticles. In control experiments, unsensitized particles are exposed to the LEDs and do not show any ferrozine color change, indicating that it is indeed the photo-excited dye which is responsible for the formation of reduced iron.
The energy of the iron absorption edge varies with the Fe oxidation state, and so spectral shifts can be correlated with a conversion of Fe(III) to Fe(II). However, all Fe atoms in the particles contribute to the observed spectrum, but only a very small minority of them are reduced to Fe(II). This requires excellent signal-to-noise ratio spectra as the change in time due to steady-state build-up of Fe(II) is very slight.
Time-Resolved X-ray Spectroscopy
Using ultrafast time-resolved x-ray absorption spectroscopy, the evolution of the reaction can be followed in real time during and after the electron injection step, following the structural and electronic rearrangements of the maghemite particles. In the first, direct picosecond-resolved measurements of its kind, the dynamics of ferrous ion formation and subsequent re-oxidation or dissolution in iron oxide have been observed (see Figure 4). These experiments, performed at the Advanced Photon Source (APS) at Argonne National Laboratory, use a 3 ps laser pulse at ~530 nm to excite the molecular sensitizer, initiating the electron transfer reaction. The Fe(III) reduction reaction is followed by monitoring the Fe K-edge at 7.12 keV as a function of time delay relative to the laser excitation pulse. We find a clear signature in the time-dependent difference signal indicating the reduction reaction occurs on a timescale of approximately 500 ps, followed by a slower reaction over several nanoseconds.
Future Work
Time-Resolved Optical Spectroscopy
It will also be important to do optical spectroscopy on the system by monitoring the dye itself. It may be possible to watch the reaction progress using transient-absorption spectroscopy to measure the rate of formation of the oxidized form of the dye (appearance of dye+ or disappearance of dye*), and possibly the return to the ground state due to a recombination reaction across the interface. The dye must undergo some reaction subsequent to electron injection to maintain electroneutrality, presumably a recombination reaction or a decomposition reaction. It may be possible to add another species in solution to slow or prevent such a reaction by reducing the dye after electron injection, regenerating the ground state, by adding a sacrificial electron donor.
Further Time-Resolved X-ray Spectroscopy
Additional time-resolved x-ray spectroscopy measurements may be help to reveal further details about the structural and electronic changes occurring during the reductive dissolution reaction. The primary challenge to overcome is the extremely high signal-to-noise ratio necessary to resolve the small Fe(II) signal over the much larger background of Fe(III). With improved data statistics, it will be possible to collect full EXAFS spectra, which would reveal a great deal more information about structural changes during the reaction.
Static X-ray Spectroscopy
We have monitored the formation of Fe(II) in a static x-ray absorption experiment using an x-ray absorption near edge structure (XANES) experiment and extended x-ray absorption fine structure (EXAFS). These experiments were performed at Stanford Synchrotron Radiation Lightsource (SSRL) at the SLAC National Accelerator Laboratory. Here, a circulating solution of sensitized maghemite was illuminated by the LEDs and continuously probed by XANES for several hours.
Figure 3: A schematic energy diagram of the possible electron transfer pathways following photoexcitation of the dye bound to the surface of maghemite nanoparticles.
page last updated 4/10
Figure 1: Scheme for performing ultrafast x-ray spectroscopy of iron reduction with the pump-probe method. A water jet, seen here in cross-section, containing iron oxide particles coated with photoactive organic molecules is formed in front of an x-ray photon detector. A short laser pulse is used to photo-inject electrons and reduce surface ferric iron sites. Spectroscopy is performed by a probing with a short x-ray pulse to investigate the sample chemistry at a well-defined time delay ∆t after the laser excitation.
Background
Reductive Dissolution of Iron in Nature
The cycling of iron in nature between its reduced and oxidized forms is a critical process in a variety of biological, chemical and geologic systems. The reduction of Fe(III) is possibly the most important chemical change that takes place in the development of anaerobic soils and sediments, and the reductive dissolution of iron-bearing minerals by microbes plays an important role in this process. Specialized bacteria transfer electrons from the oxidation of an organic substrate to an iron oxide particle, thereby reducing Fe(III) to Fe(II). Despite its importance in biogeochemistry, many questions remain about the mechanism of this electron transfer reaction. The ultimate goal of my work is to image the intermediate configurations of mineral surfaces during redox reaction to aid in the elaboration of first principles descriptions of charge transfer in iron oxides.
Iron Creek, Colorado. The brown color of the stream is indicative of the high concentration of iron minerals present in these natural waters.
To find out more, please visit the
Berkeley Nanogeoscience Center
or see our recent publication in Journal of Physical Chemistry Letters
Figure 4: XANES spectra of maghemite nanoparticles of the ground state (before the laser pulse) and the excited state (500 ps after the laser). The difference spectrum is shown along with a composite difference spectrum of mixture 2% aqueous FeCl2 and maghemite. The high degree of consistency between the two is a strong indication of the rapid formation and dissolution of Fe(II).