Diffusion Time Constants Research Articles

Analysis of the normal transition event of the LHD helical coils

Normal transitions and a subsequent quench were experienced with the pool-cooled helical coils of the Large Helical Device (LHD) during its excitation test. Although the initiated normal zone once started to recover, a disruptive transverse propagation followed and triggered an emergency discharging program. The cryogenic stability of the composite-type superconductor has been studied by sample experiments as well as by numerical calculations. Due to the rather long magnetic diffusion time constant in the pure Al stabilizer, transient stability of the conductor seems to play an important role for driving finite propagation of a normal zone. The cause of the final quench is also discussed from the viewpoint of cooling deterioration due to a possible accumulation of He bubbles.

Solvation in highly nonideal solutions: A study of aqueous 1-propanol using the coumarin 153 probe

We have investigated the anomalous behavior of aqueous 1-propanol binary solutions using a typical fluorescence probe molecule, coumarin 153. We present data on the fluorescence lifetimes, fluorescence anisotropies, and solvent reorganization dynamics, as well as the steady-state absorption and emission spectra of coumarin 153 in the binary solutions. The rotational diffusion and solvation time constants depend strongly on the content of 1-propanol, especially at low 1-propanol mole fractions. Spectroscopic results presented here are consistent with prior light scattering [G. H. Großmann and K. H. Ebert, Ber. Bunsenges. Phys. Chem. 85, 1026 (1981)], small angle x-ray scattering [H. Hayashi, K. Nishikawa, and T. Iijima, J. Phys. Chem. 94, 8334 (1990)], and dielectric relaxation [S. Mashimo, T. Umehara, and H. Redlin, J. Chem. Phys. 95, 6257 (1991)] data. The anomalous dynamics features likely arise from the effect of the preferential solvation due to the 1-propanol clustering.

Turbulence Decay in Stopped Flow Experiments by Measurements of Electric Dichroism

Various macromolecules of widely different chemistry and structure show unusual rotational diffusion time constants in stopped flow experiments. These effects have been characterized by measurements of electric dichroism in a combined stopped-flow−electric-field-jump instrument. DNA restriction fragments in the range of chain lengths up to 179 base pairs do not show anything unusual yet: the dichroism decay times correspond to the standard values reflecting unperturbed rotational diffusion. For fragments with more than 200 base pairs, the dichroism measured directly after stop of the flow revealed decay time constants clearly smaller than the corresponding values found without flow. The reduction of the time constants for overall rotational diffusion is ∼10% for a fragment with 256 bp and increases to ∼70% for DNA with 800 bp. The decay time constants return to the usual values within a few ms after stop of flow. The limiting electric dichroism of the flow induced state is virtually identical with that of standard DNA and, thus, the local secondary structure remains in the B-form. However, the polarizability of the flow induced state is clearly reduced. These results seem to indicate a flexible state of DNA induced by shear stress, resulting from breaking of stacking interactions. The kinetics can be described by a model with parallel reactions leading to the native state with a time constant of ∼1 ms. Effects observed for single stranded poly(A) are virtually identical to those of double stranded DNA for samples of corresponding hydrodynamic dimensions. The time constants for the transition to the standard state are again ∼1 ms, although stacking reactions of poly(A) occur in the time range ≤ 1 μs. Further experiments with other macromolecules or macromolecular aggregates indicate that acceleration of rotational motion is a general phenomenon under stopped flow conditions: for example bacteriorhodopsin membrane disks and colloidal suspensions of the fibrous claylike silicate attapulgite show very high accelerations of their rotational diffusion time constants after stopped flow by factors of about 3000. The return to the usual state both for bacteriorhodopsin and for attapulgite extends over a broad time range up to ∼100 ms. Although it cannot be excluded that these effects are partly due to kinking of the structures, the similarity of the data for materials of very different structures indicates that the acceleration of rotational diffusion is mainly due to the existence of turbulence. The rotational motion of small molecules, such as 95 bp DNA, is much faster than rotational motion of corresponding volume elements of the solvent and, thus, these molecules do not indicate turbulence. The data demonstrate that turbulence decays over a broad time range: small vortices with high rates of rotational motion decay first, merge into larger ones with lower rotation rates, until finally the largest ones compatible with the dimension of the observation chamber are reflected only by particles with relatively large dimensions and slow rotational diffusion. These results demonstrate the mode of turbulence in stopped flow observation cells, the scaling of its decay, and the utility of electrooptical techniques for the characterization of turbulence.

Diffusion measurements using ATR-FTIR spectroscopy: Acetone diffusion in polypropylene?use of penetrant fluid pressure to improve sample/IRE contact

In this article, we present a simple method for improving the contact between a film sample and the internal reflection element (or crystal) when diffusion into thin polymer films is measured with attenuated total reflectance (ATR)-Fourier transform infrared spectroscopy. Intimate film/crystal contact is particularly important for making measurements on premade films and materials that cannot be solution-coated onto the crystal. This method is based on controlling the penetrant fluid pressure above a threshold value (>230 kPa) in the ATR flow cell. Measurements of acetone diffusion into a commercial polypropylene film at 300 K and varied pressures indicated that the diffusion time constant was constant at pressures above this threshold. We also monitored the absorbance of a polymer band that had no overlap with the acetone spectrum to examine whether adequate sample-film/crystal contact was reached and maintained. From these observations, we concluded that an apparently good match between the experimental data and a model calculation does not alone justify confidence in the accuracy of the calculated diffusion time constant. Additionally, the practice of using a reference band to correct the uncertainty in absorbance for bands of interest (due to imperfect sample/crystal contact) yielded inconsistent results. We also report further measurements of acetone diffusivity in a polypropylene film at temperatures ranging from 278 to 308 K that yielded an estimated activation energy for diffusion of 98 kJ/mol. © 2000 John Wiley & Sons, Inc.* J Polym Sci B: Polym Phys 38: 1773–1787, 2000

Stability and frequency tuning of thermally loaded continuous-wave AgGaS_2 optical parametric oscillators

We analyze and investigate experimentally the output stability and the frequency tuning characteristics of weakly triply resonant silver gallium sulfide (AgGaS2) optical parametric oscillators. These oscillators are subject to thermally induced bistability and passive self-frequency-locking phenomena. The robust self-frequency-locking on a single-mode pair (thermal lock) originates from the material’s ability to correct external cavity-length perturbations, which would normally cause a mode hop, by increasing or decreasing the optical path length of the cavity through the thermo-optic effect triggered by the intracavity signal–idler power fluctuations. The Fourier frequency bandwidth of this passive servo is limited to ∼1 kHz by the thermal diffusion time constant, which is proportional to the ratio of the specific heat to the thermal conductivity Cp/Kc. A thermal feedback servo gain as high as 180 is obtained, owing to the large thermal figure of merit η=(dn/dT)/Kc of AgGaS2, leading to routine passive mode-hop-free operation for more than 30 min, without the need for an external cavity-length servo. Analysis of the stability range of the thermally loaded standing-wave resonator shows that thermal lensing is less critical for shorter doubly resonant optical parametric oscillator cavities employing shorter-curvature mirrors, in agreement with experimental observations. When a doubly resonant oscillator operates near the boundary of the power stability range a self-pulsing behavior is observed on a number of axial mode pairs. This self-pulsing is found to originate from the destabilization of a self-locked cw state, and the transition from self-pulsing to the stable self-frequency-locked state is found to be controlled by the pump frequency detuning. The passive stability allows the single-parameter frequency tuning to be studied. Under pure thermal lock operation the oscillators show a tendency to resist the pump frequency and temperature tuning processes. When an external cavity-length servo is implemented continuous tuning over 850 MHz, by means of the pump frequency tuning (Δνs/Δνp≈0.66), and over 100 MHz, by means of the crystal temperature (Δνs/ΔT≈250 MHz/°C), is obtained. These tuning ranges are in good agreement with calculations based on a cold doubly resonant oscillator.

Transient stability of large aluminum stabilized superconductors

Very large current composite superconductors have been considered and adopted to use in SMES coils and fusion applications, such as the Large Helical Device (LHD). These superconductors have large cross-sectional area of high purity aluminum stabilizer to improve their stability and to enhance the overall current density. Once a normal-zone is initiated in such a composite superconductor, the current transfers to the aluminum stabilizer according to the temperature distribution. The time constant of current diffusion in the stabilizes however is very long due to the low electrical resistivity of aluminum and the large conductor size. Therefore, an excess Joule heat is generated in a small area near superconducting filaments and the temperature increases locally. In this paper, to evaluate this peculiar property, we carry out some simulations with regard to quench process in the superconductor applied to the helical coil of LHD in the National Institute for Fusion Science. The simulations, by using a newly developed computer code, are compared with the experimental results of the stability tests on the short samples of LHD conductor. Furthermore, we focus on the influence of the CuNi alloy clad adopted to the LHD conductor on the normal transition and normal-zone propagation properties.

Global Structure and Flexibility of Hairpin Ribozymes with Extended Terminal Helices

Global structure and flexibility of three different hairpin ribozyme constructs have been analyzed by measuring their electric dichroism decay in various buffers at temperatures between 2 and 30°C. The hairpin ribozyme is characterized by two independently folding domains A and B that are connected through a hinge and have to interact to enable catalysis. The analyzed constructs feature extended terminal helices 1 and 4 with 27 and 25bp, respectively, to increase the sensitivity of the molecular rotational diffusion time constants with respect to the interdomain bending angle. Constructs HP1 and HP2 cannot cleave because of a G+1A change at the 3′-side of the cleavage site; in HP1 the helices 2 and 3 that flank the hinge form a continuous double helical segment; in HP2 and HP3, a six nucleotide bulge confers flexibility to the expected bending site; HP3 is a cleavable form of HP2 with a G+1-base. For comparison, a standard RNA double helix with 72bp was included in our analysis. The dichroism decay curves of the hairpin constructs after pulses of low electric field strengths can be fitted to single exponentials τs, whereas the curves after pulses of high field strengths require two exponentials. In all cases, time constants increase with RNA concentration, indicating intermolecular interactions. Extrapolation of the τsvalues measured in standard buffer (50mM Tris (pH 7.5) and 12mM MgCl2) to zero RNA concentration provide values of 112, 93, and 73 ns for HP1, HP2 and HP3, respectively, at 30°C, indicating increasingly compact structures. The 72bp RNA reference under corresponding conditions did not show a dependence of its decay time constant on the RNA concentration nor on the field strength; its time constant is 175 ns (standard buffer, 30°C). The observation of two relaxation processes for the hairpin constructs at high field strengths indicates stretching to a more elongated state; the fast process with a time constant of the order of 50 ns is assigned to reversion of stretching, the slow process to overall rotation. The overall rotational time of the stretched state at 20°C is close to that for a completely stretched rigid state; at 30°C the experimental values are around 70% of that expected for a completely stretched rigid state, indicating flexibility and/or residual bending. Bead models were constructed to simulate dichroism decay curves. The time constants observed for the 72bp RNA are as expected for a rigid rod with a rise of 2.8Å per base-pair. Based on this rise per base-pair for models of a V and a Y-shape, we estimate average bending angles of 80(±20)° and 105 (±25)°, respectively, for the catalytically active hairpin ribozyme HP3. The energy required for stretching is of the order of the thermal energy.

Solid‐State Electrochemical Kinetics of Li‐Ion Intercalation into Li1 − xCoO2: Simultaneous Application of Electroanalytical Techniques SSCV, PITT, and EIS

The electroanalytical behavior of thin electrodes is elucidated by the simultaneous application of three electroanalytical techniques: slow‐scan‐rate cyclic voltammetry (SSCV), potentiostatic intermittent titration technique, and electrochemical impedance spectroscopy. The data were treated within the framework of a simple model expressed by a Frumkin‐type sorption isotherm. The experimental SSCV curves were well described by an equation combining such an isotherm with the Butler‐Volmer equation for slow interfacial Li‐ion transfer. The apparent attraction constant was −4.2, which is characteristic of a quasi‐equilibrium, first‐order phase transition. Impedance spectra reflected a process with the following steps: ion migration in solution, ion migration through surface films, strongly potential‐dependent charge‐transfer resistance, solid‐state diffusion, and accumulation of the intercalants into the host materials. An excellent fit was found between these spectra and an equivalent circuit, including a Voigt‐type analog ( migration through multilayer surface films and charge transfer) in series with a finite‐length Warburg‐type element ( solid‐state diffusion), and a capacitor (Li accumulation). In this paper, we compare the solid‐state diffusion time constants and the differential intercalation capacities obtained by the three electroanalytical techniques. © 1999 The Electrochemical Society. All rights reserved.

Kinetics of volatile organic compound sorption/desorption on clay minerals

Soils surrounding industrial sites or at locales where industrial chemicals are utilized, frequently become contaminated through unsuitable discharge of potentially hazardous organic compounds. The fate and transport of these chemicals must be sufficiently understood to predict detrimental environmental impacts and to develop technically and economically appropriate remedial action to minimize environmental degradation. Improving our understanding of the processes involved in organic pollutant vapor transport is important because the gas phase is often the most mobile, and therefore most potentially hazardous phase. In order to gain a better understanding of the basic kinetic processes affecting soil adsorption/desorption of volatile organic compounds (VOC's) in the vapor phase, we conducted VOC adsorption/desorption experiments using oven-dry clay minerals. Transient, isothermal, gravimetric sorption experiments using volatile organic compounds (VOC's) acetone, benzene and toluene onto pure clay minerals obtained from Ward's Scientific (kaolinite, illite, and Ca-montmorillonite) suggest a biphasic sorption mechanism on these minerals. Experimental results indicate that hydrophobic sorption onto oven-dry clay minerals with negligible soil organic matter is controlled by rates of inter-particle Fickian diffusion mechanisms, intra-particle Fickian diffusion mechanisms, and sorption kinetics. Using an analytical solution to Fick's Second Law where sorption is partitioned into macroscopic and microscopic domains, each with unique diffusion time constants, enables precise prediction of experimental sorption observations. Correlation coefficients of 0.999 were found between the parameter optimized analytical solution and very large sets of experimental data. Macroscopic diffusion coefficients ranged from 10 −2 to 10 −4 cm 2/min, while microscopic diffusion coefficients ranged from 10 −12 to 10 −17 cm 2/min. Sorption rates suggest that significant fractions of VOC's adsorb onto surfaces other than the external mineral surfaces. The fraction of mass sorbed on the mineral surface ranged from 30% to 75% of the equilibrium sorbed mass. Equilibrium sorption mass correlates approximately to mineral specific surface area and, similarly, the fraction of `irreversible' sorption appears to be proportional to the specific surface area of the clay mineral. Finally, the experimental results suggest that sorption processes (once the sorbate has diffused to the sorbent) are not instantaneous. Due to the experimental inter-particle length scales and intra-particle diffusion processes, diffusional time constants associated with each process during sorption experiments were generally found to be within two orders of magnitude. Such close agreement would not be expected if particles larger than a few micrometers (i.e., aggregates) comprised part of the sorbing soil system. Although the experiments were conducted using oven-dry clays, it is hoped that these results offer an improved conceptual model of clay mineral sorption processes to better explain field observations of VOC persistence in the subsurface.

Drying of glassy polymer varnishes: A quartz resonator study

We report on desorption measurements on polymeric thin films coated onto quartz crystal resonators. Due to the high sensitivity of quartz crystal microbalances, the experiments can be performed on very thin films, which have small diffusion time constants even in the glassy state. When drying is performed slowly enough, diffusion equilibrium can be maintained through the whole process of desorption, including the glassy domain. From these quasi-stationary pressure ramps, we derived the solvent chemical potential as a function of polymer volume fraction μ(o). The results fit well to a model recently proposed by Leibler and Sekimoto. 1 In addition, we have derived the mutual diffusion coefficient D(o) from pressure step experiments. We observe a strong decrease of D(o) for high polymer concentrations typical of hypodiffusive systems like polymers. We investigated the drying of an industrial varnish that is a blend of 2 copolymers as well as the drying of its components separately. Both the solvent chemical potential μ(o) and the mutual diffusion coefficient D(o) of the blend interpolate between the respective quantities of the components.

Pulse Reverse Copper Electrodeposition in High Aspect Ratio Trenches and Vias

A theoretical model, based on a one‐dimensional approximation, for the filling of high‐aspect ratio trenches and vias with copper by pulse reverse electrodeposition is presented. The dependence of void size on the pulse waveform is demonstrated for submicron features. It is shown that the off‐time should be on the order of the diffusion‐time constant and that the deposition time should be smaller than the diffusion time constant. Reverse current, i.e., dissolution, is shown to play an important role in diminishing void size. Comparison with two‐dimensional calculations indicates that the theory is an adequate yet efficient means of simulating shape change.

Effects of nonlinear equilibrium on zero length column experiments

The assumption of linear equilibrium in ZLC experiments is investigated to evaluate possible errors in the determination of diffusional time constants in microporous materials. The shape of the response curves is obtained considering the Langmuir isotherm which adds an additional parameter (nonlinearity) to the modelling equations. The analytical solution for the limiting behaviour under equilibrium control conditions as well as a closed-form analytical approximation for long time asymptotic behaviour are derived. It is shown that equilibrium controlled systems yield response curves which are qualitatively similar to those obtained under conditions of kinetic control. In order to exclude the intrusion of nonlinear equilibrium control effect it is necessary to consider simultaneously experimental response curves at different flow rates. Integral response curves are also discussed with reference to experimental results available in the literature.

Eddy Current Decay and Resistivity Measurements on Longitudinally Grooved High-Purity Aluminum Bars

Abstract The eddy current decay (ECD) method for determining electrical resistivity uses a relationship between the time constant of magnetic flux diffusion and resistivity. This method is shown to be valid for determining the resistivity of solid high-purity aluminum cylinders with eight equally spaced longitudinal grooves machined into the surface. The effects of changes in groove dimensions for a 25.4-mm-diameter, superconducting magnetic energy storage (SMES) conductor stabilizer are reported. Observations of machining, heat treatment, and material inhomogeneity effects are discussed. It is found that the ECD time constant is dependent primarily on groove depth and that groove width has only a minor influence. An empirical equation relating groove dimensions with the time constant and with resistivity is presented.

Transient Anisotropy and Fragment Rotational Excitation in the Femtosecond Photodissociation of Triiodide in Solution

Transient anisotropy experiments on the triiodide photodissociation reaction in liquid solution are presented for excitation with 30-fs, 400-nm optical pulses. The decay of the spatial diatomic fragment distribution that is initially created through the process of bond fission occurs on three distinct time scales. On an ultrashort time scale well below 1 ps, an inertial contribution can clearly be distinguished. The corresponding Gaussian decay is described by correlation times around 450 fs. This ultrafast component reflects substantial rotational excitation of the diatomic product which results from impulsive photodissociation of the parent molecule. The contribution of the bending coordinate to the reaction mechanism in liquid solution is emphasized using a simple impulsive model for rotational excitation of the diatomic fragment. An intermediate time scale is characterized by an additional exponential contribution to the anisotropy decay with time constants around 2 ps and coincides with the time scale for vibrational relaxation of the diatomic fragment. On even longer times scales, rotational diffusion of the diatomic fragment governs the anisotropy decay with typical time constants of about 12 ps. A pronounced probe wavelength dependence of the rotational diffusion time constant is explained by an effective hydrodynamic volume that is detected by the probe pulse and originates from the anharmonicity of the diiodide potential. Finally, nuclear coherences at the diiodide vibrational frequency are observed in the inertial decay of the anisotropy. This novel effect is speculated to arise from a coupled motion of the system along all possible bending and stretching coordinates.

Modification of Resin-Type Adsorbents for Ethane/Ethylene Separation

Modified adsorbents, Ag+-exchanged resins, have been prepared and studied for ethane/ethylene separation by adsorption. On Ag+-exchanged Amberlyst 35 (36.5% exchange) at 25 °C and 1 atm, the equilibrium adsorbed amount for C2H4 is 1.48 mmol/g, and the equilibrium adsorption ratio for C2H4/C2H6 is 6.4. The adsorption capacity is completely restored at 100−105 °C, although small residual amounts exist after desorption at 25 °C and 75 °C. For the adsorption encompassing both physical adsorption and π-complexation with energy heterogeneity, the equilibrium data are correlated with an equilibrium isotherm equation employing two fitting parameters. The fitted results agree well with the experimental data. Furthermore, the isosteric heats of adsorption and the diffusion time constants are calculated from experimental data. Considering all adsorption characteristics, these adsorbents show potential for application employing cyclic adsorption processes.

In‐situ CIR‐FTIR study of the diffusion of supercritical hydrocarbons in zeolite L

AbstractCylindrical internal reflectance infrared spectroscopy (CIR‐FTIR) was used for the determination of the transport properties of hydrocarbons in zeolite L at high temperatures and pressures. This was the first time that the intracrystalline diffusivity of hydrocarbons in microporous media could be determined under supercritical conditions by using a dynamic technique. The study investigated the self‐diffusion of n‐heptane, the counterdiffusion of n‐hexane and 1‐hexene, and the codiffusion of n‐heptane and pyxlene in zeolite L under subcritical and supercritical conditions. The diffusion time constants obtained for the single and binary systems showed that the diffusion within the pore was reduced significantly under supercritical conditions (i.e., liquidlike behavior) as the hydrocarbon density within the pore was increased.

Charge dynamics of a methane adsorption storage system: Intraparticle diffusional effects

The charge of natural gas adsorption storage systems is studied numerically, With emphasis given to the impact on its dynamics of intraparticle diffusional resistances to mass transport. Besides adsorption kinetics and thermal effects, the simulation model takes into account both mass transport inside the adsorbent and hydrodynamics of flow through the packed bed. Numerical results are presented for change with methane of a 50 liter cylindrical reservoir, filled with hypothetical adsorbents with diffusional time constants in the range 10−3 s1D/Rp2 ≤ ∞. and with the adsorption equilibrium curve of a commercially available activated carbon with a good adsorptive storage capacity. An attempt is made to assemble the charge histories for different values ofD/Rp2 , in a single cure by using a modilied time scale.

Streaming potentials during the confined compression creep test of normal and proteoglycan-depleted cartilage.

The streaming potential response of cartilage in the confined compression creep configuration was assessed theoretically and measured experimentally in normal and proteoglycan-depleted tissue. The analytical solution, using the linear biphasic continuum model including electrokinetics and assuming homogeneous material properties, predicted that: (i) the peak streaming potential is delta V = ke x delta sigma, where ke is the electrokinetic coefficient and delta sigma is the change in compressive stress; (ii) the potential is maintained at 95 to 100% of the peak value for 0 < t < 0.10 tau, where tau is the gel diffusion time constant; and (iii) during short times, 0 < t < 0.01 tau, 90% of the peak streaming potential occurs over a region extending 23% into the tissue sample. Experimentally, adult bovine cartilage disks, 0.5 mm thick, were subjected to step changes of compressive stress. The measured changes in potential indicated a linear response for changes in stress up to 0.10 MPa. The ke of normal cartilage, estimated from the short time (0 < t < 2 sec) change in potential, was -1.65 +/- 1.25 mV/MPa. Digestion of cartilage by chondroitinase ABC resulted in an increased (less negative) ke of -0.75 +/- 0.70 mV/MPa and a 33 +/- 29% depletion of anionic glycosaminoglycan, whereas digestion with trypsin resulted in a further increase in ke to +1.74 +/- 0.95 mV/MPa and a 98 +/- 1 % depletion of glycosaminoglycan. The streaming potential measurement may be a useful addition to the widely used confined compression creep test to assess cartilage material properties.

A precise electron beam energy calibration for use in nuclear physics

An accurate calibration of the MIT-Bates electron beam energy has been achieved. The electron beam was injected into the MIT-Bates South Hall Ring and the spin-precession frequency, Ω s , was measured. From the spin-precession frequency the equilibrium energy of the ring was determined to be 370.3115±0.0021±0.0004 MeV, where the first error is statistical and the second is systematic. The energy of the beam delivered by the linac was determined to be 369.954±0.023±0.300 MeV. This provides a rigorous check on the accuracy of the facility's energy analyzing magnets. The first moment of the energy distribution was also measured. By observing the time constant of the spin diffusion with zero power in the South Hall Ring RF cavity, we obtained a value of σ γ γ = (3.7±0.5) × 10 −4 . We provide a lower limit to the polarization lifetime. At a beam energy of 370 MeV no measurable decrease in the polarization was observed up to 2 ms after injection.

Photocurrents in a metal-semiconductor-metal photodetector

Photocurrents in a metal-semiconductor-metal (MSM) photodetector have been analyzed in a one-dimensional structure using both time-dependent and steady-state continuity equations. Analytical solutions are presented for the carrier concentrations as well as for the currents, which form a valuable tool for the investigation of the detector behavior under various bias conditions. Applying these expressions to a GaAs device, we have studied the influence of carrier diffusion, recombination, and drift on the photocurrents as a function of the applied bias voltage. We also show that the switching time at low bias voltage is dominated by a voltage-independent diffusion time constant which is of particular interest when using the device as an optoelectronic sampling gate. On the other hand, carrier recombination is found to have minimal influence on DC characteristics and pulse response.