Many areas of research rely on the detection of radiation, in the form of single photons or particles. By measuring the photons or particles coming from an object a lot can be learned about the object under study. In some cases there is a simple need to know the number of photons coming from the source. In cases like this a simple counter, like a Geiger-Mueller survey meter, will suffice. In other cases one want to know the spectral distribution of the photons coming from the source. In cases like that a spectrometer is needed that can distinguish between photons with different energies, like a diffraction or transmission grating. The work presented in this thesis focused on the development of a new generation broad band spectrometer that has a high energy resolving power, combined with a high absorption efficiency for photon energies above 10 keV and up to 500 keV. The spectrometers we are developing are based on low-temperature sensors, like superconducting tunnel junctions or transition edge sensors, that are coupled to bulk absorber crystals. We use the low-temperature sensors because they can offer a significant improvement in energy resolving power, compared to conventional spectrometers. We couple the low-temperature sensors …

The Resource Conservation and Recovery Act (RCRA) of 1976 and its subsequent amendments have had a dramatic impact on hazardous waste management for business and industry. The complexity of this law and the penalties for noncompliance have made it one of the most challenging regulatory programs undertaken by the Environmental Protection Agency (EPA). The fundamentals of RCRA include ``cradle to grave`` management of hazardous waste, covering generators, transporters, and treatment, storage, and disposal facilities. The regulations also address extensive definitions and listing/identification mechanisms for hazardous waste along with a tracking system. Treatment is favored over disposal and emphasis is on ``front-end`` treatment such as waste minimization and pollution prevention. A study of large corporations such as Xerox, 3M, and Dow Chemical, as well as the public sector, has shown that well known and successful hazardous waste management programs emphasize pollution prevention and employment of techniques such as proactive environmental management, environmentally conscious manufacturing, and source reduction. Nearly all successful hazardous waste programs include some aspects of Total Quality Management, which begins with a strong commitment from top management. Hazardous waste management at the Rocky Flats Plant is further complicated by the dominance of ``mixed waste`` at the facility. The mixed …

Preparation, characterization, and reactivity of ({eta}{sup 2}- alkyne)(meso-tetratolylpoprphrinato)titanium(II) complexes are described, along with inetermetal oxygen atom transfer reactions involving Ti(IV) and Ti(III) porphyrin complexes. The {eta}{sup 2}- alkyne complexes are prepared by reaction of (TTP)TiCl{sub 2} with LiAlH{sub 4} in presence of alkyne. Structure of (OEP)Ti({eta}{sup 2}-Ph-C{triple_bond}C-Ph) (OEP=octaethylporphryin) was determined by XRD. The compounds undergo simple substitution to displace the alkyne and produce doubly substituted complexes. Structure of (TTP)Ti(4-picoline){sub 2} was also determined by XRD. Reaction of (TTP)Ti{double_bond}O with (OEP)Ti-Cl yields intermetal O/Cl exchange, which is a one-electron redox process mediated by O atom transfer. Also a zero-electron redox process mediated by atom transfer is observed when (TTP)TiCl{sub 2} is reacted with (OEP)Ti{double_bond}O.

A newly developed method for surface structure determination, tensor LEED, combined with automated search was used to analyze the structures. The ordered structures of S on Mo(100) which were studied formed a c(2 {times} 2), c(4 {times} 2), and p(2 {times} l) periodicities at coverages of 0.5, 0.75, 1.0 ML (monolayers, of one sulfur atom per one molybdenum atom) respectively. A MO{sub 2}S-like overlayer, which formed at coverages greater than 1.0 ML, is also discussed. Calculations for the c(2 {times} 2) structure gave a best fit geometry with S adsorbed in a four-fold symmetric hollow site and the second layer buckled by 0.09{Angstrom}. The S-Mo bond length is 2.45{Angstrom} and the Pendry R-factor is 0.21. Preliminary calculations for the c(4 {times} 2) structure did not yield an acceptable fit. The three models tried are discussed. Calculations for p(2 {times} l) data did not yield an acceptable geometry either. The types of models that were tried are discussed. Implications of this analysis are discussed along with results of a scanning tunneling microscopy (STM) investigation. The ordered structures on the RE(0001) surface studied have p(2 {times} 2) and (2{radical}3 {times} 2{radical}3)R30{degree} periodicities and occurred at S coverages of 0.25 and 0.5 ML …

A radiofrequency-gain total power radiometer measured the intensity of the cosmic microwave background (CMB) at a frequency of 1.47 GHz (20.4 cm wavelength) from White Mountain, California, in September 1988 and from the South Pole, Antarctica, in December 1989. The CMB thermodynamic temperature, TCMB, is 2.27 {plus_minus} 0.25 K (68% C.L.) measured from White Mountain and 2.26 {plus_minus} 0.21 K from the South Pole site. The combined result is 2.27 {plus_minus} 0.19 K. The correction for galactic emission has been derived from scaled low-frequency maps and constitutes the main source, of error. The atmospheric signal is found by extrapolation from zenith scan measurements at higher frequencies. The result is consistent with previous low-frequency measurements, including a measurement at 1.41 GHz (Levin et al. 1988) made with an earlier version of this instrument. The result is {approximately}2.5 {sigma} ({approximately}l% probability) from the 2.74 {plus_minus} 0.02,K global average CMB temperature.

None

Crystal fault zones exhibit spatially heterogeneous slip behavior at all scales, slip being partitioned between stable frictional sliding, or fault creep, and unstable earthquake rupture. An understanding the mechanisms underlying slip segmentation is fundamental to research into fault dynamics and the physics of earthquake generation. This thesis investigates the influence that large-scale along-strike heterogeneity in fault zone lithology has on slip segmentation. Large-scale transitions from the stable block sliding of the Central 4D Creeping Section of the San Andreas, fault to the locked 1906 and 1857 earthquake segments takes place along the Loma Prieta and Parkfield sections of the fault, respectively, the transitions being accomplished in part by the generation of earthquakes in the magnitude range 6 (Parkfield) to 7 (Loma Prieta). Information on sub-surface lithology interpreted from the Loma Prieta and Parkfield three-dimensional crustal velocity models computed by Michelini (1991) is integrated with information on slip behavior provided by the distributions of earthquakes located using, the three-dimensional models and by surface creep data to study the relationships between large-scale lithological heterogeneity and slip segmentation along these two sections of the fault zone.

Novel graphite-like materials, BC{sub x} (6>x{ge}3), have been prepared using BCl{sub 3} and C{sub 6}H{sub 6} at 800--1000C, and C{sub x}N (14>x{ge}5) have been synthesized using C{sub 5}H{sub 5}N and Cl{sub 2} at 680C--986C. Bulk and thin film characterization were used to study the structure and bonding in these solids. C{sub 8}K(NH{sub 3}){sub 1.1} was prepared by reacting C{sub 8}K with gaseous NH{sub 3}. The carbon sub-lattice is hexagonal: a = 2.47 {Angstrom}, c = 6.47 {Angstrom}. The smaller a parameter and lower conductivity are attributed to smaller electron transfer from K to the conduction band solvation of K by NH{sub 3}. A simplified liquid phase method for synthesizing Li-graphite intercalation compounds has been developed; synthesis of a lamellar mixed conductor, C{sub x}{sup +}Li{sub 2}N{sup {minus}}, has been attempted. Stability and conductivity of (BN){sub 3}SO{sub 3}F have been studied; it was shown to be metallic with a specific conductivity of 1.5 S{center_dot}cm{sup {minus}1}. Its low conductivity is attributed to the low mobility of holes in BN sheets.

The electrochemistry of the heterobimetallic complexes (fulvalene)WFe(CO){sub 5} (30) and (fulvalene)WRu(CO){sub 5} (31) has been investigated. Compound 30 is reduced in two one-electron processes, and this behavior was exploited synthetically to prepare a tetranuclear dimer by selective metal reduction. Complex 31 displayed a distinction between the metals upon reoxidation of the dianion, allowing the formation of a dimer by selective metal anion oxidation. The redox behavior of 30 led to an investigation of the use of electrocatalysis to effect metal-specific ligand substitution. It was found that reduction of 30 with a catalytic amount of CpFe(C{sub 6}Me{sub 6}) (97) in the presence of excess P(OMe){sub 3} or PMe{sub 3} led to the formation of the zwitterions (fulvalene)[W(CO){sub 3}{sup {minus}}][Fe(CO)PR{sub 3}{sup +}] (107, R = P(OMe){sub 3}; 108, R = PMe{sub 3}). Compound 31 also displayed unique behavior with different reducing agents, as the monosubstituted zwitterion (fulvalene)[W(CO){sub 3}{sup {minus}}][Ru(CO){sub 2}(PMe{sub 3}){sup +}] was obtained when 97 was used while the disubstituted complex (fulvalene) [W(CO){sub 3}{sup {minus}}] [Ru(CO)(PMe{sub 3}){sub 2}{sup +}] was produced when Cp*Fe(C{sub 6}Me{sub 6}) was the catalyst. Potential synthetic routes to quatercyclopentadienyl complexes were also explored. Various attempts to couple heterobimetallic fulvalene compounds proved to be unsuccessful. 138 refs.

I have used a scanning tunneling microscope (STM) to image simultaneously the atomic lattice and the charge density wave (CDW) superstructure in tantalum disulfide (1T-TaS{sub 2}) over the temperature range of 370-77K. In the lowest temperature (commensurate) phase, present below 180K, the CDW is at an angle of 13.9{degrees} relative to the lattice and is uniformly commensurate. In the incommensurate phase, present above 353K, the CDW is aligned with the lattice. 1T-TaS{sub 2} exhibits two other phases; the triclinic (T) phase which is present between 223K and 283K upon warming the sample, and the nearly-commensurate (NC) phase which is present between 353K and 180K upon cooling the sample and between 283K and 353K upon warming the sample. In both of these phases, discommensurate models where the CDW is arranged in small commensurate domains have been proposed. In the NC phase the CDW is rotated between 10{degrees} and 12.5{degrees} relative to the atomic lattice. Such a rotated CDW would create an interference pattern with the underlying atomic lattice regardless of the existence of a true domain superstructure. Previous work on 1T-TaS{sub 2} has not adequately accounted for the possibility of this moire pattern. However, around each fundamental CDW peak in the …

A large diameter precision air bearing rotary table with a 16.0 inch diameter through-hole was designed, fabricated and tested in the course of this research. The rotary table will be used in conjunction with a specialized, computer controlled precision inspection machine being designed for the Department of Energy`s (DOE) Nuclear Weapons Complex (NWC). The design process included a complete engineering analysis to predict the final performance of the rotary table, and to ensure that the rotary table meets the required accuracy of 4.0 microinches of total radial (3.5 microinches average radial) and 4.0 microinches total axial (3.5 microinches average axial) errors. The engineering analysis included structural deformation, thermal sensitivity and dynamic analyses using finite element methods in some cases, as well as other analytic solutions. Comparisons are made between predicted and tested values, which are listed in the rotary table error budget. The rotary table performed as predicted with measured axial and radial stiffnesses of 1.1E06 lbf/inch and 2.9E06 lbf/inch, respectively, as well as average radial, axial and tilt errors of 2.5 microinches, 1.5 microinches, and less than 0.05 arcseconds, respectively.

The divalent lanthanide complex, (Me{sub 5}C{sub 5}){sub 2}Yb, reacts with methylcopper to produce the base-free, ytterbium-methyl complex, (Me{sub 5}C{sub 5}){sub 2}YbMe. This product forms a asymmetric, methyl-bridged dimer in the solid state. The bulky alkyl complex, (Me{sub 5}C{sub 5}){sub 2}YbCH(SiMe{sub 3}){sub 2}, displays similar chemistry to (Me{sub 5}C{sub 5}){sub 2}YbMe, but at a reduced reaction rate due to the limited accessibility of the metal in (Me{sub 5}C{sub 5}){sub 2}YbCH(SiMe{sub 3}){sub 2}. Copper and silver halide salts react with (Me{sub 5}C{sub 5}){sub 2}V to produce the trivalent halide derivatives, (Me{sub 5}C{sub 5}){sub 2}VX (X + F, Cl, Br, I). The chloride complex, (Me{sub 5}C{sub 5}){sub 2}VCl, reacts with lithium reagents to form the phenyl and borohydride species. Nitrous oxide transfers an oxygen atom to (Me{sub 5}C{sub 5}){sub 2}V producing the vanadium-oxo complex, (Me{sub 5}Ce{sub 5}){sub 2}VO. The trivalent titanium species, (Me{sub 5}C{sub 5}){sub 2}TiX (X = Cl, Br, Me, BH{sub 4}), form bimetallic coordination complexes with (Me{sub 5}C{sub 5}){sub 2}Yb. The magnetic behavior of the products indicates that electron transfer has not occurred. The solid state structures of the chloride and bromide complexes show unusual bend angles for the halide bridges between ytterbium and titanium. A model based on frontier orbital …

Conformal field theories comprise a vast class of exactly solvable two dimensional quantum field theories. Conformal theories with an enlarged symmetry group, the current algebra symmetry, axe a key ingredient to possible string compactification models. The following work explores a Lagrangian approach to these theories. In the first part of this thesis, a large class of conformal theories, the so-called coset models, are derived semi-classically from a gauged version Of the Wess-Zumino-Witten functional. A non-local field transformation to the parafermionic field description is employed in the quantization procedure. Classically, these parafermionic fields satisfy non-trivial Poisson brackets, providing insight into the fractional spin nature of the conformal theory. The W-algebra symmetry is shown to appear naturally in this approach. In the second part of this thesis, the connection between the fusion algebra structure of Wess-Zumino-Witten models and the quantization of the Chern-Simons action on the torus is made explicit. The modular properties of the conformal model are also derived in this context, giving a natural demonstration of the Verlinde conjecture. The effects of background gauge fields and monopoles are also discussed.

This thesis is an investigation of methods for processing multidimensional signals acquired using modern tomography systems that have an anisotropic or spatially variant response function. The main result of this research is the discovery of a new method to obtain better estimators of an unknown spatial intensity distribution by incorporating detailed knowledge about the tomograph system response function and statistical properties of the acquired signal into a mathematical model.

The production cross section times decay branching ratio for W{sub +{gamma}} in the electron decay channel in {radical}s = 1.8 TeV {bar p}-p collisions has been measured using W {yields} e{gamma} data sample obtained from the CDF 1988--1989 Tevatron collider run. For photons in the central region ({vert_bar}{eta}{sub {gamma}}{vert_bar} < 1.1) of the CDF detector with transverse energies E{sub T}{sup {gamma}} 5.0 GeV and lepton-photon angular separation {Delta}R{sub {ell}}{gamma} > 0.7, eight electron W{sub {gamma}} candidates were observed. From these events, the production cross section times decay branching ratio for the electron sample was measured to be a {sigma} {center_dot} B(W{sub {gamma}}){sub exp} = 17.0{sub {minus}13.4}{sup +13.6}(stat. + syst.)pb. The W{sub {gamma}} cross section is sensitive to the anomalous couplings of the W boson. Using the W{sub {gamma}} cross section measurement, the absence of an excess of large E{sub T} photons accompanying the production of a W boson enables one to obtain direct limits on anomalous WW{sub {gamma}} couplings. The experimental limits on the anomalous couplings was measured to be {minus}7.2 < {Delta}{kappa} < +7.7 ({lambda} = 0) and {minus}3.5 < {lambda} < +3.4 ({Delta}{kappa} = 0) at 95% CL. These experimental limits impose contraints on possible internal structure of …

A putative transcription factor in Rhodobactor capsulatus which binds upstream of the crt and bch pigment biosynthesis operons and appears to play a role in the adaptation of the organism from the aerobic to the anaerobic-photosynthetic growth mode was characterized. Chapter 2 describes the identification of this factor through an in vitro mobility shift assay, as well as the determination of its binding properties and sequence specificity. Chapter 3 focuses on the isolation of this factor. Biochemistry of later carotenoid biosynthesis enzymes derived from the non-photosynthetic bacterium, Erwinia herbicola. Chapter 4 describes the separate overexpression and in vitro analysis of two enzymes involved in the main sequence of the carotenoid biosynthesis pathway, lycopene cyclase and 5-carotene hydroxylase. Chapter 5 examines the overexpression and enzymology of functionally active zeaxanthin glucosyltransferase, an enzyme which carries out a more unusual transformation, converting a carotenoid into its more hydrophilic mono- and diglucoside derivatives. In addition, amino acid homology with other glucosyltransferases suggests a putative binding site for the UDP-activated glucose substrate.

For the semiconductor SiC to fulfill its potential as an electronic material, methods must be developed to produce insulating surface oxide layers in a reproducible fashion. Auger electron spectroscopy (AES), low energy electron diffraction (LEED) and x-ray photoelectron spectroscopy (XPS) were used to investigate the oxidation of single crystal {alpha}-SiC over a wide temperature and O{sub 2} pressure range. The {alpha}-SiC surface becomes graphitic at high temperatures and low O{sub 2} pressures due to Si and SiO sublimation from the surface. Amorphous SiO{sub 2} surface layers from on {alpha}-SiC at elevated O{sub 2} pressures and temperatures. Both the graphitization and oxidation of {alpha}-SiC appears to be enhanced by surface roughness. Chemical vapor deposition (CVD) is currently the preferred method of producing single crystal SiC, although the method is slow and prone to contamination. We have attempted to produce SiC films at lower temperatures and higher deposition rates using plasma enhanced CVD with CH{sub 3}SiH{sub 3}. Scanning AES, XPS and scanning electron microscopy (SEM) were utilized to study the composition and morphology of the deposited Si{sub x}C{sub y}H{sub z} films as a function of substrate temperature, plasma power and ion flux bombardment of the film during deposition. High energy ion bombardment …

The newly developed optical techniques have been combined with either second harmonic (SH) diffraction or linear diffraction off a monolayer adsorbate grating for surface diffusion measurement. Anisotropy of surface diffusion of CO on Ni(l10) was used as a demonstration for the second harmonic dim reaction method. The linear diffraction method, which possesses a much higher sensitivity than the SH diffraction method, was employed to study the effect of adsorbate-adsorbate interaction on CO diffusion on Ni(l10) surface. Results showed that only the short range direct CO-CO orbital overlapping interaction influences CO diffusion but not the long range dipole-dipole and CO-NI-CO interactions. Effects of impurities and defects on surface diffusion were further explored by using linear diffraction method on CO/Ni(110) system. It was found that a few percent S impurity can alter the CO diffusion barrier height to a much higher value through changing the Ni(110) surface. The point defects of Ni(l10) surface seem to speed up CO diffusion significantly. A mechanism with long jumps over multiple lattice distance initiated by CO filled vacancy is proposed to explain the observed defect effect.

The fast excited state intramolecular proton transfer of 3-hydroxyflavone is measured and effects of external hydrogen-bonding interactions on the proton transfer are studied. The proton transfer takes place in {approximately}240 fsec in nonpolar environments, but becomes faster than instrumental resolution of 110 fsec in methanol solution. The dynamics following photodissociation of CH{sub 2}I{sub 2} and other small molecules provide the first direct observations of geminate recombination. The recombination of many different photodissociating species occurs on a {approximately}350 fsec time scale. Results show that recombination yields but not rates depend on the solvent environment and suggest that recombination kinetics are dominated by a single collision with surrounding solvent cage. Studies of sterically locked phenyl-substituted butadienes offer new insights into the electronic structure and isomerization behavior of conjugated polyenes. Data show no simple correlation between hinderance of specific large amplitude motions and signatures of isomerizative behavior such as viscosity dependent excited state lifetimes, implying that the isomerization does not provide a suitable for simple condensed phase reaction rate theories. The spectral dynamics of a photochromic spiropyran indicate that recombination, isomerization and vibrational relaxation all play important roles in photoreactivity of complex molecules. The interplay of these microscopic phenomena and their effect on …

Nuclear Magnetic Resonance (NMR) spectroscopy of xenon has become an important tool for investigating a wide variety of materials, especially those with high surface area. The sensitivity of its chemical shift to environment, and its chemical inertness and adsorption properties make xenon a particularly useful NMR probe. This work discusses the application of optical pumping to enhance the sensitivity of xenon NMR experiments, thereby allowing them to be used in the study of systems with lower surface area. A novel method of optically-pumping {sup 129}Xe in low magnetic field below an NMR spectrometer and subsequent transfer of the gas to high magnetic field is described. NMR studies of the highly polarized gas adsorbed onto powdered samples with low to moderate surface areas are now possible. For instance, NMR studies of optically-pumped xenon adsorbed onto polyacrylic acid show that xenon has a large interaction with the surface. By modeling the low temperature data in terms of a sticking probability and the gas phase xenon-xenon interaction, the diffusion coefficient for xenon at the surface of the polymer is determined. The sensitivity enhancement afforded by optical pumping also allows the NMR observation of xenon thin films frozen onto the inner surfaces of different …

The hydrogen exchange reaction H + H{sub 2} {yields} H{sub 2} + H (and its isotopic variants) plays a pivotal role in chemical reaction dynamics. It is the only chemical reaction for which fully converged quantum scattering calculations have been carried out using a potential energy surface which is considered to be chemically accurate. To improve our ability to test the theory, a `perfect experiment`, measuring differential cross sections with complete specification of the reactant and product states, is called for. In this thesis, the design of an experiment is described that aims at achieving this goal for the D + H{sub 2} reaction. A crossed molecular beam arrangement is used, in which a photolytic D atom beam is crossed by a pulsed beam of H{sub 2} molecules. DH molecules formed in the D + H{sub 2} reaction are state-specifically ionized using Doppler-free (2+1) Resonance-Enhanced Multi-Photon Ionization (REMPI) and detected using a Position-sensitive microchannel plate detector. This detection technique has an unprecedented single shot detection sensitivity of 6.8 10{sup 3} molecules/cc. This thesis does not contain experimental results for the D + H{sub 2} reaction yet, but progress that has been made towards achieving this goal is reported. In addition, …

High content of excess As is incorporated in GaAs grown by low-temperature molecular-beam-epitaxy (LTMBE). The excess As exists primarily as As antisite defects AsGa and a lesser extent of gallium vacancies V{sub Ga}. The neutral AsGa-related defects were measured by infrared absorption at 1{mu}m. Gallium vacancies, V{sub Ga}, was investigated by slow positron annihilation. Dependence of defect contents on doping was studied by Si and Be dopants. No free carriers are generated by n-type or p-type doping up to 10{sup 19} cm{sup {minus}3} Si or Be. Raman data indicate Be occupies Ga substitutional sites but Si atom is not substitutional. Si induces more As{sub Ga} in the layer. As As{sub Ga} increases, photoquenchable As{sub Ga} decreases. Fraction of photoquenchable defects correlates to defects within 3 nearest neighbor separations disrupting the metastability. Annealing reduces neutral As{sub Ga} content around 500C, similar to irradiation damaged and plastically deformed Ga{sub As}, as opposed to bulk grown GaAs in which As{sub Ga}-related defects are stable up to 1100C. The lower temperature defect removal is due to V{sub Ga} enhanced diffusion of As{sub Ga} to As precipitates. The supersaturated V{sub GA} and also decreases during annealing. Annealing kinetics for As{sub Ga}-related defects gives 2.0 {plus_minus} …

Magnetic flux noise and flux creep in thin films and single crystals of YBa{sub 2}Cu{sub 3}O{sub 7-x}, Bi{sub 2}Sr{sub 2}CaCu{sub 2}O{sub 8+x}, Tl{sub 2}Ca{sub 2}Ba{sub 2}Cu{sub 3}O{sub x}, and TlCa{sub 2}Ba{sub 2}Cu{sub 3}O{sub x} are measured with a superconducting quantum interference device (SQUID). The noise power spectrum generally scales as 1/f (f is frequency) from 1 Hz to 1 kHz, increases with temperature, and decreases in higher-quality films. It is proportional to the magnetic field B in which the sample is cooled, at least in the range 0.1 mT < B < 3 mT. A model of thermally activated vortex motion is developed which explains the dependence of the noise on frequency, temperature, current, and applied magnetic field. The pinning potential is idealized as an ensemble of double wells, each with a different activation energy separating the two states. From the noise measurements, this model yields the distribution of pinning energies in the samples, the vortex hopping distance, the number density of mobile vortices, and the restoring force on a vortex at a typical pinning site. The distribution of pinning energies in YBa{sub 2}Cu{sub 3}O{sub 7-x} shows a broad peak below 0.1 eV. The small ambient magnetic field, and the …

DX centers are deep level defects found in some III-V semiconductors. They have persistent photoconductivity and large difference between thermal and optical ionization energies. Hydrostatic pressure was used to study microstructure of these defects. A new local vibrational mode (LVM) was observed in hydrostatically stressed, Si-doped GaAs. Corresponding infrared absorption peak is distinct from the Si{sub Ga} shallow donor LVM peak, which is the only other LVM peak observed in our samples, and is assigned to the Si DX center. Analysis of the relative intensities of the Si DX LVM and the Si shallow donor LVM peaks, combined with Hall effect and resistivity indicate that the Si DX center is negatively charged. Frequency of this new mode provides clues to the structure of this defect. A pressure induced deep donor level in S-doped InP was also discovered which has the properties of a DX center. Pressure at which the new defect becomes more stable than the shallow donor is 82 kbar. Optical ionization energy and energy dependence of the optical absorption cross section was measured for this new effect. Capture barrier from the conduction band into the DX state were also determined. That DX centers can be formed in InP …