No abstract

No abstract

No abstract

We report an optical, interferometric technique for measuring the temperature of semiconductor substrates during heating or cooling, which is applicable in vacuum. The technique circumvents many of the problems associated with thermocouple or pyrometer measurements. A low-power infrared (IR) laser (e.g., λ=1.15-μm He–Ne laser) having an energy below the band gap is directed at a wafer that is polished on both sides, where either reflected or transmitted laser light is detected by a photodiode. Interference results between reflections off the front and back surfaces of the wafer. As the temperature of the wafer is either increased or decreased, the temperature dependence of the refractive index, along with a smaller contribution from thermal expansion, causes the optical path within the wafer to change by λ/2n (i.e., a full interference cycle) for every ∼3 K for a typical Si, GaAs, or InP wafer thickness of 500 μm. Consequently, temperature changes of ±0.2 K are easily detected. This technique, has been used between room temperature and 600 °C on GaAs substrates in a low-pressure metal-organic chemical vapor deposition (MOCVD) system, and in an ultrahigh vacuum thermal desorption experiment. This method can be used well below room temperature, as well as at temperatures above 650 °C with the optimum choice in laser wavelength. Application of this method to other processes such as molecular beam epitaxy (MBE), reactive ion etching, and rapid thermal processing should be straightforward. We also describe a refinement of the method for measuring the sign, as well as magnitude of temperature changes for typical, slightly tapered wafers during heating or cooling cycles. In this case the reflected laser beam contains a series of parallel lines that move toward the thinner end of the region probed by the laser beam as the temperature increases. Sensing the direction that these spatial interference fringes move can be used to determine whether the sample is heating or cooling.

We report studies of the kinetics of thermal decomposition of triethylgallium (TEGa), trimethylgallium (TMGa), and trimethylindium (TMIn) adsorbed on GaAs(100) in ultrahigh vacuum. The adsorbed layers were prepared by dosing GaAs(100) at room temperature, to either saturated coverage or coverages below saturation. The relative coverage of carbon was monitored by x-ray photoelectron spectroscopy (XPS) as the substrate temperature was slowly increased (0.6–3.2 °C/min). Products were detected at faster heating rates (0.7–6 °C/s) with a differentially pumped quadrupole mass spectrometer. The substrate temperature was measured by infrared laser interferometric thermometry. The kinetic analysis also makes use of XPS and mass spectrometric data on laser-induced, rapid thermal decomposition (heating rates of ∼1011 °C/s ). TEGa dissociatively chemisorbs on GaAs(100) at room temperature. Heating the substrate from room temperature to ∼500 °C results in desorption of a Ga–alkyl at low temperature, ascribed mostly to diethylgallium (DEGa) and possibly some TEGa. At higher temperature, C2H4 and C2H5 desorb in parallel after most of the Ga–alkyl has desorbed. The hydrocarbon desorption is described well by simple first order kinetics with an activation energy, Eact=32±4 kcal/mol, and a pre-exponential A factor of 2.5×1010±1.5 s−1. Ga–alkyl desorption is more complicated; the Arrhenius parameters for assumed first order desorption exhibit strong coverage dependences. A fit to all the data was obtained for A=5×108 s−1 and Eact=18 kcal/mol at saturated coverage, with a large decrease in Eact (or increase in A) with decreasing coverage. TMGa decomposes to yield a Ga–alkyl desorption product (either dimethylgallium, TMGa, or a mixture of the two) at low temperature, and CH3 at higher temperature. CH3 desorption has a first order activation energy of 43±2 kcal/mol for an assumed A factor of 1×1013 s−1. For the Ga–alkyl, A=108 s−1 and Eact=19 kcal/mol, with a coverage dependence similar to DEGa desorption from TEGa decomposition. TMIn undergoes a methyl exchange reaction on GaAs(100). Upon heating above room temperature, a Ga–alkyl desorbs first, followed by desorption of CH3 at higher temperature. The Ga–alkyl has with the same cracking pattern as observed for TMGa decomposition. No In–alkyls desorb, and In desorption does not occur until all carbon-containing species desorb. CH3 starts to desorb at lower temperature than for TMGa decomposition. Assuming an A factor of 1×1013 s−1, CH3 desorption over the observed wide temperature range indicates a range of activation energies from 33–43 kcal/mol. Ga–alkyl desorption is similar to that observed during TMGa decomposition. At saturated coverage, A=108 and Eact=17 kcal/mol. However, the coverage dependence is not as strong as for TMGa, so that Ga–alkyl desorption peaks at lower temperature for TMIn. Decomposition mechanisms for these group-III metal alkyls are discussed, along with implications for growth of III–V compound semiconductor films from these precursors by chemical vapor deposition and molecular beam techniques.