Our purpose is to answer fundamental questions related to technologically important chemical reactions at the gas-solid interface, e.g, Fircher-Trospch synthesis and methanol synthesis. Here, we present reflection absorption infrared spectroscopy (RAIRS) as our principal surface science technique. The UT-RAIRS set-up, which is described in detail below, operates with an outstanding sensitivity due to excellent stability and low noise level performance. Overall, the set-up is designed so that detector noise is greater than noise from all other sources combined.

The current projects focus on the thermal and non-thermal reactions of molecular adsorbates on single metal surfaces. The following examples are under investigation on Cu(100) and Ag(111) surfaces:

• Reaction kinetics of C₃ hydrocarbons.
• Formation of alkoxide surface intermediates from alkyl nitrites.
• Formation of metallocycle species from cyclopropane by electron attachment.

Experiments are performed in a two-level ultrahigh vacuum (UHV) chamber with a base pressure < 4 x 10^-10 Torr. The lower chamber is equipped with a single-pass cylindrical mirror analyzer with coaxial electron gun (Physical Electronics) for Auger electron spectroscopy (AES). A differentially pumped quadrupole mass spectrometer (Extrel C-150) for residual gas analysis and line-of-sight temperature programmed desorption (TPD) experiments.

Figure 1

The apparatus for RAIRS experiments consists of a 4 in. inner diameter cylindrical chamber in the upper level, which is coupled to a commercial FT-IR spectrometer (Nicolet, model Magna IR 860). A picture of the upper level for RAIRS studies is shown in Figure 1. Details of the RAIR set-up are given below. The upper chamber also houses a residual gas analyzer (Stanford Research Systems, SRS-RGA), an ion sputter gun (LK Technologies, model NGI3000-SE), and a hot filament electron source. A sample manipulator allows for 350 mm of travel in the vertical direction and 10 mm in two orthogonal directions. The manipulator is attached to a differentially pumped rotary seal, constructed in-house, permits a full 360 of rotation of the sample about the chamber center axis without affecting the base pressure. The sample holder, constructed in-house, allows metal substrates to be heated resistively to 1000 K and cooled to 77 K using liquid nitrogen. Recently, we also added a radio frequency (RF) plasma source to crack simple molecules such as oxygen and hydrogen. The running conditions are as follows: 50 Watts incident power, pressure in the plasma region is 150 mTorr-1.0 Torr, and efficiency is 50%. Under these conditions, the chamber’s pressure is <1x10^{^-5} Torr. After turning the source off, the base pressure of the chamber is reached in minutes.

Figure 2

As shown in Figure 2, The IR beam enters and exits the UHV chamber through differentially pumped O-ring sealed KBr IR windows. The FTIR, and the entire optical path external to the UHV chamber, are sealed for purging. Adequate purging of atmospheric H₂O and CO₂ is achieved with a compressor (Mobile International, Inc.). Complete elimination (below the detection limits of the spectrometer) of water bands is achieved during normal scanning intervals. The collimated IR beam (~51 mm in diameter) emerging from the interferometer is focused onto a flat mirror (inside) and is sent out of the spectrometer to a second flat mirror which directs the beam to an off-axis paraboloidal mirror (effective focal length of 228 mm). The beam is focused at a 82 grazing angle with respect to the surface normal of the 13 mm diameter metal substrate; the diameter was chosen to minimize the loss of infrared light. The exiting beam is p-polarizer with a ZnSe wire grid polarizer before it reaches a 82 off-axis ellipsoidal mirror (42 mm short and 228 mm long effective focal length, respectively). Finally, the beam is focused onto a detector where there is a close match between the image of the beam (~1 x 1 mm²) at the detector position and the dimensions of the detectors active area (1 x 1 mm²). Figure 2 also shows an indium antimonide (InSb) detector, with a low frequency cutoff of ~ 1850 cm^-1; and a narrow band MCT detector, with a low frequency cuttoff ~ 775 cm^-1. Experiments can be performed with either of the two detectors by simply rotating the ellipsoidal mirror by 180 to the preferred detector. The relative sensitivity in the region between 3000 and 3400 cm^-1 for each detector is as follows: MCT (narrow):InSb by about a factor of 1:3 respectively. Thus, InSb is the most sensitive detector for the region above 1800 cm^-1where the peak-to-peak noise level is ~0.003% D R/R units.

Reaction kinetics of allyl halides on Cu(100).

Figure 3

Figure 3. RAIRS spectra of allyl bromide (q ~ 0.07 ML) dosed at different temperatures on Cu(100) saturated with 1:1 mixture of chemisorbed h3-allyl and Cl. The reaction of physisorbed C3H5Br and h3-C3H5(c) produces physisorbed 1,5-hexadiene. The C4H10(p) production rate is controlled by two processes-between 77 and 110 K, the reaction between h3-C3H5(c) and C3H5Br(p) controls while between 110 and 160 K, the desorption of C3H5Br(p) controls.

Figure 4

Figure 4. Based on the RAIRS and TPD measurements, a schematic one-dimensional energy profile for the reaction, illustrated in the inset, of C3H5Br(g), denoted R(g), with h3-C3H5(c) to form C4H10(p), denoted P(g). Adsorption first forms R(p) that is physisorbed over the 1:1 mixture of the C3H5(c) and Cl(c) prepared by the thermal dissociation of C3H5Cl. The adsorption energy is 38 kJmol-1. In competition with desorption, thermal activation of R(p) forms a transition state, denote R† , from which product P(p) physisorbed 1,5-hexadiene adsorbed on Br(c) + Cl(c), is synthesized. The activation energy difference, Ed-Er, between desorption and reaction of C3H5Br(g), with C3H5(c) is 12 kJmol-1. The activation for desorption of P(p)