EMSL: Mike Henderson's Publications

Ohsawa T, I Lyubinetsky, Y Du, MA Henderson, V Shutthanandan, and SA Chambers. 2009. "Crystallographic Dependence of Visible-light Photoactivity in Epitaxial TiO2−xNx Anatase and Rutile." Physical Review. B, Condensed Matter and Materials Physics 79(8):Article number: 085401. doi:10.1103/PhysRevB.79.085401 Abstract Nitrogen-doped TiO2 materials have been shown to exhibit visible-light photoactivity, but the operative mechanism(s) are not well understood. Here we use structurally and compositionally well-defined epitaxial films of TiO2−xNx anatase (001) and rutile (110) (x~0.02) to show a qualitative difference between the visible-light activities for the two polymorphs. Holes generated by visible light at N sites in anatase (001) readily diffuse to the surface and oxidize adsorbed trimethyl acetate while the same in rutile (110) remain trapped in the bulk. In light of the low doping densities that can be achieved in phase-pure material, conventional wisdom suggests that holes should be trapped at N sites in both polymorphs. Although the detailed mechanism is not yet understood, these results suggest that the hole hopping probability is much higher along the [001] direction in N-doped anatase than along the [110] direction in N-doped rutile.

Ohsawa T, I Lyubinetsky, Y Du, MA Henderson, V Shutthanandan, and SA Chambers. 2009. "Crystallographic Dependence of Visible-Light Photochemistry in Epitaxial TiO2-xNx Anatase and Rutile." Physical Review. B, Condensed Matter and Materials Physics 79(8):Art. No. 085401. Abstract All films were grown by plasma assisted molecular beam epitaxy (PAMBE) in a custom chamber described elsewhere (1). Epitaxial films of TiO2-xNx(001) (x ≤ ~0.02) anatase were grown by PAMBE on undoped or Nb-doped (0.02 at. %) SrTiO3(001) (STO) and undoped LaAlO3(001) (LAO). Similarly, TiO2-xNx(001) (x ≤ ~0.02) rutile epifilms were grown on rutile TiO2(110). The growth and physical properties of N-doped anatase on LAO(001) and N-doped rutile on TiO2(110) have been described in detail elsewhere (2-4). In what follows, we describe the growth details for N-doped anatase on STO(001). The PAMBE chamber is connected to an x-ray photoelectron spectrometer (XPS) chamber and a photodesorption chamber. The former is equipped with a Gamma Data/Scienta SES 200 analyzer and a monochromatic AlK x-ray source. The latter includes a molecular dosing apparatus for TMAA, a Hg arc lamp, and a quadrupole mass spectrometer. The STO substrates were etched in buffered HF and annealed in flowing O2 at 1 atm. at 950oC for 8 hours. The etch dissolved SrO terraces and the oxygen anneal resulted in mass transport of the discontinuous TiO2 microterraces, resulting in an atomically flat, TiO2 terminated surface with a minimum step height of 4 Å (5). This treatment left some residual fluorine on the surface which could not be removed by annealing. The measured F 1s binding energy was ~684.0 eV, which is close to that exhibited by SrF2 – 684.6 eV (6). Based on this binding energy and the high degree of thermal stability, we conclude that F substitutes for O in the lattice. Under this assumption and using atomic photoemission cross sections (7), the F mole fraction within the anion sublattice is estimated to be ~0.05 within the probe depth of XPS at normal emission (~45 Å).

Henderson MA. 2008. "Effect of Coadsorbed Water on the Photodecomposition of Acetone on TiO2(110)." Journal of Catalysis 256(2):287-292. doi:10.1016/j.jcat.2008.03.020 Abstract The influence of coadsorbed water on the photodecomposition of acetone on TiO2 was examined using temperature programmed desorption (TPD) and the rutile TiO2(110) surface as a model photocatalyst. Of the two major influences ascribed to water in the heterogeneous photocatalysis literature (promotion via OH radical supply and inhibition due to site blocking), only the negative influence of water was observed. As long as the total water and acetone coverage was maintained well below the first layer saturation coverage (‘1 ML’), little inhibition of acetone photodecomposition was observed. However, as the total water+acetone coverage exceeded 1 ML, acetone was preferentially displaced from the first layer to physisorbed states by water and the extent of acetone photodecomposition attenuated. The displacement originated from water compressing acetone into high coverage regions where increased acetone-acetone repulsions caused displacement from the first layer. The immediate product of acetone photodecomposition was adsorbed acetate, which occupies twice as many surface sites per molecule as compared to acetone. Since the acetate intermediate was more stable on the TiO2(110) surface than either water or acetone (as gauged by TPD) and since its photodecomposition rate was less than that of acetone, additional surface sites were not opened up during acetone photodecomposition for previously displaced acetone molecules to re-enter the first layer. Results in this study suggest that increased molecular-level repulsions between organic molecules brought about by increased water coverage are as influential in the inhibiting effect of water on photooxidation rates as are water-organic repulsions.

Henderson MA. 2008. "Ethyl Radical Ejection During Photodecomposition of Butanone on TiO2(110)." Surface Science 602(20):3188-3193. doi:10.1016/j.susc.2007.06.079 Abstract The photodecomposition of acetone and butanone were examined on the (110) surface of rutile TiO2 using temperature programmed desorption (TPD) and photon stimulated desorption (PSD). In both cases, photodecomposition was proceeded by a required thermal reaction between the adsorbed ketone and coadsorbed oxygen resulting in a diolate species. The diolate photodecomposed by ejection of an organic radical from the surface leaving behind a carboxylate species. In the acetone case, only methyl radical PSD was detected and acetate was left on the surface. In the butanone case there was a possibility of either methyl or ethyl radical ejection, with propionate or acetate left behind, respectively. However, only ethyl radical PSD was detected and the species left on the surface (acetate) was the same as in the acetone case. The preference for ethyl radical ejection is linked to the greater thermal stability of the ethyl radical over that of the methyl radical. Unlike in the acetone case, where the ejected methyl radicals did not participate in thermal chemistry on the TiO2(110) surface after photoactivation of the acetone diolate, ethyl radicals photodesorbing at 100 K from butanone diolate showed a preference for dehydrogenation to ethene through the influence of coadsorbed oxygen. These results reemphasize the mechanistic importance of organic radical production during photooxidation reactions on TiO2 surface. Pacific Northwest National Laboratory is operated by Battelle for the US Department of Energy.

Henderson MA. 2008. "Relationship of O2 Photodesorption in Photooxidation of Acetone on TiO2." Journal of Physical Chemistry C 112(30):11433-11440. doi:10.1021/jp802551x Abstract Organic photooxidation on TiO2 invariably involves the coexistence of organic species with oxygen on the surface at the same time. In the case of acetone and oxygen, both species exhibit their own interesting photochemistry on TiO2, but interdependences between the two are not understood. In this study, a rutile TiO2(110) surface possessing 7% surface oxygen vacancy sites is used as a model surface to probe the relationship between O2 photodesorption and acetone photodecomposition. Temperature programmed desorption (TPD) and photon stimulated desorption (PSD) measurements indicate that coadsorbed oxygen is essential to acetone photodecomposition on this surface, however the form of oxygen (molecular and dissociative) is not known. The first steps in acetone photodecomposition on TiO2(110) involve thermal activation with oxygen to form an acetone diolate ((CH3)2COO) species followed by photochemical decomposition to adsorbed acetate (CH3COO) and an ejected CH3 radical that is detected in PSD. Depending on the surface conditions, O2 PSD is also observed during the latter process. However, the time scales for the two PSD events (CH3 and O2) are quite different, withthe former occurring at ~10 times faster than the latter. By varying the preheating conditions or performing pre-irradiation on an O2 exposed surface, it becomes clear that the two PSD events are uncorrelated. That is, the O2 species responsible for O2 PSD is not a significant participant in the photochemistry of acetone on TiO2(110) and likely originates from a minority form of O2 on the surface. The CH3 and O2 PSD events do not appear to be in competition with each other suggesting either that ample charge carriers exist under the experimental conditions employed or that different charge carriers or excitation mechanisms are involved.

Ohsawa T, I Lyubinetsky, MA Henderson, and SA Chambers. 2008. "Hole-mediated Photodecomposition of Trimehtyl Acetate on a TiO2(001) Anatase Epitaxial Thin Film Surface." Journal of Physical Chemistry C 112(50):20050-20056. doi:10.1021/jp8077997 Abstract Surfaces of titanium dioxide in both rutile and anatase polymorphs have attracted significant attention in catalysis and photochemistry. The (110) orientation of rutile, and to a lesser extent other rutile orientations, have been studied on an atomic scale, yielding information on surface structure and chemical reactivity. In contrast, the thermal and photochemistry of well-defined, single-crystal anatase surfaces had not been investigated, largely because of the metastable nature of anatase , as well as the lack of availability of high-quality surfaces. Here we describe a study of the adsorption and photoreactivity of an organic adlayer, trimethyl acetate (TMA), on structurally-excellent anatase (001) epitaxial thin films grown by oxygen plasma assisted molecular beam epitaxy (OPAMBE). High-resolution scanning tunneling microscopy (STM), x-ray photoelectron spectroscopy (XPS), and photodesorption spectrometry have been used to study the chemisorptions and ultraviolet (UV) light-induced photodecomposition of TMA in ultrahigh vacuum. UV light promotes hole-mediated photodecomposition of TMA, resulting in decarboxylation to yield tert-butyl radical and CO2. The photochemical rate constant is equal to that measured for OPAMBE grown rutile TiO2(110) surfaces.

Zehr RT, and MA Henderson. 2008. "Acetaldehyde photochemistry on TiO2(110)." Surface Science 602(13):2238-2249. doi:10.1016/j.susc.2008.04.045 Abstract The ultraviolet (UV) photon induced decomposition of acetaldehyde absorbed on the oxidized retile TIO2(110) surface was studied with photon stimulated desorption (PSD) and theral programmed desorption (TPD). Acetaldehyde desorbs molecularly from TiO2(110) with minor decomposition channels yielding butene on the reduced TiO2 surface and acetate on the oxidized TiO2 surface. Acetaldehyde absorbed on oxidized TiO2(110) undergoes a facile thermal reaction to form a photoactive acetaldehyde-oxygen complex. UV irradiation of the acetaldehyde-oxygen complex resulting in the ejection of methyl radical into gas phase and conversion of the surface bound fragment to formate.

Zehr RT, and MA Henderson. 2008. "Influence of O2-induced surface roughening on the chemistry of water on TiO2(110)." Surface Science 602(8):1507-1516. Abstract The impact of oxygen induced regrowth of TiO2 on the reduced rutile TiO2(110) surface has been studied using temperature programmed desorption (TPD) of adsorbed water multilayers. Pre-exposure of UHV annealed TiO2(110) surfaces to O2 at temperatures from 300 to 850 K induced changes in subsequent water TPDs that were interpreted in terms of the rougher surface morphologies resulting from the regrowth process. Water TPD from TiO2(110) previously oxidized at 300 K exhibited a new peak at *312 K due to reaction of water with O adatoms. These O adatoms were produced by dissociative adsorption of O2 at O-vacancy sites. Additionally, oxygen reacted (slowly) with surface Ti2O3 strands at RT. Water TPD from surfaces pre-oxidized at higher temperatures (P500 K) exhibited features reflective of desorption from rough surfaces, namely loss of peak resolution and eventual merger of the second layer and ice peaks, formation of a high temperature tail on the second layer peak, and broadening of the first layer TPD peak. The multiplicity of kinetically different adsorption sites on the roughened TiO2(110) surfaces contributed to the widening of the desorption features. Published by Elsevier B.V.

Lyubinetsky I, Z Yu, and MA Henderson. 2007. "Direct Observation of Adsorption Evolution and Bonding Configuration of TMAA on TiO2(110)." Journal of Physical Chemistry C 111(11):4342-4346. doi:10.1021/jp067264d Abstract Trimethyl acetic acid (TMAA) adsorption evolution on the rutile TiO2(110) surface from submonolayer to saturation coverages was examined at the atomic level by scanning tunneling microscopy using the same area analysis approach. Upon TMAA deprotonation, no evidence of terminal OH group formation has been found. It has been suggested that uncommon geometry associated with detached hydrogen atom takes place instead, with proton bonding to pair bridging oxygen atoms. Such a configuration is likely to be stabilized by adjacent adsorbed TMA groups and, in turn, be a factor in the formation of TMA (2x1) reconstruction at saturation coverage. Our results indicate that TMAA adsorption on reduced TiO2 is virtually not affected by bridging oxygen vacancies or other surface defects.

Henderson MA, JM White, H Uetsuka, and H Onishi. 2006. "Selectivity Changes During Organic Photooxidation on TiO2: Role of O2 pressure and Organic Coverage." Journal of Catalysis 238(1):153-164. doi:10.1016/j.jcat.2005.12.004 Abstract The selectivity of trimethyl acetate (TMA) photodecomposition on TiO2(110) as a function of O2 pressure and TMA coverage was probed at room temperature (RT) using isothermal mass spectrometry (ISOMS) and scanning tunneling microscopy (STM). The selectivity of TMA photodecomposition on TiO2(110) is sensitive to the initial TMA coverage and the O2 pressure. TMA bridge bonds to the surface via the carboxylate end of the molecule in a manner consistent with the binding of other carboxylate species (e.g., formate and acetate) on TiO2 surfaces. Under all conditions, photodecomposition of TMA was initiated via hole reaction with the electron in carboxylate’s  system resulting in opening of the O-C-O bond angle, and formation of CO2 and a t-butyl radical by cleavage of the C-C bond between these groups. The CO2 product desorbs from the surface at RT, but the t-butyl radical has several options for thermal chemistry. In ultrahigh vacuum (UHV), where the O2 partial pressure is <1x10-10 torr, the TMA photodecomposition results in a near 1:1 yield of isobutene (i-C4H8) and isobutane (i-C4H10) from surface chemistry of the t-butyl radicals. STM results show that the reaction occurs fairly homogeneously across the TiO2(110) surface. In the presence of O2, the photodecomposition selectivity switches from initially i-C4H8 to a mixture of i-C4H8 and i-C4H10 and then back to predominately i-C4H8. The latter selectivity change occurs at the point at which void regions form and grow in the TMA overlayer. At this point, the photodecomposition rate accelerates and the reaction occurs preferentially at the interface between the TMA-rich and TMA-void regions on the surface. These results illustrate both the changing dynamics of a typical photooxidation reaction on TiO2, and how factors such as O2 pressure and TMA coverage, impact the photooxidation reaction selectivity. We also present results that suggest the rate of photodecomposition of monodentate carboxylates is greater than that of bidentate (bridging) carboxylates. This implies that the structural arrangement of Ti cation sites on the surface is an important issue that influences photocatalytic rates on TiO2.