Ion-Beam-Induced Amorphization and Chemical Modification of

Jan 24, 2007 - by the high-temperature solution growth (TSSG) technique. A three-zone furnace, which is inside a water-cooled chamber equipped with an...
0 downloads 0 Views 394KB Size
Download PDF
2702

J. Phys. Chem. C 2007, 111, 2702-2708

Low-Energy Ar+ Ion-Beam-Induced Amorphization and Chemical Modification of Potassium Titanyl Arsenate (001) Crystal Surfaces C. V. Ramana,*,† V. V. Atuchin,‡ U. Becker,† R. C. Ewing,†,§ L. I. Isaenko,| O. Yu. Khyzhun,⊥ A. A. Merkulov,| L. D. Pokrovsky,‡ A. K. Sinelnichenko,⊥ and S. A. Zhurkov| Nanoscience and Surface Chemistry Laboratory, Department of Geological Sciences, UniVersity of Michigan, Ann Arbor, Michigan 48109, Laboratory of Optical Materials and Structures, Institute of Semiconductor Physics, SB RAS, NoVosibirsk 90, 630090, Russia, Departments of Materials Science and Engineering & Nuclear Engineering and Radiological Sciences, UniVersity of Michigan, Ann Arbor, Michigan 48109, Laboratory of Crystal Growth, Institute of Geology and Mineralogy, SB RAS, NoVosibirsk 90, 630090, Russia, and Institute for Problems of Materials Science, NASU, UA-03142 KyiV, Ukraine ReceiVed: October 30, 2006; In Final Form: December 8, 2006

The effect of 1.5 keV Ar+ ion irradiation on the (001) surfaces of potassium titanyl arsenate, KTiOAsO4 (KTA), has been investigated using reflection high-energy electron diffraction (RHEED) and X-ray photoelectron spectroscopy (XPS). The (001) KTA surface is very sensitive to 1.5 keV Ar+ ion irradiation which induces significant structural modification in the top surface layers. The bonds in the KTA crystal lattice are relatively less unstable when compared to ; reduction of As5+ ions with the formation of As0 and partial As loss from the top surface occurs with Ar+ ion irradiation. The formation of an unstable layer with chemically active and passive arsenic states may be a factor reducing the optical parameters and durability of nonlinear devices involving KTA.

1. Introduction Potassium titanyl arsenate, KTiOAsO4 (KTA), is a representative member of a large isostructural family with the KTiOPO4 (KTP) structure-type.1,2 KTA has been recognized as an excellent nonlinear optical crystal for nonlinear optical and electrooptical device applications. KTA possesses high nonlinear optical coefficients, appropriate birefringence, and a wide range of transparency; thus it is considered as a most promising material for nonlinear optical applications in the infrared spectral range.3-9 As compared with potassium titanyl phosphate (KTiOPO4 or KTP), KTA has electrooptic and nonlinear figures of merit higher than KTP by 30 and 60% (for second harmonic generation under pumping at λ ) 1.32 µm), respectively.9,10 The lower absorption in the 3-5 µm spectral regions when compared with KTP makes KTA an important material for midIR optical parametric oscillation applications.4,11,12 At room-temperature KTA has orthorhombic crystal structure,13 Pna21 space group, a0 ) 1.314 nm, b0 ) 0.658 nm, c0 ) 1.079 nm, Z ) 8. The crystal structure of KTA (001) surface is shown in Figure 1. The structure consists of short string chains As(1)O4-Ti(2)O6 along [010] direction and long corner linked chains of TiO6 octahedra cross-linked by AsO4 tetrahedra forming a sequence As(2)O4-Ti(1)O6-As(2)O4-Ti(1)O6 along [001] direction. Both Ti(1)O6 and Ti(2)O6 octahedra are strongly distorted with bond lengths lying in the range Ti(1)-O: 178.9-209.9 pm and Ti(2)-O: 199.6-201.2 pm. * Corresponding author. E-mail: [email protected]. Tel: 734-7635344. Fax: 734-763-4690. † Department of Geological Sciences, University of Michigan. ‡ Institute of Semiconductor Physics. § Departments of Materials Science and Engineering & Nuclear Engineering and Radiological Sciences, University of Michigan. | Institute of Geology and Mineralogy. ⊥ Institute for Problems of Materials Science.

Figure 1. The schematic representation of the crystal structure of KTiOAsO4 (KTA) for the (001) plane.

The difference between longest and shortest bonds in Ti(1)O6 octahedra is 30 pm, which is slightly lower than that (3.6 pm) found in KTP. The tetrahedra AsO4 are near symmetrical with bond lengths in the narrow range 166.0-167.6 pm. Potassium ions occupy the positions in the Z-pores, possessing a high ionic conductivity in the (001) direction.1,5,9 This feature provides a possibility for an ionexchange reaction for (001) crystal plane resulting in substitution of Rb or Cs for K in the top surface layer, and optical waveguide formation due to refractive index increase.14 Ion-beam implantation/irradiation is an attractive method for the fabrication of waveguides and/or surface modification in

10.1021/jp0671392 CCC: $37.00 © 2007 American Chemical Society Published on Web 01/24/2007

Potassium Titanyl Arsenate (001) Crystal Surfaces many optical materials, including crystals, polymers, glassy materials, and semiconductors.15-17 Ion-beam irradiation is also a versatile method for surface or near-surface modification of crystalline materials in the electronics, magnetoelectronics, and optoelectronics industries.18-20 Recently, there has been considerable interest in modifying the optical properties of KTA and fabricating waveguides using ion-beam methods.5,8,21-23 However, the information concerning variations of electronic properties and the structure of KTA surfaces subjected to external physiochemical effects, such as ion bombardment, is limited. The ion-beam irradiation-induced modifications of the surface structure may result in significant variations of optical transparency, refractive indices, and optical damage thresholds. Therefore, a fundamental understanding of the physical and chemical effects induced by the ion-beam interaction with KTA surfaces is principal for further progress in fine-tuning the optical characteristics and also optical surface preparation by ion-beam cleaning and ion implantation waveguide technologies. Improved understanding of ion-induced effects will also enhance the ability to engineer optical crystal surfaces on the atomic scale with desired properties to meet the required applications in optoelectronics. In this paper, we report the low-energy Ar+ ion-irradiation-induced amorphization and chemical modification of the (001) KTA surfaces investigated using reflection highenergy electron diffraction (RHEED) and X-ray photoelectron spectroscopy (XPS). The Ar+ ions are typically employed in technical applications, such as surface cleaning/processing, near surface doping, elemental concentration-depth profiling, surface texturing, and ion lithography, and therefore it is particularly important to study the Ar+ ion interaction with KTA surfaces. Furthermore, it is expected that the results obtained in the present investigation of KTA will also be useful in comparing its behavior with that of other members of the KTP structural family in order to reveal the common effects of ion bombardment. 2. Experimental Section A. KTA Crystal Growth and (001) Surface Preparation. An optical grade crystal of KTA was grown on oriented seed by the high-temperature solution growth (TSSG) technique. A three-zone furnace, which is inside a water-cooled chamber equipped with an automatic control of water flow, was employed. The reagents were K2CO3 and TiO2. The solvent used was K5As3O10, which is obtained from K2CO3 and KH2AsO4 by mixing them directly in a growth crucible and annealing in a two-stage process at 260 and 700 °C. The required amount of KH2AsO4 and TiO2 were mixed into the furnace with fused solvent to grow the KTA crystal. The chemicals were homogenized at 1000 °C by mixing them thoroughly for 24 h. The crystal was grown on a seed which was orientated with a-axis perpendicular to the flux surface. Typical KTA sample grown is shown in Figure 2. The (001) surface was cut from the part of the crystal that was free of any optical inhomogeneities or domain boundaries. At the finishing stage of polishing, the nanodiamond powder, 0.1 grade, lubricated by water was applied. B. RHEED and XPS Measurements. The crystallographic surface properties of KTA (001) surfaces were investigated by RHEED at an electron accelerating voltage 50 kV. The chargeneutralization flood gun was utilized in order to eliminate charging effects. Elemental composition and surface electronic parameters were determined using XPS. The measurements were performed at room temperature to characterize the pristine KTA, immediately after surface preparation, and also as a function of

J. Phys. Chem. C, Vol. 111, No. 6, 2007 2703

Figure 2. Single crystal of KTiOAsO4 (KTA).

Ar+ ion irradiation on the KTA (001) surfaces. Crystal surface bombardment has been produced employing an Ar+ beam with energy of 1.5 keV at an ion current density of 11 µA/cm2 and incident angle 45°. Total Ar+ fluence was ∼2.1 × 1016 ions/ cm2. The ion-beam irradiation was performed, keeping the KTA at room temperature. The ion beam was rastered over the area 8 × 8 mm2, and the sputtering rate was estimated at 10 Å/min. All these measurements were carried out in an ion-pumped chamber of an ES-2401 spectrometer. The chamber base pressure was ∼1 × 10-7 Pa during XPS experiments. Monochromatic Mg KR (hν ) 1253.6 eV) excitation was used as a source of X-rays. The energy scale of the spectrometer was calibrated by setting the measured Au 4f7/2 binding energy to 84.0 eV with respect to the Fermi energy of a spectrometer energy analyzer. Energy drift due to charging effects was calibrated, taking C 1s (284.6 eV) line of hydrocarbons for the pristine KTA surface and F 1s (685.0 eV) line induced by specially adsorbed fluorine for the bombarded KTA surface. 3. Results The RHEED patterns obtained for the (001) KTA surfaces are shown in Figure 3. Images a, b shown are obtained immediately after chemically cleaning the KTA surface, and image c represents the pattern obtained after Ar+ ion irradiation. The dominant components in intensity are from the monocrystal streaks accompanied by wide Kikuchi lines in the far-field and a relatively weak diffuse background (Figure 3; images a, b). The single-crystal phase on the surface was identified as KTA. The background component appeared to be due to traces of mechanical damage caused during polishing. Superstructure ordering has been observed for the surface with the relations a ) a0, b ) 2b0 and c ) 2c0 where a, b, and c are the surface ordering lattice parameters. The effect on the surface structure of the bombardment with Ar+ ions was remarkable. The welldefined streaks progressively disappeared for the KTA surface irradiated with 1.5 keV Ar+ at a fluence of 2.1 × 1016 ions/ cm2. Only a background signal was detected in RHEED pattern for an Ar+ at a fluence of 2.1 × 1016 ions/cm2 (Figure 3; image c). These results suggest that geometric structural modifications

2704 J. Phys. Chem. C, Vol. 111, No. 6, 2007

Ramana et al.

Figure 4. XPS survey scans of KTA. Curves 1 and 2 represent the pristine and ion-bombarded KTA surfaces, respectively.

Figure 3. RHEED patterns for pristine and ion-beam-irradiated KTA surfaces. Images a, b represent the pristine surface, and image c represents that of the irradiated KTA surface. The incident electron beam is parallel to [100] for image a and [-110] for image b. Complete amorphization due to Ar+ ion irradiation is evident in image c.

have occurred on the KTA surface due to the Ar+ ion-beam irradiation, resulting in complete amorphization of KTA surface at a fluence of 2.1 × 1016 ions/cm2. XPS measurements revealed changes in the chemical bonds on the KTA surface upon Ar+ ion irradiation. The XPS survey scans obtained for KTA surfaces are shown in Figure 4. The XPS spectra of pristine (curve 1) and ion-beam irradiated (curve 2) KTA surfaces are shown for comparison. A strong C 1s corelevel line for the pristine KTA surface indicates the adventitious hydrocarbon contamination covering the surface layers. The bombardment of the KTA surface with Ar+ at a fluence of 2.1 × 1016 ions/cm2 results in complete removal of the adsorbed hydrocarbons from the surface and the appearance of peaks from

constituent elemental core-levels. The Ar+ ion-beam irradiation induced changes are clearly evident in XPS (curve 2), where the photoemission intensities increase in the 200-600 eV region, and two new Auger transitions in Ti and K (Ti LMM and K LMM) appear in the higher binding energy (BE) region. The detailed scans of core-level photoemission peaks of O 1s, Ti 2p, and K 2p have been used to identify the changes in chemical composition and chemical bonding on the KTA surfaces. The core-level photoemission spectra of O 1s, Ti 2p, K 2p, As 3d, and As 3p levels obtained for the KTA crystal surfaces before and after ion irradiation are shown in Figures 5 and 6. Curves labeled as 1 and 2 represent the spectra obtained on the pristine and bombarded KTA surfaces, respectively. The measured BE of respective core-levels are presented in Table 1. The major changes observed as a function of Ar+ ion irradiation are in the core-level scans of O 1s, As 3p, and As 3d. The two important observations from the O 1s corelevel spectra (Figure 5a) are as follows: (1) the O 1s peak is slightly broader and asymmetrical for the pristine KTA surface; (2) there is a shift of the O 1s line maximum to lower BE associated with a decrease in FWHM of the peak. The slightly broader nature of the peak with a FWHM of ∼2.3 eV for the pristine KTA can be attributed to adsorbed species on the surface, but the latter change with a significant decrease in BE from 531.2 to 530.5 eV (∆EO 1s ) 0.7 eV) associated with a decrease in FWHM from ∼2.3 eV to ∼1.6 eV is a clear indication of a drastic redistribution of oxide chemical bonds in the top surface layer as modified by the interaction with 1.5 keV Ar+ ions. The detailed scan of Ti 2p core-level is shown in the middle panel of Figure 5. The core level binding energy peaks due to Ti 2p1/2 and Ti 2p3/2 are observed at 464.8 and 459.0 eV, respectively (Table 1), for the pristine KTA surface. The doublet (Ti 2p-levels), due to spin-orbit splitting, is characterized by separation energy ∆ETi 2p)5.8 eV. No significant changes are observed in the BE of Ti 2p doublet for KTA surface bombarded with Ar+ ions even at the highest dose of 2.1 × 1016 ions/cm2. The measured values for the irradiated KTA surface are Ti 2p1/2 and Ti 2p3/2 at 464.7 and 458.9 eV; ∆ETi 2p)5.8 eV. The behavior of K 2p core-level (Figure 5c) is found to be quite similar to that of Ti 2p with no observed changes except

Potassium Titanyl Arsenate (001) Crystal Surfaces

J. Phys. Chem. C, Vol. 111, No. 6, 2007 2705

Figure 5. Core-level photoemission spectra of O 1s, Ti 2p, and K 2p in the KTA crystal. (a) X-ray photoelectron O 1s core-level spectrum, (b) X-ray photoelectron Ti 2p core-level spectrum, and (c) X-ray photoelectron K 2p core-level spectrum. Curves 1 and 2 represent the pristine and ion-bombarded KTA surfaces, respectively, in all of them.

TABLE 1: Binding Energies of Core-Level Photoemission and Auger Lines for Pristine and Ar+ Ion-Irradiated KTA Surfaces (Uncertainty is ( 0.1 eV)

Figure 6. X-ray core-level photoemission spectra of (a) As 3p and (b) As 3d. Curves 1 and 2 represent the pristine and ion-bombarded KTA surfaces, respectively.

increased peak intensity. The peaks due to K 2p1/2 and K 2p3/2 are observed at 295.6 and 292.8 eV, respectively (Table 1), for

core-level photoemission peak, Auger Line

pristine KTA

Ar+ ion-irradiated KTA

O 2s K 3s Ti 3p As 3d Ti 3s As 3p C 1s K 2p3/2 K 2p1/2 Ti 2p3/2 Ti 2p1/2 O 1s O KLL Ti LMM K LMM

22.4 32.8 37.0 45.3 62.5 146 284.6 292.8 295.6 459.0 464.8 531.2 742.4

22.1 33.1 37.0 41.6 62.3 140.6 293.0 295.7 458.9 464.7 530.5 742.2 840.3 874.1

the pristine KTA surface. The K 2p doublet, due to spin-orbit splitting, are separated by ∆EK 2p)2.8 eV. The measured values for the irradiated KTA surface are K 2p1/2 and K 2p3/2 at 295.7 and 293 eV; ∆EK 2p)2.7 eV. The detailed core-level photoemission scans of As 3p for the pristine (curve 1) and ion-irradiated samples (curve 2) are shown in Figure 6a. Major changes can be seen (Figure 6a) for the As 3p region as a result of Ar+ ion irradiation of the (001) KTA surface. The As 3p3/2 is observed at ∼144 eV for the pristine KTA. A drastic decrease in peak intensity associated with a peak shift for As 3p occurs as result of Ar+ ion irradiation. The peak structure is completely disappeared at the highest dose of 2.1 × 1016 Ar+ ions/cm2 with only a minor contribution at 140.5 eV. The chemical shift δEAs 2p3/2 (As 3p3/2 BE of pristine KTA-As 3p3/2 BE of ion-irradiated KTA) is 5.5 eV. The evolution of the shape and intensity of the XPS As 3d level of the KTA surface is identical to the behavior of As 3p corelevel due to ion irradiation as shown in Figure 6b. The BE of As 3d measured for the pristine KTA surface is at 45.3 eV. Similar to As 3p, a drastic decrease in the As 3d peak intensity can be seen for the irradiated KTA surface. The minor As 3d peak for the irradiated KTA surface is found at 41.6 eV. The chemical shift δEAs 3d (BE As 3d pristine-BE As 3d irradiated) is 3.7 eV.

2706 J. Phys. Chem. C, Vol. 111, No. 6, 2007 4. Discussion The emphasis in this work is to characterize the structure and chemical bonding in as-grown KTA crystal surfaces, the effect of low-energy Ar+ ion bombardment on the structural characteristics, and to compare the behavior of both pristine and irradiated KTA surfaces with other crystals of the KTP family. For this purpose, first we discuss the RHEED measurements to confirm structural quality of the KTA surfaces and the effect of Ar+ ion irradiation. Subsequently, the XPS results will be considered to establish the chemical quality of the grown KTA surfaces and to show that the effect of Ar+ ion irradiation is remarkable on the chemical bonds on the surface. While discussing the experimental results of this work, a comparison is made, wherever appropriate, with the other members of the isostructural KTP family. RHEED measurements provide evidence for the ordered structure of KTA surfaces. Superstructure ordering is noticed with the relations: a ) a0, b ) 2b0, and c ) 2c0 where a, b, and c are the surface ordering lattice parameters and a0, b0, and c0 are the lattice parameters of the bulk crystal. Superstructure formation has also been observed and reported for TlTiOPO4, KTiOPO4, and K0.77Ti0.77Sb0.23OPO4 crystals previously.24-26 It should be pointed that in all cases the ordering with only even superstructure indices has been found for the crystals related to the KTP family. However, the effect of Ar+ ion irradiation observed in KTA is completely different when compared to other crystals. The RHEED patterns clearly demonstrate that the ion-irradiation induced geometric structural modification, leading to complete amorphization of KTA surface at a fluence of 2.1 × 1016 ions/cm2. Similar behavior of surface amorphization due to Ar+ ion irradiation was noticed only in KGd0.95Nd0.05(WO4)2 (Nd:KGW) crystal surfaces.27 Comparison of the data indicates that the KTA surface is more sensitive to the Ar+ ion beam in terms of induced changes in the surface structure, which can be understood from drastic modification of chemical bonds as revealed by the XPS results. The detailed XPS scans of Ti 2p, K 2p, and As 3d corelevels provide important information on the chemical state of cations in the KTA crystal. First we consider the case of the pristine KTA surface to establish the chemical state of cations in the crystal. The Ti 2p3/2 peak at about 458.9 eV with a ∆ETi 2p of 5.8 eV are in excellent agreement with the literature reports28-37 and characterize the chemical state of Ti ions as Ti4+ in KTA crystal. In general, the KTP-type framework is constructed by linked TiO6 octahedra and PO4 (AsO4 in KTA) tetrahedra with covalent oxide bonds and KOn polyhedra with ionic bonds. When a metal interacts with oxygen, a great redistribution of electronic density occurs as a result of valence electrons shifting from metal to oxygen resulting in noticeable variations of electronic structure of inner shells of cation and anion. The effective displacement of valence electron density away from the atomic nucleus results in reduction of electrical screening of inner shells with increasing BE of inner electrons. This effect can be detected as a variation of core level BE in XPS by a few eV if we compare the electronic parameters of pure metals and fluorides for which the ionicity of chemical bonds is the highest.38 However, the difference is not so pronounced, and more sensitive parameters are necessary to quantify the effect in oxides. The energy difference ∆(O-M) between the representative metal core level and O 1s level is a more robust parameter for chemical bonding characterization.39-41 Additionally, the variation of the BE difference is more prominent because of opposite sign of the chemical shifts of BEs of metal and oxygen core levels when valence electron

Ramana et al. TABLE 2: Comparison of BE and ∆(O-Ti) of Ti 2p Core-Level in KTA with Titanium Oxides and Other Members of the KTP Crystal Family binding energy (eV) material

O 1s

Ti 2p3/2

∆(O-Ti)

reference

KTiOAsO4 (pristine) KTiOPO4 TlTiOPO4 KTi2(PO4)3 Na2Ti6O13 TiO2 (nanoparticle) TiO2 (spray coating) TiO2 (anatase) KTiOAsO4 (Ar+ irradiated)

531.2 530.9 531.1 531.4 530.3 530.4 530.1 530.0 530.5

459.0 458.4 458.9 459.4 458.1 458.8 458.7 458.6 458.9

72.2 72.5 72.3 72.0 72.2 71.6 71.4 71.4 71.6

this work 26 26 35 36 33 34 37 this work

transfer occurs from metal to oxygen ions. The electron density shift of a bond formation can be characterized by mean value of the chemical bond length L(M-O). Therefore, the chemical nature of KTA surface may be elucidated by the parameter, ∆(O-Ti), which is the BE difference BE(O 1s) BE(Ti 2p3/2), as a characteristic parameter independent of the surface charging effects.39 A comparison of this parameter obtained for KTA with the reported ∆(O-Ti) values of other crystals of the KTP family and titanium oxides is presented in Table 2. The ∆(O-Ti) parameter is computed for other titanium oxides based on the reported values in the literature. It is evident that ∆(O-Ti))72.2 eV obtained for the pristine KTA surface is in the range typical for TiO44+ groups in the KTP type crystals (Table 2). The K 2p core-level spectrum (Figure 5c) indicates the K+ ions in the pristine KTA crystal surface. The core level binding energy peaks due to K 2p1/2 and K 2p3/2 are observed at 295.6 and 292.8 eV, respectively (Table 1), for the pristine KTA surface. The BE of K 2p core-level is reported for metallic K at 294.4 eV (2p3/2) and 297.2 eV (2p1/2).38 The K 2p doublets have been observed at 292.9 eV (2p3/2) and 295.6 (2p1/2) for KCl.38 Therefore, our XPS data can be attributed to the monovalent potassium ions in KTA. The appearance of As 3p3/2 at ∼146 eV and, particularly, As 3d at 45.3 eV indicate that As in the pristine (001) KTA surfaces is present mainly as As5+. Arsenic exhibits a wide range of oxidation states (-1 to 5+) and an extensive literature is available on XPS-based chemical characterization of As-bearing compounds.42-50 A closer examination and comparison of the literature data, as presented in Table 3, reveals that the As 3d BE occurs at: 45.3-45.6 eV for As5+, 43.9-44.8 eV for As3+, 43.0-43.1 eV for As2+, 41.8-42.0 eV for As0, and 41.2-41.3 for As-1. The As 3d peak usually shows shoulder contributions if multiple valence states (major/minor component) of As are present.42-50 Therefore, comparison of As 3d BE at 45.3 eV and the absence of shoulder contributions clearly indicate the existence of As5+ and rules out the possibility of coexistence of multiple valence states in pristine KTA. We now consider the observed changes in KTA surfaces due to 1.5 keV Ar+ ion-beam irradiation. RHEED analysis indicates the formation of altered layers with complete amorphization on the surface. The observation that there is no shift in BE of Ti 2p doublet and no shoulder formation in the Ti 2p3/2 component on the lower BE side clearly indicates that there is no change in chemical state of Ti. It is quite interesting to see that the interaction of the (001) KTA surfaces with Ar+ ions does not generate reduced Ti that is typically observed for KTP family crystals and other titanates,29-31,51-53 and practically all Ti remain in Ti4+ state. The detailed XPS scans of As 3p doublet and As 3d establishes the chemical state of arsenic as As5+ for

Potassium Titanyl Arsenate (001) Crystal Surfaces

J. Phys. Chem. C, Vol. 111, No. 6, 2007 2707

TABLE 3: Comparison of BE of 3d Core-Level and Chemical State of Arsenic material KtiOAsO4 (pristine) As2O5 on GaAs Oxidized FeAsS Oxidized GaAs As2O5 (bulk) Oxidized GaAs Oxidized FeAsS As2O3 (bulk) As2O3 AsS, As4S4 AsS on S-doped GaAs As4S4 Elemental As As in As-doped GaAs As in S-doped GaAs As in KTiOAsO4 (ion-irradiated) FeAsS NiAsS

As 3d BE (eV)

oxidation state of As

reference

45.3 45.3 45.3 45.6 46.0 44.8 44.4 44.4 44.4 43.0 43.0 43.0 41.6 41.9 41.9 41.6

5+ 5+ 5+ 5+ 5+ 3+ 3+ 3+ 3+ 2+ 2+ 2+ 0 0 0 0

this work 44 47 49 50 49 47 50 42 50 48 42, 43 42 48 48 this work

41.2 41.3

-1 -1

47 46

the pristine (001) KTA surface. It is evident from the As corelevel scans that the irradiation of the surface with 1.5 kV Ar+ ions almost completely removes Ar5+ from the surface. A minor component of As 3p at 140.5 eV and As 3d at 41.6 eV is in good agreement with arsenic in the As0 state and indicates that the remaining minor contributions in the surface region are due to mainly metallic As. On the other hand, ∆(O-Ti) ) 71.6 eV estimated for bonding in the altered surface layers is within the typical range for TiO2. The most prominent effect observed in this experiment was the near surface amorphization (from RHEED analysis) and associated broadening of core-level peaks. Therefore, the amorphous compound on the surface is, perhaps, based on TiO44+ tetrahedra with a mean chemical bond length L(Ti-O) ∼ 196.4 pm similar to that in TiO2 (Table 3). This suggests that the altered surface layer contains amorphous TiO2 as the dominant chemical component, while most of the As is removed by preferential sputtering, on the ion-irradiated KTA surface. The results of the present work suggest that the drastic variation in the surface chemistry due to Ar+ ion irradiation is due to the fact that the surface chemical bonds are highly sensitive to the Ar+ ion irradiation, which induces a preferential loss of As and O leading to the formation of As0 states in the altered surface layers. Finally, we present a simple model for understanding the Ar+ ion-induced amorphization and chemical modification of the KTA crystal surfaces. Alteration of the structure and chemical composition of materials due to energetic ions or neutral particles is the most important concern in physical chemistry of surfaces/interfaces. The mechanism of ion-induced alteration of single and/or binary component oxides is well established.29-31,51-55 The Ti-bearing oxides have been particularly investigated in the past to investigate the effect of ions, such as helium, argon, and oxygen, and neutral particles since these materials exhibit various stoichiometric states and multiple valence states of Ti.29-31,50-54 Formation of reduced states, such as Ti3+, Ti2+, and even metallic Ti, on the surfaces of TiO2 and other Ti-bearing simple oxides due to ion beam-induced alteration, depending on the conditions, of surface layers is wellknown.51-54 Chemical changes are usually attributed to the preferential sputtering. However, the mechanism of multicomponent oxide crystals in not well-developed and requires further attention. The ion-beam-induced compositional changes of multicomponent oxides, such as KTA, are not only governed by the sputtering yields of the different constituents of the target

but also by other ion-beam-induced effects like diffusion, redistribution, segregation, etc. By this mechanism there is a tendency to reach one of the most chemically stable states in the ion-irradiated altered surface layer. During irradiation, the energy of the impinging Ar+ ions is distributed within a layer thickness determined by their penetration depth. They generate cascades of energetic atoms. Therefore, the complete amorphization of the KTA surfaces is due to the primary 1.5 keV Ar+ ions. The decomposition usually occurs at the near surface layers, and the elements which are volatile or those with bonds which have relatively low free energies of formation will be more readily lost from the surface.55 Therefore, the preferential loss of As from the surface can be easily understood because it is volatile and the formation energy of bonds (301 kJ/mol) is less than that of bonds (364 kJ/mol).56 Therefore, the decomposition occurs when the dissipated energy due to irradiation exceeds the bond dissociation energies of and bonds on the KTA surface. However, under nonsteady conditions of primary ion interaction, preferential loss of As is more favorable as the volatility of arsenic increases during the bond dissociation (or at higher temperatures within the local area).57,58 As a result, more arsenic will be removed form the surface compared to Ti. The remaining As prefers to be in elemental state, in view of very low surface energy (0.633 J m-2, when compared to Ti with 2.57 J m-2),59,60 to provide the most stable configuration to the altered surface layers by having minimum surface energy. Furthermore, the surface energy of As (0.48 Jm-2 at 1267 K) is significantly less than that of Ti (1.723 J m-2 at 1933 K) at the conditions of melting.59 The XPS results are, in fact, consistent with this approximation based on the bond and surface energies. Formation of Pb0 and metallization on PbTiO3 compared to NiTiO3, due to Pb loss, as reported by Leinen et al. also provide evidence in support of this approximation. Therefore, the chemical effect due to ion irradiation is the formation of a layer on the KTA surfaces with altered bonds, along with elemental arsenic, which is found in amorphous TiO2. 5. Summary and Conclusions The effect of 1.5 keV Ar+ ion bombardment on the (001) KTA optical surfaces at room temperature was studied using RHEED and XPS. The surface structure and chemical composition of the KTA surface is quite sensitive to the Ar+ ion bombardment. The long-range order of the KTA surface is lost during irradiation, leading to the formation of completely amorphous surface layers at an ion dose of 2.1 × 1016 ions/ cm2. The bonds in the KTA crystal lattice are less stable and are sensitive to the ion irradiation. The As5+ ions are reduced, resulting in the formation of As0 states and partial arsenic loss from the top surface as result of Ar+ ion bombardment. The formation of an unstable layer with chemically active arsenic states may be a factor reducing the durability of nonlinear devices. Acknowledgment. The authors at the University of Michigan acknowledge the support of the National Science Foundation (NSF-NIRT, EAR-0403732). References and Notes (1) Merkulov, A. A.; Isaenko, L. I.; Belov, A. I. J. Struct. Chem. 2001, 42, 610. (2) Hagerman, M. E.; Poeppelmeier, K. R. Chem. Mater. 1995, 7, 602. (3) Pack, M. V.; Armstrong, D. J.; Smith, A. V. Appl. Opt. 2004, 43, 3319.

2708 J. Phys. Chem. C, Vol. 111, No. 6, 2007 (4) Jiang, Y.; Wang, K. M.; Wang, X. L.; Jia, C. L.; Wang, L.; Jiao, Y.; Lu, Q. M.; Ma, H. J.; Nie, R.; Shen, D. Y. Appl. Phys. Lett. 2006, 88, 011114. (5) Loiacono, G. M.; Loiacono, D. N.; Zola, J. J.; Stolzenberger, R. A.; McGee, T.; Norwood, R. G. Appl. Phys. Lett. 1992, 61, 895. (6) Feve, J. P.; Boulanger, B.; Pacaud, O.; Rousseau, I.; Menaert, B.; Marnier, G.; Villeval, P.; Bonnin, C.; Loiacono, G. M.; Loiacono, D. N. J. Opt. Soc. Am. B 2000, 17, 775. (7) Hansson, G.; Karlsson, H.; Wang, S.; Laurell, F. Appl. Opt. 2000, 39, 5058. (8) Liu, X.; Xia, Y.; Lu, Q.; Li, F.; Tan, Z.; Xia, H.; Song, C.; Xu, X. Mater. Sci. Eng. B 2005, 117, 355. (9) Cheng, L. K.; Bierlein, J. D. Ferroelectrics 1993, 142, 209. (10) Boulanger, B.; Fe`ve, J. P.; Marnier, G.; Me´naert, B. Pure Appl. Opt. 1998, 7, 239. (11) McGowan, C.; Reid, D. T.; Ebrahimzadeh, M.; Sibbett, W. Opt. Commun. 1997, 134, 186. (12) Kraemer, D.; Hua, R.; Cowan, M. L.; Franjic, K.; Dwayne Miller, R. J. Opt. Lett. 2006, 31, 981. (13) Mayo, S. C.; Thomas, P. A.; Teat, S. J.; Loiacono, G. M.; Loiacono, D. N. Acta Crystallogr. B 1994, 50, 655. (14) Atuchin, V. V.; Plotnikov, A. E.; Ziling, C. C.; Isaenko, L. I.; Tjurikov, V. I. Proc. SPIE-Int. Soc. Opt. Eng. 1996, 2969, 261. (15) Townsend, P. D.; Chandler, P. J.; Zhang, L. Optical Effects of Ion Implantation; Cambridge University Press: Cambridge, UK, 1994. (16) Lian, J.; Rios, S.; Boatner, L. A.; Wang, L.; Ewing, R. C. J. Appl. Phys. 2003, 94, 5695. (17) Atuchin, V. V. Nucl. Instr. Methods Phys. Res. B 2000, 168, 498. (18) (a) Ramana, C. V.; Choi, B. S.; Smith, R. J. Appl. Surf. Sci. 2003, 214, 338. (b) Ramana, C. V.; Choi, B. S.; Hutchison. R.; Smith, R. J. Nucl. Instr. Methods Phys. Res. B 2003, 212, 465. (19) Lian, J.; Helean, K. B.; Kennedy, B. J.; Wang, L. M.; Ewing, R. C. J. Phys. Chem. B 2006, 110, 2343. (20) Zhu, S.; Sun, K.; Zhang, Q. Y.; Zu, X. T.; Wang, L. M.; Ewing, R. C.; J. Appl. Phys. 2003, 94, 5648. (21) Chen, F.; Wang, X.-L.; Fu, G.; Li, S.-L.; Lu, F.; Wang, K.-M.; Lu, Q.-M.; Shen, D. Y. Nucl. Instrum. Methods Phys. Res. B 2002, 194, 355. (22) Wang, F. X.; Chen, F.; Wang, X. L.; Lu, Q. L.; Wang, K. M.; Shen, D. Y.; Ma, H. J.; Nie, R. Nucl. Instrum. Methods Phys. Res. B 2004, 215, 389. (23) Liu, X. D.; Xia, Y. Y.; Li, F.; Tan, Z. Y.; Zhao, M. W.; Xu, X. G.; Huang, B. Appl. Surf. Sci. 2003, 228, 77. (24) Atuchin, V. V.; Pokrovsky, L. D.; Kesler, V. G.; Maklakova, N. Yu.; Voronkova, V. I.; Yanovskii, V. K. Surf. ReV. Lett. 2004, 11, 191. (25) Atuchin, V. V.; Maklakova, N. Yu.; Pokrovsky, L. D.; Semenenko, V. N. Opt. Mater. 2003, 23, 363. (26) Atuchin, V. V.; Alekseeva, O. A.; Kesler, V. G.; Pokrovsky, L. D.; Sorokina, N. I.; Voronkova, V. I. J. Solid State Chem. 2006, 179, 2349. (27) Atuchin, V. V.; Kesler, V. G.; Maklakova, N. Yu.; Pokrovsky, L. D.; Sheglov, D. V. Eur. Phys. J. B 2006, 51, 293. (28) Liang, Y.; Gan, S. P.; Chambers, S. A.; Altman, E. I. Phys. ReV. B 2001, 63, 235402. (29) Jun, J.; Shin, J. H.; Dhayal, M. Appl. Surf. Sci. 2006, 252, 3871. (30) Hashimoto, S.; Murata, A.; Sakurada, T.; Tanaka, A. Surf. Interface Anal. 2002, 9, 459.

Ramana et al. (31) Padaletti, G.; Cusma`, A.; Viticoli, M.; Ingo, G. M.; Mezzi, A.; Watts, B. Mater. Sci. Eng. B, 2004, 109, 104. (32) Sodergren, S.; Siegbahn, H.; Rensmo, H.; Lindstrom, H.; Hagfeldt, A.; Lindquist, S.-E. J. Phys. Chem. B 1997, 101, 3087. (33) Song, Z.; Hrbek, J.; Osgood, R. Nano Lett. 2005, 5, 1327. (34) Tomaszek, R.; Pawlowski, L.; Gengembre, L.; Laureyns, J.; Znamirowski, Z.; Zdanowski, J. Surf. Coatings Technol., 2006, 201, 45. (35) Shpak, A. P.; Korduban, A. M.; Trachevskii, V. V.; Slobodyanik, N. S. Theor. Exp. Chem. 2000, 36, 241. (36) Meng, X. D.; Wang, D.; Liu, J.; Zhang, S. Mater. Res. Bull. 2004, 39, 2163. (37) Richter, J. H.; Henningsson, A.; Karlsson, P. G.; Andersson, M. P.; Uvdal, P.; Siegbahn, H.; Sandell, A. Phys. ReV. B 2005, 71, 235418. (38) Briggs, D.; Seach, M. P. Practical Surface Analysis by Auger and X-ray Photoelectron Spectroscopy; Wiley: New York, 1983. (39) Atuchin, V. V.; Kesler, V. G.; Pervukhina, N. V.; Zhang, Z. J. Electron Spectrosc. Relat. Phenom. 2006, 152, 18. (40) Fukuda, Y.; Nagoshi, M.; Suzuki, T.; Namba, Y.; Syono, Y.; Tachiki, M. Phys. ReV. B 1989, 39, 11494. (41) Mahony, L. O.; Curtin, T.; Zemlyanov, D.; Mihov, M.; Hodnett, K. J. Catal. 2004, 227, 270. (42) Fantauzzi, M.; Atzel, D.; Elsener, B.; Lattanzi, P.; Rossi, A. Surf. Interface Anal. 2006, 38, 922. (43) Kyono, A.; Kimata, M.; Hatta, T. Am. Miner. 2005, 90, 1563. (44) Ghita, R. V.; Negrila, C.; Manea, A. S.; Logofatu, C.; Cernea, M.; Lazarescu, M. F. J. Optoelect. AdV. Mater. 2003, 5, 859. (45) Bullen, H. A.; Dorko, M. J.; Oman, J. K.; Garrett, S. J. Surf. Sci. 2003, 531, 319. (46) Jackson, D. A.; Nesbitt, H. W.; Scaini. M. J.; Duggal, A.; Bancroft, G. M. Am. Miner. 2003, 88, 890. (47) Nesbitt, H. W.; Muir, I. J.; Pratt, A. R. Geochim. Cosmochim. Acta 1995, 59, 1773. (48) Moriatry, P.; Murphy, B.; Roberts, L.; Cafolla, A. A.; Hughes, G.; Koenders, L.; Bailey, P. Phys. ReV. B 1994, 50, 14237. (49) Flinn, B. J.; McIntyre, N. S. Surf. Interface Anal. 1990, 15, 19. (50) Schafuss, A. G.; Nesbitt, H. W.; Scaini, M. J.; Hoechst, H.; Bancroft, M. G.; Szargan, R. Am. Miner. 2000, 85, 1754. (51) Sullivan, J. L.; Saied, S. O.; Bertoti, I. Vacuum 1991, 42, 1208. (52) Ivanova, O. P.; Naumkin, A. V.; Vasilyev, L. A. Vacuum 1995, 46, 363. (53) Leinen, D.; Fernandez, A.; Espinos, J. P.; Gonzalez-Elipe, A. R. Surf. Interface Anal. 1993, 20, 941. (54) Atuchin, V. V.; Kesler, V. G.; Maklakova, N. Y.; Pokrovsky, L. D.; Semenenko, V. N. Surf. Interface Anal. 2002, 34, 320. (55) Ho, S. F.; Contarini, S.; Rabalais, J. W. J. Phys. Chem. 1987, 91, 4779. (56) Cottrell, T. L. The Strengths of Chemical Bonds, 2nd ed.; Butterworths: London, 1958. (57) Barton, R. G.; Clark, W. D.; Seeker, W. R. Combust. Sci. Technol. 1990, 74, 327. (58) Gray, D. B.; Watts, F.; Overcamp, T. J. EnViron. Eng. Sci. 2001, 18, 1. (59) Mezey, L. Z.; Giber. J. Jpn. J. Appl. Phys. 1982, 21, 1569. (60) Ramana, C. V.; Choi, B. S.; Smith, R. J. Surf. Sci. 2004, 551, 189.