Superficial oxidation of molybdenum at high pressure and low temperature: RHEED and AES analyses of the molybdenum oxide formation

Superficial oxidation of molybdenum at high pressure and low temperature: RHEED and AES analyses of the molybdenum oxide formation

Surface 1044 Science 251,252 (1991) 1044-1051 North-Holland Superficial oxidation of molybdenum at high pressure and low temperature: RHEED and AE...

2MB Sizes 0 Downloads 49 Views

Surface

1044

Science 251,252

(1991) 1044-1051 North-Holland

Superficial oxidation of molybdenum at high pressure and low temperature: RHEED and AES analyses of the molybdenum oxide formation N. Floquet and 0. Bertrand

Received

1 October

1990: accepted

for publication

7 December

1990

Numerous studies relate to the interaction of the molybdenum surface with oxygen at low pressure and high temperature. They give results about oxygen chemisorption. surface facetting and the epitaxial formation of MoOZ crystallites. This work deals with the interaction of Mo(100). Mo(ll0) and Mo(ll1) surfaces with oxygen at high pressure (10” Pa) and low temperature (620-820 K). RHEED and AES analyses results prove that. in these oxidation conditions: Moo, and neon-st(~ichiomet~c molybdenum oxide such as Mo,O,, are not evidenced in any of the molybdenum oxidation steps. The Moo, phase nucleates directly from any Mo surface. The structure and orientation of MOO, nuclei are characterized in relation with the parent molybdenum surface. Some MOO, nucleation and growth mechanisms are proposed. involving the specific properties of the MOO, phase and the structure of molybdenum surfaces

Numerous investigations relate to the interaction of molybdenum surfaces with oxygen at low pressure and high temperature. They give detailed results about oxygen chemisorption, surface facetting and the epitaxial formation of MOO, crystallites [l-22]. Only, some results of a recent work [16] establish that for long oxygen exposure. amorphous MOO, (presumably disordered) and intermediate oxides are formed, mixed with MOO, crystallites on the Mo(l12) faces created on a Mo( 111) surface. Other investigations do not relate more, about MoOj or intermediate oxide formation, because the MOO, phase is highly volatile in these particular oxidation conditions. Earlier thermodynamical studies [23,24] and our recent works [35,36] on the oxidation of molybdenum foils investigated in a temperature range from 733 to 953 K and in a high oxygen pressure (2.7 X 10’ and 8 X lo4 Pa), prove that the oxidation of molybdenum foils leads to MOO, oxide scales, with a preferential orientation of the crystalhtes. 0039~6028/91/$03.50

‘1’ 1991 - Elsevier Science

Publishers

In view of these previous data, the purpose of the present investigation is two-fold. The fundamental aim is to characterize the mechanisms of nucleation and growth of the MOO, oxide produced by interaction of oxygen with a molybdenum single crystal surface. The applied aim is to control the elaboration of the MOO, crystals for catalytic applications, considering that the catalytic properties of MOOR oxide are known to be sensitive to the superficial structures of the exposed faces of MOO, crystals.

2. Oxidation

experiments

Experimental oxidation conditions are closely linked to the therm(~dynanli~ai data for the molybdenum oxygen system 137-401. Schematically. under about 10’ Pa oxygen pressure 131,321. from 673 K molybdenum undergoes oxidation into molybdenum oxide solids. No volatile oxides are formed. From 773 K. the sublimation of MOO, solid is beginning. The most prominent species are

B.V. (North-Holland)

N. Floquet, 0. Bertrand / Superficial oxidation of MO at high pressure and low temperature

ranges at about 2.7 X lo3 Pa, both the phenomena of molybdenum oxide solid formation and the MOO, sublimation will occur. Mo(lOO), Mo(ll0) and Mo(ll1) surfaces were

the MOO, trimerics [41,42]. From 1073 K, the liquefaction of MOO, solid is arising. In the explored temperature range choosen between 620 and 820 K and the oxygen pressure

I

I

d Fig. 1. RHEED

patterns from a M@llO) surface exposure time: (a) initial Mo(ll0)

1045

Mo[ljO]

Mo[llO]

Mo[OOl] oxidized at 720 K temperature and 2.7 X lo4 Pa oxygen pressure surface, (b) after 5 min. (c) after one hour, (d) after several hours.

for increasing

1046

N. Floquet, 0. Bertrand

/ Superficial

oxidation of Mu ut htgh pressure und lm rrmperuturr

exposed to oxygen in the reaction chamber in the following way: oxygen gas first introduced at 2.7 x 103 Pa, then the crystal surface was heated by means of the calibrated infrared reflector. After exposure, heating was stopped and gas was rapidly evacuated. The sample was then transferred in the UHV chamber for AES and RHEED analyses. successively.

3003

Moo3

Whiskers

Rods

IlOO

,~ MD rough

surface al

MO facetted

surface

b) Ma03 Thick i-cds

3. Results and discussions The changing surface structures were monitored by reflection high-energy efectron diffraction (RHEED) and the variations of surface composition and metal oxidation states were measured by Auger electron spectroscopy (AES) peak ratios and peak shapes. RHEED and AES analyses of the oxidized surface are realized after each exposure lasting from several minutes to a few hours. 2.1. Oxidution on the Mo(l IO) surface After the first oxygen exposure at 720 K (P,,_ = 2.7 x lo3 Pa), the Auger spectra change markedly. which characterize the MOO, formation: M,,?VV peak splitting, new peak at 183 eV lower than the M,,,N?,.;V peak by about 4 eV. The peak height ratio Mo(220)/Mo(187) decreases to 0.56 and the ratio 0(5lO)/Mo(220) increases to 4.1. For further exposures no noticeable changes occur in these Auger spectra. RHEED patterns (fig. 1) evidence the formation of MOO, crystallites on the Mo(ll0) surface by successive structural steps. At the first exposure, streak patterns of the M~llO) surface disappear and a set of thin rings of uniform intensity appear, at any azimuth (fig. 1b). This set of rings corresponds to the (100). (110). (210) and (230) rings of the orthorhombic MOO, structure. It does not correspond to the strong rings of the MOOR structure that are (021) (110) and 1040) rings. It is interpreted as the formation of MoO,[lOO] whiskers in random orientation on the Mo(170) surface (fig. 2a). After one hour exposure. the RHEED patterns change and become differentiated for specific azimuths of the initia1 Mo(ll0) surface (fig. lc):

cl Fig. 7. Schematic the Mo(l10) whiskers normal interface.

pictures of the three main oxidation

surface: (a) formation

onto

the rough surface.

to the surface growth

facets.

of thick

(c)

of perpendicular (h) growth far

MoOJlOO]

normal to the Mo(l10)

from

steps of

MoOz[lOO]

of the whiskers the metal -oxide

rods. well structured. surface.

previous MOO, rings become of modulated intensity and more or less diffuse arcs of new MoQ, rings appear. The intensity maxima are arranged in three networks of parallel lines which have the same MoO,[lOO] periodicity (4 A) and are parallel to the Mo(lO0). Mo(lO1) and Mo(112) faces of the MO crystal, respectively. These diffraction patterns are interpreted as the growth of M00~[100] rods perpendicularly to each Mo( loo), Mo( 101) and Mo(112) facets simultaneously formed onto the Mo(lI0) surface during the oxidation process (fig. 2b). Further exposures, up to 8 h and more. result in a final modification of the RHEED patterns. The intensity maxima of the MOO, rings are rather arranged on a set of lines, with MoO,[lOO] periodicity and parallel to the Mellon surface. The patterns in the Mo[OOl] and Mo[l 101 azimuths can be distinguished (fig. Id). These last diffraction features characterize MOO, rods whose elongated MoO,[lOO] axes grow perpendicular to the initial Mo(ll0) surface and M00~[001] rows are parallel to the Mo[liO] direction of the molybdenuln surface (fig. 2~): M00,(100),,,~~~Mo(l 10),,i,,,. Compared with the RHEED patterns of Moo, single crystals. all these observed patterns reveal that the structure of the grown MoOj crystallites is never well organized. Accurate analyses specify

N. Floquet, 0. Bertrand / Superficial oxidation of MO at high pressure and low temperature

their structural evolution: at first the MoO,[lOO] rows, oriented perpendicular to the molybdenum facets, are short range ordered in the perpendicular MoO,[OlO] row, which is preferentially epi-

1047

taxied along a metallic row of the molybdenum facet. During their growth, the MoO,[lOO] rows become oriented perpendicular to the initial molybdenum surface, while MoO,(OlO) planes are

T&x a

10001

11121

(0241

Mo[02i] Mo(lOO] Mo03[010]

Mo]OOl]

b

I



Mo03]100],

,

(

li+hkl$ ’

I



I

(0031

I0021



I

-t

!

ROOI

l001q%00)

I

'U

/202/ I

I1

'Ill21I

!2001

(390)

Mo[uvw] Fig. 3. RHEED patterns from a Mo(100) surface oxidized at 720 K temperature and 2.7 X lo4 Pa oxygen pressure for increasing exposure time: (a) initial Mo(100) surface, (b) after two hours, (c) after ten hours, (d) after several tens of hours.

1048

N. Floquet, 0. Bertrand / Superfkial oxidation of MO UI high pressure und Ion’ rm~prrarurc~

building up by the periodic arrangement MoO~[~OO] rows. The periodic stacking MoO~(Ol0) never to be seems obtained. _7._. 7 Oxidation

of these of these

on the Mo(lO0) surface

For the initial and further exposures, the AES analysis shows immediate and lasting formation of MOO, oxide as observed from the Mo(l10) surface. Arising in the first few minutes of exposure. the first modification of the RHEED patterns consists of the fading of diffraction streaks of the initial Mo(100) surface (fig. 3a) into a diffuse background intensity. After about a two hours exposure. two new superimposed diffraction patterns become apparent (fig. 3b): (1) The set of MoO,(hkO) arcs ordered into lines with MoO,[lOO] periodicity (4 A) and parallel to the Mo(100) surface. It characterizes the presence of the MoO~[l~] whiskers perpendicular to the Mo( 100) surface. (2) A double network of streaks, perpendicular and oriented in relation to the Mo(100) surface. These diffraction features are due to M00~(010) plates parallel to the Mo(lOO~ surface and oriented in the two equivalent relations (fig. 4a): MoO~(OlO~~~,~~~~ Mo( 100) [Oll]and[01i]. For a ten hours exposure, a new modification in the RHEED patterns (fig. 3c) arises: the set of MoO,( hk0) arcs disappears, while the double network of streaks characteristic of the MoO~(Ol0) plates fades. A new set of large MoO,( h01) and MoO,( hll) arcs ordered into parallel lines with the MoO,[lOO] periodicity (4 A) is visible at any azimuth. For both the Mo[Oll] and Mo[Oli] azimuth. these MOO, arcs intersect precisely the MoO~(Ol0) streaks, which are in the spaced MoO,[OOl] direction. These new diffraction features characterize perpendicular MoO,(OlO) plates, whose MoO,[lOO] axes are about 10” around the normal of the Mo(100) surface (fig. 4b). After an exposure of several tens of hours, RHEED patterns (fig. 3d) consist of a set of MoO,( hkl) rings with intensity modulations while the MoO,(OlO) streaks have disappeared. The intensity maxima of the previous MoO,( h01) and MoO,( hll) rings are still visible. This last evolution is interpreted by the growth

Moo3

Platelets

b)

/ MollOf_% Cl

Fig. 4. Schematic pictures of the three main oxidation htrp.4 01 the Mo(100) surface: (a) formation of perpendicular MoO,[lOO] whiskers and epitaxied MoO?(OlO) platelets onto the bmonth Mo(100) surface, (b) orthogonal orientation of the MoO,(UlO) platelets. (cf growth of thick MoO,(OlO) plates normal to the Mo( 100) surface.

in thickness of the MoO,(OlO) plates which are oriented perpendicular to the Mo( 100) surface in the relationship (fig. 4~): MoO,(OlO) I Mo( 100): MoO~[l~] at f fO” Mo[lOO]. 3.3. Oxidation

on the Mo( I I I) surfircr

Since the first exposure, Auger spectra show the formation of Moo, oxide as observed from the Mo(ll0) and Mo(lO0) surfaces. The RHEED patterns prove the formation ctl oxide nuclei of either amorphous structure or nanometric size by the fading of Mo( 111) diffraction streaks into a diffuse background intensity. This oxidized state of the surface is changing

N. Floquet, 0. Bertrand / Superficial oxidation of MO at high pressure and low temperature

1049

of lines with the MoO,[lOO] periodicity (4 A). For the three equivalent azimuths, Mo[liO], Mo[Oil] and Mo[iOl], the two networks of lines are tilted respectively at 35 “C and 20” C to the Mo(ll1)

after about three hours exposure, as characterized by the RHEED patterns in fig. 5. A set of MoO,( hkl) rings appears with more or less diffuse intensity. Intensified arcs order into two networks

I zonr

1 de

2.22A

10001 IliOl 1250)

Mohl?l

Mo[lTO]

I

Mo(lll]

Mo[lOi] Fig. 5. RHEED

patterns from a Mo(lll) surface oxidized at 720 K temperature and 2.7 X lo4 Pa oxygen pressure for increasing exposure time: (a) inital Mo(ll1) surface, (b), (c), (d) after three hours, along various azimuths.

1050

N. Floquet, 0. Bertrand

/ Superficial

oxidation of MO at hrgh pressure and low temperature

surface, that is parallel to the Mo(ll0) and Mo(112) and their equivalent facets. For the intermediate azimuths Mo[%l], Mo[lTl] and Mo[ll?$, they are symmetric and tilted at 35O to the Mo(ll1) surface, that is parallel to the Mo(llO), Mo(Ol1) and Mo(lO1) equivalent facets. These diffraction features are comparable to the previous ones observed from the M~llO) surface. They are interpreted by the formation of crystalline MoO,[lOO] rods growing perpendicular to the main Mo(llO), Mo(Ol1) and Mo(lO1) facets and the Mo(211), Mo(121) and Mo(112) facets produced onto the Mo(ll1) surface. No further evolution of this oxidized state of the molybdenum surface appears, even after several hours of exposure. 3.4. Conclusion From these results, some important the oxidation process emerge.

features

of

3.4. I. Chemical state of the metallic surface and steps of jorm~tion of MOO., oxide The MOO, formation as an intermediate step of MOO, formation has been previously reported in thermogravimetric studies on the oxidation of polycrystalline molybdenum foils by interaction of oxygen at about 104-lo5 Pa at 673-998 K [27]. On the whole, the results seem to prove that the MOO, oxide is formed directly by interaction of oxygen with molybdenum. The only molybdenum oxide grown by interaction of oxygen on a MO surface, at 720 K temperature and 2.7 X lo4 Pa oxygen pressure, is the stoicbiometric MOO, oxide. The MoOz dioxide and the non-stoichiomet~c molybdenum oxides (even the more stable Mo,O,, oxide) are not evidenced at any stage of the oxidation process. 3.4.2. Cqstalline orientation of the molybdenum surface, oriented nucieat~on and growth of the MOO, oxide 3.4.2.1. Structural relations within metal-oxide interface: epituxial factor in the oxide nucleation. The structural features of the first MOO, nuclei

which depend on the crystalline orientation of the molybdenum surface, express the specific properties of the metal-oxide interface. The Mo(100) surface produces oriented MoO,(OlO) plates, inducing its stabilization. Only in this case, there is a 2D interfacial relation between the metallic and the oxide surface. The epitaxial relations~ps between the oxygen network and the close-lying molybdenum rows in the interfacial planes could explain the formation of MoO,(OlO) plates oriented onto the Mo(100) surface as: MoO,(O1O)l,,l]]Mo(lOO)l,,,, or 10,il. The Mo(ll0) and Mo(ll1) surfaces generate MoO~[l~] whiskers normal to the simultaneously produced facets. A short range order of the MoO,[lOO] rows is always observed in the MoO,[OlO] direction lying onto the metallic surface. From the same structural considerations. it could be explained that the lamellar MoO~(Ol0) plane is not initially built. Indeed the close-lying MoO,[OOl] rows, the zig-zag rows in the MOO, (010) planes, do not fit any MO row in the interfacial molybdenum planes (8.5% mismatch for the best parametric a~eement). 3.4.2.2. Structural relations between the oxide nucleation and its growth: &namic factors of the During their growth, from any parent reaction. metallic surface, the oxide nuclei built up in the orthorhombic MOO, structure and orientate their lamellar MoO,(OlO) planes perpendicular to the metallic surface and their close-lying MoO,[OOl] rows onto close-lying MO rows of the metallic surface. The results obtained from the Mo(100) surface emphasize these growth features, because of an orthogonal reorientation of the initial MoO,(OlO) plates epitaxied onto the Mo(100) surface.

These structural features of nucleation of growth, and their relationships, allow to distinguish between features related to the metaloxide interfacial relations and features of the dynamics of the oxidation reaction: - the interfacial forces induce the non well-structured oxide nuclei. as MoO,[lOO] rows normal to the metallic surface. Some order of the MoO,[lOO] rows along MoO~[O~O] direction could be explain

N. Floquet, 0. Bertrand / Superficial oxidation of MO at high pressure and low temperature

by structural relationships into metal-oxide interface. - the preferential diffusion of oxygen along the lamellar MoO,(OlO) planes through the orthorhombic MOO, structure, induces their orientation normal to the reaction plane. Such an interpretation of the nucleation and growth mechanisms of the MOO, oxide produced by interaction of oxygen with a molybdenum surface explains the previous [35] and coming [36] results of the morphological and structural study of the MOO, oxide layer generated from polycrystalline molybdenum foils.

References VI T.W. Haas and A.G. Jackson,

J. Chem. Phys. 44 (1966) 2921. PI J. Ferrante and G. Barton, NASA Technical Note TN D-4735 (1968) 1449. [31 K. Hayek, H.E. Farnsworth and R.L. Park, Surf. Sci. 10 (1968) 429. [41 J.C. Tracy and J.M. Blakely, Surf. Sci. 13 (1969) 313. I51 H.K.A. Kan and S. Feuerstein, J. Chem. Phys. 50 (1969) 3618. 161 G.J. Dooley III and T.W. Haas, J. Chem. Phys. 52 (1970) 461; J. Vat. Sci. Technol. 7 (1970) 1570. [71 A.E. Lee and K.E. Singer, Proc. R. Sot. London 40 A 323 (1971) 523. PI D. Tabor and J.M. Wilson, J. Cryst. Growth 9 (1971) 60. [91 H.M. Kennett, A.E. Lee and J.M. Wilson, Proc. R. Sot. London A 331 (1972) 429. WI J. Bruckner, Kris. Tech. 9 (1974) 647. IllI R. Riwan, C. Guillot and J. Paigne, Surf. Sci. 47 (1975) 183. WI H.M. Kennett and A.E. Lee, Surf. Sci. 48 (1975) 591. 1131 E. Bauer and H. Poppa, Surf. Sci. 88 (1979) 31. P41 E.I. Ko and R.J. Madix, Surf. Sci. 109 (1981) 221. P51 E. Bauer and H. Poppa, Surf. Sci. 127 (1983) 243. (161 C. Zhang, M.A. Van Hove and G.A. Somojai, Surf. Sci. 149 (1985) 326. 1171 J.L. Grant. T.B. Fryberger and P.C. Stair, Surf. Sci. 159 (1985) 333. [18] Ts.S. Marinova, P.K. Stefanov and N. Neshev, Surf. Sci. 164 (1985) 196.

1191 C. Zhang,

1051

A.J. Gellman, M.H. Farias and G.A. Somojai, Mater. Res. Bull. 20 (1985) 1129. WI J.L. Grant, T.B. Fryberger and P.C. Stair, Appl. Surf. Sci. 26 (1986) 472. WI E. Minni and F. Werfel, Surf. Interface Anal. 12 (1988) 385. WI P.K. Stefanov and Ts.S. Marinova, Surf. Sci. 200 (1988) 26. v31 E.A. Gulbransen and W.S. Wysong, Metal. Tech. 14 (1947) 2144. and E.A. Gulbransen, Metal. Tech. 14 t241 J.W. Hickman (1947) 2144. 1251 B. Lustman, Metal. Prog. 57 (1950) 629. WI R.C. Peterson and W.M. Fassel, Jr., Tech. Report VI Army Ordnance. Contract DA.04.495, ORD 237 (1954). 1271 M. Simnad and A. Spimers, Trans. AIME 203 (1955) 1011. PI M. Gleiser, W.L. Larsen, R. Speiser and J.W. Spretnak, ASTM Spec. Publ. 171 (1955) 65. 1291 ES. Jones, J.F. Mosher, R. Speiser and J.W. Spretnak, Corrosion 14 (1958) 20. 1301 R.W. Bartlett and D.M. Williams, Trans. Metal Sot. AIME 212 (1958) 280. K.F. Andrew and F.A. Brassart, J. 1311 E.A. Gulbransen, Electrochem. Sot. 110 (1963) 952. Corrosion 26 (1970) 19. 1321 E.A. Gulbransen, [331 V.Ya. Kolot, V.I. Tatus, V.F. RybaIko, Ya. M. Fogel, V.V. Vodolazhchenko and V.M. Evseev, Sov. Phys. (Solid State) 13 (1972) 1275. (341 G.M. Raynaud, J. Mater. Sci. Lett. 3 (1984) 965. [351 B. Mingot, N. FIoquet, 0. Bertrand, M. Treilleux, J.J. Heizmann, J. Massardier and M. Abon, J. Catal 118 (1989) 424. [361 N. Floquet and 0. Bertrand, Oxidation of Metals and J. Solid State Chem., to be published. [371 B.A. Stasliewicz, J.R. Trucker and P.E. Snyder, J. Am. Chem. Sot. 78 (1956) 1553. E.L. Evans and C.B. Alcok (Pergamon, 1381 0. Kubaschewski, London, 1958). 1391 L. Kihlborg, Arkiv Kemi 21 (1963) 473. WI L.L.Y. Chang and B. Phillips, J. Am. Ceram. Sot. Vol. 52 (1969) 527. [411 J. Berkowitz, M.G. Inghram and W.A. Chupka, J. Chem. Phys. 26 (1957) 842. K. Amioka and G. [421 Y. Ikeda. M. Ito, I. M&no, Matsumoto, High Temp. Sci. 16 (1983) 1.