Nuclear Instruments and Methods in Physics Research B80/81 (1993) 348-351 North-Holland
8aam intaraetions with MaterialsRAtoms
High dose implantation of aluminium into iron H. Reuther a, O. NikojOV b,l, S. Kruijer b, R.A. Brand b and W . Keune b
° Forschunoszentruni Rassendorf e. V., Postfach 19, 0-8051 Dresden, Germany b Universität Duisburg, Laboratorium für Angt .vandte Phy~ik. ! otharstr. 1-21, W-4100 Duisburg, Germany
Al' ions were implanted into iron with doses between 1 x 10 17 and 1 x 10 1s cm -2 at 50, 100, and 200 keV. The samples were investigated by integral and depth selective conversion electron M8ssbauer spectroscopy . It was found that there is an abrupt transition from the magnetic to the nonmagnetic state if the ion dose exceeds a special value . The nonmagnetic phase is only formed at the upper surface !a-, er.
1 . Introduction
2 . Experimental
Iron-aluminium alloys are of some interest both for fundamental research and for technology . Especially the iron-rich alloys show interesting magnetic properties because there is a transition from ferromagnetic Fe 3 Al to nonmagnetic FeAl at 33% Al . Iron-aluminium alloys can be produced in surface layers by ion beam techniques especially by ion beam mixing [2-a] or by direct ion implantation . In ref.  ion *,nplantation of aluminium into iror. at different energies seed different doses was investigated by conversion electron Mössbauer spectroscopy (CEMS). At low doses (vp to 1 x 10 17 em -2 ) only small changt:!; were found . Besidee the normal six-line pattern of a-iron another weak sextet appeared corresponding to the D-site of Fe 3AI (D03 structure) . With increasing implantation dose this sextet became stronger with broad lines which let conclude to the formation of amorphous fractions . Calculated hyperfine field distributions showed curves with maxima near 30 T . At high doses (5 x 1017 cm -2 ) and energies of 50 and 100 keV a singlet line with an isomer shift IS = 0.2 mm/s (referring to a-iron) appeared additionally . It was interpreted to orginate from nonmagnetic FeAI-like structures . The purpose of this paper is to determine the dose limit for the formation of the nonmagnetic phases and to find out if implantation with 200 keV can result in the formation of such fractions too . Moreover, depth selective CEMS (DCEMS) was applied to investigate which compounds or structures are formed at different depth .
Implantation conditions are given in table 1 . Current densities were chosen so that sample temperatures did not exceed 60°C during ion treatment . In the case of the 200 keV implantation the samples were cooled during the irrac`lation . For the CEMS investigations 1 mm thick no , 'nal a-iron sheets (99.99% Fe) were implanted while '..'or the DCEMS investigations a 7 wm thick a-iron foil enriched to 95 .10% 57 Fe was used . This was necessan , to provide reasonable high counting rates. The foil w is cooled during implantation too .
Permanent address: Institute of Nuclear Research, 72 boul . Tsarigradsko Shosse, BG-1784 Sofia, Bulgaria.
Table 1 Parameters for aluminium implantation in iron Sample ° 1 2 3 4 5 5a 1' 6 7 8
10 I1 12 13 14 b
Dose [cm -2 ] 1 X 10 17 2x10 17 3x10 17 4 x 10 17 5 x 10" 5 ,, j() 17 1 x 10 17 2 x 10 17 3 x 10 17 4x1017 5 x 1017
Energy [keV] 50 50 50 50 50 50 100 100 100 100 100
Current density [WA/cm2 ] 5 4 3 2 4 4 3 3 3 4 4
1 x 1017 200 3 2 x 10 17 200 3 5 x 10 17 200 3 1 x 10 1s 200 4 Samples 1, 2, 5-7, 10-13 from ref . . lion foil for the DCEMS measurement.
U16ß-58' - ' /93/$06 .00 0 1993 - Elsevier Science Publishers il.V . All rights reserved
H. Reuther et al. ;H'~;h .4,?re implantation ofAl in Fe
The CEMS measurements were performed at room temperature . The same conditions as described earlier  were used . A detailed description of the DCEMS system will be published later. 3 . Results In figs . 1-3 the CEM spectra and the calculated hyperfine field distributions of the aluminium implanted iron samples are shown. In fig. 1 is seen that for :i0 keV a dose of 5 X 1017 cm -2 is necessary to produce nonmagnetic components (singlet line). At the lowest ion dose the hyperfine field distribution is narrow with its maximum at about 3I T. With increasing dose the distribt;tion becomes broader with several peaks . For the 100 keV implantation nonmagnetic components are produced already with a dose of 4 X 1017 cm -2 . Regarding the hyperfine field distributions
w w â w
Fig. 2. CEM spectra and calculated hyperfine field d?stribu tions of iron ?mpla.:ted with aluminium with different doses at 100 keV (samples 6-10, from top to bottom).
E W W a
-4 n +4 VELOCITY (mm/s)
Fig. 1 . CEM spectra and calculated hyperfine field distribu tions of iron implanted with aluminium with different doses at 50 keV (samples 1-5, from top to bottom).
the same effects as mentioned above are seen (fig. 2). Implantation with 200 keV and doses of up to 5 X 10 17 CM-2 produces magnetic fractions (fig . 3). Only the high dose of 1 X 10 1" cm -2 let other components appear. The best fit was reached by assuming a quadrupole distribution. The results of the DCEMS measurements are shown in fig. 4. While for 6900, 6600, and for 13400 eV the spectrum consist of a main nonmagnetic component with minor magnetic fractions in the ease of 111750 and 10500 eV magnetic components predominate. 4. Discussion The mean projected ranges R P and the longitudinal straggling 8R p for aluminium implanted into iron were calculated according to ref. [7) using the TRIM program . The values for R P are 33, 67, and 140 nm and for IIb. METAL MODIFICATION (b)
H. Reuther et al. / l/igh dose implantation ofAl in Fe
SR p 20, 36, and 65 nm for implantation with 50, 100, and 200 keV, respectively. As known from Auger electron spectroscopic measurements  the peak of the aluminium concentration is in livery case closer as predicted by the theoretical calculations to the sample sur~'ace . The depth of the aluminium profiles does not exceed 200 run . So the whole implanted range car, be investigated by CEMS. It was already discussed in ref.  which possible processes could take place during implantation. Regarding the high dose aluminium implantation into iron tnese are amorphization or compound formation . From the CEM spectra i! could be assumed that amorphization was the most likely process and that depending on ion doss and ion, encrgv ranges with an Fe sAllike (up to 2 x 10 17 cm -2 Al + for the 50 and l 0U iccV implantation and up to 5 x 10 17 em -2 Al + for the 200 keV implantation) or an FeAl-like (at 5 x 10 17 cm -2 At' for 50 and Hk) keV) short range order has been famed. The results of the 50 and the 100 keV implantations confirm thc results found in ref.  . There is no slow growth with increasing dose but a sudden appearance of the nonmagnetic component. The dose for the transition is different for the two energies. Surprisingly, it
L VI w w r a w
Fi y. 4. DCEIA spectra
and calculated ryperfine field distributions of the 57Fe foil (sample 5a) at virious selected electron c nergies (energy resolution SE/E = 3 yo), plotted from top to bottom in the order of increasin , probing depths .
Fig . 3 . CEM spectra anc! calculated hyperfine field aist"iba tions of iron implanted with aluminium with different doses at 200 keV !samples 11-14, from top to bottom).
is higher for the 50 keV implant uion (5 >, 10 17 crn -2 ) than for the 100 keV impla~itacl ~r. (Q :-< 10 17 cm -2 ), although the maximum aluminium concentration reached due to the implantation should be higher for the lower implantation energy if the dose is the same. It shows that not only the concentration but also the deposired energy is responsible for the kind of the formed structures. At doses lower than those mentioned abore only a broadening of the hyperfine field distribution takes place with increasing ion dose. High dose aluminium implantation into iron with 200 keV results in other components than implantation with lower energies. At 1 x 10" cm --2 comlicaents with a quadrupole moment are produced . The main component of the distribution has a quadrupole splitting of about 0.5 mm/s and 1S = 0.2 mm/ s . The isomer shift corresponds to that of the single line found in the CEN1 spectra of the samples implanted with 50 and 11)0 keV. However, crystalline iron-rich Fe-AI compounds with such quadrupole splitting are not known
H. Reuther ii al / High dose implantation of Al in Fe
and it is not possible to identify this phase only by Mbssbauer spectroscopy. Regarding the DCEMS measurements, a similar pattern is present at all measured electron energies, involving three major components: (i) A central line with IS = 0 .2 mm/s, representing the nonmagnetic alloy phase . This line does not change its shape or relative weight when the sample is cooled down to 40 K. (ii) A magnetic hyperfine-field distribution with a maximum near 30 T and an average IS ~ 0.05 mm/s, du .- to a magnetic Fe 3 A1-tike disordered phase. The 21 .1 T hyperfine splitting of the ordered D 0 3 structure is not seen . (iii) A single Zeeman sextet with = 33 T, IS - 0 mm/s and no quadrupole line shift. It is the signal from the bulk bee (x-iron. The relative area of the nonmagnetic component in the spectra from the almost pure K-conversion-clectron range 6 .3-7.3 keV is reduced with decreasing electron energy . This provides evidence that the Alconcentration profile is highest near the surface, roughly in the topmost 50 run . The disordered magnetic Fc 3 Al-like phase lies below it, extending not deeper than 100 nm from the surface . This depth distrt:, ution is confirmed even more convincingly by the spectra from the 10 .5-13.4 keV range, due to the generally larger probing depths of the L-electrons .
The DCEMS results confirm the measurements perfo.med by integral CEMS and provide a qualitative depth resolution. A more detailed quantitative analysis will soon be published in another paper. Acknowledgements This work was prerared with the support of the Deutsche Fotschungsgemeinschaft, Grants no. Re 868/1-1 and Ke 273/10 1 References  H. Chatham, E. Galvao da Silva, D. Guenzburger and D.E. Ellis, 't'hys . Rev. B35 (1987) 1602 . [21 V.P . Godbole, S.M . Chaudhan, S .V. Ghaisas, S.M . Kanethar, S .B . Ogale and V.G . Bhide, Phys . Rev. B31 (1985) 5703 .  C . Jaouen, J .P. Eymery, E.L. Mathe and J . Deiafund, Mater. Sci. Eng. 69 (1985) 483.  M.A .Z. Vasconcellos, S.R. Teixeira, F.L. Freire, Jr ., M .C .S. Nobrega, P.P . Dionisio, W .H. Schreiner and I .J .R. B ,iumvol, Mater. Sci . Eng. A104 (1988) 159.  H . Reuther, Nucl. lnstr . and Meth. B53 (1991) X67 .  H . Reuther, Nucl. Instr . and Meth. B30 (1988) 61 .  J .F . Ziegler, J.P. Biersack and U. Littmark, The Stopping and Range of Ions in Solids (Pergamon, New York, 1985) .  Unpublished results.
lib . b4ETAL MODIFICATION (b)