Development of a time-of-flight spectrometer at the Ru der Bošković Institute in Zagreb

Description
Development of a time-of-flight spectrometer at the Ru der Bošković Institute in Zagreb

Please download to get full document.

View again

of 5
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Information
Category:

Leadership & Management

Publish on:

Views: 2 | Pages: 5

Extension: PDF | Download: 0

Share
Transcript
  Development of a time-of-flight spectrometer at the Ru  d erBosˇkovic´ Institute in Zagreb Zdravko Siketic´ * , Iva Bogdanovic´ Radovic´, Milko Jaksˇic´ Ru d  er Bosˇkovic´ Institute, P.O. Box 180, 10002 Zagreb, Croatia Received 20 September 2007; received in revised form 6 December 2007Available online 29 January 2008 Abstract A new TOF telescope has been constructed for thin film and surface analysis. The timing system consists of two electrostatic mirrortype detectors of Busch design. The detection efficiency of timing stations for very light ions was significantly improved using DLC (dia-mond like carbon) foils coated with LiF instead of the conventionally used carbon foils. Ion energy is measured by a 300 mm 2 ULTRAion-implanted silicon detector. For the ERDA measurements with heavy and energetic ion beams, a time-of-flight (TOF) spectrometer ispositioned at 37.5  . Spectrometer can be easily moved to 120   backward angle for time-of-flight RBS analysis with low energy beam of light ions. Positioning and fine adjustments of sample orientation are performed with a motorized sample stage. The same spectrometercan be also installed at the ion microprobe scattering chamber for 3D elemental imaging.   2007 Elsevier B.V. All rights reserved. PACS:  82.80.Yc; 24.10.Lx; 82.80.Rt Keywords:  Elastic recoil detection; Time-of-flight; Time resolution; Mass resolution; Depth resolution; Efficiency 1. Introduction Time-of-flight (TOF) elastic recoil detection analysis(ERDA) is a simple method for the mass separation anddepth profiling of the light elements. The energy resolutionof the silicon surface barrier detectors (SSB) [1,2], used inclassical ERDA setup, limits depth resolution for the hea-vier ions ( Z   > 3) (pulse height defects). In the TOF-ERDAsetup, energy of the recoiled particles is calculated from thetime-of-flight between the two time detectors. In this caseenergy resolution is better and achievable depth resolutionat the surface is several nm [3]. Additional application of TOF spectrometer can be also for the Rutherford backscat-tering (RBS) analysis. With TOF-RBS setup and lowenergy He ion beam, achieved depth resolution can be evenbetter than in TOF-ERDA [4].A TOF-ERDA system described here was constructed atthe accelerator facility in Zagreb. It is installed at the beamline that can accept ions from either 6 MV EN Tandem (forTOF-ERDA) or 1.0 MV Tandetron accelerator (for TOF-RBS). The first results and performance of the TOF-ERDA telescope are presented. 2. TOF-ERDA setup TOF-ERDA telescope, positioned at 37.5   scatteringangle, consists of two time detectors separated by 523 mmand ULTRA ion-implanted silicon detector of 300 mm 2 area. First time detector is 478 mm away from the targetwith 6 mm collimator placed in front of it to reduce theamount of false start signals. Also, collimator defines solidangle of the telescope to be 0.11 msr. Samples are mountedon the motorized sample stage which enables fine and pre-cise target orientation in respect to the beam direction.Time detectors are electrostatic mirror assemblies of Busch design [5]. Our assembly includes DLC foil 0168-583X/$ - see front matter    2007 Elsevier B.V. All rights reserved.doi:10.1016/j.nimb.2007.12.070 * Corresponding author. Tel.: +385 1 456 1012; fax: +385 1 4680 239. E-mail address:  zsiketic@irb.hr (Z. Siketic´). www.elsevier.com/locate/nimb  Available online at www.sciencedirect.com Nuclear Instruments and Methods in Physics Research B 266 (2008) 1328–1332 N B Beam InteractionswithMaterials&Atoms  (0.4  l g/cm 2 of C grid supported) [6], three tungsten gridelectrodes and a microchannel plate (MCP) for electrondetection. Timing signals are produced in MCP detectorsdue to electron emission induced by recoil atoms that passthrough the thin foil [7]. Time-of-flight is measuredbetween the two time detectors where the first detector pro-vides a start signal. Signals from MCP are directly sent tothe constant fraction discriminator (CFD) input. CFD pro-vides a start and stop signals for the time to amplitude con-verter (TAC) unit. Signal from the energy detector is usedin the coincidence with TOF signal. In a typical two dimen-sional (E, TOF) coincidence map, signals from differentrecoil atoms are well separated due to mass dependentvelocity (Figs. 1 and 2).  2.1. Intrinsic time resolution Correlation between the time-of-flight and energy of anrecoil ion is t  ¼ l  ffiffiffiffiffiffi m 2  E  r   ;  ð 2 : 1 : 1 Þ where  l   is distance between two time detectors,  E   is energyand  m  mass of the recoiled ion. Overall TOF time resolu-tion consists of two contributions: D t  2 ¼  lm 8  E  3  D  E  2 þ D t  2int ;  ð 2 : 1 : 2 Þ where  D t int  is the intrinsic time resolution of the entire spec-trometer and  D E   is the energy spread of ions entering thestart detector. The energy spread consists of the ion beamenergy spread  D E  beam , geometrical spread  D E  geom  due to fi-nite solid angle, energy straggling in the foil  D E  stragg  andthe spread due to non-uniform foil thickness  D E  foil  [8]: D  E  2 ¼ D  E  2beam þ D  E  2geom þ D  E  2stragg þ D  E  2foil :  ð 2 : 1 : 3 Þ Intrinsic time resolution for our system was calculatedfrom the measurements of the time peak of 15 MeV 16 O 3+ ions scattered from the 2 nm thick gold layer. Maincontribution to the energy spread (2.1.3) the geometricalspread  D E  geom  is negligible in the scattering case (othercontributions from (2.1.3) are also negligible). Taking allthis into account the main contribution to the overall timeresolution (2.1.2) is coming from  D t int . It was found thatintrinsic time resolution is  D t int  = 170 ps.  2.2. Mass resolution The mass resolution is defined as [9] D mm   2 ¼  2 D t t    2 þ  D  E  E    2 þ  2 D ll   2 ;  ð 2 : 2 : 1 Þ where  D t t   is relative width of the time peak in a projection of the TOF-ERDA spectrum corresponding to certain mass, D  E  E   is relative energy resolution of the particle detector [1]and  D ll  is path length difference contribution due to thedetector solid angle.From Figs. 1 and 2 it can be seen that mass resolution of our spectrometer is approximately 1 over the mass range1 <  m  < 28. Fig. 1 shows overlapped TOF-ERDA coinci-dence maps of elastically scattered  1 H,  4 He,  7 Li,  10 B,  11 B, 16 O and  19 F ions from the thick Au target. Fig. 2 showsoverlapped TOF-ERDA coincidence maps elastic recoiled 27 Al and  28 Si by the 35 MeV  35 Cl 6+ ions. Calculated value Fig. 1. Overlapped TOF-ERDA coincidence maps of elastically scattered  1 H,  4 He,  7 Li,  10 B,  11 B,  16 O and  19 F ions from the thick Au target. Z. Siketic´  et al./Nucl. Instr. and Meth. in Phys. Res. B 266 (2008) 1328–1332  1329  D m  = 1.1 (2.2.1) for elastic recoiled  27 Al by 35 MeV  35 Cl isconsistent with statement that  27 Al can be well separatedfrom  28 Si (see Fig. 2).  2.3. Depth resolution Depth resolution is usually defined as the depth intervalcorresponding to the energy resolution [10]: D  x ¼  D  E S  eff  ;  ð 2 : 3 : 1 Þ where  D E   is the energy resolution and  S  eff   is the effectivestopping power. To achieve the best possible depth resolu-tion, energy spread must be minimized and effective stop-ping power increased at the same time. Maincontributions to the energy spread are kinematical spread,energy straggling and multiple scattering [10,11]. Optimiza-tion of the experimental parameters (incident ion, angle,energy) depends on analyzing element, target thicknessand matrix composition. The best surface resolution is ob-tained using a low energy heavy ion beams with small inci-dence angle [3]. On the other hand, for thicker films a depthresolution deteriorates rapidly (due to pronounced multiplescattering for heavy ions in heavy matrix) and thereforebetter results are obtained with higher incident energies [3].In order to calculate surface depth resolution for oxygenin SiO 2  matrix, 10 MeV  127 I beam with  h in  = 5   (toward thesample surface) was used. The experimental energy resolu-tion was calculated by fitting the edge of the energy profilewith a Boltzmann function (time axis was used for theenergy profile). Using (2.3.1), with measured value  D E   =(67 ± 3) keV and calculated value  S  eff   = 12.12 keV/nm, itwas found that surface depth resolution for the oxygen inSiO 2  matrix is  D x  = (5.5 ± 0.2) nm.  2.4. Relative detector efficiency The relative TOF-ERDA detector efficiency is defined asthe ratio between the number of scattered particles in theTOF-E coincidence spectrum and all particles detected bySSB detector without coincidence [12]. When ion passesthrough the foil in the time detector, emitted secondaryelectrons trigger the processing electronics [7]. The TOF-ERDA detector efficiency depends on a number of elec-trons emitted from the foil and probability that a singleelectron impinging on the MCP triggers the electronics(electron optics, quantum efficiency of the MCP andCFD threshold). The detector efficiency is mainly influ-enced by the secondary electron production if the CFDthreshold and time detector voltages are fixed [12]. Thenumber of emitted secondary electrons is proportional tothe electronic stopping power [7,13,14], resulting in adependence of the detection efficiency on an ion type andenergy. Detector efficiency is therefore smaller for lighterions and its value is important for the quantitative analysis.For heavier elements efficiency is constant and close to 1.To improve relative detector efficiency for lightest elementsit is important to increase electron yield from DLC foils,which was achieved by coating them with thin LiF layer(2.5  l g/cm 2 ).Efficiency calibration of our spectrometer was per-formed by scattering of different light ions ( 1 H,  4 He,  7 Li, 16 O) from a thick Au target. CFD threshold was set at  5 mV. Fig. 3 shows measured values of the relative detec-tor efficiency in dependence with an electronic stopping    T   O   F   (  c   h  a  n  n  e   l  s   ) E (channels) Recoiled 27 Al Recoiled 28 Si Scattered 35 Cl           →  →    → Fig. 2. Overlapped TOF-ERDA coincidence maps elastic recoiled  27 Al and  28 Si by the 35 MeV  35 Cl 6+ ions.1330  Z. Siketic´  et al./Nucl. Instr. and Meth. in Phys. Res. B 266 (2008) 1328–1332  power in the LiF. Comparison with results from Zhanget al. [12] shows enhancement of detector efficiency by afactor of two for lightest elements. Saturation in efficiencyis in our case obtained for lighter ions. It can be also seenthat saturation value is less than 1 (  0.95) which is mostlikely caused by limited quantum efficiency of MCP-s.Stopping power for  1 H,  4 He,  7 Li and  16 O ions in LiF wascalculated by SRIM 2003 code [15] and energy calibration 0.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 2.25 2.500.20.30.40.50.60.70.80.91.0    R  e   l  a   t   i  v  e   d  e   t  e  c   t  o  r  e   f   f   i  c   i  e  n  c  y Electronic stopping power (keV/nm) 1 H 4 He 7 Li 16 O   fit − − − Ref. [12] Fig. 3. Relative detector efficiency as a function of electronic stopping power in the LiF for  1 H,  4 He,  7 Li and  16 O ions.    T   O   F   (  c   h  a  n  n  e   l  s   ) E (channels) Recoiled 1 H Recoiled 12 C Recoiled 16 O Recoiled 14 N ←          ←    ←  ← Fig. 4. Coincidence TOF-ERDA spectrum of Kapton  bombarded by 35 MeV  35 Cl 6+ ions. Z. Siketic´  et al./Nucl. Instr. and Meth. in Phys. Res. B 266 (2008) 1328–1332  1331  of the SSB detector was performed by multivariate analysis[16]. Experimental data were fitted to an empiricalfunction: g ¼ a  b exp   S  el t  1    c exp   S  el t  2   ;  ð 2 : 4 : 1 Þ where  S  el  is electronic stopping power in LiF. Fittedparameters are  a  = (0.940 ± 0.002),  b  = (0.95 ± 0.06),  c  =(0.59 ± 0.05),  t 1  = (0.029 ± 0.003) and  t 2  = (0.139 ±0.007). This function (2.4.1) may be used to determinethe relative detection efficiency for any recoiled ion.  2.5. H detection Fig. 4 shows ERDA spectrum of Kapton  sample asobtained by 35 MeV  35 Cl 6+ ion beam. As can be seen  12 C, 14 N and  16 O are well separated. For the lightest elementhydrogenitisvisiblethatthelinecorrespondingtohydrogenis accompanied by additional, ‘‘shadow ”  background line.The srcin of this ‘‘shadow ”  is attributed to the recoiledhydrogen ions that are energetic enough to pass throughtheCusupportinggridineitherbothDLCfoilsorinjustsec-ond DLC foil. The main line of hydrogen is partly consistedfrom hydrogen ions that passed through supporting gridonly in first DLC foil. Thus, the thickness of supportingCu grid (5  l m) limits the energy of recoiled ions at whichtheir quantitative analysis can be performed. Using SRIM2003 code [15] it is found that  1 H,  4 He,  7 Li,  9 Be and  11 Bcan be quantitative analyzed up to energies of 0.8 MeV,2.7 MeV, 4.5 MeV, 7 MeV and 9 MeV, respectively. 3. Conclusions The TOF-ERDA was constructed in order to performmultielemental depth profiling of light elements in heaviermatrices in the near surface layers. Mass resolution of 1 amu for the elements ( Z   < 14) has been achieved.Obtained surface depth resolution for oxygen is 5 nm andcan be improved if a position sensitive detector instead aconventional SSB particle detector is used. Coating of DLC foils with LiF enhanced detector efficiency by the fac-tor of two for the lightest elements. Limitations due to thethickness of supporting grid in DLC foil must be taken intoaccount. References [1] P.F. Hinrichen, D.W. Hetherington, S.C. Gujrathi, L. Cliche, Nucl.Instr. and Meth. B 45 (1990) 275.[2] L. Cliche, S.C. Gujrathi, L.A. Hamel, Nucl. Instr. and Meth B 45(1990) 270.[3] S. Giangrandi, K. Arstila, B. Brijs, T. Sajavaara, A. Vantomme, W.Vandervorst, Nucl. Instr. and Meth. B 261 (2007) 512.[4] M. Do¨beli, R.M. Ender, V. Liechtenstein, D. Vetterli, Nulc. Instr.and Meth. B 142 (1998) 417.[5] F. Busch, W. Pfeffer, B. Kohlmeyer, D. Schu¨ll, F. Pu¨hlhoffer, Nucl.Instr. and Meth. 171 (1980) 71.[6] V. Kh. Liechtenstein, T.M. Ivkova, E.D. Olshanski, R. Repnow, P.Steier, W. Kutscera, A. Wallner, R. von Hahn, Nucl. Instr. and Meth.A 561 (2006) 120.[7] E.J. Sternglass, Phys. Rev. 108 (1957) 1.[8] J.K. Kim, Y.S. Kim, G.D. Kim, H.W. Choi, H.J. Woo, S.Y. Cho,C.N. Wang, Nucl. Instr. and Meth. B 140 (1998) 380.[9] A. Razpet, P. Pelicon, Z. Rupnik, M. Budnar, Nucl. Instr. and Meth.B 201 (2003) 535.[10] E. Szila´gy, Nucl. Instr. and Meth. B 183 (2001) 25.[11] M. Mayer, K. Arstila, K. Nordlund, E. Edelmann, J. Keinonen, Nucl.Instr. and Meth. B 249 (2006) 823.[12] Y. Zhang, H.J. Whitlow, T. Winzell, I.F. Bubb, T. Sajavaara, K.Arstila, J. Keinonen, Nucl. Instr. and Meth. B 149 (1999) 477.[13] H. Rothard, K. Kroneberger, A. Clouvas, E. Veje, P. Lorenzen, N.Keller, J. Kemmler, W. Meckbach, K.O. Groenveld, Phys. Rev. A 41(1990) 2521.[14] A. Clouvas, H. Rothard, M. Burkhard, K. Kroneberger, C. Bieder-mann, J. Kemmler, K.O. Groenveld, R. Kirsch, P. Misaelidis, A.Katsanos, Phys. Rev. B 39 (1989) 6316.[15] J.F. Ziegler, Nucl. Instr. and Meth. B 219 (2004) 1027.[16] M. El Bouanani, M. Hult, L. Persson, E. Swietlick, M. Andersson, M.O¨ stling, N. Lundberg, C. Zaring, D.D. Cohen, N. Dytlewski, P.N.Johnston, S.R. Walker, I.F. Bubb, H.J. Whitlow, Nucl. Instr. andMeth. B 94 (1994) 530.1332  Z. Siketic´  et al./Nucl. Instr. and Meth. in Phys. Res. B 266 (2008) 1328–1332
Related Search
Similar documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks