- Email: [email protected]
Investigation of fullerene ions in crossed-beams experiments
Nuclear Instruments and Methods in Physics Research B 212 (2003) 67–71 www.elsevier.com/locate/nimb
Investigation of fullerene ions in crossed-beams experiments D. Hathiramani a, P. Scheier b, H. Br€ auning a, R. Trassl a, E. Salzborn L.P. Presnyakov c, A.A. Narits c, D.B. Uskov c,d a
a,*
,
Institut f€ur Kernphysik, Justus-Liebig Universit€at, Leihgesterner Weg 217, D-35392 Giessen, Germany Institut f€ur Ionenphysik, Leopold Franzens Universit€at, Technikerstr. 25, A-6020 Innsbruck, Austria c P.N. Lebedev Physical Institute, 117924 Moscow, Russia d Department of Physics and Astronomy, Pittsburgh, PA 15260, USA
b
Abstract Employing the crossed-beams technique, we have studied the interaction of fullerene ions both with electrons and He2þ -ions. Electron-impact ionization cross sections for Cqþ 60 (q ¼ 1; 2; 3) have been measured at electron energies up to 1000 eV. Unusual features in shape and charge state dependence have been found, which are not observed for atomic ions. The evaporative loss of neutral C2 fragments in collisions with electrons indicates the presence of two different mechanisms. In a first-ever ion–ion crossed-beams experiment involving fullerene ions a cross section of 2þ (1.05 0.06) · 1015 cm2 for charge transfer in the collision Cþ at 117.2 keV center-of-mass energy has been 60 + He obtained. 2003 Elsevier B.V. All rights reserved. PACS: 34.70.+e; 34.80.Kw; 36.40.Qv; 36.40.Wa Keywords: Fullerene ions; Electron ionization; Charge transfer
1. Introduction Fullerenes have been the subject of intense research within the last 15 years and are often regarded as a link between single atoms and solids. However, most collision experiments performed have studied either the interaction of a charged particle with a neutral fullerene or of a fullerene ion with a neutral particle. In this paper we report on our studies of the interaction of positively charged fullerene ions
*
Corresponding author. Fax: +49-641-99-15109. E-mail address: [email protected] (E. Salzborn).
with electrons. We show cross sections for electron-impact ionization and propose a new model to explain the observed behaviour for electronimpact induced C2 -fragmentation. We also present our very first data of collisions between fullerene ions and He2þ -ions. 2. Electron–ion collisions The cross section measurements for collisions between fullerene ions and electrons were performed employing the electron–ion crossed-beams set-up [1] which is shown in Fig. 1. A commercially available mixture of fullerenes, mainly containing C60 and C70 , was evaporated in an electrically
0168-583X/$ - see front matter 2003 Elsevier B.V. All rights reserved. doi:10.1016/S0168-583X(03)01481-2
68
D. Hathiramani et al. / Nucl. Instr. and Meth. in Phys. Res. B 212 (2003) 67–71
single-particle detector
10 GHz ECR ion source
large Faraday cup
electron-ion impact
90° sector magnet
0
0.5
Einzel lens collimating apertures
1m
electrostatic quadrupole triplet
60° spherical condenser
90° sector magnet
Fig. 1. Schematic set-up of the electron–ion collision experiment.
Single-ionization cross section measurements for fullerene ions Cqþ 60 have been performed for charge states q ¼ 1; 2; 3. The results are shown in Fig. 2. These cross sections show a behaviour that has not been observed with any atomic or light molecular ion. After the typical increase above the 3.0
C+60 2.5
cross section [10-15 cm2]
heated oven. The neutral vapor was introduced into the plasma of a 10 GHz electron cyclotron resonance (ECR) ion source [2] where the positively and negatively charged fullerene ions were produced. After mass and energy analysis the ion beam was collimated to 2 · 2 mm2 and crossed with an intense electron beam [3] with electron currents of up to 450 mA. The energy of the electrons can be varied between 10 and 1000 eV. After the electron–ion interaction, the product ions were separated from the incident ion beam by a 90-magnet and detected in a single-particle detector. The current of the parent ion beam was measured simultaneously in a Faraday cup. Employing the dynamic crossed-beams technique introduced by M€ uller et al. [4], where the electron beam is moved through the ion beam with simultaneous registration of the primary and the product ion intensities, absolute cross sections were obtained. The total experimental uncertainties are typically ±10% at the maximum of the cross sections with systematic errors of about 8% and statistical errors at the 95% confidence level.
C603+
3+
C604+
C60 C60
2.0
C602+
2+
1.5
1.0 2.5 2.0
0.5
1.5
C602+
C+60
1.0 0.5
0.0
0.0 20
10
40
60
100
80
100
1000
electron energy [eV]
Fig. 2. Single-ionization cross sections for Cqþ 60 (q ¼ 1; 2; 3). The error bars indicate the total experimental uncertainties.
D. Hathiramani et al. / Nucl. Instr. and Meth. in Phys. Res. B 212 (2003) 67–71
6
-16
2
cross section [10 cm ]
8
4
2
0 10
100
1000
electron energy [eV]
Fig. 3. Absolute cross section for C2 -fragmentation from Cþ 60 þ by electron impact: Cþ 60 + e ! C58 + . The error bars indicate the total experimental uncertainties.
ionization threshold, the cross sections remain constant over a large energy range. Furthermore,
69
the cross sections do not show any structure which could be attributed to indirect ionization processes (see inset in Fig. 2, which was measured using a high-resolution energy-scan method [5]). It should be noted that the cross section for single-ionization þ of C2þ 60 exceeds that of C60 at energies above about 60 eV. The reason for this unexpected behaviour is not yet understood. When energetic electrons collide with a Cqþ 60 molecule fragmentation may occur in addition to ionization. This mainly happens through the evaporation of neutral C2 dimers. The measured þ cross section for the process Cþ 60 + e ! C58 þ is shown in Fig. 3. Cross sections for the same kind of reaction but with doubly and triply charged ions show the same shape differing mainly in the absolute value. The energy dependence suggests the presence of two different mechanisms as shown in Fig. 4. The high-energy part of the cross section
Fig. 4. C2 -fragmentation from Cqþ 60 by electron impact.
70
D. Hathiramani et al. / Nucl. Instr. and Meth. in Phys. Res. B 212 (2003) 67–71
qþ for the reaction Cqþ 60 ! C58 can be fitted by the Lotz formula [6] which is able to describe direct ionization only (left side of Fig. 4). However, since the charge state of the reaction product is the same as that of the parent ion, the behaviour can be explained as an ‘‘unsuccessful ionization’’. This means that the outgoing electron has an energy which is low enough to be re-captured by the fullerene. The shape of the cross section at low energies is very similar to the giant plasmon resonance observed by Hertel et al. [7]. It therefore suggests that at low electron-impact energies, the fragmentation process induced by electron impact is predominantly caused by a plasmon excitation (right side of Fig. 4).
For the theoretical treatment the following physical model is employed. The electron transfer is considered as the under- and over-barrier passing from the fullerene ion to the atomic one. The angular–differential cross sections are described by means of two main mechanisms: the ion–ion Coulomb scattering and the diffraction scattering of the atomic ion upon the carbon cage of the fullerene ion. The latter is considered within the hard-sphere approach. A possible fragmentation is neglected. In the partial-wave representation, the angular– differential cross section has the form (atomic units are used) 2 1 dr 1 X ¼ 2 ð2l þ 1ÞSl Pl ðcos hÞ ; ð2Þ dX 4k l¼0 and the total cross section is given by Z 1 dr p X dX ¼ 2 r¼ ð2l þ 1ÞjSl j2 : dX k l¼0
3. Ion–ion collisions In a first-ever experiment we started to investigate charge exchange in ion–ion collisions involving fullerene ions. As a first step we chose the reaction 2þ þ He2þ þ Cþ 60 ! C60 þ He
Here k denotes the wavenumber, Sl ¼ jSl j expði2dl Þ is the transition matrix element at the given value of the orbital momentum l of the relative motion and Pl is the Legendre polynomial. Characteristic
ð1Þ
Analysing magnets: M1, M2 Quadrupole lenses: Q1, Q2, Q3 Collimators : C1, C2, C3, C4
low voltage terminal UB < 20 kV
Q2
10 GHz full-permanent ECR
ð3Þ
M2 C2
C4
Fig. 5. Schematic set-up of the ion–ion collision experiment.
D. Hathiramani et al. / Nucl. Instr. and Meth. in Phys. Res. B 212 (2003) 67–71
Fig. 6. Time coincidence spectrum for the collision 2þ 3 3 þ He2þ + Cþ 60 ! C60 + He at 117.2 keV center-of-mass energy (measurement time: 30 min).
values of the orbital momentum for fullerene–ion collisions are very large, l P 2 104 . The computational method has been discussed earlier (see Narits et al. [9,10] and references therein). For the 3 He2þ –Cþ 60 experiment the crossedbeams technique in conjunction with the coincident detection of both reaction products was employed. The Giessen ion–ion crossed-beams setup (see Fig. 5) was described in full detail by Meuser et al. [8]. Like in the e -fullerene experiment, a mixture of fullerenes was evaporated in an electrically heated oven. The vapor was introduced into the plasma of an all-permanent magnet 10 GHz-ECR ion source in the low-energy beam line. The energy of the ions was limited to 3 keV due to the 90 analyzing magnet M1. The 3 He2þ -ions were produced in a second all-permanent magnet 10 GHz-ECR source in the high energy beam line with accelerating voltages between 20 and 135 kV. We have used the 3 He isotope in order to avoid ion beam impurities from Hþ 2 ions. The two well-collimated and charge-analyzed ion beams were made to intersect at an angle h ¼ 17:5 in an ultrahigh vacuum of a few 1011 mbar. 3 þ The C2þ 60 ions and the He ions formed in the collision are separated immediately after the interaction region from their parent ion beams by electrostatic deflectors and counted individually by single-particle detectors while the parent ion beams are measured in Faraday cups. The time coincidence spectrum for the collision 2þ 3 3 þ He2þ + Cþ 60 ! C60 + He at 117.2 keV center-of-
71
mass energy is shown in Fig. 6. The width of the time peak is in the order of 5 ls (FWHM) and is the direct result of the spread in the time-of-flight of the slow fullerene ions due to the finite size of the interaction region. The total cross section at this center-of-mass energy is in the order of (1.05 ± 0.06) · 1015 cm2 . The error given is the 1r statistical error only. The systematical error is significantly larger with 25%. The reason for this large systematical error is the low energy and the therefore even lower velocity of the heavy fullerenes, which results in a large uncertainty in the determination of the absolute detection efficiency. Comparison with the theoretical calculations is not yet meaningful, with only one data point. However, the measured cross section is about 20% lower than the theoretical one. The difference is thus within the systematical error limit. A detailed investigation of the energy-dependence of the cross section is currently in progress and will soon be presented in a forthcoming paper. Furthermore, charge exchange between two fullerene ions will be the subject of future experiments.
References [1] K. Tinschert, A. M€ uller, G. Hofmann, K. Huber, R. Becker, D.C. Gregory, E. Salzborn, J. Phys. B: At. Mol. Opt. 22 (1989) 531. [2] M. Liehr, M. Schlapp, R. Trassl, G. Hofmann, M. Stenke, R. V€ olpel, E. Salzborn, Nucl. Instr. and Meth. B 79 (1993) 697. [3] R. Becker, A. M€ uller, C. Achenbach, K. Tinschert, E. Salzborn, Nucl. Instr. and Meth. B 9 (1985) 385. [4] A. M€ uller, K. Tinschert, C. Achenbach, E. Salzborn, Nucl. Instr. and Meth. B 10–11 (1985) 204. [5] A. M€ uller, K. Tinschert, G. Hofmann, E. Salzborn, G.H. Dunn, Phys. Rev. Lett. 61 (1988) 70. [6] W. Lotz, Z. Phys. 206 (1967) 205. [7] I.V. Hertel, H. Steger, J. de Vries, B. Weisser, C. Menzel, B. Kamke, W. Kamke, Phys. Rev. Lett. 68 (1992) 784. [8] S. Meuser, F. Melchert, S. Kr€ udener, A. Pfeiffer, K.V. Diemar, E. Salzborn, Rev. Sci. Instr. 67 (1996) 2752. [9] A.A. Narits, L.P. Presnyakov, E. Salzborn, Short Communications in Physics 12 (2000) 22 (in Russian). [10] A.A. Narits, L.P. Presnyakov, H. Br€auning, E. Salzborn, in: XXII International Conference on the Photonic, Electronic and Atomic Collisions, Santa Fe, USA, Abstracts, 2001, p. 613.
Investigation of fullerene ions in crossed-beams experiments D. Hathiramani a, P. Scheier b, H. Br€ auning a, R. Trassl a, E. Salzborn L.P. Presnyakov c, A.A. Narits c, D.B. Uskov c,d a
a,*
,
Institut f€ur Kernphysik, Justus-Liebig Universit€at, Leihgesterner Weg 217, D-35392 Giessen, Germany Institut f€ur Ionenphysik, Leopold Franzens Universit€at, Technikerstr. 25, A-6020 Innsbruck, Austria c P.N. Lebedev Physical Institute, 117924 Moscow, Russia d Department of Physics and Astronomy, Pittsburgh, PA 15260, USA
b
Abstract Employing the crossed-beams technique, we have studied the interaction of fullerene ions both with electrons and He2þ -ions. Electron-impact ionization cross sections for Cqþ 60 (q ¼ 1; 2; 3) have been measured at electron energies up to 1000 eV. Unusual features in shape and charge state dependence have been found, which are not observed for atomic ions. The evaporative loss of neutral C2 fragments in collisions with electrons indicates the presence of two different mechanisms. In a first-ever ion–ion crossed-beams experiment involving fullerene ions a cross section of 2þ (1.05 0.06) · 1015 cm2 for charge transfer in the collision Cþ at 117.2 keV center-of-mass energy has been 60 + He obtained. 2003 Elsevier B.V. All rights reserved. PACS: 34.70.+e; 34.80.Kw; 36.40.Qv; 36.40.Wa Keywords: Fullerene ions; Electron ionization; Charge transfer
1. Introduction Fullerenes have been the subject of intense research within the last 15 years and are often regarded as a link between single atoms and solids. However, most collision experiments performed have studied either the interaction of a charged particle with a neutral fullerene or of a fullerene ion with a neutral particle. In this paper we report on our studies of the interaction of positively charged fullerene ions
*
Corresponding author. Fax: +49-641-99-15109. E-mail address: [email protected] (E. Salzborn).
with electrons. We show cross sections for electron-impact ionization and propose a new model to explain the observed behaviour for electronimpact induced C2 -fragmentation. We also present our very first data of collisions between fullerene ions and He2þ -ions. 2. Electron–ion collisions The cross section measurements for collisions between fullerene ions and electrons were performed employing the electron–ion crossed-beams set-up [1] which is shown in Fig. 1. A commercially available mixture of fullerenes, mainly containing C60 and C70 , was evaporated in an electrically
0168-583X/$ - see front matter 2003 Elsevier B.V. All rights reserved. doi:10.1016/S0168-583X(03)01481-2
68
D. Hathiramani et al. / Nucl. Instr. and Meth. in Phys. Res. B 212 (2003) 67–71
single-particle detector
10 GHz ECR ion source
large Faraday cup
electron-ion impact
90° sector magnet
0
0.5
Einzel lens collimating apertures
1m
electrostatic quadrupole triplet
60° spherical condenser
90° sector magnet
Fig. 1. Schematic set-up of the electron–ion collision experiment.
Single-ionization cross section measurements for fullerene ions Cqþ 60 have been performed for charge states q ¼ 1; 2; 3. The results are shown in Fig. 2. These cross sections show a behaviour that has not been observed with any atomic or light molecular ion. After the typical increase above the 3.0
C+60 2.5
cross section [10-15 cm2]
heated oven. The neutral vapor was introduced into the plasma of a 10 GHz electron cyclotron resonance (ECR) ion source [2] where the positively and negatively charged fullerene ions were produced. After mass and energy analysis the ion beam was collimated to 2 · 2 mm2 and crossed with an intense electron beam [3] with electron currents of up to 450 mA. The energy of the electrons can be varied between 10 and 1000 eV. After the electron–ion interaction, the product ions were separated from the incident ion beam by a 90-magnet and detected in a single-particle detector. The current of the parent ion beam was measured simultaneously in a Faraday cup. Employing the dynamic crossed-beams technique introduced by M€ uller et al. [4], where the electron beam is moved through the ion beam with simultaneous registration of the primary and the product ion intensities, absolute cross sections were obtained. The total experimental uncertainties are typically ±10% at the maximum of the cross sections with systematic errors of about 8% and statistical errors at the 95% confidence level.
C603+
3+
C604+
C60 C60
2.0
C602+
2+
1.5
1.0 2.5 2.0
0.5
1.5
C602+
C+60
1.0 0.5
0.0
0.0 20
10
40
60
100
80
100
1000
electron energy [eV]
Fig. 2. Single-ionization cross sections for Cqþ 60 (q ¼ 1; 2; 3). The error bars indicate the total experimental uncertainties.
D. Hathiramani et al. / Nucl. Instr. and Meth. in Phys. Res. B 212 (2003) 67–71
6
-16
2
cross section [10 cm ]
8
4
2
0 10
100
1000
electron energy [eV]
Fig. 3. Absolute cross section for C2 -fragmentation from Cþ 60 þ by electron impact: Cþ 60 + e ! C58 + . The error bars indicate the total experimental uncertainties.
ionization threshold, the cross sections remain constant over a large energy range. Furthermore,
69
the cross sections do not show any structure which could be attributed to indirect ionization processes (see inset in Fig. 2, which was measured using a high-resolution energy-scan method [5]). It should be noted that the cross section for single-ionization þ of C2þ 60 exceeds that of C60 at energies above about 60 eV. The reason for this unexpected behaviour is not yet understood. When energetic electrons collide with a Cqþ 60 molecule fragmentation may occur in addition to ionization. This mainly happens through the evaporation of neutral C2 dimers. The measured þ cross section for the process Cþ 60 + e ! C58 þ is shown in Fig. 3. Cross sections for the same kind of reaction but with doubly and triply charged ions show the same shape differing mainly in the absolute value. The energy dependence suggests the presence of two different mechanisms as shown in Fig. 4. The high-energy part of the cross section
Fig. 4. C2 -fragmentation from Cqþ 60 by electron impact.
70
D. Hathiramani et al. / Nucl. Instr. and Meth. in Phys. Res. B 212 (2003) 67–71
qþ for the reaction Cqþ 60 ! C58 can be fitted by the Lotz formula [6] which is able to describe direct ionization only (left side of Fig. 4). However, since the charge state of the reaction product is the same as that of the parent ion, the behaviour can be explained as an ‘‘unsuccessful ionization’’. This means that the outgoing electron has an energy which is low enough to be re-captured by the fullerene. The shape of the cross section at low energies is very similar to the giant plasmon resonance observed by Hertel et al. [7]. It therefore suggests that at low electron-impact energies, the fragmentation process induced by electron impact is predominantly caused by a plasmon excitation (right side of Fig. 4).
For the theoretical treatment the following physical model is employed. The electron transfer is considered as the under- and over-barrier passing from the fullerene ion to the atomic one. The angular–differential cross sections are described by means of two main mechanisms: the ion–ion Coulomb scattering and the diffraction scattering of the atomic ion upon the carbon cage of the fullerene ion. The latter is considered within the hard-sphere approach. A possible fragmentation is neglected. In the partial-wave representation, the angular– differential cross section has the form (atomic units are used) 2 1 dr 1 X ¼ 2 ð2l þ 1ÞSl Pl ðcos hÞ ; ð2Þ dX 4k l¼0 and the total cross section is given by Z 1 dr p X dX ¼ 2 r¼ ð2l þ 1ÞjSl j2 : dX k l¼0
3. Ion–ion collisions In a first-ever experiment we started to investigate charge exchange in ion–ion collisions involving fullerene ions. As a first step we chose the reaction 2þ þ He2þ þ Cþ 60 ! C60 þ He
Here k denotes the wavenumber, Sl ¼ jSl j expði2dl Þ is the transition matrix element at the given value of the orbital momentum l of the relative motion and Pl is the Legendre polynomial. Characteristic
ð1Þ
Analysing magnets: M1, M2 Quadrupole lenses: Q1, Q2, Q3 Collimators : C1, C2, C3, C4
low voltage terminal UB < 20 kV
Q2
10 GHz full-permanent ECR
ð3Þ
M2 C2
C4
Fig. 5. Schematic set-up of the ion–ion collision experiment.
D. Hathiramani et al. / Nucl. Instr. and Meth. in Phys. Res. B 212 (2003) 67–71
Fig. 6. Time coincidence spectrum for the collision 2þ 3 3 þ He2þ + Cþ 60 ! C60 + He at 117.2 keV center-of-mass energy (measurement time: 30 min).
values of the orbital momentum for fullerene–ion collisions are very large, l P 2 104 . The computational method has been discussed earlier (see Narits et al. [9,10] and references therein). For the 3 He2þ –Cþ 60 experiment the crossedbeams technique in conjunction with the coincident detection of both reaction products was employed. The Giessen ion–ion crossed-beams setup (see Fig. 5) was described in full detail by Meuser et al. [8]. Like in the e -fullerene experiment, a mixture of fullerenes was evaporated in an electrically heated oven. The vapor was introduced into the plasma of an all-permanent magnet 10 GHz-ECR ion source in the low-energy beam line. The energy of the ions was limited to 3 keV due to the 90 analyzing magnet M1. The 3 He2þ -ions were produced in a second all-permanent magnet 10 GHz-ECR source in the high energy beam line with accelerating voltages between 20 and 135 kV. We have used the 3 He isotope in order to avoid ion beam impurities from Hþ 2 ions. The two well-collimated and charge-analyzed ion beams were made to intersect at an angle h ¼ 17:5 in an ultrahigh vacuum of a few 1011 mbar. 3 þ The C2þ 60 ions and the He ions formed in the collision are separated immediately after the interaction region from their parent ion beams by electrostatic deflectors and counted individually by single-particle detectors while the parent ion beams are measured in Faraday cups. The time coincidence spectrum for the collision 2þ 3 3 þ He2þ + Cþ 60 ! C60 + He at 117.2 keV center-of-
71
mass energy is shown in Fig. 6. The width of the time peak is in the order of 5 ls (FWHM) and is the direct result of the spread in the time-of-flight of the slow fullerene ions due to the finite size of the interaction region. The total cross section at this center-of-mass energy is in the order of (1.05 ± 0.06) · 1015 cm2 . The error given is the 1r statistical error only. The systematical error is significantly larger with 25%. The reason for this large systematical error is the low energy and the therefore even lower velocity of the heavy fullerenes, which results in a large uncertainty in the determination of the absolute detection efficiency. Comparison with the theoretical calculations is not yet meaningful, with only one data point. However, the measured cross section is about 20% lower than the theoretical one. The difference is thus within the systematical error limit. A detailed investigation of the energy-dependence of the cross section is currently in progress and will soon be presented in a forthcoming paper. Furthermore, charge exchange between two fullerene ions will be the subject of future experiments.
References [1] K. Tinschert, A. M€ uller, G. Hofmann, K. Huber, R. Becker, D.C. Gregory, E. Salzborn, J. Phys. B: At. Mol. Opt. 22 (1989) 531. [2] M. Liehr, M. Schlapp, R. Trassl, G. Hofmann, M. Stenke, R. V€ olpel, E. Salzborn, Nucl. Instr. and Meth. B 79 (1993) 697. [3] R. Becker, A. M€ uller, C. Achenbach, K. Tinschert, E. Salzborn, Nucl. Instr. and Meth. B 9 (1985) 385. [4] A. M€ uller, K. Tinschert, C. Achenbach, E. Salzborn, Nucl. Instr. and Meth. B 10–11 (1985) 204. [5] A. M€ uller, K. Tinschert, G. Hofmann, E. Salzborn, G.H. Dunn, Phys. Rev. Lett. 61 (1988) 70. [6] W. Lotz, Z. Phys. 206 (1967) 205. [7] I.V. Hertel, H. Steger, J. de Vries, B. Weisser, C. Menzel, B. Kamke, W. Kamke, Phys. Rev. Lett. 68 (1992) 784. [8] S. Meuser, F. Melchert, S. Kr€ udener, A. Pfeiffer, K.V. Diemar, E. Salzborn, Rev. Sci. Instr. 67 (1996) 2752. [9] A.A. Narits, L.P. Presnyakov, E. Salzborn, Short Communications in Physics 12 (2000) 22 (in Russian). [10] A.A. Narits, L.P. Presnyakov, H. Br€auning, E. Salzborn, in: XXII International Conference on the Photonic, Electronic and Atomic Collisions, Santa Fe, USA, Abstracts, 2001, p. 613.