- Email: [email protected]
First operation of a superconducting RF-gun
ARTICLE IN PRESS
Nuclear Instruments and Methods in Physics Research A 507 (2003) 314–317
First operation of a superconducting RF-gun a D. Janssena,*, H. Buttig . , P. Evtushenkoa, M. Freitaga, F. Gabriela, B. Hartmanna, a . a, T. Quasta, B. Reppea, A. Schamlotta, U. Lehnert , P. Michela, K. Moller Ch. Schneidera, R. Schuriga, J. Teicherta, S. Konstantinovb, S. Kruchkovb, A. Kudryavtsevb, O. Myskinb, V. Petrovb, A. Tribendisb, V. Volkovb, W. Sandnerc, I. Willc, A. Matheisend, W. Moellerd, M. Pekelere, P.v. Steine, Ch. Haberstrohf a
FZR-Rossendorf, Neue Beschleuniger Zentralabteilung, Postfach 510119, Dresden 01314, Germany b Budker Institute of Nuclear Physics, Novosibirsk, Russia c Max Born Institut, Berlin, Germany d DESY, Hamburg, Germany e ACCEL Instruments, Bergisch Gladbach, Germany f TU Dresden, Germany
Abstract For the first time, a superconducting RF gun where a photocathode is inside a superconducting cavity has been working stably over a period of seven weeks. The gun with a half-cell cavity has been operated in the cryostat at 4.2 K. A maximal field strength of 22 MV/m in the cavity and maximal beam energy of 900 keV have been obtained. Measurements of beam current, transmission, energy spread and transverse emittance are presented and discussed. r 2003 Elsevier Science B.V. All rights reserved. PACS: 41.60.Cr; 41.75.Fr; 42.55.Xi; 82.25.j Keywords: Cavity; Superconductivity; Photocathode; Laser; Electron gun
1. Introduction In recent years, there has been a renewed interest in high average power FELs and, as a result, high average current photoinjectors. Some effort is made, to increase the duty factor of normal conducting RF guns at the price of great cooling problems, high demands on klystron power and low energy conversion efficiency. *Corresponding author. Tel.: +49-351-260-3548; fax: +39351-260-3690. E-mail address: [email protected] (D. Janssen).
A more elegant way would be to combine the high brightness of RF guns with the low RF losses of superconducting cavity technology. Besides, superconducting cavities offer an additional merit. The good vacuum conditions should increase the lifetime the of very sensitive photo cathodes. A disadvantage of superconducting technology is its sensitivity with respect to magnetic fields. In Ref. [1] it is shown, that instead of a magnetic field which is applied in normal conducting RF guns, the RF focusing can be used in a properly designed superconducting cavity to achieve small emittance and short bunch length at high-bunch charge.
0168-9002/03/$ - see front matter r 2003 Elsevier Science B.V. All rights reserved. doi:10.1016/S0168-9002(03)00936-7
ARTICLE IN PRESS D. Janssen et al. / Nuclear Instruments and Methods in Physics Research A 507 (2003) 314–317
The first proposal on a superconducting photoelectron source has been published in Ref. [2]. The behavior of photocathodes inside a superconducting cavity has been investigated in Ref. [3]. In 1998 a cooperation between German and Russian institutes started the development of a superconducting RF gun based on a half-cell TESLA cavity. First RF measurements with this cavity in a test cryostat at DESY showed excellent values for field strength and Q value [4].
2. Description of the gun The cavity of the RF gun is a TESLA type [5] half-cell closed by a shallow cone with a centered hole in which the cathode is situated. A special support structure isolates the cathode thermally and electrically from the surrounding cavity and a four-step coaxial filter prevents the transfer of RF power through the coaxial line around the cathode stem. The cathode itself is normal conducting and is held at liquid nitrogen temperature. A detailed description of the cavity has been published in Ref. [6]. An overview of the gun is shown in Fig. 1. The gun cavity, together with the cathode support, the RF coupler and a part of the beam tube are installed in the He vessel of the cryostat. This cryostat has no direct transfer line to the refrigerator. After filling the vessel by a dewar
315
with liquid helium we operate the gun at normal pressure and 4.2 K over a time of 4–5 h. The cathode can be moved by means of a manipulator from the preparation chamber into the cavity. Through a port at the beginning of the beam line the laser light is guided to the cathode. In the preparation chamber Cs2Te layers are deposited onto the cathodes. In the chamber the pressure was 1010 mbar, but during the heating of evaporators it increased to 106 mbar. For that reason the quantum efficiency of the cathodes was not better than 0.25%. The laser consists of an oscillator which works in the additive pulse modelocking (APM) mode, a five-pass diode pumped amplifier and a UV conversion stage. The oscillator supplies a pulse frequency of 26 MHz. After amplification and conversion UV laser pulses with 263 nm wavelength and 5 ps (FWHM) length are obtained. A jitter s ¼ 2 ps (peak–peak) was measured with a stable reference signal from the RF oscillator. In order to suppress the influence of fluctuations in the cavity frequency (microphonics), the laser is synchronized to the cavity resonance frequency. In this case the jitter of laser pulses increases to 12 ps (peak–peak). Beam line diagnostics include steerers, view screens, solenoids a pepper pot mask for measuring the transversal emittance, and a dipole magnet for measuring energy and energy spread. The beam current can be measured by an insulated beam dump at the end of the line.
Fig. 1. Cryostat of the SRF gun with preparation chamber.
ARTICLE IN PRESS 316
D. Janssen et al. / Nuclear Instruments and Methods in Physics Research A 507 (2003) 314–317
3. Results of the measurements Fig. 2 shows the measurement of the quality factor of the cavity dependence on the field strength. The values are from direct RF measurement and from a comparison of measured electron energy with simulation. The maximum field strength of 22 MV/m near the cathode is limited by field emission. The insignificant difference of Q values with and without cathode shows the good performance of the four stage coaxial RF filter. Fig. 3 presents the cathode emission and accelerated (dump) current together with the corresponding electron energy as a function of the laser phase. For a phase window of 60 we obtain complete transmission and the energy reaches its maximal value at j ¼ 0 : These properties are determined by the geometry of the cavity and the field amplitude. In Fig. 4 the behavior of the energy spread with respect to the laser phase is presented. For small phase angles a spread smaller than 1% has been obtained. The measured values can be reproduced by calculation, assuming an electron bunch length between 10 and 20 ps. The influence of the bias voltage between cathode and cavity is shown in Fig. 5. This voltage gives an additional possibility to influence the beam optics and (if necessary) to suppress multipacting effects. It is interesting to remark that we can obtain both focusing and defocusing effects.
Fig. 2. Determination of the maximum field strength from RF and energy measurements.
Fig. 3. Transmission of the electron beam and the dependence of the energy from the laser phase.
Fig. 4. Energy spread as function of the laser phase.
Fig. 5. Focusing effect of the bias voltage.
ARTICLE IN PRESS D. Janssen et al. / Nuclear Instruments and Methods in Physics Research A 507 (2003) 314–317
4. Conclusion For the first time we have shown that a RF electron source, where a photo cathode is inside a superconducting cavity, can work stably over a period of seven weeks (approximately 5 h/day). At a temperature of 4.2 K we could not see any change in the quality factor Q ¼ 2:5 108 of the cavity during the whole period of operation. The field emission of the cavity, which is the reason for the limitation of the field strength, is caused by the difficult clean room handling of the big intrinsic parts of the cryostat. But nevertheless a field strength of 22 MV/m and an electron energy of 900 keV have been obtained. The maximum bunch charge was 20 pC, which corresponds to an average current of 520 mA in the cw-mode. It is limited by average power and repetition rate of the laser and by the small quantum efficiency of the photocathode. Due to the long drift space after the gun and the arrangement of optical elements, we could measure the transversal emittance for bunch charges between 1 and 4 pC only. In agreement with
317
PARMELA calculation we have measured normalized rms-values between 1 and 2.5 mm mrad.
Acknowledgements We would like to thank Prof. A. Schwettmann and Prof. T. Smith, who have made their cryostat and the RF coupler available for our experiments. Furthermore we thank Prof. E. Grosse, K. Jordan, P. Kneisel and B. Schneider for many helpful discussions.
References [1] D. Janssen, V. Volkov, Nucl. Instr. and Meth. A 452 (2000) 34. [2] H. Piel, et al., Proceedings of the 10th FEL Conference, Jerusalem, 1998. [3] A. Michalke, External report, WUB-DIS 92–5, Universit.at Wuppertal 1993. [4] E. Barthels, et al., Nucl. Instr. and Meth. A 445 (2000) 408. [5] A. Aune, et al., DESY 00-031, Hamburg 2000. [6] P.v. Stein, Internal report, FZR-227, Forschungszentrum Rossendorf 1998.
Nuclear Instruments and Methods in Physics Research A 507 (2003) 314–317
First operation of a superconducting RF-gun a D. Janssena,*, H. Buttig . , P. Evtushenkoa, M. Freitaga, F. Gabriela, B. Hartmanna, a . a, T. Quasta, B. Reppea, A. Schamlotta, U. Lehnert , P. Michela, K. Moller Ch. Schneidera, R. Schuriga, J. Teicherta, S. Konstantinovb, S. Kruchkovb, A. Kudryavtsevb, O. Myskinb, V. Petrovb, A. Tribendisb, V. Volkovb, W. Sandnerc, I. Willc, A. Matheisend, W. Moellerd, M. Pekelere, P.v. Steine, Ch. Haberstrohf a
FZR-Rossendorf, Neue Beschleuniger Zentralabteilung, Postfach 510119, Dresden 01314, Germany b Budker Institute of Nuclear Physics, Novosibirsk, Russia c Max Born Institut, Berlin, Germany d DESY, Hamburg, Germany e ACCEL Instruments, Bergisch Gladbach, Germany f TU Dresden, Germany
Abstract For the first time, a superconducting RF gun where a photocathode is inside a superconducting cavity has been working stably over a period of seven weeks. The gun with a half-cell cavity has been operated in the cryostat at 4.2 K. A maximal field strength of 22 MV/m in the cavity and maximal beam energy of 900 keV have been obtained. Measurements of beam current, transmission, energy spread and transverse emittance are presented and discussed. r 2003 Elsevier Science B.V. All rights reserved. PACS: 41.60.Cr; 41.75.Fr; 42.55.Xi; 82.25.j Keywords: Cavity; Superconductivity; Photocathode; Laser; Electron gun
1. Introduction In recent years, there has been a renewed interest in high average power FELs and, as a result, high average current photoinjectors. Some effort is made, to increase the duty factor of normal conducting RF guns at the price of great cooling problems, high demands on klystron power and low energy conversion efficiency. *Corresponding author. Tel.: +49-351-260-3548; fax: +39351-260-3690. E-mail address: [email protected] (D. Janssen).
A more elegant way would be to combine the high brightness of RF guns with the low RF losses of superconducting cavity technology. Besides, superconducting cavities offer an additional merit. The good vacuum conditions should increase the lifetime the of very sensitive photo cathodes. A disadvantage of superconducting technology is its sensitivity with respect to magnetic fields. In Ref. [1] it is shown, that instead of a magnetic field which is applied in normal conducting RF guns, the RF focusing can be used in a properly designed superconducting cavity to achieve small emittance and short bunch length at high-bunch charge.
0168-9002/03/$ - see front matter r 2003 Elsevier Science B.V. All rights reserved. doi:10.1016/S0168-9002(03)00936-7
ARTICLE IN PRESS D. Janssen et al. / Nuclear Instruments and Methods in Physics Research A 507 (2003) 314–317
The first proposal on a superconducting photoelectron source has been published in Ref. [2]. The behavior of photocathodes inside a superconducting cavity has been investigated in Ref. [3]. In 1998 a cooperation between German and Russian institutes started the development of a superconducting RF gun based on a half-cell TESLA cavity. First RF measurements with this cavity in a test cryostat at DESY showed excellent values for field strength and Q value [4].
2. Description of the gun The cavity of the RF gun is a TESLA type [5] half-cell closed by a shallow cone with a centered hole in which the cathode is situated. A special support structure isolates the cathode thermally and electrically from the surrounding cavity and a four-step coaxial filter prevents the transfer of RF power through the coaxial line around the cathode stem. The cathode itself is normal conducting and is held at liquid nitrogen temperature. A detailed description of the cavity has been published in Ref. [6]. An overview of the gun is shown in Fig. 1. The gun cavity, together with the cathode support, the RF coupler and a part of the beam tube are installed in the He vessel of the cryostat. This cryostat has no direct transfer line to the refrigerator. After filling the vessel by a dewar
315
with liquid helium we operate the gun at normal pressure and 4.2 K over a time of 4–5 h. The cathode can be moved by means of a manipulator from the preparation chamber into the cavity. Through a port at the beginning of the beam line the laser light is guided to the cathode. In the preparation chamber Cs2Te layers are deposited onto the cathodes. In the chamber the pressure was 1010 mbar, but during the heating of evaporators it increased to 106 mbar. For that reason the quantum efficiency of the cathodes was not better than 0.25%. The laser consists of an oscillator which works in the additive pulse modelocking (APM) mode, a five-pass diode pumped amplifier and a UV conversion stage. The oscillator supplies a pulse frequency of 26 MHz. After amplification and conversion UV laser pulses with 263 nm wavelength and 5 ps (FWHM) length are obtained. A jitter s ¼ 2 ps (peak–peak) was measured with a stable reference signal from the RF oscillator. In order to suppress the influence of fluctuations in the cavity frequency (microphonics), the laser is synchronized to the cavity resonance frequency. In this case the jitter of laser pulses increases to 12 ps (peak–peak). Beam line diagnostics include steerers, view screens, solenoids a pepper pot mask for measuring the transversal emittance, and a dipole magnet for measuring energy and energy spread. The beam current can be measured by an insulated beam dump at the end of the line.
Fig. 1. Cryostat of the SRF gun with preparation chamber.
ARTICLE IN PRESS 316
D. Janssen et al. / Nuclear Instruments and Methods in Physics Research A 507 (2003) 314–317
3. Results of the measurements Fig. 2 shows the measurement of the quality factor of the cavity dependence on the field strength. The values are from direct RF measurement and from a comparison of measured electron energy with simulation. The maximum field strength of 22 MV/m near the cathode is limited by field emission. The insignificant difference of Q values with and without cathode shows the good performance of the four stage coaxial RF filter. Fig. 3 presents the cathode emission and accelerated (dump) current together with the corresponding electron energy as a function of the laser phase. For a phase window of 60 we obtain complete transmission and the energy reaches its maximal value at j ¼ 0 : These properties are determined by the geometry of the cavity and the field amplitude. In Fig. 4 the behavior of the energy spread with respect to the laser phase is presented. For small phase angles a spread smaller than 1% has been obtained. The measured values can be reproduced by calculation, assuming an electron bunch length between 10 and 20 ps. The influence of the bias voltage between cathode and cavity is shown in Fig. 5. This voltage gives an additional possibility to influence the beam optics and (if necessary) to suppress multipacting effects. It is interesting to remark that we can obtain both focusing and defocusing effects.
Fig. 2. Determination of the maximum field strength from RF and energy measurements.
Fig. 3. Transmission of the electron beam and the dependence of the energy from the laser phase.
Fig. 4. Energy spread as function of the laser phase.
Fig. 5. Focusing effect of the bias voltage.
ARTICLE IN PRESS D. Janssen et al. / Nuclear Instruments and Methods in Physics Research A 507 (2003) 314–317
4. Conclusion For the first time we have shown that a RF electron source, where a photo cathode is inside a superconducting cavity, can work stably over a period of seven weeks (approximately 5 h/day). At a temperature of 4.2 K we could not see any change in the quality factor Q ¼ 2:5 108 of the cavity during the whole period of operation. The field emission of the cavity, which is the reason for the limitation of the field strength, is caused by the difficult clean room handling of the big intrinsic parts of the cryostat. But nevertheless a field strength of 22 MV/m and an electron energy of 900 keV have been obtained. The maximum bunch charge was 20 pC, which corresponds to an average current of 520 mA in the cw-mode. It is limited by average power and repetition rate of the laser and by the small quantum efficiency of the photocathode. Due to the long drift space after the gun and the arrangement of optical elements, we could measure the transversal emittance for bunch charges between 1 and 4 pC only. In agreement with
317
PARMELA calculation we have measured normalized rms-values between 1 and 2.5 mm mrad.
Acknowledgements We would like to thank Prof. A. Schwettmann and Prof. T. Smith, who have made their cryostat and the RF coupler available for our experiments. Furthermore we thank Prof. E. Grosse, K. Jordan, P. Kneisel and B. Schneider for many helpful discussions.
References [1] D. Janssen, V. Volkov, Nucl. Instr. and Meth. A 452 (2000) 34. [2] H. Piel, et al., Proceedings of the 10th FEL Conference, Jerusalem, 1998. [3] A. Michalke, External report, WUB-DIS 92–5, Universit.at Wuppertal 1993. [4] E. Barthels, et al., Nucl. Instr. and Meth. A 445 (2000) 408. [5] A. Aune, et al., DESY 00-031, Hamburg 2000. [6] P.v. Stein, Internal report, FZR-227, Forschungszentrum Rossendorf 1998.