Outline Fermi experimental setup Impact of the laser heater on seeded FEL Non-Gaussian effects 2

Transcription

1 Benefits of the laser heater induced energy spread for high harmonic conversion in HGHG FEL E. Ferrari 1,2, G. Penco 1, E. Allaria 1, S. Spampinati 1, L. Giannessi 1,3, W. Fawley 1, Z. Huang 4 1. Elettra - Sincrotrone Trieste; 2. Universita' degli Studi di Trieste; 3. ENEA C.R. Frascati 4. SLAC National Accelerator Laboratory 1

2 Outline Fermi experimental setup Impact of the laser heater on seeded FEL Non-Gaussian effects 2

3 FERMI: a Seeded High Gain Harmonic Generation FEL Modulator Dispersive section Radiator 3

4 Experimental setup LH Fermi scheme, 3, 4 LH,3,4 Laser heater setup (from Z.Huang., FEL9) 4

5 For more details on the FERMI scheme 5

6 Laser heater FEL intensity FEL intensity [µj] FEL intensity vs. Laser Heater ub instability suppression Optimal value.6.8 uj too much energy spread FEL degradation LH energy [µj] 6

7 But if we enlarge the scan range 14 FEL intensity vs. heating at 32 nm FEL intensity [µj] Appearence of local maxima at large heating (x1 with respect to optimal) LH energy [µj] 7

8 The induced energy spread is monotonic FEL intensity [µj] FEL intensity and Energy spread vs. heating LH energy [µj] LH Induced energy spread [KeV] 8

9 The effect can be dramatic! 4 FEL intensity vs. heating at 52 nm FEL intensity [µj] Many multiple peaks with decreasing intensity LH energy [µj] 9

10 Dependence on R56 and seed Normalized FEL intensity Change in width R 56 dependence (32 nm) LH energy [µj] R 56 = 49 µm, seed = 5.6 µj R = 64 µm, seed = 29.1 µj 56 R = 81 µm, seed = 21.3 µj 56 Shift in position 1

11 Dependence on FEL wavelength Normalized FEL intensity Wavelength dependence λ = 26 nm λ = 32 nm λ = 52 nm LH energy [µj] The number, position and relative intensity of secondary peaks can be tuned 11

12 How to explain this behaviour? 12

13 FEL vs. LH - Without gain 3 Without gain, the FEL intensity is almost proportional to the square of the bunching b m FEL intensity vs. heating (no gain, 32 nm) Exp. Data FEL intensity [µj] Coherent emission from three radiators only LH energy [µj] 13

14 Bunching (1) L. H. Yu FEL intensity [µj] FEL intensity vs. heating (no gain, 32 nm) Exp. Data Gaussian distribution Experimental results differ from what expected for a Gaussian energy spread distribution Dispersion Momentum compaction e - energy FEL wavelength Harmonic number Energy spread (rms) m-th order Bessel LH energy [µj] FEL energy modulation 14

15 Longitudinal Phase Space and heating 15

16 Non-Gaussian energy spread Same area and same second moment Time (ps) uj Charge Density (Arb. Units) Exp. Data Gaussian fit E E (MeV) Significant differences E E (kev) 16

17 Energy distribution and heating Simulated energy profile 17

18 Energy distribution and heating Simulated energy profile Normalized sim. energy profile 18

19 Energy distribution and heating Measured energy profile Normalized sim. energy profile The shape of the energy distribution is, as expected, independent on the heater power 19

20 Bunching with non-gaussian energy spread (2) Z. Huang, PRSTAB 7, 7441 (24) FEL intensity [µj] FEL intensity vs. heating (no gain, 32 nm) Harmonic number Energy spread (rms) m-th order Bessel FEL 1 energy modulation 1 1 LH energy [µj] Exp. Data Gaussian distribution As from Eq.(2) bunching suppression factor laser spot size (in LH) e - spot size (in LH) 2

21 (Almost) no gain FEL intensity [µj] L g = 1.3 m FEL intensity vs. heating (no gain, 32 nm) LH energy [µj] Exp. Data Exp. Data Gaussian distribution Gaussian As from Eq.(2) As from Eq.(2) L g = 1.7 m 21

22 Simulated impact on high-harmonic emission x 3!!! 22

23 Summary A Laser Heater is routinely used in FEL operations at FERMI. The non-gaussian distribution of the energy spread induced by the Laser Heater has been shown to be preserved up to the linac end and the undulators. The shape of the slice energy spread distribution has a significant impact on FEL intensity, as it ultimately determines the bunching. In particular, several FEL local maxima as a function of LH intensity have been observed, and can be controlled by tuning the machine parameters. The unexpected behavior is well reproduced by previously developed LH theory. 23

24 Perspectives Preliminary numerical simulations show that the non-gaussian energy spread can increase the FEL power at high harmonic (i.e. shorter wavelength) in a HGHG FEL. The significant increase in emission power could potentially extend the operation range of the single cascade HGHG scheme. 24

25 We acknowledge the support of the FERMI COMMISSIONING TEAM

26 Thanks for your attention! 26

27 Laser heater - COTR 2.5 x 15 OTR intensity [a.u.] LH energy [µj] As already observed, a small amount of heating is sufficient to dump the COTR at screens downstream the bunch compressor 27