1

2 Experience with coating of low-gap chambersR. Kersevan, CERN-TE-VSC-VSM With input from many people, too many to list all of them here… see individual slides… 12/09/2016

3 ALERT-2016– R. Kersevan, 16 Sept 2016Introduction: The low-emittance light sources of the new generation are among us... welcome MAX- IV! … and hopefully soon it will be another welcome to SIRIUS… and then to all of the machine upgrades which are already funded or close to being funded, ESRF-EBS, APS- U… just to name a few I am familiar with. Can we say that vacuum will never be the same? I’m not sure. MAX-IV has shown that a very compact lattice, with very small apertures, can be dealt with by using extensively the NEG-coating technology. Will it pay? For the moment it seems to be OK… see presentation of M. Grabski., with data on conditioning and lifetimes. On the other hand, there are some projects, like ESRF’s Extremely Bright Source, which are going down the old road, avoiding NEG-coating as much as possible. Will it pay? Only time will tell… for the time being the extreme complexity (and probably cost) of most chambers’ geometry is noteworthy (see D. Einfeld’s presentations). What is interesting for vacuum is that for these new machines the whole vacuum system has become a “low-gap chamber”, not only the IDs. Some novel ideas have surfaced which ask for extremely small diameter chambers, less than 10 mm ID, and for this kind of designs there is no alternative to NEG-coating: the specific conductance of a 10 mm circular profile is about 0.9 l·m/s for CO. For out-of-vacuum IDs, there is no alternative to NEG-coating, other than using some “expensive” design like the chamber-antechamber solution used at APS, where “expensive” means needing a lot of ancillary longitudinal space for installing a lumped absorber and tapers, reducing the space available to IDs. ALERT-2016– R. Kersevan, 16 Sept 2016

4 Brief history of NEG-coated chambers on light sourcesNot much has changed since this one was presented at EPAC-2000 in Vienna: ALERT-2016– R. Kersevan, 16 Sept 2016

5 Brief history of NEG-coated ChambersThe obvious conclusion was: “NEG-coating on narrow-gap extruded aluminium chambers works very well!” Proof was that SOLEIL’s design was based on massive NEG-coating (~ 60% of the ring) and it did work too, although in retrospect SOLEIL people would have avoided the pumping ports (un-coated) altogether, because they have been determined to be a major source of outgassing (scattered photons entering the pumping grids… see further below) Few years after the first paper, the situation at the ESRF was this: SOLEIL ID “Ultimate LS” concept, ESRF (30x20) “10 mm” ID (57x8) “15 mm” ID (74x11) ALERT-2016– R. Kersevan, 16 Sept 2016

6 … and another confirmation:Success explained: … and another confirmation: Conclusions: “Judging by the experiences from MAX II, NEG-coated vacuum chambers, including dipole vacuum chambers, do not appear to have any negative impact on the performance and operation of a synchrotron light source.” ALERT-2016– R. Kersevan, 16 Sept 2016

7 CERN experience and contribution:The Vacuum Surfaces and Coating group of CERN has been heavily involved in the fabrication of MAX-IV vacuum chamber, namely “the most difficult ones”, i.e. those for which industry was not ready to bid and get a contract, too risky: ALERT-2016– R. Kersevan, 16 Sept 2016

8 CERN experience and contribution:ALERT-2016– R. Kersevan, 16 Sept 2016

9 CERN experience and contribution:ALERT-2016– R. Kersevan, 16 Sept 2016

10 CERN experience and contribution:ALERT-2016– R. Kersevan, 16 Sept 2016

11 CERN experience and contribution:ALERT-2016– R. Kersevan, 16 Sept 2016

12 CERN experience and contribution:Present R&D towards coating chambers of arbitrarily small cross-sections: ALERT-2016– R. Kersevan, 16 Sept 2016

13 CERN experience and contribution:ALERT-2016– R. Kersevan, 16 Sept 2016

14 CERN experience and contribution:First results: ALERT-2016– R. Kersevan, 16 Sept 2016

15 CERN experience and contribution:First results: ALERT-2016– R. Kersevan, 16 Sept 2016

16 CERN experience and contribution:First results: ALERT-2016– R. Kersevan, 16 Sept 2016

17 CERN experience and contribution:First results: ALERT-2016– R. Kersevan, 16 Sept 2016

18 CERN experience and contribution:Temporary conclusions: ALERT-2016– R. Kersevan, 16 Sept 2016

19 CERN experience and contribution:Photodesorption experiments at KEK: (Marton Ady’s PhD thesis, EPFL/CERN) ALERT-2016– R. Kersevan, 16 Sept 2016

20 ALERT-2016– R. Kersevan, 16 Sept 2016SOLEIL contribution: SOLEIL (slides courtesy of C. Herbeaux, SOLEIL) ALERT-2016– R. Kersevan, 16 Sept 2016

21 ALERT-2016– R. Kersevan, 16 Sept 2016SIRIUS contribution: SIRIUS (slides courtesy of R.M. Seraphim, LNLS) ALERT-2016– R. Kersevan, 16 Sept 2016

22 SAES Getters contribution:SAES Getter, Milan, Italy (slides courtesy of P. Manini, see also his presentation here) Note: Specific conductance of a 4 mm pipe: 7.9E-3 l·m/s! ALERT-2016– R. Kersevan, 16 Sept 2016

23 Summary and conclusionsNEG-coating of small-gap chambers is a technology which has a very successful record, since 1999; During the years, different research groups (and industry as well, see presentation by P. Manini of SAES Getters, Milan) have found ways to deposit sufficiently uniform thin-films onto chambers with very challenging, even non uniform, cross-sections; In spite of some theoretical papers claiming that NEG-coating has/could have a potentially harmful resistive-wall impedance budget, several machines have been able to run with many NEG-coated chambers on them, without too much trouble: this point is raised again for this “run” of machine upgrades, see SLS upgrade paper presented at IPAC-16 or presentation here by M. Cox on DIAMOND’s DDBA test section; New technologies and process set-ups have allowed a number of laboratories to deposit NEG on very small vacuum chambers, see in particular ALS’s paper at IPAC-15 (10 mm ID, 1 m-long), or CERN’s “inverted” process and SAES’ results discussed here above; Some machine upgrades have adopted a completely different design philosophy, in particular ESRF, which has tried to avoid NEG-coating as much as possible: in my humble opinion, this is a choice which could potentially bring the machine to a very long conditioning phase, especially if aluminium chambers, with their attendant high photon-induced desorption yield, are going to be used (which is the case for ESRF); Same applies to studies for the SLS upgrade; There are “mixed configuration” scenarios, with a mix of coated and un-coated chambers, which are being considered by some upgrade project, namely APS-U (see nice papers at IPAC-15 and -16 by J. Carter et al. (*) ): be careful of the “SOLEIL pumping ports problem”. ... and many thanks for your patience and attention!  (*) MEDSI conference, Barcelona– R. Kersevan, 12 Sept 2016

24

25 Supporting information/bonus slidesR. Kersevan – CERN TE-VSC-IVM – “Calibrated Leak Experiment at ID6, ESRF” Fig.4 Schematics of the ID6 section, with calibrated leaks manifold (1 atm*cc/s = mbar*l/s)

26 Supporting information/bonus slidesR. Kersevan – CERN TE-VSC-IVM – “Calibrated Leak Experiment at ID6, ESRF” Fig.6 Rough data plot. NO LEAK  Ar Leak  CO Leak   CH4 Leak 

27 Supporting information/bonus slidesR. Kersevan – CERN TE-VSC-IVM – “Calibrated Leak Experiment at ID6, ESRF” Fig.9a CH4 Leak: After dumping the beam, a 5.0E-7 mbar*l/s CH4 leak is connected to the system. Note the increased lifetime due to the CH4 gas with respect to fig.7. The ID6 BS measuring ionization cell measures a slight increase of the on-axis radiation (compare with Fig.8) Only PEN1 measures part of the CO leak rate and the pressure increase induced by SR from the beam, PEN3 only the PID. This is an indication of beam- and wall-pumping. Note the increase of mass 16peak as the beam current is halved!

28 Supporting information/bonus slidesR. Kersevan – CERN TE-VSC-IVM – “Calibrated Leak Experiment at ID6, ESRF” Fig.9b CH4 Leak: Detailed view of CH4 partial pressure evolution. Major wall/beam-pumping effect!

29 Un problème non anticipé : le rayonnement produit par le NEG lui-mêmeSupporting information/bonus slides: SOLEIL (court. C. Herbeaux) Un problème non anticipé : le rayonnement produit par le NEG lui-même Conclusions: « avoid quaternary NEG alloys (Ti-Zr-V-Hf) because Hf has an even fluorescence photons of higher energy »

30 Endommagements d’équipements dû à du rayonnementSupporting information/bonus slides: SOLEIL (court. C. Herbeaux) Endommagements d’équipements dû à du rayonnement Vieillissement rapide d’équipements: Isolants de câbles devenant rigides, cassants et friables Sextupoles (amont/aval) Sondes deTemperature Cables de BPM Cables/isolants sont remplacés avec des matériaux plus résistants aux rayonnements

31 Endommagements d’équipements dû à du rayonnementSupporting information/bonus slides: SOLEIL (court. C. Herbeaux) Endommagements d’équipements dû à du rayonnement Vieillissement rapide d’équipements: Dégradations des films d’étuvage : la colle entre les couches se désagrège Leur remplacement nécessite d’ouvrir les aimants, retirer les chmabres à vide, décoller les anciens films et en recoller de nouveaux et réétuver pour activer les NEG. Ceci nécessite du temps et des ressouces ainsi qu’une solution fiable pour remplacer

32 Absorption of the radiation emitted in the bending magnetSupporting information/bonus slides: SOLEIL (court. C. Herbeaux) Absorption of the radiation emitted in the bending magnet The photons emitted in the dipole impinge the vacuum chambers on different locations : NEG coated Quad. VCs CROTCH BENDING MAGNET LONG. ABS e- trajectrory Photon distribution The crotch: first 102 mrad, 7.6 kW at 500 mA The longitudinal absorber: next 69 mrad, 5.1 kW The downstream quadrupole VC’s: last 25 mrad,1.8 kW Dph ~ 3, ph.mm-1.s-1.mA-1 c = 8,6 keV

33 Supporting information/bonus slides: SOLEIL (court. C. Herbeaux)X-ray fluorescence Chambre à vide en aluminium avec une couche de 1µm de NEG TiZrV( ) Aluminium transmission factor X-rays Fluorescence Quadrupole vacuum chamber profile Primary photons On voit une parfaite corrélation entre la distribution de dose mesurée avec les films Gafchromic films and le facteur de transmission calculé pour des rayons X d’énergie 15 keV X-rays (en fonction de l’épaisseur d’aluminium traversée)