New results on the Argon purification Grzegorz Zuzel

New results on the Argon purification Grzegorz Zuzel Max-Planck-Institut für Kernphysik, Heidelberg IDEA Meeting, 19. 11 -20. 11. 2007, Paris, France

Outlook o o o o Applications of ultra-pure gases Purification techniques Detection of 222 Rn, 39 Ar and 85 Kr Argon purity and purification A new purification plant for LAr Test of a nitrogen delivery chain Summary IDEA Meeting, 19. 11 -20. 11. 2007, Paris, France

Applications of ultra-pure gases (GERDA) o o o o Tasks: - cooling of Ge crystals - shielding against external radiation Requirements: - low boiling point - high purity Base line: LAr, option: LN 2: experience from BOREXINO LAr: better shielding properties, scintillation Removal of radio-isotopes from (L)N 2/(L)Ar crucial Developed techniques can be applied in other projects (EXO, XENON, XMASS, WARP, CLEAN, …) IDEA Meeting, 19. 11 -20. 11. 2007, Paris, France

Applications of ultra-pure gases (general) o o o o Clean rooms (222 Rn – reduced atmosphere) Clean benches - detectors assembly - 222 Rn-free atmosphere (nitrogen) Blanketing Cleaning 222 Rn emanation (cleanliness) measurements IDEA Meeting, 19. 11 -20. 11. 2007, Paris, France

Purification techniques o Distillation: n n o Adsorption: n n o Relatively cheap Successfully applied for 222 Rn removal (BOREXINO) Sparging n n o High costs and high energy consumption Big plants Pure gas needed anyway Boiling points are important Buy ultra-pure gases: n n If commercial products fulfill requirements If purity can be kept during transportation IDEA Meeting, 19. 11 -20. 11. 2007, Paris, France

Purification by adsorption Ideal situation: VN N 2/LN 2 + impur. N 2/LN 2 Real situation: N 2/LN 2 + impur. V 1 N 2/LN 2 + impur. V 2 N 2/LN 2 + some impur. . N 2/LN 2 + impur. V 3 IDEA Meeting, 19. 11 -20. 11. 2007, Paris, France N 2/LN 2 + impur.

Purification by adsorption n = H p IDEA Meeting, 19. 11 -20. 11. 2007, Paris, France

Possible contaminations Decay mode Source Concentration (STP) 222 Rn T 1/2 = 3. 8 d , , Primordial 238 U 1 - ? 00 Bq/m 3 air 85 Kr T 1/2 = 10. 8 y (687 ke. V) , 39 Ar T 1/2 = 269 y (565 ke. V) 42 Ar T 1/2 = 32. 9 y (600 ke. V) o o o 235 U fission (nuclear fuel reprocessing plants) 1. 4 Bq/m 3 air 1. 2 MBq/m 3 Kr Cosmogenic 17 m. Bq/m 3 air 1. 8 Bq/m 3 Ar Cosmogenic 0. 5 µBq/m 3 air 50 µBq/m 3 Ar Q-value of 39 Ar and 85 Kr below 700 ke. V – relevant in case of dark matter detection Dead time could be a problem when Ar scintillation is used (slow decay time: ~ 1µs) 42 Ar is naturally low IDEA Meeting, 19. 11 -20. 11. 2007, Paris, France

Detection of 222 Rn, 39 Ar and 85 Kr Detection limit 0. 5 µBq/m 3 (1 atom in 4 m 3) IDEA Meeting, 19. 11 -20. 11. 2007, Paris, France

Detection of 222 Rn, 39 Ar and 85 Kr Mass spectrometry of Ar and Kr Ar: 10 -9 cm 3 (1 ppb; ~1. 4 n. Bq/m 3 for 39 Ar in N 2) Kr: 10 -13 cm 3 (0. 1 ppt; ~0. 1 µBq/m 3 for 85 Kr in N 2) IDEA Meeting, 19. 11 -20. 11. 2007, Paris, France

Nitrogen purification from Kr and Ar IDEA Meeting, 19. 11 -20. 11. 2007, Paris, France

Nitrogen purification from Kr at -186 C IDEA Meeting, 19. 11 -20. 11. 2007, Paris, France

222 Rn emanation of carbon adsorbers Adsober Henry‘s constant [mol/Pa/kg] Synthetic carbon Carbo. Act 0. 21 ± 0. 02 0. 3 ± 0. 1 Carbosieve SIII (molecular sieve) 0. 34 ± 0. 02 0. 7 ± 0. 2 o o 222 Rn emanation rate [m. Bq/kg] Kr adsorption ability and 222 Rn emanation rate comparable Carbo. Act is final choice (grain size, availability and prize) IDEA Meeting, 19. 11 -20. 11. 2007, Paris, France

Argon purity studies o o Argon, originally as a backup solution, became the GERDA base line Scope of our project was extended to investigations of Argon purity and purification Only 222 Rn seems to be relevant (39 Ar and 85 Kr can influence only the dead time of the detector). Allowed concentration at the level of 1 Bq/m 3 (STP, B ~ 10 -4 c/kg/y/ke. V) Purification tests were done using Carbo. Act as an adsorber IDEA Meeting, 19. 11 -20. 11. 2007, Paris, France

222 Rn in Argon Company Quality 222 Rn [m. Bq/m 3] Air Liquide LINDE AG Westfalen AG Ar 4. 8 Ar 5. 0 Ar 4. 6 0. 3 0. 4 0. 7 Westfalen AG Ar 6. 0 Ar 5. 0 0. 4 8. 4 • Ar initially less pure than nitrogen (~0. 05 m. Bq/m 3) • Systematic effect due to production in air separation plants and storage (distribution) IDEA Meeting, 19. 11 -20. 11. 2007, Paris, France

222 Rn removal from nitrogen IDEA Meeting, 19. 11 -20. 11. 2007, Paris, France

60 g 150 g 222 Rn removal from nitrogen Volume [m 3] Initial conc. [m. Bq/m 3] Final conc. [ Bq/m 3] Reduction factor [1/kg] 141 0. 197 0. 004 <0. 5 >2700 80 0. 27 0. 02 0. 7 0. 3 2600 150 48 0. 15 0. 01 3. 3 0. 6 740 150 104 0. 11 0. 01 5. 6 0. 6 330 40 140 0. 21 0. 01 5. 2 0. 5 660 80 200 6. 0 0. 1 600 23 170 10 • Purity at the 1 Bq/m 3 level achieved • Gas phase purification is more efficient • 10 times bigger trap than 60 g-trap used in the liquid phase adsorption should be sufficient to stand 1 m 3 of LAr IDEA Meeting, 19. 11 -20. 11. 2007, Paris, France

Ar purification plant IDEA Meeting, 19. 11 -20. 11. 2007, Paris, France

A new adsorption column • Size: 120 650 mm • Volume: ~5 L • Carbo. Act mass: 602 g • 2 temperature sensors • Pressure gauge • Burst disc • Cryo-valves • Metal sealed (CF-100) • Column ready for tests IDEA Meeting, 19. 11 -20. 11. 2007, Paris, France

Test of the gas delivery path o o o One of the options to get ultra-pure gases is to buy them It has to be proved that it is possible to produce them and later on deliver without loosing the purity Crucial is cleanliness of the storage and transportation tanks (222 Rn emanation) IDEA Meeting, 19. 11 -20. 11. 2007, Paris, France

222 Rn emanation from storage tanks Quality of Vol. Tank from stored gas [m 3] 222 Rn activity in specific 222 Rn saturation act. [m. Bq] [m. Bq/m 3] Westfalen AG technical 3 177 ± 6 59 ± 2 LINDE 7. 0 3 2. 7 ± 0. 3 0. 9 ± 0. 1 Westfalen AG 6. 0 0. 67 42 ± 2 63 ± 3 SOL 6. 0 16 65 ± 6 4. 1 ± 0. 4 IDEA Meeting, 19. 11 -20. 11. 2007, Paris, France

Testing SOL nitrogen Initial concentrations: 222 Rn: Ar: 6 ppb Kr: 0. 04 ppt Total 222 Rn budget inside tank: 65 m. Bq Converted in 222 Rn concentration: 6 Bq/m 3 (STP) IDEA Meeting, 19. 11 -20. 11. 2007, Paris, France

Testing SOL nitrogen – initial 222 Rn IDEA Meeting, 19. 11 -20. 11. 2007, Paris, France

Testing SOL nitrogen – 222 Rn, Ar and Kr Conditions CAr [ppb] CKr [ppt] CRn [µBq/m 3] VLN 2 ~14. 2 m 3 p ~10. 5 barg 12. 3 1. 7 0. 030 0. 006 7 1 VLN 2 ~14. 2 m 3 p ~11 barg 11. 1 1. 5 0. 016 0. 004 5. 5 0. 6 VLN 2 ~13. 5 m 3 p ~11 barg 11. 7 1. 6 0. 011 0. 004 7. 6 0. 7 VLN 2 ~6 m 3 p ~15 barg 7. 4 1. 1 0. 006 0. 004 10 1 VLN 2 ~3 m 3 p ~15 barg 12. 3 1. 8 0. 024 0. 006 11 1 VLN 2 ~0. 8 m 3 p ~15 barg 16 2 0. 031 0. 008 38 5 Tank empty p ~15 barg 19 3 0. 03 0. 01 237 15 • Is it possible to keep high purity of the gas during transportation • Ar, Kr and 222 Rn do not change with the filling level of the tank • The tank should not be drained! (222 Rn) to keep high purity of a gas IDEA Meeting, 19. 11 -20. 11. 2007, Paris, France

Summary o o o Theoretical studies helped to select proper adsorbing materials and purification techniques For a new purification system Carbo. Act was selected due to low 222 Rn emanation rate, good adsorption properties and reasonable prize Project has been extended for Ar purification (gas/liquid phase) A new adsorption column has been designed and constructed according to the results obtained for a 60 -g trap Alternative concept: no purification – instead avoiding contaminations. Full delivery chain successfully tested for nitrogen. Ar, Kr and 222 Rn can be kept under control at very low level! Future: tests of the 5 -l column IDEA Meeting, 19. 11 -20. 11. 2007, Paris, France