Upgraded multiprobe sample inserts for thin film SRF cavity developments

UPGRADED MULTIPROBE SAMPLE INSERTS FOR THIN FILM SRF CAVITY DEVELOPMENTS∗ D. Seal1,2,†, C. Benjamin1, O.B. Malyshev1, K. Sian1, L. Smith1, R. Valizadeh1, ASTeC, UKRI/STFC Daresbury Laboratory, Daresbury, Warrington, UK G. Burt1, N. Leicester1, Engineering Department, Lancaster University, Lancaster, UK J. Wilson, TD, UKRI/STFC Daresbury Laboratory, Daresbury, Warrington, UK 1also at Cockcroft Institute, UKRI/STFC Daresbury Laboratory, Daresbury, Warrington, UK 2also at Engineering Department, Lancaster University, Lancaster, UK Abstract Optimisation of thin film (TF) coating parameters for producing SRF cavities requires rapid testing of superconducting properties. A dedicated multiprobe facility built at Daresbury Lab, based on a liquid He free cryocooler, allows such measurements to be performed. The facility has vacuum tubular inserts where the sample probe is loaded and cooled with He gas. The experimental inserts were either newly built or upgraded: (1) A DC resistance experiment allows measurements of critical temperature ( 𝑇c ) and residual resistance ratio ( 𝑅𝑅𝑅 ) on non-conductive substrates (e.g. sapphire). A newly designed insert allows better temperature control and easier sample change. (2) A new insert for magnetic field measurements of 𝑇c on both conductive and non-conductive substrates. (3) An existing insert for planar magnetic field penetration experiments was significantly redesigned. It operates at lower temperatures (> 5.5 K), parallel magnetic fields < 600 mT, increased sensitivity, and enables measurements of field of first flux penetration ( 𝐵fp ) and 𝑇c on various substrates: copper and sapphire, the latter of which was impossible to measure with an older design. INTRODUCTION Thin film (TF) coated superconducting radio frequency (SRF) cavities are under development in a number of accelerator laboratories worldwide [1, 2]. Before a TF coating can be produced on an RF cavity, the deposition process is optimised on small samples. These samples are then evaluated with surface analysis and cryogenic techniques. At Daresbury Laboratory (DL), three in-house developed facilities based on two stage 4 K LHe-free cryocoolers have been built. These are used to evaluate the superconducting (SC) properties of TFs deposited on both conductive and nonconductive substrates [3]. Two of these facilities measure samples under: DC conditions [4, 5] or RF conditions [6–8]. The third facility is a multiprobe cryostat for quick tests [9]. Inside this facility, two tubes, known as Vacuum Tubular Inserts (VTI), are thermally connected to the main cryogenic stages. The VTIs are filled with He gas during sample loading and removal, without the need to warm and vent the main chamber. This process enables multiple sample loads ∗ This work has been supported by: the IFAST collaboration which has received funding from the European Union’s Horizon 2020 Research and Innovation programme under Grant Agreement No 101004730. †[email protected] and tests per day. This paper describes recent upgrades on three sample inserts used in the multiprobe facility. SAMPLE INSERTS Four-Point Probe The purpose of a Four-Point Probe (FPP) is a quick, first evaluation of the deposited TF sample (on non-conducting substrate) to answer: whether the sample is SC, what is its critical temperature ( 𝑇c ) and residual resistance ratio ( 𝑅𝑅𝑅 ). The sample holder of the FPP insert was redesigned (Fig. 1) to allow a wider range of sample sizes, easier sample change, more accurate temperature readings and easier rewiring. Figure 1: FPP sample holder. (a) Sample mount, (b) PCB. The sample holder a Cu rod (20 mm dia., 60 mm length). This ensures that the whole piece is thermalised. A heater (10Ω resistor) and a Cernox thermometer (Lake Shore CX1050-CU-HT-1.4L) are located at the bottom and mounted on opposite sides. The 4-point contact piece is a square spring-loaded PCB with 2.54 mm contact pitch. Two pins apply a current ( 𝐼app ) and two measure the voltage. It is mounted on a PCB along with other wiring and cables to the top of the insert. A sample (max 7 mm × 7 mm) is mounted opposite the PCB, so its TF coated side is in contact with the 4 pins. It is held in place with a Cu plate which also ensures strong thermalisation with the Cu rod. It operates at sample temperatures in the range of 4 K ≤ 𝑇s≤ 80 K. An example of 𝑇c measurements of V 3 Si, NbN and NbTiN (at 𝐼app = 1 mA) are shown in Fig. 2. 𝑇c is measured at 50% maximum (with the 10% and 90% points defining the transition width). Here, 14.24 ± 0.10 K for V 3 Si, 𝑇c= 15.74 ± 0.25 K for NbN and 𝑇c= 16.48 ± 0.02 K for NbTiN. The optimal temperature ramp rate was found by 15th International Particle Accelerator Conference,Nashville, TN JACoW Publishing ISBN: 978-3-95450-247-9 ISSN: 2673-5490 doi: 10.18429/JACoW-IPAC2024-WEPS38 MC7.T07 Superconducting RF 2775 WEPS: Wednesday Poster Session: WEPS WEPS38 Content from this work may be used under the terms of the CC BY 4.0 licence (© 2024). Any distribution of this work must maintain attribution to the author(s), title of the work, publisher, and DOI. comparing the results of different rates of ramp up/down cycles for the same sample. It was found that a d 𝑇s /d 𝑡 = 0.1 K/min ensures consistent measurements for a ramp up/down cycle (with an accuracy ±0.01 K). The agreement between measurements of increasing and decreasing temperature sweeps also indicate that 𝑇s and the thermometer temperature are the same. Figure 2: 𝑇c measurements with the FPP for NbN and V 3 Si. It should be noted that this system can measure 𝑇c and 𝑅𝑅𝑅 on a TF sample ( < 5 μ m) only. Thicker samples would require a higher 𝐼app , another contact design, and/or a larger distance between the pins. The method is suitable only for TFs on electrically non-conductive substrates (e.g. sapphire). In general, the typical measurement time at 3 different currents (to extrapolate to 𝐼app = 0) is 2 − 3 hours. Magnetic 𝑇c The 𝑇c of the samples deposited on electrically conductive substrates (i.e. Cu and Nb) can be measured with magnetic methods. The main idea for a new insert (Fig. 3) is to place a sample (max 10 mm × 10 mm) at the end of a coil and apply a DC magnetic field ( 𝐵app ). Measurements of magnetic field are made using two Hall probes (Lakeshore HGCA3020): in the bore of the coil ( 𝐵1 ) and on the other side of the sample ( 𝐵2 ). The coil is wound with 0.1 mm wire with a total number of turns 𝑁 = 256 , allowing 𝐵 ≈ 4.5 mT (𝐼app = 100 mA). Figure 3: The magnetic 𝑇csample holder. An example of 𝐵2/𝐵1 measurements of Nb 3 Sn are shown in Fig. 4 for 𝐵app = 12 mT. 𝑇c is calculated from the 50% maximum point (with 10% and 90% points for the transition width). Here, 𝑇c= 15.49+0.66 −0.52 K. Typically, multiple sweeps of 𝑇s are recorded at different values of 𝐵app and linearly fitted as a function 𝑇2 s . Extrapolation to 𝐵app = 0 allows calculation of 𝑇c using the same method as with the magnetic field penetration insert. The typical time required for sample measurements at 3 different currents in the coil, i.e. with 3 different magnetic fields (to extrapolate to 𝐵app = 0 ) is 2 − 3 hours. Figure 4: 𝐵2/𝐵1 vs 𝑇s measured with the magnetic 𝑇c insert for Nb3Sn. Magnetic Field Penetration Magnetic Field Penetration (MFP) measurements are based on the simple idea of simultaneously applying a magnetic field: (a) parallel to the sample surface, (b) from one side of the sample (i.e. similar to the magnetic field in an RF cavity) and (c) locally (within an area < sample size) to avoid or minimise the leakage of magnetic field around the sample. Two facilities have been designed and built at DL for MFP measurements: (a) an MFP insert for the multiprobe facility [10, 11] and (b) a dedicated MFP facility [4, 5]. Figure 5: The upgraded MFP sample holder (heaters - H1−H4, thermometers - T1and T2). 15th International Particle Accelerator Conference,Nashville, TN JACoW Publishing ISBN: 978-3-95450-247-9 ISSN: 2673-5490 doi: 10.18429/JACoW-IPAC2024-WEPS38 2776 MC7.T07 Superconducting RF WEPS38 WEPS: Wednesday Poster Session: WEPS Content from this work may be used under the terms of the CC BY 4.0 licence (© 2024). Any distribution of this work must maintain attribution to the author(s), title of the work, publisher, and DOI. The layout of the MFP insert sample holder is shown in Fig. 5. The magnetic field is generated with a C-shaped magnet with 10 mm × 10 mm poles and a 2 mm gap between them. The applied magnetic field ( 𝐵1 ) is measured in the gap between the poles with a Hall probe (HP1), and the penetrated field ( 𝐵2 ) is detected on the other side of the sample (max 50 mm dia., 0.5 − 2 mm thick) with another Hall probe (HP2). The Hall probes are Lakeshore HGCT3020. Recently, the MFP insert was redesigned and tested with two main modifications: (a) Increased length of the insert to reach a lower 𝑇s , (b) magnetic shields (MS1 and MS2) around HP2. Smaller samples can also be evaluated, but not < 30 mm in any dimension. Figure 6 shows an example of the raw data obtained with a 50 mm dia. bulk Nb disk. Measurements of field of first flux penetration ( 𝐵fp ) are made at each value of 𝑇s . This is defined as when 𝐵2= 0.1 mT. A linear fit to 𝐵fp(𝑇2 s) (Fig. 7) allows for an estimation of 𝐵fp (0 K) and 𝑇c . For this sample: 𝐵fp(0 K) = 514.6 mT and 𝑇c= 9.23 K using this method. Figure 6: 𝐵1 vs 𝐵2 of a bulk Nb sample with the MFP insert. Figure 7: 𝐵fp vs 𝑇2 sfor the bulk Nb sample with a linear fit to estimate 𝐵fp(0 K) and 𝑇c. The main characteristics of this upgraded insert are: 𝑇s≥ 5.5 K, applied magnetic field 0 ≤ 𝐵app ≤ 600 mT, magnetic field sensitivity 10 μ T and typical time required for full sample evaluation of 𝐵fp and 𝑇cis 1 day. LABVIEW AUTOMATED DATA ACQUISITION Automated data acquisition is essential to standardise the measurement procedure and free up the time of the staff for data analysis. For data recording, LabVIEW is used to allow for a full data set for the sample to be collected without human intervention. Therefore, measurements can be performed continuously at any time of the day, allowing the most efficient use of the multiprobe facility. DISCUSSION Enabling SC sample evaluation is essential for developing TF superconductors for SRF cavities. The three inserts described in this paper provide a complete set for initial evaluation on small planar samples. The FPP and Magnetic 𝑇c inserts are critically important to address: (a) Is the sample superconducting? (b) If yes, what is its 𝑇c and 𝑅𝑅𝑅 ? The properties of a TF deposited on different substrates (e.g. sapphire and Cu) could be different. The FPP is good for measuring a TF sample on sapphire, but to measure a TF on Cu, Kapton tape and nitric acid must be used to remove the TF beforehand. The question of how etching affects the TF properties remains open. A new Magnetic 𝑇c insert avoids this work and makes the test faster and more reliable. The MFP insert provides evaluation at higher 𝐵 fields, but requires slightly larger samples than the other inserts. Compared to the dedicated MPF facility, the MPF insert is limited to a maximum sample size and cannot reach sample temperatures below 5.5 K (compared to 2.5 K in the MFP facility). However, the measurements can be performed in a shorter time: 1 day instead of 2 days. CONCLUSIONS Three experimental inserts were newly constructed or upgraded for operation with a multiprobe facility. (1) A DC resistance experiment allows measurements of 𝑇cand 𝑅𝑅𝑅 on non-conductive substrates, newly designed for better temperature control and easier sample change. (2) A new insert for magnetic field measurements of 𝑇c on both conductive and non-conductive substrates. (3) An existing insert for magnetic field penetration experiments was significantly redesigned. Compared to previously, it operates at lower temperatures (> 5.5 K), parallel magnetic fields < 600 mT, increased sensitivity and enables measurements of 𝐵fp and 𝑇c on various substrates. Magnetic shielding of the Hall probe allows for an improved accuracy and simpler analysis of the results. ACKNOWLEDGEMENTS The authors acknowledge support received by members of ASTeC, namely A. Vick, J. Conlon, A. Palmer, A. Wootten. We also acknowledge work carried out by staff in the Engineering Technology Centre at Daresbury Laboratory. D. Seal would like to thank UKRI for his PhD funding. 15th International Particle Accelerator Conference,Nashville, TN JACoW Publishing ISBN: 978-3-95450-247-9 ISSN: 2673-5490 doi: 10.18429/JACoW-IPAC2024-WEPS38 MC7.T07 Superconducting RF 2777 WEPS: Wednesday Poster Session: WEPS WEPS38 Content from this work may be used under the terms of the CC BY 4.0 licence (© 2024). Any distribution of this work must maintain attribution to the author(s), title of the work, publisher, and DOI. REFERENCES [1] C. Antoine, “R&D in superconducting RF: thin film capabilities as a game changer for future sustainability,” in Proc. IPAC’23, Venice, Italy, May 2023, pp. 21–26. doi:10.18429/JACoW-IPAC2023-MOYD3 [2] C. Pira et al., “Progress in European Thin Film Activities,” in Proc. SRF’23, Grand Rapids, MI, USA, Jun 2023, pp. 607– 614. doi:10.18429/JACoW-SRF2023-WECAA01 [3] D. Seal et al., “Characterisation facilities for evaluating superconducting thin films for SRF cavities,” in Proc. IPAC’23, Venice, Italy, May 2023, pp. 2983–2986. doi:10.18429/JACoW-IPAC2023-WEPA141 [4] D. A. Turner et al., “A facility for the characterisation of planar multilayer structures with preliminary niobium results,” Supercond. Sci. Technol., vol. 35, no. 9, p. 095 004, Jul 2022. doi:10.1088/1361-6668/ac7fbf [5] L. Smith et al., “Investigation, using Nb foils to characterise the optimal dimensions of samples measured by the magnetic field penetration facility,” in Proc. SRF’23, Grand Rapids, MI, USA, Jun 2023, pp. 88–91. doi:10.18429/JACoW-SRF2023-MOPMB012 [6] O. B. Malyshev et al., “The SRF thin film test facility in LHefree cryostat,” in Proc. SRF’19, Dresden, Germany, Jun-Jul 2019, pp. 610–612. doi:10.18429/JACoW-SRF2019-TUP070 [7] D. Seal et al., “RF characterisation of bulk niobium and thin film coated planar samples at 7.8 GHz,” in Proc. LINAC’22, Liverpool, UK, Aug-Sep 2022, pp. 818–821. doi:10.18429/JACoW-LINAC2022-THPOGE10 [8] D. Seal et al., “Optimisation of niobium thin film deposition parameters for SRF cavities,” in Proc. SRF’23, Grand Rapids, MI, USA, Jun 2023, pp. 253–258. doi:10.18429/JACoW-SRF2023-MOPMB062 [9] O. B. Malyshev et al., “Design, assembly and commissioning of a new cryogenic facility for complex superconducting thin film testing,” in Proc. IPAC’18, Vancouver, Canada, Apr-May 2018, pp. 3859–3861. doi:10.18429/JACoW-IPAC2018-THPAL089 [10] O. B. Malyshev et al., “A facility for magnetic field penetration measurements on multilayer S-I-S structures,” in Proc. SRF’15, Whistler, Canada, Sep 2015, pp. 716–719. doi:10.18429/JACoW-SRF2015-TUPB059 [11] O. B. Malyshev et al., “First Results of Magnetic Field Penetration Measurements on Multilayer S-I-S Structures,” in Proc. IPAC’16, Busan, Korea, May 2016, pp. 2245–2247. doi:10.18429/JACoW-IPAC2016-WEPMB057 15th International Particle Accelerator Conference,Nashville, TN JACoW Publishing ISBN: 978-3-95450-247-9 ISSN: 2673-5490 doi: 10.18429/JACoW-IPAC2024-WEPS38 2778 MC7.T07 Superconducting RF WEPS38 WEPS: Wednesday Poster Session: WEPS Content from this work may be used under the terms of the CC BY 4.0 licence (© 2024). Any distribution of this work must maintain attribution to the author(s), title of the work, publisher, and DOI.