Impact of heat treatments on the performance of low-frequency superconducting quarter-wave resonators at 4.3 K Journal: Nucl Instrum Methods Phys Res A
Jacob Brown, Sang-hoon Kim, Walter Hartung, Ting Xu · 2026
Read Impact of heat treatments on the performance of low-frequency superconducting quarter-wave resonators at 4.3 K Journal: Nucl Instrum Methods Phys Res A by Jacob Brown, Sang-hoon Kim, Walter Hartung, Ting Xu (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-00388.txt.
Introduction
Brown and three colleagues test how heat treatment changes low frequency superconducting quarter wave resonators built from bulk niobium and polished chemically. Tested near 4.3 kelvin, a two day bake at 120 degrees Celsius cuts surface resistance two to three fold by trimming one loss term. It also eases the medium field quality slope by about two fifths on average. A shorter bake at 350 degrees gives no gain, unlike results seen at higher frequencies.
Catalog record
| CREATOR | Jacob Brown, Sang-hoon Kim, Walter Hartung, Ting Xu |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 25061 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.00388
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Impact of heat treatments on the performance of low-frequency superconducting quarter-wave resonators at 4.3 K Journal: Nucl Instrum Methods Phys Res A Jacob Brown, Sang-hoon Kim, Walter Hartung, Ting Xu Abstract We applied heat treatments to 80.5 MHz quarter-wave resonators made from bulk niobium and prepared with buffered chemical polishing (BCP). We evaluated their performance at 4.3 K. We found that a 48 hour, 120 ∘ 120^{\circ} C bake-out (“low-temperature bake out”) reduces the surface resistance by a factor of 2 to 3, stemming from a reduction in the Bardeen-Cooper-Schrieffer contribution, consistent with previous findings. This decrease leads to a 38 % 38\% decrease on average in the medium-field Q Q -slope when compared to cavities which had only BCP. Mechanisms for the change in quality factor with low-temperature baking have been explored. We observed no improvement in cavity performance after a 3-hour bake-out at 350 ∘ 350^{\circ} C (“medium-temperature bake out”), in contrast to observations for higher-frequency cavities. 1 Introduction The Facility for Rare Isotope Beams (FRIB) began user operations in May 2022 46 , 47 . The driver linear accelerator (linac) consists of 324 superconducting radio-frequency (SRF) cavities to accelerate light and heavy ions (up to 200 MeV per nucleon in the case of uranium ions). Four different cavity types are needed, two quarter-wave resonator (QWR) types and two half-wave resonator (HWR) types 49 . The second cavity type is a QWR optimized for acceleration of low-velocity beams ( \beta=0.085 , 80.5 MHz). The linac includes 92 such cavities in 12 cryomodules. Figure 1 shows the field distributions for the accelerating mode calculated with CST Microwave Studio 3 . Figure 1: Intensity maps of the magnitude of the electric field (left) and magnetic field (right) for the FRIB \beta=0.085 QWR. The field amplitudes are normalized to a stored energy of U=1 Joule. The quality factor Q_{0} and surface resistance R_{s} of an SRF cavity are related via 32 Q_{0}(B_{s})=\frac{\omega_{0}U}{P_{w}(B_{s})}=\frac{G}{R_{s}(B_{s})}, (1) where U is the stored energy, \omega_{0} is the cavity resonant frequency, P_{w} is the power dissipation in the cavity, and G is the geometry factor; R_{s} is the average RF surface resistance of the cavity (weighted according to the distribution of the square of the surface magnetic field B_{s} ). The left equation is generally valid, but the right equation is strictly correct only when there are no additional loss mechanisms in the cavity, such as field emission or multipacting 11 , 33 . In the absence of additional loss mechanisms, P_{w} is proportional to R_{s} , so reducing R_{s} decreases P_{w} , and therefore decreases the load to the cryogenic plant. As such, there has been a sustained effort in the SRF community to reduce R_{s} and mitigate the Q -slope (decrease in Q_{0} as accelerating gradient E_{\mathrm{acc}} increases). Imperfections, foreign particles, and contaminants on the inner surface of the cavity can produce an increase in R_{s} and may lead to field emission or early thermal breakdown. Hence careful surface preparation is needed for a high-performance cavity to remove the damage layer from cavity forming and produce the best possible surface. Standard practice is to remove \geq 100 \mu m via buffered chemical polishing (BCP). Electro-polishing (EP) is an alternative to BCP which generally produces lower surface roughness 41 , 9 , 42 , 18 , 53 . Baking for \sim 48 hours at 120^{\circ} C has been shown to mitigate high-field Q -slope at 2 K for electropolished elliptical cavities over a range of frequencies 19 , 5 , 4 . This low-temperature baking (LTB) has hence been incorporated as a standard preparation step for the International Linear Collider 35 and the European X-ray Free Electron Laser facility 1 . Other treatments have been shown to produce higher Q_{0} and thus improve the cavity performance. For cavities with frequencies >1 GHz, nitrogen doping has been found to be an attractive treatment. Briefly stated, the cavity is heat treated in an ultra-high vacuum (UHV) furnace, nitrogen is introduced into the furnace during the treatment, and then harmful surface nitrides are removed via light EP 12 , 7 , 6 . At GHz frequencies, N-doping results in an increase in Q_{0} as a function of E_{\mathrm{acc}} (“anti- Q -slope”). This anti- Q slope has been shown to have a frequency dependence, and has not been observed in sub-GHz cavities after N-doping, even though Q_{0} improves 26 , 29 . Recently, heat treatment at 300 to 350^{\circ} C for 3 hours (medium-temperature bake, MTB) has been shown to increase Q_{0} and produce anti- Q -slope for 1.3 GHz TeSLA-style cavities at 2 K 38 , 17 , 15 , 50 . For medium-frequency cavities (500 MHz to 1 GHz), MTB results in higher Q_{0} at 2 K, in line with N-doping effects for such cavities 48 , 29 . For low- \beta cavities operating at 4.3 K, LTB is often used and is considered helpful to soften multipacting barriers. For example, LTB was applied for all of the QWRs for SPIRAL2 23 , 25 , 24 and was incorporated for ISAC-II at TRIUMF 51 . Both groups observed a lessening of the Q -slope at \sim 4.3 K with LTB. Recent studies at TRIUMF on multi-mode coaxial resonators prepared with BCP have shown that the effectiveness of MTB and LTB has a frequency dependence 21 , 20 . At 4.3 K, R_{s} was reduced by MTB for higher-frequency modes ( > 500 MHz), while MTB did not improve the performance for low-frequency modes (220 MHz, 390 MHz). Their work focused on the relationship between the effectiveness of different heat treatments and the frequency of the electromagnetic mode excited in two different coaxial resonators. In our studies, we sought to understand the cause of the improvement in medium-field Q -slope for many low-frequency quarter-wave resonators due to LTB as as well as explore the effect of MTB on a FRIB production resonator. Originally, it was planned to operate the FRIB QWRs at 2 K, co-habitating with superconducting solenoids at 4.3 K in the same cryomodules. Hence, during FRIB cavity production, LTB was not included as a routine preparation step, as it was not found to be beneficial for 2 K operation 52 (as can be seen in Fig. 2 ). However, during linac commissioning, we found that the QWRs could be stably operated at 4.3 K, which allowed for simpler cryogenic operation of the linac. (Though the 322 MHz FRIB HWR cryomodules still operate with cavities at 2 K and solenoids at 4.3 K.) For the present FRIB operating configuration, LTB would provide improved performance for QWRs, so an LTB step is now included for spare and refurbished QWR cryomodules. As can be seen in Fig. 2 , there is a clear performance improvement for FRIB QWRs at 4.3 K due to LTB. Figure 2: Cavity intrinsic quality factor ( Q_{0} ) as a function of accelerating gradient ( E_{\mathrm{acc}} ) for several \beta=0.085 QWRs after BCP (black) or BCP+LTB (red) at 4.3 K (solid lines) and 2 K (dashed lines). The peak surface magnetic field ( B_{\mathrm{pk}} ) is also shown. The star represents the FRIB 2 K cryomodule goal. 2 Cavity Preparation FRIB cavities were formed from