Direct observation of nanometer size hydride precipitations in superconducting niobium | Scientific Reports

Scientific Reports volume 14, Article number: 26916 (2024) Cite this article

Metrics details

Superconducting niobium serves as a key enabling material for superconducting radio frequency (SRF) technology as well as quantum computing devices. Niobium has a high propensity for the uptake of hydrogen. At room temperature, hydrogen commonly occupies tetragonal sites in the Nb lattice as the metal (M)–gas (H) phase. When the temperature is decreased, however, a solid solution of Nb-H begins to precipitate. In this study, we show the first identified topographical features associated with nanometer-size hydride phase (Nb1−xHx) precipitates on the surface of the metallic superconducting niobium using cryogenic-atomic force microscopy (AFM). Further, high energy grazing incidence X-ray diffraction reveals information regarding the structure and stoichiometry of these precipitates. Finally, through time-of-flight secondary ion mass spectroscopy (ToF-SIMS), we locate atomic hydrogen sources near the top surface. This systematic study clarifies nanometer scale hydrides precipitated on the surface of the SRF Nb cavity that exhibit performance degradation at a high accelerating field regime.

Niobium (Nb) is a 4d transition bcc (body centered cubic) metal that acts as a marginal type II superconductor below Tc (the critical superconducting temperature) of 9.2 K with a narrow-mixed state1,2. It has the highest lower critical field, Hc1(0) ≈ 180 mT, and the longest coherence length, ξNb(0) \(\:\approx\:\) 40 nm, compared to other practical or composite superconductors3,4. Additionally, its extended low yield strength mechanical property allows to form into various desired geometries easily5. These unique properties make niobium an enabling material applicable for 3D superconducting radio frequency (SRF) resonator (cavity)5, superconducting quantum interference devices (SQUID)6,7, and most recently for quantum computing devices8,9.

In 3D SRF Nb cavities, various engineering treatments are employed to achieve high quality factor (Q0) and sustain electromagnetic accelerating field close to the theoretically expected maximum of 50 MV/m5. For instance, baking the cavity at 120 °C/48hrs10,11 and doping the cavity by exposing it to nitrogen at 120 °C for 48 h or 800 °C for 10 min12,13,14 are two such treatments which are known to significantly reduce the surface resistance in the superconducting state and improve the overall cavity performance, associated with dirty-limit superconductivity of the nano-structured SRF surface15,16. However, there is one of the impurities that is theorized to degrade SRF cavity performance at a high RF field regime: hydrogen, for which niobium acts as a getter. Hydrogen is highly soluble in the niobium17 and upon cryogenic cooling, it forms nanometer-sized Nb hydride (Nb1-xHx) precipitates on the cavity surface at a temperature < 150 K18,19. Nb hydride precipitate becomes a normal conductor at SRF cavity operation temperature, T \(\:\approx\:\) 2 K. The presence of these precipitates has been suggested to be one of the major concerns for the local breakdown of surface superconductivity. As a result, degradation in Q0 at higher accelerating fields, called high field Q slope (HFQS), can be induced10,20.

As Nb hydride precipitates form only at cryogenic temperature, direct in-situ observation is challenging. Analytical electron microscopy and atom probe tomography21,22,23 confirmed localized segregation of the nano-sized Nb-H phases on various SRF Nb cavities. Scanning Tunneling Microscopy24,25 first showed hydride formation and growth behavior in oxide-free Nb surfaces along with density function theory calculation26,27. Surface-sensitive spectroscopy28,29 addressed hydrogen-rich layers on the top layer after chemical treatments, which is responsible for hydride formation. However, these techniques primarily provide local information from selected areas, and practical surface conditions modified by cavity treatments have not been reflected. Alternative modalities are needed to understand the three dimensional morphology and distribution of the precipitates. The initial in-situ study30,31 using laser confocal microscopy directly observed the formation of Nb hydride precipitations that left behind permanent physical surface deformation shown as µm-size seashell scars. Especially the study found that nanometer scale hydride-like surface features appeared on the surface and completely vanished when heated back to room temperature. Its size is comparable to the coherence length of niobium, ξNb(0) ≈ 40 nm. However, this finding was limited only to the Nb sample that has been exposed to hydrogen rich environment.

In SRF grade Nb, hydrogen concentrations are known to be highest in ~ 100 nm region closest to the surface32,33, underneath the topmost dense Nb2O5 layer34. Understanding the size, morphology, and distribution of the nanometer scale Nb hydride-like features requires additional techniques at cryogenic temperature, which will be helpful for evaluating the extent of localized degradation in the cavity performance. Also, importantly, the structural phase of the surface precipitates is unexplored. Thus, in this study, by directly analyzing cut-out coupons from the RF-characterized Nb SRF cavities, we explore the morphology of the nano-size Nb hydride precipitates while identifying their structural phase through a combination of cryogenic atomic force microscopy (AFM) and high energy grazing incidence x-ray diffraction (GI-XRD) technique.

The overview of a single cell SRF Nb cavity and the cryogenic-AFM system used in this study is represented in Fig. 1. The machined parts of the outside cavity depict the areas where local temperature rise occurred during RF test35. Typically, the area is called a hot spot. A 10 mm diameter cut-out coupon is placed on the top copper plate of the X-Y-Z piezo-driven stage in the AFM. The sample was cooled at a rate of 20 K/min, and contact-mode AFM scans were performed at every 50 K step from 300 K to 10 K and as the sample was warmed back up to 300 K. Due to thermal fluctuation of an AFM cantilever piezo, an individual scan was executed after allowing 3 h for the system to stabilize at each target temperature. This long stabilization likely provided a preferential environment where hydrogen atoms incorporated within the niobium lattice can form Nb-H phase with extended diffusion36.

(a) A photo of a 1.3 GHz single-cell Superconducting Radio Frequency (SRF) niobium cavity with an enlarged image of a cut-out coupon (10 mm in diameter and 2.5 mm in thickness). The ruler is 18 inches long. (b) Cryogenic AFM (atomic force microscope) system with a cut-out coupon on the sample stage of the piezo-driven X–Y–Z stage block.

Table 1 describes the characteristics of the cavities and their cut-out coupons, investigated in this study. Q0 (quality factor) vs. Eacc (accelerating field) of the cavities are represented in the supplementary Fig. 1. Hydrogen-related performance degradations are seen in the cavities. The green arrow indicates a marked Q drop due to hydrogen-disease37, and the black arrow points out high field Q slope (HFQS) associated with nano-sized Nb hydride precipitates18. Anti-Q slope behavior is seen on the N (nitrogen)-doped cavity12. 120 °C mild baking resolves HFQS that appears on the 800 °C HT’ed + EP cavity10. The N-infused cavity exhibits the highest performance, close to 45 MV/m14.

We first scanned a “hot spot” area of a non-degassed Nb cavity with cryo-AFM. This cavity surface was reset with a 40 μm EP removal without extra heat treatment, meaning that no thermal treatment effect exists in the surface inner layer of the cavity, ~ 40 μm in depth. We expected that this treatment would leave the near surface (~ 100 nm) impurity structure of the cavity rich in hydrogen. Its RF performance presents slightly earlier HFQS, compared with an 800 °C HT’ed EP cavity, as shown in Fig. 2. The hot spot represents a local area where a breakdown of superconductivity occurs due to local temperature rising38. AFM laser intensity images are shown as a function of temperature in Fig. 3. At 300 K and 250 K, no distinguishable features are seen besides scratches and/or foreign particles arising from sample handling. However, distinct surface features, such as small bumps appeared during cooling to 150 K, as indicated by the red arrows. These surface features remained down to 10 K, which are ~ 800 nm in diameter and ~ 30–50 nm in height, as described in Fig. 3b and c. As the sample is warmed to room temperature, however, the surface features disappear, as shown in Fig. 3d. Based on the similar phenomenon observed in the previous in-situ study30,31, we suspect that these surface features arise from the formation of Nb hydride segregation. In this case, residual hydrogen in the hot spot region from the non-degassed cavity could be sufficient to drive Nb hydride precipitates on the nanometric scale but insufficient to form larger precipitates on the order of a few hundreds of micrometers (µm) size that remains surface scars.

Quality factor (Q0) versus accelerating field (Eacc) of the 1.3 GHz superconducting radio frequency (SRF) niobium (Nb) cavities applied in this study: black square) the non-degassed EP cavity; only 10 μm EP was applied (no thermal treatment) after the standard 40 μm EP cavity reset process, and red circle) the 800 °C HT’ed EP cavity; 800 °C/3 h heat treatment and 20 μm EP applied.

Cryogenic AFM images on the hot spot cut-out coupon from the non-degassed EP cavity. (a) A series of the AFM intensity images acquired during cooling from 300 K to 10 K, surface scratches marked with the yellow dotted lines are to give guidance for locating the scan area. The orange square area at 150 K is enlarged at 10 K. (b) Height profile across two bumps (A & B) marked on the intensity AFM image at 10 K. (c) 3D topography of the two bumps, (d) The intensity images after warmed back to 300 K and 50 K, respectively.

Figure 4 compares the surface features observed between 300 K and 100 K in the cut-out coupons from the SRF cavities (described in Table 1 and in the supplementary Fig. 1). From the hot spot, several µm size features are observable at 200 K, and, as described before, nm-scale features started appearing at 150 K between the larger features. The scan image at 200 K slightly shifted to the right and down compared to the 300 K scan due to piezo stage movements by cooling. 800 °C/3 hrs heat treatment is empirically known to annihilate interstitial hydrogens in the Nb surface39. We found that the post 800 °C heat treated sample (No 2. 800 °C HT’ed) shows a markedly suppressed formation of the surface features. But many smaller sized features still appeared at 100 K. The 120 °C baked and N-doped cut out exhibit surface features even at 200 K. However, the N-doped cut-out shows a large density of the small size features at 100 K. Interestingly, residual particles on the 120 °C baked cut-out, marked with the green arrows, appear to increase in size from 300 K to 200 K upon cooling. It is assumed that structural defects localized near the surface may facilitate the formation of Nb hydride precipitates in this case. In contrast, the N-infused cut-out shows no distinct surface features down to 100 K. In most cases, the surface protrusions that appeared up to at 100 K remained stable while cooling down to 10 K. The small features that appeared on the 120 °C baked and N-doped cut-outs are too small to ascertain with our AFM resolution.

A series of AFM intensity images of the cut-out coupons from the differently treated SRF Nb cavities from 300 K to 100 K. The cut-out coupons applied in this study are from the non-degassed, 120 °C baked, N-doped, and N-infused SRF Nb cavities, as described in Table 1. The hot spot is the same as Fig. 3, and the yellow dotted lines are marked to guide locating the scan area. The 800 °C HT’ed sample is one of the hot spot areas from the non-degassed EP cavity, but it was post-annealed at 800 °C for 3 h and 20 μm surface chemical removal by BCP. Orange squares and circles indicate distinct surface features that appeared during cryogenic cooling. Grain boundaries (GBs) are marked with the red arrows.

After multiple scans on each cut-out upon a complete cooling and warming cycle, the first appearances of the surface features (bumps) at each scan temperature (50 K interval) were manually counted by comparing the surface topology at 300 K. If the features remained at the following temperatures after initially emerging and their dimension was less than 3–5 nm, they were not counted. Less than 5 nm is out of the AFM resolution. These simple statistics are plotted in Fig. 5a. The plot indicates the averaged appearance frequencies (density) of the surface features within a 37 by 37 μm2 area. The heights of the observed (counted) surface features are also plotted in Fig. 5. The averaged total number of the observed features per scan is presented in the last column for the frequency in Fig. 5a. The hot spot and post 800 °C HT’ed cut-out show similar high densities of the surface protrusions, compared to the others. In Fig. 5b, the hot spot exhibits large size features (> 100 nm) at 200 K, but on the 800 °C HT’ed cut-out, the features with height less than 10 nm are predominant. On the 120 °C baked, N-doped, and N-infused cut-outs, these surface features are observed at a low density, and their heights are mainly between 10 nm < < h << 50 nm. However, the surface features appeared at 250 K for the 120 °C baked cut-out, which is a higher temperature compared to that observed in other samples. The N-infused coupon shows the lowest density of the surface features.

(a) The averaged appearance frequency along with cooling temperature and (b) the averaged frequency with the heights of the surface protrusions on the cut-outs of the cavities over a unit area of 37 by 37 μm2 during cryogenic cooling from 300 K to 10 K. For each sample, an average of 2–5 AFM scan images were taken.

To verify that the surface features observed by cryo-AFM scan arise from hydride precipitations, the surfaces of the cut-out coupons from the non-degassed EP (No. 1) and 800 °C/3 hrs HT’ed EP (No. 7) cavities were probed in the temperature ranging from room temperature to 30 K by synchrotron grazing incidence X-ray diffraction (GI-XRD). However, a heavy hydrogen-loaded (H-enriched) sample was initially investigated as a reference based on the findings from the previous in-situ study [30, 31], see Supplementary Fig. 2. It aimed to identify saturated Nb-H phases and precipitation morphology as a function of temperature. Upon cryogenic cooling, the H-enriched sample showed about 0.1% reduction of the lattice parameter (Å). At 175 K, new prominent peaks appeared on the low-angle side of the (110) bcc-Nb peak. The Rietveld refinement reveals the appearance of the Nb-hydrides phase associated with orthorhombic (orth)-NbH0.89 and cubic (bcc)-NbH0.8. These hydride phases completely disappeared when warmed up to 300 K.

Figure 6 shows a series of the diffraction patterns with the lattice parameter (Å) from 300 K to 30 K for (a) the hot spot and (b) the 800 °C HT’ed cut-outs, respectively. At 150 K, the hot spot shows new reflections related to the Nb hydride phase, but their intensities are weak. The peaks are determined to be the cubic phase of Nb-Hydrides (NbH0.8) located at 2θ = 18.46 and 32.7 with d (interplanar distance) = 2.42 Å and 1.37 Å, respectively. In contrast, on the 800 °C HT’ed EP cut-out at 200 K, a sharp peak is observed at 2θ = 28.89 (d = 1.55 Å). The identified phase corresponds to the face centered orthorhombic (fco)-NbH0.75 hydride, which remains stable upon cooling to 30 K. After warming up to 300 K, the reflection related to the Nb-hydrides completely disappeared at room temperature, which agrees with the observation of cryo-AFM. In addition, the dynamics of the Nb-hydrides segregation are indirectly observed through the variation in the lattice parameter of Nb as a function of temperature. As illustrated in Fig. 6, the lattice parameter (a) vs. temperature (T) curves show a remarkable hysteresis. This hysteresis suggests an irreversible process where the atomic arrangement of H deviates from its original positions in the lattice. Such behavior agrees with previous studies30,35, where differences in the Nb-hydrides segregation (including bump size, distribution, and population) are consistent with altered cavities performance (Q0 vs. Eacc) after consecutive cooling cycles.

High-resolution XRD (X-ray diffraction) reflections with lattice parameters (Å) at 1° of grazing incidence from 300 K to 30 K. The 2θ X-ray patterns were recorded from (a) the hot spot of the non-degassed cavity (No. 1) and (b) a cut out of the 800 °C heat treated EP cavity (No. 7). Lattice parameters (Å) variation upon cooling and warming. The color codes of the measurement temperature are presented with a scale bar.

SIMS was performed to locate the sources of hydrogens leading to the segregation of the nanometric scale Nb hydrides on the surface. Figure 7 shows the depth profiles of the hydrogen, oxygen, oxyhydrogen and Nb2O5 near the surfaces of the hot spot (No. 1) of the non-degassed and the 800 °C HT’ed + EP (No. 7) cavities at 300 K. Hydrogen content increases to a maximum value immediately underneath the Nb2O5 layer in both samples, but the non-degassed hot spot cut-out shows a slight thick Nb2O5 layer. The non-degassed cut-out exhibits 25% and 45% higher counts in H- and NbH-, respectively, compared to the EP’ed sample. However, NbH- is a more decisive parameter for direct comparison because its count is already normalized with Nb signal. While in both samples, the oxygen counts peak at the surface and decay in direction of the Nb matrix, the normalized oxygen counts (Fig. 7b) decay more gradually in the EP’ed sample. Oxyhydrogen (OH-) count shows a similar trend with the peaks following with vanishing of Nb2O5 counts. Since OH- can be referred to affinity of H trapping by O, it is inferred that the amounts of residual hydrogen could be higher on the surface of the hot spot cut-out of the non-degassed cavity than the case of the EP’ed cavity cut-out.

Depth profiles of (a) hydrogen, niobium hydrogen, and Nb2O5 and (b) normalized oxygen, oxyhydrogen, and Nb2O5 on the hot spot cut-out from the non-degassed (solid) and the cut-out of the 800 °C heat treated EP (dotted) cavities at 300 K by secondary ion mass spectroscopy (SIMS). In (b), O-, OH-, Nb2O5, are normalized with Nb-.

Hydrogen atoms behave as a source of hydride precipitate in the gas-metal (H-M) environment40,41. This is a major concern for superconducting radio frequency (SRF) technology using niobium18,26,42 because the Nb hydrides can lead to thermal breakdown of superconductivity as a normal conductor38,43. In the case of the H-enriched condition like the supplementary Fig. 2, micrometer-size of the Nb-H phase segregations are favorable, thereby, resulting in extended lattice distortion such as a scar on the surface along with destruction of the topmost dense Nb2O5 layer (in the supplementary Fig. 3). However, this destructed surface feature has not been observed in the practical SRF Nb cavities, even for the cases of “H-disease” or “HFQS” where RF performance markedly drops, as described in the supplementary Fig. 1. Rather, nanometer scale Nb-H precipitates are likely associated with such degradations18. If Nb-H phase with comparable size to ξnb(0) \(\:\approx\:\:\)40 nm precipitates, the proximity effect could locally induce superconductivity suppression, which is supposed to degrade cavity performance.

Our first cryogenic-AFM scan presents a variation of surface topographical features upon cooling, associated with the segregation of nanoscale Nb hydride precipitates on the hot spot of the non-degassed cavity beyond ~ 1 μm size features. Nano-sized surface features appeared on the surface at 150 K, remained down to 10 K, and then completely vanished when warmed up to room temperature. The GI-XRD study suggests that bcc-NbH0.8 phase mainly segregates across this same temperature range. Although the overall weak peak intensities of the Nb hydrides can be interpreted as a small quant, compared to the Nb matrix, if considering the depth resolution of GI-XRD, we supposed that those weak intensities could be contributed to surface-restricted NbH precipitation. The peaks of the Nb hydride on 2θ XRD patterns disappeared when warmed back to room temperature, indicating that the nanometer-sized surface features observed by cryo-AFM stem from Nb-hydride precipitation.

The nano-size Nb hydride is suggested to be responsible for HFQS. 800 °C heat treatment is known to suppress Nb-hydrides formation. However, in Fig. 2, the HFQS of the 800 °C/3hrs HT + 20 μm EP’ed cavity shows a slightly better RF performance compared to the case of the non-degassed cavity that has no 800 °C thermal effect on the surface. HFQS still remains even after 800 °C/3hrs HT. We expected that the H-related surface condition of the hot spot with post 800 °C HT would be equivalent to that of the 800 °C/3hrs HT + EP’ed cavity. Figures 4 and 5 presents a high density of small-size Nb hydride precipitates. GI-XRD study also shows a marked increase in the peak intensity of fco-Nb-H phase on the 800 °C/3hrs HT’ed + EP cavity from 200 K to 50 K. We also observed a a-Nb(H) peak on the 800 °C/3hrs HT’ed + EP at 100 K, but its intensity was so weak that we could not completely extract this reflection from the background. However, this finding suggests another Nb hydride phase precipitate on the 800 °C/3hrs HT’ed Nb upon cooling. In contrast, the cut-outs from the 120 °C baked, N-doped, and N-infused cavities show fewer and smaller nano-size Nb hydride-like surface features, but those are slightly bigger than the nano-hydride features on the 800 °C/3hrs HT’ed. The 120 °C baked, N-doped, and N-infused cavities do not have HFQS. Thermal treatment at mild temperature (120–400 °C) facilitates oxygen (O) diffusion toward the Nb matrix44. Treatment with N gas promotes interstitial invasion of N atoms into the cavity surface45. Importantly, it was first observed that N-doping leads to the suppression of nano-Nb hydride growth52. Trapping of hydrogen atom is favorable with O (oxygen) and (N) nitrogen26,27,46. Therefore, the lowest density of surface protrusion features is not surprising on the 120 °C baked, N-doped, and N-infused cavity cut-outs. It can be, therefore, concluded that the total amount (or density) of nano-size Nb-hydride precipitations can be considered the key parameter that can determine the extent of HFQS in the SRF Nb cavity.

Unlike the micrometer scale Nb-hydride that leads to extensive irreversible surface deformation (in the supplementary Fig. 3), the nano-sized Nb-hydride follows a different formation mechanism. At ambient temperature, hydrogen atoms are mobile in the Nb lattice. Redundant hydrogen atoms segregate at interstitial (tetragonal) sites or defects such as voids47, grain boundaries48,49, or dislocations50,51. When the temperature is lowered below 300 K, hydrogen atoms diffuse to neighboring defects, and Nb hydrides start to precipitate locally. The finding that nanoscale Nb hydrides completely disappear when heated to 300 K suggests that the segregated Nb-H phases dissolve thoroughly during this process, and H atoms diffuse back to neighboring tetragonal sites. The two-state model32,33 describes H concentration as being highest near the surface (< 100 nm) rather than the bulk, and hydrogen atoms cannot diffuse through the dense Nb2O5 layer. Thus, if Nb-H precipitates form at the interface between the Nb2O5 layer and Nb matrix and those hydrides protrude through the oxide layer, we can observe the nanoscale features with AFM scan. Figure 8 presents the schematic diagram of the heterogenous segregations of Nb hydride (Nb1-xHx) precipitates, proposed based on this scenario. Secondary ion mass spectroscopy (SIMS) showed a few nm thick Nb2O5 oxide layer present on the top layer of the H-enriched sample that exhibits both irreversible (micro-sized) and reversible (nano-sized) Nb-H precipitates. Observation of the nanoscale Nb hydrides could not be made by extended diffusion of hydrogen atoms via the thick oxide layer. Thus, protrusions could be caused by Nb hydride segregation at the interface. SIMS depth profiles show that excess hydrogens are mainly located just underneath the Nb2O5 layer. Even though the hydrogen signals in Fig. 7 are constant down to a depth of 25 nm, the signals gradually decreased to 150–200 nm and reached a plateau for a 500 nm deep analysis. This trend follows the scenario of the two-layer system proposed by the kinetic studies32,33. Here, the topmost layer signals are only presented to describe the surface chemistry. The dynamics of Nb-hydrides segregation were indirectly confirmed by the hysteresis observed in the lattice parameter (a) vs. temperature (T) curves. This behavior indicates an irreversible process, where the positioning of H atoms in the Nb lattice is altered during the phase ordering. However, in the case of nanoscale hydride precipitates, the cooling cycle produced reversible lattice structure changes on the SRF Nb surfaces.

Schematic diagram of the heterogeneous segregation of nanometer size niobium hydride (Nb1 − xHx) precipitates at the interface between the Nb2O5 surface oxide layer and the Nb matrix. (a) at ambient temperature, hydrogen atoms are freely distributed, and they are favorable to create vacancy-hydrogen complexes and nitrogen-hydrogen coupling, and (b) when cooling below 300 K, hydrogen diffuses into the defects and forms hydrides.

In conclusion, by systematically combining cryogenic atomic force microscopy (AFM) with high energy grazing incidence X-ray diffraction (GI-XRD), we were able to make the direct observation on nanometric surface features associated with hydride precipitates on the SRF Nb cavities (3D resonators) upon cryogenic cooling. Such nanometer sized Nb hydrides are supposed to be responsible for degradation of RF performance in a high accelerating field regime (HFQS). It results from a local breakdown of superconductivity by proximity effect. Importantly, our study suggests that the amount (density) of nano-Nb hydride precipitation is more important parameter for the extent of HFQS. Secondary ion mass spectroscopy (SIMS) suggests that nanoscale surface pop-ups upon cryogenic cooling are developed by hydrides segregated at the interface between the Nb2O5 oxide layer and the Nb matrix. From the fact that the surface bumps completely disappeared without any detectable topographical features when warming back to room temperature, it is feasible that the precipitated Nb hydrides completely dissolved at 300 K, and H atoms diffused back to interstitial sites or defects. This study shows the formation morphology of nanometric Nb hydrides depending on the cavity fabrication processes. It suggests that the extent of proximity degradation of RF superconductivity can be mitigated with surface or thermal treatments to minimize hydrogen intakes.

To develop a high-performance superconducting radio frequency (SRF) Nb cavity (3D resonator), a series of single cell cavities are fabricated and tested with various surface and thermal treatments. To understand the quench mechanism of the cavities, mapping with an array of thermometers (T-mapping) on the outer surface of a cavity35 is conducted to precisely locate the areas showing local temperature rising during RF test at 1.8–2 K. Identification of the local area is beneficial for understanding of the cavity RF performance degradation, represented by Q0 (quality factor) versus Eacc (accelerating field), as shown in the supplementary Fig. 1. The thermometers are carbon resistors whose resistance exponentially increases with decreasing temperature on a resolution of 1 mK. The application of fast response resistors and a large number of thermometers that can span the entire cavity’s outer surface is required for precision diagnosis.

As described in Table 1, 10 mm diameter disk shape coupons were extracted from the tested superconducting radio frequency (SRF) cavities. Extraction was carried out by slow drill-machining without any water or oil based lubricant in order to minimize hydrogen intrusion and structural deformation by cutting heat load. The cut out coupon is lightly curved (concaved) due to the cavity’s intrinsic circular pillbox shape. The term “hot spot” represents a local area of the cavity surface where breakdown of superconductivity occurs due to local thermal rising during RF operation35. The non-degassed cavity refers to the cavity that is not thermally treated after a standard ~ 40 μm EP reset process. There is no heat treatment effect on the top ~ 40 μm cavity surface, even though this cavity has a history of engineering processes for high RF performance close to 40 MV/m. This means that the reset could introduce excess hydrogens to the cavity surface, which could later be saturated as a hydride phase. No. 2 (800 °C HT’ed sample) is one of the hot spots from the non-degassed cavity. After extraction, the sample was thermally treated at 800 °C for 3 h, and processed by ~ 20 min BCP (buffered chemical polishing) to remove surface damaged layers (~ 20 μm). It is supposed that the surface condition of No. 2 could be analogous to the surface chemical condition of the 800 °C/3hrs HT + 20 μm EP’ed cavity (No. 7). No. 6, H (hydrogen)-enriched sample, was prepared from a hot spot of a post 975 °C HT + N-doped (at 800 °C) Nb cavity with applying conventional cross-section mechanical polishing, finished with vibrometry polishing in H2O based colloidal silica solution. This H-enriched sample preparation is identical to the recipe applied in the initial in-situ cryogenic study30,31. Nb hydride precipitation on this H-enriched sample left behind permanent physical deformation like scars, as shown in the SEM image and EBSD data in the supplementary Fig. 3.

The cryogenic AFM applied is a commercial instrument of attocube system AG, which is incorporated with QD (quantum design) 9 Tesla (T) physical property measurement system (PPMS) that enables to cool a sample down to 1.8 K. To maximize topographical resolution, contact mode AFM was implemented with minimal surface loading force using a Nanosensors AFM tip of 0.022–0.77 N/m stiffness. Since cooling the 9T PPMS sample chamber is achieved by a flowing rate of cold helium gas, the AFM is operated under helium gas environment. 20 K/min cooling rate is applied, but ~ 3 hour staying at the target temperature was required to avoid noise from thermal fluctuation of a cantilever piezo. The total scanning area gradually shrinks with the decreasing of the system temperature because the movement of the piezo blocks reduces with temperature.

HR XRD was performed at the sector 33-BM-C beamline of the Advanced Photon Source (APS) of ANL (Argon National Laboratory) facility, which delivers high flux and monochromatic beam using a double Si (111) monochromator. The high resolution is achieved through the Huber 6-circle diffractometer and Pilatus 100 K detector, with 900 μm x 500 μm as a focused point. All the samples were cooled using a liquid helium cold finger (300 K to 30 K) under ultra-high vacuum. The selected energy used for obtaining the patterns was 16 eV (λ = 0.775 Å) and incident angle of 1º. The depth penetration by the X-rays (E = 16 keV) was calculated from the Parratt equation52, resulting in ~ 0.75 μm. Peaks of 2θ patterns were indexed by applying DicVol integrated in FullProof software53 with ICSD database to corroborate the phase identification54. The Nb-hydride NbH0.8, ICSD number 638,368, is found as a bcc (body centered cubic)-structure with a lattice parameter (å) of 0.342 nm as a \(\:Im\stackrel{-}{3}m\) space group. Orthorhombic (orth)-NbH0.89 is the face centered orthorhombic structure, ICSD number 150,604 and lattice parameters of a = 0.484 nm, b = 0.49 nm and c = 0.345 nm with \(\:Pnnn\) as space group.

Surface sensitive analysis was carried out by ToF-SIMS (time of flight secondary ion mass spectroscopy) at ambient temperature. While pulsed Cs (Cesium) ions continuously remove the topmost surface at 500 eV energy, the time-of-flight mass analyzer detects molecular ions from the surface layer with mass-to-charge ratios ranging from 1 to 10,000 in a single spectrum. Based on empirical experiments with standardized SRF-grade niobium samples, we estimated the depth profile from the Cs ion sputtering (milling) time. Negative ion mass spectra were used for most surface species, and all peaks were calibrated with those from the standard SRF Nb sample.

The datasets generated and/or analyzed during the current study are not publicly available based on guidance from the U.S. Department of Energy – High Energy Office. Still, they may be available from the corresponding author upon reasonable request. We can only share the data via the published transcripts, technical reports, or presentation files approved by the Fermi National Accelerator Publication team. In some cases, the datasets used and/or analyzed during the current study are available from the corresponding author upon reasonable request. Otherwise, all data generated or analyzed during this study are included in this published article: arXiv:2305.01022 and https://accelconf.web.cern.ch/srf2019/talks/tufub1_talk.pdf.

Saito, K. Critical Field Limitation of The Niobium Superconducting RF cavity. Proceeding 10th Workshop RF Supercond. Tsukuba Jpn. 5 (2001).

Gurevich, A. Superconducting radio-frequency fundamentals for particle accelerators. Rev. Accel Sci. Technol. 05, 119–146 (2012).

Article Google Scholar

Rose-Innes, A. C. & Rhoderick, E. H. Introduction to Superconductivity (Pergamon, 1978).

Gurevich, A. Theory of RF superconductivity for resonant cavities. Supercond Sci. Technol. 30, 034004 (2017).

Article ADS Google Scholar

Hasan Padamsee, J., Knobloch, T. & Hays RF Superconductivity for Accelerators, 2nd Edition | Wiley. (2008).

Voss, R. F., Laibowitz, R. B. & Broers, A. N. Niobium Nanobridge Dc SQUID. Appl. Phys. Lett. 37, 656–658 (1980).

Article ADS CAS Google Scholar

Qing, Z. et al. High transfer coefficient niobium nano-SQUID integrated with a nanogap modulation flux line. Meas. Sci. Technol. 32, 025001 (2021).

Article ADS CAS Google Scholar

Zmuidzinas, J. Superconducting microresonators: physics and applications. Annu. Rev. Condens. Matter Phys. 3, 169–214 (2012).

Article CAS Google Scholar

Murthy, A. A. et al. TOF-SIMS analysis of decoherence sources in superconducting qubits. Appl. Phys. Lett. 120, 044002 (2022).

Article ADS CAS Google Scholar

Ciovati, G., Myneni, G., Stevie, F., Maheshwari, P. & Griffis, D. High field Q slope and the baking effect: review of recent experimental results and new data on Nb heat treatments. Phys. Rev. Spec. Top. - Accel Beams. 13, 022002 (2010).

Article ADS Google Scholar

Romanenko, A., Grassellino, A., Barkov, F. & Ozelis, J. P. Effect of mild baking on superconducting niobium cavities investigated by sequential nanoremoval. Phys. Rev. Spec. Top. - Accel Beams. 16, 012001 (2013).

Article ADS Google Scholar

Grassellino, A. et al. Nitrogen and Argon doping of niobium for superconducting radio frequency cavities: a pathway to highly efficient accelerating structures. Supercond Sci. Technol. 26, 102001 (2013).

Article ADS Google Scholar

Dhakal, P. Nitrogen doping and infusion in SRF cavities: a review. Phys. Open. 5, 100034 (2020).

Article CAS Google Scholar

Grassellino, A. et al. Unprecedented quality factors at accelerating gradients up to 45 MVm – 1 in Niobium superconducting resonators via low temperature nitrogen infusion. Supercond Sci. Technol. 30, 094004 (2017).

Article ADS Google Scholar

Gurevich, A. Challenges and Opportunities of SRF Theory for Next Generation Particle Accelerators. arXiv:2203.08315. https://doi.org/10.48550/arXiv.2203.08315

Kubo, T. Superfluid flow in disordered superconductors with Dynes pair-breaking scattering: depairing current, kinetic inductance, and superheating field. Phys. Rev. Res. 2, 033203 (2020).

Article CAS Google Scholar

Isagawa, S. Hydrogen absorption and its effect on low-temperature electric properties of niobium. J. Appl. Phys. 51, 4460–4470 (1980).

Article ADS CAS Google Scholar

Romanenko, A., Barkov, F., Cooley, L. D. & Grassellino, A. Proximity breakdown of hydrides in superconducting niobium cavities. Supercond Sci. Technol. 26, 035003 (2013).

Article ADS Google Scholar

Romanenko, A. & Padamsee, H. The role of near-surface dislocations in the high magnetic field performance of superconducting niobium cavities. Supercond Sci. Technol. 23, 045008 (2010).

Article ADS Google Scholar

Knobloch, J., Geng, R. L., Liepe, M. & Padamsee, H. Proceedings of the 1999 Workshop on RF Superconductivity, La Fonda Hotel, Santa Fe, New Mexico, USA. tua004, 15 (1994).

Trenikhina, Y., Romanenko, A., Kwon, J., Zuo, J. M. & Zasadzinski, J. F. Nanostructural features degrading the performance of superconducting radio frequency niobium cavities revealed by transmission electron microscopy and electron energy loss spectroscopy. J. Appl. Phys. 117, 154507 (2015).

Article ADS Google Scholar

Tao, R., Romanenko, A., Cooley, L. D. & Klie, R. F. Low temperature study of structural phase transitions in niobium hydrides. J. Appl. Phys. 114, 044306 (2013).

Article ADS Google Scholar

Kim, Y. J., Tao, R., Klie, R. F. & Seidman, D. N. Direct Atomic-Scale Imaging of Hydrogen and oxygen interstitials in pure Niobium using atom-probe tomography and Aberration-Corrected Scanning Transmission Electron Microscopy. ACS Nano. 7, 732–739 (2013).

Article CAS PubMed Google Scholar

Veit, R. D., Farber, R. G., Sitaraman, N. S., Arias, T. A. & Sibener, S. J. Suppression of nano-hydride growth on Nb(100) due to nitrogen doping. J. Chem. Phys. 152, 214703 (2020).

Article ADS CAS PubMed Google Scholar

An, B. et al. Investigation of the Interaction of Hydrogen with a nb(100) surface by scanning tunneling Microscopy. Jpn J. Appl. Phys. 43, 4502 (2004).

Article ADS CAS Google Scholar

Ford, D. C., Cooley, L. D. & Seidman, D. N. First-principles calculations of niobium hydride formation in superconducting radio-frequency cavities. Supercond Sci. Technol. 26, 095002 (2013).

Article ADS CAS Google Scholar

Garg, P. et al. Revealing the role of nitrogen on hydride nucleation and stability in pure niobium using first-principles calculations. Supercond Sci. Technol. 31, 115007 (2018).

Article ADS Google Scholar

Dhakal, P. et al. Surface characterization of nitrogen-doped high purity niobium coupons compared with superconducting rf cavity performance. Phys. Rev. Accel Beams. 22, 122002 (2019).

Article ADS CAS Google Scholar

Singh, N., Deo, M. N., Nand, M., Jha, S. N. & Roy, S. B. Raman and Photoelectron spectroscopic investigation of high-purity niobium materials: Oxides, hydrides, and hydrocarbons. J. Appl. Phys. 120, 114902 (2016).

Article ADS Google Scholar

Barkov, F., Romanenko, A. & Grassellino, A. Direct observation of hydrides formation in cavity-grade niobium. Phys. Rev. Spec. Top. - Accel Beams. 15, 122001 (2012).

Article ADS Google Scholar

Barkov, F., Romanenko, A., Trenikhina, Y. & Grassellino, A. Precipitation of hydrides in high purity niobium after different treatments. J. Appl. Phys. 114, 164904 (2013).

Article ADS Google Scholar

Kim, S. W. & Sohn, K. S. Hydrogen uptake kinetics on niobium surfaces. Phys. Rev. B. 40, 1003–1007 (1989).

Article ADS CAS Google Scholar

Davenport, J. W., Dienes, G. J. & Johnson, R. A. Surface effects on the kinetics of hydrogen absorption by metals. Phys. Rev. B. 25, 2165–2174 (1982).

Article ADS CAS Google Scholar

Wu, A. T. Investigation of oxide layer structure on niobium surface using a secondary ion mass spectrometry. Phys. C Supercond. 441, 79–82 (2006).

Article ADS CAS Google Scholar

Knobloch, J., Muller, H. & Padamsee, H. Design of a high speed, high resolution thermometry system for 1.5 GHz superconducting radio frequency cavities. Rev. Sci. Instrum. 65, 3521–3527 (1994).

Article ADS CAS Google Scholar

(eds Jena, P. & Stterhwaite, C. B.) Electronic Structure and Properties of Hydrogen in Metals, edited by Springer Line Mageri, H., Rush, J. J., Row, J. M., Richter, D. & Wipf, H. Influence of substitutional and interstitial impurities on local hydrogen mode in Nb, page 171. (1983).

Knobloch, J. The Q disease in Superconducting Niobium RF Cavities. in AIP Conference Proceedings vol. 671 133–150 (AIP, Newport News, Virginia (USA), (2003).

Gurevich, A. Multiscale mechanisms of SRF breakdown. Phys. C Supercond. 441, 38–43 (2006).

Article ADS CAS Google Scholar

Visentin, B., Barthe, M. F., Moineau, V. & Desgardin, P. Involvement of hydrogen-vacancy complexes in the baking effect of niobium cavities. Phys. Rev. Spec. Top. - Accel Beams. 13, 052002 (2010).

Article ADS Google Scholar

(Metal Hydrides, Springer, U. S. & Boston, M. A. (1981). https://doi.org/10.1007/978-1-4757-5814-6

Makenas, B. J. & Birnbaum, H. K. Phase changes in the niobium-hydrogen system I: accommodation effects during hydride precipitation. Acta Metall. 28, 979–988 (1980).

Article CAS Google Scholar

Lee, J. et al. Discovery of Nb hydride precipitates in superconducting qubits. Preprint at. https://doi.org/10.48550/arXiv.2108.10385 (2021).

Article Google Scholar

Bonin, B. & Roth, R. W. Q degradation of Niobium cavities due to Hydrogen contamination, Proceedings of the Fifth Workshop on RF Superconductivity, Desy, Hamburg, Germany, srf91d01,. (1991).

Bafia, D., Grassellino, A. & Romanenko, A. The Role of Oxygen Concentration in Enabling High Gradients in Niobium SRF Cavities. in The Role of Oxygen Concentration in Enabling High Gradients in Niobium SRF CavitiesUS DOE, doi: (2021). https://doi.org/10.2172/1825287

Trenikhina, Y., Grassellino, A., Romanenko, A. & Kwon, J. Characterization of nitrogen doping recipe for Nb superconducting radio frequency cavities, Proceedings of SRF 2015, Whistler, BC, Canada, MOPBO55.

Pfeiffer, G. & Wipf, H. The trapping of hydrogen in niobium by nitrogen interstitials. J. Phys. F Met. Phys. 6, 167–179 (1976).

Article ADS CAS Google Scholar

Hautojarvi, P., Huomo, H., Puska, M. & Vehanen, A. Vacancy recovery and vacancy-hydrogen interaction in niobium and tantalum studied by positrons. Phys. Rev. B. 32, 4326–4331 (1985).

Article ADS CAS Google Scholar

Sung, Z. H. et al. Development of low angle grain boundaries in lightly deformed superconducting niobium and their influence on hydride distribution and flux perturbation. J. Appl. Phys. 121, 193903 (2017).

Article ADS Google Scholar

Antoine, C. et al. Evidence of preferential diffusion of impurities along grain boundaries in very pure niobium used for radio frequency cavities. J. Appl. Phys. 81, 1677–1682 (1997).

Article ADS CAS Google Scholar

Schober, T. The niobium-hydrogen system – an electron microscope study. I. Room temperature results. Phys. Status Solidi A. 29, 395–406 (1975).

Article ADS CAS Google Scholar

Antoine, C. Z. H in Niobium: Origin And Method Of Detection. in AIP Conference Proceedings vol. 671 176–189 (AIP, Newport News, Virginia (USA), (2003).

Parratt, L. G. Surface studies of solids by total reflection of X-Rays. Phys. Rev. 95, 359–369 (1954).

Article ADS Google Scholar

Rodriguez-Carvajal, J. & FullProf, S. (2013). https://www.ill.eu/sites/fullprof/

ICSD Inorganic Crystal Structure Database, Leibniz Institute for Information Infrastructure. https://www.fiz-karlsruhe.de/en/produkte-und-dienstleistungen/inorganic-crystal-structure-database-icsd-news

Posen, S. et al. Efficient expulsion of magnetic flux in superconducting radiofrequency cavities for high Q 0 applications. J. Appl. Phys. 119, 213903 (2016).

Article Google Scholar

Download references

This manuscript has been authored by Fermi Research Alliance, LLC under Contract No. DE-AC02-07CH11359 with the U.S. Department of Energy, Office of Science, Office of High Energy Physics. This research used resources of the Advanced Photon Source; a U.S. Department of Energy (DOE) Office of Science User Facility operated for the DOE Office of Science by Argonne National Laboratory under Contract No. DE-AC02-06CH11357. Special thanks to Dr. Balachandran and Dr. Eremeev for thoughtful review.

Fermi National Accelerator Laboratory, Batavia, IL, 60510, USA

Zuhawn Sung, Arely Cano, Akshay Murthy, Daniel Bafia, Martina Martinello, Jaeyel Lee, Anna Grassellino & Alexander Romanenko

Advanced Photon Source, Argonne National Laboratory, Lemont, IL, 60439, USA

Evguenia Karapetrova

SLAC National Accelerator Laboratory, Menlo Park, CA, 94025, USA

Martina Martinello

Center for Research and Advanced Studies of the National Polytechnic Institute, Mexico City, Mexico

Arely Cano

You can also search for this author in PubMed Google Scholar

Z-H.S, A.C, A.R, A.G conceived the experiment(s). Z-H.S performed cryogenic atomic force microscopy, A.C, E.K performed high energy grazing incident X-ray diffraction, A.C, M.M, J-Y L designed HR GI XRD and analyzed the data. A.M, A.R, Z-H.S performed secondary ion mass spectroscopy and analyzed its data, Z-H.S performed sample preparation and scanning electron microscopy. A.R, D.B, M.M performed SRF cavity tests and cavity performance data analysis. All authors reviewed the manuscript.

Correspondence to Zuhawn Sung.

The authors declare no competing interests.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Below is the link to the electronic supplementary material.

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

Sung, Z., Cano, A., Murthy, A. et al. Direct observation of nanometer size hydride precipitations in superconducting niobium. Sci Rep 14, 26916 (2024). https://doi.org/10.1038/s41598-024-77905-6

Download citation

Received: 01 March 2024

Accepted: 28 October 2024

Published: 06 November 2024

DOI: https://doi.org/10.1038/s41598-024-77905-6

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative