Surface Analysis of Tritium Substrates

Cody Fagan

University of Rochester Laboratory for Laser Energetics

I. Introduction

Deuterium and tritium are isotopes of hydrogen that are used as fuel in nuclear fusion reactions. Tritium, a radioisotope of hydrogen, is a contamination issue at the laboratory for laser energetics. When a metal is exposed to a tritium gas atmosphere, large quantities of the tritium will diffuse into the metal lattice. This provedes a mechanism for fuel loss, as well as contamination of parts exposed to a tritium gas environment. All metals support a hydroxyl layer where adsorbed water binds, these water layers can potentially hold large quantities of tritium. Recent studies have shown that the total quantity that is retained in the metal, is dependent on the surface conditions of the metal, potentially due to the different hydroxyl site densities. Samples analyzed were aimed to reduce factors that influence surface water, such as surface roughness, and a noble gold layer.

II. Methods

Samples were received from a comercial supplier and polished according to an ASTM. A subset of the polished samples were then sent to a comercial plating company, and Los Alamos National Lab for gold plating. Once the samples received their final surface preparation technique, all samples were degreased in Acetone, followed by IPA, and finally water. The samples were then exposed to a room temperature deuterium-tritium gas for 24 hours. A single sample of each surface finish is not exposed to tritium and saved for surface analysis. Once loaded, the samples are stored in individual containers and sealed in dry helium to prevent cross-contamination. The total quantity of tritium contained in a sample is measured with temperature programmed thermal desorption. This robust technique heats the sample in a decontamination chamber under dry he and encourages thermal desorption of tritium from the surface and the bulk. The liberated tritium then is captured in one of two bubblers filled with liquid scintillation fluid. The resultant fluid is counted in a liquid scintillation counter, and the total quantity of captured tritium is calculated.

Figure 1: Schematic of the temperature programmed thermal desorption facility housed at LLE.

For surface analysis, standard samples were cut in thirds using a diamond impregnated saw blade, the samples were then washed in acetone to remove lubricant used in the cutting process. The surface techniques chosen for this work included scanning electron microscope (SEM) and atomic force microscopy (AFM) due to the complementary information obtained in each technique. All images were collected on a Zeiss Auriga CrossBeam SEM-FIB and the AFM is a NTMDT AFM microscope. In order to view the deposited gold layers, the SEM was used in coincidence with focused ion beam milling to cause ablation of the surface and then image with the primary electron beam. This method was used to create a cross-section of the sample where distinct layers are positively identified using x-ray analysis. Further methods included backscatter electron imaging, secondary electron imaging, and electron flight simulations to aid in the understanding of the x-ray results.

Figure 2: Cut coupons used for surface analysis.

III Results

Results of thermal desorption are shown in Figure 3.There is a clear increase in the total quantity of tritium for samples that were coated with gold, compared to an untreated sample.

All of the samples stem from the same supplier of unmodified samples. A subset of the unmodified samples are sent for polishing to a 3-4 Ra by a comercial polisher. Samples GP1 and GP2 are from the same comercial supplier, which contains a nickel strike. These samples only differ in plating method by when they were coated. The LANL sample is an attempt to gold coat stainless-steel with no nickel strike and were supplied by Los Alamos National Lab. To fully understand these results, surface analysis is necessary.

Figure 3: Thermal desorption results showing the total quantity of tritium removed in a sample with a particular surface finish.

The first sample analyzed was the untreated sample to establish a baseline for the samples. Figure 4 shows the AFM results of the untreated sample. There are large striations present and a large peak-valley number for this sample. This is indication of a large surface area on the sample to adsorb water and be exposed to tritium gas. However, we see that compared to the gold coating, these samples retain less tritium.

Figure 4: AFM surface analysis of unmodified surface sample.

The next sample in the sequence is a polished sample. The process is not well understood but the unmodified samples are mechanically polished at a comercial metal manufacture in Rochester NY. The samples are polished until a mirror finish is established, and an estimated 3-4 Ra. These samples have less total tritium than the unmodified samples. The AFM results shown in Figure 5 that the large striations that are from the unmodified sample are still present in the polished sample, but are smoother than the previous sample. The SEM results still show the large striations on the surface as well as the development of holes in the surface. X-ray analysis of the hole show that there is stainless-steel and Si present in the metals surface.

Figure 5: AFM surface analysis of polished sample (left), SEM image of polished samples surface, using the InLens detector [WD 5.7 mm, 20 kV @ 564x].

To understand the x-ray spectrum obtained in the SEM, an understanding of the interaction volume developed in the metal is necessary. The electron flight simulator is used to determine the depth of the volume, to identify the region of x-ray emission. The results of the flight simulator are illustrated in Figure 6, where a 700 nm gold layer is deposited on a bulk stainless steel structure. We can see that with a 20 kV accelerating voltage, the interaction volume extends 3.5 microns into the sample. This would indicate that a surface x-ray scan will reveal sub-gold structure, in our case either Ni or stainless-steel because of the depth of the interaction volume.

Figure 6: Electron flight simulation of a gold-steel bilayer, showing penetration depth of a 20 kV beam and forward scattered interaction volume.

The GP1 sample was processed and the thickness of the gold layer determined to be 1.3 microns thick, on a larger 3 micron Ni strike, followed by substrate stainless-steel. The FIB produced SEM imaged cross-section is shown in Figure 7 as well as the AFM results and a zoomed view of the porous gold layer using the backscatter detector. The SEM images indicate that there are voids in the top gold layer, where tritium may congregate in a so-called trapsite, shown in Figure 7.

Figure 7: AFM surface analysis of GP1 sample (left), SEM cross-section (middle), using the SE2 detector [20 kV with a 13.2 mm WD @17 KX], Porous gold using backscatter detector [20 kV, 13 mm, 26.42 KX].

The GP2 sample which is coated in an identical fashion to the GP2 sample shows the same Au-Ni-SS layered structure as expected. The gold layer was approximated to be 1 micron thick with a 3 micron thick Ni layer and verified with EDAX x-ray analysis (not shown) shown in Figure 8. The AFM results indicate that the large striations are still contained within the gold layer, and do not extend into the Ni layer. The SEM image also indicates that there are holes present in the surface of the gold, similarly to GP1.

Figure 8: AFM surface analysis of GP2 sample (left), SEM cross-section [20 kV, Inlens, 5.1 mm WD, 5.74 KX], .

The results of the surface analysis of the LANL sample are shown in Figure 9. The FIB milled cross-section is seen with the ablation shape stil in the image. The gold layer is very rocky compared to the GP1 and GP2 samples. The gold layer is incomplete for the LANL sample indicating a mixed surface where tritium gas can adsorb. The AFM results also agree with a peak-valley ratio larger than the estimated gold thickness. Since the diffusivity of gold is slower than that of steel, the tritons can penetrate the areas around steel deeper into the bulk than the areas covered with gold. The solubility of hydrogen in gold is much higher than that of steel, so we can stack near surface areas of gold with tritium. This may explain the elevated total tritium levels for the LANL sample.

Figure 9: AFM surface analysis of LANL sample (left), SEM cross-section (middle) [20 kV Inlens, 5.1 mm WD, 8.25 KX], Zoomed area on the cross-section, showing gold-steel interface [20 kV, 5.1 mm WD, 38.93 KX].

IV Conclusions

Several studies in the group have shown that the surface conditions strongly influence the total quantity of tritium in a stainless-steel sample. Until now, we were unable to probe the surface conditions directly to make comparisons between different surface modification techniques. From these results, we can establish that the surfaces are not indeed what we expected, especially the highly porous gold layer. The determination of gold as a tritium barrier is still incomplete, and based on the analysis, several new experiments have been planned for the coming months, with a collaboration from Sandia, Savannah River, and Los Alamos national labs. The skills gained in the course will prove to be invaluable going forward in this research.

V. Acknowledgments

Gratitude to Brian McIntyre for his training on the FIB/SEM and humor in the lab.

This material is based upon work supported by the Department of Energy National Nuclear Security Administration under Award Number DE-NA0001944, the University of Rochester, and the New York State Energy Research and Development Authority. This report was prepared as an account of work sponsored by an agency of the U.S. Government. Neither the U.S. Government nor any agency thereof, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not neces- sarily constitute or imply its endorsement, recommendation, or favoring by the U.S. Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the U.S. Government or any agency thereof.

Please enter any comments, criticisms, questions, etc. below.

Your name:

Email address:

References

W. T. SHMAYDA et al., “Dependence of Tritium Release from Stainless Steel on Temperature and Water Vapor,” Fusion Sci. Technol., 68, 4, 766 (2015); http://dx.doi.org/ 10.13182/FST14-913.

M. SHARPE, W. T. SHMAYDA, and W. U. SCHRÖDER, “Tritium Migration to the Surfaces of Type 316 Stainless- Steel; Aluminum 6061, and Oxygen-Free, High-Conductivity Copper,” Fusion Sci. Technol., 70, 1, 97 (2016); http://dx.doi. org/10.13182/FST15-198.

C. Fagan, M. Sharpe, W. T. Shmayda & W. U. Schröder, "The Impact of Acid Treatments and Electropolishing Stainless-Steel Surfaces on Tritium Inventories," Fusion Sci. Technol., pre-release, published online - 4/7/17