A protocol for the measurement of outgassing rates of hydrogen from ordinary steel vacuum chambers using the rate-of-pressure rise method is presented.
Steels are commonly used materials in the fabrication of vacuum systems because of their good mechanical, corrosion, and vacuum properties. A variety of steels meet the criterion of low outgassing required for high or ultrahigh vacuum applications. However, a given material can present different outgassing rates depending on its manufacturing process or the various pretreatment processes involved during the fabrication. Thus, the measurement of outgassing rates is highly desirable for a specific vacuum application. For this reason, the rate-of-pressure rise (RoR) method is often used to measure the outgassing of hydrogen after bakeout. In this article, a detailed description of the design and execution of the experimental protocol involved in the RoR method is provided. The RoR method uses a spinning rotor gauge to minimize errors that stem from outgassing or the pumping action of a vacuum gauge. The outgassing rates of two ordinary steels (stainless steel and mild steel) were measured. The measurements were made before and after the heat pretreatment of the steels. The heat pretreatment of steels was performed to reduce the outgassing. Extremely low rates of outgassing (on the order of 10−11 Pa m3 sec−1 m−2) can be routinely measured using relatively small samples.
Steels are routinely used in construction because of their good mechanical properties. Certain steels (ferrous steels, in particular) are preferred materials for applications involving vacuum. Depending on the type and grade, these steels have sufficiently low outgassing rates essential for high vacuum (HV, 10−7 < p < 10−5 Pa) or ultrahigh vacuum (UHV, 10−10 < p < 10−7 Pa) systems. Further, extensive research has been conducted toward the development of special pretreatment procedures that reduce outgassing1-3. The pretreatment measures are designed to minimize the pumping investment or to improve the vacuum from HV to UHV or from UHV to extreme-high vacuum (p < 10−10 Pa).
Although many practical methods have been proposed to reduce the outgassing rate of ferrous steels, recent methods are focused on reducing the time and temperature required to obtain a lower outgassing rate. Heat treatment at 350 °C-450 °C rather than vacuum firing at 800 °C-950 °C, is a good example of this approach.1,4,5 Furthermore, choosing the ideal material for a specific vacuum application is critical; for example, selecting a ferritic material with a very low outgassing rate for use in magnetic field shielding.6,7
During such investigations, precise measurement of the outgassing rate is a prerequisite for the screening of candidate materials or verifying the effectiveness of various pretreatment procedures.8,9 The most common experimental techniques used for the measurement of outgassing are the throughput and rate-of-pressure rise methods.10 Recently, various experiments have been conducted to measure the hydrogen outgassing rate based on the RoR method using spinning rotor gauge (SRG).1, 11-13 The RoR method using SRG is highly suitable for measuring very low hydrogen outgassing rates that often limit the lowest pressure achievable in a vacuum system made of steel. This is because the SRG has negligible pumping or outgassing action. Further, the SRG also has excellent accuracy and good linearity in high vacuum and ultra-high vacuum range.14
Given that the published literature on RoR experiments is limited, it is worthwhile to describe the experimental details to develop a deeper understanding of the method. In this video article, we describe in detail the process of setting up the experiment and provide detailed instructions to perform outgassing measurements using the RoR method. To demonstrate the efficacy of the method, the outgassing rates of two commonly used steels (stainless steel 304 and mild steel S20C) were measured before and after a preheat treatment to reduce the hydrogen outgassing rate. The pre- and post-treatment values were compared. Typical experimental results using a rather simple setup are presented to demonstrate the efficacy of the method optimized for evaluating low hydrogen degassing rates.
Caution: Please follow all appropriate safety practices while assembling the equipment and sample chambers. Please wear personal protective equipment (safety glasses, gloves, safety shoes, etc.).
1. Fabrication of a Sample Vacuum Chamber
2. Fabrication of the Oven
3. Experimental Setup for RoR Measurements
4. Measurement of Outgassing Rates
As expected, the residual gas after the bakeout was mostly hydrogen.7 The pressure rise measured using the SRG was linear over a long period of time (Figure 5). Thus, the readsorption effect might be negligible and the intrinsic outgassing rate (q) for the steels tested in this study can be evaluated using the RoR method.10 The measured pressure rise data was analyzed using the linear least squares fitting method. The outgassing rates of the sample chambers were determined from the slope (Figure 5).
The measured outgassing rate for untreated STS304 steel (sample 1) was 5.1 × 10−9 Pa m3 sec−1 m-2, which is consistent with the reported values.1-7 A ~22-fold reduction in outgassing was achieved with a medium-temperature heat pretreatment in vacuum furnace for 36 hr at 450 °C (Table 1). This demonstrates the effectiveness of heat pretreatment in reducing the hydrogen outgassing rate of stainless steel, further indicating that the degassing of hydrogen during the heat treatment is governed by a bulk diffusion mechanism. While the outgassing rates for untreated mild steels were very low (<~4 × 10−10 Pa m3 sec−1 m−2 (samples 2 and 3), the outgassing rates were second to the rates of stainless steels after intensive heat treatment.1,3,4 In addition, the outgassing rate for mild steel (sample 2) decreased by only 66% following heat treatment at 850 °C for 12 hr in vacuum furnace (Table 1), and no significant reduction in outgassing was observed.
The findings from these measurements strongly suggest that the difference in outgassing between stainless steels and mild steels can be attributed to the differences in the steel making processes, and in particular, the secondary metallurgy processes, during which impurity gases are extracted.16 A vacuum degassing process, such as the Ruhrstahl-Hausen process, is generally employed during the production of mild steels. Thus, mobile hydrogen is completely degassed during the steel-making process. In contrast, mixed-gas refining, such as argon-oxygen decarburization at atmospheric pressure, is primarily used during the production of stainless steels. This provides a reasonable explanation for the observed lower hydrogen outgassing rate of untreated mild steel compared to untreated stainless steel.7
Figure 1. Sample chamber. An example of a vacuum chamber made of steel. A steel cylinder and two end plates with flanges (CF35) were directly welded. The area of the inner surface is ~2,400 cm2 and the volume is ~7 L. Please click here to view a larger version of this figure.
Figure 2. Oven. A bird-eye view of the oven, together with the experimental setup and the sample vacuum chamber. A simple, box-shaped oven is adequate for this experiment. Please click here to view a larger version of this figure.
Figure 3. Experimental setup. A schematic of the experimental setup for the measurement of outgassing rates using the RoR method. A cylindrical sample chamber is placed inside a simple oven and pumped through an all-metal angle valve (AV). After bakeout, the SRG pickup head is attached and is switched on. The active temperature control is then initiated. CF: flange, KF: clamp flange, RGA: residual gas analyzer, and TMP: turbomolecular pump. Please click here to view a larger version of this figure.
Figure 4. Mounting of the SRG head on the vacuum chamber. The axis of the SRG head should be vertical within ±2° (max) as shown. A level meter should be used to align the head. Please click here to view a larger version of this figure.
Figure 5. Representative raw RoR data (dotted line) measured using the SRG after bakeout. The solid line (in blue) is the least-squares fit of the data. The slope of the curve corresponds to an outgassing rate of 4 × 10−10 Pa m3 sec−1 (H2 equivalent). The solid line at the bottom (in red) shows the measured temperature variation, which is within ±0.1 °C. Please click here to view a larger version of this figure.
Figure 6. Modification of a commercial SRG flange. The flange is thinned as per the design drawing and heat treated at 400 °C for 72 hr (Fo ~6.4) to reduce outgassing. The measured gas load on the SRG flange, together with the angle valve (from the surface exposed to the SRG side), was 8.3 (± 0.1) × 10−12 Pa m3 sec−1, which amounts to 15%-28% of the outgassing from the samples after heat treatment (Table 1). This background gas load must be subtracted from the total gas load on the sample vacuum chamber. Please click here to view a larger version of this figure.
Material | Sample no. | d (mm) | D (cm2/sec) | Heat treatment | Fo | q (Pa m3 sec-1 m-2) |
Stainless steel (304) | 1 | 3.3 | – | 5.1×10-9 | ||
5×10-7 | 450 °C, 36 hr | 2.4 | 2.3×10-10 | |||
Mild steel (S20C) | 2 | 10 | – | 2.6×10-10 | ||
1×10-4 | 850 °C, 12 hr | 17 | 8.8×10-11 | |||
3 | 10 | – | 4.0×10-10 |
Table 1: Measured outgassing rates. The rates (q) are total outgassing rates, in hydrogen equivalent units, and measured after an in situ bakeout at 150 °C for 48 hr. Fo represents the intensity of heat treatment (dimensionless); Fo = 4Dt/d2, where D is the diffusion constant at the heat pretreatment temperature and d is the thickness of the chamber.12,13
Numerous methods for the measurement of outgassing rates have been reported in the literature. Experimental methods include the throughput, conductance modulation, two-path, RoR, and variations of these methods. However, no one method is ideal for obtaining the necessary outgassing data.10 The RoR method using SRG, however, became the method of choice for measurement of low outgassing materials.11-13 SRG17 is often used as a secondary standard in high vacuum systems without erroneous pumping or outgassing action. The RoR method using SRG is particularly suitable for measuring hydrogen outgassing at room temperature after bakeout. In contrast, other UHV gauges can cause significant errors generated by the gauges themselves. An extractor gauge, for instance, is a type of UHV ion gauge with low outgassing. However, the gauge itself and the surrounding walls generate a gas load as large as 1 × 10−11 Pa m3 sec−1.18 This amounts to 14%-30% of the gas load from the samples following the heat treatment (Table 1).
The outgassing from SRG flange (CF35) must be taken into account when measuring samples with a small area. Though small in size, the hydrogen outgas from the flange is as large as 7.5 × 10−12 Pa m3 sec−1 and the flange is too thick to degas hydrogen without firing. This amounts to approximately 12%-26% of the outgassing from the samples after heat treatment (Table 1). Thus, this systematic error in the measured gas load must be corrected. Thinning the commercial SRG flange (Figure 6) and performing an appropriate heat treatment in vacuum will help minimize the outgassing. However, in a real situation, the combined background gas loads from the SRG flange assembly and the angle valve must be measured and corrected before the main measurements. Furthermore, using a thimble without a flange that is directly welded on the sample chamber is another good technique for measuring outgassing from very small samples (surface area <500 cm2) using a pinch-off copper tube instead of an angle valve.12,13
In addition, proper operation of the SRG is crucial to ensure the precise measurement of extremely low outgassing rates. The pressure range that the measurement is taken over is from 10−8 Pa to 10−3 Pa. The temperature control is especially important. A slow, constant temperature change of 0.14 °C/hr causes a 10% error in the measured values.
Thus, the active temperature control unit, comprising a copper cooling coil at a constant temperature of 15 °C and a proportional-integral-derivative controlled heater, was deployed in this study. The temperature was stabilized to within ±0.1 °C during the measurements (Figure 5). At this temperature stability, RoR measurements as low as 1 × 10−3 Pa/day could be made in a single day.
Fabrication of individual parts of the sample chamber with the same thickness is another important factor affecting the outgassing rate following heat treatment (Figure 1). As stated earlier, bulk diffusion governs the degassing of mobile hydrogen, at least in the initial stage of heat treatment. In the RoR method, the outgassing rate depends not only on the duration of the heat treatment but also strongly on the sample thickness.19 Thus, reporting the outgassing rate with respect to the intensity of heat treatment (for example, Fo = 4Dt/d2, Table 1)12,13 is recommended; simply reporting the duration of the heat treatment is misleading with respect to the intensity of heat treatment.
Using the protocol reported in this study that uses commercial parts to the extent possible, an outgassing rate lower than 1 × 10−10 Pa m3 sec−1 m−2 can be routinely measured from vacuum chambers made of steel. With careful design and under optimum experimental conditions, such a low rate can be measured from samples with a relatively small area. The surface area of the vacuum chamber used in this study is only 2,400 cm2, which is one-third of the surface area of the chambers (7,600 cm2) used in previous experiments for making similar measurements.5 The equipment identified in this protocol is specific to the most suitable commercial ones. It should be noted that with a proper, carefully designed experimental setup and protocol, other equipment or methods can be used for the same purpose.
Furthermore, although ferrous steels were used in this video protocol, the same techniques are applicable to the measurement of outgassing rates from numerous other materials that can be used for the fabrication of vacuum chambers.
The authors have nothing to disclose.
This work was supported jointly by the Converging Research Center Program through the Ministry of Science, ICT and Future Planning, Korea (NRF-2014M3C1A8048817) and R&D Convergence Program of NST (National Research Council of Science and Technology) of Republic of Korea (CAP-14-3-KRISS).
Sample chamber | |||
Stainless steel, 304 | POSCO (www.posco.co.kr) | ||
Mild steel, D3752 | Xiangtan Iron&Steel co.,LTD (http://www.hnxg.com) | ||
Mild steel, D3752 | SeAh Besteel (www.seahbesteel.co.kr) | ||
Name | Company | Catalog Number | Comments |
Cleaning | |||
Cleaning bath | Samill IDS | Ultrasonic cleaning, heating, timer, concentration control | |
Acetone | Samchun Chemical (www.samchun.com) | A1759 | HPLC GRADE (99.7%) |
Tekusolv | NCH Co. (www.nch.com) | 0368-0058J | Solvents |
BN cleaner | Henkel surface technologies (na.henkel-adhesives.com) | 6610263775 | Akkaline, pH 13 |
Ethanol | Fisher Scientific (www.fishersci.com) | A995-4 | HPLC Reagent(99.9%) |
Deionized water (Electro deionizer SYSTEM) | A.T.A (www.atagroup.co) | EDI SYSTEM | |
Liquid N2 gas | Hanyoung (www.gasmaster.co.kr) | B/T 176 L | LN2 dewar, purity 99.999% |
Name | Company | Catalog Number | Comments |
Welding | |||
Tungsten Inert Gas wedling machine | Thermal Arc (www.victortechnologies.com/thermalarc) | 400GTSW | Ar gas prefllow&postflow 8 liter/min, backflow 5 liter/min |
turning jig | Vactron (www.vactron.co.kr) | Made to order | Made to order |
Ar gas | Lindekorea (www.lindekorea.com) | Purity 99.999% | |
Name | Company | Catalog Number | Comments |
Leak test | |||
Leak detector | Adixen (www.adixen.fr/en/) | ASM380 | Pumping Speed(air): 9.7 l/s |
He gas | Lindekorea (www.lindekorea.com) | Purity 99.999% | |
Name | Company | Catalog Number | Comments |
Vacuum equipment | |||
Spinning rotor gauge | MKS Instruments (www.mks.com) | SRG-3 | Controller, head, and thimble set |
Industrial level meter | MKS Instruments (www.mks.com) | SRG-3 | For SRG assemble ± 1˚ |
Oscilloscope | Tektronix (www.tek.com) | TDS2012B | |
Residulal gas analyser | Balzers | QMA200 | m/e 0-100 |
TMP(HiPace 80) | Pfeiffer Vacuum (www.pfeiffer-vacuum.com) | PMP03941 | Pumping Speed(N2): 67 l/s |
Scroll pump | Anest Iwata (www.anest-iwata.co.jp) | ISP 90 | Pumping Speed(Air): 1.8 l/s |
All-metall easy close angle valve(CF35) | VAT Inc. (www.vatvalve.com) | 54032-GE02-0002 | Rotatable flange |
Angle valve(KF25) | MDC Vacuum Inc. (www.mdcvacuum.com) | KAV-100 | |
Five-Way Crosses | MDC | Made to order | CF4-1/2 Spool-rotatable 1-way to CF2-3/4 Nipple 3ea, Vacuum degassed at 400℃ for 3 days |
Reducing Tees | MDC | Made to order | CF4-1/2 Flange to CF2-3/4 Tees(Half flange), Vacuum degassed at 400℃ for 3 days |
Name | Company | Catalog Number | Comments |
Temperature control | |||
Chiller | JEIO Tech (www.jeiotech.com) | RW-2025G | |
Cooling line | LS Metal (www.lsmetal.biz) | C1100 | Level Wound Coil, Diameter 10mm |
Heater controllers | HMT | Made to order | Bakeout program controller |
Electrical heater tapes | Brisk heat (www.briskheat.com) | BIH101080L | |
Thermocouple(K type) | miraesensor (www.miraesensor.com) | MR-2290 | |
Handheld multimeter | Saehan (www.saehan.co.kr) | 3234 | |
Data recorder(Temp.) | Yokogawa (www.yokogawa.com) | GP10-1E1F-UC10 |