Comparison of the standards for high and ultrahigh vacuum at three national standards laboratories

Abstract

A preliminary phase of an international comparison of standards for high and ultrahigh vacuum was carried out by the National Institute of Standards and Technology (NIST), the National Physical Laboratory-Teddington (NPL-UK), and the Physikalisch-Technische Bundesanstalt (PTB). A spinning rotor gauge (SRG), a Bayard-Alpert gauge (BAG) and an extractor gauge (EXG) were chosen as transfer standards. The comparison was carried out in a star-like pattern with PTB as pilot laboratory. The argon pressures generated by the standards at 9×10−4 Pa were compared by measuring the accommodation coefficient of the SRG, the argon pressures from 3×10−7 Pa to 9×10−4 Pa by measuring the sensitivity of the two ionization gauges. The accommodation coefficients determined at NIST and PTB indicate a difference between the pressures generated by the NIST and PTB standards of (PNIST−PPTB)/PPTB=(0.09±0.11)% (standard or one-sigma uncertainty). For the pressures between 3×10−7 Pa and 9×10−4 Pa the results obtained at NIST and PTB had a mean difference of 〈(PNIST−PPTB)/PPTB〉=(0.24±0.12)%, with a maximum difference of (1.2±0.6) at the lowest pressure. The NIST-PTB differences are all within the combined uncertainties of the two standards. Large transfer standard instabilities and inconsistent results in a first NPL-PTB comparison prompted a repeat set of measurements. An average of the two sets of SRG measurements indicates a significant difference between NPL and PTB standards of (PNPL−PPTB)/PPTB=(1.31±0.14)%. The two sets of ionization gauge measurements effectively repeated, but the results for the individual gauges are inconsistent. The EXG results indicate no significant pressure dependence in the difference between the standards for pressures below 9×10−4 Pa. The BAG results indicate a significant increase in the difference between the standards as the pressure is reduced, with (PNPL−PPTB)/PPTB becoming as large as 5% or 6% at the lower pressures. Several potential problems with the BAG operation were identified, which might indicate that more confidence should be placed in the EXG results. However, there is no reliable evidence that the BAG comparison results were actually affected, and the results for each gauge are so internally consistent that neither can be rejected. Further work should be directed towards resolving this discrepancy.

Final report of key comparison CCM.P-K3: absolute pressure measurements in gas from 3 × 10−6Pa to 9 × 10−4Pa

“…Past international comparisons using vacuum gauge artifacts have also used repeated measurements at the pilot laboratory during the course of the comparison to characterize the uncertainty due to long-term shifts [11,13,14,15]. One method [11,15] assumes that the pilot laboratory value of a i , at the time of the participant measurement of a i , would be midway between the pilot laboratory value taken before and after the participant measurement. According to this method, an estimate of u LTS (a i ) is taken as one-half of the difference of the before and after values.…”

“…The combined standard uncertainty 11 in the predicted gauge pressure calculated using eq. (5), for each gauge i at each NMI j and each cycle m, is estimated from the root-sumsquare of the component uncertainties [16]:…”

“…Next a pressure-independent correction is applied to the entire set of data at each NMI for each calibration cycle [11]. This normalization comes from assuming the generated pressure of the primary standard of the NMI, at 9×10 -4 Pa, was the same whether it was being measured with an ion gauge or an SRG.…”

“…The long-term stability of SRG rotors is typically 1 % or better [10], and one of the two rotors for which the pilot laboratory had repeat data showed a stability of 0.2 % over an 18 month period. Due to the superior calibration stability of the SRGs, they were used to normalize the ionization gauge data at the highest calibration pressure, an approach that was also used in an earlier comparison of vacuum standards by several of the participants [11].…”

“…(1). Since the primary standards operated at different temperatures and the ionization gauges are sensitive to gas density ( [11] and [18]), equivalent inverse correction factors were determined for the common reference temperature of 23 ºC. This was done by multiplying the individual inverse correction factors (prior to averaging of eq.…”