Vacuum 120 (2015) 139e146

Contents lists available at ScienceDirect

Vacuum
journal homepage: www.elsevier.com/locate/vacuum

Invited review

Practical requirements for the successful implementation and

subsequent dissemination of the redefined kilogram
Stuart Davidson a, *, James Berry a, Kilian Marti b
a
National Physical Laboratory, Hampton Road, Teddington, Middlesex TW11 0LW, UK
b
Federal Institute of Metrology METAS, Lindenweg 50, 3003 Bern, Switzerland

a r t i c l e i n f o a b s t r a c t

Article history: The kilogram is the last of the seven base units of the SI to be defined in terms of a material artefact
Received 1 May 2015 rather than by relation to an invariant of nature. Progress is being made towards a redefinition in terms
Received in revised form of the Planck constant realised via the watt balance and Avogadro experiments. The accuracy of these
10 July 2015
experiments is now at the level required by the mass community (2 parts in 108) and it is likely that the
Accepted 11 July 2015
Available online 14 July 2015
redefinition will be ratified in late 2018. The change to the definition of the kilogram presents issues
which need to be addressed in order to effectively implement the redefinition, to ensure continuity of the
mass scale and to efficiently disseminate the new definition to the user community. After redefinition the
Keywords:
Kilogram
realisation of the SI unit of mass will be possible via either the Avogadro or watt balance approaches. The
Mass two experiments work in a vacuum and so a link between mass standards in air (such as the Interna-
Standards tional Prototype Kilogram used to realise the current mass unit) and standards realised in vacuum will
Cleaning need to be established for initially fixing the Planck constant and subsequently for dissemination of the
Storage unit of mass. This paper describes research undertaken to prepare for the redefinition. Next generation
Transfer mass standards, compatible with use in vacuum, have been developed to improve mass stability while
optimising vacuum/air transfer characteristics. Methods for the transfer between and storage in vacuum,
inert gas and air have been investigated both gravimetrically and using surface analysis techniques such
as X-ray photoelectron spectroscopy (XPS) to characterise surface sorption effects. Additionally new
cleaning techniques for primary mass standards using UV activated ozone and low pressure plasma have
been developed to replace current manual cleaning methods. The implementation of this research will
ensure the maximum benefit is realised from the redefinition of the kilogram in terms of a fundamental
constant.
Crown Copyright © 2015 Published by Elsevier Ltd. All rights reserved.

1. Introduction invariant of nature. Progress is being made in the development and

In the next few years the world's measurement system, the SI,
will undergo a major change. Part of this change will be the
replacement of the existing definition of the kilogram, which de-
pends on the stability of an artefact made in the 19th century, with
a definition based on a constant, the Planck constant h, which lies at
the heart of modern quantum physics. This change will strengthen,
stabilise, and reunify the SI and, in addition, provide advantages to
science and engineering.
The kilogram is the last of the seven base units of the SI to be
defined in terms of a material artefact rather than by relation to an
* Corresponding author.
E-mail address: stuart.davidson@npl.co.uk (S. Davidson).
improvement of methods to realise the kilogram in terms of the
Planck constant via the watt balance and X-ray crystal density
(XRCD) (or Avogadro) experiments. The accuracy of these experi-
ments are either at or near the requirements of the mass commu-
nity (2 parts in 108) and it is likely that the redefinition will be
ratified in late 2018.
While achieving the aim of ensuring the long term stability of
the unit of mass, the change to the definition of the kilogram pre-
sents issues which need to be addressed in order to effectively
implement the redefinition, to ensure continuity of the mass scale
and to efficiently disseminate the new definition to the user com-
munity. After redefinition the realisation of the SI unit of mass will
be possible via either the Avogadro or watt balance approaches.
Due to the nature of the two experiments both will realise the ki-
logram in a vacuum and so a link between mass standards in air

http://dx.doi.org/10.1016/j.vacuum.2015.07.007
0042-207X/Crown Copyright © 2015 Published by Elsevier Ltd. All rights reserved.
140 S. Davidson et al. / Vacuum 120 (2015) 139e146
Fig. 1. Mass deviation (Dm) for copies of the international prototype kilogram, iden-
tified by number, with respect to the weight itself, measured at the Periodic
Verifications.
(such as the International Prototype Kilogram used to realise the
current mass unit) and standards realised in vacuum will need to be
established, initially in order to fix the Planck constant with refer-
ence to the current mass scale and subsequently for dissemination
of the unit of mass from the primary realisation experiments.
2. The current definition and its limitations
The kilogram is the unit of mass; it is equal to the mass of the
international prototype of the kilogram (IPK). The use of an artefact
clearly limits the ultimate stability with which the unit can be
defined. Fig. 1 illustrates the typical deviation of copies of IPK with
relation to the standard itself. Typical deviations of up to 50 mg can
be experienced although more recent calibrations at the BIPM have
suggested that carefully stored platinumeiridium standards are
more stable [1]. Nevertheless, an artefact based standard is sus-
ceptible to drift, particularly in the long term.
3. Progress towards the redefinition
The CIPM Consultative Committee for Mass and Related
Fig. 2. a: Principle of operation of the watt balance, moving phase. b: Principle of operation of the watt balance, weighing phase.
Quantities has made the recommendation that, in order for the
redefinition of the kilogram to be ratified, the following conditions
are met [2];

at least three independent experiments, including work from
watt balance and XRCD experiments, yield consistent values of
the Planck constant with relative standard uncertainties not
larger than 5 parts in 108,

at least one of these results should have a relative standard
uncertainty not larger than 2 parts in 108.
At present at least one watt balance experiment [3,4] and the
result from the International Avogadro Coordination (IAC) [5] are at
or near the uncertainty of 2 in 108, and the results for the value of h
are consistent. Further watt balance experiments are at various
stages of development and should be able to provide at least a third
independent realisation of h by 2017.
4. The primary realisation experiments
4.1. Watt balance
The watt balance [6,7] was conceived by Kibble at the National
Physical Laboratory (NPL). The principal of operation is shown
diagrammatically in Fig. 2(a) and (b).
The balance measures mass by comparing virtual electrical and
mechanical power, VI ¼ Mgu. The voltage V and velocity u are
measured during one stage of the experiment, Fig. 2(a), and the
current I and the weight Mg of the mass M are measured in a
separate stage, Fig. 2(b). The two stages are linked using a coil of
wire of length l placed perpendicular to the field B of a strong
magnet, the two stage process eliminates the need to determine
the values of B and l (they just need to be stable between the two
phases of the measurement). The electrical quantities, voltage and
resistance, can be measured in terms of the Planck constant via
quantum mechanical effects: the Josephson effect and the quantum
Hall effect (QHE) respectively. The Josephson effect occurs in
superconducting systems at low temperatures (usually about 4 K)
and can be harnessed to produce a standard voltage. The QHE oc-
curs in particular semiconductor structures at low temperatures
and provides a standard resistance. Fig. 3 shows the NPL watt
 S. Davidson et al. / Vacuum 120 (2015) 139e146
Fig. 3. The NPL Mk II watt balance (now at NRC, Canada).
balance which is now at NRC, Canada [3].
4.2. X-ray crystal density (XRCD)
The XRCD or Avogadro experiment [5] relates the kilogram to the
mass of a single silicon atom by counting the number of atoms of
silicon in a macroscopic sample (a highly polished and almost per-
fect sphere, Fig. 4). It is, of course, not possible to count individual
Fig. 4. A silicon sphere similar to those used in the XRCD measurements being
weighed on the NPL 1 kg mass comparator.
141
atoms but the task can be done indirectly by measuring the volume
occupied by an atom in a perfect crystal, measuring the volume of
the crystal and dividing one by the other. If the mass of the crystal is
also measured, the mass of a silicon atom can be calculated and the
Avogadro constant (NA) derived. The crystal volume measurement is
simplified by polishing the crystal into a near perfect sphere; a large
number of diameter measurements are then averaged to determine
the volume. The volume occupied by an atom is measured by using
X-ray interferometry to determine the spacing of planes in the
crystal lattice. In recent measurements, a sphere of almost pure
Silicon 28 has been used to avoid the limitations caused by the
difficulty of making accurate measurements of the amounts of the
different isotopes present in natural silicon.
5. Requirements for traceability to the new definition
At present the IPK is stored and weighed in air so the traceability
chain to national standard kilogram and thence to working stan-
dards is relatively straightforward. Due to the nature of the watt
balance and XRCD experiments, once redefined the kilogram will
be realised in vacuum. This adds significant complexity to the
dissemination of the mass scale and the additional uncertainty
components need to be carefully assessed. In particular the effect of
transferring a mass standard between air and vacuum must be fully
understood before robust traceability to mass in vacuum can be
achieved. As well as determining the sorption characteristic of
existing mass standards of platinumeiridium, stainless steel and
silicon new materials are being investigated with a view to
improving sorption performance and long term stability. New
cleaning techniques have also been developed to improve on the
current method of nettoyage-lavage used for primary plati-
numeiridium mass standard (including the IPK). Since mass stan-
dards will need to be used in vacuum to achieve traceability to the
primary realisation experiments, it may be that long term storage
in either vacuum or inert gas rather than air improves their stability
and this has also been investigated.
6. Airevacuum sorption characteristics
Key to achieving traceability to a primary realisation of the ki-
logram in vacuum is an understanding of the sorption character-
istics of the mass standards used to achieve the link between mass
in vacuum and mass in air. The work undertaken by National
Measurement Institutes to evaluate these sorption characteristics is
a specific application of the work undertaken by Redhead [8] and
others in the 1960s and 1970s determining the mechanisms for
desorption from surfaces in vacuum. The aim of the work reported
here is to determine, in real time, the desorption of molecules from
the surface of mass standards, at room temperature, when trans-
ferred from air to vacuum (and back).
Sets of sorption artefacts have been manufactured from plati-
numeiridium, stainless steel and silicon. The artefacts consist of
mass standards with closely matched mass values and volumes but
with different surface areas. A typical set of artefacts is shown in
Fig. 5.
By transferring the set between air and vacuum the differential
sorption of the individual artefacts can be used to calculate the
adsorption characteristics of the material in terms of mass change
per unit surface area. Fig. 6 shows a typical sorption curve for a
stainless steel mass standard transferred between air and vacuum.
The mass standard loses a total of about 35 mg when transferred
between air and vacuum. The mass loss is approximately the same
for pressures between 0.1 Pa and 0.001 Pa. The curve shows sig-
nificant hysteresis, the mass standard only fully reabsorbs surface
water when it is returned to ambient conditions (50% relative
142
Fig. 5. A typical set of sorption artefacts consisting of an integral kilogram and two and
four piece stacks (Surface area ratios 1:1.3:2.0).
humidity was used for the tests shown).
Test have been performed on the two materials typically used
for primary mass standards (platinumeiridium and stainless steel)
and sorption characteristics are broadly similar for weights with a
good surface finish (RA < 100 nm) with the surface sorption being
about 0.1 mg/cm2 [9]. A comparison between national measurement
institutes [10] has shown good agreement between sorption char-
acteristics using the same set of artefacts but other research has
shown significant variation between the sorption characteristics of
different artefacts of the same material (platinumeiridium)
[11e13]. This suggest that in order to achieve the best uncertainty
on airevacuum transfer of a mass standard real time sorption
measurements, using a set of artefacts of the same material and
surface finish, should be made.
7. New materials for primary mass standards
The use of platinum for primary mass standards dates from the
original kilogram of the archives (1799). The IPK, which currently
defines the mass scale, was manufactured from platinumeiridium
alloy in 1887 (the iridium being added to improve the mechanical
properties). Platinumeiridium turns out to be an excellent choice
for a mass standard; its density means the weight is compact so has
low air buoyancy and a small surface area. It is also (relatively) inert
and has a high hardness thus enhancing the long term stability of
Fig. 6. Sorption curve for 1 kg mass standard transferred between ambient air (at 50% RH) and vacuum (0.001 Pa).
S. Davidson et al. / Vacuum 120 (2015) 139e146
Fig. 7. Typical (initial) XPS survey scan of PteIr alloy showing oxide (O1s) and carbon
(C1s) contamination together with traces of Zinc and Copper.
artefacts. A significant issue with the more general use of plati-
numeiridium for primary mass standards is the expense of the
material. It also requires a large buoyancy correction when it is
compared (in air) with working standards of stainless steel and
although its (relative) magnetic permeability is low (typically
1.0003) this still required a significant correction when used in the
watt balance experiment where the magnetic field in the apparatus
is very strong.
Samples of several potential new materials for primary mass
standards were evaluated together with materials already in use. X-
Ray Photoelectron Spectroscopy (XPS) was used to characterise the
samples of the materials on receipt and after three months in air, to
look at the accretion of contamination (Fig. 7). All metallic surfaces
showed similar oxide thickness (approximately 0.7 nme1.0 nm)
and trace contamination from the manufacturing/polishing process
 S. Davidson et al. / Vacuum 120 (2015) 139e146 143

(mainly other metals). As expected there was also a layer of

carbonaceous contamination present on each sample which was
1e2 nm in depth. Silicon samples showed slightly thinner layers of
oxide (0.6 nm) and carbonaceous contamination (0.8 nm). After 3
months all the samples had gained approximately 1 nm of carbo-
naceous contamination but no other changes to the surfaces were
detected.
Pure iridium, gold and nickel alloys and tungsten were evaluated
as potential materials for new mass standards and were charac-
terised with respect to their magnetic properties, hardness, and
surface finish achievable. Additionally copper samples plated with
gold and with rhodium were also evaluated, with tests including
humidity, pressure and thermal cycling to assess the robustness of
the coating. Table 1 summarises the key properties of the materials
evaluated. Hard and dense materials tend to make good mass
standards are they are more durable and have a smaller surface area.
Given its good mechanical properties (hardness and density) Fig. 8. Mass of 100 g tungsten mass standard over 3 for 3 series of vacuum weighings
and its low magnetic permeability, samples and mass standards (the mass shown is a deviation from the 100 g nominal value).
have been produced from both single and poly crystal tungsten.
With the single crystal samples a good surface finish proved diffi-
determine the accuracy with which the mass unit cab ne realised.
cult to achieve as polishing resulted in nanoscale pitting. This is
The cleaning of the IPK and its official copies is performed using
likely to be a function of the quality of the crystal as the sample
a process called nettoyage-lavage [14,15] which involves rubbing
used may have contained significant inclusions, dislocations and
the weights with a clean chamois soaked in an ethanol/ether
voids resulting in issues with the polishing. A forged sample of
mixture and then spraying with a jet of steam. While it has been
poly-crystal material, with a density close to the theoretical
shown that this process is effective at removing hydrocarbon
maximum (19.227 g/cm3 compared with 19.254 g/cm3), has been
contamination from the surface of mass standards at the BIPM [1]
machined and polished to give 100 g mass standard with a surface
other laboratories have had variable results when implementing
finish of better than 10 nm (Ra), the major limit on the surface finish
this procedure [16]. Novel cleaning techniques such as UV activated
achievable being the grain structure of the material. The mass of the
ozone and low pressure hydrogen plasma have been developed to
100 g standard in vacuum was both stable (after a period of a few
improve the repeatability of the cleaning of primary mass stan-
hours in vacuum) and repeatable. Fig. 8 shows the mass of the
dards. Both techniques have the advantage that they are non-
standard for 3 sets of vacuum weighings.
contact so the effectiveness and repeatability of the cleaning ach-
ieved should be governed solely by the parameters set (UV in-
8. Cleaning of primary mass standards tensity and ozone concentration or plasma intensity and pressure).
Both METAS (Switzerland) [16,17] and NPL (UK) [18,19] have eval-
The current definition of the kilogram relates to the IPK in a uated UV/ozone and H-plasma non-contact cleaning methods in
cleaned state. After redefinition the cleanness of mass standards parallel with the traditional nettoyage-lavage method. The results
will still be crucial in maintain their stability and sorption charac- from the two laboratories are shown in Figs. 9 and 10 for untreated
teristic and in particular the cleanness of silicon spheres will samples (pre-cleaned) and after the cleaning processes identified.

Table 1
Comparison of the properties of materials investigated for use as primary mass standards.

Material Advantages Disadvantages

Platinumeiridium Well characterised material Expensive

Stainless steel
Silicon
(Tertiary/quaternary) gold alloy
Iridium
Nickel super alloy (Udimet720)
Single crystal tungsten
Poly crystal tungsten
(Gold and rhodium) plated copper
Easy to machine
Established manufacturing techniques
Used for the majority of current mass standards
Excellent surface finish achievable
Natural silicon readily available
Very low magnetic permeability
Lower sorption than metals
Good magnetic properties
Easy to machine
Dense
Hard
Good magnetic properties
Dense
Similar density to stainless steel
Hard
Good magnetic properties
Excellent magnetic properties
Density similar to that of PteIr
Excellent magnetic properties
Density similar to that of PteIr
Excellent magnetic properties
Easy to manufacture
Similar density to stainless steel
Relatively high magnetic permeability
High magnetic permeability
Complex alloys e surfaces difficult to analyse
Low density so must be weighed in vacuum
Potential static issues (attracts dust and can also affect weighing results)
Low relative hardness
Samples analysed showed anomalous sorption characteristics
Difficult to machine
Expensive
Relatively expensive and difficult to obtain
Difficult to manufacture artefacts
High quality crystals of suitable size expensive and difficult to obtain
Quality of raw material critical in obtaining magnetic properties, avoiding
porosity and surface finishing
Surface relatively soft
Quality of artefacts relies on good coating process
144 S. Davidson et al. / Vacuum 120 (2015) 139e146
Fig. 9. Cleaning results from METAS for six material samples (cycled samples have
been transferred between air and vacuum before measurement).
Note that 1 nm depth of hydrocarbon contamination adds about
0.03 mg/cm3 to the weight of a mass standard (equivalent to about
2 mg on a platinumeiridium kilogram, 4 mg on a stainless steel ki-
logram and about 6 mg on an Avogadro silicon sphere).
While the results of both METAS and NPL show advantages for
the UV/Ozone and H-Plasma cleaning methods over the nettoyage-
lavage the relative effectiveness of the methods is different at the
two laboratories, the H-Plasma being more effective at METAS and
the UV/Ozone at NPL. This suggests that the implementation of the
cleaning methods is critical in optimising their performance. For
example it is likely that the UV ozone method implemented at
METAS was not as effective since all surfaces were not exposed
equally to similar UV intensities due to the physical construction of
the apparatus. Additionally it should be noted that the imple-
mentation of the nettoyage-lavage method at METAS and particu-
larly at NPL was almost certainly not optimised in the way it has
been at the BIPM. The results do, however, illustrate the variability
of the technique when compared with the non-contact techniques.
A more general issue with the cleaning of mass standards is the
stability of the surfaces, and therefore of the mass value, after
cleaning. Fig. 11 shows the recontamination of a platinumeiridium
surface following cleaning for vacuum cycles venting to air and/or
nitrogen. It can be seen from the results that the nitrogen venting
cycle maintained the cleanliness of the surface most effectively and
that the introduction of an intermediate nitrogen stage in the air-
evacuum cycle did not significantly affect the recontamination of
the surface. However, with regard to the stability of the surface the
results show that the recontamination of the surface is most rapid
Fig. 10. Cleaning results from NPL for silicon and platinumeiridium samples.
Fig. 11. Recontamination of platinumeiridium surface following cleaning for three
different venting cycles.
following cleaning and it is also self-limiting so approaches a more
stable value once the surface has accreted approximately a mono-
layer of contamination.
9. Vacuum and inert gas storage
The current definition of the kilogram, the IPK is stored and used
in air so any benefits from storage and transfer of weights in inert
gas or vacuum would be outweighed by the added complexity in
maintaining and disseminating the unit of mass. The realisation
experiments for the redefined kilogram (watt balance and XRCD
experiments) both operate in vacuum so the storage and transfer of
primary mass standards in vacuum or inert gas may be of benefit in
the dissemination of the redefined mass unit by improving the
long-term stability of artefact mass standards. Experiments have
been undertaken to assess the relative stability of weights stored in
vacuum, air and inert gas [20]. Figs. 12 and 13 show the results of
storage tests for stainless steel and silicon mass standards stored in
air and argon and weighed in vacuum. Transfer of the weights from
the storage vessels for weighing in vacuum took place in a glove box
to ensure the weights were only exposed to the storage medium
(air or argon) and vacuum. Each point represents one series of
weighings undertaken in vacuum.
The results suggest that the masses stored in argon are more
stable than those stored in air which gained mass significantly over
the period of the test. However, previous studies [21] have shown
 S. Davidson et al. / Vacuum 120 (2015) 139e146 145

Fig. 12. Mass stability of steel kilograms stored in air and argon (Mass changes are e 25 ng/day for argon storage and þ97 ng/day for air storage).

Fig. 13. Mass stability of silicon weights (500 g) stored in air and argon (Mass changes are e 17 ng/day for argon storage and þ80 ng/day for air storage).

that masses transferred between air and vacuum show an accel-
erated accretion of surface contamination relative to masses stored
on either air or vacuum. Thus it is likely that the mass increase in
the air stored masses is more a function of the repeated transfer
between air and vacuum than the storage in air. The standard
deviation of the data is approximately 3 mg for all four sets but
changes of up to 8 mg between consecutive weighings can be seen.
This is due to the increased handling of the masses necessary for
transfer between the storage media and the vacuum balance.

Fig. 14. Post cleaning storage of nickel-alloy sample in air, vacuum and nitrogen. Fig. 15. Post cleaning storage of silicon sample in air, vacuum and nitrogen.
146 S. Davidson et al. / Vacuum 120 (2015) 139e146
Figs. 14 and 15 show the level of hydrocarbon contamination
accreted on nickel alloy and silicon samples stored in air, vacuum
and inert gas for 30 days following cleaning. It can be seen that the
samples stored in nitrogen remain the least contaminated with
samples stored in vacuum showing a significantly higher level of
hydrocarbon contamination.
10. Conclusions
The research described in this paper has been performed with
the aim of facilitating the redefinition of the kilogram via the watt
balance and X-ray crystal density experiments and to enable
maintenance and dissemination of the mass unit following the
redefinition. Optimum operating pressure for vacuum weighing
have been determined and the most effective methods of trans-
ferring mass standards between air and vacuum have been deter-
mined. Next generation mass standards of tungsten have been
developed which are both compatible with use in the watt balance
experiment and optimised for transfer to and use in vacuum. New
non-contact cleaning methods have been investigated and show an
improvement in repeatability over traditional manual cleaning
methods. Storage conditions for mass standards have also been
investigated. Long term (5 years) storage in inert gas shows the
potential to improve ion mass stability but in the shorter term the
(lack of) repeatability or transfer to and from inert gas for use in air
or vacuum limits the benefits of inert gas storage. With the benefit
of careful characterisation of the additional uncertainty compo-
nents inherent in achieving traceability to a unit realised in vacuum
the impact of the redefinition on uncertainties for the mass scale
realised at National Measurement Institutes will be minimised.
Acknowledgement
The authors gratefully acknowledge support and funding for this
work from the National Measurement Office (under the UK Na-
tional Measurement System Engineering & Flow Metrology Pro-
gramme). The research leading to these results has also received
funding from the European Union on the basis of Decision No 912/
2009/EC.
References
[1] Michael Stock, et al., Calibration campaign against the international prototype
of the kilogram in anticipation of the redefinition of the kilogram part I:
comparison of the international prototype with its official copies, Metrologia
52 (2015) 310.
[2] Recommendation of the Consultative Committee for Mass and Related
Quantities submitted to the International Committee for Weights and
Measures: Recommendation G 1 (2013) On a New Definition of the
Kilogram.
[3] C.A. Sanchez, B.M. Wood, R.G. Green, J.O. Liard, D. Inglis, A determination of
Planck's constant using the NRC watt balance, Metrologia 51 (2014)
S5eS14.
[4] S. Schlamminger, D. Haddad, F. Seifert, L.S. Chao, D.B. Newell, et al., Deter-
mination of the Planck constant using a watt balance with a superconducting
magnet system at the National Institute of Standards and Technology, Met-
rologia 51 (2014) S15eS24.
[5] Y. Azuma, et al., Improved measurement results for the Avogadro constant
using a 28Si-enriched crystal, Metrologia 52 (2015) 360.
[6] B.P. Kibble, A measurement of the gyromagnetic ratio of the proton by the
strong field method, in: J.H. Sanders, A.H. Wapstra (Eds.), Atomic Masses and
Fundamental Constants, vol. 5, Plenum, New York, 1976, pp. pp.545epp.551.
[7] R. Steiner, History and Progress on Accurate Measurements of the Planck
Constant, 2013, p. 76. Reports on Progress in Physics.
[8] P.A. Redhead, Thermal desorption of gases, Vacuum 12 (1962) 203e211.
[9] S. Davidson, Determination of the effect of transfer between vacuum and air
on mass standards of platinumeiridium and stainless steel, Metrologia 47
(2010) 487e497.
[10] J.J. Berry, et al., Report on the CCM WG TG1 Pilot Comparison to Measure
Water Vapour Sorption on Stainless Steel Mass Standards, November 2013.
NPL Report ENG 46.
[11] Y. Kobayashi, Discussion of correction for gas adsorption, Bull. NRLM 39
(1990) 582e596.
[12] M. Kochsiek, Measurement of water adsorption layers on metal surfaces,
Metrologia 18 (1982) 153e159.
[13] A. Picard, H. Fang, Methods to determine water vapour sorption on mass
standards, Metrologia 41 (2004) 333e339.
[14] G. Girard, The Washing and Cleaning of Kilogram Prototypes at the BIPM,
BIPM, Se vres, 1990.
[15] G. Girard, The third periodic verification of national prototypes of the kilo-
gram (1988e1992), Metrologia 31 (1994) 317e336.
[16] K. Marti, P. Fuchs, S. Russi, Cleaning of mass standards: a comparison of new
and old techniques, Metrologia 49 (2012) 628.
[17] K. Marti, P. Fuchs, S. Russi, Cleaning of mass standards: II. A comparison of
new techniques applied to actual and potential new materials for mass
standards, Metrologia 50 (2013) 83.
[18] James Berry, Stephen Downes, Stuart Davidson, UV/ozone cleaning of plat-
inum/iridium kilogram mass prototypes, Metrologia 47 (2010) 410.
[19] James Berry, Stuart Davidson, Pauline Barat, Richard Davis, Comparison of UV/
ozone cleaning of platinum/iridium kilogram mass standards with nettoyage-
lavage cleaning, Metrologia 48 (2011) 181.
[20] James Berry, Stuart Davidson, Evaluation of equipment and procedures for the
transfer and storage of mass standards in inert gas, Metrologia 48 (2011)
391e398.
[21] Stuart Davidson, Determination of the effect of transfer between vacuum and
air on mass standards of platinumeiridium and stainless steel, Metrologia 47
(2010) 487.