Report of key comparison CCQM – K92

Report of key comparison CCQM – K92

"Electrolytic conductivity at 0.05 S·m-1 and 20 S·m-1“

Final report

L. Vyskočil, M. Máriássy (SMU),Adrian Reyes, Marcela Monroy (CENAM), Alena Vospělová (ČMI), Bertil Magnusson,Rauno Pyykkö (SP), Elena Kardash (INPL), Fabiano Barbieri Gonzaga, I.C.S. Fraga, J.C.Lopes, W.B. Silva Jr, P.P. Borges, W.F.C. Rocha (INMETRO), Francesca Durbiano, E. Orrù(INRiM), Kenneth W. Pratt (NIST), Pia Tønnes Jakobsen, Hans Dalsgaard Jensen, JørgenAvnskjold (DFM), L.A. Konopelko, Y.A. Kustikov, V.I. Suvorov (VNIIM), Song Xiaoping,Wang Hai (NIM), Steffen Seitz, Petra Spitzer (PTB), Vladimir Gavrilkin, Leonid Prokopenko,Oleksiy Stennik (Ukrmetrteststandart), Wladyslaw Kozlowski, Joanna Dumanska - Kulpa,Izabela Grzybowska (GUM), Yury A. Ovchinnikov (VNIIFTRI), Zsófia Nagyné Szilágyi,Judit Fükö (MKEH)

July 2013

Slovak Institute of MetrologyKarloveská 63, SK-84255 Bratislava 4, Slovakia

CCQM-K92 2

Abstract

The aim of the key comparison CCQM-K92 was to demonstrate the capabilities of the par-ticipating NMIs to measure electrolytic conductivity of an unknown sample.

Two samples with nominal electrolytic conductivity values of 0.05 S·m-1 and 20 S·m-1 havebeen prepared for comparison. For the first time conductivity value larger than those given inthe IUPAC document [1] was measured in CCQM comparison. Thus no calibration standardswith similar conductivity value were available. The comparison was an activity of the Elec-trochemical Working Group (EAWG) of the CCQM and was coordinated by SMU.

In the comparison NMIs from fifteen countries took part. The higher conductivity (20 S·m-1)was measured by ten participants. Good agreement of the results was observed for the major-ity of participants.

CCQM-K92 3

Contents

ABSTRACT .............................................................................................................................................................. 2

INTRODUCTION................................................................................................................................................ 4

METROLOGY AREA .................................................................................................................................................. 4BRANCH ................................................................................................................................................................ 4SUBJECT ................................................................................................................................................................ 4TIME SCHEDULE ...................................................................................................................................................... 4PARTICIPANTS......................................................................................................................................................... 4

SAMPLE DESCRIPTION...................................................................................................................................... 5

SAMPLE PREPARATION AND DISTRIBUTION.................................................................................................................... 5REPLACEMENT SAMPLE PREPARATION.......................................................................................................................... 5CHECK OF HOMOGENEITY ......................................................................................................................................... 6CHECK OF STABILITY................................................................................................................................................. 7STATISTICAL TESTING OF STABILITY .............................................................................................................................. 9

RESULTS ......................................................................................................................................................... 10

SAMPLE WITH NOMINAL VALUE 0.05 S·M-1

................................................................................................................ 10SAMPLE WITH NOMINAL VALUE 20 S·M

-1................................................................................................................... 12

DISCUSSION ................................................................................................................................................... 14

THE FORMULAS FOR CALCULATION OF ESTIMATORS ..................................................................................................... 14CALCULATION OF THE DEGREES OF EQUIVALENCE ........................................................................................................ 14COMMUNICATION WITH THE PARTICIPANTS ................................................................................................................ 15REFERENCE VALUE FOR 0.05 S·M

-1.......................................................................................................................... 15

Degrees of Equivalence - 0.05 S·m-1

............................................................................................................. 16

REFERENCE VALUE FOR 20 S·M-1

............................................................................................................................. 17Degrees of Equivalence - 20 S·m

-1................................................................................................................ 18

CONCLUSIONS ...................................................................................................................................................... 19ACKNOWLEDGMENT .............................................................................................................................................. 19

REFERENCES ................................................................................................................................................... 19

APPENDIX....................................................................................................................................................... 20

ADDRESSES OF PARTICIPANTS................................................................................................................................... 20

CCQM-K92 4

Introduction

Metrology AreaAmount of Substance

BranchElectrochemistry

SubjectDetermination of the electrolytic conductivity of two unknown samples with nominalvalue 0.05 S·m-1 a 20 S·m-1 (Water solution of KCl).

Time scheduleDispatch of the samples 9 February 2011Deadline for receipt of the report 25 April 2011Discussion of results EAWG meeting, April 2011 Draft A report October 2011 Draft B report September 2012 Draft B report corrected February 2013

Participants

Participants are listed in Table 1. VNIIM and VNIIFTRI measured one sample each, as theirresponsibilities are in different ranges in electrolytic conductivity.

Table 1 Table of Participants

No

Acronym Institute Country Contact Person

1 CENAM CENAM MEXAdrian Reyes,Marcela Monroy

2 CMI Český metrologický ústav CZE Alena Vospělová

3 DFM Danish Fundamental Metrology DNK Pia Tønnes Jakobsen

4 GUM Central Office of Measures POL Wladyslaw Kozlowski

5 INMETRONational Institute of Metrology, Quality andTechnology

BRA Fabiano Barbieri Gonzaga

6 INPL The National Physical Laboratory of Israel ISR Elena Kardash

7 INRiM Instituto Nazionale di Ricerca Metrologica ITA Francesca Durbiano

8 MKEH Hungarian Trade Licensing Office HUN Zsófia Nagyné Szilágyi

9 NIM National Institute of Metrology CHN Song Xiaoping

10 NIST National Institute of Standards and Technology USA Kenneth W. Pratt

11 PTB Physikalisch-Technische Bundesanstalt DEU Steffen Seitz

12 SMU Slovenský metrologický ústav SVK Leoš Vyskočil

13 SP SP Technical Research Institute of Sweden SWE Bertil Magnusson

CCQM-K92 5

14 UMTS Ukrmetrteststandart UKR Vladimir Gavrilkin

15 VNIIFTRINational Research Institute Physicotechnicaland Radio Engineering Measurements

RUS Yury A. Ovchinnikov

16 VNIIM D.I. Mendeleyev Institute for Metrology RUS Prof. L.A. Konopelko

Sample description

Sample preparation and distribution

Samples consisted of water solutions of potassium chloride. Reagent grade KCl and distilledwater were used for the solution with electrolytic conductivity 20 S·m-1. The 0.05 S·m-1 solu-tion was prepared by diluting the above solution with distilled water. Both solutions were pre-pared on February 2, 2001. The solutions were filled into 500 mL HDPE bottles (Nalgene),which were subsequently sealed into aluminium laminated plastic foil to prevent compositionchange of the solutions.The samples were distributed to the participants on February 9, 2011 by courier companiesFedEx and DHL (Russia, Mexico). In one case (NIM, China) the sample was received dam-aged. A replacement bottle was sent immediately and was received intact. The sample sent toMoscow stuck at the customs and was later returned to SMU. The second shipment arrived atits destination.

Replacement sample preparation

Some participants reported instability of 0.05 S·m-1 sample after opening, and the same effectwas confirmed by the coordinating laboratory. Therefore a new batch of this solution wasprepared from high-purity KCl and deionised water and the new samples were distributed tothe participants free of charge. The stability of new samples after opening was satisfactory.

Check of the bottle integrity

Participants were requested to weigh the received bottles to verify that they were unchangedduring the transport. The mass change was smaller than 0.05 g in all cases except VNIIM,where the difference amounted up to 0.5 g, and NIM that reported one leaking bottle with KClcrystallisation in the bag. NIM was sent a replacement bottle.

Figure 1 Relative differences between the weights of bottles - 0,05 S·m-1

CCQM-K92 6

Relative differences between the weights of bottles - 0,05 S·m-1

-0,010%

-0,008%

-0,006%

-0,004%

-0,002%

0,000%

0,002%

0,004%

0,006%

0,008%

0,010%

DNK MEX CZE POL BRA ISR ITA ITA HUN CHN DEU SVK SWE UKR RUS USA

Figure 2 Relative differences between the weights of bottles - 20 S·m-1

Relative differences between the weights of bottles - 20 S·m-1

-0,010%

-0,008%

-0,006%

-0,004%

-0,002%

0,000%

0,002%

0,004%

0,006%

0,008%

0,010%

DNK DNK MEX POL BRA ISR HUN CHN DEU UKR RUS

Check of Homogeneity

Homogeneity of the solution was checked after filling. To identify possible trends, solutionsfrom the first and the last bottle were measured. Both results were identical within uncertaintyof measurement. The data are depicted in figures 3 and 4.

CCQM-K92 7

Figure 3 Check of homogeneity for 20 S·m-1 solution

Check of Homogeneity for 20,0 S/m

20,20

20,22

20,24

20,26

20,28

20,30

20,32

20,34

Co

nd

uct

ivit

y

( S

/m)

First bottle Last bottle

Figure 4 Check of homogeneity for 0.05 S·m-1 solution

Check of Homogeneity for 0,05 S/m

0,04985

0,04990

0,04995

0,05000

0,05005

0,05010

0,05015

0,05020

Co

nd

uct

ivit

y

( S

/m)

First bottleLast bottle

Check of Stability

Several bottles were selected for stability measurements. Samples were measured in irregularintervals during 80 to 90 days. The measurement results for solution with nominal conductiv-ity 0.05 S·m-1 are given in figure 5 and table 2.

Figure 5 The stability measurement for sample with nominal conductivity value 0.05 S·m-1

CCQM-K92 8

Measurement of Stability for 0,05 S/m

0,04990

0,04995

0,05000

0,05005

0,05010

0,05015

0,05020

0 20 40 60 80 100Days

Co

nd

uct

ivit

y S

/m

Table 2 The stability Measurement for Sample with Nominal Value 0.05 S·m-1

Days Date Bottle Conductivity(S.m-1)

U (k=2)(S.m-1)

0 2011-03-22 B3 0,050040 0,0000310 2011-03-22 B3 0,050041 0,00003149 2011-05-10 B3 0,050058 0,00003149 2011-05-10 B3 0,050031 0,00003159 2011-05-20 B17 0,050034 0,00003159 2011-05-20 B17 0,050033 0,00003179 2011-06-09 B10 0,050054 0,00003179 2011-06-09 B10 0,050052 0,000031

The measurement results for solution with nominal conductivity 20 S·m-1 are given in figure 6and table 3.

Figure 6 The stability measurement for sample with nominal conductivity value 20 S·m-1

CCQM-K92 9

Measurement of Stability for 20 S/m

20,240

20,250

20,260

20,270

20,280

20,290

20,300

20,310

20,320

0 20 40 60 80 100Days

Co

nd

uct

ivit

y S

/m

Table 3 The stability measurement for sample with nominal conductivity value 20 S·m-1

Days Date Bottle Conductivity(S.m-1)

U (k=2)(S.m-1)

0 2011-03-09 B9 20,288 0,0120 2011-03-09 B9 20,287 0,0120 2011-03-09 B9 20,287 0,0120 2011-03-09 B9 20,287 0,0122 2011-03-11 B9 20,288 0,0122 2011-03-11 B1 20,288 0,0122 2011-03-11 B1 20,288 0,0122 2011-03-11 B1 20,287 0,01262 2011-05-10 B1 20,287 0,01262 2011-05-10 B18 20,287 0,01272 2011-05-20 B18 20,287 0,01272 2011-05-20 B18 20,287 0,01292 2011-06-09 B24 20,288 0,01292 2011-06-09 B24 20,288 0,012

Statistical testing of stability

The statistical testing of stability is based on the fact, that for a stable sample the is the trendstatistically not significant. Either a test of correlation coefficient can be used the slope ofa regression line can be tested. The procedure for the test of the slope is following:Using least-squares regression the slope b1 of the regression line and residual variance sR arecalculated. Test criterion t is calculated as follows:

CCQM-K92 10

( )R

i

s

xxbt

21 −

= (1)

In the statistical tables for critical values of Student distribution a critical value tα(n-2) for (n-2) degrees of freedom at significance level α=0,05 (95 % probability) is looked up. If thevalue of the test criterion does not exceed the critical value, the slope is statistically not sig-nificant.

Table 4 Results of Statistical Testing of Stability

Sample Value of Test.

Criterion t

Critical

Value

Degrees of

Freedom

Significance

Level

Verdict

0,05 S/m 0,779 2,447 6 0,05No trend was

observed

20,0 S/m 1,241 2,197 12 0,05No trend was

observed

As can be seen from the data in table 4, for both solutions the stability during the given timeperiod was confirmed.

Results

Sample with nominal value 0.05 S·m-1

The measurement conditions and methods used at different institutes are given in Table 5.The results of measurements of the new sample with nominal conductivity of 0.05 S·m-1 aregiven in Table 6 and displayed graphically in Figure 7. Four laboratories from 15 useda primary measurement procedure.

Table 5 Conditions of Measurement at Various Institutes - 0.05 S·m-1

Institute CountryDate of

reportTraceability Measurement

frequency

[Hz]

DFM DNK 2011-04-13 Primary CRM DFM Jones type cell 111 - 1000

CENAM MEX 2011-04-25 SMU CRM Jones type cell 1000 - 3000

CMI CZE 2011-06-01Primary

measurementRemovable central part 1000

GUM POL 2011-04-22 DFM CRM Jones type cell 12 - 1820

INMETRO BRA 2011-04-20Primary

measurementPiston type 20 - 5000

INPL ISR 2011-04-06 SMU CRM Four electrode DC cell DC method

CCQM-K92 11

INRIM ITA 2011-04-19Primary

measurementRemovable central part 20 - 2000 000

MKEH HUN 2011-04-19 OIML R56 Four electrode DC cell DC method

NIM CHN 2011-04-21 IUPAC Jones type cell 20 - 5000

PTB DEU 2011-04-06Primary

measurementPiston type 16 000 - 20 000

SMU SVK 2011-03-24 IUPAC Jones type cell 1000

SP SWE 2011-04-06 IUPAC WTW TetraCon 325 ?

UMTS UKR 2011-04-27 Primary CRM UMTS Jones type cell 1000

VNIIFTRI RUS 2011-04-25 VNIIFTRI CRM Jones type cell ?

NIST USA 2011-04-05 Primary NIST CRM Daggett type cell 10 - 10 000

Table 6 Results for sample with nominal value of electrolytic conductivity 0.05 S·m-1

0,05 S/m

Institute Country

EC u (k=1)

INRIM ITA 0,049855 0,000035

INMETRO BRA 0,049970 0,000048

CMI CZE 0,049980 0,000049

INPL ISR 0,049980 0,000038

MKEH HUN 0,049988 0,000013

GUM POL 0,050000 0,000030

SMU SVK 0,050018 0,000016

VNIIFTRI RUS 0,050021 0,000012

DFM DNK 0,050026 0,000019

NIST USA 0,050036 0,000024

NIM CHN 0,050044 0,000014

PTB DEU 0,050051 0,000012

CENAM MEX 0,050100 0,000080

SP SWE 0,050110 0,000075

UMTS UKR 0,050116 0,000016

Figure 7 The plot of results for sample with nominal conductivity value 0.05 S·m-1

(standard uncertainty given)

CCQM-K92 12

CCQM-K92 (Electrolytic Conductivity) - Nominal Value 0.05 S/m

ITA

BR

A

ISR

CZ

E

HU

N

PO

L

SV

K

RU

S (

VN

IIFT

RI)

DN

K

US

A

CH

N

DE

U

ME

X

SW

E

UK

R

0,0497

0,0498

0,0499

0,0500

0,0501

0,0502

0,0503

Ele

ctro

lyti

c C

on

du

ctiv

ity

/ S

·m-1

Sample with nominal value 20 S·m-1

The measurement conditions and methods used at different institutes are given in Table 7.The results of measurements of the sample with nominal conductivity of 20 S·m-1 are given inTable 8 and displayed graphically in Figure 8. Three laboratories from 10 used a primarymeasurement procedure.

Table 7 Conditions of Measurement at Various Institutes - 20 S·m-1

Institute CountryDate of regis-

tered mailTraceability Measurement frequency [Hz]

CENAM MEX 2011-04-25 SMU CRM Jones type cell 1000 - 10000

GUM POL 2011-04-22 DFM CRM Jones type cell 12 - 1820

INPL ISR 2011-04-06 SMU CRM Four electrode DC cell DC method

DFM DNK 2011-04-13 Primary CRM DFM Jones type cell 3000 - 9000

VNIIM RUS 2011-04-13 IUPAC Jones type cell 1000

NIM CHN 2011-04-21 IUPAC Jones type cell 20 - 5000

MKEH HUN 2011-04-19 OIML R56 Four electrode DC cell DC method

PTB DEU 2011-04-06Primary


UMTS UKR 2011-04-27 Primary CRM UMTS Jones type cell 1000

INMETRO BRA 2011-04-20Primary


CCQM-K92 13

Table 8 Results for sample with nominal value of electrolytic conductivity 20 S·m-1

20 S/m

Institute Country

EC u (k=1)

CENAM MEX 20,123 0,025

GUM POL 20,279 0,011

INPL ISR 20,280 0,014

DFM DNK 20,283 0,007

VNIIM RUS 20,290 0,007

NIM CHN 20,300 0,006

MKEH HUN 20,306 0,005

PTB DEU 20,308 0,007

UMTS UKR 20,344 0,004

INMETRO BRA 20,412 0,423

Figure 8 The plot of results for sample with nominal conductivity value 20 S·m-1

(standard uncertainty given)

CCQM-K92 (Electrolytic Conductivity) - Nominal Value 20 S/m

BR

A

UK

R

DE

U

HU

N

CH

N

RU

S (

VN

IIM

)

DN

K

ISR

PO

L

ME

X

20,10

20,15

20,20

20,25

20,30

20,35

20,40

20,45

Ele

ctro

lyti

c C

on

du

ctiv

ity

/ S

·m-1

CCQM-K92 14

∑= iiWM xwκ

∑

= 2

2

1

1

i

ii

u

uw

Discussion

Nominal values of electrolytic conductivity of the samples were 0.05 S·m-1 and 20 S·m-1. 15laboratories took part in the measurement of the 0.05 S·m-1 sample. Sample with nominalvalue of electrolytic conductivity 20 S·m-1 was measured by 10 laboratories.

The Formulas for Calculation of Estimators

Arithmetic Mean (2)

Standard Deviation of Arithmetic Mean (3)

Weighted Mean (4)

Calculation of Weights for Weighted Mean (5)

Uncertainty of Weighted Mean (6)

Uncertainty of Median (7)

Calculation of the Degrees of Equivalence

The degrees of equivalence for each participant, Di, and its standard uncertainty, u(Di), aregiven by Eq 8 and Eq 9, respectively:

( )KCRV−= iiD κ (8)

( )( )1

2

−−

= ∑nn

u iAM

κκ

ni∑=

κκ

( )κκ ~1

858,1 −−

= imediann

u

( )( )( )∑∑

−−

=i

Rii

wn

xxwu

1

2

WM

CCQM-K92 15

22KCRVD uuu

ii+= κ (9)

The standardized deviations En are given by formula (10)

(10)

Communication with the participants

DFM and NIST highlighted the instability of the first (0.05 S·m-1) sampleINRIM and UMTS were requested to check their results; no numerical error was found, but

new results were sent in after recalibration of the cells. These new data are given indiscussion only, as the original data have to be used.

VNIIM was asked to correct the result for bottle weighing; corrected mass was sent.VNIIFTRI was requested to check its results; VNIIFTRI “has found a mistake by transfer of

value constant from primary to secondary cell“ and provided new results.CMI requested to postpone the deadline for internal reasons.

Reference Value for 0.05 S·m-1

Basic statistical data.

Arithmetic mean 0.050020 S·m-1

Standard uncertainty 0.000017 S·m-1

Median 0.050021 S·m-1


The consistency information for different datasets is given in table 9.

Table 9 Chi square values for nominal value - 0.05 S·m-1 (values in bold are criticalvalues)

22KCRV

in

UU

KCRVE

i+

−=

κ

κ

CCQM-K92 16

Participants

Degrees

of free-

dom

χχχχ2 - cal-

culated

χχχχ2222 − − − − tabu-

lated

95% prob-

ability

Weighted

mean

Stand. uncer-

tainty

(Birge treat-

ment)

Birge Ratio

All 14 78,437 23,685 0,0500284 0,0000115 2,37 1,30

ITA excluded 13 52,656 22,362 0,0500319 0,0000099 2,01 1,31

ITA, UKR excluded 12 22,171 21,026 0,0500231 0,0000070 1,36 1,32

ITA, UKR, HUN excluded 11 12,627 19,675 0,0500304 0,0000061 1,07 1,34

Numbers given in bold are tabulated data for 95 % probability and corresponding degrees offreedom. If ITA, UKR, HUN are not taken into account, the chi-squares value is smaller thanthe critical value. As a reference value KCRV = 0.0500304 S·m-1 (u = 0.0000061 S·m-1) wasagreed.

Degrees of Equivalence - 0.05 S·m-1

Table 10 Degrees of Equivalence - 0.05 S·m-1

Di UDi

Institute Country

S/m S/m

INRIM ITA -0,000175 0,000070

INMETRO BRA -0,000060 0,000096

CMI CZE -0,000050 0,000099

INPL ISR -0,000050 0,000077

MKEH HUN -0,000042 0,000028

GUM POL -0,000030 0,000061

SMU SVK -0,000012 0,000034

VNIIFTRI RUS -0,000009 0,000027

DFM DNK -0,000004 0,000039

NIST USA 0,000006 0,000049

NIM CHN 0,000014 0,000031

PTB DEU 0,000021 0,000027

CENAM MEX 0,000070 0,000160

SP SWE 0,000080 0,000150

UMTS UKR 0,000086 0,000034

Figure 9 Degrees of Equivalence for sample with nominal conductivity value 0.05 S·m-1

CCQM-K92 17

CCQM-K92 Degrees of Equivalence - 0.05 S/m

ITA

BR

A

ISR

CZ

E

HU

N

PO

L

SV

K

RU

S (

VN

IIFT

RI)

DN

K

US

A

CH

N

DE

U

ME

X

SW

E

UK

R

-0,0003

-0,0002

-0,0001

0,0000

0,0001

0,0002

0,0003

De

gre

es

of

Eq

uiv

ale

nce

Reference Value for 20 S·m-1

Basic statistical data.

Arithmetic mean 20.2924 S·m-1


Median 20.2950 S·m-1


The consistency information for different datasets is given in table 11.

Table 11 Chi square values for nominal value – 20 S·m-1 (values in bold are criticalvalues)

Participants

Degrees

of free-

dom

χχχχ2 - cal-

culated

χχχχ2222 − − − − tabulated

95% prob-

ability

Weighted

mean

(S.m-1

)

Stand. un-

certainty

(Birge

treatment)

Birge Ratio

All 9 192,287 16,919 20,3120 0,0096 4,62 1,37

UKR excluded 8 63,101 15,507 20,2945 0,0073 2,81 1,39

UKR, MEX excluded 7 15,528 14,067 20,2964 0,0039 1,49 1,42

CCQM-K92 18

Numbers given in bold are tabulated (critical) data for 95 % probability and correspondingdegrees of freedom. If MEX, UKR are not taken into account, the chi-squares value is stillslightly larger than the critical value. However, any further exclusion does not improve thechi-square value. As a reference value KCRV = 20.2964 S·m-1 (u = 0.0039 S·m-1) wasagreed.

Degrees of Equivalence - 20 S·m-1

Table 12 Degrees of Equivalence – 20 S·m-1

Di UDiInstitute Country

S/m S/m

CENAM MEX -0,173 0,051

GUM POL -0,017 0,023

INPL ISR -0,016 0,029

DFM DNK -0,013 0,015

VNIIM RUS -0,006 0,015

NIM CHN 0,004 0,014

MKEH HUN 0,009 0,013

PTB DEU 0,012 0,016

UMTS UKR 0,048 0,010

INMETRO BRA 0,115 0,845

CCQM-K92 Degrees of Equivalence - 20 S/m

ME

X

PO

L

ISR

DN

K

RU

S (

VN

IIM

)

CH

N

HU

N

DE

U

UK

R

BR

A

-0,20

-0,15

-0,10

-0,05

0,00

0,05

0,10

0,15

0,20

De

gre

es

of

Eq

uiv

ale

nce

Figure 8 Degrees of Equivalence for sample with nominal conductivity value 20S·m-1

CCQM-K92 19

Posterior work reported by DFM:„DFM has since submitting K92 results and Draft A, examined the detailed frequency beha-viour of its conductivity cells (cf. presentation EAWG/12-07 at CCQM-EAWG, Paris, April2012). On this basis, we have shifted the frequency range used for extrapolation towards hig-her frequencies (towards lower phase angles) for measurements of high conductivity in thesecondary cell, which provided traceability to the cell used at 20 S/m. The cell constant of thecell used in K92 at 20 S/m has thus been reevaluated and would today give a result with DoEof -0.004 ± 0.015.“

Conclusions

Sixteen laboratories took part in the comparison, 15 measured sample with nominal conduc-tivity of 0,05 S/m and 10 laboratories measured sample with nominal conductivity of 20 S/m.A good agreement of the results was observed for most laboratories. In some cases excessiveuncertainty was due to measurement of very small resistance (less than 0,1 Ω) due to low cellconstant of the conductivity cells used.In the case of measurements of sample with conductivity ~20 S/m in secondary cells therecould be some problems with calibration, as CRMs with such a high conductivity are notcommonly available. These measurements thus can be classified as “extrapolation measure-ments”.

How far the light shines statement:The results in this comparison on the 0.05 S/m solution can be considered to be representativefor measurement capabilities in the range from 0.016 S/m to 0.15 S/m. The results in thiscomparison on the 20 S/m solution can be considered to be representative for measurementcapabilities in the range from 6 S/m to 25 S/m. Due to the increased difficulty associated withperforming the measurement as the conductivity increases, for the range from 25 S/m to 60S/m the results from this comparison are to be complemented in the review process by a de-tailed technical evaluation of the measurement procedure.

AcknowledgmentThe coordinating laboratory gratefully acknowledges the contributions of all participants andof the members of the CCQM Working Group on Electrochemical Analysis for their valuablesuggestions concerning the measurement protocol and the evaluation process.

References

1. Pure Appl. Chem., Vol. 73, No. 11, pp. 1783–1793, 2001.2. Final Report for Key Comparison CCQM-K36, available at

http://www.bipm.org/utils/common/pdf/final_reports/QM/K36/CCQM-K36.pdf3. Final Report for Key Comparison CCQM-K36.1, available at

http://www.bipm.org/utils/common/pdf/final_reports/QM/K36/CCQM-K36.1.pdf

CCQM-K92 20

Appendix

Addresses of participants

Adrian Reyes,Marcela MonroyCENAMKm 4.5 Carretera a los Cues, Mpio. El Marques, Queretaro, Mexico.CP 76246MEXICO

Alena VospělováČeský metrologický ústav (ČMI)Okružní 31638 00 BrnoCZECH REPUBLIC

Bertil MagnussonSP Technical Research Institute of SwedenBrinellgatan 4501 15 BoråsSWEDEN

Elena KardashThe National Physical Laboratory of Israel (INPL)Danciger “A” bldg., Givat-Ram, Jerusalem, 91904,ISRAEL

Fabiano Barbieri GonzagaNational Institute of Metrology, Standardization and Industrial Quality - INMETROAv. Nossa Senhora das Graças, 50, XerémDuque de Caxias, RJ, Brazil25250-020BRAZIL

Francesca DurbianoInstituto Nazionale di Ricerca Metrologica (INRiM)Strada delle Cacce 91, CAP 10135, Torino,ITALY

Kenneth W. PrattNational Institute of Standards and TechnologyBuilding 227, Room B346100 Bureau Dr., Stop 8391Gaithersburg, MD 20899-8391U.S.A.

Leoš Vyskočil

CCQM-K92 21

Slovak Institute of MetrologyKarloveska 63, SK-842 55 BratislavaSLOVAKIA

Pia Tønnes JakobsenDanish Fundamental Metrology A/S (DFM)Matematiktorvet 307,DK-2800 Kgs. LyngbyDENMARK

Prof. L.A. KonopelkoD.I. Mendeleyev Institute for Metrology (VNIIM)19, Moskovsky pr., St. Petersburg, 190005RUSSIA

Song XiaopingNational Institute of Metrology (NIM)No. 18, Bei San Huan Dong Lu, Beijing 100013,P.R.CHINA

Steffen SeitzPhysikalisch-Technische Bundesanstalt (PTB)Fachbereich 3.1Bundesallee 10038116 BraunschweigGERMANY

Vladimir GavrilkinUkrmetrteststandartMetrologichna str.,4 Kiev, 03680UKRAINA

Wladyslaw KozlowskiCentral Office of Measures (GUM)Laboratory of ElectrochemistryElektoralna Str. 200-139 Warsaw,POLAND

Yury A. OvchinnikovVNIIFTRI (National Research Institute Physicotechnical and Radio Engineering Measurements)141570, Mendeleevo, Moscow region, RussiaVNIIFTRI, Lab 35RUSSIA

Report of key comparison CCQM – K92

Documents

Transcript of Report of key comparison CCQM – K92