ISO 16107:2007
(Main)Workplace atmospheres — Protocol for evaluating the performance of diffusive samplers
Workplace atmospheres — Protocol for evaluating the performance of diffusive samplers
ISO 16107:2007 specifies methods for evaluation of sampler performance in terms of workplace conditions: wind speed, humidity, temperature, atmospheric pressure, and analyte variation. The concise set of experiments specified aims to minimize cost to the user. The evaluation is limited to conditions commonly encountered in personal sampling in the indoor workplace setting, namely wind speeds of up to 0,5 m/s and for sampling periods typically from 2 h to 8 h. Static or area sampling, unlike personal sampling where movement of the subject is significant, may sometimes be subject to sampling-rate reduction due to stagnation at very low wind speeds. ISO 16107:2007 therefore does not apply to wind speeds of less than 0,1 m/s relative to static samplers. Samplers are also tested for compliance with the manufacturer's stated limits on capacity, possibly in the presence of interfering compounds. Given a suitable exposure chamber, the sampler evaluation protocol can be extended to cover sampler use for other sampling periods and conditions. ISO 16107:2007 indicates how to measure diffusive sampler uncertainty for characterizing concentration estimates obtained subsequent to the evaluation. It is impractical continually to re-evaluate diffusive sampler performance under various environmental conditions prevailing during application.
Air des lieux de travail — Protocole pour l'évaluation de la performance des dispositifs de prélèvement par diffusion
Zrak na delovnem mestu - Protokol za vrednotenje lastnosti difuzijskih vzorčevalnikov
Ta mednarodni standard določa metode za vrednotenje lastnosti vzorčevalnikov na področju pogojev na delovnem mestu: hitrost vetra, vlažnost, temperatura, zračni tlak in variacija analita. Zgoščeni niz specificiranih preskusov je namenjen za zmanjšanje stroškov uporabnika. Vrednotenje je omejeno na pogoje, ki so pogosti pri osebnem vzorčenju v okolju notranjih delovnih mest, zlasti pri hitrosti vetra do 0,5 m/s in običajnih obdobjih vzorčenja od 2 do 8 h. Pri statičnem vzorčenju ali vzorčenju površin se lahko v nasprotju z osebnim vzorčenjem, pri katerem je pomembno gibanje predmeta, frekvenca vzorčenja zmanjša zaradi mirovanja pri zelo nizkih hitrostih vetra. Ta mednarodni standard se zato ne uporablja za hitrosti vetra, manjše od 0,1 m/s, za katere se uporabljajo statični vzorčevalniki. Preskuša se tudi skladnost vzorčevalnikov z omejitvami glede kapacitete, ki jih je navedel proizvajalec, če je mogoče v prisotnosti motečih komponent. Če je na voljo ustrezna komora za izpostavljanje, se lahko protokol za vrednotenje vzorčevalnikov razširi na uporabo vzorčevalnikov v drugih obdobjih in pogojih vzorčenja.
General Information
Relations
Buy Standard
Standards Content (Sample)
INTERNATIONAL ISO
STANDARD 16107
Second edition
2007-10-15
Workplace atmospheres — Protocol for
evaluating the performance of diffusive
samplers
Air des lieux de travail — Protocole pour l'évaluation de la performance
des dispositifs de prélèvement par diffusion
Reference number
ISO 16107:2007(E)
©
ISO 2007
---------------------- Page: 1 ----------------------
ISO 16107:2007(E)
PDF disclaimer
This PDF file may contain embedded typefaces. In accordance with Adobe's licensing policy, this file may be printed or viewed but
shall not be edited unless the typefaces which are embedded are licensed to and installed on the computer performing the editing. In
downloading this file, parties accept therein the responsibility of not infringing Adobe's licensing policy. The ISO Central Secretariat
accepts no liability in this area.
Adobe is a trademark of Adobe Systems Incorporated.
Details of the software products used to create this PDF file can be found in the General Info relative to the file; the PDF-creation
parameters were optimized for printing. Every care has been taken to ensure that the file is suitable for use by ISO member bodies. In
the unlikely event that a problem relating to it is found, please inform the Central Secretariat at the address given below.
COPYRIGHT PROTECTED DOCUMENT
© ISO 2007
All rights reserved. Unless otherwise specified, no part of this publication may be reproduced or utilized in any form or by any means,
electronic or mechanical, including photocopying and microfilm, without permission in writing from either ISO at the address below or
ISO's member body in the country of the requester.
ISO copyright office
Case postale 56 • CH-1211 Geneva 20
Tel. + 41 22 749 01 11
Fax + 41 22 749 09 47
E-mail copyright@iso.org
Web www.iso.org
Published in Switzerland
ii © ISO 2007 – All rights reserved
---------------------- Page: 2 ----------------------
ISO 16107:2007(E)
Contents Page
Foreword. iv
Introduction . v
1 Scope . 1
2 Normative references . 1
3 Terms and definitions. 2
4 Symbols and abbreviated terms . 2
5 Summary of test protocol . 3
5.1 Factors affecting performance . 3
5.2 “Accuracy” as providing a single performance value linking to measurement uncertainty. 4
5.3 Bias, intersampler variability and the effects of environmental uncertainty. 4
5.4 Reverse diffusion. 5
5.5 Capacity — Control of effects from interfering compounds. 6
5.6 Capacity overload detection. 6
5.7 Desorption efficiency . 7
5.8 Atmospheric pressure. 7
5.9 Wind direction . 7
5.10 Simplifications . 7
6 Apparatus . 7
6.1 Exposure chamber specifications . 7
6.2 Controlled environmental conditions. 8
6.3 Inter-run variability . 8
7 Reagents and materials . 9
8 Procedure . 9
9 Sampler performance classification. 9
10 Accuracy. 10
10.1 General. 10
10.2 Accuracy range confidence limit . 10
10.3 Expanded uncertainty . 11
11 Test report . 11
Annex A (informative) Worked example — Program for diffusive sampler accuracy range
calculation . 13
Annex B (informative) Calculation of variances. 15
Bibliography . 16
© ISO 2007 – All rights reserved iii
---------------------- Page: 3 ----------------------
ISO 16107:2007(E)
Foreword
ISO (the International Organization for Standardization) is a worldwide federation of national standards bodies
(ISO member bodies). The work of preparing International Standards is normally carried out through ISO
technical committees. Each member body interested in a subject for which a technical committee has been
established has the right to be represented on that committee. International organizations, governmental and
non-governmental, in liaison with ISO, also take part in the work. ISO collaborates closely with the
International Electrotechnical Commission (IEC) on all matters of electrotechnical standardization.
International Standards are drafted in accordance with the rules given in the ISO/IEC Directives, Part 2.
The main task of technical committees is to prepare International Standards. Draft International Standards
adopted by the technical committees are circulated to the member bodies for voting. Publication as an
International Standard requires approval by at least 75 % of the member bodies casting a vote.
Attention is drawn to the possibility that some of the elements of this document may be the subject of patent
rights. ISO shall not be held responsible for identifying any or all such patent rights.
ISO 16107 was prepared by Technical Committee ISO/TC 146, Air quality, Subcommittee SC 2, Workplace
atmospheres.
This second edition cancels and replaces the first edition (ISO 16107:1999), which has been technically
revised.
iv © ISO 2007 – All rights reserved
---------------------- Page: 4 ----------------------
ISO 16107:2007(E)
Introduction
Gas or vapor sampling is often accomplished by actively pumping air through a collection medium such as
activated charcoal. Problems associated with a pump, such as inconvenience, inaccuracy and expense, are
inextricable from this type of sampling. The alternative covered by this International Standard is to use
diffusion for moving the compound of interest onto the collection medium. This approach to sampling is
attractive because of the convenience of use and low total monitoring cost.
However, previous studies have found significant problems with the accuracy of some samplers. Therefore,
although diffusive samplers may provide a plethora of data, inaccuracies and misuse of diffusive samplers
may yet affect research studies. Furthermore, worker protection may be based on faulty assumptions. The
aim of this International Standard is to counter the uncertainties in diffusive sampling through achieving a
broadly accepted set of performance tests and acceptance criteria for proving the efficacy of any given
diffusive sampler intended for use.
This International Standard is intended specifically for the large-scale evaluation of many diffusive
sampler/analyte pairs of practical application and is complementary to EN 838. An affordable, experimental
evaluation determines a single performance value indicating how a sampler performs in a typical situation. A
sampler can thereby be quickly judged as to acceptability. Additionally, sufficient data are obtained to predict
performance in many atypical situations. For example, although sampling may normally be done at room
temperature, a particular need may call for use in extreme cold. In such a case, the single performance value
would be superseded by the particular needs.
© ISO 2007 – All rights reserved v
---------------------- Page: 5 ----------------------
INTERNATIONAL STANDARD ISO 16107:2007(E)
Workplace atmospheres — Protocol for evaluating the
performance of diffusive samplers
1 Scope
This International Standard specifies methods for evaluation of sampler performance in terms of workplace
conditions: wind speed, humidity, temperature, atmospheric pressure, and analyte variation. The concise set
of experiments specified aims to minimize cost to the user. The evaluation is limited to conditions commonly
encountered in personal sampling in the indoor workplace setting, namely wind speeds of up to 0,5 m/s and
for sampling periods typically from 2 h to 8 h.
Static or area sampling, unlike personal sampling where movement of the subject is significant, may
sometimes be subject to sampling-rate reduction due to stagnation at very low wind speeds. This International
Standard therefore does not apply to wind speeds of less than 0,1 m/s relative to static samplers. Samplers
are also tested for compliance with the manufacturer's stated limits on capacity, possibly in the presence of
interfering compounds. Given a suitable exposure chamber, the sampler evaluation protocol can be extended
to cover sampler use for other sampling periods and conditions.
This International Standard indicates how to measure diffusive sampler uncertainty for characterizing
concentration estimates obtained subsequent to the evaluation. It is impractical continually to re-evaluate
diffusive sampler performance under various environmental conditions prevailing during application.
NOTE 1 In this International Standard, the confidence level for the initial method evaluation becomes an integral part of
the measurement uncertainty. This approach slightly broadens the statistical protocols given in ISO Guide 98:1995.
Furthermore, the possibility of sampler errors related to correctible sampler bias is addressed.
NOTE 2 This International Standard is an extension of previous research on diffusive samplers (References [1] to [17]
inclusive and EN 838).
2 Normative references
The following referenced documents are indispensable for the application of this document. For dated
references, only the edition cited applies. For undated references, the latest edition of the referenced
document (including any amendments) applies.
EN 838, Workplace atmospheres — Diffusive samplers for the determination of gases and vapours —
Requirements and test methods
ISO Guide 98:1995, Guide to the expression of uncertainty in measurement. BIPM, IEC, IFCC, ISO, IUPAC,
IUPAP, OIML
© ISO 2007 – All rights reserved 1
---------------------- Page: 6 ----------------------
ISO 16107:2007(E)
3 Terms and definitions
For the purposes of this document, the terms and definitions given in EN 838 and ISO Guide 98:1995 and the
following apply.
3.1
symmetric accuracy range
A
fractional range about the measurand concentration, c, within which 95 % of sampler measurements are found
NOTE See References [18] to [21] inclusive.
If the modulus of the bias is small, i.e. │∆│ < R/1,645, the symmetric accuracy range, A, can be shown to be closely
approximated (Reference [21]) by Equation (1):
22
AR =×1,960 ∆+ (1)
where
∆ is the bias, expressed relative to true concentrations;
R is the overall true relative standard deviation, expressed relative to true concentrations.
Otherwise
AR =+|∆| 1,645× (2)
2
If the bias is corrected, the expected value of ∆ in Equation (1) becomes equal to the variance of the bias correction,
reflecting an uncorrectable residual bias due to uncertainty in the correction. Then if the bias uncertainty is under control,
the low bias modulus model of Equation (1) indicates the proportionality of A to the root mean squared combined
uncertainty components.
4 Symbols and abbreviated terms
A symmetric accuracy range (3.1) in terms of bias and precision
A estimated symmetric accuracy range, A
est
A 95 % confidence level on the symmetric accuracy range, A
95 %
c true or reference analyte concentration, in milligrams per cubic metre, or parts per million as a ratio of
analyte to air molecules, per instructions of the sampler manufacturer
c mean of (four) concentration estimates (including pressure and temperature corrections), in milligrams
est
per cubic metre or parts per million as a ratio of analyte to air molecules, per instructions of the
sampler manufacturer
h humidity, partial pressure of water vapor, in kilopascals
n number of diffusive samplers tested for measuring sampler capacity
p (atmospheric) pressure
R overall true relative standard deviation of concentration estimates (dependent on assumed
environmental variability), expressed as a percentage relative to a “true” concentration as estimated
by reference sampling
2 © ISO 2007 – All rights reserved
---------------------- Page: 7 ----------------------
ISO 16107:2007(E)
R estimated true relative standard deviation, expressed as a percentage
est
R true relative standard deviation characterizing inter-run chamber variability, expressed as a
run
percentage
R intersampler component of the true relative standard deviation, expressed as a percentage
s
R estimated intersampler true relative standard deviation, R , expressed as a percentage
s est s
R pulse-induced true relative standard deviation, expressed as a percentage
t
R 95 % confidence limit on the true relative standard deviation, expressed as a percentage
95 %
s estimated standard deviation characterizing intersampler variation
t (ν) value which, at the 95 % probability level, exceeds random variables distributed according to the
0,95
Student t-distribution with ν degrees of freedom
T temperature, in degrees Celsius
u ambient wind speed, in metres per second
α concentration estimate dependence on environmental variable, x (T, h, u, or c)
x
∆ bias relative to concentration, c
∆ estimated bias, ∆
est
∆ bias associated with concentration pulse
t
∆ 95 % confidence limit on the bias, ∆
95 %
ν degrees of freedom in determining R
s
ν effective number of degrees of freedom in determining R
eff
σ assumed concentration variability
c
σ assumed humidity variability
h
σ assumed temperature variability
T
σ assumed ambient wind speed variability
u
5 Summary of test protocol
5.1 Factors affecting performance
5.1.1 Diffusive sampling may first of all suffer from error in the sampling rate as stated by the manufacturer
of the sampler. As diffusive samplers are usually used without recalibration, this error implies a bias or
systematic error in all concentration estimates made. As the bias may be in one direction, such error cannot
be minimized by averaging several measurements. If the error is correctible it is not strictly part of the sampler
uncertainty.
© ISO 2007 – All rights reserved 3
---------------------- Page: 8 ----------------------
ISO 16107:2007(E)
5.1.2 Aside from bias, several sources of random error exist. Some of these errors are associated with the
sampler itself. For example, the sampler’s dimensional tolerance may be low, leading to variation in the
sampling rate, sampler to sampler. The sampler may become overloaded or may not follow concentration
changes adequately.
5.1.3 Other errors relate to a non-constant environment which may affect the sampling rate. Potentially
significant environmental influence quantities include temperature, humidity, wind speed, and atmospheric
pressure. To predict the effect of any such parameter, an experiment on the sampler type may be carried out.
Moreover, the variation expected of the parameter requires estimation. So as to result in a single performance
measure, typical ranges of these environmental parameters are specified in this International Standard.
5.2 “Accuracy” as providing a single performance value linking to measurement
uncertainty
5.2.1 A sampler presented by a manufacturer for evaluation may be found to exhibit both imprecision and
bias. In the language of ISO Guide 98:1995, imprecision and bias together relate to the “accuracy” of a
sampler, rather than “uncertainty”. Accuracy refers to the closeness between measured values and
measurand (in this case the concentration to be measured), whereas uncertainty refers to what is known
about the measurand on the basis of the measurement. In other words, the manufacturer could recommend
use of information about the bias for correction of the measured values. This International Standard provides
means of judging a sampler as recommended for use by a manufacturer.
5.2.2 A useful statistical measure exists which summarizes the overall effect of both random variation and
bias. This function has come to be known as the symmetric accuracy range, A (3.1).
5.2.3 The definition directly implies that A increases with both random error and bias magnitude. Therefore
A is one particular quantification of accuracy as defined in ISO Guide 98:1995. The greater the disparity
between measured values and measurand, the larger the value of A.
NOTE A is not the only way of quantifying accuracy. For example, the selection of 95 % as coverage level is arbitrary.
Furthermore, other functions consistent with the qualitative idea of accuracy as described in ISO Guide 98:1995 are
possible.
5.2.4 A indicates quantitatively how far off measurements would generally be if further correction of the
sampler were not made. A further advantage of A is that a direct link to the uncertainty described in
ISO Guide 98:1995 is obtained in the case that bias is minimized, e.g. through results of the evaluation.
5.2.5 Moreover, a measurement element not covered extensively within ISO Guide 98:1995 is dealt with
explicitly. Namely, diffusive samplers are generally used as obtained from the manufacturer without
re-evaluation at each use. Therefore, uncertainty in a measured value must account for confidence in the
[single] evaluation itself as well as variation and possible residual bias at the point of use.
5.2.6 The result of approaching uncertainty from the point of view of accuracy results in an “expanded
uncertainty” (in the language of ISO Guide 98:1995) proportional via a “coverage factor” to the root mean
squared combined uncertainty components comprised of bias uncertainty, the effects of sampler variation,
and influence quantities described above. The coverage factor is determined by the confidence required [e.g.,
95 %] in the sampler evaluation and the coverage [also 95 %]. The final result is a value of expanded
uncertainty with specific [tolerance or predictive] meaning, namely, in providing the 95 % coverage range
about the [true] measurand at 95 % confidence in the evaluation.
5.3 Bias, intersampler variability and the effects of environmental uncertainty
5.3.1 This International Standard gives a procedure for assessing the effects of variability in the following
workplace variables: temperature, T; humidity, h (expressed in terms of the water vapor partial pressure to
minimize interaction with the temperature); ambient wind speed, u, across the sampler face [see (5.7)
regarding wind direction]; and concentration, c. An experiment is carried out which provides information about
the dependencies of the concentration estimates on these variables near conditions of intended sampler use
(T , h , u , and c ). Testing is required at the concentration, c , of intended use, as well as at concentrations
0 0 0 0 0
reduced at least to c /2. Furthermore, the sampler bias and the intersampler standard deviation are measured.
0
4 © ISO 2007 – All rights reserved
---------------------- Page: 9 ----------------------
ISO 16107:2007(E)
Finally, the effect of diffusion of material out of the sampler is measured. Pressure effects result in correctable
bias and are not evaluated in this International Standard (5.8).
5.3.2 Using four samplers for each of five experimental runs (the minimum possible), the dependencies α ,
T
α , α , and α (relative to the chamber reference concentration and target environmental parameters) on
h u c
changes in T, h, u, and c are measured, following the sampler manufacturer's instructions regarding pressure
and temperature corrections (if any). These experiments also give a value for the estimated sampler bias, ∆,
relative to the chamber reference concentration (defined for the target conditions). Two further runs describing
time-dependent effects (5.4.5) from diffusive loss of analyte are also carried out. The chamber reference
concentration must be traceable to primary standards of mass and volume.
5.3.3 Error in the estimates of the dependencies α , α , α , and α will exist on account of the intersampler
T h u c
component of the true relative standard deviation R and an inter-run chamber true relative standard deviation,
s
R . The latter results in part from uncertainty in the reference concentration. R is obtained by pooling the
run s
variance estimates from each run and therefore is estimated with 7 × 3 = 21 degrees of freedom (or
15 degrees of freedom if the reverse diffusion experiment is omitted). So as to avoid re-measurement at each
sampler/analyte evaluation, R is obtained by a separate characterization of the chamber with several runs
run
at (for example) fixed environmental conditions. An example in which the dependencies α and R are
s
estimated is presented in Annex A.
NOTE It is up to the user as to how traceability is established. In Reference [12], the concentration estimate, as
calculated from the chamber's analyte generation parameters, is regarded as the “benchmark”, although an independent
estimate is required and must be within 5 % of the calculated estimate. If these estimates differ, then a third independent
estimate is required to establish the reference concentration through agreement with one of the other independent
estimates. One possibility for such an independent estimate is the mean of at least five independent, active sampler
estimates per run within the chamber. Experiment (Reference [12]) on the accuracy of such reference measurements
using sorbent tubes indicates that a true relative standard deviation of the order of 2 % can be achieved for the individual
measurements. Alternatively, Reference [3] requires averaging of at least two independent methods (possibly including
calculated estimates) with at least four samples per method. EN 838 has adopted the looser requirement that calculated
and independent measurements shall agree within 10 %.
5.3.4 A further consolidation of tests may be made by observing that the dependence of concentration
estimates on the wind speed, v, is only sampler specific, i.e. does not depend on the specific analyte.
Therefore, after a single measurement for a given sampler type, the set of tests can be narrowed to five runs
with 5 × 3 = 15 degrees of freedom in the estimate of R .
s
5.4 Reverse diffusion
5.4.1 A potential problem with diffusive samplers is presented by the possibility of reverse diffusion
(sometimes denoted as “back-diffusion” or “off-gassing”) of the analyte. Reverse diffusion is generally only
significant in the case that an analyte is only weakly bound to the sorbent (Reference [6]). Therefore,
inaccuracy associated with these effects may usually be minimized through proper sorbent selection.
5.4.2 Because of reverse diffusion, estimates of a varying concentration may in some cases be biased. The
worst-case situation occurs with the concentration in the form of an isolated pulse at either the beginning or
end of the sampling period. A pulse at the beginning of the period allows the entire sampling period (4 h
to 12 h) for sample loss, possibly resulting in a low estimate relative to a pulse at the end.
5.4.3 In some cases, the time dependence of a specific workplace concentration correlates strongly with the
sampling period. For example, a cleanup operation at the end of a workday could introduce analyte only then.
This could imply a positive bias in the concentration estimates obtained from a day's sampling. For simplicity,
however, this International Standard is designed to assess performance of samplers for use in a concentration
fluctuating in a stationary manner, so that time-dependent effects are treated simply as components of
sampler variance. Specifically, the effect of an isolated 0,5 h pulse occurring at random within the sampling
period is estimated.
5.4.4 Challenging samplers to 0,5 h pulses is similar to the tests of NIOSH (Reference [3]) and EN 838.
© ISO 2007 – All rights reserved 5
---------------------- Page: 10 ----------------------
ISO 16107:2007(E)
5.4.5 Let ∆ (>0) represent one-half the bias between estimates from a 0,5 h pulse at the end versus the
t
beginning of the sampling period, relative to the mean of the estimates. Assume, conservatively (see, for
example, Reference [6]), that the bias in the estimates of 0,5 h pulse occurring at random within (for example,
2
an 8 h sampling period ranges uniformly between −∆ and +∆. Then the variance, R , associated with
t
t t
sampling a 0,5 h pulse at random within the sampling period is given by Equation (3):
22
1
R = ∆ (3)
tt
3
5.5 Capacity — Control of effects from interfering compounds
5.5.1 This International Standard provides a test for confirming a manufacturer's claimed sampler capacity
under stated conditions of use. Such conditions would normally refer to a specific sampling period and to
environmental extremes, such as 80 % relative humidity at a temperature of 30 °C. Additionally, a
manufacturer may claim a value of capacity for sampling in the presence of specific interferences at stated
concentrations.
5.5.2 For the purposes of this International Standard, capacity is defined as the sampled mass (or
equivalently as the concentration for a specific sampling period) at which concentration estimates are 10 %
low. Specifically, capacity is considered not exceeded if concentration estimates, corrected for correctable
bias, are above 90 % of the true concentration at the 95 % confidence level.
5.5.3 An example of the test is as follows: eight diffusive and eight active samplers are exposed to the
analyte of interest under the stated environmental conditions. Then, neglecting variability in the reference
sampler mean, the 95 % confidence limit ∆µ on the difference in the (unknown) mean concentration
95 %
estimates is given by Equation (4):
∆µ =−∆cs[×t(ν)/ n] (4)
95% 0,95
where
∆c is the estimated mean difference between diffusive and active results;
s is the estimated standard deviation characterizing intersampler variation, calculated from the eight
diffusive sampler estimates;
n is the number of samplers of each type (in this case, 8);
ν is the number of degrees of freedom (n − 1 = 7).
Then ∆µ shall be greater than −10 % c, where c is the mean concentration estimate from the reference
95 %
samplers.
EXAMPLE Suppose the diffusive sampler component of the true relative standard deviation, R = 5 %:
s
(/sc)×=t (ν)/ n 3.3 (5)
0,95
Therefore, in this case the mean value of the diffusive results shall be greater than 93,3 % of the reference concentration.
NOTE As capacity strongly correlates with sampled mass, a capacity limit expressed as sampled mass at one stated
sampling period is generally applicable to a range of sampling periods.
5.6 Capacity overload detection
The capability to detect capacity overload (e.g. by the use of a second sorbent or by employing paired
samplers with different sampli
...
SLOVENSKI STANDARD
SIST ISO 16107:2012
01-april-2012
1DGRPHãþD
SIST ISO 16107:2002
=UDNQDGHORYQHPPHVWX3URWRNRO]DYUHGQRWHQMHODVWQRVWLGLIX]LMVNLK
Y]RUþHYDOQLNRY
Workplace atmospheres - Protocol for evaluating the performance of diffusive samplers
Air des lieux de travail - Protocole pour l'évaluation de la performance des dispositifs de
prélèvement par diffusion
Ta slovenski standard je istoveten z: ISO 16107:2007
ICS:
13.040.30 Kakovost zraka na delovnem Workplace atmospheres
mestu
SIST ISO 16107:2012 en
2003-01.Slovenski inštitut za standardizacijo. Razmnoževanje celote ali delov tega standarda ni dovoljeno.
---------------------- Page: 1 ----------------------
SIST ISO 16107:2012
---------------------- Page: 2 ----------------------
SIST ISO 16107:2012
INTERNATIONAL ISO
STANDARD 16107
Second edition
2007-10-15
Workplace atmospheres — Protocol for
evaluating the performance of diffusive
samplers
Air des lieux de travail — Protocole pour l'évaluation de la performance
des dispositifs de prélèvement par diffusion
Reference number
ISO 16107:2007(E)
©
ISO 2007
---------------------- Page: 3 ----------------------
SIST ISO 16107:2012
ISO 16107:2007(E)
PDF disclaimer
This PDF file may contain embedded typefaces. In accordance with Adobe's licensing policy, this file may be printed or viewed but
shall not be edited unless the typefaces which are embedded are licensed to and installed on the computer performing the editing. In
downloading this file, parties accept therein the responsibility of not infringing Adobe's licensing policy. The ISO Central Secretariat
accepts no liability in this area.
Adobe is a trademark of Adobe Systems Incorporated.
Details of the software products used to create this PDF file can be found in the General Info relative to the file; the PDF-creation
parameters were optimized for printing. Every care has been taken to ensure that the file is suitable for use by ISO member bodies. In
the unlikely event that a problem relating to it is found, please inform the Central Secretariat at the address given below.
COPYRIGHT PROTECTED DOCUMENT
© ISO 2007
All rights reserved. Unless otherwise specified, no part of this publication may be reproduced or utilized in any form or by any means,
electronic or mechanical, including photocopying and microfilm, without permission in writing from either ISO at the address below or
ISO's member body in the country of the requester.
ISO copyright office
Case postale 56 • CH-1211 Geneva 20
Tel. + 41 22 749 01 11
Fax + 41 22 749 09 47
E-mail copyright@iso.org
Web www.iso.org
Published in Switzerland
ii © ISO 2007 – All rights reserved
---------------------- Page: 4 ----------------------
SIST ISO 16107:2012
ISO 16107:2007(E)
Contents Page
Foreword. iv
Introduction . v
1 Scope . 1
2 Normative references . 1
3 Terms and definitions. 2
4 Symbols and abbreviated terms . 2
5 Summary of test protocol . 3
5.1 Factors affecting performance . 3
5.2 “Accuracy” as providing a single performance value linking to measurement uncertainty. 4
5.3 Bias, intersampler variability and the effects of environmental uncertainty. 4
5.4 Reverse diffusion. 5
5.5 Capacity — Control of effects from interfering compounds. 6
5.6 Capacity overload detection. 6
5.7 Desorption efficiency . 7
5.8 Atmospheric pressure. 7
5.9 Wind direction . 7
5.10 Simplifications . 7
6 Apparatus . 7
6.1 Exposure chamber specifications . 7
6.2 Controlled environmental conditions. 8
6.3 Inter-run variability . 8
7 Reagents and materials . 9
8 Procedure . 9
9 Sampler performance classification. 9
10 Accuracy. 10
10.1 General. 10
10.2 Accuracy range confidence limit . 10
10.3 Expanded uncertainty . 11
11 Test report . 11
Annex A (informative) Worked example — Program for diffusive sampler accuracy range
calculation . 13
Annex B (informative) Calculation of variances. 15
Bibliography . 16
© ISO 2007 – All rights reserved iii
---------------------- Page: 5 ----------------------
SIST ISO 16107:2012
ISO 16107:2007(E)
Foreword
ISO (the International Organization for Standardization) is a worldwide federation of national standards bodies
(ISO member bodies). The work of preparing International Standards is normally carried out through ISO
technical committees. Each member body interested in a subject for which a technical committee has been
established has the right to be represented on that committee. International organizations, governmental and
non-governmental, in liaison with ISO, also take part in the work. ISO collaborates closely with the
International Electrotechnical Commission (IEC) on all matters of electrotechnical standardization.
International Standards are drafted in accordance with the rules given in the ISO/IEC Directives, Part 2.
The main task of technical committees is to prepare International Standards. Draft International Standards
adopted by the technical committees are circulated to the member bodies for voting. Publication as an
International Standard requires approval by at least 75 % of the member bodies casting a vote.
Attention is drawn to the possibility that some of the elements of this document may be the subject of patent
rights. ISO shall not be held responsible for identifying any or all such patent rights.
ISO 16107 was prepared by Technical Committee ISO/TC 146, Air quality, Subcommittee SC 2, Workplace
atmospheres.
This second edition cancels and replaces the first edition (ISO 16107:1999), which has been technically
revised.
iv © ISO 2007 – All rights reserved
---------------------- Page: 6 ----------------------
SIST ISO 16107:2012
ISO 16107:2007(E)
Introduction
Gas or vapor sampling is often accomplished by actively pumping air through a collection medium such as
activated charcoal. Problems associated with a pump, such as inconvenience, inaccuracy and expense, are
inextricable from this type of sampling. The alternative covered by this International Standard is to use
diffusion for moving the compound of interest onto the collection medium. This approach to sampling is
attractive because of the convenience of use and low total monitoring cost.
However, previous studies have found significant problems with the accuracy of some samplers. Therefore,
although diffusive samplers may provide a plethora of data, inaccuracies and misuse of diffusive samplers
may yet affect research studies. Furthermore, worker protection may be based on faulty assumptions. The
aim of this International Standard is to counter the uncertainties in diffusive sampling through achieving a
broadly accepted set of performance tests and acceptance criteria for proving the efficacy of any given
diffusive sampler intended for use.
This International Standard is intended specifically for the large-scale evaluation of many diffusive
sampler/analyte pairs of practical application and is complementary to EN 838. An affordable, experimental
evaluation determines a single performance value indicating how a sampler performs in a typical situation. A
sampler can thereby be quickly judged as to acceptability. Additionally, sufficient data are obtained to predict
performance in many atypical situations. For example, although sampling may normally be done at room
temperature, a particular need may call for use in extreme cold. In such a case, the single performance value
would be superseded by the particular needs.
© ISO 2007 – All rights reserved v
---------------------- Page: 7 ----------------------
SIST ISO 16107:2012
---------------------- Page: 8 ----------------------
SIST ISO 16107:2012
INTERNATIONAL STANDARD ISO 16107:2007(E)
Workplace atmospheres — Protocol for evaluating the
performance of diffusive samplers
1 Scope
This International Standard specifies methods for evaluation of sampler performance in terms of workplace
conditions: wind speed, humidity, temperature, atmospheric pressure, and analyte variation. The concise set
of experiments specified aims to minimize cost to the user. The evaluation is limited to conditions commonly
encountered in personal sampling in the indoor workplace setting, namely wind speeds of up to 0,5 m/s and
for sampling periods typically from 2 h to 8 h.
Static or area sampling, unlike personal sampling where movement of the subject is significant, may
sometimes be subject to sampling-rate reduction due to stagnation at very low wind speeds. This International
Standard therefore does not apply to wind speeds of less than 0,1 m/s relative to static samplers. Samplers
are also tested for compliance with the manufacturer's stated limits on capacity, possibly in the presence of
interfering compounds. Given a suitable exposure chamber, the sampler evaluation protocol can be extended
to cover sampler use for other sampling periods and conditions.
This International Standard indicates how to measure diffusive sampler uncertainty for characterizing
concentration estimates obtained subsequent to the evaluation. It is impractical continually to re-evaluate
diffusive sampler performance under various environmental conditions prevailing during application.
NOTE 1 In this International Standard, the confidence level for the initial method evaluation becomes an integral part of
the measurement uncertainty. This approach slightly broadens the statistical protocols given in ISO Guide 98:1995.
Furthermore, the possibility of sampler errors related to correctible sampler bias is addressed.
NOTE 2 This International Standard is an extension of previous research on diffusive samplers (References [1] to [17]
inclusive and EN 838).
2 Normative references
The following referenced documents are indispensable for the application of this document. For dated
references, only the edition cited applies. For undated references, the latest edition of the referenced
document (including any amendments) applies.
EN 838, Workplace atmospheres — Diffusive samplers for the determination of gases and vapours —
Requirements and test methods
ISO Guide 98:1995, Guide to the expression of uncertainty in measurement. BIPM, IEC, IFCC, ISO, IUPAC,
IUPAP, OIML
© ISO 2007 – All rights reserved 1
---------------------- Page: 9 ----------------------
SIST ISO 16107:2012
ISO 16107:2007(E)
3 Terms and definitions
For the purposes of this document, the terms and definitions given in EN 838 and ISO Guide 98:1995 and the
following apply.
3.1
symmetric accuracy range
A
fractional range about the measurand concentration, c, within which 95 % of sampler measurements are found
NOTE See References [18] to [21] inclusive.
If the modulus of the bias is small, i.e. │∆│ < R/1,645, the symmetric accuracy range, A, can be shown to be closely
approximated (Reference [21]) by Equation (1):
22
AR =×1,960 ∆+ (1)
where
∆ is the bias, expressed relative to true concentrations;
R is the overall true relative standard deviation, expressed relative to true concentrations.
Otherwise
AR =+|∆| 1,645× (2)
2
If the bias is corrected, the expected value of ∆ in Equation (1) becomes equal to the variance of the bias correction,
reflecting an uncorrectable residual bias due to uncertainty in the correction. Then if the bias uncertainty is under control,
the low bias modulus model of Equation (1) indicates the proportionality of A to the root mean squared combined
uncertainty components.
4 Symbols and abbreviated terms
A symmetric accuracy range (3.1) in terms of bias and precision
A estimated symmetric accuracy range, A
est
A 95 % confidence level on the symmetric accuracy range, A
95 %
c true or reference analyte concentration, in milligrams per cubic metre, or parts per million as a ratio of
analyte to air molecules, per instructions of the sampler manufacturer
c mean of (four) concentration estimates (including pressure and temperature corrections), in milligrams
est
per cubic metre or parts per million as a ratio of analyte to air molecules, per instructions of the
sampler manufacturer
h humidity, partial pressure of water vapor, in kilopascals
n number of diffusive samplers tested for measuring sampler capacity
p (atmospheric) pressure
R overall true relative standard deviation of concentration estimates (dependent on assumed
environmental variability), expressed as a percentage relative to a “true” concentration as estimated
by reference sampling
2 © ISO 2007 – All rights reserved
---------------------- Page: 10 ----------------------
SIST ISO 16107:2012
ISO 16107:2007(E)
R estimated true relative standard deviation, expressed as a percentage
est
R true relative standard deviation characterizing inter-run chamber variability, expressed as a
run
percentage
R intersampler component of the true relative standard deviation, expressed as a percentage
s
R estimated intersampler true relative standard deviation, R , expressed as a percentage
s est s
R pulse-induced true relative standard deviation, expressed as a percentage
t
R 95 % confidence limit on the true relative standard deviation, expressed as a percentage
95 %
s estimated standard deviation characterizing intersampler variation
t (ν) value which, at the 95 % probability level, exceeds random variables distributed according to the
0,95
Student t-distribution with ν degrees of freedom
T temperature, in degrees Celsius
u ambient wind speed, in metres per second
α concentration estimate dependence on environmental variable, x (T, h, u, or c)
x
∆ bias relative to concentration, c
∆ estimated bias, ∆
est
∆ bias associated with concentration pulse
t
∆ 95 % confidence limit on the bias, ∆
95 %
ν degrees of freedom in determining R
s
ν effective number of degrees of freedom in determining R
eff
σ assumed concentration variability
c
σ assumed humidity variability
h
σ assumed temperature variability
T
σ assumed ambient wind speed variability
u
5 Summary of test protocol
5.1 Factors affecting performance
5.1.1 Diffusive sampling may first of all suffer from error in the sampling rate as stated by the manufacturer
of the sampler. As diffusive samplers are usually used without recalibration, this error implies a bias or
systematic error in all concentration estimates made. As the bias may be in one direction, such error cannot
be minimized by averaging several measurements. If the error is correctible it is not strictly part of the sampler
uncertainty.
© ISO 2007 – All rights reserved 3
---------------------- Page: 11 ----------------------
SIST ISO 16107:2012
ISO 16107:2007(E)
5.1.2 Aside from bias, several sources of random error exist. Some of these errors are associated with the
sampler itself. For example, the sampler’s dimensional tolerance may be low, leading to variation in the
sampling rate, sampler to sampler. The sampler may become overloaded or may not follow concentration
changes adequately.
5.1.3 Other errors relate to a non-constant environment which may affect the sampling rate. Potentially
significant environmental influence quantities include temperature, humidity, wind speed, and atmospheric
pressure. To predict the effect of any such parameter, an experiment on the sampler type may be carried out.
Moreover, the variation expected of the parameter requires estimation. So as to result in a single performance
measure, typical ranges of these environmental parameters are specified in this International Standard.
5.2 “Accuracy” as providing a single performance value linking to measurement
uncertainty
5.2.1 A sampler presented by a manufacturer for evaluation may be found to exhibit both imprecision and
bias. In the language of ISO Guide 98:1995, imprecision and bias together relate to the “accuracy” of a
sampler, rather than “uncertainty”. Accuracy refers to the closeness between measured values and
measurand (in this case the concentration to be measured), whereas uncertainty refers to what is known
about the measurand on the basis of the measurement. In other words, the manufacturer could recommend
use of information about the bias for correction of the measured values. This International Standard provides
means of judging a sampler as recommended for use by a manufacturer.
5.2.2 A useful statistical measure exists which summarizes the overall effect of both random variation and
bias. This function has come to be known as the symmetric accuracy range, A (3.1).
5.2.3 The definition directly implies that A increases with both random error and bias magnitude. Therefore
A is one particular quantification of accuracy as defined in ISO Guide 98:1995. The greater the disparity
between measured values and measurand, the larger the value of A.
NOTE A is not the only way of quantifying accuracy. For example, the selection of 95 % as coverage level is arbitrary.
Furthermore, other functions consistent with the qualitative idea of accuracy as described in ISO Guide 98:1995 are
possible.
5.2.4 A indicates quantitatively how far off measurements would generally be if further correction of the
sampler were not made. A further advantage of A is that a direct link to the uncertainty described in
ISO Guide 98:1995 is obtained in the case that bias is minimized, e.g. through results of the evaluation.
5.2.5 Moreover, a measurement element not covered extensively within ISO Guide 98:1995 is dealt with
explicitly. Namely, diffusive samplers are generally used as obtained from the manufacturer without
re-evaluation at each use. Therefore, uncertainty in a measured value must account for confidence in the
[single] evaluation itself as well as variation and possible residual bias at the point of use.
5.2.6 The result of approaching uncertainty from the point of view of accuracy results in an “expanded
uncertainty” (in the language of ISO Guide 98:1995) proportional via a “coverage factor” to the root mean
squared combined uncertainty components comprised of bias uncertainty, the effects of sampler variation,
and influence quantities described above. The coverage factor is determined by the confidence required [e.g.,
95 %] in the sampler evaluation and the coverage [also 95 %]. The final result is a value of expanded
uncertainty with specific [tolerance or predictive] meaning, namely, in providing the 95 % coverage range
about the [true] measurand at 95 % confidence in the evaluation.
5.3 Bias, intersampler variability and the effects of environmental uncertainty
5.3.1 This International Standard gives a procedure for assessing the effects of variability in the following
workplace variables: temperature, T; humidity, h (expressed in terms of the water vapor partial pressure to
minimize interaction with the temperature); ambient wind speed, u, across the sampler face [see (5.7)
regarding wind direction]; and concentration, c. An experiment is carried out which provides information about
the dependencies of the concentration estimates on these variables near conditions of intended sampler use
(T , h , u , and c ). Testing is required at the concentration, c , of intended use, as well as at concentrations
0 0 0 0 0
reduced at least to c /2. Furthermore, the sampler bias and the intersampler standard deviation are measured.
0
4 © ISO 2007 – All rights reserved
---------------------- Page: 12 ----------------------
SIST ISO 16107:2012
ISO 16107:2007(E)
Finally, the effect of diffusion of material out of the sampler is measured. Pressure effects result in correctable
bias and are not evaluated in this International Standard (5.8).
5.3.2 Using four samplers for each of five experimental runs (the minimum possible), the dependencies α ,
T
α , α , and α (relative to the chamber reference concentration and target environmental parameters) on
h u c
changes in T, h, u, and c are measured, following the sampler manufacturer's instructions regarding pressure
and temperature corrections (if any). These experiments also give a value for the estimated sampler bias, ∆,
relative to the chamber reference concentration (defined for the target conditions). Two further runs describing
time-dependent effects (5.4.5) from diffusive loss of analyte are also carried out. The chamber reference
concentration must be traceable to primary standards of mass and volume.
5.3.3 Error in the estimates of the dependencies α , α , α , and α will exist on account of the intersampler
T h u c
component of the true relative standard deviation R and an inter-run chamber true relative standard deviation,
s
R . The latter results in part from uncertainty in the reference concentration. R is obtained by pooling the
run s
variance estimates from each run and therefore is estimated with 7 × 3 = 21 degrees of freedom (or
15 degrees of freedom if the reverse diffusion experiment is omitted). So as to avoid re-measurement at each
sampler/analyte evaluation, R is obtained by a separate characterization of the chamber with several runs
run
at (for example) fixed environmental conditions. An example in which the dependencies α and R are
s
estimated is presented in Annex A.
NOTE It is up to the user as to how traceability is established. In Reference [12], the concentration estimate, as
calculated from the chamber's analyte generation parameters, is regarded as the “benchmark”, although an independent
estimate is required and must be within 5 % of the calculated estimate. If these estimates differ, then a third independent
estimate is required to establish the reference concentration through agreement with one of the other independent
estimates. One possibility for such an independent estimate is the mean of at least five independent, active sampler
estimates per run within the chamber. Experiment (Reference [12]) on the accuracy of such reference measurements
using sorbent tubes indicates that a true relative standard deviation of the order of 2 % can be achieved for the individual
measurements. Alternatively, Reference [3] requires averaging of at least two independent methods (possibly including
calculated estimates) with at least four samples per method. EN 838 has adopted the looser requirement that calculated
and independent measurements shall agree within 10 %.
5.3.4 A further consolidation of tests may be made by observing that the dependence of concentration
estimates on the wind speed, v, is only sampler specific, i.e. does not depend on the specific analyte.
Therefore, after a single measurement for a given sampler type, the set of tests can be narrowed to five runs
with 5 × 3 = 15 degrees of freedom in the estimate of R .
s
5.4 Reverse diffusion
5.4.1 A potential problem with diffusive samplers is presented by the possibility of reverse diffusion
(sometimes denoted as “back-diffusion” or “off-gassing”) of the analyte. Reverse diffusion is generally only
significant in the case that an analyte is only weakly bound to the sorbent (Reference [6]). Therefore,
inaccuracy associated with these effects may usually be minimized through proper sorbent selection.
5.4.2 Because of reverse diffusion, estimates of a varying concentration may in some cases be biased. The
worst-case situation occurs with the concentration in the form of an isolated pulse at either the beginning or
end of the sampling period. A pulse at the beginning of the period allows the entire sampling period (4 h
to 12 h) for sample loss, possibly resulting in a low estimate relative to a pulse at the end.
5.4.3 In some cases, the time dependence of a specific workplace concentration correlates strongly with the
sampling period. For example, a cleanup operation at the end of a workday could introduce analyte only then.
This could imply a positive bias in the concentration estimates obtained from a day's sampling. For simplicity,
however, this International Standard is designed to assess performance of samplers for use in a concentration
fluctuating in a stationary manner, so that time-dependent effects are treated simply as components of
sampler variance. Specifically, the effect of an isolated 0,5 h pulse occurring at random within the sampling
period is estimated.
5.4.4 Challenging samplers to 0,5 h pulses is similar to the tests of NIOSH (Reference [3]) and EN 838.
© ISO 2007 – All rights reserved 5
---------------------- Page: 13 ----------------------
SIST ISO 16107:2012
ISO 16107:2007(E)
5.4.5 Let ∆ (>0) represent one-half the bias between estimates from a 0,5 h pulse at the end versus the
t
beginning of the sampling period, relative to the mean of the estimates. Assume, conservatively (see, for
example, Reference [6]), that the bias in the estimates of 0,5 h pulse occurring at random within (for example,
2
an 8 h sampling period ranges uniformly between −∆ and +∆. Then the variance, R , associated with
t
t t
sampling a 0,5 h pulse at random within the sampling period is given by Equation (3):
22
1
R = ∆ (3)
tt
3
5.5 Capacity — Control of effects from interfering compounds
5.5.1 This International Standard provides a test for confirming a manufacturer's claimed sampler capacity
under stated conditions of use. Such conditions would normally refer to a specific sampling period and to
environmental extremes, such as 80 % relative humidity at a temperature of 30 °C. Additionally, a
manufacturer may claim a value of capacity for sampling in the presence of specific interferences at stated
concentrations.
5.5.2 For the purposes of this International Standard, capacity is defined as the sampled mass (or
equivalently as the concentration for a specific sampling period) at which concentration estimates are 10 %
low. Specifically, capacity is considered not exceeded if concentration estimates, corrected for correctable
bias, are above 90 % of the true concentration at the 95 % confidence level.
5.5.3 An example of the test is as follows: eight diffusive and eight active samplers are exposed to the
analyte of interest under the stated environmental conditions. Then, neglecting variability in the reference
sampler mean, the 95 % confidence limit ∆µ on the difference in the (unknown) mean concentration
95 %
estimates is given by Equation (4)
...
Questions, Comments and Discussion
Ask us and Technical Secretary will try to provide an answer. You can facilitate discussion about the standard in here.