ISO 16107:1999

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

Zrak na delovnem mestu – Protokol za ovrednotenje lastnosti difuzijskih vzorčevalnikov

General Information

• Status: Withdrawn
• Publication Date: 29-Sep-1999
• Withdrawal Date: 29-Sep-1999
• ICS: 13.040.30 - Workplace atmospheres
• Technical Committee: ISO/TC 146/SC 2 - Workplace atmospheres
• Drafting Committee: ISO/TC 146/SC 2 - Workplace atmospheres
• Current Stage: 9599 - Withdrawal of International Standard
• Completion Date: 15-Oct-2007
• Ref Project: SIST ISO 16107:2002 - Workplace atmospheres -- Protocol for evaluating the performance of diffusive samplers

Relations

• Revised: ISO 16107:2007 - Workplace atmospheres — Protocol for evaluating the performance of diffusive samplers
• Effective Date: 15-Apr-2008

Standards Content (Sample)

INTERNATIONAL ISO
STANDARD 16107
First edition
1999-09-15
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
A
Reference number
ISO 16107:1999(E)

---------------------- Page: 1 ----------------------
ISO 16107:1999(E)
Contents
1 Scope .1
2 Normative reference .1
3 Terms and definitions .1
4 Symbols and abbreviated terms .2
5 Summary of test protocol .3
6 Apparatus .5
7 Reagents and materials.7
8 Procedure .7
9 Sampler performance classification .7
10 Accuracy.8
11 Test report .8
Annex A (informative) Worked example — Computer program for diffusive sampler accuracy calculation .10
Bibliography.14
©  ISO 1999
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 the publisher.
International Organization for Standardization
Case postale 56 • CH-1211 Genève 20 • Switzerland
Internet [e-mail]
Printed in Switzerland
ii

---------------------- Page: 2 ----------------------
© ISO
ISO 16107:1999(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 3.
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.
International Standard ISO 16107 was prepared by Technical Committee ISO/TC 146, Air quality, Subcommittee
SC 2, Workplace atmospheres.
iii

---------------------- Page: 3 ----------------------
© ISO
ISO 16107:1999(E)
Introduction
Gas or vapour 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
practice 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.
iv

---------------------- Page: 4 ----------------------
INTERNATIONAL STANDARD  © ISO ISO 16107:1999(E)
Workplace atmospheres — Protocol for evaluating the
performance of diffusive samplers
1 Scope
1.1  This International Standard covers the evaluation of the performance of diffusive samplers of gases and
vapours for use over sampling periods from 4 h to 12 h. Sampling periods of such duration are the most common in
workplace sampling. Given a suitable exposure chamber, this International Standard can be straightforwardly
extended to cover samplers for use over other sampling periods as well. The aim is to provide a concise set of
experiments for classifying samplers primarily according to a single numerical value representing sampler accuracy.
NOTE Accuracy estimates refer to conditions of sampler use which are normally expected in a workplace setting. These
conditions may be characterized by the temperature, atmospheric pressure, humidity and ambient wind speed, none of which
may be constant or accurately known. Furthermore, the accuracy accounts for difficulty in the estimation of time-weighted
averages of concentrations which may not be constant in time.
In addition to accuracy determination, a method is provided to test the samplers for compliance with the
manufacturer's stated limits on capacity, possibly in the presence of interfering compounds. A method is given for
classification of samplers according to their capability to detect situations in which sampler capacity may be
exceeded.
1.2  This International Standard is an extension of previous research on diffusive samplers [1-17] as well as EN
838. Essential advantages are the estimation of sampler accuracy under actual conditions of use and the reduction
in cost of sampler evaluation.
NOTE Furthering the latter point, knowledge of similarity between analytes of interest can be used to expedite sampler
evaluation. For example, interpolation of data characterizing the sampling of analytes at separated points of a homologous
series of compounds is recommended. At present the procedure in [9] is suggested: Following evaluation of a sampler in use at
a single homologous series member according to the present practice, higher molecular weight members would receive partial
validations considering sampling rate, capacity, analytical recovery and interferences.
2 Normative reference
The following normative document contains provisions which, through reference in this text, constitute provisions of
this International Standard. For dated references, subsequent amendments to, or revisions of, any of these
publications do not apply. However, parties to agreements based on this International Standard are encouraged to
investigate the possibility of applying the most recent edition of the normative document indicated below. For
undated references, the latest edition of the normative document referred to applies. Members of ISO and IEC
maintain registers of currently valid International Standards.
EN 838: Standard on workplace atmospheres — Diffusive samplers for the determination of gases or vapours —
Requirements and test methods.
3 Terms and definitions
For the purposes of this International Standard, the terms and definitions given in EN 838 as well as the following
apply.
1

---------------------- Page: 5 ----------------------
© ISO
ISO 16107:1999(E)
3.1
Busch probabilistic accuracy
A
fractional range, symmetric about the true concentration c, within which 95 % of sampler measurements are found
See [18-21].
NOTE In the case considered here, effects on sampler accuracy from environmental unknowns are all handled as
variances, leaving negligible uncorrectable bias.
A = 1,960 × CV (1)
where CV is the coefficient of variation (overall relative standard deviation).
4 Symbols and abbreviated terms
A Busch probabilistic accuracy as defined in terms of bias and precision
A estimated Busch probabilistic accuracy A
est
A 95 % confidence level on the Busch probabilistic accuracy A
95 %
c true or reference analyte concentration, in milligrams per cubic metre
c mean of (four) concentration estimates [including (p,T)-corrections], in milligrams per cubic metre,
est
obtained per instructions of sampler manufacturer
h humidity (expressed as partial pressure)
n number of diffusive samplers tested for measuring sampler capacity
p atmospheric pressure
CV coefficient of variation (overall relative standard deviation) of concentration estimates (dependent on
assumed environmental variability), expressed as a percentage
CV estimated coefficient of variation, expressed as a percentage
est
CV coefficient of variation characterizing inter-run chamber variability, expressed as a percentage
run
CV intersampler imprecision (relative to the reference concentration), expressed as a percentage
s
CV estimated intersampler imprecision CV , expressed as a percentage
s est s
CV pulse-induced imprecision, expressed as a percentage
t
CV 95 % confidence limit on the coefficient of variation, expressed as a percentage
95 %
s estimated standard deviation characterizing intersampler imprecision
t (v) value which, at probability 95 %, exceeds random variables distributed according to the Studentized
0,95
t-distribution with n degrees of freedom
T temperature, in degrees Celsius
v ambient wind speed, in metres per second
aconcentration estimate dependence on environmental variable x (T, h, v, or c)
x
Dbias relative to concentration c
2

---------------------- Page: 6 ----------------------
© ISO
ISO 16107:1999(E)
Destimated bias D
est
Dbias associated with concentration pulse
t
D95 % confidence limit on the bias D
95 %
ndegrees of freedom in determining CV
s
neffective number of degrees of freedom in determining CV
eff
sassumed concentration variability
c
sassumed humidity variability
h
sassumed temperature variability
T
sassumed ambient wind speed variability
v
5 Summary of test protocol
5.1 Bias, intersampler variability and the effects of environmental uncertainty
5.1.1  This International Standard gives a procedure for assessing the effects of variability in the following
workplace parameters: temperature T, humidity h (expressed in terms of the water vapour partial pressure to
minimize interaction with the temperature), the ambient wind speed v across the sampler face (see 5.7 regarding
wind direction), and concentration c. An experiment is carried out which provides information about the
concentration estimates' dependencies on these variables as well as the sampler bias, intersampler variability, and
concentration-dependent effects. Testing is required at a single target concentration, c , central to concentrations of
0
intended sampler use, as well as at a reduced concentration in the range c /10 to c /2. Pressure effects result in
0 0
one-time-correctable bias and are not evaluated here, aside from uncorrected bias (5.6).
5.1.2  Specifically, in terms of the known concentration c in the exposure chamber, the mean concentration
estimates c (over four samples at each condition), following p- and T-correction (if any) per the sampler
est
manufacturer's instruction, are modelled by:
c /c = 1 + D + a { (T/T 2 1) + a { (h/h - 1) + a { (v/v - 1) + a { (c/c - 1) (2)
est T 0 h 0 v 0 c 0
omitting error terms. The concentration c is the chamber "reference concentration" and shall be traceable to primary
standards of mass and volume. Estimates of the model parameters D, a, a, a and a are obtained from an
T h v c
experiment consisting of five runs, varying T, h, v and c, with four diffusive samplers each. The parameter D
characterizes sampler bias at the intermediate conditions (T , h , v , c ). Error in equation (2) will exist on account
0 0 0 0
of intersampler imprecision (characterized by CV ) together with an inter-run chamber variability (CV ) resulting in
s run
part from uncertainty in the reference concentration. CV is obtained by pooling the variance estimates from each
s
run, together with a further run describing time effects (5.2.5), and therefore is estimated with 6 × 3 = 18 degrees of
freedom. To avoid re-measurement at each sampler/analyte evaluation, CV is obtained by a separate
run
characterization of the chamber with several runs at (for example) fixed environmental conditions. An example in
which the parameters {a} and CV are estimated is presented in annex A.
s
NOTE It is up to the user as to how traceability is established. Within [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 [12] on the accuracy of such reference measurements using sorbent tubes indicates that a relative standard
deviation of the order of 2 % can be achieved for the individual measurements. Alternatively, [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 %.
5.1.3  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
3

---------------------- Page: 7 ----------------------
© ISO
ISO 16107:1999(E)
measurement for a given sampler type, the set of tests can be narrowed to five runs with 5 3 3 = 15 degrees of
freedom in the estimate of CV .
s
5.2 Reverse diffusion
5.2.1  A potential problem with diffusive samplers is presented by the possibility of reverse diffusion (sometimes
denoted as "back-diffusion" or "off-gassing") of analyte. Reverse diffusion can occur directly from the air spaces of a
diffusive sampler, depending on geometry. For example, a sampler as long as the Palmes tube (7 cm) used over
short sampling periods (15 min) can display a measurable effect of this type [2]. More commonly, reverse diffusion
may be significant in the case that an analyte is only weakly bound to the sorbent [6]. Therefore, inaccuracy
associated with these effects may generally be minimized through proper sorbent selection.
5.2.2  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.
5.2.3  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 solvent 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 for assessing performance of samplers for use in a stationarily fluctuating
concentration, 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.
5.2.4  Challenging samplers to 0,5-h pulses is similar to tests suggested by NIOSH [3] and CEN (EN 838).
5.2.5  Let D represent the corrected bias in estimating a 0,5-h pulse at the end of the sampling period relative to a
t
known concentration c, where D is the uncorrected bias in sampling over the sampling period of intended application
(e.g. 8 h). For pulses occurring at other times, assume conservatively (see e.g. [6]) that the bias D is proportional to
t
2
the interval from the centre of the sampling period to the time the pulse occurs. Then the variance CV associated
t
with sampling a 0,5-h pulse at random within the sampling period is:
2 2
CV = D/3. (3)
t t
5.3 Capacity — Control of effects from interfering compounds
5.3.1  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 equal to 30 °C. Additionally, a manufacturer may claim a
value of capacity for sampling in the presence of specific interferences at stated concentrations.
5.3.2  For the purposes of this International Standard, capacity is defined as the sampled mass (or equivalently as
the concentration at 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.
5.3.3  An example of the test is as follows: eight diffusive and eight active samplers are exposed to the analyte of
concern under the stated environmental conditions. Suppose the individual diffusive sampler inaccuracy estimate is
s. Then, neglecting variability in the reference sampler mean, the 95 % confidence limit Dm on the difference in

95 %
the (unknown) mean concentration estimates is:
Dm = Dc 2 s { t (n)/ v ; (4)
[]
95 % 0,95
where Dc is the estimated mean difference between diffusive and active results, n = 8, and n = n - 1 = 7. Then
Dm shall be greater than - 10 % c, where c is the mean concentration estimate from the reference samplers.
95 %
EXAMPLE
Suppose the diffusive sampler coefficient of variation CV = 5 %,
s
4

---------------------- Page: 8 ----------------------
© ISO
ISO 16107:1999(E)
(s/c) { t (n)/ v = 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.
5.4 Capacity overload detection
The capability of detecting capacity overload (e.g. by the use of a second sorbent or by employing paired samplers
with different sampling rates) may be advantageous in some sampling situations. In the case of active samplers,
such detection is easily effected through the use of back-up sections. Therefore, diffusive samplers with similar
features will receive a specific classification. The point is that practicality precludes testing of the samplers under all
conditions of use, such as in an arbitrary multianalyte environment. The capability of voiding a sample when
interferences become demonstrably problematic may therefore be useful. At present the efficacy of such
breakthrough detection is not evaluated. However, evaluation tests may be developed in the future for this purpose.
5.5 Desorption efficiency
5.5.1  A further control of the effects from interfering compounds is afforded by restricting the permissible
desorption efficiency. As in [3], the desorption efficiency, in the case of solvent extraction, shall be > 75 % at the
concentration of intended application of the sampler. This requirement is expected to control the potential
...

SLOVENSKI STANDARD
SIST ISO 16107:2002
01-maj-2002
=UDNQDGHORYQHPPHVWX±3URWRNRO]DRYUHGQRWHQMHODVWQRVWLGLIX]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 16107:1999
ICS:
13.040.30 Kakovost zraka na delovnem Workplace atmospheres
mestu
SIST ISO 16107:2002 en
2003-01.Slovenski inštitut za standardizacijo. Razmnoževanje celote ali delov tega standarda ni dovoljeno.

---------------------- Page: 1 ----------------------

SIST ISO 16107:2002

---------------------- Page: 2 ----------------------

SIST ISO 16107:2002
INTERNATIONAL ISO
STANDARD 16107
First edition
1999-09-15
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
A
Reference number
ISO 16107:1999(E)

---------------------- Page: 3 ----------------------

SIST ISO 16107:2002
ISO 16107:1999(E)
Contents
1 Scope .1
2 Normative reference .1
3 Terms and definitions .1
4 Symbols and abbreviated terms .2
5 Summary of test protocol .3
6 Apparatus .5
7 Reagents and materials.7
8 Procedure .7
9 Sampler performance classification .7
10 Accuracy.8
11 Test report .8
Annex A (informative) Worked example — Computer program for diffusive sampler accuracy calculation .10
Bibliography.14
©  ISO 1999
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 the publisher.
International Organization for Standardization
Case postale 56 • CH-1211 Genève 20 • Switzerland
Internet [e-mail]
Printed in Switzerland
ii

---------------------- Page: 4 ----------------------

SIST ISO 16107:2002
© ISO
ISO 16107:1999(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 3.
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.
International Standard ISO 16107 was prepared by Technical Committee ISO/TC 146, Air quality, Subcommittee
SC 2, Workplace atmospheres.
iii

---------------------- Page: 5 ----------------------

SIST ISO 16107:2002
© ISO
ISO 16107:1999(E)
Introduction
Gas or vapour 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
practice 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.
iv

---------------------- Page: 6 ----------------------

SIST ISO 16107:2002
INTERNATIONAL STANDARD  © ISO ISO 16107:1999(E)
Workplace atmospheres — Protocol for evaluating the
performance of diffusive samplers
1 Scope
1.1  This International Standard covers the evaluation of the performance of diffusive samplers of gases and
vapours for use over sampling periods from 4 h to 12 h. Sampling periods of such duration are the most common in
workplace sampling. Given a suitable exposure chamber, this International Standard can be straightforwardly
extended to cover samplers for use over other sampling periods as well. The aim is to provide a concise set of
experiments for classifying samplers primarily according to a single numerical value representing sampler accuracy.
NOTE Accuracy estimates refer to conditions of sampler use which are normally expected in a workplace setting. These
conditions may be characterized by the temperature, atmospheric pressure, humidity and ambient wind speed, none of which
may be constant or accurately known. Furthermore, the accuracy accounts for difficulty in the estimation of time-weighted
averages of concentrations which may not be constant in time.
In addition to accuracy determination, a method is provided to test the samplers for compliance with the
manufacturer's stated limits on capacity, possibly in the presence of interfering compounds. A method is given for
classification of samplers according to their capability to detect situations in which sampler capacity may be
exceeded.
1.2  This International Standard is an extension of previous research on diffusive samplers [1-17] as well as EN
838. Essential advantages are the estimation of sampler accuracy under actual conditions of use and the reduction
in cost of sampler evaluation.
NOTE Furthering the latter point, knowledge of similarity between analytes of interest can be used to expedite sampler
evaluation. For example, interpolation of data characterizing the sampling of analytes at separated points of a homologous
series of compounds is recommended. At present the procedure in [9] is suggested: Following evaluation of a sampler in use at
a single homologous series member according to the present practice, higher molecular weight members would receive partial
validations considering sampling rate, capacity, analytical recovery and interferences.
2 Normative reference
The following normative document contains provisions which, through reference in this text, constitute provisions of
this International Standard. For dated references, subsequent amendments to, or revisions of, any of these
publications do not apply. However, parties to agreements based on this International Standard are encouraged to
investigate the possibility of applying the most recent edition of the normative document indicated below. For
undated references, the latest edition of the normative document referred to applies. Members of ISO and IEC
maintain registers of currently valid International Standards.
EN 838: Standard on workplace atmospheres — Diffusive samplers for the determination of gases or vapours —
Requirements and test methods.
3 Terms and definitions
For the purposes of this International Standard, the terms and definitions given in EN 838 as well as the following
apply.
1

---------------------- Page: 7 ----------------------

SIST ISO 16107:2002
© ISO
ISO 16107:1999(E)
3.1
Busch probabilistic accuracy
A
fractional range, symmetric about the true concentration c, within which 95 % of sampler measurements are found
See [18-21].
NOTE In the case considered here, effects on sampler accuracy from environmental unknowns are all handled as
variances, leaving negligible uncorrectable bias.
A = 1,960 × CV (1)
where CV is the coefficient of variation (overall relative standard deviation).
4 Symbols and abbreviated terms
A Busch probabilistic accuracy as defined in terms of bias and precision
A estimated Busch probabilistic accuracy A
est
A 95 % confidence level on the Busch probabilistic accuracy A
95 %
c true or reference analyte concentration, in milligrams per cubic metre
c mean of (four) concentration estimates [including (p,T)-corrections], in milligrams per cubic metre,
est
obtained per instructions of sampler manufacturer
h humidity (expressed as partial pressure)
n number of diffusive samplers tested for measuring sampler capacity
p atmospheric pressure
CV coefficient of variation (overall relative standard deviation) of concentration estimates (dependent on
assumed environmental variability), expressed as a percentage
CV estimated coefficient of variation, expressed as a percentage
est
CV coefficient of variation characterizing inter-run chamber variability, expressed as a percentage
run
CV intersampler imprecision (relative to the reference concentration), expressed as a percentage
s
CV estimated intersampler imprecision CV , expressed as a percentage
s est s
CV pulse-induced imprecision, expressed as a percentage
t
CV 95 % confidence limit on the coefficient of variation, expressed as a percentage
95 %
s estimated standard deviation characterizing intersampler imprecision
t (v) value which, at probability 95 %, exceeds random variables distributed according to the Studentized
0,95
t-distribution with n degrees of freedom
T temperature, in degrees Celsius
v ambient wind speed, in metres per second
aconcentration estimate dependence on environmental variable x (T, h, v, or c)
x
Dbias relative to concentration c
2

---------------------- Page: 8 ----------------------

SIST ISO 16107:2002
© ISO
ISO 16107:1999(E)
Destimated bias D
est
Dbias associated with concentration pulse
t
D95 % confidence limit on the bias D
95 %
ndegrees of freedom in determining CV
s
neffective number of degrees of freedom in determining CV
eff
sassumed concentration variability
c
sassumed humidity variability
h
sassumed temperature variability
T
sassumed ambient wind speed variability
v
5 Summary of test protocol
5.1 Bias, intersampler variability and the effects of environmental uncertainty
5.1.1  This International Standard gives a procedure for assessing the effects of variability in the following
workplace parameters: temperature T, humidity h (expressed in terms of the water vapour partial pressure to
minimize interaction with the temperature), the ambient wind speed v across the sampler face (see 5.7 regarding
wind direction), and concentration c. An experiment is carried out which provides information about the
concentration estimates' dependencies on these variables as well as the sampler bias, intersampler variability, and
concentration-dependent effects. Testing is required at a single target concentration, c , central to concentrations of
0
intended sampler use, as well as at a reduced concentration in the range c /10 to c /2. Pressure effects result in
0 0
one-time-correctable bias and are not evaluated here, aside from uncorrected bias (5.6).
5.1.2  Specifically, in terms of the known concentration c in the exposure chamber, the mean concentration
estimates c (over four samples at each condition), following p- and T-correction (if any) per the sampler
est
manufacturer's instruction, are modelled by:
c /c = 1 + D + a { (T/T 2 1) + a { (h/h - 1) + a { (v/v - 1) + a { (c/c - 1) (2)
est T 0 h 0 v 0 c 0
omitting error terms. The concentration c is the chamber "reference concentration" and shall be traceable to primary
standards of mass and volume. Estimates of the model parameters D, a, a, a and a are obtained from an
T h v c
experiment consisting of five runs, varying T, h, v and c, with four diffusive samplers each. The parameter D
characterizes sampler bias at the intermediate conditions (T , h , v , c ). Error in equation (2) will exist on account
0 0 0 0
of intersampler imprecision (characterized by CV ) together with an inter-run chamber variability (CV ) resulting in
s run
part from uncertainty in the reference concentration. CV is obtained by pooling the variance estimates from each
s
run, together with a further run describing time effects (5.2.5), and therefore is estimated with 6 × 3 = 18 degrees of
freedom. To avoid re-measurement at each sampler/analyte evaluation, CV is obtained by a separate
run
characterization of the chamber with several runs at (for example) fixed environmental conditions. An example in
which the parameters {a} and CV are estimated is presented in annex A.
s
NOTE It is up to the user as to how traceability is established. Within [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 [12] on the accuracy of such reference measurements using sorbent tubes indicates that a relative standard
deviation of the order of 2 % can be achieved for the individual measurements. Alternatively, [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 %.
5.1.3  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
3

---------------------- Page: 9 ----------------------

SIST ISO 16107:2002
© ISO
ISO 16107:1999(E)
measurement for a given sampler type, the set of tests can be narrowed to five runs with 5 3 3 = 15 degrees of
freedom in the estimate of CV .
s
5.2 Reverse diffusion
5.2.1  A potential problem with diffusive samplers is presented by the possibility of reverse diffusion (sometimes
denoted as "back-diffusion" or "off-gassing") of analyte. Reverse diffusion can occur directly from the air spaces of a
diffusive sampler, depending on geometry. For example, a sampler as long as the Palmes tube (7 cm) used over
short sampling periods (15 min) can display a measurable effect of this type [2]. More commonly, reverse diffusion
may be significant in the case that an analyte is only weakly bound to the sorbent [6]. Therefore, inaccuracy
associated with these effects may generally be minimized through proper sorbent selection.
5.2.2  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.
5.2.3  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 solvent 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 for assessing performance of samplers for use in a stationarily fluctuating
concentration, 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.
5.2.4  Challenging samplers to 0,5-h pulses is similar to tests suggested by NIOSH [3] and CEN (EN 838).
5.2.5  Let D represent the corrected bias in estimating a 0,5-h pulse at the end of the sampling period relative to a
t
known concentration c, where D is the uncorrected bias in sampling over the sampling period of intended application
(e.g. 8 h). For pulses occurring at other times, assume conservatively (see e.g. [6]) that the bias D is proportional to
t
2
the interval from the centre of the sampling period to the time the pulse occurs. Then the variance CV associated
t
with sampling a 0,5-h pulse at random within the sampling period is:
2 2
CV = D/3. (3)
t t
5.3 Capacity — Control of effects from interfering compounds
5.3.1  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 equal to 30 °C. Additionally, a manufacturer may claim a
value of capacity for sampling in the presence of specific interferences at stated concentrations.
5.3.2  For the purposes of this International Standard, capacity is defined as the sampled mass (or equivalently as
the concentration at 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.
5.3.3  An example of the test is as follows: eight diffusive and eight active samplers are exposed to the analyte of
concern under the stated environmental conditions. Suppose the individual diffusive sampler inaccuracy estimate is
s. Then, neglecting variability in the reference sampler mean, the 95 % confidence limit Dm on the difference in

95 %
the (unknown) mean concentration estimates is:
Dm = Dc 2 s { t (n)/ v ; (4)
[]
95 % 0,95
where Dc is the estimated mean difference between diffusive and active results, n = 8, and n = n - 1 = 7. Then
Dm shall be greater than - 10 % c, where c is the mean concentration estimate from the reference samplers.
95 %
EXAMPLE
Suppose the diffusive sampler coefficient of variation CV = 5 %,
s
4

---------------------- Page: 10 ----------------------

SIST ISO 16107:2002
© ISO
ISO 16107:1999(E)
(s/c) { t (n)/ v = 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.
5.4 Capacity overload detection
The capability of detecting capacity overload (e.g. by the use of a second sorbent or by employing paired samplers
with different sampling rates) may be advantageous in some sampling situations. In the case of active samplers,
such detection is easily effected through the use of back-up sections. Therefore, diffusive samplers with similar
features will receive a specific classification. The point is that practicality precludes testing of the samplers under all
conditions of use, such as in an arbitrary multianalyte environment. The capability of voiding a sample when
interferences become demonstrably problematic may therefore be useful. At present the efficacy of such
breakthrough detection is not evaluated. However, evaluation te
...