# SIST ISO 16107:2002

(Main)## Workplace atmospheres -- Protocol for evaluating the performance of diffusive samplers

## 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

### RELATIONS

### 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

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 1999All 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 StandardizationCase postale 56 • CH-1211 Genève 20 • Switzerland

Internet iso@iso.ch

Printed in Switzerland

---------------------- 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.---------------------- 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 referenceThe 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.---------------------- Page: 5 ----------------------

© ISO

ISO 16107:1999(E)

3.1

Busch probabilistic accuracy

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,

estobtained 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 percentageCV estimated coefficient of variation, expressed as a percentage

est

CV coefficient of variation characterizing inter-run chamber variability, expressed as a percentage

runCV intersampler imprecision (relative to the reference concentration), expressed as a percentage

CV estimated intersampler imprecision CV , expressed as a percentages est s

CV pulse-induced imprecision, expressed as a percentage

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

0,95t-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)

Dbias relative to concentration c

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

© ISO

ISO 16107:1999(E)

Destimated bias D

est

Dbias associated with concentration pulse

D95 % confidence limit on the bias D

95 %

ndegrees of freedom in determining CV

neffective number of degrees of freedom in determining CV

eff

sassumed concentration variability

sassumed humidity variability

sassumed temperature variability

sassumed ambient wind speed variability

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

intended sampler use, as well as at a reduced concentration in the range c /10 to c /2. Pressure effects result in

0 0one-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

estmanufacturer'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 cexperiment 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 0of intersampler imprecision (characterized by CV ) together with an inter-run chamber variability (CV ) resulting in

s runpart from uncertainty in the reference concentration. CV is obtained by pooling the variance estimates from each

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

runcharacterization 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.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

---------------------- 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 .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

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

the interval from the centre of the sampling period to the time the pulse occurs. Then the variance CV associated

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 %,

---------------------- 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 efficiency5.5.1 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 diffusionTa 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

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 1999All 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 StandardizationCase postale 56 • CH-1211 Genève 20 • Switzerland

Internet iso@iso.ch

Printed in Switzerland

---------------------- 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.---------------------- 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 referenceThe 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.---------------------- Page: 7 ----------------------

SIST ISO 16107:2002

© ISO

ISO 16107:1999(E)

3.1

Busch probabilistic accuracy

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,

estobtained 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 percentageCV estimated coefficient of variation, expressed as a percentage

est

CV coefficient of variation characterizing inter-run chamber variability, expressed as a percentage

runCV intersampler imprecision (relative to the reference concentration), expressed as a percentage

CV estimated intersampler imprecision CV , expressed as a percentages est s

CV pulse-induced imprecision, expressed as a percentage

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

0,95t-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)

Dbias relative to concentration c

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

SIST ISO 16107:2002

© ISO

ISO 16107:1999(E)

Destimated bias D

est

Dbias associated with concentration pulse

D95 % confidence limit on the bias D

95 %

ndegrees of freedom in determining CV

neffective number of degrees of freedom in determining CV

eff

sassumed concentration variability

sassumed humidity variability

sassumed temperature variability

sassumed ambient wind speed variability

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

intended sampler use, as well as at a reduced concentration in the range c /10 to c /2. Pressure effects result in

0 0one-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

estmanufacturer'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 cexperiment 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 0of intersampler imprecision (characterized by CV ) together with an inter-run chamber variability (CV ) resulting in

s runpart from uncertainty in the reference concentration. CV is obtained by pooling the variance estimates from each

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

runcharacterization 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.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

---------------------- 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 .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

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

the interval from the centre of the sampling period to the time the pulse occurs. Then the variance CV associated

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 %,

---------------------- 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**...**

## Questions, Comments and Discussion

## Ask us and Technical Secretary will try to provide an answer. You can facilitate discussion about the standard in here.