Tiêu chuẩn iso tr 18637 2016

S ientif ic an t ech ical methodolo ies used t o set ex osur l mit ma dife from one entity t o anothe , w hich can lead t o disp rities in worke prot ection from cou try t o country [14]

Nanotechnolog ies — Overview

Na ote h olo ie s — Vue d’ensemble de s cadre s dispo ible s p ur la

définition de limite s et b nde s d’e xp sition pro e ssion e lle a plica le s

a x n n -o jets,à leurs a ré at et a glomérat (NO AA)

Refer ence n mb r

ISO/TR 1 637:2 16(E)

Fir t edition

2 16-1 -0

COPYRIGHT PROTECTED DOCUMENT

© ISO 2016, P blshed in Sw itz rlan

A ll rig hts r eserved Unles otherw ise spe ified, nopar of this p blc tion ma y be r epr od c d or utilz d otherw ise in an form

or b an me ns, ele tr onic or me hanic l, inclu in p oto opying , or postin on the internet or an intranet , w ithout prior

written permis ion Permis ion c n be req esed from either ISO at the ad r es below or ISO’s member bod y in the c u try of

F reword i v

Introduction v

1 Sc ope 1

2 Nor mati ve r eferenc es 1

3 Terms an definitions 1

4 Symbols and abbreviated terms 3

5 Des riptio of avaiable proc es es for set ing OELs and OEBs 5

5.1 General considerations 5

5.2 Desc iption of evidenc -b sed pr oc s 6

5.3 Subs anc -spe ific OELs 8

5.4 Categ orical OELs 8

5.5 Initial or default oc up tional ex osur e b n s 9

6 Substanc e-specific OELs for nanomater ials 10

6.1 General o e view 1

6.2 Av ia le subs anc -spe ific OELs 1

6.2.1 C r bon nanotubes 1

6.2.2 Nanoscale TiO 2

1 6.2.3 Fuler enes 1

6.3 Ev luation of OEL methods 1

6.3.1 Simiarities an diferenc s 1

6.3.2 Influenc of methods on de ived OEL v lues for nanomate ials 1

6.3.3 State of the scienc in supp rt of risk as es ment methods for nanomate ials OELs 14

7 Categ or ical OE s for nanomaterials 15

7.1 Summary of o tions pr op sed 1

7.1.1 United King dom 1

7.1.2 Ge many 1

7.1.3 NIOSH .1

7.1.4 Ja an ’s (A IST’s) a pr oaches 1

7.1.5 OECD 1

7.2 Ev luation of categ orical OEL 1

7.2.1 Simiarities an diferenc s 1

7.2.2 State of the scienc sup orting categ orical OELs 2

8 OEBs and c ontr ol banding for nanomater ials 21 8.1 Ove view of current hazar d an contr ol b n ing schemes 2

8.1.1 C mp rison of hazar d ban s an OEBs as a pled to in aled NOA A s

2 8.1.2 ISO hazar d b n ing scheme for NOA A s 2

8.2 Case s u ies on b nding NOA A s 2

8.3 Ev luation of the evidenc for initial (default) OE s for categ ories of NOA A s 2

8.3.1 C teg orical analyses an r ead-acr os 2

8.3.2 Uti ty of in vitr data in OEL OE develo ment for NOA A s 2

8.3.3 Options for de iving an OEL or OE for NOA A s 3

9 Fe sibiity c onsiderations in the OEL and OEB set ing pr oc es 30 A nne x A (informative)Stan ar d proc es es for OE set ing 32 Biblog raphy 62

ISO (he Int ernational Org nization for Stan ardization) is a worldwidefede ation of national s an ards

b dies (ISO membe b dies) The work of pr p ring Int ernational Stan ards is normaly car ied out

through ISO t ech ical committ ees Each membe b dy int er st ed in a subje t for w hich a t ech ical

committ ee has be n es a lshed has the right t o be r pr sent ed on that committ ee Int ernational

org nizations, g ove nmental an non-g ove nmental, in laison with ISO, also take part in the work

ISO cola orat es closely with the Int ernational Ele trot ech ical C mmis ion (IEC) on al matt ers of

ele trot ech ical s an ardization

The proc d r s used t o develo this document an those int en ed for it furthe maint enanc ar

desc ibed in the ISO/IEC Dir ctives, Part 1 In p rticular, the dife ent a pro al c it eria ne ded for the

dife ent ty es of ISO document should be not ed This document was draft ed in ac ordanc with the

edit orial rules of the ISO/IEC Dir ctives, Part 2 ( e www.iso.org dir ctives)

A tt ention is drawn t o the p s ibi ity that some of the element of this document ma be the subje t of

p t ent right ISO shal not be held r sponsible for identifying any or al such p t ent right Detais of

any p t ent right identified d ring the develo ment of the document wi be in the Introd ction an / r

on the ISO ls of p t ent de larations r c ived ( e www.iso.org p t ent )

Any trade name used in this document is information given for the convenienc of use s an does not

cons itut e an en orsement

For an ex lanation on the meaning of ISO spe ific t erms an ex r s ions r lated to conformity as es ment,

as wel as information a out ISO’s adhe enc t o the Wor ld Trade Org nization (WTO ) principles in the

Te h ical Bar ie s to Trade (TBT) se the folowing URL: www.iso.org/ iso/for word.html

The committ ee r sp nsible for thisdocument is ISO/TC2 9, Na ote h olo ies

Nano-o je t an their a gr g t es an a glome at es (NOAAs) r pr sent a subset of p rticulat e mat erials

that can be dispe sed in the air an can r pr sent health risks via in alation ex osur s NOAAs inclu e

s ructur s with one, two or thr e ext ernal dimensions in the nanoscale from a pro imat ely 1 nm t o

1 0 nm, w hich ma be sphe es, fibr s, tubes an othe s as primary s ructur s NOAAs can consis of

in ivid al primary s ructur s in the nanoscale an a gr g t ed or a glome at ed s ructur s, inclu ing

those with siz s larg er than 1 0 nm An a gr g t e comprises s rongly b n ed or fused p rticles

( tructur s) An a glome at e is a cole tion of weakly b u d p rticles ( tructur s)

[1][2][3][4]

The purp se of this document is t o desc ibe a g ene al framework for the develo ment of oc up tional

ex osur lmit (OELs) or oc up tional ex osur b n s (OEBs) for in ivid al NOAAs or cat eg ories of

NOAAs with dife ent levels of a aia le data OELs an OEBs ar important t ools in the pr vention

of oc up tional i lnes OELs ha e a long hist ory in in us rial hygiene an ar based on o se v tions

of workers or s u ies of la orat ory animals OELs ar es a l shed t o minimiz the lkelho d of

adve se efe t from ex osur t o pot ential y hazardous subs anc s in the workplac

[5][6]

An OEL is

g ene aly subs anc -spe ific (although sometimes g ene ical y ex r s ed, such as d s ) Sufficient data

t o develo an OEL ma not be a aia le, espe ial y for subs anc s such as NOAAs used in eme ging

t ech olo ies To aid in hazard communication an ex osur control de isions for subs anc s without

OELs, hazard b n ing has be n used for many years

[7][8][9]

Subs anc s ar as igned t o a hazard b n

b sed on lmit ed t oxicity data usualy from animal s u ies Hazard ban ing schemes ty ical y consis of

q altative b n s ranging from low t o high seve ity of efe t Th s, a hazard b nd r pr sent a rang e of

pot ential t oxicities for a particular subs anc or cat eg ory of subs anc s Some hazard b n ing schemes

inclu e as ociat ed OEBs

[1 ]

The t erm OEB is a g ene al t erm for ex osur conc ntration rang es used

in some hazard ban ing schemes that ar r lat ed t o the rang es of hazard p t entials In contras t o an

OE , an ex osur b n is a rang e of p t ential conc ntrations of a subs anc (or cat eg ory of subs anc s)

t o w hich workers ma be ex osed in a defined oc up tional sc nario an w hich is b sed on fact ors

such as the amou t of NOAA proc s ed or used, the natur of the proc s , an the form of the NOAA

inclu ing d s ines

[3]

In control b n ing, the hazard b n an the ex osur b n ar combined t o

det ermine the control b n for any p rticular oc up tional sc nario (e.g ISO/TS 1 9 1- 2)

OELs an OE s ar p rt of an o e al oc upational safety an health (OSH) pro ram an ar not

int en ed t o identify an ad r s al safety an health risks as ociat ed with a spe if ic proc s or task

OELs an OEBs ar int ended t o pro ide oc up tional safety an health profes ionals with a health

b sis for as es ing the efe tivenes of ex osur controls an othe risk manag ement practic s The

ex osur as es ment of nanomat erials inclu ing carb n nanomat erials [such as fule ene, gra hene,

single-waled carb n nanotube ( SWCNTs)an multi-wal ed carb n nanotube (MWCNTs)], metal o ides

(TiO

2, SiO

2, zinc o ide, iron o ide), an metals ( i ve an g old nano articles) r mains a chaleng e in

the f ield of oc up tional hy giene, as the e ha e be n r latively few s u ies on the charact erization of

workplac ex osur s t o NOAA Sampl ng an analytical methods that ha e the ca a i ties t o ac urat ely

measur nanomat erials ar s i un e develo ment Mos samplng devic s that measur airborne

particle cou t conc ntrations, such as con ensation p rticle count ers an o tical p rticle count ers,

cannot dife entiat e ambient ex osur s t o backgroun nano articles from NOAA in the workplac

environment Airb rne measur ment of carbon nanotubes (CNTs) an carb n nanof ibr s (CNF s) using

mo i ity p rticle siz rs also sometimes could pr sent a uniq e chaleng e d e t o the ar ing caused b

the charg ed airb rne CNT an CNF a glome at es in the dife ential mo i ity analy se

[1 ]

Although

seve al groups ha e att empt ed t o measur an cou t CNT s ructur s using transmis ion ele tron

mic osco y or othe mic osco ic methods

[1 ][1 ]

the e ar s i l no s an ard methods for measuring

an cou ting CNT s ructur s In ad ition, det ermining the mas conc ntration of CNTs an CNF s b sed

on measuring the elemental carb n (EC) r mains a chaleng e d e t o othe sour es of elemental carb n

in the workplac , such as org nic comp sit e mat erials an air an diesel p lution that could int erfe e

in the det ermination of CNT an CNF ex osur s

S ientif ic an t ech ical methodolo ies used t o set ex osur l mit ma dife from one entity t o

anothe , w hich can lead t o disp rities in worke prot ection from cou try t o country

[14]

The efor ,

harmonizing the scientif ic methodolo ies used in develo ing OELs, inclu ing using the bes a aia le

evidenc for int erspe ies extra olation an spe ifying the ty e of data an unc rtainties inv lved in

the OEL det ermination is ne es ary for a ro us health an safety ev luation framework for NOAAs

This document pro ides a col a orative, scienc -b sed platorm t o desc ibe an ev luat e the s at

e-of-the-art in such data and methods

Cur ent risk as es ment methods ar l kely t o a ply t o NOAAs

[1 ]

although the l mit ed health hazard

data for many NOAAs an the conside a le v riety in the ty es of man factur d NOAAs pr sent a

chaleng e t o the efficient develo ment of OELs for individ al NOAAs To dat e, few OELs an OEBs have

be n develo ed for spe ific NOAAs an none ha e be n formal y r gulat ed b a g ove nment ag ency

Stan ard OEL and OEB methodolo ies for NOAAs ar ne ded t o ev luat e the evidenc on the hazard

pot ential of NOAAs in the workplac t o pro ide a health b sis for risk manag ement de isions, inclu ing

sele tion and ev luation of engine ring control o tions One of the g oals of this document is t o identify

b th the simi arities an dife enc s in the methods used t o develo OELs This ev luation ma lead t o

impro ement in methods for set ing ex osur l mit or b n s

This document pr sent an o e view of the s at e-of-the-art in the develo ment of OELs an OEBs for

NOAAs Cur ent a pro ches for as igning default hazard b n s in the a senc of NOAA-spe if ic t oxicity

data ar desc ibed These a pro ches bui d on cur ent hazard an control b n ing s rat egies, such as

those develo ed in ISO/TS 1 9 1- 2 The cur ent s at e of the methods an data t o develo OELs an

OE s for NOAAs is desc ibed in this document, along with an ev aluation of those methods used in

develo ing thecur ent OELs for NOAAs C t eg orical a pro chest o de ive OE s for NOAAs with l mit ed

data ar also discus ed, such as those b sed on biolo ical mode-of-action (MOA) an phy sico-chemical

(PC) pro e ties The b sis for the framework desc ibed in this document is the U.S NIOSH Cur ent

Int el g enc Buletin Ap ro ches to Develo in Oc c up tio al Ex pos ur Limits or B ands fr En ine r d

Na omate ial s

[16]

This document also takes int o conside ation othe s at e-of-the-scienc r port ,

inclu ing output of the worksho “ Strat egies for S t ing Oc upational Ex osur Limit for E gine r d

Nanomat erials,” w hich was held on S pt embe 1 -1 , 2 1 in Washingt on, DC, USA

[6]

an the OECD

Working Party on Man factur d Nanomat erials Ex e t Me ting on C t eg orization of Man factur d

Nanomat erials, S pt embe 17-1 , 2 14

[17]

The primary targ et au ienc of this document is oc up tional safety an health profes ionals in

g ove nment, in us ry, an academia, w ho ha e the ex e tise t o develo OELs or OEBs b sed on the

guidanc in this document In ad ition, the evidenc -b sed a pro ch desc ibed in this document

ma be useful in the ev luation an / r ve ification of cur ent hazard an control b n ing schemes

an for identifying the key data g ps C ntrol b n ing r q ir s information on b th the a pl ca le

hazard cat eg ory an ex osur cat eg ory Ap ro riat ely ve if ied control b nding t ools would be bro dly

useful, as these t ools r q ir les spe ialz d ex e tise and r sour es ( han for a compr hensive risk

as es ment) an ar ac es ible t o a wide group of in ivid als an smal busines es The efor , this

document can be conside ed complementary t o ISO/TS 1 9 1- 2 on control b n ing for nanomat erials

as it desc ibes the s at e-of-the-art in the proc s of as igning nanomat erials t o hazard b n s /OE s

w hen the scientif ic evidenc is not uff icient t o develo an in ivid al OEL

Some of the cit ed methods lead t o r sult that ar not ne es ari y consist ent an this ma be d e t o

method sele tion biases of the authors In these cases, diverse r sult wi l also make it diff icult t o use

information t o con dently es a l sh ex osur an b n levels It is bey n the sco e of this document

t o att empt t o identify the methods w hich lead t o b th cor e t an consist ent r sult In the event that

methods lead t o dive se r sult , it is ho ed that his r p rt wi l lead t o ad itional methods develo ment

that wi lead t o impro ement an that these impro ement can be r l ed on for set ing ex osur an

b n ing levels

Theo je tives ofthis document inclu e

a) desc ibing an evidenc -b sed s at e-of-the-art ramework t o develo OELsor OE s for man factur d

NOAAs, an

b) ex mining the cur ently a ai a le data an othe a pro ches an methods used (e.g benchmark

subs anc s and benchmark ex osur levels) in the oc up tional risk manag ement de ision-making

for NOAAs

It is anticip t ed that this document wi contribut e t o the develo ment of s an ard hazard an risk

as es ment methods an faci tat e the sy st ematic ev luation of the pot ential health risk of oc up tional

Nanotechnolog ies — Overview ofavailable framewor ks

This document pro ides an overview of a ai a le methods an proc d r s for the develo ment of

oc up tional ex osur l mit (OELs) an oc up tional ex osur b n s (OE s) for man factur d

nano-o je t an their a gr g t es and a glome at es (NOAAs) for use in oc upational health risk manag ement

de ision-making

2 Normati ve r eferences

The e ar no normative r fe enc s in this document

3 Terms and definitions

F r the purp ses of this document, the te ms an def initions given in ISO/TS 80 04-2 an the

folowing a ply

ISO an IEC maintain t erminolo ical data ases for use in s an ardization at the folowing ad r s es:

— IECEle tro edia: a aia le at ht p:/ www.ele tro edia.org

— ISO Onlne brow sing plat orm:a ai a le at ht p:/ www.iso.org o p

3.1

a glomerate

cole tion of weakly or medium s rongly b u d p rticles w he e the r sulting ext ernal surfac ar a is

simi ar t o the sum of the surfac ar as of the in ivid al comp nent

Note 1 to entry: T e for es holding a glomerates t og ther are weak for es, for e ample, van der Wa ls for es or

simple p ysicalentanglement

Note 2 to entry: Ag lomerates are also t ermed secon ary particles an the original sour e particles are termed

primary particles

[ SOURCE:ISO 2 8 4:2 1 , 1.2]

3.2

a gre ate

p rticle comprising s rongly b n ed or fused p rticles w he e the r sulting ext ernal surfac ar a is

signif icantly smale than thesum of surfac ar as of the in ivid al comp nent

Note 1to entry: The for es holding an a gre ate t og ther are strong for es, for e ample,covalent or ionic b n s,

or those resulting from sintering or comple p ysical entanglement, or otherwise combined former primary

particles

Note 2 to entry: Ag re ates are also termed secon ary particles an the original sour e particles are termed

primary particles

[ SOURCE:ISO/TS8 0 4- 2:2 1 , 3.5]

conta ct with a chemical, physical or biolo ical agent b sw alowing, br athing, or touching the skin or yes

Not e 1t o entry: Exposure can b short-term (acute e posure), of int ermediate d ration,or long -term (chronic)

nano-o je t with two ext ernal dimensions in the nanoscale an the third dimension signif icantly larg er

Not e 1t o entry: T e larg st e ternal dimension is not neces arily in the nanos ale

Not e 2t o entry: T e terms nanof ibril an nanof ilament can also b used

Not e 3t o entry: S e 3.9 Not e 1 t o entry

[ SOURCE:ISO/TS8 0 4- 2:2 1 , 4.5]

3.8

nano-o ject

disc et e pie e of mat erial with one, two or thr e ext ernal dimensions in the nanoscale

Not e 1 t o entry: T e secon an third e t ernal dimensions are orthog onal t o the fir t dimension an t o e ch other

[ SOURCE:ISO/TS8 0 4-1:2 1 , 2.2]

3 9

nanoparticle

nano-o je t with al ext ernal dimensions in the nanoscale w he e the lengths of the long es and the

short es ax es of the nano-o je t do not dife significantly

Not e 1 t o entry: If the dimensions difer signif icantly ( y ical y by more than 3 times), t erms such as nanof ibre or

nanoplate may b prefer ed t o the term nanoparticle

[ SOURCE:ISO/TS8 0 4- 2:2 1 , 4.4 ]

3.10

nanos ale

length rang e a pro imat ely from 1 nm t o 1 0 nm

Not e 1 to entry: Properties that are not e trapolations from a larg r size are predominantly e hibited in this

length rang

[ SOURCE:ISO/TS8 0 4-1:2 1 , 2.1]

particle

min t e pie e of matt er with def ined phy sical b un aries

Note 1t o entry: A p ysical b u dary can also b des rib d as an interface

Note 2t o entry: A particle can move as a u it

Note 3t o entry: T is g neral particle def inition ap lies t o nano-o jects

oc u atio al ex osure l mit

ma imum conc ntration of airb rne contaminant de med t o be ac epta le, as def ined b the a uthority

ha ing jurisdiction

[ SOURCE:ISO 1 9 2:2 1 , 3.1 3]

3.1

oc u atio al ex osure band

q antitative r pr sentation of hazard b n w hich desc ibes hazard p t ential of a p rticular mat erial

or clas of mat erials in workplac air

3.1

bre thing zo e

sp c arou d the fac of a worker from w he e he or she takes his or he br ath

[ SOURCE:ISO 240 5:2 0 , 3.1.2.1]

4 Symbols and abbreviated terms

ACGIH Ame ican C nfe enc of Go e nmental In us rial Hy gienis s

AGS A us ch s für Gefahrst ofe(Ge man Committ ee on Hazardous Subs anc s)

AGW Arbeit platzgr nzwert (oc upational ex osur lmit)

AIST Ja aneseNational Ins itut e of Adv nc d In us rial S ienc an Te h olo y

BAL broncho lveolar la ag e fluid

BA uA Bu desans alt für Arbeit sch tz u d Arbeit medizin (Ge man F ede al Ins itut e for Oc

u-p tional Safety an Health)

BMDL benchmark dose es imat e, 9 % lower con denc l mit

BSI British Stan ards Ins itution

CMAR car inog enic, mutag enic, as hmag enic, or r prod ctive t oxicant

CNF carbon nanof ibr

DFG Deut che F orsch ngsg emeinschaf (Ge man R esear h F oun ation)

DMEL de ived minimum ex osur level

DNEL de ived no-efe t level

EPA Unit ed Stat es Environmental Prot ection Ag ency

EU-OSHA E ro ean Ag ency for Safety an Health at Work

GBP gran lar bio e sist ent particle

GHS Glo aly Harmoniz d Sy st em of Clas ification an L bel ing of Chemicals

IARC Int ernational Ag ency for R esear h on Canc r

IFA Ins itut für Arbeit sch tz (Ge man Ins itut e for Oc up tional Safety an Health)

ILV in icative lmit v lue

JSOH Ja an Society for Oc up tional Health

L 5 conc ntration as ociat ed with 5 % lethal ty

LOAEL lowes o se ved adve se efe t level

MAK Ma imale Arbeit platzkonz ntration (ma imum workplac conc ntration)

MOA biolo ical mode of action

MOEL K or an Minis ry of Emplo ment an L b ur

MSHA Unit ed Stat es Mine Safety an Health Adminis ration

MWCNT multi-waled carb n nanotube

NIOSH Unit ed Stat es National Ins itut e for Oc up tional Safety an Health

NOAAs nano-o je t , an their a gr g t es an a glome at es inclu ing those larg er than 1 0 nm

NOAEL no o se ved adve se efe t level

OECD Org nization for Economic Co pe ation an Develo ment

OEL oc upational ex osur lmit

OEL (PL) pe iod-lmit ed oc upational ex osur lmit

OELV oc upational ex osur lmit v lue

OSH oc upational safety an health

OSHA Unit ed Stat es Oc up tional Safety an Health A dminis ration

P L pe mis ible ex osur lmit

QRA q antitative risk as es ment

REACH R egulation, Ev luation, A uthorization an R es riction of Chemicals

SCENIHR S ientific C mmitt ee on Eme ging an New ly Identif ied Health Risks

SCOEL S ientific C mmitt ee on Oc up tional Ex osur Limit

STEL short-t erm ex osur l mit

STOT-SE Spe if ic targ et org n t oxicity — single ex osur

STOT-RE Spe if ic targ et org n t oxicity — r peat ed ex osur

SWCNT single-waled carb n nanotube

TLV thr shold lmit v lue

TSCA To ic Subs anc s C ntrol A ct

TWA time-weight ed a e ag e

VLE Valeur Limit e d’Ex osition Profes ion ele (oc up tional ex osur l mit)

WHO World Health Org nization

5 Desc iption of avai able proc es es for set ing OELs and OEBs

5.1 General c onsiderations

Ex osur t o subs anc s or mixtur s in the workplac can oc ur through in alation, a sorption through

the skin or ing es ion Mos ex osur oc urs through the inhalation of v p urs, d s s, fumes or g ses

F or some chemicals, a sorption through the skin ma also be a signif icant sour e of ex osur

The r sp nse of the b dy t o ex osur from subs anc s an mixtur s depen s on the natur of the

subs anc , the health efe t it can cause an the amount of the subs anc or mixtur a sorbed b the

b dy Individ als also ha e dife ing a i ties t o meta ol z chemicals w hich can cause conside a le

v riation in the t oxic efe t between peo le The ext ent t o w hich a pe son is ex osed mainly depen s

on the conc ntration of the subs anc or mixtur in the air an the amou t of time ex osed an , of

course, on the efe tivenes of controls Subs anc s an mixtur s ma cause immediat e acut e health

efe t or it ma be de ades befor efe t on the b dy be ome evident

Oc up tional ex osur l mit ar int en ed t o pr vent adve se health efe t in “ nearly al workers”

even with r peat ed or dai y ex osur s o e a working lfetime Some OELs ar b sed on health efe t

data only (e.g ACGIH TLV), an othe OELs also inclu e conside ation of the t ech olo ical feasibi ty

(e.g NIOSHRELs) or e onomic feasibi ity (e.g OSHA P L) of measuring an control ing ex osur s

F or a few subs anc s, usual y the mor p t ent pro a le an es a lshed h man car inog ens, it is not

cur ently p s ible t o as ign an a pro riat e ex osur lmit F or these subs anc s, ex osur should be

controled t o the lowes practica le level Biolo ical monit oring ma pro ide a mor r l a le in ication

of workplac ex osur for these subs anc s

The ev luation of hazards p sed b atmosphe ic contaminant in the working environment is oft en

a complex task, taking int o ac ount the pot ential y larg e v ria i ity of ex osur at the workplac

r q iring sou d oc up tional hy giene ex osur as es ment s rat egies F or this r ason, it is es ential

that hose pe sons r sp nsible for such as es ment ar know ledg ea le an ex e ienc d profes ionals,

w ho ar fuly awar of al is ues canv s ed in this document an ha e a pro riat e q al fications and

ex e ienc in oc upational hygiene

NOTE A k owledg a le an e perienced profes ional is an in ivid al who wil properly perform a specif ic

jo This per on utilizes a combination of k owledg , skils an b haviour to improve performance More

g eneraly, competence is the state or q ality of b ing adeq ately or wel q alif ied, having the a ility t o perform a

specif ic role

[3]

The r lationship between v rious ex osur l mit should not be used as a g ene al measur of their

r lative t oxicity This is be a use, among othe things, the v lues for dife ent subs anc s ar oft en

es a lshed with r g rd t o dife ent biolo ical efe t , such as ir itation or sy st emic t oxicity Simiarly,

the ex osur lmit should not be used as a basis for the ev luation of commu ity air q alty, or for long

t erm, non-oc upational ex osur s

Mos subs anc s used in in us ry ha e not be n as igned ex osur lmit This does not imply that

these subs anc s ar safe or non-hazardous In many cases the e is insuff icient information on the

health efe t of these u lst ed subs anc s t o alow national r gulat ory b dies t o as ign an ex osur

l mit, even on a t entative b sis In othe ins anc s, the use of the subs anc does not lead t o significant

airb rne levels of contaminant, or it use is so r s rict ed that an ex osur l mit is not war ant ed

It is a g ood g ene al p lcy t o keep the ex osur t o any subs anc as low as is practica le, ir espe tive of

w hethe pr sent information in icat es it is hazardous or not Some subs anc s pr viously thought t o be

comp ratively safe ha e subseq ently be n fou d t o p se se ious long t erm health risks

The e ar thr e ty es of ex osur l mit :

— time-weight ed a e ag e (TWA)l mit ;

— short t erm ex osur lmit ( STEL);

— peak or c i ing l mit

These l mit an othe t echnical aspe t of set ing OELs ar furthe desc ibed in A.1.2

5.2 Des r iption of evidenc e-based proc es

The methods for develo ing OELs depen on the a ai a le data S h lt e, et al

[5]

desc ibe thr e g ene al

sc narios for v rying amount of t oxicolo ical data This framework was r f ined t o desc ibe lnkag es

betwe n the evidenc b sis for these g ene al cat eg ories through benchmark subs anc s Benchmark

subs anc s ar wel -charact eriz d mat erials (e.g airb rne p rticles or f ibr s) with suff icient

dose-r sp nse data from animal and/ r h man s u ies t o develo q antitative risk es imat es an

health-b sed OELs (Figur 1)

[1 ][2 ]

Benchmark mat erials also pro ide a r fe enc (e.g as a p sitive or

neg tive control) in comp rative t oxicity as ay s with new NOAAs that ha e lmit ed t oxicolo ical data

but simi ar phy sico-chemical pro e ties an infe r d biolo ical mode-of-action (MOA)

[1 ][2 ][2 ]

The

focus of this document is on oc upational airb rne ex osur s t o nanomat erials sinc in alation is the

major rout e of ex osur t o p t ential y hazardoussubs anc s, inclu ing NOAAs, in the workplac

As shown in Figur 1, in the firs case, if dose-r sp nse data ar sufficient, an OEL for an in ivid al NOAA

can be develo ed using q antitative risk as es ment (QRA) The definition of sufficient wi l ultimat ely

be b sed on a ju gment a out the av aia le data, an ma inclu e weight of evidenc ev aluations,

inclu ing the a aia i ty of adeq at e data for benchmar k dose model ing

[2 ]

or no o se ved adve se

efe t levels (NOAELs) or lowes o se ved adve se efe t levels (LOAEL) from wel-con uct ed s u ies

S con , if data ar insuff icient for QRA for a spe ific subs anc , but adeq at e information is av aia le

on a simiar subs anc in the same mode-of-action cat eg ory, then a cat eg orical OEL ma be as igned

b q altative or q antitative methods inclu ing r ad-ac os an s ructur -bio ctivity model ng, with

comp risons betwe n NOAAs an benchmark subs anc s Third, if data ar insufficient t o develo a

subs anc -spe ific or cat eg orical OEL, then initial (defa ult) hazard an control b n s ma be de ived

b comparing NOAA pro e ties t o that of simiar mat erials in bro d cat eg ories The o je tive of this

evidenc -b sed a pro ch is t o faci tat e de ision-making a out ex osur control s rat egies for NOAAs

in the workplac b sed on bes a ai a le evidenc The framewor k alow s for it eration an r vision of an

OE or OEL as new data be ome a aia le b sed on s an ard c it eria for data an methods A t this time,

mor ex mples of OELs develo ed for NOAAs ar a aia le than of cat eg orical OELs or OE sfor NOAAs

The data a ai a le for develo ing OELsor OE sfor NOAAs ma inclu e

a) data from in v i vo an in vitro t es ing of spe if ic NOAAs (e.g from the OECD t es ing pro ram,

man factur rs of NOAAs, an non-r gulat ory g ove nment ag encies such as the NIOSH an the NTP

in the US), an

b) exis ing t oxicolo y or epidemiolo y s u ies of lu g efe t from in aled particles an f ibr s for

comp rative t oxicity analy ses

Gene al chemical hazard data ases (e.g as used in GHS

[2 ]

hazard clas ification) ar also a aia le for

some ofthe p r nt or bulk mat erials with simi ar chemical comp sition t o the NOAA for use in hazard

b n /OEB alocation and control b n ing (e.g se ISO/TS 1 9 1- 2) Ta le 1 summariz s the ty e of

data an methods ne ded t o develo OELs or OEBs

Figure 1 — E idence-based strategy to develo ex osure co trol l mits and bands for NOAAs,

based o level of evidence

Table 1 — Data and meth ds ne ded to develo ex osure l mits or bands

Guidanc v lue Level of evidenc Data, analy sis t ools an meth ds

Substance-specif ic OE Suff icient Substance-specif ic dose-response data for q antitative

risk as es men availa ility of substance-specif ic

sam-pling an analytical method

Cat eg rical OE Limit ed (focused) Comparative to icity, clust ering an cate orization

t o estimate hazard or risk b sed on p y sico-chemical

properties an biolo ical mode-of-action data

OE Minimal or inadeq at e Analo y; default hazard cate ories an e posure

con-trol options are ap lied

5.3 Substanc e-specific OELs

The subs anc -spe if ic OELs ty icaly do not take sep rat e ac ou t of the nano article siz , although

some of these OELs do spe ify the p rticle siz samplng c it eria as ociat ed with r gional r spirat ory

tract deposition These sampl ng c it eria inclu e in ala le (t otal), thoracic (airwa s), an r spira le

(pulmonary) siz fractions Nano articles ar ca a le of depositing anyw he e in the r spirat ory tract

r gion, inclu ing the pulmonary r gion w he e g s ex chang e takes plac Some of the in ivid al OELs

ar spe if ic t o the d s an / r fume forms, an fumes b natur consis of nanos ructur d particles

The OELs for fumes ma be lowe mas conc ntrations than the OELs for d s of the same chemical

subs anc (e.g the NIOSH REL an OSHA P L for co pe is 1 mg m

) It isr lev nt t o not e that those OELs v ry at leas as much b chemical comp sition as b

desc ipt ors of p rticle siz (d s , sol d p rticles g ene at ed b any me hanical proc s ing of mat erials

such as c ushing, grin ing, an han lng or fume, airb rne dispe sion consis ing of smal sol d p rticles

c eat ed b con ensation from the g seous s at e)

spe ific nanomat erials b non-r gulat ory g ove nment ag encies, comp nies, an nong ove nmental

org nizations To dat e, no r gulat ory s an ards ha e be n cir ulat ed for NOAAs

5.4 Categ or ical OELs

Hist oricaly, many airb rne particulat e mat erials we e r g rded as a “ nuisanc ” or as “low t oxicity”

d s s an cat eg orical OELs, such as a g ene ic inhala le OEL of 1 mg/ m

3

an a r spira le OEL of

4 mg m

3

we e set for many low-t oxicity p orly-soluble d s s inclu ing aluminium o ides, gra hit e,

titanium dio ide and othe s

[2 ]

In Ge many, the DFG MAK commis ion r c ntly r d c d the OEL for

bio e sist ent gran lar particles from 3 mg m

3

t o 0,3 mg m

3

(r spira le fraction), r fle ting conc rns

a out a pos ible car inog enic p t ential for this cat eg ory of subs anc s

[26]

Al these v lues, howeve ,

wer not int en ed for p rticulat e mat erials with spe ific k own inhalation or sy st emic t oxicity

(e.g asbest osan lead, r spe tively)for w hich subs anc -spe if ic OELs we e also det ermined

A dv ntag es of cat eg orical a pro ches inclu e:

— mor efficient use of data;

t o de ive OELs for NOAAs using cat eg orical a pro ches ma inclu e q antitative or q al tative r

ad-ac os

[27]

comparative pot ency analy ses of NOAAs t o benchmark (r fe enc ) p rticles in the same

mode-of-action (MOA) cat eg ory

Othe risk analy sis an cat eg orization a pro ches inclu e b th oc up tional an environmental

comp nent , such as sc e ning t ools of p t ential risks o e theNOAA l fe ycle

[34][3 ]

The multi-c it eria

de ision analy sis (MCDA) a pro ch inclu es ev luation of the risks an benef it with weightings

o tained through ex e t elcitation

[28]

This proc s has be n used t o as ign NOAAs t o q al tative risk

cat eg ories (low, medium, high)

[36]

nong ove nmental org nizations These cat eg ories ar b sed on bro d groups of phy sico-chemical

pro e ties that influenc t oxicity ( oluble, bio e sist ent low t oxicity, bio e sist ent high t oxicity, an

fibr s) The BSI an IFA cat eg ories ar pro isional ex osur lmit b sed on exis ing OELs for p rticles

an f ibr s in these cat eg ories, w hich inclu es in some cases a pr cautionary downward adjus ment for

the nanoscale form The ext ent t o w hich chemical subs anc -spe ific data ar a aia le would alow

r f inement of the cat eg orical OELs t o an in ivid al OEL that ma be mor a pl ca le t o an in ivid al

subs anc

5.5 Initial or default oc cupational ex posure bands

When data ar not suff icient t o develo an in ivid al OEL, hazard b n ing a pro ches ar oft en used

t o faci tat e de ision-making among engine ring control o tions

[5]

C ntrol b n ing ty icaly uti iz s a

matrix a pro ch t o cat eg oriz subs anc s ac ording t o their hazard and ex osur p t ential

The combination of the sele t ed hazard an ex osur b n s det ermines

the control b n an as ociat ed engine ring control o tions Howeve , the uti ty of such an a pro ch

is fr q ently lmit ed b the av ai a i ty of adeq at e t oxicolo ical data for use in hazard as es ment The

a senc of such data makes workplac risk charact erization an the subseq ent sele tion of a pro riat e

control measur s pro lematic Anothe sugg est ed a pro ch is the uti ization of initial default hazard

cat eg ories or OEBs for NOAAs b sed on the physico-chemical pro e ties as ociat ed with point-of-entry

or sy st emic t oxicity, inclu ing p rticle surfac chemis ry an ar a, sha e, diamet er, an solubi ty, as

wel as any evidenc on the mutag enicity, car inog enicity, or r prod ctive t oxicity of the nanomat erial

or par nt mat erial

2 ][ 42][ 44][ 45][4 ]

ISO/TS 1 9 1- 2 also incorporat es a aia le t oxicolo ical information an physico-chemical pro e ties

t o designat e nanomat erials int o hazard b n s In this method, nanomat erials ar grouped int o one

of five inhalation hazard groups (A t o E) ac ording t o inc easing seve ity desc ibed in GHS hazard

clas ification a plca le t o chemicals

— Cat eg ory E ( eve e hazard) has no conc ntration rang es pro ided in ISO/TS 1 90 - 2 an othe

hazard al ocation schemes (8.1)

The de ision lo ic for as igning NOAAs int o these hazard b n s inclu es conside ations of solubi ity,

fibrous natur an hazardous pro e tiesof bulk an analog ous mat erials

[3]

Hazard an control b n ing a pro ches we e develo ed t o faci itat e risk manag ement de ision-making

in smal busines A key r sear h ne d for hazard an control b n ing s rat egies in g ene al, an those

spe if ic t o NOAAs, is ev luation an v ldation of the uti ity of these s rat egies t o pro ide adeq at e

health prot ection t o workers in a v riety of jo s an workplac s A s u y ev luating the g ene al hazard

an control b n ing s rat egies fou d lt le or no margin of safety betwe n worker ex osur s an the

efe t levels in animal s u ies

[4 7]

6 Substanc e-specific OELs for nanomater ials

6.1 General overview

Cur ently, the e ar no spe if ic r gulat ory OELs es a lshed for man factur d NOAAs Int erim or

draf OELs ha e be n pro osed for c rtain NOAAs, inclu ing “benchmark ex osur levels” b sed on

analo y with OELs for othe p rticles or f ibr s

[4 ]

Sinc no epidemiolo y data ar a aia le on adve se

health efe t of ex osur s t o mos man factur d NOAAs, dose-r sponse data from animal s u ies ar

ty icaly used t o es imat e risk in h mans Ex e imental s u ies in animals an in vitro s u ies ar also

used t o ev luat e hazard an u de s an me hanisms of t oxicity

Mor r c ntly, non-r gulat ory OELs for some spe if ic NOAAs ha e be n develo ed ( e Ta le 2) These

inclu e OELs pro osed b g ove nmental ag encies, r sear he s, an prod c rs of spe ific nanomat erials

Table 2 — Examples of OE s pro osed for NOAAs

Nanomat erial

OE

(μg m3

, u les stated other

-wise)

R efe enc

Titanium dio ide (ultraf ine)

6 10a

Period-lmited (1 years ) OE

Dife enc s among OELs for the same or simiar NOAAs in Ta le 2 ar d e t o dife enc s in the data

an / r risk as es ment methods used t o de ive the OEL Some dife enc scould be r lat ed t o dife enc s

in chemical comp sitions an phy sical dimensions of simi ar NOAAs The pro osed OELs v ry b up t o

an orde of ma nitu e or mor for the same or simi ar ty e of NOAAs, in icating that dife enc s in

the data an / r methods used t o de ive the OEL can influenc the b sis for risk manag ement de isions

such as sele tion of engine ring controls Thisi lus rat es the c itical ne d t o develo s an ardiz d risk

as es ment proc d r sthat ar b sed on the bes a aia le scientific evidenc an methodolo ies

6.2 Avai able substanc e-specific OELs

6.2.1 Car bon nanotubes

Carb n nanotubes (CNTs) ar an ex mple of man factur d nanomat erials that ha e be n the subje t of

several r c nt risk as es ment prod cing int erim and v lu tary oc up tional ex osur lmit CNTs

can ha e wide v riations in s ructur , siz , sha e an chemis ry ( inclu ing impurities) afe ting their

of in us ry se t ors, inclu ing cons ruction as conc et e r infor ement, for medical tr atment , as fuel

ad itives, an so on

To dat e, the e was lmit ed information r g rding adve se health efe t in workers using or prod cing

CNT or CNF Few epidemiolo ical s u ies ha e shown that some det ecta le biomarkers we e r lat ed

t o ex osur t o CNT or CNF or othe nano articles

Howeve , the e ar s u ies of animals

ex osed t o CNT an CNF that ar informative in pr dicting p t ential h man health efe t consist ent

with way s in w hich scientis s traditionaly ha e used such data in r commen ing risk manag ement

s rat egies

In 2 1 , NIOSH r commen ed a REL of 1 μg m

3

( 8 h TWA) for CNT an CNF The NIOSH REL for CNT

an CNF is b sed on pr venting the develo ment of earle adve se lu g r sponses of pulmonary

inflammation an fibrosis o e a 45-year working l fetime The REL for CNT an CNF is b sed on

animal data, although sinc no chronic s u ies we e a aia le, the dose-r sp nse data from short-t erm

an subchronic s u ies in rat an mic we e extra olat ed t o h mans Sinc the REL for CNT was set

at the l mit of q antif ication of the samplng an analytical method for measuring airb rne elemental

carb n in the workplac , the 45-year working l fetime ex ces risk es imat es for develo ing early-s ag e

pulmonary inflammation or f ibrosis ex ce d 1 % for some en p int an as umptions

[58]

Methodolo y

used t o de ive this ex osur lmit is desc ibed in Clause 5 an furthe detais can be foun in A.1

In 2 1 , the Ja anese National Ins itut e of A dv nc d In us rial S ienc an Te h olo y (AIST) publshed

a se ies of r p rt containing risk as es ment of seve al nanomat erials:carb n nanotubes

These OELs (PL) we e de ived using inflammation as an

en p int o se ved in sub-acut e inhalation ex osur t es s The working ex osur pe iod emplo ed in

OEL calculations was set at 1 years Furthe detai s a out de iv tion of these OELscan be fou d in A.7

6.2.2 Nanos ale TiO

2

Titanium dio ide, TiO

2, is a noncombus ible, w hit e, c y stal ne, sold, odourles powde It exis s in

seve al c y stal forms of w hich anatase an rutie ar comme cialy the mos r lev nt TiO

2

is used

ext ensively in many comme cial prod ct , inclu ing paint an v rnishes, cosmetics, plas ics, p pe ,

an fo d asan anticaking or w hit ening ag ent

Titanium dio ide is prod c d an used in the workplac in v rying particle siz fractions inclu ing

fine (w hich is defined as al p rticle siz s cole t ed b r spira le p rticle sampl ng) an ultraf ine

or nano articles (def ined as the fraction of r spira le particles with a primary p rticle diamet er

of < ,1 μm [< 0 nm])

In 2 1 , NIOSH r commen ed sep rat e RELs for ultrafine (nanoscale) an fine (mic oscale) titanium

dio ide (TiO

2), i.e 0,3 mg m

TWA r spira le conc ntrations)

[5 ]

These RELs we e set at the 45-year working lfetime ex osur

conc ntration as ociat ed with a 1/ 1 0 0 ex ces risk of lung canc r ( 95 % lowe conf idenc l mit)

es imat ed from nonlnear models of dose-r sponse data from animal chronic in alation bio s a s

of f ine or ultraf ine TiO

2 The rat-b sed risk es imat es we e extra olat ed t o h mans b es imating an

eq iv lent r tained lu g burden aft er a 45-year working l fetime, with adjus ment for the dife enc s

in alveolar surfac ar a in rat an h mans an the use of h man lung dosimetry model ng t o ac ount

for int erspe ies dife enc s in the long t erm clearanc an r t ention of r spira le p rticles F or b th

lu g canc r an inflammation r sp nses, ultrafine (nanoscale) TiO

2was mor p t ent on a mas b sis

than fine (mic oscale) TiO

2, w hich is the r ason for the lowe mas -b sed REL for ultraf ine TiO

2[5 ]

Methodolo y used t o de ive this ex osur l mit is desc ibed in Clause 5an furthe detais can be fou d

In ad ition AIST con uct ed subchronic inhalation s u ies using

rat an de ived OELs b sed on pulmonary inflammation en p int The OEL (PL) was det ermined t o

be 0,6 mg m

3

(r spira le d s , 8 h TWA, 1 -year working pe iod)

[4 ]

Furthe detai s a out de iv tion

of this OEL can be foun in A.7

6.2.3 F l erenes

Fule enes ar a form of sphe ical carb n s ructur s The AIST r p rt

[5 ]

focuses on one ful e ene

conf iguration composed of 6 carbon at oms, C

60 Diamet er of C

wat er an othe p lar solvent , but can be dispe sed in the form of a glome at es betwe n 2 nm an

5 0 nm It is soluble in org nic aromatic solvent

In 2 1 , fule enes we e mos ly used as ad itives for r sins used in glas war , b w ling b l s, a metal

ad itive, a lubricant ad itive, an in cosmetics Futur a plcations ma inclu e r sin ad itives for solar

c l ele trodes an fuel c l ele trodes as wel as pharmac utical-r lat ed raw mat erial such as for drug

del ve y Fule enes ar also g ene at ed as bi-prod ct in any combus ion proc s of carb n-containing

mat erial

Simi ar t o CNTs an TiO

2, the AIST s u y used the inflammation en point an subchronic inhalation

animal data Aft er taking int o ac ount unc rtainty fact ors, the oc up tional ex osur l mit (pe

iod-l mit ed) was set at 0,3 mg/ m

3

Furthe detai s a out de iv tion of this OEL can be foun in A.7

6.3 Evaluation of OEL methods

6.3.1 Simiarities and difer enc es

Pr sently mos OELs for man factur d NOAAs ar develo ed using s an ard risk as es ment methods

inv lving f ive st eps:

a) ev luating the a aia ledata;

b) sele ting the adve se r sponse (non-r versible, clnicaly signif icant);

c) det ermining the c itical dose (e.g NOAEL or BMDL);

d) calculating the h man eq iv lent dose (ac ou ting for spe ies-spe if ic dife enc s that afe t the

targ et tis ue dose, e.g venti ation rat es an p rticle dep sition an clearanc kinetics);

e) det ermining the working l fetime ex osur conc ntration that would r sult in that dose ( inclu ing

conside ation of dep sition, uptake, an clearanc )

[5][6 ]

Use of unc rtainty fact ors is a simple, default a pro ch oft en used t o de ive an OEL from a NOAEL in the

a senc of sufficient r sour es ( information or ex e tise)t o pe form q antitative risk as es ment with

dose-r sp nse model ng an dosimetry-b sed extra olation methods Sinc h man health efe t data

for man factur d nanomat erials ar not a ai a le for risk as es ment at this time, dose-r sp nse data

in animalsar used in therisk as es ment analy ses r viewed in this r p rt

W ithout a glo al y harmoniz d a pro ch t o OEL set ing proc s , OELs es a l shed b dife ent groups

for the same man factur d NOAAs using traditional QRA methods even for the same t oxicity data can

v ry b orde s of ma nitu e F or ex mple, r commen ed OELs for nano-TiO

2inclu e 0,0 7 mg m

Such dife enc s arise from dife enc s in the int erpr tation of the

sup orting t oxicity data, sele tion of health en p int , an use of u c rtainty an modifying fact ors

[6]

Ta le 3 summariz s these dife enc s for r commen ed OELs for TiO

2

Table 3 — Parameter and uncertainty factor used in develo ing OE s for nanos ale TiO

He lth en point P lmonary inflammation P lmonary inflammation L ng tumour

Stu y data reference Bermu ez, et al 2 04

Human (oc u ational)

adjustment of rat efect

level

0,5a

0,9b

Internal dose estimation

9 % lower conf idence

limit estimate (no UF)

Adjusts for the p lmonary deposited dose in rats an h mans, b sed on the respirat ory min te volume

(RMV), time (T), deposition fraction (DF), per u it b d weight BW), in e ch species

b nchmark dose estimat e, 9 % lower conf idence limit BMDL)as ociated with

1/ 10 0 ex ces risk of lu g cancer in rats, b sed on the dose-response model averag e (MA)

NOT The value of 0,0 9 in Ta le 4-5of NIOSH 2 1

[5 ]

is not the cor ect value that was used to derive the

h man BMDL estimate in Ta le 4-6 T e cor ect value is 0,04 , which is the ac elerated bias-cor ect ed b

2

/ 0,4 m2

h man lu g mas dose of ultraf ine TiO

2

of 0,3 g, the TiO

2surface are dose (m

2

) is divided by the specif ic

surface are of ultraf ine TiO

2(48 m2

is the 45-year working lifetime me n airb rne e posure ( 8 h TWA concentration)as oci

at ed with the h man-eq ivalent retained lu g dose at a 1/ 1 0 0 ex ces risk of lu g cancer ( 95 % L L esti

mat e) T e me n airb rne e posure was estimated using the Multiple-Path Particle Deposition (MP D)h man

model

[7 ]

details in R eference [5 ]

6.3.2 Influenc e of methods o deri ved OE values for nanomater ials

As shown in Ta le 3, OEL de iv tion methods for TiO

2inclu e a simple unc rtainty fact or a pro ch;

In the firs two analy ses, a NOAEL for pulmonary

inflammation is sele t ed as the c itical health en p int from a subchronic (1 we ks) inhalation

s u y in rat ; howeve , the int erpr tation of w hich dose is the NOAEL dife ed in the analy ses in

R efe enc s [6 ] an [6 ] an R efe enc s [48] an [49] The third analysis

[5 ]

used lu g tumours as the

c itical health en p int The f irs two analy ses es imat ed a h man-eq iv lent NOAEL, w hie the third

analysis de ived a risk-b sed ex osur conc ntration (as ociat ed with 1/ 1 0 0 ex ces risk)

The f irs analy sis simply adjus s the rat NOAEL b the dife enc in the dai y ex osur d ration

in rat v s workers (6 h v s 8 h) an the dife enc in a h man r s ing v s l ght work venti ation rat e

No dosimetry adjus ment ar made, an u c rtainty fact ors

ar used for int erspe ies, intra-spe ies, an ex osur d ration

[65][6 ]

The se on analy sis inclu es

dosimetry adjus ment t o ac ou t for dife enc s in rat and h man ventiation rat es an p rticle siz

spe ific pulmonary dep sition fraction of the in aled dose; this dose is normal z d t o b dy weight

[4 ]

[49]

No dosimetry adjus ment ar made t o ac ou t for dife enc s in the rat an h man clearanc an

r t ention kinetics of the in aled dose; an an u c rtainty fact or is used for int erspe ies (t oxicokinetic)

extra olation

[4 ][49]

The third analy sis

[5 ]

uses q antitative model ng methods t o desc ibe the dose-r sp nse r lationship

in rat an t o es imat e the h man-eq iv lent working lfetime ex osur conc ntration The

dose-r sp nse model ng a pro ch used t o de ive the REL was a thr e-model a e ag e of nonl near models

inclu ing multis ag e, Weibul, an log -pro it (f it t o the rat nonlnear dose-r sp nse data for lu g

tumours) The v ria i ity in the rat dose-r sp nse data was taken int o ac ount in the confidenc int erv l

es imat es of the c itical dose The dose metric used in these analy ses was the r tained p rticle surfac

ar a dose in the lungs (es imat ed from the measur d mas of TiO

2

an p rticle spe if ic surfac ar a)

folowing chronic in alation ex osur t o eithe fine or ultrafine TiO

2 A h man lung dosimetry model

was used t o es imat e the working lfetime mean ex osur conc ntration as ociat ed with the h man

r tained lu g dose that is eq iv lent t o the c itical lung dose in rat (BMDL es imat e) An as umption

in this analy sis is that h mans would be eq aly sensitivet o the rat at an eq iv lent r tained lu g dose

No u c rtainty fact orswe e used in the de iv tion of this risk-based REL

Int er s ingly, the NIOSH dose-r sp nse model ng also used the same rat pulmonary inflammation

data (R efe enc [6 ]) as used in the firs two analy ses

but dife ent es imat es wer

de ived, r sulting in a h man-eq iv lent mean ex osur of a pro imat ely 0,1 mg/ m

3

, w hich is

the dose as ociat ed with minimal pulmonary inflammation [4 % inc ease in p lymorphon clear

leukocyt es (Ta le 4-3 of NIOSH 2 1

[5 ]

)] This working lfetime ex osur as ociat ed with pulmonary

inflammation is lower than that as ociat ed with the es imat ed 1/ 1 0 0 risk of lu g canc r — w hich is

consist ent with the evidenc that TiO

2

is an in ir ct car inog en acting through pe sist ent pulmonary

inflammation In the third analy sis

[5 ]

a major dife enc in methods compar d t o the firs an se on

analy ses

[4 ][ 49][66][73]

in Ta le 3 is that NIOSH used dosimetry model ing t o ac ou t for dife enc s in

the long t erm p rticle r t ention kineticsin the lungs of rat an h mans The NIOSH REL is int en ed t o

minimiz therisk of lu g canc r in workers ex osed t o TiO

2

o e a ful working l fetime

Despit e the dife enc s in the health en point an methods an as umptions in these thr e risk

as es ment , two of the thr e OELs ar within a fact or of two (0,3 mg m

control ban , an th s the sele tion of workplac

ex osur controls ma be simi ar b sed on each of those analy ses The third OEL ismor than an orde

of ma nitu elowe and would in icat e the ne d t o use tight er containment an control o tions

6.3.3 State of the s ienc e in suppor t of risk as es ment metho s for nanomaterials OELs

Idealy, the bes a ai a le scientific evidenc an methods should be used t o de ive OELs Howeve ,

r sour e l mitations ma pr clu e a ful q antitative risk as es ment in eve y case, even if sufficient

data ar a aia le Mor oft en, data ar lmit ed an so simple default risk as es ment methods wi l be

ne ded for many NOAAs The develo ment of a framework in w hich ful q antitative risk as es ment

ar pe formed for sele t ed benchmark (r fe enc ) particles within biolo ical mode-of-action (MOA)

cat eg ories, combined with simpl fied risk as es ment an comp rative analy ses with NOAAs in the

same cat eg ory, ma faci itat e OEL develo ment for NOAAs Such an a pro ch is desc ibed furthe in

this document Conc ptual cat eg ories ha e be n pro osed as desc ibed in Clause 7, although mor

analy ses involving s ructur -activity r lationships b sed on physico-chemical pro e ties of NOAAs

wi bene ded t o develo a r la le and pr dictive risk as es ment framework

7 Categ orical OELs for nanomaterials

7.1 Summar y of options pr oposed

7.1.1 United King dom

The BSI “Guide t o safe han l ng and disp sal of man factur d nanomat erials”

[4 ]

pro ides simple

pr cautionary risk guidanc in the form of a Publc Document for the develo ment, man factur , an

use of nanomat erials In this document, al nanomat erials ar grouped int o four hazard cat eg ories with

as igned Benchmark Ex osur Levels (BEL) Simi ar t o the IFA r commen ations, BELs ar desc ibed

as “ pra matic guidanc levels only” an ar de ived from OELs for larg er particle forms “on the

as umption that the hazard pot ential of the nano article form is gr at er than the larg e p rticle form.”

Firs , the e is the “f ibrous” cat eg ory, defined as an insoluble nanomat erial with a high aspe t ratio

(ratio > :1 an length > 0 0 nm), w hich is as igned a BEL of 0,0 f/cm

3

(one t enth of the asbest os OEL

pr sc ibed in the US an elsew he e) S con , the e is the “CMAR ” cat eg ory, defined as any nanomat erial

w hich is alr ady clas if ied in it larg er p rticle form as a C r inog enic, Mutag enic, As hmag enic, or

R eprod ctive t oxicant Nanomat erials in the CMAR cat eg ory ar as igned BELs at one t enth of the

mas -b sed OEL for it larg er p rticle form Third, the e is the “insoluble” cat eg ory, def ined as insoluble

or p orly soluble nanomat erials not in the fibrous or CMAR cat eg ory Nano articles in this cat eg ory

ar as igned BELs at one f ift eenth of the mas -b sed OEL for it larg er p rticle form or 2 0 0 p/cm

3

F ourth, the e is a “ oluble” cat eg ory, defined as a soluble nanomat erial not in fibrousor CMAR cat eg ory,

w hich is as igned a B L at one half of themas -b sed OEL for it larg er particle form

7.1.2.1 Federal Ministr y of Labo r and Social A ffair s

In the An ou c ment on Hazardous Subs anc s5 7 “ Man factur d Nanomat erials” from Ma 2 1 b

the F ede al Minis ry of L b ur and Social Afairs

[4 ]

in Ge many, nanomat erials ar grouped int o four

cat eg ories:

a) soluble nanomat erials;

b) bio e sist ent nanomat erials without spe ific t oxicity (gran lar bio e sist ent nano-p rticles –

nano-GBP);

c) bio e sist ent nanomat erials with spe if ic t oxicity;

d) bio e sist ent f ibr -lke nanomat erials

Int er s ingly, conc rning the fir t categ ory, the t erms soluble /insoluble or bio e sist ent ar commonly

used in p rticle t oxicolo y but no ex ct definition or measur ment method for these t erms is at han

The efor as a pro y the solubi ity ofnanomat erials in wat er is used an subs anc s with a solubi ty

les than 1 0 mg l ar coined “ practicaly insoluble” and nanomat erials with solubi ity in wat er gr at er

than 1 0 mg l belong t o the cat eg ory of soluble nanomat erials Ful y awar of the discus ions on the

pos ible en anc d prol fe ation or chang ed pathway s of soluble nanomat erials t o dife ent targ et in

the b dy, the advic has be n given t o pe form the risk as es ment for nanomat erials in this cat eg ory

pe defa ult in tr ating them lke bulk mat erials an negle ting p s ible nano-r lat ed pro e ties

F or the seco d categ ory (nano-GBP), r fe enc was made t o the s u y b Gebel

74]

comparing the

car inog enicity of GBP mic omat erials (mic o-GBP) an nano-GBP in chronic rat inhalation s u ies

Gebel conclu ed that the dife enc in car inog enic p t ency betwe n GBP nanomat erials an GBP

mic omat erials is low and can be desc ibed b a fact or of 2 t o 2,5 r fe ring t o the dose metric mas

conc ntration The s atis ical methods for po l ng data ac os s u ies, such as in the Gebel

7 4]

s u y,

ha e be n discus ed

In July 2 1 , the Pe manent S nat e C mmis ion for the Inves ig tion of Health Hazards of Chemical

Comp u ds in the Work Ar a of the Ge man R esear h F oun ation pro osed a r d c d v lue of

v lue This v lue is int en ed t o pr vent high conc ntrations of these d s s from ha ing a car inog enic

efe t Awar of this pro osal an the on-g oing discus ion, the r commendation is given in the

an ou c ment 5 7 t o cor e t for the sl ghtly highe p t ency of nano-GBP b a plying a fact or of 1 2 on

the cur ent oc up tional ex osur l mit or r spira le d s in Ge many, w hich is 3mg/ m

As this was r g rded t oo high a v lue in t erms of p rticle n mbe conc ntration, it was s at ed that

the REL for nano-GBP should not ex ce d 0,5 mg m

3

for a density of 2 5 0 kg/ m

3

, measur d as the

r spira le d s fraction It is highlght ed that al othe o tions t o uti z REL for the risk manag ement

ar s i l a aia le for the companies They ma use the r commendations given b NIOSH or othe

org nisations/comp nies, use the benchmark levels pro osed b DGUV-IFA or set their own in-house

s an ards

Conc rning the third categ ory (bio e sist ent nanomat erials with a spe if ic t oxicity), r fe enc is made

t o the han lng of the bulk (non-nano) forms of these mat erials W ith r g rd t o exis ing OELs for mos

of these nanomat erials, it is s at ed that comp nies of course ha e t o comply with the exis ing OELs

In Ge many these OELs ar usual y below 0,1 mg m

3

In fact, in the discus ion on de iving ex osur

-risk r lationships an the cor espon ing conc ntrations for car inog enic metals or metal comp u ds,

l ke co alt or nickel, mas conc ntrations in the rang e of 0,0 0 1 mg m

3

t o 0,0 mg m

3

ar pro osed

Announc ment 5 7 s at es that in complying with OELs in this rang eof conc ntrations a s rict r gimeof

control measur s with a high efficacy has t o be emplo ed an in the case of han lng the nanomat erial

a furthe disc imination of control measur s is not feasible

F r mat erials belonging t o the fo rth categ ory (bio e sist ent fibr -lke nanomat erials), thedis inction

is made betwe n bio e sist ent, rigid nanofibr s ad e ing t o the World Health Org nization (WHO)

f ibr paradigm, for w hich one has t o as ume an asbest os-l ke efe t, an bio e sist ent, entangled

or sp ghet i-lke nanofibr s F or the latt er asbest os-lke efe t can only be ex clu ed in the risk

as es ment, if the man factur r or sup le ofthe given nanomat erial can pro ide evidenc that the

nanomat erial does not exhibit asbest os-lke efe t Ove al comp nies ar s rongly discourag ed t o use

bio e sist ent, rigid nanofibr s and a ve y s rict r gime of control measur s has t o be folowed, if these

mat erials ar handled

In conclusion, one can summariz An ou c ment 5 7 on “ Man factur d Nanomat erials” in the wa

that at leas for the time being an for the han lng of the f irs , pas ive g ene ation of nanomat erials

these mat erials ar mos ly tr at ed at the workplac l ke ordinary hazardous subs anc s The only

ex ception being the ev luation of control measur s for nano-GBP, as in thiscase a fact or of 1 2 is a pled

t o the r spira le d s l mit v lues ma be used The An ou c ment wi l be ada t ed if new evidenc

on t oxicolo ical pro e ties of man factur d nanomat erials eme g es or if futur g ene ations of active

nanomat erials fin it wa t o the workplac an p se new hazards

7.1.2.2 Ger man Social Ac cident Insuranc e

In Ge many, the Ins itut e for Oc upational Safety an Health of the Ge man Social A ccident Insuranc

(IFA)

[7 ]

r commen ed the fol owing benchmark l mit t o be used for an 8 h work shif an t o be used

for monit oring the efe tivenes of prot ectivemeasur s in the workplac

in the rang e of measur ment

betwe n 1 nm an 1 0 nm should not be ex ce ded

— F or bio e sist ent gran lar nanomat erials with a density below 6 0 0 kg m

should not be ex ce ded, b sed up n the ex osur risk ratio for asbest os It is

r commen ed that only carb n nanotubes that ha e be n t est ed for adve se health efe t simiar

t o those of asbest os (ac ording t o the man factur r’s de laration) be used

— F or nanoscale lq id p rticles ( uch as fat , hydrocarb ns, sio anes), the a plca le ma imum

workplac lmit (MAK) or workplac lmit (AGW) v lues should be emplo ed owing t o the a senc

of efe t of sold p rticles

These r commen ed benchmark l mit ar g ear d t o minimizing the ex osur in ac ordanc with

the s at e of the art in measur ment Sinc these l mit ar not b sed on o se ved health efe t , a

health risk ma s i exis for worke s, even w he e these r commen ed lmit ar folowed The efor ,

benchmark l mit should not be confused with health-b sed OELs

[7 ]

NIOSH has not formal y pro osed a cat eg orical a pro ch t o develo ing OELs for nanomat erials,

although efort ar un e wa t o ev luat e the a aia le scienc an methodolo ies for develo ing

cat eg orical OELs or OEBs for nanomat erials (desc ibed in R efe enc [2 ]) In some r spe t , the NIOSH

RELs for nanoscale titanium dio ide

[5 ]

an carb n nanotubes an nanofibr s

[58]

ar cat eg orical sinc

these RELs ar int en ed t o a ply t o v rious forms of these subs anc s, w hich ma inclu e v ria i ity in

their hazard p t ential

7.1.4 Japan ’s (A IST’s)approaches

7.1.4.1 Biax ial appr oach

Un e a NEDO proje t P0 041) “ R esear h an Develo ment of Nano article Charact erization Methods” ,

Nakanishi

5 ][79]

ado t ed a method caled the bia ial a pro ch Figur 2 show s a conc ptual dia ram of

the bia ial a pro ch

NOTE S e R eference [49]

Figure 2 — Biaxial approach

The idea was t o f il up the w hole dia ram b con ucting detaied ex minations of the r pr sentative

t es samples along the horizontal a is, an o tained r lative v lues of many nanomat erials through

fairly simple t es s on the ve tical a is On the horizontal a is, in v ivo t es s ( intratracheal ins i lation an

inhalation ex osur t es s) using rat , biokinetics an g ene ex r s ion analy ses we e con uct ed, an the

NOAELs r g rding the in alation ex osur of rat we e comput ed Then, the k ow ledg e ac umulat ed

th s far was used t o extra olat e the ext ent of the efe t on h mans, i.e OELs we e de ived Meanw hie,

along the ve tical a is, Nakanishi, et al

[5 ][79]

con uct ed as simple a t es as p s ible t o fin out the

v lues of the mat erials r lative t o one anothe , w hich was t o det ermine harmful efe t v lues an then

the OEL of v rious mat erials They had initialy conside ed in v itro t es s for the ve tical a is t es s, but

they de ided t o usethe intratracheal ins i lation method inst ead

7.1.4.2 Equi valenc e cr iter ia

The Minis ry of Economy, Trade an In us ry of Ja an launched a f ive-year pro ram for the

“ Develo ment of In o ative methodolo y for Safety As es ment of In us rial Nanomat erials” in

S pt embe 2 1 , w hich aims t o develo fun amental hazard as es ment methodolo y leading t o a

tie ed risk as es ment a pro ch for in us rial nanomat erials The pro ram has two R&D themes:

a) es a l shment of “eq iv lenc c it eria ” of nanomat erials;

b) es a l shment of an intratracheal (IT) adminis ration method as a low-cos an convenient method

for hazard as es ment t o acq ir b sic hazard information, b th of w hich ar for r gulat ory

purp ses

[80]

TheJa anese National Ins itut e of Adv nc d In us rial S ienc an Te hnolo y (AIST) is develo ing the

“eq iv lenc c it eria ” b sed on the data from a set of IT adminis ration t es s using nanomat erials with

dife ent phy sico-chemical pro e tiessuch as siz , surfac ar a, sha e, surfac chemis ry, comp sition,

et c w hen focusing on efe t in the lungs If the t oxicity of nanomat erials is insensitive t o a pro e ty,

two nanomat erials can be r g rded eq iv lent r g rdles of larg e dife enc in the pro e ty On

the othe han , if a pro e ty is foun dominating t oxicity, sl ght dife enc in the pro e ty would

compromise eq iv lenc between two nanomat erials Figur 3 ex lains a p s ible use of “eq iv lenc

c it eria ” in an efficient hazard as es ment framework of nanomat erials

[80]

Figure 3 — Equivalence c iteria used in an ef icient hazard as es ment framework of

nanomaterials

7.1.4.3 Pos ible index of c omparati ve potency of nanomater ials

TheJa anese National Ins itut e of Adv nc d In us rial S ienc an Te h olo y (AIST) has sele t ed the

rat e ofinc ease in neutro hi in broncho lveolar la ag e fluid (BAL )as a biomarke in icating p t ency

of pulmonary t oxicity of nanomat erials Nakanishi, et al

[49]

furthe showed a futur p s ibi ity that BET

spe ific surfac ar a, a phy sical pro e ty of nanomat erials, could be used as an in ex of comp rative

pot ency of nanomat erials

7.1.5 OECD

OECD defines a cat eg orical a pro ch as folow s Chemicals w hose phy sico-chemical, t oxicolo ical an

e o-t oxicolo ical pro e ties ar lkely t o be simiar or folow a r gular patt ern as a r sult of s ructural

simi arity ma be conside ed as a group, or “cat eg ory” of chemicals

[27]

The as es ment of chemicals b

using this cat eg ory a pro ch dife s from the a pro ch of as es ing them on an in ivid al b sis, sinc

the pro e ties of the in ivid al chemicals within a cat eg ory ar as es ed on the b sisof the ev luation

of the cat eg ory as a w hole, rathe than b sed on measur d data for any one particular chemical alone

The inc easing amou t of t oxicolo y data for NOAAs pro ides o portu ities t o develo c it eria for

hazard- and risk-cat eg oriesfor NOAAs

[27][81][82][83]

The OECD Working Party on Man factur d Nanomat erials (WPMN) org niz d an ex e t me ting in

S pt embe 2 14 t o furthe develo cat eg orization a pro ch for man factur d NOAAs an t o pro ide

r commen ations on how man factur d NOAAs should be cat eg oriz d for the purp ses of t es ing,

r ad ac oss /Structur -A ctivity R elationship models, risk as es ment an risk manag ement

[17]

It

also analy sed information cole t ed b a q es ion air -b sed survey a out OECD membe countries’

a pro ches t o develo or use conc pt of grouping, eq iv lenc an r ad-ac os b sed on phy

sico-chemical pro e ties of NOAAs for their hazard as es ment in r gulat ory r gimes

[8 ]

The OECD Working Party on Man factur d Nanomat erialsex e t me ting also discus ed dife enc s in

hazard pro e ties p s ibly arising from dife enc s in sha e (e.g fibr , plat e), phy sical pro e ties (e.g

z ta p t ential, solubi ty) an bio-pe sist enc and if this could be ad r s ed in a cat eg orization scheme

7.2 Evaluation of categ orical OEL

7.2.1 Simiarities and difer enc es

Al a pro ches t o de iv tion of cat eg orical OEL pr sent ed in 7.1 ar b sed on the anticipat ed primary

mode of action an ar ive at ve y simi ar cat eg ories an v lues of cat eg orical OELs They dife in

detai s of cat eg ory desc iptions an scalng fact ors a pled t o OELs for cor esp n ing bulk mat erials

( e Ta le 4)

Table 4 — Comparison of categ orical OELs

BSI[4 ]

BA uA[4 ]

IFA[7 ]

Low t oxicity immu e efects

Cat eg ry des ription insoluble or po rly soluble

nanomaterials not in the

bioper istent gran lar

na-nomaterials with a density

of >6 0 0kg m

3

Low Density:

bioper istent gran lar

na-nomaterials with a density

To icity of dis olved ions/molecules

Cat eg ry des ription soluble nanomaterial

not in f ibrous or CMAR

cate ory

nanomat erials with

sol-ubility in wat er gre ter

than 1 0 mg l

High Density:

bioper istent gran lar

na-nomaterials with a density

of >6 0 0kg m

3

Low Density:

bioper istent gran lar

na-nomaterials with a density

b low 6 0 0 kg m

3

BSI[4 ]

BAuA[4 ]

IFA[7 ]

Cate orical OE value b sed on bulk OE b sed on bulk OE 0,1 mg m

Frustrated p a ocyt osis from bioper ist ent rigid f ibres

Cate ory des ription insoluble nanomateri

al with a high aspect

ratio (ratio > :1 an

length > 0 0 nm)

rigid nanof ibres ad er

-ing to the World He lth

Org nization (WHO) f ibre

paradigm

CNT u test ed for asb

s-tos-like to icity

Cate orical OE value 0,0 f/ cm

Scale fact or for bulk OE s 1 (asb st os) 1(asb stos) 1 (asb stos)

Specif ic t oxicity/specif ic form

Cate ory des ription nanomat erial which is

alre d clas if ied in its

larg r particle form as a

Car ino enic, Muta enic,

Asthma enic, or R

epro-d ctive t oxicant

nanomaterial which is

alre d clas if ied in its

larg r particle form for

Mos v riation in cat eg orical OELs is o se ved for p orly soluble, low t oxicity cat eg ory ( hown as

“ Low t oxicity immu e efe t ” cat eg ory in Ta le 4) It is mos ly d e t o the lack of consensus a out bes

a pro ches for measuring ex osur s t o these nanomat erials in the workplac at the time w hen these

cat eg orical OELs have be n develo ed

Org nizations w hich de ived these cat eg orical OELs s at ed that they ar not subs antiat ed

t oxicolo icaly The efor , even w he e they ar o se ved, a health risk ma s i l exis for workers and it

is furthe r commen ed t o maintain ex osur levels below cat eg orical OELs if feasible

7.2.2 State of the s ienc e su por ting categ orical OELs

Pr sently, mos a pro ches for cat eg orical OELs ar charact eriz d b very bro d cat eg ories of

nanomat erials, w hich lump t og ethe mat erials with ve y dife ent hazard pro e ties an w hich ar

b sed mor on the ex osur measuring ca a i ties than on p t ential health out comes As such they

come with a warning that they should not be r g rded as safe OELs an that ex osur s should be

minimiz d as much as p s ible Rigid bio e sist ent nanof ibr s fal ng int o the WHO definition for f ibr

p radigm is the only bett er defined consensuscat eg ory of nanomat erials Asbest os OEL b sed on fibr

count is r commen ed for this cat eg ory as a prot ective measur although chaleng es with measuring

the n mbe conc ntration of nanof ibr s in workplac air r main

Ex osur measur ment chal eng es that exis for in ivid al OELs can car y o e t o cat eg orical OELs

for the same ty es of mat erials (with simi ar phy sico-chemical pro e ties) A dditional measur ment

chaleng es ma a ply t o ENMs F or ex mple, in ivid al nanoscale-diamet er CNTs would not be visible

b phase contras mic osco y (PCM), w hich has a r solution lmit of a pro imat ely 0,2 μm diamet er

Exis ing OELs for asbest os r ly on PCM methods for airborne ex osur measur ment Asbest os OELs

(e.g the U.S OSHA P L) a ply t o s ructur s of length > μm, with length:width aspe t ratio of 3:1,

count ed b PCM

[24]

Anothe chaleng e with fibr -sha ed nanomat erials (e.g CNTs) is a measur of rigidity, w hich has be n

as ociat ed with dife enc s in the inflammat ory an car inog enic pro e ties of CNTs

[85][8 ]

L ng, Table 4 (c ontin ed)

p t ent (r viewed in R efe enc [8 ]) The rigidity of MWCNTs can be es imat ed b ISO/TS 1 8 8

The Te h ical Spe if ication pro ides methods for the charact erization of mesosco ic sha e fact ors

of MWCNTs, inclu ing sample pr p ration proc d r s In p rticular, it pro ides a s atis ical method

for charact erizing MWCNTs prod c d b the CVD method During MWCNT synthesis, a ial s ructur s

ar not pe fe tly lnear but inclu e pe manent ben p int ISO/TS 1 8 8 pro ides methods for

det ermining a s atis ical q antity, r pr senting a ma imum s raight length that is not deformed b

pe manent ben ing caled the “ tatic ben ing pe sist enc length ” ( SB L) The SBPL gives information

r g rding the r lationshipbetwe n the MWCNT mesosco ic sha e an siz

As k ow ledg e a out hazard pro e ties of nanomat erials furthe develo s it can be ex e t ed that

the a pro ch for cat eg orical OELs wi be r f ined an wi shif t owards nar ower cat eg ories of

nanomat erials Ex mples of such cat eg ories ar NIOSH an OECD a pro ches desc ibed in 7.1 These

a pro ches could be b sed on comp ring pot ency of the NOAA t o a benchmark (r fe enc ) particle

in the same mode-of-action cat eg ory F or ex mple the “ paralelo ram” a pro ch

[3 ][3 ][8 ]

could

be used t o extra olat e animal data for the NOAA t o h man hazards an t o as ign the NOAA t o the

a pro riat e cat eg orical OEL

[2 ]

A r l anc on short-t erm animal ex e iment could move us t owards

ra id t oxicolo ical prof ilng for a larg e v riety of NOAAs Such ex e imental sy st ems an data analy sis

methods ne d t o be v ldat ed befor they can r plac traditional OELs Cat eg orical OELs or OEBs, b sed

on bes a aia le information from othe mat erials in that biolo ical and phy sico-chemical cat eg ory, can

beuseful t o pro ide initial ex osur lmit es imat es The BSIguidanc is an early ex mple of four bro d

cat eg ories of nanomat erials, each as ociat ed with a “ pra matic guidanc level”

[4 ]

Such schemes could

ber f ined as the nanot oxicolo y data inc ease t o be ome suff icient for the develo ment an v ldation

of evidenc -b sed models (e.g QSAR)that ar pr dictive of the hazard an risk of a nanomat erial In the

meantime, as the s at e of the scienc r mains l mit ed for the develo ment of in ivid al OELsfor NOAAs,

hazard an control b n ing schemes ar being used t o fil the g p Some hazard b nd a pro ches

inclu e OEBs (rang e of ex osur s)as ociat ed with the hazard (natur and severity of adve se efe t)

8 OEBs and c ontrol banding for nanomaterials

8.1 Over view of current hazard and c ontrol banding s hemes

Hazard an control b nding ha e be n used for many years t o make de isions on workplac ex osur

controls w hen OELs ar not a aia le an t o sup ort hazard commu ication la el ng, for chemicals

in g ene al (HSE

[9]

GHS[2 ]

OSHA[89]

) an mor r c ntly for NOAAs (ISO/TS 1 9 1- 2, ANSES

) Control b n ing is a pra matic t ool

that can be used t o identify the ty es of engine ring controls an pe formanc ca a i ties t o achieve

the spe ified levels (e.g order-of-ma nitu e b n s) of ex osur control The ty ical control ban ing

framework is a matrix consis ing of hazard b n s an ex osur p t ential b n s t o in icat e the

a pro riat e control b n for a chemical subs anc given it pro e ties an prod ction/use (Ta le 5) In

this ex mple of the ISO control b n ing scheme for NOAAs (Ta le 5), CBs 1-3 inclu e g ene al, local, or

enclosed venti ation (CB 1, 2, or 3, r spe tively) or ful containment o tions (CB 4 or 5)

Table 5 — C ntrol band matrix with hazard and ex osure potential ban s (EBs)

Some control b n ing sy st ems for NOAAs ha e a scor -b sed hazard b n alocation sy st em that

uti iz s information on the phy sico-chemical pro e ties of the nanomat erial (and it par nt or bulk

form) along with ex e t ju gment on w hat is k own a out the hazard pot ential given those pro e ties

(CB Nanot ool an St ofenmanag er Nano ) Othe hazard b n ing schemes ha e as ociat ed

order-of-ma nitu e oc up tional ex osur conc ntration rang es; se Ta le 5 (HSE

[9]

ANSES[9 ]

an

ISO/TS 1 9 1- 2), although these conc ntration rang es we e not ne es ari y de ived spe ifical y for

NOAAs OE sar a g ene al t erm for these conc ntration rang es OEBs an OELs should not be confused

with ex osur pot ential b n s (EBs) OE s an OELs indicat e the levels of ex osur that ar conside ed

t o beadeq at e t o pr vent adve se efe t in workers an / r that ar t ech icaly feasible t o achieve E s

ar q altative desc ipt ors of p t ential ex osur levels b sed on the fact ors that influenc ex osur ,

such as the pro ensity of the mat erial t o be ome airb rne (d s ines ), the ty e of proc s , an amou t

of mat erial being han led

[3]

NIOSH is cur ently ev luating the scientif ic evidenc inclu ing nanot oxicolo y s u ies for use in

develo ing cat eg orical OELs or OEBs for NOAAs (5.1)

[16]

In ad ition, NIOSH is develo ing and ev luating

a g ene al hazard ban ing framework w hich inv lves a sy st ematic an tie ed a pro ch t o hazard

b n / OEB al ocation of chemical subs anc s; this a pro ch is also being ev luat ed for it a plca i ty

inhalation data in rat (airborne conc ntration of g ses/ a ours or d s /fumes /mis s as ociat ed with

5 % lethalty in 1 h) Naumann, et al

[8]

also pro osed order-of-ma nitu e b n s, caled pe formanc

-b sed ex osur control l mit (P -EC s), w hich l n the engine ring pe formanc b n s t o the health

efe t data an uses the mos prot ective health en p int for b nding a chemical Their scheme

ex r s es p t ency as the mas dose /da , an seve ity as a q altative rang e of acut e an / r chronic

efe t (none, sl ght, mode at e, sever ) Their clas if ication scheme is simiar t o those b Henry an

S ha e

[7]

E C[94]

w hich we e develo ed t o sup ort hazard communication la el ing

Each of these control b n ing schemes inclu e a hazard b n ing scheme with upt o four or fivehazard

groups Bro ke

[96]

show s the al gnment of order-of-ma nitu e ex osur b n s with f ive hazard ban s

(A-E) an the as ociat ed R-phrases The HSE COSHH Es entials

hazard clas es Al of these control ban ing schemes use the common matrix

a pro ch of algning the hazard b n / OEB with the ex osur or emis ion p t ential b n t o identify the

a pro riat e control b n

K ey ar as of unc rtainty inclu e the a plca i ty of the order-of-ma nitu e OEBs t o NOAAs an how

the emis ion pot ential r lat es t o actual worker ex osur s Imp rtant r sear h ne ds inclu e ev luating

the efe tivenes of these g ene al a pro ches t o as es ing the hazards an ex osur s in spe if ic jo

tasks an workplac susing nanomat erials

8.1.1 C mpar ison of hazar d bands and OEBs as appled to inhaled NOA A s

A summary of publshed hazard b n ing schemes ispro ided for sele t ed acut e an chronic health en

point in Ta le6 This summary is pro ided t o faci tat e comparison of key q altative an q antitative

element of each scheme, with a focus on in alation ex osur s (Original r fe enc s should be consult ed

for information on the ful rang e of adve se health en p int an rout es of ex osur s ) These hazard

an OEB schemes ha e common element aswel as some dife enc s Each scheme inclu es q altative

desc ipt ors of the level of severity of a hazard b sed (usualy in rat ) Some schemes pro ide b th

q altative an q antitative in icat ors of seve ity ISO hazard alocation scheme (ISO/TS 1 90 - 2:2 14,

Ta le 1) has seve al element in common with othe hazard ban ing schemes, as shown in Ta le 6 for

acut e an chronic en p int that ar r lev nt t o in alation hazards

Table 6 — Hazard an oc upatio al ex osure band (OEB)s hemes for inhaled dusts, fumes, or

mists: A cute and chronic efects (selected)

A cute t oxicity: Rat L 5 in

Lik eliho d of chronic efects

(e.g syst emic)

R epe ted e posure:

Rat in alation 6 h/d for at le st

(convert ed from mg d, as uming

h mans an oc u ational air

information is from Ta les 3.1.1 an 3.1.3 (acute t oxicity); Figure 3.8.1, Ta le 3.8.1, an Ta le 3.8.3

( STOT-SE); Figure 3.9.1,Ta le 3.9.1, 3.9.2, an 3.9.3 ( STOT-RE) OSHA

[8 ]

c iteria are es entialy the same, ex cept

that only Cate ories 1 through 4 are used; se Ta le A.1.1 (acute t oxicity); Ta le A.8.1 (single dose); Ta les A.9.1

an A.9.2 ( 90-day stu y)

do not inclu e OE s Comparison of the animal e posure concentrations ac os

s hemes sug ests that Categ ories 2 an 1 of G S

S e Ta le 3 of R eference [9] for specif ic R-p rases an H-statements that are used t o as ign hazard grou ;

al ocation b sed on Bro ke

[9 ]

Hazard grou E with “—” in icates that no airb rne concentration can b fou d

t o provide adeq ate control

[9]

e

T e EU C P R egulation [R egulation (E ) No 1 72 2 0 ]p ases in the use of H p rases inste d of R-p rases,

in most cases T e de dline for transition from R to H was 1 Ju e 2 1

f

P -E L: performance-b sed e posure control limit

g

S fety an He lth In e System ( SHIS)

F or ex mple, the HSE

[9]

COSHH Es entials hazard alocation (b n ing) scheme pro ides the same

order-of-ma nitu e ex osur conc ntration rang es for groups A through D, as wel as the a senc of

an ex osur conc ntration for Group E HSE

[9]

s at es that the groups with ex osur conc ntration

in icat es that ex osur s could be identif ied as pro iding adeq at e control given the hazards identif ied

in Groups A t o D; Group E is int en ed for se ious health hazards, w he e no a pro riat e airborne rang e

could be identified

[9]

COSHH Es entials uti iz s “ risk (R) phrases” an “hazard (H) s at ement ” t o

as ign groups A ls of R- an H-phrases used in COSHH Es entials an their as ociat ed hazard b n

as ignment ar pro ided in Ap en ix 3 of HSE

[9]

S ve al of the t oxicity data ases pro ide the

R-phrases or H-s at ement for chemical subs anc s g ene aly an for NOAAs or their p r nt mat erials,

e.g (CEC

[97]

An ex VI); an GESTIS

[98]

In conc pt, the use of H-s at ement or R-phrases should be

a pl ca le t o NOAAs This is be a use the hazard phrases desc ibe the adve se health efe t that ma

oc ur t o spe if ic org ns from ex osur t o chemical subs anc s b v rious rout es of ex osur Howeve ,

unc rtainty exis s ast o w hethe the data on w hich hazard phrases we e de ived for chemicaly simiar

mat erials ar also a pl ca le t o NOAAs Furthe ev luation is ne ded t o det ermine if the use of g ene al

hazard b nding schemes would r sult in the a pro riat e hazard b n s an OE sfor NOAAs

F or acut e t oxicity, the GHS

[2 ]

hazard cat eg ories 4 through 1 ar b sed on animal data that ar

n me icaly simiar t o HSE

[9]

an ISO/TS 1 9 1- 2 hazard cat eg ories A through D That is, the ISO/HSE

Table 6 (c ontin ed)

highes hazard cat eg ories for the ISO/HSE an the GHS hazard b nding sy st ems, r spe tively GHS

cat eg ory 5 (lowes t oxicity) does not a pear t o ha e a comp ra le HSE

[9]

or ISO/TS 1 9 1- 2 cat eg ory

( i.e the GHS cat eg ories in Ta le 5 would be shift ed t o the lef b one cat eg ory) The OSHA

uses only hazard cat eg ories of 4 through

1 Othe adve se health en p int ar cat eg oriz d dife ently F or ex mple, a chemical is cat eg oriz d

ac ording t o “ pe ific targ et org n t oxicity with r peat ed ex osur ( STOT-RE) in hazard b n A or B

for unl kely, Cfor p s ible, an D for pro a le chronic adve se health efe t

[3][9 ]

(Ta le 6)

Much of the q antitative data used in these g ene al hazard b n ing schemes is based on acut e

ex osur ( y icaly L 5 for in alation) Howeve , lt le information is a aia le on acut e efe t for

nanomat erials, in p rt be a use of the de r ased use of animal t es ing an a gr at er emphasis on earle

(mor sensitive)adve se health en p int As such, a r f inement of the g ene al hazard b n ing schemes

ma be ne ded t o ca tur the dose-r sp nse r lationships o se ved in cur ent t oxicolo y s u ies of

NOAAs, inclu ing for the earl er-s ag e adverse en p int (e.g pulmonary inflammation an early-s ag e

fibrosis, w hich ma not yet be as ociat ed with fu ctional chang es but could be with chronic ex osur

t o the bio e sist ent NOAA) As discus ed in 8.3, the ex osur conc ntration c it eria for STOT-RE b n

Cor D (≤ 00 or ≤ 0 mg m

3

in a 9 d animal s u y) ma not be p rticularly r lev nt or NOAAs It is also

of int er s that the ex osur conc ntration c it eria for hazard b n ing b sed on acut e t oxicity lethalty

ha e de r ased sinc the early hazard ban ing sy st em of Henry an S ha e

[7]

comp r d t o the mor

r c nt GHS

[2 ]

an r lat ed sy st ems (Ta le 6)

8.1.2 ISO hazar d banding s heme for NOA A s

The ISO hazard group alocation scheme (ISO/TS 1 9 1- 2:2 14, Ta le 1) r fe s t o the Int ernational

La our Org nization C ntrol Ban ing To lkit (Ta le 2 of R efe enc [9 ]) an the GHS health hazard

clas if ication

[2 ]

The ISO hazard b nding uses a de ision tr e a pro ch, as desc ibed in ISO/TS 1 9

1-2:2 14, 7.2.2 an i us rat ed in ISO/TS 1 9 1- 2:2 14, Figur 2 The de ived hazard ban is used in

control b n ing for NOAAs

[3]

A summary of the hazard b n ing st eps for NOAAs (ISO/TS 1 9 1- 2:2 14, 7.2) is as folow s

— Ques ion 1: Has the NOAA alr ady be n clas if ied and la eled ac ording t o national or r gion

legislation or GHS?

— If yes, as ign the NOAA t o the cor esp n ing hazard b n

— If no, or if la el ing isb sed on lack of information, g o t o next q es ion

— Ques ion 2: Is the NOAA solubi ity in wat er highe than 0,1 g l?

— If yes, ev luat e the NOAA as a clas ical chemical hazard using a g ene al hazard b n ing scheme

— If no, g o t o next q es ion

— Ques ion 3: Does the NOAA contain bio e sist ent f ibr s or fibr -l ke s ructur s [def ined as rigid

f ibr with length (L) > μm, diamet er (d) < μm, an L d ratio of > ]?

— If yes, as ign t o hazard b n E

— If no, g o t o next q es ion

— Ques ion 4: Ar the e hazardous in ications for the NOAA?

— Ques ion 4a: Do sc e ning t es s in icat e car inog enicity, mutag enicity, r prod ctive t oxicity, or

sensitivity b in alation (CMRS)pro e ties

— If yes, as ign t o hazard b n E

— If no, g o t o next q es ion.

— Ques ion 4b: Ar compr hensive hazard data a aia le for the NOAA?

— If yes, as ign t o mos prot ective hazard b n ( tarting with E), ac ording t o t oxicolo ical

data

— If no, g o t o next q es ion

— Ques ion 5: Is the e a hazard b nd for the bulk mat erial or an analog ousmat erial?

— If yes, an the bulk hazard b n is A, then as ign the NOAA t o hazard b n A; if yes, and the bulk

hazard b n is B, C, or D, then ad one ban an as ign the NOAA t o hazard b n C, D, or E

— If no, as ign t o hazard b n E

As desc ibed, the ISO hazard b n ing proc s is hea iy depen ent on the g ene al hazard b n ing

schemes Spe ific data on NOAA hazard (Ques ion 4 ) ar ev luat ed ac ording t o the hazard b n ing

c it eria Data on the bulk mat erial ar also used with the ad ition of one ban ( i.e ex osur is r d c d

b an orde of ma nitu e)

8.2 Case studies on banding NOA A s

It is ins ructive t o ev luat e a set ofNOAAs for how they would be as es ed for hazard ac ording t o

ISO/TS 1 9 1- 2 an r lat ed schemes This ev luation show s that he acut e t oxicity c it eria of < 0 mg m

3

or < 5 mg m

3

(GHS[2 ]

an ISO/TS 1 9 1- 2, r spe tively; Ta le 6) t o b n NOAAs ma ha e l mit ed

uti ity for hazard alocation an control b n ing sinc mos acut e in v i vo s u ies use much lowe

conc ntrations F or ex mple, an acut e in alation t oxicity s u y ofsive nano articles

si ver nano articles

is subs antialy below the lowes ex osur conc ntration v lues of 5 mg/ m

and othe hazard b n ing schemes Th s,

the Sung, et al

[1 0]

s u y ex osur s ar not informative with r g rd t o ev luating si ve nano articles

ac ording t o ISO/TS 1 9 1- 2 an GHS

[2 ]

acut e t oxicity c it eria; an those c it eria ma be t oo bro d

an nonspe ific t o adeq at ely identify an dis inguish acut e t oxicity hazards for NOAAs

Ta le 7 ls s NOAAs for w hich OELs ha e be n develo ed along with the as ociat ed hazard b n an

OE b sed on STOT-RE The pro osed OELs for CNTsrang e from 1μg m

3

t o 5 μg/ m

3

Th s, depen ing

on w hich OEL is used, CNT could be plac d in eithe hazard cat eg ory C (mode at e hazard) or hazard

cat eg ory D ( e ious hazard) ac ording t o ISO/TS 1 9 1- 2 an r lat ed schemes In ad ition, ac ording t o

ISO/TS 1 9 1- 2 hazard b n ing de ision lo ic, if the CNT mat erial contains rigid f ibr s (a fr e s an ing

fibr in cole t ed samples would a pear in ele tron-mic osco ic imag es as a s raight fibr with L > μm,

d < μm, L d ratio > ), this mat erial would be conside ed as a mat erial w hose t oxicity is driven b the

fibr p radigm an alocat ed t o the highes hazard b n (E, seve e hazard), u les toxicolo ical data

pro ide evidenc that it is not the case

Pro osed OELs for fule ene also dife subs antial y, with an as ociat ed hazard b n of eithe B or D

depending on the method used t o de ive the OEL The EU methods for es imating h man in icative

no-efe t levels (INELs)for workers as used b Aschbe g er, et al 2 1

[56]

t en t o r sult in lowe OELs than

those b sed on seve al othe hazard an risk as es ment methods (Ta le 7)

Table 7 — Hazard categ ory as ociated with NOAA OEL or as derived based o animal studies

NOAA

OE

(µg/m3

F lerene

(C

6)

As hb rg er, et al

2 1[5 ]

a

Based on c mparison of the N AE or LO E e posure c nc nt ations (e te t) used to derive the OE s to the

repe ted e posure c nc nt ation criteria (<2 mg/m

The OELs in Ta le 7 we e de ived from animal (rat) s u ies of inhaled (or ins i led) NOAAs, with

ex osur d rations of a few day s t o two years The no o se ved adve se efe t level (NOAEL) or lowes

o se ved adverse efe t level (LOAEL) r p rt ed in these s u ies inclu e the folowing:

a) Fule ene (C

60):

These animal efe t levels (NOAEL or LOAEL) ar al low mas conc ntrations comp r d t o the lowes

Those r peat ed ex osur c it eria ar for d s s, mis s, or fumes

in g ene al (not for NOAAs spe if icaly, although some s u ies ma ha e inclu ed nanoscale p rticles)

Theseve ity of the efe t ma also dife in the s u ies used t o develo the STOT-RE c it eria comp r d

t o the s u ies used t o develo OELs for NOAAs If these NOAA NOAELs or LOAELs from the rat s u ies

ar comp r d t o STOT-RE c it eria, the r sult would be t o as ign each of these NOAAs t o a Hazard

Cat eg ory D an an OE of < ,0 mg m

3

( 8 h TWA)

[3][9][9 ][96]

In summary, this ev luation indicat es that the cur ent hazard ban ing schemes ( inclu ing ISO, ANSES,

GHS, OSHA, an COSHH) ma be t oo bro d an nonspe ific t o ac urat ely group the hazards of NOAAs

in t erms of the acut e an chronic adve se health en point data from nanot oxicolo y s u ies Some of

these early-s ag e adve se health efe t ha e be n used in develo ing OELs for NOAAs, yet al of these

efe t levels ar subs antial y lowe on a mas basis than the lowes efe t levels in the cur ent hazard

b n ing schemes The acut e and chronic hazard b n ing c it eria r sult ed in the alocation of a hazard

b n D ( e ious hazard) an the as ociat ed OE (< ,0 mg/ m

3

) for al NOAAs ev luat ed Th s, the

a pl cation ofhazard b nding in this ex mple r sult ed in eq al or gr at er prot ection compar d t o the

pro osed OELsfor these NOAAs

8.3 Evaluation of the evidenc e for initial (default) OEBs for categ ories of NOA A s

8.3.1 Categ orical anal yses and r ead-acros

S veral OEL and OEB a pro ches discus the use of r ad-ac os t o f il data g ps an t o de ive OEL or

OE es imat es for data-p or subs anc s using t oxicity data from a simi ar subs anc with sufficient

data

[27][1 2]

Both q altative an q antitative r ad-ac os a pro ches ar a ai a le

[27]

A pro isional OEL could be de ived b dir ct r ad-ac os with a simiar subs anc , although the inhe ent

v ria i ity an u c rtainty ma not be wel k own The de iv tion of an OE b sed on analo y r

ad-ac os from data on a simi ar subs anc ma be mor a plca le t o the pra matic purp se of sele ting

an ev luating oc up tional ex osur controls That is, given the order-of-ma nitu e ex osur ban s

b sed on pe formanc of engine ring control o tions, a c rtain level of u c rtainty in the hazard

es imat es for NOAAs ne d not pr clu e de ision-making using a control b n ing scheme In othe

words, despit e the wide v riety of phy sico-chemical pro e ties of NOAAs, the ex osur control o tions

ar mor l mit ed, e.g four or five groups b sed on order-of-ma nitu e pe formanc c it eria

[8]

Although

data ma be insuff icient t o develo a subs anc -spe if ic OEL, information ma be sufficient t o de ive an

initial hazard b n or OEB for use in control b n ing

Bro d hazard cat eg ories that ha e be n sugg est ed b seve al ag encies or r sear he s inclu e: soluble,

po rly-soluble low t oxicity, po rly-soluble high t oxicity, an fibrous p rticles (BSI

) This MOA-b sed framework is i lus rat ed in Ta le 8 Ex mple hazard/OE

alocations of thr e NOAAs in Ta le 8 (ultrafine TiO

2, CNT/ CNF, an sive ) ar b sed on the US NIOSH

r commen ed ex osur lmit (RELs) for those mat erials

[24][5 ][58]

The REL for soluble an insoluble

sive is 0,0 mg m

3

(w hich would fal int o cat eg ory C of ISO/TS 1 9 1- 2 hazard b n ing scheme);

howeve , sinc that REL is not spe if ic t o NOAA, sive NOAA is plac d int o the next highe hazard

cat eg ory D (< ,0 mg m

3

) ac ording t o ISO/TS 1 90 - 2 c it eria The seve ity of adve se health en

point as ociat ed with r peat ed or chronic ex osur t o these thr e subs anc s v ries from r latively

benign arg ria (pigmentation of skin) in workers t o pulmonary inflammation, fibrosis an lu g canc r

in rat or mic

In the cur ent ex mple, the desc iption of cat eg ory D ( e ious hazard) for b th si ve nano articles an

CNT/CNF in the same hazard/OEB cat eg ory could be confusing given the dife ent adve se health en

point In ad ition, the REL for ultraf ine TiO

) ma be diff icult t o u de s an sinc it is b sed on lu g canc r, a seve e chronic adve se

health efe t Yet, the NIOSH REL was set at an ex osur conc ntration that would ha e a r latively low

pro a i ity (95 % lowe con denc lmit of a 1/ 1 00 ex ces risk) of lung canc r if a worker is ex osed

at the REL for up t o a 45-year working l fetime (In contras , the acut e t oxicity c it eria in ISO, GHS an

othe hazard b n ing schemes ar b sed on the L 5 , or the ex osur conc ntration as ociat ed with

key data an information on w hich the hazard b n / OEB alocation is made ar es ential t o pro iding

consist ent an transp r nt information for risk manag ement de ision-making The g oal of the hazard

b n /OE alocation is t o identify an ex osur conc ntration ex e t ed not t o be as ociat ed with

signif icant risk of any adve se health efe t

A dditional r sear h an data analy sis ne ds inclu e o taining r lev nt dose-r sp nse data for a

n mbe of othe NOAAs that fal int o the main four MOA/PC pro e ty cat eg ories C mbined analy ses

of dose-r sp nse r lationships as modified b v riations in the PC pro e ties can then be ev luat ed

t o det ermine the ne d for sub-cat eg ories b sed on en point an / r p t ency The mor data that ar

o tained from a s andard set of as ay s an en p int , the gr at er the ro us nes of the cat eg orical

OEL/OEB es imat es Statis ical models an methods wi l be ne ded that can ac ommodat e mix ed

dose-r sp nse r lationships an ac ou t for v ria i ty an het erog eneity in data from multiple as a s, en

p int , an ex e imental con itions (e.g Wang, et al 2 14

[44 ][46]

[

7]

with examples of NOAAs based o NIOSH-recommen ed ex osure

lmits to prevent adver e lung or sy stemic responses given chronic inhalation ex osures

Adver e en point inclu e arg ria in h mans ( iver), p lmonary fibrosis an inflammation in rat an mic

(CNT/ CNF) an lu g c nc r in rat (UF TiO

2)

8.3.2 Utiity of in vitr data in OE /OEB develo ment for NOA A s

Given the lmit ed data a ai a le for many subs anc s used in the workplac , inclu ing high prod ction

v lume chemicals, pharmac uticals, an NOAAs, the use of in v itro data for sc e ning hazard as es ment

an prioritization of subs anc s for tie ed t oxicolo y t es ing has be n pro osed (e.g To Cas , NexGen)

In this a pproa ch, a pro isional OEL is es imated for an u s u ied

compou d that has s ructur -a ctivity simi arity to a data -rich comp u d with a v aia le in vitro an in

vivo data an an OEL v alue In a pplying such methods to nanomaterials, the in vitro tes system, toxicity

en p int , an tes article conc ntrat ions should be car fuly sele ted for r lev anc to the anticip ted

hazards an wor kpla ce ex osur s ofth e nanomaterial() being ev aluated (Gordon, et al

[6]

)

S ve al s u ies ha e shown g ood concordanc of the r lative hazard of metal o ide an othe NOAAs

in in v itro an in v i vo as ay s of inflammation r sp nses

Howeve , othe s u ies show wide

v ria i ity of in v itro an in v i vo r sult ac os ex e imental as ay s an la orat ories

[1 9][1 0]

Stan ardiz d an v ldat ed in v itro as ay s can pro ide me hanis ic data an information on w hich t o

develo Q SAR models t o desc ibe the r lationship betwe n the dose of a subs anc an the biolo ical

r sp nse given the phy sico-chemical pro e ties QSAR models using in v itro data ha e be n used t o

clas ify or clust er metal o ide nanomat erials int o bio ctivity groups

[1 1][11 ][1 3]

The use of in v itro

dose-r sp nse data t o es imat e c itical efe t levels (e.g BMD, v s NOAEL or LOAEL) has be n pro osed,

using methods simiar t o those used for in v i vo data

[1 4]

Such q antitative analy ses could pro ide the

b sis for ev luating the evidenc for p s ible hazard subcat eg ories, w he e the r lationship betwe n

membe subs anc s can be desc ibed b a set of pr dict ors PC pro e ties in ad ition t o dose

CNT/CNF is an ex mple of a cat eg ory of NOAAs within the f ibrous MOA group for w hich sufficient data

( inclu ing in v itro data) ma be a aia le t o ev luat e the r lative hazard of v riousty es of subs anc s

within this cat eg ory of NOAAs R ec nt s u ies ha e r p rt ed wide dife enc s in the pulmonary

inflammat ory r sp nses b sed on surfac functionalzation, inclu ing r d c d inflammat ory an

fibrog enic r sp nses t o v rious ty es of CNT

[1 5][116][1 7]

or t o TiO

2nanosphe es and nano elt

[1 8]

The influenc of these dife enc s r lative t o othe sour es of v ria i ity an u c rtainty in the risk

as es ment proc s has not yet be n ev luat ed

Whie mor r sear h ne ds t o be done, in v itro as ay s ma be ca a le ofpr dicting acut e in v ivo

r sp nses ( i.e within 24 h of ex osur ) F or ex mple, acut e as ay s ma be useful for highly r active

subs anc s In ad ition, seve al r c nt s u ies ha e shown cor elation betwe n the activ tion of the

NLRP3 inflammasome an pro-fibrog enic en point in v itro or f ibrosis in v i vo as ociat ed with ex osur

t o CNT

An in v itro s u y in h man lu g smal airwa epithel al c ls showed

dis inct p tt erns of CNT neo las ic -l ke trans ormation compar d t o asbest os

[1 1]

Th s, some in v itro

as ay s ma be useful for initial sc e ning of NOAAs t o det ermine w hethe furthe t es ing is ne es ary

In summary, seve al r c nt nanot oxicolo y s u ies pro ide spe if ic ex mples of g ood concordanc of

acut e in vitro an in v i vo inflammat ory r sp nsest o carb n an metal nanomat erials In ad ition, some

in v itro as ay s ma be useful as sc e ning as ay s for p t ential chronic efe t of pulmonary fibrosis an

neo las ic lesions from oc upational in alation ex osur t o NOAAs Val dation of these f indings would

pro ide sup ort for using in vitro data from v ldat ed as ay s t o f il g ps in the hazard data ne ded t o

de ive OELs or OE s

8.3.3 Options for der i ving an OE or OEB for NOA A s

Based on thecur ent tat e of the scienc , the o tions a ai a le t o de ive OELs or OE sfor nanomat erials

inclu e the folowing:

a) use OEL develo ed for spe ific NOAA or group of NOAAs, if a ai a le (e.g Ta le 7);

b) use q altative or q antitative r ad-ac os from the OEL of a simiar subs anc t o the NOAA

(e.g benchmark particles);

[2 ][27]

c) de ive an OEB for NOAA b sed on the OEL or OE for the bulk mat erial ( i.e ad one hazard b n ;

se ISO/TS1 9 1- 2);

d) de ivean initial OEB for NOAA b sed on sc e ning data (e.g in v itro);se 8.3.2

Weight of evidenc ev luation is g ene aly r g rded as the pr fe r d a pro ch t o making hazard

) Cur ent data ases a aia le t o ev luat e the hazard of g ene al chemical

subs anc s, inclu ing NOAAs, inclu e the folowing: EPA Int egrat ed Risk Information Sy st em (IRIS)

The purp se of this document is t o desc ibe an ev luat e the s at e-of-the-art in a aia le data an

methods for develo ing OELs or OE s for NOAAs Ultimat ely, the bes a aia leevidenc should be used

t o ev luat e the hazard an risk of oc upational ex osur t o NOAAs an t o sup ort risk manag ement

de ision-making, w hich inclu esthe sele tion of efe tive ex osur controls

9 Feasibi ity c onsiderations in the OEL and OEB set ing pr oc es

Develo ment an use of OELs an OE s ar inte twined with a ai a le risk mana ement measur s t o

maintain ac epta lelevel of risk Oft en, the dete mination of an OEL inv lves conside ation of b th the

health efe t data an the t ech olo ical feasibi ty of measuring and control ing ex osur s at or below

that conc ntration R egulat ory OELs also conside e onomic feasibi ity As with othe oc up tional

hazards, r commending an OEL for a spe if ic NOAA ma be contingent on ha ing adeq at e health efe t

information, an a pro riat e sampl ng an analytical method, an the a i ity t o control ex osur s at

the OEL

[5]

The purp se of this annex is t o pro ide an o e view of thescientific methods, p l cies, an proc d r s of

a uthoritative ag encies worldwide that develo oc upational ex osur lmit These inclu e proc d r s

of b th r gulat ory an non-r gulat ory g ove nmental ag encies as wel as nong ove nmental in us rial

hy giene as ociations uti z d int ernationaly The cou tries r pr sent ed ar those that v lu t ee ed t o

pro ide a desc iption By desc ibing the simiarities an dife enc s in these proc s es, it is int en ed

that this cha t er ma faci itat e the u de s an ing and harmonization of the scientific evidenc b sis

for develo ing OELs an OEBs for NOAAs

A 1.2 OEL types and adjustment models

A 1.2.1 Time weighted averag e (TWA )ex posur e lmit

A TWA ex osur lmit is the a e ag e airb rne conc ntration of a particular subs anc pe mitt ed o e

a c rtain pe iod of work usualy ex r s ed as a t otal n mbe of hours pe da These ar the mos

common ty es of ex osur lmit

It is pr fe a le t o keep ex osur levels contin aly below the TWA ex osur lmit In practic , the actual

conc ntration of an airb rne contaminant arising from a p rticular proc s ma fluctuat e significantly

with time Howeve , d ring pe iods of contin ous dai y ex osur t o an airb rne contaminant, the TWA

ex osur lmit alow s short-t erm ex cursions a o e the ex osur l mit pro ided they ar compensat ed

for b ext en ed pe iods of ex osur below the lmit d ring the working da The TWA ex osur does

not alow ex ce danc of c i ing l mit d ring the working da

In cases w hen work shifs ex ce d 8 h, TWA lmit can be adjust ed using seve al models desc ibed in

[1 4]

A 1.2.2 Shor t-ter m e x posure l mit (STE )

A STEL is the time-weight ed ma imum a e ag e airb rne conc ntration of a p rticular subs anc

pe mitt ed o e a short pe iod of time (usual y 1 min)

Some subs anc s or mixtur s can cause int ole a le ir itation or othe acut e efe t up n brief ex osur ,

although the primary t oxic efe t ma oc ur with long t erm ex osur through ac umulation of the

subs anc or mixtur in the b dy or through grad al health imp irment with r peat ed ex osur s

TheSTEL pro ides lmit only for the control of short-t erm ex osur STELs ar imp rtant sup lement

t o the TWA ex osur l mit w hich ar mor conc rned with the t otal intake over long pe iods of time

Gene aly, STELs ar es a lshed t o minimiz the risk of

— int ole a le ir itation,

— ir eve sible tis ue chang e, an

— nar osis t o an ext ent that could pr cipitat e workplac incident

STELs ar r commen ed w he e the e is evidenc that adve se health efe t can be caused b high

short-t erm ex osur

A STEL should not be ex ce ded at any time d ring a working da even if the TWA averag e is within the

TWA ex osur lmit

A 1.2.3 P eak/c eiin lmits

Peak or c i ng ex osur lmit ar a ma imum or peak airb rne conc ntration of a particular subs anc

det ermined o e the short es analytical y practica le pe iod of time

F or some ra idly acting subs anc s an mixtur s the a e a ing ofthe airb rne conc ntration o e a

work da pe iod is not a pro riat e These subs anc s ma in uc acut e efe t aft er r latively brief

ex osur t o high conc ntrations, so the ex osur s an ard for these subs anc s r pr sent a ma imum

or peak conc ntration t o w hich workers ma be ex osed A peak ex osur lmit hould not be ex ce ded

at any time

A 1.2.4 Brief and Scala model for adjusting TWA lmits

The TWA lmit is b sed on the n mbe of hours worked pe 24 h da an the pe iod of time between

ex osur s This model is int en ed t o ensur the daiy dose of the t oxicant un e an alt er d work shif is

below that for a conventional shif t o take ac ou t of the r d c d time for el mination, i.e r co e y time

betwe n ex osur s

The Brief an S ala model is r commen ed for calculating adjus ment t o ex osur lmit This model

is pr fe r d be auseit

— is simple t o use,

— takes int o ac ou t b th inc eased hours of ex osur an de r ased ex osur fr e time, an

— is mor conse v tivethan othe models

The Brief an S ala model is based on a 40 h work we k F ormula (A.1) takes int o ac ount b th the

pe iod of ex osur an pe iod of r co e y

A 1.2.5 Phar mac okinetic models for adjusting TWA l mits

The e ar seve al dife ent pharmacokinetic models a ai a le These ar suita le for a pl cation t o

ex osur s an ards b sed on ac umulat ed b dy burden These models take int o ac ou t the ex e t ed

beha iour of the hazardous subs anc in the b dy b sed on know ledg e of the pro e ties of the

subs anc These models use information such as the biolo ical half-l fe of a subs anc an ex osur

time t o pr dict b dy burden The use of pharmacokinetic models can be complcat ed b the lack of

biolo ical half-lves for many subs anc s

The mos widely used pharmacokinetic model is the Hickey an R eis model w hich r q ir s k ow ledg e

of the subs anc ’ s biolo ical half-l fe, the hours worked pe da an hours worked pe we k The

Hickey an R eis model l ke othe pharmacokinetic models as umes the b dy is one compartment,

i.e a homog eneous mas

P armacokinetic models ar les conse v tive than the Brief an S ala model, usualy r commen ing

smale r d ctions of the ex osur lmit Whie pharmacokinetic models ar theor tical y mor ex ct

than othe models, their lack of conse v tism ma not alow adeq at ely for the u k own adve se

efe t on the b dy from night work or ext ended shif s that ma afe t how wel the b dy meta olses

an elminat es the subs anc

A 1.2.6 Q uebec mo el for adjusting TWA lmits

The Quebe Model develo ed b the Ins itut de R eche che R obe t-Sauvé en Sant é et en S curit é d

Tra ai (IRS T) uses the mos r c nt t oxicolo ical data t o as ign subs anc s int o cat eg ories Depen ing

on the cat eg ory as igned, a r commen ation is made that

— no adjus ment is made t o the ex osur lmit,

— a daiy or weekly adjus ment, or

— themos conse v tive of the dai y or we kly adjus ment w he e both a ply

The Quebe model is sup ort ed b a compr hensive t ech ical guide an a sele tion t ool t o as is in

det ermining the mos a pro riat e adjus ment cat eg ory

A 2 Austral a

A 2.1 Reg ulator y ex posure l mits

A 2.1.1 Leg islatio , or ganizatio and proc es es

In A us rala, an ex osur s an ard means a workplac e expos ur s ta dard l st ed in the Workplac e

Ex pos ur Sta dards fr Airb rne Co tamin nts

[1 5]

A us ral a ’ s model, Work Health an Safety (WHS)

R egulations, r q ir s that ex osur s an ards r pr senting the airb rne conc ntration of a p rticular

subs anc or mixtur ar not ex ce ded

The ear ex osur s an ards for 64 subs anc s an mixtur s in A us ral a The e ar , howeve , many

othe subs anc s an mixtur s hazardous t o h man health and used in workplac s that do not ha e a

man at ory ex osur s an ard es a l shed Cur ently ex osur s an ards ar not updat ed r gularly

an ma not alway s r fle t the lat es r sear h or s at e of k ow ledg e on the hazardous efe t of

chemicals Ex osur s an ards do not identify a dividing lne betwe n a healthy or u healthy working

environment Natural biolo ical v riation an the rang e of in ivid al susc ptibi ities mean some

peo le ma ex e ienc adve se health efe t below the ex osur s an ard In ad ition in some cases,

workplac ex osur s an ards ar set b sed on conside ation of b th health efe t an also w hat is

achiev ble in practic Ex osur s an ards es a lsh a s atut ory ma imum up e l mit

[1 5]

Ex osur s an ardsin A us ral a ar not designed t o be a pled t o situations out ide of a wor kplac or t o

the ex osur of peo le, lke by stan e s or nearb r sident , not dir ctly eng g ed in the wor k inv lving

the hazardous chemical Howeve , the model Work Health an Safety (WHS) A ct also r q ir s a Pe son

Con ucting a Busines or Un e taking (PCBU) t o minimiz risk t o third parties This is r g rdles of

w hethe an ex osur s andard has be n es a lshed or not The WHS R egulations r q ir that the

primary focus should alway s be on elminating or, if this is not p s ible, minimizing risk through use of

ex osur controls

The ex osur s an ards r pr sent airb rne conc ntrations of in ivid al chemical subs anc s w hich,

ac ording t o cur ent k ow ledg e, should neithe imp ir the health of nor cause u d e discomfort t o

nearly al worke s Un e A us ral a ’ s fede al sy st em, the workplac ex osur s an ards ha e leg l

s atus w hen they ar spe ifical y incorp rat ed int o Commonwealth, Stat e or Te rit ory legislation In

r commen ing a pro riat e ex osur s an ards, Safe Work A us rala an it pr de es ors ha e be n

guided b the s andards an ex e ienc of a n mbe of A us ral an and o e seas org nizations

In the many cases w he e the e is no man at ory ex osur s an ard es a lshed in A us rala, othe

es a lshed ex osur s andards or action levels can be used b PCBUs an oc up tional hy gienis s t o

as is minimizing ex osur t o chemicals

A 2.1.2 Scienc e and methods for OEL set ing

Many of the ado t ed ex osur s an ards ha e be n o tained from the Ame ican C nfe enc of