ENV /JM /M ON O(2015)15/P ART1 Unclassified - OECD

Unclassified ENV/JM/MONO(2015)15/PART1 Organisation de Coopération et de Développement Économiques Organisation for Economic Co-operation and Development 17-Dec-2015

___________________________________________________________________________________________

_____________ English - Or. English ENVIRONMENT DIRECTORATE

JOINT MEETING OF THE CHEMICALS COMMITTEE AND

THE WORKING PARTY ON CHEMICALS, PESTICIDES AND BIOTECHNOLOGY

DOSSIER ON ZINC OXIDE

- PART 1 -

Series on the Safety of Manufactured Nanomaterials

No. 52

This document is only available in PDF format.

JT03388453

Complete document available on OLIS in its original format

This document and any map included herein are without prejudice to the status of or sovereignty over any territory, to the delimitation of

international frontiers and boundaries and to the name of any territory, city or area.

EN

V/JM

/MO

NO

(2015)1

5/P

AR

T1

Un

classified

En

glish

- Or. E

ng

lish

ENV/JM/MONO(2015)15/PART1

2


3

OECD Environment, Health and Safety Publications

Series on the Safety of Manufactured Nanomaterials

No. 52

DOSSIER ON ZINC OXIDE

- PART 1 -

Environment Directorate

ORGANISATION FOR ECONOMIC CO-OPERATION AND DEVELOPMENT

Paris, 2015


4

Dossiers also published in the Series on the Safety of Manufactured Nanomaterials:

No. 44, Dossier on Gold nanoparticles (2015)

No. 45, Dossier on Cerium oxide (2015)

No. 46, Dossier on Dendrimers (2015)

No. 47, Dossier on Nanoclays (2015)

No. 48, Dossier on Fullerenes (2015)

No. 49, Dossier on Multiwalled Carbon Nanotubes (MWCNTs) (2015)

No. 50, Dossier on Single-walled Carbon Nanotubes (SWCNTs) (2015)

No. 51, Dossier on Silicon dioxide (2015)

No. 53, Dossier on Silver nanoparticles (2015)

No. 54, Dossier on Titanium dioxide (2015)

© OECD 2015

Applications for permission to reproduce or translate all or part of this material

should be made to: Head of Publications Service, [email protected],

OECD, 2 rue André-Pascal, 75775 Paris Cedex 16, France


5

ABOUT THE OECD

The Organisation for Economic Co-operation and Development (OECD) is an intergovernmental

organisation in which representatives of 34 industrialised countries in North and South America, Europe

and the Asia and Pacific region, as well as the European Commission, meet to co-ordinate and harmonise

policies, discuss issues of mutual concern, and work together to respond to international problems. Most of

the OECD’s work is carried out by more than 200 specialised committees and working groups composed

of member country delegates. Observers from several countries with special status at the OECD, and from

interested international organisations, attend many of the OECD’s workshops and other meetings.

Committees and working groups are served by the OECD Secretariat, located in Paris, France, which is

organised into directorates and divisions.

The Environment, Health and Safety Division publishes free-of-charge documents in eleven different

series: Testing and Assessment; Good Laboratory Practice and Compliance Monitoring; Pesticides;

Biocides; Risk Management; Harmonisation of Regulatory Oversight in Biotechnology; Safety of

Novel Foods and Feeds; Chemical Accidents; Pollutant Release and Transfer Registers; Emission

Scenario Documents; and Safety of Manufactured Nanomaterials. More information about the

Environment, Health and Safety Programme and EHS publications is available on the OECD’s World

Wide Web site (www.oecd.org/chemicalsafety/).

This publication was developed in the IOMC context. The contents do not necessarily reflect the

views or stated policies of individual IOMC Participating Organizations.

The Inter-Organisation Programme for the Sound Management of Chemicals (IOMC) was

established in 1995 following recommendations made by the 1992 UN Conference on

Environment and Development to strengthen co-operation and increase international co-

ordination in the field of chemical safety. The Participating Organisations are FAO, ILO, UNDP,

UNEP, UNIDO, UNITAR, WHO, World Bank and OECD. The purpose of the IOMC is to

promote co-ordination of the policies and activities pursued by the Participating Organisations,

jointly or separately, to achieve the sound management of chemicals in relation to human health

and the environment.


6

This publication is available electronically, at no charge.

For this and many other Environment,

Health and Safety publications, consult the OECD’s

World Wide Web site (www.oecd.org/chemicalsafety/)

or contact:

OECD Environment Directorate,

Environment, Health and Safety Division

2 rue André-Pascal

75775 Paris Cedex 16

France

Fax: (33-1) 44 30 61 80

E-mail: [email protected]


7

PREAMBLE

In November 2007, OECD’s Working Party on Manufactured Nanomaterials (WPMN) launched the

Sponsorship Programme for the Testing of Manufactured Nanomaterials (hereafter the Testing

Programme). The objective was to conduct specific tests, relevant to human health and environmental

safety endpoints, on a variety of manufactured nanomaterials (MN). The outcomes of the Testing

Programme were intended to assess the applicability of the existing test guidelines1 to nanomaterials, as

well as to provide useful information on any intrinsic properties of MNs, which are different from the

same bulk material with greater external dimensions. Understanding the properties of NMs is crucial to

choose appropriate strategies for hazard identification, risk assessment or risk management measures. The

Testing Programme involved delegations from OECD member countries, some non-member economies

and other stakeholders. The broad international representation, from a range of delegations enabled the

programme to pool expertise and resources without which this programme would not have been possible.

Before launching the Testing Programme, the WPMN first identified a broad list of possible

nanomaterials, and the list was later adjusted to a final selection of eleven MNs for testing2. This list

comprised: i) fullerenes (C60); ii) single-walled carbon nanotubes (SWCNTs); iii) multi-walled carbon

nanotubes (MWCNTs); iv) silver nanoparticles; v) titanium dioxide; vi) cerium oxide; vii) zinc oxide;

viii) silicon dioxide; ix) dendrimers; x) nanoclays; and xi) gold nanoparticles. One fundamental criterion

for selecting these materials was that they should be either in commercial use at the time or expected to be

in the near future. At the same time, other considerations were also given attention, such as the production

volume of the materials, the likely availability of such materials for testing and the existing information

that would readily be available on the materials.

It was also agreed that 59 endpoints would be addressed3 for each material corresponding to the following

categories: i) nanomaterial information/ identification; ii) physical-chemical properties and material

characterisation; iii) environmental fate; iv) toxicological and eco-toxicological effects; v) environmental

toxicology; vi) mammalian toxicology; and vii) material safety. These endpoints were judged to be most

important based largely on the general experience of testing chemicals, while taking into account the

potentially different or new properties of nanomaterials. It is worth noticing that it was not expected that

testing for all of the listed endpoints would be necessary for each of the selected MNs.

To assist with the Testing Programme, the WPMN developed two documents: i) a Preliminary Review of

OECD Test Guidelines for their Applicability to Manufactured Nanomaterials

[ENV/JM/MONO(2009)21]; and ii) Guidance Manual for the Testing of Manufactured Nanomaterials:

OECD's Sponsorship Programme (Guidance Manual) in 2009, which was subsequently updated in 2010

1 The OECD Test Guidelines are a collection of internationally agreed test methods used by government, industry and

independent laboratories. They are used to determine the safety of chemicals.

http://www.oecd.org/chemicalsafety/testing/oecdguidelinesforthetestingofchemicals.htm

2 Originally Iron nanoparticles, Aluminium, Carbon black, and Polystyrene were suggested but later withdrawn and replaced by

gold nanoparticles.

3 As specified in the Guidance Manual, “address” includes the term “completed” which provides that all dossiers will contain the

identified endpoint information. Note that for some endpoints (for example, solubility) it is specified that the endpoint must be

“completed”. In such instances “completed” means that all Dossiers will be providing this endpoint information.

http://www.oecd.org/chemicalsafety/testing/oecdguidelinesforthetestingofchemicals.htm


8

[ENV/JM/MONO(2009)20/REV]4. The objective of this Guidance Manual was to guide sponsors

5 in the

testing of the materials while ensuring that the information collected was reliable, accurate, consistent and

therefore also comparable. The Guidance Manual addressed a whole range of issues including the

organisation of the work.

The Guidance Manual contains detailed information on the selected endpoints for testing and

recommendations on sample preparation and dosimetry.

The Guidance Manual also described the development of Dossier Development Plans (DDPs). These

plans were prepared by Lead sponsors, Co-sponsors together with contributors to describe the specific

plan for the testing of each nanomaterial including when and where the testing will be undertaken and by

whom. The DDPs also included information on the materials to be tested as well as information on issues

such as sample preparation and dosimetry. Each of the DDPs was prepared and reviewed by the WPMN

before testing work began.

Based on the lessons learned during the Testing Programme, the WPMN also developed Guidance on

Sample Preparation and Dosimetry for the Safety Testing of Manufactured Nanomaterials

[ENV/JM/MONO(2012)40]. This latter document is an update of an earlier text first published in 2010.

The work on OECD’s Testing Programme was completed by the end of 2013. In June 2014 the WPMN

agreed that for each nanomaterial the dataset would be published in IUCLID printed format6 7

. The

document will include the protocols and methods to allow their wider use (regulators and researchers).

The dataset in this document has been declassified and made publicly available and it is expected

regulators and researchers will wish to use it. Due to a broad dissemination of the data and the exploratory

setting in which they were developed there are a number of limitations in using the data of which

potential users should be aware. The programme focused on answering scientific questions in the field of

the OECD test guidelines but not to provide conclusions on the hazard or risk of the materials selected.

The data contained within these dossiers is raw data and has not been evaluated by either the programme

sponsors or the WPMN. Any conclusions found within these dossiers are under the responsibility of the

researchers who made them. The absence of data for some endpoints may be a gap for some endpoints but

for other end points there may not if the data was not considered necessary. Although the programme

ensured a broad participation of many stakeholders it was not intended to arrive at any pre-defined

regulatory datasets requirements or risk assessment decisions. It was recognised from the beginning that

4 It is worth noting that while the Guidance Manual for Sponsors was primarily intended as a guide to WPMN’s Testing

Programme, it is also expected that it will be of value to anyone involved in testing NMs.

5 The Guidance Manual noted, for example, that there could be three levels of participation to the programme. Lead sponsors,

who would assume responsibility for conducting or coordinating all of the testing, determined to be appropriate for each of the

endpoints for a specific nanomaterial. In some cases, “joint lead” arrangements were developed. Co-sponsors conducted some of

the testing determined to be appropriate and feasible to address the endpoints for a specific listed nanomaterial. Contributors

provided test data, reference or testing materials or other relevant information to the lead and co-sponsors.

6 IUCLID is a software programme for the administration of data on chemical substances. Although it was originally developed

to fulfill requirements in the EU for the evaluation and control of the risks of existing chemical substances, it is used by many

others.

7 SIAR = SIDS Initial Assessment Report (SIDS = Screening Information Data Set)


9

the exploratory nature of the work would require subsequent follow-up work for example to review the

specific needs that may arise when performing risk assessment of nanomaterials. In this context, the

programme's ultimate goal, to add to the knowledge of the properties of nanomaterials, would form a

cornerstone.


10

FOREWORD

As part of its Programme on the Safety of Manufactured Nanomaterials, OECD launched the Sponsorship

Programme for the Testing of Manufactured Nanomaterials (hereafter the Testing Programme). The

objective was to conduct specific tests, relevant to human health and environmental safety endpoints, on a

variety of manufactured nanomaterials (MN). The Testing Programme mainly aimed to assess the

applicability of the existing test guidelines to nanomaterials, as well as to provide useful information on

any intrinsic properties of MNs, which are different from the same bulk material with greater external

dimensions.

This document presents the Dossier of Zinc Oxide. This nanomaterial has been tested for a number of

endpoints for: i) Nanomaterials Information / Identification; ii) Physical-Chemical Properties; iii)

Environmental Fate; iv) Environmental Toxicology; v) Mammalian Toxicology; and vi) Material Safety.

They have been analysed using OECD Guidelines for the Testing of Chemicals (TG)8. The data is

presented in an IUCLID9 style format and includes the protocols and methods used (see Preamble).

The Business & Industry Advisory Committee to the OECD (BIAC) via the Nanotechnology Industries

Association (NIAC) led the Testing Programme on Zinc Oxide. This included the determination of the

tests that were appropriate, performing a number of tests, as well as coordinating tests and results

obtained by other the participating stakeholders. This programme has benefited from the co-sponsorship

and the contribution of Australia, the U S Food and Drug Administration (FDA), and Spain.

This document is published under the responsibility of the Joint Meeting of the Chemicals Committee and

Working Party on Chemicals, Pesticides and Biotechnology of the OECD.

8 http://www.oecd.org/env/testguidelines

9 IUCLID is a software programme for the administration of data on chemical substances. It was originally developed to fulfil

requirements in the EU for the evaluation and control of the risks of existing chemical substances. It is specifically relevant in the

context of an international programme for the initial assessment of chemical substances.

http://www.oecd.org/env/testguidelines


11

ACKNOWLEDGMENTS

The OECD Secretariat and the Working Party on Manufactured Nanomaterials wish to thank the Business

& Industry Advisory Committee to the OECD (BIAC) for leading the Testing Programme for Zinc Oxide.

They are specifically grateful to David Carlander from the Nanotechnology Industries Association. In

addition, we appreciate the efforts made by other countries that participated in the Testing Programme, in

particular to Australia, the US Food and Drug Administration (FDA), and Spain.


12

TABLE OF CONTENTS

- Part 1 -

PREAMBLE ................................................................................................................................................... 7

FOREWORD ................................................................................................................................................ 10

ACKNOWLEDGMENTS ............................................................................................................................ 11

1. GENERAL INFORMATION .................................................................................................................. 14

1.1 Identification ........................................................................................................................................ 14 1.2 Composition ......................................................................................................................................... 14 1.3 Identifiers ............................................................................................................................................. 15 1.4 Analytical information ......................................................................................................................... 15 1.5 Joint submission ................................................................................................................................... 15 1.6 Sponsors ............................................................................................................................................... 15 1.7 Suppliers .............................................................................................................................................. 15 1.8 Recipients ............................................................................................................................................. 15 1.9 Product and process orientated research and development .................................................................. 15

2. CLASSIFICATION & LABELLING AND PBT ASSESSMENT .......................................................... 15

3. MANUFACTURE, USE AND EXPOSURE ........................................................................................... 15

4. PHYSICAL AND CHEMICAL PROPERTIES....................................................................................... 15

4.1 Appearance/physical state/colour ........................................................................................................ 15 4.2 Melting point/freezing point ................................................................................................................ 24 4.3 Boiling point ........................................................................................................................................ 24 4.4 Density ................................................................................................................................................. 24 4.5 Particle size distribution (Granulometry) ............................................................................................. 24 4.6 Vapour pressure ................................................................................................................................... 54 4.7 Partition coefficient .............................................................................................................................. 54 4.8 Water solubility .................................................................................................................................... 54 4.9 Solubility in organic solvents / fat solubility ....................................................................................... 70 4.10 Surface tension ................................................................................................................................... 70 4.11 Flash point .......................................................................................................................................... 70 4.12 Auto flammability .............................................................................................................................. 70 4.13 Flammability ...................................................................................................................................... 70 4.14 Explosiveness ..................................................................................................................................... 70 4.15 Oxidising properties ........................................................................................................................... 70 4.16 Oxidation reduction potential ............................................................................................................ 70 4.17 Stability in organic solvents and identity of relevant degradation products ...................................... 77 4.18 Storage stability and reactivity towards container material ............................................................... 77 4.19 Stability: thermal, sunlight, metals .................................................................................................... 77 4.20 pH ....................................................................................................................................................... 82 4.21 Dissociation constant ......................................................................................................................... 82 4.22 Viscosity ............................................................................................................................................ 82 4.23 Additional physico-chemical information ......................................................................................... 82 4.24 Agglomeration/aggregation ............................................................................................................... 88 4.25 Crystalline phase ................................................................................................................................ 93 4.26 Crystallite and grain size .................................................................................................................. 101


13

4.27 Aspect ratio/shape ............................................................................................................................ 109 4.28 Specific surface area ........................................................................................................................ 109 4.29 Zeta potential ................................................................................................................................... 122 4.30 Surface chemistry............................................................................................................................. 130 4.31 Dustiness .......................................................................................................................................... 147 4.32 Porosity ............................................................................................................................................ 150 4.33 Pour density ..................................................................................................................................... 159 4.34 Photocatalytic activity ...................................................................................................................... 161 4.35 Radical formation potential .............................................................................................................. 173

- Part 2 -

5. ENVIRONMENTAL FATE AND PATHWAYS ................................ ENV/JM/MONO(2015)15/PART2

6. ECOTOXICOLOGICAL INFORMATION ........................................................................................ 1092

- Part 3 -

7. TOXICOLOGICAL INFORMATION ............................................... ENV/JM/MONO(2015)15/PART3

8. ANALYTICAL METHODS ................................................................ ENV/JM/MONO(2015)15/PART3

9. RESIDUES IN FOOD AND FEEDINGSTUFFS ............................... ENV/JM/MONO(2015)15/PART3

10. EFFECTIVENESS AGAINST TARGET ORGANISMS ................. ENV/JM/MONO(2015)15/PART3

11. GUIDANCE ON SAFE USE ............................................................ ENV/JM/MONO(2015)15/PART3

12. LITERATURE SEARCH .................................................................. ENV/JM/MONO(2015)15/PART3

13. ASSESSMENT REPORTS ............................................................... ENV/JM/MONO(2015)15/PART3

14. INFORMATION REQUIREMENTS ............................................... ENV/JM/MONO(2015)15/PART3


14

Substance: CH322-00371_Nano ZnO_Revision

1. GENERAL INFORMATION

1.1 Identification

Substance identification

Chemical name CH322-00371_Nano ZnO_Revision

Reference substance

zinc oxide / oxozinc / 1314-13-2

EC number EC name

215-222-5 zinc oxide

CAS number CAS name

1314-13-2

IUPAC name

oxozinc

1.2 Composition

Substance composition

Name NM110 Zinc Oxide Nano

Brief description Uncoated

Degree of purity

> 99 % (w/w)


Brief description Coated triethoxycaprylylsilane (2%)

Degree of purity

> 96 % (w/w)


Brief description Uncoated

Degree of purity

> 99.5 % (w/w)

Name NM113 Zinc Oxide

Brief description supplier: Sigma Aldrich, uncoated non-nanosized ZnO


15

1.3 Identifiers

1.4 Analytical information

1.5 Joint submission

1.6 Sponsors

1.7 Suppliers

1.8 Recipients

1.9 Product and process orientated research and development

2. CLASSIFICATION & LABELLING AND PBT ASSESSMENT

3. MANUFACTURE, USE AND EXPOSURE

4. PHYSICAL AND CHEMICAL PROPERTIES

4.1 Appearance/physical state/colour

Endpoint summary: Appearance/physical state/colour

Administrative Data

Short description of key information

The shapes of the particle as well as the morphology of the NM 110, NM 111 and NM 113 were

comparable. The three NM were described as polyhedral with variable morphology and size. The NM 112

was determined to be distinctly different to all the other samples. NM 112 was described to appears as

near spherical and with the smallest particle size compared to the other three materials.

Key value for chemical safety assessment

Physical state at 20°C and 1013 hPa

solid

Discussion

CSIRO, 2012 characterised the ZnO nanomaterials. All samples were glow discharged in nitrogen for 30

seconds to render them hydrophilic. Samples were dispersed by briefly sonicating a few milligram of the

material in approximately 20 µl ethanol to form a milky dispersion. 5 µL of dispersion was applied to the

freshly glow-discharged grids. After 2 min adsorption time, excess dispersion was wicked off using filter

paper and the grids were air-dried for 15 min. Grids were examined using a Tecnai 12 TEM (FEI,

Eindhoven, Netherlands) operating at 120 kV, and micrographs were recorded using an Olympus

Megaview III CCD camera (Tokyo, Japan) running AnalySiS imaging software (Olympus) at a variety of

magnifications chosen to show both the aggregation/agglomeration state of the samples (lower


16

magnifications e.g. 6000x) as well as particle morphology (higher magnifications e.g. 100000x -

360000x).The four TEM images of the nanomaterials showing varied particle sizes and shapes in the

different samples. The TEM image of NM-110 indicated that the primary particles appeared polyhedral

with variable morphology and size. Two main types of morphology could be distinguished:

- Particles with aspect ratio close to 1 (typically 20 – 250 nm size and very few particles of approx.

400 nm size) and hexagonal morphology

- Particles with aspect ratio 2 to 7.5 (50 – 350 nm) with cubic, tetragonal and orthorhombic morphologies.

The TEM image of NM-111 indicated that the primary particles appeared polyhedral and with variable

morphology as observed in NM-110, but with different size distributions:

- Particles with aspect ratio near 1 (~90 % in the 20 – 200 nm range)

- Particles with aspect ratio 2 to 8.5 (~90 % in the 10 – 450 nm range

The TEM image of NM-112 indicated that primary particles were near spherical (rather than polyhedral)

with regular morphology and a relatively homogenous size distribution. Generally, particles had an aspect

ratio close to 1, with sizes varying between 20 and 50 nm and appeared distinctly different to all the other

samples (NM-110, NM-111 and NM-113).

- Particles with aspect ratio near 1 (typically in the 80 – 100 nm range)

- Particles with aspect ratio > 2 (typically in the 180 – > 200 nm range)

The TEM image of NM-113 showed that the sample was composed of polyhedral particles with sizes

ranging generally between 100 to 200 nm, and with some larger agglomerates.

Deakin University, 2012 inestigated the shape of the nanomaterials. TEM specimen was prepared by

evaporating a drop of the nanoparticle dispersion on a carbon-coated specimen grid. Grids were examined

using a JEOL JEM-2100 TEM operating at 200 kV. No further details about the sample preparation and

the method parameter were available. The TEM image of NM-110 and NM-111 indicated wide

distributions of shapes and sizes of primary particles. Furthermore the geometrical shapes indicated high

crystallinity and low surface defects. The TEM image of NM-112 indicated near spherical shape of the

primary particles and a high number of surface defects. The TEM image of NM-113 indicated

geometrical shapes and large particle sizes (>100 nm in diameter).

Conclusion

The TEM images analyses in the course of the two studies yielded comparable results. The appearances

of the particle as well as the morphology of the NM 110, NM 111 and NM 113 were comparable. The

three NM were described as polyhedral with variable morphology and size. The NM 112 was determined

to be distinctly different to all the other samples. NM 112 was described to appears as near spherical and

with the smallest particle size compared to the other 3 materials.

Endpoint study record: key_2012-03-05_Australia_CSIRO_Appearance_TEM

Administrative Data

Purpose flag key study

Study result type experimental result

Reliability 1 (reliable without restriction)

Rationale for reliability incl.

deficiencies

The documentation provide in the full study report was sufficient for the

data evaluation.


17

Data source

Reference

Reference

type

Author Year Title Bibliographic

source

Testing

laboratory

Report

no.

Owner

company

Company

study no.

Report

date

study

report

Casey,

Yin,

McCall

2012 Summary of

physicochemical

properties of

OECD Zinc

Oxide (ZnO)

samples

Commonwealth

Scientific and

Industrial

Research

Organisation

(CSIRO)

2013-

03-05

Data access

data submitter is data owner

Data protection claimed

yes, but willing to share

Materials and methods

Test guideline

Qualifier Guideline Deviations

no guideline required

Principles of method if other than guideline

For all samples, carbon-coated grids (copper, 300 mesh) were glow discharged in nitrogen for 30 seconds

to render them hydrophilic. Samples were dispersed by briefly sonicating a few mg of the material in

approximately 20μL ethanol to form a milky dispersion. 5μL of dispersion was applied to the freshly

glow-discharged grids. After 2 mins adsorption time, excess dispersion was wicked off using filter paper

(Whatman 541) and the grids were air-dried for 15 minutes.Grids were examined using a Tecnai 12 TEM

(FEI, Eindhoven, Netherlands) operating at 120 kV, and micrographs were recorded using an Olympus

Megaview III CCD camera (Tokyo, Japan) running AnalySiS imaging software (Olympus) at a variety of

magnifications chosen to show both the aggregation/agglomeration state of the samples (lower

magnifications e.g. 6000x) as well as particle morphology (higher magnifications e.g. 100 000x - 360

000x).

GLP compliance

no

Test materials

Identity of test material same as for substance defined in section 1 (if not read-across)

yes

Test material identity

Identifier Identity

CAS number 1314-13-2

EC number 215-222-5

IUPAC name oxozinc

other: OECD

Sponsorship

NM110, NM111, NM112, NM113


18

Test material form

nanomaterial

Details on test material

-Name of test material: Z-cote ®

Code: NM110Supplier: BASF SE

Surface coating: none

- Substance type: Inorganic

- Physical state: solid powder, nano-form

-Name of test material: Z-cote HP


Surface coating: triethoxycaprylylsilane (2%)



-Name of test material: Nanosun

TM Code: NM112

Supplier: Micronisers




-Name of test material: Zinc Oxide

Code: NM113Supplier: Sigma-Aldrich




Results and discussion


solid

Form

nanomaterial

Odour

odourless

Substance type

inorganic

Overall remarks, attachments

Remarks on results including tables and figures

Typical TEM images of NM110, NM111, NM112 and NM113 are shown in Figure 1.

The TEM image of NM110 indicates that the primary particles appear polyhedral with variable

morphology and size. Qualitatively, the aspect ratio of the particles is generally greater than 2, with the

width 30-50 nm and the length 100- 200 nm.

The TEM image of NM111 shows polyhedral particles. Two morphological types appear distinguishable:

1. Particles with aspect ratio near 1 (typically less than 60 nm in diameter)

2. Particles with aspect ratio between 1 and 2 (width 50-100 nm and length 100-200 nm).

The TEM image of NM112 indicates that primary particles are near spherical (rather than polyhedral)


19

with regular morphology and a relatively homogenous size distribution. Generally, particles have an

aspect ratio close to 1, with sizes varying between 20 and 50 nm.

The TEM image of NM113 shows that the sample is composed of polyhedral particles with sizes ranging

generally between 100 to 200 nm, and with some larger agglomerates.

Attached full study report


NM_110_ANNEX_A11_OECD ZnO Appearance TEM CSIRO Australia.docx / 404.61 KB (application/octet-

stream): ENV/JM/MONO(2015)15/ANN1

Illustration (picture/graph)

Applicant's summary and conclusion

Conclusions

The TEM image of NM110 indicates that the primary particles appear polyhedral with variable

morphology and size. Qualitatively, the aspect ratio of the particles is generally greater than 2, with the


20

width 30-50 nm and the length 100- 200 nm.

Executive summary

A study was conducted by CSIRO, 2012) to characterise the ZnO nanomaterials. All samples were glow

discharged in nitrogen for 30 seconds to render them hydrophilic. Samples were dispersed by briefly

sonicating a few milligram of the material in approximately 20 µl ethanol to form a milky dispersion. 5

µL of dispersion was applied to the freshly glow-discharged grids. After 2 min adsorption time, excess

dispersion was wicked off using filter paper and the grids were air-dried for 15 min. Grids were examined

using a Tecnai 12 TEM (FEI, Eindhoven, Netherlands) operating at 120 kV, and micrographs were

recorded using an Olympus Megaview III CCD camera (Tokyo, Japan) running AnalySiS imaging

software (Olympus) at a variety of magnifications chosen to show both the aggregation/agglomeration

state of the samples (lower magnifications e.g. 6000x) as well as particle morphology (higher

magnifications e.g. 100000x - 360000x).The four TEM images of the nanomaterials showing varied

particle sizes and shapes in the different samples. The TEM image of NM-110 indicated that the primary

particles appeared polyhedral with variable morphology and size. Two main types of morphology could

be distinguished:

- Particles with aspect ratio close to 1 (typically 20 – 250 nm size and very few particles of approx.

400 nm size) and hexagonal morphology

- Particles with aspect ratio 2 to 7.5 (50 – 350 nm) with cubic, tetragonal and orthorhombic morphologies.

The TEM image of NM-111 indicated that the primary particles appeared polyhedral and with variable

morphology as observed in NM-110, but with different size distributions:

- Particles with aspect ratio near 1 (~90 % in the 20 – 200 nm range)

- Particles with aspect ratio 2 to 8.5 (~90 % in the 10 – 450 nm range

The TEM image of NM-112 indicated that primary particles were near spherical (rather than polyhedral)

with regular morphology and a relatively homogenous size distribution. Generally, particles had an aspect

ratio close to 1, with sizes varying between 20 and 50 nm and appeared distinctly different to all the other

samples (NM-110, NM-111 and NM-113).

- Particles with aspect ratio near 1 (typically in the 80 – 100 nm range)

- Particles with aspect ratio > 2 (typically in the 180 – > 200 nm range)

The TEM image of NM-113 showed that the sample was composed of polyhedral particles with sizes

ranging generally between 100 to 200 nm, and with some larger agglomerates.

Endpoint study record: supporting_2012-03-

06_Australia_Deakin_Appearance_TEM

Administrative Data

Purpose flag supporting study


Reliability 2 (reliable with restrictions)

Rationale for

reliability incl.

deficiencies

The documentation provide in the full study report was sufficient for the data evaluation but there were

some limitations within the study. The procedure for subsampling and data analysis and representation

was not explained in detail. According to the ECHA Guidance it is recommended to provide a

combination of terms and/or measurands to describe shape. Geometric macroshape descriptors were

available in ISO 9276-6:2008. Thus, the study was regarded as reliable with restrictions.


21

Data source

Reference

Reference

type


source

Testing

laboratory

Report

no.

Owner

company

Company

study no.

Report

date

study

report

T Tsuzuki,

R He, R

Chaudhary

2012 Appearance

(by TEM)

Deakin

University

2012-

01-03

Data access





Test guideline


no guideline followed


No details provided.

GLP compliance

no

Test materials


yes


Identifier Identity


CAS name 215-222-5

IUPAC name oxozinc

other: OECD Sponsorship Programme NM110, NM111, NM112, NM113

Test material form

nanomaterial


-Name of test material: Z-cote ®Code: NM110

Supplier: BASF SE





Code: NM111

Supplier: BASF SE






22

TM Code: NM112






Code: NM113

Supplier: Sigma-Aldrich




Any other information on materials and methods incl. tables

Measurement principle: Transmission electron microscopy

Instrument: JEOL JEM-2100

Sample preparation: TEM specimen was prepared by evaporating a drop of the nanoparticle dispersion on

a carbon-coated specimen grid.

Measurement parameters: electron beam energy 200 kV

Results and discussion


solid

Form

nanomaterial

Odour

odourless

Substance type

inorganic



Geometrical shapes with some sintering, indicating that the synthesis technique is a vapour phase

involving a high temperature. Geometrical shapes indicate high crystallinity and low surface defects.

Wide distributions of shapes and sizes of primary particles.



NM_110_ANNEX_A12_OECD ZnO TEM Australia Deakin.pdf / 430.09 KB (application/octet-stream):

ENV/JM/MONO(2015)15/ANN1



23


Conclusions

The TEM image of NM 110 and NM 111 indicated wide distributions of shapes and sizes of primary

particles. Furthermore the geometrical shapes indicated high crystallinity and low surface defects. The

TEM image of NM 112 indicated near spherical shape of the primary particles and a high number of

surface defects. The TEM image of NM 113 indicated geometrical shapes and large particle sizes (>100

nm in diameter).

Executive summary

Deakin University, 2012 investigated the shape of the nanomaterials. TEM specimen were prepared by

evaporating a drop of the nanoparticle dispersion on a carbon-coated specimen grid. Grids were examined

using a JEOL JEM-2100 TEM operating at 200 kV. No further details about the sample preparation and

the method parameter were available. The TEM image of NM-110 and NM-111 indicated wide

distributions of shapes and sizes of primary particles. Furthermore the geometrical shapes indicated high


24

crystallinity and low surface defects. The TEM image of NM-112 indicated near spherical shape of the

primary particles and a high number of surface defects. The TEM image of NM-113 indicated

geometrical shapes and large particle sizes (>100 nm in diameter).

4.2 Melting point/freezing point

4.3 Boiling point

4.4 Density

4.5 Particle size distribution (Granulometry)

Endpoint summary: Particle size distribution (Granulometry)

Administrative Data


The basic assumption (NM 113 as “bulk material” has the largest particle size) could be confirmed by

particle size determination. Throughout all measurements NM 112 was determined to have the smallest

particle size compared to the other nanomaterials. The results of the different measurements of NM 110

and NM 111 samples showed that these 2 nanomaterials have a comparable particle size. This general

trend was also observed for the endpoints density and surface area, which are directly linked to the

particle size.The mean particle size of the nanomaterials was determined to be 77.5 ± 18 nm for NM-110,

75.2 ± 7.6 nm for NM-111, 33.75 ± 6.2 nm for NM-112 and 149.7 ± 25 nm for nm 113 respectively.

Discussion

Several studies using different analytical techniques were available for particle size of the investigated

nanomaterials and were assessed in a weight of evidence approach.

The National Physical Laboratory (NPL), 2010 investigated the particle size distribution of the 4 test

items. In accordance with the ECHA guidance size distribution was investigated using multiple

techniques. According to the OECD guidance no single suitable and validated procedure is currently

available. Therefore different analytical measures using Transmission Electron Microscopy (TEM),

Scanning Electron Microscopy (SEM), Dynamic Light Scattering (DLS), Scanning Mobility Particle

Sizer (SMPS), centrifugal sedimentation method (CPS) and X-ray diffraction (XRD) were conducted

NPL, 2010 investigated the particle size of the test items NM-110, NM-112 and NM-113 using TEM.

Therefore a nanomaterial sample dispersion of 50 mg/L was made in accordance to the OECD

recommended protocol for sample dispersion. The dispersion was allowed to settle out in order to remove

the larger particles naturally as it sediments by gravity. The purpose of this sedimentation was to remove

large micron size particles. On day 2, there was clear visible evidence of sedimentation events of the

larger micron size particles having taken place. Two microliter aliquots were extracted from the top layer

(an opaque but slightly transparent layer). The sample was placed on TEM grids and allowed to air dry

for 10 minutes; grids are formvar/carbon on 400 meshes copper. Grids were used as supplied with no

further modification. All images were recorded using a Hitachi 2300 A instrument operated at 200 kV.

An adequate magnification was chosen for image acquisition e.g. for the estimation of primary particle

mean diameter. TEM micrographs were analysed by manually tracing contours of primary particles on to

a transparency sheet. The transparency sheet was scanned for further image analysis using ImageJ

software, which automatically calculated particle diameter dimensions. Furthermore an extremely small

area of the sample could be analysed, which might not be representative enough for the whole sample.


25

The comparatively small share of evaluated particles results in limited statistical precision. The mean

primary particle size was quoted with the corresponding standard deviation which represents the

broadness of the size distribution.

Based on the measurements of 77 particles the mean particle size of the NM 110 was determined to be

75.4 ± 58.4 nm.


30.5 ± 13.8 nm.

Based on the measurements of 87 particles the mean particle size of NM 113 was determined to be 165.2

± 90.4 nm.

NPL, 2010 investigated the particle size of the test items NM-110, NM-112 and NM-113 also using DLS.

Hydrodynamic size (z-average mean) measurements were obtained using a Zetasizer Nano ZS (Malvern

Instruments,UK) equipped with a 633 nm laser. A reference standard (polystyrene, latex bead, nominal

size of 100 nm, NIST traceable) was used to qualify the performance of the instrument. Sample analysis

involved filling of a disposable capillary cell (DTS1060, Malvern). Prior to their use, these cells were

thoroughly cleaned with ethanol and de-ionised water, as recommended by the instrument vendor.

Individual cell was then filled with the appropriate sample and flushed before re-filling; measurement was

carried out on the second filling. Malvern Instrument’s Dispersion Technology software (Version 4.0)

was used for data analysis. For particle size it was the z average diameter (the mean hydrodynamic

diameter) that was reported. The protocol for sample preparation was described by Tantra, R., Jing,S. and

Gohil, D.in Technical issues surrounding the preparation, characterisation and testing of nanoparticles for

ecotoxicological studies, in Environmental Toxicology 3, V.B. Popov, C.A. , Editor. 2010, WIT Press p.

165-176. The mean particle size of three distributions was determined to be 275 nm for NM-110, 253 nm

for NM-112 and 508 nm for NM-113 respectively.

NPL, 2010 determined the particle size distribution of Aerosolised nanomaterials by SMPS consisting of

a Differential Mobility Analyser (DMA) and Condensation Particle Counters (CPC) system. The DMA

within the SMPS was calibrated using reference material polystyrene latex beads from NIST. TSI

Fluidised Bed Aerosol Generator (FBAG) was used to produce an aerosol from the dry powder sample.

After introduction of the nanomaterial into the FBAG, the aerosol generated was allowed to stabilise for a

day prior to sending the aerosol to an SMPS. The CPC within the SMPS setup were calibrated according

to NPL’s UKAS accredited (ISO 17025) procedure, using an internally calibrated Faraday Cup

Electrometer and soot generator (model CAST 2). The SMPS was set to record at 4 minute intervals; at

least 6 SMPS scans of 200 seconds each were used for analysis. The data was processed using TSI

Aerosol Instrument Management (AIM) software, in which the mean size distribution from the stable

time segment was estimated. The size distribution was also analysed using an in-house curvefitting

program (as implemented in a recent SMPS intercomparison at METAS). The range of the SMPS size

distribution measurement was 16-660 nm. The Geometric Mean diameter and Geometric Standard

Deviation for each size distribution was estimated by fitting a log-normal curve to the size distribution

data plot. Each plot represents the mean values of at least 6 SMPS scans of 200 seconds each. The

Geometric Mean Particle Size was determined to be 289 nm for NM-110, 400 nm for NM-111, 269 nm


NPL, 2010 investigated the particle size distribution of the NM-110 by SEM. SEM images were obtained

using a Supra 40 field emission scanning electron microscope from Carl Zeiss in which the optimal

spatial resolution of the microscope was a few nanometres. In-lens detector images were acquired at an

accelerating voltage of 15 kV, a working distance of ≈ 3 mm, and a tilt angle 0°. SEM instrument was

calibrated using a SIRA grid calibration set (SIRA, Chislehurst, Kent, UK). These are metal replicas of

cross ruled gratings of area of 60 mm2 with 19.7 lines/mm for low magnification and 2160 lines/mm for

high magnification calibrations, accurate to 0.2 %. For analysis of the “as received” nanoparticle powder,

a sample of the powder was sprinkled over a SEM carbon adhesive disc; one side of the carbon disc was


26

placed securely on a metal stub, whilst the other side was exposed to the nanoparticle powder. Excess

powder was removed by gently tapping the stub on its side until a light coating of powder on the surface

became apparent. An adequate magnification was chosen for image acquisition e.g. for the estimation of

primary particle mean diameter. The shape and limits of the primary particles should become apparent.

The SEM images were opened in ImageJ (a free image analysis program produced and distributed by the

National Institute of Health, US) installed on a Tablet PC (DELL XT1), and for each image at least 20

distinct particles identified, to measure at least 100 particles per replicate. The particles chosen were an

array of sizes to accurately represent the variety in the sample. Using the Pencil Tool (pencil width 2

pixels, colour black) the outlines of these particles were drawn using the supplied digital ‘pen’. The image

was calibrated with reference to the scale marking on the image. Then the threshold level adjusted so that

only the particle outlines were highlighted. The particles were then measured using the Analyze Particles

tool, which produced a results table of various measurements of the particles. This process was repeated

for each image of the particular replicate (seven in total – giving a minimum of 140 particles identified.

Feret’s diameter – also called the “maximum calliper length”, i.e. the longest distance between any two

points along the selection boundary - was recorded and the scientific graphing and analysis software,

SciDAVis, where a histogram (bin size 2) was plotted, showing the particle size distribution of the

replicate. The entire process was repeated for each replicate (r1, r2, r3) for all six sub-samples of both the

ZnO and NM-110 samples. From the SEM image a broad range of particle sizes were seen from small

(20-50 nm) circular or spherical particles, to much larger (>100 nm) rod shaped particles. Overall, the

particle size distributions between replicates per sub-sample are very similar, only differing by an average

range of 14.6nm for the ZnO-HP1 sample and 8.4nm for the ZnO sample. An exception however, is the

difference in the replicates of sub-sample 0830 for NM-100.The mean particle size for NM-110 is in the

range 70-80nm, and for the ZnO sample the range 100-111nm. Except for sub-sample 1455 which is

93.3nm, larger than that of the other sub-samples for NM-110 and for ZnO, sub-sample 0599 at 90.3nm

has a comparatively small mean particle size.

In conclusion the study conducted by CSIRO, 2012 was regarded as most reliable as the methodical

comparison of particle size lead to high reliability as the results using different analytical techniques leads

to comparable results. These results were furthermore supplemented by the other studies.

Endpoint study record: WoE_2010_06-17_NPL for PROSPECT_Particle

size,size distribution_TEM

Administrative Data

Purpose flag weight of evidence



Rationale for

reliability incl.

deficiencies

The study was regarded as in general scientifically acceptable. Nevertheless, only limited

documentation of the data evaluation was provided in the full study reports. Thus, it was not

possible to assess the validity of the provided data as the sample preparation and the data

analysis could have a significant influence to the result. In general it should be possible to

calculate the Particle size distribution from TEM data. Furthermore respective ISO guidelines

are available (ISO/TR 27628:2007. ISO/13322-1:2004 and ISO/13322-2:2006). It was not

indicated in the report if the experiment was conducted in accordance with these guidance

documents (in the absence of a validated OECD guideline). The evaluated particles (ca. 1,000

was regarded as sufficient for data evaluation as the NanoDefine project and its recently

published guidance on TEM evaluation does not require more than 1000 particles.

Furthermore the used protocol for sample preparation is not recommended by NanoDefine or

other sources, and can introduce biais towards smaller particles. Thus, the study was regarded

as reliable with restrictions.


27

Data source

Reference

Reference

type


source

Testing

laboratory

Report

no.

Owner

company

Company

study no.

Report

date

study

report

National

Physical

Laboratory

2010 PROTOCOL

for TEM

image

acquisition

and analysis

National

Physical

Laboratory

Hampton

Road

Teddington

Middlesex

TW11 0LW

2010-

06-17

Data access





Type of distribution

other: determined by Transmission electron microscopy


Nanomaterial sample dispersion of 50 mg/L was made in accordance to the OECD recommended

protocol for sample dispersion. The dispersion was allowed to settle out in order to remove the larger

particles naturally as it sediments by gravity. The purpose of this sedimentation was to remove large

micron size particles; TEM sample preparation is crucial and particles up to ~ 200 nm thick are required

for acquiring successful transmission images. On day 2, there was clear visible evidence of sedimentation

events of the larger micron size particles having taken place. 2 microliter aliquots were extracted from the

top layer (an opaque but slightly transparent layer). The sample was placed on TEM grids and allowed to

air dry for 10 minutes; grids are formvar/carbon on 400 mesh copper (purchased from Agar Scientific

UK). Grids were used as supplied with no further modification. All images were acquired using a Hitachi

2300A instrument operated at 200 kV. The instrument is based at the University of Surrey, Guildford and

analysis was carried out by NPL personnel. An adequate magnification was chosen for image acquisition

e.g. for the estimation of primary particle mean diameter. The shape and limits of the primary particles

should become apparent. TEM micrographs were analysed by manually tracing contours of primary

particles on to a transparency sheet. The transparency sheet was scanned for further image analysis using

ImageJ software, which automatically calculated particle diameter dimensions. The mean primary particle

size was quoted with the corresponding standard deviation; note that the s.d. here represents the broadness

of the size distribution (not error).

GLP compliance

no

Test materials


yes


28


Identifier Identity


EC number 215-222-5

IUPAC name oxozinc


Test material form

nanomaterial



Code: NM110

Supplier: BASF SE





Code: NM111

Supplier: BASF SE





TM Code: NM112






Code: NM113






Instruments Hitachi 2300A

Results and discussions

Any other information on results incl. tables

Sample Average Feret Diameter (nm) SD (nm)

NM112 30.5 13.8

NM113 165.2 90.4

NM110 75.4 58.4


29




NM_110_ANNEX_A15_ TEM data from NPL.doc / 1.39 MB (application/octet-stream):



Conclusions


75.4 ± 58.4 nm. Based on the measurements of 312 particles the mean particle size of the NM 111 was

determined to be 30.5 ± 13.8 nm. Based on the measurements of 87 particles the mean particle size of

NM 113 was determined to be 165.2 ± 90.4 nm.

Executive summary

NPL, 2010 investigated the particle size of the test items NM-110, NM-112 and NM-113 using TEM.

Therefore a nanomaterial sample dispersion of 50 mg/L was made in accordance to the OECD

recommended protocol for sample dispersion. The dispersion was allowed to settle out in order to remove

the larger particles naturally as it sediments by gravity. The purpose of this sedimentation was to remove

large micron size particles. On day 2, there was clear visible evidence of sedimentation events of the

larger micron size particles having taken place. Two microliter aliquots were extracted from the top layer

(an opaque but slightly transparent layer). The sample was placed on TEM grids and allowed to air dry

for 10 minutes; grids are formvar/carbon on 400 meshes copper. Grids were used as supplied with no

further modification. All images were recorded using a Hitachi 2300 A instrument operated at 200 kV.

An adequate magnification was chosen for image acquisition e.g. for the estimation of primary particle

mean diameter. TEM micrographs were analysed by manually tracing contours of primary particles on to

a transparency sheet. The transparency sheet was scanned for further image analysis using ImageJ

software, which automatically calculated particle diameter dimensions. Furthermore an extremely small

area of the sample could be analysed, which might not be representative enough for the whole sample.

The comparatively small share of evaluated particles results in limited statistical precision. The mean

primary particle size was quoted with the corresponding standard deviation which represents the

broadness of the size distribution.


75.4 ± 58.4 nm.


30.5 ± 13.8 nm.

Based on the measurements of 87 particles the mean particle size of NM 113 was determined to be 165.2

± 90.4 nm.

Endpoint study record: WoE_2010-06-17_NPL for PROSPECT_Particle size,

size distribution_DLS

Administrative Data




Rationale for

reliability incl.

deficiencies

The study was regarded as in general scientifically acceptable. It was not indicated in the

report if the experiment was conducted in accordance with available guidance documents. The

procedure of the sample preparation and subsampling was provided and regarded as

sufficient. In conclusion the study was regarded as reliable with restrictions and reliable for

the weight of evidence approach.


30

Data source

Reference

Referenc

e type

Author Year Title Bibliographi

c source

Testing

laboratory

Repor

t no.

Owner

compan

y

Compan

y study

no.

Repor

t date

study

report

R. Tantra,

D. Gohil,

S.

Kaliyappa

n and S.

Jing

2010 Nanoparticle

characterisation

for

ecotoxicologica

l studies using

Dynamic Light

Scattering,

Scanning

Electron

Microscopy and

Nanoparticle

Tracking

Analysis

Techniques.

National

Physical

Laboratory

Hampton

Road

Teddingto

n

Middlesex

TW11

0LW

2010-

06-17

Data access






other: DLS

Test guideline





Instruments, UK) equipped with a 633 nm laser. A reference standard (polystyrene, latex bead, nominal


involved filling of a disposable capillary cell (DTS1060, Malvern). The advantage of using this cell is that

zeta-potential measurements can be taken immediately after acquiring the DLS measurement. Prior to

their use, these cells were thoroughly cleaned with ethanol and de-ionised water, as recommended by the

instrument vendor. Individual cell was then filled with the appropriate sample and flushed before re-

filling; measurement was carried out on the second filling. Malvern Instrument’s Dispersion Technology

software (Version 4.0) was used for data analysis. For particle size it is the z average diameter (the mean

hydrodynamic diameter) that is reported.

GLP compliance

no

Test materials


yes


31


Identifier Identity


EC number 215-222-5

IUPAC name oxozinc


Test material form

nanomaterial



Code: NM110

Supplier: BASF SE





Code: NM111

Supplier: BASF SE





TM Code: NM112

Supplier: Micronisers Surface coating: none




Code: NM113






Sample preparation A stock dispersion of 500 mg/L was prepared and appropriate dilutions with DI water were made from

this stock. In order to obtain representative samples when sub-sampling, the dispersion was agitated

sufficiently (using a glass rod) prior to aliquoting the correct amount and diluting further with DI water.

The final concentrations were: 500, 200, 100, 50, 10, 1, 0.1, 0.01, 0.001 mg/L. To minimise any

variations in the nanoparticle dispersion, all measurements were performed simultaneously straight after

the sub-sampling step.

The protocol for sample preparation was described by Tantra, R., Jing,S. and Gohil, D.in Technical issues

surrounding the preparation, characterisation and testing of nanoparticles for ecotoxicological studies, in

Environmental Toxicology 3, V.B. Popov, C.A. , Editor. 2010, WIT Press p. 165-176.


32



Sample Mean particle size of three

distributions (nm)

NM112 253

NM110 275

NM113 508


Attached background material

Attached document Remarks

NM_110_ANNEX_A16_Cyprus2010v2.pdf / 318.37 KB (application/octet-stream):


NM_110_ANNEX_A17_FinalVersionSustainableDevPlanning2.pdf / 1.21 MB (application/octet-




NM_110_ANNEX_A17_FinalVersionSustainableDevPlanning2.pdf / 1.21 MB (application/octet-stream):



Conclusions

The mean particle size of three distributions was determined to be 275 nm for NM-110, 253 nm for NM-

112 and 508 nm for NM-113 respectively.

Executive summary

NPL, 2010 investigated the particle size of the test items NM-110, NM-112 and NM-113 using DLS.


Instruments,UK) equipped with a 633 nm laser. A reference standard (polystyrene, latex bead, nominal


involved filling of a disposable capillary cell (DTS1060, Malvern). Prior to their use, these cells were

thoroughly cleaned with ethanol and de-ionised water, as recommended by the instrument vendor.

Individual cell was then filled with the appropriate sample and flushed before re-filling; measurement was

carried out on the second filling. Malvern Instrument’s Dispersion Technology software (Version 4.0)

was used for data analysis. For particle size it was the z average diameter (the mean hydrodynamic

diameter) that was reported. The protocol for sample preparation was described by Tantra, R., Jing,S. and

Gohil, D.in Technical issues surrounding the preparation, characterisation and testing of nanoparticles for

ecotoxicological studies, in Environmental Toxicology 3, V.B. Popov, C.A. , Editor. 2010, WIT Press p.

165-176. The mean particle size of three distributions was determined to be 275 nm for NM-110, 253 nm



33

Endpoint study record: WoE_2011-10-03_NPL for PROSPECT_Particle

size,size distribution_SMPS

Administrative Data




Rationale for

reliability incl.

deficiencies

The study was regarded as in general scientifically acceptable. However, only limited

documentation of the data evaluation was provided in the full study reports. Furthermore the

procedure for subsampling and data analysis and representation was not explained in detail.

Thus, it was not possible to assess the validity of the provided data as the sample preparation

and the data analysis could have a significant influence to the result. Furthermore respective

ISO guidelines are available (ISO 15900:2009; ISO 10808:2010; ISO 28439:2011). It was not

indicated in the report if the experiment was conducted in accordance with these guidance

documents (in the absence of a validated OECD guideline). In the study report it was not

indicated how the aerosol was generated. Thus, the study was regarded as reliable with

restrictions.

Data source

Reference

Reference

type


source

Testing

laboratory

Report

no.

Owner

company

Company

study no.

Report

date

study

report

National

Physical

Laboratory

2011 Characterisation

of aerosolised

nanomaterial

powders

National

Physical

Laboratory

Hampton

Road

Teddington

Middlesex

TW11

0LW

2011-

10-03

Data access






other: SMPS


other: Scanning Mobility Particle Sizing (SMPS)(TSI 3080 SMPS), consisting of a DMA and CPC

system, was used to determine the particle size distribution. The Differential Mobility Analyser (DMA)

within the SMPS was calibrated using reference material polystyrene latex beads from NIST. The

Condensation Particle Counters (CPC) within the SMPS setup were calibrated according to NPL’s UKAS

accredited (ISO 17025) procedure, using an internally calibrated Faraday Cup Electrometer and soot

generator (model CAST 2). The SMPS was set to record at 4-minute intervals; at least 6 SMPS scans of

200 seconds each were used for analysis. The data was processed using TSI Aerosol Instrument

Management (AIM) software, in which the mean size distribution from the stable time segment was

estimated. The size distribution was also analysed using an in-house curvefitting program (as


34

implemented in a recent SMPS intercomparison at METAS).

GLP compliance

no

Test materials


yes


Identifier Identity


EC number 215-222-5

IUPAC name oxozinc


Test material form

nanomaterial



Code: NM110

Supplier: BASF SESurface coating: none




Code: NM111

Supplier: BASF SE





TM Code: NM112






Code: NM113






Instruments TSI Scanning Mobilty Particle Analyser (SMPS 3080 with a 3081 LDMA and a 3775 CPC)

Stand-alone TSI Condensation Particle Counter 3022a (used to correct for any source fluctuations)

Calibration Condensation Particle Counter internally calibrated to UKAS ISO 17025


35



Geometric Mean particle size of the different aerosolised PROSPEcT powders as measured by SMPS

Sample

Name Batch Number Geometric Mean Particle Size

Geometric Standard Deviation of log-

normal

NM112 NM-112

06022, 06023, 06024 269 nm 1.80

NM111 NM-111

06298 400 nm 1.57

NM113 NM-113

06032, 06033, 06034 300 nm 1.63

NM110 NM-110

06309 289 nm 1.68



NM_110_ANNEX_A19_NPL SMPS Data.doc / 119 KB (application/octet-stream):


NM_110_ANNEX_A02_FINALINTERIM REPORT2010corrected.pdf / 1.05 MB (application/octet-stream):



Conclusions

The range of the SMPS size distribution measurement was 16-660 nm. The Geometric Mean diameter

and Geometric Standard Deviation for each size distribution was estimated by fitting a log-normal curve

to the size distribution data plot. Each plot represents the mean values of at least 6 SMPS scans of 200

seconds each. The Geometric Mean Particle Size was determined to be 289 nm for NM-110, 400 nm for

NM-111, 269 nm for NM-112 and 300 nm for NM-113 respectively.

Executive summary

NPL, 2010 determined the particle size distribution of Aerosolised nanomaterials by SMPS consisting of

a Differential Mobility Analyser (DMA) and Condensation Particle Counters (CPC) system. The DMA

within the SMPS was calibrated using reference material polystyrene latex beads from NIST. TSI

Fluidised Bed Aerosol Generator (FBAG) was used to produce an aerosol from the dry powder sample.

After introduction of the nanomaterial into the FBAG, the aerosol generated was allowed to stabilise for a

day prior to sending the aerosol to an SMPS. The CPC within the SMPS setup were calibrated according

to NPL’s UKAS accredited (ISO 17025) procedure, using an internally calibrated Faraday Cup

Electrometer and soot generator (model CAST 2). The SMPS was set to record at 4 minute intervals; at

least 6 SMPS scans of 200 seconds each were used for analysis. The data was processed using TSI

Aerosol Instrument Management (AIM) software, in which the mean size distribution from the stable

time segment was estimated. The size distribution was also analysed using an in-house curvefitting

program (as implemented in a recent SMPS intercomparison at METAS). The range of the SMPS size

distribution measurement was 16-660 nm. The Geometric Mean diameter and Geometric Standard

Deviation for each size distribution was estimated by fitting a log-normal curve to the size distribution

data plot. Each plot represents the mean values of at least 6 SMPS scans of 200 seconds each. The

Geometric Mean Particle Size was determined to be 289 nm for NM-110, 400 nm for NM-111, 269 nm



36

Endpoint study record: WoE_2010-12-06 NPL for PROSPECT Particle sizeSEM

NM110-0305

Administrative Data




Rationale for

reliability incl.

deficiencies

The study was regarded as in general scientifically acceptable. However, only limited documentation of

the data evaluation was provided in the full study reports. Furthermore the procedure for subsampling and

data analysis and representation was not explained in detail. Thus, it was not possible to assess the validity

of the provided data as the sample preparation and the data analysis could have a significant influence to

the result. In general it should be possible to calculate the Particle size distribution from SEM data.

Furthermore respective ISO guidelines are available (ISO/TR 27628:2007. ISO/13322-1:2004 and

ISO/13322-2:2006). It was not indicated in the report if the experiment was conducted in accordance with

these guidance (in the absence of a validated OECD guideline) and no reason was given why the Particle

size distribution (D0, D50, D90) was not calculated. Thus, the study was regarded as reliable with

restrictions.

Data source

Reference

Reference

type


source

Testing

laboratory

Report

no.

Owner

company

Company

study no.

Report

date

study

report

National

Physical

Laboratory

2010 HOMOGENEITY

TEST: SEM National

Physical

Laboratory Hampton Road

Teddington

Middlesex TW11 0LW

2010-

12-06

Data access






other: SEM


SEM images were obtained using a Supra 40 field emission scanning electron microscope from Carl Zeiss

(Welwyn Garden City, Hertfordshire, UK), in which the optimal spatial resolution of the microscope was

a few nanometres. In-lens detector images were acquired at an accelerating voltage of 15 kV, a working

distance of ≈ 3 mm, and a tilt angle 0°. SEM instrument was calibrated using a SIRA grid calibration set

(SIRA, Chislehurst, Kent, UK). These are metal replicas of cross ruled gratings of area of 60 mm2 with

19.7 lines/mm for low magnification and 2160 lines/mm for high magnification calibrations, accurate to

0.2 %. For analysis of the “as received” nanoparticle powder, a sample of the powder was sprinkled over

a SEM carbon adhesive disc; one side of the carbon disc was placed securely on a metal stub, whilst the

other side was exposed to the nanoparticle powder. Excess powder was removed by gently tapping the

stub on its side until a light coating of powder on the surface became apparent. For analysis of

nanoparticles dispersed in liquid media, sample preparation requires to “fix” the nanoparticles on to a

substrate surface. This involved the deposition of an appropriate liquid sample (1 ml) on to a poly-l-lysine

coated microscope glass slide (purchased from Fisher Scientific, UK) and allowing it to incubate for a

period of 5 min at room temperature (≈ 20 C) before dipping in a beaker of water in order to remove

unbound nanoparticles. Slides were then allowed to dry under ambient conditions for ≈ 2 h before they


37

were thinly sputtered with gold using an Edwards S150B sputter coater unit (BOC Edwards, UK).

Sputtering was conducted under vacuum (≈ 7 mbar or 0.7 mPa), while passing pure, dry argon into the

coating chamber. Typical plate voltage and current were 1200 V and 15 mA, respectively. The sputtering

time was approximately 10 s, which resulted in an estimated gold thickness of not more than 2

nanometres being deposited on top of the substrate. An adequate magnification was chosen for image

acquisition e.g. for the estimation of primary particle mean diameter. The shape and limits of the primary

particles should become apparent. SEM micrographs were analysed manually; this was done by manually

tracing contours of primary particles on to a transparency sheet. The transparency sheet was scanned for

further image analysis using ImageJ software, which automatically calculated particle diameter

dimensions.

Test materials


yes


Identifier Identity

other: OECD Sponsorship programme NM110, NM111, NM112, NM113

Test material form

nanomaterial



Table summarises the primary particle size (as defined by their corresponding Feret’s diameter) of the

JRC sub-sampled powders for BASF Z-COTE and Z-COTE HP 1 samples. Results suggest that there is

no significant difference in the mean primary particle size (and corresponding SD) between the two types

of NMs and no real differentiation in the mean primary particle size between the vials in one type of

sample

Z-COTE

NM110-0305 NM110-4899 NM110-3975 NM110-1866 NM110-0286 NM110-2617

Rep mean s.d. mean s.d. mean s.d. mean s.d. mean s.d. mean s.d.

1 120.9 55.7 109.2 56.

4

111.2 72.

1

102.1 66.8 117.4 61.1 115.9 60.1

2 120.9 57.6 113.7 42.

2

115.4 60.

2

112.6 59.7 111.3 49.2 116.7 87.0

3 117.6 58.0 106.5 62.

4

114.7 48.

6

115.2 58.7 113.1 53.3 121.8 58.4

4 117.5 58.1 106.6 52.

3

123.2 60.

2

109.0 49.5 120.7 116.

9

119.7 58.7

5 111.7 64.1 105.1 47.

6

122.0 61.

0

117.6 72.2 120.8 71.5 121.6 126.1

6 104.8 57.1 105.9 46.

6

116.2 42.2 121.6 130.

9

121.8 89.4

Mean 120 60 110 50 120 60 110 60 120 90 120 80

Total weighted mean for uncoated ZnO = 120

Total pooled SD for uncoated ZnO = 60

Z-COTE HP1

NM111 -2419 NM111-1869 NM111-0486 NM111 -1017 NM111-3396 NM111 -4479

Rep mean s.d. mean s.d. mean s.d. mean s.d. mean s.d. mean s.d.

1 113.0 66.5 107.7 52.

6 114.7

79.

6 120.0 57.2 125.9

72.

3 127.7

189.

9

2 122.2 107. 114.1 48. 124.1 76. 119.0 118. 118.1 84. 120.1 53.8


38

8 0 4 6 2

3 127.7 134.

1 102.6

62.

6 112.6

57.

6 125.1 68.1 126.0

84.

2 118.5 41.8

4 123.1 42.7 102.2 54.

2 118.3

73.

0 119.8

118.

3 120.1

82.

6 127.9 65.4

5 113.2 47.0 110.1 48.

7 113.4

74.

3 124.8 79.0 127.5

81.

7 127.9 68.8

6 119.6 67.3 104.1 47.

6 123.0

89.

7 110.7 57.0 117.3

63.

6

Mean

Total weighted mean for coated ZnO = 120 Total pooled SD for coated ZnO = 80




NM_110_ANNEX_A20_ZnO HOMOGENEITY SEM.doc / 45 KB (application/octet-stream):



Conclusions

The total weighted mean for NM-110 was determined to be 120 nm and the total weighted mean for NM-

111 was determined to be 120 nm respectively. Results suggest that there is no significant difference in

the mean primary particle size (and corresponding SD) between the two types of NMs and no real

differentiation in the mean primary particle size between the vials in one type of sample.

Executive summary

NPL, 2010 determined the particle size distribution by SEM. SEM images were obtained using a Supra

40 field emission scanning electron microscope from Carl Zeiss (Welwyn Garden City, Hertfordshire,

UK), in which the optimal spatial resolution of the microscope was a few nanometres. For analysis of the

“as received” nanoparticle powder, a sample of the powder was sprinkled over a SEM carbon adhesive

disc; one side of the carbon disc was placed securely on a metal stub, whilst the other side was exposed to

the nanoparticle powder. Excess powder was removed by gently tapping the stub on its side until a light

coating of powder on the surface became apparent. An adequate magnification was chosen for image


particles should become apparent. SEM micrographs were analysed manually; this was done by manually

tracing contours of primary particles on to a transparency sheet. The transparency sheet was scanned for

further image analysis using ImageJ software, which automatically calculated particle. The total weighted

mean for NM-110 was determined to be 120 nm and the total weighted mean for NM-111 was determined

to be 120 nm respectively. Results suggest that there is no significant difference in the mean primary

particle size (and corresponding SD) between the two types of NMs and no real differentiation in the

mean primary particle size between the vials in one type of sample.


39

Endpoint study record: WoE_2012-03-

05_Australia_CSIRO_Particle_Size_TEM_XRD_BET

Administrative Data




Rationale for

reliability incl.

deficiencies

The study was regarded as in general scientifically acceptable. Particle size measurement

from TEM images were performed on 100 particles. However, only limited documentation of

the data evaluation was provided in the full study reports. Furthermore the procedure for

subsampling and data analysis and representation was not explained in detail. Thus, it was not

possible to assess the validity of the provided data as the sample preparation and the data

analysis could have a significant influence to the result. The respective ISO guidelines are

available but it was not indicated in the report if the experiment was conducted in accordance

with these guidance documents (in the absence of a validated OECD guideline). Nevertheless

the methodical comparison adds high reliability. Therefore the study was regarded as reliable

with restrictions.

Data source

Reference

Reference

type


source

Testing

laboratory

Report

no.

Owner

company

Company

study no.

Report

date

study

report

Casey,

Yin,

McCall

2012 Summary of

physicochemical

properties of

OECD Zinc

Oxide (ZnO)

samples

Commonwealth

Scientific and

Industrial

Research

Organisation

2012-

03-05

Data access






other: TEM

Test guideline




Two methods were used to estimate average particle size based on TEM results. Both Feret’s diameter

and the equivalent circular diameter were determined by measuring 100 individual particles in TEM

images. (i) Feret’s diameter (D Feret TEM) is a parameter that is widely used in imaging of irregularly-

shaped particles. Feret’s diameter can be defined as the “maximum calliper length” i.e. the longest

distance between any two points along the selection boundary.(ii) Use of the equivalent circular diameter

technique requires that a spherical shape be assigned to the particles. The equivalent circular diameter (D

circ TEM) is an estimate of the diameter of a circle that has the same area as that of the projected particle.


40

XRDTwo methods were used to estimate average crystallite size based on XRD results.(i) D S-XRD was

calculated using Scherrer’s formulaIt should be noted that a variety of factors can contribute to the width

of a diffraction peak; besides crystallite size, the most important of these usually are inhomogeneous

strain and instrumental effects. When Scherrer’s formula is applied, it is assumed that all of these other

contributions to the peak width are zero. (ii) D R-XRD was calculated using DIFFRACplus TOPAS 4.2

through Rietveld refinement of the diffraction data. Rietveld refinement is a technique that gives the Full

Width at Half Maximum (FWHM) of each diffraction peak while taking into account potential

broadening contributed from both strain and instrumental effects. BETAssuming that all particles in the

sample have a spherical shape, a smooth surface and are the same size, the average equivalent particle

size (DSA BET) was calculated from measurements of BET surface area by D =6000/(ρ•Sw) where D is

the average diameter of a spherical particle; Sw represents the measured surface area of the powder in

m2/g; and ρ is the particle density.

GLP compliance

no

Test materials


yes


Identifier Identity


EC number 215-222-5

IUPAC name oxozinc


Test material form

nanomaterial



Code: NM110

Supplier: BASF SE





Code: NM111

Supplier: BASF SE





TM Code: NM112






Code: NM113


41






Instruments:

Tecnai 12 TEM (FEI, Eindhoven, Netherlands) operating at 120 kV. Micrographs were recorded using an

Olympus Megaview III CCD camera (Tokyo, Japan) running AnalySiS imaging software (Olympus).



Comparison of crystallite/particle sizes in samples NM110, NM111, NM112 and NM113, as determined

from XRD analyses, TEM images and BET surface area measurements.

DR-XRD

(nm)

DS-XRD

(nm)

DFeret

TEM

(nm)

Dcirc TEM

(nm)

DSA BET

(nm)

Average of all

methods (nm)

NM110 113 41 70 ± 46 52 ± 33 74 77.5 ± 18

NM111 83 34 82 ± 45 60 ± 33 76 75.2 ± 7.6

NM112 43 25 30 ± 9 25 ± 7 37 33.75 ± 6.2

NM113 200 42 143 ± 47 113 ± 45 143 149.7 ± 25.0

All methods indicate that the particle sizes are in the same order as sizes provided by the suppliers. Based

on the average values from all methods, NM-112 is the smallest (~34 nm), NM-110 (78 nm) and NM-111

are larger (75 nm), and NM-113 is the largest (150 nm). The difference between the measured and

supplier-supplied particle size is largest for NM-113 (measured 150 nm, compared with a supplier-

supplied value of 95% < 3.5µm and 50% <1 µm) and smallest for NM-112 (measured 34nm compared

with a supplier-supplied value of 30-50 nm).Of the four samples analysed, NM-112 has the smallest size

distribution, with particles likely to be single crystals.

The sizes of particles determined by different methods were generally in the order of

D(R-XRD) > D(SA BET) ~ D (FeretTEM) > D(Circ TEM) > D(S-XRD )

D(S-XRD) was smallest because Scherrer’s formula ignores the contributions of width broadening due to

strain and instrument effects. If these contributions are non-zero, the crystallite sizes could be larger than

those predicted by the Scherrer formula, as revealed by D(R-XRD). The real states of particles can be

described as singlets, agglomerates or aggregates. Agglomerates are primary particles held by weak Van

der Waals force that can be overcome if sufficient energy is provided. Aggregates are primary particles

held together by strong chemical bonds that cannot be separated by conventional methods. D(Feret Tem)

provides the most reliable assessment of aggregate size, because aggregates are not broken up by

ultrasonication during sample preparation whereas agglomerates can be. D(Circ Tem) provides a smaller

value than D(Feret Tem) since this method treats all particles as spheres; it is not the most suitable

method for analysing samples containing particles with a broad or multimodal size distribution. D(SA

BET) was calculated from measurements of specific surface area and assumed that all particles are non-

porous and have the same spherical shape and size.


42




NM_110_ANNEX_A21_OECD ZnO Crystallite Size CSIRO Australia.docx / 29.24 KB (application/octet-stream):



Conclusions

All methods indicate that the particle sizes are in the same size range. Based on the average values from

all methods, the particle size was determined to be ~34 nm for NM 112, 78 nm for NM 110, 75 nm for

NM 111 and 150 nm for NM 113.

Executive summary

CSIRO, 2012 investigated the particle size distribution of the 4 nanomaterials. The particle size was

determined using XRD, TEM and Brunauer–Emmett–Teller (BET) surface area measurements. Two

XRD methods were used to estimate average crystallite size based on XRD results. The D S-XRDwas

calculated using Scherrer’s formula and D R-XRD was calculated using DIFFRACplus TOPAS 4.2

through Rietveld refinement of the diffraction data. For TEM measurements two methods were used to

estimate average particle size based on TEM results. Both Feret’s diameter and the equivalent circular

diameter were determined by measuring 100 individual particles in TEM images. The BET measurements

were performed under the assumption that that all particles in the sample have a spherical shape, a smooth

surface and are the same size. The average equivalent particle size (DSA BET) was calculated from

measurements of BET surface area. All methods indicate that the particle sizes are in the same size range.

Based on the average values from all methods, the particle size was determined to be ~34 nm for

NM-112, 78 nm for NM-110, 75 nm for NM-111 and 150 nm for NM-113. The particles size determined

by different methods were generally in the order of

D R-XRD > D SA BET ~ D Feret TEM > D Circ TEM > D S-XRD.D S-XRD

was smallest because Scherrer’s formula ignores the contributions of width broadening due to strain and

instrument effects. If these contributions are non-zero, the crystallite sizes could be larger than those

predicted by the Scherrer’s formula, as revealed by D R-XRD. The real states of particles can be



held together by strong chemical bonds that cannot be separated by conventional methods. D Feret-TEM

provides the most reliable assessment of aggregate size, because aggregates are not broken up by

ultrasonication during sample preparation whereas agglomerates can be. D Circ-TEM provides a smaller

value than D Feret-TEM since this method treats all particles as spheres; it is not the most suitable

method for analysing samples containing particles with a broad or multimodal size distribution.

D SA-BET was calculated from measurements of specific surface area and assumed that all particles are

non-porous and have the same spherical shape and size.


43


size,size distribution_CPS

Administrative Data




Rationale for

reliability incl.

deficiencies


documentation of the data evaluation was provided in the full study reports. The author of the

study erroneously described the D10 distribution that 10 mass % of the particles will have

particle larger diameter as the given value. Thus, the D10 and D90 values were reversed.

Furthermore the procedure for subsampling and data analysis and representation was not

explained in detail. Thus, it was not possible to assess the validity of the provided data as the

sample preparation and the data analysis could have a significant influence to the result.

Furthermore respective ISO guidelines (ISO 13318-1:2001; ISO 13318-2:2007; ISO 13318-

3:2004) are available (for details see comments above). It was not indicated in the report if the

experiment was conducted in accordance with these guidance documents (in the absence of a

validated OECD guideline). Therefore the study was regarded as reliable with restrictions.

Data source

Reference

Reference

type


source

Testing

laboratory

Report

no.

Owner

company

Company

study no.

Report

date

study

report

National

Physical

Laboratory

2010 Particle size

measurements

by cps disc

centrifuge

National

Physical

Laboratory,

Teddington,

Middlesex,

United

Kingdom

2010-

12-03

Data access






other: CPS

Test guideline




Particle size distribution by centrifugal sedimentation was acquired using CPS Disc Centrifuge Model DC

20000 instrument (Analytik Ltd, UK). At the start of the method, the centrifuge was brought up to speed

by partially filling the disc with a sucrose gradient fluid and dodecane cap fluid. The purpose of the

gradient fluid was to stabilise the sedimentation; the purpose of the cap fluid was to maintain the gradient

inside the disc. The disc centrifuge was then allowed to equilibrate at 6000 rpm for 1 hour; this gradient

will be stable and used within the next 6 hours. 0.2 ml of the nanoparticle sample (50 mg/L) was injected

into the disc; a calibration standard was injected after every three samples. Analysis was run against a


44

calibration standard, NIST traceable standard, PVC 0.377 micron. The Disc Centrifuge Control System

software (CPS Instruments Inc.) was used to acquire and process the data.

GLP compliance

no

Test materials


yes


Identifier Identity


EC number 215-222-5

IUPAC name oxozinc

other: OECD sponsorship programme NM110, NM111, NM112, NM113

Test material form

nanomaterial








Code: NM111

Supplier: BASF SE





TM Code: NM112






Code: NM113






Instruments: CPS Disc Centrifuge Model DC 20000 instrument (Analytik Ltd, UK

Calibration: Analysis was run against a calibration standard, NIST traceable standard, PVC 0.377 micron.


45



Sample

Name Supplier Batch number

DI water

(nm)

Fish medium

(nm)

Seawater

(nm) Daphnia (nm)

NM112 Micronisers, ZA250#30#ICP 277±7 390±70 510±40 500±200

NM113 Sigma, ZrA250#33#ICP 590±30 620±20 660±20 631±5

NM110 BASF, ZC250#37#ICP 193±3 290±20 309±10 296±16

Sample


DI water

(nm)

Fish medium

(nm)

Seawater

(nm) Daphnia (nm)

NM112 Micronis

ers, ZA250#30#ICP

D10720±30

D5040.1±0.7

D9064.6±0.6

D101000±200

D50190±17

D9093 ±4

D101180±20

D50330±70

D90130±50

D10100±200

D50400±200

D90100±50

NM113 Sigma, ZrA250#33#ICP

D10870±60

D50572±19

D90306±7

D10890±40

D50606±12

D90336±8

D10930±50

D50639±15

D90399±14

D10930±20

D50612±3

D90332±6

NM110 BASF, ZC250#37#ICP

D10286±2

D5082.8±1.9

D90107.3±1.7

D10400±30

D50270±20

D90130±30

D10417±12

D50301±8

D90193±7

D10410±20

D50285±16

D90140±30 The above tables show the CPS disc centrifugal sedimentation results, with Table a) showing the

equivalent spherical mean particle diameter and Table b) the corresponding D10, D50, D90values (oversize

percentiles). D10, D50, D90values are often used to describe the particle size distribution of the sample. If

D10= 1225 nm, then this means that that 10 mass % of the particles will have particle diameter of 1225 nm

or larger. The results show that the largest mean particle size exists when the NMs are dispersed in

seawater; this is reflected on the particle mean size as well as the corresponding D90values. Results also

show that the smallest particle size exists when the NMs are dispersed in DI water. This suggests that

larger agglomerates exist in the ecotox media, with seawater being the worst i.e. largest agglomerates

found in seawater.




NM_110_ANNEX_A22_PARTICLE SIZE MEASUREMENTS BY CPS DISC CENTRIFUGE.doc / 61.5 KB

(application/octet-stream): ENV/JM/MONO(2015)15/ANN1




Conclusions

The equivalent spherical mean particle diameter was determined to be for NM-110 193 ± 3 nm in DI

water, 290 ± 20 nm in fish medium 309 ± 10 nm in seawater and 296 ± 16 nm in daphnia media

respectively. The equivalent spherical mean particle diameter was determined to be for NM-112 277 ± 7

nm in DI water, 390 ± 70 nm in fish medium 510 ± 40 nm in seawater and 500 ±200 nm in daphnia media

respectively. The equivalent spherical mean particle diameter was determined to be for NM-131 590 ±30

nm in DI water, 620 ± 20 nm in fish medium 660 ± 20 nm in seawater and 631 ± 5 nm in daphnia media

respectively. The results show that the largest mean particle size exists when the NMs are dispersed in

seawater; this is reflected on the particle mean size as well as the corresponding D90 values. Results also

show that the smallest particle size exists when the NMs are dispersed in deionized water. This suggests


46

that larger agglomerates exist in the ecotox media, i.e. largest agglomerates found in seawater.

Executive summary

NPL, 2010 investigated the particle size distribution by using CPS Disc Centrifuge. Analysis was run

against a calibration standard, NIST traceable standard, PVC 0.377 micron. The equivalent spherical

mean particle diameter was determined to be for NM-110 193 ± 3 nm in DI water, 290 ± 20 nm in fish

medium 309 ± 10 nm in seawater and 296 ± 16 nm in daphnia media respectively. The equivalent

spherical mean particle diameter was determined to be for NM-112 277 ± 7 nm in DI water, 390 ± 70 nm

in fish medium 510 ± 40 nm in seawater and 500 ±200 nm in daphnia media respectively. The equivalent

spherical mean particle diameter was determined to be for NM-131 590 ±30 nm in DI water, 620 ± 20 nm

in fish medium 660 ± 20 nm in seawater and 631 ± 5 nm in daphnia media respectively.


size,size distribution_SEM

Administrative Data



Reliability 4 (not assignable)

Rationale for

reliability incl.

deficiencies




The evaluation of minimum of 50 particles measured in the SEM images was considered as

statistical not sufficient for data evaluation. In addition it was not specified which Feret

Diameter was reported. Therefore, the reliability could not be assigned.

Data source

Reference

Reference

type


source

Testing

laboratory

Report

no.

Owner

company

Company

study no.

Report

date

study

report

National

Physical

Laboratory

2010 THE

PROSPEcT

Interim

REPORT

and

information

related to the

SEM particle

sizing

National

Physical

Laboratory

Hampton

Road

Teddington

Middlesex

TW11 0LW

2010-

06-17

Data access






other: SEM


47

Test guideline





(Welwyn Garden City, Hertfordshire, UK), in which the optimal spatial resolution of the microscope was

a few nanometres. Inlens detector images were acquired at an accelerating voltage of 15 kV, a working

distance of ≈ 3 mm, and a tilt angle 0°. SEM instrument was calibrated using a SIRA grid calibration set

(SIRA, Chislehurst, Kent, UK). These are metal replicas of cross ruled gratings of area of 60 mm2 with

19.7 lines/mm for low magnification and 2160 lines/mm for high magnification calibrations, accurate to

0.2 %. For analysis of the “as received” nanoparticle powder, a sample of the powder was sprinkled over

a SEM carbon adhesive disc; one side of the carbon disc was placed securely on a metal stub, whilst the

other side was exposed to the nanoparticle powder. Excess powder was removed by gently tapping the

stub on its side until a light coating of powder on the surface became apparent. For analysis of

nanoparticles dispersed in liquid media, sample preparation requires to “fix” the nanoparticles on to a

substrate surface. This involved the deposition of an appropriate liquid sample (1 ml) on to a poly-l-lysine

coated microscope glass slide (purchased from Fisher Scientific, UK) and allowing it to incubate for a

period of 5 min at room temperature (≈ 20 °C) before dipping in a beaker of water in order to remove

unbound nanoparticles. Slides were then allowed to dry under ambient conditions for ≈ 2 h before they

were thinly sputtered with gold using an Edwards S150B sputter coater unit (BOC Edwards, UK).

Sputtering was conducted under vacuum (≈ 7 mbar or 0.7 mPa), while passing pure, dry argon into the

coating chamber. Typical plate voltage and current were 1200 V and 15 mA, respectively. The sputtering

time was approximately 10 s, which resulted in an estimated gold thickness of not more than 2

nanometres being deposited on top of the substrate. An adequate magnification was chosen for image


particles should become apparent.

GLP compliance

no

Test materials


yes


Identifier Identity


EC number 215-222-5

IUPAC name oxozinc


Test material form

nanomaterial



Code: NM110

Supplier: BASF SE


48





Code: NM111

Supplier: BASF SE





TM Code: NM112






Code: NM113






Instruments Carl Zeiss Supra 40 field emission scanning electron microscope

Calibration SIRA grid calibration set (SIRA, Chislehurst, Kent, UK)

State of test material: dry bulk

Any other information on materials and methods incl. tables: SEM micrographs were analysed manually;

this was done by manually tracing contours of primary particles on to a transparency sheet. The

transparency sheet was scanned for further image analysis using ImageJ software, which automatically

calculated particle diameter dimensions.



Size of primary particles, as defined by their corresponding Feret’s diameter. Mean diameter (±1 SD) of a

minimum of 50 particles measured in the SEM images; the SD here represents the broadness of the size

distribution (not error).

Sample Name Supplier Batch number Mean Feret diameter

(± 1 SD) from SEM images

NM110 BASF, Germany ZC250#37#05 151 ± 55.6

NM111 BASF, Germany ZB250#64#05 140.8 ± 65.8

NM112 Micronisers, Australia ZA250#30#05 42.5 ± 3.6

NM113 Sigma Aldrich, UK ZrA250#45#05 891.8 ± 800.0



Size of primary particles, as defined by their corresponding Feret’s diameter. Mean diameter (± 1 SD) of

a minimum of 50 particles measured in the SEM images; the SD here represents the broadness of the size

distribution (not error). Pages 5-6 of the FINALINERIMREPORT2010corrected.pdf should be amended

to SEM Feret diameter from NPL.doc


49





NM_110_ANNEX_A13_SEM Preliminary Data From NPL NM110.doc / 1.68 MB (application/octet-stream):


NM_110_ANNEX_A23_SEM Feret diameter from NPL.doc / 140 KB (application/octet-stream):



Conclusions

The mean Feret diameter was determined to be 151 nm for NM-110, 140.8 nm for NM-111, 42.5 nm for

NM-112 and 891.8 nm for NM-113 respectively.

Executive summary

NPL, 2010 determined particle size distribution by SEM. SEM images were obtained using a Supra 40

field emission scanning electron microscope from Carl Zeiss (Welwyn Garden City, Hertfordshire, UK),

in which the optimal spatial resolution of the microscope was a few nanometres. For analysis of the “as

received” nanoparticle powder, a sample of the powder was sprinkled over a SEM carbon adhesive disc;

one side of the carbon disc was placed securely on a metal stub, whilst the other side was exposed to the

nanoparticle powder. Excess powder was removed by gently tapping the stub on its side until a light

coating of powder on the surface became apparent. An adequate magnification was chosen for image


particles should become apparent. SEM micrographs were analysed manually. This was done by

manually tracing contours of primary particles on to a transparency sheet. The transparency sheet was

scanned for further image analysis using ImageJ software, which automatically calculated particle. The

mean Feret diameter was determined to be 151 nm for NM-110, 140.8 nm for NM-111, 42.5 nm for NM-

112 and 891.8 nm for NM-113 respectively.

Endpoint study record: WoE_2010_12_06_NPL for

PROSPECT_Homogeneity_NM110-0286_SEM

Administrative Data




Rationale for

reliability incl.

deficiencies




Therefore, the study was regarded as reliable with restrictions.


50

Data source

Reference

Referen

ce type


source

Testing

laboratory

Repo

rt no.

Owner

compa

ny

Compa

ny

study

no.

Repo

rt

date

study

report

National

Physical

Laborat

ory

Homo

geneit

y_NM

110

National

Physical

Laboratory,

Teddington,

Middlesex,

United

Kingdom

2010-

12-06

Data access





Test guideline





in which the optimal spatial resolution of the microscope was a few nanometres. In-lens detector images

were acquired at an accelerating voltage of 15 kV, a working distance of ≈ 3 mm, and a tilt angle 0°. SEM

instrument was calibrated using a SIRA grid calibration set (SIRA, Chislehurst, Kent, UK). These are

metal replicas of cross ruled gratings of area of 60 mm2 with 19.7 lines/mm for low magnification and

2160 lines/mm for high magnification calibrations, accurate to 0.2 %. For analysis of the “as received”

nanoparticle powder, a sample of the powder was sprinkled over a SEM carbon adhesive disc; one side of

the carbon disc was placed securely on a metal stub, whilst the other side was exposed to the nanoparticle

powder. Excess powder was removed by gently tapping the stub on its side until a light coating of powder

on the surface became apparent. An adequate magnification was chosen for image acquisition e.g. for the

estimation of primary particle mean diameter. The shape and limits of the primary particles should

become apparent.The SEM images were opened in ImageJ (a free image analysis program produced and

distributed by the National Institute of Health, US) installed on a Tablet PC (DELL XT1), and for each

image at least 20 distinct particles identified, to measure at least 100 particles per replicate. The particles

chosen were an array of sizes to accurately represent the variety in the sample. Using the Pencil Tool

(pencil width 2 pixels, colour black) the outlines of these particles were drawn using the supplied digital

‘pen’. The image was calibrated with reference to the scale marking on the image. Then the threshold

level adjusted so that only the particle outlines were highlighted. The particles were then measured using

the Analyze Particles tool, which produced a results table of various measurements of the particles. This

process was repeated for each image of the particular replicate (seven in total – giving a minimum of 140

particles identified. Feret’s diameter – also called the “maximum calliper length”, i.e. the longest distance

between any two points along the selection boundary - was recorded and the scientific graphing and

analysis software, SciDAVis, where a histogram (bin size 2) was plotted, showing the particle size

distribution of the replicate. The entire process was repeated for each replicate (r1, r2, r3) for all six sub-

samples of both the ZnO and ZnO-HP1 samples.


51

GLP compliance

no

Test materials


yes

Test material identity Identifier Identity


EC number 215-222-5

IUPAC name oxozinc


Test material form

nanomaterial

Confidential details on test material


Code: NM110

Supplier: BASF SE





Code: NM111

Supplier: BASF SE





TM Code: NM112






Code: NM113






Instruments: Supra 40 field emission scanning electron microscope from Carl Zeiss (Welwyn Garden

City, Hertfordshire, UK) Calibration: SEM instrument was calibrated using a SIRA grid calibration set

(SIRA, Chislehurst, Kent, UK).



The results are summarised in Table below with the histograms and example images given in the

appendix. The findings are: 1.From the SEM image a broad range of particle sizes were seen from small (20-50 nm) circular or


52

spherical particles, to much larger (>100 nm) rod shaped particles. 2.Overall, the particle size distributions between replicates per sub-sample are very similar, only differing

by an average range of 14.6nm for the NM-110 sample and 8.4 nm for the ZnO sample. An exception

however, is the difference in the replicates of sub-sample 0830 for ZnO-HP1. 3.The mean particle size for ZnO-HP1 is in the range 70-80nm, and for the ZnO sample the range 100-

111nm. Except for sub-sample 1455 which is 93.3nm, larger than that of the other sub-samples for ZnO-

HP1 and for ZnO, sub-sample 0599 at 90.3nm has a comparatively small mean particle size.

The size of particles for JRC sub-sampled powders ZnO and ZnO-HP1 given by the mean Feret’s

diameter, the standard deviation (represents broadness of size distribution, not error) for each sample,

sub-sample and replicate.

NM 111 ZnO-HP1 Mean

Feret’s

diameter

(nm)

S.D. NM 11O ZnO Mean

Feret’s

diameter

(nm)

S.D.

0830 r1 92.4 59.5 0599 r1 87.7 56.0

r2 82.0 52.6 r2 94.0 72.8

r3 66.8 48.8 r3 89.2 70.2

sub sample 79.4 54.2 sub sample 90.5 67.4

1455 r1 92.9 54.5 0945 r1 98.3 60.6

r2 86.3 66.6 r2 103.3 63.3

r3 99.5 72.3 r3 107.7 67.4


1951 r1 89.6 70.2 0979 r1 110.8 73.9

r2 73.6 60.1 r2 113.6 69.7

r3 70.0 57.1 r3 102.3 60.4


2017 r1 81.1 59.3 3911 r1 109.4 70.4

r2 72.7 53.0 r2 113.9 68.4

r3 76.5 57.2 r3 108.9 62.3


2100 r1 78.1 64.8 4410 r1 99.5 56.5

r2 73.7 56.4 r2 106.2 68.5

r3 88.7 63.6 r3 104.5 64.0


4414 r1 73.6 71.4 4779 r1 99.5 62.6

r2 68.2 53.7 r2 111.1 74.3

r3 73.7 51.1 r3 110.8 70.8



53



NM_110_ANNEX_A04_draft homogenity report SEM.doc / 19.43 MB (application/octet-stream):



Conclusions

From the SEM image a broad range of particle sizes were seen from small (20-50 nm) circular or

spherical particles, to much larger (>100 nm) rod shaped particles. Overall, the particle size distributions

between replicates per sub-sample are very similar, only differing by an average range of 14.6nm for the

ZnO-HP1 sample and 8.4nm for the ZnO sample. An exception however, is the difference in the

replicates of sub-sample 0830 for NM-100.The mean particle size for NM-110 is in the range 70-80nm,

and for the ZnO sample the range 100-111nm. Except for sub-sample 1455 which is 93.3nm, larger than

that of the other sub-samples for NM-110 and for ZnO, sub-sample 0599 at 90.3nm has a comparatively

small mean particle size.

Executive summary

NPL, 2010 investigated the particle size distribution of the NM-110 by SEM. SEM images were obtained

using a Supra 40 field emission scanning electron microscope from Carl Zeiss in which the optimal

spatial resolution of the microscope was a few nanometres. In-lens detector images were acquired at an

accelerating voltage of 15 kV, a working distance of ≈ 3 mm, and a tilt angle 0°. SEM instrument was

calibrated using a SIRA grid calibration set (SIRA, Chislehurst, Kent, UK). These are metal replicas of

cross ruled gratings of area of 60 mm2 with 19.7 lines/mm for low magnification and 2160 lines/mm for

high magnification calibrations, accurate to 0.2 %. For analysis of the “as received” nanoparticle powder,

a sample of the powder was sprinkled over a SEM carbon adhesive disc; one side of the carbon disc was

placed securely on a metal stub, whilst the other side was exposed to the nanoparticle powder. Excess

powder was removed by gently tapping the stub on its side until a light coating of powder on the surface

became apparent. An adequate magnification was chosen for image acquisition e.g. for the estimation of

primary particle mean diameter. The shape and limits of the primary particles should become apparent.

The SEM images were opened in ImageJ (a free image analysis program produced and distributed by the

National Institute of Health, US) installed on a Tablet PC (DELL XT1), and for each image at least 20

distinct particles identified, to measure at least 100 particles per replicate. The particles chosen were an

array of sizes to accurately represent the variety in the sample. Using the Pencil Tool (pencil width 2

pixels, colour black) the outlines of these particles were drawn using the supplied digital ‘pen’. The image

was calibrated with reference to the scale marking on the image. Then the threshold level adjusted so that

only the particle outlines were highlighted. The particles were then measured using the Analyze Particles

tool, which produced a results table of various measurements of the particles. This process was repeated

for each image of the particular replicate (seven in total – giving a minimum of 140 particles identified.

Feret’s diameter – also called the “maximum calliper length”, i.e. the longest distance between any two

points along the selection boundary - was recorded and the scientific graphing and analysis software,

SciDAVis, where a histogram (bin size 2) was plotted, showing the particle size distribution of the

replicate. The entire process was repeated for each replicate (r1, r2, r3) for all six sub-samples of both the

ZnO and ZnO-HP1 samples. From the SEM image a broad range of particle sizes were seen from small

(20-50 nm) circular or spherical particles, to much larger (>100 nm) rod shaped particles. Overall, the

particle size distributions between replicates per sub-sample are very similar, only differing by an average

range of 14.6nm for the ZnO-HP1 sample and 8.4nm for the ZnO sample. An exception however, is the

difference in the replicates of sub-sample 0830 for NM-100.The mean particle size for NM-110 is in the

range 70-80nm, and for the ZnO sample the range 100-111nm. Except for sub-sample 1455 which is

93.3nm, larger than that of the other sub-samples for NM-110 and for ZnO, sub-sample 0599 at 90.3nm


54

has a comparatively small mean particle size.

4.6 Vapour pressure

4.7 Partition coefficient

4.8 Water solubility

Endpoint summary: Water solubility

Administrative Data


It was shown that the dissolution rates were fastest when the NMs were dispersed in deionized water,

with NM 110 dissolving the fastest and NM 112 dissolving the slowest. Out of all ecotox media, fish

medium showed highest dissolution rate followed by daphnia and then seawater. Dispersing

nanomaterials in such ecotox media would mean less stable dispersion and this subsequently equates to

the reduced surface area concentrations and thus a lower dissolution rate. There was no significant

difference in dissolution rate between the uncoated ZnO nanomaterials and the bulk material.

Discussion

Four studies using different analytical techniques were available for dissolution of the investigated

nanomaterials and were assessed in a weight of evidence approach.

A dissolution study was conducted by CSIRO, 2012 in artificial soil solution of with ionic composition

typical of that found in Australian soil porewater. The solution pH was buffered to 4, 7 or 9 throughout

the dissolution study. The NM stock solution was added with 20 mL of artificial soil solution (final

concentration of ~ 300 µg Zn/L) into 50 mL centrifuge tubes. The solution pH was maintained at 4, 7, or

9 using perchloric acid (HClO4, 0.1 M) or sodium hydroxide (0.1 M) and buffered using 0.1 mM sodium

acetate (NaCH3COO), 0.1 mM 2-(N morpholino)ethanesulfonic acid sodium salt (MES), or 0.1 mM

sodium borate (Na2B4O7 x 10 H2O) solutions, respectively. The solutions were shaken on an end-over-

end shaker for 0, 24h and 7d. At individual time periods, 2 mL of solution was removed and added into 1

kDa ultra filtration devices (UF). The UF devices were centrifuged at 4000 g for 15 min and total Zn

concentrations in filtrates determined using inductively coupled plasma-mass spectrometry. The

dissolution of NM in each pH solution and time period was expressed as a percentage of the total Zn

added. The dissolution for all NM was highest at pH 4 and lowest at pH 9. As there is doubt on the

validity of the presented data, e.g the time series, the absolute values as well as the differences between

ZnO grades were regarded to be not plausible, no further results was taken into account.

A publication by Rogers et al., 2010, was available. The dissolution of NM-110 in synthetic softwater

medium without EDTA and buffered at pH 7.5±0.1 with 2 mM piperazine-N,N’-bis(ethanesulfonic acid)

was determined using the equilibrium dialysis technique described by Franklin et al., 2007. The

dissolution of NM-110 was determined to be 6.77±0.12 mg/L at pH = 7.5 in US EPA medium. Due to the

different test procedure, conditions and the lack of data from the other three nanomaterials, this result was

regarded as less relevant.

The dissolution of NM 110, NM 112, NM 113 in different media (deionized water, fish medium, daphnia

medium and seawater) were investigated by NPL, 2010 using inductively coupled plasma mass

spectrometry (ICP-MS). Dispersion was performed in accordance to the recommended PROSPEcT

protocol. A concentration of 50 mg/L was made up for each sample; a total volume of 1 L was made up


55

and stored in clean media (1 L) bottles at room temperature. After day 2, the bottles were stored in the

fridge. Several extractions from the 1 L sample were made over a period of 22 days. The extracted sample

(~ 50 mL) was then subjected to a three-step process in order to remove particles and to extract the

resultant supernatant. Dissolution rates were fastest when the NMs were dispersed in deionized water,

with NM 110 dissolving the fastest and NM 112 dissolving the slowest. Deionized water yielded the most

stable dispersions and this increase in stability will mean less aggregation/agglomeration (and subsequent

sedimentation) in the dispersion. Out of all the ecotox media, fish medium had the largest dissolution rate

followed by daphnia and then seawater. Dispersing NMs in such ecotox media would mean less stable

dispersion and this subsequently equates to the reduced surface area concentrations and thus a lower

dissolution rate.

In addition dispersion of NM-110, NM-112 and NM-113 in an ecotoxicology relevant media (fish

medium, daphnia medium and seawater) was investigated by the NPL, 2010. Deionised water was

employed as the corresponding media control and was used to prepare all aqueous solutions and

suspensions. Zinc ions were measured using Cole-Palmer® Colorimetric Test Kits (Cole Palmer, UK).

The colorimetric measurement was used to evaluate the nanomaterials when dispersed in different media

over time of 21 days; the extracted supernatant from the dispersions were obtained prior to performing the

colorimetric tests. The dispersions were stored in a refrigerator after day 2 in order to prevent degradation

of the sample e.g. minimising bacterial growth. It was shown that dissolution rates were fastest, when the

NMs were dispersed in deionized water, with NM-110 dissolving fastest and NM-112 dissolving slowest.

Deionized water yielded the most stable dispersions and this increase in stability will result in less

aggregation/agglomeration (and subsequent sedimentation) in the dispersion. The total surface area was

greater when the particles were dispersed in deionized when compared to corresponding ecotox media.

Thus, an increase in surface area means that the ion dissolution rate will also increase. An apparent

decrease in zinc concentrations from Day 6 to Day 9, for all zinc oxide NMs was observed. This effect

may be indicative of the dissolution-precipitation process occurring during this time. Out of all the ecotox

media, fish medium had highest dissolution rate followed by daphnia and then seawater. Dispersing


the reduced surface area concentrations and thus a lower dissolution rate.

Conclusion

It was shown that the dissolution rates were fastest when the NMs were dispersed in deionized water,

with NM 110 dissolving the fastest and NM 112 dissolving the slowest. Out of all ecotox media, fish

medium showed highest dissolution rate followed by daphnia and then seawater. Dispersing


the reduced surface area concentrations and thus a lower dissolution rate. There was no significant

difference in dissolution rate between the uncoated ZnO nanomaterials and the bulk material.

Endpoint study record: WoE_2012-03-02_Australia_CSIRO_Dissolution

Administrative Data




Rationale for

reliability incl.

deficiencies


documentation of the data evaluation was provided in the full study report. Furthermore the



and the data analysis could have a significant influence to the result. There is doubt on the

validity of the presented data. The time series, the absolute values as well as the differences

between ZnO grades were regarded to be not plausible. It was shown that the nanomaterials

highly agglomerate in water and that the media has a significant influence on the same. Thus,


56

due to the unknown composition of the used artificial soil solution (no reference was

provided), there is a unknown factor which have an influence on the result. Thus, the study

was regarded as reliable with restrictions.

Data source

Reference

Reference

type


source

Testing

laboratory

Report

no.

Owner

company

Company

study no.

Report

date

study

report

Cornelis

et al.

2012 Dissolution Commonwealth

Scientific and

Industrial

Research

Organisation

00001 2012-

03-02

Data access



yes


Type of method

flask method


Dissolution of NM-110 was examined in an artificial soil solution at pH 4, 7 and 9 (see attached

dissolution method).

GLP compliance

no

Test materials


yes


Identifier Identity


EC number 215-222-5

IUPAC name oxozinc


Test material form

nanomaterial







57



Code: NM111

Supplier: BASF SE





TM Code: NM112






Code: NM113






The dissolution experiments occurred in artificial soil solution (Table 1) of which the ionic composition is

typical of that found in Australian soil porewater(McLaughlin et al., 1997). The solution pH was buffered

to 4, 7 or 9 throughout the dissolution study (Figure 1). The NM-110 stock solution was added with 20 ml

of artificial soil solution (final concentration of ~ 300 µg Zn L-1) into 50 ml centrifuge tubes. The

solution pH was maintained at 4, 7, or 9 using perchloric acid (HClO4,0.1 M) or sodium hydroxide (0.1

M) and buffered using 0.1 mM sodium acetate (NaCH3COO), 0.1 mM 2-(Nmorpholino)ethanesulfonic

acid sodium salt (MES), or 0.1 mM sodium borate (Na2B4O7.10H2O) solutions, respectively. The

solutions were shaken on an end-over-end shaker for 0, 24h and 7d. The dissolution of bulk ZnO was

examined by weigh ~ 0.01 g of bulk ZnO (Sigma-Aldrich) powder into 50 ml centrifuge tubes and adding

20 ml of artificial soil solution at pH 4, 7 and 9. At individual time periods, 2 ml of solution was removed

and added into 1kDa ultra filtration devices (UF) (Pall-Gellman). The UF devices were centrifuged at

4000g for 15 min and total Zn concentrations in filtrates determined using inductively coupled plasma-

mass spectrometry (Agilent 7500ce). The dissolution of NM-110 in each pH solution and time period was

expressed as a percentage of the total Zn added.

Table 1. Composition of artificial soil solution used during dissolution experiments.

Component Concentration (mM)

Ca 6.22

Mg 6.00

K 9.55

Sulfate 6.00

Phosphate 0.02

Nitrate 22.00



Nanoparticle Time (day) pH Dissolution

(%)

NM-110 0 4 91 ± 12


58

1 4 96 ± 10

7 4 105 ± 11

0 7 53 ± 8

1 7 67 ± 4

7 7 58 ± 4

0 9 < 2

1 9 < 2

7 9 < 2

Bulk ZnO 0 4 1.4 ± 0.1

1 4 1.5 ± 0.02

7 4 7.4 ± 0.6

0 7 0.06 ± 0.001

1 7 0.11 ± 0.01

7 7 0.25 ± 0.08

0 9 0.009 ± 0.0045

1 9 0.007 ± 0.0001

7 9 0.007 ± 0.0001


(%)

NM-111 0 4 9.2 ± 1.4

1 4 14.4 ± 1.1

7 4 13.0 ± 2.3

0 7 1.4 ± 2.8

1 7 4.4 ± 3.8

7 7 6.3 ± 4.6

0 9 7.2 ± 0.7

1 9 11.8 ± 0.4

7 9 11.4 ± 0.4

Bulk ZnO 0 4 1.4 ± 0.1

1 4 1.5 ± 0.02

7 4 7.4 ± 0.6

0 7 0.06 ± 0.001

1 7 0.11 ± 0.01

7 7 0.25 ± 0.08

0 9 0.009 ± 0.0045

1 9 0.007 ± 0.0001

7 9 0.007 ± 0.0001


(%)

NM-112 0 4 89 ± 4


59

1 4 81 ± 7

7 4 82 ± 4

0 7 62 ± 3

1 7 60 ± 6

7 7 65 ± 3

0 9 2.7 ± 1.0

1 9 2.5 ± 0.5

7 9 2.8 ± 0.6

Bulk ZnO 0 4 1.4 ± 0.1

1 4 1.5 ± 0.02

7 4 7.4 ± 0.6

0 7 0.06 ± 0.001

1 7 0.11 ± 0.01

7 7 0.25 ± 0.08

0 9 0.009 ± 0.0045

1 9 0.007 ± 0.0001

7 9 0.007 ± 0.0001




NM_110_ANNEX_A24_NanoHub_NM-110_dissolution method.docx / 102.15 KB (application/octet-stream):


NM_110_ANNEX_A25_NanoHub_NM-110_dissolution_data.docx / 11.72 KB (application/octet-stream):







Conclusions

It was shown that the dissolution for all NM was highest at pH 4 and lowest at pH 9. As there is doubt on

the validity of the presented data, e.g. the time series, the absolute values as well as the differences

between ZnO grades were regarded to be not plausible, no further results was taken into account.

Executive summary

A dissolution study was conducted by CSIRO, 2012 in artificial soil solution of with ionic composition

typical of that found in Australian soil porewater. The solution pH was buffered to 4, 7 or 9 throughout

the dissolution study. The NM stock solution was added with 20 mL of artificial soil solution (final

concentration of ~ 300 µg Zn/L) into 50 mL centrifuge tubes. The solution pH was maintained at 4, 7, or

9 using perchloric acid (HClO4, 0.1 M) or sodium hydroxide (0.1 M) and buffered using 0.1 mM sodium

acetate (NaCH3COO), 0.1 mM 2-(N morpholino)ethanesulfonic acid sodium salt (MES), or 0.1 mM

sodium borate (Na2B4O7 x 10 H2O) solutions, respectively. The solutions were shaken on an end-over-

end shaker for 0, 24h and 7d. At individual time periods, 2 mL of solution was removed and added into 1

kDa ultra filtration devices (UF). The UF devices were centrifuged at 4000 g for 15 min and total Zn


60

concentrations in filtrates determined using inductively coupled plasma-mass spectrometry. The

dissolution of NM in each pH solution and time period was expressed as a percentage of the total Zn

added. The dissolution for all NM was highest at pH 4 and lowest at pH 9. As there is doubt on the

validity of the presented data, e.g the time series, the absolute values as well as the differences between

ZnO grades were regarded to be not plausible, no further results was taken into account.

Endpoint study record: WoE_2012-03-02_Australia_CSIRO_Solubility

Administrative Data




Rationale for

reliability incl.

deficiencies

Limited documentation of the data evaluation was provided in the publication but the

data was regarded as scenically sufficient. Thus, the study was regarded as reliable with

restrictions.

Data source

Reference

Reference

type

Author Year Title Bibliographic source Testing

laboratory

Report

no.

Owner

company

Company

study no.

Report

date

study

report

Angel

et al.

2012 Solubility of NM-

110

Sydney NSW,

Commonwealth

Scientific and

Industrial

Research

Organisation

00001 2012-

03-02

publication N. J.

Rogers

et al.

2010 Physico-chemical

behaviour and algal

toxicity of

nanoparticulate

CeO2 in freshwater

Environ. Chem. 2010,

7, 50–60.

doi:10.1071/EN09123

publication Franklin

et al.

2007 Comparative

Toxicity of

Nanoparticulate

ZnO, Bulk ZnO,

and ZnCl2 to a

Freshwater

Microalga

(Pseudokirchneriella

subcapitata): The

Importance of

Particle Solubility

Environ. Sci.

Technol. 2007, 41,

8484–8490

Data access

data submitter has Letter of Access


yes


Test guideline




61

Type of method

other: dialysis


The dissolution of NM-110 in synthetic softwater medium without EDTA and buffered at pH 7.5 +- 0.1

with 2 mM piperazine-N,N’-bis(ethanesulfonic acid) (PIPES: Sigma-Aldrich) (US EPA, 1994) was

determined using the equilibrium dialysis technique described by Franklin et al. (2007). Cole Parmer

Spectra/Por 7 dialysis membranes with a molecular weight cut-off of 1000 Dalton (~1 nm nominal pore

size) and 45 mm diameter were cut into 9 cm lengths with a Teflon-coated razor blade and washed

thoroughly in Milli-Q water. The dialysis cells were filled with 10 mL Milli-Q water and sealed with acid

washed (1% v/v HNO3) plastic dialysis clips. A concentrated suspension of NM-110 was prepared for

each type of nanoparticle by accurately weighing 0.12 g NM-110 into polycarbonate vials, pipetting 10

mL deionised water into the vial, and sonicating for 30 minutes. The suspension was then shaken

vigorously and quantitatively transferred into 3 L of synthetic softwater to achieve 40 mg/L ZnO in

solution. Triplicate tanks were utilised in dialysis tests so that three replicates were sampled at each time-

point. The dialysis cells were added to the test solution and continually stirred under constant light and

temperature (24ºC) conditions for 72 hours. The total volume of the dialysis cells was kept to below 5%

of the test solution in order to minimise dilution effects as dissolved zinc diffused into the dialysis cells.

A cell was removed from each triplicate tank at each sampling time andan aliquot of solution was

removed from the cells by pipette. A 5 mL volume of the external solution was also withdrawn using a

syringe and filtered through a 0.1 μm filter (Pall) at each time point to measure the zinc in this fraction.

Total zinc was measured at the start and end of the experiment only. The samples were acidified to 0.5%

v/v HNO3 (Merck tracepur) and the zinc concentration measured by ICP-MS (Agilent 7500ce).

GLP compliance

no

Test materials


yes


Identifier Identity


EC number 215-222-5

IUPAC name oxozinc


Test material form

nanomaterial








Code: NM111


62

Supplier: BASF SE





TM Code: NM112






Code: NM113







Sample I.D. NM110

Dispersion medium US EPA medium

pH 7.5

Solubility 6.77 ± 0.12 mg/L



Attached document Remarks

NM_110_ANNEX_A26_Franklin et al 2007.pdf / 465.82 KB (application/octet-stream):


NM_110_ANNEX_A27_Rogers et al 2010.pdf / 395.58 KB (application/octet-stream):



Conclusions

The dissolution of NM 110 was determined to be 6.77 ± 0.12 mg/L at pH = 7.5 in US EPA medium.

Executive summary

The dissolution of NM-110 in synthetic softwater medium without EDTA and buffered at pH 7.5 ± 0.1

with 2 mM piperazine-N,N’-bis(ethanesulfonic acid) was determined using the equilibrium dialysis

technique described by Franklin et al., 2007. The dissolution of NM-110 was determined to be 6.77 ±

0.12 mg/L at pH = 7.5 in US EPA medium. Due to the different test procedure, conditions and the lack of

data from the other three nanomaterials, this result was regarded as less relevant.


63

Endpoint study record: WoE_2010-12-09_NPL for PROSPECT_Water solubility

Administrative Data




Rationale for

reliability incl.

deficiencies

The study was regarded as scientifically acceptable. However, only limited documentation of

the data evaluation was provided in the publication. It was shown that the nanomaterials

highly agglomerate in water and that the media has a significant influence on the same. The

agglomeration of the test item was not determined by supporting analytics. Thus, the

influence of the agglomeration of the nanomaterial could not be estimated. The Powder was

not subsampled by spinning riffler. Therefore the study was regarded as reliable with

restrictions.

Data source

Reference

Reference

type


source

Testing

laboratory

Report

no.

Owner

company

Company

study no.

Report

date

study

report

National

Physical

Laboratory

2010 DISSOLUTION

OF NMs IN

VARIOUS

MEDIA

National

Physical

Laboratory,

Teddington,

Middlesex,

United

Kingdom

2010-

12-09

Data access





Test guideline



Type of method

ICP-MS


Dissolution to the mass proportion of the NM in colloidal suspension that has lost its particulate character

i.e. from a particle form to ionic or molecular form i.e. measuring the ion concentration over a specific

period of time. This involved a sample extraction followed by analysis i.e. extraction in the removal of

the NM particles and subsequent collection of supernatant using ICP-MS; the series of data below ties in

with the calorimetric testing. Dispersion was performed in accordance to the recommended PROSPEcT

video. Concentrations of 50 mg/L was made up for each sample; a total volume of 1 L was made up and

stored in clean media (1 L) bottles at room temperature. After day 2, the bottles were stored in the fridge.

Several extractions from the 1 L sample was made over a period of 22 days; prior to extraction, the

bottles were gently agitated (this was done by hand) to allow proper mixing to ensure homogeneity. The

extracted sample (~ 50 ml) was then subjected to a three-step process in order to remove particles and to

extract the resultant supernatant. This step was done immediately after extraction. First step, involved the


64

extraction of aggregates/agglomerates using filtration method, through a Millipore Express PES

membrane, 0.1 μm pore size filter (Fisher, UK) under vacuum. In the second step, the resultant filtrant

was centrifuged (Centrifuge 5430, Eppendorf, UK) (7500 rpm for one hour). Finally, the extraction of the

clear supernatant was carried out by using Peri-Star Pro peristaltic pump (World Precision Instruments,

UK); this was done carefully (so as to not disturb the pellet). Less than half of the supernatant was

collected. The resultant supernatant was then stored in the freezer for analysis using ICP-MS; this work

was subcontracted and performed by a personnel in LGC (Laboratory Government Chemist, UK). The

ICP-MS analysis was carried out using an Agilent 7500ce ICP-MS Octopole Reaction System, operating

in standard (no collision cell gas) mode for Cerium (Ce) and Helium mode for Zinc (Zn). The instrument

is UKAS accredited and was set up following standard operating procedure (SOP) INS/A1-0013. The

samples were defrosted and equilibrated to room temperature, then shaken, to ensure homogeneity. An

aliquot of 0.2g – 0.23g was taken from each sample and digested in a CEM Discover microwave, SOP

INS/A1-0014, using a mixture of HNO3/H2O2. The digested samples were then diluted to 5g prior to

analysis. The samples were digested and analysed over a period of 5 days. Validation was carried out

following SOP INS/A1-0015, this includes spiked recoveries and replicate analyses.

GLP compliance

no

Test materials


yes


Identifier Identity


EC number 215-222-5

IUPAC name oxozinc

other: OECD Sponsorship Programme NM110, Nm111, NM112, NM113

Test material form

nanomaterial













TM Code: NM112






65


Code: NM113







Data showing the ICP-MS test results for cerium. The ICP-MS measurement was used to evaluate

PROSPEcT NMs (extracted supernatant) when dispersed over time in one of four different media: a) DI water b) fish medium c) daphnia medium d) seawater

DAY Zinc concentration of the supernatant extracted (ng g-

1)

NM112 NM113 NM110

DI water 2 764 1864 2536

6 1741 3436 3360

9 1490 2813 3130

14 1808 3005 3772

22 1607 6007 5030

Nanosun ZnO,

Microniser

Micron ZnO,

Sigma Aldrich

Z-COTE ZnO,

BASF

Fish 2 1198 1780 2216

6 1632 2442 2192

9 1744 2420 3028

14 1676 2961 2697

22 1954 3239 3036

Nanosun ZnO,

Microniser

Micron ZnO,

Sigma Aldrich

Z-COTE ZnO,

BASF

Daphnia 2 1158 1465 1454

6 1458 1644 sample missing

9 1731 1515 1014

14 1052 2193 1588

22 1402 2611 2037

Nanosun ZnO,

Microniser

Micron ZnO,

Sigma Aldrich

Z-COTE ZnO,

BASF

Seawater 2 241 531 681

6 371 466 736

9 439 605 773

14 420 1089 972

22 359 1051 1155


66




NM_110_ANNEX_A02_FINALINTERIM REPORT2010corrected.pdf / 1.05 MB (application/octet-


NM_110_ANNEX_A29_AdditionalDissolutionDataICP-MS.doc.doc / 111.5 KB (application/octet-



Conclusions

Dissolution rates were fastest when the NMs were dispersed in deionized water, with NM 110 dissolving

the fastest and NM 112 dissolving the slowest. Deionized water yielded the most stable dispersions and

this increase in stability will mean less aggregation/agglomeration (and subsequent sedimentation) in the

dispersion. Out of all the ecotox media, fish medium had the largest dissolution rate followed by daphnia

and then seawater.

Executive summary

The dissolution of NM 110, NM 112, NM 113 in different media (deionized water, fish medium, daphnia

medium and seawater) were investigated by NPL, 2010 using inductively coupled plasma mass

spectrometry (ICP-MS). Dispersion was performed in accordance to the recommended PROSPEcT

protocol. A concentration of 50 mg/L was made up for each sample; a total volume of 1 L was made up

and stored in clean media (1 L) bottles at room temperature. After day 2, the bottles were stored in the

fridge. Several extractions from the 1 L sample were made over a period of 22 days. The extracted sample

(~ 50 mL) was then subjected to a three-step process in order to remove particles and to extract the

resultant supernatant. Dissolution rates were fastest when the NMs were dispersed in deionized water,

with NM 110 dissolving the fastest and NM 112 dissolving the slowest. Deionized water yielded the most

stable dispersions and this increase in stability will mean less aggregation/agglomeration (and subsequent

sedimentation) in the dispersion. Out of all the ecotox media, fish medium had the largest dissolution rate

followed by daphnia and then seawater. Dispersing NMs in such ecotox media would mean less stable

dispersion and this subsequently equates to the reduced surface area concentrations and thus a lower

dissolution rate.

Endpoint study record: WoE_2010-12-09_NPL for PROSPECT_Water solubility

Administrative Data




Rationale for

reliability incl.

deficiencies


the data evaluation was provided. It was shown that the nanomaterials highly agglomerate in

water and that the media has a significant influence on the same. The agglomeration of the

test item was not determined by supporting analytics. Thus, the influence of the

agglomeration of the nanomaterial could not be estimated. The Powder was not subsampled

by spinning riffler. Therefore the study was regarded as reliable with restrictions.


67

Data source

Reference

Reference

type


source

Testing

laboratory

Report

no.

Owner

company

Company

study no.

Report

date

study

report

National

Physical

Laboratory

2010 DISSOLUTION

OF NMs IN

VARIOUS

MEDIA

National

Physical

Laboratory,

Teddington,

Middlesex,

United

Kingdom

2010-

12-09

Data access





Test guideline


no guideline

followed

Type of method

flask method


Deionised water was employed as the corresponding media control and was used to prepare all aqueous

solutions and suspensions. Zinc ions were measured using Cole-Palmer® Colorimetric Test Kits (Cole

Palmer, UK). The colorimetric measurement was used to evaluate the nanomaterials when dispersed in

different media over time of 21 days; the extracted supernatant from the dispersions were obtained prior

to performing the colorimetric tests. The dispersions were stored in a refrigerator after day 2 in order to

prevent degradation of the sample e.g. minimising bacterial growth.

GLP compliance

no

Test materials


yes


Identifier Identity


EC number 215-222-5

IUPAC name oxozinc

other: OECD Sponsorship Programme NM110, Nm111, NM112, NM113


68

Test material form

nanomaterial



Code: NM110

Supplier: BASF SE





Code: NM111

Supplier: BASF SE





TM Code: NM112






Code: NM113






Details on results

The results of the colorimetric zinc ion tests; the aim here was to evaluate the dissolution events, for 21

days, of the PROSPEcT ZnO powders in the various media. The dispersions were stored in a refrigerator

after day 2 in order to prevent degradation of the sample e.g. minimising bacterial growth. Results show

the following trends: a) Dissolution rates were fastest when the NMs were dispersed in DI water, with Z-

COTE ZnO from BASF dissolving the fastest and Nanosun ZnO dissolving the slowest. As shown earlier

in this report, DI water yielded the most stable dispersions and this increase in stability will mean less

aggregation/agglomeration (and subsequent sedimentation) in the dispersion. Hence, the total surface area

is greater when the particles are dispersed in DI if compared to corresponding ecotox media; an increase

in surface area means that the ion dissolution rate will also increase.b) Of particular interest is the result in

Figure 2a, in which we see an apparent decrease in zinc concentrations from Day 6 to Day 9, for all zinc

oxide NMs. This effect may be indicative of the dissolution-precipitation process occurring during this

time.c) Out of all the ecotox media, fish medium had the largest dissolution rate followed by daphnia and

then seawater. Dispersing NMs in such ecotox media would mean less stable dispersion and this

subsequently equates to the reduced surface area concentrations and thus a lower dissolution rate. In

addition, the much larger ionic concentration in seawater may indirectly affect the dissolution rates,

possibly through the ability to influence “inner–sphere adsorption”, which have been known to be

important in mineral dissolution.


69




NM_110_ANNEX_A28_DISSOLUTION OF NMs IN VARIOUS MEDIA.doc / 2.91 MB (application/octet-



Conclusions

Out of all the ecotox media, fish medium had highest dissolution rate followed by daphnia and then

seawater. Dispersing nanomaterials in such ecotox media would mean less stable dispersion and this

subsequently equates to the reduced surface area concentrations and thus a lower dissolution rate.

Executive summary

Dispersion of NM-110, NM-112 and NM-113 in an ecotoxicology relevant media (fish medium, daphnia

medium and seawater) was investigated by the NPL, 2010. Deionised water was employed as the

corresponding media control and was used to prepare all aqueous solutions and suspensions. Zinc ions

were measured using Cole-Palmer® Colorimetric Test Kits (Cole Palmer, UK). The colorimetric

measurement was used to evaluate the nanomaterials when dispersed in different media over time of 21

days; the extracted supernatant from the dispersions were obtained prior to performing the colorimetric

tests. The dispersions were stored in a refrigerator after day 2 in order to prevent degradation of the

sample e.g. minimising bacterial growth. It was shown that dissolution rates were fastest, when the NMs

were dispersed in deionized water, with NM-110 dissolving fastest and NM-112 dissolving slowest.

Deionized water yielded the most stable dispersions and this increase in stability will result in less

aggregation/agglomeration (and subsequent sedimentation) in the dispersion. The total surface area was

greater when the particles were dispersed in deionized when compared to corresponding ecotox media.

Thus, an increase in surface area means that the ion dissolution rate will also increase. An apparent

decrease in zinc concentrations from Day 6 to Day 9, for all zinc oxide NMs was observed. This effect

may be indicative of the dissolution-precipitation process occurring during this time. Out of all the ecotox

media, fish medium had highest dissolution rate followed by daphnia and then seawater. Dispersing


the reduced surface area concentrations and thus a lower dissolution rate.


70

4.9 Solubility in organic solvents / fat solubility

4.10 Surface tension

4.11 Flash point

4.12 Auto flammability

4.13 Flammability

4.14 Explosiveness

4.15 Oxidising properties

4.16 Oxidation reduction potential

Endpoint summary: Oxidation reduction potential

Administrative Data


No reliable information was available.

Discussion

A study was conducted by NPL, 2010 using Cyclic voltammetric method to determine the redox potential

of NM-110. Results show two redox processes that are taking place, which was referred to as “Redox 1”

and “Redox 2”. Each redox reaction consists of two half-reactions i.e. for oxidation and reduction

reactions and these correspond to the oxidation peak and reduction peak in the C-V plot.

Furthermore it was indicated by the study director that the values should be treated with caution as:

a) Peaks were very broad, so numbers are only approximate.

b) Peak-peak separation was large, indicating that the processes were not fully reversible.

NPL, 2010 investigated the redox potential measurements, using ORP probe electrode, of various ZnO

dispersions (NM-100, NM-112 and NM-113), in various liquid media. NM-111 could not be investigated

by this method as it was difficult to disperse. The redox potential ORP electrode was calibrated against

Calibration Solution. This standard solution was also used to verify the performance of the electrode in

the beginning and end of the study. Redox potential measurements were carried out on freshly dispersed

NM in various media (deionized water, fish medium, daphnia medium and seawater). Dispersion of the

individual nanomaterial in the appropriate liquid media was carried out in accordance to the protocol as

recommended under PROSPECT/OECD (as part of the OECD guidelines). Redox potential was

determined to be highest in fish water and to be lowest in seawater.

The results showed that there were no significant differences in the redox potential between the different

nanomaterials.


71

Endpoint study record: WoE_2010-12-05 NPL for PROSPECT Oxidation

reduction potential.001

Administrative Data




Rationale for

reliability incl.

deficiencies

It was indicated by the study director that, the assignment is speculative in nature, as the

Pourbaix diagram is a potential-pH diagram and is a map of thermodynamic possibilities. The

Pourbaix diagram may well identify the lowest energy state of Zn but the exact speciation

may not exist as the corresponding redox reaction may not be favoured for kinetic reasons.

Therefore the reliable of the study could not be assigned.

Data source

Reference

Reference

type


source

Testing

laboratory

Report

no.

Owner

company

Company

study no.

Report

date

study

report

National

Physical

Laboratory

2010 Redox

potential

measurements

National

Physical

Laboratory,

Teddington,

Middlesex,

United

Kingdom

2010-

12-05

Data access






Cyclic voltammetry. Cyclic voltammetric experiments were performed using Autolab PGSTAT 12

potentiostat, equipped with a PC for electrochemical measurement. A conventional three electrode cell

was employed throughout the experiments with: NM electrode as the working electrode (WE), Hg/HgO

as a reference (RE) and platinum electrode as the counter electrode (CE). Prior to use the Hg/HgO

reference electrode was calibrated against a saturated calomel electrode (-149 mV against a SCE). Cyclic

voltammograms were acquired after exposure of the working electrode to the test solution (100 ml) in a

glass filled cell; electrical connection to the working electrode was achieved by connecting a small

mounted crocodile clip to the silver gauze that was attached to the WE. The WE was immersed so that

approximately two thirds of the disc (and no silver) was exposed to the solution. Cyclic voltammetry (5

cycles) was undertaken at a rate of 100 mV/s, between the potentials of 0.7 V and –1.8 V.

GLP compliance

no

Test materials


yes


72


Identifier Identity


EC number 215-222-5

IUPAC name oxozinc


Test material form

nanomaterial


electrode made from bulk NM













TM Code: NM112






Code: NM113







Figure shows a typical cyclic voltammogram (C-V) of PROSPEcT Z-COTE ZnO (BASF,) electrode in

seawater at a scan rate of 100 mV/s. Results show two redox processes that are taking place, which will

be referred to as “Redox 1” and “Redox 2”. Each redox reaction consists of two half-reactions i.e. for

oxidation and reduction reactions and these correspond to the oxidation peak and reduction peak in the C-

V plot. From these values, a mid-point potential was reported, which gave an indication of the

corresponding redox potential value; the information from the C-V plot is summarised on Table 10. It is

evident from the C-V plot that the redox potential values reported here should be treated with caution as: A) Peaks were very broad, so numbers are only approximate B) Peak-peak separation was large, indicating that the processes were not fully reversible A summary of reduction and oxidation potential and the corresponding redox potential values for Z-COTE


73

ZnO electrode PROSPEcT Z-COTE ZnO electrode (BASF, Germany), batch number ZC250#37#RP.

Reduction peak

potential (anodic

process)

(mV)

Oxidation peak

potential

(cathodic

process)

(mV)

Redox potential

(mV)

Redox system

Redox 1 -1550 -980 -1265 ZnO/Zn

Redox 2 -130 400 135 ?




NM_110_ANNEX_A30_REDOX POTENTIAL MEASUREMENTS.doc / 64.5 KB (application/octet-stream):




Conclusions

Results showed that redox 1 to ZnO/Zn was ascribed attentively, as interpreted from the Pourbaix

diagram. However, the assignment was speculative in nature, as the Pourbaix diagram was a potential-pH

diagram and was a map of thermodynamic possibilities. The Pourbaix diagram may well identified the

lowest energy state of Zn but the exact speciation may not existed as the corresponding redox reaction

may not was favoured for kinetic reasons. Currently, the assignment of Redox 2 had been ascribed and

there is a need to conduct a thorough review of the literature for interpretation of Redox 2.

Executive summary

A study was conducted by NPL, 2010 using Cyclic voltammetric method to determine the redox potential

of NM-110. Results show two redox processes that are taking place, which was referred to as “Redox 1”

and “Redox 2”. Each redox reaction consists of two half-reactions i.e. for oxidation and reduction

reactions and these correspond to the oxidation peak and reduction peak in the C-V plot.


74

Furthermore it was indicated by the study director that the values should be treated with caution as: a) Peaks were very broad, so numbers are only approximate. b) Peak-peak separation was large, indicating that the processes were not fully reversible.

Endpoint study record: WoE_2010-12-10_NPL for

PROSPECT_Oxidationreduction potential

Administrative Data




Rationale for

reliability incl.

deficiencies

The redox potential values acquired from ORP electrode may be indicative of the redox

state of the entire system and it is difficult to quantify the reliability of such

measurements. Thus, the reliability of the study could not be assigned.

Data source

Reference

Reference

type


source

Testing

laboratory

Report

no.

Owner

company

Company

study no.

Report

date

study

report

National

Physical

Laboratory

2010 Redox

Potential

National

Physical

Laboratory,

Teddington,

Middlesex,

United

Kingdom

2010-

12-10

Data access






Redox potential were measured using an ORP Oakton® Waterproof ORP Testr®, purchased from Cole

Palmer UK; this in effect measures the potential difference across two electrodes i.e. a Pt electrode

against a double junction Ag/AgCl reference electrode. The electrode was used in accordance to the

manufacturer’s instructions. Prior to use the electrode was pre-condition in clean tap water for 30 minutes

before rinsing in distilled water. When making measurements, the electrode was carefully placed in a vial

containing the sample; there must be sufficient liquid sample to cover the sensing element. The electrode

was carefully stirred a little and then placed in a fixed position, slightly above the bottom of the container.

The signal output was allowed to settle for 5 minutes before a reading i.e. the “field potential” was noted.

After measurement, the electrode was cleaned with tap water and final rinse was with distilled water, after

which further measurements can be made. When not in use, the electrode was stored in a solution of

Oakton® electrode storage solution, as recommended by the manufacturer.

GLP compliance

no


75

Test materials


Identifier Identity


EC number 215-222-5

IUPAC name oxozinc


Test material form

nanomaterial



Code: NM110

Supplier: BASF SE





Code: NM111

Supplier: BASF SE





TM Code: NM112






Code: NM113






The redox potential ORP electrode was calibrated against YSI® Zobell ORP Calibration Solution

(purchased from Cole Palmer); this reagent was made available in dry form and was reconstituted with

125 mL of DI water prior to use, after which the solution has ~ 6 months expiry date. This standard

solution was also used to verify the performance of the electrode in the beginning and end of the study.

For Ag/AgCl reference, the redox potential value for Zobell solution was 231 ±10 mV (depending on

temperature); at ~ 20 C, this value was ~ 237 mV. Redox potential was carried out on freshly dispersed

NM in various media; the media was the DI water and the three ecotox media chosen by the University of

Exeter (fish, daphnia, water flea and seawater). All field potential values recorded were subjected to an

additive correction factor of +206 mV; this was necessary so that the final value was reported as if the

reference electrode was a standard hydrogen reference electrode instead of the Ag/AgCl. Dispersion of

the individual nanomaterial in the appropriate liquid media was carried out in accordance to the protocol


76

recommended under PROSPEcT programme (as part of the OECD guidelines); currently this protocol is

accessible through the NIA website. Note: protocol associated with making pellets and measuring redox

potential via acquiring the cyclic voltamogram (initially proposed) was not used. Several issues were

raised, to include the fact that these are semiconducting based material and therefore has a high

resistance; current flow will be limited under such measurements. In addition, the pellets were not shown

to be not stable i.e. there were signs of disintegration.



MEDIA BLANKS ONLY Redox potential (mV) pH

DI water 405 N/A

Daphnia 425 7.94

Fish water 418 7.34

Seawater 384 8.75 Redox potential of NM dispersion in various liquid media, the value quoted is relative to the standard

hydrogen reference electrode; values quoted in mV.

NM112 NM113 NM110

DI water 398 398 396

Fish media 424 430 427

Daphnia media 415 415 422

Seawater 380 374 379




NM_110_ANNEX_A31_Redox Potential from NPL.rtf / 46.55 KB (application/rtf):


NM_110_ANNEX_A02_FINALINTERIM REPORT2010corrected.pdf / 1.05 MB (application/octet-



Conclusions

Redox potential was determined to be highest in fish water and to be lowest in seawater. The results

showed that there were no significant differences in the redox potential between the different

nanomaterials

Executive summary

NPL, 2010 investigated the redox potential measurements, using ORP probe electrode, of various ZnO

dispersions (NM-100, NM-112 and NM-113), in various liquid media. NM-111 could not be investigated

by this method as it was difficult to disperse. The redox potential ORP electrode was calibrated against

Calibration Solution. This standard solution was also used to verify the performance of the electrode in

the beginning and end of the study. Redox potential measurements were carried out on freshly dispersed

NM in various media (deionized water, fish medium, daphnia medium and seawater). Dispersion of the

individual nanomaterial in the appropriate liquid media was carried out in accordance to the protocol as

recommended under PROSPECT/OECD (as part of the OECD guidelines).Redox potential was

determined to be highest in fish water and to be lowest in seawater. The results showed that there were no

significant differences in the redox potential between the different nanomaterials.


77

4.17 Stability in organic solvents and identity of relevant degradation

products

4.18 Storage stability and reactivity towards container material

4.19 Stability: thermal, sunlight, metals

Endpoint summary: Stability: thermal, sunlight, metals

Administrative Data


All weight-losses were small (less than 6%), little / no oxidation occurred up to 1100 °C. Slight

differences in the degree of weight loss were observed between different samples; NM-110 and NM-113

were both very stable and exhibited minor losses (< 0.6 wt %) while losses from NM-111 and NM-112

were much higher (4-6wt% and 1-2wt% respectively). Mass spectrometry identified CO2 and water when

both NM-111 and NM-112 were heated in air, but not when NM-110 and NM-113 were heated in air.

When heated in argon, traces of both CO2 and water were identified in all samples. The 4-6 wt% loss by

NM-111 suggests that the quantity of coating may be higher than the nominative 2 wt %.

Discussion

CSIRO, 2012 determined the thermal properties of the different nanomaterials using a Setaram Evolution

Differential Thermal Analysis / Thermogravimetric (DTA/TG) thermal analyser with a carrier gas of air

or argon, and using a heating ramp rate of 20°C/min to a maximum temperature of 1100°C. Volatile

species were simultaneously analysed by mass spectrometry; species of interest were those with atomic

mass units of 2, 12, 16, 17, 18, 28, 44, 48, and 64. All weight-losses were small (less than 6 %),

indicating that little/no oxidation occurred up to 1100 °C. Slight differences in the degree of weight loss

were observed between different samples; NM-110 and NM-113 were both very stable and exhibited

minor losses (< 0.6 wt%) while losses from NM-111 and NM-112 were much higher (4-6 wt% and 1-

2 wt% respectively). Mass spectrometry identified CO2and water when both NM-111 and NM-112 were

heated in air, but not when NM-110 and NM-113 were heated in air. When heated in argon, traces of both

CO2and water were identified in all samples. The 4-6 wt% loss by NM-111 suggests that the quantity of

coating may be higher than the nominative 2 wt %. For NM-111, the onset of coating decomposition

occurred at similar temperatures in both air (328 °C) and argon (290 °C). However, NM-111 exhibited a

much sharper and larger (5-7 times) exotherm in air compared with argon. These observations are

consistent with thermal oxidation of the coating which occurs between 350 - 420 °C. Degradation via

thermal depolymerization occurs at higher temperatures (420 - 480 °C). The respective exotherms of

311 J/g (air) and 40 J/g (argon) reflect the oxidative nature of the reaction for NM-111, and the larger

amounts of CO2 and water detected by mass spec (compared with uncoated samples NM-110, NM-112

and NM-113) confirm coating degradation. The fact that NM-112 is the only uncoated ZnO that exhibits

CO2and water release at low temperatures combined with having the highest surface area suggests that it

is more reactive in its adsorption behavior and is more likely to be less stable in the longer term than its

larger-sized counterparts.

In conclusion it was shown that the difference in thermal properties for the 4 nanomaterials is not linked

with the material form (nano or bulk) but rather with coating.


78

Endpoint study record: Key_2012-03-

05_Australia_CSIRO_Thermal_Stability_DTA/TG

Administrative Data




Rationale for

reliability incl.

deficiencies

The study was regarded as scientifically acceptable. Non-GLP study but similar to

guideline. However, only limited documentation of the full study report was available.

Thus, the study was regarded as reliable with restrictions.

Data source

Reference

Reference

type


source

Testing

laboratory

Report

no.

Owner

company

Company

study no.

Report

date

study

report

Casey,

Yin,

McCall

2012 Summary of

physicochemical

properties of

OECD Zinc

Oxide (ZnO)

samples -

Thermal

analysis

Commonwealth

Scientific and

Industrial

Research

Organisation

2012-

03-05

Data access





Test guideline


equivalent or similar

to

OECD Guideline 113 (Screening Test for Thermal Stability and Stability in

Air)

no


Thermal properties of the ZnO samples were analysed using a Setaram Evolution Differential Thermal

Analysis /Thermogravimetric (DTA/TG) thermal analyser with a carrier gas of air or argon, and using a

heating ramp rate of 20°C/min to a maximum temperature of 1100°C. Volatile species were

simultaneously analysed by mass spectrometry; species of interest were those with atomic mass units of

2, 12, 16, 17, 18, 28, 44, 48, and 64.

GLP compliance

no

Test materials


yes


79


Identifier Identity


EC number 215-222-5

IUPAC name oxozinc


Test material form

nanomaterial








Code: NM111

Supplier: BASF SE





TM Code: NM112






Code: NM113





Details on methods

Thermal properties of the ZnO samples were analysed using a Setaram Evolution Differential Thermal

Analysis /Thermogravimetric (DTA/TG) thermal analyser with a carrier gas of air or argon, and using a

heating ramp rate of 20°C/min to a maximum temperature of 1100°C. Volatile species were

simultaneously analysed by mass spectrometry; species of interest were those with atomic mass units of

2, 12, 16, 17, 18, 28, 44, 48, and 64.


For study on stability to metals


Figure 1 and 2 present the DTA/TGresults of ZnO samples heated to 1100 °C under air or argon, and

Table 2 lists the losses in weight for these samples. All weight-losses were small (less than 6%),




80

minor losses (< 0.6 wt %) while losses from NM-111 and NM-112 were much higher (4-6wt% and 1-

2wt% respectively). Mass spectrometry identified CO2 and water when both NM111 and NM112 were



coating may be higher than the nominative 2 wt %.

For NM-111, the onset of coating decomposition occurred at similar temperatures in both air (328 °C)

and argon (290 °C) (DTA curves shown in Figure 1b and 2b, respectively). However, NM-111 exhibited

a much sharper (peak maxima occurred at 396 °C in air, and 580 °C in argon) and larger (5-7 times)

exotherm in air compared with argon. These observations are consistent with thermal oxidation of the

coating which occurs between 350 – 420 °C. Degradation via thermal depolymerization (ie non oxidative)

occurs at higher temperatures (420-480 °C)[i]. The respective exotherms of 311 J/g (air) and 40 J/g

(argon) reflect the oxidative nature of the reaction for NM-111, and the larger amounts of CO2and water

detected by mass spec (compared with uncoated samples NM-110, NM-112 and NM-113) confirm

coating degradation.

Water and CO2were detected when NM-112 was heated under both air and argon, The generation of

these species consistently occurred at two different temperatures, 250-300 °C and 500-550 °C, with the

lower temperature generating the greater portion of the total released. This presumably reflects desorption

of physically/chemically adsorbed water/CO2species at the lower temperature, while a structural sintering

or rearrangement is more likely at the higher temperature. The fact that NM-112 is the only uncoated ZnO

that exhibits CO2and water release at low temperatures combined with having the highest surface area

suggests that it is more reactive in its adsorption behavior and is more likely to be less stable in the longer

term than its larger-sized counterparts. As this sample was not supplied in a coated form it unclear

whether this could be mitigated by such

Per cent weight loss from ZnO samples heated under air or argon, as determined by DTA/TG. NM-110 NM-111 NM-112 NM-113

Weight loss under air 0.3% 5.8% 1.7% 0.2%

Weight loss under argon 0.3% 4.1 % 1.3% 0.6%

[i]Polymer Data Handbook. Oxford University Press, Inc. p 427.




NM_110_ANNEX_A32_OECD ZnO thermal stability CSIRO Australia.docx / 372.72 KB




81


Conclusions

All weight-losses were small (less than 6%), little / no oxidation occurred up to 1100oC. Slight

differences in the degree of weight loss were observed between different samples; NM-110 and NM-113

were both very stable and exhibited minor losses (< 0.6 wt %) while losses from NM-111 and NM-112

were much higher (4-6wt% and 1-2wt% respectively). Mass spectrometry identified CO2 and water when

both NM-111 and NM-112 were heated in air, but not when NM-110 and NM-113 were heated in air.

When heated in argon, traces of both CO2 and water were identified in all samples. The 4-6 wt% loss by

NM-111 suggests that the quantity of coating may be higher than the nominative 2 wt %.

Executive summary

CSIRO, 2012 determine the thermal properties of the different nanomaterials using a Setaram Evolution

Differential Thermal Analysis / Thermogravimetric (DTA/TG) thermal analyser with a carrier gas of air

or argon, and using a heating ramp rate of 20°C/min to a maximum temperature of 1100°C. Volatile

species were simultaneously analysed by mass spectrometry; species of interest were those with atomic

mass units of 2, 12, 16, 17, 18, 28, 44, 48, and 64. All weight-losses were small (less than 6 %),



minor losses (< 0.6 wt%) while losses from NM-111 and NM-112 were much higher (4-6 wt% and 1-

2 wt% respectively). Mass spectrometry identified CO2and water when both NM-111 and NM-112 were



coating may be higher than the nominative 2 wt %. For NM-111, the onset of coating decomposition

occurred at similar temperatures in both air (328 °C) and argon (290 °C). However, NM-111 exhibited a

much sharper and larger (5-7 times) exotherm in air compared with argon. These observations are

consistent with thermal oxidation of the coating which occurs between 350 - 420 °C. Degradation via


82

thermal depolymerization occurs at higher temperatures (420 - 480 °C). The respective exotherms of

311 J/g (air) and 40 J/g (argon) reflect the oxidative nature of the reaction for NM-111, and the larger

amounts of CO2 and water detected by mass spec (compared with uncoated samples NM-110, NM-112

and NM-113) confirm coating degradation. The fact that NM-112 is the only uncoated ZnO that exhibits

CO2and water release at low temperatures combined with having the highest surface area suggests that it

is more reactive in its adsorption behavior and is more likely to be less stable in the longer term than its

larger-sized counterparts.

In conclusion it was shown that the difference in thermal properties for the 4 nanomaterials is not linked

with the material form (nano or bulk) but rather with coating.

4.20 pH

4.21 Dissociation constant

4.22 Viscosity

4.23 Additional physico-chemical information

Endpoint study record: supporting_2010-12-03 NPL for PROSPECT Stability

monitoring_Turbidity

Administrative Data




Rationale for reliability

incl. deficiencies

Limited documentation of the data evaluation was provided in the full study

reports. Thus, the reliability could not be assigned.

Data source

Reference

Reference

type


source

Testing

laboratory

Report

no.

Owner

company

Company

study no.

Report

date

study

report

National

Physical

Laboratory

2010 Dispersion

stability as

determined by

turbidity

measurements

National

Physical

Laboratory,

Teddington,

Middlesex,

United

Kingdom

2010-

12-03

Data access





Endpoint investigated

other: Turbidity


83

Test guideline




Turbidity was measured using HF Scientific – Micro100 RI turbidity meter (Cole-Palmer, UK); this

meter has an infrared light source that meets the international standard ISO 7027 for turbidity

measurements. The meter was calibrated on standards, which are based on AMCO-AEPA-1

microspheres; these standards are traceable to standard formazin suspension. Standard values of 1000, 10

and 0.02 NTU were used to calibrate the meter. Prior to use, the meter was allowed to warm up for 30

minutes. Sample cuvettes (HF Scientific (USA)) were used to hold the sample. Note that glass thickness

may vary from cuvette to cuvette and within the same cuvette. Hence, individual vials were indexed;

indexing of the cuvette entails finding the point of the cuvette that light passes through that gives the

lowest reading; once indexed the holder can be marked accordingly. Prior to their use, cuvettes were

cleaned, in accordance to manufacturer’s instructions. This involved washing the interior and exterior of

the cuvette with a detergent (2% Hellmanex in DI water); it was then rinsed several times in distilled

water before finally rinsing in DI water. The cuvette was further rinsed with the sample two times before

filling (30ml) and analysed. The cuvette was placed into the meter and signal allowed to settle before

taking readings.

GLP compliance

no

Test materials


yes


Identifier Identity


EC number 215-222-5

IUPAC name oxozinc


Test material form

nanomaterial



Code: NM110

Supplier: BASF SE





Code: NM111

Supplier: BASF SE





84


TM Code: NM112






Code: NM113






Results

Table shows the corresponding “half-lives” of the PROSPEcT powders when dispersed in the various

media. The concept of “half lives” has been put forward in the OECD guidelines on NM testing and this

value is an indication of dispersion stability through time i.e. the larger the half life value the longer it

takes for the concentration to reduce by half and thus the more stable the dispersion. Results show that

overall NMs are most stable when dispersed in DI water and least stable when in an ecotox media.


Dispersion stability of PROSPEcT powders as measured by their corresponding “half lives” (the time it

takes for particle concentration to reduce by half, as determined from turbidity values).

Sample Name Supplier DI water (min) Fish media

(min)

Seawater

(min)

Daphnia media

(min)

NM-112 Micronisers, Australia 2526 498 402 444

NM-113 Sigma Aldrich, UK 966 216 228 324

NM-110 BASF, Germany 4038 816 738 768




NM_110_ANNEX_A08_DISPERSION STABILITY AS DETERMINED BY TURBIDITY

MEASUREMENTS.doc / 35.5 KB (application/octet-stream): ENV/JM/MONO(2015)15/ANN1


Conclusions

Results show that overall NMs are most stable when dispersed in DI water and least stable when in an

ecotox media.


85

Endpoint study record: supporting_2012-03-

05_Australia_CSIRO_Chemical_Analysis_ICP-AES

Administrative Data




Rationale for reliability incl.

deficiencies

The study was conducted in accordance with ISO-standardized

method.

Data source

Reference

Reference

type


source

Testing

laboratory

Report

no.

Owner

company

Company

study no.

Report

date

study

report

Casey,

Yin,

McCall

2012 Summary of

physicochemical

properties of

OECD Zinc

Oxide (ZnO)

samples -

Chemical

Analysis

Commonwealth

Scientific and

Industrial

Research

Organisation

2012-

03-05

Data access





Endpoint investigated

other: elemental analysis


Experiments to determine the elemental composition of the ZnO samples were performed in duplicate.

Approximately 0.15g of ZnO sample was dissolved in a 1:1 HNO3: H2O2 mixture with heating for 30

minutes. The solution was diluted to 100 mL, internal standard Sc was added, and the resultant solution

was analysed by Inductively Coupled Plasma-Atomic Emission Spectroscopy (Varian 730 Axial ICP-

AES). Certified multi-element solutions were also used to check the accuracy.

GLP compliance

no

Test materials


yes


86


Identifier Identity


EC number 215-222-5

IUPAC name oxozinc


Test material form

nanomaterial



Code: NM110

Supplier: BASF SE





Code: NM111

Supplier: BASF SE





TM Code: NM112






Code: NM113







The table presents the elemental analysis of the four ZnO samples. In general, Zn was the main element

detected with few secondary elements present, but there were noticeable differences between samples in

the quantities of trace elements detected. For example, NM-112 had substantially higher levels (5-60

times) of alkali metals (Ca, Na) and Al compared with all other samples. NM-113 appeared to have a

significantly higher level of Cu (8-45 times other samples). The amount of Si appeared to be relatively

constant across samples but was statistically higher in NM-111 probably due to the silicon-containing

dimethicone coating. The elevated alkali metal levels present in NM-112 were confirmed by the supplier

as a consistent analysis of their product, while the elevated amount of Cu in NM-113 was inconsistent

with (i.e. higher than) nominative published product data.

Table 2. Elemental analyses of four ZnO samples, NM110, NM111, NM112 and NM113. The data are

expressed in units of weight% and ppm.


87

Sample Reference NM-110 NM-111 NM-112 NM-113

Zn 213.857* 79.3 % 78.2 % 78.0 % 80.6 %

Al 396.152 24 ppm 27 ppm 130 ppm 7.9 ppm

Ca 422.673 44 ppm 74 ppm 680 ppm 38 ppm

Ce 446.021 <10 ppm <10 ppm <10 ppm <10 ppm

Co 238.892 <2.5 ppm <2.5 ppm <2.5 ppm <2.5 ppm

Cr 267.716 <1.5 ppm 1.9 ppm 0.63 ppm <1.5 ppm

Cu 324.754 15 ppm 4.8 ppm 2.7 ppm 120 ppm

Fe 259.940 <5 ppm <5 ppm <5 ppm <5 ppm

K 769.897 <15 ppm <15 ppm <15 ppm <15 ppm

Mg 285.213 <2 ppm <2 ppm <2 ppm <2 ppm

Mn 257.610 <0.3ppm <0.3 ppm 0.9 ppm <0.3 ppm

Na 589.592 3 ppm 17 ppm 176 ppm 4.2 ppm

Ni 231.604 <20 ppm <20 ppm <20 ppm <20 ppm

P 213.618 <90 ppm <90 ppm <90 ppm <90 ppm

Pb 283.305 <40 ppm <40 ppm <40 ppm <40 ppm

S 181.972 <200 ppm <200 ppm <200 ppm <200 ppm

Si 251.611 195 ppm 265 ppm 210 ppm 236 ppm

Sn 283.998 <1 ppm <1 ppm <1 ppm <1 ppm

Ti 336.122 6.4 ppm 7.7 ppm 8.2 ppm 6.4 ppm

V 311.070 <0.02 ppm <0.02 ppm <0.02ppm <0.02 ppm

Zr 339.198 <3 ppm < 3ppm <3 ppm <3 ppm




NM_110_ANNEX_A33_OECD ZnO Chemical Analysis ICPAES CSIRO Australia.docx / 24.63 KB



Conclusions

In general, Zn was the main element detected with few secondary elements present, but there were

noticeable differences between samples in the quantities of trace elements detected. For example, NM-

112 had substantially higher levels (5-60 times) of alkali metals (Ca, Na) and Al compared with all other

samples. NM-113 appeared to have a significantly higher level of Cu (8-45 times other samples). The

amount of Si appeared to be relatively constant across samples but was statistically higher in NM-111

probably due to the silicon-containing dimethicone coating. The elevated alkali metal levels present in

NM-112 were confirmed by the supplier as a consistent analysis of their product, while the elevated

amount of Cu in NM-113 was inconsistent with (i.e. higher than) nominative published product data.


88

4.24 Agglomeration/aggregation


05_Australia_CSIRO_Agglomeration/Aggregation_SEM

Administrative Data




Rationale for

reliability incl.

deficiencies

The documentation provided in the full study report was sufficient for the data evaluation. Subjective

assessment of sample agglomeration/aggregation based on viewing SEM images. Thus, the study was

regarded as reliable with restrictions.

Data source

Reference

Reference

type


source

Testing

laboratory

Report

no.

Owner

company

Company

study no.

Report

date

study

report

Casey,

Yin,

McCall

2012 Summary of

physicochemical

properties of

OECD Zinc

Oxide (ZnO)

samples - Dry

state-

representative

SEM pictures

Commonwealth

Scientific and

Industrial

Research

Organisation

2012-

03-05

Data access





Test guideline



Method

SEM

Details on methods and data evaluation

A Philips XL30 field emission Scanning Electron Microscope (SEM) was used for this study. The

optimal spatial resolution of the microscope was from 2-5 nm with varying accelerating voltage from 30

KV to 1 kV. Images of ZnO particles were acquired at an accelerating voltage of 5 kV, a working

distance of ≈ 10 mm, and a tilt angle 0°. An SEM metal stub was covered with adhesive conducting tape

and a small amount of “as received” ZnO powder(around 5 mg) was sprinkled over the tape. The surface

of the powder sample was flattened with a spatula. Excess powder was removed by gently tapping the

stub on its side until a light coating of powder on the surface became apparent. The nanoparticles were

thinly sputtered with iridium using a Polaron SC570 sputter coater. Sputtering was conducted under

vacuum while passing gas was argon. The coating deposition time was 20 seconds at a plate current of

50mA, giving a coating thickness of approximately 1 nm.


89

Data gathering

Instruments

Philips XL30 field emission Scanning Electron Microscope (SEM)

GLP compliance

no

Test materials


yes

Test material Identity

Identifier Identity


EC number 215-222-5

IUPAC name oxozinc


Test material form

nanomaterial - aerosol



Code: NM110

Supplier: BASF SE










TM Code: NM112






Code: NM113







Attached full

study report

NM_110_ANNEX_A34_OECD ZnO Agglommeration Aggregation SEM CSIRO

Australia.docx / 313 KB (application/octet-stream):



90



Conclusions

SEM images reveal that ZnO particles as powder without media were agglomerated. The particle size was

smallest for sample NM 112 and largest for NM 113, consistent with analyses of TEM images (please

refer to IUCLID section 4.1).

Executive summary

CSIRO, 2012 investigated the agglomeration/aggregation of the nanomaterials. Therefore a Philips XL30

field emission SEM was used for this study. The optimal spatial resolution of the microscope was from 2-

5 nm with varying accelerating voltage from 30 kV to 1 kV. Images of ZnO particles were acquired at an

accelerating voltage of 5 kV, a working distance of ≈ 10 mm, and a tilt angle 0°. An SEM metal stub was

covered with adhesive conducting tape and a small amount of “as received” ZnO powder (around 5 mg)

was sprinkled over the tape. The surface of the powder sample was flattened with spatula. Excess powder

was removed by gently tapping the stub on its side until a light coating of powder on the surface became

apparent. The nanoparticles were thinly sputtered with iridium using a Polaron SC570 sputter coater.

Sputtering was conducted under vacuum while passing gas was argon. The coating deposition time was

20 seconds at a plate current of 50 mA, giving a coating thickness of approximately 1 nm. SEM images

reveal that ZnO particles as powder without media were agglomerated. The particle size was smallest for

sample NM-112 and largest for NM-113, consistent with analyses of TEM images (please refer to

IUCLID section 4.1).


91


05_Australia_CSIRO_Agglomeration/Aggregation_DLS

Administrative Data




Rationale for

reliability incl.

deficiencies

The study was regarded as scientifically acceptable. DLS is mainly suitable for assessing

dispersed samples containing particles/agglomerates with narrow size distributions. Thus,

the study was regarded as reliable with restrictions.

Data source

Reference

Reference

type


source

Testing

laboratory

Report

no.

Owner

company

Company

study no.

Report

date

study

report

Casey,

Yin,

McCall

2012 Summary of

physicochemical

properties of

OECD Zinc

Oxide (ZnO)

samples - Wet

state-Dynamic

Light Scattering

(DLS)

Commonwealth

Scientific and

Industrial

Research

Organisation

2012-

03-05

Data access





Method

DLS


Measurements of hydrodynamic size were obtained using a Brookhaven particle size analyzer 90Plus

equipped with a 657 nm laser. Reference standards (Duke polystyrene latex, with a nominal size of 100

nm, and NIST RM8013 Au nanoparticles with a nominal size of 60 nm) were used to assess the

performance of the instrument. 10 mg as received ZnO particles were added to a measuring cuvette

containing 3 ml of deionised water. The cuvette was placed in an ultrasonic bath, ultrasonicated for 10

seconds and then shaken to ensure the particles were well dispersed before starting the dynamic light

scattering measurements. Each size distribution curve and correlation function curve that was generated

was based on 10 measurements. Experiments for each sample were performed in triplicate. The

temperature was maintained at 25oC. The cuvette was thoroughly washed with deionised water after each

experiment.

Data gathering

Instruments

Brookhaven particle size analyzer 90Plus equipped with a 657 nm laser

Calibration


92

Reference standards (Duke polystyrene latex, with a nominal size of 100 nm, and NIST RM8013 Au

nanoparticleswith a nominal size of 60 nm) were used to assess the performance of the instrument. The

measured hyrdrodyamic diameter for Duke polystyrene latex was 98 nm, and for NIST RM8013 Au

nanoparticles was 61 nm.

Test materials


yes

Reference material/nanomaterial and Sample identification number

Identifier Identity

reference material/nanomaterial Duke polystyrene latex, nominal size 100 nm

reference material/nanomaterial NIST RM8013 Au nanoparticles, nominal size 60 nm


Identifier Identity


Test material form

nanomaterial - dispersion


Remarks on results incl. tables and figures

Size distribution curves and correlation function curves for NM110, NM112 and NM113 are given in the

attached study report. DLS data were not obtained for sample NM111, as its surface coating makes it

hydrophobic and the sample could not be dispersed readily in water.


Attached full

study report

NM_110_ANNEX_A35_OECD ZnO Agglomeration Aggregation DLS CSIRO Australia.docx /

223.43 KB (application/octet-stream): ENV/JM/MONO(2015)15/ANN1


Conclusions

The mean hydrodynamic sizes for NM 110, NM 112 and NM 113 were determined to be 338 nm, 444 nm

and 466 nm respectively. These hydrodynamic diameters appeared to be independent of the primary

particle sizes, suggesting that particles from the ZnO samples were aggregated /agglomerated when

dispersed in deionized water.

Executive summary

CSIRO, 2012 determined the hydrodynamic size of the test items using Dynamic Light Scattering (DLS).

DLS data were not obtained for sample NM-111, as its surface coating made it hydrophobic and the

sample could not be dispersed readily in water. Measurements of hydrodynamic size were obtained using

a Brookhaven particle size analyser 90Plus equipped with a 657 nm laser. Reference standards (Duke

polystyrene latex, with a nominal size of 100 nm, and NIST RM8013 Au nanoparticles with a nominal

size of 60 nm) were used to assess the performance of the instrument. 10 mg as received ZnO particles

were added to a measuring cuvette containing 3 mL of deionised water. The cuvette was placed in an

ultrasonic bath, ultrasonicated for 10 seconds and then shaken to ensure the particles were well dispersed

before starting the dynamic light scattering measurements. Each size distribution curve and correlation

function curve that was generated was based on 10 measurements. Experiments for each sample were

performed in triplicate. The temperature was maintained at 25 °C. The cuvette was thoroughly washed

with deionised water after each experiment. The measured hydrodynamic diameter for Duke polystyrene


93

latex was 98 nm, and for NIST RM8013 Au nanoparticles was 61 nm. The mean hydrodynamic sizes for

NM-110, NM-112 and NM-113 were determined to be 338 nm, 444 nm and 466 nm respectively. These

hydrodynamic diameters appeared to be independent of the primary particle sizes, suggesting that

particles from the ZnO samples were aggregated /agglomerated when dispersed in deionized water.

4.25 Crystalline phase


05_Australia_CSIRO_Crystalline_Phase_XRD

Administrative Data




Rationale for

reliability incl.

deficiencies

The study was regarded as reliable without restrictions. XRD is fully ISO

standardized and all procedures and techniques were regarded to be compliant.

Data source

Reference

Reference

type


source

Testing

laboratory

Report

no.

Owner

company

Company

study no.

Report

date

study

report

Casey,

Yin,

McCall

2012 Summary of

physicochemical

properties of

OECD Zinc

Oxide (ZnO)

samples -

Crystalline

phase

Commonwealth

Scientific and

Industrial

Research

Organisation

2012-

03-05

Data access





Test guideline



Method

x-ray diffraction (XRD)


A suitable amount of particles (~0.5g) was packed in a plastic sample holder and a glass slide used to

flatten the surface. The crystallite phase was determined using a Bruker ASX-D8 X-Ray Diffractometer

(XRD) using Cu K(alpha) radiation. The operation current and voltage was 40mA and 40kV respectively.

The scan ranged from 5o to 85o with a step size of 0.02o and a scan speed of 0.40 second/step. The


94

aperture slit size directing the x-ray source was 0.2mm.

Data gathering

Instruments

Bruker ASX-D8 X-Ray Diffractometer (XRD) using Cu K(alpha) radiation

GLP compliance

no

Test materials


yes


Identifier Identity


EC number 215-222-5

IUPAC name oxozinc


Test material form

nanomaterial - powder



Code: NM110

Supplier: BASF SE





Code: NM111

Supplier: BASF SE





TM Code: NM112






Code: NM113






95


Crystalline Phase

Crystal System hexagonal (wurtzite zincite)



Attached full

study report NM_110_ANNEX_A36_OECD ZnO Crystalline Phase XRD CSIRO

Australia.docx / 244.34 KB (application/octet-stream):




Conclusions

XRD patterns for NM110, NM111, NM112 and NM113, indicating that the only detected phase was

hexagonal wurtzite zincite

Executive summary

CSIRO, 2012 investigated the crystallite phase of the nanomaterials. The crystalline phase was

determined using a Bruker ASX-D8 XRD using Cu K(alpha) radiation. The operation current and voltage

was 40 mA and 40 kV respectively. The scan ranged from 5° to 85° with a step size of 0.02° and a scan

speed of 0.40 second/step. The aperture slit size directing the x-ray source was 0.2 mm. XRD patterns of

the four ZnO samples are shown in the attached Figure and indicated that all ZnO nanomaterial samples

were in a hexagonal wurtzite zincite crystalline phase.


96


06_Australia_Deakin_Crystallinephase_XRD

Administrative Data


Study result

type

experimental result


Rationale for

reliability incl.

deficiencies

The study was regarded as reliable with restrictions. XRD secondary use was the analysis of

peak width to obtain crystallite size. Uncertainties grow above ~50nm (as is the case for

several samples, see TEM). It was concluded that this method was not ideally suited for

present samples.

Data source

Reference

Reference

type


source

Testing

laboratory

Report

no.

Owner

company

Company

study no.

Report

date

study

report

Tsuzuki,

Re

2012 Crystal phase

& crystallite

size

measurements

for OECD ZnO

samples

Deakin

University

GEELONG

VIC 3220

Australia

2012-

03-06

Data access





Test guideline



Method



Power: 40 kV, 30 mAX-rays: Cu-K-alphaSlit width: 2 00 mmScan step size: 0.02 degreeScan type:

continuousTime per step: 0.5 sec

Sampling

Dry powder (as-received) was packed in a plastic sample holder. The surface was flattened using a glass

slide

Data gathering

Instruments

Panalytical X’Pert Pro MRD


97

GLP compliance

no

Test materials


yes


Identifier Identity


EC number 215-222-5

IUPAC name oxozinc


Test material form




Code: NM110

Supplier: BASF SE





Code: NM111

Supplier: BASF SE





TM Code: NM112





Code: NM113


Surface coating: none- Substance type: Inorganic



Crystalline Phase

Common Name wuritzite

Crystal System hexagonal

Remarks on results incl. tables

Sample Crystallite size (nm)

NM110 24

NM111 21


98

NM112 21

NM113 27



Attached full

study report

NM_110_ANNEX_A37_OECD ZnO XRD Auatsralia Deakin.pdf / 85.42 KB




Conclusions

XRD patterns of the four ZnO samples are shown in the attached figure and indicated that all ZnO


99

nanomaterial samples were in a hexagonal wurtzite zincite crystalline phase.

Executive summary

The Australia Deakin University, 2012 investigated the crystalline phase of the test items. The crystallite

phase was determined using a X’Pert Pro MRD X-Ray Diffractometer (XRD) using Cu K(alpha)

radiation. The operation current and voltage was 40 mA and 40 kV respectively. The scan ranged from

10° to 80° with a step size of 0.02° and a scan speed of 0.50 second/step. The aperture slit size directing

the x-ray source was 200 nm. XRD patterns of the four ZnO samples are shown in the attached figure and

indicated that all ZnO nanomaterial samples were in a hexagonal wurtzite zincite crystalline phase.

Endpoint study record: WoE_2011-08-11_NPL for PROSPECT_XRD

Administrative Data

Purpose flag disregarded study



Rationale for

reliability incl.

deficiencies

The study does not provide information about the crystalline phase. In addition the study

has significant limitations regarding documentation and data evaluation. Thus, the

reliability of the study could not be assigned.

Data source

Reference

Reference

type


source

Testing

laboratory

Report

no.

Owner

company

Company

study no.

Report

date

study

report

National

Physical

Laboratory

2011 NPL for

PROSPECT

XRD

National

Physical

Laboratory,

Teddington,

Middlesex,

United

Kingdom

2011-

08-11

Data access




Cross-reference to same study

4.26 Crystallite and grain size


Test guideline



Method



100

Data gathering

Instruments

Siemens D5000 diffractometer

GLP compliance

no

Test materials


yes


Identifier Identity


EC number 215-222-5

IUPAC name oxozinc


Test material form

nanomaterial - aerosol



Code: NM110

Supplier: BASF SE





Code: NM111

Supplier: BASF SE





TM Code: NM112






Code: NM113






101


Crystalline Phase

Crystal System hexagonal

Space Group P63mc

Crystallographic planes a = 3.2493(38) Å, c = 5.2097(2) Å


Scherrer equation size data

Material Crystallite Diameter (nm)

NM-112 24.1

NM-111 33.8

NM-110 41.5

NM-113 41.5



This size data however should be interpreted with care. The technique is generally regarded to break

down for particle size, when the particle size is above 100nm. The size determination is also greatly

affected by the size distribution of the particles in the powder. As larger particles tend to swamp the size

contribution from the smaller size fraction. Therefore, the technique works best with particles under 100

nm and with a narrow size distribution.


Attached full

study report

NM_110_ANNEX_A39_XRD Preliminary Data from NPL.doc / 507.5 KB


4.26 Crystallite and grain size

Endpoint study record: Key_2012-03-

05_Australia_CSIRO_Crystallite_Size_XRD_Scherrer

Administrative Data




Rationale for

reliability incl.

deficiencies

The study was regarded as scientifically acceptable. Contributions to the widths of Xray

diffraction peaks from crystal strain and instrument effects are assumed to be zero. Using

XRD the average particle size can be estimated. Specific factors can influence the peak width,

such as microstrain, lattice defects and temperature factors. Due to the limited documentation

it was not possible to estimate the influence of these factors to the result. Thus, the study was



102

Data source

Reference

Reference

type


source

Testing

laboratory

Report

no.

Owner

company

Company

study no.

Report

date

study

report

Casey,

Yin,

McCall

2012 Summary of

physicochemical

properties of

OECD Zinc

Oxide (ZnO)

samples -

Comparison of

particle/crystallite

sizes as

determined using

XRD, TEM and

BET surface area

measurements

Commonwealth

Scientific and

Industrial

Research

Organisation

2012-

03-05

Data access





Test guideline



Method



D S-XRD was calculated using Scherrer’s formula. It should be noted that a variety of factors can

contribute to the width of a diffraction peak; besides crystallite size, the most important of these usually

are inhomogeneous strain and instrumental effects. When Scherrer’s formula is applied, it is assumed that

all of these other contributions to the peak width are zero.

Data gathering

Instruments

Bruker ASX-D8 X-Ray Diffractometer (XRD) using Cu K(alpha) radiation

GLP compliance

no

Test materials


Identifier Identity


EC number 215-222-5

IUPAC name oxozinc


103


Test material form

nanomaterial



Code: NM110

Supplier: BASF SE





Code: NM111

Supplier: BASF SE





TM Code: NM112






Code: NM113




- Physical state: solid powder, nano-for



Comparison of crystallite/particle sizes in samples NM110, NM111, NM112 and NM113, as determined

from XRD analyses, TEM images and BET surface area measurements.

DR-XRD

(nm)

DS-XRD

(nm)

DFeret TEM

(nm)

Dcirc TEM

(nm)

DSA BET

(nm) Average of all methods (nm)

NM110 113 41 70 ± 46 52 ± 33 74 77.5 ± 18

NM111 83 34 82 ± 45 60 ± 33 76 75.2 ± 7.6

NM112 43 25 30 ± 9 25 ± 7 37 33.75 ± 6.2

NM113 200 42 143 ± 47 113 ± 45 143 149.7 ± 25.0

All methods indicate that the particle sizes are in the same order as sizes provided by the suppliers. Based

on the average values from all methods, NM-112 is the smallest (~34 nm), NM-110 (78 nm) and NM111

are larger (75 nm), and NM-113 is the largest (150 nm). The difference between the measured and

supplier-supplied particle size is largest for NM113 (measured 150nm, compared with a supplier-supplied

value of 95% < 3.5µm and 50% <1µm) and smallest for NM112 (measured 34nm compared with a

supplier-supplied value of 30-50nm).Of the four samples analysed, NM112 has the smallest size

distribution, with particles likely to be single crystals.

The sizes of particles determined by different methods were generally in the order of


104

DR-XRD> DSA BET~ DFeretTEM> DCirc TEM> DS-XRD

DS-XRDwas smallest because Scherrer’s formula ignores the contributions of width broadening due to

strain and instrument effects. If these contributions are non-zero, the crystallite sizes could be larger than

those predicted by the Scherrer formula, as revealed by DR-XRD. The real states of particles can be



held together by strong chemical bonds that cannot be separated by conventional methods. DFeret

TEMprovides the most reliable assessment of aggregate size, because aggregates are not broken up by

ultrasonication during sample preparation whereas agglomerates can be. DCirc TEMprovides a smaller value

than DFeret TEMsince this methodtreats all particles as spheres; it is not the most suitable method for

analysing samples containing particles with a broad or multimodal size distribution. DSABETwas calculated

from measurements of specific surface area and assumed that all particles are non-porous and have the

same spherical shape and size.


Attached full study

report

NM_110_ANNEX_A21_OECD ZnO Crystallite Size CSIRO Australia.docx / 29.24 KB



Executive summary

CSIRO, 2012 investigated the crystallite size of the nanomaterials. Two methods were used to estimate

average crystallite size based on XRD results. Crystallite size of the test samples were calculated using

Scherrer’s formula (D S-XRD) and through Rietveld refinement of the diffraction data (D R-XRD). The

crystallite size of the test samples determined in two independent experiments calculated using Scherrer’s

formula were comparable for all nanomaterial samples. The crystallite sizes determined by XRD were in

the range of 24 nm (NM-112) to 42 nm. Both NM-110 and NM-113 have the same crystallite size of 42

nm. The average crystallite size determined by Rietveld refinement yielded larger crystallite sizes. This

result could be explained by the fact that the Scherrer’s formula ignores the contributions of width

broadening due to strain and instrument effects. If these contributions are non-zero, the crystallite sizes

could be larger than those predicted by the Scherrer formula, as revealed by Rietveld refinement of the

diffraction data (D-R XRD). These data were regarded as most reliable.

Endpoint study record: supporting_2010-07-02_NPL for PROSPECT_Crystallite

andgrain size_XRD

Administrative Data




Rationale for

reliability incl.

deficiencies

The study was regarded as scientifically acceptable. In general XRD is the recommended

method in the respective ECHA guidance. However, only limited documentation of the data

evaluation was provided in the full study report. Furthermore the procedure for subsampling

and data analysis and representation was not explained in detail. Thus, it was not possible to

assess the validity of the provided data as the sample preparation and the data analysis could

have a significant influence to the result Other factors can also influence the peak width, such

as microstrain, lattice defects and temperature factors. Based on the provided data neither

these influences could not be assessed nor was it investigated in the study. Thus, the study



105

Data source

Reference

Reference

type


source

Testing

laboratory

Report

no.

Owner

company

Company

study no.

Report

date

study

report

National

Physical

Laboratory

2010 X-ray

diffraction

analysis of 7

nanoparticle

samples –

direct

comparison

for

PROSPEcT

National

Physical

Laboratory,

Teddington,

Middlesex,

United

Kingdom

2010-

07-02

Data access





Test guideline



Method


Data gathering

GLP compliance

no

Test materials


Identifier Identity


EC number 215-222-5

IUPAC name oxozinc

other: OECD Sponsorship Programme NM110, NM111, NM112 , NM113

Test material form

nanomaterial



Code: NM110

Supplier: BASF SE





Code: NM111


106

Supplier: BASF SE





TM Code: NM112






Code: NM113







Scherrer equation size data

Material Crystallite Diameter

(nm)

NM-112 24.1

NM-111 33.8

NM-110 41.5

NM-113 41.5

This size data however should be interpreted with care.The technique is generally regarded to break down

for particle size, when the particle size is above 100nm. The size determination is also greatly affected by

the size distribution of the particles in the powder. As larger particles tend to swamp the size contribution

from the smaller size fraction. Therefore, the technique works best with particles under 100nm and with a

narrow size distribution.


Attached full study report NM_110_ANNEX_A02_FINALINTERIM REPORT2010corrected.pdf / 1.05 MB


Attached full study report NM_110_ANNEX_A39_XRD Preliminary Data from NPL.doc / 507.5 KB


Attached full study report NM_110_ANNEX_A38_XRD.doc / 1.21 MB (application/octet-stream):



Conclusions

The crystallite size as determined from the XRD patterns using the Scherrer equation was determined to

be 41.5 nm for NM-110, 33.8 nm for NM-111, 24.1 nm for NM-112 and 41.5 nm for NM-113.

Executive summary

NPL, 2011, investigated the crystallite size of the nanomaterials. To characterise the zinc oxide (ZnO)

nanomaterials XRD traces were obtained using a Siemens D5000 diffractometer. Crystallite size was


107

determined using Scherrer equation. The pattern matches were performed using this software linked to the

ICDD (International Centre for Diffraction Data) PDF (Powder Diffraction File) database 2005). The

crystallite size as determined from the XRD patterns using the Scherrer equation was determined to be

41.5 nm for NM-110, 33.8 nm for NM-111, 24.1 nm for NM-112 and 41.5 nm for NM-113.

Endpoint study record: supporting_2012-03-06_Australia_Deakin_Crystallite

and grain size_XRD

Administrative Data




Rationale for

reliability incl.

deficiencies

Limited number of replicate was used. In general XRD is the recommended method in the

respective ECHA guidance. However, only results were provided. Without any

documentation of the used method and data evaluation the data could not be assessed. Thus,

the reliability of the study could not be assigned.

Data source

Reference

Reference

type


source

Testing

laboratory

Report

no.

Owner

company

Company

study no.

Report

date

study

report

Tsuzuki,

He

2012 Crystal phase

& crystallite

size

measurements

for OECD ZnO

samples

Deakin

University

GEELONG

VIC 3220

Australia

2012-

03-01

Data access





Test guideline



Method



XRD, Williamson-Hall Method

Sampling

Dry powder (as-received) was packed in a plastic sample holder. The surface was flattened using a glass

slide.

Data gathering

Instruments

Panalytical X’Pert Pro MRDMeasurement parameters:Power: 40 kV, 30 mAX-rays: Cu-K-alphaSlit


108

width: 2 00 mmScan step size: 0.02 degreeScan type: continuousTime per step: 0.5 sec

GLP compliance

no

Test materials


Identifier Identity


EC number 215-222-5

IUPAC name oxozinc


Test material form

nanomaterial



Code: NM110

Supplier: BASF SE





Code: NM111

Supplier: BASF SE





TM Code: NM112






Code: NM113







Attached full study

report

NM_110_ANNEX_A37_OECD ZnO XRD Auatsralia Deakin.pdf / 85.42 KB



Sample Crystallite size (nm)

NM110 24


109

NM111 21

NM112 21

NM113 27

4.27 Aspect ratio/shape

4.28 Specific surface area

Endpoint study record: WoE_2010-06-25_NPL for PROSPECT_specific

surfacearea_BET

Administrative Data




Rationale for

reliability incl.

deficiencies


the data evaluation was provided in the full study report. Furthermore the procedure for

subsampling and data analysis and representation was not explained in detail. There are

specific sample splitting practices to be followed (cf. ISO 14488:2007). It was not indicated in

the study report that these sampling procedures were followed. Thus, there is a doubt that the

sample could be considered to be representative of the whole sample with a defined

confidence level. Thus, it was not possible to assess the validity of the provided data as the

sample preparation and the data analysis could have a significant influence to the result. The

most requirements as outlined in ECHA guidance were fulfilled. Thus, the results were


Data source

Reference

Reference

type


source

Testing

laboratory

Report

no.

Owner

company

Company

study no.

Report

date

study

report

National

Physical

Laboratory

2010 National

Physical

Laboratory,

Teddington,

Middlesex,

United Kingdom

2010-

06-25

Data access





Test guideline


according to ISO 9277:2010 Determination of the specific surface area of solids by gas adsortion -

BET method

no data

Methods

BET (BET (Brunauer, Emmett and Teller) Multipoint Specific Surface Area determination)


110


BET surface area measurements were determined using Autosorb-1 (Quantachrome Instruments). The

Autosorb-1 was calibrated using a quartz rod of a known volume, which is traceable to NIST. This

calibration was then further checked using two BAM certified reference materials: BAM-PM-102

(nominal SSA 5.41m2g-1) and BAM-PM-104 (nominal SSA 79.8 m2g-1). These two reference materials

allowed the range of SSA of the nanoparticles to be encompassed with known specific surface area

materials, thus adding confidence to the measurements. Surface area measurements were acquired using

an 11-point BET gas adsorption method, with nitrogen as the adsorbate. Prior to analysis, the powdered

sample was transferred to a sample bulb, then sealed and subsequently de-gassed overnight at 300 °C

under a high vacuum and subsequently weighed on an analytical balance in order to determine the sample

mass after the degassing step.


Each sample was degassed overnight at 300°C and under vacuum. An 11-point BET measurement using

nitrogen was then taken. This measurement was carried out on two separate aliquots of each powder.

Data gathering

Instruments

Autosorb-AS1

Calibration

Manifold calibration using a calibrated quartz rodSSA calibration carried out using BAM CRM BAM-

PM-102 and BAM CRM BAM-PM-104

GLP compliance

no

Test materials


yes


Identifier Identity

reference material/nanomaterial BAM-PM-102



Identifier Identity


EC number 215-222-5

IUPAC name oxozinc


Test material form




Code: NM110

Supplier: BASF SE



111




Code: NM111

Supplier: BASF SE





TM Code: NM112






Code: NM113







summary table Material 1stAliquot BET SSA (m2/g) 2ndAliquot BET SSA (m2/g) Mean BET SSA (m2/g)

NM-110 12.47±0.59 12.37±0.38 12.4±0.6

NM-111 15.10±0.60 15.13±0.57 15.1±0.6

NM-113 27.17±1.15 27.18±0.82 27.2±1.2

NM-113 6.14±0.29 6.29±0.20 6.2±0.3


Specific Surface Area

Specific Surface Area 12.4 m²/g

Remarks NM-110

Specific Surface Area m²/g

Remarks NM-111


Remarks NM-112


Remarks NM-113


Attached full

study report

NM_110_ANNEX_A40_NPL BETpreliminary data.doc / 652.5 KB


Attached full

study report

NM_110_ANNEX_A02_FINALINTERIM REPORT2010corrected.pdf / 1.05 MB


Attached full

study report

NM_110_ANNEX_A41_BET Interim report data.doc / 32.5 KB (application/octet-



112


Conclusions

The specific surface area was determined to be 12.4 ± 0.6 m2/g for NM-110, 15.1 ± 0.6 m2/g for NM-11,

27.92 ± 1.2 m2/g for NM-112 and 6.2 ± 0.3 m2/g for NM-113.

Executive summary

The specific surface area of the NMs was investigated NPL, 2010 using BET Multipoint Specific Surface

Area (SSA) determination. BET surface area measurements were determined using Autosorb-1

(Quantachrome Instruments). The Autosorb-1 was calibrated using a quartz rod of a known volume,

which is traceable to NIST. This calibration was then further checked using two BAM certified reference

materials: BAM-PM-102 (nominal SSA 5.41 m2/g) and BAM-PM-104 (nominal SSA 79.8 m²/g). These

two reference materials allowed the range of SSA of the nanoparticles to be encompassed with known

specific surface area materials, thus adding confidence to the measurements. Surface area measurements

were acquired using an 11-point BET gas adsorption method, with nitrogen as the adsorbate. Prior to

analysis, the powdered sample was transferred to a sample bulb, then sealed and subsequently de-gassed

overnight at 300 °C under a high vacuum and subsequently weighed on an analytical balance in order to

determine the sample mass after the degassing step. The specific surface area was determined to be 12.4 ±

0.6 m2/g for NM-110, 15.1 ± 0.6 m2/g for NM-11, 27.92 ± 1.2 m2/g for NM-112 and 6.2 ± 0.3 m2/g for

NM-113.


05_Australia_CSIRO_Specific_Surface_Area_BET

Administrative Data




Rationale for

reliability incl.

deficiencies





the study report that these sampling procedures were followed and if reference standards were

used. Thus, there is a doubt that the sample could be considered to be representative. It was

not possible to assess the validity of the provided data as the sample preparation and the data

analysis have a significant influence to the result. However, other requirements as outlined in

ECHA guidance were fulfilled.

Data source

Reference

Reference

type


source

Testing

laboratory

Report

no.

Owner

company

Company

study no.

Report

date

study

report

Casy,

Yin,

McCAll

2012 Summary of

physicochemical

properties of

OECD Zinc

Oxide (ZnO)

samples - BET

Surface Area

Commonwealth

Scientific and

Industrial

Research

Organisation

2012-

03-05

Data access



113




Test guideline


according

to

ISO 9277:2010 Determination of the specific surface area of solids by gas adsortion -

BET method

no data

Methods

BET


BET Surface Area was determined using a Micromeritics Tristar II 3020 which uses physical adsorption

and capillary condensation principles to obtain information about the surface area and porosity of a solid

material. Prior to analysis, the powdered sample was transferred to a sample bulb, which was then sealed

and de-gassed overnight at 300C under high vacuum, and subsequently weighed on an analytical balance

in order to determine the sample mass after the degassing step. Then the sample tube containing degassed

sample was cooled to 77K (the temperature of liquid nitrogen) and exposed to the analysis gas (nitrogen)

at 11 controlled pressures. With each incremental pressure increase, the number of gas molecules

adsorbed on the surface increases. The equilibrated pressure (P) is compared to the saturation pressure

(P0) and their relative pressure ratio (P/P0) is recorded along with the quantity of gas adsorbed by the

sample at each equilibrated pressure. Experiments were done in duplicate, on different days.

Data gathering

Instruments

Micromeritics Tristar II 3020

GLP compliance

no

Test materials


yes


Identifier Identity


EC number 215-222-5

IUPAC name oxozinc


Test material form








114



Code: NM111

Supplier: BASF SE





TM Code: NM112






Code: NM113







The specific surface areas of the ZnO samples, obtained using the Micromeritics instrument, are shown in

below. The values(±SD)were averaged from data obtained from duplicate experiments performed on

different days. The values are consistent with TEM data on particle size.

Specific surface areas of the ZnO samples, NM110, NM111, NM112 and NM113, obtained by the BET

gas adsorption technique and using a Micromeritics instrument. The data represent the means of values

(±SD) acquired on different days.

Sample Name Mean BET SSA (m2/g)

NM-110 11.76 ± 0.55

NM-111 13.75 ± 0.23

NM-112 27.25 ± 0.5

NM-113 5.78 ± 0.05




Remarks NM-110


Remarks NM-111


Remarks NM-112


Remarks NM-113


Attached full

study report

NM_110_ANNEX_A42_OECD ZnO Specific Surface Area BET CSIRO Australia.docx /

23.65 KB (application/octet-stream): ENV/JM/MONO(2015)15/ANN1


115


Conclusions

The specific surface area was determined to be 11.76 ± 0.55 m2/g for NM-110, 13.75 ± 0.23 m2/g for

NM-11, 27.25 ± 0.5 m2/g for NM-112 and 5.78 ± 0.05 m2/g for NM-113.

Executive summary

CSIRO, 2012 investigated the specific surface area of the nanomaterials. BET Surface Area was

determined using a Micromeritics Tristar II 3020 which uses physical adsorption and capillary

condensation principles to obtain information about the surface area and porosity of a solid material. Prior

to analysis, the powdered sample was transferred to a sample bulb, which was then sealed and de-gassed

overnight at 300 °C under a high vacuum, and subsequently weighed on an analytical balance in order to

determine the sample mass after the degassing step. Then the sample tube containing degassed sample

was cooled to 77 K (the temperature of liquid nitrogen) and exposed to the analysis gas (nitrogen) at

controlled pressures. With each incremental pressure increase, the number of gas molecules adsorbed on

the surface increases. The equilibrated pressure (P) was compared to the saturation pressure (P0) and their

relative pressure ratio (P/P0) was recorded along with the quantity of gas adsorbed by the sample at each

equilibrated pressure. Experiments were done in duplicate, on different days. The values (±SD) were

averaged from data obtained from duplicate experiments performed on different days. The specific

surface areas of the ZnO samples, NM 110, NM 111, NM 112 and NM 113, obtained by the BET gas

adsorption technique was determined to be 11.76 ± 0.55 m2/g for NM-110, 13.75 ± 0.23 m2/g for NM-

11, 27.25 ± 0.5 m2/g for NM-112 and 5.78 ± 0.05 m2/g for NM-113.

Endpoint study record: WoE_2011-12-19_NPL for PROSPECT_BET analysis

Administrative Data




Rationale for

reliability incl.

deficiencies





the study report that these sampling procedures were followed. Thus, there is a doubt that the

sample could be considered to be representative. Thus, it was not possible to assess the

validity of the provided data as the sample preparation and the data analysis have a significant

influence to the result. However, other requirements as outlined in ECHA guidance were

fulfilled.

Data source

Reference

Reference

type


source

Testing

laboratory

Report

no.

Owner

company

Company

study no.

Report date

study

report

MCA

Cambridge

2011 BET

measure

ments

MCA

Cambridge

010611

MCACa

mbridge

2011-12-19

Data access





116


Test guideline



Methods

BET


A Micromeritics TriStar II (3020) was used for the collection of nitrogen adsorption / desorption isotherm

data up to a saturation pressure of approximately 0.995 P/Po. The analysis was typically conducted to

measure 45 adsorption relative pressure points and 23 desorption relative pressure points. Samples were

outgassed overnight in vacuo at 300 degC using a Micromeritics VacPrep apparatus prior to analysis. In

order to indicate any possible microporous nature of the materials additional relative pressure data were

also collected at pressures lower than the usual starting point for analyses using this instrument. These

were in the approximate range 0.005 to 0.01 P/Po. Whilst the data reduction methods available are

unsuitable for application to the micropore range the characteristic shape of the adsorption isotherm at

these low partial pressures would provide a good indication of the presence of micropores in the sample

material. The sample tube dead space was measured for each analysis using helium (CP grade) thus

providing warm and cold freespace values. Samples requiring only BET surface area analysis were

analysed using the same equipment with the application ofthe same freespace measurement technique.

BET surface area was calculated using partial pressures in the nominal range 0.07 to 0.25.Data Reduction.

The samples for which the full adsorption/desorption isotherms were analysed graphical plots of the

adsorption/desorption isotherm and BET surface area transform plot together with tabulated data for each

are presented.

Data gathering

Instruments

Micromeritics TriStar II

Calibration

BAM standard Alumina BAM-pm-104

GLP compliance

no

Test materials


yes


Identifier Identity

reference material/nanomaterial BAM standard Alumina BAM-pm-104


Identifier Identity


EC number 215-222-5

IUPAC name oxozinc



117

Test material form




Code: NM110

Supplier: BASF SE





Code: NM111

Supplier: BASF SE





TM Code: NM112






Code: NM113







Sample Sample code Multipointsurface area by Nitrogen (m2g

-1) BETC-value

NM112 ZA250#30#ICP 27.15 ± 0.0199 252

NM111 ZB250#64#ICP 14.62± 0.0483 21

NM113 ZrA250#33#ICP 4.33 ± 0.0011 198

NM110 Zc250#37#ICP 11.91± 0.0041 234



Attached full study

report

NM_110_ANNEX_A43_010611MCACambridge.doc / 93 KB (application/octet-



The BET C-value:

The C-value is part of the characterisation – regardless of the theoretical explanations (relationship to the

heat of adsorption, residence time of the adsorbate molecules, etc.) the best practical aspect is the

relationship to the intercept on the BET plot – the higher the C-value the smaller the intercept. Also, a


118

slight change in the BET slope can lead to a significant change in the intercept (and therefore the C-value)

with negligible change in the BET surface area.


Conclusions

The specific surface area was determined to be 11.91± 0.0041m2/g for NM-110, 14.62± 0.0483 m2/g for

NM-11, 27.15 ± 0.0199 m2/g for NM-112 and 4.33 ± 0.0011 m2/g for NM-113.

Executive summary

MCA Cambridge, 2011 investigated the specific surface area of the nanomaterials. A Micromeritics

TriStar II (3020) was used for the collection of nitrogen adsorption / desorption isotherm data up to a

saturation pressure of approximately 0.995 P/P0. The analysis was typically conducted to measure 45

adsorption relative pressure points and 23 desorption relative pressure points. Samples were outgassed

overnight in vacuum at 300 °C using a Micromeritics VacPrep apparatus prior to analysis. In order to

indicate any possible microporous nature of the materials additional relative pressure data were also

collected at pressures lower than the usual starting point for analyses using this instrument. These were in

the approximate range 0.005 to 0.01 P/P0. Whilst the data reduction methods available are unsuitable for

application to the micropore range the characteristic shape of the adsorption isotherm at these low partial

pressures would provide a good indication of the presence of micropores in the sample material. The

sample tube dead space was measured for each analysis using helium (CP grade) thus providing warm

and cold freespace values. BET surface area was calculated using partial pressures in the nominal range

0.07 to 0.25. The results are presented in Table 14. The specific surface area was determined to be 11.91±

0.0041m2/g for NM-110, 14.62± 0.0483 m2/g for NM-11, 27.15 ± 0.0199 m2/g for NM-112 and 4.33 ±

0.0011 m2/g for NM-113.


PROSPECT_Homogenity_NM110_BET

Administrative Data




Rationale for

reliability incl.

deficiencies

The study was regarded as scientifically acceptable. The study was conducted under non-

GLP conditions. BET is established and standardized. Certified reference materials were

used and reported. Results were regarded as plausible and fit with TEM data. Thus, the study


Data source

Reference

Reference

type


source

Testing

laboratory

Report

no.

Owner

company

Company

study no.

Report

date

study

report

National

Physical

Laboratory,

Teddington

2011 National

Physical

Laboratory,

Teddington,

Middlesex,

United

Kingdom

2011-

11-25

Data access



119




Test guideline



Methods

BET


A Micromeritics TriStar II (3020) was used for the collection of nitrogen adsorption/desorption isotherm

data up to a saturation pressure of approximately 0.995 P/Po. The analysis was typically conducted to


outgassed overnight in vacuo at 300 degC using a Micromeritics VacPrep apparatus prior to analysis. In



were in the approximate range 0.005 to 0.01 P/Po. The sample tube dead space was measured for each

analysis using helium (CP grade) thus providing warm and cold freespace values. Samples requiring only

BET surface area analysis were analysed using the same equipment with the application of the same

freespace measurement technique. BET surface area was calculated using partial pressures in the nominal

range 0.07 to 0.25.

Data gathering

Instruments

Micromeritics TriStar II (3020)

GLP compliance

no

Test materials


yes


Identifier Identity




Identifier Identity


EC number 215-222-5

IUPAC name oxozinc

other: OECD sponsorship programme NM110, NM111


120

Test material form




Code: NM110

Supplier: BASF SE





Code: NM111

Supplier: BASF SE







Remarks NM-111


Remarks NM110


Attached full study

report

NM_110_ANNEX_A05_030211BET.doc / 69 KB (application/octet-stream):



Reference material employed BAM Replicate number SSA(m²/g) BET C-value

BAM Standard - Alumina BAM-PM-104 79.72± 0.2934 125

BAM-PM-102 5.43±0.0057 151

Sample Mean SSA(m²/g) Mean BET C-value

NM 111 ZnO Z-Cote HP1 15.41± 0.2005 21.6

NM 110 ZnO Z-Cote 11.96±0.0665 138.8

CHARACTERISATION- NM110 – ZINC OXIDE NM-110 2 g (+/-) 10 mg

Stub name Replicate number SSA (m²/g) BET C-value*

0979 rep1 11.92 ± 0.0102 140

rep2 11.93 ± 0.0075 134

rep3 11.94 ± 0.0083 138

0599 rep1 12.09 ± 0.0062 123

rep2 12.14 ± 0.0121 111

rep3 11.99 ± 0.0107 116

0945 rep1 11.97 ± 0.0079 131

rep2 12.02 ± 0.0098 125

rep3 11.98 ± 0.0087 127

4779 rep1 11.88 ± 0.0059 136


121

rep2 11.99 ± 0.0043 159

rep3 11.95 ± 0.0090 130

4410 rep1 11.89 ± 0.0094 135

rep2 11.96 ± 0.0093 130

rep3 11.91 ± 0.0085 135

3911 rep1 11.92 ± 0.0070 181

rep2 11.93 ± 0.0039 170

rep3 11.93 ± 0.0076 178

CHARACTERISATION- NM 111 – ZINC OXIDE – NM-111 ; 2 g (+/-) 10 mg

Stub name Replicate number SSA (m²/g) BET C-value*

0803 rep1 15.72 ± 0.0483 21

rep2 15.66 ± 0.0413 21

rep3 15.75 ± 0.0443 22

1951 rep1 15.52 ± 0.0466 22

rep2 15.47 ± 0.0437 22

rep3 15.47 ± 0.0472 22

1455 rep1 15.47 ± 0.0462 22

rep2 15.42 ± 0.0415 22

rep3 15.43 ± 0.0493 22

2017 rep1 15.48 ± 0.0479 22

rep2 15.44 ± 0.0454 22

rep3 15.43 ± 0.0479 22

2100 rep1 15.14 ± 0.0485 21

rep2 15.07 ± 0.0437 21

rep3 15.06 ± 0.0428 21

4414 rep1 15.30 ± 0.0467 21

rep2 15.21 ± 0.0459 21

rep3 15.28 ± 0.0414 21


Conclusions

The results for both samples imply the samples are homogenous, all the results for the repetitions of each

stub have high repeatability and are very close to the mean SSA results for NM111 and NM110: 15.41 ±

0.2005 and 11.96 ± 0.0665 m²/g respectively.

Executive summary

The specific surface area of the NMs was investigated in a study conducted by NPL, 2010 using BET

Multipoint Specific Surface Area (SSA) determination. A Micromeritics TriStar II (3020) was used for

the collection of nitrogen adsorption/desorption isotherm data up to a saturation pressure of

approximately 0.995 P/Po. The analysis was typically conducted to measure 45 adsorption relative

pressure points and 23 desorption relative pressure points. Samples were outgassed overnight in vacuo at

300 degC using a Micromeritics VacPrep apparatus prior to analysis. In order to indicate any possible

microporous nature of the materials additional relative pressure data were also collected at pressures

lower than the usual starting point for analyses using this instrument. These were in the approximate

range 0.005 to 0.01 P/Po. The sample tube dead space was measured for each analysis using helium (CP

grade) thus providing warm and cold freespace values. Samples requiring only BET surface area analysis

were analysed using the same equipment with the application of the same freespace measurement


122

technique. BET surface area was calculated using partial pressures in the nominal range 0.07 to 0.25.

The results for both samples imply the samples are homogenous, all the results for the repetitions of each

stub have high repeatability and are very close to the mean SSA results for NM111 and NM110: 15.41 ±

0.2005 and 11.96 ± 0.0665 m²/g respectively.

4.29 Zeta potential

Endpoint study record: WoE_2010-12-03 NPL for PROSPECT Zeta potential

Administrative Data



Rationale for

reliability incl.

deficiencies


the data evaluation was provided in the full study reports. Furthermore the procedure for

subsampling and data analysis was not explained in detail. There were some uncertainties

because size was used to evaluate the electroporetic mobility in terms of zeta potential, and

size was not well determined in this equipment. However, these deficiencies were regarded to

have a minor impact on the obtained results.

Data source

Reference

Reference

type


source

Testing

laboratory

Report

no.

Owner

company

Company

study no.

Report

date

study

report

National

Physical

Laboratory

2010 Zeta-

potential

National

Physical

Laboratory,

Teddington,

Middlesex,

United

Kingdom

2010-

12-03

Data access





Test guideline



Methods

Laser-Doppler (Laser-Doppler-Electrophoresi)


Electrophoretic measurements were obtained using a Zetasizer Nano ZS (Malvern Instruments, UK)

equipped with a 633 nm laser. The reference standard (DTS1230, zeta-potential standard from Malvern)

was used to qualify the performance of the instrument. Sample preparation involved filling of a

disposable capillary cell (DTS1060, Malvern). Prior to their use, these cells were thoroughly cleaned with

ethanol and de-ionised water, as recommended by the instrument vendor. For analysis, the individual cell


123

was filled with the appropriate sample and flushed before re-filling; measurement was carried out on the

second filling Malvern Instrument’s Dispersion Technology software (Version 4.0) was used for data

analysis and zeta-potential values were estimated from the measured electrophoretic mobility data using

the Smoluchowski equation

Data gathering

GLP compliance

no

Test materials


yes


Identifier Identity


EC number 215-222-5

IUPAC name oxozinc


Test material form




Code: NM110

Supplier: BASF SE





Code: NM111

Supplier: BASF SE





TM Code: NM112





Code: NM113






124


Zeta Potential

Zeta Potential 24.3 mV

Standard Deviation 0.4 mV

in medium Deionized water

Remarks NM-110




Remarks NM-112




Remarks NM-113



in medium Deionized water + 5 mM NaCl

Remarks NM-110




Remarks NM-112




Remarks NM-113



in medium Fish medium

Remarks NM-110




Remarks NM-112




Remarks NM-113


The mean values of zeta-potential (of six replicates) for different PROSPEcT nanomaterials dispersed in

various media at a concentration of 50 mg/L.;*DI water + 5 mM NaCl - this medium was employed to

compare with the DI results when in the presence of inert background electrolyte. Values are the mean

and ± 1 SD of six replicates.

Nanomateri

al Reference

Sample


DI

water

(mV)

DI

water

+ 5mM

NaCl*

(mV)

Fish

mediu

m

(mV)

Seawate

r (mV)

Daphni

a

mediu

m (mV)

NM-110

E. Z-

COTE

ZnO

BASF,

Germany

ZC250#37#IC

P

24.3±0.

4

20.8±0.

8

10.8±0.

1 N/A 1.3± 0.2

NM-112 B. Microniser ZA250#30#IC 24.6 25.2±0. 12.4±0. N/A 4.9± 0.2


125

Nanosu

n ZnO

s, Australia P ±0.4 6 3

NM-113

D.

Micron

ZnO

Sigma

Aldrich,

UK

ZrA250#33#IC

P

20.2±0.

4

13.9±0.

6 4.4±0.4 N/A

-4.6±

0.4

The measured zeta-potential values for the PROSPEcT NMs (50 mg/L) are summarised in the table

above. Results show that zeta-potential values of NMs when dispersed in seawater cannot be successfully

measured (due to high conductivity) and thus displayed as N/A on the table; such unsuccessful

measurements were reported in the corresponding “quality report” at the end of the measurement. In

general, results indicate high zeta-potential values for NMs that are dispersed either in DI water (or DI

water + 5 mM NaCl), and thus confer stability in such media. Values of zeta-potential measured were

lower when the NMs were dispersed in an ecotox media indicating much poorer dispersion stability in

such media.



Attached full study

report

NM_110_ANNEX_A66_ZETA potential ceria and zno.doc / 37 KB (application/octet-



Executive summary

NPL, 2010 investigated the surface charge (zeta potential) of the NMs. Electrophoretic measurements

were obtained using a Zetasizer Nano ZS (Malvern Instruments, UK) equipped with a 633 nm laser. The

reference standard (DTS1230, zeta-potential standard from Malvern) was used to qualify the performance

of the instrument. Results show that zeta-potential values of NMs when dispersed in seawater cannot be

successfully measured (due to high conductivity). Results indicated high zeta-potential values for NMs

that were dispersed either in deionized water (or deionized water + 5 mM NaCl), and thus confer stability

in such media. Results showed values of zeta-potential measured were lower when the NMs were

dispersed in an ecotoxicology media indicating much poorer dispersion stability in such media.

Endpoint study record: WoE_2012-03-05_Australia_CSIRO_Zeta_Potential

Administrative Data




Rationale for

reliability incl.

deficiencies

Limited documentation of the data evaluation was provided in the full study reports.

Furthermore the procedure for subsampling and data analysis was not explained in detail.


could have a significant influence to the result. The provided data are not sufficient to finally

conclude on the zeta potential of the 4 NMs or to identify differences or similarities. The

concentration used was excessive and the results were regarded to be not self-consistent (e.g.

the pH-dependence for NM111), and were not confirmed by other labs (should see positive

charge at neutral pH).


126

Data source

Reference

Reference

type


source

Testing

laboratory

Report

no.

Owner

company

Company

study no.

Report

date

study

report

Casey,

Yin,

McCall

2012 Surface

charge

(zeta

potential)

Commonwealth

Scientific and

Industrial

Research

Organisation

2012-

03-05

Data access





Test guideline



Methods

Laser-Doppler (Laser-Doppler-Electrophoresi)


Zeta potentials of samples were determined at different pH values (pH=2, 4, 6, 8 and 10) using a

Brookhaven particle size analyser 90Plus equipped with a 657 nm laser. 10 mg “as-received” ZnO sample

was dispersed in a cuvette containing 3 ml DI water, and the pH was adjusted by adding either 0.1M HCl

or 0.1M NaOH. The cuvette was placed in an ultrasonic bath for 10 seconds and then shaken manually to

ensure good dispersion of particles in the sample. The electrode was inserted into the dispersion and the

Zeta potential at each pH was measured 5 times and an average was determined. The temperature of all

measurements was maintained at 25 °C. The cuvette was thoroughly washed with deionised water after

each measurement.

Data gathering

Instruments

Brookhaven particle size analyser 90Plus equipped with a 657 nm laser

GLP compliance

no

Test materials


yes


Identifier Identity


EC number 215-222-5


127

IUPAC name oxozinc


Test material form









Code: NM111

Supplier: BASF SE





TM Code: NM112






Code: NM113






Zeta Potential

Zeta Potential -5.79 mV


at pH (give value) 6.2

in medium DI water

Remarks NM110




in medium DI water

Remarks NM111


Standard Deviation mV


in medium DI water

Remarks NM112





128

in medium DI water

Remarks NM113

Isoelectric Point

Isoelectric Point 3.9

in medium deionised water

Remarks NM110



Remarks NM111



Remarks NM112


in medium deionised

water

Remarks NM113


The table and the figure show the relationship between zeta potential and pH for the four ZnO samples

tested. The pH value at which the net surface charge is zero is called the isoelectrical point (IEP). The

IEPs are 3.9 for NM110, 6.5 for NM112 and 5.1 for NM113. IEPs in the range 4-6 are consistent with the

dissociation of water to H+ and OH- on the particle’s surface and, where there is no surface coating, the

IEP will be due solely to this dissociation. This therefore suggests that there is no specific surface coating

on NM110, NM112 and NM113.

The IEP of NM111 is nominally 2.7, based on the data acquired. This IEP, which is significantly lower

than IEPs determined for the uncoated NM110, NM112 and NM113, is suggestive of a different surface

reaction (other than water dissociation) occurring on the coated NM111. However, NM111 is

hydrophobic and observed to be very difficult to disperse in aqueous solutions; it is difficult to reconcile

this observation with measurements of large zeta potential in water at most pHs. One possible explanation

may be that the zeta potential data pertain to a small portion of the sample that is able to disperse.

Therefore these data on zeta potential for NM111 should not be considered as representative of the

sample, unless supported by other evidence. Table Zeta potentials for ZnO samples dispersed in DI water where the pH was adjusted by adding either 0.1M

HCl or 0.1M NaOH.

pH NM110

Zeta

potential

mV

NM110

SD

NM111

Zeta

potential

mV

NM111

SD

NM112

Zeta

potential

mV

NM112

SD

NM113

Zeta

potential

mV

NM113

SD

2.10 25.04 1.84 14.36 3.01 24.04 1.91 16.94 2.74

4.00 -1.50 0.6 -33.67 2.76 10.20 0.92 5.94 3.1

6.20 -5.79 0.61 -26.78 1.77 3.74 0.56 -5.51 0.72

8.10 -21.63 0.82 -28.20 1.5 -22.00 3.45 -13.50 0.76

10.00 -31.45 0.48 -19.25 1.06 -33.34 0.62 -37.38 1.25



Attached full

study report

NM_110_ANNEX_A67_OECD ZnO Zeta Potential CSIRO Australia.docx / 57.33 KB



129



Executive summary

CSIRO, 2012 investigated the Surface charge (zeta potential) of the nanomaterials. Zeta potentials of the

samples were determined at different pH values (pH=2, 4, 6, 8 and 10) using a Brookhaven particle size

analyser 90Plus equipped with a 657 nm laser. 10 mg. ZnO nano samples were dispersed in a cuvette

containing 3 mL deionized water, and the pH was adjusted by adding either 0.1 M HCl or 0.1 M NaOH.

The cuvette was placed in an ultrasonic bath for 10 seconds and then shaken manually to ensure good

dispersion of particles in the sample. The electrode was inserted into the dispersion and the Zeta potential

at each pH was measured 5 times and an average was determined. The temperature of all measurements

was maintained at 25 °C. The result showed the relationship between zeta potential and pH for the four

ZnO samples tested. The isoelectrical point (IEP) was determined to be 3.9 for NM-110, 6.5 for NM-112

and 5.1 for NM-113. IEPs in the range 4-6 are consistent with the dissociation of water to H+ and OH- on

the particle’s surface and, where there is no surface coating, the IEP will be due solely to this dissociation.

This therefore it was concluded that there is no specific surface coating on NM-110, NM-112 and

NM-113. The IEP of NM-111 was determined to be 2.7 which was significantly lower than IEPs

determined for the uncoated test samples. NM-111 is hydrophobic and observed to be very difficult to

disperse in aqueous solutions; it is difficult to reconcile this observation with measurements of large zeta

potential in water at most pHs. One possible explanation may be that the zeta potential data pertain to a

small portion of the sample that is able to disperse. Therefore these data on zeta potential for NM-111

should not be considered as representative.


130

4.30 Surface chemistry


05_Australia_CSIRO_Surface_Chemistry_XPS

Administrative Data




Rationale for

reliability incl.

deficiencies

XPS is sensitive to contaminantsThe used XPS method is recommended in the OECD

guideline for the determination of the surface chemistry. However, only limited

documentation of the data evaluation was provided in the full study report. As sample

preparation and the data analysis have a significant influence on the result there are some

concerns on the validity of the data.

Data source

Reference

Reference

type


source

Testing

laboratory

Report

no.

Owner

company

Company

study no.

Report

date

study

report

Casey,

Yin,

McCall

2012 Summary of

physicochemical

properties of

OECD Zinc

Oxide (ZnO)

samples -

Surface

chemistry

Commonwealth

Scientific and

Industrial

Research

Organisation

2012-

03-05

Data access





Test guideline



Methods

XPS


Individual wells of a powder sample holder were filled with “as received” ZnO samples (1 well per

sample). Samples were analysed using a Kratos HS spectrometer fitted with a monochromated Al

K(alpha) source, under standard conditions. The sampling depth was several nm (~10 nm maximum) and

the analysis area was ~ 0.3 mm × 0.7 mm. Samples were analysed at 2 different times, approximately 2

months apart


131

Data gathering

Instruments

Kratos HS spectrometer fitted with a monochromated Al K(alpha) source

GLP compliance

no

Test materials


yes


Identifier Identity


EC number 215-222-5

IUPAC name oxozinc


Test material form




Code: NM110

Supplier: BASF SE










TM Code: NM112





-Name of test material: Zinc OxideCode: NM113


Surface coating: none- Substance type: Inorganic



Functionalisation


The table presents the average of two data sets obtained 2 months apart. In all samples, zinc, oxygen and

carbon were the major species present, with minor traces of chlorine detected in NM110, NM112 and

NM113 but not NM111. The ratios of Zn:O were near, but not quite, stoichiometric typically with a


132

deficiency of oxygen. The significant level of carbon present is likely due to surface contamination (the

technique is sensitive to contaminations) or carbon-containing species adsorbed on the surface; however,

NM-111 had significantly more carbon than the other samples, consistent with the fact that it has

atriethoxycaprylylsilanesurface coating. There appeared to be little change over the 8 week period except

that data from the second experiment were noticeably closer to ZnO stoichiometry than the first.

Table Surface elemental composition measured in 8 weeks apart

Element NM 110

(At. %)

NM 111

(At. %)

NM 112

(At. %)

NM 113

(At.%)

Zn 42.86

(39.57)

38.55

(35.49)

45.33

(41.48)

43.46

(38.88)

O 35.92

(40.54)

34.19

(36.92)

34.93

(41.83)

35.72

(38.41)

C 20.29

(19.43)

27.25

(27.59)

18.21

(16.21)

20.00

(22.23)

Cl 0.92

(0.46)

1.53

(0.48)

0.82

(0.48)

( ) = Eight weeks later



Attached full

study report

NM_110_ANNEX_A69_OECD ZnO Surface Chemistry XPS CSIRO Australia.docx / 23.81 KB



Conclusions

For NM110, zinc, oxygen and carbon were the major species present, with minor traces of chlorine. In the

first experiment, the ratio of Zn:O was not quite stoichiometric, with a deficiency of oxygen. There

appeared to be little change over the 8 week period except that data from the second experiment were

noticeably closer to ZnO stoichiometry than the first. The significant level of carbon present is likely due

to surface contamination (the technique is sensitive to contaminations) or carbon-containing species

adsorbed on the surface.

Executive summary

The surface chemistry, in particular the elemental composition near the surface of nanomaterials, was

investigated CSIRO, 2012. Therefore an X-ray photoelectron spectroscopy (XPS) measurement of the test

items was conducted. Spectra were obtained by irradiating the sample with an X-ray beam while

simultaneously measuring the kinetic energy and number of electrons that escape from the top 1-10 nm

layer of the material being analysed. In all samples, zinc, oxygen and carbon were the major species

present, with minor traces of chlorine detected in NM-110, NM-112 and NM-113 but not NM-111. The

ratios of Zn:O were near, but not quite, stoichiometric typically with a deficiency of oxygen. The

significant level of carbon present is likely due to surface contamination (the technique is sensitive to

contaminations) or carbon-containing species adsorbed on the surface. However, NM-111 had

significantly more carbon than the other samples, consistent with the fact that it has a

triethoxycaprylylsilane surface coating. There appeared to be little change over the 8 week period except

that data from the second experiment were noticeably closer to ZnO stoichiometry than the first

determination.


133

Endpoint study record: Disregarded_2010-07-02_NPL for

PROSPECT_Surface_chemistry_XPS

Administrative Data



Reliability 3 (not reliable)

Rationale for

reliability incl.

deficiencies

Limited documentation of the data evaluation was provided in the full study report. Thus, it

was not possible to assess the validity of the provided data as the sample preparation and the

data analysis have a significant influence to the result. Due to the contamination on the

particles with carbon the results of the measurements were not reliable and thus were not

taken into account

Data source

Reference

Reference

type


source

Testing

laboratory

Report

no.

Owner

company

Company

study no.

Report

date

study

report

National

Physical

Laboratory

2010 XPS element

atomic

concentrations

National

Physical

Laboratory,

Teddington,

Middlesex,

United

Kingdom

2010-

07-02

Data access





Test guideline


according to other guideline: ISO 15472 no data

Methods

XPS (ISO 15472)


XPS measurements were obtained in ultra high vacuum using a Kratos AXIS Ultra DLD (Kratos

Analytical, UK) instrument fitted with a monochromated Al Kasource, which was operated at 15kV and

5mA emission. Photoelectrons from the top few nanometres of the surface were detected in the normal

emission direction over an analysis area of approximately 700 x 300 micrometres. Spectra in the range

1400 to –10 eV binding energy and a step size of 1 eV, using a pass energy of 160 eV were acquired from

selected areas of each sample. The peak areas were measured after removal of a Tougaard background.

The manufacturer’s intensity calibration and commonly employed sensitivity factors were used to

determine the concentration of the elements present. High resolution narrow scans of some of the peaks of

interest were acquired with a step size of 0.1 eV and 20 eV pass energy. (The manufacturer calibrated the

intensity calibration over the energy range). The energy scale was calibrated according to ISO 15472

Surface chemical analysis – X-ray photoelectron spectrometers – Calibration of energy scales. However,


134

the charge neutraliser was used when acquiring the spectra, which shifted the peaks, by several eV. The C

1s hydrocarbon peak (285 eV binding energy) was used to determine the shift for identifying the peaks.

Samples were prepared using carbon adhesive tape to affix them to 1 cm copper squares. Care was taken

to cover the tape with the powders as completely as possible but some samples had better coverage than

others and in a lot of cases there was a signal detected from the tape as well as the powder itself. The tape

contained oxygen and silicon in addition to carbon.

Data gathering

Instruments

Kratos AXIS Ultra DLD (Kratos Analytical, UK) instrument fitted with a monochromated Al Kasource

Calibration

The energy scale was calibrated according to ISO 15472 Surface chemical analysis – X-ray

photoelectronspectrometers – Calibration of energy scales

GLP compliance

no

Test materials


yes


Identifier Identity


EC number 215-222-5

IUPAC name oxozinc


Test material form




Code: NM110





Code: NM111

Supplier: BASF SE





TM Code: NM112






135


Code: NM113Supplier: Sigma-Aldrich






The elemental composition of the different PROSPEcT powders as measured by XPS is summarised

below, where the elemental concentrations of the elements: carbon (C), cerium (Ce), oxygen (O), silicon

(Si) and zinc (Zn) are shown.As expected, XPS can successfully differentiate between vials containing

cerium and vials containing zinc; and it was observed that no cross contamination had occurred between

the two types of batches within the detection limit of XPS. As evident from the results, there was a

significant contribution of carbon and this can be largely attributed to contamination on the particles.

Areas of best coverage were selected for analysis and, using XPS analysis of the carbon tape alone which

showed a composition of 74% C, 21% O and 5% Si. From the lack of any significant signal from Si on

samples, it was estimated that there was better than 90% coverage within these analysis areas. A different

sample preparation procedure could be adopted to separate background carbon signal from that on the

particles during XPS measurements. XPS results also showed the presence of Si and this was mainly

associated with Z-COTE HP 1 ZnO sample i.e. Si 2s of 3.5%. This can be attributed to the fact that this

sample was coated with triethoxycapryl silane and hence the silicon signal contribution. The silicon

contribution with the Z-COTE of 0.3 % is lower than the estimated detection limit for Si of ~ 0.5% and

can be regarded as lying within the noise level.

Sample


C 1s

(%)

O 1s

(%)

Si 2s

(%)

Zn

2p3/2(%)

NM112 Micronisers, Australia ZA250#30#03 64.7 26.9 0.0 8.4

NM111 BASF, Germany ZB250#64#03 67.9 24.3 3.5 4.3

NM113 Sigma Aldrich, UK ZrA250#33#ICP 25.6 44.3 0.0 30.1

NM110 BASF, Germany ZC250#37#03 69.0 25.1 0.3 5.6 XPS element atomic concentrations results of PROSPEcT powders; the powders were spread on to an adhesive

carbon tape.


Attached full study

report

NM_110_ANNEX_A70_XPS element concentrations.doc / 38 KB (application/octet-


Attached full study

report




Executive summary

A study, which was disregarded, was conducted by NPL, 2010 to determine the elemental composition of

the different nanomaterials as measured by XPS. XPS measurements were obtained in ultra-high vacuum

using a Kratos AXIS Ultra DLD (Kratos Analytical, UK) instrument fitted with a monochromated Al K-

source, which was operated at 15 kV and 5 mA emission. Photoelectrons from the top few nanometers of

the surface were detected in the normal emission direction over an analysis area of approximately 700 x

300 micrometres. Spectra in the range 1400 to –10 eV binding energy and a step size of 1 eV, using a

pass energy of 160 eV were acquired from selected areas of each sample. The peak areas were measured

after removal of a Tougaard background. The manufacturer’s intensity calibration and commonly

employed sensitivity factors were used to determine the concentration of the elements present. High


136

resolution narrow scans of some of the peaks of interest were acquired with a step size of 0.1 eV and

20 eV pass energy. (The manufacturer calibrated the intensity calibration over the energy range). The

energy scale was calibrated according to ISO 15472. The charge neutraliser was used when acquiring the

spectra, which shifted the peaks, by several eV. The carbon 1s hydrocarbon peak (285 eV binding energy)

was used to determine the shift for identifying the peaks. Samples were prepared using carbon adhesive

tape to affix them to 1 cm copper squares. Care was taken to cover the tape with the powders as

completely as possible but some samples had better coverage than others and in a lot of cases there were a

signal detected from the tape as well as the powder itself. The tape contained oxygen and silicon in

addition to carbon. As evident from the results, there was a significant contribution of carbon and this can

be largely attributed to contamination on the particles. Areas of best coverage were selected for analysis

and, using XPS analysis of the carbon tape alone which showed a composition of 74% C, 21% O and 5%

Si. Due to the lack of any significant signal from Si on samples, it was estimated that there was a better

than 90 % coverage within these analysis areas. XPS also showed the presence of Si mainly associated

with NM-111 sample i.e. Si 2s of 3.5 %. This can be attributed to the fact that this sample was coated

with triethoxycapryl silane. The silicon contribution with NM-110 of 0.3 % is lower than the estimated

detection limit for Si of ~ 0.5% and can be regarded as within the analytical noise level.

Endpoint study record: Disregarded_2011-12-14_NPL for PROSPECT_ToF

SIMS

Administrative Data



Reliability 3 (not reliable)

Rationale for

reliability incl.

deficiencies

In the study report, it was concluded that XPS technique provides a more informative and quantitative

chemical analysis of the NMs as the ToF-SIMS method has significant deficits at the moment. Thus,

the provided data were regarded as unreliable and were not taken into account.

Data source

Reference

Reference

type


source

Testing

laboratory

Report

no.

Owner

company

Company

study no.

Report

date

study

report

National

Physical

Laboratory

2011 Characterization

of PROSPEcT

samples by

ToF-SIMS

National

Physical

Laboratory,

Teddington,

Middlesex,

United

Kingdom

2011-

12-14

Data access





Test guideline



Methods

SIMS (ToF SIMS)


137


The ToF-SIMS instrument utilized for this work is an IONTOF (Germany) ToF-SIMS IV. ToF-SIMS

analysis was performed with a Bi+ cluster primary beam (incident at 45° from the sample normal with an

energy of 25 keV) to obtain high mass resolution mass spectra with an imaging resolution of

approximately 5 μm. The ion beam is rastered over an area of 500 μm × 500 μm using a 256 × 256 pixel

raster. The ion beam current was measured before and after each sample was analysed. The raw data that

was recorded consisted of a ToF-SIMS mass spectrum at every pixel. Data analysis was carried out

retrospectively from the raw data using ION-TOF SurfaceLab 6.1 software

Data gathering

Instruments

IONTOF

Calibration

The instrument is periodically checked for repeatability and constancy of the relative intensity scale using

ISO 23830:2008. The mass scale is calibrated following ISO/DIS 13084. The spatial scales are calibrated

using metal grids with spacings of 25 μm and 125 μm.

GLP compliance

no

Test materials


yes


Test material form




Code: NM110

Supplier: BASF SE





Code: NM111

Supplier: BASF SE





TM Code: NM112

Identifier Identity


EC number 215-222-5

IUPAC name oxozinc



138






Code: NM113






Functionalisation


Effect of sample topography Assessment of sample preparation protocol for Tof-SIMS analysis performed on ZnO nanoparticles.

Enlargement of the spectrum showing the 69Zn+ peak. The spectrum exhibits a satellite peak shifted

0.045 mass units from the Zn+ peak.

Region-of-interest ToF-SIMS spectra, regenerated from two areas of the image which are shown in the

inset. This shows that the higher intensity peak (red) originates from the central area of the sample, and

the shifted peak (blue) originates from the bright area surrounded by dark circular regions. Similar

features are observed on other samples and is a typical artefact due to sample topography.

As expected, a 69Zn+ peak is observed at 69.93 u in the mass spectra. However, it was also observed an

unknown satellite peak whose apparent mass was 0.045 u higher than the 69Zn+ peak. A detailed data

analysis was performed to elucidate the origin of the two peaks. The total ion image, reveals significant

spatial inhomoegenity on the sample. It was found that the two peaks originate from different regions of

the sample. The higher intensity peak originates from the flat central area, while the shifted peak

originates from a bright area which is surrounded by a dark circular region. Similar effects were observed

for other peaks associated to the nanoparticles, both on this and other samples. This effect was interpreted

as an instrumental artefact due to sample topography. The peak marked in red is associated to regions of

the sample where the nanoparticle powder forms a relatively flat film on the adhesive tape. The bright

areas which are surrounded by circular dark regions are interpreted as a large aggregation of nanoparticles

(> 100 μm) on the surface of the sample. Due to topography effects, secondary ions from the top of the

aggregate particle are detected with a delayed time-of-flight (higher apparent mass) compared to

secondary ions from the flat areas of the sample. These peaks are also broadened and have poorer mass

resolution compared to peaks from the flat area, due to the larger spread in ion time-of-flight. The

aggregate is also surrounded by dark areas where the topographic effects prevent the secondary ions from

reaching the mass spectrometer. For these reasons, for the purpose of this study, only signals originating

from the flat areas of the samples were considered in the analysis of the ZnO powders.

Chemical analysis of nanopowders The spectrum of each nanopowder sample was compared against that of the carbon tape. In addition to

the peaks characteristic for the carbon tape, the nanopowder samples exhibited peaks that were related to

the elemental composition of the nanopowders. ZnO nanopowder samples exhibited mainly peaks

associated with Zn and O and their isotopes (Figure 2C). In addition to H and C contaminants, N was also

observed (Figure 2D). It is interesting to note that some of the peaks associated with the presence of CH3

groups are low for Sample C (Figure 2D). This sample is the Z-cote HP1 nanoparticles, which according

to the manufacturer (BASF) has a triethoxycaprylylsilane coating at the surface. It is possible that CH3

groups are characteristic of the surface of uncoated ZnO nanoparticles. However, the silicone shell could


139

not be detected due to the presence of strong signals from the carbon tape. The strong Zn signal from

Sample C is interesting. Since ToF-SIMS is sensitive only to the outermost atomic layers of a surface,

comprising a thickness of few nanometres, the detection of Zn implies that the silicone shell is either

thinner than a few nanometers or it does not cover the surface homogeneously.



Attached full study

report

NM_110_ANNEX_A71_Minelli_Characterization of PROSPEcT samples by ToF-

SIMS.pdf / 265.25 KB (application/octet-stream): ENV/JM/MONO(2015)15/ANN13



Conclusions

The evaluation of sample purity and type of contaminations by ToF-SIMS is challenging, due to the

restrictions imposed by sample mounting requirements. The use of the carbon tape limits the extent of the

analysis that can be performed on signals related to C and Si elements. A more ideal mounting strategy is

that adopted for XPS measurements, where the analysis is performed on a thick nanopowder pellet and

therefore 100% of the signal is collected from the nanomaterials. Unfortunately, for reasons that were

explained, this approach is unsuitable for ToFSIMS at present.TOF-SIMS is by principle not quantitative,

and rather useful to identify traces, not to quantify them nor to compare samples. Thus, the study was

disregarded and the results were not taken into account.

Executive summary

A study, which was disregarded, was conducted by NPL, 2010 using Time of Flight Secondary Ion Mass

Spectrometry (ToF-SIMS). ToF-SIMS analysis was performed with a Bi+ cluster primary beam (incident

at 45° from the sample normal with an energy of 25 keV) to obtain high mass resolution mass spectra

with an imaging resolution of approximately 5 μm. The ion beam is rastered over an area of 500 μm ×

500 μm using a 256 × 256 pixel raster. The ion beam current was measured before and after each sample

was analysed. The raw data recorded consisted of a ToF-SIMS mass spectrum at every pixel. Data

analysis was carried out retrospectively from the raw data using ION-TOF Surface Lab 6.1 software. The

result was regarded as unreliable and therefore was not taken into account.


140


PROSPECT_Homogeneity_NM110-4899_XPS

Administrative Data





incl. deficiencies

The study was not conducted in compliance with GLP. The results are scientifically

acceptable. Thus, the study was regarded as reliable with restrictions.

Data source

Reference

Reference

type


source

Testing

laboratory

Report

no.

Owner

company

Company

study no.

Report

date

study

report

National

Physical

Laboratory

2010 National

Physical

Laboratory,

Teddington,

Middlesex,

United Kingdom

2010-

12-06

Data access





Test guideline



Methods

XPS



















141


Data gathering

Instruments

Instruments: Kratos AXIS Ultra DLD (Kratos Analytical, UK)Calibration: The manufacturer’s intensity

calibration

GLP compliance

no

Test materials


Identifier Identity


EC number 215-222-5

IUPAC name oxozinc

other: OECD sponsorship programme NM110, NM111

Test material form




Code: NM110

Supplier: BASF SE





Code: NM111

Supplier: BASF SE





Functionalisation


The table shows the elemental compositions of the sub-sampled BASF powders for: a) Z-COTE and b) Z-

COTE HP 1. The powders are adhered on to an (adhesive) carbon tape, in which the elemental

composition of the tape was shown to be (atomic %) 74.3% C, 20.9% O, 4.8% Si. It is clear from the

table of results that there is significant carbon and oxygen signal for both Z-COTE and Z-COTE HP1,

which potentially originates from the carbon tape on which the NMs were fixed. Although the area

(analysis area of ~ 700 x 300 µm, with information depth of ~ 8nm) was carefully chosen to obtain

maximum particle coverage, it is clear that the carbon and oxygen tape background signal is contributing

towards the XPS signal. Nonetheless, we can deduce clear significant difference in the XPS results

between the two sets of vials, which are as follows:

a) The count rate of Zn peaks were always lower from Z-COTE HP 1 samples vs. Z-COTE samples i.e. 4

to 11.5 kcps and 19 to 23 kcps, respectively. This can be attributed to the presence of a triethoxycarpryl

silane coating associated with Z-COTE HP 1 samples.


142

b) The Si level is much higher (3.1 to 4.1 %) in Z-COTE HP 1 if compared to Z-COTE (0 to ~1%). This

is consistent with the presence of a silane coating with the former sample. The silicon signal contribution

(of less than 1%) can be attributed to silicon background signal from the fixing tape.

XPS results for JRC sub-sampled powders for: a) BASF Z-COTE (NM-110) and b) Z-COTE HP 1 (NM-

111). Replicates: 1 vial, 1 replicate per vial.

NM-110/atomic%

Sample batch C 1s% O 1s% Si 2s% Zn 2p3/2%

NM110-4899 57.7 29.8 1.0 11.5

NM110-2617 45.7 35.2 0.2 18.9

NM110-1866 43.8 36.2 0.0 19.9

NM110-3795 35.7 39.7 0.0 24.7

NM110-0286 38.0 38.8 0.0 23.1

NM110-0305 36.0 39.6 0.0 24.4

NM-111/atomic%

Sample batch C 1s% O 1s% Si 2s% Zn 2p3/2%

NM111-4825 59.1 28.3 3.4 9.2

NM111-2419 68.6 23.7 3.9 3.9

NM111-1869 70.6 23.2 3.5 2.7

NM111-4779 67.7 24.3 3.8 4.1

NM111-1017 57.7 28.4 4.1 9.7

NM111-3396 70.2 23.3 3.4 3.1

NM111-0486 72.8 22.3 3.4 1.5



Attached full study

report

NM_110_ANNEX_A06_Homogeneity testing using XPS.doc / 40 KB (application/octet-



Conclusions

The count rate of Zn peaks were always lower from NM-111 samples vs. NM-110 samples i.e. 4 to 11.5

kcps and 19 to 23 kcps, respectively. This was attributed to the presence of a triethoxycarpryl silane

coating associated with NM-111 samples. Furthermore the Si level was much higher (3.1 to 4.1 %) in

NM-111 1 if compared to NM-110 (0 to ~1%). This was consistent with the presence of a silane coating

with the former sample. The silicon signal contribution (of less than 1%) can be attributed to silicon

background signal from the fixing tape.

Executive summary

NPL, 2010 determined the elemental composition of the different nanomaterials as measured by XPS.












143








The elemental compositions of the sub-sampled BASF powders for: a) Z-COTE and b) Z-COTE HP

1. The powders were adhered on to an (adhesive) carbon tape, in which the elemental composition of the

tape was shown to be (atomic %) 74.3% C, 20.9% O, 4.8% Si. It is clear from the table of results that

there is significant carbon and oxygen signal for both Z-COTE and Z-COTE HP1, which potentially

originates from the carbon tape on which the NMs were fixed. Although the area (analysis area of ~ 700 x

300 µm, with information depth of ~ 8nm) was carefully chosen to obtain maximum particle coverage, it

is clear that the carbon and oxygen tape background signal is contributing towards the XPS signal.

Nonetheless, it was deduced a clear significant difference in the XPS results between the two sets of vials

The count rate of Zn peaks were always lower from NM-111 samples vs. NM-110 samples i.e. 4 to 11.5

kcps and 19 to 23 kcps, respectively. This was attributed to the presence of a triethoxycarpryl silane

coating associated with NM-111 samples. Furthermore the Si level was much higher (3.1 to 4.1 %) in

NM-111 1 if compared to NM-110 (0 to ~1%). This was consistent with the presence of a silane coating

with the former sample. The silicon signal contribution (of less than 1%) can be attributed to silicon

background signal from the fixing tape.


PROSPECT_Homogeneity_NM110-2617_XPS

Administrative Data





incl. deficiencies

The study was not conducted in compliance with GLP. The results are scientifically

acceptable. Thus, the study was regarded as reliable with restrictions.

Data source

Reference

Reference

type


source

Testing

laboratory

Report

no.

Owner

company

Company

study no.

Report

date

study

report

National

Physical

Laboratory

2010 Homogeneity

NM110-2617

XPS

National

Physical

Laboratory,

Teddington,

Middlesex,

United

Kingdom

2010-

12-06

Data access



yes


144


Test guideline



Methods

XPS



Analytical, UK) instrument fitted with a monochromated Al Ksource, which was operated at 15kV and










the charge neutraliser was used when acquiring the spectra, which shifted the peaks by several eV. The C

1s hydrocarbon peak (285 eV binding energy) was used to determine the shift for identifying the

peaks.The pellets of the sample powders were produced using the KBr Quick Press pellet presser. The

powder was loaded from half to ¾ filled and gently pressed before the 3 pieces were inserted into the

socket of the pellet maker. The handle was carefully pressed until some resistance was felt, and when

pressed downwards there was ~ 1. 5 cm gap between the stop-screw on the handle to the central body.

After a few minutes the handle was released, and pressure re-applied twice more until the presser clicked

three times in total. The 3 piece assembly was taken out and the pellet removed.

Data gathering

GLP compliance

no

Test materials


yes


Identifier Identity


EC number 215-222-5

IUPAC name oxozinc

other: OECD sponsorship programme NM110, NM111, Nm112, NM113

Test material form







145




Code: NM111

Supplier: BASF SE





TM Code: NM112






Code: NM113


Surface coating: none- Substance type: Inorganic- Physical state: solid powder, nano-form


Functionalisation


Showing the elemental compositions from the NM111 ZnO HP1 sample as determined by XPS. Stub name replicate number C 1s % O 1s % Si 2s % Zn 2p3/2 %

0803 rep1 21.5 44.0 1.8 32.6

21.9 43.8 1.9 32.4

rep2 24.8 41.5 2.0 31.8

25.5 41.3 1.5 31.7

rep3 25.5 41.3 1.5 31.6

25.9 40.7 1.9 31.5

1951 rep1 31.2 37.1 1.4 30.3

rep2 29.2 39.0 1.7 30.1

rep3 29.9 38.5 2.0 29.7

1455 rep1 31.5 37.2 1.6 29.7

rep2 30.5 38.1 1.1 30.3

rep3 32.0 36.9 1.4 29.6

2017 rep1 29.6 38.6 1.8 30.0

rep2 30.1 38.0 1.6 30.3

rep3 30.4 37.8 2.2 29.6

2100 rep1 30.3 37.6 1.6 30.5

rep2 31.8 36.6 1.8 29.7

rep3 31.5 37.0 1.2 30.3

30.0 37.9 1.6 30.5

4414 rep1 31.1 37.6 1.2 30.1

31.7 36.9 1.6 29.8

rep2 28.2 39.4 2.0 30.5

29.0 38.9 1.9 30.1

rep3 29.2 38.4 2.2 30.2


146

Showing the elemental compositions from the NM110 ZnO sample as determined by XPS. Stub name replicate number C 1s

%

O 1s % Si 2s % Zn 2p3/2 %

0979 rep1 24.2 39.3 0.0 36.5

rep2 24.8 39.6 0.0 35.6

rep3 25.9 39.2 0.0 34.8

0599 rep1 24.1 40.2 0.0 35.7

rep2 26.2 39.1 0.0 34.7

rep3 26.6 39.0 0.0 34.4

0945 rep1 23.7 40.3 0.0 36.0

rep2 25.0 39.6 0.0 35.5

rep3 25.9 39.4 0.0 34.7

4779 rep1 23.2 36.0 0.0 36.2

rep2 26.0 39.3 0.0 34.7

rep3 25.6 39.6 0.0 34.7

4410 rep1 19.6 41.2 0.0 39.2

rep2 22.2 40.2 0.0 37.6

rep3 22.0 40.7 0.0 37.3

3911 rep1 20.3 40.8 0.0 38.9

rep2 21.5 40.3 0.0 38.2

rep3 22.8 40.2 0.0 36.9



For the NM 111 - ZnO HP1 sample:

1. Zinc, Oxygen and Silicon (the latter from the silane layer) were all detected.

2. All samples show a large amount of carbon probably adsorbed from the atmosphere. The

percentage concentration of carbon is consistent in the region 30-32% except from 0803 sample, which is

in the region 22-26%.

3. The percentage levels of the other sub-samples are remarkable consistent varying only between 37

– 40 % for Oxygen, 39-31% for Zinc and 1.2 – 2.2 % for Silicon.

4. More Oxygen that Zinc was detected due to the oxygen in the silane layer.

For the NM110 - ZnO sample:

1. Zinc and Oxygen were detected but no Silicon. Again, Carbon was detected.

2. More Oxygen was detected than expected considering the amount of Zinc (assume a 1:1 ratio from

Zinc Oxide). The amount of Oxygen varied between 39-41% and Zinc from 35-39%.

3. Carbon shows the greatest variability from 20 – 26%.


Attached document NM_110_ANNEX_A07_draft homogenity report XPS.doc / 126.5 KB (application/octet-



Conclusions

For the NM 111 - ZnO HP1 sample Zinc, Oxygen and Silicon (the latter from the silane layer) were all

detected. All samples showed a large amount of carbon probably adsorbed from the atmosphere. The

percentage concentration of carbon is consistent in the region 30-32% except from 0803 sample, which is

in the region 22-26%. The percentage levels of the other sub-samples are remarkable consistent varying


147

only between 37 – 40 % for Oxygen, 39-31% for Zinc and 1.2 – 2.2 % for Silicon. More Oxygen that

Zinc was detected due to the oxygen in the silane layer.For the NM110 - ZnO sample Zinc and Oxygen

were detected but no Silicon. In addition, Carbon was detected. More Oxygen was detected than expected

considering the amount of Zinc (assume a 1:1 ratio from Zinc Oxide). The amount of Oxygen varied

between 39-41% and Zinc from 35-39%.Carbon shows the greatest variability from 20 – 26%.

Executive summary

NPL, 2010 determined the elemental composition of the different nanomaterials as measured by XPS.


Analytical, UK) instrument fitted with a monochromated Al K alpha source, which was operated at 15kV

and 5mA emission. Photoelectrons from the top few nanometres of the surface were detected in the

normal emission direction over an analysis area of approximately 700 x 300 micrometres. Spectra in the

range 1400 to –10 eV binding energy and a step size of 1 eV, using a pass energy of 160 eV were

acquired from selected areas of each sample. The peak areas were measured after removal of a Tougaard

background. The manufacturer’s intensity calibration and commonly employed sensitivity factors were

used to determine the concentration of the elements present. High resolution narrow scans of some of the

peaks of interest were acquired with a step size of 0.1 eV and 20 eV pass energy. (The manufacturer

calibrated the intensity calibration over the energy range). The energy scale was calibrated according to

ISO 15472 Surface chemical analysis – X-ray photoelectron spectrometers – Calibration of energy scales.

However, the charge neutraliser was used when acquiring the spectra, which shifted the peaks by several

eV. The C 1s hydrocarbon peak (285 eV binding energy) was used to determine the shift for identifying

the peaks. The pellets of the sample powders were produced using the KBr Quick Press pellet

presser. The powder was loaded from half to ¾ filled and gently pressed before the 3 pieces were inserted

into the socket of the pellet maker. The handle was carefully pressed until some resistance was felt, and

when pressed downwards there was ~ 1. 5 cm gap between the stop-screw on the handle to the central

body. After a few minutes the handle was released, and pressure re-applied twice more until the presser

clicked three times in total. The 3 piece assembly was taken out and the pellet removed.

The experimentally determined elemental compositions for all materials For the NM 111 - ZnO HP1

sample Zinc, Oxygen and Silicon (the latter from the silane layer) were all detected. All samples showed

a large amount of carbon probably adsorbed from the atmosphere. The percentage concentration of

carbon is consistent in the region 30-32% except from 0803 sample, which is in the region 22-26%. The

percentage levels of the other sub-samples are remarkable consistent varying only between 37 – 40 % for

Oxygen, 39-31% for Zinc and 1.2 – 2.2 % for Silicon. More Oxygen that Zinc was detected due to the

oxygen in the silane layer.

For the NM110 - ZnO sample Zinc and Oxygen were detected but no Silicon. In addition, Carbon was

detected. More Oxygen was detected than expected considering the amount of Zinc (assume a 1:1 ratio

from Zinc Oxide). The amount of Oxygen varied between 39-41% and Zinc from 35-39%.Carbon shows

the greatest variability from 20 – 26%.

4.31 Dustiness

Endpoint study record: Key_2011-25-11_NPL for PROSPECT_Dustiness

Administrative Data




Rationale for

reliability incl.

deficiencies

The method is recommended in the respective ECHA guidance and was regarded as

suitable to determine the distribution of particles of respirable or inhalable size. However,

only limited documentation of the data evaluation was provided in the full study report.


148

Data source

Reference

Reference

type


source

Testing

laboratory

Report

no.

Owner

company

Company

study no.

Report

date

study

report

National

Physical

Laboratory

2011 National

Physical

Laboratory,

Teddington,

Middlesex,

United Kingdom

Data access





Test guideline


according

to

EN 15051: 2006 (Workplace atmospheres - Measurement of the dustiness of bulk

materials - Requirements and reference test methods)

no data

Methods

rotating cylinder


The dustiness of the sample powder will be carried out using the rotating drum method specified in the

new European standard of the dustiness of bulk samples (EN15051). This device uses two porous size-

selective foam stages and a filter to provide information about the dustiness of the powders in terms of the

inhalable, thoracic and respirable fractions of the dust dispersed. Three replicate tests of the powder will

be carried out to obtain an estimate of the precision of the measurements. The standard also requires

simple moisture content measurements to be made for each material, as dustiness has been found to be a

function of moisture content. Analysis is carried out at 50+5% relative humidity. For each measurement

35 ml will be tested.

Data gathering

GLP compliance

no

Test materials


yes


Identifier Identity


EC number 215-222-5

IUPAC name oxozinc



149

Test material form




Code: NM110





Code: NM111

Supplier: BASF SE





TM Code: NM112






Code: NM113







Table1. Mean and SD of the dustiness results (in terms of the inhalable, thoracic and respirable fractions of dust

dispersed) and moisture content of the PROSPEcT sample powders.

Sample

Name Supplier

Batch

number

Inhalable

fraction

(mg kg−1)

Thoracic

fraction(mg

kg−1)

Respirable

fraction(mg

kg−1)

Moisture

content (%)

Mean SD Mean SD Mean SD Mean SD

NM110 BASF, ZC250#78 2905 371 599 239 27 3 0.5 0.10

NM111 BASF, ZB250#64 5880 610 1340 241 138 105 0.3 0.00

NM112 Micronisers, ZA250#30 1095 222 317 37 42 8 1.2 0.20

NM113 Sigma, ZrA250#33 166 26 34 10 10 2 0.4 0.10

Table2. The calculated coefficient of variation (COV) values of the results inTable1. The coefficient of variation

is the ratio of the standard deviation to the mean. It is a measure of variability in relation to the mean and is used

to compare the relative dispersion in one type of data with the relative dispersion in another type of data. The

data to be compared may be in the same units, in different units, with the same mean, or with different means.* Sample

Name

Inhalable

fraction

(mg kg−1)

Thoracic fraction(mg

kg−1)

Respirable fraction(mg

kg−1)

Moisture content

(%)


150

COV (%) COV (%) COV (%) COV (%)

NM110 12.8 39.9 12.7 21

NM111 10.4 18.0 76.2 0

NM112 20.3 11.5 19.9 18

NM113 15.8 29.5 17.6 34



Attached full study

report

NM_110_ANNEX_A75_010611HSL_results only.doc / 68 KB (application/octet-



Executive summary

NPL, 2010 investigated the dustiness of the nanomaterials. The dustiness of the sample powder was

carried out using the rotating drum method specified in the new European standard of the dustiness of

bulk samples (EN15051). Three replicate tests of the powder were carried out to obtain an estimate of the

precision of the measurements. The standard also requires simple moisture content measurements to be

made for each material, as dustiness has been found to be a function of moisture content. Analysis was

carried out at 50 + 5 % relative humidity. For each measurement 35 mL was tested. Test results of the

dustiness studies showed a significant difference in the inhalable dustiness levels. The respirable

dustiness index, however, was quite comparable and possibly influenced by larger variation than the

inhalable dust fraction. The inhalable dustiness index is classified to be at the high end of “low” dustiness

(NM-110) to just “moderate” (NM-111). This compares approximately to the levels of nanoparticle

powders of goethite, organoclay and talc compared to dustiness data on other test nanomaterials. For

Respirable dust both samples are in the lower “moderate” dustiness range (range: 50 to 250 mg/kg).

4.32 Porosity Endpoint study record: WoE_2012-03-06_Australia_CSIRO_porosity_BJH

Administrative Data




Rationale for

reliability incl.

deficiencies

The used method is recommended by the OECD guidance and the data were analysed using

the t-plot method. However, only limited documentation of the data evaluation was provided

in the full study report. The data analysis could have a significant influence to the result, there

are some concern regarding validity. Nevertheless, the provided data were regarded as most

reliable (compared to the other provided studies).

Data source

Reference

Reference

type


source

Testing

laboratory

Report

no.

Owner

company

Company

study no.

Report

date

study

report

Casey,

Yin,

McCall

2012 Summary of

physicochemical

properties of

OECD Zinc

Oxide (ZnO)

samples -

Porosity

Commonwealth

Scientific and

Industrial

Research

Organisation

2012-

03-06


151

Data access





Test guideline



Methods

BET (BJH)


Porosity was determined simultaneously with surface area using a Micromeritics Tristar II 3020

instrument. The Barrett-Joyner-Halenda (BJH) method for analysing gas adsorption and desorption

isotherms was used to determine pore area, specific pore volume and pore size distribution independent of

the external area due to the particle size of the sample. The t-plot method is commonly used to determine

the external surface area, pore volume and pore surface area in microporous solids.

Data gathering

Instruments

Micromeritics Tristar II 3020

GLP compliance

no

Test materials


yes


Identifier Identity


EC number 215-222-5

IUPAC name oxozinc

other: OECD Sponsorship programme NM110, NM111, NM112, Nm113

Test material form




Code: NM110

Supplier: BASF SE





Code: NM111

Supplier: BASF SE


152





TM Code: NM112






Code: NM113

Supplier: Sigma

-AldrichSurface coating: none





The Table presents the external surface area, micropore surface area and volume, and a determination of

the average pore width for all ZnO samples.All samples have either low or no microporosity. The major

contribution to total surface area is from external surfaces and is thus predominantly determined by

particle size and shape rather than high internal porosity.

For pristine samples of ZnO (NM110, NM112 and NM113) the surface areas, when determined by adding

the t-Plot micropore surface area and the t-plot external surface area, are generally consistent with those

determined by the BET method. However, for NM111 (the coated sample), the t-plot calculation indicates

no micropore surface area and an external surface area considerably greater than that determined by the

BET method (~21 m²/gcf 14 m²/g). This variation is likely due to the presence of the hydrophobic

(mesoporous) silicone coating which has capacity to adsorb gas both internally and externally and imply a

greater surface area. The fact that no microporous volume is reported suggests that any porosity is likely

mesoporous. Consequently, for the coated sample only, the-plot external surface area (multi-layer)

calculation is higher than the BET specific surface area (monolayer) calculation.

Table Porosity of the ZnO samples, NM110, NM111, NM112 and NM113.

Sample

Name

t-Plot

Micropore

Surface Area:

m²/g

t-Plot External

Surface Area : m²/g

t-Plot micropore

volume: cm³/g

BJH Desorption

average pore width

(4V/A): Å

NM-110 1.79315± 0.58 9.97± 0.98 0.000805± 0.00029 89.7445± 4.5

NM-111 0 20.899±0.312 0 208.02±4.67

NM-112 5.3518± 0.85 21.9027± 1.17 0.0024255± 0.0004 157.63± 10.3

NM-113 1.38765± 0.66 4.39675± 0.11 0.000638± 0.11 107.49± 12.7



Attached full

study report

NM_110_ANNEX_A76_OECD ZnO porosity CSIRO Australia.docx / 23.98 KB



Executive summary

CSIRO, 2012, determined the porosity simultaneously with surface area using a Micromeritics Tristar II


153

3020 instrument. The Barrett-Joyner-Halenda (BJH) method for analysing gas adsorption and desorption


the external area due to the particle size of the sample. All samples have either low or no microporosity.

The major contribution to total surface area is from external surfaces and is thus predominantly

determined by particle size and shape rather than high internal porosity. For pristine samples of ZnO

(NM-110, NM-112 and NM-113) the surface areas are generally consistent with those determined by the

BET method. However, for NM-111 (the coated sample), the t-plot calculation indicates no micropore

(pores smaller than 2 nm in diameter) surface area and an external surface area considerably greater than

that determined by the BET method (~21 m²/g cf 14 m²/g). This variation is likely due to the presence of

the hydrophobic (mesoporous) silicone coating which has capacity to adsorb gas both internally and

externally and imply a greater surface area. The fact that no microporous volume is reported suggests that

any porosity is likely mesoporous (pores greater than 2 nm and less than 50 nm in diameter).

Consequently, for the coated sample only, the t-plot external surface area (multi-layer) calculation is

higher than the BET specific surface area (monolayer) calculation.

Endpoint study record: WoE_2012-03-06_Australia_Deakin_Pore size_BJH

Administrative Data



Rationale for

reliability incl.

deficiencies

limited number of replicates. However, only limited documentation was provided. The

results were only presented as data plots and no further explanation with respect to data

handling and analysis were provided. There are some concerns on validity as the data

analysis have a significant influence to the result.

Data source

Reference

Reference

type


source

Testing

laboratory

Report

no.

Owner

company

Company

study no.

Report

date

study

report

Tsuzuki,

He

2012 Crystal phase &

crystallite size

measurements

for OECD ZnO

samples

Deakin

University

2012-

03-06

Data access





Methods

BET (BJH)


N2 gas adsorption at 77K

Sampling

Sample preparation: degassing at 150 C for 1 h


154

Data gathering

Instruments

Micromeritics Tristar 3000

Test materials


yes


Identifier Identity


EC number 215-222-5

IUPAC name oxozinc


Test material form




Sample Specific Surface Area (m2/g) Pore Volume (cm3/g) Average pore width (nm)

NM110 6.6 ± 0.3 0.0100 6.1

NM111 11.8 ± 0.2 0.0289 8.2

NM112 25.9 ± 0.3 0.0652 10

NM113 4.0 ± 0.15 0.00655 6.5 Note: • Pore volume = total pore volume of pores less than 475.454 Å diameter at P/Po = 0.957701651. • Average pore width was calculated as 4 x (pore volume) /(BET specific surface area) and hence does not

necessarily reflect the true state of pore structures. The scale of average pore width indicates that the quantity of meso-pores in in primary particles are negligible and

that the measured pore size distribution reflects the pores in the agglomerates (secondary particles).



Attached full study

report

NM_110_ANNEX_A77_OECD ZnO pore Auatsralia Deakin.pdf / 264.01 KB




155


Executive summary

The porosity of the samples were investigated by Deakin Univerity,2012 using a Micromeritics Tristar

3000 apparatus. The Barrett-Joyner-Halenda (BJH) method for analysing gas adsorption and desorption



156

the external area due to the particle size of the sample. The scale of average pore width indicates that the

quantity of meso-pores in in primary particles are negligible and that the measured pore size distribution

reflects the pores in the agglomerates (secondary particles). All samples have very low microporosity.

The major contribution to total surface area is from external surfaces and is thus predominantly

determined by particle size and shape rather than high internal porosity. NM-112 has the highest surface

area and micropore volume of all the samples approximately 3-4 times greater than other samples.

Endpoint study record: WoE_2011-12-19_NPL for PROSPECT_Porosity

Administrative Data




Rationale for

reliability incl.

deficiencies

The study was regarded was reliable with restrictions. The used methodology is well-

established. However, only very limited documentation of the data evaluation was provided in

the full study report. Furthermore the procedure for data analysis and representation was not

explained in detail. There were some concerns on validity as the data analysis have a

significant influence to the result.

Data source

Reference

Reference

type

Author Year Title Bibliogr

aphic

source

Testing

laboratory

Report no. Owner

company

Company

study no.

Report

date

study

report

National

Physical

Laboratory

2011 Porosity MCACambr

idge

010611MCAC

ambridge

2011-

12-19

Data access





Test guideline



Methods

BET


InstrumentationA Micromeritics TriStar II (3020) was used for the collection of nitrogen adsorption /

desorption isotherm data up to a saturation pressure of approximately 0.995 P/Po. The analysis was

typically conducted to measure 45 adsorption relative pressure points and 23 desorption relative pressure

points. Samples were outgassed overnight in vacuo at 300 degC using a Micromeritics VacPrep apparatus

prior to analysis. In order to indicate any possible microporous nature of the materials additional relative

pressure data were also collected at pressures lower than the usual starting point for analyses using this

instrument. These were in the approximate range 0.005 to 0.01 P/Po. Whilst the data reduction methods

available are unsuitable for application to the micropore range the characteristic shape of the adsorption


157

isotherm at these low partial pressures would provide a good indication of the presence of micropores in

the sample material. The sample tube dead space was measured for each analysis using helium (CP grade)

thus providing warm and cold freespace values. Samples requiring only BET surface area analysis were

analysed using the same equipment with the application ofthe same freespace measurement technique.

BET surface area was calculated using partial pressures in the nominal range 0.07 to 0.25.Data Reduction.

The samples for which the full adsorption/desorption isotherms were analysed graphical plots of the

adsorption/desorption isotherm and BET surface area transform plot together with tabulated data for each

are presented. The pore size distribution is presented as pore size by volume and area from the adsorption

isotherm using the BJH method. The lower limit of BJH calculations in terms of pore size (by diameter)

is extended below the typical value in order to highlight any possible microporous nature of the materials.

The pore size distribution data presented in the BJH reports is applied to a maximum of 1000Å, although

data below approximately 20Å should be considered only as a guide to the full porous nature of the

materials. The total pore volume of the materials is calculated from the volume of nitrogen adsorbed at

the maximum relative pressure obtained on the adsorption branch of the isotherm and is detailed on the

summary report.

Data gathering

Instruments

Micromeritics TriStar II

Calibration

BAM Standard - Alumina BAM-PM-104BAM Standard - Alumina BAM-PM-102

GLP compliance

no

Test materials


yes


Identifier Identity


EC number 215-222-5

IUPAC name oxozinc


Test material form




Code: NM110

Supplier: BASF SE





Code: NM111


158

Supplier: BASF SE





TM Code: NM112






Code: NM113







Porosity Sample Porosity (cm3/g)

NM110 0.041538

NM111 0.071347

NM112 0.158354

NM113 0.013820



Attached full study report NM_110_ANNEX_A43_010611MCACambridge.doc / 93 KB (application/octet-



Conclusions

The porosity was determined to be 0.041538 cm3/g for NM-110, 0.071347 cm3/g for NM-111, 0.158354

cm3/g for NM-112 and 0.013820 cm3/g for NM-113 respectively.

Executive summary

As study was conducted by the MCA in Cambridge, 2011 to determine the porosity of the NMs. A

Micromeritics TriStar II (3020) was used for the collection of nitrogen adsorption / desorption isotherm

data up to a saturation pressure of approximately 0.995 P/P0. The analysis was typically conducted to


outgassed overnight in vacuum at 300 °C using a Micromeritics VacPrep apparatus prior to analysis. In



were in the approximate range 0.005 to 0.01 P/P0. Whilst the data reduction methods available are

unsuitable for application to the micropore range the characteristic shape of the adsorption isotherm at


159

these low partial pressures would provide a good indication of the presence of micropores in the sample

material. The sample tube dead space was measured for each analysis using helium (CP grade) thus

providing warm and cold freespace values. Samples requiring only BET surface area analysis were

analysed using the same equipment with the application of the same freespace measurement technique.

BET surface area was calculated using partial pressures in the nominal range 0.07 to 0.25.

The pore size distribution is presented as pore size by volume and area from the adsorption isotherm

using the BJH method. The pore size distribution data presented in the BJH reports is applied to a

maximum of 1000 Å. The total pore volume of the materials is calculated from the volume of nitrogen

adsorbed at the maximum relative pressure obtained on the adsorption branch of the isotherm. The

porosity was determined to be 0.041538 cm3/g for NM-110, 0.071347 cm3/g for NM-111, 0.158354

cm3/g for NM-112 and 0.013820 cm3/g for NM-113 respectively.

4.33 Pour density

Endpoint study record: Key_2011-12-13_NPL for PROSPECT_pour density

Administrative Data




Rationale for

reliability incl.

deficiencies


the data evaluation was provided in the full study report. It is not clear how the data of the

bulk density was determined. In the study report the experimental determination of the tap

density and the calculation of the carr index where roughly described. According to the

OECD guidance the procedure for pour density measurements is described in ASTM Standard

D1513 – 05e2. Due to the limited documentation it was not possible to assess if the data were

obtained in accordance with recognised guidance. The provided results for density are not

consistent with the data on particle size which were in general regarded as reliable. This,

inconsistence might be caused by different parameter (e.g. sample preparation, representative

sample). Thus, it was not possible to finally conclude on pour density.

Data source

Reference

Reference

type


source

Testing

laboratory

Report

no.

Owner

company

Company

study no.

Report

date

study

report

Escubed

Ltd

2011 Pour density

Analysis:

Outsourced

work for

PROSPEcT.

National

Physical

Laboratory,

Teddington,

Middlesex,

United

Kingdom

2011-

12-13

Data access





Test guideline




160

Method

other: tapped density


Using a Copley JV2000, a known mass of the dry sample was placed into a measuring cylinder to a

recorded volume and ‘tapped’ by mechanically raising and lowering by a set distance until a consistent

volume was reached, which corresponds to the maximum packing density of the material.

Data gathering

Instruments

Copley

JV2000

GLP compliance

no

Test materials


yes


Identifier Identity


EC number 215-222-5

IUPAC name oxozinc


Test material form




Code: NM110

Supplier: BASF SE





Code: NM111

Supplier: BASF SE





TM Code: NM112






Code: NM113


161







Sample Bulk Density (g/cm3) Tapped Density (g/cm3) Carr Index

NM110 0.293 0.346 15.351

NM111 0.693 0.832 16.799

NM112 0.415 0.519 20.000

NM113 0.646 0.714 9.471



Attached full study report NM_110_ANNEX_A44_pour density.doc / 45.5 KB (application/octet-stream):



Executive summary

Escubed Ltd., 2012 determined the density of the nanomaterials using a Copley JV2000. A known mass

of the dry sample was placed into a measuring cylinder to a recorded volume and tapped by mechanically

raising and lowering by a set distance until a consistent volume was reached, which corresponds to the

maximum packing density of the material. The Carr Index was calculated from the respective bulk and

tapped densities.NM-110 and NM-111 have comparable particle size. NM-112 was determined to have

the smallest particle size compared to the other 3 nanomaterials and NM-113 (“bulk material”) was

determined to have the largest particle size. As the particle size directly influence the tapped density, this

general trend should be observed in the density measurements. The provided results for density are not

consistent with the data on particle size which were in general regarded as reliable. This, inconsistence

might be caused by different parameter (e.g. sample preparation, representative sample). Thus, it was not

possible to finally conclude on pour density.

4.34 Photocatalytic activity

Endpoint study record: WoE_2010-12-05 NPL for PROSPECT Photocatalytic

activity.001 Free radical formation

Administrative Data




Rationale for

reliability incl.

deficiencies


the data evaluation was provided in the full study reports. The photocatalytic activity of the

ZnO samples depends on the particle size as well as the specific surface area and geometrical

shape. The surface treatment is essential to understand the behaviour of the nanomaterial

under consideration.


162

Data source

Reference

Reference

type


source

Testing

laboratory

Report

no.

Owner

company

Company

study no.

Report

date

study

report

National

Physical

Laboratory

2010 Free radical

formation

under

photocatalytic

conditions

National

Physical

Laboratory,

Teddington,

Middlesex,

United

Kingdom

2010-

12-05

Data access





Test guideline



Methods

other: UV-vis spectroscopy


The photocatalytic activity of materials refers to their ability to create electron-hole pairs under light

irradiation, which then generate reactive oxygen species (ROS) such as free radicals (e.g. superoxide

radicals, hydroxyl radicals, etc.), hydrogen peroxide, singlet oxygen, etc. ROS lifetime is generally below

the millisecond range, which makes their detection challenging. One strategy to detect ROS formation is

to use scavenging substances that get oxidised by ROS into stable chemical compound whose

concentration can be detected. A method for assessing the general oxidative activity of the NMs under

irradiation is by using potassium iodide (KI) test. The oxidation of iodide ions results in the production of

tri-iodide ions, whose concentration can be assessed by measuring optical absorbance at 352 nm.UV-vis

spectroscopy for the detection of tri-iodide ions. KI test under irradiated conditionsA 5 M KI (Sigma, St.

Louis, MO) solution in ultra-pure water was freshly prepared; shaking and vortexing was preferred to

sonication to dissolve KI. KI solution was added to the samples of NMs as received after dispersion

(50mg/L), to obtain a typically 1 mL volume sample, with 0.1M KI. 6 x 3 samples were prepared for each

NM/media combination. Additionally, 6 x 3 samples containing 0.1 M KI only and 50mg/L Anatase NMs

(Anatase Nanopowder, Sigma) for each media were prepared as negative and positive controls

respectively; 6 NM samples plus controls were prepared and assessed in total. All samples were contained

in individual 2mL microcentrifuge tubes. Samples were irradiated under a 1kW Solar Simulator (Newport

Corporation, Stratford, CT). The instrument posses a personal wavelength correctionTM Certificate by

Newport. The irradiance of the Solar Simulator was measured to be 1000 Wm-2 using an optical

power/energy meter (Newport, model 842-PE). Irradiation was performed on groups of 40

microcentrifuge tubes. The tubes were placed vertically under the centre of the lamp of the solar

simulator, on an in-house made polystyrene holder, their cups having been removed. The samples were

subjected to 10min periods of irradiation, followed by 5min period of non-irradiation to reduce sample

overheating. After each 10 min period, 1x3 samples for each NM/media combination and controls were


163

removed from the irradiations. Samples irradiated for 0 min, 10 min, 20 min, 30 min, 40 min and 60 min

were collected for each NM/media combination and controls. The samples containing NMs were

centrifuged at 20800 rcf for 15 min and 800 µL of supernatant was collected in a new micro-centrifuge

tube and then analysed using UV-visible spectroscopy (see protocol below). UV-visible spectroscopy for

KI testThe UV-visible spectrum (absorbance scans from 300 nm to 500 nm) was acquired for samples

that were irradiated for 60 minutes. Optical absorbance at 352 nm was acquired for all samples.

Absorption spectra were acquired with a Lambda 850 UV-Vis spectrometer supported by UV Winlab

software [Version 5.1.5] (Perkin Elmer, Waltham, MA). The instrument wavelength calibration was

checked using Holmium glass standards (Serial # 9393, Starna Scientific, Hainault, UK). For the

reference channel of the spectrophotometer a matched cell containing the corresponding dispersing media

(with no nanoparticles) was used. Absorption spectra were acquired on samples that have been irradiated

for 60 minutes. Absorbance scans from 300nm to 500nm were performed, using a slit width of 2 nm and

a scan rate of 50 nm/min. After each sample, the cuvette was cleaned with a 2% solution of Hellmanex

detergent, rinsed with pure water and ethanol and then blow-dried. Optical absorbance at 352nm was

performed using a plate-reader Victor3 1420 multilabel counter (Perkin Elmer), supported by Wallac

1420 software (Perkin Elmer). 300L of each sample (supernatant after centrifugation) was placed in the

wells of a 96-well plate. Only the wells of rows 2 to 6 and columns 1 to 10 were used, as they had the

same level of noise. The absorption at 352nm was measured using a 0.1s measurement time. Measured

absorption values were displayed on a 0 a.u. to 2 a.u. scale.

Data gathering

GLP compliance

no

Test materials


yes


Identifier Identity


EC number 215-222-5

IUPAC name oxozinc


Test material form









Code: NM111

Supplier: BASF SE





164


TM Code: NM112






Code: NM113







The figure compares the absorption measured at 352 nm for all the NM samples in 4 different media after

60 min of total irradiation; the corresponding negative control (i.e. media with no NMs) are also

shown.Results show that there was a certain level of tri-iodide (I3-) measured in the irradiated sample

containing media only. Interestingly, tri-iodide was suppressed in seawater and may be attributed to a

higher concentration of ions (potentially with some scavenging capacity either to ROS species or to

electron (or holes) at the NM surface) in this media. As expected, results for Anatase (TiO2), being the

most active photocatalytic material, show a much higher rate of tri-iodide formation than the

corresponding PROSPEcT NMs. In particular, the absorbance signal was highest in DI water, with the

lack of ionic species in the media. Again, when in seawater, the absorbance signal was reduced (as in the

corresponding blank i.e. seawater with no Anatase). There are several possible explanations for this:

a) Presence of scavengers in solution, as previously described.

b) Enhanced aggregation/sedimentation of the NMs in seawater media compared to other media.

Out of all the PROSPEcT NMs, we see Nanograin CeO2following a similar trend to Anatase, in having

the largest absorbance signal in DI water and the smallest when in seawater. Z-COTE ZnO is interesting,

in that it does not follow a similar pattern observed with Anatase and Nanograin CeO2. With Z-COTE

ZnO, the absorbance signal is much higher in seawater than when dispersed in the other three media. At

present we offer no explanation for this observation. With the other PROSPEcT NMs, the absorbance

signals were within a similar range to that of the corresponding irradiated blank. Samples that were kept

in the dark exhibited no absorption peak at 352 nm.

Lastly, a UV-visible plate reader was used to follow the cumulative production of I3-with varying

irradiation time; this was quantified by measuring absorption at 352 nm. In summary, results show that

absorbance signal generally increases with irradiation time and this can be attributed to the increase in the

amount of ROS being generated. Again, our findings are consistent with previous observations, in that:

a) Anatase gave the highest absorbance reading.

b) Z-COTE ZnO gave a higher absorbance reading in seawater than when in other media



Attached full

study report

NM_110_ANNEX_A45_FREE RADICAL FORMATION UNDER PHOTOCATALYTIC

CONDITIONS.doc / 78.5 KB (application/octet-stream): ENV/JM/MONO(2015)15/ANN1


165



06_Australia_Deakin_Photocatalyticactivity

Administrative Data




Rationale for

reliability incl.

deficiencies


Nevertheless, the study was regarded as scientifically acceptable and was regarded as reliable

with restrictions. This used dye was regarded as sufficient for photocatalytic activity

determination in organic solvent. Furthermore DPPH and Rhodamine are recommended by

OECD 2014

Data source

Reference

Reference

type


source

Testing

laboratory

Report

no.

Owner

company

Company

study no.

Report

date

study

report

Tsuzuki,

Wang

2012 Photocatalytic

activity study

of OECD

nanomaterials

Deakin

University

2012-

03-06

Data access





Test guideline




166

Methods

other: UV-vis


Decomposition of Rhodamine-B dye in the presence of nanomaterial, detected by UV-Vis spectroscopy.


For each measurement, 0.012g of the powder sample was added into 100 ml of Rhodamine B (RhB)

aqueous solution having the concentration of 0.0096g/L. The suspension was placed in a quartz beaker

with a quartz watch glass as a lid. The suspension was then stirred in the dark for 1 h to ensure the

establishment of adsorption and desorption equilibrium of RhB on the particle surface. Subsequently the

suspension was irradiated with simulated sunlight using an Atlas Suntest XLS+ instrument (equipped

with a 150 W xenon lamp) with a flux of 300 wm-2 while continuously stirred. The temperature of the

solution was regulated to 37 °C (air cooling). At given intervals, 3 mL of the suspension was extracted

and then centrifuged at 6000 rpm for 10 min to separate the nanoparticles from the supernatant. UV-Vis

absorbance spectra of the supernatant were measured with a Varian Cary 3E spectrophotometer. The

intensity of the optical adsorption peak around 554 nm was used to monitor the rate of dye degradation.

Each of the photocatalytic activity tests was repeated three times. Overall rates of the reactive dye

degradation were calculated assuming the first-order kinetics. It is assumed that extraction of a small

quantity of sample (3 mL each x max 6) does not alter the UV exposure conditions of the remaining

sample.

Data gathering

Instruments

Varian Cary 3E UV-Vis spectrophotometer:Atlas Suntest CPS+ instrument equipped with a 150 W

Xenon lamp.

GLP compliance

no

Test materials


yes


Identifier Identity


EC number 215-222-5

IUPAC name oxozinc


Test material form




Code: NM110

Supplier: BASF SE





167


Code: NM111

Supplier: BASF SE





TM Code: NM112






Code: NM113






Photocatalytic Activity equation/description

It was found that the photocatalytic activity was the highest for NM110 and NM113 and the lowest for

NM112 (Table 1). MN111 did not show the first order kinetics and hence it was not possible to estimate

the rate constant. The rate constant did not show a strong correlation with specific surface area (Figure 4).

Although NM113 is supposed to be a non-nano material, the photocatalytic activity was as high as

NM110. When the photoactivity rates are normalised with specific surface area values, NM110 showed

the highest photocatalytic activity per unit surface area (Figure 5). The results could be explained as

surface defects acting as charge recombination sites to prevent photocatalysis. NM112 has near spherical

shapes that are expected to have a large number of surface defects. On the other hand, MN110 and

MN113 have geometrical shape particles that indicate high crystallinity and a low number of surface

defects.



Attached full

study report

NM_110_ANNEX_A46_OECD ZnO Photocatalysis Australia Deakin.pdf / 291.74 KB




168


Conclusions

It was found that the photocatalytic activity was the highest for NM110 and NM113 and the lowest for

NM112. NM111 did not show the first order kinetics and hence it was not possible to estimate the rate

constant.

Executive summary

The photocatalytic activity was investigated by the Deakin University, 2012. Photocatalytic activity of the

ZnO samples was determined by monitoring the degradation of Rhodamine B (RhB) in aqueous solutions

having the concentration of 0.0096 g/L. To quantify the photo-reactivity, the absorbance at 554 nm (the

wavelength of maximum absorbance for RhB) was monitored. NM 111 did not show the first order

kinetics due to the presence of a surface coating. The rate constant did not show a strong correlation with

specific surface are. The photocatalytic activity of NM 0113 was as high as NM 110. When the

photoactivity rates are normalised with specific surface area values, NM 110 showed the highest

photocatalytic activity per unit surface area. NM 112 is the sample with the smallest primary particle size

(and highest surface area), and its photocatalytic activity is less than that of its larger counterpart NM 110,

and NM 113 with the largest particle size. The results could be explained as surface defects acting as

charge recombination sites to prevent photocatalysis. NM 112 has near spherical shapes that are expected


169

to have a large number of surface defects. On the other hand, MN110 and MN113 have geometrical shape

particles that indicate high crystallinity and a low number of surface defects results in higher

photocatalytic activities.


06_Australia_CSIRO_Photocatalytic_Activity

Administrative Data




Rationale for

reliability incl.

deficiencies


Nevertheless, the study was regarded as scientifically acceptable and was regarded as reliable

with restrictions. This used dye was regarded as sufficient for photocatalytic activity

determination in organic solvent. Furthermore DPPH and Rhodamine are recommended by

OECD 2014

Data source

Reference

Reference

type


source

Testing

laboratory

Report

no.

Owner

company

Company

study no.

Report

date

study

report

Casey,

Yin,

McCall

2012 Summary of

physicochemical

properties of

OECD Zinc

Oxide (ZnO)

samples -

Photocatalytic

activity

CSIRO 2012-

03-06

Data access





Test guideline



Methods

other: UV-vis spectroscopy


Degradation of 1,1-diphenyl-2-picrylhydrazyl (DPPH) dye in the presence of nanomaterial, detected by

UV-Visspectroscopy.


Photocatalytic activity of the ZnO samples was determined by monitoring the degradation of 1,1-

diphenyl-2- picrylhydrazyl (DPPH), following the procedure described by Dransfield et al (2000). To

quantify the photo-reactivity, the absorbance at 520 nm (the wavelength of maximum absorbance for the

purple DPPH radical) was monitored. Two 62.5ml mixtures of 1:1 Mineral Oil White Light


170

(Aldrich):Caprylic Capric C8/C10 Triglyceride (MOTG) were prepared. ZnO (0.031g) was added to one

mixture, and DPPH (0.0052g) to the other. Each was magnetically stirred for 1.5 hours in a beaker

covered on all sides with Al foil. Then the two mixtures were combined, poured into a crystallising dish

(135mm diameter x 23mm height) covered on all sides with foil, and stirred for 5 minutes. Before

exposure to UV (t=0), 3ml of the solution was withdrawn and its UV-Vis absorption spectrum was

measured using a Cary 5G UV-Vis NIR spectrophotometer. The 3ml sample was returned to the solution.

The solution was then exposed to UV using a pre-warmed Spectroline UV lamp (BIB150 P/FA 365 nm,

150W concentrated spot bulb, lamp diameter=110mm) placed 12cm from the ZnO2/dye mixture. Samples

were taken at various times and absorbance at 520nm was measured. The equipment was designed with a

sliding shield separating the sample and UV lamp so that between exposures the lamp remained on, thus

avoiding variations in lamp intensity. G. Dransfield, P.J. Guest, P.L. Lyth, D.J. McGarvey, T.G. Truscott,

J. (2000) Photochem. Photobiol. B 59 147.

Data gathering

GLP compliance

no

Test materials


yes


Identifier Identity


EC number 215-222-5

IUPAC name oxozinc


Test material form




Code: NM110

Supplier: BASF SE





Code: NM111

Supplier: BASF SE





TM Code: NM112







171

Code: NM113







Figure 1(a) presents typical absorption spectra of DPPH in a dispersion containing NM112, upon

exposure to UV irradiation for various times. With increasing exposure time, the absorbance (C)

reducesfrom its initial level (C0). Figure 1(b) presents the absorbance ratio (C/C0) as a function of time

for DPPH for various samples. The data were further analysed by comparing (i) the decay time (the time

required to bleach the dye) and (ii) the first order rate constant of the radical consumption (A) calculated

by using the linear relationship between ln(c/c0) and irradiation time t.

The decay times and rate constants for the ZnO samples are listed in the Table. A shorter decay time and

larger rate constant corresponds to higher photocatalytic activity. In addition to charge transfer and

subsequent radical formation, recombination of separated electrons and holes in/on the ZnO particle may

be effective in suppressing photocatalytic activity. Theoretically, photocatalytic activity should decrease

with increasing particle size (or decreasing surface area) because recombination processes dominate at

larger sizes. However, the expected correlation between surface area and photocatalytic activity does not

appear to be supported by the data. NM112 is the sample with the smallest primary particle size (and

highest surface area), and yet its photocatalytic activity is less than that of its larger counterpart NM110,

and comparable to NM113 with the largest particle size. For ZnO, it has been reported that as particle size

decreases to the nanoscale (40-80nm) photo-reactivity increases, but observed increases are substantially

lower than those expected from associated increases in surface area. Theoretically effects normally

associated with high surface area may somehow be contributing to this observed deviation. The influence

on photocatalytic activity of the combination of radical surface recombination and surface migration rates

with availability of surface adsorbed species may be greater than that of volume recombination.

A comparison of the data in Figure 1(b) and Table for the BASF-supplied samples NM110 (uncoated)

and NM111 (similar size to NM110 but with a surface coating), indicates that the presence of a surface

coating appears to mitigate, but not eliminate, photocatalytic activity.

Decay times and rate constants of DPPH used as a probe to monitor the photocatalytic activity of the ZnO

samples NM110, NM111, NM112 and NM113.

DPPH as probe

Decay time (min) Rate constant (min-1)

NM110 9 0.124

NM111 11 0.112

NM112 11 0.069

NM113 15 0.089

A shorter decay time and larger rate constant corresponds to higher photocatalytic activity.



Figure 1. (a) Typical absorption spectra of the dye DPPH in a dispersion of NM112 after exposure to UV

irradiation for varying times. (b) The absorbance at time t relative to the initial absorbance at time zero

(C/C0) is plotted as a function of UV exposure time for all four ZnO samples NM110, NM111, NM112

and NM113. Data denoted by DPPH are for the dye alone without ZnO. Data for the photocatalytic P25

(TiO2) are included for comparison.


172


Attached full study

report

NM_110_ANNEX_A47_OECD ZnO photocatalytic activity CSIRO Australia.docx / 95.22

KB (application/octet-stream): ENV/JM/MONO(2015)15/ANN1



173


Conclusions

Theoretically, photocatalytic activity should decrease with increasing particle size (or decreasing surface

area) because recombination processes dominate at larger sizes. However, the expected correlation

between surface area and photocatalytic activity does not appear to be supported by the data obtained for

NM110, NM111, NM112 and NM113. NM110 is more photocatalytic than the other three ZnO samples,

in the system used here.

Executive summary

The photocatalytic activity was investigated in a study conducted by CSIRO, 2012 by monitoring the

degradation of 1,1-diphenyl-2-picrylhydrazyl (DPPH) in mixtures of 1:1 Mineral Oil White Light :

Caprylic Capric C8/C10 Triglyceride (MOTG), following the procedure described by Dransfield et al.,

2000. To quantify the photo-reactivity, the absorbance at 520 nm (the wavelength of maximum

absorbance for the purple DPPH radical) was monitored. The data were analysed by comparing (i) the

decay time (the time required to bleach the dye) and (ii) the first order rate constant of the radical

consumption calculated by using the linear relationship between ln(c/c0) and irradiation time t. A shorter

decay time and larger rate constant corresponds to higher photocatalytic activity. In addition to charge

transfer and subsequent radical formation, recombination of separated electrons and surface defects in/on

the ZnO particle may be effective in suppressing photocatalytic activity. Theoretically, photocatalytic

activity should decrease with increasing particle size (or decreasing surface area) because recombination

processes dominate at larger sizes. However, the expected correlation between surface area and

photocatalytic activity was not supported by the data. NM-112 is the sample with the smallest primary

particle size (and highest surface area), and yet its photocatalytic activity is less than that of its larger

counterpart NM-110, and comparable to NM-113 with the largest particle size. The influence on

photocatalytic activity of the combination of radical surface recombination and surface migration rates

with availability of surface adsorbed species may be greater than that of volume recombination. A

comparison of the data for the samples NM-110 (uncoated) and NM-111 (similar size to NM-110 but

with a surface coating), indicates that the presence of a surface coating appears to mitigate, but not

eliminate, photocatalytic activity.

4.35 Radical formation potential

Endpoint study record: RL4_2010-12-09_NPL for PROSPECT_Radical

formation potential_Photocatalytic conditions

Administrative Data





incl. deficiencies

Limited documentation of the full study report was available. Based on the

provided data no further conclusion could be achieved.


174

Data source

Reference

Reference

type


source

Testing

laboratory

Report

no.

Owner

company

Company

study no.

Report

date

study

report

National

Physical

Laboratory

2010 Interim Report

on the Physico-

Chemical

Characterisation

of PROSPEcT

Nanomaterials.

unpublpished

report

National

Physical

Laboratory

Hampton

Road,

Teddington,

Middlesex,

TW11 0LW

2010-

09-01

Data access





Test guideline




The photocatalytic activity of materials refers to their ability to create electron-hole pairs under light

irradiation, which then generate reactive oxygen species (ROS) such as free radicals (e.g. superoxide

radicals, hydroxyl radicals, etc.), hydrogen peroxide, singlet oxygen, etc.


- Irradiation was performed on groups of 40 microcentrifuge tubes. The tubes were placed vertically

under the centre of the lamp of the solar simulator, on an in-house made polystyrene holder, their cups

having been removed. - The samples were subjected to 10min periods of irradiation, followed by 5min

period of non-irradiation to reduce sample overheating.- After each 10 min period, 1x3 samples for each

NP/media combination and controls were removed from the irradiations.- Samples irradiated for 0 min,

10 min, 20 min, 30 min, 40 min and 60 min were collected for each NP/media combination and controls.-

The samples containing NPs

Data gathering

Instruments

1kW Solar Simulator (Newport Corporation, Stratford, CT)Lambda 850 UV-Vis spectrometerPlate-reader

Victor 1420 multilabel counter (Perkin Elmer)

Calibration

The instrument wavelength calibration was checked using Holmium glass standards (Serial # 9393,Starna

Scientific,Hainault, Uk)

GLP compliance

no


175

Test materials


yes


Identifier Identity


EC number 215-222-5

IUPAC name oxozinc


Test material form




Code: NM110

Supplier: BASF SE





Code: NM111

Supplier: BASF SE





TM Code: NM112






Code: NM113







Absorbance readings at 352 nm after being irradiated with solar simulator at 1000 W/m2, for 60 minutes.



Attached full

study report

NM_110_ANNEX_A45_FREE RADICAL FORMATION UNDER PHOTOCATALYTIC

CONDITIONS.doc / 78.5 KB (application/octet-stream): ENV/JM/MONO(2015)15/ANN1

Attached full

study report




176


Executive summary

NPL, 2010 investigated the radical formation potential. The test items were dispersed in four different

media in the presence of KI (deionized water, seawater, daphnia and fish media) and the dispersions were

exposed for 60 minutes, under 1000 W/m2 white light irradiation. Anatase (TiO2) was used as positive

control. Results showed that there was a certain level of tri-iodide (I3-) measured in the irradiated sample.

Tri-iodide was suppressed in seawater and may be attributed to a higher concentration of ions in this

media. Results for Anatase (TiO2), being the most active photocatalytic material, show a much higher

rate of tri-iodide formation than the corresponding zinc oxide NMs. In particular, the absorbance signal

was highest in deionized water, lower in daphnia and fish media and when in seawater, the absorbance

signal was reduced (as in the corresponding blank i.e. seawater with no Anatase). It was shown that the

absorbance signal of NM 111 is much higher in seawater than when dispersed in the other three media. In

the course of the present study no explanation was found. In general, for the ZnO nanomaterials the

absorbance signals were within a similar range to that of the corresponding irradiated blank (Samples that

were kept in the dark exhibited no absorption peak at 352 nm) indicating no differences Radical

formation potential between bulk and nano ZnO.

ENV /JM /M ON O(2015)15/P ART1 Unclassified - OECD

Documents

Transcript of ENV /JM /M ON O(2015)15/P ART1 Unclassified - OECD