Meeting Summary Report

Material Characterization of Nanoscale Materials

(September 6 and 7, 2007

Peer Consultation Meeting)

Prepared for:

U.S. Environmental Protection Agency

Office of Prevention, Pesticides and Toxic Substances

1200 Pennsylvania Avenue, NW

Washington, D.C. 20460

Prepared by:

Eastern Research Group, Inc.

14555 Avion Parkway 

Suite 200

Chantilly, VA 20151-1102

October 15, 2007

EPA Contract No. EP-W-05-014

Work Assignment 3-06

TABLE OF CONTENTS

Page

  TOC \o "1-3" \h \z    HYPERLINK \l "_Toc179695393"  1.0	Introduction	 
PAGEREF _Toc179695393 \h  1-1  

  HYPERLINK \l "_Toc179695394"  1.1	Panel Members	  PAGEREF
_Toc179695394 \h  1-1  

  HYPERLINK \l "_Toc179695395"  1.2	Meeting Background and Purpose	 
PAGEREF _Toc179695395 \h  1-2  

  HYPERLINK \l "_Toc179695396"  2.0	Day One Summary (September 6, 2007)	
 PAGEREF _Toc179695396 \h  2-1  

  HYPERLINK \l "_Toc179695397"  2.1	Introductions	  PAGEREF
_Toc179695397 \h  2-1  

  HYPERLINK \l "_Toc179695398"  2.2	General Discussion	  PAGEREF
_Toc179695398 \h  2-3  

  HYPERLINK \l "_Toc179695399"  2.3	Session 1: Discussion – Types of
Nanoscale Materials &

Their Structure and Chemical Composition	  PAGEREF _Toc179695399 \h  2-4
 

  HYPERLINK \l "_Toc179695400"  2.3.1	Question 1:  Are there any other
significant categories, based on structure and chemical composition,
that should be included in this discussion because they are
substantially different from the categories mentioned (e.g., hybrids,
self- assembly devices, others)?	  PAGEREF _Toc179695400 \h  2-4  

  HYPERLINK \l "_Toc179695401"  2.3.2	Question 2:  For the different
categories of nanoscale materials, what is the current state of
knowledge about structure and

chemical composition?	  PAGEREF _Toc179695401 \h  2-7  

  HYPERLINK \l "_Toc179695402"  2.3.3	Question 3:  Can structures and
chemical composition be correlated to specific properties and is this
correlation

quantifiable?  Chemical Identification Elements (e.g.,

which ones? for which substances? test methods?)	  PAGEREF _Toc179695402
\h  2-9  

  HYPERLINK \l "_Toc179695403"  2.3.4	Question 12:  How important are
impurity identity and

impurity levels to the understanding and characterization of nanoscale
materials?	  PAGEREF _Toc179695403 \h  2-10  

  HYPERLINK \l "_Toc179695404"  2.3.5	Question 13:  Are there routine
purification procedures that can effectively control or remove
impurities, when desirable, for certain classes of nanoscale materials?	
 PAGEREF _Toc179695404 \h  2-11  

  HYPERLINK \l "_Toc179695405"  2.4	Session 2- Discussion:  Physical
Chemical Properties	  PAGEREF _Toc179695405 \h  2-11  

  HYPERLINK \l "_Toc179695406"  2.4.1	Question 4:  Which
physical-chemical properties are

relevant to nanoscale materials and how?  Which are

known or reasonably ascertainable and which have

data gaps?	  PAGEREF _Toc179695406 \h  2-11  

  HYPERLINK \l "_Toc179695407"  2.4.2	Question 5:  Are there properties
that would have little or no relevance under the NMSP?	  PAGEREF
_Toc179695407 \h  2-13  

  HYPERLINK \l "_Toc179695408"  2.4.3	Question 6:  Which properties are
associated with

aggregated or agglomerated nanoscale materials, as opposed to properties
that are inherent to the material regardless of

physical form?	  PAGEREF _Toc179695408 \h  2-14  

  HYPERLINK \l "_Toc179695409"  2.4.4	Question 7:  Are there routine
manipulations of nanoscale materials that result in physical-chemical
properties changes or other defining characteristics (e.g., surface
modifications of nanotubes to enhance solvent dispersibility)?	  PAGEREF
_Toc179695409 \h  2-14  

  HYPERLINK \l "_Toc179695410"  3.0	Day two Summary (September 7, 2007)
  PAGEREF _Toc179695410 \h  3-1  

  HYPERLINK \l "_Toc179695411"  3.1	Session 2 - Discussion:  Physical
Chemical Properties (Continued)	  PAGEREF _Toc179695411 \h  3-1  

  HYPERLINK \l "_Toc179695412"  3.1.1	Question 8:  How should
physical-chemical property data be prioritized for the NMSP?  Based on
availability, effect

(toxicity or exposure criteria), or other factors?	  PAGEREF
_Toc179695412 \h  3-1  

  HYPERLINK \l "_Toc179695413"  3.2	Session 3 - Discussion:  Nanoscale
Manufacture and Processing	  PAGEREF _Toc179695413 \h  3-2  

  HYPERLINK \l "_Toc179695414"  3.2.1	Question 9:  What are the common
processes used to

manufacture nanoscale materials?	  PAGEREF _Toc179695414 \h  3-2  

  HYPERLINK \l "_Toc179695415"  3.2.2	Question 10:  How are processes
used to produce specific characteristics or properties?	  PAGEREF
_Toc179695415 \h  3-3  

  HYPERLINK \l "_Toc179695416"  3.2.3	Question 11:  Which methods reduce
particle size but do

not result in property changes?  Which methods reduce

particle size and result in property changes?	  PAGEREF _Toc179695416 \h
 3-3  

  HYPERLINK \l "_Toc179695417"  3.3	Session 4 - Discussion: 
Methodologies for Characterizing

Nanoscale Materials Analytical Methods	  PAGEREF _Toc179695417 \h  3-4  

  HYPERLINK \l "_Toc179695418"  3.3.1	Question 14:  Are validated
methods available for the

different categories of nanoscale materials?	  PAGEREF _Toc179695418 \h 
3-4  

  HYPERLINK \l "_Toc179695419"  3.3.2	Question 15:  Are there techniques
that can be

universally applied?	  PAGEREF _Toc179695419 \h  3-4  

  HYPERLINK \l "_Toc179695420"  3.3.3	Question 16:  For small quantities
of materials, are there

sampling, handling, and collection techniques as well as

sample integrity, accuracy and precision QA/QC

methodologies available?	  PAGEREF _Toc179695420 \h  3-5  

  HYPERLINK \l "_Toc179695421"  3.3.4	Question 17:  What is the status
of standardization efforts?

Are these efforts focused on broadly applicable

characterization methods or category-specific methods?	  PAGEREF
_Toc179695421 \h  3-5  

  HYPERLINK \l "_Toc179695422"  3.3.5	Question 18:  What alternative or
innovative methods or technologies can be applied to nanomaterial
analysis?	  PAGEREF _Toc179695422 \h  3-6  

  HYPERLINK \l "_Toc179695423"  3.3.6	Question 19:  Are there models
that are currently used to

obtain property data for nanoscale materials?  For which

properties and which nanoscale materials?	  PAGEREF _Toc179695423 \h 
3-6  

  HYPERLINK \l "_Toc179695424"  3.3.7	Question 20:  Has any validation
work been conducted that compares predicted values with measured data? 
For which properties and which nanoscale materials?	  PAGEREF
_Toc179695424 \h  3-7  

  HYPERLINK \l "_Toc179695425"  3.3.8	Question 21:  Are there current
significant characterization

needs for which the NMSP should investigate model

development?	  PAGEREF _Toc179695425 \h  3-7  

  HYPERLINK \l "_Toc179695426"  3.4	Session 5 - Discussion: 
Prioritization of Data Gaps	  PAGEREF _Toc179695426 \h  3-8  

  HYPERLINK \l "_Toc179695427"  3.5	Closing Remarks	  PAGEREF
_Toc179695427 \h  3-9  

 

Appendix A:	MEETING AGENDA

Appendix B:	DISCUSSION PAPER

Appendix C:	PRELIMINARY PANEL COMMENTS ON DISCUSSION PAPER

Introduction

	This report summarizes the Panel discussions and public comments during
the Peer Consultation Public Meeting on The Material Characterization of
Nanoscale Materials organized by the U.S. Environmental Protection
Agency (EPA).  The meeting took place in Arlington, VA, on September 6
and 7, 2007.  Eastern Research Group, Inc. (ERG), a contractor to EPA,
organized the logistics, provided facilitation support, and prepared
this summary report.  Meeting minutes were not prepared and a transcript
was not recorded.  The intent of this report is to provide an overview
of the discussion that occurred.  No attempt has been made to analyze or
evaluate any portion of the discussions.  The discussion and comments
presented in this summary reflect individual opinions of the commenters
and should not be considered to be the opinion or belief of EPA.

Panel Members

Dr. Joseph BelBruno, Professor, Department of Chemistry, Dartmouth
College.

Dr. Ahmed Busnaina, William Lincoln Smith Professor and Director, NSF
Nanoscale Science and Engineering Center for High-rate Nanomanufactruing
and the NSF IURC Center for Microcontamination Control, Northesastern
University.

Dr. Richard Canady, DABT, Office of the Commissioner, Food and Drug
Administration.

Dr. Alan Cassell, Project Scientist/Task Manager, Ames Research Center,
National Aeronautics and Space Association.

Dr. Vicki Colvin, Director, Center for Biological and Environmental
Nanotechnology (CBEN), Rice University (participated on Day 2, September
7th only).

Dr. Joerg Lahann, Professor, Department of Chemical Engineering,
University of Michigan (participated on Day 1, September 6th only).

Dr. Vladimir Murashov, Special Assistant to the Director, National
Institute for Occupational Safety and Health (NIOSH).

Dr. James Murday, Associate Director for Physical Sciences, Office of
Research Advancement, University of Sourthern California.

Dr. John Small, Group Leader, Center for Nanoscale Science and
Technology, National Institutre of Standards and Technology (NIST).

Dr. David Warheit, DuPont Haskell Laboratory.

Meeting Background and Purpose

	

	As part of EPA’s initiative to address growing interest in the
potential health and safety issues of engineered nanoscale materials
(NMs), EPA is developing the voluntary Nanoscale Materials Stewardship
Program (NMSP).  This program is being designed to encourage responsible
commercial development of NMs.  The NMSP will enable EPA, the relevant
industry, and interested stakeholders to better assess potential risks
to human health and the environment from NMs and to identify risk
management practices which may reduce such potential risks.

	The NMSP is intended to include but is not limited to engineered NMs
manufactured or imported for commercial purposes as defined in 40 CFR
720.3 (r).  The NMSP is envisioned to have two parts: 1) a basic
reporting program and 2) an in-depth program.  This public peer
consultation meeting is intended to support the NMSP by identifying
material characterization data that participants could submit under the
Basic Program if they are known or are reasonably ascertainable as
defined in 40 CFR 720.3 (p).  The data and experience generated by the
basic reporting phase will help to inform the types of in-depth data to
be developed.  

	The Scientific Peer Consultation Public Meeting on The Material
Characterization of Nanoscale Materials was organized to help clarify
which data and elements should be included in the NMSP basic and/or
in-depth program (see Appendix A for the meeting agenda).  The goal was
to have an applied discussion that considers 1) the current
understanding of material characterization as it relates to NMs and 2)
how this understanding can be used to guide the Agency’s thinking
regarding the material characterization data elements that would be most
useful and important to include in the NMSP

	Prior to the meeting, EPA developed a paper to facilitate discussion on
topics related to material characterization (Appendix B).  EPA then
convened a Panel to review the discussion paper.  The Panelists reviewed
the discussion paper and submitted comments prior to the meeting, which
were summarized and circulated for review to aid the meeting discussions
(Appendix C).  The requested approach for the Panel during the meeting
was to discuss what data are known or reasonably ascertainable to
characterize NMs.  This was then to be followed by a discussion on the
methodology used to obtain and use characterization endpoints of
interest.  There was no attempt prior to, or during the meeting, to
arrive at joint decisions, reach consensus, or provide majority advice.

	The meeting was organized into a series of sequential sessions over a
two-day period.  After initial introductions from EPA, an open,
facilitated discussion occurred regarding the questions posed to the
Panel in the topical areas per the discussion paper.  After Panelists
concluded their discussion, meeting observers were asked to provide
comment.  As time permitted, the Panel then responded or continued
discussion based on the public comments.

	Additional information pertaining to the specific charge to the Panel,
a summary of their written responses prior to the meeting, and the
meeting agenda can be found in the public docket and appendices of this
report 

Day One Summary (September 6, 2007)

Introductions

	Opening Remarks and Background

	Jim Alwood (EPA, Office of Pollution Prevention and Toxics (OPPT),
Chemical Control Division).  Mr. Alwood greeted Panelists and members of
the public.  Mr. Alwood stated that the purpose of the meeting was to
solicit input from individual Panelists and the public.  EPA would then
consider all input, but there would be no specific attempt during the
meeting to come to consensus on any discussion topics.

	

	Greg Fritz (EPA, OPPT, Economics, Exposure, and Technology Division
(EETD)).  Dr. Fritz addressed the group and summarized the scope of the
meeting.  He stated that the Panelists had been asked to review a
document titled, “Peer Consultation Public Meeting on The Material
Characterization of Nanoscale Materials:  Discussion Topics, August 7,
2007” prior to the meeting and consider the corresponding series of
questions.  Dr. Fritz stated that the Panel discussions surrounding the
questions during the meeting would help inform EPA as the Nanoscale
Materials Stewardship Program (NMSP) is structured, specifically
regarding the data elements that may be requested for participation in
the basic and in-depth programs.

	Dr. Fritz then summarized the content of the discussion paper and the
specific questions that EPA asked the Panelists.

	Nanoscale Materials Stewardship Program Overview and Status

	Jim Willis (EPA, Office of Pollution Prevention and Toxics; Director,
Chemical Control Division).  Mr. Willis provided additional remarks to
set the stage for the meeting, including a background and history on
EPA’s effort to develop the NMSP and how material characterization
fits into the program.

	He stated EPA’s intent is to develop and implement the NMSP in an
open and transparent manner.  The meeting is the third in a series of
public meetings held to solicit input on various components of The
Program.  The first was held on October 19th and 20th, 2006 regarding
Risk Management Practices; and the second was held August 2nd, 2007 to
discuss EPA’s, “Concept Paper for the Nanoscale Materials
Stewardship Program under TSCA” and the “TSCA Inventory Status of
Nanoscale Substances - General Approach”.

	Mr. Willis also discussed the overall status of the NMSP.  EPA is
currently in the “design phase” of the program.  The current meeting
is intended to help finalize the structure and design.  After completion
of this first phase, the NMSP will be implemented such that participants
will submit information to EPA.  After the “implementation phase”
has been functional for a time period (still to be determined), an
“evaluation phase” will begin.  During the evaluation, EPA will
review data submitted, modify the program as appropriate, and provide a
summary to the public.

	Facilitated Introductions and Charge to Panelists

	Jan Connery (ERG) facilitated the meeting.  The Panelists were first
asked to briefly introduce themselves.  After introductions, Ms. Connery
stressed to all participants and observers that there would be no
attempt to reach consensus during the discussion.  Rather, the intent of
the meeting is to provide EPA with input from a broad, scientific
stakeholder community on the various discussion topics.  All comments
and suggestions were to be recorded as individual opinions for EPA to
consider while proceeding with development of the NMSP.  However, to aid
EPA’s evaluation, the meeting summary document would note instances
when multiple Panelists concurred on any topic.

	Ms. Connery then noted six general topical areas that were not intended
for discussion during the meeting. 

EPA Policy

Classification and Nomenclature

Environmental Fate or Transport

Risk Management or Hazard Assessment

Solutions to Perceived Data Gaps

Assessment of the NMSP or other EPA programs

	Dr. Canady asked for clarification regarding the topic of nomenclature.

	Tracy Williamson (EPA, OPPT, EETD; Chief, Industrial Chemistry Branch)
provided a response.  Dr. Williamson agreed that a standardized
classification scheme and standardized nomenclature are important and
relevant.  However, she noted that there are ongoing efforts to discuss
these topics in great detail.  Therefore, EPA desired not to devote
significant portions of the Material Characterization meeting to address
nomenclature, recognizing that there may be some topics that require
some discussion.

Public Comment and Q/A Session

	Public Comment #1:  Scott Sweeney (Safer Nanomaterials and
Nanomanufacturing Initiative, SNNI) provided a prepared statement that
included three primary points, briefly summarized below.  

He noted his organization has developed a library with information
pertaining to nanomaterials and suggested this type of repository may be
very useful.

He noted that the purity of nanomaterials is important and applicable to
environmental assessments.  He also provided examples of test data that
may show purity has a considerable toxicological impact.

His organization has access to “rapid characterization methods” to
evaluate purity, noting that electron microscopy has been used
historically; but it is not useful data in many circumstances.  Mr.
Sweeney suggested the use of other techniques such as nuclear magnetic
resonance spectroscopy (NMR), Thermogravimetric Analysis (TGA), and
X-ray photoelectron spectroscopy (XPS).

Mr. Sweeney’s statements resulted in extensive dialogue from Panelists
regarding the test data and SNNI’s characterization methods.  Mr.
Sweeney stated SNNI is developing a data base that could be made
available upon request.  Mr. Sweeney was asked to provide written
comments to EPA.

	Public Comment #2:  Shekhar Subramoney (DuPont Company) asked EPA to
clarify the rationale for excluding discussions regarding hazard
assessment from this meeting.

	Dr. Williamson noted that NPPTAC recommended a public meeting
specifically devoted to material characterization, which she
acknowledged helps inform risk management and hazard assessments.  Dr.
Williamson also reminded the audience of a previous meeting (held
October 19th and 20th, 2006) that was focused on the Risk Management of
Nanomaterials.  Portions of that meeting were specifically devoted to
discussing hazard assessments.  

	Dr. Williamson also clarified that EPA was interested in all media
(e.g., dry state, air, water) regarding material characterization.

	Public Comment #3:  Shaun Clancy (American Chemistry Council, ACC))
requested that EPA clearly state the test methods that will be required
of NMSP participants to provide appropriate data when the NMSP is
ultimately developed.

	

General Discussion

	Ms. Connery asked each Panelist to provide their general thoughts and
comments on the discussion paper and issues surrounding material
characterization of nanomaterials before delving into the detailed
session topics.

	Dr. Warheit commented that the questions posed in the discussion paper
were generally appropriate.  He also noted that from a toxicological
view, many people have attempted to generate lists of appropriate tests
for characterizing nanomaterials.  His view is that this exercise
typically results in an extensive list that is too long because a great
number of tests will provide potentially-useful information.  One reason
is the nearly infinite number of potential nanomaterials and a
corresponding infinite number of uses.  He suggested developing a
smaller, prioritized list of required information and the tests that are
needed to provide that information.  Dr. Warheit provided a suggested
list in his pre-meeting comments.

	Dr. Warheit also stated his belief that purity of the nanomaterial
being evaluated is very important.

	Finally, Dr. Warheit discussed his view that exposure assessments of
nanomaterials are highly dependent on:

Whether the material is in a wet or dry state;

The composition of the matrix being evaluated; and,

Whether assessments are conducted on the core or the shell of the
nanomaterial

	Dr. Canady asked for clarification from EPA regarding the rationale for
establishing the categories that were posed as starting points for
grouping nanomaterials in the discussion paper.

	Dr. Williamson responded that the categories were established to
facilitate discussion and that they could be altered if appropriate. 
After additional discussion, Dr. Williamson acknowledged that the Toxic
Substances Control Act (TSCA) does not include categories of chemicals
on the TSCA Inventory.  She noted, however, that categorizing chemicals
sometimes aids EPA’s internal review process when determining how to
assess and evaluate chemicals.  Therefore, it may be useful to establish
categories.

	Dr. Busnaina noted that he would have suggested different categories
because the existing categories were developed for macro materials based
on the distinguishing characteristics of their properties.  He stated
that these characteristics are not appropriate for distinguishing
between different nanomaterials.  Dr. Busnaina suggested developing
categories based on:  chemicals with inert surfaces, active surfaces,
and functionalized surfaces.

	Dr. Busnaina also stated that EPA should recognize qualifying
terminology such as “simple structures” may not be appropriate.  As
an example, he noted that even discussing nanoscale titanium dioxide can
be very complex.

	Dr. Murday initiated a clarifying discussion regarding composites that
include a macromaterial core and a nanomaterial attached to the surface.

	Dr. Williamson stated the group may not want to exclude these
materials, but she suggested the discussion should focus on
free-standing nanomaterials.

Session 1: Discussion – Types of Nanoscale Materials & Their Structure
and Chemical Composition

	Ms. Connery opened this session by stating that there were five related
questions in EPA’s discussion paper that would be discussed (questions
1, 2, 3, 12, and 13).  She reviewed each question and then asked
Panelists to discuss their responses.  The questions and subsequent
discussions are presented below.

Question 1: 	Are there any other significant categories, based on
structure and chemical composition, that should be included in this
discussion because they are substantially different from the categories
mentioned (e.g., hybrids, self- assembly devices, others)?

	Dr. Murday suggested that using the term “molecule” when discussing
nanomaterials may not be appropriate.  He commented that this term
implies uniformity, whereas many nanomaterials are not uniform and are
not based on a specific chemical formula (for example, the number of
atoms in a gold nanomaterial is flexible).  Dr. Murday agreed with the
comments provided by Dr. Murashov in his pre-meeting comments.

	Dr. Murashov reiterated the statements made in his pre-meeting
comments.  He noted a particular concern regarding “purity”.  He
noted that if a nanomaterial were defined based on a chemical structure,
EPA would need to determine an acceptable “level of purity”.  For
example, EPA would need to specify a range regarding the number of atoms
for various nanomaterials.

	Dr. Murashov also suggested that EPA should consider the difference in
chemical composition between the core and the surface of the
nanomaterial (and the associated purity of these two different regions
of the nanomaterial).  He noted that this is important because the
effects based on the content of the core are different than those based
on the surface.  The surface chemistry may result in more immediate
concerns related to exposure; however, the potential hazards associated
with the core are also important because the surface may eventually
degrade.

	Dr. Murashov suggested that if EPA must develop categories and
corresponding characterization, the Agency could group nanomaterials
based on the structure of the core, recognizing surface chemistry and
function will have effects.

	Dr. Murashov pointed out inconsistencies in the proposed grouping into
categories of nanomaterials described as “organic” and
“inorganic”.  He suggested “carbonaceous” is a more suitable
category to describe carbon-based nanomaterials, noting that pure carbon
is not “organic”.

	Dr. Canady reviewed his written, pre-meeting comments.  He then
expanded on a few topics.  First, Dr. Canady recognized the need to
develop categories.  However, he proposed that EPA recognize that the
categories could be modified in the future as more information is
acquired (possibly modifying the definition of existing categories, or
adding and deleting categories).

	Dr. Canady suggested EPA consider nanomaterials that may exist as
“combinations” of the existing categories (he suggested some could
fall into multiple categories as they are written). 

	Dr. Canady also suggested a separate category for emulsions, vesicles,
and micelles.  He stated they are omitted from the categories in EPA’s
discussion paper; however, there are drug applications that may
eventually be utilized in industry.

	Dr. Joerg Lahann agreed with Dr. Canady’s comments regarding
emulsions, vesicles, and micelles.  He also agreed with the comment
regarding the potential for “combinations”, and suggested
development of a “hybrid” category (or categories, if appropriate).

	Dr. Lahann stated that the effects of nanomaterials are often dependent
upon functionality; not their specific size, structure, or composition.

	Dr. BelBruno commented that it is appropriate to develop initial
categories based on the chemical structure; however, functionality is
very important and should be considered for future categorization.

	Dr. Cassell stated that organometallic materials should be a separate
category.  This initiated a discussion regarding the potential effects
from a metal atom that is surrounded by various functional groups. 
Various Panelists noted potential concerns depending on the specific
metal and the functional groups.

	Dr. Canady suggested that organometallic materials could be a
sub-category or could be a “hybrid” as previously discussed.

	Dr. Murday suggested the following categories and subcategories:

Free-Standing nanomaterials

Materials whose core is always capped

Materials whose core consists of a uniform shell (e.g., quantum dots)

Hybrid materials

those with multiple shells

those with multiple nanostructures linked together 

Nanomaterials embedded in a composite material

Thin-film nanostructures (e.g., those used in electronic devices)

	Dr. Murday suggested this categorization would allow analysis in a
hierarchy from simple to more complex.

	Dr. Lahann agreed with Dr. Murday’s proposed categories.

	Dr. Murashov also agreed.

EPA Response, Public Comment, and Q/A Session

	Public Comment #1: Dan Ewart (NanoProducts Corp.) noted that TSCA is
based on chemical structures.  Therefore, he commented the EPA may need
to consider implications of grouping nanomaterials in categories that
are not based on chemical formulae.

	Dr. Williamson reiterated her previous remarks that categories of
chemical are not included on the TSCA inventory, but EPA utilizes
categories (based on structure) to assist internal review.

	She also presented a brief overview of EPA’s current vision for the
NMSP.  This vision includes the need to understand how to characterize
nanomaterials based on chemical identification information and
physical-chemical properties and the test methods needed to acquire the
characterization data.  Dr. Williamson recognized that the list of
properties and tests may be lengthy; therefore, EPA is hoping to receive
input on how to prioritize the most important, needed information.

	Public Comment #2: Mark Herwig (General Electric) asked if other
organizations had already developed categories for nanomaterials.  This
question resulted in significant discussion regarding other
categorization attempts.

	Mr. Willis specifically discussed current activities of the
Organization of Economic Cooperation and Development (OECD) Working
Party on Manufactured Nanomaterials and of the International
Organization for Standardization (ISO) Technical Committee (TC) that
promotes international standardization in nanotechnology (TC229). 

	Dr. Canady reiterated his view that it is appropriate to establish
categories for initial discussion and evaluation.  However, EPA should
allow for category modification as more information becomes available.

Question 2: 	For the different categories of nanoscale materials, what
is the current state of knowledge about structure and chemical
composition?  

	Dr. Small indicated that the structure and composition of some
nanomaterials is known and can be measured “very accurately.” 
However, a problem occurs when nanomaterials are purchased from
manufacturers.  He noted that the actual composition, core, and
sometimes functional groups vary from batch to batch; even when product
specifications are well-defined.

	Other Panelists agreed and brought up this as an issue during several
discussions throughout the remainder of the meeting.

	Dr. Warheit commented that from a toxicological perspective the
structure and composition is well understood for specific
nanomaterial-types that have been studied; however, this is not
necessarily true for nanomaterials as a whole.

	Dr. BelBruno agreed with and expanded upon Dr. Small’s assertion that
there are differences between batches of the same material.  He noted
that given a sample, laboratories can conduct tests to characterize the
nanomaterial.  However, the results may only apply to that particular
batch.

	Dr. Murashov stated that there are only two examples of nanomaterials
he feels are well-characterized (fullerenes and dendrimers).  He noted
that even titanium dioxide has multiple states and is not completely
characterized.

	Dr. Busnaina commented that it may not be possible to acquire all the
information needed to fully characterize each nanomaterial from a
particular supplier.  He questioned whether the suggested categories are
appropriate.  Dr. Busnaina then suggested selecting a few properties to
prioritize the characterization (such as surface activity) then conduct
follow-up evaluations with secondary properties (such as oxidation
potential).

	Dr. Murday commented that nanomaterials are very complex; therefore,
the proposed properties for characterization may not apply.  For
example, he noted that the core is often very different in its chemistry
and functionality than the surface.  Therefore, he suggested it may be
more appropriate to consider properties of concern rather than the
actual chemical structure.

	Dr. Murashov agreed with Dr. Murday’s previous comment.  He also
agreed with Dr. Small’s comment about variability between batches.

	Dr. Lahann suggested that although there may be variability between
batches, there may be appropriate ways to characterize the purity based
on statistical analyses.  He commented on current methods of
characterizing uncertainty when evaluating the purity of polymers.

	Dr. Canady observed that based on other Panelist’s comments regarding
this question it seemed there is some level of knowledge regarding
certain specific nanomaterials; however, it is not appropriate to state
there is significant knowledge for nanomaterials in general and there is
no knowledge for most.

EPA Response, Public Comment, and Q/A Session

	Dr. Williamson provided comments on the Panel discussions.  She stated
the discussion was very informative and asked Panelists and observers to
provide additional information regarding the variability between
batches.  Dr. Williamson also asked Panelists to elaborate on the
specific characteristics being studied (e.g. for metal oxides). 
Finally, she asked if there are certain characteristics that are more
important than others for characterization.  The following summary of
discussions ensued.

 

	Dr. Murashov indicated that X-Ray crystallography is commonly conducted
on metal oxides.

	Dr. Warheit agreed with Dr. Murashov, but noted that this was typically
done to evaluate the core.  He noted there are different grades of TiO2
for which his laboratory had tested (and published information).  His
work considered different formulations for first time (as far as he was
aware).  He also stated there are different degrees of surface coatings,
which may not be common knowledge.

	Dr. Murashov addressed another question from Dr. Williamson by stating
he felt primary data needs to include a measurement of the distribution
of material in the sample, the nanomaterial’s size, and its shape.

	Dr. Fritz asked Panelists to elaborate on their experience with
variability between samples.

	Dr. Busnaina confirmed this was a significant issue, elaborating on his
specific experiences.  For example, he observed considerable variability
when purchasing nanomaterials with the same product specifications from
the same supplier.  He specifically noted that processing steps that a
manufacturer believes are routine can change certain properties and
chemical characteristics of nanomaterials.

	Dr Murashov then initiated a discussion regarding “surface
activity”.  He commented that although surface activity is an
important consideration, it was not clear how to define or measure this
characteristic (e.g., redox potential or something else?).

	Dr. Fritz asked Panelists to elaborate on potential differences or
concerns regarding tests conducted on dry nanomaterials vs. emulsions.

	Dr. Warheit initiated an extended discussion among multiple Panelists
and EPA representatives on the importance in recognizing differences in
nanomaterials in dry vs. wet matrices.  Differences will include the
toxicological and environmental effects as well as the most important
characteristics to measure the effects and the corresponding measurement
techniques.

	Public Comment #1:  Shekhar Subramoney (DuPont Company) commented that
his experience is that analytical tools are available to identify the
structure and composition of nanomaterials.  He agreed there is
variability between batches, but disagreed with most Panelists’
assertion that this is a primary concern.

Question 3: 	Can structures and chemical composition be correlated to
specific properties and is this correlation quantifiable? Chemical
Identification Elements (e.g., which ones? for which substances? test
methods?)

	Dr. Warheit referenced a paper that was recently published on the lung
toxicity of pristine and fully hydroxylated fullerenes.  The original in
vitro study showed increases in toxicity of at least three orders of
magnitude between the pristine (more toxic) and the fully hydroxylated
fullerene samples.  However, in the in vivo pulmonary toxicity study,
few toxicity differences were measured between the two samples.  The
different results between the two studies indicate that the structure
and chemistry and other in vitro analyses may not be predictive of the
inherent toxicity of the nanomaterial and that each sample has to be
evaluated on a case-by-case basis.

	Dr. Murday felt there is a correlation between structures and chemical
composition, and properties.  He also felt computer modeling based on
structure and composition could be developed to predict properties.  Dr.
Murday also suggested that EPA contact industry stakeholders that
produce catalysts regarding their work on this topic.  Finally, Dr.
Murday noted that we still do not fully understand everything that
impacts a living system, whether or not the material is nanoscale.  This
may affect correlations.

	Dr. Murashov referred to his pre-meeting comments, reiterating them and
stating he felt there were correlations.

	Dr. Lahann also agreed it is possible to develop models and
correlations once data are obtained.  However, he noted there are still
many unknowns.

	Additional discussion between Panelists followed these specific
remarks.  Multiple Panelists stated that theoretical modeling could be
conducted; however, the model goals would need to be clearly defined. 
Further, Panelists acknowledged modeling results would need to be
validated and tested for scale-up to production levels.

Question 12: 	How important are impurity identity and impurity levels to
the understanding and characterization of nanoscale materials?  

	Dr. Cassell replied that the effects of impurities are extremely
important.  However, he noted examples where the impurities result in
more risk than the nanomaterial itself (e.g., when heavy metals are used
as catalysts).  Dr. Cassell also stated that the term “impurity”
must be defined when referring to nanomaterials (referencing previous
discussions on this topic).  He suggested that small changes in
structure, aspect ratio, and location of ligands may not result in
different chemical formulae, but will result in very different
characteristics.  These could be considered impurities.

	Dr. Murashov agreed with these comments.  He noted that it is important
to agree on what is considered the “pure material” before an
evaluation of impurities can be conducted.  Further, Dr. Murashov
suggested establishing an acceptable range for a material to be
considered “pure”.

	Dr. Small noted that measuring purity may be difficult (depending on
the characteristic that is selected to define purity).  He also noted
that tests must be designed to distinguish between the bulk material and
the catalysts within it.  

	Dr. Small’s comment regarding catalysts resulted in an extended
discussion about what other chemicals in the commercial nanomaterial
product should be evaluated (besides the nanomaterial itself). 
Panelists noted examples where specific catalysts are expected to remain
with the nanomaterial and other examples where contamination of unknown
chemicals results from processing operations.  Panelists suggested this
is of primary concern during full-scale production opposed to Research &
Development where samples are typically tested and monitored more
closely.  Multiple Panelists suggested the focus should be on
characteristics that can impact environmental health and safety.

EPA Response, Public Comment, and Q/A Session

	Dr. Williamson asked the Panelists and observers if EPA should ask for
more information on impurities for the NMSP.

	Public Comment #1:  Shaun Clancy (ACC) stated that the issue of
impurities may be very important in context of the NMSP.  He recognized
that EPA may request data on specific chemical impurities in the
commercial mixture.  He also recognized the potential for multiple NMSP
participants to submit data on the same nanomaterial.  Test results may
be affected by impurities; therefore, concentration data and data
regarding the specific impurities will be needed to compare the results.

	Public Comment #2: Donald Ewert (NanoProducts Corp.) commented that his
customers require information regarding impurities.  Therefore, he
believes this is an important topic.  He further stated that his company
routinely characterizes their products and the corresponding impurities,
believing that measurement techniques are available to acquire
appropriate information.

Question 13: 	Are there routine purification procedures that can
effectively control or remove impurities, when desirable, for certain
classes of nanoscale materials?

	Dr. Canady responded that these procedures are available.  However,
they vary depending on the specific nanomaterial.  There is not one
procedure that is universally suitable.

	Dr. Cassell referred to his pre-meeting comments for more detail.  He
also specifically noted that the methods currently used for
nanomaterials are the same as those used for macromaterials.

	Each Panelist was asked for input and all concurred with the statements
made by Dr. Canady and Dr. Cassell.

Session 2- Discussion:  Physical Chemical Properties

	Ms. Connery opened this session by stating that there were five related
questions in EPA’s discussion paper that would be discussed (questions
4 through 8).  She reviewed each question and then asked Panelists to
discuss their responses.  The questions and subsequent discussions are
presented below.

Question 4: 	Which physical-chemical properties are relevant to
nanoscale materials and how?  Which are known or reasonably
ascertainable and which have data gaps?

	Prior to the discussion on this topic, Dr. Williamson clarified that
relevant properties include those that may impact the function of the
material in its intended use, or those that may impact the material’s
effects.

	Dr. Murday commented that virtually all properties are relevant to some
degree; however, he suggested some have a more dramatic influence.  Dr.
Murday referred to his pre-meeting comments for a list of those he
believes are most significant.

	Dr. Murashov agreed with Dr. Murday’s list but suggested a few
additions, referenced in his pre-meeting comments for a list of the most
relevant properties.

	Dr Warheit reminded the group of the earlier discussion that the
biological matrix from which the nanoparticle is dispersed can
significantly affect characteristics (e.g., wet (water, culture media)
or dry).

	Dr. Lahann commented that EPA should try to identify properties that
are unique to nanomaterials; however, he did not suggest what those
properties may be.

	Dr. Canady noted that there are differences in observed properties, and
the corresponding characteristics, when the nanomaterial is aggregated.

	Dr. Busnaina agreed with the lists that were provided by Dr. Murday and
Dr. Murashov, and with the discussion regarding aggregates.  He added
that surface energy is also an important property (recognizing it is not
easy to measure).

	Dr. BelBruno agreed with the lists discussed above.

EPA Response, Public Comment, and Q/A Session

	Dr. Fritz asked Panelists to expand on comments regarding
agglomeration, specifically asking whether surface energy was a factor
in agglomeration.

	Various Panelists commented that surface energy (or perhaps
“interfacial energy”) does affect the agglomeration potential, but
the degree is case-by-case.  Also, some Panelists noted that
nanomaterials can be coated to prevent agglomeration.  This is desirable
in some situations and can be done to reduce the hazard.  However, other
Panelists noted that coatings do not fully mitigate concerns.

	Public Comment #1:  An observer asked the group if there are concerns
related to silver in the nanoscale that are different from those in the
macroscale.  Panelists did not provide a response.

	Public Comment #2:  Shaun Clancy (ACC) suggested that EPA should revise
their investigation on this topic to only include an analysis of
properties that are important to specified toxicological end points.

	Mr. Clancy also urged EPA to develop the required elements for the NMSP
based on information that will help fill data needs (information that
will inform a risk assessment opposed to general data gaps).

	Public Comment #3: Steve Hayes (Gobbell Hays Partners, Inc.) suggested
that EPA should revise their investigation on this topic to focus on
properties that are relevant:

Compared to materials in the bulk form

Performance

For toxicological effects

For fate & transport analyses

Persistence (because it relates to considerations during cleanup)

He also suggested that EPA consider topics that may affect clean up of
unintended releases (e.g., spills).

	Public Comment #4: Tanya Spellman (Washington Council of Government)
commented that EPA should consider the ultimate fate of nanomaterials;
specifically noting that they may be transferred to Publicly–Owned
Treatment Works and these facilities will need to understand appropriate
handling and cleanup procedures.  Ms. Spellman also concurred with Mr.
Clancy’s remarks that EPA should only request information that will
help inform a risk assessment.

Question 5: 	Are there properties that would have little or no relevance
under the NMSP?

	Multiple Panelists discussed their view that virtually all properties
have some importance.  Some Panelists commented that because there is a
lack of information regarding potential effects associated with exposure
to nanomaterials, EPA should attempt to acquire as much information as
possible rather than limiting the effort to investigate certain
properties.  Other Panelists acknowledged that all properties could
potentially result in an impact; however, they felt it would be more
appropriate to prioritize efforts and focus on properties that are most
likely to result in an impact.  They acknowledged that the costs
associated with testing may have an impact on which tests should be
performed, if only limited useful information would be produced.

	Multiple Panelists suggested specific properties that may not be
relevant for certain aspects of a risk evaluation.  These are briefly
summarized below.

	Dr. Murashov referred to his pre-meeting comments for specific
suggestions.  During the discussion he suggested that refraction may not
be relevant.

	Dr. Lahann commented that elastic properties may affect cell uptake of
nanomaterials; therefore, they could be relevant.

	Dr. Murday agreed with Dr. Murashov’s list in general but stated that
optical properties can sometimes be important.

	Dr. Warheit reiterated his previous statement that virtually all
properties have some relevance. 

	Thermal properties and superconductivity were also specifically
mentioned as properties that may not be relevant.

EPA Response, Public Comment, and Q/A Session

There were no public comments or questions.

Question 6: 	Which properties are associated with aggregated or
agglomerated nanoscale materials, as opposed to properties that are
inherent to the material regardless of physical form?

	Dr. Murashov reviewed his pre-meeting comments.  He stressed that
biological activity can be affected by agglomeration.

	Dr. Small stated that agglomeration affects the fate and transport of
nanomaterials in the environment. 

	Other Panelists agreed with Dr. Murashov’s initial comments and the
subsequent discussion that ensued during his review.  It was noted that
larger aggregated particles are easier to remove from a sample by
filtration.

EPA Response, Public Comment, and Q/A Session

	There were no public comments or questions.

Question 7: 	Are there routine manipulations of nanoscale materials that
result in physical-chemical properties changes or other defining
characteristics (e.g., surface modifications of nanotubes to enhance
solvent dispersibility)?

	EPA added a clarifying example:  Coatings can be applied to minimize
agglomeration.  How are these applied?

	Multiple Panelists indicated there are many processes that industry
uses, but each is developed and implemented case-by-case, depending upon
the specific situation.  Multiple Panelists indicated there is not a
“routine” process.  However, Panelists offered the specific examples
presented below.

	Dr. Warheit stated that sonication can reduce agglomeration.

	Dr. Murday stated that the surface of many nanomaterials must be
“capped” during production.  Depending on the complexity of the
material and the end use, the cap may include a ligand that imparts new
functionality.

	Dr. Cassell stated there are many processes that manufacturers use to
package materials for transport to customers.  These include annealing
and application of coatings.  He referred to his pre-meeting comments
for additional information.

EPA Response, Public Comment, and Q/A Session

	Public Comment #1: Mark Herwig (GE) stated that routine manipulations
include applying coatings to alter the functionality of the
nanomaterial.  However, the type of coating and desired functionality
affects the specific processes.

Day two Summary (September 7, 2007)

Session 2- Discussion:  Physical Chemical Properties (Continued)

	The meeting was reconvened on September 7th, with Ms. Connery reminding
Panelists and observers that Day One concluded with a discussion
pertaining to Question 7 from the discussion paper.  Ms. Connery then
asked Panelists to continue their efforts beginning with Question 8.

Question 8: 	How should physical-chemical property data be prioritized
for the NMSP?  Based on availability, effect (toxicity or exposure
criteria), or other factors?

	Dr. Warheit stated that there is a very long list of physical-chemical
data that may be useful.  However, he provided a list of “minimal
essential criteria” in his pre-meeting comments.  All Panelists agreed
in general with Dr. Warheit’s list (with some specific comments as
listed below).

	Dr. Warheit then provided some additional points for EPA to consider
when finalizing the list:

Particle size

Wet vs. dry state

Differences in the core vs. the shell (the surface is often where most
activity occurs)

Dr. Warheit also referred EPA to five studies that indicate the surface
activity is often more important than the particle size (see his
pre-meeting comments).

	Dr. Canady agreed with Dr. Warheit’s list and comments.  However, he
noted that the list focuses on data to inform a hazard assessment. 
Additional data may help inform other environmental effects that have
delayed or indirect effects on humans or ecosystems, for example, as are
recognized in the ozone depletion and global warming issues.  He also
noted that the current Panel did not include an expert on ecotoxicity. 
Additional items may be important from an ecotoxicity perspective.

	Dr. Colvin concurred with Dr. Warheit’s list; however, she suggested
acquiring all data and information would be useful (rather than limiting
the effort to a prioritized list).  Other Panelists agreed that it may
not be appropriate to “eliminate” certain data elements and that
acquiring as much information as possible may be appropriate.

	Multiple Panelists agreed that surface reactivity is very important. 
An extended discussion regarding surface reactivity, the type of
information needed to appropriately define it, and the corresponding
tests that would be required ensued without specific resolution.  The
types of chemical reactions associated with the surface were discussed. 
In addition, surface reactivity was related to hazard and biological
activity.  One Panelist encouraged EPA to consult NIH on this topic.

	Dr. Busnaina commented that EPA should recognize that Confidential
Business Information (CBI) concerns may make it difficult to acquire
this information.

EPA Response, Public Comment, and Q/A Session

	Public Comment #1: Andy Atkinson (Environment Canada) initiated a
discussion by asking if agglomeration impacts the proposed list of
primary properties to consider.  Multiple Panelists stated agglomeration
would affect the properties (particularly surface reactivity). 
Follow-up discussions resulted in a suggestion to add a study of
persistence because agglomeration affects persistence in the environment
(and there is the potential for agglomerated nanomaterials to
disagglomerate over time).

Session 3 - Discussion:  Nanoscale Manufacture and Processing

	Ms. Connery opened this session by stating that there were three
related questions in EPA’s discussion paper that would be discussed
(questions 9, 10, and 11).  She reviewed each question and then asked
Panelists to discuss their responses.  The questions and subsequent
discussions are presented below.

Question 9: 	What are the common processes used to manufacture nanoscale
materials?

	Dr. Colvin referred to a recent workshop devoted to this specific
topic.  She stated the summary will be available soon.  In general,
participants developed three primary methods for each of which there are
many specific processes:

Gas phase processes

Liquid phase processes

Top-down manufacturing

Dr. Colvin noted agglomeration concerns are relevant to methods (1) and
(2) and that coatings are applied to prevent agglomeration for type (2).

	Dr. Murashov suggested that biological processes leading to the
production of nanomaterials, such as nanoscale metal oxide particles,
could be identified as a major distinct method.  This method can produce
nanomaterials with unique compositions, properties, and impurities and
therefore can be characterized by very unique hazard properties.

	

	Dr. Murday commented that the above discussion refers to chemical-based
processes.  He noted that physical-based processes such as crushing and
grinding are also commonly used.

	Dr. Cassell suggested the following processes:

Chemical vapor deposition

Laser ablation

Thin film deposition

Self assembly (there could be “hundreds” of specific techniques

	Dr. Murday indicted that various metallurgical processes using
mechanical deformation are used.

EPA Response, Public Comment, and Q/A Session

	Comment #1:  Shekhar Subramoney (DuPont Company) stated he is aware of
a large-scale production process (thousands of pounds per year) using
mechanical/chemical exfoliation.

Question 10: 	How are processes used to produce specific characteristics
or properties?

	Dr. Cassell commented that chemical vapor deposition can be used to
form one dimensional linear materials where the reaction time can be set
to control length.  Also, self-assembly techniques (with micelle-based
approaches) can be used to control diameter.  Finally, annealing can be
used to modify the surface area and porosity.

	Dr. BelBruno indicated that “normal chemical methods” are used for
liquid synthesis.

	Dr. Murday stated that there are different processes used for the
following three steps:

Manufacture the core

Cap the core

Add a functionalized shell

This statement resulted in a discussion as to whether these general
steps are typical.  Multiple Panelists indicated these are typical and
presented some examples.

EPA Response, Public Comment, and Q/A Session

	There were no comments relevant to this topic.

Question 11: 	Which methods reduce particle size but do not result in
property changes?  Which methods reduce particle size and result in
property changes?

	Multiple Panelists asked for clarification from EPA.  During the
resulting discussion the Panelists and EPA agreed that the question was
answered to EPA’s satisfaction during discussions of previous
questions.

Session 4 - Discussion:  Methodologies for Characterizing Nanoscale
Materials Analytical Methods

	Ms. Connery opened this session by stating that there were eight
related questions in EPA’s discussion paper that would be discussed
(questions 14 through 21).  She reviewed each question and then asked
Panelists to discuss their responses.  The questions and subsequent
discussions are presented below.

Question 14: 	Are validated methods available for the different
categories of nanoscale materials?

	Dr. Colvin stated that 10 methods are currently being validated and
another 15 are in the pre-validation phase.  She referred to her
pre-meeting comments for additional information.

	Dr. Small stated that NIST is developing standards.  He also noted that
it is difficult to develop Standard Reference Materials.

	Dr. Murashov commented that ISO is in the process of developing
standards.  He also noted that there are no current industrial hygiene
methods to assess nanomateials and suggested that some bulk-phase
methods may be applicable (recognizing that further validation would be
required).

	Dr. Murday commented that equipment and tools are available and are
used to measure and characterize nanomaterials on a regular basis;
however, their use is not “validated”.  He also noted that certain
analytical methodology (NMR & surface analytical tools) require
“care” in interpreting data.

EPA Response, Public Comment, and Q/A Session

	There were no public comments or questions.

Question 15: 	Are there techniques that can be universally applied?

	All Panelists agreed that there are no techniques that can be
universally applied.  However, several Panelists provided specific
examples of techniques that can be used in various situations.  These
examples are summarized below. 

	Dr. Murashov suggested that electron microscopy is widely used and
accepted.  Multiple Panelists agreed but noted some limitations.  For
example, Dr. Small commented that this technique works well to
characterize the core, but not the shell of nanomaterials.  Also, Dr.
Canady noted that it does not provide statistical information (he
referred to the previous discussions regarding the analysis of
impurities).

	Dr. Colvin provided EPA with citations for studies involving
measurements of aerosols and indicated she would provide references.

	Dr. Murday commented that there are other techniques that are
available; however, many are cost-prohibitive.

EPA Response, Public Comment, and Q/A Session

	There were no public comments or questions.

Question 16: 	For small quantities of materials, are there sampling,
handling, and collection techniques as well as sample integrity,
accuracy and precision QA/QC methodologies available?

	Dr. Colvin referred to her pre-meeting comments for specific
information.  In addition, Dr. Colvin commented that she does not
believe there is a need for nano-specific guidelines on this topic.  She
noted that temperature control is important (particularly when measuring
and handling samples that may aggregate).  She also stated that
sonication can alter surface coatings and should be used appropriately.

	Dr. Busnaina agreed with Dr. Colvin’s comments, particularly
regarding temperature control.  He initiated a discussion between
multiple Panelists where examples of problems when samples were shipped
occurred due to extreme changes in temperature during flights to or from
hot and cold geographic locations.

	Multiple Panelists noted that best management practices for ultrafine
materials may be appropriate for nanomaterials.

EPA Response, Public Comment, and Q/A Session

	No comments on additional topics were provided.

Question 17: 	What is the status of standardization efforts? Are these
efforts focused on broadly applicable characterization methods or
category-specific methods?

	Dr. Small commented that individual researchers develop and use their
own standards (referring to standard reference materials).  He also
noted an upcoming NIST workshop that will be focused on developing a
prioritized list of standard reference materials for NIST to develop.

	Dr. Canady stated that OECD is also investigating this issue and may be
developing a work group.

EPA Response, Public Comment, and Q/A Session

	Public Comment #1: An observer asked for clarification regarding the
phrase “NIST Certified” that he had observed on raw materials he had
received from suppliers.

	Dr. Small responded that, in general, this means that the specific
sample can be traced to show it was compared to a NIST Standard
Reference Material.

Question 18: 	What alternative or innovative methods or technologies can
be applied to nanomaterial analysis?

	Dr. Colvin suggested that a combination of mass spectroscopy and
electro ionization could be appropriate.

	Dr. Small indicated that helium ion microscopes may be appropriate to
characterize materials greater than 10 nm.

	Dr. Murday suggested that Atomic Force Microscopy (AFM) probes can be
used to analyze individual particles.

EPA Response, Public Comment, and Q/A Session

	Public Comment #1:  Shekhar Subramoney (DuPont Company) stated that
X-Ray absorption spectroscopy allows for analysis down to the
atom-to-atom interactions.  However, the equipment may be cost
prohibitive.

Question 19: 	Are there models that are currently used to obtain
property data for nanoscale materials?  For which properties and which
nanoscale materials?

	Dr. Murashov referred to his pre-meeting comments for specific
examples.

	Dr. Cassell stated models exist to predict mechanical and electrical
properties of carbon-based nanomaterial systems.

	Multiple Panelists mentioned that models can be developed; however,
they need to be specific for every situation.  Therefore, there are no
standard or general models.  Other Panelists agreed they would have a
concern regarding model output because it would be difficult to verify
the results.  Still, some Panelists agreed that appropriate models have
been (or could be) developed for specific nanomaterials.  

EPA Response, Public Comment, and Q/A Session

	No comments on additional topics were provided.

Question 20: 	Has any validation work been conducted that compares
predicted values with measured data?  For which properties and which
nanoscale materials?

	Dr. Murday stated that some modeling had been conducted and verified
but only for “simple” molecules; he was not aware of modeling for
more “complex” molecules.

	No other Panelists were aware of existing models that had been formally
validated.  However, multiple Panelists indicated there are efforts in
academia to develop models.  Panelists noted that these models may not
be published because many scientific journals require validation and
there are no accepted validation methods.  Dr. Cassell stated one
example is modeling of the thermoconductivity of carbon nanotubes that
has been developed at the University of California at Berkeley. 

EPA Response, Public Comment, and Q/A Session

	No comments on additional topics were provided.

Question 21: 	Are there current significant characterization needs for
which the NMSP should investigate model development?

	Dr. Murashov reviewed his pre-meeting comments.

	Dr. Warheit agreed with Dr. Murashov and added that accurate models
generally cannot be developed until more information is acquired via
data generation.  Dr. Busnaina agreed.

	Dr. Murday suggested that EPA coordinate with other organizations for
development of specific types of models:

Reactivity (consult industry’s catalyst manufacturers)

Toxicology (consult with NIH)

EPA Response, Public Comment, and Q/A Session

	No comments on additional topics were provided.

Session 5 - Discussion:  Prioritization of Data Gaps

	Ms. Connery asked each Panelist to reflect upon the discussions that
occurred during the meeting on each specific question and general topic,
then identify the most important data gaps from their perspective.  A
summary list of Panelist responses is provided below.

	Dr. Canady:

What materials will be developed in the future (5, 10, and 15 years from
now)?

Specifically what needs to be validated?

What are the priorities for modeling?

	Dr. Warheit:

How can surface reactivity be measured?

What would be the disparity between measuring in the wet and dry states
for each particle type in each category?

	Dr. Murday:

What parameters associated with health effects are most important?

What is the ability to get “reliable” validated measurements?

How can we determine the composition and structure of the shell of
nanomaterials?

	Dr. Busnaina:

There is a lack of data regarding surface reactivity (from a
toxicological perspective).

There is a lack of data on functionalization (and effects from a
toxicological perspective) specifically for commercial products.

What are the effects of impurities and the correlation to toxicity and
properties (and at what concentration are they important)?

	Dr. BelBruno:

There is a lack of information regarding the morphology (differences
between the core, shell and crystal structures).

There is a lack of information regarding surface reactivity and
correlated with #1 above.

There is a lack of information regarding impurities:

How can consistency between batches be ensured?

How are impurities defined?

	Dr. Small:

How can surface reactivity be defined and measured?

There is a need to correlate data from wet vs. dry tests.

There is a data gap regarding the use of standards.

Determining criteria for categories.

	

	Dr. Cassell:

Standardization.

Validation of characterization methods.

There is a need for a publicly-available information repository.

Surface reactivity.

Relationship between size and other properties.

	Dr. Colvin:

There is a lack of understanding regarding wet nanoparticles and their
interaction in the environment (micelles and complexation).

A mechanism for data mining and sharing of information is lacking.

Fully understanding the classification of commercial products

	Dr. Murashov:

There is a need to harmonize and standardize:

Common definitions (e.g., what does “impurity” mean?)

A list of physical-chemical properties that are critical for a risk
assessment.

Protocol for collecting data.

Closing Remarks

After Ms. Connery confirmed that all Panelists had provided appropriate
closing comments, EPA thanked the Panel and observers and the meeting
was adjourned.

Appendix A

MEETING AGENDA

	

	 United States 	

	 Environmental Protection Agency

Peer Consultation on the Material Characterization of

Nanoscale Materials - Agenda

Holiday Inn Rosslyn at Key Bridge, Arlington, Virginia

September 6-7, 2007

Day One: Thursday, September 6

8:30 a.m.	Registration

9:00 a.m.	EPA Opening Remarks and Background	EPA

Meeting Purpose and Objectives

9:30 a.m.      Nanoscale Materials Stewardship Program (NMSP) Overview
and Status	EPA

Summary of EPA’s Concept for the NMSP 

NMSP Relationship to OPPT TSCA Programs

10:00 a.m.    BREAK

10:15 a.m.	Introductions, Meeting Agenda, Format and Charge	ERG

10:45 a.m.	Public Comment Period	ERG

11:45 a.m.      LUNCH

1:00 p.m.	Discussion: Types of Nanoscale Materials & Their Structure and
Chemical Composition

Question 1: Are there any other significant categories, based on
structure and chemical composition, that should be included in this
discussion because they are substantially different from the categories
mentioned (e.g., hybrids, self- assembly devices, others)?

Question 2: For the different categories of nanoscale materials, what is
the current state of knowledge about structure and chemical composition?
 

Question 3: Can structures and chemical composition be correlated to
specific properties and is this correlation quantifiable? Chemical
Identification Elements (e.g., which ones? for which substances? test
methods?)

Question 12: How important are impurity identity and impurity levels to
the understanding and characterization of nanoscale materials?  

Question 13: Are there routine purification procedures that can
effectively control or remove impurities, when desirable, for certain
classes of nanoscale materials?  

4:30 p.m.        Wrap-up and Prepare for Day Two

5:00 p.m.	ADJOURN

Day Two: Friday, September 7

8:30 a.m.	Discussion: Physical-Chemical Properties

Question 4: Which physical-chemical properties are relevant to nanoscale
materials and how?  Which are known or reasonably ascertainable and
which have data gaps?

Question 5: Are there properties that would have little or no relevance
under the NMSP?

Question 6: Which properties are associated with aggregated or
agglomerated nanoscale materials, as opposed to properties that are
inherent to the material regardless of physical form?

Question 7: Are there routine manipulations of nanoscale materials that
result in physical-chemical properties changes or other defining
characteristics (e.g., surface modifications of nanotubes to enhance
solvent dispersibility)?

Question 8: How should physical-chemical property data be prioritized
for the NMSP?  Based on availability, effect (toxicity or exposure
criteria), or other factors?

	Discussion:  Nanoscale Manufacture and Processing	Methodologies for
Characterizing Nanoscale Materials

Question 9: What are the common processes used to manufacture nanoscale
materials?

Question 10: How are processes used to produce specific characteristics
or properties?

Question 11: Which methods reduce particle size but do not result in
property changes?  Which methods reduce particle size and result in
property changes?

Noon	      LUNCH

1:15 p.m.	Discussion:  Methodologies for Characterizing Nanoscale
Materials

	Analytical Methods:

Question 14: Are validated methods available for the different
categories of nanoscale materials?

Question 15: Are there techniques that can be universally applied?

Question 16: For small quantities of materials, are there sampling,
handling, and collection techniques as well as sample integrity,
accuracy and precision QA/QC methodologies available?

Question 17: What is the status of standardization efforts? Are these
efforts focused on broadly applicable characterization methods or
category-specific methods?

Question 18: What alternative or innovative methods or technologies can
be applied to nanomaterial analysis?

     	Models:

Question 19: Are there models that are currently used to obtain property
data for nanoscale materials?  For which properties and which nanoscale
materials?

Question 20:  Has any validation work been conducted that compares
predicted values with measured data?  For which properties and which
nanoscale materials?

Question 21: Are there current significant characterization needs for
which the NMSP should investigate model development?

Discussion:  Prioritization of Data Gaps

2:30 p.m.	BREAK

2:45 pm	Public Comments

3:15 p.m.	Panelist Closing Remarks 

3:45 p.m.	Next Steps and EPA Closing Remarks

4:00 p.m.	ADJOURN

Appendix B

DISCUSSION PAPER

Peer Consultation Public Meeting on

The Material Characterization of Nanoscale Materials

Discussion Topics

August 7, 2007

Peer Consultation Public Meeting on

The Material Characterization of Nanoscale Materials

Discussion Topics

Meeting Background and Purpose

	EPA is convening a public scientific peer consultation meeting on
material characterization for nanoscale chemical substances
(“nanoscale materials”) to inform the development of its Nanoscale
Materials Stewardship Program (NMSP) under the Toxic Substances Control
Act (TSCA).  The peer consultation is one of several actions EPA is
taking to better understand the potential risks and benefits of
nanotechnology.

	 On October, 18, 2006, EPA invited the public, industry, environmental
groups, other federal agencies and other stakeholders to participate in
the design, development and implementation of a stewardship program for
nanoscale materials.  On July 12, 2007, EPA announced the availability
of a NMSP concept paper and related documents and a public meeting on
August 2, 2007 to discuss and receive comments on these materials
(http://www.epa.gov/oppt/nano/nmspfr.htm).

 	The NMSP will complement and support the Agency's new and existing
chemical programs under TSCA and will help provide a firmer scientific
foundation for regulatory decisions by encouraging the development of
key scientific information and appropriate risk management practices for
new and existing chemical nanoscale materials.  The NMSP is intended to
include but not limited to engineered nanoscale materials manufactured
or imported for commercial purposes as defined in 40 CFR 720.3 (r).  The
NMSP is envisioned to have two parts: 1) a Basic Reporting Program and
2) an In-depth Program.  This discussion paper and the public peer
consultation meeting are intended to support the NMSP by identifying
material characterization data that participants could submit under the
Basic Program if they are in the participant’s possession or are
reasonably ascertainable as defined in 40 CFR 720.3 (p).  The data and
experience generated by the basic reporting phase will help to inform
the types of in-depth data to be developed.  In-depth data development
could begin at any time and would entail, among other types of data,
development of material characterization data in a greater amount of
detail.  In-depth data development could also include additional types
of material characterization data if they are identified.

	EPA received input in November, 2005 from the National Pollution
Prevention and Toxics Advisory Committee (NPPTAC) regarding a voluntary
stewardship program for nanoscale materials.  A NPPTAC ad-hoc work group
on nanoscale materials developed an overview paper that stated that a
voluntary stewardship program should:  

Give EPA, and the public to the extent possible recognizing legitimate
CBI issues, a better understanding of the types of engineered nanoscale
materials; the physical, chemical, hazard and exposure characteristics
of such substances; the volume of such substances; and the uses of such
substances;

Help EPA develop capacity and a process to identify and assess risks of
engineered nanoscale materials;

Help EPA determine what information it needs about engineered nanoscale
materials and articulate those information needs to industry and other
stakeholder groups;

Help EPA understand what risk management practices are being used at
production, processing, use and disposal stages, and what additional
risk management practices  need to be implemented;

Prompt or reinforce the implementation of risk management practices; and

Provide the information and experience needed to develop an overall
approach to the treatment of nanoscale chemical substances under TSCA
that builds public trust in nanoscale materials while enabling
innovation and responsible development.

	EPA will utilize all public input, including that from NPPTAC, other
stakeholders, public meetings and peer consultations to further inform
the development of its Nanomaterials Stewardship Program and TSCA
program for nanoscale materials.

Meeting Objectives

	The EPA public peer consultation meeting on material characterization
needs for nanoscale materials will help clarify which data and elements
should be included in the NMSP Basic Program and/or In-depth Program. 
The goal is to have an applied discussion that considers 1) the
currently available understanding of material characterization as it
relates to nanoscale materials and 2) how this understanding can be used
to guide the Agency’s thinking regarding the material characterization
data elements that would be most useful and important to include in the
NMSP.  The specific objectives of the public peer consultation meeting
are as follows:

To inform industry and the public of EPA’s level of understanding of
the material characterization needs for nanoscale materials in general
and for the NMSP;

To further develop EPA’s understanding of how nanoscale materials are
engineered or manufactured to achieve specific properties and
characteristics; 

To further develop EPA’s understanding of which chemical
identification elements and physical-chemical property data are
generally relevant in characterizing nanoscale materials and which
identification elements and property data are most important in
characterizing specific classes of nanoscale materials;

To discuss what analytical procedures and test methods are available for
acquiring these material characterization data, and where procedure and
method validation or development is needed;

To discuss how these material characterization data needs should be
prioritized for the NMSP Basic and/or In-depth Program; and

To discuss potential nomenclature needs for specific classes of
nanoscale materials based on material characterization outcomes.

Discussion Overview

Despite the rapid advancement of nanotechnology, the breadth of
nanoscale material types coupled with the limited hazard data available
for many of these materials pose a challenge in understanding and
measuring their benefits and risks.  Numerous efforts are underway to
begin to address these challenges.  The International Life Sciences
Institute (ILSI) Research Foundation/Risk Science Institute, for
example, convened an expert working group to develop a strategy for
identifying hazards associated with engineered nanoscale materials. 
Focusing on the limited data available, the working group developed a
screening strategy for hazard identification (rather than a detailed
testing protocol) that includes a broad data gathering effort.  

	The ILSI report describes the characterization of nanoscale materials,
in addition to in vitro and in vivo screening, as a key third aspect of
an overall screening strategy due to the likely dependence of the
biological activity of nanoscale materials on physical-chemical
properties not often considered in toxicity screening studies. 
Additionally, given the difficulties associated with characterizing many
nanoscale materials, nanoscale material characterization is a subject
appropriate for detailed investigation and discussion.

Numerous national and international standards organizations, including
the American National Standards Institute (ANSI) and the International
Organization for Standardization (ISO), have also convened committees to
begin to address many of these same challenges and in particular the
need for methods standardization.  Several of these committees have
indicated that, given the breadth of nanotechnology and its data issues,
the initial products the committees develop may be in the form of best
practices rather than actual test protocols.  

	The remainder of this paper discusses a proposed approach for the EPA
scientific peer consultation meeting on nanoscale material
characterization.  EPA recognizes that the different chemical classes of
nanoscale materials would make universal application of any particular
characterization endpoint or methodology impossible.  The premise of the
Basic Reporting phase of the NMSP is that some information is known or
reasonably ascertainable.  The approach for this Panel is therefore to
discuss what data are known or reasonably ascertainable to characterize
nanoscale materials.  This will be followed by a discussion on the
methodology used to obtain and use characterization endpoints of
interest.  

Discussion Topics

Characterization of Nanoscale Materials 

1.	Description of nanoscale materials 

Types/categories of nanoscale materials

2.	Physical-chemical properties of potential interest

Particle size and distribution

Particle shape and dimensions

Agglomeration and aggregation

Surface area

Surface charge

Surface chemistry

Chemical composition

Crystal structure

Impurity identification and levels

3.  Design to achieve unique properties

Manufacturing and processing methodologies

Chemical transformations

Methodologies for Characterizing Nanoscale Materials

Obtaining characterization data for nanoscale materials 

Analytical methods for detecting and quantifying nanoscale materials

Analytical methods measuring physical-chemical properties (measurement
techniques and testing protocols)

Models to predict properties and effects

Metrology 

Methods validation

Standards and harmonization

Prioritization of characterization data and data gaps 

Miscellaneous – Do Panelists have additional topics to discuss?

	    

Discussion

	This section provides additional information on specific technical
issues to facilitate discussion at the meeting.  The information will
include literature findings as well as questions on the specific
discussion topics.

Types of Nanoscale Materials & Their Structures and Chemical
Compositions

	Based on structure and chemical composition, EPA has grouped
nanomaterials into 4 distinct categories for purposes of this
discussion: 1) simple organic molecules; 2) simple inorganic molecules;
3) polymeric substance (including dendrimeric substances); and 4)
composites.  A fifth category, biological compounds will not be
addressed in the peer consultation.  While all of these categories can
be divided further, only the organic category will be divided further
into molecules based predominantly on carbon (e.g., fullerenes,
nanotubes) and all other organic substances (e.g., salts of carboxylic
acids).  This grouping is similar to the American National Standards
Institute Nanotechnology Standards Panel approach presented at the
September 2004 meeting at the National Institute of Standards and
Technology in Gaithersburg Maryland.  

Question 1: Are there any other significant categories, based on
structure and chemical composition, that should be included in this
discussion because they are substantially different from the categories
mentioned (e.g., hybrids, self- assembly devices, others)?

Question 2: For the different categories of nanoscale materials, what is
the current state of knowledge about structure and chemical composition?
 

Question 3: Can structures and chemical composition be correlated to
specific properties and is this correlation quantifiable?

Physical-Chemical Properties

	The importance of nanoscale materials is due to their potential for
unique or greatly enhanced properties.  	EPA routinely uses a base set
of physical-chemical property data (e.g., melting point, boiling point,
vapor pressure, water solubility) for a variety of programs (e.g., High
Production Volume Challenge, New Chemicals) and decision-making.  As
mentioned previously, certain material properties are of significant
importance in characterizing nanoscale materials.  Recent research
suggests that particle size, surface area, and surface chemistry (or
surface activity) are initially some of the most important properties to
measure.  

	As expected for most chemicals, class 1 substances having specific
molecular structures and formulas may be more readily studied and
characterized at the nanoscale than the polymer and composite
categories.  For example, carbon-based nanoscale materials as well as
metal oxide nanoscale chemicals are often well characterized. 
Structural and physical-chemical property data therefore are likely to
be well documented for these types of materials.

	

Question 4: Which physical-chemical properties are relevant to nanoscale
materials and how?  Which are known or reasonably ascertainable and
which have data gaps?

Question 5: Are there properties that would have little or no relevance
under the NMSP?

Question 6: Which properties are associated with aggregated or
agglomerated nanoscale materials, as opposed to properties that are
inherent to the material regardless of physical form?

Question 7: Are there routine manipulations of nanoscale materials that
result in physical-chemical properties changes or other defining
characteristics (e.g., surface modifications of nanotubes to enhance
solvent dispersibility)?

Question 8: How should physical-chemical property data be prioritized
for the NMSP?  Based on availability, effect (toxicity or exposure
criteria), or other factors?

Nanoscale manufacturing and processing

	The number of manufacturing and processing methods for generating
nanoscale material continues to grow and become more sophisticated.  The
two primary areas for this discussion include physical reduction methods
(milling) and engineering methods (e.g., particle stabilization, vapor
deposition, self assembly).  

Question 9: What are the common processes used to manufacture nanoscale
materials?

Question 10: How are processes used to produce specific characteristics
or properties?

Question 11: Which methods reduce particle size but do not result in
property changes?  Which methods reduce particle size and result in
property changes?  

Impurities

	Impurity content is a growing area of interest in nanotechnology due to
improved performance observed in some cases (e.g., solar cells and
semiconductors) and deleterious effects observed in others (e.g.,
quantum dot quantum computers).  The confounding effects that impurities
have with respect to toxicological endpoints are also being studied
(National Nanotechnology Initiative 2006 Environmental Health and Safety
research report).

Question 12: How important are impurity identity and impurity levels to
the understanding and characterization of nanoscale materials?  

Question 13: Are there routine purification procedures that can
effectively control or remove impurities, when desirable, for certain
classes of nanoscale materials?  

Obtaining characterization data 

	Determining identity, quantifying the nanoscale particle range, and
measuring physical-chemical properties for that identity and particle
range are essential to the characterization of nanoscale materials. 
Because of the challenges associated with size, shape, surface
characteristics, and possibly other aspects of nanoscale materials, an
evaluation of existing measurement techniques is critical to nanoscale
material characterization.  The National Nanotechnology Initiative
report stated that “...Accurate and useful measurement techniques are
also important because agglomerated nano materials may either retain or
lose their emergent properties - or take on new properties - thus
affecting the potential biological response.”  

Question 14: Are validated methods available for the different
categories of nanoscale materials?

Question 15: Are there techniques that can be universally applied?

Question 16: For small quantities of materials, are there sampling,
handling, and collection techniques as well as sample integrity,
accuracy and precision QA/QC methodologies available?

Question 17: What is the status of standardization efforts? Are these
efforts focused on broadly applicable characterization methods or
category-specific methods?

Question 18: What alternative or innovative methods or technologies can
be applied to nanomaterial analysis?

Modeling

Empirical modeling can be a useful approach to predict physical-chemical
properties when experimental data are not known or ascertainable.  The
initial problem with modeling is that, to accurately predict property
endpoints for a given category of substances, there must be some
experimental data available in the tool’s database for at least some
representative substances in that category.  For newly discovered or
studied materials, the minimum but necessary quantity and type of
experimental data often is not available to sufficiently populate a
tool’s database and allow accurate prediction by the tool.  Some
estimation methods have been developed for specific property endpoints,
but many others are lacking.

Question 19: Are there models that are currently used to obtain property
data for nanoscale materials?  For which properties and which nanoscale
materials?

Question 20:  Has any validation work been conducted that compares
predicted values with measured data?  For which properties and which
nanoscale materials?

Question 21: Are there current significant characterization needs for
which the NMSP should investigate model development?

Appendix C

PRELIMINARY PANEL COMMENTS ON DISCUSSION PAPER

Comment Compilation of Panelists’ Preliminary Responses to Panel
Discussion Questions Posed in the Discussion Paper for the Public
Meeting on Material Characterization of Nanoscale Materials

Draft Compilation as of August 31, 2007

A compilation of Panelists’ preliminary comments received in response
to “Panel Discussion Questions” contained in the Discussion Paper
for the Public Meeting on Material Characterization of Nanoscale
Materials is provided below.  This compilation is organized as follows:

General Observations

Discussion Category Responses

Types of Nanoscale Materials and their Structures and Chemical
Compositions

Physical-Chemical Properties

Nanoscale Manufacturing and Processing

Impurities

Obtaining Characterization Data

Modeling

Supporting Figures

Supporting References

General Observations

Dr. Warheit:

Thank you for giving me the opportunity to provide some preliminary
comments on the Discussion Topics for EPA’s Peer Consultation Public
Meeting on the Material Characterization of Nanoscale Materials.  The
forthcoming meeting should provide useful information, both for EPA and
for stakeholders/outside participants.  I would caution EPA, that in my
opinion, many of the answers to the questions contained in the
“Discussion Topics” document are case-by-case and/or
nanoparticle-type specific (e.g. those pertaining to metal oxides vs.
carbon structures) and a universal response cannot be given (e.g.
questions 9 and 10 – “What are the common processes used to
manufacture nanoscale materials?  And “How are processes used to
produce specific characteristics or properties”).  Indeed the various
subsets of nanoparticle-types are likely to require different answers to
the same questions.

Discussion Category Responses

Types of Nanoscale Materials & Their Structures and Chemical
Compositions

	Based on structure and chemical composition, EPA has grouped
nanomaterials into 4 distinct categories for purposes of this
discussion: 1) simple organic molecules; 2) simple inorganic molecules;
3) polymeric substance (including dendrimeric substances); and 4)
composites.  A fifth category, biological compounds will not be
addressed in the peer consultation.  While all of these categories can
be divided further, only the organic category will be divided further
into molecules based predominantly on carbon (e.g., fullerenes,
nanotubes) and all other organic substances (e.g., salts of carboxylic
acids).  This grouping is similar to the American National Standards
Institute Nanotechnology Standards Panel approach presented at the
September 2004 meeting at the National Institute of Standards and
Technology in Gaithersburg Maryland.  

Question 1: Are there any other significant categories, based on
structure and chemical composition, that should be included in this
discussion because they are substantially different from the categories
mentioned (e.g., hybrids, self- assembly devices, others)?

Dr. BelBruno:

The classification scheme presented seems to be the most practical.  To
do otherwise, would overwhelm the issue at a time when we are trying to
get a basic set of data together.  Most other nanoparticles are
combinations or composites of the classes already listed, for example,
polymer coated iron nanoparticles for imaging.  While the combined
material will need to (eventually) be examined for its own possible
unique properties, the components need to be understood first.

Self-assembled and hybrid materials, cited in the topic description,
fall into the same composite or “combination” category as the
imaging particles noted above.

Dr. Canady:

Perhaps the categories, or EPA’s thinking about “mechanisms to help
clarity of approach,” should not be limited by structure AND
composition.  A useful classification scheme (including cross-referenced
sub-classifications across categories) should probably be more amenable
to identifying attributes that are important to regulatory clarity by
being open to structure, composition, or combinations of structure and
composition.  From asbestos and PM data it seems there may be
combinations of structure and composition that are important and there
may also be structures that are important independent (to a large
degree) of composition.  Function (e.g., OP compounds, endocrine active
compounds) may also be a useful classification attribute. 
Classification should be amenable to a matrix approach with composition,
structure, and function as dimensions of the matrix.  It seems likely,
as more and more NMs are developed, that the very broad categories
presented here by EPA will begin having very little practical utility
because we will be discovering categories and classes within them that
are more important in understanding how to regulate, and that confuse
the boundaries. 

If EPA stays with large bins, here are some other categories to
consider:

Combinations: There are already intersections of
inorganic/organic/polymer/composite materials in the form of
specifically constructed nanoscale materials or particles that will
challenge any category-based approach to nomenclature or regulation.  At
some point the combinations may comprise the bulk of the applications as
molecular assembly comes into full use. 

Emulsions, vesicles, liposomes, micelles: This would be a structure (or
form) rather than composition category with either amphiphillic
molecules forming “filled balloons” or small collections of
molecules forming nanoscale particles in stasis in a liquid.  These
kinds of particles are being developed for use in food and drug
applications, biocides, and, most likely, pesticides.  Based on the
utility of the approach in drug delivery and nutrient stabilization in
foods, it’s not hard to imagine that industrial uses would also be
developed (for chemical processes to control/specify reactions,
solubilization of insoluble intermediates, and transport/stabilization
of compounds).

Dr. Cassell:

I’m confused at this terminology defined in the introductory
paragraph.  For instance, simple organic molecules would not be
inclusive of multiwalled carbon nanotubes (this is implied that it can
be done this way).  I think there needs to be more thought put into
classifying these materials correctly.

Hybrid structures in particular need to be considered.  For example,
there are emerging reports detailing the use of engineered nanomaterials
that could be placed into multiple categories (organic, inorganic and
polymeric).  Filled carbon nanostructures (e.g. metal filled multiwalled
carbon nanotubes) are a class of materials that could be grouped in such
a way.  Another common class of materials is hybrid inorganic/organic
polymers.

Dr. Lahann:

In principle, the four categories are sufficient for broad
classification on a materials basis. However, hybrid materials have also
been observed increasingly in recent times and are distinct from
composites.  For instance, a particle with an inorganic core and an
organic shell should be considered a hybrid material rather than a
composite.  Thus, adding a 5th category may be wise.

Dr. Murashov:

The concept paper proposes the following grouping of nanomaterials: 1)
simple organic molecules; 2) simple inorganic molecules; 3) polymeric
substances (including dendrimeric substances); 4) composites; 5)
biological (nanomaterials assigned to the latter group are not
considered under the proposed stewardship program).  The first category
is further subdivided into molecules based predominantly on carbon (e.g.
fullerenes and nanotubes) and all other organic substances (e.g. salts
of carboxylic acids). 

This categorization suffers from inconsistency with the conventional
usage of chemical terms. Specifically:

Carbonaceous materials including allotropes of carbon such as fullerenes
and carbon nanotubes are considered inorganic materials.

It is not clear why nanoscale formulations of simple organic molecules
such as salts of carboxylic acid were made into a separate category of
nanomaterials given their little significance as engineered
nanomaterials at this point.

Traditionally, “simple inorganic molecules” means molecules such as
CO2, SO2, NO, H2O.  It is not clear why nanoscale formulations of these
materials would be made into a separate category of nanomaterials given
their little significance as engineered nanomaterials at this point. Was
this category meant to describe “simple inorganic compounds” to
include solid materials with a high degree of ionic or metallic bonding
such as TiO2 and Ag, respectively? 

It is not clear what is meant by “composite” category of
nanomaterials.  For example, would nanoparticles composed of cores and
shells of different chemistries be considered composites? Would
functionalized nanoparticles be considered composites? 

An alternative classification of simple nanomaterials can be found, for
example, in the draft ISO TC229 Technical Report on “Health and Safety
Practices in Occupational Settings Relevant to Nanotechnologies” and
includes the following categories: 1) carbonaceous; 2) oxides; 3)
metals; 4) semiconductors; 5) organic polymeric; 6) bio-inspired.  This
classification is better aligned with the chemistry of simple
nanomaterials and their hazard properties.  More complex nanomaterials
could be categorized further along the lines of nomenclature tree being
developed by ISO TC229 which includes distinction between nano-objects
(includes nanoparticles) and nanostructured materials (includes
nanocrystals and complex assemblies of nano-objects).

Dr. Murday:

The presently identified categories apparently presume freestanding
nanostructures (NS).  These certainly pose a greater risk for pervasive
dissemination.  One might also think about NS embedded physically in a
matrix (i.e. no chemical bonding), embedded chemically in a matrix (i.e.
chemical bonding) and what might be called “thin film” where NS are
etched out of deposited films (that are chemically bonded to the
substrate).

There is some danger in working from a molecular perspective of NS. 
Molecules are largely well defined entities – one knows exactly what
atoms are present and their first order bonding structure (some second
order effects like chain folding, cis/trans folding might not be known).
 Nanostructures will almost never be that precisely known – if for no
other reason than the number of atoms can be too large for effective,
specific nomenclature.

There is benefit from utilizing a molecular perspective.  There is
policy/procedure in place to handle hazardous molecules.  That
policy/procedure should be adequate (albeit over constraining) to handle
NS safely.  NS are not some brand new form of matter that presents a
totally new set of problems.

Dr. Small:

EPA should probably include a category for “elemental”
nanoparticles, i.e. Au, Ag, Cu, or are these considered simple inorganic
molecules.  To me they seem to be a separate class from say oxides like
TiO2.

Which category includes layered nanomaterials?

Is EPA considering only those materials with 3 nanodimensions i.e.
particles?

Where do functionalized surfaces fit in this scheme?

Dr. Warheit:

No specific comments provided.

Question 2: For the different categories of nanoscale materials, what is
the current state of knowledge about structure and chemical composition?

Dr. BelBruno:

Pure materials (perhaps better defined as single component?) can be
well-defined.  For example, pure carbon-based materials such as
fullerenes or nanotubes may be well-characterized by size, microscopy
and chemical functionality.  Generally impurities consist of remnants of
metal catalysts, which can be detected and (usually) removed. 
Similarly, inorganic materials such as TiO2 nanoparticles may be
characterized by size and structure using standard techniques (described
elsewhere in response to subsequent questions).  

Polymer nanoparticles can present problems with respect to chemical
composition, but may be characterized by size distribution using
standard methods.  Chemical composition is more difficult since, even
for macroscopic polymers, there are impurities and mixed compositions. 
This problem can probably be solved, but requires more attention. 
Composite materials fall into this same category.

Dr. Canady:

I don’t have this detailed knowledge but based on the breadth of the 4
categories, not sure the question is answerable.  If the question were
about carbon nanotubes versus quantum dots or dendrimers you could start
forming a basis for the state of knowledge, but where do you start and
end in a description of the state of knowledge about particles of
inorganic materials?

Dr. Cassell:

For newer nanomaterials systems, knowledge concerning structure and
composition varies widely.  Some nanomaterials systems are very well
characterized whereas in other systems, I would say there is little if
any knowledge.  The extent to which characterization is performed is
often determined by the particular study involved and whether the
researchers/users of the material required this additional
characterization for understanding their system.

Dr. Lahann:

For categories 1 and 2, chemical particle compositions and often also
structures are in many cases well understood.  Mostly, the role of
impurities is often hard to assess and their influence on structure
maybe underestimated in cases.  How composition and structure relates to
function is however often unclear and more work will need to be done. 
Nanomaterials in categories 3 and 4 are often less well characterized. 
Synthetic polymers are associated with variations of their molecular
weight and structure, which equally holds for nanomaterials comprised of
polymers. The extent of this variability may change between different
polymer materials, but is always present at least to some extend. 
Again, it may be important to assess the role of this variability with
respect to function.  Similarly, composite materials can be associated
with substantial variability in chemical composition and structure.

Dr. Murashov:

It is possible to characterize relatively well almost any nanoscale
material.  The main challenge is that this characterization is time
consuming, expensive and limited to only a small sample.  At the present
state of the development, many nanomaterials have wide distributions of
chemistries and structures within the same product batch.  This
heterogeneity of chemical composition and structure results in a range
of properties which are not fully characterized by conventional
measurements.

Dr. Murday:

Presumably NMSP is presently focusing on free standing nanostructures.

In the chemistry/physics research laboratories, where there is generally
careful attention to the identity of the NS being measured, the state of
knowledge is okay.  But there remain limitations to the analytical tools
available for full characterization of NS (as is true generally for all
materials).

There is generally poor knowledge from the standpoint of large scale NS
batches where separation/purification problems are more
difficult/expensive to handle.  This problem is clearly illustrated in
carbon nanotubes.   A couple of years ago one researcher purchased CNT
from two vendors and multiple batches.  He carefully analyzed the
procured material.  See Figure A.  The worst batch had only 40% carbon
in it – and not all of that carbon was likely to be CNT.

NS will frequently (usually) have “surface treatments” or
“shells” to make them compatible with exposure to their intended
environment.  In the case of medical/health applications, the shell will
likely impart recognition/transport functions.  Those same two functions
clearly have environmental implications.

Another level of challenge is the role of “interphase” – roughly
speaking a layer of material enveloping/separating the NS.  

Dr. Small:

Simple organic molecules: The structure and chemical composition of the
fullerenes and C nanotubes are well known for the specific
nanomaterials. However when these materials are sold as a product the
product can be very inhomogeneous and the purity of the commercial
material (structure and chemical composition) is often unknown.  

Simple inorganic molecules: The structure and chemical composition of
simple inorganic molecules like salts and metal oxides are for the most
part well known.  The purity of the commercial product is more
homogeneous that commercial nanotubes but can still be an issue
depending on how the nanoproduct is made. 

Polymeric materials: For materials like dendrimers, lipids the
theoretical structures are known to the degree this structure applies to
what is sold is unknown i.e. particularly for functionalized systems %
coverage. For polymeric materials that “contain” specific compounds
there is very little validation of structure.  

Composites: For more complex materials multifunctional, layered, phase
separated etc. the current state of structure and chemical composition
is much less well known if at all. Several initial studies of complex
nanomaterials indicate that there is significant variation with respect
to structure and composition.

Dr. Warheit:

No specific comments provided.

Question 3: Can structures and chemical composition be correlated to
specific properties and is this correlation quantifiable?

Dr. BelBruno:

As a general rule, we are not at the point where we can correlate
physical properties to chemical structure/composition.  The properties
have the potential to be size-dependent and until theory can provide a
rigorous method to describe the change in properties with size, this
desirable goal cannot be claimed.  For example, silver nanoparticles, in
a select size range, may have antibacterial activity (C. N, Lok, et al,
Journal of Biological Inorganic Chemistry (2007), 12, 527), but this is
not a prior, predictable.  Magnetic properties of nanoparticles also
show a size and structure dependence (S Kar, et al, Journal of
Nanoscience and Nanotechnology (2006), 6, 771; Y.T. Jeon, et al,
International Journal of Modern Physics B (2006), 20, 4390).

Dr. Canady:

Yes and yes.  Why is this considered even potentially uncertain?

Dr. Cassell:

Yes to a certain extent.  For instance, we know that surface area of
nanoscale materials increases as diameter of the material decreases. 
However, direct correlation between chemical composition and reactivity
for instance is not as well understood for nanoscale systems.  I do
think it is quantifiable in a qualitative sense, but perhaps not in a
quantitative sense.

Dr. Lahann:

Structure and chemical composition are critical aspects of every
nanomaterial, and as such contribute to the specific properties of a
material. In some cases, there is a direct correlation to a specific
property, which then is quantifiable. In other cases, structure and
composition influence specific properties through secondary mechanisms.
For instance, nanomaterials of category 3 may be water-soluble and as
such highly resorbable, unless the polymer chains are further
crosslinked. The extent of crosslinking may directly correlate with the
solubility of a material and its resorption. Similarly, molecular weight
or chirality of polymers can influence resorption and degradation
kinetics significantly.

Dr. Murashov:

Yes. Structure and chemical composition can be quantifiably correlated
to specific risk-related properties. Examples are: 

Aerodynamic properties as a function of particle aerodynamic diameter
(Lee and Liu, 1982);

Surface area as a dose metric for low solubility low toxicity
particulate matter (Maynard and Kuempel, 2005);

Biological response as a function of surface charge (e. g. for
dendrimers see Jevprasesphant et al, 2003; Mecke et al, 2004);

Biological response and pharmacokinetics as a function of surface
chemistry (e. g. for functionalized fullerenes and carbon nanotubes see
Sayes, 2004; Singh et al, 2006).

Once nanomaterial’s structure approaches atomic-level consistency
(nano-molecular materials?), it should be straightforward to develop and
reliably employ Quantitative Structure Activity Relationship - like
models. Presently, we have very limited number of nanomaterials which
have such an atomic-level consistency. An example of nanomaterials where
we do have such atomic level control over chemical structure is
dendrimers.

Dr. Murday:

As an academic question, the answer is definitively yes – composition/
structure will define the chemical/physical properties – on NS as well
as any other form of material.  As a practical question the correlation
is difficult because samples are generally not the same level of purity
as we expect for chemical reagents.

I have included a preliminary table that identifies selected properties
and “critical scale lengths,” i.e., properties that will likely vary
significantly from predictions of the solid state (larger scale) /
chemical (smaller scale) models we presently use.  See Figure B. There
could be real value for EPA to sanction a careful examination of
“critical scale lengths” to identify when problems would be more
likely to occur.

One area that is very important is chemical reactivity – clearly an
environmental issue.  Surface science has developed reasonable
understanding of reactions on atomically flat surfaces, and some
understanding of vicinal surfaces (where atomic scale “roughness” is
higher).  But NS have surfaces more complex than these.  The research on
catalysis does address some of that complexity, but the understanding -
because of the difficulty in preparing well defined samples,
characterizing them, and modeling the higher level of complexity – is
not as good.

Dr. Small:

For some properties the answer is yes, for example the color or
absorption of light by specific sized Au particles. The correlation for
most properties related to risk assessment and EHS is not well known.
Efforts to understand these types of correlation are hindered by lack of
validated “standards” and methods.

Dr. Warheit:

No specific comments provided.

Physical-Chemical Properties

	The importance of nanoscale materials is due to their potential for
unique or greatly enhanced properties.  	EPA routinely uses a base set
of physical-chemical property data (e.g., melting point, boiling point,
vapor pressure, water solubility) for a variety of programs (e.g., High
Production Volume Challenge, New Chemicals) and decision-making.  As
mentioned previously, certain material properties are of significant
importance in characterizing nanoscale materials.  Recent research
suggests that particle size, surface area, and surface chemistry (or
surface activity) are initially some of the most important properties to
measure.  

	As expected for most chemicals, class 1 substances having specific
molecular structures and formulas may be more readily studied and
characterized at the nanoscale than the polymer and composite
categories.  For example, carbon-based nanoscale materials as well as
metal oxide nanoscale chemicals are often well characterized. 
Structural and physical-chemical property data therefore are likely to
be well documented for these types of materials.

	

Question 4: Which physical-chemical properties are relevant to nanoscale
materials and how?  Which are known or reasonably ascertainable and
which have data gaps?

Dr. BelBruno:

The prelude to the questions really has it all there.  The important
properties (aside from chemical structure/composition) are particle
size, surface area and reactivity.  Size, surface area and structure are
measurable and standard procedures are possible, but reactivity is more
difficult to quantify and often particle specific.  Techniques such as
dynamic light scattering, BET procedure and x-ray diffraction,
respectively, are routinely employed.

Dr. Canady:

It seems you have to first express the decisions that need to be made
about the materials and then what is minimally sufficient for decisions
and therefore relevant.  I am afraid I am not familiar enough with TSCA
to begin this process. 

There is a risk here of developing lists of what we can do or can think
of and then letting them stand forever as requirements.  The list of
relevant properties should flow from the decision needs.  For example,
some recent reviews focus on surface area because of the thought that it
may correlate better with toxicity than mass does.  However, other
measures may be enough to capture the listing decision needs or even
reportable quantity determinations without the need to measure surface
area in any specific way. 

Basing property needs on a minimally sufficient set to make decisions
could more easily consider both cost and utility so that unnecessary
costs can be avoided. 

On the other hand, at least initially and until we get a better sense of
what is useful for decisions, the response to this question could be
rather long for the general case of all nanoscale material so that we
can begin to sort through the utility functions. 

There are likely to be properties relevant to decisions about specific
classes of materials but not others.

Dr. Cassell:

The most relevant physical-chemical properties are crystallinity,
surface properties (charge, roughness, and area), and solubility state
of agglomeration because these properties directly influence the direct
assessment of toxicological risk.  I think that most new materials the
crystallinity has been characterized well but the state of agglomeration
is probably not as well understood.

Dr. Lahann:

As for all other materials, nanoscale materials have specific sets of
properties that can be used for characterization.  In addition, these
materials often have specific properties due to the nanoscale
information encoded in their structure.  This information may be
transient and alter with temperature.  For instance, for a given CdSe
nanoparticle material, its optical properties are – within a certain
range – a function of the particle diameters.  Another example is the
magnetic properties of iron oxide particles, which alter with diameter. 
In general, properties that are associated with the nanoscale structure
of a material are better suited for characterization of nanomaterials. 
Often, the characterization of individual nanoobjects on the basis of
size and size distribution, shape and shape distribution, chemical
composition, structure, etc., may be more insightful than “bulk
characterization”.

Dr. Murashov:

All listed physical-chemical properties in the discussion paper are
relevant to risk potential of nanomaterials:

Particle size and distribution

Particle shape and dimensions

Agglomeration and aggregation

Surface area

Surface charge

Surface chemistry

Chemical composition

Crystal structure

Impurity identification and levels

In addition, the following are important parameters for characterization
(see also draft ISO TC229 TR “Guidance on physico-chemical
characterization of engineered nanoscale materials for toxicologic
assessment”):

Solubility;

Porosity;

Stability under different experimental conditions (biopersistence,
durability and dispersibility).

In particular, since interactions between biological species and
nanomaterials occur primarily at the nanomaterial surface, it is
critical to characterize not just the chemistry of nanomaterial core,
but also chemical species represented on the surface of nanomaterials. 
For example, if one considers titania nanoparticles coated with silica
and made to reflect UV light, the commercially important active
ingredient is titania.  However, since silica is on the surface, the
biological response is expected to be to silica at least initially until
silica coating is dissolved or otherwise altered. 

All properties can be measured at a research level.  However, presently
atomic-level description is limited to individual nanoscale features
such as individual nanoparticles.  There is a lack of standardized
methods and techniques to conduct characterization of representative
nanomaterial samples routinely that is cost- and time-effective.

Dr. Murday:

All physical/chemical properties are relevant to NS.  Perhaps the more
important question is which of those properties might vary significantly
at the nanoscale and are those properties of environmental concern.

For the NS itself one can address this question by examining property
vice critical scale length – two generic specific issues to watch for
are quantization (size) and correlation (many atoms) effects.

Many, if not most NS, will have a core/shell(s) structure – i.e., they
will not be a monolithic entity.  The shell(s) will impart
different/additional physical and chemical properties.  As we move more
to “smart” structures – those that sense and respond - the
complexity of those added properties will grow.

In addition, NS can have an “interphase.”  This will be a tougher
problem since the “interphase” may be generally even more complex
and hard to characterize.

Dr. Small:

I think it will be important to define relevance to what.  In general,
properties such as size, surface area, chemical structure, surface
characteristics, functionalization, and charge are relevant to the
properties of most nanoscale materials.  Different materials will have a
different set of properties that are relevant to its use or hazard
evaluation.  Most methods for the determination of physical-chemical
properties of nanomaterials have large gaps.  Critical to minimizing
these gaps is to provide information that is consistent and reproducible
from one lab to the next.  Accuracy may well be a secondary
consideration.

Dr. Warheit:

No specific comments provided.

Question 5: Are there properties that would have little or no relevance
under the NMSP?

Dr. BelBruno:

At this point, it is not at all clear which, if any, properties have
little relevance.  I do not believe that any measurements should be
ignored or discouraged by the standards that might be set.

Dr. Canady:

There is probably a list of things that are not relevant from the
chemical realm, which EPA might have included in a list of what to ask
for out of habit.  For example, volatility measures that you might
require for chemicals may be valueless. 

I can’t think of a case for a categorical exclusion on a property
other than things like volatility (i.e., property measures that just
come up blank because they don’t apply to particles).  We probably
don’t know what will have little or no relevance in all cases.  There
would need to be a theoretical basis to “prove a negative” and this
may be particularly difficult while we are in the middle of discovering
behaviors and uses as a result of manipulating properties. 

Perhaps when SAR or other methods of relating properties to function
evolve to a point of understanding null points in chemical and
biological interaction we can begin to develop maps of where there are
not effects.  Seems likely they would be multivariate likelihood
statements rather than “this property never interacts” with biology.

Dr. Cassell:

I think all of the information is useful.  In certain contexts, optical
properties (photoelectric properties for instance) would not be relevant
to NMSP.  All of the well-accepted physical chemical properties should
be relevant to NMSP however.

Dr. Lahann:

See response to question 4.

Dr. Murashov:

At this point, it appears that the following properties have little
direct relevance to the risk potential of nanomaterials:

thermal properties such as thermal conductivity, heat capacity and
thermal diffusivity;

optical properties such as refractive index, optical rotation,
luminescence;

elastic properties such as Young modulus and elastic constants.

Dr. Murday:

Presumably the emphasis is on impacting living systems.  There is not
enough NS volume to worry about unsightly/unwieldy waste piles.

Some likely properties with little/no relevance from an environmental
protection perspective - thermal conductivity, plasmonics,
superconductivity.

Dr. Small:

It is reasonable to assume that so little is known regarding the
relationship between properties and impacts that most properties have to
be considered relevant until definitive studies are conducted to show
they are not relevant.  That said, I would think the determination of
mass alone will not be relevant to NMSP.

Dr. Warheit:

No specific comments provided.

Question 6: Which properties are associated with aggregated or
agglomerated nanoscale materials, as opposed to properties that are
inherent to the material regardless of physical form?

Dr. BelBruno:

If we are dealing with aggregates or agglomerated materials, we need to
measure those properties cited above, size, surface area, reactivity)
for both the components and the aggregates.  Essentially, we need
“core and aggregate” measurements.  Obviously, the chemical
composition/structure issue arises in these cases as well.  This issue
of properties and agglomeration affecting dosage rates for in vitro
studies was covered in a recent review paper: J.C Teeguarden, et al,
Toxicological Sciences, (2007) 95 300.

Dr. Canady:

I don’t know.

Dr. Cassell:

Surface charge, surface area, surface chemistry, geometry, and
solubility are properties that come to mind for aggregates and
agglomerates.

Dr. Lahann:

Most properties are depending on the dispersion state of the
nanoobjects.  Some times, individual nanoobjects can self-assemble into
well-defined structures leading to novel functions and properties.

Dr. Murashov:

Transport properties will depend greatly on the state of aggregation/
agglomeration. Specific examples are

diffusion coefficient;

precipitation rate;

surface deposition rate (lungs, skin);

rate of penetration of filters, membranes, fabrics;

translocation rates in vivo.

Effect of aggregation/agglomeration on biological activity will be
determined by the mechanism of biological activity.  For example, for
poorly soluble low toxicity nanomaterials if aggregation/agglomeration
results in changes in the total surface area available for interaction
with biological species, then the biological activity of such materials
is expected to change.  For example, pulmonary exposure of rats to a
well dispersed a sample of ultrafine carbon black was more toxic and
inflammatory than an equal dose of a less dispersed sample (Shvedova et
al., 2007).

Dr. Murday:

Buoyancy (size/weight); capillarity; Ostwald ripening; grain growth;
pore size distribution; hybrid structures – link a variety of
different NP each with different properties (nano Au color in solution
as an example); electrical/thermal transport in compact; mechanical
behavior of compact; supported catalysis.

Dr. Small:

Agglomeration may affect size distributions from indirect methods such
as DLS and surface area measurements.  In addition agglomeration will
also affect properties such as respirability, movement in environment
and possible uptake.

Dr. Warheit:

No specific comments provided.

Question 7: Are there routine manipulations of nanoscale materials that
result in physical-chemical properties changes or other defining
characteristics (e.g., surface modifications of nanotubes to enhance
solvent dispersibility)?

Dr. BelBruno:

Obviously, the cited case of surface modification is the main issue
here.  As may be inferred from responses to previous queries, such
modification necessitates re-evaluation of the nanoparticle properties. 
Surface modifications include functional groups directed towards other
applications, for example, to stealthly transport materials in
biological systems. Usenko, et al, in an attempt to establish a rapid
screening assay, have shown that C60(OH)24 was “significantly” less
toxic than C60. (Carbon, (2007) 45 1891)

Dr. Canady:

I don’t have specific knowledge of this but there are surely lots of
data being developed and discarded in development of materials now that
would speak to this.  There is probably not a finite set or a small list
of key manipulations, and the manipulations that result in relevant
change will vary based on the material and the matrix or setting it is
used in.

Dr. Cassell:

Yes, there are a number of physical and chemical treatments that can
alter the defining characteristics of nanoscale materials.  Vacuum
treatment, thermal annealing, drying can all drastically influence the
defining characteristics of nanoscale materials.

Dr. Lahann:

Fundamentally, dispersion requires the offset of interparticulate
interactions with solvent/particle interactions.  Changing the surface
chemistry is an effective way of changing the surface energy of the
particle and can therefore be used to tune the balance between these
competing interactions.  Similarly, change the solvent system can have
similar effects.  Finally, the application of electrical or magnetic
fields may be used to tune dispersibility.

Dr. Murashov:

My understanding is that depending on application some nanomaterials are
routinely modified to 

improve biocompatibility (e. g. surface functionalization of medical
nanoapplications with polyethylene glycole);

protect from dissolution and improve photoluminescence by surface
passivation (e. g. coating of quantum dots with ZnS in CdSe/ZnS and/or
organic ligands);

improve dispersibility and reduce agglomeration (e.g. use of surfactants
to disperse carbon nanotubes).

Dr. Murday:

“Routine” varies depending on NS and application, i.e., it is not
clear that the word “routine” is valid in this context.

It is likely that most NS will require a “shell” to make them
compatible with their intended application(s).  The creation of that
shell can happen by many variants of either gas (including low pressure
– vacuum) or liquid deposition.  Since the environment interacts
directly with the outermost shell, it should receive careful attention.

Dr. Small:

Materials may be modified in the purification process i.e. CNT can be
acid washed to remove catalyst which may functionalize surface. 
Chemicals may be added during production to modify surface
characteristic to prevent aggregation.  Milling will change the surface
characteristics.

Dr. Warheit:

No specific comments provided.

Question 8: How should physical-chemical property data be prioritized
for the NMSP?  Based on availability, effect (toxicity or exposure
criteria), or other factors?

Dr. BelBruno:

The preferred method should be toxicity.  However, it is not at all
clear that such data would be available for each nanoparticle.  The best
approach would be to begin with those nanomaterials for which we do have
data and construct what could be a prototype nanomaterial MSDS. 
Obviously, the test cases would be “pure” materials such as carbon
nanotubes or TiO2 or many of the quantum dot materials.  However, we
must establish consistent (and meaningful) standards for in vitro
testing.

Dr. Canady:

Not sure what is meant by priority in this case but there is probably a
set that is relevant to decisions and a broader set where the intention
is to build knowledge of what might be relevant to decisions.  Variables
that would be informed by the data might include 

Classification for regulatory purposes like things that would change
inventory status or inclusion in a SNUR,

Fate and transport (and exposure or bioaccumulation or magnification or
sequestration in odd places as a result), and

Toxicity or environmental effect

The degree to which data inform decisions in consideration of these
variables would seem to be the main prioritization criteria.

Dr. Cassell:

The NMSP should take a careful and comprehensive approach to
understanding nanoscale materials.  Priortization of data should be made
based upon the detailed context of the question and the specific
situation.  There are also many poor sources of information or
information that may not be valid.

Dr. Lahann:

Based on effect.

Dr. Murashov:

Nanoscale Materials Stewardship Program goal is to better understand the
potential risks and benefits of nanotechnology, therefore,
physical-chemical property data should be prioritized according to its
relevance to risk assessment and risk management of nanomaterials.
However, it is also important to collect all the available information
relevant to risk assessment (hazard and exposure) and risk management.

Dr. Murday:

Risk is roughly the product of exposure and harm.

Free standing nanostructures can lead to exposure; embedded might be
released if not covalently bonded, otherwise parts of matrix will likely
be attached; “thin film” nanostructures likely will never be
available for incorporation into living systems.

Higher priority properties might be:

Chemical reactivity

Solubility/Partition (largely governed by surface properties and Hamaker
constants)

Dr. Small:

The prioritization of the physical-chemical property data to be relevant
needs to be based first on measurements that can be made in a
reproducible manner both intra - and inter - laboratory. Next once these
methods have been identified then their effect/application to risk
assessment, (toxicity, exposure etc.) needs to be applied to the
prioritization process.

Dr. Warheit:

With regard to human health effects or ecotoxicity testing, there are
however, certain generalized points relating to material
characterization for all nanoparticle-types that should be emphasized. 
These are the following:

Many scientific organizations or task forces have strongly recommended
that toxicologists adequately characterize physicochemical properties of
the nanoparticle-types that are being evaluated for hazard testing. 
However, too often this recommendation becomes a “laundry list” of
physicochemical characteristics and does not have adequate
prioritization.  As a consequence, in order to adequately describe the
physical characteristics of the nanoparticle-type being evaluated, I
would recommend that toxicologists characterize the following
(prioritized) physicochemical properties prior to conducting hazard
studies with nanoparticle-types:

Particle size and size distribution (wet state) and surface area (dry
state) in the relevant media being utilized – depending upon the route
of exposure;

Crystal structure/crystallinity;

Aggregation status in the relevant media;

Composition/surface coatings;

Surface reactivity;

Method of nanomaterial synthesis and/or preparation including
post-synthetic modifications (e.g., neutralization of ultrafine TiO2
particle-types);

Purity of sample;

This represents a focused approach concomitant with a minimum,
standardized assessment of physicochemical properties that should be
investigated prior to the development of toxicity testing with
nanoparticles.

Additional points to consider for material characterization:

Depending upon the exposure scenario, it may be appropriate to conduct
assessments of nanoparticle characteristics in both the wet state (in
the relevant media) and the dry state.  Therefore a BET surface area
measurement may not suffice – but should be carried out in parallel
with different types of measurements such as cryo TEM (dry state) and
DLS (dynamic light scattering - particle sizing in the wet phase).

The chemical composition of the nanoparticle core vs. shell (hard or
soft) should be evaluated.

In this regard, the surface chemistry could be specific for each
nanoparticle-type.

Below, I have listed 5 recently-published nanoparticle toxicity studies
from our Laboratory concomitant with the specific physicochemical
characteristics that were identified from each of the studies. 
Accordingly, we have concluded that characteristics of nanoparticle
surface reactivity appeared to be the best predictors of in vivo
pulmonary toxicity in rats.  However, it should be pointed out that
surface reactivity evaluations are particle-type specific (e.g. Vitamin
C assay for TiO2 particles; erythrocyte hemolysis for silica particles).
 Therefore, the same tests for surface reactivity cannot be utilized for
each nanoparticle-type.

A Listing of Material Characterization Methods utilized in the 5 studies
listed below

“Pulmonary Instillation Studies with Nanoscale TiO2 Rods and Dots in
Rats: Toxicity is not dependent upon Particle Size and Surface Area”.
Warheit DB, Webb TR, Sayes CM, Colvin VL, and Reed KL. .  Toxicol Sci.
91: 227-236, 2006

Material characterization employed in this study:

synthesis method

crystal structure

particle size

surface area

composition/surface coating

aggregation status

cryo TEM

crystallinity

purity (TGA) 

“Pulmonary bioassay studies with nanoscale and fine quartz particles
in rats: Toxicity is not dependent upon particle size but on surface
characteristics”. Warheit DB, Webb TR, Colvin VL, Reed KL and Sayes
CM.  Toxicol Sci. 95:270-280, 2007;  

Material characterization employed in this study:

synthesis method

crystal structure/crystallinity (XRD)

median particle size - particle size (range)

purity (% Fe content)– ICP-AES

surface area

TEM

aggregation status

purity

surface reactivity (erythrocyte hemolysis)

reactive oxygen species (ESR) 

“Pulmonary Toxicity Study in Rats with Three Forms of ultrafine-TiO2
Particles:  Differential Responses related to Surface Properties”.
Warheit DB, Webb TR, Reed KL, Frerichs S, and Sayes CM. Toxicology
230:90-104, 2007; 

Material characterization employed in this study:

crystal phase

median particle size and size distribution in water and PBS

pH in water and PBS

surface area (BET)

TEM 

aggregation status

chemical (surface) reactivity – (Vitamin C assay)

surface coatings/composition, purity

Assessing toxicity of fine and nanoparticles: Comparing in vitro
measurements to in vivo pulmonary toxicity profiles Sayes CM, Reed KL,
and Warheit DB.  Toxicol Sci, 97:163-180, 2007.

Particle-types utilized in this study:

Fine-sized carbonyl iron

Fine-sized crystalline silica

Fine-sized amorphous silica

Nano ZnO

Fine ZnO

Particle characterizations conducted both in the “dry state” and
“wet state” 

Material characterization employed in this study:

Particle characterization in the dry state

particle size

surface area

density

calculated size in dry state (based on surface area determinations)

crystallinity

purity

 

Particle characterization in the wet state

particle size  in solutions – PBS, culture media, water 

average aggregated size in solutions

% distribution

surface charge

aggregation status

 

Conversion and comparisons of in vitro and in vivo doses for dosimetric
comparisons

Comparative Pulmonary Toxicity Assessments of C60 Water Suspensions in
Rats: Few Differences in Fullerene Toxicity In Vivo in Contrast to In
Vitro Profiles.    Sayes CM, Marchione AA, Reed KL, and Warheit DB. Nano
Lett. 2007 Aug;7(8):2399-2406. 

Material characterization employed in this study:

particle size and size distribution

surface charge

crystallinity

TEM

Composition

oxidative radical activity (ESR measurements)

surface reactivity (erythrocyte hemolytic potential)

Nanoscale manufacturing and processing

	The number of manufacturing and processing methods for generating
nanoscale material continues to grow and become more sophisticated.  The
two primary areas for this discussion include physical reduction methods
(milling) and engineering methods (e.g., particle stabilization, vapor
deposition, self assembly).  

Question 9: What are the common processes used to manufacture nanoscale
materials?

Dr. BelBruno:

The list in the paragraph above covers the general “bottom up” and
“top down’ methods.

Dr. Canady:

A list was developed in the ICON workshop last year at NIH – I don’t
have the full list, but it is extensive.

Dr. Cassell:

Solution based methods are quite common and include a number of
approaches (self-assembly, colloidal preparation), chemical vapor
deposition is also used extensively.  All techniques conventionally used
in organic and inorganic synthesis are also adapted for nanoscale
materials manufacture.

Dr. Lahann:

Organic materials: self-assembly

Inorganic materials: CVD

Polymers: staged synthesis, self-assembly, suspension polymerization,
grafting, electrified jetting

Composites: similar to polymers

Dr. Murashov:

The draft ISO TC229 Technical Report on “Health and Safety Practices
in Occupational Settings Relevant to Nanotechnologies” lists the
following categories of major processes used to manufacture nanoscale
materials:

Gas phase processes such as flame pyrolysis, high temperature
evaporation and plasma synthesis;

Vapor deposition;

Liquid phase methods: colloidal, self-assembly, sol-gel;

Supercritical fluid;

Electro-spinning for polymer nanofiber synthesis;

Mechanical processes including grinding, milling and alloying;

Bioreactors.

Dr. Murday:

Since NS, by definition, have large surface to volume ratios, quality
manufacturing processes must utilize a controlled environment.

For large scale bulk NS the processes are largely either a variant on
colloid chemistry or gas phase nucleation and growth.  Fumed silica and
carbon black are examples of the latter – both can be NS and have been
made in volume for decades.

Dr. Small:

Common methods would include: lithography, vacuum and spray coating,
CVD, Vapor deposition, Plasma vaporization, laser ablation,
arc-discharge, gas-phase catalytic growth of CNT from carbon monoxide,
and Sol gel.  There are of course many “variations” on these methods
some of which are proprietary.  In the case of the more complex
materials these would probably start with one of the above and then have
proprietary additional steps.

Dr. Warheit:

No specific comments provided.

Question 10: How are processes used to produce specific characteristics
or properties?

Dr. BelBruno:

The size of synthesized particles can be controlled through the
chemistry, either by limiting reagents or by the specific conditions of
the chemistry (heat, solvent, reaction mechanism, etc).  To the extent
that properties reflect size (and surface area), then the properties are
controlled by the chemistry.  Of course, that relationship
(size-properties) is not yet well-defined.  Similarly, self assembly can
be chemically controlled and the surface characteristics, therefore,
influenced by the process.  Finally, chemistry can be used to modify or
functionalize the particles.  We know that functionalized nanotubes, for
example, have different properties from the naked nanotubes.

Dr. Canady:

I don’t know.

Dr. Cassell:

Self assembly for instance can be used to control nanoparticle diameter.
 Polymerization techniques (stoichiometric control, temperature,
pressure, etc.) can be used to control engineered polymer systems. 
Thermal/chemical treatment can be used to increase surface area or
surface chemistry.  There are a wide variety of techniques borrowed from
conventional disciplines that are used in the manipulation of nanoscale
materials.

Dr. Lahann:

By exactly controlling chemical composition and structure of
nanoparticles/nanoobjects, a specific characteristic or property can be
achieved.  For instance, in the case of CdSe nanocrystals, the particle
diameter defines important optical properties and developing processes,
which can generate monodisperse particles with defined diameters, will
give a unique function.

Dr. Murashov:

This depends on the process, materials and application.

Dr. Murday:

After the formation of most NS it will be necessary to tailor the
surface to make the NS compatible with further processing.  The tailored
surfaces can be thought of as a shell.  There may be several shells as
we progress toward “smart” NS.

As an example, the first shell of a semiconductor QD is designed to trap
the electron-hole pair in the core, a second shell might have a
therapeutic function and a third shell a recognition function.

Processes used to produce the shells will likely vary considerably, but
would again be mostly variants on colloid/gas phases processing.

Dr. Small:

No specific comments provided.

Dr. Warheit:

No specific comments provided.

Question 11: Which methods reduce particle size but do not result in
property changes?  Which methods reduce particle size and result in
property changes?  

Dr. BelBruno:

See response to question 10.

Dr. Canady:

First, the question should be about change not reduction.  Size going up
from 50 to 200nm will cause different disposition effects in the body,
so increase in size in the nanoscale range is relevant as well. 

But I don’t understand this question.  Size is a property that has
relevance in biology on its own.  If “property” as a function of
size means lack of biological or environmental impact, then how could
you know that properties are not changed without testing?  How could we
know the answer to this question unless the answer is that all methods
to change size will result in some unknown likelihood of a change in
biological or environmental response?  This is a smaller set of the
question of whether you can assume that any modification will have no
effect (by limiting the modification to “size change” rather than
“any modification”).

Dr. Cassell:

This depends on the system under question.  Cutting a carbon nanotube in
half reduces its particle size, but does not greatly effect other
properties.  Reducing gold particle diameters dramatically alter melting
point for instance.

Dr. Lahann:

If a material’s change in particle size results in fundamental
property changes, the outcome is more a function of the material itself
and less of the method used to reduce the size.  However, the process
used to reduce the size of a material may determine polydispersity and
may still be influencing properties.

Dr. Murashov:

If the reduction in size results in changes in properties of interest,
then it does not matter how the reduction in size was achieved.  If
properties of interest also depend on surface characteristics, then the
production method can affect these properties and additional steps such
as surface passivation might be required (see also answer to question
7).

Dr. Murday:

The question is not meaningful – all methods may/may not change
properties.

Processes without a controlled environment are likely to lead to wider
variation in surface composition and thereby different behaviors.  For
instance, the technique of ball milling larger particles into smaller
frequently is inadequately controlled; compacted powder studies have
suffered from inability to reproduce results.

To control size, processes growing NS will need to have a nucleation
event separate from the growth step.  If nucleation and growth proceed
in parallel, there is an inevitable distribution in NS size.

Dr. Small:

Many of the production methods are bottom up so this question does not
apply to those. In general the property change results for the nanoscale
characteristics of the materials not on how they are produced.  Not sure
what this question refers to.

Dr. Warheit:

No specific comments provided.

Impurities

	Impurity content is a growing area of interest in nanotechnology due to
improved performance observed in some cases (e.g., solar cells and
semiconductors) and deleterious effects observed in others (e.g.,
quantum dot quantum computers).  The confounding effects that impurities
have with respect to toxicological endpoints are also being studied
(National Nanotechnology Initiative 2006 Environmental Health and Safety
research report).

Question 12: How important are impurity identity and impurity levels to
the understanding and characterization of nanoscale materials?  

Dr. BelBruno:

Impurity level and identity are important factors, since the impurities
are also nanometer scale and their own properties add to the difficulty
in assessing the impact of the target particles.  Again using nanotubes
as an example, the catalytic metals that are sometimes incorporated in
the final product need to be identified, quantified and, often, removed.
 This has been accomplished in a number of different ways.  One example
is provided by Y. Wang, et al, Chemical Physics Letters (2006), 432,
205.

Dr. Canady:

Impurity identity is critical and not clear at this point.  It could be
defined relative to the set of nanoscale properties of interest which
would also be the specification of the material, or of the nanoscale
particle.  Something could be an impurity if it has slight topographic
variation from the optimal and conversely, something may not be an
impurity if it has dramatically differing composition but similar enough
structure so that the effect does not depend on composition. 

Impurity could also be defined based on unwanted side effects of a
variation in nanoscale material even if the nanoscale material that is
“impure” (in the sense of its unwanted side effects) is “pure”
with respect to its intended effects.

Impurity could also just be a solvent or carryover from the
manufacturing in the typical sense, such as the PAHs in CNTs.

The need to measure impurity level is a function of effect and would be
case-dependent.

Dr. Cassell:

Extremely important.  I would say that impurities become more relevant
as you scale your system to smaller dimensions.  Impurities could be
even more important in the context of toxicological effects.

Dr. Lahann:

Depending on the application, the effects can either be hugely important
or can be neglected. However, impurities always introduce an additional
source for structural variability into a nanomaterial.

Dr. Murashov:

There is a need to clarify what is meant by “impurity”.  Term
“impurity” when it is applied to nanomaterials is more similar to
molecular chemicals (e. g. organic polymers) than to macroscopic
materials with a high degree of ionic or metallic bonding (e.g. SiO2,
Si, NaCl).  Specifically, target nanomaterials can be contaminated by
other nanomaterials and chemicals (example is a raw carbon nanotube
material composed of a mixture of carbon nanotubes and nanoscale metal
catalyst particles).  Nanomaterials can be also “contaminated” by
atomic/molecular species covalently bonded to nanomaterials.  If such a
“defect” or “contamination” results in significant changes of
nanomaterial properties, it could be considered a new nanomaterial
rather than “contaminated” or “defective”, just as replacement
of an atom in a molecule for a different chemical element creates a new
molecule/chemical.  Presently, in many instances nanomaterials are
produced as mixtures of nanoparticles with a range of particle sizes,
shapes and chemistries, which is, to some degree, similar to a mixture
of oligomers obtained during polymer synthesis. 

The importance of “impurity” or “impurity” level (or the amount
of other molecules and nanomaterials in the mixture) for a particular
application will be strongly dependent on that application.  The
importance of “impurity” or “impurity” level on risk of a target
nanomaterial will be strongly dependent on risk-related properties of
the “impurity” and the target nanomaterial.

Dr. Murday:

Impurity identify and level can be very important, i.e., doping of a
semiconductor NS.  The NS size is sufficiently small that a single
dopant atom will have dramatic impact on its electrical properties. 
Doping at the nanoscale might be better accomplished by injecting charge
through surface bond compensation.

CNT manufacture provides an excellent case study in quality control (or
lack thereof).  Figure A shows the results of an analysis of eight
different nominal CNT samples – the worst has only 40% carbon (as a
fraction of C, H, O; all other elements were excluded from the
analysis), and that carbon might not be NT.

Dr. Small:

This is a critical area when characterizing these materials. For
example, the catalyst in NT production is nanosized metal (Fe, Ni)
particles. Some studies indicate that these may be more toxic than the
NTs.  The ability to determine the amount and the physical-chemical
character of the impurities contained in ENM relates directly to
understanding and evaluating exposure and risk.

Dr. Warheit:

No specific comments provided.

Question 13: Are there routine purification procedures that can
effectively control or remove impurities, when desirable, for certain
classes of nanoscale materials?  

Dr. BelBruno:

For nanotubes, the response is yes (see Q12).  Removing impurities from
inorganic materials is more difficult because procedures could also
affect the desired particles.

Dr. Canady:

I don’t know.

Dr. Cassell:

Yes.  There are many techniques that are borrowed from those known in a
number of disciplines (chemistry, materials science, biology) and used
in nanoscale materials manipulation.  They are too numerous to list
here, but include techniques such as annealing, chemical treatment,
separation, chromatography, sedimentation, plasma treatment.

Dr. Lahann:

Processes to remove impurities from nanoscale materials are often
available.

Dr. Murashov:

The following broad classes of purification processes can be
distinguished:

chemical modification to reduce heterogeneity;

dissolution of ionic, molecular and particulate species from target
nanomaterials (e.g. acid-washing has been reported as a means of
removing metal catalytic particles from raw carbon nanotube material);

mobility sizing using difference in mobility of nanomaterials of
different sizes, masses, charges and shapes in gas and liquid phases,
across membranes and subject to force-fields. 

Dr. Murday:

Chemistry provides a rich source of purification processes that can be
exploited for NS.  One limitation could be difficulty in sustaining
solubility/suspension in fluid media.

Dr. Small:

Acid washing for CNT along with other steps has been used for
purification of CNTs.  There are several lab-based methods these are not
routine.

Dr. Warheit:

No specific comments provided.

Obtaining characterization data 

	Determining identity, quantifying the nanoscale particle range, and
measuring physical-chemical properties for that identity and particle
range are essential to the characterization of nanoscale materials. 
Because of the challenges associated with size, shape, surface
characteristics, and possibly other aspects of nanoscale materials, an
evaluation of existing measurement techniques is critical to nanoscale
material characterization.  The National Nanotechnology Initiative
report stated that “...Accurate and useful measurement techniques are
also important because agglomerated nano materials may either retain or
lose their emergent properties - or take on new properties - thus
affecting the potential biological response.”  

Question 14: Are validated methods available for the different
categories of nanoscale materials?

Dr. BelBruno:

Particles are easily sized using commercially available equipment.  This
measurement is a bit more difficult (larger uncertainty) for coated
nanoparticles.  Force and electron microscopy may also be used to
measure the size of particles.  These are essentially universal, but
tedious, methods.  X-ray spectroscopic techniques can be used to
identify metallic impurities (in many electron microscopes).  Surface
area measurement is also a well-developed technique. (see Q4)

Dr. Canady:

No specific comments provided.

Dr. Cassell:

I am unaware of validated methods for nanoscale materials in particular,
but there may be methods (e.g. ISO) that are inclusive of nanoscale
materials.

Dr. Lahann:

Some validated methods are available, but further efforts are needed.

Dr. Murashov:

There are no validated methods specific to engineered nanoparticles
(NIOSH 2006).  There are validated methods for materials that exist in
the nanometer scale, such as welding fume, carbon black; but these
methods do not discriminate by size.  There are some methods that have
been validated for the larger (bulk) form of nanomaterials that may be
applied to the nano form; for instance dust sampling with a cyclone
(ISO, 2007); sampling for a specific element with analysis by NIOSH 7300
(NIOSH, 1994); fiber sampling and analysis by Transmission Electron
Microscopy (TEM).  These methods have not been validated for nanoscale
material, but there is potential for reapplication and they can provide
useful information for evaluating occupational exposures with respect to
particle size, mass, surface area, number concentration, composition,
and surface (NIOSH, 2006).  The challenge is demonstrating that the data
generated by the measurement (the metric) was created by the engineered
nanoparticle rather than by incidental nanoparticles of similar chemical
composition.  Variations of the sampling process for mixtures of
nanoscale and fine TiO2 are being tested (NIOSH, 2005).  One of the
challenges of that method is collecting enough material for the first
mass-based step in the analysis.  Often the electron microscopy analysis
is used to verify that the material collected has the expected elemental
composition and is in the nanometer size range though there is always
some degree of agglomeration.

Dr. Murday:

MS and light scattering are the closest to validated.

NIST and NCL are working this issue – EPA should partner.

Dr. Small:

There are several methods under development be few in publication.  One
I am aware of is IEEE P1650 standard test method of meas. of electrical
properties of CNTs adopted in 2005.

Dr. Warheit:

No specific comments provided.

Question 15: Are there techniques that can be universally applied?

Dr. BelBruno:

See response to question 14.

Dr. Canady:

No specific comments provided.

Dr. Cassell:

I would think so, yes.  Especially the more generic particle
characterization techniques (surface area, particle size determination,
etc.) that have been utilized in other materials characterization work.

Dr. Lahann:

Particle size and shapes can be characterized using Transmission
electron microscopy, scanning electron microscopy and scanning probe
microscopy independent of the materials composition.

Dr. Murashov:

The collection and analysis of an air sample by TEM is an example of a
universal technique which can be applied across several types of
nanomaterials.

Dr. Murday:

In a broad context, there are no techniques that provide all the
necessary information, all the time.  NS can be found in free form and
embedded in matrices.  The NS itself is likely to have
composition/structure that varies with distance from its center.

Light scattering is likely to be useful for size/shape measurement.  It
might be useful/important to extend the photon energy in the far UV to
make the wavelength more comparable to the NS size (however, this may
invoke a problem for the optics).

Mass Spectrometry has been developed to handle very large molecules –
100s of KDa.  Fragmentation patterns can illuminate substructures.  This
technique could prove very useful in probing core/shell NS.

MS and light scattering require multiple copies of a NS and yield
ensemble averaged results.  Techniques to probe individual NS properties
are also very useful.  Some variant on HRSEM with energy dispersive
analysis for composition, coupled with AFM, is likely to be the most
versatile.

SERS might be developed with enough sensitivity to analyze one or a few
NS for a vibrational approach to composition.

Dr. Small:

Not that I am aware of.  Perhaps some of the microscopy methods could be
used to characterize multiple classes.

Dr. Warheit:

No specific comments provided.

Question 16: For small quantities of materials, are there sampling,
handling, and collection techniques as well as sample integrity,
accuracy and precision QA/QC methodologies available?

Dr. BelBruno:

Many nanoparticles are available from commercial suppliers (quantum
dots, fullerenes, nanotubes, inorganic cluster molecules), with
specified purity levels and chemical composition.  Clearly, commercial
suppliers can arrange QA procedures and we should look to these methods
as a means of introducing standards.

For example, from American Elements
(http://www.americanelements.com/fenp.html): “Iron (Fe) Nanoparticles,
nanodots or nanopowder are spherical or faceted high surface area metal
nanostructure particles. Nanoscale Iron Particles are typically 20-40
nanometers (nm) with specific surface area (SSA) in the 30 - 50 m2/g
range and also available in with an average particle size of 100 nm
range with a specific surface area of approximately 7 m2/g. Nano Iron
Particles are also available in Ultra high purity and high purity and
coated and dispersed forms.”

Dr. Canady:

No specific comments provided.

Dr. Cassell:

I would say that there may be generic techniques published in the
technical literature, but these are probably not standardized by a
leading organization.  Quality control procedures for nanoscale
materials are largely dependent upon the organization that handles the
materials and most of these are based upon the needs of that particular
organization.

Dr. Lahann:

Some specialized procedures are available, but further efforts are
needed.

Dr. Murashov:

At this point, the greatest QA/QC challenge is the lack of methods
specific for nanoscale particles and the lack of reference materials
that accurately represent the material being evaluated.

Dr. Murday:

The traditional handling of chemicals provides proven precedents.  But
they will have to be adapted to the broader range of size/properties
incumbent in NS.

Testing on NS shows (at least in some circumstances) that they can be
collected effectively on filters – van der Wall forces or surface
charges attract to the filter material; NS are small enough that fluid
drag forces are small and don’t displace them from the filter.

Dr. Small:

No.  I think best-practices methods (handling of ultrafines) are used
although there are several efforts by government agencies and industry
to develop safe handling practices for ENMs that would apply to
“small” quantities.

Dr. Warheit:

No specific comments provided.

Question 17: What is the status of standardization efforts? Are these
efforts focused on broadly applicable characterization methods or
category-specific methods?

Dr. BelBruno:

Mostly category specific, as the response to Q19 indicates.

Dr. Canady:

No specific comments provided.

Dr. Cassell:

NIST is active in generating standards for single walled carbon
nanotubes see:    HYPERLINK
"http://polymers.nist.gov/Nanotube3/Agenda.pdf" 
http://polymers.nist.gov/Nanotube3/Agenda.pdf .  There are a variety of
techniques described for characterizing single walled carbon nanotubes.

Dr. Lahann:

Depending on the field and material, standardization efforts vary and
further activities are needed.

Dr. Murashov:

Standardization efforts are focused on both broadly applicable
characterization methods and category-specific methods. For example, the
ISO TC 229 (Nanotechnologies), Working Group 3 (Health, Safety and the
Environment) is working on five projects:

Technical Report “Health and Safety Practices in Occupational Settings
Relevant to Nanotechnologies” – broadly applicable characterization
methods;

International Standard “Endotoxin test on nanomaterial samples for in
vitro systems” – broadly applicable;

International Standard “Generation of silver nanoparticles for
inhalation toxicity testing” – specific to silver nanoparticles;

International Standard “Monitoring silver nanoparticles in inhalation
exposure chambers for inhalation toxicity testing” - specific to
silver nanoparticles;

Technical Report “Guidance on physico-chemical characterization of
engineered nanoscale materials for toxicologic assessment” - broadly
applicable.

The ISO TC 229 Working Group 2 (Metrology and Instrumentation) is
working on a number of standards for characterizing carbon nanotubes.

Dr. Murday:

No specific comments provided.

Dr. Small:

Physical standards: Standardization efforts are in the process of
ramping up both from an activity as well as a funding view point.
Government, industry, and academia, in the US as well as other areas
like Europe are in the process of defining and measuring candidate
materials. Efforts are focused on both category-specific i.e. CNT and on
broader definitions such as metal oxides or on schemes used for
classifications of materials like toxicity, commercial impact, and
environmental impact.

Documentary Standards (methods): As with the physical standards these
efforts are also increasing although development of “standard methods,
in most cases, requires the availability of well characterized physical
standards.

Dr. Warheit:

No specific comments provided.

Question 18: What alternative or innovative methods or technologies can
be applied to nanomaterial analysis?

Dr. BelBruno:

No specific comments provided.

Dr. Canady:

No specific comments provided.

Dr. Cassell:

Atomic force microscopy along with some of the latest scanning probe
microscopy techniques could be useful for nanomaterials analysis.  I’m
unsure if standards for characterizing with these techniques have been
set.

Dr. Lahann:

Transmission electron microscopy, scanning electron microscopy and
scanning probe microscopy, fluorescence microscopy, X-ray microscopy.

Dr. Murashov:

Some examples of alternative or innovative methods or technologies that
could be applied to nanomaterial analysis as part of safety and health
assessments are:

A number of instruments providing on-line size-resolved chemical
speciation of aerosols, such as aerosol mass-spectrometry, have been
developed.  These techniques can provide compositional information on
single particles or ensembles of size-selected particles in real time.

Size distribution and number concentration of nanoparticles in liquid
phase can be characterized using light scattering, laser diffraction,
size exclusion chromatography, acoustic techniques and field flow
fractionation (Powers et al, 2006), while space-resolved spectroscopic
techniques can be useful in obtaining information about chemical
composition and structure of nanoparticles.

Dr. Murday:

ASTM interlab standards not yet in place.

Tip characterization for proximal probes is a major challenge.

Absence of standard materials another factor.

Lab on a chip and variants on microarrays might offer new approaches
that could be readily adapted to field use.

To the extent that NS impact on living systems is paramount, cell based
microarrays (being developed for drug candidate screening) might provide
a “canary on a chip.”

Dr. Small:

There are many novel technologies, e.g. microscopy-based, as well as
others, which are applicable to nanoscale regimes.  Is the EPA looking
for methods that are likely to be widely used (cheap and fast) or for
very specific research areas?  Many of these methods are currently
difficult and expensive to make and are confined to laboratory studies
some may eventually become routine others will not.  Examples of some
techniques include: super-resolution optical, 3D chemical tomography,
TGA on chip, size-exclusion chromatography.

Dr. Warheit:

No specific comments provided.

Modeling

Empirical modeling can be a useful approach to predict physical-chemical
properties when experimental data are not known or ascertainable.  The
initial problem with modeling is that, to accurately predict property
endpoints for a given category of substances, there must be some
experimental data available in the tool’s database for at least some
representative substances in that category.  For newly discovered or
studied materials, the minimum but necessary quantity and type of
experimental data often is not available to sufficiently populate a
tool’s database and allow accurate prediction by the tool.  Some
estimation methods have been developed for specific property endpoints,
but many others are lacking.

Question 19: Are there models that are currently used to obtain property
data for nanoscale materials?  For which properties and which nanoscale
materials?

Dr. BelBruno:

I believe that we need to have a long-term view for this.  Simulations
provide agreement with already established data, but are not predictive.
 Quantum mechanical calculations provide a basis for the emission in
quantum dots.  However, structure function type relationships of the
type used in the pharmaceutical industry do not exist.

Dr. Canady:

I don’t know.

Dr. Cassell:

There are some atomistic approaches that have been used to predict
properties of nanoscale materials and some of these models have been
extended to molecular dynamics and mesoscale modeling approaches.  I am
not an expert in the field, but I am aware of these techniques.  The
properties modeled are typically physical in nature such as mechanical,
thermal or electrical in nature.  Primarily these approaches are
utilized to understand transport phenomena.

Dr. Lahann:

A series of different approaches is currently under development
depending on specific materials categories.

Dr. Murashov:

Only few risk-related models exist that are currently used to obtain
property data for nanoscale materials.  Examples are:

Several models currently used to obtain property data for nanoscale
materials are models based on aerodynamic properties of nanoparticles
such as penetration through filter media, coagulation and deposition in
the lung (Lee and Liu, 1982; Maynard and Kuempel, 2005; ICRP, 1994).

Research data indicate that health hazard properties of poorly soluble,
low-toxicity nanoparticles can be modeled using their total surface area
(Maynard and Kuempel, 2005; Duffin et al., 2007).

Dr. Murday:

It is possible to extrapolate from smaller (atomistic) and larger
(continuum) scale models, but caution must be exercised.

First principle models of CNT were able to predict accurately many of
their properties well before the experimental community could measure
them (see Figure C).  However, CNT is a special case of NS – CNT is
all C, has only on type of bond, and has a high degree of symmetry.

Dr. Small:

No specific comments provided.

Dr. Warheit:

No specific comments provided.

Question 20:  Has any validation work been conducted that compares
predicted values with measured data?  For which properties and which
nanoscale materials?

Dr. BelBruno:

See response to question 19.

Dr. Canady:

NCL has published standard methods through ASTM for some assays that
have validation data.

Dr. Cassell:

Individual carbon nanotube based thermal conductivity has been predicted
by theory and subsequently validated using experimental approach.  Prof.
Arun Majumdar’s group at UC Berkeley performed these elegant
experiments.

Dr. Lahann:

Some of the models have been validated.  A good example is the work of
Nick Kotov and Sharon Glotzer at University of Michigan that
experimentally validated the formation of nanoparticle sheets (Science
2006).  More work is needed.

Dr. Murashov:

Validation work has been conducted for data related to penetration of
filter media by nanoscale particles (see for example, Rengasamy et al.,
2007).

Dr. Murday:

CNT model predictions have been verified by experiment (see Figure C).

QD model predictions have been verified by experiment.

Fluid flow in nanopores has been examined by Navier Stokes (continuum)
and molecular dynamics approaches to understand the modeling
limitations.  Results have been partially validated by experiment.

Dr. Small:

No specific comments provided.

Dr. Warheit:

No specific comments provided.

Question 21: Are there current significant characterization needs for
which the NMSP should investigate model development?

Dr. BelBruno:

See response to question 19.

Dr. Canady:

Yes – but an entire workshop or similar effort should be devoted to
developing ideas and projects for modeling.  The most immediate one may
be developing a modeling approach that allows updating of the
information needs relative to the decision needs.  As more is learned
about the materials and the properties that are most relevant to
decisions there will need to be a process for adjusting the information
requirements.

Dr. Cassell:

I’m not sure if there are aggregation/agglomeration models that could
be used for various types of nanomaterials, but I see this as being a
valuable method to augment characterization approaches.  Any predictive
tool that is useful in a contextual setting as opposed to idealized
models would be extremely useful.

Dr. Lahann:

Specifically in the area of correlation between nanoparticle properties,
such as size, shape, composition, and biological properties, such as
cell uptake or cytotoxicity, there would be a great potential for model
developments.

Dr. Murashov:

Examples of current significant characterization needs for which the
NMSP should investigate model development are:

Risk assessment models for nanoscale materials are needed. 
Specifically, models to extrapolate toxicity data from in vitro to in
vivo; from in vivo to human; external exposure levels to internal dose
should be further developed.

There is a need to develop models correlating biomarkers of exposure
with specific nanomaterials.

There is also a need to determine if certain physico-chemical properties
of nanoparticles are predictive of biological activity.

Dr. Murday:

Chemical reactivity – partner with catalysis efforts.

Toxicology – partner with NIH, NIOSH, NIST, NCL.

Fate and effect in soils – a topic more closely aligned with EPA than
other agencies.

Dr. Small:

No specific comments provided.

Dr. Warheit:

No specific comments provided.

Supporting Figures

Figures provided by Dr. Murday:

Figure   SEQ Figure \* ALPHABETIC  A 

FIELD	PROPERTY	SCALE LENGTH

ELECTRONIC	ELECTRON WAVELENGTH

INELASTIC MEAN FREE PATH

TUNNELING	10 

1

1	--

--

--	100nm

100nm

10nm

MAGNETIC	DOMAIN WALL

EXCHANGE ENERGY

SPIN-FLIP SCATTERING LENGTH	10 

0.1

1	--

--

--	100 nm

1nm

100nm

OPTIC	QUANTUM WELL

EVANESCENT WAVE DECAY LENGTH

METALLIC SKIN DEPTHS	1

10

10	--

--

--	100nm

100nm

100nm

SUPERCONDUCTIVITY	COOPER PAIR COHERENCE LENGTH

MEISSNER PENETRATION DEPTH	0.1

1	--

--	100nm

100nm

MECHANICS	DISLOCATION INTERACTION

GRAIN BOUNDRIES	1

1	--

--	1000nm

10nm

NUCLEATION/

GROWTH	DEFECT

SURFACE CORRUGATION	0.1

1	--

--	10nm

10nm

CATALYSIS	LOCALIZED BONDING ORBITALS

SURFACE TOPOLOGY	0.01

1	--

--	0.1nm

10nm

SUPRAMOLECULES	PRIMARY STRUCTURE

SECONDARY STRUCTURE

TERTIARY STRUCTURE	0.1

1

10	--

--

--	1nm

10nm

1000nm

IMMUNOLOGY	MOLECULAR RECOGNITION	1	--	10nm

Figure   SEQ Figure \* ALPHABETIC  B : Characteristic Lengths in Solid
State Science Models

Figure   SEQ Figure \* ALPHABETIC  C 

Supporting References

References provided by Dr. Murashov:

Duffin, R., Tran, L., Brown, D., Stone, V., and Donaldson, K. (2007).
Proinflammogenic effects of low-toxicity and metal nanoparticles in vivo
and in vitro: highlighting the Role of Particle Surface Area and Surface
Reactivity, Inhal. Tox. 19(10), 849-856.

ICRP, Human respiratory tract model for radiological protection, 1994,
ICRP publication 66.

ISO, Workplace Atmospheres – Ultrafine, nanoparticle and
nano-structured aerosols – Inhalation exposure characterization and
assessment, ISO/TR 27628:2007, 2007.

Jevprasesphant, R., Penny, J., Jalal, R., Attwood, D., MckEown, N. B.,
and D’Emanuele, A. (2003). The influence of surface modification on
the cytotoxicity of PAMAM dendrimers, Int. J. Pharm. 252, 263-266.

Lee, K. W. and Liu, B. Y. H. (1982). Theoretical study of aerosol
filtration by fibrous filters, Aerosol Sci. Technol. 1(2), 147-162.

Maynard, A. D. and Kuempel, E. D. (2005). Airborne nanostructured
particles and occupational health. J. Nanopart. Res. 7, 587-614.

Mecke, A., Uppuluri, S., Sassanella, T. M., Lee, D. K., Ramamoorthy, A.,
Baker, J. R., Jr., Orr, B. G., and Banaszak Holl, M. M. (2004). Direct
observation of lipid bilayer disruption by poly(amidoamine) dendrimers.
Chem. Phys. Lipids 132, 3-14.

NIOSH (1994). NIOSH Manual of Analytical Methods (NMAM®), 4th ed. DHHS
(NIOSH) Publication 94-113 (August, 1994), Schlecht, P.C. & O'Connor,
P.F., Eds., 1994. Available on line at   HYPERLINK
"http://www.cdc.gov/niosh/nmam/chaps.html" 
www.cdc.gov/niosh/nmam/chaps.html .

NIOSH (2005). Evaluation of Health Hazard and Recommendations for
Occupational Exposure to Titanium Dioxide. Available on line at  
HYPERLINK "www.cdc.gov/niosh/review/public/TIo2/" 
www.cdc.gov/niosh/review/public/TIo2/ .

NIOSH (2006). Approaches to Safe Nanotechnology: An Information Exchange
with NIOSH. Available online at:   HYPERLINK
"http://www.cdc.gov/niosh/topics/nanotech/safenano/" 
http://www.cdc.gov/niosh/topics/nanotech/safenano/ . 

Powers, K. W., Brown, S. C., Krishna, V. B., Wasdo, S. C., Moudgil, B.
M., and Roberts, S. M. (2006). Research Strategies for Safety Evaluation
of Nanomaterials. Part VI. Characterization of Nanoscale Particles for
Toxicological Evaluation. Tox. Sci. 90(2), 296-303.

Rengasamy, S., Verbofsky, R., King, W. P., and Shaffer, R. E. (2007).
Nanoparticle Penetration through NIOSH-approved N95 Filtering-facepiece
Respirators, J. Int. Soc. Resp. Protec. 24, 49-59.

Sayes, C. M., Fortner, J. D., Guo, W., Lyon, D., Boyd, A. M., Ausman, K.
D., Tao, Y. J., Sitharaman, B., Wilson, L. J., Hughes, J. B., West, J.
L., and Colvin, V. L. (2004). The differential cytotoxicity of
water-soluble fullerenes, Nano Lett. 4 (10), 1881-1887.

Shvedova, A.A., Sager, T., Murray, A.R., Kisin, E., Porter, D.W.,
Leonard, S., Schwegler-Berry, D., Robinson, V.A., and Castranova, V.
(2007). Critical issues in the evaluation of possible adverse pulmonary
effects resulting from airborne nanoparticles.  In: Nanotoxicology –
Characterization, Dosing and Health Effects. N.A. Monteiro-Riviere and
C.L. Tran (eds), Informa Healthcare, NY, Chap14, pp 225-236.

Singh, R., Pantarotto, D., Lacerda, L., Pastorin, G., Klumpp, C., Prato,
M., Bianco, A., Kostarelos, K. (2006). Tissue biodistribution and blood
clearance rates of intravenously administered carbon nanotube
radiotracers, Proc. Nat. Acad. Sci. 103 (9), 3357-3362.

Wexler, A. S., and Johnston, M. V. (2001). Real-time single-particle
analysis. In Aerosol Measurement: Principles, Techniques and
Applications, Baron, P. A. and Willeke, K. eds., John Wiley & Sons, New
York, pp. 365-385.

References provided by Dr. Warheit:

References for Nanoparticle Material Characterization Strategies for
Toxicity Studies

International Life Sciences Institute – Panel – “Principles for
characterizing the potential human health effects from exposure to
nanomaterials: elements of a screening strategy” - Oberdorster G,
Maynard A, Donaldson K, Castranova V, Fitzpatrick J, Ausman K, Carter J,
Karn B, Kreyling W, Lai D, Olin S, Monteiro-Rivere N, Warheit DB, and
Yang H.  Particle and Fibre Toxicology 2005, 2:8.

“Physicochemical properties that may be important in understanding the
toxic effects of test materials include particle size and size
distribution, agglomeration state, shape, crystal structure, chemical
composition, surface area, surface chemistry, surface charge, and
porosity”. 

Testing strategies to establish the safety of nanomaterials: 
Conclusions of an ECETOC workshop.  Inhal. Toxicol., 19: 631-643, 2007.
Warheit DB, Borm PJA, Hennes C, and Lademann J.  Inhal. Toxicol., 19:
631-643, 2007.

“In summary, it was concluded that nanoparticle composition,
dissolution, surface area and characteristics, size, size distribution,
and shape are parameters needed for any target organ toxicity
assessment.  Depending on the type of toxicological study undertaken,
other physicochemical parameters would also be required

 http://www.particleand fibretoxicology.com/content/2/1/8 

 http://www.ansi.org/standards_activities/standards_boards_panels/nsp/ov
erview.aspx?menuid=3

    HYPERLINK
"http://www.iso.org/iso/en/stdsdevelopment/tc/tclist/TechnicalCommitteeD
etailPage.TechnicalCommitteeDetail" 
http://www.iso.org/iso/en/stdsdevelopment/tc/tclist/TechnicalCommitteeDe
tailPage.TechnicalCommitteeDetail ? COMMID=5932

    HYPERLINK
"http://publicaa.ansi.org/sites/apdl/Documents/Forms/AllItems.aspx" 
http://publicaa.ansi.org/sites/apdl/Documents/Forms/AllItems.aspx   

 Class 1 substances are distinct chemicals with known, non-variable
molecular structures  

  http://www.nano.gov/NNI_EHS_research_needs.pdf

    HYPERLINK
"http://publicaa.ansi.org/sites/apdl/Documents/Forms/AllItems.aspx" 
http://publicaa.ansi.org/sites/apdl/Documents/Forms/AllItems.aspx   

 Class 1 substances are distinct chemicals with known, non-variable
molecular structures  

  http://www.nano.gov/NNI_EHS_research_needs.pdf

 PAGE   

 PAGE   i 

TABLE OF CONTENTS (Continued)

Page

 PAGE   1-1 

 PAGE   2-1 

A- PAGE   2 

									

 PAGE   

 PAGE   8 

									

B-  PAGE  7 

Compilation of Panelists’ Preliminary Comments	

C-  PAGE  39 

