45
9 th California Unified Program Training Conference Garden Grove, California February 13, 2007 – Green Technology Course Green Chemistry in California: A Framework for Leadership in Chemicals Policy and Innovation Michael P. Wilson, Ph.D, MPH Center for Occupational and Environmental Health University of California, Berkeley [email protected]

9 th California Unified Program Training Conference Garden Grove, California February 13, 2007 – Green Technology Course Green Chemistry in California:

  • View
    215

  • Download
    0

Embed Size (px)

Citation preview

9th California Unified Program Training ConferenceGarden Grove, California

February 13, 2007 – Green Technology Course

Green Chemistry in California: A Framework for Leadership in Chemicals Policy and Innovation

Michael P. Wilson, Ph.D, MPH

Center for Occupational and Environmental Health

University of California, Berkeley

[email protected]

2

• Toxicology

• Epidemiology

• Industrial hygiene

• Environmental health policy

• Occupational and environmental medicine

• Occupational health nursing

• Ergonomics

• Labor occupational health education

• Continuing professional education

Center for Occupational and Environmental Health

Est. 1978 (AB 3414)

Berkeley, Davis, San Francisco (northern California).

The University of California, Berkeley

3

The challenge of chemicals policy: Hexane/acetone-induced neurological disease in the California vehicle repair industry.

Harrison et al. MMWR, Nov 16, 2001, Vol 50 #5The University of California, Berkeley

4

Ladefoged et al. 1994. Pharm and Tox. 74:294-299.

The University of California, Berkeley

5

8h TWA breathing zone exposure concentration, mg/m3 (n=23)

0 10 20 30 40 50 60 70 80

Hexane

Acetone

TolueneToluene

Hexane

Acetone

The University of California, Berkeley

6

Introduction of hexane, 1988:

… unintended consequence of listing chlorine-contaminated oil as hazardous waste.

The University of California, Berkeley

Introduction of hexane-acetone blends, 1997:

…unintended consequence of exempting acetone from California VOC rules.

7

1970: Stoddard solvent Fire hazard1978: CFCs Ozone depletion1990: Methylene chloride Carcinogen1990: 1,1,1-Trichloroethane Ozone depletion1991: Perchloroethylene and hexane Dioxin emissions1997: Hexane/acetone blends Neurotoxin2003: 1-Bromopropane Repro toxin

Chemical management in the CA vehicle repair industry

The University of California, Berkeley

8

Stage 1: Disposal and dilution

Stage 2: Waste treatment and pollution control

Stage 3: Toxics policy (chemical-by-chemical approaches; e.g. hexane case)

Stage 4: Chemicals policy (transparency, accountability, chemical design, markets, life

cycles)

Chemical management over the last 30 years

The University of California, Berkeley

Commissioned January 2004 by:• Byron Sher (Chair, SEQC) • John Laird (Chair, ACESTM)

UC Chemicals Policy Report• Assesses problems and opportunities in chemicals policy• Proposes broad policy goals

Released to Legislature March 14, 2006 to:• Joseph Simitian (Chair, SEQC)• Ira Ruskin (Chair, ACESTM)

Download etc: http://coeh.berkeley.edu/news/06_wilson_policy.htm

The University of California, Berkeley

10

John R. Balmes, MDSchool of Medicine, UC San Francisco

 Carl F. Cranor, PhD Department of Philosophy, UC Riverside S. Katharine Hammond, PhDSchool of Public Health, UC Berkeley

Bill E. Kastenberg, PhDCollege of Engineering, UC Berkeley Ann Keller, PhDSchool of Public Health, UC Berkeley

Amy D. Kyle, PhD, MPHSchool of Public Health, UC Berkeley

Geoff Lomax, DrPHDepartment of Health Services 

 

Timothy Malloy, JD School of Law, UC Los Angeles Thomas E. McKone, PhD Lawrence Berkeley National Laboratory  Dara O’Rourke, PhD College of Natural Resources, UC Berkeley Julia Quint, PhDDepartment of Health Services Christine Rosen, PhD Haas School of Business, UC Berkeley David J. Vogel, PhD Haas School of Business, UC Berkeley

Advisory Committee Members

11

Methods

• Literature review

• Key informant interviews

• Survey data from 37 electronics companies

• 35 conferences

• Presentations at 17 conferences

• Advisory Committee review

The University of California, Berkeley

12

Plastic resins

Organics

Industrial chemicals

Fertilizers

Consumer products

Inorganics

Basic chemicals

Specialty chemicals

Other Inorganics

Alkalies & chlorine

Inorganic pigments

Acids

Coatings

Synthetic materials

Others

Industrial gases

Adhesives & sealants

Pharmaceuticals

Crop protection

Consumer use

Professional & industrial use

Man-made fibers

Synthetic rubber

Petrochemicals & derivatives

Life sciencesSource: American Chemistry Council, Guide to the

Business of Chemistry, 2003

Scope

13

Report’s findings are similar to those of:

• National Academy of Sciences 1984

• U.S. General Accounting Office 1994

• Congressional Office of Technology Assessment 1995

• Environmental Defense 1997

• U.S. EPA 1998

• former EPA officials 2002

• RAND Science and Technology Institute 2003

• U.S. Government Accountability Office 2005

• National Academy of Sciences 2005

The University of California, Berkeley

14

                     

Environmental Policy Developments in the European Union

- Cosmetics Directive: 2004

- Waste in Electrical and Electronic Equipment (WEEE): 2005

- Restriction on Hazardous Substances (RoHS): 2006

- Registration, Evaluation, Authorization of Chemicals (REACH): 2007

The University of California, Berkeley

15Electronic waste recovery workers, Guiyu, China

Courtesy Basel Action NetworkThe University of California, Berkeley

16

Registration: tiered tox and use data

Authorization: for CMR, PBT, vPvB

Evaluation: >100 tons/year

About 30,000 chemicals

Data requirements depend on volume

3, 6, 11-yr phase-in

About 1,400 – 2,000 “chemicals of very high concern” (no 1 ton trigger for these)

De-authorization: inadequate control

Authorization

De-authorization: benefit too small

Chemicals produced or imported at >1

ton/year/producer

E.U. REACH

About 5,000 chemicals

17

AB 121 (Vargas) AB 263 (Chan) AB 289 (Chan) AB 319 (Chan) AB 342 (Baca) AB 597 (Montanez)AB 623 (Aanistad) AB 639 (Aghazarian) AB 752 (Karnette) AB 815 (Lieber) AB 816 (Lieber) AB 848 (Berg)

AB 908 (Chu) AB 912 (Ridley-Thomas) AB 966 (Saldana) AB 985 (Dunn) AB 990 (Lieber) AB 1125 (Pavley) AB 1337 (Ruskin) AB 1342 (Assem ESTM) AB 1344 (Assem ESTM) AB 1354 (Baca) AB 1415 (Pavley) AB 1681 (Pavley)

35 chemically-related bills in California, 2005-2006

SB 419 (Simitian) SB 432 (Simitian) SB 484 (Migden) SB 490 (Lowenthal) SB 600 (Ortiz) SB 838 (Escutia) SB 849 (Escutia) SB 982 (Sen EQ comm) SB 989 (Sen EQ comm.) SB 1067 (Kehoe) SB 1070 (Kehoe)

The University of California, Berkeley

18

20

25

30

35

40

45

50

55

60

1990 2000 2010 2020 2030 2040 2050

Mill

ions

California’s expected population growth, 1990-2050

2006 = 36 million

Source: California Dept of Finance, CA pop. trends, 1990 – 2050 The University of California, Berkeley

19

1) What is the problem?

2) What is the cause of the problem?

3) How can we respond to these problems?

The University of California, Berkeley

Green Chemistry in California: A Framework for Leadership in Chemicals Policy and Innovation.

20

Public environmental

health

Workers

Children

Consumers

Industry & business

Downstream users

The Legislature

Municipal agencies

The University of California, Berkeley

Chemical producers

GovernmentState

agencies

Green chemistry

21

1) What is the problem?

– Data Gap: The market “undervalues” the hazardous properties of chemicals relative to their function, price, and performance.

– Safety Gap: Government is unable to assess chemical hazards and control those of greatest concern.

– Technology Gap: The U.S. private sector is investing minimally in cleaner chemical technologies, such as green chemistry.

The University of California, Berkeley

Green Chemistry in California: A Framework for Leadership in Chemicals Policy and Innovation.

22

2) What is the cause of the problem?

– We have federal statutes…

– that do not require transparency and accountability.

– This has produced a chemicals market…

– that allows potentially hazardous chemicals to remain competitive.

The University of California, Berkeley

Green Chemistry in California: A Framework for Leadership in Chemicals Policy and Innovation.

23

• Data Gap: – TSCA (and FFDCA) do not require producers to generate and

disclose chemical hazard data for government or downstream users.

• Safety Gap:– TSCA (and FFDCA) have constrained government’s ability to

efficiently assess and control chemical hazards.

• Technology Gap:– The resulting chemicals market has dampened private sector interest

in green chemistry, as reflected in R&D funding and in education.

The federal Toxic Substances Control Act of 1979.

The University of California, Berkeley

24

Function

HazardsPerformance

Price

The University of California, Berkeley

Example 1. Data Gap

Downstream users

To make informed decisions about chemicals, buyers need four pieces of information:

25

Function

SafetyPerformance

Price

The University of California, Berkeley

In the present market, hazard information (e.g. toxicity) is largely missing.

26

The University of California, Berkeley

Identity

HazardsUses

Sales volume

Example 2. Data Gap

Agencies

To assess & prioritize chemical hazards, state agencies need at least four pieces of information:

27

The University of California, Berkeley

This information is unavailable to agencies in California and other states.

28

Scientific suspicion of harm

Reasonable grounds for concern

Balance of evidence

Clear evidence of risk; risks outweigh benefits

The University of California, Berkeley

Example 3. Safety Gap

• Under TSCA (& FFDCA), government carries the burden of proving risk…

• yet producers are under no obligation to provide the information necessary for government to do so…

• which produces a logical paralysis.

29

The University of California, Berkeley

Why? Because university teaching and research in chemistry reflect conditions in the chemicals market.

Example 4. Technology Gap.

With very few exceptions, one can earn a Ph.D in chemistry in the U.S. without demonstrating a basic understanding of toxicology…

…or the principles of green

chemistry. Sather Gate, UC Berkeley

30

                                             

• Total daily U.S. chemical production and importation: = 42 billion pounds (623,000 tankers) (TSCA IUR).

• Total daily California sales of chemical products alone: = 644 million pounds (inc. water) (2,700 tankers) (CA ARB).

The University of California, Berkeley

31

The workplace Consumer products Waste streams

Global trade

Ecosystems: air, water, food, soil

The University of California, Berkeley

32

Public health issues related to chemicals

• Hazardous waste: 600 new hazardous waste sites needed each month in U.S. up to 2033 (77,000 today) ($250 billion). CA DTSC: 975 employees working on hazwaste, 25 on poll prevention; emerging “dumping ground” issue.

• Workers: Yearly CA morbidity (23,000) and mortality (6,500). PELS for only 193 of 2,800 HPV chemicals (~7%).

• Bioaccumulative chemicals: 688 found by EPA; CDC; CA SB 1379 (Perata).

• New pediatric morbidity estimates: Asthma (10-35%), certain cancers (2-10%), neurodevelopmental disorders (5-20%) related to chemical exposures.

• Endocrine disruption: Early human evidence, substantive animal evidence.

• Umbilical cord blood: Early studies showing presence of many chemicals.

The University of California, Berkeley

33

Global chemical production is expected to double every 25-years.

0

100

200

300

40019

92

1994

1996

1998

2000

2002

2004

2006

2008

2010

2012

2014

2016

2018

2020

2022

2024

2026

2028

2030

Production index = 131 today, where 1997 = 100

Production index = 231 in 2022, where 1997 = 100

The University of California, Berkeley

34

3) How can we respond to these problems?

– Chemicals policy to re-orient the market to green chemistry.

The University of California, Berkeley

Green Chemistry in California: A Framework for Leadership in Chemicals Policy and Innovation.

35

                     

Will E.U. policies result in E.U. competitive advantage in cleaner technologies, such as green chemistry? Other issues:

- U.S. competitiveness in these areas?

- Potential for U.S. to become “dumping ground?”

- Will E.U. policies create “uneven playing field” among U.S. companies?

- Will states begin to leverage E.U. databases to implement new policy?

- Increased activity by U.S. NGOs for chemicals policy reform?

- How will the chemicals market respond to new hazard information?

- Will pressure increase to reform TSCA, FFDCA?

The University of California, Berkeley

Cosmetics,WEEE, RoHS, REACH, others China RoHS, 2007

Korea RoHS, 2007

Japan RoHS, 2006

CA RoHS, 2007

36

“G.E. Chief Points to ‘Green’ Handicap”

“…the deregulatory agenda favored by the U.S. business community – particularly on environmental issues – is not providing American companies with a competitive advantage over their European

counterparts.”

Financial TimesMay 10, 2005

Stephanie Kirchgaessner in Washington

Jeffrey Immelt, Chairman and CEO (Ecomagination)The University of California, Berkeley

37

0

2,000

4,000

6,000

8,000

10,000

12,000

14,000

1960

1962

1964

1966

1968

1970

1972

1974

1976

1978

1980

1982

1984

1986

1988

1990

1992

1994

1996

1998

2000

2002

2004

United States

California

A systems approach is needed to produce enduring changes in the chemical sector. Example: electricity use.

Total electricity use per capita, kWh, 1960 - 2002

The University of California, BerkeleyCourtesy John Wilson, CA Energy Commission

38

The design of chemical products and processes to reduce and/or eliminate substances hazardous to human health and the environment.

12 principles proposed by Anastas &Warner.

Anastas, P.T. and J. Warner. 1999. Green Chemistry Theory and Practice

The University of California, Berkeley

39

The University of California, Berkeley

• From a policy perspective,

• green chemistry links public & environmental health…

• with new business & investment opportunities.

40

UC chemicals policy analysis

Barriers to green chemistry Drivers of green chemistry

The University of California, Berkeley

41

UC chemicals policy analysis

Barriers to green chemistry Drivers of green chemistry

Data Gap

TSCA, other statutes

Buyers: no haz data Sellers: no case for GC

Safety Gap Tech. Gap

Markets Government

Inability to assess haz Inability to control haz

- Hazard undervalued against price, function - Hazardous chemicals competitive in market - Green chemistry innovation impeded

42

UC chemicals policy analysis

Barriers to green chemistry Drivers of green chemistry

E.U. Directives

Downstream users

Industry leaders

Public opinion

Green chem.

entrepreneurs

California Legislature

Government procurement

NGO activity

Climate change

National Academy

The University of California, Berkeley

43

UC chemicals policy analysis

Policy objectives: Close the Data, Safety & Technology Gaps: * Improve the flow of information in the chemicals market. * Improve government capacity to act. * Implement other incentives for green chemistry.

Issues, models, mechanisms: * Leverage market forces. * Address chemical life cycle. * Place least demands on government. * Motivate technology innovation and diffusion, etc.

Recommendations

Barriers to green chemistry Drivers of green chemistry

The University of California, Berkeley

44

UC chemicals policy analysis

Policy objective

Ideal mechanisms

Recommendations

In choosing not to act: * Existing problems will expand in California. * U.S. & California could become “dumping ground.” * CA will cede leadership in green chemistry.

Drivers of green chemistryBarriers to green chemistry

The University of California, Berkeley

Thank you!

The University of California, Berkeley