Environmental impact of steelworks' emissions

(1)

Delft University of Technology

Environmental impact of steelworks' emissions

An integrated approach

Murias Gomes Lage, J.

DOI

10.4233/uuid:9839c2a7-3c7e-4705-9c1f-1868da8c7c1a

Publication date

2016

Document Version

Final published version

Citation (APA)

Murias Gomes Lage, J. (2016). Environmental impact of steelworks' emissions: An integrated approach.

https://doi.org/10.4233/uuid:9839c2a7-3c7e-4705-9c1f-1868da8c7c1a

Important note

To cite this publication, please use the final published version (if applicable).

Please check the document version above.

Copyright

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons. Takedown policy

Please contact us and provide details if you believe this document breaches copyrights. We will remove access to the work immediately and investigate your claim.

This work is downloaded from Delft University of Technology.

(2)

(3)

Stellingen behorende bij het proefschrift

"Environmental impact of steelworks' emissions: an integrated approach" door Joana Lage

1 .Staalproduktie is essentieel voor de mens om tot een duurzame toekomst te komen.

This Tiiesis (Cimpter 3)

2. Humane gezondheid is sterk afhankelijk van de bodemgesteldheid.

This Thesis (Ciiapler 5)

3. Het meest belangrijke doel van studies naar bronherkenning is de identificatie van de oorzaken van overschrijding van wettelijke grenswaarden.

Tiiis Thesis (Cimpter 6)

4. Biomonitoring is een goede manier om het milieu-effect van lucht-verontreinigende stoffen vast te stellen.

5. Verontreiniging is niets anders dan de rijkdommen die niet geoogst worden. We staan toe dat deze verspreid worden omdat we onwetend zijn van hun waarde.

R. Bucicminster Fidier

6.Intelligente ideeën worden slechts dan geaccepteerd als er iets doms in aanwezig is.

Fernando Pessoa

7. Als je een schip wilt bouwen moet je geen mensen optrommelen om hout te verzamelen en je moet ze geen taken en werk opleggen, maar je moet ze een hevig verlangen naar de eindeloze uitgestrektheid van de zee bijbrengen.

Autoine de Saint-Exupéry

8. De geest is als een parachute. Het werkt het best als het geopend is.

Daiai Lama

9.SteiTen zijn gidsen voor reizigers. Voor anderen zijn het slechts kleine lichtjes.

Antoine de Saint-Exupér

10. Definiëren is hmiteren.

Oscar Wilde

Deze stellingen worden opponeerbaar en verdedigbaar geacht en zijn als zodanig goedgekeurd door de promotor Prof.dr, H.Th. Wolterbeek.

(4)

Propositions accompanying tiie thiesis

"Environmental impact of steelworks' emissions: an integrated approach" by Joana Lage

1. Steel production is essential in enabling man to move towards a sustainable future. This Tliesis (Ciiapter 3)

2. Human health is strongly dependent on the status of the soils.

Tins Tliesis (Cimpter 5) 3. The most important purpose of source apportionment studies is to identify the

causes for exceedance of legislated thresholds.

Tills Tiiesis (Ciiapter 6) 4. Biomonitoring is a good approach to assess environmental impact of air

pollutants.

5. Pollution is nothing but the resources we are not harvesting. We allow them to disperse because we've been ignorant of their value.

R. Buckininster Flitter 6. No intelligent idea can gain general acceptance unless some stupidity is mixed in

with it.

Fernando Pessoa 7. I f you want to build a ship, don't drum up people to collect wood and don't assign

them tasks and work, but rather teach them to long for the endless immensity of the sea.

Antoine de Saint-Exupéry 8. The mind is like a parachute. It works best when it is open.

Dalai Lama 9. For travelers, stars are guides. For others they are nothing but tiny lights.

Antoine de Saint-Exupéry 10. To define is to limit.

Oscar Wilde

These propositions are regarded as opposable and defendable, and have been approved as such by the supervisor Prof.dr. H.Th. Wolterbeek

(5)

ENVIRONMENTAL IMPACT OF

S T E E L W O R K S ' EMISSIONS: AN

(6)

(7)

Environmental impact of steelworks' emissions: an integrated approach

Proefschrift

ter verkiijging van de graad van doctor aan de Technische Universiteit Delft,

op gezag van de Rector Magnificus Prof.ir. K.C.A.M Luyben, voorzitter van het College voor Promoties,

in het openbaar te verdedigen op 2 februari 2016 om 10:00 uur

door

Joana Mürias Gomes L A G E

MSc Environmental Engineering

Universidade Lusófona de Humanidades e Tecnologias, Lisbon, Portugal BSc Environmental Health,

Escola Superior de Tecnologias da Saüde de Lisboa, Lisbon, Portugal geboren te Lisbon, Portugal

(8)

This dissertation has been approved by the

Promotor: Prof. dr. H.Th. Wolterbeek, Delft University of Technology Promotor: Dr. S.M. Almeida, 1ST, University of Lisbon

Composition of the doctoral committee: Rector Magnificus, Chairman

Prof. dr. H.Th. Wolterbeek, Deflt University of Technology Dr. S.M. Almeida, 1ST, University of Lisbon

Independent members:

Dr. F. Drewnick, Max Planck Institute, Germany Dr. D.R. Anderson, Tata Steel, UK

Prof.dr. A. van de Wiel, Delft University of Technology Prof dr. R. Samson, University of Antwerpen

Prof dr. E.H. Briick, Delft University of Technology

The research described in this thesis was performed in Faculty of Applied Sciences, Depai tnieiit of Radiation Science and Technology, Delft University of Technology, and in (fTN, Instituto Superior Técnico, Univeristy of Lisbon, Portugal, with the contribution of European Community's Research Fund for Coal and Steel (RFCS) under grant agreement RFSR-CT-2009-00029. The author acknowledges Fundagao para a Ciencia e a Tecnologia (FCT) for funding her grant (SFRH/BD/79084/2011).

ISBN:

Keywords: Industrial air pollution, Steelworks, PMIO monitoring, Biomonitoring, Soil contamination. Source apportionment, Neutron Activation Analysis

Cover design: Tiago Martins

Printed & Lay Out by: Proefschriftmaken.nl || Uitgeverij BOXPress

Published by: Uitgeverij BOXPress, 's-Hertogenbosch

LEGAL NOTICE

The publisher is not responsible for the use which might be made of the following information.

(9)

To my dear mum, dad and sister and to the memory of Greg

(10)

(11)

1. Introduction

1.1. Motivation and objectives

Over the last decades interest was growing to improve air quality not only by the general

public, but also by governmental entities (Calvo et al, 2013). Countries are developing and

applying abatement strategies to reduce population exposure to air pollutants. However,

cost-effective strategies require substantial knowledge on the contribution of emission

sources to pollutant concentrations. This prompted an important increase in atmospheric

pollution research (Taiwo et al., 2014).

The industry is one of the main emission sources contributing to the reduction of air

quality and steel industry is a major contributor to local airborne particle (PM)

concentrations in various regions of the world (Taiwo et al., 2014). On the other hand, steel

production is a key factor in Europe's economy and competitiveness (Hleis et al, 2013):

steel plays a significant role in modern society due to the diverse and increasing

applications of steel products (Mohiuddin et al, 2014).

The EU Air quality directive (OJL, 152/2008) emphasises the need to achieve improved air quality standards in order to improve the health and longevity of citizens living in the EU.

In this context it is essential that steelworks play their part in ensuring that the required

targets are achieved by the application of the best available technologies (BAT) for

pollution control (EIPPCB, 2013).

This reasoning prompted the set-up of the project "Assessment of Emissions and Impact of Steel Production Processes", supported by the European Research Fund for Coal and Steel

and developed by two major steel making coinpanies (Tata Steel UK and ArcelorMittal

-Spain) and three research institutes (Instituto de Soldadura e Qualidade (Portugal), the Max-Planck-Institute for Chemistry (Germany) and the Instituto Superior Técnico

(Portugal)). The main objectives of this project were 1) to provide novel methods and strategies to assess ambient air quality in the vicinity of steelworks; 2) to complete and

(16)

refine inventories of particulate emissions from steelworks; and 3) to assess the air quality

impact of these industries in their surroundings.

The study described in the present PhD thesis was conducted within the framework of this European project. The study focused on the ArcelorMittal plant located in Gijon, in

Asturias, Spain, which is an integrated steel plant with various processes, including iron sintering, cokemaking, iron and steelmaking, rolling mills and coating lines. The purpose of

the work was to provide improved understanding on steelworks' emissions and apportion

the impact of the major sources to the ambient air quality, thus enabling cost-effective abatement measures to be properly targeted and implemented. Following this line of

reasoning, the overall major objectives of this thesis were set as:

1) to assess ambient air quality in the vicinity of steelworks;

2) to increase knowledge on particulate emissions from processes, areas and

operations within integrated steelworks;

3) to quantify sources' contributions;

4) to produce data that can be used for future regional control programs.

1.2. Ambient Air Quality

1.2.1. General introduction

Air pollution has been recognized as a problem throughout history, although not with the

priority it has today. As early as in 1165 to 1204, Egyptian historical records show a

philosopher Jewish leader, Moses Maimonides, writing " [ . . . ] Comparing the air of cities to

the air of deserts and arid lands is like compai'ing waters that are befouled and turbid to

waters that are fine and pure [ . . . ] " (Gaffney and Marley, 2009). In France (1382), King

Charles V I implemented rules regarding the emission o f "nauseating fumes" in Paris. In

1661 John Evelyn presented Charles I the pamphlet Fumifugium, the first document dealing

with atmospheric pollution by particulate matter (Calvo et al., 2013).

Air pollution really came to public attention with the London sinog episode of December

1952 in which thousands of deaths have been attributed to the dense fog of SO2 and

particles. This event led to legislation and regulation, that over the past 50 years have

(17)

produced significant improvements in air quality v/ith concomitant decreases in health

effects and ecological damage (Hopke, 2009).

The industrial revoludon contributed to a huge development of the cities and of the

industrial centres worldwide and influenced exponentially the population densities. In the

megacities the demand for energy is extremely high and combustion processes in mobile

and stationary sources are the main factors contributing to local air pollution (Gaffney and

Marley, 2009).

During the last 50 years, air pollution in the industrialised world has undergone drastic changes (Fenger, 2009), being nowadays the most important pressing environmental

concern for the world's nations (Hsu et al., 2013). In highly-developed countries

manufacturing skills and procedures improved significandy, at high costs, so that manufacturing is sometimes shifted to less-developed countries with less stringent

emissions standards. According to Hopke (2009), a shift in eiuissions is occurring from developed to developing countries, and new sources of air pollution may arise, due to

deliberate emission of toxic, biological or radioactive material as a result of war or terrorist acts, posing new risks to human and environmental health and climate (Hopke, 2009).

According to Fenger (2009), there are four ways to regulate air pollution: 1) by defining emission standards; 2) by establishing air quality standards; 3) by creating emission taxes;

and 4) by performing a cost benefit analysis. The most commonly used approach is to limit

the emission from a source, a sector or an endre country.

1.2.2. Air pollution effects on human health

According to Kampa and Castanas (2008), an air pollutant is any substance in the air which

may harm humans, animals, vegetation or material or which can cause or contribute to an increase in mortality or serious illness. Air pollutants vary in their chemical composidon,

reaction properties, emission, persistence in the environment, ability to be ti'ansported in

(18)

In this thesis special focus is given to particulate matter (PM) since this pollutant is

considered the most important environmental challenge of steel producüon.

Pardculate pollution is currently the focus of many air quality management and health

protection policies. Aerosols originate from a wide variety of natural and anthropogenic sources. Due to their importance, the scientific community has intensified its efforts to

investigate aerosol's physico-chemical properties. Particles largely vary in their chemical

composition and size.

Particles can absorb and transfer a multitude of pollutants, but their major components are

metals, organic compounds, material of biologic origin, ions, reactive gases, and the particle carbon core. The penetradon of the particles in the respiratory tract differs

according to their granolometry. Coarse particles (aerodynamic diameters (AD) > 2.5 (xm) mainly deposit in the upper respiratory tract, while fine (AD < 2.5 [im) and ultrafine (AD <

0.1 |.im) particles are able to reach lung alveoli (Kampa and Castanas, 2008). Therefore

there is strong evidence to support that ultrafine and fine particles ai'e more hazardous than larger ones. Epidemiological studies have described adverse health effects associated with

exposure to PM, with well-studied health endpoints such as cause-specific mortality, and hospital admissions (Englert, 2004). The exposure to ambient fine particles was recently

estimated to have contributed to 3.2 million premature deaths woddwide in 2010, due largely to cardiovascular diseases, and 223,000 deaths from lung cancer (lARC, 2013).

Epidemiological and toxicological investigations have sought to establish the chemical

components primarily responsible for particle toxicity (Harrison and Yin, 2000; Dall'Osto et al., 2008).

These data supported the announcement, at 17* October 2013, by the specialized cancer

agency of the World Health Organization (WHO), the International Agency for Reseaixh

on Cancer (lARC), that classified outdoor air pollution and PM as carcinogenic to humans.

(19)

1.2.3. Impact of steelworks

Steel plays a fundamental role in the development of modern societies with ever-increasing

demand due to the diverse applications of steel products. According to the World Steel

Association reporting data of 2014, in the 1950-2014 period steel production increased

from 184 million tonnes up to 1,665 Mt. From these millions of tonnes of produced steel,

51.2% was used by the construction sector, followed by the sectors mechanical machinery,

metal products and automotives with 14%, 12% and 12%, respectively (Word Steel

Association, 2015).

Europe is a major producer of steel and the steel industry is dominated by large

multinational companies (Riccardi et a l , 2015) from various European countries such as

Germany, Ukraine, Italy, France, Spain, United Kingdom, Poland and Austria (World Steel

Association, 2015).

The integrated iron and steelmaking process is the main production route used worldwide and is so-called because it involves a number of linked processes: 1) iron ores and coke are

agglomerated in iron ore sintering and cokemaking facüities to use in blast furnace (BF), 2)

iron is extracted from its ores in the BF, 2) the resulting molten iron is subsequently refined

into liquid steel by the basic oxygen steelmaking (BOS) process (EIPPCB, 2013, Dall'Osto

et a l , 2008). Emissions from these facilities ai-e a complex mixture of local and fugitive

emissions.

The impact of steel plants is associated with a number of significant environmental

challenges. Earlier works (Moreno et al., 2004; Dall'Osto et al., 2008) have identified the

steelworks as major contributors to local PMIO concentrations and as major emission sources of metals such as chromium, copper, lead, cadmium, ai'senic, zinc, manganese,

iron, nickel, vanadium and selenium (Taiwo et al., 2014). These metals are mainly

originated from mechanical processes such as moving piles of iron ore, coal and iron ore

loading, and reactions of fine particles in high temperature processes such as cokemaking,

sintering, BF and basic oxygen furnaces (BOF) (Mohiuddin et a l , 2014). Next to the emissions of PM and metals from the steelworks, there are other emissions with

(20)

impact of steelworks also comprises CO2 emissions, due to their energy - and hence CO2

-intensive activity, which accounts for about 7% of the world's anthropogenic CO2 emission

(Riesbeck et al., 2013).

Previous studies demonstrated that inhalation of iron dust is linked to chronic bronchitis, breathlessness, chronic cough, chronic phlegm, pneumoconiosis, reduced lung function,

and can lead to chronic obstructive pulmonary disease (COPD) (Mohiuddin et al., 2014). Such studies have frequendy implicated the metal content (particularly water-soluble

metal) as a possible harmful component of PM, with repercussions for human health (Lippmann et al., 2006; Lippmann and Chen, 2009; Bollati et al., 2010) and ecosystems (de

Vries et a l , 2007).

This relationship between industrial emissions and negative effects for human health was demonstrated by Pope (1996) who reported that during the closure of a steel mill in the

Utah valley, the reduction in PMIO concentrations and changes in PM composition were associated with decreases in morbidity and mortality of the local population. Also

Hutchison et al. (2005) showed evidence of this reladonship, who reported an increase in

PM metal content when samples were taken downwind of a steelworks, which in turn were associated with an increased inflammation of rat lung tissues (Hutchison et al., 2005;

Dall'Osto et a l , 2008).

The above underlines that knowledge on the contribution of steelworks to auborne PM concentrations as well as the identification of the predominant sources within the

steelworks should be increased (Taiwo et al., 2014).

1.3. Soil Contamination

Global industrialization and human activities have an effect on soil contamination not only

due to direct releases to the soil but also due to their emissions to the atmosphere and

subsequent deposition in the soil. This factor contributed to the increase of soil pollution being nowadays a widespread problem in Europe (Science Communication Unit, 2013).

(21)

Metals are the main pollutants present in soils, and their main origin are anthropogenic sources as mining, smelting procedures, steel and iron industry, chemical industry, traffic,

agriculture and domestic activities (Elbagermi et al., 2013). Cd, Cr, Cu, Fe aie the elements of highest concern due to their toxic characteristics: human health can be affected by soil

contamination through pathways as inhalation of dust, ingestion of soil, dermal contact, or ingestion of contaminated food by metals accumulation (Abrahams, 2002; Parra et al.,

2014). Up to today, there is no European Union legislation specifically targeting the

protection of soil. Recently a soil framework directive was proposed (COM/2012/46), but it is not yet officially established. Surveys on soil contamination are also important however,

to better understand air quality assessment surveys.

1.4. Aerosol monitoring

In aerosol (PM) monitoring, PM characterisdcs are of importance in impact studies, for

source identificadon, and in establishing control strategies (Alfarra, 2004). In the majority

of these works an instrumental approached is applied, through the use of appropriate

equipment to measure particle characteristics such as size, number concentration, shape,

mass concentration, and composition, or to sample particles in filters that are subsequently

analysed in order to determine their chemical composition. It should be noted here that the

adopted filter-sampling does not allow the capturing of transient emission events: the

selected sampling protocols do not have a time resolution sufficiently short to precisely

pinpoint non-continuous steelmaking processes.

In contrast to biological approaches however, the filter technique permits the assessment of

PM concentration and mass.

1.5. Biological monitoring

Bioindicators and biomonitors have prominendy been used in air quality monitoring,

through the assessment of the response of living organisms to different levels of pollution, due to their capability to indicate the presence and amount of atmospheric pollutants

(22)

of anatomical, morpliological and pliysiological characteristics, or the analysis of trace

elements in plants, lichens, mosses, small plants or tree leaves (Hofman et al., 2014).

A bioindicator is an organism (or a part of an organism or a coiimiunity of organisms) that

provides qualitative information on the environment (or part of it), while a biomonitor is an

organism that gives quantitative information on various aspects of the enviromnent

(Markert, 2007).

The biomonitoring technique, which has gained increased interest in the last decades, bears

numerous advantages, such as the ability to perform high-density sampling at virtually any

desired spatial scale and to ineasure a wide range of pollutants simultaneously, with low

costs and manpower (Wolterbeek, 2002a; Lage et al., 2015).

Lichens are a slow-growing association of fungi (mycobionts) and green algae or

cyanobacteria (photobionts). This symbiotic association forms a common thallus that does not possess roots or waxy cuticles and depends mainly on an atmospheric input of mineral

nutrients. Lichens are considered the best bioindicator of air pollution due to their

advantageous physical characteristics combined with their capability to accumulate airborne chemical elements, owing to their metabolism, which is strictly dependent on the

atmosphere, and their resistance to metal toxicity. Additionally, they grow in highly variable habitats, thus may cover large geographical ai'eas,and may accumulate mineral

elements to levels that are far above their physiological needs (Gaity, 2001; Abril et al., 2014).

According to Markert et al. (2007), lichens can be used in biomoiutoring surveys in an

active or a passive way. The first term means that lichens are collected in a specific site and transplanted to the area of investigation for a defined period of dme, whilst the second

deals with lichens that exist in-sitit, being native lichens. The use of transplants in surveys have additional important advantages such as the standardization of experimental material

in terms of physiological condition and capacity for bioconcentration (Godinho et al.,

2008).

(23)

Lichens may retain pollutants by their deposition on their surface as dry particulates or as

material dissolved and/or suspended in precipitadon. According to Tyler (1989), elements

can be retained by particulate entrapment, physiochemical processes such as ion exchange,

as well as by passive and active indacellular uptake. This enforces the thought that lichens

are not only passive samplers, but that their accumulation may be viewed as a dynamic process, involving uptake and release processes until equilibrium with the surrounding

environment is reached (Godinho et al., 2008).

Although biomonitoring is generally seen as valuable for the implementation of

environmental policies on air quality (Pirintsos and Loppi, 2008), employing biomonitors

in assessing the atmospheric quality in industrial areas should be compatible with the

necessary spatial resolution (Hofman et al., 2014).

1.6. Presently used analytical techniques

The impact of the steelworks emissions was assessed in this study using three different

approaches: PMIO monitoring, biomonitoring, and topsoil monitoring. In this thesis the

epiphytic lichen Pannelia sulcata was used as biomonitor.

Sampled filters, lichens and soil were analysed using Insdumental Neutron Acdvadon

Analysis (INAA), in the Portuguese Research Reactor (PRR), to determine the elemental

contents of the samples. Earlier work showed that INAA is very useful for environmental pollution control and monitoring studies due to its ability to analyse solid phase samples for

many elements, without the need for sample dissolution or digestion and with a high degree of sensidvity and selecdvity. Also, Particle Induced X-Ray Emission (PIXE), Ion

Chromatography, Indophenol-blue Spectrophotometry, and Atomic Absorption Spectroscopy were used during this work. Data generated by these techniques were used as

(24)

1.7. The case study: the Gijon steelworks

The present study focuses on an integrated steelworks situated in the north of Spain, in

Gijon, Asturias. In the Gijon district, near the urban centre, there is an industrial complex

composed by a steelworks, a cement factory, a power plant and a harbour (Figure 1.1).

Figuj-e 1.1- Area of study of the present work.

(25)

The steelworks includes the following facilities: 1 coal yard; 1 ore yard; 1 coke oven plant;

2 blast furnaces; 1 steel plant; 1 heavy plate mill; 1 wire rod mill; 1 rail mill, and the Gijon

harbour installations (Figure 1.1). The main products manufactured are heavy plate, wire

rod and rail (Figure 1.2).

Figure 1.2 - Steelworks facilities localization.

Details of the steelworks processes are explained in the reference document for iron and

steel production, from the Journal Research Center Reference Report (EIPPCB, 2013).

1.8. Thesis outHne

To achieve the main objectives of this thesis, which are 1) quality control of the of the

INAA methodology, 2) development and application of integrated methodologies to characterize the environment impact of integrated steelworks, and 3) application of source

apportionment techniques and mapping to identify the main emission sources of pollutants,

(26)

I N D U S T R I A L E^^ssIO^•s I M P A C T rv . A M B I E N T .AIR

QU.ALnY

Uliatis the iiülueiice of die steelworks' eirassioiis in its

Inipro\ement

3. QI-.U.1IY .iSSl-RVNCE OF I.N.A.A

. y P L l I D TO H L I I R S LICHINS .V(D SOILS QUALiTA' C O N T R O L

Evaluating qualit\' control of I N A A for the diflerent

matrixes. 1 A X R O S O L M O N T T O R I N C Temporal information on element concentration in aerosols. 3. C H O n C . U . C R i R A C T I R I U T l O N Or.tlMOSPHIRIC F.\RTICL£S I'SING I N S T R n n N T . U METHODS B I O L O G I C A L M O N T T O R I N G Spatial information on element concentration in lichens. 4. MO.MTORI.VG S P . l T l i l DISTRIBITIO.V O F E U C I R I C CONDVCTlMTV.i.\D .iTMOSPHIRlCELnilNTS USIXG TR4.\SPL.4.NT IICHI.VS I N T E G R A T E D .APPROACH

Complete characterization of the integrated steelworks' emissions. Steelworks impacts' in lire \icinities.

S O I L

CON-T.A-MIN.ATION

Spatial infonnation on element concentration in

soils

5. ASSISSMIXT OF THI ELEMiyT C0.\T.i-\DN.iT10.\ IN SlTiE.iCE SOIL

.Aplicatiott of mitigation measures to impro^-e the steel production sector.

Figure 1.3. Outline of the thesis.

The thesis is organized in 6 chapters:

Chapter 1 gives the modvation and the objecdves of the study.

Chapter 2 is focused on I N A A as analytical technique, and screens INAA results obtained

by the I N A A laboratory on reference materials, to verify analytical approaches and quality.

Chapter 3 addresses the assessment of the steelworks' emissions, based on element

concentrations measured in aerosols, thereby considering daily temporal patterns,

fine/coarse PM ratios, crustal enrichment factors, and receptor models.

Chapter 4 addresses air quality as assessed by biomonitoring techniques. Emission sources

are identified by the study of the geospatial distribution of elemental contents in

(27)

transplanted lichens, enrichment and contamination factors and Principle Component

Analysis (PCA).

Chapter 5 is dedicated to soil contamination, assessed by using an Index of

geoaccumuladon, enrichment and contaminadon factors and muldvariate stadstical

analysis.

(28)

(29)

= C H A P T E R 2 =

Q U A L I T Y A S S U R A N C E O F N A A A P P L I E D T O F I L T E R S , L I C H E N S A N D SOILS

2. Quality assurance of INAA applied to filters,

lichens and soils

Based on article: Assessment of tiie Portuguese iig-INAA laboratory performance by evaluating internal quality control

data SM. Almeida, M. Almeida-Silva, C. Galinlia, CA. Ramos, J. Lage, N. Caniia, A.V. Silva, P. Bode

Journal of Radioanalytical and Nuclear Clieniislry (2014) 300:581-587 DOI: 10.1007/sl0967-014-2987-3

2.1. Abstract

The Portuguese INAA laboratory processes approximately one thousand of multi-matrix

samples per year, generating fifteen thousand of results in the same period, using the Icq methodology. In order to ensure that the data produced meets the required quality any

sainple analysed is processed together with a reference material. Therefore, every year a large amount of results of many reference materials are generated. This work analysed a

large database created with the results from the reference materials irradiated in the period

2009-2013. Zeta-scores were calculated and different control charts were created as function of the date of the analysis, the irradiated mass, the reference material and the

operator. The objective of this work was to recognize human errors, to identify deficiencies in the protocols and to improve the quality of the results generated by the laboratory.

2.2. Introduction

For nearly three decades, INAA using the Icq standardizadon method is being applied in the Instituto Superior Técnico (formerly: Instituto Tecnológico e Nuclear) using the Portuguese

Research Reactor (PRR), a 1 MW pool type reactor (Dung et al., 2010a). The irradiations are performed using regular in-pool devices or using fast moving pneumatic devices,

essential for measurements of short-lived radionuclides. The laboratory has several facilities for counting of the radiadon from the induced acdvities, such as germanium

detectors, automatic sample changers and a Compton suppression system (Dung et al., 2010b). The major research lines being pursued using ^o-INAA are related to the

(30)

= C H A P T E R 2 =

Q U A L I T Y A S S U R A N C E O F N A A A P P L I E D T O F I L T E R S , L I C H E N S A N D S O I L S

The Portuguese research group has already shown the advantages of the technique in aerosol studies, such as the many elements that can be measured, the high degree of

accuracy, whereas also little sample preparation is necessary (Almeida et al.,2013a; Freitas et al., 2003, 2004). Currently, the chemical characterization of the indoor (Canha et al.,

2010, 2014a; Almeida et al., 2011) and outdoor (Farinha et al., 2004; Freitas et al., 2005;

Almeida et al., 2006a) particles is used to elucidate the sources of the pollutants and the processes associated with their formation (Almeida et al., 2009a, 2013b, 2014a; Vicente et

al., 2013). This will allow insight in local, regional and long-range transport (Almeida et al., 2008; Almeida-Silva et al., 2013, 2014) and, finally, to identify mitigation options

focusing on the reduction of the air pollutant concentrations (Silva et al., 2012; Almeida et al., 2012a). The attractiveness of ko-lNAA for biomonitoring studies is reflected by 1) the

large number of biomonitoring surveys performed by the group throughout international, national and regional levels (Almeida et al., 2012b; Lage et a l , 2015a), 2) its widespread

use in the identification and characterization of emission sources (Almeida et al., 2012c)

and 3) more recently, its application in the realm of human epidemiology (Canha et al., 2012a).

The objective of the epidemiology research line is to establish unequivocal associations between pollution and morbidity and mortality. Respiratory problems, cardiovascular

diseases and cai-cinogenic incidence in the Portuguese population have been studied in

association with chemical elements measured by /TQ-INAA (Almeida et al., 2007, 2009b). This research line also focuses on the assessment of occupational exposure to chemical

elements and in the development of human bioindicators to be applicable for occupational exposure (Almeida et al., 2010; Félix et al., 2013).

In the nutrition field, different types of supplementation and their efficacy are studied

(Galinha et al., 2011, 2013).

A wide variety of samples is processed in all these lines of research, typically in the order of one thousand per year using specific optimized analytical protocols depending on the

matrix. Variables are 1) the mass of the samples; 2) the packaging procedure; 3) the

irradiation position and 4) the uTadiation, decay and counting times. The work process

from the sample preparation until the spectrum analysis is performed by several students

(31)

= C H A P T E R 2 =

and scientists but obviously, results obtained by these different operators, under different

protocols, using different detectors should be consistent with each other.

Therefore, the laboratory has to anticipate continuously on situations that can affect the

quality of the results (Bode and van Dijk, 1997). In order to ensure that the data produced meets the required quality, thus making it fit for the intended purpose, quality assurance

and quality control has to be implemented. Control charts are a valuable tool in this for monitoring some of the variables (Bode and van Dijk, 1997; Koster-Ammerlaan et al.,

2009). The analysis of reference materials and the resulting patterns in control charts is therefore the onset to establish i f systematic errors may have been made, and the associated

need for acdons aiming the improvement of the performance (Bode and Blaauw, 2012).

The objective of the work described in this paper was to recognize human errors, to identify

deficiencies in the protocols and to evaluate the quality of the analysis results of different

reference materials.

2.3. Experimental

For internal quality control, in the Portuguese /TQ-INAA laboratory any sample analysed is

processed together with a reference material, which is chosen according to 1) the known

amount of the elements of interest; 2) the suitability for irradiation and measurement of the

material and 3) the detectability of the radionuclides of the elements of interest under these

conditions. Table 2.1 presents an overview of the variety and frequency of the principal

(32)

= C H A P T E R 2 =

Q U A L I T Y A S S U R A N C E O F N A A A P P L I E D T O F I L T E R S , L I C H E N S A N D SOILS

Table 2.1. Irradiation and measuring conditions for the reference materials analysed in the period 2009-2013.

Irradlialion Conditions Measuring Conditions

Reference Mnterial No. of Associated Irradiated Tiieniial analysed R M samples mass (nig) Flux (cin'^s"^) f (I

I r r . T Wait Meas. Wiiil Meas. (h) T ( d ) T ( h ) T ( d ) T ( h ) NIST-16.«a CoalHyA.sh 119 AcriKols 11-181 7.0x10" 50 O.mS

IAEA-336 Lfchens .16 Ljcheas 134-474 3.9xl0'' 69 OMUS IAKA-Soil7 Soil 12 Soils 1.19-207 3.9xUl" 69 OMW IGGE-GBW07406 Son 11 Soils öS-l.'il 3.9x10'' 69 O.OWS IGGE-OBW074M Soil 5 Soils 6.5-111 3.9x10'' 69 0.0045 N1ST-I.568a Rice nour 14 Coreak 49-246 3.9xl0" 69 0.0M5 NI.ST-1567a Wlieat Flour 3 Ceiïals 77-126 3.9xl0" 69 0,(XW5 NIST-1.572 Citrus Leaves 10 I'lmls 135-143 3.9xl(l" 69 0,(KM5 I N C T - 0 D T L 5 OrtnlalTiiliacco Leaves 4 Ptints 149-152 3.9x10'' 69 0.0(M5

3-4 3-4 3-4 3-4 3-4 3-4 3-4 3-4 3-4 7 2 L 5 L 5 1.5 4 4 1 3 28 28 28 28 28 28 28 28 2K 2.5 2.5 2.5 4 4 3 3 Irr, T -Irnidi;ition liiuc;\V;iil T - Wiiiiiaij tiiuc; Meas. T - Mcasurini! Tiine; NIST - National Institute of Standards and Teclinoloiry: IAI':,'\ - International Atomic Energy Agency; IGGE - Institute of Geophysical and Geoclienrical Exploration; I N C T - Institute of Nuclear Cltetuistry and Technology

The reference materials were co-irradiated with the samples in two different positions near

the reactor-core depending on the matrix of the samples. The certified reference material

NIST SRM-1633a (Coal Fly Ash), was irradiated with aerosol filters in Cell 55, which has a thermal neutron flux of about 7.0 x 10'^ c m ' V (/=50 and a=0.005). The other reference

materials were irradiated with soils, plants, lichens and cereals in Cell 56, which has a thermal neutron flux of about 3.9 x lO'^ cm"^s"' (/=69 and a=0.005). Irradiation times are

also presented in Table 2.1.

The induced activity of the samples and the reference materials was measured with two calibrated Ge detectors, each with a F W H M of approximately 1.85 keV at 1332.5 keV and

a reladve efficiency of 30%. Measuring conditions are presented in Table 2.1. The gamma-ray spectra were interpreted by ko-lABA program.

A large database was created with the results for 9 quality control reference materials,

obtained by 6 operators and under different analytical protocols, and Zeta scores were calculated following the Equation 2.1:

(33)

= C H A P T E R 2 =

in which X i a b is the measured mass fraction of the element in the reference material, x^si is

the certified/indicative mass fraction, »/oi) is the combined standard uncertainty of the

measured result and « r e f is the combined standard uncertainty of the certified value. The

combined standard uncertainties are derived from the listed expanded uncertainties,

ignoring, i f applicable, the uncertainty from the bias since this value is not given by the

supplier. For non-cerdfied values, which are not furnished with their uncertainties, the

expanded uncertainties were set to as 10% reladve of the values given in certificate

(Kubesova et a l , 2011). The results were interpreted according the following classes: \Q<2,

considered as a satisfactory level; 2<|^|<3, classified as a questionable level and |^|>3,

which is an unsadsfactory level (ISO/IEC, 2010).

2.4. Results and Discussion

Figure 2.1, which presents the percentage of measurements classified by Zeta-score level,

indicates that 73% of the values are at a sadsfactory level, 9.7% are at a questionable level

and 17% are at an unsadsfactory level. Results also show that 67% of the Zeta-scores

present a negative value indicating the existence of a systematic bias. This could be a

systematic error due to the measurement and/or preparation of the flux monitor since this

affects all elements in the same way.

The percentage of values in the interval -3<^<3 was calculated for each element. Mo, Se, U

and Zn presented percentages higher than 90%; Ba, Br, Ca, Cs, Rb, Sb, As, Ce, Co and Sm

had percentages between 80% and 90% and K, La, Na and Sc were within the interval

75-80%. Cr and Fe presented the lowest percentage of values in the interval -3<^<3 (66% and

(34)

= C H A P T E R 2 =

<-3 [-3,-2] [-2,0] [0,2] [2,3] >3 Zeta-scores

Figure 2.1. Percentage of measurements classified by Zcta-Scores class {\Q<2, satisfactory level; 2<|g<3, questionable level and Q>3, unsatisfactory level) for all data and only for data associated with masses higher than the prescribed by the reference material.

In order to identify the sources of non-conformance (|^|>3) data fi-om the reference

materials were displayed in various control charts where Zeta-score was displayed as a

function of 1) the analysis date; 2) the mass of the element; 3) the reference material and 4) the operator. These charts were made in order to check the results from a single analysis

(such as in graphs analysing the Zeta-score in function of the analysis data and of the mass

of the elements) or from a group of analysis (in case of the graphs where Zeta-score was

plotted in function of the reference material and of the operator).

Figure 2.2 shows the Zeta-score as a function of the date of the analysis discriminated by reference material. This control chart was created in order to indicate incidental deviations

and trends. Results showed that in the year 2012, the performance of the laboratory

decreased, especially for the certified reference material NIST SRM 1633a (Coal Fly Ash), which presented \Q>3 for the elements As, Br, Ce, Co, Cs, Fe, La, Sb, Rb, Sb, Cr, Sm, U

(35)

= C H A P T E R 2 = Q U A L I T Y A S S U R A N C E O F N A A A P P L I E D T O F I L T E R S , L I C H E N S A N D SOILS As 1)9 10 11 12 13 ft O c/) N Ce 9 10 11 12 13 Fe 2 ^ III 0 -10 -2( 10 11 12 13 19 10 I I 12 13 Ó9 10 11 12 13 Ba Br 10 I I 12 13 . Co 1)9 10 11 12 13 16l .1^ C r 9 10 11 12 13 1)9 10 11 12 13 C a V. • N I S T 1572 18 N I S T 1 6 3 3 a NTST1567a • N I S T I 5 6 8 a G B W 0 7 4 ( M GBW07-106 I A E A 3 3 6 C s 09 10 11 12 13 L a )9 10 11 12 13 Rb 9 10 11 12 13 Sm 09 10 11 12 13 1 6 1 -1Ó» 5 _ 0 -5 1(1 -i f 6 0 -6 -13, 10 11 12 13 Sb 19 10 11 12 13 •Mo i9 10 11 12 13 Sc 1)9 10 11 12 13 u T)9 10 11 12 13 Zn '9 10 11 12 13 Years

Figure 2,2, Control chart showing the Zeta-score as a function of the analysis date (grossly indicated by the years in this millennium) discriminated by reference material and element.

Figure 2,3 presents the Zeta-score as a function of the mass of the element that can indicate problems associated with detection limits and inhomogeneity of the sample. Information

can be obtained on the performance of the technique for the determination of an element at

a certain level in a given maü'ix. Results from this control chart show that, for the elements identified previously and for the Coal Fly Ash, |^|>3 were associated with samples which

had the lowest element masses. Table 2,1 shows that the irradiated masses of this certified reference material varied between 11 and 181 mg, which is an amount lower than the

iTunimum required by the producer. The use of small amounts of masses was done in order to allow the co-irradiadon of this reference material with aerosol filters without becoming

excessively activated. However, the irradiation of small amounts of material can pose

problems, due to the limited homogeneity of the sample or due to the proximity to the detection limits, and this was reflected in the control chart presented in Figure 2.3.

(36)

= C H A P T E R 2 = Q U A L I T Y A S S U R A N C E O F N A A A P P L I E D T O F I L T E R S , L I C H E N S A N D SOILS • N1ST1572 • N I S T 1 6 J i a NIST1567a • N I S T 1 5 « l a GD\VU74(M Ü B W 0 7 4 0 6 Soil? • O B T L 5 IAKA3J6

o.onon 0.0004 o.oooso.ooo o.ooi

Mass of Elements (mg)

Figure 2.3. Control chart showing the Zeta-score as a function of the sample mass of the elements discriminated by reference material and element.

Therefore, it is probable that the analytical results for the real samples co-irradiated with the reference material, in this case the aerosol filters, are correct but not revealed due to the

low amount of reference material that it was used. This fact reflects a problem that exist in the quality control of the aerosol filters analysis, which is the lack of adequate reference

materials (with a similar matrix) to be co-irradiated with these samples. The few reference materials that exist on the market cannot be irradiated with all batches of fdters due to

economic reasons. In order to overpass this restriction, the Portuguese kg-WAA laboratory

applies three methodologies in order to control the quality of aerosol filters analysis: 1) the preparation of simulated air-filters by spiking known amounts of standard solutions onto

Nuclepore polycarbonate filters (Almeida et al., 2003a); 2) the analysis of different parts of aerosol filters by ^o-INAA (Almeida et al., 2003b), and 3) the parallel analysis of different

parts of the same fdter by ko-lNAA and PIXE (Almeida et al., 2003b).

(37)

= C H A P T E R 2 =

The highest Zeta-values for Cr were observed for IAEA336 and NIST SRM 1572, which

are the materials that presented the lowest mass fraction for this element (1.06 mg.kg"' and

0.8 mg.kg'', respectively). Table 2.1 indicates that the irradiated masses of IAEA336 varied

between 134 and 474 mg and results showed that for masses varying between 135 and 207

mg the value was 10 whereas for masses vai^ying between 392 and 474 mg the

Zeta-value was 0.5. Therefore, also for this element the low sample masses ai'e the probable

reason for |^|>3.

Since the control charts present in Figure 2.3 indicated that the higher were obtained

when operators used masses lower than those indicated by the reference materials providers, the percentage of measurements classified by Zeta-score level was re-calculated

only considering the reference materials that were analysed with masses higher than the prescribed amount (Figure 2.1). Results indicated that results improved significantly being

91% of the values at a sadsfactory level, 4.7% at a quesdonable level and 4.3% at an

unsatisfactory level.

Figure 2.3 shows that for Fe high Zeta-scores were obtained for the reference material

GBW07406 which has a Fe indicadve value of 61015 mg.kg"'. Results showed a very good

precision (average - 52191 mg.kg"' and standard deviation = 3360 mg.kg"') and counting statistical en'ors lower than 2%. Therefore, differences between measured and reference

values seem to be related with the quality of the indicative value.

Figure 2.4 presents a control chai't with the Zeta-score given as a funcdon of the reference material discriminated by element. In this case it was considered X|ab as the average of the

mass fraction obtained for the each reference material and Ui^b the respective standard deviation. With this type of chart, a systematic bias for all elements in one specific

reference material may indicate an erroneous way of e.g., moisture content correction,

drying or storage. Results showed that the results obtained for the reference material 0 B T L 5 (Oriental Tobacco Leaves) are the worst: Ca, Ce, La and Sc at the unsatisfactory

(38)

= C H A P T E R 2 = Q U A L I T Y A S S U R A N C E O F N A A A P P L I E D T O F I L T E R S , L I C H E N S A N D SOILS o r. k As T J C F Ce Ba Br r i i Co Cr Ca rt Cs Cs Cs -|J U Fe Na •

Se

T- V V- < \-'i^ '{l 'iz. 'L. ^ — •• — 7 7 y y. -z z C - y 7. 7 7 -z z

Figure 2.4. Control chart showing the Zeta-score as a function of the reference material discriminated by element.

A contr ol chart displaying the Zeta-score as a function of the user was made for NIST SRM

1633a, which was the most frequendy used reference material, aiming the identification of

analyst depending results. Figure 2.5 showed that in general there isn't any analyst dependency.

(39)

= C H A P T E R 2 = Q U A L I T Y A S S U R A N C E O F N A A A P P L I E D T O F I L T E R S , L I C H E N S A N D S O I L S 3| 1 3l 1 3| 1 3 2 1 0 -1 _2 As 1—1 2 _Ba 2 1 (1 -1 _2 Br 2 1 0 -1 _2 Ca 2 1 0 -1 _2 L - J l _ l ' — ' L J -1 2 1 (1 -1 _2 2 1 0 -1 _2 ^ u _3 3 3 2 1 0 -1 J2 3 2 3 2 1 0 -1 J2 Pn 3 2 _Cr 1 1.—, 3 2 _Cs 3 2 1 0 -1 J2 c e X—O 1 0 -1 _2 Cr 1 1.—, 1 0 -1 -2 -3 12 8 4 Cs 3 2 1 0 -1 J2 L J " L J ' - ^ 1 0 -1 _2 •—• 1 ^1 i 1 0 -1 -2 -3 12 8 4 U ' L - i y u -3 3 2 1 ,3 1 0 -1 -2 -3 12 8 4 -3 3 2 1 3 2 1 0 -1 -2 -3 12 8 4 -3 3 2 1 C o I f 3 2 _La 1 0 -1 -2 -3 12 8 4 Mo -3 3 2 1 r e 1 La 1 0 -1 -2 -3 12 8 4 Mo 0 -1 J2 f—i ₀ -1 -2 3 (—1 ₀ -4 -8 -12 0 -1 J2 L J • - ' • L J 1 1 1—1 l-_J 0 -1 -2 3 U U L J L J 0 -4 -8 -12 I 1 1 1'—' ,3 0 -1 -2 3 0 -4 -8 -12 I 1 1 1'—' 3 2 1 0 -1 -2 3 2 1 0 -1 -2 3 0 -4 -8 -12 3 2 1 0 -1 -2 Na [—1 — -1 3 2 1 0 -1 -2 3 Sb 3 2 1 0 -1 -2 Sc 3 2 1 0 -1 -2 Na [—1 — -1 K D 3 2 1 0 -1 -2 3 Sb 3 2 1 0 -1 -2 Sc 3 2 1 0 -1 -2 1 ^ 1 — 1 ' — • -1 L J ^ L J ^ 3 2 1 0 -1 -2 3 3 2 1 0 -1 -2 1 \' ^1—^1 ^1 1 3 2 1 0 -1 -2 -1 3 2 1 0 -1 -2 3 3 2 1 0 -1 -2 3 2 1 II -1 -2 1 3 2 1 0 -1 -2 Se 3 2 1 II -1 -2 1 '—' I T 3 2 1 0 -1 -2 Zn i—1 3 2 1 0 -1 -2 Se r i i - i 3 2 1 II -1 -2 1 u 3 2 1 0 -1 -2 Zn i—1 3 2 1 0 -1 -2 -2 —< 1 i i — 1 3 2 1 II -1 -2 1 3 2 1 0 -1 -2 3 2 1 0 -1 -2 -2 3 2 1 II -1 -2 1 3 2 1 0 -1 -2 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 ^ 1 2 3 4 5

User

Figure 2.5. Conti'ol ciiart showing the Zeta-score as a function of the user discriminated by element for the certified reference material N I S T S R M 1663a.

2.5. Conclusions

In the Portuguese /TQ-INAA laboratory, all users analyse their samples together with a

reference material to assess i f no systematic errors have been made. The results of these

analyses can be used in more advantageous way when they are all put together in a

database, and condol charts are created to sort and correlate data.

In the present work the assessment of control charts identified sources of errors. The mass

of the irradiated samples was identified as the main cause for weaker results for the

reference material irradiated in 2012, due to the increase of inhomogeneity of measurands

in smaller analysed masses. It is therefore essential to use the reference materials

(40)

= C H A P T E R 2 =

(41)

= C H A P T E R 3 =

C H E M I C A L C H A R A C T E R I Z A T I O N O F A T M O S P H E R I C P A R T I C L E S U S I N G I N S T R U M E N T A L M E T H O D S

3. Chemical characterization of atmospheric

particles using instrumental methods

Based ou articles: Levels and Spatial Distribution of Airborne Ciiemical Elements in a Heavy Industrial Area Located

in tlie North of Spain J. Lage, S.M. Almeida, M.A. Reis, P.C. Chaves, T. Ribeiro, S. Garcia et ai. Journal of Toxicology and Environmental Health, Part A (2014)77:14-16, 856-866

DOLIO. 7152873941080.2014.910156

Chemical ciiaracterization of atmospheric particles and source apportionment in the vicinity of a steelmaliing industiy S.M. Almeida, J. Lage, B. Fernandez, S. Garcia, M.A. Reis, P.C. Chaves

Science of tiie Total Environment (2015) 521-522:411-420 DOI: 10.1016/j.scitotenv.2015.03.ll2

3.1. Abstract

The objective of this work was to provide a chemical characterization of atmospheric particles collected in the vicinity of a steelmaking plant and to identify the sources that

affect PMIO levels. A total of 94 PM samples were collected in two sampling campaigns that occurred in February and June/July of 2011. PM2.5 and PM2.5-10 were analysed for a

total of 22 elements by Instrumental Neutron Activation Analysis and Particle Induced X -Ray Emission. The concentrations of water soluble ions in PMIO were measured by Ion

Chromatography and Indophenol-Blue Spectrophotometry. Fine/coarse ratios were

performed as a first approach to identify natural and anthropogenic sources of particles, as well as the crustal enrichment factor. Positive Matrix Factorization receptor model was

used to identify sources of particulate matter and to determine theu' mass contribution to PMIO. Seven main groups of sources were identified: marine aerosol identified by Na and

Cl (22%), steelmaking and sinter plant represented by As, Cr, Cu, Fe, N i , Mn, Pb, Sb and Zn (11%), sinter plant stack identified by NH4^, K and Pb (12%), an unidentified Br source

(1.8%), secondary aerosol from coke making and blast furnace (19%), fugitive emissions from the handling of raw material, sinter plant and vehicles dust resuspension identified by

A l , Ca, La, Si, T i and V (14%) and sinter plant and blast furnace associated essentially with

(42)

= C H A P T E R 3 =

3.2. Introduction

Steel production is a key sector for Europe's economy and competitiveness and its use is

critical in enabling man to move towards a sustainable future. However, production of steel

is associated with a number of significant environmental challenges, one of which is the

emission of airborne particles (PM) to the atmosphere (Ciaparra et al., 2009).

The integrated iron and steelmaking process, the main production route used in Europe and

worldwide, is so-called because it involves a number of linked processes: 1) the sintering

process is a pretreatment - step where the raw material composed by fine iron ores (mainly

hematite (Fe203), magnetite (Fe304) and goethite (FeOOH) and flux materials are

aggloinerated by cooking, heating/sintering; 2) coke is produced from coal in coke making

plant ovens at the absence of oxygen; 3) sinter and coke are subsequently introduced in

blast furnace where the reduction of iron ores takes place at high temperatures leading to

molten iron; 4) the sulfur impurities are removed from the hot metal, which is then

transported to the steelmaking plant; 5) the carbon content is lowered to less than 1% in a

basic oxygen furnace resulting in steel; 6) downstream ladle metallurgy of the steel is

applied in order to produce steel with the required quality; 7) lasdy, the liquid steel is solidified mainly using condnuous casting process (EIPPCB, 2013).

Although many of the operations consider an abatement of primary and secondary

emissions, important emissions are still produced. The sintering process, mainly the stack of the sinter strand accounts for about 45% of all PM emissions from the iron and

steelmaking process and is also a major emitter of heavy metals such as Pb, which is the major tracer of the sintering process (Menad et al., 2006; Sammut et al, 2010). Two other

important sources of dust emissions are the blast furnace and the steelmaking process. Iron-based particles are released into the air during hot metal casting and are also emided during

the reloading and desulfurization of the hot inetal, basic oxygen furnace charging, oxygen

blowing and tapping, and finally in the secondary metallurgical processes. Diffuse emissions occur from all of these processes whenever the emissions are not fully captured.

An important amount of particles is still emitted through the roof openings. Different types of slag are generated in the iron and steel industry: granulated blast furnace, basic oxygen

(43)

= C H A P T E R 3 =

furnace, secondary metallurgie and desulfurization slag. The latter contains important

amounts of fine particles (Hleis et al., 2013).

PM emissions from integrated steelmaking facilides are therefore a complex mixture of stationary sources and diffuse emissions associated with the main processes and with the

general site operations such as the stocking, blending and transportation of raw materials.

There are processes which operate continuously while others are batch processes, some of

which may be quasi-continuous. Owing to the non-continuous nature of many steelmaking

processes it is difficult to capture transient emission events arising from specific operations.

Moreover, the major steelworks processes are located in very close proximity to each other,

being also close to other emission sources like traffic or other industries, making it difficult

to distinguish between the contributions of the processes. Atmospheric PM generated from

steel industries have high concentrations of As, Cd, Cr, Cu, Fe, Mn, N i , Se, V, and Zn

(Taiwo et al., 2014). However, information on emissions from the steel processes are still

needed due to the scarce data existent.

The health effects of PM have been subject of intense study in recent years (Dominici et al,

2005; WHO, 2013; EEA, 2014; Almeida et al, 2014a; Cruz et al., 2015). The European Directive 2008/50/EC defined legally binding limits for the concentrations of PMIO and

PM2.5. The European Member States were required to draw up plans to guarantee compliance with the defined limit values over the period up to 2020. However, the current

policy efforts, at EU and national level, have not fully delivered the expected results. The development of cost-effective sdategies through efficient air quality management depends

upon a quantitative knowledge of the condibution of different sources to airborne PM concendations. Receptor modelling is a widespread approach to identify emission sources

and to resolve their contribution to PM mass.

Since protecting the safety and health of everyone who works in or around steelworks is of vital importance to the steel industry, the present study was developed to investigate the

chemical composition of particles in the vicinity of steelworks and to improve the understanding on the dominant marker elements and on the contribution of this industry to

(44)

= C H A P T E R 3 =

ions in atmospheric particles collected in the vicinity of a steelworks located in Spain were

analysed and Positive Matrix Factorization receptor model was used to identify and

apportion emission sources of particles in the studied area. The objective of this study was not only to identify the steelworks emission profiles, but also to attribute emissions to

specific production units in the steelworks complex.

3.3. Materials and methods

3.3.1. Studv Area

This study was carried out in the north coast of Spain, in Asturias (coordinates latitude

4 3 ° 3 4 ' 5 3 " N , longitude 5°51'07" W at the northwest corner, and latitude 43°25'24" N , longitude 5°36'04" W at the southeast corner). This area within 18 km x 20 km includes a

heavy industrial area formed by a steelworks, a harbour, a cement factory and a power plant, as well as the urban area of Cujon with 25,700 inhabitants, rural areas and heavy

traffic roads. This region has a strong influence of the coast line and the local topography.

3.3.2. Aerosol sampling and analysis

A total of 94 PM samples were collected adjacent to an integrated steelworks located in the

North of Spain (ladtude 43°31'7"N, longidide 5°43'51"W, height 20 m above sea level). Sampling was performed in two periods that occurred from 3"* to 15* February 2011 and

from 22"^" June to 4"' July 2011. The sampling station was located 4 km fiom the Atlantic

Ocean and 4 km from the urban ai'ea of Gijon. Figure 3.1 presents the localization of the steelworks in Spain and the location of the different processes inside the factory. Wind

roses are also presented for the sampling campaigns. Meteorological data was obtained froin the model TAPM (Hurley et al. 2005), developed by Australia's Commonwealth

Scientific and Industrial Research Organization (CSIRO). The meteorological inodule of TAPM uses 1) global databases of terrain and land use from the Earth Resources

Observation Systems, 2) surface temperature from the US National Center for Atmospheric Research, 3) synoptic conditions from the Limited Area Prediction System and Global

(45)

= C H A P T E R 3 =

conditions for tlie model outer grid, TAPM uses large-scale weather information (synopdc

analyses or, potentially, weather forecasts) typicaUy available at a horizontal grid spacing

of 100 km. Next, TAPM 'zooms i n ' to model local scales at a fmer resolution using a

nested approach, predicting local-scale meteorology. Results show that during the winter

sampling campaign the wind was predominandy from S, SE and NE whereas during the

summer sampling campaign the wind was predominantly from N and NW.

i m i ' l '"1* 'I

Figure 3.1. Sampling site localization at local and regional scales. Wind roses for winter and summer campaigns.

Environmental impact of steelworks' emissions

Delft University of Technology

Environmental impact of steelworks' emissions

An integrated approach

Murias Gomes Lage, J.

DOI

10.4233/uuid:9839c2a7-3c7e-4705-9c1f-1868da8c7c1a

Publication date

2016

Document Version

Final published version

Citation (APA)

Murias Gomes Lage, J. (2016). Environmental impact of steelworks' emissions: An integrated approach.

https://doi.org/10.4233/uuid:9839c2a7-3c7e-4705-9c1f-1868da8c7c1a

Important note

To cite this publication, please use the final published version (if applicable).

Please check the document version above.

ENVIRONMENTAL IMPACT OF

S T E E L W O R K S ' EMISSIONS: AN

Table of Contents

1. Introduction

1.1. Motivation and objectives

1.2. Ambient Air Quality

1.2.1. General introduction

1.2.2. Air pollution effects on human health

1.2.3. Impact of steelworks

1.3. Soil Contamination

1.4. Aerosol monitoring

1.5. Biological monitoring

1.6. Presently used analytical techniques

1.7. The case study: the Gijon steelworks

1.8. Thesis outHne

2. Quality assurance of INAA applied to filters,

lichens and soils

2.1. Abstract

2.2. Introduction

2.3. Experimental

2.4. Results and Discussion

Se

User

2.5. Conclusions

3. Chemical characterization of atmospheric

particles using instrumental methods

3.1. Abstract

3.2. Introduction

3.3. Materials and methods

3.3.2. Aerosol sampling and analysis