General Chemistry, Sampling, Analytical Methods, and Speciation

C H A P T E R
General Chemistry, Sampling, Analytical

Methods, and Speciation*
DONALD R. SMITH AND MONICA NORDBERG
ABSTRACT The commonly used term “heavy metals” to describe

metals or metalloids that can give rise to toxicity was
This chapter provides an introduction to the general brought to the attention of the participants of work-
chemistry of metals with particular emphasis on their shops held by the Scientific Committee on Toxicology
role in biology and toxicology. This is followed by a of Metals (SCTM) under the International Commission
description of aspects of analytical chemistry relevant on Occupational Health (ICOH) in the 1970s. It was
to trace element analysis. Because total elemental anal- concluded by the Task Groups that the term should
ysis is strongly complemented by sophisticated metal not be used (Task Group on Metal Accumulation,
speciation and fractionation analyses, information 1973; Nordberg, 1976). The improper use of the term
is also given about useful strategies in sampling and “heavy metals” to design a group of metals and semi-
sample preparation, followed by separation techniques metals (metalloids) associated with contamination and
and detection methods for elemental species. Finally, potential toxicity or ecotoxicity was brought to further
the important topics of calibration and quality assur- attention, and the misuse of the term was critically com-
ance strategies are also discussed. mented upon (Duffus, 2002; Duffus et al., 2007). Never-
Of 114 identified elements—90 occurring naturally theless, the term continues to be commonly (mis)used
on Earth—most (67) form metals, and all but one of in toxicology and legislation to encompass the pure
these (Hg) are solid. Eleven elements make up atomic metal and all its chemical species. This meaningless ter-
or molecular gases, and 12 more elements form solid or minology totally ignores the fact that pure metals and
liquid nonmetals. The chemistry of metals represents a their compounds do not have the same physiochemical,
major part of inorganic chemistry. Understanding the biological, and toxicological properties. It is obvious
toxicology of metal species has advanced substantially that metal species need to be addressed in each case.
during the past few decades, thanks to the consider- Considering the persistence of the misuse of the term
able contributions of bioinorganic chemistry, e.g. the “heavy metals,” it is interesting to learn something more
discovery that metal species undergo biomethylation about the origin of the term. An excellent historic over-
and thus can form organometallic compounds. This view of the terminology “heavy metals” has been pub-
discipline goes hand in hand with chemical speciation, lished by J. Duffus (2002), to which the reader is referred
which is instrumental for understanding metal toxicol- to find the historic references. A brief summary might be
ogy and related adverse health effects. helpful to illustrate the confusion that surrounds the term
and avoid its further use in matters concerning chemistry,
* Partly based on Chapter 2, General Chemistry, Sampling, Ana- policy, and regulations. Before 1936, the term was used
lytical Methods, and Speciation, by Rita Cornelis and Monica Nord- with the meanings “guns or shot of large size” or “great
berg, Handbook on Toxicology of Metals, 3rd Ed. 2007. ability.” The oldest scientific use of the term to be found
Handbook on the Toxicology of Metals 4E

http://dx.doi.org/10.1016/B978-0-444-59453-2.00002-0 15 Copyright © 2015 Elsevier B.V. All rights reserved.
16 Donald R. Smith and Monica Nordberg
in the English literature is cited in Bjerrum’s Inorganic speciation in relation to human health risk assess-
Chemistry, published in 1936. The Bjerrum definition is ment (2006).
based on the density of the elemental form of the metal,
and he classifies heavy metals as those metals with ele-
mental densities > 7g/cm3. Over the years, this definition 1 DEFINITION OF METALS
has been modified by various authors, and there is no
consistency. There were various classifications according Metals are usually defined on the basis of their com-
to the specific gravity: either > 4, or > 5, then 4.5, also 6, mon physical properties in the solid state: they have
and even 3.5. It is evident that it is impossible to come up high electrical conductivity, high thermal conductiv-
with a consensus on the basis of specific gravity, because ity, metallic luster due to their high reflectivity, and
it yields nothing but confusion. At some point in the his- mechanical properties of strength and ductility. These
tory of the term, it has been realized that specific gravity are, of course, the physical properties of greatest tech-
is not of great significance in relation to the reactivity of nological significance.
the metal. Accordingly, definitions have been formulated Metals in the solid state are also characterized by
in terms of atomic mass. The criterion was still unclear their crystal structure, by the specific chemical bond
as some scientists opted for atomic mass greater than in which electrons are delocalized and mobile, and by
23 (sodium); this means magnesium and higher. Others their magnetic properties. These physical properties
take 40 as a criterion, thus starting with scandium (Sc). have only limited value for understanding the sys-
Another suggestion is the ability of the element to form temic toxic effects, although some may be important
soaps with fatty acids as a criterion of “heaviness.” Still for understanding the local effects of metal aerosols
another group of definitions is based on atomic number and nanoparticles.
and suggest citing metals above sodium (11) as heavy. A more useful definition of metals to make it pos-
One more group of definitions is based on chemical prop- sible to explain the toxic effects is based on their prop-
erties, with little in common: density for radiation screen- erties in aqueous solutions. This definition is “a metal
ing, density of crystals, and reaction with dithizone. With is an element which under biological conditions may
this in mind, there is no clear and consistent basis for react by losing one or more electrons to form a cation.”
deciding which metals should be included in this “cat- In the following text, the discussions will be based on
egory” of “heavy metals.” The term is hopelessly impre- the behavior of metals/metal ions in solution and,
cise, leads to confusion, and is useless to describe toxic where applicable, in biological media.
properties. In light of this, in the field of toxicology the The distinction in metal toxicology between metals
term “heavy metal” has been replaced with “toxic metal” and nonmetals, whether on the basis of their physical
(Duffus et al., 2007). or chemical properties, is not sharp. In metal toxicology,
Detailed information about the chemistry of met- some strictly defined metalloids are included because
als is given in a number of textbooks (Cotton et al., they produce adverse health effects in humans, either
1999; Heslop and Jones, 1976; Parish, 1977), and the by themselves or by interaction with other elements.
introductory text by Pauling (1970). A good source They exhibit certain properties that are typical of met-
of information on the properties of ions in solution, als, whereas other properties make them similar to
an important field for understanding metal toxicol- nonmetals. In general, in some groups of the periodic
ogy, is the introductory monograph by Pass (1973). system, a gradual transition of properties occurs from
The principles of bioinorganic chemistry were pre- nonmetals to metals when descending from the lighter
sented in the 1970s by Hughes (1972), Phipps (1976), to the heavier elements (e.g. C, Si, Ge, Sn, and Pb in
Williams (1976), and Fiabane and Williams (1977). group 14). Borderline elements such as As, Ge, Sb, Se,
More advanced aspects about bioinorganic chemis- and Te are sometimes called metalloids. As per com-
try can be found in Dessy et al. (1971), Addison et al. mon definition, metalloids present more metallic prop-
(1977), Fraústo da Silva and Williams (1991), and erties by increasing valency, such as is the case with Te.
Williams and Fraústo da Silva (1996). Theoretical and
chemical aspects on the toxicology of metals can be
read in Hill and Matrone (1970), Brakhnova (1975), 2 THE PERIODIC TABLE
and Goyer et al. (1995). Two handbooks on elemen-
tal speciation covering analytical techniques, meth- The periodic table consists of seven horizontal rows
odology, and element-by-element review were fairly called periods and 18 vertical columns, as presented in
recently published by Cornelis et al. (2003, 2005). Table 1. The element with the lowest number of pro-
Also, the 2006 World Health Organization Interna- tons is H, with one proton. Increasing the number of
tional Programme on Chemical Safety (WHO/IPCS) protons increases the atomic number and yields a dif-
publication highlighted the importance of elemental ferent element. With an equal number of protons, the
2 General Chemistry, Sampling, Analytical Methods, and Speciation 17
TABLE 1 The Periodic Table1
Group 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Period
1 2
1 Key: He
H
Atomic #
3 4 Symbol 5 6 7 8 9 10
2
Li Be B C N O F Ne
11 12 13 14 15 16 17 18
3 Na Mg Al Si P S Cl Ar
19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36
4
K Ca Sc Ti V Cr Mn Fe Co Ni Cu Zn Ga Ge As Se Br Kr
37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54
5
Rb Sr Y Zr Nb Mo Tc Ru Rh Pd Ag Cd In Sn Sb Te I Xe
55 56 * 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86
6
Cs Ba Hf Ta W Re Os Ir Pt Au Hg Tl Pb Bi Po At Rn
87 88 ** 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118
7
Fr Ra Rf Db Sg Bh Hs Mt Ds Rg Cn Uut Fl Uup Lv Uus Uuo
57 58 59 60 61 62 63 64 65 66 67 68 69 70 71
*Lanthanoids
La Ce Pr Nd Pm Sm Eu Gd Tb Dy Ho Er Tm Yb Lu
89 90 91 92 93 94 95 96 97 98 99 100 101 102 103
**Actinoids
Ac Th Pa U Np Pu Am Cm Bk Cf Es Fm Md No Lr
Claims for the discovery of elements 113, 115, 117, and 118, for which no assignments have yet been made, are being considered by an IUPAC
and IUPAP joint working party.
Source: Table updated from the IUPAC 1 May 2013 version (http://www.iupac.org/fileadmin/user_upload/news/IUPAC_Periodic_Table-1
May13.pdf).
1 As of February 2014, claims for the discovery of four additional elements, atomic numbers 113, 115, 117, and 118, for which no assignments
have yet been made, are being considered by an IUPAC and IUPAP joint working party.
number of neutrons for each element determines the either in ionic or covalent bonds. Intermediate forms
isotope. Elements can thus occur as several isotopes— are also seen. Dissolved in water, the metallic com-
a property of great usefulness in studies that have pounds dissociate into metal ions, mostly as cations.
used stable isotope abundances to assess sources of In some cases, such as the permanganate ion (MnO4−),
human exposure to toxic metals such as lead and ura- an oxoanion is formed. Metallic ions can form com-
nium (Gwiazda and Smith, 2000; Gwiazda et al., 2004, pounds with other metallic ions, forming alloys with
2005). Some have an unstable nucleus, so they also dis- two or more metals in varying proportions. Binary and
play radioactivity besides their chemical properties. multicomponent systems also exist in the crystalline
The electron configuration of elements is described in phase (FeS is an example).
orbitals assigned as s, p, d, and f shells or subshells,
thus indicating the spatial configuration of the elec- 3.1 Covalent and Ionic Bonds
trons. The s-orbital is spherically symmetrical around
the nucleus of the atom of the element. The orbitals p, Two major types of chemical bonds exist (i.e. cova-
d, and f, are not spherically symmetrical. The chemical lent and ionic). The covalent bond is defined as a
properties of an element depend on the specific elec- region of relatively high electron density between
tronic configuration of the atom, which varies in a sys- nuclei that arises, at least partly, from the sharing
tematic way according to the atomic number. of electrons and produces an attractive force and
characteristic internuclear distance (McNaught and
Wilkinson, 1997). Covalent bonds exist as homopolar
3 COMPOUNDS OF METALLIC ELEMENTS and heterocovalent. When the two atoms of the diatomic
molecule are the same (e.g. two hydrogen atoms), the
Metallic elements and metalloids form compounds electron density is distributed symmetrically between
in various oxidation states, yielding inorganic com- the two nuclei, and the covalent bond is homopolar. If
pounds such as salts and salt-like products, metal com- the two atoms are not the same, the electron distribu-
plexes (coordination compounds), and organometallic tion will be asymmetrical, and the electron density will
compounds. In metallic compounds, the atoms bind be displaced toward the atomic nucleus that is more
electronegative (i.e. that has a higher capacity to attract electron pairs that were shared with the central atom
electrons). This is a heteropolar covalent bond. The (McNaught and Wilkinson, 1997). It is represented by
greater the difference in electronegativity between two a roman numeral, for example, CrIII and CrVI.
atoms forming a bond, the more uneven the distribu- The oxidation state is a measure of the degree of
tion of the electrons will be. In an extreme case, a com- oxidation of an atom in a substance. It is defined as
plete transfer of electrons from one atom to another the charge an atom might be imagined to have when
occurs, thus forming an ionic bond. Metallic elements electrons are counted according to an agreed-on set of
have low electronegativity. Chemical bonds are rarely rules: (l) the oxidation state of a free element (uncom-
entirely covalent or entirely ionic. Ionic bonds are pre- bined element) is zero; (2) for a simple (monoatomic)
dominantly formed in metal salts like chlorides (NaCl) ion, the oxidation state is equal to the net charge on
or nitrates [Ca(NO3)2]. Covalent bonds are predomi- the ion; (3) hydrogen has an oxidation state of +1 and
nantly, but not exclusively, formed between metals and oxygen has an oxidation state of −2 when they are
carbon atoms, as in organometallic compounds such as present in most compounds (exceptions to this are that
dimethylmercury (CH3-Hg-CH3). hydrogen has an oxidation state of −1 in hydrides of
active metals (e.g. LiH) and oxygen has an oxidation
state of −1 in peroxides (e.g. H2O2); (4) the algebraic
3.2 Oxidation Number
sum of oxidation states of all atoms in a neutral mol-
Oxidation can be defined according to the following ecule must be zero. In ions, the algebraic sum of the
three criteria (McNaught and Wilkinson, 1997). oxidation states of the constituent atoms must be equal

to the charge on the ion. For example, the oxidation
(1) Oxidation is the complete removal of one or more
states of sulfur in H2S, S8 (elementary sulfur), SO2, SO3,
electrons from a molecular entity, for example, the
and H2SO4 are, respectively, −2, 0, +4, +6, and +6. The
Zn2+ ion derives from the atom Zn, of which two
higher the oxidation state of a given atom, the greater
electrons have been removed.
is its degree of oxidation; the lower the oxidation state,
(2) This definition can be extended to chemical reac-
the greater is its degree of reduction (McNaught and
tions in which complete transfer of an electron
Wilkinson, 1997).
does not occur and which, by custom and in cur-
Of course, the oxidation state of a metal is an impor-
rent usage, are called oxidations. In this applica-
tant factor in the coordination chemistry and reactiv-
tion, oxidation numbers are considered. Oxidation
ity of the metal, and hence its potential toxicity (see
now is an increase in the oxidation number
Section 5.2); for example, in their reduced oxidation
of any atom within a substrate, for example,
states transition metals such as FeII, CuI, and MnIII may
Fe2+ − e− ↔ Fe3+. The oxidation number in an ion
participate in Fenton chemistry reactions and the gen-
or a molecule is the charge the atom would have
eration of reactive oxygen species.
if the polyatomic ion or molecule was composed
entirely of ions. Thus, for example, in MnO4−,
manganese is considered to be in the oxidation 3.3 Inorganic Compounds
state +7 (MnVII), and oxygen is assumed to exist as
Metallic elements form a great variety of inorganic
an O2− ion.
compounds. They can be classified into binary and
(3) Oxidation is also the gain of oxygen and/or loss
multielement compounds. The most important binary
of hydrogen of an organic substrate. All oxida-
compounds, both from the technological and the toxi-
tions meet criteria 1 and 2, and many meet crite-
cological viewpoints, are oxides and sulfides. These
rion 3, but this is not always easy to demonstrate.
are the chemical forms in which most metals appear
Alternatively, oxidation can be described as a trans- in nature: the minerals and ores. The metal-containing
formation of an organic substrate that can be rationally aerosols produced in metallurgical processes often
dissected into steps or primitive changes. The latter occur as metal oxides. However, in experimental toxi-
consist of removal of one or several electrons from the cological studies, chlorides and acetates are the most
substrate followed or preceded by gain or loss of water commonly used metal compounds because of their
and/or hydrons or hydroxide ions, or by nucleophilic high solubility in water and biological systems.
substitution by water or its reverse, and/or by an intra-
molecular rearrangement.
3.4 Metal Complexes
Oxidation number is used to define the state of oxi-
dation of an element. The oxidation number of a cen- A metal complex or coordination compound is
tral atom in a coordination entity is the charge it would formed by the association of a metal atom or ion and
bear if all the ligands were removed along with the another chemical species, called ligand, which may
be either an anion or a polar molecule. Of course, it which are acidic (pH 2-6) in the stomach and almost
is the ability to form specific (or in some cases non- neutral (∼pH 6.8) in the intestines. Biological fluids
specific) coordination complexes in biological sys- typically also contain a variety of organic ligands that
tems that underlie many of the essential functions that may influence the solubility of metal compounds.
metals serve in biology, as well as their toxicological Generally, solubility of a substance is given as solu-
effects. Moreover, organic ligand compounds such as bility in water, solvents, and acids. The solubility of
ethylenediaminetetraacetic acid (EDTA), 2,3-dimer- metal compounds/species in water may be a useful
captopropanol, d-penicillamine [(2S)-2-amino- indicator of solubility of these compounds in biologi-
3-methyl-3-sulfanyl-butanoic acid] and succimer cal fluids. A simple rule used in chemistry divides vari-
(2,3-dimercaptosuccinic acid), with reactive carboxylic ous substances into “soluble” or “insoluble.” “Soluble”
acid, amino, and thiol functional groups, form rela- substances have a solubility in water > 1 g/100 mL
tively specific and high affinity complexes with a num- (10 g/L), while “insoluble” substances have a solubil-
ber of toxic metals such as lead and mercury, and as ity of < 0.1 g/100 mL (1 g/L). This distinction may not
a result have received considerable use as therapeutic be meaningful if a substance is highly toxic. Within
agents in the treatment of metal poisoning (Aposhian each group of the periodic table, the solubility of metal
and Aposhian, 1990). compounds/species generally decreases with increas-
ing atomic number.
Usually, nitrates, acetates, and all chlorides, bro-
3.5 Organometallic Compounds mides, and iodides of metals are soluble, except for
Organometallic compounds are classically com- those of silver, mercury (I), and lead. All sulfates apart
pounds having bonds between one or more metal from those of barium, strontium, and lead are also
atoms and one or more carbon atoms of an organyl soluble. All salts of sodium, potassium, and ammo-
group. Organometallic compounds are classified by nium are soluble, except for sodium antimonite,
prefixing the metal with “organo” (e.g. organopalla- potassium hexachlorplatinate, and potassium cobalt-
dium compounds). In addition to the traditional met- initrite. Mainly insoluble are hydroxides (except for
als and semimetals, elements such as boron, silicon, those of alkali metals, ammonium, and barium), all
arsenic, and selenium are considered to form organo- normal carbonates, and phosphates apart from those
metallic compounds. The status of compounds in of alkali metals, ammonium, and the alkaline earth
which the canonical anion has a delocalized structure metals.
in which the negative charge is shared with an atom
more electronegative than carbon, as in enolates, may 5 PROPERTIES OF METAL IONS
vary with the nature of the anionic moiety, the metal
ion, and possibly, the medium. In the absence of direct
5.1 Formation of Metal Ions
structural evidence for a carbon-metal bond, such
compounds are not considered to be organometallic Metal ions are formed by the removal of one or
(McNaught and Wilkinson, 1997). more outer electrons from the neutral atom. The
energy required for ion formation depends on the
environment in which the metal exists. The formation
4 SOLUBILITY of ions in the gas phase requires a considerable amount
of energy. Much less energy is required when the pro-
The solubility of metal compounds in aqueous and cess takes place in water because part of the ionization
lipid environments is of great toxicological importance energy is provided by the energy of hydration (i.e. the
because it is one of the major factors influencing the energy that is gained when a positively charged metal
availability and absorption and disposition of met- ion binds dipolar water molecules). The number of
als in the body. The solubility of metal compounds water molecules that are bound directly to the metal
in aqueous environments depends largely on the oxi- ion (first hydration sphere) depends on the size and
dation state of the metal and chemical nature of the the charge, i.e. charge radius, of the metal ion and var-
compound, the pH of the environment and presence ies from 4 for Li+ to approximately 10 for Ra2+. Because
of other ions, and the rate of oxidation-reduction con- further polarization occurs on water molecules con-
versions (see Section 5.2). Hence, solubility may vary tained in the first hydration sphere, additional water
widely, depending on whether the aqueous solvent is molecules will be attracted to form a second hydra-
“pure” water or a biological fluid. In mammals, the tion sphere. The extent that hydration sphere layering
biological fluids are slightly alkaline (∼pH 7.4), with occurs decreases rapidly with distance from the ion.
the exception of the fluids in the gastrointestinal tract, Thus, the size of the hydration sphere will depend on
the polarizing power of the ion, which in turn depends Fe2 + + H2 O2 → Fe3 + + OH + OH −
on its charge radius. The hydrated ion is a dynamic
This is the iron-salt-dependent decomposition of
system in which water molecules in the hydration
dihydrogen peroxide, generating the highly reactive
sphere rapidly exchange with those in the bulk phase
hydroxyl radical, possibly by means of an oxoiron(IV)
of the solution.
intermediate. Addition of a reducing agent, such as
ascorbate, leads to a cycle that increases damage to
5.2 Redox Potential biological molecules (McNaught and Wilkinson, 1997).
Redox, or oxidation-reduction, processes involve
the transfer of electrons from one reactant to another. 5.3 Metal Ions as Lewis Acids
The two processes, oxidation and reduction, are
always coupled. This means that when one substance The categorization of metals as Lewis acids or
is reduced (oxidizing agent) another must be oxidized bases is based on the definition that a Lewis acid is an
(reducing agent). Oxidizing or reducing power of an electron-pair acceptor and that a base is an electron-
oxidation-reduction system is measured in terms of pair donor. This means that all positively charged
the standard electrode potential (i.e. the value of the metal ions can be classified as Lewis acids or elec-
standard electromotive force (emf) of a cell in which tron acceptors. In the same way, the hydration sphere
molecular hydrogen under standard pressure is oxi- water molecule that is formally an electron donor can
dized to solvated protons at the electrode). If the be classified as a Lewis base. The following equations
standard electrode potential, E0, of a metal is large illustrate this.
and negative, the metal is a powerful reducing agent
metal ion + water → hydrated metal ion
because it loses electrons easily. Examples of standard
electrode potentials of some metals of biological and
(Lewis acid + Lewis base → complex)
toxicological importance are given in Table 2. In vivo, hydrated ions + ligand → complex + water
the true reduction potentials will depend on the con-
centration of metal ions, the temperature, and the pres- The International Union of Pure and Applied Chem-
ence of other ligands that can displace water from a istry (IUPAC) compendium of chemical terminology,
hydrated ion. the Gold Book, defines these terms as follows:
Oxidation-reduction reactions are of fundamental
A Lewis acid is a molecular entity (and the cor-
importance in biology and toxicology. Transition met-
responding chemical species) that is an electron-
als with redox potentials in the biologically relevant
pair acceptor and therefore able to react with a
range, such as iron, copper, and manganese, play very
Lewis base to form a Lewis adduct, by sharing
important roles in biological processes by accepting
the electron pair furnished by the Lewis base.
and donating electrons in a catalytic fashion. These
metals are also believed to participate in Fenton reac- For example,
tion chemistry within biological systems, which results
in generating potentially damaging oxidative species Me3 B + : NH3 ↔ Me3 B − NH3 + H3
in the cell, as follows using iron as an example:
Lewis acid + Lewis base ↔ Lewis adduct
(Me, methyl; B, boron)
TABLE 2 Standard Electrode Potentials
Reaction E0 in volts at 25°C
5.4 Hydrolysis
Mn3+ + e− ↔ Mn2+ +1.500
Hydrolysis is a reaction that can occur between the
Ag+ + e− ↔ Ag +0.799
Fe3+ + e− ↔ Fe2+ +0.771 metal ion (M) and one or more water molecules in the
Cu2+ + 2e− ↔ Cu(s) +0.337 coordination (solvation) sphere, in which a proton
Cu2+ + e− ↔ Cu1+ +0.159 (hydrogen ion) is released and the solution becomes
Ni2+ + 2e− ↔ Ni(s) −0.230 acidic.
Cd2+ + 2e− ↔ Cd(s) −0.403
Cr3+ + e− ↔ Cr2+ −0.410 (n − 1) +
Zn2+ + 2e− ↔ Zn(s) −0.763 (M(OH2 )x )n + ↔ (M (OH) (OH2 )x − 1 ) + H + (aq)
Zn2+ + 2e− ↔ Zn(s) −0.763
Hydrolysis may proceed in several stages until the
Source: Skoog et al., 2007; http://www.webelements.com (accessed last coordinated water molecule is removed. The pro-
29 October, 2013). cess of hydrolysis may be interrupted if at one stage an
insoluble compound is produced. Hydrolysis occurs either part of the structure or essential cofactors for
most readily with metal ions that strongly polarize the the biological activity of many biomolecules such as
coordinated water molecules. enzymes and vitamins (Fraústo da Silva and Williams,
2001). Iron, for example, plays an essential role within
the heme moiety cofactor of cytochrome enzymes; zinc
6 OTHER ASPECTS OF METAL CHEMISTRY is an essential constituent in the enzyme alcohol dehy-
OF BIOLOGICAL AND TOXICOLOGICAL drogenase; manganese, copper, and zinc are essen-
INTEREST tial cofactors in superoxide dismutase enzymes; and
cobalt is essential to vitamin B12 (Fraústo da Silva and
6.1 Main Group and Transition Metals Williams, 2001; Hughes, 1972).
Metalloproteins (also called conjugated proteins) con-
According to the IUPAC Nomenclature of Inorganic sist of a protein and a prosthetic group or cofactor that
Chemistry (2005), the main group metals are those belong- consists of a metal. Metalloenzymes with a metal-con-
ing to the periodic system group 1 (alkali metals; Na, K, taining prosthetic group or cofactor are defined as the
Rb, Cs, Fr), group 2 (alkaline earths; Be, Mg, Ca, Sr, Ba, holoenzyme with the metal cofactor intact, and the apo-
Ra), group 13 (Al, Ga, In, Tl), group 14 (Ge, Sn, Pb), group enzyme when the metal cofactor is dissociated or lost.
15 (Sb, Bi), and group 16 (Po). The transition metals are It has been estimated that approximately a quarter
those belonging to group 3 (Sc, Y, La, Ac), group 4 (Ti, to a half of all proteins contain a metal that is essential
Zr, Hf), group 5 (V, Nb, Ta), group 6 (Cr, Mo, W), group to their biological function (Thomson and Gray, 1998;
7 (Mn, Tc, Re), group 8 (Fe, Ru, Os), group 9 (Co, Rh, Ir), Waldron and Robinson, 2009). For example, hemoglo-
group 10 (Ni, Pd, Pt), group 11 (Cu, Ag, Au), and group 12 bin, hemerythrin, and myoglobin carry oxygen bound
(Zn, Cd, Hg). to iron. The iron sulfur proteins cytochromes c and
Elements with partly filled d- or f-orbitals are usu- b5 are examples of redox electron shuttling proteins,
ally defined as transition elements. A broader defini- while cytochrome P450, catalase, and peroxidase are
tion would be elements in any oxidation state in which examples of redox enzymes. Additional examples of
they form compounds with partly filled d- or f-orbit- metal-containing proteins are shown in Table 3.
als. By this definition, Cu, Ag, and Au would also be
included. There are 56 transition elements of the d- and 6.2.1 Metalloporphyrins
f-block. All have the same common properties:
The metalloporphyrins include two important cat-
( 1) They are all metals. egories: the chlorophyll molecule and the molecules
(2) They exhibit variable oxidation states, with a few carrying the heme group. The ability of chlorophyll to
exceptions. absorb light is related to the conjugated polyene struc-
(3) Because of partly filled d- and f-orbitals, they ture of the porphyrin ring. Magnesium ions that are
form at least some paramagnetic compounds. coordinated to the nitrogen atoms of the four pyrrole
(4) Their ions and compounds are colored in one or rings have at least two functions. They provide the
all oxidation states. necessary structural rigidity and they increase the rate
of conversion of the singlet-excited state resulting from
Properties (1) and (3) are of great biological impor- photon absorption into the triplet state that enables the
tance because of their role in biological catalysis and transfer of the excitation energy into the redox chain.
their electron transport function. The two main functions of heme iron-containing
The transition elements are further subdivided into proteins are the transport of oxygen and the mediation
three main groups,

( 1) The main transition elements, or d-block elements TABLE 3 Examples of Metal-Containing Proteins
(2) The lanthanide elements Protein Metal
(3) The actinide elements

Phosphoproteins Ca
The lanthanides and actinides are classified as ele- Ceruloplasmin Cu
ments of the f-series. Superoxide dismutase Cu, Zn, or Mn
Transferrin Fe
Cytochrome P450 Fe
6.2 Metal-Containing Biological Molecules Glutamine synthetase Mn or Mg
Metallothionein Zn, Cd, Hg, Cu
Many metals, most notably iron, copper, manganese, Albumin Zn, Cd
zinc, magnesium, and cobalt, play important roles as
of electron transfer reactions. The heme group is in all hydroxyl radicals. Ascorbic acid oxidase and various
cases associated with a protein molecule, as in hemo- tyrosinases are additional examples of copper-contain-
globin, myoglobin, cytochromes, and enzymes such as ing metalloenzymes, while the molybdenum- and iron-
catalase and peroxidase. containing nitrogenases, which catalyze the formation
of ammonia from elemental dinitrogen and dihydrogen,
6.2.2 Nonheme Iron Proteins play an important role in nitrogen fixation.
Metal ions can be bound to proteins in a reversible
Nonheme iron proteins (e.g. rubredoxins, ferredox- way. This is the case with metal-activated enzymes,
ins, hemerythrin, aconitase, and high-potential iron such as enzymes associated with phosphate group
proteins) contain strongly bound functional iron atoms transfers or hydrolysis that are activated by Mg2+. Such
attached to sulfur, but they do not contain porphyrins. systems are much less amenable to study than metallo-
All of them have a role in electron transfer and, in enzymes because they cannot be easily isolated in the
the case of iron regulatory proteins (IRP1 and IRP2), holoenzyme form with the metal in place.
play a role in cellular iron homeostasis (Wang and
Pantopoulos, 2011). Ferritin and hemosiderin are
important iron-containing biological structures that 6.2.5 Metallothioneins
store iron in a protein structure. Transferrin binds fer-
Metallothionein (MT) is a low molecular mass metal-
ric iron and transports it from ferritin to cells.
binding protein of approximately 6500 Da, with a com-
position of approximately 30% thiol-containing cysteine
6.2.3 Cobalt-Containing Biological Molecules residues. Two distinct metal-binding domains, the α- and
The best-known cobalt-containing biological mol- β-clusters, have been characterized in MT and are formed
ecules are the vitamin B12 coenzymes (cobalamins). by the 61 amino acids in human MT. Most organisms
Cobalamins contain a cobalt atom, a macrocyclic possess MT-encoding genes, and both essential (e.g. zinc
ligand corrin, and a complex organic part compris- and copper) and nonessential (e.g. cadmium and mer-
ing a phosphate group, a sugar, and an organic base cury) metals readily induce the synthesis of MT. MT is
also coordinated to the cobalt atom. Methylcobalamin important in the metabolism and kinetics of cadmium
is involved in bacterial methane production and has and copper because these metals are transported by MT
been shown to transfer the methyl (CH3) group to a in the organism. Non-MT-bound cadmium is toxic and
number of metals and metalloids, including Hg(II), causes a toxic insult to the cell. MT also serves various
Te(III), Pt(II), Au(I), in vitro. It is considered the most important functions for zinc and mercury. Thus, MT pro-
likely methylator of mercury in vivo. teins have a role in the metabolism of essential metals
and in protection against the toxicity of metals. The four
6.2.4 Metalloenzymes and Metal-Activated Enzymes major groups of MTs are MT-1, -2, -3, and -4, with MT-1
existing in several isoforms. Mammalian MT-1 and MT-2
As noted above, metalloenzymes are enzymes that are present and expressed in almost all tissues. MT-3 has
require metal ions as cofactors in their normal structure seven additional amino acids (a total of 68) and is some-
and function. Carbonic anhydrase and carboxypep- what different in charge characteristics compared with
tidase are zinc-containing metalloenzymes in which MT-1 and MT-2. MT-3 was identified as a growth inhibi-
the zinc ion is bound in a distorted tetrahedral con- tory factor (GIF) in the brain. MT-4 consists of 62 amino
figuration, with two histidine nitrogen atoms and one acids, with a glutamate residue comprising the addi-
glutamate carboxyl oxygen atom from the protein and tional amino acid over MT-1 and MT-2. MT-4 is specific to
a water molecule as coordinating ligands. Delta-ami- squamous epithelium and is expressed in keratinocytes.
nolevulinic acid dehydratase (ALAD), a zinc-containing The 14 human MT genes are localized on chromosome
enzyme essential in the heme synthesis pathway, is nota- 16q13-22. Of these, six are functional, two are not, and six
ble because it is a sensitive molecular target of elevated have not been characterized (Nordberg, 1998; Nordberg
lead exposure, in which lead readily displaces the zinc and Nordberg, 2000; Dabrio et al., 2002; Lu et al., 2005).
cofactor coordinated with vicinal thiols in the enzyme’s
active site, leading to inhibition of enzyme activity.
Superoxide dismutase enzymes, which may contain 7 METALLOMICS, TOTAL ELEMENT
copper, zinc, iron, or manganese as cofactors, catalyze ANALYSIS, AND ELEMENTAL SPECIATION
the dismutation of superoxide anion to dihydrogen per-
oxide. As noted above, the latter may serve as a potential New concepts are emerging in relation to the func-
substrate for transition metal-catalyzed Fenton chem- tional role that metal species at trace and ultratrace lev-
istry reactions, leading to the production of damaging els play in biological processes such as signaling, gene
expression, and catalysis. Therefore, the chemistry of a an analyte or a group of analytes from a sample
cell needs to be characterized not only by its genome according to their physical (e.g. size, solubility) or
in the nucleus and by its protein content, the proteome, chemical (e.g. binding, reactivity) properties.

but also by the entirety of the metals and nonmetals
In reality, it is often not possible to determine the
among the different species in a biological system (the
concentrations of the different chemical species that
metallome), and the examination of the metabolism
sum up to the total concentration of an element in a
of the elemental and chemical species (metabolomics)
given sample (Templeton et al., 2000). Often, chemical
that play roles in biological and toxicological processes
species present in a given sample are not sufficiently
(Suzuki, 2005).
stable to survive sample collection, storage, process-
The metallome is a term used to describe the dis-
ing, and analysis. Individual chemical species may be
tribution and function of metal ions within biological
changed by, for example, a change in pH necessitated
systems (Williams, 2001). It follows then that the field
by the analytical procedure or by intrinsic properties of
of metallomics may be defined as the “comprehensive
the measurement methods that affect the equilibrium
analysis of the entirety of metal and metalloid species
between species.
within a cell or tissue type” (Szpunar, 2005). Within
the context of metal toxicology, these definitions may
be subdivided to emphasize the distribution, action, 8 SAMPLING AND SAMPLE PREPARATION
and analysis of toxic metals in their various chemical
forms (species) within biological systems. Indeed, the
8.1 General Considerations
adverse effects and toxicity of metals depend to a large
extent on the quantity and the chemical form (species) The exposure of humans to toxic metals may be
of the metal within the biological system. estimated from measurements of the concentrations in
The following paragraphs will cover different aspects environmental media (e.g. air, food, and water), or bio-
of the identification and quantification of metal species, logical samples (e.g. blood, hair, and urine) that have
in addition to methods used for total element determi- been shown to reflect exposures. Preliminary invento-
nations in samples appropriate for toxicology. Address- ries of sources in both occupational and environmental
ing the chemical form of the element in addition to the settings are essential to the design of adequate sam-
total trace element concentration provides complemen- pling and analysis (Kneip and Friberg, 1986). Similarly,
tary information that, together, better informs risk for some understanding of the pharmacokinetics (i.e. rates
toxicity. This is because toxicity risk may be specifically of uptake, disposition, excretion, etc.) of metals within
related to the levels of a particular chemical species of the body is needed to design a biomonitoring program
the metal, rather than the total metal concentration, and interpret the results.
while in other cases total metal levels may be highly Regardless of whether samples are environmentally
associated with well-defined health effects. Mercury or biologically based, sample collection, subsampling
and tin may be good examples of the former: the inor- for various analyses, and sample handling/storage
ganic forms of these elements are considered less toxic must satisfy several conditions. First, samples should be
compared to their alkylated forms. collected and stored in a manner that will preserve (not
The IUPAC has defined elemental speciation in alter) the elemental composition and speciation of ana-
chemistry as follows (Templeton et al., 2000). lytical interest. For elemental composition, this implies
strict and informed control of potential contamination
(1) Chemical species. Chemical element, specific form sources. For elemental/chemical speciation, this addi-
of an element defined as to isotopic composition, tionally implies that samples must be managed in a way
electronic or oxidation state, and/or complex or to avoid alteration/degradation of species. Second, the
molecular structure. sampling design should take into account the sources of
(2) Speciation analysis. Analytical chemistry, analyti- variance in the analyte(s) of interest within the sample,
cal activities of identifying and/or measuring the and the desired accuracy and precision associated with
quantities of one or more individual chemical the measurement of those analytes. Often, samples col-
species in a sample. lected for analyses are believed to accurately represent,
(3) Speciation of an element; speciation. Distribution of in terms of chemical composition and concentration,
an element among defined chemical species in a the population or range of samples that could be col-
system. lected with unlimited resources, although in practice
(4) Fractionation. When elemental speciation is not this may not be the case due to heterogeneity within
feasible or desired, the term fractionation may be the sample population. Thus, having an understanding
used to describe the process of classification of of sample population heterogeneity is important for
optimizing sampling and analytical designs. Related to specific metal and chemical species. In the case of
this, the variance within the laboratory sample should particulate-borne metals, particle size distribution and
be an unbiased estimate of the variance of the total- chemical properties such as solubility are important in
ity of the material in the collected sample, i.e. that the determining the site of deposition in the respiratory
subsamples removed for analyses accurately reflect tract and the degree of absorption. These factors have
any heterogeneity in the composition of the entire been addressed in a number of studies (Baron, 2003;
sample. This implies that when the concentration of Valavanidis et al., 2008; Moreno et al., 2009; Schmid
an elemental species has an estimated variance σ2, the et al., 2009).
calculated variance on the result given by a laboratory Environmental surveys using stationary samplers
should fall within the 95% confidence interval for the has allowed establishing the concentration of several
population σ2. metals in a number of cities and rural and remote
As a rule of thumb, the accuracy of all sampling locations (Van Dingenen et al., 2004). More relevant,
operations together should be of the same order of subject-specific, estimates can be obtained by the use
magnitude as the accuracy of the subsequent analytical of personal samplers (Lucchini et al., 2012b). The use
procedure (Sansoni and Iyengar, 1978). For example, of such devices for the determination of workplace
it is pointless to carefully determine the concentration exposures has become common practice (Schwela
of an element within a sample that was contaminated et al., 2002; WHO, 1982). An interesting survey
10-fold during collection with the element of inter- about sampling systems can be found in Dabek-Zlo-
est (e.g. contamination with lead) (Flegal and Smith, torzynska and Keppel-Jones (2003). The choice of the
1992). Similarly, it would be pointless to very carefully filter substrate is very important. Filters should be
determine the chemical speciation of an element in a mechanically and thermally stable and should not
sample that was collected and stored in a manner that interact with the deposit, even during analyses and
did not protect against species alteration/degradation. possible chemical extraction. The rule of thumb is
In both cases, and perhaps unknowingly, the results that when no data are available from reliable studies
would be meaningless. by other research groups, the effect of sampling and
This chapter does not intend to give in-depth details storage conditions on the stability of the species in
of sampling and analysis methods for every possible the matrix should be determined before large-scale
element in any feasible matrix. Such information is sample collections. Many species are thermodynam-
given in the specific chapters, metal by metal, and in ically unstable. The simple act of sampling and stor-
the peer-reviewed scientific literature. For example, ing the species may alter them. The information is
considerations for the collection and analysis of lead then irreversibly lost.
in blood have been discussed extensively (Flegal and
Smith, 1992; NRC, 1993) and, more broadly, sample 8.2.1.1 Ambient Air
collection guidelines for trace elements in blood and Metal concentrations in ambient air are generally
urine have been published by IUPAC (Cornelis et al., low, and intake through air is usually small in rela-
1995). The latter article describes harmonized guide- tion to oral intake (e.g. diet). Exceptions may occur
lines for collection, preparation, analysis, and quality in the vicinity of industrial operations emitting large
control. The aim was to assist scientists worldwide to amounts of particulate metals (e.g. smelters) and in
produce comparable data to be useful on a regional, areas with heavy vehicular traffic. In the past, the
national, and international scale. Avoidance of con- intake of lead through inhalation exceeded that from
tamination is a major issue when determining the trace food because of the use of leaded automobile fuels.
elements in body fluids and tissues (Versieck, 1985; Notably, historically deposited airborne particu-
Flegal and Smith, 1992). Informative chapters on the lates enriched in metals may also be resuspended
collection/sampling of clinical samples for trace ele- and pose respiratory exposure risks long after the
ment speciation purposes can be read in the Handbook primary environmental inputs have ceased (Harris
of Elemental Speciation (De Cremer, 2003; Muñoz Olivas and Davidson, 2005). Today, a new problem may be
and Cámara, 2003). emerging through the use of automobile exhaust
catalysts containing Pt, Pd, and Rh (platinum group
elements, or PGE) (Colombo et al., 2008). Emission
8.2 Air, Water, and Food rates of PGEs are estimated to be in the ng/km range.
The forms in which the PGEs are emitted are still
8.2.1 Air
unclear; however, a significant soluble fraction has
Metals exist in ambient or workplace air in both been measured in automobile exhausts. Analysis of
vapor and particulate forms, depending on the exhaust particles revealed the occurrence of metallic
Pt(0) attached to aluminum oxide together with a material collected on the filter [e.g. FeII (magnetite) and
small amount of Pt(IV) (Rauch and Morrison, 2005). AsIII-containing components] (Dabek-Zlotorzynska
and Keppel-Jones, 2003). Storing the samples in closed
8.2.1.2 Industrial Air polypropylene vessels under an inert atmosphere
Occupational exposure takes place mainly by inha- (nitrogen or argon) may minimize such changes
lation. Appropriate sampling techniques and accurate (Christensen et al., 1999, Dyg et al., 1994). The shorter
analysis are thus essential for evaluating the expo- the time between sample collection and analysis, the
sure in workplaces. Concentrations in industrial air better.
are usually much higher than in ambient air, which
makes it easy to collect sufficiently large amounts of 8.2.2 Water
particles for accurate measurements. Continuous per-
As noted above, sample collection procedure must
sonal samplers are the preferred method, since they
minimize contamination and yield samples represen-
best integrate the exposure over entire work shifts.
tative of the original sample population in terms of
Special attention is needed concerning sampling tech-
elemental composition and speciation (Emons, 2003).
niques and the storage of airborne metal species in the
Sampling guidelines are available from WHO (1997).
workplace (Dabek-Zlotorzynska and Keppel-Jones,
For example, in areas with municipal tap water, it may
2003). The choice of the filter media plays a prepon-
be necessary to determine metal concentrations and
derant role. General criteria that must be considered
speciation in the cold as well as in the hot water supply
in filter selection are (1) representative sampling for
because water pipes may contain certain toxic metallic
particulates of ≥ 0.3 μm; (2) low hygroscopicity, because
species that can be released in larger amounts in warm
hygroscopicity exceeding 1 mg per piece leads to seri-
water. In addition, the total time that water has been
ous errors in weight concentration measurements and
in contact with suspected contamination sources, e.g.
hence to an improper estimation of the environmental
plumbing pipes and fixtures, should also be consid-
concentration; and (3) absence of impurities that might
ered. For example, if fixtures internal to a building are
interfere with the analysis. As an example of the latter,
suspected, then water samples should be collected in
glass fiber or Teflon filters were found to be unsuitable
both a “first flush” sample first thing in the morning
for the sampling of airborne dust with low platinum
after water has been sitting in contact with the internal
content (Alt et al., 1993). Only polycarbonate and cel-
plumbing undisturbed overnight, and again after run-
lulose gave sufficiently low blank values (e.g. as low as
ning (flushing) the water for some time. The chemical
5 pg Pt per total filter).
properties of the water, like hardness and pH, should
The absence of interaction between species and fil-
also be considered since they influence the water’s
ter substrate is particularly relevant in the case of the
corrosive nature and its ability to leach metals out of
analysis of Cr(III)/(VI) in air particulate matter. Spini
the water transport system. Similarly, when water is
et al. (1994) have reported the reduction of Cr(VI)
collected from private or natural sources (e.g. well
to Cr(III) when cellulose filters were extracted with
water), the chemical composition of the water may be
an alkaline solution containing a known amount of
an important factor influencing elemental speciation
Cr(VI). The same thing was encountered by acid dis-
(Schaider et al., 2007).
solution (H2SO4) of the filters, which can be explained
by cellulose’s well-documented reducing properties.
8.2.3 Food
These results indicate that cellulose filters should not
be used for chromium speciation in airborne particu- Food sampling strategies depend on the purpose of
lates. Polycarbonate membrane filters (Scancar and the study. One strategy may focus on single food items
Milačič, 2002) and borosilicate microfiber glass disks that are analyzed for a given trace element, followed
(Christensen et al., 1999) are suitable for this type of by an estimate of the amount of the element that could
analysis. be ingested by people with different food consump-
Also crucial is maintaining sample integrity dur- tion habits, using available food consumption statis-
ing storage of particulate matter. Some changes can be tics. A second method, often referred to as a “market
anticipated [e.g. reduction of Cr(VI) because of inter- basket” strategy, is to collect representative samples
action not only with the collection substrate but also from different classes of foods (e.g. vegetables, dairy
with the air and with other compounds in the collected products, fish and meat products) in relative amounts
dust]. Erroneous results may occur because of redox similar to those that are actually consumed (as esti-
reactions. The enrichment of particles on the filter mated for a nation, a region, or population group),
gives rise to enhanced contact of the Cr species with and then analyze each class and make estimates from
gaseous species (e.g. SO2, NOX, O2, O3) and/or with that. Before analysis, each food would be treated as it
would normally be in preparation for consumption environmental exposure and health risk by compari-
(e.g. cooked, cured, or fried). Market basket surveys son with appropriate reference values. Based on this
have been conducted in the United States (Mahaffey definition, biological monitoring may be grouped
et al., 1975), Italy (Lombardi-Boccia et al., 2003), Spain into biomonitoring of exposure(s), and biomonitor-
(Urieta et al., 1996), and Japan (Maitani, 2004), among ing of health (biological) effects. Ideally, biomonitor-
other countries and regions. A third method is called ing should build upon knowledge of the probable
“duplicate diet” sampling. During a certain period, relationship(s) between ambient exposure and resul-
the people under study put meals identical to the ones tant adverse health effects (Duffus, 1993; Nordberg
they have eaten into a storage container. Trace ele- et al., 2010). The purpose of biomonitoring is to obtain
ments can be determined in the duplicate meals after an integrated estimate of the uptake of metal species
homogenization to provide a total dietary intake figure through all pathways and media of exposure. The
per subject. For details on sampling and analysis of ele- design of biomonitoring programs and subsequent
ments and their species in food, see Emons (2003) and interpretation of the data should build upon knowl-
Brereton et al. (2003). edge of the absorption, metabolism, and excretion of
Elemental speciation in addition to total ele- the metal and metal species in question, particularly in
ment determinations in nutrition has contributed cases where metal species may behave differently from
to improvements in human health risk assessments. one another.
Because this topic will be dealt with in great depth A good example is arsenic, which is typically
in the chapters dedicated to each element, only a encountered and absorbed by humans as inorganic
few examples are cited here. Detailed knowledge is arsenic. It is methylated to a large extent first to mono-
available about the different inorganic AsIII and AsV methyl arsonic acid and next to dimethylarsinic acid.
species in food, as well as about the various organic Inorganic and methylated species are the compounds
As species (Gallagher et al., 1999). There are sub- to be specifically monitored in the urine of people
stantial differences in toxicity between these com- exposed to inorganic arsenic from drinking water or
pounds, going from the extremely toxic arsine to the through inhalation, and to distinguish arsenic expo-
relatively inoffensive arsenobetaine (AsB), arseno- sure through shellfish. This will allow discrimina-
choline (AsC), and arsenosugars. A more detailed tion between arsenic uptake from potentially toxic
description can be found in Chapter 28 on arsenic. sources and that from eating fish and seafood, where
Mercury contained in foods, most notably higher tro- the element is mainly present as nontoxic AsB and
phic level marine and freshwater fishes, is largely in arsenosugars (Buchet, 2005; Francesconi, 2002). AsB
the form of organomercury compounds (e.g. meth- progresses unaltered through the gastrointestinal tract
ylmercury), which produce very different toxicity and, if absorbed, is excreted in the urine. Arsenosug-
profiles compared to inorganic mercury. For dietary ars, however, are metabolized and approximately 12
cadmium intake, approximately 70% originates from metabolites have been documented (Raml et al., 2005).
eating vegetables and meats such as liver and kid- Before starting on speciation of the arsenic species,
ney. Relevant analytical data should reflect the cad- one should be aware that this element is easily sub-
mium species in the edible parts of plants. Cadmium ject to contamination from reagents, dust, and labora-
is bound to high and low (< 5000 Da) molecular mass tory ware at the μg/L level. If contamination occurs, it
compounds (Günther and Kastenholz, 2005). In will most probably be in the form of inorganic arsenic
meat, Cd is mainly bound to MT. A third example is and not organic arsenic. Another aspect to consider is
tin. Organotin compounds (OTCs), such as tributyl the stability of the arsenic species. An extensive study
tin, are known to be very toxic to marine organisms. by Feldmann et al. (1999) on the stability of common
They originate from the use of antifouling paint and arsenic species such as arsenite (AsIII), arsenate (AsV),
are now being monitored in crustaceans (oysters) monomethylarsonic acid (MMA), dimethylarsinic acid
and fish. Eventually, they could become harmful to (DMA), and AsB in urine shows that low temperature
man (Rosenberg, 2005). conditions (4°C and −20°C) are suitable for the storage
of samples for up to 2 months. For longer periods (4-8
months), stability of the arsenic species was dependent
8.3 Biological Monitoring
on the urine matrix. Whereas the arsenic species in
Biological monitoring consists of continuous or some urine samples were stable up to 8 months at both
repeated measurement of potentially toxic substances 4°C and −20°C, other urine samples showed substan-
or their metabolites or their biochemical/molecu- tial changes in the concentrations of arsenite, arsenate,
lar effects in tissues, secreta, excreta, expired air, or DMA, and MMA. The use of additives did not improve
any combination of these to evaluate occupational or the stability of the arsenic species in urine. Moreover,
the addition of 0.1 mol/L HCl to urine samples pro- 9 SEPARATION TECHNIQUES

duced relative changes in the inorganic arsenite and
arsenate concentrations. Another research group (Jokai Analyses of metals based on their speciation
et al., 1998) investigated the effect of storage at room within an environmental or biological sample, or on
temperature and 4°C on the stability of solutions of their partitioning within different components of a
some organic and inorganic arsenic species. The results sample, have been shown to yield valuable informa-
indicated that organic arsenic species are stable during tion that richly informs our understanding of the
short-term storage at both temperatures, whereas solu- metabolism of the metal and the potential health
tions of inorganic species were only stable in refriger- risks associated with exposure. In the former, sepa-
ated conditions. ration of metal species is often required for quantita-
Another example in which determination of metal tion of individual species. Species separation is often
speciation has contributed greatly to understanding achieved by one of the following well-known tech-
health risk is the monitoring of mercury. Exposure to niques, liquid chromatography (LC), gas chromatog-
the toxic alkylated species must be discerned from raphy (GC), capillary electrophoresis (CE), and gel
that of exposure to elemental mercury or its inorganic electrophoresis (GE). The choice will be determined
salts (Horvat and Gibičar, 2005). Methylmercury may by the chemical properties of the species, the avail-
bioaccumulate to mg/kg levels in the top predators able skills and infrastructure in the laboratory, and,
of marine and freshwater food chains, making up last but not least, by the available resources. In the
90-100% of the total Hg concentration in those ani- case of the latter (i.e. partitioning of metals within a
mals. On one hand, the measurement of total mercury sample), current methodologies such as laser abla-
in those species might be sufficiently informative to tion inductively coupled plasma-mass spectrom-
establish health risk, since the vast majority of mercury etry (ICP-MS), particle-induced X-ray emission
is known to exist as methylmercury; however, in cases (PIXE) spectrometry, and extended X-ray absorption
where this is not known with certainty, it is essential fine structure spectrometry (XAFS) allow for the
to establish the speciation of mercury first. Another analysis of metals and/or metal species in samples
example of where this may be essential is in the case in situ.
of indigenous small-scale gold mining communities
in developing countries, where elemental mercury
9.1 Liquid Chromatography
is widely used in the gold amalgamation process. In
these settings, mercury exposure may occur via inhala- In LC, the sample is introduced into a chromato-
tion of elemental mercury vapor, ingestion of inorganic graphic column packed with a stationary phase, usu-
mercury, or ingestion of methylmercury from fish spe- ally a chemically modified silica or polymer, while a
cies in the contaminated watershed. Exposure to mer- liquid mobile phase is continuously pumped through
cury vapor is highly toxic because it is easily absorbed the column. The analytes interact (i.e. partition) to dif-
from the lungs into the bloodstream, from where a ferent extents with the stationary phase and the mobile
major share crosses the blood-brain barrier and even phase, which determines the length of time each ana-
the placenta barrier (see Chapter 46). lyte resides in the column, resulting in separation of
Tin is another example where the measurement of compounds. Usually the LC system is coupled to
elemental species is most relevant to assessing health a specific detector. Such a setup is perhaps the most
risk. The widespread use of organotin compounds common in elemental speciation analysis. High per-
(OTCs) has led to their entrance into various ecosys- formance liquid chromatography (HPLC) columns
tems and into the food chain. Because of their high form a widely used subset of LC, with small diameter
toxicity at even very low levels, tributyltin and triphe- (3-5 μm) particles as the stationary phase providing a
nyltin have received substantial attention. These com- substantial stationary phase surface area to facilitate
pounds, as well as the complete family of OTCs, are separation of sample components. A good overview of
very persistent in the environment and may represent its role in elemental speciation can be found in Ackley
a significant ecosystem health risk problem. Analyti- and Caruso (2003). The most common types of LC are
cal methods to monitor the very low concentrations size exclusion chromatography, ion exchange chroma-
of these compounds typically encountered in humans tography, reversed-phase chromatography, and affin-
exist, but at the moment these analyses require sub- ity chromatography. Today, it is also possible to couple
stantial preconcentration and sample cleanup. There- an LC setup to a soft ionization system, such as electro-
fore, they are time consuming, costly, and not readily spray mass spectrometry (ES-MS) (Chassaigne, 2003)
amenable to large-scale monitoring of blood or tissue and tunable plasma spectrometry (Leach et al., 2003),
samples (Rosenberg, 2005). to obtain molecular species structural information.
Figure 1 shows an example of nickel speciation analy- Ni on a Superdex peptide HR (30 cm × 10 mm) column
sis in the hyperaccumulating plant Sebertia acuminata also coupled to ICP-MS. Next, electrospray ionization
(Schaumlöffel et al., 2003). In that study, the plant extract mass spectrometry (ESI-MS) was done on the fractions
was first subject to size exclusion chromatography and corresponding to the major peak. This allowed the
nickel was measured with ICP-MS, followed by chro- identification of the nickel compound as a complex of
matography of the mass/charge fraction containing nickel with nicotinamide.
(A) 7 104
5
intensity/cps
3
58Ni
0
0 5 10 15 20 25 30
Time, min
(B)
106
210 M (I) 105 359.9 (II)
2.0
58Ni
5.0
361.9
1.5 M+2 2.5
60Ni
Intensity, cps
1.0 360
193 402
360 407
425
0.5 279
249
304
200 300 400 m/z, u 500
270 COOH COOH COOH

(C) 213
104
10 N N NH2
H
Intensity, cps
8 Nicotianamine
215
6 10 4 197 226 299 360
5 252
4 316
4 272
2
3
362
228
2 199 254 300
1 318
150 200 250 300 350 m/z, u
FIGURE 1 Ni speciation analysis in the hyperaccumulating plant Sebertia acuminata. (A) Size exclusion chromatography HPLC-ICP-MS
chromatogram of the latex extract, elution with 5 mM ammonium acetate buffer (pH 6.8) from the Superdex peptide HR (30 cm × 10 mm) column.
(B) ESI-MS spectrum of the fractions corresponding to the major peak (shaded area) in (A) inset (I), theoretical Ni isotopic pattern, inset (II), ob-
served Ni isotopic pattern. (C) ESI-MS/MS spectrum of the m/z 360 and 362 ions leads to the identification of a Ni-complex with nicotinamide,
inset, structure of the ligand. Source Schaumlöffel et al., 2003. Reproduced by permission of The Royal Society of Chemistry.
Another example of coupled chromatographic and sensitivity of ICP-MS to detect arsenic species in the
ICP-MS coupled analysis is the separation and analy- low μg/L range.
sis of arsenic compounds in urine. Sakai and Wilbur Multiple procedures are described in the literature for
(2006) report the performance of a newly developed the separation of specific elemental species or groups of
HPLC ion exchange column coupled directly to a species. As an example, approximately 100 chromato-
quadrupole ICP-MS for the simultaneous separation graphic conditions have been listed in the literature for
and quantification of inorganic and organoarsenic the separation of organotin compounds (Harrington
species in human urine (Figure 2). In this case, the et al., 1996; Cui et al., 2011; Vahčič et al., 2011; Sano et al.,
method has the distinct advantage of overcoming 2010), selenium species (Montes-Bayón et al., 2000; Lu
potentially significant interference from urine chlo- et al., 2012), arsenic species (Ali and Jain, 2004; B’Hymer
ride content, coupled with the unparalleled elemental and Caruso, 2004; Niegel et al., 2012; Sakai and
FIGURE 2 Analysis of arsenic species in urine using HPLC coupled with ICP-MS. (A) Chromatogram of a 5-μL injection of undiluted
NIES CRM No.18 urine standard. (B) A 5-μL injection of 5 μg/L As species standard. Chromatographic separation of As species used an Agilent
G3288-80000 4.6 × 250 mm ion exchange column. Reprinted from Sakai and Wilbur, 2006.
Wilbur, 2006), mercury species (Harrington, 2000; high-voltage electrical field (typically 20-30 kV) is
Leopold et al., 2010), elemental species bound to pro- applied along an open tube column with small internal
teins (Templeton, 2005; Finney et al., 2010; Rogers diameter (e.g. < 100 μm) (Michalke, 2003).
et al., 2010), elemental species bound to humic acids Because of the low sample amounts needed for CE
(Heumann, 2005), etc. Variations in the choice of the analysis, and its tremendous separation resolution
column and the elution conditions are substantial. capabilities, CE has received relatively widespread use
Enrichment and derivatization of the species have been in the biomedical and toxicological sciences.
comprehensively outlined by Bouyssiere et al. (2003). Different separation modes exist in CE: capillary
These and the many other resources available in the zone electrophoresis, with separation on the basis of
peer-reviewed scientific literature provide a wealth of the charge/mass ratio; capillary isoelectric focusing,
knowledge upon which to base a species separation and based on the isoelectric point; capillary isotachopho-
analysis strategy based on the combination of matrix, resis, based on analyte conductivity; and micellar
analyte, and the available resources of the laboratory. electrokinetic capillary chromatography, based on
hydrophobicity.
CE analysis offers high resolution and high speed,
9.2 Gas Chromatography
and it is easily adaptable for automation and quanti-
GC separates only volatile and thermally stable spe- tative analysis. It has been successfully used for the
cies (Szpunar et al., 1996). Since few metal compounds speciation of many compounds (Alvarez-Llamas et al.,
of interest naturally satisfy these criteria, samples are 2005), including metallomic profiling (Mounicou et al.,
typically chemically processed or “derivatized” within 2009), metallodrugs (Timerbaev and Sturup, 2012),
the laboratory to transform nonvolatile compounds Cr(III)/Cr(VI) (Jung et al., 1997), selenium, and arse-
into volatile, thermally stable ones (García Alonso and nic compounds (Sun et al., 2004), Se in human milk
Encinar, 2003; Snow, 2010). Liu and Lee (1999) have (Michalke, 2000), and MTs of Cu, Cd, and Zn (Pröfrock
written a comprehensive review on the chemical modi- et al., 2003).
fication or derivatization of analytes in speciation anal- This technique can be used as a primary or as a sec-
ysis, and numerous studies have been published since ondary separation technique (e.g. after HPLC); the lat-
then presenting methodologies for the analysis of met- ter case is referred to as a two-dimensional technique.
als or metal compounds using GC (e.g. Snow, 2010; Lin Taking into account the very small loading capacity of
and Whang, 2007; Ito et al., 2008; Zúñiga et al., 2009). CE, the two-dimensional approach will yield far more
Some naturally occurring volatile metal species are detailed results, bringing the high resolution of CE
dimethylmercury (Me2Hg), dimethylselenium (Me2Se), to its full potential. The system is often connected to
tetramethyltin (Me4Sn), trimethylantimony (Me3Sb), ultraviolet (UV) detection for molecular information,
trimethylbismuth (Me3Bi), methylated arsines, tetra- but also to ICP-MS or ES-MS for either elemental or
alkylated lead compounds in sewage sludge, and molecular information.
many more gases from municipal waste disposal sites.
This list is not exhaustive. Feldmann (1997), and more
9.4 Gel Electrophoresis
recently Zúñiga et al. (2009) and Peñalver et al. (2011),
among others, have reported innovative ways to extract The use of GE for separation of metal-complex spe-
and derivatize metal species into volatile species for cies relies upon the differential migration rate of the
subsequent separation by GC and mass spectrometry complexes within the gel matrix under an electric field.
or atomic emission detection. These have included The metal-macromolecular complexes can include
methods of cryogenic trapping and sequential ther- proteins, humic acids, or other samples such as DNA
mal desorption from the packed columns, solid-phase or dyes. There are practical limitations because of the
extraction, and supercritical fluid extraction meth- small amount of material that can be brought into the
ods to name a few. ICP-MS and inductively coupled gel matrix, which may limit separation and detection
plasma atomic emission spectrometry (ICP-AES) are of the species (Chéry, 2003).
very common detectors for the analysis of separated An important factor to consider during GE of metal
volatile metal species, as discussed in detail below. complexes is the stability of the metal in the complex.
GE is typically divided into native (nondenatured)
and denatured categories, which refers to whether the
9.3 Capillary Electrophoresis
sample is processed to preserve metal-complex native
The principle of separation by CE is based on dif- conformation, or treated with reducing agents and/
ferences in the electrically driven mobility of charged or detergents to denature the complex for more pre-
analytes, similar to conventional electrophoresis. A dictable separation based on the molecular size. Thus,
choice of sample processing methods and their impact by the use of ultrafiltration. It turned out that at only
on the metal-complex stability are important consid- around physiological pH is the percentage of vana-
erations. Other critical parameters are the choice of dium bound to transferrin high, whereas in extreme
electrophoresis gel composition, buffer, and pH. More acidic and basic media the binding is completely dis-
about the precautions to be taken during such separa- rupted. This behavior can be explained by the pro-
tions can be found in Section 9.5. tonation and deprotonation of some amino acids at
In cases where the metal is covalently bound to the the protein-binding site. From these results, it was
complex, such as Cu in ceruloplasmin, denaturing concluded that the addition of an acid for preserva-
conditions can be used during electrophoresis. This tion purposes (e.g. to urine) would lead to mislead-
is not the case for noncovalently bound elements, for ing results. Contrary to these findings, a study of the
which nondenaturing conditions or native electropho- vanadate-albumin complex showed that the addition
resis should be applied to prevent the loss of the basic of acid increased the binding capacity of vanadium to
structure of the complex and stripping of the metal. albumin at lower pH values. However, the addition of
Another factor that may even jeopardize the stability high amounts of salt caused a similar disruption of the
of covalently bound elements is oxidation of amino transferrin-vanadate bond. Typically, salts are added
acid residues of proteins, as has been documented in to the buffer during chromatographic techniques such
the case of selenoproteins (Chéry et al., 2001, 2005). as anion exchange or hydrophobic interaction chro-
Notably, GE methods can be combined with pow- matography to optimize the elution behavior of the
erful detection methods such as matrix-assisted laser analytes from the column. In the case of separating the
desorption ionization mass spectrometry (MALDI-MS) transferrin-vanadate complex, however, it was con-
for molecular detection or with laser ablation dynamic cluded that the use of hydrophobic interaction chro-
reaction cell inductively coupled plasma-mass spec- matography in combination with high salt levels in the
trometry (LA-DRC-ICP-MS) for elemental detection elution buffer (high salt concentration at the starting
(Chassaigne et al., 2004). Laser ablation ICP-MS is also point) is not recommended.
an attractive alternative for the in-gel analysis of met- Another example is manganese, and the challenges
als associated with separated complexes (Fan et al., associated with investigating manganese speciation
2002; Becker et al., 2009). and oxidation states in biological systems. Both the
essential biological and toxicological roles of man-
9.5 Precautionary Measures in Elemental ganese depend on its oxidation state. Manganese
can exist in the (II), (III), and (IV) oxidation states in
Speciation
biological systems, although the latter has not been
Before embarking on a separation procedure, it is found in mammalian systems. Manganese(II) exhib-
essential to investigate the stability of the trace elemen- its chemistry similar to calcium(II) (Andersson et al.,
tal species in the different media. If not, the chromato- 1997) and magnesium(II) (Vermote et al., 1992), while
grams will fail to give the original distribution of the manganese(III) is similar to iron(III) (Fraústo da
metal species and, worse, they will only yield mean- Silva and Williams, 2001; Aisen et al., 1969). While
ingless artifacts. Thus, accurate speciation analysis manganese(III) is considered a more potent prooxi-
requires that the chemical species composition of the dant, consistent with the large reduction potential of
sample be preserved during sample collection, storage, “free” Mn(III) (E0 = +1.51 V), existing evidence sug-
and analysis (De Cremer, 2003). With this in mind, it is gests that the vast majority of manganese in biologi-
necessary to define the conditions under which chemical systems exists as manganese(II) (Reaney et al., 2002;
cal species of interest remain stable before separating Günter et al., 2005, 2006). Definitive evidence elucidat-
and analyzing those species. In practice, these condi- ing the role of manganese(II) versus manganese(III)
tions should often mimic the conditions occurring in species in the toxicology of manganese has been lim-
the original sample. ited by challenges associated with accurately deter-
A well-documented example is the noncovalently mining the speciation of manganese without altering
bound vanadium (V)-transferrin complex, the prin- its oxidation state. Manganese(III) is unstable unless
cipal vanadium species in serum with a log K value well-coordinated with stabilizing ligands (Reaney
of 6.5 mole−1. This is a relatively low stability constant et al., 2002), while the stability of manganese(II) is
compared with those of Fe(III) and Al(III) with transfer- strongly affected by sample oxygen content and pH;
rin (log K(Fe3+) = 22.7 mole−1 and K(Al3+) = 12.9 mole−1). for example, manganese(II) may be readily oxidized to
The V-transferrin complex was studied at varying manganese(III) under atmospheric oxygen conditions,
pH, salt molarity, and acetonitrile concentrations (De commonly present in most sample processing and stor-
Cremer et al., 1999). All experiments were carried out age methodologies (Reaney et al., 2002). Thus, studies
investigating manganese speciation in biological sys- done versus standards in a similar matrix calibrated
tems face the challenging fact that processing samples for the element(s) of interest and undergoing the same
under typical laboratory conditions probably disturbs procedure. This method has been developed (e.g. the
the very speciation one may be hoping to study. measurement of Hg in solid samples by ETV-ICP-MS).
As mentioned in Section 9.2, derivatization (chemi- In this particular case, the prerequisite for a similar
cal modification) of a compound is commonly needed matrix calibrated for the element can be circumvented
before GC separation of analytes. When studying ele- by the use of isotope dilution mass spectrometry. The
mental species, care should be taken that the origi- direct determination of Hg in solid samples, such as
nal moiety in which the trace element occurs is not hair and tissue, was done with a 200Hg-enriched gas-
disrupted as a consequence of the derivatization. Liu eous phase for calibration based on isotope dilution
and Lee (1999) warn about a problem that is some- (Resano et al., 2005).
times encountered by using chemical modification In elemental speciation analysis, the most common
for speciation analysis because of the loss of specia- approach is to use separate separation and detection
tion information in the original sample. For example, systems that are integrated, or “hyphenated,” for real
when the fraction of free versus bound ions is of inter- time analyses. The most convenient methods for online
est or when the species and its complexes originally coupling are ICP-MS, hydride generation AAS (HG-
existing in the sample are less stable than the com- AAS), and atomic fluorescence spectrometry (AFS).
plexes formed as a result of the added complexing ICP-MS is the most versatile and sensitive multielement
agent (Olesik et al., 1995), the use of that complexing detection system, but also the most expensive. ICP-MS
agent will cause problems in identifying the original can be coupled to LC systems, GC, and CE (Houk, 2003).
species. Therefore, attention must be paid to choosing
modification methods and appropriate reagents and
in controlling operational conditions governing the 10.2 Current Methods for the Detection of
modification process. Metals
In all these endeavors a main impediment is the lack
10.2.1 Atomic Absorption Spectrometry
of suitable standards for identifying and quantifying
metal-binding biomolecules, which commonly appear The theory and practice of AAS is presented and
during toxicological investigations. discussed in many different handbooks and texts, such
as Welz and Sperling (1998) and Skoog et al. (2007),
among others. The basic principle underlying AAS is
that ground-state analyte atoms vaporized into a gas
10 DETECTION METHODS
will absorb electromagnetic radiation characteristic of
their ground-state energy levels. A light source with
10.1 General Aspects
the spectral composition of the ground-state spectral
There continues to be substantial progress made in lines of the element is used to generate nearly mono-
improving the sensitivity and specificity of the detec- chromatic light, which will be specifically absorbed by
tion methods of the commercially available instrumen- the ground-state atoms of that element. If the intensity
tation. In brief, detection methods may be grouped into of the incident light source is compared to the inten-
those that detect total element levels in a sample versity of the light after it has passed through the sample,
sus those that may selectively detect specific elemen- the relative absorption can be readily determined. For
tal species. The majority of detection methods detect example, ground-state cadmium atoms absorb electro-
total element levels in a sample, thus requiring sepa- magnetic radiation at a characteristic wavelength of
ration of selective species prior to detection. As with 228.8 nm, one of the ground-state cadmium lines, and
nearly all analytical methods, quantitation of analytes lead atoms absorb at 283.3 nm. The light source is a
in a sample relies upon comparison with an external high-intensity, ground-state line-emitting hollow cath-
calibration curve. In the case of liquid samples of dif- ode lamp of the specific element.
ficult matrices (e.g. salts or organic compounds such There are two main methods for atomization of a liq-
as proteins), it may be necessary to apply the standard uid sample in AAS, the flame (F-AAS) and the graphite
addition method. For direct analysis of solid samples, furnace (GF-AAS) AAS methods. The flame method
such as tissues, it may be feasible to apply electrother- uses different gas mixtures for creating a high temper-
mal vaporization (ETV) of the sample in a graphite ature atomization flame (e.g. air-acetylene). Different
furnace and introduction of the vaporized sample into types of flames with different oxidizing/reducing con-
the element detector, such as in ETV-atomic absorption ditions and different temperatures are used for differ-
spectrometry (AAS) and ETV-ICP-MS. Calibration is ent metals. The liquid sample is aspirated directly into
the flame. Because of the relatively low temperature of developing the technique of argon-segmented flow in
the air-acetylene flame (approximately 2300C) and the the postcolumn eluent, a substantial improvement in
relatively short residence time of the atomized analyte chromatographic resolution for the separation of these
in the light beam (e.g. < 1 sec) the method may not be four arsenic species was obtained. The LC separation,
sensitive enough to measure most elements in bio- photooxidation, HG, and AAS measurement can be
logical materials. In the graphite furnace, analyte resi- completed online within 10 min. The detection limits
dence times in the furnace/light beam are increased for MMA, DMA, AsB, and AsC in serum were 1.0, 1.3,
to 4-5 sec or more, while the furnace temperature pro- 1.5, and 1.4 μg/L of arsenic, respectively. However, with
file may be programed to volatilize interfering matrix the advancement of ICP-MS technologies coupled with
components (e.g. organics) before the analyte volatil- efficient chromatographic separation methods, arsenic
izes, and then apply very rapid temperature increases speciation is now more commonly achieved with these
up to 2500°C to rapidly volatilize sample analytes for analytical tools rather than AAS.
detection. Typically, the analytical detection limits for Cold vapor AAS is the most widely used tech-
metals by F-AAS are in the sub-μg/mL range, and sev- nique for measuring Hg species. Direct coupling of
eral orders of magnitude lower by GF-AAS (i.e. sub- solid-phase microextraction and quartz tube AAS has
ng/mL range). been used for the selective and sensitive determina-
Accurate determination of sample analyte concen- tion of methylmercury in seafood (Fragueiro et al.,
trations is often affected by nonatomic (i.e. nonspecific) 2004). It can be anticipated that these simple, rapid,
absorption due to other components in the sample and inexpensive procedures based on AAS detection
matrix. Nonspecific absorption may occur because will become more routine in laboratory settings that
of the presence of matrix atoms and molecules in the may not have the resources to dedicate to more costly
flame or the furnace that interact with the radiation at instrumentation, such as ICP-MS.
the analyte wavelength. Salts such as sodium chloride
and phosphates are especially apt to cause interfer- 10.2.2 Atomic Fluorescence Spectrometry
ence. Nonatomic absorption is typically corrected for
When species can be converted into hydrides, such
using background correction methods that rely upon
as is routinely the case for Hg, Se, As, and Sb, AFS
comparison of absorption of a broad spectrum beam
becomes a very economical elemental detection tech-
generated from a deuterium lamp versus the analyte-
nique. This method is based on measuring the inten-
specific source lamp (in F-AAS and GF-AAS), or by
sity of the specific resonance fluorescence of the atom
using the more sophisticated and powerful Zeeman
(Kirkbright and Sargent, 1974; Skoog et al., 2007). It is,
background correction (GFZ-AAS). If background cor-
however, necessary to keep in mind that conversion of
rection does not adequately compensate for matrix
the different species of an element into hydrides does
interferences, the method of additions is often used
not happen to the same extent and at the same rate,
in addition to background correction. In severe cases
as has been documented in the case of As. The con-
of matrix interference, it may be necessary to perform
version of methylated arsenic species into methylated
sample extraction to either selectively remove the inter-
hydrides gives a different response than the conver-
ferences or remove the analyte of interest into a sample
sion of inorganic arsenite or arsenate to AsH3 (Zhang
free of interferences, e.g. the metal-chelating agent
et al., 1996).
ammonium pyrrolidine dithiocarbamate extracted
into methylisobutyl ketone.
10.2.3 Atomic Emission Spectrometry
Alternatively, elements such as As, Se, Sb, and Hg
that are readily transformed into hydrides can be ana- ICP-AES, sometimes referred to as ICP-optical emis-
lyzed in complex matrices, where the volatile hydride sion spectrometry (OES), has become the most com-
metal complex is separated from the aqueous sample monly employed emission spectrometric technique.
matrix and analyzed directly. HG-AAS has been applied The argon-based plasma is compatible with aqueous
for the speciation of arsenic in the serum of persons aerosols and provides very high thermal energy for
with abnormally high serum arsenic concentrations, drying, dissociation, and atomization of the analytes.
such as dialysis patients (Zhang et al., 1996, 1998). In The temperature of a typical argon ICP is 5500-6500K,
these studies, an online method was developed for the which is high enough to destroy to a great extent the
speciation of arsenic species in human serum, includ- molecular bonds and to atomize and ionize most ele-
ing MMA, DMA, AsB, and AsC. The method is based ments, albeit to different degrees of efficiency. The rela-
on cation-exchange LC separation, UV-photooxidation tively high sensitivity of atomic emission spectrometer
for sample digestion, and continuous HG-AAS for detectors yields very high sensitivity and a very broad
the measurement of arsenic in the LC eluent. By dynamic range.
The standard configuration of an ICP includes a (Dean, 2005; Pröfrock and Prange, 2012). Typically, ion-
pneumatic nebulizer for the formation of aerosols and ized elements are drawn into the mass spectrometer
a spray chamber, which acts as a selector for aerosol- and focused using electrostatic “lenses” into the
ized droplets with a maximum cutoff diameter. When magnetic field of the mass analyzer. The resolving
aerosolized analytes enter the plasma, they are rap- power (R) of a mass spectrometer is calculated as
idly desolvated and a fraction of the atoms are tran- R = m/(|m1 − m2|) = m/Δm, where m1 is the mass of
siently excited. When the excited electrons return to one species or isotope and m2 is the mass of a second
their ground state, they emit electromagnetic radiation species or isotope it must be separated from.
characteristic of the change in energy states, which Quadrupole mass spectrometers utilize four parallel
in today’s instruments is detected by charge coupled metal rods; a radio frequency (RF) voltage is applied
device detectors capable of monitoring a large range of between a pair of rods while a DC voltage is super-
wavelengths simultaneously. Typically, multiple ana- imposed on the RF voltage. Typical quadrupole mass
lytes and multiple wavelengths per analyte are mea- spectrometers used in ICP-MS have resolutions of
sured, and one or several internal standard elements 0.7-1.0 amu (i.e. atomic mass units), which is sufficient
are mixed with the sample online during analysis to for many routine applications. There are, however,
compensate for sample-matrix interferences and fluc- many instances where this resolution is insufficient
tuations in instrument sensitivity over the course of to separate overlapping molecular or isobaric inter-
an analytical run. More information on ICP-AES can ferences from the elemental isotope of interest. For
be found in specialized handbooks (Boumans, 1987; instance, when measuring 52Cr (most abundant Cr iso-
Dean, 2005). tope θ = 83.8%), mass 52 will experience interference
ICP-AES is a multielemental analytical platform, from the isobars of 40Ar12C+, 35Cl16OH+, and 36S16O+
although in some cases of elemental speciation it may occurring in samples having a high C, Cl, S content.
be used as a single element-multispecies detector. It is Accurate quantitation of 50Cr and 54Cr suffers from
easy to couple ICP-AES online with LC because it can interference from high background counts because of
accept a continuous flow of eluent. A limitation may 36Ar14N+ and 38Ar16O+, respectively.
be the overall inefficiency of the nebulizer (i.e. the frac- To overcome these interferences, high-resolution
tion of the consumed sample that actually enters the ICP-MS is needed. High-resolution (HR)-ICP mass
plasma) and the plasma’s sensitivity to organic sol- spectrometers utilize stronger and more stable mag-
vents. The poor tolerance of the plasma source to com- netic fields to generate mass-resolving powers of up to
mon mobile phases, such as ion-pair reagents, limits 10,000. For example, the Thermo Scientific Element XR
the applicability of the technique. Moreover, the fact HR-ICP-MS is capable of resolving powers of 300, 4000,
that many ion exchange chromatography elutions are and 10,000 in low, medium, and high-resolution modes,
not isocratic (i.e. the elution is performed under vari- respectively, with a dynamic range of 1012 and a detec-
able, usually increasing, ionic strength) requires special tion power of < 1 ppq (i.e. parts per quadrillion) for
protocols to circumvent the problem of varying analyte uninterfered isotopes. In addition, sample introduction
response during the elution (Zhang and Zhang, 2003). technologies, upstream of the argon plasma, are capa-
ble of reducing interference. For example, the dynamic
10.2.4 Mass Spectrometry reaction cell allows chemical reactions in a collision cell
such that the interfering isobars are neutralized or the
10.2.4.1 Inductively Coupled Plasma-Mass Spectrometry analyte is transformed into another heavier polyatomic.
ICP-MS is a remarkably powerful technique for Besides spectral interferences, there may be numer-
trace and ultratrace element determinations. It is char- ous nonspectral interferences, which emerge as matrix-
acterized by extremely low limits of detection and a induced signal suppression or enhancement. One must
wide linear dynamic range, multielement capabil- also consider that instrument sensitivity may change
ity, and high sample throughput (Vanhaecke and over the course of analysis of a set of samples due to signal
Köllensperger, 2003; Pröfrock and Prange, 2012). This instability and/or drift, differences in sample matrices
analytical platform relies upon the ionization of ana- from one sample to another, build-up of sample-matrix
lytes in the argon plasma, and the introduction and components on the nebulizer or ICP-MS sample intro-
separation of those ions in a magnetic field based on duction cones, etc. To adjust for these issues, a carefully
their m/z ratios. The very low analytical detection lim- selected internal reference standard is typically used.
its, often in the low pg/mL range for many elements, For this, an element is selected as internal standard that
are due to the very high degree of atomization and is absent from or in low abundance in the samples and
ionization of elements in the argon plasma (∼7000K), that is of a similar mass and ionization behavior as the
and the sensitivity and stability of modern detectors analyte of interest. The internal standard is added to all
external standards, samples, blanks, etc., and the counts of organotin compounds and the oxidation states of
monitored in each sample across the analytical run. The various elements.
assumption is that changes in instrument sensitivity or
suppression/enhancement of the analyte counts will be 10.2.4.3 Glow Discharge Plasmas as Tunable Sources for
reflected in the counts of the internal standard, allow- Elemental Speciation
ing an adjustment to be made to the analyte counts in Glow discharge plasmas offer a number of possible
the samples. It is recommended to consult specialized advantages as speciation detectors for gaseous and liq-
literature because much research has already been done uid sample analysis (Marcus, 2003). The plasma works
for each specific element to define optimal conditions at sufficiently low temperatures (kinetic temperatures
and the selection of optimal internal standards. in the range of 100-500K) so as not to cause dissociation
Laser ablation (LA)-ICP-MS is emerging as an of the molecular species. The detector can be OES or
extremely powerful tool for the analysis of solid MS. The technique has been successfully applied for
samples or evaporated sample films (Koch and the speciation of organotin compounds, among other
Günther, 2011; Arora et al., 2011, 2012). LA-ICP-MS things (Marcus et al., 2011).
relies upon a pulsed laser source and beam delivery
optics, a sampling cell, and an aerosol transport line into 10.2.4.4 Spark Source Mass Spectrometry
the ICP-MS. Ideally, the laser aerosolizes a fraction of the Spark source mass spectrometry (SSMS) was the
sample into an aerosol of homogenous particle sizes that most sensitive widely used multielement technique
are carried with a carrier gas (e.g. argon) into the plasma until the 1960s. The basis for SSMS is the formation of
for degradation and ionization of analytes. Important ions when the sample is subjected to a high-energy dis-
conditions for accurate analyses are (1) an aerosol com- charge (Morrison, 1979). The apparatus uses a vacuum
position representative of the sample; (2) high transport spark in which a high-voltage RF discharge (20-100 kV)
efficiencies; and (3) complete decomposition of particles is produced between two closely spaced electrodes
that reach the ICP (Koch and Günther, 2011). Commonly, containing the material to be analyzed. This spark-
a Nd:YAG (i.e. neodymium-doped yttrium aluminum ing results in vaporization and ionization of sample
garnet; Nd:Y3Al5O12) laser source, which emits nanosec- constituents. The repetition rate and duration of the
ond (ns) pulses (5-10 ns) in the mid- and far-UV spectral RF spark source is variable to meet the various ana-
range down to 213 nm, have been most commonly used. lytical requirements. Nearly all masses are integrated
Recently, though, 193 nm ArF-type excimer laser sys- simultaneously over a period of time to provide high
tems are being used to produce relatively homogeneous sensitivity. For SSMS, the sample has to be electrically
sample aerosols in terms of mean particle size and com- conductive. This is achieved for nonconducting bio-
position; ArF-type lasers use a mixture of argon and flu- logical materials by blending them with high-purity
orine gas under high pressure, which, when electrically graphite followed by briquetting to form electrodes
stimulated, emit radiation in the UV range. that sustain the vacuum spark. Two major problems
ICP-MS is a well-suited analytical platform to pair arise in these applications, however. First, the graph-
with online HPLC, or other forms of LC (Figure 2) ite used for making the electrodes also contains trace
(Ponce de León et al., 2002; Pröfrock and Prange, 2012). element impurities, and they may exceed the concen-
Difficulties caused by the influence of the HPLC eluent trations present in the biological sample. Second, there
on the plasma, as noted in the section on ICP-AES, can may be interference caused by organic ions of differing
be anticipated and need careful consideration for ICP- complexity obtained from many possible combinations
MS as well. Alternative sample introduction methods of C, H, O, N, S, P, etc. Therefore, to take full advantage
may be considered, including solid-sampling electro- of this multielement method, samples are ashed before
thermal vaporization followed by ICP-MS detection. analysis. This will, however, cause the loss of volatile
This approach was applied for the direct determination elements such as mercury and selenium, among others.
of methylmercury and inorganic mercury in fish tissue
with nonspecific isotope dilution (Gelaude et al., 2002). 10.2.4.5 Electrospray Mass Spectrometry
This technique offers soft ionization of metal-con-
10.2.4.2 Plasma Source Time-of-Flight Mass Spectrometry taining species followed by tandem MS for the precise
Plasma source time-of-flight mass spectrometry determination of the molecular mass of the original
is a powerful tool for elemental speciation analysis species and of the individual fragments. This method
through the use of a modulated or pulsed ioniza- is ideal to obtain structural molecular information
tion source that provides both atomic and molecular about the species. There is, however, a need for sub-
fragmentation information (Leach et al., 2003). Its use stantial sample cleanup to obtain high sensitivities
has been documented, among others, for the analysis (Chassaigne, 2003).
Of note, this method is highly amenable to online other methodologies for the characterization of metal
coupling with HPLC, which has been successfully or metal species levels in aqueous samples.
applied for the speciation of organoarsenicals, sele- Spectrophotometry is still used for the rapid deter-
nium species, MTs, etc. mination of copper, iron, and of other metals such as
Cr(VI) in environmental samples. In the latter case, the
10.2.5 Electrochemical Methods selective determination of Cr(VI) is based on the for-
mation of a complex between Cr(VI) and ammonium
Electrochemical methods are based on the measure- pyrrolidinedithiocarbamate (ADPC) (Andrle and
ment of electrically based signals associated with ele- Broeckaert, 1993). A major disadvantage of the ADPC
mental or molecular properties, or interfacial processes method is that interference by other colored species
of chemical species (Town et al., 2003). Because of the such as Fe(III) or Cu(I), or species forming complexes
direct transformation of the desired chemical informa- with ADPC such as V, Mo, and Hg, are possible.
tion (concentration, activity) into an electrical signal
(potential, current, resistance, capacitance) by the meth- 10.2.7 Biosensors for Monitoring Metal Ions
ods themselves, the analytical platforms and method-
ologies are often relatively easy and inexpensive. The There has been enormous progress since the year
two major difficulties in the application of electroana- 2000 in the development of biosensors for metal ions
lytical techniques to complex real-world samples have in environmental and biological samples (Kim et al.,
been lack of selectivity and the susceptibility of the 2012; Amaro et al., 2011; Lan and Lu, 2012; Saha et al.,
electrode surface to fouling by surface-active materials 2012). Metal ion biosensors can be grouped into sev-
in the sample. However, with emerging technologies eral categories: whole-cell biosensors, enzyme- and
that continue to refine electrochemical methods, their apoenzyme-based biosensors, antibody- or protein-
performance characteristics continue to improve and based sensors, and protein-based capacitive sensors
applications in real-world settings continues to expand (Bontidean et al., 2003; Kim et al., 2012; Amaro et al.,
(Wanekaya, 2011; Yantasee et al., 2007). A variety of 2011; Lan and Lu, 2012). They are rapidly evolving and
electroanalytical techniques that differ in the mode of versatile tools for monitoring metal ions in physical and
excitation signal-response characteristics are currently biological environments, including within intact whole
being used, including potentiometry, fixed potential cells or organisms, with high sensitivity and selectivity.
methods, amperometry, various forms of voltam- Moreover, important parameters such as bioavailabil-
metry, and electrochemical detection in LC and flow- ity and cellular toxicity can only be evaluated (or best
injection analysis. These methods have been applied evaluated) using intact living cells. Whole-cell biosen-
for the quantification of various oxidation states sors use the intact functioning cell as a reporter system
of several elements [(Fe(III)/Fe(II), Cr(VI)/Cr(III), that incorporates both the metal receptor/sensor and
Tl(III)/(I), Sn(IV)/(II), Mn(IV)/(II), Sb(V)/(III), As(V)/ reporter/transducer components. For example, Tetra-
(III), and Se(VI)/(IV)]. Perhaps most exciting are the hymena thermophile protozoa containing the MTT1 and
recent development of metal sensor applications for MTT5 MT promoters linked with the eukaryotic lucifer-
the rapid, highly sensitive, and specific nanoscale ase gene as a reporter were successfully used to detect,
detection of metals and metal species in environmen- in near real time, metal levels in their environment (con-
tal and biological samples (Wanekaya, 2011; Yantasee taminated soils or water) (Amaro et al., 2011). Another
et al., 2007). notable class of metal ion biosensors is the metal-selec-
tive enzymes, such as DNAzymes. DNAzymes are a
relatively new class of metalloenzymes that have been
10.2.6 Spectrophotometry isolated based on their ability to catalyze many differ-
ent types of reactions in the presence of different metal
Spectrophotometry usually involves the forma-
ions. Because of their high metal ion selectivity, they are
tion of a complex of the trace element with a selective
being used in highly sensitive and selective fluorescent,
ligand reagent that, as a metal-ligand complex, absorbs
colorimetric, and electrochemical sensors for a wide
light in the visible and UV regions. Selective ligands
range of metal ions, including lead, uranium, and cop-
have been used for measuring many trace elements
per (Lan and Lu, 2012).
and also for distinguishing between oxidation states of
certain elements. Most of them, however, suffer from
10.2.8 Direct Measurement of Metals in Solid
interference from colored substances, from substances
Samples (Particle Characterization)
that also form complexes with the chosen ligand, or
from oxidizing-reducing agents. That said, they still The characterization of aerosol particles, including
offer relatively simple and inexpensive alternatives to their size distribution and chemical composition, is of
great importance in occupational and environmental NAA is divided into two broad categories: (1)
health monitoring to evaluate health hazards for peo- instrumental neutron activation analysis (INAA) con-
ple exposed to metal-containing dust from industrial sisting of irradiation, measurement, and evaluation of
emissions, resuspension of airborne urban particu- the spectra; and (2) radiochemical activation analysis
lates, calcination ovens, powder handling, milling, etc. (RNAA), including irradiation of the samples, fol-
(Ortner, 2003; Lucchini et al., 2012a). The procedures lowed by radiochemical separation of the elements,
often used to characterize collected particles are high- measurement of the induced radioactivity, and evalu-
resolution scanning electron microscopy (HR-SEM) ation of the spectra. In the case of INAA, advantage
and electron probe microanalysis (EPMA), although is taken of the decay properties of the isotopes so that
recently the development of total reflection X-ray the short-lived ones are measured first. As they pro-
fluorescence (XRF) and portable XRF methodologies gressively decay, the longer-lived isotopes can be mea-
(e.g. Niton XRF analyzer, Thermo Scientific) have sured. INAA is feasible for long-living isotopes, such
provided opportunities to simultaneously determine as 65Zn, 75Se, and 124Sb; RNAA requires highly skilled
the levels of a vast array of metals in collected solid technicians, is time consuming, and requires signifi-
samples, such as filter-collected airborne particulates cant infrastructure. More on NAA can be found on
and soil, etc. (Borgese et al., 2012; Zacco et al., 2009). the International Atomic Energy (IAEA) website (as of
In addition, there is the possibility of direct speciation 2012, http://www-nds.iaea.org).
of solids with XAFS, an element- and species-selective
method (Welter, 2003). The region above the edge
in the spectrum is normally divided into two subre- 11 CALIBRATION
gions: the first 50-100 eV above the edge is called the
X-ray absorption near edge structure (XANES); and The vast majority of analytical platforms and meth-
the region above the XANES region is called EXAFS odologies discussed above, with the notable exception
(Welter, 2003). Between explicit methods for particle of isotope dilution mass spectrometry (IDMS), rely
speciation, there exist valence band X-ray spectrom- on an external calibration curve to quantitate sample
etry by EPMA-wavelength-dispersive X-ray (EPMA- analyte concentrations (Heumann, 2003; Skoog et al.,
WDX) detection, transmission electron microscopy, 2007). As such, accurate calibration standards are criti-
and X-ray-induced photoelectron spectrometry cal to accurate results, assuming sample matrix issues
(Ortner, 2003). are properly dealt with. External calibration is based
on determining the relationship between analyte con-
centration and instrumental response (signal) in a
10.2.9 Neutron Activation Analysis
standard or set of standards of known analyte concen-
Neutron activation analysis (NAA) is a very sensi- tration. Because of the practical advantages of linear
tive and reliable multielement method (Ehmann and calibration functions, they are strongly favored. Linear
Vance, 1993). Today, it is only sparingly used as a con- calibration graphs can be obtained by measuring only
sequence of the somewhat limited access to nuclear a few calibration standards and, in addition, are easily
research reactors. NAA is based on the principle that described by a simple mathematical function, S = kc × b
when materials are irradiated in a nuclear reactor, or (i.e. y = mx + b), where S is the signal response of the
in another neutron source, some of the atoms are con- detection method, k the calibration factor, c the concen-
verted into radioactive isotopes. The type and energy tration of the calibrant, and b the y-axis intersect. The k
of their radiation and their decay rate are element value is the slope of the calibration line and reflects the
specific. A quantitative estimation can be made by sensitivity: the higher the k value for a given concen-
comparing the element’s radioactivity with suitable tration range, the greater the sensitivity.
standards irradiated simultaneously. Gamma spec- External calibration modes, where the sample and the
trometry with high-resolution germanium detectors corresponding calibrant are separately measured, can
coupled to multichannel analyzers has become the only be used when there is no effect (either quenching
method of choice for measurement of the induced or enhancement) by the sample matrix on the analyte
radioactivity. The standardized ko method offers signal. If significant matrix issues exist, then internal
unique possibilities because it has modeled NAA into calibration (i.e. standard addition method) should be
a very flexible procedure (Decorte et al., 1994). NAA used to compensate for the influences of matrix com-
has historically played an important role in trace ele- position on the signal intensity (Skoog et al., 1994,
ment analysis because sample processing is minimal 2007). In fact, it is generally recommended that at least
and the danger of sample contamination and loss of several samples be analyzed using the method of addi-
elements can be virtually eliminated. tions to demonstrate whether matrix interferences are
present in samples quantitated using the external stan- gives the following definitions for reference materials
dard curve. If they are, then suitable measures must (ISO, 1991).

be taken to avoid or compensate for the matrix inter-
Reference material (RM): a material or substance
ferences. For the standard additions method, three or
one or more of whose property values are suf-
more aliquots of the sample are transferred to separate
ficiently homogeneous and well established to
sample vials, and different quantities of analyte stan-
be used for the calibration of an apparatus, the
dard are added to each vial such that each vial contains
assessment of a measurement method, or for as-
the same fraction of sample, the same final volume, but
signing values to materials.
different amounts of added standard analyte (e.g. 0, 2,
Certified reference materials (CRM): a reference
4 nmol). The standard concentration versus measured
material, accompanied by a certificate; one or
signal data are then plotted (or the appropriate math-
more property values are certified by a proce-
ematical formula used) and the resulting linear regres-
dure that establishes traceability to an accurate
sion extrapolated to zero signal on the negative x-axis,
realization of the unit in which the property
which gives the concentration of the analyte in the
value is accompanied, and for which each certi-
measured sample.
fied value is accompanied by an uncertainty at a
Calibration (external or internal) for speciation anal-
stated level of confidence.
ysis requires that standards of the chemical species of
interest are certified or well characterized and stable. For determining total element and even some ele-
Whereas most of the relevant inorganic elemental spe- mental species in matrices of toxicological importance,
cies are commercially available (e.g. selenite, selenate, a relatively wide choice of CRMs is available. There
arsenite, arsenate), there is often a lack of certified are several serum, blood, urine, tissue, water, work-
organoelemental species. There is also a near total lack ing place dust, and environmental matrices issued
of isotopically labeled elemental species needed for by internationally recognized bodies. The complete
the species-specific isotope dilution technique, which inventory of these products can be found on the web
contains the elemental species to be determined in a pages of these institutes: European Virtual Institute for
different isotopic composition from that of the sample. Speciation Analysis (EVISA), IAEA, Institute for Ref-
As a result, the synthesis and characterization of the erence Materials and Measurement of the European
corresponding chemical species standard or “spike” Commission, National Institute for Standards and
must be carried out in the laboratory. The spiked Technology of the USA, National Research Council
calibrant is added to the sample before all analytical Canada, National Institute for Environmental Studies
sample processing steps. This method will allow for in Japan, and Virtual Institute for Reference Materials.
coping with any loss of analyte because the isotopic When choosing a reference material, care should be
ratio of the isotope-diluted elemental species does not taken not only to use a matrix that is the same or very
change. It also guarantees a similar response to that of similar to the matrix of the sample to be monitored
the analyte from the sample. When no spiked calibrant but also to ensure that the concentration of the ana-
is available, external calibration in ICP-IDMS may still lyte in the reference material is within the concentra-
be worthwhile using a species-unspecific standard tion range expected during the analysis of the actual
added continuously just before the detection. Such a sample. For example, a liver reference material with a
spike does not allow for any losses during the sepa- concentration of an element in the mg/kg range is not
ration/processing procedure, but it does allow correc- suitable as a reference material for analysis of blood or
tion of any variation in response of the analyte in, for urine, where the concentration of the analyte is in the
example, the eluent of the LC. Because the amount of μg/L range.
spike is constant, any variation in response of the spike
will be used to recalculate the response of the analyte
from the sample. 13 QUALITY ASSURANCE
The development and strict adherence to a compre-

12 REFERENCE MATERIALS hensive quality assurance (QA)/QC scheme is essential
for demonstrating analytical accuracy and reproduc-
The use of certified reference materials is an essen- ibility. Moreover, since the analyst is typically charged
tial tool for evaluating procedural accuracy and repro- with making multiple decisions over the course of
ducibility in a reference sample matrix that, ideally, is method development and sample analyses, it is imper-
well matched with the unknown samples. The Interna- ative that s/he is able to ensure that systematic errors
tional Organization for Standardization (ISO) Guide 30 or bias have been minimized, and that the decision
process and rationale are recorded in detail into the applied to a set of results, involves a combina-
permanent record (Prichard, 1995). This written record tion of random components and a common
should cover the several steps related to procedures systematic error or bias component.
before analysis (e.g. preanalytical QA, including sam- Trueness: the closeness of agreement between the
ple collection, avoidance of contamination or losses average value obtained from a large series of test
during sampling, and storage of the sampled material). results and an accepted reference value.
Chemical (e.g. speciation) changes of a substance after Precision: the closeness of agreement between
sampling may cause problems if the specific chemi- independent measurements obtained by apply-
cal species of a substance is to be monitored. Detailed ing the experimental procedure under stipulated
recording and treatment of data are to be included in conditions. Precision can be evaluated in many
both the preanalytical and analytical phase. different contexts, e.g. within sample, between
It should also be the policy to deliver analytical samples, within an analytical run, or across
results that are fit for purpose. This implies a thor- many analytical runs. Therefore, it is important
ough discussion with the end user of the results before to consider measures of precision across multi-
starting the analysis. It is vital to understand why the ple contexts. The smaller the random part of the
work is being done, what will happen to the results, experimental errors that affect the results, the
and what decisions will be taken, depending on the more precise the procedure.
reported numerical values. The highest level of sen- Method validation: this applies to a defined proto-
sitivity and precision may not always be required or col for the determination of a specified analyte
justified. However, the results should be precise and and range of concentrations in a particular type
accurate enough for the intended purpose. Every of test material, used for a specific purpose
result (or set of results) should be accompanied by an (Thompson et al., 2002).
estimate of the uncertainty of the result, in which the
uncertainty associated with steps in sample processing
13.2 Sources of Error
and analyses are appropriately propagated to the final
uncertainty surrounding the final result. There are multiple possible sources of error and
incorrect results in any analytical scheme, including
unsuitable methodology, contamination, interferences,
13.1 Definitions calibration errors, sampling errors, losses and degra-
dation, incompetence, or lack of care (Prichard, 1995).
The terminology and symbols of practical statisti-
All sources of error should be carefully considered and
cal methodology required to process and interpret
plans developed to minimize them before beginning
results assessed from a sample can be found in the
analysis. For example, the avoidance of contamination
ISO Standards Handbooks, “Terminology and Sym-
is crucial in trace element analysis. Potential sources
bols, Acceptance Sampling” (1995a) and “Interpreta-
of contamination may include the laboratory environ-
tion of Statistical Data and Process Control” (1995b).
ment (including the analyst), the reagents, and the
It is beyond the scope of this chapter to give a detailed
laboratory ware (Flegal and Smith, 1992). Similarly,
account of the important actions to be taken. The com-
losses of the substance to be monitored can occur as
ments about quality are therefore limited to some prac-
a consequence of adsorption to sampling equipment,
tical concepts,
volatilization, decomposition of the species, or other
QC: the planned activities designed to verify the problems.
quality of the measurement. There is no definite method that always gives accu-
QA: the planned activities designed to ensure that rate and precise results. Even so-called definitive
the quality control activities are being properly methods such as IDMS and NAA may give inaccurate
implemented. These should fully reflect the results if not properly performed. A definitive method
need of the research and user. is a method of exceptional scientific status that is suf-
Quality system: a set of procedures and responsi- ficiently accurate to stand alone in the determina-
bilities that an organization puts into place to tion of a given property. Such a method must have a
make sure that the analyst has the facilities and firm theoretical foundation so that systematic error
resources to carry out the measurements that is negligible relative to the intended use (McNaught
will meet the needs of the research or end user. and Wilkinson, 1997; Skoog et al., 2007). Erroneous
Accuracy (trueness and precision): the closeness results may be obtained even with approved methods
of agreement between a test result and the ac- if they are wrongly executed, used outside the tested
cepted reference value. The term accuracy, when calibration range, or used with matrices that were
not included in the original validation process. Many tissues published by IUPAC (Cornelis et al., 1995), the
analysts have a tendency to introduce subtle changes Codex Alimentarius Commission guidelines on the
into an approved or well-validated procedure, includ- sampling of food (2004), and the guidelines about sam-
ing changes to sample mass, reagent ratios, times pling of drinking water issued by WHO (1997). When-
and temperatures of sample processing, and recom- ever applicable, they should follow legal and statuary
mended purity of the reagents or columns. The extent requirements, such as have been issued for the sam-
to which a method may be modified without change pling of food.
in performance is often referred to as the method’s
robustness.
13.5 Statistical Considerations
It is common practice to report a result together
13.3 Results of Interlaboratory Testing
with its uncertainty (i.e. an estimate that characterizes
The need for well-defined QA/QC procedures the range of values within which the true value is esti-
is particularly evident in interlaboratory testing, in mated to lie). This is described by the standard devia-
which participants are asked to analyze samples con- tion or the standard error of the result. When analyzing
taining (for the participating laboratories) unknown a certified value in a reference material, the reported
concentrations of analytes. In interlaboratory testing value should lie between the boundaries indicated
programs, control of the actual concentration of the by the standard error on the certified value. As noted
analyte in the samples for external quality control above, it is important to appropriately propagate mea-
(EQC) should be established by reference laboratories, sures of error or uncertainty to the final result (Skoog
preferably using multiple different methods. Inter- et al., 2007).
laboratory testing schemes for routine measurements
are now very well established in different countries
and even on an international scale. New initiatives 13.6 Reporting of Quality Assurance Data
are underway for elemental speciation measurements Today all published data have to be accompanied
(EVISA, www, speciation.net). by QA data. These QA data are helpful in explaining
variations in results obtained in different studies by
methodological errors rather than by actual differ-
13.4 Elements of Quality Assurance ences in sample concentrations. Laboratories may well
A QA plan should be developed for each element/ meet acceptance criteria on one occasion and for one
elemental species and in each particular sample metal, whereas one month later or for another metal
matrix. This plan must specify in detail the differ- results may be far from the true values. It is therefore
ent internal QC (IQC) and EQC procedures that will important that QA procedures form an integral part of
be used to guarantee an accurate and sufficiently any project in which trace metal analyses take place. It
precise result. IQC is conducted by the laboratories should be up to the individual investigators to prove
themselves, whereas an outside laboratory or agency that their data are reliable. Moreover, the data should
is involved in EQC procedures. Both IQC and EQC be documented in a form that is easily understood by
should include the analysis of a number of reference the layman.
materials certified for the elements or species of inter- It is also compulsory to report the method’s detec-
est. To test the quality of the laboratory, the control tion limit and other “figures of merit” for the analyti-
samples should be analyzed with the methods used cal method (Skoog et al., 2007). The method’s detection
for routine analysis, and the CRM samples should be limit, expressed as the concentration (cL) or the quan-
treated in the same way as the other samples. If more tity (qL), it is derived from the smallest measure (xL)
care and control were to be given to these samples that can be detected with reasonable certainty with
than to the others, the results would not reflect the that analytical procedure. The value of xL is given by
quality of the routine analyses, often being more pre- the equation,
cise and accurate. xL = Xbi + ksbi
QA procedures should also exist for sampling and
–
sample handling. The sampling plan should be devel- where X bi is the mean of the blank measures, sbi is
oped in consultation with the analyst. The method the standard deviation of the blank measures, and k is
of sampling may be laid down in international and a numerical factor chosen according to the confidence
national standards or in a set of guidelines. Examples level desired, e.g. three (McNaught and Wilkinson,
are the guidelines for sampling blood, urine, and 1997; Skoog et al., 2007).
14 CONCLUSIONS Baron, B.A., 2003. Factors Affecting Aerosol Sampling, NIOSH Man-
ual of Analytical Methods, fourth edition, third supplement. Na-
tional Institute for Occupational Safety and Health, Cincinnati,
Accurate and predictable assessment of health risk OH. pp. 184–207. http://www.cdc.gov/niosh/nmam/pdfs/
from exposure to potentially toxic metals or metal spe- chapter-o.pdf (accessed 29.10.13).
cies relies heavily upon accurate exposure assessment, Becker, J.S., Lobinski, R., Becker, J.S., 2009. Metallomics 1 (4), 312–316.
which in turn relies upon accurate and precise analyses B’Hymer, C., Caruso, J.A., 2004. J. Chromatogr. A 1045, 1–13.
Bontidean, I., Csöregi, E., Schuhmann, W., 2003. Biosensors for
of metals or metal species in environmental or biologi-
monitoring of metals. In: Cornelis, R., Crews, H., Caruso, J., et al.
cal samples. There are now analytical techniques that (Eds.), Handbook of Elemental Speciation, Techniques and Meth-
allow for the accurate measurement of total element odology. John Wiley & Sons, Chichester, pp. 471–483.
concentrations in most matrices, as well as measure- Borgese, L., Salmistraro, M., Gianoncelli, A., et al., 2012. Talanta 89,
ment of a number of elemental species, even at very 99–104.
Boumans, P.W.J.M., 1987. Inductively Coupled Plasma Emission
low concentrations (e.g. sub-μg/L). Moreover, analyti-
Spectroscopy. Part 1, Methodology, Instrumentation and Perfor-
cal platforms and methodologies continue to evolve mance. John Wiley & Sons, New York.
towards greater sensitivity, selectivity, and versatility. Bouyssiere, B., Szpunar, J., Potin-Gautier, M., et al., 2003. Sample
In all cases, standardized sampling procedures and preparation techniques for elemental speciation studies. In:
well-developed QA programs for the control of ana- Cornelis, R., Crews, H., Caruso, J., et al. (Eds.), Handbook of El-
emental Speciation, Techniques and Methodology. John Wiley &
lytical results are essential. Such practices are common
Sons, Chichester, pp. 95–118.
place in most laboratories, and should be expected in Brakhnova, I.T., 1975. Environmental Hazards of Metals. Consul-
all analytical laboratories. Metal speciation analysis is tants Bureau, New York.
already well developed for many species of anthro- Brereton, P., Macarthur, R., Crews, H.M., 2003. Food, sampling with
pogenic origin, but there remains a need for further special reference to legislation, uncertainty and fitness for pur-
pose. In: Cornelis, R., Crews, H., Caruso, J., et al. (Eds.), Hand-
development, particularly in the case of metal species
book of Elemental Speciation, Techniques and Methodology.
noncovalently bound to biological ligands. The devel- John Wiley & Sons, Chichester, pp. 47–58.
opment of new technologies such as metal biosensors Buchet, J.-P., 2005. Arsenic speciation in human tissues. In: Hand-
and improvements in existing ones such as hybrid ana- book of Elemental Speciation. II. Species in the Environment,
lytical methodologies, will help fill this need, and in so Food, Medicine and Occupational Health. John Wiley & Sons,
Chichester, pp. 86–93.
doing further decipher the metallome and the highly
Chassaigne, H., 2003. Electrospray method for elemental analysis.
varied and complex role that metals play in biology In: Cornelis, R., Crews, H., Caruso, J., et al. (Eds.), Handbook of
and disease. Elemental Speciation, Techniques and Methodology. John Wiley
& Sons, Chichester, pp. 356–377.
Chassaigne, H., Chéry, C.C., Bordin, G., et al., 2004. J. Anal. At. Spec-
References trom. 19, 85–95.
Chéry, C.C., 2003. Gel electrophoresis for speciation purposes.
Ackley, K., Caruso, J.A., 2003. Liquid chromatography. In: Cornelis, In: Cornelis, R., Crews, H., Caruso, J., et al. (Eds.), Handbook of
R., Crews, H., Caruso, J., et al. (Eds.), Handbook of Elemental Elemental Speciation, Techniques and Methodology. John Wiley
Speciation, Techniques and Methodology. John Wiley & Sons, & Sons, Chichester, pp. 224–239.
Chichester, pp. 147–162. Chéry, C.C., Dumont, E., Cornelis, R., et al., 2001. Fresenius J. Anal.
Addison, A.W., Cullen, W.R., Dolphin, D., et al. (Eds.), 1977. Bio- Chem. 371, 775–781.
logical Aspects of Inorganic Chemistry. John Wiley & Sons, Chéry, C.C., Dumont, E., Moens, L., et al., 2005. J. Anal. At. Spectrom.
New York. 20, 118–120.
Aisen, P., Roland, A., Refield, A., 1969. J. Biol. Chem. 244, 4628–4633. Christensen, J.M., Byrialsen, K., Vercoutere, K., et al., 1999. Fresenius
Ali, I., Jain, C.K., 2004. Int. J. Environ. Anal. Chem. 84, 947–964. J. Anal. Chem. 363, 28–32.
Alt, F., Bambauer, A., Hoppstock, K., et al., 1993. Fresenius J. Anal. Codex Alimentarius Commission, 2004. Joint FAO/WHO Food
Chem. 346, 693–696. Standards Programme, “Report of the twenty-fifth session of the
Alvarez-Llamas, G., de la Campa, M.D., Sanz-Medel, A., 2005. codex committee on methods of analysis and sampling” ALI-
Trends Anal. Chem. 24, 28–36. NORM 04/27/23. ftp://ftp.fao.org/docrep/fao/meeting/008/
Amaro, F., Turkewitz, A.P., Martín-González, A., Gutiérrez, J.C., j2291e.pdf.
2011. Microb. Biotechnol. 4 (4), 513–522. Colombo, C., Monhemius, A.J., Plant, J.A., 2008. Ecotoxicol. Environ.
Andersson, M., Malmendal, A., Linse, S., et al., 1997. Protein Sci. 6, Saf. 71 (3), 722–730.
1139–1147. Cornelis, R., Caruso, J., Crews, H., et al. (Eds.), 2003. Handbook of
Andrle, C.M., Broeckaert, J.A.C., 1993. Fresenius J. Anal. Chem. 346, Elemental Speciation, Techniques and Methodology. John Wiley
653–658. & Sons, Chichester.
Aposhian, H.V., Aposhian, M.M., 1990. Ann. Rev. Pharmacol. Toxi- Cornelis, R., Caruso, J., Crews, H., et al. (Eds.), 2005. Handbook of
col. 30 (1), 279–306. Elemental Speciation. II. Species in the Environment, Food, Med-
Arora, M., Bradman, A., Austin, C., et al., 2012. Environ. Sci. Technol. icine and Occupational Health. John Wiley & Sons, Chichester.
46, 5118–5125. Cornelis, R., Heinzow, B., Herber, R.F.M., et al., 1995. Pure Appl.
Arora, M., Hare, D., Austin, C., et al., 2011. Sci. Total Environ. 409 Chem. 67, 1575–1608. http://www.iupac.org/publications/pac/
(7), 1315–1319. 1995/pdf/6708x1575.pdf (accessed 29.10.13).
Cotton, F.A., Wilkinson, G., Murillo, C., et al., 1999. Advanced Inor- Gelaude, I., Dams, R., Resano, M., et al., 2002. Anal. Chem. 74,
ganic Chemistry, A Comprehensive Text, Sixth ed. John Wiley & 3833–3842.
Sons Inc, Chichester, UK. Goyer, R.A., Klaassen, C.D., Waalkes, M.P., 1995. Metal Toxicology.
Cui, Z., Zhang, K., Zhou, Q., et al., 2011. Talanta 85 (2), 1028–1033. Academic Press, San Diego, New York, London, Sydney, Tokyo,
Dabek-Zlotorzynska, E., Keppel-Jones, K., 2003. “Sampling, Col- Toronto.
lection, Storage—Occupational Health. In: Cornelis, R., Crews, Günther, K., Kastenholz, B., 2005. Speciation of cadmium in the en-
H., Caruso, J., et al. (Eds.), Handbook of Elemental Speciation, vironment. In: Cornelis, R., Crews, H., Caruso, J., et al. (Eds.),
Techniques and Methodology. John Wiley & Sons, Chichester, pp. Handbook of Elemental Speciation. II. Species in the Environ-
59–72. ment, Food, Medicine and Occupational Health. John Wiley &
Dabrio, M., Rodriguez, A.R., Bordin, G., et al., 2002. J. Inorganic Sons, Chichester, pp. 94–106.
Biochem. 88 (2), 123–134. Günter, K.K., Aschner, M., Miller, L.M., 2006. Neurobiol. Aging.
Dean, J.R., 2005. Practical Inductively Coupled Plasma Spectroscopy. 27 (12), 1816–1826.
John Wiley & Sons, Chichester. Günter, K.K., Aschner, M., Miller, L.M., et al., 2005. Free Radic. Biol.
Decorte, F., Simonits, A., Lin, X.L., et al., 1994. Biol. Trace Elem. Res. Med. 39 (2), 164–181.
43, 19–31. Gwiazda, R., Campbell, C., Smith, D.R., 2005. Environ. Health Per-
De Cremer, K., 2003. Sampling of clinical samples, collection and spect. 113, 104–110.
storage. In: Cornelis, R., Crews, H., Caruso, J., -Heumann, K.G. Gwiazda, R., Smith, D.R., 2000. Environ. Health Perspect. 108,
(Eds.), Handbook of Elemental Speciation, Techniques and Meth- 1091–1097.
odology. John Wiley & Sons, Chichester, pp. 23–46. Gwiazda, R., Squibb, K., McDiarmid, M., et al., 2004. Health Phy. 86
De Cremer, K., De Kimpe, J., Cornelis, R., 1999. Fresenius J. Anal. (1), 12–18.
Chem. 363, 519–522. Harrington, C.F., 2000. Trends Anal. Chem. 19, 167–179.
Dessy, R., Dillard, J., Taylor, L., 1971. Bioinorganic Chemistry. Ad- Harrington, C.F., Eigendorf, G.K., Cullen, W.R., 1996. Appl. Or-
vances in Chemistry Series, vol. 100, American Chemical Society, ganomet. Chem. 10, 339–362.
Washington, DC. Harris, A.R., Davidson, C.I., 2005. Environ. Sci. Technol. 39, 7410–7415.
Duffus, J.H., 1993. Pure Appl. Chem. 65, 2003–2122. http://pac.iu Heslop, R.B., Jones, K., 1976. Inorganic Chemistry. A Guide to
pac.org/publications/pac/pdf/1993/pdf/6509x2003.pdf (ac- Advanced Study. 2. Elsevier, Amsterdam.
cessed 10.12.13). Heumann, K.G., 2003. Calibration in elemental speciation analysis.
Duffus, J.H., 2002. Pure Appl. Chem. 74, 793–807. http://www.iup In: Cornelis, R., Crews, H., Caruso, J., et al. (Eds.), Handbook of
ac.org/publications/pac/2002/7405/7405x0793.html (accessed Elemental Speciation, Techniques and Methodology. John Wiley
29.10.13). & Sons, Chichester, pp. 547–562.
Duffus, J.H., Nordberg, M., Templeton, D.M., 2007. Pure Appl. Heumann, K.G., 2005. Metal complexes of humic substances. In:
Chem. 79, 1153–1341. Cornelis, R., Crews, H., Caruso, J., et al. (Eds.), Handbook of
Dyg, S., Cornelis, R., Griepink, B., et al., 1994. Anal. Chim. Acta 286, Elemental Speciation, Techniques and Methodology. John Wiley
297–308. & Sons, Chichester, pp. 621–637.
Ehmann, W.D., Vance, D.E., 1993. Radiochemistry and Nuclear Hill, C.H., Matrone, G., 1970. Fed. Proc. 29, 1474–1481.
Methods of Analysis. John Wiley & Sons, Chichester. Horvat, M., Gibičar, D., 2005. Speciation of mercury, environment,
Emons, E., 2003. Sampling, collection, processing and storage of en- food, clinical, and occupational health. In: Cornelis, R., Crews,
vironmental samples. In: Cornelis, R., Crews, H., Caruso, J., et al. H., Caruso, J., et al. (Eds.), Handbook of Elemental Speciation,
(Eds.), Handbook of Elemental Speciation, Techniques and Meth- Techniques and Methodology. John Wiley & Sons, Chichester,
odology. John Wiley & Sons, Chichester, pp. 7–22. pp. 281–304.
Fan, T., Pruszkowski, E., Shuttleworth, S., 2002. J. Anal. At. Spec- Houk, R.S., 2003. Elemental speciation by inductively coupled plasma-
trom. 17, 1621–1623. mass spectrometry with high resolution instruments. In: Corne-
Feldmann, J., 1997. J. Anal. At. Spectrom. 12, 1069–1076. lis, R., Crews, H., Caruso, J., et al. (Eds.), Handbook of Elemental
Feldmann, J., Lai, V.W.M., Cullen, W.R., et al., 1999. Clin. Chem. 45, Speciation, Techniques and Methodology. John Wiley & Sons,
1988–1997. Chichester, pp. 378–416.
Fiabane, A.M., Williams, D.R., 1977. The Principles of Bioinorganic Hughes, M.N., 1972. The Inorganic Chemistry of Biological Processes.
Chemistry. The Chemical Society, London. John Wiley & Sons, New York.
Finney, L., Chishti, Y., Khare, T., et al., 2010. ACS Chem. Biol. 5 (6), IAEA (International Atomic Energy), 2004. Analytical applications of
577–587. nuclear techniques. http://www-pub.iaea.org/MTCD/publicat
Flegal, A.R., Smith, D.R., 1992. Environ. Res. 58, 125–133. ions/PDF/Pub1181_web.pdf (accessed 29.10.13).
Fragueiro, S., Lavilla, I., Bendicho, C., 2004. J. Anal. At. Spectrom. ISO (International Organisation for Standardization) Guide 30, 1991.
19, 250–254. second ed. Terms and Definitions Used in Connection with Ref-
Francesconi, K.A., 2002. Appl. Organomet. Chem. 16, 437–445. erence Materials, ISO/REMCO Resolution 1.091, International
Fraústo da Silva, J.J.R., Williams, R.J.P., 1991. The Biological Chem- Standardization Organisation, Geneva, Switzerland.
istry of the Elements—The Inorganic Chemistry of Life. Oxford ISO, 1995a. Fourth ed. Statistical Methods for Quality Control, vol.
University Press, Oxford, England. 1, Terminology and Symbols Acceptance Sampling., Geneva,
Fraústo da Silva, J.J.R.F., Williams, R.J.P., 2001. The Biological Chem- Switzerland.
istry of the Elements, The inorganic chemistry of Life, second ed. ISO, 1995b. fourth ed. Handbook. Statistical Methods for Quality
Oxford University Press, Oxford, England. Control, vol. 2, Measurement Methods and Results. Interpreta-
Gallagher, P.A., Wei, X., Shoemaker, J.A., et al., 1999. J. Anal. At. tion of Statistical Data Process Control, Geneva, Switzerland.
Spectrom. 14, 1829–1834. Ito, R., Kawaguchi, M., Sakui, N., et al., 2008. J. Chromatogr. A 1209
García Alonso, J.I., Encinar, J.R., 2003. Gas chromatography and (1-2), 267–270.
other gas based methods. In: Cornelis, R., Crews, H., Caruso, J., IUPAC (International Union of Pure and Applied Chemistry). Gold
et al. (Eds.), Handbook of Elemental Speciation, Techniques and Book. Version 2.3.2, 2012-08-19. http://goldbook.iupac.org/index.
Methodology. John Wiley & Sons, Chichester, pp. 163–200. html (accessed 28.10.13).
IUPAC Nomenclature of Inorganic Chemistry, 2005. http://old.iupac. Nordberg, G., 1976. Permanent Commission and International As-
org/publications/books/rbook/Red_Book_2005.pdf (accessed sociation on Occupational Health. Subcommittee on the Toxicol-
28.10.13). ogy of Metals. Effects and Dose-response Relationships of Toxic
Jokai, Z., Hegoczki, J., Fodor, P., 1998. Microchem. J. 59, 117–124. Metals. In: Proceedings from an International Meeting Organized
Jung, G.Y., Kim, Y.S., Lim, H.B., 1997. Anal. Sci. 13, 463–467. by the Subcommittee on the Toxicology of Metals of the Perma-
Kim, H.N., Ren, W.X., Kim, J.S., et al., 2012. Chem. Soc. Rev. 41 (8), nent Commission and International Association on Occupational
3210–3244. Health, Tokyo, November 18-23, 1974. Elsevier Scientific Pub. Co,
Kirkbright, G.F., Sargent, M., 1974. Atomic Absorption and Fluo- Amsterdam, New York.
rescence Spectroscopy. Academic Press, London, New York, San Nordberg, M., 1998. Talanta 46, 243–254.
Francisco. Nordberg, M., Duffus, J.H., Templeton, D.M., 2010. Pure Appl.
Kneip, T.J., Friberg, L., 1986. Sampling and analytical methods. In: Chem. 82 (3), 679–751.
Friberg, L., Nordberg, G.F., Vouk, V. (Eds.), Handbook on the Nordberg, M., Nordberg, G.F., 2000. Cell. Molec. Biol. 46 (2), 451–463.
Toxicology of Metals, second ed. Elsevier Science Publishers, NRC (National Research Council), 1993. Measuring Lead Exposure
Amsterdam, New York, Oxford, pp. 36–67. in Infants, Children, and Other Sensitive Populations. National
Koch, J., Günther, D., 2011. Appl. Spectroscopy 65 (5), 155A–162A. Academy Press, Washington D.C.
Lan, T., Lu, Y., 2012. Met. Ions Life Sci. 10, 217–248. Olesik, J.W., Kinzer, J.A., Olesik, S.V., 1995. Anal. Chem. 67, 1–12.
Leach, A.M., McClenathan, D.M., Hieftje, G.M., 2003. Plasma source Ortner, H., 2003. Characterization of individual aerosol particles
time-of-flight mass spectrometry, A Powerful tool for elemental with special reference to speciation techniques. In: Cornelis, R.,
speciation. In: Cornelis, R., Crews, H., Caruso, J., et al. (Eds.), Crews, H., Caruso, J., et al. (Eds.), Handbook of Elemental Spe-
Handbook of Elemental Speciation, Techniques and Methodol- ciation. Techniques and Methodology. John Wiley & Sons, Chich-
ogy. John Wiley & Sons, Chichester, pp. 313–333. ester, pp. 505–525.
Leopold, K., Foulkes, M., Worsfold, P., 2010. Anal. Chim. Acta 663 Parish, R.V., 1977. The Metallic Elements. Longman, London.
(2), 127–138. Pass, G., 1973. Ions in Solutions—Inorganic Properties. Oxford Uni-
Lin, M.Y., Whang, C.W., 2007. J. Chromatogr. A 1160 (1-2), 336–339. versity Press, Oxford, England.
Liu, W., Lee, H.K., 1999. J. Chromatogr. A 834, 45–63. Pauling, L., 1970. General Chemistry. W.H. Freeman and Co, San
Lombardi-Boccia, G., Aguzzi, A., Cappelloni, M., et al., 2003. Br. J. Francisco.
Nutr. 90, 1117–1121. Peñalver, R., Campillo, N., Hernández-Córdoba, M., 2011. Talanta
Lu, J., Jin, T., Nordberg, G.F., et al., 2005. Toxicol. Appl. Pharmacol. 87, 268–275.
206 (2), 150–156. Phipps, D.A., 1976. Metals and Metabolism. Clarendon Press, Oxford.
Lu, Y., Rumpler, A., Francesconi, K.A., et al., 2012. Anal. Chim. Acta. Ponce de León, C.A., Montes-Bayón, M., Caruso, J.A., 2002. J. Chro-
731, 49–59. matogr. A 974, 1–21.
Lucchini, R.G., Dorman, D.C., Elder, A., et al., 2012a. Neurotoxicol- Prichard, F.E., 1995. Quality in the Analytical Chemistry Laboratory.
ogy 33 (4), 838–841. John Wiley & Sons, Chichester.
Lucchini, R.G., Guazzetti, S., Zoni, S., et al., 2012b. Neurotoxicology Pröfrock, D., Leonhard, P., Prange, A., 2003. Anal. Bioanal. Chem.
33 (4), 687–696. 377, 132–139.
Mahaffey, K.R., Corneliussen, P.E., Jelinek, C.F., et al., 1975. Environ. Pröfrock, D., Prange, A., 2012. Appl. Spectrosc. 66 (8), 843–868.
Health Perspect. 12, 63–69. Raml, R., Goessler, W., Traar, P., et al., 2005. Chem. Res. Toxicol. 18,
Maitani, T., 2004. J. Health Sci. 50, 205–209. 1444–1450.
Marcus, R.K., 2003. Glow discharge plasmas, Tunable sources for Rauch, S., Morrison, M., 2005. Importance of platinum group ele-
elemental speciation. In: Cornelis, R., Crews, H., Caruso, J., et al. ments and gold speciation in the environment and medicine. In:
(Eds.), Handbook of Elemental Speciation, Techniques and Meth- Cornelis, R., Crews, H., Caruso, J., et al. (Eds.), Handbook of El-
odology. John Wiley & Sons, Chichester, pp. 334–356. emental Speciation II, Species in the Environment, Food, Medi-
Marcus, R.K., Quarles Jr, C.D., Barinaga, C.J., et al., 2011. Anal. Chem. cine and Occupational Health. John Wiley & Sons, Chichester,
83 (7), 2425–2429. pp. 328–337.
McNaught, A.D., Wilkinson, A., 1997. second ed. IUPAC Compen- Reaney, S., Kwik-Uribe, C., Smith, D., 2002. Chem. Res. Toxicol. 15,
dium of Chemical Terminology. http://goldbook.iupac.org/ (ac- 1119–1126.
cessed 29.10.13). Resano, M., Gelaude, I., Dams, R., et al., 2005. Spectrochim. Acta,
Michalke, B., 2000. Spectroscopy 15, 30–39. Part B 60, 319–326.
Michalke, B., 2003. Capillary electrophoresis in speciation analysis. Rogers, D.A., Ray, S.J., Hieftje, G.M., 2010. Metallomics 2 (4),
In: Cornelis, R., Crews, H., Caruso, J., et al. (Eds.), Handbook of 280–288.
Elemental Speciation, Techniques and -Methodology. John Wiley Rosenberg, E., 2005. Speciation of tin. In: Cornelis, R., Crews, H.,
& Sons, Chichester, pp. 201–223. Caruso, J., et al. (Eds.), Handbook of Elemental Speciation II.,
Montes-Bayón, M., DeNicola, K., Caruso, J.A., 2000. J. Chromatogr. Species in the Environment, Food, Medicine and Occupational
A 1000, 457–476. Health. John Wiley & Sons, Chichester, pp. 422–463.
Moreno, T., Querol, X., Alastuey, A., Gibbons, W., 2009. Biomarkers Saha, K., Agasti, S.S., Kim, C., et al., 2012. Chem. Rev. 112 (5), 2739–2779.
14 (Suppl. 1), 17–22. Sakai, T., Wilbur, S., 2006. Application report. http://www.agilent.
Morrison, G.H., 1979. CRC Crit. Rev. Anal. Chem. 8, 287–320. com/chem.
Mounicou, S., Szpunar, J., Lobinski, R., 2009. Chem. Soc. Rev. 38 (4), Sano, T., Takagi, H., Nagano, K., et al., 2010. J Chromatogr. A
1119–1138. 1217 (26), 4344–4346.
Muñoz Olivas, R., Cámara, C., 2003. Sample treatment for speciation Sansoni, B., Iyengar, V., 1978. Sampling and Sample Preparation
analysis in biological samples. In: Cornelis, R., Crews, H., Caru- Methods for the Analysis of Trace Elements in Biological Materi-
so, J., et al. (Eds.), Handbook of Elemental Speciation, Techniques als. IAEA technical report, IAEA-SM-227/107.
and Methodology. John Wiley & Sons, Chichester, pp. 73–94. Scancar, J., Milačič, R., 2002. Analyst 127, 629–633.
Niegel, C., Pfeiffer, S.A., Grundmann, M., et al., 2012. Analyst Schaider, L., Senn, D., Brabander, D., et al., 2007. Environ. Sci. Tech-
137 (8), 1956–1962. nol. 41 (11), 4164–4171.
Schaumlöffel, D., Ouerdane, L., Bouyssiere, B., et al., 2003. J. Anal. Vanhaecke, F., Köllensperger, G., 2003. Detection by ICP-mass spec-
At. Spectrom. 18, 120–127. trometry. In: Cornelis, R., Crews, H., Caruso, J., et al. (Eds.),
Schmid, O., Möller, W., Semmler-Behnke, M., et al., 2009. Biomarkers Handbook of Elemental Speciation, Techniques and Methodol-
Suppl 1, 67–73. ogy. John Wiley & Sons, Chichester, pp. 281–312.
Schwela, D., Morawska, L., Kotzias, D., 2002. Guidelines for Concen- Vermote, C.L.M., Vipond, I.B., Halford, S.E., 1992. Biochemistry 31,
tration and Exposure-Response Measurement of Fine and Ultra 6089–6097.
Fine Particulate Matter for Use in Epidemiological Studies. EUR Versieck, J., 1985. CRC Crit. Rev. Clin. Lab. Sci. 22, 97–184.
20238 EN. http://whqlibdoc.who.int/hq/2002/a76621.pdf Waldron, K.J., Robinson, N.J., 2009. Nat. Rev. Microbiol. 7 (1), 25–35.
(accessed 29.10.13). Wanekaya, A.K., 2011. Analyst 136 (21), 4383–4391.
Skoog, D.A., Holler, F.J., Crouch, S.R., 2007. sixth ed. Principles of Instru- Wang, J., Pantopoulos, K., 2011. Biochem. J. 434 (3), 365–381.
mental AnalysisThomson Brooks/Cole Publishing, Belmont, CA. Welter, E., 2003. Direct speciation of solids, X-ray absorption fine
Skoog, D.A., West, D.M., Holler, F.J., 1994. Analytical Chemistry—An structure spectroscopy for species analysis in solid samples. In:
IntroductionSaunders College Publishing, Philadelphia, PA. 431. Cornelis, R., Crews, H., Caruso, J., et al. (Eds.), Handbook of El-
Snow, N.H., 2010. Adv. Chromatogr. 48, 373–388. emental Speciation, Techniques and Methodology. John Wiley &
Spini, G., Profumo, A., Riolo, C., et al., 1994. Toxicol. Environ. Chem. Sons, Chichester, pp. 526–545.
41, 209–219. Welz, B., Sperling, M., 1998. “Atomic Absorption Spectrometry.”
Sun, B.G., Macka, M., Haddad, P.R., 2004. J. Chromatogr. A 1039, 3rd completely revised edition. John Wiley & Sons, Ltd.,
201–208. New York.
Suzuki, K.T., 2005. Anal. Chim. Acta 540, 71–76. Williams, D.R. (Ed.), 1976. An Introduction to Bioinorganic Chemis-
Szpunar, J., 2005. Analyst 130 (4), 442–465. try. Charles Thomas Publishers, Springfield, Illinois.
Szpunar, J., Schmitt, V.O., Lobinski, R., et al., 1996. J. Anal. At. Williams, R.J.P., 2001. Coordination Chem. Rev. 216-217, 583–595.
Spectrom. 11, 193–199. Williams, R.J.P., Fraústo da Silva, J.J.R., 1996. The Natural Selection
Task Group on Metal Accumulation, 1973. Environ. Physiol. Bio- of the Chemical Elements. Clarendon Press, Oxford.
chem. 3, 65. WHO (World Health Organization), 1997. Guidelines for drinking-
Templeton, D.M., 2005. Selected examples of important metal-pro- water quality, second ed. Surveillance and Control of Com-
tein species. In: Cornelis, R., Crews, H., Caruso, J., et al. (Eds.), munity Supplies, vol. 3, WHO, Geneva, Switzerland. Chapter
Handbook of Elemental Speciation II., Species in the Environ- 4, Water Sampling and Analysis, pp. 51–72. http://www.who.i
ment, Food, Medicine and Occupational Health. John Wiley & nt/water_sanitation_health/dwq/en/2edvol3d.pdf (accessed
Sons, Chichester, pp. 638–649. 29.10.13).
Templeton, D.M., Ariese, F., Cornelis, R., et al., 2000. Pure Appl. WHO, 1982. Estimating human exposure to air pollutants. WHO off-
Chem. 72, 1453–1470. http://www.iupac.org/publications/pac/ set publication. No.69. http://whqlibdoc.who.int/offset/WHO_
2000/7208/7208pdfs/7208templeton_1453.pdf. OFFSET_69.pdf. (accessed 29.10.13).
Thompson, M., Ellison, S.L.R., Wood, R., 2002. Pure Appl. Chem. 74, WHO/IPCS, 2006. Environmental Health Criteria 234. Elemental
835–855. http,//www.iupac.org/publications/pac/2002/pdf/ speciation in Human Health Risk Assessment, Geneva, Switzer-
7405x0835.pdf (accessed 29.10.13). land.
Thomson, A.J., Gray, H.B., 1998. Curr. Opin. Chem. Biol. 2, 155–158. Yantasee, W., Lin, Y., Hongsirikarn, K., et al., 2007. Environ. Health.
Timerbaev, A., Sturup, S., 2012. Curr. Drug Metab. 13 (3), 272–283. Perspect. 115 (12), 1683–1690.
Town, R.M., Emons, H., Buffle, J., 2003. Speciation analysis by elec- Zacco, A., Resola, S., Lucchini, R., et al., 2009. J. Environ. Monit.
trochemical methods. In: Cornelis, R., Crews, H., Caruso, J., et al. 11 (9), 1579–1585.
(Eds.), Handbook of Elemental Speciation, Techniques and Meth- Zhang, X., Cornelis, R., De Kimpe, J., et al., 1996. Anal. Chim. Acta.
odology. John Wiley & Sons, Chichester, pp. 427–460. 319, 177–185.
Urieta, I., Jalon, M., Eguileor, I., 1996. Food Addit. Contam. 13, Zhang, X.R., Cornelis, R., De Kimpe, J., et al., 1998. Clin. Chem. 44,
29–52. 141–147.
Vahčič, M., Milačič, R., Sčančar, J., 2011. Anal. Chim. Acta. 694 (1-2), Zhang, X., Zhang, C., 2003. Atomic absorption and atomic emission
21–30. spectrometry. In: Cornelis, R., Crews, H., Caruso, J., et al. (Eds.),
Valavanidis, A., Fiotakis, K., Vlachogianni, T., 2008. J. Environ. Sci. Handbook of Elemental Speciation, Techniques and Methodol-
Health C Environ. Carcinog. Ecotoxicol. Rev. 26 (4), 339–362. ogy. John Wiley & Sons, Chichester, pp. 241–260.
Van Dingenen, R., Raes, F., Putaud, J.P., et al., 2004. Atmos. Environ. Zúñiga, M.C., Jover, E., Arancibia, V., et al., 2009. Talanta 80 (2),
38, 2561–2577. 504–510.

General Chemistry, Sampling, Analytical Methods, and Speciation

Uploaded by

Copyright:

Available Formats

General Chemistry, Sampling, Analytical Methods, and Speciation

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

General Chemistry, Sampling, Analytical Methods, and Speciation

Uploaded by

Copyright:

Available Formats

C H A P T E R

General Chemistry, Sampling, Analytical

ABSTRACT The commonly used term “heavy metals” to describe

Handbook on the Toxicology of Metals 4E

TABLE 1 The Periodic Table1

the addition of 0.1 mol/L HCl to urine samples pro- 9 SEPARATION TECHNIQUES

200 300 400 m/z, u 500

270 COOH COOH COOH

150 200 250 300 350 m/z, u

The development and strict adherence to a compre-

You might also like