Electromagnetic radiation, of which solar flux is composed, encompasses visible light but also extends far beyond it to shorter and to longer wavelengths than our eyes can respond to. When light is passed through a prism it is resolved into components, as is well known from studies of elementary physics or, equivalently, from observation of a rainbow.
The light so resolved is referred to as a spectrum. The interaction of electromagnetic radiation with matter is a major part of how nature itself works. Where by human ingenuity the interaction of radiation with matter is used as a means of analysing the ‘matter’, the term spectroscopy applies. The term should be understood as a generic one, as there are very many forms of spectroscopy and electromagnetic radiation used in spectroscopic techniques ranges from radio waves on the long-wavelength side to X-rays on the shortwavelength side. The etymological link between ‘spectrum’ and ‘spectroscopy’ should be noted.
The read-out from a spectroscopic measurement is usually referred to as a ‘spectrum’, though here it does not have precisely the same meaning as that used previously to describe light resolved by a prism. ‘Spectra’ in the second sense can be classified into one of two types: emission spectra and absorption spectra. The terms are fairly self-explanatory: in an emission spectrum the matter is releasing radiation and in an absorption spectrum it is receiving it.
The type of spectroscopy chosen for further discussion in this article is atomic absorption spectroscopy. The choice has been made for two reasons. First, atomic absorption in its several forms finds many environmental applications. Secondly, and perhaps more importantly, it combines the emission and absorption spectra principles. It is called atomic absorption because the basis of the measurement is radiation absorbed by particular a chemical element, but the radiation absorbed had previously come from an emission spectrum of atoms from the same element.
The level of originality in the development of atomic absorption spectroscopy was therefore very high indeed and both the intrinsic scientific content and the scope for application make it one of the most important advances in analytical chemistry ever.
Background on AA
About 25 years ago I worked in a government research establishment in Melbourne. I soon observed on the wall of one of the laboratories a group photograph taken a few years earlier marking a visit by Sir Alan Walsh FRS. Sir Alan was the developer of the very widely used atomic absorption (AA) method. He had spent his formative years in the UK and had gone to live in Australia in early adult life. He was well established in Australia by the time he developed AA, which, therefore, is reasonably seen by many as an Australian contribution. Sir Alan died, at a very good age, in 1998.
The element of interest is got into atomic form, usually in a flame. Light from the same element, having been produced in a hollow cathode tube, is directed at the flame, as noted in the previous section. So, for example, in analysing for copper in the flame one would use a hollow cathode tube containing copper. Light entering the flame is absorbed by the target element, and this is the basis of the signal. Most metal and ‘metalloid’ elements can be analysed for in this way. In many cases more than one spectral line is produced by the element in the cathode tube, so there is a choice of which one is selected for absorption measurement by the corresponding atoms in the flame.
Different detection limits apply according to which spectral line is chosen. A notable example of this is arsenic.
Variants on the basic AA technique include the following:
- Where a graphite furnace is used instead of a flame to atomise the element of interest, the term electrothermal atomic absorption (ETAA) applies
- The well known Zeeman effect can be used to resolve the signal into a number of components by application of a magnetic field, in which case the term Zeeman Atomic Absorption (ZAA) applies
- The element being analysed for can initially be taken up into solution, atomic vapour above the solution developing as a result of reduction of positive ions to atoms by a suitable reagent in the solution. The vapour enters a transparent cell at room temperature where it receives and responds to light received; for example with mercury a spectral line at 253.6 nm will be used. This is called cold vapour atomic absorption (CVAA)
In discussions of atomic absorption the acronym AA therefore tends to mean flame atomic absorption in contrast to the extensions of the method described above. However, the abbreviation FAAS – flame atomic absorption spectrometry (or spectroscopy) – is sometimes used to denote AA in this its simplest form in which case it is probably reasonable to use ‘AA’ as a generic term for all of the techniques involving atomic absorption.
Examples of the application of atomic absorption to atmospheric contaminants
Seven metals are discussed below in terms of their presence in the atmosphere and analysis by atomic absorption methods. The US Environmental Protection Agency identifies twenty-two metals as air pollutants requiring monitoring: these include the seven chosen for the discussion that follows:
Mercury in air arises largely from the burning of coals that contain trace amounts of mercury. The ambient level of mercury recommended by the World Health Organisation 5 ng m-3 but in most countries, even those where industrial hygiene is not of the highest standard, ambient levels are only 1 to 2 ng m-3. Close to a releaser such as a power station, levels of the order of a microgram per m3 are expected and these can readily be measured with atomic absorption devices of various types. ZAA is often preferred for ambient level measurement distant from a releaser. ZAA devices for mercury analysis are available which have a detection limit down to 2 ng m-3, the spectral line to which the magnetic field is applied being at 254 nm.
The background level of antimony in an urban area depends on the volume of traffic and can be up to ≈20 ng m-3. In certain industrial environments in which antimony is a component of a catalyst, levels in the air can be over an order of magnitude higher than ambient levels distant from a single major source, and in such situations monitoring is needed. ETAA has found application to such monitoring. The spectral line used in analysing for antimony has wavelength 217.6 nm.
Amounts of zinc in air vary very widely and depend on factors including rainfall. Typical ambient atmospheric zinc contents would be 10 to 100 ng m-3 from trace amounts in coal and also in petroleum fuels. FAAS measurements in parts of urban Nigeria have revealed levels up to 171 mg zinc per kg air, equivalent to 200 mg m-3, several orders of magnitude higher than the levels previously given as being typical.
This element is present in minute amounts in coals and petroleum fractions and finds its way into the atmosphere on combustion. On account of its toxicity, cadmium needs to be monitored in the atmosphere and atomic absorption techniques have been widely applied. FAA, ETAA and ZAA are all routinely applied to such measurements. In each case, detection limits of about 0.01 ng cm-3 (0.01 mg m-3) apply. Exposure limits for cadmium in air are typically two to four times this level.
The organisms from which crude oil was formed contained a chemical resembling haemoglobin in blood, but having a vanadium atom at its centre whereas haemoglobin of course has an iron atom. This vanadium is retained in heavy fuel oils in particular, and this is one way in which vanadium finds its way into the atmosphere. FAAS and ETAA are both applied to vanadium in the atmosphere, with detection limits of about 0.25 ng m-3.
Barium in the atmosphere arises from its occurrence at very low levels in fuels and also from industrial operations that involve barium compounds, for example the manufacture of fluorescent lamps. The atmosphere distant from a source of barium contains about 0.0015 p.p.b. of barium and this challenges atomic absorption methods. In places where barium is being used there is a need for monitoring of levels well in excess of the ambient. In addition to the manufacture of fluorescent lamps already referred to, barium compounds find application to the paint, rubber and glass industries. FAAS is amongst the techniques that can be used to measure atmospheric levels of barium in such industries, and the detection limit is 20 p.p.b. (≡0.1 mg m-3).
Levels of arsenic in the atmosphere are about 0.02 μg m-3. As noted above, arsenic displays more than one spectral line. If that at 228.8 nm is used in atomic absorption spectroscopy there is the difficulty that cadmium also has a spectral line at this exact wavelength. ‘Slits’ are sometimes incorporated into atomic absorption devices which can resolve closely similar wavelength emissions from different atoms in the flame, but such a slit can only separate lines at least half a nanometre apart so cannot resolve the arsenic and cadmium lines referred to. The spectral line from arsenic at 193.7 nm is often used in atomic absorption. The line at 189 nm might be used as an alternative.
Our discussion has been in terms of the presence of these elements in the atmosphere. A reader will obviously have realised that atomic absorption methods are by no means restricted to the gas phase but are widely used for analysing for particular elements in condensed phase materials including water. Two applications from the huge number that there are will be mentioned: measurement of amounts of mercury in fish harvested as seafood, often by CVAA, and measurement of amounts of manganese in quality control of steel manufacture. With some such applications sample preparation is an important part of the procedure and preliminary experimentation on such preparation might be necessary.
During the writing of this piece I called an eminent inorganic chemist who had taught me when I was a student in the 1970s and had some background on the development of atomic absorption. My former teacher expressed the view that atomic absorption had revolutionised analytical chemistry. A huge amount of related literature has appeared over the decades since Walsh’s discovery, some of it concerned with atomic adsorption per se . There is also a great deal of literature in which atomic absorption features as a means to an end, the analytical information from it being put to some use. How can one possibly do justice to such a topic in an article on the scale of this one? The physics of the technique has been outlined and application to seven selected elements briefly explained and commented on. A reader will hopefully have become aware that applications of atomic absorption are multitudinous and that development of new applications is continuous. At the very least this article will, I hope, have given the reader not previously having encountered atomic absorption sufficient confidence to investigate the topic further for him/herself.
Other types of spectroscopy for possible future coverage
In AA it is electronic transitions within an atom that lead to the spectral lines. By transition we mean the raising of an electron to a higher energy level by acceptance of a quantum of light (absorption spectrum) or release of a quantum of light with the accompanying transition of an electron to a lower level (emission spectrum). We have seen that AA is unique in that it involves emission and absorption. In general ‘electronic spectroscopy’, with atoms or with molecules, will involve one or the other but not both. Gaps between electron energy levels in atoms and molecules are large, necessitating radiation in the visible or the ultra-violet region of the electromagnetic spectrum for a transition, and sometimes the term ‘u.v./visible spectroscopy’ is used.
However, functions of molecules other than their electrons are also amenable to examination according to the interaction of radiation with matter and spectroscopic techniques based on these are consequently in wide use. The transition of a molecule from one vibrational level to another requires light in the infrared region, and infrared spectroscopy has been in wide use, with progressively more advanced equipment, for very many years. (The infrared spectrum of ‘heavy water’ when first obtained during World War II turned out to be surprisingly difficult to interpret.)
A principle very important in the understanding of spectroscopic behaviour of molecules is that the time taken for an electronic transition is so short as to be negligible in comparison with that required for a molecular vibration. The reason is not difficult to understand: the atoms, connected by bonds, which are vibrating are much heavier than electrons are, making the vibration a relatively sluggish phenomenon. This is known as the Born- Oppenheimer approximation.
Moving to longer wavelengths still than infrared, microwaves bring about transitions in rotational energy and this is the principle of microwave spectroscopy, which is very widely practised. Radio waves also find a place in spectroscopy, but in a somewhat more restricted way. Some atomic nuclei when placed in a magnetic field form energy levels, and a transition from one level to another can be brought about by absorption of radio waves.
This is called nuclear magnetic resonance (n.m.r.) and, like AA, it has had a major impact on analytical chemistry over the last four decades or so. The usual procedure when running an n.m.r. spectrum is to use a single radio frequency and to vary the field, as the separation of the levels depends on the field strength. There will be a response when the field is such that the spacing of the levels corresponds to the energy of the radio waves, so a signal is said to be at ‘high field’ or at ‘low field’ according to where it occurs as the field is varied. Applications of n.m.r. are legion, and there are certain very important medical examples. Selected environmental examples of the types of spectroscopy outlined in this paragraph will form suitable future material for AWE International.
Published: 10th Jun 2008 in AWE International