There have been articles in this periodical on atomic absorption 1, nuclear magnetic spectrometry (NMR)2 and mass spectrometry3. To these will be added in this article information on infrared spectroscopy, on ultra-violet/visible spectroscopy, and on fluorescence spectroscopy. This will supplement the earlier material and enable the series to congeal into a set of guidance notes on spectroscopy.
It is helpful to reproduce the electromagnetic spectrum (pictured above). It can be correlated with the content of the earlier articles, for example by noting that NMR uses radiation in the radio waves wavelength and frequency range.
Infrared spectrometry with special reference to carbon dioxide
Forty years ago when I was a teaching physical chemistry at Macquarie University in Sydney, I had to supervise a laboratory exercise which used a model of the carbon dioxide molecule. The atoms were represented by plastic spheres and the bonds by springs. Unlike most such molecular models then and now, it was not static. There was a device for making it move in a fashion that simulates the vibrational modes of the carbon dioxide molecule. It was my job to show the students which vibrational modes lead to infrared absorption and are therefore ‘infrared active’. Those which are provide a means of detecting and measuring carbon dioxide by infrared spectroscopy.
“information on infrared, ultra-violet/visible, and fluorescence spectroscopy will supplement earlier published articles”
To be infrared active a molecular vibration needs to effect a change in the dipole moment, which is why the symmetric stretch of carbon dioxide is not infrared active whilst asymmetric stretch and bending are. This is well known, and summarised in the diagram below4. It is because the carbon dioxide molecule is linear that some vibrations are not infrared active. By contrast the water molecule is bent and all of its vibrations are infrared active.
Infrared radiation interacting with a molecule has to be of a precise energy which corresponds to a spacing of adjacent vibrational energy levels. These are often expressed in terms of wave number, the number of waves in a centimetre, of the infrared radiation having been absorbed. The asymmetric stretch in carbon dioxide results in a peak at 2,350 cm-1. That corresponds to a wavelength of 4.26 μm. It is stated in ‘Infrared Spectroscopy’ Royal Society of Chemistry (2009)5 that carbon dioxide is ‘probably the molecule most people associate with the absorption of infrared radiation’. That is of course because of the international interest in atmospheric carbon dioxide referred to in the previous section of this article. Global warming through carbon dioxide and measurement of amounts of carbon dioxide by infrared have a common physical basis.
“1896 is often given as the time when scientific awareness of the absorption of heat by gases in the atmosphere began6.”
“about 60 years earlier, the French physicist Joseph Fourier expressed belief in retention of heat by gases in the atmosphere7”
In these times carbon dioxide, its formation, release and capture, sometimes appears to dominate world affairs. It is a mistake to believe that such matters are the intellectual property of our own generation. The year 1896 is often given as the time when scientific awareness of the absorption of heat by gases in the atmosphere began6. About 60 years earlier, in a largely intuitive way, the French physicist Joseph Fourier expressed belief in retention of heat by gases in the atmosphere7. Greenhouse gas measurement devices based on infrared are in wide use and will obviously remain so.
Absorption by a molecule of visible or UV radiation, of shorter wavelength than infrared, results in transitions between electronic energy levels. Imagine an aqueous solution of copper sulphate which, as is well known, has a blue colour. That is because when it encounters visible radiation it absorbs some of it, and the emergent radiation is depleted in some wavelengths. Whiteness is lost and the solution takes on colour. The absorption spectrum of the solution will therefore have peaks corresponding to those wavelengths which have been absorbed. The interested reader can easily enough access the absorption spectrum of copper sulphate solution online and relate it to the visible part of the electromagnetic spectrum. He or she will observe that most of the absorption is in the region of longer wavelength than the blue. The UV spectrum of gaseous carbon dioxide has peaks at 147.5 nm, 133.2 nm and 112.1 nm8.
Applications of UV visible spectroscopy are of course numerous. Five recently reported ones are described in the table below, which is followed by comments.
With reference to row 1 of the table, absorption in the 250 to 300 nm range is attributed to dissolved aromatic hydrocarbons. The absorption spectra showed changes with the season as would be expected, and opacity due to bacteria had a particularly marked dependence on time of year. Chlorophyll is not considered to be water soluble, but enough of it to show up in the absorption spectrum was present in the samples and was manifest by a signal at about 700 nm. Reference to the electromagnetic spectrum above shows that this would cause emergent visible light to shift towards green, the colour one associates with chlorophyll.
The natural products known as tannins (next row of the table) occur very widely. They contribute to the flavour of a cup of tea or a glass of wine. As noted in the table and a little surprisingly the tannins in example 10 showed no absorption in the visible region, therefore the authors concluded that they are ‘transparent in the visible region’. The tannins did absorb at some wavelengths in the UV region, for example at 205 nm where the absorption was attributed to aromatic compounds. Tannins vary in structure, and gallic acid is often seen as a representative compound. Its structure is shown below. That tannins will display absorption characteristic of aromatics can be inferred from that.
In the application of UV/visible spectroscopy to foodstuffs (next row) peaks such as were referred to in the previous two rows of the table are not the basis of the analysis as much as the wavelength range, over which there is absorption and the broad profile of the spectrum. The range given for egg white, consistently with what has been said about ‘whiteness’ previously in this article, encompasses the entire visible region of wavelengths. Where the overall features of the spectrum, but not close details of it are the basis of deductions, the term ‘fingerprinting’ can reasonably be applied. One would look for the ‘fingerprint’ of egg white or whatever food material is being examined. Others such as in ‘Advances in Food Authenticity Testing’11 include salmon fillet, which absorbs over the range 400 to 1100 nm. Pork absorbs between 400 and 1,000 nm. A reader will understand from that how food substances have a fingerprint in UV/visible analysis. In Australia in 1981, in what became known as the ‘roo in the stew affair’, there were allegations that kangaroo meat had been sold as beef. The investigators would have benefited from the UV/visible methods now used to identity foodstuffs. It is quite reasonable to expect that beef and kangaroo meat would have had easily distinguishable spectra.
“in Australia’s ‘roo in the stew affair’, the investigators would have benefited from the UV/ visible methods now used to identity foodstuffs”
Moving on to the dyes in the next row, that the dye chosen as an example is absorbing in the visible is clear and again the wavelength range is consistent with absorption at short wavelengths, causing the red colour of the emergent or reflected light. One of the dyes, described as being brownish, absorbed at 510 nm, 410 nm and 380 nm. That also can easily be interpreted by consultation of the electromagnetic spectrum. Application to wines, and the interesting matter of tracing their provenance, is discussed in the final row. Minor constituents like the phenols in a Spanish white wine discussed in the table are markers towards identifying the source of the parent grapes and their location. Anthocyanins are another class of substance which can be used as a basis for such characterisation, and are reported as giving peaks at 500 nm and at 560 nm13. Whisky also came within the scope of the work. A Scotch whisky from a Highland location and one from the western Isles had peaks in the visible region separated in wavelength by 100 nm making distinction of them very straightforward and quite unequivocal.
The phenomenon of fluorescence is emission of a photon by a molecule having previously received a photon of higher energy. Fluorescence spectroscopy is therefore an example of emission spectroscopy. The principles will be discussed further with reference to selected contemporary applications.
Fluorescence spectroscopy was applied to specially prepared blends of two natural oils – sunflower oil and tahini oil – so that emissions characteristics of each could be observed and noted14. Signals at 348 to 355 nm were attributed to tocopherol, the structure of which is shown below. Chemical compounds capable of displaying fluorescence often contain an aromatic nucleus, and tocopherol is typical of such compounds in that respect.
A signal at about 315 nm from the blend of natural oils14 was attributed to phenols. An application of UV/visible spectroscopy to wine was given in the previous section. An oenological application of fluorescence spectroscopy will be given here. The point was made at the beginning of this discussion that fluorescence means emission at a lower energy than the preceding absorption. Lower energy means longer wavelength, and a sweet (‘botrytized’) wine having displayed electronic excitation at 400 nm emitted at 492 nm. Having been excited at 460 nm it emitted at 536 nm15. These wavelengths are all in the visible region.
In ‘Fluorescence spectroscopy as a tool for in vivo discrimination of distinctive skin disorders’16 samples of human skin were examined using fluorescence spectroscopy. Emissions at 500 nm and at 600 nm were the basis of diagnostic distinction of basal cell carcinoma lesions from psoriasis lesions. A further histological application has been to cervical tissue17. From fluorescence emissions in the range 385 to 700 nm it was possible to distinguish pre-cancer cervical tissue from normal cervical tissue.
That fluorescence is about differing wavelengths, absorbed and emitted, is emphasised in an account of recent applications to dairy products18. Milk shows excitation at 330 nm and emission at 420 nm. This is so across a variety of forms of milk including skimmed milk, condensed milk and ‘infant formula’. An environmental application was to polyaromatic hydrocarbons (PAH) in Arctic waters19, having occurred there from historic oil tanker leaks. Simulated samples made by adding particular PAH to water in concentration of the order of milligrams per litre were examined by fluorescence spectroscopy and emissions in the wavelength range 303 to 464 nm were observed. These could often be linked to particular PAH compounds, e.g. anthracene.
As with UV/visibility, five applications have been outlined, and each time the observations have been linked to principles. All of the applications are thoroughly up to date.
“the scope of spectroscopy has extended beyond simple molecules to complex substances not having a molecular identity”
The eminent physical chemist G.M. Barrow (1923-2002) of the Case Institute in Ohio wrote a number of scholarly texts on spectroscopy and in one of the earlier ones said that spectroscopic techniques enable us to ‘see’ molecules20. The scope of spectroscopy has extended beyond simple molecules to complex substances not having a molecular identity. Spectroscopic methods enable us to ‘see’ something of those, as the examples used in this article show. There is evidence that Barrow himself was conscious of this expansion of scope in spectroscopy. A book from his pen published just over a decade after the one previously cited is entitled ‘Physical Chemistry for the Life Sciences’21. Applications have advanced and expanded enormously since then, and recent applications have featured in this article. A reader is asked to note that with one exception all of the spectroscopic applications in this article are all from the last three years. A follow-up article to this dealing inter alia with electron spin resonance spectroscopy is expected.
1 Jones J.C. ‘Spectroscopic techniques: atomic absorption spectroscopy’ Air, Water and Environment International June 2008 30-37.
2 Jones J.C. ‘Spectroscopic techniques: 2. Nuclear magnetic resonance’ Air, Water and Environment International June 2009 24-29.
3 Jones J.C. ‘Mass spectrometry in environmental applications’ Air, Water and Environment International March 2012 48-57.
5 ‘Infrared Spectroscopy (IR)’ Royal Society of Chemistry, London (2009).
6 Jones J.C. ‘Global Trends and Patterns in Carbon Mitigation’ Ventus Publishing, Fredricksberg (2013).
7 Jones J.C. ‘Fourier and Napoleon’ International Journal of Mechanical Engineering Education 35 182 (2007).
8 M., A., E. ‘Reaction Mechanisms in Carbon Dioxide Conversion’ Springer (2015).
9 Bastos de Oliveira D., Russo M.R., Caires A.R.L., Rojas S.S. ‘Fish farming water quality monitored by optical analysis: The potential application of UV–Vis absorption and fluorescence spectroscopy’ Aquaculture 490 91-97 (2018).
10 Grasel F., Ferrao M.F., Wolf C.R. ‘Ultraviolet spectroscopy and chemometrics for the identification of vegetable tannins’ Industrial Crops and Products 91 279- 285 (2016).
11 M.J., Vázquez M. ‘Advances in Food Authenticity Testing’ 35-70 Woodhead Publishing (2016).
12 Shindy H.A. ‘Synthesis, characterization and visible spectral behaviour of some novel methine, styryl and aza-styryl cyanine dyes’ Dyes and Pigments 75 3344- 350 (2007).
13 Urícková V., Sádecká J. ‘Determination of geographical origin of alcoholic beverages using ultraviolet, visible and infrared spectroscopy: A review’ Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 148 131- 137 (2015).
14 H.T., U. A., yaci I.H. ‘Synchronous fluorescence spectroscopy for determination of tahini adulteration’ Talanta 167 557-562 (2017).
15 Sádecká J., Jakubíková M., Majek P. ‘Fluorescence spectroscopy for discrimination of botrytized wines’ Food Control 75-84 (2018).
16 Maciel V.H., Correr W.R., Kurachi C., Bagnato V.S., Souza C da S. ‘Fluorescence spectroscopy as a tool to in vivo discrimination of distinctive skin disorders’ Photodiagnosis and Photodynamic Therapy 45-50 (2017).
17 Zhu H., Morris B., Wei F., Cox D.D. Multivariate functional response regression, with application to fluorescence spectroscopy in a cervical pre-cancer study Computational Statistics & Data Analysis 88-101 (2017).
18 Shaikh S., O’Donnell C. ‘Applications of fluorescence spectroscopy in dairy processing: a review’ Current Opinion in Food Science 17 16-24 (2017).
20 Barrow G.M. ‘The Structure of Molecules’ McGraw-Hill (1963).
21 Barrow G.M. ‘Physical chemistry for the Life Sciences’ McGraw-Hill (1974).