MOLECULAR FLUORIMETRY, AND THE ANALYSIS OF POLYCYCLIC AROMATIC HYDROCARBONS (PAHs) IN CIGARETTE SMOKE
INTRODUCTION Following the initial absorption of radiation, a molecule may de-excite in a variety of ways, one of which is the fluorescent re-emission of a photon. As a spectroscopic technique, fluorescence has the advantage over absorption in that it has a greater sensitivity.
One class of molecules in which fluorescence is the dominant relaxation mechanism is the polycyclic aromatic hydrocarbons (PAH). These are of special interest not only because many of them are confirmed carcinogens, but also because they are quite commonly formed (in minute quantities) in the combustion of cellulosic materials. PAHs are discharged into our environment as the by- products of the combustion of fossil fuels. Other sources of PAHs include industrial processes, biomass burning, waste incineration, oil spills and cigarette smoke. While low molecular weight PAHs are gases, most PAHs have low vapour pressures and hence are adsorbed on airborne particles. These organic compounds have also been found in aquatic environments. The Environmental Protection Agency (EPA) has identified 16 PAHs as priority pollutants. Therefore, understanding the properties of PAHs, particularly as related to their qualitative and quantitative analysis, is important and highly relevant. It is thus not surprising that a substantial effort has gone into developing a sensitive analytical technique for their detection, nor (from what has been said above) that the method-of-choice should be fluorimetry.
AIM This experiment quantitatively determines the relative PAH concentration in low, medium and high tar cigarettes and examines the efficiency of the cigarette filter using fluorescence spectroscopy.
gln/jun2000/jul2001cfh/Mar2007, ed/er/May2008 2.
THEORY General Aspects Reference  describes the basic spectroscopic principles, while  outlines examples of analytical applications. An adequate introduction to the theoretical aspect of molecular fluorescence and a detailed account of the molecular-structural factors relevant to the fluorescence of organic molecules, can be found in Harris (CH18). Other general classes of compounds from which fluorescence is observed. are the inorganic rare-earth compounds and various organo-metallic chelates (see  Chap 5 Section F). The inorganic rare-earth compounds fluoresce (from 300 to 600 nm, depending on the element) because their excited states are f-orbital based. As these electrons are sufficiently screened from their surroundings, their excited states do not suffer degradation either by chemical reaction or internal-conversion/vibration back to the ground state as, by contrast, do the transition metals. Organo-metallic chelates are compounds in the form of a heterocyclic ring, which contains a metal ion attached by coordinate bonds to at least two non-metal ions. They therefore exhibit structural properties similar to those of fluorescent pure organic compounds (see  ). Their fluorescence is typically observed in the 500 to 600 nm region.
Quantitative Aspects  Transmitted intensity (It) can be found through the Beer-Lambert Law where; the light of intensity (Io) is incident on a cell (length (l) = 1cm) the cell contains a solution of absorbing substance (concentration (c) = M) the absorbing substance has a molecular extinction coefficient (ε = dm cm-1 mol-1) at a specific wavelength log10(Io/It) = εcl or alternatively, I t = Io e–2.3A where A = εcl is the absorbance (or optical density) of the solution. Now by definition the quantum efficiency of fluorescence is given by:
φf = Fluorescent quanta (per sec) Absorbed quanta (per sec) = If Ia
so that I f = Ia φf = (Io – It) φf = Io (1 – e–2.3A) φf Thus provided the absorbance is not too large I f = Io (2.3εcl) φf (1) i.e. for a dilute solution the intensity of the fluorescence observed is proportional to both the concentration and the intensity of the exciting light.
NB; you will need to use this equation to calculate Io and subsequently find the concentration of your cigarette samples.
The error introduced by assuming the validity of this relationship at all concentrations is commonly referred to as the “inner filter effect”, i.e. the solution at the back of the cell is exposed to a lower intensity of exciting light than that at the front, owing to adsorption (or “filtering out”) of a part of the exciting light by the intervening solution. The error varies
gln/jun2000/jul2001cfh/Mar2007, ed/er/May2008 3.
proportionately to the absorbance, with an absorbance of 0.01 introducing an error of 1% in If. In some experimental geometries the fluorescent light is not gathered over the full excitation path length, but only from a central region, which reduces the error. In practical terms, one must ensure that the fluorescent intensity varies linearly with concentration i.e. before the onset of saturation.
The previous equations also indicate how spectrofluorimetry can be much more sensitive than absorption spectrometry. In absorption spectrometry the concentration is proportional to log10(Io/It). The instrumental factor governing the minimum detectable concentration is the difference between Io and It. The strength of the light, (absolute intensity Io) is irrelevant. To achieve high sensitivity Io/It must be measured with a high degree of precision. In practice this can be done to somewhat less than one part in a thousand, giving log (Io/It) = 10–3. The maximum value of ε (for a strong band corresponding to a fully-allowed transition) may sometimes be as large as 105. Thus an estimate for the minimum detectable (molar) concentration, assuming the usual 1 cm cell, is cmin /M ~ 10–3/1×105 = 10–8 For a spectrofluorimeter, the instrumental sensitivity is limited in principle only by the maximum intensity of exciting light available, and not by the precision with which a light intensity can measured. With photomultiplier detectors exceedingly low light intensities (If) can be measured. Thus, providing If is not vanishingly small, by using very high exciting light intensities (Io) extremely low concentrations can be detected. With typical source intensities and other conditions (e.g. Фf, ε) favourable, concentrations as low as 10–12 M can be detected. Actual values measured both by fluorescence and absorption for a selection of PAHs are given in the following table .
DETERMINATION OF ENVIRONMENTAL PAHs BY FLUORESCENCE A model study on the fluorescent determination of specific PAHs in water is described in , while in  the partial identification and determination of PAHs in atmospheric particulate matter is discussed. An extensive study (though not by fluorescence) of PAHs in tobacco smoke is reported in . Taken together these papers give an adequate background to the present experiment;
Table 1: _________________________________________________________________ Compound (PAH) MW cmin Fluor /M cmin Abs /M φf ε _________________________________________________________________ Anthracene 178 2.1×10–12 3300×10–12 0.31 130,000 1,2-Benzanthracene 228 5.5×10–12 4800×10–12 0.19 91,000 Fluorene 166 1.5×10–10 220×10–10 0.53 20,000 Naphthalene 128 4.3×10–10 790×10–10 0.13 5,500 Phenanthrene 178 2.0×10–11 710×10–11 0.10 61,500 _________________________________________________________________ Avg. (geom. mean) 173 2.7×10–11 114×10–10 0.21 38,000 _________________________________________________________________
EXPERIMENTAL You are supplied with 2 low tar filtered (LTF), 2 medium tar filtered (MTF), 2 high tar filtered (HTF) and 2 high tar unfiltered (HTU) cigarettes, as well as a couple of excess cigarettes to calibrate the system Mark all cigarettes 5 cm from the unfiltered end. The burning is done on a suction line which includes 2 cold traps to collect the condensate
gln/jun2000/jul2001cfh/Mar2007, ed/er/May2008 4.
from the smoke. The traps are cooled by liquid nitrogen. Insert an unlit cigarette before cooling the traps.
Initially you need to calibrate the suction system, set a good flow of water through the suction pump, insert one of the cigarettes and light it, remember to time the burn time (to the 5cm mark)
The suction should be sufficiently great that smoke is not lost, but at the same time the cigarette should not burn too fast, otherwise condensate may be lost. 2 to 3 mins is satisfactory. This is adjusted by slightly altering the air bleed (leaving the water flow constant) – once the system is adjusted to your satisfaction clean out the smoke traps and tubes before starting your actual samples.
Start your sample collection – remember- • Each “sample” consists of two cigarettes • Cool the traps for a minute before you light the 1st cigarette, and let the traps warm up after the second before touching them to avoid cold burns • Remember to rinse out the cigarette holder and trap fitting
When the 2nd cigarette has burned to the mark, remove it with the tweezers supplied. Disconnect the suction and lower the dewar flasks, allowing the traps to warm up.
Using pasteur pipettes and alternate small aliquots of hexane and water/methanol mixture, carefully wash the condensate from both traps into a 100mL separating funnel. The last wash should be water/methanol and the traps should be clean. Don’t forget to wash the cigarette holder and include the washings in the sample.
To separate the PAH’s from the water/methanol/hexane mixture: • 1) Stopper the separating funnel and shake to mix the layers • 2) Allow the layers to separate – the top layer contains PAH’s/Hexane, bottom layer methanol/water. • 3) Drain the bottom layer into a clean 50mL volumetric flask. Do not throw out yet. • 4) Drain the top remaining layer into another clean, labelled (LTF, MTF etc) 50mL volumetric. • Take the bottom layer, and re-add it to the separating funnel. Add 15mL hexane, and repeat steps 1-4 two or three times, or until the top, hexane layer is clear. • Make the flask containing the combined hexane top layers up to the mark using hexane. Set aside for measurement later. • Remember – Rinse the separating funnel between samples! Dispose of solvents in the waste bucket in your fume cupboard. Don’t overfill the 50mL flask containing the hexane/sample mixture. This entire process needs to be completed for each sample (four times in total).
Because of their high concentration, further dilution of the HTF and HTU samples is required. Make a one-in-five dilution of both these samples by taking 2mL of the sample and making it up to 10mL with hexane in 10mL volumetric flask.
You should now have 6 samples – LTF-MTF-HTF-HTU-dilutedHTF-dilutedHTU.
CAUTION – METHANOL IS A HAZARDOUS CHEMICAL AND IS ABSORBED THROUGH THE SKIN. EXERCISE CARE WHEN USING IT.
gln/jun2000/jul2001cfh/Mar2007, ed/er/May2008 5.
Calibration curve construction: Take four 10 cm3 standard flasks. Add to them 2, 4, 6 and 8 mL of the standard anthracene solution supplied. Make up to the mark with hexane. You will also need and pure hexane and a pure anthracene standard solution.
FLUORIMETRY Before making each measurement wash the four sided cell with a little of the solution to be measured. Take great care not to get any solution on the outside of the cell, especially when emptying. Do not touch any of the four cell faces. Only handle the cell by the edges – if any solution is spilt on the outside, rinse with a little hexane and wipe carefully with tissue paper. Take care to not spill ANY solutions in or on the instrument. HANDLE THE FLUORESCENCE CELL WITH EXTREME CARE AT ALL TIMES. When you are finished with the cell, thoroughly rinse the cell and return it to your lab instructor.
Open the Cary Eclipse program called “Scan”. It is first necessary to record an excitation spectrum in order to determine the optimal wavelength for excitation( ) ex max λ . Here the excitation wavelength is scanned while the emission monochromator is held at a constant value. Partially fill a sample cell with the undiluted anthracene solution. .. Select Excitation and set the excitation wavelength scan range from 300 to 395 and the emission wavelength to about 400 nm. Run an excitation spectrum and decide on a value of ( ) ex max λ . Next, select Emission and run an emission spectrum between 365 nm and 480 nm using this value of ( ) ex max λ to find the wavelength of maximum emission( ) em max λ . You also need to decide on the best excitation / emission filter settings [auto?], the scan speed, the slit widths and the PMT voltage. Finally, re-run the excitation spectrum using this value of ( ) em max λ .
Using the optimal settings, measure the emission spectra of the remaining samples in the following order: (i) pure hexane (ii) 2/10 diluted anthracence (iii) 4/10 diluted anthracence (iv) 6/10 diluted anthracence (v) 8/10 diluted anthracence (vi) undiluted anthracence (vii) pure hexane (viii) undiluted LTF (ix) undiluted MFT (x) 2/10 diluted HTF (xi) 2/10 diluted HTU (xii) undiluted HTF (xiii) pure hexane If you do not get very similar results for the three pure hexane measurements then repeat ALL measurements until you get a set of results which are consistent.
Check that the sample results (viii)-(x) lie within the linear portion of the anthracene calibration plot. If any one does not, make an appropriate dilution to obtain a result that does. If you are satisfied with your results, then get them initialled by the Demonstrator (or the Lab Technician). Discard all samples (including water/methanol waste) into the organic waste bottle in the fume cupboard. Carefully rinse the cell and all glassware with acetone.
gln/jun2000/jul2001cfh/Mar2007, ed/er/May2008 6.
REPORT 1. Sketch the optical system of the fluorimeter used in this experiment. Show the source lamp, the lens, the filters, the sample cell and the detector. 2. Why is the path for the fluorescence at right-angles to the source beam? 3. Compare the excitation spectrum with the emission spectrum, comment on and explain any similarities / differences in shape and wavelength. 4 Tabulate your results for emission Intensity at ( ) em max λ vs. Concentration and the wavelength of each peak in the Excitation and emission spectra. Which peaks correspond to the 0-0 transitions?
5. From the diluted HTF and HTU fluorescence values, calculate the corresponding ‘dilution-adjusted’ values you would expect from undiluted solutions. How does this calculated value for HTF and HTU compare with the measured, undiluted value? Explain any discrepancy you observe.
6. Plot the anthracene calibration curve (nett fluorescence versus molar concentration; include zero concentration). Add a trend line (linear) and equation to the graph. Comment on the shape of the curve, and the result obtained for the undiluted anthracene sample. How does this plot relate to equation (1), and the discussion of the inner filter effect? How can you now account for any difference between the dilution-adjusted and the measured undiluted HTF values?
7. Using equation (1), the data for Anthracene in table (1), and any point from the linear section of your calibration curve calculate your reference Io value. Now, again using eq. 1, calculate the concentration “average PAH” in each cigarette sample – hint – use the Io reference value calculated above, data from table (1) and your experimental results. Present this data in two ways – (a) compare these relative results with the tar values stated on the packets (mG per cigarette) – what are some factors that could account for errors or have effected your results? And (b), normalise your results so that HTU = 1.0. What conclusions can be made as to the effectiveness of the filters? What assumptions are you making using the data in table (1) to calculate the average PAH in the sample?
8. From these estimates of PAH in cigarettes, and making any appropriate assumptions, comment on the amount of PAHs (in grams) released per year into (a) a smoker’s lungs, and (b) the atmosphere of the city of Melbourne. ‘
Remember your report must include an Introduction, aim, methods/materials, results/calculation, discussion, and conclusion and must be properly referenced and formatted.
REFERENCES 1. Chang, R; Basic Principles of Spectroscopy, (1971) Chapter 12. 535.84 C4566. 2. Browning, D.R., Spectroscopy (1969); Chapter 2. 3. Hercules, D.M., Fluorescence and Phosphorescence Analysis (1966). 545.812 H539f.
gln/jun2000/jul2001cfh/Mar2007, ed/er/May2008 7.
4. Parker, C.A., Photoluminescence of Solutions (1968). 535.35 P28p. 5. Cetorelli, J.J., McCarthy, W.J. and Winefordner, J.D., J. Chem. Ed., 45 (1968) 98. 6. Schwarz, F.P. and Wasik, S.P., Anal. Chem., 48 (1976) 524. 7. Fox, M.A. and Staley, S.W., Anal. Chem., 48 (1976) 992. 8. Severson, R.F., Snook, M.E., Arrendale, R.F. and Chortyk, O.T., Anal. Chem., 48 (1976) 1866. 9. Harris, C.E., Quantitative Chemical Analysis – 7th Ed. (2007) Chapter 18.
EXPERIMENT 25 MOLECULAR FLUORIMETRY, AND THE ANALYSIS OF POLYCYCLIC AROMATIC HYDROCARBONS (PAHs) IN CIGARETTE SMOKE