272 ANALYTICAL PROCEEDINGS, JUNE 1993, VOL 30 Precursors to and Evolution of Elemental Organic Tube Combustion - Analysis Over the Last Two Hundred Years D. Thorburn Burns Department of Analytical Chemistry, The Queen’s University of Belfast, The tube combustion procedures for organic elemental analysis go back to the pioneering studies of Lavoisier using vari- ous oils. The problems of early workers with the determination of hydrogen, caused by difficulties with the absorption of water, were solved elegantly by Prout for the analysis of carbohydrates and organic acids using a known volume of oxygen for the combustion. The princi- ples of modern automatic CHN analysers can be seen to derive from the work of earlier analysts such as Rigg, who used solid oxidants, coupled with the micro- practices developed by Pregl in the post- Leibig era.Introduction Since the time of Robert Boyle it has been stated that ‘an element is the ultimate product of analysis’. Early chemists’ methods of analysis were not completely specific and it was a long time before the concepts of the Aristotelian elements, earth, air, fire and water, were refined into the chemical elements as now under- stood.’ Fire was at one stage both an element and a resolving agent or agent for analysis: Boyle’s ‘Sceptical Chemist’ is an essay on the validity of fire assay,2 Lavoi- sier solved the problem of the nature of combustion, reaction with oxygen, and considered that organic compounds were com osed of carbon, hydrogen and oxy- gen! Later, Berthollet dicovered the presence of nitrogen in organic com- p o u n d ~ , ~ the determination of which was surveyed by Stephen’ and Thorburn Burns6 at the Kjeldahl Centenary Meet- ing in 1983.Antoine Laurent Lavoisier (1743-1794) Lavoisier, born in Paris in 1743, became an able scientist and public administrator; however, the latter was the cause of his death by guillotine in 1794 after trial by the Revolutionary Tribunal.’ He is justly remembered for his discovery of the role of oxygen in chemical reactions and for his reform of chemistry, including the elimination of the concept of phlogiston. Lavoisier, as noted by Szabadvary’” and by Belcher,’ was the first to attempt to determine the composition of organic compounds. First, he examined various oils by combustion in a somewhat compli- cated apparatus (Fig.l),37’0>1’a the description and operation of which is readily available in Kerr’s translation of ‘Elements of Chemistry’.’* The bulk of the water produced was condensed and collected (at 16 in Fig. l), the remainder being removed by an absorption tube (19- 20) containing deliquescent salts. Carbon dioxide was absorbed in a series of bubblers (22-25) containing caustic alka- line solution, a total of at least nine, the later ones filled with lime water (to check for complete absorption). The air was collected in a gazometer (connected at 30) and small portions withdrawn and assayed for residual oxygen by absorption with potassium sulphide. The original plates were drawn by Madame Lavoisier who, as can be seen from her portraits, was elegant as well as useful. (Marie Anne Paulze married Lavoisier in 1771, aged 13.) The original plate of the combustion apparatus contains an error: the tubes in the carbon dioxide absorbers were con- nected the wrong way round; Ken cor- rected this mistake. l2 Lavoisier wisely did not attempt to combust volatile com- pounds such as alcohols in this a aratus because of the risk of explosionR‘>” but used much simpler apparatus and obtained 29% C for spirit of wine.Lavoi- sier, in later experiments, made use of solid oxidants for the analysis of sugar.’“ Although his results were only approxi- mate his work is important in showing the correct approach to the problem of or- ganic elemental analysis. Richard Rigg (1799-1861) Rigg was one of the earliest British organic elemental analysts.Very little is Belfast BT9 5AGf Northern Ireland known about him; Boase gives but a brief biography13 based on Lonsdale’s ‘Worth- ies of Cumberland’.14 The Royal Society, to which he was elected in 1839, has some archival material mainly concerned with his publications. His Royal Society nomi- nation certificate was signed by Faraday , Daniell, Graham and Phillips, amongst others. Rigg published 16 papers in the period 1836-1846, dealing with chemical changes in the fermentation and germina- tion of seeds, growth of plants and organic analysis. His method of analysing or anic compounds appeared in Phil. in 1838 and in the Arcana of Science,17 although it had been shown to the Royal Society about 2 years earlier. The appara- tus consisted of two small glass tubes connected by a caoutchouc (rubber) collar (C) supported on a wire frame (Fig.2). The compound to be analysed was mixed with black copper oxide [cop- per(r1) oxide] packed into tube A and covered with 1 inch or more of the same copper oxide. The end of the tube was filled with dry amianthus (fibrous asbes- tos) or cotton wool. This section of the tube was heated to drive off the moisture. The tube was cooled, weighed and con- nected up to the capillary tube B, the bent end of which was placed in a mercury trough under the graduated collection tube. The pure copper oxide section was brought to red heat with a spirit lamp, then the whole tube was heated to white heat and rotated in the flame. The com- bustion was carried out slowly in order to T R A I T E E L E M E N T A I R E D E C H I X I E Fig.1 Lavoisier’s apparatus for the combustion analysis of oils (1784)ANALYTICAL PROCEEDINGS, JUNE 1993, VOL 30 273 x \ Fig. 2 Rigg’s elemental analysis apparatus (1838) avoid the formation of carbon monoxide. The tube was next allowed to cool. The loss in mass was equal to the mass of N2 + C02. The analysing tube was then reheated to drive off water, cooled and re- weighed. The nitrogen and carbon diox- ide were collected in the graduated tube. The carbon dioxide was removed by absorption with potassium hydroxide solution. A correction was made for the gaseous products left in the capillary tube and in the analysing tube (2/10 of its interstices). The ultimate analysis was calculated from the weights of water, carbon dioxide and nitrogen, and the volumes of C 0 2 and N2.Rigg’s technique and results were praised by J. B. Readel’ who, it should be noted, also signed his Royal Society Nomination Certificate. The results were, in fact, far from reliable and led Rigg, by now an FRS, to the erroneous conclusion that carbon was not an element but a compound body made by plants.” An editorial footnote to Riggs’s paper15 on the preparation of black copper oxide drew attention to Prout’s contribution to organic elemental analysis. William Prout (1785-1850) Prout was born on a farm in Horton, Gloucestershire, but spent most of his life in London, where from 1812 he practised as a physician until his death in 1850.2@22 He studied medicine in Edinburgh (1808- 1811) where he came under the influence of the distinguished chemist T.C. Hope. He made significant contributions to physiological chemistry including the dis- covery that the stomach secretes free hydrochloric acid, and is remembered for his unitary matter theory, i.e., that the chemical elements possess atomic weights which are integral multiples of the atomic weight of hydrogen,23 now known as Prout’s Law or Prout’s Hyp~thesis.~~-~’ This law, published anonymo~sly,~~ gave a great impetus to the determination of accurate atomic weights. Prout’s contri- butions to organic analysis have been overlooked. Szabadvary’ dismisses them as ‘difficult to understand’ and comments that the results were unexpectedly precise. Prout’s first contribution to the analysis of organic substances,28 published in Thornson’s Annals of Philosophy, con- cerned the use of Wollaston’s chemical equivalents in calculating formulae and in drying samples prior to analysis, for which purpose he designed and used a vacuum desiccator wherein the sample is dried at 212°F and exposed to concentrated sul- phuric acid (Fig.3). He produced excel- lent results for the analysis of urea29 and reported that black copper oxide was a better oxidant than potassium chlorate when compounds contained nitrogen. He then gave a detailed description of his well-engineered apparatus3’ in which the combustion tube was vertical and a circu- lar burner was used to heat the tube evenly (Fig. 4). He determined water either by differences (which he thought was best) or by condensation coupled with a calcium chloride drying tube.Early elemental analysts found considerable difficulties in drying substances and hand- ling hygroscopic powdered substances. Prout solved the problem in a most elegant manner, by noting that if a compound containing hydrogen, carbon and oxygen is burnt in oxygen, the volume of oxygen remains unchanged if the hydrogen to oxygen proportion is the same as in water; the determination of carbon dioxide then gave a complete analysis.31 Corrections could be made if the volume of the combustion product was greater or less than the original Fig. 3 Prout’s vacuum dessicator (1815) volume of oxygen. His second elemental analysis apparatus used a horizontal com- bustion tube with vertical mercury man- ometer tubes, which also acted as pumps to drive the oxygen back or forth over the sample (Fig.5). The results were amazingly accurate (Table 1). Little is known of Prout’s personal life in London but he was a successful physician who specialized in digestive and urinary com- plaints, as indicated by much of his chemical publications. Unfortunately, deafness made him avoid scientific con- tacts after 1830. He was extremely religious and was invited to write one of the eight Bridgewater Treatises32 which had the general title ‘On the Power, Wisdom and Godness of God, as Mani- fested in the Creation’. He dealt with his personal interests in ‘Chemistry, Metero- logy and the Function of Digestion’.33334 Tube combustion methods were improved by Leibig and Dumas, then refined and transformed to the micro- scale by Pregl .8,9 Fritz Pregl (1869-1930) Pregl was born in Laibach, Austria (now Ljubljana, Yugoslavia) in 1869.35,36 Edu- cated at the local Gymnasium, he entered the University of Graz to read Medicine, where he gained an intense interest in physiology which led, as with Prout, earlier, to the need to develop appropri- ate analytical methodology for his prob- lems.Whilst working on bile acids and protein chemistry he found the then analytical methods too complicated, lengthy and inexact, and in addition requiring large samples which were diffi- cult to obtain. He thus focused on organic microanalysis, which gradually claimed his full attention. He was appointed Professor of Medical Chemistry at Inns- bruck in 1910. His first task was to obtain a sufficiently sensitive microbalance.This was made for him by W. H. Kuhlmann of the P. Bunge Works, based on an earlier balance made for F. Emich. Pregl scaled down the methods of Leibig (Fig. 6) and Dumas and, after a deal of effort and close attention to detail, developed reli- able methods which spread world-wide. He extended the work, developing micro- methods for sulphur, halogen, carboxyl and other functional groupings. The results of these studies were published in 1917: ‘Die Quantitative Organische M i k r o a n a l y ~ e ’ . ~ ~ , ~ ~ The impact was such that in 1923 he was awarded the Nobel Prize for Chemistry, the first to be awarded for accomplishment in Analyti- cal Chemistry. In 1913 he had been recalled to Graz, where he remained active in research till his death in 1930. He is commemorated in the Fritz Pregl prize for Microchemistry, awarded by the Aus- trian Academy of Science and by the Fritz Pregl medal of the Austrian Society for Analytical Chemistry.274 ANALYTICAL PROCEEDINGS, JUNE 1993, VOL 30 L ~ ~ Y z v .~ /rorBfl&m& ,&/ah’- hrB&vm &u%& & Ji, 2&T#68b f i r . J/&L iuaa Fig. 4 Prout’s elemental analysis apparatus (1820) Post-Pregl to Full Automation of Analysis Following the establishment of the Pregl methods modifications became necessary as the range of organic and organometal- lic compounds requiring analysis increased. The developments and modifi- cation may be summarized under four main heading^.^^,^' 1. Combustion tube fillings (oxidation fillings, nitrogen oxide absorbents, etc. ) .2. Combustion procedures (partial auto- mation and rapid procedures). 3. Apparatus design. 4. Determination of carbon dioxide and water (chemical, manometric and chromatographic procedures). Fig. 5 (1 827) Prout’s elemental analysis apparatus The next major development following Simon’s work4* in the early 1960s was Perkin-Elmers’ production of the Model 240 Elemental Analyser for C, H and N. Its semi-automatic operation is well docu- mented.40.42 The sample (2-3 mg) is com- busted in oxygen at about 950°C. After oxidation interfering components are removed and the products H20, C02 and oxides of nitrogen are swept to a reduc- tion stage by means of helium. The final combustion products H20, COz and N2- He are homogenized. The detection system measures thermal conductivity before and after the absorption of H20 and CO,; finally only N2 and He remain, which are related to pure helium to give a signal for N2.Modifications to combus- tion tube packings, developed to deal with fluorinated compounds43 and for use when large numbers of halo en and be analysed, are now recommended. Operational modifications such as to the ladle45 and the digitization of output data46 have been described. sulphur containing compounds4 ‘F have toANALYTICAL PROCEEDINGS, JUNE 1993, VOL 30 1 St n Fig. 6 Pregl’s micro elemental analysis train (1917) Full automation following sample en- capsulation is achieved in the Carlo Erba l102.40,42 It uses the same basic Leibig chemistry4’ but the final products of combustion/reduction are separated by gas chromatography and measured by a katharometer.Provided that flow rates are carefully controlled and widely differ- ent sample composition standards are used excellent results can be attained.48 1 2 3 4 5 6 7 8 9 10 11 12 References Read, J . , Through Alchemy to Chemistry, Bell, London, 1961. Boyle, R., The Sceptical Chemist. . . , J. Cadwell for J. Crooke, London, 1661. Lavoisier, A. L., Trait6 Elementaire de Chemie.. .. Cuchet, Paris, 1781. Berthollet, C. L., J. Phys., 1786, 28, 272. Stephen, W. I., ‘Determination of Nitrogen in Organic Compounds in the Years before Kjeldahl’s Method’, Anal. Proc., 1984, 21, 215. Thorburn Burns, D., ‘Kjeldahl, the Man, the Method and the Carlsberg Laboratory’, Anal. Proc., 1984,21,210. Guerlac, H ., ‘Antoine-Laurent Lavoi- sier’, in Dictionary of Scientific Bio- graphy, ed. Gillispie, C. C.. Scribner, New York, 1973, vol. 8, p. 66. Szabadvary, F., History of Analytical Chemistry, Pergamon Press, Oxford, 1966; 8a p. 285; 8b p. 289. Belcher, R., ‘The Elements of Organic Analysis’, Proc. Anal. Div. Chem. Soc., 1976, 13, 153. Lavoisier. A. L., ‘MCmoire sur la Com- binaison du Principe Oxygine avec L’Espirit-du-vin, L’Huile & Differens Corps Combustibles’, Mem. Acad. Sci., Paris, 1784, 593. Lavoisier, A. L., Oeuvres de Lavoisier, Imp. Imperiale, Paris, 1964-93, vols. 1- 6. l l a vol. 1, p. 346; l l b vol. 2. p. 586; Lavoisier, A. L. (trans. Kerr, R.), Elements of Chemistry . . ., Wm. Creech, Edinburgh, 1790 (5th edn., 1802). Ilc vol. 2, p. 773. 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 Boase, F., Modern English Biography .. ., Netherton and Worth, Truro, Lonsdale, H., Worthies of Cumberland, G. Routledge & Sons, London, 1867- 1875, vols. 1-6. Rigg, R., ‘On a Method of Analysing Organic Compounds’, Phil. Mag., 1838, 12, 31. Rigg, R., ‘Further Observations on the Ultimate Analysis of Organic Com- pounds’, Phil. Mag., 1838, 12, 232. Rigg, R., ‘Method of Analysing Organic Compounds’, in Arcana of Science and Art, J. Limbard, London, 1838. Reade, J . B., ‘On the Chemical Compo- sition of Vegetable Membrane and Fibre: With a Reply to the Objections, Professor Hemslow and Professor Lind- ley’, Phil. Mag., 1837, 11, 421. Rigg, R., Experimental Researches; Chemical and Agricultural Shewing Car- bon to be a Compound Body, Made by Plants and Decomposed by Putrefac- tion, Smith.Elder & Co, London, 1844. Brock, W. H., ‘William Prout’. in Dic- tionary of ScientiJc Biography. ed. Gil- lispie, C. C., Scribner, New York, 1973, vol. 174. Glasstone, S., ‘William Prout (1785- 1850), J. Chem. Educ., 1947, 24, 478. Copeman, W.S., ‘William Prout M.D. F.R.S., Physician and Chemist’, Notes Rec. Roy. SOC., 1969, 247. 273. (Prout, W.), ‘On the Relation between the Specific Gravities of Bodies in their Gaseous State and the Weights of their Atoms’, Ann. Phil., 1815, 5 , 321. Siegfried, R., ‘The Chemical Basis for Prout’s Hypothesis’, J. Chem. Educ., 1956, 33, 263. Benfey, 0. T., ‘Prout’s Hypothesis’, J. Chem. Educ., 1952, 29, 78. Kendall, J . , ‘The Adventures of an Hypothesis’, Proc.Roy. SOC. Edin- burgh.. 1949-50, 53A, 1. Brock, W. H., ‘Studies in the History of Prout’s Hypotheses Parts I and 11’. Ann. Sci., 1969, 25, 49 and 127. 1892-1921, VO~S. 1-6. 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 275 Dr. Prout, ‘Some Observations on the Analysis of Organic Substances’, Ann. Phil., 1815, 6, 269. Prout, W., ‘Observations on the Nature of Some of the Proximate Principles of Urine . . .’, Ann. Phil., 1818, 11, 352. Prout, W., ‘Description of an Appara- tus for the Analysis of Organised Sub- stances’, Ann. Phil., 1820, 15, 413. Prout, W., ‘On the Ultimate Compo- sition of Simple Alimentary Substances; with some Remarks on the Analysis of Organised Bodies in General’, Phil. Trans., 1827, 355. Brock, W., ‘Prout’s Chemical Bridge- water Treatise’, J.Chem. Educ., 1963, 40, 652. Prout, W., Chemistry, Meteorology and the Function of Digestion Considered with Reference to Natural Theology, W. Pickering, London, 1834; (2nd edn., 1834, 3rd edn., 1845, 4th edn., 1855). Brock, W. H., ‘William Prout and Barometry’, Notes Rec. Roy. Soc., 1969, 24, 281. Lieb, H., ‘Fritz Pregl’, Berichte, 1931, 64b, 113. Lieb, H., ‘Fritz Pregl’, Mikrochemie, 1931, 3, 105. Pregl, F., Die Quantitative Organische Mikroanalyse, J. Springer, Berlin, 1917. Pregl, F., (transl. Fylman, E.), Quanti- tative Organic Microanalysis, J . & A. Churchill, London, 1924. Ingram, G., Methods of Organic Ele- mental Microanalysis, Chapman and Hall, London, 1962. Bance, S . , Handbook of Practical Organic Analysis, Ellis Horwood, Chi- Chester, 1980. Simon, W., Sommer, P. F., and Lyssy, G. H., ‘Complete Automation of the Microdetermination of Carbon and Hydrogen in Organic Compounds’, Microchem. J., 1962, 6, 239. Belcher, R., Instrumental Organic Ele- mental Analysis, Academic Press, Lon- don, 1977. Macdonald, A. M. G., and Turton, G. G., ‘The Automated Analysis of Highly Fluorinated Organic Materials. A Note’, Microchem. J . , 1968, 13, 1. Gustin, G., and Tefft, M. L., ‘Improved Recovery of Rapid Micro Carbon and Hydrogen Method by Modified Com- bustion-Absorption Techniques’, Mi- crochem. J.. 1966, 10, 236. Thorburn Burns, D., McKnight, H. B., and Swindall, W. J., ‘Improved Com- bustion Ladle for an Elemental Ana- lyser’, Lab. Pract., 1978, 27, 650. Thorburn Burns, D., McKnight, H. B., Quigg, K. K., and Swindall, W. J . , ‘Automated Data Handling System for the Perkin-Elmer 240 Elemental Ana- lyser’, Analyst, 1980, 105, 544. Pella, E., and Colombo, B., ‘Study of Carbon, Hydrogen and Nitrogen Deter- mination by Combustion Gas Chroma- tography’, Mikrochimica Acta, 1973, 697. Swindall, W. J., and Thorburn Burns, D., ‘Improvements to the CHN Perfor- mance of a Carlo-Erba 1106 Elemental Analyser by Blank Evaluation, Drift Correction Using a Pair of Dissimilar Standards and Modifications to the Gas Flow system’, 2. Analyt. Chem., 1988, 31, 730.