ISSN:0003-2654    

DOI:10.1039/AN96186FX017

出版商:RSC    

年代:1961

数据来源: RSC

ISSN:0003-2654    

DOI:10.1039/AN96186BX019

出版商:RSC    

年代:1961

数据来源: RSC

ISSN:0003-2654    

DOI:10.1039/AN96186FP069

出版商:RSC    

年代:1961

数据来源: RSC

ISSN:0003-2654    

DOI:10.1039/AN96186BP081

出版商:RSC    

年代:1961

数据来源: RSC

摘要:

APRIL, 1961 THE ANALYST Vol. 86, No. 1021 PROCEEDINGS OF THE SOCIETY FOR ANALYTICAL CHEMISTRY DEATH WE record with regret the death of Albert Arthur Henly. SCOTTISH SECTION A JOINT Meeting of the Scottish Section with the Glasgow and West of Scotland Section of the Royal Institute of Chemistry was held at 7.15 p.m. on Friday, February 24th, 1961, in the Royal College of Science and Technology, George Street, Glasgow, C.l. The Chair was taken by the Chairman of the Section, Mr. A. F. Williams, B.Sc., F.R.I.C. The following papers were presented and discussed : “Applications of Ion-exchange in Inorganic Analytical Chemistry,” by R. A. Wells, B.Sc., F.R.I.C., M.I.M.M. ; “Ion-exchange in the Electroplating Industry,” by V. E. Gripp, B.Sc., A.R.C.S., A.R.I.C. (see summaries below).APPLICATIONS OF ION-EXCHANGE IN INORGANIC ANALYTICAL CHEMISTRY MR. R. A. WELLS discussed the fundamental properties of ion exchangers, ion- exchange equilibria and ion-exchange kinetics in the context of inorganic analytical chemistry. The various ways in which the inorganic analyst might employ ion-exchange were illustrated by examples. He proposed possible further extensions of the use of ion-exchange in analytical chemistry, with particular reference to the use of modified cellulose exchangers, liquid ion exchangers, ion-exchange membranes and selective resins. ION-EXCHANGE IN THE ELECTROPLATING INDUSTRY MR. V. E. GRIPP discussed the elementary principles of ion-exchange and gave a brief review of their historical development and their use in large-scale plants for the purposes of water treatment and so on.This industrial usage of the ion-exchange processes was further amplified by reference to specific applications in the electroplating industry. The process had found extensive practical use in the metal-finishing trade from two entirely different stand-points, namely the purification of metal-treatment liquors, e.g., chrome anodising baths, and also as a means of decontaminating the emer- gent swill waters from metal-finishing shops. The purification of electroplating solutions represented an important economic advantage that had been well proved in practice] and the speaker amplified this theme by reference to several specific examples, during which he conveyed some idea of the scale upon which they were practised.The treatment of dilute swill waters from, e.g., chromium-plating processes, permitted several advantages to be gained at the same time by using a demineralising technique. Water was largely recovered in a purified condition for re-use, and the capital cost of effluent treatment was greatly reduced because the volume of the plant outflow was drastically curtailed. Several particular points of technical interest arose during the consideration of swill-water treatment] and he enlarged upon these during the lecture. 209210 PROCEEDINGS [Vol. 86 WESTERN SECTION THE sixth Annual General Meeting of the Section was held at 5.45p.m. on Thursday, January 12th, 1961, at the University of Bristol. The Chair was taken by the Chairman of the Section, Dr. G. V.James, M.B.E., F.R.I.C. The following appointments were made for the ensuing year : Chairman-Dr. G. V. James. Vice-Chairman-Dr. F. H. Pollard. Hon. Secretary and Treasurer-Dr. T. G. Morris, :Brockleigh, Clevedon Avenue, Sully, Glamor- gan. Members of Committee-Dr. R. G. H. B. Boddy, Mr. R. E. Coulson, Mr. S. Dixon, Mr. E. A. Hontoir and Mr. R. F. Stephens. Mr. C. H. Manley and Dr. W. J. Williams were appointed Hon. Auditors. The Annual General Meeting was followed by a Joint Meeting with the Bristol and District Section of the Royal Institute of Chemistry at which the Chair was taken by the Chairman of the Bristol and District Section, Dr. R. Woodcock, F.R.I.C. A lecture on “The Analysis of Plastics” was given by J. Haslam, D.Sc., F.R.I.C. A JOINT Meeting of the Western Section and the Bristol and District Section of the Royal Institute of Chemistry was held at 7.30 p.m.on Thursday, March 2nd, 1961, at the Technical College, Brunswick Road, Gloucester. The Chair was taken by the Chairman of the Bristol and District Section, Dr. R. Woodcock, F.R.I.C. The following paper was presented and discussed : “Titrations in Non-aqueous Media, * ’ by E. Minshall, MSc., F.R.I.C. A JOINT Meeting of the Western Section and the South Wales Section of the Royal Institute of Chemistry was held a t 6 p.m. on Friday, March loth, 1961, in the Large Lecture Theatre of the New Chemistry Building of University College, Singleton Park, Swansea. The Chair was taken by Mr. P. F. Ellis, M.B.E., B.Sc., A.R.I.C., past Chairman of the South Wales Section.The following paper was presented and discussed : “Vapour-phase Chromatography,’’ by A. Verdin. MIDLANDS SECTION A JOINT Meeting of the Midlands Section with the Association of Clinical Biochemists was held at 6.30 p.m. on Thursday, February 16th, 1961, in the Lecture Theatre, General Hospital, Steelhouse Lane, Birmingham, 4. The Chair was taken by the Chairman of the Midlands Section, Dr. S. H. Jenkins, F.R.I.C., F.1nst.S.P. The following papers were presented and discussed : “The Use of Vapour-phase Chromato- graphy in the Study of Human Lipid Metabolism,” by H. G. Sammons, BSc., Ph.D.; “Some Recent Developments in Paper Chromatography,” by C. S. Knight, M.Sc., Ph.D., A.R.I.C. AN Ordinary Meeting of the Section was held a t 6.30 p.m. on Wednesday, March 8th, 1961, at the Technical College, Park Square, Luton.The Chair was taken by the Chairman of the Section, Dr. S. H. Jenkins, F.R.I.C., F.1nst.S.P. The following paper was presented and discussed : “Some Newer Reagents in Analytical Chemistry,’’ by Professor R. Belcher, Ph.D., D.Sc., F.R.I.C., F.1nst.F. MICROCHEMISTRY GROUP THE Seventeenth Annual General Meeting of the Group was held at 6.45p.m. on Friday, February 24th, 1961, in the Meeting Room of the Chemical Society, Burlington House, Piccadilly, London, W.l. The Chair was taken .by the Chairman of the Group, Mr. F. Holmes, BSc., A.R.I.C. The following Officers and Co:mmittee Members were elected for the forth- coming year : Chairman-Mr. C. Whalley. Vice-Chairman-Miss M. Corner. Hon. Secretary -My. D. W. Wilson, Department of Chemistry, Sir John Cass College, Jewry Street, Aldgate, London, E.C.3.Members of Committee-Mr. S. Bance, Mr. C. B. Dennis, Miss J. P. Dixon, Mr. F. Holmes, Dr. J. A. Hunter and Mr. H. C. J. Saint. Dr. L. H. N. Cooper and Mr. H. Childs were re-appointed as Hon. Auditors. The Annual General Meeting was followed by an Ordinary Meeting of the Group a t which the Chair was taken by the new Chairman of the Group, Mr. C. Whalley. The following paper was presented and discussed : “The Detlermination of Metals by Organic Reagents,” by F. Holmes, B.Sc., A.R.I.C. (see summary below). Hon. Treasurer-Mr. G. Ingram.April, 19611 PROCEEDINGS 21 1 THE DETERMINATION OF METALS BY ORGANIC REAGENTS MR. F. HOLMES reviewed briefly the factors influencing the formation of metal complexes and then described how the nature of the complexes affected their solubility ; for insoluble complexes the effect of pH was also mentioned.He then surveyed the possibilities of chelate ring formation in the light of the ligand atoms involved and the flexibility of the rings formed; both bidentate and polydentate ligands were dealt with. After this introduction the lecturer passed on to a consideration of unselective sequestering and precipitating agents. These were surveyed in general terms, then EDTA and structural modifications of it were considered, from both the practical and theoretical points of view. He pointed out the consequences of changing the molecule in various ways in attempts to make it more selective, or its complexes more stable.Following this he made some mention of nitrogen- and sulphur-containing ligands, which were more selective for transition metals and B-sub-group metals. The relation- ship between stability and configuration were indicated for different types of co-ordination and different ligand molecules, and he discussed the limitations imposed by the Irving - Williams order. The rest of the lecture was devoted to a consideration of two highly selective groups of reagents-those based on two oxime groups and used for the gravimetric determination of nickel and palladium, and those containing the -N=C-C=N- group and used for the spectrophotometric determination of ferrous, cuprous and other reducing cations. He gave an indication of the charge-transfer spectra associated with these latter reagents.Mr. Holmes dealt at some length with modern work on substituted bipyridines and o-phenanthrolines and discussed the properties of the chelates formed by these com- pounds, He concluded by saying that, although these properties might be understood once they were known, it was not generally possible to predict them quantitatively. PHYSICAL METHODS GROUP THE Seventy-fifth Ordinary Meeting of the Group was held at‘6 p.m. on Tuesday, February 28th, 1961, in the Cuthbert Wallace Lecture Room, Royal College of Surgeons of England, Lincoln’s Inn Fields, London, W.C.2. The Chair was taken by the Chairman of the Group, Dr. G. W. C. Milner, F.R.I.C., A.1nst.P. The subject of the meeting was “Physical Methods of Analysis Used in Medical Research’’ and the following papers were presented and discussed : “Gas Chromatography and Anaesthetic Research,” by D.W. Hill, MSc., A.Inst.P., A.M.I.E.E. ; “The Measurement of the Oxygen Tension of Blood by Means of a Covered Platinum Electrode,” by J. M. Bishop, M.D., M.R.C.P. ; “Application of Infra-red Spectroscopy to Some Problems of Medical Science,” by A. E. Kellie, B.Sc., Ph.D., A.R.I.C. (see summaries below). The meeting was preceded a t 3.15 p.m. by a visit to the Research Departments of the Royal College of Surgeons of England. GAS CHROMATOGRAPHY AND ANAESTHETIC RESEARCH MR. D. W. HILL said that the outstanding advantage of gas chromatography in anaesthetic and respiratory research lay in its versatility. He concentrated on the various techniques that had been employed rather than on the applications.In anaesthetic research, in particular, one was faced with the need to analyse a considerable variety of gases and vapours. These included oxygen, nitrogen, carbon dioxide, nitrous oxide, cyclopropane, diethyl ether, halothane, chloroform, trichloro- ethylene and methoxyfluorane. Using packed columns, the anaesthetic vapours were separated on a short column of dinonyl phthalate on firebrick, run at 80” C. In parallel with this was run a 20-foot column of dimethyl sulphoxide on firebrick, at room temperature, to separate oxygen, nitrous oxide and carbon dioxide. In addition to the use of packed columns, he described an experimental capillary- column gas chromatograph with a cathode-ray tube display. An electrically driven sampling valve was used.212 PROCEEDINGS [Vol.86 The flame-ionisation detector had been extensively used by Dr. Butler of the Research Department of Anaesthetics for the analysis of anaesthetic concentrations in whole blood. This technique offered considerable promise for use in pharmacological investigations. THE MEASUREMENT OF THE OXYGEN TENSION OF BLOOD BY MEANS OF A COVERED PLATINUM ELECTRODE DR. J. M. BISHOP said that the measurement of the oxygen tension of blood had considerable value in physiological studies. Owing to the shape of the oxyhaemoglobin dissociation curve, the oxygen tension was a more precise measure of the state of oxy- genation of arterial blood in normal subjects, and deviations from the normal in disease were more readily detected.The oxygen tension of venous blood might be of importance in a study of the metabolism of a region of the body. Estimation of oxygen tension indirectly from the dissociation curve was inaccurate for the arterial blood of normal subjects. The microtonometer method of Riley was a difficult technique and was not suitable for tensions greater than 90 mm Hg. Polaro- graphic measurement therefore offered many advantages, providing a direct estimate and theoretically being suitable for all oxygen tensions. Unfortunately many complicating factors arose, and for a long time they had not been satisfactorily overcome. The dropping-mercury electrode was unpredictably influenced by the presence of red cells, so that estimates had to be made on separated plasma. Bare platinum electrodes were similarly affected, and when covered with cello- phane or agar they were still sensitive to most of these difficulties, An instrument em- ploying this type of electrode was described, together with the results of tonometer experiments to establish its accuracy and reliability.APPLICATION OF INFRA-RED SPECTROSCOPY TO SOME PROBLEMS OF MEDICAL SCIENCE DR. A. E. KELLIE said that for the development of a practical infra-red spectrometer in the early forties they were largely indebted to the recognition of the value of infra-red spectroscopy in the oil industry. During the following decades improvements in in- strumental performance as a result of developing techniques-double-beam operations, multiple-pass systems, gratings and grating replicas, as well as improvements in detector response-had kept the field in a state of continuous change, yet almost from the outset the uniqueness of the steroid series as a medium for the application of infra-red spectro- scopic techniques was appreciated, and under the direction of the late Dr.Dobriner spectroscopic methods had been applied .with outstanding success to the study of the metabolism of endogenous steroid hormones. In the animal kingdom, many somatic cells possessed the ability to biosynthesise cholesterol, which could be considered as a major precursor of the animal steroid hormones. Steroids of the pregnane (C,,), androstanle (C,,) and oestrane (C18) series formed in the endocrine glands-adrenal cortex, placenta, ovary and testes-were secreted into the blood system, which carried them round the body to centres of biological action.They exerted their influence on the metabolic processes of the body, exercising control over carbohydrate metabolism (gluco-corticosteroids), electrolyte balance (mineralo-cortico- steroids), the development of secondary sex characteristic in males (androgens) and females (oestrogens), and were responsible for changes that were essential to the main- tenance of pregnancy (progestins) . The survival of these biologically active compounds in the blood stream as measured by their biological half-lives was short, and they under- went multiple changes leading to the foirmation of many new compounds, which were eventually excreted as a complex mixture in the urine. By reference to specific examples, Dr. Kellie illustrated how chromatographic techniques used in conjunction with infra-red spectroscopy contributed largely to the formidable task of separating and identifying the components of this mixture. The application of infra-red methods to other problems of medical biochemistry had been less striking. He used the failure to make equal progress in these fields to illustrate the limitations and difficulties of the metlhod.April, 19611 OBITUARY 213 BIOLOGICAL METHODS GROUP AN Ordinary Meeting of the Group was held at 7 p.m. on Wednesday, February 15th, 1961, in the Meeting Room of the Chemical Society, Burlington House, Piccadilly, London, W.l. The Chair was taken by the Honorary Assistant Secretary of the Society, Mr. S. A. Price, B.Sc., F.R.I.C. The following paper was presented and discussed: “The Use of Enzymes in Analysis,” by C. J. Threlfall, BSc.

ISSN:0003-2654    

DOI:10.1039/AN9618600209

出版商:RSC    

年代:1961

数据来源: RSC

摘要:

April, 19611 OBITUARY 213 Obituary WILLIAM HERBERT SIMMONS WILLIAM HERBERT SIMMONS was born in North London on May 24th, 1880, and died at his home in Addiscombe, Surrey, on December Znd, 1960. He received his early education at the Grocers’ Company’s School, Hackney Downs. He later studied chemistry and physics a t Finsbury Technical College, Birkbeck College and University College, taking his degree of BSc. in 1904. In the meantime, in 1899, he had entered the laboratory of Dr. Samuel Rideal as a pupil and, after serving for three years, became assistant chemist to the soap firm of Vinolia Co. Ltd., later succeeding as chief chemist, and in this period also taking up an appointment as consulting chemist to Ozonair Ltd. I t was this branch of chemistry, soap and essential oils, that was to become his life work, but, nevertheless, in 1907 he took up an appointment as research chemist to Messrs.Thomas Morson & Son Ltd., manufacturers of pharmaceutical chemicals, and remained with this firm until, in 1911, he established a private consulting practice in Victoria Street, London, specialising in the chemistry and analysis of soap, fixed oils, waxes and essential oils. In this field he became an internationally acknow- ledged expert and continued as a consultant until his retirement through ill health in June, 1959. In 1917 he was elected to the Fellowship of the Institute of Chemistry and somewhat later he included the field of petroleum and petroleum products in his consulting practice, becoming a Member of the Institute of Petroleum in 1934.For many years W. H. Simmons was Lecturer in Soap Manufacture, Waxes and Lubrica- ting Oils at Battersea Polytechnic. In 1914 he was appointed Examiner in Soap Manufacture to the City and Guilds of London Institute, and later was, for over 20 years, a member of the Advisory Committee of the same Institute on Petroleum and Petroleum Products. He also served on a number of the Technical Committees of the British Standards Institution and Sub-committees of the British Pharmacopoeia Commission and of the British Pharmaceutical Codex Revision Committee. Simmons joined our Society in 1904 and was first elected a Member of Council in 1918, serving many terms thereafter, including a Vice-presidency in 1938-9. For 29 years, from 1923, he was a conscientious member of the Publication Committee, but his most outstanding contribution to the work of the Society was that of Honorary Secretary to the Standing Committee on the Uniformity of Analytical Methods from the time of its formation in Nov- ember, 1924, until the name was changed in 1935 to the Analytical Methods Committee, and thereafter until 1946.Twenty-one years as Honorary Secretary to one of our most important Committees is indeed a fine record, but, not content with this, Simmons served as Chairman of the Essential Oils Sub-committee from 1933 until 1956, when, in recognition of his long service to the Society, he was appointed Honorary Member of the Analytical Methods Committee. Simmons was sole author of a book entitled “Soap: its Composition, Manufacture and Properties” (4th edition, 1936) and joint author with H.A. Appleton of “The Handbook of Soap Manufacture” (1908); with C. Ainsworth Mitchell of “Edible Fats and Oils” (1911) and with G. H. Hurst of “Textile Soaps and Oils” (3rd edition, 1921). Most of his original papers were contributed to The Perfumery and Essential Oil Record; upwards of 40 appeared over the years and many were abstracted in The Analyst, including a useful one on the Detection of Petroleum in Cassia Oil (Analyst, 1936, 61, 51). One of Simmons’s special interests was the use of formic acid in essential-oil analysis, and in this connection he contributed an important paper to The Analyst on the Determination of Alcohol in Geran- ium Oils (1915,40, 491) in which, for the first time, many valuable data were presented. The214 OBITUARY [Vol.86 formylation method has been widely used and was recommended in a recent A.M.C. Report on The Determination of Linalol in Essential Oils (Analyst, 1957, 82, 325) and was only amended last year in favour of the Fiore acetylation method (Analyst, 1960, 85, 165). The foregoing account indicates that Simmons was a man completely devoted to his profession, but nevertheless, as a freemason, he held London Grand Rank, was an active member of the University of London Graduates Association and a zealous member of the Methodist Church. W. H. Simmons was highly respected by all who knew him; his dominant characteristics were complete loyalty to all persons and causes, coupled with extreme modesty of outlook, these qualities being ever embellished with gracious courtesy.NOEL L. ALLPORT DONALD FORD PHILLIPS DONALD FORD PHILLIPS, whose tragic death in February of this year came as a great shock to all his friends, was born at Stourbridge and educated at Bablake Secondary School, Coventry. He held various appointments in the metallurgical industry, and I first met him in 1943, when he was Chief Chemist at High Duty Alloys Limited, Slough, and was enthusiastic enough to travel from Slough to Yorkshire every week to attend my evening course in microanalysis. This formed the basis of an intimate friendship which lasted until Phillips’s death. In 1949 he joined the Chemical Inspectorate, Department of Atomic Energy, Springfield, as Senior Experimental Officer, where he was in charge of a group of laboratories responsible for the analysis of uranium and ancillary materials by spectrographic, radiometric and fluorometric methods.In 1947, Phillips and I were selected as B.I.O.S. investigators to study developments in microanalysis in Germany during the war; this involved living together for about a month, sometimes under very primitive conditions. Throughout this time, Phillips’s sense of humour transformed what could have been a penance into an enjoyable outing. In 1949 he became Honorary Secretary of the Microchemistry Group and in 1957 its seventh Chairman. In both capacities he worked tirelessly to promote the activities of the Group, which still bears the stamp of his efforts. Phillips had two main interests outside chemistry: one .was transport and, in particular, railways.He published various articles on the subject in transport journals. His other main interest was languages; he spoke French and German fluently and had an astonishing knowledge of their idioms. Phillips was not only a good friend; he was that rarity, a loyal friend, as those who were privileged to know him remarked many times. He was generous and kind; his good humour became proverbial and nothing ever made him lose his temper. Those of us who knew him well will miss him sorely; but such was his character that he would not have us mourn him. One can say of him as of Chaucer’s knight : He held this position until his death. “And of his port as meek as i s a maid He never yet no villiany had said In all his life unto no manner of wight” R.BELCHER THOMAS TICKLE THOMAS TICKLE died in London on October 29th, 1960, in his ninetieth year. He was born on June l l t h , 1871, and was the only son of a Devon farmer. He started school in the small village of Patchacott and then proceeded to Shebbear School in Devon as a boarder. Upon leaving school he was apprenticed to a phanmaceutical chemist in Exeter, and in 1892 he entered the School of the Pharmaceutical Society as Jacob Bell Scholar. Here he had a highly creditable record, gaining a bronze medal in botany, silver medals in chemistry and pharmacy and also the Pereira Medal awarded for the most meritorious performance in the Society’s “Major” qualifying examination. Tickle was awarded a Salters’ research fellowship for the period 1896-98, when he made a study of y-pyrone a t University College, London, the results of his work being publishedApril, 19611 OBITUARY 215 in joint authorship with the late Professor J.Norman Collie. He then graduated as a B.Sc. of London University, became an Associate of the Institute of Chemistry in 1900, and in 1901 he passed the examination of the Institute in the branch which, at that time, covered Food and Drugs, Fertilisers and Feeding Stuffs, Soils and Waters, as well as Therapeutics, Pharmacology and Microscopy. It was in 1901 that Tickle was appointed Public Analyst for the City of Exeter, where he established a laboratory and began to practise as an analytical and consulting chemist. In the following year he was appointed Public Analyst for the County of Devon, and later also for the City of Plymouth and the Borough of Torquay.He was also Agricultural Analyst for Devonshire, Exeter and Plymouth. Tickle, always affectionately referred to by his many friends as “Tommy” Tickle, joined our Society (then the Society of Public Analysts) in 1902, and he was a Member of Council in 1914-15. He also served the Institute of Chemistry in a similar capacity in 1916-19. Although he was an inveterate experimentalist (he would not take anything on trust) and faithfully recorded the results of his work in his laboratory note-books, he did not publish much of i t ; but in 1907 he contributed a paper to the Pharmaceutical Journal on the assay of morphine, and in 1921 published a note in The Analyst on the determination of lead in peaty waters.In view of Tickle’s training it is not surprising that outside the work arising from his official appointments, his professional interests had a strong pharmaceutical bias. Much of his work involved analysis for the National Pharmaceutical Union of samples taken under the National Health Service drug-testing scheme, and his work in this connection resulted in the amendment of a number of official standards. In 1947 he took into partnership Dr. C. V. Reynolds, who had been his chief assistant for nearly 20 years and who now carries on the practice. Tickle served for many years as a member of the Board of Examiners of the Pharmaceutical Society, and it has been placed on record that in this capacity “he showed a characteristic sympathy, fairness and wide outlook which earned him the respect of all his colleagues.” He was a member of the British Pharmacopoeia Commission, 193248, and in 1948 he gave the inaugural address at the opening of the 107th session of the School of Pharmacy, which was the first session under the aegis of the University of London.A great lover of the out-of-doors, Tickle found relaxation from his work in swimming, rowing and sailing. Up to within a year of his death he regularly bathed in the sea a t Dawlish, and as a young man, he knew Dartmoor well. He was a Past Master of the Galen Lodge of Freemasons, London, of which he was a member for over sixty years. Thomas Tickle always found much pleasure in his family, to whom he was devoted, and at the time of his death he was visiting his son in London in order to be present a t the christening of his great-grandson. Two days before this event he arranged a dinner party for his relatives and some old friends, and, after spending a most enjoyable evening with them, he returned to his son’s home in Wembley and happily retired to rest. This tall, upright, bearded, sharp-featured and keen-eyed gentleman, with his warm pleasant personality, will be much missed by those members of the Society for Analytical Chemistry who were privileged to know him. He was elected a Fellow of the Institute in 1903. However, that was not to be. That night he died in his sleep. Tickle had been a widower for 25 years. He leaves two sons and two daughters. STAN LEY DIXO N

ISSN:0003-2654    

DOI:10.1039/AN9618600213

出版商:RSC    

年代:1961

数据来源: RSC

摘要:

216 LIGHTBOWN : BIOLOGICAL STANDARDISATION [Vol. 86 Biological S tandar disation and the Analyst A Review* BY J. W. LIGHTBOWN (Department of Biological Standards, National Institute for Medical Research, Mill Hill, London, N . W.7) SUMMARY OF CONTENTS Introduction Direct measurement of biological activity Comparative measurement of biological activity Comparison between non-biological and biological assay methods Requirements for valid biological assay Biological assay of heterogeneous material International co-ordination of biological standardisation Classification of international biological preparations International Standards International Reference Preparations Author's Preparations Authentic Chemical Substances Preparation of an International Standard International collaborative assays Development of international biological standardisation Expression of biological activity in weight units Expression of biological activity in international units Future developments Appendix MOST quantitative analytical procedures are based on the assessment of known chemical reactions or physical constants ; this necessitates a precise knowledge of the chemical structure and properties of the substance under examination.In a few instances the analyst has to estimate amounts of a substance of unknown or complicated structure, perhaps in a complex mixture for which chemical and physical methods cannot be used. Two examples illustrate the problems that then arise. A solution of dextrose prepared for therapeutic use can be analysed and the concentration of dextrose determined precisely.If, however, it has been carelessly prepared and is adminis- tered by injection it may produce an undesirable rise in body-temperature, owing to the presence of pyrogen, a material of bacterial origin with complex but uncertain structure. The presence of pyrogen cannot be determined by any known chemical or physical means. Blood serum contains a complex mixture of proteins, which can be fractionated chemically and physically. The globulin fraction of certain sera has the property of combining specifically with toxic materials elaborated by bacteria and neutralising the toxicity. A particular globulin fraction may contain many different antitoxin proteins, but these cannot be identified and determined by chemical and physical means. As the use of antitoxic sera is often the only method of treatment for certain toxaemias, their reliable standardisation is desirable.Many other useful therapeutic agents, such as the enzymes hyaluronidase and fibrinolysin, the hormones insulin and corticotrophin and the antibiotics nystatin and neomycin, pose similar problems to the analyst. Some of these drugs are extremely active and have only a small margin of safety between the useful and toxic doses, so that the activity must then be precisely controlled. In the future, chemical and physical methods for analysing these substances may become possible, but at present other methods must be used. One property common to all these materials is their biological activity; they are all capable of modifying in some way the activities of living organisms.Usually, the effects produced by them are specific: hyaluronidase causes the hydrolysis of hyaluronic acid, a material important in maintaining the structural organisation of certain tissues ; insulin, when introduced into the blood-stream of an animal, produces a decrease in the amount of glucose in the blood; nystatin interferes with the metabolism of yeast cells and prevents * Reprints of this paper will be available shortly. For details please see p. 275.April, 19611 AND THE ANALYST. A REVIEW 217 their growth; neomycin behaves similarly towards bacteria. In each substance we have a specific property producing an effect that can be measured quantitatively. The measurement of biological activity so as to quantify the substance producing the effect requires the use of biological standardisation by one or other of several techniques based on well established principles.DIRECT MEASUREMENT OF BIOLOGICAL ACTIVITY When measurement of a physical property, such as optical rotation, is used in analysis, it is possible to define conditions for agreement among different operators; the specific rotation of glucose is the same in London and New York. Experience has shown that conditions for measuring specific biological activity cannot be so defined. Insulin may be highly purified, but any attempt to define the decrease in blood-sugar produced by a given dose of insulin when injected into a rabbit would be unsuccessful. If a group of rabbits of the same age, weight and sex maintained under uniform conditions were all injected with a fixed dose simultaneously, the responses would be variable.Even the mean response for a large group of animals varies between experiments. When the concentration of an antibiotic that inhibits the growth of a given micro- organism is determined, many millions of individual cells are exposed simultaneously to the antibiotic, and a reliable mean value might be expected. In practice, the values obtained on successive occasions may still vary greatly. Variability is a characteristic of living organisms; variation in characters such as shape and size is self-evident, but variation in degree of response to certain stimuli is not always so obvious. The response of a living organism to the stimulus of a biologically active drug is the result of complex interaction of a series of related processes, e.g., absorption into the animal, transport to the sensitive site and penetration into the cell and components of the cell.Variation at each stage will occur between animals and even in the same animal at different times. Some small success can be obtained in standardising the biological system by using closed colonies of mice or standard bacterial strains, but the level of a given response still varies with changes in environment, such as temperature and nutrition, and also with other undefined variable factors ; effective standardisation is therefore impracticable in terms of a biological response. COMPARATIVE MEASUREMENT OF BIOLOGICAL ACTIVITY Ehrlich was the first to show how these difficulties could be overcome in standardising diphtheria antitoxin.Attempts had been made to standardise preparations of antiserum directly on the basis of the amount of diphtheria toxin they neutralised; but preparations of toxin were variable and unstable, and results varied from day to day. Ehrlich chose one particular sample of antiserum, which he designated as the standard reference preparation, and all other preparations of antiserum were compared directly with it. If the abilities of standard and unknown to neutralise a toxin preparation were compared directly in a single experiment, all variables affected both estimates equally, and a reproducible ratio for the activities of test and standard could be obtained. Ratios of the activities of a number of different preparations with the standard were directly comparable with each other.By defining the activity of the standard preparation in arbitrary units, the potencies of the unknown preparations could be stated in terms of these same units. This principle of calibrating the test system at every determination is now universally practised in biological assay. Standard preparations are available nationally and internationally for many biologi- cally active substances. Each has its activity defined in terms of units, and each standard preparation is a yardstick for measuring the activity of a particular substance. Many assays based on colorimetry involve use of this principle, as do determinations of arsenic and lead,. The fundamental difference is that in the physico-chemical assays the standard curves of samples are prepared with reagents of known identity and purity, whereas in many biological assays the purity and even the identity may be unknown.If agreement between laboratories is to be obtained under these conditions one unique standard preparation must be used, although working standards calibrated against the master standard may be used in routine analysis. In a few instances biological assay may be applied to materials of known structure and readily obtainable in a pure condition, perhaps because biological assay is more sensitive The use of a standard of comparison in analysis is not confined to biological assay.218 LIGHTBOWN BIOLOGICAL STANDARDISATION [Vol. 86 than a chemical or physical method or more convenient.Determinations of a number of vitamins and amino acids are often made in this way; a unique standard preparation is then unnecessary. Whether or not a unique standard preparation is required, the principles involved in designing and interpreting the assay are the same. COMPARISON BETWEEN NON-BIOLOGICAL AND BIOLOGICAL ASSAY METHODS All assays, including chemical and physical ones, are based on standards, and certain assumptions are made. The constants used to interpret effects such as light absorption, optical rotation and refractive index have been previously recorded for pure samples; as certain assumptions were made in estimating the purity, the constants have no absolute validity. In chemical assays, the specificity of the method also varies, and once again the determination is usually based on a standard of assumed purity, except in gravimetric determinations, which are referred ultimately to the standard gram.The assumptions made about the purity of the standards used in chemical and physical assays can for most purposes be ignored. Assumptions about the purity of material being assayed are also commonly made in routine analysis. Pharmacopoeia1 monographs are specifically designed to control impurities likely to be present, e.g., the assay described in the British Pharmacopoeia for tartaric acid is a determination of free acid and in itself is cornpletely nonspecific. Only when considered with tests for identity and with limit tests can it be considered to provide a valid quantitative measurement. Even then the assumption is made that the tartaric acid has been prepared in a normal way, thereby limiting the kinds of impurities that might be present. A sample of tartaric acid could contain 10 per cent.of citric acid and still comply with the B.P. requirements for tartaric acid. Although tartaric acid is produced biologically, it is chemically a relatively simple substance, and the impurities likely to be present after purification are known. For many naturally occurring substances that can only lie assayed biologically, there cannot be the same certainty as to the efficiency and specificity of the purification processes used. Often the substances occur naturally in families of closely related compounds in different proportions, eg., the penicillins and the neomycins.It may be possible to control the fermentation by choice of a suitable strain and growth medium, so that only a single member of the family is produced, as in the manufacture of penicillin, when mainly benzylpenicillin is obtained. Sometimes this control may be more difficult, and selection of a single component can only be obtained by chemical and physical fractionation, e.g., in the manufacture of neomycin. The difference in the assumptions that can be safely made in assaying materials by the different methods can be seen from the designs of the assays. This is illustrated by the methods of assay used for penicillin, which originally could only be assayed biologically, but may now be assayed by all three means. (a) Physical assay is conveniently carried out by determining the optical rotation at a single concentration.The result is interpreted by reference to a published value for reputedly pure material. (b) Chemical assay may be made by measuring the amount of iodine that combines with the penicilloic acid produced by hydrolysis of a known weight of penicillin. The nature of the reaction with iodine is not fully understood, and, since the amount of iodine combining with a fixed weight of penicilloic acid varies with conditions such as temperature, pH and exposure to light, it is necessary to calibrate the assay by means of an authentic sample. Assays of the unknown sample and the standard are made at a single concentration, as the number of atoms of iodine combining with one molecule of penicilloic acid is independent of the concentration of this acid.Penicillin may also be assayed by forming a coloured hydrorramate after hydrolysis, the intensity of the colour being measured photometrically. The assay is calibrated with a sample of known purity by plotting a graph relating optical d.ensity to concentration. Two concentrations of the standard are sufficient to demonstrate adherence to Beer’s law. The activity of unknown samples examined at a single concentration may be read from the line; it is a safe assumption that test and standard behave identically if conditions are st andardised. All these methods estimate total penicillin, and the results are calculated on the assumption that only benzylpenicillin is present. If a sample contained 10 per cent. of The specificity of physical methods may vary greatly.April, 19611 AND THE ANALYST.A REVIEW 219 heptylpenicillin this would not be apparent from the results, but similar results would be obtained in different laboratories. Molecule for molecule the different penicillins react in the same way. (c) Biological assay could be carried out by means of a design similar to that of the hydroxamate method, i.e., with two concentrations of a standard preparation to define the calibration graph and a single concentration of the unknown sample. This procedure, which might be expected to measure total penicillin, is unsound. Many factors, mostly ill defined, govern the shape of the calibration graph, which may be complex, and, although a portion of the line may be straight under certain conditions, these factors may be difficult to control.This is particularly so since one important factor may be the presence or absence of heptylpenicillin. Since it cannot be assumed that the standard and test will react with the test organism in an identical fashion, each assay must be constructed so that its validity can be assessed from its own internal evidence. REQUIREMENTS FOR VALID BIOLOGICAL ASSAY Three requirements that must be satisfied for valid biological assays have been discussed by Jerne and Woodl and, with special reference to closely related antibiotics, by Miles.2 The first is that differences between responses in the several dose groups of an assay are wholly caused by differences in dosage and by random sampling; the random sampling applies both to the variable test organism and to variations in environment that cannot be controlled.The second is that the response must be a determinable function of the dose; usually a trans- formation of the response is sought, which gives a straight line when plotted against the logarithm of the dose. Thirdly, the response of the standard and test materials must be due to a single active principle. If more than one active principle is present, the proportions in test and standard must be the same. This last requirement applies also to materials exerting an antagonistic or potentiating effect or affecting stability of the active principle. When a biological assay is constructed with three dose levels each of standard and test and with a sufficient number of test organisms or measured responses in the several dose groups, its validity can be checked statistically, and the potency ratio is obtained from the distance between the two dose-response lines. If the requirements described are met, valid bioassays can be obtained, and the potency expressed in international units per unit of weight or volume will be a measure of the content of biologically active principle as well as of the biological activity.If, however, any one of the requirements is not satisfied, then statistically valid assays may still be obtained or they may not. If they are obtained, then the potencies found for the unknown may vary from assay to assay. The potency in international units per milligram in these circumstances is not a measure of content of active principle, but only of the biological activity under the specific conditions of each assay.BIOLOGICAL ASSAY OF HETEROGENEOUS MATERIAL The requirement for valid bioassay that standard and test substances shall be homo- geneous is often not met in practice. Even when standardised procedures of isolation and purification are used, a variable product must be expected, since the starting material, being of natural origin, is itself variable in composition. The standard and test material may contain less than 1.0 per cent. of contaminating active material, as with certain antibiotics, hormones and vitamins, or the heterogenity may be much greater. The implications should always be remembered, particularly when biological and chemico-physical assays are being compared.Two neomycins have been identified, B and C; their chemical structures are similar, although not yet fully defined, and both are biologically active, but their relative activities differ for different test organisms. .Nearnine, or neomycin A, which is a degradation product of either material may also be present, as may other hitherto unidentified biologically active components. This complex, normally called “neomycin BJJ may contain anything from 80 per cent. upwards of neomycin 13, perhaps up to 15 per cent. of neomycin C, 1 or 2 per cent. of neamine and traces of the other biologically active components resembling neomycins. The biological assay of such material poses quite a problem; in fact, an accurate and reproducible assay is fundamentally impossible.If the standard of comparison has exactly the same composition These implications may be illustrated by reference to neomycin B.[Vol. 86 as the material being tested, valid comparisons can be made, but the biological assay would not be necessary were this fact known in advance. The standard preparation for neomycin currently used in Great Britain has a com- position similar to that described above; commlercial material assayed against it may vary in composition, and this will become obvious in the results obtained by bioassay. Each biologically active component of the complex contributes to the over-all activity; the activity of each component is affected to different degrees by changes in environment, e.g., pH, Eh, composition of medium and temperature, and also by the age of the test organism.When comparisons are made against the standard, the result obtained will therefore vary from assay to assay, and difficulty is experienced in objtaining agreement between successive assays in either the same or different laboratories. There is also difficulty in getting statistically valid assays with parallel dose-response lines. Chemical and physical methods of control are not specific and may not show up the variations in composition, so that disagreement arises between chemical and biological findings. The inability of bioassayists to obtain valid and reproducible assays from time to time and from place to place is too often inter- preted as a failure in technique, and much wasted time and effort is directed to correcting this.The shortcomings of the assay method are, however, fundamental and cannot be overcome by manipulating the technique. By carefully controlling some of the variables known to be important, such as strain of test organism and composition of medium, precise and reproducible results may be obtained for a while, but this situation cannot last. If it were possible to control the assay in this way, then the potency of the drug could be measured directly in terms of the response of the biological system, without reference to a standard; such an approach has always proved abortive. The attempts to standardise assay conditions so that the biological activity of unlike substances can be reproducibly compared cannot be better illustrated than by the attempts that have been made for over 30 years to define conditions for the Rideal - Walker test, so that a wide range of disinfectants could be compared with phenol.It is a safe prediction that attempts will still be in progress after a further 30 years. The discrepancies in bioassay of such materials as current neomycin have to be accepted. Instead of trying to ignore the implications, it should be recognised that these very discrepan- cies afford valuable evidence. When two samples of neomycin are found to have non-parallel dose-response lines, it shows that they have different compositions. This difference is not eliminated by adjusting conditions so that the dose-response lines are parallel. The correct approach is to develop methods of separating the active components before determining them.Here, the bioassayist is dependent on the chemist. Too often this seems to be a challenge that is not met or even recognised. During attempts to develop methods of separation and purification, biological assays, with their internal evidence, prob- ably give a much more sensitive indication of‘ progress than do changes in chemical and physical criteria. To be able to obtain reprod.ucible and precise biological comparisons of two samples under varied conditions of environment and test organism is probably as good evidence of their common identity as is a mixed melting-point. An obvious conclusion to be drawn is that, when “perfect” biological assays can be obtained, the material will be so pure that there will be no need to use them, and it will be possible to exert analytical control by chemical and physical methods.It has been suggested that the ultimate aim of bioassayists concerned with analysis should be the abolition of bioassay. As will be seen later, this situation has arrived for certain of the vitamins and almost for certain antibiotics. m e n this stage has not been reached, biological assay must be used, whatever its limita- tions, since it is needed to control the quality of material bought and sold and to ensure that substances used clinically have a defined activity. In defining quality for the purposes of the Therapeutic Substances Regulations or for conformity to B.P. standards, account is taken of the difficulties in assay when the acceptable minimum potency is laid down.The limits of error acceptable in the assay are also related to the difficulties of assay; thus, in the British Pharmacopoeia, fiducial limits of SO to 135 per cent. for the stated potency of penicillin are recognised, whereas for corticotrophin, fiducial limits of 50 to 200 per cent. of the stated potency are considered satisfactory. For heterogeneous materials controlled biologically and assayed against a heterogeneous standard, it must be recognised that there is no true potency for any particular sample. A sample will have a family of potencies depending on the conditions of assay. These may be distributed about a mode, but the modal value has 220 LIGHTBOWN : BIOLOGICAL STANDARDISATIONApril, 19611 AND THE ANALYST. A REVIEW 22 1 no intrinsic superiority over any individual value.Probably, the potency obtained under conditions of assay most closely resembling the conditions under which the material is to be used would be the most useful. These conditions are, however, virtually indeterminable for biological products intended for use in man. They might theoretically be determined for products assayed on the whole animal, but with antibiotics there is little guide to the true physiological conditions under which the drug acts in v i m ; even such factors as pH and oxygen tension cannot be defined, and both these factors could markedly affect the potency obtained in a biological assay of, for example, neomycin. When products giving variable potencies on assay are to be bought and sold, obvious difficulties will arise if the unit is used directly as a unit of quantity instead of a unit of activity.The drug should be dealt with on a weight basis, with the specification that it should meet certain minimum requirements. These should be framed so as to take into account the possible variation, as with B.P. minimum requirements. INTERNATIONAL CO-ORDINATION OF BIOLOGICAL STANDARDISATION Ehrlich’s original standard reference preparation was distributed internationally by him up to 1914, thereby allowing agreement to be reached between different national labora- tories and hence rationalising development and research in this field. After the first World War it became obvious that similar standard preparations were necessary for other drugs, and the League of Nations accepted responsibility for organising their collection and distribution. The service of developing and supplying the standards was based mainly on two laboratories, The State Serum Institute, Copenhagen, which was made responsible for international standards for sera and antigens, and the National Institute for Medical Research, London, which was made responsible initially for hormones and later for other non-immunological products.International reference preparations for certain arsenical drugs were maintained by the Frankfurt Institute in Germany until the withdrawal of Germany from the League of Nations in 1935, when these standards were transferred to the Department of Biological Standards at the National Institute for Medical Research. When the League of Nations was disbanded, the responsibility for international biological standards was taken over by the World Health Organisation, which formed an Expert Committee on Biological Standardisation.The committee has met annually in Geneva since 1947. I t has no fixed membership, but draws each year from a panel of international experts, chosen according to the subjects under consideration. The committee decides what International Standards are needed and initiates their setting up. I t has permanent administrative offices in Geneva, but no laboratory facilities. The practical work is organised by the two departments at Copenhagen and Mill Hill, as in the past, with the same division of interests. CLASSIFICATION OF INTERNATIONAL BIOLOGICAL PREPARATIONS INTERNATIONAL STANDARDS- These are preparations to which an international unit has been assigned on the basis of extensive international collaborative studies.Each standard is a unique preparation defining the international unit and is intended primarily to be used for calibrating National Standards. INTERNATIONAL REFERENCE PREPARATIONS- These are preparations that originally served as standards in limit tests for toxicity and minimum therapeutic activity, but, at its twelfth session in 1958 the Expert Committee re-defined3 the preparations as follows : “An International Reference Preparation is a prepara- tion to which an International Unit has not been assigned. The reason for this may be that the completion of a full international study, which must precede the establishment of an international standard, would delay the availability of an international reference preparation for which there is an immediate demand.The reason may also be that an international unit would not serve a useful purpose or that extensive laboratory studies have failed to provide a satisfactory assay method.” When the preparations are likely to be used in biological assay, each is assigned a provisional potency by the department distributing them, usually on the basis of information supplied by the manufacturer.222 LIGHTBOWN : BIOLOGICAL STANDARDISATION [Vol. 86 AUTHOR’S PREPARATIONS- At its fifth and sixth sessions in 1951 and 1952, the Expert Committee on Biological Standardisation decided to set up a new class of substances, “Author’s Preparations,” as defined in the Eighth R e p ~ r t .~ I t was held that there was a need for international reference preparations to facilitate research and development in certain fields involving biologically active substances that might eventually prove useful clinically, particularly in the antibiotic field. It was considered impractical to set up international standards for these substances because of the volume of work involved, but it was decided that the International Centres for Biological Standards would hold and distribute Author’s Preparations. These were to be supplied to the Centres by the authors, already filled into ampoules and in a stable form. It was expected that the authors would also define the unit. Although this type of prepara- tion was set up as the result of a number of relquests, only two substances were submitted up to 1958, namely, primisterin and dextran sulphate, and in 1958 this category was dis- continued when the class of international reference preparations was re-defined.AUTHENTIC CHEMICAL SUBSTANCES- The Expert Committee on Biological Standardisation at its fifth5 and sixth6 sessions in 1951 and 1952 made recommendations that led to the setting up of a new class of international preparations designated as Authentic Chemical Substances. This collection, which is held and issued by the World Health Organisation Centre for Authentic Chemical Substances, in Stockholm, is the responsibility of the Expert Committee on the International Pharmacopoeia. The collection includes substances that, although they can be completely characterised by chemical and physical tests, are in demand as authentic chemicals or as convenient standards for biological assay.A number of international biological standards that have been dis- continued, e g . , oestrone, progesterone, vitamin A and tubocurarine, were transferred to the collection at Stockholm, as was the internationa.1 reference preparation for chloramphenjcol, and these are now available as Authentic Chemicals. These materials clearly do not define a unit of activity; they are simply authentic specimens having a high degree of purity. PREPARATION OF AN INTERNATIONAL STANDARD When a new standard is to be set up, a suitable sample is chosen as a result of preliminary testing. Occasionally, several samples prepared in different laboratories may be blended, e.g., insulin or digitalis.Usually these samples are presented by industrial laboratories or individual research workers. The material is d.istributed in small amounts into individual glass containers and dried in a vacuum over phosphorus pentoxide; the containers are filled with pure dry nitrogen, sealed by fusion of the glass and stored in the dark at -10” C. The material is thus protected as far as is practicable from oxidation, moisture and light and is kept at a low temperature-conditions conducive to the highest stability.’ The new standard is then distributed to a number of laboratories i n different countries for examination and for comparison with any existing national or laboratory standard. As a result of this study, the proposed International Standard is assigned a potency, and the international unit of activity is defined as the activity contained in a given weight of the standard.Other samples can then be compared with the International Standard and their activity expressed in inter- national units per unit of weight or volume. ‘The comparison may be direct or performed through a secondary national standard previously calibrated against the International Standard. The International Standards are u:mally limited in amount and are normally issued only to national laboratories with the intention that they should be used to establish and check national standards. However, it is hoped to set up part of the new International Standard for corticotrophin to be freely available to all laboratories as an International Working Standard.Should the standard consist of the purest sample available or should it be a sample corresponding more closely in composition with the material to be assayed against i t ? In practice both types of standard are used. The International Standard for vitamin D8 serves as a standard for determining vitamin D in a wide range of extremely complex natural materials. The International Standard in this instance was the purest sample available at the time it was set up. The International Standard for b;t~itracin,~ on the other hand, is a complex mixture of bacitracins. It would have been possible to obtain a purer sample of a single The choice of sample for use as Internationd Standard is not always easy.April, 19611 AND THE ANALYST.A REVIEW 223 bacitracin, but most commercial bacitracin, at the time of setting up the standard in 1953, consisted of mixtures of approximately the same composition as the standard. There did not appear to be any immediate likelihood of one particular bacitracin becoming freely available, and therefore the sample chosen for use as an international standard resembled current production material. Supplies of this particular standard will be exhausted in a year or so, and for the second International Standard a sample is being sought that will contain a greater proportion of a single bacitracin, since such material is now more readily available. Successive International Standards for insulin have increased in purity, and a third Inter- national Standard for corticotrophin, now being set up, will be much purer than the first and second.INTERNATIONAL COLLABORATIVE ASSAYS It is of interest to consider the international collaborative assays carried out over the last 15 years in establishing new standards, whether the setting up of a first Internatioiial Standard or of a replacement. The activities of two samples were compared in a number of different laboratories, often by a variety of methods. One of the samples was often the purest that could be currently obtained; the other may have been of a similar degree of purity or, if it had been produced at an earlier period, less pure. The laboratories chosen to take part in the assays were all highly experienced and used techniques with which they were familiar. Of eighteen collaborative studies organised for this purpose and dealing with “Pharmacological Substances,’’ only five produced results that were homogeneous between laboratories.In the other collaborative studies, the variation occurring between laboratories had different degrees of significance. The cause of these variations could not be defined. In the assays on antibiotics, such as penicillin and dihydrostreptomycin, which were highly precise, small weighing and dilution errors may have been sufficient to account for the discrepancies. In the collaborative assay of the International Standard for penicillin, for example, the limits of error for the final potency were less than t-1 per cent.1° Collaborating laboratories are always asked to carry out their assays in a way such that weighing and dilution errors may be assessed, but unfortunately the results received do not always comply with this request. In many collaborative assays the discrepancies are probably due to some degree of heterogeneity between the two samples being compared.This was obviously so in the collaborative study of dextran sulphate, since an attempt was made to assay this material against the International Standard for heparin, both substances being anticoagulants, but chemically unrelated.11 Perhaps the most interesting result of this study was that a proportion of the assays carried out were statistically valid, with no significant departures from parallelism of the two dose-response lines. In a number of col- laborative studies, e.g., on thyrotrophin12 and heparin,13 the variations were greater than would be likely from weighing and dilution errors, and it seems possible that there was heterogeneity between the samples being compared, in spite of the fact that statistically valid assays were obtained within each laboratory.As has been pointed out by Miles,2 statistically valid assays of penicillin G against penicillin K or of streptomycin against dihydrostreptomycin are possible, but the potency ratios obtained may differ with different test organisms. When the results of a collaborative assay are heterogeneous, the choice of a potency for the proposed International Standard is difficult and arbitrary. Formally, the collaborative assay is invalid and should be rejected; in practice, this course is usually avoided.There is often no reason to believe that a further collaborative study would be more successful, and the difficulty in reaching agreement between laboratories emphasises the need for an Inter- national Standard. The value chosen is therefore a compromise arrived at in various ways according to the particular conditions. Even if the collaborative assay is statistically valid, the potency is only estimated within certain limits, and an arbitrary choice of a particular value has to be made. In either event, once the potency of the proposed International Standard has been agreed on and the inter- national unit defined, the uncertainties of the collaborative assay have only a historical interest . DEVELOPMENT OF INTERNATIONAL BIOLOGICAL STANDARDISATION The development of international biological st andardisation up to 1939 was reviewed by Dale14 and up to 1945, in much greater detail, by Gautier.15 This later review, which gives224 LIGHTBOWN : BIOLOGICAL STANDARDISATION [Vol.86 a complete bibliography, describes the work carried out by the Permanent Commission on Biological Standardisation of the League of Nations from 1935 until the functions of the Commission were taken over by the World Health Organisation. It discusses the practical difficulties that arose in this period, many of them, as might be expected, being due to heterogeneity in the materials to be standardised. Much useful and fundamental knowledge came directly from attempts to obtain satisfactory standard preparations and to rationalise the assay procedures. As a direct result of the attempts to standardise gas-gangrene anti- toxins the nature of many of the toxins involved was elucidated and their relative importance in vivo was determined.This work would have had much greater importance during the period of the second World War had it not been for the development of penicillin. The establishment of an International Standard for penicillin was one of the last functions of the League of Nations Commission on Biological Standardisation in 1944.16 When International Biological Standards were reviewed by Dale1* in 1939, there were thirty available. There are now fifty-three established Standards with defined international units of activity, together with some fifty-three International Reference Preparations, which do not involve a unit of activity.17 In 1939 most International Standards were for immuno- logical products, vitamins and hormones.These groups are still represented, but thirteen International Standards and Reference Preparations have been discontinued since 1939, mainly because these substances can now be characterised by chemical and physical means. These include provitamin A, vitamin B (thiamine), vitamin C (ascorbic acid), vitamin A, vitamin E (a-tocopherol), oestrone, oestradiol monobenzoate, androsterone and progesterone. There is now only one vitamin with a unit of activity defined in terms of an international standard, i.e., vitamin D,. An International Reference Preparation for vitamin B,, was established in 1959.18 It was originally to be set up as a standard for biological assay, but this is no longer needed, as Vitamin B,, can readily be standardised by chemical and physical means.The preparation will probably be considered for transfer to the collection of Authentic Chemicals. The only standard antiserum discontinued has been staphylococcus B antitoxin. The advent of penicillin and other antibiotics active against Staphylococcus diverted attention from immunological products for the treatment of S. uzcyeus infections. In recent years the problems produced by antibiotic-resistant strains of this organism have re-focused interest on the staphylococcal toxins and antitoxins, and the Expert Committee on Biological Standardisation is re-considering the need for International Standards and reference prepara- tions for staphylococcal products, including leuc ocidins and antileu~ocidins.1~ In 1939 there was only one standard antigen, namely, tuberculin; now there are nine.They include the International Standard for pertussis vaccine, the first International Standard for a vaccine. The standardisation of potency of vaccines is particularly difficult, although it does not nor- mally concern the analyst; great strides have 'been made in this field in the last 20 years, much of the success being due to work in the United Kingdom. More than half of the reference preparations are antisera, most of which have been set up recently in an attempt to rationalise diagnosis of certain infectious diseases in man. A new class of International Standards that has come into existence since 1939 is for antibiotics, a rapidly expanding group.There are at present ten standards for antibiotics, and nine International Reference Preparations, three of which it is expected will shortly be established as International Standards. Two of the antibiotics, nystatin and amphotericin, are specifically active against mycoses. The twenty or so antibiotics dealt with are only a minority of those discovered since the introduction of penicillin; the policy in the past has been to set up standards only for antibiotics fii-mly established and widespread in their use. One of these, heparin, was established in 1942 ; the others are thyrotrophin, growth hormone and cortico- trophin. The last-named has presented many difficulties of standardisation. The first International Standard, set up in 1950, was replaced in 1955 by the second International Standard,,* there being little difference in purity between these two preparations.A third standard is now in course of preparation; it has appreciably greater purity and should help to improve the assays of this hormone. An International Reference Preparation for human menopausal gonadotrophin has also been esta blished.21 Among the miscellaneous international preparations are those reference preparations serving as standards in limits tests for t0xicit.y or minimum therapeutic activity. Two of the earliest international preparations established were of this nature, for neoarsphenamine Among the hormones four new standards have been created since 1939.April, 19611 AND THE ANALYST.A REVIEW 225 and sulpharsphenamine ; for neither was a unit of activity defined. Similar preparations now exist for the melaminyl trypanocides Me1 B and MSb, established in 1954,22 and for dimercaprol, established in 1952.23 It is perhaps surprising that, in spite of modern analytical techniques, biological methods are still necessary to control the purity of such chemicals. In 1951 an International Standard for tubocurarine was e~tablished~~; however, this was discontinued by 1955, since the material could be controlled by chemical and physical means. In 1958 an International Reference Preparation for pyrogens was e~tablished.~~ Limit tests for pyrogens are widely carried out, and their inadequacies are well appreciated. In 1953 a small-scale collaborative study was organised by the World Health Organisation. Two samples of pyrogenic material were compared in a number of laboratories. Great difficul- ties were experienced in obtaining reasonable dose-response curves or reproducible results from day to day and from laboratory to laboratory. I t was impossible to make a valid quantitative comparison of the activities of the two preparations; this was probably due to heterogeneity between the samples, one of which was prepared from Proteus vulgaris and the other from Serratia marcescens.It was soon recognised that the standards for toxicity of neoarsphenamine and sulpharsphenamine could not be used for other arsphenamines, although they were fairly closely related chemically. It is perhaps not surprising that even crude comparisons of different pyrogens proved impossible.The International Reference Preparation for pyrogens remains, however, as a common reference point for workers in different laboratories. The preparation has no defined unit of activity and is made available in the hope that those laboratories using the material will submit information and comments about its use. The results so accumulated may eventually point the way to better methods for controlling this activity. A list of the available International Standards, Reference Preparations and Authentic Chemicals is given in the Appendix to this review (p. 229). EXPRESSION OF BIOLOGICAL ACTIVITY IN WEIGHT UNITS The concept of a unit of biological activity based on a defined weight of a unique prepara- tion was generally accepted as both meaningful and useful by analysts concerned with biologi- cal products at the time that the first antibiotic standard was developed.The unit of activity for penicillin was defined in the traditional way and permitted research, control and therapy in the penicillin field to develop without confusion. With the advent of streptomycin came the use of the “microgram equivalent” notation to express the potency of the product. This method of expressing activity, which has developed widely in the U.S.A. in particular, has produced much confusion, not least in the analytical field. The aim was to express the purity of a sample of antibiotic in terms of the number of micrograms of theoretically pure material contained in 1 mg of the sample. Estimates of purity were made in a biological assay by comparison with a standard preparation, this standard preparation having its own potency expressed as micrograms of theoretically pure material per milligram of standard.In the U.S.A., the standard preparation is usually a unique preparation held by the Food and Drug Administration Laboratories or the United States Pharmacopoeia. Its “potency” is defined on the basis of the best estimate of purity made by chemical and physical means, either directly on the standard itself or on a previous standard, perhaps a manufacturer’s Master Standard, used to establish the first U.S.A. National Standard. The assessment, by chemical and physical means, of the purity of the standard preparation is often difficult; after all, it is precisely because of this difficulty that a biological standard is required. However, an assessment is made, sometimes when the structure of the anti- biotic is not fully known, and the material may only be available in an amorphous form.A sample of the highest activity obtainable is taken, determinations are made of known impurities, such as salts, water and solvent, and it is then assumed that the remainder is pure antibiotic. Thus, if the total of known impurities is x per cent., the sample contains lO(100 - x) pg of pure antibiotic per mg. The Expert Committee on Biological Standardisation at its third session26 discussed the question of alternative methods for expressing potency; it was stated “that the expression of potency in gram-equivalents is valid, though not always desirable, when the active principle in the standard preparation is known to be homogeneous and free from inert material, and the active principles in preparations to be assayed may be heterogeneous.” However, this nomenclature has been little used, and in fact the microgram notation used in the U.S.A.226 LIGHTBOWN : BIOLOGICAL STANDARDISATION [Vol.86 is based on a different concept. The potency of an unknown sample may be expressed in terms of a standard reference preparation as .v microgram equivalents per mg, indicating that 1 mg of the unknown contains the activity of x pg of the standard preparation. This notation has meaning; if the unit of activity is defined as the activity of 1 pg of the standard reference preparation, then 1 unit is equivalent to 1 microgram equivalent.Both are measures of activity. In the U.S.A., potency is expressed not in microgram equivalents per milligram, but in micrograms per milligram. The potency has become a measure of weight content, not activity, and the weight measure is of a theoretical substance, not of a standard preparation. The difficulties of assessing the purity of the standard in absolute terms are considered relatively unimportant in the U.S.A. If it becomes obvious that a mistake has been made, as when samples are found on assay to have a higher than “theoretical” content, then the potency of the master standard is adjusted in the light of the new kn~wledge.~’ One of the aims of unit notation is to maintain continuity; 1 unit of the present International Standard for diphtheria antitoxin has the same activity as 1 unit of the first International Standard for this substance, established in 1922.The aclivity of “1 pg” may be changed from year to year in the hope that each change will be the final one. It is argued that in most instances the first estimate is usually close enough for all practical purposes and that, if re-evaluations have to be made, they are insignificant in clinical importance. This is probably often so, but it does not justify the resultant ambiguity and confusion. Further complications arise from the fact that no uniform policy was adopted in choosing content of base, acid or salt to express the potency of the standard preparation of several antibiotics. In some instances the “potency” of the standard is expressed in terms of theoretical base content, e.g., streptomycin, dihydrostreptomycin and oxytetracycline, or acid, e.g., novobiocin.In others it is expressed in terms of content of a particular salt, e.g., aureomycin hydrochloride and tetracycline hydrochloride. In the last-named instance this leads to figures being quoted for the potency of “theoretically pure” base, which are difficult to comprehend; thus, the potency of theoretically pure tetracycline base is 1082 pg per mg and of chlortetracycline base is 1076 pg per mg.28 The relationship between the inter- national unit and the “microgram” for the International Standards and Reference Prepara- tions is shown in the Appendix to this review (Tables I11 and IV, respectively).When a statement is made that a sample contains 1600 units of penicillin per mg or a solution 100 units per ml, the meaning is clex and unequivocal, whereas the statement that a sample contains 650 pg of neomycin per nig or a solution 100 pg per ml may give rise to doubt. Does this refer to a weight of neomycin sulphate or neomycin base? Does it refer to a weight of neomycin B or C or a mixture of both? To add to the confusion, it was realised at one stage that the potency of the U.S.A. Standard for neomycin was based on a false assessment of the purity, and its value in micrograms per milligram was changed in 1953. Much of the value of Waksman’s monograph on n e ~ m y c i n ~ ~ is lost because of the difficulty of knowing in a particular context what is meant by x p g of neomycin.One microgram has an internationally accepted meaning, viz., one millionth part of a gram; it is a unit of weight. Although the aim of “microgram” notation is to express the content of active ingredient of an antibiotic on a weight basis, this is impossible when the active ingredient cannot be defined, and so the unit of weight is being used as a unit of activity; in such circumstances only the author of a statement relating to amounts of this antibiotic can know what he means by x pg. The main reason put forward in favour of microgram” notation for expressing potency of antibiotics is that the physician is used to prescribing on a weight basis and is confused by units. The procedure used in the British Pharmacopoeia in certain instances, however, allows dosage in units of weight, so accommodaing the physician, but exercises control of quality in terms of units of activity. The mininium permitted potency is expressed in units per milligram, e g ., 900 units per mg for chlortetracycline hydrochloride and erythromycin. The dosage is expressed in weight of B.P. material. This method may be used when the antibiotic is homogeneous and when the minimum permitted potency represents a high degree of It is indeed possible that a given dose may vary in its content of active ingredient, but a variation of 5 or even 10 per cent. is therapeutically unimportant. This would be the maximum variation that could exist with tablets of chlortetracycline or oxytetracycline of B.P. quality. When it is possible to define quality in terms of weight of a defined chemical substance, then the need for a biological standard with a defined unit of activity no longer exists.ThisApril, 19611 AND THE ANALYST. A REVIEW 227 situation has been reached for phenoxymethylpenicillin. The B.P. monograph for this antibiotic does not make use of bioassay and defines the minimum permitted quality in terms of C,,H,,N,O,S ; control is exercised through chemical and physical tests. EXPRESSION OF BIOLOGICAL ACTIVITY IN INTERNATIONAL UNITS It is common practice to refer to the potency of a biologically standardised drug in units per milligram or millilitre without any qualification. The Therapeutic Substances Regulations require that, in the United Kingdom, units of activity shall be the international units when these exist, but it is rare for a labelled potency to be stated in international units or with the recognised abbreviation, i.u.In the U.S.A., potencies are often expressed in U.S.P. units; these are again usually identical with the appropriate international unit, but this is not always made clear. Some of the advantages of having an international unit are lost in this way, and it would be worth-while for units to be described as international units when appropriate. The expression of activity simply in the form of “international units” abbreviated to i.u. is in itself not particularly satisfactory. There are many international units, e.g., for weight, temperature, heat, electrical resistance; these all have specific names, so that confusion is avoided.In biological standardisation we deal with international units of activity; unfortunately, the qualification “of activity” is often omitted, and reference is usually made to international units of the substance, implying a quantitative measurement of the material. It would obviously be difficult to use a specific name for international units of activity, since all the units of activity are unrelated, but the recognition of an abbreviation i.u.a. instead of i.u. would make the position clear. The International Union of Biology and the International Union of Pure and Applied Chemistry have formed an International Enzyme Commission, one of whose terms of reference is the definition of international units of activity for enzymes.31 The international unit is to be the amount of an enzyme that will under certain standard conditions bring about the conversion of 1 pmole of substrate per minute.I t seems unlikely that it will be possible to define units of activity in this way for most of the enzymes used therapeutically, but the possibility can readily be foreseen of confusion arising if two international units of activity are defined-one on the basis of a material standard and the other on the basis of standard conditions-the two being unrelated, but both abbreviated i.u. or u. Such a situation might arise with trypsin and chymotrypsin. Although it would be difficult to use a specific name for international units based on material standards, since these are not related to one another, it might be desirable to introduce a specific name for the international unit of activity proposed by the Enzyme Commission, as these units will be directly comparable one with another.FUTURE DEVELOPMENTS In the past 10 years, International Standards for many vitamins and hormones have been discontinued, and it is likely that a number of antibiotic standards will soon follow suit. In 1959 the Expert Committee on Biological Standardisation considered32 a French proposal that the international unit of activity for benzylpenicillin should be discarded as a basis of dosage for this material.33 It also considered the possibility of discontinuing the Inter- national Standard for benzylpenicillin and transferring it to the collection of Authentic Substances. Of eleven different National Control Laboratories whose opinions were sought on this point, only five were in favour of discontinuing the Standard, and the Committee decided not to make a change.Even when an International Standard becomes unnecessary, biological assay may stiU be used, -because of its greater sensitivity. It it unlikely, for example, that chemical or physical methods for estimating antibiotics in animal feeding stuffs or in blood serum will be a practical possibility in the near future. It might be expected, however, that the use of the international unit would cease under these conditions and that dosage and so on would be expressed in terms of weight. Unfortunately, the use of units may be so firmly established that such a change is difficult. This has been the situation with vitamin A, so that, at the request of the International Union of Pure and Applied Chemistry, the Expert Committee on Biological Standardisation re-defined the unit of activity for vitamin A as being the activity of 0.000344 mg of pure all-trans vitamin A acetate.= This therefore became the first international unit of biological activity defined solely in terms of the pure substance and not in terms of a Standard Preparation. This procedure, which can only be justified228 LIGHTBOWN BIOLOGICAL STANDARDISATION [Vol.86 on grounds of expediency, might cause confusion, and it is to be hoped that official publications, such as pharmacopoeias, will change from the use of units as soon as this is possible. New methods of biological assay are continually being developed, some of which may be used routinely for analytical purposes in the future.Naturally occurring anti-tumour substances can be assayed by plate-diffusion techniques with selected mutant strains of bacteria that are reputed to resemble tumour Anti-viral agents, such as interferon, may also be assayed by a similar technique, in which the test organism is a layer of tissue cells infected with Of more immediale interest to the analyst are the biological methods of assaying insecticide residues in foodstuffs, recently the subject of a report by Needham.37 These methods are more sensitive than available physico-chemical methods, and there is promise of obtaining specific assays by the use of strains of test organisms rendered resist ant to selected insecticides. In the biological standardisation of antisera, basic methods have changed little since the first International Standard was established. However, methods of analysis based on immuno-electrophoresis, which are currently used in immunological research , may well make possible the quantitative measurement 'of specific antibodies by essentially physical methods.38 Although new methods of biological standardisation should extend the scope of the analyst, practice of even well established assay techniques is often neglected. Most analysts using bioassay techniques in this country are probably working in laboratories associated with the manufacture of the drugs, and many of the methods now widely used have originated in these laboratories.The procedures of biological standardisation in analysis have been avoided by most laboratories dealing with control of the quality of drugs under the Food and Drugs Act, probably because of high cost and because the discipline is strange to the traditional analyst.In particular, the need for statistical evaluation of the results seems to create a barrier restricting the use of these methods. Many biological-assay methods are expensive, particu- larly those requiring the use of experimental animals, but others, e.g., antibiotic assays, are not. The precision that can be obtained in bioassay of antibiotics is as great as that obtainable in analysis of most drugs by chemical and physical means. The methods of statistical evalu- ation required are relatively simple and well d.ocumented. The control of quality of those biologically active drugs scheduled under the Therapeutic Substances Act is adequately exercised.Certain similar drugs supplied under the National Health Service are also carefully controlled , but many important drugs requiring biological assay are sold t o the public without any safeguard, except the integrity of the manufacturer. Legislation relating to the control of drugs is at present under review, but this deficiency in existing control is due not so much to inadequacy of the present laws, as to inadequacy of existing laboratories responsible for testing under the Food and Drugs Act. It would be unreasonable to expect every such analytical laboratory to be equipped to deal with the full range of biological assays, but there is no reason why regional laboratories should not perform this function, each specialising in a different field.1. 2. 3. 4. 5. 0. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. REFEFIENCES Jerne, N. K., and Wood, E. C., Biornetrics, 1949, 5, 273. Miles, A. A., Bull. World Health Organisation, 1952, 6, 131. Expert Committee on Biological Standardisation, Twelfth Report, World Health Organisation -, Eighth Report, Ibid., 1955, 96, 18. -, Fifth Report, Ibid., 1952, 56, 12. -, Sixth Report, Ibid., 1953, 68, 25. Jerne, N. K., and Perry, W. L. M., Bull. World Health Organisation, 1956, 14, 167. Coward, K. H., and Irwin, J. O., Ibid., 1964, 10, 875. Humphrey, J . H., Lightbown, J . W., Mussett, M. V., and Perry, W. L. M., Ibid., 1953, 9, 861. Humphrey, J . H., Mussett, M. V., and Perry, W.L. M., Ibid., 1953, 9, 15. Mussett, M. V., and Perry, W. L. M., Ibid., 1956, 14, 647. -- , Ibid., 1955, 13, 917. Baniham, D. R., and Mussett, M. V., Ibid., 1959, 20, 1201. Dale, Sir Henry, Analyst, 1939, 64, 554. Gautier, R., Bull. Health Organisation, League of Nations, 1945, 12, 1. Commission on Biological Standardisation, Ibid., 1945, 12, 181. Expert Committee on Biological Standardismation, Thirteenth Report, World Health Organisation Tech. Re+. Ser., 1959, 172, 20. Tech. Rep. Ser., 1960, 187, 26.April, 19611 AND THE ANALYST. A REVIEW 229 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. Bangham, D. R., and Mussett, M. V., Bull. World Health Organisation, 1960, 22, 555. Expert Committee on Biological Standardisation, Thirteenth Report, World Health Organisatioiz Mussett, M.V., and Perry, W. L. M., Bull. World Health Organisation, 1956, 14, 543. Expert Committee on Biological Standardisation, Thirteenth Report, World Health Organisation -, Eighth Report, Ibid., 1955, 96, 16. -, Sixth Report, Ibid., 1953, 68, 18. Davies, M. G., Miles, A. A., and Perry, W. L. M., Bull. World Health Organisation, 1949, 2, 65. Humphrey, J. H., and Bangham, D. R., Ibid., 1959, 20, 1241. Expert Committee on Biological Standardisation, Third Report, World Health Orgunisation Tech. Humphrey, J. H., Lightbown, J. W., and Mussett, M. V., Bull. World Health Organisation, 1959, Grove, D. C., and Randall, W. A., “Assay Methods of Antibiotics,” Medical Encyclopaedia Inc., Waksman, S. A., Editor, “Neomycin,” Ballikre, Tindall & Cox Ltd., London, 1958.Miles, A. A., and Perry, W. L. M., Bull. World Health Organisation, 1953, 9, 1. International Commission on Enzymes, Nature, 1958, 181, 452. Expert Committee on Biological Standardisation, Thirteenth Report, World Health Organisation Commission d’gtudes des Antibiotiques, Thkrapie, 1959, 14, 9. Expert Committee on Biological Standardisation, Thirteenth Report, World Health Organisation Gause, G. F., Biol. Rev., 1959, 34, 378. Porterfield, J. S., Lancet, 1959, ii, 326. Needham, P. H., Analyst, 1960, 85, 792. Grabar, P., Adv. Protein Chem., 1952, 13, 1. Tech. Rep. Ser., 1960, 187, 19. Tech. Rep. Ser., 1960, 187, 9. Rep. Ser., 1950, 2, 12. 20, 1221. New York, 1955. Tech. Rep. Ser., 1960, 187, 7. Tech. Rep. Ser., 1960, 187, 10.NOTE-Annual Reports of the Expert Committee on Biological Standardisation of the World Health Organisation can be obtained from Her Majesty’s Stationery Office, London. Since the Tenth Report, published in 1957, an Appendix has been included; this gives a complete list of Inter- national Biological Standards and Reference Compounds available from the two Control Centres and also contains a complete bibliography relating to these materials. Received November 21st, 1960 APPENDIX TABLE I INTERNATIONAL STANDARDS, EXCLUDING ANTIBIOTICS Antigens- Old tuberculin Tetanus toxoid, plain Diphtheria toxoid, plain Diphtheria toxoid, adsorbed Schick test toxin (diphtheria) Pertussis vaccine Swine erysipelas vaccine Tetanus antitoxin Diphtheria antitoxin Antidysentery serum (Shiga) Gas-gangrene antitoxins- Antibodies- Clostridiurn welchii, types A, B and D Vibrion septique Oedematiens Histolyticus Sordelli Staphylococcus a-antitoxin Scarlet fever streptococcus antitoxin Anti-streptolysin 0 Swine erysipelas serum (anti-N) Antipneumococcus serum, types 1 and 2 Anti-Brucella abortus serum Anti-Q-fever serum Anti-rabies serum Anti-A blood-typing serum Anti-B blood-typing serum Syphilitic human serum Hovmones- Oxytocic, vasopressor and antidiuretic sub- stances (previously named posterior pituitary lobe) Prolactin Corticotrophin (previously named adrenocortico- trophic hormone) Thyrotrophin Growth hormone Serum gonadotrophin Chorionic gonadotrophin Insulin Heparin Miscellaneous- Vitamin D, H yaluronidase Digitalis230 LIGHTBOWN [Vol.86 TABLE I1 INTERNATIONAL REFERENCE PREPARATIONS, EXCLUDING ANTIBIOTICS A ntigens- Hoymones- Cholera antigen (Inaba) Cholera antigen (Ogawa) Cholera vaccine (Inaba) Miscellaneous- Cholera vaccine (Ogawa) Cardiolipin Vitamin B,, Lecithin (beef heart) Neoarsphenamine Lecithin (egg) Sulpharsphenamine Oxoarsphenamine Antibodies- Me1 B Cholera agglutinating serum (Inaba) MSb Cholera agglutinating serum (Ogawa) Dimercaprol Diphtheria antitoxin for flocculation test Antityphoid serum (provisional) Pyrogen Antipoliomyelitis sera, types 1, 2 and 3 Anti-Leptospira! sera, representing 19 strains Human menopausal gonadotrophin Opacity reference preparation Pro tamine TABLE I11 INTERNATIONAL STANDARDS FOR ANTIBIOTICS Calculated purity Defined potency, Equivalence of 1 i.u. of Standard on basis Substance i.u. per mg to American “pg” ofiAmerican “pg,” % Penicillin (sodium salt) . . .. 1670 Not used > 99* Phenoxymethylpenicillin (free acid) . . 1695 Not used > 99” Streptomycin (sulphate) . . .. 780 1 “pg” of base 97.5 Tetracycline (hydrochloride) . . .. 990 1 “pg” of hydrochloride 99.0 Chlortetracycline (hydrochloride) . . 1000 1 “pg” of hydrochloride 100 Erythromycin (base) . . .. I . 950 1 “pg” of anhydrous base 95 Dihydrostreptomycin (sulphate) . . 760 1 “pg” of base 95.1 Bacitracin . . .. .. .. 55 Not used Not known Oxytetracycline (base dehydrate) . . 900 1 “pg” of anhydrous base 97.1 Polymixin B . . .. .. . . 7874 Not used Not known * Independent: estimate. TABLE IV INTERNATIONAL REFERENCE PR:EPARATIONS FOR ANTIBIOTICS Provisional potency, Substance units per mg Amphotericin B . , .. ,. Kanamycin (sulphate) .. .. Vancomycin . . .. .. .. Viomycin (sulphate) . . .. .. Neomycin (sulphate) . . .. .. Nystatin . . .. .. .. Novobiocin (sodium salt) . . .. Penicillin Kt . . .. .. .. Oleandomycin (chloroform adduct) . . 960 812 1007 7 30 680 2855 835 845 - Equivalence of Calculated purity of provisional unit to American “pg” Reference Preparation on basis of American “pg,” % 1 ( ( p g l l 96 1 “pg” of free base 97.4 Not used Not known 87 * Not known Not used Not known 86.5 97.3 1 “pg” of free base 1 “pg” of free base 1 “pg” of free acid 1 “pg” of free base - - * On the assumption that viomycin is (&basic. No unit; preparation is intended for use as a marker in chromatography. DISCONTINUED INTERNATIONAL STANDARDS AVAILABLE AS AUTHENTIC CHEMICALS Oestrone; progesterone ; vitamin A (acetate) ; tubocurarine (D-tuboCUrarine chloride) ; chloramphenicol

ISSN:0003-2654    

DOI:10.1039/AN9618600216

出版商:RSC    

年代:1961

数据来源: RSC

摘要:

April, 19611 HOLNESS 231 The Determination of Citronellol in Admixture with Geraniol Further Studies of Formylation Reactions by Gas - Liquid Chromatography BY D. HOLNESS Work undertaken on behalf of the Essential Oils Sub-Commiteee of the Analytical Methods Committee (Pvoprietary Perfumes Ltd., 117-121 Grange Road, London, S.E.l) Previous studies of hot formylation proved tedious and difficult, owing to the lack of a suitable technique for investigating the course of the reactions and the nature and proportions of the end-products. This paper describes experiments in which gas - liquid chromatography was used for that purpose and from which it was concluded that formylation cannot be made the basis of an accurate analytical method no matter what changes be made in working conditions.WHEN the method known as “hot formylation” was first used, it was supposed that boiling anhydrous formic acid converted citronellol to its formate and geraniol to hydrocarbons,l so that the ester value of the fonnylated material could be used to determine the percentage of citronellol in an essential oil containing both citronellol and geraniol. Later work indicated that these assumptions were not wholly true. Pfau2 reported that the chief reaction products from citronellol were the mono- and diformates of 3,7-dimethyloctane-1 ,7-diol. It was also found that polymers of uncertain composition could originate from the alcohols. A previous communication3 from the Essential Oils Sub-committee of the Analytical Methods Committee of the Society for Analytical Chemistry reported results of comparative tests with four variants of the method. Method A, under consideration by the International Organization for Standardization, gave answers nearest to the true figures, but it was judged too inacciirate to justify its recommendation for determining citronellol in essential oils.Changes in working conditions obviously caused significant variations in the results, but it was not possible to deduce from available information whether some specific conditions existed under which a hot-formylation method could be relied upon to give results of acceptable accuracy. The same report mentioned that gas - liquid chromatograms of the formylated samples had shown peaks that might correspond with the substances described by Pfau, although they failed to support his views about their relative proportions.In this investigation, the first objective was to determine the nature of the compounds giving rise to the principal observed peaks. Gas - liquid chromatograms were then used to trace the effects of variations in reaction conditions and to relate changes in the observed proportions of reaction products to deviations from the correct figures in analytical results. The findings of these experiments were used to decide whether or not the principle of hot formylation is a sound basis for an analytical procedure for determining citronellol in essential oils. Fig. 1 shows a chromatogram of the more volatile end-products after a mixture of equal weights of geraniol and citronellol had been treated by method A.3 These were followed by a number of compounds having low volatility, which gave rise to a series of low drawn-out peaks.It was evident from the retention times of the peaks traced that the assumptions that geraniol is converted almost quantitatively to hydrocarbons and citronellol to its formate could not both be true, and consideration of the areas under the peaks in relation to the known composition of the mixture suggested that neither assumption was correct, but that both alcohols undergo complex changes. Consequently, it was decided to study the reactions and reaction products of citronellol separately from those of geraniol. In all experiments, specially purified samples were used, whose content of citronellol or geraniol exceeded 99 per cent., as determined by acetylation and by gas-liquid chromatography.The formic acid used was of analytical-reagent grade.232 HOLNESS : THE DETERMINATION OF CITRONELLOL I N ADMIXTURE [VOl. 86 REACTION PRODUCTS- CITRONELLOL A 10-ml portion of citronellol and 20 ml of 98 per cent. formic acid were heated together at boiling-point for 1 hour. The washed product had an ester value of 359, equivalent to 118 per cent. of citronellyl formate. Acid diluted with water to 76 per cent. gave a citronellyl formate equivalent of 93 per cent. Chromatograms of these two reaction mixtures are shown in Figs. 2 and 3. Peaks (1) and (2) have the retention times of citronellyl formate and citronellol. Peaks (4) and (5) may be produced by the mono- and diformates of 3,7-dimethyl- octane-1,7-diol.Saponification of the second reaction mixture gave a product whose chrom- atogram is shown in Fig. 4, in which peak (2) corresponds to citronellol and peak (3) is pre- sumably produced by 3,7-dimethyloctane-l,7-diol. Fig. 2 Fig. 3 Fig. 1. A mixture of equal parts of citronellol and geraniol after hot formylation Fig. 2. Citronellol after reaction with 98 per cent. formic acid at boiling-point for 1 hour Fig. 3. Citronellol after reaction with 75 per cent. formic acid at boiling-point for 1 hour In Figs. 1, 2 and 3, (S) = start, (T) = a group of terpene hydrocarbons, (I?) = the start of a series of compounds of very low volatility, (1) = citronellyl formate, (2) = citronellol, (4) = glycol monoformate, ( 5 ) = glycol diformate Next, 200g of citronellol were formylated with 99 per cent. formic acid.The product was saponified, after which it gave a chromatogram similar to Fig. 4. It was then fractionally distilled under reduced pressure, and the two major components were separated. The first collected was identified as citronellol. The other gave the chromatogram shown in Fig. 5, which indicated a purity of about 97 per cent. Its infra-red spectrum was consistent with the supposed glycol structure. Comparison with the spectrum of citronellol confirmed the increase in hydroxyl content and the loss of the isopropylidene linkage. Its chemical behaviour strongly indicated the presence of a tertiary hydroxyl group, which could hardly arise except in the 7-position. A comparison was made between this compound and a sample (also having a purity of about 97 per cent .-see Fig.6) of 3,7-dimethyloctane-l,7-diol prepared by hydrogenation of “hydroxycitronellal,” which is known to be 7-hydroxy-3,7-dimethyloctan-l-al. Their refrac- tive indexes were 1.459 and 1.460, respectively, their retention times with silicone oil, Apiezon L and polypropylene sebacate as stationary phases were identical and their infra-red spectra were nearly indistinguishable. The correctness of the supposed structure of the compound giving rise to peak (3) seemed to be adequately established. Attempts to prepare the formates from the glycol led to chromatograms somewhat resembling Fig. 2, with three peaks whose retention times equalled those of peaks (l), (4) and (5). Peak (1) is produced by citronellyl fclrmate, which suggests that addition to the isopropylidene linkage is reversible.Ester values were highest when the relative size ofApril, 19611 233 peak (5) was greatest, from which it would seem that peak (5) corresponded to the diformate. The compound giving rise to peak (4) was isolated by fractionally distilling a batch of formylated citronellol under reduced pressure. The first portion of distillate consisted mainly of citronellyl formate. From later fractions, a sample whose chromatogram indieated that it contained about 95 per cent. of the compound giving rise to peak (4) was selected. It had an ester value of 282; clearly, peak (4) is produced by the monoformate (ester value 278). Too little material remained in the flask to permit distillation of the diformate without con- siderable decomposition.It was not considered important to identify the minor peaks on chromatograms of formylated citronellol. WITH GERANIOL : FURTHER STUDIES OF FORMYLATION REACTIONS ( 3 1 Fig. 4 Fig. 5 Fig. 6 Fig. 4. Fig. 5. Fig. 6. In Figs, 4, 5 and 6, in addition to symbols used in Figs. 1, 2 and 3, (3) = 3,7-dimethyloctane-l,7-diol Citronellol after hot formylation and then saponification Glycol obtained after fractional distillation of the mixture from Fig. 4 Purified 3,7-dimethyloctane-l17-diol obtained by hydrogenation of “hydroxycitronellal” EFFECTS OF REACTION CONDITIONS- The chromatograms in Figs. 2 and 3 illustrated how greatly a change from 98 to 75 per cent. formic acid altered the composition of the reaction product, so six similar experiments were carried out in which the concentration of acid was increased in even steps from 76 to 99 per cent.In each experiment, 10 ml of citronellol and 20 ml of acid were mixed and gently boiled for 1 hour, after which the ester value was determined; the results are shown in Table I. TABLE I EFFECT OF CONCENTRATION OF FORMIC ACID Concentration of Experiment formic acid, 1 75 2 80 3 85 4 90 5 95 6 99 No. % w/w Ester value found 287 299 311 323 342 362 Equivalent, as citronellyl formate, % 94 98 102 106 112 119 Chromatograms of the reaction products of experiments Nos. 3, 4, 5 and 6 are shown in Figs. 7, 8, 9 and 10, and Fig. 3 shows a chromatogram after an experiment like No. 1. These confirm Pfau’s findings that relatively more glycol diformate is produced as the con- centration of acid is increased, but this occurs mainly at the expense of glycol monoformate and not citronellyl formate. Citronellyl formate is in fact the major component when the234 HOLNESS: THE DETERMINATION OF CITRONELLOL IN ADMIXTURE [VOl.86 concentration of acid used is above 80 per cent.; below this concentration, the glycol mono- formate is found in greatest amount. Hydration of the 6,7 double bond would seem to occur readily when plenty of water is present, whereas subsequent formylation of the 7-hydroxy group (or, alternatively, direct addition of formic acid to citronellyl formate) proceeds best when the acid is least diluted. Some citronellol always remains unchanged, and the amount becomes greater as less concentrated acid is used.The chromatograms therefore reveal three sources of significant error in determinations of citronellol by hot formylation and show that the nearly correct results obtained in experiments Nos. 2 and 3 are only the outcome of a fortuitous balance of positive and negative errors. A true answer cannot be obtained, since no adjustment of the concentration of acid can lead to the postulated single reaction in which citronellol is completely converted to its formate. Fig. 7 Fig. 8 Fig. 9 Fig. 10 Effects of different concentrations of acid on the reaction between citronellol and boiling formic acid: Fig. 7, 99 per cent. formic acid; Fig. 8, 95 per cent. formic acid; Fig. 9, 90 per cent. formic acid; Fig. 10, 85 per cent. formic acid Figs. 7 to 10.In Figs. 7 to 10, symbols used are as for Figs. 1 to 6 Fig. 11 Fig. 11. Fig. 12. In Figs. 11 and 12, symbols used are as for Figs. 1 to 10 Citronellol pZus 99 per cent. formic acid ;after 1 hour a t 60" C Citronellol plus 99 per cent. formic acid ;after 1 hour a t 20" C Fig. 12 Variations in temperature and reaction time were equally ineffective in eliminating these sources of error. Experiment No. 6 was repeated at 60" and 20" C; the results were- Reaction time, minutes . . .. .. .. 10 60 120 Ester value after reaction at 60" C . . .. 356 373 388 Ester value after reaction a t 20" C . . .. 247 346 375April, 19611 WITH GERANIOL : FURTHER STUDIES OF FORMYLATION REACTIONS 235 The chromatograms of all six reaction products resembled those in Figs. 7, 8, 9 and 10 in that the chief components were citronellyl formate and the two glycol formates, although at 20” C the presence of small amounts of other unidentified materials was revealed; two of these chromatograms are shown in Figs.11 and 12. A decrease in the amount of acid used, however, tended to decrease the ester value. An experiment similar to No. 6, but in which only 10 ml of 99 per cent. formic acid were used, gave an ester value of 340 and a chromato- gram like that obtained with 20ml of 95 per cent. acid. Presumably, this is because the water formed during the reaction more effectively dilutes the smaller amount of acid. In practice, ideal conversion to citronellyl formate could be achieved only by slowly adding 99 per cent. formic acid to citronellol at 100” C.So long as citronellol is in excess, the acid seems to be esterified with sufficient rapidity to prevent it from catalysing hydration or additive formylation. Unfortunately, such a procedure is useless for determining citronellol in the presence of geraniol, as geraniol also gives its formate under these conditions. The chromatograms suggested that the proportion of reaction products having low volatility increased with the concentration and amount of formic acid added. Determinations of residues after evaporation supported this view, as shown by the results in Table 11. How- TABLE I1 EFFECT OF AMOUNT AND CONCENTRATION OF FORMIC ACID ON PROPORTION In each experiment a 10-ml portion of citronellol was formylated for 1 hour OF RESIDUE Formic acid Residue A A f I I \ Concentration Volume Proportion, used, yo used, ml yo w/w Ester value 80 10 6 280 80 20 11 316 90 10 6 296 99 10 8 318 ever, the residues still contained glycol formates, which cannot entirely be removed by heating on a steam-bath.The true percentages of “polymers” and their real ester values are lower than those found; nevertheless, small amounts of these compounds do occur, and they may give rise to yet another source of error inherent in hot-formylation methods. GERANIOL REACTION PRODUCTS- When a mixture of geraniol with excess of concentrated formic acid is heated to boiling- point, the geraniol is rapidly transformed into other compounds. The main reactions are complete in so short a time that samples withdrawn at intervals after the liquid has begun to boil give chromatograms that hardly differ from each other.Fig. 13 shows a typical exam- ple, obtained after treatment by method A.3 Its chief feature is a group of peaks (1) that were shown to be produced by unsaturated hydrocarbons. There is a small amount of an unidentified compound, indicated by peak (2). No significant amounts of esters are shown, for similar chromatograms were obtained after the reaction mixtures had been saponified. It is also clear that little or no geraniol is present. At first sight, the chromatographic evidence seems to confirm the supposition that geraniol is converted mainly to hydrocarbons, but closer examination reveals that the areas under the recorded peaks are far too small to account for the full volume of sample placed on the column.Further, although the chromatograms were hardly changed after saponi- fication, the reaction mixtures had ester values greater than 30. Evidently, compounds having low volatility must be produced, some of which must presumably contain formyl groups. This was confirmed, first by running a sample on a non-polar stationary phase a t a higher column temperature, when a group of large peaks having retention times correspond- ing to very low vapour pressures was traced (see Fig. 14), and secondly by determining the proportion of residue after evaporation (88 per cent. w/w) and the ester value (42) of this residue. There is no doubt that, under the conditions recommended for hot-fonnylation analysis, geraniol-like citronellol-fails to react in conformity with the theory upon which the method is based.The previously reported “polymers” are in fact the principal products.236 EFFECTS OF VARIATION IN REACTION CONDITIONS- HOLNESS : THE DETERMINATION OF CITRONELLOL IN ADMIXTURE [Vd. 86 If the reaction temperature is decreased, chromatograms of the volatile end-products are similar to Fig. 13, although the rate of reaction is slower; at 20" C the reaction is sufficiently slow to be followed in stages. Fig. 13 Fig. 14 Fig. 15 Fig. 13. Geraniol treated by method A Fig. 16 Fii. 14. Higher temperatuie chromatogram of hot-formylated geraniol recorded a t slow chart speed Fig. 16. Reaction between geraniol and 99 per cent. formic acid after 10 minutes a t 20" C Fig. 16. Reaction between geraniol and 99 per cent. formic acid after 2 hours ar 20" C In Figs, 13 to 16, (S) = start, (C) = compounds of low volatility, (1) = unsaturated hydrocarbons, (2) = unidentified material, (3) = formate of unidentified alcohol, (4) = geranyl formate, (5) = geraniol ....................................0 0 C ''1, , , , , , , -OD 0.- ._.- ~ _.-__.-._.-._._.._ ~ _.-.-____. 0 -......... I 0 I 2 3 4 24 Reaction time, hours Fig. 17. Effects of time and temperature on the ester values for reaction between geraniol and 99 per cent. formic acid: curve A, 20" C; curve B, 60" C; curve C, on water bath at 100' C; curve D, on sand-bath under reflux A 10-ml portion of geraniol and 20ml of 99 per cent, formic acid were mixed andset aside at 20" C. Samples were withdrawn at intervals, and each was divided into two parts, one for the determination of ester value and the other for chromatography.The peaks of group (1)-the hydrocarbons-were in evidence at the start; so was that for unchanged geraniol, peak (6). Between them appeared several peaks, the largest of which were thoseApril, 19611 WITH GERANIOL : FURTHER STUDIES OF FORMYLATION REACTIONS 237 for geranyl formate, peak (4), and the fonnate of an alcohol that seems to be a cyclic isomeride of geraniol. These peaks are shown in Fig. 15, a chromatogram of a sample taken after reaction for 10 minutes, when the ester value was at its highest. From this time on, all peaks were seen to decline in size, with the exception of those in group (1) and peak (2). After a lapse of 2 hours, none but these remained of the volatile compounds, other than in trace amounts.This is seen in Fig. 16, which closely resembles chromatograms of reactions at boiling-point (e.g., Fig. 13). No further changes were observed during the next 22 hours. Fig. 17 shows graphs in which ester values are plotted against the corresponding reaction times. Curve A, derived from the experiment described above, behaves initially as would be expected from the indications given by chromatograms of the volatile products, in that the ester value increases sharply during the first 10 minutes and then decreases; it does not, however, diminish to zero, but approaches a steady value of 70. Further experiments suggested that a limiting value is approached at all temperatures between 20” C and boiling- point and that the magnitude of this limiting value vanes inversely with temperature.Succesive ester values at 60” C are plotted in curve B of Fig. 17, the steady figure in this instance being 62. On a boiling-water bath it was 42, and tests under reflux over an electric heating mantle gave results between 32 and 34. Evidently, a higher proportion of “polymers” having ester groups is produced as the temperature is decreased. The concentration of formic acid, from 75 per cent. upwards, was found to have far less influence on this ultimate ester value. Samples of geraniol heated with equal volumes or more of 75, 80, 90 or 95 per cent. formic acid at boiling-point all gave values between 32 and 36. Similarly, the concentration of acid had little effect on the volatile end-products. Even a mixture of geraniol with 75 per cent.formic acid at 20” C gave, after being set aside for 24 hours, a chromatogram essentially like those in Figs. 13 and 16. All the experiments point to the conclusion that no adjustment of working conditions can prevent the formation of end-products having significant ester values. In the best circumstances, a pure geraniol must be expected to give an apparent-citronellol content equal to nearly 10 per cent. of its own weight. DISCUSSION OF RESULTS The reactions of citronellol and geraniol are far more complex than the original theory of hot-formylation analysis supposed. From citronellol, some glycol diformate is always produced, and this causes high results at high concentrations of acid. Decreases in the con- centration of acid progressively decrease the content of glycol diformate and increase the proportions of monoformate and unchanged citronellol.The ester value of the glycol mono- formate is slightly lower than that of citronellyl fonnate, and the value for citronellol is nil; a point therefore exists at which a “correct,” but spurious, answer is obtained. Further decreases in concentration of acid yield results lower than the true figure. With geraniol, hydrocarbons are certainly produced in the reaction, but large amounts of several other compounds having low volatility are also formed. One or more of these compounds has an ester value, so that treatment by hot formylation invariably shows an apparent content of citronellol. This, together with the glycol diformate, could cause the high results reported from analyses of the mixed alcohols.Fig. 1 shows that the same principal end-products occur when a mixture of geraniol and citronellol is subjected to hot formylation, although how far the presence of the one may influence the extent of each reaction undergone by the other is not clearly revealed. A check was made by applying the above findings to the figures given in the interim report of the Essential Oils S~b-Committee.~ The four methods tested embodied three variations in working conditions: (a) formic acid at 90 or 100 per cent., (b) the use of 10 or 20 ml of acid and (c) heating on a water bath or sand-bath. It was shown that neither (a) nor (b) has much influence on the ester value for geraniol, but both affect the percentages of the glycol formates from citronellol.Variation (c) is not important for citronellol, but it appreciably alters the result with geraniol. It is therefore deduced that- (i) All four methods should give figures exceeding the true content of citronellol. (ii) Method C should give the highest result, as 20ml of 100 per cent. formic acid favours the biggest yield of glycol diforrnate, and heating on a water bath increases the apparent-citronellol content of geraniol.238 HOLNESS : THE DETERMINATION OF CITRONELLOL IN ADMIXTURE [Vol. 86 (iii) Method A should show the lowest result, a s 90 per cent. formic acid gives the smallest proportion of glycol difonnate, and heating on a sand-bath reduces the ester value of the geraniol. (in) Methods B and D should give results intermediate between those of methods A and C.In method B (10 ml of 100 per cent. formic acid), there will be more glycol diformate than in method D or A, but less than in method C, in which 20ml of 100 per cent. acid are used. On the other hand, the ester value due to geraniol by method D (water bath) will be greater than those by methods B and C (sand- bath). Methods B and D should consequently yield rather similar results; those of method B would tend to be higher when citronellol predominated and that of method D when geraniol predominated. Since these deductions agreed so well with the observed trends in Table I of the interim report,3 a quantitative comparison was made between the results for mixtures and the expected values calculated from the experiments with individual alcohols as described above.By taking 325 and 362 as the means of ester values recorded for citronellol with twice its volume of 90 and 99 per cent. formic acid, respectively, 340 as the value with an equal volume of 99 per cent. acid and 33 and 42 as the respective values for geraniol when heated on a sand-bath or water bath, the results in Table IIC emerged. TABLE III COMPARISON BETWEEN OBSERVED AND EXPECTED RESULTS Method A Method B Method C Method D Citronello1 -- fvAv, content of Citronellol Citronellol Citronellol Citronellol Citronellol Citronellol Citronellol Citronellol mixture, found, calculated, found, calculated., found, calculated, found, calculated, 79 84.9 85 87.8 89 95.1 96 86.0 86 13 20-5 21 24.0 21 25.9 24 24.4 23 37 42.5 42 46.3 44 50.0 49 46.8 44 57 63.2 62 67.1 65 71.7 71 65.5 63 o/, w/w Yo % % % % % % % Within the limits of likely experimental error, these results are sufficiently concordant to strengthen the supposition that neither alcohol greatly influences the reactions of the other with formic acid and that the findings for each one taken separately may safely be extended to mixtures of both.CONCLUSIONS The theory of hot formylation takes no account of several important side-reactions that seem to be unavoidable whatever adjustments be made in working conditions. Every result happening to coincide with the true percentage of citronellol is the outcome of a chance combination of self-cancelling errors. For any given mixture, within a limited range of compositions, a particular concentration of acid might be selected at which a low figure for citronellol would nullify the false positive contribution from geraniol and so lead to the right answer ; obviously, however, the same conditions would not yield a coincident answer with a blend of different proportions. Formylation can never lead to genuinely accurate results, nor can it even be made the basis of a generally applicable method giving superficially correct ones. For instance, the presence of additional alcohols must furthfer complicate the reactions, each alcohol according to its structure, and additive formylation of double bonds in compounds other than citronellol is not beyond the bounds of possibility. Such matters would demand detailed investigation were it not that the experiments dlescribed appear to be sufficiently adequate to condemn the use of formic acid in analytical procedures for determining citronellol in every type of mixture. The decision of the Essential Oils Sub-committee not to support any method of hot formylation is, on the evidence, entirely justified. It is even less reliable as a guide to the citronellol content of essential oils.April, 19611 Xr. D. Welti for his help with infra-red spectra. WITH GERANIOL : FURTHER STUDIES OF FORMYLATION REACTIONS 239 I thank the Directors of Unilever Ltd. for permission to publish this paper and REFERENCES 1. Walbaum, H., and Stephan, K., Ber., 1900, 33, 2302. 2. Pfau, A. S., J . prakt. Chem., 1921, 102, 276. 3. Analytical Methods Committee, Analyst, 1959, 84, 690. Received November 25th, 1960

ISSN:0003-2654    

DOI:10.1039/AN9618600231

出版商:RSC    

年代:1961

数据来源: RSC

摘要:

April, 19611 WITH GERANIOL FURTHER STUDIES OF FORMYLATION REACTIONS 239 The Detection of “ Additional Elements ” in Plastic Materials the Oxygen Flask Combustion Method BY J. HASLAM, J. B. HAMILTON AND D. C. M. SQUIRRELL (Imperial Chemical Industries Ltd., Plastics Division, Bessemer Bead, Welwyn Garden City, Herts.) A method has been devised for detecting “additional elements” in polymers, polymeric compositions and plasticisers. The sample is burnt in oxygen and the products of combustion are absorbed in sodium hydroxide solution. Some of the newer colorimetric methods have been applied to detect such elements as nitrogen, chlorine, fluorine, phosphorus and sulphur in the combustion products. The test is so designed that semi-quantitative information can be obtained about the proportions of these additional elements, particularly when present in the lower concentrations.ONE of the essential preliminaries in the examination of polymers, polymer compositions, plasticisers, etc., in the plastics industry involves the detection of the additional elements, i e . , elements other than carbon, hydrogen and oxygen. In the past this has been accom- plished by means of the sodium fusion test, the aqueous extract of the fusion product being examined for (a) nitrogen by the ferric ferrocyanide test, (b) sulphur by the nitroprusside test, (c) phosphorus by the ammonium molybdophosphate test, (d) halogen (usually chlorine) by the silver nitrate test, after removal of cyanide if necessary, and (e) fluorine by either a colorimetric alizarin or etching test.The sodium fusion test is hazardous and often far from satisfactory when only small proportions of the additional elements are present. It is probably useful in teaching labora- tories, but, in our experience in the plastics industry, the analyst often wishes to know if relatively small amounts of additional elements are present in a material under examination. The ability to detect these small amounts of additional elements will often provide evidence of small proportions of additives and make the analysis of a given material all the more complete and informative. It has to be admitted that on many occasions the sodium fusion test has failed to provide this preliminary guidance. Within the past year or so we have had considerable experience in opening up organic substances, such as polymers, polymer compositions and plasticisers, by combustion in oxygen,lS2 and we have found that electrical firing gives us a simple means of initiating the combustion.The procedure adds no objectionable impurities to the combustion products, and it was thought that if it could be linked with some of the newer colorimetric methods for the detection of chlorine, phosphate, etc., then it might be possible to dispense with the older sodium fusion test in these laboratories. Further, we had in mind that a combustion process of this kind could be carried out in a fairly reproducible manner, and hence we might make the test not only qualitative but also semi-quantitative in character, particularly when the amount of each additional element present was less than 2 per cent.Moreover, we considered that, with this kind of information in his possession, it might be much easier for the analyst to fashion correct procedures for the subsequent quantitative determination of any of the additional elements. Our first efforts were directed to finding a suitable absorbent for the combustion products. Examination of the literature indicated that, when the oxygen combustion method had240 HASLAM, HAMILTON AND SQUIRRELL : THE DETECTION [Vol. 86 been applied in quantitative tests, different absorbents had been used for different elements, e.g. , sodium hydroxide and sodium sulphite for chlorine,lS2 alkaline hydrogen peroxide for halogens other than fluorine,3 distilled water for fluorineJ3 aqueous hydrogen peroxide for sulphuI-4 and nitric acid for phosph~rus.~ For our purpose, however, we considered it desirable to use only one absorption solution so that all the tests for additional elements could be effected on the products of a single combustion, but it was manifestly impossible to combine all the desirable characteristics of the different a-bsorbents given above in a simple single solution.It was fairly obvious after our preliminary work that the choice of absorbent lay between water and sodium hydroxide. In the tests for Lhlorine, sulphur, phosphorus and fluorine, both these absorbents were of almost equal value, the products of combustion being absorbed in fairly high yield. Nitrogen, however, was a special case, and we realised that a test for this element in the combustion products of an organic compound would almost certainly have to be one for either nitrate or nitrite.We required the test to be fairly rapid, and, as we considered the test for nitrate6 to be rather too lengthy for our purpose, we chose to apply a direct test for nitrite. It was thus necessary to choose the absorption solution giving the highest yield of this anion. Accordingly, tests were carried out on the combustion products obtained from an organic mixture containing 1 per cent. of nitrogen absorbed in (a) water and (b) N sodium hydroxide, testing for nitrite by the Griess - Ilosvay method7 and for nitrate by Buckett, Duffield and Milton’s method, which involves removal of nitrite, reaction of the nitrate with 2,4-xylenol in sulphuric acid media and final colorimetric examination of the extracted nitration products in alkaline media.Our results were- Nitrogen as nitrite in Nitrogen as nitrate in absorption solution, absorption solution, calculated a s % of calculated as yo of Absorbent total N in compound total N in compound Distilled water * . .. 9 12 Sodium hydroxide, N , . .. 25 5 These results indicated that with sodium hydroxide as absorbent we secured a greater proportion of nitrite and less nitrate than with water, and this helped to convince us that sodium hydroxide was the better all-round absorbent for our tests. It should be realised that in all the work described above, and indeed in all our subsequent work, no attempt has been made to secure complete absorption of the products of combustion in the sodium hydroxide soluti,on by complicated shaking procedures.Our aim throughout has been to keep the combustion process and all subsequent operations as simple and as reproducible as possible. Although only about 25 per cent. o:f the total nitrogen in a sample is finally detected, we are satisfied that the whole procedure is reasonably reproducible. Most of our work has dealt with the examination of solid and non-volatile viscous liquid samples. We have found that these materials are most satisfactorily examined by wrapping in Whatman No. 541 filter-paper for the combustion process and, moreover, that materials such as “Melinex” film, polythene film, cellophane and methylcellulose possessed no advan- tages over filter-papers for this purpose.B.P.C.. super-quality cotton-wool manufactured by the Bardsley Wadding Company proved to be satisfactory for wick preparation. This combination of Whatman No. 541 filter-paper and cotton-wool wick has been shown to give satisfactory and complete combustion and the requisite low blank values in the tests for the individual elements. In essence, the final procedure we have devised for the qualitative detection and semi- quantitative determination of the additional elements in plastic materials involves pre- liminary combustion in oxygen and absorption of the combustion products in sodium hydroxide solution. After dilution to a definite volume, the tests for the individual elements are made on suitable aliquots of this solution. Before describing the complete procedure, certain observations on the selected colorimetric tests should be made.NITROGEN- Much of our preliminary work on the detection of nitrogen was carried out by using the principle of the Griess - Ilosvay test.7 Although this test could be most useful in the detection of small amounts of nitrogen in, for example, filter-paper, we considered that, for normal day-to-day work in the plastics industry, it was too sensitive. Great care wasApril, 19611 OF “ADDITIONAL ELEMENTS” IN PLASTIC MATERIALS 241 necessary in the choice of filter-paper for holding the sample for combustion and in the preparation of reagents to avoid relatively high blank colours. The ultra-violet absorption method for determining nitrite described by Altshuller and Wartburg* was also considered, but was found to be too insensitive for our requirements. We finally tried Peach’s r n e t h ~ d , ~ in which resorcinol is nitrosated and the product treated with ferrous salt to give a green complex proportional to the nitrite in the test solution.This method proved to be satis- factory for our purpose, and, although the recoveries of nitrite varied slightly with the func- tional form of the nitrogen in the sample, good positive tests were obtained with all the nitrogen compounds tested. These included amides, amines, nitroso-compounds, azo- compounds and nitro-compounds, including cellulose nitrate ; a stabilised diazonium salt gave only a faint positive test. SULPHUR- The test that we recommend for sulphate in the combustion products is that described by Milton, Hoskins and Jackman.lo This test depends on precipitation of the sulphate by barium chloride in the presence of a protective colloid. In the course of our work, however, information has also been accumulated about (a) the barium chloranilate methodl1 and (b) the 4-amino-4’-chlorodiphenyl hydrochloride method12 for the detection of small amounts of sulphate, and this information is also included in the paper.In our experience, the blank value with the barium chloranilate method is too high and the colour for a given amount of sulphate too low for real satisfaction. The 4-amino-4’-chlorodiphenyl hydro- chloride method is the more sensitive method, and, under our conditions of test, no inter- ference was encountered from fairly high proportions of phosphates. This method may be used when extreme sensitivity is required in the test.PHOSPHORUS- to stannous chloride for the reduction of the molybdophosphate. FLUORINE- The method proposed here is the alizarin complexan - cerous nitrate test of Belcher, Leonard and West,14,15 details of which were kindly supplied to one of us (J.H.) some time ago. HALOGENS- Chlorine is detected by the mercury thiocyanate test described by Sergeant and Thomp- son.16 This test is also given by iodides and bromides, although the presence of iodine and bromine in most organic plastic materials would at present be regarded as abnormal. Never- theless, we have included information about the fluorescein test1’ given by both iodides and bromides and the starch test for iodine. It should be noted that only a small amount of fluorescein is used in the above test in order that there shall be no difficulty whatever in distinguishing between a positive and negative test for these elements.Moreover, we take the view that if iodine and/or bromine are detected in a plastic material then this detection will almost certainly involve the analyst in subsequent quantitative determination of these constituents. Our experience of the behaviour of the various additional elements in the combustion process, when they are present in the sample as different functional groups (e.g., nitrogen as amide or nitro-groups), is obviously limited. We have, however, as a guide, given in the final methods described in full below the order of optical densities obtained under our condi- tions of test when (a) nitrogen is present as acetanilide, ( b ) sulphur as sulphonal, (c) fluorine as fluorobenzoic acid, (d) phosphorus as triphenylphosphine and (e) chlorine as chlorobenzoic acid.In addition, we have shown the influence of the other elements on each specific test by carrying out combustions on composite samples. Fogg and Wilkinson’s test13 for phosphate is used, in which ascorbic acid is preferred COMBUSTION OF SAMPLE AND PREPARATION OF TEST SOLUTION APPARATUS AND REAGENTS- Combustion u.nit-The electrically fired combustion unit previously described2 was used in this work, although we see no reason why other forms of oxygen flask should not be used, with appropriate calibration.242 HASLAM, HAMILTON AND SQUIRRELL : THE DETECTION [Vol.86 FiZter-$aper-Whatman No. 541 filter-paper. When not in use, store the filter-paper in a sealed container out of contact with the laboratory atmosphere. Cotton-wool-B. P. C. Super Quality. Sodium hydroxide, N. PROCEDURE- Weigh out approximately 20 mg of sample, and transfer it to the centre of a small piece of the Whatman No. 541 filter-paper weighing approximately 0.1 g. Fold the filter-paper so that the sample is completely enclosed, and, before making the final fold, insert a small wick of cotton-wool weighing about 6 mg. Carry out the combustion as described previously,2 with 5ml of N sodium hydroxide in the bottom of the flask as absorption solution. Set the flask aside for 15 minutes to allow the gases formed in the combustion to be absorbed, and then wash the contents of the flask quantitatively into a 25-ml measuring cylinder with distilled water. Dilute the solution to 25m1, and mix well.This constitutes the test sol.ution, aliquots of which are taken for the detection of the individual elements by the colorimetric methods detailed below. For com- parison purposes in the tests, prepare a blank test solution by carrying out the combustion procedure on the filter-paper and cotton-wool only. NITROGEN REAGENTS- Resorcinol-AnalaR. Acetic acid, glacial. Ammonium ferrous sulphate-AnalaR. PROCEDURE- Weigh 0-1 g of resorcinol into a clean dry 50-ml beaker, and dissolve in 0-5 ml of glacial acetic acid. Add 5 ml of the test solution, and, after mixing, add 0.1 g of ammonium ferrous sulphate. Carry out the same test on the blank test solution. The development of a green colour in the sample test solution, compared with a pale yellow in the blank, indicates the presence of nitrogen in the sample. If a semi-quantitative estimation of the nitrogen content of the sample is required, set both sample and blank solutions aside for 20 minutes.Add 10 ml of distilled water to each, mix, and measure the optical density of the sample solution against the blank at 690 mp in a 4-cm cell. The blank in this test is low, giving an optical density of 0.07 measured against water at 690 mp in the 4-cm cells. TABLE I TYPICAL OPTICAL DENSITIES OBTAINED WITH MIXTURES CONTAINING 1 AND 2 Mixture tested .. .. .. .. Acetanilide, 1-9 mg Benzoic acid, 18-1 mg Acetanilide, 3.8 mg Benzoic acid, 16.2 mg Acetanilide, 1.9 mg Chlorobenzoic acid, 0.9 mg (1% C1) Sulphonal, 0.7 mg (1% S) Triphenylphosphine, 1.7 mg (1 yo P) Fluorobenzoic acid, 1.5 mg (1% F) Benzoic acid, 13.3 mg Acetanilide, 3.8 mg Chlorobenzoic acid, 1.8 mg (2yo C1) Sulphonal, 1.4 mg (2% S) Triphenylphosphine, 3.4 mg (2% P) Fluorobenzoic acid, 3.0 mg (2yo F) Benzoic acid, 6.6 mg per Cent.OF NITROGEN Nitrogen in Optical density mixture, ( D&$mm,r measured against blank % .. 1 0.65 .. 2 1.0 .. 1 0.65 2 1.0April, 19611 OF “ADDITIONAL ELEMENTS” IN PLASTIC MATERIALS 243 Typical results obtained with organic mixtures containing 1 and 2 per cent. of nitrogen present as acetanilide both in the presence and absence of other additional elements are shown in Table I. SULPHUR RECOMMENDED BARIUM CHLORIDE METHOD REAGENTS- Hydrochloric acid, x Precipitating reagent solution A-I>issolve 0.2 g of peptone in 50 ml of 1 per cent.barium chloride (BaC1,.2H20) solution. Buffer to a pH of 5-0 with 0.02 N hydrochloric acid, add 10 g of sodium chloride (AnalaR), and dilute to 100 ml. Heat on a water bath for 10 minutes, and add a few drops of chloroform. Filter if necessary. Precipitating reagent solution 23-Dissolve 0-4 g of gum ghatti in 200 ml of distilled water by warming slightly. When solution is complete, add 2.0 g of barium chloride (BaC12.2H,0). Filter if necessary. Store solutions A and B separately, and prepare the final reagent just before use by diluting 10ml of solution A to 100ml with solution B. Hydrogen peroxide, 100-volume-AnalaR. METHOD- Transfer 5ml of the test solution to a 6-inch x 1-inch test-tube, and add 2 drops of 100-volume hydrogen peroxide and then 1.2 ml of N hydrochloric acid.Mix well, and add 2.0 ml of precipitating reagent with continued shaking. A distinct turbidity will be produced in the mixed solution if sulphur is present in the sample; the blank test under the same conditions will be perfectly clear. If a semi-quantitative estimation of the sulphur content is required, add 5ml of distilled water to both blank and test solutions, mix, and set aside for 30 minutes. Mix the solutions, ,and measure the optical density of the test solution in a 4-cm cell at 700mp with the bhnk solution in the comparison cell. OTHER METHODS FOR SULPHUR For information , we give below details of the 4-amino-4’-chlorodiphenyl hydrochloride and the barium chloranilate methods, :in the form that they can be applied to the combustion test solution if required.4-AMINO-4’-CHLOROl~IPHENYL HYDROCHLORIDE METHOD REAGENTS- 4-Amino-4’-chlorodiPhenyl hydrochJoride - peptone solution (solution A )-Dissolve 0.025 g of peptone (Hopkin and Williams Ltd.) and 0.125 g of 4-amino-4’-chlorodiphenyl hydrochloride (L. Light and Co. Ltd.) as completely as possible by warming with 50 ml of 0-05 N hydrochloric acid. 4-Amino-4’-chlorodiphenyL hydrochloride - gum ghatti solution (solution B)-Dissolve 0.20 g of gum ghatti (Hopkin and Williams :Ltd. ; finely ground) and 1.00 g of 4-amino-4’-chlorodi- phenyl hydrochloride as completely as possible by warming to about 70” C with 400 ml of 0.05 N hydrochloric acid.Cool, and filter through a double Whatman No. 40 filter-paper. Final precipitation reagent-Dilute 10 ml of solution A to 100 ml with solution B. This solution should be freshly prepared. Hydrochloric acid, N. METHOD- Transfer 5ml of the test solution to a 6-inch x 1-inch test-tube, and add 1.2ml of N hydrochloric acid. Mix well, and add 10.0 ml of the prepared final precipitation reagent, A distinct turbidity will be produced in the mixed solution if sulphur is present in the sample. The blank test under the same conditions will be perfectly clear. If a semi-quantitative estimation of the sulphur content is required, set the solutions aside for 30 minutes, then mix, and measure the optical density of the test solution against the blank in 2-cm cells at 700mp.Cool, and filter through a double Whatman No. 40 filter-paper.244 HASLAM, HAMILTON AND SQlJIRRELL THE DETECTION BARIUM CHLORANILATE METHOD REAGENTS- Barium chloranilate-Obtained from Hopkin and Williams Ltd. Potassium hydrogen 9hthalate-AnalaR. Ethanol, absolute. METHOD- pol. 86 Transfer 5ml of the test solution to a stoppered 100-ml conical flask, and add 1-0mt of N hydrochloric acid and then 0.2 g of potassiuin hydrogen phthalate and 6.25 ml of absolute ethanol. Add 0.04g of barium chloranilate, insert the stopper, and shake the flask for 15 minutes on a mechanical shaker. Filter the solution through a small Whatman No. 540 filter-paper. Treat the blank solution in a siimilar manner. When sulphur is present in the sample a red-purple colour will be developed in the test solution compared with the lighter coloured blank. If a semi-quantitative estimation of the sulphur present is required, measure the optical density of the sample against the blank solution a t 530mp in a 4-cm cell, 30 minutes after beginning the filtration.RESULTS Typical results for the optical densities obtained with organic mixtures containing 1 and 2 per cent. of sulphur present as sulphona.1 are shown in Table I1 for all three methods. The effect of the presence of other elements on the sulphur figure is also shown at both the 1 and 2 per cent. levels. The blank solutions measured against water under the conditions described had optical densities of the following order : recommended barium chloride method, 0.02 ; 4-amino- 4’-chlorodiphenyl hydrochloride method, 0.01 ; barium chloranilate method, 0.13.The results in Table I1 show clearly the relative insensitivity of the barium chloranilate method, although it is possible to increase the sensitivity in a particular test by absorbing the combustion products from 20 mg of sample in 5 ml of 0.2 N sodium hydroxide and adding the 1.0 ml of N hydrochloric acid and 0.2 g of potassium hydrogen phthalate to the solution in the combustion flask. The 6.25 ml of ethanol is used to wash down the sides of the flask, and, when mixed, all the solution is poured into a stoppered 100-ml conical flask for treatment with the barium chloranilate and colorimetric measurement as previously described. Under these conditions, the optical densities obtained for sulphur contents of 1 and 2 per cent.are increased to 0.3 and 0.6, respectively, but no test for other elements can be made on the same combustion products. TABLE I1 TYPICAL OPTICAL DENSITIES OBTAINED WITH MIXTURES CONTAINING 1 AND 2 per cent. OF SULPHUR Optical density by Optical density by Optical density recommended 4-amino-4’-chloro- by barium barium chloride diphenyl hydrochloride chloranilate Sulphur in method (D$hrnm, method (D;&mmw method (D&T& mixture, measured against measured against measured against Mixture tested % blank) blank) blank) Sulphonal, 0.7 mg Benzoic acid, 19-3 mg Sulphonal, 1.4 mg Benzoic acid, 18.6 mg Composite mixture as Composite mixture as in Table I in Table I 0.2 0-4 0.2 0.45 0.35 0.70 0.35 0.70 0.06 0.13 0.04 0.09 PHOSPHORUS REAGENTS- [(NH,),Mo,0,,.4H20] in about 70 ml of water, and dilute to 100 ml.stirring to a cooled mixture of 150ml of sulphuric acid and 150ml of water. Ammonium molybdate solution-Dissolve 10 g of AnalaR ammonium molybdate Add this solution with A scorbic acid-B .D. H. laboratory-reagent grade.April, 19611 OF “ADDITIONAL ELEMENTS” IN PLASTIC MATERIALS 245 METHOD- Add 40ml of distilled water and 4 ml of ammonium molybdate solution. Mix thoroughly, then add 0.1 g of ascorbic acid, and boil the solution for 1 minute. Cool in running water for 10 minutes, and dilute to 50ml with distilled water. When phos- phorus is present in the sample, a blue colour will be developed in the test solution as compared with a pale yellow in the blank. If a semi-quantitative estimation of the phosphorus is required, measure the optical density of the test solution against the blank solution at 820 mp in 2-cm cells.Transfer 2ml of the test solution to a 100-ml beaker. Treat the blank solution in a similar manner. TABLE I11 TYPICAL OPTICAL DENSITIES OBTAINED WITH MIXTURES CONTAINING 1 AND 2 per cent. OF PHOSPHORUS Phosphorus in Optical density Mixture tested mixture, (DiGrnmlL measured against blank) % .. .. . . 1 0.45 .. .. .. 2 0.95 .. .. .. 1 0-5 1.0 i .. .. .. 2 Triphenylphosphine, 1.7 mg Benzoic acid, 18.3 mg Triphenylphosphine, 3.4 mg Benzoic acid, 16.6 mg Composite mixture as in Composite mixture as in Table I Table I Typical results for the optical densities obtained with organic mixtures containing triphenylphosphine at 1 and 2 per cent.phosphorus levels are shown in Table 111. The effect of the presence of other elements on the phosphorus figure is also shown at both 1 and 2 per cent. levels. The blank solution measured against water under the conditions described had an optical density of the order of 0.01 to 0.015. FLUORINE REAGENTS- Buflered alizarin complexan solution-Weigh 40.1 mg of 3-aminomethylal izarin-“- diacetic acid (Hopkin and Williams Ltd.) into a beaker, and add 1 drop of N sodium hydroxide and approximately 20ml of distilled water. Warm the solution to dissolve the reagent, cool, and dilute to 208 ml. Weigh into another beaker 4-4g of sodium acetate (CH3COONa.3H,0), and dissolve in water. Add 4.2 ml of glacial acetic acid, and dilute to 42 ml. Pour this sodium acetate solution into the alizarin complexan solution, and mix to give the final buffered alizarin complexan solution. Cerous nitrate, 0.0005 M-Dissolve 54.3 mg of cerous nitrate [Ce(N03),.6H,0] in water, and dilute to 250ml.METHOD- Transfer 20ml of distilled water and 2.4 ml of buffered alizarin complexan solution to a 50-ml beaker. Add 1 ml of test solution, and mix by swirling the solution. Finally, add 2 ml of cerous nitrate solution, and mix again. Treat the blank solution in a similar manner. When fluorine is present in the sample a mauve colour will be developed in the test solution compared with the pink-coloured blank solution. If a semiquantitative estimation of fluorine is required, set the solutions aside for 10 minutes, and measure the optical density of the test solution against the blank solution at 600mp in 1-cm cells.Typical results for the optical densities obtained with organic mixtures containing fluoro- benzoic acid at 1 and 2 per cent. .fluorine levels are shown in Table IV. The effect of the presence of other elements on the fluorine figure is also shown at both 1 and 2 per cent. levels.246 HASLAM, HAMILTON AND SQIJIRRELL : THE DETECTION [Vol. 86 The blank solution measured against water .under the conditions described had an optical density of about 0.08. TABLE IV TYPICAL OPTICAL DENSITIES OBTAINED WITH MIXTURES CONTAINING 1 AND 2 per cent. OF FLUORINE Fluorine in Optical density Mixture tested mixture, (DtgFmw measured against blank) % . . . . .. 1 0.07 . . . . . . 2 0-135 . . .. , . 1 0.08 i . . . . . . 2 0.135 Fluorobenzoic acid, 1.5 mg Benzoic acid, 18.5 mg Fluorobenzoic acid, 3.0 mg Benzoic acid, 17.0 mg Composite mixture as in Composite mixture as in Table I Table I CHLORINE REAGENTS- in water, and add 40 ml of AnalaR nitric acid. crystallised from ethanol in 100 ml of absolute ethanol. Ammonium ferric suZ#hate solution-Dissolve 12 g of AnalaR ammonium ferric sulphate Mercuric thiocyanate solution-Dissolve 0.4 g of mercuric thiocyanate that has been re- Dilute to 100 ml, and filter. METHOD- Transfer 5 ml of the test solution to a 50 ml beaker, and add 1 ml of ammonium ferric sulphate solution. Mix the solution, and add 1.5 ml of mercuric thiocyanate solution. Mix, and dilute to 10ml. When chlorine is present in the sample an orange colour will be developed in the test solution compared with the yellow-coloured blank solution.If a semi-quantitative estimation of chlorine is required, set the solutions aside for 10 minutes, and measure the optical density of the test solution against the blank solution at 460mp in 2-cm. cells. Treat the blank solution in a similar manner. TABLE V TYPICAL OPTICAL DENSITIES OBTAINED WITH MIXTURES CONTAINING 1 AND 2 per cent. OF CHLORINE Mixture tested Chlorine in Optical density mixture, ( Dz&&p measured against blank) n / 70 . . . . . . 1 .. .. . . 2 Chlorobenzoic acid, 0.9 mg Benzoic acid, 19.1 mg Chlorobenzoic acid, 1.8 mg Benzoic acid, 18.2 mg . . . . . . 1 . . . . . . 2 Composite mixture i s in Composite mixture as in Table I Table I 0.4 0.75 0.4 0.75 Typical results for the optical densities obtained with organic mixtures containing chloro- benzoic acid at 1 and 2 per cent.chlorine levels are shown in Table V. The effect of the presence of other elements on the chlorine figure is also shown at both 1 and 2 per cent. levels. The blank solution measured against water under the conditions described had an optical density of 0-15.April, 19611 OF “ADDITIONAL ELEMENTS” IN PLASTIC MATERIALS 247 BROMINE REAGENTS- and dilute to 100ml with water. acid. Fluorescein solation-Dissolve 0.1 g of fluorescein in 25 ml of 0.1 N sodium hydroxide, Sodium acetate bufler solution-Mix 100 ml of N sodium acetate with 15 ml of N acetic Chloramine-T solution-Dissolve 1 g of chloramine-T in 100 ml of distilled water. Sodium thiosulphate solution-Prepare a 0.5 per cent.w/v solution of sodium thiosulphate Hydrochloric acid, N. Transfer 5 ml of the test solution to a 50-ml beaker. in 5 per cent. w/v sodium hydroxide solution. METHOD- Add 1 ml of N hydrochloric acid and then 0 6 m l of sodium acetate buffer solution and 1 drop of fluorescein solution. Mix thoroughly, and then add 1 drop of chloramine-T solution. Mix by swirling, and set aside for 30 seconds; then stop the reaction by adding 2 drops of alkaline thiosulphate reducing agent. Treat the blank solution in a similar manner. When bromine is present in the sample a rose-pink colour will be developed in the test solution compared with the yellow-green blank solution. NOTE-This t e s t is also given by iodine. IODINE REAGENTS- Starch solution-Dissolve 0 .2 g of soluble starch in 100ml of distilled water. Sulphwic acid, dilute-Prepare an approximately 10 per cent. v/v solution of concen- trated sulphuric acid in distilled water. METHOD- Mix the solution, and then acidify with dilute sulphuric acid. manner. will be developed in the test solution compared with the colourless blank solution. Transfer 5 ml of test solution to a 50-ml beaker, and add a few drops of starch solution. Treat the blank in a similar When iodine is present in the sample, the characteristic blue colour of starch iodide APPLICATION OF THE METHODS For most purposes, application of the procedure as described provides sufficient informa- tion about the sample under test, but, if required, the sensitivity of any of the individual tests may be significantly increased.This may be accomplished by repeating the combustion, absorbing the products in 5 ml of 0.2 N sodium hydroxide and making a direct test for the particular element on this solution without dilution. An example of this has been given for the barium chloranilate test for sulphur. The principle of the combustion procedure may be extended, if necessary, to the detection of other elements, such as boron, arsenic,18 radioactive carbonlg and, sometimes, metals such as zinc, cadmium, magnesium,2o barium21 and mercury. 22 Although most of our work has dealt with the examination of solids and non-volatile viscous liquids, which can be wrapped in filter-paper for combustion, we see no reason why the test should not be extended to the examination of volatile liquids.In certain tests, materials such as methylcellulose or gelatin capsules could be used to hold the sample, provided careful attention is paid to the blank values obtained with these containers. Alterna- tively, the glass-capillary method of Bennewitz23 might be used. It may be useful if we indicate here some of the problems to which we have applied the procedures described. A polymer received for analysis was suspected to contain small amounts of both iodine and fluorine. Application of the qualitative tests indicated the presence of both iodine and fluorine, and, moreover, it was shown that these qualitative tests could be readily adapted to yield immediate semi-quantitative information about the proportions of the elements present, which were of the order of 2 per cent.of fluorine and 5 per cent. of iodine. This problem might not have been solved so easily at this level by conventional procedures.248 HASLAM, HAMILTON AND SQUIRRELL pol. 86 In another instance an acrylic sheet was shown to possess abnormal solubility charac- teristics. A portion of the sheet was submitted to the general tests outlined above, which quickly indicated that the sample contained of the order of 1 per cent. of combined nitrogen. A figure of 1.0 per cent. of nitrogen was obtained by the lengthy Kjeldahl procedure. This detection of nitrogen implied the presence in the sheet of a copolymerised nitrogen-containing compound, which was later confirmed. A vinyl chloride copolymer was suspected from the infra-red evidence to contain a proportion of a butadiene - styrene - acrylonitde copolymer, there being evidence of an absorption band corresponding to a -C=N grouping at 4-4 p.This portion of the spectrum is particularly free from interference from other polymer groupings. Repeated sodium fusion tests indicated that nitrogen was absent from the composition, but application of our procedure indicated the definite presence of about 0.25 per cent. of nitrogen, equivalent to approximately 1 per cent. of acrylonitrile in the sample. The more sensitive form of the test we have mentioned has proved most useful for the confirmation that certain additives have been incorporated in polymer compositions. In this connection the detection of additives equivalent to less than 0.1 per cent. of sulphur, nitrogen or phosphorus has proved possible. We thank Miss M. Clark and Mr. J. L. Sharp for their assistance in this investigation. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. RE FE RE. NC E s Haslam, J., Hamilton, J. B., and Squirrell, D. C. M., J . AppZ. Chem., 1960, 10, 97. Schoniger, W., Mikrochim. Acta, 1955, 123. BOOS, R, N., Analyst, 1959, 84, 633. Fleischer, K. D., Southworth, B. C., Hodecker, J. H., and Tuckerman, M. M., Anal. Chem., 1958, Buckett, J., Duffield, W. D., and Milton, R. F., Analyst, 1955, 80, 141. “The Determination of Toxic Substances in Air,” I.C.I. Manual, p. 196. Altshuller, A. P., and Wartburg, A. F., Anal. Chem., 1960, 32, 174. Peach, S. M., Analyst, 1958, 83, 644. Milton, R., Hoskins, J. L., and Jackman, W. H. F., Ibid., 1944, 69, 299. Bertolacini, R. J., and Barney, J. E., A n d Chem., 1957, 29, 281. Belcher, R., Nutten, A. J., Parry, E., and Stephen, W. I., Analyst, 1956, 81, 4. Fogg, D. N., and Wilkinson, N. T., Ibid., 19158, 83, 406. Belcher, R., Leonard, M. A., and West, T. S., Talanta, 1959, 2, 92. Sergeant, G. A., and Thompson, P. B., Analyst, 1959, 84, 251. Hardwick, P. J., Ibid., 1942, 67, 223. Corner, M., Ibid., 1959, 84, 41. Gotte, H., Kretz, R., and Baddenhausen, H., Angew Chem., 1957, 69, 561. Relcher, R., Macdonald, A. M. G., and West, T. S., TaZanta, 1958, 1, 408. Barney, J. E., Bergmann, J. C., and Tuskan, W. G., Anal. Chem., 1959, 31, 1394. Southworth, B. C., Hodecker, J. H., and Fleischer, K. D., Ibid., 1958, 30, 1152. Bennewitz, R., Mikrochirn. Acta, 1960, 54. 9 1 , Analyst, 1960, 85, 556. --- 30, 152. , I , J . Chem. SOC., 1959, 3577. --- Received November 17th, 1960

ISSN:0003-2654    

DOI:10.1039/AN9618600239

出版商:RSC    

年代:1961

数据来源: RSC

摘要:

April, 19611 LAWS AND WEBLEY 249 The Determination of Organo-phosphorus Insecticides in Vegetables A General Method for Insecticide Residues BY E. Q. LAWS AND D. J. WEBLEY (Department of Scientific aizd Industrial Research, Laboratory of the Government Chemist, Clement's Inn Passage, Strand, London, W.C.2) A method is described for the extraction of insecticides from treated cabbages and other vegetables and for their separation into water-soluble and petroleum-soluble groups. The further separation of members of each group from associated plant material is described for eighteen organo-phos- phorus insecticides, and the quantitative separation of both groups is discussed, with particular reference to mixtures of (a) phorate with its oxygen-analogue sulphone and (b) parathion-methyl with Phosdrin.THE method described by Laws and Webleyl for extracting and determining demeton-methyl is applicable to a number of water-soluble insecticides. It is not applicable, as it stands, to insecticides preferentially soluble in light petroleum, since the solvent systems used fail to extract the hydrocarbon-soluble insecticides from the macerated plant material, and no attempt is made to utilise the light petroleum washings. Modifications have consequently been made to the extraction procedure, and these ensure that both types of compound are extracted from the plant tissue before they are separated by partition. EXPERIMENTAL EXTRACTION AND SEPARATION- The first extraction is carried out by macerating the sliced vegetable with dichloromethane.After filtration, washing and then evaporation of the solvent, partition is effected between light petroleum and a 15 per cent. solution of methanol in water. At this stage, the insecticide is distributed as shown in the diagram below (solutions A and B)- Sliced vegetable material I Macerate with dichloromethane (All common insecticides extracted) I I Partition between light petroleum and 15 per cent. methanol - water I Petroleum-soluble organb-phosphorus insecticides Water-soluble okgano-phosphorus insecti- (solution A) cides (solution B) I I Chromatography on alumina; light petroleum as eluting agent I 1 I I Chlorthion* Wash column successively Disyston* with light petroleum and a Diazinon 15 per cent. solution of diethyl Fenchlorphos ether in light petroleum Phenkapton Gusathion* Phorate (Thimet*) Malathion S1752 Parathion-m ethyl I * Trade name. Chromatography on carbon; chloroform as eluting agent I Demeton-methyl and metabolites Dime fox Morphothion Phorate oxygen-analogue sulphone Phosdrin" Phosphamidon Rogor* Trichlorphon From this point in the determination the two groups are treated separately by different methods.The petroleum-soluble portion (solution A) is chromatographed on a column of graded alumina, light petroleum being used to elute the insecticides, and the water-soluble250 LAWS AND WEBLEY: THE: DETERMINATION OF [Vol. 86 portion (solution B) is examined by Laws and Virebley's chromatographic method1 involving elution with chloroform from a carbon column. The final eluates from the columns in both groups contain the insecticides substantially free from organic phosphorus of plant origin.TABLE: I RECOVERIES OF PETROLEUM- AND WATER-SOLUBLE INSECTICIDES The amount of each insecticide present was 50 to 70 pg. Petroleum-soluble insecticides were eluted with light petroleum and water-soluble insecticides with chloroform Petroleum-soluble group A r Insecticide Chlorthion* .. .. Disyston" . . .. .. Diazinon . . .. .. Fenchlorphos . . .. Gusathion* .. .. Malathion . . .. .. Parathion-methyl . . .. Phenkapton . . . I Phorate (Thimet*) . . S1752 . . .. .. Recovery from alumina column, 64 72 76 90 90 66 95 91 75 100 % Water-soluble group - A I 1 Recovery Recovery of Recovery of from insecticide insecticide added to carbon added to cabbage, Insecticide column, cabbage, % % % 66 Demeton-meth yl 95 and metabolites .. 85 76 89 Dimefox .. . . . . 98 80 89 Morphothion . . .. 91 84 79 Phorate oxygen-analogue 75 sulphone , . .. 80 70 65 Phosdrin* . . .. . . 90 65 74 Phosphamidon . . .. 78 70 75 Rogor* . . .. .. 80 65 70 Trichlorphon . . . . 75 57 * Trade name. TABLE I1 RECOVERY OF ADDED PHORATE AND ITS OXYGEN-ANALOGUE SULPHONE FROM CABBAGE The insecticides were added to 50-g portions of cabbage (range 0.25 to 4-0 p.p.m.) Insecticide added & Oxygen- analogue Phorate, sulphone, r g 209.3 11 1.6 111.6 55.8 55.8 55.8 27.9 27.9 13.9 Nil 55.8 r g 200.0 106.6 55.3 106.6 53.3 26.7 53.3 26.7 13.4 53.3 Nil Insecticide recovered Recovery r Phorate, rg 146.5 83.7 74.8 27.9 47.4 55.8 24.0 17.9 11.7 Nil 37.4 (Oxygen- analogue sulphone, r g 128-0 64.0 19.7 90.6 53.3 21.6 38.4 17.4 8.7 33.0 Nil Average .. Phorate, 70 76 67 50 85 100 86 64 84 Nil 67 75 % Oxygen- analogue sulphone, % 64 60 37 85 100 81 72 65 65 62 Nil 69 As this procedure isolates the insecticides in a reasonably pure condition, the method used to determine the residue is a matter of choice; determination of phosphorus, cholin- esterase assay, infra-red spectrography, gas chromatography or specific chemical or colori- metric methods present themselves as possibilities. The spectrophotometric determination of phosphorus as the molybdenum-blue complex was used to obtain the results reported in this paper. The solubility of the insecticides in light petroleum or water was first investigated, and then the behaviour on the appropriate chromatographic column was studied.Quantita- tive recovery experiments were next carried out on solutions of the commercially pure insecticide, and, finally, the compounds were added to raw cabbage and the complete method was applied.April, 19611 ORGANO-PHOSPHORUS INSECTICIDES I N VEGETABLES 251 In all the determinations reported here, the two phases of the partitition were completely andysed to ensure that the observed losses were not due to incomplete separation at this stage. In considering the recoveries of the various insecticides, the difficulties of residue work must be borne in mind. There are systematic losses at various stages of the procedure, and these are well known to practitioners in this field. Extraction from the plant rarely exceeds 95 per cent., and recovery from the chromatographic column varies with the purity of the insecticide; commercially pure insecticides may contain from 90 to 100 per cent.of pure active ingredient. Further, some insecticides are volatile, and, although the proposed method includes precautions to avoid losses due to volatility, it is a factor to be reckoned with at all stages. It is generally accepted that, for this type of analysis, an over-all recovery of 70 to 80 per cent. is satisfactory. The results of the experiments are shown in Table I. APPLICATIONS OF METHOD- To illustrate the application of the method to mixtures of insecticides, two series of recovery experiments were carried out on cabbage; two insecticides were used in each series, one component of each pair being petroleum-soluble and the other water-soluble. The chosen mixtures were (a) phorate (Thimet) with its oxygen-analogue sulphone and (b) parathion-me t hyl with Phosdrin .TABLE I11 RECOVERY OF ADDED PARATHION-METHYL AND PHOSDRIN FROM CABBAGE The insecticides were added to 50-g portions of cabbage (range 0.5 to 2-0 p.p.m.) Insecticide added Insecticide recovered s Parathion- methyl, Phosdrin, Pg Pg 97.5 112-0 97.5 56-0 58.5 112.0 58.5 56.0 29.3 56.0 58-5 33.6 29-3 33.6 97.5 Nil Nil 112.0 97-5 112.0 58.5 Nil Nil 56.0 I 1 Parathion- methyl, Phosdrin, Pg Pg 84.8 75.0 60.5 38.1 49.3 80.6 37.4 42.6 17.9 26.9 33.9 21.5 25.5 20.5 59.5 Nil Nil 62.7 -* 73.9 30.4 Nil Nil 37-0 Average . . * Solution lost. Recovery methyl, Phosdrin, G O ? % % 87 62 84 64 61 58 87 60 Nil 52 Nil 65 -* 67 68 72 76 48 64 61 Nil 56 66 Nil 66 65 The compound 00-diethyl S(ethylthiomethy1) phosphorothiolothionate, I, is known as I t undergoes a series of oxidative changes, either by chemical or enzyme phorate (Thimet).systems, the main product of which is the oxygen-analogue sulphone, 11. Samples of compounds I and 11, provided by Cyanamid of Great Britain Ltd., had a purity of 98 per cent., based on phosphorus content. Suitable standard solutions were prepared, each containing approximately 15pg of insecticide per ml of the final dilution, and the amounts of insecticide added and recovered are shown in Table 11. Compound I is typical of the petroleum-soluble group and compound I1 of the water-soluble group.252 LAWS AND WEBLEY: THE DETERMINATION OF [Vol.86 The compounds 00-dimethyl O-@-nitrophenyl phosphorothionate, 111, and 2-methoxy- carbonyl-l-methylvinyl dimethyl phosphate, IV, are commercially known as parathion-methyl and Phosdrin, respectively ; the former is petroleum-soluble and the latter water-soluble. S CHa-0 A sample of compound I11 having a purity of :LOO per cent. (by phosphorus content) was provided by Fisons Pest Control Ltd., and one of compound IV (purity of 96 per cent. by phosphorus content and 88 per cent. by extraction with chloroform) by Shell Chemicals Ltd. Two standard solutions, one containing 19 pg of compound I11 per ml and the other 11 pg of compound IV per ml were prepared, and the results of recovery experiments with these solutions are shown in Table 111. To show that the method was capable of wider application, some experiments were carried out on other vegetables and some fruits; the blank values found are shown in Table IV.TABLE :[V BLANK VALUES FOUND FOR 50-g SAMPLES OF VARIOUS VEGETABLES AND FRUITS Elution with light petroleum from alumina column Elution with chloroform from carbon column f Sample Cabbage .. . . .. Apples . . . . . . Peas. . . . . . . . Plums . . .. . . Cherries . . .. .. Lettuce . . .. .. Tomatoes . . . . .. Potatoes (tubers) . . .. Brussels sprouts . . .. Carrots . . . . .. Optical-density blank value 0.040 0.070 0.010 0.015 0.017 0.025 0-055 0.017 0-018 0.044 - r - Equivalent amount of phosphorus, Pg 0.64 1.12 0.15 0.23 0.26 0.42 0-89 0.26 0.27 0.68 A Optical-density blank value 0.045 0.032 0.016 0.0 13 0.018 0.015 0.015 0.043 0.040 0.040 - Equivalent amount of phosphorus, Pg 0.72 0.50 0.23 0.22 0.29 0.23 0.23 0.69 0.64 0.64 METHOD The method consists in the extraction of the insecticides from the plant material, their separation into petroleum- and water-soluble groups , chromatography of the groups on alumina and activated carbon, respectively, anti then determination of the phosphorus by spectrophotometric measurement of the rnolybdenum-blue complex.APPARATUS- Unicam SP500 absorption spectrophotometer. Glass tubes for chromatography-Tubes 1.5 c:m in diameter and 14 cm long, fitted with a tap at the lower end and a B19 ground-glass' joint at the upper end for attachment to a reservoir of solvent. M.S.E. homogeniser, with 100-ml Vortex beakei.s-0btainable from Measuring and Scientific Equipment Ltd., London, S.W.l. REAGENTS- All reagents must be of recognised analytical grade and phosphate-free. Chloroform. Dichloromethane. Diethyl ether. Light petroleum, boiling range 40" to 60" C. Methanol, absolute.April, 19611 ORGANO-PHOSPHORUS INSECTICIDES I N VEGETABLES 253 Alumina-Heat chromatographic aluminium oxide at 500" C for 2 hours to ensure conversion to the y-form. Adjust the activity to Brockmann grade V by adding 15 per cent. w/w of water. Active carbon-Heat 14- to 22-mesh carbon (grade 207, type B, obtainable from Messrs. Sutcliffe Speakman, Leigh, Lancashire) at 600" C in closed crucibles for 2 hours to remove organic impurities, boil twice with concentrated hydrochloric acid for 30 minutes on each occasion, wash free from acid with water, and dry in an oven at 100" to 110" C.Perchloric acid, N. Sulphuric acid, 10 and 1 N. Nitric acid, sp.gr. 1.420. Hydrochloric acid, s9.g. 1.180. Ammonia solution, sp.gr. 0.880. Ammonium molybdate solution-Dissolve 50 g of ammonium molybdate in 400 ml of Stannous chloride solution, concentrated-Dissolve 10 g of stannous chloride dihydrate Stannous chloride solution, dihte-Dilute the concentrated solution 200-fold with 1 N Ethanolic sulphuric acid-Mix 5 ml of concentrated sulphuric acid with 245 ml of absolute Isobutyl alcohol - benzene mixture, (1 + 1, v/v). Standard insecticide solutions-Prepare from the pure active ingredients, and suitably dilute. Potassium dihydrogen orthophosphate. EXTRACTION AND SEPARATION- Shred a representative sample of the plant in a slicing-and-grating machine, and transfer 5 0 g of the shredded material to the homogeniser beaker.If an addition of insecticide is required, add an aliquot of standard insecticide solution, and set aside for 30 minutes. Add 100 ml of dichloromethane, macerate for 15 minutes, filter the mixture through a Buchner funnel, and wash the solid on the filter-pad with 60ml of dichloromethane. Transfer the combined filtrate and washings to a separating funnel, and run the lower organic layer into a 260-ml conical flask. Wash the small aqueous layer with two 15-ml portions of dichloro- methane, and add the washings to the main extract. Reject the aqueous layer, and heat the dichloromethane solution on a hot-plate in a current of air until all the solvent has evaporated, taking care not to heat the flask overmuch in removing the last traces.Im- mediately add 5ml of methanol to the residue, pour the methanol solution into a 100-ml separating funnel, wash the flask with 30 ml of light petroleum and 25 ml of water, and add the washings to the contents of the separating funnel. Shake the funnel vigorously, allow to settle, and run the lower aqueous layer back into the 250-ml conical flask. Re-extract the light petroleum with 5 to 10 ml of water, run the aqueous layer into the conical flask, and pour the light petroleum layer from the top of the funnel into a 100-ml flask. Return the aqueous layer to the funnel, rinsing the flask with 30 ml of light petroleum and 10 ml of water. Shake vigorously, and separate as before.Rinse the funnel with a little light petroleum, and add the rinsings to the petroleum extract. Again return the aqueous portion to the funnel, extract it with four 20-ml portions of chloroform, and combine the chloroform extracts. CHROMATOGRAPHY- Petroleum-soluble insecticides-Evaporate the light petroleum extract to about 10 ml by heating on a water bath in a current of air. Prepare a column 1.5 cm in diameter from 8 g of the Brockmann grade V alumina; use light petroleum as liquid phase. Transfer the extract to the column, and elute with 150 ml of light petroleum a t about 1.5 ml per minute. To recover Gusathion and malathion, elute with 150 ml of light petroleum - diethyl ether mixture (85 + 15 v/v). Water-soluble insecticides-Evaporate the chloroform extract to about 10 ml by heating on a hot-plate; blow a current of air across the surface of the solvent t o cool the liquid and so minimise loss of volatile insecticides.With chloroform as liquid phase, prepare a column 10 M sulphuric acid, and dilute to 1 litre with distilled water. in 26ml of hydrochloric acid, spgr. 1.180. sulphuric acid. Prepare a fresh solution daily. ethanol .254 LAWS AND WEBLEY: THE DETERMINATION OF vol. 86 from 4 g of active carbon in a tube similar to that used for the alumina. Transfer the extract to the column, and elute with 100 ml of chloroform at 1.5 ml per minute. TREATMENT OF ELUATES- Evaporate the light petroleum, the light petroleum - diethyl ether and the chloroform eluates from the columns in separate conical flasks fitted with ground-glass joints; heat the first two eluates on a water bath and the third on a hot-plate, and assist evaporation with a stream of air to avoid loss of volatile insecticides.The chloroform solution need not be taken completely to dryness, as the last traces cd chloroform will boil away from the mixed acids and water in the final stages of the determination. Wet combustion of separated insecticides-To each evaporated eluate add 10 ml of water, 4 ml of N perchloric acid, 5 ml of nitric acid, sp.gr. 1-420, and 1 ml of hydrochloric acid, sp.gr. 1-180, and heat the solution under reflux beneath a Liebig condenser for 40 minutes. Drain the cooling water from the condenser, and leave until brown fumes begin to appear. At this point, remove the condenser, and evaporate the solution until fumes of perchloric acid are evolved.Rinse the condenser with 6 to 7 mi of water, add the rinsings to the contents of the flask, and again evaporate until white fumes are evolved. Add 3ml of water and 4ml of ammonia solution, sp.gr. 0.880, and remove the excess of ammonia by boiling. If it is known that volatile insecticides are not present, the heating under reflux may be omitted; the solution is heated to fumes with the acids as described above and then with 6 ml of water, with 2 ml of nitric acid and again with 5 ml of water before the ammonia solution is added. Place the test solution containing the phosphorus in a 100-ml separating funnel, and bring the volume to 10 ml, including any washings. Add 3.5 ml of ammonium molybdate solution, and mix after adjusting the total volume to 14ml.If the molybdate solution is added before transfer to the separating funnel, the final volume should still be 14 ml. Add 10 ml of isobutyl alcohol - benzene mixture, and shake the funnel vigorously. Allow the layers to separate, and discard the lower layer. Wash the remaining organic layer with 5 to 10 ml of N sulphuric acid, and again discard the lower layer. Add 15 ml of the dilute stannous chloride solution, shake for 5 to 10 seconds, and allow the layers to separate. Discard the lower aqueous layer, and allow the organic layer to fill the bore of the tap, but do not allow any of this layer to run to waste. Remove any of the aqueous layer left in the stem of the separating funnel with a piece of cotton-wool, and run the organic layer, which is blue if phosphorus is present, into a 10-ml measuring cylinder fitted with a stopper.Wash the separating funnel with 1 to 2ml of the ethanolic sulphuric acid, add the washings to the contents of the cylinder, and adjust the volume to 10ml with ethanolic sulphuric acid. Insert the stopper, and mix thoroughly. Measure the optical derisity of the solution in 1-cm cells at 730 mp with a spectrophotometer; use as blank solution a mixture of 4 parts of isobutyl alcohol - benzene mixture and 1 part of ethanolic sulphuric acid. The volume at this stage is about 8ml. NOTES- For successful production of molybdenum blue from orthophosphate, the acidity should be between 0.5 and 1.5 N and the amount of neutral salts other than ammonium molybdate should not exceed 0.5 g in 10 ml of test solution.To fulfil these conditions, the initial phosphate solution, whether derived from a wet combustion during a deter- mination or from the standard solution in the preparation of a calibration graph, is made up to 14m1, including 3 5 m l of the ammonium molybdate solution, which is in 4 N sulphuric acid. The time of shaking with the stannous chloride solution is restricted to 10 seconds, because a longer time results in a decrease in the intensity of the final colour. I t is important that the procedure should be carried out as rapidly as possible from the addition of the stannous chloride solution to the production of the final solution. The final colour is stabilised by the addition of the ethanolic sulphuric acid and remains unchanged for a considerable time.The method is essentially that of Berenblum and Chain2 and was more recently described by Field and Laws3 Minor chan,pes in procedure have been included in the above.April, 19611 ORGANO-PHOSPHORUS INSECTICIDES IN VEGETABLES 265 DISCUSSION OF RESULTS A general picture of the application of the method to eighteen different insecticides is shown in the diagram on p. 249 and in Table I. Recoveries of mixtures of phorate with its oxygen-analogue sulphone a t various levels are reported in Table I1 and those of mixtures of parathion-methyl with Phosdrin in Table 111. Blank values obtained on vegetables and fruit are summarised in Table IV. From Table I it can be seen that recovery of insecticide is sometimes higher in the presence of the vegetable matter than in its absence.We believe that this is caused by displacement of the adsorbed insecticide by some part of the extracted plant material, which results in the insecticide leaving the column earlier than it would if no such substance were present. In general, the results are reasonable for the type of system studied. The plant material contains phosphorus many hundreds of times in excess of the amount contributed by the insecticide, and the greater part of this is removed by the column. It is very important that the stated rates of elution are not exceeded, as experiment has shown that the amount of interfering plant substance passing through the column grows rapidly as the rate of elution is increased.Most of the experiments were carried out on cabbage, as this material is readily available and is typical of the crops likely to be examined. Since fruit is also likely to be sprayed with insecticide and is often eaten without preparation, some results were obtained to illustrate the wider application of the method (Table IV). Gusathion and malathion are unlike the other petroleum-soluble organo-phosphorus insecticides studied in that they are not eluted from the alumina column by light petroleum alone. They are, however, eluted by a mixture of light petroleum and diethyl ether (85 + 15 v/v). Some of the malathion is sometimes eluted by the light petroleum before the application of the mixture, but the mixture removes it all. The “plant blank” is not excessive when the mixed solvent is used.Preliminary examination of the petroleum-soluble insecticides on the alumina column demonstrated that the various compounds were eluted by different amounts of solvent. The amount chosen (150 ml) is the minimum that will effectively elute all the insecticides studied except malathion and Gusathion. I t is possible that separation of the individual insecticides could be achieved by using a second column of alumina and a fraction collector, with the purified eluate from the first column as starting material. The-general utility of the method lies in the fact that it divides the insecticides into two distinct groups and is applicable as a general method of analysis to a t least eighteen well known insecticides; it is potentially useful for a far greater number of compounds. Although it is accepted that methods are available for determining chlorinated insecticides, such as DDT, BHC and dieldrin, it should be mentioned that these compounds will be found in the light petroleum solution obtained by the proposed method and that this solution could in fact be used as a starting point in the determination of such residues. With one exception, optical-density blank values obtained on the vegetables so far tested have not exceeded 0.050, which corresponds to about 0.10 p.p.m. of insecticide. It would be possible to deter- mine residues of the order of 0.10 p.p.m. in a 50-g sample. This paper is published by permission of the Government Chemist, Department of Scientific and Industrial Research. REFERENCES 1. 2. 3. Laws, E. Q., and Webley, D. J., Analyst, 1959, 84, 28. Berenblum, J., and Chain, E., Biochem. J., 1938, 32, 295. Field, K., and Laws, E. Q., Analyst, 1957, 82, 667. Received Octobev 13th, 1960

ISSN:0003-2654    

DOI:10.1039/AN9618600249

出版商:RSC    

年代:1961

数据来源: RSC