|
1. |
Proceedings of the Chemical Society. September 1959 |
|
Proceedings of the Chemical Society ,
Volume 1,
Issue September,
1959,
Page 241-284
Preview
|
PDF (5193KB)
|
|
摘要:
PROCEEDINGS OF THE CHEMICAL SOCIETY SEPTEMBER 1959 ~~~~ ~ -~ CHEMISTRY IN THE ANCIENT WORLD By 3. R. PARTINGTON IN considering the subject of “Chemistry in the Ancient World” two questions arise. In the first place it must be stated what is to be understood by “chemistry” in such a remote period and in the second place the parts of the ancient world of particular interest must be defined. Ancient applied chemistry included the extrac- tion of metals from their ores the formation of alloys the production of glazed pottery and other glazed articles and of glass and coloured pastes the preparation of pigments inks and salts dyeing with the use of mordants the ex- traction of vegetable oils the compounding of ointments and perfumes the production of in-toxicating drinks and the preparation of drugs.All these operations were carried out and the methods and results show that the ancient peoples were often very highly skilled in the tech- nical arts and that they were not at all primitive in this respect. In some cases a high standard of work was reached by mere patience as when a block of hard stone such as granite or porphyry was shaped into a thin-walled vase by the use of stone grinders and abrasives and without the assistance of the wheel but in many other cases the processes were executed in a way which leaves nothing to be supplied in order to make them rank with the best achievements of the modern craftsman. The culture of the Bronze Age was on a high level.The manual worker it is true was not very highly esteemed and his hard life is graphically described in some Egyptian papyri which show that he often wished to become a scribe or clerk in which capacity he had the opportunity of rising to a higher level of comfort and appreciation. The goldsmiths of Egypt were better thought of than other manual workers and even the god Ptah was not above assuming the title of “master of arts and artisans,” his priest being called “fore- man of artisans,” and “master of art.” It seems as if the finer work was actually carried out in temple workshops and the Egyptian priests prob-ably busied themselves in producing ornamental materials in much the same way as the monks during the Middle Ages in Europe.The outstanding achievements in applied chemistry in the earliest period were made in 241 some special areas. The principal regions of the ancient world we shall consider are Egypt Mesopotamia and Crete. Although iron was known in Egypt and Mesopotamia before 3000 B.c. and was in quite extensive use from about 1700 B.c. almost the whole period dealt with was in the Bronze Age since the beginning of an Iron Age is properly associated with the use on the large scale of iron weapons. The date 1000 B.C. B.C. Predynast ic gold silver lead cop-per (iron) glaze (glass) 3400 I Dynasty 3000 (tin-bronze) 2500 2000 XII Dynasty (tin) iron tools 1700-1500 X VIII Dynasty tin-bronze common tin glass factories cobalt glass indigo useful iron Ebers Papyrus (medicine) 1350-1200 loo0 Nubian iron industry Assyrian iron common in Egypt 650 I PROCEEDINGS marks the beginning of an intrusion of an iron- using race or races into Mediterranean sites long previously occupied by peoples of higher culture with a Bronze Age civilisation.This immigration called by the Greeks the Dorian invasion as far as their particular experience was concerned marked a definite break in the continuity of craftsmanship and a lowering of standards in many directions. The Classical period which lies Mesopotamia 1 Aegean etc. Sumer-akkad I Lagash Kish Ur gold silver lead (iron) cqper7 tin-bronze7 glaze (glass) lead-antimony bronzes (tin-bronze) (metallic antimony) Early Minoan Knossos copper gold silver lead tin-bronze purple dye glaze Mycenae glass tin-bronze coxnmon useful iron Amarna The Hittites letters steel weapons iron working in the tin oxide glaze (cobalt Black Sea region glass) (“Chalybes”) Tiryns lead glass coloured with copper Assyrian iron common Iron Age Classical Greece (zinc) Assyrian chemical tab- lets imitation gems I SEPTEMBER 1959 in the Iron Age is in some respects a period of decadence in the technical arts.Of the three regions to be considered Crete was least important. The older Cretan culture had greater affinities with Egypt than with later Greece; it was continued in the Mycenaean civilisation on the mainland and a similar culture in later Cyprus.As a whole this region was lack- ing in originality in the technical arts most of which were borrowed from Egypt in a period when there was close contact with the Aegean. The materials and processes known in antiquity are shown in the Table from which it is clear what an extensive range of metallic and non-metallic products was available. The Table gives a view of the Occurrence of metals and other materials requiring skill in applied chemistry for their preparation. Materials found only very rarely are enclosed in brackets. The dates of the earliest periods are provisional. In considering the materials we may begin with the metals. The earliest discovery of metals is still in many ways mysterious and has been supposed to go back to 4000 B.C.at least in Egypt. The old authors preserve fragments of legends in which the great inventors are gods and kings. In Ancient Egypt the god Ptah was re- garded as the inventor of metallurgy. In an inscription of Rameses I1 (1250 B.c.) at Abu- Simbel the god says to the king “I have wrought thy body of gold thy bones of copper thy vessels of iron”; this is the first definite reference to iron in a text in Egypt although the metal was known long before. King Sennacherib the Assyrian in 700 B.C. says he made great clay moulds at the command of his god and cast in them great bronze statues of bulls and lions weighing over 10,000 talents “as if making half-shekel pieces.” Other legends or near-history suggest that the working of metals was in the hands of members of special races feared but despised as practi- tioners of magic.The earliest known working in metals appears before 3500 B.C. in Egypt and Mesopotamia and rather later (2800 B.c.) in Crete and Cyprus. The Egyptians and Sumerians are rivals in the claim for the origin of metallurgy. The earliest known metal was probably gold encountered in alluvial washings although mining from the rock was soon in use. Copper has been found in some of the oldest Egyptian graves without gold. The earliest specimens of gold are unrefined and con- tain varying amounts of silver from 3 to 16% and usually some copper. When the silver reached 20% or more the resulting pale yellow natural alloy now called electrum which is harder than gold was given a special name in Egypt and this name asem occurs on an ebony tablet of King Menes (about 3400 B.c.).Electrum was a metal of peculiar sanctity in Egypt and was associated with the sun. The kings of the XVIIIth dynasty covered the summit-pyramids of obelisks with sheets of the metal. Queen Hatshepsut built “two obelisks of electrum whose points mingled with heaven their summits being of electrum of the best of every country which are seen on both sides of the river their rays flood the Two Lands [Upper and Lower Egypt] when the sun rises between them as he dawns in the horizon of heaven.” One is still standing (without electrum) at Karnak. Asem is also mentioned in the in- scription on the obelisk of Thothmes 111 now standing on the Thames Embankment and incor- rectly called “Cleopatra’s needle,” which may at one time have been plated with the metal.This king once received about four tons of electrum as tribute from Asia and Nubia. Gold seems to have been used in great pro- fusion in Egypt during the XVIIIth dynasty since a gold coffin in Tutankhamen’s tomb was over 6 ft. long and weighed 110.4 kg. and in the Amarna Letters (1375 B.c.) the King of the Mitanni in the Euphrates region writes to Pharaoh begging him to “send me so much gold that it cannot be measured for in my brother’s land gold is as common as dust.” He also men- tions that “thou sentest my father a great deal of gold.. . but the tablet thou sentest me was as if it were alloyed with copper.” The King of Babylon wrote to the Egyptian King Ikhnaton (1375-1350 B.c.) saying his gold was not as good as his father’s; from 20 minas of this gold put in the furnace only 5 minas of fine gold remained. These letters show that some kind of refining process was in use in Asia Minor about 1400 B.c. i.e. long before its use in Lydia in 550 B.c. the traditional date for the introduction of the process of refining gold. Pharaoh sent the base metal in the hope that it would pass as gold and the processes of testing and refining gold were perhaps not well known in Egypt in his time; almost pure gold first occurs in Egypt at PROCEEDINGS the beginning of the Persian period about 525 B.C.The method of purification of the gold ob- tained in Egypt by stamping the quartz rock and washing is thus described by Agatharchides (about 113 B.c.) “At last others take it by weight and measure and putting it with a fixed propor- tion of lead salt a little tin and barley bran into earthen crucibles well closed with clay they leave it in a furnace for five days and nights together after which it is allowed to cool. The crucibles are opened and nothing is found in them except the pure gold a little diminished in quantity.” The use of crude vitriols known at an early period in Egypt seems possible. The powder used in fairly recent times for the purification of gold and called the “royal cement,” consisted of brick dust calcined green vitriol and common salt.Silver was of little importance in Egypt but occurs in fairly large amounts in early Sumerian remains at Ur. It does occur in small quantities in predynastic remains in Egypt and generally contains appreciable amounts of gold and cop- per. The silver at Ur is in the form of useful objects such as tumblers as well as ornamental so that the metal was probably in fairly extensive use. Most early silver came from Asia Minor. The Hittites were particularly important in the development of the metallurgy of silver the ores of which occur in large amounts in Asia Minor although some of the old mines are now worked out. In the XVIIIth dynasty Egypt received much silver from this source and so probably did the Babylonians and Assyrians.Silver was used fairly early in Crete and there are interesting finds of artistic Phoenician silver on the coast at Byblos (modern Jebeil) going back to 2000 B.C. A silver coinage was early in use in Phoenicia and Palestine; the Egyptians used weighed rings of gold silver and copper which are shown as tribute. The metallurgy of copper in Egypt goes back to the early predynastic period. The metal was probably not native copper but was smelted from malachite which was used as an eye-paint from a very early period. The malachite was obtained from the peninsula of Sinai. It is readily reduced in charcoal fires and abundant remains of an early copper industry were found in this region. The wrinkled slag on the surface of molten copper is mentioned in a papyrus of 3000 BC.There is a copper axe-head of 34 Ib. of the middle predynastic period (4000 B.c.?). In the IInd dynasty the art of working copper had reached a high stage of perfection as the actual articles show. Equally remarkable are the highly artistic works in copper from early Sumerian sites per- haps as early as 3000 B.C. In both regions the copper was very pure and very good castings were made as well as hammered work. We do not know enough about the actual processes used by the metallurgists since the reports of excavations are often confused and in- accurate and many so-called smelting furnaces are probably merely the remains of fire-places or bakers’ ovens. We are better informed of the working of metal as there are good representa- tions of metal-workers’ workshops in Egypt for example one of the Old Kingdom (2980-2475 B.c.) showing the weighing of precious metals and malachite a furnace with men blowing the fire with mouth blow-pipes (probably reeds tipped with clay and formerly mistaken for glass- blowers’ blow-pipes) the cutting and hammering of metal and the putting together of necklaces and costly ornaments.Blow-pipes were used to a much later period down to the XVIIIth dynasty where they are shown although bellows were in use in Egypt at least as early as 1580 B.C. and are also represented. These were dish bellows still used in Africa being wooden or clay dishes covered with leather the skins being pressed down by stepping on them and raised by pulling strings.There are no valves the air entering through leaky places. The metal was melted in clay crucibles actual specimens of which are known and poured through funnels probably also of clay into the moulds. Quite large castings were made when a quantity of metal was poured simultaneously from a number of crucibles. Really good castings were first possible with the alloy bronze containing copper and tin which has a lower melting point than pure cop- per. Bronze occurs in small amounts in Egypt in the Old Kingdom but first became abundant in the Middle Kingdom (XIIth dynasty). It seems that the metals copper and tin were used separately and melted together to form bronze since they are shown brought to the furnaces in blocks of different shapes.The copper was in large hide-shaped blocks which were traded in the Aegean in the Cretan period actual speci- SEPTEMBER 1959 245 mens being known. In some Egyptian representa- tions smalkr blocks of a material called dhty perhaps tin are shown carried by Cretans. The Cretans are often shown in the XVIIIth dynasty bringing precious metals made into vases etc. into Egypt. It is possible that they brought the tin. Later on the trade in tin appears to have been in the hands of the Phoenicians but mari- time trade was flourishing before they appeared and their importance in the early period has been over-estimated. The source of the tin for all early bronze is a matter of speculation but it is possible that it came at first from some place in KhurBs2n in Persia called by the older authors Drangiana.Strabo (A.D. 7) says tin was found there and there are later accounts of it although the mines are worked out. It is also possible that the early tin used in Egypt came from the mountains of the Kasrwan district in Syria not far from the very ancient port of Byblos which traded with Egypt at least as early as the 1st dynasty. It has also been suggested that the tin came ultimately from Central Germany where it is supposed to have been exploited before 2500 B.C. This theory may hold for the bronzes of Troy 11 which was apparently in close relation through the north of the Balkans with the Danube basin and Hungary.Bronzes from the Second Town of Troy (2400- 1900 B.c.) contain up to 11% of tin. British tin may have been used from about 1500 B.C. The early Egyptian bronzes are poor in tin and normal bronze containing about 10% of tin came into general use there about 2000 B.C. Later there was a tendency to replace tin by lead a metal known from the predynastic period and some early Egyptian bronzes contain lead. Later Mesopotamian bronzes contain lead and anti- mony in place of tin. Tin itself occurs in Egypt from the XIIth dynasty (2000-1785 B.c.) and was perhaps known earlier although no actual specimens of still earlier tin are known. Since the Sumerian culture is closely related to if not the same as that at Mohenjo-daro and HarappB in the Indus Valley the possiblity of the source of early tin being in Drangiana which connects Chaldaea and India seems to be strengthened.The Sumerian and Egyptian metal workers had different techniques. The early Sumerians’ axes are cast in the more fusible bronze and are socketed whilst the Egyptians used flat ham- mered axes of the more difficultly fusible copper not socketed but bound to the haft by thongs and such copper axes were also used by the Sumerians. Copper and bronze occur in Crete later than in Egypt and Mesopotamia so that Crete presents nothing noteworthy in this respect. Cyprus which was in relation with the very old site of RBs Shamra on the mainland and thus formed a bridge for the transmission of materials and ideas into Palestine had a very old copper industry going back at least to 3000 B.c.the ores being smelted on the island. Bronze was made there somewhat later. Even more important was the brass industry of Cyprus. Egyptian brass of the predynastic period was found by Petrie. There is no definite knowledge of early brass in Mesopotamia but some of about 1400-1000 B.C. was found at Gezer in Palestine containing up to 23% of zinc. Brass is barely mentioned by Plato (429-347 B.c.) who calls it oreichalkos literally “moun- tain copper,’’ which he says was only a name in his time. Aristotle (384-322 B.c.) did not seem to know of it but it is mentioned more than once as an Indian or Persian gold-coloured metal in the book On Marvels written by some member of Aristotle’s school and probably completed in the first century A.D.Brass was well-known to the Romans the best authors using Plato’s name for it (orichalcum) although it was also called aurichalcum “golden copper.’’ In the Roman period the brass industry of Cyprus is mentioned by Pliny Dioskourides and Galen although they could not obtain any detailed information about it (the process being apparently kept a trade secret). The rather extensive use of brass by the Romans who established the brass industry at Cologne about the beginning of the Christian era and made brass known to the rest of Europe may have been derived from Cyprus. The zinc ore in Cyprus was probably called cadmia a name also given to varieties of zinc oxide produced in the smelting furnaces and used in medicine.The smelting furnace was built into a house of two storeys the roof being open to the air. In the wall was a charging door and the apertures for the bellows which were in a separate house. The fuel was put in first and kindled then the cadmia broken into small pieces and more fuel. The zinc oxide driven off was collected in the upper storey like fleeces of wool. This process is the one used for making zinc oxide not brass. Galen was told the brass furnaces were not working when he happened to be there. The same type of furnace was used for making brass in Rammelsberg in the seventeenth century A.D. and small quantities of metallic zinc were collected there from the cooler walls of the furnace.The same process seems to be men- tioned by Strabo. The passage which seems to be the first account of the production of metallic zinc is very obscure and is almost certainly cor- rupt. With two slight emendations however it becomes perfectly intelligible. It reads as follows “There is an ore found at Andeira which when roasted produces an ash. When this is put into a furnace with coal it goes away as mock-silver which when added to copper forms the alloy called brass.” The last statement shows that Strabo’s “mock-silver” could not have been any- thing but zinc since brass was well known at the time when he wrote and only zinc forms brass with copper. The Greek silver mines at Laureion are rich in zinc ore and zinc was probably ob- tained there more or less accidentally.The pro- duction of zinc by merely heating the ore with coal in a crucible is described as late as 1743 by an English author. The early history of zinc is much in need of revision modern historians of chemistry having added to its obscurity by misplaced scepticism and imperfect knowledge of the processes they describe. At least seven early specimens of zinc are reported most of them analysed. There are hollow silver bracelets filled with zinc from the island of Rhodes of about 500 B.c. a plate of nearly pure zinc from Athens of about 250 B.c. a large bar of zinc found in the ruins of a Roman villa in Champagne an idol of over 80% of zinc found in a Dacian hoard of the first century A.D.at Tordosch (Siebenbiirgen) a piece of zinc found in a Roman house in Bulgaria zinc plates from Bosnia and the pediment of a fountain at Pompeii (which has disappeared). There seems no reason to doubt that the metal was occasion- ally obtained. There may have been an early brass industry in Persia where there are rich deposits of calamine but the accounts of the industry there are quite late. The working of iron probably originated neither in Egypt nor in the Euphrates valley; old PROCEEDINGS Greek and Jewish legends point to the Chalybes tribes related to the Hittites and living in the regions between the Caspian and the Black Sea as the first workers in iron which they traded with the Egyptians and Babylonians.Iron occurs in Egypt in the predynastic period as a valuable metal and since the rusted beads found by Petrie are rich in nickel they are prob- ably of meteoric iron. An early Sumerian iron blade found at Ur contained over 10% of nickel and was probably of meteoric metal. Early iron found at Tall Asmar however is free from nickel and is probably terrestrial iron. The process used in making iron is not known but it probably involved the reduction of pure oxide ore with charcoal in a shaft furnace pro- vided with bellows a spongy mass of metal being obtained which was afterwards hammered into bars whilst hot. This process is still carried on by African tribes no flux being used the metal being subjected to heat treatment after form- ing into small bars.Ancient steel was probably obtained by some kind of cementation process the iron being heated in contact with car-bonaceous material. Micrographical examina- tion of Egyptian iron of 1200-800 B.C. indicates that the metal was carburized and quenched. Although there are some Greek references to cast iron it seems improbable that the iron could ever have been melted on the large scale in the furnaces then available. Iron was in use in Egypt from about 2000 B.C. but only sparingly. The source was probably the Hittites who were very skilled in the working of iron. A text from Boghazkoi of the fourteenth century B.C. speaks of “black iron from heaven,” and iron smelting was well known in Syria Palestine and Asia Minor in an early period.Petrie in 1927 found extensive remains of an iron foundry at Gerar near Gezer in Palestine dating from about 1200 B.c. and a steel dagger Of 1350 B.C. Rameses 11 obtained presents of iron from Kitzwatna probably the Black Sea region. In a Boghazkoi text of about 1250 B.c. Rameses asks the Hittite king for iron and receives the reply “As for pure iron [steel] about which you have written there is no iron in my warehouse at Kitzwatna now and to make it would be in- convenient but I have given written orders that it shall be made. As soon as it is ready I will send SEPTEMBER 1959 some; at the moment I send only a dagger.” The Pharaoh had come to know the value of steel weapons in his wars against the Hittites.There was a well-developed iron industry among the later Assyrians and the remains of the palace of Ashurnazirpal at Nimrud (885-860 B.c.) in-cluded large masses of iron. This king also ob-tained large amounts of iron as tribute from Hittite kings. The later Assyrians probably owed their period of military success to their possession of iron weapons. The prehistoric Egyptian pots of soft body faced with red “haematite” (the Egyptologist’s name for a clay rich in ferric oxide) were usually baked upside-down with the brim covered with fuel the process giving a black rim and interior partly due to carbon from smoke and partly to reduction of ferric oxide to magnetic oxide Fe,O,. It has been said that most pots are blackened by carbon but some contain both carbon and magnetic oxide of iron as I have found by analyses.In some experiments by Mercer a well-defined red with a sharply defined black top never achieved except in Egypt and at one period in Cyprus under Egyptian influence was obtained as follows. The pot rubbed with “haematite” and water was set with the mouth in fine sawdust in the centre of which just under the vessel was placed a piece of resin the size of a chestnut. Over the whole was an arch of wire netting and over this a fire of dry rye straw which burned for forty-five minutes. Two kinds of heat smokeless and smoky are required and when the pot becomes red-hot the sawdust ig- niting last and smothering its own flame ends the baking in a smoky heat.Primitive industries have often proved very difficult to imitate but this was achieved by Mercer after many others had failed. The ancient pigments were mostly natural an exception being the Egyptian blue. This was used in the predynastic period but its preparation is first given by Vitruvius from an Alexandrian source. He says “Sand and natron (native Egyptian soda) are powdered together as fine as flour and copper is grated by coarse files over the mixture. This is made into balls by rolling with the hands. The dry balls are put in an earthen jar and this jar put into a furnace. When the copper and sand have coalesced in the intense heat and the separate things have disappeared the colour caerulium is made.” This colour was carefully examined in specimens from Pompeii by Sir Humphry Day who obtained a similar product by heating sodium carbonate silica and copper filings.This did not give the true Egyptian blue which has been shown to contain a definite crystalline compound Ca0,Cu0,4Si02 which is formed by heating fine sand calcium carbonate copper carbonate and a little alkali carbonate as a flux for several hours in the temperature range 830-90O0. Below and above this limited range it is not formed and if too much alkali is used only a slag results. There was no glazed pottery in Egypt. Pottery was glazed in Assyria in the 15-12th centuries B.C. in the so-called egg-shell ware and was brightly coloured but this product died out later only coarse pottery being made.Lead glaze ap- pears both in Egypt and Mesopotamia about 950 B.C. In Assyria also in the period 900-600 B.c. very artistic glazed bricks were made the blue colour being copper silicate the red cuprous oxide the white a tin oxide enamel (the earliest known example of enamel) and the yellow lead antimoniate (“Naples yellow”). The bright yellow of some Egyptian glass may have been obtained with lead and antimony. Glazed pottery was made in Crete about 2200 B.c. and this is perhaps the earliest known. In Egypt the body was of powdered quartz; the glaze was probably first made as a frit then powdered mixed with water and painted on the object which was then dried and fired. This is the process used by modern forgers in Cairo who carry out the firing in closed boxes of copper or in pots to prevent discoloration by smoke and their products are the same as the ancient Egyptian being made by the same process.Glass itself was known very early in Egypt and Mesopotamia. There are finds of glass in both localities before 3000 B.C. in the form of beads and a large piece of blue glass of about 2400 B.C. was found at Abu Shahrain (ancient Eridu) in Mesopotamia. The Egyptians were very skilled in the making of glass; the remains of a factory with a profusion of glass objects was found at Amarna (1350 B.c.). Apparently quartz sand and alkali were made into a frit in clay pans some of which were found. The alkali was native soda found in Egypt.The frit was then melted in small crucibles 2-3 in. deep and wide. Petrie says the glass was never cast in moulds but was pressed when viscous into the moulds; others say it was pressed over clay forms or the clay forms dipped into the glass or glass threads were wound round over the mould and then “dragged” while soft so as to give a wavy pattern. The colours are very good. The blue was usually copper and the red cuprous oxide; the yellow may have been lead antimoniate lead and anti- mony being found in XIXth dynasty yellow glass. Cobalt occurs in some Egyptian glasses of the fourteenth century B.C. Colourless glass was made in Egypt from about 1400 B.c. probably from very pure materials although some glass found at Amarna contains manganese and had probably been decolorised by pyrolusite.The so-called murrine vessels (vasa murrina) of the Roman period about which so much has been written were probably of two kinds a “natural” made from fluorspar and an “artificial” made from coloured glass. Glass objects were ground but the wheel was used in Egypt only in the Roman period. Analyses indicate that Egyptian glasses are of good quality and of relatively high melting point. Blowing glass was unknown in ancient Egypt and seems to have been invented in Sidon about the beginning of the Christian era. Both Seneca and Pliny speak of blown glass as something new in their time. Assyrian glass is represented by the glass bottle of Sargon I1 (722-705 B.c.) which was almost certainly of native manufacture.Blue glass coloured with cobalt occurs in Mesopotamia. About two-thirds of the so-called lapis lazuli found in Nippur probably of the fourteenth century B.c. was a blue lead glass coloured with copper (1.94% CuO) and cobalt (0.93% COO). This blue glass imitating lapis lazuli was a speciality in Babylonia where it was called uqnu. Assyrian cuneiform tablets of 668-628 B.c. found in the library of Assurbanipal and now in the British Museum contain lists of stones liquids etc. Some deal with the production of false gems and coloured glass pastes. They have been rather hastily called “alchemical” texts but in reality they are technical treatises. One tablet contains instructions for building the furnace with an appropriate ritual and for the produc- tion of coloured glasses.It seems as if the tradi- PROCEEDINGS tions preserved in these chemical tablets go back to much earlier times since a similar tablet of the seventeenth century B.c. found at Tall Umar on the Tigris is written in a difficult style perhaps intentionally cryptic and describes the produc- tion of a blue glaze containing lead and copper by an unnecessarily complicated process. The collection of Egyptian recipes in the Greek chem- ical papyri of Leyden and Stockholm written in Alexandria about A.D. 300 may go back to much earlier but still unknown originals. On the side of organic chemistry perhaps the use of dyes is the most interesting. Dyeing was very old in Egypt where red yellow and green clothing is shown in the oldest pictures.Ancient Egyptian dyes established by analysis are indigo and safflower (Carthamus).The use of indigo is especially interesting since the Romans who then obtained it from India were unable to get it into solution and used it only as a pigment (Pliny puts it among the minerals). The Egyptians seem to have been acquainted with the use of mordants since Pliny mentions that they first impregnated the fabric with chemicals (probably alum and vitriols) which did not give it any colour but when it was put into a cauldron of boiling dye it was drawn out showing various colours although there was only one colour in the cauldron. Parti-coloured fabrics some printed were found in Egypt at Ikhmim from about Pliny’s time and the process probably went back to an earlier period.The use of the so-called Phoenician purple (a dibromoindigo) obtained from mollusca appears to have begun in Crete where large heaps of the murex shells have been found. In later times dyeing in purple was carried out on the large scale in Tyre and Sidon. The process is described by Pliny and has been successfully imitated in modern times. Two kinds of mollusc were used the buccinum (with trumpet-shaped shells) and the pelagium. The juices were combined mixed with a little salt and boiled in a lead cauldron. The liquor was tested by putting in a piece of washed wool and the boiling continued until this came out the required shade which was a very dark red or crimson rather than what we now call purple.The wool was then put in left for five hours taken out and immersed in a new bath producing the so-called “double-dyed” purple which was very expensive. Pliny says 1 lb. of double dyed wool cost E31 5s. It had a rather unpleasant fishy smell. SEPTEMBER 1959 In Mesopotamia we have the beginnings of the petroleum industry. Bitumen occurs in some of the oldest Sumerian remains and was obtained from Hit on a river of the same name flowing into the Euphrates about a hundred miles north of Babylon. It was extensively used by the Babylonians for cement for asphalt drains floors etc. Bitumen was also obtained in other places. From these sources (including Kirkuk) was also obtained naphtha (Babylonian naptu) or crude petroleum used for lamps.There were two kinds black and white. Strabo says the naphtha of Susiana when brought near a fire ignited and could only be extinguished by a large quantity of water or by mud vinegar alum and glue (an ancient “foamite” process). Pliny speaks of a white naphtha which Dioskourides calls “filtered asphalt.” Perhaps it was refined by filtration through kaolin or some kind of ab- sorbent earth although the possibility of distilla-tion at this time cannot be dismissed. As in the case of zinc there is no good history of distillation. Glass distillation apparatus (alem- bics the Greco-Egyptian word ambix being so read by the Arabs) was apparently invented early in the Christian era in Alexandria by Mary the Jewess.She belonged to one of the small gnostic circles there in which the so-called Divine Art which became Chemistry arose. She first separated the distilling flask cooled in its upper part from the receiver by a tube. A primitive form of distillation in which vapour from a liquid heated in a pot condensed on a cool lid is described rather earlier by Pliny and Dioskour- ides and may have been known to Aristotle if indeed it does not go back to the Babylonians. We may be sure that Mary and the early chem- ists distilled all kinds of liquids perhaps includ- ing wine although the early history of alcohol is still very imperfectly known. There is no definite indication that alcohol was known in the period we are considering.The ancient achievements in applied chemistry present an impressive picture. A history of chem- istry which omits this part is incomplete and if the ancient world had little to teach us in theory its contributions to the practical side of chemical knowledge are worthy of re cognition. COMPUTERS IN CHEMICAL PROCESS DEVELOPMENT By R. V. THOMAS (IMPERIAL INDUSTRIES CENTRAL LABORATORY, CHEMICAL LIMITED INSTRUMENT READING) IN less than two decades the needs of the defence and nuclear industries of the U.S.A. and Britain have resulted in the development of automatic computers of tremendous power and flexibility. These machines were required to help in the solution of a wide range of technical problems such as stress flutter control and interception calculations in the aircraft and missile fields and criticality resonance escape probability and power and temperature transients for the nuclear industry.Their use has made possible calculations that would otherwise have been quite impracticable. Appreciation of the potentialities of these machines by the Chemical Industry has grown rapidly in the last five years and there have been many successful applications many of which unfortunately are unpublished. Early applications in the Chemical Industry included the analysis of experimental data to deduce gas-law constants kinetic calculations for chemical reaction systems reduction of mass-spectrometer readings and quantum-mechanical predictions of thermodynamic properties from molecular structure.This article endeavours to survey the uses that have been made of computers in chemical process develop- ment. However as many chemists will have little or no knowledge of computers Part I will attempt to explain the very different principles of operation of the two types of computer analogue and digital and how these differences affect their use and application. PART r Analogue computation is based on the idea of finding a physical system which behaves in an analogous manner to the one whose behaviour is to be investigated but which is easier to build and investigate than the original system. There are many more or less specialised analogues that have been used widely in fields as diverse as electron optics and aerodynamics ; the conduc- tion of electricity in solids or electrolytes is the basis of many of these.When a particular type of problem for which a specialised physical analogue has been developed occurs frequently it is sometimes worth considering the purchase or construction of the analogue although the latter process often calls for great ingenuity and may prove to be fraught with pitfalls. At some stage in comparing the behaviour of the pro- posed analogue and the original system the equations governing the two will be formulated and compared. If these equations are identical then the behaviour of the analogue will represent exactly that of the original system. Any differ-ences however will result in deductions from the behaviour of the analogue being more or less misleading.Any physical system whose equations correspond sufficiently closely to those of the original system can be used as an analogue providing it is easy to set up and use. Many physical systems are expressible by simultaneous ordinary differential equations and the general purpose analogue computer (usually abbreviated to analogue computer) or electronic differential analyser is designed to be adaptable to represent such systems. Although best suited to these systems it can also be used for many other purposes e.g. for the solution of non- differential and partial-differential equations. In the computer the independent variable of the problem is represented by time and the dependent variables by voltages the variations of the latter being displayed by a cathode-ray tube or plotted on paper by some electromechanical device.Digital computers are essentially counting machines and operate by counting discrete events. All numerical operations such as multi- plications division the evaluation of trigono- metric and other functions differentiation and integration can be expressed in terms of the two basic operations of addition and subtraction and a modern electronic digital computer can per- form these basic operations with extreme rapid- ity and great accuracy e.g. pairs of ten-decimal digit numbers can be added at the rate of about a million per minute. This very high speed of calculation can only be used efficiently by design- ing the machines so that they can store the numbers required and the sequence of instruc- tions to be obeyed in a calculation-the me.PROCEEDINGS It is important to stress that whether an analogue or a digital computer is used the first stage is the formulation of the problem in precise mathematical terms. Once the problem has been formulated in this way it is comparatively easy for an expert to decide whether analogue or digital computation is most suitable. If the resulting equations are a set of simul- taneous ordinary differential equations and an analogue computer is to be used the approach is to establish on the computer a physical system the behaviour of which is governed by the same equations as the problem.The independent vari- able is represented by time in the computer with a suitable scale factor and the time variations of selected voltages in the computer must be the same as those of the problem variables with respect to their independent variable apart from known scale factors. It must therefore be pos- sible in the computer to perform all the opera- tions on voltages that are performed on the original variables in the original equations of the problem the operations required are addition and subtraction multiplication and division by constants and other voltages and the production of functions of voltages and their integration with respect to time which last as will be seen later replaces differentiation with respect to the independent variable in the original equations.The most essential part of an analogue com- puter is the high-gain drift-corrected electronic amplifier which is used with suitable feed-back arrangements for addition subtraction multipli- cation by constants and integration and without feed-back has many other uses. Without feed- back this amplifier usually has a d.c. gain of about lo7to los dropping to about lo4to lo5at 100cycles per second. It draws neghgible current from the source of the input signal although it is capable of supplying a considerable output cur- rent. It is usually represented by the symbol shown in Fig. 1. If the input voltage viis applied FIG.1. Symbol for a high-gain amplifier. through the resistance ri and the output voltage v is fed back to the input terminal through the SEPTEMBER 1959 resistance Y as in Fig.2a then it can be shown that v = -(r,/r,)v = av . . . (1) Thus apart from the reversal of sign the output voltage bears a fixed ratio to the input voltage; this ratio being determined only by the ratio of two resistors can be altered very easily. This arrangement is represented by the symbol of Fig. 26. With equal resistors the output and in- put voltages have opposite signs and the amplifier is then called an “inverter”. If several input voltages are applied to the amplifier through separate input resistors as in Fig. 2c with feed-back as before it can be shown that I? -Krolr3v1 + (ro/rdv2 + - (ro/~m)vrnl = -[a,v + a2v2+ .. . a,v,] . . . (2) The arrangement is called a “summing” amplifier or “summer’y and is used to add voltages and to multiply them by constants before addition. On some computers a, a2 etc. can be altered by changing r, rz etc. On others a, a, etc. are fixed e.g. at 1 1 1 5 5 10 10 and adjustable potentiometers are used; this is more convenient and accurate though expensive the symbol is shown in Fig. 2d. Subtraction of a voltage is achieved by inverting it before feeding it to the appropriate summer input terminal. Both of the circuits of Figs. 2a and 2c can be made to integrate the inputs with respect to time by replacing the feed-back resistors by capacitors C. The equations are then V = -(l/rlC) v,.dt = -k v,.dt + v,’ (3) J 5 I= and 17 -r(klv + k2v2+ .. . . + 0 k,v,)dt + v,’ . . . . . . (4) where k = l/rlC k = 1/r2C,etc. and yo‘ is the initial condition i.e. the value of v at t = 0 which is set by ensuring that the capacitor has the appropriate charge immediately before in- tegration commences. Symbols for integraters with fixed gains are shown in Figs. 2e and 25 Multiplication by constants smaller than unity can also be performed by potentiometers (Figs. 2g and 2h) for which the equation is v,=cvf . . . . . . (5) 25 1 Multiplication by voltages requires more com-plicated equipment. In the servo-multiplier one of the input voltages is applied to the winding of a potentiometer the slider of which is moved by a servo-mechanism to a position proportional to the other input voltage with the result that the output voltage tapped from the slider is then proportional to the product of the two inputs.Several Potentiometers can be driven by one servo-mechanism so that several voltages can be multiplied by a single voltage if required. The symbol of a servo-multiplier is shown in Fig. 2i. Its equation is v a vlv2 . . . . . . (6) There are also several types of purely electronic. multiplier in wide use which because of the absence of moving parts are faster in operation than servo-multipliers. Division of voltages can be achieved by combining a multiplier and a high-gain amplifier in a feed-back circuit. There are several methods of function generation some purely electronic others involving mechanical movement the problem is to produce a voltage which is a specified function of another voltage.In one common arrangement regularly spaced taps into the winding of a servo-multiplier potentiometer are used to produce a voltage distribution representing the desired function. Multipliers and function generators are much more expensive than amplifiers and potentio- meters; however most of the problems for which chemists and chemical engineers might consider using an analogue computer would require them. As an example we will take the equations for the catalytic oxidation of ethylene solved on the N.P.L. mechanical differential analyser in what was possibly the first application of a mechanical differential analyser to a problem in chemical kinetics.The mechanical differential analyser which was first suggested by KelvinP2 was developed by Bush in the late 1920’~.~ It is very accurate but cumbersome and expensive and is now largely superseded by the modern electronic analogue computer. Fig. 3 shows the equa-tions and outlines of a suitable form of analogue; the precise form would depend on the numerical values of the constants and the ranges of the A. M. Wood J. Crank and G. H. Twigg Trans. Farachy SOC.,1948 44 256. a W. Thomson (Lord Kelvin) Proc. Roy. Soc. 1875,24,269. a V. Bush J. Franklin Inst. 1931 212,447. PROCEEDINGS n9 c-s-4 3I $ h U t3 L h s n2 h !? I" Le h SEPTEMBER1959 253 APPENDIX “Mercury” autocode programme for integration of equations of isothermal catalytic oxidation of ethylene.Programme tape Chapter 0 a-+9 Variable directives These specify that a to a,,yl to y3 z1and zz Y-3 will be used in the programme. z+2 Note.-The computer reads and stores the whole Programme Tape before commencing to obey the stored programme at the first instruction. (1) halt 1st instruction Computer halts for operator to insert separate data tape. Restarted manually by operator. i = 1(1)9 Computer reads first nine numbers on data tape i.e. Ao Al At, read (a,) B Po,P,,P2,8 and R-l and stores them in a,,a .. . a, repeat i i = 1(1)3 Yt = %+4) yl yz and y3 set to initial values of po p, and pa repeat read (x) x set from data tape (initial value of t) read (h) h set from data tape (increment of t i.e.step length) read (4) q set from data tape (qh = print-out interval) read (r) r set from data tape (rqh = period of integration) i = l(l)r j = l(l)q From the present values of po pl and p2 at t the intstep (2) computer calculates* new values at t +h. Repeated repeat q times j = 1(1)3 Repeated Prints values of po pl p2 at t + qh ’rtimes ?Y3 = Y3 repeat ?zl= 2a -2y2 -2ys ?z2= a + a6a9-ylag Calculates and prints values of p3 and 8 at t + qh -0-5y3ag-1*5z1ag repeat jump 1 Computer jumps to instruction “( 1) halt”-end of klculation. (2) z1= 2a -2y2 -2y3 P3 = 2p1 -2P1 -2P2 z2= a8 + a6an-ylaQ 8 = 8 + P,/R -Po/R -PJ2R -3p $2R -0.5y3ag-1*5z,a Function fl = 2alYIZ2 -alY1 subroutine b = a2y2z2 f2 = -a4y2z2z2 -b f3 = b -a3y3 592 0 close Indicates end of Programme Tape (see previous note) Data Tape This consists of the numerical values of A, A, A2,B Po p1,p2 80 R-l x h 4,and r * For this the computer uses a Runge-Kutta numerical integration procedure in conjunction with the function sub- routine specifying the method for calculation of derivatives.variables. Three integraters five summing ampli- fiers eight potentiometers and a servo-multiplier would be needed. This is a small problem by modern standards. Chemical process develop- ment problems may need fifty or more amplifiers and nuclear energy and defence problems not infrequently need several hundred amplifiers. Several American chemical companies have analogue computer installations with 100 or more amplifiers and 20 or more servo-multipliers and there are several commercial analogue com- puters having 50 or more amplifiers available on a daily rental basis at computing centres in this country or nearby.The time scale used on an analogue computer depends on limitations of the analogue equip- ment and the nature of the problem. In this case po,pl and pz change slowly the reaction taking about 1 hour for completion so a time scale of about 100 1 would be appropriate i.e. 1 minute of computer time would represent 100 minutes of reaction time so that a run would take about 35 seconds. The alteration of parameters and initial conditions in a small analogue such as this is extremely easy and it would take only a few minutes to calculate and alter the potentiometer settings.The results of each run i.e. curves of po,pl pz p3 and 8 against time would be re-corded on paper by an electromechanical plotter. It would take only a few hours to run through a very large number of combinations of para- meters and initial conditions. The accuracy would be of the order of 1 yo,though this depends on the computer and recording equipment used. The accuracies quoted by manufacturers for in- dividual analogue computer components such as feed-back amplifiers potentiometers and multipliers vary from 0.01yo to ca. 1yo,and the higher accuracy though expensive is necessary to reduce accumulations of error in large problems involving perhaps 100 or more amplifiers.Once designed or programmed an analogue can be set up on the computer whenever required. This setting-up splits into three stages (1) making the connections between units; (2) setting the potentiometers and function generators; (3) eliminating mistakes and faults (“de-bugging”). On some computers the connections are made on a patch-panel which can be removed PROCEEDINGS from the computer and stored for further use. This obviously eliminates stage 1 and much of stage 3 when the analogue has to be set up again on the computer and is very useful for large problems. Further facilities which are very desir- able for large problems are automatic setting of potentiometers and methods of preserving function-generator settings.There are several good textbooks on analogue comp~ters.~ For engineering reasons nearly all digital com- puters use the binary system which has only two digits 1 and 0. This is because a device to repre- sent or store decimal digits has to be capable of assuming any one of ten distinguishable and stable states whereas two such states suffice for the binary system e.g. conduction and non- conduction in a thermionic valve or transistor or saturation in two directions in a magnetic device. In the binary system the equivalent of a number such as 5-5 is 101.1 for 101.1 = (1 x 1010) +(0 x 101) + (1 x 100) +(1 x 10-1) in binary notation = (1 x 22) +(0 x 21) + (I x 20) +(1 x 2-1) in decimal notation = 5.5 Fortunately all binary computers are pro-grammed to make the necessary conversions between the decimal and binary systems during the input and output of numbers.Within a typical computer numbers are represented by 30 binary digits known as a “word,” to a precision of about eight decimal places. A typical computer has the following sections the store; the arithmetic unit which is capable of adding or subtracting two thirty-digit numbers in lo4 second; input and output facilities using magnetic tape punched cards or punched paper tape; and interconnections between these and a control unit. The store is divided into a large number of addressed locations each capable of storing a word. The control unit arranges the interconnections between the several sections of the computer e.g.it can route incoming numerical data from the input device to the re- quired storage locations send two numbers from the store to the arithmetic unit and send the result of the arithmetical operation performed there back to another storage location and print E.g. G.A. Korn and T. M. Korn “Electronic Analog Computers,” 2nd edn. McGraw-Hill New York 1956. SEPTEMBER 1959 out the results at appropriate points in the calculation. The calculations needed to solve the problem on the computer must first be broken down into a sequence of simple instructions or “pro-gramme” for the control unit to obey. These are then coded in the standard numerical form used in the machine so that they can be stored and transported in it in the same way as numbers.The programme is then fed into the machine through the input device and stored sequentially together with the numerical data needed in the calculation. The control unit then refers to the store for instructions commencing at a location specified by the programer and working on through the sequence. It is often necessary when programming to break into this blind obedience of the stored sequence of instructions to send the computer to repeat a section of the programme with different numbers or forward to miss a section. Whenever this is necessary instructions of the form “instead of going to the next storage location in the sequence go to storage location x for your next instruction,” or “if the number last calculated was negative go to storage loca- tion x for your next instruction,” where x is a number specified by the programmer are written into the programme.These are known as “jump instructions” and together with a further facility which allows the programmer to use a form of algebraic notation for describing the addresses of stored numbers the “variable address” facility enable him to write flexible and econom- ical programmes. All machines have “library” sub-routines of instructions for such common operations as multiplication division square- roots logarithms trigonometric functions etc. which the programmer can use. There are also ‘‘library” programmes for such mat hema t ical operations as the solution of simultaneous linear equations matrix operations statistical work etc.The extent of the library varies from machine to machine. Storage of numbers and instructions is normally at two levels of accessibility in com- puters designed for scientific work the high- speed store e.g. magnetic cores from which a word can be taken in about second and the slow store e.g. a magnetic drum whence re- trieval may take about second. High-speed storage is expensive and is usually kept to a minirnum especially in small computers hence a small computer with a total storage of about 500 words may have a high-speed store of only about 10 words whereas a large computer costing about five times as much as the small one might have a total storage capacity of 20,000words with a high-speed store of lo00 words.Because of this as well as being capable of dealing with larger problems than small ones large computers are usually faster. Most computers are fixed- (binary) point machines i.e.,although the precision of numbers is about nine decimal places it is necessary to keep their magnitudes between 5 1. It is then necessary for the programmer to choose suitable scale factors for the problem to ensure that all numbers arising during the calculation satisfy the above limitation-and this can be a tedious business. Many Computers can be programmed to adjust the scaling automatically but this usually slows down their operation appreciably. A few computers have been designed for floating operation.In one the Ferranti “Mercury,” numbers can have any magnitude between 1070 with a precision of eight to nine decimal digits. Preparation of a problem for a digital com- puter consists of two distinct stages numerical formulation and the actual programming. In the first stage the problem is expressed in mathe- matical form e.g. as a set of differential equa- tions numerical data are collected a suitable numerical procedure for solving the problem is selected and the most suitable computer chosen. The second stage consists of expressing the chosen method of solution in computer language or machine code and may take anything from a few days to several weeks. Machine codes vary in complexity but even the simplest takes several days to learn and several weeks to become familiar with and the time taken if the originator does it himself or its cost if he passes it on to a professional programer can be a serious obstacle especially if the programme when written is only to be used once or perhaps a few times as is the case in many research projects.This difficulty has been recognised and for several machines there are now available simplified coding procedures which are both quick to learn and easy to use. As an example the appendix gives a Mercury Autocode programme for the numerical integration of the reaction kinetic equations considered previously which took only 40 minutes to write. This code is designed to resemble normal mathematical language as far as possible and can be learned in a day or two.S The computer uses a conversion programme to enable it to translate the autocode programme into precise machine instructions.The penalty paid is that the speed of the machine is approxi- mately halved but in many cases this is in- significant compared with the advantages result- ing from quick and easy programming by the originator of the problem. Digital computers are capable of a very much higher order of accuracy than analogue com- puters and this is especially noticeable when a problem variable covers a very wide range of values. On a digital computer this normally presents little difficulty but on an analogue com- puter the maximum signal voltage is limited to about 100 v and trouble is experienced with very small signal voltages so that a range of 100 1 is about the limit that can be used and even then gross errors may appear.It is possible that this range can be extended artificially by using the logarithms of problem variables rather than the variables themselves. The equations programmed in Fig. 3 are dp,/dt = -Aopo(1 -20) . . . . . . (4 dpl/dt = -(Alp16 + Bp18'). . . . . (B) dp2ldt = -(AZp2 -AlpiB) . . . . . (a p3 = -2(p1 +p2 -Pi). . Re = -(PO +~2/2+ 3~312--(D) Po -ROO) . . . . . . (a where po = partial pressure of oxygen (initially Po) p1 = partial pressure of ethylene (initially Pl) p2 = partial pressure of ethylene oxide (initially P2=0),p3= partial pressure of carbon dioxide 8 =fraction of catalyst surface occupied by oxygen (initially go) R = equivalent pressure of absorbed oxygen when 0 = 1 and A, A 1 A 2 and B = constants.The application of eqn. (2) to the summers Sl and S2 shows that these satisfy equations (D) and (E). Application of eqn. (4) to the summer-integrators Zl 12 and 13 shows that equations (A) (B) and (C) are satisfied e.g. applying R.A. Brooker Compiitcr J. 1958 1 15. PROCEEDINGS equation (2) to 11 we have Po = -5 MOP -2AOPOQdt + Po which is the required integral of equation (A). It should be emphasised that Fig. 3 is only a preliminary scheme the final arrangement and the amplifier gains and potentiometer settings used would depend on the numerical values of the constants and the variables i.e.on the scaling necessary. In the investigation reported by Wood et a2.l various combinations of values of A,, Al A, and B were tried with the object of finding the combination giving best agreement between the equations and experiments performed with different starting values of Poand P,. The value of 8 corresponding to each combination was found by preliminary runs. PARTI1 Computers have proved to be of great value in all stages of chemical process development from the initial laboratory research to the improve- ment of operating plant. In the earliest stages of the development of a new chemical process the required end product is known but it is necessary to investigate suggested alternative methods of producing it.These methods may use the same or different raw materials. The chemist may be able to find sufficient published information about some of the suggested processes to enable them to be rejected out of hand but an experi- mental programme is usually necessary to pro- vide sufficient data for the choice to be made between the more promising processes. At this as at all stages of process development it is necessary to make measurements of chemical composition by methods such as infrared ultra- violet and mass spectrometry emission spectro- graphy and vapour-phase chromatography. The calculation of data from these measurements can take a long time e.g. the calculation of sample composition from mass-spectrometer readings involves the solution of a set of linear simultan- eous equations (one equation for each com-ponent of the sample) and may take 1-2 hours per sample.The solution of such sets of equa-tions is a problem for which digital computers are ideally suited and programmes for the pur-pose have existed for several years. As the mass SEPTEMBER1959 calibration matrix for a particular type of sample is semipermanent and the calibration is usually used for many samples the inverse of the calibra- tion matrix is often formed first and used in the subsequent sample calculations thereby reducing these to straightforward matrix by vector multi- plications which for a twenty-component sample can take less than one minute including the input of data and the output of results.There is no difficulty in extending these programmes to per- form further calculations based on the sample analyses e.g. of specific gravity gross calorific value etc. Mass-spectrometer readings have also been reduced on analogue computers. If the effects of varyingseveral parameters such as temperature pressure time and the initial concentrations of reactants catalysts and sol- vents have to be explored careful design of the laboratory experiments is necessary and inter- pretation of the results may be difficult. The objective is often to find the optimum conditions e.g. for maximum yield of product. G. E.P.Box and his co-workers have developed iterative experimental procedures for this type of problem.An example has been given6 of the application of the method to a reaction of the type A + B + C -+D + E + other products The objects were to maximise the yield of D and minimise that of E. “Starting from a yield of 64% of D and 15% E 92% was obtained for D after two cycles of the procedure the product containing less than of E. At this stage it was suggested that the same reaction might be used for obtaining E and a second investigation was begun in this case maximising E; three cycles of the procedure led to conditions which gave a yield of about 70°/,for E.” In the simpler case of straightforward maximisation of one quantity e.g.,yield of a single product the iterative experi- ments are begun at the best known conditions and an exploratory experimental design is per- formed in this neighbourhood.A polynomial function relating yield to the process parameters is then fitted to the results by the method of least squares. Knowledge of the local geography pro- vided by the polynomial then permits the choice of a direction for simultaneous parameter changes which should lead to higher yield. This direction is followed until the yield commences to decrease when a further exploratory experi- mental design is performed yielding a new poly- nomial the process being repeated until the maximum is reached. Often two or three repeti- tions suffice. The polynomials used are of the lowest order for satisfactory interpretation of the results and at the early stages the first order is usually sufficient while it is rarely necessary to go beyond the second order which is of the form y = b + bxx1 + b2x2 + -.+ b,Xi2 + b,+1~2~ + -+ bj+1~1~3 + b3~1~2 + . -+bkX,X +a - where y is the yield and x1 x2,x3 etc. are t& process parameters. With five parameters there are 21 b’s to be evaluated for a second-order polynomial and the least-squares procedure in- volves the solution of 21 simultaneous linear equations in 21 unknowns. As in mass spectro-metry a digital computer can be used to solve these equations and standard matrix programmes have also proved useful in the interpretation of the resulting p~lynomial.~ This technique has also been applied to find the optimum operating conditions of chemical processes at full-scale plant level.s The chemist may wish to test and develop his ideas on possible mechanisms of reaction.The methods for numerical solution of the differential equations for the reaction described in Part I can be used in cases where the course of the reaction can be followed experimentally. In many cases however measurements of composition during the reaction are impracticable and only the final compositions reached in a set of laboratory ex- periments may be available. One approach to this problem has been described in the literature.‘ Briefly the method is to guess initial trial values for the reaction rate constants and to integrate the differential equations repeatedly once for each experiment. The mean-square error S be-tween predicted and measured final yields is then calculated for the whole set of experiments.An iterative procedure is then followed the reaction rate constants being altered until S is reduced to a minimum. Again a digital computer can be of 0. L. Davies (ed.) “Design and Analysis of Industrial Experiments,” 2nd edn. Oliver and Boyd London, 1956 chap. 11. ’ G.E. P. Box and G. A. Coutie Proc. Inst. Electrical Engrs. 1956 By103 Suppl. 1 100. B. D. Dagnafl and P. Mayers J. Inst. Petroleum 1957 43 115. great assistance. Statistical investigations of significance and confidence limits are of course desirable in this type of work particularly when a choice has to be made between several possible mechanisms. The development of the most promising process to a full-scale chemical plant is usually tortuous.The choice of process even is often difficult and it is often necessary to prepare tenta- tive plant flow sheets for more than one process to consider economic and chemical engineering design aspects control and operating problems and then refer back to the laboratory for further experimental work before the final choice can be made. The calculations made at this stage are usually limited in their scope and it is possible that access to a digital computer with autocode facilities would help the engineer and enable him to consider a wider range of alternatives in greater detail than is possible at present. The design of the chemical plant is in a broad sense a stage in the development of the chosen chemical process and a very important one.The technical decisions made at this stage may make or mar the final process. Lack of time and tech- nical manpower have often in the past resulted in these decisions’ being based on past experience and the semiquantitative assessment of alterna- tives rather than detailed calculation. The potential value of computers at this stage is enormous. In a case reported by Hoerl,lo “Less than two hours of computation (on a digital computer) served as the basis for the elimination of a costly plate column from a new plant design.” Indications have since been given that the saving in capital was approximately $300,000 though this has not been confirmed in the literature.Most plant design calculations seem at present to be more suitable for digital than analogue computation and widespread use is now being made of digital computers by chemical plant designers especially in the U.S.A. A valuable account of the experience of the Esso Research and Engineering Company has been given.ll This company has encouraged its engineers to use a medium-sized digital computer in the planning and design of new refineries and for * Chan-Hui Chou. Ind. Em. Chem. 1958.50. 799. PROCEEDINGS making major improvements and additions to existing refineries. It was reported that savings of up to a month had occurred on several design projects and that there was an increasing tendency to use the computer to get the best design in the available time.It was also found that the saving in manpower resulting from automatic computa- tion more than made up for the manpower needed for programming. This company has also found that use of the autocode facilities of a large digital computer can cut programming and “debugging” time to between 50 and 70% of that needed for the medium-sized computer. Modern chemical plants make increasing use of automatic control and the importance of con- sidering control aspects of the plant at the design stage with the control engineers working in con- junction with the chemical engineers is being realised. Analogue computers are particularly useful to the control engineer in his investiga- tions of the control requirements of a chemical plant.The computer is used to represent the plant and the proposed control systems and voltages representing disturbances are applied at appropriate points of the analogue ;the response of the analogue is then an indication of the way the plant may be expected to behave. The great advantages of the analogue computer for this work are the rapidity with which results can be obtained and the ease with which alternative plant layouts and control systems can be evalu- ated. Analogue computer studies have also been made into control and operating difficulties ex-perienced on existing chemical plants indeed most of the studies reported have been made in connection with operating plants rather than at the design stage of new plants.One of these12 was an investigation into the poor control of a con- tinuous exothermic catalytic hydrogenation pro- cess in a distributed reaction system. The analogue was used to investigate several pos- sibilities for improving the control such as in- creasing the height of the catalyst bed decreasing the size of catalyst particles and increasing the flow of recycled material and it is claimed that changes made to the system as a result of the investigation have since proved satisfactory. Apart from the obvious change in scale the loA. E. Hoerl Chem. Eng Progr. -1956,52 464. l1 E. J. Higgins J. W. Kellett and L. T. Ung Ind. Erg. Chem. 1958 50 712. l2 T. L. Bathe R. L. Franks and E. W. James Insfr. SOC.Amer. J. 1957 4 14. SEPTEMBER 1959 final chemical plant is often very different from the laboratory equipment with which the pre- liminary investigations are made and it is common for a pilot plant to be erected to bridge the gap between the two.The pilot plant follows the proposed design of the full-scale plant but is as small as possible consistent with the require- ment that deductions made from its behaviour should be applicable to the larger plant. Apart from confirming that the proposed plant design is satisfactory a well designed and instrumented pilot plant can serve other important purposes even when the main plant is in operation it is commonplace for example to find that the inevitable differences between the main plant and the laboratory experiments result in the optimal conditions for the plant being very different from those found in the laboratory; furthermore these optimal conditions may change with raw material quality and throughput exhaustion of catalyst fouling of heat-transfer surfaces and so on.The pilot plant can be used to explore the effect of these changes on operation and to find operating conditions that will minimise adverse effects. Extensive instrumentation is often neces- sary and the problem of collecting the readings and extracting the required information from them during a prolonged series of tests can then become serious. The calculations needed for each set of readings are often quite simple such as integration of flow rates and material and heat balances but may have to be repeated hundreds of times.An application of a digital computer to the processing of routine data from a catalyst testing unit has been described;13 raw data from records sheets were fed into the computer which first rejected bad data on statistical grounds and compensated for the occasional missed reading before calculating the results. It was claimed that the 3-4 hours required for a single manual calculation were reduced to 2 minutes. The problem of collecting sets of simultaneous readings from dozens of different instruments has been solved by the introduction of the data logger. Flows pressures temperatures levels compositions etc. measured on the plant are transmitted as pneumatic or electrical signals to the data logger which scans these signals auto- matically at preselected intervals and records the measurements on a typed log sheet.To avoid the labour of converting typed log sheets to punched tape or cards for subsequent computer calcula- tions some loggers are arranged to produce punched tape output as well as a typed log sheet. An automatic pilot plant in which a small digital computer is linked permanently to the output of a logger scanner coupled to the pilot plant has been described.14 The computer corrects and con- verts instrument readings before they are printed out and can continuously integrate data and per- form material balance calculations. The system has also been designed so that the computer can alter controller set-points according to a pre- arranged time schedule or when the plant reaches equilibrium so that large evaluation tests can be programmed for unattended operation with the computer initiating catalyst regeneration when necessary.The instrumentation is elaborate and includes continuous composition analysers for hydrogen oxygen water and carbon dioxide and vapour-phase chromatographs. The application of the Box type of iterative experiment to improve the operation of chemical plants and the value of digital computers in this type of investigation has already been referred to. Another method of investigating the opera- tion of chemical plant that has proved of value in some cases uses actual plant operating records. Disturbances are always present on an operating plant and produce changes in the plant operating conditions so that it should in principle be pos- sible to deduce useful correlations between plant operating conditions and say yield of product from the plant records.Investigations of this type necessitate the analysis of large sets of data and again for this reason rather than because of the complexity of the calculations involved a digital computer is essential. The method is at- tractive and results have been claimed in particular applications.1° The attraction of the method possibly arises because it seems to be a way of getting something for nothing-at any rate without experimental work. Box has drawn attention to serious objections to the method.15 These methods for the improvement of the operation of a plant obviously require no formu- lation of the precise equations governing the operation of the chemical process.An alternative l3 L. S. Stanton and E. B. Reid Ind. Eng. Chem. 1958,50 719. J. K. Walker and C. K. Hines Instruments and Automation 1958 31 1688. l5 G. E. P. Box and G. A. Coutie Proc. Inst. Electrical Engrs. 1956 B 103 Suppl. 1 111. approach is to formulate these equations con- firm them and then use them to find ways of improving the operation and even the design of the process. An early investigation of this typelg was an investigation of the fouling in a non- catalytic homogeneous gas-phase reaction vessel. A digital computer being used the effects of different feed rates and jacket temperatures on the development of fouling were investigated and the accuracy of the mathematical model was checked by comparing computed temperature profiles with those actually measured on the vessel.It was claimed that a predicted 20-25% increase of production between shut-downs was obtained when plant conditions were changed approximately to the optimal ones selected from the computed results. The computer took 42 hours for the calculations compared with esti- mates of 29 man-years for hand calculations or several years for actual semi-technical plant duplication of the runs. The difficulties en-countered in formulating confirming and using a mathematical model of a reaction vessel can be formidable but as this example shows the rewards may be considerable.Without computers this approach would rarely be possible for generally only by making gross simplifying assumptions can analytical solutions be found for the equations of a reaction system and the labour of manual numerical solutions would be prohibitive. A chemical reaction system is a complex of interacting chemical and physical processes such as consecutive chemical reactions complex circulation and diffusion effects heat and mass transfers between phases etc. and the process of developing a satisfactory and tractable mathematical model can be diffi- cult. Laboratory and plant experiments to pro- vide reasonably accurate information to base the model on comparisons of the results of com- puter studies of its behaviour with that of the plant and several major revisions of the model and many minor ones are inevitable except in relatively simple cases Several successful ana- logue- and digital-computer studies of batch and perfectly-mixed continuous reaction systems have been made.Distributed reaction systems have also been studied but the difficulties en- countered in obtaining even numerical solutions PROCEEDINGS for the resulting equations are formidable and at present can only be overcome when simplifying assumptions such as perfect transverse mixing are justified. The effect of such an assumption is to eliminate two of the three spatial-independent variables so that the resulting partial differential equations have only two independent variables distance down the vessel and time.Unfortunately an analogue computer has only one independent variable time so that it Is necessary to rewrite the equations as difference-differential equations by replacing the continuous variation of one in- dependent variable usually distance by a finite- difference representation with the result that conditions are calculated only at a number of chosen positions in the system; the form of the resulting equations is suitable for analogue solu- tion but the amount of equipment needed may be prohibitive if the original equations are com- plex and if it is necessary to use small differences. On digital computers the method of reduction to difference-differential equations and subsequent solution by a numerical-integration procedure has been used successfully though difficulties are encountered and research into quicker integra- tion procedures than those now available and to develop possible alternative methods is very necessary.At present the time taken to obtain a single solution even on a large computer is often considerable and may make a thorough explora- tion of the behaviour of a model extremely ex- pensive. The next generation of digital computers now being developed to deal with correspond- ingly complex calculations in the nuclear field will be 10-100 times as fast as the present ones and may transform the situati0n.l' For many purposes it is sufficient to be able to calculate the steady-state behaviour of a dis-tributed reaction system and the necessary differential equations can easily be derived from the complete equations of the system once these have been formulated or can be formulated in- dependently.The effect of a simplifying assump- tion such as perfect transverse mixing is now to leave only one independent variable distance down the system and in this case the resulting equations are ordinary differential ones. With time representing distance these equations can be integrated on an analogue computer with R. E. Gee W. H. Linton jun. R. E. Maier and J. W. Raines Chem. Eng. Progr. 1954,50,497. l7 J. Howlett Proc. Inst. Electrical Engrs. 1958 B 105 331. SEPTEMBER1959 much less equipment than is needed for the cor- responding difference-differential equations though if the boundary conditions are specified at more than one position in the system a trial- and-error approach is necessary.la This can be- come difficult if several independent physical quantities are being computed.Changes of para- meters along the length of the reaction vessel complicate the analogue while the extreme case of actual discontinuities such as occur when catalyst beds are separated by cooling coils make it necessary to interrupt the integration at these points and alter the analogue before con- tinuing. More thorough exploration of these difficulties and methods of overcoming them is needed. Digital computers can also be used to calculate steady-state profiles one approach is to use a numerical integration procedure [which was the method used by Gee Linton Maier and Rainesls (see also ref.19)] though difficulties similar to those met in the analogue solution can be encountered. Another is to rewrite the equa- tions as difference equations in distance which with the addition of the boundary conditions gives a set of simultaneous non-linear equations in a large number of variables. Unfortunately methods for the solution of such sets of equations are very much the province of specialists in numerical analysis and there seems to be very little experience of this type of problem. The possibility of using computers to control chemical processes is a fascinating prospect and several pioneer projects have been reported.In one of these the time required by a batch hydrogenation process was reduced by nearly 25 %.20 At short intervals throughout the opera- 261 tion a small high-speed analogue computer was supplied with measurements from the process and calculated the changes in operating condi- tions required to reduce the batch time to a minimum. Even a comparatively modest project in this field is likely to require considerable re- search effort ;the rewards however may be con-siderable. Extreme reliability will be demanded of any computer that is to be in control of a chemical plant and this must of course be achieved without the cost's being raised to an uneconomic level. However at present the principal technical obstacle seems to be that of deciding how in any particular application the computer should interpret and use the data that can be fed to it from the plant.21 It is possible that for some applications it may be necessary for the computer to be able to remember past patterns of plant behaviour and learn by ex- perience how best to deal with them.The last decade has seen the development of general-purpose analogue and digital computers of great power and the beginning of the explora- tion of this power by the chemical industry. Part I1 of this article has considered some of the avenues along which this exploration has pro- ceeded and some of the problems encountered. That the next decade will see the solution of many of these problems is clear; what cannot be foretold is which of these avenues will be found to be culs-de-sac which will become major high-ways and what alternative routes will be found.Perhaps however sufficient has been said to make clear the rewards likely to result from the more widespread application of computers in chemical process development. l8 J. A. Beutler and J. B. Roberls Chem. Eng. Progr. 1956 52 69~. J. A. Beutler Chem. Eng. Progr. 1954 50 569. 2o D. P. Eckman and I. Lefkowitz Control Engineering 1957 4 197. 21 E. W. James and A. S. Boksenbom Control Engineering 1957 4 148. CHEMICAL SOCIETY MEETING At a Scientific Meeting of The Chemical Society held at Burlington House on Thursday June 4th 1959 with the President (Professor H. J. Emelkus) in the Chair the following papers were read and discussed.Kinetics and Orientation of Some Epoxide Ring- opening Reactions. By N. B. CHAPMAN, N. s. ISAACS, and R. E. PARKER. When a monosubstituted ethylene oxide reacts with a nucleophilic reagent ring-openingmay occur in two ways corresponding to attack of the reagent at the CH carbon atom (normal attack) or at the CH carbon atom (abnormal attack) /O\ R-CH-CH,+ HA +R.CH(OH).CH,A+ R-CHA-CH,-OH Normal Abnormal In order to investigate the effect of a substituent group R in the epoxide on the reactivity at each of these positions it is necessary to determine the rate constant for the overall reaction and to carry out a product analysis. Provided that the two reactions have the same kinetic order the following relation will then hold kobs.= k f kA where km. is the observed rate constant k is the rate constant for normal attack and kAthat for ab- normal attack. Furthermore the ratio kN/kAwill be equal to the ratio of the amount of normal to that of abnormal product. The reactions in ethanol solution between styrene oxide 3-phenylpropylene oxide and phenyl glycidyl ether on the one hand and piperidine morpholine diethanolamine and benzylamine on the other have been investigated in this way at a series of tempera- tures using infrared spectroscopy to determine the product ratios. In addition the rates of the reaction of styrene oxide with 2- 3- and 4-methylpiperidine have been measured. The reactions are all of the second-order and have been shown to be free from various possible complications.The results indicate that all the reactions give entirely or almost entirely normal products with the exception of that between styrene oxide and benzylamine. The reasons for this will be discussed and the results will be analysed in terms of the Taft linear free-energy relationship. Work is in progress on the reactions of benzyl- amine with 4bromostyrene oxide and 4-methyl- styrene oxide. These two oxides together with styrene oxide form a series in which steric effects are likely to be constant and it is hoped that these studies will provide information concerning the polar effects of substituents on the rate of attack at the normal and at the abnormal position of an unsymmetrical epoxide.Professor N. B. Chapman made the following addition at the meeting. He said that the “normal” products of ring-opening of epoxides by amines have been unambiguously synthesised by the interaction of phenacyl bromides with amines followed by re- duction of the amino-ketone with isopropyl alcohol and aluminium isopropoxide and thus the orienta- tion of ring opening as to major product has been established in numerous cases. It has also been found that branched chain amines do not readily react with styrene oxide so the above synthesis is an important route to the corresponding products. The solvolysis of NN-dialkyl- C-phenylethyleneiminium ions form- ally resembles epoxide ring opening but probably pursues a mechanism corresponding to acid-catalysed hydrolysis of epoxides.Aromatic Reactivity. Part HI. Cleavage of Substituted Phenyltrimethylsilanes by Sulphuric Acid in Acetic Acid-Water. By F. B. DEANS and C. EABORN. We have measured the rates of cleavage (“proto- desilylation”) of substituted phenyltrimethylsilanes by aqueous sulphuric acid in acetic acid at 50” PROCEEDINGS X*CIHI*SiMe + H,O -+ XC,H,.H + Me,Si.OH Hf The relative rates of cleavage are (X = ) H 1; p-MeO 1010; p-M%Si*CH, 202; p-Me 18.0; o-Ph 5.85; p-Ph 2.83; p-F 0.95; m-MeO 0-38; nz-Ph 0.33; p-C1 0.190; p-I 0-101; p-Br 0.104; rn-CI 1.2 x p-CO,H 1.48 x lo-,; p-NMe,+ 3-8 X 1O4;p-NO, 1.2 x lov4. These figures taken with data previously obtained for cleavage by aqueous-methanolic perchloric acid provide an unusually complete picture of effects of substituents on an electrophilic aromatic substitu- tion.Apparent activation energies and probability factors have been determined. The activation energy appears to be unaffected by change in the substitu- ent but this is an accidental result of the variation of the sulphuric acid content of the medium which is necessary for compounds of widely differing re- activities. If allowance is made for this the trend of activation energies is consistent with the trend of reactivities. Dr. Gerrurd suggested that considering the overall equation it would be helpful if Dr. Eabom would give a more detailed scheme showing the career of the reactive centres (or atoms) during the substitu- tion the proton formally shown free will of course be attached to something before during and after the substitution.The possible formation of the tri- methylsilicyl cation should be considered in con- junction with the difficulty of establishing the exist- ence of the triphenylsilicyl cation. Having regard to the large effect of the p-Me,SiCH group comparison of the influence of the neopentyl group in this position would be most interesting. Dr. Eaborn replied that the simplest mechanism consistent with the facts is one in which a relatively fast proton transfer from oxonium ion to the carbon of C-SiMe bond is followed by ratedetermining nucleophilic attack of a solvent molecule on silicon in the intermediate [ArHSiMe,]f.The transition- state of this slow step will however not be far re- moved along the reaction co-ordinate from this inter- mediate so that overall the reaction involves pre- dominant electrophilic attack at carbon. Separation of a Me,%+ ion in the slow step is also consistent with the data but we do not postulate this in absence of evidence that such an ion exists. (There is clear evidence that the Ph,Si+ is not formed under conditions in which the Ph,Ci+ on forms readily.) More complicated mechanisms can be written particularly those involving entry of the proton and exit of the Me,Si group via .rr-complexes with the aromatic ring. The activating effect of a p-neopentyl group is SEPTEMBER 1959 unlikely to be very different from that of a p-Me or p-Bu group and thus it is almost certainly much smaller than that of a Me3Si*CH group.The Heats and Entropies of Ionisation of Some Aromatic and N-Heteroaromatic Amines. By J. J. ELLIOTT and S. F. MASON. The ionisation constants of a series of unsubsti- tuted polycyclic aromatic amines have been measured at 0" and 20" in 50% ethanol-water solution and those of a series of N-heteroaromatic amines have been obtained at 5" and 35" in aqueous solution entropies and enthalpies of ionisation being derived from the results. This dissociation of the conjugated acids of the aromatic amines is found to be governed equally by the entropy (TAS) and the enthaIpy factor (AH) both TAS and AH varying by some 2 kcal./mole in the series covered.Thus the linear relation observedl between the free energy of ionisation of the aromatic amines and the n-electron energy change due to the protonation of the amines (an internal energy term) is largely fortuitous. The dissociation entropies fall into two main groups TAS having the value of Hush J. 1953 684. Longuet-Higgins,J. Chem. Phys. 1950 18 275. f0-4kcal./mole for the unhindered aromatic arnines and 0-9-1.7 kcal./mole for the peri-amines a difference which may be due to the loss of the en-tropy of rotation of the NH3+ group in the un-hindered amines on dissociation to the free base. In contrast the ionisation constants of the N-heteroaromatic amines studied vary by a factor of lo5owing to differences in the enthalpy of ionisation in the series but by a factor of only 50 because of variations in the entropy of ionisation.Thus it is more probable in the N-heteroaromatic2 than in the aromatic1 series of amines that the v-electron energy change due to the dissociation of the conjugate acid of the amines primarily determines the relative pK values. However the relative values of the H-N-H bond angles and of the heats and entropies of ionisa- tion of the N-heteroaromatic amines suggest that 0-bond-energy changes and solvation-energy changes on cation formation inductive and steric effects and intramolecular hydrogen-bonding all have a sub- sidiary influence upon the relative values of the ionisation constants in the series of N-heteroaro- matic amines.COMMUNICATIONS Stereochemistry and Synthesis of Phytol Geraniol and Nerol By J. W.K. BURRELL, L. M. JACKMAN,and B. C. L. WEEDON OF CHEMISTRY OF SCIENCE (DEPARTMENT IMPERIALCOLLEGE AND TECHNOLOGY SOUTHKENSINGTON S.W.7) LONDON PREVIOUS publications1 allow no conclusion concern- ing the absolute configuration of phytol (VII). Professor C. Djerassi has kindly informed us of optical rotatory dispersion studies which suggest a D centre at position 7. It has now been shown that phytol is 3,0-7,D-11,15-tetramethylhexadec-trans-2-en-1-ol,* and this structure has been confirmed synthetically. D'-Citronellolf. (I) was converted by reduction and subsequent nitrile synthesis into D'-4,8-dimethyl-nonanoic acid (II) which was anodically cross- coupled with (&)- (-)-D-and (+)-&(methyl hydrogen p-methylglutarate)2 (III).The resulting 3,7,1l-trimethyldodecanoicacids (IV)were similarly coupled with laevulic acid to give DL-6,D'-10,14- L-6,0'-10,14-and 0-6,D'- 10,14-trimethyIpenta- decan-Zone (V) respectively. The optical rotations TABLE 1. Optical rotations of the CI8ketones (I = 1 homogeneous liquid) Configuration UzI; DL-6 D'-10 -0.20"* L-6,D'-10 -0.98 0-6,D'-10 +0.66 Natural +0.57 * For a sample prepared by an alternative route Karrer et al. (Helv. Chim.Acta 1943 26 1741) report (YD -0.24". * With trisubstituted double bonds the terms cis and trans are used to designatz the relative positions of the two largest substituents. The configurational symbols D and L are those defined by Linstead Lunt and Weedon.2 t The citronellol a + 3-95" wasprepared by PonndoriT reduction of (+)citronella1 from Java citronella oil and had an optical purity of ca.80%. Citronellol has been correlated with ~-glyceraldehyde;~ the symbol D' is used rather than D to indicate the lack of optical purity at the asymmetric centre in the citronellol and the corresponding centres in the compounds derived from it. Cf.Simonsen and Barton "The Terpenes," Cambridge Univ. Press 1952 VoI. 111; Lukd and ZobAEovB Chem. Listy 1957,51,330; Weichet HodravA and Kvita ibid.,p. 568; Sarycheva Vorob'eva Kuznetsova and Preobrazhenskii Zhur. obshchei Khim. 1958,28 647; Nazarov Gusev and Gunar ibid. p. 1444. Linstead Lunt and Weedon J.1951 1 130. PROCEEDINGS of these C, ketones (Table 1) show that a D-con- methyl cis-and methyl trans-geranate which were figuration at C(6) contributes f082” f0.04”,and reduced to nerol and geraniol respectively. This con-one at C(lo)-0-25” f0.04”,to the rotation. firms the structural assignments previously made to Ozonolysis3 of phytol (a + 0.03”) gave a C, these C, alcohols mainly on the basis of their (+)-ketone which by treatment with methylenetri- relative rates of acid cyclisation.s phenylph~sphorane~ and reduction of the resulting The authors thank Messrs. Boake Roberts and olefin was converted into meso-2,6,10,14-tetra-Co. Ltd. for a research bursary (to J.W.K.B.) and methylpentadecane (a 0”). The DL-6,D’-10,14-for a generous gift of (+)-citronellol Messrs.ketone was similarly converted into the 2,DL-6,D’- Stafford-Allen and Sons Ltd. for liberal supplies of‘ 10,14- (~2,L’-6,DL-1OY14-) hydrocarbon (a -phytol and Imperial Chemical Industries Limited 0.18’). The “natural” ketone must therefore have Paints Division for gas-liquid chromatograms on a 0-6,D-10 configuration. Pye instrument. (Received July loth 1959.) TABLE2. /%Methyl absorption of $-unsaturated esters* Me ester cis trans crcis-~trcans Phytenoate 8-14 7.89 025 Geranate 8-27 8.02 0.25 Me0,CCMe :CH-C0,Me5 7-96 7-72 0-24 MeO,CCH,CMe CHC02Me5 8-04 7.80 0.24 * Spectra were determined on carbon tetrachloride solutions (ca. 10%) at 40 Mc. and calibrated against tetramethylsilane as internal standard. Band positions are given a walues defined by Tiers (J.Phys. Chem. 1958, 62 1151). Reaction of the “natural” C, ketone with methoxyacetylene and treatment of the resulting alcohol with acid gave a mixture of methyl phyteno- ates (VI) which was separated by chromatography on alumina into (+)-cis- and (-)-trans-isomers. The unambiguous assignment of geometrical configura- tion was made from a study of the @-methyl absorp- tion~~ in the nuclear magnetic resonance spectra (Table 2). Reduction of the esters with lithium aluminium hydride gave the corresponding cis- and trans-alcohols (VII). These like natural phytol gave only one band on gas-liquid chromatography in argon on a polyethylene succinate column. In mixed chromatograms natural phytol separated from the cis- but not from the trans-alcohol.Similarly 6-methylhept-5-en-2-one yielded both Fischer Annalen 1928 464,69. * Wittig and Schollkopf Chern. Ber. 1954 87 1318. Jackman and Wiley Proc. Chem. SOC.,1958 196. Cf. Simonsen and Owen “The Terpenes,” Cambridge Unh. Press 1947 Vol. I. Cf. Klyne “Progress in Stereochemistry,” Butterworths London 1954 Vol. I. OpticaI Rotatory Dispersion Studies. Part XXIX.l Absolute Codgumtion of Phytol CRABBE E. J. EISENBRAUN, By PIERRE CARLDJERASSI and SHWENLIU (DEPARTMENT 2 MICHIGAN,) OF CHEMISTRY WAYNE STATE UNIVERSITY DETROIT UNTIL now no information has been available on the element of various natural products such as chloro-absolute configuration of the diterpene alcohol ~hyll.~ Indeed the very low rotational values reported phytol (II)which represents an important structural for specimens of phytol from different natural Part XXVIII Aeli and Djerassi Helv.chim. Acta in the press. * Present addresses Department of Chemistry StanfordUniversity Stanford California. Simonsen and Barton “The Terpenes,” Cambridge Univ. Press 1952 Vol. 111 pp. 345-349. SEPTEMBER 1959 sources have even raised some doubt as to the optical activity of this s~bstance.~,~ Both problems have now been solved by taking advantage of information accumulated during extensive investigations on the optical rotatory dispersion behaviour of ketones and aldehydes. Since optically active carbonyl compounds show Cotton effect curves in the ultraviolet spectrum with greatly increased rotational values it was felt that a rotatory dispersion study of appropriate carbonyl- containing degradation products of phytol would provide an answer to its optical activity.Further it has been shown6 that while dextrorotatory (29-2-methylbutanal' (I;n = 0) which is configurationally related to L-glyceraldehyde exhibits apositive Cotton effect its next higher homologue (I;n = 1) shows a negative one. This characteristic behaviour can be used for assignments of absolute configuration since the nature of the hydrocarbon substituent attached to the asymmetric centre is usually of no great importance in this respect.* Dehydration of a benzene solution of phytol (chlorophyll-derived samples of phytol from L. Light and Co.Ltd. Sandoz Ltd. and Stafford Allen and Sons Ltd. were employed with substantially identical results) with phthalic anhydride and toluenep-sulphonic acid afforded phytadiene-C (III),' b.p. 135-140"/1~0 II~II~.,Amax. (in 95% EtOH) 235 my_L (log E 4.40) Amax. (cap.) 10.38 p nz 1.4705 a -5.7" (neat). Ozonolysis of the diene (LII) at -70" in methylene chloride and decomposi- tion of the ozonide with ferrous sulphate led to the C, aldehyde (IV) which exhibited a positive Cotton effect curve (peak in octane solution at [a]32o + 60"). Conversion into the en01 acetate (V) b.p. 110-1 15"/ 3 mm. Am. (cap.) 559,591 and 8-13p nz 1-4454 [a]2,9 -1-17' (c 1-9 in chloroform); a second ozonolysis led to the C, aldehyde with a nega- tive Cotton effect (trough in octane solution [a1335 -68").The aldehydes (IV) and (VI) therefore show the opposite rotatory dispersion behaviour from that of the reference aldehydesg (I; n = 0 1) of the (S)-series. Consequently C 7) of phytol (II)must have the (R)-configuration and this conclusion has now been confirmed by synthetical studies by Weedon and col- laborators1° who also settled the configuration of the second asymmetric centre. 7% cps 7 [CH,I,CHO H (1) (a) &cHiOH MR (a)A (IY)FHO (a>,CH*OAc / (rn) CHO We are grateful to Dr. B. C. L. Weedon (Imperial College London S.W.7) for exchange of information and for supplies of phytol. Samples have also been provided through the courtesy of Professor G. W. Kenner (University of Liverpool) and Dr.E. Wiedemann (Sandoz Ltd. Basle). Financial assist-ance was provided by the National Cancer Institute of the U.S. PublicHealth Service. One of us (P.C.) is indebted to the U.S.Educational Foundation in Belgium for a Fulbright travel grant. (Received Jury loth 1959.) Karrer et al. Helv. Chim. Acta 1943 26 1741;1944 27 313 1006. For reviews see Djerassi Bull. SOC. chim. France 741 (1957); Record Chem. Prog. 1959,20 101. Levene and Rothen J. Chem. Phys. .1936,4,48; Djerassi and Geller J. Amer. Chem. SOC.,1959 81 2789. This isomer is designated L on the LFstead-Lunt-Weedon convention (J.,1951 1130). * See for instance Eisenbraun Osieclu and Djerassi J. Amer. Chem. SOC.,1958 80 1261. Rowland ibid.,1957 79 5007. lo Burrell Jackman and Weedon Proc.Chern. Soc. accompanying paper. Fluorine Cyanide By E. E. AYNSLEY, R.E. DODD and R.LITTLE ("HE CHEMICAL KING'SCOLLEGE UPON TYNE, LABORATORIES NEWCASTLE 1) PREPARATION of fluorine cyanide FCN was reported infrared absorption spectrum of the fraction of the by Cosslettl in 1931 but a repetition of Cosslett's products subliming in the range -120" to -130" at method2 evidently did not yield fluorine cyanide. So 5 mm. shows it to be a mixture containing among far we have also failed to obtain this compound but other things carbonyl fluoride and sometimes we have good indications of its presence among the carbon dioxide and dinitrogen monoxide. There are products of fluorination of cyanogen. The complex two bands which appear to belong to the same Cosslett 2.mrg.Chem. 1931 201 75. a Calloman Thompson Anderson and Bak J. 1953 3709. PROCEEDINGS species presumably fluorine cyanide. That at 1077 cm.-l has been well resolved (Grubb-Parsons spectrometer GS2 operated with a single beam) and shows no Q-branch the rotational line separation corresponds to a moment of inertia of 47.5 amu A2. This implies a linear molecule which must be tri- atomic (for no tetra-atomic molecule not containing hydrogen could have so small a moment of inertia). It is not carbon dioxide or dinitrogen monoxide. Further the moment of inertia corresponds to reasonable pairs of C-F and C-N bond lengths e.g. 1.26 and 1.16 8,respectively if account is taken of the shortening of the C-F distance which may be expected by analogy with the behaviour of the C-Cl bond in chlorine cyanide.We have not been able to resolve the somewhat stronger band at 2290 cm.? but its contour shows no sign of a @branch and the separation of the P-and the R-maxima is in agree-Orville Thomas J. Chem. Phys. 1952 20 920. ment with the above moment of inertia. We believe therefore that these are the parallel bands of linear fluorine cyanide and they agree with Orville Thomas's predictions3 (1052 and 2294 an.-:) obtained by extra- polation from the other halogen cyanides. We are now trying to isolate the substance from other products in order to determine its physical properties to make a confirmatory analysis and to locate the perpendicular band v2.The substance responsible for the above two bands is stable we have kept it for 2-3 weeks although mainly at -I SO",without evident loss. We are grateful to Dr. V. E. Cosslett and Dr. L. H. Long for valuable and helpful correspondence and we thank the Department of Scientific and Industrial Research for a maintenance grant (to R.L.). (Received June 1 st 1959.) The Magnitudes and Relative Signs of Hydrogen Spin-Spin Coupling Constants in Hydrocarbon Groupings By C. N. BANWELL N. SEEPPARD, A. D. COHEN and J. J. TURNER CHEMICAL LENSFIELD (UNIVERSITY LABORATORY ROAD,CAMBFUDGE) CONSIDERABLE interest in the coupling constants .IHH, between hydrogen nuclei obtained from nuclear magnetic resonance spectra of hydrocarbon group- ings has been revived as a result of recent theoretical paper~.l.~*~ We summarise in the Table some pertinent results from recent experimental work in this laboratory in which particular attention has been paid to the determination of the relative signs of coupling constants.The spectra were obtained by using a Varian Associates V-4300 B spectrometer operating at 40 Mc. Coupling constants between chemically equi- valent hydrogen atoms of the substituted ethanes were obtained by analysis of the hydrogen spectra of 13CH groups.* Solvents of different dielectric con- stant caused variations of the proportions of rota- tional isomers and the spectra then enabled separate values for the trans- and gauche-constants to be determined. The spectra of the vinyl derivatives and the trisubstituted benzene were analysed by using ABC theory5 (ABCX for vinyl fluoride) and those for the dichloropropenes by ABCztheory.6 Karplus and Anderson J.Chem. Phys. 1959 30 6. Karplus ibid. p. 11. McConnell. ibid.. D. 126. The values given for the trans-and gauche- coupling constants seem to be the first to have been published for simple substituted ethanes; they com- pare fairly well with values previously quoted for ring compo~nds,4.~ and the relative magnitudes are in agreement with Karplus's calculated values of f9.2 and +1.7 c/sec. respectively.2 The values for the ethylenic compounds agree well with those ob- tained by Alexandefl for but-l-ene and 3,3-dimethyl- but-l-ene except that with the help of spectra ob-tained at 16.2 Mc/sec.we have beenable to solve the important uncertainty concerning which of the coupling constants JZ3or J, is of opposite sign to the others. Our result is analogous to that obtained recently by Elvidge and Jackman for a conjugated ~ystem.~ The Jzzconstant for the vinyl compounds is negative* in all cases that we have measured except for vinyl cyanide and styrene. The relative magni- tudes of the cis-and trans-coupling constants again agree well with Karplus's predictions of +6.1 and +1 1 *9 CIS-. respectively.2 From the trichloroben- zene spectrum we have been able to determine the Cohen Sheppard,-and Turner Proc. Chem. SOC.,1958 118. Bernstein Pople and Schneider Cud.J. Chem. 1957,35 65. Cohen and Sheppard unpublished work.'Lemieux Kulling Bernstein and khneider J. Amer. Chem. Soc. 1957 79 1005. Alexander J. Chem. Phys. 1958,243,358. Elvidge and Jackman Proc. Chem. SOC.,1959 89. * We understand that the same conclusion has been reached for vinyl chloride by Dr. W. A Associates. Anderson of Varian SEPTEMBER 1959 267 JHHcoupling constants of some hydrocarbon groupings in clsec. Substituted ethanes ClH2&6H2Cl Cl2HC.CH2Cl C12HCCHC12 BrH2C.CH2Br J12(trans) = 18-lo* J12(gauche) = +1 to +3-5 12 Substituted ethylenes vinyl compounds XHC=CH2 (X= Br C1 F CN OMe OCH:CH, Ph). J12(trans) = l2-l8*; J12(cis) = +5 to +11; Jz2 = -3 to +2 12s cis-and trans-1 :3-dichloroprop-l-ene ClHC=CHCH,Cl cis compound J, (cis) = 7.0*;J2, = +7.4; J, = -1-2 trans compound:J12(trans) = 13.1*; Jzs = +7.3; Jla = -1.2 Substituted benzene 1,2,4trichlorobenzene J (ortho) = 8-7*;J (rnetu) = +2.4; J (para) = +0*2 * These coupling constants are arbitrarily assumed to be positive; other signs quoted for the same molecule are experimentally determined relatively to this.sign of the para-coupling constant relative to the are all small in fair agreement with an earlier predic- other two .lo-ll tion that all .IHH constants are likely to be positive.12 Only the relative signs of the coupling constants It has however been more recently envisaged that for a given molecule are obtainable from spectra and in certain cases coupling constants between hydrogen so absolute signs must be decided by comparison nuclei separated by an even number of bonds may with theoretical calculations.There seems however be negative;11*13 this is the situation for both types to be a close correspondence between the observed of experiment ally determined negative constants . and the calculated relative magnitudes of the trans Detailed analyses of the spectra will be published gauche and cis constants and it therefore seems likely later. these are all positive in sign as calculated by Karplus2 Our work has been greatly helped by the use of the If this is so the signs given in the Table are correct; digital computer EDSAC 2 and financial aid from it is then seen that the negative coupling constants the Institute of Petroleum. (Received June 8th) 1959.) lo Pople Schneider and Bernstein Canad.J. Chem. 1957 35 1060. l1 Bak Shoolery and Williams J. Mol. Spectroscopy 1958,2 525. l2 McConnell J. Chem. Phys. 1956,24,480; Ann. Rev.Phys. Chem. 1957,8 105. l3 McConnell J. Chem. Phys. 1955 23 2454. Reactions of Tricarbonyltropyliumchromium Perchlorate with Anions :a Novel Rearrangement By J. D. MUNRO and P. L. PAUSON (THEUNIVERSITY 10) SHEFFIELD and HONNEN~ DAUBEN have described the prepara- course yielding tricarbonylbenzenechromiumfrom tion of tricarbonyltropylium-molybdenumfluoro-the unsubstituted and tricarbonyltoluenechromium borate. We have used their method to obtain the from the methyl-substituted tricarbonyltropylium- analogous tricarbonyltropylium- and tricarbonyl-chromium salt. These facts suggest that the benzene methyltropylium-chromium perchlorates and in-ring in the product is derived from the seven-vestigated their reactions with HCO,- CN- OMe- membered ring by ring contraction C,H,- C,H,- and C,H,CMe,-.These reactions will be described in detail elsewhere but the most [C,H,R.Cr(CO),]+ + C,H,-+ general course can be indicated by the expression C*H,R-Cr(CO) + (CJW?) We are attempting to verify this conclusion and to establish the nature of the C,H fragment which is eliminated by use of tritium-labelled starting materials. C;r Cr oc’l‘co oc/ I‘co CO CO We are indebted to the Ethyl Corporation for The reaction with the unsubstituted cyclopentadienyl ion C,H,- however takes a radically different (Received June 15th) 1959.) Dauben and Honnen J. Amer. Chem. Soc.1958,80,5570. PROCEEDINGS Structure of the Cyclododecane Molecule Amendment By J. D. DUNITZ and H. M. M. SHEARER (ORGANIC LABORATORY INSTITUTE ZURICH) CHEMISTRY SWISSFEDERAL OF TECHNOLOGY WEare compelled to amend our earlier interpretation of the results of our X-ray analysis1 in the light of new calculations on the effect of overlapping on the accuracy of the positional parameters. The crystal which is disordered contains two kinds of molecule randomly distributed with respect to statistical mirror 7.56 planes at y = 0 and y = 3. The atomic positions are I thus grouped in pairs and the previously described molecular model corresponds to one of a number of possible ways of selecting a “half-atom” from each 1.5$ 761’ I pair.The alternative models derived by interchang- ing the signs of the Y-co-ordinates of atoms 2 and 5 and of 3 and 4 (see the Figure accompanying our previous communication) were rejected on the 70 i ground that they gave rise to GC distances of up to 7‘ 1.63 A which were regarded as unreasonable in view of the estimated standard deviation of about 0.014 A in atomic position calculated by the usual methods?*s Our new calculations which will be described in detail elsewhere show that while this standard devia- 0 tion is acceptable for the atoms 1 3,4 and 6,it must be drastically increased for atoms 2 and 5 to take Revised representation of cyclododecane molecule. The account of overlapping of these atoms which are A molecular axis points upwards from the paper and separated only by 0.30 A in the previously described the X-co-ordinates (in A) are given in parentheses.disordered structure. The calculated structure factors Bond lengths and angles are shown in the upper part are particularly insensitive to the magnitude of this torsional angles (about the bond indicated) in the lower separation and are even compatible with complete part. This model difers from the earlier one in the superposition of these two “half-atoms”-a change alteration of the X-co-ordinates of atoms 2 and 5 and of 0.15 A in the X-co-ordinates of both. This large in the change in sign of the Y-co-ordinates of atoms uncertainty means that the alternative structures 3 and4. cannot be rejected. If atoms 3 and 4 are left unaltered the conforma- electron A-9 at calculated positions for atoms tion of the molecule is not changed although the attached at positions 2 and 5 is 0.10,0.07 0.12 0.14 bond distances valency angles and torsion angles for the 422 model and -0.07 0.07,0.12 -0.05 for are somewhat different from those previously given.the 42m model. The expected height of a “half The point symmetry of the molecule is still nearly hydrogen” atom with B = 14 A2(assumed) is 0-14 ;12m(D2J. If however the signs of the Y-co-electron A-3 but the estimated standard deviation .in ordinates of these atoms are interchanged we have the observed electron density is 0.07 electron A-3 so another conformer which approaches the point sym- that although this test would favour the 422 model metry 422(0,) the principal difference being that it cannot be considered conclusive.the torsion angles about the bonds 1-2,2-3,4-5 and A rough calculation indicates that the Pitzer 5-6 are now close to 60” (syn-skew conformation) strain4g5 amounts to about 23 and 5 kcal. mole-l for instead of about 105” (nearly anti-skew conforma-the 42m and the 422 model respectively relative to tion) in the 42m model. The change of 0.15 A in the that in cyclohexane and suggests that the latter (con- X-co-ordinates of atoms 2 and 5 leads to a model taining 8 syn-skew conformations) would be the shown in the annexed Figure with reasonable bond more likely on energetic grounds. The observed distances and valency angles. thermochemical value of the strain energy of cyclo- The Werent dispositions of hydrogen atoms in the dodecane relative to cyclohexane is 3-4 &-1.2 kcal.two conformers might suggest that a distinction be-mole-l (ref. 6). tween them might be possible by direct placing of the Thus although our general description of the hydrogen atoms in the final three-dimensional differ- molecule as being “built from four nearly planar ence synthesis. The value of the electron density (in units each of four atoms with an atom shared in Dunitz and Shearer Prm. Chem. SOC.,1958 348. * Pitzer Science 1945 101 672. Cochran Acta Cryst. 1951 4 408. Kuhn J. Chem. Phys. 1947,15 843. Cruickshank ibid. 1949 2 65. * Van Kamp Thesis Amsterdam 1957. SEPTEMBER 1959 common between successive units" stands unaltered the conformations about the C-C bonds formed by the common atoms are probably syn-skew rather than anti-skew.This work was carried out with the financial sup- port of the Schweiz. Nationalfonds mr Forderung der wissentschaftlichen Forschung. (Received July 9th 1959.) Epicalciferol By I. T. HARRISON, R. A. A. HURST and B. LYTHGOE (DEPARTMENT CHEMISTRY LEEDS 2) OF ORGANIC THEUNIVERSITY INour partial synthesis of vitamin D,Z the mixture ofthe 3 a-and the 3P-epimer obtained was separated by crystallisation of the 3,5-dinitrobenzoates which gave the known ester of calciferol together with a well-crystalline 3,5-dinitrobenzoate m.p. 148"; the latter then regarded (incorrectly) as that of pure epicalciferol gave on hydrolysis esterification with p-nitrobenzoyl chloride and crystallisation epi- calciferyl p-nitrobenzoate m.p.122-123O.l Other morkersY2 who obtained oily epicalciferol from a partial synthesis characterised it as an amorphous 3,5-dinitrobenzoate m.p. 100-1 lo" and suggested that our product of m.p. 148" contained equal parts of the epimeric 3,5-dinitrobenzoates. We confirm that it is in fact a molecular complex less soluble than either of itscomponents.Theirsuggestionthatwe failed to achieve a separation of an epicalciferol derivative is however incorrect. Epicalciferyl p-nitrobenzoate m.p. 122-123" [a] + 7" (in ben-zene) identical with our synthetic material has now been obtained from epilumisteroP by ultraviolet irradiation. Hydrolysis gave epicalciferol m.p. 85-86" [a] -12" (in benzene) Amax.265 mp (E 18,000) now obtained crystalline for the first time. We thank Dr. B. A. Hems of Messrs. Glaxo Laboratories Ltd. for a generous gift of lumisterol. (Received June loth 1959.) Harrison and Lythgoe Proc. 1957,261 ;J. 1958 837. Inhoffen Iqnscher Hirschfeld Stache and Kreutzer Chem. Ber. 1958,91 2309; J. 1959 385. Barnett Hedbron Jones and Verrill J. 1940 1390. Biflavonyls. The Structures of Kayafiavone and Sotetsufiavone By W. BAKER and K. W. ROBINSON W. D. OLLIS, (THEUNIVERSITY, BRISTOL) WEhave recently established biflavonyl structures (I) for ginkgetin (R = Me R' = R"= H) isoginkgetin (R = R = H R" = Me) and sciadopitysin (R = €2" = Me R' = H),l and in the last two cases com- plete degradative evidence was given.In ginkgetin the location of one of the two 0-methyl groups depended upon a study of the effect of base upon its ultraviolet spectrum. Kawano simultaneously re- ported2 his studies on sciadopitysin and although a complete structure was not proposed by him his evidence was compatible with formula (I; R = R" = Me R' = H). Kawano has now obtained structure.3 evidence supporting this sciadopitysin synthetical3 RO~~/O,/ \gMe Rb// I /-\ OR#* \s" HO d (I) Additional evidence concerning the structures of these biflavonyls has now been obtained. Alkaline hydrolysis of sciadopitysin yields various products including p-anisic acid 2,6-dihydroxy-4-methoxy-acetophenone and a ketoflavone C23H1405(OMe)2 (m.p. 266-267"; carbonyl bands vmax.1675 and 1650 cm.-l) which has been previously isolated and described? We now propose for this ketoflavone a new structure (II; R' = H R" = Me) compatible with all the evidence. The same ketoflavone is ob-tained by hydrolysis of isoginkgetin thus confirming R the structure of isoginkgetin as (I;= R' = H R" = Me). A similar hydrolysis of ginkgeth gives 4-hydroxy acetop henone 2,6-dihydroxy4me thoxy- ace tophenone and a ke t oflavone C23H OMe) m.p. 286-287" which must be (LI; R' = R" = H) Me*co& 'Me-CO80Me / R'O 1 \ OR" Me0/ OH \ COMe HO Ho 0 (a) (W because the position of the methoxyl group has al-ready been established1 by the formation of 4-methoxyisophthalic acid by the oxidation of gink-getin. This therefore confirms the above structure for Baker,Finch Ollis and Robinson Proc.Chem. SOC.,1959 91. Kawano Chem. and Id. 1959 368. Kawano ibid. in the press. Kariyone and Kawano J.Pharm. Soc. Japan 1956,76,451,453. Kariyone and Sawada J. Pharm. SOC.Jupun 1958,78,1010,1013 1016. ginkgetin which was proposed earlier on spectro- scopic evidence. Kariyone and Sawada5 have shown that naturally occurring kayaflavone and sotetsuflavone are bi-flavonyls closely related to ginkgetin and sciadopity- sin but the structures formerly suggested6 must be modified in the light of the new and established structure of ginkgetin and sciadopitysin. Kaya- flavone C30H,204(OH)3(OMe)3 is an isomer of sciadopitysin and both yield the same trimethyl ether.Alkaline hydrolysis of kayaflavone5 yields several products including a ketoflavone C2,HI3O4(OMe) which must be (IT;R’ = R” = Me) so that kayaflavone itself must be (I; R = H R’ = R” = Me). Another product of hydrolysis C1,H,,06 must be the diketone (III). We have con- firmed this structure for kayaflavone using material isolated from the leaves of Torreya nucifera by treat- ing it with alkaline hydrogen peroxide to obtain p-anisic and LGmethoxyisophthalic acid; since sciadopitysin is (I; R = R”= Me R’ = H)it follows that its isomer kayaflavone must be (I; R = H R’ = R” = Me). This structure is supported by the infrared spectrum (Nujol) which shows one carbonyl band (vmax. 1662 cm.-l) and by the effect of base on PROCEEDINGS its ultraviolet spectrum Amax.(E) in EtOH 271-5 (44,lOO) 329 mp (41 ,OOO); in ~/50-NaoEt-EtOH 281 (58,000) 378 mp (20,800); in ~/500-NaOEt- EtOH 276 (51,600) 353 mp (24,200); in ~/5000-NaOEt-EtOH 274 (48,000) 308 (32,000) 350 mp (27,300). The infrared spectrum shows that kaya- flavone is a biflavonyl containing two hydroxyl groups in the 5-and the 5”-position and the ultra- violet spectrum shows that a hydroxyl group from which it is not difficult to remove a proton is located para to a carbonyl group. This places the third hydroxyl group of kayaflavone in position 7. Sotetsuflavone C3,H,20,(0H),.0Me gives a pentamethyl ether identical with ginkgetin tetra- methyl ether and gives on alkaline hydrolysis a ketoflavone C2,H1,O6.OMe identical with that shi-larly derived from ginkgetin.This we now know to be (11; R’ = R” = H) and sotetsuflavone must therefore be (I; R = R’= R”= H). We thank Dr. Kawano for a pre-publication copy of one of his papers (ref. 3) Dr. W. B. Whalley for a specimen of 2,6-dihydroxy-4-methoxyacetophen-one and Mr. R. F. Wood of the Forestry Commis- sion for plant material. (Received Jury 2nd 1959. Kariyone and Sawada “CompIete publication in memory of Professor T. Kariyone,” 1956 p. 16; T. Kariyone, Proceedings of the Phytochemical Symposium Kuala Lumpur 1957 p. 160 quoted by €3. Erdtman Fourth Internat. Congr. Biochem. Vienna 1958; “Biochemistry of Wood,” Pergamon Press Vol. 11 p. 1. Aldobiouronic Acids from Catalytically Oxidised Polysaccharides By G. 0.ASPINALL and A.NICOLSON I. M CAIRNCROSS (DEPARTMENT OF CHEMISTRY UNIVERSITY OF EDINBURGH) PRIMARY alcoholic groups in carbohydrates may be selectively oxidised to carboxylic acids by gaseous oxygen in the presence of a platinum cata1yst.l We have now shown that this type of oxidation may also be carried out onpolysaccharides. Both glycofurano- siduronic and glycopyranosiduronic acid linkages in the oxidised polysaccharide thus formed resist acid- hydrolysis and aldobiouronic acids may be isolated as partial hydrolysis products. The formation of aldobiouronic acids by this pro- cedure provides a useful new method for establishing the mode of linkage between certain sugar residues in polysaccharides. Many plant polysaccharides con- tain Garabinofuranosyl residues as non-reducing end groups but it has seldom been possible to obtain conclusive evidence for the mode of attachment of such residues to the adjacent sugar residues.The particular value of the new approach in providing solutions to this type of structural problem may be illustrated from experiments on rye-flour arabino- xylan. In previous studies2 of this polysaccharide it was not possible to obtain direct evidence for the Cf. Mehltretter Ah. Carbohydrate Chern. 1953 8 231. presence of L-arabinofuranosyl residues attached as single-unit side-chains (I) although a comparison of the original and a degraded polysaccharide provided indirect evidence in favour of structure (I),in prefer- ence to structure (11) in which the arabinose units terminate xylose-containing side-chains.Whereas in the arabinoxylan (I) controlled acid-hydrolysis re- sults in selective cleavage of L-arabinofuranosyl units with the formation of a linear degraded xylan (In) in the oxidised polysaccharide (TV) the L-arabino- furanosiduronic acid linkages are the most resistant to acid hydrolysis and the aldobiouronic acid (3-~- xylose L-arabin0furanosid)uronicacid (V),has been characterised as a product of partial hydrolysis. The formation of this acidic disaccharide provides direct evidence for the presence in the arabinoxylan of L-arabinofuranosyl residues linked to the basal xylan chain as single unit side-chains. Aldobiouronic acids may be formed in a similar manner from hexose-containing polysaccharides but the value of the method is clearly limited to those hexosans containing only a limited number of free Aspinall and Sturgeon J.1957 4469. SEPTEMBER 1959 -.*4Xylp 1-4 Xylp + Arabim -*.4Xylp 1-4 3 Xylp I... L ;YIP t t ArafA I (IV) ArafA I (V) Aspinall Hirst and Ramstad J. 1958 593. 271 primary hydroxyl groups namely polysaccharides with a high proportion of 1,6-linkages. A poly-saccharide containing such a high proportion of 1,6-linked hexose (D-galactose) residues and also some L-arabinofuranose residues (although not as end groups) is the arabinogalactan (E-galactan) from European larch wood? Two aldobiouronic acids (6-~-galactoseP-D-galactopyranosid)uronicacid and (6-D-galactose L-arabinofuranosid)uronic acid have I ...(m> been isolated on graded hydrolysis of the oxidised polysaccharide and their formation in this way pro- vides further evidence for the detailed structure of the arabinogalactan. We thank Professor E. L. Hirst C.B.E F.R.S. for his interest and advice. (Received June 8th 1959.) The Structure of Nyctanthic Acid By G. H. WHITHAM (CHEMISTRY UNIVERSITY DEPARTMENT OF BIRMINGHAM) PREVIOUS work1 on nyctanthic acid a constituent of Nyctanthes arbor-tristis seeds favoured the empirical formula C3oH480 and showed the presence of a carboxyl group and of two double bonds thereby indicating a tetracyclic nucleus. One of the double bonds was a methylene group and the other was tri- substituted. A skeleton of the lanostane type was assumed by analogy with other tetracyclic triterpene acids though no evidence bearing on this point was presented.Oxidation of di hy dron yc t an t hin y1 acetate (n yc t -anthic acid; -CO,H +-CH,.OAc and :C=CH -+ :CH-CH,) with selenium dioxide in acetic acid under reflux gave a crystalline diene acetate which had Amax. (in EtOH) at 243 251 and 260 mp (E 24,600 28,700 and 18,800 respectively). This absorption pattern is characteristic of an oleana- 1 1,13( 18)-diene,2 and the reaction parallels the con- version of #3-amyrin acetate into 3/3-acetoxyoleana- 11,13( 18)-diene under the same conditions (whereas or-amyrh acetate is ~naffected).~ The biogenetically plausible structure (I),which readily accounts for the hitherto puzzling absence of a hydroxyl orketo-group thus appeared probable for nyctanthic acid and it was confirmed by partial synthesis from /3-amyrin as follows.Treatment of b-amyrenone oxime4 (II) with toluene-p-sulphonyl chloride in pyridine afforded a mixture of a lactam presumably (IU) on account of its relatively easy alkaline hydrolysis to an amino- acid and an unsaturated nitrile (IV). Structure (IV) was assigned on the basis of (a) the infrared spectrum which had vmax (in CCl,) 3050,1635 and 899 cm.-l (CH,=C<) and (b) the ready alkaline hydrolysis to an unsaturated acid. The identity of the latter acid with nyctanthic acid was shown by m.p. and infrared data and by comparison of the methyl esters. I thank Dr. J. H. Turnbull for Nyctanthes arbor- tristis seeds and Dr.T. G. Halsall for a gift of b-amyrin benzoate. (Received July 8th 1959.) Turnbull Vasistha Wilson and Woodger J. 1957 569. Barton and Brooks J. 1951 257. Ruzicka Miiller and Schellenberg Helv. Chim. Acta 1939 22 767. Rollett Monafsh. 1922 43 413. PROCEEDINGS The Synthesis of Hydrazine by Glowdischarge Electrolysis of Liquid Ammonia* By A. HICKLING and G. R. NEWNS (UNIVERSITY OF LIVERPOOL) A SERIES of studies of glow-discharge electrolysis in for different time intervals. It is seen that the initial aqueous systems1 has shown that this form of ion yield is about 2.5 moles of hydrazine per faraday of bombardment gives primarily the hydroxyl radical electricity passed and this falls only slightly with in-which subsequently undergoes reaction and under creasing quantities of electricity.The yield of some conditions can produce hydrogen peroxide. hydrazine was practically independent of the pres- This work has now been extended to systems in sure in the gaseous phase (70-175 mm.). It tended liquid ammonia and with an inert electrolyte it has to drop somewhat with increased current (0.015-41 been found that the chief product of glow-discharge amp.) and with increase in the concentration of the electrolysis is hydrazine. The conditions for its electrolyte used. Temperature-variation was limited formation are not critical and its purity seems solely by the freezing point of the ammonia and the dependent upon that of the ammonia used from feasibility of maintaining glow-discharge at high which it is readily separated by fractional distilla- pressures but an increase of temperature of 16" tion; this is in contrast to the usual methods of resulted in a small increase of yield.making hydrazine which result in a dilute aqueous In a number of experiments alternating (50 cycles) solution from which concentration to an anhydrous was used instead of direct current; hydrazine was product is difficult.2 Samples of hydrazine prepared again found but the initial yield was decreased to 1.3 by the present method with liquid ammonia direct as against 2.5 moles per faraday suggesting that it from a cylinder had m.p.s ranging from 1.0" to 1.5" is mainly during the anodic half cycle that hydrazine vapour pressure of 10.5 mm.at 20" and n22 1-4609. is formed. It has also been found that glow-discharge These values of physical properties correspond to electrolysis can be carried out in liquid ammonia at > 97 % purity which agrees with chemical analysis. atmospheric pressure with a small immersed anode In the cell used the anode was a platinum wire if the current passing exceeds a critical value. This suspended in the gas phase above the solution in causes the anode to become surrounded by an en- which a platinum cathode was immersed and a cur- velope of vapour through which a discharge at rent was passed by maintaining a discharge between several hundred volts takes place and the process the anode and the solution surface. The cell was kept again leads to hydrazine as the main product.at a reduced pressure and cooled by immersion in a It is probably premature to discuss the mechanism bath of solid carbon dioxide in acetone. The experi- of hydrazine formation. Work in the aqueous systems mental conditions usually employed were electro-(Denaro et all) has shown that in glow-discharge lyte 20 ml. of 0-Olwammonium nitrate in liquid gaseous ions with energies of about 100 ev enter the ammonia exposing a surface of 9 sq. cm.;anode to liquid and bring about dissociation of solvent mole- surface distance 0-5 cm.; pressure 100 mm. mer- cules and hydrogen peroxide probably arises by di-cury; current 0-025 amp.; voltage drop across the merisation of hydroxyl radicals. It is tempting by discharge ca. 600 v. analogy to suggest that hydrazine is produced in For studying the formation of hydrazine under liquid ammonia by a reaction of the type'2NH2 -+ various conditions the chosen current was passed for N,H, but the results so far obtained favour rather a measured time; the ammonia was then evaporated a reaction of the form NH + NH -+ N2H4.It is from the cell and the residue was acidified and hoped to discuss the mechanism fully in a later analysed volumetrically for hydrazine by the indirect communication.iodate method.2 In Fig. 1 are shown the results of a We thank the Department of Scientific and In- number of electrolyses under the above conditions dustrial Research for a maintenance grant to G.R.N. (Received,June 22nd 1959.) * This research provides the subject matter of Patent Appln.15,744159 filed by the National Research Development Corp. Davies and Hickling J. 1952,3595; Hickling and Linacre J. 1954,711 ;Denaro and Hickling J. Electrochem. Soc. 1958,105 265. Audrieth and Ogg "The Chemistry of Hydrazine," Wiley New York 1951. NEWS AND ANNOUNCEMENTS Carlsberg-Wellcome Fellowships.-Travelling Fel-head of the Statistical Department of the State lowships for 1959-60 of the Carlsberg Foundation Serum Institute Copenhagen and Mr. John (Copenhagen) and the Wellcome Trust (London) Anthony Hunt at present working in the Medical have been awarded to Mr. Michael Weis Bentzon Research Council's research unit for molecular SEPTEMBER 1959 biology at the Cavendish Laboratory Cambridge. Mr. Bentzon will work with Professor M.S. Bartlett at Manchester University on the application of modem statistical theory to biological problems and Mr. Hunt will study methods of protein chemistry with Professor M. Ottesen at the Carlsberg Labora- tory Copenhagen. The Perkin Centenary Trust.-The Perkin Cen- tenary Fellowship has been awarded to Mr. Brian Whitear A.R.I.C. of Plaistow London E. 13 and will be tenable from October lst 1959 in the Department of Chemistry at the University of Sou thamp t on. Perkin Centenary Scholarships have been awarded to Mr. Ronald R. Cox of Birmingham tenable at the University of Birmingham; to Mr. B. T. Lawton of Walkden near Manchester tenable at the Royal Technical College Salford; and to Mr. D. J. Pearson of Horsforth near Leeds tenable at the Bradford Institute of Technology.The Perkin Centenary Trust was established as a lasting memorial to the discovery in 1856 of the first important synthetic dyestuff Mauveine by William Henry Perkin. The centenary of this historic event was acknowledged by Celebrations held in London in 1956 and widely supported by many organisa- tions having an interest in chemistry including the Royal Society The Chemical Society The Society of Chemical Industry The Society of Dyers and Colourists The Royal Institute of Chemistry and The Association of British Chemical Manufacturers. The Trustees are Mr. H. Jackson (Chairman) Dr. D. W. Hill Mr. J. M. Leonard Mr. W. R. Mathers Mr. M. W. Perrin and Sir Robert Robin- son. The Secretary to the Trustees is Mr.J. R. Ruck Keene to whom enquiries relating to the awards should be addressed c/o The Chemical Society Burlington House London W. 1. British Association Granada Lectures.-A new series of annual lectures is organised by the British Association for the Advancement of Science and sponsored by Granada TV Network Limited on the theme of “Communication in the Modem World”. The broad purpose of the Granada Lectures is to explore the impact of communications in the field of mass media-Press Radio Television and Film- in their sociological political scientific and technical aspects. The Lectures will be delivered to an invited audience to include representatives of science industry politics education local government the mass media and others.The Inaugural Series of the Lectures will be given in the Guildhall London on the evenings of October 13th 19th and 27th. Sir Edward Appleton G.B.E. K.C.B. F.R.S. Principal and Vice-Chancel- lor of Edinburgh University will speak on the signi- 273 ficance of long-range communication and the exploration of space; the impact of television and radio in the field of politics and public affairs in a democracy will be the subject of Dr. Edward R. Murrow the well-known American television and radio commentator and news analyst; and Sir Eric Ashby Vice-Chancellor of Queen’s University Belfast will talk about methods of presenting scientific information to the public and the extent to which the role of the mass media in this field might be extended or improved.A limited number of tickets have been made avail- able for Fellows of The Chemical Society wishing to attend these Lectures. Application should be made before Saturday October 3rd to the Secretary The British Association Granada Lectures 36 Golden Square London W. I. International Congress on Corrosion.-The 1st International Congress on Metallic Corrosion will be held in South Kensington London during the week April 10-15th 1961 under the Presidency of Sir Harry Melville. Mr. E. Leslie Streatfield has been appointed Chairman of the Executive Committee and Dr. J. Ferguson Chairman of the Finance Com- mittee. The Honorary Secretary is Lt.-Col. Francis J. Griffin. Further information may be obtained from the Honorary Secretary at 14 Belgrave Square London S.W.1.Election of New Fellows.42 Candidates whose names were published in the Proceedings for June/ July have been elected to the Fellowship. Deaths of Fellows.-The deaths of the following Fellows are announced with regret Mr. Richard Selwyn Haskew (17.8.59) Chairman of the General Chemical and Pharmaceutical Co. Ltd. ;Mr. Stanley Augustus Mumford (July 1959) late Chief Super- intendent of Porton Experimental Station; and Mr James Francis Ronca (10.8.59) Director G. Wilson Gas Meters Ltd. and Radiant Heating Ltd. Personal.-Mr. D. L. Baulch of Melbourne Uni- versity has been awarded an I.C.I. Research Fellow- ship for post-doctorate work with Professor Dainton at the University of Leeds.Dr. David A. Brown of Queen Mary College London has been appointed Lecturer in Inorganic Chemistry at University College Dublin. Professor Melvin Calvin received the Honorary Degree of D.Sc. at the University of Oxford on August 1st. Mr. P. C. Chaumeton managing director of Styrene Products an associate of the Shell Chemical Company has retired. Mr. Gilbert Do& of Monsanto Chemicals Ltd. has been appointed a member of the Board of R. H. Cole & Co. Ltd. Dr. G. Malcolm Dyson has been appointed research director for Chemical Abstracts Service the publication of the American Chemical Society. He has been one of their consultants for several years. Dr. P. J. Garner has been appointed director of research at Shell Chemical Company’s Carrington Research Laboratory.Dr. Garner was formerly deputy laboratory manager at the Thornton Research Centre Cheshire. Mr. K. H. Handy works manager at the new Monsanto factory at Fawley near Southampton for the last two years has been appointed works manager at the Ruabon factory. Dr. E. W.Bodycote acting works manager at Ruabon has been ap- pointed works manage! at Fawley. A Stothert Research Fellowship has been awarded by the Council of The Royal Society to Mr. J. A. Hunt of Peterhouse Cambridge to work on the chemical structure of proteins. The British Columbia Research Council has offered to build a one million dollar biochemistry institute if Dr. H. G. Khorana will remain at the University of British Columbia.PROCEEDINGS Dr. M. F. Lappert Dr. J. E. Davies and Dr. J. Lee have been appointed Lecturers in chemistry at the Manchester College of Science and Technology. Mr. G. J. Leigh has been appointed an Assistant Lecturer. Mr. W.Lloyd Jenkins has been appointed Lecturer in the Department of Chemistry University College of Rhodesia & Nyasaland with responsibility for teaching Agricultural Chemistry . Dr. D. A. Sutton has been appointed Director of Research of the British Gelatine and Glue Research Association. Mr. G. V. TayZor has been appointed works manager at the Newport Monmouthshire factory of Monsanto Chemicals Ltd. in place of Dr. N. B. Dyson. Sir Owen Wansbrough-Jones is relinquishing his post as Chief Scientist to the Ministry of Supply and will join the board of Messrs.Albright & Wilson Ltd. on October 1st. Professor R. B. Woodward (Honorary Fellow) Morris Loeb Professor of Chemistry at Harvard University has been appointed Morrell Lecturer at the University of Cambridge for the academical year 1959-60. PROGRAMME OF MEETINGS* OCTOBER 1959 TO JANUARY 1960 * Reprints of this programme can be obtained from the General Secretary The Chemical Society Burlington House, Piccadilly London W.l. London Thursday October 15th 1959 at 7.30 p.m. Niels Bjerrum Memorial Lecture. To be given by Professor E. A. Guggenheim M.A. Sc.D. F.R.S. in the Rooms of the Society Burlington House London W.l. Thursday November 12th at 7.30 p.m. Meeting for the Reading of Original Papers.To be held in the Rooms of the Society Burlington House w.1. Thursday December loth at 7.30 p.m. Centenary Lecture “Some Recent Advances in Fluorocarbon Chemistry” by Professor G. H. Cady. To be given in the Rooms of the Society Burlington House W.l. Thursday January 14th 1960 at 7.30 p.m. Tilden Lecture “Progress in the Study of Hetero- geneous Catalysis” by Professor C. Kemball M.A. Ph.D. F.R.I.C. To be given in the Lecture Theatre The Royal Institution Albemarle Street W. 1. Aberdeen (Meetings will be held in the University Union.) Thursday October 15th 1959 at 8 p.m. Lecture “Milk and Milk Products in Under-developed Countries” by Professor H. D. Kay C.B.E. F.R.S. Joint Meeting with the Royal Institute of Chemistry and the Society of Chemical Industry.Monday November 16th at 8 p.m. Lecture “Stereochemistry of the Complex Halides of the Transition Metals” by Professor R. S. Nyholm D.Sc. F.R.I.C. F.R.S. Joint Meeting with the Royal Institute of Chemistry and the Society of Chemical Industry. Thursday December loth at 8 p.m. Lecture “Chemical Kinetics in Relation to Large Scale Production” by Professor K. G. Denbigh. Joint Meeting with the Royal Institute of Chemistry and the Society of Chemical Industry. Thursday January 21st 1960 at 8 p.m. Lecture “Synthetic Detergent Washing Powders” by Mr. L. N. Savidge. Joint Meeting with the Royal Institute of Chemistry and the Society of Chemical Industry. SEPTEMBER 1959 Birmingham (Meetings will be held in the Large Chemistry Lecture Theatre The University.) Friday October 23rd 1959 at 4.30 p.m.Lecture “Some Recent Developments in the Porphyrin Field” by Professor G. W. Kenner Ph.D. Sc.D. Joint Meeting with Birmingham Uni- versity Chemical Society. Friday November 13th at 4.30 p.m. Lecture “Spectra and Reaction Kinetics of Some Organic Anions” by Dr. E. Warhurst M.Sc. Joint Meeting with Birmingham University Chemical Society. Friday January 15th 1960 at 4.30 p.m. Lecture ‘‘Alkali Metal Derivatives of Organic and Organometallic Compounds” by Professor G. E. Coates. Joint Meeting with Birmingham University Chemical Society. Bristol (Meetings will be held in the Department of Chemistry The University unless otherwise stated.) Thursday October lst 1959 at 6.30p.m.Lecture “Some Recent Developments in Sterol Hormones” by Dr. B. A. Hems F.R.I.C. Joint Meeting with the Royal Institute of Chemistry and the Society of Chemical Industry. Thursday October 8th at 7.30 p.m. Lecture “Polyurethanes” by Mr. J. M. Buist and Mr. R. Packer. Joint Meeting with the Royal Institute of Chemistry and the Society of Chemical Industry to be held at the Technical College Brunswick Road Gloucester. Thursday October 15th at 5.15 p.m. Lecture “Chromatography” by Dr. F. H. Pollard. Joint Meeting with the Student Chemical Society. Thursday October 15th at 6.30 p.m. Lecture “Dusts” by Dr. P. F. Holt F.R.I.C. Joint Meeting with the Royal Institute of Chemistry and the Society of Chemical Industry to be held at British Cellophane Bridgwater.Thursday October 22nd at 6.30 p.m. Lecture “Ferrocene as an Aromatic System” by Professor P. L. Pauson. Joint Meeting with the Royal Institute of Chemistry and the Society of Chemical Industry. Thursday October 29th at 7.30 p.m. Dinner and Dance jointly with the Royal Institute of Chemistry and the Society of Chemical Industry to be held at the Cavendish Rooms Hawthorns Hotel Bristol 8. Friday October 3Oth at 5.15 p.m. Lecture “Chemical Aspects of the Origin of Life” by Professor J. D. Bernal F.R.S. Joint Meeting with the Student Chemical Society. Thursday November 19th at 5.15 p.m. Lecture “Fungi as Molecular Architects” by Dr.W. B. Whalley F.R.I.C. Joint Meeting with the Student Chemical Society. Thursday November 26th at 5.15 p.m. Lecture by Dr. M. F. Perutz. Joint Meeting with the Student Chemical Society. Thursday December 3rd at 6 p.m. Lecture “Beryllium Production Properties Ap- plications” by Dr. G. A. Wolstenholme. Joint Meeting with the Chemical Engineering Group the Institute of Metals the Royal Institute of Chemistry and the Society of Chemical Industry. To be followed by Dinner at 8 p.m. Thursday December loth at 5.15 p.m. Lecture “Rockets” by Dr. J. Black. Joint Meeting with the Student Chemical Society. Friday January 8th 1960 at 6.30 p.m. Lecture “Radiochemical Analysis” by Dr. J. N. Andrews A.R.I.C. Joint Meeting with the Royal Institute of Chemistry the Society for Analytical Chemistry and the Society of Chemical Industry to be held at the College of Technology Ashley Down Bristol 7 Thursday January 14th at 6.30 p.m.Lecture “The Dyeing of the Newer Synthetic Fibres” by Mr. J. G. Graham B.Sc. Joint Meeting with the Royal Institute of Chemistry and the Society of Chemical Industry. Thursday January 28th at 6.30 p.m. Lecture by Professor A. L. Roberts. Joint Meeting with the Institute of Fuel the Royal Institute of Chemistry and the Society of Chemical Industry. Cambridge (Meetings will be held in the University Chemical Labor at or y ,Lensfield Road.) Monday October 12th 1959 at 5 p.m. Lecture “Diterpene Synthesis” by Dr. J. A. Barltrop M.A. Friday October 30th at 8.30 p.m.Lecture “Developments in the Chemistry of Bac-terial Walls” by Professor J. Baddiley Ph.D. D.Sc. Joint Meeting with the University Chemical Society. Monday November 2nd at 5 p.m. Lecture “Quadrupole Moments” by Dr. D. Buckingham. Friday November 6th at 8.30 p.m. Lecture “Fast Halogenation Reactions in Solution” by Mr. R. P. Bell M.A. F.R.S. Joint Meeting with the University Chemical Society. Monday November 16th at 5 p.m. Lecture “Maytenone a Tetraterpene?” by Dr. T. J. King. Friday November 27th at 8.30 p.m. Lecture “Some Studies on Peptides of Cystine” by Professor H. N. Rydon Ph.D. D.Phil. F.R.I.C. Joint Meeting with the University Chemical Society. Monday December 7th at 5 p.m. Lecture “Fluorescence of Organic Vapours” by Dr.B. Stevens M.A. Monday January 18th 1960 at 5p.m. Lecture “Chelatometry” by Dr. T. S. West. Tuesday January 19th at 11.30 a.m. Lecture “The Analysis of Microgram Amounts of Organic Compounds” by Dr. T. S. West. Friday January 29th at 8.30 p.m. Lecture “The Structure of Myoglobin” by Dr. J. C. Kendrew M.A. Joint Meeting with the University Chemical Society. Cardiff (Meetings will be held in the Chemistry Department University College Cathays Park.) Monday October 26th 1959 at 5.30 p.m. Lecture “Recent Developments in the Crystallo- graphy of Vitamin BIZ” by Dr. Dorothy M. Hodgkin F.R.S. Monday November 16th at 5.30 p.m. Lecture “Synthetic Studies in the Vitamin D Field” by Professor B.Lythgoe Ph.D. F.R.I.C. F.R.S. Joint Meeting with the Student Chemical Society. Monday January 25th 1960 at 5.30 p.m. Lecture “Some Models of Physical Adsorption” by Professor D. H. Everett M.B.E. Durham (Meetings will be held in the Science Laboratories The University.) Monday October 19th 1959 at 5 p.m. Lecture “Applications of Colloid Science to Problems of Lubrication” by Dr. J. B. Matthews F.R.I.C. Joint Meeting with the Durham Colleges Chemical Society. Monday November 2nd at 5 p.m. Lecture “Recent Advances in Infrared Spectro-scopy” by Dr. L. J. Bellamy. Joint Meeting with the Durham Colleges Chemical Society. PROCEEDINGS Monday November 16th at 5 p.m. Lecture “Reaction Mechanisms” by Professor E. D. Hughes F.R.P.C.F.R.S. Joint Meeting with the Durham Colleges Chemical Society. Monday November 30th at 5 p.m. Lecture “Acyl Trifluoroacetates and Related Com- pounds” by Professor E. J. Bourne Ph.D. F.R.I.C. Joint Meeting with the Durham Colleges Chemical Society. Tuesday January 26th 1960 at 5 p.m. Lecture “Reactions in Liquid Dinitrogen Tetro- xide” by Dr. C. C. Addison F.Inst.P. F.R.I.C. Joint Meeting with the Durham Colleges Chemical Society. Edinburgh Thursday October lst 1959 at 5.30 p.m. Society of Chemical Industry Lister Memorial Lecture entitled “Chemical Structure and Action of Morphine-like Analgesics and Related Substances” by Dr. N. B. Eddy. Joint Meeting with the Royal Institute of Chemistry and the Society of Chemical Industry to be held in Adam House The University Chamber Street.Thursday October 29th at 7.15 p.m. Lecture “Stereospecific Polymerisation and Tactic Polymers” by Dr. M. Gordon. Joint Meeting with the Royal Institute of Chemistry and the Society of Chemical Industry to be held at the Arthur D. Little Research Institute Inveresk Musselburgh. Thursday November 12th at 7.30 p.m. Lecture “Nitrogen Fixation” by Dr. E. R. Roberts. Joint Meeting with the Royal Institute of Chemistry and the Society of Chemical Industry to be held at the North British Station Hotel. Tuesday December Sth at 7.30 p.m. Lecture “The Biosynthesis of Porphyrins” by Professor A. W. Johnson Ph.D. Sc.D. A.R.C.S. Joint Meeting with the Royal Institute of Chemistry and the Society of Chemical Industry to be held at the North British Station Hotel.Friday January Sth 1960 at 7.30 p.m. Lecture “Chemistry Applied to Criminal Investiga- tion” by Detective Chief Inspector J. K. McLellan. Joint Meeting with the Royal Institute of Chemistry and the Society of Chemical Industry to be held at the North British Station Hotel. Exeter (Meetings will be held in the Washington Singer Laboratories Prince of Wales Road.) Friday October 23rd 1959 at 5 p.m. Lecture “Some Photochemical Rearrangements” by Professor D. H. R. Barton F.R.S. SEPTEMBER 1959 Friday November 27th at 5 p.m. Lecture “The Acetylenic Approach to the Synthesis of Natural Products” by Professor R. A. Raphael DSc. Ph.D. F.R.I.C.Glasgow (Meetings will be held in the Chemistry Department The University.) Thursday October 22nd 1959 at 4 p.m. Lecture “Chemical Propellents” by Mr. W. S. Wood B.Sc. F.R.I.C. A.M.1.Chem.E. Joint Meet- ing with the Alchemists’ Club and the Andersonian Chemical Society. Friday November 20th at 4 p.m. Lecture “Reaction Mechanisms” by Professor E. D. Hughes F.R.I.C. F.R.S. Joint Meeting with the Alchemists’ Club. Friday January 22nd 1960 at 4 p.m. Lecture “Energy Transfer in Gases” by Professor T. L. Cottrell. Joint Meeting with the Alchemists’ Club. Hull (Meetings will be held in the Organic Lecture Theatre Chemistry Department The University.) Thursday November 12th 1959 at 7.30 p.m. Lecture “Liquid Crystals in Solutions of Polyep- tides and Other Substances” by Dr.Conrnar Robinson. Joint Meeting with the Royal Institute of Chemistry. Tuesday December lst at 5 p.m. Lecture to be given by Professor D. D. Eley Sc.D. Ph.D. Joint Meeting with the University Student Chemical Society. f rish Republic Friday November 13th 1959 at 7.45 p.m. Lecture “The Stereochemistry of Some Metal Ions” by Dr. L. E. Orgel M.A. Joint Meeting with the Werner Society to be held in the University Chem- ical Laboratory Trinity College Dublin. Wednesday January 6th 1960 at 5.30 p.m. Lecture “Aescigenin” by Mr. J. B. Thomson. To be given in the Chemistry Department University College Upper Merrion Street Dublin. Leeds Thursday November 26th 1959 at 6.30 p.m. Lecture “Chemical and Biogenetical Studies on Emetine” by Dr.A. R. Battersby. Joint Meeting with the University of Leeds Union Chemical Society to be held in the Chemistry Lecture Theatre The University. Leicester (Meetings will be held at The University.) Monday November 2nd 1959 at 4.30 p.m. Lecture “The Organic Reactions of Molecular Oxygen” by Dr. A. G. Davies. Joint Meeting with the University of Leicester Chemical Society. Monday November 30th at 4.30 p.m. Lecture “Pure Metals” by Dr. J. C.Chaston. Joint Meeting with the University of Leicester Chemical Society. Liverpool (Meetings will be held in the Department of In-organic and Physical Chemistry The University.) Thursday October 29th 1959 at 5 p.m. Lecture “The Growth of Fluorocarbon Chemistry” by Professor R.N. Haszeldine D.Sc. Ph.D. F.R.I.C. Joint Meeting with the University Chemical Society. Thursday November 26th at 5 p.m. Lecture “Experimental Methods in Chemical Kinetics” by Professor J. C. Robb D.Sc. Ph.D. A.R.I.C. Joint Meeting with the University Chemical Society. Thursday January 28th 1960 at 5 p.m. Tilden Lecture “Progress in the Study of Hetero- geneous Catalysis” by Professor C. Kemball Ph.D. F.R.I.C. Joint Meeting with the University Chemical Society. Newcastle upon Tyne (Meetings will be held in the Chemistry Department King’s College.) Friday October 30th 1959 at 5.30 p.m. Bedson Club Lecture “Ubiquinone and Vitamin A” by Professor R. A. Morton Ph.D. D.Sc. F.R.I.C. F.R.S. Friday November 20th at 5.30 p.m.Bedson Club Lecture “Organic Semi-conductors“ by Professor D. D. Eley Ph.D. Friday December 4th 1959 at 5.30 p.m. Lecture “Bridged Rings” by Professor R. C. Cookson. Friday January 29th 1960 at 5.30 p.m. Bedson Club Lecture “Activation of Carbon-Carbon Double Bonds by Cationic Catalysts” by Professor A. G. Evans Ph.D. D.Sc. Northern Ireland (Meetings will be held in the Department of Chemistry Queen’s University Belfast.) Tuesday October 27th 1959 at 7.45 p.m. Lecture “ Some Properties of Heteroaromatic Amines” by Dr. K. Schofield. Joint Meeting with the Royal Institute of Chemistry and the Society of Chemical Industry. Tuesday December lst at 7.45 p.m. Lecture “Oxidation of Organic Sulphides” by Dr.L. Bateman. Joint Meeting with the Royal Institute of Chemistry and the Society of Chemical Industry. Thursday January 26th 1960 at 7.45 p.m. Lecture “Electron Resonance of Free Radicals” by Dr. D. H. Wen M.A. Joint Meeting with the Royal Institute of Chemistry and the Society of Chemical Industry. North Wales (Meetings will be held in the Department of Chemistry University College of North Wales.) Thursday October 29th 1959 at 5.45 p.m. Lecture “Recent Developments in the Study of Ionic Solutions” by Professor K. W. Sykes M.A. D.Phi1. Joint Meeting with the University College of North Wales Chemical Society. Thursday January 28th 1960 at 5.45 p.m. Lecture “Synthesis and Dissolution of Starch in Plants” by Professor Helen K.Porter D.Sc. F.R.S. Joint Meeting with the University College of North Wales Chemical Society. Nottingham (Meetings will be held in the Chemistry Department The University unless otherwise stated.) Tuesday November loth 1959 at 5 p.m. Lecture “Alkyl and Aryl Derivatives of Transition Metals” by Dr. J. Chatt M.A. F.R.I.C. Joint Meeting with the University of Nottingham Chemical Society. Tuesday December Sth at 8 p.m. Lecture “The Study of Knock and Anti-knock by the Method of Kinetic Spectroscopy” by Professor R. G. W.Norrish Sc.D. Ph.D. F.R.I.C. F.R.S. Joint Meeting with the Society of Chemical Industry and the University of Nottingham Chemical Society. Thursday December loth at 7.30 p.m. Lecture “Infrared Spectroscopy” by Dr.L. J. Bellamy. Joint Meeting with the Royal Institute of Chemistry to be held at Nottingham and District Technical College. Tuesday January 26th 1960 at 8 p.m. Official Meeting and Tilden Lecture “Hydrocarbon Metal Carbonyls” by Professor P. L. Pauson. oxford (Meetings will be held in the Inorganic Chemistry Laboratory.) PROCEEDINGS Monday October 19th 1959 at 8.15 p.m. Lecture “Absolute Configuration” by Dr. W. Klyne. Joint Meeting with the Alembic Club. Monday November 9th at 8.15 p.m. Lecture “Nuclear Magnetic Resonance” by Dr. R. E. Richards M.A. F.R.S. Joint Meeting with the Alembic Club. Monday November 30th at 8.15 p.m. Lecture by Dr. E. Schlittler. Joint Meeting with the Alembic Club. St. Andrews and Dundee (Meetings will be held in the Chemistry Department St.Salvator’s College St. Andrews unless otherwise stated.) Friday October 3&h 1959 at 5.15 p.m. Lecture “Experimental Methods in Chemical Kinetics” by Professor J. C. Robb Ph.D. A.R.I.C. Joint Meeting with the Univerity Chemical Society. Friday November 13th at 5.15 p.m. Lecture “Some Recent Developments in the Porphyrin Field” by Professor G. W. Kenner Ph.D. Sc.D. Joint Meeting with the University Chemical Society. Tuesday November 24th at 5 p.m. Lecture “Chemotherapy” by Dr. F. L. Rose O.B.E. F.R.I.C. F.R.S.To begiveninthechemistry Department Queen’s College Dundee. Friday November 27th at 5.15 p.m. Lecture “Aromatic Character and the Inorganic Aromatics” by Professor D.P. Craig. Joint Meeting with the University Chemical Society. Tuesday January 12th 1960 at 5 p.m. Lecture “The Chemical Contribution to Cancer Research” by Professor A. Haddow F.R.S. To be given in the Chemistry Department Queen’s College Dundee. Friday January 15th at 5.15 p.m. Lecture “Some Effects of Remotely Placed Groups upon Reactions of Organic Molecules” by Professor H. 13. Henbest Ph.D. A.R.C.S. A.R.I.C. Joint Meeting with the University Chemical Society. Sheffield (Meetings will be held in the Chemistry Department The University.) Thursday October 22nd 1959 at 4.30 p.m. Lecture “The Inorganic Chemistry of the Nitrate Group” by Dr. C. C. Addison F.Inst.P. F.R.I.C. Joint Meeting with the Royal Institute of Chemistry and the University Chemical Society.Thursday November 19th at 7.30 p.m. Lecture “Chemistry in Fruit and Vegetable Can-ning” by Dr. D. Dickinson. Joint Meeting with the SEPTEMBER 1959 Royal Institute of Chemistry and the University Chemical Society. Thursday November 26th at 4.30 p.m. Lecture “Infrared Investigation of the Structure of High Polymers including those of Biological Interest” by Dr. G. B. B. M. Sutherland F.R.S. Joint Meeting with the Royal Institute of Chemistry and the University Chemical Society. Thursday January 21st 1960 at 4.30 p.m. Lecture “Some Recent Studies with Natural Products” by Professor E. R. €3. Jones Ph.D. D.Sc. F.R.I.C. F.R.S. Joint Meeting with the Royal Institute of Chemistry and the University Chemical Society.Southampton (Meetings will be held in the Chemistry Depart- ment The University.) Friday October 16th 1959 at 5 pm. Lecture “Applications of Chemistry in the Interpro tation of Infrared Spectra” by Dr. L. J. Bellamy. Joint Meeting with the University Chemical Society and the Royal Institute of Chemistry. Friday November 13th at 5 p.m. Lecture “Chemical Applications of Oxygen-18” by Dr. C. A. Bunton. Joint Meeting with the Uni- versity Chemical Society. Friday November 27th at 5 p.m. Lecture “Solvent Extraction of Inorganic Sub-stances” by Dr. A. G. Maddock. Joint Meeting with the University Chemical Society. Swansea (Meetings will be held in the Department of Chemistry University College.) Tuesday October 27th 1959 at 5.15 pm.Lecture “The Chemistry of Vitamin B,,” by Professor A. W. Johnson Sc.D. Ph.D. A.R.C.S. Joint Meeting with the University College of Swansea Chemical Society. Monday November 9th at 5.15 p.m. Lecture “ Interhalogen Compounds and Poly-halides” by Dr. A. G. Sharp M.A. E.R.I.C. Joint Meeting with the University College of Swansea Chemical Society. Friday December 4th at 3 p.m. Lecture “Fun with Free Radicals” by Professor D. H. Hey D.Sc. Ph.D. F.R.I.C. F.R.S. Joint Meeting with the University College of Swansea Chemical Society. Tees-side Monday September 28th 1959 at 8 p.m. Lecture “Pharmacology of Polymethylenes” by Dr. H. R. Ing M.A. D.Phil. F.R.S.To be given at the Constantine Technical College Middlesbrough.Monday November 2nd at 8 p.m. Lecture “Recent Developments in the Chemistry of the Less-common Elements” by Professor R. S. Nyholm D.Sc. F.R.I.C. F.R.S. Joint Meeting with the Society of Chemical Industry to be held at the William Newton School Norton Stockton-on- Tees. Wednesday December 9th at 8 p.m. Film Show. To be given at Spark’s Cafe High Street Stockton-on-Tees. APPLICATIONS FOR FELL0WSHIP (Fellows wishing to lodge objections to the election of these candidates should communicate with the Honorary Secretaries within ten days of the publication of this issue of Proceedings. Such objections will be treated as confidential. The forms of application are available in the Rooms of the Society for inspection by Fellows.) Beddall Kenneth Arthur Toller B.Sc.4 Woodgrange Gardens Bush Hill Park Enfield Middlesex. Bottomley John Neville B.Sc. 56 Lightwoods Hill Warley Woods Smethwick 41 Staffs. Bunch Kenneth B.Sc. Flat 1 17 Pacific St. Manly Sydney New South Wales Australia. Coldrey James Michael BSc. Ph.D. 19 Park Valley The Park Nottingham. Davison Peter Sinclair B.Sc. Ph.D. Halheath House nr. Dunfermline Scotland. Evans Peter Richard B.A. University Chemical Labora- tory Lensfield Road Cambridge. Evans Robert John B.Sc. Ph.D. 220-7th Avenue N., Texas City Texas U.S.A. Hugo Jennifer Mary M.Sc. Pius XI1 University College P.O. Roma Basutoland South Africa. Inward Peter William B.Sc. 252 Cowdray Avenue Colchester Essex. Jeffs. Peter Walter A.R.I.C.University of Natal Durban South Africa. Kampmeier Jack A, A.B. 219 Noyes Laboratory Uni-versity of Illinois Wrbana Illinois U.S.A. Kirby Edward Cameron B.Sc. 44Harringworth Corby Northants. Lane David Charles B.Sc. 26 Hilfield Lane Aldenham Watford Herts. Leonard Dennis Reigh. 89 Hardy Street Dover Heights New South Wales Australia. Ramsay George Caird. 8 Selby Street Grassmere South Australia. Robbins Derek A.R.I.C. 28 Woodgate Drive Birstall Le ice ster. Schmeisser Martin. Institut fur Anorganische Chemie und Elektrochemie Technische Hochschule Aachen Germany. Taylor John Bodenham B.Sc. 4 Park Crescent Merthyr Tydfil Glamorgan. Vickars. Michael Austin. B.Sc.. Ph.D. Colonial Pesticides Research P.O. Box 204 kusha Tanganyika East Africa.Walker Thomas B.Sc. Ph.D. 60 Paxford Road North Wembley Middlesex. Westley John William B.Sc. Department of Chemistry The University Nottingham. 280 PROCEEDINGS OBITUARY NOTICES SIR JAMES ECKERSLEY MYERS 1890-1958 JAMESECKERSLEY MYERS was born in Bolton Lancashire on June 24th 1890 the elder son of William and Sarah Myers. He received his education at the Manchester Grammar School and the Univer- sity of ManChester where he graduated with First Class Honours in Chemistry in 1910 and was elected a Fellow of the Society in the same year. His appoint- ment as Assistant Lecturer in Chemistry at Man- Chester in 1912 was followed within two years by the outbreak of World War I and the temporary eclipse of most academic research.Throughout the war period he with many of his colleagues devoted his energies to routine analysis and research connected with the war. For this work he was awarded the O.B.E. In 1918 he returned to pure research and following up the early interest of his research training in ex-perimental kinetic studies he published in collabora- tion with E. S. Hedges a number of papers in the Journal on periodical chemical reactions and a mono- graph on “The Problem of Physico-Chemical Periodicity.” Myers however had a taste and a genius for administration. He was appointed Senior Lecturer in Chemistry and Secretary and Tutor to the Faculty of Science at Manchester in 1920. As the years passed his administrative duties grew steadily and he was elected to one high office after another serving always with distinction.Among his appointments may be cited Assistant to the Vice-chancellor of the University of Manchester Justice of the Peace for the County of Cheshire Chairman of the Stockport Division Juvenile Court ,Chairman of the Northern Joint Matriculation Board President of the Association of University Teachers Member of the Norwood Committee on Secondary Education and many others. In 1935 he was appointed Principal of the Manchester College of Technology where thirteen years earlier his father had held the Chair of Textile Technology. The Second World War drew Myers inevitably into a multitude of committees connected with the war effort. He was knighted in 1950.The strain of the war years and the death of his only son seriously undermined his health and he resigned from his Principalship in 1951 to return to the University of Manchester as its Director of the School of Education. Here he remained until his retirement at the age of 65 in 1955. He died rather suddenly on December 5th 1958 at his home in Colwyn Bay. He is survived by his widow formerly Miss Elsie Ingram whom he married in 1917. Though Myers’s researches were few in number his services to chemists and the community were very many and he will be remembered by all who came into contact with him for his unvarying kindliness and tolerance and deep interest in education. F. FAIRBROTHER. THOMAS HAROLD DURRANS 1885-1 958 THOMAS DURRANS HAROLD was born in 1885 of a Huguenot stock his ancestors having migrated from France during the times of religious persecution.He studied chemistry at Birkbeck College London and after taking his degree joined the staff of A. Boake Roberts and Co. Ltd. of Stratford London in 1907 and remained with them until his retirement. He be- came a Fellow of the Chemical Society in 1906. In 1907 Abrac were a small firm producing sulphites and a few other products mainly for the brewing trade. Durrans’s arrival practically coincided with the formation of a new section known as the “Fine Chemicals Department” for making products of odorous character now known as “aromatics”. In its early stages this Department was concerned -entirely with the production mainly by fractional distillation of “isolates” from essential oils.Durrans realised that many of these could be made as “syn- thetics” but his first venture into this field was a simple esterification plant to make “essence of pear,” alias amyl acetate. In the early part of the 1914-1918 war this plant was greatly enlarged at Government request as solvents were needed and other war-time manufacture commenced. In 1916 the firm now somewhat grown established a Research Depart- ment with Durrans as chief but under the direction of Dr. W. H. Perkin jun. who had recently trans- ferred from Manchester to Oxford as Waynflete Professor of Organic Chemistry. The new Depart- ment was established in one of the rooms at the new Dyson Perrins Laboratory at Oxford.There Durrans and his colleagues worked on synthetic aromatics such as musk xylene war requirements such as triphenyl phosphate and solvent esters. SEPTEMBER 1959 During this period Durrans became interested in fractional distillation and had considerable corres- pondence with Professor Sydney Young of Dublin who was at that time the leading authority. This re-sulted in his contributing a chapter on Essential Oils -written after a visit to most of the essential-oil distilleries in Europe-to Young’s book “Distilla- tion Principles and Processes” (Macmillan 1922). In later years this interest resulted in several com- munications to the Perfumery and Essential Oil Record. In 1923 the Oxford venture was closed and Durrans established a Research Department at Stratford.Under his guidance which continued throughout his active career it laid the foundations of most of the manufactures which the firm began to develop. The war-time demand for aeroplane dope was followed by the beginnings of the plastics in- dustry and thus tritolyl phosphate followed triphenyl phosphate as a plasticiser or high-boiling solvent and various esters followed amyl acetate as a low-boiling solvent. The production of materials for use in the plastics and surface-coating industries in- creased rapidly. Durrans soon became an inter-nationally recognised authority in this field and his book entitled “Solvents” first published in 1930 (Chapman and Hall London) has run through seven editions and is regarded as a classic.Durrans was also regarded in his later active 28 I years as an authority on Fine Chemicals in general. A. J. V. Underwood in the chair at a lecture which he gave on the Fine Chemicals Industry made an apt quotation from Goldsmith “And still they gazed and still the wonder grew; That one small head could carry all he knew.” In 1936 Durrans became a director of the firm. During 1939-1941 he was chairman of the London Section of the Society of Chemical Industry and for the term 1941-1944 he was Vice-chairman of the Society itself. During his career he had also become an F.R.I.C. and had obtained a D.Sc. of London University by an investigation into chlorination by means of sulphuryl chloride.In 1946 he resigned and went to live in his beloved Cotswold country where he had nearly always spent his holidays but he did not rusticate completely. He retained a live interest in the progress of chemistry and kept his “Solvents” up to date until last year. His wife whom he had married early in life died a few months ago and he did not long survive her; they had been greatly attached to each other. They leave two sons. His death occurred on November 25th. All who worked with him found him to be a considerate helpful colleague and an inspiring enthusiast. It would be difficult to find anyone who could not speak well of him. A. L. BLOOMFIELD. WOLFGANG PAUL1 1900-1958 THEdeath of Wolfgang Pauli on December 15th 1958 at the age of 59 has deprived the world of one of its greatest scientists.His work will be remembered as long as science lives and long after his obituary notices have crumbled to dust. Though in our time he has taken his place among the great masters of theoretical physics his greatest discovery had no less an impact on chemical theory and it was in recogni- tion of this that he was elected an Honorary Fellow of the Chemical Society in January 1958. Born in Vienna and later a pupil of Sommerfeld and a colleague of Bohr Pauli reached scientific maturity at the time when the quantum theory in its modern form was coming to birth. The Bohr- Rutherford theory of the hydrogen atom had achieved triumphant success in accounting for the line spectrum of atomic hydrogen and the theory had been extended by Sommerfeld to more complex atoms.An outstanding problem however was to find the principles determining the distribution of atomic electrons in their Bohr orbits. In 1924 Pauli proposed an essentially correct solution to this prob- lem namely that not more than two electrons can occupy the same Bohr orbit at the same time. The Exclusion Principle as it cam: to be called was to be regarded not as a consequence of any existing physical postulates but as an independent law of nature in its own right. It made immediate sense of the Periodic Table of the elements and provided a theoretical basis for the concept of the electron pair which was one of the cornerstones of the new elec- tronic theory of valency.The famous experiment by Stern and Gerlach had already shown that univalent atoms possess magnetic moments in their ground states and in 1925 Goudsmit and Uhlenbeck ad- vanced the suggestion that the electron possesses an intrinsic angular momentum and an associated mag- netic moment which can be aligned in either of two ways relative to an external magnetic field. If account were taken of the two alternative spin alignments the Exclusion Principle could be summed up by saying that two electrons occupying the same orbit must have their spins antiparallel and this explained the diamagnetism of the vast majority of organic substances. It was not however until the discovery of the new PROCEEDINGS quantum mechanics that Pauli’s law could be stated in its most general form.The concept of an individual electron orbit was clearly unsatisfactory in view of the repulsion between electrons which results in a disturbance of their individual motions; the new wave mechanics replaced this concept by that of a many-electron wave function which represents a “probability amplitude” associated with every con- ceivable set of positions and spins for the various electrons. This amplitude according to the new theory could be positive or negative or indeed com- plex; Pauli’s Principle could then be formulated exactly by saying that the wave function for a set of electrons is reversed in sign if the positions and spins of any pair of electrons are interchanged.This more exact if more abstruse statement of the Exclusion Principle could be shown to reduce to Pauli’s original generalisation in so far as electrons could be assigned to individual orbits and has now taken its place as one of the two basic postulates of chemical theory the other being the wave equation itself. Pauli’s second great contribution to science was the neutrino hypothesis which he advanced in 1931. The dilemma which led to this conjecture was the apparent breakdown of the laws of conservation of energy momentum and angular momentum in the decay of p-radioactive nuclei. Briefly it had been ex- pected that since the initial and the final states of the emitting nucleus must have sharply defined energies the emergent /?-particle must always have the same energy.Pauli showed that if a very light neutral particle is supposed to emerge from the emitting nucleus at the same time as the electron then not only energy but also spin and angular momentum can be conserved on the assumption that the “neutrino” moves at something approaching the speed of light. Nowadays we know of so many funda- mental particles that it is easy to forget the courage which must have been necessary to advance such a revolutionary suggestion. Needless to say the neutrino is now an essential link in the modern theory of the fundamental particles and is indeed held responsible for some of the phenomena referred to as the “non-conservation of parity.” Pauli’s last contribution to theoretical physics was his joint work with Heisenberg on the so-called “wave equation for matter.” This was an attempt to obtain a fundamental equation from which could be extracted the masses spins and interaction con- stants of all the fundamental particles.This work however is still incomplete and its significance is as yet uncertain. Pauli’s influence on the recent development of theoretical science would be hard to overestimate. In matters experimental it is claimed that his influence was of a different kind and has indeed been referred to as the “Pauli effect.” It is said that a delicate experiment at the University of Zurich was yielding absurd and erratic results one day and had to be abandoned. On the following day the apparatus worked perfectly but it was later discovered that Pauli had passed through the city in a train a few hours earlier.This effect was however disregarded by the Committee which awarded him the Nobel Prize for Physics in 1945. H. C. LONGUET-HIGGINS. ADDITIONS TO THE LIBRARY Monographs on the Physics and Chemistry of Materials. Experimental techniques in low-temperature physics. G. K. White. Pp. 328. Clarendon Press. Oxford. 1959. (Presented by the publishers.) The determination of molecular structure. P. J. Wheatley. Pp. 263. Clarendon Press. Oxford. 1959. (Presented by the author.) Infrared absorption spectra index for 1945-1957. Compiled by H. M. Hershenson. Pp. 111. Academic Press. New York. 1959. Mass transfer between phases.T. K. Sherwood. Sponsored by the Pennsylvania State University. Pp. 86. Pennsylvania State University. University Park Penn- sylvania. 1959. (Presented by the publishers.) Precipitation from homogeneous solution. L. Gordon M. L. Salutsky and H. H. Willard. Pp. 187. John Wiley and Sons Inc. New York. 1959. (Presented by Chapman and Hall.) The structure of electrolytic solutions. Edited by Walter J. Hamer. A symposium sponsored by the Electrochemical Society Inc. and the National Science Foundation and held in Washington 1957. (Electrochemical Society Series.) Pp. 441. John Wiley and Sons Inc. New York. 1959. (Presented by Chapman and Hall.) Free radicals an introduction. A. F. Trotman-Dickenson. (Methuen’s Monographs on Chemical Sub-jects.) Pp.329. Methuen and Co. Ltd. London. 1959. Vistas in free-radical chemistry. Published in memoriam to Morris S. Kharasch. Edited by W. A. Waters. (Tetra- hedron. 1959 Supplement No. 3.) (International Series of Monographs on Organic Chemistry. Vol. 1.) Pp. 251. Pergamon Press. London. 1959. (Presented by the editor.) Substitution at elements other than carbon being the Fifth Weizmann Memorial Lecture Series May 1958 by C. K. Ingold. Pp. 52. The Weizmann Science Press of Israel. Jerusalem. 1959. (Presented by the publishers.) Boron fluoride and its compounds as catalysts in organic chemistry. A. V. Topchiev S. V. Zavgorodnii, and Ya. M. Paushkin. Translated from the Russian by J. T. Greaves. Pp. 326. Pergamon Press. London. 1959. The sequestration of metals theoretical considerations and practical applications.R. L. Smith. Pp. 256. Chapman and Hall Ltd. London. 1959. (Presented by the pub- lishers.) The molecular arrangement of cellulose nitrates. J. Trommei. Issued by the N. V. Koninklijke Nederlandsche Springstoffenfabrieken (Royal Dutch Explosive Manu- factories Ltd.) (Communication No. 15.) Pp. 24. N.V. SEPTEMBER 1959 Koninklijke Nederlandsche Springstoffenfabrieken. Am- sterdam. 1958. (Presented by the publishers.) Principles of organic chemistry. T. A. Geissman. Pp. 635. W. H. Freeman and Company. San Francisco. 1959. (Presented by the publishers.) Methoden der organischen Chemie (Houben-Weyl). 4th edn. Eugen Muller. Vol. I. Part 2. Allgemeine Labora- toriums.11. Pp. 1017 Georg Thieme. Stuttgart. 1959. Cahiers de synthkse organique mkthodes et tableaux d’application. J. Mathieu et A. Allais. Edited by L. Velluz. Volume 4. Pp. 272. Masson et Cie. Paris. 1958. (Presented by the publishers.) Die atherischen Ole. E. Gildemeister and Fr. Hoffmann. 4th edn. Wilhelm Treibs and K. Bournot. Vol. 5. Pp. 766. Akademie-Verlag. Berlin. 1959. Nitration of hydrocarbons and other organic com-pounds. A. V. Topchiev. Translated from the Russian by C. Matthews. Pp. 329. Pergamon Press. London. 1959. Chemie der Azofarbstoffe. H. Zollinger. Pp. 308. Birkhauser Verlag. Basle. 1958. Synthesis of beta-amino alpha beta-unsaturated and bis-(aminoaryl) sulphones thesis. M. Balasubramanian. Pp. 89. Annamalai University.Annamalainagar South Arcot Madras State. 1954. (Presented by the publishers.) Recent advances in the chemistry of cellulose and starch; edited by J. Honeyman. Based on lectures arranged by the Manchester College of Science and Tech- nology. Pp. 358. Heywood and Company Ltd. London. 1959. Chemistry of Heterocyclic Compounds. Vol. 12. Six-membered heterocyclic nitrogen compounds with three condensed rings. C. F. H. Allen G. M. Badger B. Graham G. A. Reynolds J. H. Richmond J. R. Thirtle J. A. Van Allan and C. V. Wilson. Pp. 624. Interscience Publishers Inc. New York. 1958. Heterocyclic chemistry an introduction. A. Albert. Pp. 424. The Athlone Press. London. 1959. (Presented by the publishers.) Syntheses and properties of some thiophene oligomers academic thesis.A. Bantjes. Pp. 68. Uitgeverij Excelsior. The Hague. 1959. (Presented anonymously.) Progress in biochemistry a report on biochemical problems and on biochemkal research since 1949. F. Haurowitz. Pp. 357. S. Karger. Basle. 1959. (Presented by Interscience Publishers Ltd.) Biochemistry and the central nervous system. H. Mcllwain. 2nd edn. Pp. 288. J. & A. Churchill Ltd. London. 1959. The viruses biochemical biological and biophysical properties. Edited by F. M. Burnet and W. M. Stanley. Vol. 3. Animal viruses. Pp. 428. Academic Press. New York. 1959. Comprehensive analytical chemistry. Edited by C. L. Wilson and D. W. Wilson. Volume IA. Classical analysis. Pp. 577. Elsevier Publishing Company.Amsterdam. 1959. Chemical Analysis. Vol. 3. 3rd edn. Colorimetric determination of traces of metals. E. B. Sandell. Pp. 1032. Interscience Publishers Inc. New York. 1959. Colorimetric methods of analysis including photo- metric methods. Volume IIA. 3rd edn. F. D. Snell C. T. Snell and C. A. Snell. Pp. 793. D. Van Nostrand Inc. Princeton New Jersey. 1959. The measurement of impurities in helium. Part 1. The katharometer as a continuous analyser for total im- purities. M. Wikins and J. D. Wilson. Issued by the United Kingdom Atomic Energy Authority Research Group. Pp. 14. (A.E.R.E. C/R 2808.) Atomic Energy Research Establishment. Harwell. 1959. The determination of oxygen and hydrogen in helium. W. R. Marsh. Issued by the United Kingdom Atomic Energy Authority Research Group.(A.E.R.E. C/M 377.) Pp.7. Atomic Energy Research Establishment. Harwell. 1959. Chemical and spectrographic analysis of magnesium and its alloys. A. Mayer and W. J. Price. Issued by Magnesium Elektron Limited. Additions and revisions (second series). (“Loose-leaf” pages.) Magnesium Elek- tron Ltd. Manchester. 1958. (Presented by the publishers.) Analytical method The determination of phosphorus in mixtures of sodium potassium and magnesium chlorides. E. Booth and A. Parker. Issued by the United Kingdom Atomic Energy Authority Research Group Chemistry Division. Pp. 2. (A.E.R.E. AM 4.) Atomic Energy Research Establishment. Harwell. 1959. An ultra-micro method for the estimation of the rare earths by complexometric titration.E. A. C. Crouch and I. G. Swainbank. Issued by the United Kingdom Atomic Energy Authority Research Group. Pp. 6. (A.E.R.E. C/R 2843.) Atomic Energy Research Establishment. Harwell. 1959. Methods of silicate analysis. H. Bennett and W. G. Hawley. Issued by the British Ceramic Research Associa-tion. Pp. 159. The British Ceramic Research Association. Stoke-on-Trent. 1958. Stoichiometry for chemical engineers. E. T. Williams and R. Curtis Johnson. Pp. 350. McGraw-Hill Book Company Inc. New York. 1958. Sourcebook on atomic energy. S. Glasstone. 2nd edn. Pp.641. D.V an Nostrand Company Inc. Princeton New Jersey. 1958. Nuclear power reactors. By J. K. Pickard. (Geneva Series on the Peaceful Uses of Atomic Energy.) Pp. 388.D. Van Nostrand Company Inc. Princeton New Jersey. 1957. A tentative index of the correspondence of the Honour- able Robert Boyle. R. E. W. Maddison. (Notes and Records of the Royal Society of London 1958 13 128-201 .) Royal Society. London. 1958. (Presented by the author.) Poggendorff ’s Biographisch-literarisches Handworter-buch der exakten Naturwissenschaften. Vol. VlIa Part 3 L-R. No. 5. Pp. 112. Akademie-Verlag. Berlin. 1959. Poggendorffs biographisch-literarisches Handworter-buch der exakten Naturwissenschaften. Vol. VIIa. Part 3 L-R. Part 7 & 8. Pp. 870. Akademie-Verlag. Berlin. 1959. Poggendorffs biographisch-literarisches Handworter-buch der exakten Naturwissenschaften. Vol. VIIa Part 3 L-R. No. 6. Pp. 1 12. Akademie-Verlag.Berlin. 1959. Landolt-Bornstein Zahlenwerte und Funktionen aus Physik Chemie Astronomie Geophysik und Technik. Vol. 11. Eigenschaften der Materie in ihren Aggregatzu- standen. Part 6. Elektrische Eigenschaften I. Pp. 1018. Springer-Verlag. Berlin. 1959. Gmelins Handbuch der anorganischen Chemie. Magnestische Werkstoffe. Zugleich 2. Erganzungsband zu Eisen. Teil D. System-nummer 59. Pp. 580. Verlag Chemie GmbH. Weinheim. 1959. Encyclopedia of chemical reactions. Compiled by C. A. Jacobson. Edited by C. A. Hampel. Vol. 8. Pp. 533. Reinhold Publishing Corporation. New York. 1959. May’s chemistry of synthetic drugs. 5th edn. G. Malcolm Dyson and P. May. Pp. 678. Longmans. London. 1959. (Presented by the publishers.) Handbook of toxicology.Volume I11 Insecticides a compendium by W. 0. Negherbon. Prepared under the direction of the Committee on the Handbook of Bio- logical Data Division of Biobgy and Agriculture the National Academy of Sciences the National Research Council. Pp. 854. W. B. Saunders Company. Philadelphia. 1959. Handbook of toxicology. Volume IV Tranquilizers compiled by Maxwell Gordon R. F. J. McCandless and S. W. Lipsman. Edited by Rudolph M. Grebe. Prepared under the direction of the Committee on the Handbook of Biological Data Division of Biology and Agriculture the National Academy of Sciences the National Research Council. h.120. W. B. Saunders ComDanv. Phila-delphia. 1959. Handbook of toxicolom. Volume V Funnicides com- piled by Everett F.Davii,-Barbara L. Tuma,and Lucy C. Lee. Edited by D. S. Dittmer. Prepared under the direc- tion of the Committee on the Handbook of Biological Data Division of Biology and Agriculture the National Academy of Sciences the National Research Council. Pp. 120. W. B. Saunders Company. Philadelphia. 1959. A symposium on the evaluation of drug toxicity held at Alderley Park Cheshire 1957. Edited by A. L. Walpole and A. Spinks. Sponsored by Imperial Chemical Industries (Pharmaceuticals Division). Pp. 138. J. & A. Churchill Ltd. London. 1958. Trabajos de la Tercera Reunion Tnternacional sobre Reactividad de 10s Solidos Madrid 1956. Vol. 11. Pp. 693. C. Bermejo. Madrid. 1958. (Presented by the publishers.) The structure and function of subcellular components a symposium held in London 1957.Organised and edited by E. M. Crook (Biochemical Society Symposia. No. 16.) Pp. 100. ‘CJniversity Press. Cambridge. 1959. (Presented by the Biochemical Society.) The physico-chemical properties of proteins with special reference to wheat proteins a symposium organised by the Food Group of the Society of Chemical Industry held in London 1957. (S.C.I. Monograph No. 6.) Pp. 92. Society of Chemical Industry. London. 1959. (Presented by the publishers.) Proceedings of the Third Conference on Carbon held at Buffalo New York 1957; sponsored by the University of Buffalo the National Science Foundation and the U.S. Office of Naval Research. Pp. 718. Symposium Publica- tions Division Pergamon Press. London.1959. Gas chromatography proceedings of the First Inter- national Symposium on Gas Chromatography held at the Kellogg Center Ann Arbor Michigan 1957; under the auspices of the Analysis Instrumentation Division of the Instrument Society of America. Edited by V. J. Coates H. J. Noebels and I. S. Fagerson. Pp. 323. Academic Press. New York. 1958. Glove boxes and shielded cells for handling radio- active materials a record of the proceedings of the sym- posium on glove box design and operation held at Harwell 1957. Edited by G. N. Walton et al. Sponsored by the United Kingdom Atomic Energy Authority. Pp. 5 15. Butterworths Scientific Publications. London. 1958. Proceedings of the Second United Nations Inter-national Conference on the Peaceful Uses of Atomic Energy held in Geneva 1958.Vol. 18. Waste treatment and environmental aspects of atomic energy. Pp. 624. United Nations. Geneva. 1958. Proceedings of the Second United Nations International Conference on the Peaceful Uses of Atomic Energy held in Geneva 1958. Vol. 19. The use of isotopes industrial use. Pp. 366. United Nations. Geneva. 1958. Proceedings of the Second United Nations Inter-national Conference on the Peaceful Uses of Atomic Energy held in Geneva 1958. Vol. 24. Isotopes in bio- chemistry and physiology. Part 1. Pp. 297. United Nations. Geneva. 1958. Proceedings of the Second United Nations Inter-national Conference on the Peaceful Uses of Atomic Energy held in Geneva 1958. Vol. 25. Isotopes in bio-chemistry and physiology.Part 2. Pp. 301. United Nations. Geneva. 1958. Proceedings of the Second United Nations Inter-national Conference on the Peaceful Uses of Atomic Energy held in Geneva 1958. Vol. 28. Basic chemistry in nuclear energy. Pp. 686. United Nations. Geneva. 1958. Proceedings of the Second United Nations Inter-national Cunference on the Peaceful Uses of Atomic Energy held in Geneva 1958. Vol. 29. Chemical effects of radiation. Pp. 475. United Nations. Geneva. 1958. Comptes rendus du Congrks International de Physique NuclCaire Paris 1958; present& par P. Gugenberger. Interactions nucltaires aux basses energies et structure des noyaux. Sous le patronage de I’Union Internationale de Physique Pure et Appliquee avec ie concours de I’U.N.E.S.C.0.et de la SociCtC FranCaise de Physique. Pp. 950. Dunod. Paris. 1959. (Presented by the pub- lishers.) A symposium on the chemical basis of development held at Baltimore 1958; sponsored by the McCollum- Pratt Institute of the Johns Hopkins University and the National Science Foundation. Edited by W. D. McElroy and B. Glass. Pp. 934. The Johns Hopkins Press. Balti-more. 1958. Seventh Symposium (International) on Combustion, held at London and Oxford 1958 and published for the Combustion Institute. Organised by the Institute of Fuel. Pp. 959. Butterworths Scientific Publications. London. 1959. Sulfur in proteins proceedings of a symposium held at Falmouth Massachusetts 1958 ;organised and edited by R. Benesch R. E. Benesch P.D. Boyer I. M. Klotz, W. R. Middlebrook and A. G. Szent-Gy8rgyi. Pp. 469. Academic Press. New York. 1959. Eleventh Pakistan Science Conference held at Karachi 1959. Abstracts of the section of Chemistry and Applied Chemistry. The role of agriculture in national economy general presidential address by Muhammad Afzal. A brief discussion on the development of minerals in the tribal areas of North West Pakistan presidential address given to the section of Chemistry and Applied Chemistry by Mirza Anwar Beg. Pp. 76. Pakistan Association for the Advancement of Science. Lahore. 1959. (Presented by he publishers.) Symposium on the Chemistry of Co-ordination Com- pounds held at Agra 1959; organised by the National Academy of Sciences India. Symposium handbook.India’s contribution to co-ordination chemistry by Arun K. Dey. Presidential address recent advances in the chemistry of co-ordination complexes by Priyadaranjan RPy. Recent researches on co-ordination compounds in Japan by Kazuo Yamasaki. Pp. 220. National Academy of Sciences India. Allahabad. 1959. (Presented by the publishers.) NEW JOURNALS Acta Biochimica Polonica from 1958 5. Izvestiia Akademii Nauk Armianskoi CCB Khim-icheskie Nauki from 1958 11. Clemio Stosowana from 1957 1. Doklady Akademiia Nauk Armianskoi CCP from 1956 22. Syntheses of Heterocyclic Compounds from 1956 1. Journal of Molecular Biology from 1959 1. Advances in Inoraanic Chemistrv and Radiochemistrv. .,I from 1959 1. 1 Planseebericht fur Palvermetallurnie.from 1959. 7. Progress in Industrial Microbiology; from 1959; 1. Electrochimica Acta from 1959 1. Journal of Less Common Metals from 1959 1.
ISSN:0369-8718
DOI:10.1039/PS9590000241
出版商:RSC
年代:1959
数据来源: RSC
|
|