July 7, 1892] 



NATURE 



231 



discoveries made only within the last few years by following up 

 Liehig's idea. 



Carbonic oxide, composed of one atom of carbon and one atom 

 of oxygen, is a colourless gas, without taste or sm.ell, which I 

 have here in this jar. It burns with a blue flame. When it 

 acts as a radical combining with other bodies, we term it car- 

 bonyl, and its compounds with other elements or radicals are 

 termed carbonyls. 



Liebig defined a radical as a compound having the charac- 

 teristics of a simple body, which would combine with, replace, 

 and be replaced by simple bodies. In more modern times 

 a. radical has been defined as an unsatiated body. I am 

 of course speaking of chemical radicals. If we look at it 

 from the modern point of view, carbonyl should be the very 

 model of a radical, because only half of the four valencies of the 

 carbon atom are satiated, the other two remaining free. Car- 

 bonic oxide should even be a most violent radical, because, 

 amongst all organic radicals, it is the only one we know to exist 

 in the atomic or free state. All the other organic radicals, even 

 such typical ones as cyanogen and acetylene, are known to us as 

 molecules composed of two atoms of the radical, so that the 

 •cyanogen gas and acetylene gas we know should more properly 

 be called di-cyanogen and di-acetylene ; they consist of two 

 atoms of the radical cyanogen or of the radical acetylene, the 

 free valencies or combining powers of which satiate or neutralize 

 €ach other. On the other hand, carbonic oxide gas, as I stated 

 before, makes the sole exception. Its molecule contains only 

 -one atom of carbonyl moving about with its free valencies un- 

 fettered by a second atom. For all that, carbonic oxide is by 

 no means a violent body, but the very reverse, and instead of 

 being ready to attack with its two free valencies anything coming 

 in its way, until very recently we only knew it to interact and 

 to combine with substances possessing themselves extreme 

 attacking pov\ ers, such as chlorine and potassium. Although 

 Liebig had so long ago proclaimed it as a radical, the chemical 

 world was startled when, two years ago, I announced in a paper 

 I communicated to the Chemical Society in conjunction with 

 Drs. Langer and Quincke, that carbonic oxide combines at ordi- 

 nary temperature with so inactive an element as nickel, and 

 forms a well-defined compound of very peculiar properties. 



The fact that carbonic oxide does not possess the chemical 

 activity one would suppose in a radical composed of single 

 atoms may, I believe, be explained by assuming that the two 

 valencies of carbon which are not combined with oxygen do 

 satiate or neutralize each other. Everybody admits that the 

 valencies of two different carbon atoms, which are all considered 

 of equal value, can neutralize each other. I see, therefore, no 

 reason to question the possibility of two valencies of the same 

 carbon atom neutralizing each other. On this assumption car- 

 bonic oxide may be looked upon as a self-satisfied body — one 

 which keeps in check its free affinities within itself. 



You have here the typical carbon radicals containing one 

 atom of that element, acetylene, . methylene, methyl, cyanogen, 

 and carbonyl. In the second column you have these substances 

 as they are known to us in the free state. You see the carbonyl 

 is the only one which exists in the free state as a single atom, 

 while all the others only exist as molecules, composed of two 

 atoms the free valencies of which neutralize each other. The 

 carbonyl I have represented in the last formula, with the two 

 valencies not combined with oxygen neutralizing each other, so 

 that in this way it also becomes a satiated body. I will try to 

 make this still plainer to you by means of the models I have 

 before me. 



The paper published by Liebig in 1834, from which I have 

 already quoted, was entitled '* On the Action of Carbonic Oxide 

 on Potassium." In it Liebig fully described the preparation 

 and properties of the first metallic carbonyl known — a compound 

 of pota>sium and carbonic oxide. Liebig obtained this com- 

 pound by the direct action of carbonic oxide upon potassium at a 

 temperature of 80° C, and proved it to be identical with a sub- 

 stance which had been previously obtained as a very disagreeable 

 by-product of the n anufacture of potassium from potash and 

 carbon by Brunner's method. It forms a grey powder which is 

 not volaUle, and which on treatment with water yields a red 

 solution, gradually turning yellow in contact with air, and from 

 which on evaporation a yellow salt is obtained, called potassium 

 croconate, on account of its colour. Liebig showed this salt to 

 consist of two atoms of potassium, five of carbon, and five of 

 oxygen, and not to contain any hydrogen, as had previously 

 lieen supposed. 



NO. I 1 84, VOL. 46] 



Since the publication of Liebig's paper, potassium carbonyl 

 has been studied by numerous investigators, amongst whom Sir 

 Benjamin Brodie deserves particular mention ; but it has been 

 reserved to Nielzki and Benkiser to determine finally in the year 

 1885, by a series of brilliant investigations, its exact constitu- 

 tion, and its place in the edifice of chemistry. They have proved 

 that it has the formula KgCgOg ; that the six carbons in this 

 compound are linked together in the form of a benzole ring ; 

 that, in fact, the compound is hexhydro.xylbenzole, in which all 

 the hydrogen is replaced by potassium. By simple treatment 

 with an acid it can be converted into the hexhydroxylbenzole, 

 and from this substance it is possible to produce, by a series of 

 reactions well known to organic chemists, the whole wide range 

 of the benzole compounds. The body which Liebig obtained 

 by the direct action of carbonic oxide on potassium has thus 

 enabled us to prepare synthetically in a very simple way from 

 purely inorganic substances — to wit, from potash and carbon, or 

 if we like even from potash and iron — the whole series of those 

 most important and interesting compounds called aromatic 

 compounds, including all the coal-tar colours, which have fur- 

 nished us with an undreamt-of variety of innumerable hues and 

 shades of colour, as well as many new substances of great value 

 to suffering humanity as medicines. Surely a startling result, 

 which alone would have fully justified Liebig's prediction of 1834 ! 

 Speaking of coal-tar colours, everybody will be reminded of 

 the great loss the scientific world has recently sustained by the 

 death of August Wilhelm Hofmann, their first discoverer, 

 Liebig's greatest pupil. Hofmann will ever be remembered in 

 this Institution, where he so often delighted the audience by his 

 lucid lectures, and in whose welfare he took the greatest in- 

 terest, of which he gave us a fresh proof only last year, in the 

 charming letter he wrote on the occasion of his election as an 

 honorary member. 



Looking back upon the wonderful outcome of Liebig's idea 

 I have referred to, it seems surprising indeed that others should 

 not have followed up his work by attempting to obtain other 

 metallic carbonyls. 



A very few experiments were made with other alkaline metals : 

 sodium, otherwise resembling potassium so closely, has been 

 shown not to combine with carbonic oxide ; lithium and cjEsium 

 are stated to behave similarly to potassium. But metals of other 

 groups received little or no attention. The very important r$U 

 which carbonic oxide plays in the manufacture of iron did lead 

 a number of metallurgists (among whom Sir Lowthian Bell 

 and Dr. Alder Wright are the most prominent) to study its 

 action upon metallic iron and other heavy metals, including 

 nickel and cobalt at high temperatures. They proved that these 

 metals have the property to split up carbonic oxide into carbon 

 and carbonic acid at a low red heat, a result of great importance, 

 which threw a new light upon the chemistry of the blast furnace. 

 None of these investigators, however, turned their attention to 

 obtaining compounds of these metals with carbonic oxide, and, 

 owing to the high temperature and the other conditions under 

 which they worked, the existence of such compounds could not 

 come under their observation. In order to obtain these com- 

 pounds, very special conditions must be observed, which are 

 fully described in the papers I have published during the last 

 two years in conjunction with Dr. Langer and Dr. Quincke. 



The metals must be prepared with great care, so as to obtain 

 them in an extremely fine state of division, and must be treated 

 with carbonic oxide at a low temperature. The best results are 

 obtained when the oxalate of the" metal is heated in a current of 

 hydrogen at the lowest temperature at which its reduction to the 

 metallic state is possible. 1 have in the tube before me metallic 

 nickel prepared in this way, and over which a slow current of 

 carbonic oxide is now passing ; the carbonic oxide before enter- 

 ing the tube burns, as you see, wiih a blue non-luminous flame. 

 After passing over the nickel it burns with a highly luminous 

 flame, which is due to the separation of metallic nickel from the 

 nickel carbonyl formed in the tube, which is heated to incandes- 

 cence in the flame. (In passing the gas issuing from our lube 

 through a glass tube heated to about 200°, we obtain a metallic 

 mirror of pure nickel, because at this temperature the nickel 

 carbonyl is again completely resolved into its components, nickel 

 and carbonic oxide. If we pass the gas through a freezing mix- 

 ture, you will observe that a colourless liquid is condensed, of 

 which I have a larger quantity standing in this tube. This 

 liquid iormed is pure nickel carbonyl, and has the formula 

 Ni(CO),. 



If cooled to — 25° C, it solidifies, forming needle-shaped 



