ADDRESS

BY

PROFESSOR CHARLES E. MUNROE,

VICE PRESIDENT, SECTION C.

SOME PHASES IN THE PROGRESS OF CHEMISTRY.

I am deeply impressed with the honor which you have conferred upon me in selecting me to preside over your section during the present session; but, as I look back upon the eminent men and distinguished chemists who have preceded me in this office, and around upon those who form this section, I feel unequal to the responsibility which the position entails and the obligations which it imposes, and especially so as regards your annual address. For where one is actively and constantly engaged in a variety of pursuits of a highly technical nature, it is difficult to possess one's self of but a very partial knowledge of the advances made in chemistry, since in these busy modern days the science has developed so many special phases, and has shown in many directions so marked a physical and mathematical tendency; while, at the same time, its investigators have continued to add an enormous annual increase to the bodies with which chemistry is to deal. Hence, in reviewing the progress made in chemistry, I must content myself with a glance at some few phases and leave many as important ones untouched.

Since the isolation and recognition of oxygen by Priestley, the search for new elements, like that for new heavenly bodies, has

formed for many a most entertaining pursuit, the facilities for which have been much increased by the use of the spectroscope and the delicate spectroscopic methods as developed by Bunsen and Kirchoff, and perfected by Nobert, Rutherford and Rowland in construction, and by Crookes, Boisbaudron, Liveing and Dewar and others in methods, and the result has been to extend continually the list of bodies which are grouped under this head. The announcement of new discoveries during the last ten years has been especially large. According to a recent statement of Dr. Bolton,1 who has kept a careful record of these announcements, over seventy bodies have been added to the list during this time, though it will be observed that many of these so-called elements are obtained by the resolution of others which are included in the above-mentioned number. The largest number added by any observer has resulted from the joint labors of Krüss and Nilson2 on the absorption spectra of the rare earths and reaches to over twenty. Should these discoveries be verified and the elementary character of the substances, as we now use the term elementary, be established, the possible number of compounds which would result is something enormous; but, judging from experience, few of them are likely to survive a very searching criticism. Still two in this list, scandium and germanium, have already passed the tests, and the latter, which was discovered by Winkler in 1886, has been accepted as the missing element in Mendeléeff's scheme, whose existence and properties he predicted under the name of ekasilicon.

Among the fundamental constants of chemistry there are none which occupy a more important place than the weights of the atoms of the elementary substances; and it is only natural that, from the time when the present chemical elements were recognized as such, and especially since the adoption of the atomic hypothesis as enunciated by Dalton, strenuous efforts should have been made to determine this constant for each element with all the precision of which chemical art permits. Apart from the evident necessity and advantage of knowing this constant for purposes of analysis and in chemical processes in general, an added zest was given to the pursuit by the hypothesis of evolution as developed by Prout, and extended and modified by Dobereiner, Dumas, Cooke, Pettenkofer, Odling and Gladstone in their demonstrations of the numerical

1 J. Anal. Chem., 2, 282-283; 1888.

2 Ber. d. chem. Ges., 20, 2134; 1887.

relations existing between the atomic weights of elements belonging to the same natural group; while, more recently, a deeper interest has been imparted through the discussions by Newlands, Mendeléeff, Lothar Meyer, Carnelley, Mills, Reynolds and others of the data already collected which leads to the conclusion that the properties of the elements are functions of their atomic weights and that the various weights are related according to some law of nature. Thanks to the labors of Becker, Clarke, Lothar Meyer and Seubert, the literature of the subject of the determinations of the atomic weights has been carefully reviewed, the processes collated and the data discussed so that the whole matter is now readily accessible to any one who desires to engage in further research in this field.

Since the unit weight of hydrogen is taken as the standard for comparison, while the determination of the atomic weights of a large number of the elements has been made only through the intervention of oxygen, the ratio of the atomic weights of these two elements is the most important one to be determined, for any error which may occur here will be magnified when repeated through a moderate series of other ratios. Three methods were employed by the earlier investigators for fixing this constant: (1) through the synthesis of water, which was effected by passing hydrogen gas over hot copper oxide; (2) through the exact determination of the relative densities of the two gases; (3) by weighing the quantity of water formed through the direct union of a known volume of hydrogen with oxygen. The first method was employed by Dulong and Berzelius, by Dumas, and by Erdmann and Marchand; the second by Dumas and Boussingault, and by Regnault; and the third by J. Thomsen. Of these researches, that of Dumas by the first method is by far the most important, and it constitutes one of the most memorable investigations in the history of chemistry. In this he burnt an undetermined amount of hydrogen by means of copper oxide, the amount of oxygen consumed being determined by the loss of weight of the tube containing the copper oxide, and the water formed being collected and weighed directly. The greatest care was taken to insure the purity of all the materials used, every known experimental means were employed to secure accuracy, and all the corrections which could be conceived of were applied to the results, while the experiments were carried out on a very large scale, the amount of water produced being in some

cases as high as seventy grams. As a result, the value of the atomic weight of oxygen was, as the mean of nineteen determinations, found to be 15.9607 with a probable error of .007.

Notwithstanding the great intelligence and skill displayed by Dumas in the devising and execution of this research, the process is open to serious criticism, inasmuch as that the weight of the hydrogen in the water obtained is estimated by difference and hence any errors in the process are accumulated in the value obtained for the hydrogen and as this is the lighter body such errors may become very appreciable. Dumas himself recognized this, for he says: "Of all analyses presented to the chemist, that of water is the one which offers the greatest uncertainty. Indeed, one part of hydrogen unites with eight parts of oxygen to form water, and nothing would be more exact than the analysis of water if we could weigh the hydrogen as well as the water which results from its combustion. But the experiment is not possible under this form. We are obliged to weigh the water formed, and the oxygen which was consumed in producing it, and to determine by difference the weight of the hydrogen which has entered into combination. Thus an error of in the weight of the water, or of do in the weight of the oxygen, is equivalent to an error of or in the weight of the hydrogen. Let these two errors be in the same direction and the total error will amount to."

Dr. J. P. Cooke1 and T. W. Richards have, however, sought to solve this problem and have contrived a very ingenious apparatus by which the weight of hydrogen burnt and of water obtained is known, while the weight of the oxygen contained is estimated by difference. For this purpose, pure, dry hydrogen was first collected in a large, glass balloon, which had previously been exhausted and weighed, and then carefully weighed again, the container being counterpoised by a second balloon of similar material and of exactly the same external volume. The balloon used had a capacity of 4961.5 cubic centimetres, weighed 570.5 grams and held about .42 grams of hydrogen. The method of compensation adopted in the weighing of the large globe was found to be so accurate that under good conditions the weight of the globe, when filled, did not vary more than one-tenth of a milligram through large changes of temperature and pressure. The filled and weighed

1 Amer. Chem. Journ., 10, 81-110, 1888.

balloon was then inserted between a set of combustion furnaces so arranged that pure, dry atmospheric nitrogen should be drawn into the balloon so as to sweep the hydrogen before it. The hydrogen was thus passed over ignited copper oxide and the water produced collected in suitable condensers. Finally, pure, dry air was drawn through the whole apparatus, so that at the end of the operation the apparatus and contents were in the same condition as at the start except that the hydrogen was all in the form of water. The operation was conducted with all the precision possible, and the mean of the fifteen determinations gave 15.953 ± 0.0017 as the atomic weight of oxygen.

Dr. Scott has repeated the experiments of Gay-Lussac and Humboldt for ascertaining the composition of water by volume, using an improved eudiometer in the place of the Volta eudiometer used by them, and besides seeking to attain greater accuracy by preparing purer gases, using larger volumes, measuring both gases in the same vessel and by analyzing the residue after each explosion and determining the impurities present in each experiment. The apparatus used was entirely of glass with the exception of the rubber connections to the receivers employed for holding the mercury used in exhausting the apparatus. The oxygen employed was produced from potassium chlorate or mercuric oxide and the hydrogen by electrolysis. The largest amount of impurity found present in any one of the twenty-one experiments was zz parts while it fell in another case to 7 parts. As a result of these experiments Scott finds the most probable ratio of the gases by volume in water to be H: 0=1.994:1. Hence taking the density of oxygen referred to hydrogen as 15.9627, he arrives at an atomic weight for oxygen of 16.01. A more recent determination of this ratio by Scott under improved conditions gave H:0= 1.9965: 1.

In 1882 Lord Rayleigh, animated by the same motive as Dr. Cooke, namely, the desire to examine whether the relative atomic weights of hydrogen and oxygen really deviated from the simple ratio of 1:16 as demanded by Prout's hypothesis, planned an investigation for determining the relative densities of these two gases, the results of which he held when combined with the determinations of the relative atomic volumes, measured by eudiometric methods such as Mr. Scott employed, ought to lead to a true

1 Chem. News, 56, 173-175, 1887.