287
Mar., 1955
COMMUNICATIONS TO THE EDITOR A COMPARISON O F SORPTION CHARACTERISTICS OF SOME MINERALS HAVING A LAYER STRUCTURE Sir: The report by Young and Healey’ showing that unactivated chrysotile adsorbed approximately twice as much water as nitrogen, expressed in units of surface area, is of interest to the science and technology of some minerals possessing a layer structure. It has recently been reported2 that the mineral tobermorite (4Ca0.5sio2*5H20) the probable binder of autoclaved concrete block, also adsorbed approximately twice as much water as nitrogen. Because tobermorite has a layer structure the large adsorption of water compared to that of nitrogen became of interest regarding the drying shrinkage of concrete. Theoretical considerations indicate that the shrinkage is not a capillary phenomenon. Berna13 suggested that tobermorite may undergo the swelling-shrinkage changes similarly as do some clays which also have a layer structure. Such volume changes in clays are easily demonstrated by X-ray results on clays before and after drying. A similar hypothesis on concrete shrinkage was studied in these laboratories but it could not be demonstrated2 that the water involved in volume change was inter-layer water. X-Ray results obtained subsequently4 show that phases closely related to tobermorite may undergo significant structural changes during in-vacuum drying and subsequent re-saturation, but it has not yet been shown how such structural changes may be due to the movement of water or what effect they produce on the properties of concrete. The present comments are being presented because similar sorption results on minerals having the layer structure have been interpreted differently by the writer, and Young and Healey. The sorption behavior of tobermorite (and antigorite as mentioned by these authors) shows that a portion of the water is accommodated differently than is nitrogen. Because these minerals generally occur as flat plate-like crystals explanations based on capillary phenomenon do not apply. Young and Healey pointed out that chrysotile occurs in the form o,f capillaries having a bore of apprgximately 150 A. and a diameter of about 350 A. They postulate formation of water plugs a t the ends of the capillaries which are permeable to polar substances, water and ammonia, but not to non-polar adsorbents as nitrogen. The water (1) G. J. Young and F. H. H e d e y , THISJOURNAL, 66, 881-884 (1954). ( 2 ) G . L. Kalousek, J. Amer. Concr. Inslitule, 2 6 , 233-248 (1953). ( 3 ) .J. D. Bernal, “Structures of Cement Hydration Compounds,” Proc. Srd International Symp. o n Chemistry o j Cemenls, p. 216-230 (1952). (4) G. L. Kalousek, “Toberrnorite and Related Phases in the Systeni CaO-SiOz-H-0,” J. Amer. Concr. Institute, acceyted for publication.
required to fill the pores would account for the much larger sorption of water than nitrogen. Results on properties of chrysotile and antigorite components of serpentine, now being prepared for publication, included density measurements on several kinds of chrysotile in form of relatively large “bundles” of fiber and also in highly micronized form; the samples were tested as received. Water and carbon tetrachloride were used as immersion media. The density results were the same within h i t of experimental error for both media and regardless of the degree of subdivision of the fiber. The spread in values amounted to about 2% but generally the differences were less than 1%. If the ends of the capillaries were plugged with water, the density results obtained with carbon tetrachloride, a non-polar medium, would be approximately 15 to 20% lower than in water. These results, and others on the crystal habits of antigorite and chrysotile which will be submitted for publication, suggest that water plugs probably do not form in the tube-shaped crystals of chrysotile. Antigorite and chrysotile are either the same in compositions and atomic structure or very closely related. It would seem, therefore, that the sorption characteristic should be the same for both. The mechanism of adsorption-desorption of these two minerals may be the same as that for tobermorite, presumably that of movement of structural water. As mentioned, however, this hypothesis remains to be resolved experimentally. It would appear that careful studies on sorption characteristics and surface energy changes such as those being conducted at the Surface Chemistry Laboratories a t Lehigh University, should provide an unequivocal proof on the manner in which the excess water is adsorbed by the layer structure minerals. Such basic information will be of great help to concrete technology, to mention only one field of application. OWENS-ILLINOIS COMPANY 1700 NORTHWEBTWOOD G. L. KALOUSEK TOLEDO, OHIO RECEIVED JANUARY 27, 1955
FORMATION OF DIBORANE DURING T H E SLOW OXIDATION OF PENTABORANE Sir: When dry air was admitted slowly to a gas absorption cell containing pure pentaborane at 25” and a pressure of 50 mm., partial oxidation took place and diborane, hydrogen and a white solid were formed. The reaction was conducted in a cell 10 cm. in length, with a volume of 150 ml. Absorption measurements were made with a Perkin-Elmer Model 21 double-beam infrared recording spectrometer. The progressive disappearance of pentaborane and the simultaneous accumulatioii of diborane were followed by infra-
COMMUNICATIONS TO THE EDITOR
288
4
8 W~VE LEmerm
m
12
I4
1
WCRONSI.
Fig. 1.-Infrared spectra of pentaborane and of the products formed during its slow oxidation.
red absorption measurements, made possible because of the differing spectra of the two hydrides.' At the end of the reaction, after six hours, when the total pressure approached 1 atm., the partial pressure of diborane was approximately 10 mm. (1) L. V. McCarty, G. C. Smith and R. S. McDonald, Anal. Chem., 26, 1027 (1954).
VOl. 59
The initial absorption spectrum of the pentaborane and the final absorption spectrum of the reaction mixture after the reaction was complete are shown in Fig. 1. It is evident that initial reaction of oxygen with pentaborane results in incomplete oxidation. Unstable residues of the partially oxidized pentaborane molecules may be the source of the diborane formed. At the temperature and pressure at which the reaction took place, diborane-oxygen mixtures are well below the spontaneous explosion limit.2Ba The spectrum of the white solid reaction product, which absorbs strongly a t 7.9 and 13.8 microns, differs from that of boric acid and the higher boron hydrides. The solid is presumed to be HBOz or Bz03. DEPARTMENT OF CHEMISTRY HARRY C. BADEN POLYTECHNIC STEPHEN E. WIBERLEY RENSSEIAER INSTITUTE WALTER H. BAUER TROY,N. Y. RECEIVED FEBRUARY 5, 1955 (2) W. Rot11 and W. H. Bauer, Fifth Svmposium on Combuslion, John Wiley and Sons, Inc., following publication. ( 3 ) F. P. Price, J . Am. Ckem. Soc., 72, 53G1 (1950).
I.
*
