Measurements of vapor pressures and boiling points, over the range 48 to 780 millimeters of mercury, and above about 11 0 C, were made on 60 purified hydrocarbons. The apparatus consisted of an electricall y heated boiler, a vapor space with a vertical reentrant tube containing a platinum thermometer having a resistance of 25 ohms, and a condenser. Measurements of the temperat ure of the liquid-vapor equilibrium were made at 20 fixed pressures maintained automaticall y. The values of the fixed pressures were determined by calibration of the apparatus with water using the vapor pressure-temperature tables prepared at this Burcau.The experimental data on the hydrocarbons were correlated, the method of least squares being u ed, w ith t he three-constant Antoine equation for vapor pressures, log P = A -B/ (C+O or I= B/(A-log P )-C. Experimental data, together with the values of the three constants of the Antoine equation applicabl e over the range of measurement, are reported for 60 API-NBS hydrocarbons, including 17 paraffins, 14 alkyJcyclopentanes! 8 aJkylcyclohexa nes, and 21 a lkylbenzenes.

mum values of M o (eq 29 ) assuming minimum and maximum values for both R m ax and~. The range of ,~ may be taken as 0.7 to 0.9. Compu te the corl'esponding valu es for J.I. M o and mark out the vertical band corresponding to these limits in figure 1. 5. For the m aximum and minimum limits assumed for both L and ~, compute the corresponding limits for L o (eq 28 ). Compute the corresponding values ·of QLo/t:.p and mark out the horizontal band C01'-l'esponding to these limits in figure 1. Steps (4) and (5) result in a design rectangle on figure 1 within which a solution is possihle.6. Further limit this design r ectangle by excluding regions of figure 1 representing greater and lesser area A (really J.l.A ) than desired.7. For gas flow, compute the maximum tolerable value of the coefficient of the Knudsen term band the corresponding minimum val ue of fib er diameter d. Exclude regions of figure 1 r epresenting smaller values of d (really d! p,). One may then choose design parameters corresponding to any point in the design region that has not been excluded.8. When the flowm eter is built and tested, adjustment of the resistivity can then be made by the principal techniqueiof changing the weight of glass wool used. Journal of Research of the Nationa l Bureau of StandardsFinancial support for this investigation was provided by the Office of Naval Research under a project on Basic Instrumentation of Scientific R esearch. Grateful acknowledgement is also due W . A. Wildhack, at ~vhose suggestion and under whose sup ervision the development of the glass wool flowm eter was carri ed on. D e nsity (at 20°, 25°, and 30° C), refractive index (at se ven wav elengths at 20°, 25°, and 30° C), vapor pressure, and boiling point (from 48 to 778 mm Hg) of 16 h~ghly purifi ed samp les of hydrocarbons of the API-N BS series wer e m easured for 8 monoolefin (l-a lkene), 6 pe ntadiene, and 2 cyclomonoolefin hydrocarbons.The data on r efractive index were adju sted by m eans of modified Cau ch y and Hartmann equations, and values of t he constants are given for each compound.The data on vapor press ure were adjusted by mea ns of the method of lea st squares and the three-constant Antoine equation. The values of th e constants are given for each compound .Values were calcula ted for the specfic dispersions,As a cooperative investigation of the National Bureau ' of Standards, the U. S. Office of Rubber R eserve, and the American Petroleum Institute R esearch Proj ect 6, measurem ents of density, refractive index, vapor pressure, and boiling point wer e made on highly purified samples of eight monoolefin (I-alkene), six pentadiene, and two cyclomonoolefin hydrocarbons of the API-N BS series.The compounds m easured were mad e available

This report gives the results of the purification and measurements of r efractive index (nD at 20° and 25° C), density (at 20° and 25° C), boiling point and pressure coeffi cient of t he boiling point (at 1 atm) and, except for four of the compounds, tije freezing point, t ogether wit h the calculated amount of impurity, of samples of 51 hydrocarbons, including 29 paraffins, 4 alkylcyclopentanes, 10 alkylcyclohexanes, and 8 alkylbenzenes.

Measurements of the refractive index at 20°, 25°, and 30° C and at se ven wavelengths, from 6678.1 to 4358.3 Angstrom units , were made on 60 purified hydrocarbons of t he API-NBS series. The apparatus consisted of a precision Abbe-type refractomcter, calibrated by m eans of NBS Standard Samples of hydrocarbons certified with respect to refractive index at thc th ree tcmperatures and seven wavelengths uscd. The light SOU l'ces used were a mercury arc, a sodium vapor lamp, and h ydrogen and helium discharge tubes, w ith suitable filL er . A constant t empe rature bath maintained the te mpe rature of the refractomete r prism s constant within ± 0.02° C.The experimental daLa on the hydrocarbons were adjusted by m ean s of a four constan t Hartmann equation , nA=n ",+C/(A-A*) 1.6, and a two constant modified Cauch y equation , fln, =a+ b/ X2• (fln)., t he change in t he refracti ve index a t t he wavele ngth, A, when the t e mpem t ure is chan ged from 20° to 25° C or 25° t o 30° C, is used instead of n, of the Cauchy eqllation).The compu ted refractive index a L 20°, 25°, an I 30° C, t oge Lher with the values of t h e specific di spers ions, and the constants of the HarLmann and m odified Cauchy equat ions, applicable over the ran ge of m ea l11'e ment, a re rep or ted for 60 API-NBS h yd rocarbo ns, comprising 17 paraffins, 14 alkylcyclop entanes, 8 alkylcyclohexanes, and 21 alky lbenze nes.

A simple method is described for determining the amount of aromatic hydrocarbons in a mixture of hydrocarbons, as in the gasoline fraction of petroleum. The mixture to be analyzed is filtered through a column of solid adsorbent. An aromatic-free filtrate is obtained which contains the paraffin, naphthene, or olefin hydrocarbon which was associated in the original solution with the quanti ty of aromatic hydrocarbon which has been adsorbed. The concentration of an aromatic hydrocarbon in an unknown solu t ion is d et ermined by m eans of a calibration curve, established from experiments on known solutions which show the amount of aromatic-free fil trate produced by the standard adsorbent from solutions of various concentrations of the aromatic hydrocarbon. R esults of experiments are given for several concentrations of eight binary solutions of an aromatic h ydrocarbon with a paraffin or naphthene h ydrocarbo n and for three concentrations of a solution consisting of an aromatic hydrocarbon with a paraffin and an olefin. These experiments show that, if the temperature is controlled to wit.hin 1 0 C, the amount of aromatic hydrocarbon can be determined with an accuracy corresponding to 0.10 or less in the percentage by volume. A general procedure is given for determining the aromatic hydrocarbons in a "straight-run " gasoline and in a gasoline containing oletins.