2 THE ANALYST. A N A U T O M A T I C B U R E T T E . BY EL DROOP RICHMOND. (Read at the Meeting, June 1, 1898.) SOME years ago it was the custom in my laboratory to measure off the sulphuric acid for the Leffmann-Beam process by means of a 100 C.C. burette, which was filled by means of a siphon with a tap to it ; the acid ran in at the top of the burette, and the operator had to stand by the burette till the acid reached the zero mark, and then to turn off the tap. To prevent the possibility of accidental overflow of the acid, it occurred to me to L Fro. 1. F ~ G . 2. * -I - - -3 -h F I G . 3. F-G. 4. fill the burette from the bottom, using a three-way tap, and to cork up the top with acork carrying a tube of diameter so small that the volunie of the acid in the tube could be neglected (Fig.1). A trial convinced me that the method was slow, owing to the time that sul- phuric acid took to run out of t h e narrow tube, and a little consideration revealed the fact that the error was much larger than was at first glance apparent. The following demonstration will show the nature of the errors intro- duced. The level of the acid a de- pends not only upon the level of the end of the tube, but on the pressure on the air in the space above ; if /I, and h1 represent the heights to which the acid rises when the stock-bottle is full and empty respectively, the volume of air in the space will vary in the proportion ofTHE ANALYST. 3 760 to x -____~__--- 760 760+ahx - 13.6 d 760 + n,h, x 1s a xx- where x = volume of air at 760 mm. pressure, d = density of liquid filling tube ; taking - - FIG.5. this it is forced x as 10 c.c., the distance from h to hl as 250 mm., and d as 1.82, which were the approxjmate values of the trial apparatus, a variation of nearly 0-5 c.c between the volume delivered when the stock-bottle is full and nearly empty is found ; this variation is much greater than that due to the actual volume of acid in the'tube, which was less than 0.2 C.C. ; both variations go in the same direction, and the combined error quite unfits this apparatus for any- thing more than rough work. A fairly obvious modification was to place the air-chamber above, instead of below, the effective zero point ; this was accomplished by drawing out the end of the burette to a narrow tube, and en- closing this in an air-chamber.The following diagrams will show two forms of accomplishing this end. In Fig. 2 the chamber and an exit-tube for the excess of liquid are made all in one piece ( A form); in Fig. 3 the exit-tube passes through a cork (B form). A further modification, which was not so accurate, but was convenient where great accuracy was not desired (e.g., the measurement of sulphuric acid for the Leffmann- Beam process, for which I have used it for the last four years), is shown at Fig. 4 (C formj. This enables an ordinary burette to be used, and, provided that the lower air-space is kept as small as possible, is quite accurate enough, and has the advantage that the zero point can be altered if desired, though it must be borne in mind that an alteration of the zero point will cause an alteration in the extent of the error, Of these forms B is pre- ferable, and is capable of greater accuracy than C, and is not so liable to break as A .In the case of corrosive liquids the connection between the stock-bottle and the burette is made by a siphon and the well- known air-trap, which enables an indiarubber connection to be used, and diminishes the risk of breakage. The disgram (Fig. 5 ) shows a burette (C form) with all connections ; after a, measurement has been performed the tap is turned half round, and the burette is automaticdly filled to the zero point ready for use when required. The liquid siphons over from the stock-bottle, fills the burette to the top, and then runs over into the air-chamber ; from UD bv the Dressure of the incorninn liauid UD the ascending tube till4 THE ANALYST. the level is equal to that in the stock-bottle; the zero is quite automatically adjusted and requires no attention. When the liquid is run out, air enters through the ascending tube and bubbles through the liquid in the air-chamber. When the burette is again filled the liquid in the air-chamber is forced back into the bottle.