The Preparation of Carbon Monosulphide and Hydrocyanic Acid by the Use of Activated Carbon Citation Vesper, Howard Gockley (1922) The Preparation of Carbon Monosulphide and Hydrocyanic Acid by the Use of Activated Carbon. Bachelor's thesis, California Institute of Technology. doi:10.7907/HXNR-JN80. https://resolver.caltech.edu/CaltechETD:etd-07162004-142151 Abstract No Abstract. Item Type: Thesis (Bachelor's thesis) Subject Keywords: Chemical Engineering Degree Grantor: California Institute of Technology Division: Chemistry and Chemical Engineering Major Option: Chemical Engineering Thesis Availability: Public (worldwide access) Research Advisor(s): Unknown, Unknown Thesis Committee: None, None Defense Date: 1 January 1922 Record Number: CaltechETD:etd-07162004-142151 Persistent URL: https://resolver.caltech.edu/CaltechETD:etd-07162004-142151 DOI: 10.7907/HXNR-JN80 Default Usage Policy: No commercial reproduction, distribution, display or performance rights in this work are provided. ID Code: 2902 Collection: CaltechTHESIS Deposited By: Imported from ETD-db Deposited On: 22 Jul 2004 Last Modified: 03 Oct 2019 23:21 Thesis Files Preview PDF (Vesper_hg_1922.pdf) - Final Version See Usage Policy. 1MB Repository Staff Only: item control page

THE PREPARATION of CARBON MONOSULPHIDE AND HYDROCYANIC ACID by the use of ACTIVATED CARBON Thesis by Howard G. Vesper In partial fulfillment of the requirements for the degree of Bachelor of Science in Chemical Engineering CALIFORNIA INSiITUTE OF TECHNOLOGY Pasadena, California 1loze (Page 1) INTRODUCTION During the late war a great deal of work was done in this country on the development of an "active" form of charcoal for use in gas masks as an adsorbent for certain toxic gases. A large amount of this work was done in the laboratories of the National Carbon Company, with the result of a large scale method for producing from whatever source of C-bearing material might be cheapest or most desirable, an amorphous form of carbon called "activated carbon". Active carbon is formed whenever carbon is deposit- ed at a relatively Low temperature by a chemical or ther- mal decomposition of C-bearing materials, generally below 600~700 degrees ¢. A Similar decomposition carried on under higher temperature conditions gives another form of carbon called “inactive carbon" because of its sta- bility toward oxidizing agents and its low specific ad- sorptive power. The first product occuring in the low temperature distillation of carboniferous materials is a stabilized complex of hydrocarbons adsorbed on a base of active carbon. Active carbon is obtained from this by freeing the hydrocarbons, which may be accomplished in two dif- ferent ways: a) by a process of controlled distillation, and 1 N. Ke Chaney, Trans. Em.ELectrochem.Soc. ,36, pweprint (1919) (Page 2) b) by a process of differential oxidation. The latter gives the most satisfactory results and may be applied to a wide variety of primary carbons. The active carbon used in the work described by this paper was selected from a sample furnished by the U. S. Fixed Nitrogen Research Laboratory and Was ac- tivated from a cocoanut shell base. It was thought that this new form might have special reduction effects that would not be found in an ordinary wood charcoal reduc- tion, and hence its effect on two substances was tried. (1) croS, =2cs ) ) (Ideal) (2) C+NHg = HON+H, ) The value of the product in (1) is more theoretical than practical, because of the fact that it has never been prepared in quantity. However, its nature would lead one to surmise that it might find application in gas war- - fare as well as in medical and fumigation work if it could be produced on a commercial scale. ‘The value of HCN from (2), however, is distinct, particularly in Southern California, where a large mar- ket for the liquified form of the gas for use in citrus tree fumigation has grown up in the past few years. The HCN produced by the treatment of MaCN with HpS0, has a market value of about 75¢ per lb. If large enough yields at reasonably low temperatures could be secured by the reduction of NH, the relatively low cost of the (Page 3) two raw materials and the absence of any waste product (there is a good market for He) such as impure NaoSO,4 would furnish a more economical method of HCN production. (‘Page 4) PART A-THE REDUCTION OF CARBON DISULPHIDE BY ACTIVE CARBON I - PREVIOUS WORK. A number of attempts to prepare CS have been made in the last 65 years. The record of this work is very unusual in that it is simply a succession of revorts of supposedly successful attempts to make CS, followed a few years later by other work which disproves the previous assertions. In 1857 Baudrimont= » @ Frenchman, did the first work on CS. He claimed that he produced the gas in a number of ways, chief among which were the decomposition of CSp at high temperatures either by leading it as a gas over glowing Pt or pumice or by treating with hot Ho eas, and the decomposition of CS» vapors by wood charcoal and all kinds of animal charcoal when broken into small bits. | Several years later, Berthelot’ and Playfair® dis- proved the above assertion of CdSe decomposition by Pt and pumice at red heat, and held that Uso, when heated strongly, gave only C, 8, and CS_ vapor. In 1875, Sidot” declared that he had obtained cS in a solid form by the decomposition of CSp in sunlight, 1 Baudr imont » Compt. Rend., 44, 1000, (1857) @ Berthelot, L'institute, (1559), 365 5 Playfair, Jour. Chem. Soc. London, 13, 248 4 Sidot, Compt. Rend., 81, 32, (1875) (Page 5) using certain metals as catalysts. §S. Kerm in 1876 showed that the solid reddish substance obtained thus was due to the iron catalyst used. Deninger” claimed to have prepared gaseous CS by treatment of chloroform and iodoform with NajS and Ages respectively, and also by the action of sodium on a mix-~- ture of CSe and analine. The resulting gas was identi- fied by explosion and subsequent measurement of the ratio of S0g to Co, but this work was disagreed with by Russell and smith® and by Dunn“. Julius Thomsen” tried to carry out the reaction be- tween CSp and Cu in two stages, the first giving CS and the second © and CuS. He claimed to have obtained a large yield of CS, but Stock and Kuchler® showed this be- lief to be due to an error in the construction of his apparatus. The most recent work has been done by Dewar and Jones’, who obtained solid CS from the reaction X NL (CO)4 + X CSClo = NiGlop + 4X CO (GS)... 1 this experiment of Sidot's was repeated qualitatively without the use of a catalyst, but using a quartz con- tainer in order to get ultra-violet light on the CSe. After several weeks of exposure, a fairly heavy, reddish- brown precipitate was observed which answered the des- gription given by Sidot of his solid CS. Deninger, Jour.Pr.Chem., 51, 346, (1895-2) 5 Russell & Smith, Jour.London Chem. Soc. 81, 1538 Dunn, Proc. Shem. Soc. of London, 26, 116, 5 Thomsen, Zeit, Inorg. Chem., 34, 187, (1903) 6 Stock & Kuchler, Berichte, 36, 4336. 7 Dewar & Jones, Proc. Royal Soc. of London, (A) 83,408. (Page 6) L. OG. Martint gave evidences in the form of photo- graphs and wave lengths of some bands supposed to be due to CS formed in the vacuum discharge in CS5 and by S in ac arc. from the above summaries and from work done in 1915 by Lewis and Lacey”, it seems probable that Cs exists in both Baseous and solid form, the latter being a polymer of the gas. Of the various methods of preparation of the gas, one of the few which was not refuted was the original one of Berthelot by which he decomposed C55 vapor with charcoal. Hence, it was decided to try this methd under more modern conditions andiesing activated carbon as the reducing agent. II - Me THOD It is probable that CS exists and corresponds to CSg as CO corresponds to COe, and also that 0S is an easily polymerized gas which is, however, not condensable at liquid air temperatures“. The difficulty in forming gaseous CS lies in the endothermic character of both sul- phides. The heat of formtion of gaseous CSo from amor- phous C and solid S is given by Landolt-Bérnstein as - 26000 cal. and that for CS under similar conditions by Thomsen” as - 28000 cal. Such compounds require high temperatures for pro- 1 L.c.Martin, Proc .RoyalSoc. of London, (A), 89, 127. 2 Lewis & Lacey, Jour. Am. Chem. Soc., 37, I976, (1915) 5 Julius Thomsen, Zeit. Inorg. Chem., 34, 187, (1903). (Page 7) duction from the elements, but they may also be obtained by breaking down other compounds. It seems probable, therefore, that the best way to form CS is by the suit- able breaking down of CSe- To accomplish this end through the reaction CSotC = 2 6S, using activated carbon as the reducing agent, the apparatus shown in Figures 1, 2, and 3 was designed and set up. The proposed use of the apparatus was as follows: In tube X was placed the liquid CS and in tube Y¥ the active carbon. Tubes F and G comprised a mercury pump by which the mercury in F was raised and lowered by alter- nately turning conpessed air and vacuum into G. As it was raised and lowered, the three-way stopcocknwas first turned so that the air in F was expelled through o and then so that the air in the rest of the apparatus was sucked into F, after which it was again opened to o and this air expelled. By repeating this cycle a number of times a fairly good vacuum was obtained within the apparatua. (Page 8) Tr ambly 223) | (/\ST7 a { *S> J axnvd «z0dde A = eernn2en Ky Vapgusou sary x) CS aenvw§ daddop L ry —, J iL t y & poco scluroo \ (Page 9) (Page 10) One essential to the success of this apparatus is that no O2 be present, as it would complicate matters by contaminating the product with CO, C09, COS, and other un- desirable gases which would render the separation and measurement of the CS very difficult. The active carbon contained a large amount of absorbed materials such as HeO and Og which had to be removed before the CSp was in- troduced. In order to get ria of all the Og in the apparatus, the CSe in tube X was first frozen by means of the liquid air in bath B, which was pumped by compressed air from container A. Then the reaction chanber Y was heated to 300 degrees C for about six hours by means of the Hg bath C, the Hg vapors being drawn off at v. Dur- ing this time the whole apparatus was evacuated as com- pletely as possible by the merarry pump. There still re~ mained, however, a small amount of oxygen, which was re- moved by heating tube D, which contained copper gauze, to about 350 deg. C by means of a tubular metal air bath. This caused the Og to be removed as black CuO and Cug0d, which was deposited on the copper gauze. The two-way stopcock 1 was then closed and that at k opened, and the liquid air container removed from the CSg tube, which was then gradually brought to a tempera- ture of 35 deg. C by immersion in a di-ethyl ether bath. This gave a high vapor pressure of CSo (whose b.p. is 46.2 deg.C) throughout both tubes X and Y. Stopceock k was then closed, thus giving a definite vapor pressure of (Page 11) CSp and weight of active carbon in the reaction chamber Y. “this mixture was then heated at constant tempera- ture for several hours, during which time any reaction between the two substances took place. After this, stop- cock k was re-opened and the CSeq reservoir again surround- ed with liquid air, which drew over any CS vapors that did not react and disposed of them as solid CSe. cs is not affected at liguid air temperatures according to Lewis and Lacey. Then stopcock 1 was opened and the CS resulting from the reaction was drawn by means of the mercury pump into the small tube u (the copper gauze in D had been al- lowed to cool), whence it was forced into the reservoir E& by reversing a stopcock m. ‘The process was repeated several times until all of the gas had been drawn over. The gas may then be measured and led out through the exit tube o and identified by explosion and conse- Quent measurement of the ratio of COo and 300. A rough indication of the presence of CS may be afforded by the heating of the copper gauze in tube E, which is similar to that in DBD. If a previous blank run shows no black coloration on the copper due to C&o, it is reasonable to suppose that any black color which may result is due CS, which is known from the work of Lewis and Lacey to react with Cu: cS +Cu = CuS+C. (Page 12) The only other possibilities ror contamination of the C8 produced would be from the decomposition of the CSe in the reaction tube to give carbon and gaseous sul- pher and the decomposition of the oxides of coprer formed in De The following caluclations show the absence of any possibility of the presence of S, gas from the decomposi- tion of the CS, and even if some were formed in this way, it would be removed as solid by the liquid air, CS_ (gas) = S>(gas) + © (amorph.) + 12500 calories. K (823 deg.c) for this reaction =(Seicse)= 7.8 x 107° Then, from van't Hoff's law, K (300 deg.c)= 8-5 x 1076 Thus, if the pressure of the CS, were assumed to be 600 mm in tube Y, the maximam pressure of any sulpher vapor from its decomposition would be 5.1 x 107° mm, which conclusively shows the absence of any chance of contamina- tion from this source. Any probability of the CuO and Cup0 decomposi ng to give off Op is precluded by the data of Foote and Smith on the subject. CuO dissociates 1% at 1000 degrees C. Cuo @issociates <1% at 900 degrees C. Iti = EXrRRIMENTAL The apparatus as outlined was completely set up and 1 Foote & smith, Jour.Am.ChemSoc.,50, 1344, (1908). (Page 35) tests made as to the efficiency of the mercury pump and everything was in readiness to begin making runs when it was learned that liquid air could not be ob- tained from the expected source. As the essential sep- aration of CS from CSp depended on the freezing of the CSq (which can not be accomplished by any ordinary freezing medium because the f.p. of the substance is - 116 deg.C), 1t was decided to abandon this problem for the time being and to take up the work on active carbon and ammonia. IV - SUMMARY l - A complete apparatus was designed and con- structed for the carrying out of the reduction of carbon disulphide by active carbon, but no experimental work could be done because of the failure of the liqid air supply. — (Page 14) PART B - THE REDUCTION G? AMMONIA BY ACTIVE CARBON. I - PREVIOUS WORK There seems to have been very little work done on the reaction between NHz and C, about the only work of any importance on this subject having been reported by G. A. Voerkelius!, He describes the results obtained by leading a mixture of NHz and Ho or illuminating gas in varying proportions and at varying speeds over wood charcoal heated to different temperatures. The purpose of the Hp and illuminating gas was simply to dilute the ammonia. He first observed the formation of HCN at 700 deg.C and obtained a maxinmm yield of 38% HCN at 1000 deg. C. The percentage of HCN and undecomposed NHz in- ereased with the rapidity with which the gas mixture was passed over the charcoal and also with increasing dilu- tion of the ammonia. With these results as a basis for comparisons, very interesting conclusions can be drawn as to the relative effects of activated and ordinary charcoals on the re- action. II - METHOD When NHz is led over hot charcoal, a number of reactions are possible: 1 yoerkelius, Chem. Zeit., 33, 1090, (1909) (Page 15) (1) C+NHgz = HON + Ho (2) C+NHg = 1/2(CN) + 3/2H (3) NHg = 1/2Ng+ 3/2Ho (4) HON C +1/2Ng + 1/2Heg of these possibilities, it was decided because of lim- jted time to analyse only for NHs, HCN,and (CGN)e. fo this end, the apparatus shown in Figures 4 and 5 was designed. Into tubes A, B, and GC are pipetted definite amounts of standard acid, AgNOs, and NaQH solu- tions respectively. One gas of the mixture coming from the combustion chamber F is absorbed in each of the three tubes. NHg is taken out in A,HCN in B, and (CN)o inc. The acid selected for A was subhuric because of its low vapor pressure as compared with the other common acids. There is consequently less liability of sulphate ion passing over into the silver solution and causing con- tamination. AgCl would be almost sure to precipitate if HCl were used in A. When the gases are passed through Aa, only the NHz will be absorbed because of the acid character of both the HCN and the (CN)o- In 3 all of the HCN will be ppt. ag AgCN, with the (CN)o passing on to be absorbed in the NaOH solution in G. The absorption tubes used were of special design (D of Fig. 4) which proved very effective. The incoming gases were led down nearly to the bottom of the entering tube where they wre bubbled out through . 70 V, AK. — * . ’ Rheos tat pinch-clamp ; mi (CEs L \ Vf waste Gases FE ° ‘le o ag 3 “ye 2] e a PR alo o (yy D rs) © @ Y A B C Liquid 7 NY Figure S (Page 17) small holes into another alightly larger enclosed space, at the top of which were more holes which still further proke the gas up as it escaped finally into the main body of the tube. These upper exit holes were made ale ternate with the lower ones in order that better break-~ ing up of the gas might be obtained. By means of a small capillary osning in the tube just before entering the absorption tubes, supplemented by a pinch-clamp, quite even bubbling was obtained. As a cylinder of liquid NHgz was used for the NHz source, the back pressure necessitat- ed by this arrangement was no drawback. The methods of analysis used for the resulting solutions in tubes A and C were volumetric, and that for B was gravimetric. The excess acid of A was titrated ag against standard NaOH and from this the amount of acid used inneutralizing the NH, and the amount of NHg were calculated. In obtaining the amount of (CN)> caught in the NaOH solution inc, AgNnOz was used to titrate against it. When this is added with constant stirring to a basic solution containing (CN)o, no permanent precipitate of AgCN occurs until all of the (CN)> has combined with the Sodium and silver ions to form a soluble double salt, {(NaCH, AgCN). With any excess of silver ion, a permanent precipitate of AagCN occurs the appearance of which marks the end point of the titration. ‘The basic solution of the cyanogen is diluted from 10 to 20 times with water (Page 18) before the titration in order to prevent too great a concentration of base. The mixture resulting from B was filtered and washed, thus removing all of the agCN, The remining Ag in the filtrate is then precipitated as AgCN in a slight- ly acid solution and measured gravimetrically in a Gooch filter. From this and the known amount of Ag added to tube B at the beginning of the run, the amount of HCN may be readily calculated. The furnace used was of the simple electric re- sistance type with two units connected in parallel across 70 volts A.C. A 12 amp. rheostat was connected in series ‘with the line for temperature control. With all the re- Sistance in, a temperature of 425 deg. &. was obtained as a minimum, while a maximum of 750 deg. ¢ could be reached. The combustion tube used was of Pyrex glass, and the ac~- tive carbon, in order to obtain the best contact possible, was packed loosely in a space about 6" long and com- pletely filling the cross-section of the tube, being held in place by asbestos plugs. This part of the tube was then centered in the furnace where the temp. was the most uniform. The tube was packed and sealed in at the end of the furnace by shredded asbestos in order to pre- vent circulation of air and consequent uneven heating in the furnace. (Page 19) TIT - EXPERIVENTAL ‘a total of seven runs under varying conditions was all that time allowed. in tabular form, the results ob- tained appear as follows: Run Duration Temp. Total tp in HON # (min.) (degC) ee HCN (cN) No as % of Kemarks (eg) (eg) (g) Totaly 1 75 425 ee ee we) aoe won Qualita- tive run 2. 90 425 2.765 Trecees None --- Traces re 3 75 500 2.725 OO115 " 2.251 0.265 — 4 70 600 2.395 00141 " 1.978 @.370 a 600 2.680 00205 " 2.2 o 2s Bsterrate 5 60 0 . 5 217 0.47 of bbb ling 6 60 600 2.215 a0187 " 1.835 0.528 fitipas” 4 60 725 1.740 0284 " 1.580 9.325 one The first run was primarily a gqualitdtive one for the purpose of determining the concentrations of the solutions necessary for the three absaption tubes. It was found, using N/2 HeS04 in A, that a slight white ppt. formed in B at the start of the run which later became fainter and finally disappeared entirely. This was vorobably due to the fact that all of the acid in A was used up in neutral- izing part of the NHz, after which the NHzg passed on to tube B, where it made the solution alkaline, under which condition AgGN does not ppt. Hence, it was found nec- essary to make the HoetO, up about 6N in order to absorb all of the Nig in a run of about one to one and one half hour's duration. A concentration of n/20 was found to (Page 20) be satisfactory for the AgNOgz solution and one of N/10 for the NaOH. The temperature of the carbon for the runs was varied fom 425 degrees G to a maximum of 725 degrees, and the effects on the yield of HCN of running the NHz through the combustion tube faster and of diluting with illuminat- ing gas were also tried. The results tabulated above all appear reasonable except the percentage ybld of HCN in run #7. In this case it is probable that the high temper- ature used caused a partial decomposition of the NHs as well as a greater reaction with the active carbon to form HCN. The No produced from such a partial decomposi- tion would not be measured by the avparatus used, and hence the amount of total No indicated is probably low. On the basis of the previous amounts of HCN obtained, the amount resulting from this run is about an average of fif- teen times as large for the same length of time and rate of bubbling. Hence, the correct percentage of HON is probably somewhat lower than 9.525 and would probably fall in the region of about 7-8%. Ordinarily there is prac- tically no decomposition of NHz at any temperature below 650-700 degrees CG, and hence the above effect does enter into the other results, although there is a pessibility that the active © might lower this enough to cause slight decomposition of the NHg at 500-600 degrees c. From the tabulated results, it apvears that the (Page 21) the process which occurs in the reaction zone tends to be a balance between the action of C on NHg to give HON and the decompositions of this HCN and of NHz. Any conditions which tend to remove the products of the first reaction (to form HCN) as soon as it is formed (such as dilution with illuminating gas and increased rate of passing the NHz over the carbon) increase the yield of HON by tending to get it out of the reaction zone before the decomposi- tion reaction has had a chance to get very far. The ef- fect of coiling the gases by wrapping a wet towel around the exit tube from the reaction zone to the first absorp- tion tube was also tried, and it was found that this also increased the HCN yield by cooling the gas below its de- conposition temperature. IV - SUMMARY 1 - In general, the same results were found for the reaction of NHz and active carbon as were determined for NHs and wood charcoal by Voerkelius, but the active carbon causes Slight yields at temperatures below 700 degrees C, and seems to have the added effect of causing larger yields above 700 deg. than wood charcoal. 2 - No cyanogen is formed in the reaction,which tends towards being a balance between the formation of HCN and the decomposition of NHg and HON. Methods of checking the decompositions by removing the HCN as fast as it is form- markedly increase the yields of HCN.