CATALYSIS IN THE DECOMPOSITION OF POTASSIUM CHLORATE BY MYRON ALONZO SNELL THESIS FOR THE DEGREE OE BACHELOR OF SCIENCE CHEMISTRY COLLEGE OF LIBERAL ARTS AND SCIENCES UNIVERSITY OF ILLINOIS 1922 Digitized by the Internet Archive in 2015 https://archive.org/details/catalysisindecomOOsnel ) 922 Sh 2 UNIVERSITY OF ILLINOIS l-lay__3_5 --IQ2A THIS IS TO CERTIFY THAT THE THESIS PREPARED UNDER MY SUPERVISION BY _ _Al_onzo_ _Snell_ _ entitled C at I s_ _ In _ jbhe_ JDe oo si t_i on _ o f _ Pot as siuin __Chlorate_. IS APPROVED BY ME AS FULFILLING THIS PART OF THE REQUIREMENTS FOR THE DEGREE OF P£_3_Cc£_*v.C,e* d»v__C_ RenUwSTtxa _Cj_ Instructor in Charge Approved HEAD OF DEPARTMENT OF TABLE OF CONTENTS. Aokno wl e dgrae n t I. Introduction II. Historical III. theoretical The Course of the Reaction IV. Experimental a. Materials b. Apparatus V. Experimental Results a. Qualitative Tests b. Nature of the Catalyst c. Relative Activity of Catalysts d. Effect of Varying Percent of Catalyst e. Promoter Action VI. Summary VII. Bibliography Page 1 3 3 6 7 8 9 10 11 11 13 33 - 1 ~ I INTRODUCTION The decomposition of potassium chlorate is facilitated by certain substances which act as catalysts in the reaction. The catalytic effect includes a considerable lowering of the temper- ature at which the decomposition begins, and a tremendous increase in the rate at which the oxygen is given off* When potassium chlorate is heated alone, oxygen begins to come off at about 340 o- 350®, and manganese dioxide, heated alone, gives off oxygen, beginning at about 400® (Alex Smith), but with a mixture of the two, the evolution commences at about 225®, more than 100® lower. The rate of evolution of the gas increases very slowly as the temperature is raised until, at 305®, it suddenly becomes rapid, and at about 335® as the chlorate begins to melt, it is toe great to be measured. < - . - 2 - II HISTORICAL It is not known who was the first chemist to observe that certain catalytic agents accelerate the decomposition of potass ium chlorate to give oxygen, but several men have added some- thing to the knowledge of the subject in the last seventy- five years. The first contribution was that of Wachter, in 1843. Other prominent contributors were Baudrimont and Jungfleisch in 1871; Schulze in 1860; Hodgkinson, Lowndes, and Veley in 1888-89; Fowler and Grant in 1890; McLeod in 1889, 189 4 and 1896; Brunck in 1893; and Sodeau in 1900; 1901, and 1902. (The mo3t important results of these investi- gators is outlined below.) Nothing much of importance has been done on it since the work of Sodeau. . . - 3 - III THEORETICAL The Course of the Reaction. A great many different reactions have been suggested for the decomposition and there still exists a great deal of disagreement on this question. McLeod in 1896 considered the decomposition to be a chemical reaction, which involved the catalyst (approximate reactions) : 2 KC10 3 + 2 MnO s * 3 KMn0 4 4 Cl 8 + Og 2 KMn0 4 * K 8 Mn0 4 + MnO a + 0 8 KgMn0 4 + Cl* * 2 KC1 + MnO a + Og He found, also, that the particles of MnOg are broken up during the decomposition by examining them with a micro- scope. Baudrimont and Veley claim than an evolution of oxygen can be produced without fusing the chlorate and with no catalyst present. And since many inert substances, such as Si0 2 do not have a catalytic effect, Soaeau stated that what actually took place was probably a chemical reaction involving the catalyst, rather than a physical efxect due to contact. Sodeau also reasoned that the chlorate probably under- goes self-oxidat ion when heated alone, and the perchlorate * . » is formed: - 4 - 4 KC10 3 + 61,300 cal. * KC1 4 3KC10 4 His idea is confirmed by the following facts: 4 KClOa - 39,000 cal., 4 KC1 + fo 4 KCXO3 cal .> 3 KCIO* 4 KC1 Then by Hess's law: 3 KC10* 4 KC1 l~i 00 » 500 Cal - > 4 KOI 46 0* Since any react ion ^will always take place in the direction in which heat is absorbed, the above reaction will go through the formation and decomposition of the perchlorate at the expense of the other one. Consequently unless a catalyst is present, the perchlorate is always formed as an intermediate compound. This reaction occurs only at a temperature of 340® or higher. At this temperature, only a small amount of the total oxygen present in the ohlorate is given off, as shown by Remsen's equation, and confirmed by experimental results: 8 KC10 a = 5 KCIO4 + 3 KC1 + 3 0 S But when a catalyst is present, no perchlorate is formed, even temporarily from the chlorate, according to Sodeau, since 1 > t - 5 - Ecoles found none in the residue on analysis, and McLeod showed that if it was once formed, it would not be decomposed under the conditions of the experiment, so that the reaction is probably 3 KC10 3 = 2 KC1 + 3 0g (Melior) or one similar to it. - 6 - IV EXPERIMENTAL a* Materials. C. P. KC10 3 was used in all of the determinations. The crystals were very fine and white. The commercial MnO a used was found on analysis to contain 91.07$ MnO s , and 6.83$ Fe 2 0 3 . The pure MnO s was prepared after the method of Georgen. C. P. Manganese carbonate was added in excess to nitric acid. The crystals of Mnjho^were washed and changed to MnO a by heating at 300® . The other catalysts employed were the impure compounds: A charge was prepared as follows: The KC10 3 and catalyst in weighed proportions, were mixed thoroughly, ground in a porcelain mortar, moistened with distilled water and dried. In this way, the catalyst was deposited on the surface, and in the pores of the KC10 3 in a finely divided form, and in intimate contact with it. A charge consisted of 6 g. of KC10 3 , plus an amount of the catalyst calculated as a certain percent-* age of the total weight. For example, in the table, KC10 3 plus 25$ of MnOg, refers to 6 g. of the chlorate mixed with 2 g. of the dioxide. - 7 - to* Apparatus. The apparatus used to study the decomposition was essen- tially as shown in Fig. 1. The reaction tutoe was a Pyrex glass test-tube 3.5 toy 30.5 cm. The charge occupied a length of from 3 to 3 cm. in the bottom of the tutoe. The flask was also of Pyrex glass 5 toy 35 cm. The gas was collected toy downward displacement of water in a 100 cc . eudiometer. The furnace was connected to 110 A.U. terminals, the current toeing cut down to a suitable amount toy means of a cartoon resistance. The temperature was measured with a 360° thermom- eter, with the bulb placed in the reacting mixture. The temperature of the charge was kept constant toy means of a con- stant boiling liquid in the bottom of the flask. Benzoic acid was used for a temperature of 335©, acetanilid for 300<> , and anthracene for 338° . The variation in temperature was . - 8 - V EXPERIMENTAL RESULTS* a* Qualitative tests. The temperatures chosen were 235©, 300©, and 328© • Below 235©, the decomposition was too slight to he appreciable, and above 328©, too violent to be controlled or measured. The boiling points and suitability of the liquids also had to be considered. As a basis for comparison, a charge was run without any catalyst present. This was followed by experiments on silica, sodium chloride, sodium sulfate, sodium carbonate, alumina, cobalt sulfate, zinc oxide, molybdenum oxide, stannic oxide, barium dioxide, bismuth oxide, and mercuric oxide, none of which exerted a catalytic influence on the reaction. Cr 2 0 3 , Co a 0 3 , and Ni 2 0 3 each caused the chlorate to give off both oxygen and chlorine at a tremendous rate, even below 235©, due to the fact that they reacted with the chlorate, liberating part of the oxygen and all of the chlorine. The probable reaction for the Cr 2 0 3 is given below. K 2 Cr 2 0 7 was tested for, and found in the residue. Cr 2 0 3 + 2 KCiO^ — K 2 Cr 2 0 3 4- Cl 2 + 0 2 CuO, MgO, Pr 4 0 7 , Nd 2 0 3 , Mn0 2 , and Fe 2 0 3 ail facil- itated the reaction, but the last two, only, showed a consider- - e - able catalytic effect, and were the only ones that were studied in detail* The results are shown in the accompanying table* The data on MnO a and Fe 8 0 3 is represented graphically in figures 2, 3, 4, 5, 6,7 and 8. The effect of varying the temperature, and the amount of catalyst is clearly indicated in these curves. The rate of the evolution of oxygen was measured in cubic centimeters per minute at 22®, and 760 mm* pressure • b* Nature of the catalyst. Sodeau showed that any reaction other than one including alternate oxidation, and deoxidation of the catalyst would involve setting free of all of the chlorine, and its subsequent complete reabsorption, and since the amount of chlorine given off is very small, the catalyst is the oxygen carrier. An intermediate compound is probably formed between the catalyst and the oxygen which is a surface complex of indefinite and variable composition. This substance is unstable and is almost immediately broken up again into the catalytic agent and oxygen. The only substances which could act as catalysers, would be then, oxides of metals, capable of more than one valence. This was actually found to be the case, since the oxides of copper, magnesium, praseodymium, neodymium, manganese, and iron were . ' T < - 10 - the only catalysers, and all of these are oxides of metals with at least two valences. There seems to be some doubt as to the existence of another oxide of magnesium besides MgO, but R030Q3 and Schorlemmer state that Mg0 2 exists, but is unstable. Magnesium would not, then, be an exception. How- ever, there are metals which, it would seem, ought to catalyse the reaction, such as barium and calcium, both of which form two oxides. For some unknown reason such metals are exceptions to the rule. c. Relative Activity of Catalysts. The table shows that Mn0 2 and Fe 2 0 3 are the only efficient catalysts for the reaction, and that Mn0 2 is almost twice as active as Fe 2 0 3 in increasing the rate of gas evolution. Praseodymium oxide i3 next most active, then the oxides of copper, magnesium, and neodymium in the order mentioned. Manganese dioxide seemed to have the greatest effect in lower- ing the temperature of decomposition. With Mn0 2 , the oxygen began to come off at about 235®, with Fe 2 0 3 , the evolution began at a trifle higher temperature, and so on for the others. So the temperature at whioh the decomposition begins is in*--- veraely proportional to the activity of the catalyst in question, in increasing the rate of gas evolution. 11 d* Effect of Varying Percent of Catalyst* The effect of varying the amount of catalyst is shown clearly in Figures 4, 5, 6 and 7 for the three temperatures 235©, 300©, 328©* The optimum concentration for both MnO a and Fe 2 0 3 was about 33$ • Experiment showed also that the optimum concentration did not vary with the temperature, and that the amount of Oatalyst present had no effect in lowering the temperature of decomposition* e* Promoter Action. Impure MnO s was found to be more active than the pure MnOg* Analysis showed this to be due to the presence of some Fe 2 0 3 as an impurity* So a charge was made up of 22*8$ pure MnO s and 2.2$ pure Fe 2 0 3 , the proportions existing in the impure Mn0 2 * As shown in the table, this mixture gave essen- tially the same results as did the impure Mn0 2 . This is evi- dence for the belief that the Fe 2 0 3 exerts a promoter action on the MnO a in catalysing the reaction and vice versa. This action was studied with various concentrations of both catalysts at the three different temperatures. The results are shown graphically in Figure 8* The optimum concentration was found to be 50$ of each catalyst* Thi 3 mixture was much more active than 100$ of either one. - 12 - VI SUMMARY 1. A number of s^ustances have been tested for a catalytic effect upon the decomposition of pota33ium chlorate. Only certain metallic oxides are active in this respect. 2. The two best accelerating agents, MfiOg and Fe 2 0 3 have been studied at several temperatures, and in varying concentra- tions. The optimum concentration of the catalyst has been found to be approximately 33 $ by wei^it. 3. Promoter action is indicated by the fact that the activity of Mn0 2 is improved by an admixture of Fe 2 Q 3 , and vice versa* The superior activity of pyrolusite to pure MnO s is thus explained by the presence of Fe 2 0 3 as an impurity in the ore • 4. A sudden, abnormal increase in the rate of the evolution of oxygen at about 305© (as illustrated in the curves) indicates that this is a critical point in the de comp os it ion » 5. It appears from available data that the effect of heating KC10 3 alone may be expressed by two concurrent reactions: (1) 4 KC10 3 = 3 KC10 4 * KC1 (2) 2 KC10 3 = KC1 + 3 0 8 of which the first is of much greater magnitude, especially if the temperature is low. The XC10 4 subsequently decom- poses at a higher temperature into KC1 and oxygen. - 13 - 6* In the presence of a catalytic agen$, the de compos it ion of KC10 3 proceeds by the second equation, at the expense of the first, or more probably, by an entirely different route, in which an intermediate compound is forma* between oxygen and the catalyst* This compound is un- stable and is probably of indefinite and variable compo- sition - of the nature of a surface complex. This con- clusion would seem to be supported by the fact that the only substances which are active accelerating agent 3 are the oxides of metals capable of more than one valence* - 14- Material $ Catalyst 335° CC. 0 2 per minute 300° 328° KClOs (6 g. ) — 0.35 1.0 7.7 MnO 2 1.9 3. 55 33.5 14.3 MnO g 3.55 5.0 59.0 35.0 Mn0 2 5. 95 8.0 94,5 39. 4 MnO 2 7.8 8.6 113. Q 33.3 MnO 2 8.3 13.4 119.5 40.0 MnO g 8.1 13.0 115.5 50.0 MnO 2 7.5 11.5 104.3 1 4. 3 Fe gO 3 2.1 3.3 34.0 3a. 0 Fe gO 3 3.6 5.3 51 . 5 35,3 Fe 2 O 3 4.85 6.65 62.0 4 a ,0 £ e g 9 3 4.3 6.4 59. 0 a 0 . 0 Fe 2 0 3 3,5 5.85 51.0 18.75 MnO s ) 6.35 Fe 3 0 3 ) 7.0 11.3 99.1 18.75 Fe 2 O 3 ) 6.35 MnO 2 ) 6.0 9.5 - - 89.0 ... V 13.5 MnO 2 ) 13.5 Fe gO 3 ) 7.9 13.35 ii&.s 35.0 CuO 0.95 1.95 34.0 35.0 Pr 4 07 3.35 3.75 38.0 35.0 Nd 3 0 3 0045 0.65 5.05 35.0 MgO . 45 .8 7.35 35.0 MnOgCC.P. ) 1.8 3, 6 37.5 33.8 Mn0 2 (C.P. ) 5.7 7.8 86. 0 • 3 F© gQ g ^ t 1 ft 15 Te. MPE RATURE. Fioune 2j 16 - - 17 - o »o Zo zo 40 Caocentratioo tn percent FtQ U RE 4 CC. GAS PER MIN U X E. -IS- CC.' . P£ K H I M U T t. ~| 5 ~ C e . G A C) R r^i I - 20 - o - Z I - 22 - VII. BI3LI0GRAPHY Baudrimont - J. Pharm. Chim. 1871 (IV) _14, 81 and 161. .Jungf lelsoh - Ibid 1871 (IV) .14, 130. H. Schulze - J. Pr. Chem. 1880 (II) 21, 426. Hodgkinson and Lowndes - Chem. News, 1888, J58, 309; 1889, _59, 63 Veley - Phil. Trans. - 1888, A . 271. Fowler and Grant - Trans. 1890, .57, 272. McLeod - Trans. 1889, 55 , 184; 1894, .64, 202: 1896, 69, 1015. Brunck,- Ber. 1893 , 26 , 1790; Zeit. Anorg. Chem. 1895, 10, 222. Wachter - J • Pr. Chem. 1843, 30. 325. Sodeau - J. Chenu Soc. 1900, 77 , 137 and 717; 1901, 79 , 247 and 939; 1902, 81, 1066. Mellor , J. W. - Modern Inorg. Chem. Alex Smith - Inorg. Chem. Remsen - Chemistry (1906) Georgen - Compt. rend. 1879, J58, 797. Roscoe and Schorlemmer - Treatise on Chemistry, Vol. 2.