Studies on the adsorptive powers of certain activated carbons

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content STUDIES OK THE ADSORPTIVE POWERS OP CERTAIN ACTIVATED CARBONS G 7>f £ :A Thesis Presented to the Faculty of the Department of Chemistry University of Southern California In Partial Fulfillment of the Requirements for the Degree Master of Arts hy Edward Conniff Lanphier May 193^ UMI Number: EP41455 All rights reserved INFORMATION TO ALL USERS The quality of this reproduction is dependent upon the quality of the copy submitted. In the unlikely event that the author did not send a complete manuscript and there are missing pages, these wilLbe noted. Also, if material had to be removed, a note will indicate the deletion. Otesflrtattoft RaMithfag UMI EP41455 Published by ProQuest LLC (2014). Copyright in the Dissertation held by the Author. Microform Edition © ProQuest LLC. All rights reserved. This work is protected against unauthorized copying under Title 17, United States Code ProQuest LLC. 789 East Eisenhower Parkway P.O. Box 1346 Ann Arbor, Ml 48106- 1346 This thesis, w ritte n under the direction of the candidate’s Faculty Committee and approved by a ll its members, has been presented to and ac cepted by the C ouncil on Graduate Study and Research in p a rtial fu lfillm e n t of the require ments fo r the degree of . * + * ‘ * ( <j?-i" ¥ acidly Committee C haft man TABLE OF CONTENTS CHAPTER PAGE I. INTRODUCTION Historical n o t e s .................. 1 Some uses of activated carbons............... 4 II. THE NATURE OF ADSORPTION....................... & The nature of surfaces....................... 9 Elastic and non-elastic collisions ........... 11 Process of condensation ....................... 12 Surface forces independent of temperature . . . 13 Process of evaporation ....................... 13 Adsorption defined 14- Adsorption a chemical process ................. 15 Evidence of its chemical nature ............... 16 Discussion of the work of Gurvich ....... 17 Monomolecular adsorption layer ............... 21 “Specific action" of adsorbents ............ 22 Equations worked out for plane surfaces .... 24 Langmuir*s work done at low pressures........ 26 Langmuir's equations for smooth plane surfaces o n l y ............ 27 Polanyi’s theory . .......................... 2$ Freundlich equation.............. 30 iv CHAPTER PAGE II. (CONTINUED) Patrick - adsorption and capillarity .... 31 Adsorption lay porous adsorbents akin to solu tion ........................................ 33 III. STRUCTURE OF THE CARBON MOLECULE AND OF CHAR COAL ........................................ 35 Drift in adsorption and in density determina tions ............................... 39 Alpha and beta carbons of N. K. Chaney .... ^1 IV. MANUFACTURE OF ACTIVATED CARBON Chaney’s primary carbon ................. * 4 - 6 Activation defined .......................... 4-7 “Selective"and "limited" oxidation ......... ^7 Activation by air, carbon dioxide or steam . . Early wartime activation ............... . ^9 Failure of anthracite activation in 191$ . . . 50 Methods of activation.............. 51 High temperature calcination. Its relation to air activation ....................... 51 Impregnation with chemicals ................. 5^ Removal of hydrocarbons by solvents ......... 55 "Selective" and "limited" oxidation (more on) 55 Some mechanical factors.......... 57 Properties required for gas and vapor adsorbers 57 V CHAPTER PAGE IV. (CONTINUED) Properties required for deodorizers .... 60 V. INFLUENCE OF SUBSTANCES OTHER THAN CARBON . . . 62 More modern work ............................ 62 Action of minerals present ................. 64 Selection of raw materials ................. 67 Presence of nitrogen ................. 70 Effect of the solvent................. 72 Influence of hydrogen ion concentration . . . 73 VI. EXPERIMENTAL PART Purpose ................................ 77 Materials used ................. 77 Procedure.................................... 7S Titration procedure ......................... 79 Handling of the carbons . ............... SO Computing results ................... .... Si Results and conclusions..................... Si VII. SUMMARY...................................... 92 BIBLIOGRAPHY ...................................... 93 LIST OF TABLES TABLE PAGE I. Comparison of Darco DC, Nuchar DC and Carbac . . S2 II. Effect of adsorbed moisture on the adsorption of iodine by Darco DC, Nuchar DC and Carbac . . . S3 III. Effect of three hours heating at 900°C on the ad sorption of iodine by Darco DC, Nuchar DC and Carbac................... .. ........... .. S6 IV. Determination of (k) and (l/n) of Freundlich*s equation for the activated carbon, Darco DC . . S7 V. Determination of (k) and (l/n) of Freundlich*s equation for the activated carbon, Nuchar DC . 6S VI. Determination of (k) and (l/n) of Freundlich*s equation for the activated carbon, Carbac . . S9 LIST OF FIGURES FIGURE PAGE 1. Carbon structure ................................. 3$ 2. Effect of pH on the adsorption of different types of colloids ......................... 7^- 3. Graphs for the comparison of three carbons .... 90 Logarithmic form of the graphs for the comparison of three carbons . .............................91 CHAPTER I HISTORICAL HOTES The use of charcoals as decolorising agents dates hack to the beginning of the modern era in chemistry. The first use of charcoals for this purpose occurred within a few years of the discovery of oxygen. In 1777> Scheele and Fontana"*" found that charcoal, heated to incandescence and subsequently cooled and evacu ated, had the power to adsorb gases. In 17&3, Morrozzo^ used a wood charcoal, prepared by cooling glowing charcoal under mercury (to cut off the air), to adsorb the gases, C02» H2, O2, air, NH^, HC1, H2S, SOg, and U02* Lowitz of St. Petersburg, in 17&5 used wood charcoal to remove colored solutes from solution-"* and shortly after ward, in 179^> there is a record of charcoal being used as i j . a deodorizer in an English sugar refinery • 1 J. B. Garner, "Charcoal as an Adsorbent." natural Gas, 5:3, 192^. 2 Ibid., p. 3. ^ Ibid., p. 3» k F. W. Zerban, Vegetable Decolorizing Carbons and Their Use in the Sugar Cane Industry (Louisiana State Experiment Station, Bulletin 161, Louisiana State University, Baton Rouge, La. , p. 5. 2 In 1202, Delessert was so successful in using charcoal in the beet sugar industry that Napoleon subsidized the start of several other refineries-*. c In 1210, Fiquier used boneblack for the first time in decolorizing sugar. Its activity was so much greater than the activity of ordinary charcoal that it came to be used everywhere in the sugar industry and charcoal was dropped as inefficient. Bussy, Payen and Desfosses, in 1222, noted that animal chars, prepared by carbonizing in the presence of potassium carbonate, which substance was removed by washing the char, gave extremely fine results as a decolorizing agent^. How ever, no attention was paid to their work and it lead to no commercial application. The Hindus of India refused to use sugar prepared 2 with animal chars and so, in 1236, Boettcher suggested the use of lignite for decolorizing the sugar sold in their country. Following this incident, there seems to have been a 5 Ihid., p.5. c ° J. B. Garner, loc. cit. ^ M. A. Schneller, “The Vegetable Decolorizing Carbons’ 1 Louisiana Planter, :15^> September, 1917. 2 F. W. Zerban, loc. cit. 3 period during which boneblack was the unquestioned "king" of the deodorizers as it is, even today, in some sections. In 1830, a controversy9 concerning decolorizing agents arose between Casamajor, chief chemist of an American sugar refining company, and Remmers, an English chemist. Casamajor recommended finely ground crude sawdust as a fil tering medium while Remmers advised the use of powdered wood charcoal. Later, Remmers switched to pulverized lignite. From then until 1887» a series of trials were made using these various substances. Test runs were made in privately owned refineries and by the Louisiana State Exper iment Station, a state owned organization. Final reports favored charcoal as giving a purer product, tho lignite was found to have about ten times the decolorizing power of charcoal. Hereafter, interest in charcoals seems to have waned and they next come to one’s notice in 1910, when a new type, "activated carbons", appeared. In 19H> ‘ two of them with the trade names, Epohit and Norit, were on the market* These chars had adsorbing power many times the adsorbing power of ordinary charcoal. With the use of gas in the world war, activated car- 9 Ibid., p. 6 k - bons became an essential part of the adsorbents used in gas masks. Much experimenting was done to find the best raw materials and to determine the best methods of activation. This work will be discussed later. After the war, industry made extensive use of the lessons learned and, today, there are many activated carbons on the market used in a wide variety of ways. Thus, infor mation ,concerning adsorption and factors influencing the efficiency of carbons, has become important to industry. SOME USES OF ACTIVATED CARBONS Chaney, Ray and St.John^ give lists of uses for the activated carbons. Many of these are being applied in the various industries today. The lists, published in 1923 > are worthy of attention. Under gas adsorption: I Recovery of gases, solvents, etc., from dilute mixtures. (A) (a) Gasoline from natural gas. (b) Recovery of oxides of nitrogen. 10 A. G. Fieldner, R. E. Hall, and A. E. Galloway, Study of the Production of Activated Carbon from Various Coals and' 'Giber Materials, tJ. S. Bureau of Mines, Paper' k-79. 11 N. K. Chaney, A. B. Ray and A. St.John, "Properties of Activated Carbon Which Determine Its Industrial Applica tion." , American Institute of Chemical Engineers, Transactions, 15*33^, 1923* I (A) continued’ 7" (c) Recovery of solvents, ethers, alcohols, etc., from the air. (d) Recovery of acetone and alcohols from the mois ture laden gases from fermentation vats. (e) Removal of sulfur compounds, benzene and light oils from illuminating gas. (B) (a) Abatement of stenches and odors from rendering plants and the like. (b) Purification of the air of submarines. (c) Gas masks - military and industrial. II Purification of gases: (A) Refining helium. (B) (a) Purifying carbon dioxide from fermenting vats for making soda water. (b) Purifying hydrogen and nickel carbonyl for hydrogenation purposes. (c) Purifying ammonia gas before catalytic oxidation. Ill Catalysis in gas reactions: (A) Decomposition or oxidation of hydrogen sulfide gas to give free sulfur. (B) Decomposition of phosphine in ammonia and acetylene purifications. (@) Oxidation of nitric oxides by the air, 6 III continued (d ) (a) Chlorination of hydrocarbons. (b) Chlorination carbon monoxide to make phosgene. (e ) A general means of facilitating counter current gas - liquid reactions. IV Storage of compressed gases: More of certain fixed gases can be stored in cylinders containing highly active carbon than in empty ones. V Evacuation of vessels: (a) Cheaply and simply evacuating commercial heat insulating containers. (b) In various apparatuses where low pressure must be maintained in spite of the slow formation of a gas in their use. Under adsorption from solution: (a) Making white sugar direct from cane juice. (b) Purification of organic and inorganic acids (carbon of extremely low ash content required). (c) Purification of a great variety of organic liquids. (d) Decolorizing waxes, gelatin, etc. (e) Removing objectionable colors from edible oils and fats. (f) Decolorizing and purifying petroleum oils. Purifying water - removal of tastes, odors and bacteria. Recovery of rare metals from dilute solutions. Recovery of alkaloids from solution. CHAPTER II THE NATURE OP ADSORPTION For many years, the term, "adsorption", was covered by the term, "absorption"• There was no differentiation between the taking up of moisture by a sponge and the taking in of gases by charcoal. In a vague way, it was realized that there was a difference. The charcoal or other sub stance doing the "absorbing" was apt to show a preference in taking up things from mixtures or solutions. Gradually, the idea of two different types of Pabsorb ing" processes developed. One seemed purely physical and due to capillary action while the other was seen to be more specific (influenced by the nature of the materials involved). It was recognized that the second type went on best where porous or finely divided materials were used. In the devel opment of the subject, the term, "absorption", has come to cover the process due strictly to capillarity while the more specific type of action is called "adsorption". There have been many theories advanced as to the nature of adsorption. Generally, it is agreed that adsorp tion is a surface phenomenon, but ideas of the actual process differ. One group would have it a purely "physical" process, the adsorbed matter being held by "mass attraction" and the like, while the other postulates a chemical reaction between 9 the material adsorbed and the adsorbing surface* Much has been written by both sides in this controversy* THE NATURE OF SURFACES Since adsorption occurs at surfaces, it would seem to be necessary to an intelligent investigation of the process that a clear idea be had of the make up of the surfaces in volved. In his work, van der Waals assumed surfaces to vary continuously and uniformily from one state to the other with no sharply defined line of demarcation. Polanyi, according to Svedberg^, assumes a dense poly molecular layer of adsorbed material, varying gradually out ward to the density of the surrounding gas or liquid. How ever, he does not do surface structure in detail. p Langmuir , making use of the Bragg space lattice, assumes the adsorbent to consist of atoms, ions or molecules, arranged in definite three dimensional patterns. He assumes that there is no continuous gradation to a gas, but that an abrupt change in arrangement exists. The surface of the close packed crystal lattice lies in contact with the more 1 The Svedberg, Colloid Chemistry, A. C. S. Monograph (New York: The Chemical Catalog Co., 192S) 2nd. ed., p. 16S. 2 Irving Langmuir, 1 1 Constitution of Solids and Liquids” Jour. Amer. Chem. Soc., j&:22k-9f 1916. 10 widely spaced gas molecules and there is no transition layer. In fact, Langmuir states, that at the actual contact surface, the difference in the degree of dispersion of the solid and of the gas is greater than the difference between the interior of the adsorbent and the gas. ". . .no contin uous gradation from solid to gas but an abrupt change with a more densely packed surface layer of adsorbent than the density of the interior." If charged bodies are brought into proximity, they move in such a way as to minimize the fields of force about them. However, the atoms of the surface are not acted on equally in all directions. He says, "Then the surface layer of atoms must rearrange themselves, tho the distance moved may be slight compaxed to the distance between centers. This abnormal arrangement is limited to the surface layer only." He assumes that, thus, a regular surface pattern is created and that the rearrangement is not sufficient to bal ance all the forces acting. The surface is described as, "... a sort of checker board, containing a definite number of atoms, of definite kinds, arranged in a plane lattice formation. The space between and immediately above (away from the interior) these atoms is surrounded by a field of electromagnetic force more intense than between the atoms inside the crystal.^ Others have discussed the condition of the adsorbing 2 Langmuir, loc. cit., (All references this page) 11 il surface. Donnan that the adsorbing surfaces cannot be smooth, the surface atoms and molecules probably being in chains and filaments. He thinks that the activation of a smooth metallic surface is due to the roughening of it (the creation of such chains and filaments). Practically all those taking up the subject state that even the smoothest surface must be somewhat rough and that the effective adsorb ing surface is very likely much greater than the measured “apparent surface". However, none go into detail or furnish logical reasons for their assumptions as does Langmuir. ELASTIC AND NON-ELASTIC COLLISIONS The kinetic theory assumes that collisions of gas mol ecules with surfaces are elastic, the molecules striking the surface and being reflected. Langmuir holds that such colli sions are wholly inelastic. According to him, the molecules striking the surface come under the control of surface forces and are held by the surface for intervals of time depending on various factors. In other words, the collision of a gas molecule with a surface consists of two separate and distinct processes, the condensation of the molecule upon the surface 1 5 and its evaporation therefrom. ^ F. G. Donnan, Faraday Soc., Transactions, 12:607,1917. ^ Langmuir, oj>. cit., p. 2250. 12 THE PROCESS OF CONDENSATION < 5 At a distance of approximately 2x10 cm., molecules approaching the surface come under the influence of the sur face forces and are accelerated towards the surface. The net force acting increases rapidly to a maximum and dimin ishes thereafter with equal rapidity as the molecules move in. By the time this force has diminished to zero, the mol ecules have acquired a kinetic energy sufficient to carry them on toward the surface. Thus, they enter a field of repulsion. Generally, this repulsion will he due to one atom. However, the acceleration inward was due to several atoms. Thus, several atoms were disturbed in the acceleration and only one is concerned in the retardation. Therefore, the incoming molecule will not be repelled strongly enough to acquire sufficient kinetic energy to rebound and escape from the influence of the surface field. Tho several atoms, con cerned in the retardation in some cases, they would not have enough energy to cause reflection as some of their energy would be passed to their neighbors under the surface. There fore, it is logical to assume that the molecules will be held near the surface. 6 Ibid., p. 22h6. 13 SURFACE FORCES INDEPENDENT OF TEMPERATURE According to Langmuir, the intensity of the surface field is independent of the temperature. WhereT = the free surface energy and% * the total surface energy, % = *r-T^f (i) There are good reasons for believing that the inten sity of this field of force is substantially independent of the temperature. The energy in the surface field is measured by T0 . Now, the work done during the formation of a fresh surface by the thermal agitation of the mole cules (namely, - T -gvp) should be approximately propor tional to the temperature. Therefore, “§5*" should be in dependent of the temperature. Thus, we may place nr= a - bT (2) where (a) and (b) are constants. Combining (1) and (2), we findri« a. In other words, , the total energy of the magnetic field should be independent of the temperature.' Thus, since condensation on the surface is due to the intensity of the surface field, the rate of condensation of molecules on the surface should be independent of temperature. THE PROCESS OF EVAPORATION As we have seen, the condensed molecules, instead of being repelled out of the surface field, lose a part of their energy until an equilibrium position is reached. There they remain and engage in vibration due to thermal agitation. In order to have such a molecule leave the surface, 7 Ibid., p. 22^7. lH- (evaporate from the surface), it must receive increments of energy from several neighboring atoms or molecules simulta neously, Since the ability to give up energy on the part of these particles is determined by their own energy content, the evaporation rate is influenced by the temperature. In event that condensed molecules ionize on the sur face, they must recombine before the molecule can evaporate. Langmuir calls attention to the fact that many more complex polar compounds are freed after condensation as simpler mol ecules. This is due, probably, to the difficulty that the ions of the original compounds have in recombining. With polar compounds, apparently, primary valence combinations occur with the atoms of the surface as different polar mole cules show differences of behavior toward a given adsorber more marked than in the case of non-polar molecules in gen eral. In the case of the simpler non-polar organic molecules, apparently, only secondary valences are involved as their behavior is similar to that of the monatomic gases. ADSORPTION DEFINED If condensation occurs, a certain time interval must elapse before the condensed molecules can leave the surface. Therefore, if an adsorber is exposed, due to the Hlagw of the evaporation process, a certain quantity of molecules will always be on its surface after equilibrium has been reached. 15 Thus, there will he a higher concentration of molecules on the surface than in the medium adjacent to it. In other words, adsorption will have occurred. ADSORPTION A CHEMICAL PROCESS Since condensation is unaffected by the temperature, the evaporation rate largely determines the tendency of the molecules to be adsorbed on a surface. It, the evaporation rate, in turn depends upon the magnitude of the forces acting between the atoms of the surface and those of the adsorbed substance. Langmuir considers these forces to have the same nature as those holding solid bodies together. We may, therefore, profitably look upon them as chem ical forces and apply our knowledge of the chemical pro perties in studying the phenomena of adsorption.3 In the surface layer, because of the asymmetry of the conditions, the arrangement of the atoms must be slightly different from that in the interior. These atoms must be unsaturated chemically and thus they are surrounded by an intense field of force.9 He says that the so called "physical" phenomena are chemical but that they involve secondary valences only,i.e.- forces due to the fields set up by the displacement of the surface atoms in adsorbers or the like. g Ibid*, p. 2269. 9 Irving Langmuir,"Adsorption of Gases on Plane Sur faces of Glass, Mica and Platinum." J. Amer. Ohem. Soc.* ^ 0: 1362, 191S . ~ -- -- -- EVIDENCE OF ITS CHEMICAL NATURE 16 He offers experimental evidence to support his ideas* I Atomic hydrogen Atomic hydrogen, because of its lightness, should be almost wholly MreflectedM according to HphysicalB theories. Atomic hydrogen, as a matter of fact, is exceptionally well adsorbed. Further, the theory of a highly compressed layer of gas on the surface would call for far less atomic hydro gen that is actually held there. Atomic hydrogen will pass thru glass tubes several feet long at room temperatures. At the temperature of liquid air, none passes thru indicating complete adsorption^ II Lack of relation between vacua and amounts adsorbed In high vacua, the adsorption should be low and in proportion to the vacuum obtained. Langmuir found that the quantity adsorbed, up to a certain limit, was independent of the vacuum and that this adsorption was remarkably stable. Ill Oxygen adsorbed on tungsten At 2770°K, fifteen percent of all the 02 molecules striking a tungsten filament are adsorbed. At 3300°K, this percentage is increased to fifty percent. Evidence shows that at least half the tungsten surface must be covered with oxygen in some form. Since the pressure used is not over five bars, the 17 film of oxygen is so unusually stable as to justify the idea of a chemically combined layer with the tungsten atoms. The film does not consist of WO3, since this would distill off at 1200°K. At 500°K, WO3 is reduced quite rapid ly by hydrogen at atmospheric pressure. At as high as 1500°K, hydrogen does not react with the adsorbed oxygen film. These data indicate, first, that the primary valence of the oxygen must be satisfied as the hydrogen cannot find a place to attach itself to the oxygen and, second, that the tungsten atoms are unsaturated, that is - they are attached to other tungsten atoms below them (toward the interior). If this were not so, the WO3 formed would distill off at the temperature of the experiment. DISCUSSIOU OF THE WORK OF GURVICH10 Langmuir^apparently chose the work of Gurvich because it seemed to embody the views of so many earlier workers on adsorption. Each of the main points made by Gurvich are taken up by Langmuir and arguments and evidence are offered against them. Gurvich rejected the idea of ”chemical” adsorption 10 Quoted by Langmuir, Jour. Russian Phys. Chem., k7:&05, 1915- 1} Irving Langmuir, ”Fundamental Properties of Solids and Liquids.” Jour. Amer. Chem. Soo., 39:1S4&, 1917. I Amounts of liquids adsorbed by a given quantity of adsorbent were not in agreement with stoichiometric re lations. He found the volumes of the various liquids adsorbed very nearly equal. His data was obtained by keeping one gram of adsorbent in the presence of the saturated vapor of the liquid until no further adsorption was apparent (15 - 20 days). Most of the adsorbents used were porous. II He concluded that adsorption forces were effective at greater than molecular distances from a surface with a gradually diminishing intensity. He determined the c effective distance as 3*10 cm. His data was obtained by placing silver foil or glass wool in the presence of saturated vapors until weighings showed no further gain in weight. The thickness of the adsorbed film was calculated from the weight of the liquid adsorbed and the area of the surface exposed. Ill Gurvich found the heat of adsorption to be greater where similar substances were involved than in the case of disimilar ones. Since ordinary chemical reactions occur between disimilar substances and the heats of adsorption indicated reactions between similar substances, he concluded that adsorption was not a chemical process. 19 Langmuir answers each item in turn: I Langmuir says that in his work with films on water,12 he found stoichiometric relations common. He proved that where oust enough liquid was used to cover the water with a film, the films were one molecule thick. With similar compounds, practically the same number of molecules were found per unit area. He states, that with porous adsorbents, the bodies of those molecules adsorbed in small cavities would be large enough to prevent the complete coverage of the surface by other molecules. Being in contact with saturated vapor, the larger cavities would be filled by capillary action. Therefore, the volume adsorbed was about the same in all cases, as Gurvich measured what amounted to the volume of the larger cavities only. II Where foil was crumpled or glass wool packed in, the surfaces were close enough to cause capillary action. This would cause quantities of liquids to be held in excess of the amount actually adsorbed. Naturally, the weight figures would be too large and, thus, computation would show adsorbed layers more than one molecule thick. Ill In first order reactions, primary valences are in volved and unlike substances react. 12 Ibid., p.1364. 20 All thru chemistry, there are evidences of another type of valence - reactions between oxides, between halides, between metals, etc. He calls this type of valence, secon dary valence. He, therefore, concludes that the solubility of like substances and the adsorption tendencies of like substances are due to secondary valences, which seem to be most active between like substances. In discussing earlier theories, it is well to note that earlier proponents of 1 1 chemical" adsorption believed that a layer of molecules of a definite compound was formed on the adsorbing surface. Langmuir presents evidence of chemical combination of the adsorbed material with the sur face atoms of the adsorber, these surface atoms still retain ing their chemical union with the atoms within the adsorber. Also, earlier workers like Gurvich, used saturated vapors and made no provision to avoid capillary spaces. In his work, Langmuir has avoided possible condensations of liquid and the resulting capillary action by using vapors at extremely low pressures. As a further preventive of capil lary action, he has spaced his adsorbing surfaces at proper distances apart. He and Sweetser obtained results indicating layers of adsorbed molecules one molecule thick where surfaces were spaced sufficiently far apart to avoid capillary action. THE MOFOMOLECULAR ADSORPTION LAYER 21 Working with atomic hydrogen and glass surfaces, Langmuir found direct evidence of an adsorbed layer one atom thick. If the diameter of the hydrogen atom is assumed to be, approximately, the same as that of its molecule, namely - 2.5x10 cm., then the number of hydrogen atoms required to 15 p form a layer one atom thick is about 1.6x10 per cm . of surface. This corresponds to 0.032 cm.*' of molecular hydro gen at 20°0 and one megabar pressure. The maximum amounts of atomic hydrogen actually adsorbed correspond to the same order, namely - 0.01 to 0.03 om.^ per cm.^ Other work, with oil films on water, has shown similar results. Due to the fact that adsorptive forces do not act thru distances equal to the length of an ordinary oil mole cule, the films were found to consist of single layers of oriented molecules, tightly packed, their active portions in contact with the water. The only exceptions to this tight packing occurred where extra large active groups, such as _ nil FOg or I caused steric hindrance. ^3 Irving Langmuir, "Constitution of Solids and Liquids" Jour. Amer. Chem. Soc., 3^12270, 1916 . . . . , "Fundamental Properties of Solids and Liquids", Jour. Amer. Chem. Soc., 39:1^9# 1917 THE "SPECIFIC ACTION" OF ADSORBENTS 22 "In a heterogeneous chemical reaction, the activity of a surface depends in general upon the nature of, the arrange ment of, the spacing of the atoms forming the surface layer." The surface activity of the same adsorbent has been found to vary markedly for various substances adsorbed. Langmuir*s theories account for this phenomenon. The plane faces of a crystal consist of atoms forming a regular lattice structure. Necessarily, atoms in the cleavage surfaces of crystals like mica have the weakest stray fields of force of any of the atoms of the crystals concerned. In mica, hydrogen atoms must cover most if not all of the surface, since hydrogen atoms when saturated by such atoms as oxygen, possess only weak residual valences. In the case of glass and other oxygen compounds like quartz or calcite, the surfaces are possibly lattices of oxygen atoms. The sur face of a crystal thus resembles to some extent a checker board. The molecules adsorbed by such a surface take up def inite positions with respect to the surface lattice, thus tending to form a new lattice above the original one. Thus, a unit area of any crystal surface has a definite number of "elementary spaces", each capable of holding one adsorbed molecule of a definite kind. These spaces will not necessa rily be alike. There will be many cases where two or three 23 diffexent kinds of spaces occur. For instance, it is possi ble that both hydrogen and oxygen may occur in a surface, making various surface combinations. If the adsorbed molecules HO H HOH HOH HHOH HO HH O take up positions over the OHHOHHOHH H OHHOHH OH centers of groups of four of H HOHHOHHO OHHOHHOHH these surface atoms, then there HOH HO HH O H HO H HOHHOHHO are two kinds of groups, H H OHHOHHOHH HO and 0 H . It will be noted that, for each of the latter, two of the first kind occur. In event that atoms arrange themselves above individual sur face atoms, there are still two kinds of surface spaces to take into consideration. Thus, a surface may have spaces of only one kind or two or three or more different kinds, repre- 15 senting simple fractions of the surface. ^ It is reasonable to suppose that different kinds of spaces will have different tendencies to adsorb various mol ecules. In the case of the same molecules, it is reasonable to assume that different spaces will vary in their adsorp tion efficiency. At very low pressures, gas molecules would naturally be adsorbed by the most efficient spaces first. As the pressure is increased, making more molecules available, 15 ^ Irving Langmuir, "Adsorption of Gases on Plane Surfaces of Glass, Mica and Platinum.", Jour. Amer. Chem. Soc. kOiljGK, 191S. 2 * J - the less efficient spaces will fill up, thus, giving a pro cess of adsorption in steps, Langmuir*s work, tho not con clusive, gives some evidence of such adsorption. When different gases are adsorbed alternately on the same surface, often the quantities adsorbed are related in simple stoichiometric ways. Thus, if each space holds one molecule of each gas, equal numbers of molecules of each gas will be adsorbed. However, if three molecules of one gas are able to occupy the space taken by one molecule of the other gas, the amounts adsorbed will be to each other as one to three. The size of the adsorbed molecules, also, will have an influence on the number of molecules adsorbed per unit area. Amorphous surfaces, where the surface atoms are not in a geometric pattern, will present an endless variety of spaces. It becomes obvious that any equations, worked out to cover adsorption relations, will fail to apply in all cases, due to the very large number of factors involved. EQUATIONS WORKED OUT FOR PLANE SURFACES Let it be assumed that there is one kind of space only and an adsorption of only one molecule per space. Of all the molecules striking a surface, a certain fraction, cc , will condense and be held for some interval of time until evaporation takes place. If the number of gram- 25 molecules, striking a square centimeter of surface per second is designated by jx * then the rate of condensation on a bare surface will bea^u However, very shortly after condensation begins, a large portion of the surface will no longer be bare because of the adsorbed molecules and so, the actual rate of of condensation becomes where © is the fraction of the surface which is bare. The rate of evaporation will be w*©‘, where u* is the rate of evaporation from a completely covered surface and 0* is the fraction of the surface which is covered. When a state of equilibrium is reached, flf/A© = v'O • However, 0 + 0* = 1 so, therefore, 0* = at = — I + c r t f Now, the number of gas molecules adsorbed cannot ex ceed the number of elementary spaces on the surface. Let No be the number of such spaces per square centimeter. If rj is the number of gram-molecules of gas adsorbed per unit area, Let the ratio,-^r = Then, 0 1 a . < r ' t X~m . . then we have O' - where N is Avagadro's number. This gives an equation with items which may be either measured or calculated. If the surface contains more than one kind of space, let B', B'*, B,,,» etc. represent the fractions of the total 26 number, No> for each kind, so that B* + B*' B'1' + * * = 1. With crystal surfaces, there should be a simple whole number relationship between the numbers of the various kinds of spaces, so that the adsorption could be regarded as the sum of several adsorptions on different surfaces. In t ^ „ aces are very likely all unlike. The surface may be considered to be di vided into infinitesimally small fractions, dB, each space having its corresponding value of C • Integrating, where c r is a function of B. This equation, however, is not a practical one to apply without having a greater knowledge of amorphous sur faces than is now possessed. Most of Langmuir*s experiments were carried on at ex tremely low pressures, approximating a few bars. This condi tion of pressure is one seldom encountered in "practical1 1 work. Condensation and capillary effects are disregarded. In fact, they are not apt to be present at the pressures used. Langmuir admits that they may exist at higher pressures. Then, the equation can be written LANGMUIR*S WORK DONE AT LOW PRESSURES 27 LANGMUIR'S EQUATIONS FOR SMOOTH PLANE SURFACES ONLY Langmuir worked with plane surfaces. He did not try to cover adsorption by porous bodies such as charcoal or silica gel or even roughened plane surfaces. His experiments were performed, in all eases, with materials prepared to give smooth plane surfaces. Every precaution was taken to prevent surface oxidation or any other accident which might roughen the surfaces used. 16 G. H. Latham found that very few surfaces could be considered plane. He worked on the assumption that, if the "apparent surface" divided into the quantity adsorbed showed a surface layer many molecules thick, then the surface was not a true plane but much roughened, very much as described by Donnan.^ He found polymolecular layers on most of the ordinary surfaces used, the two apparently absolute plane surfaces (showing monomolecular adsorption layers) used by him being electrically amalgamated silver and soft glass thoroughly melted and cooled in the presence of dry air. These two surfaces gave monomolecular layers with both polar and non-polar substances. One interesting feature of Latham's work is that he found monomolecular adsorption layers with freshly annealed G. H. Latham, "The Thickness of Adsorbed Vapor Films.", Jour. Amer. Ohem. Soc., 50:29^7> 1J2&. 17 Of. ante, "The Nature of Surfaces.", p. 11. 23 glass surfaces and layers, apparently, about fifty molecules thick with thoroughly washed or acid treated glass. Langmuir ascribed the apparent adsorption on glass of a polymolecular layer to solution in the glass. Latham’s work leads one to believe that a roughened surface layer rather than solution is the cause of this phenomenon. However, adsorption on an extremely roughened surface seems to have many of the charac teristic aspects of true solution. How, practically all commercial adsorbents are either porous or have activated (roughened) surfaces. Thus, the actual wpracticalw application of Langmuir’s equations is rather small. However, his theory is tremendously important in explaining the “specific action" of adsorbents and the mechanism of adsorption and catalysis and is, therefore, a very important contribution to chemical theory. POLAHYI’S THEORY Polanyi postulates a region of adsorptive forces bounding a solid adsorbent.^9 There is a sort of atmosphere of adsorption. The adsorbed molecules are densely packed at i p i The Svedberg, Colloid Chemistry, A. C. S. Monograph (Hew York: The Chemical Catalog Co.,1928) 2nd. ed., p. 163. E. o. Kraemer, A Treatise on Physical Chemistry, edited by Hugh S. Taylor,“(Hew York: Van Hostrand Co., 1931) II, Chapter XX, p. l661 4 - . 29 the surface and the density of the adsorption layer varies continuously outward until the density of the surrounding medium is reached. The idea of adsorption , , potentialH is used. The ad sorption potential is the energy required to transport a mass unit of the adsorbed material from a particular point in the adsorption layer outward to infinity. Thus, the potential decreases gradually outward from its maximum at the surface. The potential at a given point is independent of the temper ature, pressure and the presence of adsorbed molecules. How ever, the forces acting are specific for a given interface, that is - the strength of the field varies with the particular sample of adsorbent and the material adsorbed. Using van der Waal's equation and, for very low pres sures, Boyle's Law, and the adsorption potential, Polanyi gets rather close checks in his results. The potential dis tribution is determined from a known isotherm and it is then used to predict other isotherms corresponding to other con ditions. The use of the gas laws is presumed to be valid and they are not supposed to be affected by the adsorption forces acting. His equations are energy equations. He is less con cerned with the actual mechanism of adsorption than with the energy changes involved. In many cases, his equations give very accurate results. 30 McBain2^ seems to think that the Polanyi formulas are to some extent empirical. The evidence for the thick compressed film of Polanyi is weak. The unknown constants of the formula based up on the assumption of a multimolecular film are determin ed by using some of the actually measured sorption data; the expression then fits the remaining sorption data at other temperatures, etc. However, any one of the other theories will do as much. He says that thick films are generally calculated from experiments using tarnished, oxidized or otherwise roughened surfaces where the computation of the actual sur face area is impossible. THE FREUNDLICH EQUATION Freundlich worked out an empirical equation which seems to cover a great number of cases of adsorption. x = k (p)l/n m x « the amount of gas adsorbed m = the weight of the adsorbent p » the gas pressure at equilibrium k and n are constants specific for each particular case of adsorption For adsorption from solution, the equation has been modified. J. W. McBain, wTheories of Adsorption and the Technique of Its Measurement.w, Nature, 117:551, 1926. where c is the equilibrium concentration of the solute. Tho it is a purely empirical thing, this equation does very well in covering a range of practical applications. In its logarithmic form, it becomes log x j. log k + l/n log p m ~ This is an equation giving a straight line. The slope of the line is given by l/n while k is the amount adsorbed when the pressure is unity, l/n varies downward from unity to a quite small fraction. This equation is applicable over small ranges of con centration change only. It cannot be used to predict adsorp tions except at quite close to the concentration at which the constants are determined. Many modifications of this equa tion have been tried, but no form has been discovered to cover the whole range of concentrations. PATRICK - ADSORPTION AND CAPILLARITY Patrick and McGavack^, working with silica gel, found very little evidence of specific action of the adsorbent. The amounts of liquid adsorbed were quite high, in spite of P I W. A. Patrick and John McGavack,MAdsorption of Sul fur Dioxide by Gel of Silicic Acid11, J. Amer.Chem.Soc., *1-2: 946, 1920. ------------------ 32 the fact that silica gel is rather inert chemically. Also, he found that equal volumes of liquid were adsorbed, regard less of the liquid used. These results led them to believe that adsorption was purely “physical” and that the adsorption of gases and vapors might be predicted from a knowledge of the physical constants of the gases or vapors alone. There has been worked out an exponential equation which covers the work with silica gel rather well. V = k ^ppr^1/11 V = the volume of the condensed liquid phase uncorrected p * the pressure of the gas phase pQ= the vapor pressure of the liquid phase <f - the surface tension of the liquid phase k and n are constants H. K. Chaney22 says that Patrick is wrong in neglect ing the specific action of the adsorbents and that his con clusions are due to the fact that he worked with silica gel, which has a very low specific action for any substance and which acts almost entirely because of its pores, that is - by capillarity. pp N. K. Chaney, American Institute of Chemical Engin eers, Transactions, 15:292, 1923» 33 McBain, in commenting on Patrick*s theory, makes two very pertinent points, namely - that Patrick*s theory ignores the fact of adsorption on plane surfaces and also that it fails to account for the adsorption of the layer of molecules of liquid in actual contact with the solid adsorbent. He says, also, Furthermore, calculation from the well known thermo dynamic formula, connecting vapor pressure with pore diameter for a liquid in a capillary, shows that the greater part of the adsorption is observed at pressures so low as to correspond with pores of diameters compara ble with molecular magnitudes. Hence, even from this; one may argue that here true adsorption is responsible because the adsorbed molecules are all in direct contact with the molecules of the surface. Of course, whenever a surface covered with a monomolecular film is shaped as a pore and exposed to nearly saturated vapor, liquid may condense in such a capillary. 23 ADSORPTION BY POROUS ADSORBENTS AKIN TO SOLUTION In the case of adsorbents like charcoal, where the pores or interstices are of molecular dimensions or slightly larger than molecular, the adsorbed molecules may be evenly distributed thruout the mass of the adsorbent. In such a case, we have a state of affairs closely approximating true solution of the substance in the porous adsorbent. The idea of solutions, containing “solvated" solute molecules or ions, is generally accepted. These solvated particles are surround- ^ J. W. McBain, loc. cit. 3^ ed by a group of solvent molecules attached to them by secon dary valence forces. Certainly, this appears to be a thing closely related to adsorption. If one imagines these solvated aggregates forming chains of "dissolved1 1 molecules, these molecules being actu ally bonded one to the other, but retaining in the inter stices between the chains formed by them a large portion of the solvent molecules, then one should have a rather good picture of the state of affairs in a piece of saturated por ous adsorbent, In many quarters, this view is now held and it seems a rather logical one. CHAPTER III THE STRUCTURE OF THE CARBON MOLECULE AND OF CHARCOAL All the various forms of carbon are made up of units containing more than one atom. The high volatilization tem perature of all the forms indicates a molecular set up. The distinctly different physical properties and the gradation of intensity of chemical properties among the various forms indicate a variation in molecular arrangement. Finally, Debuye and Scherrer’s X-ray patterns indicate the existence of molecules. Kekule^, in 1299 > considered the carbon molecule to contain twelve atoms. Barlow and Pope^, in 1906, suggested the arrangements of atoms shown in sketches (a) and (b).^ Note that while (b) implies separate molecules, (a) would allow a continuous linkage thruout the carbon mass. Dewar**’ , in 1902, due to the formation of melletic racid, ^ Maurice Copisarow, "The Allotrophy of Carbon." Chemical Hews, 112:301, 1919. 2 ILid., p. 301. 3 All diagrams referred to on this page and those of this chapter are shown on page 3^* i f W. D. Bancroft, "Charcoal Before the War.", Jour. Phys. Chem., 2^:201, 1920. 36 O^COOOH)^* from charcoal (see formula e ), conceived the idea of the double hexagon as shown at (d). Aschan5, in 1909» put forward the continuous linkage of sketch (f), but this does not satisfy the carbon valence of four. In 1913» fk® Braggs^, from X-ray patterns, proposed the type shown in (g), atoms being located at each corner and the center of each face of the cube. This arrangement might be fitted into a continuous space lattice. From Laue diagrams, Debuye and Scherrer7 concluded that diamond and graphite are the only distinct forms of car bon, amorphous carbon, according to them, being a form of finely divided graphite. Copisarow^ remarks that their unification of graphite and amorphous carbon seems untenable, since no matter how finely graphite is divided, it still retains different chemi cal properties than those of amorphous carbon. Acheson*s Aquadag, colloidal graphite, still shows the characteristic chemical behavior of graphite which differs from that of the amorphous types of carbon. 5 Maurice Copisarow, 0£. oit., p.301. ^ . . . . , oj>. cit., p. J01. « « • « , op. cit., p. £ > . . . . , oj). cit., p. 302. 37 Copisarow assumes a "basic valence of four for carbon and considers structural possibilities on the basis of link age between atoms. Type I Non-rigid molecular configuration, some val ences of which are free. See diagrams (h) and (i). Type II A rigid configuration with some free valences. See diagram (j). Type III A rigid configuration with all valences satis fied. See diagrams (b), (g) and (k). Thus, on theoretical grounds, he arrived at three classes of carbon linkages. Also, he points out that of the three classes of carbons, which have been recognized in the past, each has its own different heat of combustion, tho the amount of carbon dioxide produced is the same in each case, weight unit for weight unit. Thus, the difference in energy liberated must be due to a variation in the arrangement and strength of bonding between the atoms of carbon making up each type. Langmuir*s ideas^ support Copisarow*s theories of amor phous carbon. He says that the fibers of cellulose of which charcoals are made ’ 'consist of almost endless chains of car bon atoms attached to hydrogen and hydroxyl groups.” 9 Irving Langmuir, "Constitution of Solids and Liquids" Jour. Amer. Chem. Soc., 3&:22S6, 1916. 32 \ J \ / — c— c— c— — c— c _ / J \ A b^ c_ V \ yi v v- V — c — — c — 'c-- COOH flooc- COOH HOOC- -coori COOK — c --'I Fig. I ~ Carbow S-frnclur«g. 39 H H H H H H H H - C - C - C - C - C - C - C - O- OH OH OH OH OH OH OH OH "When charcoal is formed, hydrogen and oxygen are driven out, leaving chains of carbon atoms with occasional cross linkages. The porosity, thus, undoubtedly extends down to atomic dimensions." Thus, Langmuir arrives at the Type I of Copisarow. See diagram (h). According to Latimer and HildebrandlO, diagram (g) is the accepted arrangement for the diamond and (1) for graphite. For interatomic linkage, both (g) and (l) may be resolved in to (c). It will be noted that where (g) may be extended in three dimensions, (l) is capable of extension in two dimen sions only, which leads to the formation of flat graphite crystals. These forms coincide with Copisarow*s Type III, but Type II would require something like (d) or possibly (a) to fill its requirements. THE "DRIFT" IN ADSORPTION AND IN DENSITY DETERMINATIONS Howard and Hulett^ found, that in determining the density of charcoal, there was a variation in results with ® W. M. Latimer and J. H. Hildebrand, Reference Book of Inorganic Chemistry, (New York: The Macmillan Co., 1930)» pages 212 and 376. ^ H. 0. Howard and G. A. Hulett,"Study of the Density of Carbon." Jour. Phys. Chem., 2S;10S2, 192*1-. k - 0 the liquid employed. They found, also, a change in density with a change in the time of immersion, that is - a drift oc curred. They concluded that the variation in results with the liquid employed was due to molecular size, that is - the smaller liquid molecules were able to penetrate capillaries too small for the larger molecules. The 1 1 drift" was ascribed to the time required for the liquid molecules to penetrate to and thru the innermost capillaries. ip Firtfcrc observed that the bulk of the material adsorb ed on charcoal was taken on quite rapidly, but that a gradual increase in the quantity adsorbed continued for years. Oude and Hulett^-3 record a similar drift in adsorption. These results substantiate the ideas of charcoal struc ture given in previous sections, particularly as to the exis-. tence of tremendous numbers of tiny capillary passages. It is conceivable that a liquid might have to take a very long period of time to penetrate thru to the innermost of such a maze of tiny pores with the resulting "drift" in adsorption. J. B. Firth, "Sorption of Iodine by Carbon." The Faraday Society, Transactions, iSikjkj 1921. ^ h . E. Cude and G. A. Hulett, "Some Properties of Charcoals.", Jour. Amer. Chem. Soc., 42:391» 1920. k - 1 THE ALPHA AND BETA CARBONS OF N. K. CHANEY Chaney postulates the existence of two forms of carbon other than graphite or diamond. He divided amorphous carbons into two. classes, alpha carbon or active carbon capable of catalytic action and beta carbon or inactive carbon which is relatively inert. He states that carbonization at a temper ature below 600°-700°C produces a carbon which may be acti vated while carbonization at higher temperatures produces an 14 inert carbon incapable of activation. He does not claim actually to have proved the existence of two allotropic modifications of amorphous carbon. He says, It would be premature to assert that these two forms of carbon are true allotropic modifications. It is not yet established that both forms are amorphous .... This much is established, the two forms are characteris tically distinct and easily differentiated both by their properties and conditions of formation.15 He states that alpha carbon is easily oxidized while beta carbon is oxidized with more difficulty. Beta carbon seems to be composed largely of a graphite like material* It is referred to as “amorphous graphite”. Chaney finds that crystalline forms of carbon, excepting the diamond, ik N. K. Chaney, “Activation of Carbon.", American Electrochemical Society, Transactions, 36:91* 1919* 15 J N. K. Chaney, A. B. Ray and A. St.John, "The Properties of Activated Carbon Which Determine Its Industrial Application.", American Institute of Chemical Engineers, Transactions, 15:316, 1923. k-2 tend to fall into two groups, 1 1 graphitic" and Mpseudo-graphi tic” by all X-ray and chemical evidence available. Normal graphites are sharply crystalline, have a characteristic X-ray pattern common to all and give graphitic acid by the Brodie reaction. Pseudo-graphites, also, are definitely crystalline, but have a distinctly modified X-ray pattern and seem to yield graphitic acid with difficulty. The proportions of these forms present may be measured by X-ray means within five to ten percent. Pseudo-graphite has been found to have from five to ten times the electrical resistance of true graphite. It is stated that all forms of amorphous carbon are converted to one of these two types on heating to quite high temperatures. Heating active alpha carbon gives a distinct pseudo-graphite while beta carbons yield almost pure normal graphite. These facts support the classification of amor phous carbon into alpha and beta groups. Quoting the work of W. B. Dexter; .... he has started with normal graphite, produced graphitic acid, reduced it to finely divided carbon using mild reducing agents and reconverted it to normal gra phite by electric baking at usual temperatures. By a more violent decomposition, he has obtained a finely di vided carbon which seems to yield pseudo-graphite when treated in exactly the same manner as the first carbon. Since these changes of graphitic acid to two forms of graphite take place in the solid phase, they are particu larly instructive, indicating that a partial disintegra tion gives beta carbon, restored to graphite by heating, ^3 but that a more vigorous disintegration gives alpha car bon and pseudo-graphite on heating.1» This suggests that there is an element of complexity in the beta carbon which determines its crystallization as graphite. It, also, suggests that beta carbon is more com plex than alpha carbon. In support of Chaney's ideas, the work of Ruff and Mautner may be noted.1? They obtained a decrease in adsorp tive power with an increase in the amount of graphite present according to space lattice evidence by X—ray. Chaney states that alpha carbon probably is entirely amorphous and of a loose open structure, possibly having a large proportion of unsatisfied valence bonds by reason of the loosely packed arrangement. This view of Chaney's seems entirely logical. It will be noted that his alpha carbon could easily coincide with Copisarow's Type I and with Langmuir's idea of charcoal struc ture. It seems logical that the alpha modification could be produced by a temperature, which would decompose cellulose without causing a great disturbance in the original arrange ment of the carbon atoms in the cellulose chains. The origi nal decomposition would leave twisted chains of carbon atoms 16 P« 312. lg Ruff and Mautner, "Active Charcoal: Amorphous Nature.", Ohem. Abstracts, 22:281*4-, 1928. I t f . with myriads of interstices of molecular dimensions or very slighty larger. According to Langmuir-1 -^, atoms and molecules do not affect each other very strongly at distances greater than the order of 10 cm. Therefore, it is reasonable to assume that the greater portion of these carbon chains would remain as such. Only at contact points would there be a chance for a three dimensional rearrangement into a crystalline structure* Further, Chaney*s assumption that, at higher, tempera tures, the volatile products of cellulose decomposition aid in forming beta carbon seems quite reasonable. He assumes that, at higher temperatures, the volatile decomposition pro ducts are themselves “cracked”, depositing an inactive form of carbon, beta carbon. A slight modification of these ideas gives a very log ical concept. If it is assumed that this process of cracking results in the deposition of carbon atoms from the volatile material in the interstices between the original carbon chains, it is conceivable that numerous groups of carbon atoms thus, are brought into close enough contact to lead to the formation of the space lattices of many graphite nuclei. The formation of these nuclei would produce a material which ^9 Irving Langmuir, *Constitution of Solids and Liquids” Jour. Amer. Ohem. Soc., 3*5:2221, 1916. would be very close to Chaney's beta carbon. Briggs of E d i n b u r g h ^ ® takes exception to the alpha and beta classification. He then ascribes the difference between active and inactive carbons to the difference in the "degree of polymerization". It is quite apparent that, in reality, he objects to the new nomenclature only as his statements imply quite the same things as do Chaney's. 20 Henry Briggs, The Royal Society of London, Transactions, 1001921. CHAPTER IV THE MANUFACTURE OF ACTIVATED CARBON CHANEY*3 PRIMARY CARBON Of special interest in the activation of carbon, is the work of N. K. Chaney and his associates. It is Chaney who first stressed the temperature of the actual carboniza tion as the controlling factor in making activated carbon. To him goes the credit for the concept of activatable or '•primary*' carbon as active or alpha carbon covered by a uniquely stable adsorption complex of hydrocarbons. These hydrocarbons are very strongly attached to the surface. When a carbon containing material such as wood is car bonized, volatile hydrocarbons are formed. If the tempera ture is kept low enough, generally below 500°“600°C, a large part of these hydrocarbons leave the charred material as gases. The balance of them are strongly adsorbed by the car bon surfaces, giving an exceptionally stable adsorption com plex. This carbon with its adsorbed hydrocarbons is the "primary carbon". It is readily activated. The formation of this complex accounts for the rela tively low activity of ordinary charcoal. The presence of N. K. Chaney, "Activation of Carbon.", American Electrochemical Society, Transactions, 36:91t 1919* 4 - 7 this complex is used to explain the difference between ordi nary charcoal and carbon made from carbon monoxide or other non-hydrocarbons. The latter is quite active without further treatment. Chaney offers as a proof of the formation of this com plex the fact that chlorine passed over primary carbon gives halogen derivatives of hydrocarbons. ACTIVATION DEFINED Chaney must be credited with giving us an idea of what activation really is, namely - the breaking up of the hydro carbon adsorption complex and the removal of the hydrocarbons from the mass. This leaves the carbon bare and ready for useful adsorption. "SELECTIVE AND LIMITED OXIDATION" The hydrocarbon complex is broken up, partly by dis tillation, but largely by the chemical action of air, carbon dioxide or steam. The temperature is regulated to produce a reaction with the hydrocarbons, but not with the carbon beneath. With air, the temperature used is from 350°-^00°C, while with steam, it is kept at from S50o~ H 00°C.^ p F. Bonnet, Jr., "Activated Carbon, Its Evaluation, Manufacture and Uses.", Ohem. Age, N. Y., jL» 327, 1923. 4S There seem to he two useful stages in the oxidation. The first stage consists of a 1 1 selective1 1 or ’ 'differential' 1 oxidation during which the activity of the carbon changes as the adsorbed hydrocarbons are eliminated. The adsorptive power increases at a rate all out of proportion to the in crease in porosity as indicated by density determinations. The second stage is one of "limited” oxidation in which the carbon surface is attacked with an increase in por osity and available surface approximately proportional to the increase in adsorptive power. Further heating must cause a loss of available surface by oxidation of carbon, as it produces a drop in adsorption ■ % per unit volume of carbon.y ACTIVATION BY AIR, CARBON DIOXIDE OR STEAM ? The use of air for activation has the advantage of lower temperature. However, the reaction is exothermic and even with careful regulation,"hot spots"are produced in the carbon mass. This causes cracking of the hydrocarbons and the formation of graphitic carbon which is almost impossible to remove. It has been found that the surrounding active 2 N. K. Chaney, A. B. Ray and A. St.John, "The Pro perties of Activated Carbon Which Determine Its Industrial Use.", American Institute of Chemical Engineers, Transactions, 15; 309 > 1923* 1 1 9 carbon is oxidized first. Steam and carbon dioxide require higher initial tem peratures. This introduces difficulties in finding the pro per materials for constructing the apparatus used, it has the advantage of giving an endothermic reaction which can be controlled more easily. Steam is more or less generally accepted as the best • ‘oxidizing1 1 medium. EARLY WARTIME ACTIVATION According to Dorsey**", the government chemists prepar ed activated carbon for gas masks in April, 1917 by an ini tial distillation at from S>50°-900°C, followed by an “air treating' 1 at 350o-**-00°0» Thus, they divided the process into two parts, the initial distillation followed by acti vation. In the latter part of 1912> the National Carbon Co. replaced the “air treating" with a steam treatment at 900°C. This method of activation gave the best results. The carbonization and activation were carried on in iron cylinders or catridges placed in holders in a furnace. The cylinder caps were tapped to permit the passage of air, etc., thru the carbon during the process. It is interesting to note high initial temperatures, considering later work. * * " Dorsey, F. M. , “Development of Activated Charcoal." Ind. Eng. Chem., 11:231, 1919- 50 FAILURE OF ANTHRACITE ACTIVATION IN 191S According to Chaney5, the government chemists failed to activate anthracite because of the high carbonization temperatures used by them. As the hydrocarbons were distil led off, a large portion of them were adsorbed on the active carbon surfaces. The high temperatures and uneven heating of the retorts caused “hot spots" and the formation of in active carbon on the surfaces. Then, what should have been a "selective" oxidation of adsorbed hydrocarbons, actually became an oxidation of active carbon, leaving the harder graphitic carbon unchanged. By careful control of the temperature, the steam con centration and the circulation of the hydrocarbon vapors, Chaney has produced fairly active carbon from anthracite. Apparently, Chaney*s ideas on the control of carbon ization temperatures, first published in 1919> took quite a while to spread or else were skeptically received. Ardagh, in 1921, directs that carbonization take place at the high est possible temperature with prolonged heating.^ This is a part of his summary of the available knowledge of carbon activation up to that year. 5 Chaney, Ray and St,John, op. cit., p. 309* 6 E. G. R. Ardagh, "Activated Carbon.", Jour. Soc. Ohem. Ind., lK>:230T, 1921. 51 Firth, in 1923» quoting from a committee report to the Royal Society, states that their work was done with chars carbonized at near white h e a t .7 METHODS OF ACTIVATION Chaney divided all methods of activation into four classes. I High temperature calcination. II Impregnation of carbonaceous materials with minerals before carbonization, the minerals being removed later in \ the process. Ill Removal of hydrocarbons by special solvents. IV “Selective” and "limited” oxidation. TYPE I HIGH TEMPERATURE CALCINATION ITS RELATION TO AIR ACTIVATION Ray states that it is claimed that a long calcination at 650°C produces some active carbon from primary carbon when the calcination is carried out in a neutral atmosphere. He says that the neutral atmosphere is not necessary and really does not exist.^ According to him, such activation as occurs ^ J. B. Firth, ‘ ‘Sorption Activity of Carbon.“, Jour. Soc. Cham. Ind., ^2:2^2T, 1923- A. B. Ray, “Manufacture of Activated Carbon.”, Chem. Met. Eng., 2S:977> 1923* 52 is due to the oxygen of the air held in the pores of the car bon or.; to air which enters the apparatus during the process. Both Chaney and Ray assert that the use of very high temperatures alone actually destroys some active carbon due to the deposition of carbon from cracked hydrocarbons and the formation of graphitic carbon as a result of this deposit. In support of Ray's oxidation ideas, the work of Philip and his associates may be quoted. The conditions under which charcoal is heated can be widely modified, air being excluded in varying degrees according to the method of packing and covering, and it was proved conclusively that the greater the facilities for the access of air to the heated material, the great er is the decrease in bulk density and the greater the increase in adsorptive power. . . . . that the degree of activation is intimately con nected with the extent to which charcoal is oxidized during heating is strengthened by experiments. .... the heating of charcoal results in widening, by progressive oxidation, of the capillary channels with which the granules are riddled. In this way, tho the external volume of a granule is practically unaltered, its bulk density is diminished and the effective surface is enormously increased.9 They found that longer heat treatments seem to cause increased activity. Page has done some interesting work on the rate of oxidation and activation. He carbonized and treated material 9 J. 0. Philip, S. Dunhill and 0. Workman, ’ 'Activation of Wood Charcoal by Heat Treatment." Jour. Chem. Soc., 117:362, 1920. 53 at 400°C and found: That varying the ratio, 02/^2* in a mixture of these gases, gave varying oxidation rates. That charcoals prepared in a fast stream of gas were less active than those from a slow stream. That there was little or no variation in activity if the same quantity of oxygen per unit weight of charcoal was used, provided the rate of furnishing the gas was kept con stant and the same in all trials. That changes in activity were observed with equal quantities of oxygen furnished at different rates. That for a given temperature and rate of flow, the activity diminished with an increase in the quantity of oxygen used.*® This last item does not seem to check with the work of Philip and others previously mentioned, but if it is realized that the oxygen percentages ran higher than those of air, the results are not irreconcilable. The furnishing of oxygen at too rapid a rate and in higher concentrations, no doubt, causes the oxidation to go on in spots near to the surface which are heated by the rapid reaction,. The oxygen is very likely used up before it can penetrate far into the granules. *® A. B. P. Page* “Activation of Wood Charcoal by Progressive Oxidation in Relation to Bulk Density and Iodine Adsorption.", Jour. Ohem. Soc., 130:1^76, 1927. 5^ Also, it is very likely that quite high temperatures are reached, causing a fusion into graphitic carbon about the reaction centers. In this connection, it may be noted that Firth, work ing with cocoanut charcoal, carbonized at the lowest possible temperature and activated by heating at from 600° to 900°C, found that activity increased with the temperature of treat ment within certain limits.^ However, he found that “too much heating1 1 increased density and decreased activity. This might be explained on the basis of “fusion11. TYPE II IMPREGNATION WITH CHEMICALS This type of carbon has been given a tremendous amount of attention. The presence of the chemicals is secured, either by using raw materials containing them or by impreg nation of the carbonaceous matter before carbonization. Boneblack, for a century the “only* 1 decolorizing car bon, is an example of the first type while most of the decol orizing vegetable chars, now on the market, are of the impregnated variety. In general, the process of making the second type con sists of: 11 J. B. Firth, “Some Factors Governing the Sorptive Capacity of Charcoal.", Jour si Ohem. Soc., 119:926, 1921. 55 Impregnation of the carbonaceous matter with the desir ed chemical. Carbonization and calcination. Removal of the chemical by washing. The effectiveness of the various chemicals will be discussed later. In attempts to imitate boneblack, carbonaceous matter has been made into a paste, mixed with lime, silica, etc., and calcined. The results have not been very satisfactory. TYPE III REMOVAL OF THE HYDROCARBONS BY SOLVENTS This method of activation has not shown very satis factory results* The great weakness of the method lies in the fact that the solvent has to be removed as well as the hydrocarbons. Thus, a solvent capable of thoroughly dis solving the hydrocarbons and very poorly adsorbed by the car bon would have to be found. According to Ray, the use of SeOClty has been patented, but no report has been made as to its effectiveness. TYPE IV "SELECTIVE" AND "LIMITED" OXIDATION This method has been discussed already in some detail. Briefly, it consists of carbonization at from 500°-600°C or lower and the removal of the hydrocarbons formed by the chem ical action, generally of steam, at about 900°C. 56 According to Bonnet, a treatment of the finished car bon with acid to remove ash, followed by a washing with water and a thorough drying, gives a more active carbon.12 Ray states that, in activating the carbon, steam must be in excess and the products of the oxidation be rapidly conducted away. The complete removal of the hydrocarbons can only be accomplished by a very high concentration of the oxidation material. As a particular advantage of the process there is the fact that by varying the steam concentration and temperature, the character of the carbon may be varied at will. By control of these factors, it is possible to make activated carbons to exact specifications and to dupli cate previous runs in their manufacture.^-^ Kosakavich and Ismailov have worked with temperature and time of steam treatment. They found a slow but continu ous increase of activity with the time of treatment between 700°0 and 2>00°0. At £>50°C, the maximum activity was reached after a treatment of fifteen minutes - at 900°C, in half that time. The person who abstracted their article (the original is in Russian) failed to give the quantity of steam used. They found that the slower the rate of furnishing the steam, the lower the activity of the carbon obtained. Also, 12 F. Bonnet, 033. cit. , p. 322. A. B. Ray, o£. cit., p.977* 57 graphitization and activation go on concurrently, the velo city of graphitization increasing rapidly between 2>50°C and, 900°C.lij' SOME MECHANICAL FACTORS The materials should be carbonized and activated in thin layers. This is particularly important in carboniza tion. Thick masses will be unevenly heated and the hydro carbons from one portion of the char will be adsorbed by other portions above them, making activation more difficult. A current of inert gas should be passed to carry off the volatile products of the carbonization. A rotating, inclined cylinder, possibly with blades set lengthwise in it, enclosed in a furnace to maintain the proper temperature, seems to be indicated by the conditions set forth above. PROPERTIES REQUIRED FOR GAS AND VAPOR ADSORBERS Chaney gives three essential characteristics of a good gas adsorbing carbon,^ Good mechanical strength. P. P. Kosakevich and N. A. Ismailov, ’ ’Activation of Charcoal by Steam.”, Ohem. Abstracts, 23:^7S0, 1929* •*•5 ohaney, Ray and St.John, o£. cit., p. 33S. 5^ Close approximation of a definite optimum density. High intrinsic activity. The necessity for the possession of sufficient mechan ical strength to avoid crushing, packing and dusting, when the carbon is used in gas adsorbing towers, is obvious. The efficiency of gas adsorbing equipment is based upon adsorption per unit volume of carbon as the space occu pied is an important factor. This makes for a definite ideal porosity which corresponds, according to Ray, to a density of 0.66 . In the purification of gases, the specific adsorption of the carbon, as defined by its retentivity, is the most important property. Adsorbed impurities taken on must not be released into the gas stream later. This selective ac tion must be carefully determined, because carbons sI kht marked tendencies to take on some substances before others. A carbon may have a low retentivity for a compound and yet take up considerable quantities of it by capillary action. When capillarity is involved, almost invariably the quantity of vapor which can exist in equilibrium with the adsorbed substance is too large to give satisfactory puri fication. Such an adsorption is too easily reversed. Where adsorbed vapors are to be recovered, a high retentivity is not a hindrance, as it rapidly diminishes as a certain critical temperature is approached and the volatile 59 compounds are easily recovered. Wood chars made by selective oxidation and the vari ous chars made from impregnated materials make poor gas adsorbers* The first have low densities and the others have too little selective adsorption capacity, that is - too little active carbon. Very good gas adsorbers of the proper density have been made from coals by crushing the carbonized mass and briquetting before activation.^6 Burrell says that charcoal given fifty minutes of lim ited oxidation (burning away of actual carbon) is best for gasoline vapor recovery, because of the large pores which are formed. These are of the right size to admit the gaso line molecules. In addition, the char must show a selective action for the desired hydrocarbons and not hold the smaller molecules.^ Both Chaney and Ardagh stress the fact that fine gran ules are the most efficient adsorbers,- but in gas adsorption the size is limited by the necessity of avoiding the packing of the char , which would block the passage of the gas.^^ 16 A. C. Fieldner, R. E. Hall and A. E. Galloway, Study of the Production of Activated Carbon from Various Coals' and'"either Materials, U. S. Bureau of Mines, Paper ^79• ^ G. A. Burrell, ’ 'Gasoline by the Charcoal Adsorp tion Process.", Chem. Met. Eng., 2^:156, 1921. lg E. G. R. Ardagh, o]D. cit., p. 23OT. PROPERTIES REQUIRED FOR DEOOLORIZERS Bonnet gives as the properties of an ideal deodori zer: High adsorptive capacity. No acid or water soluble content. Easily filterable. Tough enough to keep handling loss to a minimum. Easily and cheaply revivified.^9 Most of these are perfectly obvious, but the third would indicate that Bonnet was thinking in terms of a single process. Chaney says that the nature of the carbon used depends upon the work to be done. He names as factors to be consid ered: Size of the particles to be adsorbed. Viscosity of the liquid. Nature of the methods for the recovery of the adsorb ed material or the revivification of the carbon. The engineering methods of handling the carbon. The degree of dispersion of the carbon used depends upon the size of the particles to be adsorbed. For molecu lar adsorption, a dense carbon in granules like the gas F. Bonnet, ojd. cit. , p. 330* 61 carbons is satisfactory. Suck carbons will adsorb from sol ution in proportion to their activities as determined by gas adsorptions, provided that the solution is not too viscous. The carbon need not be finely ground. With an increase in particle size or viscosity or both, the carbon must be more dispersed, that is - be in a smaller state of division or have larger pores. Ray says that Chaney*s active carbons (prepared by selective oxidation) make excellent decolorizers. Carbons prepared by impregnation methods are good decolorizers, as a rule, because of their porosity. Chaney suggests that porous granules are better than very finely ground dense carbons, because of their better filtering qualities. Such carbons are soft, but as long as they do not disintegrate too far with handling, they are satisfactory. However, some of these carbons will form colloidal material and" slime**the filters under adverse conditions. The type of carbon to be used cannot be considered apart from the engineering methods of handling it in the par ticular process under consideration. Mechanical stremgth and structure are as significant as the content of active carbon itself, in determining the suitability of a particu lar decolorizer in industrial applications. CHAPTER V INFLUENCE OF SUBSTANCES OTHER THAN CARBON As early as 1S22, Bussy, Payen and Desfosses^ impreg nated animal and vegetable materials with potassium carbon ate, carbonized them and washed out the inorganic compounds afterward. In this manner, they got very good decolorizers, but their work was not followed up by the industries. 0 Bancroft mentions earlier work by many investigators. Among others, Stenhouse impregnated with aluminum sulphate and Richter moistened charcoal before calcining. Both report improved decolorizing ability. MORE MODERN WORK Schneller^ and Zerban^" give lists of patents granted for the manufacture of decolorizers. These patents cover a wide range of raw materials. Some contain minerals and others M. A. Schneller, "The Vegetable Decolorizing Carbons” Louisiana Planter, :15^> September, 1917* p W. D. Bancroft, "Charcoal Before the War.", Jour. Phys. Chem., 2^:201, 1920. 3 Schneller, loc. cit. F. W. Zerban, E. C. Freeland and D. D. Sullivant, Studies on Preparation of Vegetable Decolorizing Carbons for the Cane Sugar Industry'TLouisiana State Experiment Station, Bulletin 167» Louisiana State University, Baton Rouge, La.) pages 5 and. 6. are impregnated with various acids, bases or salts. Zerban, in an attempt to get relationships between the inorganic substances used and decolorizing power, has treated sawdust with a long list of chemicals.-* He has test ed the resulting chars for decolorizing power against a standard molasses solution. These chars were standardized against Norit as 100. Zerban made a number of very efficient chars. Magnesium hydroxide, calcium oxide and zinc bromide gave chars about four times stronger than Norit. Zinc chlor ide gave one approximately five times stronger, while magne sium chloride with an excess of ammonium chloride completely decolorized the test solution and rated a score of 3200. In making up the first char, one mole of calcium oxide was mixed with an equal weight of sawdust of the long leaf yellow pine. In the other experiments, the chemicals were used on an equivalent basis, that is - two equivalents of each of the compounds used were mixed with the same weight of sawdust as was used with the calcium oxide. The chemicals were wetted and boiled down to dryness with the sawdust. After carbonization, the chars were treated and washed to remove minerals before they were tested. Schneller particularly mentioned zinc chloride as a good impregnating agent. Three parts by weight of zinc 5 IMd-» P- 9 Sk chloride to one of sawdust were mixed thoroughly with excess water and hoiled to dryness. The resulting mass was carbon ized, gently at first, and then heated up to the distilling point of the zinc chloride. The char was treated with hydrochloric acid to take care of any zinc oxide or carbon ate formed and the zinc chloride was removed by washing with water. Analysis showed no trace of zinc left in the carbon. The expensive zinc chloride was almost entirely recovered. It could be concentrated for reuse.^ He, also, described his work with rice hulls. He carbonized the hulls at red heat in a closed retort and ob tained a char containing about fifty percent silica. Its activity was low, but treatment with sodium hydroxide solu tion to remove some of the silica gave a more active carbon. He digested the carbon with solutions of varying strength. It was found that as the concentration increased up to twenty percent, the activity increased. Beyond that, there was no further gain. The best carbon obtained was about fifty percent better than Norit. ACTION OF THE MINERALS PRESENT There is considerable difference in the various ideas of what part these inorganic substances take in the activa- Schneller, op. cit., p. 157» 65 tion of the cartoons. Bonnet thinks that their presence causes the hydro gen and oxygen of the raw material to toe removed almost entirely as water with a minimum formation of hydrocartoons.7 According to Chaney, the chemicals either prevent the adsorption of hydrocartoons or else promote their decomposi- S tion without the formation of inactive cartoon. Ray asserts that they either prevent the formation of hydrocartoon complexes during carbonization or aid in "break ing down and eliminating the adsorbed hydrocartoons during calcination.9 There is a fair amount of agreement among these three. However, Philip says that the materials mixed with the char are not so effective except as they modify structure during carbonization.This seems rather vague. According to Swiderek11, charcoal activated toy car- ^ F. Bonnet, Jr., ‘ ‘Activated Cartoon, Its Evaluation Manufacture and Uses.”, Chem. Age, H. Y., 31:327, 1923* ° N. K. Chaney, “Activation of Cartoon.”, American Electrochemical Society, Transactions, 36:91j 1919. ^ A. B. Ray, “Manufacture of Activated Cartoon.” Chem. Met. Eng., 26:977, 1923. 10 J. C. Philip, S. Dunhill and 0. Workman, “Activation of Wood Charcoal toy Heat Treatment.", Jour. Chem. Soc., 117:362, 1920. "LI AA Swiderek, “Charcoal Activated toy Mineral Substances” Chem. Abstracts, 21:3506, 1927. 66 bonization in the presence of mineral salts, owes its acti vity to the extension of its surface by the fine division of the charred mass on the mineral bases. He found the activi ty of such chars to vary with the percentage of carbon in them. This last idea would seem to cover boneblack which is only 10-11$ carbon. The arrangement would resemble to some extent platinized asbestos in structure. However, in cases where chemicals are entirely removed from the char, it seems that they could have acted in this manner only to a limited extent, as there is no marked shrinkage in volume of the char when the minerals are removed. Incidently, there is some shrinkage when boneblack is revivified. The process 12 causes a reduction in mineral content. Zerban thinks that the dehydrating powers of the chem icals used are involved in the activation. -^3 He found, to a certain extent, correlation between the heat of hydration and the resulting char. He notes that those chemicals, which remain solid at the temperature at which calcination ends, give the best results. Apparently, their ability to take on water causes the decomposition of the carbonaceous W. D. Bancroft, ' ‘Charcoal Before the War.", Jour. Phys. Chem., 24:201. 1920. Zerban, Freeland and Sullivant, oj>. cit., p. IB. 67 matter without the formation of hydrocarbons. Nearly all the investigators note that such carbons are very porous and, possibly, a large part of the benefi cial effect of the chemicals lies in the fact that they do disperse the carbon to some extent, leaving a larger useful surface when they are removed. There is evidence for both views. The fact that these carbons generally lack specific activity, to a large degree, leads one to suspect that the chemicals do cause decomposition without hydrocarbon formation. The carbon atoms, which otherwise, would go off with the hydrocarbons, are left, very likely, to form more saturated (less active) surfaces or, possibly, graphite nuclei. There is no con clusive evidence for either view. SELECTION OF RAW MATERIALS Natural carbonaceous matter, having a mineral con tent, behaves very much as does impregnated matter. Some materials, with extremely high non-organic content, do give chars in which the carbon is widely dispersed. The impreg nated chars, in general, do not fall into this class. Ordinary organic substances, unimpregnated and pre pared by selective oxidation or similar methods, give car bons with varying activities. Firth carbonized a number of carbohydrates. He found 6g i i i glucose carbon to be the most active of those tried. ^ He reports that there is no great difference in the carbons from cedar and willow woods.^5 As a result of carbonizing thirty-nine different sub stances, including saturated hydrocarbons, carbon dioxide and aromatic hydrocarbons and their derivatives, he concludes that there is no direct relation between molecular complex- ity and the activity of the char obtained. The carbons fell into two groups, one approximately twice as active as the other. According to Chaney, wood gives carbons which are not dense enough for good gas adsorption.^ He says that pul verizing and briquetting is necessary if wood is to be used. Ray says that the original structure of the material determines whether its char may be put to a particular use ^ J. B. Firth, "Sorption of Iodine by Carbons Pre pared from Carbohydrates.", Jour. Chem. Soc., 123:323» 1923* ^ J. B. Firth, “Sorption Activity of Carbon.", Jour. Soc. Chem. Ind., 42:243T, 1923* J. B. Firth, W. Farmer and J. Higson, "Sorption of Iodine by Carbon Prepared from the Paraffin Hydrocarbons, Carbon Dioxide, Aromatic Hydrocarbons and Derivatives, and from the Products of Oxidation of Wood Charcoal with Fuming Nitric Acid.", Jour. Chem. Soc., 125:46S, 1924. 17 N. K. Chaney, A. B. Ray and A. St.John, "The Pro perties of Activated Carbon Which Determine Its Industrial Application.", American Institute of Chemical Engineers, Transactions, 15:316, 1923. 6 9 and that carbons with the right physical properties are ob- tained by the selection of the right raw material. The chemical engineers of the Bureau of Mines at Washington have experimented with a very large number of raw materials.list includes various dense woods, lignite, coals of all grades, pitch, carbon black and lamp black. They found a large number of chars to be too porous and bulky and remedied this fault by grinding after carboniza tion, briquetting with pitch or one of the caking coals, activating and then grinding to proper granule size. Quite possibly, the arrangement of the molecules in the carbonaceous matter is the deciding factor in determin ing the activity of the carbon obtained. If the atoms are so placed as to facilitate the formation of small volatile hydrocarbon molecules or molecules of water, leaving inter stices of molecular dimensions with walls composed of enough carbon atoms to give maximum adsorption, then a good activa ted carbon is obtained. Where large hydrocarbon molecules are formed with a resulting flow of melted material, no doubt the walls of the pores are scoured out, forming a por ous, light carbon. Deposition of many atoms gives an inert one. 13 A. B. Ray, 0£. cit., p. 977. ^ A. 0. Fieldner, R. E. Hall and A. E. Galloway, Study of the Production of Activated Carbon from Various Coals and Other Materials, U. S. Bureau of Mines, Paper ^79* THE PRESENCE OF NITROGEN 70 For years, it was thought that deodorizers owed their action to the presence of nitrogen in some form. Schneller says that the idea that cyanogen compounds or the nitrogen content of the raw material was the source of de colorizing power, was due to the fact that hones, with a high nitrogen content, were the source of boneblack for so long a time , the one deodorizer. Also, there was the fact that degelatinized hones from the manufacture of glue make very poor boneblack. ^ However, there is more behind the idea than just these facts. Farmer and Firth got their most active carbons 21 from compounds containing basic nitrogen. Zerban*s best 22 deodorizer was made with ammonium chloride. Also, he made a fair deodorizer by using a mixture of casein and sawdust. Horton could discover no mathematical relationship between activity and nitrogen content. Following is a table giving the results of his work. The decolorizing power of 20 Schneller, ojd. cit. , p. 156. 21 W. Farmer and J. B. Firth,"The Catalytic Activity of Carbons from Aromatic Hydrocarbons and Some Derivatives." Jour. Phys. Chem., 2^:1136, 192*1-. 22 Zerban, o£. cit., p. $, 71 each carbon is compared with that of Norit taken as one.23 Name Nitrogen content Activity Norit O.I36 1.00 Darco O.265 0.99 Supchar 0.070 1.01 Kelpchar 2.000 0.50 Carbrox 0.590 0.95 Bone char 1.060 0.30 Purified bone char 2.570 0.46 C. P. Sucrose char 0.000 0.212 The sucrose char was made by selective oxidation and there was no nitrogen present. Obviously, the results show no correlation. Bancroft mentions the work of Patterson^, who claim ed to have isolated a nitrogenous substance from boneblack far more active as a decolorizer than boneblack, itself. On this he based a statement that decolorizing action was due to the presence of nitrogen compounds. In checking over this work, Horton found that, if the solutions were kept neutral, the nitrogenous extract was entirely inactive. If the solutions were acid, the hydrogen ions caused the passage of colloidal carbon thru the filters into the extract, giving a very active material. From his results, Horton concludes that, while the ^ P. M. Horton, nDecolorizing Action of Boneblack.”, Ind. Eng. Ohem., 15:519, 1923. pit Bancroft quotes from the Jour. Soc. Chem. Ind., 22:60S, 1903. 72 nitrogen compounds undoubtedly influence the activation of of the carbon, as do other chemicals, the actual decoloriz ing action is an intrinsic property of the carbon, itself. THE EFFECT OF THE SOLVENT As one considers the phenomena of solution and of adsorption, it becomes evident that the solvation of solute particles and the adsorption of these same particles are like processes. Bonnet speaks of the”competition” between solute and solvent for places on the adsorption surface. This is common usage. However, if one remembers the matter of sol vation, it becomes apparent that there is also competition between the adsorbent and the solvent for the solute parti cles. This, seemingly, is ignored in the literature. It is well known that the adsorption of the same substance by the same adsorbent varies widely from solute to solute. Adsorption will vary from two solutions of the same kind, but of different strengths. Bonnet quotes for water - alcohol mixtures: A 94.7$ alcohol solution, exposed to activated carbon, became a 9$*9$ alcohol solution. A 9»l6$ alcohol solution, under similar conditions, W Bonnet, o£. cit.,p. 329* 73 "became a 7.0$ alcohol solution. The factor of competition is very evident. A number of problems remain to be solved. If the ad sorbed particle carries the better part of its solvent coat into the adsorbent, then there must be considerable steric hindrance with a resulting decrease in adsorption. It would seem that the intensity of the forces causing solvation, as well as those causing adsorption, should be considered in setting up a satisfactory adsorption equation for adsorption from solution* THE INFLUENCE OF HYDROGEN ION CONCENTRATION Chaney noted that the adsorptive power of a carbon for charged particles diminished,if the carbon was given a like charge, and increased by an unlike charge on the carbon. He found that the active carbon might be neutral or become charged, either positively or negatively, by the adsorption of hydrogen or hydroxyl ions, respectively.2^ He explained the activity of carbon, in decolorizing acid sugar solutions, by its adsorption of hydrogen ions and its becoming positively charged, the caramel colloids of the coloring matter being mostly negative ones. He found that the addition of acid to cottonseed oil decreased the Chaney, Ray and St.John* op* cit., p. 336. 7^ decolorizing power of the carbon in it, due to the fact that the cottonseed oil colorings were mostly positive colloids. Hauge and Willaman have studied the effect of pH on decolorizing power. ^7 am OH" Fig. II The effect of pH on the adsorption of different types of colloids. Using the negatively charged caramel colloids, they • found a very decided increase in decolorizing power as high er acid concentrations were reached. With positive methylene blue, a similar effect was noted with increasing concentrations of hydroxyl ion, but the adsorption was somewhat lower at all concentrations. For amphoteric substances (proteins were used), a zone of maximum adsorption was found on the acid side be- ^7 s. m . Hauge and J. J. Willaman, ”Effect of pH on Adsorption by Carbons.”, Ind. Eng. Chem., 19:9^3* 1927. 75 tween pH = 3 and pH ■* 6. The maximum for all carbons fell within these limits, tho no two maximums were at exactly the same pH* With non-polar substances, the adsorption was found to be irregular, varying from pH to pH without any definite trend. All showed a tendency to be less adsorbed than the charged particles. It was found that, for a number of different carbons, the ranking on a basis of the quantity adsorbed had to be shifted with changes in pH. Note the ranking of the following. The effect of the adsorbent is, thus, largely depend ent on the difference in electrical potential between the carbons and the material adsorbed. Miller asserts that all activated carbons behave ex actly alike, if all impurities are removed and that the vary ing results of previous workers may be duplicated with pure ash free carbons by adding impurities. He found a strong tendency for ash free carbons to adsorb acids and a negative adsorption for strong bases. At pH ■ 7 At pH - ^ 1. Norit 2. SuperfiItchar 1. 2. 1. Boneblack 2. Norit 4. Applechar 5* Boneblack 6. Blood char 7* Sugar char 3* Applechar 4. Carbrox 5* Superfiltchar b. Blood char 7. Sugar char 76 He states that the adsorption of neutral salts is exclusive ly hydrolytic. The acid taken up "by the char was found to "be exactly equal in quantity to the amount of "base set free pel in solution. His results and those of Hauge and Willaman are im portant in testing the adsorptive capacity of carbons and in interpreting the results. ^ E* J. Miller, Fifth Colloid Symposium Monograph, (Hew York: The Chemical Catalog Co., 1927; p. 55* CHAPTER YI EXPERIMENTAL PART PURPOSE First, there was made a comparison of the carbons, Nuchar DC, Darco DC and Carbac and the effect of three sol vents on their adsorption capacity for iodine was determined. Second, an attempt was made to determine the extent to which the presence of moisture would cut down their activities. Third, the effect of heating at about 900°C for three hours and cooling in desiccators over freshly calcined cal cium chloride was tried. Finally, using carbon tetrachloride as a solvent, a series of samples at various concentrations were run in order to draw the isotherms and determine (l/n) and (k) for the case of each adsorbent. MATERIALS USED For the first test, the solvents, chloroform, carbon bisulphide and carbon tetrachloride were used. These were of 0. P. quality as furnished by the J. T. Baker Chemical Company. In the other experiments, carbon tetrachloride was 7g used. This solvent was purified by an adaptation of the method of McClendon. Instead of the bromine treatment, the waste solutions of iodine were allowed to stand in the sun for several days. The excess iodine was reduced with sodium thiosulphate. The carbon tetrachloride was dehydrated over freshly calcined calcium chloride for twenty-four hours after repeated washings with distilled water to remove the sodium thiosulphate. The dry carbon tetrachloride was. filtered thru a layer of cotton and distilled twice, the first and last fractions being discarded. In every case, a boiling point within the range, 76.5°-77»0°C, was observed. The iodine was pure, resublimed iodine furnished by the company previously mentioned. The carbons were obtained from commercial packages. The Darco DC and the Nuchar DC were extremely fine powders. The Carbac was ground to a size which would pass thru a six ty mesh sieve. The iodine was determined by titration against solu tions of sodium thiosulphate, Baker’s C. P. quality. PROCEDURE Samples of the carbons, approximating one gram each, were weighed in small flasks. (Weighings to 1/10 mg.) J. McClendon, Jour. Biol. Chem., 60:2&9, 192^ 79 Solutions of approximately the strength desired were made by dissolving weighed amounts of iodine in the proper volumes of solvent. At the time that these solutions were used, their actual strengths were determined by titration as described below. In the first experiment, ^ > 0 ml. portions were used. In the balance of the work, this was cut to ko ml. portions. The solutions were pipetted into flasks and the flasks sealed. The carbon was left in contact with the sol ution for twenty-four hours in a dark cabinet. At the end of this time, the greater portion of the contents of each flask was filtered quickly thru a small plug of cotton wool to remove suspended carbon. Two 10 ml. portions were pipetted from this filtrate and titrated to ascertain the equilibrium concentration of the iodine. In computation, the average of these two titrations was used. TITRATION PROCEDURE The stronger solutions were titrated against N/lO sodium thiosulphate solution, prepared and standardized ac cording to Scott.2 For weaker solutions, N/20 and N/100 sodium thiosulphate, freshly prepared and standardized, were 2 W. W. Scott, Standard Methods of Chemical Analysis, (New York: D. Van Nostrand Co.) Vol. I, p. £¥o. go used. The sodium thiosulphate solution was shaken with the iodine solution. Due to the immiscibility of the two liquids, it was necessary to shake fox a considerable period after each addition of thiosulphate. It was found that the disap pearance of the iodine color gave an end point within the error limits of the weighings, provided a white background and a strong light were used. HANDLING OF THE CARBONS Where the purpose was simply to compare the carbons, (in the first three sets of trials), the samples were all made up at the same time and allowed to stand together under the same conditions of temperature and light. In determining the constants, all samples were kept for the usual twenty-four hours in a constant temperature bath at 25°C. The carbons were exposed to moisture by placing the samples in a desiccator containing water, instead of the usual dehydrating agent, for twenty-four hours. The heated carbons were treated in a section of three inch pipe with loosely screwed caps, in a furnace at approx imately 900°C for three hours. As soon as these containers had cooled sufficiently to be handled, their contents were placed in desiccators and allowed to cool over night. COMPUTING RESULTS Si Prom the loss of iodine from solution, the adsorption per gram of carbon was determined. In getting (l/n) and (k), simultaneous equations were set up "between consecutive samples by using the logarithm form of the Freundlich equation, log (x) = log (k) - l/n log (c) In this manner,(l/n)was determined for each pair of results and from them, an average value was obtained which checked fairly well with lines drawn thru the plotted points. The average value of (l/n) was substituted in each of the equations and (k) was found for each case. These values of (k) were then averaged. RESULTS AND CONCLUSIONS The first set of samples run showed that the activity of the same carbon varied from solvent to solvent. The or der of activity as between carbons did not change, the car bons ranking from Carbac, the highest, to Darco, the lowest, in the case of each solvent. See Table I. The presence of much moisture was found to cut down the activity of Darco and Nuchar, but the Carbac showed an increased capacity after its wetting. See Table II, page S3. A number of possible causes of the peculiar behavior 32 TABLE I A COMPARISON OF DARCO DC, NUCHAR DC AND CARBAC BY THE ADSORPTION OF IODINE Sample Solvent Weight sample grams Cone. original solution grams/ml. Cone. after adsorption grams/ml. Iodine adsorbed grams/gram Darco CHCI3 1.1151 0.009639 0.004214 0.2454 i t 1 1 ^ 1.0946 0.009639 0.004325 0.2450 Nuehar 1 1 1.0602 0.009639 0.003934 0.2714 i i 1 1 1.0469 0.009639 0.004077 0.2630 Carbac 1 1 1.0964 0.009639 0.003293 0.2916 i i 1 1 1.0153 0.009639 0.003123 0.3229 Darco os2 1.0394 0.010131 0.006465 0.1763 i i 1 1 I.O625 0.010131 0.006426 0.1743 Nuchar 1 1 1.0463 0.010131 0.005926 0.2003 ii 1 1 I.OO56 0.010131 0.006142 0.19S3 Carbac 1 1 1.0362 0.010131 0.004335 0.2433 n 1 1 1.0126 0.010131 0.004975 0.2545 Darco CCllj. 1.1394 0.009715 0.002375 0.3001 i t I I 1.2063 0.009715 0.002744 0.2339 Nuchar I I 1.1340 0.009715 0.001979 0.3266 n I I I.0513 0.009715 0.002352 0.3501 Carbac I I 1.0054 0.009715 0.002614 0.3531 i i I I 1.1046 0.009715 0.002379 0.3320 «3 TABLE II THE EFFECT OF ADSORBED MOISTURE ON THE ADSORPTION OF IODINE ON DARCO DC, NUCHAR DC AND CARBAC Weight Cone, original Cone, aftei Iodine Sample sample solution adsorption adsorbed grams grams/ml. grams/ral. grams/gram Darco dry 1.0094 0.010141 0.002393 0.2370 i i n 1.0110 0.010141 0.002911 0.2360 H wet 1.0001 0.010141 O.OO3636 0.2601 n n 1.0137 0.010l4l 0.003505 O.2595 Nuchar dry 1.0474 0.010141 0.001396 0.3159 n i i 0.9997 o.oioi4i 0.002013 O.3250 w wet 1.0035 o.oioi4i 0.003043 0.2312 n i i 1.0165 0.010141 0.002930 0.2337 Carhac dry 0.9930 0.010115 0.002102 0.3211 1 1 1 1 1.0035 0.010115 0.002357 0.3077 " wet 1.0027 0.010115 0.001459 0.3453 1 1 1 1 0.9999 0.010115 0.001407 0.3433 g& of the Carbac suggest themselves. The actual solution of the pore clogging minerals and their crystallization into more compact masses is one possibility* The hydrolysis of minerals and the resulting charge on the carbon surface is another* Much more work would be required before a definite conclusion could be reached. All of the carbons showed a decided increase in activ ity after being heated to 900°C for three hours. The average increase in adsorption by the Darco was 27*9* J > ; by the Nuchar, 17• £ > $ > ; by the'Carbac, g.lThis finding was expected in the case of the Carbac as it was the most recently activated of the three. The other carbons were older. See Table III. It would seem that users of activated carbons should subject them to a heating process before using in order to drive out the adsorbed moisture and gases. Of course, it is quite possible that the presence of moisture might be an advantage in cases, where the reaction with the water might give a more easily adsorbed material. This would be determined by experiment. However, the results leave no doubts, that in most cases, the heat treatment would promote adsorption, particu larly where the carbon has been stored for any length of time. The cases of adsorption investigated seem to be cover ed fairly well by the Freundlich isotherm. The values found 25 for l/n were 0.3333> 0.3757 and 0. *1-926, while the values of (k) were found to he 1.91&3> 3*1626 and 10.2753 for the carbons Darco, Nuchar and Carbac, respectively. The values of l/n, as determined by experimental figures, gave lines with slopes which seemed to agree with the plotted log points to a fair degree. While there is some divergence from the actual plot, the two are in close enough accord to warrant the statement that the Freundlich isotherm covers the cases under consideration. For the actu al figures, consult Tables IV, V, VI on pages 27, 22 and 29. The plots of the curves will be found on pages 90 and 91* 36 TABLE III THE EFFECT OF THREE HOURS HEATING AT 900°C ON THE ADSORPTION OF IODINE BY DARCO DC, NUCHAR DC AND CARBAC Sample Weight sample grams Cone. original solution graras/ml. Equilibrium cone. grams/ml. Iodine adsorbed grams/gram Darco untreated 1 1 1 1 1.0001 1.0002 0.009936 0.009936 0.002956 0 . 00 2 992 * 0.2791 0.2776 M heated 1 1 i t 1.0000 1.0000 0.009936 0.009936 0.001021 0.001059 O.3566 0.3551 Nuchar untreated 1.0000 « « 1.0001 « heated 0.9997 « « 1.0006 0.009936 0.009936 0.009936 0.009936 0.002132 0.002201 0.000319 0.000319 0.3102 0.3093 0.36^7 0.3644 Carbac untreated 1.0000 " « 1.000*1- heated I I 1.0007 1.0003 0.009936 0.009936 0.009936 0.009936 0.000343 0.000329 0.000255 0.000261 0.3637 0.3641 0.3969 0.3367 TABLE IV THE DETERMINATION OF (k) AND (l/n) OF FREUNDLICH'S EQUATION FOR THE ACTIVATED CARBON, DARCO No. Weight Darco used grams Cone. original solution grams/ml. Equilibrium Iodine conc. adsorbed (0) (x) grams/ml. grams/grair Log (c) I Log (x) (l/n) 00 1 1.0012 0.014-095 0.005020 0.3420 -2.2345 -0.4650 0.7509 1.9046 2 1.0005 0.012991 O.OO5163 O.313O -2.2071 -0.5045 0.3000 1.0104 3 1.0005 0.011900 0.004490 0.2996 -2.3470 -0.5232 0.3529 1.0214 4 1.000*1 0.010971 0.003762 0.2002 -2.4246 -O.55O3 0.3156 1.0104 5 1.0002 0.010076 0.003311 0.2705 -2.4000 -0.5670 0.4l42 1.0143 6 1.0002 0.009234 0.002701 0.2501 -2.5556 -0.5992 0.1009 1.7090 7 1.0001 0.000002 0.002000 0.2429 -2.6972 -0.6146 0.2975 1.9250 0 1.0002 0.007244 0.001603 0.2256 -2.7951 -0.6467 0.2759 1.9270 9 0.9999 0.005037 0.000957 0.1952 -3.0191 -0.7095 0.2005 1.9003 10 1.0005 0.004964 0.000632 0.1732 -3.1993 -O.7615 o.4i40 2.0174 11 1.0001 0.003910 O.OOO305 0.l4l0 -3.4145 -0.0500 0.1722 1.9360 12 1.0002 0.003097 0.000139 0.1103 -3.6570 -0.9270 2.2030 Average value (l/n) = 0.3333 Average value (k) = I.9103 TABLE V THE DETERMINATION OF (k) AND (l/n) OF FREUNDLICH'S EQUATION FOR THE ACTIVATED CARBON, NUCHAR No. Weight Nuchar used grams Cone. original solution grams/ml. Equilibrium Iodine conc. adsorbed Log (c) (c) (x) grams/ml. grams/gram Log (x) (l/n) 0 0 1. 1.0009 0.014-095 0.004676 0.3764 -2.3301 -0.4244 0.4243 2.8249 2 1.0000 0.012991 0.004096 0.3558 -2.3876 -0.4488 0.4382 2.8062 3 1.0003 0.011988 0.003465 0.3306 -2.4603 -0.4807 0.0692 2.7771 4 1.0003 0.010971 0.002820 0.3259 -2.5498 -0.4869 1.2993 2.958O 5 1.0007 0.010076 0.002629 0.2976 -2.5S02 -0.5264 O.O855 2.7733 6 1.0006 0.009105 0.001873 0.2891 -2.7275 . -O.5390 0.4014 3.0599 1 1.0006 0.008082 0.001488 O.2636 -2.8274 -0.5791 0.1742 3*0423 8 1.0000 0.007244 0.001047 0.2479 -2.9801 -0.6057 0.4844 3. 265O 9 1.0001 O.OO5837 0.000708 0.2051 -3.1500 -0.6880 0.1620 3.1295 10 1.0004 0.004964 0.000362 0.1840 -3.4413 -0.7352 0.3609 3.6116 11 1.0008 0.003910 0.000199 0.1483 -3.7012 -0.8289 0.2340 3.6440 12 1.0003 0.003097 0.000082 0.1205 -4.0862 -0.9190 .---- 4.1323 Average value (l/n) = 0.3757 Average value (k) s 3.1686 TABLE VI THE DETERMINATION OF (k) AND (l/n) OF FREUNDLICH’S EQUATION FOR THE ACTIVATED CARBON, CARBAC No. Weight Carbac used grams Cone. original solution grams/ml. Equilibrium conc. (0) grams/ml. t Iodine adsorbed Log (c) (x) grams/gram Log (x) (l/n) 00 1 1.0006 0.014-095 0.002473 0.4-64-0 -2.6059 -0.3335 0.2340 3.9166 2 0.9993 0.012991 0.00204-0 0.4-390 -2.6904- -0.3575 0.2517 9.2354 3 1.0002 0.01193S 0.001623 0.4-145 -2.7397 -0.3225 0.3444 9.2107 4 1.0004 0.010971 0.001320 0.3353 -2.3794- -0.4-136 0.9946 10.1090 5 1.0003 0.010076 0.001212 0.3544 -2.9165 -0.4-505 0.2200 9-6370 6 1.0003 0.009105 0.000363 0.3293 -3.0615 -0.4-324- 1.4405 10.6100 7 1.0001 0.003032 0.000797 0.2913 -3.0935 -0.5357 0.2911 9.7353 3 1.0001 0.007244 0.000536 0.2663 -3.2321 -0.5746 0.4393 10.4110 9 0.9999 O.OO5337 0.000339 0.2179 -3.4-101 -0.4-101 0.3025 10.4115 10 1.0009 0.004-96*1 0.000214 0.1319 -3.6696 -0.74-02 0.6033 11.6710 11 1.0000 0.003910 0.00014-0 0.140S -3.2539 -0.3514- 0.1971 11.1430 12 1.0000 O.OO3097 0.000059 0.1215 -4-. 2292 -0.9154- 11.6625 Average value (l/n) = 0.4-926 Average value (k) = 10.2753 f c f c b ' t n ? . E£Ei ft ft* iis s s - - - -hTv1 • « v - i s T O ?3: i SLtiXj- : i ' jip+.irpiffiaxfcp: : j. fff Hy ’ P- - r n ffty?r4 M-.^+-r I FrffPgrTOy^ m m MiM-t p r o 3 $Mm a s ffPiis®3 U*»aa*BlB HHlllin 3 j - - i - : - U 4 - q « ' 4 4 t l r j = ; 4 r H r F - M T 4 - : - T - r + r - - - f - ■w : ; - . i 4 ± t ' TO m ffl I S W f M p f t f t ftw TOff bi+ H w& |TO§§ i + i X t :ax»’.u±i±= Mill H H : :£ M : M--1 TO I TOR 'm H T O iit> 'ft •Hvv u + H - y , - m SETO m TO mm ftftror ' - hT O & t o t o i t ft TO xd S i it ft :#xt-:+4+ ftftTO Iplpfl WM F8’ ftft ^pfofS+ift+g ft 3 > ttH m r r . T f t lf t EF t : i Hi f t Ft ft if f ft ffl * : Hi 33 X iff m t o TO A Q W I A T T t n «5TTTTrt?VTK5* Q T rtD B R ff P lflfT J fY T A I .I F I 1 &ro j f o ’ o m ASSOCIATED STUDENTS’ STORE. BERKELEY. CALIF. SUMMARY 92 1. The activity of a given carbon varies from one sol vent to the other. 2. While the activities of the carbons tested varied with the solvent used, the rank in degree of activity did not vary, that is - the best adsorber in one solu tion was the best in the others, etc. 3. The presence of excess moisture decreased the activi ty of two of the carbons, but increased the activity of the third. 4-. Heating the carbons for three hours at 900°0 increas ed the activity of each carbon tried. 5. The poorest carbon showed the most improvement under the heat treatment. Feel that this was due to the fact that the poorest carbon was the least active because it had been open to the air longest. Incidently, the new est carbon used, showed the best tests originally and the least improvement on heating. 6. The cases of adsorption investigated seem to be covered fairly well by the Freundlich isotherm. BIBLIOGRAPHY BIBLIOGRAPHY 93 Ardagli, E. G. R., "Activated Carbon." Journal of the Society of Chemical Industry, 40:230Tt33T* 1921. Bancroft, W. D., "Charcoal Before the War." Journal of Phys ical Chemistry, 24-: 127-4-6, 201-24-, 34-2-657 1920. ....,, "Second Report of the Committee on Contact Cataly- sis." Journal of Physical Chemistry, 27:801-94-1* 1923. Bonnet, F., Jr., "Activated Carbon, Its Evaluation, Manu facture and Uses." Chemical Age, Hew York, 31:327-31*1923* Burrell, G. A., "Gasoline by the Charcoal Adsorption Process." Chemical and Metallurgical Engineering, 24-:156-60, 1921. Chaney, H. K., "The Activation of Carbon." American Electro chemical Society, Transactions, 36:91-111* 1919* . . . . , A. B. Ray and A. St.John, "The Properties of Activ ated Carbon Which Determine Its Industrial Application." American Institute of Chemical Engineers, Transactions, 15: 309-4- 6, 1923. Copisarow, M., "The Allotropy of Carbon." Chemical Hews, 118:301-4-, 1919. Cude, H. E., and Hulett, G. A., "Some Properties of Charcoals" Journal of the American Chemical Society, 4-2:391-401,1920. Dorsey, F. M., "Development of Activated Charcoals." Indus trial and Engineering Chemistry, 11:281-7, 1919* Farmer, W., and J. B. Firth, "The Catalytic Activity of Car bons From Aromatic Hydrocarbons and Some Derivatives." Journal of Physical Chemistry, 28:1136-4-6, 1924-. Fieldner, A. C., R. E. Hall and A. E. Galloway, A Study..of the Production of Activated Carbon from Various Coals and Other Raw Materials., U- S. Bureau of Mines, Paper 4-79* 30 p^ Firth, J. B., "Some Factors Governing the Sorptive Capacity of Charcoal." Journal of the Chemical Society, 119:926-31, 195X1 9 4 Firth, J. B., “Sorption Activity of Carbon." Journal of the Society of Chemical Industry, 42:242T-44t, 1923. . . . . , "Sorption of Iodine by Carbon." Faraday Society, Transactions, 16:434-52, 1921. . . . . , "Sorption of Iodine by Carbon Prepared from Carbo hydrates." Journal of the Chemical Society, 123:323-7, 1923. . . . . , W. Farmer and J* Higson, "Sorption of Iodine by Carbon Prepared from the Paraffin Hydrocarbons, Carbom Dioxide, Aromatic Hydrocarbons and Derivatives and from Products of Oxidation of Wood Charcoal with Fuming Nitric Acid." J ournal of the Chemical Society, 125:433, 1924. Garner, J. B., "Charcoal as an Adsorbent." Natural Gas, 5:(PP« 3>4,4g,50,54 and 56), 1924. Hauge, S. M., and J. J. Willaman, "Effect of pH on Adsorption by Carbons." Industrial and Engineering Chemistry, 19:943-53, 1927: Horton, P. M., "Decolorizing Action of Boneblack." Industrial and Engineering Chemistry, 15:519-20, 1923• Howard, H. C., and G. A. Hulett, "Study of the Density of Carbon." Journal of Physical Chemistry, 23:1032-95, 1924. Kosakevich, P. P., and N. A. Ismailov, "Activation of Char coal by Steam." Chemical Abstracts, 23:4730, 1929. Langmuir, Irving, "Adsorption of Gases on Plane Surfaces of Glass, Mica and Platinum." Journal of the American Chem ical Society, 40:136l-l403,~ I9l3. “ . . . . , "Constitution of Solids and Liquids." Journal of the American Chemical Society, 3^:2221-95, 1$16. . . . . , "Fundamental Properties of Solids and Liquids." Journal of the American Chemical Society, 39:1343-1906, wrr.------------------ Latham, G. H., "The Thickness of Adsorbed Vapor Films." Jour- of the American Chemical Society, 50:2937-97, 1923. McBain, J. W., "Theories of Adsorption and the Technique of Its Measurement." Nature, 117:550-1, 1926. 95 Me Bain, J. W., ''Theories of Occlusion and the Sorption of Iodine by Oarbon." The Faraday Society, Transactions, 1^:202-12, 1919. McClendon, J., "The purification of Oarbon Tetrachloride." Journal of Biological Chemistry, 60: 269, 192^. McGavack, John, and W.A. Patrick, "Adsorption of Sulfur Di oxide by the Gel of Silicic Acid." Journal of the Amer- ioan Chemical Society, *J-2:9^6-76, 1920. Miller, E. J., "Adsorption by Ash Free Charcoal." Article in Fifth Colloid Symposium Monograph, (Hew York, The Chemi cal Catalog Co., 1927)PP* 55-30. Page, A. B. P., "Activation of Wood Charcoal by Progressive Oxidation in Relation to Bulk Density and Iodine Adsorp tion." Journal of the Chemical Society, 130:l^-76-9^»1927. Patrick, W. A., "The Capillary Theory of Adsorption from Solution." American Institute of Chemical Engineers, Transactions, 15‘2S3-9it - » 1923. Philip, J. C., S. Dunhill and 0. Workman, "Activation of Wood Charcoal by Heat Treatment." Journal of the Chemical Society, 117:362-69, 1920. Ray, A, B., "Manufacture of Activated Carbon." Chemical and Metallurgical Engineering, 26:977-62, 1923* Ruff and Mautner, "Active Charcoal: Amorphous Nature, Chemi cal Abstracts, 22*2614-, 1^26. Schneller, M. A., "The Vegetable Decolorizing Carbons." Louisiana Planter, :154--64-, Septemner, 1917* Scott, W. W., Standard Methods of Chemical Analysis, 2 Vols.; New York; D. Van Nostrand Co. 1930* P• 2 4 - 0 ' . Svedberg, The, Colloid Chemistry, A. C. S. Monograph, New York: The Chemical Catalog Co., 2nd. ed., 1926. Swiderek, "Charcoal Activated bp Mineral Substances." Chemical Abstracts, 21:3506, 1927. Taylor, H. S., Editor, A Treatise on Physical Chemistry, 2 Vols., New York,~D. Van Nostrand Co., 2nd. ed., 1931* Reader referred to Vol. II, Chap. XX, Colloids,Kraemer, 0. 96 Zerban, F. W., Vegetable Decolorizing Carbons and Their Use in the Sugar- C ane Industry, Bullet in 161, Louisiana State Agricultural Experiment Station, Louisiana State University, Baton Rouge, La., 39 P« . . . . , E. 0. Freeland, and D. D. Sullivant, Studies on the Preparation of the Vegetable Decolorizing Carbons for tb'e Cane Sugar Industry, Bulletin 167, Louisiana State Agri cultural Experiment Station, Louisiana State University, Baton Rouge, La., **4 p. TO all i g S b ^ w S f i i T t T -" t::J in :r^St;'im::t. t.3 K :::-* -* :r .ts :E fe t*“3 ^ -Z ^-53 + + • + ttm rr: s * - + • + i« V : u x u

Asset Metadata

Creator Lanphier, E. C (author)

Core Title Studies on the adsorptive powers of certain activated carbons

Contributor Digitized by ProQuest (provenance)

Degree Master of Arts

Degree Program Chemistry

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag chemistry, organic,OAI-PMH Harvest

Language English

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c17-785252

Unique identifier UC11348353

Identifier EP41455.pdf (filename),usctheses-c17-785252 (legacy record id)

Legacy Identifier EP41455.pdf

Dmrecord 785252

Document Type Thesis

Rights Lanphier, E. C.

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA