UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The kinetics of the reduction of mercuric salts by molecular hydrogen in aqueous solution Korinek, George Jiri 1956

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-UBC_1956_A1 K6 K4.pdf [ 4.65MB ]
JSON: 831-1.0106272.json
JSON-LD: 831-1.0106272-ld.json
RDF/XML (Pretty): 831-1.0106272-rdf.xml
RDF/JSON: 831-1.0106272-rdf.json
Turtle: 831-1.0106272-turtle.txt
N-Triples: 831-1.0106272-rdf-ntriples.txt
Original Record: 831-1.0106272-source.json
Full Text

Full Text

THE KINETICS OP THE REDUCTION OP MERCURIC SALTS BY MOLECULAR HYDROGEN IN AQUEOUS SOLUTION by George J i r i Korinek A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OP PHILOSOPHY in METAL CHEMISTRY We accept this thesis as conforming to the required standard T H E U N I V E R S I T Y OF B R I T I S H C O L U M B I A August, 1956 %\\t H n m e r s r i t i ; at Pritislj ( J L u l r n n l t i a Faculty of Graduate Studies P R O G R A M M E O F T H E J f f i t r a l © r a l ^ E x a m i n a t i o n for t l j e ]jbt%xtt i t f p a c t a ai ^\}ihsoTff\}\i of GEORGE JIRI KORINEK B.A. (University of British Columbia) 1953 M.Sc. (University of British Columbia) 1954 W E D N E S D A Y , A U G U S T 15, 1956, at 10:00 a.m. I N T H E PHYSICAL M E T A L L U R G Y B U I L D I N G R O O M 204 C O M M I T T E E I N C H A R G E D E A N G. M . S H R U M , Chairman W. A. B R Y C E C. A . M C D O W E L L F. A. FORWARD C. S. SAMIS J . H A L P E R N L . W. S H E M I L T R. D. J A M E S F. H . SOWARD External Examiner: PROFESSOR C. A. W I N K L E R McGil l University THE KINETICS OF T H E R E D U C T I O N OF MERCURIC SALTS B Y M O L E C U L A R H Y D R O G E N I N AQUEOUS SOLUTION ABSTRACT The kinetics of the homogeneous reduction of mercuric salts by molecular hydrogen in aqueous solution have been examined over a wide range of solution composition, temperature and partial pressure. In perchlorate solution, where Hg++ and H g 2 + + are uncomplexed, the kinetic results were found to be of the form: - d [ H 2 ] / d t = k, [H 2 ][Hg+ + ] - f k 2 [ H 2 ] [Hg 2++] where kt = 4.2 x 10!° exp [—18100/RT] 1. mole"1 sec.' 1 and k 2 = 1.2 x 10" exp [—20400/RT] 1. mole-' sec. - ' It was concluded that the rate-determining process involves the biomolecular interaction of one H 2 molecule with either Hg++ or H g 2 + + , i.e. Hg++ + H 2 - Hg + 2H+ Hg 2++ + H , ~* 2Hg - f 2H+ The Hg atoms thus formed undergo further rapid reactions to yield the observed products ( H g 2 + + or metallic mercury) whose nature is determined by thermodynamic considerations. Mercuric complexes also react homogeneously with hydrogen but in most cases, more slowly than the simple Hg++ ions. The order of the decreasing re-activity of the complexes, HgAc, > HgCl,> HgBr 2> Hg(ethylenediamine)2+ + is the same as the order of their increasing stability. This is attributed to the re-duction of the electron affinity of Hg++ through electron donation from the complexing ligand. The reactivity of mercuric complexes such as HgAc 2 and Hg(ethylenediamine)2++ is increased by anions such as O H ~ , C0 3 =, Ac - etc., whose influence increases in the same order as their basicity. This is attributed to stabilization, by the anions, of the H+ ions which are released in the rate-determining step of the reaction. Some conclusions are drawn concerning the mechanism of heterogeneous activation of hydrogen by solid catalysts. PUBLICATIONS Homogeneous Reaction Between Molecular Hydrogen and Mercuric Acetate: J. Halpern, G. J. Korinek and E. Peters. Research (London), 7:s6l -2(1954). Kinetics of the Reaction of Molecular Hydrogen with Mercuric and Mercurous Perchlorates in Aqueous Solution: G. J. Korinek and J. Halpern. /. Phys. Chem., 60, 285-9, (1956). Effects of Complexing on the Homogeneous Reduction of Mercuric Salts in Aqueous Solution by Molecular Hydrogen. G. J. Korinek and J. Halpern. Can. ]. Chem. in press. G R A D U A T E STUDIES Field of Study: Metal Chemistry Metallurgical Thermodynamics and Kinetics C. S. Samis Theory of Metal Reactions J. Halpern Structure of Metal Compounds J. Halpern Metal Surface Chemistry H . G. V . Evans Theory of Alloys H . P. Myers Other Studies: Topics in Physical Chemistry J. G. Hooley and B. A . Dunell Chemical Kinetics W . A. Bryce Theory of Chemical Bond . C. Reid Quantum Mechanics O. Theimer Magnetic Properties of Metals H . P. Myers Phase Transformations in Metals W. M . Armstrong and J. G. Parr ' Molecular Spectroscopy..... O. Theimer ABSTRACT i i The kinetics of the homogeneous reduction of mercuric salts by-molecular hydrogen in aqueous solution have been examined over a wide range of solution composition, temperature and hydrogen partial pressure. In perchlorate solutions, where Hg and Hg2 are uncomplexed the kinetic results can be expressed by the equation: -d[H2J/dt = k 1[H 2][Hg + +] + k ^ K B g ^ J where = 4.2 x 10 1 0 exp [-18100/RT] l.mole"1 sec.""1 and k 2 a 1.2 x 10 1 1 exp [-20400/RT] l.mole"1 sec." 1 It was concluded that the rate-determining process of the reaction involves the bimolecular interaction of one EL molecule with either Hg or Hg , i.e., k Hg** + H 2 - i * Hg + 2H+ Hg 2 + + + K2-2> 2Hg + 2H+ The Hg atoms thus formed undergo further rapid reactionsto yield the observed products (Hgg4"*" or metallic mercury) the nature of which is determined by •' thermodynamic considerations. Mercuric complexes also reacted homogeneously with hydrogen but in most cases more slowly than the simple Hg ions. The order of decreasing reactivity of the complexes, HgAc,,^ . HgClg ^  HgBr2 y Hg(ethylenediamine)2 is the same as the order of their increasing stability. This is attributed to the reduction of the electron affinity of Hg through electron donation from the complexing ligand. The reactivity of mercuric complexes such as HgAc2 and HgCethylenediamine)^ is increased by anions such as 0H~, C0^~, Ac", etc., the influence of which increases in the same order as their basicity. This is attributed to stabilization, by the anions, of the H + ions which are released in the rate-determining step of the reaction. i i i Some conclusions are drawn concerning the mechanism of heterogeneous activation of hydrogen by solid catalysts< iv ACKNOWLEDGMENT The author is grateful for the assistance and encouragement given by the members of the Department of Mining and Metallurgy. He wishes to express his indebtedness primarily to Dr. J. Halpera for his inspiring and able direction of this project. The author also wishes to thank the National Research Council of Canada for financial support of the research and to the Consolidated Mining and Smelting Company for the Cominco Fellowship, held in 1955-56. V TABLE OF CONTENTS Page INTRODUCTION 1 GENERAL ASPECTS OF HYDROGENATION REACTIONS . . . . . . 1 HETEROGENEOUSLY-CATALYZED HYDROGENATION . . . 3 HOMOGENEOUS HYDROGENATIQN REACTIONS IN SOLUTION . . . . ,7 OBJECT AND SCOPE OF THIS INVESTIGATION . 10 EXPERIMENTAL 12 MATERIALS . 12 ANALYTICAL T . . . 12 PROCEDURE . . . . . . . . . 12 RESULTS AND DISCUSSION . . . . . 15 PERCHLORATE SYSTEM . . . . . . . . 15 CLASSIFICATION OF OTHER SYSTEMS . . . . . . . . . T . . 33 CATEGORY A 35 (1) Nitrate System . . . . . . . . . . . 35 (2) Sulphate System . . 37 CATEGORY B . . , , . . 37 (1) Acetate System 37 (2) Propionate System . . . . . . . . 50 (3) Chloride System . . . . . 50 (4) Bromide System 57 CATEGORY C 60 (1) Ethylenediamine £pA) System 60 (2) Ethylenediaminetetraacetic acid (EDTA) System 64 CONCLUSIONS . 65 MECHANISM OF THE REACTION , . . . 65 v i Page (1) Perchlorate System . . . . . . . . . . . . . . . 65 (2) Effect of Complexing . . . . . . . . . . . . . 68 GENERAL CONCLUSIONS ABOUT THE ACTIVATION OP H IN HOMOGENEOUS AND HETEROGENEOUS SYSTEMS . . . . . . . . . 71 R]j)yj(jjPl^MQT*jFi o a o o o o o o o a o i t o o o o o e o o o o o o o 75 APPENDIX A. Isotopic Exchange between Hg(l) and Hg(ll) . . 78 APPENDIX B. Summary of Selected Experimental Results . . . 81 v i i LIST OF TABLES Table No. Page I SUMMARY OF HOMOGENEOUS HYDROGENATION REACTIONS .... 9 II VALUES OF K AND a USED IN CALCULATION OP k^ AND kg 23 III THE INFLUENCE OF SOLUTION COMPOSITION ON THE RATES OF REACTION OP Kg AND Hg2 WITH H 2 29 IV THE INFLUENCE OF TEMPERATURE AND HYDROGEN PRESSURE ON THE RATES OF REACTION OF Hg AND Hg2 WITH Hg 30 V CLASSIFICATION OP REACTIONS 34 VI REDUCTION OF MERCURIC NITRATE BY HYDROGEN 36 VII COMPARISON OF EXPERIMENTAL AND CALCULATED VALUES OF C 57 VIII RELATIVE REACTIVITIES AND STABILITY CONSTANTS OF VARIOUS MERCURIC COMPLEXES 70 v i i i LIST OF FIGURES Fig. Noo • Page 1. Diagram of Stainless Steel Pressure Vessel and Internal 2. Typical Rate Plots for the Reduction of HgCciO^) by H 2 16 3. Rate Plots Showing Effect of Varying Initial Mercuric Perchlorate Concentration . . . . . . . . . . . . . . . 20 4 . Rate Plots Depicting the Variation of Hg i n Experi-ments with Different Initial Hg Concentrations . . . 21 5. Rate Plots for the Reduction of Hg ( c i 0 . ) 2 at Different Hydrogen Partial Pressures . . . . . . . . . . . . . . 25 6. Dependence of the Mercuric and Mercurous Perchlorate Reduction on the Hydrogen Partial Pressure . 26 7 . The Solubility of Hydrogen in Water as a Function of Temperature . . . . . . . . . . . . . . . . . 27 8. Rate Plots for the Reduction of Mercuric and Mercurous Perchlorate at Different Temperatures . . . . . . . . . .31 9. Arrhenius Plots- Showing the Temperature Dependence of 1 0 . Rate Plots Showing the Effect of Sulphate on the Reduction of Hg ( l l ) . . . . . . . . . . . . . . . . . . 38 11. Effect of Sulphate on . . . . . . . . . . . . . . . 39 12. Rate Plots for the Reduction of Mercuric Acetate . . . 41 13. Rate Plots for the Reduction of Mercuric Acetate at Different Hydrogen Partial Pressures . . . 43 14 . Rate Plots for the Reduction of Mercuric Acetate at Different Temperatures . . . . . . . . . . . . . . . . 44 15« Arrhenius Plot Showing the Dependence of k^ for Mercuric Acetate on Temperature . . . . . . . . . . . . 46 16. Rate Plots for the Reduction of Mercuric Acetate at Different Sodium Acetate Concentrations . . . . . . . . 47 1 7 . Dependence of the Reduction of Hg ( l l ) on the Concentr-ation of Sodium Acetate and Sodium Propionate . . . . . 48 18. Dependence of the Reduction of Mercuric Acetate on the Acetic Acid Concentration . . . . . . . . . . . . . . . 49 ix Fig. No. Page 19• Rate Plots for the Reduction of Mercuric Propionate at Different Sodium Propionate Concentrations 51 20. Rate Plots for the Reduction of Mercuric Chloride and Mercuric Bromide by Hg . . . . . . . . . . . . . 52 21. Rate Plots for the Reduction of Mercuric Chloride at Different Hydrogen Partial Pressures . . . . . . . . . . . 53 22. Effect of Excess Chloride Concentration on the Rate of Reduction of Mercuric Chloride . . . . . . . 55 23. Rate Plots for the Reduction of Hg(ll) in Solutions with Various Initial [Cl~]/[Hg ] Ratios . 56 24* Rate Plots for Reduction of Mercuric Chloride at Different Temperatures . . . . . . « . . « . . 58 25. Arrhenius Plot Showing the Dependence of k^ for Mercuric Chloride on Temperature . . . . . . . . . . . . 59 26. Rate Plots for the Reduction of Hg(ll) in Solutions Containing EDA o » » < - » * « o o * » o « » o * * > « * « < t » > * 61 27. Rate Plots for Reduction of Hg(EDA)g++ at Different Hydrogen Partial Pressures . . . . . . . . . . . . 62 28. Effects of Various Added Bases and Salts on the Rate of Reduction of Hg(ll) in Solutions Containing EDA . . . . . 63 THE KINETICS OF THE REDUCTION OF MERCURIC SALTS BY MOLECULAR HYDROGEN IN AQUEOUS SOLUTION INTRODUCTION GENERAL ASPECTS OF HYDROGENATION REACTIONS The reactions of molecular hydrogen have been studied extensively in the past both because of their practical significance, and because of the great scientific interest which attaches to them- The practical interest arises from the numerous well-known and important industrial applications, both organic and inorganic, of hydrogenation reactions. On the other hand, contributing to the fundamental scientific interest in such reactions is the fact that hydrogen i s the simplest molecular species; hence i t s reactions should be more readily amenable to theoretical treatment than those of most other molecules, and should serve as particularly good models against which to test c r i t i c a l l y theories of kinetics and reaction mechanisms. Furthermore, the marked sensitivity of hydrogenation reactions to catalytic influences has made their study of special interest in connection with the elucidation of various catalytic phenomena. For the purpose of the present discussion, i t is convenient to classify the reactions of hydrogen into two categories. The reactions in the f i r s t category (among which are most organic hydrogenation reactions) are those in which the molecule is split, and the H atom becomes covalently attached to other atoms.. The most common examples involve the addition of hydrogen across unsaturated bonds, e.g. (1) = 0 + Hg — - OH H (2) etc. Such reactions (usually carried out in the presence of a catalyst) are widely used industrially, in the manufacture of synthetic gasoline, the production of fats from oils, etc. In the reactions of the second category the H;2 molecule acts strictly as a reducing agent, by giving up i t s electrons to another species which i s , in turn, reduced to a lower oxidation (or valence) state. Most inorganic hydrogenation reactions, particularly in aqueous solution, are of this type. Examples include the reduction of metallic ions in solution to ions or compounds of lower valence, or to the metallic state, e.g. C r 2 0 7 = + 3H2 + 8H+—-» •2Cr + + + + 1B£ (3) U0 2(C0 5) 3 S + H 2 — » U02 + 2HC05~ + C0= (4) Ni** + H 2 —> Ni + 2H+ (5) Reactions of this type have found widespread application in recent years, particularly in the metallurgical industry (l,2 ,3,4). It is clear that such reactions can proceed only in a system where there is present a suitable base (which may be a solvent molecule such as RVjO) to stabilize the protons which are released when H 2 transfers i t s electrons to the oxidant. The reactions of hydrogen with mercuric salts, the study of which forms the main subject of this thesis, belong in the latter category. Reactions such as those listed above, even when very favorable thermodynamically, usually proceed only very slowly at moderate temperatures in the absence of catalysts;, indeed this is characteristic of most reactions of molecular hydrogen. This apparent inertness of the R"2 molecule, is attributable to it s high dissociation energy (ca. 103 kcal./mole) and to it s closed shell electronic configuration. Because of these properties H 2 is repelled strongly on approaching most other species (particularly when the latter is also electronically saturated) and it s reactions are therefore normally characterized by high activation energies. -3-Some basis for understanding this behaviour is provided by quantum mechanics and by current theories of reaction kinetics. However, such treatments are thus far, for the most part, qualitative and incomplete. In particular, two features of hydrogenatlon reactions stand out as being in need of further elucidation: 1. The ability of certain catalysts, for the most part solids (i.e. heterogeneous catalysts), to effect a very marked lowering of the activation energy of many hydrogenation reactions. Thus the activation energy for the hydrogenation of ethylene is 40-50 kcal./mole in the gas phase (5) but only about 10 kcalo/mole when the reaction proceeds on the surface of metallic nickel (6). 2. The fact that a limited number of hydrogenation reactions proceed homogeneously in solution, in the absence of solid catalysts, at moderate temperatures and with relatively low activation energies (7,8,9). A high degree of selectivity appears to characterize such reactions. Thus MnO ~ is reduced readily by hydrogen in aqueous solution at room temperature, while Cr^Orj- is not reduced even at 150°C Because these two related phenomena have an important bearing on the work to be described in this thesis (and indeed on any consideration of hydrogenation reactions) they will be commented on here, in turn, in some detail. HETEROGENEOUSLY-CATALYZED HYDROGENATION In general, the role of a catalyst in a reaction i s to provide an alternative mechanism, often quite different from that of the uncatalyzed reaction, which has a lower activation energy than the latter. Because the same catalysts are generally effective in a large variety of hydrogenation reactions, the view has been widely held that in each case the function of -4-the catalyst i s primarily to activate the Hg molecule a, i.e. by forming a reactive complex with i t ( l O , l l ) . Furthermore i t has often been assumed that a l l hydrogenation catalysts achieve this by essentially the same mechanism. Both these views probably require some qualification, particularly in connection with hydrogenation reactions of the f i r s t category designated above, where the catalyst is probably involved in the activation not only of hydrogen but of the other reactant as well (i.e. GgH^ , Kg, etc.). In general the most effective catalysts for hydrogenation reactions are the transition metals such as Ni, Pd, Pt and certain semi-conducting metallic oxides such as ZnO, CrgO^ and their mixtures. Various attempts have been made to account, both qualitatively and quantitatively, for the catalytic activity of these substances, but these attempts have so far met with only moderate success. Two main approaches have been used which emphasize respectively the structural (i.e. geometric) and the electronic properties of the catalysts (12). The Geometric Factor; Many attempts have been made to relate the activity of hydrogenation catalysts to the spacing and geometric arrangement of atoms on the surface of the catalysts (12). Experimental evidence for such a relation has been provided through studies of the hydrogenation of ethylene (13), benzene (14,15) etc., on transition metal catalysts. Some theoretical support for this approach has also been provided by the semi-quantitative calculations of Sherman and Eyring ( l6) which indicate that the activation energy for the chemisorption of hydrogen by carbon, i.e. H - H H Hi C - C * A - A (6) a The concept of "activation of hydrogen" has assumed great importance in connection with both catalyzed and uncatalyzed hydrogenation reactions and extensive reference to i t will be made throughout this thesis. -5= varies with the distance between adjacent carbon atoms9 having a minimum value for a separation of about 3°5 Objections to this general approach have been based on the claims that (a) the activation of hydrogen does not necessarily involve covalent bonding of the H atoms to atoms of the catalyst, as in the above model9 (b) metals such as copper and zinc which have favorable lattice spacings are nevertheless catalytically inactive, and (c) in catalysis by binary alloys (l7,18) the variation in. catalytic activity with alloy com-position bears no relation to geometrical factors. Finally i t should be noted that even as with the pure transition metals, where the catalytic activity can apparently be correlated with variation in the lattice parameter of the catalyst, i t can also be correlated with certain electronic properties such as the d-character of the metallic bonding, and i t is possible that the former correlation is only of secondary significance (l9)° The consensus of current opinion seems to be that the geometry of the catalyst is probably of importance in the hydrogenation of molecules such as benzene (and possibly ethylene) which have a high degree of geometrical specificity and whose chemisorption on the catalyst surface involves attachment at more than one site. However, i t seems unlikely that the geometry of the catalyst, as such, is important in connection with the catalytic activation of hydrogen (l2). Electronic Factors; The weight of modern evidence is consistent with the view that electronic factors play a large part in determining the catalytic activity of solids. These factors are not as yet a l l fully understood, but i t is generally accepted that the ability to activate hydrogen catalytically is associated with the presence of an incompletely f i l l e d d-band in the catalyst. There is some difference of opinion as to whether the interaction between the catalyst and the reactant molecules in the chemisorbed layer is primarily covalent or ionic. According to the former view (18), H„ molecules are split homolytically on the catalyst surface and - 6 -the H atoms become covalently bonded to catalyst atoms. This approach emphasizes the relationship between catalytic activity and the availability of unfilled regions in the d-band of the catalyst (Mott-Jones theory) or vacant atomic d-orbitals (Pauling theory). The incomplete f i l l i n g of the d-band in metallic catalysts results in the presence of unpaired electrons and is reflected in the value of the paramagnetic susceptibility of the metal. Maxted (20) has pointed out that for the transition metals, which are particularly effective catalysts in hydrogenation and denydrogenation reactions, paramagnetic susceptibility and catalytic activity follow parallel sequences. Generally the activity rises to a maximum for the last metal in each transition series (i.e. Ni, Pd and Pt respectively). At least in the last two cases, where there is no complication due to ferromagnetism, the paramagnetic susceptibility also increases to a maximum for those metals. When transition metals are alloyed with electron-donating metals of the IB group (Cu, Ag, Au) their activity diminishes as the number of holes in the d-band decreases to zero. The paramagnetic susceptibility of the alloy follows a similar trend. Similarly, the catalytic activity of palladium is reduced by dissolved hydrogen which donates electrons to the incompletely-filled d-band. The paramagnetic susceptibility i s reduced to zero, and the catalytic activity f a l l s sharply as the atomic ratio of H to Pd approaches a value of 0 , 6 , at which point the d-band of palladium (originally containing 0.6 vacancies) is just f i l l e d ( l 8 ) . This phenomenon provides a basis for explaining the action of catalyst poisons, most of which contain strongly electron-donating atoms or groups ( 2 l ) . It also suggests that adsorbed reactant molecules may become aotivated through donating or transferring electrons to the catalyst. This view has received support both on theoretical (22) and on experimental grounds. Particularly convincing experimental evidence has been provided by Schwab (23) who studied the kinetics of the dehydrogenation of formic acid on silver alloys, and found that the activation energy increased linearly with the square of the electron concentration of the Hume-Rothery a-phase. This was interpreted to imply that the catalytic activation of formic acid i n -volves the entry of two electrons from the adsorbed molecule into>the-conduction band of the metal. Among the intermetallic Hume-Rothery phases, the y-phase (in which the conduction band is nearly f u l l ) showed a maximum activation energy. Similar considerations have been applied to the interpretation of the catalytic activity of metal oxides, the most active of which are defect semiconductors such as ZnCrgO^ which can readily accept electrons (24). Experimental information, although of a conflicting nature, about the interaction of adsorbed gas films with metallic catalysts has also been provided by surface potential measurements (25,26). Some progress has also been made, in the theoretical understanding of these phenomena, particularly of the factors which influence the tendency for electron transfer between adsorbed molecules and the underlying surface. Dowden (22) has shown that when the rate of reaction is affected by the rate of formation or the concentration of a chemisorbed positive ion, the activity of the catalyst increases with the value of i t s electronic exit work function [©], the energy density of electron levels at the Fermi surface [g(E) ] E = —9 and the gradient of the latter [dg^/dE] ^ ^ . When negative ions assume the dominant role (i.e. electron transfer is from the catalyst to the adsorbed reactant molecule) then the catalytic activity should depend in-versely on these properties of the catalyst. At the present time quanti--tative application of these ideas is s t i l l d i f f i c u l t . HOMOGENEOUS HYDROGENATION REACTIONS IN SOLUTION The limited progress which has been made, particularly at the level of quantitative theory, in understanding the kinetics and mechanisms of heterogeneously-catalyzed hydrogenation reactions, reflects in large measure the complexity of these catalyst systems and the inherent difficulties associated with the study of surface reactions. In view of this, reoent demonstrations that some substances—notably the salts of certain metals-possess the property of activating molecular hydrogen homogeneously in solution have evoked considerable interest. Because the understanding of properties of molecules and ions in solution i s generally in a more advanced state than those of solids, i t might be expected that the study of such systems will lead to a more detailed knowledge of the nature of the activation process than has been derived thus far from the study of hetero-geneous hydrogenation reactions. In particular, some modification can be expected of the view, s t i l l commonly held, that catalytic activity refleoted in the ability to activate hydrogen is confined to solids and depends cr i t i c a l l y on some characteristic property (e.g. lattice arrangement or electronic band structure) of the solid state. To date more than fifteen systems have been discovered and studied in which hydrogen i s activated and reacts homogeneously in solution (9) . The principal ones together with the available data about the kinetics and mechanism of each are listed in Table I. Most of the reactions (although not without exception) belong to the second category of hydrogenation reactions defined above, i.e. those in which Bg acts only as a reducing agent or electron-donor. In each case, the rate-determining step of the reaction can be identified with the activation of hydrogen. In some cases, the species responsible for activating Hg functions only as a catalyst, while in others i t i s the reactant (i.e. oxidant) i t s e l f . Almost a l l the reactions involve more than one step/ with the consequent participation of an "activated' intermediate" which may be either a hydrogen-carrying complex (i.e. CuH+, AgH*, HCo(CO)^, etc.) or a reduced form of the reactant or catalyst (e.g. MnCv,-) (27). In the light of these results, the question of whether a -9-TABLE I SUMMARY OP HOMOGENEOUS HYDROGENATION REACTIONS Hydrogen Activating Species Solvent , Reaction Studied Temp, Kinetics for Range -d[Hj/dt • OC. AH* kcal. AS* "e,u Intermediate Species Proposed Mechanism Refs. CuAc Quinoline Reduction of Cu(ll) or quihone; p-Hg convention 25-117 kCH^HCu1]2 15-16 (-20) CvpR 2CuI + H -^-i 2CuI.H fast 2CuI.H + Substrate —* 2CuI + Products • 28,29 30 CuAc Pyridine Reduction of Cu(ll) 100 ktHgltCu 1] v» CuH Cul + H 9 — C u H + H+ fast CuH + Substrate > Cul + Products 51 AgAc Pyridine' Agr-> M. 25-78 k[H 2KAgIj 12-14 (-25) AgH AgAc + H9—^+ AgH + HAc AgH + AgAc 2Az + HAc 31.32 Co 2(CO) 8 Benzene, Bther Hydrogenation etc. 90-200 HCo(C0)4 Co 2(C0) 8 + Hg—^2HCo(C0)4 2HCo(C0)4 + Substr.—>Products + Co ?(C0) R 33,54 Ethylene platinous chloride Toluene, Acetone C 2H 4 + H 2 - 4 C 2H 6 (PtCl 2C 2H 4) + 2C 2H 4 >2Pt(C 2H 4) 2 2Ptfc 2H 4) 2 • 2H 2—>PtCl 2(C 2H 4) 2 + C 2H 6 35 CuAc g Aqueous HAc C u 1 ^ . Cu 1 or 80-140 k[H 2][CuAc 2] 24 (-7) CuH+ CuAc. + Hj-^ CuH+ + HAc + Ac" + fast CuH + Substrate ——> Products + CuACg 56 Cu^ Aqueous HC104 Reduction of Cr20y~, 10^" etc. 80-140 k 1[H 2][Cu + +] 2 26 (-10) CuH+ Cu** + H 2 ?=^ CuH+ + H + [Cu + +] + (k^/k 2)[H +] CuH+ + 'Cut ? 2Cu+ + H + 2Cu + Substrate 2Cu + Products 37 Ag + Aqueous HC104 Reduction of Cr 20^~, etc. 50-70 k[H 2][Ag) 2 15 (-22) AgH* 2Ag+ + H 2 2AgH+ + fast + 2AgH + Substrate ——> Products + 2Ag 38 MnO " 4 Water Mn04~ +5/2 H 2 + H+ —> Mn02 + 2H20 50-70 k[H2J[Mn04~] 14 (-17) MnV Mn04~.+ Hg-iE^MnO 3 + 2H* . or — ^ MnO " + HgO Mn043(orMn05~) -^T MnOg 39 Ag + + MnO^ ~ Aqueous HC104 Mn04" + 5/2 H 2 + H + —*Mn0 2 + 2H20 50-60 k[H2][Ag+][Mn04"] 9 (-26)' Mn04s + AgH* Ag + + MnO." + H _^AgH + + MnO.= + H* + - fast ^ 1 AgH + MnO Products M n 0 4 = - ^ ? Products 39 Cobaltous cyanide Water H 2 absorption 25 k[H 2][Co(H)) 2 ( 0 0 ^ ( 0 0 ) ^ ^ *2*> 8 hydrogenation reaction i s , strictly speaking, catalyzed or not, does not appear to be of great theoretical importance. Fundamental significance attaches rather to the definition of the circumstances under which Hg can approach another species (which may be either a reactant or catalyst) and become activated, or undergo reaction with relatively low activation energy. Even the role of heterogeneous hydrogenation catalysts would seem to be best considered within the framework of this more general question. OBJECT AND SCOPE OF THIS INVESTIGATION Preliminary experiments, leading up to this investigation, indi-cated that of the common metal ions only a few, notably Cu , Ag , MnO^  and Hg , possessed the ability to activate hydrogen homogeneously in aqueous solution. Of these, Hg + + and MnO^ ~ are themselves reduced by hydrogen, while Cu and Ag catalyze the hydrogenation of other dissolved sub-strates such as CrgO^-, 10 ^  etc. This thesis describes detailed kinetic studies of the reduction of various mercuric salts by hydrogen in aqueous solution, made with the object of elucidating the mechanism of the activation of hydrogen in this system. Thereby i t was also hoped to obtain information of a fundamental nature which would throw some further light on the problem.of the activation of hydrogen in general. It was considered of particular interest to determine the ++ effect of complexing Hg with various inorganic and organic complexing agents, on i t s rate of reaction with Hg. During the course of this investigation i t was discovered that Hg is activated in aqueous solution, also by Hgg , and the kinetics of this process were also elucidated. It should be noted that most of the homogeneous hydrogenation reactions listed in Table I have been studied, and the results on them reported, only since the investigation described in this thesis was under-taken (only the activation of H_ by Cu(l), Cu(ll) and Cd_(C0)o had previously been reported). The kinetic examination of a number of these reactions, particularly those in aqueous solution, was made by other workers in this laboratory, concurrently with this work. EXPERIMENTAL MATERIALS Ethylenediamine (EDA) and ethylenediaminetetraacetic':acld•• (BDTA )V were Eastman white label products. The perchlorate salts were obtained from G . F . Smith Chemical Co. A l l other chemicals were Baker and Adamson reagent grade. Hydrogen and nitrogen gases were supplied by Canadian Liquid Air Co. Distilled water was used in the preparation of a l l solutions. ANALYTICAL The solutions were analyzed for Hg(l) by potentiometric titration with KMnO^ . In most cases the total Hg concentration was determined by titration with KCNS using a ferric indicator, following acidification and oxidation with KMnO^  where necessary. The concentration of Hg( l l) , estimated from the difference between these two values, always agreed well with that determined directly by the cyanide method (40). With solutions containing chloride or bromide, Hg(ll) determinations were made by the iodide method (4l)- An excess of KI (beyond that necessary to convert a l l the Hg(ll) to the stable Hgl^"" complex) was added, and back-titrated with HgClg to the f i r s t appearance of a red precipitate of Hg'Ig. PROCEDURE The kinetic experiments were made in a cylindrical autoclave (Fig. l ) which was constructed entirely of No. J16 stainless steel. The inside dimensions of the vessel were abput 6 in. diameter by 9 in. height. A stirrer shaft, thermometer well, thermoregulator well and sampling tube, extending nearly to the bottom of the autoclave, were connected through the l i d . When corrosive solutions (such as those containing chloride or bromide) were used, a titanium liner was placed in the autoclave and the stainless steel stirrer, thermocouple well and sampling tube were replaced by i n A - IMPELLER B - GAS INLET TUBE 0 - SAMPLING TUBE D - THERMOMETER WELL B - THERMOREGULATOR WELL P - GAS OUTLET TUBE Pig. 1. Diagram of Stainless Steel Pressure Vessel and Internal Parts -14-titanium counterparts. The solution was stirred with an impeller of 3 in. diameter which was generally rotated at 970 r.p.m. It was ascertained that the stirring was adequate to keep the solution saturated with hydrogen throughout the reaction. The partial pressure of was controlled by a standard gas regulator. The autoclave was heated externally with a gas burner, connected through a solenoid valve which was activated by an electronic relay. Using a mercury thermoregulator the temperature of the solution could be controlled to within i 0.1BC. A smaller autoclave of similar design was used for the experiments at pressures above 10 atmospheres. A solution of desired composition (usually three or four l i t e r s ) , was placed in the autoclave which was then sealed, flushed with nitrogen and heated to the reaction temperature. The solution was allowed to remain in the autoclave under nitrogen for a period of time comparable with the duration of the subsequent reaction to ascertain i t s stability in the absence of hydrogen. Where any reduction of Hg(ll) (i.e. as a result of side reactions with the metal autoclave or with organic complexing agents) •was detected, i t s rate was determined and subtracted from the subsequently measured rate of reduction in the presence of hydrogen. It can be seea from the rate plots that this correction was generally small. The nitrogen was then discharged and hydrogen was introduced and maintained throughout the experiment at a constant partial pressure. Zero reaction time, recorded at the instant of introduction of the hydrogen, may be subject to an uncertainty of up to a few minutes because of the time required for the solutions to become saturated. To follow the reaction, the solution was sampled periodically and analyzed as described earlier. A few isotopic electron exchange experiments, using radioactive mercury, were also made. The procedure employed in these, together with the results obtained, is recorded in Appendix A. -15-RESULTS AND DISCUSSION PERCHLORATE SYSTEM 4"T" The kinetics of the reaction between H^  and Hg were f i r s t +4* investigated in perchlorate solutions, since cations such as Hg or Hg2 are believed to be essentially uncomplexed in this medium. The results of a typical rate experiment are depicted in Pig. 2. No change in solution composition occurred i n the absence of hydrogen. The reaction in the presence of hydrogen can best be described in terms of two clearly defined stages. Stage A corresponds to the reduction of Hg + + to +4* Hgg according to the stoichiometric equation 2Hg++ + H 2 > Hg 2 + + + 2H+ (7) It i s characterized by a progressive decrease in the mercuric ion con-centration, [Hg + +], and an equivalent increase in the mercurous ion con-centration, [Hg 2 + +], the total concentration of dissolved Hg remaining o st nt. This state persists u til about 99% of the Hg has been reduced to Hgg**. 4*4* I I During the following stage, B, the remaining Hg , and the Hg2 formed in the f i r s t stage, are both reduced to metallic Hg, the net reactions being Hg + + + H 2 > Hg. + 2H+ - - - - (8) and Hgg** + H 2 r> 2H£ + 2H+ - r - - (9) 4*4* 44* Thus the concentration of both Hg and Hg2 decrease progressively. The transition between the two stages may be defined unambiguously by the time, t^, at which [Hg2 J, determined analytically, i s a maximum. Visually, t^, whose significance will be discussed later, was characterized by the f i r s t appearance of metallic mercury. Fig. 2 shows that a plot of log [Eg^] against time, during stage B of the reaction, is linear, corresponding to apparent f i r s t order kinetics. -16-TIME - MIN. Pig. 2, Typical Rate Plots for the Reduction of Hg(C10 ) by H in 0.05 M. HCIO^ Solution. 74 . 8 0 C . , 4.0 atm. Hg. (Note compression of time scale for Stage B.) On the other hand the plot of log [Hg ] against time, during stage A, diverges from its i n i t i a l linearity, the " f i r s t order constant" increasing progressively as the reaction proceeds. It will be shown subsequently that ++ this i s due to the catalytic influence of the Hgg which is being formed. The significance of the two reaction stages i s readily understand-able in terms of the relative reduction potentials of Hg and Hgg (B*H^*' ) Hgg** = 0 , 9 1 0 v-» B 8Hg tL^Hg = 0 , 7 9 6 v*) (42), which suggest, in agreement with the observations just described, that the reduction of Hg to Hgg should proceed nearly to completion before the reduction of Hgg to Hg cpmmences. The conditions determining the relative tendencies of these two reactions to proceed, at any point, are expressed by the following equilibrium, which appears to be readily reversible ' Hg(l) + Bs*%=± Hg,,""" (10) whence [Hg 2 + +]/[Hg + +] » K ( l l ) Thus i t seems likely that t^, marking the onset of stage B and the f i r s t appearance of metallic Hg, corresponds to the point at which the ratio [Eg^+]/[Eg++], which increases progressively during stage A, f i r s t attains the value K. The ratio must remain constant at this value throughout stage B, during which the solution i s in contact with metallic Hg. It will now be shown that these results as well as other features of the kinetics are entirely consistent with a mechanism in which Hg is activated i n i t i a l l y by homogeneous interaction, with either dissolved Hg or Hg + + , i.e. ++ k l H 2 + Hg -h X - - - - (12) H2 + ^2 Y (--3) These two independent bimolecular processes are rate-controlling and give rise to active intermediates, X and Y, which undergo further rapid reactions to give the observed products. The total rate of reaction of hydrogen (formally designated as -dtHgl/dt) at any time during either stage A or B, is thus given by -d[H2]/dt = k^Hg^KHg] + k 2[Hg 2 + +][H 2] (14) or, since the Hg concentration was constant throughout each experiment, -d[H2]/dt = ^'[Hg**} + kg'tHgg**] (15) where k^' and kg' are constants for each experiment and are given by V = k l [ H 2 ] = - - - - (16) V = k 2 [ H 2 ] a-k2a!BE2 ( l 7 ) where a is Henry's constant denoting the solubility of hydrogen. The final products are determined by the thermodynamic consider-ations discussed earlier and are different for the two stages of the reaction. Steffi? A: The reaction occurring during this stage is represented by equation 7« Application of equation 15 gives -d[H2]/dt = -VgdtHg+^/dt = k^ [Eg + +] + k 2' [Hg,,**] = k^CHfe**] + k £ ( [ H g + + ] o - [Hg + +]) (18) 2 since [Hg + + J Q = [Eg**] + 2[Hg 2 + +] - - - - ( l9) where [Hg J q is the i n i t i a l Hg concentration. On rearrangement, equation 18 becomes " d [ l f e + + ] : . dt (20) ^ [ H g * * ] + k 2'([Hg + +] o - [Hg**)) j' which yields, on integration " 2 , ? ° ? log (k '[Hg + +] n + (2k,' - k ')[Hg**]) = t + c - (21) ( 2 ^ ' - k2') 2 o I 2 Since [Hg+*j = [Hg**] , at t = 0 C = ^ P J L Q G ( 2 , [ H G + + ] } ( 2 2 ) ( 2 ^ ' - k 2») ^ ° log K ' / c y -v)][*ri0 ] u ( 2 y - y ) t ( 2 3 ) [[Hg**3 + [ ^ 7 ( 2 ^ ' - k2»)][Hg**3o ] 2.303 -19-This corresponds to a linear relation between log 1 [Hg ] + [kg'/(2k1' - k 2')][Hg + +] oJand time. Typical plots of this function are shown in Figures 2 and 3* In no case could any deviation from linearity be detected, although in each experiment the reduction of Hg was followed to well over 90$. As suggested earlier, t^ corresponds to the time at which the ratio [Hg 2 + +]/[Hg + +] f i r s t attains the value K. = K = I * " ' . - ^ - - - ( 3 4 ) [ % + + J Zing**] Hence at t = t„, M 2[Hg++] K = [Eg**] - [Hg + +] o [Hg**] = [Hg + +] 1 + 2 K Substitution in equation 23, gives i o g h 2 k i ' / ^ i ' " V ) 3 f e + % ) = ( 2 V - y ) t [ [Hg + +] 0/(l + 2 K) + [k 2 ' / (2k 1 t - k 2')][Hg + +] o J 2^03 ~ (25) +. 2.303 , (k ' ( l + 2 K) ) ( n r\ \ = log j 1 j . - (26) [2^' -k 2'] (k x ! + K k2* ; The form of this equation is in accord with the observation that t,. is inde-W pendent of the i n i t i a l Hg + + concentration (Fig. 4). Stage B; The reactions occurring during this stage are represented by equations 8 and 9. Application of equation 15 gives -d[H2] . . U[g,~] + dig,, 4*! ) . , r H ~ ! , , r ~ , ( 2 7 ) dt J dt dt j d < L Combining this with equation 11, which defines the ratio [Hg 2 + +]/[Hg + +] throughout this stage, the following expression is obtained -19-( ++ This corresponds to a linear relation between log / [Hg ] + [k 2'/(2k 1' - k2')][Hg++]oJand time. Typical plots of this function are shown in Figures 2 and 3» In no case could any deviation from linearity be detected, although in each experiment the reduction of Hg was followed to well over 9C$. As suggested earlier, corresponds to the time at which the ratio [Hg 2 + +]/[Hg + +] f i r s t attains the value K. teg"! = K = [Hg^J. - [Hg++3 - - - - ( 2 4 ) [Hg + +] 2[Hg++3 Hence at t = t.. M 2[Hg++3 K = [Hg + +3 Q - [Hg++3 [Eg*] = [Hg + +3 P 1 + 2 K Substitution in equation 23? gives log [ [ 2 ^ 7 ( 2 ^ ' -k 2')J[Hg + + 3 Q j = ( , _ f ) , I [Hg""30/(1 + 2 K) + [k2»/(2k1' - k 2 ' )3[Hg + + 3 O ] 2.303 (25) t M = _ ± ^ L _ l o g [ k l ' ( l + 2 K ) j} (26) [2^' - k 2 *3 (k1» + K kg' j The form of this equation is in accord with the observation that t^ is inde-pendent of the i n i t i a l Hg + + concentration (Fig. 4). Stage Bt The reactions occurring during this stage are represented by equations 8 and 9. Application of equation 15 gives -dtV = . U[Hg++3 + *[Hg 2~3 ) = v . [ Hg + + 3 + k '[Hg -3 (27) dt I dt dt j ^ Combining this with equation 11, which defines the ratio [Hg2++3/[Hg"H"3 throughout this stage, the following expression is obtained 1 "2 STAGE A -1.5 STAGE B 5 0 TIME - MINUTES Fig. 3. Rate plots showing effect of varying the i n i t i a l Hg(C10j 2 concentration at 74.8°C.; 4.0 atm. H 2. (Note compression of time scale for stage B). 0 / 25 50 150 250 350 „ TIME - MIN. Pig. 4. Rate plots depicting the variation of Hgg + + in 4*4* experiments with different i n i t i a l Hg ' concentrations. 0.05 M HC104, 4 atm. Hg, 74.8»C. (Note compression of time scale for stage B.) -22--(1 + i) d [ H g 2 + + ] = ( ~v~ + V ) & 0 (28) * dt On integration this becomes i«[*2"'o.<V* + V> t , i V l i l I ) t ( 2 9 ) [Hg 2 + +] 2.303 ( l + 1/K) 2.303 (1 + K) This is in agreement with the first-order disappearance of Hg2 , which was always observed during stage B of the reaction, as shown by the linear plots of log [Hg 2 + +] against time in Figures 2 and 3» The following procedure was employed in evaluating the rate constants k^1 and kg' from the kinetic data. (1) A preliminary value of k^1 was estimated from the i n i t i a l slope of a plot of log [Hg**] vs. _t for stage A. Since i n i t i a l l y [Hg 2 + +] [Hg + +] equation 18 reduces to -d[Hg + +]/dt = 2k^'[Hg++], which becomes on integration, log ([Hg + +] o/[Hg + +]) = 2 k^/2.303 t. The fact that k^ * ^> kg' favours this procedure. (2) This preliminary value of k^' was used to calculate k 2' from the slope of the stage B plot of log [Hg2 ] vs. t, which corresponds according to equation 29 to - (k^ + kg' K)/2.303 ( l + K). Values of K used in this calculation are listed in Table II. Since k 2 K y k^ and K ^  1, values of k^ determined in this way are not very sensitive to errors in either k^ or K. (3) Using this value of kg', the function log [ [Hg + +] + [k 2 V(2k 1' - k 2 ' ) ] [ H g + + ] J was calculated and plotted against time for stage A, giving a straight line whose slope corresponds, according to equation 23, to -(2k^' - k2')/2.303« The final value of k^' was calculated from this slope. If i t differed sig-nificantly from the preliminary value, the whole procedure (i.e. steps 2 and 3) was repeated to self-consistency. -23-TABLB II VALUES OF K AND a USED IN CALCULATIONS OF k. AND k, Temp., °C a ^ (mole 1. 1 atm. 1) 64.4 69.7 7.2 x IO"4 74.8 67.1 7.3 x 10~4 84.9 65.0 7-5 x 10~4 95.5 62.8 7.8 x 10~4 99.5 62.1 8.0 x 10" 4 123.0 - 8.6 x 10~4 a Estimated by a linear extrapolation of log K vs. l/T based on data for the range 0 - 40 9C, of Schwarzenbach and Anderegg (50). b From data for the solubility of H_ in water (43,44). Kinetic experiments were made using different Hg partial pressures ranging from 0 to 5 atm. The rate plots are shown in Fig. 5. Fig. 6 shows that, in agreement with equations (16) and (l7)»values of k^' and kg', estimated from the slopes of these plots by the procedure described above, are directly proportional to the partial pressure of Hg. k^ and kg were calculated from the experimentally determined values of k^1 and kg1 by means of equations (16) and (17) using the values of a listed in Table II. These values were obtained by interpolation from the plot in Fig. 7, which is based on the data of Wiebe and Gaddy (43) and Pray, Schweickert and Minnich (44) for the solubility of hydrogen in water. The small effect on the solubility of the electrolytes present was neglected. Evidence that the solution remained saturated with Hg throughout the reaction and hence that equations (16) and (l7) apply, was provided by the observation that varying the stirring velocity between 870 r.p.ra. and 1000 r.p.m. was without effect on the rate (see Table III). Other aspects of the kinetics also support this conclusion. Values of k^ and kg determined by the above procedure are listed 4*+ in Table III, and are seen to be independent of the i n i t i a l Hg concentrations. Duplicate determinations of and kg generally agreed to within - 5$ and - 3$ respectively. The experimental values of t^ are also listed in Tables III and IV, and are seen to be i n excellent agreement with the values calculated by means of equation (26). Direct evidence for homogeneous character of the reaction is fur-nished by the results of an experiment (see Table III) in which 40 g. of stainless steel powder, of the same composition as the autoclave, was added to the solution; no effect on the reaction rate could be detected. In another experiment (see Table III) 80 g. of metallic mercury ++ was added to the solution i n i t i a l l y . It reacted immediately with Hg STAGS A STAGE B 0 120 240 360 TIME - MINUTES Fig. 5 . Rate plots for the reduction of Hg(C10 / {.)2 at different H 2 partial pressure. 0 .O5 M HClO^j 7 4 . 8 5 C . (Time scales for A and B stages are independent.) [Hg] x 1CT - M/L 0 1 2 3 4 1 i. • '• ! 1 • ~ r i • : V - d LJ --1 1 r i i 0 1 2 3 4 5 P u - ATM. H 2 • Pig. 6. Dependence of the rates of reduction of mercuric and mercurous perchlorate on hydrogen partial pressure. 0.0047 M/L Hg 2(C10 4) 2; 0.05 M/L HC104; 74.8»C. -21-50 / P 40 30 20 10 KhO-O-O- 0 r*r*< P Wiebe and Gaddy - Q Pray et a l . - • 0 • • i 1 1 50 100 150 ' 200 250 300 t, TEMP. - »C. Fig. 7. The Solubility of Hydrogen in Water as a Function of Temperature. Based on Published Data (43, 44). according to equation (10), forming Hgg++, until the equilibrium defined by equation (l l ) was established. The subsequent rate of reaction with Hg was identical with that normally observed for stage B. Results are listed in Table III which show that varying the HCIO^ concentration between 0.025 and 0.10 M, and the NaClO^ concentration between zero and 0.6 M had substantially no effect on either or kg. Addition of larger amounts of NaClO^ up to 1.0 M, appeared to increase k^ slightly (about 10$) but was without effect on kg. Addition of up to 1.5 M HCIO^ had an even more pronounced effect. In general these results are con-sistent with the postulated bimolecular rate-determining steps, represented by equations (12) and (l3)» in which one of the reactants, i.e. Hg, is uncharged. They also provide support for the conclusion that the reactions are those of simple Hg and Hgg ions, rather than of OH or 010^" complexes. Rates were measured at five temperatures ranging from 64.4 to 99»5BC. Good Arrhenius plots, shown in Fig. 9, were obtained when log and log kg were plotted against l/T. These are fitted by the equations k x = 4.2 x 10 1 0 exp [-18100/RT] 1. mole~1sec71 and kg = 1.2 x 10 1 1 exp [-20400/RT] 1. mole^secT 1 These frequency factors and the corresponding entropies of activation, AS = -12.2 - 2 e.u., and AS = -10.2 - 2 e.u. (each at 74.8»C.), are c l c2 normal for simple bimolecular reactions such as those represented by equations (12) and (l3)« TABLE III THE INFLUENCE OF SOLUTION COMPOSITION ON THE RATES OP REACTION OP Hg + + AND Hg 2 + + WITH H 2 Temperature, 7 4 . 8 ° Q . ; Hg Partial Pressure^ 4 atm; Stirring Velocity^ 970 r.p.m. Initial Concentrations - M k l l.mole sec. k2 i i - l -1 l.mole sec. tM~ min. Hg(C104) 2 H C 1 0 4 NaC104 Calc. exp. 0.0055 0.05 0.0 0.168 0.0187 49.6 50 0.0094 II it 0.161 0.0188 51.6 50 0.0148 II ti 0.161 0.0187 51.7 51 0.0268 II it 0.161 0.0188 51.6 50 0.0094 0.025 0.0 0.161 0.0180 51.4 51 ii 0.05 11 0.161 0.0188 51.6 50 1! 0.075 II 0.162 0.0188 51.3 50 II 0.10 tt 0.164 0.0187 50.6 50 II 0.50 tt 0.185 0.0196 46.1 47 II . . 1.00 II 0.190 0.0205 47.1 46 tl 1.50 tt 0.219 0.0210 40-0 42 0.0094 • 0.05 0.0 0.166 0.0185 50.2 50 it it 0.2 0.163 0.0193 51.2 51 II ii 0.4 0.169 0.0188 49.1 47 II it 0.6 0.169 0.0192 48.5 49 I I tt 0.8 0.179 0.0196 46.2 46 n II 1.0 0.183 0.0183 42.8 46 0.0094 0.05 0.0 0.163 0.0185 51.2 52 a " tt n 0.162 0.0192 51.4 53 b " n •t 0.0180 - — c " it it 0.167 0.0185 49.2 51 •d " II it 0.161 0.0188 51.6 50 a 40 g. stainless steel powder added b 8Q g. metallic mercury added c stirring velocity 1000 r.p.m. and baffle added for better mixing d stirring velocity 875 r.p.m. -30-TABLE I? TP INFLUENCE OF TEMPERATURE AND HYDROGEN PRESSURE ON THE RATES OF REACTION OF Hg + + AND Hg + + WITH H [Hg(C10 4) 2J = 0.0094 M., [HC104J = 0.05 M. Temperature H 2 Pressure _ k l _± _k2 lK " m i n * °C. atm. l.mole" sec." l.mole sec." Calc. exp. 64.4 440 0.077 0.0077 112 113 74.8 it 0.166 0.0185 50.0 50 84.9 II 0.361 0.0451 21.8 23 95.5 II 0.697 0.101 10.4 11 99.5 II 0.952 0.129 7.5 8 74.8 1.0 0.166 0.0198 202 196 n 2.0 0.162 0.O19O 103 104 it 3.0 0.163 0.0188 68.1 68 it 4.0 0.173 0.0190 47.8 48 II 5.0 0.170 0.0200 38.9 40 STAGE A STAGE B _ _ . TIME - MINUTES Fig. 8. Rate plots for the reduction of mercuric and mercurous perchlorate at different temperatures. 0.0094 M HgfClO^)^; 0.05 M HC104; 4 atm. H2. (Time scales-for A and B stages are independent.) CLASSIFICATION OF OTHER SYSTEMS The subsequent results refer to the reduction of mercuric salts by-hydrogen in aqueous media other than perchlorate. In general, the mercuric ion was complexed in these systems and, furthermore, the nature of the pro-ducts varied from system to system. In considering the results, i t is convenient to divide the systems studied into three categories depending on whether the reduction product is (A) a soluble mercurous salt, (B) an insoluble mercurous salt or (c) metallic mercury. Each of these categories is characterized by different stoichiometric and kinetic patterns which are described in Table V. Reactions in the f i r s t category are distinguished by the fact that Hg is activated by interaction not only with the mercuric salt but also with the mercurous ions which result from i t s reduction. The resulting kinetics have been described in detail for the perchlorate system. In the reactions of the other two categories, only the mercuric salts contribute to the activation of Hg, in each case apparently through a bimolecular rate-determining step (characterized by the rate constant k^) involving a dissolved mercuric ion or complex and the Hg molecule. As in the case of the perchlorate system ,described above^ the concentration of Hg was kept constant throughout each experiment and hence the observed kinetics were generally f i r s t order (i.e. in Hg(ll)). Values of k^' and k^ were estimated from the slopes of linear kinetic plots, using the appropriate rate equations. In order to f a c i l i -tate kinetic comparison of different systems (since the activation of Hg is rate-limiting in a l l cases) the rate laws defining k^ were uniformly expressed in terms of the equivalent rate of reaction of Hg(-d[Hg]/dt). The latter could be calculated from the measured rate of reduction of Hg(ll) and the known stoichiometry of the reaction. A l l forms of Hg(ll) were found TABLE V CLASSIFICATION OF REACTIONS Category A B C 3. 3. Systems Perchlorate; Nitratej Acetate , Propionate , EDA and EDTA Complexes Sulphate Chloride, Bromide in basic solution Reaction 2Hg(ll) + H 2 -* 2Hg(l) (soluble) 2Hg(lI) + H 2 -* 2Hg(l) Hg(Il) + H2 — Hg Stoichiometry (ppt.) _ Rate-determining Hg(II) + H 2 .-* kj_ k]_ steps k 2 Hg(Il) + H 2 - Hg(II) + H 2 -Hg(l) 2 + H 2 -Rate Law -dOfe(ll)] - 2k7 [Hg(ll)] T H J * -dfog(Il)] = 2k 1[Hg(lI)J [H?1 -dfHR(llJl = k1 (jte(Il)]M dt dt dt zkzfagiDJ [H2] Integrated Rate .- (2k 1/(2k 1 -k 2)[Hg(II ) J 0 ) . ;[Hg(Il)] 0 J H g ( I l ) ] 0 ^ U a t i ° n l 0 g ( [ H g ( I l ) ] ^ [ ^ ( 2 ^ 3 ^ ( 1 1 3 , 1 . L 0 G L ^ " ' ^ [ H ^ l T - (2k rk 2)[H 2] t - 2 k l f o j t = k l [ H J t 2.3 2.3 2.3 a Although mercurous acetate and propionate are appreciably soluble, they do not contribute to the activation of H2. Hence the kinetics conform to those of Category II. -35- . to contribute to the activation of Hg, although with varying rates. Hence with solutions containing more than one mercuric species the apparent value of k^, determined in this way, obviously represents a weighted average of vthe rate constants for a l l the species present. The investigation of the individual systems, the results of which are described below, generally involved the determination of (i) the products and stoichiometry of the reaction (i i ) the kinetic dependence on the Hg(ll) and Hg concentrations ( i i i ) the effects of varying the solution composition, pH, concentration of the complexing agents, etc. CATEGORY A In the reactions of this category the mercuric salt is reduced to a soluble mercurous salt which also contributes to the activation of hydrogen. The kinetics are the same as in the perchlorate system and are described by equation - d [ H j 1/ d[Hg(ll)J ~"dt = " 2 dt = k j H g C l l J H H g ] + k 2 [ H g ( l ) 2 ] [ H 2 ] - T (30) (l) Nitrate System The kinetics of the reduction of Hg(ll) to Hg(l) in nitrate solution were found to be identical with those for perchlorate. k^ was unaffected by variations in solution composition in the range 0.05 - 0.75 M HNO, and 3 0 - 1.0 M NaNO^  (see Table Vi). The extent of complexing in this system (i.e. between Hg** and NO^-) is known to be slight (45) and, as in the perchlorate system, H 2 is apparently activated by interaction with simple ++ —1 Hg and Hg2 ions. At 86°C., the measured value of lc^, 0.397 l.mole sec.-"'", was in good agreement with that determined earlier for perchlorate solutions. The apparent decrease in the value of kg with increasing HNO^  and NaNO^  concentrations, i s probably due to the re-oxidation of metallic -36-TABLE VI REDUCTION OF MERCURIC NITRATE BY HYDROGEN [Hg(NOj)2] = O.Ol M, Temp. = 868C., Hydrogen Partial Pressure = 4 atm. Concentrations - M HNO, NaNO, Lmole^sec. ^ l.mole "''sec: ^  3 3 0.05 0.0 0.428 0.0449 0.10 " 0.394 0.0424 0.25 " 0.388 0.0377 0.50 " 0.416 0.0316 0.75 " 0.386 0.10 0.0 0.394 0.0424 11 0.125 0.397 0.0329 " 0.25 0.395 0.0299 11 0.75 0.397 0.0322 " 1.00 0.387 0.0330 mercury by nitrate. (2) Sulphate System Although the form of the kinetics remained substantially unchanged, the addition of sodium sulphate to a solution of mercuric perchlorate was found to increase the rate of reduction of Hg(ll) to Hg(l) (reflected in an increased value of k^) as shown in Figs. 10 and 11. This is attributed to complexing between Hg + + and SO^- (45) and, by analogy with similar observ-ations in the cupric system, i t seems reasonable to conclude that the constant value of k, (0.69 1.mole-1 s e c - 1 at 86°C.), attained at SO.1- con-1 4 centrations above 0.25 M, represents the rate constant for the activation (of Hg by the undissooiated HgSO^ molecule. Its reactivity i s about 80$ higher than that of uncomplexed Hg ion (See Table a in Appendix B). CATEGORY B In the reactions in this category the mercuric salt is reduced to an insoluble mercurous salt which does not contribute to the activation of hydrogen. The kinetics are described by the equation -d[Hj d[Hg(ll)] 2_ = 1, = k [ H g ( l l ) ] [ H J ( 3 1 ) dt d dt (l) Acetate System The product of the reduction of mercuric acetate by hydrogen is mercurous acetate, .(HggACg), whose solubility was found to range from 0.0015 m/l at room temperature to about 0.02 m/l at 1009C. Under the experimental conditions used, mercurous acetate usually commenced to precipitate from the solution before the reduction of mercuric acetate was complete. The pre-cipitated product was identified analytically a (Found: Hg = 74.0$ C = 10.38$ The carbon and hydrogen analyses were made by Drs. G. Weiler and F.B. Strauss, Oxford, England. -38--4 0 10 TIME - MINUTES Fig. 10. Rate Plots Showing the Effect of Sulphate on the Reduction of Hg(ll) at 86», 4 atm. Hg, in Solutions Containing Initially 0.005 M HgCciO^) 0.02 M HC10,. -39--40-H = 1.46$. Theoretical for E ^ k y Hg = 77.2$ C = 9.25$ H = 1.16$) and by X-ray diffraction measurements 8 (lines found(d-values in &): 11.60, 3.83, 3.45, 2.96, 2.87- Given for Hg 2Ac 2 (46): 13.01, 11.50, 3.86, 3.49, 3.00, 2.90). Some rate plots depicting the reduction of mercuric acetate are shown in Pig. 12. As in the perchlorate system, the reaction is character-ized by two distinct kinetic regions. The i n i t i a l f i r s t order region corresponds to the reaction: 2HgAc2 + H 2 -7 Hg2Ac2: + 2HAc . - (32) As this reaction proceeds,, the Hg(ll) - Hg(l) potential is lowered (as a result of the decreasing'Hg(ll) concentration) and ultimately reaches the point where the reduction of Hg(l) to metallic mercury:becomes preferred thermodynamically. The horizontal portion of the rate plot apparently represents a region in which only the reaction Hg2Ac2 + H 2 — + 2Hg * 2HAc (33) is occurring. The concentration of Hg(l) retains a constant value in this region (determined by the solubility of mercurous acetate) and hence no further Hg(ll) reacts until a l l the precipitated mercurous acetate has been reduced to metallic mercury. The constant ratio of the concentration of Hg(l) 2 to that of Hg(ll), in this region was found to be about 4 at 90.2°C. This is about twenty times smaller than the value of the corresponding ratio in perchlorate solution (see Table Ii) and presumably reflects the fact that a greater proportion of Hg than of Hg2 is complexed by acetate. This is in keeping with the general tendency for Hg(ll) to complex to a greater extent than Hg(l) (47). The X-ray diffraction measurements were made with a Phillips machine using a copper source and a nickel f i l t e r . The constant Hg(ll) concentration in this region is subsequently desig-nated as the "limiting Hg(ll) concentration." :* 0 60 120 180 240 300 360 420 TIME - MINUTES Fig. 12. Rate plots for the reduction of mercuric acetate at 90.2°C., H 2 partial pressure = 4 atm. and [HAc] = 0 . 5 M. -42-In accord with the above interpretation, i t was found (see Fig. 12) that the limiting Hg(ll) concentration ( 0.007 M at 90.2°C.) was essentially independent of the i n i t i a l Hg(ll) concentration (except where the latter was less than solubility of mercurous acetate, i.e. 0.0187 m/l at 90.2°C.), and of the Hg partial pressure (see Fig. 13) • On the other hand, the limiting Hg(ll) concentration increased slightly with temperature, reflecting pre-sumably the increasing solubility of mercurous acetate. Although i t s solubility at the reaction temperatures is appreciable, i t seems f a i r l y clear from the kinetics, that the mercurous acetate product (in contrast with the uncomplexed Hgg + + ion) does not contribute significantly to the activation of hydrogen in this system. In some experiments (notably those at high mercuric acetate con-; centration, low sodium acetate concentration, and low Hg partial pressure) a short "induction period" was noted at the beginning of the experiment before the slope of the f i r s t order rate plot (i.e. k^) attained a constant value (see especially Figs. 12, 13, and 16). The explanation of this effect is not entirely clear. Several possible contributing factors are (a) the presence of impurities in the solution whose reduction competes with that of mercuric acetate, (b) the slow i n i t i a l dissolution of hydrogen and (c) the presence of some incompletely complexed mercuric species, e.g. HgAc+, whose reactivity is lower than that of HgACg. Rate plots for the reduction of mercurous acetate at various Hg partial pressures ranging from 0 to 6 atm. are shown in Fig. 13. Values of k^, evaluated from the slopes of the linear portions of these plots, were found to be independent of the Hg partial pressures (see Table c, Appendix B). Rate plots obtained at different temperatures ranging from 69 to 1159C. are shown in Fig. 14. Values of k,, determined from the slopes of -43-• • ' ' „ * i.' I -20 0 160 320 480 TIME - MINUTES Fig. 13* Rate plots for the reduction pf mercuric acetate at different hydrogen partial pressures. 0.05 M KgAc^; 0.5 M HAc; 90.2«C. -44--1.5. -2 . 0 --40 320 160 TIME - MIN. Pig. 14. Rate plots for the reduction of mercuric acetate at different temperatures. 0.05 M/L HgAc0; 0.5 M/L HAC; 0.25 M/L 480 NaAc; 4 atm. Hg. these plots, yielded a good Arrhenius plot, shown in Fig. 15, which i s fitted by the equation = 1.3 x 10 1 0 exp [-19400/RT] l.mole" 1sec." 1 The apparent activation energy is about 1300 cal./mole higher than that for mercuric per-chlorate (consistent with lower rate) and the frequency factor is again in the range which is considered normal for a simple bimolecular reaction in solution. It seems likely that the above expression for k^ represents the rate constant for the bimolecular activation of Hg by the HgACg molecule, the extent of whose dissociation in aqueous solution i s known to be small (48). At 90.2°C, the value of k^ (0.02 l.mole^sec." 1) is about 25 times smaller ++ than that for uncomplexed Hg ion. Addition of sodium acetate to a solution of mercuric acetate was found to increase the rate as shown in Figs. 16 and 17. The increase- is linear in the acetate concentration and may reflect the formation of higher^ mercuric acetate complexes such as HgAc^" and HgAc^" which are more reactive than HgACg, or, alternatively, i t may be a purely kinetic effect, the possible significance of which will be discussed later. The limiting Hg(ll) con-centration increased slightly with the sodium acetate concentration, ++ presumably due to increased complexing of Hg' . The influence of the concentration of acetic acid on the rate is depicted in Fig- 18 and is seen to be of a fa i r l y complicated form. The rate increases i n i t i a l l y , passes through a maximum at a concentration of acetic acid of about 1.5 M and then decreases with further addition of acetic acid. The reason for this dependence i s not clearly understood. It may involve a multiplicity of factors including a purely medium (i.e. solvent) effect, since i t is not unlikely that the rate of the reaction is sensitive to variations in such properties of the medium as its polarity, basicity (see subsequent discussion) etc. Changes in the solubility of hydrogen would also be expected to influence the rate. -46-I I , 1 , r—. 1 _ . 2.5 2.7 2.9 1000/T°K Fig. 15. Arrhenius plot showing the dependence of for mercuric acetate on temperature. 0.05 M/L HgACg-; 0.5 M/L HAc; 0.25 M/L NaAc; 4 atm, H . -47--30 "0 60 120 180 240 300 TIME - MIN. Fig. 16. Rate plots for the reduction of mercuric acetate by molecular hydrogen at different sodium acetate concentrations. 0.05 M/L HgAc2; 0.5 M/L HAC; 90.2°C; 4 atm. Hg. 0 0.5 1.0 NaAc or NaPr CONCENTRATION - M/L Pig. 17. Dependence of the reduction of Hg(ll) on the concentr-ation of sodium acetate and sodium propionate. Hg(ll) 0.05 M/L; 0.5 M/L HAC or HPr; 90.2°C; 4 atm. H . -.49-* 0 6 ' 12 18 Acetic Acid M/L Pig. 18. Dependence of the reduction of mercuric acetate on the acetic acid concentration. 0.05 M/L HgACgj 0.25 M/L NaAc; 90.2»C 4 atm. H_. The limiting Hg(ll) ^ concentration showed a slight tendency to increase (Table f in Appendix B) with the acetic acid concentration, possibly reflecting an increase in the solubility of mercurous acetate or a change in the Hg(ll) -Hg(l) potential, due to a solvent effect. (2) Propionate System The results of a few experiments on the reduction of Hg(ll) in propionate solution were analogous to those for the acetate system. The reduction product was a . precipitate whose analysis corresponded to that of mercurous propionate. a (Found: Hg = 72.9$ C = 13.47$ H = 1.91$. Theoretical for Hg 2Pr 2: Hg = 73*4$ C = 13.15$ H = 1.83$,) At 90.2°C, the value of (0.021 l.mole ^sec. "'"J-is close to that for mercuric acetate. The effect of the sodium propionate concentration on the rate is depicted in Figs. 17 and 19. The linear increase in the rate with increasing sodium propionate concentration is analogous to, although somewhat greater than, the effect observed in the acetate system (Fig. 17). The limiting Hg(ll) concentration is approximately the same as that in the acetate system. (3) Chloride System In chloride-containing solutions, the reduction of Hg(ll) yields a precipitate of mercurous chloride, whose solubility is negligible. When the ratio of the i n i t i a l concentration of Cl~" to that of Hg(ll) in the solution was 2 or greater, the kinetics of the reaction were consistently f i r s t order in Hg(ll) (see Fig. 20) and in Hg (Fig. 21 and Table i in Appendix B). At 123 QC and a [Cl~] to [Rg(ll)J ratio of 2, ^  was found to have a value of 0.0075 l.mole '''sec. 1 Since complexing in this system is probably complete (49), this presumably represents the rate constant for the bimolecular reaction between H_ and the HgCl^ molecule and is about The carbon and hydrogen analyses were made by Drs. G. Weiler and F.B. Strauss, Oxford, England. -51-I i i NaPr - M/L -40 0 80 160 TIME - MIN. Pig. 19. Rate plots for the reduction of mercuric propionate at different sodium propionate concentrations. 0.05 M/L HgPr2J 0.5 M/L HPr; 90.2*0; 4 atm. Hg. -53-T -40 0 60 .120 180 TIME - MIN. Fig. 21. Rate plots for reduction of mercuric chloride at different hydrogen partial pressures. 0.02 M/L Hg(ll); 0105 M/©92i^; 123»C. -54-++ 400 times smaller than the corresponding value for uncomplexed Hg ion. With further addition of sodium chloride the rate increased slightly (Fig. 22) approaching a constant value for ratios of CI to Hg(ll) greater than about 4-5• The value of in this region, which presumably represents ' the rate constant for the reaction of H^  with HgCl^-, is about JCffo higher than that for HgCl^. With i n i t i a l [Cl~]/[Hg(ll)] ratios between 1 and 2, the kinetics were more complicated, as shown by the rate plots in Fig. 2J. The high i n i t i a l rate is presumably due to a reaction of H^  with incompletely cbmplexed Hg(ll) species such as, HgCl +, i.e. 2HgCl+ + H 2 > Hg 2Cl 2 + 2H+ (34) As the reaction proceeds, the ratio of CI to Hg(ll) in the solution increases until i t reaches a value of 2. Beyond this point the reaction is essentially 2HgCl2 + H 2 * HggCl + 2HC1 _ _ _ _ (35) and, as expected, the f i r s t order rate plots are linear and parallel. If, in solution containing i n i t i a l [CI ]/[Hg(ll)] ratios between l a n d 2, the Hg(ll) is assumed to be present as a mixture of HgCl+ and HgCl 2 a, then the i n i t i a l HgCl 2 concentration is equal to (b - a) where a is the i n i t i a l total concentration of Hg(ll) and b is the i n i t i a l total concentration of Cl~. From equations (34) and (35) i t follows that (b -r a) also represents the total Hg(ll) concentration when a l l the HgCl + has been reduced and only HgCl 2 remains in the solution. (b - a) should therefore represent the Hg(ll) concentration (c) at which the linear portion of the f i r s t order rate plot commences. Table VII shows good agreement between a ++ x The presence of some uncomplexed Hg in the solution as well does not affect the subsequent calculation of C, provided that i t is also reduced to Hg oCl 0. - 5 5 -S 10 3 9 IH O H • H 9 -— oo o — / 0 o — — i 1 1 1 2 3 4 5 6 7 INITIAL RATIO - [Cl~] / [Hg(ll)] Fig. 22. Effect of excess chloride concentration on the rate of reduction of mercuric chloride at 123°Cj 10 atm. Hg. 2 . 3 1 - H 2 INTRODUCED 1 1 INITIAL RATIO [C1-] / [Hg(lljf 1 -1 10 20 30 TIME - MINUTES Fig. 23. Rate plots for the reduction of Hg(ll) in solutions with various i n i t i a l fil-/Hg(H) ratios at 123°C, 10 atm. H2. -57-the values of C calculated as above and the values determined experimentally (i.e. from the plots in Fig. 23) for a series of experiments with different i n i t i a l ratios of [Cl~]/[Hg(ll)]. TABLE VII COMPARISON OF EXPERIMENTAL AND CALCULATED VALUES OF C a b (b-a) C (expt) M. M. M. M. 0.02 0.038 0.018 0.0186 0.02 0.035 0.015 0.015 0.02 0.030 0.010 0.011 Rate plots for a series of experiments, made with an i n i t i a l [Cl""]/[Hg(ll)] ratio of 4.5, at several temperatures ranging from 103 to 130°C, are shown in Fig. 24. Values of k^, calculated from the slopes of these plots by means of equation (31) were found to conform to a good Arrhenius plot, shown in Fig. 25, which is fitted by the equation k x = 2.5 x 10 8 exp [-19100/RT] l.mole" 1sec." 1 (4) Bromide System The results in bromide-containing solutions parallel those for the chloride system. Hg(ll) is reduced to an insoluble precipitate of mercurous bromide. The value of k^ for HgBrg (0.005 l.mole "'"sec."1 at 123*C.) is about 30$ lower than that for HgClg, as seen dm Fig. 20 and in Table 1 in Appendix B. -58w TIME - MIN. Pig. 24. Rate plots for reduction of mercuric chloride at different temperatures. 0.02 M/L Hg(ll); Oi09-M/JS"Cl~j 10 atm. H . 2.45 1 2.55 1 ~ ~ 2.65 1000/T«K Pig. 25- Arrhenius plot showing the temperature dependence of k± for mercuric chloride. 0.02 M/L Hg(ll);, 0.09 M/Ll Cl~ 10 atm. H-. -60-CATEGORX C In the reactions in this category the mercuric salt is reduced directly to metallic mercury which does not contribute to the activation of hydrogen. The kinetics are therefore represented by -d[Hj -d[Hg(ll)] _ _ 2 _ = =k[Hg ( l l)][Hj (36) dt dt (l) Ethylenediamine (EDA) The effect of EDA on the reduction of Hg(ll) was examined in basic solutions where the Hg^DA^^"4" complex is stable. Mercurous salts disproportionate under these conditions so that metallic mercury i s the only product of the reaction. The rate of reduction was found to be f i r s t order in Hg(ll) (see Fig. 26), f i r s t order in Hg (Fig. 27) and independent of the ratio of EDA to Hg(ll) in the solution provided the latter exceeds 2. At 123°C and low OH*" concentrations, k^ was found to be about 0.003 l.mole"*1 sec. - 1 It seems reasonable to conclude that this value (which is about 1000 times smaller than that for the Hg ion; represents the rate-constant for the bimolecular reaction between Hg a n d Hg(EDA)++. It was found, even when an excess of EDA was present, that the rate increased on the addition to the solution of salts of certain anions such as 0H~, Ac"*, Pr~ and CO^-, as shown in Fig. 28. The effects resemble those which were observed in the acetate, chloride and propionate systems^ on increasing the concentration of the anion (X ) beyond that required to convert a l l the Hg(ll) to the stable complex HgXg. In those systems the increased rate seemed attributable to the formation of higher complexes, i.e. HgX^ ~ and HgX^-. However, such an interpretation does not f i t the present case, since the solution already contains an excess of EDA and a l l the Hg(ll) is presumably present as the very stable Hg(EDA) + + complex. -61-TIME - MIN. Fig. 26. Rate plots for the reduction of Hg(ll) at 123<>C., 5 atm Hg, in solutions containing i n i t i a l l y 0.01 M/L HgCClO^Jg, 0.05 M/L EDA and various concentrations of NaOH. -62-TIME - MIN. Pig. 27. Rate plots for the reduction of Hg(EDA) 2 + + at different hydrogen partial pressures. 0.01 M/L Hg(ll); 0.05 M/L NaOH; 0.05 M/L EDA; 123°C. -63-10 O - NaOH A - NaAc V - NaPr • - Na2C05 O- NaC104 0 0.5 1.0 CONC. OF ADDED SALT - MOLE LITER - 1 Fig. 28. Effect of various added bases and salts on the rate of reduction of Hg(ll) at 123°C, in solutions containing i n i t i a l l y 0.01 M/L Hg(C104)2, 0.05 M/L EDA and 0.05 M/L NaOH (in addition to cone, of NaOH or salt shown). Further complexing with anions such as OH , CO^  or Ac is unlikely, although i t is possible that weak ion-pairs of the type Hg(EDA)2++ X are formed. Alternatively i t is possible that the promoting influence of these anions is due to a purely kinetic effect. (2) Ethylenediaminetetraacetic acid (EDTA) A few experiments were made under similar conditions to those above, using EDTA instead of EDA to complex the Hg + +. The resulting rate of reaction between Hg(ll) and H^  could not be measured accurately (because of the com-peting reduction of Hg(ll) by EDTA) but was apparently of the same order as that of the EDA complex. Catpn and Prue (5l) have recently reported evidence for the formation of an analogous ion-pair, C o ( N H 3 ) O H . They determined a value of 0.014 mole l . " " 1 for the equilibrium constant of the reaction Co(NHJg 5* 0H"f=f CotMHJg5* + 0H~ CONCLUSIONS MECHANISM OF THE REACTION (l) Perchlorate System It has been shown that in perchlorate solution the rate of reaction of Hg is determined by two alternative and independent bimolecular processes ++ ++ in which an Hg molecule is activated by interaction with a Hg or Hgg ion. respectively. Prom the observed stoichiometry, i t is clear that the overall reactions (represented by equations 7, 8 and 9) cannot be accomplished in these steps, but that intermediates must be fprmed which undergo further / "f4" 4"+" \ reactions (i.e. with other Hg and Hgg ions) to yield the final stable products. The nature of these products is determined by purely thermo-dynamic considerations. In equations 12 and 13 which have been used to depict the two rate-determining steps of the reaction, these intermediates have been represented as X and Y. It is clear that the kinetic data alone do not permit the identity Of these intermediates to be conclusively established. In cases involving the homogeneous activation of Hg by other metal ions (e.g. Cu+, Cu + + and Ag+) i t has been concluded (9,27) on both kinetic and energetic grounds that the analogous intermediates are hydrogen-carrying species such as CuH+ and AgH*. It i s possible that a similar species, HgH+, is formed in the present system, through reactions such as Hg + + + Hg- > HgH+ + H + (37) Hgg+* + Hg—T> 2HgH+ (38) However i t should be noted that, in contrast to Cu and Ag, the formation ++ ++ of free Hg atoms, through the reaction of Hg and Hgg; with Hg is also very favourable on energetic grounds. Thus available thermodynamic data a indicate that the. following reactions Hg"^ + A Hg+ 2 H+ (39) Hg 2 + + + H 2 }• 2 H g + 2 H + ' (40) are exothermic by 27 and 12 kcal./mole respectively, and i t seems likely that these constitute the rate-determining steps of the reactions. They may be ++ regarded formally as involving a two-electron transfer from to Hg and Hg 2 + + respectively. A two-electron transfer step, leading to the formation of a Mn(v) intermediate, has also been suggested in the reaction between H, and MnO^- (39). In line with the above suggestion, the reduction of Hg4 2 during stage A of the reaction is depicted as occurring through the following -mechanism. (i) Uncatalyzed Reaction Hg + + + H g Hg + 2 H + (slow) Hg + Hg + + — — * Hg 2 + + (fast) Overall Reaction; 2 Hg + + + H 2 » Hg 2 + + + 2 H + (41) (i i ) Hg 2 + + - Catalyzed Reaction Hg 2 + + + H 2 2 Hg + 2 H (slow) 2 Hg + 2 K g " " — 2 Hg 2 + + (fast) Overall Reaction: 2 Hg + + + H 2 r Hgg"1"4" + 2 H* (42) The reactions leading to the formation of metallic mercury during The following values of the standard heats of formation were used in these calculations: Hg*4^ j = 41.6 kcal. (50); Hgg4"4"^^ = 40.0 kcal. (50); H g ^ = -14.6 kcal. (50); = 0 kcal. (50). The solvation energy of the uncharged Hg atoms and Hg molecules was assumed to be negligible. -67-stage B are depicted as follows: k l Hg + + + H. — > Hg + 2 H (slow) Hg > Hg (l) (fast) Overall reaction: Hg + + + H g > Hg (l) + 2 H + - - ^ - (43) Hg 2 + + + H 2 — ^ 2 Hg + 2 H + 2 Hg • » 2 Hg (1) Overall reaction: Hg 2 + + + H,2 » 2 Hg (l) + 2 H + (44) During stage B, the following rapidly-established equilibrium is also operative ^ Hg (l) + Hg + + ; ^ Hg 2 + + (45) During both stages,the rate of reaction of hydrogen, in agreement with the observed results, is represented by -d[H2]/dt = k 1[Hg + +][H 2] + ^[Hgg^lCHg] (46) Evidence for the existence of Hg atoms in appreciable concentration in aqueous solution has been provided by direct measurements (53) of the solubility of •metallic mercury in water (estimated to be about 10 m/l) and by spectro-' scopic observation of the 2537 A" absorption band of Hg in solution (54). Hg atoms have also been postulated as intermediates in. at. least.; two other reactions in aqueous solution. (i) Wolfgang and Dodson (55) have suggested that the very rapid isotopic electron exchange between Hg and Eg^ , i . e i TT + + T T " X ' + + V TT + + T T # + + Hg2 + Hg » Hg + Hg2 a In view of this connection i t seemed of interest also to determine the effect of various complexing anions such as Ac"", CI*", Br~ etc. on the Hg(ll) - Hg(l) exchange reaction. It has previously been shown (55,56) that this exchange is immeasurably fast in perchlorate solution but is slowed down appreciably in the presence of CN~. Measurements were also -68-proceeds through the mechanism ++ Hg + Hg (XI) The reaotion between Hgg< . and Tl In aqueous solution, i.e. Hgg Hg + Hg (rapid equilibrium) Hg + T l * * * — * Hg** + Tl* (alow) The last two reactions are seen to resemble, in reverse sequence the steps whioh have been proposed for the reduction of Hg by Hg to Hffg** (equations 41 and 42). ( 2 ) Bffeot of Compjlexing In view of the uncertainties concerning the nature of the Hg(ll) species whioh are present in complexing media, the kinetic results are some-what more diffioult to interpret in detail than those for perohlorate solutions* Nevertheless, it seems reasonable to oonolude, from the oon» tinuity and overall similarity of the pattern of kinetio results! that the various Hg(XX) oomplexes all activate Hg by essentially the same mechanism as does the simple Hg** ion, although with widely varying rates* the transfer of eleotrons from Hg to the Hg(Xl) species, it would be expected that this step would be influenced by the following considerations* The unfilled low-lying eleotronio orbitals of Hg**, on whioh this eleotron transfer depends, may be used up to varying degrees in forming attempted of the rate of Hg(ll) - Hg(l) exohange in both water and methanol solution, in the presence of Ac", Cl", Br" and l " . The results were negative, sinoe in eaoh oase the exchange was found to be oomplete within the time required for separation* Nevertheless it seemed of interest to record (in Appendix A) the details of these measurements* Since the rate-determining step presumably involves in eaoh oase covalent bonds with complexing ligands, with a resulting decrease in the electron affinity of the Hg(ll) species and hence in i t s reactivity towards hydrogen. The inverse correlation, apparent in Table VIII, between the stability of the complexes and their reactivity, is consistent with this interpretation. The HgSO^ complex, which is more reactive than the simple Hg ion, is relatively unstable and is probably of ionic type. This is •also the case for the various cupric complexes (e.g. GuSO^ , CuACg, CuPr^, CuCl^ etc.) whose reactivities towards Hg are greater than that of the ++ simple Cu ion and which are also much less stable than the corresponding mercuric complexes (37)»a However, the very stable covalent cupric complexes (e.g. with glycine and ethylenediamine) are less active than Cu + +. Apart from the above considerations, the reaction should be favoured by the presence of reagents of high basicity capable of stabilizing the H ions which are released in the rate-determing step. It is probable that this explains the general promoting influence of basic anions on the reaction under conditions such that Hg(ll) i s already complexed to the point where any additional reduction of i t s electron affinity, through further com-plexing, is unlikely. This seems to be the case for the chloride, acetate and propionate systems, when the ratio of the concentration of complexing anion to that of Hg(ll) is greater than 2, and for the EDA system where the promoting influence on the reaction of various anions was found to follow the same order as their basicities, i.e. OH ") ^°3~ ^ ^>r""> Ac" CIQ^ "". To be effective in this way the anion may participate i n the reaction either as a separate entity or by complexing with Hg(ll). In the former case, the This is in accord with the generallyr-recognized principle that ions of metals of the second and third row transition series have a greater tendency to form complexes of the inner orbital type than do those of the f i r s t row transition series (58). TABLE VIII RELATIVE REACTIVITIES AND STABILITY CONSTANTS OF VARIOUS MERCURIC COMPLEXES Complex Stability Constant Reactivity HgS04 22 (45) 1.8 Hg 1 HgAc2 2.7 X i o 8 (43) 4 x IO"2 HgPr2 4 x IO"2 HgCl 2 1.7 X IO 1 5 (49) 2.5 x IO"5 HgBr2 1.2 X IO 1 5 (49) 1.7 x 10~5 Hg(EDA) 2 + + 2.6 X 1025 (63) 1 X IO"5 Equilibrium constant at 25° for the reaction: Hg + + + n A^ 1 Hg(ll).An (Concentrations expressed in mole l i t e r " 1 ) b Value of k^ relative to that for Hg . rate would increase linearly with the anion concentration, while in the latter case a saturation effect might be expected. Both types of behaviour have been observed. ++ Parallel behaviour was observed in the case of Cu where the ++ enhancing influence of the negative ions on the catalytic aotivity of Cu also follows the order of increasing basicity (27). In several other systems (9) involving the activation of hydrogen by metal ions in solution, the activity was found to increase with the basicity of the anion and of the solvent. The fact that the reaction of mercuric acetate with Hg in con-centrated acetic acid is slower than in aqueous solution may also be due to this factor. An alternative interpretation, not readily distinguishable from that just given,has been proposed to explain the catalytic influence which certain anions (e.g. OH*" and Cl~) exert in other electron transfer reactions, such as the exchange between isotopic metal ions of different valence. It has been suggested that the anion forms a polarizable bridge between the two reacting species, thereby lowering the electrostatic repulsion barrier to their mutual approach and facilitating electron transfer between them (59). It is possible that some of the anion influences observed in the present reaction should also be regarded in this light. GENERAL CONCLUSIONS ABOUT THE ACTIVATION OF Hg IN HOMOGENEOUS AND  HETEROGENEOUS SYSTEMS Information is now available about the detailed mechanisms by which Hg is activated homogeneously by a number of species in solution (see Table i ) , and the work in this thesis has contributed such information about two further species (i.e. Hg + + and Hgg + +). At this time i t appears difficult to advance any single type of "mechanism" for the activation of hydrogen which will apply without serious qualifications to a l l the systems that have been - 7 2 -studied. Thus, there are indications that Hg can be activated by a variety of processes including homolytic fission (i.e. Ag+, Cu+, Co0(CO)0) or hetero-lyt i c fission (i.e. Cu + +) of the H-H bond or electron transfer from the Hg molecule to the activating species (i.e. MnO^  , Hg + +, Hgg**). It even appears that superficially similar species, or in some cases (i.e. Ag +, Cu+) a given species under different conditions or in different solvents, can activate Hg by different mechanisms.-If any property emerges which appears to be common to a l l the species which have been observed to activate Hg, i t is that they have high electron-affinities, implying the presence of energetically low-lying, unfilled electronic orbitals (usually of d- or hybrid character). In each case the activation process appears to involve some measure of displacement (not necessarily complete transfer) of electrons from the Hg molecule to the catalyst. The explanation for this probably lies in the fact that the formation of an activated complex involving an electronically saturated molecule such as Hg, generally entails the promotion of electrons into anti-bonding orbitals. Hence a lowering of the activation energy i s to be expected i f the activated complex is coupled with a suitable electron acceptor. An interpretation of catalytic activity along these lines has previously been proposed by Eyring and Smith (6l). It is of interest to consider to what extent these factors also have a bearing on the heterogeneous activation of hydrogen by solid cata-lysts. The suggestion has already been made that the heterogeneous cata-l y t i c activation of hydrogen may involve the displacement of electrons from the adsorbed hydrogen molecule to the catalyst and in several cases detailed mechanisms for the activation process have been suggested (l8) which parallel those which have been proposed for homogeneous systems. In connection with this approach to the explanation of heterogeneous -73-catalysis, particular emphasis has been placed on the availability of vacant low-lying electronic levels in the catalyst (which is almost without exception a metal or semi-conductor) which can act as electron acceptors. In accord with this view i t is observed that the metals with the highest electronic work functions (i.e. the highest electron affinities) such as Ni, Pd, Pt and Rh, are also among the most efficient hydrogenation catalysts. This approach also favours the view that electronic factors are more important than geometric factors in determining the ability of a species to activate Hg. This view receives particularly convincing support from the observations, made in homogeneous systems, that even monoatomic species, notably metallic ions such as Cu and Hg , can activate Hg. Presumably the most favourable electronic configurations, from the standpoint of cata-lyt i c activity, are those which result in the highest.electron affinity, level density etc. , In this connection, i t is of interest to observe that those metal ions which have been found to activate Hg homogeneously in solution, i.e. Cu , Cu , Hg and Ag , are isoeleetronic with the atoms of elements which, in metallic form, are among the outstanding hydrogenation catalysts, i.e. CU -» Co Cu + > Ni Hg*4-—> Ag > Pd In this connection i t is of interest to recall the suggestion made elsewhere (60) that the surface states of metal crystals can sometimes approximate to those of free atoms, and in such cases catalytic activity would be more closely related to properties of atoms than of bulk crystals. Thus i t is difficult to escape the conclusion that (catalytic activity i s linked to electronic factors. In conclusion i t seems worth-while to make some observations about -74-the phenomenon of poisoning of hydrogenation catalysts,, and i t s relation to the effect of complexing on the reactivity of metal, ions such as Hg toward Hg. The most effective poisons for metallic catalysts such as Ni, Pd, Pte etc., are molecules containing atoms of the elements of groups ¥ A and VI A of the periodic table (e.g. As, Sb, S etc.) which possess unshared electron-pairs and are strongly electron-donating. It seems likely that these poisons act by donating their electrons to the catalyst and thereby f i l l i n g up the electronic levels which are essential to catalytic activity. Support for this has been obtained by Maxted and Moon (62) who found that the magnetic susceptibility of palladium decreases on adsbfption of dimethyl sulphide, indicating f i l l i n g of the d-band. The analogy between this interpretation, and that which has been proposed i n this thesis to account for the inhibition of the homogeneous reaction between H and Hg by complexing agents such as Ac*", CI , EDA etc., is readily apparent. -75-REFERENCES 1. F.A. Schaufelberger, J.Metals, 8, 695 (1956) 2. F.A. Forward, Bull.Inst.Metals, 2, 113 (1954) 3. F.A. Forward;and J. Halpern, Trans.Can.Inst.Mining Met., £6, 355 (1953) 4. R.N. O'Brien, F.A. Forward and J. Halpern, Trans.Can.Inst.Mining Met. 56, 359 (1953) 5. R.N. Pease, J.Am.Chem.Soc. 54_, 1876 (1932) R.N. Pease and A. Wheeler, ibid, 5J., 1144 (1935) A. Wheeler and R.N. Pease, ibid, 5.8, 1665 (1936) • 6. D.D. Eley, in "Catalysis" Ed. by P.H. Emmett, Vol. 3. Reinhold Publishing Co. N.Y. (1955). p. 49 7. S. Weller and G.A. Mills, Adv. Catalysis, 8_ (1956) in press 8. M. Winfield, Rev.Pure Appl.Chem. (Australia), 5_, 217 (1955) 9. J. Halpern, Quart.Rev., in press 10. M. Polanyi, Sci.J.Roy.Coll.Sci., X» 21 (1937) 11. D.D. Eley, Adv.Catalysis, 1, 157 (1943) 12. B.M.W. Trapnell, Quart.Rev., 8, 404 (1954) 13. 0. Beeck, Rev.Mod.Phys., 17, 61 (1945) 14. A.A. Balandin, Z.phys.Chem., 2B, 289 (1929) ibid. _3_B, 167 (1929) 15. 0. Beeck and A«W. Ritchie, Disc.Faraday Soc, 8, 159 (1950) 16. A. Sherman and H. Eyring, J.Am.Chem.Soc., %±, 2661 (1932) 17. P.W. Reynolds, J.Chem.Soc, 265 (1950) D.A. Dowden and P.W. Reynolds, Disc.Faraday Soc, 8, 184 (1950) 18. A. Couper and D.D. Eley, DiscFaraday Soc, 8, 172 (1950) 19. 0. Beeck, Disc.Faraday Soc, 8_, 118 (1950) 20. E.B. Maxted, J.Chem.Soc, 1987 (1949) 21. M.H. Dilke, E.B. Maxted and D.D. Eley, Nature, 161. 804 (1948) 22. D.A. Dowden, Research, 1., 239 (1948) J.Chem.Soc, 242 (1950) - 7 6 -23* G.M. Schwab, Disc.Faraday Soc, 8, 166 (1950) 24. D.A. Dowden, Disc.Faraday Soc, 8, 296 (1950) 25. J.CP. Mignolet, DiscFaraday Soc, 8_, 105 (1950) Bull.Soc.Chim.Beigea, 6±, 126 (1955) J.Chem.Phys., 753 U955) 26. M. Boudart, J.Chem.Phys., 2£, 753 (1955) 27. J. Halpern, Adv. Catalysis, v o l . 9 . i n press 28. M. Calvin, Trans.Faraday Soc, .%L, 1181 (1938) J.Am.Chem.Soc, 61, 2230 (1939) 29. S. Waller and G.A. M i l l s , J.Am.Chem.Soc, 2 5 . , 769 (1953) 30. W.K. Wilmarth and M .K . Bareh, J.Am.Chem.Soc, 2 5 . » 2237 (1953) M. Calvin and W.K. Wilmarth, i b i d , 2 8 , 1301 (1956) W.K. Wilmarth and M.K. Barsh, i b i d . J8,, 1305 (1956) 31. L. Wright, S. Weller and G.A. M i l l s , J.Phys.Chem., 5£ , 1060 (1955) 32. W.K. Wilmarth and A.F. Kapuan, J.Am.Chem.Soc, 2 8 , 1308 (1956) 33> H. Adkins and G. Krsek, J.Am.Chem.Soc, 2 P , » 393 (1948) i b i d . 2 1 , 3051 (1949) yv. I . 'v'-snder, M. Orchin and R. Levine, J.Am.Chem.Soc, 72, 4375 (1950) I. Wander, M. Orohin and K.H. Storoh, fold. 22., 4842 (1950) . M, Orchin, Adv. Catalysis, 385 (1955) 35. J.H. Flynn and H.M. Hulbert, J.Am.Chem.Soc, 2&, 3393 (1954) 56. R.G. Dakers and J. Halpern, Can.J.Chem., 969 (1954) E. Peters and J. Halpern, i b i d . 2 1 , 356 (1955) 37. E. Peters and J. Halpern, J.Phys.Chem., 5<3, 793 (1955) J. Halpern, E.R. Maogregor and E. Peters, i b i d , i n press 38. A.H. Webster and J. Halpern, J.Phys.Chera., 60, 280 (1956-) 59. A.H. Webster and J. Halpern, Trans.Faraday S o c , i n press 40. l.M. Kolthoff and V.A. Stenger, "Volumetric Analysis", v o l . 2, Inter-science Publishers, Inc., New York, N.Y., (1947) p. 197 41. I-M. Kolthoff and V.A. Stenger, "Volumetric Analysis", v o l . 2, Inter-science Publishers, Inc., New York, N.Y., (1947) p. 205 42. W.M. Latimer, "The Oxidation States of the Elements and their Potentials i n Aqueous Solution", Second Edition, Prentice-Hall Inc., New York, N.Y., (1952) pp. 175-179 43. R. Wiebe and V.L. Gaddy, J.Am.Chem.Soc, £6, 76 (1934) -77-44. H.E. Pray, C.E. Schweichert and B.H. Minnich, Ind.Eng.Chem., 44. 1146 (1952) 45. C Infeldt and L.G. Sillen, Svensk.Kem.Tid., J|X» 104 (1946) 46. "Cumulative Alphabetical and Grouped Numerical Index of X-ray Diffraction Data". Published by ASTM, 1955, Philadelphia, Pa. 47. N.V. Sidgwick, "The Chemical Elements and their Compounds", Oxford University Press, 1950, p. 292 48. .j. P. Mahapatra, S. Aditya and B. Prasad, J. Ind. Chem. Soc, ^0, 509 (1953) 49. L.G. Sillen, Acta Chem.Scand., ± t 539 (1949) 50. F.R. Bichowsky and P.D. Rossini, "The Thermochemistry of the Chemical Substances", Reinhold Publishing Corp., New York, N.Y., (l95l) p. 69 51. J.A. Caton and J.E. Prue, J.Chem.Soc, 671 (1956) 52. G. Schwarzenbach and G. Anderegg, Helv.Chim.Acta, JJj 1289 (1954) 53- A- Stock, Z.Anorg.Chem., 217, 241 (1934) 54. H. Reichardt and K.P. Bonhoeffer, Z.Electrochem., ^ 6, 753 (1930) Z.Physik, 67, 780 (l93l) 55. R.L. Wolfgang and R.W. Dodson, J.Phys.Chem., £6, 872 (1952) 56. E.L. King, J.Am.Chem.Soc, 71, 3553 (1949) 57. W.C.E. Higginson, A.M. Armstrong and J. Halpern, in press 58. H. Taube, Chem.Revs., 50, 69 (1952) 59. B. Zwolinski, R.J. Marcus and H. Eyring, Chem.Revs., £5_, 1 5 7 (1955) 60. K.G. Pickup and B.M.W. Trapnell, J.Chem.Phys., 25_, 182 (1956) 61. H. Eyring and R.P. Smith, J.Phys.Chem., ^ 6, 972 (1952) 62. E.B. Maxted and K.L. Moon, J.Chem.Soc, 2171 (1949) 63. J. Bjerrum, Chem.Revs., 46, 381 (1950) APPENDIX A ISOTOPIC EXCHANGE BETWEEN Hg(l) AND Hg(ll) EXPERIMENTAL Radioactive mercuric oxide (48d Hg20^, E^ = 0.208 Me?, E^ = 0.279 MeV) of specific activity 4.5 mc./g. and high radiochemical purity was obtained from the Atomic Energy of Canada Ltd., Chalk River, Ont. The sample was dissolved in perchloric acid and diluted with water. For the purpose of these experiments the resulting solution, which was used to intro duce activity into the stock reagent, could be considered as carrier free. Mercurous perchlorate was prepared by shaking together for one hour 30 g. of Hg, 23 g. of HgO, 33.3 g. of 60$ HC104 and 50 ml. of water. It was diluted with water to 2 li t e r s and stored in a dark bottle. A 0.01 M solution of picrolonic acid was made from reagent supplied by Eastman Kodak Co. A l l chemicals used were of reagent or C.p. grade. Glass-distilled water was used in a l l experiments. It is not feasible to separate Hg(l) and Hg(ll) by precipitating the Hg(ll) since this causes dismutation of the Hg(l). Accordingly the separation was made by precipitating the Hg(l). Picrolonic acid and potassium chloride were used as precipitating agents. In each experiment the appropriate amount of mercurous perchlorate was poured into a reaction vessel already containing the requisite quantities of active mercuric perchlorate and complexing agent (i.e. KBr, NaAc etc.). The experiments were made at room temperature and 0°C. After a suitable time interval an aliquot of the solution was poured into a 2-10 fold excess of 0.01 M picrolonic acid or of 0.01 M KC1. The precipitate of Hg(l) was filtered on paper, washed, dried, mounted on an aluminum support and counted. The counting equipment consisted of a standard flow-type proportio-nal counter which was connected with a Tracerlab amplifier and "Ampliscaler". The gas in the counter was P10(mixture of argon and methane) obtained from Nuclear Instruments Corp., Chicago, 111. The geometry of the counter was nearly 2 If. Over a l l counting efficiency was about 20-30%. The counter was operated at atmospheric pressure. The operating voltage was 1600? and the sensitivity 1 mV. The length of the plateau 200 V (1500-1700 V/) with a slope of 2%/l00 V. The counter displayed exceptional stability. (Changes in count from day to day were less than 0.5%') The background count was usually 20-30/min. RESULTS Medium [Hg + +] [Hgg**] Precipitant Temp. Rate of exchange M M °C. H20 5,10"4 5.10"5 Picrolonic acid 0 complete a it 1.10"5 1.10"5 KC1 0 „ II l . K f 5 1.10""5 KC1 0 n 1.10"5 1.10"5 2 KC1 23 II II II 2 K 2Cr 20 7 23 n ti II 2 KC1 0 tt 1.10"4 II 2 KC1 0 n 1.10"4 2.10"4 4 K 2Cr 20 ? 0 » II 1.10"4 2.10^4 2 KC1 0 it -5 1.10 J 1.10"5 2 KC1 0 it 5.10~4 5.10"5 4 Picrolonic 0 t. Complete, in time necessary for separation.( ~ 2 min.). -80-Medium M [Hg 2 + +] [Ac]/ M [Hg**] Precipitant Temp. °C Rate of exchange H20 5.10"4 5.10"5 2 Picrolonic 0 complete II 5 . K T 4 5.10~5 1 11 0 ti II -4 5.10^ 5.10"5 2 11 0 n CH_0H 7 -4 7.10^ 3.10"5 3 KC1 0 11 [CN-]/[Hg++] H20 5.10-4 5.10"5 2 Picrolonic 0 slow it 5.10-4 5.10~5 2 ti 25 slow 1^2^184 min. II 6.10-4 3.10""3 4 ti 25 slow CH,0H 5 6.10-4 5.10"5 3 KC1 0 slow \ / 2 ^ 2 8 0 rain* [Br-J/tHg**] H20 1.10~5 3.10"5 4 Picrolonic 25 complete it 1.10"5 3.10"5 4 11 25 11 it -5 1.10 5 3-10"5 4 11 0 it GH^ OH 6.10-4 3.10"5 2 11 0 it ii 6 . H T 4 3.10"5 2.5 KC1 0 11 [l']/[Eg + +] CH,0H 5 5-10-4 5.10"4 2 Picrolonic 0 11 II 5.10-4 3.10"*5 3 KC1 0 11 [oi'Vtsg**] CH-OH 3 5.10-4 5.10~4 2 Picrolonic 0 tt n 7.10"4 -2 1.10 2 KC1 0 ti it 7.10"4 -2 1.10 * 3 KC1 0 11 II 7.10-4 -2 1.10 c 3 KC1 0 tt -81-APPENDIX B SUMMARY OP SELECTED EXPERIMENTAL RESULTS (a) Effect of Sodium Sulphate on the Reduction of Hg(ll). [Hg(C104)2] = 0.005 M; [HCK>4] = 0.02 M; Temp. = 86°C; Hg partial pressure =4 atm. (Figs. 10 and l l ) [Na2S04] M k l l.mole - 1sec. 1 k2 l.mole sec. 0 0.377 0.0482 0.005 0.410 0.0474 0.01 0.431 0.0475 0.10 0.560 0.0482 0.25 0.695 -O.50 0.690 0.0705 (h) Effect of Initial Mercuric Acetate Concentration on the Rate of i t s Reaction with H . [HAc] = 0.5 M; [NaAc] = 0.25 Mj Temp. = 90.2»C; H 0 partial pressure =? 4 atm. (Fig- 12) [HgAc2] k l M , -1 -1 l.mole sec. 0.01 0.0259 0.02 0.0259 0.04 0.0257 0.05 0.0244 -82-(b) contd. 0.07 0.0258 0.10 0.0241 0.20 0.0249 (q) Effect of Hydrogen Partial Pressure on the Reaction Rate of Mercuric  Acetate with R„. [HgACg] = 0.05 M; [HAc] = 0.5 Mj [NaAc] = 0.25 M5 Tempi = 90.29C. (Fig. 13) Hydrogen partial pressure atm. k l l.mole "'"sec."1 2 0.0230 3 0.0252 4 0.0244 5 0.0240 6 0.0262 (d) Effect of Temperature on the Reaction Rate of Mercuric Acetate with H^. [HgAc2] = 0.05 M;: [HAc] s 0.5 M;. [NaAc] = 0.25 H; Hg partial pressure = 4 atm. (Figs. 14 and 15) Temperature «C. *1 l.mole sec. 68.9 0.0044 78.9 0.0119 90.2 0.0244 100.3 0.0574 U4.9 0.1326 -83-Effect of Sodium Acetate on the Reaction Rate of Mercuric Acetate  with H£. [HgAc2] = 0.05 M; [HAc] = 0.5 M; Temp. = 90.2°C; H 0 partial pressure = 4 atm. (Figs. 16 and 17) [NaAc] k l M -1 -1 l.mole sec. 0.05 0.0199 0.125 0.0204 0.25 0.0244 0.50 0.0307 1.00 0.0393 1.50 0.0510 Effect of HAc on the Reaction Rate of Mercuric Acetate with H_. — — , — , T 11 — 1 ! ! 2— [HgAc2] = 0.05 M; [NaAc] = 0.25 H; Temp. = 90,2°C; H 2 partial pressure = 4 atm. (Fig. 18) [HAc] Limiting [Hg(ll)} M l.mole ''"sec. 1 M 0.25 0.0233 • • 0.50 0.0244 0.0080 1.00 0.0247 0,0080 1.05 0.0270 1.50 0.0298 2.00 0.0288 0.0087 2.50 0.0282 0.0095 3.00 0.0257 5.00 0.0249 -84-contd. .. 7.00 0.0238 0.0995 8.75 0.0219 0.0119 10.5 0.0207 12.2 0.0214 0.0119 14.0 0.0200 15.8 0.0197 0.0120 17.5 0.0178 Effect of Sodium Propionate on the Rate of Reduction of Mercuric • • • > • • . . . . . . . i i i ' i Propionate by H,. [HgPTg] = 0.05 M; [HPr] = 0.5 M; Temp. = 90.2»C. H Q partial pressure = 4 atm. (Pigs. 17 and 19) [NaPr] M - i - i l.mole sec. 0 0.0206 0.25 0.0293 0.50 0.0373 0.75 0.0449 1.00 0.0521 Effect of Initial Concentration of Mercuric Chloride on i t s Rate of  Reaction with H^ . [HCIO^] = 0.1 M; Temp: = 123°C; Hg partial pressure = 10 atm. (Fig. 20) [HgClg] * 1 M -1 -1 l.mole sec. 0.02 0.0979 0.04 0.0945 -85-Effect of Hydrogen Partial Pressure on the Rate of Reaction of Mercuric Chloride with H-. [HgClg] = 0.02 Mf [HC1043 = 0.1 M; [NaCl] = 0.05 M; Temp. = 123°C. (Fig. 2l) Hg part.press. k l atm. 1. mole - 1 sec."*1 5 0.01028 10 0.00979 15 0.01012 Effect of C l ~ Concentration on the Rate of Reduction of Hg(ll) by H. i i . . . . . . , ( | [Hg(ll)] = 0.02 M; Temp, s 123 9C; Hg partial pressure s 10 atm. (Pig. 22) ten k l M -1 -1 l.mole sec. 0.04 0.00736 0.045 0.00741 0.05 0.00756 0.06 0.00804 0.07 0.00845 0.08 0.00879 0.09 0.00950 0.10 0.00911 0.12 0.00988 0.14 0.00903 -86-Effect of Temperature on the Rate of Mercuric Chloride Reduction by H, [HgClg] = 0.02 M; [NaCl] = 0.05 M; Temp. = 123«C; Hg partial pressure = 10 atm. (Figs. 24 and 25) Temp. k l °c. l.mole "''see."*1 103.8 i " 'ii- 1 0.00290 113.8 0.00532 123.0 0.00979 130.2 0.01363 Effect of Br Concentration on the Rate of Reduction of Mercuric Bromide by H„. [HgBrg] = 0.015 M; Temp. = 123°C; H g partial pressure = 10 atm. [KBr] k l M -1 -1 l.mole sec. 0 0.00505 0.01 0.00659 0.03 0.00591 0.06 0.00423 Effect of Hydrogen Partial Pressure on the Rate of Reduction of  Hg(EDA) 2 + + Complex. [Hg(ll)] = 0.01 M; [EDA] = 0.05 M; [NaOH] =? 0.05 M; Temp. = 123°C. (Fig. 27) (m) contd. -87-Hg part, press. atm. -1 -1 l.mole sec. 5 0.0210 8 0.0225 10 0.0241 (n) Effect of Concentration of EDA on the Rate of Reduction of Hg(ll) by H. [Hg(ll)] = 0.01 M; [NaOH], = 0.02 M; Temp. = 122»C; Hg partial pressure = 5 atm. [EDA] h M - i - i l.mole sec. 0.01 0.0402 0.02 0,0393 0.05 0.0452 0.10 0.0464 (o) Effect of Sodium Hydroxide on. the Rate of Reduction of Hg(EDA) by H, [Hg(ll)] = 0.01 M; [EDA] = 0.05 H; Temp. = 123°C; Hg partial pressure = 5 atm. (Figs. 26 and 28) [NaOH] M - i - i l.mole sec. 0 0,0027 0.025 0.0199 0.05 0.0210 -88-contd. 0.10 0.0302 0.20 0.0464 0.30 0.0532 0.40 0.0668 0.50 0.0893 Influence of Different Anions on the Rate of Reduction of HgfEDA)-,4"1"  by Hr. [Hg(ll)] = 0.01 M; [EDA] = 0.05 M; [NaOH] = 0.05 M; Temp. = 123°C; Hg partial pressure = 5 atm. (Fig. 28) [Anion] k^ M l.mole "'"sec.""'" 0.15 NaAc 0.0247 0.45 tt 0.0258 0.95 it 0.0260 0.95 NaPr 0.0267 0.25 NagCO^ 0.0453 0.45 11 0.0597 0.70 ti 0.0606 0.95 11 0.0672 0.45 NaC104 0.0232 0.95 u 0,0230 Effect of Concentration of Sodium Hydroxide on the Rate of Reduction of  Hg(EDTA)2t"f by H^ [ H g ( n ) ] * 0.01 M; [EDTA] = 0.05 M; Temp, = 123°C.;. Hg partial pressure = 5 atm. (q) contd. -89-[NaOH] k l M -1 -1 l.mole sec. 0.3 0.00272 0.5 0.00756 0.7 0.01244 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items