THE DETEFiMIHATIOH OF THE STABILITIES AHD El\fTHALPIES OF DISSOCIATION OF SOME AMMINE COMPLEXES OF PLATINUM(II) By Charles P. Knop A THESIS Submitted to the School of Advanced Graduate Studies of Michigan State Iftiiversity of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Chemistry 1958 ProQuest Number: 10008597 All rights reserved INFORMATION TO ALL USERS The quality of this reproduction is dependent upon the quality of the copy submitted. In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if material had to be removed, a note will indicate the deletion. uest ProQuest 10008597 Published by ProQuest LLC (2016). Copyright of the Dissertation is held by the Author. All rights reserved. This work is protected against unauthorized copying under Title 17, United States Code Microform Edition © ProQuest LLC. ProQuest LLC. 789 East Eisenhower Parkway P.O. Box 1346 Ann Arbor, Ml 48106 - 1346 ACKNOWLEDGMENT The author wishes to express his sincere appreciation to Assistant Professor Carl H* Brubaker for encouragement, advice, and guidance throughout this investigation* Greatful acknov/ledgment is also made to other members ofthe Chemistry Department, faculty, service staff and colleagues, for their cooperation in this investigation and to the Atomic Energy Commission for financial aid. Gratitude is also deeply expressed to his wife, Patricia, and to the other members of his family, including his parents, for their patience and understanding and for the many sacrifices v/hich they endured throughout this investigation. In behalf of each of these, the author humbly offers thisand all succeeding work to our Heavenly Father so that they may receive many blessings, as were benignantly bestowed upon him. VITA Charles P, Knop v/as born in Detroit, Michigan, on May 23, 1927# He received his elementary and secondary education at St# Bernard School in Detroit. He received further education at Aquinas College in Grand Rapids, Michigan and was granted a Bachelor of Science Degree in mathematics in 1952. In 1953 he enrolled at Michigan State University and, subsequently, became a candidate for the Degree of Doctor of Philosophy. His fields of major and minor study were inorganic chemistry, physical chemistry and mathematics, respectively. His professional experience includes 15 months as a chemist at the Haviland Products Company in Grand Rapids, 3 years as a Graduate Assistant and 2 years as a Special Graduate Research Assistant at Michigan State University. He is a member of the American Chemical Society and of the Society of Sigma Xi. He is married and is the father of three children. THE DETERMINATION OF THE STABILITIES AND ENTHALPIES OF DISSOCIATION OF SOME AMMINE COMPLEXES OF PLATINUM(II) By Charles P, Knop AN ABSTRACT Submitted to the School of Advanced Graduate Studies of Michigan State University of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Chemistry Year Approved 1958 ^ ^ > ABSTRACT Dichloro{2 ,2-blpyridine)platinxmi(II), [pt(bipy)CIg], suspended in chloroform, was treated with various anhydrous ammines and a series of complexes were formed which are represented by the general formula [pt(blpy)(An)^]cig where Am is ammonia, methylamlne, ethylenediamine and 1^3-diaminopropane and n is 2 for the former two amralnes and 1 for the latter two. ‘ Tliese complexes, upon heating, dissociate into the original components. The pressure of the dissociated ammine vapor over the solid complex was measured as a func­ tion of temperature and the thermodynamic quantities A H°, and A A P^, S^ for the dissociation v/ere obtained for each complex. Comparison of these thermodynamic values shows that the four complexes fall into two groups with the complexes containing the monoammine ligands in one group and those of the diamine ligands in the other group. The first groun is characterized by large values of A a n d A a n d latter group by relatively small values. By a semi- the quantitative consideration of lattice energies and entropies it is shovm that the large differences in the A a n d values for the two groups cannot arise from this source AS*^ unless the compounds in these groups differ in some way not usually considered, A model is proposed whereby, during the process of heating the complexes, one of the platinumamine bonds dissociates in the complexes containing the diamine ligands and a chloride ion in the complex becomes coordinated. The thermodynamics for this process are not measured or detected by the experimental methods used in the study and the net result is that the thermodynamic values which are obtained for these two complexes refer to the dissociation of the h^rpothetical (at 25°) complexes. Difference in lat­ tice energies and entropies between the two groups, under the assumptions of this model, can account for part of the differences in A K ° and A but a further comparison cannot be made. Minor differences in A a n d A S ° within each of the groups are the result of differences in lattice energies and entropies and in the enthalpies and entropies of bonding of the various ligands to the platinum ion. At 25°C the order of decreasing stability for these complexes is given as CH5NH2 , en and tn, where these symbols refer to the complexes containing these ligands (en is ethylenedianiine and tn is 1,3-diaminopropane), AH° and A This same order is given for From the standard free energies and enthalpies of formation of the free ainmines and their standard entropies and from the thermodynamic values obtained for the dissocia­ tion of the complexes, relative values for the standard free energies and enthalpies of formation (from the elements) and the standard entropies of the complexes v/ere obtained. The order of decreasing AP°|. for these complexes is given as ilH^, MeNHg, en, tn and for en and tn. AH°*^ and S° as CH^NHg, These orders refer to the latter two complexes in a hypothetical state. TABLE OF CONTENTS Page BACKGROUND ............................ The Proble m ........ Th e r m o c h e m i s t r y ......... The Chelate Effect ................. . . . . . . PlatinumCll) Ammine Compounds ................ Dichloro(2,2 »-bipyridlne)platinum(II) ......... 1 1 2 5 12 16 EXPERIMENTAL ..................................... 20 Reagents ............................ Preparation of Compounds ...... Apparatus ...... Vapor Pressure M e a s u r e m e n t ........ Apparatus Calibration .............. .......... Errors 20 20 26 50 35 34 RESULTS ..................... 36 DISCUSSION ............................................ 45 SUIÆMARY........................ 63 LITERATURE C I T E D .......'.............................. 66 APPENDIX ....................... 70 X-Ray Patterns .......................... 78 LIST OP TABLES TABLE PAGE I Factors which determine the magnitude of the bond energy ..... * II Thermodynamic constants for the reaction M(NH5)2 + en =: M(en)**'^ = BNH^ in aqueous solution at 25°C, ionic strength 2.1 III Thermodynamic c o n s t a t s for the reaction M(en)"*"^ f tn = M(tn) + en in aqueous solution at 25°G, ionic strength 0*15 5 10 11 IV Enthalpy changes for the reactions ............ 15 V Enthalpies of solution and lattice energies of some tetrammineplatinum(II) salts .... 16 VI Thermistor calibration data ..... 30 Tq 3 II Constants for ±he equation log P^nm = A-B(—^ ) for compound |pt(bipy) (Am)j^jGl2 .............. 38 VIII Thermodynamic constants for the thermal dissociation of the ammine complexes of [Pt(blpy)01g]at 100®C ......................... 41 IX Thermodynamic constants for the thermal dissociation of the ammine complexes of [Pt(bipy)Cl2] at 25°C ........... ............. 41 X Thermodynamic constants for the reaction at 100*^0 Pt(bipy )Amnl Clp(s ) f mBn(p.) [Pt(bipy )BhjJci2 (3 ) •• ^9 XI Thermodynamic constants for the reaction at 25°C [Pt(bipy)Am_lci2 (s) f mBn(«) [Pt (bipy ) ( 3 )4 n Am{g) ,, XII Lattice energies calculated from the Kapustinskii equation using n = 5,"VpVp = 2 rp = 1 *81and various values of rp ........ 49 50 LIST OP TABLES - continued XIII Standard free energies, enthedpies and entropies of formation for gaseous ammines at 25°C and 100°C ............................... XIV Differences i n A F°f A ,A 3 % and 3° for the pairs rPt(bipy)(Bn )nJ Clo and |Pt(bipy)(Am)_lClp at 25® and 100®C ...... ....................... . XV Observed vapor pressures of diarrmilne (2 ^2 ’-bipyridlne)platinum(II) chloride at various temperatures XVI Observed vapor pressures of diraethy1amine(2, 2 *-bipyridine)platinum(II) chloride at various temperatures ......................... XVII Observed vapor pressures of ethylenediamine(2;2 ^-bipyridine)platinura(II) chloride at various temperatures ....................... XVIII Observed vapor pressures of 1-2 diaminopropane (2y2 *-bipyridine)platinum(II) chloride at various temperatures .......... XIX X-ray powder pattern data ....................... Page 56 57 70 72 74 76 79 LIST OF FIGURES FIGURES PAGE I Apparatus used in preparation of the complexes containing monoammine ligands ...... II Apparatus used for the measurement of the vapor pressure of the solid complexes ........ III Log P versus IV Log P versus V Log P versus VI Log P versus i for [pt (bipy )( i Gig ....... for |[pt (bipy )(GH5NH2 ...* 25 29 71 73 Gig........ 75 for [pt (bipy )trQ Gl^.. .......... 77 i for [pt(bipy )erQ TACKGRCIJKD The Problem Chelate groups, those which are linked thro’igh two atoms to the central atom and so form a. ring, give complexes of exce;'tional stability. Thus, as uerner showed, ethylenedi amine has a much greater pov/er of coordination than ammonia . .. Again, vfhile phenol is a very weak donor - weaker probably than alcohol catechol (o-dihydroxybenzene) is a very strong one, and has been shown to be capable of foiuming complex anions v.hth 27 different metals. Oxalic acid forms far more numerous end stabler co'nnlexes than the mono­ car boxy lie acids such as acetic. The list might be continued indefinitely. So general a tendency must be capable of explanation. Now with one Important class of chelate compounds, the derivatives of ^-diketones, ^ -keto-esters, salicylic acid and the like, there is a special reason for the stability, since the forma­ tion of the ring-complex introduces new possibilities of resonance. But v/ith such groups as ethylenedi amine, catechol and the oxalate group, no new resonance pos­ sibilities are brought in, and some other explanation must be sought. ... N, Sidgv/ick (1) For further development in this field (coor­ dination chemistry) the nost in?gent need is for more quantitative data. ... the heats of formation of many more series of complexes are required. This informa­ tion will be particularly valuable if investigations are confined ... to series of complexes in which only one factor is varying from one complex to the next, ... R . Nyholm (2 ) Solid dichloro(2^£-bipyridine)platinum(II) can be treated with various monodentate and bidentate ligands and a series of complexes can be formed, v/hich v/lll evolve the ligand groups on gentle heating, but the 2,2-bipyrldine will remain bound and the original compound will persist to about 200°C. It v/as proposed to measure the stability of a series of monodentate and bidentate ammine*''* complexes by measuring the dissociation pressures of the solid complexes as a function of temperature. This will provide informa­ tion concerning the stabilities and enthalpies of dissociation of these complexes* Thermochemis try Standard free energy, enthalpy, and entropy changes for a chemical reaction are mutually related to the equi­ librium constant for the reaction, as shown in the follow­ ing equat ions: AF® = - RT'lnK « AH°- T A S ® dlnK = A H ° dT 1 2 The determination of an equilibrium constant is, therefore, of use in obtaining the thermodynamic constants for a '''By the term ammine, it is implied to mean one or more sub­ stances containing one or more nitrogen atoms (in the -3 oxidation state) per molecule, including ammonia. By the term amine, the same is implied with the exclusion of ammonia* reaction. In the reaction for the formation of a complex ion orcompound,ïvIL, from and some ligand, L, the simple ion or compound, M, M 4- L = IÜj 3 Kz 4 v/e define K as ^ML where a is the activity of the species involved. In aqueous solution the activity is a function of the concentration, c, a = y-c 5 and K becomes 6 where Z is the activity cooefficient. Tie value of a function of ionic strength and, hence, practice is also. ^ is In is determined as a function of the ionic strength and upon extrapolation to Infinite dilution the equilibrium constant, K, is obtained (3,4). and A S ® ^obtained by using The ''AF°, AH® in equations 1 and 2 are not thermodynamic values and, as such, they should be used only in comparison with values of other systems obtained under the same condition of ionic strength. It should be remembered that, in the study of the formation of complex ions in aqueous solution, all of the ions are hydrated to some extent and the actual process being studied is the following (5 ): [M(HgO)J + L(aq) = rML(HgO)pl + (n-p)HgO 7 Enthalpy of bonding of the ligand to the metal ion is de­ fined as the enthalpy change for the reaction M*“ (g) + L (g) = [ml]*“1(3 ) Enthalpy e of bonding may differ significantly from the en­ thalpy change for the former process due to possible dif­ ferences in the hydration energies of the ions involved. Differences may also appear in the formation (equilibrium) constants and in the entropy and free energy changes for the two processes. There are several factors Influencing the bond energy. These are listed in Table I, For their studies of the che­ late effect many investigators have chosen the metal-ammine systems. In choosing a given metal-ammine system most of the factors remain constant or nearly so. Factors which may vary are those which involve the ligand molecule. These are (i) the base strength of the ligand and (ii) steric effects of the ligand. TABLE I FACTORS AuICII DETERMINE THE MAGNITUDE OF THE BOND ENERGY (5) 1. The bond hybrid of the metal atom; 2. The relative electronegativities of the donor and acceptor atoms; 3. The underlying electronic structure of the metal atom in its bonded state; 4. The electronic structure of the ligand (possible con­ jugation, etc.); 5. The TDossibility of double-bonding between the metal atom and the ligand; 6. Steric requirements of the ligand and of the metal ion in a given bonding state; 7. Charge on the complex ion and net charge on the central metal ion; 8. Base strength or proton affinity of the donor atom. The Chelate Effect The chelate effect may be described as the increased stability ( A F ° ) of complexes that contain polydentate or chelate ligands over complexes containing corresponding monodentate or simple ligands. By corresponding, it is meant that the simple ligands contain the same donor func­ tional groups as the chelate ligand but there are as many simple ligands as there are donor functional groups on the chelate ligand. This description can be extended to include complexes with tetradentate and corresponding bidentates and the l i k e . Schwarzenbach (6 ) defined the chelate effect quantitatively as chel : log where and - log 9 are the formation constants for the che­ late complex MZ and simple complex IvIAp respectively, Sidgwick (1) was one of the first to offer an explana­ tion for the chelate effect. The most probable (explanation) is the very simple one that if one of the two co-ordinate links of a chelate group to the central atom is broken, the other will still keep it in place so that the broken linli can be re-formed, where as an atom or group which is attached only through a single link will drift away if that link is broken. Since this is a question of probability, it should appear in the entropy term for the system. Schwarzenbach (6 ) and Spike (7) utilized the model suggested by Sidywick as the basis for a kinetic treatment of the chelate effect. These treatments have been summar­ ized by Parry (S) in the following: the forr/iation and dissociation of the nonchelated complex ÎÆA.2 s.nd of the chelated complex MAA are con­ sidered to be step-v/ise processes. It is then logi­ cal to assume that the chelate molecule AA reacts v/ith or dissociates from the metal ion in tv/o steps. 'The intermediate form is a complex in which the che­ lating ligand is bound by only one donor atom. By application of simple collision theory of reaction rates, by assuming a comparable energy of activation for the reaction of chelate and nonchelate structures, and by using the best available data on sizes of molecules, one can estimate the order of magnitude of the entropy term in the chelate effect. It appears from the above models that the rate of the reaction [iIa ] ** * a - [lIAg] * can be related to the size of the volume of element containing one free amine molecule and the rate of the cornu arable re act-ion = (M-A-A )'*'■*' can be related to that volume inside the sphere of radius r * which is available to the second end of the chelating ligand. Commenting on this treatment, Parry states: The above model suggests that the stabiliza­ tion due to chelation should decrease rapidly as the chain of the ligand is lengthened. Schwarzenbach has shown that the difference in free energy of forma­ tion between chelate and nonchelate structures de­ creases rapidly and even reverses in sign as the chain is lengthened. One also arrives at a iustification for the stability of five-membered rings. As a result of steric strain the energy of bond formation is low for small rings but increases as increasing size of the ring relieves strain. On the other hand, the stabilizing influence of chelation, which appears in the entropy term, is greatest for small rings. These two terms, working in opposite directions, pro­ duce a stability maximum in a five-membered saturated ring and in a six-rnembered unsaturated ring, the stereochemistry of v/hich is further restricted by double bond formation. 8 The model also indicates (i) that further re­ striction on the mobility of the second ligand should enhance the stability of the complex if the size of the metal ion is such as to fit into the space be­ tween the binding atoms* (ii) that multiple ring formation should result in enhanced chelate stability, *,, (iii) that the chelate effect should be quite independent of the metal except insofar as special steric requirements of the metal are concerned ,.. , In criticism of this treatment, Irving et a l . (9) showed that it correctly predicts the observed decrease in stability with ring size enlargement but underestimates its magnitude. They state that the decrease in chelate effect of 1-3 diæninopropane compared to ethylenediamine is pre­ dicted to be 0.4 log units which is much less than the ob­ served values of 1,0 for copper (II) and 1,3 for nickel (II) complexes. Also, the predicted change in chelate effect on passing from a six to a seven-membered ring is about 0.3 log units but the values found experimentally are larger, A reaction in which the two simple ligands. A, coor­ dinated to a metal ion, M, are replaced by a corresponding bidentate ligand, Z, may be represented as MA2 + Z « MZ 4- 2A, 10 Irving et a l . pointed out that in this treatment there is the assumption that the enthalpy change for this reaction should be nearly zero. That is, the enthalpy of formation 9 of MA2 and MZ are nearly the same. Prom such an assumption it follows that the magnitude of the chelate effect should be independent of the metal ion involved. Experimental data do not support this general­ ization. Among the more stable chelates of the trans­ ition metals the magnitude of the chelate elect can be closely related to their individual stabilities. Thus, for Go(II), Ni(Il), and Gu(II) we find the linear relations chel(l) = log chel (2 ) =" log K q (e n ) = .20 log K^KgCNH^) + 1.40 K 2 (en) = .32 log K^KqCNH^) 4 2.43 %K^(NH^) For the three metals chel(l) is materially less than chel(2). For Go(Il) and N i (II) we have chel(3) equal 3.54 and 3,56, respectively, which represents still further increase in stability. Such variation can scarcely be coincidental ,,, (which suggests that the treatment) is over-simplified. Spike and Parry (7) measured the entropy and enthalpy changes for reactions of the type + en = M(en )*-2 girgg where M is Gd, Zn and Gu, and en is ethylenediamine, n An examination of the data they obtained (Table II) shov/s that for the cadmium and zinc systems the chelate effect is due to the entropy term. However, in the copper system, the enthalpy term accounts for a large portion of the chelate 10 effect indicating a possible increase in bond strength in the chelate complex as compared to the nonchelate complex, TABLE II THERMODYNAMIC CONSTANTS FOR THE REACTION M(NH^)g + en = M( en ) 4- 2NH? IN AQUEOUS SOLUTION AT 25®G, IONIC STRENGTH 2.1 M AF°(kcal/mole) AH°(kcal/mole) A S ® (e,u.) chel Gd -1.20 0.1 4.3 0.89 Zn -1.55 0.1 5.66 1.14 Gu -4.30 -2.6 5.7 3.15 Cotton and Harris (10) measured the relative effects of entropy and enthalpy for reactions of the type M(en)‘*'^ + tn M(tn)"*’^ + en 12 where M is Gd, Gu, and Ni and tn is 1,3-diaminopropane. the complex ion M( en a f ive-membered ring exists while in the ion M (t n )+ ^ a six-membered ring exists. An examina­ tion of the data they obtained (Table III) shov/s that the five-membered ring is more stable and is favored by the entropy term. The enthalpy term favors the formation of the six-membered ring. In 11 TABLE III THERMODYNAMIC CONSTANTS FOR THE REACTION M( en + tn = M(tn)+^ f en IN AQUEOUS SOLUTION AT 25®C, IONIC STRENGTH 0.15 M AF°(kcal/mole) A H ° (kcal/mole) AS®(e.u.) Cu 1.4 -0.6 -7 Cd 1.2 0.0 -4 N1 1.3 -0.9 -11 For the process shown in equation 10 in aqueous solu­ tion Adamson (11) has proposed that the standard states of the solutes be changed from the hypothetical one molal state, which is conventional, to a state with mole fraction equal to one. The entropy ( A S ® ) and free energy ( A F ® ) changes obtained by use of the former standard state can be con­ verted to those ( AS*® and A F * ® ) of the latter standard state using the following equations: AS*® = AS® A F* ® = AF® where - AnR In 55.5 = AnRT In 55.5 = AS® - 7.9A t| (25®C) 12 AF® - 2360AW (25®C) 13 A n denotes the moles of product minus the moles of reactants. For a reaction in which there is no change in ionic charge, the use of the hypothetical mole fraction unity standard state tends to minimize the translational entropy effect due to A n values. 12 However, as Schwar zenbach (12) has pointed out, dif­ ficulties would be encountered with such a standard state in the case of ionic ligands and, in general, there would be no advantage to such a change. tropy on making the change The decrease in the en­ in standard state does empha­ size that the chelate effect is primarily an entropy effect, Platinum(II) Ammine Compounds The stereochemistry of tetracovalent platinum(II) compounds has been well established by both chemical and physical studies. In almost all such compounds the square planar configuration is formed (13, 14), i.e., the bonds o f the platinum atom to the four groups are directed to the four corners of a square and the ligand atoms of the four groups are coplanar with the platinum atom. The chemistry of divalent platinum, v/hich is essen­ tially the chemistry of its complex compounds, is well known and is discussed extensively in several inorganic chemistry reference books (15, 16, 17). Especially well known is the chemistry of tetracovalent platinum(ll) com­ pounds containing chloride ions and ammines such as ammonia, ethylenediamine, pyridine, etc. These compounds are also y 13 discussed extensively in review articles by Me11or (13) and Quagliano and Schubert (18), The general formula for a series of tetracovalent platinum(Il) complexes containing only chloride ions and ammonia molecules is [ptCrai3)nCl4_n] where n is a positive integer less than five. 14 In general, for a given compound, n may be increased by treatment with aqueous ammonia and decreased by treatment with hydrochloric a.cid or by heating the solid chloride salt. For n equal to two, several isomeric forms of the com­ pound can be prepared, each dependent upon the mode of pre­ paration, Gis-dichlorodiamminenlatinum( II ), p ’t(îîH-)2C1q] (A), where the chioro groups occupy adjacent positions in the square plane configuration, can be prepared by the addi­ tion of ammonia to tetrachloroplatinate(I I ) ion in aqueous solution, Tlie trans form (B), v/here the chi or o groups occupy nonadjacent positions, can be prepared by the re­ moval of ammonia from tetraairLineplatinum( II) ion, [pt (HNg )q] by treatment with hydrochloric acid or by heating the dry claloride salt of the ion (19). Tetraimnine- platinum(II) tetrachloroplatinate (II ) ^Pt (NHg ^PtCl^J (C), 14 a polymerization isomer of [Pt{ÎIH3 jgClg] > can be prepared by mixing solutions containing the respective ions (2 0 ), It has been shoivn that cis^^t fNH3 )oClo] and [ptCNH^)^^ ^PtCl^ are converted to trans ~[Pt (I1IÎ3 )oClo] at about Cl Cl :Pt NH (21). ,NH. Pt. Cl Cl A BE3j ^ NHgj Cl B Ammonia can be replaced by various amines including methylamine, ethylamine, and pyridine and by diamines such as ethylenediamine, 1,2-diaminopropane, 2 ,3-diaminobutane, 1 j3-diaminopropane, and bipyridine. The amine groups in the diamines occupy adjacent positions in the platinum atom coordination sphere. The 1,4-and Ij5-diaminoalkanes, in which the anine groups are separated by more than three methylene groups, form ill-defined products with platinum(II) (19). The chioro groups can also be replaced in the series by other coordinating anions. The thermal stabilities of the tetrammineplatinum(II) halides and the sulfate, chrornate, nitrite, and thiocyanate have been studied (22). The order of decreasing stability for these tetrammines is given as follows: , CrO^ , XI 15 Cl“ , Br"*, NOg, I", SON’*. The order of stability for the diaimaines is about the same. Some thermodynamic data have been obtained for various diaimnines and tetrammines of platinum(II ) by Chernyaev and co-workers. They determined the heat capacities of the cis and trans isomers of |^t ( ) gClg] . The heat capacities follow the equation Cp s 0.00125t + 0.118 15 over the temperature ranges 22-78^G and 22-45^G, respectively(23) Enthalpies of reaction with aqueous ammonia of various isomers having the empirical formula Pt(NH^)2Cl2 have been determined and enthalpies of transition of each isomer to the trans isomer have been calculated. These are summarized in Table IV (24). TABLE IV ENTHALPY CHANGES FOR THE REACTIONS [Pt(KH3)2Gl2]n + SnHHs = n [pt(NHg)^Gig Ah£ [PtCNHsjgClgJn = n ^Hg ^ 1°(e al/g) Gompound trans- fpt( trans - [pt(HH3 jgClg] )oCl-j 52,2 ^Hg(kcal/mole) -- [pt(NH3 )gGlH 62.4 3.0 ± 0 . 2 [ p t ( m 3 )3Gl] jptmisGls] 61.8 5.8 ± 0.3 [pt(lŒ3 )3Cl]g [ptClJ 54.9 2.5 ± 0.3 [pt(HH3 )4] [ptGlJ 59.5 6,6 Î 1.2 16 Ilnthalpies of solution and lattice energies of several platinum complexes have been determined. Tliese are sunmar- ized in Table V for the tetraiTmiineplatinum( II ) salts (25). TABLE V EÎJTHALPIES OP SOLUTION AND LATTICE ENERGIES OP SOME TETRAMIilNEPLATINUM(II) SALTS Salt Enthalpy of Solution ______ (cal/g )__________ Crystal Energy (kcal/mole) Pluoride 388 Chloride -20.95 351 Iodide -28,9 340 Bromide Nitrite 325 -37.1 342 Dichloro(2^ 2lbipyridine)platinum(II) The preparation of dichloro(2,21bipyridine)platinum(II) was first reported by Rosenblatt and Schleede (26) and a short time later by Morgan and Bur stall (27). Morgan and Burstall also described many of the reactions of this sub­ stance with various ammines and anions in aqueous solution. Dlchloro(2,21bipyridine)platinum(II) (henceforth re­ ferred to by the formula |pt (bipy )Clg] where bipy means 2,2 lbipyridine) is a yellow or red fibrous crystalline 17 compound, insoluble in water but slightly soluble in chloro­ form and methylene chloride giving only yellow solutions. The color of the solid material is dependent upon the mode of preparation. The yellow form, which is the most common form, is prepared by the addition of 2 ^2lbipyridine to a hot, slightly acid solution of potassium tetrachloroplatin a t e (I I ), As with almost all tetracovalent platinum(Il) compounds (14), the structure of [pt(bipy)ClgJis square planar, and the ligand atoms of the bipyridine occupy adja­ cent coordinating positions (D). Pt Cl Cl D NH- Clg-xHgO NH. E Diamanine(2,2ibipyridine)platinum(II) chloride can be prepared by dissolving The hydrated solid [pt(bipy)Cl2] in aqueous ammonia. |pt(bipy)(NH^)2]Glp'xHgO (3) is obtained by crystallization from a solution containing excess ammonia or by precipitation from a concentrated solution by the 18 addition of alcohol; a dihydrate and a monohydrate, respec­ tively, are formed. Other complexes with [pt(bipy)Cl2j can also be formed with ethylenediamine, pyridine, and 2 y2 lb ipyr id ine, V/ith the latter two substances the chloride salts were not obtained, but the tetrachloroplatinate(I I ) salts [Pt (bipy )pyg] [ptCl4j and p t (bipy ) \vere. By various treatments the chioro groups, either inside or out­ side the coordination sphere of the platinum atom, can be replaced by other anions such as bromide, iodide, and nitrate, The addition of ammonia and other amines to j^Pt (bipy )C1^ is reversible since they can be removed from the complex with hydrochloric acid, reforming the original compound. color of the The fpt(bip y )Gig] ivhich is reformed varies from yellow to red and is dependent upon the amraine being removed and the concentration of the acid. A bright red color is obtained if concentrated hydrochloric acid is used on the ethylenediamine complex. The bonds of the nitrogen atoms of the bipyridine to the platinum atom are very stable as shov/n by their persist­ ence throughout the various treatments mentioned. The lone exception to this is in the case of the b i s (2^ 21bipyridine) platinum(II) ion from which one molecule of the bipyridine is easily removed by treating it with hydrochloric acid. 19 However, the other molecule remains firmly bonded. Bipyridlne also forms strong complexes with other group VIII metal ions, especially the iron group (iron, ruthenium, and osmium) (28). Burstall and Nyholm (29) sug­ gest that the major factor responsible for the strong bonds found in the iron group is due to the formation of strong double bonds using the outer d electrons of the metal ion and the p orbitals of the nitrogen atoms. This v/ould give rise to resonance structures shown in F and G, Brandt, et al. (28) also add the structures li and J. It is reasonable to assume that some of these resonance structures occrn in these Pt-ll bonds, giving rise to their great stability. P G H EXPERIMENTAL Reagents Reagent grade chemicals were used in the preparation of all compounds listed except those whose grade and/or method of purification are indicated. Gommercially avail­ able anhydrous ammonia and methylamine were condensed on an excess of alkali metal (sodium and lithium, respectively) immediately before being used to remove all trace of water. Ethylene diamine and 1, 3-diaminopropane’“ were purified by the method described by Rolllnson and Bailar (30). 2 ^2lbipyridine The and anhydrous sulfur dioxide were used without further purification. U.S.P. grade chloroform was stored over anliydrous calcium chloride and v/as filtered prior to u s e . Preparation of Compounds Potassium tetrachloroplatinate(I I ), Kg^PtCl^J, was prepared by the reduction of the hexachloroplatinate(IV) salt. Kg [ptClgJ , with sulfur dioxide in aqueous solution as ■'‘ 'Obtained from K and K Laboratories, Long Island, New York. '''■■'‘ 'Obtained from Matheson, Coleman and Bell Division of The Hatheson Co., Inc. 21 described by Keller (31), although the procedure is tedious and time consuming, A more rapid procedure employs hydra­ zine sulfate (32) as the reducing agent. Other suitable methods make use of potassium oxalate (3 3 ) and hydrazine hydrochloride (34) as reductants, Dlchloro( 2,-2ib ipyr idine )platinum( II ), [pt (bipy )Cl2'], was prepared by the method described by Morgan and Burstall (27). A solution containing 0.01 mole of 2^ 2.1bipyr id ine, 0,01 mole of potassium tetrachloroplatinate(II), and 0,04 mole of hydrochloric acid was brought to a boil, whereupon the yellow form of the compound precipitated. The precipi­ tate was removed by filtration, washed with acetone and ether, then allowed to dry in air. This compound was re­ covered from its various ammine complex compounds by pre­ cipitation from a hot aqueous solution of the ammine com­ plex with hydrochloric acid. The product so formed varied in color from yellow to red depending upon the ammine in the complex and upon the concentration of the acid. This is in accordance with the observations of Morgan and Burstall• Diainmine (2^ 2jbipyridine )platinum( II ) chloride, Q"t (bipy ) ( ) g^Gl^, and dimethylainine (2^21bipyridine )platinum(Il) chloride, |pt (bipy )(C % N E o )^ Clg, were prepared 22 by condensing the appropriate anhydrous ammine on a chloro­ form suspension of 1-2 grams of [pt(bipy)GI2] in the appara­ tus shovm in Figure I, The ammine was condensed and dried in Tube A and then recondensed in the reaction Tube B con­ taining the chloroform suspension. had condensed, After 2-3 milliliters the dry ice bath used as the refrigerant was removed from 3 and the mixture in B was allowed to warm to room temperature. During this time any excess ammine dis­ tilled from, the system tlirough a drying tube while the mix­ ture was stirred with a magnetic stirrer. No apparent re­ action occurred until the mixture was near room temperature. After the reaction had occurred (as indicated by a slight, but distinct, color change) the suspended solid was collected on a fritted glass filter, washed successively with chloro­ form, acetone and ether and then air dried. The compoimds obtained by this method contained up to tv;o moles of the ammine in excess of the theoretical quantity required for a diammine complex. This excess ammine was loosely bound in the compound and was readily removed by placing the compound over phosphorous pentoxide in a vacuum. requires Pt, 42.78/; G, 26.32/; N, 3.09/. C, 25.68/; H, 3.08/, ( [pt (bipy )(^11^)2] Gl^ Found Pt, 42.92/; [pt(bipy)(GH3NH2 )2I Clg requires Pt, 40.30/; Cl, 14.64/. Found P t , 40.28/; Gl, 14.66/.) 23 Ethylene diamine (2 y2JLbipyr id ine )platinum( II ) chloride, ^Pt (bipy )en]ci2 , Eind 1, 3-diaminopropane (2,2i.bipyridine ) platinum(Il) chloride, |^t(bipy)tn]Glg,were prepared by the addition of 1-2 milliliters of the diamine to 1-2 graiTis of Q^t(bipy)G l ^ suspended in chloroform. The mixture was stirred for 1-2 hours during vjhich time there was a slight color change in the suspended substance. The solid was collected on a fritted glass filter, washed and dried as with the previous ammine complexes. The compounds prepared by this method contained up to one mole of the diamine in excess of the theoretical quantity required for the com­ pounds with the above formulae. This excess diamine was relatively loosely bound and v:as removed by placing the compound over phosphorous pentoxide in a vacuum. ( [pt(bipy)eiQ Gig requires Ft, 40.47/; Gl, 14.70/, Pt, 40.37/; 01, 14.41/. 01, 14.28/. Pound [pt(bipy)tn] Gig requires Pt, 39.32/; Pound Pt, 39.25/; 01, 14.30/.) Attempts were made to add phosphine, carbon monoxide, pyridinef and acetonitrile"^^ to (pt(bipy)01g] but these were unsuccessful. Phosphine, in a stream of nitrogen, was '"'‘ ’These attempts v/ere made by J . A. Sincius. 24 bubbled through a chloroform suspension of the compound with no apparent reaction. placed Two containers of the compound v/ere in a bomb with relatively pure carbon monoxide at 1.5 atmosphere pressure for several hours with no apparent reaction. In both cases mentioned the lack of reaction was determined by the lack in color change which accompanied the reactions previously described. The procedure for the attempted addition of pyridine and acetonltrile to the com­ pound was identical to that described for the preparation of the diamine complexes. Again, there were no apparent reactions as determined by the lack of color change and weight increase of the starting material. The apparatus shown in Figure I was used for the pre­ paration of the compounds which required gaseous ammines. The system was partially evacuated; then the airmiine was con­ densed in A with a dry ice bath, v/here it was dried with an alkali metal. The dry ice bath was removed and placed at B which contained the chloroform suspension. After all the ammine was recondensed in B, the dry ice bath was removed and the ammine was allov/ed to distill from the container through stopcock 1 and, subsequently, through a drying tube or a cold trap. Also included in the apparatus was a U-tube G CD U'l Cl C3 s < 2 I ë ^ 12 C cl nH % C' o tr^ i-u < >j C c< :r u • C P-i and mercury leveling bulb''/ which acted as a "rough" mano­ meter and a pressure safety device, and a bulb D, which acted as a mercury trap in the event of an unexpected pres­ sure decrease. The stopcocks 2 , 3, and 4 facilitated the handling of the gas and the evacuation of the system. Apparatus The apparatus used for the measurement of the vapor pressure of the various solid ammine complexes of ^ t ( b i p y )Gig] is shown in Figure II, major sections. It consists of three These sections will be referred to as the Sample Section A, the Pressure Measurement Section B, and the Manifold Section G, The sample section consists of a short section of a male S 45/50 sealing tube which has been sealed at both ends. Two side arms have been attached to the tube. One side arm is a S 24/40 sealing tube fitted with a cap through which the sample is introduced, Tne other arm is of smaller diameter and is connected to the manifold section for the evacuation of this section. Each side arm. is sealed off after serving its purpose and resealed to the apparatus prior to the introduction of the next sample, A double bulb sickle gauge is sealed to the tapered end of the tube and serves as an isotenoscope in the system. •X'Not shown. 27 The sickle gauge bulbs were constructed primarily according to the directions given by Phipps, et (35), Two sickle bulbs in series are incorporated in this isoteno­ scope as suggested by the work of Foster (36), The use of more than one bulb permits sturdier construction without sacrificing sensitivity. The sanple section is sealed in a long female 45/50 sealing tube with a high melting vacuum sealing resin. tube is the pressure measurement section. This It is connected to the manifold section through a metal needle valve. A large bore mercury manometer is also connected to this tube. The function of this section is to maintain a pressure identi­ cal to that of the sample section as indicated by the posi­ tion of the tip of a long thin rod attached to the end of the sickle bulb. The pressure in this section is controlled by the use of the needle valve, A reference point for the sickle indicating pointer was constructed on the outside of the pressure measurement section near the tip of the pointer. This reference con­ sists of two narrow slits in a black band painted around the tube, These two slits are aligned with the tip of the nointer so that the alignment is perpendicular to the travel 28 of the pointer, A burette telescope mounted in front of the slits aids in the alignment of the pointer during the pressure regulation. The manifold section is connected to a vacuum pump and to a supply of dry nitrogen gas. It is through this section that the pressure in the other sections is changed, either in their evacuation prior to the vapor pressure determina­ tions, in pressure regulation during the determinations, or in bleeding gas back into them after the determinations are completed . The entire apparatus, as shown in Figure II, is mounted on a movable rack for convenience in Immersing the sajnple section into a constant temperature bath. Several liquid media were tried and used in the con­ stant temperatune bath. Te clinical grade tritolyl phosphate was found to be the most suitable bath media, A thermistor was used for the measurement of the bath temperature. The resistance of the thermistor was measured with a Wheatstone bridge and a sensitive galvanometer. resistance of a thermistor is related to its absolute temperature by the equation (37) IS The FIGURE 11 manomefe*’ APPARr\TUS USED FOR THE MEASUREMENT OF THE VAPOR PRESSURE OF THE S O LID COMPLEXES 30 where R and R q are the resistances at temperatures T and and B is a constant characteristic of the particular ther­ mistor, The constant 3 v/as evaluated by the calibration of the thermistors at 25® and 100®G. These data are summarized in Table VI, TABLF VI THERMISTOR CALIBRATION DATA Thermistor Number b 2 Temperature (® G ) 24,98^ 99.28^ 24,85^ 99.13^ Resistance (ohms) 101300 102600 B (deg)(calculated) a 1 5786 4278 6002 4240 Measured by a Beckmann thermometer calibrated against a platinun resistance thermometer, Temperature of boiling water corrected for barometric pressure. Vanor Pressure Measurement The following is a general description of the proce­ dure which was followed to obtain the vapor pressures of the platinun complex compounds under study. The letters used are in reference to the corresponding letters in Figure II, A 50-100 milligram sample of the solid compound was nlaced in A and the entrance port, a, was sealed off at b , 31 The entire apparatus was then evacuated techniques when possible. using high vacuum The evacuation process was con­ tinued for several hours while the bottom portion of A v/as partially immersed in a dry ice bath to prevent the removal of the ammine ligands from the sample. This precaution, however, prevented proper degassing of the system. After several hours of evacuation the dry ice bath was removed and A was sealed at the constriction at c • was closed and a cap was placed at e. The stopcock d The apparatus was then lowered into a constant temperature bath. The bath was heated to about 50®G and allowed to re ­ main at 50®C for a short time, after which it was increased 5^, The temperature was increased 15-20® daily, in 5^ intervals until a pressure sufficient for measurement was obtained. At this point the temperature was kept constant for several hours before pressure measurements were made. After the pressure measurements were obtained, the tempera­ ture was increased 3-5®, maintained constant for several hours and the pressure v/as again measured. This process was repeated until further readings were either unnecessary or impractical. The pressure in A was measured at each temperature by adjusting the pressure in B, using the needle valve, f, 32 ■until the tip of the reference pointer on the sickle bulb was aligned in the reference slits on the outside of B at g. The press'ure in B was then obtained by measuring the dif­ ference in the heights of the mercury in the two arms of a mercury manometer. These heights were measured with a high precision cathetome ter placed several feet from the mano­ meter, After the pressure in B was obtained, the section was partially evacuated and the process was repeated. Several such independent pressure adjustments in B were obtained and the pressure in B for each adjustment was measured and recorded. Several independent measurements were also made of the resistance of the thermistor in the constant temperature bath, from which the temperature of the bath was calculated, Vi/hen the series of meas-urements were completed for a given sample the apparatus was removed from the oil bath and rinsed with carbon tetrachloride. To open the apparatus one end of a piece of rubber pressure tubing was connected to the system at e while the other end was placed securely over the constriction at h. ('This constriction was pre­ viously scratched with a file edge,) The manifold section, including the rubber tubing, was then evacuated, after which 33 the stopcock i to the vacuum pump was closed. The constric­ tion at h was broken, and the valves d and f v/ere immediately opened. With A, B and G open nitrogen v/as bled into the system through stopcocks j and k until atmospheric pressure was obtained, 'The side arms were resealed to the apparatus at b and c and a new constricted seal was placed at h in preparation for a new sample. Apparatus Calibration The apparatus was calibrated in the follov/ing manner. The entire apparatus was completely evacuated after v/hich a small amount of nitrogen was allowed to enter the apparatus until a pressure v/as obtained which was comparable to the pressures of the samples. With the sections A, B and G open the pressure was measured and recorded. Section A v/as then isolated from B and G by closing the stopcock d. The pres­ sure in B was then adjusted until the references were aligned. This pressure was measured and recorded* Several independent measurements in A were obtained at a given pressure and such measurements were repeated at different pressures. In all, 30-50 independent measurements were made at 3-5 different pressures for each calibration. The differences in the true pressure in A and the apparent pressure in A, as measured by 34 the adjusted pressure in B, were calculated. The average difference was added to the pressures obtained in the actual experimental measurements. Errors For each series of measurements of pressure or tempera­ ture at a given temperature the mean (or average) and stand­ ard deviations v/ere calculated. The standard deviation v/as calculated according to the equation s = Î1 n-1 '(X. -X)® 17 \l where X is the average of n measurements. In almost all cases the value of s for the temperature measurements was found to be 0.3®, which was also its maximum value. The value of s for the various pressure measurements varied from 0.10 to 0,30 millimeter. There is no correlation betv/een the absolute pressure measured and the value of s obtained. It is believed that the variability of s is due to visual and mechanical errors in the pressure adjustments. It was found that if the temperature of the apparatus exceeded 100® by more than a few degrees, the calibration correction value changed by a slight amount. This is believed to be due to a slight shift of the sample section at the 35 standard taper joint when the sealing resin in the joint was sufficiently soft. The sealing resin was found to begin to soften at about 100®C. ITie apparatus v/as calibrated before and after each series of vapor pressure measurements. practice, In if the calibration correction changed, the correc­ tion v/hich was used for that series of measurements was depen­ dent upon the temperature at which the measurements v/ere ob­ tained, For example, for measurements at lower temperatures the calibration correction used was that obtained before the series of measurements was begun, v/hile for the measurements at higher temperatures the calibration correction used was the one obtained after the measurements were made. Corrections due to the difference in the heights of the menisci of the mercury v/ere employed only in the case of the NH3 complex, A small bore manometer (7,7 millimeters) was used for the pressure measurements of this compound. The corrections used were interpolated from the tables listed by VVeissberger (38) and ranged from 0.1 to 0.3 millimeter. For the vapor pressure measurements of the other compounds a wide bore manometer (14 millimeters) was u sed. It was found that menisci corrections for this manometer ranged from 0,01 to 0.03 millimeter. These v/ere ignored. In all cases, the pressures obtained were corrected to the density of mercurv at 0 ®G , RESULTS The vapor pressures of the following four compounds were measured as a function of temperature by the method desci’ibed; (b Ipy )( (bipy )(en )j Clg and )g] Gl g , jpt (bipy )(CHjKHg )gjcig, (bipy )(t n H Clg, v;’nere on is ethyl- enedi amine and tn is 1/o-diaminopropane, These four com­ pounds will generally be indicated by the formula Q^t (bipy )(Am)j^^Clo and specifically as the and tn complex, respectively. CH^NHg, en The vapor pressure data ob­ tained are listed in Tables XV through XVIII in the Appendix. These data include the individual measurements and their averages, the temperature and the inverse absolute tempera­ tures , the calibration corrections which had been applied to the pressure data, and the standard deviations of the aver­ ages which are listed. For the UHg and complexes, the pressures were not measured until a pressure of about 10 millimeters of Hg was obtained. with the en and tn complexes, the pressures obtained v/ere much lov/er than those of the simple complexes, and it was necessary to obtain all the measurements below 15 millimeters. Attempts to obtain higher pressures with 3‘ the en complex at much higher temperatures did not yield satisfactory results. The average of the pressure measurements obtained at each temperatui’e was plotted on a log scale versus the in­ verse absolute temperature (Figures III through VI in the Appendix). The data for that part of the curve which ap­ proaches a straight line were treated according to standard statistical methods (39). The equations obtained are of the form g log P = A - B letting ) 18 10^ X = - r 19 Y = log P 20 and then and A = Y i BX 22 where X and Ÿ are average values and n is the number of ob­ servations. Estimates of the standard deviation of a single measurement (s) and of B ( ) were obtained using the equations (n-2 )s- - 1 X l S r i )^ 38 and ns' 24 The coefficients A and B and the standard deviations s and Sb obtained are listed in Table VII. TABLE VII CONSTANTS THE EQUATION P ™ -= aA JX O rFOR u n X IIX L x L '< ^ u x i 1 ± v i \ JLOG - iU U FOR COMPOUND [Pt (bipy )(Am)npîl2 C omp ound A NH3 11.659 CH3NH2 17.486 B - Sb 3.878 + .059 + 5.791 .118 en 5.675 1.951 tn 3.247 .813 + + .114 .051 Valid Temperature Range +• 0.022 87-110®C 0.023 78-89®C 0.017 114-1250C 1 0.009 97-108®C + + It was found that upon heating [^t (bipy )Gig], from which the various complexes were prepared, there was no significant loss in weight until the temperature exceeded 200®C. It was also found that upon heating the pale yellov/ NHg complex at 100®C, there v/as a loss in weight and a deeper shade of yellow, comparable to that of the [pt(bipy)Gig], was acquired by the complex. ‘ .'Vlth these observations in mind it is assumed that upon heating the 39 complex dissociates according to the equation [p t{b ip y )(IIH 3 )2jc i2 (s) =■ [ p t f b l p y i C l g j t s ) + 2KH3(g) and that the pressure which is measured over this complex is due to the vapor liberated. By analogy, it is assumed a similar dissociation occurs with each of the other complexes studied and that the pressure measured over the complex is due to the liberated ammine vapor. This dissociation process can be represented by the general equation [pt(bipy )(Am) J GlgC Sj) = [ptfbipyJClgQts^ t nAm(g) 26 where n is 1 or 2 , If the process represented in equation 26 is reversible and at equilibrium when the pressure is measured, an equili­ brium constant can be defined for the process as K = where a .A a(g^) ^(.Sp ) 2 is the activity of the species represented. ing theusual gas atunit fugacity 27 Apply­ choice of standard states(hypothetical ideal for gases and pure substanceat one atmosphere pressure for solids), the equilibrium constant expression reduces to the form K = p^ 28 25 40 (where p is measured in atmospheres) if the activity of the solids are unity or are equal. From the equations A F ^ = -RT In K log K = 29 - A 2 .3O 0RT + const. AP^ = AH° - TA 30 31 the thermodynamic constants can be obtained for the disso­ ciation process in equation 26. form of equation 2, assuming A similar in form to equation 18, Equation 30 is the Integrated is constant, and it is Comparison of these tv/o equations (30 and 18) shows that AH° = 2.303-(lO^)-nRB 32 Using equations 18, 29, 31, 32 and the data in Table VII, the thermodynamic constants were calculated at 1 0 0 and 25° and are tabulated in Tables VIII and IX, The data presented in Tables VIII and IX are valid only if each of the following are true: I - the dissociation process is reversible; II - the activity of the two solid substances involved in the dissociation process are equal or nearly equal, and the fugacity of the gas is equal to its pressure; 41 TABLE VIII THERMODYNAMIC CONSTANTS FOR THE THERMAL DISSOCIATION OF THE AMMINE COMPLEXES OP [Pt{bipy)Cl2]at 100°C Log -imnuP ■V 1.265 .022 AF° (kcal/mole) AH° (kcal/mole) 5.52 t .08 35.49 - .54 CH3NH2 1.966 + .027 3.13 1 .09 53.00 - 1.08 en 0.446 .023 4.16 - .04 8.93 t .52 12.8 tn 1.068 .009 3.10 Î .02 3.72 - .23 1.66 Complex + + As°( e .u. ) 80.31 133.6 f 1.47 + 2.9 + 1.4 + .62 TABLE IX ÏHERMODYUAFIG CONSTANTS FOR THE THERFiAL DISSOCIATION OF THE AMvlINE COMPLEXES OF [Pt (bipy )ClgJ at 25°C Complex AF° (kcal/mole) Fog P.m-m ■ + .045 -1.348 11.54 + ,069 13.16 en -0.867 + .093 5.12 tn 0.520 t .036 3.22 NH3 CH3NH2 -1.938 + 4 .12 AH° (kcal/mole) + 35.49 .54 .19 53.00 .13 8.93 .05 3.72 t f + 1.10 AS"^(e.u. ) 80.32 t 1.85 133.6 •52 12.8 .23 1.68 f 3.7 + 1.8 ♦ .81 ‘''"The estimate of the standard deviation s-, is given as i ntl n n X. where X q is defined by equation 19 using the corresponding value of T q . 42 III - the extrapolations of the data for the and en complexes to 100°C are valid (Table VIII only); and IV — the extrapolation of the data for all compounds to 25°C are valid (Table IX only). It was not unequivocally demonstrated that the dis­ sociation processes are reversible. Several times during the vapor pressure determinations, at least once for each different compound, the temperature of the system was lov/ered. The pressure in the system decreased but not to the extent to show the process to be reversible. However, the decrease in pressure was greater than the amoimt predicted by the ideal gas law, indicating that some of the gas mole­ cules were either adsorbed or absorbed by the solid present. This adsorption or absorption can occur by either of two different methods. It may be due (1) to the adsorption of the ammine on the surface of the [pt (bipy )Glg] crystals by reaction with them to form the original complex, with slow diffusion of the ammine molecules through the crystal to react with the inner molecules; or (2 ) to the absorption of the ammine by the undissociated crystals to form solvates, which are readily formed in the preparation of these materials If the first occurs, then the dissociation is reversible 43 but not rapidly so, shov/n by the slow attainment of equili­ brium on temperature decrease. If the second occurs, then it is necessary that the process is reversible; otherwise, the ammine complex would dissociate completely leaving no ammine complex with which to form the solvated complex. Solvates of fpt(bipy)Cl2] are not known to exist. The change in free energy of a pure substance with a change in pressure at constant temperature is given by the equation (^)t= V where V is the molar volume. 33 Integration of this equation, assuming V to be constant, gives AF for a given pressure change. - VAF 34 The error in A F ° for the dissociation process, under the assumption that the activi­ ties of the solids are equal, is given by the expression A F q = (Vi-V2 ) A P 35 where Vq and Vg are the molar volumes of the two solid sub­ stances involved in the process. With AP of approximately one atmosphere and an assumed value of 0.1 liter for Vq-Vg (corresponding to a difference in density of 2-3 grams per cubic centimeter at an average density of 5 grams per cubic 44 centimeter) the error in ^ F ° is 2.4 calories per mole. This is insignificant in this system. The fugacity approaches the pressure as the pressure decreases so that at low pres­ sures they may be assumed to be equal. The short extrapolation of the data to lOO^G for the GHÿTHg and en complexes is not unreasonable and Implies the assumption that A is constant in this temperature range. The extended extrapolation of the data to 25°G for all the complexes also Implies that temperature range. discussed later. AH° is constant over this The validity of this assumption will be DISCUSSION In aqueous solution, it has been shown (7) that ethy1enediamine forms more stable complex ions with various metal ions than does ammonia and me thy 1 amine, Comparison of avail­ able data for 1^ 3-diaminopropane complexes with correspond­ ing ammonia complexes (7, 10) shows that 1^3—diaminopropane complexes are of equal or greater stability than the ammonia complexes but they are not as stable as the ethylenediamine complexes. On the basis of these observations and on the basis that methylamine is a stronger base than ammonia and should form stronger bonds to the metal ion, the following order of increasing stability would be expected in the series of complexes in this study: NH3 < CH^NHg < tn < en. From an examination of the data in Tables VIII and IX, the fol­ lowing orders were found experimentally : at lOO^C, tn < GH5IIH2 < e n < N H 3 ; at 25^0, t n < en < N % < GH^^NH^. differ significantly from that expected. These orders It should be noted from the data that while there is relatively little dif­ ference in A F ^ for these complexes, there is a much larger difference in A and As°. These differences are es­ pecially noteworthy in comparing the values of the CH3 NH2 complexes with those of the en and tn complexes. and An explanation for these large differences should also explain the anomalous stability order. 46 In studying the stabilities and thermodynamic con­ stants of complex ions or compounds, it is desirable to compare the thermodynamic constants for the process in which the gaseous complex metal ion dissociates into the gaseous metal ion and gaseous ligand. This process is hypothetical but the thermodynamic constants can be obtained with the use of a thermochemical cycle (40) if the other values in the cycle are knov/n or can be estimated. All statements concerning enthalpies, entropies and free ener­ gies of bonding are in reference to this process. This re­ action is, in general, endothermie. The thermodynamic constants for the process represented by equation 26 can be treated by a thermodynamic cycle and subsequently shovm to be the sum of several component steps. Such a cycle is given belov; for the enthalpy of dissociation, [pt (bipy )(Am)^]ci2 (s) ---- ^ [^Pt(bipy )(Am)];^]'*'^(g)42Cl-(g) nAam AH°^ [pt(bipy)] + r ^ ( g ) + 2Cl-(g) 37 \/ [^Pt(bipy )Clq] (s ) + nAm(g) Vl/\1/- 2A Cl |]pt (bipy )Cl2] (g ) nAm(g) 47 From this cycle A ^n ■ ^ H ^°A == TT. - TT U J- r^A * n A‘am _ _- 2A„, Cl 38 where Ua and U q are the lattice energies of the solid compounds, and and A^^ are the enthalpies of bonding ner ligand molecule or ion. The component steps of this cycle are endothermie unless indicated by a negative sign preceeding the term* Prom equation 38 we may note that the enthalpy of dissociation is the sum of the differences in lattice ener­ gies and in the enthalpies of bondina of the ligands to the platinum ion. Qualitatively in this series, the enthalpy of bonding of the chloride ligand is probably greater than that of the ammine due to electrostatic considerations, but a positive value in A is obtained because the lattice energy of the ammine complex is greater than that of the [pt(b i p y )C12] compound• If we consider differences in the dissociation thermo­ dynamic constants for the various complexes in this series, we obtain the thermodynamic constants for the following type of reaction, [pt(bipy) (Am) J c i g ( s ) A f mBn(g) = [pt (b i p y } (Bn )jjci2 (s ) + B 48 where Am and Bn are two different ænmines and m or n is 1 or 2. Tne enthalpy change for this reaction is given as Ah tp = A H^a "" ^ H°g 40 = Ua - Ug + nAg^ - ^^bn By analogy, similar equations can be obtained for free energy, AF°rp, and entropy, of A P^qi, A A S°ip, changes. H^^ and A The experimental values S°^ are listed in Tables X and XI, Kapustinskii (41) has given the following empirical equation for lattice energies, independent of crystal structure U = 287.2 (nV-]-Ops rq+rg^ .54 \ " ^1+^2 42 where n is the number of ions per molecule, Vq and '^p are ionic valences and rq and r^ are ionic radii. F'rom this relation one would expect Ua and Ug to be nearly equal if their cationic radii are nearly equal. From lattice energy data for the tetrammineplatinum(II) salts in Table IV and the Goldschmidt values for the halide ionic radii (42) (1.33, 1.81, 1.91, 2,20 Î. for F“ , Cl“ , Br“ and I", respectively) a value of 2.73 was calculated for the ionic radius of the tetrammineplatinum(II) ion. Using the Kapustinskii equation with n = 3, '^q rg = 1.81A and a minimvuii value of 2.70A for r^, lattice ener­ gies v/ere calculated for increasing values of r q . are listed in Table XII. These values 49 TABLE X THERMODYNAMIC GONSTAH :S PGR THE REACTION [Pt(bipy)Aran]ci2 (s) + mBn(g) = [pt (bipy )BnJ Glg( s ) + nAm^ ) AT lOO^C - mBn A F^,j AH°^ 2NH3 - 2CH3NH2 2.39 -17.51 -53.3 2NH3 - en 1.36 26.56 67.5 2NH3 - tn 2.42 31.77 73.6 nj^ A 2GH3N H 2 - en -1.03 44.07 120,8 2GH3ÎTH2 - tn 0,03 49.28 131.9 1.06 5.21 11.1 en - tn TABLE XI THERMODYNAMIC CONSTANTS FOR THE R U C T I O N [Pt (bipy )AmJ GlgC s ) + mBn(g) = [pt (b ipy )BniJ Clg (3 ) + nAm(g) AT 25°C A F % AH°T -1.62 -17.51 -53.3 2NH3 - en 6.42 26,56 67.5 2NH3 - tn 8.32 31.77 78.6 2CH3NH2 - en 8.04 44.07 120.8 2CH3NH2 - tn 9.94 49,28 131,9 en - tn 1.90 5.21 11.1 nAm - iTiBn 2NH3 — 2CH3NH2 AS°^ 50 TABLE XII LATTICE ETIEROIES GALCBLATED PROM THE ITVPUSTINSKII EQUATION USING n = 3, = 2, = 1,81 AMD VARIOUS VALUES OF Pq U(kcal/mole) ^1 rq U( kcal/mole) 2,70 353 3.10 327 2,80 346 3.20 321 2.90 340 3.30 315 3.00 333 3.40 309 3.50 304 These calculations show that where cationic radii dif­ fer by 0 .loX, the lattice energies will differ by only 5-7 kilocalories 0.7-0.8/t per mole while a difference in radii of is necessary to give a lattice energy difference of 40-50 kilocalories per mole, Altshuller (43) has calculated lattice entropies for several simple salts. He has shown that lattice entropies vary from 40 entropy units per mole for the alkali halides to 85 entropy units per mole for salts containing oxyanions, He has also shown, however, that for a given series of salts the lattice entropies vary by 5 entropy units or less. Lattice entropy is the change in entropy upon evapora­ tion of an ionic crystal to gaseous ions. The entropy of a 51 gaseous substance can be partitioned into translational, rotational and vibrational entropies. Differences in lat­ tice entropies will be reflected by differences in these en­ tropies, assuming the entropies of the solid to be equal. Differences in translation entropies under the same conditions of temperature and pressure are proportional to the differ­ ence in In M where M is the molecular (or ionic) weight. Differences in rotational entropies of substances with the same symmetry and at the same tempera.ture are proportional to the difference in In Iqlglg where Iq, Ip and are the moments of inertia about the tliree axes of rotation. Dif­ ferences in vibrational entropies depend in some way on the difference in vibrational frequencies and on the number of degrees of freedom within the molecule. [pt(bipy )(Am) J t r a n s l a t i o n a l For the series entropies differ by a maximum of 0,3 entropy units and rotational by less than 5 entropy units assuming a difference in radii of 1 5 and a spherical ball model for the ion in calculating mo­ ments of inertia. Vibrational frequencies are unknown for this series; how^ever, vibrational entropies are relatively small and the difference in them will be smaller. On the basis of these observations and upon the calculations of 52 Altshuller*, it is reasonable to assume that lattice entrop­ ies of the series under study will agree within 5 entropy units * Because of the relative constancy which one would ex­ pect to find in the lattice entropies of the series l^t(bipy)(Am)jJci2 , bhe difference in lattice free energies should be similar to the difference in lattice energies. On the basis of these observations, the large dif­ ferences in A H ® values listed in Tables VIII and IX can­ not be accounted for alone in terms of lattice energy dif­ ferences since correspondingly large differences would also occur in A P®. Some other explanation must be sought for the apparent abnormal stability order found in this series. To explain the large differences in AH® and A S® between these two groups of complexes (the NII3 and CHgNHg complexes in one group and the en and tn complexes in the other group), it is necessary to propose that the two groups are two different types of compounds or modifications of the same type, so that the process which occurs on heating dif­ fers in some way. It is possible to consider one of these groups as normal complex compounds, as previously assumed, and the other group as ammine solvates of the parent 53 £pt(bipy)Gig], in which the ammine ligands are more loosely bound. However, in this model the vapor pressure of the sol­ vate compound should be higher at a given temperature while the A h ® should be lower than those of the complex com­ pounds, a combination which is not found in this series. It is possible that the four complexes in this series have formulas as previously assumed with the exception that the diamine ligands are bonded to the platinum ion through only one amine group. In this model the diamine would be functioning as a monodentate ligand and, as such, these complexes should exhibit higher vapor pressures in the temp­ erature range v;here the NH3 and GH^HHg complexes were stud­ ied, Tliis model may be modified slightly by assuming that the diamine ligands function as bidentate ligands at lower temperatures but at a higher temperature, near but lower than the temperature range in T/hich these complexes were studied, one of the amine bonds is broken. Above the temp­ erature at which this occurs, the diamine ligand functions as a monodentate. At this temperature the heat capacity and entropy of the solid complex should increase. would account for the much lower AH® and A This S° for these compounds since these increases are not detectable by the 54 experimental methods used in this study. This latter model is proposed as the most likely to explain the data obtained in this study. In this model it can be further assumed that when the one bond of the diamine to the platinum is broken, one of the chloride ions in the compound becomes coordinated to the platinum. Under this assumption a thermochemical cycle treatment would modify equation 41 to the form - Up + 2Aq^ ” =• Ua bn “ 4]1 when comparing the ITH3 or GHpUHp (A) complex to the en or tn (B) complex. However, the coordination of the chloride ion occurs before the experimental measurements are made so that these measurements do not include in this process. The term, A*p^, refers to the breaking of only oneamine bond and, in this consideration, equation 41 should be modi­ fied by substituting A ’ for A for the terms representing the diaiîiine ligands. Lattice energy and entropy considera­ tions discussed previously are not valid for all four com­ plexes but only within either of the two separate groups, Hence, lattice energies and entropies for the l^THp and GH5H H 2 complexes are similar and also those en and tn com­ plexes are similar but a simple relation cannot be made between the two groups. 55 By use of Hess* law, the standard thermodynamic con­ stants for the formation of the free ammines in the gas phase and their entropies and the thermodynamic constants for the process in equation 39 ( A A H®^ and AS°^), the difference in the standard thermodynamic constants of formation for the various pairs of complexes can be calcu­ lated along with the difference in standard entropies. The equation to be followed for the standard enthalpies of forma­ tion is given as example A hQ b - A Some thermodynamic h P^ = A H ° ^ - n A n q ^ ^ t n A P i P g ^ . 44 data are available for the pure ammines in the literature (44), Data not available were calculated by empirical methods of estimating standard enthalpies of formation and entropies, and heat capacities (45), data, listed in Table XIII, appearing in Table These were used in the calculations XIV from the above equation and the analo­ gous equations for the standard free energies and entropy differences. 100° For the free arar;iines, the estimated data at are less reliable than those at 25°C and the A F®^. estimates are relatively poor since they contain an accumu­ lation of the errors found in the estimates of and A S®p* A H®^ 56 TABLE XIII STANDARD FREE ENERGIES, ENTHALPIES AND ENTROPIES OF FORMATION FOR GASEOUS AMMINES AT 25°C AND 100®G Ammine AF®p AH®p (kcal/ (kcal/ mole) mole) Ammonia -3.98 -11.04 Ammonia -2.16"'" -11.42''""" AS®^ (e.u,) -23.69 3° (e.u.) 46.01 T®C 25 -24.82""''' 47.98""" 100 Methyl amine 6.6 -6.7 Methylamine 10.0"" -7.3" -46.27"" 60.71" 100 25 -44.54""*"' 57.73 25 Ethylenediamine 26.8"" -5.7"‘ -109.0“ 64.3 ' Ethylenediamine 35,1"" -6 .8 “ -112.4“ 69,7' 100 l-3Diaminopropane 28,3"" -10.9“ -131.4"" 74,5' 25 l-3Diaminopropane 39.2“ -11.3“ -135.4“ 81,3' 100 “Estimated by empirical methods (45) ■“"■“Gal cud.a ted from available data (44) 57 TABLE XIV DIFPERSMCES IN A P U , A H U , A S°^_ AND S° FOB THE PAIRS |_Pt(bipy) ( B n ) q c i 2 |Pt(blp3 )(Am)^]Clg AT 25° AND 100°C A AH®q.T Complex Pair Bm - An kc^ll 25® 100® ^kâir 25® 100® A sQ b; 25° 100® 25® L) ^ 100° CH3NH2 “ NH3 19.6 16.7 -8,8 -9,3 -95.0 -96.2 -29,8 -27,9 en — NH3 41.1 40.8 43.1 42,6 5.9 4.7 39.8 41.2 tn - NH3 44.6 45.9 43,0 43.3 -5.4 -7,2 61.1 63.9 en — GH3NH2 21,6 14.2 51,8 51,9 100.9 100.9 69,6 69.1 tn - CH3NH2 25.0 19.2 51.0 52.6 89,6 89.0 90.9 91.8 3.4 5,2 0,0 0,7 -11.3 -11.9 21,1 22.7 tn - en Of the various factors affecting -the enthalpy of bond­ ing of a ligand in a complex (Table I) all are constant in the series of complexes in this study, with the exception of the basicity and steric requirements of the four ammine ligands. As the base strength of the ligand increases the enthalpy of bonding increases and the bond tends to be more stable. In aqueous solution the order of increasing basicity of the free aimnines is given as (46) e n t n < GH3NH2 • Consideration of steric factors should be included in a 58 consideration of lattice energies since these ligands should have no apecial steric ninderances in foririing bonds in the gaseous platinum complex. A comparison of the enthalpies of bonding in this series can be obtained from equation 41 if the lattice energ­ ies involved in the relation are known or can otherwise be treated in a semi-quantitative way. This comparison can, at best, only lead to relative values of the enthalpies of bonding. For the NI13 and CH3NH0 complexes, as 3inning the lattice energies are equal, the enthalpy of bonding of the CH3NHP ligand is 8.75 killocalories (J-AK®p) greater than that of the WH3 ligand. This is predicted qualitatively by the relative basicities of these aimnine ligands. Using the Kapustinskii equation as a good estimate for lattice energies, the ass'oinption that the lattice energies of these two complexes are equal implies that their cationic radii are also equal. Since the CH3NH2 ligand is slightly larger than the NII3 ligand, it can be assumed that the cation in which it is bound is also larger than corresponding ion in the NIig complex and the lattice energy of the CH3NII2 complex should, therefore, be slightly less than that of the NH3 complex. The 8,75 kilocalories difference in the 59 enthalpy of bonding is a lower lirait and the actual limit may oe several units larger, (The lower lattice energy of the GH3NH2 complex may be considered as a consequence of some steric hindrance of the ligand in the solid complex.) For the on and tn complexes, again assuming equal lattice energies, the enthalpy of bonding of the on ligand is 5.21 kilocalories ligand. ( A H°q,) greater than that of the tn This is in disagreement with the relative enthal­ pies of bonding predicted by basicity considerations, Hov/- ever, the lattice energy of the tn complex should be less than that of the en complex. The 5.21 kilocalories dif­ ference is an upper limit and may be completely compensated by a larger difference in the lattice energies. Under the assunptions of the proposed model of this system it is not possible to compare the enthalpies of bond­ ing of the simple ammine ligands to those of the diamine ligands. Assuming equal cationic radii in the two types of complex, the resulting lattice energies would be in a ratio of 3 to 1 , with that of the simple ammine complexes being greater, showing that lattice energy differences are a major contribution to AH°ip, Further, if the lattice energy 60 differences could be accounted for quite accurately the re­ sulting value Y/ould represent the difference in the enthalpies of bonding of tv/o simple ammine ligands and one bond of the diamine ligand. Prom Table XIV the follov/ing order of increasing standard enthalpies of formation is found for the complexes in this series: CH^NHg < N H ^ < en < tn at both 25® and 100° C . In the dissociation process for these complexes, a major contribution to A S ® the liberated aimnine. should be due to the entropy of Prom Table XIII the entropies for the number of moles of ammine liberated per mole of solid com­ plex are 115,5, 92.0, 64.5 and 74,5 for GH3NH0 , NH3 , en and tn, respectively. Comparison of these values v/ith A 8° for the dissociation of the corresponding complex shows that for the GH3NH2 and NII3 complexes the major contribution to As® is, indeed, due to the entropy of the liberated ammine, dith the other two complexes, the entropy of the liberated amines are offset by a large increase in the entropies of the solid substances involved in the process. The order of increasing entropy of the solid complexes, obtainable from Table XIV, is CÎI3NH2 larger entropies of the latter two complexes are a result of entropy increase 61 v/hich accompanied the breaking of one of the amine bonds before the dissociation process occurs. The relative entropy of bonding (analogous to the enthalpy of bonding) can be obtained from Tables X and XI ( A S°qi) if the lattice entropies are assumed to be equal. The lattice entropies should be approximately equal only when comparing the NH3 complex to the GH^ITHg complex and the en complex to the tn complex. A i s In the latter comparison relatively small and may be compensated by the difference in lattice entropies of the solid complexes. However, as a first approximation, we can assume that -|-As°;p is the difference in the entropy of bonding of NH3 and GH3NH2 ligands in the complex and A S°q the corresponding difference for the en and tn ligands. Hie relative free energy of bonding, applying the same assumptions, can also be obtained from Tab lex X and XI. As a first approximation, jAF°rp can be taken as the difference in the free energy of bonding for the MH3 and CH3NH2 ligands while A F®q is the corresponding difference for the en and tn ligands. Note that the difference in the free energy of bonding indicates that at 100®G the ITN3 complex is more stable than the CH3NH2 complex while at 25®C the reverse is 62 true. Prom Table XIV the order of increasing standard free energy of formation for the complexes is given as NII3 < GÎÎ3IIH2 < e n < tn at both 25®C and 100®G. It must be emphasized that in these discussions con­ cerning the en and tn complexes, the amine ligands are bound by only one amine group to the platinum ion. At 25®G this model is hypothetical and the discussions have no relation to the compounds as they actually exist at this temperature. However, at, or near, 100®C these complex forms and the pro­ cesses concerning them are real. It is for this reason that all data have been reported at both 25® and 100®G, Since the various enthalpy and entropy values have little or no temperature dependence all discussions concerning them are valid at both temperatures under the expressed assumptions. Under the as sumptions of the proposed model for the en and tn complexes, it is obvious that comparisons con­ cerning the chelate effect and the effect of ring size in the chelate effect cannot be made. SUTiMARY Dicliloro(2^ 2J,bipyridine )platlnun( II), ^Pt (bipy , suspended in cbloroform, was treated with various anhydrous ammines and a series of complexes were formed which are represented by the general formula [pt(bipy )(Am) J Clg where Am is ammonia, methylamine, ethylenediamine and 1^ 3-diaminopropane and n is 2 for the former two ammines and 1 for the latter two. These complexes, upon heating, dissociate into the original components. The pressure of the dissociated ammine vapor over the solid complex was measured as a func­ tion of temperature and. the thermodynamic quantities A f ® , All®, and As® for the dissociation were obtained for each complex. Comparison of these thermodynamic values shows that the four complexes fall into two groups with the complexes containing the monoammine ligands in one group and those of the diamine ligands in the other group. characterized by large values of A H ® and latter group by relatively small values. The first group is A S ® and the By a semi- quantitative consideration of lattice energies and entropies it is shov/n that the large differences in the A H ® and values for the tv/o groups cannot arise from tnis source A 3 64 unless the compounds in these groups differ in some way not usually considered* A model is proposed whereby, during the process of heating the complexes, one of the platinumamine bonds dissociates in the complexes containin-'^ the diamine ligands and a chloride ion in the complex becomes coordinated. The thermodynamics for this process are not measured or detected by the experimental methods used in the study and the net result is that the thermodynamic values which arc obtained for these two complexes refer to the dissociation of the hypothetical (at 25^) complexes. Difference in lat­ tice energies and entropies between the two groups, under the assumptions of this model, can account for part of the differences in A H ° and A but a further comparison cannot be made. Minor differences in A E° and A within each of the groups are the result of differences in lattice energies and entropies and in the enthalpies and entropies of bonding of the various ligands to the platinum ion. At 25^0 the order of decreasing stability for these complexes is given as CHsî'Iiîo, AH^ and en and tn. This sæiie order is given for AS^, From the standard free energies and enthalpies of formation of the free aimnines and their standard entropies and from the thermodynamic values obtained for the dissocia­ tion of the complexes, relative values for the standard free energies and enthalpies of formation (from the elements) and the standard entropies of the complexes were obtained. The order of decreasing for these as IIIÎ5 , Iviei'Ho, en, tn and for A en and tn. and complexes is given as CH-Î'ïïig, These orders refer to the latter two complexes in a hypothetical state. LITERATURE CITED 1, Sidgwlck, N. V., J. Chem. Soc., 433 (1941). 2, ihfholm, R. S., Revs. Pure and Appl. Chem. 4^15 (1954). 3, Martell, A. E. , Calvin, M., -’Chemistry of the Metal Chelate Compounds”, Prentice-Hall, Inc., New York , 1952. 4, G-lasstone, S., "Thermodynamics for Chemists", D. Von Nostrand Co., Inc., Nev/ Yorkjl947. 5, Busch, D. H, , J. Chem. Education, ^3 ^376 (1956), 6 , Schwarzenbach, G . , Helv. Ghim, Acta., ^ (Australia), ^2344 (1952), 7, Spike, G, G . , Parry, R. W . , J. Am, Chem. Soc,, 75 2726 (1952), ibid.. 3770. ' Soike, 0. G , , Ph.D. Dissertation, University of Michigan 1952, 8, Parry, R. , "The Chemistry of the Coordination Compounds", .Ed. by Bailar, J. C ., J r ,, Re inhold Publishing Corp., New YorkJ 1956, 9, Irving, H , , Williams, R. J. P., Ferrett, D. J,, Williams, A. E. , J. Chem, Soc., 3494 (1954), 10, Cotton, F. A,, Harris, F. E., J. Phys. Chem,, 59,1203 (1955), 11, Adamson, A. VV. , J. Am. Chem. S o c %6 ^1578 (1954), 12, Schwarzenbach, G ., "Proceedings of the International Conference on Coordination Compounds", Koninklyke Nederlandse Ghemische Vereniging, Amsterdam, 1955, p. 169. 13, Mellor, D. P., Chem. Revs.^33^137 (1943). 14, Harris, C. M. , Nyholm R. S., Stephensen, N. C., Nature^ 177,1127 (1956), 67 15. Sldgwick, N. V., "The Chemical Elements and Their Com­ pounds", Oxford University Press, London, 1950, Vol. II. 16. Mellor, J. W ., "A Comprehensive Treatise on Inorganic and Theoretical Chemistry", Longmans Green and Co., New York, 1937, Vol. XVI. 17. Pascal, P. "Traite de Chimie Minérale", Masson et Cle, Paris, 1932, Vol. XI. 18. Q,uagliano, J. V., Schubert, L . , Chem. Revs., 50 ,201 (1952). 19. Sidgwick, N. V., l o c . cit. , p. 1595. 20. Keller, R. N., Inorganic Syntheses, ^^251 (1946), 21. Nikolaev, A. V., Gompt. rend, acad. sci. U.R.S.S., 2 0 , 571 (1938); C. A., ^ , 6 7 4 4 (1938). 22. Grinberg, A. A., Ann. inst. platine, £,73 (1932); C, A., 27,1290 (1932); Chem. Zentr., 104 11,198 (1933). 23. Chernyaev, I. I., Sokolov, V. A., Schmidt, N. S., Muraveiskaya, G. S., Doklady Akad. Naul-c S.S.S.R., 62; 235 (1948); G. A., 43,29 (1948); Chem. Zentr., 102 II, 1168 (1949). 24. Chernyaev, I. I., Palkin, V. A., Sokolov, V, A., Izvest. Akad, Nauk S.S.S.R., Oldel Khim Nauk, 215 (1953); C, A., 48,3782 (1953); Izvest Sektora Platiny i Drug. Blagorod. Métal., Inst. Obshchei 1 Neorg. Khim. Akad, Nauk 3.S.S.R., 28,142 (1954); C. A., £9,2790 (1954). 25. Chernyaev, I. I., Palkin, V. A., Izvest. Sektora Platiny i Drug. Blagorog. Métal., Inst. Obshchei i Neorg. Khim, Altad. Nauk S.S.S.R., 30,92 (1955); C. A., 50,9848 (1955); Zhur. Neorg, Khim., 890 (1956); C. A,, 3266 (1956). 26. Rosenblatt, L., Schleede, A., Ann., 505*51 (1933). 27. Morgan, G. L., Burstall, P. H., J. Chem. Soc., 965 (1934). 28. Brandt, W. W., Dv;yer, F. P., Gyarfas, E . , 54,959 (1954). Chem. Revs., 68 29* Bur stall, P* H . , Nyholm, R, S., J. Chem, Soc,, 953 (1950), 30. Rollinson, C, L , , Bailar, J* G,, Inorganic Syntheses. 2)197 (1946). 31. Keller, 32. Gih*dengerhel, K. I., Shagesultanova, G. A., Zhur. Khim., 26^ 222 (1953); G. A., 47^11061 (1954). 33. Mikhelis, Y. L., ibid., 221; C. A., ibid. 34. Khyucknikov, N. G., Savel»eva, R. N . , Zhur. Neorg. Khim., 1,2764 (1956); C. A., 50,10288 (1957), 35. Phipps, Ï. E.,Spealman, M. L . , Cook, Education, £2,321 (1935). 36. Poster, L. M ., J. Am. Chem. Soc., 72,1902 (1950). 37. Dowell, K. P., Electrical Manufacturing, August 1948. 38. Weissberger, A., "Physical Methods of Organic Chemistry, I", Interscience Publishers, Inc., New York, 2nd Ed., 1949, p. 153. 39. Yoden, W. J., "Statistical Methods for Chemists", John Wiley and Sons, Inc., New York, 1951, Chapter V. Mood, A, M . , "Introduction to the Theory of Statistics", McGraw-Hill Book Co., Inc., New York, 1950, Chapter XIII. 40. Busch, D. H . , J. Chem. Education, £ 3 y376 (1956), 41. Kapustinskii, A. P., Acta Physicochim. U.R.S.S., 18, 370 (1943); C. A. 38^5705 (1943). 42. Pauling, L., "The Nature of the Chemical Bond", Cornell University Press, Ithaca, N. Y . , 1940, p. 350. 43. Altshuller, A. P., J. Chem. Phys., 26,404 (1957). 44. Rossini, P. D . , -Wagner, D. D . , Evans, W. H . , Levine, S., Jaffe, I., "Selected Values of Chemical Thermodynamic Properties", N. B. S. Circular 500, 1952. Kelley, K. K , , "Contributions to the Data on Theoretical Metallurgy", U. S. Bureau of Mines Bull. 476, 1949. R. N . ,ibid.. 247, Preklad T. G . ,J. Chem. 69 45, Klotz, I. M . , "Chemical Thermodynamics", Prentice Hall, Inc., Nev; York, 1950. 46. "Handbook of Chemistry", S d , by Lange, N, A., Handbook Publishers, Inc., Sandusky, Ohio, 6th Bd., 1946, p. 1581-5. APPENDIX 70 TABLE XV OBSERVED VAPOR PRESSURES OF DIAIvlMINE (2yBiBIPYRIDIWE)PLATINUM(II) CHLORIDE Aï VARIOUS TEMPERATURES t°c 10^ T P(mm) corrected P (ave.) 87.4 t .1 94.3 Î .1 99.1 t .1 104.4 1 .1 109.4 ' 2.773 2.721 2.686 2.648 2.614 8.36 12.30 15.94 24.51 35.27 8.49 12.65 16.37 24.93 34.94 8.37 12.69 16.49 24.59 34.99 8.72 12.66 16.12 24.38 35.04 0.49 12.37 16.23 24.48 35.15 8.48 12.73 16.06 24.90 35.23 8.72 12.48 16.32 24.61 34.94 8.46 12.85 16.20 24.76 35.10 8.52 12.70 16.44 24.53 35.01 12.87 16.24 8.51 t .17 Applied Correc­ tion -1.11 Î .13 35.08 12.63 - .19 16.24 - .17 24.63 ± .19 35.08 -1.11 - .13 -1.11 - .] 3 -1.11 Î .13 -1.11 FIGÜHE III 71 MM.H6 30 20 10 2.60 2.64 LOG P VKRSUS T FOR 2.68 2.72 H’ c ( b t py ) ( NH3 ) ,1 Cl , 2.76 ,3 72 TABLE XVI OBSERVED VAPOR FRESSURES OF DIMETHYLAMINE(2^21BIPYRIDINE) PLATINUI,UII) CHLORIDE AT VARIOUS TEMPERATURES t^C 10^ T P (mm) Corrected 78*7 t ,3 85.6 * .3 89.0 - .3 2.842 2.814 2.787 2.761 10.33 16.26 22.90 29.45 10.14 16.26 23.01 29.51 10.40 16.18 23.05 29.90 9.87 16.58 23.24 29.93 10.26 16.11 23.28 29.68 9.98 16.12 23.20 29.57 9.94 16.30 23.20 29.83 10.16 16.17 23.19 29.70 10.03 16.24 23.17 29.80 10.12 16.54 23.30 29.74 P(ave.) 10.12 t .17 Applied Correc­ tion 82.2 - .3 1,91 - .19 16.27 t .16 23.16 t .13 29.71 t .16 1.91 - .19 1.91 - .19 1.91 : .19 F IG tia IV 73 MM.HG 30 20 2.76 2.78 I.OG P V E R S U S T FOR 2.80 2.82 jpt < bi py ) ( C H 3 N H 2 ) 3! G !2 2.84 3 iû 74 TABLE XVII OBSERVED VAPOR PRESSURES OF ETHYLENEDIAMINE (2.21BIPYRIDIÎTE)PLATIHU>I(II) CHLORIDE AT VARIOUS TEÎ/IPERATURES toe 114.2^.3 10' T 2.581 2.562 2.543 2.516 2.487 2.465 4.37 4.82 5.23 5.86 6.45 6.89 4.31 4.87 5.06 5.93 6.45 6.79 4.82 4.85 5.09 5.89 6.87 6.85 4.51 4.91 5.15 6.22 5.94 6.70 4.05 4.62 5.23 5.76 6.69 6.62 4.68 4.87 4.91 5.97 6.16 6.79 4.16 4.59 5.12 6.00 6.70 6.52 4.61 4.81 5.01 5.70 6.05 6 .49 4.02 4.63 5.15 5.69 6.66 6.75 P(mm) Corrected 117.it.3 120.Ot.3 124.2^.3 129.ot.3 132.5^.3 6.32 4.30 6.34 6.51 P (ave.) 4.39^.27 4.75-.13 5.11^,10 5.89-.17 Applied Correc­ tion 1.91Î.19 1.91Î.19 1.91-.19 1.911.19 6.441.29 2.10±.13 6.71-.14 2.10l.l3 75 lO 2 |i 00 ID cJ CD lO (\i r \i lO cvi L CM in CM c!> o Ip CM 00 CM CD CM O ID an (X P. 76 TABLE XVIII OBSERVED VAPOR PRESSURES OP 1^3-DIAMIN0PR0PAKÏÏ (2,21BIPYRIDINE)PLATINWi(II) CHLORIDE AT VARIOUS TEMPERATURES t^c 89.it.2 97.0i.2 10^ T 2.760 2.701 2.663 2.624 2.589 9.45 11.17 11.73 13.20 13.82 14.31 9.12 11.38 11.99 12.75 13.96 13.92 9.37 11.11 11.80 13.14 13.43 13.64 9.26 11.05 12.20 12.76 13.72 14.32 9.09 11.49 12.69 13.03 13.38 14.28 9.23 11.21 12.34 13.00 13.96 9.21 11.40 12.00 12.68 14.02 11.13 12.13 12.90 14.15 11.11 12.64 13.20 13.72 10.87 11.68 12.86 14.07 11.53 12.40 P(imn) Correct ed 102.3t.2 107.9t. 1 113.Ot. 2 122.2t 2.529 12.13 P(ave.) 9.25t.l3 11.22t.20 Lpplled îorrection 1.39t.l6 1 .39t.16 12.14^. 25 1 .39t.16 12.95Î .19 1 .39t. 16 13. 66 t. 25 J. 1.91-.19 1- + 1.91Î.: 77 K) M i l l ?; CM CM M CM GO (O CM CL W .C CM N O (P CM (-0 .'3 LT) CL. L? CD lO cJ Si CM e> X 3 2 CO CJ 00 78 X-RAY PATTERNS In an effort to be more specific concerning lattice energies. X-ray powder patterns were obtained for the fouir ajiimino complexes and the parent compound |pt (bipy )C1^ . The d values for the more intense lines of each pattern are listed in Table XIX. intensity are omitted. The values for the lines of weaker For each compound the d values were plotted on a log scale versus relative intensity. The re­ sulting graph was then compared with Hull-Davey charts for the tetragonal and hexagonal structures and a similar type chart for the cubic structure. The d values for the NH^ complex, as listed in Table XIX, were found to compare with those for a tetragonal structure with & = .90, a - 21.7^ and c = 19.5^. Similarly, those for the en complex compared with the tetragonal structure with c 3 26,4^. - 1.10, a = 24.oS and No comparisons could be found for the d values of the remaining three compounds. Since the structures for all four complexes were not obtained no attempt is made to discuss these data in the lattice energy considerations. Also structures obtained by this method are for the real complexes at 25°C and are not valid for any hypothetical forms which may be assumed. 79 TABLE XIX X-RAY POWDER PATTERN DATA ^ t ( b l p y ) (NHg)g]cig (pt(blpy)(C%NHg)2]ci2 d d [pt{bipy )fcn]ci; d I 9.47 VS 9.92 S 10.59 VS 8.24 VS 8.74 S 8.28 VS 6.51 s 8.27 VS 6.06 vs 6.23 vs 6.04 vs 5 .36 s 5.14 s 4.17 VS 5.02 s 4.88 s 3.78 S 4.37 s 3.76 s 3.55 s 4.21 s 3.56 s 3.43 s 3.55 vs 3.44 s 3.32 s 3.36 s 3.16 vs 3.12 s 3.19 s 3.00 s 3.00 s 3.12 s 3.01 s 2.92 s Pt(blpy)Cl2] I®- (pt (bipy }eri] Cl2 d pa 10.63 VS 10.41 VS 8.12 VS 8.05 VS 7.17 S 6.46 VS 6.68 S 5.05 VS 5.04 VS 4.81 VS 5.14 S 3.54 S 3.57 S 3.41 S 3.18 S 3.30 S 2.46 S a Relative Intensity; VS Very strong; S Wealcer intensity lines have been omitted. Strong 80 NOTE ADDED IN PROOF In the model proposed to explain the anointilous nature of the data obtained in this investigation the assumption was made that 0ne of the platinum amine bonds is broken prior to the dissociation of the diamine ligand (page 53 f f .) An attempt has been made to verify this assumption experi­ mentally by differential thermal analysis (D.T.A.), With D.T.A. a sample is heated at a fairly rapid rate (several degrees per minute) and the temperature of the sample is compared to that of a reference material placed near the sample in the furnace. The difference in the rate of heat­ ing of the sample and the reference material is approximately constant until a phase change or decomposition occurs in the sample. (The reference material is chosen so that this does not occur within it in the temperature range under investi­ gation. ) At this point the rate of change in the temperature of the sample rapidly increases or decreases depending upon whether the accompanying enthalpy change is exothermic or endothermie. The difference in temperature of the sample and reference material is recorded as a function of time. In this investigation, a sample of the tn complex was heated with aluminum oxide as the reference material.'*' ''Experimental D.T.A. by (loro Uehara of the Soil Science Department. The 81 resulting D.T.A. curve indicates two endothermie enthalpy changes occur, one beginning at 80-90^0 and the other at 130-140^0, The latter change is attributed to the dissocia­ tion of the diamine ligands from the complex and the former to the breaking of the first platinum amine bond, in support of the assumption of the proposed model.