a II.-O?2 the Action of Bydrogelz on Xilver N,ftrafe. By W. J. RUSSELL, Ph.D. F.R.S. ITis generally acknowledged that hydrogen to a small extent decom- poses silver nitrate and separates from it metallic silver but this decomposition does not appear to have been minutely studied. I noticed some time ago that aftcr a current of hydrogen had passed for a considerable length of time through a wash-bottle containing silver nitrate a very appreciable amount of deposit was produced ; also that this deposit had a metallic appearance and when examined under the microscope appeared entirely crystalline in structure. Other experiments were made on the subject and it was found that when hydrogen was pcssed through a dilute solution of silver nitrate even for a very long time an exceedingly small amount of precipitate was formed and this of a greyish colour and only by the aid of a micro- scope could any tiace of crystalline struct *re be seen ; on the other hand if a saturated solution of the silver nitrate was taken then com-paratively a considerable amount of precipitate was formed that coming down first being somewhat dull and grey whereas that which precipi- tated afterwards was beautifully bright and crystalline.30 pms 10 grams and 2 grams of silver nitrate were dissolved each in 120 C.C. of water and hydrogen passed through the solutions for about a fortnight. There was then found to be a precipitate which weighed 0.365 in the 30 gram solution one weighing 0.1229 in the 10 gram solution andan amount too small to weigh in the 2 gram solution.The precipitate formed in this kind of way has been examined both by igniting and re-weighing it and by converting it into chloride and weighing it as such ; by both methods it gave results showing it to be pure silver. In the above experiments and in some of the following ones the hydrogen was generated by the action of a solution of copper sulphate on granulated or powdered zinc or of a solution of copper sulphate and zinc sulphate on zinc. In this way a very regular current of hydrogen is produced and one tlhat will coktinue for many weeks and at most the addition of a few crystals of the copper sulphate is all the attention it will require. Another mixture which acts very well indeed and gives a remarkably uniform current for a very great length of time is that of zinc and tin both in the state of powder simply put into a small flask of water at drdinary temperature 10 grams of zinc and 5 grams of tin give off' with ciirious regularity 3 C.C.to 4 C.C. of hydro- gen every day. In one case this was continued for 44days and then only stopped on account of the hydrogen not being required and in a more recent experiment tlie action has gone on for six months and the n2 ON THE ACTION OF HYDROGEN ON SILVER NITRATE. evolution of gas appears to be still as regular as when it was first set up. In most of the following experiments when the hydrogen was required only for comparatively short times it was generated from zinc and sul-phuric acid in a Kipp’s apparatus the gas being always very carefully purified and very thoroughly washed with silver nitrate before using it.First in order to make sure that this precipitation was produced by the hydrogen and not by any impurity it might contain the following experiments were made. The hydrogen was washed and then passed through and over an excessive amount of silver nitrate solution ;it first passed through two large Wolffs bottles filled with pumice-stone wetted with silver nitrate and then through some 12 feet of tubing filled with glass also wetted with the silver solution. The hydrogen after this treatment still precipitated the silver as freely as when purified in the ordinary way; further when a series of separate solutions of silver nitrate had hydrogen passed through them the pre- cipitations took place with apparently the same facility in the last solution as in the first.Other experiments were made with zinc which was free entirely from arsenic and lastly hydrogen was obtained from reactions in which zinc took no part in all cases the same preci- pitation of metallic silver took place. To obtain the hydrogen without using zinc magnesium and platinum both in the form of powder were put into a flask of water; at first a very violent action ensues but afterwards a tolerably regular stream of hydrogen comes off. It appears then that hydrogen is capable of precipitating silver from a solution of the nitrate the precipitation occurring much more readily in a saturated than in a dilute solution.Even from a saturated solution the precipitation takes place only after some time and the amount pre- cipitated as compared with that in solution is always very small. When a saturated solution is used hydrogen must be bubbled through it for nearly half an hour before precipitation takes place; if the current be continued a dull greyish substance separates out at first afterwards the precipitate is perfectly crystalline and bright exb ibiting a beaut,iful appearance. If for instance 10 grams of silver nitrate were taken dissolved in about 8 C.C. of water and hydrogen bubbled through the liquid for 18 honr.s only about 0.1308gram of silver would be preci- pitated. At first the amount of silver precipitated is nearly proportional to the time to which it has been exposed to the action of the hydrogen but after the first 48 hours the amount of silver precipitated increases at a slower and slower rate and after a time the amount of silver present will often begin to diminish.This precipitation of silver by hydrogen is independent of the action of light. Two similar solutions both contain- ing 20 grams of silver nitrate were acted on by hydrogen the one in bright light the other in the dark all other circumstances being made as similar as possible; from the solution in the light 0.2630 gram of ON THE ACTION OF HYDROGEX ON SILVER KITRATE. silver were precipitated and from the one in the dark 0.2703 gram. The precipitation is however very much expedited by heat. On passing hydrogen through the silver nitrate solution only for a minate and then heating ih it very visible although not very heavy precipi- tat,ion immediately appears ;it is amorphons and of a brownish colour.With a cold solution the following happens hydrogen may be passed through it for about 25 minutes witlho.;lt producing the slightest alte- ration in appearance. If the solution be then exposed to the air SO that the gas above the liquid diffuses away silver begins to precipitate after a short time and a considerable amount will in this way sepa- rate out. In one experiment 10 grams of silver nitmte were dissolved in 9 C.C. of water and the hydrogen was passed through the solution for 25 minutes. It was then perfectly clear ; it was allowed to stand all night and there was then found to be 0.01 gram of silver precipf- tated ; this weight of silver corresponds to 1.037 C.C.of hydrogen and if hydrogen precipitates silver equivalent for equivalent from the nitrate this bulk of the gas must have been dissolved in the 9 C.C. of liquid when the passage of the hydrogen through it ceased ; this same bulk of water would dissolve only 0.174 C.C. of hydrogen that is ith of the total amount of gas ; the other +ths must therefore have been dis- solved in the liquid owing to the presence of the silver nitrate or some chemical change must take place previous to the precipitation of the silver. A neutral solution of silver nitrate becomes acid as soon as the silver separates out ; and it has always been found that the silver is precipitated in larger crystals and brighter in appearance from acid than from neutral solutions.When examined by the microscope the silver is found sometimes to have an arborescent appearance at others to be a close network of very fine filaments so intertwined that tho whole of the metal cm be lifted out of the'solution in a connected mass and sometimes it is seen to bo in distinct and separate triangular plates of brilliant metal. The silver prepared by this means is as might be expected remark- ably pure ;any gold in the solution would be precipitated with the first portions of silver and the other metals likely to be present would not be separated out by the hydrogen. There is another way in which the hydrogen may be made to act upon the silver solution that is by merely keeping an atmosphere of the gas above the solution instead of bubbling it through.If a saturated solution of the silver nitbrake be put into a tolerably capacious bottle and the bottle be then filled with hydrogen the action will com- mence rather sooner than if the gas was merely bubbled through it and in 18 or 24 hours a somewhat larger amount of silver will be pre- cipitated than by the other process. With regard to the nature of the reaction which takes place is it a ON THE ACTION OF HYDROGEN ON SILVER NITRATE simple replacement of silver by hydrogen and if so why should it not be possible by continuing the action of the hydrogen to continue the pre- cipitation of silver ? and further if hydric nitrate easily dissolves silver how is it that hydrogen can separate the nitrate into the acid and metal ? First it was proved that the hydrogen was abaorbed by silver nitrate and no other gas evolved during the reaction.By taking a tube some 7 inches long and about &ths of an inch in diameter filling it with pure hydrogen and allowing it to stand in a saturated solution of the nitrate it was found that the solution gradually rose in the tube and at last completely filled it. The following method was adopted for estimating the relatJion between the amount of hydrogen absorbed and silver precipitated. A piece of tubing was bent into this form i:-’ and the horizontal end drawn out but not to a very fine point and bent upwards ; down it was poured the silver nitrate ; then the vertical end was placed ina vessel of water and hydrogen passed through the tube until the whole of the air was expelled ;the drawn out end was then sealed up.The water soon begins to rise in the tube a.nd when the experi- ment is over by marking off where the water stands and weighing the amount of silver separated out at least a tolerably accurate approxima- tion may be obtained to the amount of gas absorbed and silver preci- pitated. If an experiment of this kind be carried on for about aweek it will be evident that some secondary reaction has set in; for large yellow crystals will separate out from the solution ; these crystals were proved by analysis to be silver nitrite consequently it is not to be expected that the hydrogen absorbed and silver precipitated will be in simple atomic proportion.The following experiments show the relation which was found to exist between them. 20 grams of silver nitrate in solution were put into the bent tube with the hydrogen; after 23 hours it was found that 33.12 C.C. (corrected) of the gas had been absorbed and that 0.3024 gram of silver had been precipitated :-33.12 C.C. of hydrogen correspond with 0.3205 gram of silver Silver found ...................... 0.3024 ?> -0181 A quantity of hydrogen equivalent to this weight of silver is there-fore unaccounted for. In another and similar experiment 15 grams of silver nitrate were used and left in contact with the hydrogen for 27 hours; in this case 31.25 C.C. of the gas were absorbed:- 37.25 c c.hydrogen correspond with. . 0.3024 gram of silver. Silver found ...................... 0.2857 , -0167 R ON THE ACTION OF HYDROGEN ON SILVER NITRATE. again then the weight of silver found was less than the equivalent of the hydrogen absorbed. If the experiment be continued for tt much longer time there is a still greater disappearance of silver as compared with the hydrogen absorbed and as before mentioned after about four days the nitrite will begin to crystallise out. In an experiment which went on for six months the following results were obtained :-182*3 C.C. of hydrogen were absorbed which are equivalent to 1.758 of silver whereas only 0.2011 gram was found in the liquid. In another experiment 136.9 C.C.of hydrogen were absorbed equivalent to 1,3197 gram of silver and only 0.1277 of silver was found. Another experiment on a larger scale gave a similar result. 20 grams of pure silver nitrate were taken and exposed to the action of hydrogen from June 28th till August 11th ; it was then found to have absorbed llOX*21 C.C. (cor-rected) of hydrogen which corresponds to 10.72 grams of silver whereas in the liquid there was only 0.7079 grm. of metallic silver. In reality the amount of hydrogen absorbed in this and foregoing experiments was greater than above stated for in all cases a small amount of nitric oxide is evolved. It appears then that after a time a secondary reaction certainly sets in some of the silver which has been precipitated re-dissolving and nitrite separating out.That by the action of the hydrogen on the nitrate there is formed a solution capable of dissolving silver is shown by putting some weighed pieces of silver foil into the neutral nitrate solution and exposing it to the hydrogen ; it will be found that the pieces of silver will graduallylose weight. In order to stop this secondary action by neutralising the acid as soon as formed and thus be enabled to ascertain whether as regards the hydro- gen it simply replaced the silver in the nitrate the following experi- ment was made:-A bent tube was taken similar to those before clescribed ; a saturated solution containing 5 grams of silver nitrate was introduced into it and along with it 3 *5gram of silrer oxide ; the tube was filled with hydrogen aad the action allowed to go on for 20 days 105.3 C.C.of hydrogen had then been absorbed 1.0089 gram of silver precipitated and 0.3958 of the oxide had dissolved up :-105.3 C.C. of hydrogen correspond with 1.0184 gram of silver. Silver found . . . ,. . . ,. . . . . . . . . . . . 1.0089 ?7 that is 1part of hydrogen ia equivalent to 10'7 parts of silver. Again the weight of hydric nitrate corresponding with the amount of silver found is 0.5885 and that calculated from the weight of oxide whicth had dissolved is 0-5994. This seems satisfactorily to shorn that the action of the hydrogen on the silver nitrate is to replace the-silver thus precipitating the metal and forming hydric nitrate and that in the former experiments some secondary reaction is afterwards set up which ON THE ACTION OF HYDROGEN ON SILVER NITRATE.causes the solution of the silver and the formation of silver nitrittb. The questioii is then what is the nature of the secondary action ? And how is it that any such action is set cp ? If the hydrogen takes the place of the silver forming hydric nitrate and silver how can it be that on simply continuing the action of hydrogen they recombine ? Dilute solutions of the ordinary nitric acid have little or no action on silver; if an excess of silver be added to this acid the solution will always remain very acid even if very finely divided metal be used ; a 1 per cent. solution of HNO has the power of dissolving only the slightest trace of the metal; further if a dilute solution of hydric nitrate be saturated with silver nit4rate and then exposed to the action of hydrogen the silver is precipitated as it would be from a neutral solution.From a solution containing 7 per cent. of HN03and saturated with 10 grams of the silver nitrate hydrogen precipitated in 18 hours 0,1275 gram of silver very nearly as much as would have been precipitated from a neutral solution. Even in a solu-tion containing as much as 17.5 per cent. of HNO, silver is precipitated oy hydrogen but there is this difference between what takes place with acid and with neutral solutions stop the hydrogen passing through the acid solution and in a few hours all the silver will have disap- peared. With the neutral solutions this does not happen.This is shown very well with asolution coritaining 7 per cent. of acid. Satu-rate it with silver nitrate; then on passing hydrogen through it the silver is precipitated; stop the current of gas for the first three minutes there is no action; then a string of bubbles rise from the precipitated silver and in a very short time the whole of it has dis- appeared. On again passing in hydrogen the silver is precipitated as before and the whole experiment may be repeated over and over again thus dissolving up and precipitating in the same liquid. Thc explanation of the dissolving up of the silver in these acid solutions is appnrently the presence of a trace of nitrous acid. It has long been believed that the great activity of nitric acid towards many metals was owing rather to its containing a small amount of nitrous acid than to the affinities of the acid itself; and with regard to silver thin un- doubtedly is the case and that it holds good even with very dilute solutions of nitric acid the following experiments show.Two similar pieces of silver foil both weighing 0.3 gram were placed in similar tubes each with 5 C.C. of dilute nitric acid containing only 1.6 per cent. of HNO3. Through one of these solutions only two small bubbles of nitrous acid were passed ; both liquids were then corked up and left to stand for 46 hour^ and the amount of silver dissolved was estimated in the solution through which the nitrous acid had passed 0.0262 gram was found; in the other one 0.0093 gram. If stronger acid solutions be used this action of the nitrous acid is still ON THE 2iCTION OF HYDROGEN OX SILVER SITRATE.more marked. Two similar pieces of silver foil weighing about 2 grams each were put as in the above experiment into dilute nitric acid but containing 1'7.5 per cent. of HN03,in place of 1%per cent. Two or three bubbles of nitrous acid were passed through one liquid only ;immediately the silver in this liquid became covered with minute bubbles and after twenty minutes the whole of the silver was dissolved whereas the other piece of foil after exactly the same length of time was hardly a.t all attacked for the liquid on treating it with a chlo-ride gave only a slight cloud of silver chloride. Nitric oxide passecl into the solution gives of course similar results.There is another curious reaction with regard to the dissolving up of the silver in nitric acid. Two similar pieces of silver were placed in diluted nitric acid but still so strong as to dissolve silver that containing 17.F;per cent. of HN03,for instance and hydrogen was made to bubble through one of the tubes with silver solution while through the other the hydrogen streamed without bubbting through the liquid. In one experiment it was found that after an hour the silver in the tube through which the hydrogen did not bubble was entirely dissolved and in the other one the metal was not attacked to an appreciabIe extent by the acid €or the liquid gaye only a faint cloud on adding a chloride the bubbling of the hydrogen through the liquid having therefore protected the silver.This protective action is however in no way peculiar to hydrogen ; air will act in the same manner. The gas sweeps away any nitric oxide and keeps the nitric acid pure. The dissolving up of the silver after it has been precipitated by the hydrogen from silver nitrate may then be satisfactorily explained if there be any reason for the formation of only a trace of nitrous acid. A trace of this oxide of nitrogen is suEcient to start the action of the acid on the silver and once started the action very rnpidiy increases in intensity. The reaction which takes place is probably something of this kind. The nitrous acid attacks the silver form- ing silver nitrite nitric oxide and water the nitric oxide thus formed reducing hydric nitrate and forming more nitrous acid.Thus the action gets quicker and quicker and may be. started by the addition of a lower oxide of nitrogen. The point still remaining to be explained is then how the first trace of nitrous acid is formed. Assume that to be formed and with this cumulative action the rest is easy of explanation; and certainly if as appears to be the case pure hydric nitrate has little or no action on silver at all events when dilute it is not obvious why it should be formed. The dissolving up of the silver seems however to be directly con- nected with the presence of hydrogen; for instance if a volume of pure nitric acid be taken so dilute that it does not act appreciably upon silver and if it be exposed to an atmosphere of hydrogen instead of 10 ON THE ACTION OF HYDROGEN ON SILVER NITRATE.air then more or less action will be found to have taken place. If for instance a solution be taken containing l+p. c. of HN03,and pre- cipitated silver or silver foil be placed in it after they have remained in contact for a long time hardly the faintest cloud of silver chloride can be obtained from the liquid whereas if an exactly similar experi- ment be made but the air above the liquid be driven out by hydrogen and the vessel then closed on examining the liquid it will be found to contain sufficient silver to yield a dense precipitate of chloride. T’wo portions of precipitated silver were taken each weighing 0.157 gram and placed in separate bottles with 10 c.c. of dilute nitric acid con- taining 1.6 p. c. HNOs ; one bottle was filled with hydrogen the other left full of air. After two days the silver in each was again weighed ; in the bottle with the hydrogen 0.0039 gram had dissolved and-in the bottle with air 0*0006. Again if still stronger solutions of nitric acid be used this action is still more obvious. Two pieces of roll silver as similar as possible in every way weighing about 1gram each and exposing a surface of about 2 square inches each were placed in two large bottles and 50 C.C. of dilute nitric acid containing 7 p. c. of HNOs added. Pure air had previously been passed for half an hour through the dilute acid. One bottle was now filled with hydrogen the other left full of air.After two days it was seen that nearly all the silver in the hydrogen bottle was dissolved; and on examining the liquid in each bottle it was found that in the one case there was 0.9438 gram of silver in solution and in the other (where only air had been) there was only sufficient silver to give a faint cloud of silver chloride. To gain some idea how long this absorbingof the hydrogen by silver nitrate would continue and what would be the ultimate products 0.3 gram of nitrate was dissolved in water and put into a bent tube with the hydrogen. The experiment was started in December and stopped in the following March when it appeared that no further ab- sorption of gas was taking place. Supposing that no trace of nitric oxide remained with the unabsorbed hydrogen then 40.94 C.C.of hydrogen had been absorbed 0.0593 gram silver found in the liquid and 0.1234 gram of silver nitrite thus leaving 0.0446 of silver nitrate undecomposed. The amount of hydrogen which had disappeared was capable of precipitating 0.3932 gram of silver ; that is just twice as much as there was in the nitrate used. Direct experiment shows that the silver nitrite is not decomposed by hydrogen ; consequently under the above circumstances the silver nitrate is not a stable salt but indirectly by the action of the hydrogen it passes over into nitrite which is stable. The nitrite as is well known is very slightly soluble in water but it is soluble to a very considerable extent in a solution of uitrate hence in these tubes a considerable amount is always found in ON THE ACTION OF HYDROGEN ON SILVER NITRATE.solution besides what has crystallised out. On passing nitrous acid into a solution of silver nitrate if tho solution be dilute the nitrite separates out at once. But if the solution be a saturated one no separa- tion takes place ; the nitrite formed at first remains dissolved up and from the decomposition which the nitrous acid in greater part under- goes sufficient nitric acid soon accumulates in the solution to decom- pose the nitrite and an abundant evolution of nitric oxide takes place. In the foregoing experiments it has been shown that as the silver dissolves up silver nitrite separates out ; this is indeed an ordinary product of the action of nitric acid on silver when such reaction takes place out of contact with air.For instance take ordinary nitric acid add an excess of silver to it and replace the air above it with carbonic acid and in a short time lai=ge crystals of silver nitrite will separate out ; or if nitric acid with an excess of silver be simply placed in tt bottle and then the stopper put in the nitrite will form very readily. The series of reactions then which appear to take place when hydrogen is left in contact with silver nitrate are first the replacement of silver by hydrogen the silver separating out in a crystalline form ; then after a certain time by the aid of the hydrogen present a trace of nitrous acid is probably formed and this is capable of attacking the precipitated silver and combining with it.At the same time nitric oxide is given off which by acting on more nitric acid forms more nitrous acid thus accelerating the solution of the silver. If much nitric acid be present the silver nitrite naturally passes back again to nitrate and this may again be acted upon by the hydrogen. The silver nitrite on the contrary is unacted on by hydrogen and will therefore be the end-product formed by this reaction. With regard to other metals platinum palladium and gold are corn- pletely precipitated from solution by hydrogen in the metallic state and at ordinary temperatures and pressures. A solution of copper nitrate by long standing in contact with hydrogen becomes appuently from the change of colour converted into nitrite. Mercuric nitrate seemed to be acted on and a basic nitrate thrown down while bismuth nitrate seems not to be at all attacked.