摘要:

JOURNAL OF THE CHEMICAL SOCIETY. CONTAINING THE PAPERS READ BEFORE THE SOCIETY AND ABSTRACTS OF CHEMICAL PAPERS PUBLISHED IN OTHER JOURNALS. Ph.D. F.R.S. C. W. HEATON H. E. ARMSTICONG F.C.S. Ph.D. HUGO M~~LLER J. ATTFIELD Ph.D. F.R.S. D.C.L. F.R.S. W. H. PERKIN, E. FRANKLAND F.R.8. Ph.D. F.R.S. J. H. GILBERT,Ph.D. F.R.S. W. J. RUSSELL Ph.D. F.R.S. R. WARINGTON, J. H. GLADSTONE F.C.S. C. E. GROVES, F.C.S. &%tar HENRY .WATTS,B.A. F.R.S. @istractor# Ph.D.. G. T. ATHINSOS. E. W. PR~VOST Ph.D. E.C. BABER. W. RAMSAY ROBINSON. P. P. BEDSON,B.Sc. JOHN A. BELL,M.B. R. ROUTLEDGE, CHICHESTER BSc. D. BENDIX. WALTER SAISE D.Sc. F. D. BROWN. C. SCHORLEMMER, F.R.S. D.Se. L. T. O'SHEA. C. A. BURGHARDT D.Sc. WATSON T CARNELLY SMITH. FRANK TAYLOR. CLOWES,D.Sc. JAMES THOMSQN. A. J. COWNLEY. J. MILLAR ROBERTWARINGTON. 0. F. CROSS. C. E. GROVES. c. w. WATTS. WATTS,D.Sc. F. J. LLOYD. JOHN MUIR. W. C. WILLIAMS. M. M. PATTISON E. NEISON. R. C. WOODCOCK. B.A. BSc. C. R. A. WRIGHT,D.Sc. J. H. POYNTTXG 1877. Vol. 11. LONDON J. VAN VOORST I PATERNOSTER BOW. 1877. LONDON 1IlRfllSOS AND SONS PRINTERS IN OllUINABP TO HER MAJESTT ST. MARTIN’S LANE.

ISSN:0368-1769    

DOI:10.1039/JS87732FP001

出版商:RSC    

年代:1877

数据来源: RSC

摘要:

2s TI-IOXXS ON SOME POINTS IN GAS ANALYSIS. By J. W. THOMAS. HAVIXG expericnccci considerahlo difficulty in detcrmining the naturc of a gaseous mixture which I obtained during the examination of the gnses occluded in Bovey lignite owing to the results bcing discorclard I uiidertook some experiments in order to determine the action OP t!ie reagents used for absorption in psanalysis more especially the action of different gases upon pyrogallic acid in the presence of (KHO) pot-ash and potassic carbonate (COKO?). In the instance of the gaseous mixture referred to there was a€tw the removal of carbonic acid (CO,) an absorption by pyrogallic aci(l eqnal to from 1.3 to 3.8 per cent. in differcnt experiments on the same mixture. This discrepancy together with an almost absolute certainty that no oxygen wais present led me to make the following experi- ments the results of which tended to prove as vas afterwards con-firmed that the absorption by pyrogallic acid and RHOwas not cluc trj oxygen but nitric oxide (N20,).Some gases which I have bem examining recently consisted almost entirely of CO, and in orclei-to THOMAS ON SOME POIXTS IN GAS ANALYSIS. 29 arrive at the nature of the remainder some 20 30 and even 50 C.C. of gas was used. This entailed the use of a rather large quantity of KHo and being unable to find any decisive record whether pyrogallic acid did or did not liberate CC) from alkaline carbonates under varj- ing conditions I began by trying the effect of pyrogsllic acid upon a concentrated solution of COKoZ.About 1 C.C. of COKoz sol. was passed up by the usual pipette int,o a laboratory tube of a gas appara-tus somewhat similar to McLeod’s and about -3 C.C. of a concentrated solution of pyrogallic acid added. After the lapse of tm7o hours no gas was liberated and apparently no action whatever had taken place. About 1C.C. more pyrogallic (in future abbreviated to “ pyro.”) acid was introduced and left for four hours ; no action was perceived and no gas was liberated. As sodic hydrate is sonietimes employed for absorbing CO in gas analysis (I however prefer KHo and always use it) I passed up 1C.C. of a saturated solution into the laboratory (lab.) tube and placed it under like conditions to the COKo2 with pyro. and obtained a like result.The proportions of alkaline carbo- nate and pyro. were varied but in no case was any CO or any other gas liberated. In “Watts’s Dictionary ” (vol. iii page 759) it is stated that pyro. liberates COz from the alkaline carbonates and also that Rosing states that it doesnot liberate C02 from any carbonates. My experiments confirm those of RJosing. The action of 0 upona solution of COKo and pyro. was next tried and to this end 7.3 C.C. of pure 0 was measured in thr eudiometer and tmnsferred to the lab. tube into which the mixture of COKop aid pyro. was previously introduced. Absorption of the 0 took place but only at a slow rate. After the expiration of an hour only a small bubble remained and this very shortly disappeared almost entirely leaving a far from measurable quantity.After carrying out a num-ber of experiments in this direction I found that when the volume of COKo2 sol. was in large excess over the pyro. the absorption of 0 took place more rapidly and with very good results. It was not however sufficiently rapid for use to replace the hycirate even when practicable. An orange-brown precipitate is formed by the action of 0 on pyro. in the presence of COKo or CoNRo,. The per cent. of 0 in air was determined by absorbing with pyro. and COKo, the quantity of the acid being in excess. After one hour 20.1 per ceut. of 0 had been absorbed and although the absorption was continued €or several hours afterwards the per cent. of 0 did not exceed thnt given but after four hours appeared slightly less.-4nother experi- ment was tried with air using again an excess of pyro. over the COKo,. The absorption was slow and there was evidently a quantity of gas evolved by the pyro. as the per cent. of 0 after 18 h0~1-s’ ab-sorption reached only 19.05; there was no free 0 left. Two other THOMAS ON' SOME POIh'TS IN GAS ANRLSSIS. analyses were made with air and the same reagents but twice as much COKo as pyro. was used. The 0 was absorbed more rapidly than in the previous experiments and the results were very concor- dant vie. 20.97 and 20.98 per cent. The air was left in contact with the absorbents all night. In order to ascertain the effect of using an excess of pyrogallic acid to absorb 0 in a gaseous mixture from which a large volume of COahad been previously removed by KHo some COKoz was intro-duced into the lab.tube then a small quantity of RHO and aftcr- wards a solution of pyro. equal to rather more then the vol. of COKo? and KHo. A measured quantity of air was used for the experiment and the absorption of 0 continued for one-quarter of an hour a slow current of Hg dropping into the lab. tube during the time The per cent. of 0 absorbed was 19.63. The remaining gas was retransferred to the lab. tube and subjected to the action of the absorbents for ten minutes and then measured and the per cent. of 0 absorbed was found to be 20.5. In a second experiment with air about 4 parts of COKo, 2 parts RHO,and 2 parts pyro. were used for the absorption of 0;after 15 minutes the per cent.of 6) absorbed was found to be 26-53; after another 10 minutes' absorption 20.7 per cent. but did not exceed this- by prolonged contact. In a third experiment with air about 3 parts COKo2 2 parts KHo and 2 parts pyro. wcre em- ployed to absorb the 0 ; after five minutes the per cent. absorbcd was 19.71; after five minutes more 20.6 per cent. ; and after another five minutes 20.72 per cent. which was not exceeded. In a fourth expe1-i- ment with air 2 parts COKoz 2 of KHo and I of pyro. were used and the absorption continued for ten minutes when 20.9 per cent. of 0 was found to be absorbed. The above experiments were repeated and the results of a great number of trials showed that it is pwsible $0 obtain good results ia the presence of COKo, when an excess of KHo over the pyro.used is present. To test the above results more completely pure 0 was used in quantity about 10 c.c. and the results obtained confirmed those enumerated above. Having some years ago observed that the absorption of 0 took place more rapidly and more absolutely too when the quantity of KEQwas in excess of that of the pyrogallic acid used I repeated the experiments under varying conditions. C a1v er t B o us sin g au1t and Cloez (Jnlwesb. 1863 p. 389) found that althongh the absorp-tion of 0 by KHo and pyro. was complete there was always formed simultaneously more or less CO; when pure 0 was used the GO amounted to from 3 to 4 per cent. of its volume ; and when air was used about 2.5 parts of its 0 was replaced by CO.1found as the following experiments show that CO is generated under nearly all circumstances but the quantity formed is dependent TIIOMAS ON SOME POINTS IS GAS ANALYSIS. upon the proport'ions of the absorbents used and the rapidity of the ab- sorption. When 25 C.C. of oxygen was taken for the absorption and twice as much pyro. as KHo employed the per cent. of CO and traces of nitrogen which appears to be held in solution in the pyro- gallic acid amounted to 3.16 per cent. In the foregoing experiment the absorption proceeded slowly. When the lab. tube mas agitated and a current of Hg allowed to drop through the gas the absorption was naturally accelerated and the per cent. of CO left was rcduced to 2.73. When the quantity of KHo was equal to that of the pyro.used and the absorption was left at rest the residual gas measured 2.83 per cent. and in another experiment by agitating the lab. tube &c. as before it was reduced to 2.06 per cent. In another experiment with 0 twice as much KHo as pyro. was used and the Hg allowed to drop through without agitating the lab. tube the residual gas (original vol. of 0 = 15 c.c.) was found to be 1.26 per cent. When a small quantity of 0 (5 e.c.) was taken and the absorbents in the proportion of 2 parts KHo and 1part pyro. and the lab. tube agitated the resi- dual gas was too small to be transferred to the eudiometer for measure-ment. I consequently employed an absorption tube having a long capillary neck. Using as I did liquid absorbents which could not be removed from the contracted end of the absorption tube the percent- age volumes in the numerous experiments which I carried out are more comparative than absolute.I found however that it is possible to reduce the error caused by the generation of GO to less than 0-5per cent. by using at leasf twice as much RHO as pyrogallic acid the absorption tube being continually agitated during the absorption. In all instances whether the volume of KHo be greater or less than that of the pyro. the quantity of CO is always less when the absorption is accelerated by agitating the absorption tube. Again when large volumes of 0 are taken the absorption is not so complete and this is always the case no matter how much of the absorbents are usecl.This is also the case when only just enough pyro. to absorb the 0 present is introduced into the lab. tube. The result of the experi- ments showed that the absorption of 0 by pyro. and KHo takes place most rapidly when there is a large excess of the latter reagent present. The results of a further series of experiments carried out for the purpose of replacing KHo or NaHo by some other compound to assist the absorption of 0 by pyro. in special cases were not worth recording. Mention may be made however that KCy and pyro. and KCy satu-rated with CuHo2and pyro. absorb 0 slowly and give good results. The residual gas in two experiments measured 1-36and 0.98 per cent BaHozdoes not materially assist in the absorption of 0 by pyro or CaHoz unless very concentrated.THOMAS ON SOXE POINTS IN GAS ANALYSIS. Eaving mistaken N,02for 0 in a gaseous mixture containing but a very small percentage OF that gas (N20,) some experiments were made with KHo and pyro. in order to determine the nature of the re- action which takes place when N20 is present. I had previously noticed that N,O was absorbed by those reagents. I am not aware that the fact that N202is absorbed by an alkaline solution of pyro. is widely known and if it be known at all it is unfortunate that it is not recorded in the books on gas analysis as its presence in small quantity in a gaseous mixture may be the cause of unneces-sary trouble and annoyance to those who have to determine gaseous mixtures of unknown composition. N202 is absorbed most rapidly xvlien the KHo is present in excess of the pyro.used in like manner to 0 but it is slowly absorbed when those reagents are used in any proportion. It is also absorbed by pyro. in the presence of COKo,. In all instances however a portion of gas remains and the percent- age of the residual gas depends upon the time the absorption is allowed to proceed the quantity of the absorbents employed and the relative proportions of the absorbents. Six experimen ts were carried out with N,O, using varying quantities of KHo and pyro. to absorb trhe gas. The N,O was made from pure NOzKo in the manner in which N as N02Hois estimated by Frankland and Armstrong's process." No. of Approximate proportion of Time the xbsorp-Percentage of experiment.re-agents used. tion proceeded. gas absorbed. 1 .... IiKHo ..1pyro (dark) 35 hours 58.20 2 .... 2KHo ..1pyro (fresh) 16 55.92 7? 3 .... 2KHo ..2 pyro , 14 ? 57.90 4 .... 2KHo ..lpyro , 14. 9 55.54 5 ..*. 2KHo ..l&pyro , 15+ , 57-92 6 .... 24KHo.. 1pyro , 40 60.65 99 The gas which remains after the absorption which so far as the N,Oz is concerned is absolute and complete consists entirely of nitrous oxide so that the reaction appears to be simply this- 2N202= N,Os -+ N2O. * Some time ago I communicated the rcsult of my experiments on nitric oxide to Dr. Frankland who suggested that the re-action between the X20 and the alka-lille (caustic) solution of pyrogallic acid was simply tlie reduction of X202to ONz by the abstraction of 0 and tlie consequent loss of one-half its volume thus :-I found by experimci~tsthat no NO3o or NOICo was formed.Dr.Frankland's view of t,Iiereaction is therefere corred. THOXAS ON SOME POINTS IN GAS ANALYSIS. During the determination of the nature of the residual gas T ob-served that when N20 is present in a gaseous mixture containing H with a large excess of 0 much of the N,O yields 0 to the I3 when a spark starts combustion but some of the N,O is always converted into higher nitrogen oxides. N,02 is absorbed by pyro. in the presence of COKo, but the ab-sorptiou is much slower ; when KHo and pyro. was used the ahsorp- %ion was always complete after from 7 to 10 hours; but when COlio is substituted for RHO,the absorption is not complete for 15 hours or more.The gaseous residue is much less when COKo is wed in place of KKo as the following experiments will show. About 2 parts COKo and 1 part pyro. was uwd in both experiments. In Experiment 1the absorption was continued for 36 hours. Small bubbles of gas were congregated around the upper portion of the Hg in the tq9 of the lab. tube where some of the violet compound formed was deposited which seems to indicate that the Hg acted upon it. A11 the N,O was absorbed. The residual gas which measured 14.63 per cent. of the original volume was found to be N,O. In Experi-ment 2 the absorption was continued for 15i hours when 93.87 per cent. of the N,Oz was absorbed. The absorption was therefore incom- plete 6.13 per cent. of the N202being present in the residual gas.The remainder of the gas which was found to be N,O occupied 18 per cent. by volnme of the N20,employed. If N20 is known to be present in a gaseous mixture then no one monld look for 0 but the absorption of Xz02by au alkaline solution of pyi’o. is suffjciently rapid to prevent those reagents being used to determine which of the two N202or 0,is present in a gaseous mix- ture of unknown cornpositlion. When the percentage of N,O present is large the brown colour produced when 0 is added will reveal the presence of what is left after the partial absorption by KHo arid pyro. but if present in small quantity my 3 per cent. no such indication would be given. The gas absorbed by KHo and pyro. would be put down as 0 and the small quantity of N20,left (owing to the absorp- tion being incomplete) would be the means of preventing the opera- tor from obtaining any concordant results-at least this has been my experience.Every mixture containing pyro. which I have exanlined absorhs N,02 and I have not yet succeeded in fiuding an absorbent for 0 or N20zwhich will yemove only one of those gases. The method which I now adopt to detect ihe presence of N,O is the following but it is not applicable when Bunsen’s apparatus are employed. A portion of the gaseous mixture is introduced into the lab. tube of Or. Frankland’s or McCleod’s apparacVus transferred into the eudiometer measured and after the CO is absorbed the gas is passed over into the lab. tube which is removed by placing a ground VOL XXXI1.I) THOMAS ON SOME POINTS IN GAS ANALYSIS. glass plate under it into another Hg trough. A second lab. tnlw is then connected to the apparatus some 0 passed up measured in the eudiometer and let stand therein. Lab. tube No. 2 is removed and replaced by No. 1,containing the original gas less CO,. There would be 0 in the eudiometer and the gas in the lab. tube both known volumes. The 0 is passed over into the gas. mixed by pass-ing mercury over into the lab. tube and the absorption continued for fiveminutes. The gaseous mixture is then measured in the endio- meter and if the last measurement is less than thc sum of the 0 + the original gas (less CO,) then the presence of N,O may be in-ferred. The excess of 0 can now be absorbed by pyro.and the vol. of N,Oz noted and the analysis proceeded with. If it is only required to know which gas is present N,02 or 0 then after the removal of CO the gas (after measurement) is passed over into the la,b. tube which is simply disconnectfed. A quantity of air is sucked into the eudiometer and measured the lab. tube again connected the .air passed over and any loss in volume shows the presence of N,O,. Some further experi- ments are in process of completion. The experiments with 0 absorptions by KHo and pyro. in the pre-sence of more or less COKo may be briefly summarised thus :-1. An excess of KHo over the pyro. employed should always be present and the absorption accelerated by agitation. 2. When the absorption of much CO by KHo has preceded that of 0 by pyro.it is advisable to add some more KHo so as to have twice as much of the caustic alkali as the acid present and if an excess of the latter has been added by accident more KHo should be passed up. The presence of COKoz does not then interfere with the absorption. 3. The alkaline solution of pyro. should be used in moderate excess as the absorption takes place more quickly and with the production of less GO. 4. If the reagents are used in the proportions recommended and the absorption accelerated by agitation the absorption will be complete in five minutes; but when much COKoL is present it is advisable to extend the time to 7; or even 10 minutes according to circumstances while the absorption is not complete in 15 minutes if the pyro. is in excess. I use a saturated solution of KHo and 1part of pyro. to 5 parts of OHz,and the word “excess ” as used in this paper is not intended to convey any idea as to the combining proportion of pyrogallic acid with KHo but simply a measurement bulk for bulk. So far as my experiments throw a light on the subject pyrogallic acid does not appear to form any definite compound with K€€CI.

ISSN:0368-1769    

DOI:10.1039/JS8773200028

出版商:RSC    

年代:1877

数据来源: RSC

摘要:

35 111.-Experiments on the Decomposition of Nitric Oxide by Pyrogdlate of Potash. ByW. J. RUSSELL, Ph.D. F.R.S. and W. LAPRAIK. WHILEmaking some experiments on gas analysis we were led to tl~e conclusion that nitric oxide was decomposed by pyrogallate of potash and on further experimenting with this gas we found that this decom- position invariably bakes place. In the following experiments the nitric oxide was prepared by the action of sulphuric acid on nitrate of potash and ferrons sulphate and the purity of the gas tested by its total absorption in a solution of ferrous sulphate. Different volumes of gas were taken and generally from 12 to 15 hours elapsed before the decomposition of the nitric oxide was found to be absolutely complete. We naturally expected that the action of the pyrogallate would be simply to convert the nitric oxide into half it#volume of nitrous oxide but we think the following experiments clearly sliow that this is not the only action which takes place.In the first three experiments the nitrous oxide was absorbed by alcohol and in the last one it was estimated'by explosion with hydro- gen. 1st expt. 2nd expt. 3rd expt. Absorbed by pyrogallate of potash .. 59.43 58-57 58.03 alcohol (N,O) ........ 37.67 39.21 7 99 nitrogen.. ......-..... 2-90> 41'43 2-76 ~ 100~00 100.00 100-00 Absorbed by pyrogallate of potash.. ...... 58.42 After explosion with hydrogen. ........... 38.20 , , nitrogen ............ 3.38 100~00 In all these cases more than 50 per cent.of the volume of tho gas disappeared and in all cases free nitrogen was found to be present. It wiil be seen; however that the nitrous oxide and nitrogen together only amount to a little more than 40'per cent. of the volume of nitric oxide taken still t'heir'volume is very fhirly constant viz. 40.57 41.43 41.97 and 41-58. In hopes of obtaining some further insight into this reaction we. next tried the action of pyrogallate of potash which had been fully-saturated with oxygen on nitric oxide. The solution of pyrogallate D2 3G RUSSELL AND LAPRAIK’S EXPERIMENTS ON THE was for five months exposedin a large basin to the air and afterwards placed in a large flask full of oxygen. Some of this liquid as in the former experiments was allowed to act on 6.773 C.C.of nitric oxide in a eudiometer. The reaction went on from January 6th to February 12th and then ceased. Vol. of gas absorbed by this pyrogallate. ... 76-01 Vol. absorbed by alcohol N,O.. .......... 14.24 Nitrogen.. ............................ 9.75 100~00 In th<s case the unabsorbed gas amounted to only 23.99 per cent. and consisted of 59.36 per cent. of nitrous oxide and 40.64 per cent. of nitrogen. We now took potash alone and tried its action on nitric oxide. It is stated in Gmelin’s Chemistry (vol. ii pap 378) on the authority of Gay-Lussac that when nitrous oxide is kept in contact for three months with strong potash it is resolved into one-fourth its volume of nitrous oxide and into nitrous acid which combines with the potash.The following experiments agree with this statement. 130th esperi-ments were started on January 23rd 1876. In August the action in the first experiment was still going on and the last reading was not taken until January in this year. The second experiment where there was a smaller bulk of gas was completed in May last 1st cxpt. 2nd expt. Volume of gas absorbed by potash .... 7-5-23 77.02 , unabsorbed ................ 24.72 22.95 100*00 100~00 The amount of gas unabsorbed was then very nearly the same as in the last experiment. In order to try whether this action could be much expedited by bent we placed a tube with the gas and potash in a water-bath arid heated them continuously until the action ceased. This took 13 days.The experiment was made in a non-graduated tnbe ;the result was- Volume of gas absorbed.. ........ 76.7 99 unabsorbed ........ 23.3 100.0 To ensure the air taking no part in these long exposures we have tried the same action in sealed tubes and in these cases have carefully examined the residual gas. The amount of contraction produced br the potash in the tubes could be but approximately estimated but the DECOMPOSITION OF NITRIC OXIDE. ETC. results are sufficiently precise to show that it is as in the foregoing experiments about three-quarters of the total volume. In the first experiment the diminution amounted to nearly 80 per cent. and in tl~e second to 77.3 per cent. 5-51 C.C.of the residual gas was taken in one case and 4.09 C.C.in the other.On analysis hhey were found to consist of- 1st expt,. 2nd expt. Nitrous oxide ............ 92.02 89.68 Nitric oxide .............. 1.45 2.66 Nitrogen ................ 6-53 7-66 100*00 100~00 In both cases the tubes were kept sealed for four months. Crystals of potassic nitrite separated out adhering to the inside of the tube. Essentially then the nitric oxide as Gay-Liissac states is by the slow action of potash converted into nitrous acid and nitrons oxide at' tlie same time another decomposition takes place possibly identical with that which gives rise to the nitrogen in the pyrogallate experiments but apparentiy the action is much slower. Two similar experiments were made in which the sealed tubes were placed in water-baths and heated for a fortnight but kept unopened for four months ; the amount of eontraction in one case was about 73 per cent.In the other the composition of the residual gas was-Nitrous oxide .............. 82.34 Nitric oxide .............. 2.56 Nitrogen.. ................ 15-10 100.00 The action of watcr alone on nitric oxide was also tried. The expe- riment was made in a sealed tube and was in every way like the last one only water was substituted for the solution of potash. The tube was heated in a water-bath for a fortnight and remained iinopened for four months. The contraction the gas had undergone was 46.7 per cent. but as the decomposition was not complete nitric oxide remaining in the tube the residual gas had the followiiig coin- position :-Nitrous oxide.............. 68.10 Nitric oxide .............. 18.38 Nitrogen.. ................ 13.52 100~00 In this residue there is a great increase in the amount of nitrogen as compared to the potash experiments but the relative quantities of nitrous oxide and nitrogen is nearly the same as in the experiment :!j8 PRUEN AND JONES ON THE ANALYSIS Ok’ CARBONATES wi%h the saturated solution of potassic pyrogallate. We find that the pyrogallic acid alone has no action on either of the oxides of nitrogen and that alkaline pyrogallate is without action on the nitrous oxide ; it swms then probable that the action of the alkaline pyrogallate is to convert the nitric oxide into half its volume of nitrous oxide but either from the simultaneous action of the excess of potash present or of certain compounds formed by the action of the oxygen on the pyrogallate another and more obscure action occurs very possibly a formation of some hyponitrite takes place.Our experiments arc not however sufficient to establish the nature of this reaction and for the present we wish merely to record the above results. Practically in the analysis of the gases obtained by burning the residue from waters it must not be assumed that a contraction of volame on the introduction of alkaline pyrogallake is a proof of the presence of oxygen.

ISSN:0368-1769    

DOI:10.1039/JS8773200035

出版商:RSC    

年代:1877

数据来源: RSC

摘要:

:!j8 PRUEN AND JONES ON THE ANALYSIS Ok’ CARBONATES IV.-Xhe Arzalyysis of Carbonates by means of the Carbometer. By S. T. PRUEN LL.D. and G. JONES TNEcarbometer for the volumetric estimation of carbonic anhydride is a modification of Sch ei b 1 e r’s well-known calcimeter and consists of two equal and graduated glass tubes one (A) stationary the other (B) capable of moving in a vertical direction. These tubes are con- ntAcrerl hy a piece of vulcanized india-rubber tubing attached to t’he lower extremity of each. The flask (G) in which the carbonic anhy- dride is generated contains a gutta-percha tube (H) for the hydro- chloric acid. The two tubes are each filled with water to zero aLow which is placed a thin layer of oil of equal depthin both. The reason we measure the column of gas from the point where the layers of oil and ~atermeet is because globules of oil cling to the side of the tube in descending and thus the oil alters its relative level.The water being below the oil does not cling to the sides and thus retains its relative level. The sliding tube (B) is attached to a thin slip of brass which slides along a vertical groove (C) made in the board. k cord (Id) fastened to this movable slip at one end is passed over a pulley and connected at)the other end to a metal weight (M) just capable of balancing the slide when the tube attached is filled with water to zero. To this weight which rests at the foot of the apparatus is attached a chain band (N),wound round a roller and equal in weight to but twice the length of the column of water between the points 100 C.C.mid zero in the sliding tube. A small thermometer (D) is fitted BY MEANS OF THE CARBOMETER. inside the upper portion of the first tube (A) passing through the india-rubber stopper to register the temperature of the gas. A stop-cock (E) is also fitted in the same place to allow the escape of any superfluous air which may be forced in when the stop- per is fixed in the flask after it has been charged. A chloride of calcium drying tube F intervenes between the flask and tube A. Let us now consider the action of the apparatus during use. Taking the figures of an analysis we made in the laboratory with pure calcic carbonate. 0-2 gram of pure calcic carbonate is weighed out and introduced into the flask ; the acid tube is then charged with 2 C.C.hydrochloric acid ;the stopper fitted into the flask and the stopcock E opened for a few seconds to equalise the pres- siire. The t'emperature and pressure is A then noted down 12" C. 756 mm. bar. ; and the flask gradually inclined until the wliole of the acid has mixed with the sample care being taken to grasp the flask by the neck only in order to avoid warming it by coiitact with the hand. The liberated gas after passing through A-the drying tube enters tube A and de- presses the column of liquid in it ; and a correspotiding amount of liquid is forced into B which thus increases in weight and consequently descends until it has drawn up by means ot'the cord and pulley a corresponding weight of chain.For inctance snpposing the liquid to have been depremed 1 centimeter in A it will have risen 1 centimeter in B. The di€ference between the two levels will therefore be 2 centimeters ; but I3 will have increased in weicht by a column of liquid 1centimeter in height and will consequently sink until it has pulled up a corresponding weight of chain which will be 2 centimeters long ; and thus the liquids in both tubes are kept nearly at the same level during the whole of the operation. The two tubes having been accurately adjusted the column of gas is read off. It is found to be 46.8 C.C. ; to this -16 C.C.is added being tl~ecorrection for the carbonic anhydride dissolved by the 2 C.C.of hytlrochlohc acid.This gives 46.96 c.c. the temperature being 12" C. arid the baroztieter stariding at 7'56 mm. Iteducing this. by the ordi- MUIR ON CERTAIS BISJfUTH COMPOUXDS. nary method to 0" and 756 mm. barometer we obtain 44-16 C.C. as the true reading. Now 1C.C. GO at 0" and 760 mm. barometer weighs e00196592 grm.; consequently when -2 gram is taken for snalvsis each cubic centimeter of gas at 0" C. and 760 mm. is equal to *!3h2!6 per cent. ; and multiplying 44.76 by -98296 we obtain its the result' of analysis 43.99 per cent. as against the theory 44.00 per cent. show-ing a loss of -01per cent. We have found as the result of many analyses that it is most c0n-venient to use hydrochloric acid 1 C.C. for each decigram of sample and to calculate according to Scheibler -08 C.C.as the amount of carbonic anhydride dissolved in each C.C. of hydrochloric acid.

ISSN:0368-1769    

DOI:10.1039/JS8773200038

出版商:RSC    

年代:1877

数据来源: RSC

摘要:

MUIR ON CERTAIS BISJfUTH COMPOUXDS. By 11.M. PATTISON MUIR,F.R.S.E. Assistant-Lecturer on Chemistrg The Owens College. 1. INthe fourth part of the present series of papers I described the action of potassium ferrocyanide upon a nearly neutral solutlion of bismuth nitrate. This action results in the formation of bismut)h femocyanide Bi45B’eCy6 a salt which when moist and containing traces of nitric acid readily undergoes decomposition by the action of the air. The results of the analyses of the greenish-blue substance pToduced by acting on bismuth ferrocyanide with very dilute nitric acid were most readily explained on the assumption that this substance is toler-ably pure bismut,h ferricyanide Bi,5PeCy6 (Glum. SOC.J. vol. i 187’7,p. 655). 2. I have now succeeded in preparing pure bismuth ferricyanide by precipitating a nearly neutral cold solution of bismuth nitrate with an excess of potassium ferricyanide washing the precipitate with cold water by decantation and drying over sizlphuric acid in vacuo.3. The salt as thus prepared presented the appearance of a tawny-yellow amorphous powder with a shade of green. It.was analysed by the process described in my former paper viz. decomposition by strong sulphuric acid separation of bismuth by precipitation with much water and titration of iron after reduction by means of per-manganate. I find that instead of separating bismuth from the whole of the liquid obtained by adding water to the mass from which sulphuric acid has been for the most part expelled it is preferable to add to this mass such a quantity of dilute sulphuric acid as suffices MUIR 0s CERTAIN BISNUTH COMPOUNDS.to form a clear solution to make this solution up to a dctermirintc volume (say to + a litre) to withdraw an aliquot portion for the wti- mation of iron to add ammonia to t'he remainder until the precipitate which is produced has a pale yellowish-red tint and then to add three or four drops of acid and a considerable quantity of water. After stand- ing some time the prccipitnted bismuth salts are collected washed once or twice and dissolved in the minimum quantity of hot nitric acid. The bismuth is then determined by the ordinary gravimetric process or by the improved volumetric method described by me in vol. I of the Society's Journal for 1877 p.658. As the analyses given in my former paper of the somewhat decom- posed bismuth ferricyanide evidently pointed to the fori~ula,Bi35FeCy, I have not thougbt it necessary to make more than a single arialysis of the purer preparation. 0.2793 gram gave 0.112 gram Bi,O = 0.1005 gram Bi. 0.468 , used 8.5 C.C. permanganate = 0.0744 gram Pe. (I C.C. permanganate = 0.00876 gram Fe). Cdculated for Bi,5FeCyG. Found. Bismuth = 37.29 37.18 Iron = 16.59 1591 Cyanogen = 46.12 46.91 (by difference)) 100*00 100~00 4. Ferricyanide of bismuth whether in a dry or moist condition is unaltered in air even after many days. If it be suspended in water and the water be boiled hydrocyanic acid is evolved the water acquires an acid reaction and a green then a greenish-blue and finally a blue salt (or salts) is produced.The presence of small quantities of acid more especially of nitric acid materially hastens this decomposition of' the ferricyanide. When the salt is dried at 100" it undergoes partial decomposition with evolution of hydrocyanic acid and production of' greenish-yellow substances. 5. The action of chlorine and of bromine upon ferro- arid ferri- cyanide of bismuth respectively is interesting. In the hope of being able to prepare ferri- from ferrocyanide of bismuth by a method other than that already described viz. boiling with dilute nitric acid I suspended a quantit,y of the nearly pure ferrocyanide (still contsining traces of potmh) in cold water and passed chlorine into the mixture until the blue colour of the original salt had entirely disappeared.The liquid was now of a brownish-re(l colour and contained a quantity of nearly white solid mattcr which quickly settled to the bottom of the vessel. After filtration arid wash- MLJIR ON CERTAIN BISMUTH COMPOUNDS. ing this solid mat'ter was found to contain bismuth and chlorine but not a trace of iron either in the form of ferro-or ferricyanide or of oxide. The filtrate showed the reactions of hydroferricyanic acid but was free from hydrofcrrocymic acid ; on evaporation over sulphuric acid it yielded cryst'als of potassium ferricyanide. No bismuth could be detected in the filtrate. 6. I performed a similar experiment keeping t,he water in which the ferrocyanide was suspended at the boiling point ; the result was altogether different from that detailed in par.5. The colour of the light yellow ferrocyanide slowly changed to pale blue then to green- blue and finally to deep blue. The supernatant liquid showed the reactions of hydroferm- and hydroferricyimic acids it contained no bismuth. An examination of the mloured substances produced 'in this reaction showed that they probably consisted of more or less pure bismuth ferricyanide mixed with varying quantities of some form of prussian blue and also with bismuth oxycbloride. 7. Another quant'ity of the ferrocyanide which contained traces of potash was suspended in cold water to which bromine was added the whole being shaken up from time to time.Judqing from the colour of the solid and liquid portions there appeared to be a slow change of ferro- into ferricyanide of bismuth ; after some hours however the main resultant of the action was a white precipitate containing his-muth and bromine and a supernatant reddish-yellow liquid which exhibited the reactions of hydroferricyanic acid.. 8. An experiment similar to that described in par. 6 was carried out substituting bromine for chlorine ; the result was analogous with that obtained when chlorine was employed. 9. The action of chlorine upon bismuth ferricyanide when suspended in cold and in boiling water respectively is analogous with the action of the same gas upon bismuth ferrocyanide ; the decomposition of tlie ferricyanide is not however in either case so rapid as that of the ferrocyanidc.The action of bromine upon the ferricyanide when sus-pended in boiling water is analogous with but less energetic than the action of the same element upon the ferrocyanide under the same con- ditions. If however ferriqanide of bismuth be suspended in cold water to which bromine is added a complete decomposition of the salt slowly ensues a reddish-white precipitate is produced which contains the whole of the iron and bismuth originally present in tlie ferri- cyanide while the supernatant liquid shows no traces either of hydro-ferro-or hydroferricyanic acid. 10. I have examined the action of chlorine and bromine respectively upon ferro- and ferricyanide of bismuth in the presence of cold and of hot solutions of caustic soda.The ferrocyanide is conipletely decomposed with the precipitation MUIR OX CERTAIN BISMUTH COMPOUNDS. of the whole of t'he bismuth and iron which it originally contained. This decomposition is more rapid with bromine than when chlorine is employed ; it is also more rapid in presence of hot than of cold caustic liquids. The first portions of the precipitate wbich is produced contain the whole of the bismuth the solution at the same time giving the reactions for hydrofwricyanic acid. If the supernatant liquid be decanted and be warmed with addition of more chlorine or bromine the decomposition is more rapid than if the action be allowed to pro-cecd in contact w-ith the precipitate which is at first produced.Bisniuth ferricyanide is not so readily decomposed as the ferrocyanide by either chlorine or bromine in presence of caustic soda whether hot or cold. Indeed I did not succeed in accomplishing the complete decomposi- tion of the salt by means of chlorine even after many hours' treat+ ment. By using a considerable excess of alkali and digesting with bromine in the cold $hen pouring off the supernatant liquid adding more bromine and genkly warming and repeating this process several times I succeeded in effecting the complete decomposition of bismuth ferricy anicle. 11. These reactions of the two double cyanides with chlorine and bromine may obviously be applied for the analysis of these salts. Bismuth ferrocyanide may be readily andysed by adding a con-sidcrable excess of caustic soda followed by addition of bromine and heating until the supernatant liquid when poured off from the pre- cipitate and boiled becomes colourless ; the precipitate is tlieri col- lected washed and dissolved in sniphuric acid.The solution is divided into two portions; in one iron is determined by means of permanganate after reduction by zinc ; in the other bismuth is sepn-rated fipom iron as already described and is dctermined. The alka-line filtrate (containing bromine) separated from the precipitated iron aiid bismuth oxides niay then be employed for the deterniirmtion of cyanogen. This determination is most conveniently performed by the use of a standard solution of silver nitrate if it be desired to analyse bismuth ferricyanide by a similar pro-cess it is advisable to digest tlie salt with water and brmiine in the cold for some time then to add caustic soda to continue the digcstioii to pour off the clear liquid to add niore bromine and tinally to gently warm this liquid until tlie decomposition is complete.The process is more applicable for the analysis of the ferro- than of the ferricyenicle of bismuth. 12. Strong nitric acid when added to bismutli ferrocyanide witliout warming quickly causes the formation of bisinutli ferricyanide ; if the temperature be raised hydrocyanic acid is el olved and the greater part of the salt present passes into solution evidently however only after undergoing decoiriposition. MTrIR ON CERTAIN BISNUTH COJIPOU?;DS.13. If bismuth ferricyanide be suspended in water to which sodium amalgam is added the colour of the suspended salt gradually becomes paler until an alniost colourless substance is produced which from its general reactions I have no doubt is bismuth ferrocyanide. 14. The action of chlorine and bromine upon ferro- and ierricyanide of bismuth respectii-ely as also the action of nitric acid upon the for-mer and of sodium anialgam upon the latter salt is I think in keep- ing with the general analogies existing between ferro- and ferri- cyanides. Most probably the first action of the halogens upon tlie ferrocyanide is to withdraw part of the bismuth with the production of hydroferrocyanic acid and bismuth ferricyanide. Hydroferrocyan ic acid it is m7ell known readily undergoes decomposition in the air yielding some form of prussian blue.In the manufacture of ferri-cyanide of potassium the formation of a green salt is frequently noticed. This green salt or a salt of analogous composition would appear to be produced in many of the actions which I have described in the preccAding paragraphs. The decomposition of bismuth ferrocyanide by chlorine in presence of cold water appears at first sight to be more completc than the cle-composition of the same salt by chlorine in presence of boiling m-atrr. T believe however that the apparent anomaly is explaiii~d by regard-ing-as indeed analysis shows that m-e must regard-thc former dc-composition as resulting in the production of bismutliyl chloride and hydroferrocyanic acid while at the high temperature of the second reaction the hydroferrocyanic acid is quickly decomposed w-i th thc production of some form of prussian blue which colours the precipi- tated hismuthyl chloride.The rapid decomposition of the hydro- ferrocyanic acid produced with subsequent precipitation of a form of prnssian blue will naturally tend to prevent the action of the chlorine upon the undecomposed ferrocyanide. The blue or green substanco produced when chlorine is passed into boiling water in which bismuth ferrocyanide is suspended will therefore probably contain hismn th ferricyanide undecomposed ferrocyanide bismuth oxychloride and a form of prussian blue. 15. Ferrocyariide and ferricyanide of bismuth when heated in closed crucibles over a Bumen lamp yield a black-brown maw containing iron bismuth carbon and small quantities of cyanogen.

ISSN:0368-1769    

DOI:10.1039/JS8773200040

出版商:RSC    

年代:1877

数据来源: RSC

摘要:

45 No. 'VI.-OY~ (x iV&od for Detecting Xmall Quantities of B73is~rufh. By M. M. PATTISON MUER,F.R.X.E. Assistant Lecturer on Chemistry The Owens College. 1.INvol. lxxxviii of Pogg e n d orf f '9 Animken page 45,there is a paper entitled "Researches upon Bismuth," by R. Schneider. The paper is mainly concerned with a detailed account of the author's investiga- tions upon an oxide of bismuth containing less oxygen than bismuthous oxide (Bi203). In describing one of the methods adopted for prepar-ing this oxide Xchneider mentions the fact that a solution of bismuth in nitric acid is precipitated by addition of tartaric acid but that the precipitate (tartrate of bismuth) so produced is soluble in caustic potash. He also shows that stannous chloride when mixed with excess of tartaric acid and then with such a quantity of potash as is sufficient to neutralise the acid yields a clear liquidwhich is unaltered on boiling or on dilution with water.He further shows that by mixing the prepared bismuth with the prepared tin solution, and adding an excess of potash a very dark brown liquid is obtained. Schneider proves that this liquid owes its colour to the presence of dissolved hypobismuthous oxide (Bi,O j contained in it. 2. As I have been lately and am still engaged with a study of the properties of this hypobismuthous oxide (I hope to be able to lay the results before the Society at an early date) it occurred to me that a colorimetric process for estimating small quantities of bismuth might be based upon the reaction described by Schneider.On trial how-ever I found that the reaction was not delicate enough to serve as the basis of a quantitative colorimetric method but I think we have in Schneider's reaction the means of detecting very small quantities of bismnth. 3. The fest liquid I prepare as follows:" 4 grams of stannous chloride are dissolved along with 12 grams of crjstallised tartaric acid in such a quantity of tolerably strong caustic potash solution as serves to produce a clear liquid having a distinctly alkaline reaction. This liquid must remain perfectly clear when it is warmed for some time to a temperature of 60" or 70". In order to prepare the solution which is supposed to contain bis- muth for the addition of the test liquid I add a considerable quantity of tartaric acid warm and then make alkaline by addition of potash.To t>his liquid I add a few C.C. of the test solution warm the mixture to 60" or 70" for a few minutes and place the beaker upon a piece of f Schneider's directions somewhat modified. 46 MCIR ON DETECTING SMALL QUANTITIES OF BISNUTH. white paper. The production of a brownish-black colour shows tlie presence of bismuth. 4. One part of bismuth in 210,000 parts of liquid may be thus detected. 5. I have examined the action of the stannous chloride solution upon solutions of metals other than bismuth prepared by adtling. tartaric acid to an acid or neutral solution of each metal followeJ by addition of excess of potash. The following are the results :-Cudmiurn yields a white precipitate without.any shade of brown or black. Mercury yields a grevish-black precipitate. Lead yields no precipitate nor is the liquid darkened. Copper yields a yelIowish-red precipitate. Arsenic yields 110 precipitate the liquid is not darkened. A?itimon!y yields no precipitate the liquid is not darkened. Irorc yields no precipitate the liquid is not darkaed. Cob& yields no precipitate the liquid is not darkened. Nickel yields no precipitate the liquid is not darkened. Chromium yields no precipitate the liquid is not cl'nrkeneci. Mangaaese yields a yellowish-red solution which gradually darkens. Mercury copper and manganesc can therefore alone interfere with the reaction the first-named metal if present must certainly be removed before applying the test although the others yield precipi- tates or coloured solutions yet this is only the case when they are present in comparabively large amount. 6. I had hoped to have perfected a volumetric method for deter-mining bismuth by precipitation as hypobismuthous oxide by means of the prepared tin solution employing mercuric:chloride as an indicator but as yet I have not succeeded. I would propose to call the prepared tin solution " Schneider's Reagent."

ISSN:0368-1769    

DOI:10.1039/JS8773200045

出版商:RSC    

年代:1877

数据来源: RSC

摘要:

47 VII.-Ooiztributions to the Histo7-y of the Naphthalene 8el-ie.x. Xo. I. Nitroso-@-naj&hol. STENHOUSE, By JOHN LL.D. F.R.S, and CHARLESE. GROVES. FTJCHS (Deut. Clzem. Ges. Rejp. viii 625) has shown that as in the case of phenol an atom of hydrogen in cc-naphthol can readily be directly replaced by the nitrosyl group NO giving rise to two isomeric nitroso- a-naphthols. He has also obtained a nitroso-plnaphthol (ibid. 1026) by a similar process but on repeating the experiment we found that not only was the dirty greenish coloured precipitate containing the nitrosonaphthol very difficzllt to pnrify but that the yield of the nitroso compound was far from satisfactory. It seemed probable however that the use of nitrosyl sulphate which had answered so admirably as a nitriting agent for the preparation of nitroso-orcin and nitroso-resorcin might also give better results in this instance A pre-liminary trial at once showed this to be the case.Nitroso-@-naphthol. On mixing dilute aqueous solutions of sodium /3-naphtholate and of nitrosyl sulphate a brilliant yellow precipitate of the nitroso compound was at once formed but on standing a portion of this soon rose to the surface and gradually became darker in colour probably owing to secondary action between the nitroso-naphthol and anattacked naph-thol ; the portion which remained suspended in the liquid retaining its bright yellow colour. After several trials the following process was found to yield the best resulks 1part of pure @-naphthol was dissolved in 10 parts of boiling water by means of 1 part by measure of sodium hydrate solution of sp.gy.. 1*323,* cooled and poured into 100 parts of water. A nitriting solution was then prepared by pouring 2 parts by weight of the 15 per cent. nitrosy1 sulphate solution previously described (this Journal 1877 vol. I p. 545) into 200 of water with constant stirring and the dilute aqueous solution of sodium P-naphtholate above mentioned was at once added and thoroughly and intimately mixed which when small quantities were operated on was most conveniently done by pouring the liquid from one vessel ink0 another. After stand- ing 12 to 20 hours the precipitate of crude nitroso-p-naphthol was collected on a linen filter and washed’with cold water until the wash- ings ceased to be acid.This precipitate when dried weighed rather more than the @-naphthol originally taken. * A solution of this strength contains 30 grams sodium oside in each 100 c,c. 4% STENHOUSE AND GROVES OX THE Puchs purified his crude uitrosc,-P-coinpound by dissolving it in petroleum and concentrating by distillation. Tlris however is ex- ceedingly inconvenient as the slight solubility of the substance in petroleum renders it necessary to use very large quantities of the latter; moreover the product obtained in this manner is far from pure. h modification of this method however obviating the distilla- tion of large quantities of petroleum was adopted in the first instance for the partial purification of small quantlities of the nitroso-compound.The crude product after being carefully dried at the ordinary tem- perature or at a very gentle heat was agitated with 200 times its weight of light petroleum at about 40" C. and filtered from the insoluble portion. A few drops of an alcoholic solution of ammonia was then added to the clcar petroleum solntipn which produced a bright green precipitate consisting of the ammonium compound of nitroso-p-naphtl-iol. This WRS collected and the petroleum after removal of the excess of ammonia by a,gitntion with dilute sulphuric acid employed again to extract the crude product. The green ammo- nium compound when exposed to the air to allow the pet8roleum to evaporate gradually lost part of its ammonia becoming yellow. The remainder of the ammonia was removed by moistening it with acetic acid and washing with water but the nitroso-naphthol thus obtainccl even after erystallisatiori from boiling petroleum was still impure and had to be converted into the sodium salt and precipitated in the manner subsequently described.As Fuc hs states that he obtained two isomeric derivatives from a-naphthol the nitroso-P-naplithol was submitted to a careful fractional crystallisation from petroleum but no evidence could be obtained of the existence of a second isomeric nitroso compound. The attempts made to purify the crude product by crystallisation from various solvents such as alcohol etlier benzene carbon bisul-phide acetic acid &c. met with but indifferent success as did also the precipitation of the alcoholic solution by ammonia or soda.The m&hod ultimately adopted was based on the insolubility of the barium compound which is thrown down as a bulky green precipitate on adding a solution of barium chloride to a solution of the nitroso com- poud in a dilute alkali. The moist crude product mwtioned above as obtained from 1part of /3-naphthol was suspended in 35 parts of water that is sufficient water was added to the moist crystalline paste to make it up to 35 parts by measure ; 1 part of the 30 per cent. solution bf sodium hydratle diluted with 35 of water was added to this and the mixture agitated occasionally and filtered after standing about mhour." * In this and the subsequent operations distilled water was used 8s the lime in ordinary vater forms an insoluble compound with a portion of the nitroso naphthol occasioning loss.HISTORY OF THE NAPHTHALENE SERIES. The clear olive-brown filtrate was tben precipitated by a slight excess of a dilute solution of barium chloride,-about 1.5parts of a cold saturated solution diluted with twice its volume of water,-and the green pre-cipitate collected on a fine cambric filter. The reddish-brown filtrate when neutralised with an acid gave a dark-brown precipitate contain- ing a crystalline substance this has not been further examined as yet. The green barium compound after being thoroughly washed was suspended in about 35 parts of water and decomposed with hydro- chloric acid in excess. The brownish-yellow partially purified nitroso- cornpound was then thoroughly washed until free from barium chl ride again dissolved in dilute soda,-% parts water and 0.5 parts :)(-.per cent. solution of sodium hydrate,-filtered precipitated as barium compound washed decomposed by acid &c.,a,? before and the same process repeated a third time. In this way the nitroso-@-naphthol was obtained in minute needles of a brilliant yellow colour containink water of crystallisation. They retain their colciur when dried at the ordinary temperaiure but may readily be converted mto the anhydrous compound by crystallisation from hot spirit (5 parts) the solution must not however be boiled for any length of time as decompositioii would take place. The best method of recovering the nitroso-com- pound dissolved in the spirituous mother-liquors is to add one-tenth of its bulk of the solution of sodium hydrate; this throws down R green precipitate consisting of the sodium compound of the nitroso- @-naphthol.As the barium compound is bulky and very finely divided it was found better when large quantities were to be operated on to slightly modify the process just described. The barium compound first obtained (from 1part of naphthol) after being decomposed with hydrochloric mid and thoroughly washed was suspended in 15 parts of water 0.5 parts 30 per cent. solution of sodium hydrate added arid the mixture heated in a water-bath. On filtering and adding onc-tenth of its bulk of the soda solution the sodium compound of thc nitroso-naphthol being almost insoluble even in comparatively dilute solutions of sodium hydrate was thrown down as a green precipitate ; this was collected pressed and .decomposed by acid the process of precipitation as sodium compound being repeated if necessary.The amount of pure nitrsso-@-naphthol obtained in this manner is rather more than half the weight of the naphthol originally taken. A full account of the method of preparation and purification of this sub-stancs has been given as success depends in a grcat measure on careful attention to details. The anhydrous substance crystallised from hot alcohol and dried at the ordinary temperature iir VCLCWO,was submitted to an il-pis with the following results :-VOL. XXXII. E STENHOUSE AND GROVES ON THE T.*318 gram substance gave -810 gram carbonic anhydride and ,119 gram waker. IT. .263 gram substance gave -672 gram carbonic anliydride and *lo0gram water. 111. *386gram substtance gavc 27.2 cub. ccnt. of nitrogen at 13.5"C. and under a pressure of 751.6 mm. (corr. at 0' C.) equivalent to 25.239 cub. cent. dry nitrogen at 0" C. or .OX662 gram of nitrogen Theory. I. 11. 111. Mean. C, = 120 69.36 69.47 69.68 -6'3.57 H7 = 7 4.05 4.16 4.2'2 -4.10 N= 14 . 8.09 -8-20 8-20 I O2 = 32 18.50 ~ 173 100*00 These numbers correspond to those required by the formula ClOH7NO2, or C,,H,(NO).OH that of nitroso-naphthol. The hydrated nitroso-6-naphthol is of a brilliant yellow colour and crystallises in minute necdles. It is obtained in this state when any of its salts me decomposed by dilute acids or when an alcoliolic solu-tion of the compound is poured into water.The yellow necdles losc water at a gentle heat and become brown. The anhydrous compound ob- tained by crystallisation from alcohol ether benzene carbon bisulphide or petroleum forms thin plates or short thick prisms of an orange-brown colour melting at 109.5". It is only slightly soluble in water even when boiling and separates almost entirely as the solution cools in long yellow needles. It dissolves very readily in carbon bisulpliidc benzene ether acetic acid and hot alcohol and requires ahout 42 parts of alcohol at 13" for solution. It is comparatively slightly soluble in light petroleum even when boiling. Nitroso-B-naphthol dissolves in cold conccritrated sulpliuric acid with development of heat forming a bright red solution from which the substance is precipitated unaltered on the addition of water.On strongly heating the acid solution the colour changcs to brown and the addition of water then no longer produces a precipit:tte probably owing to the formation of a sulphonic acid. The nitrosc,-corriponncl is dccomposed when heated with concentrated nibric mid yielding a tarry substance which dissolves on long continued boiling with the strong acid. The compounds which nitroso-6-naphthol forms with ammonia and the alkalis are crystalline and of a bright green colour when in a finely dirided state. although the masses of crystals deposited fi-om a hot concentrated solution appear black by reflccted light.The derivatives containing the mctals of the alkaline earths are also of various shades of green. HISTORY OF THE NAPHl'HALENE SERIES. ~~n,on,it.ro-P-N~t~ol. Although as stated above nitroso-B-naphthol when heated mith concentrated nitric acid is decomposed and converted into tarry pro-ducts the action is different when it is submitted to the rcgulatecl action of the dilute acid at the ordinary temperature. The pure yellow crystalline hydrate waq suspended in ten times its weight of water and mixed with an equal bulk of nitric acid of sp. gr. 1.25. It soon began to change in character the yellow iieedles aggregating in flocks whilst the odour of nitrous acid became distinctly apparent.In the course of an hour the precipihte had again become dense and lion-coherent the colow having changed to a yellowish-grey. It was at once col-lected washed thoroughly with water and treated with 50 parts of R cold very dilute solution of soda (1part 30 per cent. sol. to 50 of water) and filtered from the insoluble residue. A slight excess of acetic acid was then added to the decp orange-coloured filtrate ; ad t'he bright yellow precipitate collected washed and crystalliscd orice or twice from boiling alcohol in wliich it is very soluble. Dried in 5 vacuum and analysed it gave the following results :-I. *215 gram substance gave *499 gram carbonic anhydride and -074gram of water. 11. -649 gram of substance gave 40.6 cub. cent. of nitrogen at 10.5" C.and under a barometric pressure of 770*.2mm. (coi~~ at 0') equivalent to 39.134 cub. cent. dry nitrogen at 0" or *049094gram of nitrogen @lo = 120 Theory.63.49 I. 63.30 11. - H = 7 3.70 3.82 - N = 14 7.41 - 7.56 03 = 48 25.40 - - 189 100-00 From these results it will be seen that the new substance as miglit have been expected is a monohitro-p-naphthol formed from the corrc'-sponding nitroso-compound by the displacement of the NO group by NO,. This nitro-naphthol is precipitated from its solutions as a pult yellow crystalline powder apparently a hydrate. It crystallises from alcohol in orange- brown plates scarcely distinguishable in appearance from the nitroso-compound. It melts at 96O and forms orang(.-coloured crystalline compounds with the alkalis and ammonia.It is more soluble than the nitroso-compound but behaves very like the latter with conceiitrated sulphuric or nitric acids This nitro-P-naphthol however is not the only compound formetl Eg STENHOUSE AND GROVES ON THIC by the action of dilute nitric acid on the nitroso derivative as a portion of the crude prodmt is insoluble in dilute alkaline solutions; it is crystdline and of a pale-yellow colour. Again if the nitroso-g-naphthol be suspended in spirit instead of water and then treated with dilute acid and allowed to stand a considerable time a greenish crystalline substance is formed almost identical in appearance with that above described. On examination however it was found to consist of two compounds t'reatment with a dilute alkaline solution left an inso- luble crystalline residue of an orange colour whilst the addition of an mid to the filtrate threw down a pale-coloured flocculcnt precipitate considiing of minute slender needles.With ammonia the latter forms a sparingly soluble salt crystallising in deep yellow rhomboidal plates These substances have not been further examined as yet. /3-naphthapuione. When the ammonium compound of nitroso-&naphthol was treated with hydrogen sulphide it was reduced to an amido-compound. It was found more convenient however to pass hydrogen sulphide into the barium compound suspended in dilute ammonia as the finely- divided precipitate was more readily and completely attacked than the crystalline ammonium salt ; 2 parts of pure nitroso-&naphthol were dissolved in a mixture of about 100 parts of water with 2 of the soda solution filtered and precipitated with barium chloride in slight excess.The precipitate after being collected and washed was stls-]'ended in 140 parts of water ; 3 parts of ammonia solution (S.G.. 0.880) were added and hydrogen sulphide passed to saturation. At first the barium compound assumed a more intense green colour which how- ever rapidly disappeared as the solution became saturated with hydrogen sulphide the green precipiiate being replaced by almost colourless scales of the amido-compound. After passing a gentle current of hydrogen sulphide for about an hour longer the 'vessel containing the mixture was tightly corked and allowed to stand several hours witli occasional agitation.The precipitate was then collected waslied with a very dilute solution of hydrogen sulphide and dissolved by sus-pending it in 140 parts of water and adding dilute sulphuric acid in excess. It was also found advisable to add a little sulphurous acid solintion as otherwise the amido compound oxidised rapidly during fil- tration. The clear filtrate containing the stroiigly acid solution of fimido-~-napkttliol sulphate was immediately poured into a filtered solution of two parts of potassium dichrornate in 20 of water when the quinone at once separated in slender needles of a bright orange colour. It was collected without delay thoroughly washed with dis- tilled water and dried ati the ordinary temperature.It was found on HISTORY OF THE NAPHTHALENE SERIES. incineration to contain traces of chromium ; but as it decomposes very readily it was not practicable to remove the chromium by crystallisa- tion from any solvent. When the solutions are carefully filtered ant1 the operations rapidly performed the quinone is quite pure with the exception just mentioned. It melts at 96". The substance was dried in waxcuo and analysed with the following results in which a correction has been made for the chromium pre,sent. I. ~142gram of substance gave -396 gram of carbonic anhydride and -030 gram of water. 11. -139 gram of substance gave -387gram of carbonic anhydricle and -050 gram water. Theory. I. 11. Mean. c, = 120 75.95 76.05 75.93 75-99 H = 6 3-80 3-91 4.00 3.95 042 = 32 20.25 158 100*00 The analytical results correspond very closely with those required by the formula CI0H6O2, showing it to be an isomericie of the naphtho- quinone previously discovered by one of us (Groves this Journal xxvi 209).As its general characters and reactions moreover correspond with those of the quinones we have called it /3-mphthnquinone,distin-guishing the readily volatile yuiuone formed by the direct oxidatioii of naphthalene or af amido-a-naphthol as a-rLa~~~tliaqicino,ie. Like the ordinary quinones 6-naphthaquinone when treated with hydriodic acid yields a hydroquinone crgstallising in colourless needles which appeared to be reconverted into the quinone by the action of oxidising agents.The same compound is produced by the action of sulphurous acid ; the quinoiie dissolves readily in aqueou.3 sulphurous acid forming a colourless solution which deposits the. hydroquinone when concentrated by evaporation at a gentle lieat. When alcoholic solutions of the quinone and hydroquinone are mixed water added and the solution gently heated dark-coloured cr;stals are deposited resembling those obtained from the corresponding a-corn- pounds under similar circumstances The 6-naplitbaqainane dissolves when heated with dilute nitric acid and the solution on cooling deposits a crystalline substance of a magnificent crimson colour. If however the quinone be boiled with nitric acid for some time it is completely oxidised t'o phthalic acid which may be obtained by evapo- rating the solution to dryness and recrystallisiag the residue from water.The phthalic acid readily yields a sublimate of phthalic aiihy- dride which with resorcin gives the characteristic fluorescin reaction. Adopting Graebe's symbol for naphthalene it is evident that both 54 ACWORTH AND ARMSTRONG ON THE REDUCTION OF oxygen atoms in the (3-quinone are in the same benzene nucleus since it yields phthalic acid on oxidation; moreover one of the oxygeu atoms occupies the &position since it is a derivative of 6-naphthol. The relation of the two oxygen atoms to one another is undetermined although considering the readiness with which the quinone is attacked and decomposed and the difficulty with which it is volatilised it is moye probable that the oxygen atoms have a meta-relation than an ortho one.This substance is of special interest as being the first instance of an isomeric quinone derived from the same hydrocarbon although it should not be forgotten that in the benzene series we have a brominated compound C6HBr3O2,the “tribromoresorquinone” of Liobermann and Dittler which unless the formula be doubled is isomeric with tri bromoquinonc. We purpose continuing our investigation of the new substances mentioned in tbis paper and also of instituting a careful comparison of the two isomeric naphthaquinones.

ISSN:0368-1769    

DOI:10.1039/JS8773200047

出版商:RSC    

年代:1877

数据来源: RSC

摘要:

VIII.-C?OMMUNICATIONS FROM THE LABORATORY OF THE LONDON INSTITUTION. XeAearcjzes om the Beduction of Nitric Acid nrd the Oxides of Nitrogen. Pccrt I. On the Qases Evolved by the Actiom of Metals on Nitric Acid.* By J. J.ACWORTH E.ARMSTRONG, and HENRY F.R.S. CONTENTS. Page 9 I. Introductory ............................................ 56 $ 11. Theory of the formation of the various products obtained on dis-. solving metals in nitric acid .............................. 56 5 111. Method of experiment .................................... 57 $ IT. Mode of stating the results ................................ 60 (i V. Action of copper on nitric acid.. ............................ 60 § YI. ) silver , ..............................71 $ VIL. , zidc > .............................. 73 3 VIII. , cadmium and magnesium on nitric acid,. ............ '17 $ IX. .. iron on nitric acid.. .............................. 79 4 X. , nickel cobalt indium and aluminium on nitric acid . . 81 § XI. .. tin lead and thallium on nitric acid ................ 84 (i XII. .. alloys on nitric acid .............................. 86 * The substance of this paper was communicated to the Society on June 30th) 1876 and the experiments described in it were made previous to this date NITRIC ACID. § I. Int?~odzcctory.-Although it is a generally recognised fact that the composition of the gases evolved when metals are dissolved in iiitric acid depends not only on the metal employed but also on the strength of the acid and the temperature at which the reaction takes place no sufficient examination of the influence of these various factors on the nature of the gaseous product has ever been made and what is even more remarkable no quantitative determination of the amount of gas produced by the dissolution of a giver1 weight of any metal in acid of a certain strength appears to have been at- tempted.It is not easy to understand how so important a problem as that of the action of metals on nitric acid should have so long been re- garded we may almost say with complete apathy by chemists. Doubtless the difficulty of collecting the gaseous product without loss and in a pure state and of analysing ths somewhat complex mixtures obtained-difficulties which are now entirely removed by the invention o€ those most important engines of research the Sprengel-pump and the gas-analysis apparatus-has in part contributed to this ; pro-bably however it is chiefly attributable to the influence of the dogmatic teaching to which we have long been subject in these matters for in scarcely any of our text-books is the least attempt made to explain the changes which result in the formation of the various oxides of nitrogen and of nitrogen itself.We find merely the statement that they are the products of the more or less complete deoxidation of the acid and a single equation is invariably used to express that which undo~ibtedly involves a series of changes and should therefore be represented by a series of equations ; indeed as a rule explanatory equations are conspicuous by their absence from our text-books and hence one of the most effective means of awaken-ing a desire in the student to analyse the ~ihenomena brought under 1~isnotice remains undeveloped.The views which wem long entertained with regard to the action of metals on nitric acid are succinctly stated in the following extract from Gmelin's Chemistry (English translation vol. ii 1849 p. 397) :-" Nitric acid,* at ordinary temperatures or at the boiling point oxidises all metals excepting siliciurn titanium tantalum platinum rhodium iridium and (under ordimry circumstances) gold. The resulting metallic oxides (except t,hose of tungsten tellurium tin and arsenic) combine with the undecomposed portion of the acid arid form salts.. . . . . . In this reaction the portion of acid which oxidises the metd is converted sometimes into hyponitric acid sometimes into iiitric oxide nitrous oxide or nitrogen gas or-if the metal at the * By nitric acid is here meant the compound represented by the formula NO5 (0 = 8). ACWORTH AND ARMSTRONG ON THE REDUCTION OF same time decomposes water the hydrogen of which then combines with the nitrogen of the acid-into ammonia. Which of the above products is formed depends partly on the affinity of the metal for oxygen partly on the temperature and concentration of the acid.” The explanation herein contained although perfectly sufficient as long as the older theory of the nature of acids and of the forma-tion of salts prevailed ceased to be so when the acids came to be re-garded as hydrogenised compounds (“ hydrogen salts ”) ; nevertheless the language used in most chemical works even at the present day;is almost invariably such as to lead the reader to attribute the reduction which nitric acid undergoes when acted upon by metals to the metal itself whereas it can scarcely he do.ubted that it is due not to the direct action of the metal but to the action of the hydrogen displaced from the acid by the metal.Miller’s Elements of Chemistry (5th ed. 1874 vol. ii p. 154) however is the only text-book in which it is to our knowledge that this latter explanation is given. 8 11. Theory of the Formatiom of the various Products obtained 0% dissolvirLy Metals in Nitric Acid.-From what is already known on the subject and from our own experiments we believe the following to be the most consistent and probable explanation which can be given of the forrnation of the gaseous and other products obtained by dissolvirig various metals in nitric acid.We regard the action of the metal as consisting simply in the dis- placement of the hydrogen of the acid and the formation of the cor- responding nitrate in the mamer expressed by the following equation in which B denotes the amount of any metal which is equivalent in combining or displacing power to two atoms of hydrogen 1. R + 2HNO3 = R(N03)2 + 2H. Under no circumstances however is the hydrogen thus displaced evolved as such ; it at once acts on the free acid present (and even in some cases as we shall show later on also bn the metallic nitrate which has been formed) reducing it more or less completely to nitrous acid nitrosidic acid,* hydroxylamine or ammonia as shown by the following equations 2.HNO + 2H = HN02 + OH,. 3. HNO3 + 4H = HNO +‘=OHZ. 4. HN03 + 6H = NH,(OH).+ z 9H 5. NHOs + 8H = NH3 +fiOHp Q We employ this name to designatc the acid-if indeed it be an acid-the salts of which mere first described by Diver 6 in 1870 (Roy.SOC.Proc.,p. 426) in prefer- Piice to that of hyponitrous acid as it nq be regarded a8 a compound of nitrosgl (NO) with hydrogen. NITRIC ACID. The gases which are evolved on dissolving metals in nitric acid result we believe from the decomposition of these reduction products and from their actian upon each other.The nitric oxide is chiefly if not entirely. formed by the decomposi- tion of the nitrous acid in accordance with the equation 6. 3HN02 = 2N0 + HNO.3 + OH,. We say chiefly if not entirely as it is possible that some nitric oxide may be produced from the nitrosidic acid since according to Di vers although silver nitroside (AgNO) is soluble in dilute nitric acid with- out immediate decomposition it is immediately oxidised by the 'corn-centrated acid copious red fumes making their appearance ; and moderately diluted nitric acid decomposes it with evolution of nitrogen and production of apparently both nitrous and nitric acids in the solution.It is also possible that hydroxylamine may furnish nitric oxide on oxidation with nitric acid. The nitrous oxide is. in all probability chiefly the product of two distinct changes viz. a of the spontaneous decomposition of nitrosidic acid which as Divers has shown breaks up with the greatest readi- ness into nitrous oxide and water ; thus 7. 2HNO =NZO + OH,; and b of the action of nitrous acid on hydroxylamine which as V. M ey er has shown takes place in the following manner 8. NH,(OH) +HNO =NZO +20HZ. Finally we regard the nitrogen' as chiefly the product of the action of the niirous acid on ammonia 9. NH3 +HNO =N +SOH, but it is perhaps in some cases in part and in others entirely tlie pro-duct of the action of nitric acid on nitrosidic acid.The explanation we have advanced cannot be regarded as cornplete until the action of the variaus reduction-products upon each other under all the conditions which can occur in practice has been rigor- ously investigated-a task of no slight difficulty and magnitude ; but before undertaking this it appeared desirable to gain more definite information as to the behaviour of various metals and especially as to the quantitative relation which the amount of gas produced bears to the amount of metal dissolved. 9 111. Method of Ezperiment.-In all our experiments we have em-ployed weighed quantities of metal and before bi-inging the latter into contact with the acid the air has been exhausted as completely as possible from the apparatus ; we have then allowed the action to take ACWORTH AND ARMSTRONG ON THE REDUCTION OF place and have afterwards collected not only the gas evolved spon- taneously but also that given off on repeatedly heating the nitric acid solution in a vacuum.The total quantity of gas obtaiiied has alwaFs been measured and its composition dehermined ; in most cases the gas evolved during the dissolution of the metal and on subse-quently exhausting previous to Beating has been collected apart from that evolved on heating and the two portions separately analysed ; sometimes the gas has been pumped off at various intermediate stages of the operation and the several portions separately analysed. In this way much information has been obtained which is of value in the dis- cussion of the sources of the gases evolved.The apparatus employed in most of our experiments consisted of a glass flask formed by expanding the one end of a piece of half-inch tubing about 4 inches loug into a bulb of about 25 C.C. capacity at a short distance above which was blown on the side of the tube a small bulb or pocket for the reception of the metal the acid being placed in the lower bulb. The flask was closed by a caoutchouc plug through which passed a narrow glass tube provided with a stopcock bent twice at right angles ; by means of this tube-the flask was connected with a U-tube filled with pieces of solid potassium hyclroxicle the U-tube being in connection with the Sprengel-pump. It was impossible with this apparatus to employ concentrated nitric acid OIL account of its action upon the caoutchouc-plug and also on account of the metal becoming acted upon by the vapours of the acid before the exhaustion was complete.A more perfect form of appa-ratus was therefore devised consisting of a flask similar to khat just described but having a ground-glass stopper through which passed the tube of a dropping funnel provided with a stop-cock. It is repre- sented by the figure on p. 59. Although it is not to be expected that with an arrangement of the kind described the ingress of air can be entirely prevented we have found that if all the joints are kept covered with glycerin the leakage is so slight in the case of the glass apparatus that practically the result is uninfluenced thercby ; but in the case of khe simpler form of apparatus there is undoubtedly an appreciable error introduced in consequence probably of the diffusion of air through the csoutchouc- Plug.111 making an experiment the weighed piece of metal (as a rule foil was used) was placed in the pocket of the flask and the acid -of which 15 C.C. were usually employed-was either poured into the lower bulb of the flask or sucked into the dropping funnel ; the various parts of the apparatus having been connected together the Sprengel- pump was set in action and the pumping continued until the exhaus- tion was judged sufficient wheu the metal was tipped into the acid. NITRIC ACID. 59 In most cases pumping was not re-commenced until the metal was entirely dissolved ; it was 'then carried on as long as an appreciable amount of.gas was delivered. The stop-cock in the tube csniiecting the flask with the U-tube containing potasainm hydroxide was now closed and the bulb of t,he flask surrounded with water heated to about 50"-60" ; after heating it in this manner for a few minutes it was surrounded with cold water and when cold the stop-cock mas opened and the gasfswhich had been evolvcd from the solution were pumped off. This alhernate heating cooling and removal of' the gases evolved was then several times repeated until in fact the quantity of gas obtained was almost inappreciable-which was usually the case after the second or third heating. 60 ACWORTH AND ARMSTRONG ON THE REDUCTION OF The mixture of nitric oxide nitrous oxide and nitrogen obtained was analysed in the usual manner the nitric oxide being determined by the addition of an excess of oxygen in presenceof potassium hy-droxide solution and subsequent removal of the unabsorbed oxygen by means of pyrogallol ; the nitrous acid by explosion with hydrogen.It was frequently necessary to add a known volume of oxygen with the hydrogen in order to secure an explosion when the mixture con-sisted chiefly of nitrogen. The excess of hydrogen remaining after the explosion was usually determined in opder to check the determina- tion of nitrous oxide. The analyses were almost iiivaiiably performed in duplicate. The results of the nitric oxide deberminations were almost uniformly satisfactory buk khe estimation of the nitrous oxide in the extremely small amount of gas frequently remaining after the removal of the nitric oxide was not always satisfactorily effected ; no difficulty however was experienced in the analysis of mixtures rela- tively rich in nitrous oxide of which a fain amount was available for combustion.8 IT.Node of stating the .ResuZts.-In order to facilitake the corn-parison of the results obtained with various metlals and under various conditions they are exhibited in the form of tables in which are given the weight of metal taken ; the temperature of the acid st the com- mencement of the experiment; the st'rength of the acid of which unless otherwise stated 15 C.C. were always employed ; the number of C.C.of gas obtained ; the percentage composition of the gas ; and the number of C.C.of gas per twit-weight of metal i.e. the quantity of metal equivalent to 2 grams of hydrogen. 8 V. Action of Copper on Nitric Acid-The results of our experi-ments on the action of copper on nitric acid of various strengths are exhibited in Tables I and 11. In discussing these results it will be necessary to consider a the cornposition of the gas evolved with various strengths of acid it being remembered that unless otherwise stated whatever the strength of the acid the same number of C.C.(15) of liquid was always placed in the bulb of the flask ; b the influence of temperature on the composition of the gas as to which however we have hitherto made very few experiments; c the volume of gas measured at 0" C.and under a pressure of 760 mm. of mercury ob-tained per unit-weight of metal ; and d the alteration in the composi- tion of the gas due to the presence of varying quantities of cupric nitrate. If the results of the experiments with acid of strength 1 2 (Nos. IT1 to X inclusive) are compared it will be seen that on the whole the composition of the gas evolved is remarkably constant from about I. Action of @qpron Nitric Acid. Percentage coinposition C.C. of gas per unit-weight of C.C. of gas calc. Weight 8trength C.C. of gas. metal. as NO per uiiit- No. of of metal T. of acid of gas weight of metal. ri expt. taken. used. obtained. lheory = NO. N20. Total. NO. N20. 1480 C.C. I .. . . . -35 15" 1:Q 2 -4 54 '0 10 '3 35 *7 434 TI....* . . ,3215 15" 1:l 43 *48 98 -17 -92 -91 8574 8417 78 78 8891 62 022 97 '9'7 1'33 90 42 422 98 *33 -87 -86 2 43*28 97 -85 1-45 -70 s 42 -01 98 '01 1-21 -18 2i 41.90 99 '26 -'74 3- 15789 E .....I IT1 -020 13" 1:2 98 -26 *99 9.5 15272 15007 151 114 U 47 *16 97 *63 1-69 -68 40.41 97 *82 1-47 -71 42 $7 9'7 *54 1.55 '91 40 9.3 97 '84 1*64 '52 45-18 99 .lo -*90 0. IV .... *918 loo 1:2 216.15 97 -99 1*21 *80 14928 14628 180 120 15508 v ... . . * . '920 11" 1:2 201-99 97.15 1.55 1'30 13919 13522 206 191 14707 VI ..*. *. -395 11" 1:2 '79 *55 98 '12 -95 *93 12768 12327 121 120 13189 VTI .. . ... -1525 8" 1:2 20.65 96 *89 -30 2 -81 8584 8316 27 241 9193 vm .... '1512 11" 1:2 33 *lo 97 .83 1.55 1-18 13879 13508 215 156 14267 IX ......-150 50" 1:2 22 -91 96.98 2 -27 -75 9683 9390 219 74 10220 c w I. Actioiz of Copper 0% Nitric Acid-continued Percentage composition C.C. of gas per unit-weight of C.C. of gas calc. Weight Strciigtli C.C. of gas. metal. as NO per unit-No. of of metal T. of acid of gas aTeight of metal. expt. tatken. used. obtairi cd. Theory = NO. NiO. Ns. Total. NO. NZO. 142380 C.C. ~ x ...,... . -2285 goo 1:2 51 -40 97 -23 1.a2 -95 14261 13866 259 136 15009 46 -53 93.68 4.43 1'89 16 *74 95 93 1*18 2 -89 XI ..... -329 123 1:4 63 *27 94 -28 3.57 2 -15 12192 11494 435 263 13530 40 *43 73 -30 19 .90 6 -80 15 -03 67.40 22.30 10 *30 10.70 72 *91 21 .so 5 -29 XI1 ....,. .325 16" 1:s 66 '16 71 -89 20 -74 7.3'1 12845 9234 2664 947 (?)19492 3'7.34 80 -07 16 -07 3 -86 10.45 96 *53 1*57 1*90 XI11 ....-2895 20" 1:8 47 -79 83 *67 12.88 3.45 10466 8756 1348 361 13552 NITRIC ACID ri3 97 to 98 per cent. being nitric oxide the residue in all cases consisting of nitrous oxide and nitrogen. The eomposition of this residue appears indeed to be somewlmt variable :but in considering the com- position of the residue it is to be remembered that thc apparatus cannot be completely exhausted of air before tlie metal is broiiglit into contact with the acid; that probably there is a slight leakaye in of air during the experiment ; and that it is difficult to avoid errors in the analysis of the gas remaining after the removal of the nitric oxide as the quantity which remains is extremely small even if as much as 10 C.C.be analysed-and more than this was seldoni employed froni 5 to 7 C.C. being the quantity usually taken for an analysis. Taking the a\-erage of the experiments me are probably not far wrong in assuming that slightlly niore nitrous oxide than nitrogen is produced. Turning now to the experiments witli more concentrated and with weaker acid it will be seen (Experiment IT) that having regard to the composition of the gas produced the 1 1acid has about tie same action as the weaker 1 2 acid but the result obtained with the more concentrated 1 0 acid (Experiment I) is very different prob:rL,ly in consequence of the retention of the nitrous acid by the nitric acid. The 1 4 acid (Experiment XI) furnishes a gas containing only about 94 per cent.of nitric oxide but about twice as rich in nitrous oxide as that furnished by the 1 2 acid. With ths 1 8 acid (Experiments XI1 and XIII) tjhe diminution in the proportion of nitric oxide and increase in the proportion of nitrous oxide and of nitrogen is very marked indeed; so that commencing with the 1 2 acid it appears that nitrous oxide is produced in an increasing proportion as the strength of the acid diminishes ; at the same time the rate at which the copper dissolves also diminishes but of this we shall say more presently. The only observations we have instituted as to the influence of tcm-perature have been with the 1 2 acid. Of eight experiments wit11 this acid six were made by allowing the metal to dissolve in it at the atmospheric temperature ; but in the other two (IX and X) the acid was heated to about 50" and 90" respectively before introducing tile metal into it aftcr which the heating was not continued.The metal dissolved much more rapidly in the hot acid especially in thzt Iir~tecI to about 90"; but the pas obtained scarcely differs in composition from that gexerated on allowing the metal to dissolve at the ordinary temperature. The volume of gas obtained per unit-weight of metal (63.4 grams) dissolved varies considerably even in experiments with the same strength of acid thus of two experiments with 1 2 acid (IV ancl VII) the one gave 14928 c.c. and the other only 8584 C.C. This how-ever is probably due to the retention of the nitrous acid in solution 64 ACWORTH AND ARMSTRONG ON THE REDUCTIOX OF for we have found that the amount of gas obtained by decomposing silver nitrite by nitric acid varies according to the strength of the acid being greater the weaker the acid.It is always less however then is required on the assumption that the nitrous acid is decom-posed in accordance with the equation 3HN02 = 2N0 + HNO + OH,. For example in an experiment in which ~4493gram silver nitrite was decomposed by 15 C.C. 1 2 acid 11,975 C.C.of nitric oxide were obtained instead of 14,880 C.C. required by theory if 1 litre of hy-drogen at 0" and under the pressure of 760 mm. of mercury weigh *089578 gram (Regnanlt) and 30 grams of nitric oxide are con- taiiied by a space of 22.320 litres.We may mention that the dccom- position was effected in the glass apparatus shown on p. 59 and that the gas as collected contained 99.25 per cent. nitric oxide ; whereas in three similar experiments made at the commencement of the inves- tigation in which the apparatus provided with a caoutchouc stopper was employed the gas as collected contained nearly two per cent. of nitrogen. That the decomposition of nitrous acid is especially retarded by nitric acid there can be no doubt ;for if sulphuric acid be employed the nitric oxide is given off with greater facility and in larger amount. Indeed in three experiments which furnished closely accordant results in m-hich silver nitrite was decomposed by a 10 per cent. solution of sulplzuric acid we actually obtained more than the required quantity of nitric oxide viz.15800 C.C. The amount of nit'ric oxide obtained depends howerer vory much on the manner in which the experiment is conducted. Gas is at once given OEwhen the acid and nitrite are brought into contact and a further quantity is evolved as the apparatus is exhausted ; but after a time the pump ceases to deliver any appre-ciable quantity of gas and in order to decompose any portion of the nitrous acid remaining in solution within a reasonable time it is neces- sary alternately to heat the liquid and then to cool it and pump off the gas in the manner previously described. If weak nitric acid be employed it is possible by repenting the heating &c. a sufficient number of times entirely to decompose the nitrous acid in solution but apparently no amount of heating will effect this if the acid be concentrated ; it remains to be investigated whether the nitrous acid is merely held in solution by the nitric acid or whether it is not actually combined with it-a supposition which cannot be regarded as altogether unwarranted.When copper is dissolved in nitric acid the gas is evolved in a precisely similar manner a portion being given 08 during the dissolution of the metal a further portion on exhausting and a very considerable proportion on beating &c. ; but the amount of nitric oxide furiiished previous to the application of heat is as a NITRIC ACID. rule less than when silver nitrite is decomposed by nitric acid and the gas is also given off with much less facility on heating.In fact the presence of the copper salt appears to exercise a special retarding influence on the decomposition of the nitrous acid such as no other metal possesses of which we have examined the action on nitric acid. Thus in the case of zinc we have found that after the second or third heating the amount of gas evolved is inappreciable? whereas with copper a measureable quantity of gas is often obtained at the eighth or tenth heating. In Experiment VIII for example only about one- third of the gas collected was given off in the cold and the heating had to be ten times repeated before gas ceased to be evolved. It is not improbable therefore that in certain cases in which low numbers were obtained (Experiments VII and IX) the heating &c.was not repeated sufficiently often to entirely decompose the nitrous acid pro-duced during the dissolution of the metal. If the nitric oxide be as we have supposed derived from nitrous acid the unit-weight of metal should furnish 14880 C.C. ; thus CU + 2HNO3 = 2H + Cu(N0,)2. 2H + HNO = HNO + OH,. 3HN02 = 2N0 + HNO + OH,. .* . 3Cu = 2N0 or 44.640 litres. and Cu = 4-4-+4-0 or 14880 C.C. NO. It will be noticed however that without taking into account the relatively small quantity of nitric oxide and nitrogen obtained slightly more than this amount of nitrous oxide appears actually to have been produced in one case (Experiment 111); while in three others (Experiments IT X and XIV) the nitric and nitrous oxides and nitrogen together are somewhat ih excess of the amount which according to our hypothesis can be produced.This is probably in the case of Experiments I11 and IV at least in part due to the accu- mulation of a number of positive errors of experiment. In both the gas was collected in five portions; of each of these a portion was analysed and if the analysis was satisfactory the remaining portion was merely measured; while if the analysis appeared doubtful the remaining portion was also analysed SO that the total volume is deduced from ten measurements. Moreover as in the one experiment *92and in the other -918 gram of copper was taken any error in the mea-surement of the gas is multiplied nearly seventy times in calculating the volume of gas per unit weight of metal ;so that a small initial error becomes magnified into a large one.This explanation however does not apply to Experiments X and XIV and for the present we must hesitate to attribute the excess entirely to errors of experiment espe- cially since as was above stated we have also obtained more than the VOL. XXXII. I? ACWORTH AND ARMSTRONG ON THE REDUCTION OF theoretical amount of nitric oxide on decomposing silver nitrite by dilute sulphuric acid ; and on this account it is obviously undesirable also to attempt any further explanation until the decomposition of nitrous acid under various conditions has been carefully studied. One of us has previously shown (J. J. Acworth this Journal 1875 p.830) that when nitric oxide is prepared in the ordinary manner by acting on copper with nitric acid the gas evolved gradually becomes less pure and especially richer in nitrous oxide as the amount of cupric nitrate dissolved in the liqcid increases ; and it was found that if cupric nitrate were added in the first instance to the acid the amount of nitrous oxide in the gas produced was very much greater in all stages of the experiment than when the acid alone was employed. But in these experiments relatively large quantities of acid and of copper were used and portions only of the gas evolved were collected at intervals and analysed. In order therefore if possible to ascertain the nature of the influence exerted by cupric nitrate we have made several experiments in the manner already described but dissolving nitrate in the acid before bringing the latter in contact with the metal.The results we have obtained are exhibited in Table 11. Apparently the reaction which leads to the production of the nit,rous oxide does not take place to any great extent unless a relatively large amount of the nitrate be present or the solution be very concentrated. Thus the gas evolved by the action of 1 2 acid satwated with cupric nitrate (Experimentt XVII) contained nearly 10 per cent'. of nitrous oxide and a somewhat higher proportion of nitrogen than would have been obtained in the absence of the nitrate ;whereas in an experiment (XV) in which acid of the same strength but only 5 grams cupric nitrate were taken the gas evolved contained about twice as much nitrous oxide as would have been obtained without the addition of the nitrate.Experiment XVI in which the conditions were precisely similar to those in Experiment XV furnished a soniewhat different result; but the determination of nitrous oxide in this case was un- satisfactory and we bclieve that the number given in the table is con-siderably too low. It should be mentioned that the copper was very mid more rapidly dissolved by the saturated solution of cupric nitrate in nitzic acid than by the acid alone. The results of Experiment XIV compared with those of Expe-riments 111 to X show however that the quantity of acid employed is of consequeace as well as the strength of the acid about double the proportion of nitrous oxide having been obtained when only 5 c.c instead of the usiml amount of 15 C.C.was taken. E'rom this experi- ment and those with cupric nitrate it may with much probability be concluded that the extent to which the nitrate takes part in the re- action becomes greater in proportion as the amount of nitric acid IT. Action of Coppe~o~bNitric Acid. -Percentage composihion C.C. of gas per unit-weight of C.C. of gas calc. Weight Strength C.C. of gas. metal. as NO per unit- No. of of metal T. of acid of gas weight of metal. expt. taken. used. obtained. Theory = NO. NzO. Nq. Total. NO. NgO. Na. 14880C.C. -52 *40 97 -38 1'88 94 41.97 96 -21 3 -24 *55 36 '51 92.98 5 243 1*59 XIV .... -660 11" 129 '88 95 -79 3 -28 -93 14704 14085 482 137 15926 1:2 41 -26 91 -83 5-71 2 -46 tllld 5 45 '56 99 -60 -*50 $-grams xv a,....-4725 13" CuN,Ofj 86 *82 2 -71 1-44 11649 11140 314 215 12693 2 XVI ..... -1520 12O ditto 24 *36 1-09 3 *15 10160 9729 111 320 11091 P 41 *74 11*28 2 '40 1:2 38 -42 8-08 2 -49 sat. with XVII 0. -459 10" CuN,Ofi 80 -16 87.80 2 -46 11072 9721 1078 2'73 13505 43 -77 22 *lo 74 -47 +4 1:2 37 '39 20 -80 77 *27 to alla 26 TO 64 *10 35 *no 5 grams 6.66 12 *84 87 *15 NH4NO:3 XVITI .*. *41 13" 114.52 30 *92 67.12 17708 5475 348 11885 - ACWORTH AND ARMSTRONG ON Tl3X REDUCTION OF diminishes and it may be expected that results of some interest will be obtained by extending the investigation in this direction. It certainly appeared to be somewhat remarkable that the presence of an excess of cupric nitrate should lead to an increased production of nitrous oxide and apparently also though in a minor degree of nitrogen.For a long time we were in doubt how to explain its action until it occurred to us that probably the only way in which it could take part in the reaction would be by undergoing reduction to cuprous nitrate and that the nitrous oxide was perhaps more or less directly a product of the subsequent re-oxidation of the cuprous nitrate by the nikric acid. This was the more probable as Mr. Acworth had found that the presence of potassium nitrate caused little or no alte- ration in the composition of the gas evolved from nitric acid and copper. In order to test the validity of our hypothesis as cuprous nitrate wa8 not procurable we dissolved cuprous oxide in nitric acid and the result obtained appears entirely to justify our assumptions.The amount of cuprous oxide taken was ~5845of a gram and 15 C.C.of 1 2 nitric acid at a temperature of 20" were employed ; the oxide dis- solved rapidly and 26.19 c.~.of gas were collected without the appli- cation of heat 11C.C. being subsequently given off on heating &c. The percentage composition of the two portions of gas was as fol- lorn-s 64.21 96.06 26.19 C.C. NZO 28-32 11.0 C.C. { 1-59 K 7.47 2.35 Hence the percentage composition of the total gas (37.19 c.c.) col-lected is NO = 73.70 N,O = 20.38 N = 5.92 and the volume obtained per unit-weight of oxide (142.8 grams) NO = 6696 C.C.N,O = 1852 C.C. N2= 537 C.C. Total = 9085 C.C. The evidence afforded by this experiment as compared with those in which the metal was dissolved in 1 2 acid is of the most definite character and in our opinion necessitates the conclusion that the hydrogen displaced from nitric acid by copper has less reducing power than cuprous nitrate. It also renders it extremely probable if it does not actually prove that the nitrous oxide obtained on dissolving copper in nitric acid of various strengths is not derived from a compound or compounds formed by the reduction of a portion of the acid by the hydrogen displaced fi-om another portion of acid by the metal but that it is a more or less direct product of the reduction of the acid by the cuprous nitrate ;in other words we incline to the belief that whatever NITRIC ACID.the reduction-product or products may be from which the nitrous oxide is directly derived the hydrogen displaced from nitric acid by copper has not the power of reducing nitric acid to such an extent that nitrous oxide is ultimately produced. But it will be remembered that the amount of nitrous oxide produced increases as the strength of the acid diminishes; hence if we suppose the reduction to be effected entirely by the hydrogen displaced from the acid by the metal the hydrogen displaced when the weaker acid is employed would appear to have a greater reducing power than that displaced when the metal is acted upon by the stronger acid which is obviously contrary to our proposifion.Not only however is it theoretically improbable that this shonld be the case but as we shall show later on when discussing the results of our experiments with other metals and especially with zinc where an action similar to that which we have supposed is ex- erted by the cuprous nitrate is inconceivable such a conclusion would be entirely opposed we believe to the evidence at our disposal. The dissolution of copper takes place as we have already stated with increasing slowness asthe strength of the acid diminishes ; a quantity of metal for example which is dissolved by 1 2 acid within 5-6 hours requiring 36-48 hours to dissolve in 1 8 acid ; and probably this circumstance is the chief if not the only cause of the difference in the results observed with acids of different strengths the explana- tion which we venture to advance being that a greater number of molecules of cupric nitrate come within the sphere of action of the hydrogen atoms displaced from the acid and that consequently a greater number of molecules of cuprous nitrate are formed when the metal slowly dissolves in weak acid than when it more rapidly dis- solves in stronger acid.It may be suggested as a more probable hypothesis that the increased production of nitrous oxide is due to the presence of a relatively greater number of molecules of cupric nitrate when weak than when strong acid is used affording opportunity for a greater number of molecules of the former to undergo reduction ; our observations (Experiments XV XVI and XVII) showing that the amount of nitrous oxide produced on dissolving copper in 1 2 acid in presence of an excess of cupric nitrate ie.more than is formed from the metal itself employed in relatively small quantity is not much in- creased unless a relatively very large amount of the nitrate be present would seem however to negative such an assumption. It remains only to consider the evidence afforded by our experi-ments with copper and nitric acid as to the manner in which the gases evolved are produced. In relation to this question the amount of gas evolved per unit-weight of metal is of especial importance. The amount of nitric oxide theoretically obtainable supposing this gas to be derived sslely from the decomposition of nitrous acid it will be '70 ACWORTH AND ARMSTRONG ON THE REDUCTION OF remembered is 14880 C.C.per unit-weight of metal ;if the nitrous oxide be derived from the decomposition either of nitrosidic acid or of hydr-oxylamirie nitrite (p 57) the unit-weight of metal should furnish 3580 C.C. ;and iiS the nitrogen result from the action of nitrous acid on ammonia at most 4464 C.C. of this gas could be evolved per unit- weight of metal dissolved. The amount of nitric oxide equivalent to the nitrous oxide produced may therefore be calculated by multiplica- tion of the latter by 2.666 and in a similar manner by multiplying the amount of nitrogen evolved per unit-weight of metal by 3.333 we obtain the eqDivalent amount of nitric oxide.The results of the various experiments thus reduced are exhibited in the last column of Tables I and 11. It will be seen that in the majority of cases the amount of gas obtained was below that which can be produced if the nitric and nitrous oxides and nitrogen are formed in the manner we have supposed and the decomposition be complete; and if Experi-ment XI1 be excepted that in those cases in which the theoretical quantity is exceeded (Experiments 111,IV X and XIV) the excess is comparatively slight. Experiment XI1 was one of those in which nitrous oxide was largely produced and the result which it furnished is apparently altogether abnormal as the amount of gas obtained (19492 c.c.) is nearly one-third in excess of the theoretical quantity. As the result of a second experiment (XIII) under what we believe to have been similar conditions mas very different it might be thought that some error had been committed in the measurement of the gas in the first experiment ; but we do not think this probable especially as the two experiments differ considerably in other respects.Thus it will be observed that in Experiment XI1 the last portion of gas col- lected which consisted almost entirely of that given off on heating &c. contained a very considerable proportion of nitrous oxide (21.8 per cent.) whereas in Experiment XI11 the portion of gas col- lected in a similar manner contained very little nitrous oxide (1.57 per cent.). In other words in the one experiment practically the whole of the nitrous oxide was evolved without the application of heat while in the other a considerable proportion was evolved only on heating.From the observations we have made in the course of our experiments with other metals as to the manner in which nitrous oxide is evolved we are inclined to believe that that given off in the cold and that given off on heating are products of essentially distinct reactions and that the former is probably derived chiefly from nitro-sidic acid and the latter chiefly from hydroxylamine nitrite. If this be the case the difference in the results of the two experiments above referred to would appear to indicate that although the conditions seemingly were the same the reduction had taken place to a greater extent in one than in the other.But this conclusion interesting as NITRIC ACID. it is in itself in no way serves to explain the great " excess " of gas obtained in Experiment XII; in fact it must be admitted that although the explanation we have advanced holds good in the majority of cmes especially with regard to the manner in which the nitric oxide is formed there is in others distinct evidence of its insufficiency. It also must not be forgotten that we have determined only the gaseous products and that before a complete explanation of -&he reaction in all its phases can be given we require in addition to this a knowledge of the amount of the reduction-products remaining in solution. The importance of determining these latter will be evi- dent when it is recollected that as me have already pointed out a considerable quantity of nitrous acid may escape decomposition into nitric oxide &c.and is rendered still more obvious by the fact that ammonia may be present in considerable quantity together with nitrous acid without much nitrogen being produced and that there- fore the amount of nitrogen obtained-supposing it all to be derived from the decomposition of ammonic nitrate-is but an imperfect mea- sure of the amount of ammonia formed. This is shown to be the case by Experiment XVIII Table 11,where although 5 grams of ammo-nium nitrate was added to the 15 C.C. of 1 2 acid in which the copper was dissolved no less than 30.92 per cent. of the gas obtained con- sisted of nitric oxide; hence if it be supposed that an amount of nitrous acid was formed corresponding to the quantity of copper dis- solved less the amount equivalent to the nitrous oxide produced as much as 39.2 per cent.of the nitrous acid produced underwent the normal decomposition into nitric oxide &c. notwithstanding the re- latively very large amount of ammonium salt present. The amount of nitrogen evolved is within 428 C.C. of that which could be formed from the remaining nitrous acid supposing that it was completely decom- posed in accordance with the equatiou HNO + NH3= N + 20112. § TI.Action of Silver on Nitric Acid.-The number of cxpcriments we have made with silver is very small but they appear to us to afford confirmation of the conclusions we have drawn from those with copper.On reference to the following table (111) it will be seen that the gas collected consisted only of nitric oxide and nitrogen :in two of the experiments however the amount of nitric oxide obtained was considerably less and the amount of nitrogen considerably greater than in the third experimeut. This difference in the results is we believe due to the fact that in Experiment XIX the gas was pumped off immediately the metal dissolved whereas in the two other experi- ments at least twelve hours elapsed after the metal dissolved before the apparatus was exhausted in which period a certain amount of the nitric oxide evolved during the dissolution of the metal was ACWORTH AND ARMSTRONG ON THE REDUCTION OF 111. Action of Silver on Nitric Acid.-I Percentage composi-C.C. of gas per unit-weight tion of gas. of metal. T. NO. NzO. Na. Total. NO. NzO. N,. ~~~~ I--~~ XIX ... '4 10' 98.15 1.85 12511 12280 231 271 xx .... -4 11° 97 *18 2'82 9601 9330 XXI . . . -4 13' 96 90 3'30 9342 9033 309 -absorbed by the nitric acid and reconverted into nitrous acid 2N0 + HNO + OH = 3HN0,. We may mention that the quantity of silver employed was about four hours in dissolving and that the dis- solution proceeded more and more rapidly as the amourit; of nitrous acid in solution became greater which is in accordance with the obser- vations of Russell and others. In Experiment XIX about half the gas was given off previous to heating ; while in Experiment XX of the 17.78 C.C.of gas obtained 9.5 C.C. were evolved in the cold ; and in Experiment XXI only 5.92 C.C. were collected before heating and 11.38 C.C. after heating several times. The gas collected after heat-ing in Experiments XX and XXI contained about the same proportion of nitrogen as that collected in a similar manner in Experiment XIX and as the apparatus employed was that provided with a caoutchouc plug it is highly probable that the increased amount of nitrogen obtained in Experiments XX and XXI was due to leakage ;we are in fact almost inclined to regard the whole of the nitrogen obtained in these experiments as derived from external sources rather than as the product of the action of the silver on the acid the amount actually collected in Experiment XIX being only *43c.c.a quantity which is easily accounted for when it is remembered that the exhaustion of the air from the apparatus is necessarily incomplete in the first instance and that even if actual leakage does not occur at any of the joints t'here is undoubtedly a slow diffusion of air through the caoutchouc' Plug-The non-production of nitrous oxide affords strong evidence in favour of the explanation we have advanced of the manner in which this gas is formed when copper is dissolved in nitric acid for there is no probability in the case of silver that a reaction can take place corre- sponding to the reduction of cupric to cuprous nitrate ; and from its general behnviour and especially from Tho m sen's thermochemidal investigations (J.pr. Chew,.[el xii 27l) there is every reason to believe that the energy of silver is much below that of copper ; conse-quently if the hydrogen displaced from nitric acid by copper be inca- NXTRIC ACID. pable as we have suggested of directly reducing nitric acid to the extent sufficient for the production of nitrous oxide it is extremely improbable that the hydrogen displaced by silver should have this power which is in accordance with our observat.ions. 9 VII. Action of Zinc on Nitric Acid.-The remlts of our experiments with zinc are exhibited in Table IT,from which it will be seen that whatever the strength of the acid the gas evolved is always a mixture of nitric oxide nitroils oxide and nitrogen. As in the experiments with copper the amount of nitric oxide evolved per unit-weight of metal varies not only with acid of different strengths but also with acid of the same strength ; probably however this is at least in part due to the difficulty with which as we have before pointed out the nitrous acid produced is entirely decomposed especially in presence of concentrated nitric acid.In our opinion the most important element for consideration in the case of the metal under discussion is the amount of nitrous oxide pro-duced as this probably corresponds strictly with the amount of nitro-sidic acid and of hydroxylamine-supposing these to be the two bodies from which the nitrous oxide is more or less directly derived-formed in the reaction the decomposition of the former into nitrous oxide and water being doubtless immediate while the production of nitrous oxide by the action of nitrous oxide on the latter according to IT.Meyer’s observations (Ann. Chem Phnrrn. clxxv 1411,takes place with great facility. Now it will be observed that the amount of nitrous oxide evolved on dissolving the metal in acid varying in strength from 1:8 to 1 2 at the ordinary temperature that is to say in Experiments XXII to XXXII inclusive but excluding Experiment XXIX is on the whole remarkably constant the lowest amount being 2422 C.C. (Experiment XXVIII) and the highest 2862 C.C. (Experiment XXII) ;and as there is no great variation in the amount of nitrogen these results would appear to indicate that within the specified limits of concentration the extent to which the various reactions take place is influenced in a comparatively slight degree by alterations in the concentration of the acid.From the increased amount of nitrogen produced in Experi- ments XXXV and XXXVI it would almost appear that considerably more ammonia had been produced than in the experiments with weaker acid. This explanation is not a necessary one however as another explanation may be given of the increase in the amount of nitrogen. Thus we have already pointed out that the amount of nitro-gen evolved is but an imperfect measure of the amount of ammonia formed in the reaction as a considerable proportion of the nitrous acid fails to act on the ammonia present even if the latter be in great 4 &-IV. Action of Ziizc on Nitric Acid Percentage composition C.C.of gas per unit-weight of 2- Weight St,rength C.C. of gas. metal. Per cent. of 0 No. of of metal T. of acid of gas 4 metal 0 expt. taken. used. obtaiiied. accounted for. NO. N,O . Total. NO. N2O. ri XXII . . . -4 5" 1:8 35 '33 46 *05 49 *86 4 *on 5741 2G43 2862 236 74.33 XXIII.. .. '4 6" 1:s 35 '80 47 -95 48 -08 3 -97 5817 2780 2797 231 74 *G3 XXIV.. .. -4 6" 1:4 37 52 53.45 43.35 3 -24 6097 3268 2643 196 73 .64 XXV *.. . *4 8" 1:4 35 *33 49 *34 46 -10 4 -56 5741 2832 2636 263 72 '33 XXVI.. * '4 5" 1:4 34 -39 49 -08 47 'GO 3 *32 5588 2742 26'59 18'7 70.25 XX.VII .. *415 '7" 1:4 34 *a7 46 *7l 50.30 2 *99 5461 2650 2748 163 70 *02 XXVIII .. -226 6" 1:4 15 -80 38 -08 53 -30 8 $2 4544 1730 2422 302 63 .so XXIX ...~283 go* 1:4 25 .I7 57.44 35 '20 7 -36 5781 3320 2034 427 68 *32 xxx . . . '4 7" 1:2 31.30 45 *30 49 -90 4.80 5086 2304 2538 244 65 97 XXXI ... *4 8" 1:2 31 -63 47 95 48 *08 4 -17 5140 2454 2411 215 65 58 23 93 33.52 60 '1'7 6 -31 4.06 84 '27 5.95 9 '78 XXXII .. -386 12" 1:2 27 -79 40 *91 52 -21 6 m 4679 1014 2442 323 63 *85 4.7'64 22 *G2 70 -79 6 -59 40 *68 18 .OO 73.78 8.22 14*52 32 *27 61.23 6.50 'I*64 72 *14 19.54 8 '32 XXJII ,,. 1.6 10" 1:2 110'48 25 -59 67 *08 'I*33 4488 1148 3010 330 68 *88 XXXIV ,* *221 90" 1:2 19 '31 51 *88 39 -99 8.13 5679 2946 2271 462 70 *83 IV. Action of Zinc om hTitric Acid-continued. Percentage composition C.C. of gas per unit-weight of Weight Strength C.C. of gas. metal. Per cent.of No. of of metal T. of acid of gas metal expt. taken. used. obtained. accounted for. NO. N2O. Total. NO. N20. 19.62 66 -66 8*64 2 *76 8 *88 13 -28 XXXQ .. -3455 15" 1 :1 22*38 31 '23 59 -56 9 -21 4210 1314 250'7 389 62-46 XXXVI . . '321 15" l:o 15m -95 78 -29 20.76 3175 31 2485 659 59 '49 1. 4 and 5 E granis 0 XXXVII . *4 6' NWdNO 41 -4.3 41 '98 3'7 -48 20.54 6'732 2826 2523 1383 95 -18 * XXXVIII. -242 6" ditto 21 *16 41 *59 41 *25 17 -16 5928 2466 2445 1017 83 *16 5? P 39 -00 -3.69 96 *31 sat. sol. 9.35 -100*oo NH,NO, XI; -179 12O 48.35 -97 -03 17557 -521 17036 85 -64 .I.**. ACWORTH AND ARMSTRONG ON THE REDUCTION OF excess ; that this is the case is further proved by Experiments XXXVII and XXXVITI where notwithstanding 5 grams of ammonium nitrate were added as much nitric oxide was obtained as in many of the experiments in which no such addition was made.But as the amount of nitrous acid escaping decomposition into nitric oxide &c. is greater the more concentrated the acid we may expect that if a greater amount of nitrous acid remain in solution or indeed if it undergo dissociation more slowly when concentraied than when dilute acid is employed the amount of nitrogen evolved will be greater in the former than in the latter case even supposing the amount of ammonia pre- sent in both cases to be the same. In Experiments XXIX and XXXIV in which the acid was heated to about 90’ before dropping in the metal considerably more nitrogen was obtained than in corresponding experiments in which the acid was not thus heated.* A somewhat similar explanation as the above may be advanced in regard to these results for supposing the same amounts of ammonia and of nitrous acid to be formed under the two sets of conditions it is probable that more of the ammonia would suffer decomposition by the nitrous acid at the higher temperature.We are not however at all inclined to regard this explanation as sufficient but believe that our experiments rather justify the conclu- sion that reduction does take place to a greater extent and proceed further the more concentrated the acid and the higher the tempera- ture. Still this is but a mere opinion and must remain such until not only the gaseous products but also those which remain in solution in the acid shall have been determined.The last experiment with zinc and nitric acid to which we haire to direct attention is that numbered XXXIII where the amount of metal dissolved was four times as great as in the other experiments under otherwise similar conditions. It will be observed that in this case considerably less nitric oxide and considerably more nitrous oxide was obtained than in any of the experiments with smaller quantities of metal. This probably is due to the fact that when a larger quantity of metal is employed there is a greater opportunity for the nitrous acid produced by the action of the first portions of metal dissolved to undergo reduction side by side with the nitric acid. It is worthy of note however that this “division of labour,” so to speak cannot take place with dl metals for in the case of silver no nitrous oxide is produced notwithstanding the extreme slowness with which the metal dissolves and the richness of the solution after a time in nitrous acid.* Even when cold acid was employed the metal (very thin foil) dissolved with great rapidity much heat being developed so that the mean temperature at which the dissolntion of the metal was effected was considerably higher than that of the acid at the commencement of the experiment. NITRIC ACID. Probably the extent to which nitric and nitrous acids undergo reduc- tion when simultaneouslypresent will be found to be a function of the ‘‘energy ” of the hydrogen displaced by the action of the metal.‘En the Table of Results on p. 75 an experiment with zinc and ammonium nitrate solution is included which may here be briefly mentioned. Zinc dissolves readily in a concentrated solution of this salt especially if ammonia be added much heat being developed ; if a moderate amount of metal be taken and the solution be carefully cooled no gas is evolved ; but on heating the solution a mixture of nitrous oxide and nitrogen is given off the former increasing in amount with the quantity of metal dissolved. When after repeated heating gas is no longer given off the whole of the nitrite and (?) nitroside appears to have been decomposed as the solution is im- mediately coloured by permanganate even when acidulated and as the whole of the metal is not accounted for in the form of these gases it is probable that ammonia is also produced.It is proposed to ex- tend these experiments and also to observe the action of other metals on ammonium nitrate as well as on other nitrates. It should here be stated that M. Saint-Claire Derille has pub-lished a most valuable account of a number of experiments on the action of zinc on very dilute solutions of nitric acid (solutions con-taining an amount of acid corresponding to from 2 to 20 grams of the anhydride per litre) and also on mixtures of nitric and hydrochlo- ric acids and of nitric and sulphuric acids (De 1’Btat naissant Conzpt. rend- 1870 lxx 22 550). M. Deville however appears to regard the various products as formed from the acid by the direct action of the metal a view which is iir opposition to our own.One of us pro-poses 60 discuss this subject on a future occasion. § VIII. Action. of Oadmium and Magnesium on Nitric Acid.-The results of our experiments with these metals are exhibited in Tables v and TI. It will be observed that with cadmium considerably more nitric oxide and less nitrous oxide was obtained than on dissolving zinc in acid of She same strength. There is a considerable difference between the two experiments in the amounts of nitrogen produced but it will be noticed that less nitrous oxide and more nitric oxide was obtained in that which yielded the larger amount of nitrogen and it is possible therefore that the reduction had proceeded further in this case. The difference may be at least in part due to the fact that a larger amount of the metal was employed in the second experiment.The behaviour of magnesium as compared with that of cadmium and zinc presents several points of interest. Much less nitric oxide was obtained with this metal than with either of the latter neither of which however furnished so large an amount of nitrous oxide ; and V. Action of Cadmima OIL hTitric Acid. Percentage composition C.C. of gas per unit-weight of Weight Strength G.C. of of gas. metal. Per cent. No. of of metal T. of acid gas of metal expt. taken. used. obtaincd. accounted for. NO. N,O. Total. NO. N,0. 21.14 67 *53 28 -53 3 -94 13 -91 96 -51 1*38 2 -11 xm..... . -3655 13" 1:2 35 -05 79 *oo 17.74 3 .26 10740 8485 1905 350 98 *99 28 *13 72.27 25 -88 1-85 9 .oo 96.95 2 -45 '60 XLII ....*416 11" 1:2 37 -13 78 -21 20.17 1*62 9996 '7818 2015 163 92 *30 4'7.22 6 '37 69-22 24 -41 13*82 26.40 59 '31 14 *29 6 *77 79.94 12 *87 7.19 XLIII.. . . -3085 12" 1:2 67 *81 17 -8'7 61 -55 20.58 5275 94!4 3246 1085 88 '81 40 *52 62 *10 26 -96 25 '34 57 *32 21 '40 .3.84! 5 90 19.65 XLIV a. . . *314 13" 1:2 69 90 17 -38 58 -09 24 *53 5327 926 3094 1307 90 '93 - NITRIC ACID. the amount of nitrogen produced is far in excess of that ohtained with either zinc or cadmium ; and there can therefore be little doubt that magnesium is a far more "active " metal than either zinc or cad-mium ; but the production of so large an amount of nitrogen appears somewhat remarkable and tends perhaps to throw some doubt on the sufficiency of our explanation of the mode of formation of the nitro- gen as certain of our experiments with zinc (XXXVII and XXXVIII) show that the influence exercised even by a relatively very large quantity of an ammonium salt in increasing the production of nitrogen is by no means great.As with cadmium there is a noticeable difference between the amounts of nitrous oxide and nitrogen obtained in the two experi-ments ; and in the ca'se of both metals the lesser quantity of nitrous oxide corresponds with the greater quantity of nitrogen a lower per-centage of metal also being accounted for in the experiments which yield the smaller quantity of nitrogen. This may be in part due t'o the fact that the heating &c.was more often repeated in the one case than in the other. There is however internal evidence especially in the case of magnesium tending to show that this explanation is in- sufficient; for it will be observed that in the one experiment with this metal the portion of gas last collected-that given off entirely on heating-was far richer in nitrous oxide and poorer in nitrogen than in the other. A similar difference may be observed on comparing the results of two of the experiments with zinc (XXXII and XXXIII). Thus the gas given off on heating contained in the one case only 5.95 per cent. of nitrous oxide but in the other no less than 19.54 per cent. the only difference bet'ween the two experiments being that the amounts of metal taken were not the same (*386and 1.6 grams).The results of the two experiments with cadmium also differ in a similar but mu'ch less marked manner. If as we have previously suggested the decomposition of nitrosidic acid into nitrous oxide and vater is immediate and the nitrous oxide given off on heating is due to the de- composition of hydroxylamine the differences here referred to indicate apparently that the extent to which the various reactions take place may be considerably affected by what seem to be very slight altera- tions in the conditions of experiment. Lastly we may point out as noteworthy that the percentage of metal accounted for in the form of gaseous products is far higher with both cadmium and magnesium than with zinc.It is scarcely possible at present to give any satisfactory explanation of this especially as zinc would appear to be intermediate in its properties between cadmiurn and magnesium. § IX. Action of Iron on Nitric Acid-The gases obtained on dip- VII. Actiota of Irou of? Nitric Acid. Percentage composi tioii C.C.of gas per unit-weight of Weight Strength C.C. of of gas. metal. Per cent. No. of of metal T. of acid p of metal expt. taken. used. obtained. accounted for. NO. N,O. Total. NO. &,O. N,. 40 *79 95 *77 1.71 2 52 13.43 79 -29 -20.71 XLV. .... *251 16" 1:l 54-22 91 -68 1.28 '7 -04 8064 7392 104 567 64:23 35 -70 83 *35 11-21 5 *44 20 -38 97 *49 *30 2.48 7 -78 90 -74 -9 .26 XLVI a. 43975 13" 1:2 63 -86 88 -77 6 *26 4 -97 5997 5323 375 299 49 -18 a.41 *33 83 -24 12 -17 4.59 6 -58 93 -76 4.80 1-44 9 *21 94.52 1-48 4 *OO XLVII ... *400 10" 1:4 86 '27 9 .5Y 4.14 5330 4598 551 221 45 '72 86 *11 10-29 3 -60 95 *37 1-98 2 %5 XLVIII .. '379 13" 1:4 87 -50 9 -02 3-48 5456 4773 492 191 45-15 92 -36 4 *58 3 *06 97 *15 *99 1-86 XLIX .... *318 12O 1:8 93 *87 3 '43 2.70 5904 5542 202 160 443-44 L........ ,200 9" 1 12 91 *28 4.45 4.27 5638 5146 232 242 44 '51 NITRIC ACID. solving this metal" in nitric acid &c. consist chiefly of nitric oxide the amounts of nitrous oxide and nitrogen evolved being reluti vely small. It will be observed however that as the concentration of the acid diminishes the amount of nitrous oxide produced per unit-weight of metal (two-thirds of 56 grams) at first increases and afterwards diminishes ; whereas the nitric oxide at first considerably diminishes in amount as the concentration of the acid sinks hut at a'certain point increases as the concentration is still further reduced.These differences in the amounts of nitric and nitrous oxides evolred with different strength of acid correspond with perceptible differ- ences in the bchaviour of the metal,.during dissolution. In 1and 11 1 2 acid the iron dissolves very rapidly producing it. pale greenish- yellow coloared liquid; .in 1 4 acid however it dissolves somewha,t less rapidly and the solution is of a dark brown colour as also is that produced'onjdissolvingin1:-8and 1 12 acid which the latter espe- cially ac4 upon it still more slowly.From this there can be no doubt that by the action of the more concentrated acid the iron is imme- diately converted into ferric salt whereas1 when weaker acid is em-ployed ferrous salt-is first formed and is converted into ferric salt only on afterwards heating the liquid ;? and it would appear probable that the hydrogen displaced in the formation of the ferrous salt is niore ''active '' than that displaced in the formation of the ferric salt. The manner in which the nitrous oxide is evolved appears also to indicate tliat the hydrogen displaced from weaker solutions of nitric acid is more active in effecting reduction. Thus it will be noticed that in Experiment XLVI with 1 1acid the last portion of gas col- lected-that evolved on heating-is free from nitrous oxide which is also the case in Experiment XLVII with 1':2 acid; in this latter experiment; moreover tlie second portion of gas collected-that ex-tribeted by continuing the action of the pump after practically the whole of the gas evolved during the dissolution of the metal had been removed from the appamtus-is almost free from nitrous oxide.A second experiment with 1 2 acid not quoted here gave precisely similar.results. In the case of the weaker acid howcver the gas evolved on heating contains a considerable proportion of nitrous oxide which as we have before suggested is probably derived from the decomposition of hydroxylamine whilst that immediately evolved without the appli- cation of heat is probably in great -part; if 2 not entirely the product of the decomposition of nitrosidic acid.$ X. Action of Nickel Cobalt Indium and Aluminium on Nitric Acid. * The iron employed was tt portion of the pure iron prepared by the late Dr. Matthiesen for which weare indebted to Dr. Russell F.R.S. + A mixture of ferrous an'd ferric salt is doubtless produced with certain strengths of acid in proportions varying with the concentration of the acid. VOL. XXXII. G 82 ACWORTH AND ARMSTRONG ON THE REDUCTION OF -The results obtained with these metals are represented in Table VITI; although few in number they appear to us to possess very considerable interest. NickeL-This metal would seem to be far more “active ” than iron as it yields a much smaller amount of nitric oxide and a considerably larger amount of nitrous oxide ;the low percentage of metal accounted for is probably to be regarded as evidence of the production of a con- siderable amount of ammonia and affords additional proof of the “activity ” of nickel.The amounts of nitric and nitrous oxides pro- duced in the two experiments differ considerably but different amounts of metal were diasolved and here again it will be noticed that the lesser amount of nitric oxide and the greater amonnt of nitrous oxide is obtained from the larger amount of metal which may be accounted for as we have before said if we suppose that the nitrous acid pro-duced in the first instance undergoes reduction side by side with the nitric acid. Nickel at least in the form in which we have employed it,* dissolves with extreme slowness in nitric acid the time occupied in effecting the dissolution of the quantities taken in Experiments LI and LII having been upwards of 50 and 70 hours respectively.* I am indebted to Dr. Russell for the nickel and cobalt used in these ex-periments. They were portions of the pure metals prepared for his researches on the atomic weights of these elements (this Journal xvi 51 ; xxii 294). The nickel was in the form of hard dull greyish lumps. On analyzing the gaseous mixture evolved on dissolving it in nitric acid it was found to contain hydrogen of which 1.4’7C.C. was obtained in Expt. LI and 2.28 C.C. in Expt. LII or 625 C.C. end 617 C.C. respectively from 100 grams of nickel. As in the course of this inrestigation the production of hydrogen has never been observed on dissolving nietals in 1:2 acid or indeed in nitric acid of any strength and as moreover no hydrogen was ob- tained on dissolving fused commercial nickel (Expt.LIII),there can be no doubt that the hydrogen had been occluded by the metal during its preparation from the oxide by reduction in hydrogen. This fact that occluded hydrogen is without action on nitric acid is however of very considerable interest as it may be possible by the action of nitric acid to ascertain whether as has been suggested a portion of the hydrogen occluded by palladium is combined with the metal. I also propose to apply the method to the study of the gases in meteorites and it may here be remarked that this discovery of the power of nickel to occlude a very considerable volume of hydrogen explains the occurrence of such large amounts of this gas in meteorites which previously on account of the slight power of occluding hydrogen which iron alone appeared to possess was somewhat difficult of explanation.No hydrogen was obtained from the cobalt although it had also been prepared from the oxide by reduction in hydrogen. But the piece dissolved was the only one in the bottle from which it was taken preserving a metallic appearance the remainder of the specimm had assumed a brown friable condition having evi- dently undergone oxidation. I am inclined to believe from this that the cobalt had originally been charged with hydrogen with the exception of the piece used by us which was extremely dense and compact.The spontaneous oxidation of metals prepared by reduction in hydrogen is in fact very probably always due to the presence of occluded hydrogen.-H. E. A. VIII. Percentage Composition C.C. of gas per unit-weight of Strength C.C. of gas. metal. Per cent. of No. of Weight of metal T. of acid of gas metal expt taken. used. obt,ained. Nickel. LI ...... *2356 13" 1 2 7-76 5.37 83.31 11.32 1935 104 1612 219 34 -52 LII ..... I 13-38 I 2-70 87'89 9.41 2127 1869 200 38 *36 LIII.. .. i:' 1 iii 1 -1'86 1 82.35 I 15.79 1 -I -58 I -1 -1 -Cobnlt. LIV .... 1 *4295 I 18' I 1 2 I 7.98 I 5-71 I 79-23 I 15 '06 I 1090 63 ! 863 I 164 I 19-55 Indium. LV ..... I -265 1 15" I 1:2 I 29.43 I 90'57 1 4.49 I 4.94 I 8395 7603 I 376 I 416 I 67-13 A1 uniinburn.ACWORTH AND ARMSTRONG ON THE REDUCTION OF Cobalt.-This metal like nickel dissolves with extreme slowness in 1:2 acid. From the one experiment we have made it would appear to be as much superior in "activity " to the allied metal nickel as this latter is to iron. Irdium.-A small quantity of this metal was kindly presented to us by ,Dr. Gl ads t one F.R.S. It is acted upon with extreme slowness by 1 2 acid the time occupied in dissolving the *265gram employed in the experiment of which the results are exhibited in Table VIII having been upwards of 70 hours. It appears to be possessed of con-siderable "activity," as from the comparatively low percentage of metal accounted for and the comparatively large amount of nitrogen obtained it is probable that a considerable amount of ammonia is formed.Alzcminiurn.-This metal was so slowly acted upon at the ordinary temperature that we found it necessary to heat the acid to about 60"-65O when it dissolved with moderate rapidity. It appears to resemble silver in its action on nitric acid and most probably it furnishes only nitric oxide. $ XI. Action of Tin Lead and Thallimz om Nitric; Acid.-The results of our experiments with these three metals are given in Table IX. Tin.-In the experiments with 1:0 and 1 I acid white (?) meta-stannic acid separated out as the metal dissolved but in the remaining experiments clear solutions were obtained from which no separation took place until they were heated when however the liquid became almost semi-solid.It will be observed that not only the nitric oxide but also the nitrous oxide and nitrogen increase in amount and then diminish as the concentration of the acid diminishes and it is probable therefore that as in the case of iron there is a difference in the '' activity " of the hydrogen according as the product is a stannic or a stannous salt. Judging from the composition of the gaseous mixture and the quantities of gas obtained before and after heating when 1 2 acid is employed (Experiment LIX) it appears probable that in the dissolu- tion of the metal to (?) stannous salt it relatively small amount of nitrous acid but considerable amounts of nitrosidic acid and ammonia are produced ; and that in the conversion of the stannous into sta,nnic salt which takes place on heating chiefly those reactions occur which lead to the production of nitrous oxide.Lead.-The most noticeable feature with regard to the action of this metal on nitric acid is that comparatively large amounts of nitrous oxide and nitrogen are produced and as a very considerable proportion of the metal is unaccounted for it is very probable that much ammonia IX. Percentage composition C.C. of gas per unit-weight of Weight Strength C.C. of gas. No. of of metal T. of acid of gas metal. Per cent. of metal expt. taken. used. obtained. I I I accounted for. ~ ~ I NO. NsO. N,. Tota,l. NO. N20. N2. Ti%. LYII .,.. -4135 14" 1:0 21 '54 1'08 85 -14 13-78 3073 34 2616 423 56.58 LTIlI ...'414 15" 1:l 31-02 16 -38 73 *82 9 *80 4420 724 3263 433 73 -03 5 -32 2 '57 79 -17 18 *26 29 -85 25 *93 65 *77 8 -30 3 LIX.. * ... ,404 11" 1:2 35 -17 23 '37 67 "78 9 -85 5136 1148 3481 506 81-20 LX .... *4145 12O 1:2 33 '09 14 '47 75 -55 9.98 4710 781 355s 470 79 -52 b-a. LXI ..* * -415 11" 1:s 4'66 3 *27 85 *02 11*so 3312 106 2816 390 71 -76 Lead. LXII .. . . 0381 14" 1:2 10 -56 51 -23 41 -47 7 *30 5737 2939 2379 419 71*76 5 -06 2-45 81-90 15.65 64 *85 4 -82 -41-89 3 '26 I_-LXIII ... . '403 11" 1:2 0 -88 28 -06 62 '37 9 *57 5074 1423 3164 487 77 -16 2.88 34-92 88'66 26'42 4.04 90.38 7.72 1-90 ~~~ bo LXIV .... -3065 30" 1 2 6'42 69*'78 19'15 11.07 8646 5963 1636 94'7 90 -59 'a 86 ACWORTH AND ARMSTRONG ON THE REDUCTION OF remains in solution.From the composition of the gases collected be-fore and after heating (Experiment LXIII) it would seem that much of the nitrous oxide is derived from the decomposition of hydroxyl-amine. ThnZZiwm-Although much more nitric oxide and considerably less nitrous oxide is furnished by this metal than by lead the amount of nitrogen produced is innch greater ; but as the percentage of metal accounted for is high the results we have obtained do not appear to warrant the conclusion that thallium is a more ‘‘active ” metal than lead in respect of its action on nitric acid. fi XII. Action of AZloys on Nitric Acid.-As yet we have only snb-mitted two alloys to examination vix. brass and gun metal; the results obtained are exhibited in Table X.X. Percentage composi- C.C. of gas per unit- tion of gas. weight of metal. T. I N2. I Total. NO. N20. N2. 1 Brass .... -374 15’ 1:2 1.34 1277312392 210 171 90 -74 7-04 4.31 Gun metal. -317 16” 1:2 45.19 1 61 93 32.41 5.66 9038 5597 2929 512 101*57 --.1--Our object in making these experiments was if possible to throw further light on and to obtain some explanation of the differences observed in the behaviour of various metals towards nitric acid. Thermochemical investigation has shown that as a general rule a considerably greater amount of heat is developed in the formation of tl-e magnesium compounds than in the formation of‘the corresponding compounds of either zinc or cadmium and more in the formation of those of zinc than of those of cadmium ; and as from the composition of the gases obtained on dissolving these metals in nitric acid it would appear that the first-mentioned is the most and the last-mentioned the least “active,” we are inclined to believe that the extent to which reduction takes place on dissolving a given metal in nitric acid is R function of the energy developed in the reaction R’ + nHN03 & Aq = nR’N03 =I= Aq + nH i.e.in the displacement of the hydrogen ~lfthe acid by the metal including that developed subsequently by the -1-1-1-1- NITRIC ACID. action of the water present on the salt which is produced. This how- ever we regard at present merely as a working hypothesis for it must not be forgotten that we have not yet suflicient data to enable us to decide absolntely as to which of two related metals is possessed of the greater " activity," inasmuch as only the qaseous products of reduction have been determined ;this being the case even with regard to magnesium and zinc as although less nitric oxide and considerably larger quantities of nitrous oxide and nitrogen are furnished by mag-nesium than by zinc the percentage of metal account3ed for in the form of gaseous products is considerably less in the case of the latter.Moyeover in many instances at least reaction takes place not only between the metal and nitric acid but also between the former and the nitrous acid which is the first product of seduction of the latter ; and that this may have an important influence on the final result and may much complicate the discussion of the phenomena conccmed is evident from the fact that platinum for example will dissolve in nitrous but not in nitric acid; and that silver if not insoluble in nitric acid is certainly much more readily dissolved by nitrous acid.It will be remembered also that the amount of metal employed is not always without influence on the result and that we have suggested that this is a consequence of the accumulation of nitrous acid in the solution. Roscoe however has shown that zinc and cadmium are con-siderably more " active " than magnesium in reducing the vanadic compounds a solution of the pentoxide in sulphuric acid being reduced to a salt of the tetroxide by sulphurous acid and hydrogen sulphide ;to a salt of the trioxide by magnesium ; and to salt of the dioxide by zinc cadmium and sodium amalgam; and it may be said that these observations are in opposition to our conclusions as to the relative " activity " of magnesium zinc and cadmium.But the dis- cussion of the phenomena involved in these changes is doubtless also complicated by the occurrence of a series of reactions ; so that in fact until we are acquainted with the entire series and are able to deduce the dynamic values corresponding to the several reactions we must refrain from drawing conclusions as to the reason of the superior activity of zinc and cadmium in reduciiig vanadium pentoxide and of the superiority of magnesium in reducing nitric acid.Moreover if the activity of a given metal depend as we suppose on the amount of "available " energy associated with the hydrogen dis- placed by the metal from the acid our present knowledge would lead us to expect that the same metal may produce very different results with different acids for the reason that probably not only are different amounts of heat developed by the mere displacement of the hydrogen in different acids by the same metal but the compounds formed by 88 ACWORTH AND ARMSTRONG ON TEE REDUCTION OF the displacement of the hydrogen of the acid by the metal doubtless enter into reaction with the water present and in other ways and probably very different amounts of heat are developed in these reac- tions by different salts of the same metal.For the same reason we may expect also that alterations in the conditions of experiment viz. in the degree of concentration of the acid and of the temperature at which the metal is dissolved will perceptibl3 influence the result in many cases. The rapidity with which the metal dissolves may also influence the result; there can be little doubt of this in the case of copper as we have pointed out and Deville has shown (Cotnpt. rend. lxx 20 550) that a relatively much larger arnount of nitric acid is required to prevmt the evolution of hydrogen when zinc is dissolved in a mixture of nitric and hydrochloric acids than when it is dissolved in a mixture of nitric and sulphuric acids a result which he attributes in great measure to the fact that the zinc is dissolved more rapidly by the hydrochloric than by the sulphuric acid.It is well known that in the formation of many alloys from their constituent metals much heat is developed. From this it follows that the energy of the associated metals must be reduced and it appeared to us that if our hypothesis above stated were correct the comparison of the behaviour of a given alloy with that of its constituent metals towards nitric acid should afford evidence of this :that in fact reduc- tion of the acid would take place to a less extent on dissolving a given amount of the alloy4han on dissolving corresponding amounts of its constituent metals. Only two experiments have as yet been made one with brass andl the other with gun metal.The brass employed contained 30.2 per cent. zinc 69-1 per cent. copper and -7per cent. lead and tin ; the gun metal 10.90 per cent. tin 88.5 per cent. copper and a little lead. It will be seen on reference to hhe table that the brass behaves exactly as copper the percentage of nitrous oxide and nitrogen being no greater than with copper and the amount of nitric oxide per unit- weight of metal (regarded as copper) is as high as was obtained in most of the experiments with coppe? alone under similar conditions. So far as its behaviour towards nitric acid is concerned the zinc may indeed be said truly to have undergone transmutation into copper and this result appears to us to afford much support to the hypothesis we have advanced in explanation of the difference in’the behaviour of diffcrent metals.The result oMained with gun metal is more difficult of interpreta-tion ;it would seem however from the large amount of nitrous oxide produced and the manner in which it isevolved-chiefly ic the cold- that whilst the “activity”’ of the tin is reduced by its association with copper that of the copper on the other hand is considerably NITRIC ACID. raised by its association with tin ; but whether the gain to the copper is less than the loss to the tin it is impossible to decide from this one experiment. A most interesting series of observations on the action of acids on alloys of zinc and copper and of tin and copper in various proportions were published some years ago in this Journal by Messrs.Calvert and Johnson (1866 xix 434) which appear to us further to support our hypothesis. The method they adopted was to determine the loss experienced in a given time .by a given surface of the alloy exposed to the action of a given quantity of the acid at a known temperature and the results obtained in this way indicate that the irifluence of a metal such as copper in reducing the “activity ” of metals such as zinc and tin is very considerable. Thus whereas an alloy of zinc and copper containing 67.26 per cent. of zinc and 52.74 per cent. of copper lost 2550 grams per square metre of surface when exposed to tlie action of nitric acid of sp. gr. 1.10 an alloy containing 50.95 per cent. of zinc and 49.05 per cent. of copper lost only 45 grams per square metre of surface ; and the loss experienced by allop stil,l richer in copper was even less.Again whereas an alloy of tin and copper in tlie pro- portion corresponding to the formula Sn,Cu containing 90.27 per cent. of tin and 9-73 per cent,. of copper lost 1883 grams per square metre when exposed to the action of nitric acid of sp. gr. 1.25 the alloy represented by the farmula SnCu containing 65.02 per cent. of tin and 34.98per cent. of copper lost only 183grams per square metre ; alloys containing a larger proportion of copper were found to be some- what more acted upon that of the formula SnCu6losing 808 grams per square metre. Although much importaat information has been gained as to the behaviour af a large number of the metals with nitric acid the results narrated in the previous pages are chiefly of value as indicating the directions in which the investigation must be continued in order to obtain the requisite data for the complete discussion of the problem under consideration.In the first place it will be necessary to study the action upon each other in presence of nitric acid of all the possible reduction products- nitrous acid nitiwsidic acid hydroxylamine ammonia-under con-ditions as similar as possible to those which obtain when the metals are dissolved in nitric acid. Then the action of metals on nitric acid must be again submitted to examination and both the gaseous products arid those remaining in solution determined ; and as very slight variations in the conditiotis of experiment often appear to exert an appreciable influence on the final result it will be iiecessary to determine the two series of products wherever possible in one and the same experiment.PERKIN ON DIBROMACETIC AND GLYOXYLIC ACIDS. In the case of those metals which furnish two series of salts it will he requisite to examine separately the action of the ous salts-of tlie ferrous and stannous salts for example as these may behave veiy differently from the metals themselves. Lastly a series of experi-ments on the action of various metals on the same mixture of nitric and hydrochloric or sulphuric acids will doubtless throw much ligli t on the question of the diEerence in behaviour of the different metals. I hope very shortly to be able to give an account to the Society of the results obtained in continuation of this investigation.-H.E. A.

ISSN:0368-1769    

DOI:10.1039/JS8773200054

出版商:RSC    

年代:1877

数据来源: RSC

摘要:

PERKIN ON DIBROMACETIC AND GLYOXYLIC ACIDS. 1X.-On Dibronzacetic and Glyoxylic Acids. By W. H. PERKIN, F.R.S. SEVERAL years since (Jou,vn. Chenz.. Xoc. xi 22) in conjunction with the late M:r. Duppa I gave an account of bromo- and dibromo-acetic acids and some of their derivatives ; amongst these were the products obtained by the decomposition of their silver salts when heated with water that of the former acid yielding silver bromide and glycollic acid that of the latter silver bromide and an acid which we believed to be bromo-glycollic acid. Our reasons for believing this to be bromoglycollic acid were various ; one was from the apparent analogy of the reaction by which it is formed from silver dibromacetate to that by which glycollic acid is produced from silver bromacetate thus- C2H2BrAg02+ OHz = C,H403 + AgBr.Silver bromacetate. Glycollic acid. C2HBr2Ag02+ OHz = C2H3Br03+ AgBr. Silver dibromacetate. Bromoglycollic acid. A second reason was because its silver salt when boiled with water yielded silver bromide and glyoxylic acid apparently thus- C2H2BrAg03+ H20 = C2H,04 + AgBr. Silver bromoglycollate. Glyoxrlic acid. From the difficulty of preparing dibromacetic acid in large quanti- ties at that time we were prevented from investigating this matter very fully; nor did we consider it important to do so as we had no reason to doubt the correctness of the conclusions we had arrived at. Since then new processes for the preparation of bromo-and dibrom-acetic acids have been discovered by which they can be obtained PERKIN ON DIBROMACETE AKD GLYOXYLIC ACIDS.more easily; I therefore thought it would be interesting to make a more complete examination of this so-called bromoglycollic acid. The process adopted for the preparation of dibromacetic acid con- sisted simply in the use of acetic anhydride in the place of glacial acetic acid for the production of the broma.cetic acid required; the anhydride being readily acted upon by bromine the use of sealed tubes is thus avoided. The bromacetic acid so obtained was con-verted into the dibromo acid by treatment with bromine in sunshine whilst heated to its boiling point as previously described. Consider-able quantities were thus prepared and the acid was usually in the crystalline state.A portion converted into the silver salt gave the following results on analysis :-*6695of substance gave -3833 of AgBr = 32.87 p. c. Ag. Theory for C2HBr2Ag02requires 33.23 p. c. Ag. Decomposition of Silver Dibromacetate whew heated with Water. Having obtained a supply of dibromacetic acid a considerable quan- tity of its silver salt was prepared and a portion of it boiled with water until silver bromide ceased to be formed. The decomposition takes place with considerable energy when quantities of about 20 grams of silver salt are used the ebullition continuing for some time after the removal of the source of heat. The resulting dilute acid liquid was separated from silver bromide by filtration concentrated over the water-bath nearly neutralised with sodium carbonate and converted into the silver salt by precipitation with silver nitrate.It was then washed and dried in a vacuum. A silver determination gave the following numbers :-*2447of substance gave -177 of silver bromide = 41.55 p. c. Ag. The calculated percentage for silver bromoglycollate C2H2BrAg0, is 41.22 p. c. Ag. This result appeared to confirm the correctness of the equation already given ; but by experimenting under different conditions very variable results were obtained. For example when dilute solutions of the acid product obtained by boiling silver dibromacetate with water were used for the preparation of a silver salt the resulting compound always contained a smaller percentage of silver than when strong and neutralised ones were employed.To get a clue to this unexpected result fractional precipitation was resorted to which soon showed that the product was not a definite body but a mixture the first crops giving low and the last ones high PERKIN ON DIBROMACETIC AND GLYOXYLIC ACIDS. percentages of silver. The following is an example selected from several experiments :-First crop contained 33.06 p. c. silver. .,f Third 45.86 , It was soon remarked that the first crops often gave percentages of silver closely corresponding with those required by silver dibrom-acetate viz, 33-23 p. c. and on examinirg them they were found to consist of that substance. Having established this curious fact it was necessary to find out the nature of the product which accompanied the dibromacetic acid in this decomposition.After making several experiments it was believed to be glyoxylic acid. To determine this about 18 grams of dibromacetic acid were made into its silver salt and deccmposed by boiling with water and the re- sulting acid solution was neutralised with calcium carbonate and fil-tered. On cooling the solution was inclined to gelatinise but on leaving it in a vessel surrounded by hot sand so as to cause it to cool slowly it deposit'ed about 3 grams of crystals and when further con- centrated a further small quantity was deposited. These were un- doubtedly caZcimb glyoxylate. Nevertheless a calcium determination was made. 0292of substance gave -1795 of CaS04 = 18.07 p. c. calcium.The formula C4H,Ca"0 requires 18.0 p. c. The mother-liquors from these crystals contained nothing but cal- cium dibromacetate. From these result9 it is evident that silver dibromacetate when boiled with water yields only dibromacetic and glyoxylic acids thus-2(GHBr2Ag02)+ 2H20 = C2H2Br202 + C,H40a + 2AgBr. Silver dibromacotate. Dibromacetic acid. Qlyoxylic acid. As this decomposition is remarkable it was thought desirable to see if the products were really produced in the ratios indicated by the above equation. For this pnrpose recourse was had to the well known decomposition of glyoxylic acid into oxah and glycollic acids when treated with calcium hydrate it having been found previously from experiment that dibromacetic acid did not yield oxalic acid under these circumstances.The following quantitative experiment was made :-4.586 grams of silver dibromacetate were boiled until silver bromide ceased to form and the solution after filtration mixed with an excess of pure calcium hydrate. It was then heated until it boiled and afterwards acidified with acetic acid. The calcium oxalate which had PERKIN ON DIBROMACETIC AND GLYOXYLIC ACIDS. been produced was collected washed ignited &c. In this manner -363 gram of calcium carbonate was obtained equivalent to 7.1 per cent. of oxalic acid from the silver dibromacetat,e used. Xow as two molecules of glyoxylic acid are required to produce one of oxalic acid iti follows from the above equation that four molecules of silver dibromacetate would be necessary to produce one of oxalic acid.This if calculated is found to be in the proportion of 100 parts of silver salt to 6.9 parts of oxalic acid which closely corre- sponds with the above experiment and therefore proves the correct- ness of the equation. These experiments show how the mistake oceurred in supposing that bromogly~ollic acid resulted on boiling silver dibromacetate with water because a mixture of dibromacetic acid md glyoxylic acid in the proportions produced in this reaction has the composition of that substance. C H2Br2 O2 CzH4 04 C4H6Br206 = 2( C2H,Rr0,) Bromoglycollic acid. It is difficult to understand the manner in which silver Zbrom-acetate yields these products when boiled with water. It may perhaps be that bromoglycollic acid is first formed as the salt is decomposing and being nnstable is immediately acted upon by the undecompssed portions of the silver dibromacekate thus :-I.C2HBnAgO2+ HzO= C2H,Br03+ AgBr. 11. C2H3Br03+ C2HBr2Ag0,+ H20= C2H404 + C2H,Br20 + AgBr. Decompmition of Silver Dibmmacetate when heated with Alcohol. Perfectly dry silver dibromacetate when heated to 100" C. with absoltite alcohol quickly decomposes with formation of silver bro- mide. On filtering this off and allowing the alcoholic fluid to evapo-rate gently a product is obtained which consists of an ethereal liquid and an acid. The latter was removed from the oily product by means of water ad when converted into a silver salt gave the following numbers :-*0979of substance gave -0574 of AgBr = 32-68 p.c. silver. showing it to be dibromacetic acid. The ethereal product was distilled and was found to boil at about 193-195". It was analysed but no definite results were obtained. It contained from 32 to 33 p. c. bromine. It is probably a mixture of dibromacetic and diethylglyoxylic ether for when treatcd with PERKIN ON DIBROMACETIC AND GLYOXYLLC ACIDS. ammonia it yields dibromacetamide and on evaporating the mother- liquors from this brilliant plates like diethylglyoxylamide are ob-tained. Decomposition of Xiluer Dibromacetate when heated with Dry Eth.ey-. When heated to 100' C. in a sealed tube with ether the perfectly dry salt decomposes with formation of silver bromide a small quantity of carbonic acid being also produced.On filtering off the ethereal fluid and evaporating a thick oily product is left behind. It ap-pears to be impossible to obtain this product in a perfectly pure state as it cannot be distilled without decomposition. It is generally con- taminated with a small quantity of a substance having a very irri- tating odour ; this is removed partially by passing dry air through it whilst heated to 100" C. In order to purify it further it was redis- solved in ether and kept over dry sodium carbonate for several hours filtered the excess of ether distilled off and further separated by passing dry air through the oily product whilst heated in the water- bath. On analysis it did not give very satisfactory numbers. The following are the best obtained:- I.*2950of substance gave .4017 of AgBr. 11. -3495 of substance gave -2293 of COz and -0558 of H20. The formula C4H2Br204 requires the following percentages :-Calculated. Found. Cq H2 = = 48 2 17.72 -73 17.89 1.77 Br = 160 58.39 67.94 O4 = 64 23-36 - 274 100*00 The above formula represents this substance as formed by the de-composition of two molecules of silver dibromacetate with separation of two molecules of silver bromide. 2(CzHBr2AgOa) = C4H2Br204+ 2AgBr. This formula is also confirmed by the products of decomposition of this body. If a quantity be mixed with alcohol and a little sulphuric acid on dilution an oil separates having an irritating odour which when treated with ammonia @Tea dibromacetamide showing it to be dibromacetic ether.PERKIN ON DIBROM-4CETIC AND GLYOXYLIC ACIDS. If left in contact with water it gradually dissolves forming an acid solution. On fractionally precipitating this with silver nitrate the first crop gave the following numbers.:- -29.26of substamce gave -1765 of AgBr = 34.68 p. c. silver evidently silver dibromacetate containing a small quantity of gly-oxy lat e. Another portion was dissolved in water and the solution neutralized with calcium carbonate ; after concentration a crop of crystals of cal-cium glyoxylate separated. A calcium determination of this gave the following results :--0424 of substance gave -0262 of Ca804 = 18.18 p. c. calcium. calcium glyoxylate requires 18.01p.c. Therefore this oily substance when decomposed with water is resolved into dibromacetic and glyoxylic acids thus :-+ 2H20= C2H2Br20z CaH2Br20* + C2H404. When distilled it is decomposed yielding carbonic oxide and dibro- mncetic acid ; a quantity of a giimmy product is also formed somewhat like tartaric anhydride. The first part of this decomposition may be represented thus :-C4H,BrZO4 = 2CO + C2H2Br2O2. The formation of this substance from silver dibromacetate is difficult to understand ; half only of the dibromacetyl contained in the silver dibromacetate used being decomposed the rest remaining unchanged ; thus :-COAg 0 I CH Br COAg 0 I CH Br2 This curious reaction may perhaps be explained if we assume that a bromoglycollide is first formed and as quickly as it is produced reacts upon the undecomposed silver dibromacetate thus ;-I.C2HBr2Ag02 = C2MBr0 + AgBr. Silver dibromacetate. Bromoglycollide. 11. C2HBr02+ C2HBrzAgOz= CaH2Br204-t AgBr. New product. PERKIN ON DIBROMACETIC AND GLPOXYLIC ACIDS. This new product appears to possess the properties of an anhydride and may be written either as a dibromacetoglyoxyllide or as a double anhydride thus :-CO-0-co I I CHBrz CHO. 0(C,RBr 0). Silver dibromacetate undergoes decomposition epontaneonsly hen kept and the decomposed salt when treated with water yields a solu-tion of dibromacetic and glyoxylic acids. Gl!jox:ylic Acid. Being desirous of making a further examination of glyoxylic acid and some of its derivatives it was necessary to prepare a qnantity of this substance and the following process was employed :-Forty or fifty grams of silver dibromacetnte were boiled with water until decomposed and the resulting solution of dibromncebic and gly-oxylic acids filtered from the bromide of silver formed.Carbonate of silver was then carefully added to the clear solution until it was no longer acted upon and the resulting mixture of silver salt boiled until silver bromide ceased to be formed. After this latter substance had been separated by means of a filter the acid licpid was concentrated on the waher-bath. It usually contained a small quantity of silver in solution owing to the addition of a slight excess of silver carbonate in the previous operation.This was not found to be injurious and in fact was useful in removing the last traces of bromine which it sepa- rated as the solu.t;ion concentrated. When the product was reduced to a somewhat small bulk hydrochlbric acid was cerefulIy added just in sufficient quantity to remove the last traces of silver ; and after filtra-tion the solution was further concentrated under a bell jar over sul-phuric acid. In the course of a week or so the liquid which had become syrupy gradually deposited crystal? ; these slowly increased in quantity ; when a considerable crop had formed the thick mother- liquor was decanted and the crystals afterwards pressed strongly between bibulous paper the resulting crystalline cake was then broken up and dried in a vacuum over sulphuric acid.The mother-liquors mere again placed under a bell-ja'r over sulphuric acid and in the course of a week or so became a neady solid mass of crystals. The following combustions were made of the product after drying for fourteen days in a vacuum :-I. -29405 of substance gave -2814 of GOz and ,1149 of H20. PERKIN ON DIBROMACETIC AND GLYOXYLIC ACIDS. 11. -2428 of substance gave -2317 of COz and -0970 of water. These numbers show that the crystals were pure glyoxylic acid having the formula CzH404. Calculated. Found. -/ I. 11. C = 24 26-08' 26.09 26.02 H* = 4 4-34 4.34 4.43 04 = 64 69.58 -92 -100.00 Several preparations were made of this acid and it was always obtained in the cryst8a,lline condition.Pure glyoxylic acid crystallises apparently in oblique rhombic prisms; it is however very difficult to observe the form of the crystals as they are small and rather confused and also produced in a very thick fluid in which it is only possible to view them because when removed they deliquesce very rapidly. This acid is very soluble in water and alcohol. It tastes very like tartaric acid. When heated it melts to a sympp liquid ; it cannot be distilled wihhout decomposi- tion. Its aqueous solution if quickly evaporated over sulphuric acid in a vacuum does not usually crystallise but becomes a gummy mass ; if however the vacuum be dispensed with so that the evaporation takes place slowly crystals are obtained. The process for the preparation of this acid is easily understood by the following equations :-2C2HBr2Ag02+ 2HZ0= C2H404+ C2H2Brz02+ 2SgBr.The resulting mixture of dibromacetic and glyoxylic acids being then converted into silver salts by the silver carbonate decomposes as follows :-C,H,Ag04 + C2HBr2Ag02+ 2H20= 2C2H404+ 2AgBr. A portion of crystallised glyoxylic acid mas heated with absolute alcohol in a sealed tube for five hours at 120"-130" in the hop of obtaining the ether- CO(OC2H5) A{&& The product was a colourless liquid which gave numbers showing it to have the composition- VOL. XXXII. H. PERKIN ON DIBROMACETIC AND QLYOXYLIC ACIDS. c0(0CZH,) The following aye the numbers obtained 1-I. -1983of substance gave- *3900of COz,and *1625 of H20.11. 2929 of substance gave- -5758 of GOz,and '23570f H2O. Calculated. Found. &--I. 11. Cs = 96 54'54 53-63 53.61 Hi = 16 9-09 9-10 8.94 04 = 64 36.37 -176 100*00 Although these results are not so good as could be desired they undoubtedly show that the product is diethylglyoxylate of ethyl any other possible ethylic derivative requiring much lower percentages of carbon. The amount of product at my disposal being small I was unable to purify it so thoroughly as I could have wished ; otherwise I have no doubt better numbers would have been obtained. Sodlimn Glyoxy1nte.-A solution of crystallised glyoxylic acid when neutralised with sodium carbonate and concentrated yields a hard crystalline sodium salt. This when recrystallised and dried at 100" C.gave the following numbers on analysis :-I. -3038 of substance gave- -1928 of N&SO4 = 20.55 p. c. sodium. II. -232 of substance gave- -146 of Na2S04= 20.38 p. c. sodium. Theory for C2H3Na04= 20.17 p. c. Potassium GZyoxyZate.-This salt prepared as the above but substi- tuting potassium for sodium carbonate is obtained as an easily soluble crystalline salt. It cannot be dried at looo,as it swells up and decom- poses at that temperature. Dried in a vacuum it gave the following numbers on analysis :-I. -2658 of substance gave-*1776of K,S04 = 29.95 p. c. potassium. PERKIN ON DIBROMACETIC AND GLYOXYLIC ACIDS. 11. ,2046 of substance gave- ,1375 of K,SO = 30.12 p. c. potassium. 111. -2.36 of substance gave- -155 of K,SO = 29.44 p.c. potassium. Theory for C2H3KOa= 30.0 p. c. Glyoaylic acid ad Rmmonia.-The product obtained by Dr. D ebus by the decomposition of calcium glyoxylate with ammonium oxalate and generally known as ammonium glyoxylate being the compound that most favours the formula CzH,03for glyoxylic acid and C2HROJ,H20 for the glyoxylates is necessarily of considerable interest ; further experiments were therefore made to determine whether or no it be a true salt. I. A quantity of calcium glyoxylate obtainea from the acid pro-duced by oxidizing alcohol with nitric acid was dissolved in water and the theoretical quantity of a neutral solution of ammonium oxalate added both solutions being cold. On filtering off the calcium oxalate a clear neutral fluid was obtained.This was evaporated over sulphuric acid in a vacuum but it was observed that it gradually became acid to test-paper the acidity increasing until crystals separated out. The crystals were washed with cold water but when dissolved in cold water were still found to be decidedly acid. They were dissolved and recrystallised again by evaporation in a vacuum but were still acid to tes t-paper . XI. Another quantity of this product was prepared in the same manner using calcium glyoxylate prepared from dibromacetic acid and the solution before evaporation in a vacuum rendered slightly alkaline with ammonia. The crystalline product was also in this case acid to test-paper 111. A quantity of a solution of crystallised glyoxylic acid was neutmlised slowly with dilute ammonia and evaporated in a vacuum ; the solution gradually became acid and deposited crystals which were acid to test-paper even after recrystallisation.As this derivative of glyoxylic acid dissolves but slowly in cold water its acidity is most readily seen by placicg a little of it in powder on wet blue litmus paper. With ammonia it behaves like an acid inasmuch as it dissolves in it with great facility. The following analyses were made of different specimens of this substance :-*I. -2389 of substance gave- -2311 of COZ. * The hydrogen was lost. 100 PERKIN ON DIBROMACETIC AND GLYOXYLIC ACIDS. 11. -1480 of substance gave- *1416 of CO, and -0753 of H,O. 111. -2115 of substance gave- a2020 of CO and -1082 of HZO.Found. Theory for C2H,N03. I. 11. 111. Carbon ......... 26.37 26-38 26.09 26.04 Hydrogen ...... 5.49 - 5-58 5.68 These numbers correspond with those obtained by Dr. Debus but as they agree almost equally with an acid ammonia salt and also glyoxylic acid it was thought desirable to determine the nitrogen in this body. Dr. Frankland kindly had a determination made for me by the pro-cess he employs for water analysis and the ratio of carbon to nitrogen was found to be practically as 2 1; therefore the above formula is evidently the correct one for this substance. Seeiug that the glyoxylates as those of potassium and sodium are neutral to test-paper and also that the freshly prepared solution of the ammonium salt obtained by double decomposition is also neutral some change must evidently take place in this salt during its evapora- tion in a vacuum over sulphuric acid seeing it becomes acid without loss of nitrogen.This would indicate that the ammonium left the CO(0H) group for some other. At first sight this may appear strange but we must remember that Dr. Debus has already shown that neutral glyoxylates will remove ammonia from its salts. From these considerations I am led to infer that this so-called am- monium glyoxylate is not a true salt but a product resulting from its decomposition. The change which the ammonium salt which is undoubtedly at first formed undergoes may perhaps be as follows :-CO(0NH.J CO(0H) = 1 IH +OHz. C NH OH \OH Ammonium glyoxplate.Amidoglyoxylic acid. This last compound would be a substance analogous to aldehyde-ammonia in which CH is replaced by CO(0H)-CH3 A{f2 Aldehyde-ammonia. PERKIN ON DIBROMACETIC AND GLYOXYLIC ACIDS. 101 A portion of the ammonia derivative of glyoxylic acid was mixed with dilute ammonia in which it is very readily soluble far more so than in water. The solution was evaporated in a vacuum. It dried up to a glassy uncrystalline mass and on analysis gave the following numbers :--200 of substance gave -2040 of CO and * -1147of HZO showing it to contain 27-81per cent. of carbon and 6.37 per cent. of hydrogen. A determination of carbon and nitrogen showed it to contain t'hese elements in the ratio of 4 3.The nearest formula corresponding witlh these numbers is-or a cornpound of amido- and diamido-glyoxylic acid ; but the sub-stance was of too indefinite a nature to form any decided conclusions from The following derivative obtained with aniline and glyoxQcL acid favours the view that the ammonia derivative of this acid has the con- stitution assigned to it namely that it is an amido acid. This sub- stance has already been referred to in the paper by Mr. Duppa arid myself. If instead of using ammonium oxalate in the preparation of the so-called ammonium glyoxylate oxalate of aniline be employed calcium oxalate separates ; and on filtering this off a clear solution is obtained doubtless glyoxylate of aniline. It however quickly changes be- comes yellow and deposits an orange-coloured powder.This pecu- liar substance does not possess the properties of a salt but is both an acid and a base so that the aniline must have left the COOH group. This substance is easily soluble in ammonia and when the solution is evaporated in the water-bath it forms an orange glassy substance which is a soluble ammonia salt. With nitrate of silver a solution of the compound forms an amorphous pale yellow bulky silver salt which can be washed only on a vacuum filter. Several specimens were analysed but the results were not very concordant the percentage of silver generally being between 32 and 36 per cent. It also dissolves in dilute hydrochloric acid and when evaporated to dryness the solution leaves a soluble liydrochloride containing about 20 per cent.chlorine. This gives with platinum tetrachloride ;t plati-num salt containing about 18 per cent. platinum. This peculiar aniline derivative does not crystallise ; a great many analyses of it have been made but no satisfactoryconclusions as to its 102 PERKIN ON DIBROMACETIC AND GLYOXYLIC ACIDS. formula have as yet been obtained. When distilled with strong potas- sium hydrate it decomposes and aniline distils over. From the foregoing results and considerations it appears that un- doubtedly the formula of glyoxylic acid is- CO(0H) A {EH OH and its anhydrous salts- CO(0R) :H OH and therefore it contains two hydroxyls in union with one atom of carbon. The substance- COOH is at preseiit unknown though its existence is of course probable.We have the following lists of substances confirming this view Glyoxylic acid ............... (lzH,04 Sodium glyoxylate. ........... CzH,Nn04 Potassium glyoxylate ......... C2H3K04 37 Calcium ......... C4H,Ca08 Silver 77 ......... CzHsAgOa EthyldiethylglFoxylate ........ CZH(C2H5)3O4 Diet hylglyoxylamide .......... C2H3( C2H5) ZN03. The compound of sodium bisulphite and glyoxylic acid and the ammonia derivative might also be added The bisulphite of sodium cornpound- CO( ONa) C NaSO3 '.{F Ammonia compoqnd- CO.OH It is very possible that mesoxalic acid has the formula C3H*06,and noti CsH,O j its so-called ammonium salt being possibly a product of decomposition like that of glyoxylic acid.This appears not impro- bable as mesoxalio acid may be considered as glyoxylic acid with one atom of hydrogen replaced by CO( OH) PEREIN ON DIBROMACETIC AND OLYOXYLIC ACIDS. 103 CO(0H) CO(0H) I I 1 G{E* c{E I H CO(0H) Glyoxylic acid. Mesoxalic acid. The metallic derivatives of this acid and apparently its ether favour this view. The ammonia derivative being the only exception. This may however be constituted thw- CO(ON&) I CO(0H)

ISSN:0368-1769    

DOI:10.1039/JS8773200090

出版商:RSC    

年代:1877

数据来源: RSC

摘要:

104 ABSTRACTS OF CHEMICAL PAPERS PUBLISHED IN BRITISH AND FOREIGN JOURNALS. General and Physical Chemistry. Effect produced by the Admixture of Foreign Substances with Charcoal in the production of Carbon Points for the Electric Light. By M. GAUDUIN ,(Cow@. red. lxxxiv 218-219) .-The following bodies were introduced into the carbon pencils to the extent of 5 per cent. :-Calcium phosphate chloride borate and silicate ; precipitated silica ; magnesia ; magnesium borate and phosphate ; alumina and aluminium silicate. Calcium phosphate was completely decomposed and the oxide partly reduced ; the resulting fumes of phosphoric anhydride and lime rendered the light twice as brilliant as that produced by ordinary gas- carbon poles. Calcium chloride borate and silicate were also decom- posed and the anhydrides volatilised.The introduction of silica into the pencils reduced their conducti- bility and diminished the light produced. The magnesium salts were decomposed and reduced; the fumes Pormed by the burning metal and by the phosphoric anhydride in- creased the intensity of the light but the augmentation was not so great as with calcium phosphate under similar circumstances. Alumina and aluminium silicate were decomposed with greater diffi- culty and required therefore a stronger current. The aluminium vapour burned with a bluish feebly luminous flame. The flame and smoke which always accompanied these electro-chemical lights ap-pcared so great an obstacle to ttheir use for illuminating purposes that the experiments were discontinued.J. W. The Electric Conductivity of Acids in Aqueous Solution. By F. KOHLRAUSCH (Pogg. Ann,.; clix 233-275).-The acids ex- perimented upon were sulphuric nitric hydrochloric hydrobromic hydriodic phosphoric oxalic tartaric and acetic. The numerous series of determinations detailed in the paper have reference to two points chiefly-the relation of the conductivity to the strength of the solution and the influence of temperature. Some interesting and un-looked for results are recorded especially as regards szclpl2uric acid. Fused SO3 on the orie hand and water on the other are equally non- conductors of the current. Between these extremes the author has measured the conductivity of solutions containing known percentages of sulphuric acid and he finds that the curve which expresses the relation of conductivity to percentage of acid in the liquid has at least three maxima and two minima.One of the latter occurs when the solution contains 84.3 per cent. of acid. Now 84.48 per cent. corresponds with the easily crystnllised hydrate H,SO1 + H,O. The highest maximum occui's at 30.4 per cent. when the conductivity is GENERAL AND PHYSICAL CHEMISTRY. 105 80 times greater than that of H,S04. As to the supersaturation of €€,SO with SO3,thg conductivity increases ; one maximum at least must fall between H,S04 and SO,. At and near the strength 73.18 per cent. corresponding with H,S04 + 2H,O the curve of conduc-tivity shows however no special character.Phosphoric acid gives a curve differing from the rest a8 it rises to the maximum (about 35 per cent.) in a line almost straight and falls in a nearly symmetrical manner. For acetic acid the maximum is attained when the solution contains 1part by weight of C2H402,to 5 parts of water ; and then the conductivity is at least 2,000 times better than that of water and 38,000 times better than that of the concentrated acid. The maximum conducting powers at 18" of various liquids take the following order beginning with the highest. Nitric hydrochloric and sulphuric acids ; potash ammonium chloride ammonium nitrate potassium bromide potassium chloride soda ammonium sulphate po-tassium carbonate sodium chloride. Below these come phosphoric acid and the chlorides of the earth-metals ; poltassium nitrate copper chloride potassium sulphate oxalic acid zinc sulphate copper sul-phate tartaric acid acetic acid ammonia.When the liquid has a minimum of conductivity an increase of temperature raises the minimum proportionately. When the percentage content of solutions of nitric hydrochloric hydrobromic or hydriodic acids is multiplied by the specific gravity and divided by the molecular weight the result -represents the number of molecules in equal volumes ; and it is very remarkable that when these are equal in the case of the four acids named above the conduc- tivities are also equal or nearly so. R.R. Spectra of Metals at the Base of Flames. By M. GOUY (Compt. rend. lxxxiv 231-234).-When a mixture of gas and air burns as in Bunsen's arrangement the flame has an interior cone for its base at the surface of which combustion begins.This surface is brilliant of a blue colour and like the flame proper gives the carbon spectrum. If however the combustible mixture holds in suspension any saline particles the surface of the cone gives a spectrum which differs materially from that of the flame of which it forms the base. A moderately strong solution of a salt is pulverised by means of a jet of compressed air and the saline dust introduced into the flame in a suitable manner. Two spectra may then be observed t'he one above the other. T'he rays forming the lower spectrum are all of thc same height and are produced by the light of the blue surface ; the upper spectrum is formed by the flame itself and overlaps the lower spec- trum by reason of the peculiar form of t,he flame.The spectra of fourteen salts are recorded of which the details re-specting lithium may be taken as an example. The upper spectrum of this salt shows a brilliant red ray and a weaker orange ray ; the red appears equally brilliant throughout its length while the orange ray at the point where it meets the lower fipectrum becomes much more distinctly defined. In the lower spectrum a blue line is visible (7of the electric spectrum) which terminates at the same height as the carbon rays and is altogether wanting in the upper spectrum. ABSTRACTS OF CREMICAL PAPERS. Inkhismanner it is shown that the base of the flame givesfor a short distance a spectrum which resembles the elbctric spectrum of the same metal.J. W. Contributions to the Theory of Luminous Flames. By R. HEUMANN (fiebig’s AmnuZen clxxxii 1-29 ; clxxxiii 102-141 ; and clxxxiv 206-254).-1n the first and second sections of this paper the author considers the effect of the withdrawal of heat upon luminous flames ; in the third he brings forward experimental proof of the pre- sence of solid carbon particles in the luminous flames of hydrocarbons. He first shows that; the burner exercises a cooling action upon the ignited gas issuing from it and that the point at which the luminosity of the flame begins is therefore situated at some distance from the orifice of the burner. The same reason explains the fact that a candle flame does not actually touch the wick and also the fact that a flame does not directly come into contact with a solid body held within it.When the combustible gas is mixed with an indif- ferent gas or is caused to issue from the burner under high pres- sure the cooling action is increased and the distance between flame and burner is therefore rendered greater. If the velocity of the issuing gas be greater than the velocity of propagation of ignition through- out the gas no ignition takes place and it is only when the former velocity has diminished and has become equal to the latter that the flame makes its appearance. If the burner be heated or if the issuing gas be itself raised in temperature then the rate of propagation of ignition is increased and the flame appears at a point nearer to the burner than that at which it had been previously seen.If a metallic wire be held between the flame and burner and be moved towards the latter the space intervening between the flame and burner is diminished. This is accounted for by the protecting influence exer- cised by the advancing wire upon the gaseous particles behind it; these purticles are sheltered from the cooling action of the rushing stream of gas and hence they become ignited. The author suggests that determinations of the velocities of igni-tion of various gases should be made under conditions such that the velocity of propagation of ignition should be equal to the velocity of the stream of gas at the point where the flame begins.A cold object brought into a luminous flame causes a suspension of the process of combustion in its immediate neighbourhood and at the same time materially diminishes the luminosity throughout a considerable space around itself. The deposition of soot upon such a cold object is generally supposed to be proof of decreased temperature but it is shown that a decrease in the temperature of the flame causes a diminution in the quantity of soot deposited. Further it is shown that soot is deposited upon a heated surface held in an ordinary lumi- nous flame ;this soot is however quickly burned away inasmuch as it is impossible to altogether prevent admission of air. The surface collecting the soot is compared to a redoubt stopping the progress of the balls fired against it.The hypothesis which the author puts forward to account for these and other facts is its follows. Carbon containing luminous materials GENERAL AND PHYSICAL CHEMISTRY. may burn with luminoiis flames i.e. with separation of carbon ; or without luminous flames i.e. without separation of carbon according as a certain temperature which differs for each material is or is not maintained. If the combustible material be diluted with indifferent gases a higher temperature must be maintained in order that a lumi- nous flame shall be produced than if it be not so diluted. A porcelain rod was held in the upper (hot) portion of a luminous flame ; soot was deposited. The same rod was brought into the lorn-er (cooler) portion of the same flame ; no soot was deposited.In the former case the cooling action of the porcelain was not sufficient to reduce the temperature of the flame below the point at which carbon was sepa- rated; in the latter case the temperature was reduced below this point. The action of burners upon decreasing luminosity of flames is con- sidered. Tt is shown experimentally that burners constructed of a material having a high conductivity for heat cause a greater diminution in the luminosity of the Aame of the gas issuing from them than is caused by burners having a lower conductivity for heat. If the burner be heated the flame becomes more luminous because the flame is itself enlarged thereby and carbon is also sooner separated in the flame and separated at a higher temperature.Any change in the chemical com- position of the gas which may be brought about by heating the burner or by heating the gas itself is without appreciable effect upon the luminosity. The author draws a distinction between " light effect " of the whole flame and " intensity of light " of the various constituent parts of the flame. He suggests that in practical photometric observations the total light effect should be determined and also the maximum quantity of light obtained by allowing the rays to pass through a small and accu- rately measured opening in a shade placed between the flame and the diaphragm of the photometer. The numbers so obtained might be re- garded as approximative values of the relative " intensities of light '' of the most brilliant parts of the various flames.The principal facts adduced in proof of the presence of solid car- bon particles in ordinary luminous flames are as follows :-Chlorine when admitted into a non-luminous or feebly luminous flame causes a marked increase in luminosity. But chlorine decom- poses hydrocarbons at a red heat with separation of carbon. A rod held in a luminous flame becomes covered with soot only on that surface against which the issuing gas impinges. Were the depo- sition of soot occasioned by the cooling action exerted by the rod upon the flame this deposition would take place equally on all sides of the rod. Heated surfaces become covered with soot when held in luminous flames. This would not be the case if the deposition of soot were a result of the condensat'ion of heavy vapours.If two flames be caused to impinge upon each other with a proper velocity and at a proper angle or if a flame be caused to play against the convex side of a platinum basin placedverticnlly the concave side being strongly heated the solid carbon particles in tbe flame gather themselves together into larger masses and the flame becomes filled ABSTRACTS OF CHEMICAL PAPERS. with glowing points. The soot deposited from such a flame is very coarse-grained. The luminous portion of an ordinary flame is not more transparent than the layer bf smoke which rises from burning turpentine and which certainly contains solid matter. The flame of hydrogen rendered lumiuous by the presence of solid chromic oxide is as transparent as an ordiiiary hydrocarbon flame.Those flames which undoubtedly owe their luminosity to the pre- sence of solid matter throw characteristic shadows when viewed in sunlight against a white screen. Luminous flames free from solid matter cast no shadows. Ordinary luminous hydrocarbon flames cast distinct shadows therefore these flames contain solid matter. If the soot which is deposited upon a cold body held in a luminous flame existed in that flame in the form of vapour it would be possible aga'in t,o vaporise it by applying heat. Stein has shown that this can- not) be done. Stein has also shown that the soot which is deposited from ordinary luminous flames does not cont'ain more than 0.9 per cent.of hydrogen. M. M. P. M. On a Remarkable Regularity in the Volume-Relations of Defi-nite Series of Compounds. By H. SCHRODER (Ueut. Chenz. Qes. Ber. ix 1888-1893).-When an element like silver and a series of its compounds or a component like magnesia and a series of its coin-pounds have volumes which stand exactly in simple relations to each other the author expresses the fact by saying that they have equal volume-masses or equal steres. Their volumes may be represented in fact as multiples in whole numbers of a common volume-mass or common stere. There is as a rule one of the elements of a compound which impresses its own volume-mass or stere on the whole compound and makes itself the prevailing determining or controlling element in a whole series of otherwise very different compounds ;and thus it would seem as if one constituent of a compound exerted a dominating or assimilating influence on the volume of all the remaining constituents.The author's investigations show that in silver oxide carbonate acetate benzoate chloride iodide sulphide telluride in miargyrite and red silver ore the volume mass or stere of the silver dominates. In corundum beryl and andalusite the aluminium stere prevails and this is identical with the stere of silver gold tellurium and graphite. In quartz and cyanite (disthene) the atere of silicon prevails. In periclase chrysolite augite and magnesite the volume-mass of mag-nesium dominates. It is identical with the stere of zinc man-ganous oxide calcite and several other compounds of the magnesium series.The isomerism of andalusite and cyanite is referred to the fact that in andalusite aluminium in cyanite silicon is the determining element while the other constituents are assimilating. G. T. A. Determination of the Specific Gravities of Fats at High Temperatures. By G. W. WIGNER(Analyst 1876 145).-The method coiisists in noting the temperature at which glass bubbles siuk INORGANIC CHEMISTRY. in melted fats. The bubbles are previously tested in liquids of known gravity. For the details of procedure the original paper must be con-sulted. M. M. P. M.

ISSN:0368-1769    

DOI:10.1039/JS8773200104

出版商:RSC    

年代:1877

数据来源: RSC