ON PEROXIDE OF HYDROQEN A N D OZONE.
147
sprue, down which considerable air was carried along with the metal. The reeult of the analysis leaves little doubt that the bottom, made of red bricks and fire-clay, was not thoroughly dry. T h e gases were collected over water, the ingot being immersed in an iron box, with a drill passing through a packing box into its side. T h e boring wae done as rapidly as possible. This method of collecting the gases is riot free from objections, but was the best a t command. The average of two closely accordant analyses was as follows : Carbonic oxide. ................. 2.08 Hydrogen.. .56.42 Nitrogen .39.86 Oxygen ....................... 2.14
................... .....................
A few of the cavities near the surface evidently communicated with the air, hence a little free oxygen, and a larger percentage of nitrogen. The specific gravity of the solid steel, about one foot from the top of the ingot, was 7.866.
xvII~.-oN P E R O X I D E
OF
BY ALBERTR.
HYDROGEN AND
LEEDR,
OZONE.
PxD.
[Second Paper.] (With Figures on Plate I.) Read April Ist, I.-UPON
1880.
T H E R E L A T I V E AMOUNTS O F P E R O X I D E OF H Y D R O G E N A N D
OZONE, OBTAINED I N T H E OZONATION
O F A I R B Y MOIST PHOS-
PHORUS, AS D E T E R M I N E D BY T H E METHOD
O F DECOMPOSITION
O F T H E S E BODIES, THROUGH T H E AGENCY O F HEAT.
I n preceding articles, and more especialty in the first paper with the above title, it has been pointed out that both hydrogen peroxide ;md ozone are formed by the action of air upon moist phosphorus, Moreover, a number of experiments, protracted for several months (in which as many as 481 liters of purified air were used in a single experiment, and every devisable precaution was employed), were brought forward in support of the proposition that the percentage o f hydrogen peroxide in the ozonised air bears a constant ratio to t h a f of the ozotze. T h e conclusioii drawn from these results was, that the production of hydrogen peroxide, like that of ammonium nitrate, depends upon the same series of chemical changes as those which are
148
O S PEROXIDE O F HYDROGEN A N D OZONE.
concerned, in this instance, in t h e generation of ozone itself. N o r e over, that t h e connectioii between the production of t h e three bodies is a necessary one, and t h a t any hypothesis which explains t h e generation of t h e ozone, and ignores the contemporaneous generation of hydrogen peroxide and animonirim nitrate, is of necessity imperfect, and, .to a certain extent, erroneous. T h e object of the present article is t o demonstrate by a method entirely different from t h a t previously employed : I. T h a t both hydrogen peroxide arid ozone are generated b y t h e action of air upon moist phospliorus.
11. T h a t after very prolonged contact with one another, in a series of wash-bottles and dessicating tubes, both bodies survive in amounts capable of satisfactory estimation. 111. T h a t when t h e current of ozonised air, containing hydrogen peroxide, is passed through a tube heated to various temperatures, t h e amount of water obtained by the decoinposition of the hydrogep peroxide increases with the increment of teniperatore.
IV. l'hat under these circumstances, t h e amount of t h e residual g a s capable of decomposing a neutral wlutioii of potassium iodide, regularly diminislies, and a t 200' no such decomposition whatsoever OccLIrS. V. T h a t after this point has been attained, if a solution of potaasiuni iodide (entirely free from iodate) which has previously been acidified with snlphnric acid, be substituted for the neutral solution, i t will undergo slow decomposition. This result is due, not t o ozon(~, which is completely destroyed by continued exposure t o a temperature of Zoo0, but t o the spontaneous decomposition of an acidified solution of potassium iodide in presence of oxygen. T h e objects kept prominently in view, ill devising t h e method of t h e experiment, were : 1st. T o bring filtered and purified air in contact with a large surface of phosphorus, the phosphorus being partly immersed in tlifitilled water, qnite free from nninionincal and nitrous compounds, and maintained during the course of t h e experiment, a t t h e temperature of maximum evolution of ozone (24-25' c.) 2d. T o wash out of t h e ozonised air, as large a n amount of hydrogen peroxide as possible, by an extended series of wash-bottles. 3d. T o free t h e ozonised air, after its escape from the wash-bottles, from every trace of moisture.
ON PEROXIDE OF IITDROQEN A N D OZONE.
149
4th. T o decompose the hydrogen peroxide and ozone, a t gradually increasing temperatures. 5th. T o weigh any water, and to titrate any ozone, remaining after the ozonised air had been subjected t o the action of heat. These ideas were embodied in the arrangement of apparatus represented in the accompanying drawing (see Plate I). The train of purifiers, A, B, C, D, E, of which A was filled with the “ absorbent hygroscopic cotton ” (prepared according to von Bruhns), B and C, with glass beads drenched with soda solution, D and E, with sulphuric acid, was connected with the “ phosphorus ozonator.” T h e water in the jars of the ozonator W;LS maiiitained throughout the course of all the experiments a t 24-25’ C. The ozonator was connected with a wash-bottle by means of kerite tuhing, and the washbottle to four Geissler bulbs, F, G, H and I, the entire five being partly filled with water. T h e last bulb was connected with an empty wash-bottle, J, and this again with a wash-bottle, K, containing sulytruric acid. The drying was completed by three dryiiig tubes, L, filled with beads dreiiched with sulphuric acid. T h e entire length of these driers was 23 meters. From the driers, the ozonised air passed into a curved glass tube, N, dipping down into an oil-bath, M. T h e middle portion of this tube for a length of 0.25 meter, was filled with amianthus, which had previously been ignited. The object of this amianthus filling, was to cause the ozonised air to filter through a great extent of heated air passages. After this followed a weighed sulphuric acid drying tube, P, a sulphuric acid guard tube, T, and a Geissler bulb containing a neutral solution of potassium iodide, W. Between P and T,an empty tube, S, closed with corks a t both ends, was interposed, for convenience in dipping out the drying tube, P. All the connections were made either with strips of cotton cloth, bound tightly about the tubes with flower-wire and coated with melted paraffine applied with a brush, or by corks boiled in paraffine. A number of preliminary experiments were tried, in order to determine whether on aspirating air only through the apparatufi, the weighed drying tribe, P, would change weight, or the potassium iodide solution would undergo decomnosition. The former was found to be the case, a result attributed to the failure to detain all the moisture contained in the air before its reaching the tube, P. In these experiments, instead of the three drying tubes figured a t L, but one was employed, having a length of 0.85 m. It was therefore supplemented by two more tubes, filled with phosphoric anhydride, and having together a length of 1.75 m.
150
ON PEROXIDE O F HYDROGEN A N D OZONE.
T h i s arrangement answered perfectly, so far as the detention of moisture was concerned, t h e drxing tube,, P, not altering in weight in repeated trials, when air alone was drawn through t h e apparatus. The oil-bath was then raised t o a temperature of gooo, and t h e ozonator thrown into actioii. Six liters of air were t11.awn through t h e apparatus a t this temperature, and six more at 1.50'. No change occurred 'in the potassium iodide solution in both experiments. T h i s did not excite surprise, inasmuch as t h e complete (kcomposition at these teniperatnres of any ozone which niight be present, was highly probable. B u t on repeating the trials, upon tlie same amounts of air a t 100" and 2Po, it. was found, most i ~ n c x ~ ~ e ~ t ethat, d l y tlw , iodide solrition did not change. 'l'liere was n o question as t o the clestruction of t h e ozone, during the coiirse of its passage through t h r apparatus, inasmuch as it was plentifillly evolvetl b y the ozotiat.or. Three ways of accounting for this deuoniposition were probable. Either t h e ozone was destroyed by the slow action ~ i thc ' hydi.ogen peroxide, generated sinitiltaneously by the action of moist air i i p n ph~SphOrUS, or it was expended in acting npon t h e paraffine oorinections; or' it was decomposed by the phosphoric atllipdride. Since t h e arrangement of the apparatbs differed only b y the introduction of phosphoric anhydride from many similar arrangenients hithei~to cmployccl i n experimenting upon ozone, the last siipposition appeared the one most probaLle. On exatmitring the anhydride, it was found t o contain no unconibined phosphorus, b u t a notable amount of phosphoroiis anhydride, tlie raising of which to the highest stage of oxidation, had been t h e cause of t h e entire disappearance of the ozone. T h e phosphoric anhydride in the drying tubes was therefore replaced by siilpburic acicl, when i t was found that the irregularities above noticed entirely disappeared. Paraffine does not perfectly withstand the actioii of ozone, being slou.Iy decoitiposed, with the formation of cat~bonicacid. I n the course of long continued use, t h e parafine joints employed by the author in connecting t h e ozolrising eletiients of his large electrical ozoniser (.JOCRNAL AJIER.CHEX.Soc., I, 410), were destroyed, and had t o be renewed from time to time. T h e amount of action on the joints was progressive, t h e paraffine connection between t h e first and second ozonisirig element requiring to be changed very seltlom, while that between the eleventh and twelfth elements needed frequent removal. This increase of action demonstrated t h e effect of 1)rolonging t h e time during which the same portion of oxygen was subjected t o the influence of t h e s'ilent discharge. 111 experimenting with
151
ON PEHOXIDE OF HYDROGEN A N D OZONE.
ozonised air, containing so small a n amount of ozone a s that employed in t h e investigation described in t h e present article, t h e connections may b e made of paraffine without entailing serioiis error, b u t in researches of greater nicety, where not merely the relative but t h e absolute amounts of ozone have t o be determined, paraffine connectione cannot be employed. T h e following experiments were performed under as nearly as possible identical conditions. Twelve liters of ozonised air were drawn through t h e apparatus in each experiment, at t h e rate of six litera per hour. T h e ozonator was maintained a t t h e temperature of 24' C. T h e increase in weight of t h e drying tube, P, corresponded t o the water formed by the decomposition of t h e hydrogen peroxide, when heated to the temperatures indicated in the table. T h e amounts of iodine set free in t h e potassium iodide solution are calculated into ozone according t o t h e equatiotis : 2KI 0, 2HCI = 2KC1 I, II,0 0, and 2KI0, 2HCI = 2KCl I, 0, H,O, each molecule of ozone corresponding t o two atoms of iodine. A large number of trials were made without throwing the ozonator into action, a n d i t was only when t h e sulphuric acid drier, P, both at 0' and when t h e oil-bath was heated t o 200', became constant in weight, that t h e experiments recorded in t h e table were performed.
+ + +
+ + + + + +
Table shoroing the relative amounts of hydrogen. peroxide and ozone formed during the ozonatioii of air by moist phosphorus : 1st Series.
Experiment, Temperature. Water. I. 100' 0 0015 grm
2nd Series. 3rd Series.
4th Series.
11. 1111.
1:. i
V.
VII. VIII. IX.
1 &. XII.
80"
24" 200' 200" 150" 100" 50' 21" 200" 100'
22O
0.0010 0.0000
0.0010 0.0011
"
" " "
0.000~ " 0.0015 ''
0.0003 " 0.0000 0.0018 '' 0.OOO2 " 0.0000 " ' I
He02
0.0028grm 0.0019 " 0.0000 " 0.0019 " 0.00'20 0.0004 0.0028 ' * 0.0006 " 0.0000 " 0.0034 I ' 0.0004 " 0.0000 " ' I
' I
Ozone. Not determ.
0.0000 grm 0 0000 " 0.0011 " 0.0013 " 0.0014
"
0.0037 " 0.0000 * ' 0.0011 " 0.0051 "
These experiments, it appears t o us, establish the truth of the first, second and fourth propositions. They show t h a t both hydrogen peroxide and ozone are formed, and that on heating t h e ozonised air, the proportion of water thus produced regularly increases, while t h a t of t h e ozone diminishes, until at 200' all t h e hydrogen peroxide is converted into water, and all the ozone into ordinary oxygen. At t h e close of t h e twelfth experiment, the immediate train of driers,
152
OS
1'EROYII)E
OF IIYDROOES .4ND OZOSE.
etc., was thrown out, and t h e Geksler bulb containing the potassium iodide sollition, 1%ar connected directly with the first wash-bottle (that preceding F). T h e same volume of ozonised air, a t the same rate, being drawn over uiider these conilitious, g a l e a rcsuit oorresponding to 7.94 mgrm, instead of 5 . 1 mgrm, :tb i l l experiment S I I . In both experinierits, tlic temperature of the ozoiintor \mi L $ - , and of the iodide solution, 22". T h e difference of 2.b-C mgriii, which is 36 per cent. of the amount of o ~ u i i eoriginally evolved, rcprdsents t h e lors due t o t h e decomposition effected by the 4multaneous presence of hydrogen peroxide. This establishes t h e second proposition, and shows not only that peroxide of hydrogen and ozone ran c o - e A t for a Iong time, in dilute condition, without great loss of either, b u t also t h a t the amount of hydrogen peroxide in the ozonised gas, bears a no't inconsiderable proportion t(J that of the ozone itself. I n fact, nrglccting t h e small amounts of hydrogen peroxide, which later on are shown to have heen held back in t h e wash-water, the ratio of the hydrogen peroxide t o the ozone exceeds one to thier. To determine the truth or falsity of proposition V.-" T h a t t h e air, after being deprived of its hydrogen peroxide and ozone, could bring about a decomposition of an acidified solution of potassium iodide," the experiment was again repeated with t h e tube N, maintained a t a temperature of 2 0 0 O . I t will be noted that in every preceding trial a t this temperature, a neutral solution of t h e iodide heing used, ) i o liberatiuic 01% i o d i occurred ~~ ; but in this experiment, a tlecoiriposition took place correspoiidirig to 0.21 mgrm of ozone. T h e increase in weight of the sulphuric acid drying tube, P, due to the decomposition of hydrogen peroxide, and the fixation of water, was 0.0022 grm. T h i s experiment prokes that t h e oxygen contained iii a current of air, from which every trace of hydrogen peroxide arid of ozone, has prcviously been rcniob ed by strong heating, ma)- produce apparently t h e ozone reaction, in ccrse the potassitoil iodide solzction irsetl ,foi* titrution flus been ctcitl~fietl. It may be objected, although the objection is riot supported by the above results, that even after passing through an amianthus plug 0.25 m. in length, maintained a t the temperature of 200°, all the hydrogen peroxide had not been decomposed, and t h e residual peroxide had brought about t h e change noted. It may also be said that t h e result had been due t o t h e use of potassium iodide, containing iodate. To show that a far simpler supposition accounts for the decomposition, the experiment was performed with air alone. and po-
153
ON PEROXIDE OF HYDROGEN AND OZONE.
tassium iodide, from which every trace of iodate had been removed, was used. To prepare b y the niethod we have found most convenient* any amount of the crystallized iodide, ahsol'utely free from traces of iodate, is so troublesome, that we preferred to employ a salt containing iodate, but from which the iodate had been carefully removed before using. This we attempted t o do by throwing down t h e iodic acid with baryta, and then getting rid of the excess of baryta from the boiling solution by means of carbonic acid. But a notable amount of barytic iodate remained in solution. It was found better t o saturate the neutral solution of potassium iodide, containing iodate, with hydrosulphuric acid, and then drive off the last traces of H2Sby heating. W i t h this precaution, the potassium iodide sold as chemically pure, may be used in investigations similar to the foregoing. 2 0 O.C. of such a potassium iodide solution, containing 10 per cent. of iodide, was placed in one GeiRsler bulb, and the same amount slightly acidified, in another bulb, the two being connected together, and 4+ liters of air drawn through them. A t the expiration of 1+ hours, the n e u t m l solution was entirely colorless; the acidQied solution contained 1.4 m g r m of,free iodine. This experiment points out too emphatically t o need any comment, the magnitude of the error brought about by the use of acidified solutions in various titrations, and especially where it is used to estimate the ozoiie contained in a slowly moving current of ozonised air. 11.-AMOUNT
OF
HYDROGEN
PEROXIDE
ABSORBED
BY
THE
WASH-
WATERS.
In earlier experiments, undertaken first, t o determine the relation between the temperature and volume of the ozone given off by the action of moist phosphorus upon air, and second, to estimate the amounts of by-products, the quantities of hydrogen peroxide had been determined in the wash-water only. This was done, after testing with metadiamidobenzole had established the absence of nitrous acid. I n one series of experiments, the hydrogen peroxide thus determined, was the -rbvth part of the total weight of ozone evolved. Moreover, only the wash-water in the first wash-bottle was titrated, inasmuch as that in the bottles placed later in the series, contained too little hydrogen peroxide to permit of satisfactory estimation. But it was found by these later experinrents, that this method gave c1 much too low result $br the relutive perceiztage of hydrogen p e r o x ide, since the greater part of it was not held baok by the washwater. *Contributions to the Chemistry of the Atmosphere,Ann. N.Y.Acad. Eki.,
I
r'l
1 he bulb F, containing l i C.C. of water, after evaporation t o twnthirds its original volunie, in order to expel any dissolved nzon(~,was found to contain 0.2 nigrrn H2O2. T h e bulb G, containing 2ti C.V. of water, after being similarly evaporated, gave a reaction corresponding to 0.08 nigrni H,02. T h e bulb H, to 0.01 mgrm I-120g. 'l'hv falling off in t,he amount of hydrogen peroxide cwntained i n F, i n comparison with that in Q, is pcrhaps greater than should be, for thr reason that H arid I were changed during the course of the experimerits, and a much smaller volurne of ozoriiued air passed through them, than through the two preceding bulbs. Hut, the striking feafure i n the experiment is, that only 0 . 3 rngrm of 11,Q in all was absorbed by the wasli-watci,s, the remainder passing on i n tlic S ~ I ~ C R I I I of ozonised air. II.-UPON
IHE
RELATIVE
AMOUSTS
OF H Y D R O G E S PEROXID+:
AS11
0 2 0 NE.
T h e considerations above detailed, render manifest that the solution of this problem is attended with great difficulties. For the titration must be effected either in the neutral or acid solution ot potassium iodide. In case the foriner is employed, there is no reason f o r supposing t h a t any of the ozone itself escapes without decornposition. T h e statements which ha\ e h e n made t o t h e contrary, liavc probably been based on the fact, t tiat the potassium iodide solution has been employed to absorb n mixture of ozone and of hydrogel1 peroxide held in aerial suspension. .4nd, moreover, the observation of von &bo has been lost sight of, that wlienever ozone is deconiposed by potassium iodide solution, hydrogen peroxide is set free. B u t t h e Case is different with the hydrogen peroxide p ~ e s e n ti n t h r ozonised gas. A very small portion nf this undergoes absorption by a neutral solution of potassium iodide, as is strikingly shown by tlie following experiment. ,4 Geissler biilb containing an 0.8 p e r ceiit solution of hydrogen peroxide, was connected with another bulb filled with a 10 per cent. solution of neutral potassium iodide, and 1%liters of air were ahpirated through t h e twn bulbs, the current flowing from t h e former through t h e latter. If t h e current of air merely deconiposed a portion of the peroxide into water and free oxygen, 110 iodine would be liberated, inasmuch as oxygen alone is without action upot1 pure neutral potassium iodide. But, if any H,O, were vaporised and carried over, it might be expected to act, although not energetically, iipon the neutral solution. This latter supposition was the one which accorded best with the results of the experiment. T h e first bulb of thri
ON PEROXIDE OF IITI)KOOES AND OZOSE.
155
Geissler apparatus became decidedly, t h e second bulb slightly, yellow. On titration. it was found t h a t 0.02 nigrni of H,O, had been absorbed. W h a t would occur if t h e titration of t h e ozonised g a s were performed in an acidified solution of potassium iodide? I n t h e first place, a decomposition would result from the action of the air alone, which would vitiate t h e results. I n t h e second place, even an acid solution acts very slowly upon hydrogen peroxide, when t h e latter i s present in extremely small amount. Thus, Schoene* has shown t h a t it is only in somewhat concentrated solutions t h a t the deccmposition is complete after a short period. I n a solution which contained in 1 liter newly 1 g r m H,O,, 24 hours were required for t h e completion of the reaction. When about one-quarter g r m H,O,, in a liter, was present, it was not until after 30 hours that 98.2 per cent. of t h e total amount of hydrogen peroxide had been decomposed. W i t h greater dilution, still less action ensued, and with only 5 mgrms H,O, t o t h e liter, after 48 hours, less than three-quarters of this amount had been decomposed. I\-.-PRODUCTION
O F H Y D R O G E N PEROXIDE, D U R I S G T H E DECOYPOSI-
TION O F POTASSItJM IODIDE BY OZONE, A N D T H E ORIGIN O F T H E W H I T E CLOUD FORMERLY THOUGHT TO INDICATE T H E P R E S E N C E O F ANTOZONE.
One of t h e most singular phenomena connected with t h e action of moivt phosphorus upon air, is the dense white cloud which makes its appearance after t h e ozoiiised air has issued from t h e potassium iodide solution. If this cloud ifi drawn over into an aspirator bottle, i t gradnally disappears on standing. T h e phenomena are of such a character t h a t it is difticult t o avoid framing t h e hypothesis, t h a t we have t o deal, not with one, b u t with two substances, one of which is absorbed by the potassium iodide solution, while the other passes on, and asserts itself b y t h e production, in contact with water, of this mysterious white cloud. I n fact, the hypothesis of t h e dual nature of active oxygen, the ozone and atmizone of Meissner, t h e ozone and antozone of Schoenbein, had come t o be regarded as well established, and has only recently been overthrown. T h e credit of so doing is gencrally assigned t o Nassep and Engler, who demonstrated t h a t t h e so-called antozone, is merely hydrogen peroxide, diffused through a large quantity of aqueous vap0r.t B u t this merit rather belongs, it
* Zeitschr.
fuer analgt. Chemie, 18, 145. 154, 230.
t Ann. der Chim. und Pharm.,
seems t o me, t o von I3abo,* since his work hardly appears t o have o b tained the consideration which it deserves, and sincci some i)t' his most important conclusioris :ire, eve11 :it t l i r present time, generally overlooked by writers upon this siibject. I t h i n k i t will be ivcll t o re-state tliem hcw. Ii(x says, " the forniation of tlie (:lor111is always t o be noted, wlien i n any n i a n n c ~ivliatso~~vcr oxon(' is d e c o n i p o ~ e dwater , being present. 'I'he c~lontls i i c c i i r i n the gre;itcAst intensity, wlicii n portion of the ozone escapes ilecoiii~~i~sitioii, anti this portion, as Jleissncr lias quite correc.tly observetl, a r t s i l l the d r y eontlition ripori t h e aqueous solrit ion, in ~ v h i c hit protliices osidatiori. '' Meissiier 1)elieves that the clontls arc' not Ii!iiropcri peroxide, hecaiisc this is riot volatile, ~ n t theixafore l are tlricx t o :i peculiar nioditication of o s y g e ~ (;intozone). i " IYheii ozoriiseil o x y g e n acts iipoii oxidiza1,le bodies, a reaction ~ in:iriy analogies. For instance, i n t h r oxioccurb for whicli \ v Iiavc. dation of a siil~stance,n o t i)rily is this hody acted u1)on l)>- t h e oxidizing agent, 1)nt i n H C ) I I I ~ instances, tlio operation of this agent exteiitls t(J otlier hodicq, wliicli otlierivisc :LIT riot :it tarked. It niay siiifict: t o iiiataiii.(: tlic osiiliitioii I)[ n i t r o p > n in t l i i , bnriiing of airinionin, or t h e solutic.)n of ~il~itiiiurii, :illoycci with bilvcar, i n nitric maid. blveri so, ozont! b y i t w l f C ~ O Wnot act upon water, but oxidize> it when, at t h e sanie t i m e , i t acts \ 1 1 ) 0 1 1 otlier re:idily ositlizn\)lc 1)odi es. " In likc manner, by itb action U ~ O I Ipotassium iodide, togetlic~ with free iodine, potassium pcrcixide anti potassium iodate, hytlt*o!/c~l ~ m o z i t l eis $ ~ ~ * m e t l . ' ' ( ' h e italic5 :ire the authors,) " \Vlien ;L dry current of oxone goes throuqh a solution cont;iiniiig pot:issiuin iodidr, t h e clry gas readily takes up v:ipor (if ivater, ~ v l i i c * l itlif'fiises itself in t h e gaseous current, and thus, at least in part, I)ec:oriic. with water vapor. Uy the action of tlits ozone, 0 1 1 t l i v [)ne liand, iodine is made free, arid on t h e otlier h i i d , tlie i-:iiior of watcr i k ox id i zed. illoii g-with tli est, a v 01 ,zt i 1(, coni 1, i n:1 t i on c I i' i od i ne i i r i (1 ozonc may he formed. Tlic forin:ttion o f tlie Iiy,lrog(~ii p w x i t l c ~ may be lirought about by the oxidstioii of the tliiitl a s \vel1 :is I J t~l i r ~ gaseous water. " Since, however, the hydrogen peroxide can only tlsist i i i the fliiitl condition, t h e port>iori which is formed iq the gas condcnses no fog vesicles, t h a t is t o say, a liquid suspended in the gas. l'liis can swim f o r some time in t h e gns withont decomposition, in:ismriuli as it finds itself in an atmosphere more or lcss satnr:tted with water vapor, t h c ~
____.______.
.- . ~
.
-
-
----
* Reitr. zur Kentnias iles Ozons. Ann. dci ('lieni,, Suppi, 1 i\nd 2. 1863,290.
O N PEKOXIDE OP HYDROGEN AND OZONE.
15’1
reason of t h e rapid decomposition (the abstraction of water) not bei n g present. How far such a cloud can be conducted through an apparatus, how difficultly condenbible it is, is not surprising, when one reflect8 on t h e well-known behavior of other solid or fluid bodies suspended in gases. T h e destruction of t h e cloud, t h e so-called ’ nbk?ii?gen’ of t h e atmizone, can happen in part through the gradual precipitation of t h e hydrogen peroxide, and in part through its decompobition, as when, for example, the gas is not sufficiently saturated with water vapor.” W e believe t h a t the theory advanced by von Baho is the correct one, and t h a t it affords a satisfactory explanation of the phenomena associated with ozonation, whether the method by phosphorus, or by electrolysis, or by t h e silent discharge, be employed. For when in t h e ozonation by phosphorus, t h e white clouds (of phosphoric anhydride and peroxide of hydrogen) which a t first accompany the evolved gas, are completely absorbed by repeated washi n g with water, they make their appearance again at t h e nzoinant LEU the ozone is decomposed b y a potassium iodide qolution. T h c phenomenon is a very striking one. T h e de-ozonised air, laden with t h e suspended hydrogen peroxide, thus generated, will present the appearance of a white cloud for a long time, eken when absorption of t h e peroxide is slowly taking place a t the surface of t h e water in the vessel into which t h e air has been aspirated. Still more convincing are the phenomena attendant upon the ozonation b y t h e silent discharge. To study these with more care, t h e following experiment was performed: perfectly pure and d r y oxygen was ozonised by one of t h e “ozonising elements” (see JOURNAL AMER.CHEJI.Soc., I , 441), and then drawn through three bulbs, the first containing solution of neutral potassium iodide, the second water, and the third Fhromic acid. T h e latter was connected with an aspirator. As soon as t h e strongly ozonised oxygen came into contact with t h e potassium iodide, a very slow current being employed, a yellow coloration made its appearance on the surface of t h e solution in the ,first b d h , and above t h e surface the characteristic white cloud. On increasing t h e rapidity of the current, this white cloud was drawn through the water and chromic acid, into the aspirator, and remained for one or more hours before its absorption was complete. A f t e r drawing over 6 liters of oxygen, containing 0.104 mgrm of ozone, the water was titrated, and found t o have absorbed 0.31 mgrni of peroxide of hydrogen. T h e cliromil: acid Aolution was unaffected, shoa ing that the suspended hydrogen peroxide may pass
through it without effecting a decomposition. T h e same remark applies t o a neutral potassiuni iodide solution, put in the placv of t h e chromic acid. The excessively dilute hydrogen peroxide held i n ;L stat,e of aerial aqueous suspension, was not ;hie t o Jecornpose t h e neutral iodide during t h e coursc o f tlie t.xpei*irrtt~nt. Of course, : L I ~ ricitZ{fieietZ solution c o d t i not be enipioyetl to absorl) the peroxide, since, as above ~ I I O W I Iits , cieconilmsition (n'ciirs iriider t h e action of oxygell alonc. A final series of esperiments \va5 then tried with t h e following objects : 1st. l'o repeat t h e experirnent of Andrews (It. Sot. Proc., 7, 4;s; Ann. der Chem. 11. Pharm., 97, 3 i l ) , and to determine whether the amount of ozone, as determined by titration, was precisely the same as that determined by weighing, Tvith. suitable pwcautions, the potassium iodide solution iised for absorption, before1 and after t h e experiment. Also, to discover i f possible, t h e so~:rces of tile anomalies observed by Aiidrewa, when a neiitr;tl solution of iodide was employed. a d . T o establish whether, the same volume of oxygen being employed in two experiments performed under otherwise identical conditions, the amount of ozone as dcterinined by the titration and neighing of an acidified solution of potassium iodide, would not be in excess of t h e amount as determined by the titration and neighing of a nciitral sol ut ion, : j d . After drying t h e oxygen, on its leaving the potassinni iodide h o l n t i o n , to find tlie amount of hydrogen peroxide, generated by t h e cIecompositiori of t h e ozone i n presence of water, and of an osidizable tlodj-. 4th. T o discover the alterations in compositioii \vhich t h e potassium iodide solution, itself, had undergone ; tu find out what proportion of the potasli sct free by t h e liberation of tlie iodine, had by ;x. further process of oxidation, been converted into iodate, and in what condition the remainder was left, so that it was without action on the liberated iodine. 'Yo perform these experiments, t h e apparatus employed above was modified by t h e introduction of a sulphurio acid drier, a f t e r the potassium iodide bulbs, and t h e drier was again protected by L guard from t h e latter part of t h e train. Before attaching the apparatus, similar t o t h a t shown in the wood-cut, f o r determining the hydrogen peroxide evolved from the potassium iodide bulb, the experiments included under t h e first head were performed :
159 First, with neutral solution of iodide.
T h e weights of t h e solution and the drier were carefully determined, before and after each experiment. Aniount of ozone, as found by increment of weight. 0.114 grm. 66 66 66 t i t r a t i o n . . .......... 0.108 "
--
Excess (as determined by weighing). ............. 0.006 " 2 . 7 5 liters of oxygen were drawn over in 2 hours and 40 minutes.
Rcpated. Amount of ozone, as found by increment of weight. 0.0972 grin. 66 i i thc reaction i n the ncwtral solution niay be rq)rcsentcd by the c q w t i o n : lIiI 211,o + 6 0 , = I r:mc on the addition of acid ; a fresh portion of thc potassiuni iodide u ~ o u l d be tlecoruposetl, with liberation of iodine. 1 h t a yriantitative imalysis showed that tlie amount i ) f pot;issiiini cumI)inod as iodate, was not sufficient t o account for all the iidiiie set free oil acidifying. Hence, we believe that the explanation of von Babo is tlie correct one, and that, i n addition to potassiurri iodate, small amounts of potassium peroxidc and hydrogen peroxide are also for1ned.
+
+
+
+
+
