Mutual Influence of Water and Heavy Water

MUl^UAL INFLUENCE OF WATER AND HEAVY WATERS

By I, RAMAKRISHNA KAO AND

Y. PAR AM ASIVA RAO Andhra University, Waltair.

{Kcccicvcd fo) puldicahoji, ‘lp}il /y/o) riaic V

ABSTRACT. The Kaiiuui bauds ni aii<l IM) fnr Lli< pnii’ liijuids and Ibcir oniitninin^ c(|ual |)rr)poilions iiu- dasn-ibcd. 1 lii' LbaiiLHs in llu* s and thr dislrilmlicjii of intensity o{ lliest* bands aii‘ explained, partly, on the Ijasis of the dt polymerisation of lliese liquids in the mixture, aiifl parth on tln^ hypollusis ol tlie formal ion (jf tlu> HDO moleeides.

I N T R O I) U C r 10 N

The Raman speclrmn oflieavy water was lirsl studied by R, W. Wood, ’ who reported a broad and diffuse band with a maximum of intensity at 2517 cm.“ ^ Later, Aiianlhakrisiian found that this band consisted of tliree components with Raman frctiueucics of 2363, 2515 and 2662 cm.“ ’ . He reported additional bauds, feeble in intensity, at ^{i i k j} and 1250 cm."'. IJauer and Magal observed three more low frc(juency liaiids at 170, 35''and 500 cm. ', besides the one at 1207. Thej could, however, find only two components at 23SQ and 2509 (“jj-j “ 1 in the principal band, but could not confiim tlic m o liand of Aniinllia- krishnati. Rank, Larson and Bordiier ^ working with vapom containing 34'X, proportion of water recorded a fairly sharp hue at 2Mi> cm. ', uhidi tlie) attribut’d to the pure D^t 1 uioleculc. Itiieof u.s with Kotesuaiam studied tlie temperature variation in the stiuclnre and intensity distribution of the band foi heavy water and explained the changes as due to splitting up «'f associated molecules with increase of temperature. The low frequency hands at r7o, 350, 500, ir 10 and 1250 ctn.'L reported by other workers, were not notreed m these investigations. On account of the similarity in the Raman spectra and otlier properties of D-^U and H2O, the authors have undertaken the study of the Raman liands in mixtures of the two liquids, as a preliumiary to the more extensive investigation on the effect of dissolved electrolytes and non-electrolytes upon the Raman band of D-^O.

Coininniiiciiled !>} the Imliaii Phy.Moil Soeiety.

17

136 1. K. Rao and Y. P. Rao

Mixtures uf water and heavy water ueie studied only by R. W. Wood. ^ He woiked with two sami>les ('ontaiUiiijLi i8 and tSo% of heavy uaiei. The intensity inaxiniuni at j.517 in the So'/n iniKUne shifted to 2623 in the iS%

sample. The i<S% niixlnre was estimated to have a coiiiposilion of o'C)34 U-3),

■ 30 I)( )H and o’bf) >■ the eoi resi)ondin^ ]n'oi>orti()ns for tlie So% mixture bein^

()■ (),I 0-32 ])()II and Of)] lI.jO. In tlie former, the number of DOII moleniles is nine times tliat of DoO, while in the latter it is only halt. So he coneluded that the band with its maximum of intensit>' at 2()23 obtained with the iS% sample is mostly due to 1)( )H molecules, while the band with its maximum at 2517 has its orirun in as it is obtained wuth tlie So% sample whicli has a larj^er j)! oportion of 1)^() molecules. Wood never studied the Raman band for pure and hence could not compare his observations for the inixtuie v\ ith this- For exam[)!e, he had not observed the two eomponents of the i)nnci]'al hand of l)j() reported by other workers. Thus his investigations were not complete. The authors have studied the Raman spectrum t)f a 50-50 mixture of heavy and ordinary waters and coiiijiaied the results thus obtained w ith the Raman spectra of OQ'v-1% and imre

K X U 1^ R I AI p: N T a I,

The a])paraiiis used is similar to that already described by one of us ' aiicl consists of a Wood’s lube made of (iLiartz, provided willi a melai cooling jacket and su])]>orte(l on a suitable stand. A mercury arc 15 cm. in length was })laced close to the tnl;>e. A steady current of water flowing through the nietal jacket prevents the liquid in the lube frcmi getting heated by the radiation of the mercury lami>. The water Ilow' also serves as a sensitive temperature control for the contents of the lube- During the e.xiicrimeiits llie temiJeialure did not \':iry by mure than +1 L. Aftei i>ruj)ei collimalion, iJie vScatteied radiation w as focussed oil to the slit (jf the S])Cctrograt>li.

With water atid heavy water an exj)usnie of tom Jioins was enough to gel the corresponding Raman liaiid with siifhcient intensity. ¥<>i- ihe Raman siieetrum of the mixture, Jiowevci, the time of ex[)osuie \n .is dnulaied to compensate for the dimimition of the iiumlier of molecules of each com])oiieiit to half. The mixtine eontaining ecpiai vn:)lumes of llr2t f and D2^ h c ontains equal nnml)ers of these molecules, as tlie ratios of molecular weight to density for li^O and D^O arc almost identical.

'riie three spectra were pliolographed cjii the same plate, together wdth a series of density marks given by varying slit widths. It was ensured that the Raman spectra wutli the pure liquids and tlie mixture were taken under identical conditions with respect to the position of Wood’s tube relative to the arc, the temperature of the liquid contained in it and the running voltage and current of Uic aic, so that the changeSj, that are observed in the intensity of the bands from

D ,0

^ ' %

\ /

\

\ uf

* VaV>,*»/'^

/

V V

w

H ,0

/

■ / \

Microphot

H ,0

Mutual Influence oj Water and Heavy Water 137

the pure stale to the inixliuc, are due to llie imitual iiilliieiKV of the lupiids and not due to any variations in tlie external oonditions either of illinnination, or of temperature. The densities of the bands were determined from a mierophotometric curve and the intensities of the bands calculated in the usual manner, Tlie Raman fre(iuencey at any point on tlie mierophotometric cmve is deteniiincel by extrapolating from a dispersion curve drawn with the inercnry lines as the standards.

u V, s r L T ,s

The bands excited by .1047A of the mcicury arc are studied. The mierophotometric cinves for the D^() ami H j ( ) bands in the pure state and in the mixture are reproduced in Plate V The heavy water band shows two distinct maxiuia, aud one inflexion at a longer frciiueucv . The maximuni at the low'er frequency is less intense. These are the three components of the principal band reported l.iy Ananthakrishnan ' 'I'he shayic of lire band is very much altered in the mixture, the ma.xinimn with the smaller Raman frequency, so prominent in pure Djhb being completely absent, and replaced by an inflexion. Wood ' could not study this interesting change from the puie liquid to the mixture, as he did not work with pure D^O.

160 140

1 -iO

60 40

— /V-

> --

“f

/ ..1

\ ^

— -- --

i /

\

\ t-

i —

---

- rt

w

— - -

—

i ^

! /

/

•

-

-

—- ----, 1 r

j

—- ' \'\

- - -

■

— /'

IL--

' / tr ■

/ - V \

Raiiiai)

Distriliulinn ihe Kanian B;jnd of licavv

\Naler in tlu.‘ ]niu' .slnlr ( ^ - I ami ill thr nii\tiirt - )-

Fio, I

B8 I. R. Rao and Y. P. Rao

111 pure water, there is nu inaxiinum at shorter fre(iueiicy. But the inflexion is distinct enou^^h to indicate that there is superposition of a second handover tlie central band of water Tliere is no trace of this component m the mixture.

Hut, in both lieavy and ordinary wateis, tlie sliape of the band on tlie side farther away from the Kayleij^h line is not altered perceptibly. The mutual influence of HijO and D2O, in this resi>ect, seems to resemble the effect of dissolved acetone upon the water band as ofiserved by C. vS. Rao He remarks therein that dissolved non-electrolytes do not sharpen the band on the lower frequency side.

Figure 1 ^ives the intensity curves for the D^O band in pure heavy water and in the mixture. Besides the absence of the maximum of smaller frequency, there is a large increase in the intensity of tlie second maxinmin in the mixture. The band is sharpened and its extent diminished. Also, the i^osition of the maximuir has shifted slightly towards smalier frequency.

.igno 3100 33e(> 3500

Raiiiaii Pmjueiicv

IniLii^hv I>istribnlinii in the Ranum Hiiml of IT^O ill the pure state (--- ) aiul in the mixture uitli 1\0 ).

Fig. 2

«

Figure 2 gives the intensity curves of tlie HgO band in pure water and in the mixture. As iu the case of 1)^0, the baud is sharper in the niixluie than £01 the pure state, and is less in extent. The position of the maxirauni has sliifted

s l i g h t l y towards longer frequency. The intensity of the maximum has .slightly

increased but it is not comparable to the increase in the intensity of the D^O band in the mixture.

Mutual Influence of Water and Heavy Water 139

Table i gives the important characteristics of the H3O band in tlie mixture and ill the pure liquid.

Tablk 1

Frequencies and Intensities of the II^O Band

1

l^xtcnt Central

JMaxiinum First

Component vSeeorid Component

Pure II2O Raman

f requeueV 2971-3S17

(840 cm. ') 3*143 3632

Intensil’^ 27'5

HaO in the

mixture Raman

frequency (7jS cm ) 3461 3640

Tntensit} fH, i8 32

The corresponding table for DjO is given below*

Tabui; II

Frequencies and IniensUics of the Ru})ian Baud of D^O

!

lixtent Central t'irst Second

i inaximuni Component Component

Pure DjjC^ Raman 2104-2844 253S 2400 2698

frequency (650 cm.’ b seeond maxi-

inuml

Tn tens! tv log 93 ⁴⁰

D2O in the Raman 2244—2810 2523 2394 2685

mixture frequency <566 em b

i

Intensity ^54 ^Ii i)4

__ _______ _________ _____ _ ___

The intensity of the maximum of the ) band has increased by 50% in the mixture, while that of the water band has I'eniained practically the same.

This is a highly interesting aspect of the problem the significance of which will be discussed in the next section. The position of the maximum in 1)20 is not shifted to longer frequency, but slightly in the opposite direction. To allow for any errors of location of the maxima, the authors have studied tw^o sets of photographs and have consistently found that the shift is not to a greater fre­

quency but the other way and of the order of 15 cm."^

To sum up, in both cases the bauds are sharpened and the components of smaller frequency suppressed. The difference in their behaviour, however,

140

/. R. Rao and Y. R. Rao

lies in the increase oi intensity in the case of l)o() from the pure slate to the mixture.

1) T S C IT S S I O N

Tlie princii)al bands of 1)^0 and IloO have each three components. One of us, ^ working with water at various temperatures, found that, with increase of temperature, the component of smaller frequency gradually l>ecomes less intense, the central one remains almost the same, the third component showing a slight increase in intensity. Tlie three components were attriliutcd to HoQ^ (Ili-Oja and 'file trii>le molecules are less stable and arc depnlymerised with increas­

ing tem])erature. C. S Rao, studying the inlluence of dissolved electrolytes and Tion-electrol>'tes iqion w'ater, found that the proijortion of trihydrols was very much less in solutions tljan in pure water Increase of temperature or presence of any foriegn matter seems to split up these higher polymers. The pheno­

menon, in so fur as temperatine elTect is concerned, is found to be similar in heavy water.

Wood Meports that he eoiild not lind any struetiire for the water band.

Mis inability to note llic lower frequency components of D J ) band, which is consj)icLious in our picture w ith i)ure M l’ ), is probably due to the fact that he worked witli mixtures only, and even a small percentage of ILO siip])resses this component. But his failine to get tlie sti iictnre of the w’ater hand is perplexing es])ecially in view of the fact that he must have W'orked at inucli lower tempera­

tures than ours, wliere the lower fiequetiey band of water ought to be more intense- This may probably he due to want of suifieient lime of exposure in his case, and a inicrophotonietric study of the l)and by him would have revealed the other eompoiieiits. rure water shows that it is not a single symmetrical band but is the result of the superposition of at least two bands.

The general changes in the structure of the bands of both Hot ) and DaO are capable of being explained on the basis of the changes in the equilibrium between their polymers, which both these coiiipounds are supposed to consist of.

These changes are similar to those brought about by change of slate or of tem­

perature. 'riic diminution of the intensity of tlie II/ ) band on the kwver frequency side in the mixture indicates the depolymerisalion of the

molecules, to wdiich this portion of the band is supposed to correspond. This change in the structure of the H / ) band lends su]jpoit to tlie view tliat the (H/))3 molecules are less stable thetn the other polymers as postulated by one of us.^

A similar explanation serves for the changes in ihe stiucture of the ) baud. The maximum at .3400 corresponds to the trihydrols of heavy water, and these are less stable in the presence of the water in the mixture. So the component due to them is diminished in intensity.

Mutual Influence of Water and Heavy Water 141

If this were the r>Jily the large inciease in intensity of tliel)a(>

Ixiiul in the mixture cannot be reconciled witli the absence of any coiies])oiuling change in water unless it is assumed that (D-n).-! is less stable than :i l)OStiilate not warranted from results of temperature inilucnce. I'or water alsct, there must be an iiiereasc on account of the breaking up of (ll.a )).> into lower polymers, unless the equilibrium is stich as to keep the number of (H^t ))n mole­

cules unchanged, the highei polymers dissociating directly to single molecules.

But the intensity of the higher frequency conq)onent in ^^atel does not show any ai:»preciable increase to support such a conclusion.

Tile other possibility is the formation ofH D O . The fi eijiiencies of this coni[)omid as caleiilated l>y Van Vleck and Cross are npu), and ^^7,so cm." *, the experimental value reported l)y Rank, Larson and Bardenei for the vapour being e7 iS cn i."L The latter found no trace of any lines ca)iresp(aiding to cm ’ and i,|oo cm.^’* for HDt) ^riioiigh it is surprising to note the absence of O il oscillation in HJ)f) wliilc the corres])Oiiding O 1) oscillation is cons])!- cuons, tlic above residts seem to be confirmed by oiir work also. While tlieiu is a considerable increase in tlic inlensily of the ^ band from the pure state* to the mixture, tlie corresponding TLO band incre^ases only sliglitly in iiitetisiiy.

As required l)y the law of mass action, the pioiiorlion of H 1)( ) molecules formed in a 50‘ 50 mixture of kLO ami 1)^0 studied by us should lie .|SY)%, while IIA) and Dot) are each 25'7%, if tlie eeinilibrinm constant is assumed to be ..'S as given 1.»y I'ivy and Ritten)>urg.'' The abnormal ineaease in llie intensity of the D.O band nnist be due to llie superposition of the HJ)() band over it, while in the ease of the Hot > band the slight increase in intensity is due only tr)thu splitting up of higher iiolymers of water into the lower wliicli thus increases the number of the double molecules giving rise to tlie maximum at the centre of the H^O band, there being no ('ontribntion to this inciease by the O— II oscillation of the IID () molecules as no baud due to this oscillation was detected by Rank, L^ii'son and Bordner foi the vapour and is therefore probably absent for the liquid state as well.

The DoO band should show a decrease in intensity due to conversion of some molecules into HDtJ on combination with ITT). But the change in the equilibrium between its polymers may increase the number of the D^O molecules giving rise to the maximum of the band for the mixtures upon which is sni^erposcd tlie band due to IIDU molecules. 'Ihiis while there are two factors, viz., increase in double molecules of D /) and snneriKisition of the HIM) l)and, which increase tlie inten.sity of the band in the case of D Y ), after counterbalancing the diiniimtion in the number of the D^(.) molecules on account of the conversion into H D () molecules, tliere is only one factor which -ends to increase* the intensity of the water band, viz., the increased number of ll20)o molecules due to depolyrnerisatioii of molecules which are just

8

142 I. R. Rao and Y P, Rao

more tlmii counterbalancing tlieir diminution due to conversion into HDO molecules.

There is vStil] tlie question of the Kaman frequency of HDO band. Wood ^ gives it a frequency of shifted towards longer frequency by luo cin."'^ from the c'cntral maximum of D^.n. The ealculaled value of this band is 2720 and Ihe frequency ol)Served by Rank and co-workers is 2718. Wood observes that when the i^roportion of HDO molecules to D:,.0 is g:i, the shift is to 2623 from 25J5. In tile mixture studied I>y us the ratio is 2:1 and not cmly do we not reeord any shift towards loiigei frequency, but there is a decisive shift in the opposite direction. It is likely that the IIDU frequenc'y m the Hciuid state is much closer to the cenlral comiionent of D.O, than found by R. W. Wood.^

We ex[)ect that the further study contemplated by us will settle the doubts about the Raman frecjueiicy of HDO in the lupiid state and throw' greater light on tile eciuilibriiiin beUvc^eii DHO and D.t). 'I'he absence of the ()-II oscillation in IIJX) is a very iuteresling obsei valion.

R 1^{C I-} 1^{C R} 1^{C N C} 1^{C S}

U VV. W'nnd, 45, (e\vp

R Anniifhiikn’sljiian, /'/c Imi. Ira] Srn'air'^ :or (lo.v"-'’

lialiL‘1 and MaLjal, C. R, 231^{^ Shv}

Rank, and rdner, ( Ihtn 2, jS.j Ugxh-

T. Rainkiisliin Ra > and I' K')1rs\\ ;iram, / ;// f(an I^JiV

C. S Ra.), PluL Mai’ ,, 20, (1933).

T. Rainkrislnia R ao, P)(u. }\(>y ,S‘nr., A, 145, 4^^ (ig^.p Van Vlia'l: and CVn^s, Join n i n u /V/y,s , I, 337 (193

UrcN and Rilteid)crg, Join. ( hem. Phys., 1, 137 Ij(;33)

12.