Hurd Et Al, 1954. Hidroxidos

7/27/2019 Hurd Et Al, 1954. Hidroxidos

http://slidepdf.com/reader/full/hurd-et-al-1954-hidroxidos 1/5

Subscriber access provided by CONICET NACIONAL INSTITUTES

Journal of the American Chemical Society is published by the American ChemicalSociety. 1155 Sixteenth Street N.W., Washington, DC 20036

The Mechanism of the Base-catalyzed Rearrangement of Organopolysiloxanes1

Dallas T. Hurd, Robert C. Osthoff, and Myron L. CorrinJ. Am. Chem. Soc., 1954, 76 (1), 249-252• DOI: 10.1021/ja01630a064 • Publication Date (Web): 01 May 2002

Downloaded from http://pubs.acs.org on April 27, 2009

More About This Article

The permalink http://dx.doi.org/10.1021/ja01630a064 provides access to:

• Links to articles and content related to this article• Copyright permission to reproduce figures and/or text from this article

http://dx.doi.org/10.1021/ja01630a064



Jan. 5 , 1%4 BASE-CATALYZEDEARRANGEMENTF ORGANOPOLYSILOXANES 240

[CONTRIBUTIONROM THE RESEARCHABORATORY,ESERAL LECTRICo. ]

The Mechanism of the Base-catalyzed Rearrangement of Organopolysiloxanes

BY DALLAS . HURD, OBERT . OSTHOFFAND MYRON . CORRIN

RECEIVED UGUST 4 1953

The mechanism of the pol)-merization of polydimethylsiloxanes with strong bases is discussed. This mechanism involvestwo distinct although related steps: (a ) the entry of the catalyst molecule into th e siloxane system, and (b) the continued

action of the catalyst molecule, or some part of it , in the reshuffling of siloxane linkages. It is shown tha t a rate-controllingfactor in the polymerization of octamethylcyclotetrasiloxane is the base strength of the catalyst. A study of the base-cata-lyzed polymerization of dimethylsiloxanes at room temperature in homogeneous alcoholic systems also is discussed; thebase-str ength effect is clearly demon strat ed. Th e nat ure of th e silicon-oxygen-alkali metal bond has been investigatedby measureinents of conductivity in siloxane and alcoholic media, both at room and elevated temperatures.

I. Introduction

There are two general methods tha t are used forthe rearrangement or polymerization of polyorgan-osiloxanes catalysis with acid^,^-^ or with strongbases6 Th e acid catalysis has been used for therearrangement or so-called “equilibration” of sili-

cone fluids. Base-catalyzed polymerization hasbeen employed for the preparation of silicone rub-ber gum stock (polydimethylsiloxanes of high mo-

lecular weight) from low molecular polydimethylsil-

oxanes6 In particular base-catalyzed polymeriza-tion of octamethylcyclotetrasiloxane (I) has beendescribed.‘j Th e strong base which has been cus-tomarily used is anhydrous potassium hydroxide7and reaction times up to several hours are desirablefor the preparat ion of polymeric diorganosiloxanes.

11. Mechanism of the Rearrangement Reaction

No ma tter what the mechanism of the poly-merization of polydimethylsiloxanes, it is apparentthat the process must involve an opening and clos-ing of silicon-oxygen bonds, that is, a reshufflingof siloxane linkages. In t he absence of chain-ter-minating units and any thermodynamically favoredmolecular form, a random shuffling of siloxanelinkages would be expected to lead to high molecular

weight forms as the most probable forms. On thebasis of this random reshuffling, there should be adistribution of molecular weights; this distributionhas been observed in practice by fractionationtechniques. At equilibrium, increased amountsof chain-terminating groups would be expected to

yield products of lower average molecular weight.This effect also may be observed in practice.

There appear to be two distinct although re-lated steps in t he mechanism whereby a strong basesuch as potassium hydroxide can catalyze the poly-merization of a polysiloxane: (a) th e entr y of thecatalyst molecule into the siloxane system, and(b) the continued action of the ca talyst molecule,or some part of it , in the reshuffling of siloxane link-

ages.

We have postulated that the first step in thebase-catalyzed reaction is the coordination of the

(I) Presented before the Division of Physical and Inorganic Chemis-

try , American Chemical Society Meeting, Chicago, Ill., September,

1953.

(2) W. I. Patnode and D. F. Wilcock,THISOURNAL,8 , 360 (1946).

(3) M. J. Hunter, 5. . Hyde. E. L. Warrick and H. J . Fletcher, i b id . ,

(4) M. J. Hunter, E. L. Warrick, J. F. Hyde and C. C. Curie, i b id . ,

5 ) D . W. Scott, i b i d . , 68 2244 (1446).

(6) J . F. Hyde, U. S. Patent 2,443,353, June 15, 1948.

7) E. L. Warrick, U. S. Patent 2 634 2.52 April 7, 19.53.

68 667 (1946).

68 2284 (1946).

oxygen atom in the potassium hydroxide moleculeto a silicon atom in a siloxane chain, and a subsequentrearrangement of the act ivated complex with chain

Si-0-Si + KOH Si-0-Si

tKOH

cleavage. (Itwill be understood tha t, unless they areotherwise designated, the silicon atoms will also belinked to two methyl groups and to one other oxygen

atom as parts of large molecules.)Once the potassium hydroxide has entered the

siloxane chain in the form of the potassium silano-late- iOK, it is proposed th at this is the activespecies in causing further siloxane rearrangements.This assumption is substantiated by the fact thatmetal silanolates have been shown to bring aboutsiloxane rearrangements.8

The silicon atom appears to be a relatively goodelectron acceptor. It is known that silicon cancoordinate electron donor groups such as, for ex-ample, fluoride ion in the formation of the fluosili-

cate ion, and that certain reactions of silicon com-pounds do proceed through intermediate coordina-tion co mp lex e~ .~ here are, in fact, compounds

known in which silicon coordinates as many as sixoxygen atoms.1°Now, one would expect that if coordinat ionof the

oxygen atom as an electron donor to the siliconatom were the first step in reaction12 ncreasing theefficiency of the oxygen as an electron donor shoulddecrease the temperature a t which t he base be-comes soluble in th e siloxane and, possibly, increasethe rate of solution. As evidence for this, we haveobserved that whereas potassium hydroxide doesnot dissolve appreciably in I until a temperature ofabout 130’ has been reached, cesium hydroxidedissolves a t 100’ or lower. Sodium hydroxide, aweaker base than potassium hydroxide, does notdissolve a t 150’ and, in fact , seems to be ineffectivein catalyzing the polymerization reaction a t this

temperature. Lithium hydroxide similarly is in-effective; it does not appear to dissolve, and i t pro-duces no apparent polymerization of I in 16 hours

a t 175’. Potassium oxide, in which the oxygenatom presumably is a strong electron donor, dis-solves quite rapidly in the siloxane a t 150’ and doespromote t he polymerization reaction.

Si-0-K + H-0-Si (1)

8) J. F. Hyde, U. S. Pat ent 2,634,284, April 7, 1953.

(9) C. G. Swain, R. M. Esteve and R . H Jones, THISOURNAL , ‘71,

(10) A. F. Wells. “Structural Inorganic Chemistry,” Oxford lini-

943 (1949).

versity Press, Oxford, 1945, p. 459, et seq.



2.50 DALLAS. HURD, O ERT . OSTHOFFAND MYRON . CORRIN Vol. 76

We have observed th at certain compounds con-taining strong electron donor groups other thanoxygen also are capable of coordinating to, and cat-alyzing the polymerization of, siloxane systems.For example, potassium amide, in which the nitro-gen atom is a strong electron donor, dissolved veryrapidly in I a t 150" and caused the formation of

high molecular weight polydimethylsiloxaneThe reaction delineated by equation 1 above

undoubtedly is reversible, and it should be notedth at th is type of reaction could be used alone to ex-plain the reshuffling of siloxane linkages. How-

ever, we consider that it probably plays a minor

role in the redistribution process, and that theamount of reversibility in this reaction actually issmall, particularly a t t he low catalyst concentra-tions normally employed.

We believe the second step in the reaction to bea rapid exchange reaction. (The stars are used

Si-0-Si + Si-0-K -- Si-0-Si + Si-0-K ( 2 )* * X

here to differentiate individual silicon atoms, and it

should be remembered that these units ar e parts oflarger molecules.) This reaction, like the initialreaction step, also must depend on t he formation of

intermediate coordination complexes, in this casethrough t he action of the oxygen atom in th e Si-0-K as an electron donor, and also will be influencedby the electropositivity of the alkali metal compo-nent. This reaction is, of course, reversible, bu twith a large concentrat ion of siloxane and a smallconcentration of c atalyst, the probability t ha t the

particular molecular units designated as Si-0-Si

and Si-0-K would reac t again would be small in

comparison to th e probability tha t the Si-0-K unitwould react with a different siloxane linkage. Fromthe present study it is not possible to determine

whether reaction 1 or reaction 2 is the rate-determin-ing step, although it is expected t ha t th e choice may

vary depending on the particular siloxane-catalystcombination employed.

The authors have observed qualitatively th at th epolymerization of I is brought about at the samerate b y potassium amide and potassium hydroxidea t 150". This might be expected, since in a largeexcess of the siloxane, both these materials shouldyield the same active intermediate, ;.e., -SiOK.

On the other hand, if potassium is replaced by amore electropositive metal, such as cesium, thepolymerization reaction not only proceeds muchmore rapidly, but will proceed at a much lowertemperature. I t appears thus that the enhancedability of the oxygen as an electron donor is main-tained in the Si-0-Cs groups throughout the course

of the polymerization. Conductivity experimentsvide infra indicate th at the Si-0-Cs unit is iiot

appreciably ionized in t he siloxane medium a t150. Thus it has been established the order of re-activity of bases toward I a t elevated temperaturesis CsOH > RbOH > KOH > NaOH, or in otherwords the reactivity is related directly to the si7e ofthe cation.

A further consequence of the proposed mecha-nism is th at the molecular weight of the polymer

should depend on the nature of the catalyst. If,

*

*

for example, the hydroxy group that becomes at-tached t o one end of a siloxane chain during the firststep of t he process is relatively unreactive, it willact more or less as a temporary chain-terminatingun it; this will tend to keep the average molecularweight of the polymer lower tha n t ha t which wouldbe expected in the absence of chain-terminatinggroups. However, hydroxyl groups can he lost

by condensation reactions between Si-0-13 units.

Thus, for polymerizations conducted in open sys-tems and over long periods of time, it is probable

that most of the hydroxyl groups become lost in thisfashion, and so are ineffective as permanent chainstoppers ; high molecular weight gums eventuallyare obtained. On the other hand, if large and pre-sumably inactive groups such as the methoxygroup are present (polymerization with potassiummethoxide, see Experimental section), these groups

ac t as permanent chain-terminators, and soft gumsof low molecular weight are obtained. These softgums do not become firm on extended heating sothe effect is not one dependent on the rate of poly-merization.

111. Rearrangement Reactions in Alcoholic Media

lye observed that the base-catalyzed siloxane re-

arrangement process could be studied a t roomtemperature by polymerizing I with various basesin a medium of anhydrous methanol or ethanol.The polydimethylsiloxanes of low molecular weightincluding I are soluble in methanol over a widerange of concentrations. The polydimethylsilox-anes of high molecular weight, however, are not ap-

preciably soluble in methanol; and if a solution of Iin methanol is caused to polymerize, the highermolecular weight siloxanes separate as a distinctphase. Although we made no att em pt to determinethe relation of siloxane molecular weight to solu-bility in methanol, the separation of the higher mo-lecular weight products as a new phase was useful

in comparing the relative effectiveness of differentbases inopromoting the polymerization of I. It will

be noted that this technique, in which the base inalcohol solution was added to the alcohol solutionof I , eliminated the uncertainty of any inadequatedispersal of a difficultly soluble, powdered solid inhot siloxane. I t is assumed, however, that thecoordination of the alkali metal hydroxide, ormethoxide, to the siloxane still occurred as the pre-

liminary step in the arrangement and polymeriza-tion reaction.

U'e found that the time required for the appear-ance of the second phase in the alcoholic polymeri-zation reaction was dependent on both the concen-tration of I and the concentration of the base in

the methanol solution, as well as on the nature ofthe base. Under certain conditions, it was pos-sible to obtain systems which exhibited turbidityalmost immediately on addition of the alcoholicbase, and the appearance within a few minutes of adistinct second phase which gradually increased involume as the polymeriiation reaction proceeded.*llthough it was recognized th at the new phase insuch cases was not entirely polymeric product,being swollen with I and methanol, i t was indeedsurprising that the polymerization reaction pro-



Jan. 5 , 1954 BASE-CATALYZEDEARRANGEMENTF ORGANOPOLYSILOXANES 251

ceeded so rapidly a t room temperature. Th e baseconcentrations used were too small for any "salting-

out" effect, as indicated by t he fact th at homogene-ous solutions were obtained initially.

By set ting up a series of similar systems with dif-ferent bases, it was possible to make some fairlyquantitative comparisons of the relative effective-ness of different bases in promoting the polymeriza-tion reaction (see Experimental). As we expected,

both cesium and rubidium hydroxides caused amore rapid polymerization of I than potassium hy-droxide. Tetramethylammonium methoxide wasalmost as effective as cesium hydroxide. This was

expected, however, since quaternary amine basesare observed to be quite effective in catalyzing thepolymerization of organosiloxanes a t elevatedtemper atu res 6

It should be recognized that these polymeriza-

tion experiments in alcohol are not strictly com-parable with the polymerization of I with base atelevated temperatures. For one thing, the poly-meric siloxanes obtained, although of considerablyhigher molecular weight than I, would be chain-terminated with methoxy groups for the most part.Their average molecular weight would depend on

the relative concentration of alcohol and siloxane,b u t in any case would be very low relative to theaverage molecular weight of a firm silicone gum.The experiments do, however, correlate basestrength with catalytic activity ; in this respect,the observations agree well with those previouslymade on the polymerization of pure I.

IV. Conductivity Measurements

Conductivity measurements were first made todetermine whether the active silanolate interme-diate (Si-0-M where M is Cs, Rb or K) was appre-ciably ionized in the siloxane medium during therearrangement reaction. A number of experi-ments made with the most sensitive equipment

available, indicated clearly tha t an y ionization wasso small as to be negligible. Very small and transi-tory rises in conductivity were observed immedi-ately following the solution of the base in t he silox-ane a t the elevated temperature, bu t this effect wasfound to be caused by the small amount of water

originally present in the hydroxide, or that formedby silanol condensation.

After the unusually rapid polymerization of I inabsolute methanol had been observed, a few con-ductivity measurements were made on this systemto obtain, if possible, additional information on therate of the reaction. This sor t of information wasconsidered to be useful, since little is known aboutthe effect of traces of catalytic impuri ties on thephysical and chemical behavior of silicone materials.These experiments corroborated the observationsmade visual ly; the reaction of I with rubidium hy-droxide, for example, was found to occur irnniedi-ately on mixing, and to proceed quite rapidly.Of par ticular interest was the observation thatlinkages between trimethylsiloxane units were notaffected by base in alcohol at room temperature,although such linkages are attacked a t 150".This behavior probably is indicative of a stericeffect acting to inhibit the coordination reaction.

V. Experimental

A. Polymerization of Polyorganosiloxanes. PotassiumHydroxide Polymerizations.-Powdered potassium hy-droxide 0.05y0 by weight) added to I a t 150' dissolvedslowly and, over the course of 45 minutes, catalyzed thepolymerization of th e siloxane to a soft gum. The timerequired to reach approximately the same gum stage at 175'was about 10 minutes.

Cesium Hydroxide Polymerizations.-In a typic al ex-periment, 0.03% by weight of powdered CsOH JFai rmontChemical Co.) was added to 220 g. of I a t 100 Within

about five minutes, I was polymerized to a stiff gum. Infact, the increase in viscosity was so rapid that the gumsta te was reached before all of the hydroxide could be stirre dinto solution.

Rubidium Hydro xide Polymerizations .-An amoun t ofrubidium hydroxide (Kahlbaum) equal to 0.05% by weightadded to I a t 150" caused th e polymerization of I within15 minutes to a stiff gum. I t was noted that the rate atwhich gum was formed was roughly about intermediatebetween that of the very rapid CsOH polymerization andthe slower KOH polymerization.

Effect of Lithium and Sodium Hydroxides.-Xeither ofthese materials appeared to catalyze the polymerization ofI within 16 hours at the reflux temper ature of 175". It isreported, however, that these materials do cause some poly-merization of I under more drastic conditions.ll

Potassium Methoxide Polymerization .-Potassium meth-oxide was prepared by dissolving potassium in anhydrousmethanol, The solution was evaporated t o dryness n

vucuo. A 0.1 solution of KOCH8 in I (t he compound issoluble in I a t 70') a t the reflux tempe ratu re of 175" causedthe polymerization, within about 10 minutes, of I to a verysoft gum. This gum showed no increase in viscosity on ex-tended heating, indicating that the methoxy groups prob-ably were acting as permanent chain-terminators and t husdepressing the average molecular weight.

Sodium Methoxide Polymerization.-Eimer an d Amendreagent grade sodium methoxide (0.16%) in I produced,upon refluxing for about two hours at 175", a very soft andsemi-fluid gum. Prolonged refluxing caused no apparentincrease in viscosity. This experiment was considered in-teresting, however, as an example of t he effect of solubilityon the polymerization reaction; sodium hydroxide did notreact a t all under these conditions.

Potassium Oxide Polymerization.-Potassium oxidewas prepared by burning potassium metal in air in a nickelcrucible. The material obtained was colored green, owingto th e presence of nickel, and undoub tedly contained higheroxides of potassium as well as KzO. An amount of thi smaterial equivalent to 0.02% dissolved rapidly in I a t 175'and produced a hard gum in about five minutes a t this tem-perature.

Pota ssium Amide Polymerization.-Potassium amidewas prepared by heating potassium metal in ammonia gasa t 300 , and allowing the amide to cool in ammonia. About0.17 g. of KNHz in 60 g. of I produced a gum in two to threeminutes a t 175'. It was observed that the amide dissolvedvery rapidly in the hot siloxane.

Polymerization of Tetramer in Alcohol.-Polymeri-zations of I in anhydrous methanol were made a t a numberof different concentrations an d with different bases. Al-though it was possible to obtain systems which exhibitedevidence of polymerization within a very few minutes af teradding the base to the siloxane-alcohol mixture, we usedslower acting systems to measure comparative rates withdifferent bases.

Four solutions were prepared, each of which contained13 cc. of I diluted to 25 cc. with methanol, and containing a

small amount of brom thymo l blue (soluble in methanol butnot in I) . Each solution also contained 5 X 10-4 mole ofbase, added in alcoholic solution. The bases were (a) Cs-OH, (b) RbOH, c) KOH, (d) (CH3)dN-O-CHa. Elapsedtime movies were taken of these four solutions contained instoppered 25-cc. graduates; one frame was taken per min-ute . In 15 hours a meniscus was noticed with CsOH,RbOH and th e quaternary base. This meniscus occurreda t the 19-cc. mark and appare ntly was due to the formationof a siloxane phase containing a large quan ti ty of dissolvedmethanol; the dye was distributed between the two phases.

B.

The experiment below is typical:

(11) J. F Hyde. U. S . Patent 3 400 357 December 6 1949



252 DALLASL. Hu m, ROBERT. OSTIIOFFAND MYRON . CORKIN Vvl. 76

At 22 hours, a bott om phase free of dye appea red in theRb OH solution; the volume of this phase increased from 6cc. at 22.75 hours to 14 cc. in 23 hours. The same volumeof phase free of dye formed in the CsOH solution in 35hours and a similar volume of the dye-free phase formed inthe quaternary base solution in 36 hours. I11 the KOHsolution, a meniscus was barely visible in 43 hours an d in 52hours at th e conclusion of t he experime nt, no dye-free phasehad formed. IVe can offer no good explanat ion why RbOHcaused a more rapid polymerization than did CsOH in themethanol solut ion, and this sa me effect was observed in otherexperiments in methanol.

A mixtureof 260 c c. of I , 140 cc. of methanol and 20 cc. of 0.1 iV CsOHwas placed in a stoppered flask and allowed to stand sixdays at room temperature. The silicone layer was thenseparated, washed three times with water and allowed tostand over Amberlite Resin 100H. The neutral productwas washed again with water and dried over anhydrous so-dium carbonate. A 200-cc. sample was subjected to dis-tillation. Approximately 25y0 of I was recovered un-

changed; 5 of the sample boiled from 100-120" at 47mm., 27% from 100 to 150' a t 21 mm., 8 from 120 to150" at 1mm., and the remaining 3570 consisted of materialboiling a t a temperature higher than 160 a t 1 mm. I t isobvious tha t a considerable amount of high molecular weightmaterial was formed in the room temperature reaction withcesium hydroxide.

C . Conductivity Measurements. Measurements inOctamethylcyclotetrasiloxane. .4 small beaker of I wasset on a hot-plate and equip ped with a pair of pl atin um elec-

trodes. The resistance of the liquid was measured con-tinuously with a General Electric photoelectric recorder;the circuit constants were such that a full-scale deflectionwas equivalent to 100 megohms resistance, and conductivi-ties as sma ll as those corresponding t o resistances of 2 X 10l2ohms would clearly be indicated. The applied measurin gvoltage was ca. 150 v. d.c. The resistance of I remainedpractically at infinity as th e liquid was heated to ca. 100".At this point, a small amoun t of Cs OH (0.1%) was added tothe I. No immediate change in the resistance of the mix-ture was noted, but after several minutes, as the ohydroxidedissolved and the temperature rose to ca . 120 , a veryslight increase in conductivity was observed, then a rapiddecline back to essentially infinite resistance as I polymer-ized to a stiff gum. Th e lowest resistances measured in theprocess still were thousands of megohms.

Since there was some doubt abou t th e possibility of con-duction in such a stiff gum, experime nts also were made in amedium of hexaethyldisiloxane, using the photoelectric

recorder. Whenca.

0.50% of powdered potassium hy-droxide was adde d to 20 ml. of hexaethyldisiloxane a t 130"(in test-tube fitted with Pt electrodes and immersed in con-st an t temperat ure-bath of silicone oil), a small increase inconductivity was observed immediately. The conduc-tivity increased slowly for about ten minutes to a maximumof ca . 1 X 1O-IO ohms, then decreased over an approximatelyequivalen t time to almost infinite resistance. In a similarexperiment, the addition of C C I 0 5yo f CsOH to hexaethyl-disiloxane caused a very sharp increase in conductivity;the initial rat e of change of conduct ivity with time was ap-proximately 20 times as great as tha t with KOH. The de-cline in conductivity was at about the same rate as withKOH, however. We observed tha t the addition of onetiny drop of water to the system either with KOH or withCsOH, after the resistance had assumed its original lowvalue, caused a sharp rise in the conductivity followed by aSlow decay t o essentially zero conductivit . a s the waterdistilled out of t he hot siloxane.

The very low level of conduct ivit y of the solutiolis of

KOH or CsOH in I or hexaethyldisiloxane at temperaturesof 120-150" certainly indicates t hat the intermediate metalsilanolate complexes are not appreciably ionized. A4 .5%solut ion of e ither of t hese hydroxides in anh ydrou s meth-anol, for example, exhibits relative ly high conductivi ty.

It should be pointed out that bot11 KOH and CsOH nor-mally are not completely anhydrous materials, thus somewater is present before the solubilizing reaction can begin.However , tbe amount of wa ter formed by reaction of the

One large-scale experiment was carried out.

hydroxide a nd condensation of the silanols thus formed isconsiderably in excess of the amo un t of water normall ypresent in the hydroxide. -41~0, he conductivity does notincrease appreciably until the hydroxide begins to dissolvein the hot siloxane. This argument does not affect th e con-clusion, previously drawn, that the metal silanolate com-plex is essentially un-ionized in the siloxane medium.

Measurements in Alcohol-Siloxane Systems.-The volt-age source for these resistance measurements was an audio-frequency generator with adjustable out put . The signalfrom the generator was applied to the test cell ( Pt electrodes)

through a suitable resistance (20-200 ohms dependiiigon

the conductivity of the liquid in the cell). Th e voltage de-veloped across this resistance then was rectified with a

15 34 crystal and measured on a photoelectric recorder Iinicroamp. sensitivity, but sensitivity was adjusted with rc-sistors to fit experimen t). The frequency used \vas usually2 kc. or 20 kc. and did not seem to be at all critical; i1 c

was used only to :rvoid the polarization effects tha t iverenoticed when measurements were made with d .c . inof such relatively high conductivity.made to measure absolute values of resi stance or coiiduct iv-ity, only the changes in these quantities.

All the solutions of s alts or base in alcohol wcre ~ n a d eseveral days prior t o thc m easureme nts since it was observedthat the resistance of such solutions changed somcxvhatduring the first fen. hours after mixing. Solutions iii ahso-lute ethanol were used for the most part since such sexhibit a loner change in specific resistivity withthan d o methanol solutions; the results with methaltions were approximately the same, however.

1.-10 ni l . of O.Syosolution of R bO H in EtO H was])laced in cell, input adjusted to give CQ . half-scale readingon recorder and apparatus allowed to run for a few minutesuntil recorder reading became const ant. Then 1 inl. of Iwas added. An immediate drop in conductivity to about50% of the original value occurred due to dilution a nd in-crease in solution viscosity, then a slow decline in contluc-tivity to a lower and constant value, about l: ~ of originalvalue, iii c u . 3,14 hr . The voltage then vas raised to bringreading back up-scale on the recorder. Th e addition of :t

second 1 ml. amount of I again produced ;t very rupitcline in conductivity due to dilution and viscosity incrbut no slow decline following this drop. This indicated th atthe first ml. of I had reacted with all the available RbOHin the system.

2.-This experiment was conducted under the i : m c con-ditions as experiment 1 except that a solution of CsCI iiiEtOH (ca. 0.25%) was used instead of RbOH . Additioiiof I produced immediate drop to ca. 50% conductivity due

to dilution and viscosity increase, but then recorder rc-mained constant at this value. Thus, there appcuri to hi.n o reaction of I with CsCl in Et OH .

3.-Experiments 1 and 2 were repeated with 8 d r o p of

water added to the ethanol solutions before :iddition of I .In the CsCl experiment no differences n-ere noted. Ii'ithRbOH, however, the water seemed to acceleratc the rw c -tion of the siloxane iyith the base; th e slow decline t o ( (i.

one-sixth of the original conductivity was cornplele in 20minutes. The p-stern showed the same conductivity afte rstanding 24 hours as it did a fter 20 minutes.

RbOH solution in ethanolin test cell (10 ml ,) . Then 1 ml. of hexaethyldisilosaneadded. Immediate drop in conductivity to 50 value

noted, but no further decline. This indicated that this sil-oxane did not react with RbOH t room temperLiture iiiethanol. To shoiv tha t RbOH still was active, 1 1111. of Iwas added to the \ohti oil. This produced the c a n i c s11:irl)drop and suhsequent slow decline to much lower v;llui. ovcran extended period as n as observed with I iu experinlellt 1 .

.i.-Experiineiit siiniiar to 1, but rvith hesar~iethyltiisilo\;-:nee. ilgaiii no reaction \\-as obscrvcd but r eacti ol~withadded later did occui-. Froin the evidence ob.rt-\-etI, i t

appears that trimethylsilosnnc uliits do not re:ict, ; i t le:ik t

at an y appreciable r atc , wit11 IIbOH l i c>th: i i in l I rooiu t c l i l

perature. Thus a (CII:JYSiO c l r i i t i, I I IO~( re.it.tivc u ~ ~ t i c r

these conditions than a (CHi)&O,i,- uiiit.

SCHEXECTADY.WORK

T o atternp

4.-Experiment similar to 1.

Documents

Hurd Et Al, 1954. Hidroxidos