402

JOURNALOF THE SOCIETY OF COSMETIC CHEMISTS

AEROSOL

PROPELLANTS
B.A., B.Sc.*

P. DYSON,

Presented at the Symposium on "Aerosols",organised by the Societyat Southport, Lancs.,on 25thApril 1965.
The major lroleries, advantages and disadvantagesof the established and the lesser used lrolellantsare reviewed. Some detail is also given of work done on the thermal decomlositionof halocarbons, which has a bearing
on their use in aerosols.
INTRODUCTION

MucI oF rIE informationin this paper is well-known; the objecthas been to collectit into oneplace for easy reference. In addition, it is hopedthat the paper may put into perspective someof the pros and consthat have often been brandishedrecklessly by the protagonists of one or other class of propellant. Vinyl chloridehas anaestheticproperties,but not at concentrationsremotely involved with its use of aerosols. The chloro- and chlorofluorocarbons can produce highly toxic productson decomposition in naked flamesand on red hot surfaces, but it is virtually impossible by using an aerosol to createconditions wherethis vouldpresenta hazardto the user. Althoughit is my personal opinionthat the uncontrolled useof flammable propellants coulddonothingbut irreparable harm to the industry,I feeljust as stronglythat there is no logicalcasewhatever for a complete ban on the use of a flammablegas as a constituentof an aerosol. Commercialaerosols were first producedin the U.S.A. in 1946, and in Europe and elsewhere from about 1951. Today, some1,300million containersare beingpackedeveryyear and there is no indication,evenin the U.S.A., that the market is becoming saturated. Hair sprayshave beenthe best seller in the North American market for some years; in the last year for which figureswere published,1961, some150 million units accounted for nearly 20% of American aerosolproduction. In the same year, cosmeticsas a class-hairsprays, shaving creams, colognes, perfumes and other personalproducts-comprised about one-third of U.S.A. aerosols, and a similar pattern is rapidly developingin Europe. The majority of the world's aerosols contain chlorofiuorocarbons and the safe properties and high standardof purity of thesepropellants have had muchto do with the publicacceptance of aerosols as safeand efficient commodities. Indeed,whenthe pioneers of aerosols chose the chlorofluorocarbons as propellants, they did so only after considering the properties of many other liquefiedgases;they decidedthat they alone had properties approachingcloselyto those of the ideal propellant.
*General ChemicalsDivision, Imperial Chemical Industries Ltd., Runcorn, Ches.

AEROSOL PROPELLANTS

403

However, in sucha progressive and competitiveindustry, attention will alwaysbe directedto the useof alternative propellants and diluents,either
for technical reasonsor to reduce the cost of the propellant element in
aerosols.

LIQUEFIED GAS SYSTEMS

A liquefied (orliquefiable) gasis defined officially in the United Kingdom

as one which has a boiling point below 30C but a more liberal description is "a gas which is gaseous at ordinary temperaturebut can readily be liquefiedby the application of pressure". Thus a singleliquefiedgasin a closedcontainerexerts a pressure which depends only on temperatureand is independent of the quantityof liquidpresent. A mixtureof two liquefied gasesexerts a pressure intermediatebetweenthe pressures of the two components at anytemperature. For "ideal" systems, Raoult's Law defines the total pressure of suchmixtures as:
P=m p d- m2 P2

where m and m2 are the mole fractionsof the two components p and p2 are the vapourpressures of the two components at the temperature
considered.

Mixtures of the liquefiedgases usedas propellantsfollow Raoult's Law

quite closely over the temperature rangeinvolvedin filling and in use. If then a liquefiedgas, or a mixture of two or more suchgases, is discharged as liquidfrom a container, the pressure in the container(and the composition of the mixture) will remain virtually constantuntil all the
liquid has been discharged.

(a) In two phaseaerosols the liquefiedgas propellantis completely

misciblewith the rest of the formulation,e.g. a spaceinsecticide consisting

of propellant 12/11mixtureanda concentrate basedon odourless distillate. The liquid contents are driven out by the pressure in the gasphase. The composition of the propellant/concentrate mixture and the internal pressure
of the aerosolremain virtually the same until all the contentshave been

discharged, ensuring a uniformspray performance throughout the life of

the container.

When the valve is opened the liquefiedgas,as it leavesthe spray head, expands rapidlyto produce 200 or 300 timesits volumeof gas. This effect, coupled with the actionof the valveorifices, etc.,atomises theproduct being dispensed.The degree of atomization and hence the coarseness or particle sizeof the sprayis determined both by the composition of the propellant and by its proportion in the formulation, aswell as by the structure of the valve used. Temperature is, of course,an important consideration;a productformulated to give a suitable sprayat 50-70Fwouldnot dispense

404

JOURNALOF THE SOCIETY OF COSMETICCHEMISTS

satisfactorily at temperatures below freezingand similarly it might develop an undesirablyhigh pressurefor its containerin the tropical sun. Most spacesprays,hairlacquers,colognes and surfacecoatingsare dispensed by the two-phase system,employing from say85% downto 30-40% by weight liquefiedgas prope!lantin the formulation.
(b) Threephaseaerosols containingtwo liquid and one gaseous phaseare most commonlywater-based products. Water is immiscible with all the fluorinatedhydrocarbon and hydrocarbon propellantsusedin aerosols, but it can be emulsifiedwith them to varying degreesby using emulsifying agents and by shaking the container before use. Shaving creams, foams and shampoos canvirtually be regarded as two phasesprayscontaining just sufficientpropellantto expel the contentsand expand the emulsioninto a foam. Water-based furniture polishes and starchsprayson the other hand containa small proportionof prope!lantto expel the contentsand rely to a large extent on the assistance of a mechanica! break-up valve to produce a coarsespray. Foams and water-basedpolishes of these types usually contain only 5-15% of liquefied gas propellant.

(c) More recently spacesprayssuch as air fresheners have been formulated with a different water-basedsystem which uses a higher liquefied gas propellantcontent of the order of 30-40%. Emulsifiersand shaking provide mixing of the aqueousand propellant phases. The dip tube is sometimesrestricted to limit the rate of dischargeand an orifice in the valve housing insidethe vapour phaseallowsgaseous propellantto join the liquid phaseand assistatomization. The valve is of the mechanical breakup type.

(d) Powderaerosols are a!sothree phasesystems, with the finely divided solid phasesuspended in the liquid propellantphase,e.g. talcs.
COMPRESSED GAS SYSTEMS

In this type of aerosol,the propellant is almostentirely presentas gas in the head space, usual!y about one-third of the tota! volume. As the contentsare discharged the pressure fal!s progressively. Higher pressures are thus required than with liquefied gases-for example a typical pack wou!d be chargedinitially with nitrogen gas to a pressure of 90-100 p.s.i.g. to ensurethat sufficient pressure remainedto expelthe last of the contents. Viscous products, suchas toothpastes and handcreams, are dispensed in this way with nitrogen,where the main object is to expel the contentswithout appreciable change. The cost of the propellantis negligible. Where a degreeof solubility of the propel/antin the formulationcan be achieved,compressed gasesare usedfor foam productsor, with the aid of mechanical break-up valves, for very coarsesprays. The outstandinguse

AEROSOL PROPELLANTS

405

(largely in the U.S.A.) is of nitrous oxide/carbondioxide mixtures for dispensing whippedcreamtoppings. Other food useshave beenslowto follow. Nitrous oxide is occasionally used in shaving creams, and carbon dioxide in de-icingand similar coarsesprays; both have been usedin motor starter

sprays in the U.S.A. but it is unlikelythat the useof compressed gases will increase for applications whereany real degree of atomization is required. Some"mother-and-daughter" packsusenitrogento boostthe pressure of the largerunit and so facilitatetransferof its contents to the handbag
container.

A comprehensive accountof compressed gas propellantsfor non-food productsis given by Webster 1.

CItLOROFLUOROCARBON PROPELLANTS

The useof dichlorodifluoromethane as a hazard-free refrigerant wasfirst developed in the U.S.A. in 1931. Other fiuorinatedhydrocarbons followed as the refrigeration and air-conditioning industries expanded. Thesecompounds are now made in countriesthroughout the world to the extent of over 300,000 tons a year and for someyears the demand for use in aerosols has exceeded that for refrigeration. They are manufactured in the United Kingdom under the trade names "Arcton" and "Isceon".

The outstanding advantages of this class of liquefied gaspropel!ant are their freedom from hazardand their high degree of stabilityand chemical inertness. The liquidsare colourless, giving colourless yapourswith faint and not unpleasant odours. They do not damagefurnishings or fabrics. Someof the physical properties of the threepropellants in mostcommon use
(12, 11 and 114) are given in Table 1.

The versatility of these threepropellants is almost unlimited. A propellant can be madeto measure for almostevery non-food application by selecting the right compound or mixture. The mostwidely usedsource of pressure is propeRant 18; it exertsa pressure of about70 p.s.i.g.at 21C (70F)and is thusonly usedalonewherethe overallpressure is reduced by the other components of the formulation, e.g. in aerosol paints. More commonly it is mixedwith propeRant 11 to reduce the pressure; this also increases the solvent powerof the propel!ant withoutthe risk of damage to gaskets, etc. Throughoutthe world the most widely suppliedmixture is propellant12/11 50/50--equalweightsof 12 and 11--with a pressure of about37 p.s.i.g.at 21C(70F). It is the basis of many space insecticides,
air freshcriers and hair lacquers.

Althoughthe chlorofiuorocarbons are extremelystable,the most asymmetric moleculeof the three, propellant11 (CClaF) is lessresistantto
hydrolysisthan the others and can alsoreact under certain conditionswith

the loweralcohols.In addition, it can affectcertainclasses of perfume

406

JOURNALOF THE SOCIETYOF COSMETIC CHEMISTS

AEROSOL PROPELLANTS

407

adversely. Thus for water-based productsand a number of cosmetics, such as colognes and perfumes,propellant 12 is used alone or in mixtures with propellant114 which is characterizedby its outstandingstability and is practically odourless. Propellant 12/114 mixtures in proportionssuch as 10/90, 20/80 and 40/60 are commonlyused. The gaugepressures of a range of 12/11 and 12/114 mixtures are given
in Table 2.
Table 2

Vapour pressures of propellant mixtures at 21C (70F.)

Propellant 12/11 mixtures 35/65

Gauge pressure p.s.i.g. 27

Propellant 12/114 mixtures 10/90

Gauge pressure p.s.i.g. 20

40/60
50/50
60/40

30
37
44

20/80
30/70
40/60

27
34
40

65/35

47

50/50

46

Flammability The propellants are non-flammable and have flame suppressantproperties; indeed,it is sometimes possible to formulatenon-flammable products (judged by, say, the flame-projectiontest) containingflammable solvents, by incorporatingchlorofluorocarbon propel/ants. The propel/ants do not form explosivemixtures with air in any proportion. Toxicity The (U.S.) Underwriters LaboratoriesClassifications for the vapours of propellants12, 11 and 114 are Group6, 5a and 6 respectively. Group 6 is defined as "Gasesor vapours which in concentrations up to at least about 20% by volume for durationsof exposureof the order of 2 hours do not appearto produceinjury", i.e. virtually free from toxicity under all normal conditions of use. For comparison purposes, carbondioxide is classified in Group5a and hasan MAC (maximumallowableconcentration for continuous working) of 5,000 ppm (0'55o) by volume.

ThermalStability Propel/ants11, 12 and 114 begin to decompose in contact with open flamesor red-hot surfaces to form acidicproductsand sometimes tracesof phosgene,which would be hazardous if inhaled in sufficient quantity. Fortunately, the sharp, acrid odour of the acidic productsgives sufficient

408

JOURNAL OF THE SOCIETY OF COSMETICCHEMISTS

warningof their presence. Nevertheless, work areas nearfillingplant should be adequatelyventilated when, for example,weldingis in progress and heating installations should avoid red hot surfacesor naked flames. From the aerosol userpoint of view, it is considered that the quantities involvedmake the possibility of hazard from suchdecomposition extremely slight and this is borneout by the trouble-freerecordof fluorocarbon propellants and refrigerantsin this respect over many years. A number of cases wherepropellant12 was cited s as a cause of poisoning due to thermal decomposition were discussed by Downingand Madinabeitia ain 1960. They pointed out that the possibility of decomposition of the fluorinated compoundscouldnot be ignored,and that there was no intentionof minimizing

thetoxicity of some oftheproducts formed; but theyconcluded that there

was insufficient evidence to implicate propellant 12. Quite independently,work has been carried out in our laboratoriesto simulate the use of three types of domestic convectorheater in a small unventilatedroom containingthe vapour of a number of halocarbons. An atmosphere maintainedat a constant composition was fed to the intake ports of the heatersand the effluent air analysedfor breakdownproducts. The
convection rate of each heater was measured. The results obtained with

an atmospherecontaining1,000 ppm (0.1%) by volume have been usedto estimatethe concentrations of toxic breakdownproducts which couldresult after 10 minutesin a smallsealed roomof 1,000ft.a (28 ma). Theseestimates are given in Table 3.
Consider first the estimated concentrations from the most toxic de-

compositionproduct, phosgene, which has a maximum allowable concentration (MAC) for continuous working of 1 ppm by volume.*

(i) With one exception, phosgene was not detectedin the effluentgases from the electricand gas convectorheaters,but the paraffin heater favoured its formation. Only with methylene chloride was the
M.A.C. reached or exceeded.

(ii) With the paraffinheater, 5.7 oz of propellant 11 in the atmosphere produced about the same concentrationof phosgeneas 2.6 oz of vinyl chloride. Propellant 12 producednone and propellant 114 a negligible amount.
*The M.A.C. is the maximum average atmospheric concentration of contaminants to which persons may be

exposedfor an 8-hour working day without injury to health. Thesevalues are basedon the best available information from industrial experience, from experimental studies and when possible,from a combination of the two. They should be used as guides in the control of health hazards and should not be regarded as fine lines between safe and dangerousconcentrations. They represent only conditionsunder which it is felt that workers may be repeatedly exposed,day after day, without adverseeffect on their health. The figureslisted refer to weighted
average concentrations of an 8-hour working period rather than a maximum which is not to be exceeded even

momentarily. M.A.C. figuresare publishedby the American Conference of GovernmentalIndustrial Hygienists and are reviewed annually. They are usedby the International Labour Office,and have alsobeen usedsince1960 as a basis for the "maximum permissibleconcentrations"publishedin the United Kingdom by the Ministry of
Labour.

AEROSOL PROPELLANTS

409

410

JOURNALOF THE SOCIETY OF COSMETICCHEMISTS

(iii) In all caseswhere phosgenewas produced, appreciably greater concentrations of the acidic gaseswere formed, confirmingthat these would have given adequate warning. All the compounds tested have been usedin aerosols and in most cases the quantities involved could conceivably be discharged from standard aerosol containers in the home. Under certain circumstances toxic products are undoubtedly produced with all the compounds, but considerationof (i)-(iii) above,and the fact that deliberatelyunfavourable conditions were created, suggests that this would not constitute a hazard to health.
CHLOROCARBON PROPELLANTS

MethyleneChloride(CH2Cl2) Methylene chlorideis a non-flammableliquid boiling at 40.2C. It is one of the least toxic of the chlorinatedsolvents(MAC=500 ppm v/v, Underwriters Laboratories Group 4-5). Although it has an anaesthetic effectif breathedat high concentrations, the quantitiesinvolved in aerosols do not constitutea real user hazard in this respect. It is usedin aerosols in its own right as a so!vent,to depressthe flammability of other solvents,as a co-solvent to increase the solubility of some ingredientsin propellantsand as a cheapersubstitutefor part of the propellant 11 in 12/11 mixtures. As a pressuredepressant for propellant 12 its effect is very similar to that of propellant 11. It has, however, a number of technical disadvantages which limit its use in aerosols. It hydrolysesmore readily than propel!ant 11 which can increasecorrosion problems. It has a more definedsmell and this, coupled with the higherhydrolysisrate, can seriously affect fragrances and perfumes usedin aerosols. It has a marked solventand swellingeffect on elastomers 5, and resistant materials must be used in valves, etc. Its strong solvent
action also occasions caution when used in household aerosols; we have

found that the misuseof an aerosolcontainingmore than 15% methylene chloride,by applying from a short distance,can damage synthetic fibres such as acetate and triacetate rayons. Although these disadvantageslimit its range of applications and the proportionwhich may be used,methylenechloridehas already established itself as a valuable raw material in the aerosolindustry. 1.1.1-Trichloroethane (CCl.CH) Stabilisedgradesof this non-flammablesolventare sold under the trade names"Genklene"(ICI) and "Chlorothene" NU (Dow). 1.1.1-trichloroethane

hasa similarlowtoxicityto that of methylene chloride (MAC=500 ppm,v/v) but its higherboilingpoint (ca.76C) makesit a rather safersolventfor cold
cleaning. It is believed, in general, to have less effect on the elastomers

AEROSOL PROPELLANTS

411

used in valves and little or no effect on textiles, but this apart, it sharesthe disadvantages of methylene chloridelisted above. Its pressurecharacteristicsin mixtures with propellant 12 are very closeto thoseof propellant 11. It is used as a solvent in aerosols(e.g. for cold cleaning,spotting) and to reduce the flammability of other solvents. Appreciably more costly than

methylene chloride, it is not likely to be considered solely as a cheaper substitute for propellant 11.
Vinyl Chloride (CH:CHC1) Vinyl chlorideis a liquefiedgas,boiling at -13.9C. Its vapour pressure is very similar to that of propellant 12/11 mixture. It has an MAC of 500

ppm v/v and, althoughnot classified by the UnderwritersLaboratories, its toxicity would appear to place it between Groups 4 and 5. At higher concentrations it has an anaestheticeffect but the quantities which might be used in aerosolsare never likely to create a user hazard in this respect. Vinyl chloride is cheaperthan the fluorocarbonpropellants,is a stronger solvent and has a higher expansionratio on evaporation. On the debit side, it is highly flammable and forms explosivemixtures with air in the range 4-22% by volume. It has a marked swellingeffect on elastomers and possesses a pronounced odour. It is manufacturedalmost exclusively for polymerizationto PVC and for the productionof allied co-polymers; publishedwork6 has recommended that it shouldcontain a stabiliserif used
in aerosols.

To the best of my knowledgevinyl chloride has not been used commerciallyin aerosols in the United Kingdom. In a few Europeancountries, and in Japan,it hasbeenusedasa primary propellantand two manufacturers in the U.S.A. offer blendsof propellant 12 or 12/11 and VC. One of these firms6 reports that vapour mixtures of vinyl chlorideand propellant 12/11 50/50 are non-flammable if the VC contentis below45% v/v, i.e. 29.4% on a liquid weight basis; becauseof fractionation the actual amount of VC that can be toleratedis 22/0 w/w, hencetheir blend is offeredas 78% 12/11 50/50+22% VC.
HYDROCARBON PROPELLANTS

Propane(Calla),n-and isobutane(CH0)are liquefiedgases. They are virtually non-toxic, being classified by the Underwriters Laboratoriesin Group5b. Commercial grades vary in composition according to source, but mixtures of two or more of the hydrocarbons provide pressures suitable for aerosol dispensing. For example, a mixtureof 50% nbutane, 25% isobutane and 25% propanegivesabout the samepressure as propellant12/11 50/50. The hydrocarbon propellants are low-priced. They are immiscible with, and have a goodstabilityin, water and, unlike the fluorocarbon propellants,

412

JOURNAL OF THE SOCIETY OF COSMETICCHEMISTS

have densities lessthan 1. Theseproperties all favour the hydrocarbons for water-based applications suchas iurniture polishes, wheresmall proportionsof propellantare usedin three phasesystems.

The majordisadvantage of the hydrocarbons is their highflammability

and the fact that they form explosive mixtureswith air at quite low levels

(<2% by volume). TheU.S.Bureau of Mines 7found that theflammability of

purenbutanecouldbe suppressed by mixingit with three timesits volumeof prope!!ant 12. Work whichwe carriedout with commercial butanehasshown that about the same figure appliesfor propellant12/11 mixture; 3:1 by volume is equivalent to about 13% butane by weight in the mixture. A second serious disadvantage is that the hydrocarbons often presentodour problemsdue to impurities which are sometimes difficult to remove from containers,etc., as well as affectingfilled aerosols.
I am not aware of any non-aqueousaerosolsmarketed in the United Kingdom with hydrocarbons as the only propellant. Such aerosols are, however, packedin Franceand a numberof Europeanfillersusea proportion of hydrocarbon with fluorocarbon propel!ants to reducecost. In the U.S.A. one manufacturermarkets two blends: 12/11/isobutane45/45/10, and 12/114/butane9.1/82.4/8.5--this last mixture being offeredfor perfume aerosols. Propane is occasionally usedin the U.S.A. mixedwith propellant 11 or methylene chloride to depress its pressure and suppress flammability.

As yet, few of the water-based spacesprayscontaining larger amounts of hydrocarbon propellant(Liquefied GasSystem(c) above)have appeared on the United Kingdom market and it is too early to say whether their
performancewill equal that of the established spacesprays or how much the extra cost of valve, lacquered container and formulation work will be outweighed by savingsin solventand propellant,also the extent to which specialstorageand transport conditions will be necessary in the distributive
chain.
FLAMMABLE VERSUS NON-FLAMMABLE PROPELLANTS

A paper on the flammability of propellants is being read at this Symposium,so that I have deliberatelyavoidedenteringthe lists and starting
what our American friends call a "Hassle". National Aerosol Associations,

Governmentbodies, fillers and propellant manufacturersthroughout the world are debating what should or shouldnot be done to ensurethe safe transportof aerosols and the protectionof the user. It is foolishto emulate the ostrich,but it is equally important to ensurethat the goodrecordof aerosols over many years is not endangered by departingtoo rapidly or radically from well tried practices.

(Received: 1!tthMarch 11165)

AEROSOL PROPELLANTS
REFERENCES

413

Webster, R.C. Aerosol Age 6 20 (June 1961) Nord. Hyg. ridskr. 39 11 (1958)

Downing, R. C. and Madinabeitia, D. AerosolAge $ 25 (September1960) Toxic Substances in Factory AtmospheresMinistry of Labour, Safety, Health and Welfare New Series, No. 8 (March 1960) (H.M. Stationery Office, London) Brown, J. A. Proc. Chem. Specialties3lifTs.Assoc.46th Annual Meeting 80 (1959)
Scott, R. J. and Terrill, R. R. Aerosol Age 7 18 (January 1962)

Jones, G. W. and Scott,F.E.

Report RI $908 (1946)(U.S. Bureauof Mines)

I)ISCUSSION

MR.C. BLOOM: Propellants142band 152a have beenclaimedto be good solventsfor certain drugs and they might, therefore, serve a quite useful role in the cosmeticand pharmaceuticalfields. I)o they have any serious disadvantages and what progress, if any, has been made towards their commercial exploitation? THE LECTURER: Propellants152a (CH3.CHF2)and 142b (CH3.CC1F2) boil at -24C. and -9.4C. respectively and have the relatively low densities of 0.91 and 1.12 g/cc at 21C. I believethat 152awas first manufactured in the U.S.A. for use as a refrigerantin an azeotropic mixture with 12, designated refrigerant500. To the best of my knowledge propellants152a and 142b are only offeredin commercialquantities by one American supplier who claims that 152a has unique solvent characteristics and recommends 142b for cosmetics suchas perfumes,either alone or in mixtures with 114. It is noteworthythat both propellants can form flammablemixturesill air (152a: 5.1-17.1%, 142b: 9.0-14.8% v/v). There has beenno demandfor thesepropellants in the United Kingdomand the high costassociated with small scaleproductionwould only warrant their adoption where they had outstandingadvantages.
MR. A. I-IERZKA: IS it likely that your companywill supply,in the near future, propellantblendssimilarto the two listed at the end of your paper?
THE LECTURER: NO.

MR. A. HERZKA:You refer to odour problemswhen usinghydrocarbon propel!ants. Is it not possible for suchmaterialsto be suppliedfree from
obnoxiousodours, as is the case in the U.S.A.?

THE LECTURER: My company does not supply, or use, unstenched hydrocarbons for aerosols, but it is my understanding from fillersthat such hydrocarbons suppliedon the United Kingdom market still have some residual odour. Perhaps another participant would care to comment
further.

MR. I). S. RANDALL: In destenched butane there is a residual odour, but we

consider our products to be almostcomparable with Americanand European

414

JOURNALOF THE SOCIETY OF COSMETIC CHEMISTS

hydrocarbons, whichwe havetested. Work continues on this problem,and we hopeto have thoroughly acceptable products in the near future.
MR. H. F. FROST: I assume,from the nature of the resultsin Table $, that the gas convector heater usedin the experimentwas not the modern type in which the combustion productsgo up the flue and hot air only is circulated,but the older free-standing type where all the combustion products were released into the room. THE LECTURER: This is true. is this so? The convector heaters used in these

experiments were standarddomestic appliances which rely on convection of air andwhichwerenot fittedwith fluesor chimneys.
DR. P. H. WITJENS: Are any quantitative data known on the formation of COC12 from CH2Ci vapour in the presence of a radiant bathroom heater (e.g. metal coated ceramicheating bar)? This problem is of significance in Holland as some new buildings have small bathroomcubicles so heated,with poor ventilation,wherehair sprayscontaining methylene chloridemay be used.

THE LECTURER: i know of no data referringto thesespecific conditions. I would think, however, that sincethe quantity of propel/ant,or methylene chloride,involved in one applicationto a woman'shair would be small, the concentration of phosgene produced would be correspondingly minute.

DR. H. K/JBLER:The toxicity values for vinyl chlorideand 1,1,1-trichloroethane in Table1 do not agree with those I have quoted. i must
admit that I did not determine the value for the latter, but I know that the

vinyl chloridevalue is proven. THE LECTURER: I presumeyou are referring to the U.S. Underwriters Laboratory's classification groups for vinyl chloride and 1,1,1-trichloroethane. To the bestof my knoMedgeneither compound hasbeenclassified by the Underwriters Laboratoriesand the figuresI quoted were estimated
from a review of available data.

The data on vinyl chlorideis somewhatconflicting,e.g. the findingsof Torkelson,which do not agreewith thoseof the Battelle Institute, quoted by you. Toxicity data on the 1,1,1-trichloroethane is limited. However, because of the anaesthetic propertiesof vinyl chlorideand bearingin mind the fact that the current MAC figure for each compound is 500 ppm, as for methylenechloride,it seemed reasonable to me to suggest that both would fall within the sameUnderwriters Laboratories Groupas methylenechloride, i.e. Group4-5.
DR. H. K/JBLER: You are quoting estimated concentrations of decompositionproductsof halocarbons, in Table 3, whereasthe figuresquoted by

AEROSOL PROPELLANTS

415

me are exact. I consider temperatureto be decisivefor decomposition, which must be observedonly above 500C.
TaE LECTURER: I would like to assure Dr. Kiibler that the actual

measurement of the decomposition productsleaving the convector heaters was carried out very preciselyand was not "estimated". Phosgene, for example, was measured by individually calibratedDrtger tubes and by absorption in aqueous anilineto form diphenylurea. I haveusedthe word estimatedonly in so far as I have interpretedtheseresultsin terms of an atmosphere in a sealed room,i.e. attemptedto relate the scientific teststo extreme conditions under which an aerosolmight be discharged. Temperature is indeedan important factor but other effectssuchas humidity and metal surfaces are alsoimportant sofar as decomposition is concerned. }V[R.J. C. THORNTON: With regard to the thermal decomposition of halocarbons detailed in Table $ of your paper, were the temperaturesof the
elements in the three different convector heaters measured?

TE LECTURER: NO. The experiments were carriedout with standard domestic appliancesto give realistic conditionsrather than to obtain
additional academic information.

1VtR. H. F. FROST:Have you examined the decomposition of propellantsdrawnthrougha lightedcigarettein the "domestic" experiments?
TE LECTUre;R: No. We have, however, carried out work on the

problem of smoking cigarettes in atmospherescontaining the vapour of chlorinatedhydrocarbons and the resultswere published by Little [Brit. J. Ind. Med. 12 304 (1955)]. In no casewas phosgene found in the vicinity of the glowingtip of a cigarettesmoked in the contaminated atmospheres, and evenwhena traceof phosgene wasaddedto the atmosphere, it wasdestroyed by passage throughthe cigarettebeing smoked.