Tenacity and Fixing of Aromatic Chemicals

By WOLFGANGSTURM and GERD MANSFEIJJHammann & Reimer GmbH

The qwdity of a perfume oi~ is determined byseveral factors. So far as its acceptance by tbe con-sumer is concerned, decisive importance attachesnot only to the actual odor characteristics and sta-bility under the particular conditions of use, butalso to the odor intensity and tenacity.

Experienced perfumers have long made use intheir compositions of so-called fixatives, which can,to a greater or lesser degree, increase the period forwhich tbe odor of a perfume oil or specific compo-nents of the composition remain perceptible whenused,

The experiments here described should make acontribution of practical value to tbe elucidation ofthe tenacity and fixative characteristics of somecomponents commonly used in perfume oils. It wasinitially attempted to determine regularities in thebehavior of aromatic chemicals in relatiou to theirtenacity as well as to detect facto~s through whichthis tenacity could be influenced. In addition, ananalysis was made of the fixative properties of cer-tain substances generally used for this purpose. Onthe basis of the results obtained, an attempt wasalso made to determine an inherent law governingtbe fixative action of these substances.

Tenacity of homogeneousaromatic chemicals

These experiments were carried out chiefly onthose classes of substance most frequently encoun-tered in perfumery, such as hydrocarbons, alcohols,phenols, ethers, aldehydes, ketones, and esters orlactones.

To obtain precise, reproducible results, tbe tenac-ity of aromatic chemicals was determined by mea-surement of the weight loss. In most cases, wherethe weight loss progressed continuously, the gravi-metric result was identical with the tenacity deter-mined purely by odor. This comparison will be dis-cussed in greater detail later on.

A piece of bibulous paper with a substance ofabout 264 g/m2 was used as tbe support. 0.6 g ofthe compound to be tested was slowly dripped ontoa strip of tbe paper 2 g in weight and 75.84 cmz inarea (15.6 cm x 4.8 cm). In all cases this amount ofcompound was sufficient to completely saturate thepaper with liquid, so that tbe surface area was tbesame in all the experiments. In the case of viscoussubstances, tbe time needed to saturate the paperuniformly was considerably longer than in tbe case

6/Perfumer and Flavorist

of aromatic chemicals of lower viscosity. The papersprepared in this way were freely suspended in aroom at about 20” C with ordinary ventilation andhumidity, so that evaporation could proceed uni-formly on all sides.

The weighings were usually carried out after 1,3, and 10 hours; 1, 3, and 10 days; 1 month and 3months, using an automatic balance that could beread to an accuracy of 10 mg. In certain roses itwas necessary to make weighings at times inter-mediate to those scheduled,

Continuous weight loss, The ~esults obtained hadto be divided into two groups, one consisting ofaromatic chemicals whose weight loss was continu-ous and the other consisting of those whose weightloss was discontinuous, Table I lists the membersof the first group, together with tbe tenacities of theindividual aromatic chemicals.

Figures 1-3 show the change with time of theweights lost by some substances, selected from T;~-ble 1. Tbe graphs show the amount of odorant stillpresent at the time of each wcigbing, The linearforms of these curves show clearly the uniform ratesof weight loss, or uniform progress of evaporation,

Of interest and practical importance are com-parisons of the times for which aromatic chemicalsof different stmcture classes and molecular sizes areretained on tbe paper. For example, the two C,,,hydrocarbons, ocimene and dipentene, were re-tained for only ca. 1 hour, whereas the C,” alcohols,Iinalool, tetrahydrolinalool, dihydromyrcenol, gera-niol, citronellol, and isopulegol, took from 4 hoursto 3 days to completely evaporate from the paperstrips (see Table I and Fig. 1). It was thus imme-diately apparent that the functional groups have alarge influence on the tenacity of aromatic chemi-cals, just as they do on that of dyestuffs. Moreover,the tenacity is dependent not only on the kind offunctional group (e.g., alcohol, aldehyde, or ester)but also on the molecular configuration adjacent toa specific functional group (e.g., primary, second-ary, or tertiary alcohol). The above-mentioned ter-tiary alcohols, linalool, tetrahydrolinalool, and dihy-dromyrcenol, required 4 to 6 hours to evaporateunder the experimental conditions; the secondaryalcohol isopulegol took ca. 8 hours; and the primaryalcohols, geraniol and citronellol, took between 2and 3 days.

Vol. 1, February 1976

T +

—tt — —

I — — — —t

‘t

fig. 2: Progress with time of the evaporation ❑f arorn.licchemicals(UPto 30 days),

geranyl methyl ether decreases the tenacity from2days to 4 hours, Esterification to geranyl formatelikewise brings about a reduction in the tenacity, to13 hours, If now the size of the acid radical is in-creased, the tenacity increases from geranyl formateto acetate to butyrate, with times of 13 hours, 28hours, andca,3 days (see Fig. 1). The series citro-nellol (3 days), citronellyl acetate (ea. I day), ci-tronellyl propionate (ea. 3 days) displays a similarrelationship (Table I) and so also do eugenol (ea.2 days), eugenol methyl ether (ea. 4 days), andeugenyl acetate (ca, 1 month), (see Fig. 2).

There is likewise a clear parallelism of tenacityand molecular weight in the case of the formatesstudied. With increasing molecular size, the te-nacity increases steadily from hexenyl formate (ea.1 hour) via benzyl formate (ea. 2 hours), phenylethyl formate (ea. 3 hours), geranyl formate (ea. 12hours), to decalinol formate ( ca. 1 day). (See TableT). ..

The introduction of an additional functionalgroup can considerably retard the evaporation of anodorant and thus prolong its tenacity (cf. ethanoland ethylene glycol). Thus, for example, the twostructurally similar compounds dihydrojasmone and2-hexyl.cycIopentanone require about 1% and 3days respectively for complete evaporation, whereasmethyl dihydroiasmonate is retained for over 3monihs (see”Fig: 2).

All thcperfumery ’’fixatives’’ that were examinedshowed a tenacity of atleast3 months in the experi-ments made. Nevertheless. the individual resultswere very different. Diethyl phthalate was almosttotally evaporated after 3 months, but about 30% ofthe isopropyl myristate and musk ambrette werestill present. About 80% of the Iinalylbenzoate hadmmporated, whereas tbe weight of the benzyl ben-zoate had only decreased by 25% (see Fig. 3).

In practically all the experiments, the gravimet-ric and the odorous determinations of the tenacitygave identical results in the case of aromatic com-pounds whose evaporation progressed at a uniformrate, so that the present results provide data thatcanalso reutilized in perfume oil compounding.

8/Perfumer and Flavorist

fig. 3: Progress with time of the wap.ar.ti.n of aromaticchemicals(up to 3 months).

The relationship found between tenacity andstructure class, functional group, and molecularweight of the aromatic chemicals shows a regularbehavior of these substances and thus allows pre-diction of their tenacity inthe rational synthesis ofnew odorants, The parallelism between tenacityand molecular weight can, however, be primarilyattributed to connectiona between, on the one hand,the rate of evaporation of an aromatic chemical and,on the other, specific molecular properties, such asvapor pressure, boiling point, polarity, density, andviscosity

One of aperfumer’s skills is the ability to souseodorants with similar smell but different tenacitiesthat a particular fragrance characteristics empha-sized throughout a prolonged period of evaporationof a perfume oil. Examples of these are the pre-viously mentioned genmiol derivatives and the se-ries of compounds, hexyl methyl ether, hexanol,hexyl acetate, and hexyl tiglinate.Nonuniform rate of weight loss. The behavior ofthose odorants that evaporate at nonuniform rates isconsiderably less predictable and systematic. Intbese cases, either amolecular alteration takes placeduring evaporation, or the substances tested arenot homogeneous but consist of mixtures of variouscomponents.

It was remarkable that none of the aldehydestested evaporated at a uniform rate. In the course ofthese investigations. noodorant containing an alde-hydic group-was found whose evaporatio~ rate didnot diminish with increasing time.

In correspondence with this observation, hepta-nal and decanal, respectively, decreased in weightby ca. 70% and ca. 40% in the first 3 hours, yet hadnot completely evaporated after3 months (see Figs.4 and 6). Both substances were olfactorily percep-tible for about 3 days. By means of NMR spec-troscopyit was found that, under the experimentalconditions, after about 2days the residues consistedpredominantly of the corresponding carboxylicacids, as was also manifested by a change in theodor. Under the particular conditions, atmosphericoxidation of the aldehydes commenced as soon as

v.I. 1, February 1976

$0.—

2t/i(l, o12fi(6 ,s 2022,’,!,,,, (,,,,,,, )

fro. 4 w,<..I;...us evaporation of .ad.arants(up to 24ho”,,].

the experiment started. The relatively long tenaci-ties of up to 3 days, which were determined hyolfactory assessment, can be explained as due tothe considerably greater tenacities of the carboxylicacids formed. These are therefore exceptionallygood fixatives for the aldchydes. In addition, be-cause of the great intensity of the odors of heptawdand deczmal, extremely small amounts of these sub-stances can he clearly perceived.

Citral behaved in a similar fashion, although inthis cwe almost 90% of the applied weight of ma.terial had already evaporated in the first 24 hours(see Fig. 4). The residual 10% consisted chiefly ofgertmic acid. In conformity with this finding, the te-nacity of citral by olfactory determination wasabout 1 day.

Alpha-amyl cinnamaldehyde was odoromly per-ceptible for shout 10 days. After 10 days the gravi-metric measurements showed that shout 7070 of theoriginal weight applied ww still retained hy the pa-per ( sw Fig. 5). The residual small amount ofa-amyl cinnamaldehyde was so greatly reduced inodor intensity by the carboxylic acid formed that itcould scarcely he smelled, The a-amyl cinnamicacid formed hy oxidation was retained for about 21Amonths (see Fig. 6).

The two bomolugues, cyclamen aldehyde andp- tert -butyl a -methyl -hydrocinnamaldehyde, re-vealed practically no loss in weight over a period of3 months, so that their evaporation could he de-scribed as approximately continuous (see Figs. 5and 6). Odorously, cyclamen aldehyde can be de-tected for up to 10 days, while some outstandinglysensitive noses could identify p-tert-huty l-a-methyl-hydmcinnamaldehyde even up to 3 months. In bothcases, after only a few days, the amount of carboxy -Iic acid was greater than that of aldehyde, thougheven after 3 months more than 30% of the originalamounts of aldehydc were still present. The readyoxidizahility of p-tert-butyl-a-methyl-hydmcinna-maldehy de, in particular, is well known to everyexperienced perfumer, since there is scarcely onebottle containing this odorant that does not have

fig. 5 bs.o.finw.s evaporation .+ .dora.fs (.P t. 30days).

,0

9

8. ,,,,

“ IQ~,o. ““. ,,, ,,, ,,, ,,, ,,, ,,, ,,, .,

2 60

: 5,,0-30-20.

,,,!, (,”O”,,, )

fig, 6: Dix.a.tim..us e.ap.ration of odore.fs (.P t. 3month,).

numerous colorless crystals around the joint of theclosure, consisting simply of the corresponding car-boxylic acid. Apparently the crystals of the acidocclude some aldehyde, so that even after monthssome traces of ddehydc are released to the sur-rounding atmosphere.

Even though these experimental results were pre-cise and plausible, a systematic basis for the predic-tion of the tenacity of aldehydes can scarcdy bediscerned, since in the case of aldehydes the tenac-ity is dependent not only on the compound itself,but also on the rapidity of oxidation, the tenacity ofthe carboxylic acid formed, and the fixative abilityof this acid for the aldehyde.

Nonuniform rates of evaporation were also ob-served, in isolated cases, among odorants that didnot have an aldehydic structure. p-Methyl quino-line, for example could he readily recognized byodor for 1 month, although 97% of this compoundhad evaporated within 3 days ( see Figs. 4 and 5).The residual 3% retained a constant weight forITIOR than 1 month, excluding the possibility ofweighing error. The polymerized residues formed

Vol. 1, February 1976 Perfumer and Flavorist/9

in the case of this and other compounds with apyridine structure apparently have the ability tostrongly fix residues of the readily volatile odorantsfor a prolonged period.

Within 10 days, about 90% of a-ionone had evap-orated, the violet odor also being perceptible up tothat time. The residual 10% remained for over amonth (see Fig. 5). Interestingly, as the time theionone had been retained on the paper increased, anot insignificant increase in the peroxide numberwas detectable, suggesting that the residue left after10 days was, at least in part, a peroxidic derivativeor condensation product of ionone. This suppositionwas also supported by an odor reminiscent of de-composing substances,

Tbe small, but reproducibly determinable, devia-tion of 5-phenyl-3-methyl-pent-2-en-l acid nitrilefrom a uniform evaporation rate (see Fig. 5) can beattributed to the fact that this odcmmt was z mix-ture of the cis. and trans-isomers. The lower-boil-ing cis-compounds evaporated preferentially withintbe first day, so that the content of the higher boil-ing trans-isomer was progressively inc~cased andthe rate of evaporating reduced. The alteration oftbe cis:trans ratio as evaporation pmgrmscd couldbe followed by the slight change in the odor. Thetenacity of 5-phenyl-3-methyl-pent-2 -en-l acid ni-trile was about l% months. in good agreement withthe value determined by olfaction.

The related compounds with extremely weakodors used as fixatives, methyl abietatc and Abitrd,are mixtures and in consequence their evaporationrates were far from uniform, especially in the initialphase. The methyl ester lost about 10% of its initialweight within 3 hours and then remained at con-stant weight over the whole 3-month period of thetest (see Figs. 4-6). Abitol, in contrast, evaporated18% of its weight in the first day, the weight 10SSin-creasing to 3,5% within the first month, its residualweight thereafter remaining approximately con-stant (Figs. 5 and 6). In agreement with these find-ings, the tenacity of these substances, as determinedby olfaction, was more than 3 months.

The amber body well-known m “Fixateur 404’”lost weight at a uniform rate, after the first fewhours (see Figs. 4 and 6). After 3 months, more than75% of the original weight of substance was stillpresent. According to tbe olfactory determination,the tenacity was about 2 months. These findings be-come comprehensible if it is taken into account thatthe odorous component, amber epoxide, is presentasasolution invarious almost odorless fixatives, andin addition, to round out the odor complex, thereare a few percent of readily volatile odorants.

Geranyl benzoate should behave like linzdyl ben-zoate (see Fig. 3). Remarkably, however, the gera-nyl benzoate lost about 259. of its weight in thefirst 3 days and thereafter evaporated only veryslowly, so that about 70% of the initial amount wmstill present after3 months (see Fig. 5). The olfac-torily determined tenacity of geranyl bcnzoate wasabout 10 days. Analysis of the sample used showedthat the material declared as geranyl benzoate con-tained about 227. of gcraniol. The rapid loss in

weight during the first few days was therefore al-most exclusively due to the evaporation of geraniol,The odorous assessment also related to geraniol,Pure geranyl benzoate bas a tenacity far greaterthan 3 months and is practically odorless,

In the main, two causes for the nonuniform mtcof evaporation of odorants were found: the modifi-cation of the molecule by the surrounding atmos-phere and the presence of mixtures,

Atmospheric oxygen oxidized, to a greater or less-er extent, all the aldehydes studied. Oxygen was al-so responsible for molecular alterations in othercases, e.g., the polymerization of quinoline and pe-mxidc formation by ionone. In practice, these oxida-tirm reactions can to some extent be considerablyretarded hy an optimal addition of suitable antioxi.dants.

A nonuniform rate of evaporation was naturallyrevealed by mixtures of components with differentvapor pressures. Bymeasuring thechange in weightloss with time, one can readily detect the presenceof mixtures of substances, such m geometric iso-mers and “up-graded” odorants as well as such im-purities as starting materials and residues.

Fixation of homogeneous odorants

In order to obtain specific effects in perfume com-positions, it is often necessary to increase the tenaci-ties of odorants, to a greater or lesser extent, beyondthe tenacities indicated by experiments such asthose described. For this purpose, perfumers havelong made successful use of so-called fixatives, aninvestigation of whose effect on the evaporation ofhomogeneous aromatic chemicals will he describedin some detail in the following. The experimentswere carried out using different aromatic chemicals,to each of which 10% of a fixative was added. Thefixatives usedwere: methyl ahietate. Abitol, benzylbenzoatc, diethylphthdate (in part), Fixateur 404,isopropyl myristate, methyl [Iillyclrojasnlonate, andmusk ambrette. In each case, the weight lost by theparticular aromatic chemical was determined, with-out taking the weight of tbe fixative into considera-tion. If the fixative itself evaporated to a significantextent in the period of the test, the weight-loss dataof the odorant were appropriately corrected.

Evaporation of fixed aromatic chemicals. On thebasis of the data plotted in Figure 7, it is clear thatIinalool, for example, evaporated in 5 to 6 hours.The addition of 10% (reckoned on the weight ofodorant) or Abitol nearly doubled the tenacity ofIinalool, that is, the evaporation rate was almosthalved. However, it was far from possible to achievea similar result with all the substances generallyconsidered to be fixatives. The addition of a corres-ponding amount of isopropyl myristate had prac-tically no effect on the tenacity of the odorant.From this, it was apparent that for each aromaticchemical, or each class of aromatic chemical, therewas one or only a few fixatives with optimal fixingproperties.

Wrmenich, Geneva.

Vol. 1, February 197610/Perfumer .nd Flworist

The tenacity of citronellyl acetate was almostdoubled by Abitol, whereas methyl abietate causedno increase, Citronellyl propionate was exceptional-ly well fixed by benzyl .benzoate. In this case also,methyl abietate had .fhe least effect of all the 8 8xa-

tives tested, even though its use almost doubled thetenacity of the citronellyl propionate (see Fig. 7).

The rate of evaporation of an aromatic chemicalwas not always uniformly retarded by the additionof a fixative, In the course of these investigations itwas observed that different fixatives exerted an ef-fect on the progress of evaporation at times thatgreatly differed, but small deviations from thestraight lines shown in the graphs were ignored inorder to show the major effects more clearly,

The tenacity of the valuable “Rosenoxid(methyl-propenyl-methyl-tetrah ydropyran) wastripled by means of Fixateur 404. Isopropyl myris-tate caused a doubling (see Fig. 8).

The gravimetric measurement of the evaporationof fixed aldehydes yielded results that could not bereproduced in practice, i.e., as an odorous effect in

-===2x‘$ “~..;. ““....

‘, ““.:,.‘, “’,::, ;..

Fin. & Fixing of h.m.gene.aus aromatic chemicals(.P to 24h..,,].

perfume oil. It could certainly be accurately deter-mined that, e.g., heptanal was many times betterfixed by Abitol than by Fixateur 404, and that tbcrate of evaporation of citral was approximatelyhalved hy the addition of benzyl benzoate, where-as the effect of Abitol was considerably. smaller inthis case (see Fig, 8), but the gravimetric resultsgave no indication whether the aldehyde was fixedor whether even larger amounts of the correspond-ing carboxylic acids, formed by atmospheric oxida-tion, were present.

The tenacity of gerarriol was determined as beingabout 2 days. Fixation with diethyl phthalate gave awdue of 3% days, and with methyl dihydrojasmotr-ate the tenacity was increased to 5 days. The addi-tion of Abitol had practically no effect on the evap-oration rate of diphenyl oxide, but with musk am-brette its tenacity could be more than doubled.Doubling of the tenacity was also achieved by J3x-ing 2-bexyl-cyclo-pentanone with methyl dihydro-j=monate. Abitol produced only a 50% increase intenacity (see Fig. 9).

I

Fig. 10: Fixin# d hmnos.nee.s ar.m.ti< chernicak (.P t. 10days).

vol. 1, February 1976 Perfumer and Flavorist/13

A clear graduation was also shown by the fixingof diphenyl methane (tenacity ca. 2 days) with di-ethyl phthalate (3 days) and Fixateur 404 (4 days).The evaporation rate of eugenol was reduced toless than half the unfixed rate by the addition ofFixateur 404, more than doubling the tenacity,whereas musk ambrette caused only a 50% increasein the tenacity (see Fig. 10). In the case of ionone,as in that of the aldehydes, conclusions of practicalvalue could be drawn only with difficulty, becauseafter a few days a residue was formed, which couldhe measured gravimetrically, but had an odor thatdiffered considerably from the original material,For these materials, the olfactory studies describedin the next section yielded usable results,

These experiments showed that the tenacity ofany aromatic chemical that evaporated at a uniformrate could, as a rule, be doubled by the use of themost effective of the 8 fixatives studied, and in afew cases the tenacity could be tripled.

OcimenDipenteneOiphenylmethaneLinaloolTetrahydrolin.loolDihydromyrce”olStyrolyl alcoholIsop”lerjolTrimethyl cyclohexanolBenzyl alcoholPhenyl ethyl alcoholHexa”olCitron ellolGeraniolDimethyl benzylcarbinolPhenylethy ldimethylcat-binolm- KresolEugenolPhenylethy lisoarmy letherGeranylmethyl ethepRosen OxideEugenolmethyl etherIsoeugenol methyl etherDiphenyl oxideDibenzyl. therAceto. henonnCvcl Iohexyl cyclohexanone2~Hexyl -cyclopenta”oneDihydrojasmo”ea- lon onei“enzyl formatePhenylethyl fortnateGeranyl formateDecal inolformateNonyl acetateIsobornvlacet. teGera”ylicetateCitron ellol acetateDimethylben zylcai’bi nolacetateCitron ellolpropionateGeranylbutyrateAn7syl acetatePhenylethyl isovalerianateIsoarnylcapryl atef-Octal actoneHeptanalDecanalCitral

Comparison of tbe tenacities determined by gravi-Ym?tric and olfactory methods To be of practicaluse, the gravimetrically determined tenacities mustbe identical or at least parallel to the values deter-mined by olfaction. Figure 11, for example, showsthe comparative values for diphenyl oxide.

The tenacities indicated by olfactory testing wereless than those measured by gravimetry, but thegravimetric and olfactory values were in the samemutual relation in each case. The olfactory determi.nation of the duration of tenacity also showed thatdiphenyl oxide could be well fixed by muskambrette, but scarcely fixed at all by Abitol.

In the case of Rosenoxid the values were identi.cal when the substance was unfixed and when itwas fixed with isopropyl myristate. The tenacity ofthe odorant fixed with 107. of Fixatellr 404 ~a~rated considerably higher by olfactory than bygravimetric assessment (see Fig, 12). Apparentlysmall unevaporated amounts of Rosenoxid are very

A is Abieti”sSu vemethyl ester; 8 is Abitol; C is 8enzylbenzoate;D is Diethylphthalate; E is Fixative 404; F is lsop Popylmyris -tate; G is Methyl dihydrojasmonate; and H is Musk Ambrette.

Table 11:fixation of hemog.n.o.s w.m.t b chemic.h- = N. de+wminati.. mad..

14/Perfumer and Flavorist

Fig. 11: Tenwsities of diphenyl mid. asdetermined hy grwimetry (solid bar) andby olfaction (striped bar). A is MuskAmbrette, B is Abitcd.

I.,0% A . 10% B

l-lFig. 12: Temwities of 3Cmenoxidas deter-mined by gmvinmtry (solid bar) and byolfaction (striped bar). A is Fixative 404,B is Iwpropylmyristate,

vol. 1, February 1976

Fixative

Abietinsa. remethylesterAbitolBenzylbenzoateDiethylphthalateFixative 404Isopropyl myristateMethyl d{hydrojasmonateMusk Ambrette

● ✎

Kohl en-wasserstof

6.05.04.73.76.75.05.31.7

‘fe Alcohol

4.74.34.34.55.33.55.25.2

Phenol

2.56.54.0

6.53.05.02.5

4.8 4.6 4.3

Ether Ketone Ester Aldehyde *

3.7 3.4 1.5 2.0 3.34.1 3.6 3.2 3.7 4.04.9 4.6 4.5 6.0 4.7

4.54:7 4;2 5:1 2:3 5.04.6 3.5 4.9 5.3 4.33.4 5.2 3.8 4.7 4.53.3 5.2 5.8 5.3 4.8

4,1 4.2 4.1 4.2 4.4

~ge rating of fixative over all odorants.** A,e, ag, ,atf, gs of all fixatives for each class of odorant.

Table Ill: fixation of womati< chemical. co.lai.i.e the same +unctic..d group.- = Not determined.

readily perceptible to tbe sense of smell.In the gravimetric measurements, furthezrnore,

the evaporation, especially when at nonuniformrates, was often extrapolated to zero when 17. or2% of the substance still remained (sensitivity ofthe balance used ).

In the case of citronellyl acetate, the tenacitiesdetermined gmvimetrically and by olfaction werethe same for the unfixed and both fixed samules

highly tenacious residue, except in a few cases, hasno odorous similarity to the odorant originally used.This is illustrated in Figure 14 for tbe case ofa-ionone. The olfactory test, in contrast, showedthat Abitol was a considerably better fixative forc--ionone than was meth y] abietate.

In these comparative experiments, the gravimet-rically determined tenacities were, for the mostDart, confirmed by the olfactory determination.

(see Fig. 13),In tbe case of aromatic chemicals that evaporate Statistical evaluation. By a purely statistical evalua-

at nonuniform rates, the endpoint of the tena- tion, an attempt was made to find, ideally, a fixativecity is practically impossible to determine by the that possessed better fixing properties than all thegmvimetric method, because in these cases the other fixatives tested, or at least to find one fixative

Vol. 1, February 1976 Perfumer and F1.vorist/15

with optimal effect for each class of substance, forexample, for alcohols, aldehydes, or ketones,

Table II summarizes the results of alf the experi-ments on the fiaing of homogeneous aromatic cbem-icak. This list gives the most effective fixative ineach series 8 or 7 points (depending on the numberof fixatives tested) and that with the least effect 1point. The ratings of the other fixatives are inter-mediate, according to their effectiveness,

Table HI is an evaluation of Table II, showingthe efficiency of the different fixatives for odorantscontaining the same functional group,

For some classes of substance, clear differenceswere found. Thus, most hydrocarbons, for example,were well fixed by Fixateur 404 and by methylabietate, whereas musk ambrette had an extremelyweak effect. Almost all the fixatives tested weremoderately effective with alcohols. Only Fixateur404 (most effective) and isopropyl myristate (leasteffective) diverged slightly from the average, Withphenols, the best fixing was achieved with Fixateur404 and Abitol, while methyl abietate and muskambrette had only slight effects. Since only a few

w

illFig. 14: Tenecities af dwme as determined bygrav~e~y (,olid b~) ~d b.Y olfaction (s~iwd bd.A is Abikd, B is Methyl Ab]etate.

L----K=-

1-.lo~,. s

Fig. 13: Tmacities of Cilmmellol Acetate m deter-mined by gravimetry (solid bar) and by olfaction(striped bar). A is Abitol, B is Methyl Abietate.

16/Perfumer and Flavorist

aromatic chemicals with a phenolic strucfme weretested, these data should not be overrated. For themost part, moderate effectiveness was also found incompounds containing an ether group. Both henzylbenzoate, the most effective, and musk ambrette,the least effective, had average values that divergedonly slightly from the overall mean. Ketones couldbe well fixed with methyl dihydrojasmonate andmusk ambrette, whereas methyl abietate had only aslight effect on these substances. A good rating wasobtained hy musk amhrette as a fixative for esters,whereas methyl abietate was very poorly rated..4romatic chemicals containing an aldebyde groupwere outstandingly well fixed by benzyl benzoate,while methyl abietate had a particularly weak effecton these compounds also.

Reckoned over all the aromatic chemicals tested,scarcely one fixative could be nominated as havingparticularly good properties, even though Fixateur404 gained the highest rating. Apart from methylahietate, which had the leaat fixative properties bysome margin, all the fixatives studied were moreor less moderately effective.

Discussion of results

For each of numerous aromatic chemicals, at leastone fixative could be determined that possessedgood fixing properties for any particular substance.The practical significance of this is that the tenacityof an aromatic chemical can he increased by a notinsignificant amount by the addition of a suitablefixative.

Prerequisites for the effectiveness of a substancem a fixative are, among others, a certain molecularsize and a vapor pressure lower than that of theodorant to be fixed. The fixing effect is also influ-enced by viscosity.

From these relationships, it is clear that them isno clear borderline between odorants and fixatives.A good fixative can also have good odorous char-acteristics, for example, musk ambrette and methyldihydrojmmonate, and any odcmmt has a usefulfixative effect if its vapor pressure is lower than thatof the substance that is to he fixed.

Fixation reduces the vapor pressure of an odor-ant, As a consequence, its tenacity is increased; ofnecessity, however, the intensity of its odor will bereduced.

On the hasis of the statistical evaluation of the ex-periments carried out, it was not possible to deter-mine a general relationship between the effective-ness of a fixative, on the one hand, and its chemi-cal structure, vapor pressure, viscosity, molecularweight, dissolving- or absorbing-power, or the stTuc-ture of the fixed odorant, on the other. Although tbeeffectiveness of a fixative is largely dependent onthese material characteristics, it is not possible tostate the fixative by which aromatic esters or ter-pene alcohols, for example, will be particularly wellfixed.

In accordance with practical experience, these re-sults also indicate that a perfume oil compoundedfrom numerous odorants with different molecularstructures will be better fixed by a mixture of sev-eral fixatives than by any single fixative.

W. 1, February 1976