J. Soc. Cosmet.Chem. 22 401-410 (1971} ¸

1971Society of Cosmetic Chemists of GreatBritain

The mechanicalpropertiesof virgin and treated human hair fibres; a

studyby meansof the oscillating beam

method P.J. HUCK and C. B. BADDIEL*

$ltnopsis--A dynamic mechanical technique using an OSCILLATING described

which will determine

both the ELASTIC

MODULUS

BEAM method i8

and the LOSS

MODULUS

in single fibres of HUMAN HAIR KERATIN. The method is particularly useful in that it can be applied at very low extensions,< 2 •o, and can, therefore, realistically evaluate changesin the fibre moduli which are likely to be of the same order of magnitude as those experiencedin the in vivo situation where strains are largely in the HOOKEAN region. The data obtained from virgin hair fibresat 75 •o r.h. are usedas a standard set or reference frame and the effects of dihydroxynaphthalene,resorcinol formaldehyde polymer, a commercial waving lotion (Pin-Up), a commercial setting aid (Textra), and sodium dodecyl sulphateon the moduli of thesefibresare discussed,wherepossible,in terms of the KERATIN STRUCTURE.

INTRODUCTION

Investigations into the effects of mechanical and/or chemical damageon keratin fibres,generallywool, have been largely confinedto an examination of the tensile or torsional properties of the material, e.g. elastic modulus,yield point and breaking stress(1-10). While noticeable changescan be detectedin fibresby measurements which give the above mentioned quantities, the stressesimposed bear little relation to the situation with regard to human hair keratin, becausethe in vivo stresses which normally occurin human hair are imposed during treatments,in *Unilever Research Laboratory, Isleworth, Middlesex 401

402

JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS

movementsof the head and in combingand the wearing of hats, etc. The time scale of these stresses is often of the order of one second or less and the

extensions(strain), if any, do not usually exceed1 or 2%, i.e. arein the Hookean region. An experimental technique which could be used to measurechangesin thesetransient stressesat low strain valuesbeforeand after treatment would yield usefulinformation about the state of the fibre. A rheologicaldevice which most closelysimulatessuch in vivo conditions in hair is the rockingbeam oscillator.An apparatus,developedby Tokita (11), imposesa low frequencysinusoidalstresson hair in the longitudinal direction.The use of a dynamic mechanicaltechniqueof this type has a distinct advantage over the static test methodsin that two independent parameterscan be derived from one experiment.These are the elastic modulus(E'), and the lossmodulus(E") whichis a measureof the irreversible lossin energywhen the fibre is extended. Changesin one or both of these terms after a treatment may well be related to the bulk propertiesof the keratin fibre, and by experimenting with suitably chosenmodel compoundsthesechangescould be assignedto molecularmotions.However, as a standardtechniquefor the evaluationby rank of treatments the method offers a relatively fast and inexpensive alternative to static tensilestressexperiments. EXPERIMENTAL

Materials

Virgin Italian Blue Stringbrownhair, suppliedby RaoulLtd., London, wasether degreased, and usedthroughouttheseexperiments. Unless otherwise stated, all chemicalsused were Analar© grade*.

Pin-Up© waving lotiont and Textra© settingaid+ + were obtained from the manufacturersand usedaccordingto the instructionson the pack. Apparatus

The apparatusis illustratedin Fig. 1. The sampleof hair is clampedto the brassbeam (7 cm long) at two pointscloseto the pivot and passed throughthe eye of a verticallyadjustableneedleto which it is lightly ,

*BDH ChemicalsLtd., Poole, Dorset, BH12 4NN, and Hopkin and Williams Ltd., Freshwar r Road, Chadwell Heath, Essex.

tElida Ltd., 43 Portman Square, London, W. 1. :[ClynolLtd., 22 Old Bond Street, London, W. 1.

MECHANICAL PROPERTIES OF VIRGIN AND TREATED HUMAN HAIR FIBRES 403

Revolving Photographic Recorder

•Mirror

Beam

Light

Source

HairSample

Figure 1.

Tokita's oscillating balance

glued.This arrangementallowsthe fibre to be slightly extended(< 1%) prior to the experimentbut ensuresalso that no slippageoccurswhich would introduce an additional damping factor; hhe two portions of the samplethus act independentlyand identically. To start the oscillationthe beam is deflectedby the electromagnet.The frequencyof oscillationis determinedby the momentof inertia of the beam and by the viscoelasticpropertiesof the fibre. The oscillationis allowedto decayfreelyandthe decayis followedphotographically. Thisis achievedby reflectingvisible light off a small mirror attached to one end of the beam; the light is directedon to a revolving drum whichis coveredwith photographicor light sensitivepaper.The wholeassembly,with the exceptionof the moving drum, is housedin a constant temperature and humidity environment fitted with a transparent window. The time taken for one determinationis of the order of a few minutes although the hairs are left to equilibratefor two hoursat the environmentalr.h.

404

JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS Theoretical

The elasticmodulus,E', is derivedfrom the expression(11):

8 •2I 1 (v2-Vo 2)

E'=

wherevo is the naturalresonantfrequencyof the freely vibratingbeam, and v is the frequencywith the fibre in position;1 is the length of the fibre, b1 is the distancebetweenthe pointsof suspension of the fibre ends; • is the cross-sectionalarea of the fibre; I is the moment of inertia of the beam.

The moment of inertia of the beam can be determinedby usingfibres the elastic modulus and dimensions of which are known; it can also be cal-

culatedby summingthe momentsof the parts, and for this purposethe clampson the bar may be neglected. The lossmodulus,E", cannot be gained directly from these measurements. To obtain this quantity a parametercalled the losstangent term

(tanb or Q- •) isderivedfromthe decayin the oscillationwhichis attributed to the viscouslossof energyin the fibre. In physical terms this decay

represents the fact that the sinusoidalstresslags behind the sinusoidal strainby the angleõ, and E' followsimmediatelyfrom the relationship: tan b =•

(II)

E'

The expression relatingtan õ to the beamfrequencyis givenby (11):

v2 •,c (1_ •,•) tanS--(v2-vo

(III)

wherev andVoare definedas in equation I; •,p is the logarithmic decrementof freevibrationand •,cis the decrementwith the fibre in place.

If • isconsidered to bezerothenequation III reduces to tan b ---

v2

•,c

(v2- Vo

(IV)

RESULTS AND DISCUSSION

All experiments were conducted at 25 +2øCand at 75 +2% relative humidity. The latter was maintainedat this value by a saturatedsalt solution(NaC1)and monitoredby a humidity transducer. The data obtainedin each experimentwere from severalfibres.The resultingspreadof modulivalueswereanalysed according to the statistical techniques of significantdifferenceand correlation.

MECHANICAL PROPERTIES OF VIRGIN AND TREATED HUMAN HAIR FIBRES 405

Virgin hair To act as a referencebasefor comparisonpurposesmeasurementswere made on twenty samplesof virgin hair; the values of the elastic and loss moduli, E' and E" respectively,and their averagestandard deviationsare presentedin TableI. Thesevaluesagreewell with thoseobtainedpreviously (12). Table

I

The visco-elasticproperties of virgin hair at 75 •orh at 25øC

*E'

*E" 0.214 0.115 0.160 0.201 0.148 0.181 0.161 0.181 0.174 0.161 0.152 0.180 0.218 0.166 0.202 0.167 0.166 0.223 0.205 0.172

3.51 3.20 4.00 5.92 4.12 4.12 3.51 3.61 3.55 4.12 4.00 4.28 3.70 4.60 3.74 4.92 5.04 4.96 3.94 4.20

Sulphur content % 5.9 5.7 5.9 6.3

5.1 6.5 5.1 4.4 4.9 4.7 4.6 4.8 7.0 5.9 6.5 5.5

6.0 6.4 5.2 --

Average 4.10

0.177

5.6

S. Der.

0.027

0.7

*Ncm-2

0.66 X

10 -15

However, considerable scatter was observed. This is not unusual and

can probably be accountedfor by the biologicalinhomogeneityof each individual fibre (the reproducibilityof the instrument was of the order of z2% and thereforecan be eliminated as a sourceof scatter). This inhomogeneitycan be traced to two most likely causes:(i) difficulty in determiningaccuratelythe cross-sectional area of the keratin fibre,and (ii) variation in the sulphur content of each hair. The sulphur content of each fibre was determined using a polarographicmethod and these results are also included in Table I. While no clear correlationbetween the sulphur contentand the elasticmodulus,E', emerged(13) there was a correlation

4O6

JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS

with the lossmodulus,E", significantat the 5•/olevel. This resultindicates that the increasein lossmoduluscouldbe attributed to the differingcrosslinked densitiesbetweenindividual fibres.Although an explanationdoes not springto mind readily, it is possiblethat even at theselow extensions someresidualdisulphideinterchangeis occurring;this would be facilitated

in thosehairswith highersulphurcontent,andperhapscontributemoreto the value of E". This concentrationdependence of mechanicalproperties with sulphurcontenthasbeenreportedpreviouslyby Ripa and Speakman (14)for creepexperiments in thepostyieldregion.Althoughthemechanism must be different at greater stressesit is significantthat variation in the concentrationof sulphurcan determinechangesin the mechanicalproperties.

Measurements on treatedfibres

Having establisheda set of valuesfor E', E" and tan • for virgin hair keratin, we are in a positionto examinetreatedfibresand relate changesin thesequantities,wherepossible,to structuralmodifications broughtabout by these treatments. 1:8 Dihydroxynaphthalene (DHN)

The binding of organicmoleculesto hair keratin has been shownto fayour thosecompoundswhich are unchargedand contain both hydrophilic and hydrophobicgroups;sucha materialis DHN. The absorption isothermof DHN on hair indicatesthat there is a considerable uptake at

20øC(1.25 X 10-3 moleDHN g-• hair) (15)and that theremay wellbe corresponding changesin the values of the storage (E') and lossmoduli

(E*); accordinglyhairs treated with DHN (15) were examinedby the rockingbeam oscillatortechnique.The data are listed in TableII. It is Table

1I

The visco-elasticpropertiesof treated hair at 75 • r.h at 25øC Treatment

*E'

S. Dev.

*E"

S. Dev.

1,3 D.H.N.

3.9

0.3

0.26

0.05

2 •o Thioglycolate $5.2 •o Polymer $9.2 •o Polymer $21.5•o Polymer

3.5 3,7 3.8 3.8

0.2 0.7 0,2 0.8 1.2 0.6

0,17 0.24 0.27 0.27

0.32 0.22

0.02 0.04 0.03 0.08

0.05 0.04

Pin-Up

4.5

0.5

0.20

0.0S

S.D.S. on cuticle free hair Cuticle free hair Textra

5.8 4.3 4.7

*lq cm-2 x 10-15

1.0

0.21

?Percentageuptake of polymer

0.05

MECHANICAL PROPERTIES OF VIRGIN AND TRE.•TED HUMAN HAIR FIBRES 407

clear that the elasticmodulus,E', is unchangedbut the lossmodulusincreasesby nearly 40%. It is not entirely unexpectedthat E' is unchanged at this strain level as the value of this modulus reflects the H-bond

and salt

linkagestrengthsand there is no reasonwhy DHN would specificallyreact to affect either. However, it is believed that the mode of interaction of

DHN with hair keratin is to disrupt the intrinsic water structure of the fibre thus changingthe local hydrophobic-hydrophilic balance whilst not complexingcovalentlywith the protein.This breakdownin water structure would account for the increase in E" which is a sensitive

indicator

of the

viscosityof the material under test. It is interestingto note that this in-

formation wouldbeverydiœficul• to obtainin the Hookean regionusing static tensile testing methods.

Impregnationof hair with polymers The problemof introducingpolymersinto wool and hair has beenunder investigationfor many years (1t3),and recently the mechanicalproperties of wool fibrescontainingpolyacronitrilewere describedby Simpson(17). These studieswere carried out using the oscillatingbeam technique presumably becausethe conventionalload extensioncurve is severely disrupted by the presenceof polymer. In the presentwork, hair which had containedincreasingquantitiesof resorcinol formaldehyde condensatewas examined. In order to secure adequateuptake of polymer (maximum of 21.5ø/0by weight of dried hair, introducedas a precondensate) mild reductionof the hair was a primary requirement(2% thioglycolateat pH 8). Reductionof the hair decreases slightlythe valueof E' from 4.1-3.5 X 105N cm-2. Thisresultis predictable but not really significantand it is interestingto note that the addition of increasingquantities of polymer showsa trend towards higher E' values (cf virgin hair); in contrastto the small changesexhibitedby E' the loss modulusof the keratin increasesconsiderablywith polymeruptake. It is not possiblefrom this data to accuratelyassessthe changestaking placebut it would appearthat the polymeris probablylocatedin the noncrystallineareasof the cortex,i.e. the matrix, as any directinteractionwith the microfibrilswould showup in large changesin E'. The changein E" togetherwith the observedswellingof the fibressupportthe view that the polymer further disorganizesthe matrix as an increasein this quantity is indicative of easierflow propertieswithin the system.Theseresults,with regardto E', appearto conflictwith thoseof Simpson(17) who showedfor

4O8

JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS

the wool/polyacrylonitrile systemthat E' decreased linearlywith increasing polyacrylonitrilecontent.The discrepancyin the value and directionof the quantity E' in thesetwo casesmay be attributed to the way in which the polymer was introducedinto the fibre. In the present work large precondensatemoleculeshad to penetrate first the cuticle and then into the cortex, etc. whereas in Simpson'sstudy the relatively small monomer moleculeacrylonitrile, was taken up by the fibre from aqueoussolution without prior reductionor pretreatmentand the polymerizationtook place in situ. It is not unreasonableto assumethat there would be physical restrictionsto the freedom of diffusion of large moleculesboth into and within the hair whereasa small moleculecould diffusemore readily inside the fibre and perhapsinteract directly with the crystallinecomponentsof the cortex (protofibrils)beforepolymerizing.If theseinteractionsweakened the protofibrillar structureit would be reflectedin the decreasein E' Sodiumdodecylsulphate

The role of the cuticlein hair fibresmay be to act as a diffusionbarrier. Sodiumdodecylsulphate(S.D.S.)interactsstronglywith globularproteins but affectsthe mechanicalpropertiesof woolandhair onlyto a slightextent. Thus, Zahn et al (18) showedthat wool fibresextendedin saturatedsolutions of S.D.S. were weakenedby about 6% on the basisof the 25% work index (a measureof the behaviourof the elasticmodulusin the Hookean and the yield regions).For hair the effectswere much less.This suggests that the cuticle of the wool is implicated in controllingthe rate of the diffusionprocessas its layer thicknessis considerablylessthan that of hair cuticle.It wouldbe expected,therefore,that hairs from which the cuticle wasremovedwould react more effectivelywith detergentssuchas S.D.S. Hairs strippedof cuticle (by mechanicalabrasion)were immersedin saturatedS.D.S. (0.7 M at pH 7.0) for 100 h and after light rinsingwere examinedon the oscillatorat 75% r.h. and comparedwith the data from cuticle-freevirgin fibres.It was found that there was an increaseof 35% in the averageelasticmodulusand about 50% in the lossmodulus(TableII). It is clear that in the absence of cuticle the reaction between hair keratin

(i.e. cortex)and S.D.S. is quite severe(18). The changein mechanicalpropertieson reactionmay be accountedfor by postulatingthat the headof the S.D.S. moleculereactswith polar side chaingroups,particularlyin the microfibril areas,andthe tail portionof the moleculesticks into the amorphousand more hydrophobic-likecortex.

MECHANICAL PROPERTIES OF VIRGIN AND TREATED HUMAN HAIR FIBRES 409

Effectively, the S.D.S. could be forming a quasi-salt bridge which is strongerthan those present in the untreated fibre, and as a result of its hydrophobicinteractionit driveswater from the structureleadingto lower moistureregainsat given r.h. comparedwith the untreated fibres. Both theseeffectswill increaseE'. There is someevidenceto supportthis point of view from low angle X-ray diffractionstudieson wool (19). This indicated that the detergentmoleculeswere lying perpendicularlyto the fibre axis with the head portion closeto cystinecrosslinks. The marked increasein the lossmodulusis ascribedto a disruption of the matrix by the non-polarparts of the S.D.S. molecule.This would be in accordwith the X-ray studiespreviouslyreferredto and alsoto the work of Weigmannand Chen (20) who showedthat at high pH (> 6) there is a considerableincreasein entropy accompanyingthe diffusion of S.D.S. into wool.

In conclusionit may be stated that althoughthe action of the S.D.S. is not clearly determinable,the absenceof the cuticle doesallow increased accessto the cortex with a correspondingmarked effect on the mechanical propertiesof the fibre (comparecontroldata TableII).

A wavinglotion(Pin-up©)andsetting aid (Textra©) It was of interest to examinechangesin the mechanicalpropertiesof singlehair fibresafter applicationof thesecommercialtreatments.Neither producedsignificantchangesin E' or E". This resultfor Textracanbe explainedsimplyby the fact that the setting aid providesa surfacecoatingto the fibre which is too thin to be mechanically important and its set propertiesprobablyderive from enhancement of interfibre

friction.

In the case of hair fully treated with Pin-Up waving lotion, which involvesa reductionfollowedby a reoxidation,the systemhas returned to its original mechanical state even though its internal and external configuration may have changed.Thus the 'no change'result indicatesthat internal fibre damageis at a minimum which confirmsthe earlier work of Hamburgerand Morgan (6). CONCLUSION

The measurement

of the elastic and loss moduli

in human

hair keratin

fibres by the oscillatingbeam techniquehas been shown to be reliable,

410

JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS

inexpensiveand relativelyquick. In this contextit canbe of particularuse in assessing varioushair treatments. The data appearto indicatethat the value of the elasticmodulus,E', is dependenton the morecrystallinecomponents in keratinand alsoon the degreeof cross-linking. The lossmodulus,E', however,seemsto vary with the extra degreeof disorderwhich is superimposed on the cortexby some of the reactive materials

introduced

into the hair.

ACKNOWLEDGEMENT

We thank MissT. A. Howard for her invaluableassistancein obtaining someof the experimentalresults.

(Received: œ1stDecember 1970) REFERENCES

methodsfor usewith animal fibres.J. TextileInst. (1) Speakman,J. B. Mechano-chemical Trans. 118 102-26 (1947). (2) Mitchell, T. W. and Feughelman,M. The bending of wool fibres. Textile Res. J. 11õ 311-14 (1965). (3) Guthrie, J. C., Morton, D. H. and Oliver, P. H. An investigation into bending and torsionalrigidities of somefibres.J. Textile Inst., Trans. 45 912-929 (1954). (4) Beyak, R., Meyer, C. F. and Kass, G. S. Elasticity and tensilepropertiesof human hair. I. Single Fibre Test Method. J. Soc.Cosmet.Che,n.20 615-626 (1969). (S) Rebenfeld, L., Weigmann, H. D. and Dansizer, C. Temperature dependence of the mechanicalpropertiesof humanhair in relation to structure.ibid 17 525-538 (1966). (6) Hamburger, W. J. and Morgan,H. M. Someeffectsof waving lotionson the mechanical

propertiesof hair. Proc.Sci.Sect.ToiletGoods Assoc.18 44-8 (1952).

(7) Feughelman,M. and Reis, P. J. The longitudinalmechanicalpropertiesof wool fibres and their relationshipto the low sulphur keratin fraction. Textile Res.J. 87 334-6 (1967). (8) Feughelman,M. and x•Vatt,I. C. Wool fibres treated with ninhydrin or formaldehyde:

mechanicalproperties.TextileRes.J. 34 643-4 (1964). (9) Deam, D. E. and Rieger, M. M. Mechanical hysteresis of chemically modified hair.

J. Soc.Cosmet.Chem.19 395-410 (1968).

(10) Feughelman,M. and Robinson,M. S. The relationshipbetweensomemechanicalpro-

pertiesof singlewoolfibresandrelativehumidity.TextileRes.J. 117441-6 (1967).

(11) Tokita, N. The effects of crystallization and drawing on the visco-elastic properties of

fibres. J. Polymer Sci. 9.0515-536 (1956). (12) Khayatt, R. M. and Chamberlain,N.H. The bendingmodulusof animal fibres. J. Textile Inst., Trans. 119185-197.

(13) Alexander, P., Fox, M. and Hudson, R. F. The reaction of oxidizing agents with wool. Blochem.J. 129-138 (1951). (14) Ripa, O. and Speakman,J. B. The plasticityof wool.TextileRes.J. 9.1215-22(1951). (15) Breuer,M. M. The bindingof phenolsto hair IV. The binding of 1,3dihydroxynaphthalene

(DHN) to hair. J. ColloidInterfaceSci. 9.4577-583(1967).

(16) Wolfram, L. J. Modificationsof hair by internal depositionof polymer. J. Soc.Cosmet. them. 20 539-553 (1969). (17) Simpson,W. S. Int. WoolText. Res.Conf.,3rd, Paris 359-375(1965). (18) Zahn, H., Stein, W. and Blankenburg, G. The influence of surfactants on the physical

properties of keratinfibres.l•'ette,Seifen,Anstrichm 70 (10)757-760(1968).

(19) Spei, M. Influenceof anionicdetergentson low angle X-ray diffractionpatternsof a-keratin. Int. Wool Text. Res. Conf. •tth, Berkeley,Calif. 125-6 (1970). (20) Weigmann,H. D. and Chen, J. C. Chemisorptionof an alkyl sulphateby wool. Int. Wool Text. Res. Conf., 4th, Berkeley,Calif. 126 (1970).