1

Kinetics of the changes imparted to the main structural components of human hair

by thermal treatment

F.J. Wortmann a, *, G. Wortmanna and C. Popescub, #

a School of Materials, The University of Manchester, Manchester M13 9PL, United

Kingdom

b DWI – Leibniz-Institute for Interactive Materials, Forckenbeckstrasse 50, D-52056

Aachen, Germany

* Corresponding author:

Prof. Franz J. Wortmann, School of Materials, University of Manchester, Sackville

Street Bldg. C47, Manchester M13 9PL, United Kingdom. Tel: +44 161 306 4158, e-

mail: franz.wortmann@manchester.ac.uk

# Current address:

Prof. Crisan Popescu, Kao Germany GmbH, Pfungstaedterstrasse 98-100, D-64297

Darmstadt, Germany

Keywords:

human hair, thermal treatment, structural damage, keratin protein denaturation, DSC

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

2

Abstract

Thermal straightening of hair is a wide-spread consumer practice, which will impart

specific hair damage. For practically relevant, cumulative conditions of thermal

treatment (straightening iron, 200 oC, 100 - 800 s) untreated and oxidized (bleached)

hair were investigated by DSC in water to determine the time-dependent changes of

protein denaturation enthalpy (ΔHD) and temperature (TD). Assuming the standard two-

phase morphology, the parameters are associated in specific ways with the α-helical

proteins (ΔHD) in the intermediate filaments (IF) and their associated matrix (TD),

respectively. Both parameters show systematic decreases with treatment time with

synergistic effects of oxidation. The decrease can in all cases be described by a 1 st-order

type kinetic model. These predict that ΔHD, and thus the contents of α-helical material in

the IFs, will approach zero for longer times of thermal treatment. The half-life time for

the process is consistently about 20-25 min. A two-level, 1st-order approach shows that

TD approaches limiting lower values with comparatively short half-times (untreated: 5

min, oxidized: 1 min). The approaches thus succeed to provide specific kinetic models

for the thermal degradation in intermediate filaments (IFs) and matrix, including the

synergistic effects of bleaching. The kinetic approaches are expected to be useful in the

context of a range of further analytical investigations of thermal hair treatments.

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

3

1. Introduction

1.1 Practical context of the study

The well-established practice of treating slightly curly or frizzy European hair with flat,

heated ‘irons’ provides an effective grooming tool to achieve straighter hair.

Due to the relevance of this grooming practice in the market place, the expert working

group ‘Hair Care Products’ of the DGK (Deutsche Gesellschaft fuer Wissenschaftliche

und Angewandte Kosmetik e.V.) has been conducting a comprehensive study to

contribute to our understanding of the objective as well as subjective changes of

European hair through thermal straightening. Part of the study has been an extensive

consumer survey.

The study shows that the majority of consumers applies temperature settings on their

devices between 150oC and 190oC, where there seems to be an acceptable trade-off

between styling effects and undesirable hair changes (1, 2).

1.2 Structural and molecular basis of non-permanent hair deformation

Human hair, similar to all other α-keratinous fibres, exhibits a complex set of structures

from the microscopic down to the molecular level (3). The two main morphological

components on the microscopic scale are the cuticle, as outer protective layers, and the

fibre core (cortex). Due to the limited amount of cuticle (approx. 10%) a specific

contribution is usually neglected in the context of mechanical and thermal analyses. The

cortex, in a first but rather plausible and successful approximation, going back to

Feughelman (4), is described as a two-phase, filament/matrix composite.

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

4

In this model the intermediate filaments (IFs) or rather their α-helical structures are the

filaments and the matrix consists of the IF-associated proteins (IFAPs). Applying the

model for the whole fibre, the term ‘matrix’ is extended to comprise all amorphous fibre

components, that is IFAPs as well as the cuticle and other minor components, such as

the cell membrane complex. For specific experimental contexts, (5, 6, 7, 8, 9, 10) more

complex models are necessary for satisfactory data analysis and discussion.

The straightening effects of the heat treatment are primarily due to ‘cohesive set’

relating to the breakage and reformation of hydrogen bonds in the hair proteins (11, 12,

13, 14). The rearranged bond network is reasonably stable until exposure to humid

conditions or washing (1, 15, 16). This lack of stability is due to the sensitivity of the

hydrogen bonds to moisture, as reflected in the humidity dependent glass transition and

viscoelastic properties of keratin fibres, such as human hair (13, 17, 18, 19). A related

practice uses curved heated surfaces to impart more or less pronounced curls to straight

hair.

Hair straightening or other heat treatments at moderate or high temperatures may impart

extensive changes to the various morphological components of human hair (20, 21, 22,

23, 24, 25). In practice about 190oC are considered as a viable compromise between the

straightening results and hair damage (26, 27). However, in order to achieve faster and

longer lasting grooming results, higher temperatures are available in straightening

devices (28), which approach (≈ 210oC) or even enter (≈ 230oC) the range of keratin

denaturation and pyrolysis in dry hair (20, 29, 30, 31).

64

65

66

67

68

69

70

71

72

73

74

75

76

77

78

79

80

81

82

83

84

85

5

1.3 Basis of the study and objectives

The material for the study consisted of commercial, brown hair tresses in their natural as

well as in strongly oxidized (bleached) form. The tresses were treated with a

commercial straightening device set to 200 oC for cumulative times of 100 s, 300 s, and

800 s, respectively. A group consensus had been reached that this would be expected to

represent roughly 2, 6, and 16 months of normal grooming practice. The temperature is

somewhat above the usual practical range, but still well below the pyrolysis range (20,

29, 30, 31) and was chosen to increase the probability that effects would in fact be

detectable, while avoiding the excessive temperature range (up to 230 oC).

Though effects of cohesive set, being related to the humidity-dependent glass transition

of hair (19) are expected to be fully reversible, it is nevertheless expected that practical

relevant heat treatments of human hair, namely with repeated use, will impart a variety

of substantial, cumulative and irreversible changes to human hair.

Here we report investigations of untreated and oxidized hair samples by Differential

Scanning Calorimetry (DSC) in water. This method specifically enables to assess

changes of the thermal stability of the main morphological components in the two-

component composite structure of the cortex of hair (4), namely, of the helical proteins

in the IFs and of the amorphous matrix (32, 33, 34).

The objective is to develop generally applicable approaches to model the time-

dependence of the parameter changes for a representative treatment temperature. These

will be linked to changes of the thermal stability of IFs and IFAPs and will furthermore

enable the assessment of the synergism of the oxidative pre-treatment.

86

87

88

89

90

91

92

93

94

95

96

97

98

99

100

101

102

103

104

105

106

107

108

109

6

2. Materials and Methods

Hair material: Commercial, Caucasian mixed hair, untreated, medium brown (Kerling,

Backnang, Germany) was used for the investigations, in what follows referred to as

‘natural’. The hair was in the form of tresses (19 cm long, 1.5 cm wide). An overall

number of 12 tresses was split into two groups. The first group was left chemically

untreated, while the second group was subjected to a bleaching process twice

(bleached).

Chemical treatment: The bleaching process (hair oxidation) was performed by

applying a commercial product (Wella, Darmstadt, Germany) based on an alkaline

solution (pH 10.5) of hydrogen peroxide (9%) and ammonium persulfate, applied for 30

min and at room temperature. This treatment was followed by rinsing and air-drying.

The treatment was repeated after a 24 h rest period.

Thermal treatment: Prior to the thermal treatment the hair was washed with 10%

sodium laurylether sulphate (SLES) solution, rinsed and dried by pressing between

paper towels. On these initially ‘towel-dry’ tresses the thermal treatment was applied,

using a commercial straightening iron set to a digital reading of 200 oC. The authors are

well aware that this is a nominal temperature. However in view of the objectives of the

study, no steps were taken at this stage to determine the actual temperature on the

surface or inside the tress, which is, however, expected to be reasonably homogeneous

(21). To control the conditions of the treatment, a tress was clamped into a tensile

testing machine (Instron, UK) and drawn through the iron such that a total contact time

along the tress of 1.67 s was achieved for each pass. For two tresses each, from the

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131

132

133

7

groups of untreated as well as bleached tresses, total contact times of 60, 300, and 800 s

were realized through repetitions. In view of practical modes of applying thermal

straightening, after 30 repetitions each tress was washed, brought to a towel-dry state

and the thermal treatment procedure was restarted.

DSC-measurements: All investigations by Differential Scanning Calorimetry (DSC)

were carried out in water, for reasons described elsewhere (33, 35) The measurements

were conducted on a power-compensated instrument (DSC-7, Perkin-Elmer, USA),

using stainless steel, large volume pans, which are pressure resistant up to 25 bar. The

temperature range was 50–190 oC with a heating rate of 10 K min-1. From the hair

tresses small subsamples were taken (approx. 100 hairs) and cut into snippets, about

2mm in length. The snippet samples were stored under standard room conditions and

would thus contain about 12% of water (3, 36, 37). Under these conditions hair snippets

(4-7 mg) were weighed into the DSC-pans and 50 μl of water added. The pans were

sealed and stored overnight prior to the DSC-measurement. Data obtained from the

measurement were denaturation temperature TD (peak location) and denaturation

enthalpy ΔHD (determined from the peak area on the basis of a linear baseline). No

corrections were applied with respect to the water content of the material. For each hair

type 22 data pairs were acquired across the experimental time range. Data analyses were

conducted using SPSS (Version 20, IBM Corp., 2011) and Statistica (Version 13, Dell

Software, 2015).

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

154

155

156

8

Codes for samples: According to the various steps of preparation, samples are coded as

follows:

N0 = natural hair, no thermal treatment (0 s)

N1 = N0 + 100 s cumulative thermal treatment

N2 = N0 + 300 s

N3 = N0 + 800 s

For oxidized, bleached hair sample coding is analogous (B0 – B3).

3. Results and Discussion

3.1 General Observations

After the heat treatment the tresses in the two groups were investigated with respect to

their denaturation performance by DSC in water. Figure 1 gives a typical DSC curve for

the natural material (commercial, untreated Caucasian human hair). Upon oxidation, the

peak becomes smaller and moves to lower temperatures (34).

157

158

159

160

161

162

163

164165166167168

169

170

171

172

173

174

9

Figure 1: Typical DSC curve for untreated Caucasian hair in water with a linear

baseline (heating rate: 10 K min-1). As indicated, the peak location signifies the

denaturation temperature TD, while the area gives the denaturation enthalpy ∆HD. The

inset shows the two-phase model (adapted from Ref.4) and illustrates the connection

between the DSC-parameters and the structural components.

DSC-curves for hair in water generally show a single peak with the general occurrence

of asymmetry or even a more or less prominent shoulder to higher temperatures. The

peak location as such, that is the denaturation temperature TD is generally associated

with the properties and, namely, the viscosity of the matrix, which kinetically controls

the denaturation of the alpha-helical material in the IFs, while the enthalpy ΔHD relates

to the energy required for this denaturation (31, 32). The assumption of a linear baseline

is an adequate simplification for the current investigation in view of the smallness of the

heat capacity change which is underlying the denaturation transition (31).

175

176

177

178

179

180

181

182

183

184

185

186

187

188

189

10

The asymmetry of the DSC-peak is attributed to the inhomogeneity of the cortex on the

cellular level, containing two major cells groups (ortho- and para-type-cells) with

different thermal performance (31, 38). The fractions of these cell groups, as well as

that of a further minor cell type, may be estimated by deconvolution of the DSC-curves,

as has recently been shown (39).

Figure 2 summarizes the results for the denaturation temperature and enthalpy for

thermally treated (200 oC) natural and oxidized hair samples, respectively. Figures 2A

and 2B show for both denaturation temperature and enthalpy generally a systematic

decrease with increasing time of exposure to the thermal treatment. For both parameters

values are off-set to lower values through the bleaching treatment, in line with previous

investigations (33, 34).

190

191

192

193

194

195

196

197

198

199

200

201

202

11

Figure 2: Denaturation temperatures TD (Fig.2A) and denaturation enthalpies ΔHD

(Fig.2B) for natural and oxidized (bleached) hair samples, respectively, and for various

times of cumulative thermal treatment (0, 100, 300, 800 s) at nominally 200 oC.

Arithmetic means (symbols), standard errors (boxes), 95% confidence limits (whiskers).

203

204

205

206

207

12

Since TD and ΔHD are assumed to be related to the properties of matrix and IFs,

respectively, plotting the parameter values against each other, as realized in Figure 3, is

an especially straightforward first step (33) to assess a potentially differentiating effect

of the thermal treatments. In view of the comparatively high precision of TD, this

parameter was chosen as independent variable for the plots.

Figure 3: Plot of denaturation enthalpy ΔHD vs. denaturation temperature TD for natural

and oxidized hair. The linear regression line through the data for natural hair is given.

The curved lined for oxidized hair is purely empirical and meant to emphasize the

apparent non-linear relationship.

Figure 3 shows the systematic and correlated decrease for TD and ΔHD for both types of

hair. For natural hair the data are well fitted by a straight line (Coefficient of

208

209

210

211

212

213

214

215

216

217

218

219

220

221

222

13

Determination, R2=0.89). The linear relationships implies that for natural hair thermal

treatment leads to equivalent changes to IFs and matrix, which are associated with

thermally induced structural damage.

The suggested non-linear relationship for oxidized hair indicates for lower TD -values,

that is for longer thermal treatment times, a faster drop for the denaturation enthalpy.

This may at first sight be interpreted as a specific sensitivity of the IFs in bleached hair

towards the thermal treatment. However, this deduction is not supported by the further

more detailed analysis of the data in what follows, which rather indicate a lower

limiting value for TD. Such non-linearity in the ΔHD vs. TD relationship, though into the

opposite direction, has previously been observed for hair after reductive + oxidative

treatment (permanent waving or straightening) (33).

3.2 One-step kinetic analyses

Though the results in Figures 2A and 2B are presented on what is essentially an ordinal

scale, the course of the data may be taken to suggest a generally exponential decay for

both TD and ΔHD. To progress beyond the essentially empirical considerations

underlying Figure 3, it is assumed in a first analysis step that in line with previous

investigations (33, 40):

TD = TD0 exp (- aT t) (1)

and

ΔHD = ΔHDo exp (- aH t) (2)

so that

223

224

225

226

227

228

229

230

231

232

233

234

235

236

237

238

239

240

241

242

243

244

14

ln TD = ln TD0 - aT t (3)

and

ln ΔHD = ln ΔHDo - aH t (4)

TD0 and ΔHD

o are the denaturation temperature and enthalpy, respectively, prior to the

thermal treatment and t is the treatment time. If Equation 3 applies, the data should form

a straight line in a ln TD vs t plot with an intercept of ln TD0 and a slope of –aT. Figure

4A shows the appropriate plot for both natural and oxidized hair. The analogous applies

for ΔHD according to Equation 4 with an intercept of ln ΔHDo and a slope of –aH (see

Figure 4B).

245

246

247

248

249

250

251

252

253

15

Figure 4: Plots for (A) denaturation temperature TD and (B) enthalpy ΔHD according to

Equations 3 and 4 for natural (♦) and oxidized hair samples (■) thermally treated at

200 °C for various times

All data in Figure 4 appear to be reasonably well described by straight lines. Table 1

summarises the parameter values for the linear regressions.

254

255

256

257

258

259

260

16

Table 1:

Linear regression parameter values for the straight line fits in Figures 4A and B

according to Equations 3 and 4. SE is the standard error associated with a parameter. R2

is the coefficient of determination for the regression. N=22 for both hair types.

Parameter Natural Hair Oxidized Hair

ln TDo ± SE 6.056 ± 0.0006 6.019 ± 0.0014

aT ± SE, 10-4 s-1 0.22 ± 0.014 0.15 ± 0.035

R2 0.920 0.480

ln ΔHDo ± SE 2.83 ± 0.016 2.55 ± 0.027

aH ± SE, 10-4 s-1 5.2 ± 0.37 4.4 ± 0.67

R2 0.902 0.676

The linear fit for denaturation temperature is good for natural hair but actually quite

poor for oxidized hair. For the denaturation enthalpy the fit is again good for natural

hair, but less so, but still highly significant well beyond the 95%-level, for oxidized

hair. This is attributed to the decrease of data precision for longer treatment times, that

is for smaller ΔHD –values, in the latter case.

As qualitatively suggested by the overlap of the standard errors for the aH-values (see

Table 1), it can be shown (41) that the slopes of the lines in Figure 4B are not

261

262

263

264

265

266

267

268

269

270

271

272

273

274

275

17

significantly different on the 95%-level (0.1 > p > 0.05). The common slope for both

lines is aH = 4.8 * 10-4 s-1.

From the equality of the slopes and assuming the denaturation enthalpy to be a property

of the helical material in the IFs, it can be concluded that the oxidative pre-treatment

systematically lowers the enthalpy compared to the natural material. However, by itself

the oxidative pre-treatment has no accelerating effect on the further development of

thermally induced damage in this structural component of hair, as may have possibly

been expected.

The behaviour of TD in this respect is different from ΔHD, in that the slopes for the lines

in Figure 4A are significantly different (p<0.05). For natural hair TD appears to decrease

faster with time than for oxidized hair. With TD considered to be associated with the

viscosity of the matrix and thus its cross-link density through disulphide bonds (32, 33),

the decrease of TD reflects a decrease of the concentration of these bonds with thermal

treatment time. In practice this corresponds to the slight sulphide smell when thermally

treating, namely, wet hair. Oxidation reduces the concentration of disulphide bonds

through breakage of cystin and the formation of cysteic acid (3). This changes the

course of the thermally induced damage in the matrix.

To proceed beyond the very general considerations underlying Equations 1-4 and based

on previous investigations (33, 34, 40) it is assumed that the value of ΔHD is related to

the amount of unchanged α-helical material in the IFs of hair, so that the relative helix

content HXrel is given by:

276

277

278

279

280

281

282

283

284

285

286

287

288

289

290

291

292

293

294

295

296

297

298

299

18

HXrel = ΔHD/ ΔHD0 (5)

Combining Equations 2 and 5 yields:

HXrel = exp (- kH t) (6)

where kH is the rate constant of the process at 200 °C and, in view of Equation 2, is

equal to aH. Equation 6 is in fact the integrated 1st-order rate law for the decrease of the

amount of native α-helix content in hair during the thermal treatment. Similar 1 st-order

processes have been observed for the thermally induced chemical changes in wool (42)

and hair (43).

As outlined above, the slopes of the lines in Figure 4B and thus the reaction rate

constants kH,200°C (= aH) are not statistically significantly different for natural and

oxidized hair. From the common value for kH = 4.8*10-4s-1 we can calculate the mean

half-life time of the α-helical material under 200 °C thermal treatment t1/2, 200°C = ln(2)/kH

~ 1400 s, with an obviously substantial error range (see Table 1). This means that under

the chosen thermal conditions (straightening iron, intermittently wet hair, nominally 200

oC) 50% of the α-helical material in untreated as well as oxidized hair has become

undetectable by DSC roughly after 20-25 min of cumulative treatment. This effect is

attributed to thermally induced cross-linking and/or denaturation in the IFs. The course

of the process is independent of chemical hair pre-treatment.

300

301

302

303

304

305

306

307

308

309

310

311

312

313

314

315

316

317

318

319

320

321

322

19

3.3 Two-level kinetics for denaturation temperature

The empirical approach to analyse the time-dependent changes of TD (Equations 1 and

3) is of limited success (see Table 1). However, a model-based approach, equivalent to

Equation 6, cannot directly be applied for TD, since the denaturation temperature will

not approach zero at long times, but a finite value (see Figure 2), which may be

approached well outside the experimental time range. This consideration suggests the

application of a two-level process, in which TD decreases from the initial value TD0 to a

final value TD∞ according to a 1st-order process, leading in analogy to Equation 6 to:

TD = TD∞ + (TD

0 – TD∞) exp (-kT t) (7)

where kT is the rate constant for the TD-drop between TD0 and TD

∞. Parameter values for

Equation 7 were determined through non-linear regression analysis (SPSS, IBM) and

are summarised in Table 2.

The results of the non-linear fit according to Equation 7 predict a substantial, but in

effect rather limited, ultimate decrease of the denaturation temperature for natural hair

by 9.3 oC from 154.9 oC to 145.6 oC. This decrease, for the chosen 200 °C

treatment conditions, is in general agreement with the results of Zhou et al. (28). The

difference between temperature levels (see Equation 7) is expected to become more

pronounced for higher treatment temperatures. There is also an expectation for a

synergistic effect of hair water content (40, 44).

323

324

325

326

327

328

329

330

331

332

333

334

335

336

337

338

339

340

341

342

343

344

345

20

Table 2:

Values for the initial (TD0) and limiting value of denaturation temperature (TD

∞), as well

as rate constants kT at 200 °C, according to Equation 7 for natural and oxidized hair.

Values are given with their 95% confidence ranges. t1/2 = ln(2)/kT is the mean half-life

time for the 200 °C treatment. R2 is the coefficient of determination of the fit.

Parameter Natural Hair Oxidized Hair

TD0, oC 154.9 ± 0.37 140.6 ± 0.60

TD∞, oC 145.6 ± 1.07 134.1 ± 0.54

kT, 200°C, 10-3 s-1 2.4 ± 0.64 10.3 ± 3.33

t1/2, 200°C, min 4.8 1.1

R2 0.981 0.939

The treatment time to achieve half of the TD-drop is estimated to be about 5 min. This is

much faster than the damage development for ΔHD, that is in the IFs. The decrease of TD

for bleached hair is smaller with 6.5 oC from 140.6 oC to 134.1 oC, arriving at a

substantially lower TD∞ (134.1 oC). This is attributed to the lower start concentration of

cystine cross-links in the pre-oxidized hair.

Figure 5 gives the TD-values and their fit on the basis of the parameter values in Table 2.

346

347

348

349

350

351

352

353

354

355

356

357

358

21

Figure 5: Denaturation temperature values TD for natural and oxidized hair vs.

treatment time at 200 °C. The solid lines through the data are based on Equation 7,

applying the parameter values in Table 2.

In contrast to the empirical 1st-order fit (see Table 1), the more realistic two-level model

(see Table 2) now shows, that the decrease of TD between its limiting values is in fact

substantially faster for oxidized (t1/2, 200°C = 1.1 min) compared to natural hair (t1/2, 200°C =

4.8 min). For both types of samples the final value of TD (TD∞) is essential reached

within the experimental time range of the thermal treatment (800 s). Oxidized hair thus

exhibits not only pre-damage in the matrix but is also prone to much faster thermal

damage development in this component, in pronounced contrast to the performance of

the IFs. This highlights the special, chemical and thermal sensitivity of the disulphide

bonds, which are concentrated in the matrix proteins.

359

360

361

362

363

364

365

366

367

368

369

370

371

372

22

4. Conclusions

For the example of a single, rather high, but still practically relevant application

temperature, the development of the changes in IFs and matrix as the two main

morphological components of hair was investigated by DSC in water. It is shown that

the analytical approach enables the detection of specific types of damage and their

development with treatment time within the two components.

Analysing the data through two stages, it is shown for both natural and oxidized hair

that denaturation temperature and enthalpy decrease with thermal treatment time,

essentially according to 1st-order kinetics. While ΔHD is expected to go to zero, TD is

plausibly predicted to assume a finite value for the chosen treatment conditions (200 oC)

largely within the maximum treatment time of 800 s. The decrease of the parameter

values is considerably faster for TD, that is for the matrix properties, than for ΔHD (IFs)

and accelerated by the oxidative pre-treatment. Oxidation thus induces specific

synergistic effects for the development of thermal damage in the morphological

components. The various aspects of the analyses emphasize the need for plausible

kinetic models to reliably systematize the data.

It is, namely, the cumulative damage in the IFs, which is expected to reduce the ability

of a hair fibre to recover from the imposed deformation and will thus lead to permanent

effects through thermal straightening (1, 45, 46). A strict relationship between

permanent straightening effect and denaturation of the helical material in the IFs, as

reflected in a decrease of the denaturation enthalpy, has been shown for the relaxation

of African hair (47, 48).

373

374

375

376

377

378

379

380

381

382

383

384

385

386

387

388

389

390

391

392

393

394

23

Against the background of earlier research (40), it is suggested that the theoretical

models to describe the time-dependent decrease of denaturation enthalpy as well as

temperature will be generally applicable and are expected to enable the systematic

analysis and possibly even prediction of the damage formation in hair at other

temperatures, measured by DSC or other analytical approaches, including the expected

synergistic contributions of hair moisture content. These aspects are part of a wider

investigation to further understand and assess hair damage from thermal treatment, as

well as to develop reliable tools to determine the efficacy of protective ingredients and

products.

Acknowledgements

The authors are grateful to the working group ‘Hair Treatments’ (Chairman: Dr. J.

Wood) of the DGK (Deutsche Gesellschaft fuer Wissenschaftliche und Angewandte

Kosmetik e.V.) for the opportunity to use data, which were generated by DWI – Leibniz

Institute for Interactive Materials (Aachen, DE) as part of a group project on the effects

of thermal treatments on human hair. In this context, the authors wish to specifically

acknowledge the contribution of Dr. P Augustin (DWI) for sample treatment, data

acquisition and organization.

395

396

397

398

399

400

401

402

403

404

405

406

407

408

409

410

411

412

413

414

24

References

1) D. Harper, J.C. Qi, P. Kaplan, Thermal styling: Efficacy, convenience, damage tradeoffs, J.Cosmet.Sci. 62 (2011) 139-147.

2) J. Quadflieg, Presentation to DGK Working Group `Hair Treatments`, priv.com. (2015).

3) C.R. Robbins, Chemical and Physical Behavior of Human Hair (5th edn.), Springer, Heidelberg, DE, 2012.

4) M. Feughelman, A two-phase structure for keratin fibers, Text.Res.J. 29 (1959) 223-228.

5) J.W.S. Hearle, A critical review of the structural mechanics of wool and hair fibres, Int. J. Biol. Macromol. 27(2) (2000) 123-138.

6) H. Liu, W.G. Bryson, A three-component model of the wool fibre–effects of morphology, elasticity and intermediate filament arrangement on fibre stiffness, J. Text. Inst. 93(2) (2002) 121-131.

7) M. Baias, D.E. Demco, C. Popescu, R. Fechete, C. Melian, B. Blumich, M. Moller, Thermal denaturation of hydrated wool keratin by 1H solid-state NMR, J. Phys. Chem. B 113(7) (2009) 2184-2192.

8) D. Istrate, C. Popescu, M.E. Rafik, M. Moller, The effect of pH on the thermal stability of fibrous hard alpha-keratins, Polym. Deg. Stab. 98(2) (2013) 542-549.

9) F.J. Wortmann, G. Wortmann, H.M. Haake, W. Eisfeld, Analysis of the torsional storage modulus of human hair and its relation to hair morphology and cosmetic processing, J. Cosmet. Sci. 65(2) (2013) 59-68.

10) F.J. Wortmann, G. Wortmann, H.M. Haake, W. Eisfeld, Model-based analysis of the torsional loss modulus of human hair and of the effects of cosmetic processing, J. Cosmet. Sci. 68 (2017) 173-182.

11) M. Feughelman, The change in stress on wetting and drying wool fibers, Text. Res. J. 29 (1959) 967-970.

12) M.Feughelman, Mechanical Properties and Structure of Alpha-Keratin Fibres, University of New South Wales Press, Sydney, Australia, 1997.

13) F.J. Wortmann, M. Stapels, L. Chandra, Humidity-dependent bending recovery and relaxation of human hair, J.Appl. Polym.Sci. 113 (2009) 3336-3344.

14) R.R. Wickett, Practical Modern Hair Science, Chap.5, Changing the shape of hair. (T. Evans, R.R. Wickett, Eds.) Allured Business Media, Carol Stream IL, USA, 2012.

15) P. Diaz, M.Y.M. Wong, Set relaxation of human hair, J.Soc.Cosmet.Chem. 34 (1983) 205-212.

415

416417

418419

420421

422423

424425

426427428

429430431

432433

434435436

437438439

440441

442443

444445

446447

448449

450

25

16) U. Assmus, S. Bielfeldt, H. Hensen, P. Hoessel, G. Lang, H. Leidreiter, A. Markowetz, V. Martin, B. Noecker, M. Pfaffernoschke, E. Poppe, H. Schmidt-Lewerkuehne, A. Schwan-Jonczyk, J. Wood, F.J. Wortmann, Multicenter study on the effect of various test parameters on the curl retention results of hairsprays, IFSCC Mag. 17 (2014) 29-35.

17) F.J. Wortmann, B.J. Rigby, D.G. Phillips, Glass transition temperature of wool as a function of regain, Text. Res. J. 54(1) (1984) 6-8.

18) F.J. Wortmann, S. De Jong, Analysis of the humidity-time superposition for wool fibers, Text. Res. J. 55(12) (1985) 750-756.

19) F.J. Wortmann, M. Stapels, R. Elliot, L. Chandra, The effect of water on the glass transition of hair, Biopolymers 81 (2006) 371- 375.

20) S.P. Detwiler, J.L. Carson, J.T. Woosley, T.M. Gambling, R.A. Briggaman, Bubble hair: Case caused by an overheating hair dryer and reproducibility in normal hair with heat, J. Americ. Acad. Dermatol. 30 (1994) 54-60.

21) R. McMullen, J. Jachowicz, Thermal degradation of hair. I. Effect of curling irons, J. Cosmet. Sci. 49 (1998) 223-244.

22) H. You, L. Yu, Atomic force microscopy as a tool for study of human hair, Scanning 19 (1997) 431-437.

23) M. Gamez-Garcia, The cracking of human hair cuticles by cyclical thermal stresses, J. Cosmet. Sci. 49 (1998) 141-153.

24) A.C.S. Nogueira, C. Scanavez, C. Carnelo, A. Gaspari, I. Joekes, I., Hair color changes caused by dyeing and thermal treatments, J. Cosmet. Sci. 55 (2004) 437-447.

25) Y.L. Lee, Y.D. Kim, H.J. Hyun, L.Q. Pi, X. Jin, X., W.S. Lee, Hair shaft damage from heat and drying time of hair dryer, Ann. Dermatol. 23 (2011) 455-462.

26) L.H. Wright, C.S. Brown, S. Cockcroft, A.A. Gandhi, A. Fuentes Diaz, Y. Deng, S.G. Latham, Identifying the optimum temperature range for using a styler on human hair, Proc. 19th Int.Hair-Science Symp., Trier, GER, 2015.

27) E. Marzec, The influence of hydration and ionizing radiation on denaturation process in keratin, Biolelectrochem. Bioenergetics 46 (1998) 279-284.

28) Y. Zhou, R. Rigoletto, D. Koelmel, G. Zhang, T.W. Gillece, L. Foltis, D.J. Moore, X. Qu, C. Sun, The effect of various cosmetic pretreatments on protecting hair from thermal damage by flat ironing, J. Cosmet. Sci. 62 (2011) 265-282.

29) M. Spei, R. Holzem, Thermoanalytical determination of the relative helix content of keratins, Colloid & Polym. Sci. 267 (1989) 549-551.

30) V.F. Monteiro, A.P. Maciel, E. Longo, Thermal analysis of Caucasian human hair, J. Thermal Anal.Calorim. 79 (2005) 289-293.

451

452453454455456

457458

459460

461462

463464465

466467

468469

470471

472473

474475

476477478

479480

481482483

484485

486487

26

31) F.J. Wortmann, G. Wortmann, J. Marsh, K. Meinert, Thermal denaturation and structural changes of α-helical proteins in keratins, J. Struct. Biol. 177 (2012) 553-60.

32) F.J Wortmann, H. Deutz, Characterizing keratins using high-pressure differential scanning calorimetry (HPDSC), J. Appl. Polym. Sci. 48 (1993) 137-150.

33) F.J. Wortmann, C. Springob, G. Sendelbach, Investigations of cosmetically treated human hair by differential scanning calorimetry in water, J. Cosmet. Sci. 53 (2002) 219-228.

34) F.J. Wortmann, C. Popescu, G. Sendelbach, Nonisothermal denaturation kinetics of human hair and the effects of oxidation, Biopolymers 83 (2006) 630-635.

35) C. Popescu, C. Gummer, DSC of human hair: a tool for claim support or incorrect data analysis?, Int. J. Cosmet. Sci. 38 (2016) 433–439.

36) F.J Wortmann, A. Hullmann, C. Popescu, Water management of human hair, IFSCC Mag. 10 (2007) 317-320.

37) S. Pye, P. Paul, Trehalose in hair care: Heat styling benefits at high humidity, J. Cosmet. Sci. 63 (2012) 233-241.

38) F.J. Wortmann, H. Deutz, Thermal Analysis of Ortho- and Para-Cortical Cells Isolated from Wool Fibers, J. Appl. Polym. Sci. 68 (1998) 1991-1995.

39) F.J. Wortmann, G. Wortmann, A somewhat unexpected result from the deconvolution of DSC-curves for human hair: There is no apparent relation between cortical cell fractions and hair curliness, Exp.Dermatol. (2017) DOI: 10.1111/exd.13462.

40) H. Schmidt, F.J. Wortmann, High pressure differential scanning calorimetry and wet bundle tensile strength of weathered wool, Text. Res. J. 64 (1994) 690-695.

41) J.H. Zar, Biostatistical Analysis. (3rd edn,. Prentice Hall, Upper Saddle River, NJ, USA, 1974.

42) L. Lu, I.L. Weatherall, The thermal degradation of wool, Proc. 9th Int. Wool Text. Res. Conf. Biella, Italy II (1995) 265-271.

43) R.L. McMullen, J. Jachowicz, Thermal degradation and protection of hair, J. Cosmet. Sci. 49 (1998) 190-191.

44) P. Christian, N. Winsey, M. Whatmough, P.A. Cornwell, The effects of water on hair-styling damage, J. Cosmet. Sci. 62 (2011) 15-27.

45) A. Dussaud, B. Rana, H.T. Lam, Progressive hair straightening using an automated flat iron: Function of silicones, J. Soc. Cosmet. Chem. 64 (2013) 119-132.

46) S.B. Ruetsch, Y.K. Kamath, Effects of thermal treatments with a curling iron on hair fiber, J.Cosmet.Sci. 55 (2004) 13-27.

47) J. Quadflieg, Fundamental properties of Afro-American hair as related to their straightening/relaxing behaviour, PhD-dissertation, RWTH Aachen, DE, 2003.

488489

490491

492493494

495496

497498

499500

501502

503504

505506507508

509510

511512

513514

515516

517518

519520

521522

523524

27

48) F.J. Wortmann, F.J., J. Quadflieg, L.J. Wolfram, Mechanistic aspects of African-hair relaxation, Proc. 17th Int. Hair-Science Symp., Irsee, DE, 2011.

525526

527