Effect of shampoo, conditioner and permanent waving on the molecular structure of human hair

Effect of shampoo and conditioner on hair

The main purpose of shampoo is to remove dirt and oil from the surface of the hair fibres and the scalp, while the main purpose of conditioner is to ensure that the hair is smooth for combing. Some commercial shampoos may also have additional components to control dandruff and condition hair. Similarly, conditioners may also prevent static electricity, improve the cosmetic shine and increase protection. Due to the multitude of purposes of these hair care products, they contain a long list of ingredients with various effects on the hair. Shampoos typically contain a primary and a secondary surfactant for thorough cleaning, a viscosity builder, a solvent, conditioning agents, pH adjuster and other non-essential components such as fragrance and colour for commercial appeal. Conditioners usually contain silicone polymers to increase shine and soften hair, cationic polymers such as quaternized nitrogen compounds to reduce static electricity, bridging agents to increase absorption, viscosity builder, pH adjuster and components for commercial appeal (Robbins, 2012; Preedy, 2012).

Since shampoo removes soil and lipids at the surface level, and the active conditioner ingredients only adsorb to the surface, most of the shampoo and conditioning interactions can be expected to occur at the first few layers of the cuticle. The cortex will only be impacted if damage to the cuticle is extensive and the cortex has been exposed. There is evidence that shampooing can result in hair damage in this case; however, the effect is believed to be mostly restricted to the cuticle level. For instance, abrasive action of the shampooing process can damage the keratin and non-keratinous structures at the surface of the hair (Gould & Sneath, 1984). Yet, other studies suggest that the application of conditioner after shampoo increases the protection at the surface of hair fibres, preventing damage from destructive processes such as grooming and bleaching (Tate et al., 1993).

Studies have shown that application of shampoo once can extract approximately 50% of total extractable lipids in hair (Shaw, 1979), and that 70–90% of total lipid can be extracted with repeated shampooing (Koch, 1982). However, it is believed that largely surface level lipids are removed. The cell membrane complex at the cuticle level contains covalently bonded 18-methyl eicosanoic acid (18-MEA) lipids attached to a proteinous cell membrane. At the hair’s outer surface, free lipids within the 18-MEA lipid layers are removed during shampooing (Robbins, 2012). Internal lipids found deeper within the hair shaft are not affected to the same extent as the surface lipids. Internal lipids can travel to the surface layers via the process of diffusion after repeated shampooing (Robbins, 2012).

In summary, shampooing has the effect of extracting surface lipid matter and repeated shampooing could result in damaging structures at the hair surface. These changes could be reflected in the X-ray profiles, particularly the lipid signals. Conditioner ingredients adsorb to hair surface and act as a protective layer preventing this damage. Thus conditioner application, either alone or after the use of shampoo could also affect the X-ray signals.

Effect of permanent waving on hair

Permanent waving is the process of converting straight hair into curled hair using a chemical process. This process takes many steps to complete and involves the breaking and restructuring of the disulfide bonds within hair. In the first step, disulfide bonds are cleaved using reducing agents such as thioglycolates and bisulfites. The reduction permeates throughout the cuticle and into the outer layers of the cortex. Alkaline reductive solutions are used to lift the scales of the cuticle such that the reducing agent can effectively diffuse into the cortex. In a second step, the hair fibres are shaped into the preferred curl and held using curlers. During this time, molecular reorganization occurs in the cuticle layers, as well as in the outer levels of the cortex. Lastly, the fibres are neutralized with an oxidizing agent such as hydrogen peroxide (Robbins, 2012; Bolduc & Shapiro, 2001).

The damaging effects of permanent waving are widely known. Particularly, the use of hydrogen peroxide in the oxidation step increases surface damage and decreases the stiffness of the fibre (Tate et al., 1993). Studies have shown structural changes in both protein composition and lipids after perming (Johnson, 1997). A recent study suggests that prolonged exposure of over 30 min to reducing/oxidizing agents during the permanent waving process reduces the contents of α-helical proteins. This effect is observable in the wide angle X-ray diffraction (WAXD) pattern as a reduction in the intensity of signals related to packing at the level of individual keratin fibres (Nishikawa et al., 1998). The decrease in keratinous protein has also been observed by another group of researchers using a novel method for quantitative analysis of hair components (Kon et al., 1998). Additionally, the permanent waving process was found to extract internal lipids from the cell membrane complex of the fibre, as one study discovered a significant decrease in polar internal lipids after perming treatment (Hilterhaus-Bong, 1989). Moreover, the permanent waving process converts the hydrophobic hair surface to a more hydrophilic one. Thus, in interactions with shampoos, this would result in greater lipid removal from the cuticle (Robbins, 2012).

In summary, permanent waving produces more significant changes to hair structure than shampooing and conditioning, and its effects permeate into the cortical layers of hair fibres. Observable changes to the X-ray profile are, therefore, expected to incur as a result of perming.

Preparation of hair samples

This research was approved by the Hamilton Integrated Research Ethics Board (HIREB) under approval number 14-474-T. Written consent was obtained from the participating individual. Samples were obtained from a single subject to eliminate the effect of other potential factors on the observed signals, since the X-ray profiles of various individuals were observed to be different in the previous study (Yang, Zhang & Rheinstädter, 2014). Using samples from a single subject ensures that any changes observed in the signals are the result of the hair treatments. The subject is female, of Asian descent and naturally has slightly wavy and black hair.

The shampoo and conditioner used in this experiment were a common, top-selling brand found in drugstores in US and Canada. When studying the list of ingredients we found that many shampoos contain conditioner ingredients and vice versa, such that a clear distinction is often difficult (Robbins, 2012). The particular brand was chosen for relatively little overlap between shampoo and conditioner ingredients. The active ingredient in the shampoo was sodium laureth sulfate and the active ingredients in the conditioner included cetearyl alcohol, stearamidopropyl dimethylamine and dimethicone. The permanent waving product used was an acid wave product with the main ingredient being glycerolmonothioglycolate (GMT). Since GMT is not stable in water, the reducing formula consisted of two parts that were mixed right before application to produce GMT (Robbins, 2012). The reducing formula was applied to virgin hair, which was subsequently held in position with curling rods. Though the particular product did not require heat application, the hair dresser chose to use an electric heater for thirty minutes to improve treatment result. After heat treatment, the reducing formula was washed off with warm tap water. Then neutralizing lotion was applied to entirety of the head and allowed to settle in for five minutes. Afterwards, the curling rods were removed and hair was rinsed with warm tap water. Lastly, the hair was blow dried.

At the beginning of the study, around 200 hair strands were obtained from the subject’s frontal scalp region, less than 10 cm from the roots. These 200 strands were divided into four approximately equal portions with around 50 strands each. In preliminary experiments where we optimized the X-ray setup and hair preparation we found that a sample that contained ∼50 hair strands gave the best signal. Fewer hairs resulted in faint signals, thicker hair bundles led to absorption and a reduction of signal and increase of background.

Shampoo and conditioner were applied differentially to each of the samples according to Table 1. Then the subject underwent permanent waving treatment in all regions of her scalp hair at a local barber shop using the procedure described above. Another 200 strands were obtained from the same area of the head. These 200 strands were again divided into four 50 strand portions and differential amounts of shampoo and conditioning was applied to each. The treatments applied to each sample can be found in Table 1.

Prior to shampoo and conditioning treatments for both non-permed and permed series, the subject did not wash her hair for the duration of one week to facilitate the build-up of lipids on the hair surface. A previous study noted that one application of shampoo can extract 50% of total extractable lipids in hair after not shampooing hair for 2.5–3 days (Shaw, 1979). Thus, the period of a week was determined to be enough time for lipid build-up. Shampoo and conditioning treatments were performed after the hair strands had been collected from the subject’s scalp. Solutions of shampoo and conditioner were prepared in a 1:9 ratio with tap water to give a 10% solution. To shampoo or condition the sample, the hair was held using a tweezers at one end and rinsed in the corresponding 10% solution for 2 min continuously. Then the solution was replaced by tap water and the sample was rinsed for another 2 min. Each sample was allowed to dry for at least 1 hour to eliminate the effect of water on the keratin signals, since water is known to cause swelling of hair (Robbins, 2012). This process was designed to mimic the normal application of shampoo and conditioner during a single shower.

The treated samples were then cut into strands ∼3 cm long. Care was taken to prevent stretching or deforming the hair strands during this process. Each sample was mounted onto a flexible cardboard apparatus as shown in Fig. 2A. The cut-out at the middle of the apparatus allows for the scattering of X-ray signals on the hair sample. The cardboard apparatus is then mounted vertically onto the loading plate of the Biological Large Angle Diffraction Experiment (BLADE) using sticky putty as shown in Fig. 2B. All hair samples were scanned at ambient conditions, i.e., a room temperature and humidity of 28 °C and 50% RH.

X-ray diffraction experiment

X-ray diffraction data was obtained using the Biological Large Angle Diffraction Experiment (BLADE) in the Laboratory for Membrane and Protein Dynamics at McMaster University. BLADE uses a 9 kW (45 kV, 200 mA) CuKα Rigaku Smartlab rotating anode at a wavelength of 1.5418 Å. Focusing multi-layer optics provides a high intensity parallel beam with monochromatic X-ray intensities up to 10¹⁰ counts/(s × mm²). By aligning the hair strands in the X-ray diffractometer, the molecular structure along the fibre direction and perpendicular to the fibres could be determined. We refer to these components of the total scattering vector, $\overrightarrow{Q}$, as q_z and q_‖, respectively, in the following. An illustration of q_z and q_‖ orientations is shown in Fig. 1.

The result of such an X-ray experiment is a 2-dimensional intensity map of a large area of the reciprocal space of −2.5 Å⁻¹ < q_z < 2.5 Å⁻¹ and −2.5 Å⁻¹ < q_‖ < 2.5 Å⁻¹. The corresponding real-space length scales are determined by d = 2π/|Q| and cover length scales from about 2.5 to ∼200 Å, incorporating typical molecular dimensions and distances for primary and secondary protein structures and lipid structures.

Integration of the 2-dimensional data was performed by a script developed in our lab using MATLAB (MathWorks, Natick, Massachusetts, USA). By integrating of the peak intensities along the q_z and the q_‖ directions, one-dimensional data along each of the two directions were produced. The q_z intensity was integrated azimuthally for an angle of 25° over the meridian. The q_‖ intensity was integrated azimuthally for an angle of 25°over the equator.

The instrumental background was determined from an empty scan, where no hair was mounted on the apparatus and resulted in a sharp ring at ∼3 Å. This area was, therefore, excluded from the hair data analysis in the following. The corresponding area is covered by a red circle in the 2-dimensional data.

Effect of shampoo and conditioners

Figure 3 show split comparisons between the original untreated hair and hair with various combinations of shampoo and conditioner treatments. The 2-dimensional intensity maps of the respective samples were cut in half and combined such that the left half shows one sample and the right half a different one. This presentation makes a qualitative comparison between the scans easy. The red rings in each of the 2-dimensional plots mark the position of a known background from the apparatus used, as discussed under Materials and Methods. There are no observable differences in the scattering from untreated hair and hair treated with shampoo or conditioner from untreated and shampooed or conditioned hair samples within the resolution of this experiment, suggesting that likely no internal lipid or keratin structures were significantly altered during the application of shampoo and conditioner. A single application of shampoo, conditioner or the combination of the two does not seem to have an effect on the observed signals within the resolution of this experiment.

Effect of perming

The effect of permanent waving on hair structure is illustrated by the split comparison 2-dimensional intensity map in Fig. 4A. Similar to the shampoo and conditioner trials, position of peaks associated with individual keratin and lipids agree between samples. Differences in peak intensities are observed suggesting that perming effects the molecular organization within the hair.

There are two observable differences between the untreated and the permed samples at the small angle q_‖ region, as shown in Fig. 5. The signals corresponding to this region are best visualized using the small angle X-ray diffraction (SAXD) technique to quantitatively analyze the position and amplitude of signals. The signals at 90 Å, 47 Å and 27 Å were observed in the q_‖ direction of the SAXD data, in agreement with the results of the previous study (Yang, Zhang & Rheinstädter, 2014). As seen in Fig. 5, the difference between untreated and permed hair observed on the 2-dimensional maps and the line graphs is represented by the disproportional increase in the 47 Å signal and slight reduction of the 27 Å signal after perming. This is expected to be the region were keratin intermediate filaments demonstrate packing along the cross-sectional plane of the hair into macro-fibrils (Yang, Zhang & Rheinstädter, 2014). This suggests that structural changes do occur at the intermediate filament level after permanent waving.

Another change is observed after integrating the WAXD data along both the q_‖ and q_z directions. In the q_‖-direction, the peak at 9.5 Å is noticeably decreased in comparison to the 4.3 Å peak after perming. In the q_z-direction, the peak at 5.0 Å is also decreased in comparison to the peak at 4.3 Å. This change is visualized in Fig. 6. Both of these peaks refer to the coiled-coil protein structure, i.e., the α-helical keratin packing distances, while the peak at 4.3 Å refers to lipid packing. The peak at 2.08 Å⁻¹ or ∼3 Å is the result of instrumental background and has no relation to hair structure.

Shampoo and conditioner usage is often advertised to repair damage to hair (“liquid hair repair,” “advanced care and repair” or “keratin repair shampoo”). In this study, we applied combinations of shampoo and conditioner to permanently waved hair to examine whether these hair care products had an impact on the changes incurred during perament waving. As shown in Figs. 4B–4D, the signal in the SAXD region at 47 Å remained after a single application of shampoo or conditioner. Thus, a single application of shampoo and conditioner was found to have no restorative effect to the changes sustained by hair fibres during permanent waving. Moreover, the wide angle region of the 2-dimensional data remains unchanged. As with untreated hair, a single application of shampoo or conditioner to permed hair did not result in changes to internal hair structure on the length scales of the intermediate filaments or keratin proteins studied in this experiment.