Localization of Human Hair Structural Lipids Using Nanoscale Infrared Spectroscopy and Imaging

Localization of Human Hair Structural Lipids Using Nanoscale Infrared Spectroscopy and Imaging Curtis Marcott, a, * Michael Lo, b Kevin Kjoller, b Françoise Fiat, c Nawel Baghdadli, c Guive Balooch, d Gustavo S. Luengo c, * a Light Light Solutions, LLC, P.O. Box 81486, Athens, GA 30608-1484, USA b Anasys Instruments, Inc., 325 Chapala Street, Santa Barbara, CA 93101 USA c L Ore al Research and Innovation, 93600 Aulnay sous Bois, France d L Oréal Research and Innovation, Clark, NJ 07066, USA Atomic force microscopy (AFM) and infrared (IR) spectroscopy have been combined in a single instrument (AFM-IR) capable of producing IR spectra and absorption images at a sub-micrometer spatial resolution. This new device enables human hair to be spectroscopically characterized at levels not previously possible. In particular, it was possible to determine the location of structural lipids in the cuticle and cortex of hair. Samples of human hair were embedded, cross-sectioned, and mounted on ZnSe prisms. A tunable IR laser generating pulses of the order of 10 ns was used to excite sample films. Short duration thermomechanical waves, due to infrared absorption and resulting thermal expansion, were studied by monitoring the resulting excitation of the contact resonance modes of the AFM cantilever. Differences are observed in the IR absorbance intensity of long-chain methylene-containing functional groups between the outer cuticle, middle cortex, and inner medulla of the hair. An accumulation of structural lipids is clearly observed at the individual cuticle layer boundaries. This method should prove useful in the future for understanding the penetration mechanism of substances into hair as well as elucidating the chemical nature of alteration or possible damage according to depth and hair morphology. Index Headings: Infrared microspectroscopy; IR; Atomic force microscopy; AFM; Hair. INTRODUCTION Understanding the morphology of keratin fibers at a chemical level is of great importance from the textile (i.e., wool) and cosmetic industries to forensic research. 1 5 In the case of cosmetics, the knowledge of the structure and composition of keratin fiber is essential to explain the observed specific properties of hair (high tensile strength, resistance, shininess, luster, hydrophobic character, humidity response, etc.). From simple weathering, brushing, and surface care (cleaning, styling, etc.) to more chemical actions (perming, dyeing, bleaching, etc.), the researcher aims at optimizing the process or product effects to assure either the integrity or the modification of a particular hair structure. Hair is an extremely complex structure. As in skin, apart from exogenous sebum from scalp, bound lipids in the hair are remnants of the membranes of mother cells that fully keratinized to produce the intricate structures of hair. Weathering, sun exposure, and strong Received 21 October 2013; accepted 23 December 2013 * Authors to whom correspondence should be sent. E-mail: marcott@ lightlightsolutions.com, gluengo@rd.loreal.com. DOI: 10.1366/13-07328 chemicals can affect the integrity of these lipids. Several studies have attempted to pinpoint the location of lipids in hair fibers. 6 11 The shaft of a typical human hair fiber is approximately 50 100 lm in diameter and consists of a barrier cuticle layer (outer 5 lm), a middle cortex layer (40 95 lm), and an inner medulla when present (5 10 lm). 6 11 The outer cuticle consists of several dense layers of flat, keratinized cells that protect the hair fiber. The cortex, the fibrous component that makes up the bulk of a hair fiber and determines its strength, consists of long embedded cortical cells. It also contains the hair pigment, melanin. Melanin can also be found occasionally in small amounts in the cuticle (endocuticle), giving rise to a local distortion (a bulge) in the normal parallel alignment of the cuticle cell layers. The inner medulla is composed of loosely packed cells in a more open structure with holes; and the proteins contain very little cystine and are cross-linked using isodipeptide bonds. It helps distribute moisture and nutrients to the hair fiber. The medulla may be discontinuous along the hair length or completely absent. It is believed that many ingredients in shampoos and hair conditioners pass through the cuticle barrier layer via a winding route through a thin lipid matrix phase that surrounds the keratinized cells in the cuticle. 10 Thus, the ability to visualize the distribution of lipophilic components in the cuticle is important for gaining a better understanding of penetration pathways of chemical substances into the hair cortex. Vibrational spectroscopy represents an important set of characterization tools that have been applied to the study of hair fibers. Fourier transform infrared (FT-IR) attenuated total reflection (ATR) 11 17 spectroscopy and Raman 11,18 spectroscopy have been used by many researchers to chemically characterize the surface of hair fibers. Fourier transform infrared microspectroscopy is a powerful technique for characterizing the chemical nature of biological tissues. It provides information about the chemical functional groups present and the way they interact with the surrounding matrix. One shortcoming is that spatial resolution of conventional transmission FT-IR microspectroscopy is typically greater than several micrometers. This is due to fundamental diffraction limitations associated with the wavelength of light used to make measurements in the mid-ir spectral range (2.5 14 lm). Several recent studies using ATR objectives coupled with FT-IR microscopes equipped with large format focal-plane array (FPA) mercury cadmium telluride (MCT) detectors have demonstrated 564 Volume 68, Number 5, 2014 APPLIED SPECTROSCOPY 0003-7028/14/6805-0564/0 Q 2014 Society for Applied Spectroscopy

it is possible to monitor the diffusion of substances into hair. However, even when a high refractive index internal reflection element (IRE) such as Ge is used as the ATR objective, it is yet not possible to reach submicrometer spatial resolution. 19,20 Several studies involving high intensity synchrotron radiation sources have reported IR maps of hair cross-sections with spatial resolution up to 3 lm 3 3 lm. 6 9 Recently, highresolution FT-IR spectral images have been reported using an array of 12 synchrotron beams as the source of a spectral imaging system equipped with a 64 3 64 element FPA detector. 21 This system produces excellent signal-to-noise ratio spectra with higher image fidelity than spectral images collected with conventional FT-IR imaging systems. Even though a single pixel of the detector represents a 0.54 lm 3 0.54 lm area on the sample, the results are still diffraction limited. Since many biologically important cellular and histological features are often less than 1 lm in size, obtaining IR spectra and spectral images at higher spatial resolution is highly desirable. For example, it has not been possible to spatially differentiate individual cell layers in the cuticle of hair using any FT-IR microspectroscopic method, including with devices using synchrotron sources. Through combining an atomic force microscope (AFM) with a tunable IR laser source in a single instrument, sub-micrometer-scale measurements of IR absorption as a function of wavenumber can now be obtained and used to help characterize hair crosssections at spatial resolution levels out of reach until now. This approach has been made possible using the photothermal induced resonance effect 22,23 that we refer to here as AFM-IR. In the AFM-IR technique, the radiation from a tunable IR laser source illuminates the sample from underneath through a high refractive index IR transparent prism (IRE) in a configuration similar to ATR. 11 17,19,20 Unlike conventional ATR FT-IR spectrometry, where a highsensitivity MCT single-element or FPA detector senses the internally reflected IR beam, the AFM-IR approach uses a sharp AFM tip attached to a cantilever to detect the rapid thermal expansion of the sample caused by absorption of nanosecond-long pulses of IR radiation at a given wavelength. Wavelengths where the sample absorbs light cause greater heating, and thus quicker expansions, which then kick up the AFM tip and cause the cantilever to ring at its natural vibration frequencies. These motions are sensed with a separate photodetecting sensor. The induced resonance amplitude in the cantilever is proportional to the amount of IR radiation absorbed by the sample. 24 The resulting spectrum is obtained by measuring the ringdown amplitudes while tuning the IR source over a range of wavenumbers. The source can cover a broad range of the mid-ir spectrum, allowing continuous measurements to be made over the range from 900 cm 1 to 3600 cm 1. More details regarding the AFM-IR experiment can be found elsewhere. 24 28 AFM-IR spectra are virtually identical to those obtained with conventional transmission FT-IR spectroscopy, which means they can be digitally searched against readily available IR spectral databases. FIG. 1. End view image of the sample block used to prepare the thin hair cross-sections for analysis. The hair fibers were introduced together tightly bundled inside 1 mm diameter thin plastic cylinders. EXPERIMENTAL Human hair native samples of Caucasian origin were obtained from L Ore al Research Laboratories. The fibers were first cleaned with regular shampoo, rinsed, and dried at room temperature using standard procedures. Special care was taken during the preparation of the hair sections. The hair fibers were introduced together tightly bundled inside 1 mm diameter thin plastic cylinders. This method allowed us to prepare hair sections without any need for the resin to be in direct contact with the hair fibers, thus helping prevent any contamination or interference (Fig. 1). Very smooth, flat, and semi-thin cross-sections of 400 nm thickness were obtained with an UltraMicrotome (Leica Microsystems, Nanterre, France) using a diamond knife (Diatome Ltd., Biel, Switzerland) adapted for histological cuts. The crosssections were then transferred to the ZnSe prism surface using water as a transfer medium. The water subsequently evaporated in ambient air and the hair sections made contact with the prism surface by means of weak intermolecular interactions. The prism was then mounted in the nanoire instrument (Anasys Instruments, Inc., Santa Barbara, CA) for analysis. Only areas deemed to be in good contact with the prism were examined. All AFM topographic images were obtained in contact mode with a CSC37-ALBS lever B (Mikromasch, Tallinn, Estonia). Analysis Studio software (version 3.4, Anasys Instruments) was used for data collection. For the collection of local spectra, a cantilever frequency of 135 khz was used with a width of 60 khz. The IR laser produced 10 ns pulses at a repetition rate of 1 khz. The power levels at 3300 and 2900 cm 1 incident IR radiation on the ZnSe prism were set to 0.5 mw, and the focused laser spot size was about 15 20 lm. Local spectra were collected over a 3200 2800 cm 1 spectral range, using a APPLIED SPECTROSCOPY 565

FIG. 2. Contact mode AFM height image and ten AFM-IR spectra of a human hair cross-section. The colored marker locations on the AFM image are locations where the corresponding AFM-IR spectra of the same color were obtained. The two black spectra at the bottom were collected from the outer cuticle of hair. The four blue spectra were collected from the blue marker locations in the hair cortex. The four red spectra were collected from the red marker locations in the inner medulla of hair. The IR band labeled with an asterisk at 2924 cm 1 is due to the absorbance of the long-chain CH 2 antisymmetric stretching vibration of a lipid component. data point spacing of 4 cm 1. Ringdowns from 256 laser pulses were co-averaged per data point in a given spectrum. Spectral resolution was determined using the laser line width and estimated to be about 8 cm 1. Spatial resolution of the IR spectra collected using AFM-IR is limited by mechanical and thermal properties of the sample. In addition, it can deteriorate for thicker samples. For the 400 nm-thick hair cross-sections used in this study, we estimate that spatial resolution of the IR spectra reported here is approximately 100 nm. Infrared laser power was optimized to minimize any softening due to IR heating, and three to five spectra were collected and co-averaged for each spectrum shown in the figures. All spectra were smoothed using a three-point smoothing routine and normalized using Anasys Instrument s Analysis Studio software. Infrared images were also obtained for the hair crosssections by continuously scanning the AFM tip across the surface while irradiating the same area with a single wavelength of light (2960, 2930, 2924, or 1525 cm 1 ) from the pulsed laser source. The IR images were generated using the peak-to-peak deflection of the ringdown signal after a bandpass filter centered at a frequency of 140 khz and using a frequency window of 60 khz. The power level of the laser incident light on the ZnSe prism was set to 0.5 mw and a scan rate of 0.1 Hz per image line was used. All of the AFM-IR single-wavenumber images shown correspond to an array of 512 3 128 data points. A total of 16 co-averages were collected at each position. Ratios of two IR images are calculated by dividing data points at the same pixel location in Excel and plotting them in Gwyddion (version 2.25, http://gwyddion.net/). Ratio images were constructed by dividing the intensities of the map obtained at 2924 cm 1 by those collected at 2960 cm 1, or by dividing the intensities at 2930 cm 1 by those at 1525 cm 1. The bands at 2924 and 2930 cm 1 are assigned to antisymmetric CH 2 -stretching vibrations in long-chain lipid molecules. The band at 2960 cm 1 is mainly due to a CH 3 antisymmetric stretching vibration in the keratin component of the hair. The 1525 cm 1 band is assigned to the amide II vibration in keratin. Since keratin is present everywhere in the hair, it is helpful to divide out this component in order to better highlight the location of lipids. Gold, Code-V, and Spring gradient scales (version 2.25, http://gwyddion.net/) were used for visualizing AFM, IR, and IR ratio images, respectively. RESULTS AND DISCUSSION Figure 2 shows a contact mode AFM height image and 10 AFM-IR spectra of a human hair cross-section. All of the IR spectra are normalized and offset for clarity, and each spectrum in the figure represents the co-addition of three scans. The colored marker locations on the AFM image are locations where the corresponding AFM-IR spectra of the same color were obtained. The two black spectra at the bottom were collected from the outer cuticle of the hair. The four blue spectra were collected from the blue marker locations in the cortex of the hair, with the bottom blue spectrum being collected closest to the outer cuticle. The four red spectra were collected from the red marker locations in the inner medulla of the hair. The fact that the medulla region of the IR spectrum is white indicates it has the highest elevation. The IR band labeled with an asterisk at 2924 cm 1 is due to the absorbance of the long-chain CH 2 antisymmetric stretching vibration of a lipid component. There are clearly more lipids in the medulla of the hair than anywhere else, consistent with previous studies. 7,9 We also note some differences in the spectral region between 3000 and 3200 cm 1 as a function of spatial position. Bands in 566 Volume 68, Number 5, 2014

FIG. 3. (a) Contact mode AFM height image, (b) AFM-IR single-wavenumber image at 2924 cm 1,(c) AFM-IR single-wavenumber image at 2960 cm 1, and (d) the ratio of the 2924 cm 1 and 2960 cm 1 single-wavenumber images. Higher IR absorbance is observed in the inner medulla and outer cuticle as compared to the middle cortex layer. this region are due to aromatic CH-stretching bands from phenylalanine, tyrosine, and tryptophan groups in the keratin component of the hair. Figure 3a shows a contact mode AFM height image, Fig. 3b shows an AFM-IR single-wavenumber image at 2924 cm 1, Fig. 3c shows an AFM-IR single-wavenumber image at 2960 cm 1, and Fig. 3d shows the ratio of the 2924 cm 1 and 2960 cm 1 single-wavenumber images. Ratioing these two images helps to clarify the location of lipids, as keratin is present everywhere. The 2960 cm 1 band is mainly due to keratin, while the 2924 cm 1 band is mainly due to lipids. The red pixels in Fig. 3d indicate locations of highest lipid concentration. Each pixel in these images corresponds to a 156 nm 3 273 nm area of the sample. Higher IR absorbance at 2924 cm 1 for the lipid component is found in the inner medulla and outer cuticle as compared with the middle cortex layer. There are also some pockets of higher lipid concentration distributed throughout the cortex. Lipids are evidently present in the medulla as expected. 7,9 As for the pockets of lipids in the cortex, these might be present in the intermacrofibrillar matrix. It is worth bearing in mind that as the hair hardens with the laying down of the keratin composite (microfibrils þ matrix) in the cortex, the debris, including remnant cell nuclei, get pushed into the space between the macrofibrils. These debris FIG. 4. Contact mode AFM height image and 12 AFM-IR spectra of a human hair cross-section near the interface between the outer cuticle and cortex layers. The colored marker locations on the AFM image are locations where the corresponding AFM-IR spectra of the same color were obtained. The spectra from bottom to top correspond to colored markers from bottom left to the top right along the ordinate axis of the image. The IR band labeled with an asterisk at 2924 cm 1 is due to the absorbance of the long-chain CH 2 antisymmetric stretching vibration of a lipid component. APPLIED SPECTROSCOPY 567

FIG. 5. AFM-IR single-wavenumber intensity ratio images (2930 cm 1 /1525 cm 1 ) collected from cross-sections of three different human hair samples from the same source. The IR band located at 2930 cm 1 is due to the absorbance of the long-chain CH 2 antisymmetric stretching vibration of the lipid component, while the band at 1525 cm 1 is due to the absorbance of the amide II vibration of the hair keratin protein. The white-colored areas are locations of higher lipid concentration. contain numerous lipids, which likely originate from the lipid-containing mitochondria, endoplasmic reticulum, and other cell organelles. Figure 4 shows a contact mode AFM height image and 12 AFM-IR spectra of a human hair cross-section obtained near the interface between the outer cuticle and cortex layers. The colored marker locations on the AFM image are locations where the corresponding AFM- IR spectra of the same color were obtained. The IR band labeled with an asterisk at 2924 cm 1 is due to the absorbance of the long-chain CH 2 antisymmetric stretching vibration of a lipid component. The spectra show some variation as a function of spatial location, which can be more clearly visualized using single-wavenumber images with closer data point spacing. It also appears that the aromatic CH-stretching bands in the cortex (between 3000 and 3200 cm 1 ) are somewhat sharper than those same bands are in the cuticle. This suggests that the keratin component may be more disordered in the cuticle. Figure 5 shows three AFM-IR single-wavenumber intensity ratio images (2930 cm 1 /1525 cm 1 ) collected from cross-sections of three different human hair samples from the same source. The IR band located at 2930 cm 1 is due to the absorbance of the long-chain CH 2 antisymmetric stretching vibration of the lipid component, while the band at 1525 cm 1 is due to the absorbance of the amide II vibration of the hair keratin protein. The white-colored regions are locations of higher lipid concentration. Each pixel in these images corresponds to a 20 nm 3 40 nm area of the sample. Clear evidence of higher accumulation of lipids at the interface between the individual cuticle layers is observed in all three of the images. White lipidic regions in the three pictures of Fig. 5 are consistent with being in the cuticular cell membrane complex (CMC) and endocuticle of each cell layer, i.e., correct location and thickness of each layer. 10 Lipids are not only present in the CMC, but also extend to some extent into the endocuticle. These could be organelle lipids. It is still difficult for lipids to be separately detected from the CMC immediately adjacent to the endocuticle. The thickness of the layers in the images displayed in Fig. 5 is too great to be just the lipids of the CMC. Additional experiments at higher spatial resolutions will be needed for deeper analysis. In any case, this level of spatial detail with IR microspectroscopy is unprecedented, even when using devices with high brightness synchrotron radiation sources. 8,29 Future detailed studies in the amide I region between 1720 and 1600 cm 1 should show differences in the keratin secondary structure as a function of position with much higher spatial resolution than is obtainable with synchrotron FT-IR experiments. 29 CONCLUSION A new approach using an AFM tip to detect light absorption from a pulsed tunable IR laser source enables human hair to be spectroscopically characterized at unprecedented levels. Study of thin crosssections of human hair samples indicates that medulla have a higher lipid content than outer cortex and cortex that make up the bulk of hair. AFM-IR images, which compare absorbances at 2930 cm 1 (lipid) and 1525 cm 1 (keratin), indicate that the lipid component in the cuticle barrier layer is at its highest at the boundary between the individual cuticle layers. This work suggests that future AFM-IR studies may prove useful for understanding penetration pathways of specific ingredients in topically applied drugs, dyes, or hair care products through the hair cuticle and the cuticle cortex boundary. ACKNOWLEDGMENTS We thank NIST-ATP Award 70NANB7H7025 and the National Science Foundation award NSF-SBIR 0750512 for providing the financial 568 Volume 68, Number 5, 2014

support. We also thank Anne Potter and Céline Farcet (L Ore al) for their kind support for the realization of this project. 1. R.P. Oertel. Protein Conformational Changes Induced in Human Stratum Corneum by Organic Sulfoxides: An Infrared Spectroscopic Investigation. Biopolymers. 1977. 16(10): 2329-2345. 2. J. Strassburger, M.M. Breuer. Quantitative Fourier Transform Infrared Spectroscopy of Oxidized Hair. J. Soc. Cosmet. Chem. 1985. 36(1): 61-74. 3. K.S. Kalasinsky, J. Magluilo, T. Schaefer. Hair Analysis by Infrared Microscopy for Drugs of Abuse. Forensic Sci. Int. 1993. 63(1-3): 253-260. 4. K.S. Kalasinsky, J. Magluilo, T. Schaefer. Study of Drug Distribution in Hair by Infrared Microscopy Visualization. J. Anal. Toxicol. 1994. 18(6): 337-341. 5. K.S. Kalasinsky. Drug Distribution in Human Hair by Infrared Microscopy. Cell. Mol. Biol. 1998. 44(1): 81-88. 6. J.L. Bantignies, G. Fuchs, G.L. Carr, G.P. Williams, D. Lutz, S. Marull. Organic Reagent Interaction with Hair Spatially Characterized by Infrared Microspectroscopy Using Synchrotron Radiation. Int. J. Cosmet. Sci. 1998. 20(6): 381-394. 7. P. Dumas, L. Miller. The Use of Synchrotron Infrared Microspectroscopy in Biological and Biomedical Investigations. Vib. Spectrosc. 2003. 32(1): 6-21. 8. P. Dumas, M.L. Tobin. A Bright Source for Infrared Microspectroscopy: Synchrotron Radiation. Spectrosc. Eur. 2003. 15(6): 7-23. 9. L. Kreplak, F. Briki, Y. Duvault, J. Doucet, C. Merigoux, F. Leroy, J.L. Leveque, L. Miller, G.L. Carr, G.P. Williams, P. Dumas. Profiling Lipids Across Caucasian and Afro-American Hair Transverse Cuts, Using Synchrotron Infrared Microspectrometry. Int. J. Cosmet. Sci. 2001. 23(6): 369-374. 10. J.A. Swift. Morphology and Histochemistry of Human Hair. In: P. Jolles, H. Zahn, H. Hocker, editors. Formation and Structure of Human Hair. Basel, Switzerland: Birhauser Verlag, 1997. Pp. 166. 11. G.J. Zhang, L. Senak, D.J. Moore. Measuring Changes in Chemistry, Composition, and Molecular Structure Within Hair Fibers by Infrared and Raman Spectroscopic Imaging. J. Biomed. Opt. 2011. 16(5): 056009. doi: 10.1117/1.3580286. 12. K.L.A. Chan, K.G. Kazarian. New Opportunities in Micro- and Macro-Attenuated Total Reflection Infrared Spectroscopic Imaging: Spatial Resolution and Sampling Versatility. Appl. Spectrosc. 2003. 57(4): 381-389. 13. K.L.A. Chan, K.G. Kazarian, A. Mavraki, D.R. Williams. Fourier Transform Infrared Imaging of Human Hair with a High Spatial Resolution Without the Use of a Synchrotron. Appl. Spectrosc. 2005. 59(2): 149-155. 14. K.L.A. Chan, F.H. Tay, C. Taylor, K.G. Kazarian. A Novel Approach for Study of In Situ Diffusion in Human Hair Using Fourier Transform Infrared Spectroscopic Imaging. Appl. Spectrosc. 2008. 62(9): 1041-1044. 15. K. Lau, M.A.B. Hedagaard, J.E. Kloepper, R. Paus, B.R. Wood, V. Deckert. Visualization and Characterisation of Defined Hair Follicle Compartments by Fourier Transform Infrared (FTIR) Imaging Without Labeling. J. Dermatol. Sci. 2001. 63(3): 191-198. 16. K.M. Elkins. Rapid Presumptive Fingerprinting of Body Fluids and Materials by ATR FT-IR Spectroscopy. J. Forensic Sci. 2011. 56(6): 1580-1587. 17. X. Wang, Z. Qi, X. Liu, S. Wang, C. Li, G. Liu, Y. Xiong, T. Li, J. Tao, Y. Tian. The Comparison of Hair from Gastric Cancer Patients and from Healthy Persons Studied By Infrared Microspectroscopy and Imaging Using Synchrotron Radiation. Cancer Epidemiol. 2010. 34(4): 453-456. doi:10.1016/j.canep.2010.03.016. 18. V.F. Kalasinsky. Biological Applications of Infrared and Raman Microscopy. Appl. Spectrosc. Rev. 1996. 31(1-3): 193-249. 19. K.G. Kazarian, K.L.A. Chan. Micro- and Macro-Attenuated Total Reflection Fourier Transform Infrared Spectroscopic Imaging. Appl. Spectrosc. 2010. 64(5): 135A-152A. 20. H.J. Gulley-Stahl, S.B. Bledsoe, A.P. Evan, A.J. Sommer. The Advantages of an Attenuated Total Internal Reflection Infrared Microspectroscopic Imaging Approach for Kidney Biopsy Analysis. Appl. Spectrosc. 2010. 64(1): 15-22. 21. M.J. Nasse, M.J. Walsh, E.C. Mattson, R. Reininger, A. Kajdacsy- Balla, V. Macius, R. Bhargava, C.J. Hirschmugl. High-Resolution Fourier-Transform Infrared Chemical Imaging with Multiple Synchrotron Beams. Nature Methods. 2011. 8(5): 413-416. 22. A. Dazzi, F. Glotin, R. Carminati. Theory of Infrared Nanospectroscopy by Photothermal Induced Resonance. J. Appl. Phys. 2010. 107(12): 124519. 23. A. Dazzi, R. Prazeres, F. Glotin, J.M. Ortega, M. Al-Sawaftah, M. de Frutos. Chemical Mapping of the Distribution of Viruses into Infected Bacteria with a Photothermal Method. Ultramicroscopy. 2008. 108(7): 635-641. 24. B. Lahiri, G. Holland, A. Centrone. Chemical Imaging Beyond the Diffraction Limit: Experimental Validation of the PTIR Technique. Small. 2013. 9(3): 439-445. 25. C. Prater, K. Kjoller, D. Cook, R. Shetty, G. Meyers, C. Reinhardt, J. Felts, W. King, K. Vodopyanov, A. Dazzi. Nanoscale Infrared Spectroscopy of Materials by Atomic Force Microscopy. Microscopy Analysis. 2010. 24(3): 5-8. 26. A. Dazzi, C.B. Prater, Q. Hu, D.B. Chase, J.F. Rabolt, C. Marcott. AFM IR: Combining Atomic Force Microscopy and Infrared Spectroscopy for Nanoscale Chemical Characterization. Appl. Spectrosc. 2012. 66(12): 1365-1384. 27. C. Marcott, M. Lo, K. Kjoller, C. Prater, I. Noda. Spatial Differentiation of Sub-Micrometer Domains in a Poly(hydroxyalkanoate) Copolymer Using Instrumentation that Combines Atomic Force Microscopy (AFM) and Infrared (IR) Spectroscopy. Appl. Spectrosc. 2011. 65(10): 1145-1150. 28. C. Marcott, M. Lo, K. Kjoller, Y. Domanov, G. Balooch, G.S. Luengo. Nanoscale Infrared (IR) Spectroscopy and Imaging of Structural Lipids in Human Stratum Corneum Using an Atomic Force Microscope (AFM) to Directly Detect Absorbed Light from a Tunable IR Laser Source. Exp. Derm. 2013. 22(6): 419-421. 29. L. Kreplak, J. Doucet, P. Dumas, F. Briki. New Aspects of the a- Helix to b-sheet Transition in Stretched Hard a-keratin Fibers. Biophys. J. 2004. 87(1): 640-647. APPLIED SPECTROSCOPY 569