We present extended spectroscopic analysis of pharmaceutical tablets in the close near infrared spectral range performed using broadband photon time-of-flight (PTOF) absorption and scattering spectra measurements. We show that the absorption spectra can be used to perform evaluation of the chemical composition of pharmaceutical tablets without need for chemo-metric calibration. The spectroscopic analysis was performed using an advanced PTOF spectrometer operating in the 650 to 1400 nm spectral range. By employing temporal stabilization of the system we achieve the high precision of 0.5% required to evaluate the concentration of tablet ingredients. In order to further illustrate the performance of the system, we present the first ever reported broadband evaluation of absorption and scattering spectra from pure and doped Spectralon®.
© 2013 Optical Society of America
Diffuse optical spectroscopy (DOS) applied to evaluation of absorption and scattering in turbid materials is an indispensable component in the modern photonics toolbox. Applications of DOS range from biomedical diagnostics [1–5] and medical treatment monitoring , to quality control and product analysis in food [7–10], pharmaceutical [11–14], and timber [15, 16] industries.
The key advantage of near-infrared (NIR) DOS spectroscopy in industrial applications is that, combined with chemo-metric analysis, it is capable of resolving subtle variations in the chemical composition, and in the physical, structural, and morphological properties of different kinds of turbid samples, without the need for costly and lengthy sample preparation [7, 13]. Furthermore, NIR DOS spectroscopy is relatively cheap, fast, and easy to implement, and can be set up for remote operation. As a result, it is suitable for on-line process monitoring and quality control .
The main challenge associated with DOS measurements is differentiating between the effects of absorption and scattering in evaluated optical attenuation. The precise measurement of absorption is essential for the accurate evaluation of the chemical composition of a sample. Scattering can be utilized for the characterization of the structural and morphological properties of the sample. In practice, the problem of intermixing of absorption and scattering in steady-state extinction measurements is often circumvented by employing data pretreatment algorithms (such as multiplicative scatter correction and standard normal variate) in combination with elaborate chemo-metric modeling . Although efficient in many cases, this approach lacks generality and, moreover, requires development and maintenance of costly calibration databases. The cost and complexity of chemo-metric analysis could be reduced, and its robustness with regard to unexpected variations in sample properties improved, by the use of advanced DOS techniques, which allow the independent evaluation of absorption and scattering spectra from turbid samples.
Many approaches have been used for the evaluation of absorption and scattering in turbid samples, including spatially and spectrally resolved CW techniques [18–21], frequency  and time domain  measurements, as well as combinations of these . The question regarding superiority of any of them is still open and likely be dependent on a particular application. The time domain technique, which is also known as photon time-of-flight (PTOF) spectroscopy, is currently prevalent for implementation in broadband instruments, and numerous systems have been demonstrated in the past decade [25–29]. In particular, a wide bandwidth PTOF system, based on singe-photon counting in an InGaAs avalanche photodiode, operating at wavelengths up to 1700 nm, was recently demonstrated .
PTOF measurements on small, highly scattering samples such as pharmaceuticals [31, 32] in numerous aspects differ from more conventional measurements on comparatively low-scattering, tissue-like samples [33, 34]. Firstly, tight control over the experimental arrangement is considerably more critical, as relatively small (a few tens of microns) uncertainties in fiber positioning result in comparatively large errors (up to a few percent) in the evaluated optical parameters. This is important for the analysis of pharmaceuticals where even small variations in absorption coefficient affect the estimates of the chemical composition. Secondly, PTOF measurements are typically performed in the time correlated single-photon-counting (TCSPC) regime, which is a rather slow technique. Besides PTOF measurements require accurate monitoring of the instrument response function (IRF) . Temporal system stability is thus critical to avoid the need for repeated calibration, while maintaining the high measurement precision required for pharmaceutical analysis. Thirdly, there is a lack of “gold standards” with calibrated scattering properties that can be used for the verification of system performance.
Although the potential of PTOF spectroscopy in pharmaceutical analysis was recognized long ago [31, 32], little has been done so far to assess the performance and potential of novel broadband PTOF systems  for the analysis of solid pharmaceuticals. In order to facilitate the development of the advanced optical techniques minimizing the need in extensive chemo-metric calibrations here we apply PTOF technique for performing spectroscopic analysis on pharmaceuticals. We show that tablet absorption spectra evaluated using PTOF technique can be used for direct calibration-free evaluation of the tablet chemical composition, whereas the scattering spectra are directly correlated to the tablet microstructure. In order to obtain the measurement precision required to perform pharmaceutical analysis, the original system  was upgraded with a timing stabilization scheme that allowed the suppression of temporal drift. Additionally, special care was taken to maintain precise control over the measurement arrangement, which further reduced measurement errors. We also used the instrument to evaluate the optical spectra from doped and pure Spectralon®, as a prospective candidate for a standard calibration material.
2. Materials and methods
2.1 System design
The operation principle of the spectrometer is based on monitoring of the PTOF distribution through a turbid sample. Fitting the PTOF distribution with the appropriate model of the light propagation in turbid media enables to determine the absorption coefficient and the reduced scattering coefficient . The reduced scattering coefficient is defined as where g is scattering anisotropy, and is scattering coefficient which is reciprocal to scattering mean free path. Scanning the wavelength for which PTOF distribution is measured and fitted allows and spectra to be evaluated.
A schematic of the spectrometer is given in Fig. 1. A photonic crystal fiber supercontinuum source (PCF SCS) (Model SC500-5, Fianium Ltd, Southampton, UK) driven at an 80 MHz repetition rate is used to provide probe pulses. Depending on the wavelength interval, one of two computer-controlled acousto-optical tunable filters (AOTFs) is used to slice a spectrally narrow probe pulse from the broadband supercontinuum. The operating ranges of the two AOTFs are 650 nm–1000 nm and 950 nm–1800 nm. The time width of the resulting probe pulse is shorter than 30 ps. The spectral width of the probe pulse is wavelength-dependent, and is determined by the AOTF parameters. It increases from ~3 nm at 650 nm to approximately 12 nm at 1300 nm. The output of the AOTF is coupled to a short piece of narrow-core (10 µm) fiber (SMF-28) that is used as an AOTF pinhole. The SMF-28 fiber is connected to a custom-made gradient index multimode fiber (400 µm/640 µm core/cladding diameter, Leoni Fiber Optics, Germany). The gradient index fiber is used to guide the signal to and from the sample, while restricting the multimode dispersion to a minimum. A small fraction of the probe signal is split off from the probe light before the sample and routed directly to the detector via dedicated optical path. This timing reference pulse (TRP) is used to ensure synchronization of the IRF and PTOF measurements taken at different times. Both the signal and the timing reference light paths include dedicated attenuators (OZ Optics Ltd, Ottawa, Canada), which are used to adjust the power to a level appropriate for single-photon counting (SPC) measurements. Depending on the wavelength range chosen, one of two SPC detectors is used in combination with the TCSPC electronics (SPC-130 Becker & Hickl, Berlin, Germany) for time-resolved acquisition of PTOF distribution through the sample. In the range 650 nm –1000 nm we employed a silicon SPC avalanche photodiode (APD) (PD1CTC Micro Photon Devices, Bolzano, Italy). A NIR-extended micro-channel plate (MCP) photomultiplier tube (PMT) (R3809U-68 Hamamatsu Photonics, Hamamatsu, Japan) is used for detection in the 950 nm – 1400 nm range. Both detectors are protected by shutters (OZ Optics, and Thorlabs, Newton, USA). A trans-impedance (TI) amplifier (HFAC-26, Becker & Hickl) is used to match MCP-PMT to the TCSPC electronics. The TI amplifier also controls PMT protecting shutter. The setup is controlled by a personal computer (PC), which is also used for data collection. The overall spectral range of the present setup is 650 nm to 1400 nm.
The absorption and reduced scattering coefficients spectra were evaluated using the original Matlab Script implementing the Levenberg–Marquardt fitting algorithm. Depending on the experimental geometry (infinite, semi-infinite, or slab) and the range of optical parameters, either the analytical diffusion model [35, 36] or a model based on a predefined Monte-Carlo simulation database [37, 38] was used to fit the PTOF distribution. To account for the finite temporal width of the probe pulse, the finite detector response time, and residual fiber dispersion, the fitting model was convolved with IRF measured with a piece of double-sided, black-printed office paper instead of the sample .
The linearity of the system was verified by carrying out adding absorber and scattering solution series measurements similar to those described previously . Furthermore, the optical properties of MEDPOT phantoms  evaluated with the present system are consistent with previously reported values .Spectral stability of AOTF was monitored on everyday basis while operating the setup and residual drifts were compensated.
2.2 Tablet holder
Tight control over experimental arrangement is a key factor in ensuring high precision and repeatability of the measurements. In order to ensure the highest possible precision and repeatability in fiber positioning, a tablet sample holder was constructed from a black metal tube and two black plates with FC/PC fiber connector terminals (Thorlabs, parts SM1L10 and S120-FC). The tablet was centered inside the tube using a homemade spacer, and then sandwiched between the FC/PC connector plates providing an interface for source and collection fibers terminated with FC/PC connectors. The holder provides a fiber positioning precision of ~10 µm.
2.3 Pharmaceutical samples
The pharmaceutical samples used in the present study were part of a large pharmaceutical test tablet set  prepared from ibuprofen, mannitol and magnesium stearate mixed in different proportions. In the experimental design ibuprofen was used to represent the active pharmaceutical ingredient (API) and mannitol was used as the filler. In order to vary the scattering properties of the tablets irrespectively of their absorption, ibuprofen and mannitol were taken with the different average particle size. The average particle sizes of the ingredients were evaluated by laser diffraction spectroscopy assuming spherical particle shape. Table 1 provides further details on the tablet ingredients used and Table 2 lists the composition of the multi-component tablets used within this study. In order to enable concentration evaluation five single-component tablets were prepared from pure ingredients. All the multi-component tablets were manufactured with a single punch press (Korsch EK 0, Korsch AG, Berlin, Germany) equipped with flat, round 10 mm punches using compression forces 8-16 kN. The single-component tablets were compressed in a manual IR tablet press (Specac, Oprington, UK) using a compression force of 2 ton. All tablets were flat discs, 10 mm in diameter, had a thickness of ~3 mm, and weighed ~300 mg. Further details on the tablet preparation can be found in reference .
The actual concentration of the tablet samples used within the present study was estimated in the following way. First Raman spectra were measured for all the samples within the original large tablet set. Latter on a subset consisting of one third of the samples (138 tablets) that accurately represented parameter distribution within the original set was sacrificed in order to accurately determine the chemical composition by dedicated procedure including dissolution, filtration and UV-VIS concentration analysis . Finally the Raman spectra and the chemical composition of the subset were used to design a chemo-metric model that was eventually used to estimate the ingredient concentration in the remaining two thirds of the tablets of the large set including the tablets used in the present study. Based on the residual error of the original chemo-metric model the error in the determined ingredient concentration was estimated to be less than 1% w/w. In the following we refer to this estimated ingredient concentration as the reference concentration. For evaluation of the optical properties we assumed tablet refractive index of 1.5.
2.4 Spectralon samples
Pure and doped samples of Spectralon® were acquired from Labsphere (2” WCS-MC-INSERT p/n SM-00556-400). Spectralon is the trade name for a type of optical Teflon with exceptionally high diffuse reflectance, which means it is highly scattering and has low absorption. The material is also available doped with rare-earth oxides, which results in sharp absorption lines in the visible and close NIR regions. The supplier claims high sample stability and minimal batch-to-batch variation, which makes Spectralon a good candidate for calibration samples. Further precautions like storing in dry atmosphere may be recommended to avoid residual reference drifts. For the present experiments we used 13 mm diameter discs with thicknesses 2.88 mm and 1.3 mm for pure and doped Spectralon® samples, respectively.
3.1 Temporal response of the system
The wavelength dependence of the full width at half maximum (FWHM) of the IRF for the present setup is shown in Fig. 2. The IRF FWHM decreased from ~60 ps at 650 nm to ~45 ps at 800 nm, and then remained constant over the whole measurement range of the setup up to 1400 nm. In the wavelength range 950 nm to 1050 nm the PMT detector causes IRF broadening up to 95 ps. This does not cause any noticeable deterioration in system performance as the faster silicon SPC APD can be used in this range.
3.2 Evaluation of SCS drift and timing stabilization
Precise monitoring of the IRF is crucial for implementing high-quality PTOF measurements . By monitoring the temporal position of the IRF we observed that the time delay between the SCS optical pulse and the SCS electronic synchronization output drifted at a rate of ~3 ps / hour. In case of pharmaceutical samples this caused noticeable fluctuations in the absorption and reduced scattering coefficients. In order to suppress this temporal drift, and to enhance the precision of the present setup, we implemented a double-path optical scheme that enables synchronization of the IRF and PTOF distribution via a TRP incorporated into both signals. While evaluating our PTOF data we ensured synchronization of the IRF and PTOF distribution by precisely matching the temporal position of the TRP in both signals. The results of this procedure are depicted in Fig. 3, which shows the absorption and scattering coefficients evaluated from continuous measurements of a tablet over two hours. In this evaluation we used a single IRF determined directly prior to the tablet measurements. It is apparent from the Fig. 3 that the uncertainty in IRF timing caused by the temporal drifts resulted in a precision in the determination of optical parameters of ~3%. When the timing reference pulse was used for IRF and PTOF distribution synchronization a precision of 0.5% was obtained.
3.3 Spectroscopic characterization of pharmaceutical samples
In order to apply PTOF technique for the characterization of pharmaceutical samples, we measured and analyzed spectra of the absorption coefficient and the reduced scattering coefficient for a test set consisting of three single-component and eleven multi-component tablets. Typical examples of and spectra corresponding to the single-component and multi-component tablets are shown in Fig. 4a and 4b, respectively. Slab extension of diffusion model  was used to evaluate the data. The absorption spectra corresponding to multi-component tablets were fitted with linear combination of the absorption spectra of the pure ingredients. The scattering spectra of all the tablets were fitted by dependence ( = 1 nm), typical for Mie scattering by particles with narrow size distribution. The respective scattering fit examples are also depicted in Fig. 4(b).
3.4 Evaluation of absorption and reduced scattering coefficient spectra of Spectralon
Optical absorption and reduced scattering coefficients spectra of pure and doped Spectralon samples are depicted in Figs. 5(a) and 5(b), respectively. In both cases, a refractive index of 1.35  was used for data evaluation. As expected the pure Spectralon exhibited high scattering and very low absorption. The reduced scattering coefficient of the sample is in the range 350–450 cm−1 and can be well approximated by dependence, as shown in the Figs. 5(a) and 5(b).
The absorption spectra of the Spectralon doped with rare-earth oxides (cf. Figure 5(b)) exhibited relatively narrow spectral absorption lines, corresponding to the absorption of rare-earth elements. The reduced scattering coefficient of the doped sample is larger than that of the pure sample, suggesting a higher concentration of smaller scattering centers in the doped sample as a result of doping procedure. The spectrum of the reduced scattering coefficient spectrum of the doped sample shows artifacts caused by the limited spectral resolution of the present setup. It has previously been pointed out that highly non-uniform sample transmission, caused by steep variation in sample absorption, amplified by high scattering, results in considerable spectral distortions in the propagating probe pulses . In extreme cases, spectral side-lobes of the probe pulse leaking through the AOTF, rather than the main peak itself, become the dominant detected component. Fitting the shape of the spectrally distorted probe pulse results in the characteristic artifacts observed in absorption and reduced scattering coefficients spectra. Similar to conventional CW absorption spectra, finite spectral resolution results in spectral broadening of the absorption lines. The corresponding effect in scattering spectra is seen as characteristic deeps corresponding to sharp absorption lines, as can be seen in Fig. 5(b). It is worthwhile noting at this point, that high Spectralon scattering (>300 cm−1) and sufficient sample thickness >1 mm justify the use of the diffusion model for the evaluation of the optical parameters, as was done in this study.
4.1 General aspects
The aim of the present study was to apply the broadband PTOF spectroscopy for the analysis of pharmaceuticals. Below we first discuss the characteristics of the broadband PTOFS system we designed for pharmaceutical analysis and the corresponding measurements procedure. Since the general factors determining the spectroscopic performance of PTOF instruments have been well discussed in dedicated literature [27, 39] here we only focus on the aspects that were essential to achieve sufficiently high precision of PTOF measurements to enable the evaluation of tablet properties. Further on we discuss how decoupled absorption and scattering spectra obtained with PTOF spectroscopy can be used for the evaluation of the chemical composition and morphology of tablets. We conclude with a discussion of the spectra obtained from the Spectralon samples.
It is well known that precise timing between a recorded PTOF distribution and the IRF is essential for high precision in PTOF measurements . The problem is aggravated in the case of measurements on small pharmaceutical samples, where the typical PTOF distribution has a FWHM of about 150 ps, and an uncertainty in the IRF position of even a few picoseconds may cause considerable errors in the absorption and scattering coefficients. Furthermore, it is desirable, for practical reasons, to be able to use one IRF for evaluation of several PTOF distributions acquired during a day. We found that a temporal drift of 3-4 ps/hour between the electronic synchronization signal and the SCS optical pulses was common in the SCSs currently available on the market. Such drift may not pose problems when a measurement configuration resulting in a broader PTOF distribution can be used , or when a 3%-5% fluctuations in and can be neglected. In the present case, using small tablets and small tolerances on and , the SCS drift is indeed deleterious. The timing stabilization scheme utilized in our system enables stable synchronization between IRF and PTOF measurements, thus enabling accurate measurements, as reported above.
Tight control over experimental arrangement is another crucial factor in ensuring high precision and repeatability of the PTOF measurements . This becomes especially critical in experiments with small, highly scattering samples, such as pharmaceutical tablets with a typical thickness of ~3 mm, as even a very small uncertainty in the positioning of the source and collection fibers (~100 µm) will result in comparatively high relative errors in the determination of the optical parameters. The use of FC/PC connectors on the tablet holder provided a ~10 µm repeatability in fiber positioning, which greatly improved the precision of the measurements.
4.2 Analysis of tablet chemical composition
The linear decomposition of the multi-component tablet absorption spectrum by absorption spectra of pure ingredients enables to estimate partial light-paths in each ingredient and eventually determine ingredient concentration. Indeed, the absorption measured in the multi-component tablet at a wavelength is given by:Fig. 6(a) versus the reference API concentration. The root mean square of the absolute difference between the evaluated and reference concentrations is 1.2% which corresponds to ca. 6% relative error at the middle of concentration span.
It is important to note high fitting quality of multi-component tablet absorption by absorption of the ingredients. This as well as an accurate correspondence between the evaluated and the reference concentrations enables to verify linearity of the present setup. As the absorption spectra from the tablets were measured at very different levels of scattering, possible nonlinearities and residual absorption-scattering coupling would introduce considerable distortions in the absorption spectra, preventing accurate fitting and evaluation of the concentration. The high accuracy of evaluated tablet concentration suggests that the present setup enables linear spectra evaluation with the minimal spectral distortion.
The chemical composition of typical commercial medicines may be far more complex than in the simple three-component tablets investigated here [12, 13]. For this reason, the linear decomposition analysis presented here may not be generally practical and evaluation of the chemical composition of the tablet may still require standard calibration-based chemo-metric modeling [11, 14]. Nonetheless, the extent and complexity of chemo-metric calibration required for the modeling will be greatly reduced due to decoupling the absorption and the scattering from each other. This will lead to reduced cost and enhanced robustness of the analysis.
4.3 Tablet scattering analysis
The scattering spectrum of a sample is related to the sample micro-structure [43, 44]. All the scattering spectra observed in this study could be fitted by dependence, typical for Mie scattering. As could be anticipated from Mie theory, scattering from the multi-component tablets made from the ingredients with the smaller particle size showed steeper dependence on the wavelength (i.e. larger ) than the tablets made from ingredients with larger particle size. Furthermore, the tablets made with the smaller particles exhibited stronger scattering than those made with the large particle ingredients. The single-component tablets prepared from pure ingredients using other equipment than multi-component tablets fell out from this trend. In order to obtain better insight into these effects is plotted vs. for all the fitted scattering spectra (Fig. 6(b)). The average size of the particles used in the tablets is also given in the Fig. 6(b). It is apparent that the data for the multi-component tablets is well-fitted by A = 143.8 and = 2094 nm. This is a peculiar observation, however further experiments certainly needed to completely rule out possible coincidence. Alternatively it may be hypothesized that the trend is a result of the compression of the granules of different sizes prepared from very similar material. Irregularities in single component samples may thus be attributed to differences in parameters of the tablet compression. It is known from rigorous Mie scattering theory that the parameters A and for a poly-disperse set of scatterers such as the pharmaceuticals considered here, are indeed related to scatterer concentration and size distribution . This can be used for the analysis of sample micro-structure. The particles prepared from typical pharmaceutical ingredients have highly complex scattering structure (rather than being simple homogeneous spheres) therefore extensive computational effort may be needed to relate observed scattering spectra to particle parameters from the first principles.
4.4 Spectralon spectra
High stability and the low batch-to-batch variability claimed by the supplier  make Spectralon and its analogs (Flourilon® by Avian Technologies  and Zenith Polymer® by SphereOptics GmbH ) an interesting candidate for system calibration and inter-laboratory comparison standard in the range of optical parameters of interest for analyzing pharmaceuticals. The material is also available doped with spectrally uniform and non-uniform absorbers. Doped Spectralon is particularly interesting for the assessment of system performance as it is a rather demanding sample to analyze. Indeed, the combination of moderate absorption and high scattering leads to very large variation of the FWHM of the PTOF distribution in the visible and close NIR range. Figure 2 shows the FWHM of the PTOF distribution for the 1.55 mm thick Spectralon sample together with the IRF FWHM. In order to enable measurements in the vicinity of the 1270 nm absorption peak, it is necessary to optimize the thickness of the sample, while taking the parameters of the system into account. The optimal sample thickness is a trade-off between the maintaining acceptably low sample extinction while achieving sufficiently broad FWHM of the PTOF distribution, both of which increase with increasing sample thickness. Precise evaluation of the absorption and scattering in the rare-earth doped spectralon sample challenges multiple PTOF system parameters such as probe pulse spectral line-width, IRF time-width and probe pulse power. The material is thus suitable for integral assessment of system performance and for inter-laboratory equipment comparison.
Here we present first ever reported spectroscopic analysis of the pharmaceutical tablets in the 950 nm - 1350 nm wavelength range performed on the basis of the decoupled absorption and scattering spectra. The spectra of absorption and reduced scattering coefficients are evaluated using the broadband PTOF spectrometer optimized for pharmaceutical analysis. The superior linearity and precision of the spectrometer enable us to evaluate drug concentration in the tablets with an average accuracy of ~1.2 (% w/w). The concentration analysis is based on linear decomposition of the mixed tablets absorption spectra by absorption of individual ingredients and notably does not require any type of chemo-metric calibration. The tablet scattering spectra observed during the study show clear correlation with the tablet ingredients particle size and fabrication parameters that may serve as a basis for novel pharmaceutical analysis techniques. We also present, for the first time, absorption and scattering spectra from pure and rare-earth-oxide-doped Spectralon, demonstrating that these materials can be used as standards in both system calibration and inter-laboratory comparisons.
We gratefully acknowledge Madeleine Hartman for tablet samples preparation. Doug Baxter is gratefully acknowledged for help in getting doped spectralon samples. D. Khoptyar gratefully acknowledges AstraZeneca for financial support.
References and links
1. T. Durduran, R. Choe, W. B. Baker, and A. G. Yodh, “Diffuse optics for tissue monitoring and tomography,” Rep. Prog. Phys. 73(7), 076701 (2010). [CrossRef]
2. P. Taroni, A. Pifferi, G. Quarto, L. Spinelli, A. Torricelli, F. Abbate, A. Villa, N. Balestreri, S. Menna, E. Cassano, and R. Cubeddu, “Noninvasive assessment of breast cancer risk using time-resolved diffuse optical spectroscopy,” J. Biomed. Opt. 15(6), 060501 (2010). [CrossRef] [PubMed]
4. A. Pifferi, A. Farina, A. Torricelli, G. Quarto, R. Cubeddu, and P. Taronia, “Time-domain broadband near infrared spectroscopy of the female breast: a focused review from basic principles to future perspectives,” J Near Infrared Spectrosc. 20(1Spec.), 223–235 (2012). [CrossRef]
5. D. Contini, L. Zucchelli, L. Spinelli, M. Caffini, R. Re, A. Pifferi, R. Cubeddu, and A. Torricelli, “Brain and muscle near infrared spectroscopy/imaging techniques,” J Near Infrared Spectrosc. 20(1Spec), 15–27 (2012). [CrossRef]
6. Q. Luo, B. Li, Z. Qiu, Z. Huang, Y. Gu, and X. D. Li, “Advanced optical techniques for monitoring dosimetric parameters in photodynamic therapy,” Proc. SPIE 8553, 85530F (2012).
7. D. W. Sun, Infrared Spectroscopy for Food Quality Analysis and Control (Elsevier 2009).
8. L. Čurda and O. Kukačková, “NIR spectroscopy: a useful tool for rapid monitoring of processed cheeses manufacture,” J. Food Eng. 61(4), 557–560 (2004). [CrossRef]
9. H. Huang, H. Yu, H. Xu, and Y. Ying, “Near infrared spectroscopy for on/in-line monitoring of quality in foods and beverages: A review,” J. Food Eng. 87(3), 303–313 (2008). [CrossRef]
10. B. M. Nicolaï, K. Beullens, E. Bobelyn, A. Peirs, W. Saeys, K. I. Theron, and J. Lammertyn, “Nondestructive measurement of fruit and vegetable quality by means of NIR spectroscopy: A review,” Postharvest Biol. Technol. 46(2), 99–118 (2007). [CrossRef]
14. Y. Roggo, P. Chalus, L. Maurer, C. Lema-Martinez, A. Edmond, and N. Jent, “A review of near infrared spectroscopy and chemometrics in pharmaceutical technologies,” J. Pharm. Biomed. Anal. 44(3), 683–700 (2007). [CrossRef] [PubMed]
15. S. Tsuchikawa, “A Review of Recent Near Infrared Research for Wood and Paper,” Appl. Spectrosc. Rev. 42(1), 43–71 (2007). [CrossRef]
16. S. S. Tsuchikawa and M. Schwanninger, “A review of recent near infrared research for wood and paper (Part 2),” Appl. Spectrosc. Rev. 48(7), 560–587 (2013). [CrossRef]
18. R. M. P. Doornbos, R. Lang, M. C. Aalders, F. W. Cross, and H. J. C. M. Sterenborg, “The determination of in vivo human tissue optical properties and absolute chromophore concentrations using spatially resolved steady-state diffuse reflectance spectroscopy,” Phys. Med. Biol. 44(4), 967–981 (1999). [CrossRef] [PubMed]
19. R. Nachabé, B. H. Hendriks, A. E. Desjardins, M. van der Voort, M. B. van der Mark, and H. J. Sterenborg, “Estimation of lipid and water concentrations in scattering media with diffuse optical spectroscopy from 900 to 1,600 nm,” J. Biomed. Opt. 15(3), 037015 (2010). [CrossRef] [PubMed]
20. T. H. Pham, F. Bevilacqua, T. Spott, J. S. Dam, B. J. Tromberg, and S. Andersson-Engels, “Quantifying the absorption and reduced scattering coefficients of tissuelike turbid media over a broad spectral range with noncontact Fourier-transform hyperspectral imaging,” Appl. Opt. 39(34), 6487–6497 (2000). [CrossRef] [PubMed]
21. F. Foschum, M. Jäger, and A. Kienle, “Fully automated spatially resolved reflectance spectrometer for the determination of the absorption and scattering in turbid media,” Rev. Sci. Instrum. 82(10), 103104 (2011). [CrossRef] [PubMed]
22. T. H. Pham, O. Coquoz, J. B. Fishkin, E. Anderson, and B. J. Tromberg, “Broad bandwidth frequency domain instrument for quantitative tissue optical spectroscopy,” Rev. Sci. Instrum. 71(6), 2500–2513 (2000). [CrossRef]
23. L. Spinelli, F. Martelli, A. Farina, A. Pifferi, A. Torricelli, R. Cubeddu, and G. Zaccanti, “Calibration of scattering and absorption properties of a liquid diffusive medium at NIR wavelengths. Time-resolved method,” Opt. Express 15(11), 6589–6604 (2007). [CrossRef] [PubMed]
24. F. Bevilacqua, A. J. Berger, A. E. Cerussi, D. Jakubowski, and B. J. Tromberg, “Broadband absorption spectroscopy in turbid media by combined frequency-domain and steady-state methods,” Appl. Opt. 39(34), 6498–6507 (2000). [CrossRef] [PubMed]
25. C. Abrahamsson, T. Svensson, S. Svanberg, S. Andersson-Engels, J. Johansson, and S. Folestad, “Time and wavelength resolved spectroscopy of turbid media using light continuum generated in a crystal fiber,” Opt. Express 12(17), 4103–4112 (2004). [CrossRef] [PubMed]
26. A. Bassi, J. Swartling, C. D’Andrea, A. Pifferi, A. Torricelli, and R. Cubeddu, “Time-resolved spectrophotometer for turbid media based on supercontinuum generation in a photonic crystal fiber,” Opt. Lett. 29(20), 2405–2407 (2004). [CrossRef] [PubMed]
27. A. Pifferi, A. Torricelli, P. Taroni, D. Comelli, A. Bassi, and R. Cubeddu, “Fully automated time domain spectrometer for the absorption and scattering characterization of diffusive media,” Rev. Sci. Instrum. 78(5), 053103 (2007). [CrossRef] [PubMed]
28. T. Svensson, E. Alerstam, D. Khoptyar, J. Johansson, S. Folestad, and S. Andersson-Engels, “Near-infrared photon time-of-flight spectroscopy of turbid materials up to 1400 nm,” Rev. Sci. Instrum. 80(6), 063105 (2009). [CrossRef] [PubMed]
29. S. Tsuchikawa and S. Tsutsumi, “Application of time-of-flight near-infrared Spectroscopy to wood with anisotropic cellular structure,” Appl. Spectrosc. 56(7), 869–876 (2002). [CrossRef]
30. I. Bargigia, A. Tosi, A. B. Shehata, A. D. Frera, A. Farina, A. Bassi, P. Taroni, A. D. Mora, F. Zappa, R. Cubeddu, and A. Pifferi, “Time-resolved diffuse optical spectroscopy up to 1700 nm by means of a time-gated InGaAs/InP single-photon avalanche diode,” Appl. Spectrosc. 66(8), 944–950 (2012). [CrossRef] [PubMed]
31. J. Johansson, S. Folestad, M. Josefson, A. Sparen, C. Abrahamsson, S. Andersson-Engels, and S. Svanberg, “Time-resolved NIR/Vis spectroscopy for analysis of solids: Pharmaceutical tablets,” Appl. Spectrosc. 56(6), 725–731 (2002). [CrossRef]
32. C. Abrahamsson, J. Johansson, S. Andersson-Engels, S. Svanberg, and S. Folestad, “Time-resolved NIR spectroscopy for quantitative analysis of intact pharmaceutical tablets,” Anal. Chem. 77(4), 1055–1059 (2005). [CrossRef] [PubMed]
34. J. P. Bouchard, I. Veilleux, R. Jedidi, I. Noiseux, M. Fortin, and O. Mermut, “Reference optical phantoms for diffuse optical spectroscopy. Part 1--Error analysis of a time resolved transmittance characterization method,” Opt. Express 18(11), 11495–11507 (2010). [CrossRef] [PubMed]
35. M. S. Patterson, B. Chance, and B. C. Wilson, “Time Resolved Reflectance and Transmittance for the Non-invasive Measurement of Tissue Optical Properties,” Appl. Opt. 28(12), 2331–2336 (1989). [CrossRef] [PubMed]
36. D. Contini, F. Martelli, and G. Zaccanti, “Photon migration through a turbid slab described by a model based on diffusion approximation. I. Theory,” Appl. Opt. 36(19), 4587–4599 (1997). [CrossRef] [PubMed]
39. A. Pifferi, A. Torricelli, A. Bassi, P. Taroni, R. Cubeddu, H. Wabnitz, D. Grosenick, M. Möller, R. Macdonald, J. Swartling, T. Svensson, S. Andersson-Engels, R. L. P. van Veen, H. J. C. M. Sterenborg, J. M. Tualle, H. L. Nghiem, S. Avrillier, M. Whelan, and H. Stamm, “Performance assessment of photon migration instruments: the MEDPHOT protocol,” Appl. Opt. 44(11), 2104–2114 (2005). [CrossRef] [PubMed]
40. A. Sparén, O. Svensson, M. Hartman, M. Fransson, and J. Johansson, “Matrix Effects in Quantitative Assessment of Pharmaceutical Tablets, Using Transmission Raman and NIR Spectroscopy,” Appl. Spectrosc., to be submited (2013).
41. LabSphere, Inc. “A Guide to Reflectance Coatings and Materials.”
42. A. Farina, A. Bassi, A. Pifferi, P. Taroni, D. Comelli, L. Spinelli, and R. Cubeddu, “Bandpass Effects in Time-Resolved Diffuse Spectroscopy,” Appl. Spectrosc. 63(1), 48–56 (2009). [CrossRef] [PubMed]
43. C. F. Bohren and D. R. Huffman, Absorption and Scattering of Light by Small Particles (Wiley, 1983), pp. xiv, 530 p.
44. H. C. d. Hulst, Light Scattering by Small Particles (Dover Publications, 1981), p. 470 p.
45. R. Graaff, J. G. Aarnoudse, J. R. Zijp, P. M. A. Sloot, F. F. M. de Mul, J. Greve, and M. H. Koelink, “Reduced Light-Scattering Properties for Mixtures of Spherical Particles: a Simple Approximation Derived from Mie Calculations,” Appl. Opt. 31(10), 1370–1376 (1992). [CrossRef] [PubMed]
46. LabSphere, Inc.”, retrieved http://www.labsphere.com/.
47. Avian Technologies, LLC”, retrieved http://www.aviantechnologies.com.
48. SphereOptics Gmbh”, retrieved http://www.sphereoptics.de/en/.