Measuring milk fat content by random laser emission

The luminescence spectra of milk containing rhodamine 6G are shown to exhibit typical signatures of random lasing when excited with 532 nm laser pulses. Experiments carried out on whole and skim forms of two commercial brands of UHT milk, with fat volume concentrations ranging from 0 to 4%, presented lasing threshold values dependent on the fat concentration, suggesting that a random laser technique can be developed to monitor such important parameter.

results were nearly the same, just those of one brand are presented. The composition of milk was determined with two techniques: the Gerger method 13 and dynamic light scattering (DLS) 14 . The results of the Gerger method showed that the presence of lipids (fat globules) in whole milk were 4.00 and 3.70 grams per each 100 g for each of the two brands. For skim milk, no fat globules were detected with this procedure in each brand. DLS results in whole milk revealed two gaussian size distributions of particles. The average diameter of particles corresponding to the higher intensity distribution (73%) was around 227 nm and for the lower intensity distribution (27%) was 937 nm with a standard deviation of 226 nm. The skim milk presented only one gaussian size distribution with particles of diameters around 220 nm with a standard deviation of 54 nm. These results confirm the presence of fat globules in the whole milk samples, as well its diameter (approximately 930 nm) and its percentage per 100 g (approximately 4%), which are in good agreement with typical features of milk reported previously 10 .
Rhodamine 6G (R6G) dye was purchased in analytical solution from Sigma-Aldrich and used with a concentration of 0.1 mM in all solutions. The spectroscopic measurements were performed in fused quartz cuvettes with 10 mm long optical path. The experimental setup is shown in Fig. 1.
The luminescence in our samples was produced with the second harmonic (532 nm) of a Q-switched Nd:YAG laser, delivering 8 ns pulses with a 10 Hz repetition rate. The pulse irradiance was controlled by passing the optical beam through a half-wavelength plate located between two polarizers. The optical beam waist inside the cuvette was estimated to be around 3 mm and the incidence angle was approximately 45 degrees. The light emitted was collected normally to the cuvette surface by a set of lenses coupled to an optical fiber connected to a CCD-compact spectrometer.

Results and Discussion
At low average powers, the emission spectra of R6G in water and milk are nearly identical, as shown in Fig. 2, with a broadband having a peak located at 564 nm at this concentration 15 . However, the bands in the two brands of whole milk containing R6G get narrower as the average pumping power increases. The luminescence was analyzed by deconvolution of the broadband with two Gaussians. The RL regime observed is associated to the contribution of  the monomer emitter 16 . For pumping powers above about 7 mW, the maxima of the milk spectra slightly shifts to longer wavelengths. Furthermore, besides the bandwidth decrease, an increase in the peak intensity is also present.
The results, shown in Fig. 3, indicate that the RL behavior starts first in whole milk, as expected, because of its larger number of scattering centers owing to the fat presence. This behavior is typical of random laser action and must depend on the number of scattering centers present in the emulsion.
In order to confirm this dependence of random laser properties by changing the scattering centers density, we diluted five different ratios of whole milk and skim milk. These dilutions allow us to have five distinct fat concentrations samples, with a constant concentration of 0.1 mM of R6G dye. The diluted ratios were selected to achieve 3.70, 2.78, 1.85, 0.93 and 0.00 grams of fat globules per 100 grams of milk. Because the content of proteins is nearly the same in both whole and skim milks (about 3% in volume), it contributes with the same amount of scattering in all samples. The results revealed that minor scattering center density results in different linewidth and peak intensity thresholds, as expected from the RL theory and previous studies 17 .
Hence, in order to develop a practical instrument to determine the fat content we analyzed the peak height of the band centered around 564 nm of those five samples, employing a power of 60 mW, shown in Fig. 4, which is approximately above the intensity threshold of all samples. These results demonstrate that the measurement of the RL emission peak intensity would provide a relatively cheap and practical approach for the development of a sensor to determine fat concentration in milk. Although we used here a spectrometer, a bandpass filter centered around 564 nm could alternatively be employed to determine how the RL emission is sensible to the milk fat concentration.
We observed RL behavior in UHT milk and since the results presented lasing threshold values dependent on the fat concentration, a technique to monitor such parameter could be developed. Further experiments still need to be done to understand how this method applies to non-homogenized natural milk.  Method Gerber method is based on milk emulsion breaking by the addition of sulfuric acid (Specific gravity 1.820-1.825) and isoamyl alcohol (Specific gravity 0.814-0.816). The sulfuric acid digests proteins and carbohydrates. The isoamyl alcohol generally prevents the charring of sugar.