Non-intrusive studies of gas contents and gas diffusion in hen eggs.

A detailed study of the condition of eggs was performed using tunable diode lasers to monitor free gas in hen eggs. We detected oxygen and water vapor signals from 13 unfertilized eggs and studied the growth of the egg air cell over a time period of 3 weeks. We also studied the gas exchange through the egg shell, which is of particular interest for fertilized eggs. Four fertilized and five unfertilized eggs were followed over 3 weeks, the hatching period for hen eggs, and significant variations were found both in time and for the two types of eggs. Our results indicate that the techniques could be developed for automatic control of egg freshness, as well as for monitoring the hatching progress of fertilized eggs.


Materials
Unfertilized eggs were bought from an Anhui province supplier. The eggs were sent by express delivery to our laboratory in Guangzhou directly after being laid, and we had an exact knowledge of the date. We stored the eggs at room temperature of 24°C and with 60 percent of relative humidity (RH). Figure 1 (a) illustrates the interior of a hen egg, and the detector position and the point of laser light entry are indicated. Figure 1(b) is a photograph of the arrangement. As illustrated in the figure the detector is placed at the egg shell above the air chamber with the light-injection fiber located at a fixed lateral position. Oxygen and water vapor absorption was detected in 13 eggs. Each egg was measured every second day for 10 minutes for oxygen and water vapor during a time period of 21 days. In addition, five unfertilized eggs and four fertilized eggs were selected for an experimental study of gas exchange to determine the diffusion time constant of the eggs. The fertilized eggs were bought from the same egg supplier as mentioned. The fertilized eggs were kept in a hatching machine, with the temperature regulated close to 37°C and with an RH value at 70 percent, which was enabled by adding 50 ml water to the machine every day. Each unfertilized egg was put in a plastic bag, which was flushed with nitrogen for 30 minutes before the measurements for gas diffusion. In contrast, each fertilized egg was taken out of the hatching machine and put in a bag containing 30% oxygen for about 30 minutes before performing the gas diffusion measurements. These steps could ensure that the egg interior reaches a steady-state gas situation before the bags were opened, and the transition back to ambient atmospheric condition was monitored. The detection geometry was the same as described above. Signals from each fertilized egg were detected for 30 minutes every second day during its incubation period, and likewise, the unfertilized eggs were subject to the same type of measurements during 3 weeks.

Experimental set-up
A schematic diagram of the experimental system used in the study is shown in Fig. 2. One distributed feedback (DFB) laser (#LD-0760-0040-DFB-2, TOPTICA) operating at 760 nm was used as the spectroscopic light source to monitor the oxygen absorption lines. The other DFB laser (#LD-0937-0100-DFB-2, TOPTICA) operating at 937 nm, was used for water vapor detection. Each laser was connected to a laser driver (LCD201C, Thorlabs) and a temperature controller (TED200C, Thorlabs). The lasers had nominal output powers of 40 mW and 100 mW, respectively. In our Labview-controlled program, we used two analogue waves to modulate the laser outputs. A 5 Hz ramp wave was used to ensure that the laser could scan through the gas absorption line. A sinusoidal wave with a frequency of f = 10295 Hz and 9015 Hz, respectively, for the two gases, was used for acquiring the harmonic signals in the lock-in detection process. Those signals were sent to modulate the DFB diode lasers via an analog output (AO) channel of a data acquisition (DAQ) card (PCI6120, National Instruments). The wavelength modulation depth is optimized for maximum sensitivity, since GASMAS signals are always weak [26,27]. The laser light was transmitted through a fiber with a core diameter of 600 μm to the egg shell surface. We used a photodiode (S3204-08, HAMAMATSU) with an area of 18x18 mm 2 to monitor the scattered light intensity. Thereafter, the signal was boosted with a low-noise current amplifier (DLPCA-200, FEMTO) and converted into a voltage signal, which was fed to an analog input (AI) of the data acquisition (DAQ) card with a sampling rate of 400 kHz and a buffer size of 80000. Digital lock-in detection was used for retrieving signals with optimum signal-to-noise ratio [26].

Measurements and results
Before the measurements, we ensured that the eggs chosen for the study had no obvious anomalies. This was performed by candling, a traditional method which employs transillumination of the eggs [28,29]. Thirteen unfertilized eggs were retained for the study of oxygen and water vapor signal development, and 9 eggs were used for gas exchange experiments. In order to make sure every measurement had the same geometry, we marked the position of light injection in each egg and the detector position employed in all measurement, as show in Fig. 3 (a). From our earlier study [25], we know that the signal is not very sensitive to the detailed positions. To make sure that the recorded signal solely comes from the egg we used some wax to cover the gap between the detector and the egg surface.
GASMAS is a variety of diode-laser-based gas absorption spectroscopy [27]. Absorption measurements are governed by the Beer-Lambert law relating the transmitted intensity I to the incident intensity I 0 according to where k is a p and l is the intensity norm We note, that known. How concentration GASMAS me the L eq value imprint as the is normally a temperature a oxygen in the path length th expressed by recalculated to

Measu
The oxygen a starting on da were kept at a for oxygen an and water rec is straight-for length in amb volume of a concentration corresponds to proportionality path length [2 malized GASM t for evaluatio wever, in a sc n and the mea easurements ar denotes the di e scattering me air with 21% o and relative hu e unfertilized e hrough the air y the empirica o other temper 3. (a). One of the on for laser light in detector was place high-quality liner signal from an ent air. The L eq i ase of the laser, an 15

Measu
In order to stu in an experim minutes. The oxygen signa exposed to th gradually incr exchange is s exponential fi

Measure
The gas excha started by firs exposed to a minutes. The like in the me was recorded development evaluated to 5 rmed this type ks. The develo l eggs. We obs the four first a ys) we obtain th 6 and not drastically for physiological reasons considering fertilized eggs, the recordings have rather low quality, and the development of the time constant over time could not be followed in a detailed way; no special tendency could be observed, though. Instead, we formed the average time constant as evaluated for the eggs, measured on 7 occasions. The average time constant for the fertilized eggs was thus found to be 6.0 (3.5) minutes, while the corresponding values for all the data shown in Fig. 6 was found to be 9.0 (2.3) minutes.

Discussion and outlook
We have demonstrated that GASMAS is a useful non-intrusive technique to study the contents of free gas in hen eggs. We could show that a steady size of the air chamber of unfertilized eggs with increasing storage time occurs by studying the increase of the oxygen as well as the water vapor signal. While the air cell development progresses smoothly as seen in the readily calibrated oxygen data, the situation is more complex for the water vapor data. However, the general trend is very similar for both gases, with an increase of the values over the study period by a factor 2.14 and 2.32 or 2.16, respectively, where the two water vapor values correspond to the two different types of calibrations, as discussed above. We note, that all three factors are basically the same, and reflect the continuously increasing size of the air cell. Data for oxygen display a particularly smooth age dependence, and would thus be the most precise age indicator.
Still, it should be acknowledged, that an age determination would work only for a specific type of eggs, for which the corresponding time-dependent L eq curve has been established. Different types of eggs most likely have different curves, while still showing steadily increasing values. Further, an improved signal-to-noise ratio, achievable, e.g. by using a higher laser power, would clearly increase the precision attainable.
Under the assumption of 100% RH in the presence of free liquid water, a deviation between the measured equivalent path values for the two gases might be explained by differences in scattering and absorption properties of the bulk medium, which contains the gas-filled pores or cavities. It has been shown in other contexts that such deviations caused by the use of somewhat different laser wavelengths are small when the measurements are made in a geometry of transmission type, and all detected photons are known to have passed the air containing volume. Deviations were, e.g., not observed, in a study of human sinuses [31].
Deviations could also be caused by so called water activity, modifying the vapor pressure calculated by the Arden-Buck relation [32][33][34], effectively reducing the RH value. Water activity means, that the vapor concentration above a liquid water surface is changed if the water contains dissolved salts. While observable in precision measurements [34], the effects have been shown to be small for most wet foods, such as meat or milk [32,33]. Water activity is an important aspect in food science, since the growth of bacteria requires a high RH value [32].
It should be noted, that L eq data for water vapor, evaluated assuming the presence of free water inside the egg, would have resulted in a RH of 100 percent. In Fig. 4 we show one data set evaluated under such assumption, and leading to data considerable below the oxygen ones. Since the calibration now relies on the measurement of the ambient air relative humidity for the recalculation to the case of RH = 100%, inaccuracies in the RH readings could result in a somewhat jagged curve, which is actually observed. This is in contrast to the situation for oxygen, where the curve is perfectly smooth corresponding to constant values of 21% oxygen.
In contrast, if we assume the egg white to be a homogenous substance without free water, the ambient air RH, propagating also into the air cell gas would be the situation, but now instead observed to lead to values above the ones for oxygen. Our conclusion is, that the water vapor conditions related to the egg interior is complex and possibly influenced by all the factors mentioned above.
We also note, that the scattering and absorption of the bulk medium (egg white and yolk) may be changing over the study period, in which case the GASMAS signal development could also partially reflect this aspects, apart from the increase of the air cell with time.
However, our measurements show, that for a particular type of eggs, the GASMAS technique can be used for quite good assessment of the age of the egg based on oxygen as well as water vapour monitoring, where a signal increase by a factor of about 2.2 is observed over a 3-week period.
As demonstrated, GASMAS provides a unique way to study gas diffusion through the egg shell. We repeatedly measured the time constant over a three-week period for unfertilized eggs and observed that it is decreasing substantially on a general time scale of about 10 minutes for the type of eggs studied. Of particular interest are of course the conditions for the fertilized egg during the hatching period, when the chicken embryo clearly is in demand for oxygen. Our study shows, that fertilized eggs have a faster gas diffusion than the unfertilized eggs (by about a factor of 1.5), which clearly helps the oxygenation process for the developing chicken inside the shell. Given the limited signal-to-noise ratio levels obtained in the assessment of gas exchange rates, this parameter is not a useful indicator for routine assessment of freshness for unfertilized eggs. Our study rather gives insight into the interesting time-dependent physiological changes in the permeability of the shell.
Our study indicates, that GASMAS might be developed into a valuable tool for the objective assessment of egg freshness and for studying gas contents and gas exchange during hatching. Thus, GASMAS might develop into a useful non-invasive tool in poultry science, expanding the applicability of biophotonics techniques.