Light backscatter fiber optic sensor: A new tool for predicting the stability of pork emulsions containing antioxidative potato protein hydrolysate

    Light backscatter fiber optic sensor: A new tool for predicting the stability of pork emulsions containing antioxidative potato protein hydrolysate Gema Nieto, Youling L. Xiong, Fred Payne, Manuel Castillo PII: DOI: Reference:

S0309-1740(14)00468-9 doi: 10.1016/j.meatsci.2014.10.020 MESC 6580

To appear in:

Meat Science

Received date: Revised date: Accepted date:

7 January 2014 10 July 2014 20 October 2014

Please cite this article as: Nieto, G., Xiong, Y.L., Payne, F. & Castillo, M., Light backscatter fiber optic sensor: A new tool for predicting the stability of pork emulsions containing antioxidative potato protein hydrolysate, Meat Science (2014), doi: 10.1016/j.meatsci.2014.10.020

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT LIGHT BACKSCATTER FIBER OPTIC SENSOR: A NEW TOOL FOR PREDICTING THE STABILITY OF PORK EMULSIONS CONTAINING ANTIOXIDATIVE POTATO

PT

PROTEIN HYDROLYSATE

Department of Food Technology, Nutrition and Food Science, Faculty of Veterinary

NU

a

SC

RI

Gema Nieto* a,b,c, Youling L. Xiong c, Fred Payne b, Manuel Castillo b, d

Sciences, University of Murcia. Murcia, Spain

Department of Biosystems and Agricultural Engineering, University of Kentucky, 128

MA

b

C.E. Barnhart Building, Lexington, KY 40546-0276, USA c

Centre Especial de Recerca Planta de Tecnologia dels Aliments (CERPTA), Departament

AC CE P

d

TE

Building, KY 40546-0275, USA.

D

Department of Animal and Food Sciences, University of Kentucky, 206 W.P. Garrigus

de Ciència Animal i dels Aliments, Facultat de Veterinària, Universitat Autònoma de Barcelona, 08193-Bellaterra, Spain.

*

Corresponding author: Gema Nieto, Tel.: +34-868884798. Fax: +34-868884147

E-mail address: [email protected]

ACCEPTED MANUSCRIPT Abbreviation:

PT

L*, lightness; a*, redness; b*, yellowness; T, temperature; CL, cooking loss; CLnor,

RI

normalized cooking loss; RFL, fat/lean ratio; D, distance between optical fibers (D2 = 2 mm; D2.5 = 2.5 mm; D3 = 3 mm); I, normalized light backscatter intensity; I3D2, normalized light

SC

backscatter intensity peak 3 distance 2mm; I3D2.5, normalized light backscatter intensity

NU

peak 3 distance 2.5mm; I3D3, normalized light backscatter intensity peak 3 distance 3mm; λ3D2, wavelength at the maximum intensity for peak 3 distance 2mm; λ3D2.5, wavelength at

MA

the maximum intensity for peak 3 distance 2.5mm; λ3D3, wavelength at the maximum intensity for peak 3 distance 3mm; OD, optical density; OD2-2.5 optical density between

TE AC CE P

Hue.

D

distances 2 and 2.5; OD2-3, optical density between distances 2 and 3. C*ab, Chroma; H0ab,

ACCEPTED MANUSCRIPT ABSTRACT The objective of this study was to determine whether light backscatter response from

PT

fresh pork meat emulsions is correlated to final product stability indices. A specially

RI

designed fiber optic measurement system was used in combination with a miniature fiber optic spectrometer to determine the intensity of light backscatter within the wavelength

SC

range 300-1100 nm (UV/VIS/NIR) at different radial distances (2, 2.5 and 3 mm) with

NU

respect to the light source in pork meat emulsions with two fat levels (15, 30%) and two levels (0, 2.5%) of the natural antioxidant hydrolyzed potato protein (HPP). Textural

MA

parameters (hardness, deformability, cohesiveness and breaking force), cooking loss, TBARS (1, 2, 3, and 7 days of storage at 4ºC) and CIELAB color coordinates of cooked

D

emulsions were measured. The light backscatter was directly correlated with cooking

TE

losses, color, breaking force and TBARS. The optical configuration proposed would

AC CE P

compensate for the emulsion heterogeneity, maximizing the existing correlation between the optical signal and the emulsion quality metrics. KEYWORDS: Fiber optic, light backscatter, sensor, frankfurters, emulsion stability.

ACCEPTED MANUSCRIPT 1. Introduction Meat emulsions such as frankfurters and bolognas are finely comminuted and cooked

PT

products composed of water, muscle proteins, fat particles, salt and small amounts of non-

RI

meat ingredients, where meat proteins serve as natural emulsifier. According to Barbut (1998), fat stabilization during chopping is due to the formation of a proteic film around the

SC

fat particles that allows retaining fat inside the protein matrix. During chopping, certain

NU

attractive forces contribute to hold the raw materials together and create a homogeneous matrix structure (Allais, Christophe, Pierre, & Dufour, 2004). Excessive reduction of fat

MA

particles size and inadequate soluble protein extraction or fat to protein ratio could lead to reduced emulsification ability and increased fat oxidation tendency.

D

Therefore, the development of novel techniques for the meat emulsion quality

TE

analysis is required by the food industry in order to complement the traditional analysis

AC CE P

methods (Rodriguez-Otero et al., 1997). The two most prominent measurement configurations used for optical sensor development in food applications are transmission and light backscatter. As a result of the optical properties of meat emulsions particles, they tend to scatter the most light backward, and subsequently, light backscatter was chosen for the optical measurement configuration in this study. During the last two decades, a variety of optical sensor technologies has been developed for food process monitoring and control applications. Several technologies based on infrared spectrophotometry and using optical fibers to measure backscatter have been widely used for the analysis of food systems to evaluate different sources of paleness in meat (Swatland, 1982, 1983) to monitor and control the functional properties of comminuted meat batters used in meat processing (Swatland & Barbut, 1990). Specifically, light-scatter measurements have been applied for measuring the degree of meat emulsion

ACCEPTED MANUSCRIPT stability during hot dog manufacturing and monitoring milk coagulation, predicting cutting time and determining thermal denaturation of β-lactoglobulin and other whey proteins in

PT

milk (Fagan et al., 2007a, b; Álvarez et al., 2009; Lamb et al., 2013).

RI

The use of a real-time meat sensor technology having the ability to determine the optimum fat emulsification degree and emulsion stability (i.e., obtaining meat emulsion

SC

with the maximum stability and minimum cooking losses) in meat could be a good strategy

NU

to improve the current control over the emulsification process. The aim of this study was the study the optical properties (UV/Visible/NIR range:

MA

300 - 1100 nm) of pork emulsions having different fat/lean ratio, and presence or not of hydrolyzed potato protein percentages (natural ingredient with antioxidant properties) at

D

several distances between the emitting and detecting optical fibers, in order to detect

TE

changes in comminuted meats that may be correlated with those technological parameters

and color).

AC CE P

associated with emulsion stability (e.g., textural parameters, cooking losses, lipid oxidation

2. Material and Methods

Data analyzed in this study correspond to the data set presented previously (Nieto et al., 2009), where details of the material and methods, including the procedure for manufacture of pork meat emulsions, were explained. Hence, only a brief description of the main aspects of special relevance is provided here. For further details, the reader should refer to the previous publication. 2.1. Experimental design A completely randomized factorial design with two factors and three replications was used. Two different HPP (hydrolysed potato protein) levels (0 –control – and 2.5% w/w) were tested within a range of emulsion breakdown and lipid oxidation tendencies that were

ACCEPTED MANUSCRIPT induced by using two different levels of fat (15 and 30% w/w; i.e., fat/lean ratios, RFL, of 0.18 and 0.43, respectively). A total of 12 tests (N = nab = 3·2·2) were conducted with this

PT

design. A variety of final product quality indices (technological dependent variables) were

RI

determined to establish the degree of lipid oxidation (thiobarbituric acid-reactive substances or TBARS) and emulsion stability (cooking losses, hardness, cohesiveness, deformability

SC

and breaking force) of the cooked meat emulsions. Reflection photometry parameters

2.2. Meat samples and HPP preparation

NU

(CIELAB color coordinates) were also collected from fresh emulsions.

MA

The pre-weighed amounts of lean and fat, salt, crushed ice, water, and HPP solution were mixed according to the target treatment formulation. Treatment formulations for the

D

frankfurters were adapted from Feng, Xiong, and Mikelr (2003) and are detailed in Nieto et

TE

al. (2009). Potato protein concentrate was suspended in aqueous solution to obtain a 15%

AC CE P

w/v protein concentration, and the pH was adjusted to 8.0 with 1 N NaOH. Protein was hydrolyzed with alcalase for 1 h at 50ºC, using an enzyme/substrate ratio of 0.01. For further details, refer to Wang and Xiong (2005). Raw material mixtures were chopped using a 9 kg capacity bowl chopper (CM-14, Mainca, St. Louis, MO, USA). Knife and bowl speeds of 3000 and 10 rpm, respectively, were used. The raw emulsion was immediately split into two homogeneous aliquots. The first aliquot was used to measure light backscatter and CIELAB coordinates (L*, a*, b*) of the raw emulsions. The second aliquot was stuffed into 27 mm diameter frankfurter, weighed and cooked for 90 min in an Alkar smokehouse (450 U, Alkar- RapidPak Inc., Lodi, WI, USA) to an internal temperature of 71ºC (60 min). After cooking, the frankfurters were immediately cooled with cold water for 2 min, packed in polystyrene trays B5-37 (Aerpack), overwrapped with oxygen-permeable polyvinyl chloride (PVC) film (650 cm3 m-2 h-1 at 23ºC; Resinite

ACCEPTED MANUSCRIPT Packaging Films, Borden, Inc., North Andover, MA), and stored at 4ºC for 7 days to evaluate the inhibitory effect of HPP on lipid oxidation.

PT

2.3. Color CIELAB color coordinates, L*, a*, and b*, were measured 1 h after the emulsion

RI

was prepared using a hand held tristimulus Chroma Meter (CR-310 Minolta Camera Co.,

SC

Ltd., Osaka, Japan). Coordinates a* and b* were used, according to Hunt (1977), to

NU

calculate both Chroma, C*ab, and Hue, H0ab, values as follows: 2

C*ab  a*  b*

2

MA

H 0 ab  arctan(b* / a* ) 2.4. Cooking losses

Eqn. 1

Eqn. 2

TE

D

Once the chopping process was completed, cooking losses (CL) of each emulsion sample was measured in triplicate. CL was calculated from the weight of the final cooked

AC CE P

emulsion (WF) and the initial weight (W0) of the sample before cooking as follows:

CL  100(1  WF Wo )

Eqn.3

In addition, normalized cooking losses (CLnor) were calculated, taking into account the initial moisture content of the frankfurters. 2.5. Texture profile analysis (TPA) The influence of HPP and fat concentration on textural properties of frankfurters was investigated by uniaxial compression tests using an Instron Universal Testing Machine (Model 4301; Instron Corp., Canton, MA, USA) as described by Xiong et al. (1999). Cylindrical samples of 1.5 cm length were cut and compressed, using a 100 N load cell at a crosshead speed of 50 mm min-1, to 80% of its original height (strain, ε = ΔL/L0 = 0.2, were L0 is the initial length of the cylinder) in a two cycle compression with 15 s delay between

ACCEPTED MANUSCRIPT cycles. Hardness (H) of the sample was measured as the force (N) of the first compression peak (F1). The force of the second compression peak was designated as F2. The percent

PT

reduction in the compression force between the first and second compression peaks was defined as structure ‘‘Deformability” (D) and was calculated as D (%) = 100(F1-F2)/F1.

RI

Cohesiveness (C), as defined by Bourne (1978), was estimated as (F2/F1)2 (dimensionless),

SC

assuming that peaks of the first and second compression form similar triangles with the

NU

baseline. Another set of nine samples was compressed to 20% of its original height (ε = 0.8) to determine the breaking force (FB) (N).

MA

2.6. Thiobarbituric Acid-Reactive Substances (TBARS) TBARS were measured on days 0, 1, 3 and 7 of storage at 4ºC, according to the

AC CE P

TE

kg-1), was calculated as follows:

D

method described by Wang and Xiong (2005). The TBARS value (mg malonaldehyde

TBARS  9.48( A532 / Ws )

Eqn. 4

where A532 was absorbance at 532 nm, Ws was the sample weight (g), and 9.48 was a constant derived from the dilution factor and the molar extinction coefficient of the red TBA reaction product.

2.7. Light backscatter measurement of raw emulsions Light backscatter profiles were obtained from the fresh pork emulsions using a dedicated laboratory-scale optical sensor prototype (Figure 1), which was designed to set the radial distance between the optical emitting and detecting fibers by means of a micrometer as proposed by Álvarez et al. (2009). A plastic fiber holder was machined to hold the optical fibers in a fixed vertical position and attach to the micrometer for radial position adjustment as shown in Figure 1. One optical fiber was connected to the detector, a

ACCEPTED MANUSCRIPT High-Resolution Fiber Optic Spectrometer (HR4000, Ocean Optics, Inc., Dunedin, FL, USA). The other optical fiber was connected to the light source, a tungsten halogen

PT

(UV/VIS/NIR range: 300-1100 nm) bulb (LS-1, Ocean Optics, Inc.). Before measurement,

RI

the terminating ends of the fibers were aligned vertically to the same elevation and adjusted horizontally to minimal separation between fibers (2 mm centerline distance). The fiber tips

SC

were immersed into the emulsion samples (~25 mm of depth) up to a final depth of ~12.7

NU

mm from the surface. Fiber optic cables were manufactured using 600 µm diameter fibers (Spectran Specialty Optics, Avon, CN, USA). Scans were taken beginning with the fibers

MA

touching; a radial distance of 2 mm. This procedure was repeated in sequences of 0.5 mm radial increments until collect two additional scans at 2.5 and 3 mm of radial distance. As it

D

can be observed, light from a tungsten halogen bulb was transmitted to the probe tip using

TE

a fiber optic cable of 0.6 mm of diameter. Light reflected from the meat emulsion matrix

AC CE P

particles was transmitted through the receiving fiber to a high-resolution fiber optic spectrometer, with a bandwith of 300–1100 nm, an optical resolution of 2.5 to 2.7 nm and the grating is 300 line composite blaze. The temperature of the sample was controlled by means of connecting the sample holder to a water bath (Lauda Ecoline RE220. Brinkman Instruments Inc. NY. USA; ± 0.01ºC of accuracy). A series of measurements at three radial distances between optical fibers were performed for each experimental design treatment. Three distances between fibers, respectively 2.0, 2.5 and 3 mm, were considered in the study. Light scatter intensity at the target radial distances was measured at an integration time (IT) ranging from 19 to 60 s, where IT was the detector light exposure time. The light scattering spectral scans, I(λ), were automatically processed by subtracting the respective background signal and dividing by the IT to give the light scattering normalized spectral scans, I(λ) (bits s-1). Light scattered by the meat emulsion matrix particles was transmitted

ACCEPTED MANUSCRIPT through the receiving fiber to a high-resolution fiber optic spectrometer. The data acquisition system consisted of a PC connected by a USB port to the spectrometer and

PT

programmed for data acquisition with SprectraSuit Spectroscopy platform software (Ocean

RI

Optics, Inc.).

Figure 2 shows the typical normalized spectral scan obtained from pork meat

SC

emulsions with 0% and 2.5% HPP (media for the three distances and two percentages of

NU

fat). As it can be observed, three peaks were consistently observed in all the spectral scans: peak number 1 corresponded to the blue-green region of the spectrum (P1: 493-533 nm),

MA

while peaks number 2 and 3 corresponded to the green-yellow (P2: 560-584 nm) and red (P3: 636-720 nm) regions, respectively. A number of optical parameters were obtained from

D

the normalized spectral scans based on the observed peaks. A representative response, Ii,

TE

was calculated for each peak by numerically integrating the response over a selected

AC CE P

waveband about each peak. Each spectral scan was reduced to three intensity averages by dividing them into three wavebands (one waveband per peak), averaging the normalized intensity for the wavelengths constituting each waveband to yield the average waveband normalized intensity (Ii, i = peak number), and assigning the average values to the corresponding waveband at the maximum intensity for Peak i (λi, i = peak number). For each peak (i.e., waveband) and for each given Fat and HPP treatment combination, the calculated normalized intensity averages corresponding to different distances were used to calculate the optical density (OD; log Ix/ Iy, where x and y were two different fiber distances from the light source and x < y) according to Álvarez et al. (2010). Note that OD is not used here in place of absorbance as both light absorbance and scatter contribute to the extinction of light inside the emulsion samples. 2.8. Statistical Analysis

ACCEPTED MANUSCRIPT For the analysis of data SAS software (SAS, version 9.1, 2002-2003, SAS Institute, Cary, NC, USA) was used. Pearson correlation coefficients, r, were determined by the

PT

correlation (CORR) procedure of SAS. The analysis of variance (ANOVA) was performed

RI

using the general linear model (GLM) procedure of SAS. The least squares means (LSM) and significance of treatments were calculated using type IV sum of squares. LSM were

SC

considered to be statistically different when P < 0.05.

NU

3. Result and discussion. 3.1. Analysis of variance of dependent variables

MA

Dependent variables tested for monitoring the degree of meat emulsification in fresh ground pork were classified as optical (I, OD), color (L*, a*, b*, Hºab and C*ab), and meat

D

emulsion quality parameters (cooking loss, hardness, cohesiveness, ¨deformability¨,

TE

breaking force and TBARS). An ANOVA was independently conducted for each one of the

AC CE P

spectral scan regions studied (blue-green, green-yellow and red regions) to determine the main sources of variation in the dependent variables. Replicate, Fat and HPP were selected as main effects in the preliminary ANOVA model. The main interaction ‘‘HPP×Fat” was also included. Replication effect was not significant and was removed from the model. In Table 1, only ANOVA and F statistics corresponding to the studied light backscatter intensity parameters for the red region were shown to avoid unnecessary redundancy, because the response of parameters was similar in the three spectral scan regions studied. The ANOVA model was highly significant for I3D2, I3D2.5, I3D3, λ3D2, and λ3D2.5 but not significant for λ3D3. HPP was found to have a statistically significant (P <0.001) effect on all optical dependent variables, except for λ3D3 (significant at P < 0.05). On the contrary, Fat and the interaction HPPxFat were not statistically significant.

ACCEPTED MANUSCRIPT In Table 2, only ANOVA and F statistics corresponding to the studied optical density parameters for the red region were shown to avoid unnecessary redundancy, because the

PT

response of parameters was similar in all spectral scans studied. The ANOVA model was

RI

significant (P <0.05) for ODD2-2.5 but not significant for ODD2-3. HPP was found to have a statistically significant effect on both ODD2-2.5 and ODD2-3. Fat and the interaction HPPxFat

SC

were not found to have a statistically significant effect (P >0.05) on OD.

NU

3.2. Effect of distance between fibers on light backscatter spectral scans Figure 3 shows the effect of distance between the emitting and detecting optical fibers

MA

on the spectral scan. As it can be observed in the figure, the normalized light intensity decreased logarithmically with increasing distance between fibers.

D

The higher intensity detected at 2 mm, and the larger differences between treatments

TE

observed at this distance suggests the use of short distances between the detecting and

AC CE P

emitting optical fibers to design an optical sensor technology able to monitor the meat emulsion process. In this same way, Álvarez et al. (2009), determined the importance of using an appropriate spacing between fibers for the design of a meat emulsion stability inline sensor. These authors suggested that a light backscatter signal obtained using multiple pairs of emitting and detecting fibers spaced to collect an optical signal over a large sample area would compensate for the large heterogeneity of the matrix during chopping. According to our data, in this proposed design the distance between the emitting and detecting fibers within each detection unit would need to be smaller than 2 mm, while the distance separating different detection units would need to be four to five times larger than that.

ACCEPTED MANUSCRIPT 3.3. The effect of HPP on optical parameters The effect of HPP on normalized intensity of fresh meat emulsions at the maximum

PT

intensity for red region (P3: 636-720 nm) is shown in Table 3 and Figure 2. A significant (P

RI

<0.05) decrease (~ 55%) of intensity was observed with the presence of HPP. The Figure 2 shows clearly the mentioned tendency, not only for peak 3 but for the whole spectrum

SC

including peaks 1 and 2. In addition, the wavelength at which the maximum intensity value

NU

was observed within peak 3 region shifted significantly (P < 0.05), increasing to a larger value with the presence of HPP.

MA

Note that OD is not used here in place of absorbance as both light absorbance and scatter contribute to the extinction of light inside the emulsion samples.

D

The light that is “traveling” through the sample extinguished by absorption of soluble

TE

components, and as a result of light scatter, and components that mostly scatter the light are

AC CE P

fat globules and the myofibrillar and connective tissue, which are not completely dissolved after the chopping.

On the other hand, light scattering intensity is not only proportional to light scattering particle concentration but also depends, among other properties, on the nature of the particle (i.e., the complex index of refraction) and its size parameter (πd/λ, where d is the diameter of the particle) (Modest, 2003). The effect of decrease of light scatter (higher at distance 2 than 2.5 or 3) that is observed to add HPP is attributed to the strong absorption of visible light as consequence of the presence of potato protein (noted that during the preparation of the hydrolyzed potato protein was dark and it is related with the color noted in the visible region). And by that in a particulate media, light which is not absorbed by the media is either scattered or transmitted.

ACCEPTED MANUSCRIPT Although the optical properties of individual muscle fibers are well known (Martin Jones et al., 1991), the optical properties of meat are difficult to predict. On-line sensors are

PT

needed for prediction of the optical properties.

RI

A maximum (0% HPP) and a minimum (2.5% HPP) normalized intensity values were observed in the vicinity of 630 and 694 nm, respectively (Figure 5).

SC

Regarding the optical density (OD), the effect of HPP at the maximum intensity for

NU

red region (P3: 636-720 nm) is shown in Table 4. It was observed that optical density (OD2D2-2.5, OD2D2-3) increased significantly (P < 0.05) with HPP, which is likely originated

MA

by the intense absorbance caused by the presence of HPP (with the presence of HPP, lower normalized intensity (I), higher OD and higher light extinction, therefore less light is

D

detected by the sensor). This trend can be also observed in the Figure 5, where the LSM for

TE

OD increased by 12% with the presence of HPP.

AC CE P

In addition, deoxymyoglobin shows an absorption peak at 557 nm, and the oxymyoglobin at 582 nm (Tang et al., 2004). Thus, the shape of the OD profiles suggests that the peak observed in the optical density value of meat emulsions at wavelengths 582 nm (Figure 5) could result from oxymyoglobin absorbance. The presence of these peaks suggests the prediction on line of the percentage of metmyoglobin and oxymyoglobin during the shelf life of the meat. 3.4. The effect of fat concentration on optical parameters The effect of fat concentration on light backscatter parameters studied at the maximum intensity for red region (P3: 636-720 nm) (red region) is showed in Table 3 and Figure 4.

ACCEPTED MANUSCRIPT As it can be observed in Figure 4, light scatter through the sample increases with increasing fat levels. This is logical as light scattering intensity depends, among other

PT

properties, on particle concentration (Modest, 2003). Further increase of fat percentage

RI

produced an increase of light intensity. The observed pattern is considered to be related to the aqueous nature of meat fat dispersion in the emulsion matrix. Note that the primary

SC

components of meat emulsions associated with their light scattering properties are fat

NU

globules and broken microstructural constituents of lean tissue such as rests of myofibrillar proteins, sarcoplasmic proteins, connective tissue, etc., (Swatland, 2004). Thus, it is

MA

expectable that at low fat concentration, meat emulsion light backscatter intensity decreases. In absence of HPP (reduced light extinction as compared to samples with HPP –

D

note that HPP is absorbing light intensely–), increasing fat concentration induced, as

TE

expected, an increase in light backscatter normalized intensities at short distances from the

AC CE P

emitting source (2 and 2.5 mm) (Figure 4). Under conditions where light extinction is very intense (i.e., 3 mm distance from the emitting source and/or presence of HPP), this effect was less apparent. For instance, in samples with HPP and different fat percentages, as fat concentration increases, the intense light extinction observed decreased as a consequence of the HPP absorption, apparently made more difficult to observe the expected proportionality between concentration and intensity, which no longer holds. A very similar pattern was observed by Castillo et al. (2005). Working within a range of milk fat concentration in whey, these authors found that, once a certain concentration of fat was reached (0.4%), sidescatter intensity measured at 425 nm saturated and declined even though the concentration increased. Payne et al. (1997), working with milk having different fat levels, also observed a decrease in light penetration distance with increasing milk fat level.

ACCEPTED MANUSCRIPT Also it was observed that optical density (Table 4) increased not significantly as fat percentages increased. Although the effect was not significant, the observed trend agrees

PT

with the results obtained by Payne & Danao (2004) using a light extinction sensor in milk

RI

samples with fat values <50%. The authors observed that a ratio of light intensities, measured at two radial distances from the light source (I1/I2), was linearly and positively

SC

correlated with the milk fat content.

NU

In agreement with our results, Franke & Solberg (1971) conclude that fat concentration has a large effect on reflectance measurements at all visible wavelengths,

MA

especially when fat is comminuted together with lean muscle. The effect intensity on fat is shown in table 3: although the trends are decreasing the

D

intensity, however the decreased is not significant. As result of the increase of fat, the light

TE

intensity decrease, however are not significant due to that the light intensity is saturated as a

AC CE P

result of multiple scattering. As already mentioned, the main particles responsible for light scattering in meat emulsions are the fat globules. At low fat concentration (i.e., within the single scattering region), sidescatter intensity is linear with concentration (if particle concentration is low enough). However, as fat concentration increases, a point is reached when proportionality between concentration and intensity no longer hold (Castillo, 2005) 3.5. Relation between emulsion quality metrics and light backscatter According to Table 5, Pearson correlation coefficients between the emulsion quality metrics and optical measurements were strong in all the parameters studied. As expected, most of the parameters characterizing changes in texture, color and lipid oxidation were correlated with many of the optical measurements obtained in meat emulsion. The cooking losses (CL and CLnor) observed in this study were significantly and positively correlated with the following optical parameters derived from light backscatter measurements: I3D2-CL

ACCEPTED MANUSCRIPT (R= 0.696, P <0.05), I3D2.5-CL (R= 0.636, P <0.05) and I3D3-CL (R= 0.631, P <0.05). These correlations suggest that changes in CL, could be predicted by optical changes detected

PT

during the emulsification process, which would prevent a potential loss of water and fat

RI

during cooking. According to Hedrick et al. (1994) larger CL might be associated to lower meat emulsion quality.

SC

According to Barbut (1998) and Álvarez et al. (2007), fat and cooking losses during

lightness (L*) during the chopping process.

NU

the meat emulsion heat treatment can be predicted from the change on the emulsion

MA

The colour parameters were strongly correlated with the optical parameters: I3D2.5C*ab (R= -0.874, P <0.001), λ3D3-Hºab (R= - 0.518, P <0.001), I3D3-L* (R= 0.430, P <0.001),

D

I3D2.5-b* (R= 0.707, P <0.001), I3D2.5-a* (R= 0.813, P <0.001).

TE

Furthermore, considerable changes in light reflection (P < 0.05) were observed as a

AC CE P

result of the different fat levels in fresh frankfurters (Figures 4; arrow indicates the effect of increasing Fat concentration). Changes in L* with the fat content were as expected since the increase in the proportion of the whitish fat contributes to the increase in L*. Logically, the redness values, a*, were inversely proportional to fat content. Reduced protein content (i.e. increased fat to protein ratio) resulted in dilution of myoglobin and consequently a less intense red colour. According to Bañón, Díaz, Nieto, Castillo, and Álvarez (2008), high fat concentration emulsions (0.66 fat/lean ratio) show greater light reflection (larger L*), and less yellowness/redness than low fat concentration emulsions (0.25 fat/lean ratio). Allais et al. (2004) have also found that fat reduction results in darker and redder meat products. Therefore, the high correlations observed between optical and colour parameters, especially with L* values, suggest that light backscatter could have potential as indicator of emulsion stability during finely comminuted meat product manufacturing.

ACCEPTED MANUSCRIPT This finding shows the feasibility of an optical sensor technology to predict the optimum chopping end-point.

PT

Breaking force values (FB) were strongly (P <0.001) correlated with both normalized

RI

intensity (I3D3, R = 0.903, P <0.001) and optical density (OD2D2-2.5, r = -0.893, P <0.001).

SC

Similarly, a strong (P <0.001) correlation was observer between I3D3 and TBARS7 (r = 0.899). These findings suggest the feasibility of an optical sensor technology for on-line

NU

monitoring of physicochemical changes occurring during the meat emulsification process

MA

that would improve the meat emulsion stability. 4. Conclusions

D

In this study, light backscatter response from fresh pork meat emulsions with different

TE

distances between fibers, fat and hydrolyzed potato protein concentration and its correlation to final product stability indices have been evaluated. Light propagation was found to

AC CE P

decrease as distance, HPP percentage increased, and Fat percentage decreased. High correlation between light backscatter parameters generated during the comminuting process with color coordinates(C*ab - I3D2.5 : R= -0.874, P <0.001; L*- I3D3 : R= 0.430, P <0.001), breaking force ((FB- I3D3 : R= 0.903, P <0.001), TBARS (TBARS- I3D3 : R= 0.899, P <0.001) and cooking losses (I3D2-CL: R= 0.696, P <0.05) of the frankfurters were found. The results obtained suggest that the optimum sensor configuration for an optical meat emulsion stability sensor would use multiple measurements groups (i.e., detecting fibers surrounding an emitting fiber with a radial distance < 2.5 mm). It could be concluded that the use of a fiber optic light scatter sensor to monitor meat emulsification during the chopping process allows correlating optical parameters obtained from the fresh emulsion with cooking losses, textural parameters and lipid oxidation

ACCEPTED MANUSCRIPT tendency in the cooked product. The development of an inline light scatter sensor and the proposed optical measurement system will maximize the final cooked sausage quality.

PT

Acknowledgements

RI

The authors wish to thank the University of Murcia for the postdoctoral contract of Gema Nieto and Department of Biosystems and Agricultural Engineering (University of

SC

Kentucky) and the Departments of Animal and Food Sciences and Biosystems and

NU

Agricultural Engineering (University of Kentucky) for the economical support of this research.

MA

REFERENCES

Allais, I., Christophe, V., Pierre, A., & Dufour, E. (2004). A rapid method based on front-

D

face flurescence spectroscopy for the monitoring of the texture of meat emulsions and

TE

frankfurters. Meat Science, 67, 219-229.

AC CE P

Álvarez, D., Payne, F.A., Castillo, M., & Xiong, Y.L. (2007). Application of backscatter Light extintion tyo determine the stability of beef emulsions with different fat/lean ratios.

ASABE,

Annual

International

Meeting,

Minneapolis,

Minnesota,

USA.backscattering tracking of the transitional phase inversion of emulsions. Colloids and Surfaces A: Physicochemical and Engineering Aspects, 338, 148-154. Álvarez, D., Castillo, M., Payne, F.A., & Xiong Y.L. (2009). A novel fiber optic sensor to monitor beef emulsion stability using visible Light scattering. Meat Science, 81, 456466. Álvarez, D., Castillo, M., Payne, F.A., Cox, Y.L., & Xiong Y.L. (2010) Application of light extinction to determine stability of beef emulsions. Journal of Food Engineering, 96, 309-315.

ACCEPTED MANUSCRIPT Barbut, S. (1998). Use of a fiber optic probe to predict meat emulsion breakdown. Ital. Journal Food Science. 3, 253.

PT

Bañón, S., Díaz, P., Nieto, G., Castillo, M., & Álvarez, D. (2008). Modelling the yield and

ratio and starch. Meat Science, 80 (3), 649–655.

RI

texture of comminuted pork products using color and temperature. Effect of fat/lean

SC

Bourne, M. C. (1978). Texture profile analysis. Food Technology, 41(2), 163-178.

NU

Castillo, M., Payne, F.A., López, M.B., Ferrandini, E., & Laencina, J. (2005). Optical

Food Engineering, 71, 354-360.

MA

sensor technology for measuring whey fat concentration in cheese making. Journal of

Fagan, C. C., M. Castillo, C. P. O’Donnell, D. J. O’Callaghan, and F. A. Payne. (2007a).

TE

Science, 90, (Suppl. 1):145.

D

Optical measurement of curd shrinkage during cheese manufacturing. Journal of Dairy

AC CE P

Fagan, C. C., C. P. O’Donnell, D. J. O’Callaghan, G. Downey, E. M. Sheehan, C. M. Delahunty, C. Everard, T. P. Guinee, and V.Howard. (2007b). Application of midinfrared spectroscopy to the prediction of maturity and sensory texture attributes of cheddar cheese. Journal of Food Science, 72, 130-137. Feng, J., Xiong, Y. L., & Mikelr, W. B. (2003). Textural properties of pork frankfurters containing thermally/enzymatically modified soy protein. Journal of Food Science, 68, 1220-1224. Franke, W. C., & Solberg, M. (1971). Quantitative determination of metmyoglobin and total pigment in an intact meat sample using reflectance spectrophotometry. Journal of Food Science, 36, 515–519.

ACCEPTED MANUSCRIPT Hedrick, H. B., Aberle, E. D., Forrest, J. C., Judge, M. D., & Merkel, R. A. (1994). Principles of meat processing. In Principles of Meat Science. Dubuque, IO:

PT

Kendall/Hunt Pub. Co.

RI

Hunt, R. W. G. (1977). The specification of colour appearance: I. Concepts and terms. Color Research and Application, 2, 55–68.

SC

Lamb, P., Payne, F., Xiong, Y.L., Castillo, M. (2013). Optical backscatter method for

NU

determining thermal denaturation of β-lactoglobulin and other whey proteins in milk. Journal of Dairy Science, 96 (3), 1356-1365.

MA

Martin Jones H, Baskin R J, Yeh Y 1991 The molecular origin of birefringence in skeletal muscle. The contribution of myosin subfragment S-1. Biophys J 60, 1217- 1228.

D

Modest, M.F. (2003). Radiative he 434 at transfer. Massachusetts: Academic Press.

TE

Palombo, R., Van Roon, P. S., Prins, A., Koolmees, P. A., & Krol, B. (1994).

AC CE P

Changes in lightness of porcine lean meat batters during processing. Meat Science, 38, 453–476.

Nieto, G., Castillo, M., Xiong, Y.L., Álvarez, D., Payne, F., Garrido, M.D. (2009). Antioxidants and emulsifying properties of alcalase-hydrolyzed potato proteins in meat emulsions with different fat concentration. Meat Science, 83 (1), 24–30. Payne, F. A., Y. Zhou, R. C. Sullivan, and S. Nokes. (1997). Radial backscatter profiles in milk in the wavelength range of 400 to 1000 nm. ASAE Paper No. 976097. St. Joseph, Mich.: ASAE Payne, F.A., Danao, & M.G. (2004). Measuring Particulate Concentration with a Fiber Optic Light Extinction Sensor. 9th International Congress on Engineering and Food. Montpellier.

ACCEPTED MANUSCRIPT Rodriguez-Otero, J. L., M. Hermida., & J. Centeno. (1997). Analysis of dairy products by near-infrared spectroscopy: A review. Journal Agricultural Food Chemistry, 45,

PT

2815–2819.

RI

SAS (2002-2003). Statistical analysis system, version 9.1, 2002-2003. Cary, NC: SAS Institute Inc.

SC

Swatland, H.J. (1982). Fiber optic spectrophotometry and the wetness of meat. Journal of

NU

Food Science, 47, 1940–1942.

Swatland, H. J. (1983). Infrared fiber optic spectrophotometry of meat. Journal of Animal

MA

Science, 56, 1329–1333.

Swatland, H.J. (2004). Absorbance of light by mitochondria in pork muscle fibres differing

D

in myoglobin content. Meat Science, 67 (3). 371-375.

TE

Swatland, H.J., & Barbut, S. (1990). Fibre-optic spectrophotometry for predicting lipid

AC CE P

content, pH and processing loss of comminuted meat slurry. International Journal of Food Science and Technology, 25. 519-526. Tang, J., Faustman, C., & Hoagland, T.A. (2004). Krzywicki Revisited: Equations for Spectrophotometric Determination of Myoglobin Redox Forms in Aqueous Meat Extracts. Journal of Food Science, 69 (9), 717–720. Wang, L. L., & Xiong, Y. L. (2005). Inhibition of lipid oxidation in cooked beef patties by hydrolyzed potato protein is related to its reducing and radical scavenging ability. Journal of Agricultural and Food Chemistry, 53, 9186–9192. Xiong, Y. L., Noel, D. C., & Moody, W. G. (1999). Textural and sensory properties of lowfat beef sausages with added water and polysaccharides as affected by pH and salt. Journal of Food Science, 64, 550–554.

ACCEPTED MANUSCRIPT

Figure captions: Figure1. Schematic figure of the optic sensor used to measure light backscatter in meat

PT

emulsions at different radial distances from the light source (D2 = 2 mm, D2.5 = 2.5 mm and D3 = 3 mm), using a fiber optic of 0.6 mm diameter. The fiber tips were immersed 12.7 mm

RI

into the sample, which had a total depth of 25 mm.

Figure 2. Light backscatter intensity spectral scan from meat emulsions with 0% and 2.5%

SC

HPP (media for the three distances and two percentages of fat). I, normalized light backscatter intensity for peak 1, 2, 3.λ, wavelength at the maximum intensity for peaks 1, 2,

NU

3.

Figure 3. Light backscatter intensity spectral scans (IN) from meat emulsions with 30% fat

MA

(media for the two HPP percentages: 0% and 2.5%) at peak 3 (red region) measured at different radial distances (mm): D2 = 2 mm, D2.5 = 2.5 mm and D3 = 3 mm. Figure 4. Average values of normalized intensity (IN) at 678 nm (peak 3, red region)

TE

HPP (0%, 2.5%).

D

measured at different radial distances (mm) and as function of fat level (15%, 30%) and

AC CE P

Figure 5. Optical density spectral scans from meat emulsions with 0% and 2.5% HPP (media for the three distances and two percentages of fat).

PT

ACCEPTED MANUSCRIPT

RI

Figure1.

SC

LIGHT SOURCE

SPECTROPHOTOMETER

D1 D2

MA

NU

EMITING FIBER OPTIC

DETECTING FIBER OPTIC

Fat Fat

TE

D

D3

AC CE P

MEAT EMULSION PARTICLES

Fat Connective tissue

Fat

Muscle fiber

SCHEMATIC DIAGRAM MEAT EMULSION

ACCEPTED MANUSCRIPT

600

SC

O% HPP

NU

500 400

MA

2.5% HPP

300

P3

TE

D

200 100 0 -100

AC CE P

Normalized intensity I (bits s-1)

RI

PT

Figure 2.

0

200

P2 P1

400

600

Wavelenght (nm)

800

1000

1200

ACCEPTED MANUSCRIPT

SC

RI

PT

Figure 3.

600

NU

D2

MA

400

D 2.5

TE

D

300

200

100

0 0

AC CE P

Normalized Intensity (bit s-1)

500

200

400

D3

600

Wavelenght (nm)

800

1000

1200

ACCEPTED MANUSCRIPT Figure 4.

350

70-30

70-30 HPP

85-15

85-15 HPP

PT

300

RI

200

SC

150

NU

100 50

1.8

2.3

MA

0

2.8

TE

D

Radial distance (mm)

AC CE P

IN (bits s-1 )

250

3.3

ACCEPTED MANUSCRIPT Figure 5.

PT

0,7 0,6

RI

0,9 0,8

SC

2.5 % HPP

NU

0,5 0,4 0,3 0,2

0 % HPP

0,1 blue-green 0 550 570

MA

O.D.dimensionless

1

green-yellow

590

610

630

red

650

AC CE P

TE

D

Wavelenght (nm)

670

690

ACCEPTED MANUSCRIPT Table captions: Table 1. Analysis of variance and F statistic for normalized intensity and corresponding

PT

maximum wavelenght at the maximum intensity for peak 3 (red region).I Table 2. Analysis of variance and F statistic for optical density (OD) at corresponding

RI

maximum wavelength at the maximum intensity for peak 3 (red region).I

SC

Table 3. Influence of main effects, HPP and Fat concentrations, on normalized intensity of fresh

NU

meat emulsionsI at the maximum intensity for peak 3 (red region).I Table 4. Influence of main effects, HPP and Fat concentrations, on optical density (OD) of

MA

fresh frankfurtersI at the maximum intensity for peak 3 (red region). Table 5. Pearson’s correlation coefficients between dependent variables obtained by light backscatter at the maximum intensity for peak 3 (red region), color and meat emulsion

AC CE P

TE

D

quality metrics.I

ACCEPTED MANUSCRIPT Table 1. Variation Source

R2

F

F

I3D2

0.923

14.5**

62.5 ***

I3D2.5

0.950

22.8***

I3D3

0.963

31.7***

λ3D2

0.930

16.1 **

λ3D2.5

0.962

30.4 ***

λ3D3

0.647

2.20 ns

RI

F

HPPxFat(DF = 1) F

0.70 ns

103 ***

4.39 ns

0.15 ns

153***

0.62 ns

0.07 ns

72.8 ***

0.09 ns

5.34 ns

147 ***

0.11 ns

4.42 ns

7.96*

0.25 ns

1.75 ns

MA

NU

SC

2.03 ns

N = 12; DF, degrees of freedom; R2, determination coefficient; F, ANOVA F-statistic; *P<0.05, **P<0.01, ns

not significant. I3D2, normalized light backscatter intensity peak 3, distance 2mm; I3D2.5,

TE

***P<0.001,

D

I

Fat (DF = 1)

PT

HPP(DF = 1)

Model

normalized light backscatter intensity peak 3, distance 2.5 mm; I3D3, normalized light backscatter intensity

AC CE P

peak 3, distance 3mm; λ3D2, wavelength at the maximum intensity for peak 3, distance 2mm; λ3D2.5, wavelength at the maximum intensity for peak 3,distance 2.5 mm; λ3D3, wavelength at the maximum

intensity for peak 3, distance 3mm.

ACCEPTED MANUSCRIPT Table 2. Variation Source

F

ODD2-2.5

0.833

6.01*

21.2**

ODD2-3

0.591

1.74 ns

6.28*

F

0.50ns

2.10ns

0.71ns

0.69ns

N = 12; DF, degrees of freedom; R2, determination coefficient; F, ANOVA F-statistic; *P<0.05, **P<0.01, ***P<0.001, nsnot significant. ODD2–2.5, optical density between distances 2 and 2.5 mm; ODD2-3, optical density between distances 2 and 3 mm.

AC CE P

TE

D

MA

NU

I

HPPxFat(DF = 1)

F

RI

F

SC

R2

Fat (DF = 1)

PT

HPP(DF = 1)

Model

ACCEPTED MANUSCRIPT

Parameters II

PT

Table 3. Main effects

I3D2

787a

I3D2.5

525a

I3D3

371a

λ3D2

645a

15

30

472b

658

601

295b

434

387

180b

282

245

683b

664

663

651a

704b

677

678

662a

698b

677

683

MA

λ3D2.5

D

λ3D3 I

FatIV (%)

SC

2.5

NU

0

RI

HPP III (%)

AC CE P

TE

N = 12; number of observations; LSM with the same letters were not significantly different (P<0.05). II I3D2, normalized light backscatter intensity peak 3 distance 2 mm; I3D2.5, normalized light backscatter intensity peak 3 distance 2.5 mm; I3D3, normalized light backscatter intensity peak 3 distance 3mm; λ3D2, wavelength at the maximum intensity for peak 3 distance 2 mm; λ3D2.5, wavelength at the peak 3distance 2.5 mm; λ3D3, wavelength at the maximum intensity for 3 distance 3mm. III LSM value for each HPP concentration was based on average of 6 trials over a range of two fat concentration levels. IV LSM value for each Fat concentration was based on average of 6 trials over a range of two HPP concentration levels.

ACCEPTED MANUSCRIPT

Table 4. Parameters II

2.5 a

OD2D2-2.5

0.31

OD2D2-3

0.61a

I

0.35

b

SC

0

0.69b

FatIV (%)

15

30

0.33

0.34

0.63

0.66

RI

HPP III (%)

PT

Main effects

AC CE P

TE

D

MA

NU

N = 12; number of observations; LSM with the same letters were not significantly different (P<0.05). II OD, optical density; ODD2–2.5, optical density between distances 2 and 2.5 mm; ODD2–3, optical density between distances 2 and 3mm. III LSM value for each HPP concentration was based on average of 6 trials over a range of two fat concentration levels. IV LSM value for each Fat concentration was based on average of 6 trials over a range of two HPP concentration levels.

ACCEPTED MANUSCRIPT

CLnor

H

FB

TBARS7

L*

a*

b*

CR

CL

IP

T

Table 5.

Hºab

C*ab

0.696*

0.594*

-0.592*

0.817 **

0.792**

0.328***

0.750**

0.642*

0.722 **

-0.824***

I3D2.5

0.636*

0.527ns

-0.562ns

0.875***

0.822**

0.312***

0.813**

0.707**

0.787**

-0.874***

I3D3

0.671*

0.594*

-0.681*

0.903 ***

0.899***

0.430***

0.771**

0.656*

0.745**

-0.827***

λ3D2

-0.421ns

-0.389ns

0.543ns

-0.832*** -0.658*

-0.521*

-0.637*

-0.490ns

-0.598*

0.757**

λ3D2.5

-0.481ns

-0.424 ns

0.583*

-0.873*** -0.750**

-0.486*

-0.726**

-0.591*

-0.692*

0.825**

λ3D3

-0.381ns

-0.312ns

0.381ns

-0.545 ns

-0.283**

-0.545ns

-0.455ns

-0.518***

0.651*

OD2D2-2.5

-0.272ns

-0.219ns

0.404 ns

-0.893

-0.752**

-0.237**

-0.761**

-0.681*

-0.750 **

0.742**

OD2D2-3

-0.387ns

-0.811**

-0.111**

-0.730**

-0.686*

-0.727**

0.687*

MA N

TE D

0.384 ns

-0.474ns

-0.707 **

CE P

-0.243ns

N = 12. nsNot significant, *P<0.05, **P<0.01, ***P<0.001. L*, lightness; a*, redness; b*, yellowness; Hºab, hue angle; C*ab, chroma; CL, cooking losses; CLnor, normalized cooking losses; H, hardness; FB, breaking force; TBARS7, TBARS day 7. I3D2, normalized light backscatter intensity peak 3 distance 1; I3D2.5, normalized light backscatter intensity peak 3 distance 2.5; I3D3, normalized light backscatter intensity peak 3 distance 3; λ3D2, wavelength at the maximum intensity for peak 3 distance 2; λ3D2.5, wavelength at the maximum intensity for peak 3distance 2.5; λ3D3, wavelength at the maximum intensity for peak 3 distance 3. OD, optical density; OD2D2–2.5, optical density between distances 2 and 2.5; OD2D2–3, optical density between distances 2 and 3.

AC

I

***

US

I3D2

ACCEPTED MANUSCRIPT Highlights  We designed a fiber optic measurement system in combination with a miniature fiber

optic spectrometer. We examine whether light backscatter from meat emulsions is correlated to final stability

PT



SC

Light backscatter has potential as an early predictor of meat emulsion stability during

TE

D

MA

NU

manufacturing.

AC CE P



RI

indices

35