S £19.: .. kflahnmrz; r. :1 “Exxwmmumflfirémrh infirm $5; a... a? Ta. .53.. .1 {it}: ‘ . 1.2:: Stu 3.... :1 $2. .5 £25.? 51.5.3 I! .124 1:: 5:23.} I. :V ||l||l|l|||llllllllllllllllIllllllllllllllllllllllllllll 3 1293 02079 920 This is to certify that the thesis entitled Yield Stress Of Cream Cheese presented by Spencer L. Breidinger has been accepted towards fulfillment of the requirements for M.S. degree in Food Science 3a Au 1.554%. Major professor Date 7/31/00 / / 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution LEBRARV Mieh'gan State University YIELD STRESS OF CREAM CHEESE By Spencer L. Breidinger A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Food Science and Human Nutrition 2000 ABSTRACT YIELD STRESS OF CREAM CHEESE By Spencer L. Breidinger Two rheological methods for measuring yield stress, the controlled stress and vane method, were used to evaluate the spreadability of cream cheeses with varying fat contents at refrigeration temperature (5°C) and room temperature (22°C). The vane method was found superior to the controlled stress method due to lower cost, ease of use, and higher quality data produced. While both methods were capable of quantifying the influence of temperature on spreadability (results at 5°C were 1.5 to 3 times higher than at 22°C), results from the controlled stress method were skewed by excessive sample handling prior to testing. Fat free samples were found to fracture at the yield stress, especially at 50C. Controlled stress data were influenced by fracturing more than vane method data because it occurred prior to achieving the yield stress. Fat free samples also showed less sensitivity to temperature than others. having similar results at both room and refrigeration temperatures. Apparent yield strain versus yield stress was plotted to produce “texture maps of spreadability”. Results show a trend in sample spreadability giving a “quality control window” which can be used in quality control or product development efforts. These texture maps are excellent tools for quickly determining acceptable levels of cream cheese spreadability. DEDICATION To my wonderful wife, Kathy. who made countless sacrifices so that I could succeed. iii ACKNOWLEDGEMENTS I cannot thank Dr. James Steffe enough for the support, advice, and guidance he has given me. If it were not for Dr. Steffe, this thesis would not have been possible. I would also like to thank Dr. Denise Smith and Dr. Zeynep Ustunol for serving on my committee. Special thanks to Mr. Richard Wolthuis for constructing the vanes used in this work. Table of Contents List of Tables List of Figures Nomenclature Introduction 1. Literature Review 2. 1.1. Sensory Spreadability and Instrumental Techniques 1.2. Cream Cheese 1.3. Yield Stress Measurement 1.3.1. Vane Method 1.3.2. Controlled Stress Method 1.4. Objectives Theoretical Development 2.1. Yield Stress Calculations 2.1.1. Controlled Stress Method 2.1.2. Vane Method 2.1.2.1. Single Point Method 2.1.2.2. Slope Method 2.2. Yield Strain Calculations 2.2.1. Controlled Stress Method 2.2.2. Vane Method . Materials and Methods 3.1. Equipment 3.1.1. Controlled Stress Method 3.1.2. Vane Method 3.2. Experimental Materials 3.3. Data Collection 3.3.1. Controlled Stress Method 3.3.2. Vane Method 3.4. Data Analysis 3.4.1. Controlled Stress Method 3.4.2. Vane Method Page vii 4. Results and Discussion 4.1. 4.2. 4.3. 4.4. 4.5. Vane Method Controlled Stress Method Comparison of Methods and Temperature Effects Yield Strain Fat Level and Yield Stress 5. Conclusions and Recommendations 5.1. 5.2. Appendix Summary and Conclusions Recommendations for Future Research References vi 36 36 52 65 70 79 81 81 82 84 95 Table 1.1. 3.1. 4.1. 4.2. 4.3. 4.4. 4.5 4.6. 4.7. A1 A2. A3. List of Tables Page Typical composition (in percent) of commercial unripened soft cheeses (Kosikowski and Mistry, 1997). 7 Percentage of total fat and carbohydrate of cream cheese samples. 31 Vane method (single point) yield stress results (Pa) at 5°C and 22°C. 37 Vane method (slope method) yield stress results (Pa) at 5°C and 22°C. 50 Yield stress (Pa) at 5°C and 22°C determined from the controlled stress method. 64 Yield stress (Pa) results at 5°C for the controlled stress method (CS), single point vane method (SP), and slope vane method (SL). 68 Yield stress (Pa) results at 22°C for the controlled stress method (CS), single point vane method (SP), and slope vane method (SL). 69 Controlled stress method yield strain results (dimensionless) at 5°C and 22°C. 71 Vane method yield strain results (radians) at 5°C and 22°C. 72 Controlled stress and single point vane method data for Kraft Philadelphia regular cream cheese. 85 Controlled stress and single point vane method data for Kraft Philadelphia Neufchatel cheese. 86 Controlled stress and single point vane method data for Kraft Philadelphia light cream cheese. 87 vii A.4. A5. A6. A7. A8. A9. A.10. Controlled stress and single point vane method data for Kraft Philadelphia fat free cream cheese. Controlled stress and single point vane method data for Kraft Philadelphia whipped cream cheese. Controlled stress and single point vane method data for Store Brand regular cream cheese. Controlled stress and single point vane method data for Store Brand Neufchatel cheese. Controlled stress and single point vane method data for Store Brand fat free cream cheese. Controlled stress and single point vane method data for Bruegger’s regular cream cheese. Controlled stress and single point vane method data for Bruegger’s light cream cheese. viii 88 89 90 91 92 93 94 Figure 1.1. 3.1. 3.2. 3.3. 4.1. 4.2. 4.3. 4.4. 4.5. 4.6. List of Figures Vessel and vane for yield stress determination. Cream cheese loaded into Haake RSlOO for controlled stress test (note digital meter indicating gap width of 2 mm). F our-bladed stainless steel blades (scale in cm) used in vane method testing. Example of vane dimensions (cm) for the medium-sized vane. Typical plot from a vane method test showing torque versus time for Kraft Philadelphia Regular cream cheese at 22°C (M0 denotes peak torque achieved). Typical plot from a vane method test showing torque versus time for Kraft Philadelphia Neufchatel cheese at 22°C (MO denotes peak torque achieved). Typical plot from a vane method test showing torque versus time for Kraft Philadelphia Light cream cheese at 22°C (M0 denotes peak torque achieved). Typical plot from a vane method test showing torque versus time for Kraft Philadelphia Fat Free cream cheese at 22°C (M0 denotes peak torque achieved). Typical plot from a vane method test showing torque versus time for Kraft Philadelphia Whipped cream cheese at 22°C (M0 denotes peak torque achieved). Typical plot from a vane method test showing torque versus time for Store Brand Regular cream cheese at 22°C (M0 denotes peak torque achieved). Page 12 25 27 28 38 39 4O 41 42 43 4.7. 4.8. 4.9. 4.10. 4.11. 4.12. 4.13. 4.14. 4.15. 4.16. 4.17. Typical plot from a vane method test showing torque versus time for Store Brand Neufchatel cheese at 22°C (Mo denotes peak torque achieved). Typical plot from a vane method test showing torque versus time for Store Brand F at Free cream cheese at 22°C (M() denotes peak torque achieved). Typical plot from a vane method test showing torque versus time for Bruegger’s Regular cream cheese at 22°C (M0 denotes peak torque achieved). Typical plot from a vane method test showing torque versus time for Bruegger’s Light cream cheese at 22°C (M0 denotes peak torque achieved). Example of slope vane method for Store Brand fat free cream cheese at 5 °C (a = 0.962). Typical plot from a controlled stress method test on Kraft Philadelphia Regular cream cheese at 22°C showing shear strain versus shear stress. Typical plot from a controlled stress method test on Kraft Philadelphia Neufchatel cheese at 22°C showing shear strain versus shear stress. Typical plot from a controlled stress method test on Kraft Philadelphia Light cream cheese at 22°C showing shear strain versus shear stress. Typical plot from a controlled stress method test on Kraft Philadelphia Fat Free cream cheese at 22°C showing shear strain versus shear stress. Typical plot from a controlled stress method test on Kraft Philadelphia Whipped cream cheese at 22°C showing shear strain versus shear stress. Typical plot from a controlled stress method test on Store Brand Regular cream cheese at 22°C showing shear strain versus shear stress. 44 45 46 47 51 53 54 55 56 57 58 4.18. 4.19. 4.20. 4.21. 4.22. 4.23. 4.24. 4.25. Typical plot from a controlled stress method test on Store Brand Neufchatel cheese at 22°C showing shear strain versus shear stress. Typical plot from a controlled stress method test on Store Brand Fat Free cream cheese at 22°C showing shear strain versus shear stress. Typical plot from a controlled stress method test on Bruegger’s Regular cream cheese at 22°C showing shear strain versus Shear stress. Typical plot from a controlled stress method test on Bruegger’s Light cream cheese at 22°C showing shear strain versus shear stress. Texture map for the controlled stress method showing yield stress versus yield strain for all cream cheeses at 5°C. Texture map for the controlled stress method showing yield stress versus yield strain for all cream cheeses at 22°C. Texture map for the vane method showing yield stress versus yield strain for all cream cheeses at 5°C. Texture map for the vane method showing yield stress versus yield strain for all cream cheeses at 22°C. xi 59 60 61 62 74 75 76 77 40:357. >6 70,3 70. v iupl Nomenclature slope ofline defined by Eq. [2.8] (N) vane diameter (m) shear modulus (Pa) gap width of parallel plate (mm) height of vane (m) consistency coefficient (Pa 3") constant defined by Eq. [2.3] (dimensionless) torque (N 1n) peak torque (N m) radius of vane (m) radius of parallel plate (1n) time (s) shear strain (dimensionless) yield strain, controlled stress method (dimensionless) yield strain, vane method (radians) shear rate (3") plastic viscosity of a Bingham fluid (Pa s) Newtonian viscosity (Pa 3) Shear stress (Pa) xii yield stress (Pa) shear stress over top and bottom surfaces of cylindrical sample (Pa) rotational speed (rpm) sweep angle (radians) xiii Chapter 1 Introduction Rheology is the study of the deformation and flow of matter, or how material responds to applied stress or strain. Its application in the food industry includes product development, process engineering calculations, and quality control. When used for quality control purposes, rheological properties can be correlated to consumer acceptance (sensory panels), then used to set guidelines to ensure consistency of product flow behavior. The simplest approach for modeling flow behavior is the Newtonian model, 0: w [1.1] which states that Shear rate is directly proportional to shear stress. The classic example of a fluid exhibiting this behavior is water, but honey, vegetable oil, and corn syrup are also Newtonian. A power law model 0 2 Kit” [1.2] describes the behavior of fluids that decrease in apparent viscosity (n < 1) with increased shear rate (Shear-thinning), Or increase in apparent viscosity (n > 1) with an increase in shear rate (shear-thickening). Other constitutive models take into account the presence of a yield stress (00): Bingham plastic (0:00 +Hp17); Herschel-Bulkley (o = 90 + K?" ), and Casson (oo'5 = 0005 + K705 ). Yield stress can be defined as the minimum shear stress required to initiate flow. A commonly observed example of a substance with a yield stress is toothpaste. Toothpaste does not readily flow from the tube when inverted, but flows easily when sufficient force is exerted on the tube. The minimum amount of force needed to get toothpaste to flow from the tube can be thought of as the yield stress. Barnes and Walters (1985) questioned the existence of a yield stress, but others have shown that it is an engineering reality (Harnett and Hu, 1989) which can have a large influence on process engineering calculations. The yield stress of a semi-solid or fluid food can give insight into its texture and flow behavior. The measurement of yield stress in foods such as butter, margarine, cream cheese, and ketchup can be used to quantify spreadability. For products such as these, spreadability is the most important textural property (Elejalde and Kokini, 1992). It has been shown that subjective spreadability is inversely proportional to the amount of shear stress on the knife (Kokini, 1985). Therefore, the yield stress can be used to characterize spreadability by measuring the force required to get a viscoelastic food product to flow (spread). Mortensen and Danmark (1982) found that yield stress alone is a sufficient measure for spreadablity of butter samples, and correlates better with sensory spreadablity than apparent viscosity. Yield stress has also been shown to correlate well with sensory spreadability data from untrained panelists (Rohm, 1990); and according to Daubert et a1. (1998), yield points of spreadable foods can provide manufacturers with useful parameter for quality improvement and control. 1. Literature Review 1.1. Sensory Spreadability and Instrumental Techniques The perception of food texture from a sensory perspective is very complicated process that is not completely understood. Attempts to emulate this process with simple instrumental methods are very difficult and not always exact, Since they typically measure only one property where sensory perception of texture can be a combination of many different stimuli. An advantage of using instrumental methods is that they are less expensive, quicker, and more reproducible than sensory tests (Elejalde and Kokini, 1992). Instrumental techniques used to measure food texture can be categorized into three groups: empirical tests, imitative tests, and fundamental tests (Scott- Blair, 1958). Empirical tests typically measure a physical property under well- defined conditions. Examples of empirical test methods include the Bostwick Consistometer, which is used to evaluate the flow properties of pureed foods; dough testing equipment such as the Farinograph, Mixograph, and Alveograph; and the cone penetrometer. Imitative tests attempt to simulate the conditions to which food is subjected to in the mouth. A good example of this type of test is Texture Profile Analysis, which is the most recognized instrumental method of evaluating the texture of solid and semi-solid foods (Steffe, 1996). Fundamental tests measure physical properties, such as viscosity, that are independent of the measurement apparatus. Data from these tests are expressed in well-defined scientific terms, and can be useful in predicting values of other physical properties and for making comparisons between different instruments. Spreadability is a textural property of foods that is difficult to characterize. Kokini and Dickie (1982) performed sensory tests where panelists Spread numerous food products on crackers with a knife to evaluate their spreadability. Rohm and Ulberth (1989) used a similar method to evaluate the Spreadability of butter, except the samples were spread on filter paper. Most of the work done investigating spreadability has focused on butter and margarine. The spreadability of these products is very important to the consumer (Fearon and Johnston, 1989). Most attempts to evaluate Spreadability instrumentally have been with the cone penetrometer. This method is the commonly accepted manner in which the texture of spreadable products is compared. As previously stated, this method falls under the category of empirical testing. The ability to measure a fundamental property that relates to spreadability, but is still as easy and inexpensive as empirical testing, could be a valuable tool for both quality control and product development applications. The use of the vane method and controlled stress method for evaluating yield stress (the focus of the current research) are both attempts to find such a tool. 1.2. Cream Cheese Cream cheese is a soft, unripened cheese formed by lactic acid fermentation. By federal standards, cream cheese must contain at least 33 percent fat, and not more than 55 percent moisture (Kosikowski and Mistry, 1997). Cream Cheese is made by pasteurizing and homogenizing a mixture of whole milk and cream, or whole milk, cream, skim milk, and skim milk powder; then cooling to a temperature suitable for lactic acid bacterial growth. Starter culture is added, and an acid curd is produced after incubating for a set period of time. The curds in cream cheese manufacturing differ from other harder cheese in that they do not need to be cut; breaking the curds with mechanical agitation is adequate. The curds are then cooked in water, drained in nylon or muslin bags, and made into either hot-pack, or cold-pack form. The difference between hot-pack and cold-pack cream cheese is established by how curds are treated after draining. In cold-pack cream cheese, the drained curds are salted, stabilized, kneaded, then packaged for consumption. In hot-pack cream cheese (the predominant procedure in large operations), salt and stabilizers are added to the curds in large vats or kettles. The mix is standardized, homogenized, then pumped in to packages hot. Neufchatel cheese is a product very similar to cream cheese, only lower in fat content (Kosikowski and Mistry, 1997). By legal definition, Neufchatel cheese should contain at least 20 percent, but not more than 33 percent, fat and not more that 65 percent moisture. Neufchatel cheese is manufactured in the same manner as cream cheese, only starting with a lower fat content mix. Neufchatel cheese gets its name from the town in northeastern France where it originated. Low-fat and fat free cream cheeses are manufactured in a similar manner to Neufchatel and cream cheese, but contain ingredients not found in regular cream cheeses. Added ingredients typically include artificial flavors, colors, and thickeners used to mimic the characteristics of the absent fat. When formulating low-fat or fat free cream cheeses, the goal is to make the product texturally similar to regular cream cheese. Whipped cream cheese is a product injected with gas, usually nitrogen, so it will have a lighter, fluffier texture. The procedure for whipping cream cheese is similar to producing overrun in ice cream. Bakers’ cheese is a product similar to cream cheese, but is practically fat free because it is made from skim milk only. Table 1.1 displays the composition of various unripened. soft cheeses similar to cream cheese. Little research has been done on the spreadability of cream cheese. Sanchez et al. (1996a) looked at physical properties of double cream cheese containing whey protein concentrate. Double cream cheese is similar to cream cheese, only with a higher fat content. Rheological and microstructural characteristics were also examined by Sanchez et al. (1996b). Textural qualities of cheeses similar to cream cheese were examined by steady shear concentric- cylinder testing at various shear rates (Omar et al., 1995). When attempts were made to correlate sensory apparent viscosity data with experimental data, Table 1.1. Typical composition (in percent) of commercial unripened soft cheeses (Kosikowski and Mistry, 1997). Cream Neufchatel Bakers’ Quark Ymer Cheese Fat 33.5 20.5 0.20 0.2 3.0 Moisture 54.0 64.0 74.0 79.0 85.0 Protein 9.80 12.0 19.0 15.0 6.7 Salt 0.75 0.75 0.0 0.7 0.0 Gums‘ 0.30 0.35 0.0 0.0 0.0 *Gums include karaya, tragacanth, guar, xanthan, carrageenan, and locust bean. problems occurred due to the question of what value of the apparent viscosity Should be used for comparison, since it is dependent on the shear rate. Cone and plate rheometry was utilized to evaluate the consistency coefficient (K) and flow behavior index (n) by Massaguer et al. (1984). Yield stress values obtained by a step-stress method using cone and plate geometry, a technique where the stress on the sample is increased in small steps until visible flow is observed ranged from approximately 71 to 350 Pa. 1.3. Yield Stress Measurement The evaluation of textural qualities of foods can be carried out by a variety of methods: empirical, imitative, and fundamental. Typically, the methods used to evaluate the spreadablity of foods are highly empirical. This means that results obtained from these tests depend upon the test conditions and type of instrumentation used. In addition, test results are not always in well-defined scientific units, making comparisons of results difficult. Barnes (1999) provides a very thorough review of the history and applications of the measurement of yield stress. He found that the terms “yield stress” or “yield point” were referenced almost 2500 times within the past dozen years. Clay, toothpaste, drilling mud, molten chocolate, creams, and ketchup are a few examples of materials reported having a yield stress. An interesting concept discussed by Barnes is that if measured at a sufficiently low speed, no materials would exhibit a yield stress. He concludes that measurable flow will occur in any material given sufficiently sensitive equipment. However, he does agree that in many applications the use of an “apparent” yield stress, i.e., a yield stress found under the applicable shear rate of the process, is very important in determining the material’s flow behavior. Many methods can be used to evaluate the yield stress of semi-solid or fluid materials. A very common procedure is to extrapolate the shear stress versus Shear rate flow curve. Using the Bingham plastic, Casson, or Herschel—Bulkley models, yield stress is determined by extrapolating the flow curve to the point where the value of the shear rate is zero. Cone and plate, parallel plate, and concentric cylinder data have all been used for the extrapolation method. Yoshimura et a1. (1987) examined this technique in comparison with others using model oil-in water emulsions. A problem with this technique is that the yield stress values obtained depend on the rheological model used, and the shear rate range of the test (Ofoli et al., 1987). It is also difficult to use this technique on highly viscous materials, because the torque capacity of many rheological instruments can be exceeded, even at low shear rates. Cone penetration data has been used extensively to determine the yield stress of many foods. This method can be run in two modes: one where a known force is applied to the probe, and the distance the probe penetrates into the sample is measured; and the other where the probe penetrates the sample to a known depth, and the force required to reach that depth is recorded. Often the yield stress results from these tests are expressed as apparent yield values (Mortensen et al., 1982; Rohm and Ulberth, 1989), which is defined by the same units as yield stress, but is in fact an empirical parameter. Butter and margarine are most often the substances analyzed by the cone penetrometer (DeMan et al., 1990, 1991; Borwankar et al., 1992; Fearon and Johnston. 1989; Kawanari et al., 1981). Typical yield values from these tests range from 20 to 150 kPa. Tanaka et al. (1971) used the cone penetrometer to evaluate the texture of various foods: soft margarine, cheese spread, pudding, peanut butter, and jelly. Another method to evaluate Spreadability, which is similar to cone penetrometry, is the squeezing flow method. The two modes of operation in squeezing flow are the same as that of the cone penetrometer; however, in this method the probe is a cylindrical flat plate, rather than a cone. Shukla et a1. (1995) used lubricated squeezing flow to evaluate the spreadability of butter. The yield stress of tomato paste, ketchup, mustard, and mayonnaise was determined from squeezing flow data by Campanella and Peleg (1987). Hoffner et a1. (1997) used lubricated squeezing flow data to evaluate the flow behavior index and consistency coefficient of mayonnaise and mustard. Measuring the minimum stress required to initiate flow is another frequently use method of determining yield stress. Various sensors and geometries can be utilized in these tests. The controlled rate vane method is one technique used in this research, and will be discussed later. The controlled stress method for determining yield stress is the other technique investigated in this work and will also be discussed later. 1.3.1. Vane Method Nguyen and Boger (1983) investigated the use of a vane to determine yield stress in red mud suspensions, which are residues from the processing of bauxite to extract alumina. The same researchers defined the limits of the vane and vessel dimensions (Figure 1.1): h/d < 3.5; Zl/d > 1.0; Zz/d > 0.5; and D/d > 2.0 (1985). Nguyen and Boger (1985) used vanes with 4 blades rotated at a constant speed. Vanes with 4, 6, and 8 vanes have been Shown to give similar results (Yoo and Rao, 1995; Yoshimura etal., 1987). The vane method can be run in two modes: controlled rate and controlled stress. In the controlled rate mode, which was used in this work, the vane is rotated in the sample at a constant angular speed while measuring the torque on the vane shaft. Angular speeds of 0.1 to 8.0 rpm have been shown to give comparable data (Nguyen and Boger. 1983), while speeds below 1.0 rpm are preferred (Steffe, 1992). Yoo et a1. (1995) compared the controlled rate method with the controlled stress method. The controlled stress method is performed by applying increasing, incremental levels of torque on the vane, then noting the torque at which the vane begins to move freely. Both methods give comparable results, but errors in the controlled stress method are common (Yoo et al., 1995), making the controlled rate method the preferred technique. One of the assumptions made when determining yield stress by the vane method is that the yielding surface is a cylinder that has the same dimensions as the vane. Keentok (1982) ll wd—a Z2 E D 417T Figure 1.1. Vessel and vane for yield stress determination. 12 found that the actual diameter of the sheared cylinder could be larger than the diameter of the vane. By measuring the diameter of the sheared cylinder of sample after testing, a ratio of the sheared cylinder diameter to the vane diameter was found. to be as high as 1.04. The vane method has been used to determine the yield stress of many foods, including apple sauce (Missaire et al., 1990; Usiak Godfrey et al., 1995), ketchup (Missaire et al., 1990; Yoo et al., 1995), ice cream (Briggs et al., 1996), tomato concentrates (Yoo and Rao, 1995). Most recently, Daubert et al. (1998) utilized the vane method to evaluate the spreadability of peanut butter, jelly, margarine, sour cream, full fat cream cheese, and fat free cream cheese. These researchers observed that the spreadability of these products might not relate to yield stress alone, but also to the yield strain. A material such as grape jelly was found to have a low yield stress, but could withstand a large degree of strain before yielding. Consequently, even though the low yield stress value indicated a very spreadable product, the ability of the material to resist deformation at low strains made uniform spreading more difficult. Alderman et al. (1991) evaluated yield strain from the vane method of bentonite gels by dividing the yield stress by the elastic modulus (G). For this calculation, the material was assumed to behave as a Hookean solid below its yield stress. This means that the material’s yield strain varied linearly with the yield stress. The method for evaluating yield strain by relating the time to the yield stress to strain (Daubert et al., 1998) was used instead of this more tedious method because of its simplicity, and the fact that it was not assumed that the cream cheeses behave in a Hookean manner. 1.3.2. Controlled Stress Method Yoo et a1. (1995) examined the yield stress of various foods with the vane method at a controlled shear stress. James et a1. (1987) used traditional geometries (cone and plate, concentric cylinder, etc.) to measure the minimum stress required to initiate flow in cohesive suspensions. The yield stress of toothpaste gel was determined with a cone and plate apparatus by Steffe (1996) by subjecting the sample to a linear change in shear stress while measuring the percentage of shear strain. Virtually no studies on foods have been reported using the controlled stress method with these geometries. A possible explanation for lack of information of this method is the comparatively high cost of controlled stress rheometers relative to controlled rate instruments. An advantage of the controlled stress method with a traditional sensor (cone and plate, parallel plate, etc.) is the capability to determine true strain at the yield point. This yield strain point can be found by simply observing the strain value that corresponds to the yield stress on the strain versus stress curve. Since yield strain might be a factor in determining spreadability, this method may receive further attention in the future. 1.3. Objectives The objectives of this research are: 1. to compare the single point vane method, slope vane method, and controlled stress method of evaluating the yield stress of cream cheese; 2. to examine the influence of fat content, temperature, and manufacturer on the variability of cream cheese yield stress; 3. to provide benchmarks for what is considered “good spreadability” for cream cheese; 4. to develop a texture map showing the relationship between yield stress and yield strain for common types of cream cheese. 15 Chapter 2 Theoretical Development 2.1. Yield Stress Calculations 2.1.1. Controlled Stress Method When the controlled stress method for determining yield stress is used, a shear strain versus shear stress curve that readily gives an approximate value for the yield stress is obtained. A problem arises, though, when an exact yield stress is desired. The complete curve may be considered as two separate curves: one where the applied stress is lower than the yield stress, and another where the applied stress is greater than the yield stress. As in Figure 2.1, the transition between the horizontal and vertical segments of the curve makes the exact determination of the yield stress difficult. The point of intersection of both segments gives the yield stress. This has been found in the past by using a ruler and pencil to draw “best-fit” lines, and subsequently the intersection point. A different method for finding the point of intersection was utilized in this work. Software (Haake RheoWin Pro 2.5) was applied to determine the best-fit line of both segments of the curve, then give the x-axis value of the intersection as the yield stress. This was accomplished by including an increasing number of points in the linear regression of the horizontal segment of the curve that produced the highest R2 value for correlation. Once the equation for this line was found, it was stored for later calculations. The equation for the vertical portion of the curve 10.0 ' I 9.0 I 8.0 7.0 I 6.0 I —<>-— Krafi Philadelphia Regular (22°C) ’ 5.0 I o Shear Strain (-: 4.0 I 3.0 Re ion 1 2.0 g 1.0 . (‘- -..'.~' ‘2"‘°§‘@® w . .ESW0WF‘ e 2 saw “3° ' 4" ’ .éfifiwéfi‘vv A“ 0.0 ‘ 1 . . ,, 1000 1050 1100 1150 1200 1250 1300 1350 1400 Shear Stress (Pa) Figure 2.1. Typical plot from a controlled stress method test on Kraft Philadelphia Regular cream cheese at 22°C showing shear strain versus shear stress. 17 was found in a similar manner, with the exception being that points with decreasing stress values were included until a maximum R2 value was obtained. The equations for each line were combined, and the x-value of the intersection was found. 2.1.2. Vane Method 2.1.2.1. Single Point Method An assumption made when using the vane method is that test material yields along a cylindrical surface having the outer dimensions of the vane. A simple torque balance can be constructed to find the total torque (M0) to overcome the yield stress of the material: M0 = (Whig—jg + 2jd/2272rzacdr [2.1] 0 where d, r, and h are the diameter, radius, and height, respectively, of the vane in meters (Steffe, 1996). Also, ac is the shear stress over the top and bottom surfaces of the cylindrical sample. Simplification of Eq. [2.1] gives (l/Z M = 7:th () 0'” + 472'] 0 flogdr [2.2] A reasonable assumption is made that 0}, varies with the radius according to a power-law relationship: 0', = f(r) = (221:) 00 [2.3] where m is constant. Substituting Eq. [2.3] into Eq. [2.2] gives _zzhd2 0 2 M 0 0'0 + 472' [Mrz[%) and)” [2.4] Integration and simplification of Eq. [2.4] provides an expression for the yield SII‘CSS: 2M h 1 " a, = ——[— + ] [2.5] 7rd" d m + 3 The assumption that m = 0 is acceptable when making quality control comparisons between samples or when quick, straightforward evaluations are needed (Steffe, 1996). This assumption has been shown to be reliable in numerous studies involving the vane method (Daubert et al., 1998; Yoo et al., 1995; Yoo and Rao, 1995). In this work, the top surface of the vane was loaded flush to the top surface of the sample which eliminates error due to the upper end effect. When using this orientation and the assumption that m = 0, the following equation is used to calculate yield stress: 6M = 0 2.6 0" 7rd2(3h + d) [ ] Since the vane dimensions are known for any particular test, Eq. [2.6] reduces to a very simple expression: 0 = constant M .1 [2.7] where the constant only depends upon the dimensions of the vane. Eq. [2.7] was used in all single point yield stress calculations in the current work. 2.1.2.2. Slope Method Equation [2.2] is needed to calculate yield stress using the slope method: Irhd 2 {/2 M0 =———0' +47rf r20",dr 2 () O ( Using this expression, M0 versus 17 can be plotted as M n = ah + constant [2.8] The slope, a, of the best-fit line through the torque versus vane height data includes the yield stress: d2 a z ” 0.. [2.9] 2 therefore, 2 a“ = a, [2.10] 7rd“ To determine a, raw data of peak torque versus vane height are plotted (Figure 2.2), and the slope is found using a standard linear regression program. At least three vanes, having the same diameter and different heights, are needed for this analytical method. The advantage of the slope method is that no end effect 20 0.035 0-030 * Vane Method - Slope @ Store Brand F at Free (5°C) 9 0.025 - E a 0.020 O) :5 0" S 1— f? 0.015 0.. y = 0.962x - 0.006 0.010 R2 = 0.96 0.005 0.000 . . 0.015 0.020 0.025 0.030 0.035 0.040 Vane Height (m) Figure 2.2. Example of the slope vane method for Store Brand fat free cream cheese at 5°C (a = 0.962). 21 assumptions are required because the integral portion of Eq. [2.2] is not needed to determine the yield stress. 2.2. Yield Strain Calculations 2.2.1. Controlled Stress Method Calculation of the yield shear strain (ym) at the yield stress point in the controlled stress method is straightforward. Observing the raw data from a shear strain versus shear stress graph, the y-axis value of strain (dimensionless) at the yield stress point is the true yield strain. The yield strain can be calculated as (Steffe, 1996) p = ._ 2.11 12,. h l l where R is the outer radius of the plate, '1’ is the sweep angle (rad), and h is the vertical distance between the parallel plates. When the yield stress value fell between points in the raw data table, the yield strain was found by extrapolation. Finding the yield strain by this method may introduce a small amount of error, but is preferred over finding an exact solution from a best-fit equation for the data. 2.2.2. Vane Method The strain found at the yield stress point from the vane method is not an actual shear strain, but a simple approximation based on the angular rotation (in 22 radians) of the vane at the yield stress. Vane yield strain (yav), or apparent yield strain, was defined as _ ’_~‘?_ 70.27! [2.12] where t is the time required to reach the yield stress point (seconds), [2 is the rotational speed (revolutions per second), and the factor 27: is used to convert the result from revolutions to radians. Equation [2.12] was also used by Daubert et a1. (1997) to determine yield strain. The time to the yield stress point can be taken directly from raw data. If an instrument lacks the capability of displaying a data table, a simple timer could be used to obtain the needed information. 23 Chapter 3 Materials and Methods 3.1. Equipment 3.1.1. Controlled Stress Method A Haake Model R8100 RheoStress controlled stress rheometer (Haake USA, Paramus, NJ) equipped with a 20 mm serrated parallel plate sensor connected to a load cell with a 5 N cm transducer (Figure 3.1) was used to determine yield stress by the controlled stress method. The rheometer was interfaced with a computer for data acquisition and measurement control, using Haake RheoStress software. A serrated plate apparatus was employed to eliminate any effects due to slip. A Haake F3/CH water bath was used to control sample temperature at 5°C (refrigeration temperature) or 22°C (room temperature) within a variation of i0.5°C. 3.1.2. Vane Method A Haake Viscotester VT550 with a torque capacity of 3 N cm equipped with a four-bladed stainless steel vane was used to determine yield stress by the vane method. Three 10.0 mm diameter vanes with height dimensions of 20.0. 25.0, and 35.0 mm were machined from a single piece of stainless steel. The vane blades were no more than 0.7 mm thick, and there was no center Shaft because vanes were machined from a single piece of stainless steel. 24 Figure 3.1. Cream cheese loaded into Haake RS100 for controlled stress test between serrated parallel plates (note digital meter indicating gap width of 2 mm). 25 Figures 3.2 and 3.3 illustrate the geometry of the vanes. These sensors were attached to the Haake VT550 with an adapter chuck. Precise temperature control for the vane testing was not possible because each sample was tested in the original container. Like the controlled stress method, two temperatures were used: 5°C and 22°C. Samples kept at refrigeration temperature were monitored with a solid state mini-thermometer to ensure all tests were run at temperatures of 5.0i0.5°C. Before a sample reached the upper temperature limit, it was returned to the refrigerator. Samples to be run at room temperature were brought up to 22.0:0.5°C without supplemental external heating (i.e. waterbath or oven) to eliminate any effects heating rates or high temperatures would have on the tests. This was done to better imitate what might be experienced by a consumer. The Haake VT550 was equipped with a torsion bar system, which contributed negligible wind-up error to yield stress calculations. Such a system has been shown to introduce errors less than two percent (Steffe, 1996). 26 Hgmm isgi‘ttqltiufi’t‘a’izi:1,2mpm!111mmpilipinpnqnwimpunmipiupmu 3 , t ,1 1 ‘ I . ' - ,t m Q If) :1 12 1}; I“. m I». 1, :4 Figure 3.2. Four-bladed stainless steel blades (scale in cm) used in vane method test. 27 2.5 4.5 /\ \/ 1.0 \ 0.05 0.40 Figure 3.3. Example of vane dimensions (cm) for the medium-sized vane. 28 3.2. Experimental Materials Ten plain (unflavored) cream cheese samples of varying fat levels from three manufacturers were analyzed by both the controlled stress and vane methods. The total fat and carbohydrate contents of the samples are summarized in Table 3.1. Five varieties of Kraft Philadelphia (Glenview, IL), three private label store brand (RASKAS, Inc., St. Louis, MO), and two restaurant chain (Bruegger’s Corp., Mishawaka, IN) cream cheeses were used to represent a wide variety of cream cheeses. Each brand of cream cheese was chosen to depict a product from a different type of manufacturer. Kraft Philadelphia varieties reflect a well known national brand, with very large-scale production, sold in many retail outlets. Originally, Meijer and Spartan store brands were regarded as representing different regional manufacturers, but were found to be identical samples manufactured by Raskas, Inc. Therefore, results from these samples could be pooled together. Cream cheeses manufactured by Raskas, Inc. were chosen to represent typical private label, or generic brands. Raskas, Inc. produces approximately eighty percent of the nation’s private label brands. Bruegger’s cream cheeses were purchased at the local Bruegger’s Bagels restaurant, and represent a national brand available only through the company’s retail outlets. Only plain, unflavored cream cheese was used in testing. This was done to avoid problems that flavoring and particulates such as berries, onions, or other materials may cause during testing. Also, sample homogeneity was important to 29 acquire reliable yield stress results. Tub cream cheese was selected over block cream cheese whenever possible. All Kraft Philadelphia varieties excluding Neufchatel, and both Bruegger’s varieties were purchases in tubs (either eight or sixteen ounces). All store brands, as well as Kraft Philadelphia Neufchatel, were purchased in bricks (eight ounces). 30 _ 9 .o n. Table 3.1. Percentage of total fat and carbohydrate of cream cheese samples: Sample Total Fat Total Carbohydrate Kraft Philadelphia Regular 33.3 3.33 Kraft Philadelphia Neufchatel 21.4 3 .57 Kraft Philadelphia Light 15.6 6.25 Kraft Philadelphia Fat Free 0.00 6.06 Kraft Philadelphia Whipped 33.3 4.76 Store Brand Regular 33 .3 6.67 Store Brand Neufchatel 20.0 6.67 Store Brand Fat Free 0.00 13.3 Bruegger’s Regular 34.6 7.69 Bruegger’s Light 17.3 11.5 *All values obtained from package information. 31 3.3. Data Collection 3.3.1. Controlled Stress Method Prior to loading and testing samples, the parallel plate apparatus was attached to the Haake RS100, the temperature was allowed to equilibrate for approximately one half hour. Samples were loaded onto the bottom (stationary) plate. Great care was taken to ensure that samples were not excessively sheared prior to testing. Small cylindrical slabs of cream cheese, just larger than the 20 mm diameter of the plate and approximately 0.5 mm thicker than 2 mm, were loaded onto the plate with a flat-bladed Spatula. The bottom plate was raised automatically to a gap width of 2 mm. Gap widths of 1.0 and 1.5 mm were used in preliminary tests, but found to cause too much sample Shearing prior to testing. Samples were allowed to rest between the plates for ten minutes before starting the stress ramp. This allowed the sample to come to the correct temperature, and relax compression induced strain caused by loading. In preliminary testing, a stress sweep from 1000 Pa to 3000 Pa was run on each sample at each temperature for 600 seconds while measuring strain. An approximate yield stress was found from the resulting strain versus stress plot to determine the appropriate stress range to run subsequent tests. Stress values of 1000 Pa above and 1000 Pa below the estimated yield stress were chosen as the end points of the sweep. For example, if the preliminary result from the stress sweep was approximately 1500 Pa, the subsequent stress sweeps would be run from 500 to 2500 Pa. This procedure ensured that each strain versus stress plot 32 had approximately the same number of data points, giving each graph the same resolution. During a regular 600 second test, 200 data points were collected. By properly adjusting the starting and ending stress values of the sweep, approximately the same number of data points could be taken during each test. Overall, three stress sweeps were used: 500 to 2500 Pa; 1000 to 3000 Pa; and 2000 to 5000 Pa. Each sample was evaluated in the same manner for fifteen replicates. The raw data from each test was saved and analyzed to determine precise yield stress and yield strain values. 3.3.2. Vane Method Vane tests were run at a constant rotational speed of 0.5 rpm for thirty seconds, while collecting 100 data points. While in the original container, cream cheese was raised up to the vane using a scissors stand. The sample was raised until the top surface of the sample was even with the top of the blades. This was done with minimum sample disturbance. The effect of the vane entering the sample is negligible because the actual yielding surface is on the outer edge of the cream cheese, not the surface of the blade. Also, differences between sample temperature and vane temperature are insignificant because of the short testing time and highly insulating nature of the cream cheese. The test was started immediately after the sample was loaded. Torque on the vane shaft was measured and recorded for later analysis by either the single point or slope method. 33 Five replicates using each vane were run on each sample for a total of fifteen replicates. When using the largest size vane to analyze the Kraft Philly fat free cream cheese, the maximum torque capacity of the VT550 was exceeded; therefore, five additional replicates were run with the medium-sized vane to bring the total number of replicates to fifteen. The same was done with both Bruegger’s varieties because the packaging of the Bruegger’s cream cheeses was not deep enough to accommodate the large vane. Attempts were made to bypass this problem, but excessive handling of the product deemed this method impractical. To allow for testing brick samples with the large vane, the bricks were carefully cut in half and the halves were stacked to provide a sample deep enough to accommodate the length of the large vane. 3.4. Data Analysis 3.4.1. Controlled Stress Method Haake RheoWin Pro software (Version 2.5) was used to analyze the data from the controlled stress method plots. After plotting the strain versus stress curve, the software was used to find the best-fit lines of the horizontal segment of the curve and the vertical segment of the curve. After finding the equations for these lines, the shear stress value corresponding to the x-value of the intersection point of the lines was calculated. This point is the yield stress. Strain at the yield point was determined in a similar manner, using the y-value of the intersection as 34 the yield strain. All yield stress and strain values were tabulated and evaluated using Microsoft Excel spreadsheets. 3.4.2. Vane Method Haake RheoWin Pro software (Version 2.5) was used to evaluate torque versus time curves from vane method tests. The highest torque achieved during the controlled rate test was used in calculating yield stress by the single-point method using Eq. [2.6]: 6M 0 0' : " 7rd2(3h + d) Each set of fifteen peak torque readings (five for each vane height) were used in calculating yield stress by the slope method. The strain at yield stress was calculated by using the time to reach the yield stress to determine the sweep angle required in Eq. [2.1 1]. Chapter 4 Results and Discussion 4.1. Vane Method Figures 4.1 through 4.10 Show typical plots of torque versus time for each cream cheese variety tested. All examples are from tests run at 22°C, but also reflect the trends observed in data from tests run at 5°C. No Significant differences were observed in curve shapes between temperatures. All curves show a clear maximum torque, which is required to calculate yield stress by Eq. [2.6]. Most curves are very Similar; however, the curves for the Kraft Philadelphia light (Figure 4.3), Kraft Philadelphia fat free (Figure 4.4), and store brand fat free (Figure 4.8) have the sharpest peaks. This indicates that once the yield stress is reached, the sample was quickly broken down and flowed easily. The graphs for the Kraft Philadelphia Regular (Figure 4.1), Store Brand Neufchatel (Figure 4.7), and Bruegger’s Light (Figure 4.10) samples are similar, with a evenly rounded peak. The Kraft Philadelphia fat free cream cheese had the highest yield stress values at both 5°C and 22°C (Table 4.1), while having the curve with the sharpest peak. This is due to the extensive fracturing observed in this cheese when the shear strain exceeded the yield strain. Spiral fracture lines, approximately 0.5 to 1 cm in length, extended from the edges of the blades outward as the yield stress point was exceeded. This variety was observed to be much harder, and to have 36 Table 4.1. Vane method (single point) yield stress results (Pa) at 5°C and 22°C. Sample 5°C 22°C Kraft Philly Regular 4407 (524.9)‘3 1891 (175.4)a Kraft Philly Neufchatel 2724 (3 87.9)b 1248 (92.80)b Kraft Philly Light 5023 (467.4)8 1482 (130.7)C Kraft Philly Fat Free 6587 (726.8)° 5144 (466.1)d Kraft Philly Whipped 1296 (159.3)b 461.1 (4150)6 Store Brand Regular 4556 (182.0)a 1875 (156.9)1l Store Brand Neufchatel 3291 (149.1)a 1462 (115.2)C Store Brand Fat Free 4273 (470.9)‘11 2882 (139.9)f Bruegger’s Regular 6270 (874.7)C 1518 (99.80)C Bruegger’s Light 3837 (492.8)3 1255 (136.2)b ab“Means with different superscripts are significantly different (p<0.05). Comparisons made in same column only. Results shown as mean (standard deviation); n=15 for all samples. 37 0.0070 0.0060 ~ F—T, M0 40 09 0.0050 0.0040 (9 Torque (N m) 0.0030 0.0020 - 9 0.0010 . 0.0000 '-*v 0 5 10 15 20 25 30 Time (s) Figure 4.1. Typical plot from a vane method test showing torque versus time for Kraft Philadelphia Regular cream cheese at 22°C (M0 denotes peak torque achieved). 38 0.0045 0.0040 -‘ 0.0035 0.0030 0.0025 .9 0.0020 Torque (N m) 9 Kraft Philadelphia Neufchatel (22°C) 9 0.0015 ‘ 0 0.0010 ea 0.0005 ’ 0.0000 -- Time (s) Figure 4.2. Typical plot from a vane method test showing torque versus time for Kraft Philadelphia Neufchatel cheese at 22°C (M0 denotes peak torque achieved). 39 0.0070 0.0060 4’ 6 0.0050 0.0040 Torque (N m) 0.0030 0.0020 9 Kraft Philadelphia Light (22°C) 0.0010 0.0000 . 0 5 10 15 20 25 30 Time (s) Figure 4.3. Typical plot from a vane method test showing torque versus time for Kraft Philadelphia Light cream cheese at 22°C (Mo denotes peak torque achieved). 40 0.0140 0.0120 - 3 w 6 Q; 9 0.0100 . fl 0 A ‘96 E 0.0080 * 9. a «to g N». 2' \ 0 0.0060 n 1— N “0 0.0040 N O Q. N” - . . at 0-0020 ‘ o Kraft Phlladelphla Fat Free (22°C) "W 0.0000 0 5 10 15 20 25 30 Time (s) Figure 4.4. Typical plot from a vane method test showing torque versus time for Kraft Philadelphia Fat Free cream cheese at 22°C (M0 denotes peak torque achieved). 41 0.0018 0.0016 - 0.0014 ' ”mm“ cm,“ W 0 Wine» 0.0012 0.0010 I 0.0008 Torque (N m) .9 . . ' 0 0.0006 9 Kraft Philadelphia Whipped (22 C) 0.0004 430 0.0002 0.0000 Time (s) Figure 4.5. Typical plot from a vane method test showing torque versus time for Kraft Philadelphia whipped cream cheese at 22°C (Mo denotes peak torque achieved). 42 0. 0070 0.0060 “WW M0 g . 0.0050 ‘ 36‘. ° *9 «o 3 0.0040 , " \ a “M“. 3 <0 2' is 0.0030 <6 0.0020 .. . _, 9 Store Brand Regular (22°C) 0 0.0010 9‘ ‘. 0.0000 0 5 10 15 20 25 30 Time (s) Figure 4.6. Typical plot from a vane method test showing torque versus time for Store Brand Regular cream cheese at 22°C (M0 denotes peak torque achieved). 43 0.0060 W Mo 0.0050 - *9 {’0 . a.“ 0.0040 - t E 3 a 3 0.0030 ~ 0" ‘5 1— 9. 0.0020 9 Store Brand Neufchatel (22°C) 0.0010 ' 0 0.0000 . . 0 5 10 15 20 25 30 Time (s) Figure 4.7. Typical plot from a vane method test showing torque versus time for Store Brand Neufchatel cheese at 22°C (M0 denotes peak torque achieved). 44 0.0120 0.0100 ¢ W.— Mo , ‘09 0.0080 ~ " ’9 0.0060 - ' 0‘ 9 . \‘NW . . 0.0040 ’ NM I 9 Store Brand Fat Free (22°C) Torque (N m) 0.0020 0.0000 Time (s) Figure 4.8. Typical plot from a vane method test showing torque versus time for Store Brand Fat Free cream cheese at 22°C (M0 denotes peak torque achieved). 45 0.0060 0.0050 F 0‘ K; M° ,6 3‘ ‘ . 0.0040 o ”if 4. E g 0.0030 2' o O [— 0.0020 9 0 Bruegger's Regular (22°C) 0.0010 ‘ 0.0000 -. 0 5 10 15 20 25 30 Time (s) Figure 4.9. Typical plot from a vane method test showing torque versus time for Bruegger’s Regular cream cheese at 22°C (M0 denotes peak torque achieved). 46 0.0045 0.0040 0.0035 0.0030 * 0.0020 Torque (N m) 0.0015 0.0010 0.0005 0.0000 0.0025 - CO 0 Bruegger's Light (22°C) 10 15 20 Time (s) 25 30 Figure 4.10. Typical plot from a vane method test showing torque versus time for Bruegger’s Light cream cheese at 22°C (Mo denotes peak torque achieved). 47 a more crumbly texture than the others, especially at 5°C. The curve for the Store Brand regular cream cheese (Figure 4.6) shows a double peak. In almost all cases, the torque value at the first peak was higher than the second one. This sample was the only one to exhibit this type of curve, which is odd since it has practically the same composition as the Kraft Philadelphia regular cream cheese. The Store Brand variety may have a slightly different structure induced during processing, as all the curves for this type of cream cheese showed this double peak. The curve for the Kraft Philadelphia whipped cream cheese (Figure 4.5) was slightly different from the others as it was fairly flat after the yield stress had been attained. This is likely due to product processing to make the yield stress very low compared to other varieties of cream cheese (Table 4.1). By incorporating air into a full fat cream cheese, a very light and porous structure can be obtained. The curve for the Kraft Philadelphia Neufchatel (Figure 4.2) and the Bruegger’s Regular (Figure 4.9) had a similar shape. The curves produced by these products Show a material where the sample breakdown produces a yielding flow, like a Bingham plastic, and exhibit little fracture. The coefficients of variation for each vane size are all similar (5 to 9 percent). Coefficients of variation (in percent, defined as the standard deviation divided by the mean times 100) were used to make comparisons of the variability of results. Using these values allowed direct comparisons to be made between results at different magnitudes. At both test temperatures, the large vane gave the 48 highest variation. This was not expected because a larger surface area for testing with the large vane should lead to more repeatable results. One explanation for this is that the preparation of samples prior to testing with the large vane could have disturbed the structure more than the preparation method employed when using the small and medium sized vanes. For the smaller vanes, no sample handling was required prior to running tests, as opposed to the cutting and stacking of brick cream cheeses needed to have an adequate thickness for testing. The medium sized vane had a slightly lower coefficient of variation for all samples than the small vane. Figure 4.11 illustrates an example of how the slope method examination of vane data produces a value of the yield stress. In this example the fit of the curve is very good (R2>0.96). A value of 0.9621 N is found for the slope (a) which is used in Eq. [2.10] to calculate a yield stress of 6125 Pa. Results from the slope method analysis are summarized in Table 4.2, as well as the correlation coefficient of the best-fit lines utilized to obtain values of a. Correlation coefficients range from a low of 0.56 to a high of 0.98. Good correlation, however, does not guarantee that results using the technique will compare well with the averages from the Single point tests. This is discussed further in Section 4.3. 49 Table 4.2. Vane method (slope method*) yield stress results (Pa) at 5°C and 22°C. Sample 5°C 22°C Kraft Philly Regular 3648 (0.71) 1944 (0.86) Kraft Philly Neufchatel 1667 (0.61) 1445 (0.92) Kraft Philly Light 3518 (0.78) 874.1 (0.58) Kraft Philly Fat Free ” 4365 (0.86) Kraft Philly Whipped 994.4 (0.69) 586.3 (0.94) Store Brand Regular 4393 (0.97) 1518 (0.84) Store Brand Neufchatel 3659 (0.98) 1066 (0.84) Store Brand Fat Free 6125 (0.96) 2464 (0.96) Bruegger’s Regular m Bruegger’s Light m ... i Results calculated by Eq. [2. 10]. Correlation coefficients (R2) in parentheses. H Use of large vane exceeded torque capacity of instrument. mContainers too shallow to allow use of large vane. 50 0.035 0-030 * Vane Method - Slope @ Store Brand Fat Free (5°C) a 0 0.025 -< E a 0.020 1 d.) :3 3 O [_. "E43 0.015 a. y = 0.962x - 0.006 0.010 _. R2 = 0.96 0.005 , 0.000 1 . , 0.015 0.020 0.025 0.030 0.035 0.040 Vane Height (m) Figure 4.11. Example of the slope vane method for Store Brand fat free cream cheese at 5°C (a = 0.962). 51 4.2. Controlled Stress Method Figures 4.12 to 4.21 show typical shear strain versus shear stress curves from 22°C controlled stress tests. These curves also represent tests run at 5°C, as no differences in the shape of the curves were observed. The solid line in Region 1 indicates the best-fit line for that segment of the strain versus stress curve, and the dashed line in Region 2 indicates the best-fit line for the vertical segment. The intersection of these two lines, illustrated by the thick vertical arrow, shows the point where both the yield stress and yield strain are determined. Each curve Shows an obvious transition from solid-like behavior (Region 1) to shearing flow (Region 2). In Region 1, the cream cheese can be thought of as a solid material, where if the applied stress was removed, the sample would return to its original shape. Between Regions 1 and 2 there is a transitional zone where the cream cheese begins to yield, the structure breaks down, and flow begins. The yield stress lies in this transitional portion of the curve. Upon inspection of each curve, it is difficult to arrive at a precise value for the yield stress. A crude method of finding the yield stress from this type of graph (Figure 4.12 to 4.21) is to visually determine a best-fit line for each region and draw these lines using a ruler. A major problem with this manual method is that results are inconsistent from person to person. In addition, only an approximate number could be found since the final value was merely found from marking a point on the x-axis of the graph. The technique of determining the yield stress by statistically finding the best-fit lines in each region provides more precise, consistent results, 52 Shear Strain (-j 10.0 I I I 9.0 I 9 I 8.0 7.0 I, I I. 6.0 I "0" Kraft Philadelphia Regular (226C) I, 5.0 I 9 I Go I 4.0 - I, e I I .. 3.0 2.0 Region 1 Region 2 .. w rte-we're“ 9‘ “*9 ‘ I 1.». 1.34:. at: - w .r . I 0.0 - ! 1000 1050 1100 1150 1200 1250 1300 1350 1400 Shear Stress (Pa) Figure 4.12. Typical plot from a controlled stress method test on Kraft Philadelphia Regular cream cheese at 22°C showing shear strain versus shear stress. 53 10.0 . $1 9.0 ' 1 8.0 l I I. 7.0 f I ;; 6.0 — 1' .E l S 55 5.0 ~ 8 ~ * 90 a 4 0 Kraft Philadelphia Neufchatel (22°C) , , 3.0 i Region 1 _ Region 2 00 A z r 1.9 T I _ 1000 1200 1400 1600 1800 2000 Shear Stress (Pa) Figure 4.13. Typical plot from a controlled stress method test on Kraft Philadelphia Neufchatel cheese at 22°C showing shear strain versus shear stress. 54 10.0 9.0 - —-=1£; 8.0 - 7.0 " 6.0 5.0 1' . N 77 60 ' Kraft Philadelphia Light (22°C) : . V I I I l Shear Strain (-: 4.0 ' 3.0 2.0 Region 1 Region 2 00 A W75: xii/7'5"» '" "“' i 1 1000 1200 1400 I600 Shear Stress (Pa) 1800 2000 Figure 4.14. Typical plot from a controlled stress method test on Kraft Philadelphia Light cream cheese at 22°C showing shear strain versus shear stress. 55 10.0 I» I 9.0 .. l I I 8.0 I I '0 I 7.0 I I I Q) 6.0 ~ 1" .S I a l 5 5.0 - I 26‘ ‘ . -- ._ fl 0'0 I g 4 0 _ —<>- Kraft Philadelphia Fat Free (22°C) I 3.0 Region 1 2.0 1.0 .3 xx ”: 7. 3:12: L-Zat}.wast.s\-\’1~\"§{>t¢“it“ ' I 0.0 33‘1“” ‘ l ' 1000 1200 1400 1600 1800 Shear Stress (Pa) Figure 4.15. Typical plot from a controlled stress method test on Kraft Philadelphia Fat Free cream cheese at 22°C showing shear strain versus shear stress. 56 10.0 I Shear Strain (-j 500 600 700 800 900 1000 1 100 Shear Stress (Pa) Figure 4.16. Typical plot from a controlled stress method test on Kraft Philadelphia Whipped cream cheese at 22°C showing shear strain versus shear stress. 57 10.0 I I I 9.0 I I 8.0 .' l I 7.0 I l I g; 6.0 I? c I .§ ' m 5.0 ,' 0°: '55 4.0 -<>— Store Brand Regular (22°C) : . I I 3.0 ' 2.0 Region 1 l 2 1.0 0.0 " \ 1000 1200 1400 1600 1800 2000 2200 2400 Shear Stress (Pa) Figure 4.18. Typical plot from a controlled stress method test on Store Brand Regular cream cheese at 22°C showing shear strain versus shear stress. 58 10.0 9.0 __—- 8.0 7.0 6.0 5.0 Shear Strain (-j 4.0 —<>- Store Brand Neufchatel (22°C) 3.0 2.0 Region 1 Region 2 l l l 1000 1200 1400 1600 1800 2000 2200 2400 Shear Stress (Pa) Figure 4.19. Typical plot from a controlled stress method test on Store Brand Neufchatel cheese at 22°C showing shear strain versus shear stress. 59 10.0 9.0 8.0 7.0 6.0 5.0 Shear Strain (-: —<>— Store Brand Fat Free (22°C) i 4.0 I 20 Region 1 1500 1700 1900 2100 2300 2500 2700 Shear Stress (Pa) Figure 4.20. Typical plot from a controlled stress method test on Store Brand Fat Free cream cheese at 22°C showing shear strain versus shear stress. 60 10.0 9.0 8.0 7.0 6.0 5.0 Shear Strain (-3 4.0 + Bruegger's Regular (220C) 3.0 Region 1 2.0 1.0 00 _ <31" -...’.’. :1: .._‘..'.._..':\~.;\.. (“argue-c, .. I x 1000 1200 1400 1600 1800 2000 Shear Stress (Pa) Figure 4.21. Typical plot from a controlled stress method test on Bruegger’s Regular cream cheese at 22°C showing shear strain versus shear stress. 61 10.0 4 9.0 8.0 - 7.0 6.0 a 5.0 Shear Strain (-: 4.0 . . 0 ° —<>— Bruegger s nght (22 C) ° 3.0 ' \ Region 1 ’ Region 2 2.0 1.0 - 500 600 700 800 900 1000 1 100 1200 1300 Shear Stress (Pa) Figure 4.22. Typical plot from a controlled stress method test on Bruegger’s Light cream cheese at 22°C showing shear strain versus shear stress. 62 regardless of the person evaluating the graph. Table 4.3 summarizes the yield stress values determined from the controlled stress method at both test temperatures. In many of the strain versus stress graphs, the vertical portion of the curve (Region 2) extends beyond the boundaries of the graph. In Figures 4.16 and 4.17, for example, the last data point of the curve stops well before the maximum strain of 10.0 (1000%). This occurred when the cream cheese flowed easily, and the maximum rotational speed of the instrument was exceeded. At this point, the structure of the cream cheese is completely broken down and the fluid behavior dominates. The Kraft Philadelphia Whipped cream cheese had the lowest yield stress at both 5°C and 22°C. Figure 4.16 shows how this causes the maximum recorded strain to be much lower (less than 5.0) than that found with the other varieties of cream cheese. If the structure broke down quickly, it was important to ensure that enough data points are collected in Region 2 to find a best-fit line. Figure 4.21 shows a typical curve for Bruegger’s Light cream cheese. It is different from the other curves in two ways: 1) the transitional segment of the graph is larger, and 2) the segment of the graph in Region 2 is much less vertical. This indicates the material has a poorly defined yield stress. The structural differences could be explained by considering the many additional ingredients contained in the Bruegger’s Light cream cheese which are not in the other cream cheeses (whey protein concentrate; caseinates; citric, phosphoric and acetic acids). The addition of these ingredients seems to decrease the sharp contrast in behavior 63 Table 4.3. Yield stress (Pa) at 5°C and 22°C determined from the controlled stress method. Sample 5°C 22°C Kraft Philly Regular 4705 (474.7)a 1195 (114.6)al Kraft Philly Neufchatel 3645 (130.8)b 1799 (122.4)b Kraft Philly Light 2933 (207.9)c 1910 (124.8)b Kraft Philly Fat Free 3160 (164.7)d 1781 (178.8)b Kraft Philly Whipped 2073 (145.3)6 898.1 (76.20)° Store Brand Regular 7289 (687.7)f 2224 (246.4)d Store Brand Neufchatel 4760 (464.8)a 2054 (1 19.7)bd Store Brand Fat Free 2939 (216.7)C 2409 (238.1)6 Bruegger’s Regular 5270 (383.3)g 1863 (136.6)b Bruegger’s Light 2131 (218.1)6 1083 (110.0)8 albcMeans with different superscripts are significantly different (p<0.05). Comparisons made in same column only. Results shown as mean (standard deviation); n=15 for all samples. 64 between Region 1 and Region 2. By creating a product with this rheological behavior, Bruegger’s has produced a light cream cheese with spreadability second only to the whipped Kraft Philadelphia variety since it has the next lowest yield stress (Table 4.3). This is a good indication of the importance of the link between spreadability and yield stress. 4.3. Comparison of Methods and Temperature Effects Materials such as butter, margarine, and cream cheese are assumed to soften with increasing temperatures. Margarine has been shown to be softer at 10°C than at 5°C by cone penetrometry (DeMan et al., 1991). Consumers can readily distinguish the difference in softness (or hardness) between samples at refrigeration and room temperatures. While these differences are easily perceived by consumers, finding an instrumental method that can quantitatively evaluate hardness of a semi-solid is a challenge. Traditional methods of quantifying textural differences are empirical, and results from these methods are not easily compared with others. Results from the current study show that both the controlled stress and vane methods are capable of distinguishing differences in spreadability due to temperature changes. Also, both methods determine yield stress results in fundamental units (N m'z); hence, results from each can be compared directly. Table 4.3 shows the results from controlled stress tests at both 5°C and 22°C. Large differences in yield stress were found in all samples. Each yield 65 stress at 5°C was at least 1.5 times greater than results at 22°C, with the exception being the private label store brand. Values reported in Table 4.3 Show that the controlled stress method is an acceptable technique for detecting differences in spreadability due to the influence of temperature. For all controlled stress tests run at 5°C, the average coefficient of variation (CV) was 9.79%. For tests run at 22°C, the CV was 8.52%. The single point vane method showed similar success in detecting the influence of temperature on spreadability (Table 4.1). Again, most yield stress values at 5°C were at least twice as high as those at 22°C. The exceptions to this observation were Kraft Philadelphia fat free and Store Brand fat free cream cheese. Also, the CV for all the vane method tests run at 5°C was 10.5%, which is comparable to the value calculated for the controlled stress method. The CV for vane method tests run at 22°C was 8.09%, which again was comparable to the controlled stress method at the same temperature. Other factors which could have influenced results were the effects of aging, lot variability, and variation of major constituents (e.g. fat, protein) from the label declarations. The effects of these factors were not examined in this study, but could be investigated in future work. Table 4.4 summarizes the results from both the controlled stress and vane methods at 5°C. The only results found not to be significantly (p<0.01) different between methods are those for Kraft Philadelphia regular cream cheese. There is no observable trend for the other sample types. For samples such as Kraft Philadelphia light and fat free, and store brand fat free the vane method results are 66 considerable higher (1.5 to 2 times). In other samples, such as Kraft Philadelphia whipped, and store brand regular and Neufchatel, the controlled stress method is higher (1.5 to 1.6 times). In both Bruegger’s varieties, the vane method results are higher (1.2 and 1.8 times). All the slope method results with correlation coefficients less than 0.80 did not relate well to either the controlled stress or single point vane method. All of the slope method results from the Store Brand samples had very high correlation coefficients (R2 2 0.96), but only results for the regular and Neufchatel varieties compared well the single point vane method results. At 22°C, controlled stress and vane method results Show much less variation than those at 5°C (Table 4.5). The exception to this is the Kraft Philadelphia fat free yield stress, where the vane method values are nearly three times higher than the controlled stress results. Fracture lines were clearly observed in both methods for the fat free samples, and vane results are higher because the cream cheese tended to fracture much closer to the yield stress. Fracturing of the cream cheese was observed well before (approximately 300 to 500 Pa) the yield stress using the controlled stress method. The vane results using the slope method all had fairly high correlation coefficients (R2 2 0.84), with the exception of the Kraft Philadelphia light variety (R2 = 0.58). However, the high correlation coefficients did not guarantee a good comparison to the other methods. Steffe (1996) observed that yield stress data obtained from different techniques are often very different. This is most likely due to the different manner in which 67 Table 4.4. Yield stress (Pa) results at 5°C forthe controlled stress method (CS), single point vane method (SP), and the s10pe vane method (SL). Sample CS Vane (SP) Vane (SL)* Kraft Philly Regular 4705 (474.7) 4407 (524.9) 3648 (0.71) Kraft Philly Neufchatel 3645 (130.8) 2724 (387.9) 1667 (0.61) Kraft Philly Light 2933 (207.9) 5023 (467.4) 3518 (0.78) Kraft Philly Fat Free 3160 (164.7) 6587 (726.8) " Kraft Philly Whipped 2073 (145.3) 1296 (159.3) 994.4 (0.69) Store Brand Regular 7289 (687.7) 4556 (182.0) 4393 (0.97) Store Brand Neufchatel 4760 (464.8) 3291 (149.1) 3659 (0.98) Store Brand Fat Free 2939 (216.7) 4273 (470.9) 6125 (0.96) Bruegger’s Regular 5270 (383.3) 6270 (874.7) Bruegger’s Light 2131 (218.1) 3837 (492.8) IMHO! Results shown as mean (standard deviation) for CS and Vane (SP); n=15 for all samples. .Results shown at calculated yield stress (correlation coefficient). Use of large vane exceeded torque capacity of instrument. Containers too shallow to allow use of large vane. 68 Table 4.5. Yield stress (Pa) results at 22°C for the controlled stress method (CS), single point vane method (SP), and the slope vane method (SL). Sample CS Vane (SP) Vane (SL)* Kraft Philly Regular 1195 (114.6) 1891 (175.4) 1944 (0.86) Kraft Philly Neufchatel 1799 (122.4) 1248 (92.80) 1445 (0.92) Kraft Philly Light 1910 (124.8) 14812 (130.7) 874.1 (0.58) Kraft Philly Fat Free 1781 (178.8) 5144 (466.1) 4365 (0.86) Kraft Philly Whipped 898.1 (76.20) 461.1 (41.50) 586.3 (0.94) Store Brand Regular 2224 (246.4) 1875 (156.9) 1518 (0.84) Store Brand Neufchatel 2054 (119.7) 1462 (115.2) 1066 (0.84) Store Brand Fat Free 2409 (238.1) 2882 (139.9) 2464 (0.96) Bruegger’s Regular 1863 (136.6) 1518 (99.80) '” Bruegger’s Light 1083 (110.0) 1255 (136.2) ” Results shown as mean (standard deviation); n=15 for all samples. *Results shown at calculated yield stress (correlation coefficient). Containers too shallow to allow use of large vane. 69 strain is applied in different techniques. The lone sample to Show similar results for all methods was the Store Brand fat free cream cheese at 22°C. 4.4. Yield Strain Tables 4.6 and 4.7 summarize yield strain values from the controlled stress and vane methods. In general, the repeatability of these results was not as good as the yield stress results. The average CV for the controlled stress method at 5°C was 23.2%, while at 22°C the average CV was 24.0%. For the vane method, the average CV’s for yield strain results at 5°C and 22°C were 21.8% and 15.95%, respectively. Since the yield strains obtained from controlled stress and vane testing are not the identical property, no direct comparison can be made. The yield strain from the controlled stress method is an actual maximum shear strain at the outer edge of the plate. The yield strain obtained in a vane method test is simply a measure of the distance the vane rotated (in radians) from the start of the test until the yield stress point was reached. For both quantities to be compared directly, the rate of strain would have to be the same, and the material would have to be time- independent. If the materials to be tested were completely time-independent, varying the rotational speed in the vane method test and rate of stress increase in the controlled stress method test would not have a large influence on results. However, it has been found that yield stress can be a time—dependent property (Cheng, 1986). Other problems in calculating strain for the vane method are that 70 Table 4.6. Controlled stress method yield strain results (dimensionless) at 5°C and 22°C. Sample 5°C 22°C Kraft Philly Regular 0.743 (0.160)a 2.091 (0.363)“1 Kraft Philly Neufchatel 3.060 (0.688)b 3.036 (0.590)a Kraft Philly Light 1.254 (0.292)a 1.440 (0.541)a Kraft Philly Fat Free 1.414 (0.140)a 3.158 (0.876)8 Kraft Philly Whipped 2.908 (0.685)b 1.609 (0.305)3 Store Brand Regular 1.328 (0.269)8 1.810 (0.464)81 Store Brand Neufchatel 1.555 (0.528)21 2.047 (0.430)a Store Brand Fat Free 3.775 (0.964)b 1.658 (0.453)° Bruegger’s Regular 0.947 (0.264)8 1.601 (0.239)a Bruegger’s Light 2.137 (0.442)b 1.673 (0.569)a abcMeans with different superscripts are significantly different (p<0.05). Comparisons made in same column only. Results shown as mean (standard deviation); n=15 for all samples. 71 Table 4.7. Vane method yield strain results (radians) at 5°C and 22°C. Sample 5°C 22°C Kraft Philly Regular 0.239 (0.045)8 0.229 (0.071)a Kraft Philly Neufchatel 0.431 (0.107)” 0.614 (0.063)” Kraft Philly Light 0.225 (0.045)a 0.299 (0.075)8 Kraft Philly Fat Free 0.182 (0.046)8 0.194 (0.042)a Kraft Philly Whipped 0.629 (0.109)” 0.589 (0.064)” Store Brand Regular 0.462 (0.168)” 0.353 (0.071)3 Store Brand Neufchatel 0.456 (0.095)” 0.342 (0.054)81 Store Brand Fat Free 0.373 (0.046)” 0.387 (0.050)3 Bruegger’s Regular 0.274 (0.073)8 0.287 (0.053)a Bruegger’s Light 0.343 (0.058)” 0.418 (0.048)” abcMeans with different superscripts are significantly different (p<0.05). Comparisons made in same column only. Results shown as mean (standard deviation); n=15 for all samples. 72 the top, bottom, and side strains are different, and the material thickness must be known. Also, matching the rate of strain increase in both methods cannot be done, since the vane method is operated in a controlled rate mode, and in the controlled stress method the rate of strain is constantly changing. Therefore, comparing yield strain results from both methods is impractical. Daubert et a1. (1998) applied the use of a texture map of spreadability to simultaneously look at the effects of yield stress and yield strain on the spreadablity of various foods (peanut butter, jelly, cream cheese, margarine, sour cream, whipped toppings). These maps were used to categorize products by their degree of spreadability. Their work focused on the controlled rate vane method, and it remains to be shown if the spreadability map concept can also be applied to controlled stress data. Figures 4.22 through 4.25 are spreadability maps Showing yield stress versus yield strain values for both methods at both temperatures. Each point represents the mean of fifteen measurements, while the standard deviation is shown by error bars. For the vane method, Figures 4.24 and 4.25 show an observable trend where an increase in the yield strain corresponds with a decrease in yield stress. While the fit of the trend lines for the vane method at 5°C (Figure 4.24) and 22°C (Figure 4.25) was poor (R2 < 0.46 and R2 < 0.38, respectively), a trend was observed in the data that allowed the construction of an arbitrary “quality control window” (shown by the dotted oval region for illustration purposes). This window can be used in determining if a cream cheese has 73 9000 e Kraft Philadelphia 8000 — [:1 Store Brand 7000 — .4 Bruegger’s 6000 - E g. 5000 ' i I a y = -565x + 4700 a ‘ r \ x T R2 = 0.13 g 4000 ~ \ ‘ \ \ \ s ‘\\-t 8 3000 -- ff: ,\\ T \. 2000 . T 1% T 1000 ~ 0 . 0.0 1.0 2.0 3.0 4.0 Yield Strain (-) 5.0 Figure 4.22. Texture map for the controlled stress method showing yield stress versus yield strain for all cream cheeses at 5°C. 74 Yield Stress (Pa) 3000 FT 7 <> Kraft Philadelphia 2500 - . l T . 1:1 Store Brand : Li: 1' i it. Bruegger's 2000 - %f‘i—%—* " h 1500 - y = 46.8x +1627 2 : . T+ R 0.003 I A l 1000 - {1: 500 - 0 . 1 i . 0.0 1.0 2.0 3.0 4.0 5.0 Yield Strain (-) Figure 4.23. Texture map for the controlled stress method showing yield stress versus yield strain for all cream cheeses at 22°C. 75 8000 k TEST] .- eKr . . 7000 _ I aft Philadelphla \ FT . DStoreBrand 6000 - \ \ _ .:':': :-:-:-:-:-. ‘2 Bruegger's ,\ 5000 _ .ggjiz'flzhffizfiffizfiffi:fi; (6 III: ' - - - . . . . . . . a . . . . . 8 4000- ' U3 uni-PUD:-I~I'I'I'I~. E 23555353335533} 353333333333335333353533} ” 3009 "':':':':':':‘:‘: '2 $2 2‘:‘::::::::: ”=133i;;;;;;;:;;.;.;.;.;‘.;::"212‘y=-7098x+6765 R2=0.39 2000 _ . 0.0...I .' r I I fi l 1000 0 . r . 0 0.2 0.4 0.6 0.8 Apparent Yield Strain (rad) Figure 4.24. Texture map for the vane method showing yield stress versus apparent yield strain for all cream cheeses at 5°C. 76 6000 Yield Stress (Pa) <> Kraft Philadelphia 5000 ~ [:1 Store Brand A Bruegger's 4000 ~ 3000 2000 , 1000 1 0 0.2 0.4 0.6 0.8 Apparent Yield Strain (rad) Figure 4.25. Texture map for the vane method showing yield stress versus apparent yield strain for all cream cheeses at 22°C. 77 desirable spreadability. The windows are shown only as suggestions for how the texture maps can be utilized, not to define specific values for yield stress and yield strain values. Windows and trend lines do not include the results from the Kraft Philadelphia Fat Free cream cheese because this material exhibited large degrees of fracturing at the yield stress, rather than flowing easily. This could lead to atypical results when comparing this variety to other cream cheeses that do not fracture at the yield stress point. The whipped cream cheese results were also omitted because they do not represent typical cream cheese. For the controlled stress method at 5°C (Figure 4.21) the observed trend is similar to that of the vane method, but the fit of the trend. line is very poor (R2 < 0.13). At 22°C, the results from the controlled stress method display no apparent trend, as the fit of the trend line is almost zero. These results indicate that a texture map of spreadablity based on the data from the controlled stress method would not be productive. Variability in the controlled stress results can be explained by the varying degrees of stress the cream cheese is subjected to prior to achieving the yield stress. When loading the samples, they undergo different amounts of deformation depending on the sample size, and during the test are continuously deformed up to the yield stress. These factors, along with the higher cost and longer analysis time, make the controlled stress method an undesirable technique to evaluate yield stress when compared to the controlled rate vane method. For these reasons, only the vane method results will be discussed in the following section. 78 4.5. Fat Level and Yield Stress While there is no direct correlation between the percentage of fat contained in cream cheese and the yield stress, there are useful trends which emerge from this study. For all brands of cream cheese tested, the regular fat level cream cheese is less spreadable than its reduced fat counterparts (Neufchatel and light varieties) at 5°C except for Kraft Philadelphia varieties, where the regular and light cream cheeses are not significantly different. Borwankar et a1. (1992) found that products with the same amount of fat can Show very different rheological behavior. This was found to be particularly true for margarines, and was due to the emulsion characteristics of the products. The rheological properties of lipid based spreadable products are determined by the amount of fat the product contains in crystal and liquid form, not simply the total amount of fat present (DeMan, 1990). These concepts can be applied to cream cheese, except instead of looking at emulsion characteristics and phase properties, the influence of stabilizers and texture modifiers is the important factor. Changes in the rheological properties of regular cream cheese and Neufchatel cheese are the result of different processing conditions, since the ingredients are essentially identical and the addition of large amounts of stabilizers is not allowed. In light and fat free varieties of cream cheese, the rheological properties are largely influence by the amounts and types of stabilizers added to produce a reduced fat product with Similar textural properties to a full fat product. Overall, the trend is for the light cream cheese to have better spreadability than the 79 full fat versions. Cream cheese manufacturers have succeeded in finding the correct stabilizers allowing them to provide a reduced fat product with similar or superior spreadability. The fat free varieties of Kraft Philadelphia and Store Brand cream cheese Show completely different results. The Kraft Philadelphia fat free cream cheese was found to have the worst spreadablity of all Kraft varieties at both temperatures (Tables 4.1 and 4.3). This fact can be attributed to the tendency of the fat free samples to fracture at the yield stress point rather than begin to flow. Kawanari et al. (1981) found that in studying the texture of butter, the flow properties obtained. for fractured products are not good predictors of sensory spreadability. The large degree of fracturing in this product caused results that are not representative of the poor spreadability a consumer would perceive. The Store Brand variety did not exhibit this severe fracturing behavior, and overall had much better spreadability than the Kraft product. This Shows the importance of stabilizer types and quantity in a fat free product. While both products have similar ingredients, slight changes in these texture modifiers can have a large influence on the rheological properties of the final product. 80 Chapter 5 Conclusions and Recommendations 5.1. Summary and Conclusions AS an indicator of spreadability, the controlled stress and vane methods (single point and slope) were utilized to determine the yield stress and yield strain of various cream cheeses with different fat levels. The current work suggests the vane method is a superior technique for determining these properties. Controlled rate vane tests can be carried out more quickly, more economically, and with minimal sample disturbance prior to testing. The controlled stress method simply is not practical for characterizing yield stress, when such a superior method is available. The influence of temperature on spreadability was easily quantified by both methods, with results at 5°C being from 1.5 to 3 times those at 22°C. The exception to this trend was the fat free samples, which had similar results at both temperatures. There was no trend in the data showing the influence of the level of fat in the samples. Fat free samples tended to have higher yield stresses, while also exhibiting the highest degree of fracturing. The single point vane method data was used to construct “texture maps of spreadability” that can be used in a quality control or product development setting. These “maps” can quickly Show if product spreadability is within an acceptable range. Cream cheese yield stress data at 5°C (Figure 4.23) indicates that 81 acceptable yield stress values range from approximately 2000 Pa to 5500 Pa, while apparent yield strain values range from 0.2 rad to 0.6 rad. At 22°C (Figure 4.24) the data points move towards the origin because all of the samples are more spreadable. For routine testing of cream cheese, the following is recommended: use the controlled rate vane method at a rotational Speed of 0.5 rpm, with a 10 mm by 25 mm vane in an instrument having a torque capacity of at least 3 N cm. Any controlled rate rotational viscometer can be set up for this test. The size vane shown provides for the largest surface area possible (given the size of most commercial cream cheese containers) for testing without exceeding maximum instrument torque capacity. 5.2. Recommendations for future research The following would be excellent topics for future research and study: 1. Perform thorough sensory panel analyses on cream cheese samples and correlate results with rheological measurements, particularly the vane method texture maps. 2. Formulate and manufacture cream cheeses to evaluate how different manufacturing techniques and ingredients influence the spreadability of the product as determined by the vane method. 82 3. Apply the vane method to other spreadable foods. Find optimal vane sizes for different products and instruments. Establish benchmark values for “good” spreadability for each product using texture maps. 4. Construct texture maps from vane method data at various temperatures to predict spreadability at any temperature. 83 APPENDIX Table A.1. Controlled stress and single point vane method data for Kraft Philadelphia regular cream cheese. Vane Method (Single point) Controlled Stress Method 5°C 22°C 5°C 22°C 70w 0'0 Yo,v 0.0 Yo,s OIo 70,5 C3'0 Repetition (rad) (Pa) (rad) (Pa) (-) (Pa) Q) (Pa) 1 0.1369 4158 0.3104 1719 0.658 4806 2.055 1320 2 0.2166 4163 0.2339 1742 0.636 4456 2.202 1287 3 0.2297 4208 0.1305 1907 0.771 4230 2.579 1331 4 0.2475 4528 0.1678 1710 0.733 4706 2.599 1286 5 0.1662 4249 0.3807 1687 0.943 4031 1.532 1320 6 0.2449 5503 0.2894 1948 0.868 4462 2.060 1094 7 0.2040 5460 0.1683 2005 0.674 4536 2.626 1058 8 0.2380 4507 0.2800 2106 1.106 4400 1.585 1245 9 0.2779 4402 0.2496 1974 0.926 4284 2.532 1291 10 0.2317 4278 0.1610 2170 0.816 4656 1.930 1290 11 0.2522 4354 0.1673 1791 0.606 5769 2.137 1066 12 0.2784 4134 0.1893 1885 0.524 5534 1.691 1068 13 0.2842 4022 0.1893 1845 0.603 4774 1.851 1087 14 0.2716 3411 0.2984 2200 0.643 4762 2.115 1091 15 0.3015 4726 0.2155 1669 0.635 5169 1.716 1084 Average 0.2390 4407 0.2288 1891 0.743 4705 2.081 1195 St. Dev. 0.0447 525 0.0707 175 0.160 475 0.373 115 85 Table A.2. Controlled stress and single point vane method data for Kraft Philadelphia Neufchatel cheese. Vane Method (Single point) Controlled Stress Method 5°C 22°C 5°C 22°C Yo,v O’0 Yo,v O'0 Yo,s 60 1,0,5 0'0 Repetition (rad) (Pa) (rad) (Pa) (-) (Pa) (-) (Pa) 1 0.5503 3238 0.6210 1164 3.387 3763 3.828 1893 2 0.5086 3335 0.5964 1164 3.184 3700 3.459 1734 3 0.4856 3153 0.6634 1237 2.072 3790 3.255 1715 4 0.4735 3112 0.6775 1237 2.784 3625 2.985 1707 5 0.4221 2971 0.6142 1352 1.691 3681 2.766 1705 6 0.5149 2760 0.6493 1196 3.470 3521 3.635 1632 7 0.4840 2562 0.6624 1194 3.540 3359 3.686 1633 8 0.5246 2564 0.6299 1215 2.818 3737 3.287 1653 9 0.4898 2363 0.6288 1184 3.233 3676 2.806 1971 10 0.5529 2256 0.7273 1203 3.703 3711 2.762 1994 11 0.2957 2893 0.5686 1229 2.730 3659 3.655 1908 12 0.2816 2362 0.5233 1280 4.267 3448 3.097 1839 13 0.2978 2748 0.5676 1192 3.443 3493 2.090 1881 14 0.3130 2491 0.4724 1496 2.154 3790 2.162 1836 15 0.2684 2048 0.6058 1377 3.442 3720 2.070 1878 Average 0.4309 2724 0.6139 1248 3.061 3645 2.993 1799 St. Dev. 0.107 388 0.0634 92.8 0.688 131 0.565 122 86 Table A.3. Controlled stress and single point vane method data for Kraft Philadelphia light cream cheese. Vane Method (Single point) 5°C 22°C Controlled Stress Method 5°C 22°C Yo,v O‘0 Vow 60 Yo,s Go Yo,s 0'0 Repetition (rad) (Pa) (rad) (Pa) - (Pa) - (Pa) 1 0.1846 5392 0.1672 1933 1.061 3037 1.663 1818 2 0.2784 4443 0.2957 1666 1.082 3090 1.493 1794 3 0.2784 4813 0.2548 1836 1.077 3201 0.263 1841 4 0.2234 4854 0.2082 2001 1.152 2867 2.912 1782 5 0.1919 5177 0.2066 1904 1.423 2661 1.320 1975 6 0.1683 5197 0.2695 1962 1.600 2968 1.332 1854 7 0.1694 5668 0.2266 2132 0.847 3032 1.347 1864 8 0.2187 5551 0.3755 1619 1.651 2775 1.472 1838 9 0.2334 5278 0.2795 1590 0.891 3021 1.448 1980 10 0.1720 5565 0.3466 1576 1.121 3172 1.367 2040 11 0.2915 4457 0.4011 1491 1.766 3016 1.116 2032 12 0.2674 4373 0.3435 1369 1.192 3060 1.214 2190 13 0.1804 4270 0.3660 1699 1.264 2526 1.138 2012 14 0.2674 5009 0.3455 1395 1.647 2577 1.757 1907 15 0.2501 4354 0.3938 £124 1.044 2990 1.760 1723 Average 0.2250 5023 0.2987 1709 1.25 2933 1.44 1910 St. Dev. 0.0450 467 0.0750 239 0.292 208 0.541 125 87 Table A.4. Controlled stress and single point vane method data for Kraft Philadelphia fat free cream cheese. Vane Method (Single point) Controlled Stress Method 5°C 22°C 5°C 22°C 70w 0'0 70w 0'0 70,5 0'0 Yo,s Go Repetition (rad) (Pa) (rad) (Pa) (-) (Pa) (-) (Pa) 1 0.1363 5924 0.1672 4689 1.513 3342 2.091 1375 2 0.1196 5506 0.1599 4816 1.370 2958 3.987 1757 3 0.1515 5195 0.1741 5045 1.268 3231 4.442 1697 4 0.1211 6126 0.1222 5198 1.794 3020 4.476 2060 5 0.2171 6708 0.1825 6156 1.384 2877 2.772 1899 6 0.1825 6014 0.2160 5192 1.337 3235 2.565 1753 7 0.1814 6341 0.2323 5579 1.324 3325 3.300 1801 8 0.1505 7057 0.1515 5677 1.443 3094 3.994 1669 9 0.1672 6496 0.1368 5713 1.516 3080 3.160 1667 10 0.2475 7606 0.1814 4510 1.395 3107 3.831 1832 11 0.2323 6907 0.2344 4974 1.195 3416 3.660 1981 12 0.1363 7558 0.2197 5180 1.488 3129 2.102 1865 13 0.2018 7129 0.2517 4904 1.339 3104 1.875 2047 14 0.2354 6935 0.2344 5000 1.343 3432 2.305 1706 15 0.2459 7298 0.2485 4527 1.498 3054 2.810 1606 Average 0.1818 5892 0.1942 5181 1.414 3160 3.158 1781 St. Dev. 0.0458 582 0.0423 580 0.140 165 0.876 179 88 Table A.5. Controlled stress and single point vane method data for Kraft Philadelphia whipped cream cheese. Vane Method (Single point) Controlled Stress Method 5°C 22°C 5°C 22°C Yo,v 00 Yo,v 60 Yms 0.0 Yo,s Go Repetition (rad) (Pa) (rad) (Pa) (-) (Pa) (-) (Pa) 1 0.5508 1507 0.6660 437.8 2.722 1743 1.713 990.3 2 0.4824 1243 0.5519 452.5 4.208 1999 1.357 998.6 3 0.5383 1213 0.6314 402.5 4.393 2012 2.427 728.9 4 0.5550 1111 0.6529 426.0 3.237 2018 1.582 884.5 5 0.4850 1522 0.5843 417.2 2.735 2085 1.485 958.8 6 0.6440 1411 0.6336 499.0 3.114 2090 1.662 959.1 7 0.5985 1318 0.5055 494.2 3.137 2215 1.585 884.7 8 0.7273 1502 0.6603 405.8 2.518 2410 1.564 763.6 9 0.7053 1354 0.6372 453.6 1.955 2172 1.543 912.5 10 0.7069 1239 0.5404 448.8 3.114 2051 1.606 924.6 11 0.7718 1173 0.4730 501.0 2.193 2083 1.069 873.4 12 0.8090 1164 0.4866 549.9 2.402 2064 1.998 848.0 13 0.5953 1095 0.5906 457.4 2.968 2172 1.563 871.5 14 0.5154 1096 0.6131 469.7 2.737 2044 1.664 954.0 15 0.7535 1496 0.6032 501.0 2.193 1934 1.311 918.6 Average 0.6292 1296 0.5887 461.1 2.908 2073 1.609 898.1 St. Dev. 0.109 159 0.0636 41.6 0.685 145 0.305 76.2 Table A.6. Controlled stress and single point vane method data for Store Brand regular cream cheese. Vane Method (Single point) Controlled Stress Method 5°C 22°C 5°C 22°C Yo,v 0'0 Vow 0'0 Yo,s 0'0 Yo,s O'0 Repetition (rad) (Pa) (rad) (Pa) (-) (Pa) - (Pa) 1 0.5969 4490 0.4431 1933 0.885 7456 2.215 2107 2 0.6011 4584 0.3403 1745 1.492 6419 1.742 2163 3 0.6131 4557 0.4856 1942 1.291 7009 1.788 2216 4 0.5545 4230 0.3492 1816 1.092 8189 1.828 1928 5 0.5335 4769 0.2805 1845 1.018 7045 1.462 2247 6 0.1846 4352 0.2690 1886 0.925 8446 1.561 2331 7 0.6461 4574 0.2947 2189 1.220 8249 1.875 2078 8 0.5351 4751 0.3319 1981 1.538 7442 1.943 2164 9 0.6095 4732 0.3109 2141 1.593 6923 2.045 1957 10 0.3429 4727 0.2643 1853 1.450 7414 2.055 2242 11 0.4845 4581 0.4053 1564 1.815 6969 2.473 2069 12 0.1856 4780 0.4258 1770 1.218 8114 2.504 2154 13 0.1861 4574 0.4310 1929 1.586 6644 1.825 2305 14 0.4845 4349 0.3728 1753 1.442 6659 0.997 2424 15 0.3566 4289 0.2805 1781 1.358 6362 0.836 2973 Average 0.4610 4556 0.3523 1875 1.323 7289 1.810 2224 St. Dev. 0.167 182 0.0714 157 0.269 688 0.464 246 90 Table A.7. Controlled stress and single point vane method data for Store Brand Neufchatel cheese. Vane Method (Single point) Controlled Stress Method 5°C 22°C 5°C 22°C Yo,v O'0 Yo,v Co Yo,s 0'0 Yo,s O.0 Repetition (rad) (Pa) (rad) (Pa) (-) (Pa) (-) (Pa) 1 0.4216 2944 0.4284 1416 1.649 3964 1.789 2160 2 0.4079 3444 0.2837 1551 1.367 4470 2.604 2165 3 0.4562 3200 0.3928 1490 1.206 4095 2.3 86 2062 4 0.5013 3203 0.3691 1484 1.138 4400 2.177 1950 5 0.4897 3306 0.3010 1548 1.841 4414 1.765 1898 6 0.5377 3275 0.3282 1521 1.824 5203 1.818 2133 7 0.6152 3352 0.2894 1568 1.665 5064 1.543 2026 8 0.5639 3252 0.2952 1540 1.025 5052 1.790 2133 9 0.5964 3154 0.2994 1542 1.723 4411 1.651 2234 10 0.4751 3192 0.2826 1542 1.141 5287 3.201 2181 11 0.3917 3303 0.4226 1309 1.861 5293 2.087 2005 12 0.2921 3532 0.3241 1145 1.203 5135 1.759 1995 13 0.3801 3516 0.3182 1463 1.470 4651 2.164 2069 14 0.3550 3378 0.4085 1367 3.127 4592 1.846 2008 15 0.3581 3319 0.3907 1440 1.082 5374 2.112 1786 Average 0.4561 3291 0.3356 1462 1.555 4760 2.047 2054 St. Dev. 0.0952 149 0.0533 115 0.528 465 0.430 120 91 Table A8 Controlled stress and single point vane method data for Store Brand fat free cream cheese. Vane Method (Single point) Controlled Stress Method 5°C 22°C 5°C 22°C Yo,v Go Yo.v O'o Yo,s OIo Yo,s 0'0 Repetition (rad) (Pa) (rad) (Pa) (-) (Pa) (-) (Pa) 1 0.3549 3708 0.3770 2915 3.032 3051 1.562 2348 2 0.3770 4046 0.3241 3112 4.910 3001 1.443 2757 3 0.2952 3896 0.4310 2859 3.424 3134 2.533 2275 4 0.3120 3687 0.3592 3018 3.122 2809 1.301 2470 5 0.3125 3743 0.3471 3091 3.280 3097 2.194 2030 6 0.3451 4300 0.3550 2882 3.348 3141 2.678 2148 7 0.3728 4331 0.3487 2893 3.538 3002 1.451 2272 8 0.4037 4180 0.3917 2910 3.584 2992 1.336 2367 9 0.3445 4209 0.3907 2846 2.091 3247 1.374 2203 10 0.4226 4082 0.3728 2762 3.254 2948 1.843 2372 11 0.3938 4242 0.4221 2961 4.812 2610 1.348 2464 12 0.3744 5166 0.5341 2950 6.138 2712 1.612 2351 13 0.4289 4846 0.4037 2655 3.756 2704 1.264 2879 14 0.4069 5119 0.3906 2737 4.089 3114 1.404 2888 15 0.4530 4541 0.3587 2643 4.248 2523 1.526 2498 Average 0.3732 4273 0.3871 2882 3.775 2939 1.658 2409 St. Dev. 0.0462 471 0.0500 140 0.964 217 0.453 238 92 Table A9 Controlled stress and single point vane method data for Bruegger’s regular cream cheese. Vane Method (Single point) Controlled Stress Method 5°C 22°C ° 22°C Yo,v 0'0 Yaw 0'0 Yo,s 00 10,5 0'0 Repetition (rad) (Pa) (rad) (Pa) (-) (Pa) (-) (Pa) 1 0.2302 7607 0.2627 1540 1.236 5032 1.662 1792 2 0.3597 7061 0.2354 1519 1.267 4559 1.513 1552 3 0.3611 6526 0.2952 1554 1.381 5499 1.614 1808 4 0.2868 6185 0.2942 1601 0.931 4824 1.868 2022 5 0.3670 6209 0.2805 1543 1.115 5176 1.453 1797 6 0.2207 7047 0.3597 1683 0.964 5100 1.719 1663 7 0.2963 7274 0.2459 1552 0.651 5248 1.811 1899 8 0.2983 7090 0.3629 1590 0.643 5126 1.903 1944 9 0.1830 6897 0.2160 1511 0.816 4859 1.381 1861 10 0.2349 6856 0.3581 1674 1.234 5655 1.661 2010 11 0.2978 5324 0.2533 1393 0.711 5709 1.630 1750 12 0.1395 5491 0.2323 1410 0.742 5626 1.021 1999 13 0.2381 5606 0.2910 1352 1.129 5331 1.902 1964 14 0.2684 5433 0.2465 1422 0.561 5294 1.365 1903 15 0.2281 5988 0.3749 1428 0.822 6007 1.506 1983 Average 0.2673 6270 0.2872 1518 0.947 5270 1.601 1863 St. Dev. 0.0659 875 0.0533 99.8 0.264 383 0.239 137 93 Table A.10. Controlled stress and single point vane method data for Bruegger’s light cream cheese. Vane Method (Single point) Controlled Stress Method 5°C 22°C ° 22°C Yo,v O‘0 , Vow 60 70,5 00 1,0,5 60 Repetition (rad) (Pa) (rad) (Pa) (-) (Pa) (-) (Pa) 1 0.2905 4352 0.3901 1211 1.979 2106 1.375 952.8 2 0.3141 3908 0.3440 1208 2.044 2157 1.156 975.5 3 0.2790 4069 0.3907 1108 2.506 2320 1.215 1115 4 0.3618 3908 0.3602 1343 1.463 2123 1.090 987.9 5 0.2480 3852 0.4206 1196 2.748 2406 1.111 994.7 6 0.3335 3834 0.4316 1369 1.627 1916 1.245 1035 7 0.2805 3373 0.4698 1519 2.794 2021 1.200 998.2 8 0.3288 3297 0.4221 1316 2.252 2010 1.297 969.0 9 0.2978 3111 0.3775 1355 2.553 1940 1.573 1036 10 0.4221 2982 0.4724 1296 1.670 2235 2.712 1225 11 0.3570 4216 0.4541 1060 1.989 2519 2.318 1185 12 0.4158 4613 0.4247 1132 1.648 2493 2.523 1261 13 0.3 807 4152 0.4939 1062 2.231 1944 2.069 1150 14 0.4200 3479 0.4688 1098 1.862 1887 2.152 1260 15 0.4090 4413 0.3451 1103 2.682 1892 2.057 1102 Average 0.3426 4018 0.4177 1225 2.137 2131 1.673 1083 St. Dev. 0.0578 204 0.0484 136 0.442 218 0.569 110 94 LIST OF REFERENCES m,— .13.; V ——____d LIST OF REFERENCES Alderman, G.H., Meeten, G.H., and Sherwood, J .D. 1991. Vane rheometry of bentonite gels. Journal of Non-Newtonian Fluid Mechanics 39: 291-310. Barnes, HA. 1999. The yield stress — a review of “Iron/tor p81” — everything flows? Journal of Non-Newtonian Fluid Mechanics 81: 133-178. Barnes, HA. and Walters, K. 1985. The yield stress myth? Rheologica Acta 24: 323-326. Borwankar, R.P., Frye, L.A., Blaurock, A.E., and Sasevich, F .J . 1992. Rheological characterization of melting of margarines and tablespreads. Journal of Food Engineering 16: 55-74. Briggs, J .L., Steffe, J .F., and Ustunol, Z. 1996. Vane method to evaluate the yield stress of frozen ice cream. Journal of Dairy Science 79: 527-531. Campanella, OH, and Peleg, M. 1987. Determination of the yield stress of semi- liquid foods from squeezing flow data. Journal of Food Science 52: 214- 215, 217. Cheng, D. C-H. 1986. Yield stress: a time-dependent property and how to measure it. Rheologica Acta 25: 542-554. Daubert, C.R., Tkachuk, J .A., and Truong, V.D. 1998. Quantitative measurement of food spreadability using the vane method. Journal of Texture Studies 29: 427-435. DeMan, L., Postmus, E., and DeMan, J .M. 1990. Textural and physical properties of North American stick margarines. Journal of the American Oil Chemists Society 67: 323-328. DeMan, L. Shen, CF, and DeMan, J .M. 1991. Composition, physical and textural properties of soft (tub) margarines. Journal of the American Oil Chemists Society 68: 70-73. Dixon, ED. and Parekh, J .V. 1979. Use of the cone penetrometer for testing the firmness of butter. Journal of Texture Studies 10: 421-434. Elejalde, CC. and Kokini, J .L. 1992. The psychophysics of pouring, Spreading and in-mouth viscosity. Journal of Texture Studies 23: 315-336. 96 Fearon, AM. and Johnston, DE. 1989. A comparison of three instrumental techniques to evaluate butter spreadability. Journal of Food Quality 12: 23- 38. Hoffner, B., Gerhards, C, and Peleg, M. 1997. Imperfect lubricated squeezing flow viscometry for foods. Rheologica Acta 36: 686-693. James, A.E., Williams, D.J.A., and Williams, PR. 1987. Direct measurement of static yield properties of cohesive suspensions. Rheologica Acta 26: 43 7- 446. Kawanari, M., Hamann, D.D., Swartzel, K.R., and Hansen, AP. 1981. Rheological and texture studies of butter. Journal of Texture Studies 12: 483-505. Keentok, M. 1982. The measurement of the yield stress of liquids. Rheologica Acta 21: 325-332. Kokini, J.L. 1985. Fluid and semi-solid food texture and texture—taste interactions. Food Technology 39: 86-92. Kokini, J .L. and Dickie, A. 1982. A model of food spreadability from fluid mechanics. Journal of Texture Studies 13: 211-227. Korolczuk, J. and Mahaut, M. 1989. Viscometric studies on acid type cheese texture. Journal of Texture Studies 20: 169-178. Kosikowski, F .V. and Mistry, V.V. 1997. Bakers’, Neufchatel, cream, quark and ymer. Chapter 9. In: Cheese and Fermented Milk Foods, Volume 1, 3rd Edition. F.V. Kosikowski. Westport, Connecticut. Luciasano, M., Casiraghi, E., and Pompei, C. 1989. Optimization of an instrumental method for the evaluation of spreadability. Journal of Texture Studies 20: 301-315. Massaguer, S., Rizvi, SSH, and Kosikowski, F.V. 1984. Characterization of thixotropic behavior of soft cheeses. Journal of Food Science 49: 668-670, 684. Missaire, F ., Qiu, CG, and Rao, MA. 1990. Yield stress of structured and unstructured food suspensions. Journal of Texture Studies 21: 479-490. 97 Mortensen, B.K. and Danmark, H. 1982. Consistency characteristics of butter. Milchwissenschaft 27: 530-532. Muengo-Gnanasekharan, K.F. and Resurreccion, A.V.A. 1993. Physicochemical and sensory characteristics of peanut paste by processing conditions. Journal of Food Processing and Preservation 17: 321-336. Nguyen, OD. and Boger, D.V. 1983. Yield stress measurement for concentrated suspensions. Journal of Rheology 27: 321-349. Nguyen, OD. and Boger, D.V. 1984. Direct yield stress measurement with the method. Journal of Rheology 29: 335-347. Ofoli, R.Y., Morgan, R.G., and Steffe, J .F. 1987 . A generalized rheological model of inelastic fluid foods. Journal of Texture Studies 18: 213—230. Omar, Z.B., Raphaelides, S., and Kestleloot, R. 1995. Texture evaluation of French acid-type cheese. Journal of Texture Studies 26: 325-338. Prentice, J .H. 1972. Rheology and. texture of dairy products. Journal of Texture Studies 3: 415-458. Rohm, H. 1990. Magnitude estimation of butter spreadability by untrained panelists. Lebensmittel Wissenschaft und Technologie 23: 550-552. Rohm, H. and Raaber, S. 1991. Hedonic spreadability optima of selected edible fats. Journal of Texture Studies 6: 81-88. Rohm, H. and Ulberth, F. 1989. Use of magnitude estimation in sensory texture analysis of butter. Journal of Texture Studies 20: 409-418. Sanchez, C., Beauregard, J.L., Bimbenet, J.J., and Hardy, J. 1996. Flow properties, firmness and stability of double cream cheese containing whey protein concentrate. Journal of Food Science 61: 840-843, 846. Sanchez, C., Beauregard, J .L., Bride, M., Buchheim, W., and Hardy, J. 1996. Rheological and microstructural characterization of double cream cheese. Nahrung 40: 108-116. Scott-Blair, G. 1958. Rheology in food research. Advances in Food Research 8: 1-56. 98 Shukla, A., Rizvi, SSH, and Bartsch, J .A. 1995. Rheological characterization of butter using lubricated squeezing flow. Journal of Texture Studies 26: 313- 323. Steffe, J.F. 1992. Yield stress: phenomena and measurement. Chapter 29. In: Advances in Food Engineering. Singh, RP. and Wirakartakusumah, eds. CRC Press. Boca Raton, Florida. Steffe, J .F. 1996. Rheological Methods in Food Process Engineering, 2nd Edition. Freeman Press. East Lansing, Michigan. Usiak Godfrey, A.M., Boume, M.C., and Rao, MA. 1995. Blanch temperature / time effects on rheological properties of applesauce. Journal of Food Science 60: 1289-1291. Yan, J. and James, AB. 1997. The yield surface of viscoelastic and plastic fluids in a vane viscometer. Journal of Non—Newtonian Fluid Mechanics. 70: 237-253. Yoo, B. and Rao, MA. 1995. Yield stress and relative viscosity of tomato concentrates: effect of total solids and finisher screen size. Journal of Food Science 60: 777-779, 785. Yoo, B., Rao, M.A., and Steffe, J .F . 1995. Yield stress of food dispersions with the vane method at controlled shear rate and controlled shear stress. Journal of Texture Studies 26: 1-10. Yoshimura, A.S., Prud’Homme, R.K., Princen, H.M., and Kiss, AD. 1987. A comparison of techniques for measuring yield stress. Journal of Rheology 31: 699-710. 99