Novi Wijaya (A0068355X) Yong Wai Fen (A0068349R)

Mathematical Modelling and Simulation for the Cooking of Potato

By early 1990s, the petroleum and petrochemical industries have had computer simulators for their processes for years[1]. This means that the processes happening in the industries are well-understood. The same cannot be said for the food industries; a major difficulty with food systems is the lack of mathematical models to describe mass and energy balances in the unit operations.

There had been many papers published in the field of food processing, particularly the potato processing. Potato may have caused many interests because it is globally consumed, cheap and easily available in various sizes. Potato have a water content of about 80 wt%, a starch content of 16 wt% and cellulosic material content of 4 wt%[2]. Starch is a long chain of sugar molecules called a polysaccharide[3] and it composes of two polymeric units, i.e. amylase (a linear polymer) and amylopectin (a highlybranched polymer). The size of each amylopectin group is about 9 to 10 nm. The amylose fraction exists naturally and is randomly interspersed among the amylopectin molecules in both the amorphous and crystalline regions of starch. Thus, small amylose molecules located at the periphery are free to leach out of the granule and undergo gelatinization. Two obvious changes occur in potato tissue during cooking: the starch is gelatinized which occurs at 65 0C and, owing to a marked decrease in cell adhesion, the tissue is softened so that it can be readily mashed. These two alterations are those prerequisite to the conversion of raw tissue into a product having the physical qualities associated with cooked potato[4].

The cooking of potato involves heat transfer and change in its material content due to gelatinization. In this assignment we will model the heat transfer and the gelatinization in a potato, and simulate the cooking of a potato based on the model.

Kozempel, M., Craig, J. C., Jr., Sullivan, J. F. and Damert, W. Computer simulation of potato processing. Biotechnology Progress, 4, 63-67 (1988). 2 Chen, X.D. Cooking Potatoes Experimentation and Mathematical Modeling. 2002. 3 P. Barham. The science of cooking. 2001. 4 Personius, C. J. and P. F. Sharp, 1939. Simulation, by chemical agents, of cooking of potato tissue, Food Res. 4. 469.

Novi Wijaya (A0068355X) Yong Wai Fen (A0068349R)

To be able to derive a first principle mathematical model, we made several assumptions. The assumptions made are: 1. The potatoes have spherical shapes. 2. The potatoes are homogenous throughout. 3. Cooking corresponds to the gelatinization. 4. The gelatinization moves on a uniform front. 5. Little swelling occurs during the gelatinization process and we therefore assumes a constant radius while cooking. 6. The boundary between gelatinized to ungelatinized starch is relatively small compare to the size of the potato. 7. We conclude that the potato is cooked based on the watermark dividing the cooked and uncooked regions. 8. At temperature above 65oC, the gelatinization occurs at its maximum rate. 9. Since the gelatinization is very fast, the heat transfer becomes the limiting step. Hence the rate of cooking is determined by the rate at which heat arrives at the cooking interface. 10. Constant driving force for heat transfer. 11. All the heat conducted through the shell to the interface is consumed by the interface cooking reaction.

For this simulation we will use a laptop computer with MATLAB, to solve the partial differential equations in our model.

Novi Wijaya (A0068355X) Yong Wai Fen (A0068349R)

First Principle Model The equation of energy in terms of q, for spherical systems is:

For one-dimensional, conductive heat transfer, however, the equation above can be simplified into: ( )

Fouriers law of heat conduction for one dimensional heat transfer in a spherical coordinate:

The equation of heat transfer can be written as: ( )

The properties for potato is taken from Food Properties Handbook, and5, and is shown in Table 1 below. TABLE 1
Properties of Potatoes

cp k starch

(J/kg.K) : (W/m.K) : (kg/m3) :

3500 0.624 + 1.19x10 T (oC) 1500

It is also stated in literature that a potato consists of 20% starch and 80% water.

At the surface, convective heat transfer takes place: ( )

Chen 6 observed that the convection is strong at the boundary, hence the boundary temperature can be assumed to be constant, i.e.:

5 6

Rahman, S., Food Properties Handbook, CRC Press, Boca Raton, FL (1995). Chen, X. D. Cooking potatoes: Experimentation and mathematical modeling. Chemical Engineering Education, Winter 2002.

Novi Wijaya (A0068355X) Yong Wai Fen (A0068349R)

The change in temperature will stop at the center, hence:

The initial condition for the potato is: ( ) We would assume that the initial temperature of the potato is 20oC. According to Chen, the reacting interface temperature is 65oC. So when the temperature of the potato reaches 65oC, the gelatinization occurs so fast that the rate of heat transfer becomes the limiting rate. Then we can simulate the heat profile with different cooking time. We use pdepe in MATLAB to solve our partial differential equation for T.

Novi Wijaya (A0068355X) Yong Wai Fen (A0068349R)

Experiment We next design an experiment that can be use to validate our model.

The objectives of the experiment are: To quantify the gelatinization of a potato visually by cooking potato samples at the same temperature for different intervals of time. The gelatinization front can be distinguished as a watermark between cooked and uncooked potato (Figure 1). However, this watermark may be hard to distinguish, so we propose the use of iodine to mark the barrier between the gelatinized and the ungelatinized potato. Iodine only stained starch purple, so only the ungelatinized potato will be colored purple by the iodine. We predicted that the gelatinization front will move closer to the center of the potato when time increasing.

Figure 1 Watermark between the cooked and uncooked regions of the potato after being heated in water for few minutes Materials used are: Spherical potatoes with a radius of 25 mm Iodine Distilled water, to avoid mineral interaction with the potato Latex gloves Vernier calipers

Novi Wijaya (A0068355X) Yong Wai Fen (A0068349R)

Experiment procedures: 1. Fill an aluminium tank with 50 % of distilled water. Attach heating element at both sides of the tank. Place small bucket into the tank and fill with distilled water until the water level in the bucket is greater than the water level in the tank. Turn on the heater and heat until 65oC (gelatinization start). 2. 3. 4. 5. Place the potato samples into the bucket. Remove potato samples at every interval of 10 minutes. Slice the potato samples through the circumference. Wear latex gloves and wet end of index finger with iodine. Rub the iodine onto the potato cross-sections. 6. 7. Place the potato for 3 minutes and rinse the potato with water. Take pictures of the potato cross-section and compare the distance of the iodine barrier mark between the different interval times. It will look like the illustration in Figure 2.

Figure 2 Illustration of the partially cooked potato

Novi Wijaya (A0068355X) Yong Wai Fen (A0068349R)

Simulation We will do several simulations to observe how our model behaves. Several variations we will use are: The length of cooking time The radius of potato The temperature of water used to cook the potato

Several things that we will observe: The temperature profile of the potato The radius of uncooked potato The cooking time

Temperature plot for a potato with a radius of 2.5 cm and cooking time 15 minutes (900 seconds) Figure 3 is the contour plot of cooking time, potato radius, and potato temperature.
The Cooking of Potato

380 360

Temperature (K)

340 320 300 280 1000 0.025 500 0.01 Time (s) 0 0.005 0 Distance from center (m) 0.02 0.015

Figure 3 The contour plot of a simulation of the cooking of a potato

Novi Wijaya (A0068355X) Yong Wai Fen (A0068349R)

Various length of cooking time Figure 4 shows how temperature changes throughout the potato, with different cooking time.
Temperature profile at different time 380 370 360 350 t=0.2 min t=3 min t=5 min t=6 min t=10 min t=15 min

Temperature (K)

340 330 320 310 300 290

0.005

0.01 0.015 Distance from center(m)

0.02

0.025

Figure 4 Temperature-distance profile at different cooking times (potato radius=2,5 cm and the initial temperature=20oC) Because we assume that gelatinization occurs at 65oC, then we can conclude that a potato is cooked once it reaches 65oC (or 338.15 K). Hence, from the data we obtained from MATLAB, we can plot a graph of uncooked radius versus time (Figure 5).
0.03 0.025 Uncooked radius (m) 0.02 0.015 0.01 0.005 0 0 2 4 Time (min) 6 8 10

Figure 5 Uncooked radius of the potato versus cooking time

Novi Wijaya (A0068355X) Yong Wai Fen (A0068349R)

Various radius of potato Figure 6 shows how temperature changes with time, in potatoes of different sizes.
Temperature profile at different radius 380 370 360 350 r=2.5 cm r=3 cm r=3.5 cm r=4 cm r=4.5 cm r=5 cm

Temperature (K)

340 330 320 310 300 290

500

1000 Time (s)

1500

2000

2500

Figure 6 Temperature-time profile at different potato radius (cooking time=35 minutes and the initial temperature=20oC)

From the data we obtained from MATLAB, we can plot a graph of cooking time versus the radius of potato (Figure 7).
6 5 Radius of potato (cm) 4 3 2 1 0 0 2 4 6 8 10 12 14 16 Cooking time (min)

Figure 7 How the radius of potato influences cooking time

Novi Wijaya (A0068355X) Yong Wai Fen (A0068349R)

The temperature of the water used to cook potato We used potatoes of radius 2.5 cm and cooking time is 10 minutes. The water used to cook potato is set to be 60oC, 70oC, 80oC, 90oC and 100oC. The temperature profile of the potato for each water temperature can be seen in Figure 8 below.
Temperature profile at different time 380 T=100 C T=90 C T=80 C T=70 C T=60 C

370

360

Temperature (K)

350

340

330

320

310

0.005

0.01 0.015 Distance from center(m)

0.02

0.025

Figure 8 Temperature-distance profile at different cooking water temperature (potato radius=2,5 cm and the initial temperature=20oC) From the simulations results, we can conclude several things: 1. To cook a potato with a radius of 2.5 cm in boiling water, we need about 9 minutes. 2. The time needed to cook a potato relates to the size of the potato; an increase in the size of potato will increase the time needed to cook the potato, roughly linearly. 3. To cook a potato with a radius of 2.5 cm in 10 minutes, the water temperature must be 90oC or higher. (Note: the first two simulations are written in pdepotato.m, while the last simulation is written as pdepotatoinwater.m)

Novi Wijaya (A0068355X) Yong Wai Fen (A0068349R)

Conclusion/Insight of the Problems We tried to derive a first principle mathematical model of potato cooking process. We found the thermal properties of potato (i.e. thermal conductivity and the specific heat capacity) and the physical properties of potato (composition and density) from papers and textbooks. Then we determine that the gelatinization temperature is 65oC, based on literature. From there we can calculate the cooking time of a potato, also the radius of gelatinization as a function of time.

We assume that the gelatinization front moved through potato in a uniform way. The longer the cooking time, the faster the gelatinization occurs and the farther the gelatinization front move from the boundary towards the center of the potato. This logic agrees with our model and simulations. So we can say that to some extent, our model successfully represents the physical phenomenon that happens inside a potato when it is cooked.

In our mathematical model, there is no free convection term; free convection term at high temperature can greatly influence the conduction of heat into a submerged sphere. Convection within a fluid increases at high temperature. This proportionally increases heat transfer in the potato. Our mathematical model can simulate the heat transfer due to convection. However, to get a more accurate representation of the phenomenon of starch gelatinization inside a potato, other variables need to be taken into account, such as water absorption and the hardness of the potato. The change of the hardness should be studied as well, as this is the result of starch gelatinization in excess water. In addition, our experiment procedure should able to define a good visual guide for the amount of starch gelatinization that occurs. One of them is, in order to have a better accuracy, the amount of the iodine used should be quantified.

In our project, we use pdepe in MATLAB to solve our mathematical model. Other than pdepe, we can also use the long and gruesome finite differential method such as Euler method. In the finite differential method, the sample is subdivided into elements of arbitrary size and shape. The temperature field at a given time in each element is expressed as a function of the temperatures at a small number of nodes at the corners

Novi Wijaya (A0068355X) Yong Wai Fen (A0068349R)

and edges of the element. Spatial discretization of the heat transfer equation will give a system of ordinary differential equations. From this system of ordinary differential equations, we can calculate the unknown temperatures at the nodes. A simple representation can be seen at Figure 9

Figure 9 Finite element grid of the potato sample to calculate starch gelatinization process.