The term ‘anomaly’ has been used in science to denote a phenomenon, which is completely out of the ordinary, and thus the initial observers do not immediately know its cause; often it takes the application of a combination of scientific theory and/or experimentation to find a probable explication. An example of this is ‘St. Elmo’s Fire’; a bluish glow, which appears on the ends of ships masts or airplane’s wings during storms. It wasn’t until the development of electrostatics that the cause of this was understood. (The very large charge in the storm clouds induces a similarly enormous opposite charge in the ground below. Because charge tend to accumulate on points, this charge is even greater in a ship’s mast (or other object). The potential difference between the two becomes enough to ionize the air around the object, and ionized air has a bluish tinge because this is the characteristic frequency of light given off by ionized oxygen/nitrogen, i.e. our atmosphere.)

This essay attempts to deal with a similar mystery, one which has only been noticeable for a short length of time, as the technology that gives rise to it did not exist in a form to allow notice of this phenomenon until this century. In 1940, Percy Spencer was walking by an experimental high frequency radio wave communications device when the chocolate bar in his pocket melted. Ten years later, microwave ovens came onto the market. Sometime after that, the following anomaly was observed. If a grape is placed in a microwave oven, sliced in two and spread (butterflied in such a way that the two hemispheres are directly beside each other and just touching, ideally with a small flap of skin still connecting them) and then the microwave is turned on, within seconds a spark/flame appears between the two halves. This seems bizarre, but has been observed by many different people and documented several places: on the Internet, on science question-and-answer periods on CBC, and probably by others, who haven’t bothered to publish their observations.

I have chosen to examine this occurrence for several reasons. The science show in question was not able to find a satisfactory answer to the question. In fact, no matter where I went or whom I asked, I have been unable to find an individual who could do more than hypothesize, and often not well, on the cause of these grapes’ behavior. Thus, it became somewhat of a personal challenge to be able to find a satisfactory explanation.

This essay consists of my attempt to treat this subject and is divided into the following three parts:

• A more detailed description of phenomenon itself, and observations made when the experimental design elaborated modified;

• A theory pertaining to why this behavior is observed, based on both observations and knowledge of the physics involved in the setup;

• An evaluation of the validity of this theory and of the limitations present in the experimentation, and recommendations for further experiments which could further clarify the origins of this phenomenon.

This phenomenon is something that I first encountered on a visit to my grandmother’s house in the spring of 1997. She had heard about on a show where people would phone in science questions and then a panel would try to answer them. Someone phoned in with this question, and the rest of the show was spent with both the panel and callers making hypotheses about why it was happening. This section of the essay details both the basic procedure used to make the sparks and every variation on this procedure which was tried in an attempt to limit the possible explanations. It will also discuss limitations in equipment, which are the main reason this essay can discuss only probable causes instead of offering a hard-fast answer to the research question.

A microwave with a stirrer (600W) Seedless green grapes

A microwave with a turntable (600W) Stainless steel knife

Seedless red grapes Ceramic plate

1. Cut grape in two equal halves, leaving the two halves attached by a small flap of skin. Open up so that they resemble figure 1, and place on the ceramic plate (it seems to be irrelevant whether or not the grapes are placed cut side down as in figure 1, or if they are placed cut side up.)
2. Place the grape halves inside a microwave oven. If it has a turntable, be sure to set them in the exact center so they move as little as possible. (The reasoning behind this placement will be explained in the next section.)
3. Turn the microwave on high for 10 seconds.
4. Record all observations. (Because you are microwaving such a small amount of matter, most of the microwaves will not be absorbed and there is a chance they can reflect back to the magnetron and damage it. If sparking does not result after 5-7 seconds, or if sparking ceases for any reason, abort the experiment to avoid more harm then is necessary to the microwave oven for the purposes of science.)
5. Repeat as many times as deemed necessary for uniformity of data.

After approximately 3-5 seconds of microwaving, a bright light and loud buzzing sound was observed at the junction between the grape halves, where they were connected by the flap of skin. The light seems to be primarily sparks but partly flame. After about 2 seconds of this, (7 seconds into the procedure) smoke started to be observed rising from the junction between the grapes (see figure 2). One of two things would happen next.

Either the juice of the grapes would boil over and extinguish the flames (this usually happened when the grapes were placed cut side down), or the skin bridge would break and the grapes would fall apart, also ending the sparks. (This happened when the grape was placed cut side up, but not if the grapes were placed so the leaned against each other instead of relying on the skin to hold them together. Thus, because the sparks lasted longer, most variations on this procedure were performed cut side up.) The post experimental grapes were singed along the edge where the skin bridge had been. Usually the skin had burnt off the surface of the grape 0.5 mm to 0.7 mm from the bridge. (See figure 3.) If the grapes were in a turntable microwave, the sparks were more constant then if they were in a stirrer microwave. There was no noticeable difference between the red and green grapes’ performances.

Cutting Skin Bridge:

It was found that if the skin bridge was not left (the grape was cut completely in half) and then the two halves were placed just touching each other, the sparks were not significantly affected. Observations were identical to those when the skin bridge was left.

Multiple Grapes:

1. If two grapes were placed in the microwave at the same time separated by several centimeters form each other, the effect was less intense for each grape, and the sparks occurred at different times for each pair. (i.e. 4 and 6 seconds into the microwaving period respectively)
2. If several grape halves were placed in a circuit around the center of the microwave, so all the grapes are touching, but half are connected by skin and half are not, the sparks occur equally at the junction of every grape. The amount of burning is equal at the end.

Removing Skin:

If the skin is removed and then the grape is microwaved as per the no skin bridge experiment, no visible burning or sparking occurs.

Substitution of Aluminum Foil for Grapes:

In step 1 of the procedure, two balls of aluminum foil were substituted for grapes, made into about the same size as the grapes and then placed close together. When microwaved, these balls of foil behaved very similarly to the grapes. After 3-5 seconds, sparks and slight flame appeared between the two foil balls. This was following by a slight smoke from their junction. When they were removed from the microwave and unfolded, two holes about 0.5 cm in diameter had been burnt into the aluminum foil where the balls were next to each other.

One of the most important procedures to prove the theory elaborated in the next section would be to be able to measure the potential difference between the grape halves in the two microwave types. However, this proved impossible to do for several reasons. First, getting leads for a voltmeter into the microwave without microwaving the voltmeter itself is difficult, as a safety feature of microwaves mean they will not operate without the door completely closed. Furthermore, microwaves consisting in part of magnetic fields tend to induct currents in metals, so the leads to the grape would be affected by this radiation and thus a current might occur even with no electrons moving in the grape. A further experiment that could prove my theory will be discussed in the evaluation section of this essay. In the meantime, a rather tenuous link had to be made between the grapes and the aluminum foil to try and find a value for the potential difference created by the microwave.

Since I started researching this subject, I’ve heard many theories about why it could be happening. In the beginning I registered this essay in chemistry, having heard so many stories about how it was the sugar burning, because sugar resonates with the frequency of the microwaves. Later, I received an interesting e-mail from someone in Austria who thought it was the grape acting as a satellite dish due to it’s parabolic shape, and this was causing them to catch the RF energy. After much deliberation and research into microwaves, I decided to switch into physics. My explanation of the observations in the previous section is thus based on a synthesis of 3 main areas of research/knowledge:

• The functioning of a microwave oven;
• The ability of microwaves to create a potential difference and the capability of grapes to act as a capacitor;
• The resistance in the skin flap causing flames.

We learned in the introduction that microwave ovens were first conceived of in the 1940s, after a United States military project using extremely short-length radio waves for communication found that the ‘microwaves’ had other potential applications.

Microwave ovens operate at a frequency of 2.54 GHz. This was chosen for several reasons. First of all, it gives them a wavelength of about 12 cm, which is long enough that they can easily be contained within the oven without fear of radiation into the surrounding environment (i.e. a kitchen). Secondly, it resonates mildly with water, but not enough that the water in the outer shell of food would absorb all the energy and leave the inside raw: it is absorbed enough that it heats polar molecules quickly, but penetrates enough to heat foods evenly.

Once the operating frequency was determined, it was necessary to find a way to distribute the microwaves evenly in the operating chamber. Because they are waves and are contained inside the oven, they will reflect and thus set up point of destructive and constructive interference. If this were not compensated for somehow, the placement of food in the microwave would determine whether it took 30 sec. or 30 min. to cook. Two main ways were devised to heat food evenly – one by moving the microwaves, one by moving the food. The first case involved the placement of metal fan blades between the magnetron (which emits the microwaves) and the chamber, scattering the radiation and makes the interference migrate throughout the chamber. The second involved the placement of a turntable in the microwave, to rotate the food though the areas of high and low concentration of microwave radiation. Very rarely is a turntable placed in a microwave with a fan as well.

The implications of these differences are twofold. First of all, they imply that if the sparks are not due merely to the heating of sugar but to electrodynamics of some variety, then there should be a difference between the sparking of grapes in the stirrer oven, compared to the ones in the turntable oven. (The microwaves in the stirrer oven are not constant; therefore the sparks they generate would be too. The grape placed at the center of the turntable, however, should be subject to constant radiation, as it’s turning in a region which has a constant intensity of radiation in it, and as its frequency of rotation, 1x10^-1 Hz, is negligible compared to the microwaves’ frequency, 2.541 x10⁹ Hz.) This difference was in fact noted during experimental procedures – this, (along with the presence of sparks in and of themselves) was the first sign that physics was primarily responsible for the light show being investigated. Secondly, this implied that the placement of the grapes in the turntable oven was important. Only if placed in the center would the grape stay in a region of constant microwave radiation instead of rotating though many different areas; this is why the procedure called for the grape to be placed there.

The next important step was to find out how the grapes sparked. Obviously, if there are sparks, there is an ionization of the air occurring due to a high potential difference, and the constant transfer of electrons, very much like what was going on in the example of St. Elmo’s Fire discussed in the introduction. Because of the constant buzz in the background during sparking, it should be obvious that the sparking was at a fairly high frequency. (If each separate spark generated a sound, for them to blend into a single constant noise it would be necessary that they be happening fairly often.)

This can only mean that the electrons are jumping back and forth between the grapes. (If they were only going one way, the spark would eventually be forced to go through the ceramic plate or the air to the metal side of the oven. Since this did not happen, we must presume that the charge is transferring from one grape half to the other then back again due to induction from the microwaves.)

If the grapes were aluminum foil balls, this would be a relatively easy thing to explain. As aluminum foil is a conductor, it allows easy movement of electrons across its surface (only the surface, as the microwaves don’t penetrate far. In fact, it is the very movement of electrons that allows metal to reflect microwaves. All the microwaves’ energy goes into moving the electrons near the surface of the metal, and since moving charges make electromagnetic waves, the metal reflects exactly the same wave as it received, (minus a slight amount of energy that goes into heat in the metal due to resistance.) But back to our story.) The microwaves cause the movement of electrons back and forth in the metal. However, assuming that the two metal balls are close enough together that they are in the same region of influence of the wave (and seeing as both foil balls in the experiment were less then 5 cm in total length, and microwaves have wavelengths of 11.8 cm, this would often be the case), then the charges in the balls would be pushed to the same side, say, right, or left, and there would be a potential difference between the two. (See figure 4.) If this difference were enough to cause the air between them to ionize and allow the electrons to flow from one ball to the other (to initiate sparkover) then a spark would occur. Then, when the electromagnetic field reversed itself, one of the balls would already be charged and a greater potential difference between the two balls would be easier to create, and sparkover would occur again. It is relatively easy to calculate whether or not the microwaves would have enough energy to make this happen. We can calculate the potential difference that must exist between the two balls to ionize the air by treating the foil balls in a similar way as we might treat parallel plate capacitors. Assuming that the balls are near circular, a small enough section of their areas will be nearly flat (just think of the earth for an example of this!). We could say, for instance, that a 0.5 cm² area on the side of each of the balls were 0.2 mm apart, and this would be a quite close approximation of the actual scenario. The capacitance of a parallel plate capacitor is given by the equation

C = Ke_oA/d

where the C is capacitance, K is the dielectric constant of the material in between the plates, e_o is the permittivity of free space, A is the area and d is the distance between the plates. Because the dielectric constant of air is 1.0006, we can find out the C of the balls: 1.11 x 10^-13 F. We also know that capacitance (C) is equal to the charge on the plates (Q) over the potential difference between them (V), so

1.11 x 10^-13 F = Q/V

Furthermore, we know that work (DE) is equal to VQ. Power (P) is equal to DE/DT. DT can be termed the time during which a certain direction of push will be felt by the charged particles, i.e. for half of the wavelength of the microwave, or it’s frequency inverted and halved. ((2.54 x 10⁹ GHz) ^–1/2, or 1.97 x 10^-10s.). This will allow us to calculate the maximum potential difference that could be produced by the microwave before the wave passed and the electromotive force started being in the opposite direction. By substituting work over potential difference into the capacitance equation above, we are left with the final equation:

C=Q/V

C=(DE/V)/V

C=DE/V²

C=PDT/V²

P=CV²/DT

V=(PDT/C)^1/2

We can now find the potential difference induced by the microwave oven over the time: it takes one half wavelength to pass by the balls, using the capacitance and time found above, and taking 600 W, the power output of the microwave ovens used for the experiments, as P. The value comes out to be:

V=(600 W x 1.97 x 10^-10s /1.11 x 10^-13 F)^-1

V= 1.03 x 10³ V

Now, this is all very well – what we have yet to do is to apply this information to grapes. Aluminum is a very conductive material. Its resistivity is on the order of 2.65 x 10^-8WM. Grapes, on the other hand, are a much less conductive material; consisting mostly of sugar and water, they have a resistivity somewhere in the same range as water itself (0.5 Wm - 300Wm.) However, if you take Ohm’s law:

V=IR

and substitute in the equation for resistivity (R=rL/A), assuming that we can treat the grape as a wire, (giving it a value of 1.5 cm for L and an average cross sectional area of about 0.7 cm² – these are not exact, but even given that they will vary ± 1 cm, it shouldn’t affect the calculations too much given the magnitude of V), we can calculate the approximate current which will run though the grapes:

V=I (rL/A)

I = VA/rL

I = (1.03 x 10³ V)(0.007 m)/ (300 Wm^*)(0.015 m)

I = 1.60 A

Now, we take this back into the capacitance equation derived for the aluminum foil balls. The equation should be about the same for the grape halves as for the aluminum foil, so we can say that

1.11 x 10^-13 F = Q/V

If Q/V for the grapes is bigger then the capacitance of the system, then the grapes should spark, and this theory of a cause for the grapes sparking is correct. We know that I = DQ/DT, so if we take DT to be the 1.97 x 10^-10s found above to be appropriate for this system, then we get

Q/V= IDT/V

Q/V=(1.60 A)(1.97 x 10^-10s)/(1.03 x 10³ V)

Q/V= 3.06 x 10^-13 F

Thus the grapes spark.

The obvious explanation for the flames which are observed along with the sparks is that the skin bridge at the beginning (during the first 5 seconds!) provides a pathway for the electrons to move though, allowing the microwaves to push electrons in the grape all the way to one side and all the way back, though the bridge, without having to cause sparks. However, because (as we can easily imagine!) the resistivity of the skin is quite high and because it is so thin, having 1.6 A of current pushed though it quickly heats it and causes it to burst into flames. This explains the lack of sparking at the beginning, and the presence of flames as well as of sparks. Once the gap between the grapes is burnt far enough apart, the capacitor effect no longer happens, causing the sparking to stop. Similarly, the overheating of the grapes can cause the resistivity of the water inside them to skyrocket, making it much harder for the microwaves to induct current, thus sparks are no longer observed.

This investigation seems to come together into a cohesive whole. The experiment identified several qualities of the phenomenon in question: the sparks, the flame, the buzzing. The theory that each separate spark is caused by charge separation inside the grape itself due to the large potential difference, and that the oscillating field caused by the microwaves makes the sparks leap back and forth between the grape halves seems to make sense. It also rationalizes the burning of the skin bridge early in the experimental procedure. The buzzing is explained by the rapid series of jumps made by the charge between the grapes.

However, there are some questions not addressed by this theory. If in fact this is all there is to this occurrence, why is it that when the grape was peeled and microwaved, it did not spark? Could it be that the burning of the skin is necessary for the sparks to manifest themselves and if that is the case, why did the aluminum foil spark? Perhaps it has something to do with the composition of the grape skins, so that they act like metal does and shield the grape, becoming charged on the outside, while the cut surface is the only part that absorbs radiation.

Furthermore, this procedure was unable to experimentally determine whether or not a potential difference existed between the grape halves by any method other then theoretically, and generalizations were necessary. Further experiments involving a microwave generator aimed directly at the grapes, so that potential could be measured without microwaving the measurement equipment, and so that the spinning turntable, scattering of the microwaves, and constructive/destructive interference between the waves could be eliminated would be useful, as well as an experimental determination of the resistivity of grapes.

Even taking into account these flaws however, the research has accomplished it’s main goal: to establish some sort of sensible explanation for why a simple grape, cut in half and microwaved makes one of the most entertaining light shows you can get without leaving your kitchen.

Appliance Corner – Microwave Ovens – History

(http://www2.whirlpool.com/html/homelife/cookin/cookmwav5.htm)

Beaty, Bill. Unwise Microwave Oven Experiments

(http://www.eskimo.com/~billb/weird/microexp.html)

Carter, G.W. The Electromagnetic Field and it’s Engineering Aspects. Longmans’, 1967.

Cutnell, John D. Physics (2^nd Ed.) John Wiley & Sons, 1992.

Effect of Solute on Resistivity of Water

(http://www-classes.usc.edu/engr/ms/125/water/tsld015.htm)

Giancoli, Douglas C. Physics (4^th Ed.). Prentice Hall, 1995.

How things work: Microwave Ovens

(http://landau1.phys.virginia.edu/Education/Teaching/HowThingsWork/microwave_ovens.html)

The Lightening Page

(http://www.wvlightning.com/)

Michaud, Patrick. Fun with grapes – a case study. (http://www.sci.tamucc.edu/~pmichaud/grape/)

Resistivities for Common Earth Materials

(http://gretchen.geo.rpi.edu/roecker/AppGeo/lectures/res/resoutline.html)

Tulasidas, T.N., The dialectric Properties of Grapes and Sugar Solutions at 2.54 GHz, International Microwave Power Institute, 1995.