The biennial Conference for Food Protection provides a formal process for all interested parties to influence food safety guidance. At a recent conference, an issue was raised culminating in a formal request to the U.S. Food and Drug Administration to change its Food Code recommendation for safe cooking of seafood using microwave energy when steaming was also employed. The request was to treat microwave steam cooked seafood as a conventionally cooked raw animal product rather than a microwave cooked product, for which the safe cooking recommendation is more extensive owing to the complex temperature distributions in microwave heating. The request was motivated by a literature study that revealed a more uniform temperature distribution in microwave steam cooked whole lobster. In that study, single-point temperatures were recorded in various sections of the whole lobster, but only one temperature was recorded in the tail, although the large size of the tail could translate to multiple hot and cold points. The present study was conducted to examine lobster tail specifically, measuring temperatures at multiple points during microwave steam cooking. Large temperature differences, greater than 60°C at times, were found throughout the heating period. To compensate for such differences, the Food Code recommends a more extensive level of cooking when microwave energy, rather than conventional heat sources, is used. Therefore, a change in the Food Code regarding microwave steam heating cannot be recommended.

In dealing with the complete cooking of raw animal food, the U.S. Food and Drug Administration (FDA) Food Code (9) distinguishes between food cooked conventionally (e.g., contact heating or radiant heating) and food cooked in a microwave oven. Section 3-401.11 of the Food Code stipulates that conventional cooking must raise temperatures throughout the food to a minimum of 63°C (145°F) for 15 s. Implicit in this recommendation is the knowledge that the coldest point in a heating food is at its geometric center and therefore easily measured. Section 3-401.12, however, stipulates that for microwave cooking of raw animal foods, the minimum temperature should be 74°C (165°F) followed by a 2-min stand time where the food is covered or otherwise insulated from heat loss. This more extensive recommendation takes into account the inability to locate the coldest point in the microwave heated food owing to the inescapable complex temperature distribution caused by the nonuniform energy distribution in the microwave oven (10, 11). Thus, when a food preparer testing various portions of a microwave cooked food finds no temperatures less than 74°C, the 2-min hold time would allow possibly missed colder portions of the food to come up to at least 63°C owing to the higher temperatures elsewhere.

Microwave heating combined with steaming offers the potential for improved temperature distribution over that of microwaving without steam. Steam heating by itself is purely a surface phenomenon, imparting heat via condensation at the surface, with conduction transporting heat into the cooler interior of the food. Microwaves can impart heat within the food, with the possibility of leaving the surface much cooler than the interior. Therefore, combining the two should lead to more uniform temperature distributions than what microwave heating could achieve alone. This conclusion was reported by Li et al. (7), who investigated the temperatures obtained during microwave heating of whole lobster. Using thermocouples, the authors measured final temperatures in seven sections of the body, including at one location in the tail. Comparing temperatures using a specific protocol for heating lobster with and without steam followed by a 2-min hold time, Li et al. found that a minimum temperature of 63°C could be attained at the seven points within the lobster. Based on these experimental results, Issue 021 of Council III of the Conference for Food Protection (2) recommended that a letter be sent to the FDA requesting that the minimum cook temperature for microwave steam cooked seafood be lowered to match that for conventionally cooked seafood. However, given the complex temperature distribution in microwave heated foods, one temperature measurement point in the tail does not suffice to determine whether microwave steam heating has sufficiently heated the entire tail. Therefore, to address this request, the present study was conducted to take a closer look at microwave steam cooking in seafood by examining multiple temperatures within the lobster tail. Real-time temperatures were measured with fiberoptic temperature sensors at various points within the tail during heating with microwave energy within a steam environment also created with microwave energy. The purpose of this study was not to determine whether microwave steam heating results in more uniform heating than microwave heating alone but to evaluate overall heating behavior, which could lead to a reassessment of the Food Code guideline for microwave heated food.

Flash frozen cold water lobster tails 170 to 220 g and 13.5 to 14.0 cm long, with the greatest width 6.3 to 7.1 cm and the greatest height 4.8 to 5.2 cm, were obtained from a local seafood warehouse (Lobster Gram, Chicago, IL) and upon arrival were placed in a freezer at −20°C for storage. On the day before the experiment, holes were drilled into the frozen lobster tail for eventual placement of fiberoptic sensors. Two points at the top and three points along the side of the shell were made with a hobby drill (Dremel, Racine, WI) fitted with a 1.9-mm-diameter drill bit. The top holes were approximately 2.2 and 2.7 cm in depth, depending on where along the shell they were drilled, to reach the approximate center of the cross-sectional area perpendicular to the long dimension of the meat. The side holes were drilled to a constant 1.2 cm. In both groups of holes, account was taken of the sensitive portion of the sensor being displaced 7 mm from the tip (5). The tail was then placed in a refrigerator at 4°C for overnight thawing.

Five fiberoptic temperature sensors were attached to a UMI8 signal conditioner (FISO, Montreal, Quebec, Canada), which translated the light signals to temperatures and had been calibrated against a traceable Kaye 400 Intelligent RTD secondary standard (Amphenol Advanced Sensors, St. Mary's, PA) at five temperatures: 15, 30, 45, 60, and 75°C. Sensors were inserted into the predrilled holes of the thawed tail, taking care to not push the sensors beyond the end of the hole. The tail was placed in a half-size (32 by 26 by 4 cm) microwave-transparent high-heat polyamide H-Pan (Cambro USA, Huntington Beach, CA), 45 ml of cold water was added to the pan (following a published procedure (7)), and the fiberoptic cables were taped to the pan rim to keep them in place. The lid was placed on the pan, and the pan placed into the microwave oven. Figure 1 shows an example of one lobster tail fitted with five sensors and placed in a pan with water just before heating.

FIGURE 1.

Lobster tail just before heating, with fiberoptic temperature sensors (1 through 5) in place. The tips of sensors 1 and 2 are located at a depth of 2.7 and 2.2 cm, respectively, and the tips of sensors 3 through 5 are all located at a depth of 1.2 cm.

FIGURE 1.

Lobster tail just before heating, with fiberoptic temperature sensors (1 through 5) in place. The tips of sensors 1 and 2 are located at a depth of 2.7 and 2.2 cm, respectively, and the tips of sensors 3 through 5 are all located at a depth of 1.2 cm.

Close modal

A commercial nonsteaming microwave oven (model NE-12521, Panasonic, Secaucus, NJ) operating at 2,450 MHz was used for all experiments. This oven distributed microwave energy through two counter-rotating antennas, one above and one below the stationary food platform within the oven cavity. Experiments took place at full power (1,200 W) for 2 or 4 min. During the heating and hold time, temperatures were logged every 5 s with the UMI8 unit. Four experiments were conducted: three 2-min experiments served as replicates, and one 4-min experiment elucidated the extended heating behavior within the lobster tail.

Figure 2 shows the temperature measurements of the five sensors during microwave steam heating of lobster tail. Figure 2A shows the wide span of temperatures recorded. The sensor recording the coldest point among the five sensors was not the same throughout the heating period, switching from sensor 1 to sensor 2 just after 100 s of heating. Figure 2B shows the greatest temperature difference among the five sensors at each time point, indicating large temperature differences throughout heating, the highest of which was over 60°C.

FIGURE 2.

Temperatures obtained from a 4-min experiment in which a lobster tail was heated in a microwave field with microwave-induced steam. (A) Temperatures recorded by all five sensors. (B) Maximum-to-minimum temperature difference at each time point. Microwave power was enabled approximately 20 s after temperature logging commenced.

FIGURE 2.

Temperatures obtained from a 4-min experiment in which a lobster tail was heated in a microwave field with microwave-induced steam. (A) Temperatures recorded by all five sensors. (B) Maximum-to-minimum temperature difference at each time point. Microwave power was enabled approximately 20 s after temperature logging commenced.

Close modal

To better isolate the temperature differences before temperatures approached 100°C, 2-min experiments were used. Figure 3 shows the results of temperatures averaged over three replicate runs at sensors 1, 2, and 5. Sensors 1 and 5 were chosen to examine the temperature difference between two points in the same cross-sectional plane of the lobster tail. Sensor 2 was chosen to examine, along with sensor 1, the difference between different center temperatures where temperature measurement for doneness may take place. Included in Figure 3 is an additional y axis on the right side for assessing temperatures at the end of heating. The temperature increased linearly in each place as a function of time. The heating rate local to each sensor is proportional to the slope of these lines. When the heating rates are different, which they usually are, then temperatures continue to diverge as long as microwave heating continues. This fundamental problem prevents uniformity of heating in microwave ovens.

FIGURE 3.

Temperatures averaged over three replicate experiments at sensors 1, 2, and 5 as a function of time for 2-min heating trials. Error bars represent one standard deviation. The three replicates used to construct this figure did not include the results shown in Figure 2.

FIGURE 3.

Temperatures averaged over three replicate experiments at sensors 1, 2, and 5 as a function of time for 2-min heating trials. Error bars represent one standard deviation. The three replicates used to construct this figure did not include the results shown in Figure 2.

Close modal

To obtain the graph in Figure 3, temperatures from the three replicate experiments were averaged. However, averaged temperatures can compromise food safety assessment because complete cooking must be achieved every time, not an average number of times. Therefore, the maximum temperature differences possible in a food during microwave cooking must be documented. Figure 4A shows the temperature difference between sensors 1 and 5 in one of the three replicate experiments used to create Figure 3; this experiment was chosen for illustration because it documented the greatest temperature difference between the two sensors. Figure 4B shows the temperature difference between sensors 1 and 2. The final temperature difference in Figure 4A is greater than 30°C at the end of 2 min of microwave steam heating, and the final temperature difference in Figure 4B is greater than 10°C. In both cases, temperature differences persisted even though steam heating was also occurring.

FIGURE 4.

Temperatures as a function of time obtained from sensors 1 and 5 (A) and sensors 1 and 2 (B). Each data point was taken from one of the three experimental runs used to create Figure 3, chosen based on the greatest temperature difference in the two sensors. The data shown in the two graphs were from different experiments.

FIGURE 4.

Temperatures as a function of time obtained from sensors 1 and 5 (A) and sensors 1 and 2 (B). Each data point was taken from one of the three experimental runs used to create Figure 3, chosen based on the greatest temperature difference in the two sensors. The data shown in the two graphs were from different experiments.

Close modal

In food heated in consumer and commercial microwave ovens, the physics of 2,450 MHz microwave heating creates the complex nonuniform temperature distributions associated with this type of heating (11). Turntables in consumer ovens (1) and mode stirrers and counter-rotating opposed antennas in commercial ovens (1, 4) may reduce this nonuniformity but cannot eliminate it.

Another approach to reducing nonuniformity is to combine microwave heating with conventional heating. Conventional heating involves uniform surface heating, after which heat is transferred into the food via thermal conduction. Conduction is a slow process (10), but it uniformly transfers heat into the food, creating an easily definable cold spot at the geometric center of the food. With microwaves providing fast internal heating and conventional heating providing more uniform surface heating, a greater overall uniformity of heating can be achieved by combining these modes than can be achieved with either mode alone. In addition to steam, other conventional heating modes have been combined with microwaves: infrared heat, hot air, induction heating, and combinations of these modes (3). The particular combination of microwave and jet impingement was studied by Geedipalli et al. (6). In that study, surface heating by hot air impingement complemented internal heating by microwaves creating a measurably more uniform temperature distribution than could be achieved with microwave heating alone. Based on the performance of jet impingement microwave heating, condensed steam (another type of surface heating) should give similar results. However, these results depend on how well microwaves penetrate the food being heated.

Microwave penetration is a function of the free space wavelength of the microwaves and the relative dielectric constant, ε′, and the relative dielectric loss, ε″, of the food. Although ε′ of lobster is generally similar to that of other foods at low and high temperatures, ε″ is not (Table 1). This observation was the first indication that microwave heating would occur closer to the surface of lobster meat than in other foods. A penetration depth (dp) also can be calculated with these parameters (1). This depth is shown in Table 1, with values relative to that of lobster meat at 25 and 60°C, further indicating that microwave penetration is noticeably greater in other foods than in lobster. The reduced penetration of microwaves in lobster undergoing microwave steam heating apparently maintains, if not increases, the temperature difference between the surface and the interior of the food (Figs. 2 through 4). In contrast, microwave penetration in potatoes, which were used by Geedipalli et al. (6), is 1.7 times that in lobster and apparently sufficient for microwave heating to complement the jet impingement surface heating, resulting in the increased temperature uniformity these authors observed.

TABLE 1.

Dielectric properties of various raw foods and foods cooked at 2,450 MHz

Dielectric properties of various raw foods and foods cooked at 2,450 MHz
Dielectric properties of various raw foods and foods cooked at 2,450 MHz

Li et al. (7) examined temperature uniformity in microwave steam heated shrimp and with lobster; therefore, the request for the Food Code modification was for both lobster and shrimp. Twelve large shrimp (mean, 23.6 g per shrimp) were arranged in one undefined configuration, and temperatures were measured at one undesignated point in each shrimp after heating: all temperatures were above 62.8°C. However, microwave steam heating may succeed with shrimp even when it does not succeed for lobster. Microwave penetration is much greater for shrimp than for lobster (Table 1), indicating more efficient internal heating with microwaves. Because shrimp are smaller than lobsters, heat transfer by conduction, although slow, will be more successful because the heat does not have as far to travel to penetrate the shrimp completely. Nevertheless, multiple shrimp will experience different rates of heating depending on their placement in the oven and how they are arranged respective to each other (e.g., in a pile or separated). These variables confound any attempt to generalize an approach to measure minimum temperature. Therefore, without more extensive data, microwave steam heated shrimp should not be treated differently than any other microwaved raw animal food.

In much of the published research, the desire to improve temperature uniformity in microwave heating is driven by the desire for increased quality. Yet in microwave cooking, especially where raw animal foods are concerned, food safety is the primary issue. Even using various methods to increase temperature uniformity, the persistence of complex temperature distributions and temperature differences, including the large ones recorded in the present study, reinforces the need to keep the recommendation of 74°C with a 2-min hold time in the Food Code for all raw animal foods cooked with microwave heating.

1.
Buffler
,
C. R.
1993
.
Microwave cooking and processing: engineering fundamentals for the food scientist
,
p
.
150
151
.
John Wiley & Sons
,
Hoboken, NJ
.
2.
Conference for Food Protection
.
2014
.
2014 Biennial meeting, Council III
. .
3.
Datta
,
A. K.
, and
V.
Rakesh
.
2013
.
Principles of microwave combination heating
.
Compr. Rev. Food Sci. Food Saf
.
12
:
24
39
.
4.
Edgar
,
R. H.
, and
J. M.
Osepchuk
.
2001
.
Consumer, commercial, and industrial microwave ovens and heating systems, chap. 7
.
In
A. K.
Datta
and
R. C.
Anantheswaran
(
ed.
),
Handbook of microwave technology for food applications
.
Marcel Dekker
,
New York
.
5.
FISO Corporation
.
2007
.
FOT-L temperature sensors, product data sheet
.
Available at: http://fiso.com/admin/useruploads/files/fot-l.pdf. Accessed 2 August 2016
.
6.
Geedipalli
,
S.
,
A. K.
Datta
, and
V.
Rakesh
.
2007
.
Heat transfer in a combination microwave–jet impingement oven
.
Food Bioprod. Process
.
86
:
53
63
.
7.
Li
,
Y.
,
H.
Su
,
J. J.
Specchio
,
J. P.
Schrade
,
S.
Chung
, and
M.
Unanski
.
2014
.
Utilization of steam heat generated via microwave energy in seafood cooking
.
J. Nutr. Health Food Eng
.
1
:
1
6
.
8.
Tanaka
,
F.
,
P.
Mallikarjunan
, and
Y.-C.
Hung
.
1999
.
Dielectric properties of shrimp related to microwave frequencies: from frozen to cooked stages
.
J. Food Process Eng
.
22
:
455
468
.
9.
U.S. Food and Drug Administration
.
2013
.
Food Code
. .
10.
Vadivambal
,
R.
, and
D. S.
Jayas
.
2008
.
Non-uniform temperature distribution during microwave heating of food materials—a review
.
Food Bioprocess Technol
.
3
:
161
171
.
11.
Zhang
,
H.
, and
A. K.
Datta
.
2001
.
Electromagnetics of microwave heating: magnitude and uniformity of energy absorption in an oven
,
p
.
33
67
.
In
A. K.
Datta
and
R. C.
Anantheswaran
(
ed.
),
Handbook of microwave technology for food applications
.
Marcel Dekker
,
New York
.