The amplified microwave signal was fed through a coaxial isolator and a coaxial double bridge into the sample holder, a rectangular cavity (110 x 55 x 184 mm) made of two coax to rectangular waveguide adapters.
The narrow side of the waveguide and the E field were horizontal, parallel to the culture liquid layer (10 ml contained in a 25 cm² rectangular Falcon plastic flask). Due to small differences in the volume and positioning of the samples, the incident power was adjusted at the beginning of each exposure to obtain nearly the same power loss in all 18 experiments (nine CW and nine GMSK).
A thermostated (35 ± 0.1 °C) water jacket surrounded the sample holder cavity. The sample temperature, on its own, was allowed to rise from 35 ± 0.1 °C to no more than 35.7 ± 0.1 °C at the end of the exposure. No sham exposure was performed. Unexposed control cultures were left in the incubator at 37 ± 0.2 °C.
The incident, reflected, and transmitted RF powers were measured during the actual exposure using power sensors and power meters, and the power lost in the sample holder unit (sample included) was evaluated. From this, the power lost per unit mass of the sample (m = 10.4 g) was 5.02 ± 0.22 W/kg for the CW exposures and 4.99 ± 0.24 W/kg for the GMSK exposures. These evaluations were checked by means of calorimetric measurements, performed following the technique suggested by Allis et al. . Heating and cooling curves were recorded at nine different points of a dummy sample (culture medium alone) using point like thermocouples and temperature monitors. From the nine temperature curves, nine local SARs were calculated: the mean value was 2.25 W/kg, with an SD of 0.87 W/kg (minimum value 1.17 W/kg and maximum value 3.93 W/kg). These SAR values were considerably less than the total power lost per unit mass of the sample (see above), that included the losses of the sample holder structure. Due to the large uncertainties of SAR evaluations based on temperature measurements [Burkhardt et al., 1996], the mean value of power lost (~5 W/kg) can be assumed as an upper bound, actually never exceeded by the power absorbed by the sample.
Esmekaya MA et al.
Mutagenic and morphologic impacts of 1.8 GHz radiofrequency radiation on human peripheral blood lymphocytes (hPBLs) and possible protective role of pre-treatment with Ginkgo biloba (EGb 761)
Maes A et al.
Cytogenetic investigation of subjects professionally exposed to radiofrequency radiation
Scarfi MR et al.
Exposure to radiofrequency radiation (900 MHz, GSM signal) does not affect micronucleus frequency and cell proliferation in human peripheral blood lymphocytes: an interlaboratory study