CSIRO - June 1994 BIOLOGICAL EFFECTS AND SAFETY OF EMR |
10.2.2 Cellular Telephones
Exclusion ClauseThe Australian Standard (SAA 1990) and the IRPA Standard (IRPA 1988) contain exclusions for devices that have output powers of less than 7 Watts and transmission frequencies of less than 1000 MHz. The two types of hand held mobile telephones currently used in Australia are:- (a) (AMPS) Advanced Mobile Phone Systems operating with a maximum radiated power of 600 mW and frequencies between 825 and 845 MHz (b) (GSM) Global System Mobile with voice information digitally encoded radiates 0.8 and 2.0 W peak power in a frequency band 890 to 915 MHz. The transmission is pulsed at a repetition frequency of 217 Hz with pulse widths of approximately 0.6ms. Therefore, these hand-held mobile telephones are excluded from compliance with the Australian or international, IRPA, standard. Germany has dropped the exclusion clause and instead requires certification of compliance based on "worst case" spatial peak SAR, as exposures from "low power hand held devices has been strongly under-estimated in the past" (Kuster 1993). The ICNIRP standard is about to be amended following agreement to delete the exclusion clause. Pulsing EffectsThe GSM digital telephones have stronger peak electromagnetic fields than the analogue telephones and have been shown to cause electromagnetic interference in a range of electronic medical equipment (Clifford et al 1994, Bassen et al 1994). Increasing the power of the GSM phone did not change the symptoms. Many of the cellular responses, including transmembrane ionic flow, are elicited by RF emission that is modulated at around 100 Hz, whereas exposure to continuous wave at the same fundamental RF frequency has no effect. There may be some reason to consider that the relatively high instantaneous power that causes EM interference may be the parameter that elicits responses in sensitive biological cell membrane receptors. This issue of critical exposure parameters is fundamental to an evaluation of potential health issues and requires urgent investigation. The process of investigating underlying mechanism of interaction has had little direct attention to date. DosimetryA recent study in Australia (Fleming & Joyner 1992) assessed the RF radiation dose in a physical phantom of a human head when exposed to an AMPS Telecom cellular telephone. The telephone emitted 600 mW continuous wave radiated power which produced an average SAR of 0.7 W/kg in tissue-equivalent gel within the eye. At the same distance of 5 cm from the antenna the measured power density was 0.27 mW/cm2. No increase in temperature was detected. Using numerical estimates the same group (Fleming 1994) reported peak SAR of 1.77 W/kg /W in the skin muscle and bone adjacent to the ear. Information received from the CTIA shows similar estimates in situ from cellular telephones used in the USA. Balzano was reported to have tested cellular phones with a physical model designed to give a worst-case estimate. In the model, signal attenuates rapidly, and the peak areas of energy absorption occurred in the cheek and mastoid area with peak exposures of 0.5 W/kg and brain exposures of 0.3 - 0.4 W/kg. When the antenna was collapsed the peak exposure was 1.1 W/kg and the brain exposure was 0.5 W/kg. Energy can be absorbed from near field, as well as far field, and the absorbed energy is a function of the medium, as well as geometry. Gandhi originally described the anatomy of SAR from cellular telephone exposures using a magnetic resonance imaging model of a male volunteer. The average SAR of the whole body was less than 0.08 W/kg with a peak of 0.7 W/kg at the level of the handset. Approximately 10-12% of the power is absorbed in the head and neck region. The peak SAR (behind the ear) for any 1 gram of tissue was approximately 0.2 W/kg, and the average for brain tissue was 0.014 W/kg. However, other groups including NRPB and Kings College in the UK and Swiss Federal Institute of Technology have used numerical models and direct measurement to show substantially greater values for SAR in the brain when a cellular phone is located next to the head. The UK groups have greatly refined the procedure of MRI based phantom and numerical techniques to obtain resolution for dielectric constants for specific tissues (Gabriel et al 1989; Dimbylow 1993). It seems that an error in Gandhi's model was partly due to incorrect estimate of the dielectric constants for bone and brain tissue. Using FDTD calculations on 5 x 105 cells in 2 mm3 voxels in an MRI acquired image of a human head, Dimbylow (1994 BEMS presentation) showed SAR values 3.1 W/kg averaged over 10 gm tissue inside the head. The SAR averaged over 1 gm of tissue was 4.7 W/kg for a quarter wave monopole operating at 900 MHz. When operating at 1.8 GHz the maximum SAR values along the side of the head were 4.6 and 7.7 W/kg for 10 and 1 gm of tissue, respectively. Based on these estimated SARs the maximum power that would be required to be emitted to meet the ANSI Safety Standard for the uncontrolled population (1.6 W/kg) would be 0.34 W at 900 MHz. For 1.8 GHz the maximum power value required to reach the ANSI safety limit is 0.21 W. Lovisolo et al (1994) reported estimated SAR values of 1.9 W/kg averaged over 10 gm of liquid brain-equivalent material in a cylindrical phantom head exposed to 0.6 W cellular phone operating at 900 MHz. Using data supplied by Gabriel an anatomically correct (in terms of dielectric properties and dimensions) phantom head was used for the direct measurement of the worst-case exposure (Meire & Kuster 1994) as it was claimed that many cellular phones exceed the ANSI Standard for SAR per 1 gm of tissue. Their measurement show peak SAR levels of 3.5 and 2.5 W/kg at depths in the head of 5 and 10 mm, respectively. Bone of 5 mm thickness reduced SAR by less than 15%. 10.3 REGULATIONWhile cellular telephones on the market do meet an existing ANSI standard, FDA questions the adequacy of the standard. FDA administers the Radiation Control for Health and Safety Act. Its purpose is to protect the public from unnecessary radiation from electric devices, and it covers many consumer products. Unlike the situation for medical devices, there is no premarket screening. FDA acts after marketing by recalling products, issuing civil penalties, and setting performance standards. Currently, there are insufficient data to establish cellular telephones as hazardous or to even set a safe standard. If FDA did have to go through notice and comment to set a new standard, it likely would be below today's ANSI standard (personal communication). FDA needs research data that characterises the potential hazards, if any, of cellular telephone use as soon as possible, including emissions, biological effects and risks.
[CONTENTS] [GLOSSARY] [REFERENCES] Current Research Abstracts Index
Please e-mail comments, information and updates to DON MAISCH: |