Initial response to the cesium-137 accident: how Goiânia discovered the contamination
The initial response to the Goiânia cesium-137 radiological accident began on September 28, 1987, when M.F.1 carried the remains of the radioactive source to the Vigilância Sanitária (public health surveillance office), convinced that the glowing powder was killing her family. Until that moment, physicians had diagnosed the symptoms of vomiting, diarrhea, and skin lesions as an allergic reaction to food or a tropical disease. What unfolded over the next 24 hours completely changed the scale of the emergency.
For the full context of how the source was removed and dispersed, see our complete guide to the Goiânia radiological accident.
The cesium-137 source: essential technical data
The source involved in the Goiânia accident consisted of cesium-137 in the form of highly soluble cesium chloride salt. Manufactured around 1970 at the Oak Ridge National Laboratory (USA), the source was doubly sealed within two stainless steel capsules inside an international standard capsule with an external diameter of 50.6 mm and a window thickness of 1 mm.
By September 1987, the source activity was 50.9 TBq (1,375 Ci), producing a dose rate of 4.56 Gy/h at 1 meter. Cesium-137 emits 0.66 MeV gamma radiation (84% of disintegrations) and beta particles with maximum energies of 0.51 MeV (95%) and 1.17 MeV (5%), with a half-life of 30 years.
Radiological data and protection limits
The following table summarizes the key parameters of the source and relevant radiological protection data for managing the accident.
| Parameter | Value |
|---|---|
| Gamma emission | 0.66 MeV (84%) |
| Maximum beta emission | 0.51 MeV (95%); 1.17 MeV (5%) |
| Mean beta energy | 0.187 MeV |
| Half-life | 30 years |
| Specific gamma ray constant | 8.9 × 10⁻² mGy·h⁻¹ at 1 m per GBq |
| Activity (Sept 1987) | 50.9 TBq (1,375 Ci) |
| Dose rate at 1 m | 4.56 Gy/h |
| Material volume | 3.1 × 10⁻⁵ m³ |
| Mass | 0.093 kg |
| Specific activity | 0.55 TBq/g (15.1 Ci/g) |
| Dose per intake (ingestion) | 1.2 × 10⁻⁸ Sv/Bq |
| Dose per intake (inhalation) | 8.7 × 10⁻⁹ Sv/Bq |
| Annual limit of intake (oral) | 4.0 × 10⁶ Bq |
| Annual limit of intake (inhalation) | 6.0 × 10⁶ Bq |
| Derived air concentration | 2.0 × 10³ Bq/m³ |
Source: The Radiological Accident in Goiânia (IAEA, 1988) — Table I
From ignored symptoms to alert: a chain of misdiagnoses
The path to identifying the radiological contamination was slow and marked by repeated misdiagnoses. W.P., the first person to directly handle the rotating assembly, sought medical attention on September 15 — just two days after removing the assembly from the abandoned clinic. His symptoms of vomiting and hand swelling (edema) were attributed to an allergic food reaction.
M.F.1, wife of the junkyard owner where the capsule was opened, went through the same cycle. Admitted to São Lucas Hospital between September 21 and 23 with vomiting and diarrhea, she received the identical diagnosis: a reaction to contaminated food. She was sent home to rest.
W.P. returned to hospital on September 23 and was admitted to Santa Maria Hospital until September 27. The radiation-induced skin lesions were interpreted as symptoms of an infectious disease, and he was transferred to the Tropical Diseases Hospital. By that point, several patients with similar symptoms had already been seen at the same hospital — none correctly diagnosed.
The turning point came from Dr. R.P., a physician at the Tropical Diseases Hospital, who began to suspect that the patients’ skin lesions were consistent with radiation damage. He contacted Dr. A.M., who worked at both the hospital and the Toxicological Information Centre. In practice, it took 15 days and multiple incorrect diagnoses before anyone connected the symptoms to radiation exposure.
September 28: the source reaches the health authorities
On September 28, M.F.1 took the initiative that triggered the entire response chain. Convinced that the luminous powder from the junkyard was causing her family’s illness, she went with G.S. (an employee of D.F.) to Junkyard III, where the remains of the rotating assembly had been sold. G.S. placed the fragments in a bag, and they traveled by bus to the Vigilância Sanitária.
G.S. carried the bag on his shoulder during the journey — resulting in a significant radiation burn and an estimated whole-body dose of 3.0 Gy, with an intake of approximately 100 MBq (2.7 mCi).
At the Vigilância Sanitária, M.F.1 placed the bag on Dr. P.M.’s desk and told him the material was “killing her family.” Dr. P.M. left the source remnants on his desk for some time but, growing worried, moved them to an external courtyard where they sat on a chair against a wall for one day. His estimated dose was 1.3 Gy, with negligible intake since the source remained in the bag.
M.F.1 and G.S. were sent to a health centre, where the initial diagnosis was — once again — a tropical disease. They were then transferred to the Tropical Diseases Hospital, joining other contaminated patients already there.
September 29: the medical physicist confirms radiation
The morning of September 29 was decisive. Dr. J.P. from the Department of the Environment of Goiás State was contacted by the physicians and proposed that a medical physicist examine the suspicious package. By coincidence, W.F., a licensed medical physicist, happened to be visiting Goiânia at the time.
W.F. borrowed a scintillometer from the NUCLEBRÁS offices in Goiânia — an instrument normally used for uranium prospecting, with a dynamic range of 0.02 to 30 µGy/h and high sensitivity to radiation. Approaching the Vigilância Sanitária, still some distance away, he switched on the monitor. The needle deflected full scale regardless of which direction he pointed it. Assuming the meter was defective, he returned to NUCLEBRÁS for a replacement.
With the second instrument, he arrived at the Vigilância Sanitária at 10:20, by then convinced that a major radiation source was nearby. He arrived just in time to dissuade the fire brigade from their initial plan of picking up the source and throwing it into a river.
By 11:00, W.F. persuaded the occupants to vacate the premises. At noon, he and Dr. P.M. went to Junkyard I, where they found widespread contamination — the monitor reading exceeded its scale over a wide area. With considerable difficulty, they persuaded D.F., his family, and many neighbours to evacuate.
Mobilization of authorities and the scale of the emergency
At 13:00 on September 29, W.F. and others went to the offices of the Secretary for Health of Goiás State to inform the authorities. As the IAEA report describes, officials were incredulous of the account and the potential scale of evacuation required. It took considerable persuasion before W.F. was permitted to see the Secretary for Health.
At 15:00, the Director of Nuclear Installations at CNEN was reached as the designated coordinator for nuclear emergencies (NEC). He suggested contacting the hospital physicist at the IGR for assistance, given the wider range of instruments available. The source was tentatively identified as possibly originating from the IGR.
Between 16:00 and 20:00, multiple actions occurred simultaneously:
- The Tropical Diseases Hospital was informed that patients were contaminated and should be kept isolated.
- Civil defence elements — police, fire brigade, ambulances, and hospitals — were placed on alert.
- Known contamination sites (the Vigilância Sanitária and Junkyard I) were resurveyed with IGR equipment.
- The Secretary for Health held a meeting and planned to receive contaminated persons at the city’s Olympic Stadium.
By this time, the press was taking an interest in events, amplifying both the institutional response and public alarm.
Overnight triage at the Olympic Stadium
At 22:00 on September 29, Z.S. — who had earlier offered to cut up the assembly pieces with an oxyacetylene torch but forgot to do so — found W.F. and explained how the source had been broken up and where the pieces had been taken. This information enabled the monitoring team to identify additional major contamination sites and evacuate more people.
During the night of September 29-30, 22 people were identified at the stadium as potentially having been highly exposed. They were placed in tents, segregated from others, based on contamination measurements, medical symptoms, and their family groupings at the major contamination sites. Dr. A.M. and a colleague performed initial screening; those with lesions were sent to the Tropical Diseases Hospital.
As the IAEA report notes, the interest aroused by the blue glow emanating from the cesium chloride significantly affected the course of the accident. The compound’s high solubility contributed to extensive contamination of persons, property, and the environment. Without these factors, the incident might have developed similarly to the 1983 cobalt-60 accident in Ciudad Juárez, Mexico, which resulted in little contamination and no deaths.
The blue glow phenomenon
The blue glow emanating from the cesium-137 chloride was what attracted onlookers, spread the contamination, and — paradoxically — was also what led M.F.1 to connect the powder to her family’s illness. D.F. noticed the glow on the night the remnants were brought to his home and invited neighbours, relatives, and acquaintances to see the “curiosity” over three days. People even daubed the radioactive powder on their skin, as they would with carnival glitter.
The phenomenon was subsequently observed by researchers from ORNL and the REAC/TS (Radiation Emergency Assistance Center/Training Site) during the decapsulation of a cesium-137 chloride source in early 1988. The glow is thought to be associated with fluorescence or Cherenkov radiation due to moisture absorption by the source. Further research was underway at Oak Ridge to determine its exact nature.
For the complete chronology, from source removal to material dispersal, read our detailed article on how the Goiânia radiological accident happened.
Lessons from the discovery phase
The initial response to the Goiânia accident reveals patterns that recur in radiological emergencies worldwide. The systematic failure in diagnosis — with at least four patients receiving tropical disease or allergy diagnoses — demonstrates the importance of including radiation exposure in the differential diagnosis when atypical skin lesions present without clear aetiology.
The speed at which contamination spread between September 13 and 28 — fifteen days without detection — also highlights the risk posed by orphan sources in decommissioned facilities. A 50.9 TBq source remained accessible in an abandoned clinic long enough to be removed with simple tools.
M.F.1’s decisive action in bringing the fragments to the health authorities, and the coincidence of W.F. being in Goiânia with access to instrumentation, were the two factors that accelerated detection. Without them, the gap between dispersal and discovery could have been even longer, with potentially more severe consequences.
