Megabecquerel to Kilocurie

Diagnostic imaging doses fall neatly in the MBq range — a PET scan uses 185–370 MBq, a bone scan 500–800 MBq. Using becquerels would mean writing hundreds of millions; using gigabecquerels would mean awkward decimals like 0.37 GBq. MBq is the Goldilocks unit for the hospital pharmacy: large enough to avoid scientific notation, small enough to express a single patient dose as a tidy number on a syringe label.

That depends entirely on the isotope. Technetium-99m, the workhorse of diagnostic imaging, has a 6-hour half-life — so a 740 MBq injection drops to 370 MBq in 6 hours, 185 MBq in 12, and becomes negligible within 2 days. Fluorine-18 (used in PET) has a 110-minute half-life and is essentially gone in a day. Iodine-131 (used in therapy) lingers for about 8 days per half-life. Hospitals choose isotopes partly based on how fast they want the activity to vanish.

Most diagnostic isotopes (Tc-99m, F-18) have half-lives under a day, so hospitals simply store waste in shielded bins and let it decay. After 10 half-lives — about 3 days for Tc-99m — the activity is down to less than 0.1% of the original and can be disposed of as normal clinical waste. Longer-lived therapeutic isotopes like I-131 require weeks of decay storage. The vast majority of nuclear medicine waste is never shipped to a radioactive disposal site; it just sits in a locked closet until physics solves the problem.

A patient injected with 370 MBq of F-18 for a PET scan emits gamma rays at a dose rate of roughly 5–6 µSv/hr at one meter. That means sitting next to them for two hours gives you about 10–12 µSv — less than a chest X-ray. Staff handle dozens of patients daily so they follow time-and-distance protocols, but for family members the exposure from a single visit is trivially small. The activity halves every 110 minutes, so by evening the patient is barely distinguishable from background.

Molybdenum-99, which decays into the technetium-99m used in 30+ million scans per year worldwide, can only be produced in a handful of aging research reactors. It has a 66-hour half-life so it cannot be stockpiled — you have to make it, ship it, and use it within days. When a reactor goes down for maintenance (as happened in 2009 when both the Canadian NRU and Dutch HFR shut down simultaneously), hospitals worldwide face scan cancellations within a week. New production methods using particle accelerators and LEU targets are slowly diversifying supply.

In 1987, scrap metal scavengers in Goiânia, Brazil broke open an abandoned caesium-137 teletherapy source containing about 1,375 Ci (50.9 TBq). The glowing blue Cs-137 powder fascinated locals — they rubbed it on skin, gave it to children, and spread it across multiple homes. Four people died, 249 were contaminated, and the cleanup produced 3,500 m³ of radioactive waste. The incident became the textbook case for why sealed sources must be tracked and securely stored throughout their entire lifecycle, and why the IAEA created its Code of Conduct on the Safety and Security of Radioactive Sources.

Yes, multiple times. In Ciudad Juárez, Mexico (1983), a stolen Co-60 teletherapy source was sold as scrap and melted into rebar, contaminating 4,000 tonnes of steel and exposing thousands. In Samut Prakan, Thailand (2000), a junked Co-60 source killed three scrap workers who pried it open. In Yanango, Peru (1999), a welder pocketed an Ir-192 industrial radiography source and carried it in his pocket for hours — his leg was amputated. The IAEA documents over 30 serious radiation accidents involving orphaned or stolen sources since the 1960s, collectively killing dozens and injuring hundreds.

Cobalt-60 has a 5.27-year half-life, so a 500 kCi source drops to 250 kCi after five years and becomes too weak for industrial throughput after about 15–20 years. The spent source pencils are returned to the manufacturer (typically in Canada or Russia) for reprocessing or secure storage. Transport uses heavily shielded Type B casks certified to survive a 9-meter drop and 30-minute fire. The manufacturer often offers a swap program: deliver fresh sources and take back decayed ones in the same shipment, minimising the number of high-activity transports.

The Fukushima Daiichi disaster released an estimated 10–30 PBq (10,000–30,000 TBq) of caesium-137 directly into the Pacific Ocean between March and July 2011 — the largest single marine radioactive release in history. For comparison, the Sellafield reprocessing plant in the UK discharged about 40 PBq of Cs-137 into the Irish Sea over decades of operation (1952–2000). Soviet dumping of entire reactor compartments from nuclear submarines in the Arctic added further inventory. Despite these numbers, ocean dilution is vast: Pacific Cs-137 levels from Fukushima peaked at about 50 Bq/m³ near the plant and dropped below 2 Bq/m³ within a few hundred kilometers.

This is exactly why the IAEA, NRC, and national agencies track high-activity sources so aggressively. A kilocurie Cs-137 or Co-60 source dispersed by conventional explosives would contaminate a few city blocks — not causing acute radiation casualties (the blast itself is deadlier) but creating a costly, panic-inducing cleanup lasting months. The actual health risk to the public would be low, but the economic and psychological damage would be enormous. Post-9/11 programs like the US GTRI (now NNSA OSRP) have recovered or secured thousands of orphaned high-activity sources worldwide.