Anatoli Bugorski: Head Inside a Particle Accelerator

Anatoli Bugorski was a Soviet physicist who, in 1978 at the U-70 synchrotron in Protvino, put his head in the path of a proton beam widely cited as 76 GeV, and survived. The reconstructed dose was about 2,000 Gy at the entry point and about 3,000 Gy at the exit. He lived because the beam was far too energetic to stop in a head, so a mean 42 MeV of its 76,000 MeV went into 17 cm of tissue, and because it crossed a channel only a few millimetres wide, sparing the bone marrow and gut whose whole-body failure causes fatal radiation sickness (PDG stopping-power formula, read 30 October 2026).

Two thousand grays. That is the reconstructed radiation dose at the entry wound in Anatoli Bugorski’s skull, the figure still cited for the instant a live proton beam bored through his head at the U-70 synchrotron in 1978. Set that beside the roughly 4.5 sievert that kills half of untreated people exposed across their whole body, and the Bugorski number looks like it belongs to a different universe. It does. The mismatch between the two is the actual story.

Every retelling reaches for the same move: divide his dose by a lethal dose and marvel at the ratio. That ratio is the wrong tool, because a radiation dose without a stated volume and tissue means almost nothing. Correct for what was actually irradiated, a few cubic centimetres of brain and inner ear rather than a whole human body, and the “miracle” stops looking miraculous. It starts looking like physics doing exactly what the numbers predict.

The Accident: Anatoli Bugorski and the U-70 Beam

In 1978, Bugorski was a researcher at the Institute for High Energy Physics in Protvino, home to the U-70 synchrotron, then the Soviet Union’s largest particle accelerator. On the day, he was checking a malfunctioning piece of beamline equipment. The door interlock had been switched off during an earlier experiment and never switched back on, and the warning light over the door was dark because its bulb had burnt out. Eight years later the Space Shuttle Challenger was lost to another overridden safeguard, a known temperature margin; in both cases people ended up where the system assumed no one would be.

Bugorski leaned in with his head in the beamline. A proton beam moving at 99.99% the speed of light, carrying an energy widely cited as 76 GeV (the U-70’s nominal design maximum was 70 GeV), passed through the back of his skull and out through his face in a fraction of a second. He later described the moment as a flash brighter than “a thousand suns”, a line repeated in nearly every account since, though no primary transcript has surfaced. There was no pain; the exposure was over before his nervous system could register it.

What followed over the next days was visible and frightening: the left side of his face swelled, and the skin along the beam’s exit track peeled where the tissue had been destroyed. Doctors who examined him expected him to die within days, the standard prognosis for anyone who had absorbed a dose anywhere near what the beam physics implied. He did not die, and understanding why requires throwing out the framing doctors reached for first.

Soviet Secrecy and a Story That Took Decades to Surface

Bugorski was transferred to Moscow specialists who studied him as a singular case: a living data point on what a beam of that energy does to a human head. His treatment records were classified immediately and were still hard to obtain twenty years later, a common pattern for anything touching the USSR’s nuclear and accelerator programs; in the 1990s he said he would welcome study by Western researchers but could not afford to leave Protvino. It sits alongside other closed chapters of Soviet big science, where unprecedented results stayed unpublished for reasons that had nothing to do with the science itself.

Although the beam damaged his face and hearing, his intellectual abilities were virtually undamaged, though mental work now tired him much faster, and in 1980 he defended the candidate of sciences dissertation (the Soviet equivalent of a PhD) that he had prepared before the accident. That fact alone told Soviet physicians something important about how localized the damage really was.

Popular accounts often claim the story broke in the late 1980s, with the world supposedly oblivious for nearly fifteen years. That timeline does not hold up: the earliest accounts we could verify are Masha Gessen’s article in Wired (December 1997) and a long Izvestiya piece in January 1998, and Bugorski did not speak publicly about the accident for more than a decade. Once outside researchers did learn the details, Bugorski’s case became one of the only human data points anywhere for what a body does with an extreme, sharply localized dose from a high-energy proton beam.

The Numbers: Anatoli Bugorski’s Dose in Context

The “hundreds of thousands of times fatal” claim comes from setting a local dose beside a whole-body one. A gray (Gy) is energy absorbed per kilogram of tissue, whether that tissue is a few cubic centimetres or the whole body. A sievert (Sv) weights dose for radiation type and organ sensitivity and is meant for long-term cancer risk; thresholds for acute radiation sickness are whole-body absorbed doses. For protons the two numbers are close (relative biological effectiveness near 1.1). What separates Bugorski’s figure from a lethal dose is the volume irradiated. To convert the gray and sievert figures quoted here, or the older rad and roentgen units, use the radiation units converter.

The Bugorski entry and exit figures below are local Gy values through a few cubic centimetres of tissue. They come from a widely repeated reconstruction, not a primary dosimeter reading; no named calculating physicist or original dosimetry report has been independently verified. The two radiotherapy rows are also local doses; the remaining rows are whole-body figures, drawn from sources such as the HHS Radiation Emergency Medical Management program and UNSCEAR. Reading down the table as one continuous scale is the mistake; reading it as two separate scales is the point.

Measurement	Value	Type	Source
Bugorski, exit point (localized)	3,000 Gy (from 300,000 roentgen)	Local dose, Gy	Medical record as quoted in Izvestiya (1998); converted, approximate
Bugorski, entry point (localized)	2,000 Gy (from 200,000 roentgen)	Local dose, Gy	Medical record as quoted in Izvestiya (1998); converted, approximate
Full head/neck radiotherapy course	60–70 Gy total	Local tissue, Gy	Standard external-beam oncology (ICRU guidance), ~30–35 fractions
Typical single radiotherapy fraction	~2 Gy	Local tissue, Gy	Standard external-beam oncology
Hisashi Ouchi, Tokaimura criticality accident, 1999	~17 Sv (STA estimate)	Whole-body, Sv	Published clinical case reports; died 83 days post-exposure
Human whole-body LD50/60, no medical care	~4.5 Sv	Whole-body, Sv	REMM/HHS; NRC-cited (~6–7 Sv with aggressive supportive care)
Annual background radiation, global average	~2.4 mSv (0.0024 Sv)	Whole-body, Sv	UNSCEAR

The Bragg Peak: Why Proton Therapy Works, and Why This Beam Didn’t Stop

Here is the piece of physics that actually explains the survival, and it starts with how charged particles lose energy as they travel through matter. A proton slows as it plows through tissue, and it does not lose energy at a steady rate the whole way. It dumps the overwhelming majority of its energy in a sharp spike right near the end of its range, called the Bragg peak. Everywhere before that peak, the particle deposits a relatively low, flat dose; at the peak itself, the dose jumps dramatically before falling to almost nothing beyond it.

Proton therapy for cancer exploits this directly. Clinical beams run at roughly 70 to 250 MeV, tuned so the Bragg peak lands precisely inside the tumour, concentrating damage there while sparing healthy tissue in front of and behind it. It is a deliberately engineered version of what makes protons attractive for radiotherapy in the first place, a level of energy control that has pushed accelerator engineering in directions as extreme as keeping magnets colder than deep space at the LHC, in the same broad family of machines as the U-70.

A proton of that energy is roughly 300 to 1,000 times more energetic than a clinical proton-therapy beam, with a range in dense matter measured in metres, not centimetres. A human head is roughly 15 to 20 centimetres across. The beam that hit Bugorski was so far past the range where a Bragg peak could form inside a human body that it never got the chance. It crossed his skull and brain on the flat, near-minimum-ionizing part of the energy-loss curve and kept going, exiting through his face still carrying most of its original energy. Proton therapy puts the peak inside the target; Bugorski’s accident is the exact inverse, a beam too energetic to stop inside a person at all.

Worked example: what 76 GeV does inside a head

Take the widely cited 76 GeV. Such a proton has a Lorentz factor of 82 and moves at 99.993% of the speed of light. At that energy the mean stopping power of soft tissue is 2.47 MeV per g/cm². A head 17 cm across, at a density of 1 g/cm³, therefore takes about 2.47 × 17 = 42 MeV. That is 0.06% of the beam energy. If the whole path were dense bone at 4.3 MeV/cm, a 20 cm path would take 86 MeV, still 0.11%.

The dose figure follows from the same number. One MeV absorbed per gram is 1.602 × 10^-10 Gy. For 2,000 Gy at the entry point, the fluence is 2,000 / (2.47 × 1.602 × 10^-10) = 5.0 × 10¹² protons per cm². This treats all lost energy as absorbed on the spot, so it is an upper estimate of the dose per proton. The reconstructed 2,000 Gy remains a quoted figure, not a measured one.

Change the particle, energy and material in the energy-loss calculator, embedded at the end of this page, to see how the loss changes when the proton is slowed to clinical energies. For the signal a similar particle leaves in a thin silicon sensor, use the detector signal calculator. The other sourced physics tools are listed under calculators.

Why Radiation Sickness Never Set In

The second half of the explanation is biological. Acute radiation syndrome, the classic radiation sickness that kills within days to weeks, is a disease of the whole body rather than of any one wound. It kills by wiping out tissues that depend on constant, rapid cell renewal: bone marrow, which produces blood cells (the hematopoietic syndrome, roughly 1 to 10 sievert whole-body), and the lining of the gut (the gastrointestinal syndrome, above roughly 10 sievert whole-body).

Bugorski’s beam was a channel only a few millimetres wide, passing through the back of his skull, through brain tissue, and through the inner ear. It never touched his femur, pelvis, sternum, or vertebrae, the bones that house the marrow acute radiation syndrome depends on destroying. It never touched his gut lining at all. Averaged across his entire body, the dose that actually governs survival from radiation sickness was small, even though the dose along that narrow track was, in local terms, catastrophic.

The damage that did occur maps precisely onto what the beam actually crossed. Injury to the facial nerve, cranial nerve VII, produced the lasting paralysis down the left side of his face. Injury to the cochlear-vestibular nerve, cranial nerve VIII, produced permanent hearing loss in his left ear. Focal scarring in the brain tissue the beam passed through created a seizure focus, which is why Bugorski has lived with recurring epilepsy ever since. None of it is mysterious once you know exactly where the beam went.

What Happened to Anatoli Bugorski After

He returned to the Institute for High Energy Physics about eighteen months after the accident and later coordinated physics experiments on the U-70 beam that had injured him: a scientist who kept working in the same field that had nearly killed him, in Protvino, the closed research town where the accident happened.

We found no reliable report of his death; born on 25 June 1942, he would now be 84. The last first-hand accounts of his health, from 1997 and 1998, describe recurring seizures, deafness in the left ear and paralysis of the left side of his face: the same injuries that mapped directly onto the beam’s path. The Anatoli Bugorski case reads, on paper, like it should have ended in 1978; the injury was real and permanent, and it stayed local to the beam’s track.

Anatoli Bugorski: Frequently Asked Questions

Did Anatoli Bugorski survive the particle accelerator accident?

Yes. Despite taking a proton beam, widely cited as 76 GeV, through his head at the U-70 synchrotron in 1978, Anatoli Bugorski survived, defended his candidate of sciences (PhD-equivalent) dissertation in 1980, and went back to work at the institute where the accident happened.

Why didn’t the proton beam kill him?

Two separate facts explain it. The beam was so energetic that it never formed a Bragg peak inside his head. It crossed his skull and brain on the flat part of its energy-loss curve and exited through his face rather than depositing a lethal concentrated dose and stopping. And acute radiation syndrome, the disease that normally kills after severe exposure, requires whole-body damage to bone marrow and gut lining; the beam’s narrow track missed both entirely.

Is it true he got hundreds of thousands of times the fatal radiation dose?

No. That figure, though widely repeated, does not hold up. The commonly cited comparison is closer to 200 to 600 times a conventional whole-body lethal dose, and even that comparison is misleading, because Bugorski’s reconstructed dose is a local Gy figure for a few cubic centimetres of tissue, while lethal-dose thresholds are whole-body figures. See the numbers above for the full comparison.

What were the long-term effects of the accident?

The left side of his face was permanently paralysed, he lost hearing in his left ear, and he has experienced recurring seizures in the years since. A popular claim that the paralysed side of his face wrinkled more slowly with age is a plausible but clinically unverified anecdote, not a documented finding.

When did the world learn about the accident?

Not for a long time. Soviet authorities kept the case out of view for years after 1978. The earliest accounts we could verify are Masha Gessen’s article in Wired (December 1997) and an Izvestiya feature in January 1998; claims pinning it to the late 1980s are not well supported.

Where did the accident happen?

At the U-70 synchrotron at the Institute for High Energy Physics in Protvino, then the Soviet Union’s most powerful particle accelerator, when Bugorski entered the beamline with the beam still live: the door interlock had been left switched off and the warning light’s bulb had burnt out.

Sources: Particle Data Group, Passage of particles through matter, Review of Particle Physics (2024); NIST ESTAR and PSTAR stopping-power tables; Izvestiya (January 1998) and Wired (December 1997) for the accident; REMM/HHS and UNSCEAR for the dose thresholds.

@online{scienceshot-the-man-who-put-2024,
  author       = {Turner, Alex},
  title        = {{Anatoli Bugorski: Head Inside a Particle Accelerator}},
  organization = {ScienceShot},
  date         = {2024-12-22},
  year         = {2024},
  url          = {https://scienceshot.com/post/the-man-who-put-his-head-in-a-particle-accelerator}
}