A team at the Radiation Application Research School of the Nuclear Science and Technology Research Institute in Karaj, Iran, has developed a real-time X-ray dosimeter built from silicone rubber mixed with bismuth oxide nanoparticles, according to a study published in Scientific Reports, a Nature portfolio journal. The work, reported in October 2026, describes sensors that respond to diagnostic-range X-rays with a signal roughly three times stronger than that of the unmodified silicone. The authors present the material as a lead-free candidate for low-dose radiation monitoring.
The devices were made by solution casting, a fabrication route in which the nanoparticle filler is dispersed into a room-temperature-vulcanized silicone matrix, designated RTV-2. The filler, monoclinic bismuth oxide (Bi₂O₃) nanorods, was varied from zero to 60 percent by weight across the sample set.
Why it matters
Dosimeters used in medical imaging need to register small radiation exposures quickly and reliably. Many existing real-time sensors depend on materials that are costly to process or contain lead, a toxic heavy metal. A sensor built from an inexpensive, flexible silicone loaded with a bismuth compound would sidestep both problems if it holds up outside the laboratory.
The study's headline number is a normalized sensitivity of about 0.35 nA/(mGy/min), which the authors express as 878 nC·mGy⁻¹·cm⁻³ on a volumetric basis. Nature reports that this figure compares favorably with values published for commercial MOSFET dosimeters, which fall in the range of roughly 0.1 to 1 nA/mGy, and with earlier bismuth oxide–polymer systems.
How the material was built and checked
Before testing the sensors electrically, the team characterized the composites structurally and chemically. X-ray diffraction, scanning electron microscopy, energy-dispersive X-ray spectroscopy with elemental mapping, and elemental analysis together indicated that the bismuth oxide nanorods were spread evenly through the amorphous silicone, with no sign of clumping at the scale the instruments could resolve. The authors also ran Fourier-transform infrared and Raman spectroscopy alongside thermogravimetric analysis, which they say confirmed that the composites retained their chemical structure, thermal stability and mechanical strength.
The dosimetric tests used a COMET MXR 320/26 X-ray source operating at diagnostic energies, with dose rates between 2.8 and 12 mGy/min. Current–voltage measurements taken at 4.8 mGy/min showed a highly linear response across a bias range of plus or minus 400 volts — meaning the output tracked the applied voltage predictably rather than drifting.
At 400 volts, the photocurrent rose from about 0.56 nanoamperes in pure RTV-2 to about 1.66 nanoamperes in the composite containing 60 weight percent bismuth oxide. That is the roughly threefold gain the authors highlight, and it points to the filler as the active ingredient: bismuth's high atomic number makes it an efficient absorber of X-rays, which the polymer alone is not.
The sensors also performed well on two practical measures. Across ten consecutive readings, the standard deviation was 2.3 percent, which the authors describe as good short-term repeatability. At 400 volts, the signal-to-noise ratio reached approximately 1900, a level that would matter for resolving low-dose exposures where the signal is faint.
What remains before clinical use
The paper is candid about its limits. The authors state that long-term stability and mechanical durability still need evaluation before the material can be considered for clinical application. Those are not incidental concerns for a device meant to sit in a radiation beam: repeated exposure can alter a polymer's properties over time, and a flexible sensor must survive handling and wear without losing its calibration.
The repeatability figure covers ten measurements, a short window. How the composite behaves over hundreds of exposures, and whether its response stays linear as dose accumulates, are open questions the study does not answer.
There is also the matter of voltage. The strongest results were recorded at a 400-volt bias, which is a substantial operating voltage for a sensor that might one day sit near a patient. Whether useful sensitivity can be retained at lower bias is not addressed in the reported findings.
The authors position the work as an early demonstration rather than a finished instrument, and the framing is consistent with the evidence presented: a fabrication method, a materials characterization, and a set of bench measurements under controlled exposure.
Watch for follow-up studies from the group and others testing whether these silicone–bismuth oxide composites hold their sensitivity and physical integrity over extended use, since that is the threshold any new dosimeter must cross before it reaches a clinic.
