Supercritical carbon dioxide, often abbreviated scCO2, is carbon dioxide held above its critical temperature of about 31 degrees Celsius and its critical pressure of roughly 74 bar. In this state, the gas and liquid phases merge into a single fluid with an unusual combination of properties: it diffuses into materials like a gas but dissolves substances like a liquid. Crucially, its solvent power can be tuned simply by adjusting pressure and temperature, and at the end of any process the fluid simply evaporates, leaving behind a dry, residue-free product. Because carbon dioxide is non-toxic, non-flammable, inexpensive, and generally recognized as safe, it has long attracted attention as a green solvent in industries ranging from decaffeination of coffee to dry cleaning of textiles.

The review’s central message is that scCO2 is not just a niche laboratory curiosity but a technology that can be applied at nearly every stage of implant manufacturing. The first stage is cleaning. Polymer devices fabricated by molding, extrusion, or additive manufacturing can carry residues of organic solvents such as chloroform or tetrahydrofuran, which are toxic and strictly regulated by bodies like the European Medicines Agency. Conventional drying and vacuum treatments struggle to remove these residues from deep within polymer matrices, but scCO2, with its high diffusivity and tunable solvency, can extract them efficiently. Studies cited in the review show that carbon dioxide extraction removed residual chloroform from biodegradable polymers and eliminated solvents from PLGA microspheres, producing products that meet stringent regulatory limits on residual solvents.

Cleaning with supercritical fluids also addresses a subtler challenge: the intricate geometry of modern implants. Devices such as angioplasty catheters and microfabricated components contain narrow channels and microholes that liquid cleaning agents cannot reliably penetrate. Pulse cleaning with supercritical carbon dioxide, in which pressure is cycled to drive the fluid in and out of confined spaces, has been demonstrated for deep microholes and is now used industrially. The review notes that companies in the medical device sector have already adopted supercritical cleaning lines, and pharmaceutical contract manufacturers operate GMP-certified supercritical units, evidence that the technology has moved beyond the pilot scale.

The second major application is impregnation, the loading of drugs or bioactive compounds into polymer implants. Traditional methods involve soaking devices in solvent solutions, which can leave toxic residues and offer limited control over how much drug is loaded and where it ends up. Supercritical impregnation works differently: the drug is dissolved in scCO2, which swells and plasticizes the polymer, allowing the active compound to diffuse deep into the matrix. When pressure is released, the carbon dioxide escapes and the drug remains trapped inside. Because scCO2 is a weak solvent for most polar drugs, small amounts of co-solvents such as ethanol can be added to boost loading, and these co-solvents are far less problematic than the chlorinated solvents used in conventional approaches.

The Marseille group and collaborators have applied this technique to a striking range of devices. Foldable intraocular lenses made of hydrophobic acrylic polymers have been loaded with antibiotics and anti-inflammatory drugs such as gatifloxacin, with in situ spectroscopic monitoring of carbon dioxide sorption and polymer swelling used to control and optimize the process. Contact lenses have been impregnated with antifungal and antibacterial agents for the treatment of keratitis. Bioresorbable suture threads have been loaded with drugs in a proof-of-concept study explicitly aimed at industrial scale-up, and biodegradable ureteral stents have been made to elute ketoprofen to reduce patient discomfort. Wound dressings, electrospun polyurethane fibers, and chitosan sponges have all been successfully impregnated with therapeutic agents ranging from natural extracts to hormones.

The third application is foaming, which is central to tissue engineering. Porous scaffolds provide a three-dimensional template for cells to colonize and regenerate tissue, but conventional methods for creating porosity, such as solvent casting with particulate leaching or freeze-drying, often leave toxic residues or produce poorly connected pore networks. In supercritical foaming, carbon dioxide is dissolved into the polymer under pressure and then rapidly depressurized, causing the gas to nucleate into countless bubbles that expand into pores. The process is solvent-free, operates at relatively low temperatures compatible with biodegradable polyesters, and yields scaffolds with interconnected porosity. Researchers have produced foams of polylactic acid, polycaprolactone, and their composites with hydroxyapatite for bone regeneration, and combined foaming and impregnation in a single step to create drug-loaded porous patches and scaffolds in one continuous operation.

Underpinning all of these applications is a body of physical chemistry that the review examines in detail. Carbon dioxide sorption into polymers lowers the glass transition temperature and, in semicrystalline polymers, can depress the melting point, effects that are exploited in foaming but must be controlled to avoid unwanted deformation during cleaning or impregnation. Techniques such as in situ ATR-FTIR spectroscopy, high-pressure differential scanning calorimetry, and quartz crystal microbalance measurements allow researchers to follow swelling, sorption, and thermal transitions in real time. Solubility data for drugs in scCO2, modeled with density-based correlations and increasingly with machine learning approaches, guide the selection of process conditions. This chemical engineering perspective, the authors emphasize, is what turns promising laboratory results into robust, reproducible industrial processes.

The final and perhaps most consequential application is sterilization. Conventional methods each carry drawbacks for polymer implants: gamma and electron-beam irradiation can degrade polymer chains and alter mechanical properties, ethylene oxide leaves toxic residues and raises environmental and worker-safety concerns, and steam autoclaving requires temperatures that destroy many biodegradable materials. Supercritical carbon dioxide sterilization works at near-ambient temperatures. The mechanism is multifactorial: the pressurized fluid penetrates cells and spores, rapid depressurization can rupture membranes, and carbon dioxide acidifies intracellular environments. Additives such as hydrogen peroxide, water, or ethanol dramatically enhance spore kill rates, allowing the process to achieve the sterility assurance levels required by international standards such as ISO 14937.

Evidence for clinical relevance is accumulating. Decellularized heart valves sterilized with supercritical carbon dioxide retained their biomechanical properties, and a study of anterior cruciate ligament reconstruction using scCO2-sterilized allografts reported favorable clinical outcomes. Sensitive biomaterials including collagen-based products, polysaccharide membranes, and hydrogels have been sterilized at low temperature without loss of structure or bioactivity. The technology has even been applied to inactivate SARS-CoV-2 on personal protective equipment. Regulatory momentum is building: the U.S. Food and Drug Administration has run innovation challenges to identify new sterilization methods, partly driven by concerns over ethylene oxide emissions, and a commercial supercritical sterilization system has received NIH funding for further development. The review concludes that while challenges remain, particularly in validating processes across diverse device geometries and scaling high-pressure equipment, supercritical carbon dioxide stands out as a sustainable, gentle, and remarkably flexible platform that could one day serve as a standard across the whole life cycle of polymer implants, from the first wash to the final sterile pouch.

Subject of Research: Use of supercritical carbon dioxide in the cleaning, impregnation, foaming, and sterilization of implantable polymer-based medical devices

Article Title: Implantable Polymer-Based Medical Devices Go Supercritical: From Cleaning to Sterilization

Article References: Laggoune, Z., Masmoudi, Y., Pepe, V., & Badens, E. (2026). Implantable Polymer-Based Medical Devices Go Supercritical: From Cleaning to Sterilization. Annals of Biomedical Engineering. https://doi.org/10.1007/s10439-026-04392-3

Image Credits: AI Generated

DOI: 10.1007/s10439-026-04392-3

Keywords: supercritical carbon dioxide, implantable medical devices, biomedical polymers, sterilization, drug impregnation, foaming, tissue engineering scaffolds, green solvents, residual solvent removal, intraocular lenses, biodegradable polymers, chemical engineering

The research team, led by Neha Sharma of the University Hospital Basel and Duke-NUS Medical School together with colleagues at the University of Basel, Iuliu Hatieganu University of Medicine and Pharmacy in Romania, and the University Center for Dental Medicine Basel, set out to answer a deceptively simple question: what happens to the dimensional accuracy and mechanical strength of modern 3D-printed photopolymer resins when they are exposed to clinically realistic sterilization cycles? The stakes are higher than they might appear. Standard steam autoclaving runs at 121 degrees Celsius, but prion inactivation protocols demanded by infection-control authorities require an extended cycle holding a searing 134 degrees Celsius for at least 18 minutes. That thermal assault had never been systematically evaluated on the contemporary biocompatible resins now flowing through hospital 3D print labs.

To close that gap, the investigators selected four commercially available biocompatible photopolymer resins: P pro Surgical Guide Clear, ProArt Print Splint, BioMed Clear Resin, and BioMed Amber Resin. Two additive manufacturing platforms were represented, stereolithography, which cures liquid resin with a precisely steered ultraviolet laser, and digital light processing, which cures entire layers at once using a projected light pattern. The experimental design deliberately worked at two scales. First, the team printed 240 standardized ISO specimens for rigorous benchtop testing of dimensional accuracy and flexural strength, the classic three-point bending measure of how much load a material can carry before fracturing. Second, they printed 72 clinically relevant patient-specific guides representing two real surgical scenarios: cranial resection guides and oral implantology guides. This dual approach ensured the findings would speak both to materials scientists and to surgeons.

The specimens were randomly allocated to three experimental groups. A baseline group remained non-sterilized, a steam group underwent the extended prion-inactivation autoclave cycle at 134 degrees Celsius for 18 minutes, and a plasma group was treated with low-temperature hydrogen peroxide plasma sterilization at a gentle 55 degrees Celsius for 19 minutes. Dimensional accuracy was quantified using root mean square deviation analysis, a method that compares the printed object against its digital design point by point and expresses mismatch as a single deviation value in micrometers. Statistical evaluation relied on two-way analyses of variance followed by Bonferroni post-hoc testing, with the significance threshold set at 0.05.

The accuracy results told a nuanced story. Across the board, deviation was lowest for the simple ISO specimens, averaging 61 plus or minus 24 micrometers, followed by the cranial guides at 119 plus or minus 29 micrometers and the oral implantology guides at 147 plus or minus 45 micrometers. Geometry, in other words, matters: the more complex and curved the anatomy a guide must capture, the harder it is to print true. Material selection proved equally decisive. The ProArt Print Splint and P pro Surgical Guide Clear resins delivered significantly higher dimensional accuracy than the two BioMed resins across every geometry tested, with the difference reaching statistical significance at p less than 0.001. Sterilization method also significantly shifted dimensional accuracy overall, although the magnitude and direction of the change depended on the shape of the object, confirming that there is no single sterilization penalty that applies uniformly to every printed device.

The mechanical findings delivered the study’s most dramatic surprise. Counterintuitively, steam sterilization increased the flexural strength of the ISO specimens to 41 plus or minus 16 megapascals, compared with a baseline of 33 plus or minus 19 megapascals, while plasma sterilization reduced strength to 29 plus or minus 16 megapascals, a statistically significant divergence at p less than 0.001. The likely explanation lies in polymer physics: the intense heat of autoclaving drives additional post-cure crosslinking within the photopolymer network, effectively finishing a polymerization reaction that room-temperature light curing had left incomplete. Steam exposure, in effect, acts as an aggressive post-cure. But that apparent strengthening came with a catastrophic catch that standardized specimens alone could never have revealed.

When the researchers turned to the clinically relevant patient-specific guides, the picture darkened. Every single steam-sterilized guide printed from BioMed Clear and BioMed Amber resins developed visible cracks across its structure. These were not hairline cosmetic flaws; they were structural failures that rendered the devices clinically unusable, undermining exactly the geometric fidelity that makes surgical guides worth printing in the first place. The thinner, anatomically contoured walls of the guides behave differently from the stocky ISO bars, concentrating thermal stresses from rapid heating and cooling into fracture lines. The lesson is one the field is learning repeatedly as additive manufacturing enters medicine: benchtop material data cannot be extrapolated naively to real device geometries. A resin that tests beautifully in standard specimens can fail in the operating room.

Plasma sterilization told a more cautious tale. The low-temperature hydrogen peroxide treatment avoided crack formation entirely; all plasma-treated guides remained structurally intact. Yet the method exacted its own price in reduced flexural strength of the specimens, and the authors flag a further unresolved regulatory and microbiological question: hydrogen peroxide clearance, the ability of the sterilant to penetrate the device and then dissipate without leaving cytotoxic residues, has not been validated for these particular resins. The researchers therefore stop short of endorsing plasma sterilization as a default substitute for autoclaving. An intact guide is worthless if residual oxidant could compromise tissue healing, and the data simply do not yet exist to rule that risk out.

The study’s central conclusion is a shift in how surgical 3D printing workflows should be conceived. Material and sterilization method, the authors argue, must be selected and validated as a pair, matched both to the specific sterilization cycle available at the point of care and to the mechanical and accuracy demands of the intended procedure. A high-accuracy resin paired with an incompatible sterilization cycle is not a workflow; it is a latent failure waiting to reach the operating theater. For hospital print labs producing guides under time pressure, the practical message is to consult validated material-sterilization combinations before a case, not after a guide emerges from the autoclave visibly fractured on the morning of surgery.

Beyond the immediate clinical implications, the work resonates with a broader trend in biomedical additive manufacturing. As regulators, including those administering the European Union Medical Device Regulation, scrutinize hospital-produced devices ever more closely, standardized evidence of how fabrication variables interact with post-processing steps becomes essential. This study provides exactly that kind of evidence for the prion-inactivation steam cycle, an extreme but legally mandated condition that most prior research ignored. It also highlights the enduring value of testing devices and specimens side by side. The 61 micrometer accuracy of an ISO bar, the 147 micrometer deviation of an implantology guide, the crosslinking that strengthens a specimen while splitting a guide apart: these are the granular, quantitative facts on which the safety of point-of-care 3D printing will ultimately rest. As hospital printers multiply from Basel to Bangkok, the pairing of resin and sterilization cycle may prove to be one of the most consequential choices a surgical team never knew it was making.

Subject of Research: Effects of steam and plasma sterilization on the accuracy and strength of 3D-printed surgical guide resins

Article Title: Dimensional accuracy and flexural strength of 3D-printed photopolymer resins for point-of-care surgical guides after steam and plasma sterilization: an in-vitro comparative study

Article References: Sharma, N., Burde, A. V., Manea, A., Fischer, J., Thieringer, F. M., & Rohr, N. (2026). Dimensional accuracy and flexural strength of 3D-printed photopolymer resins for point-of-care surgical guides after steam and plasma sterilization: an in-vitro comparative study. 3D Printing in Medicine. https://doi.org/10.1186/s41205-026-00352-8

Image Credits: AI Generated

DOI: 10.1186/s41205-026-00352-8

Keywords: 3D printing, surgical guides, sterilization, photopolymer resins, dimensional accuracy, flexural strength, steam sterilization, plasma sterilization, craniomaxillofacial surgery, dental implants, point-of-care manufacturing, biomaterials

The appeal is obvious. Commercial multiwell plates are inexpensive enough at small scale, but custom formats, unusual geometries, and specialized fixtures for microscopy or mechanical testing are either unavailable or prohibitively costly. As 3D printers have proliferated in wet laboratories, researchers have increasingly printed jigs, holders, and even culture chambers on demand. The catch is that a printed object is not automatically a culture device. It must survive sterilization, remain mechanically stable, and, crucially, prove non-toxic to living cells. Those three requirements, the new work shows, do not always travel together.

The team, led by Sergio Noé and Núria Gavara of the Universitat de Barcelona with collaborators at the Universitat Politècnica de Catalunya-BarcelonaTech, focused on two of the most common 3D printing materials: Poly-Lactic Acid (PLA) filament, the workhorse of fused filament fabrication, and Acrylonitrile Butadiene Styrene (ABS)-like photopolymer resins used in vat photopolymerization. Specimens printed from each material were subjected to the three sterilization methods most familiar to cell biologists: ultraviolet light exposure, immersion in ethanol, and steam autoclaving. The researchers then interrogated the treated samples at both macroscopic and nanoscopic scales, using tensile testing to measure bulk mechanical behavior and nanoindentation to probe local stiffness and surface integrity.

The results revealed a striking material-specific pattern of vulnerability. ABS-like resin, which measured 0.31 ± 0.07 GPa in stiffness in its baseline state, remained mechanically viable after UV illumination, which actually raised its measured modulus to 1.19 ± 0.02 GPa, and after autoclaving, at 0.36 ± 0.05 GPa. Ethanol, however, proved catastrophic: submerged specimens deteriorated to a modulus of just 0.01 ± 0.00 GPa, effectively losing their structural integrity. For a material that otherwise tolerates heat and radiation, this solvent sensitivity is a decisive limitation, since ethanol immersion is one of the simplest and most ubiquitous sterilization methods in any laboratory.

Mechanical survival, however, turned out to be only half of the story. Even when ABS-printed devices were successfully sterilized by methods they could tolerate, they proved toxic to cultured cells. Switching to Formlabs Grey resin, an inert photopolymer, did not rescue the situation; the cytotoxic effect persisted. This finding carries a cautionary message for the growing community of laboratory makers: a printed device can pass every mechanical test and still quietly kill the cells it is meant to house. Residual monomers, unreacted photoinitiators, or surface chemistry introduced during printing and post-processing may leach into culture medium in ways that standard material characterization does not detect.

PLA told a different story. Untreated PLA specimens measured 1.22 ± 0.09 GPa and remained essentially unchanged after ethanol immersion, at 1.23 ± 0.15 GPa, and after UV exposure, at 1.21 ± 0.01 GPa. The autoclave, however, was its undoing: steam sterilization dropped the modulus to 0.75 ± 0.20 GPa, a substantial degradation consistent with the hydrolytic and thermal sensitivity of PLA at autoclave temperatures. Here the researchers found an elegant workaround. By switching to Heat-Treated PLA, or HTPLA, a filament that has been thermally annealed to improve its thermal and dimensional stability, the autoclave problem disappeared. HTPLA specimens retained a stiffness of 1.07 ± 0.13 GPa after autoclaving, remaining well within the range compatible with structural use in a culture device.

With a material that could survive the full sterilization arsenal in hand, the team assembled complete glass-bottomed multiwell devices and subjected them to a sequential sterilization protocol combining autoclaving, ethanol treatment, and UV illumination. The resulting platforms were then validated in the most direct way possible: by growing cells on them. Using A549 human lung epithelial cells, the researchers confirmed that the sterilized HTPLA devices supported healthy long-term culture. Cell morphologies appeared normal, and the cells proliferated with a doubling time of 21.06 ± 4.66 hours, a figure consistent with expectations for this cell line in standard culture vessels. The devices also proved compatible with immunostaining workflows, extending their utility beyond simple culture to fixed-cell imaging and molecular labeling experiments.

The combination of a glass optical bottom with a printed polymer body is central to the design’s value. Glass remains the gold standard surface for high-resolution microscopy, offering optical clarity, low autofluorescence, and well-characterized cell adhesion properties that many printed polymers cannot match. By bonding standard glass coverslips into printed well frames, the researchers created devices that behave optically like commercial glass-bottom dishes while retaining the geometric freedom of additive manufacturing. Laboratories can now print multiwells with custom well counts, spacings, or integrated features tailored to specific microscopes, assays, or experiments, and sterilize them with equipment already present in the facility.

The broader significance of the work lies in its systematic approach. Rather than assuming that any sterilization method will suit any printing material, the study provides quantitative mechanical data across a matrix of material-sterilization combinations, exposing failure modes that would otherwise be discovered the hard way, through warped devices, failed cultures, or unexplained cell death. It also demonstrates that the sterilization protocols already standard in cell biology laboratories, UV light, ethanol, and autoclaving, can be applied directly to 3D-printed materials when the material is chosen wisely. The authors frame their contribution as a set of open designs and protocols: in-house 3D-printing and assembly instructions for glass-bottomed multiwells based on HTPLA, paired with optimized sterilization sequences validated for long-term culture and immunostaining.

For laboratories weighing the cost of specialized culture formats, the message is empowering but disciplined. A few hundred euros of printer and filament can replace custom-machined or commercially unavailable devices, but only if researchers respect the material science underneath. Ethanol will destroy ABS-like resins; the autoclave will weaken standard PLA; and even a mechanically sound, sterilized resin print may still poison cells. Heat-treated PLA, processed through a deliberate sequence of autoclave, ethanol, and UV sterilization, currently offers the most reliable path from the printer bed to the incubator. As 3D printing continues its march into wet laboratories, studies like this one supply the evidence base that turns a promising workshop trick into dependable laboratory practice, allowing researchers to print, sterilize, and culture with confidence that their custom devices will protect, not compromise, the cells within.

Subject of Research: Design and sterilization of 3D-printed glass-bottomed multiwell devices for cell culture and immunostaining

Article Title: Design and sterilization of 3D-printed glass-bottomed multiwells for cell culture and immunostaining

Article References: Noé, S., Barberá-Flichi, F., Sanz-Fraile, H., Padilla, J. A., Jorba, I., Buj-Corral, I., Xuriguera, E., Jiménez-Piqué, E., & Gavara, N. (2026). Design and sterilization of 3D-printed glass-bottomed multiwells for cell culture and immunostaining. Applied Microbiology and Biotechnology. https://doi.org/10.1007/s00253-026-14032-4

Image Credits: AI Generated

DOI: 10.1007/s00253-026-14032-4

Keywords: 3D printing, cell culture, sterilization, HTPLA, PLA, ABS-like resin, glass-bottomed multiwells, immunostaining, additive manufacturing, biomaterials, nanoindentation, autoclaving