Buying organ-on-chip plates and chip consumables: price per chip, chip material, and the ecosystem you cannot leave

Three of the major organ-chip platforms publish no pricing of any kind. That would be an inconvenience in most consumable categories. Here it is a structural problem, because there is no second source for an OrganoPlate, a Chip-S1 or a PhysioMimix plate. Choosing a platform commits you to one vendor’s consumable for the working life of the assay, and you make that commitment without ever seeing the number that determines whether the platform is affordable to run rather than merely affordable to own.

The purchase decision therefore has to be reframed. You are not buying a plate. You are buying a per-chip cost, multiplied by every condition you will ever run, from a supplier you cannot switch away from. This page assembles what can actually be evidenced about that number, states plainly which figures are list prices and which are derived upper bounds, and sets out the specification lines that make quotes comparable across platforms with wildly different formats.

The published prices

AIM Biotech is the transparent end of this market and deserves credit for it: the full range carries live public list prices, including, unusually, both the empty and the pre-seeded version of the same plate. All figures below were retrieved from the public store API on 2026-09-01 and are USD list prices.

Two things fall out of that table that you will not get from any other vendor in the category.

The assay-ready uplift is measurable. AngioPrime is an idenTx 40 plate pre-loaded with endothelial cells at $1,470.00 against $444.00 for the empty plate, an uplift of $1,026.00, or 231 percent. That is a real, published number for what it costs to buy the plate already seeded rather than seeding it yourself. Whether it is worth paying depends entirely on what a failed seeding costs you in calendar time, but at least here the trade is visible.

Volume discounting is modest and visible. organiX moves from $410.00 at box-of-one to $370.00 each at box-of-three, a 9.8 percent reduction. That is the shape of the curve in this category: real, but not the order-of-magnitude effect buyers sometimes assume.

MatTek (Sartorius) distributes the AIM Biotech range with its own published prices, which lets the distributor spread be measured directly rather than guessed at: a consistent 7.4 to 8.9 percent across eight SKUs. If you are routing through a distributor for procurement-system reasons, that is roughly what it costs, and it is worth pricing both ways before you decide the convenience is free.

The derived figures, and what they are not

MIMETAS and Emulate publish nothing. What exists instead are US federal contract awards where the record names the product and states the quantity. Dividing the obligation by the quantity gives a per-unit figure.

Say this out loud before using any of them: these are DERIVED from named federal awards and they are UPPER BOUNDS, not list prices. The obligated amount may include shipping, warranty, training or line items the description does not break out, and federal buyers may hold negotiated schedule pricing that is unavailable to other customers. A derived figure is a ceiling you should be able to argue below, not a benchmark you should be relieved to match.

The MIMETAS pair is the strongest derived evidence on this site, for a reason worth stating: two independent awards, different years, different order sizes, produce exactly the same per-plate figure. Twenty-five plates and twenty plates both derive to $1,163.00. That consistency does not make it a list price, it is still an upper bound and still may include freight, but it does make it very unlikely to be an artefact of one unusual transaction.

At 64 chips per plate on the Graft format, $1,163.00 per plate is $18.17 per chip. That is the number to compare against AIM Biotech’s $11.10, and it is the comparison that no vendor specification page will ever let you make.

Where we decline to derive. Emulate instrument pricing cannot be derived responsibly from the records we found. Two awards describing an identical two-Zoe-CM2-plus-two-Orb configuration differ by 48 percent, $218,100 under PIID 75F40122P00202 in 2022 against $146,915 under PIID 75F40123P00171 in 2023. When the records disagree that badly about the same configuration, the honest output is an envelope rather than a number: the awards establish roughly $147,000 to $354,000 for a Human Emulation System installation and nothing more precise. We are not going to pick the convenient one.

Chip material is a scoping decision, not a detail

This is the technical fact that should govern platform choice for anyone doing compound work, and it is routinely treated as an implementation detail.

Kemas and colleagues fabricated identical microdevices from eight different polymers and measured compound absorption. The absorption differed by more than 1000-fold across the materials. More than 99 percent of chlorpromazine partitioned into the PDMS bulk within 24 hours, against roughly 1500-fold lower absorption in PMMA and in thiol-ene epoxy.

Read that as a procurement statement rather than a materials-science one: if your compound is lipophilic and your chip is PDMS, the concentration your cells see is not the concentration you dosed, and the discrepancy is large enough to invert a dose-response conclusion. Emulate sells Chip-R1 explicitly on a low-absorption profile for ADME and toxicology work, which is the vendor acknowledging the same problem.

So the RFQ line is: name the chip material, and justify it against the compound class. Not “what is the chip made of” as a curiosity, but a written justification that the material is appropriate for what you intend to dose. A vendor who will not put that in writing has told you something. The full material matrix is on PDMS vs COP vs PMMA.

What separates the platforms, in buying terms

• AIM Biotech. Publishes everything, including the assay-ready uplift. The idenTx architecture is a 3D gel region flanked by media channels; idenTx 40 carries 40 chips per plate, idenTx 9 carries 9, and idenTx 3 is a microscope-slide format with 3 experiments per chip. Lowest per-chip cost in evidence at $7.60 to $11.10. Also the platform where you can model the recurring bill before committing, which is worth something on its own.
• MIMETAS. Publishes complete platform specifications and no prices. The OrganoPlate Graft 3-lane format carries 64 chips per plate with apical and basolateral access, is membrane-free using surface-tension PhaseGuide patterning, sits on a 150 µm microscope-grade glass bottom in a microtitre footprint, and is perfused by the OrganoFlow rocker rather than by pumps or tubing. Removing the membrane removes it from the optical path, which is the substantive argument for the architecture. Removing tubing removes the commonest overnight failure modes.
• Emulate. Proprietary chips tied to the Zoe culture module. Chip-S1 additionally applies cyclic mechanical strain, which no plate-format platform does; Chip-R1 is sold on low absorption. The most closed ecosystem in the category, and the one where instrument pricing is least derivable. The published Liver-Chip performance and economic assessment by Ewart and colleagues is the most substantial third-party economic analysis anyone has published on an organ-chip platform, and is worth reading before a purchase decision on this scale.
• CN Bio. Publishes no consumable pricing and we derive none here.

Per-experiment arithmetic, with the caveats attached

A per-chip figure is not a per-experiment figure. Divide by replicates and controls first.

MODELLED, this is arithmetic on the figures above, not a quotation. And it excludes cells, media, matrix, coating, imaging time and technician time, every one of which is real. The point of the table is that the platforms are within a factor of two of each other on consumable, which means consumable cost alone should rarely be the deciding factor. What should decide it is format, material and whether the architecture answers your question.

The two things buyers most often miss

Assay-ready plates ship on a production calendar, not on demand. A pre-seeded plate contains living cells, which means the vendor makes it to a schedule. Ordering one three days before you need it is not a purchasing problem you can solve with expedited freight. Build the production cadence into the project plan.

Instrument, module, software licence and service contract are four lines. Ask for them separately. The Emulate award records show why: a Zoe culture module update was procured as its own award, which means the module and the instrument are separately purchasable and therefore separately priceable. A single bundled number hides which of the four is escalating at renewal.

What to put in the RFQ

1. Tissue model and the endpoint the chip must support: barrier, perfusion, vascularisation, grafting, ADME.
2. Chips or wells per plate required, and conditions per year after replicates and controls.
3. Chip material, justified against the compound class. PDMS, COP, PMMA, thermoplastic, glass-bottom.
4. Perfusion mechanism the consumable requires: rocker, pumped recirculation, or none. See perfusion hardware.
5. Imaging requirement and bottom thickness.
6. Empty plate or assay-ready pre-seeded, with the uplift stated and accepted.
7. Direct purchase or distributor routing, priced both ways. The measured spread on this range is 7.4 to 8.9 percent.
8. Volume curve at three annual quantities rather than one.
9. Instrument, module, software licence and service contract as four separate lines.
10. Whether a finished-tissue alternative would answer the question at lower total cost.

Line 10 is the one that saves the most money and is asked the least. If the question is “does this compound damage hepatocytes”, a ready-made liver microtissue in a plate may answer it without a platform commitment at all. See liver microtissues.

What this page does not establish

It does not rank the platforms on biological performance; no head-to-head study across all of them under matched conditions was found. It does not give a list price for any MIMETAS or Emulate product, because none exists publicly. It does not derive Emulate instrument pricing, for the reason stated. And every per-experiment figure excludes the largest costs in a real programme, cells, media and people.

How we can help

We quote these per chip rather than per box, so platforms with 9, 40 and 64 chips per plate can actually be compared. We hold the derived procurement anchors and will use them to sanity-check a quote you have been given. And we will ask the questions that are awkward to ask a vendor you are about to become dependent on: the volume curve at three quantities, the distributor spread priced both ways, and whether the four commercial lines are really four lines.

The paired datasheet, with the full specification field list and supplier evidence notes, is at organ-on-chip plates and chip consumables. Send us the specification.