Buying MEA plates and electrode arrays: format, reuse and the cost you cannot see before you commit

The buying problem in this category is unusually clean to state: every multielectrode array vendor publishes a complete electrode specification and none of them publishes a price. You can find out the pitch, the sensing area, the electrode material and the well count in about ninety seconds. You cannot find out what a plate costs without a sales conversation, and by the time you are having that conversation you have usually already chosen the platform, which is precisely when your negotiating position is worst.

That matters more here than in most consumable categories, because the plate is the largest lifetime cost of an electrophysiology programme, not the instrument. A laboratory running one 48-well plate a week spends more on plates over three years than most groups spend on their first recording system. The instrument is the visible decision. The plate is the larger one, and it is the one made blind.

This page is the procurement side of that decision. The physics comparison, resolution against throughput, which platform answers which question, lives on the MEA platforms comparison and the compatibility matrix. What follows is about money, contract structure and what to put in a request for quotation.

What we could evidence, and what it is worth

We could not find a list price for any MEA plate from any vendor. So we went looking for transactions instead, and there are two kinds worth having.

Derived figures from US federal contract awards. Where a public award record names the product and states the quantity, dividing the obligated amount by the quantity gives a per-unit figure. These are real transaction prices for that buyer on that date. They are upper bounds, not list prices: the obligation may include shipping, handling or line items the description does not break out, and federal buyers may hold negotiated schedule pricing that other customers cannot access. Treat every derived number below as a ceiling you should be able to beat, not as a target you should be pleased to hit.

Stated unit prices from a published order confirmation. Stronger evidence, because the supplier wrote the number. Rarer, because it depends on a buyer being subject to a contracts-transparency regime.

The two Axion rows are three years apart and differ by 28 percent, and we are not going to present that as a price-inflation series: the 2024 record names a product number rather than a plate variant, and the two obligations may bundle different things. What the pair establishes robustly is that a 48-well MEA plate is a high-three-figure item. Budgets built on an assumed $150 plate are wrong by roughly a factor of three, and that error compounds every week of the programme.

The same MCS order confirmation carries a term that is easy to miss and expensive to discover: a minimum order value of EUR 500.00 net, with an additional EUR 25.00 charge on any order below it. A laboratory that orders two arrays at a time is paying a surcharge on most of its orders.

Format is the decision, not price

The strongest reason not to shop these on price is that the formats are not substitutes. Three representative configurations, all manufacturer-published:

Those are different experiments. A 96-well plate at 8 electrodes per well and a high-density array at 26,400 electrodes per well are not a cheap option and an expensive option; they answer different questions and one of them cannot be substituted for the other by spending more or less money.

The decision rule that survives contact with a real programme: if the endpoint is a population-level pharmacological response across many conditions, buy well count. If the endpoint is what an individual neuron or a network topology is doing, buy electrode density. Groups that try to do both on one platform generally end up doing neither well, and the plate spend on the wrong format is unrecoverable.

The cross-site CiPA pilot study on iPSC-derived cardiomyocytes is a useful reference point for what population-level multiwell MEA can deliver when the protocol is controlled: it is a blinded, multi-laboratory reliability exercise, and its existence is the reason multiwell MEA is taken seriously as a safety assay rather than as a descriptive one. At the other end, the high-density work embodying cultured neurons in a simulated game-world depended on routing stimulation to specific electrodes, a capability that only exists on a high-density platform, and that no amount of well count substitutes for.

The two commercial structures that surprise buyers

Neither appears on any specification page. Both change the total materially.

Consumables sold as a subscription rather than by the box. The MaxWell record is a single $19,600 obligation described as a plate consumables subscription. That is an annual commitment, not a unit price, and it behaves completely differently in a budget: you cannot flex it downward in a slow quarter, and if you under-consume you have paid for plates you did not use. Ask explicitly whether a subscription is the only route, what the annual commitment is, and what happens on under-consumption.

Assay capability licensed as separate software. The same buyer obligated $12,450 for stimulation assay software under a separate award. A stimulation capability you assumed came with the instrument may be a licence. This is the single most common cause of a platform costing more in year two than the quote implied.

Ask for four separate lines in every quote: instrument, plates at three annual volumes, software with the module list, and the service contract as an annual figure. A vendor who will not break those out is telling you something.

Reuse changes the three-year number more than negotiation will

This is the largest single variable in the category and it is a property of the platform, decided before you buy anything.

A reusable glass array at EUR 328.00 used twenty times is EUR 16.40 per experiment. A single-use 48-well plate at the derived upper bound of $488.80, run at four wells per condition with a vehicle and a positive control, is a ten-condition experiment at roughly $49 per condition, every time you run it. Over three years that gap is larger than most instrument price differences between platforms, and it never appears in a platform comparison because the plate prices needed to calculate it are not published.

Reuse is not free. Cleaning takes technician time, carries a contamination risk, and progressively degrades electrode impedance. But it is a cost you control, and single-use is a cost you do not. If reuse is on the table, the line to get in writing is how many cleaning cycles the vendor will stand behind on electrode impedance, because that is the number that converts a reuse claim into a budget.

What separates the vendors, in procurement terms

• Axion BioSystems. The most complete public specification in the category, catalogue numbers, well counts, electrodes per well, bottom material, wall colour, and no prices at all. The broadest published multiwell range, which is why it dominates pharmacological screening. The plate is single-use by design, so model the full recurring cost before you commit. Both derived figures on this page are Axion, which makes it the only vendor here with two independent transaction anchors.
• Multi Channel Systems. Operates a public online shop, and then restricts it to its direct sales territory of Europe, Australia and New Zealand and requires an account. Glass single-well arrays that are routinely cleaned and reused, which is the strongest structural argument in the category for a low-volume laboratory. The published order confirmation makes MCS the only vendor with a stated unit price in evidence, and also the only one with a documented minimum order value.
• MaxWell Biosystems. High-density arrays with a published specification that is genuinely unmatched at 26,400 electrodes per well and 17.5 µm pitch. Commercially the most complex: the one vendor where we found consumables sold as a subscription and assay software licensed separately. Model those as two recurring lines, not one.
• 3Brain and Alpha MED Scientific. Both publish complete specifications and no prices. Alpha MED’s MED Probes are reusable, which puts them in the same structural category as MCS for low-volume work.

Total cost, honestly

The plate is the visible recurring item. It is not the whole recurring cost, and every figure on this page is an underestimate of what an experiment actually costs to run. What we could not attach to a verified price from a named source, and therefore do not state:

• Coating. Poly-D-lysine, PEI, poly-ornithine, laminin. Mandatory on every plate, every time. The published Axion 24-well co-culture protocol work uses a specific coating regime alongside a defined neuron-to-astrocyte ratio, and neither the reagent nor the labour is free.
• Media across maturation. A neuronal network is not recordable on the day it is plated. The published co-culture characterisation records spontaneous firing at 11 days in vitro with network synchrony developing over roughly six weeks; a 3D high-density MEA study reports spontaneous firing and network bursting from seven-week-old constructs. Half-changes every two to four days across that window frequently cost more than the plate.
• Cells, including glia. Astrocytes are effectively mandatory for network activity, which makes every MEA experiment a two-vial purchase. See MEA-ready neurons.
• Service contracts. Quoted separately, and annual.
• Incubator occupancy and technician time across a four-to-seven-week maturation, which is a standing commitment rather than a project task.

The alternative that is frequently cheaper

If what you need is the recording rather than the capability, one institutional core publishes a rate card for exactly that: $49.00 internal, $64.00 external academic and $84.00 external industry per MEA recording, with a project planning fee of $290.00, $377.00 and $493.00 respectively per new project. Those are published figures captured 2026-09-01.

Set $64.00 for an external academic recording against roughly $49 per condition in consumable alone, before instrument, technician, coating, media or maturation, and the arithmetic for buying the service below about one plate a month is not close. The project planning fee is the entire fixed cost of finding out. We treat this seriously enough to give it its own page: electrophysiology characterisation as a service.

What to put in the RFQ

Ten lines. The first three are technical and the rest are commercial, which is the ratio that usually gets inverted.

1. Platform and instrument model. Axion Maestro Pro, Edge or Volt; MaxWell MaxTwo or MaxOne; MCS MEA2100; 3Brain; Alpha MED.
2. Plate format and vendor catalogue number, not a description. CytoView MEA 6, 12, 24, 48 or 96 are different parts.
3. Electrodes per well required, and whether single-cell resolution or population activity is the endpoint. This is the question that decides the platform.
4. Bottom material and wall colour, transparent SU-8, glass, white or black polystyrene, and whether optical multiplexing is needed alongside the electrical recording.
5. Single-use or reusable, and if reusable, the number of cleaning cycles the vendor will stand behind on electrode impedance.
6. Annual plate volume quoted at three quantities, so the discount curve is visible. One quantity gives you one number and no information.
7. Whether a consumables subscription is offered, its annual commitment and the under-consumption penalty.
8. Whether analysis software is perpetual or subscription, and which assay modules are separately licensed. Name stimulation explicitly.
9. Coating requirement and whether the vendor applies it.
10. Minimum order value and freight terms. The MCS document shows why this belongs in the RFQ rather than on the invoice.

Two more that are not price lines but decide schedule: lead time, and whether plates are made to order. Two to eight weeks is normal in this category, and a stock-out does not delay an experiment, it ends a maturation you cannot restart.

What the datasheet does not settle

Four things this page deliberately does not claim. Nobody publishes a list price, so nothing here is one. No derived figure is a quote. No comparison here establishes which platform produces better biology, because no head-to-head study across all five under matched conditions was found. And the per-condition arithmetic excludes everything except the plate, which means the real cost of an MEA experiment is higher than any number on this page.

How we can help

We source MEA consumables against a named platform and well format, and we bring the derived transaction data into the conversation so the quote can be checked against something rather than accepted against nothing. If you tell us the platform, the format and the plates you expect to consume in a year, we will put the same specification to the vendor and to their distributors, ask for the curve at three volumes rather than a single-quantity price, and tell you where the recurring cost lands against the alternatives, including when the honest answer is that a core facility recording at $64 beats owning anything.

The paired datasheet, with the full specification field list and supplier evidence notes, is at MEA plates and electrode arrays. Send us the specification.