North American Electronics Adhesives, Tapes and Thermal Materials: Supply Base and Application Map

There is no adhesive on the bill of materials. Open the costed BOM for a router, a handheld scanner, a fast charger, or a patient monitor and you will find the processor, the board, the connectors, the enclosure, and the display module, each with a part number and a supplier. The thermal pad under the processor, the conformal coating over the board, the conductive gasket around the shield, the tape holding the display in its frame, and the stretch-release strip under the battery are there too, in ones and twos of a percent, spread across half a dozen line items nobody owns.

That invisibility is the commercial fact of this lens, and it cuts both ways. It means nothing here is a headline negotiation, and that suppliers who show up with a price list get ignored. It also means these materials are specified by engineers rather than bought by category managers, which is a far better conversation to be in if you can get to it.

The part that is not small is thermal. Power density in North American electronics is rising faster than the mechanical means of dealing with it. A data center accelerator board dissipates more in a slot than a whole desktop did fifteen years ago. A 350 kW DC fast charger puts kilowatts of loss into a cabinet that has to survive a Phoenix summer. An industrial motor drive, a solar inverter, a 5G radio unit, and a laptop chassis are all running into the same wall: the heat sink cannot get bigger, the fan cannot get louder, and the only remaining variable is how efficiently heat crosses the interfaces between the silicon and the metal. That is what a thermal interface material does, and it is why thermal gap fillers, greases, phase change preforms, and graphite spreaders are the fastest growing category in this report by a wide margin.

The second thing worth knowing up front is who signs. In automotive, the company that specifies a structural adhesive usually also buys it. In electronics it almost never does. The material gets qualified during new product introduction at the brand owner, whose engineers write it onto an approved materials list, and then it gets purchased in volume by the electronics manufacturing services provider that actually builds the product, usually through a distributor, against a part number the brand owner chose. Apple does not buy tape. Jabil buys tape, to Apple specifications, in Guadalajara.

Third: this is a converter market as much as a chemistry market. Almost nothing in this lens ships to an assembly line as a roll. It ships as a die-cut part with a part number, a liner, a kiss cut, and a tab, kitted in the order the operator needs it. The companies that do that work, Boyd, Marian, JBC Technologies, CGR Products, are a tier of their own, and a tape producer that has not built those relationships is selling into a channel it does not control.

Finally, volumes here are strange. A consumer program can consume more square metres of one die-cut part in a quarter than an entire industrial product line will in five years, and then end. An industrial or medical program moves a hundredth of the volume and runs for a decade with a change-control process that makes displacement nearly impossible. Those are two different businesses with two different sales motions, and the same material sells into both.

Four tiers, and the first one is unusual enough to be worth stating plainly: Tier 1 in this lens is not a systems integrator. It is the contract manufacturer that builds the product, plus the converters that feed it parts.

The brand owner sits at the top and does not buy. Fourteen OEM organizations hold position rows in this industry, from Apple, Dell, HP, and Cisco through Honeywell, Rockwell Automation, Emerson, and Zebra to L3Harris and Stryker. Most of them manufacture very little themselves. What they own is the design, the specification, and the approved materials list, and that last item is the entire commercial question on this lens. An engineer at the brand owner decides during new product introduction that a given gap filler, coating, or tape is the material for a given interface. That decision is recorded in a drawing, a spec, and an approved vendor list, and it persists for the life of the platform.

Tier 1 is the EMS provider, which holds the purchase order. Jabil, Flex, Sanmina, Benchmark Electronics, and Plexus build the products the brand owners sell. They run the dispensing equipment, the selective coating machines, the pick-and-place heads that lay die-cut thermal pads, and the potting lines. They hold the supplier relationship, the incoming inspection, and the payment terms. They also have their own opinion about materials, which shows up as a substitution request the moment a specified material causes a yield problem or a lead time problem. A supplier who is on the brand owner drawing but unknown at the EMS finds out about substitutions after they happen.

Tier 1 also includes the converters. Boyd, Marian, JBC Technologies, and CGR Products buy roll and sheet stock from the material producers and ship the EMS a part: die-cut, kiss-cut, on a liner, in a kit, in the right orientation. Boyd is the largest and is more than a converter, because it also designs thermal solutions, which puts it in the specification conversation rather than downstream of it. The others compete on tooling speed, tolerance, and how much assembly work they will take on. For a material producer these companies are the channel, not the competition.

Tier 2 is the material producers, and it is crowded at the top and thin at the edges. 3M and Henkel are the two broad lines: between them they cover thermal, conductive, bonding, coating, and label materials, and they are the default on most approved materials lists. Dow sells Dowsil silicones as formulated products. Parker Chomerics and Laird Performance Materials sell EMI and thermal together, which is a genuine differentiator because those two problems are solved at the same seam. Then there are the specialists: Fujipoly on thermal, Saint-Gobain on foam gaskets, Dexerials on display bonding, Master Bond on small-lot high-specification epoxies, and Nitto, tesa, and Avery Dennison on tape.

Tier 3 is silicone chemistry, and it is short. Dow, Momentive, Wacker, Shin-Etsu, and Kaneka supply the polymers, fluids, gums, and bases that the formulators above turn into products. The concentration matters: a very large share of what this lens sells is filled silicone, which means five companies sit underneath a market of several hundred part numbers.

Distribution is not a footnote here. DigiKey and Mouser are where a design engineer finds and samples a material during the design phase, which makes them part of the design-in path rather than just a logistics channel. Ellsworth Adhesives and Krayden are the specialty side, carrying Henkel, Dow, 3M, Momentive, and Master Bond lines with dispensing equipment and application engineering, and they are how an EMS buys chemistry by the case with support attached. Arrow matters mostly because it is already on every approved vendor list for everything else.

The fourteen brand owners in this industry buy these materials in three recognisably different ways, and the difference is not about size. It is about how much manufacturing they still do and how tightly their industry regulates change.

The fabless brand owners are pure specification accounts. Apple, Dell, HP, and Cisco design in North America and build through EMS and ODM partners, mostly in Mexico and Asia. There is no plant to call on and no purchase order to win at the brand owner itself. What there is, is an engineering organisation that decides what goes on the drawing, and a materials or reliability group that runs the qualification. Getting on an approved materials list at one of these companies is worth years of volume at an EMS you have never met. Losing the engagement means finding out that the next platform specified somebody else and there is no recovery until the platform after that.

The industrial and instrumentation brand owners still build. Rockwell Automation assembles in Wisconsin and Ohio, Emerson in Minnesota and Colorado, Honeywell across Minnesota, Arizona, and Kansas, and Garmin is vertically integrated in Olathe. These are the most accessible accounts in the industry for a materials supplier, because the specifier and the buyer are in the same building and the volumes are large enough to matter without being large enough to attract every competitor. They also run hotter and dirtier environments than consumer electronics, which means more conformal coating, more potting, more silicone, and more thermal content per unit.

The regulated brand owners are slow and then permanent. Stryker on the medical side and L3Harris on the defense side both run change control that treats a material substitution as a design change. For medical devices under ISO 13485 and FDA design control, an approved adhesive is written into a device master record and changing it means revalidation. For defense work, ITAR registration, domestic sourcing, and qualified-parts documentation gate the conversation before the material is even discussed. Qualification into either takes a year or more. Once in, incumbency is close to absolute, which is exactly why the qualification effort is worth funding.

The controls-and-appliance case sits slightly apart. Whirlpool appears in this roster for control boards and user interface electronics, not for the appliance. That content is high-volume, cost-driven, coated for humidity, and bought either from an EMS or from a captive Mexican plant. It is the closest thing in this industry to an automotive purchasing conversation, and it should be approached that way.

Two practical notes on all fourteen. First, the specification conversation and the purchasing conversation happen in different states, sometimes different countries, and they are rarely introduced to each other by the customer. The supplier has to carry information in both directions itself. Second, a brand owner that outsources manufacturing has not outsourced the specification, and treating the EMS as the decision maker is the single most common way a materials supplier loses a design-in it had already won.