The Silicone Sourcing Checklist I Run Before Any Order Above $5,000
-
Step 1: Write your own spec before a rep writes one for you
-
Step 2: Split commodity SKUs from specialty SKUs
-
Step 3: Build a TCO sheet, not a unit-price sheet
-
Step 4: Verify grade claims at the source, not on the label
-
Step 5: Audit the small parts — that's where certainty lives
-
Step 6: Re-check your "silicone is safer than plastic" assumption
-
Step 7: Set a re-verification date before you close the PO
-
Common mistakes I still see
This checklist is for B2B buyers who are about to place a silicone order — compounds, sealants, dispersions, O-rings, septa — and who want to avoid the "cheap quote, expensive year" outcome. It assumes engineering or QA has already given you the technical requirements. If they haven't, stop here. A procurement team negotiating without a spec is just guessing with extra steps.
I've run this sequence on roughly 240 purchase orders across six years, at two mid-size manufacturers. Where my sample is thin, I'll say so. Seven steps, in the order I actually use them.
Step 1: Write your own spec before a rep writes one for you
If you send a supplier an email saying "we need silicone rubber, similar to what we bought last time," you have just outsourced your specification to the sales department. Whatever they quote becomes your spec by default.
Before you request quotes, put together a one-page requirement sheet. At minimum: hardness (Shore A), tensile and tear values referenced to a test method (ASTM D412, ASTM D624), compression set (ASTM D395), temperature range, media contact list, and the regulatory pathway you actually need — FDA 21 CFR 177.2600 for repeated-use food contact, USP Class VI or ISO 10993 for medical devices, or none of the above.
The step most buyers skip: they treat the supplier's data sheet as their own spec. It isn't. A data sheet describes what the material can do in a test lab. Your spec describes what your application requires. Those are different documents, and only one of them protects you when something fails.
One practical note on naming. If you're still searching for "dow-corning" part numbers, you're not alone — the Dow Corning name was retired after Dow took full ownership in 2016, and most products now carry the Dow brand. The part numbers and spec sheets are still circulating in project documents, tender packages, and archived drawings. That's fine for continuity, but confirm with your distributor that the current product name, current CoA format, and current revision of the data sheet match what's in your files. I've seen a tender package reference a 2014-era document that no longer reflected the shipped material. Nobody caught it until the incoming inspection.
Checkpoint: could two different distributors quote off your spec sheet and produce comparable numbers? If not, the spec is still too vague.
Step 2: Split commodity SKUs from specialty SKUs
Sort your list into two buckets before you talk to anyone. Bucket A is standard parts you can specify by a published standard — O-rings to ISO 3601, general-purpose rubber to ASTM D2000 line callouts. Bucket B is anything where geometry, compound, or application context matters more than the catalog number.
The reason matters for pricing strategy. Bucket A responds to competitive bidding. Bucket B does not — three quotes on a specialty part usually just gives you three different interpretations of your requirement.
This is also where most cross-material substitution errors happen. A polyurethane o-ring is not a drop-in for a silicone o-ring, and the reverse is also true. Polyurethane wins on abrasion resistance and oil resistance. Silicone wins on temperature range and compression set at temperature extremes. If a buyer forces a polyurethane o-ring into a high-temperature application to chase a lower unit price, the failure won't show up on the invoice — it'll show up as a leak, a complaint, or a line stoppage. (In other words: same shape, different job.)
Step 3: Build a TCO sheet, not a unit-price sheet
Unit price is the least useful number in silicone procurement. What you want is a total cost of ownership column set. Mine has seven columns: material cost, freight, minimum order quantity effect, tooling or setup, incoming inspection labor, expected scrap or rework, and a risk premium line for early failure.
Take a building sealant decision. If you're specifying a dow corning 790 silicone building sealant for a curtain wall joint, you'll get competing quotes from products described as "equivalent." On paper, one might come in 15–20% lower on material. Now add the columns. Does the lower-cost product carry the same movement capability at the specified joint width? Does it come with the same adhesion test data on your actual substrate? What happens to the warranty position if you switch?
The upside of the cheaper sealant is real money, usually in the low thousands on a mid-size project. The risk is a bond failure at the 18-month mark, and the rework cost on a curtain wall is not measured in material percentages — it's measured in scaffolding, labor, and schedule. I've been on the wrong side of that math once. Once was enough.
The question isn't "which sealant is cheaper." It's "which sealant is cheaper after I price the probability of doing this twice."
Step 4: Verify grade claims at the source, not on the label
"Medical grade" is not a protected term. Neither is "pharmaceutical grade." Anyone can print either on a box. This matters most when you're buying a dow corning medical grade dispersion — the current Dow dispersion line carried over from the old Dow Corning portfolio — because the price premium is substantial and the failure mode is regulatory, not mechanical.
For any medical or regulated-use dispersion, ask for four things and check them against each other:
- Batch-specific certificate of analysis (CoA), with the lot number matching the container label.
- Current quality system certification — ISO 13485 for medical device materials, not just ISO 9001.
- Biocompatibility test reports referenced to USP <87>/<88> or ISO 10993, with the actual report number, not a marketing claim.
- A formulation lock statement — how the supplier notifies you of changes to the compound.
I went back and forth for about three weeks between buying direct from the manufacturer's authorized channel and buying through a regional distributor at roughly 12% less. The authorized channel offered full traceability and a clean CoA chain. The distributor offered savings and faster lead times. I ultimately chose the authorized channel, because the project was for a device submission and I couldn't afford an audit question I couldn't answer. That was a judgment call, not a universal rule — for non-regulated industrial use, the distributor math might look different.
Step 5: Audit the small parts — that's where certainty lives
Small components are where procurement attention goes to die. A silicone/PTFE bonded septa for a chromatography or analytical instrument line might be 2% of your annual spend. It might also be the part that stops a production line for half a day when the bond delaminates or the silicone layer thickness drifts out of tolerance.
For bonded septa and similar small assemblies, three checks matter more than price:
- Silicone layer thickness tolerance — the datasheet nominal value is not the controlled parameter. The tolerance is.
- Bond integrity data — ask what test the supplier runs to confirm the silicone-to-PTFE bond, and how often.
- Extractables and sterilization compatibility — if the septa go into a sterile or analytical environment, the extractables profile and the sterilization method need to be specified, not assumed.
Batch-to-batch consistency beats unit price here by a wide margin. A supplier quoting 8% less but shipping with wider lot variation is not cheaper. It's a future investigation report.
Step 6: Re-check your "silicone is safer than plastic" assumption
This one is aimed at buyers whose material selection logic was set five or six years ago and hasn't been revisited. The question "is silicone safer than plastic" gets asked a lot, and the honest answer is that it's the wrong question in that form.
What was best practice in 2020 doesn't automatically apply in 2025. Material grades have changed, regulatory pathways have clarified, and the testing data available from suppliers is better than it was. But the fundamentals haven't changed: the answer depends on temperature, contact media, contact duration, sterilization method, and what regulatory pathway the end product sits in. Silicone is generally well-suited to high-temperature and bio-contact applications. Plastics like polypropylene or PEEK win in other contexts — chemical resistance, structural rigidity, cost. None of these materials is universally "safer."
So the checklist item is this: instead of asking which material class is safer, ask which material has valid, current, application-specific evidence for your exact conditions. If a supplier can't produce that evidence for your scenario, the safety claim is marketing, not engineering.
Honestly, I'm not sure how the newer thermoplastic elastomer grades will hold up in long-duration implant-adjacent applications — the published data sets are still short. If someone has longitudinal data on that, I'd like to see it. My procurement decisions in that space would change if the evidence did.
Step 7: Set a re-verification date before you close the PO
The last step is the one that keeps all the other steps from expiring silently. When you approve a silicone material, record four dates: the CoA date, the last data sheet revision date, the next scheduled supplier audit, and a hard re-verification date — typically 12 months out, or immediately on any material, process, or regulatory change.
I also recommend dual-sourcing only what you can actually qualify. Two approved suppliers with verified documentation is worth more than four quotes from suppliers you've never audited. But don't dual-source to chase price — dual-source to protect supply, and let price follow qualification.
Common mistakes I still see
- Anchoring to a promotional price. The first quote isn't the price. The annual price adjustment clause in the contract is the price. Read it before you compare anything.
- Assuming tooling costs transfer. If you switch suppliers, molds and fixtures usually don't come with you — or they do, and they need requalification. Price that in.
- Treating "FDA compliant" as "FDA approved." The FDA does not approve materials. It lists them under regulations like 21 CFR 177.2600. Anyone who tells you their silicone is "FDA approved" is telling you they don't know the difference, or they're hoping you don't.
- Sharing one supplier's quote with another. It feels like leverage. It usually damages the relationship and the information flow you actually need.
My experience here is built on mid-volume industrial and medical device purchasing — a few hundred SKUs, mostly North American and European supply chains. If you're buying at automotive volumes or sourcing from a different region, some of the leverage dynamics will differ. The seven steps should still hold. The politics around them won't.
Leave a Reply