CAPEX optimization that respects polysilicon purity.
A four-week CAPEX review that exposes margin stacking, duplicated contingency, and overconservative sizing — without touching any decision that matters for purity, uptime, or maintainability.
Cutting capital cost without touching purity, reliability, or operability.
CAPEX optimization on polysilicon plants is one of the most misunderstood services in the industry. Done badly, it means deleting redundancy until the plant has none left, downgrading materials until the plant corrodes, or stripping instrumentation until the plant flies blind. Done properly, it means exposing where capital cost is being created by margin stacking, overconservative sizing, unnecessary rework provisions, and duplicated contingency — and removing those layers without touching any decision that affects purity, uptime, or maintainability.
NEXARSiL's CAPEX optimization work is driven by the same engineers who run our owner's engineering and plant design reviews. That's deliberate. A CAPEX number is only as good as the engineer signing off on the risks behind the number. We will not reduce a capital cost we wouldn't ourselves accept as owner's engineers.
Seven places we routinely find savings.
Margin stacking
Licensor margin on EPC margin on vendor margin. We split the estimate by layer and re-price each one.
Duplicated contingency
Contingency buried at project level, package level, and vendor level. We rationalize it to a single line.
Overconservative sizing
Equipment sized for a nameplate the plant will never run at. We recalculate against real production envelopes.
Oversized utilities
Power, cooling, nitrogen, and instrument air systems sized on worst-case stacked peaks. We run them through operating plant data.
Scope misallocation
Scope that belongs in an existing site utility showing up in the project. Often recoverable with a scope clarification.
Spare philosophy
Unnecessary 2x100% spares on equipment where 2x50% + strategic spares is safer and cheaper.
Weld & metallurgy
Weld class and material grade pushed above what polysilicon actually requires. Savings without purity impact.
Four weeks, one number, zero compromises on purity.
Baseline
Ingest the full CAPEX estimate line by line, with all supporting vendor quotes and assumptions.
Layer split
Separate vendor cost, EPC margin, licensor margin, and contingency. Benchmark each against real projects.
Engineering pass
Challenge sizing, spare philosophy, metallurgy, and scope allocation — with the owner's engineering team involved.
Decision memo
A clean CAPEX number, a list of reductions with specific justifications, and the risks the owner is taking for each.
What we review — and what you can do with the findings
When this service is needed
- EPC or licensor quotes appear high and lack line-item transparency
- Lender or finance committee is questioning cost assumptions
- Project is over budget before it starts
- Pre-FID estimate has not had independent review
- Detailed design is in progress and scope creep is occurring
What usually goes wrong in polysilicon CAPEX
- Duplicated contingency layers across licensor, EPC, and owner estimates
- Conservative equipment sizing beyond the actual design basis
- Margin stacking through multiple subcontract tiers
- Overspecified metallurgy in non-critical systems
- Undefined scope items carried as contingency instead of properly defined scope
What NEXARSiL reviews
- Equipment lists and take-off quantities
- Vendor quotes and procurement assumptions
- Contingency methodology at each tier
- Utility specifications and spare-parts philosophy
- Metallurgy and cleanliness specifications vs. purity requirements
What you receive — and what you can do with it
- Line-by-line CAPEX review identifying specific reducible items with justification
- Written findings suitable for negotiations with EPC or licensor
- A defensible floor estimate you can present to lenders or the board
- Clear flags where proposed savings would affect purity or reliability (owner's decision)
Questions we often get about this
How much CAPEX saving is typical?
It depends heavily on the project and the baseline estimate. On estimates that haven't had independent review, we routinely find mid-single-digit to low-double-digit percentage savings. On already-optimized estimates, the value shifts from savings to risk reduction and defensibility.
Is CAPEX optimization separable from owner's engineering?
Yes, but they pair well. A standalone CAPEX review is valuable before FID. Combined with full owner's engineering, CAPEX optimization continues through detailed design and procurement.
Will you cut corners on cleanliness or metallurgy to hit a number?
No. Every proposed saving is reviewed against purity and reliability impact. If a saving touches a decision that matters for purity, we flag it and leave the call to the owner.
Talk to an independent polysilicon technology partner.
Confidential conversation. No obligation. Real production experience from the first call.