Chemical manufacturers can claim the federal research credit for formulation work and for process chemistry when each meets the four requirements of section 41(d)(1), and the two are tested as separate business components. A new or improved formula is one business component; the plant process used to make it is another under section 41(d)(2)(C). For R&D tax credits in the chemical industry, the close calls usually involve where scale-up becomes commercial production, customer-specific formulations, routine batch testing, and which raw material costs count.
Formulation and Process Chemistry as Separate Business Components
Section 41(d)(2)(B) defines a business component as any product, process, computer software, technique, formula or invention held for sale, lease or license, or used in the taxpayer’s trade or business. Section 41(d)(2)(C) treats any plant process, machinery or technique for commercial production as a separate business component, not part of the product being produced. Our overview of R&D tax credit services for chemical manufacturing lists where qualifying work is commonly found; the rules below decide which of it counts.
Formulation. Developing a coating, adhesive, resin, additive package or specialty blend with a new or improved function, performance, reliability or quality may qualify where the composition needed to reach the target is uncertain at the start. Research related to style, taste, cosmetic or seasonal design factors does not meet the qualified purpose test (section 41(d)(3)(B)).
Process chemistry. Developing a new route, catalyst system, set of reaction conditions, separation step or purification method may qualify as process research, even for a product that is already sold.
Experimentation, not repetition. Under Treas. Reg. 1.41-4(a)(5), a process of experimentation evaluates one or more alternatives to achieve a result where the capability or method of achieving it, or its appropriate design, is uncertain at the start. Designed experiments, screening studies and iterative reformulation can meet that standard when they record the alternatives tested and the results. Under Treas. Reg. 1.41-4(a)(6), at least 80 percent of the research activities have to be elements of a process of experimentation. Our guide to the four-part test covers each requirement.
Scale-Up From Bench to Pilot Plant to Production
Section 41(d)(4)(A) excludes research conducted after the beginning of commercial production of the business component. Under Treas. Reg. 1.41-4(c)(2)(i), that point arrives when the component is ready for commercial sale or use, or meets the taxpayer’s basic functional and economic requirements. The regulation deems preproduction planning for a finished component, tooling-up, trial production runs, troubleshooting faults in production equipment or processes, accumulating data on production processes, and debugging to occur after commercial production begins (Treas. Reg. 1.41-4(c)(2)(ii)).
The process has its own timeline. Under Treas. Reg. 1.41-4(c)(2)(iii), the exclusion applies separately to the product and to the process for making it. A formula settled at the bench can be a finished product design while the process to make it at scale is still uncertain: heat removal, mixing, addition rates, impurity profile or yield at larger volumes. Process work may still qualify until the process itself meets the company’s basic functional and economic requirements or is ready for commercial use.
| Stage | May qualify where | Likely excluded where |
| Bench or lab | Composition or route is uncertain and alternatives are tested | A known recipe is mixed to order |
| Pilot plant | Runs test whether a process design works at larger scale | Runs make material from a settled process |
| First plant campaigns | The process design is still being tested against targets | Runs are trial production of a settled process |
| Ongoing production | A separate process change raises new uncertainty | Troubleshooting, data collection, routine adjustment |
Where each stage ends turns on the facts, and the line between process experimentation and trial production runs is drawn differently from case to case.
Customer Specifications, Toll Manufacturing and Funded Research
Adaptation versus development. Section 41(d)(4)(B) excludes research related to adapting an existing business component to a particular customer’s requirement or need. The exclusion does not apply merely because a component is intended for a specific customer (Treas. Reg. 1.41-4(c)(3)). Adjusting a standard grade within known ranges to meet a customer’s viscosity or color specification is likely adaptation; developing a formulation to meet a performance target the company does not know how to reach may be qualified research. Reproducing an existing product from analysis of the product itself, or from its specifications, is excluded as duplication (section 41(d)(4)(C)).
Toll and contract manufacturing. When a customer pays for development work, the funded research rules in section 41(d)(4)(H) apply. Under Treas. Reg. 1.41-4A(d), payments that are not contingent on the success of the research count as funding, and a manufacturer that retains no substantial rights in the results is treated as fully funded. Agreements that bill development at cost, or assign formulas and process know-how to the customer, need review project by project; our article on funded research and the R&D credit covers the tests.
Regulatory-driven reformulation. Reformulating to remove a restricted substance or meet an emissions or exposure limit is neither qualified nor excluded because of the regulatory reason itself. The four requirements still apply: the work may qualify where it is uncertain how to keep performance with the new chemistry and alternatives are evaluated, and it is likely excluded where a known substitute drops in.
Batch Release Testing Versus Analytical Method Development
Section 41(d)(4)(D)(v) excludes “routine or ordinary testing or inspection for quality control.” The section 174 regulations describe quality control testing as testing or inspection to determine whether particular units of materials or products conform to specified parameters, not whether the design of the product is appropriate (Treas. Reg. 1.174-2(a)(7)). Batch release testing, certificate of analysis testing and inspection of incoming raw materials against specifications generally fall on the excluded side.
Testing inside a process of experimentation is treated differently. Performance and stability testing that evaluates candidate formulations, or that tests whether a new process makes material within targets, can be part of the experimentation. Developing a new analytical method may be research where it is uncertain how to measure a property or impurity and alternative methods are evaluated; whether the method is its own business component depends on the facts. Running an established method on routine samples is not research.
Costs: Experimental Batches, Pilot Equipment and Energy
Section 41(b)(2)(C) counts supplies used in the conduct of qualified research and excludes land, improvements to land and depreciable property from supplies.
Raw materials and catalysts. Raw materials, reagents, solvents, catalysts and substrates consumed in experimental batches can be qualified research expenses (QREs) as supplies. In a CSSI Services chemical manufacturing case study on waterborne resin and additive formulations, the page reports QREs of $4,450,250: $3,460,000 in wages, $740,000 in supply costs and $250,250 in contractor expenses.
Experiments on production equipment. Union Carbide Corp. v. Commissioner, 697 F.3d 104 (2d Cir. 2012), was a chemical manufacturer’s case. One of its projects tested a substitute catalyst in the normal production process for polyethylene. The court held that supplies that would have been used in the manufacturing process regardless of any research were not creditable. Experimental runs on a production line need to identify what the experiment itself consumed.
Pilot plants and lab equipment. Reactors, columns, analyzers and other pilot plant or lab equipment are generally depreciable property, so they are not supplies, and depreciation is not an in-house research expense under section 41(b)(2). Wages to design and operate them in qualified research may count.
Energy-intensive research. Utilities used in the building where research is performed are generally treated as general and administrative expenses. Under Treas. Reg. 1.41-2(b)(2)(ii), if the taxpayer can establish that the special character of the research required additional extraordinary utility expenditures, the additional amounts are treated as supplies; the regulation’s example is electricity for high-energy equipment such as laser or nuclear research. Whether a pilot reactor’s heating or cooling load meets that standard depends on the facts and the metering records.
Records That Support a Chemical Manufacturer’s Claim
Claims in the chemical industry usually rest on records kept as the work happens:
- lab notebooks, paper or electronic, recording the formulation or conditions tried and the results;
- designed experiment plans and analyses showing the alternatives evaluated;
- batch records for experimental and pilot batches, with material issues kept apart from production;
- stage-gate documents showing what was uncertain at each gate, and when the formula and the process were each released;
- customer agreements and statements of work for any funded or toll work;
- time records for the chemists, engineers and technicians involved.
Section G of Form 6765 asks for information by business component, including its type; the instructions list formula among the examples under the “All Others” type. Section G is optional for tax years beginning before 2026 and required for tax years beginning after 2025 unless an exception applies (Instructions for Form 6765, Rev. December 2025). Our Form 6765 guide and our page on documentation for R&D tax credits cover the reporting and records in more detail.
To see how your lab, pilot plant and process work measures up under these rules, you can request a free analysis from CSSI.
Frequently Asked Questions
Are formulation projects and process improvements separate claims?
They are separate business components. A formula is tested on its own, and the plant process for making it is tested separately under section 41(d)(2)(C). Each has to meet the four-part test, and the commercial production exclusion applies to each separately.
Does pilot plant work qualify for the R&D credit?
It may, where pilot runs test whether a process design can meet its targets at larger scale. Runs that make material from a settled process, and trial production runs, are generally treated as occurring after commercial production begins.
Are customer-specific formulations excluded as adaptation?
Not merely because they are made for one customer. Adjusting an existing product within known ranges to meet a customer’s specification is likely adaptation, while developing a formulation to reach a performance target the company does not know how to reach may qualify. If the customer pays for the work, the funded research rules also apply.
Do raw materials used in trial batches count as supplies?
They can, if used in the conduct of qualified research. In Union Carbide, supplies that would have been used in production regardless of the research were held not creditable, so trials on production equipment need to separate what the experiment consumed.
Is quality control testing ever part of the credit?
Routine or ordinary testing for quality control, such as batch release testing, is excluded. Testing that evaluates whether a new formulation or process design works can be part of a process of experimentation, and developing a new analytical method may be research where the method itself is uncertain.
This article provides general information about federal tax rules and is not tax advice. Speak with your tax adviser about your specific circumstances.
