- with readers working within the Technology industries
"The Federal Circuit did not mince words about Phillips's litigation strategy, describing it as 'sandbagging' Magēmā with a 'bait-and-switch.'"
In Magēmā Technology LLC v. Phillips 66, No. 24-1342 (Fed. Cir. Sept. 8, 2025) ("Magēmā, slip op."), the U.S. Court of Appeals for the Federal Circuit reversed a jury verdict finding no infringement of plaintiff Magēmā's patent by defendant Phillips. At first glance, the case appears to involve an ordinary patent dispute over methods for "hydroprocessing" marine fuel oil. But beneath the surface lies a compelling story about how Magēmā's patented invention sought to clean up the dirtiest fossil fuel on earth that drove the billion-dollar trans-ocean shipping industry for over a century, and how Phillips evaded Magēmā's proof of infringement through textbook discovery abuse and trial misconduct, requiring the Federal Circuit's intervention.
The legal drama centers on a classic litigation bait-and-switch strategy as dirty as the marine fuel at issue in the case. During discovery, Magēmā (pronounced "Maggie Mae," like the Beatles song) demanded certain refinery flashpoint data—the temperature at which fuel vapors can ignite—from Phillips's marine fuel production facility needed to prove infringement of the patent claims. Phillips, however, convinced the magistrate judge presiding over the dispute that collecting the requested flashpoint data from inside the refinery's active processing units would be "too dangerous" and assured the court that Magēmā could rely instead on flashpoint estimates prescribed by industry standards.
Then, at trial, Phillips reversed course, telling the jury that Magēmā's flashpoint estimates were inherently unreliable and worthless in assessing Magēmā's proof of patent infringement. Without actual refinery flashpoint data, Magēmā was set adrift with no meaningful way to chart a course to victory. When Magēmā sought a curative instruction to explain the evidentiary bind Phillips had created, the district court refused to throw a lifeline. And after trial, while acknowledging Phillips's obvious misconduct, the court ruled that any error arising from Phillips's tactics was harmless.
Just as Magēmā's patent sought to clean up the shipping industry's dirtiest fuel, the Federal Circuit found itself tasked with cleaning up Phillips's dirty trial tactics. The appellate court's analysis of general verdict forms and harmless error doctrine reveals how discovery gamesmanship can pollute an entire trial—and demonstrates the Federal Circuit's commitment to maintaining fairness in patent litigation. But first, one must understand the marine fuel at the core of this dispute and the regulatory earthquake that shook its century-old market, leading to Magēmā's invention.
The Sludge That Powered Globalization
We have all seen footage of quarter mile-long container ships sliding through international waters, their hulls stacked with thousands of steel containers bound for ports across the globe. Unseen is what sits deep in the vessels' engine rooms, a massive two-stroke engine the size of a four-story building churning steadily forward. These marine giants don't burn the clean diesel fuel familiar to motorists. Instead, they run on what looks and feels like liquid asphalt—a black, tarry residue left behind after refineries strip higher-grade fuels like diesel, gasoline, and jet fuel from crude oil.
This heavy marine fuel oil ("HMFO"), also known as bunker fuel, has been the fuel of maritime giants for more than a century. When crude oil enters a refinery, the high value products are drawn off through distillation, leaving behind a thick, viscous residue that refineries once considered waste. The shipping industry turned that waste into an economic advantage by designing enormous slow-speed two-stroke engines capable of burning the cheapest, and dirtiest, fuel on earth. See Magēmā, slip op. at 3.
The economics are compelling. For container ships hauling 20,000 boxes across oceans, where fuel represents 40-60% of operating costs, HMFO's price advantage is decisive. While distilled fuels like marine diesel oil ("MDO") and marine gasoil ("MGO") are cleaner and easier to handle, they cost significantly more. Although cheap, HMFO requires specialized handling—heating systems to keep it pumpable, purification equipment to remove water and contaminants, and colossal engines designed specifically for this low-quality fuel. The cost savings from HMFO justifies this complexity in vessels large enough to accommodate these cumbersome systems.
The dividing line has always been scale. Smaller vessels—ferries, offshore supply ships, naval craft—typically use MDO or MGO. But above roughly 20,000 to 30,000 deadweight tons, the economics shift decisively toward HMFO. And so HMFO became known as "the fuel of giants"—vessels so large that the cost and complexity of handling tar-like fuel is outweighed by its bargain-basement price. These economies of scale in maritime fuel selection have persisted across decades of technological improvements in ship design and propulsion.
The Regulatory Framework: ISO 8217 and MARPOL Annex VI
For decades, two international standards have defined HMFO's use, creating the invisible rulebook that governed maritime fuel worldwide.
ISO 8217:2017 sets the physical property requirements for marine fuels. Table 2 of this standard, which governs residual fuels like HMFO, specifies minimum and maximum limits for characteristics such as viscosity, density, stability, and crucially, flashpoint—the lowest temperature at which a fuel can vaporize to form an ignitable mixture. For safety reasons, ISO 8217 requires HMFO to have a flashpoint of at least 60°C (140°F). See Magēmā, slip op. at 3.
MARPOL Annex VI, incorporated into U.S. law through the Act to Prevent Pollution from Ships ("APPS") at 33 U.S.C. §§ 1905–1915, sets limits on sulfur content in marine fuels. The U.S. Environmental Protection Agency promulgated regulations implementing APPS, which incorporate by reference the full text of Annex VI. See 40 C.F.R. § 1043.100(b)(1). Before 2016, this standard allowed a generous 3.5% sulfur by weight. Fuels meeting both ISO 8217's physical properties and Annex VI's sulfur limits were considered compliant and could be sold globally as merchantable HMFO.
This regulatory framework tolerated remarkably dirty fuel. While ISO 8217 ensured minimum operational efficiency and safety standards, Annex VI's sulfur cap was so high that refineries could sell their dirtiest residue with minimal processing. The industry built its infrastructure accordingly with bunkering facilities, engine designs, and global supply chains all centered on burning this loosely regulated refinery sludge.
The 2016 Sulfur Reckoning
In October 2016, the International Maritime Organization ("IMO") announced revisions that sent shockwaves through this regulatory landscape. Beginning January 1, 2020, the sulfur cap in Annex VI was required to plummet from 3.5% to 0.5% by weight—a reduction of more than 85%. See Magēmā, slip op. at 3-4. Known as the "IMO Sulfur Cap," this single revision reshaped the entire economics of maritime fuel. And the 2016 announcement gave the industry less than four years to adapt. Shipowners were faced with two options.
The first option was switching to very low sulfur fuel oil ("VLSFO"), a higher-grade distilled fuel, to meet the new regulatory requirements. But this option comes at a high price. On average, VLSFO costs $150-200 per ton more than HMFO, which translates to millions of dollars annually for large vessels.
The other option involved a more direct solution—installing exhaust scrubbers to remove sulfur. These systems allow shipowners to continue using low-cost high-sulfur HMFO but are themselves expensive to install and operate. Installation costs alone range from $5-10 million per ship.
The industry's response was swift and dramatic. By early 2025, approximately 4,730 ships worldwide—representing about 24.5% of global fleet tonnage—opted to install scrubbers. See Ship & Bunker, "Quarter of Global Tonnage Now Using HSFO at Record Discount to VLSFO" (2024). Container ships led that adoption, with over 36% of global boxship capacity now scrubber-equipped. See Ship & Bunker, "Scrubber-Equipped Ships Now Take Up 36.4% of Global Boxship Capacity" (2023). The remainder of the fleet largely switched to low-sulfur blends, fundamentally reshaping global marine fuel markets as shipowners sought alternatives to traditional HMFO.
Magēmā's Innovation: Cleaning Sludge at the Source
Against this backdrop of regulatory disruption and industry scrambling, Magēmā developed what appeared to be an elegant third option. Rather than installing expensive scrubbers on individual ships or switching to costly low-sulfur distilled fuels, Magēmā's approach targeted the problem at its source: the refinery.
U.S. Patent No. 10,308,884 ("the '884 patent"), issued to Magēmā in 2019, teaches a method of producing low-sulfur HMFO through hydroprocessing. The process begins with a high-sulfur HMFO that meets ISO 8217's Table 2 physical property requirements but contains more than 0.5% sulfur by weight. This fuel is then exposed to hydrogen at elevated temperatures and pressures in the presence of a catalyst, stripping away sulfur atoms and producing a low-sulfur HMFO that complies with both ISO 8217 and the new Annex VI limits. See the '884 patent, col. 25, ll. 27-36.
The inventive insight was not hydroprocessing itself, which refineries had long used to clean lighter petroleum fractions. Rather, Magēmā's innovation lay in applying this established technology to residual HMFO—a notoriously difficult feedstock due to its viscosity, impurities, and relatively low value—and specifying how to produce a compliant, merchantable fuel that preserved the economic advantages of sludge while satisfying the new environmental requirements.
Claims 1 and 5 of the '884 patent specifically require that "prior to hydroprocessing," the high-sulfur HMFO must be "compliant with ISO 8217:2017" and, in the case of Claim 1, "of merchantable quality as a residual marine fuel oil." This meant the feedstock had to meet all of ISO 8217's physical property requirements, including the 60°C flashpoint minimum, before undergoing desulfurization. Id.
Magēmā pitched its technology to Phillips 66 in 2017 and 2018. While the parties met to discuss potential licensing arrangements, no agreement was reached. Unbeknownst to Magēmā, Phillips was quietly modifying its hydrotreater units at the Bayway Refinery in New Jersey and Wood River Refinery in Illinois to produce low-sulfur HMFO. When Phillips later highlighted these upgrades in investor materials, touting its ability to meet the IMO Sulfur Cap, Magēmā filed suit for patent infringement. Magēmā, slip op. at 5-6.
The Discovery Battle: Safety Concerns or Strategic Posturing?
The litigation quickly focused on a crucial technical dispute: where exactly Phillips's feedstock should be tested to prove it met ISO 8217's flashpoint requirement. In refinery terminology, "battery limits" define the boundaries of a processing unit—imagine a fence around the core equipment where the actual chemical reactions occur. Phillips contended that compliance should be determined "outside the battery limits" (OSBL) of its DSU-1 hydrotreater reactor at Bayway, essentially at the perimeter before materials enter the processing area, and it provided test samples from that location. Magēmā argued for testing "inside the battery limits" (ISBL) but before the feed entered the actual hydrotreater reactor—a location much closer to where the chemical processing begins. Magēmā, slip op. at 7.
This distinction was important because feedstock properties can change as materials move through a refinery's complex piping and heating systems. If the fuel met ISO 8217's requirements at the OSBL location but degraded by the time it reached the ISBL location, Magēmā might not be able to prove that Phillips's actual feedstock complied with the patent's requirements. Phillips's safety argument also made more sense in this context because sampling from inside the battery limits means collecting material from active processing areas where high temperatures, pressures, and flammable vapors create genuine hazards.
Thus, when Magēmā moved to compel production of flashpoint data from the ISBL location, Phillips mounted a defense based on safety concerns. At the hearing before the magistrate judge, Phillips argued that collecting samples from inside the battery limits would be "too dangerous." More importantly for the subsequent jury trial, Phillips assured the court that such samples were unnecessary because Magēmā could use the established "Riazi Formula"—a generally accepted industry method—to estimate flashpoint from otherwise available data. Id. at 7-8.
The magistrate judge, while acknowledging that he was "not keen on ordering sampling" that posed safety risks, recognized that Magēmā needed some method of proving its case. Relying on Phillips's representation that Riazi Formula estimates would suffice, the court denied Magēmā's motion to compel actual test samples. Id. at 8.
This discovery ruling set the stage for the dramatic reversal that would follow at trial.
Phillips's Bait-and-Switch: From Estimates to Actual Testing
The landscape shifted dramatically after the close of fact discovery, and the parties each moved for summary judgment. Despite Phillips's earlier representations about safety concerns and the adequacy of flashpoint estimates, it became clear that Phillips actually possessed the capability to test flashpoint inside the battery limits. Indeed, Phillips moved to supplement the summary judgment record with evidence from a new ISBL sample station it had constructed. Magēmā, slip op. at 8.
The district court denied Phillips's motion to supplement, finding that Phillips had "failed to provide a reasonable explanation for [its] failure to sample" the ISBL feed before the discovery cutoff. J.A. 25 (quoted in Magēmā, slip op. at 8). The court noted that Phillips was on notice of Magēmā's desire for testing samples from that location after Magēmā filed its motion to compel, and that allowing supplementation would be "unduly prejudicial" given Phillips's earlier representations that actual testing samples were unnecessary and that Magēmā could rely on Riazi Formula estimates. Id.; J.A. 25.
These discovery and summary judgment rulings should have resolved the issue. Phillips had successfully argued that estimates were sufficient, and actual testing was both dangerous and unnecessary. The case could proceed to trial with Magēmā relying on its expert's Riazi Formula calculations to prove that the Bayway feedstock met ISO 8217's flashpoint requirement.
But on the eve of trial, Phillips revealed an entirely new strategy. Right before jury selection, Magēmā learned that Phillips planned to use a demonstrative slide during opening statements to show the jury that—despite Phillips's pretrial representations to the contrary—Riazi Formula estimates of flashpoint were insufficient to prove infringement. Instead, Phillips planned to argue that ISO 8217 requires actual testing data. Magēmā, slip op. at 9.
Magēmā objected on the record, recounting the discovery proceedings before the magistrate judge and the district court's denial of Phillips's motion to supplement the summary judgment record with the late-produced ISBL testing data. Even as the district court reviewed its summary judgment opinion denying Phillips's motion to supplement, it overruled the objection, stating that it did not "understand [Magēmā's] argument" and that it was "not going to exclude" Phillips from presenting its actual-testing theory to the jury during opening statements. Id. at 9-10; J.A. 846.
Trial: Prejudicial Arguments and General Verdict Problems
Phillips wasted no time implementing its new strategy. During opening statements, counsel told the jury that ISO 8217 "requires the use of actual specified ISO test procedures to prove compliance with the ISO 8217 specifications" and that jurors would "not find anything in ISO 8217 stating, suggesting, or implying that you can use an estimate to show compliance." The message was clear: "If there's no compliance, there's no infringement." Id. at 10; J.A. 855.
Phillips reinforced this theme through cross-examination of three Magēmā witnesses, repeatedly emphasizing that "there is nothing in ISO 8217 that says it's okay to use an estimate as a substitute for [actual testing data]." When one witness attempted to explain that "Phillips told us it was okay to use" estimates, Phillips's counsel cut him off, demanding a simple acknowledgment that estimates were not mentioned in the ISO standard. Id. at 10; J.A. 894.
The prejudicial nature of this strategy became even more apparent when Magēmā sought a curative instruction. The district court initially seemed not to grasp the problem, but after Magēmā briefed its motion, the court began to understand the bind that Phillips had created. As the court observed, "how in the world" could Magēmā satisfy the burden of providing actual test data when it was Phillips who "didn't produce tests or provide a test method that could be used by Magēmā before the discovery cutoff." Id. at 11-12; J.A. 1065.
Nevertheless, the district court denied the motion for a curative instruction, suggesting that the "problem" could be resolved by allowing Magēmā to argue that no actual testing was done at the ISBL location while Phillips could argue that the Riazi Formula estimates were unreliable. The court also determined that a curative instruction was unnecessary because the jury was unlikely to remember "a three-minute discussion" about ISO 8217 requiring actual testing. Id. at 12.
The court did warn Phillips against further arguments that ISO 8217 required actual testing, stating it would be "attuned [to any] attempt from any creative lawyer to get around" the court's instructions not to argue to the jury that the ISO 8217 standard requires actual testing. Id.
Phillips ignored this warning. In closing arguments, counsel told the jury it should find no infringement because there was "no actual test data that shows compliance" with ISO 8217: "And you answer 'no' because that's consistent with there's no actual test data that shows compliance, none." Id.; J.A. 1086.
The jury returned a general verdict of noninfringement on all claims, but the verdict form provided no insight into the basis for this finding. The form simply asked whether "Magēmā prove[d] by a preponderance of the evidence that the Bayway DSU-1 hydrotreater process or the DSU-1 product infringed" the specified claims, with spaces for "Yes" or "No" answers. The jury wrote "no" in both spaces. Magēmā, slip op. at 13; J.A. 80.
District Court's Post-Trial Analysis: Improper but Harmless?
In post-trial proceedings, the district court made several significant findings. First, it explicitly acknowledged that Phillips's actual-testing arguments were both "improper and prejudicial." The court recognized that Phillips had argued it was too dangerous to gather actual testing data and that Magēmā could meet its infringement burden using Riazi Formula estimates calculated from otherwise available data. Magēmā, slip op. at 13; J.A. 111.
Despite this finding, the district court denied Magēmā's motion for a new trial, concluding that Phillips's improper arguments constituted harmless error. The court reasoned that the jury could have based its verdict on other permissible grounds related to two different claim limitations: whether the high-sulfur fuel was "compliant with ISO 8217" at the correct location ("prior to hydroprocessing") or whether it was "of merchantable quality." Id; J.A. 112-13.
Regarding the "prior to hydroprocessing" limitation, the district court found that Phillips had successfully impeached Magēmā's expert on the proper location for testing flashpoint compliance. As for merchantability, the court concluded that Phillips had presented "undisputed evidence that the Bayway feed was not merchantable because its viscosity was too low." Id. However, only one of the two asserted patent claims included the "merchantability" limitation, and a lack of proof for this claim element could not fully explain the verdict in favor of Phillips.
This harmless error analysis would prove to be the district court's critical mistake.
Federal Circuit's Reversal: The Perils of General Verdicts and Discovery Gamesmanship
The Federal Circuit's analysis began with established principles of harmless error review under Fifth Circuit law. A district court abuses its discretion in denying a motion for a new trial if it is "reasonably clear that prejudicial error has crept into the record or that substantial justice has not been done." Streber v. Hunter, 221 F.3d 701, 736 (5th Cir. 2000) (quoted in Magēmā, slip op. at 15). An error is harmless only if the court is "sure, after reviewing the entire record, that the error did not influence the jury or had but a very slight effect" on the verdict. Harris v. FedEx Corp. Servs., Inc., 92 F.4th 286, 303 (5th Cir. 2024) (quoted in Magēmā, slip op. at 15-16).
The court emphasized a crucial principle that district courts sometimes overlook: when a case is submitted on a general verdict form, prejudicial error under one theory generally requires reversal because "the reviewing court cannot determine whether the jury based its verdict on a sound or unsound theory." Muth v. Ford Motor Co., 461 F.3d 557, 564 (5th Cir. 2006) (quoted in Magēmā, slip op. at 16). This rule reflects the practical reality that appellate courts cannot read jurors' minds or reconstruct their deliberations. See also Jamison Co., Inc. v. Westvaco Corp., 530 F.2d 34, 37 (5th Cir. 1976) (cited in Magēmā, slip op. at 16).
The General Verdict Problem
The Federal Circuit identified the core problem: "We cannot tell why the jury answered 'no.'" The verdict form provided no basis for determining whether the jury relied on permissible theories (such as lack of merchantability or incorrect testing location) or Phillips's improper actual-testing theory. Magēmā, slip op. at 17.
This uncertainty was not merely academic. The jury could have permissibly found that Magēmā failed to prove infringement because: (1) the feedstock was not merchantable prior to hydroprocessing, or (2) compliance with ISO 8217 should be measured outside the hydrotreater battery limits where Phillips had provided actual test data. But the jury could also have impermissibly relied on Phillips's repeated assertions that ISO 8217 required actual testing data, making Magēmā's estimates worthless. Id. at 17-18.
The court noted that Phillips had "repeatedly and continuously urged its improper and prejudicial actual-testing theory throughout trial"—not just in opening statements, but through cross-examination of three witnesses and, most problematically, in closing arguments after the district court had specifically warned against such tactics. Id. at 18.
Rejecting the District Court's Harmless Error Analysis
The Federal Circuit systematically dismantled the district court's reasoning that alternative theories could explain the verdict. Regarding merchantability, the appellate court noted there was actually disputed evidence on this issue, with Phillips's own employee admitting on cross-examination that "for the majority of time, [the viscosity on the feed and product] meet [IMO] spec." Moreover, only Claim 1 included a merchantability limitation, so some other theory would be needed to explain the jury's rejection of Claim 5. Id. at 19-20; J.A. 1080.
As for the "prior to hydroprocessing" limitation, the Federal Circuit acknowledged that Phillips may have successfully impeached Magēmā's expert on the proper testing location, but emphasized this was a credibility determination that the court could not assume the jury made. The district court had itself noted that the parties "vigorously disputed" the testing location and presented "conflicting evidence on [the] issue"—hardly the stuff of certain jury findings. Magēmā, slip op. at 20; J.A. 103-04.
The Bait-and-Switch Analysis
The Federal Circuit did not mince words about Phillips's litigation strategy, describing it as "sandbagging" Magēmā with a "bait-and-switch." Magēmā, slip op. at 16. The court traced the timeline: Phillips first argued that collecting samples would be too dangerous and that estimates would suffice, successfully defeating Magēmā's motion to compel. Phillips then attempted to supplement the record with newly obtained test data, which the district court properly rejected as prejudicial given Phillips's earlier representations. Finally, Phillips announced on the eve of trial that it would argue ISO 8217 required actual testing—the very data Phillips had prevented Magēmā from obtaining.
This conduct went beyond ordinary adversarial tactics. Phillips had actively created the evidentiary gap it would later exploit, preventing Magēmā from obtaining evidence during discovery and then arguing that the absence of this same evidence defeated infringement. The Federal Circuit recognized that allowing such manipulation would undermine the integrity of the discovery process and the fairness of trials.
The Manifest Injustice Standard
Applying Fifth Circuit precedent, the Federal Circuit concluded that "manifest injustice [would] result from letting the verdict stand." Learmonth v. Sears, Roebuck & Co., 631 F.3d 724, 731 (5th Cir. 2011) (quoted in Magēmā, slip op. at 18-19). The court could not be "reasonably certain" that the jury's verdict was free from the taint of Phillips's improper theory, particularly given how "repeatedly and continuously" Phillips had urged this theme despite explicit judicial warnings. Id.
Claim Construction: Lexicography in Technical Patents
Beyond the discovery and trial procedure issues, the Federal Circuit's claim construction analysis provides guidance for patent prosecution and litigation in highly technical fields. Phillips had argued that "HMFO" should be limited to fuels containing "process residues"—specifically, "fractions that don't boil or vaporize even under vacuum conditions." Magēmā, slip op. at 22; '884 Patent, col. 1, ll. 28-30.
The district court rejected this narrow construction, and the Federal Circuit affirmed, emphasizing Magēmā's role as its own lexicographer. The patent specification explicitly stated that certain terms, including HMFO, were "having a specific intended meaning" and provided an express definition: "a petroleum product fuel compliant with the ISO 8217:2017 standards for the bulk properties of residual marine fuels except for the concentration levels of the Environmental Contaminates." Magēmā, slip op. at 22; '884 Patent, col. 7, ll. 50-54.
Phillips pointed to language in Column 1 describing HMFO as comprising "process residues," but the Federal Circuit held that this background description of the prior art could not override the express definition in the specification. The court's analysis emphasized the principle from Martek Biosciences Corp. v. Nutrinova, Inc., 579 F.3d 1363, 1380 (Fed. Cir. 2009), that when patentees act as their own lexicographers, clear and explicit definitions control over general descriptive language elsewhere in the patent.
The Remand: A Fresh Start Without Gamesmanship
The Federal Circuit's remand order includes a specific prohibition: "At that re-trial, Phillips is not to argue that the ISO 8217 standard requires actual testing data." Magēmā, slip op. at 20-21. This directive ensures that the new trial will proceed based on the discovery record as it actually developed, rather than the distorted version Phillips attempted to create.
The remand also preserves the district court's claim construction, meaning the parties can focus on the merits of infringement without relitigating the meaning of key terms. Magēmā will have the opportunity to prove its case using the Riazi Formula estimates that Phillips originally agreed were sufficient, while Phillips can challenge the reliability of those estimates and argue other non-infringement theories.
Conclusion: Fairness in High-Stakes Patent Litigation
Magēmā Technology LLC v. Phillips 66 began with a story about Magēmā's efforts to innovatively clean up the world's dirtiest fuel in response to the IMO's 2016 sulfur cap. Thematically, the outcome of its appeal turned on similar aims, cleaning up a dirty case by ensuring the integrity of the discovery process and the necessity of fairness at trial.
Phillips's attempts to have it both ways—blocking Magēmā's access to evidence during discovery while exploiting that same evidentiary gap at trial—crossed the line from zealous advocacy into prejudicial misconduct. In a case about cleaning up the sludge that powered globalization for over a century, the Federal Circuit ultimately cleansed the procedural missteps that tainted the district court proceedings. In remanding for a new trial, the court ensured that fundamental principles of litigation fairness will be properly served.
The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.
