dissenting.
The majority concludes that the district court erred in its construction of the term “substantially pure” in claim 1 of U.S. Patent No. 5,750,703 (“the '703 patent”) and therefore reverses the district court’s judgment. I would uphold the district court’s construction of that term, and I therefore respectfully dissent. -
I
The district court construed the term “substantially pure” to mean “at least 98% *1378purity with respect to all impurities.” In arriving at that construction, the court first determined that “substantially pure” has the same meaning whether it refers to the piperidine derivative end product or the para-CPK intermediate. The court noted that the “inventor uses the phrases ‘substantially pure’ and ‘substantially pure regioisomers’ indiscriminately to refer to both final products and intermediates. There is no evidence that the inventor intended the term to mean different things.” Based on the prosecution histories of the '703 patent and related U.S. Patent No. 5,578,610 (“the '610 patent”),1 the court then held that the term “substantially pure” refers to pharmaceutical grade purity, i.e., 98% pure. Finally, the court determined that the required purity level was to be'measured with respect to all impurities, not just the unwanted meta-CPK, except that the 98% purity level did not include “intended elements of solutions such as solvents, catalysts, or other compounds that are not considered impurities.”
Averitis’s primary argument on appeal is that the term “substantially pure” should be given a different meaning when it refers to the para-CPK intermediate, which the claim describes as a “substantially pure regioisomer” of CPK, than when it refers to the piperidine derivative end product. Aventis essentially concedes that if the term at issue were ^substantially pure end product,” 98% purity with respect to all impurities would be an accurate construction. But since the term “substantially pure” is used in claim 1 to refer to the “substantially pure regioisomer” — i.e., the para-CPK intermediate— Aventis argues that a different construction is required. Aventis contends that the evidence relied upon by the district court pertained only to the purity of the end product, and that the term “substantially pure” has a different meaning when used to refer to intermediates than when used to refer to end products. As used in reference to the regioisomer, Aventis argues that the term “substantially pure” means “largely but not wholly [para-CPK], as compared to [meta-CPK].”
The majority embraces Aventis’s proposed construction, holding that the patent gives the term “substantially pure” different meanings when referring to the terms “substantially pure regioisomer” and “substantially pure piperidine derivative.” As the district court ruled, however, the intrinsic evidence does not distinguish between the way “substantially pure” is used with respect to those two terms, and for that reason I would uphold the district court’s claim construction.
In seeking to distinguish between the meaning of the term “substantially pure” when it is applied to the intermediate re-gioisomer as opposed to when it is applied to the piperidine derivative end product, Aventis relies on the argument that one of ordinary skill in the art would know that the purity of intermediates may be different from the purity of end products, and it offers expert testimony in support of that proposition. But even if a person of ordinary. skill in the art would not necessarily regard purity as meaning the same thing for an intermediate as for an end product, the analysis does not end there.
In at least two places, the intrinsic record uses the term “substantially pure” in the same way with regard to the regioi-somer and the end product. First, the specification states:
Although the second mixture of regioi-somers [an intermediate] and the third mixture of regioisomers [the final piperi-*1379dine derivative product] can be analyzed by HPLC experiments, a practical separation to obtain gram quantities of substantially pure regioisomers has not been achieved.
Each mixture (including the first [also an intermediate]), would be expected to contain 33% of the para isomer and 67% of the meta isomer. Since these components are inseparable, it has not been possible to obtain either of the regioi-somers in each [first, second, and third] mixture in substantially pure form.
'703 patent at col. 4 11. 16-24 (emphasis added). Second, in an interference involving the related '610 patent, the patentee wrote:
When read in light of the specification, one skilled in the art would have understood that the phrase “substantially pure”, as used in claims 1-17 of the D’Ambra Patent [the '610 patent], to mean that the subject compound has pharmaceutical grade purity and is in a form purer than that attained by the prior art (e.g., U.S. Patent Nos. 4,254,-129, 4,254,130, 4,285,957, and 4,285,958 to Carr (collectively, “the Carr Patents”). As demonstrated, infra, those skilled in the art recognized that pharmaceutical grade purity requires an impurity level no greater than 2%, and the Carr Patents were unable to achieve such purity.
Importantly, that response refers to claims 1-17 of the '610 patent; one of those claims, claim 12, recites, “a piperi-dine derivative compound produced by a process comprising: providing a substantially pure regioisomer....” (emphasis added).
Aventis concedes that in the first passage the patentee failed to distinguish between the use of “substantially pure” as applied to an intermediate and to an end product, but it claims that the passage is irrelevant because it concerns the prior art Carr process. In fact, however, both the discussion of the prior art and the discussion of the claimed invention use the term “substantially pure” when referring to re-gioisomers; one of the “mixture[s]” referenced in the second paragraph is CPK, while another is the piperidine end product. Thus, the patentee fails to distinguish between “substantially pure regioisomer” and “substantially pure [end product],” and in fact affirmatively suggests that the meaning of “substantially pure” does not turn on whether it modifies “regioisomer” or “piperidine derivative [end product].”
Regarding the second reference, Aventis argues that it is clear in context that the passage concerns the purity of the end product. Aventis also argues that the “subject compound” described in that passage is the end product, making clear that the discussion of “substantially pure” in that passage applies only to the end product. The problem with Aventis’s position is that the quoted language expressly refers to “ ‘substantially pure’ ... as used in claims 1-17,” and claim 12 of the '610 patent refers to a “substantially pure re-gioisomer.” Thus, “‘substantially pure’ ... as used in claims 1-17” unequivocally includes “substantially pure regioisomer.” That reference thus rebuts Aventis’s claim that the patentee was careful to distinguish between “substantially pure regioi-somer” and “substantially pure piperidine derivative [end product].” The patentee could have written “ ‘substantially pure pi-peridine derivative’ ... as used in claims 1-17,” but it chose not to, referring only to “ ‘substantially pure’ ... as used in claims 1-17.” Those two examples show that the patentee did not intend for the term “substantially pure” to have a different meaning depending on whether it was describing an intermediate or an end product.
Beyond those two passages, the intrinsic record provides little else of help in eon-*1380struing the term “substantially pure.” However, general principles of claim, construction are instructive here. “[W]e presume, unless otherwise compelled, that the same claim term in the same patent or related patents carries the same construed meaning.” Omega Eng’g, Inc. v. Raytek Corp., 334 F.3d 1314, 1334 (Fed.Cir.2003); see also Paragon Solutions, LLC v. Timex Corp., 566 F.3d 1075, 1087 (Fed.Cir.2009) (“We apply a presumption that the same terms appearing in different portions of the claims should be given the same meaning unless it is clear from the specification and prosecution history that the terms have different meanings at different portions of the claims.”). Starting with the presumption that “substantially pure” means the same thing when describing “regioisomer” as it does when describing “piperidine derivative,” it is clear that Aventis has not put forth “compelling]” evidence to the contrary.2 Indeed, as noted above, the intrinsic record supports the district court’s finding that “substantially pure” has the same meaning throughout the patent, and Aventis has pointed to nothing compelling in the record to suggest otherwise.3' Aventis’s reliance on expert testimony that one of ordinary skill in the art would know that “substantially pure” can mean different things when describing intermediates than when describing end products is not enough to overcome the persuasive intrinsic record in this case. See Kara Tech. Inc. v. Stamps.com Inc., 582 F.3d 1341, 1348. (Fed.Cir.2009) (“While helpful,, extrinsic sources like expert testimony cannot overcome more persuasive intrinsic evidence.”).
Aventis’s expert testimony does not indicate that “substantially pure regioisomer” is a term of art that connotes a specific level of purity relative to that of the end product. Thus, although “substantially pure” certainly could have different meanings in different contexts, there is no evidence indicating that it must mean something different when used to describe a regioisomer as .opposed to an end product. The majority bases its construction of the term “substantially pure regioisomer” on the general definition of the term “substantially,” which is taken from an unrelated case that in turn cites a general dictionary definition. That appeal to extrinsic evidence from outside the art underscores the fact that Aventis has offered nothing in the intrinsic record, or even in the state of the art, to define the term that is the focus of the parties’ dispute. Instead of resorting to extrinsic evidence as to the general meaning of the term “substantially” standing on its own, we should interpret the claim term that Aventis did define: “substantially pure.”
There is no basis for ignoring the intrinsic record and the presumption that “substantially pure” is a discrete claim term with a consistent meaning throughout the *1381patent. Aventis apparently believes that for “substantially pure” to be construed to have the same meaning each time it is used in the patent, the patentee would have to explicitly “link” the purity of the para-CPK intermediate to that of the end product. But Aventis has it backwards: If the patentee wanted “substantially pure” to have different meanings when applied to different elements, it needed to explicitly “unlink” them.
II
Aventis’s other arguments are easily disposed of. It is clear (and essentially undisputed) that “substantially pure” means “at least 98% pure” when describing end product. In the prosecution history of the '708 patent, the applicant equated substantially pure piperidine derivative with “a purity level suitable for pharmaceutical use.” The district court found that “[i]t is essentially undisputed that pharmaceuti-cally acceptable purity is 98%.” The pat-entee’s statements made in the course of an interference proceeding involving the '610 patent also support the district court’s conclusion that the term “substantially pure,” as used in the '906 patent and its relatives, means “at least 98% pure.” In that interference proceeding, the patentee equated the term “substantially pure” with “pharmaceutical grade purity” and expressly agreed that “pharmaceutical grade purity requires an impurity level no greater than 2%.” The prosecution histories also support the district court’s conclusion that the required purity level referred to purity with respect to all impurities, not just with respect to a single other component, such as meta-CPK.
Aventis argues that this evidence is irrelevant because it pertains to the purity level of the end product. But because the patent does not distinguish between the meaning of “substantially pure” as applied to an end product and as applied to an intermediate, it follows that if a “substantially pure [end product]” means a product that is at least 98% pure with respect to all impurities, then the same meaning attaches to “substantially pure [para-CPK].”
The cases that Aventis cites in support of its position are unhelpful to it. Aventis cites several cases for the proposition that the term “substantially” need not have a strict numerical boundary. E.g., Playtex Prods., Inc. v. Procter & Gamble Co., 400 F.3d 901, 907 (Fed.Cir.2005); Anchor Wall Sys. v. Rockwood Retaining Walls, Inc., 340 F.3d 1298, 1310-11 (Fed.Cir.2003); Cordis Corp. v. Medtronic AVE, Inc., 339 F.3d 1352, 1360 (Fed.Cir.2003); Ecolab, Inc. v. Envirochem, Inc., 264 F.3d 1358, 1366 (Fed.Cir.2001). Those cases have no application here, however, because in this case the intrinsic evidence establishes that the term “substantially pure” was given a strict numerical meaning, as the district court found.
In sum, I conclude that “substantially pure” means “at least 98% purity with respect to all impurities” and that it has that meaning with respect to both regioi-somers and the end product. I would therefore affirm the district court’s judgment.
. The application that issued as the '610 patent was filed as a division of the application that issued as the '703 patent, and the two specifications are essentially the same.
. The majority cites two of our cases for the proposition that the same term in a patent can have different meanings—Microprocessor Enhancement Corp. v. Tex. Instruments Inc., 520 F.3d 1367, 1376 (Fed.Cir.2008), and Epcon Gas Sys., Inc. v. Bauer Compressors, Inc., 279 F.3d 1022, 1030-31 (Fed.Cir.2002). In both of those cases, however, there was a clear basis in the intrinsic record for applying different meanings to the same term. By contrast, there is no compelling evidence that "substantially pure” was intended to mean different things with respect to “regioisomer” and "piperidine derivative.”
. The majority also suggests the presumption of consistent claim construction does not apply in this case because "there is no explicit 'substantially pure' limitation placed on the piperidine derivative end product in the relevant claims of the '703 patent.” However, the term “substantially pure” limits the claimed end product in the related '610 patent, so the presumption applies here. See Omega Eng’g, Inc., 334 F.3d at 1334 (applying presumption of consistency to “the same claim term in ... related patents”).