dissenting.
I do not agree with the majority’s interpretation of “the combustion ... is a substantially isothermal process” limitation as recited in claim 1. In particular, I find the disputed claim language ambiguous, and the specification reveals that claim 1 does not require combustion as a result of the second fuel fraction to be substantially isothermal from the beginning until the end of that combustion. I therefore respectfully dissent.
The district court initially construed “the combustion ... is a substantially isothermal process” to mean that “average cylinder temperature remains substantially constant after the second fraction of fuel is supplied.” J.A. 5335-37. Subsequently the district court set aside that construction and narrowed its interpretation of the term, construing the limitation to mean “the combustion resulting from the second fuel fraction results in average cylinder temperature that remains substantially constant from the beginning until the end of that combustion.” J.A. 9-12. Accordingly, the district court narrowed the limitation with a durational requirement— from the beginning until the end of that combustion — which the majority now affirms.
I agree with the majority that claims should be construed based on precepts of English grammar, that the doctrine of claim differentiation is not a hard and fast rule, and that claims need not encompass every disclosed alternative embodiment. See Majority Op. 949, 951-52. I also agree that this court is not required to rewrite claims to preserve their validity. See id. at 952.
Independently, these judicial doctrines aiding in claim construction are not dispos-itive, and are frequently rebutted. However, when a particular interpretation of a claim simultaneously violates all of these doctrines, such a construction may not be tenable. In fact, a construction that excludes disclosed embodiments, violates the doctrine of claim differentiation, and invalidates a dependent claim cannot stay true to the claim language and the written description of the invention. Power Inte*957grations, Inc. v. Fairchild Semiconductor Int’l, Inc., 711 F.3d 1348, 1361 (Fed.Cir.2013) (“The construction that stays true to the claim language and most naturally aligns with the patent’s description of the invention will be, in the end, the correct construction.”).
The majority holds that the only permissible reading of the limitation “the combustion ... is a substantially isothermal process,” requires a substantially constant temperature for the entire second fraction combustion. According to the majority, “is” ties the subject (“combustion”) to the description following (“a substantially isothermal process”). Because the subject is introduced with the definite article “the” (“the combustion of the second fraction”) the majority concludes that the ordinary and customary meaning requires “the” entire combustion — from the beginning to the end — to be isothermal. The majority holds the word “substantially” in the “substantially isothermal process” limitation modifies only the “isothermal process.” Although that interpretation is reasonable, the claim language also supports an at least equally plausible alternative -interpretation.
“Substantially” can also refer to the duration of the combustion. For example, combustion initiated by a fuel fraction may begin at a certain temperature which may subsequently increase over time eventually arriving at an isothermal state. In that instance, the average cylinder temperature may remain substantially constant even if the cylinder was not in an isothermal state from the beginning of the combustion. Indeed, the district court’s earlier construction encompassed this interpretation of “substantially isothermal process.” By including the durational requirement, however, the majority effectively negates the patentee’s use of the approximate term “substantially.” Without this term, the claim would read: “the combustion ... is a[n] ... isothermal process,” which is more parallel to the majority’s construction requiring the isothermal process to occur from the beginning until the end of the combustion. The introduction of “substantially” modifies this strict requirement by allowing the combustion as a result of the introduction of the second fuel fraction to fluctuate prior to the combustion settling into an isothermal state. This claim language is ambiguous.
Recourse to the specification resolves this ambiguity. Claims do not stand alone, and the remaining intrinsic evidence aids in construing the disputed term. Phillips v. AWH Corp., 415 F.3d 1303, 1315 (Fed.Cir.2005) (en banc). The specification is highly relevant and usually dispositive in any claim construction analysis making it the single best guide to the meaning of a disputed term. Id. (internal quotation marks and citation omitted).
Unlike the claim language, the specification provides ample guidance and context, which necessarily conflict with the majority’s claim construction. For instance, the “ideal” combustion processes, '562 col. 11 11.18-19, depicted in Figures 4(A)^4(C) are instructive. Figure 4(B) is shown below:
*958[[Image here]]
In this figure, Path 3 to 4 represents an ideal isothermal combustion process as a result of a second fuel fraction that uses the remaining fuel not expended during Path 1 to 2 and Path 2 to 3. Id. col. 7 11. 4-6. The specification explains how “[r]eal engine paths will depart to some extent from the ideal cycles shown [in Figure 4] due to timing, heat and friction losses. These factors will manifest themselves in the cycle diagram as, for example, rounded corners and displacements of the process.” Id. col. 11 11. 19-24. The “rounded corners” and “displacements” as applied to Path 3 to 4 will introduce, inter alia, shifting and/or concavity to the depicted straight line altering the depiction of a perfectly isothermal state to one that is “substantially isothermal.” Figure 8, an embodiment of the invention at issue, illustrates such a deviation from the “ideal” cycle.
In reference to Figure 8, the specification discloses combustion as a result of the second fuel fraction that can be divided into two sequential states: (1) combustion occurring at constant pressure, and (2) combustion occurring at constant temperature. Id. col. 11 1. 64-col. 12 1. 2. Figure 8 is depicted below:
*959[[Image here]]
As illustrated, the combustion at Path 4 to 5 occurs at constant pressure and is provided by the introduction of a second fuel fraction. Id. col. 11 11. 64-66. The combustion during Path 5 to 6 occurs at constant temperature — isothermally. Id. col. 11 1. 66-col. 12 1. 2. This figure and its corresponding disclosure in the specification clearly indicate that the combustion need not be isothermal from the beginning of the combustion, as the majority’s construction requires. Rather, Figure 8 shows that the isothermal process may begin at some later time after the second fuel fraction is introduced. This combustion process reasonably may be characterized as “substantially isothermal.” Accordingly, a departure from a perfectly isothermal process encompasses combustion that may be “substantially” isothermal — one that may not necessarily begin with and end in such a state — much like the embodiment depicted in Figure 8 and those described as deviations from the “ideal” cycles illustrated in Figure 4.
The majority nevertheless reasons that a claim construction need not encompass every embodiment, and determines that the “ordinary meaning” of “substantially isothermal process” excludes the embodiment illustrated by Figure 8. Majority Op. 951. Even if the claim language were as clear as the majority believes, the majority’s construction excludes, not only the embodiment depicted in Figure 8, but also those embodiments described in the '562 patent that encompass “real” engines. As discussed above, the patent is directed to embodiments that depart from “ideal” combustion processes, which may not be perfectly isothermal from the beginning until the end of the combustion. Thus, the majority’s requirement that the isothermal process occur from the beginning until the end of the combustion excludes these disclosed embodiments as well. That result cannot be correct. Nellcor Puritan Bennett, Inc. v. Masimo Corp., 402 F.3d 1364, 1368 (Fed.Cir.2005) (“The fact that the construction adopted by the district court and advocated by [appellee] would have the effect of placing all the embodiments of the invention outside the scope of the claims is powerful evidence that the court’s construction is incorrect.”).
*960In addition to improperly excluding these disclosed embodiments, the majority’s construction does not give sufficient weight to the doctrine of claim differentiation. Figure 8 is particularly claimed in dependent claim 4, which reads in part:
[T]he second fraction of the total fuel is supplied so as to provide, first, constant pressure combustion until a preselected maximum combustion temperature is reached, and secondly, isothermal combustion at said preselected maximum temperature.
'562 patent col. 13-11. 1-5. Claim 4 depends from claim 3, which depends from claim 1 including the disputed “substantially isothermal process” limitation. “It is axiomatic that a dependent claim cannot be broader than the claim from which it depends.” Alcon Research, Ltd. v. Apotex Inc., 687 F.3d 1362, 1367 (Fed.Cir.2012); AK Steel Corp. v. Sollac & Ugine, 344 F.3d 1234, 1242 (Fed.Cir.2003) (“Under the doctrine of claim differentiation, dependent claims are presumed to be of narrower scope than the independent claims from which they depend.”). By construing “substantially isothermal process” in claim 1 to exclude the processes expressly recited in dependent claim 4, the majority broadens the scope of a dependent claim over the independent claim from which it depends, violating the doctrine of claim differentiation.
Nonetheless, the majority reasons that the presumption the doctrine of claim differentiation provides is rebutted “where the ordinary meaning of the words in an independent claim do not admit to a construction that would embrace a claim dependent therefrom.” Majority Op. at 951. At the same time, however, the majority renders claim 4 superfluous. That cannot be. See, e.g., Rhine v. Casio, Inc., 183 F.3d 1342, 1345 (Fed.Cir.1999) (“Claims should be so construed, if possible, as to sustain their validity.”) (internal quotation marks, citation, and altered capitalization omitted).
Specifically, because a dependent claim narrows the claim from which it depends, it must “ ‘incorporate ... all the limitations of the claim to which it refers.’” Alcon Research, 687 F.3d at 1367 (quoting 35 U.S.C. 112 ¶ 4 (2006)). Indeed, context of the claims as whole informs the proper construction of terms. Abbott Labs. v. Syntron Bioresearch, Inc., 334 F.3d 1343, 1351 (Fed.Cir.2003). Claim 1 recites broadly that combustion as a result of the introduction of the second fraction is a substantially isothermal process. Consequently, the combustion resulting from the second fuel fraction recited in dependent claim 4 while describing two sequential states — isobaric and isothermal — as a whole must still be “substantially isothermal.” Requiring the combustion to be isothermal from beginning to end negates claim 4 and threatens its validity. A construction which simultaneously violates the doctrine of claim differentiation and writes out claim 4 must be erroneous.
For these reasons, I would construe “the combustion ... is a substantially isothermal process” to mean that “average cylinder temperature remains substantially constant after the second fraction of fuel is supplied.” Accordingly, I would reverse the district court’s claim construction ruling and remand for further proceedings.