O'MALLEY, Circuit Judge.

OSRAM SYLVANIA, Inc., ("OSRAM") appeals the district court's grant of summary judgment of invalidity of claims 1, 17, 25, 27, and 32 of U.S. Patent No. 5,834,905 (the "'905 patent"), issued to Valery A. Godyak et al., and assigned to OSRAM. Because we find genuine issues of material fact that preclude a finding of anticipation and obviousness on summary judgment, and because we find that the district court erred in failing to consider objective indicia of nonobviousness, we reverse and remand for proceedings in accordance with this ruling.

BACKGROUND

I. Technology at Issue

OSRAM is the assignee of the '905 patent, which claims a closed-loop tubular electrodeless lamp, having certain and specific discharge current and pressure parameters. Claim 1, which is representative, recites:

An electric lamp assembly comprising:

'905 patent col. 8 ll. 7-19.

This invention generally relates to "a low pressure, high intensity fluorescent light source that can produce considerably more light per unit length than conventional electroded fluorescent lamps." Electrodeless fluorescent lamps were first disclosed in several patents issued to and articles authored by John Anderson of G.E. in the 1960s and 1970s. For example, as described in the '905 patent's "Background of the Invention," Anderson claimed in U.S. Patent No. 3,987,334 (the "Anderson '334 patent") a lamp having a torodial discharge tube that forms a continuous closed electrical path that passes through a torodial ferrite core. Applying a voltage to the wire wrapped around the ferrite core creates a magnetic flux that in turn induces a discharge voltage along the tube. In effect, the wire around the ferrite core operates as the primary of a transformer with the lamp tube operating as the secondary of the transformer. Thus, "[t]he inner surface of the discharge tube is coated with a phosphor which emits visible light when irradiated by photons emitted by the excited mercury gas atoms." '905 patent col. 2 ll. 14-16. The lamps described in the Anderson '334 patent operated with a buffer gas pressure of "approximately 1 torr or less." The '905 patent specification describes Anderson's lamp as inefficient and impractical, however.

The '905 patent also describes "a closed-loop, tubular lamp envelope enclosing mercury vapor ... a transformer core disposed around the lamp envelope, an input winding disposed on the transformer core and a radio frequency power source coupled to the input winding" but it specifies a buffer gas pressure of "less than about 0.5 torr." '905 patent col. 2 ll. 31-37. "The radio frequency source supplies sufficient radio frequency energy to the mercury vapor and the buffer gas to produce in the lamp envelope a discharge having a discharge current equal to or greater than about 2 amperes." Id. col. 2 ll. 36-41. Operating a lamp at the parameters specified in the '905 patent — namely a buffer gas pressure of less than 0.5 torr and a discharge current equal to or greater than 2 amperes — allegedly results in a lamp with high output, high efficiency, and long operating life as compared to conventional electroded fluorescent lamps. In its preferred embodiment, the lamp specified by the '905 patent operates with a buffer gas pressure equal to or less than about 0.2 torr, and a discharge current equal to or greater than about 5.0 amperes. The specification of the '905 patent makes clear that it is this novel relationship — high discharge current combined with low buffer gas pressure — that results in the increased efficiency and small power loss in the ferrite core.

II. Procedural History

On November 30, 2009, OSRAM filed a complaint alleging that American Induction Technologies, Inc.'s, ("AITI") 120-watt induction lamp infringes OSRAM's '905 patent. AITI responded with counterclaims seeking declarations of non-infringement, invalidity, and unenforceability. OSRAM moved for summary judgment that the '905 patent was infringed and not invalid and AITI cross-moved for summary judgment of invalidity due to anticipation and obviousness based on the Anderson '334 patent. On February 14, 2011, the district court denied OSRAM's motions and granted AITI's motion with respect to claim 32. The district court ruled, from the bench, in relevant part:

A4200-01. With respect to the remaining claims, the district court characterized the parties' positions as a "battle of the experts" and found that genuine issues of material fact precluded summary judgment in favor of either party.

After the court's finding that the Anderson '334 patent anticipated claim 32 of the '905 patent, AITI sought reconsideration of its motion for invalidity of the remaining independent claims — claims 1, 17, 25, and 27. AITI argued that these other claims differed from claim 32 only with respect to a specified discharge current of greater than 2 amperes and, therefore, were obvious combinations of lamps disclosed in the Anderson '334 patent and those described in an article by D.O. Wharmby, Ph.D., entitled "Electrodeless Lamps for Lighting: A Review," Vol. 140 IEEE Proceedings-A no. 6, November 1993 ("Wharmby"). Notably, AITI relied on no other references or prior art to support its renewed motion for summary judgment. In opposition, OSRAM contended that AITI failed to raise this specific combination of prior art in its original motion for summary judgment and that, regardless, OSRAM's expert presented testimony that the different shapes of the lamps disclosed in the Anderson '334 patent and Wharmby precluded applying the teachings of one to the other. OSRAM, in particular, argued that Anderson '334 patent does not disclose all the shape limitations of claim 17 — requiring two parallel tubes joined by two lateral tubes at each end. OSRAM also noted that, in denying OSRAM's motion for summary judgment, the district court ruled that material issues of fact existed with respect to the teachings of Wharmby. AITI acknowledged OSRAM's arguments regarding the shape of the prior art lamps and removed claim 17 from its motion for reconsideration in reply "to avoid additional distraction from the discharge current teachings of the prior art." J.A. 4142. The District Court granted AITI's motion and declared the remaining independent claims, including claim 17, invalid as a matter of law. The court's ruling was again read from the bench and states, in relevant part:

J.A. 4286-86 [sic throughout]. OSRAM requested clarification regarding the inclusion of claim 17 in the order and the court responded that "shape is irrelevant" and claim 17 was, therefore, invalid as obvious. The district court ordered AITI to submit proposed findings of fact and conclusions of law relating to the independent claims but AITI did not comply with that order and the district court failed to issue its own.

After the independent claims were invalidated, OSRAM granted AITI a covenant not to sue as to the dependent claims, and the district court dismissed AITI's counterclaims concerning the validity of the '905 patent's dependent claims. The parties then proceeded with a three-day bench trial, from September 20-22, 2011, on AITI's counterclaims of inequitable conduct, after which the district court found that no inequitable conduct was committed during the prosecution of the '905 patent. The district court entered final judgment on November 28, 2011. OSRAM timely appealed and we have jurisdiction pursuant to 28 U.S.C. § 1295(a)(1).

DISCUSSION

On appeal, OSRAM argues that the district court erred in granting summary judgment that claim 32 of the '905 patent is anticipated by the Anderson '334 patent and that claims 1, 17, 25, and 27 of the '905 patent are obvious in view of Wharmby in combination with the Anderson '334 patent. According to OSRAM, the district court resolved a disputed issue of material fact in its anticipation finding, namely that the Anderson '334 patent disclosed the operating condition of less than 0.5 torr through its reference to a lamp with "a pressure of approximately 1 torr or less," and resolved numerous disputed issues of fact in its obviousness finding relating the interchangeability of lamp shapes. Additionally, OSRAM asserts that the district court's failure to address evidence of secondary considerations of nonobviousness alone warrants reversal. AITI counters that the earlier disclosed genus — 1 torr or less — necessarily anticipates the later species — less than 0.5 torr — and that the claims are otherwise obvious in light of various combinations of prior art.1 AITI also contends that OSRAM failed to present evidence of a nexus between the claimed invention and the alleged secondary considerations and we, therefore, may overlook the district court's failure to address them below.

I.

This court reviews a district court's grant of summary judgment without deference. Dynacore Holdings Corp. v. U.S. Philips Corp., 363 F.3d 1263, 1273 (Fed. Cir.2004). Summary judgment is appropriate if no genuine issues of material fact exist such that the moving party is entitled to judgment as a matter of law. FED. R.CIV.P. 56(c); see also Ethicon Endo-Surgery, Inc. v. U.S. Surgical Corp., 149 F.3d 1309, 1315 (Fed.Cir.1998). This court reviews the evidence in the light most favorable to the non-moving party. Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 255, 106 S.Ct. 2505, 91 L.Ed.2d 202 (1986).

Anticipation, though a question of fact, may be resolved on summary judgment if no genuine issue of material fact exists. Ormco Corp. v. Align Tech., Inc., 498 F.3d 1307, 1319 (Fed.Cir.2007). We review de novo the determination of whether the evidence in the record raises a genuine issue of material fact. Id. "Summary judgment is proper if no reasonable jury could find that the patent is not anticipated." Zenith Elecs. Corp. v. PDI Commun. Sys., 522 F.3d 1348, 1357 (Fed.Cir. 2008).

Similarly, "a district court can properly grant, as a matter of law, a motion for summary judgment on patent invalidity when the factual inquiries into obviousness present no genuine issue of material facts." Ryko Mfg. Co. v. Nu-Star, Inc., 950 F.2d 714, 716 (Fed.Cir. 1991). "When the facts underlying an obviousness determination are not in dispute, we review whether summary judgment of invalidity is correct by applying the law to the undisputed facts." Tokai Corp. v. Easton Enters., 632 F.3d 1358, 1366 (Fed.Cir.2011).

II.

We begin by addressing the district court's finding that the Anderson '334 patent anticipates claim 32 of the '905 patent. A patent claim is anticipated if each and every limitation is found in a single prior art reference. 35 U.S.C. § 102. Patents are presumed to be valid and invalidity must be proven by clear and convincing evidence. Microsoft Corp. v. i4i Ltd. P'ship, ___ U.S. ___, 131 S.Ct. 2238, 2242, 180 L.Ed.2d 131 (2011). OSRAM argues that AITI faced an enhanced burden in establishing anticipation because the examiner considered the Anderson '334 patent during the prosecution of the '905 patent. While prior consideration of a reference during prosecution may carry some weight, the burden to prove invalidity does not change; at all times, it remains a showing "by clear and convincing evidence." Id.; see also Tokai, 632 F.3d at 1367.

The parties' dispute centers on whether the Anderson '334 patent discloses a buffer gas pressure of less than 0.5 torr. AITI primarily relies on the legal argument that "the disclosure of a small genus may anticipate the species of that genus even if the species are not themselves recited." Bristol-Myers Squibb Co. v. Ben Venue Labs., Inc., 246 F.3d 1368, 1380 (Fed.Cir.2001). Referencing our decisions in Atofina v. Great Lakes Chemical Corp., 441 F.3d 991 (Fed.Cir.2006) and ClearValue, Inc. v. Pearl River Polymers, Inc., 668 F.3d 1340 (Fed.Cir.2012), AITI asserts that the range in the Anderson '334 patent, which completely encompasses the range in the '905 patent, is sufficient to disclose the claimed buffer gas pressure in the '905 patent.2

We disagree. While it is true that an earlier disclosed genus may, in certain circumstances, anticipate a later species, this inquiry necessarily includes a factual component. See, e.g., Sanofi-Synthelabo v. Apotex, Inc., 550 F.3d 1075, 1083 (Fed. Cir.2008) ("[W]hether a generic disclosure necessarily anticipates everything within the genus ... depends on the factual aspects of the specific disclosure and the particular products at issue."); Atofina, 441 F.3d at 999; In re Baird, 16 F.3d 380, 382 (Fed.Cir.1994). For example, in Atofina we analyzed a patent claiming a method of synthesizing difluoromethane at a temperature between 330-450 degrees Celsius. Atofina, 441 F.3d at 993. The prior art, in contrast, disclosed a broad temperature range of 100-500 degrees Celsius. Id. at 999. In Atofina, we reversed the district court's finding of anticipation following a bench trial and we explained that our holding was premised on the "considerable difference between the claimed [temperature] range and the range in the prior art." Id. Our decision in Atofina clarified that the prior art's teaching of a broad genus does not necessarily disclose every species within that genus. Id. Under the circumstances presented in Atofina, "no reasonable fact finder could conclude that the prior art describes the claimed range with sufficient specificity to anticipate this limitation of the claim." Id.

Contrary to AITI's assertions, our decision in ClearValue does nothing to alter the unequivocal statements in Atofina. There, we analyzed a patent relating to clarifying low alkalinity water and found that the jury lacked substantial evidence to find that certain prior art — disclosing a chemical process with a range of 150 ppm or less — did not anticipate the asserted patent or disclose the claimed limitation of 50 ppm. ClearValue, 668 F.3d 1340, 1344-45. In ClearValue, we explained that, in contrast to the patentee in Atofina, ClearValue did not argue that the 50 ppm limitation was critical, or that the claimed method worked differently at different points within the prior art range of 150 ppm or less. Id. And, ClearValue did not allege that one of ordinary skill would not have recognized 50 ppm as an acceptable value for the range provided in the prior art. Id. at 1345.

A correct application of Atofina and ClearValue to the evidence presented here reveals clear disputed material issues of fact and mandates reversal of the district court's finding of anticipation. OSRAM contends — and presented expert testimony and evidence to support its contention — that the limitation of less than 0.5 torr is central to the invention claimed in the '905 patent and that a lamp would operate differently at various points within the range disclosed in the Anderson '334 patent. Specifically, the patent itself claims that prior lamps, including those disclosed in the Anderson '334 patent, are inefficient and impractical as compared to the '905 lamp precisely because they use pressure well above that claimed in the '905 patent.

A trial may reveal a minimal difference between the Anderson '334 patent prior range of "approximately 1 torr or less" and the less than 0.5 torr claimed in the '905 patent, or that one of ordinary skill would interpret "approximately 1 torr or less" as clearly disclosing less than 0.5 torr as an acceptable choice within that range, but those questions cannot be resolved as a matter of law on this record. Indeed, AITI made no effort to rebut the testimony of OSRAM's expert regarding the importance of the claimed pressure to the invention or regarding what one of ordinary skill would have understood from the disclosures in the Anderson '334 patent. AITI relies, instead, on the conclusory claim that less than 0.5 torr necessarily falls within "approximately 1 torr or less" as a matter of fact. While true, the inquiry does not end there. How one of ordinary skill in the art would understand the scope of the disclosure or, stated differently, how one of ordinary skill in the art would understand the relative size of a genus or species in a particular technology is of critical importance. And the district court's failure to explain or support its statement that "the buffer gas pressure disclosed in the Anderson '334 patent completely encompasses that disclosed in the '905 patent within a relatively narrow range," or to justify its complete rejection of OSRAM's expert testimony, complicates our review. We have no choice but to reverse the district court's finding of anticipation on this record.

III.

The district court's obviousness rulings also fail because they depend on resolution of disputed issues of fact. A patent may not issue "if the differences between the subject matter sought to be patented and the prior art are such that the subject matter as a whole would have been obvious at the time the invention was made to a person having ordinary skill in the art to which said subject matter pertains." 35 U.S.C. § 103(a). Obviousness is a question of law based on underlying factual findings: (1) the scope and content of the prior art; (2) the differences between the claims and the prior art; (3) the level of ordinary skill in the art; and (4) objective considerations of nonobviousness. Graham v. John Deere Co., 383 U.S. 1, 17-18, 86 S.Ct. 684, 15 L.Ed.2d 545 (1966). Generally, a party seeking to invalidate a patent as obvious must demonstrate "by clear and convincing evidence that a skilled artisan would have been motivated to combine the teaching of the prior art references to achieve the claimed invention, and that the skilled artisan would have had a reasonable expectation of success in doing so." Pfizer, Inc. v. Apotex, Inc., 480 F.3d 1348, 1361 (Fed.Cir.2007); see also Amgen, Inc. v. F. Hoffman-LA Roche Ltd., 580 F.3d 1340, 1362 (Fed.Cir.2009) ("An obviousness determination requires that a skilled artisan would have perceived a reasonable expectation of success in making the invention in light of the prior art." (citations omitted)). The Supreme Court has warned, however, that, while an analysis of any teaching, suggestion, or motivation to combine known elements is useful to an obviousness analysis, the overall obviousness inquiry must be expansive and flexible. KSR Int'l Co. v. Teleflex, Inc., 550 U.S. 398, 415, 419, 127 S.Ct. 1727, 167 L.Ed.2d 705 (2007).

A.

The district court did no more than recite the test of KSR and draw general conclusions based thereon in its ruling on obviousness. The court did not make any specific findings of fact and gave no basis — other than an admission that it previously failed to consider submissions by the parties — for reversing its prior statement that disputed issues of fact existed. OSRAM recognizes — correctly — that Rule 52(a)(3) states that a "court is not required to state findings or conclusions when ruling on a motion under Rule... 56." But, as OSRAM correctly notes, both this Circuit and the Ninth Circuit repeatedly have made clear that a trial court must at least provide its analysis and grounds for entering judgment somewhere in the record. AITI counters that the district court's findings are ascertainable from the parties' own arguments and again contends that, because we review judgments not opinions, we are free to affirm on any ground supported by the record.

It is not our role to scour the record and search for something to justify a lower court's conclusions, particularly at the summary judgment stage. Whether dealing with an issue of law like claim construction or an issue of fact such as infringement, this court must be furnished "sufficient findings and reasoning to permit meaningful appellate scrutiny." Nazomi Communs., Inc. v. Arm Holdings, PLC, 403 F.3d 1364, 1371 (Fed.Cir.2005) (quoting Gechter v. Davidson, 116 F.3d 1454, 1458 (Fed.Cir.1997)). Where, as here, the record is devoid of meaningful analysis, we will not conduct such an analysis in the first instance. Id.; see also Research Corp. Techs. v. Microsoft Corp., 536 F.3d 1247, 1254 (Fed.Cir.2008) ("Microsoft's motions for summary judgment were granted without a proper analysis regarding inequitable conduct. This court therefore lacks findings and conclusions for adequate review."). We previously have had occasion to remind this very district court of these important principles; we do so once more here.

In TriMed, Inc. v. Stryker Corp., 608 F.3d 1333, 1343 (Fed.Cir.2010), we reversed this district court's obviousness ruling for failing to address, among other things, what the asserted prior art teaches:

Id. AITI tries to limit TriMed to scenarios where the district court explicitly references only "common sense" as the basis for its obviousness findings. There is no justification for limiting TriMed in that fashion, however. The point made there, as here, is that the trial court must explain how it reached the conclusions it does, particularly where there is evidence in the record supporting the non-movant's position. AITI's request that we forgive the lower court's failure to explain its ruling — and insistence that we should review the record in the first instance to ascertain what the lower court might have been thinking — rings particularly hollow given AITI's failure to respond to the trial court's order to submit findings of fact and conclusions of law.

B.

With respect to the substance of the district court's ruling on obviousness, OSRAM primarily takes issue with the district court's pronouncement that "shape is irrelevant" and that "a person of ordinary skill in the art would recognize that increasing radio frequency discharge to 2 or more amperes would improve tubular lamps in the same way that [sic] globular lamps and lamps of other topologies." J.A. 4286. AITI responds that claims 1, 25, 27, and 32 of the '905 patent are prima facie obvious based on Wharmby. According to AITI, Wharmby discloses a lamp meeting the physical configurations of the '905 patent lamp by way of reference to an earlier Anderson article regarding operating conditions within the scope of those claimed by the '905 patent.

Substantial factual disputes exist, however, as to whether one of ordinary skill would consider the lamp shape to be irrelevant or, more specifically, whether one of ordinary skill would have been motivated to combine the Anderson lamp with Wharmby or operate the Anderson lamp at the conditions disclosed in Wharmby. OSRAM's expert, Dr. Lawler, testified that Wharmby "teaches away from the closed-loop tubular design of the '905 patent" and AITI's expert, Mr. Edward Hammer, acknowledged that Wharmby disparages the very external ferrite core, closed-loop tubular design of the Anderson patent that AITI argued would be obvious to combine with Wharmby. J.A. 4101; A4106-07. OSRAM also references Dr. Lawler's opinion that "discharge plasmas are inherently unpredictable and may behave differently in different lamp topologies.... One could not predict results or success simply by taking known discharge plasma behavioral characteristics from globular-style lamps and trying to apply them to a closed-loop tubular lamp with external ferrite cores." J.A. 4108-09.

Additionally, at the inequitable conduct trial, AITI focused on the purported failure of the '905 patent's inventors to disclose certain prior art references about globular lamps, including two articles written by the inventors and Wharmby. The district court, after hearing live testimony, reached the conclusion that the inventors of the '905 patent credibly believed that Wharmby was not material to the '905 patent. While the question of intent and credibility as it relates to inequitable conduct is certainly different than the scope of a prior art reference, this finding highlights the inconsistent and unsupportable nature of the district court's ruling on obviousness and lays bare a disputed issue of fact with respect to the teachings of Wharmby. For at least this reason we reverse the grant of summary judgment and remand.

C.

Separately, and in addition, nothing in the record indicates that the court considered OSRAM's evidence of secondary considerations of nonobviousness. At summary judgment, OSRAM submitted evidence of long felt need, failure of others, and industry praise, evidence which AITI did not rebut below. AITI makes no effort to argue that the district court considered this evidence, or even to claim that it did rebut OSRAM's proffered evidence. Instead, AITI asks that we reject OSRAM's evidence of secondary considerations by finding in the first instance that OSRAM's evidence is not tied adequately to the full scope of the asserted claims. AITI's arguments are without merit. Despite AITI's contentions, we have emphatically rejected any formal burden-shifting framework in evaluating the four Graham factors and the district court's failure to consider the evidence requires reversal. See Wm. Wrigley Jr. Co. v. Cadbury Adams USA LLC, 683 F.3d 1356, 1365 n. 5 (Fed.Cir.2012), In re Cyclobenzaprine Hydrochloride Extended-Release Capsule Patent Litigation, 676 F.3d 1063, 1076 (Fed.Cir.2012), Ruiz v. A.B. Chance Co., 234 F.3d 654, 667 (Fed. Cir.2000); see also Transocean Offshore Deepwater Drilling, Inc. v. Maersk Drilling USA, Inc., 699 F.3d 1340, 1349 (Fed. Cir.2012) (emphasizing the importance of objective indicia of nonobviousness to the ultimate conclusion of obviousness and making clear that "to determine obviousness, the decisionmaker must be able to consider all four Graham factors."); TriMed, 608 F.3d at 1343 ("We have repeatedly held that evidence of secondary considerations must be considered if present.... There is no indication in the record that, as required by our precedent, the court considered at all the evidence of secondary considerations offered by TriMed."). On remand, the district court must, in the first instance, consider secondary considerations before finding a claim obvious.

D.

The parties disagree as to the propriety of the district court's decision to include claim 17 in its ruling on obviousness after AITI expressly removed that claim from its initial summary judgment motion. While a trial court may invalidate a claim sua sponte, where no motion regarding that claim is before it, it may not do so without prior notice to the non-movant and an opportunity for the non-movant to address such an unanticipated action. See Portland Retail Druggists Ass'n v. Kaiser Found. Health Plan, 662 F.2d 641, 645 (9th Cir.1981). Because we reverse the court's determination of obviousness on other grounds, we need not and do not reach this issue.

CONCLUSION

For all of the reasons discussed above, we reverse the judgment of invalidity and remand this matter for trial. While the scope and content of the prior art, the motivation to combine certain references, and the impact of OSRAM's evidence of secondary considerations, all may be resolved in AITI's favor at trial, the district court's treatment of the record on summary judgment was both inadequate and improper. We will not undertake an analysis of anticipation and obviousness that should have been performed by the district court — but was not — in the first instance.

REVERSED AND REMANDED.

COSTS

Each party shall bear its own costs.