After a jury reached a verdict in favor of plaintiff Grail Semiconductor, Inc. (Grail), on Grail's action for breach of contract, the trial court ordered a new trial on damages, at the request of defendant Mitsubishi Electric & Electronics USA, Inc. (Mitsubishi). Both parties appeal. Mitsubishi contends that (1) the court should have ordered judgment notwithstanding the verdict (JNOV) instead of a new trial because Grail failed to establish the element of damages, and (2) the court should at least have granted a new trial on liability for breach of contract because it had prejudiced Mitsubishi's
The issues in this action revolved around the disclosure of a new technology in memory chip design created by inventor Donald Stern, cofounder of Grail. Before Stern's invention there were three kinds of memory: SRAM (static random access memory), which is fast but large and expensive; DRAM (dynamic random access memory), which is small and cheap but requires more power and, like SRAM, is volatile (i.e., it loses data when electric power is removed); and FLASH (or "flash") memory, which is nonvolatile (i.e., retains the information even if the electrical power is removed), but relatively slow. Stern's idea was to combine the strengths of those three memory types — "the density of DRAM, [the] speed of SRAM and the non-volatility of FLASH" — in a single chip. Witnesses expressed the view that if the design worked, it would be revolutionary, using words like "disruptive" (a "unique" innovation with "great utility" which "totally changes the market") and the "Holy Grail" of memory technology.
Stern's invention used the property of inductance, which occurs "when current passes through a wire, [causing] a magnetic field to form. And then when the current stops, the magnetic field collapses." "Inductance" had always been considered a "bad thing," an effect to be eliminated; it created extra signals and caused electrical noise and errors to occur in the circuits. Stern's departure from this "conventional wisdom" involved "figuring out how to take the inductive force and actually using it to accelerate the storing of the charge into the layers to allow the device to operate much, much faster." Stern obtained a patent for his design, which he called an "inductive storage capacitor."
In January 2000 Stern and his cousin Robert Stern incorporated a startup company in order to market Stern's technology. Between the last quarter of 2000 and the first quarter of 2003 Grail unsuccessfully tried to recruit investors in the company. By March of 2001 Grail had run out of money.
Grail then gave a PowerPoint presentation describing the technology, including the concept of the use of inductive force. Witnesses gave conflicting testimony about whether any of the information presented at the meeting was confidential. Both Terry and Mark Speizer maintained that Stern had always refused to disclose the details of his technology to anyone, including them; indeed, he appeared to be extremely concerned that someone would try to steal his ideas, and he would not agree to have it tested for investors.
Stern testified that in the afternoon the group met again in Grail's San Jose office. Both Terry and Mark Speizer asserted that there was no afternoon meeting in Grail's office; by that time the office had already been closed to save money. According to Stern, they split up into two groups, and he met privately with Hirayama while the Speizers met with Matsuo about the business plan. During this second meeting, Stern said, he disclosed to Hirayama 16 items of confidential information. Stern showed Hirayama drawings that had been submitted with the patent application and explained the "concept of [using] inductance to store electronic charges." Hirayama took extensive notes in Japanese and made copies of Grail's drawings, including some of the patent application drawings.
Grail's efforts did not culminate in any outside investment or product development. In June 2004, however, Stern read an article regarding technology promoted by Renesas Technology, which had been formed in April 2003 through a joint venture between Mitsubishi Electric Company (Mitsubishi Japan), the parent company of defendant Mitsubishi USA, and Hitachi, Ltd. (Hitachi).
The 2004 article Stern saw discussed a new Renesas memory chip which Stern believed incorporated Grail's technology in violation of the NDA. In June 2007 Grail initiated this action, alleging misappropriation of trade secrets, unfair competition, breach of contract, and related causes of action. Its fifth amended complaint, the operative pleading at trial, filed in late 2009, stated only one cause of action, for breach of contract; it sought not only damages but injunctive and declaratory relief and a constructive trust.
During the 2012 trial extensive testimony was presented on two products publicized in 2004 by Renesas: a new memory cell called SuperSRAM and its use of MONOS (metal, oxide, nitride, oxide, silicon)
Yuji Kihara, the designer of SuperSRAM, testified that he did not communicate with Hirayama or Matsuo about the design of SuperSRAM during his time at either Mitsubishi or Renesas. Kihara also stated that SuperSRAM did not combine the features of flash, DRAM, and SRAM. It improved the performance and solved the problems of conventional SRAM at reduced cost, but could not achieve "as high a density as flash or DRAM." Kihara did state that SuperSRAM had the same capacity as DRAM (larger than SRAM), and that "circuit wise it mainly uses DRAM." But he also explained the reduced soft error rate,
A critical part of Kihara's testimony regarding the SuperSRAM design was whether inductance was involved. Kihara stated that he did not contemplate its having an inductive effect, nor did testing reveal any. His product, he insisted, had "nothing to do with inductance."
The other semiconductor invention at issue was the new MONOS technology, codeveloped by Shinichi Minami, another Hitachi employee who had transferred to Renesas in 2003, and his colleague, Professor Kamigaki. Minami brought with him the work he had focused on at Hitachi in
One of plaintiff's expert witnesses, Joseph McAlexander, examined the structures of both Renesas products and compared them to the Grail devices, with the assistance of Chipworks, a respected company that used advanced optical techniques to analyze the components of electrical devices. In McAlexander's opinion, Mitsubishi had breached the NDA by disclosing Grail's confidential information to Renesas, which then used it for its products. SuperSRAM, he explained, appeared to be a composite device combining the features of DRAM and SRAM. Ten of the 16 items of confidential Grail information disclosed by Stern had appeared in the SuperSRAM chips, a "highly improbable" result. As to the MONOS MCU, while the design had, according to Minami, been finalized in late 2000, McAlexander explained that Renesas had "evolved it" so as to make it faster, with a "much smaller footprint." The MONOS memory structure, he stated, included an inductive capacitor, which allowed less energy to be used. Overall, in the MONOS product Renesas had "really developed a considerable competitive edge in terms of speed and power and retention as compared to competitors." And those critical changes were made to existing technology after the transition to Renesas in 2003. Of the 16 items of confidential information disclosed to Mitsubishi, 11 had been implemented in the MONOS embedded flash MCUs. Again, McAlexander said, it would be "highly improbable" that these 11 items would be found in a specific design without some prior knowledge.
Grail offered three measures of damages through the testimony of Joseph Gemini. Gemini first suggested that damages should be based on Grail's expectations consistent with its business plan — that is, the projected net cash flow, discounted by 15 percent, if Mitsubishi had bought the technology. That amounted to $123,898,889. A second approach, Gemini explained, was to calculate the royalty Mitsubishi would have paid for use of the technology, based on the Renesas sales resulting from that use. Discounted by 7.5 percent, that royalty (estimated as a lump sum) amounted to $203 million. A third measure was unjust enrichment, or "the benefit that Mitsubishi would have realized for taking ... and using the technology unauthorized." That amount was a portion of the operating profit that Renesas had realized from using the Grail technology, which was $252 million.
The jury found that Mitsubishi had breached the NDA Matsuo had signed at the outset of the April 19, 2001 meeting, and that Grail was harmed by the breach. In its verdict the jury determined damages to amount to $123,898,889.00. Judgment on the verdict was entered on May 16, 2012.
The trial court orally denied the motion for JNOV, finding substantial evidence to support the verdict on the issue of breach. It also denied Mitsubishi's new-trial motion on each ground except one: the jury had used an incorrect measure of damages, resulting in an excessive award. Mitsubishi then filed its notice of appeal.
In its May 16, 2012 judgment on the jury verdict the court awarded prejudgment interest and injunctive relief. Mitsubishi moved to vacate the judgment, as it had been entered without support in either the law or the facts and because it had not had an opportunity to be heard on such relief. At a hearing on July 24, 2012, the court suggested that the grant of a new trial made unnecessary an order for interest or an injunction.
Grail challenged the court's finding as to the injunction, citing a provision for injunctive relief in the NDA and asserting the necessity of the remedy for Grail's future commercial opportunities. At a hearing on Grail's renewed motion for an injunction on September 7, 2012, the court agreed with Mitsubishi that Grail had not shown the likelihood of irreparable harm that could not be compensated by damages, particularly since Grail had agreed to an injunction followed by an immediate stay. The court filed its written order denying the request on September 18, 2012. Grail then filed a notice of cross-appeal.
Mitsubishi first contends that the trial court granted the wrong remedy by ordering a new trial instead of JNOV. In its view, Grail simply failed to present sufficient evidence of damages, which required a defense judgment as a matter of law.
If the trial court had granted a new trial based on inadequacy of Grail's proof of the damages element of its claim, Mitsubishi would have a more tenable argument. (See Carter v. CB Richard Ellis, Inc. (2004) 122 Cal.App.4th 1313, 1327 [19 Cal.Rptr.3d 519] [where weight of evidence did not support jury finding that contract existed, proper remedy was JNOV, not new trial]; Garretson v. Harold I. Miller (2002) 99 Cal.App.4th 563, 575 [121 Cal.Rptr.2d 317] [JNOV properly granted where defect was in plaintiff's failure of proof rather than erroneous ruling by trial court].) But the court here made it clear that the excessive damages award resulted from the jury's application of the wrong measure of damages, not from insufficient evidence that Grail suffered damage.
Mitsubishi argues that Grail has "waived any challenge" to the measure of value used by the jury by not cross-appealing on that ground. But Grail is not making any such challenge. Grail's point is only that substantial evidence of damages was presented through both parties' witnesses. As the issue will be retried, whether the original jury received sufficient evidence of damages under any particular measure is moot.
Mitsubishi attempted in its JNOV motion to refute the viability of Grail's second measure of damages, the amount of a royalty Mitsubishi would have paid for use of the technology. On appeal, Mitsubishi again raises that issue, contending only that the trial court "properly recognized that the evidence offered to support Mr. Gemini's opinion of `what Mitsubishi would have paid
The third measure of damages, unjust enrichment, was also determined to be incorrectly used by Gemini in his calculations. Again it was the measure, not the sufficiency of evidence, that the court rejected. The $252 million figure Gemini cited was based on profits from the sale of Renesas products containing Grail's technology, which, in the court's view, was not "a reasonable basis for estimating what [Mitsubishi] would have paid for a license to use technology disclosed subject to the NDA." No issue is presented contesting this ruling.
Mitsubishi next argues that the erroneous admission of two exhibits offered by Grail was prejudicial and necessitated a new trial on liability. At trial Grail called William Keeley, a senior director at Renesas, to testify as the Renesas company representative. Keeley was unfamiliar with the MONOS chip; he was therefore unable to confirm that Renesas's MONOS chip was an "inductive capacitor" or that it trapped electrons through "inductives"; indeed, he could not even "say for certain that the devices [he] sold used a
Grail then showed Keeley PTX-19, a similar document derived from Renesas's American Web site. Like PTX-20, this exhibit represented that the MONOS chip had an inductor. Mitsubishi did not object to the admission of PTX-19, though Renesas, which was monitoring the trial, did object, because the witness could not state "how the marketing department created [the document], whether the person who created it had a duty to create it in a certain way or to meet the business record exception." The court overruled the objection. Joseph McAlexander subsequently testified that PTX-19 showed that an inductor was part of Renesas's MONOS technology.
Mitsubishi responded with evidence suggesting that the word "inductor" in the Web sites was a mistranslation from the Japanese version, which should be translated as "dielectric," an insulating material that does not conduct electricity. Keeley explained that Renesas news releases were normally written in Japan and later translated before being sent to the United States and Europe, with various degrees of preliminary editing and occasional inaccuracies. In this case, he believed that a translation error had led to the use of the word "inductor" in PTX-20. An alternative translation contained in DTX-267 made more sense to Keeley; it used the word "dielectric" rather than "inductor." Shinichi Minami disagreed with even the representation that Renesas's MONOS chip trapped electrons by using an inductor; the word "inductor" in PTX-19, he insisted, was a mistranslation from the Japanese. Harry Tredennick, a defense expert, had obtained a certified translation and agreed with Minami and Keeley about the translation error. Tredennick suggested that the confusion resulted from the similarity of the Japanese characters used in "dielectric" and "inductor"; two of the characters were the same in the two words, while the third was different. Tredennick was convinced that there was no inductive storage capacitor in the MONOS MCU, though he agreed that any memory cell that did would be new and useful.
In its motion for a new trial Mitsubishi argued that PTX-19 and PTX-20 were erroneously admitted as business records. Keeley, it maintained, was not qualified to establish the trustworthiness of the documents, because (1) he
On appeal Mitsubishi renews its argument that the court should not have admitted these two "smoking gun" exhibits, because they were the "cornerstone of Grail's case" and contained highly prejudicial hearsay for which no exception existed.
Evidence Code section 1271, the hearsay exception cited by the trial court in its ruling, provides: "Evidence of a writing made as a record of an act, condition, or event is not made inadmissible by the hearsay rule when offered to prove the act, condition, or event if: [¶] (a) The writing was made in the regular course of a business; [¶] (b) The writing was made at or near the time of the act, condition, or event; [¶] (c) The custodian or other qualified witness testifies to its identity and the mode of its preparation; and [¶] (d) The sources of information and method and time of preparation were such as to indicate its trustworthiness."
The superior court found the documents admissible under this provision because the qualifying witness, William Keeley, had testified that PTX-20 had the "look and feel" of a Renesas marketing document and that PTX-19 "generally look[ed] the same" to him as PTX-20. Moreover, the URL on the documents indicated that they "did indeed come from Renesas's North American and European websites."
In this case the jury heard abundant evidence, independent of PTX-19 and PTX-20, that Renesas's MONOS chip used an inductor, particularly through the testimony of Joseph McAlexander. In addition, the challenged exhibits pertained exclusively to the MONOS chip, while McAlexander's analysis of the SuperSRAM technology also suggested use of Grail's confidential information. After admitting the exhibits the court allowed contrary defense evidence indicating that the word "inductor" in the exhibits was a mistranslation from the Japanese word for "dielectric." The jury thus heard both sides of the issue, received documents purporting to be the same but in different languages, and reached a conclusion favorable to Grail. Having examined the asserted error in light of the entire record, we cannot find prejudice from the admission of the challenged documents. Consequently, the trial court did not err in declining to grant a new trial on liability.
As noted earlier, the trial court's initial judgment awarded Grail injunctive relief as requested in the second cause of action of the fifth amended
On appeal, Grail contends that it was entitled to an injunction upon the denial of Mitsubishi's motion for JNOV because (1) the NDA provided for this remedy,
"The first two grounds `embody the requirement that to obtain an injunction a plaintiff ordinarily must show that the defendant's wrongful acts threaten to cause irreparable injury, meaning injury that cannot adequately be compensated in damages.'" (DVD Copy Control Assn., Inc. v. Kaleidescape, Inc. (2009) 176 Cal.App.4th 697, 722 [97 Cal.Rptr.3d 856] (Kaleidescape),
Grail acknowledges the standard of review. "The grant or denial of a permanent injunction rests within the trial court's sound discretion and will not be disturbed on appeal absent a showing of a clear abuse of discretion." (Horsford v. Board of Trustees of California State University (2005) 132 Cal.App.4th 359, 390 [33 Cal.Rptr.3d 644]; see Shapiro v. San Diego City Council (2002) 96 Cal.App.4th 904, 912 [117 Cal.Rptr.2d 631].) The burden is on the party challenging the ruling to demonstrate such abuse. (Jameson v. Five Feet Restaurant, Inc. (2003) 107 Cal.App.4th 138, 146 [131 Cal.Rptr.2d 771].)
Here Grail failed to carry its burden to show that damages for Mitsubishi's breach would be insufficient to prevent any future harm. It did not demonstrate that Mitsubishi was continuing to disclose information in violation of the NDA; its focus was on Renesas's use of the technology for Renesas products. That conduct was itself the subject of a separate patent infringement action pending in federal court.
The orders are affirmed. The parties shall bear their own costs.
Rushing, P. J., and Premo, J., concurred.
Mitsubishi's expert, Brian Nappert, also testified that the harm to Grail was "the value ... of the state of the technology at the time of the April 2001 meeting." His estimate of damages was $2.1 million to $3.1 million. Under an unjust enrichment theory, he calculated the same amount.