The opinion of the court was delivered by: MARILYN PATEL, Chief Judge, District
Claim Construction Memorandum and Order for United
States Patent Numbers 6,233,041, 6,377,336, 6,392,740,
6,008,500, and 5,810,832
On December 21, 2001, plaintiffs Nikon Corporation and Nikon
Precision, Inc. (collectively "Nikon" or "plaintiffs") brought a patent
infringement action against defendants ASM Lithography B.V. and ASM
Lithography, Inc. (collectively "ASML" or "defendants").*fn1
things, plaintiffs' complaint alleges infringement of four patents:
United States Patent Number 6,233,041 ("the `041 patent"), United States
Patent Number 6,377,336 ("the 336 patent"), United States Patent Number
6,392,740 ("the `740 patent"), and United States Patent Number 6,008,500
("the `500 patent"). All four patents pertain to photolithographic and
microlithographic machines used in the manufacture of integrated
Defendants timely answered plaintiffs' complaint, later asserting
inequitable conduct and antitrust counterclaims. Some of these
counterclaims grew from plaintiffs' alleged infringement of United States
Patent Number 5,801,832 ("the `832 patent"), an ASML-held patent also
pertaining to photolithographic machinery. Nikon then filed a motion to
dismiss defendants' counterclaims. On July 19, 2002, the court denied
without prejudice plaintiffs' motion to dismiss. Now before the court are
the parties' memoranda regarding claim construction of the disputed
patent terms. The court has considered fully the parties' arguments and
submissions, and for the reasons set forth below, the court enters the
following memorandum and order.
Both plaintiffs*fn2 and defendants*fn3 develop and produce systems
and components used in photo- and micro-lithography. Used to manufacture
semiconductor integrated circuits, lithographic machines create extremely
small and precise patterns of electronic circuitry on integrated circuit
chips. A subset of photolithographic technology, microlithographic
machines*fn4 transfer minute*fn5 pattern features to a
substrate e.g., a silicon wafer and are the only machines
capable of creating the kind of integrated circuit chips needed in the
construction of electronic equipment.*fn6 Nikon, ASML, and Canon are the
only manufacturers of microlithographic machines.
To construct a working integrated circuit chip, a microlithographic
machine prints a circuitry pattern on each layer of a circuit chip. This
printing process often utilizes projection lithography, a type of
lithographic patterning that relies on a light source and a pair of
optical systems to transfer circuitry patterns. At the start of a
projection lithography process, a light-sensitive covering (a
"photoresist") is applied to a "wafer" (or "substrate"), a grouping of
thin layers of circuitry situated on a base of silicon.*fn7 Also at the
start of the process, a transparent piece of glass or quartz (a "mask" or
"reticle") is coated partially with chrome such that a pattern of opaque
and transparent features emerges. Both the mask and the wafer are then
placed in a projection exposure apparatus, the wafer deposited on a
"wafer stage," the mask on a "mask stage." Using an optical illumination
system, light is then cast onto the mask Light shines through the
transparent portions of the mask into a projection optical system. This
projection optical system which is made up, in pertinent part, of
a precision lens focuses the pattern of light features onto the
wafer, leaving an image of the pattern in the photoresist layer. The
process is repeated for each chip and each layer of a chip-on the
Photolithography machines are complex and expensive devices, but they
do not embody entirely new technology. Pioneered in the 1950s and 1960s,
photolithography devices are the subject of numerous inventions and a
comparable number of patents. Some of these inventions and patents
address the "periodic structure" of mask patterns i.e., the size
and series of the transparent and opaque spaces along a mask. When light
is projected onto a mask, some light passes through the surface of a mask
without diffraction, creating "zero-order diffracted" light; as light is
shined onto a mask, other light (viz., "non-zero-order diffracted"
light) changes path after contact with the edges of the opaque portions
of the mask As the periodic structure of a pattern grows increasingly
fine, two things occur: first, non-zero-order diffracted light exits the
mask at increasingly large angles; second, more light traveling along the
optical axis will strike the mask perpendicularly, diffracting at such
large angles that it cannot be captured by the projection optical
"Off-axis" illumination attempts to address this loss of light In
off-axis illumination, light strikes the mask at a non-perpendicular
angle, i.e., from a direction set-off from the optical axis itself. In
this way, zero-order diffracted light is inclined to a degree between the
zero-order and first-order diffraction, permitting more of both types of
light to enter the projection optical system than wold be possible
through "on-axis" illumination. Per wavelength of illuminating light,
then, off-axis illumination produces greater diffraction and allows the
use of finer mask patterns.
Even with off-axis illumination, a photolithographic machine requires
thousands of components and parts to function properly. The machines
require, inter alia, an adequate light source, an illumination
optic assembly (including, e.g., lenses, mirrors, and the like), a
projection optic system to focus the light pattern, and a system or
technique to limit the vibrations that occur as a consequence of wafer
and mask movement. The five patents at issue in this action address
putative advances on a number of these machine components; four of these
patents are held by Nikon, and one is held by ASML.
Titled "Exposure Method Utilizing Diffracted Light Having Different
Orders of Diffraction," the `041 patent was issued on May 15, 2001.
See `041 Patent at 1. Put generally, the `041 patent addresses
a particularized method of transferring a fine pattern from a mask to a
substrate through photolithographic projection exposure. Nikon describes
the `041 patent as a marked and distinct advance in preceding off-axis
technology, technology that used "annular illumination" (that is,
illumination in the shape of a complete ring) centered on the optical
axis. The `041 patent, Nikon details, uses symmetrical, off-axis pairs of
higher-illumination intensity areas in lieu of full ring illumination,
adjusting the pair-spacing to suit particular mask patterns. See
id. at 14:48-16:36.
The parties dispute the meaning of terms in two independent `041 claims
(viz., claim 1 and claim 12) and in six dependent `041 claims (viz.,
claims 2, 3, 4, 7, 8, and 13).*fn8 In an "Initial Determination on
Violation of Section 337 and Recommended Determination on Remedy and
Bond," the United States International Trade Commission ("ITC") addressed
some, but not all, of the `041 claims raised here. See In the Matter
of Certain Microlithographic Machines and Components Thereof
Investigation No. 337-TA-468 (January 29, 2003) (Bullock, A.L.J.).
Labeled "Projection Exposure Apparatus," the `336 patent was registered
on April 23, 2002.
See Patent `336 at 1. Building on the `041 patent, the `336
patent concerns a particular photolithographic apparatus that uses,
inter alia, an irradiation optical system, "fly-eye" optical
integrators, and a Fourier transform plane. Id. The parties
dispute the meaning of particular terms in four independent `336 claims
(viz., claim 1, claim 14, claim 17, and claim 25) and in three dependent
`336 claims (viz., claim 2, claim 8, and claim 18). The parties agree
that the same meaning and construction should apply to terms shared by
the `041 and `336 patents.
Like the `336 patent, the `740 patent is titled "Projection Exposure
Apparatus." See `740 Patent at 1. Also like the `336 patent,
the `740 patent covers an off-axis illumination system that uses prisms
and lenses to redirect light into pairs of intensely lit,
symmetrically-spaced areas. Through adjustment of both annular and
multipole illumination, the `336 and `740 apparatuses both work to
achieve the optimal angle of illumination for particular patterns. Terms
used in four independent (viz., claim 3, claim 6, claim 8, and claim 10)
and two dependent (viz., claim 7 and claim 9) `740 claims are disputed by
the parties. The parties agree that terms used in the `041, in the `336,
and in the `740 patents should be construed in the same manner in each
The `500 patent, titled "Exposure Apparatus Having Dynamically Isolated
Reaction Frame," was registered on December 28, 1999. See
Patent `500 at 1. Focused on the vibrations caused by acceleration and
deceleration of the wafer and mask stages, the `500 patent depicts a
"reaction frame" constructed to receive reaction forces generated by the
mask stage drive. In two embodiments in the `500 patent, the reaction
frame is shown*fn9 somehow isolated from the frame in which the core
photolithography apparatus sits. The parties dispute terms used in three
`500 claims: claim 1, claim 4, and claim 6.
One of the many challenges in the photolithographic process is
achieving proper alignment of the mask pattern relative to particular
substrate layers. For many years, lithography machines relied on a
process of "global alignment" to achieve adequate mapping; the process of
"global alignment" sought to align in one step the entire mask pattern
with the entire substrate plane.
As related aspects (e.g., reduction lens systems) of photolithographic
technology evolved, "global alignment" proved an unsatisfactory alignment
alternative; it took too long, for example, and it inadequately
accommodated the myriad, often mimscule, variations along a large wafer
surface. For a time, a process of
"field-by-field leveling" replaced "global alignment";
"field-by-field" alignment permitted recalibration and realignment for
each discrete substrate sub-area, thus eliminating some but not
all of the minute discrepancies overlooked in the "global
alignment" process. In some ways, "field-by-field leveling" proved a
market improvement over "global alignment" methods, but this type of
"field-by-field leveling" was not without flaw. Repositioning of the
substrate demands time and space, both precious commodities in the
circuit-chip manufacture process; put another way, "field-by-field
leveling" left room for technological advances regarding both cost and
time of production.
The `832 patent aimed to improve the relatively slow and costly
"field-by-field" alignment method. Like prior art, the invention covered
by the `832 patent sought to permit accurate exposure of a mask pattern
on a substrate. Unlike the prior art, however, the `832 patent art
employed precise laser interferometer positioning to do so. Rather than
by mapping mask marks and (substrate) sub-area marks during the exposure
process, the `832 patent art uses of a system of lasers and
mirrors-which, together, constitute "interferometers" to achieve
sufficient alignment and to correct for infinitesimal tilts in the
surface of the substrate sub-area. "Interferometers" measure substrate
displacement along five separate axes (i.e., directions) of
movement.*fn10 Through this kind of interferometer-based positioning,
both the space and time loss incident to "global" and "field-by-field"
alignment would be better minimized.
The parties dispute terms used in six `832 patent claims: Claim 1,
claim 5, claim 8, claim 15, claim 16, and claim 17.
Under Markman v. Westview Instruments. Inc., 52 F.3d 967, 979
(Fed. Cir. 1995), aff'd 116 So. 0 Ct. 1384 (1996), the court
"has the power and obligation to construe as a matter of law the meaning
of language used in the patent claim." When determining the meaning of
claim language, the court principally considers three types of intrinsic
evidence: the language of the claim, the patent specification, and the
relevant prosecution history. See, e.g., Vitronics Corp. v.
Conceptronic. Inc., 90 F.3d 1576, 39 U.S.P.Q.2d 1573 (Fed. Cir.
1996). These three types of intrinsic evidence provide the foundation
for and a kind of order of operations in-claim construction
generally. In construing the meaning of claim language, the court looks
first at the language of the claims themselves; where the claim language
is not sufficiently instructive, courts may then refer to the relevant
patent specifications, turning, if necessary, to the prosecution history,
if in evidence. Id. at 1582-83. Where claim language is "clear
on its face," the court's
"consideration of the rest of the intrinsic evidence is restricted
to determining if a deviation from the clear language of the claims is
specified," Interactive Gift Exp., Inc. v. CompuServe Inc.,
256 F.3d 1323, 1331 (Fed. Cir. 2001), and Federal Circuit doctrine directs
courts to construe disputed claim language according to "an objective
test of what one of ordinary skill in the art at the time of the
invention would have understood the term to mean." Markman, 52
F.3d at 986; see also Teleflex, Inc. v. Ficosa N. America
Corp., 299 F.3d 1313, 1324 (Fed. Cir. 2002) ("The words used in the
claims are interpreted in light of the intrinsic record, including the
written description, the drawings, and the prosecution history, if in
evidence."); Bell Communications Research, Inc. v. Vitalink
Communications Corp., 55 F.3d 615, 620 (Fed. Cir. 1995) (noting that
the words of the claims themselves drive construction of claims). Unless
a patentee invests a particular claim term with a different definition,
patent language is understood to convey its ordinary meaning to one
skilled in the art. See Intellical, Inc. v. Phonometrics, Inc.,
952 F.2d 1384, 1387 (Fed. Cir. 1992). When courts look to the
specifications for clarification of ambiguous claim terms, courts must
still avoid reading "limitations appearing in the specification . . .
into [the] claims." Intervet Am., Inc. v. Kee-Vet Lab., Inc.,
887 F.2d 1050, 1053 (Fed. Cir. 1989).
In most cases, intrinsic evidence will be sufficient to resolve
ambiguities and to determine the meaning of the claim terms.
Vitronics, 90 F.3d at 1583. Only when intrinsic evidence proves
inadequate may the court refer to extrinsic evidence, e.g., expert
testimony and germane textbooks. Even then, courts must use extrinsic
evidence only as an aid in "coming to the proper understanding of the
claims" and the underlying technology, id, and only to the
extent the evidence helps illuminate the language of the patent
documents. Markman, 52 F.3d at 979-81. And even when
"enlightened by  extrinsic evidence," "[t]he district court's claim
construction . . .[must] still [be] based upon the patent and prosecution
history." Id. at 981 (noting that courts may not use extrinsic
evidence to vary or to contradict claim language). When considering
extrinsic evidence, the Federal Circuit has evinced a preference for
dictionaries and prior art documents, generally eschewing reliance on
expert testimony except as a last resort. Id. at 1585.
In an action for patent infringement, claim construction is the first
part of a two-part analysis. See Intellectual Property
Development Inc. v. UA-Columbia Cablevision of Westchester. Inc.,
336 F.3d 1308, 1313 (Fed. Cir. 2003). The second part of this test is the
determination of whether the accused device or method infringes the
claims at issue. Id. On this second step, the burden of proof
typically rests on the party claiming that its patent has been infringed,
see Wilson Sporting Goods v. Davis Geoffrey & Assoc.,
904 F.2d 677, 685 (Fed. Cir.), cert. denied. 498 U.S. 992 (1990),
though only the infringement
portion of the action is tried to a jury or to a finder of fact.
Claim construction issues are questions of law, see Markman v.
Westview Instruments Inc., 517 U.S. 370, 389-90 (1996), and are not
subject to traditional burdens of proof. The duty of the court is to
consider all appropriate evidence-regardless of who produced
it when assessing the proper interpretation of claims.*fn11
See Vitronics. 90 F.3d at 1576.
The parties dispute a host of terms used in the `041 patent claims. The
court addresses each of these claims, and discrete claim terms, below.
The parties dispute the meaning of seven terms used in claim 1: (1)
"pattern"; (2) "mask"; (3) "projection optical system"; (4) "illuminating
the pattern with at least a first light beam and a second light beam from
different directions"; (5) "a first light beam and a second light beam";
(6) "0-order" and "non-0-order" diffracted beams; and (7)
"substrate."*fn12 Each term is assessed separately below.
The parties agree that the word "pattern" has a well-established
"ordinary meaning," Vitronics, 90 F.3d at
1582 specifically, "an arrangement of lines or shapes; a design
according to which something is to be made" and the parties appear
to agree that the use of the indefinite article "a" connotes "one or
more." See Crystal Semiconductor Corp. v. TriTech Microelectronics
Int'l. Inc., 246 F.3d 1336, 1347 (Fed. Cir. 2001); Pall Corp.
v. Micron Separations. Inc., 66 F.3d 1212, 1216 (Fed. Cir. 1995).
The parties do not agree, however, about how the ordinary meaning of
"pattern" fits within the context of claim 1.
Plain language suggests that "pattern" denotes a design or series of
marks in a semiconductor integrated circuit that is to be transferred to
a photoresist layer of a substrate. In the relevant art, the term
"pattern" signifies a series often of a particular design or
repetition-of circuit features in a semiconductor integrated circuit to
be transferred to a photoresist layer of a substrate, and when it uses
"pattern," the claim language depicts the exposure of a particular kind
of pattern-viz., a circuitry pattern-onto a mask or circuitry chip.
See Pitney Bowes. Inc. v. Hewlett-Packard Co., 182 F.3d 1298,
1311 (Fed. Cir. 1999) (requiring each "term [to] be read to correspond to
the only plausible meaning in each context"). Specification language, in
turn, references the "patterns" imposed on semiconductor memory devices
and liquid crystal components through the process of photoresist-based
photolithographic projection. See, e.g., `041 Patent at
1:19-24. As it is used in the relevant claim and specification language,
possesses specified meaning, not by virtue of improperly imported
limitations, cf. N. Telecom Ltd. v. Samsung Elecs. Co.,
Ltd., 215 F.3d 1281, 1290 (Fed. Cir. 2000), but through claim and
specification language itself. See Pall Corp., 66 F.3d at 1216.
The term "pattern" is used to denote "a design or series of marks in
a semiconductor integrated circuit that is to be transferred to a
photoresist layer of a substrate." Neither party offers a more
appropriate construction, and the court thus construes "pattern"
Construction of the term "mask" follows similar lines. As with
"pattern," the parties agree that the word "mask" has an established
meaning in both common and scientific idioms. Read as a part of claim 1,
in fact, "mask" carries a specific and well-established meaning: It
denotes an item containing a circuit pattern in a semiconductor
integrated circuit. See, e.g., `041 Patent at 1:19-24. Both the
context of the claim, see Pitney Bowes, 182 F.3d at 1311, and
the specification language buttress this construction. See `041
Patent at 3:15-37 (discussing the manufacture of semiconductor and liquid
crystal devices). By comparison to the parties' over-and
under-generalized definitions, the court's construction better captures
the meaning of "mask" while leaving related-but distinct-claim terms,
purposes, and connotations outside the definition of this oft-used term.
The court construes "mask" to mean "an item, in a semiconductor
integrated circuit, on which a circuit pattern is placed"
3. "projection optical system"
During previous stages of this litigation, the parties did not dispute
the meaning of the term "projection optical system"; indeed, they
apparently stipulated before the ITC that the term should be construed as
"a lens system or other component or components that project or expose a
pattern onto an object." Before this court, ASML still favors this
construction, but Nikon now proposes an alternative, asking the court to
read the term to mean "a collection of optical components for forming an
image of a pattern onto a photo resist layer on a substrate."
Neither party disputes that the "projection optical system" constitutes
part of the larger lithographic exposure apparatus. See Patent
`041 at 5:29-37. Nor does either party dispute that the role of the
"projection optical system" is to project an image of a pattern onto a
specific substrate. Id. But by comparison to this limited (and
particularized) role, the parties' proposed constructions either
overspecify or overgeneralize the operation of the relevant art.
Cf. id.; see also id. at 3:17-37 &
11:60-12:45 (discussing the image formation process). Nonetheless,
without resort to the parties' imprecise definitions,
the court can construe the claim term by reference to the words'
ordinary meaning and by reference to the intrinsic evidence. As the claim
phrase itself suggests, a "projection optical system" is "a component or
combination of components that projects (i.e., transfer) a mask pattern
onto a substrate." The specification language, in turn, repeatedly refers
to "projection optical system" as the mechanism by which mask patterns
are translated to a substrate. See, e.g., Id. at
12:55-13:62; 15:42-16:53. Consistent with plain meaning and the lessons
of the specification language, the court construes the term "projection
optical system" to mean "a component or combination of components
that transfers or translates a mask pattern onto a substrate."
4. "illuminating the pattern with at least a first light beam
and a second light beam from different directions"
Because the court has construed the term "pattern," the court need only
construe two portions of the phrase "illuminating the pattern with at
least a first light beam and a second light beam from different
directions": one, "first . . . and  second light beam"; and, two, "from
different directions." The court addresses the two in turn below.
a. "first . . . and  second light beam"
In the photolithographic pattern-transfer process, two light beams
(viz., the "first light beam" and the "second light beam") illuminate a
pattern from no fewer than twelve different directions. Upon contact with
the pattern, the two beams are diffracted into distinct orders of
diffracted beams; these diffracted beams, in turn, travel through the
projection optical system along a shared optical path. To differentiate
between these two beams, the claim refers to "first" and "second" "light
beams." According to Nikon, the "first . . . and  second light beam"
claim language should be construed to denote "beams of light that are, at
least during some portion of their paths, separate and discrete." ASML
does not contest inclusion of the concluding phrase of Nikon's
construction, but ASML offers a substantially more specific alternative;
according to ASML, the court should construe "first" and "second light
beams" to mean "separate and discrete beams, as are produced by the
disclosed spatial filter, emanating from discrete areas on the Fourier
transform plane (like holes in the spatial filter)."*fn14
The court does not disagree with ASML's presentation of the relevant
technology; the beams do indeed emanate from discrete areas, interacting
at a point with spatial filters. But the court cannot adopt ASML's
expansive reading of "first . . . and  second light beam" to mean
"[s]eparate and discrete beams, as are produced by the disclosed spatial
filter, emanating from discrete areas on the Fourier transform plane
(like holes in the spatial filter)." As the court understands the
intrinsic record and the claim language, all that
is at issue is two separate light beams that are, for at least part
of their lengths, separate. See, e.g., `041 Patent at
13:55-14:17. Nothing in the patent specifications undercuts the thrust of
this understanding, and the court thus construes "first . . . and 
second light beam" to mean "two beams of light that are, for some
portion of their paths, separate and discrete."
b. "from different directions"
The parties agree that the court's construction of the term "from
different directions" should begin with the phrase "[t]he `first light
beam' and the `second light beam' illuminate the `pattern' at different
angles of incidence." The parties do not agree, however, regarding where
the definition of the claim should end: ASML believes that "at different
angles of incidence" itself offers a sufficient construction of the claim
term; Nikon, by contrast, asks the court to detail the kind of different
angles of incidence at issue, appending the apparently non-exhaustive
example, "which include angles having the same magnitude but different
directions." Cf. In the Matter of Certain Microlithographic Machines
and Components Thereof Investigation No. 337-TA-468 at 285
(rejecting a similar, though notably different, Nikon-proffered
As the court reads them, the parties' constructions are not mutually
exclusive, whether linguistically or logically. The term "different
angles of incidence," if understood as an incorporative category,
undoubtedly includes some angles "having the same magnitude but different
directions," so what Nikon seeks to add plainly falls within the
technological capacity of the invention, if only as an example. But the
limitation Nikon seeks to import is unsupported by the claim text and the
relevant specification language. See `041 Patent at 3:31-3:50
& 12:13-45. Where claim language conduces to ready explication, the
Federal Circuit has long reminded, courts should construe claim terms to
mean precisely what they say. See Vitronics, 90 F.3d at 1582.
However innocuous (and technologically valid) Nikon's proposed
illustration, there is no reason to venture from the plain meaning of the
claim terminology here. The claim expressly discusses beams emerging
"from different directions"-i.e., from distinct and unshared angles of
incidence. Cf. Teleflex, Inc. v. Ficosa North America Corp.,
299 F.3d 1313, 1328 (Fed. Cir. 2002) ("We have `cautioned against
limiting the claimed invention to preferred embodiments or specific
examples in the specification.'") (citation omitted). Nothing in the
intrinsic evidence controverts this claim language, and Nikon posits no
compelling reason to read into the claim an example (viz., "which include
angles having the same magnitude") of something the claim's language
already if implicitly embraces. Consistent with the
intrinsic evidence, the court construes "from different directions"
to mean "at distinct and unshared angles of incidence."
5. "O-order" and "non-O-order" diffracted beams
In the photolithographic process, when the "first" and "second light
beams" strike a mask pattern, some light diffracts directly back along
the same axis; some does not. Claim 1 places these different types of
diffracted light into two categories viz., "0-order diffracted
beams" and "non-0-order diffracted beams" noting what role these
beams play in the microlithographic projection procedure. Claim 1 does
not otherwise define the "0-order diffracted beam" and "non-0-order
diffracted beam" terms, and Nikon now asks the court to construe the
"0-order diffracted" and "non-0-order diffracted" terms to mean "light
beams, rays or components formed when light from respective localized
areas of relatively higher light intensity diffracted by a mask pattern."
Before the ITC, Nikon offered an identical construction, and, at that
time, ASML apparently acceded to it.
Before this court, however, it appears that ASML has altered its
position, asking the court to define "O-order diffracted beams" and
"non-0-order diffracted beams" as nothing more than "0-order diffracted
ray[s] of light and other higher order diffracted rays of light such as
the 1st-order, 2nd-order," and the like. The court is mindful that the
"0-order diffracted" and "non-()-order diffracted" terms carry, in
certain contexts, purely descriptive, adjectival meaning, though not
necessarily the circular meaning ASML suggests. In reference to a light
beam, for example, "0-order diffracted" means simply that the beam is not
diffracted to a particular (or any) order of magnitude. Since
"diffraction" has a readily ascertainable meaning in the art (namely, the
phenomenon exhibited by wave fronts that, passing the edge of an opaque
body, are modulated, thereby causing a redistribution of energy), it
follows that a "0-order diffracted beam" is a beam in which the energy
has been modulated and redistributed to the "0-order." In the context of
claim 1, moreover, the "0-order diffracted" and "non-0-order diffracted"
modifiers are used only vis-a-vis specific light beams, specifically
those formed when light from localized areas of higher-intensity light is
diffracted by a mask pattern. See `041 Patent at 18:21-26.
Specification language buttresses this understanding of the claim
terms, and it allows the court to avoid resort to ASML's largely
tautological approach. See`041 Patent at 3:31-50 & 9:32-41. "0-order
diffracted beams" and "non-0-order diffracted beams" have specific
meaning in the context of the claim, and the court must construe the
terms to this end See Pitney Bowes. 182 F.3d at 1311. For these
reasons, the court construes "0-order diffracted beam" to mean "a
light beam or ray formed when a
mask pattern diffracts light back directly along the
illumination axis"; in turn, "non-O-order diffracted beam" means
"a light beam or ray formed when a mask pattern diffracts light off
the illumination axis."
As with "pattern" and "mask," the parties agree that the word
"substrate" has an established meaning in scientific parlance. See,
e.g., McGraw-Hill Dictionary of Scientific and Technical Terms
(defining "substrate" as, inter alia, "the physical material on
which [a] microcircuit is fabricated"). As with "pattern" and "mask,"
though, the parties disagree regarding the breadth the construction of
"substrate" should take. ASML contends, and the court does not disagree
entirely, that "substrate" signifies "an item such as a photosensitive
member that is exposed with a pattern." But however initially valid
ASML's broad phrasing, the term "substrate" does more than signify a
purely generic type of device. "Substrate," as it is used in claim 1,
denotes an item (e.g., a wafer) to which a photoresist layer is affixed;
the transfer of the pattern through the photolithographic exposure
process has nothing to do with the meaning of the bare "substrate" term.
A narrower, more particularized usage comports with ordinary meaning of
the term and with the specification language, see `041 Patent
at 1:19-24, two things the court plainly cannot ignore. See
Vitronics, 90 F.3d at 1582.
When construing the "substrate" term, moreover, the court need not
articulate a definition that reiterates the meaning of an entire claim or
that revisits the function of an entire invention. To a significant
degree, Nikon's attempt to affix "during a lithographic operation" as a
modifier to its definition like its "exposed with a pattern"
addition aims to fold a general description of the lithographic
process into the meaning of the unadorned "substrate" term, spurring
unnecessary redundancy in the name of claim construction. The claim
language suggests that a "substrate" is an item on which a photosensitive
layer is placed, and the specification language supports this simple
understanding. See, e.g., `041 Patent at 1:19-24 & 3:15-37.
Nothing more need be added. Thus, consistent with this intrinsic
evidence, the court construes "substrate" to mean "an item on which
a photosensitive layer or pattern is formed or placed"
B. Claims 2, 3, 4, 7, and 8
The parties dispute the meaning of three terms found in dependent
claims 2, 3, 4, 7, and 8: one, "Fourier transform plane"; two,
"illumination optical system";*fn15 and, three, "fineness of said
pattern." The court addresses the terms seriatim.
1. "Fourier transform plane"
The parties appear to agree that the term "Fourier transform
plane" itself a term of art has a plain and ordinary meaning: The
mathematically calculable grouping of points generally corresponding to
or substantially near the pupil plane of a projection optical system. The
patent discusses a Fourier transform plane in both spatial and functional
ways, noting both where such a plane might lie and what purpose such a
plane may serve in the lithographic process. Id.
As Nikon correctly suggests, of course, the relevant specifications
discuss more than the Fourier transform plane when delineating the
construction of the optical system. "[O]ptical paths," the specifications
teach, are placed "substantially equidistant from the optical axis of the
projection optical system at or in the vicinity of the Fourier
transform plane," see `041 Patent at 3:42-46 (emphasis added);
see also id. at 4:45-48 (". . . equal distance from
the optical axis of the projection optical system at or in the vicinity
of the Fourier transform plane"), and "a spatial filter" is "arranged at
the Fourier transform plane or the illumination optical system or in
the vicinity of the exit end of the integrator element."
Id. at 11:3-11 (emphasis added). With these lessons in mind,
the court cannot doubt that the specification language contemplates both
a Fourier transform plane and a plane conjugate to the pupil plane,
adding that planes optically conjugate to particular Fourier transform
planes operate analogously to the Fourier transform plane in the
photolithographic process. Were the court required to assess the
operation of the Fourier transform plane and all of its
functional equivalents, then, Nikon would be correct that any definition
of the "Fourier transform plane" term should embrace the appendix "or a
plane conjugate to the pupil plane."
But it is not for the court to evaluate the lithographic function of
the Fourier transform plane and all of its operational equivalents.
Rather, the court's task is to define "Fourier transform plane" as a
distinct and independent claim term.*fn16 See SRI Int'l v.
Matsushita Elec. Corp., 775 F.2d 1107, 1121 (Fed. Cir. 1985) (en
banc); see also DeMarini Sports. Inc. v. Worth. Inc.,
239 F.3d 134, 1314 (Fed. Cir. 2001). "Fourier transform plane," as noted,
has long been understood to denote a mathematically calculable plane
generally corresponding to or substantially near the pupil plane of a
projection optical system.*fn17 Nothing in the specification contravenes or
expands this plain definition of "Fourier transform plane," see `041
Patent at 14:25-36; 15:23-30, and nothing in the intrinsic record
suggests that, as a discrete phrase, "Fourier transform plane" includes
distinct planes, whether optically conjugate to the pupil plane or not.
Bounded by the scope of the parties' dispute, the court thus construes
"Fourier transform plane" to mean "a mathematically calculable plane
generally corresponding to or substantially near the pupil plane
of a projection optical system" That the plane is so
calculable does not require that the finder of fact perform the
complex calculation, nor does it mean that such a plane is not measurable
optically; it only attaches to the term the standard, widely-accepted
definition of "Fourier transform plane."
2. "illumination optical system"*fn18
ASML and Nikon agree that the definition of the term "illumination
optical system" should denote, in some way, a system constituting a
component, or group of components, that directs or otherwise acts on an
illumination beam. The parties' disagree, however, about the generality
with which this definition should be posited. ASML favors a broad
construction of the term; Nikon advocates a narrow one.
Specification language refers to and details
collections of tangible optical components, many of which work on light
from an illumination source such that they produce a region of
illumination light having a particular distribution at the mask-pattern
surface. See Patent `041 at 10:51-11:12; see
id. at 1:26-3:37. To the extent Nikon says as much in its
description of the operation of an "illumination optical system," the
court agrees. But to the extent Nikon seeks to fold this language into a
construction of the relevant term, the court cannot agree; that the
system should function properly is inherent to the invention itself
whether in the form of an "illumination optical system" or any
other and the court need not include as much in its construction
of the "illumination optical system" term. All the court need determine
is what an "illumination optical system" is.
The "best mode" description for the `041 patent makes clear that an
"illumination optical system" includes a host of component
parts e.g., an ellipsoidal mirror, a relay lens, and a condenser
lens all organized such that light is directed toward a mask
See id. at 10:64-11:12. When construing the
"illumination optical system" phrase, the court need not venture beyond
what the claim says and what the specification language teaches. As the
claim language says and the specifications teach, an "illumination
optical system" is "an optical component, or combination of optical
components, that directs light from a light source onto and through a
mask pattern." The court construes "illumination optical system" to
mean precisely that.
3. "fineness of said pattern"
According to the parties' claim construction memoranda, ASML and Nikon
now agree on a construction of this term. As a result, the court need not
construe this phrase.
The parties dispute the meaning of two terms in claim 12: (1) "defining
an intensity distribution of the
illumination light on the Fourier transform plane in the
illumination optical system with respect to a pattern on the mask to have
increased intensity portions apart from an optical axis"; and (2)
`'determining the position of the increased intensity portions."*fn19
1. "defining an intensity distribution of the illumination
light on the Fourier transform plane in the illumination optical system
with respect to a pattern on the mask to have increased intensity
portions apart from an optical axis"
Distilled to its essence, the parties' disagreement over this prolix
claim term centers on two words: "defining" and "on." ASML reads "on" to
modify "defining" such that the phrase "defining an intensity
distribution of the illumination light on the Fourier transform plane"
necessarily places the act of defining "on" the Fourier
transform plane itself; for its part, Nikon argues that the claim
language demands no such placement, connoting only the existence of "an
intensity distribution of the illumination light on the Fourier transform
plane in the illumination optical system."
At least in part, basic principles of grammar*fn20 contradict Nikon's
parsing of the claim text. The claim language's use of a gerund-form verb
(viz., "defining") unequivocally implies an act. See Anhydrides
& Chemicals. Inc. v. United States. 130 F.3d 1481, 1483 (Fed.
Cir. 1997) (applying the basic rules of grammar when evaluating a
statute); Gen. Foods Corp. v. Studiengesellschaft Kohle mbH.
972 F.2d 1272, 1274 (Fed. Cir. 1992) ("[E]ach claim is an entity that
must be considered as a whole."). When claim 12 uses the word "defining,"
it denotes the act of assigning a definition to an intensity distribution
of an illumination light; Nikon cannot construe the claim such that this
act does not occur.
Yet precisely where this act of "defining" occurs presents a
substantially more difficult question. In its proposed construction, ASML
asks the court to place the act of "defining" directly on the "Fourier "
transform plane," arguing that the location of a particular prepositional
phrase (viz., "on the Fourier. . . .") in the claim text compels the
court to place the act on the relative position of the plane between the
pattern and the light source. As the court reads claim 12, however, the
pivotal prepositional phrase could be read to modify the act of
"defining" (as ASML suggests), or it could be read to modify the
occurrence of the illumination light, but not necessarily one, the other,
or both. See `041 Patent at 19:14-20:3 (". . . defining  an
intensity distribution of the illumination light on the Fourier transform
plane"). It is simply not clear that the act must occur where ASML says
it does, nor that it ever does.
In fact, intrinsic evidence and the core technology suggest that it
does not, i.e., that the prepositional phrase modifies the illumination
light, not the act of defining alone. Indeed, much in the intrinsic
record suggests that the Fourier transform plane functions as a location
at which light may exhibit particular
characteristics, not that it is where the "defining" must take
place. See, e.g., `041 Patent at 3:30-40 &
4:20-5:11. And to say as much does not exclude the embodiment in figure
2; the disclosed embodiment is sufficiently incorporated through a
definition denoting the core act itself. Consistent with the claim text
and the lessons of the intrinsic record, the court construes "defining an
intensity distribution of the illumination light on the Fourier transform
plane in the illumination optical system with respect to a pattern on the
mask to have increased intensity portions apart from an optical axis" to
mean "quantifying or shaping of increased intensity portions of the
illumination light, as those portions appear on the Fourier transform
plane, in the illumination optical system"
2. "determining positions of the increased intensity
Much of the foregoing explication of the "defining . . . axis" term
applies to this claim term as well. Like "defining," "determining"
connotes an act; like "defining," "determining" must occur at some
location; and like "defining," "determining" is not hitched to a
particular physical location, including the Fourier transform plane. The
plain language of claim 12 requires only that positions of increased
intensity portions be determined in accordance with the mask pattern such
that a first- and second-diffracted light passes through a common area of
the projection optical system. In this, claim 12's focus is largely
spatial, articulating, inter alia, a general structure of planes
and portions and patterns. But the claim does not demand that the
"determining" occur at any particular location. See,
e.g., `041 Patent at 3:52-4:17; id. at 13:55-14:43;
see also id. at figs. 2-4. Where no limitations exist
in the claim language, and where no limitations are otherwise required to
construe a claim properly, the court is reluctant to import claim-text
limits. For this reason, the court will not import the "on the Fourier
transform plane" language ASML sets forth. Instead, adhering to the plain
meaning of the claim term, the court construes "determining the position
of the increased intensity portions" to mean "setting or
ascertaining the spatial arrangement of increased intensity portions of
D. Dependent Claim 13: "substantially conjugated"*fn21
The parties seem to agree that "substantially conjugated" possesses
meaning customary to those of ordinary skilled in the art, viz., the
near-complete mapping of the points of one plane in an optical system to
a second plane. As the court reads their proposed constructions,
moreover, the parties generally agree that, in the context of claim 13,
"substantially conjugated" denotes the relation of a given plane to the
second portion of increased light intensity. All that remains,
then, is semantics, with ASML asserting that "substantially conjugated"
should be read in one detailed manner, Nikon arguing that the term should
be read in another, equally detailed manner.
Despite the parties' attempts to fill "substantially conjugated" with
overflowing detail, neither party offers a construction that readily fits
the context in which the term is used. In the relevant claim context,
"substantially" denotes the degree to which two locations or positions
are conjugated; that is, as claim 13 uses the term, to be "substantially
conjugated" is to be conjugated to a significant degree. Cf. Cordis
Corp. v. Medtronic Ave, Inc., 339 F.3d 1352, 1360 (Fed. Cir. 2003)
(discussing the meaning of "substantially"); Epcon Gas Sys., Inc. v.
Bauer Compressors. Inc., 279 F.3d 1022, 1031 (Fed. Cir. 2002)
(same). In the relevant claim context, in turn, "conjugated" describes
any pair of locations positioned such that points of the first
map or are amenable to mapping or imaging to the
neighboring points of the second. The teachings of the specification
language are in accord. See `041 Patent at 11:60-12:11. Under
the terms of claim 13, it is an "area" that to be "substantially
conjugated," so "conjugated" denotes an "area" positioned such that its
points map the neighboring points of "one of . . . the increased in
intensity portions." Taking the definitions of "substantially" and
"conjugated" together, the term "substantially conjugated" means
"positioned such that the area's points map or image to a
significant degree to the corresponding points of another area or
plane,"*fn22 Nikon's proposed appendix concerning "all light rays"
is simply extraneous to the relevant term.
As with the `041 patent, the parties dispute the meaning of a myriad of
terms used in the `336 patent claims. The court ...