Frederick L. Ikenson, Esq.
Blank Rome LLP
600 New Hampshire Ave., NW
Washington, DC 20037

RE: Domestic Interested Party Petition, 19 U.S.C. § 1516; Tariff Classification of Glass Preforms for Optical Fibers; HQ Ruling 967058 dated April 21, 2006

Dear Mr. Ikenson:

This is in response to the domestic interested party petition, dated August 4, 2006, filed by you on behalf of Corning Incorporated (the petitioner), pursuant to Section 516 of the Tariff Act of 1930, as amended (19 U.S.C. § 1516). At issue is whether CBP has correctly classified imported merchandise of “the class or kind” termed “glass preforms for optical fibers.” 19 C.F.R. § 175.0. The petitioner states that it is a manufacturer in the United States of merchandise of the same “class or kind,” and such qualification is a necessary prerequisite to file this petition. 19 C.F.R. § 175.2(a). The petitioner specifically challenges HQ 967058, dated April 21, 2006, an administrative ruling wherein CBP classified certain imported glass preforms for optical fibers in subheading 7002.20.1000 of the Harmonized Tariff Schedule of the United States (HTSUS), which provides for, among other things, glass rods that are unworked.

Pursuant to 19 U.S.C. § 1516 and the U.S. Customs and Border Protection (CBP) Regulations, 19 C.F.R. § 175.21(a), notice of receipt of the domestic interested party petition was published on March 28, 2007, in the Federal Register, Volume 72, Number 59. The petitioner has requested that CBP initiate a proceeding under Section 625(c) of the Tariff Act of 1930, as amended (19 U.S.C. § 1625(c)), to revoke HQ 967058 and reclassify the subject merchandise in either of subheadings 7020.00.60 or 9001.10.00, HTSUS. Subheading 7020.00.60 is a basket provision that provides for other articles of glass not elsewhere covered in Chapter 70 of the HTSUS. Subheading 9001.10.00 provides for, inter alia, optical fibers. Two comments were received in response to this notice, both of which agreed with CBP's position that the glass preforms described in HQ 967058 should be classified in subheading 7002.20.1000, HTSUS. Heading 7002, HTSUS, is an eo nomine provision that covers unworked glass rods. CBP has classified the designated imported glass preforms for optical fibers in this heading based on its finding that these goods are solid multi-layer glass rods of fused silica that are unworked in their condition as imported. The petitioner’s fundamental argument against CBP’s classification of the glass preforms in HQ 967058 is that the visible differentiation between the two layers of these glass rods is de facto proof that the core area is a “worked” rod of glass. CBP disagrees with this superficial standard. The appearance of an article can be evidence of further working, but it is not dispositive. The actual test for whether an article has been “worked” requires a factual inquiry into its manufacture and any subsequent processing prior to importation. For the reasons set forth in this letter, CBP affirms that the designated imported merchandise described below is properly classified as an eo nomine unworked glass rod of heading 7002 of the HTSUS.

The description of the manufacture of the glass preforms described in HQ 967058 was incorporated into that ruling by reference to prior HQ Ruling 960948, dated September 11, 1998. Specifically, HQ 960948 provided the following description: The preforms under consideration are produced by a two-step process. In the first step, the core layer of the preform is produced by a method called "Vapor Axial Deposition" (VAD). Extremely fine "dusts" or "soots" of silica tetrachloride and additional chemicals are grown or deposited on the end of a "target" rod, forming a column of the "dust" or "soot" material. The column is drawn through a furnace, fusing it into a rod and releasing the chlorine. In the second step, the cladding layer of the preform is added by fusing to the outside of the core rod a layer of silica dioxide powder. Such a two-step production process is described in Kirk-Othmer, supra, 616, as follows:

... Sometimes a two-step process can be employed for efficiency. A preform is made which is roughly half core and half cladding. The sintered preform is then drawn into rod and then overclad with pure silica soot to obtain the appropriate core/ clad ratio.

After deposit of the silica dioxide powder or soot to form the cladding layer of the preform, the "target" rod on which the core layer was deposited or grown is removed. This phase of the process is described in Fiber Optic Reference Guide, supra, 16, as follows: ... When the deposition is complete, the rod is removed, and the deposited material is placed into a consolidation furnace. The water vapor is removed, and the preform is collapsed to become dense, transparent glass.

The core and cladding of the preform consist of glass with different refractive indexes. ***** The resulting article, in the form of a rod approximately 62 millimeters in diameter and 1500 millimeters in length, may be flame polished using an oxyhydrogen flame to achieve a smooth surface, if necessary.

The glass preforms described above are solid multi-layered rods made from fused silica. The visible differentiation between the core and cladding results from the different refractive indices of the glass. The core is the light-guiding region of an optical fiber, while the cladding, which has a different index of refraction than the core, ensures that the light signal remains within the core as it is carried along the fiber’s length. See e.g., HQ 964879, dated March 21, 2002. The cladding also functions as a protective layer for the core. See HQ 964879.

The threshold issue in the petitioner’s challenge to CBP’s classification of these glass preforms in heading 7002, HTSUS, turns on whether this visible differentiation evidences the existence of what the petitioner describes as “a glass core rod which has been ‘worked’ by the addition to it of a layer of cladding glass.” After addressing this issue, the remainder of this letter responds to the petitioner’s legal arguments regarding the classification of the class or kind of merchandise consisting of glass preforms for optical fibers within heading 7002 and other provisions of the HTSUS.

A. The Designated Imported Merchandise Is an Unworked Glass Rod of Heading 7002

Classification under the HTSUS is made in accordance with the General Rules of Interpretation (GRIs). GRI 1 provides that the classification of goods shall be determined according to the terms of the headings of the tariff schedule and any relative section or chapter notes. In the event that the goods cannot be classified solely on the basis of GRI 1, and if the headings and legal notes do not otherwise require, the remaining GRIs 2 through 6 may then be applied in order. GRI 6 provides that the classification of goods in the subheadings of a heading shall be determined according to the terms of those subheadings, and any related subheading notes and, mutatis mutandis, to GRIs 1 though 5, on the understanding that only subheadings at the same level are comparable. Additionally, the Harmonized Commodity Description and Coding System Explanatory Notes (ENs) constitute the official interpretation of the Harmonized System. While not legally binding on the contracting parties, and therefore not dispositive, the ENs provide a commentary on the scope of each heading of the Harmonized System and are thus useful in ascertaining the classification of merchandise under the system. CBP believes the EN's should always be consulted. See T.D. 89-80, 54 Fed. Reg. 35127, 35128 (Aug. 23, 1989).

The terms of heading 7002 cover “[g]lass in balls (other than microspheres of heading 7018), rods or tubes, unworked.” It is uncontroverted that the glass preforms for optical fibers at issue in HQ 967058 are rods of glass within the scope of this heading. The petitioner’s legal argument concerns the meaning of the term “unworked.” There are no applicable section or chapter notes for this heading. “When a tariff term is not defined in either the HTSUS or its legislative history, the term’s correct meaning is presumed to be its common meaning in the absence of evidence to the contrary.” Timber Products Co. v. United States, 515 F.3d 1213, 1219 (Fed. Cir. 2005) (citing Rohm & Haas Co. v. United States, 727 F.2d 1095, 1097 (Fed. Cir. 1984)). The petitioner has not provided any such evidence to the contrary or otherwise asserted a commercial meaning in this case. Furthermore, the ENs do not provide any commentary on the scope of this “unworked” for purposes of heading 7002, HTSUS.

The Federal Circuit Court of Appeals has already determined the common meaning of “work” under the HTSUS as “to form, fashion, or shape an existing product.” Winter-Wolff, Inc. v. United States, 966 F. Supp. 1258, 1265 (Ct. Int’l Trade 1998). An existing product is one that “already exists as a commercial product.” Id. It follows that in the context of heading 7002, a glass rod is “unworked” if it is a commercial product of glass that has not been formed, fashioned or shaped beyond its manufacture into the shape of a rod.” To prevail in its argument that CBP erroneously classified the designated imported merchandise in heading 7002, the petitioner must demonstrate as a matter of fact that before the cladding soots were applied, the core area (1) existed in the shape of a rod; and (2) constituted a commercial product.

1. The Designated Imported Merchandise Described in HQ 967058 Is a Glass Preform for an Optical Fiber That Was Not Preceded by a Core Rod of Glass

CBP ruled in HQ 967058 that the glass preform was not formed, fashioned or shaped beyond its manufacture into the shape of a rod. Based on past rulings, CBP explained that it has uniformly considered the working of glass to be a process that has been performed on an extant article. The factual inquiry in that case was simply whether the core existed in the form of a solid glass rod before the “cladding” layer was applied. With respect to the manufacturing process described above, the core was not in the form of a solid glass rod until after the target rod was removed and the two layers of glass soots sintered onto the target (core and cladding layers) were simultaneously consolidated into a single, solid multilayered rod in a furnace. Until this point of consolidation, the core was merely a layer on the target rod. CBP concluded that the imported glass preform was not further formed, fashioned or shaped after this consolidation, and was “unworked” in its condition as imported.

Our decision in HQ 967058 dispelled the gravamen of the petitioner’s argument that a rod with an observable layer of “cladding” distinct from its core is de facto “worked” within the meaning of heading 7002. The uniqueness of the VAD process described in that ruling yields an intermediate product in the manufacture of an optical fiber, but it is nonetheless an “unworked” glass rod. The complexity of the manufacturing process does not preclude classification of the resulting product in heading 7002. While the VAD process and the resulting glass preform may not have been envisioned when the eo nomine provision for glass rods originated under the HTSUS, Congress could not have intended to foreclose future innovations in [goods] from classification under the [eo nomine] provisions. Simmon Omega, Inc. v. United States, 83 Cust. Ct. 14, C.D. 4815 (1979).

An interested party that files a petition under 19 U.S.C. § 1516 designates imported merchandise of the same class or kind as that which is manufactured by the interested party. 19 U.S.C. § 1516(a). The manufacturing processes vary,  but the class or kind of merchandise which shares the same general physical characteristics as the designated imported merchandise in HQ 967058 consists of highly specialized rods of pure silica that exhibit precise variation between the refractive indexes of their core and cladding regions. The designations of the “core” and “cladding” regions in the technical literature which describes glass preforms are not evidence that the glass preform was manufactured in a sequential process of discrete steps that began with an extant core rod to which cladding is subsequently applied. As one commenter pointed out, these “technical references” to the core and cladding areas of a preform do not support the conclusion that the core was an extant glass rod prior to manufacture of the preform. CBP finds that the mere fact that “cladding” is a technical designation for the outer layer of a glass preform for an optical fiber does not support the ipse dixit conclusion drawn by the petitioner that the cladding was applied to a pre-existing core rod. Empirically, the petitioner’s conclusion is disproven by the scenario set forth in HQ 967058, as described above.

Another commenter specifically addressed the petitioner’s claim that “there is a pervasive commercial understanding that the preform core is produced as a rod.” The commenter takes the position that the petitioner’s “efforts to speak on behalf of the ‘industry’ in this case are … factually incorrect.” According to the commenter, “the Modified Vapor Chemical Deposition (MVCD) manufacturing process is generally accepted to be the most commonly used method of preform production.” The commenter describes the MVCD method as follows:

[T]he core in a preform manufactured in the MVCD process is not a pre-existing glass rod to which cladding is added. Rather, the cladding and core are made in a continuous process of layered soot deposition and sintering, with the different indices of refraction between the cladding and core achieved through modification of the gas mixture used to create the soot. The reason why the core remains visibly distinct from the cladding in a finished preform is not because the core rod was a pre-existing commercial entity, but because the core glass has a different refractive index. (Emphasis in original).

The MVCD method supports CBP’s conclusion that the petitioner has failed to establish as a matter of fact that a glass preform for an optical fiber is preceded by an extant core rod simply because of the visible differentiation between its layers. In light of the foregoing, the petitioner’s argument for the classification of glass preforms for optical fibers as a “class or kind of merchandise” that is “worked” beyond the scope of heading 7002, HTSUS, must fail. No evidence has been presented which would merit reconsideration of CBP’s factual conclusion in HQ 967058 that an extant core rod did not precede the existence of the cladding. The classification of the designated imported merchandise in heading 7002, HTSUS, was correct; and, the “class or kind” of merchandise consisting of glass preforms for optical fibers also falls to be classified in the same heading as explained below.

2. The Core Region Is Not a Commercial Product Even When It Is Created Before the Cladding Region of Glass Preforms of the Same Class or Kind as the Designated Imported Merchandise

The petitioner has cited to technical literature which suggests that an extant rod of glass may be an input in some manufacturing processes for glass preforms for optical fibers. However, the manipulation of a manufacturing input is considered a “working” for tariff purposes only if that manufacturing input stands on its own as a commercial product. Winter-Wolf, 966 F. Supp. at 1265 (observing that only a “commercial product” is subject to “working” for tariff classification purposes). In order to demonstrate that an article is a commercial product, there must be evidence of “commercial transactions or a market for them.” Drexel Chemical Co. v. United States, 27 Ct. Int’l Trade 804, 809 (2003); F. F. Zuniga a/c Refractarios Monterrey, S.A. v. United States, 16 Ct. Int’l Trade 459, 465-66 (1992). The evidence proffered by the petitioner does not establish that unclad core rods are “commercial products,” as explained by CBP’s analysis below for each of these examples.

HQ 967059 dated April 21, 2006. This ruling recognized that the petitioner refers to its precursor to a glass preform for an optical fiber as “cane.” However, this ruling sets forth the fact that”[a] protective layer of pure silica cladding material is added to protect the core from scratching and to reflect inward or contain within the core pulses of light passing through the glass core.” The core rod that was intended for importation in that ruling was covered in a glass cladding, and classified by CBP in heading 7002, HTSUS.

HQ 562754 dated August 11, 2003. This ruling recognized that 3M Co. refers to its glass preform for optical fiber as “seed.” This ruling clearly indicates that the “seed” is a glass rod that is clad before it is transported internationally. According to this ruling, “[t]he seed is a multi-layered glass rod. The rings, or layers, or glass that comprises the seed are melded together and light travels through the layers of glass, all of which have different refractive indexes.” The “seed” is clad prior to international shipment for further manufacture into an optical fiber.

HQ 960948 dated September 11, 1998. The petitioner cites CBP’s description of the manufacture of the glass preforms as a process whereby “a rod of the core soots is created in the first step of manufacture and that rod is then “worked” by the addition to it of cladding soots that are fused onto it.” (In HQ 967058, CBP revoked this ruling, and specifically stated that the fusing of cladding soots onto a target rod covered in core soots is not a “working” for purposes of heading 7002, HTSUS.) CBP’s acknowledgement of steps in the manufacturing process is not proof that any specific manufacturing inputs utilized are commercial products.

Application Note by Heraeus Tenvo. The note describes the process of jacketing an optical fiber preform that consists of both a core and a cladding. While the note recognizes the core as a component of a preform for an optical fiber, it does not indicate that the core is a glass rod that is traded internationally. Quite the contrary, the note indicates that the proposed “RIC preform technology” system is utilized on a preform for an optical fiber (and not merely its core).

Promotional Brochure for Shin-Etsu Quartz Products Co.: “Synthetic Quartz Glass for Fiber Optics SUPRASIL-F300.” The SUPRASIL-F300 is a specialized synthetic quartz glass. It is used in the manufacture of “substrate tubes,” which are tubes used in the manufacture of preforms for optical fibers in the MCVD and Plasma Chemical Vapor Deposition processes. The preform is manufactured inside of the substrate tube, where cladding material is deposited and then consolidated onto the core after the core is inserted into the tube. There is no evidence which suggests that the core rods used in this process are commercial products.

Paper Entitled “Next Generation Fiber Manufacturing for the Highest Performing Conventional Single Mode Fiber.” This paper describes a method of manufacturing preforms for optical fibers that is identified as “Rod-in-Cylinder and Overclad-during-Draw (RIC-ODD).” This method involves the insertion of a rod produced through VAD into a tube that will become its cladding and another tube that will act as an additional layer of overcladding. Both layers of cladding are applied to the core upon insertion into a furnace. There is no evidence which suggests that the highly specialized core rod is available as a commercial product. U.S. Patents 6,205,268 B1 and 6,928,841 B2. These patents refer to techniques that involve the etching of the core rod with a torch prior to the application of the cladding soots that will be consolidated onto the rod. The etching process “purifies the surface of the core prior to the application of the cladding soots. According to the second patent, an advantage of the method is that “a generic center core rod can be produced in large quantity, and inventoried for later use in making a wide variety of optical fiber preform designs.” There is no evidence that suggests that these generic center core rods are traded commercially.

All of these examples describe various methods of manufacturing glass preforms for optical fibers. Examples 1-4 are too general to be construed as suggesting that a separate core rod precedes the existence of a cladding layer. These examples are merely technical references to the core “region” of a rod, and, on page 6 of this letter, CBP has explained why such references are not proof of the existence of an independent core rod. Further, CBP notes that these examples do not include evidence that the petitioner itself even engages in the commercial trade of core rods for optical preforms which contain no cladding.

The methods described in examples 5, 6 and 7 utilize a pre-existing core rod that is unclad, but none of the examples suggest that these core rods are commercial products. The technical designation of an entity called a core rod is not proof that these entities are actually commercial products. CBP has examined each example provided by the petitioner. These three examples suggest that separate core rods are manufacturing inputs for preforms for optical fibers, but they do not demonstrate that these manufacturing inputs are bought and sold in the marketplace. These alleged “functional equivalents” are legally insignificant because there is no proof that they are the subject of commercial transactions. F.F. Zuniga, 16 Ct. Int’l Trade at 465-66 (noting that an alleged “functional equivalent” must be shown to be both substantially similar to the merchandise at issue and a commercial product). CBP notes that one commenter, who is also a global manufacturer of optical fibers, specifically stated that it is “unaware of the marketing by any supplier of preform core rods for sale apart from completed preforms.”

The dearth of evidence as to sales of any “functional equivalent” suggests that manufacturers of glass preforms for optical fibers who utilize pre-existing glass core rods create the core rods not for commercial usage, but for their own use in manufacturing the glass preforms for optical fibers. In this regard, CBP notes that, in Azteca Milling Co. v. United States, the court held that the intermediate products were not separate articles of commerce. The court concluded that the intermediate products were not readily marketable since they were only "materials in process, advancing toward the finished product." 890 F.2d 1150, 1160 (Fed. Cir. 1989). A review of technical literature contained in the petitioner’s examples 5, 6, and 7 suggests that the core rod is an intermediate material that is fragile, and its unclad surface is susceptible to contamination that would diminish its capacity to transmit light. The Court of International Trade (CIT) has recognized that articles which are merely a “transitional stage” of manufacture are not commercial products. F.F. Zuniga, 16 Ct. Int’l Trade at 466 (noting that the relative instability of a product is also proof that it is not suitable for commercial trade).

The petitioner has provided insufficient evidence to demonstrate that a separate glass core rod is anything more than a transitional stage in the manufacture of a glass preform for an optical fiber. The examples cited by the petitioner do not demonstrate that separate glass core rods are a distinct class or kind of merchandise that is commercially traded. In contrast, a glass preform for an optical fiber, which contains the core and cladding layers of glass, is a commercial product that is internationally bought and sold. See, e.g., Phototonics Buyer’s Guide, which identifies thirteen companies in the United States, Canada, England, France, Germany, and Taiwan that manufacture and/or supply optical fiber preforms in the marketplace. “Fiber Optic Preforms.” Phototonics Buyer’s Guide, available at (last visited March 21, 2008). It is noteworthy that this industry resource does not identify any manufacturers of separate core rods for commercial sale. The direct legal consequence is that glass preforms for optical fibers, even if manufactured using an extant core rod could not be considered “worked” within the meaning of heading 7002.

B. The Scope of Heading 7002 Covers Multilayered Glass Rods

The petitioner raises a number of arguments on the basis of certain HTSUS Notes, headings, and ENs to demonstrate that multilayered glass rods, like the designated imported merchandise, are categorically excluded from classification in heading 7002, HTSUS. This letter has already explained why, as a matter of fact, this visible differentiation is not conclusive evidence of “working.” CBP nonetheless addresses the reasons why, as a matter of law, the petitioner’s arguments on the scope of heading 7002 do not prima facie exclude glass preforms for optical fibers from eo nomine classification as unworked glass rods of heading 7002, HTSUS.

Note 2(a) to Chapter 70. The petitioner claims that CBP has impermissibly interpreted Note 2(a) to “impose a time constraint on when a glass rod may be ‘worked’ and thus excluded from heading 7002.” CBP’s position that “the ‘working’ of glass articles occurs after their creation” is not derived from text of Note 2(a), but from the broader principles of customs law. HQ 967058. See Winter-Wolff, 966 F. Supp. at 1265. Note 2(a) provides additional specificity for the delineation between an extant glass sheet and its working for headings 7003, 7004, and 7005. There is no comparable benchmark set forth in the Notes to Chapter 70 for glass rods of heading 7002. CBP’s references in previous rulings to Note 2(a) are merely exemplary—and do not create any specific time constraint on when a glass rod exists as a commercial product. Furthermore, CBP has previously clarified for the petitioner its position on the irrelevance of Note 2(a) to the definition of “worked” for purposes of heading 7002, HTSUS. HQ 968479, dated on March 21, 2002 (citing Blakley Corp. v. United States, 15 F. Supp. 2d 865 (Ct. Int’l Trade 1998) for the proposition that notes which apply to enumerated headings of a chapter of the HTSUS were intended by Congress to be inapplicable to all other unenumerated headings of that chapter to the HTSUS).

Note 2(c) to Chapter 70. The petitioner claims that Note 2(c) to Chapter 70, which defines the phrase “absorbent reflecting or non-reflecting layer,” supports its position that the addition of a layer of glass cladding is a working because it coats a core rod with additional material.  Headings 7003, 7004, and 7005 cover glass “in sheets or profiles, whether or not having an absorbent reflecting or non-reflecting layer, but not otherwise worked.” The addition of the aforementioned coating therefore constitutes a “working” for purposes of these headings. The implications drawn by the petitioner on the basis of Note 2(c) are impermissible for the same reason that, under Blakley, Note 2(a) is irrelevant to the scope of heading 7002. Furthermore, the obvious distinction between these coatings and the cladding layer of the glass preforms is that the coatings specified in Note 2(c) consist of metal or chemical compounds, but not glass. The plain language of Note 2(c) does not suggest that the application of a layer of glass is a working of an article of glass.

U.S. Note 2 to Chapter 72. The petitioner cites this U.S. Note, which provides, in relevant part, that “[f]or the purposes of this chapter [72], unless the context provides otherwise, the term ‘further worked’ refers to products subject to any of the following surface treatments: … cladding.” Contrary to the petitioner’s claim, this U.S. Note does not underscore the relevance of the ENs to Heading 7006, which are discussed below. Neither those ENs nor this U.S. Note support the petitioner’s claim that the presence of a glass layer identified as a “cladding” is a working for purposes of heading 7002. There is no legal relevance attributable to the reference to the term “cladding” in Chapter 72 for the purposes of the classification of a good of Chapter 70. The terms of this U.S. Note 2 to Chapter 72, beginning “[f]or purposes of this chapter,” specifically restrict the designation of cladding as a further working to goods of that chapter. Cladding, in the context of Chapter 72, is the molecular interpenetration of the surfaces of metals of different colors or natures, which includes a mechanical working of the article by various rolling processes. This cladding process, performed on a finished product of metal of Chapter 72, is not similar to the creation of the cladding layer of a glass preform for an optical fiber. The formation of a glass preform does not involve a finished product, nor is there any rolling or other mechanical process which “works” a core rod of glass to apply an additional layer of glass.

General ENs to Chapter 70. The petitioner claims that certain examples of glass manufacture, specifically cutting out and lampworking, result from the “working” of glass. While CBP does not doubt that the manufacturing of glass inherently involves its working, these ENs are of no specific relevance to heading 7002, HTSUS, which precludes working glass beyond the shape of a ball, rod or tube. The petitioner cites the example in the General ENs to Chapter 70 of lampworking as a means of working a glass rod or tubing to produce a glass ampoule. An ampoule is a small glass vial that is no longer in the shape of its precursor rod or tube. There is no basis for comparison with a glass preform, which is not worked beyond its shape as a glass rod.

ENs to Heading 7002. The petitioner notes that the ENs exclude glass tubes coated on the inside with fluorescent material from classification in heading 7002, HTSUS. The petitioner claims, by analogy, that this exclusion manifests intent to exclude pre-existing glass core rods coated with a layer of glass cladding from classification in heading 7002. CBP finds this analogy to be flawed because the ENs’ specificity to a coating of fluorescence (which is not a glass) does not stand for the broad proposition that visible differentiation between two layers of glass is prima facie proof of “working.” Furthermore, this exclusion is preceded by the statement that tubes with fluorescent material added to them in the mass are within the scope of heading 7002. This statement suggests that when an article is glass, the fact that other non-glass materials are added to it in the mass during manufacture does not render them “worked.” It follows that the addition of glass materials to other glass materials in the manufacture of a glass preform is not a “working” for purposes of heading 7002 of the HTSUS.

ENs to Heading 7006. The petitioner references one of the examples of “glass which has been otherwise worked” for purposes of heading 7006: “glass which has been surface worked after manufacture, for example, … enameled glass (i.e., glass decorated with enamel or vitrifiable colors)….” (emphasis added). The petitioner explains that “adding material capable of being converted into glass to the surface of existing glass is a surface working of that existing glass.” The understanding of surface working in the context of this EN is not relevant to heading 7002 for two reasons. First, the scope of heading 7006 is limited to “worked” glass sheets and profiles of headings 7003, 7004, and 7005. Those glass sheets are extant prior to their decoration with enamel or vitrifiable colors. In contrast, the glass preform described in HQ 967058 is not an extant glass rod when the cladding layer is formed. Second, vitrifiable materials are not “glass,” and are inherently different pigment materials from the actual glass sheets.

Third, the ENs do not identify the application of glass on glass as a surface working. A multilayered glass sheet is still an article that is not worked beyond its manufacture as a glass sheet. CBP rejects the petitioner’s argument for the categorical exclusion of multilayered glass rods from classification in heading 7002 on the basis of the cited HTSUS Notes, headings, or ENs. C. An Unworked Glass Rod of Heading 7002 Is Not Classifiable in Heading 7020

The petitioner claims to argue for classification in heading 7020, HTSUS, in the alternative, but the petition contains no arguments which support that position. Heading 7020 is a basket provision for “[o]ther articles of glass” that are not elsewhere classifiable in Chapter 70, HTSUS. CBP is of the position that the glass preforms for optical fibers are classified in heading 7002, for the reasons set forth above. Heading 7002 is an eo nomine provision that specifically covers unworked glass rods. By application of GRI 1, the glass preforms fall to be classified in this heading. Accordingly, the glass preforms are excluded from classification in heading 7020 because the terms of heading 7020 exclude articles elsewhere classifiable in Chapter 70.

D. An Unworked Glass Rod of Heading 7002 Is Not Classifiable in Heading 9001

GRI 1 provides that goods should be classified in accordance with the terms of the headings and any relevant section or chapter notes. Glass preforms for optical fibers are not classifiable in heading 9001 pursuant to GRI 1 because this heading, which covers “optical fibers,” does not include glass rods of any kind. There are no relevant section and chapter notes for heading 9001 that direct the classification of any glass rods or intermediate products in the manufacture of optical fibers in this heading, nor do the relevant ENs. Note 1(a) to Chapter 70, HTSUS, which excludes optical fibers of Chapter 90 from classification in Chapter 70, does not operate in this case because the glass preforms are not optical fibers. The petitioner’s argument that the glass preforms are classifiable in heading 9001 by application of GRI 2(a) is unpersuasive. Resort to GRI 2(a) is unnecessary when goods are classifiable in an eo nomine provision of the HTSUS by application of GRI 1. The glass preforms are classifiable in heading 7002, an eo nomine provision for unworked glass rods, pursuant to GRI 1. Accordingly, the petitioner’s arguments with respect to the essential character of the designated imported merchandise under GRI 2(a) are legally irrelevant.

Nonetheless, this letter addresses the reasons why the petitioner’s classification of the imported merchandise in heading 9001 by application of GRI 2(a) is otherwise erroneous. Classification by application of GRI 2(a) rests upon an essential character analysis of the article in its condition as imported. GRI 2(a) provides, in relevant part, that “[a]ny reference in a heading to an article shall be taken to include a reference to that article incomplete or unfinished, provided that, as entered, the incomplete or unfinished article has the essential character of the complete or finished article.” CBP has previously conducted an analysis of the essential character of glass preforms described in HQ 967058 pursuant to GRI 2(a). In HQ 960948, CBP determined that these relatively thick rods do not embody the very thin and flexible nature of an optical fiber, which constitute its essential character. The glass preform is 62 millimeters (almost 2 1/2 inches) in diameter and 1500 millimeters (almost 50 feet) in length. In contrast, the optical fiber will be subjected to an intricate drawing process to produce kilometers of hair-like optical fiber. The glass preform is 62,000 times larger in width than an optical fiber. With respect to length, the optical fiber is no less than 1073 times longer than the glass preform.

The petitioner dismisses CBP’s determination, and argues that the glass preforms have the essential character of an optical fiber because the glass preforms are for “a single predetermined use” in the subsequent manufacture of an optical fiber. It is undisputed that the high degree of purity of the glass and the presence of specific variation in the refractive indices of the layers of a glass rod dedicate its use to the manufacture of an optical fiber. However, dedicated use is not the legal basis for an essential character determination under the GRIs. The longstanding position of CBP is that the term "essential character" for purposes of GRI 2(a) means the attribute which strongly marks or serves to distinguish what an article is; that which is indispensable to the structure, core or condition of the article. See HQ 956538, dated November 29, 1994. The essential character for purposes of GRI 2(a) is determined on a case-by-case basis based on the nature of a given article. After importation, to produce an optical fiber, the glass preform is heated and drawn into a continuous "hair-thin" strand that may be many miles in length. The Fiber Optic Reference Guide, David R. Goff (1996), 15-16, explains that an optical fiber consists not only of the core and cladding, but also a buffer coating of plastic that “protects the fiber from damage and moisture.” According to this source, “[t]he light in a fiber-optic cable travels through the core (hallway) by constantly bouncing from the cladding (mirror-lined walls), a principle called total internal reflection. Because the cladding does not absorb any light from the core, the light wave can travel great distances.” This source also explains that the optical fibers are “arranged in bundles called optical cables and used to transmit light signals over long distances.”

In HQ 960948, CBP described the nature of an optical fiber, which is undoubtedly related to its function of transmitting information in the form of light through very thin flexible strands. CBP cited dictionary, government, and industry sources for their definitions of optical fibers, which provided in relevant part:

“…a very thin flexible glass or plastic strand along which large quantities of information can be transported in the form of light pulses.” (citing Random House Unabridged Dictionary (2d ed. 1993)).

“…[a] long thin strand of transparent glass, plastic, or other material usually consisting of a fiber optical core and a fiber optical cladding capable of conducting light along its axial length by internal reflection” (citing U.S. International Trade Commission (USITC) Publication 2851, February 1995, Industry & Trade Summary, Optical Fiber, Cable, and Bundles, B-2).

“…extremely thin strands of ultra-pure glass designed to transmit light from a transmitter to a receiver.” (citing Fiber Optic Reference Guide, 11).

All of these definitions begin their description of what constitutes an optical fiber with the physical quality of the optical fiber as a very thin strand. This form is the dominant characteristic of the physical appearance of an optical fiber. The thin nature of the strand is also integral to the flexibility of the fiber when performing its function of transmitting light. See HQ 960948, wherein CBP cited a publication by the petitioner, Just the Facts, A basic overview of fiber optics, Corning (1995), at 15, which identified “the recognized industry-standard bend diameter.” The industry standard identified in HQ 960948 provided for the looping of fiber with a bend diameter as small as two inches. Together, thinness and flexibility are physical characteristics that are integral to the essential character of an optical fiber.

CBP reached the correct result in HQ 960948 when it determined that a relatively thick and rigid multilayer glass rod does not have the essential character of the miles of thin, flexible fiber into which it will be drawn as a fiber. While both the unfinished and finished articles contain the critical optical properties for the transmission of light , the glass preform is not recognizable as a fiber and cannot transmit light over long distances. This relatively thick glass rod does not embody the essence (thinness and flexibility) of an unfinished fiber. Indeed, an optical fiber is both “optical” in its properties” and “fibrous” in its form. The very name of the product reveals a dual nature in its essential character. The petitioner’s argument that the presence of certain optical properties is dispositive of the essential character neglects the inherent significance of the form of this particular good, whose shape is integral to the transmission of light over long distances.

The petitioner disputes CBP’s finding that the glass preforms do not have the essential character of finished optical fibers of heading 9001, HTSUS, pursuant to GRI 2(a). The petitioner claims that CBP utilized an impermissible standard requiring that an article must have “the essential characteristics necessary for practical use.” The petitioner’s argument is that such a standard undermines the very existence of GRI 2(a) because, “[v]irtually by definition” no article unfinished or incomplete possesses [these characteristics].” CBP disagrees. The petitioner’s claim denies the reality that there is a direct relationship between the form and function of an article. Indeed, the form of an article is often designed to facilitate its actual function. The CIT has clearly recognized this interrelationship, indicating that an unfinished article which is devoid of the physical attributes or components essential to its function is not classifiable as a finished article by application of GRI 2(a). See, e.g., Sharp Microelectronics Technology, Inc. v. United States, 932 F. Supp 1499, 1505 (Ct. Int’l Trade 1996) (finding that the essential functional components, but not all functional components, of a finished automatic data processing display output unit must be presented together for classification of an unfinished unit as a finished one by application of GRI 2(a) because “it is the ability to process data that gives the essential character to articles under Heading 8471 [an eo nomine provision for automatic data processing machines and units thereof]).

The CIT has previously addressed the merits of the petitioner’s legal claim that dedication to a specific use is sufficient to establish essential character of an imported good for purposes of GRI 2(a). In Filmtec Corp v. United States, 293 F. Supp 2d 1364, 1369 (Ct. Int’l Trade 2003), the plaintiff argued that “because AWA No. 10 is dedicated to filtration, its “essential character” is fixed as being that of a straining cloth, and should thus be classified [as a finished article] under heading 5911, HTSUS” pursuant to GRI 2(a). Filmtec, 293 F. Supp 2d at 1369. The CIT recognized that AWA No. 10, “a nonwoven textile fabric sheet consisting of 100 [percent] polyester fibers[,]” was dedicated for use in the finished article at the time of importation, but it did not have “the essential character of ‘the complete or finished article’ – the ability to strain salt from water.” Id. The CIT clearly defined the essential character by virtue of the absence of certain physical attributes essential to the finished article’s function. Indeed, to strain salt, the AWA 10 required certain chemical layers that were not cast until after importation. Id. at 1366. Therefore, in its condition as imported, the CIT concluded that the AWA 10 was not classifiable as the finished article pursuant to GRI 2(a). Id. at 1369.

Similar to Filmtec, the petitioner’s argument that the glass preforms have the essential character of the optical fibers because the glass preforms are dedicated to use in the manufacture of an optical fiber is legally insufficient. The ability to transmit light over long distances is owing directly to the thinness and flexibility of the optical fiber, attributes which do not exist until after importation when the optical fiber is drawn from the glass preform. The statement in EN 90.01 that "optical fibres are usually presented on reels and may be several kilometers in length" supports the treatment of thinness and flexibility as essential characteristics of optical fibers. As CBP explained in HQ 960948, thinness and flexibility are necessary for the usages of optical fibers (see USITC Publication 2851, supra, at 1, wherein it is observed that "optical fiber systems now carry the bulk of long-distance telecommunications traffic in the United States").

For the aforementioned reasons, the petitioner has not provided any evidence that would merit reconsideration of the essential character determination in HQ 960948 pursuant to GRI 2(a). CBP reaches this conclusion notwithstanding the petitioner’s reliance on CBP’s application of the substantial transformation test to glass preforms in HQ 560660, dated April 9, 1999, and HQ 561774, dated January 29, 2001. In both cases, CBP found that a glass preform embodies the critical optical properties of the optical fiber, and that once those properties exist, there is no further substantial transformation of the glass preform for country of origin purposes. The substantial transformation test is distinct from a GRI 2(a) analysis, which may take into account more than the dedicated use of a good. Here, a glass preform possesses the optical properties of an optical fiber, but it does not in any way resemble a fibrous material. See supra pp. 16-17. That physical characteristic is indispensable to its function of transporting light over long distances; and, in its absence, a glass preform lacks the essential character of an optical fiber of heading 9001, HTSUS, for purposes of GRI 2(a).

CBP has also considered whether the criteria set forth in EN (II) to GRI 2(a) provide an independent basis for classification of the glass preforms as blanks for optical fibers. In HQ 967908, CBP determined that the GRI 2(a) discussion of “blanks” at EN (II) to GRI 2(a), "provides an independent, objective criteria for determining whether an incomplete or unfinished good is to be classified as if complete or finished." EN (II) to GRI 2(a) further defines a "blank" as "an article, not ready for direct use, having the approximate shape or outline of the finished article or part, and which can only be used, other than in exceptional cases, for completion into the finished article or part." The EN establishes two criteria for articles that are "blanks" for GRI 2(a) purposes: approximate shape or outline and sole use for completion into the finished article or part. The glass preforms must satisfy both of these criteria in order to qualify as “blanks” within the scope of the EN. The glass preforms do not have the approximate shape or outline of optical fibers. The petitioner has cited a USITC publication, which refers to the preform as a “magnified version of the fiber to be drawn from it.” On this basis, the petitioner concludes that “[i]f the shape were not approximate, the magnified version of the fiber would not be perceptible.” An extremely thin fiber is not, in fact, perceptible by looking at a glass rod that is 62,000 times its width and no less than 1073 times smaller in length. By order of magnitude, the discrepancy in size between the glass preform and optical fiber is not analogous to the parenthetical exception for bottle preforms of plastics in the first paragraph of EN (II) to GRI 2(a), as argued by the petitioner. The bottle preform is described as a plastic product “having tubular shape, with one closed end and one open end threaded to secure a screw-type closure, the portion below the end being intended to be expanded to a desired size and shape.” The bottle preform already embodies part of the final shape of the finished article, i.e., the threaded open end, and the scale of the entire preform does not vary substantially from the finished bottle. The glass preform which has not been worked beyond the basic shape of a rod is not, in any way, partially completed, or otherwise differentiated to reflect the fibrous form of the finished optical fiber.

The second paragraph of EN (II) to GRI 2(a) excludes products from the definition of a blank. For exemplary purposes, that same paragraph also explains that semi-manufactures which do not qualify as blanks generally include “bars, discs, tubes, etc.” Rods and tubes are both cylindrical bodies, and a rod is certainly ejusdem generis with these products. The bottle preform is clearly distinguishable because it was not merely a tube (owing to the presence of the threaded opening); its essential shape was partially that of the finished article and its size was on the same scale as the finished article. In contrast, the relatively thick glass preform is many orders of magnitude larger in diameter and length than the finished article and no aspect of its physical shape (beyond being cylindrical) reflects a finished optical fiber. Accordingly, the glass is not classifiable as blank for a finished optical fiber pursuant to GRI 2(a).

CBP now concludes its consideration of the petitioner’s arguments and other public comments concerning the classification of glass preforms for optical fibers. CBP has allocated significant time and resources in responding to the petitioner, not only in this letter, but also in numerous rulings and in consideration of the petitioner’s public comments submitted on multiple occasions pursuant to 19 U.S.C. § 1625(c)(1). Many of the arguments addressed in this letter have been presented to CBP by the petitioner over the past 10 years in its capacity as both a domestic and global manufacturer of this class or kind of merchandise.

This letter also officially notifies the petitioner that CBP correctly classified the glass preform described in HQ 967058, by application of GRI 1, under heading 7002, HTSUS, and specifically under subheading 7002.20.1000, as glass rods, unworked, of fused quartz or other fused silica. Pursuant to 19 U.S.C. § 1516, the petitioner may file notice of its desire to contest this decision not later than thirty days from the date of issuance of this letter. See 19 C.F.R. § 175.23.


Myles B. Harmon, Director
Commercial and Trade Facilitation Division