OT:RR:CTF:EPDR H333263 IPW
Center Director Machinery
109 Shiloh Dr., Ste. 300
Laredo, TX 78045 Attn.: Derrick M. Logan, Import Specialist
RE: Application for further review of Protest No. 4601-22-132081; Golden Spectrum LLC.; scope of antidumping and countervailing order; certain quartz surface products
Dear Center Director:
The above-referenced Protest was forwarded to this office for further review and was received on September 27, 2022. We have considered the points raised by your office and the Protestant. Our decision follows.
FACTS:
From February 24, 2020, through October 30, 2020, Golden Spectrum LLC (“Golden Spectrum” or “Protestant”) entered 26 shipments of slabs from the People’s Republic of China (“China”). Pursuant to instructions from the Department of Commerce (“Commerce”), Customs and Border Protection (“CBP”) assessed antidumping and countervailing duties (“AD/CVD”) on the slabs and liquidated the entries on April 1, 2022. Golden Spectrum now protests CBP’s assessment of antidumping and countervailing duties.
CBP determined the goods to be subject to AD/CVD orders on certain quartz surface products from China (Case No. A-570-504 and C-570-085) (“Orders”). Certain Quartz Surface Products From the People's Republic of China: Antidumping and Countervailing Duty Orders, 84 FR 33053 (July 11, 2019). The scope of the orders covers certain quartz surface products. Quartz surface products “consist of slabs and other surfaces created from a mixture of materials that includes predominately silica (e.g., quartz, quartz powder, cristobalite) as well as a resin binder (e.g., an unsaturated polyester).” Id. Specifically excluded from the scope of the orders are crushed glass surface products. To be excluded from the orders as crushed glass surface products, the merchandise must meet each of the following criteria:
(1) The crushed glass content is greater than any other single material, by actual weight; (2) there are pieces of crushed glass visible across the surface of the product; (3) at least some of the individual pieces of crushed glass that are visible across the surface are larger than one centimeter wide as measured at their widest cross-section (glass pieces); and (4) the distance between any single glass piece and the closest separate glass piece does not exceed three inches.
Id.
To verify if the merchandise fell into the exception, CBP’s Laboratories and Scientific Services Division (“LSSD”) conducted a series of tests on a sample of the entered merchandise. The sample was a slab measuring 52 inches (in.) by 4 in. by .75, with item number QBP007. LSSD broke up the sample into smaller pieces and tested the sample as follows:
LSSD conducted an ignition test for plastic binders, whereby a piece of the sample was heated over a Bunsen burner, which caused it to burn leading to the conclusion that the sample contained organic binders.
LSSD performed an ash test on three 5-gram pieces, whereby the pieces were heated in a crucible at 595 degrees Celsius for one hour. Between the three pieces, the average loss by weight on ignition was 10.92%, leading to the conclusion that the sample consisted of 10.92% organic material, which another test concluded to be plastic.
One gram of the ashed sample was mixed with one gram of Corundum in a ball chamber for one minute and loaded into a sample holder three times to be analyzed for its phase composition through X-ray diffraction. The composition of the ashed sample was calculated to be 62.83% quartz, 36.27% amorphous Silicon Dioxide (“SiO2”) (“crushed glass”) and .9% Titanium Dioxide (“TiO2”). Factoring in that 10.92% of the sample was plastic, the overall composition of the sample was found to be 55.97% quartz, 32.31% amorphous SiO2, 10.92% plastic, and .8% TiO2.
Other remnants of the ashed sample were ground in a ball chamber for 10 minutes and then pressed into a pellet to be analyzed under an X-ray fluorescence spectrometer. The purpose of the test was to check the X-ray diffraction test by also determining the sample’s composition. The sample was found to be predominantly composed of silicon with lesser amounts of calcium, magnesium, aluminum, iron, titanium, and sulfur.
A crushed piece of the sample, along with some of the ashed sample that had been ground, underwent a Fourier Transform Infrared Spectroscopy (FTIR) analysis to identify organic, polymeric, and inorganic materials. Analysis of the crushed piece indicated presence of polydiallyl phthalate (plastic), and the analysis of the ashed sample indicated presence of silica.
Lastly, the sample was cut into smaller sections for examination under a microscope. A steel ruler was used to measure the diameters of the glass pieces. Two pieces were found to exceed 1 centimeter.
Because the Orders exclude crushed glass surface products when the crushed glass content is greater than any other single material, by actual weight, and the tests determined that quartz was greater than the crushed glass, at 55.97%, LSSD found the sample to be within the scope of the orders. On September 27, 2022, Golden Spectrum protested CBP’s determination of AD/CVD liability on the grounds that LSSD’s determination is erroneous and the merchandise is exempt from the orders as crushed glass surface products
According to Golden Spectrum in its protest, LSSD made the following errors in its determination:
Not Incorporating the Large Glass Chips – The sample obtained by CBP to run the analysis was a 1-gram sieved sample. The size of the sample obtained means it is impossible for CBP to consider the glass chips when measuring the glass content as a whole. This significantly understates the percentage of glass as the large chip contains no quartz at all.
Sample Size – As stated above, the sample obtained to run the analysis was comprised of 1 gram of a sieved sample. However, CBP’s report / methodology was silent on how the 1 gram was obtained. CBP did not obtain multiple 1-gram samples and test for repeatability. Due to variations in particular distribution, it is possible and very likely that any given 1-gram sample has a different quartz and glass percentage content even if no large glass pieces are present in the samples.
Incorrect Interpretation of the Scope – CBP’s reports determined that the sample contains 55.97% quartz, 32.31% amorphous SiO2, 10.92% plastic, and 0.9% TiO2. Furthermore, CBP’s reports found that quartz is the primary ingredient. CBP obtained an incorrect ultimate result because it ignored all of the components of the glass that are not amorphous SiO2. Golden Spectrum’s manufacturer uses recycled float glass or soda-lime glass, which is the most prevalent type of glass manufactured. A typical composition of soda-lime glass is 70% SiO2 (in the form of amorphous silica), 15% Sodium Dioxide (“Na2O”), 9% Calcium Oxide (“CaO”), and smaller amounts of other oxides. Assuming amorphous SiO2 is 70% of glass, the addition of the other components would change the results to 49% quartz, 41% glass, 10% plastic and 1% TiO2. This in combination with the below points greatly throws CBP’s methodology into question.
Incorrect Presumption of 0% Na20 Level – CBP’s laboratory assumed that there is no Na2O in the glass. However, it is well known in the industry that float glass contains Na20. CBP’s laboratory report stated that the instrument used by CBP did not measure any sodium in the samples, and assumed a 0% Na20 content in the mixture. However, this would lower the amount of glass detected in CBP’s calculations by 15% assuming a typical composition of soda-lime glass.
Normalization of Results – The results describe how the values were normalized in order for the percentages of SiO2 to be converted into quartz and glass contents. However, these calculations are not shown and in their expert analysis, Golden Spectrum could not recreate the results shown given the data provided. Accordingly, Golden Spectrum doubts the calculations used by CBP in order to reach its conclusion.
Margin of Error / Verification of Equipment / Appropriateness and Accuracy of Methodology – CBP’s lab ran a quantitative verification of a 50%/50% mixture of quartz and glass beads in order to validate their equipment and methodology. However, this was done two months prior to the date when the samples were tested. Golden Spectrum has concerns about whether and how the equipment was validated during the time the measurements were taken. Additionally, during the validation, the listed margin of error was 9.8%. This may significantly impact the results of the experiment. Golden Spectrum has concerns regarding the validity of the quantification of X-ray diffraction (“XRD”) methods to determine crystalline vs. amorphous silica. This is because no calculations pertaining to this are presented in the report.
Revisions – There were three revisions to the lab report summary which concern Golden Spectrum regarding the reliability of the lab report. Pages 1-5 are dated 8/19/2021 (with additional revisions made on 8/23/21), pages 6-10 are dated 4/9/2021, and pages 11-15 are dated 4/9/2021 – marked as the original. However, in the 8/19/2021 version, there are multiple changes including:
Changes in verbiage from quartz
Addition of a quantitative verification step of a 50/50 mix of quartz/soda-lime glass mixture performed on 7/2/2021, indicating that CBP ran the original testing in April 2021 without proper validation of its methodology.
Addition of a normalization of oxides step to the procedure, suggesting that CBP disregarded any component of glass other than amorphous SiO2 in their original analysis and later attempted to correct for this by determining the amount of oxides such as CaO. However, in doing so, CBP never measured for Na2O. In addition, their conclusion of 55.97% quartz, 32.31% amorphous SiO2 was not revised, meaning that even though they acknowledged the issue of not measuring the amount of other glass components, they did not change their conclusion.
Additionally, Golden Spectrum sent two samples to be tested by Clemson University’s Bishop Materials Laboratory. One sample was a slab with item number QM2101L. The other was a slab with item number QB1102, which, according to the protestant, is a part of the same glass stone series tested by LSSD but differs only in colorant. For the QM2101L sample, the private lab found that the glass content was 47.98% and the quartz was 39.89%. For the QB1102 sample, the private lab found the glass content to be 53.18% and the quartz was 35.10%. No evidence has shown that the samples were from the same entry as sampled by LSSD.
ISSUE:
Whether CBP properly liquidated the entries with the assessment of antidumping duties.
LAW AND ANALYSIS:
As an initial matter, we find that this protest meets the criteria for further review. Pursuant to 19 C.F.R. § 174.24(b), this protest involves questions of law and fact which have not previously been ruled upon. We also find that, pursuant to 19 U.S.C. § 1514(c)(3)(A), this protest was timely filed on September 27, 2022, within 180 days after the liquidation date of April 1, 2022.
Antidumping and countervailing duties properly assessed by CBP are generally not protestable because CBP’s role in liquidating entries of merchandise subject to an AD/CVD order is “merely ministerial.” See Mitsubishi Electronics America, Inc. v. United States, 44 F.3d 973, 977 (Fed. Cir. 1994). However, the factual findings made by CBP to determine whether merchandise is subject to an AD/CVD order, and the amount of duty owed, fall within CBP’s protestable ministerial role. See Xerox Corp. v. United States, 289 F.3d 792, 794 (Fed. Cir. 2002); HQ H192395 (January 6, 2014); HQ H189455 (February 19, 2014). CBP’s factual findings receive a presumption of correctness pursuant to 28 U.S.C. § 2639(a)(1). Consequently, a protestant bears the “burden to prove by a preponderance of the evidence” that CBP’s findings were incorrect. Ford Motor Company v. United States, 157 F.3d 849, 855 (Fed. Cir. 1998); see also Aluminum Co. of America v. United States, 477 F.2d 1396, 1398 (C.C.P.A. 1973) (“the methods of weighing, measuring, and testing merchandise used by customs officers and the results obtained are presumed to be correct”). The challenging party can rebut this presumption “by showing that such methods or results are erroneous” or if they submit “evidence of analysis they applied to the merchandise which gave a result different from that claimed by the Government.” Aluminum Co., 477 F.2d at 1398-99. In both situations, whether there is a showing of error or of contrary evidence, a prima facie case is established that destroys the presumption of correctness and the Government bears the burden of going forward with the evidence. See id.
The protestant argues that CBP’s findings are incorrect in two ways. First by showing that the methods documented in LSSD’s lab report were faulty, and second by submitting a private lab’s analysis that gave a different result. Alternatively, Golden Spectrum compels CBP to apply a scope ruling received from Commerce to the subject entries. We address Golden Spectrum’s arguments in order.
Golden Spectrum argues that LSSD’s use of a 1-gram sample made it impossible to consider large glass chips, which could understate the percentage of glass. While 1-gram of the sample was used to determine the composition of the whole, this 1-gram was sieved after 5-grams were ashed and ground. According to LSSD, glass chips were not excluded, nor did any glass chip exceed 1-gram. As a result, the 1-gram sieved sample should be representative of the whole.
Golden Spectrum argues that LSSD’s report does not explain how the 1-gram was obtained, nor did CBP obtain multiple 1-gram samples to test for repeatability. As stated earlier, the 1-gram was obtained through ashing and grinding a 5-gram piece of the sample. According to LSSD, the correct method to determine if a sample of a homogeneous product is statistically representative of the whole is based on testing of sufficient particles, not a sufficient percentage of the whole sample. While the 5-gram piece was approximately .08% of the total sample, it contained enough particles to test a homogenous product. The sample had a homogeneous appearance and LSSD has been given no reason to believe that the manufacturing process produces a non-homogeneous product. Even if the sample was not entirely homogeneous, the deviation of the composition of the 1-gram sieved sample from the 5-gram piece is unlikely to be large.
Golden Spectrum argues that LSSD ignored all components of the crushed glass that are not SiO2. In the protestant’s view, because their glass contains more than just SiO2, LSSD could not appropriately calculate the crushed glass content. LSSD states that this is a misinterpretation of the report. The calculation for the amorphous SiO2 includes the materials that are not crystalline, which removes the quartz and the TiO2. Thus, the calculation for the amorphous SiO2 does include more than just the SiO2. As the composition of the amorphous SiO2 is irrelevant to the calculation, it is better presented as one figure.
Relatedly, Golden Spectrum argues that LSSD assumed there was no Na2O in the glass, which would affect the calculation since the type of glass used in the merchandise often contains Na2O. As stated directly above, LSSD calculated the glass to include all non-crystalline materials, which would include the Na2O.
Golden Spectrum also questioned how LSSD normalized the values when calculating the sample’s composition with X-ray fluorescence (“XRF”) because the value is missing from the report. The value used is approximated. According to LSSD, XRF is not the appropriate method to reliably calculate the composition as it does not measure sodium. For that reason, LSSD did not use XRF to reach a definitive conclusion on the composition, but rather to check the results of the XRD analysis. The use of an approximated value and its non-inclusion in the report is immaterial as it is only used to check against the XRD analysis.
Golden Spectrum notes that LSSD ran a quantitative verification of a 50% quartz / 50% glass beads to validate their equipment and methodology two-months prior to the sample testing. Golden Spectrum has concerns that the equipment was not validated during the time the measurements of the sample were taken. They also note that the margin for error was listed at 9.8%, which could impact the results of the experiment. In response, LSSD states that the 50/50 verification was run after the sample was analyzed. They also validated the methodology with a 77% quartz / 23% amorphous silica sample and a 50% quartz / 50% amorphous silica sample. The XRD instrument is routinely verified with a silicon standard and was verified to be working properly at the time of use. The “% error 9.80” stated in the report represents a comparison between experimental intensity ratios and a reference intensity ratio—it does not represent the accuracy of the methodology. The methodology has since been fully verified and the validation study indicates that a single measurement may vary by 5% from the true value.
Golden Spectrum’s final challenge to LSSD’s methods regard the revisions made on the lab report. Specifically, they state their concern regarding the replacement of the word “quartz” to “silica” but do not articulate why this is problematic. They also note the addition of a quantitative verification step that they believe indicates there was no proper validation of its methodology around the time of the testing, and the addition of a normalization of oxides that they believe suggests CBP failed to consider any component other than SiO2 in the glass. These concerns have been addressed above.
None of Golden Spectrum’s challenges to LSSD’s methodology have shown “by a preponderance of the evidence” an error that destroys CBP’s presumption of correctness.
Turning to Golden Spectrum’s second argument, they claim the samples sent to a private lab show a different result from LSSD’s and therefore should rebut the government’s presumption of correctness. When presenting evidence of analysis that proffers different results, it is necessary to demonstrate that the tested sample is 1) removed from the actual shipment under investigation, and 2) comparable to the sample tested by the Government. HQ H279772 (February 22, 2017); see also HQ H114760 (February 14, 2011). In HQ H279772, the protestant similarly attempted to refute LSSD’s results by providing private lab test results. The protestant produced no evidence that its privately tested candles were from the same shipment that LSSD sourced their sample, nor did the samples share item numbers. As a result, we held that the candles analyzed by the private lab was neither from the same shipment nor comparable to the sample analyzed by LSSD and therefore could not rebut the government’s presumption of correctness. Likewise, here, Golden Spectrum produced no evidence that the sample slab tested by the private lab was sourced from the same shipment as the sample tested by LSSD. Further, the item numbers that the private lab tested, QM2101L and QB1102, do not match with sample tested by LSSD, item number QBP007. Golden Spectrum claims that QB1102 is comparable to QBP007 because it only differs by color, but even if considered comparable, the sample would still fail to meet the two-part requirement for private lab results to rebut the government’s presumption.
Lastly, the protestant compels CBP to apply a scope ruling by Commerce. As stated earlier, Commerce determines whether a particular kind of merchandise is covered by existing AD/CVD orders. Xerox, 289 F.3d at 794. Commerce issues scope rulings to determine whether specific merchandise is included within the scope of AD/CVD orders. 19 C.F.R. § 351.225. While Commerce has the sole authority to determine whether specific merchandise is subject to AD/CVD orders, CBP makes the “factual findings to ascertain what the merchandise is and whether it is described in an order.” Sunpreme, Inc. v. United States, 190 F. Supp. 3d 1185, 1195 (Ct. Int’l Trade 2016) (citing Xerox, 289 F.3d at 794). Accordingly, “where CBP can conclude that a product falls within the words of the order, both the affirmative scope language and any exclusions, CBP properly requires an importer to enter its goods as subject to an order.” Id. at 1202.
After the subject entries were liquidated, on September 27, 2022, Golden Spectrum requested Commerce to issue a scope ruling on their slab products. In their request, Golden Spectrum argued that their merchandise was excluded from the scope of the orders as crushed glass products by including, among other things, the private lab’s test results that showed the samples had a greater crushed glass content than any other single material. As a result, Commerce found that the specific merchandise was outside the scope of the orders by falling into the clear language of the order’s exception. See Certain Quartz Surface Products from the People’s Republic of China: Scope Ruling on Golden Spectrum LLC, June 29, 2023. While Commerce is within their sole authority to interpret the scope of the orders, this ruling does not affect the factual determination made by CBP in this case. As stated in Sunpreme, CBP is required to make factual findings pertaining to a product at the time of entry to ascertain that the product falls within the order. Our fact finding revealed that the sample from the subject entries had less crushed glass content than quartz, and thus was within scope of the orders. While Commerce’s scope ruling recommends that slabs such as the one in Golden Spectrum’s request fall outside the scope of the orders, the evidence here shows that the slab tested by LSSD is materially different from the ones in Commerce’s scope ruling. The case then rests on the protestant’s ability to show by a preponderance of the evidence that LSSD’s determination was incorrect, which they did not. Therefore, CBP was in its authority to liquidate the subject entries with AD/CVD duties.
HOLDING:
CBP properly liquidated the entries with the assessment of antidumping duties. Therefore, the protest is DENIED in full.
You are instructed to notify the Protestant of this decision no later than 60 days from the date of this decision. Any reliquidation of the entry or entries in accordance with the decision must be accomplished prior to this notification. Sixty days from the date of the decision, the Office of Trade, Regulations and Rulings will make the decision available to CBP personnel and the public on the Customs Rulings Online Search System (“CROSS”) at https://rulings.cbp.gov/, or other methods of public distribution.
Sincerely,
For Yuliya A. Gulis, Director
Commercial & Trade Facilitation Division