Elkem Metals Company — Hazardous Materials Safety Interpretation

Page 1- of Transportation U.S. Department 400 Seventh Street, S.W. Washington, D.C. Special Programs Research and 20590 Administration MAR - 2 1998 Ms. Sharon L. Thompson Manager-Traffic and Distribution Elkem Metals Company Post Office Box 72 Marietta, Ohio 45750-0072 Dear Ms. Thompson: This is in response to your letter dated October 30, 1997, requesting confirmation that your company's ferrosilicon is not subject to the Hazardous Materials Regulations (HMR; 49 CFR Parts 171-180). You state that your product was tested in accordance with the test method described in Appendix E to Part 173, and found it does not meet the definition of a Division 4.3 (Dangerous When Wet) material. As provided in § 173.22, it is the shipper's responsibility to properly classify a hazardous material. However, based on the test data you provided, we concur that your company's ferrosilicon with 50% ferrosilicon does not meet the definition of a Division 4.3 material. Therefore, if your company's material does not meet any other hazard class definition of Part 173 it is not subject to the HMR. I hope this satisfies your request. Sincerely, Delon Hilling Delmer F. Billings Chief, Standards Development Office of Hazardous Materials Standards#

Page 2• & Elkem Dale PAi! 700. 173,124 1322 October 30, 1997 Ms. J. Suzanne Hedgepeth Director Office of Hazardous Materials U.S. Department of Transportation +00 Seventh Street. S.W. Research & Special Programs Administration Washington. D.C. 20590 Dear Ms. Hedgepath: In accordance with the rules set forth in U.S. DOT 49 CFR 173, Appendix E, Elkem performed in Division 4.3. Based on the attached findings. we do not believe our product meets the criteria as it did the test as described on our 50% Ferrosilicon to determine if our product met the criteria for classification not spontaneously ignite nor was there any gas evolution observed. not meet the 4.3 definition. We would appreciate your review of the attached report and confirmation that our product does Sincerely, Star 1. Thompan Manager - Traffic and Distribution Sharon L. Thompson Enclosures ELKEM METALS COMPANY MARKETING SERVICES; P.O. BOX 72; MARIETTA, OHIO 45750-0072 Quality i Ош Telephone: (800) 848-9795 Fax: (614) 374-1043 PRIORITY#

Page 3i x Results of the tests performed on Elkem's regular 50 % FeSi (50 wt. % Si) using DOT procedures. (09/29/97) E. G. Allain, F. X. He, D. G. C. Robertson Center for Pyrometallurgy Department of Metallurgical Engineering University of Missouri-Rolla Summary was used to test Elkem's 50 % ferrosilicon material. The sample was sieved to determine the particle size distribution after handling and transportation. The percentage of particles of less than 500 micrometers was 0.23%. Therefore, according to the DOT instructions, the material did not need to be ground for testing. Results of the three tests recommended by the U.S. DOT show that no gas evolution or spontaneous ignition was observed when Elkem's 50 % ferrosilicon was put in contact with distilled water. Aims The aim of this report is to provide to Elkem the results obtained during the tests of Elkem's 50 % FeSi (50 wt. % Si) in applying rigorously the instructions indicated in the U.S. DOT 49 CFR 173, appendix E, entitled "Guidelines for the Classification and Packing Group Assignment of Class 4 Materials", division 4.3. This division gives the test conditions for the classification of "dangerous when wet" materials. 1. Material The Elkem's 50% Fesi material was sent to UMR in a drum. As received, it mainly contained pieces measuring from about one to ten centimeters. However, it also included a relatively low amount of powder. The total weight of this material was 11,362 grams.#

Page 42. Tests procedure and results 2. 1. Preliminary sieving test The U.S. DOT 49 CFR 173, appendix E, division 4.3 gives in section "a). Test method" the following specifications: "For a solid material, the package should be inspected for any particles < 500 um diameter. If that powder constitutes more than 1 percent (mass) of the total or if the material is friable, then the whole of the sample should be ground to a powder before testing to allow for a reduction in particle size during handling and transport, otherwise the material should be tested in its commercial state" Therefore, the totality of the as-received Elkem's 50 % FeSi material was sieved three times, during 10 minutes, using the following equipment: - A series of 8 sieves of the Tyler Standard Screen Scale (From 3 Mesh to 28 Mesh). The 500 um opening (32 Mesh Tyler equivalent) sieve was taken from a U S standard sieving series. Openings of these sieves are given in Table I. - A shaking apparatus model "COMBS gyratory sifting machine", manufactured by Greet Western Mfg. Co., Kansas. As indicated in Table II, the amount of particles ‹ 500 um constituted less than 0.23 % of the weight of the as received 50% FeSi material. It was also found that this material was not friable (at least to handling in the laboratory). Therefore, the further tests recommended by the DOT should be performed in its commercial state. Table I: Characteristics of sieves used to sift the Elkem's 50 % FeSi Meshes to the inch Opening (in) Opening (mm) 3 0.263 6.680 4 0.185 4.699 6 0.131 3.327 8 0.093 2.362 10 0.065 1.651 14 0.046 1.168 20 0.0328 0.083 28 0.0232 0.589 Tyler equivalent 32* 0.0197 0.500* Sieve taken from a US standard sieving series. :#

Page 5in it • • : Table II: Size distribution of the as received Elkem's 50 % FeSi # Test 1 # Test 2 #Test 3 Average Screen Size Weight (g) retained Wt %, Weight Wt%, Weight Wt%, Weight Wt %, (g) retained (g) retained (g) retained +3M 11318.60 99.634 11319.38 99.642 11319.39 99.644 11319.12 99.640 -3M+4M 6.25 0.055 4.90 0.043 5.470 0.048 5.540 0.049 4M+6M 0.67 0.006 1.25 0.011 0.660 0.860 0.008 SM+8M 1.400 0.012 1.08 0.010 1.010 0.010 -8M+10 M 1.19 0.010 1.38 0.012 1.470 0.012 -10M+ 14 M 1.33 0.012 1.28 0.011 1.230 0.011 -14M+ 20 M 1.53 0.013 1.45 0.013 1.470 0.013 -20M+ 28 M 1.85 0.016 1.92 0.017 1.840 0.016 0.016 -28M+ 32 M 1.25 0.011 1.24 0.011 1.290 0.011 1.260 0.011 -32 M* 26.12 0.230 26.13 0.230 25.990 0.229 26.080 0.230 Total 11360.19 100 11360.01 100 11359.82 100 11360.00 100 * <500 um#

Page 6?. 3. Tests to detect spontaneous ignition or gas evolution. 3. 1. Tests Procedure The U.S. DOT 49 CFR 173, appendix E, division 4.3 specifies four different tests to determine whether the reaction of a material with water leads to the development of a dangerous amount of gases which may be flammable or toxic. The first three tests are described in terms indicated below: Test (1): "A small quantity (approximately 2mm diameter) of the test material is placed in a trough of distilled water at 20 °C. It is noted whether any gas is evolved and if it spontaneously ignites." Test (2): "A small quantity of the test material (approximately 2mm diameter) is placed in the center of a filter paper which is floated flat on the surface of distilled water at 20 °C in a 100 mm diameter evaporating dish. The filter paper is to keep the material in one place, under which condition the likelihood of spontaneous ignition of any gas is greatest. It is noted whether any gas is evolved and if it spontaneously ignites." Test (3): "The test material is made into a pile approximately 2 cm high and 3 cm in diameter with an indentation in the top. A few drops of water are added to the hollow. It is noted whether any gas is evolved and if it spontaneously ignites." 3. 2. Results 3. 2. 1. Test (1) Three samples of the as-received Elkem's 50% FeSi material were submitted in triplicate to this test. They were put in a trough containing about 80 cm of distilled water, during 20 minutes. The sample characteristics are the following: Sample 1: A piece of about 2mm diameter taken from the + 3 Mesh fraction size. Sample 2: A piece of about 2mm diameter taken from the - 3 Mesh to + 32 Mesh fraction size. Sample 3: 0.5 gram of the powder from the - 32 Mesh fraction size. Table III summarizes the results obtained during test (1). It shows that no spontaneous ignition or gas evolution occurred in this test condition. Table III: Results of test (1) Sample/Test N° 1 N° 2 N° 3 Sample 1 N. S. I.*, N.G. E.** N. S. I., N.G. E. N. S. I., N.G. E. Sample 2 N. S. I., N.G. E. N. S. I., N.G. E. N. S. I., N.G. E. Sample 3 N. S. I., N.G. E. N. S. I., N.G. E. N. S. I., N.G. E. * N. S. I.: No Spontaneous Ignition, ** N.G. E.: No Gas Evolution#

Page 7... • • 3. 2. 2. Test (2) Three samples of the as-received Elkem's 50 % FeSi material were submitted in triplicate to this test. These samples have the same characteristics as those described above for test (1). They were put on a filter paper floating on distilled water contained in a 10-cm diameter evaporating dish, during 20 minutes. Table IV shows the results obtained during this test. It indicates that no spontaneous ignition or gas evolution occurred during the three tests. Table IV: Results of test (2) Sample/Test N° 1 N° 2 N° 3 Sample 1 N. S. I.*, N.G. E.** N. S. I., N.G. E. N. S. I., N.G. E. Sample 2 N. S. I., N.G. E. N. S. I., N.G. E. N. S. I., N.G. E. Sample 3 N. S. I., N.G. E. N. S. I., N.G. E. N. S. I., N.G. E. * N. S. I.: No Spontaneous Ignition, ** N.G. E.: No Gas Evolution 3. 2. 3. Test (3) Two samples of the as-received Elkem's 50 % FeSi material were submitted in triplicate to this test. The sample characteristics were the following: Sample 1: A piece of about 2 om high and 3 cm large taken from the + 3 Mesh fraction size of the Elkem's 50 % FeSi material. A hollow of about 3- mm depth was drilled at the top of this piece. Then, some drops of water were added. Sample 2: Some pieces of the Elkem's 50 % FeSi material were ground.to <80 um The resulting powder was packed into a small refractory crucible of 2 cm high and 2 cm. A hollow of about 3-mm depth was made at the top of this pile. Then, some drops of water were added. Table V groups the results obtained during Test (3). It shows that no spontaneous ignition or gas evolution occurred during the three tests. Table V: Results of Test (3) Sample/Test N° 1 N° 2 N° 3 Sample 1 N. S. I.* , N.G. E.** N. S. I., N.G. E. N. S. I., N.G. E. Sample 2 N. S. I., N.G. E. N. S. I., N.G. E. N. S. I., N.G. E. * N. S. I.: No Spontaneous Ignition, ** N.G. E.: No Gas Evolution#

Page 83. 2. 4. Test (4) Test (4): "Water is put into the dropping funnel and enough of the material up to a maximum weight of 25 g) to produce between 100 cm' and 250 cm' of gas is weighed and placed in a conical flask. The tap of the dropping funnel is opened to let the water into the conical flask and a stop - watch is started. The volume of gas evolved is measured by any suitable means. The time taken for all the gas to be evolved is noted and, where possible, intermediate readings are taken. The rate of evolution of gas is calculated over seven hours at 1-hour intervals. If the rate of evolution is erratic or is increasing after 7 hours, the measuring time should be extended to a maximum time of 5 days. The 5 day test may be stopped if the rate of evolution becomes steady or continually decreases and sufficient data has been established to assign a packing group to the material or to determine that the material should not be classified in Division 4.3. If the chemical identity of the gas is unknown, the gas should be tested for flammability and toxicity." The samples taken for this test weighed 25g of ground material. The experimental apparatus used to measure the volume of gas that could evolve from the 50 % FeSi consisted in a flask of 500 cm' connected to a U-tube having a 1/16-inch ID. This U-tube was filled with water colored by the addition of a small amount of red ink. From the results of previous tests the gas volume was expected to be very low or even equal to zero. So, the change of the liquid position in the U-tube, due to the variation of the pressure in the flask was expected to indicate us the gas volume evolved as function of the time. As specified by the DOT 49 CFR 173, the 25 grams 50 % FeSi Material were put into contact with water in the flask. No ignition or gas evolution were observed. As reported in table VI no variation of the water height was noticed in the U-tube, from the beginning of the experiment through 7 hours. Hence, as recommended by DOT specifications, the measurements were stopped. Table VI: Results of Test (4) Time (hours) 1 2 3 4 5 6 7 Ah* (mm) 0 0 0 Gas volume (mm?) 0 0 0 0 0 * Variation of the colored water height in the U-tube.#

Page 93.2.5. Chemical analysis A sample weighing about 200 grams was taken from the as received 11.4 kg Elkem's 50% FeSi. This sample was ground and sieved to a powder of less than 80 micrometers. Then, four samples of five grams each were taken and sent to the MINTEK Laboratories, Randburg, South Africa (See Appendix I). The four samples were analyzed by emission spectroscopy to measure their Fe, Si, Al, Ca and Mg contents. Results are given below in Table VII and in Appendix I. Table VII: Results of chemical analysis of Elkem's 50 % FeSi. Sample Si % Fe % A1 % Ca % Mg % 2% * 1 50.5 47.9 0.65 < 0.2 < 0.1 99.35 2 50.2 47.7 0.62 < 0.2 < 0.1 98.82 3 50.7 47.6 0.61 < 0.2 < 0.1 99.21 4 50.1 47.6 0.62 < 0.2 < 0.1 98.62 Average. 50.4 47.7 0.63 < 0.2 < 0.1 99.03 * Total assuming that Ca = 0.2% and Mg = 0.1% 4. Conclusion Elkem's 50% FeSi material was tested following the U.S. DOT method procedure, specified in the U.S. DOT 49 CFR 173, appendix E. Test results lead to the following conclusions: The percentage of particles of less than 500 micrometers in the as-received material was 0.23%, No gas evolution or spontaneous ignition were observed when small quantity of Elkem 50% FeSi is placed in a trough of distilled water at 20 °C, No gas evolution or spontaneous ignition were observed when a small quantity of Elkem 50% Fesi is placed in the center of a filter paper which is floated flat on the surface of distilled water at 20 °C, No gas evolution or spontaneous ignition were observed when few drops of water were added at 20 °C into a hollow made in the top of a piece of Elkem 50% FeSi, No gas evolution could be measured or even detected when water was put in contact with a weight of 25 g of Elkem 50 % FeSi in a conical flask at 20 °C. The chemical analysis of the Elkem 50% FeSi give an average content of 50.4% Si, 47.7% Fe, 0.63 %Al, less than 0.2% Ca and less than 0.1% of Mg.#

Page 10APPENDIX I#

Page 112509-94- 08:20 792 6650 MINTEK ASD ->0915733416934 Page 1 of → Page 01 TELEFAX MESSAGE Analytical Science Division MINTEK Facsimila (011)-792-6650 (Local) Tolophone (011)-709-4111 (2711)-782-6650 (Int') Faksimileo (011)-723-2413 (Plans) (2711)-792-6650 (Int') Private Bag X3015, Randburg, Telefaon (011)-709-4111 2125 South Africa Privatsak X3015, Randburg, 2125 Suid-Afrike CONFIDENTIALITY WARNING The information in this facsimile message is privileged and confidential information intended solely for the use of the addressed person or entities. If you are not such persons or antitles, you are hereby notified that any distribution, dissemination or reproduction of this facsimile message, including the taking of any action in response to the contents thereof, is strictly prohibited. If you have received this message in error, please call us Immediately at the above number to arrange for the documents) to be returned at our cost. Dnte: 97/09/25 09l .... To: UNIVERSiTY OF MISSORI- ROLLA Fax No: (573) 34) 6934 :: For attention: DAVID ROBERTSON Please find attached results for the ferrosilicon sambles submitted or analysis. Plase acknowledge receift of this fax. • .... Kind regards: •.... Sheena Maharaj#

Page 12i:rii... .-..::. ... ---- ... "ri::. ....... . •• tintiniv Sistaal Nr. ictl No. Project No. Projek Nr. Soction EMISSION SPECTROSOOPY 2K 739 [13189701 Seksic EMISSIE- SPEKTROSKOPIE Rep MINTEK Ver ATTI DAVID ROBERTSON Spocialists in mineral and metallurgicad tocanology Tochnique/Plate No. SE002 Tegniek/Plaat Nr. Results reported in Resultate gerapporicer in % Date starled Datum begin 97/09/01 Date reported Datum gerapporteer 97/09/22 11g Ae Ca Te 1 <0.1 0.65 1 D. 2 47.9 2 ->0915733416934 0,62 3 0.62 47.7 0.59 47.6 4 V 0.62 V 47.6 5 6 MINTEK ASD 7 8 9 10 sample/Verw. monster A167 50.1 value/Aanv. waarde 50.05 B ample/Verw. monster wAany. waarde Analyst 08:21 Analis Time No of determinations 1. s, tu arsland. Tyd Getal bepalings 20 Checked by Nagesien deur .. 1. .....Hrs/uur Total time: Hrs Approved by 25-09-97 2 Pobuhozr z ....Hrs/uur Totale Tyd: uur Goedgekeur deu#

Page 13.... * Workshet no. Werkstaat nr. LAK 739 AITI DAVID ROBERTSON MINTEK Group- CHEMISTRY Results reported in Resultate gerapporteer in %½ Project Projok 1131897011 Datur begin 97/9/23 Date received Date reported Datur gerapporteer 97/9/2 Monster Sample Si 1 49,8 150,3 2 50,2 50,7 4 50,1 5 7 8 MINTEK ASD 10 Technique Tegniek Lab metode L 64 Lab, method. Analyst Analis Tyd h Time Verwysingsmonster Ref. sample Bestanddeel Constituent Resultate Results 1..A. I VAN LENDER.. 1 .... A167 S: 48,07 792 6650 2 • ... 3.... 3... 4.... 4... Total time Totale tyd No. of determinations 25-09-97 Aantal bepalngs Goedgekeur Approved#