UNITED STATES OF AMERICA

OCCUPATIONAL SAFETY AND HEALTH REVIEW COMMISSION

April 27, 1983

DECISION

Before: ROWLAND, Chairman; CLEARY and COTTINE, Commissioners.

BY THE COMMISSION:

  As a result of an inspection of the Logans Ferry Works of Aluminum Company of America (‘Alcoa’), the Secretary of Labor (‘the Secretary’) issued two citations to Alcoa for violations of the Occupational Safety and Health Act of 1970, 29 U.S.C. §§ 651–678 (‘the Act’). One of the citations contained an item alleging a violation of section 5(a)(1) of the Act, 29 U.S.C. § 654(a)(1),1 in that:

Precautions were not taken to avoid the leakage and accumulation of water on the floor near the powder dust loading area where it could come into contact with aluminum powder in the Number 4 Ball Mill. Reference Paragraph 7–1.1, Chapter 7, N.F.P.A. [National Fire Protection Association] Standard Number 651–1974.

  The Secretary proposed a penalty of $420. Administrative Law Judge Henry K. Osterman affirmed that citation item and assessed a penalty of $420. Pursuant to section 12(j) of the Act, 29 U.S.C. § 661(i), the judge’s decision is before the Commission. The only issue on review is whether the judge’s ruling on that citation item was in error. We reverse and vacate this item of the citation.

I

  The ball mill building, the location of the alleged violation, was approximately 65 feet long, 25 feet wide, and 19 feet high and contained approximately 30,875 cubic feet of air. The inside of the building consisted of a large room that was completely open except for where the ball mills were located. In the ball mills, aluminum powder was ground into a slurry as a step in the process of manufacturing paint pigments. Aluminum powder was loaded into a chute in order to ‘charge’ the ball mill. The building had three overhead exhaust fans in operation that together were able to remove 6,000 to 9,000 cubic feet of air per minute. The portion of the room where the cited condition occurred was between two ball mills and measured 18 feet long and 25 feet wide and had one of the exhaust fans overhead.

  During his inspection of the ball mill room, Compliance Officer John T. Seifert observed that on the floor there were two puddles of water, measuring a few feet in diameter and amounting to approximately one or two gallons, located about 7 or 8 feet from ‘a couple shovelfuls’ of aluminum powder. He discovered that the puddles were the result of water leaking from the improper seal around the door to the Number 4 Ball Mill and from a crack near that door. The floor was sloped so that the aluminum powder, which was at the loading or ‘charging’ area, was approximately 2 inches higher than the water puddles, which were in turn higher than the doors. The leaks that Seifert observed had been there at least three years.

  It was undisputed that at the time of the inspection the water did not come into contact with the aluminum powder. It was further undisputed that, if such contact did occur, heat and hydrogen gas would be released. The parties also agreed that if the hydrogen gas reached the level of at least 4% of the volume of air in the ball mill room, the hydrogen gas would ignite and explode.

II

A

  According to Seifert,2 the cited condition presented two recognized hazards: the possibility that the hydrogen gas would ignite or explode, and the possibility that the heat produced by the reaction of water with aluminum powder would ignite the aluminum powder itself. He testified that aluminum powder is ‘relatively unstable’ and ‘readily ignitable.’ On cross-examination, Seifert admitted that he did not know approximately how much heat would be released by that chemical reaction. The Secretary introduced several exhibits into evidence which, in his view, supported Seifert’s position. Exhibit C–1 is a data sheet on aluminum powder issued by Alcoa. It states that, when exposed to water, the aluminum powder ‘[g]enerated hydrogen slowly. Care required.’ Another part of the sheet that Seifert considered to be applicable stated that, when aluminum powder is exposed to ‘[s]trong oxidizers (such as liquid oxygen or potassium permanganate),’ the result would be a ‘[v]iolent reaction with much heat generation.’ Exhibit C–2 consists of several pages of National Fire Protection Association (‘NFPA’) standards on ‘Aluminum or Magnesium Powder’ including section 7–1.1, which was referred to in the citation item at issue. That section states: ‘The leakage of water in or into any building where it can come into contact with any light metal powder shall be prevented to avoid consequent spontaneous heating and ignition therefrom.’ Exhibit C–3, a brochure entitled ‘Recommendations for storage and handling of aluminum pigments and powder,’ was prepared by the Aluminum Association. The only portions of that exhibit noted by the Secretary at the hearing regarding the item at issue read: (1) ‘General . . . Avoid all water in storage rooms. Leaks in steam lines, radiators or roofs should be promptly repaired;’ (2) ‘Aluminum powder storage . . . Avoid all water in storage as aluminum powder can evolve hydrogen gas and sufficient heat to cause ignition if exposed to water resulting in drum rupture and subsequent fire.’

  Seifert testified that the condition could be abated by fixing the leaks, cleaning up the water puddles, and removing the aluminum powder. Unrebutted testimony established that Alcoa had repeatedly tried to stop the leaking at the cited location and had instructed its employees to clean up the water and clear away the aluminum powder. Harry L. Cribbs, an Alcoa employee who had worked at the cited location and accompanied Seifert during his inspection, testified that Walter D. Helmick, Jr., Alcoa’s Safety Director for the Logans Ferry Works, and Cribbs’ immediate supervisors had always told employees that water and aluminum powder should not come in contact with each other because ‘it would blow up.’

B

  Safety Director Helmick testified that the information on ‘strong oxidizers’ in Exhibit C–1 was designed to warn Alcoa’s customers who were not familiar with the chemical reactions involved that they should not allow aluminum powder to contact strong oxidizers. He stated that water is not a strong oxidizer and that this section of the data sheet does not refer to water. Helmick testified that the main reason that he had instructed employees not to mix water and aluminum powder was to prevent such a mixture from occurring in a confined area, such as a drum, which could explode as a result of the chemical reactions in such a small area. Helmick further testified that he agreed with the contents of Exhibit 3, and he stated that the section quoted above under ‘Aluminum powder storage’ refers to a confined space, which was not the situation at the cited location.

  Both Helmick and Thomas Kondis, an Alcoa staff scientist for 19 years, presented unrebutted testimony that, given the relatively small amounts of water and aluminum powder present in the instant case, there was no possibility that in the room at issue or in any corner of that room hydrogen gas could have reached the 4% minimal atmosphere needed for the hydrogen gas to ignite and explode because the room was large, open, and well-ventilated by the overhead fans. Kondis also testified that ‘gases diffuse very rapidly.’ He stated that ventilation in the building was not limited to the exhaust fans because the building was ‘fairly loosely constructed’ so that drafts came through openings in the walls, windows, and doorways. Kondis testified that when aluminum powder comes into contact with water, there is an induction period when there is no reaction while the water diffuses through the aluminum oxide coating on the powder in order to reach the basic metal, with which it will react to form hydrogen gas. He stated that the results of experiments conducted at Alcoa laboratories showed that when aluminum powder that was finer and thus more reactive than the powder at issue in the instant case was added to an amount of water ‘far in excess of that needed to react with the aluminum,’ there was an induction period of 4 days before a significant amount of hydrogen was noticed, and it took 2 or 3 more days before all the hydrogen was produced. He further stated that in order to create an atmosphere of 4% hydrogen in the room at issue, it would take 30 to 50 pounds of aluminum powder reacting with 10 gallons of water for ‘[s]everal weeks, maybe months.’ Kondis testified that, besides the release of hydrogen gas and heat, the reaction of aluminum powder with water results in the production of a hydrated oxide of aluminum, which is a solid. When asked if anything in the area could spontaneously ignite as a result of this exothermic, or heat-producing, chemical reaction, Kondis replied that only the hydrogen gas could ignite, but it would not be able to reach the minimum 4% level under the circumstances at issue.

III

  In his decision, Judge Osterman found that the unrebutted testimony of Alcoa’s witnesses had established that ‘in a building the size of Respondent’s plant having exhaust fans in operation capable of removing between 6,000 and 9,000 cubic feet of air per minute, the danger of an explosion [from a 4% hydrogen gas concentration] is practically non-existent.’ Nevertheless, he further stated that ‘a hazard exists when water is permitted to come into contact with aluminum powder. A thermal reaction is created with the ever-present potential that spontaneous ignition can result.’ After noting that Alcoa recognized that danger as evidenced by its unsuccessful attempts to stop the water leakage, he further declared: ‘Although there is some evidence in the record that the thermal reaction would be slow, the fact is that the uncontrolled accumulation of water in close proximity to unspecified amounts of aluminum powder creates the potential for combustion.’ He concluded: ‘Evidence that the concentration of hydrogen in the atmosphere of the mixing room could not reach explosive proportions does not eliminate the danger of fire’ and that in a plant such as Alcoa’s in which ‘volatile materials’ are produced the consequences of a fire could be serious injury or death. The judge affirmed item 7 and assessed the $420 penalty proposed by the Secretary.

IV

  In order to establish a section 5(a)(1) violation, the Secretary must prove that: (1) the employer failed to render its workplace free of a hazard; (2) the hazard was recognized by the cited employer or the employer’s industry in general; (3) the hazard was causing or was likely to cause death or serious physical harm; (4) there was a feasible means by which the employer could have eliminated or materially reduced the hazard. E.g., Gearhart-Owen Industries, Inc., 82 OSAHRC 66/A2, 10 BNA OSHC 2193, 1982 CCH OSHD ¶26,329 (No. 4263, 1982).

  In the instant case, the Secretary’s inspector, Seifert, testified that the mixture of water and aluminum powder presents two possible hazards: the ignition and explosion of hydrogen gas resulting from contact between aluminum powder and water, and the ignition of aluminum powder itself from the heat resulting from contact between aluminum powder and water. With regard to the first potential hazard, the judge found that the concentration of hydrogen gas could not reach explosive proportions under the conditions present in Alcoa’s plant, and that finding is not disputed on review. The judge predicated the affirmance of the citation entirely on his finding that a danger of fire existed from mixing aluminum powder with water. On review, the parties dispute whether the record justifies this finding.

  As Alcoa indicates in its brief, it is unclear whether the judge’s finding that a danger of fire existed from mixing aluminum powder with water is based on a danger that the aluminum powder could ignite so as to cause a fire or on an inference that hydrogen gas is combustible even at a concentration less than the 4% necessary for an explosion. There is no evidence to distinguish between the concentration of hydrogen gas required for an explosion to occur and the concentration necessary for a fire. Indeed, all witnesses including Seifert used terms relating to both fire and explosion when discussing the 4% level of hydrogen. Accordingly, we conclude that the hydrogen was not shown to constitute a fire hazard in the circumstances here.

  Both parties also presented evidence on whether aluminum powder could spontaneously ignite from the heat produced by the reaction of aluminum powder and water. The only evidence presented on behalf of the Secretary that expressly stated that aluminum powder could ignite was Seifert’s testimony to that effect. The documentary exhibits introduced by the Secretary indicate only that some substance could ignite when water contacts aluminum powder; they may only be referring to the undisputed possibility of hydrogen gas ignition when hydrogen is present in a concentration of 4%. On the other hand, Kondis, the Alcoa chemist, testified that hydrogen would be the only substance that could ignite from the heat resulting from the contact between aluminum powder and water.

  Kondis was a trained chemist with 19 years of experience in the aluminum industry. He was familiar with experiments conducted at Alcoa involving the mixture of water and aluminum powder.3 Seifert, although a highly qualified safety professional, had little formal training in chemistry and no personal experience with experiments involving the mixture of aluminum powder with water. Accordingly, we conclude on this record that there is insufficient evidence that the aluminum powder itself could spontaneously ignite. Therefore, the Secretary did not prove by a preponderance of the evidence that any hazard due to this reaction existed in Alcoa’s workplace. See United States Steel Corp., 82 OSAHRC 35/A2, 10 BNA OSHC 1752, 1982 CCH OSHD ¶26,123 (No. 77–1796, 1982).

  Accordingly, the citation alleging noncompliance with section 5(a)(1) of the Act and the accompanying proposed penalty are vacated.

SO ORDERED.

FOR THE COMMISSION:

Ray H. Darling, Jr.

Executive Secretary

DATED: APR 27, 1983

COTTINE, Commissioner, concurring:

  The Secretary established the existence of a recognized hazard in this case, but the citation is properly vacated for lack of proof that an incident endangering employees was reasonably foreseeable. The Commission and reviewing courts have repeatedly pointed out that, in determining the existence of a recognized hazard, the proper inquiry is whether the general hazard alleged in the citation was recognized, not whether a particular incident constituted a violation of section 5(a)(1). E.g., General Dynamics Corp., Quincy Shipbuilding Div. v. OSHRC, 599 F.2d 453 (1st Cir. 1979); Western Massachusetts Electric Co., 81 OSAHRC 63/B13, 9 BNA OSHC 1940, 1981 CCH OSHD ¶25,470 (No. 76–1174, 1981); United States Steel Corp., 82 OSAHRC 35/A2, 10 BNA OSHC 1752, 1982 CCH OSHD ¶26,123 (No. 77–1796, 1982) (Cottine, Commissioner, concurring). In this case the Secretary alleged that precautions were not taken to avoid water accumulation on the floor in an area where it could come in contact with aluminum powder. The documentary evidence and the testimony of witnesses for both parties demonstrate that the general hazard of fire and explosion resulting from the combination of water and aluminum powder was recognized both by Alcoa and its industry. There is also evidence that Alcoa took precautions against contact between the two substances. Accordingly, the Secretary established a recognized hazard.

  The only evidence introduced by the Secretary concerning aluminum powder accumulations in proximity to water was that the compliance officer observed ‘a couple shovelfuls’ of the powder in the large, open ball mill room. There was no evidence that any greater amount of aluminum powder could accumulate or that spills of aluminum powder could occur in smaller or confined areas. The evidence establishes that the single spill of aluminum powder occurring in this case could not result in a 4% atmospheric concentration of hydrogen gas, and no other source of ignition or explosion involving that spill was proven. Accordingly, the evidence fails to establish that fire or explosion was reasonably foreseeable on the limited facts of this case. Accordingly, the citation is properly vacated. See U.S. Steel, supra.

UNITED STATES OF AMERICA

OCCUPATIONAL SAFETY AND HEALTH REVIEW COMMISSION

April 5, 1979

DECISION AND ORDER

Appearances:

Joseph T. Crawford, Esq. U.S. Department of Labor Office of the Solicitor, Region III

3535 Market Street, Room 14480 Philadelphia, Pennsylvania 19104 for the Complainant

Robert Cattaneo, Esq., and K. Jacqueline Bernat, Esq. Aluminum Company of America

No. 1501 Alcoa Building Pittsburgh, Pennsylvania 15219 for the Respondent

Marie Malagreca

400 Lock Street Tarentum, Pennsylvania 15084 for the United Steelworkers of America

John McIntyre Local 302

227 5th Avenue New Kensington, Pennsylvania for the United Steelworkers of America

OSTERMAN, Judge; OSHRC

  This is a proceeding initiated by the Respondent pursuant to Section 10(c) of the Occupational Safety and Health Act of 1970 [hereafter the Act] to contest two Citations and Notices of Proposed Penalties issued to Respondent on or about June 28, 1978.

  The standards which Respondent is alleged to have violated and which the Respondent challenges are the following:

Citation No. 1, Serious Violations.

Item 1. 29 C.F.R. 1910.178(c)(2)(vii) [Non-approved power-operated industrial trucks were being held in an atmosphere containing hazardous concentrations of metal dust]. Proposed penalty: $560.

Item 5. 29 C.F.R. 1910.309(a), Section 502–7(a) NEC, NFPA 70–1971 [Control transformers solenoids, impedance coils and resistors and overload devices, etc., were not dust-ignition proof]. Proposed penalty: $490.

Item 7. Section 5(a)(1) of the Act [Respondent failed to provide a place of employment free of recognized hazards-water leakage in No. 4 Ball Mill]. Proposed penalty: $420.

Item 8. Section 5(a)(1) of the Act [Respondent failed to provide a place of employment free of recognized hazards-use of spark producing shovel on Blending Room Balcony]. Proposed penalty: $490.

Citation No. 2, Non-Serious Violations.

Item 1. 29 C.F.R. 1910.178(a)(6) [Nameplate(s) or markings for powered industrial truck No. 6 in use in Filter Mixer Room of Ball Mill not in place]. Proposed penalty: none

Items No. 2, 3, 4, and 6 of Citation No. 1 were not contested and have become final by operation of law. Items 1, 5, 7 and 8 were contested.

At the commencement of the hearing the parties stipulated on the record as follows:

  Citation No. 1, Item 1. The Secretary will reduce the proposed penalty from $560 to $400 and the Respondent, in consideration of this reduction, withdraws its Notice of Contest to this item.

  Citation No. 1, Item 2. The Secretary modifies the abatement date to read January 19, 1979 (instead of August 28, 1978) and the Respondent agreed to withdraw its Notice of Contest to the abatement date for this item.

  Citation No. 1, Item 5. The Secretary agrees to reduce the classification of ‘serious’ to ‘non-serious’ and reduce the penalty from $490 to zero. Respondent agrees to abate the violation and withdraw its Notice of Contest.

  Citation No. 2, Item 1. The Secretary agrees to withdraw this item. No penalty was proposed (Tr. 3–4).

  Items 7 and 8 of Citation No. 1 remained for determination at the hearing.

  Item 7 charges that Respondent violated Section 5(a)(1) of the Act in that ‘precautions were not taken to avoid the leakage and accumulation of water on the floor near the powder dust loading area where it could come into contact with aluminum powder in the No. 4 Ball Mill. Reference Par. 7–1.1, Chapter 7, NFPA Std. No. 651–1974.’ The proposed penalty is $420.

  Item 8 charges a violation of Section 5(a)(1) of the Act in that ‘Employees were permitted to use a spark producing tool (20 inch steel shovel) on the Blending Room Balcony of a dust-making building. Reference Par. 7–6.1, Chapter 7, NFPA Std. No. 651–1974.’ The proposed penalty is $490.

  Section 5(a)(1) of the Act provides that each employer:

‘shall furnish to each of his employees employment and a place of employment which are free from recognized hazards that are causing or are likely to cause death or serious physical harm to his employees.’

  Evidence given at the hearing by the compliance officer shows that on May 19, 1978, an inspection was made at Respondent’s Logan Ferry plant after a complaint was received by the Area Director (Tr. 10). The investigation disclosed that water was leaking from a crack near a door leading from the No. 4 Ball Mill and from an improper seal around the door and accumulating in puddles on the floor of the adjoining powder dust loading area. The compliance officer observed two puddles of water, approximately 1 or 2 gallons, several feet in diameter, 7 or 8 feet from a chute where a ‘couple of shovel fulls’ of aluminum powder was lying on the floor (Tr. 10–12, 14, 30, 37–38). The compliance officer testified that contact between aluminum powder and water causes a thermal reaction and a potentially explosive condition which results from the release of hydrogen. If the concentration of hydrogen should reach 4% of the volume of air in the room, an explosion is possible (Tr. 13, 16, Ex. C–1, C–2). The compliance officer stated also that Respondent was aware of the water leakage and had been advised by one Mr. Helmick, the Safety Supervisor, that management had tried repeatedly to stop the leakage and had issued instructions to its employees to keep the area of the water cleaned up and clear of powder (Tr. 14–15). The condition had existed since 1975 (Tr. 38). At the time of the inspection, two employees were working in the area of the mixing room (Tr. 21).

  With respect to Item 8, the compliance officer testified that upon inspection of the blending room he observed a steel shovel in the general area; that a steel shovel could cause a spark on contact with the concrete floor; and that a spark could ignite the aluminum powder (Tr. 21–28, Ex. C–2). The compliance officer further stated that the Respondent’s management ‘were under the impression that they [the shovels] were all the non-ferrous’ (Tr. 24) and that ‘they were not aware of that particular shovel’ (Tr. 25).

  On cross-examination, the compliance officer conceded that he did not know how much hydrogen is released when water and aluminum powder are mixed or what the reaction rate would be (Tr. 31–32). The compliance officer also admitted with respect to Item 8 that a steel shovel was found on the balcony of the blending room; that it was standing in a corner; and that no one was using it (Tr. 33). When asked the question, the compliance officer stated that there may have been traces of aluminum powder on the shovel (emphasis added) (Tr. 34).

  Testimony by Respondent’s witnesses indicates with respect to Item 7 that the building itself is approximately 65 feet long, 25 feet wide and 19 feet high (Tr. 45) and that the dimensions of the mixing room are approximately 18 by 25 feet (Tr. 56–57). The building contains three overhead exhaust fans which are in operation and are capable of removing between 6,000 and 9,000 cubic feet of air per minute (Tr. 45). One exhaust fan is located directly above the mixing area (Tr. 57). Except for the Ball Mills, the area of the mixing room is not tightly enclosed (Tr. 66). The floor in the charging area where the powder was observed by the compliance officer is approximately 2 inches higher than the surrounding area where the puddles of water were seen by the compliance officer (Tr. 46).

  Further testimony by Respondent’s witnesses indicates that the potential for an explosion from contact between the aluminum powder and water would exist only in a confined space, e.g. water seeping into a metal drum packed with aluminum powder which could result in a rupture of the drum from the pressure of the hydrogen gas which is being produced (Tr. 48, 53–54, 67; Ex. C–3). It was uncontradicted by the Secretary that the potential for an explosion would exist only if the concentration of hydrogen in the room reached a level of 4% of the total volume of air and that in a building the size of Respondent’s plant having exhaust fans in operation capable of removing between 6,000 and 9,000 cubic feet of air per minute, the danger of an explosion is practically non-existent (Tr. 62–67).

  To sustain a violation of Section 5(a)(1) of the Act, the Secretary must establish by a preponderance of the evidence: (1) That the hazard charged is one which is generally known in the particular industry of which Respondent is a part or one which can be readily detected by the senses, and (2) that the hazard charged is causing or is likely to cause death or serious harm to Respondent’s employees.

  It seems clear from the evidence in the record that a hazard exists when water is permitted to come into contact with aluminum powder. A thermal reaction is created with the ever-present potential that spontaneous ignition can result. The Respondent has recognized this danger ad indicated by the fact that for some time prior to the compliance officer’s inspection unsuccessful efforts were made to eliminate the water leakage in Respondent’s mixing room (Tr. 14–15). Although there is some evidence in the record that the thermal reaction would be slow, the fact is that the uncontrolled accumulation of water in close proximity to unspecified amounts of aluminum powder creates the potential for combustion. In a plant such as the one operated by Respondent, which is devoted to the production of volatile materials, the consequences of any fire could be disastrous and ‘likely to cause death or serious injury’ to employees. Evidence that the concentration of hydrogen in the atmosphere of the mixing room could not reach explosive proportions does not eliminate the danger of fire. I conclude that Respondent was in violation of the general duty clause because of its failure to eliminate the leakage of water in its mixing room.

  With respect to Item 8, the record indicates that Respondent had a firm policy of requiring the use of non-sparking shovels in its mixing room (Tr. 50, Ex. C–1, par. 4.4.2). The steel shovels found by the compliance officer were not in use during his inspection (Tr. 33) and there is evidence that they had been left on the balcony of the mixing room by an employee engaged in carpentry or bricklaying (Tr. 50). It was the compliance officer’s testimony that the Respondent’s management was alarmed when they learned that a steel shovel had been found in the mixing room and that ‘they were under the impression that the(y) were all the non-ferrous’ (Tr. 24). In these circumstances I conclude that the evidence is insufficient to support a finding that ‘sparking’ tools were used to shovel aluminum powder in Respondent’s mixing room.

CONCLUSIONS OF LAW

  1. On the day of inspection the Respondent was in violation of Section 5(a)(1) of the Act for the reasons described in Item No. 7 of the Citation herein.

  2. The record herein does not support the conclusion that on the day of inspection Respondent was in violation of Section 5(a)(1) of the Act as alleged in Item No. 8 of the citation.

ORDER

  Pursuant to Section 10(c) of the Act and Rule 66 of this Commission’s Rules of Procedure, it is ORDERED:

  1. That Item 7 of the Citation and the penalty proposed for this violation be, and the same hereby are AFFIRMED.

  2. That Item 8 of the Citation herein be, and the same hereby is, DISMISSED. The penalty proposed is VACATED.

HENRY K. OSTERMAN

Judge, OSHRC

Dated: APR 5, 1979

Hyattsville, Maryland

1 That section provides that ‘[e]ach employer . . . shall furnish to each of his employees employment and a place of employment which are free from recognized hazards that are causing or are likely to cause death or serious physical harm to his employees.’

2 At the time of the hearing, Seifert had been a compliance officer for five years. He testified that he had taken ‘numerous’ courses at the Occupational Safety and Health Administration (‘OSHA’) training institute, including a course on explosive safety. He further stated that, prior to his employment at OSHA, he had had experience with the National Fire Protection Association standards in his jobs as Safety Director at two Veterans Administration hospitals for seven years and as Safety Officer with the Air Force for 17 years. When asked on cross-examination if he had any formal education with regard to chemistry or physics, especially light metal powders such as aluminum powder, Seifert replied that he had not, other than the two-week to one-month courses that he had taken which ‘touched on’ chemistry and physics, but ‘not in any detail.’ He also testified that he had never inspected a powder plant before.

3 As noted above, Kondis testified that Alcoa had conducted experiments in which water was mixed with aluminum powder that was finer, and therefore more reactive, than the powder present on the floor of the ball mill room. Although the primary purpose of this testimony was to demonstrate the slow rate at which the reaction would produce hydrogen gas, his familiarity with the experiments supports his testimony that the only substance that could ignite from the mixture of aluminum powder and water is hydrogen gas. Furthermore, Alcoa’s counsel sought to question Kondis on experiments Kondis had conducted with pyroforic aluminum. Upon objection to this line of questioning by the Secretary, Alcoa’s counsel explained that pyroforic aluminum is more reactive than the oxidized aluminum present in the ball mill room and that Kondis would testify that he had been unable to create spontaneous ignition by mixing pyroforic aluminum with water. Judge Osterman sustained the Secretary’s objection and did not permit Alcoa to place this evidence in the record. Inasmuch as the case involves the ability of powdered aluminum to spontaneously ignite when mixed with water and the evidence proffered by Alcoa was highly relevant to this issue, it was error for the judge to exclude the evidence. However, in light of our disposition vacating the citation, the error was harmless.