United States Court of Appeals FOR THE DISTRICT OF COLUMBIA CIRCUIT
Argued March 30, 2026 Decided June 12, 2026
No. 25-1173
TCP SPECIALISTS, LLC, PETITIONER
v.
SECRETARY OF LABOR, RESPONDENT
On Petition for Review of an Order of the Occupational Safety & Health Review Commission
Darren S. Harrington argued the cause for petitioner. With him on the briefs was Brian L. Hurt.
Joseph J. Quick, Attorney, U.S. Department of Labor, argued the cause for respondent. With him on the brief were Jonathan Berry, Solicitor of Labor, Edmund C. Baird, Associate Solicitor for Occupational Safety and Health, Heather R. Phillips, Counsel for Appellate Litigation, and Anne E. Bonfiglio, Attorney. Amy S. Tryon and Louise Betts, Attorneys, entered appearances. 2 Before: HENDERSON and CHILDS, Circuit Judges, and RANDOLPH, Senior Circuit Judge.
Opinion for the Court filed by Circuit Judge HENDERSON.
KAREN LECRAFT HENDERSON, Circuit Judge: After a pressurized pipe ruptured and struck employees at an oil well, the Secretary of the Department of Labor (Labor) issued a citation to petitioner, TCP Specialists, LLC (TCP). TCP did not provide the piping and did not direct the depressurizing operation that led to the accident. But TCP’s employees were standing unnecessarily close to the depressurizing well, the Secretary alleged, creating a risk that they would be struck by equipment or caught in an explosion. He also contended that TCP feasibly could have abated that hazard by establishing a buffer zone around the well. Following a three-day evidentiary hearing, an administrative law judge (ALJ) agreed and upheld the citation. TCP petitioned for review, maintaining, among other things, that the ALJ violated Occupational Safety and Health Review Commission (Commission or OSHRC) precedent in defining the hazard; that substantial evidence does not support the ALJ’s factual findings; and that one provision of the Occupational Safety and Health Act (OSH Act) is unconstitutional as applied to it. Finding each claim meritless, we deny the petition.
I. BACKGROUND
A
In the 1970 OSH Act, Pub. L. No. 91-596, 84 Stat. 1590, the Congress sought to promote “safe and healthful working conditions” for “every working man and woman in the Nation.” 29 U.S.C. § 651(b). To attain that goal, the Act divided implementing authority between the Labor Secretary and the 3 Commission. The Secretary enforces health and safety standards by promulgating rules and issuing citations. Martin v. OSHRC, 499 U.S. 144, 147 (1991); see 29 U.S.C. §§ 655, 658. The Commission “act[s] as a neutral arbiter and determine[s] whether the Secretary’s citations should be enforced over employee or union objections.” Cuyahoga Valley Ry. Co. v. United Transp. Union, 474 U.S. 3, 7 (1985) (per curiam); accord 29 U.S.C. §§ 651(b)(3), 661. The employer may contest the Secretary’s allegations before a Commission ALJ. Atlas Roofing Co. v. OSHRC, 430 U.S. 442, 446 (1977); 29 U.S.C. §§ 659, 661(j). Absent the Commission’s discretionary review, the ALJ decision becomes a final order of the Commission. Atlas Roofing, 430 U.S. at 446; 29 U.S.C. § 661(j).
Section 654 equips the Secretary with two bases of enforcement. First, the Secretary may allege a violation of the specific health and safety standards promulgated under the Act. 29 U.S.C. § 654(a)(2); Nat’l Realty & Constr. Co. v. OSHRC, 489 F.2d 1257, 1261 (D.C. Cir. 1973). Courts sometimes refer to this basis as the “[S]pecific [D]uty [C]lause.” E.g., Carlyle Compressor Co. v. OSHRC, 683 F.2d 673, 676 (2d Cir. 1982). Second, the Secretary may allege a violation of the “General Duty Clause,” which is the one of the two clauses relevant here. 29 U.S.C. § 654(a)(1); Nat’l Realty, 489 F.2d at 1261. The General Duty Clause functions as a gap-filler, providing “an enforcement mechanism when no specific OSHA standard applies to a particular hazard.” Interpretation of the General Duty Clause, 90 Fed. Reg. 28370, 28371 (proposed July 1, 2025) (to be codified at 29 C.F.R. pt. 1975). Under it, each employer must “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.” 29 U.S.C. § 654(a)(1). We have interpreted that directive to encapsulate four elements: 4
(1) an activity or condition in the employer’s workplace presented a hazard to an employee, (2) either the employer or the industry recognized the condition or activity as a hazard, (3) the hazard was likely to or actually caused death or serious physical harm, and (4) a feasible means to eliminate or materially reduce the hazard existed.
BHC Nw. Psychiatric Hosp., LLC v. Sec’y of Lab., 951 F.3d 558, 563 (D.C. Cir. 2020) (citation modified).
B
The accident underlying this petition unfolded at the Blackstone B1 well, a gas well in San Augustine County, Texas. C6 Operating (C6), an oil and gas producer, operated the wellsite. In October 2022, C6 hired contractors to perform a “workover,” J.A. 18, aimed at “maintaining or restoring the productivity of [the] well,” 30 C.F.R. § 250.601. Several contractors are key to the story. One contractor, Jaguar Energy Services, operated a “frac stack,” which is an assemblage of valves, handwheels and other equipment used to control a well’s flow rate and pressure. J.A. 18. Jaguar also provided the piping used to transport fluids and pressure to and from the well. Another contractor, Reliance Well Services, operated a “pump tank,” an apparatus used to draw pressure from the well, through the frac stack and piping and, finally, to itself. J.A. 18. Generally speaking, then, Jaguar and Reliance controlled the well’s flow rate and pressure.
Petitioner TCP provided wireline services. In the oil and gas industry, these “involve using a specialized cable to lower 5 tools and equipment into a well[].” Panuswee Dwivedi, Wireline Services, ADI Analytics (May 29, 2025), https://adi- analytics.com/2025/05/29/wireline-services-supporting-the-oil -gas-industry-through-evolving-depths/ [https://perma.cc/CN6W- L4JQ]. Often, contractors use wireline services to complete routine maintenance in or gather geophysical data from a well. See Domingue v. Ocean Drilling & Exploration Co., 923 F.2d 393, 394 n.3 (5th Cir. 1991).
On December 5, 2022, a TCP crew consisting of three employees arrived at the well. They agreed to perform two wireline operations. The first of these, lasting approximately two hours, they executed without issue. Before the second operation, however, Jason Walker—a TCP supervisor— noticed that the well’s pressure had increased. He notified three non-TCP employees of this fact and the group agreed to depressurize the well during the second wireline operation. As TCP undertook that operation, however, Walker realized that the well’s pressure had nearly doubled, meaning that no one had depressurized it. Walker sent another TCP employee, Tristan McLelland, to apprise the wellsite general supervisor.
After McLelland relayed the message, a small group of employees congregated at the wellhead. At this point, several factors combined to produce catastrophe. The wellsite general supervisor instructed a Jaguar supervisor to depressurize the well by opening a valve on the frac stack. Contrary to protocol, however, the Jaguar supervisor opened one valve without ensuring that another was fully shut. Instead of slowly depressurizing the well as the supervisor had intended, this sequence released a flood of pressure from the well into the pipe. Jaguar had borrowed that pipe from Reliance earlier in the day but, unbeknownst to either, the pipe had corroded. The sudden pressure caused the corroded pipe to rupture and flail violently, striking several employees and throwing one nearly 6 sixty feet. As a result of the accident, one TCP employee was hospitalized; two other employees—the wellsite general supervisor and a Jaguar employee—died.
C
The Secretary cited TCP for a serious violation of the OSH Act’s General Duty Clause, 29 U.S.C. § 654(a)(1). He alleged that TCP’s employees had been standing unnecessarily close “to the frac stack and pressured piping” during the depressurization, which had exposed them to “fire, explosion, and struck-by hazards.” J.A. 6. He also noted two abatement measures that could have prevented the accident. The first of these, sensibly enough, was to establish a buffer zone around the wellhead. The second was to restrain the pressurized piping to ensure that it did not whip. TCP contested the citation and the parties proceeded to an evidentiary hearing.
At the hearing, the ALJ considered testimony from the three TCP employees who had been at the wellsite. He also heard testimony from two experts—Paul Luker, whom the Secretary called, and Jarold Elgin, whom TCP called. Luker’s testimony constituted much of the Secretary’s evidence on the issues relevant to TCP’s petition and, accordingly, we focus on his testimony. At the time, Luker was an independent safety consultant in the oil and gas industry. Within that industry, he had around thirty-five years of safety-related experience. He spent almost all of his time visiting job sites to review their safety protocols and devoted about thirty per cent of that time to workovers. The Secretary tendered Luker “as an expert in the oil and gas industry[,] specifically on workover operations . . . and the safety policies and procedures of those operations,” J.A. 644, and the ALJ admitted Luker as an expert in those fields. 7 To reach his opinion, Luker considered technical literature, industry publications, witness depositions, the Secretary’s description of the accident and reports filed by the on-site contractors. He drew three relevant conclusions. First, he concluded that a buffer zone around the wellhead would have materially reduced the risk of harm to TCP’s employees. Citing evidence that no employee farther than 100 feet away had been injured, Luker testified that a buffer zone of that size would have sufficed. Second, he concluded that a “reasonably prudent employer” in TCP’s shoes would have established such a buffer zone. J.A. 737–38. He explained that employers in the industry “routinely” employ 100-foot buffer zones during comparable operations and that industry standards, training and risk assessments recommend such buffer zones. J.A. 786. Third, Luker testified that TCP feasibly could have implemented a buffer zone because it was not necessary for its employees to stand near the wellhead during the depressurization. Tristan McLelland, who notified the wellsite general supervisor of the elevated pressure, could have held that conversation almost anywhere on the jobsite. And Brian Walker, asked to monitor pressure, did not need to do so until later in the operation and he could have carried out his task away from the wellhead. In sum, Luker testified, a buffer zone was an effective means of abatement and one that TCP feasibly could have implemented.
The ALJ determined that the Secretary’s evidence satisfied the General Duty Clause’s four elements. TCP Specialists, LLC, No. 23-1002, 2025 WL 1735784, at *5–18 (OSHRC May 12, 2025) (ALJ). First, he concluded that a “hazard” existed. Id. at *8. That hazard, he explained, was the risk posed by the employees’ “close proximity to the frac stack and pressurized piping during depressurization of the frac stack.” Id. Second, the ALJ concluded that the hazard was “recognized” and a supervisor’s hazard recognition can be imputed to the 8 employer. Id. Supervisor Jason Walker testified that he knew depressurizing a well created serious risks, including the pipe “exploding, rupturing, and whipping around.” Id. Third, the ALJ found that the hazard caused or was likely to cause “death or serious physical harm.” Id. at *9 (citation modified). For evidence of that, the ALJ reasoned, one need look no further than the underlying accident itself, as well as the deaths and injuries it caused. Id. Fourth, the ALJ found that the hazard was “preventable.” Id. at *9–18. Recall that the Secretary had proposed two measures by which TCP could have abated the hazard—(1) establishing a buffer zone or (2) restraining the pressurized piping. The ALJ rejected the latter, concluding that TCP had no authority to impose line restraints at the well. Id. at *10–11. But he accepted the former, determining that TCP could have feasibly established a buffer zone around the depressurizing well and that such a buffer zone would have materially reduced the risk of harm to its employees. Id. at *11–18. Finding the General Duty Clause’s elements met, the ALJ upheld the citation and assessed a corresponding penalty of $6,250. Id. at *23.
OSHRC denied TCP’s request for discretionary review. The ALJ’s decision therefore became a final order of the Commission. See 29 U.S.C. § 661(j). From that order, TCP filed this timely petition for review.
II. ANALYSIS
We have jurisdiction of TCP’s petition under 29 U.S.C. § 660(a). We review the ALJ’s factual findings for substantial evidence, id., and will overturn the ALJ’s application of the law to the facts only if it is “arbitrary, capricious, an abuse of discretion, or contrary to law,” 5 U.S.C. § 706(2)(A). Otis Elevator Co. v. Sec’y of Lab., 762 F.3d 116, 120–21 (D.C. Cir. 2014) (citation modified). 9
TCP asserts that the ALJ deviated from Commission precedent in defining the underlying hazard. That precedent imposes two relevant requirements. The Secretary must define the hazard “in terms of the physical agents that could injure employees,” rather than in terms of the absence of a proposed abatement. Chevron Oil Co., 1983–1984 CCH OSHD ¶ 26507, 1983 WL 23864, at *1–2 nn.5–6 (No. 10799, 1983). And the Secretary must identify a hazard composed of “conditions or practices” over which the employer can exercise control. Arcadian Corp., 2005 CCH OSHD ¶ 32756, 2004 WL 2218388, at *7 (No. 93-0628, 2004). We consider each in turn.
TCP first contends that the ALJ impermissibly defined the hazard as the absence of an abatement. It emphasizes that the Secretary alleged that TCP had exposed its employees to a hazard because they were standing in “close proximity” to the wellhead during depressurization. Pet’r’s Br. 17 (quoting J.A. 6). But, TCP asserts, that is just another way of saying that TCP had failed to establish a buffer zone, which doubled as the Secretary’s proposed abatement measure in this case.
TCP misreads the record. No one defined the hazard as the absence of a buffer zone. Instead, the Secretary and ALJ both identified the hazard by reference to the physical agents composing it—the frac stack and pressured piping. This much is plain from both the citation itself as well as the ALJ’s order. In the citation, the Secretary alleged that TCP’s employees “were exposed to fire, explosion, and struck-by hazards” during the depressurization. J.A. 6. He then specifically identified the “frac stack and pressured piping” as the physical agents underlying that risk. J.A. 6. The ALJ confirmed this understanding, explaining that the “hazardous condition” was 10 the risk created by the employees’ proximity to the depressurizing frac stack—proximity that exposed the employees “to fire, explosion, and struck-by hazards.” TCP Specialists, 2025 WL 1735784, at *8. In other words, the Secretary and ALJ accurately described the hazard by reference to objects composing it—the frac stack and pressured piping. Both are “physical agents” within the meaning of Commission precedent. Chevron Oil Co., 1983 WL 23864, at *2 n.6; see id. at *2 (corroded piping); cf. Arcadian Corp., 2004 WL 2218388, at *8 (pressurized reactor). And neither can be described as the absence of a particular abatement measure.
Granted, the citation and order also described the employees’ “close proximity” to the wellhead as part of the hazard. J.A. 6. But that is not inconsistent with Commission precedent. Like most potentially hazardous operations, the depressurization posed a risk of harm only to those standing nearby. See, e.g., Coastal Drilling E., LLC, 2019 CCH OSHD ¶ 33701, 2018 WL 7080227, at *3, *5 (No. 17-1179, 2018) (ALJ) (crushing hazard); cf. Mo. Basin Well Serv., 2018 CCH OSHD ¶ 33648, 2018 WL 1309482, at *2–3 (No. 13-1817, 2018) (“unsafe distance” between two objects). Nothing in Commission precedent forbade the ALJ from recognizing that fact.
TCP next argues that, in defining the hazard, the ALJ failed to identify “conditions or practices” over which TCP exercised control. Pet’r’s Br. 29 (citation modified). In TCP’s view, it lacked any control over the depressurization, which Jaguar performed, or the corroded pipe, which Reliance provided. And, it maintains, those were the only conditions making up the hazard.
As noted above, however, the employees’ proximity to the wellhead was a necessary component of the hazard. TCP 11 exercised control over that proximity: the ALJ made a factual finding, which TCP has not challenged, that TCP controlled its employees’ conduct and location during the depressurization. TCP Specialists, 2025 WL 1735784, at *6, *8. Because the ALJ identified a condition or practice over which TCP did, in fact, exercise control, it did not run afoul of Commission precedent.
TCP next challenges the evidence supporting the ALJ’s finding that it could have prevented the hazard. Specifically, TCP assails the ALJ’s conclusions that a buffer zone would have materially reduced the risk of harm to its employees and would have been feasible to employ. It also challenges the ALJ’s finding that its existing safety procedures were inadequate.
The substantial-evidence standard is not exacting. Biestek v. Berryhill, 587 U.S. 97, 103 (2019). It requires only “such relevant evidence as a reasonable mind might accept as adequate to support a conclusion”—a threshold the Supreme Court has described as “more than a mere scintilla.” Consol. Edison Co. v. NLRB, 305 U.S. 197, 229 (1938). In applying the standard, we must adopt “a very high degree of deference,” Island Architectural Woodwork, Inc. v. NLRB, 892 F.3d 362, 370 (D.C. Cir. 2018) (citation modified), affirming the agency’s findings “unless no reasonable factfinder could find as [it] did.” Hood River Distillers, Inc. v. NLRB, 130 F.4th 204, 216 (D.C. Cir. 2025) (citation modified). Nevertheless, substantial-evidence review demands more than “inordinate faith in the conclusory assertions of an expert.” Sea Robin Pipeline Co. v. FERC, 795 F.2d 182, 188 (D.C. Cir. 1986). 12 This standard is easily met here. First, substantial evidence supported the ALJ’s finding that a buffer zone would have materially reduced the risk of harm to TCP’s employees. Luker testified that this was so. As he explained, the “further back” an employee is from the hazard, “the less likely” he is “to suffer impact” and, even if he does, the “severity of the impact will be greatly reduced.” J.A. 810. In Luker’s experience, risk assessments for operations like the one here call for a 100-foot buffer zone around the wellhead for nonessential personnel. And the industry arrived at the 100- foot figure in particular, he went on, because it is sufficient both to “provide a safe zone” and to obviate the need for employees to execute complex, back-of-the-napkin calculations while out in the field. J.A. 785. Indeed, TCP’s own expert agreed that a 100-foot buffer zone, if used, would have reduced the risk of harm to TCP’s employees. The underlying facts provided direct evidence bolstering that conclusion: no employee outside Luker’s proposed buffer zone was injured in the accident. This combination of uncontested expert testimony and direct evidence constitutes substantial evidence. See Am. Wrecking Corp. v. Sec’y of Lab., 351 F.3d 1254, 1262 (D.C. Cir. 2003). 1
TCP emphasizes that pipe fragments were found more than 100 feet away. But the ALJ considered that potentially contradictory evidence and explained why he did not find it persuasive. TCP Specialists, 2025 WL 1735784, at *13–15. The ALJ is entitled to choose between conflicting interpretations of the evidence and where, as here, that choice
1 We do not conclude that a 100-foot buffer zone is required during all depressurizing operations but only that substantial evidence supports the ALJ’s determination that it was necessary here. 13 is supported by substantial evidence, we must respect it. Perdue Farms, Inc. v. NLRB, 144 F.3d 830, 838 (D.C. Cir. 1998).
TCP also asserts that no substantial evidence supported the ALJ’s finding that a buffer zone was compatible with its employees’ essential tasks such that employing a buffer zone would have been feasible. We disagree. The Secretary asked Luker expressly whether either of the TCP employees inside his proposed buffer zone was “performing essential work” during the depressurization. J.A. 747. Luker testified that both were not. McLelland could have spoken to the wellsite general supervisor anywhere at the jobsite. And Brian Walker did not need to be monitoring the pressure during that stage of the operation and could have stood away from the wellhead while doing so. Luker’s conclusion finds support in industry practice, which can be “highly probative” of “feasibility.” L.R. Willson & Sons, Inc. v. OSHRC, 698 F.2d 507, 514 (D.C. Cir. 1983). The oil and gas industry often employs or recommends buffer zones in situations where, as here, a pipe is being pressurized. In conjunction with Luker’s expert testimony, that industry practice serves as substantial evidence that a buffer zone was feasible.
In addition, TCP maintains that the Secretary failed to show that its existing safety measures were inadequate. As it observes, we have stated that if “an employer has existing safety procedures, the burden is on the Secretary to show that those procedures are inadequate” to abate the hazard. SeaWorld of Fla., LLC v. Perez, 748 F.3d 1202, 1215 (D.C. Cir. 2014). But the key predicate is “existing safety procedures.” TCP apparently had none. Its employees repeatedly acknowledged that TCP had no policies or procedures in place that were responsive to the hazard. The employees did hold a small safety meeting at the start of the 14 workday. But they did not even discuss pressurized equipment at the meeting. It is unclear, then, what relevant safety procedures TCP believes that it had in place. And it cites nothing in the record to support its argument. Thus, uncontradicted employee testimony supports the ALJ’s finding that TCP had established no “rules or an alternative abatement method” for the hazard at issue here, TCP Specialists, 2025 WL 1735784, at *13, and for that reason we reject TCP’s claim.
Finally, TCP contends that the General Duty Clause is unconstitutionally vague as applied to it. It does not deny that we have repeatedly rejected vagueness challenges to the General Duty Clause because there is little fair-notice concern if the hazard is “preventable” and “a reasonably prudent employer in the industry would have known that the proposed method of abatement was required.” BHC Nw. Psychiatric Hosp., 951 F.3d at 566 (citation modified). Rather than confronting that caselaw head-on, TCP advances a slightly different argument. Specifically, TCP contends that the definition the Secretary and ALJ gave the hazard was so “broadly worded” that it failed to provide TCP fair notice of its obligations. Pet’r’s Br. 24. In the citation, TCP emphasizes, the Secretary stated that TCP’s employees were “standing in close proximity to pressurized equipment.” Pet’r’s Br. 24 (quoting J.A. 6). And in his order, the ALJ at times seemed to define the hazard in the same way. TCP Specialists, 2025 WL 1735784, at *7. But if simply standing near pressurized equipment—which is ubiquitous in the oil and gas industry— violates the General Duty Clause, then, TCP says, the clause itself is unconstitutional.
Again, TCP misreads the record. The Secretary’s allegations have consistently been keyed to a discrete 15 operation—depressurization—rather than the dangers of pressurized equipment in general. As the Secretary explains: “The hazard is not, as TCP repeatedly phrases it, ‘working in proximity to pressurized equipment.’ The depressurization activity is key to the existence of the hazard.” Secretary’s Br. 25 n.10 (citations omitted). That understanding has been common ground throughout the proceedings, as reflected in the citation, the Secretary’s opening statement, witness testimony and the ALJ’s order. 2 Indeed, TCP itself seems, once, to have shared that understanding. In its petition for discretionary Commission review, TCP described the alleged hazard as its employees’ “expos[ure] to fire, explosion and struck-by hazards while depressurizing a frac stack.” J.A. 89. It is only now that TCP—and TCP alone—seems to misunderstand the contours of the hazard for which it was cited. Its misunderstanding cannot be traced to any defect of notice, and, accordingly, we reject TCP’s claim.
* * *
TCP also argues that it lacked knowledge of the violative conditions and that the Secretary failed to establish hazard recognition. We have considered and rejected these arguments. To the extent TCP raises other arguments, they are either plainly groundless or undeveloped and thus forfeited.
2 Even if the ALJ used shorthand once or twice in suggesting that pressurized equipment itself created the risk, he left little doubt about the nature of the hazard when he said: “The hazardous condition here was TCP’s employees standing in close proximity to the frac stack and pressured piping during depressurization of the frac stack, which exposed them to fire, explosion, and struck-by hazards.” TCP Specialists, 2025 WL 1735784, at *8. 16 For the foregoing reasons, TCP’s petition for review is denied.
So ordered.