Fumigation, Semiconductors, Human Self-Ignition and Methamphetamine
Steven Curry, M.D.*
Eugene Ngai, M.S.#
Greg Makar, D.O.*
*University of Arizona College of Medicine
Phoenix Department of Medical Toxicology Banner
University Medical Center Phoenix Phoenix, AZ
In the previous Fellow Friday on arsine, we introduced metal hydrides, their general physical properties, and how they are used in the semiconductor industry. Our arsine discussion explored the toxicity of hydrides of germanium, antimony, selenium and tellurium. We also noted how garlic odors were associated with various methylated elements surrounding arsenic in the periodic table, and how humans methylate some metals. The flammability and explosive nature of various hydrides was emphasized using silane as an example.
We now turn our attention to phosphine (PH3), the hydride of phosphorus. Phosphine can be somewhat confusing to toxicology fellows who hear of phosphine as a semiconductor gas, but also hear about deaths on ships related to phosphine used in grain holds as a fumigant. And they hear of children who have died from phosphine that, in some way, was used to kill rodents near their homes. There are also reports of phosphine being involved in methamphetamine laboratories – why would meth cooks need phosphine? And finally, there are those stories of human self-ignition. After all, every toxicologist should have a differential diagnosis of spontaneous combustion. I again am honored to have Eugene Ngai, an international expert in the production, transportation and use of metal hydrides join me, as well as Dr. Greg Makar, one our medical toxicology fellows, as we attempt to simplify phosphine toxicity.
Phosphine is a colorless gas that is slightly heavier than air. Pure phosphine is without an odor, but a garlic odor is commonly noted during phosphine releases from methylated phosphorus contaminants in many instances, especially with the release of phosphine from metal phosphides (described below).
Like silane and diborane, phosphine is pyrophoric (autoignites in 1 atm air at 130°F), which leads to fire/explosion hazards when atmospheric concentrations rise high enough, including autoignition when leaking from a cylinder.
Phosphine gas is produced and stored in cylinders for use in fumigation or in the semiconductor industry by the dissociation of polyphosphoric acid. Solvay, the world’s largest manufacturer of phosphine, converts pure white phosphorus to polyphosphorous acid. The polyphosphorous acid then, is heated to high temperatures at which it dissociates into phosphine and phosphoric acid.
Pure phosphine in cylinders is commonly used for metal organic chemical vapor deposition (MOCVD) processes as discussed in the arsine post. It frequently, but not always, is mixed with other gases for use as a dopant or to grow an electrical insulating layer on semiconductor chips.
Phosphine from cylinders is also used around the world for fumigation of grains and commodities. In commodity fumigation, phosphine released from cylinders is meant to remain at a relatively low, but adequate, constant concentrations for at least 2-3 days in order to kill pests. In this setting, phosphine is diluted in CO2 or nitrogen, in part, to lessen flammability.
A second major source of phosphine is the liberation of phosphine from aluminum, magnesium or zinc phosphide. These metal phosphides are solids, and when they react with water, including ambient humidity, phosphine is released. Of course, acids produce even more rapid phosphine formation. These reactions are similar to arsine generation from metal arsenides discussed in the previous post.
Metal phosphides, especially AlP and Mg3P2 , are used for commodity fumigations (grain, lumbar, etc.), such as during transportation by ship, but also when products require significant storage time, such as in silos. The solid phosphides (e.g., tablets, sachets of granules, dust, etc.) are dispensed into sealed atmospheres (e.g., ship holds, grain elevators) or under plastic tarps. Over time, phosphine is released as the phosphide reacts with ambient humidity. The phosphide products commonly contain other ingredients to slow liberation of phosphine and to lessen spontaneous flammability.
When used as a rodenticide, the metal phosphide is placed in burrows and typically covered with dirt. Care must be taken to use these products well away from occupied structures to prevent egress of phosphine into the buildings. Tragic deaths have also resulted from spraying metal phosphide rodenticides with water in an effort to wash them away, with liberation of large amounts of phosphine gas.1 Zinc phosphide, which does not release phosphine as readily from ambient humidity, is also used as a bait for rodents – ingestion is quickly followed by phosphine formation.
Illicit methamphetamine production
Several methods are used for the illicit production of methamphetamine from ephedrine/pseudoephedrine. Methamphetamine synthesis requires reduction of the alcohol group on ephedrine. One means of reduction is the phosphorus-iodine method. In the final steps of synthesis, ephedrine, red phosphorus and iodine are placed in a round-bottom flask. Water is typically added, although the reaction will proceed without it.
In the round-bottom flask, red phosphorus mixes with iodine to form hydroiodic acid (HI) and phosphorous acid, as below.
Hydroiodic acid, then, reduces the alcohol group in ephedrine to form methamphetamine.
You may ask, wouldn’t it be a lot simpler to just acquire hydroiodic acid? The US Drug Enforcement Administration has tightly regulated hydroiodic acid since 1993, making it difficult to obtain.
As long as the temperature in the flask does not become too elevated, ephedrine will undergo reduction to methamphetamine without much phosphine formation. But two bad things can happen during refluxing of heated iodine, phosphorus and ephedrine. First, recall those red phosphorus particles on the side of the glass as the contents are mixed in figure 13. If they are allowed to dry and overheat, the red phosphorus will convert to white (yellow) phosphorus, which will spontaneously ignite in air. Historically, this has been the second most common cause of illicit drug lab fires in Arizona.
Second, phosphorus acid that was produced as a byproduct in the synthesis of HI (Figure 14), when overheated, will be converted to phosphine, just as occurs in commercial phosphine production.2–4
Cellular actions of phosphine
The cellular mechanism by which phosphine causes toxicity is not completely understood.5 One action is that phosphine inhibits complex IV (cytochrome oxidase) of the electron transport chain in the mitochondrial inner membrane, thus decreasing electron transport, oxygen consumption and oxidative phosphorylation. As we have discussed in previous posts on cyanomythology, microvesicular steatosis, and the origins of protons in lactic acidosis, electrons from the iron of cytochrome c on the external surface of the mitochondrial inner membrane are first shuttled to a copper atom in complex IV. As shown in Figure 17, electrons then move onto a second copper atom before shuttling to an iron in the heme moiety of cytochrome a. From cytochrome a, electrons move to the binuclear center of cytochrome a3, made up of an iron and another copper atom, and then, finally, combine with oxygen to form water. Most complex IV inhibitors medical toxicology fellows study work by binding, in their undissociated form, to the binuclear center of cytochrome a3. Spectral absorption studies suggest that phosphine acts to keep the iron in cytochrome a reduced, impairing electron transport within complex IV, though some lesser activity at the binuclear center cannot be complete excluded.6
However, inhibition of complex IV seems unlikely to explain much or most of phosphine’s toxicity.7 Phosphine inhibits complex IV much more in in vitro than in vivo. It is also difficult to extrapolate from mechanisms of toxicity in insects which encounter low phosphine concentrations over hours to weeks compared to acute phosphine poisoning in mammals. Humans with phosphine poisoning do demonstrate decreased complex IV activity, but such inhibition is similar in those who die or live, and is somewhat similar to patients in shock from other causes.8 Cholinesterase inhibition is found in vitro, but there is no convincing evidence of a cholinergic crisis or meaningful depression of cholinesterase activity in poisoned humans. Thus, other mechanisms for phosphine toxicity have been sought, with examples shown in Figure 18. But oxidant stress has gained lots of attention and will be addressed next.
We noted in the post on microvesicular steatosis that when electron transport is inhibited, NADH will transfer electrons onto oxygen to form superoxide (O2–) rather than onto complex I. And electrons can also move from other members of the transport chain (e.g., ubiquinone, complex III) onto oxygen to form superoxide, as well. Superoxide (O2–) is a substrate for mitochondrial superoxide dismutase, which coverts O2– to H2O2, another reactive oxygen species. Phosphine increases concentrations of reactive oxygen species, and high and sustained concentrations of H2O2 and other species produce various damaging effects, including peroxidation of lipids in the mitochondrial inner membrane, causing further mitochondrial insult and cell death. The oxidant stress may be accentuated by phosphine’s inhibition of catalase and peroxidase.5 But there is probably more to phosphine toxicity than our limited understanding of its action on electron transport and oxidant stress.
Phosphine toxicity results from inhalation of phosphine gas or ingestion of metal phosphide.9 Death has resulted from inhalation of phosphine liberated from metal phosphides in buildings, on ships, and elsewhere. Of particular concern in shipping is leakage of phosphine from “sealed” grain holds into other areas of the ship from defects in seals or in the walls of holds, themselves.10,11 Stow-away passengers hiding in holds have been found dead from phosphine toxicity when holds were opened for unloading, usually about a day prior to reaching port. While inhalation of phosphine released from cylinders in fumigation operations or from metal phosphides reacting with water can produce lethal poisoning, neither Eugene nor I am aware of any fatality from inhalation of phosphine gas in the semiconductor industry over the last 50 years.
In general, phosphine produces a relatively rapid onset of symptoms with metabolic acidosis and death, mainly from cardiac failure. Pulmonary edema is commonly present in those who die, though it is not always clear how much is cardiogenic versus ARDS.
Ingestion of phosphide products quickly produces vomiting, epigastric and chest pain, hypotension, shock, cardiac arrhythmias and corrosive injury to the esophagus and stomach, sometimes with hematemesis.7,9 A garlic odor may be noted.12 Foaming about the mouth and throat can be seen. Impaired ventricular contractility is noted on echocardiography during times of shock. Self-combustion is possible from liberated phosphine. In one instance, moving the body of a deceased phosphine victim was accompanied by an explosive sound with flames exiting the mouth.13 Facial burns from vomiting of hot/smoking stomach contents (including activated charcoal) has occurred.14 Flames may appear at the ends of NG tubes.12,15
Liver injury is usually not severe in most patients who survive, but can be found at autopsy. Occasional oxidant hemolytic anemia is noted (especially in patients with G6PD deficiency), and methemoglobinemia is possible.7,16–18 Hypoglycemia has been reported. Other nonspecific findings include rhabdomyolysis, adrenal hemorrhage, and kidney failure from tubular necrosis.19 Tripathi and Pandy described brain congestion, edema, petechial hemorrhages and cerebral and cerebellar necrosis in 239 cases of phosphine poisoning.20 Liang and colleagues described similar brain changes in 8 children who died from phosphine inhalation as the gas spread from fumigated grain into buildings where they resided.17
In countries where ingestion of phosphides are common, diagnostic screening for phosphine involves placing gastric contents or exhaled air in contact with silver nitrate paper, which will turn dark from formation of metallic silver (Figure 19).21,22 However, sulfur compounds such as H2S will also produce a positive test.23 Additional screening of gastric contents with ammonium molybdate can confirm presence of phosphine.23
Commonly performed procedures following ingestion of phosphides include gastric lavage with potassium permanganate and administration of activated charcoal, but their effectiveness is unproved.7 ECMO has been used as a supportive measure with survival. There are no roles for cyanide antidotes at present.24
Some standard occupational medicine sources list phosphine inhalation as a cause of delayed-onset pulmonary edema. This may be somewhat true in patients who are already symptomatic from phosphine/phosphide poisoning, but I can’t locate convincing reports of asymptomatic patients who developed ARDS, alone, minutes to hours after an acute phosphine exposure (unlike phosgene). I’ve evaluated quite a few semiconductor workers over the years who have accidentally inhaled phosphine and have never seen any become ill as long as they were never symptomatic (usually just smelled the garlic odor).
We’ve seen that many metal hydride gases are quite toxic and/or combustible. What do emergency response crews do when confronted with a leaking cylinder of a metal hydride gas? Frequently the solution is to place the cylinder into a specially constructed emergency response containment vessel (ERCV), as shown below.
These ERCVs are commonly found at facilities that manufacture, store or use cylinders of metal hydrides as well as other toxic gases, such as boron trichloride, hydrogen, chlorine, and ammonia. A full-sized cylinder of a metal hydride is placed in the ERCV and then the vessel door is sealed closed. The leaking cylinder is then safely transported to a site that is setup for safe disposal of the gas.
After the door is closed the nitrogen from the small cylinder on top is flowed into the ERCV through the rear valve and out the front valve. The ERCV must be purged of air since phosphine is pyrophoric and the leak could create an explosive mixture inside. As a final step, the nitrogen pressure is allowed to build up so that the ERCV can be leak-checked prior to transport.
We’ve moved through metal hydrides now with the exception of the hydrides of boron.
So, next time we will finish up this series with boron hydrides, including, the green dragon. Fellows, please leave brief feedback and comments so we can be sure we are providing information that is of value to those of you in training. Thanks.
Medical toxicology fellows, here are some items for consideration and discussion.
- Reduction of the alcohol group in ephedrine results in formation of methamphetamine. What substance results from oxidation of that same alcohol group?
- It’s always best to have a differential diagnosis of human self-ignition apart from spontaneous human combustion. We’ve seen that ingestion of metal phosphides can result in self-ignition of phosphine as the gas mixes with oxygen in air and exhaled breath while exiting the mouth and nose. The ingestion of what other substance can produce smoking and flaming stools?
- Darkening of silver nitrate paper can occur when it comes in contact with phosphine in exhaled breath or gastric contents. What substance on the breath of a comatose victim will darken lead acetate paper?
- Don’t confuse phosphine and phosgene. We discussed how phosphine really doesn’t produce delayed-onset pulmonary edema in the absence of being symptomatic with moderate to severe toxicity. But inhalation of phosgene gas certainly can. What is the chemical formula for phosgene and what odor does it possess?
- 1.Akhtar S, Rehman A, Bano S, Haque A. Accidental phosphine gas poisoning with fatal myocardial dysfunction in two families. J Coll Physicians Surg Pak. 2015;25(5):378-379. doi:05.2015/JCPSP.378379
- 2.Wright J, Edwards J, Walker S. Exposures associated with clandestine methamphetamine drug laboratories in Australia. Rev Environ Health. 2016;31(3):329-352. doi:10.1515/reveh-2016-0017
- 3.Burgess J. Phosphine exposure from a methamphetamine laboratory investigation. J Toxicol Clin Toxicol. 2001;39(2):165-168. doi:10.1081/clt-100103833
- 4.Willers-Russo L. Three fatalities involving phosphine gas, produced as a result of methamphetamine manufacturing. J Forensic Sci. 1999;44(3):647-652. https://www.ncbi.nlm.nih.gov/pubmed/10408124
- 5.Anand R, Binukumar B, Gill K. Aluminum phosphide poisoning: an unsolved riddle. J Appl Toxicol. 2011;31(6):499-505. doi:10.1002/jat.1692
- 6.Kashi KP, Chefurka W. The effect of phosphine on the absorption and circular dichroic spectra of cytochrome c and cytochrome oxidase. Pesticide Biochemistry and Physiology. Published online August 1976:350-362. doi:10.1016/0048-3575(76)90045-6
- 7.Proudfoot A. Aluminium and zinc phosphide poisoning. Clin Toxicol (Phila). 2009;47(2):89-100. doi:10.1080/15563650802520675
- 8.Singh S, Bhalla A, Verma S, Kaur A, Gill K. Cytochrome-c oxidase inhibition in 26 aluminum phosphide poisoned patients. Clin Toxicol (Phila). 2006;44(2):155-158. doi:10.1080/15563650500514467
- 9.Misra U, Tripathi A, Pandey R, Bhargwa B. Acute phosphine poisoning following ingestion of aluminium phosphide. Hum Toxicol. 1988;7(4):343-345. doi:10.1177/096032718800700408
- 10.Loddé B, Lucas D, Letort J, Jegaden D, Pougnet R, Dewitte J. Acute phosphine poisoning on board a bulk carrier: analysis of factors leading to a fatal case. J Occup Med Toxicol. 2015;10:10. doi:10.1186/s12995-015-0050-0
- 11.Crawford J. Fumigation of cargo on board ships: the invisible killer. Presentica. Accessed November 14, 2021. https://www.presentica.com/doc/10991685/fumigation-of-cargo-on-board-ships-the-invisible-killer-pdf-document
- 12.Wahab A, Rabbani M, Wahab S, Khan R. Spontaneous self-ignition in a case of acute aluminium phosphide poisoning. Am J Emerg Med. 2009;27(6):752.e5-6. doi:10.1016/j.ajem.2008.09.045
- 13.Yadav J, Athawal B, Dubey B, Yadav V. Spontaneous ignition in case of celphos poisoning. Am J Forensic Med Pathol. 2007;28(4):353-355. doi:10.1097/PAF.0b013e31815b4bed
- 14.Mirakbari S. Hot charcoal vomitus in aluminum phosphide poisoning – A case report of internal thermal reaction in aluminum phosphide poisoning and review of literature. Indian J Anaesth. 2015;59(7):433-436. doi:10.4103/0019-5049.160952
- 15.Shadnia S, Soltaninejad K. Spontaneous ignition due to intentional acute aluminum phosphide poisoning. J Emerg Med. 2011;40(2):179-181. doi:10.1016/j.jemermed.2009.05.028
- 16.Soltaninejad K, Nelson L, Khodakarim N, Dadvar Z, Shadnia S. Unusual complication of aluminum phosphide poisoning: Development of hemolysis and methemoglobinemia and its successful treatment. Indian J Crit Care Med. 2011;15(2):117-119. doi:10.4103/0972-5229.83021
- 17.Liang Y, Tong F, Huang F, et al. Pathological changes induced by phosphine poisoning: a study on 8 children. Int J Legal Med. 2020;134(1):217-228. doi:10.1007/s00414-019-02169-z
- 18.Mostafazadeh B, Pajoumand A, Farzaneh E, Aghabiklooei A, Rasouli M. Blood levels of methemoglobin in patients with aluminum phosphide poisoning and its correlation with patient’s outcome. J Med Toxicol. 2011;7(1):40-43. doi:10.1007/s13181-010-0121-7
- 19.Yogendranathan N, Herath H, Sivasundaram T, Constantine R, Kulatunga A. A case report of zinc phosphide poisoning: complicated by acute renal failure and tubulo interstitial nephritis. BMC Pharmacol Toxicol. 2017;18(1):37. doi:10.1186/s40360-017-0144-7
- 20.Tripathi S, Pandey S. The effect of aluminium phosphide on the human brain: a histological study. Med Sci Law. 2007;47(2):141-146. doi:10.1258/rsmmsl.47.2.141
- 21.Etemadi-Aleagha A, Akhgari M, Iravani F. Aluminum Phosphide Poisoning-Related Deaths in Tehran, Iran, 2006 to 2013. Medicine (Baltimore). 2015;94(38):e1637. doi:10.1097/MD.0000000000001637
- 22.Chugh S, Ram S, Chugh K, Malhotra K. Spot diagnosis of aluminium phosphide ingestion: an application of a simple test. J Assoc Physicians India. 1989;37(3):219-220. https://www.ncbi.nlm.nih.gov/pubmed/2768165
- 23.SKRai P. IMPORTANCE OF AMMONIUM MOLYBDATE TEST IN ALUMINIUM PHOSPHIDE POISONING: A CASE STUDY. IJAR. Published online September 30, 2019:01-04. doi:10.21474/ijar01/9622
- 24.Mehrpour O, Amouzeshi A, Dadpour B, et al. Successful treatment of cardiogenic shock with an intraaortic balloon pump following aluminium phosphide poisoning. Arh Hig Rada Toksikol. 2014;65(1):121-126. doi:10.2478/10004-1254-65-2014-2393