On Carbon Monoxide Poisoning

M. L. Levontin

From the Ural Regional Institute of Hygiene and Occupational Diseases

Received by the editors April 29, 1932

Alongside the enormous growth of the workforce employed in Ural industry, we have seen some increase in the absolute number of registered occupational poisonings and diseases, including carbon monoxide poisoning. Thus, in 1928–29, of 1,250 registered cases of occupational poisoning, 676, or more than 50%, were due to CO poisoning; in 1931, of 1,859 registered cases, 7,34, or 40% of all poisonings, were due to CO.

To combat carbon monoxide poisoning successfully in the metallurgical industry of the Urals, it is therefore necessary to develop and implement a well-defined program of measures.

At a number of enterprises, CO poisonings, especially cases with less pronounced clinical manifestations, are not entered in the mandatory register. Medical personnel, particularly mid-level staff on duty, who in most cases are the first to observe the patient and provide first aid, are sometimes unable to distinguish CO poisoning from overheating. Deaths directly attributable to poisoning are observed, as are subsequent deaths following poisoning (that is, “late deaths”). Although poisonings evidently account for a relatively small proportion of sickness absences, they nevertheless disrupt production, causing disorganization in particular phases of operations. Especially consequential in this respect are “accidents”—cases of mass poisoning observed at two enterprises in 1931. These so-called “accidents” are not chance occurrences but arise from the general organization of the gas system. They are therefore highly instructive for developing a program to combat occupational poisoning. Such cases should be analyzed and published, both to accumulate material for sharing experience and to provide a basis for discussion at local meetings of engineering and technical personnel and active workers.

With the current reconstruction of plants, when blast-furnace and producer gas are used almost entirely for industrial purposes (in open-hearth furnaces, boiler houses, foundries, and so forth), carbon monoxide poisoning cannot be tackled separately and in isolation in blast furnaces and open-hearth shops. The entire gas system must be viewed as a single whole, a single system with numerous branches. The prevention of poisoning must be addressed within the gas department’s system. This makes it possible to establish a unified organization for managing, supervising, and repairing the entire gas system, regardless of where the gas is produced or which shops and installations receive it. It makes it possible to combat the lack of clearly assigned responsibility for supervising the unified gas network, parts of which are distributed among different shops. Unified organization should encourage specialization and the accumulation of experience among gas-network personnel. At present, the head of any shop supplied with gas must be a specialist in that shop’s field and at the same time have experience in supervising hazardous gas systems and in the techniques used to control the hazard.

At the end of January of this year, we had occasion to investigate a mass carbon monoxide poisoning as part of the institute’s planned work on combating occupational CO poisoning.

The circumstances and conditions of the poisoning may be of interest to those concerned with improving working conditions.

On January 21, 1932, during the morning shift at TZ, two groups of workers were cleaning the gas ducts of the open-hearth shop: one group of three tar workers was cleaning the flue (see diagram), and another group of three gas-producer stokers was cleaning the producer-gas pipeline. They had been assigned to the work by the foreman of the open-hearth shop. The first group spent 10–15 minutes in the basement near the flue while an electrician installed lamps for lighting. One worker then descended into the flue through a manhole that had been open the entire time. The flue lies beneath the floor of the building; it is approximately 700 mm wide and up to 1,280 mm high, and its bottom is covered with tar over most of its length. Immediately after descending, the worker gave a muffled cry. The second worker lay down on the ground above the manhole in order to—

…help his fellow worker, and remained lying with his head and arm lowered into the flue access hatch. A third man, standing about 1.5 m from the flue, collapsed unconscious at the same time. Meanwhile, the entire second group had descended through an access hatch into a horizontal pipeline. All began to feel ill. One managed to climb out through the hatch and, sitting on the pipeline beside it, called for help. A blacksmith who happened to come out of the smithy, located in the room where the first group was working, heard the cries for help and raised the alarm. Workers rushed over to rescue the two men remaining in the gas pipeline. Returning to the smithy, the same worker noticed the two tar workers from the first group lying there and ran out onto the open-hearth working platform to call for help; more fellow workers came running to assist. In addition to the six men being rescued, another 10 rescuers were poisoned. The rescue was carried out with bare hands, without even the most rudimentary equipment. Almost all were comatose. Medical staff from the first-aid station, who had been summoned at the same time, provided care first on the spot and then at the station, after which the victims were taken to hospital. There, two members of the first group died without regaining consciousness. The first group consisted of older, highly experienced tar workers, aged 56–62; the second consisted of young men aged 17, 20 and 23.

The circumstances of the poisoning were characterized by: (a) discovery by chance; (b) a lack of supporting arrangements at the hazardous workplace; and (c) the absence of basic rescue equipment at hazardous workplaces. As a result, many rescuers were poisoned.

The “incident” described above was preceded by the following circumstances of a sanitary and technical nature. On 17 January, one of the two open-hearth furnaces, furnace No. 2, was shut down for a major overhaul. Two gas lines lead to the furnace: one carries blast-furnace gas, the other producer gas made from charcoal. The pipelines are above ground. Producer gas is supplied from separate gas producers for each of the two open-hearth furnaces through separate pipelines. These pipelines, however, have a connection isolated by a blind flange. The connection was installed to allow both open-hearth furnaces to use a common coal-gas producer shaft. The latter is not in operation, and the addition of coal gas to the charcoal gas has been discontinued. The installation, however, and most importantly the connection, remain. The two types of gas are mixed at one end of the flue (diagram—1). The second open-hearth furnace (1) is served by a blast-furnace gas pipeline, as a separate branch of the main pipeline, and a producer-gas pipeline from separate producer furnaces, as noted above. Each open-hearth furnace has its own chimney and its own flue, with a gas-distribution bell above it (diagram—5, 1, 4).

Before demolition of the open-hearth furnace (17, 18, 19, 1), the following measures were taken: (a) closure of the shutoff valve on the blast-furnace gas pipeline; (b) filling the throttle valve with material—three wheelbarrow loads of ground chloride iron ore were poured in (diagram—7); (c) closure of the producer-gas valve (diagram—8); (d) blowing out the gas producers; and (e) setting the distributors of the gas bell above the flue to the central position, that is, for draft from the flue into the chimney.

In October of the previous year, the same furnace had undergone a cold repair. The same cleaning work had been carried out with the same precautions, but no cases of poisoning…

…had occurred. In November and December of the previous year, the same cleaning work had been carried out during a one-day shutdown of the furnace while still hot, and no poisoning had occurred then either. This was according to the shop management. The management’s account, in view of the poisoning cases that had occurred under supposedly identical conditions, underscores once again the danger of work around gas installations and the need for thoroughly considered precautions.

After the poisonings, the following additional precautions were taken before cleaning resumed: 1) a blanking plate was installed upstream of the riser; 2) a pipe was connected to the horizontal section of the gas line (diagram—10) to link it to the fan in the smithy (used for the forge), and the flue was blown through toward the chimney; 3) an additional blanking plate was inserted where the two producer-gas pipes joined the coal gas-producer shaft; 4) moreover, with a team from the institute now present, preliminary air analyses were conducted, attendants equipped with a Dräger apparatus were posted, and so forth.

All this shows that it is not enough to confine the system of precautions to a standard based on past experience. In the specific conditions of each workplace, every possibility must be explored and mobilized to strengthen the measures and make them fully comprehensive. The possibility of poisoning in gas-handling operations must be regarded as an extreme danger requiring every safeguard.

Let us consider the immediate cause of the poisonings and whether the measures taken provided adequate safeguards. As is well known, merely closing the shutoff valve on the blast-furnace gas line (diagram—6), without covering it with a layer of sand of the prescribed thickness, is unacceptable and insufficient. One cannot rely entirely on the valve fitting tightly against its seat and providing a complete seal. Covering the throttle valve with chromite ore had no effect. An inspection immediately after the accident showed that the fill material was beneath the valve. This could have been foreseen, since the throttle valve was inclined, and hence in unstable equilibrium, so that it was bound to shift when the material was poured in. Closing the producer-gas valve when the gas-producer furnaces are shut down and blowing out the gas producers would be an effective measure were it not for the connection between the producer-gas pipes of both furnaces at the coal gas-producer shaft. In essence, this connection is a superfluous addition, an unnecessary remnant. Reliance on a draft carrying the residual gas from the flue into the chimney could not be justified in the absence of a thermal or mechanical driving force. Thus, the poisonings and the seepage of gas were caused by the ineffectiveness of the isolation and sealing measures taken: the supposed “seal” was not real. Sealing remains a formality unless it is tested, for example by illumination, water, and so forth.

It should be noted that the same cleaning of the producer-gas lines and flue had been performed the day before, January 20, without incident. This must be explained by the fact that freezing temperatures had prevailed through January 20, followed by a thaw beginning on January 21. Condensate could have frozen and sealed, like putty, the small gaps and openings around the poorly fitting valve in the blast-furnace gas line. With the onset of the thaw, the gaps opened, and gas under pressure in the main blast-furnace gas line seeped into the downstream sections of the system.

The question of rational, fail-safe isolation and sealing systems is thus a very serious one from the standpoint of occupational hygiene. Of the various ways of interrupting the flow of gas, two appear to be more rational: a water seal and partial mechanical removal of a section of the line to cut off the gas.

The first principle is shown schematically in the figure. In the Urals, the water is naturally supplied heated in winter. Section “б,” after blanking plates have been installed at “а” and “а₁,” is removed when isolation is necessary, as for repairs or cleaning. The latter method is advisable for pipelines exposed to the open air.

We must also consider certain conditions we found in the organization of work on the gas pipelines. The overhead pipelines were put into operation without working platforms or ladders. Access to them is therefore extremely difficult. This was partly why the valve on the blast-furnace gas pipeline was not covered with sand. The absence of this equipment made it difficult to climb up to the horizontal producer-gas pipeline to assist the victims and to bring them down. The advisability of placing the cleaning hatch on the upper surface of the producer-gas pipes mentioned above should be questioned, since this greatly impedes the escape of a worker who feels weak and shows signs of carbon monoxide poisoning. The reason given for placing the hatch on top rather than on the side, where entry into and exit from the pipe would be more convenient, is that it is safer and offers less resistance to the escape of gas in the event of an explosion. If this is so, the question should be studied with a view to meeting hygiene requirements. Another important local feature is the location of the flues serving both open-hearth furnaces and the four pipelines leading to them. They are located in a dark basement beneath the furnaces’ working platform; the area where the pipes intersect is cluttered with unnecessary objects (the smithy is also located in this room). There are no approaches, walkways, railings, or similar aids to movement. There is no proper staircase into the room itself, only a makeshift passage made of unsteady boards. The entire room is a piece of the “Demidov” legacy of the plant’s unfinished reconstruction.

Without dwelling on a number of matters that were significant in the case described—such as the lack of a clear allocation of responsibilities and of effective supervision of dangerous work by those responsible for it (plant-wide and shop safety personnel); the absence of personnel on duty; the failure to give advance notice or summon the medical station’s duty staff; the lack of equipment such as ropes and belts; and the failure to use all available means of providing medical aid—we shall draw several conclusions:

1. The organization of management, supervision, and repair must be unified across the entire gas system, and a gas department must be established with personnel of all the necessary qualifications.

2. The gas department should include a rescue unit equipped with all the equipment and apparatus needed for rescue work. Its personnel should be trained in simple methods of determining carbon monoxide concentration, both chemical and biological.

3. There is a pressing need to develop a rational design for gas-tight sealing.

4. Persons over 50 years of age should not be permitted to work in hazardous areas of gas pipelines, cleaning or repairing them.

5. Instructions should be issued for medical-station physicians on selecting personnel for work in the gas system and on the use of rescue apparatus.