There are some questions about whether the SARS-CoV-2 virus is more or less deadly than the 1918 influenza virus. It’s not really possible to accurately compare the two pandemics’ case and fatality data, for one very big reason. Oxygen!

Today, when someone has any sort of respiratory problem the first likely action is to provide supplementary oxygen. The air we breathe contains about 21% oxygen, which for a healthy person is obviously quite adequate. But for sick people, raising the concentration to near 100% can take the load off their heart and lungs and help prevent other problems like organ failure.

In the 1918 pandemic, the use of supplemental oxygen was not widely known or accepted (as illustrated in the picture, where no one is getting oxygen). There had been some experimentation and use in prior decades (and in treating some chemical weapons victims of the First World War), but it was not yet widespread. Even if it had been enthusiastically embraced, the supplies of oxygen were very limited and it was not readily available on an industrial scale. Therefore we read stories of even young and previously healthy people succumbing to the influenza virus within hours, turning blue due to lack of oxygen in their bloodstream. If oxygen had been available, many of them may have had a reasonable chance to survive and recover. Of course, today we have many other pharmaceutical interventions like steroids and monoclonal antibodies, none of which were available in 1918. But the oxygen is still needed to keep the patient alive long enough so that the pharmaceuticals can have a chance at working.

Where do we get the medical oxygen? There are smaller scale purifying units that take air and concentrate the oxygen in it using membranes to separate the oxygen from nitrogen. These are fine for portable use, smaller scales, or lower flows, but they are not readily scalable to the huge volumes required in a hospital and high-flow oxygen therapy. Likewise, oxygen cylinders are not very practical in hospitals due to their limited capacity. For example, the large “T” size cylinders (about 24 cm diameter by 130 cm tall) only contain about 9,000 L of oxygen once it is depressurized. For patients needing high flow oxygen therapy at up to 60 L per minute, the cylinder would only last about 3 hours or less. If there are a lot of patients, it would take a small army of people constantly moving cylinders in and out of rooms and the hospital. Unfortunately, in poorer and less industrialized parts of the world these options are often the only ones available.

The big industrial oxygen supplies are typically provided in the form of liquid oxygen, shipped and stored in specialized trucks and tanks. The liquid oxygen stored at the hospital is then vapourized into its gaseous form and piped to the patient rooms as required. This is a much more compact and efficient delivery system, since one hundred litres of liquid oxygen expands into about 85,000 L of gaseous oxygen for breathing purposes.

The large scale production and supply of liquid oxygen is a chemical and mechanical engineering accomplishment dating back to the early 1900’s. It took several decades for many plants to be built, with continuous improvements over the years to improve the process and reduce energy requirements. The industrial process uses distillation to separate oxygen from nitrogen (and argon and other trace gases) in air. Since oxygen and nitrogen have quite different boiling points (-183^oC for oxygen, and -195.8^oC for nitrogen), separation by distillation is a reasonably straightforward approach. However, distillation requires that air be liquified through a combination of pressure and low temperature, and this presents some significant engineering challenges.

Modern plants, often called “Air Separation Units” or ASUs, operate at pressures up to about 6 atmospheres and temperatures in the -170 to -190 range. Clearly it takes some significant compression and refrigeration equipment to carry this out, and the plants are carefully designed to be as energy efficient as possible. The video below, from one major manufacturer, gives a simple overview of the ASU process. Of course, ASUs are built not only for medical oxygen, but also for the many other industrial uses of oxygen such as in steel production, metal cutting, water treatment and chemicals manufacturing.

Bibliography:
Graninge, C. Breath of life: the evolution of oxygen therapy. J.R. Soc. Med. 97: 489-493 (2004).
Heffner, J.E. The Story of Oxygen. Respiratory Care, 58: 18-31 (2013).
Tellier, N. Air Separation and Liquifaction. (https://cryogenicsociety.org/resources/cryo_central/air_separation_and_liquefaction/)