Let’s start with a quick video from May 14th (mostly in English):
Oxygen is all around us
As everybody knows, supplemental oxygen is a main treatment for covid-19 pneumonia, and it saves lives. In May and into June, there have been many reports coming from Low Income Countries in Asia about the scramble to obtain cylinders of the oxygen, and the usual horrifying outcome was that the sick person often died before it could be obtained. Particularly in Nepal, this led to a rush of international donors sending oxygen-related equipment to fill the gap.
Get your shot
The best therapy for covid 19 of course, is to vaccinate. This diary will not address the issues of vaccine supply in Low Income Countries, but we all need to acknowledge that it is an overarching issue.
Since 2007 I have travelled extensively in Nepal to teach critical care skills to doctors and nurses, and long-time readers of DKOS will recall that I write about it here especially when I think there are parallels to be drawn between their health system and ours.
Back to Oxygen
On May 31st I published a DKOS diary about Medical Grade Oxygen in Low Income Countries and this is a followup. www.dailykos.com/… If you click on that link, be sure to read the comments; I was particularly impressed with several commenters who, it turned out, were experts in setup of oxygen manufacturing plants including the cost, startup time, and alternative systems to pull the stuff out of the atmosphere. ( oxygen already exists it’s not “manufactures” per se but separated out. Most of earth’s atmophere is Nitrogen).
There were two comments by laniusX that added a lot to the conversation:
Those PSA plants are fairly small volume, but they can be decent for small users and point-of-use consumption. But they’re not practical or cost effective for large scale production. They’re also relatively inefficient. They’re really designed for small local cylinder sellers. To use them as on-site supply, you’d need to have one at each hospital. And with unreliable electrical supply, that wouldn’t work very well without back-up O2 supply.
The more practical solution is to have a big cryogenic plant and distribute the O2 either in cylinders or as liquid (preferably the latter). The problem with that, of course, is it takes 18 months to 2 years to build one. That doesn’t work as a response to the current crisis. If India were not in such dire straits, I suppose they could possibly ship O2 from there, but with their situation, I can’t see them sharing.
FWIW, I worked for Air Products for 30+ years and designed cryogenic oxygen plants for 13 of them.
When I read that I queried as to the cost and he wrote:
I’ll try my best, but my expertise is the cryogenic side and less so on the non-cryo side, but I’ll try to answer your questions.
1) From what I can see online, a small PSA unit is probably on the order of a few hundred thousand dollars. But then you have to include installation, tie-ins to the utilities, etc., so let’s say a million or so.
But the ongoing operating costs are much higher. A PSA system at best recovers about 33% of the O2 in air. A big cryogenic system recovers over 99% of the O2 in the air. That says that a PSA plant uses 3X the power as a cryogenic plant. So while your capital investment is low, your ongoing costs are much higher.
Also, PSA technology is just not practical for large plants. They’re probably limited to 100 Ton/day of O2. Modern cryogenic plants which make liquid O2 can make 600 T/D or more. They also make a lot of liquid N2 which helps defray the cost of the plant. Modern gaseous O2 plants can be as large as 3000 T/D. Those are actually limited by the size of the equipment that can be shipped.
Also remember, on a PSA plant you’re compressing and then throwing away 79% of the air. A big cryogenic plant will recover mush of the N2 and Ar which helps with the unit cost of product.
2) There certainly are some hazards. Pretty much anything will burn in pure O2 including metal. And those tanks probably also include liquid N2 also, which is a hazardous asphyxiant. There may also be liquid Helium tanks for the MRI units which is EXPENSIVE and an asphyxiant. That’s why they tend to be fenced off — just so people can’t screw with them. But part of the reason why those tanks are separated is for access. The tanks are typically owned by the O2 supplier, not the hospital. The supplier needs to get a semi-truck close to the tanks to fill them. And depending on how much they use, that could mean 2 trucks per week of each material.
3) It’s been a while since I was in that business, but 20 years ago a 750 T/D O2 plant which made both gaseous and liquid O2, gaseous and liquid N2, and liquid argon was about 25 million dollars. If you don’t have the gas customer, and don’t have the argon recovery the cost goes down since the gas production requires extra compressors and instantaneous back-up systems. Let’s say a large modern liquid O2/N2 plant is about 15 million $. But that will make 1200 tons/per day of liquid products. And remember, the power cost per ton (essentially the only significant operating cost) is much, much less.
4) A typical industrial plant makes 99.5% O2 or better. Medical grade is 99% or better. So inherently an industrial plant makes the same quality. In fact, the plants at Air Products that supplied hospitals were the same ones that supplied industrial users. The difference is FDA certification. That requires extra paperwork, extra analyses, and extra risk (if something happens at one customer, the FDA could, in theory, shut down your entire plant until the issue is resolved and root cause determined). So at AP, only certain plants were certified for FDA use. It wasn’t worth the risk to certify them all.
Hope this helps!
This week two articles appeared in Nepali media. The first was about plans to build a liquid oxygen plant in Bhairawaha. It gave more detail to the overall landscape of supply and demand. I did not know until now, that there is not one plant in Nepal that uses the liquid oxygen method to generate the gas, and the only source is India.
Political Economy of Oxygen
The second article appeared in Nepali Times. This one focused on the market issues and is titled “Pandemic and the political-economy of oxygen supply” I highly recommend it and you need to read the whole thing since all of it is enlightening. The journalist did a good job of finding out why and how this happened. Here’s just a short quote:
Families with Covid-19 patients hoarded full cylinders at home, and desperate ones rushed to the gates of oxygen factories to buy the life-saving gas. Municipalities, especially in areas with spare capacity, prevented the transport of oxygen beyond district borders.
Most of Nepal’s oxygen plants are in central and eastern Nepal. The Morang-Sunsari industrial corridor alone had five large oxygen factories with surplus capacity, but these were prevented from sending supplies to hospitals in Kathmandu and other areas with shortages.
Lack of training in oxygen flow management and ventilator use did not help matters. Makeshift local isolation centres with non-medical staff were using full flow to treat patients with mild symptoms. Hoarding and black marketing was rampant, and while large hospitals ran out of oxygen supply, local isolation centres had one cylinder per patient.
There are many other contributing factors to the way that Nepal will get through the pandemic, but I think this helps the reader to consider the systemic issues involved.