Commodity chemicals are often sold with very small profit margins. Yield is very important.

There are several similar process technologies to make an intermediate which is converted in a second process to a large volume commodity chemical. The major yield loss is a high molecular weight mixture stream from the distillation train, which is about 0.9 mass percent of the total feed.

A new technology was developed in a pilot plant with a 3″ diameter reactor. This heavy byproduct stream was sampled and analyzed regularly. The rest was collected in a vessel and weighed to determine that stream’s contribution to the yield loss. One of the benefits of this technology was that this byproduct was only about 0.6 mass percent of the total feeds.

I was on the design team of the first commercial plant, a large scale up from the pilot plant. The reactor was several meters diameter, with plant capacity proportional to reactor cross sectional area. Sadly, the commercial plant stream was about 0.9 mass percent of the total feed. The yield benefit had disappeared in scale up.

After much searching it was found that the pilot plant samples of that stream were about 0.3 mass percent of the total feeds. Taken upstream of the collection vessel, the samples had not been included when calculating the pilot plant yield. When they were included the yield loss was about 0.9 mass percent.

I told this story to a good friend who worked with me on many projects. He said that he had the identical experience scaling up a different process at a previous employer.

The moral of the story is to consider all the “ins” and “outs” when doing a mass balance.