Wednesday, February 16, 2011

Process Wednesday: operational qualification

Validation is an aspect of working in a manufacturing facility that's a little foreign to a novice industrial chemist. Our Process Wednesday mentor-by-literature, Neal Anderson, writes in the 1st issue of Organic Process Research and Development of 2011* about how to validate a process (defined by Anderson et al. as "the cumulative efforts to demonstrate reliable process and product quality.)

In part of that article, Anderson covers the steps necessary to ensure that the equipment works. He calls these steps "design qualification", "installation qualification" and "operation qualification." It's the last that I wish to focus on for today's post, because it's so simple and yet so important. Operational qualification, Anderson writes, is "confirming that the equipment installed operates correctly. OQ can include checking that required temperature ranges and stirrer speeds can be achieved, that nitrogen and air are supplied suitably, or that materials can be transferred from one part of a system to another without leaks. Often water or solvent is used as a surrogate for process streams to demonstrate that the equipment can perform as required."

He then goes on to describe a few situations in which OQ was key:
With virgin manufacturing facilities OQ is critical. For example, prior to a manufacturing startup, unsuccessful attempts to transfer water from one reactor to another identified a plugged transfer line. The transfer pipe in question had been blanked off for effective welding during construction, and the blank had not been removed. [snip]

Similarly, to prevent condensers, including glass lab equipment, from plugging during reflux it is important to select coolant temperatures above the freezing points of solvents such as t-BuOH (mp 25°C) and dioxane (mp 12°C). The transfer of acetic acid (mp 16°C) from a tanker into a process facility building was stopped when the ambient temperature became unusually cool; the transfer line was lagged with heat tape to melt the plug of AcOH and complete the transfer. Often physicochemical characteristics have been overlooked.
It never ceases to amaze me how Mr. Murphy will hide away in a reactor or a pipe somewhere in your facility. It's the job of the process chemist to find him and chase him somewhere else.

Finally, I wish to thank all the people who contributed to make last week's cGMP post such a wonderful read. I am so grateful to everyone who commented, especially those who took the time to explain such an arcane subject to a willing audience.

*Anderson, N.G.; Burdick, D.C.; Reeve, M.M. "Current Practices of Process Validation for Drug Substances and Intermediates." Org. Process. Res. Dev. 2011, 15, 162-172.

2 comments:

  1. You are correct Validation is foreign to novice industrial chemists but this is back to cGMP discussion where even though as a chemist one "knows" a process works and how to run it it still has to be thoroughly demonstrated and documented within a specific framework. Even though still incumbent to run at facility and scale there has been a positive shift in past few years to providing support with sound process development and statistical experimentation rather than having to do everything on full scale.
    The FDA just published official guidance which has been out in draft for many years.
    http://www.fda.gov/downloads/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/UCM070336.pdf

    In terms of Andersons OQ examples while I agree such is important as part of validation I see these particular situations as big potential Safety issues and therefore needs to be foremost in Process/Manufacturing chemists mindset and which too can be a foreign concept to those straight out on academia.

    CMCguy

    ReplyDelete
  2. Thanks for the educational comment, CMCguy.

    And thanks also to reader XWP for bringing attention to the article.

    ReplyDelete

looks like Blogger doesn't work with anonymous comments from Chrome browsers at the moment - works in Microsoft Edge, or from Chrome with a Blogger account - sorry! CJ 3/21/20