A stubborn problem...

[bisnettrj2]

My stubborn problem is the guy that used to run all the HPLCs in my section. He's difficult to deal with, and he's resistant to change (to say the least). He still runs the LC/MS instrumentation (HPLCs and MSs). I've managed to pry all the HPLC-UV instruments from his control, but he's still holding on to the LC/MS instruments. The problem is, he's using old, abused columns (he doesn't even cap the ends after use!), he's not controlling temperature, and his gradients (excuse me, gradient) starts at 90:10 H2O:MeOH and he shoots 100% methanol samples. Also, he has his mass spec sources sitting anywhere from 3 to 6 linear feet away from the ends of his columns, connected by long lengths of SS tubing.

My question: How do I go about showing him that the way he is running his instruments is a.) insane; b.) inefficient;, and c.) exceptionally annoying?

[HW Mueller]

Well, your adjectives seem to fit all of us? Maybe the only adjectives about which you need to be cncerned is "incorrect", "false", . . . (regarding results).

[bisnettrj2]

True, and that's not my issue. However, it's frustrating that there are some very easy ways to improve his chromatography and increase his efficiency, and he's just not interested. He found one way to analyze, and he'll be damned if he's going to do anything differently, no matter what common sense or good business sense should be telling him.

[Bruce Hamilton]

This may come as a shock, but he was young once, and probably learnt from bitter experience that change sucks.

That doesn't imply that he's immune to change, just that advantages have to outweigh all of the disadvantages he's encountered in the past.

"If it ain't broke - don't fix it is " often the most effective and efficient method for routine analyses in situations when the resources are readily available and customers are satisfied.

If you want to convince him to change, don't start with the assumption he's a curmudgeon, but demonstrate to him that some incremental change doesn't impact on whatever he holds important in his process, and will improve the process. Assertions don't count, only real demonstrations will do it.

Also, he's a workmate, so ensure you don't cause animosity and distrust by telling others of your opinions. Workplace problems are often due to miscommunication, so try to avoid personalising issues.

I've encountered situations where analytical changes have been catastrophic - including causing the demise of a 50-year-old company because new instrumentation couldn't cope with a variable, dirty process, and the experienced staff had been made redundant and moved on.

Bruce Hamilton

[Uwe Neue]

Once upon a time, a long, long time ago, I hired some MS people because I had no clue about MS. When they set up their instruments, they did exactly as your colleague does. Of course, this was not the way I thought that this should be done. Rather than force them to change their ways, we discussed chromatography, plus discussed some experiments that would prove the point that one can do better. For example, we turned the MS around, reduced the tubing to 10 cm or so, and low and behold, the fast peaks were much narrower. I even got MS people to buy on to the idea that one can run a 2 mm column at 1 mL/min without post-column splitting.

Of course, you need to do all such things in your spare time, but if you can demonstrate once that your ideas have some merit, I am sure that he will listen to your ideas in the future. Plus, don’t forget to listen to him too. You might learn a thing or two as well.

[bisnettrj2]

Thank you, kind sages of the chromatography world.

Uwe - In line with your suggestions, I have been doing my own work at attempting to optimize his systems in my own time. In fact, I've been doing that since I posted this 'rant'. I really do hope I can constructively suggest some improvements to his systems, and from there perhaps convince him of other changes. However, I am limited in what I can demonstrate because of some spatial restrictions he has placed on his instrumentation (if I had a picture, I would post it, but I'll say this... I'm 6'4" tall, with a 6'8" wingspan, and I can barely vertically reach the sample tray on his Agilent 1100, and the MS/MS it is coupled to is sitting above my head. If I could explain why the instruments ended up in this arrangement, I would. All the people in my lab just scratch their heads and walk away.). This is the other part of the "the way he runs his instruments is insane" rant.

Bruce - "I've encountered situations where analytical changes have been catastrophic - including causing the demise of a 50-year-old company because new instrumentation couldn't cope with a variable, dirty process, and the experienced staff had been made redundant and moved on."

Bruce - thanks for scaring the hell out of me. However, if a company needed a small change to affect its demise, it was probably headed there anyway. In any case, he's not just a workmate, but the co-owner, and I really don't want to cause the demise of the company I work for... My wife might not be happy to hear we can't pay the mortgage anymore because I was picky about fundamentals.. I think, in this case, that yours and Uwe's advice is good - show him how change can be beneficial, and incrementally change his point of view.

I do understand the "If it ain't broke" code, but I have always advocated one caveat to that belief system - Sometimes you don't know it's broke until you're shown that it is, and until then you're an ostrich with its head buried in the sand. My motto - Always learn, always read, always question, always get better - "if you aren't growing, you're dying."

Incidentally, the motto for the few people in my section of the lab is "Cans of worms opened 24/7", in that we're seemingly the only people in the lab who look at regulations, requirements, and processes from a "worst-case" point of view, tear them apart, and then present those 'worst-case' points to the boss. So, we often discuss amongst ourselves the pros and cons of best practices versus "what works". It's a frustrating but constructive effort to become better at what we do.

I will take your collective advice on trying to constructively implement incremental changes. It's just very frustrating when you want to do it all at once, and instead you need to drag it out over a couple months or more like I probably have to do in order to effect the desired change...

Thanks for the suggestions and comments!

[HW Mueller]

If it is not "your issue" whether that guy is delivering wrong or correct data, then why are you concerned?
Also, I understand Bruce´s statement on "isn´t broke" as meaning, if it is working. One of my profs from student times used to say in his lectures: If you built a machine into which you can load manure (he used a 4 letter word which Tom will not allow me to use here), crank it up, and out comes gold, you will keep on turning the crank rather than stopping to find out how it works.
One time I asked someone who had just developed a HPLC analysis for catecholamines to run a few samples for me. I got suspicious and then gave him some standards, disguised as samples. They were way off. He didn´t get any more samples from me, I gave hints that there is a problem. As far as I could tell, the catecholamine analysis died there.

[lmh]

You have my sympathy: To be honest, I put it down to the long-term effects of acetonitrile fumes.

While many chromatographers are lovely (and highly professional) people, when I encounter someone in science who is rigid, unable to change any of his/her working practices, illogical, and determined to stick to what they know, even if the world has moved on in the last 30 years or so, I can bet my bottom dollar they are a chromatographer of some flavour.

Unfortunately I know of no solution. Elute your woes away (sensibly!), or investigate a large sample of chocolate; all you can do is do your job as best you can, and avoid feeling responsible for those bits of the world that are bad, but that you cannot reasonably hope to change. We all make our own little corner of the world better if we can...

[WK]

Hi,
I've been through the mix with colleagues who IMO didn't shift their own weight. Once the respective managers get involved - I wished I hadn't raised the issue. Very stressful. Looking forward to a less stressful holiday period! When I expire I can go with the knowledge I put some effort into all I do.
Sometimes - not in your case - it can appear that chromatographers are stuck in the mud a bit - in our case we have had no capital approved for a few years now - so with manual injections how am I to get 60 samples done in a day?! I have to do them the same way - and the only better way thought out involves capital expenditure!
Have fun!
WK

[Ron]

As the co-owner of the company your coworker has different information and a different perspective than you do. If taking the time to develop and validate new methods impacts productivity and profits short term and there does not appear to be a substantial long term gain there is no real incentive to do this from a business point of view. Any time methods are changed there can be unexpected consequences. I remember one company that wanted to move some of their QC checks to megabore columns from packed columns, and discovered with the higher resolution they found more impurities and no longer met their specifications for shipping product. They dropped the idea of changing the methods as the product worked perfectly fine as it was, and if they changed their QC methods they would have to change the process to produce a purer product or change the specifications to lower purity.

Have you tried just asking why some things are done the way they are and if alternate ways have been explored in the past? If you do this with the right attitude you may find out there are good reasons things are done they way they are, or at least at the time the reasons were good. If a good dialog is established the conversations can progress to suggestions on things to try to improve the work flow and quality of the analysis.

I'll bet with the autosampler tray around 8 feet above the floor nobody does anything with it that isn't necessary, and no one disturbs anything that is running while they walk by.

[lmh]

Ron, I think the case-study you give is not a good one to follow!

If the reason a company sticks to an old method is merely that it gives them the result they want to see, but they know that the result is actually wrong, then they are guilty of deceit. Their product is fraudulent, and they know it, and yet they still sell it. This sort of behaviour is utterly unacceptable.

[Ron]

The example I gave is really very common, it is a fairly general rule that as separation and analysis techniques get better that there are more impurities found in any given product. If the specification is for 99.85% purity and the product for years has routinely assayed at 99.86% pure, then the method is changed and the assay is now 99.84% pure is the problem with the product or the method of analysis?

If the product serves the intended purpose and has for years it is a good product for the intended use. If a new assay shows a slight increase in impurities that have always been there below the detection limit and have not had a negative effect there is no real benefit to changing the assay, changing your specifications, and then telling your customers that the product they have used successfully for years is less pure than they thought. That is a good way to go out of business.

If there is gross impurity that has health, safety, or other product quality issues then selling that knowingly is fraud, but if there is a change in the second decimal place that only affects the assay and not the performance why go there? Your competitors won't, and will be glad to take your business.

[JJG]

Ron, in your example, you ask, "is the problem with the product or the method of analysis?" If your new assay gives you 99.84% because you are resolving/identifying new peaks which before were hidden, or an increase in peaks, then the problem is probably with the product, and not the method of analysis. The new method is likely more accurate because it is identifying something else. The product has a specification for a reason. If you are below that, it is a concern, otherwise, what use does the specification serve?

The question I think you are leading us to think about is this: The product tested under the old method may be the exact same as the product tested under the new method, but it now does not pass the specification. You then need to compare old product and new product under both methods and determine if the old product also reads 99.84%. If so, you may have reason to change your specification to 99.84% because of past history. The change in specification is now made more accurate. This depends on the purpose of the specification (i.e. safety, purity, efficacy, etc.).

I hope you want as accurate of a method as possible, within reason. Once you have knowledge your method may not be accurate, I think you have an obligation to investigate.

[Ron]

In the example I gave there is no new product, the process and the product have not changed. A modification was made to reduce the analysis time to increase the number of samples per day that could be analyzed without increasing the number of instruments and instrument operators. Samples were run in parallel using both the old method and the modified method. The new method not only reduced analysis time, the increased resolution revealed impurities that co-eluted with the product in the old method and lowered the product purity assay below acceptance criteria. The product passed QC standards using the old method.

This is a continuous reactor process producing product by the rail car tank load using an industry standard process. The old method was an industry standard method, the modifications in the analytical method were made to attempt to gain a competitive advantage by reducing costs of QC on the finished product.

Now that the new method shows the product is not as pure as previously believed the company faces choices as to how to handle the issue.

1. The company could sell the product as a lower grade product while costs remain the same, reducing the company net income.

2. Modify the process so the product meets the specifications using the new method, increasing costs while the selling price remains the same, reducing the company net income.

3. Keep on using the old analytical method the way all the competitors do. The product is exactly the same as it has been for years, and meets the customers' needs.

A lot depends on a person's perspective. I learned a long time ago that in most cases good enough is all that is needed, once you meet all quality requirement for a method the job is done. Most methods can be improved, but eventually you hit a point of diminishing returns. That may be the case in the original post, even though the methods may not be great they may meet the needs of the company, and in many cases the resources required to change a validated method are very significant.

A question I often ask in meetings when there is a discussion about changing a method, instrument, etc. is "Is this a real problem, or is it just something that could be done differently?" That may be the perspective of the "stubborn" co-worker, he doesn't see problems, just different ways to do something that may not actually improve the process from the company's perspective.

[Peter Apps]

Is the problem with the product ? - No, because it does the job that it is supposed to do.

Is the problem with the new analysis ? - No, because it gives more accurate results than the old one.

So where is the problem ? With the origninal spec, since a product that is less pure than it was determined to be is suitable for purpose, the original spec must have been too stringent.

Solution that keeps everyone happy ? Switch from packed to megabore columns, and make the columns so short that they match the separating power of the packed columns. The analysis will then run very quickly, and the results from the two methods will be equivalent. Product that is suitable for purpose will still meet the spec, QC costs are reduced, profits rise.

Peter