Addressing labor shortages, workplace safety, and slow QC feedback loops with NIR.

BUCHI NIR Specialist Jason Corbiere met up with BUCHI Market Manager Ryanne Palermo to discuss how benchtop or in-line NIR can address three major areas of concern for QC labs: labor shortages, workplace safety, and slow QC feedback loops. Watch the video discussion here or read the transcript below to learn more.

Transcript

RP: Hi, Jason. Thanks for joining me for this BUCHI TenTalk to compare benchtop NIR systems and NIR process analyzers. 

Jason is one of our NIR specialists at BUCHI. He’s been helping match our customers with quality control solutions for the lab, and as we’ll dive into today—the production floor.  

JC:  That’s right. I’ve worked with folks in the food, chemical, pharmaceutical, and other industries to identify QC solutions that meet their goals.  

RP: Before we get into an apples-to-apples comparison of benchtop and process analyzer NIR systems, can you let our audience know why NIR in general is a popular QC tool in that broad array of industries you just mentioned? 

JC. Sure. QC labs are often home to equipment like HPLC, LC-MS, GC-MS, titration, balances, extractors, steam distillation systems, and various types of spectroscopies, like UV-Vis, FT-IR, Raman, and of course, NIR.  

The traditional analytical equipment, or wet chemistry methods, have been the gold standards for the QC profession from the 20^th century and into the 21^st. And while much of that technology has seen significant upgrades over that time, limitations remain.  

RP: What problems are QC labs facing? 

JC. The biggest problems categories are usually related to labor, safety, and slow feedback loops.   

RP: Let’s unpack those three categories starting with labor. 

JC. Now, more than ever, labs are having to do more with fewer people. Have you tried to hire anyone recently?  

RP: There are more jobs open than there are people to fill them.   

JC. Exactly. On the other side of the coin, we’re just seen more being demanded of smaller work groups due to budget constraints and shrinking profit margins due to factors like inflation. So efficient methods are almost non-negotiable for some companies that can’t afford to increase their headcount.  

RP: So what’s the issue with these traditional methods? 

JC: For a lot of the traditional methods, samples need to be ground, dissolved, concentrated, or otherwise “prepared.” That preparation sucks up a lot of time and resources. The wet chemistry methods also typically require a more skilled workforce. 

RP: In a competitive market. 

JC: Right. No easy task. 

RP: What about the problem of workplace safety? 

JC: Really anytime chemical solvents are involved, the workplace risks go up. Certainly there are ways to mitigate risks. For example, the BUCHI extraction and  Kjeldahl equipment have automation and safety sensors that limit user contact with chemicals, and that helps. But you can’t get away from the necessity of solvents for many of the methods we mentioned earlier.  

RP: For sure. Liquid chromatography, extraction and Kjeldahl are good examples of solvent being a necessary evil.  

JC: NIR, on the other hand, is a non-destructive technique, so no solvents are required. There’s no sample preparation required. 

RP: Well, to be fair, some NIR methods work best if you grind the samples up first. 

JC: That’s true, we have some applications that are more precise when ground samples are used. For the most part, especially for labs working with powdered samples, there is no sample prep at all.  

RP: I guess moving away from solvent use can also support green chemistry initiatives. 

JC:  NIR is certainly a green technique. Reusable sample cups and contact sampling with fiber options or online sensors make most applications zero waste.  

RP: I like that. 

JC: our customers do too.  

RP: Let’s look at that last category of customer problems with traditional wet chemistry methods. 

JC: Yes, low frequency data collection and long feedback loops. Even the fastest traditional wet chemistry methods may take north of 10 minutes.  

RP: I guess a lot can happen between sampling points.  

R: How long are benchtop NIR system measurements? 

JC: Assuming no sample prep, measurements can usually be collected in less than 30 seconds.  Not only that, but you also get to capture multiple properties at once. For example, they can measure fat, protein, ash, moisture, and color, all at the same time.   

RP: What do our customers do with that data? 

JC: With an at-line system like ProxiMate NIR, plant managers and their staff can do things like check the quality of incoming ingredients right in the warehouse. They can do end-point determinations for mixing, drying, and other processes. The idea is to flag deviations and make decisions to save a batch from rework or the dumpster.  

RP: So, is it safe to say that the biggest difference between benchtop NIR and online NIR sensors is that the feedback loop goes from fast and convenient to faster and even more convenient? 

JC: Definitely. Process analyzers, or inline sensors, go right into the process. That can be hoppers, blenders, mixers, dryers, or hovering over conveyor belts. Let’s use a food manufacturer as an example. Measurements like fat, proteins, and moisture, are collected nearly continuously, giving the operators a real-time visual of what’s happening at any stage in their production.  

RP: So let’s get back to that original question: how would someone know if they should do online or benchtop NIR? 

JC: Most quality professionals intuitively know how often they see batch failure in their manufacturing processes. They know the cost that’s associated with reworking that batch or having to dispose of it completely. 

The decision for in-process monitoring usually comes down to ROI. Is lack of access to real-time data costing your organization money? Would having more data either on your raw materials or manufacturing line enable you to catch or prevent mistakes? Once you know the cost and frequency of errors, it is easy to make a choice. 

And realistically, for some organizations, the answer is no. There’s not enough variation in either their raw materials or process for the investment of real-time monitoring to be worth it. 

But most organizations fall into the group where the ROI argument is a clear winner. 

RP: And let’s not forget, you’re still able to capture a lot of quality data with at-line NIR. Staying on the topic of data quality, what can we expect regarding the quality of data when we compare benchtop and online analyzers?  

JC: Well, both benchtop and online NIR are what we call secondary methods. That means that the calibrations we build to run NIR applications require matching NIR data—which is spectra—to the properties the calibration measures—like fat or protein. The fat and protein data, for example, come from the primary methods of extraction and Kjeldahl. They will never be as accurate as ‘the gold standard.’   

RP: So NIR methods could have similar accuracy but not better accuracy than a reference method. Could we expect benchtop systems to have different accuracies than online systems? 

JC: Online analyzers typically experience much harsher environments than benchtop systems, whether that is vibration from the processing equipment it’s tethered to, or more extreme product temperature fluctuations. Those could limit the performance of online sensors compared to an equivalent spectrometer in the QC lab or on the production floor.    

RP: You mentioned earlier that NIR process sensors are collecting data nearly continuously. Is data storage or processing something companies need to consider?  

JC: Yes, managing the reams of data produced by online systems is something companies need to prepare for. Any potential issue can be mitigated by regular backups, but data storage and security should still be a consideration when researching these concepts for your company. 

RP: How about the cost of online vs. benchtop NIR systems? 

JC:  From a capital budget perspective, the startup costs for online monitoring systems tend to be slightly higher, because in addition to the instrumentation costs there are also engineering cost that comes with adapting the sensor to process. 

RP: What does a typical installation look like for a process analyzer? 

JC: There are a lot of different ways to mount a sensor. In the case of something like installing a sensor on an existing pipe with a weld-in flange, production may need to shut down for a short period of time.  

Some customers opt for a bypass option instead of putting the NIR sensor directly in the process. This option is especially useful when sample flow is discontinuous, or there is no good option for consistent product presentation to the sensor. You can also opt for  

Conveyor belt mounts are the least invasive to the manufacturing process, although product height on the belt may need to be a consideration. In this case, the sensor is mounted right above (or on the side of) the belt.  

RP: I think some of the listeners today are probably weighing the pros and cons of moving forward with a benchtop or online NIR solution. If you were to describe the ideal customer profile for a NIR-Online system, what are some key attributes that might describe them? 

JC: The customers I’ve worked with that have benefited most from NIR-Online are the ones who have process deviations that recognize that the cost of reworking or tossing a batch pays for the system within the first one to two years of ownership.  

RP: It probably also saves them uncomfortable conversations with their upper management! 

J: So true. At the end of the day, real-time insight into most manufacturing processes can alert QC professionals to process deviations that might otherwise go unnoticed. 

RP: Thanks for this enlightening conversation, Jason.  For our listeners, if you’re interested in learning more about the specifics of how to implement NIR-Online analyzers in your process, tune into our next episode! 

If you want to learn more about BUCHI NIR solutions for either benchtop or NIR-Online, you can reach out to us from the Contact page at www.buchi.com.  

Thanks for joining!