Traditional fat analysis methods such as Soxhlet and Mojonnier remain scientifically robust, but they can also create practical limitations for modern food, feed, and pet food testing programs – including long turnaround times, solvent use, manual workflows, and limited scalability.
This webinar explains how CPMG TD-NMR and rapid compositional analysis workflows can help laboratories and manufacturers generate accurate, defensible data while improving speed, sustainability, and operational efficiency.
Designed for QA, QC, R&D, laboratory, and regulatory professionals, the session connects the science behind NMR-based fat analysis with the practical decisions teams need to make when evaluating faster, more sustainable testing approaches.
Watch our panel of speakers, Alicia Stell, Ph.D. (Market Development Manager, CEM Corporation), Eileen Stochl (Process Product Manager, CEM Corporation), Jay Alappat, Ph.D. (Chief Science Officer for Chemistry, Certified Group) and Michelle Kelly (Business Development Executive, Certified Group) discuss rapid fat analysis for fast results. and read the transcript below.
Why is Fat Analysis Important? (Alicia Stell, Ph.D.)
Okay, so we are talking about fat analysis. And fat analysis is important because it is directly related to nutritional labeling, product quality, regulatory compliance, process control. So they’re all really important things. And at the end of the day, the food industry’s pet food feed to optimize your manufacturing process, that’s what’s going to be the biggest savings at the end of the day.
And so we want to be able to achieve that, that this modern method, to be able to do that and used to be accurate and defensible data operational efficiency, speed, sustainability. Okay. So this is really what I want you guys to take away as we talk through this to come back and say, okay, yeah, they did that. They presented something that is an optimized method [and] that is meeting these needs.
For what I have pictured there is the AOC food triangle, right? So it breaks up the different components in food from fat to protein to carbohydrates and the different percentages. And essentially our food feed, pet food is going to fall in this. And we’re going to be able to run all of this. So we have a very optimized method for regardless of the sample that you’re looking at.
So our goal is to show you that method.
What is the History of the Soxhlet and Mojonnier Methods of Fat Analysis? (Alicia Stell, Ph.D.)
All right so with that we’re going to talk a little bit about the history of fat testing to kind of bring us up to speed and where we’re at today. And it’s always staggering to me when I look at the fact that Soxhlet was developed in 1879, and we’re still running that in our labs today, right? If we think about where our world was in 1879 and where we are now, and the fact that we’re running a, you know, methodology that was run then today is, like, kind of mind blowing. But we are, so that says a lot to that method! Says a lot to how, you know? That method gave us what we needed. And amongst time we developed some other methods for fat testing: the Babcock test, the Gerber test, and then, as normal with history, we started getting some more rapid methods. So you had an FTIR method, and NIR method, and NMR method.
So, I’m going to reference these a little bit in the next slide in a little bit more detail. But if we move over to the next slide, what I really want to talk about is the future of fat testing, right? We’re not going to talk about the past anymore. Let’s not do a method that is over 150 years old at this point. We’re going to talk about the Oracle system, which I hopefully can convince you guys that [it] really does answer all the needs within the industry. So let’s move forward.
Okay! So we talked about some of the different methodologies, these reference methods and rapid methods. So eventually it falls into these two areas right. The general fat methods that are out there right now you’ve got reference methods.
Those [reference methods] are going to be your Soxhlet your Mojonnier that type of thing. They are reliable accurate accredited, right? This is why we’re running them still today. But there’s a lot of cons with them to, you know. They’re antiquated methods. They use hazardous chemicals. They are difficult to run. They’re expensive to run. So that’s why people started developing these rapid methods.
What are the Benefits of NIR, FTIR, and TD NMR Methods for Fat Analysis? (Alicia Stell, Ph.D.)
And so with methods like NIR and FTIR and TD NMR, they are more rapid. So and they’re easier and they cost less. But the big catch here is [that testing] still requires calibration with these reference methods. So the reference methods don’t go away. You still need them. You maybe just don’t have to run them as often, but you still have to calibrate back to it.
They can be variable because of this calibration. They can be high maintenance and they’re giving you [an] indirect analysis. You’re not directly getting the value of the fat that you can in these reference methods. So, with that, [there] really is a need to bridge these two.
We need a reference method that is also rapid. That’s the goal at the end of the day.
And we’re going to achieve that with the Oracle system that I’ll be talking about here today! And, go to the next slide… Okay! So before I get into the actual technology, so it’s [called] Carr-Purcell-Meiboom-Gill (CPMG) Pulsed Time-Domain Nuclear Magnetic Resonance (TD-NMR) — okay that’s a mouthful! All right. We’re going to get into all of that in a moment. On the details I’m going to talk about the benefits first and really [about] how it does, indeed, create that bridge there that we were talking about.
- It is reliable, so it gives us this reliability to the reference methods. But it doesn’t require any calibrations. It doesn’t require the maintenance. It’s accurate. And, later down the road, obviously, data is important and Jay is going to cover a lot of that to really bring that point across.
- [The technology is] green. I think for a modern world that’s so important, it does not require these solvents that some of these other methods.
- And most importantly, it is accredited! So it is a reference method. We do have an ISO approval that’s listed there for this methodology. So it gives all the benefits of the reference technology as well as that rapid technology without the cons. So let’s move on and watch a quick video on how this works.
How Does the CEM Oracle System Work for Fat Analysis? (Alicia Stell, Ph.D.)
Okay. So in the Oracle is an NMR. So you’re going to drop your sample into the magnet. This is going to cause your protons to align up and down, okay? So, when those protons were specifically looking so you see the sample kind of going in there. And I’ll cover how to put the sample together in a moment. So, basically, magnetic field going to line up and down. We’re specifically looking at the protons on the hydrogen molecules of the lipid molecules themselves. So it’s a direct technology. And so those protons on those hydrogen atoms when we apply a specific radio frequency it’s going to cause those to invert. When that inverts it’s then going to relax. And that relaxation energy, which is kind of visualized by those dots going out right here.
That’s what we measure. And that is directly related to the fat content in your sample. So, again, we’re looking at the proton on the hydrogen atoms on the specific lipid molecules themselves as a direct technique requires no calibration. And you get rapid total fat determination. Now that was really fast. So let’s dig in a little bit more to that technology on the next slide.
Okay. So, again, this is CP TD NMR at the end of the day, right? And so that’s what’s going to allow us to get that direct measurement. So what that’s using is a specific pulse frequency that you see pictured here the 90 degrees followed by the echo of 180 degrees. That kind of goes one after another and kind of a train of those echoes.
And that is specifically going to with that specific radio frequency that we have, we can hone in on the lipid molecules. So how do we do that, right? If you’re like “Okay, well, you talked about these protons. Where are the protons in a sample?” Well, the sample is going to have protons in moisture, fat, protein, [and] carbohydrates, right? How are we separating this out?
We’ll cover this in exact moment: We are going to remove the moisture from the sample, okay? So the moisture goes away. We still have proteins and carbohydrates that could interfere. Well, the proteins and carbohydrates, when looking at this specific CPM TD NMR, are going to relax at a slower pace than those carbohydrates and those proteins. We’re able to separate them out and specifically look at the fat molecules.
But then people are like “Well, there’s a lot of fat molecules! You got your fatty acids, your poly saturated, unsaturated, so forth.” Well, within this we normalize that effect. And so we are getting a total fat value of all the fat specifically in a sample.
Okay. So let’s move on and talk a little bit about the workflow.
What is the Lab Workflow When Using the Oracle System for Fat Analysis? (Alicia Stell, Ph.D.)
So, we mentioned that we are going to remove the moisture. You see that kind of pictured there with a standard oven. So you remove the moisture, you draw your sample, and then you do need to prepare it. You saw that in the video, where you had your little tube that you put into the Oracle system. And basically you draw your sample on a sample pad. Just kind of fold it up. We’ve got a little station that you see there that helps you prepare that sample together. It goes together very quickly, in a matter of seconds. And then it does need to be conditioned at a proper temperature. That is important, so that we have a conditioning block to quickly do that. And then you drop in the Oracle. In a matter of seconds, you have the total fat value.
Now, not picture here, for those who have a high throughput laboratory, we do have a row pot! We’ll essentially take it and put from that conditioner block and drop it into the oracle itself. So it’s basically going to do that for you. So it does have automation as a possibility for those high throughput labs.
If we go on to the next slide, we can see another option for moisture, okay? So, in the previous picture we saw that oven and standard technique. We’re just going to do a loss on drying to get our moisture analysis. Well, here, the CM Smart system is a microwave infrared loss on drying system. And all the benefits we talked about with the Oracle where it was rapid, easy and compliant? They apply here as well! The Smart 6™ System is ten times faster than traditional drying methods. It’s very flexible. You can use it without any of the products we’re talking about. You can use with the smart system, and it pairs directly with that oracle. So they kind of talk to each other. It’s a nice little setup! So you’re getting your drawing in a matter of minutes and you’re fat in the matter of seconds. So, in a very fast time frame you can achieve this entire process.
We move on. Okay, I am going to let Jay cover the data in general, right? So he’s going to take a deeper dive. But I can say all of these wonderful things about the Oracle system and what it does. And if you don’t have the data to support that doesn’t really mean much.
How Accurate is the Oracle System for Fat Analysis Compared to Reference Methods? (Alicia Stell, Ph.D.)
So here’s just kind of a very high level snapshot of some data. This is thousands of samples that were studied, from CRM to outsourcing data — a variety of different samples, from zero to all the way up to total percent fat. And what you see here is the Oracle [captured] percent fat, graphed around the reference [captured] percent fat. And you see that R-squared value of 0.995. It is really accurate, right? And this is what led us to be able to get that ISO approval. Right?
So, it is a reference technology. It is very accurate data. And again Jay is going to talk more to that and through the reproducibility as well. So we move on to the next slide. And so I said I’d reference this AC food triangle again because as I started my portion of this talk I talked about how we want to be able to have a universal method for all of these different types, a very optimized method.
And I hope I’ve shown you how optimized the Oracle is for a both rapid and. Gets rapid and reference methodology. But Jay’s going to dig in deeper here to cover sample types throughout this triangle. So whether it’s food, feed, pet food you can run it and you don’t have to do the calibration. It’s universal. Just move on to the next slide.
So I just wanted to talk a bit about some of the savings that you get here before I pass it over to Jay. And so here it kind of walks you through a chemical extraction versus a oracle extraction. And again, no hazardous solvents at all. Very rapid. Doesn’t require a lot of bench space. It has less consumable cost. And you’re going to get very reproducible data. So, well, yeah there’s a little bit upfront cost for the instrumentation itself. You’re very quickly going to get your return on investment there with running this type of methodology.
How Does Ash Testing Work Using the Phoenix Black System? (Alicia Stell, Ph.D.)
Go next slide. So I’d be remiss if I didn’t make just a quick moment and talk about the Phoenix Black system. So up till now we’ve talked about total fat and moisture.
But most of you in your lab probably have a muffle furnace. It’s kind of a commodity, right? And you’re probably doing ashing for some sort of application. And you may not know that CM has a next generation muffle furnace. And all the benefits that you’ve just seen for the Smart and the Oracle system ring true here as well.
So it’s rapid. It’s 90% faster than traditional methods. It brings a lot to the table as far as safety is concerned. And, like everything else, [it is] compliant. So you can drop this right into having reference methods and compliant data so we can move forward. All right. So, full circle, I hope that I’ve convinced you that this oracle system sits as that bridge between reference technology and rapid technology. It is a reference technology. It has ISO approval. It is universal regardless of sample size. No calibration, no maintenance. But it’s also a simple workflow. You get rapid results, you don’t have to use harmful chemicals, and it’s easy to use. So, with that, I’m going to pass it over to Jay to really dive deep into the data and prove this concept even further.
How Much Faster is Fat Analysis Compared to Traditional Methods? (Jay Alappat, Ph.D.)
Thanks a lot, Alicia. Thanks a lot. Michelle.
So, I’m Jay Alappat. I’m the Chief Science Officer for Chemistry for Certified Group. Now, I would like to shift the discussion from the science, which Alicia talked about, to the business or the operational aspects of this analysis. Now as scientists, we are trained to focus on analytical performance, the accuracy, the precision, repeatability.
But when we are making decision about a technology investments, we also have to think about the throughput, the labor, the safety, the sustainability and the return of investment, of course. Now if you compare Soxhlet with Mojonnier you’ll see that, in the TDNMR, the difference is striking! Traditional methods typically require anywhere from two hours to up to 18 hours, depending on the matrix.
Oracle, on the other hand, can generate results in approximately 3 to 5 minutes. That is one of the greatest advantage I see in this, and you will see a lot of data proving my point.
How Does the Oracle System Reduce Solvent Use? (Jay Alappat, Ph.D.)
The next one is that this also has a sustainability advantage over a traditional solvent extraction method. Sources consume significant quantities of hazardous solvents that require storage, handling, disposal, and regulatory oversight. The Oracle eliminates the solvent completely and improving laboratory safety while supporting corporate initiatives like our organization.
Now, Alicia mentioned the initial investment is a little higher. So, as I said, it’s a fully validated method. You see the data to prove that it is fast, it’s level friendly, no is generated, and customers are seeking more sustainable green solutions. Then one of the cons is growing in the adoption. I will say why people are a little concerned on this technology, and I will show you how we can dispel those myths on why this is a primary technology or secondary technology.
Is it compared to the reference method? We see a lot of data later. Next slide please.
How Does Oracle Compare to Reference Methods for Fat Analysis? (Jay Alappat, Ph.D.)
Now one main concern we hear from our customers is whether moving this newer technology means sacrifice regulatory acceptance. The answer is a solid no. At certified group, our philosophy has always been that scientific innovation should complement, not replace, validated analytical method. That’s exactly the reason we resort to this technology and the method. Now, you can slide a few more points.
Yes. Next one. Yes. So as you can see the modern technologies are now available for moisture. For in the table you see CM smart replaces not complement the machine analysis. Then the Phoenix back which Alicia mentioned about ash. Then for rapid NMR and modern technologies are now available for all of these. And there are other analytes we can do with modern technologies.
Now let us look into this actual data. The slide compares Oracle Smart against reference method for raw chicken as well as for beef blend. The first observation is that remarkable agreement between the modern method and the reference, where they can see the error is 0.06 for chicken fat and moisture 0.23 then beef blend it is one and moisture 0.05.
So for again the average difference only only about 1/600 of a percentage. And that’s very insignificant. So those differences are essentially whether it’s a fat or moisture is insignificant. So what we we are not talking about a rapid screening technology. We are talking about a validated analytical method capable of producing results that agree extremely well with established reference procedure.
We’ll go to one more data on the fat and moisture. Next slide please.
So in the previous slide what we saw was raw materials here. What you see is some finished products. So finished product producers are analytically challenging because they contain complex mixtures of proteins carbohydrates, fats and minerals. So by the way I’ll be talking a lot of pet food matrices. We have done a lot of work on food and other matrices.
But here. Our focus is on the pet food. Yet again there is an excellent agreement in the fat between the reference method and the moisture reference method for Cantwell food, as well as the can’t the finished kibble. So the point here is see this level of agreement across different product types provides confidence that I require I need before we release the results to the customers.
Next slide please.
Now accuracy that is what you have seen in the previous two slides is only part of the one part of the story. The the one is the method performance. And the critical part in the method is precision. What does that mean? If I do the analysis ten times, I should get expect essentially the same answer every time. That’s exactly what precision is.
Precision and Repeatability Across Product Types (Jay Alappat, Ph.D.)
That is what the slide demonstrates across a wide range of products, including cheeses, meat or snacks, bakery products, beverages and dairy. The repeatability is excellent. Low standard deviation. What you see on the last column a 0.130118, 0123, etc. indicates that the instrument produces highly consistent results. The more producible the result, the greater the confidence me and our team and the lab has before we release the data to the customers.
So we see excellent repeatability for the Oracle fat analysis. Let’s see what is going on with the repeatability for moisture. Next slide please. So the moisture the same story. Whether we are looking at cheese sausage putting beverage meets data products. The repeatability remains consistently strong. Most analysts often serves as the first analytical results. Any manufacturer’s need for the releasing.
The products and operators make processing decisions based on the moisture content if those results fluctuate between because between the analysis, that means the poor repeatability and products in production decisions become less reliable. Consistent moisturizers measurements translate directly into more consistent manufacturing, more consistent results to support our customers. Next slide please. Now, what you have seen so far is the accuracy of the method.
Accuracy, Precision, and Method Validation (Jay Alappat, Ph.D.)
How accurate are the results are with what is actually present in the matrix. Then we saw the the precision how repeatable these tests are. So we also we also do intermediate position or reproducibility etc. etc. in the lab. This slide particularly shows how it is comparing with the reference materials. But remember certified reference material provides one of the highest level of confidence in analytical chemistry because the acetate values have been carefully established.
So that is how when we develop a validated method, first thing is accuracy precision, robustness, repeatability, reproducibility. That’s the. Then the next one is compare with a reference material across a broad spectrum of products, including we flour mill, milk powder, dairy products, processing meats, seafood, infant formula, pet food. The smart results are closely matched with the certified values.
This tells us that the technology performs well not only with the routine production samples that we’ve seen in the previous slides, but also against internationally recognized reference material. Now we will go to so you can see the numbers are how close they are and see the standard deviation very small. That is the one you see in the last column in the slide.
Next slide please. Now, again, this one is a kind of educational slide for me and our team. So what you see here is the compare the the directly the traditional chemistry across multiple food matrices. Overall you will see excellent agreement between the two methods. But there are some differences, particularly in products such as cheese, milk powder and liquid.
I mean, these differences are actually very educational. So when I do the animal analysis then we do the traditional vacuum analysis, use of differences, the difference for dry milk as well as for milk one and two and three is pretty high. So that is a that’s how we work with our team to figure out what’s going on is as a systemic bias.
Or is there anything with the sample preparation? In most of the cases, it’s about the sample preparation, which I will show you in one of the later slides. So now Alicia mentioned that the sock slit as well as many dimensions, the total fat inside. But here in the master we are measuring the NMR, which is the total hydrogen associated with fat.
So depending on the matrix sample preparation, extraction efficiency there may be some systematic differences, systemic differences. But it is also is possible to find out to bring these data closer and closer. That means higher accurate numbers by using our sample preparation protocol protocol. So the important point here is the consistency Oracle method provides reproducible results by eliminating many of the variables associated with the solvent extraction.
Next slide please. Again this shows the this slide expands the previous slides of additional products including breakfast sandwiches as well as a pet food. Again the agreement between the methods are really high. And for pet food especially, the average difference is essentially negligible. This is important because, as I mentioned before, pet food represent one of the fastest growing analytical markets.
That is the reason we focused pet food in this presentation. So as I mentioned before, the manufacturer, near the rapid releases of the products while maintaining nutritional label accuracy technology is what we discussing here. Oracle helps laboratories to meet both objectives. Next slide please.
All right. Now Alicia mentioned that these methods, Oracle as well as Smart 6™, are validated against the reference method. So these slides next slides necessary for slides. What I’m going to show you is is how these methods compare with the reference method. Whether the application is a meat as you see in the ingredients are oat is a raw material, deboned chicken, a rice brand.
The workflow remains essentially the same and the measure of the moisture rapidly. With the smart as well as with Oracle, you see how comparable the numbers and as you can see, the references are written below. For example, the oat. The moisture was done using AUC 930 401. Then fat was done using AUC 950 402. So the point here is whether we are using a reference method or the Oracle or the CM technology is a very accurate, precise method data.
So go to the next slide please. In this case it is the treats that you can treat Patriots Chicken strip biscuit as well as the same. And you can see the average between these two the reference and the Oracle Smart 6™ methods are very close. And you can also see the range. In the case of a moisture for chicken strip still .08.
It is, as I mentioned, negligible. That’s what we see across for other matrices. Biscuit Sam and Jackie next slide please.
Again here we did. So we did the raw materials. We did the the treats. We did the wet pet foods. Chicken Philomena on prime rib. Go to the next one.
This is dry pet foods finished pet for a different pet food. So we did the finished once dry dry matrix. So you can see that how comparable these numbers are with the reference methods. Go to the next slide please. Now as I mentioned, I would like to finish what I consider one of the most important scientific message in this presentation.
Sample Preparation Drives Analytical Variability (Jay Alappat, Ph.D.)
And be always the less. And the scientists often talk about the instrument performance. In reality, sample preparation frequently contributes more analytical variability than the instrument itself. The study demonstrates exactly that. So on the left side you see homogenized separately. What does that mean? As we have the sample sent to the lab which did the majority in salt slit, and we do the other lab doing the other location, all of the lab doing the NMR.
That was one case. The other case is where we homogenized in one location and split between the labs. So when bacon samples were thoroughly common and then split between analysis, both slit and oracle produced excellent precision. However, when separate portions are homogeneous independently, variability increased substantially for both methods. So the the lesson is straightforward. Good sample preparation is fundamental regardless of which analytical technology you’re going to use.
So no instrument can compensate for the non representative sample. That’s the message here. So we instead of a group we make extreme care to make sure the samples are prepared in a homogeneous way so that the results we are producing are accurate and precise. As analytical scientists we should always remember that quality of the results begins with the quality of the sample.
So there’s a garbage in, garbage out. So if you don’t put the right sample inside the sample preparation inside, we are not going to get the right process. So next slide please. So I will close with a few three 1 or 2 key takeaways. First, as you have seen, the modern analytical technologies have matured to a point where they can deliver reference quality data well, dramatically improving laboratory productivity.
Second, laboratories no longer have to choose or compromise between speed and scientific differentiability with validated methods like I was in those slides. With the data, we can achieve both. Finally, the future of the food and list analysis is in about replacing good science. As I said, the good science always complement the traditional chemistry methods. It is about using better tools to strengthen high quality, faster data sustainably and at lower overall cost.
So it’s all about turnaround time, the cost as well as the sustainability. And with that will end the presentation today. Thank you for joining us. And we’ll be more than happy to answer any of the questions. So Michelle can look into the questions being addressed in the chat. Then we can answer one by one. Absolutely.
Q&A on Rapid Analysis for Fat, Moisture, and Ash in Food Testing
Michelle Kelly
Thank you, Dr. Jay and Dr. Alicia. That was very informative.
We do have a few questions in the Q&A section, and we also received several questions prior to today’s presentation. If we don’t get to your question today, please know that we will follow up with a personalized response afterward.
The first question is:
“The NMR analysis time is listed as three to five minutes. Does that include the drying step?”
Eileen Stochl
I’ll take that question.
Yes, the three-to-five-minute analysis time does include the drying step if you’re using the Smart 6 moisture analyzer.
If you’re using a conventional drying oven, the drying portion would take significantly longer. However, when Smart 6 and Oracle are used together, the total time for both moisture and fat analysis falls within that three-to-five-minute window.
Michelle Kelly
Thank you.
I have a few questions regarding the cost of the instrument. While pricing is typically discussed on an individual basis, could you explain the first step for someone interested in acquiring this equipment?
Eileen Stochl
Absolutely.
If you’re interested in pricing information, please contact me directly at CEM. I can connect you with the appropriate regional representative who can provide pricing, configuration options, and any additional information you need regarding acquisition and implementation.
Michelle Kelly
I have another question regarding implementation.
“What is the projected lead time to get the system installed and operational?”
Eileen Stochl
For the moisture and fat analyzer systems, our typical lead time is approximately six weeks.
One of the advantages of Oracle is that it does not require calibration development or ongoing calibration maintenance. As soon as the system is installed and verified, it’s ready to run samples.
Installation generally takes about two days.
Dr. Jay Alappat
I can add to that from the laboratory side.
When we installed the technology in our laboratory, the practical installation process was completed very quickly.
What typically takes additional time is our internal validation process. Before implementing any new method for customer testing, we perform comparison studies between traditional methods such as Soxhlet and Mojonnier and the Oracle method.
We want to understand any relative percent differences and confirm performance for each customer matrix before using it routinely.
Most customers have historical data generated using traditional methods, so generating comparison data is extremely important to ensure confidence in the transition.
Michelle Kelly
Thank you, Jay.
We have several questions regarding maintenance of the Oracle system.
One attendee mentioned that they purchased an Oracle last year and would like to know when preventative maintenance should begin and what the maintenance process involves.
Eileen Stochl
Typically, preventative maintenance begins about one year after installation, around the time the warranty period expires.
One of our service engineers will visit the site and perform preventative maintenance procedures that include:
- Cleaning the system
- Reviewing software performance
- Evaluating instrument history and system diagnostics
- Verifying proper functionality
The goal is to ensure that everything continues operating as expected and to identify any potential issues before they become problems.
Michelle Kelly
I have a question regarding sample matrices.
“Are there any sample matrices or additives that can interfere with the NMR measurement and skew results?”
Eileen Stochl
Generally speaking, no.
Because Oracle is a primary method and directly measures hydrogen associated with lipid molecules, we are specifically targeting the fat signal.
Over the years, we’ve become very good at isolating that signal, and we generally do not see additives causing significant interference in normal food, feed, or pet food applications.
Dr. Jay Alappat
I actually have a related question of my own.
Have you observed any issues with matrices that contain paramagnetic ions or elevated concentrations of metals? Would those affect the measurement or require modifications to the pulse sequence?
Eileen Stochl
We really haven’t seen issues associated with those types of materials.
If there were an unusually high concentration of magnetic materials or heavy metals, it could theoretically become a consideration, but we haven’t encountered significant problems with that in routine applications.
Michelle Kelly
Another attendee asks:
“Have you ever seen underestimation of fat content in dry milk samples?”
Dr. Jay Alappat
Yes, and that’s actually one of the examples I discussed during the presentation.
When we investigated those differences, we found that most were related to sample preparation and storage conditions.
If the NMR analysis and the traditional chemical extraction analysis are performed close together in time, the results are very comparable.
However, if the powder has been stored for an extended period, moisture uptake or moisture loss can occur, potentially affecting results.
That’s why we recommend performing comparison studies as closely together as possible to minimize additional sources of variability.
Michelle Kelly
Another question concerns sweetened condensed milk.
“How long does it take to obtain solids and fat results for sweetened condensed milk?”
Dr. Jay Alappat
The analysis time is the same.
The workflow remains unchanged, so you can expect total moisture and fat results within approximately three to five minutes.
Michelle Kelly
I have a procedural question.
“How long does the sample need to remain in the conditioning block? Is conditioning required when the Smart 6 is being used?”
Eileen Stochl
If you’re using Smart 6, additional conditioning in a heater block typically isn’t necessary.
Our Smart 6 methods are designed so that samples exit the instrument at the appropriate temperature for subsequent Oracle analysis.
Additionally, Oracle includes a QuickPrep heating station that can be used if needed.
If a traditional drying oven is used instead of Smart 6, then conditioning becomes necessary. We need to ensure that the sample enters the magnet at an appropriate and consistent temperature because temperature consistency is important for NMR measurements.
Michelle Kelly
Another attendee asks:
“How can users periodically verify calibration?”
Eileen Stochl
We typically recommend using certified reference materials to verify system performance.
Dr. Jay Alappat
That’s exactly what we do in our laboratory.
We maintain certified reference materials that represent different portions of the AOAC food triangle. We use those materials routinely to verify instrument performance.
I also noticed a related question asking about expected standard deviation.
Generally speaking, the standard deviation depends on concentration. For most samples, we’ve observed repeatability corresponding to approximately 2% to 5% relative variation.
At very low concentrations, variability can be somewhat higher, occasionally reaching around 15%.
Michelle Kelly
Another simple question:
“How should used sample pads be disposed of?”
Eileen Stochl
The dried sample pads can simply be discarded.
Michelle Kelly
We have a question about method development.
“Can the method be adjusted or calibrated differently for specific product types?”
Eileen Stochl
No adjustment is necessary on the Oracle side.
Oracle uses a universal method because it directly measures fat. No calibrations, product-specific adjustments, or ongoing method maintenance are required.
The only customization that may occur involves the Smart 6 drying method to ensure the sample reaches the appropriate temperature before entering the Oracle.
But for Oracle itself, there is no matrix-specific calibration development.
Michelle Kelly
A follow-up question asks about the significance of the Smart 6 cooling step and the settings used when Smart 6 is paired with Oracle.
Eileen Stochl
The cooling step is primarily intended to bring the sample to the appropriate handling and measurement temperature.
Once the sample exits the Smart 6, it is prepared and transferred directly into Oracle.
The cooling process simply helps optimize sample temperature before measurement.
As for specific instrument settings, I would probably need additional clarification from the attendee because the Smart 6 and Oracle perform different functions. Smart 6 determines moisture, while Oracle determines fat.
Michelle Kelly
Dr. Alicia, this next question is for you.
One attendee asks about sample placement on the Smart 6 pad.
They were originally trained to place sample dots in a swirl pattern without touching. However, they’ve also seen people place all of the sample in the center.
What do you recommend?
Dr. Alicia Douglas Stell
With earlier systems, there were several accepted placement techniques.
For Smart 6 specifically, we recommend puddling the sample into the center of the pad.
This placement works best with the infrared temperature sensor and allows the system to control temperature most effectively throughout the drying process.
Michelle Kelly
Another question:
“Can the Oracle distinguish between saturated and unsaturated fats?”
Dr. Jay Alappat
No.
Oracle provides a total fat measurement only.
If you’re interested in saturated fat, unsaturated fat, CLA, or individual fatty acid profiles, those analyses still require gas chromatography.
The Oracle reports total fat content, not fatty acid composition.
Michelle Kelly
Another attendee asks:
“If we switch from our historical method to Oracle, how much difference should we expect compared to historical data?”
Dr. Jay Alappat
That’s an excellent question.
If historical testing was performed correctly using AOAC-compliant Soxhlet or Mojonnier methods, I would not expect major differences.
Typically, the relative percent difference may fall somewhere in the range of approximately 5% to 10%.
The exact value depends on the sample type, how the historical analyses were performed, and sample-preparation practices.
The key is to conduct side-by-side comparison studies during implementation so that any differences are fully understood.
Dr. Alicia Douglas Stell
Hey Michelle, I just wanted to add one quick comment.
I’ve mostly stayed quiet during the Q&A because Eileen is our product manager and works directly with these systems every day. She’s truly the expert regarding operation and technical details.
I’m commonly the person customers meet at conferences and industry events, and I’m always happy to serve as a resource. But for detailed technical questions, Eileen is definitely the best person to answer them.
Michelle Kelly
Thank you for clarifying that.
For our final question, I’d like both organizations to answer.
“What differentiates you from your competitors?”
Eileen Stochl
From the CEM perspective, the biggest differentiator is the technology itself.
Many competing rapid methods rely on indirect technologies such as NIR. Oracle uses TD-NMR and specifically the CPMG pulse sequence to perform direct fat measurement.
That allows Oracle to function as a primary analytical method rather than an indirect calibrated method.
Another major differentiator is our service organization. We maintain support personnel around the world and focus heavily on customer accessibility and technical support.
Dr. Jay Alappat
From the Certified Group perspective, our primary differentiator is our commitment to innovation while maintaining scientific rigor.
We believe advanced technologies should complement traditional analytical science.
We focus on:
- Automation
- Innovation
- Reduced turnaround times
- Improved sustainability
- Accurate and precise results
Most importantly, we devote substantial effort to understanding each customer’s specific products and methods.
Our laboratory teams perform comparison studies, optimize workflows, and validate performance before implementation.
Unlike many organizations, the scientists who generate the validation data are available to interact directly with customers and answer technical questions.
We don’t simply generate data and release results. We work closely with customers to ensure the methods are optimized and fully understood.
That’s a major differentiator for us.
Michelle Kelly
Thank you both for those comments.
That brings us to the end of our webinar.
I’d like to thank everyone for attending today and for participating with so many excellent questions.
Don’t forget to download the white paper available in the chat.
We hope this collaborative presentation provided useful insights that you can apply to the important work you do every day.
Have a wonderful afternoon, everyone.
