 into tetrahydrofuran, so this is a complete dissolution. If you sonicate this sample while the, when you first put the PVC toy in, typically with a 15-30 minute sonication, you'll have a completely dissolved solution. So we'll get a solid precipitate that leaves a phthalates in solution, makes it a little bit easier to analyze. So we'll move into the analysis step with GCMS. And I'll mention here that the CPSC method allows for several alternative methods within it. This is to impart flexibility for testing labs, if a lab is already familiar or likes to use a certain test method, that method, those that are listed are available to use instead. So generally, the alternative methods, usually what differs is the extraction technique. And most of them fall into one of the categories I have listed here. A couple of them fall into just a typical fluid extraction, where you're cutting up the sample and you're putting it into a solvent, usually dichloromethane, and letting it sit, the California Department of Toxic Substances method asks just for a 15-minute time, and the Health Canada method is much longer rate, a 12-hour time. So even with this step, you have a long range of sample times. Soxial extractions are more of an aggressive fluid extraction. Here you have an apparatus that I have shown on the right, where you have a ground-up sample, and the solvent is heated and then cooled, the solvent will come down through the sample. So you're having a continuous extraction with fresh solvent. Usually the methods call for about six hours. Methods under this category are the Chinese method and European methods, which I have listed, and also EPA methods. There's some other older methods that we see using this as well. There are a couple other techniques typically found within the EPA methodologies. There we're looking at using different tools for your extraction, such as using a microwave, pressurized fluid extraction, or ultrasonic extraction. And the common bond here we're looking for is an analysis by GCMS. We think this is important. GCMS is our preferred analysis technique, because it's qualitative and quantitative. Here I have a chromatogram shown of an analysis with a standard of all six, of the specifically regulated phthalates, and generally we're getting a pretty good separation, especially for the lower molecular weight ones. And obviously the DNOP, DINP, and DIDP, which you see in the later half of the chromatogram, around 11, 12 minutes. That's where things get a little bit trickier, and we'll be talking a little bit more about that today. We specify GCMS, because some older methods call for GCFID, and simply put, if you're having any overlap in your chromatogram with different phthalates, FID is just not strong enough to get a full qualitative and quantitative assessment. So I'm going to briefly go over some common problems. I'm going to give you very simple problems and very simple solutions that probably don't do these justice, but just to get the ideas flowing for discussion later. Some common questions we get. Often people are asked, is there a standard reference material for phthalates and PVC? Another common problem is false positive results. A manufacturer or someone else might say, I did not put phthalates in this item. I'm sure there's no phthalates in this item, but the test lab still said that there were phthalates in this. So that could either come under a false positive result, or a result of contamination somewhere along the lines. Another issue is the identity of diastonone and diastodecal phthalate. Both of these phthalates actually have multiple cast numbers. So there's different ways of making these compounds, and there's a different chemical formula of these compounds. So trying to summarize some of these issues and solutions as best we can. There are multiple issues of the CPSC method. It's been updated a few times. We're now on the third version. Sometimes people with problems, they'll contact us and say, well, I was following the first version of this, and this is the problem I've had. There are significant advancements between the versions, so make sure you're definitely using the most current one. We have an interagency agreement with NIST to develop standard reference materials. This has been an ongoing project. The last update has been that they're hoping to have something within the next calendar year. We'll see how that develops, but this is a promising idea. There's also some companies that we've seen. I believe SpexCertiprep has started issuing certified reference material. They have a specific level of phthalates in polyethylene. So that's another frontier. You can look with standards companies. They may also be developing their own or looking for something along those lines. I mentioned contamination before. This is possible, and this is a good question to ask if you're a manufacturer or a seller. Essentially, from the starting point to the end point, have I looked at every possible area where phthalates can come into my line? We've heard of recycling plastics. Sometimes people use recycled plastics that had phthalates in before, and that level of phthalates in the new material could be enough to lead to a failure. Also, is the equipment that's being used? Have they run phthalates? Have they been properly cleaned? Essentially, you need to break down and look at every step. Are you sure that every step of the way is... there's no way that contamination can come into this issue? And finally, a lot of our suggestions and emphasis is on qualitative analysis by technical staff, and I'll be going into this more. But we have a few key points here. Our CPSC method calls for a retention time match with a standard phthalate. So make sure when you're looking at the chromatogram and the chromatogram of the sample, the retention time must match or else that should be a red flag. Also, a mass spectrum match. This is not just a general mass spectrum, but the ion ratio is very important and can help in identifying. Blank analysis is crucial for looking for contamination within the lab result itself. If you run a blank and you're seeing phthalates in there, that means you have contamination somewhere along the line. If people are reusing glassware, we found in our lab that our GCMS inlet liner sometimes can carry these phthalates over. So they're persistent enough that they can come along the line even within the lab. And I mentioned before the DINP and DIDP have multiple cast numbers. If you look closely and you get standards of both, you'll see their chromatogram peak shape is a little different. So this is a great way to identify which material you're working for. And you should be if you're quantifying one specific cast number of DINP or DIDP, your standard should be that same cast number so you get a fully well-developed match. So I'll just give you a few examples of why a strong qualitative assessment will help. Here I have two chromatograms. One is of DINP and the other is of a common phthalate alternative, DINCH. If you're looking at the peak shape, they mostly look the same. If you're looking specifically at peaks and valleys, you can kind of tell they're a little different. But our experience and I can see this translating to other labs is if you have sort of an automated quantification procedure where the computer will sort of do the work for you, if it's looking for a broad peak in this time range, it may not know the difference between DINP and DINCH. So here this is where a simple mass spectrum look can really elucidate the differences. The DINCH base ion is 155 versus DINP, which is 149. So all really takes us a quick look at the mass spectrum and you'll be able to tell the difference. Another issue we've run into is DNOP versus dioptal phthalate, another common alternative. Here I have chromatograms of the two of those. You can see the retention time is awfully close, 11.15 minutes versus 11.10 minutes. And here I've noticed that our previous settings for our computer quantification program, if the tolerances are empty enough or wide enough, sorry, it may not be able to tell a difference, especially since our mass spectrum is pretty similar. If you're running a sim scan and looking at the typical DNOP mass ions, DOTP covers those as well. But if you're thorough in follow through with your qualitative assessment, you'll see here's our DNOP typical mass spectrum base peak again is 149. There's a little bit of 167 and 279 mass ions. We'll compare that directly with DOTP and here you can see the jump in the 261 ion. It's almost 50% relative to 149. So if you're looking, again, if you're doing an automated program, you might get a false positive here if you had DOTP. Okay, so we'll start to shift into the rest of the talk here, the rest of the symposium. With a discussion about emerging technologies, we'll have a block of speakers today discussing portable spectroscopy. We've started using this as a phallate screening tool, a general kind of yes or no. Might this item have phallates? When this law was passed, people commonly asked, well, what's the XRF, if we're looking for an XRF for lead, what's the phallate comparable option we have? And this is so far the most promising technology. Also, I'll refer you to this paper by Zang et al. last year from Chemical Communications. They developed a colorimetric test. It's a different application. They're looking for phallates in water. But I imagine with the right ingenuity and the right research, this might be adapted for toys or childcare articles. They use modified gold nanoparticles that actually changed color in water when phallates were present. And also, we're big fans of direct analysis of real-time mass spectrometry. We'll have a talk on that later, so I won't steal too much of that thunder. But it's promising for a screening tool, and we think with the right research, possibly also a quantification method. So just a couple more sources for information. CPSC has its own phallates website where you can get information, common questions. Also, we have a lab Frequently Asked Questions page there. If you're interested more in the Chronic Hazard Advisory Panel, there's also a website for that. They should be issuing a ruling, I believe, within the calendar year. There's currently an ASTM workgroup focused specifically on phallate analysis for low-level percentages. So I have this information here. And finally, EPA has a design for an environment program, and they're looking at alternatives to phallates. Generally, the idea is anyone that's interested in the phallate realm and they're interested in alternatives, they can propose alternatives. I think they'll be looking at some toxicology issues. So any company that's looking to move, this might be a good website to follow. And with that, I'll take any questions. Just one quick question. What is NIST? Oh, sorry, he asked what NIST is. NIST is the National Institute for Standards and Technology. Yes. Do you have a definition for screening? We consider screening just essentially a tool or a method to kind of give you a generic yes or no answer if phallates are present. When we look as a screening tool, if we're looking for lead, we use X-ray fluorescence as a screening tool. This will give us essentially 30 seconds to a minute. We have a general idea of if lead is present or there. So that's sort of the itinerary we're looking for with phallates. With the screening, is there any percent level of false positives or false negatives that you consider an acceptable point for the screen to be successful? He asked if there's an acceptable level for screening as far as false positives or not. I think that's to be determined. We're trying to work with H-technology will have its own limitations. I think the best we can get on any regards we'll work with for now. For a generic tool, if the limit is 0.1% and I think for a final screening tool, it will need to be able to reach that. But there might be different ways to adapt that technology or different applications to try and use that to your best advantage. Hi, I'm Joel Rectum, the Chemistry Division Director here. I just wanted to add to that the what's needed from a screening tool depends somewhat on what it's being used for as well. So a firm, we have a speaker here from BASF today, BASF produces phallates. Their need to screen for products for perhaps impurities in their line are different than the needs that we have as a regulatory agency or a manufacturer who might look to do screening in between their full tests. So there can still be room for these screening methods where it makes sense in a particular end use. I'm sorry, do you get that? Oh, yeah, we consider screen printing applicable to the regulation. Yes? Going back to your definition of the screening, do you think that's going to be market data or just market data? Well, I think we'd like to work towards that point. It would be ideal, but we don't think it's at that point yet. Are these becoming more common alternatives to the local phallates? Yeah, we think they've been in use for... Oh, sorry. Has DOTP, Dialytal Terafellate, been in use often? Yes, we've been familiar with it. We think it's been in use for a while. Does it have any... That's not my... I can't answer that. That would be a good question for the EPA design for alternatives and the chronic hazard advisory panel. All right, great. Why don't we move along? Our next speaker is Rafael Alcosta from Agilent. Thank you, Matt, for inviting me here to this symposium. It's very exciting for us to be here. And what I'm going to do today is share with you some of the different mass spectrometry techniques that may aid you in solving that problem of contamination or library matching or retention times matching up. What are some of the tools that we could use and what are some of the applications that Agilent or our scientists in the field have developed to meet this growing need of phallates and consumer products and specific toys in particular? So I'll go through some of the GCMS portfolio, the different technologies of different flavors of mass spectrometry that could aid you in identifying phallates in a way that gives you that much more confidence in your lab. I'll go through the application notes developed by our scientists in the field and then we'll go through some of the ongoing work that we internally at Agilent are working with to solve this problem of trying to reach the levels required and the different technologies and level of confidence in your results. For example, the Agilent GCMS portfolio includes not only single quads, iron traps, mobile mass spectrometers as well as high-end triple quads and now QTOF technology for mass spectrometry. What I will do is I'll say how each of these different techniques apply to your particular analysis. What are some of the extra information that you get from each individual technique? So for example, we have here the usual suspects of the phallates of interest. There are cast numbers, the molecular weights and there are corresponding ions for SIM analysis. The typical GC conditions that was used in this particular application was a 30-meter column, 0.25, 0.25 micron DB5 ultra inert column that this will allow you to have more, the ultra inertness allows you to have more runs per sample, less bleed in your chromatograms. Since our mass specs are very sensitive, there is a split ratio of 20 to 1 to avoid contamination and column overload. Carrier gas is helium. However, we're noticing that customers are asking us, hey, can we use hydrogen for analysis? And that's something that you definitely can do with your single quads GCMS. However, we're trying to see how that could aid you more. The cost of helium, as you may or may not know, is going up drastically recently and we're trying to address that in terms of how can we take advantage of the properties of hydrogen as a carrier gas, which is not a new thing. In Europe, the currently used, that's the carrier gas of choice and we're trying to make our US customers aware of this as well. Not only do you get the benefits of maybe having a reducing your cost of analysis by having hydrogen through a hydrogen generator in your lab, but also have faster chromatography. So we definitely want to ensure that you're aware of the possibility that that could bring to your labs. For example, here we have a total ion chromatogram of a pacifier extract unspiked and spiked with two parts per million thalic mixture on top. You'll see the butylated hydroxy toluene as the main peak in the pacifier. However, when we spike it at 2 ppm, we're still able to see this in matrix very nicely. So again, the DIMP and the DIP are very a little bit less sensitive to particular analysis of EI. However, there are things that I'll mention throughout my presentation that may allow you to see higher levels of these particular compounds and we'll go through that as well. Okay, so for example, you're talking about sample cleanup and contamination. One possible way to avoid contamination may be to use SPME, or solid-face micro-extraction in your analysis, which reduces some of the non-analyte of interest or the matrix of your particular toys or analysis that would allow you to have cleaner chromatograms in less contamination and more throughput per samples running through your GCMS. Okay, so for example, for these particular compounds of interest, these thalates of interest, we are able to reach in the levels of from 0.25 milligrams per liter to 10 with a linear dynamic range and we have individual retention times for these so we could definitely identify them not only based on retention time but also on the spectrum as well with the GCMS. One of the things that you may have is if you're doing high throughput analysis in your lab and you want to ensure, have one method throughout your lab, you want to ensure that you implore something that we call retention time locking. How many of you use GCMS for analysis of thalates today in your labs? How many of you have an Agilent system in your lab and how many of you are using retention time locking in your instruments so that each retention time in each instrument that you have with the same method is exactly the same? One, okay, I'll talk through that. I'll go through that in my presentation and how that could aid you in having that much more confidence. It's one of the hidden secrets that we have and my job here today is to make you aware of these features that may make your life a little easier in the lab when you analyze and thalates or other compounds of interest as well. Okay, so I'll go here. So for example, let's say you want to take your lab method and you want to take it in the field in the back of a mobile lab and you want to test at the point of sale of a particular product. You could do that with our Agilent 5975T for transportable, which implores LTM technology. Essentially what you have now here is you have the same mass spectrometer that you're used to in your lab and you're able to take it in the field at a third of the footprint and half the power consumption you could run it as part of a mobile lab if you're looking for thalates in the field, for example. And essentially, as I mentioned before, the LTM technology, what it does is essentially a heated oven. So essentially your capillary column is inductively coupled with a heater that allows the maximum amount of heat transfer very efficiently. So now you have faster heating and cooling times so your chromatography may also, it will also improve. So you have more sample through, but now chromatograms that will take you 20 to 25 minutes for analysis may take you five or less minutes for the same analysis and still having the same resolution and sensitivity. So again, that's something for you if you're not aware of, for you to think about it, if you're thinking about doing mobile lab applications of thalates in the field. Okay. So one of the things that we recently were excited about is the introduction of ion trap technology for GCMS. And one of the benefits that you receive with ion trap technology is the ability to do multiple ionization techniques within the same chromatogram. So for example, for an ion for a particular thalates that may not be as responsive to EI analysis, you could do CI analysis, chemical ionization in the same run in the same chromatogram. So you could switch from EI to CI without changing the source in the same run, which will allow you greater sensitivity and specificity for analytes of interest. I'll show a particular example of that. In this case, we have DINP and DIPP, and we have them in EI and CI. If you notice, the EI spectrum shows the classical, mass 149 as the parent peak. However, if we use CI with MTBE, liquid MTBE, we're able to differentiate the two now. You see, have the molecular ion be the parent peak, and now we have 419 and 447. The molecular ions in the higher intensity. So if this is something that may be interest of you for you to identify the two, you have the flexibility of doing CI, liquid CI with MTBE to allow you to differentiate those two and give you that much more degrees of confidence. And again, this is coupled to the same 7890 GCC pneumatics that you're used to on the front end, just a different ionization technique, ion trap technology on the back end. So here, for example, in terms of calibration curve, for BBP and DBP, we're able to go down to levels of one parts per billion or less. So now we mentioned the other flavor of mass spectrometry that could be applied to the Thales analysis involves the use of a triple quadruple GCMS. And I'll go briefly into how that operates. Currently, in our lab in Santa Clara, one of our applications chemists in conjunction with we're Frontier, we have a prioritizer system, and you might hear some of that today. But essentially, we're analyzing Thales using triple quadruple GCMS, and that gives us different degrees of confidence and also lower sensitivity. There's an application note coming soon with that information. But I'll talk briefly of some of the preliminary results that we have obtained. So let me give you a little brief overview of how triple quadruple GCMS operates. For example, you have your typical spectrum. So you ionize your samples. They give you any EI mode. In this particular example, you get a classical spectrum. Everything is ionized, broken down into individual components. However, through the first quadruple, we isolate one ion. In this case, we select the most intense ion, which is 210. We'll select that ion in this particular case. And you're sending that ion, you pass that ion through a collision cell, hit it with some energy and some collision gases. You're able to break down that ion even further into two individual components, and then you can monitor multiple ions from that secondary fragmentation. In this case, it'll be a transition of, for example, 210 going to 158 and 191, which may give you that much more fragmentation and identification for some of the trouble and the likes of interest. So in this case, we have an MRM TIG, total ion chromatogram of phthalates and lemon oil. This is the MRM TIG. The EI spectra of the NOB in the triple quad. You can see that the ion ratios are very similar to your NIST libraries that you're used to. And we have developed transition. We have developed MRM transition, multiple reaction monitoring transition for our different phthalates of interest, where now if you're interested in having more information on your phthalates and hitting that one particular ion with some energy and a collision flow, you're able to get two individual transitions, one for quantitation and one for qualification, and you're able to develop, set up ion ratios for each particular transition. Therefore, they'll meet a particular requirement for your quantitation. This decreases your false positives. We're talking about false positives being a big problem. MSMS could potentially solve that for your reduces, the amount of false positives in your phthalates analysis. Now, for higher quality data and for more interesting results, for example, you could take the power of QTOF mass accuracy and high resolution for the analysis of phthalates and high matrix. You have the ability to do MSMS and EI as well. And in this case, what we have is our traditional triple quad, which is based on your single quad mass spectrometer that you're used to currently using in your lab. We took that technology a step further and incorporated our triple quad mass spectrometer to our TOF mass spectrometer that we had in our LC world and we made this marriage of QTOF. So now you have the ability of doing full scan, high resolution, high mass accuracy, less than 5 ppm mass accuracy, and I'll show you what level of information that provides you in a second. So, for example, if you're used to looking at your typical single quad mass spectrometer and we look at mass 604, what you see in the red is your typical spectrum. However, with the TOF technology, you're able to really resolve the 613 and the 614, and you get accurate masses to the point that you have more resolution. And when we talk about resolution, we're talking about the space between the two different isotopes there that you see. So that's one level of confidence that will also decrease your false positive and negative. So if you have a high matrix and things that may be interfering, QTOF technology with the high mass accuracy and resolution may give you that extra level of confidence of being able to see the needle in the haystack a little bit better. So, for example, your typical mass spectrometer has an uncertainty of, let's say, 0.3 AMUs and for mass 271.98 AMU, there are 7,600 compounds that are possible based on that mass uncertainty of 1,000 ppm. As the uncertainty decreases, let's say to a level of 1 ppm, parts per million, now you have 11 possible compounds that are possible with that particular ion of interest. So that's one of the, again, reduces your false positives at a high level of degree of confidence in your analysis. So to those of you that are using our mass spectrometers for the analysis of phthalates, retention time locking is very important. And as Matt discussed earlier, retention time is one of the requirements for the methods of analysis. Having retention time locking allows you to have one method on one mass spectrometer, transfer that method to a second mass spectrometer, and they'll have the exact retention time. And whether you perform column maintenance or not, you're able to just relock the method. It's very simple to do. It just takes five injections. You can do it overnight or over a lunch break. Or you can do it actually while you are commissioning your instruments for analysis during your performance check. You take one analyte of interest. You look for that ion. And what it does is retention time locking, what it does, it incorporates all the possible variations in column length, pressure, or whether you have a difference of a few millimeters going into the inlet or not. It takes all the possible errors, incorporates them into a pressure error, and allows you to match your retention times for method transfer. So in any system across the, so if you have a lab in the east coast, the west coast, the retention times will be the same if you're trying to match your phthalates and the likes of interest. So yes, you have a question. Can you do that within the same instrument if you were to replace the column? Say, could you pull the data up and then switch up the column and then run that software and apply runs and then have the retention times match exactly what it is? Yes, yes. You can do that. Well, you can change the column. That retention time, if you want the AIP to be at 11 minutes, 11.001 minutes, it'll be at 11.01 minutes. And this instrument and that instrument, whether you change the column every day, all day, year after year, and it's free. It's included in your chem station software. It's a button. If you go in your method, there'll be an option instrument retention time lock and it'll ask you to put a vial in a particular position, position one. It'll do five runs at five different pressures and it'll make a calibration of pressure versus retention time and then it'll retain that calibration file and then your method will be locked to a particular compound. So if you want to lock on your internal standard or whatever it is, you can lock on your internal standard and those retention times will match. That way, you won't have to question, is it this phthalate or is it that phthalate? No, because the retention times will match all day, every day. Again, retention time lock and it's available, everything from our 7975 e-single quad to our QTOF, even to your mobile lab, if you want retention time in your mobile lab, you have that flexibility. So again, one of the things I want to emphasize is the other speakers will go into different other front ends that may aid you in analysis, but I wanted to focus on the different capabilities is the take-home message here is that if you're looking for routine analysis and you want to do NIST and this is the convolution, our single quad will do a great job. If you're having problems with high matrix interference, then that's when a triple quad may be more suitable for your analysis. So we'll give you that MRM transition of one ion going to two different transitions with two different ion ratios, so you get more degrees of confidence. And if going from EI to CI is something that you really want to do for your analysis, to give you that piece of mind and give you that much more confidence in your results, that's when an ion trap with liquid CI with MTBE may aid you even more. I just want to, the references, you can look these up in our application notes and there's more to come, specifically on the QTOF and the triple quad analysis. And I want to give a special thanks to the scientists that have worked on this, Stephen Bowman, Fred Feierherm, Robert Kubis, and Matt, thank you for inviting us to be here today. Any questions? Yes. One quick. Several years ago, I guess there was an application published by Usit of ammonia, I think it's a CI source. Take a very good resolution in the parent ions and so forth. And is that possible to use these methods to differentiate the ion PIDP? As Matt pointed out, they tend to overlap in the heart. Yes. Please repeat the question. Oh, yes. The question was whether we could use ammonia CI reagent for the analysis of phthalates. And the answer is yes, you can. In this particular case, MTBE was used for the analysis, but you could definitely use ammonia or even methane if that wash. But ammonia tends to be more sensitive in terms of CI from what we've seen in particular phthalates. Yes. Good question. Any other questions? Yes. Does retention time lock work for multiple analytes? The answer is yes. What retention time locking does is it takes one particular analyte typically in the middle of your chromatogram. So you pick whatever analytes and you lock on that and everything else is going to be relative to that. If this analyte match, everything else should match. Depending on you have a good column, your column is not bad. So yes. So all the components will be retention time locked, but you only lock one compound, everything else is relative. Thank you. How does the retention time locking work for some of the uglier chromatograms like DIMP and DIDP where it's a broader peak? Is that an issue or is that handled pretty well? With broader peaks, retention time locking, what it's going to do is you're still going to have that broader peak. However, that broader peak will be at the same time. So that's essentially what's going to happen. It's not going to aid your chromatography in terms of resolution, but it will give you that same time for that blob of gram that you have for DIMP. Yes. Good question. Yes, another question. Is that trapping and if so, various amounts of ions? Okay. The question is, is our triple quad trapping or ions? And the answer is, it's not really trapping, it's scanning. It's scanning very fast and it's taking a selectively passing ions that we select that we want. So for example, in this particular example that I showed, let's say you want to isolate 149. Ion 149 will go through the quad. It will reach a collision cell. Once it reaches the collision cell, it will be hit with a particular energy, typically between 10 and 15 volts, electron volts, and a collision gas. And from there, we'll have two different transitions that we can monitor, typically the most, the two strongest transitions to give you that much more degrees of confidence. The question was, is there any gain function, automatic gain function? The answer is, there is no automatic gain. However, for compounds of interest, you could adjust the gains in the different time segments. So for example, for analytes that are maybe less sensitive, you may adjust the gain at that particular moment in time. Yes. Yeah. Oh, thank you so much. Thanks for everything. You're welcome. Up next, we have Bob Freeman from Frontier Labs. Okay. All right, good morning. I'd like to thank Matt for the opportunity to talk about the method that's been in development now for about two or three years. About three years ago, we started at the ASTM D-20 form to committee to look at low-level phthalates and polyvinyl chloride. And for the past two or three years, this is what we focused on. We've done two sets of around-robin samples and we've done an independent validation study with some laboratories. So I'd like to give you just a status report on where that method is to give you a little bit of data. And then I'd also like to kind of address the issue of phthalates in whole because you'll notice that for those people who work on methods, phthalates are particular troublesome because the fact is that the method no longer is what it was designed to do. We started off with toys and we started off with six phthalates. And in the last two or three years, now all of a sudden we're looking at electronic products, we're looking at food containers, we're looking at a whole variety of different matrices. And you'll notice that we've gone from six phthalates to now there's interest in more than six and in some cases, 15 and 16 phthalates. So whereas we started with toys and consumer products, now we have to work, the stakeholders include formulators, recyclers, manufacturers, retailers, consumers, all of a sudden everybody's interested in phthalates. And what we find out is interest in phthalates all over the world. And this is all well and good, except for now you notice that the list is starting to expand a little bit. And so some of the electronics companies have included things like diisobutyl phthalate, which is a plasticizer that's not too uncommon. It's a low cost replacement for DBP. And so now we've gone from three regulated phthalates to a possible six phthalates, seven phthalates. And you can see we're having the typical encounter here and that is as we develop a method, all of a sudden we include more and more matrices. And as we do that we want more and more compounds. And as we do that, of course, we want more and more sensitivity. And so those of us that work on methods, this is a continuing problem. There's even individual countries that are writing general methods that have promulgated most of them in 2011. And these are just mentioning things like all products containing phthalates. I mean, that's a pretty general description for a method to address. And because now you hear more and more about the countries not being important, but global, global international corporations really being the boundaries. There are many corporations that work on a global scale that have their own in-house methods and in-house lists of compounds and in-house detection limits for a various number of phthalates. And so these are just some of the electronics companies that we know and are familiar with that are working on phthalates and recycle plastics and different lists than we work on right now. So we've gone from in the beginning the ASTM method as PVC and six compounds. So now we're talking about maybe 15 or 16 and way beyond PVC. So let me back up a little bit and talk about the ASTM methods. We're looking for a standard practice for the determination of low-level phthalates and polyvinyl chloride. And the method that seems to give us the best results is thermodisorption GC mass spec. So there's really five steps in this method and this is the most difficult step, this step right here called sample homogeneity because when you take a toy, for example, the different parts of that toy are made up of different plastics and each of those different parts of plastics can have a different level of phthalate. And so I would say if I were to look at the overall scope of analyzing phthalates in materials, the biggest problem is a homogeneous sample. Because if I take this toy apart, I can get three different numbers. Some of them are good and some of them are a little bit higher. A circuit board poses the same problem. A food container poses the same problem and so on and so forth. So this is the single most difficult thing to do and that is to get a homogeneous sample. Now it sounds nice to say crowd milling. You go into laboratories to do crowd milling and contamination, crosstalk between crowd mills is ever present. And so one of the problems with phthalates, as we mentioned here, is contamination will run off the bat if you're using one crowd mill five toys. The possibility for cross-contamination is very high. This is the toughest step. The method that the the ASTM method for based on thermodisorption GCMS spec uses two solutions. One is just a standard solution of the phthalates of interest and the other is the sample itself which is dissolved in tetrahydrofuran. So there's only two solutions involved. The injection is simply placing these solutions in a small container, raising the temperature of the container such that the target compounds evolve or bake out or are extremely extracted and analyzing just that portion of the sample that contains the phthalates. Separation is done by a standard 30-meter dB5 or 5% fennel column and detector you can pick them. Most of the work we've been done has been done with scanning MS. Several of the validation labs use the SIM MS. You can use triple quads. You can use negative behind CI. This is the area where the method seems to be the most applicable and that is this is unlike the first talk. This is how to get the sample prepared and how to get the sample into the column. Okay. What column you use is up to you. It's independent of the column choice. What sort of detection algorithm or system you want to use is up to you. So this is the sample prep. The whole idea here is to get rid of the solvents and get rid of the glassware. Because I've gone into laboratories who do this and they'll tell me oftentimes they spend more time cleaning socksless than using socksless. Because that way it's a difficult to get rid of. So what we do here is we start off with the product some sort of material could be a circuit board could be a duck. We all like ducks. You ground mill it you want the smallest pieces you can if possible you want dust because you want a homogeneous bottle of dust. You take a small bottle of dust it's a 10 milliliter volumetric so now you've got so much weight you add to that THF you make a solution now so now you have a solution a THF solution of your sample. The method calls for taking 10 microliters of that put it in a small cup evaporating the solvent such that you end with a very thin film of the sample on the inside surface of the cup at the bottom. So to prepare a sample like this once the crowd is gone it takes about five minutes the only solvent you're going to use is 10 milliliters of the THF you're going to use disposable glassware so there's no worry about contamination or cross-contamination that little cup then is ready to be analyzed that if you wanted to do a screening method it's not necessary to go through the thin film process and that is you can take this sample dust from your jar and just weigh in 100 micrograms like that internal stand at that point right now just to monitor your performance that's easily done so either way you end up with a cup that's ready to analyze now the difference between a thin film that is making a THF solution put it on the cup and then evaporating the solvent I'm using a solid material that's basically reflected in the precision so if I use a THF solution this is a half a milligram of sample now you can see for this DEHB a percent relative standard deviations for injections if I use a solid form for a screening method where I just take the dust if you will 0.5 milligrams of dust and put it in the cup and analyze it directly you can see my precision goes down and of course the reason for that is if I look at this collection of dust that I have the chances of having the same amount of particles in each sample are remote so I give it a more scatter in the data where if I make a solution I can analyze exactly the same precision but for a screening technique the solid analysis seems to work just fine can we hold to the end here? thank you so the sample sits in a little cup it sits on top of a liner to make my injection I drop the cup in the furnace the furnace is at 100 degrees I heat the furnace to 320 degrees and as I heat it up the phthalates will evolve and those are carried onto the column these little cups then can be placed in an auto sample and I can do a number of samples now one sample per hour the reason for that is because the calibration the quantitation is done by standard addition that requires two analysis one of the sample and one of the samples that's been spiked with a known amount so that's two analysis so about one an hour if you're going to do a screening sample one minute if you will sample prep or extraction and then I can do about three per hour in the screening mode so the sample is heated I thermally extract the phthalates of interest they go on to the column and I get my chromatogram the conditions I'm using from my survey motor basically 100 to 700 at a fairly high rate that's going to 12 and then I do my thermoters option my thermoters option basically is 100 to 320 at 20 degrees a minute if you're using a thin film you can do 40 degrees a minute really won't make much difference this goes on to a column at 80 degrees so here's what it looks like the top trace is my screen I don't have to do the screening for every sample the importance of the screen is it characterizes the sample to it once but if I'm doing circuit boards or if I'm doing different kinds of materials then maybe I have to do my screen to find out if my phthalates have shifted but you can see if on the top trace this is a PVC sample dinsch as a plasticizer I get a large group of compounds out if that's the phthalates any dinsch may be present there's also a peak for HCl because that's a decomposition product of polybond and chloride out there where I've polymerized or pyrolyzed the PVC so I find my zone by looking at ions and you can see the lower trace where it says phthalates 149 you can see all the 149 all my phthalates are out by 320 and so my thermal desorption zone is 100 to 320 at 20 degrees a minute the nice thing about this is by doing this I'm going to extract those phthalates and then put those on a column for separation of the polymer so I don't have to worry about baking the system out to clean up the polymer I don't have to worry about contamination of my system as I backflesh the polymer just stays in the cup and when I'm done I wash the polymer out and reuse the cup or get a new cup all I'm going to inject onto the GC it's just that fracture of the sample that's extracted at 320 degrees so I can identify different compounds by this is probably more interesting right here this is the reproducibility this is six injectors of pvc dense sample and it's a little hard to read but what I have here is the percent relative standard deviation of the precision of a multiple runs of the same sample and you can see the precision across the board is less than 2% so I can take one solution and analyze it six times and my precision is pretty darn good I mentioned so I'm going to run my sample and get an area number for my compound and then I'm going to spike the sample with a known amount in this case 0.15% concentration and I'm going to rerun the sample so this is just the spike sample I just show three of these right here so this is the pvc dense sample that's been spiked with a known amount of each of the phthalates and again you can see the precision across the bottom of the spike numbers again I end up with a series of curves like this so these are the six calibration curves for the six compounds the one in the middle is at the zero that's the sample number the one on the far right that's the spike number and there's some mathematics that go through and I can calculate the amount of amount of each of the phthalates in the original sample and if you do that again you can see the precision there is less than 2% and it's a little hard to read the concentration of this of each phthalate of this particular one was about 980, 990 nanograms so the accuracy is pretty good now the IEC in Korea is also interested in this method this is some of their work these are their six calibration curves only in this case they spiked with three different levels and all you do is take that one solution and you can see from your upper right trace one case you put two microliters the next case three microliters the next case three microliters that's how you get your multiple points so it's still one solution and you can see their recoveries of all the six phthalates are pretty good and their precision is also less than 1% in this case now we did a method validation study that is we went out and found five laboratories who are willing to try this these are laboratories that hadn't done this before in one case one of the laboratories had a new hire do the study for us and this is what we found from the validation samples to them with the same sample the thermo-desorption temperature is system dependent by that I mean if you have a frontier system or a PVT in the system or a CDS system or a gyrsel system whatever system you're using for your thermo-desorption you have to be very sure that what you're looking at is the sample temperature and not the thermocouple temperature so there's some differences in some systems between the set point and the actual point all you really do is the EGA and you can determine the temperature range of a wood set system operates best some of the validation labs use SIM some of the validation labs you scan both produce comparable data Thinfield columns are definitely preferred over thick film columns we don't want a whole lot of retention and a whole lot of time in there the standard edition everyone use standard edition the one point standard edition in the R squared was always greater than 0.99 if you recall back to this study this is three points sample and the R squares and all these are one but these five laboratories all did 0.99 are better the inner lab injection to injection reproducibility is less than 5% the data I've shown here is all about 2% one of my five labs with the newbie they were about 30% higher than everybody else in the whole wide world so there was some systematic issue there but I included all their data that's how it comes out to be 5% the inter lab sample to sample reproducibility is always less than 10% and the laboratory to laboratory accuracy is 25% now again if I take out the data from that one lab it drops down to about 10% but I can't find a reason to do that so these are the numbers we came up with from the validation study of these five laboratories now all of a sudden we're getting requests for extended target compound list so here's the extended target compound list we've been asked to look at next to this method I've highlighted and read the six compounds we started with in order to do this chromatographically it's a simple thing you just drop the initial temperature from 100 to 80 degrees and you try to do a little bit better separation on that front end but still you notice it only goes up to DIDP so it doesn't go any further and so the analysis time is still on the order of about 20 minutes so it has already lengthened your analysis time and since your sample prep time is making up a 10 milliliters solution of THF it doesn't affect any of your sample prep and your operational parameters so it looks sounds like it's almost so good to be true we've done this now for three years the TD method is easy it takes one syringe and a little bit of glass where it's clean there's no cleaning and cleaning and cleaning and socks and extractors and that sort of thing it's green because I'm using almost no solvent at all and it's fully automated that sounds really really good it's precise and accurate the procedure is always less than 2% the accuracy is plus or minus 10% in your laboratory and it accommodates an extended compound list so this is all well and good there is a downside though and the downside is this run-to-run contamination issue is a real problem at 1,000 per million it's not too much of an issue that's bucket chemistry but as the sensitivity requirements get higher and higher more and more sensitivity we have to find a way to really monitor this contamination the contamination is in the injection port there's no doubt about that also the quant ions have to be verified as we change matrices that's one thing to look at PVC which is what the ASTM method is focused on and we know probably probably and probably other ethylene and we've done a number of different polymers but what happens with the circuit board what happens with some kind of material you haven't really looked at so you really have to verify you have no interference when you start changing matrices and the third thing is when you deal with these compounds that are isomeric the quantitation becomes a little bit more difficult and it could be although my mother will kill me for saying this it could be HP, LC maybe a better way to do these multiple these compounds that have a lot of different isomers so in the end those of us that develop methods are in the deep trouble here because now we've got a method now that was developed for six compounds in PVC and we're asked to do more and more phthalates and we're asked to do more and more different matrices and we're being asked to get lower and lower and higher I guess higher sensitivity and in the case of phthalates this could be a real challenge for the laboratory Any questions? We have time for one or two questions Yes Your relative standard deviation is that one signal? I think that was how was the RSD calculated? Yeah, that's one and a half signal Yes The first cryogenic milling The The question is is the precision and accuracy depended upon the cryogenic bill into the sample Great All the work here was done with PVC standards that were prepared by one of the members of the ASTM committee and so it was literally a sheet of PVC and so Yes See now you're getting into legal issues I mean obviously you are as a citizen I think you are the methodology that we're familiar with it calls for a number for the entire product and so yes you are deluding the beak but on the on the other hand you know how much the beak is of the duck, you know but that's the critical issues that homogeneity sample homogeneity Joe, how's the that Just quickly to add on to that that we're really not here to talk about the policy and the regulations but we have issued guidance that it's the component parts that the Thali's limit applies to so where our very initial guidance was different based on the initial read of the rule we have guided those the individual component parts but regardless of that then the analysis is still the same alright let's thank Bob Up next we have Chris Steele from Bureau Veritas just press forward or back or you can press one of these buttons just make sure you speak clear enough alright thanks Matt yes I'm filling in for Lisa Clarice our global technical consultant specialist because she is there's a touch of a flu bug going spreading around at her Buffalo office so hopefully I don't get it after we leave town and they get back to Buffalo but so I'm speaking on behalf of Bureau Veritas in Lewapur so yes no problem so thank you Matt for letting me represent our company here and I'm here today to talk about using a dual instrumentation for Thali's analysis and plastics and what I mean by that is two systems two mess spectrometers the first of course GCMS is I think a lot of you are familiar with and that the CPSU method revolves around but secondly the LCMS system using both in conjunction with each other just to kind of rule out any interferences that GCMS may pose basically to get like a second set of eyes looking at a particular sample scan Bureau Veritas has had about I'm guessing at least ten years of experience testing for Thali's a variety of consumer products many different plastics and I'm here to try to share with you hopefully some helpful information that has worked for us in our testing approach to resolve these tricky samples as we know as time goes on and regulations are expanding manufacturers seem to be trying to get more and more clever as trying to find workarounds to find alternatives plasticizer alternatives to have their samples compliance so in effect we see it we see a lot of different scans in our Thali's extracts because of this because they're using alternatives they can't hear you the back use speak a little louder sure, sorry so that's my purpose here today is to try to share with you our Thali testing approach so just as a quick question I'm going to go over the test method that we use the instrumentation that we use GCMS and LCMS of course results in interpretation and I'll try to answer any questions you may have about our approach our test method extremely similar to what's in the CPSC essentially we cut the sample into very small pieces of the component we place it into a disposable glass reaction vessel or vial then we add tetrahydrofuran THF and sonicate it at 40 degrees centigrade for 30 minutes which in most cases will be long enough and aggressive enough to get that into the organic solution and sonicate in additional 30 minutes if that sample does not visibly dissolve sometimes that happens with hydroplastics after which acetonitrile is what we use acetonitrile ACN is added drop wise to precipitate out the polymer that's been very useful for us I know other solvents I think CPSC mentions hexane using hexane otherwise maybe using methanol they may work in particular applications or plastics PVC probably works but we've found that acetonitrile will precipitate out more polymer more types of polymers than the other solvents in the end it results in a cleaner layer on top of your precipitated plastic precipitated polymer so all that junk and interfering things that may interfere with the analysis will come out of that solution and then you can inject much more clean solution into your instrument so once that's done you'll have to stand for 30 minutes and take that solution and put it into an auto sampler vial and it's ready for analysis actually add anthercene to the rate of the anthercene is an internal standard then you can filter it and analyze it by GCMS and or LCMS and currently LCMS is not listed in the CPS C method currently I will go most of you probably are already aware of the function of gas chromatography you have a carrier gas usually helium the sample is injected into the injector volatilized goes to the column the intellect separate out from one another as well as any interfering compounds they separate out from one another then it enters the mass spec detector fragmentation occurs mass spectrum is produced and the signal is plotted on the chromatogram what we're looking at here is just an example of many different types of thalates that are out there obviously it's not a complete list but it's just here to demonstrate what we could recognize a particular sample extract any one of these or combination of these could be found in your sample so a lot of them are nice clean needle peaks of course as Matt has mentioned and many of you already know we have the broad finger peak type thalates two that are regulated DINP DIDP you can't see it too much on this slide but it's buried in there between the range so this is just to demonstrate what we could be looking at in a sample like I said this is not a complete list by any means some of these are regulated some are not so go on to our next slide this is just an example of a mass spectrum of DDP just as an example we go back a slide DDP is just one that's highlighted it's a mass spectrum creating that response this is it of course base peak 149 is very common qualifying IN could be 223 which is unique to DDP GC mass spectrum of DINP and DIDP those troublesome finger peaks DINP would be 293 you could extract that out of the chromatogram but even doing so it still can sometimes be a challenge depending on what you might have interfering with that I'm going to talk a little bit more about that in the upcoming slides as well this is an example of a non-regulated phthalate currently non-regulated you never know what the future holds diso-octal phthalate basically this is a finger peak as well we don't know what manufacturers are putting in we've certainly seen it many times before in samples diso-octal phthalate it's a it's like a couple of carbon carbon chains last of course in DINP and DIDP the non-osin decals so here you can see the retention time it's a range it's a broad range it's an isomeric mix of course and that's why it looks the way it does so you can see it spans across the timeline the retention timeline between like 11.25 minutes to 12.5 so if you see something like this in a sample it can be challenging to properly integrate the regulated phthalates or even see if they're there DINP, DIDP even after ion extraction honing in on a particular ion that may be unique to that particular phthalate so here is an example of DINP on a GC chromatogram highlighted there it's if you want to call that napax it's peaking out around 13 minutes but the range of isomers begins at about 12.5 and extends out to like 13.5 minutes so anyone doing phthalate analysis by GC this is second nature I'm sure see this all the time so again it's very difficult to integrate and try to accurately capture the area counts of that DINP peak if you have other interferences or other phthalates present it just it can be challenging on the GC so here we're just looking at some chemical structures of some regulated phthalates we have DINOP DINP DIDP DINOP is unbranched so that's a single peak but all three of these have roughly the same retention time on GC DINOP is an octal DIOP which I mentioned before is an iso-octal so you're going to see 279 IN279 even if you extract it out for DINOP if you have DIOP in your sample and what do you do you see this big unregulated 279 IN chromatogram and then you may or may not see this little sharp peak come out of the that may or may not be DINOP if it is DINOP and you're confident it is okay now how do you quantitate it it's very hard to baseline integrate when it's coming out of a mess so DINP and DIDP of course are isomeric finger peaks they actually overlap on GC even though they have quantifying or qualifying ions associated with them they can be troublesome as well and of course the reason that they are broad finger peaks is because isononal DINP and isodacyl DIDP have it could be a multitude of arrangements of the groups on them methyl groups that could be located in many different statistical possibility and that's why you see that broad finger peak of course it's not just a single analyte it's a mixture of isomers I'm sure you're all familiar with why they look the way they do on GC now let's get into LCMS instrumentation liquid chromatography coupled with a mass spectrometer pretty much basically the same principles GCMS you're just using liquid as the mobile phase or carrier to carry along your analytes they get separated by a column the same manner enters spray chamber mobile phase is evaporated fragmentation occurs and then the resulting ions are separated out by mass to charge ratio and responses created as well as a spectrum so this is an example this is a DINP at 1 ppm on the LCMS so what we're looking at here is we run in scan mode this is actually EIN 419 I believe DINP and this is only 1 ppm we do run standards a little bit lower than this but the point is you can see how much better it looks to a chemist I mean this I like the way it looks it's very integratable the signal to noise ratio is very large it's just just a great responding signal for DINP it just jumps out of the baseline it's great DIDP at 1 ppm similar similar look to it comes right out of the baseline you know you get a much better response of course compared to GC there's no broad finger peaks and actually DIDP is fully resolved from DINP on the LCMS are you using reverse phase? I'm sorry? is this reverse phase HVLC? yeah it's reverse phase yep and this slide is just kind of giving you a side by side DINP on the upper right on the LCMS DINP on the lower left on the GCMS you can see it's hard to see of course because DINP on GCMS relative to the single analytes it just kind of is not very responsive because it's just a isomeric mix so that's one advantage that we found with using LCMS is really for for those finger peaks so our approach is that we use both GCMS and LCMS in conjunction with one another if the lab decides to use GCMS primarily it's kind of a flow chart showing what should be done if you find the Thales DNOP DIDP DINP they should be confirmed on LCMS just to ensure a good quantitation and just to confirm that they're there and now if you're going to primarily use LCMS you can do that as well but then you have there's things you need to know about that though if you find DBP Di-butyl phthalate which is regulated it also it will co-alute with an unregulated phthalate Di-isobutyl phthalate so if you're getting a detection on DBP it may really be Di-isobutyl phthalate they essentially co-alute with one another so that's why if you do find DBP it's pretty recommended it's pretty urgent that you confirm it by GCMS to see is it the regulated form or the unregulated or a mixture of both and we've seen all three cases there so that's pretty much the flow that we found to work for us so in conclusion a dual instrument system works for us it's been a useful approach enhancing our data integrity with challenging samples and like I said before as time goes on just when you think you've seen it all you haven't we've seen new plastics come out with different compounds and adding new challenges it's like an ongoing problem it's like an ongoing process the learning never ends so this is what Bureau Veritas has found to work for us in our long history of phthalate testing in order to provide clear results from an otherwise messy scan just because of this we feel much more confident in the data that we issue basically we were having two detectors give us data if need be it's also important to note that you don't need the RGCMS or LCMS I mean you can use models that have been around for years it's not like you have to go out and buy the most sensitive instrument to take this approach on in that sense it helps labs that I'm looking for cost savings of course so in the end it's all about just feeling confident about the results and data that your lab is issuing with these tricky samples are there any questions? do you have time for a couple of questions? yes the labs in Asia will use the dual method or is it just unpopular? yes the question was do our labs in Asia use the dual method approach I'm not sure from my understanding I know they have both okay my consultant there is saying yes they do I knew they had both instrumentations but I just got confirmation that yes they do the question was is this our standard practice in testing for DIMP and DIDP yes it is just because it's just so tricky on GCMS we just wouldn't feel that confident issuing results if we get something that would really challenge the chromatography like we often see on GCMS so this is our our good practice here go ahead on LCMS sure actually I can go back that's fine actually if we overlaid them it would show better but this is there are two separate ion extractions as well this is actually four I think it's 419 on the LCMS you want to take a look at this you know starting off at you know 10 and a half going to 12 and here we got 12 to like 14 for the DIDP so how long is your total run time for the LCMS LCMS total run time including the post run I believe is 22 minutes something like that 24 minutes yep yep yep depending on what what we have in the child we use that as well okay thank you okay so have you done correlation with other labs or within your labs? yeah we're involved in internal correlations with our own company as well as external correlations to see if the methodologies that other labs are using are comparable with ours so yeah that's part of our quality control alright let's thank Chris thank you up next we have Luke Ackerman from the FDA thank you all for coming and thanks for okay okay so today I'm going to be discussing Dart MS or direct analysis in real time which is an ambient ionization mass spectrometry technique as a possible screening method for phthalates specifically we've been doing a lot of work on food contact polymers and looking for rapid screening techniques phthalates just happens to be one of the model additives that we look at whenever we evaluate methodologies so I thought perhaps I could share a little bit of our work with you here and I think just to kind of give you the conclusion I think Dart can be used as a very effective screening technique it has limited capabilities otherwise but I'll be going over the details of that and I work at the center for food safety and applied nutrition at the FDA research arm of the FDA so I'm not conducting any regulatory enforcement actions at our facility we're developing methodology and trying to advance the science at the agency so I won't be discussing any legal matters either so how does Dart work well Dart is as I said an ambient ionization technique it's just one commercial application of many different gosh what else nano desi supersonic electro spray ionization these are all the academic terms for various techniques of producing gas phase ions off of condensed materials so solids and liquids and turning them into gas phase ions such that mass spectrometry can be used to help analyze the stuff in question in this particular case Dart is unique and was probably granted a patent it is a plasma it generates a plasma of helium in the glow discharge area of the diagram here and then it quenches them on the grounded electrodes in further downstream it then subsequently heats that gas and the quenched plasma generates excited helium or metastable helium which is just an electron shifted up in energy level in the orbitals of helium once you heat this helium stream you now have a hot excited helium it exits the Dart source and into the atmosphere of the lab where you place your sample the atmosphere in the lab not surprisingly contains water vapor and water has a good match with the energy of the excited helium and it produces protonated water clusters and these protonated water clusters in turn ionize any gas phase molecules in the vicinity of the source there's been lots of studies on how big of a gap you need and for different compounds different distances between the exit of the Dart and the entrance of the mass spectrometer will yield better results but in general you're protonating whatever gas phase ions you can get and the thermal heating of the helium gas also allows thermal transfer to the surfaces so you're doing thermal desorption so Dart is thermal desorption atmospheric chemical ionization similar to the negative and positive chemical ionization possibilities with standard GCMS techniques but this is done in the open lab atmosphere and the advantage of that is it allows you to stick odd shaped objects in front of a mass spectrometer so you don't have to do sample preparation in order to get some sort of representative spectra of the surfaces of the samples that you're looking at as you can imagine that creates a whole host of other questions as what kinds of surfaces are all close what compounds how hot those are all parameters that have to be dictated according to each application but this is the basic approach that Dart takes to producing gas phase ions for mass spectrometry and what does Dart actually look like in practice we've generated a little movie here one of the advantages of Dart is how fast it is and this will show you a mass spectrum of a little sample that was introduced maybe saw in an upper right-hand corner a little glass capillary came down and was stuck into the Dart stream and you see a peak there at m over z 127 this was our demonstration of how melamine is detected in milk products so we coated a little glass capillary with a milk product and if there was melamine present it was ionized and m over z 127 mass of melamine plus h or plus proton jumps up on the screen there and you can see the time frame there in the order of 30 seconds or less per sample so you get an immediate spike in signal and it drops off rather quickly as fast as your robot can go and pick up another sample or you can go grab another sample it returns to baseline and you're ready for the next analysis so what does Dart MS measure well it measures mass spectrum because Dart is just an ionization technique for a mass spectrometer in this case we interfaced Dart with a water's ultima which is just a simple triple quad and we looked at the mass spectrum because of the way that we set up the the Dart and because of the mass spec you can see a couple of diagnostic ions here for in this case diethylhexyl phthalate you see the molecular ion the m plus h some of the characteristic fragment ions we were running this under some slightly fragmenting conditions and if you use spectrometers you get different results so the bottom trace here is the mass spectrum of the same standard using a Joel acutoff which is just a low end time of flight mass spectrometer about 170k for one of those and as you can see there even in the presence of a mixed standard this happened to have a little dinp a little didp and some de-hp you can see the dominant ion because of the chime on the trace that I picked well the octal phthalate I should say signal and the nice thing about using a toff is of course the added mass accuracy as you can see you get high mass accuracy and to a certain degree a good mass resolution there's certainly other mass spectrometers that can achieve better but the point being that when you're doing direct analysis or atmospheric ionization analysis you're going to get mixtures and you no longer need to help separate your chemicals previous to analysis so anything that you can do to help you differentiate one molecule from another can help and accurate and high res mass back can help with that so what does a dart ms response look like well you're not using chromatography so you don't have a chromatogram now you just have a time trace this is a typical total ion current it's not a total ion chromatogram there's no chromatography in front of a dart and in this case this is from a triple quad the waters that I was mentioning and you see the tick rise and fall but rather noisily and rather irreproducibly but when you go to look at a particular compound of interest you see that the trace is a lot more easy to understand and in this case the DEHP standard at various concentrations gave a very predictable response as a function of time so again we're looking at a mass spectrum as a function of time so what are all the different dart configurations that I've been discussing well we started off in our laboratory with a first generation dart which is kind of a clunky looking thing and in this particular case we are using a robotic autosampler to place little strips of particular polymers into the dart stream you'll see a little glass capillary tube sticking out that helps transfer the gas ionization region into the mass spectrometer and because the manufacturer of the dart ion source had not interfaced this particular mass spectrometer we literally just terminated the glass tube immediately adjacent to the mass spectrometer inlet it wasn't even on axis and so in that case our sensitivity was quite poor and the results that we generate with this configuration weren't really translatable to a lot of the people who had purchased the Joel Akutoffs which was the current company of the ion source manufacturer dart is manufactured by IonSense out of PeabodyMass we subsequently purchased two other configurations a dart orbitrap and a dart Akutoff so this is the Joel Akutoff and the newest version of their dart source that goes with the Akutoff again as you can see here the silver cone inlet the silver and white cones coming off the dart source that's where the gas exits we happen to be running a little homemade rail linear rail through the dart source at this particular case we were mapping some chemical concentrations across the surface of a piece of packaging and we just thought we'd try a little linear rail in that particular case but as you can see the dart source can be moved you can lower it angle it at different angles and so you really can adjust the dart source to fit whatever geometries you need for your particular samples the beautiful thing about dart is you can stick really crazy shaped objects in front of it and you can still get reproducible mass spectrum you don't have to mill you don't have to homogenize now you lose all that information you lose reproducibility and quantitation and the like because you're not looking at the same object with in a typical regulatory analysis but if you're going to screen stuff and if you want to have an idea of what where on a particular object phthalates or any other compound might be for the surfaces of those objects then dart will allow you to do that by looking at only one part of the object at a time or if you really want to cover every surface you could systematically do there but in either case dart is a quick way for a laboratory worker to look at a sample without having to do prep and without cross-contamination because there's nothing to contaminate here the entrance to the mass spectrometer which is going to be less contaminated than if you inject a plug of solvent from a GC system so it's got some advantages that way here's another angle of packaging you can do it manually you can use linear rails they've got transmission configurations where you can deposit extracts onto little grid wire meshes and the helium flows through the wire mesh and dissolves whatever was on the solution that deposited on the grid there's lots of configurations for introducing samples via dart so can it quantitate that's a big question so these are some just some straight up solvent calibration solutions and just measuring the peak height of the mass spec signal for diethylhexyl phthalate very simple task not at all linear not very reproducible at that and the reason is is that various amounts of materials are deposited onto the tip of the glass capillary that we used to sample the liquids in GC you're using a syringe to deliver exactly one microliter every single time whereas in dart the current auto sampler configuration dips a melting point capillary into your liquid solution and depending on how clean that capillary is how far it's dipped in how quickly you know what was there a drop shaken off the capillary in the robot's movements over to the dart you get various quantities deposited on the surface and then how fast you scan it through there disturbances in the lab that shifted gas flow patterns it desorbs a different amount of material off of that and a different amount makes it into the mass spec and a different ion signal but one way to account for all those differences is internal standards you know in GCMS we often do that you get very nice linear calibration curves when you internal standard normalize your responses can also it also helps you with troubleshooting issues in your methodology recovery correction you don't have to take you don't have to do quantitative transfers all of those things are helped by internal standards and so is dark calibration when you include just two internal standards in this case for these suite of phthalates that you see on the bottom right I included a labeled standard of diethyl phthalate and a label of diethyl hexyl phthalate and then normalized to one of those two the relative response is quite linear and you know the squares that we go on about all better than 0.995 but when you look at them on a log scale you see GCMS is still much more linear than dark and that makes sense you're just a lot more controlled sample introduction a lot more controlled ion transfer into the inlet not to say that it can't be useful you can't good semi-quant data especially out of liquid solutions but it's never going to be the same as GCMS and I would like to point out this is over five orders of magnitude so clearly a very broad range here and but still GCMS is the gold standard for quantitation so it can quantitate in solutions but we're not looking at solutions we're looking at surfaces so how sensitive is Dart when it comes to looking at surfaces well we had another project where we were transferring from the print side of a printed piece of food packaging to the food contact side of an adjacent piece of food packaging it's called set-off and we decided to look at Ergicure 184 a common photo initiator by producing standard samples with various levels of the photo initiator on the surface of the polymer this I believe was an LDPE film and what we saw was just some of the concentrations of the samples that we prepared to see that the surficial concentrations which Dart is able to differentiate from the blanks is right around 0.25 nanograms per square centimeter so definitely down in the low fractions of a hundredth of a percent level you're going to see surficial concentrations orders of magnitude higher if you're looking at even 0.1 percent phthalate in a polymer so clearly it's way more than sensitive enough for these types of compounds we haven't done this exact procedure with phthalates but the responsiveness of Dart and Ergicure to Ergicure and phthalates is very comparable they're both small molecules I think this vapor pressure of this compound is almost exactly the same as diethylhexylphthalate vapor pressure is a pretty big factor when you're thermally disorbing something off of a surface so we see pretty comparable responses of sensitivity for phthalates to be very comparable and then we also need the technique to identify the phthalates and in particular the nice thing about Dart as I showed you back at the beginning is it produces almost exclusively molecular ion that is M plus H one of the reasons why we like to use chemical ionization it's a much softer ionization technique it doesn't give you as reproducible mass spectrum specific molecular ion to work with you can also include a little ammonia hydroxide in the lab environment or swab it onto the surface of a polymer and you can produce ammonia instead of protonated molecule ions so instead of M plus H it's M plus NH4 it's often a common technique to look at both of those ions as a way of confirming that the 195 ion you see is really corresponding to that or whatnot and as as we all know this M plus H ion is very useful the fragment ions help a little bit more but isomers produce all the same ions in theory and they should because they're isomers they're positional isomers so diethylhexyl phthalate and octyl phthalate are a perfect example and other work by other folks including Rothenbacher and Schwach it published in rapid communication and mass spec took a close look at this I think kooky out of hungry also did the same thing in international mass spec journal and what they found is the 261 to 279 the the minus ethylene and minus ethanol fragments are produced in differential ratios and that really froze up and you can differentiate ethylhexyl and octyl that way the trouble is and pretty much any isomeric um octyl phthalate can do that to varying degrees and what if you have mixtures of these phthalates so when we looked at a particular P.P. and L.D.P.E. film for various phthalates we see the molecular ions of ethylhexyl or an octyl and also diisononal phthalate when we look at the ion ratios of the fragments 279 clearly exceeds 261 suggesting it's ethylhexyl phthalate um you know the ratio is 1.7 but in the standards that ratio was 3.3 so what does this mean and in our initial analysis of this sample which is the cookie bar at the bottom of the chart there we said that we thought there's low hundreds of nanograms per square centimeter of diethylhexyl phthalate and we didn't think there was an octyl phthalate um we then subjected the same samples to extraction L.C.M.S. analysis here's the chromatogram and we focused a lot on the isononal phthalate in that particular case and what we saw was um a very odd shape L.C.M.S. peak we expect nice sharp L.C.M.S. peaks for the isononal and we suspected it was an n-octodesyl phthalate which would be a particular um structural isomer of of the nonal phthalates and sure enough when we spike it with an additional 100 ppb of dinup that small shoulder goes up is the shoulder and n-octodesyl phthalate was the dominant compound in this particular package similar analysis of ethylhexyl and n-octyl in this particular uh and compound the cookie bar showed as you can see on the bottom that there was actually nearly equivalent concentrations of n-octyl phthalate and ethylhexyl phthalate that's why our ion ratios of those fragments switched by Dart.ms and we have another sample. So what can we conclude from all of this? Well sample positioning affects Dart response. It's more than sensitive enough for um phthalate analysis. It can be quantitative in certain situations when you use internal standards and especially with solvents. It can be semi-quantitative off of polymer surfaces especially if you can deposit an internal method. They've been working on printing standards on surfaces um and it cannot reliably identify different isomers of different phthalates not in mixtures. Um and one last note is the time response of Dart. So this is a typical tick and this is what that this is a total ion current of a typical standard for Dart and what ethyl acetate excuse me um came off the surface of the tip and then the diethylhexylphthalate in the last second and the time scales of these are about one six to one ninth of a second. So you need to be measuring your mass spectral traces at six to nine hertz at least so cycles per second in order for Dart to really be useful uh in capturing the actual quant signal. So these are all really deep in mind but again the main points it cannot reliably identify the different isomers it can quant under certain circumstances but it can certainly screen for the presence of various phthalates. Thank you. Do you have any questions? I have a one question. When you're doing solids can you comment on the spatial resolution of this technique? Well if you can reliably move what we were doing with that little linear rail and the surface then you can assign a spatial map that will be our next publication where we reproduce the set off image of a print on the adjacent piece of packaging you can see the same trademark logos and you know images but just recreated in a chemical map. So here let's Well so the width of the dart beam is on the order of about a centimeter or three quarters of a centimeter somewhere around there I know that they're working on different iterations that have narrower helium beam widths but really double or triple that and that's probably double that and that's probably the smallest scale of resolution you're going to achieve and then of course you can screw it up really fast or reproducibly stuff like that sorry let's thank Luke yeah why don't we go ahead we'll break for lunch now and we'll start back up again at one o'clock