Session G – Laboratory and process optimisation

Session chair: Paul Krüsemann, Eurofins Materials Science (NL)

The session “Laboratory and process optimisation” will take place on Wednesday 19 June 2019 and Thursday 20 June 2019 with the following lectures:

G.01Improve your processes, treat your surfaces by Jeroen de Jong, Interscience (NL)

Sampling and analysis of trace and active compounds is getting more and more important. The material choice of a sample pathway can have an effect on the sample composition. Due to adsorptive effects or catalytic reactions of stainless steel surfaces, the integrity of a sample is sometimes difficult to maintain. Coating of a surface with an inert layer prevents chemical reactions and also outgassing of the substrate at the same time. Using a technique called chemical vapour deposition, several different coatings with their own characteristics can be applied on several types of surfaces like glass, steel and ceramics. Different examples and applications will be shown, e.g. coated tubing provides nearly instant sample response, requiring no “priming” of the analytical system. PPB level of sulfur containing compounds are that are sampled and contained in a surface treated sample vessel stay airborne for days instead of hours compared with vessels that are untreated. The inert coatings even prevent corrosive attack and extend the life of components by up to 10x or more.

G.02Predicting separations using ProEZGC GC modelling: a new free web-tool offering the most cost-effective way to optimize and speed up analysis, without using an instrument by Jaap de Zeeuw, Restek (NL)

There is always great interest in faster GC separations. This can be achieved in different ways. From using narrow bore and shorter columns to using faster carrier gas/flow or temperature program. The oven does have a limitation in program rate. By reducing oven volume, one can apply faster programming, reducing run time. Additionally, it is possible to predict (model) separations using a free web-tool called ProEZGC. With this tool one can optimize separations of your components of interest. You can change type of carrier gas, column dimensions, flow and temperature programming and immediate see what it will do for your separations. For many applications, optimization of analysis time is a matter of making the model and enter the new conditions in your GC. Results are remarkably close.

G.03Recent innovations of nanoplus devices and corresponding spectroscopic application by Lars Hildebrandt, nanoplus (DE)

Among the numerous methods for gas analysis, tunable diode laser absorption spectroscopy (TDLAS) excels, e.g. in accuracy and speed, allowing real-time measurements with detection limits down to the ppb level. With the interband cascade laser (ICL) approach, the entire wavelength range between 3 and 6 microns became commercially accessible in 2014. In this spectral region the fundamental absorption features of many technologically and industrially relevant gas species are located, including e.g. formaldehyde, nitric oxide as well as hydrocarbons like methane. In this talk, we report on the newest developments of these application-grade distributed feed back (DFB) lasers based on the ICL concept with tuning ranges above 20 nm, high sidemode suppression ratios and low power consumption. These laser devices enable high accuracy sensors for industrial process analysis as well as for applications related to health, safety and environment. In addition, we will give a general overview for the whole TDLAS-field, plus the latest product developments of nanoplus concerning MIR-IC-LEDs.

G.04Performance evaluations of inferential devices by Adam Lomax, EffecTech (GB)

The performance of online instruments used to measure the composition and properties of natural gas is critical to industry. The existing standard ISO 10723 ‘Performance evaluation for analytical systems’ is used to assess the performance of online gas chromatographs. However, the method described in ISO 10723 is only applicable to instruments with simple well-defined measurement equations which can be modelled. A new generation of instruments which do not have easily modelled measurement equations are now being used more widely; these instruments have complex algorithms which infer properties by direct measurement of gas properties. Often these methods are proprietary and not published by the manufacturers meaning producing a model of instrument performance is not practical or even possible. This new approach is designed to allow the performance of those instruments to be assessed using a black box approach which does not require knowledge of the measurement equation. The principle used in this approach is to measure a large number of well-defined reference gases. The measurement of a large number of reference gases would be logistically difficult, costly and time-consuming, so instead a small number of well-defined reference gas mixtures can be used to produce a large number of dynamic reference gases using a dynamic blending system. The very rapid response time (typically less than a minute) of inferential devices allows for a large sample of reference gases to be measured in a timely manner. The details of how the performance evaluation is carried out, along with the background of why the new performance evaluation is necessary, will be presented with real data from performance evaluations of inferential devices. This work is part of the development of a new standard, ISO 23568 ‘Natural gas – Measurement of properties – Inferential devices – Performance evaluation’ under ISO/TC 193.

G.05Opportunities of advanced process control (APC) applications by Martin van der Veer, Shell (NL)

Over the past years, computing power has significantly increased. This opened up the possibilities to extend process control applications from a traditional first order control loop to advanced process control (APC) applications for process unit optimization making it more efficient (lower energy use) and/or improve capacity (higher yield). The development of these APC systems increased both the importance and the amount of product quality data. This presentation will outline how new measurement technologies like sensors and analyzer based on chemometric modelling can be used in addition to the existing installed base. It focuses on the impact for both the online analyzer and laboratory organizations to ensure model development, measurement performance and response time meets the new operational requirements for applying APC.

G.06Development of a compact trace-moisture sensor based on cavity ring-down spectroscopy by Hisashi Abe, AIST (JP)

We developed a compact trace-moisture sensor based on cavity ring-down spectroscopy (CRDS). The sensor has the dimensions of 25 cm × 20 cm × 20 cm and the mass of approximately 3 kg. The sensor unit includes ring-down cavity, fiber optic collimator, and photo diode. A continuous-wave distributed-feedback laser diode placed outside the sensor unit was connected to the collimator using an optical fiber. The SI-traceable standard gas mixture of water in nitrogen generated using a magnetic suspension balance/diffusion-tube humidity generator at National Metrology Institute of Japan (NMIJ) was introduced into the ring-down cavity. The laser light exited from the collimator was injected into the ring-down cavity. The ring-down signal detected with the photo diode was transferred to a digitizer. The digitized signal was transferred to a personal computer and fitted with an exponential function to obtain the ring-down time. The mole fraction of water in nitrogen inside the cavity calculated from the ring-down time was in good agreement with the standard value in the range of 10 ppb to 1 000 ppb. We attained a standard deviation of 1 ppb at 10 ppb for an averaging time of 20 min.

G.07Improving sensitivity and speed of response of parts-per-trillion level moisture detection in UHP gases by Florian Adler, Tiger Optics (US)

Semiconductor device manufacturers continually demand higher quality materials and gases for their processes. In turn, the need for more sensitive and precise analytical technologies requires instrument makers to continuously improve detection capabilities. Laser-based gas analyzers utilizing cavity ring-down spectroscopy (CRDS) have become the preeminent equipment for monitoring H2O impurities in bulk gases for semiconductor manufacturing due to their combination of high sensitivity and speed of response. The requirements of the industry, however, represent a challenge for even the most advanced analytical technology. We present data from the latest generation of CRDS moisture analyzers with a detection limit below 100 ppt in typical bulk gases. Improved optical and electronic noise owing to an entirely new analyzer platform allow for a dramatic improvement in sensitivity. However, H2O impurities in the gas at such low levels post many additional challenges that must be addressed to provide this level of performance in real-world applications. We will present on the challenges that low-level H2O pose to the analysis. For instance, moisture’s tendency to stick to surfaces requires special consideration regarding materials to maintain good speed of response at sub-ppb levels. In addition, minute outgassing rates of stainless-steel gas lines can no longer be neglected, particularly when analyzing H2O in oxygen. These contributions must be studied and considered in the analysis. We will also elude to the important role of CRDS’s absolute zero, which does not rely on a zero reference. Small but real contributions to the moisture in the gas, which have adverse effects on the industrial process, can easily be masked by a false zero that other technologies frequently rely on.

G.08Extremely quick response of a ball SAW trace moisture sensor by Yusuke Tsukuhara, Ball Wave (JP)

We have developed a novel trace moisture sensor utilizing a unique characteristic of surface acoustic waves (SAW) propagating on a spherical single-crystal quartz. The ball SAW sensor responds to an impulsive injection of moisture into a pipeline of dry gas in less than a second. The response time is less than a second, too, when a dry gas is suddenly introduced into a wet gas flow. This characteristic enables, for the first time in industry, the real-time monitoring of trace moisture in the pipeline system, in the factory environment, and even in the vacuum dryers. By the real-time monitoring of trace moisture, we can observe accurately how the inner surface of a pipeline and a chamber becomes dry while a dry gas is introduced for the purpose of dry-down. Another interesting application is the evaluation of a dryer that is to remove the moisture from a specialty gas. When inert gas containing water molecules flows into a metal pipe, the water molecules cannot exit instantaneously from the outlet of the pipe but are captured at adsorption sites on the inner surface of the pipe until most of the sites are occupied. A theoretical model and a subsequent experiment using the ball SAW sensor show that the delay time depends on the amount of moisture level; the higher the moisture-level, the shorter the delay time. Based on the result, we developed an equipment for the validation of a standard moisture generator to be used in the field measurement such as in factories and pipe lines.

G.09Automated real-time monitoring of trace compounds with PTR-MS by Bea Rosenkranz, IONICON Analytik (AT)

Proton-transfer-reaction – mass spectrometry (PTR-MS) is a well-established technology for real-time quantification of trace compounds in atmospheric chemistry, food and flavor research, indoor air monitoring, and many other fields of application. Particularly, time-of-flight (TOF) based instruments in combination with well-controlled ion chemistry have proven not only to be extremely rapid and sensitive, but also highly selective. However, so far controlling the ion-chemistry, i.e. changing the reagent ions for chemical ionization of the analytes and modifying the reduced electric field (E/N), and interpreting the product ions had to be performed by a skilled scientific operator. Here, we present a novel PTR-TOFMS device, which is housed in two interconnected 19″ cubic racks with a total weight of about 80 kg, for maximum flexibility. By analyzing a certified TO-14a gas standard, we found sensitivity values of about 2 000 cps/ppbv and a limit of detection for 1 s integration time of 10 pptv. Most importantly, the instrument is equipped with a fully automated control and data evaluation software, which adjusts the ion chemistry and performs pattern matching algorithms according to substances entries in a database. For the very first time, we utilize a novel method of producing NH4+ reagent ions (in addition to H3O+, NO+ and O2+), which does not need any ammonia as source gas, but forms NH4+ at very high purity levels of 98 % via reactions of N2 and H2O in a hollow cathode ion source. The instrumental setup including the automated data evaluation software has successfully been employed for online monitoring and quantification of e.g. BTEX in clean-room environments. Moreover, we present very recent data obtained at a specialized chemical warfare agent (CWA) laboratory, where we could detect and unambiguously identify all investigated CWAs, including sarin, sulfur mustard, soman and VX, thus making the instrument an ideal tool for homeland security.

G.10Sub-pptv sensitivity in on-line analysis of VOCs by a novel PTR-ToF reaction cell superposing DC and RF fields by Felipe Lopez, Tofwerk (CH)

Proton transfer reaction – mass spectrometry (PTR-MS) is a technique of increasing popularity for online monitoring of volatile organic compounds (VOCs). The working principle of PTR-MS is chemical ionization upon reactions between primary H3O+ ions and the analyte VOCs. Such reactions occur within a reaction cell having very stable and controlled electric fields, pressure and temperature. The electric fields within the cell move the primary and product ions towards the detector. Conventional reaction cell designs are based on a linear, DC field along the drift axis. Ion losses occur at the reaction chamber cell walls because of diffusion and scattering. This problem affects the instrumental sensitivities and limits of detection (LOD). In order to overcome the problem of ion loss, a novel approach to the drift cell design is proposed. The solution involves adding oscillating RF fields on top of the linear field. In order to allow this addition, the material of the reaction cell has also been changed, introducing the use of resistive glass. The ions are not only moved along the cell by the DC field, but are also simultaneously focused by the oscillating RF fields, so that more ions can reach the detector. The chemistry of ion-molecule reactions within the cell is similar to that of the old design, while the spatial distribution of primary and product ions is changed to achieve the extra focusing. The net gain in sensitivity exceeds one order of magnitude and previous boundaries in LODs are overcome thus providing unprecedented performances for a PTR-MS. The new reaction chamber has been coupled to state-of-the-art mass analyzers, such as time of flight mass spectrometers reaching 15 000 in mass resolution. In the case of benzene, the achieved sensitivity is 20 000 cps/ppbv and the LODs are < 1 pptv and < 10 pptv in 1 s and 1 min integration time, respectively.

G.11Systematic evaluation of condensation particle counters by Krzysztof Ciupek, NPL (GB)

Condensation particle counters (CPCs) are widely used to measure aerosol particle number concentration and, when combined with a size-selection device, to determine their size distribution. When calibrating these instruments, two main properties are being tested: the linearity of the measurement response with concentration, and the particle detection efficiency at different particle sizes. The first feature provides one or more calibration factors to improve the accuracy of the CPC’s results over a concentration range (concentration dependence) whereas the latter shows the instrument’s size dependence. Determinations of both properties are usually provided by the manufacturer, however, their values depend on the particle material and might change over a period of using the instrument. Additionally, the values may be stated for each model, but they might be different for each particular instrument, and this will have direct impact on the quality of the measurements. In this study, a comparison of three CPC models is shown with the focus on their linearity. The results are juxtaposed with the manufacturer’s specification, showing limitations in the use of such instruments and how their properties might vary among the same type of instruments. Each CPC was calibrated in the National Physical Laboratory (NPL), the UK’s National Measurement Institute (NMI). The NPL airborne nanoparticle metrology laboratory maintains a UKAS accredited facility to ISO/IEC 17025, for the calibration of such instruments. This is conducted through comparison with NPL’s standard aerosol electrometer that has been calibrated against electrical standards. Similarly, flow rates are also calibrated against NPL flow standards providing full traceability. The results of this study has been compared with the criteria that CPC instrument have to fulfil in accordance with the requirements given by ISO 27891:2015 ‘Aerosol particle number concentration – Calibration of condensation particle counters’.


+31 (0)15 2 690 147