What about process analytical chemistry

Process analytical applications differ from laboratory analytical applications in that the analyzer hardware and software are in a more hostile environment, the maintenance and operations personnel in the analyzer area are typically not trained like lab technicians, and the purpose of the analyzer is most often for rapid process control rather than troubleshooting. As a result, process analytical applications tend to focus more on automation, minimization of maintenance, and long-term reliability. In this more pragmatic context, the instrument specialization function of chemometrics is used even more exclusively than the information extraction function. 

An additional feature of chemometrics that is appealing to process analytical applications, is the use of qualitative models to detect (and possibly even characterize ) faults in the analyzer s calibration, the analyzer system itself, and even sample chemistry and process dynamics. Such faults can be used to trigger preventive maintenance and troubleshoot specific analyzer or process problems — thus supporting the long-term reliability of the analyzer. Specific examples of such fault detection are given by Miller.3,4 

---

Devices. The tools described thus far in this paper present a physical view of polymer wear. What about the chemistry of the polymer wear process In studying chemistry it is desirable to have the analytical tool incorporated directly into the wear experiment. An example of such a device is presented in the schematic of Figure 4 from the authors laboratory (O. 

In the pilot plant, on a reaction scale of 24 kg of 11, we were unable to reproduce this yield. The major problem with this chemistry is the quality of the dibutylboron triflate. Fresh samples behaved well, giving in the lab around 70% yield. If the triflate is older the yield drops drastically to 50% or less. Just what older means has not been defined. Certainly after about 3-4 weeks the yield starts to drop. Triflate from different sources (quality ) also had an effect, between no reaction and similar yields to our standard one. The process takes some 6 days of pilot plant time and as it is very dilute, due to the triflate being delivered as a 1. OM solution, we were unable to significantly increase the throughput. The triflate is pyrophoric at higher concentrations or in other solvents. Consequently, we did not want to prepare it nor distil it ourselves. These reagent quality problems along with an analytical method to measure the quality still need to be solved. 

---