Double Down on
We specialize in online optical measurements for the accurate analytical determination of process stream composition and properties. We design complete NIR and UV/VIS instrument systems, process probes and flow cells, and sample conditioning systems that are typically used in continuous monitoring mode, online, in real-time under the rigors of the manufacturing plant environment.
Two of our key GUIDED WAVE™ NIR UV-VIS product lines (NIR-O™ and ClearView® db) are dual (double) beam analyzer systems. We use the concept of dual beam measurements to provide the most precise, drift free data possible for use in control of a process stream or chemical reaction. In dual beam systems, the effects of changes in the light source, detectors, electronics and other associated components are removed from each measurement. This ensures that data of a very high quality is available for process control purposes.
Laboratory spectrometers can be referenced frequently and thus are not prone to performance deterioration caused by long term drift. On the other hand, process systems must operate for days to months between analyzer system references. Hence there are definite benefits gained when long term system stability is achieved. These include minimal operator intervention, increased analyzer up-time, and a reduction in the demand for lab analysis.
Figure 1: Single Beam vs. Double Beam Drift 218 Hours
There are multiple ways to implement dual beam measurements. Early methods split the light beam in two, sending the separate beams through parallel cells and then on to a pair of “matched” detectors. This is referred to as the “intensity division” method. All components after the beamsplitter are unique to that optical path. That includes optics, cells and detectors. In this case the signals are simultaneous, and a differential amplifier can be used to measure the difference in intensities. However, there are two detectors and optical trains with their independent drift characteristics. Not to mention the drift component of the splitting ratio of the beamsplitter which is critical. Nonetheless, this is the preferred method for use with pulsed light sources and it also works well for continuous wave sources.
One common alternative to this is time division. This usually involves a reflective chopping mirror system where the beam is alternately sent through parallel cells and then recombined onto a single detector. In this arrangement, the independent drift of a second detector is removed, but the separate beam paths introduce drift associated with the optics. The imperfect reflective chopper also adds some noise to the system. This method also involves the use of moving optics. Since the sample and reference beams are measured at different times, this method is best suited for continuous wave energy sources with slow drift characteristics such as tungsten-halogen lamps.
Our method differs from both methods discussed above, but still qualifies as dual beam. The design consists of intensity division at the output of the spectrometer or light source implemented via bifurcated fibers, and then time division (alternating scans) at the multiplexer/optical switch. The lamp (or source), spectrometer, detectors, and electronics are all common to the optical path. The objective of our dual beam method is to remove long term drift that commonly arises from the lamp, detector, and other components of the analyzer system. Continuously operating analyzer system drift is due mostly to relatively slow thermal changes in the lamp and detector. But, perhaps surprisingly, drift may also be compounded by the absorbance of atmospheric water vapor and CO2 present in any open air optical paths! This dual beam method addresses and removes these drift modes and does so in a cost effective way without moving optics.
Following are examples demonstrating the increased stability provided with the ClearView db dual beam design. Figure 1 shows a comparison of drift in micro-absorbance units (10-6 AU) over a 218 hour time period for a single beam versus double beam configuration. Both measurements are made using the same filter wavelength. When multiple wavelengths are combined to calculate concentrations of parameters, the magnitude of the impact of drift can be increased depending on the equation being used in the calculation. Typically a single beam system must be referenced more frequently to compensate for the changes in lamp intensity over time. This translates to increased cost since operator intervention is required for the referencing task (block sample cell / remove probe, clean and zero). The overall measurement specifications obtained from the long term measurement are shown in Table 1. It is clear that from both a drift and noise perspective that the double beam approach provides more stable drift and noise-free data for the long term. This increased stability will translate into better long-term performance for online process measurement which translates into tighter process control.