Instrument and Software Development
Understanding the role of atmospheric particles requires the ability to model and measure particles both in natural environments and in detailed laboratory studies. CAS for many years has invested extensively in the development of the tools needed to achieve this. In the following sections, the development of instruments and associated software are discussed. Detailed information can be accessed by following the links in the text. CAS owns many more instruments than are listed below, but these are more commercial instruments that CAS has not had extensive involvement in developing. Full details of all CAS equipment can be found in the Research Tools section. For details on the development of modelling tools, click on the modelling tools development link.
Background
As the group’s interest in aerosol science grew, so did the suite of instruments used to study the various properties of aerosols. The most basic parameter that can be measured is number. How many particles are there at any given time? One of the main instruments used to count aerosol particles is a Condensation Particle Counter (CPC), or CPC. However, merely knowing the total number of particles is often not sufficient. Far more instructive is to know how many particles there are as a function of size.
Instrumentation – Physical Properties
The earliest developments within the group were in the area of number size distribution measurements: How many particles are there at a given size? Historically, all the aerosol sizing instrumentation within the group was based on optical methods employing Mie theory. These methods are limited to a minimum size of ~100nm. During a project run at Great Dunn Fell in 1992, new particle formation was observed by a German group participating in the experiment. The group from Leipzig were using a Differential Mobility Analyser (DMA) and CPC arrangement to monitor particles as small as 3nm. This DMA-CPC arrangement is known as a Differential Mobility Particle Sizer (DMPS), which separates out charged particles in the DMA, which are then subsequently counted by the CPC. From this, a number-size distribution can be generated.
This data set was the starting point of a PhD project in 1995 which included the development of a DMPS system. Paul I Williams, now the NCAS FGAM aerosol instrument scientist at CAS, built the first UK DMPS system. Since 1995, the DMPS system has been improved and re-built several times. A photo can be seen in fig 1. Furthermore, the original DMPS system consisted of 2 DMAs and 1 CPC. There are now more than 14 CPCs and 14 DMAs within the group.
The DMPS system works by selecting a single mobility particle (which is inverted to a diameter in the analysis), counting the number of particles at that size and then stepping to the next size. The mobility selected by a DMA is highly mono-disperse. This feature of a DMA is extremely useful as it can be used to generate single sized particles as well as building a size distribution. The ability to size select particles lead to the development of the Hydroscopic Tandem Differential Mobility Analyser, HTDMA. A HTDMA is used to investigate the water uptake properties aerosol particles.
The HTDMA uses two DMA in series and a single counter. Aerosol particles are drawn into the instrument and dried so that the RH is less than 10%. They are passed to the first DMA, which selects a single size, D0. These particles are then passed to the second DMA via a humidifier, which raises the RH to ~90%. Depending on the size and chemical composition of the particles, they may take up water which will cause them to grow. The second DMA then scans, the same as a DMPS, to determine the new sizes, D1. The ratio of D1/D0 is known as the Growth Factor (GF). If the particles are hydrophobic (do not absorb water) then the GF will = 1. If they hydroscopic (do absorb water), then the GF > 1. Once the second DMA has completed the scan, the first DMA changes D0, and the procedure is repeated so that the GF as a function of size can be determined.
Instrumentation – Chemical Properties
CAS for many years had the ability to measure the bulk chemical properties of aerosol particles using a variety of impactor and filter methods. These included using filter papers with different pore sizes, and a stacked cascade impactor such as the Sierra Anderson impactors. Both methods require off line analysis and generally have long (several hours) integration times. In 1999, Prof. Hugh Coe received a grant which included the purchase of an Aerosol Mass Spectrometer from Aerodyne Research Inc. This system, a quadruple based mass spectrometer (Q-AMS), allowed real time, size resolved chemical composition measurements of the sub-micron, none refractory aerosol to be made. None refractory refers to volatile and semi volatile material such as sulphate, nitrate and organics, but excludes sea salt, most metals and mineral material such as dust. Two of the PhD students working with the AMS, James Allan and Rami Alfarra, are now employed at Manchester by NCAS.
Although the AMS is a commercial instrument, the AMS CAS purchased in 1999 was number 2 of 2. CAS was instrumental in field testing and validating the capability of the AMS as well as developing the analysis software for the instrument. This development work is described in more detail on our Aerosol Mass Spectrometry page.
In 2002 (I think), a second Q-AMS was bought for the UK research aircraft, the BAE 146-301. The Q-AMS, along with a 3025 UCPC, were fitted to a rack and were the first aerosol sampling system on the aircraft. Further details are given in the Aircraft Sampling of Aerosol section. Since then, both the ground based and aircraft based Q-AMS have been upgraded to the Compact Time-of-Flight AMS (C-ToF-AMS), and a new High Resolution Time-of-Flight AMS (HR-ToF-AMS) has been purchased.
Software Development
Instrument development has two aspects: Hardware and software. It is equally important to develop the software tools necessary to control and interpret the data. Both the DMPS and the HTDMA were bespoke builds, and as such, they required custom control software. As the number and complexity of the both the DMPS and HTDMA increased as well as the different applications and configurations, then a similar, if not more, amount of time and effort was spent developing the data acquisition software. The control software for these instruments is written in LabVIEW from National Instruments (http://www.ni.com/labview/). LabVIEW is a graphical based programming language that now forms a large part of the group’s data acquisition software. Many of the commercial instruments now have code written by members of the group in LabVIEW. One of the latest developments has been to write a central logging program that can record numerous aerosol and meteorological instruments on one PC.
Comparatively speaking, the AMS is orders of magnitude more complex and produces much more data than any of the DMA based instruments. For example, the new HR-ToF-AMS generates over 10GB of data in a 3-4 week campaign. The data needs to be post processed before it can be analysed or even plotted. Unlike the data from the DMPS and HTDMA, this is not the sort of task that a program like EXCEL can perform. This lead to the single largest software development in the aerosol group, which was undertaken by James Allan. He was instrumental in developing the analysis code for the Q-AMS and the newer time-of-flight AMS. The scripts, written in IGOR PRO by Wavemetrics (http://www.wavemetrics.com), are used by nearly all AMS users world wide.