Neutron Imaging of Structured Packing
Principal Investigator: Dr. Bruce Eldridge
Graduate Research Assistant: Mike Basden
Status: Completed
Introduction
The Eldridge group has previously utilized x-ray computed tomography (CT) scans to analyze the performance of structured packing. In particular, Green was able to determine the hydraulic properties (e.g. liquid holdup and wetted area) of a lab scale column packed with Mellapak 250Y. (1)
The previous x-ray work utilized a linear array of detectors and a fan beam source for data acquisition. While a powerful and well understood method, there were several areas noted for improvement. In particular, with several elements of packing, acquisition time for a single experiment was on the order of several hours, and any transient effects of the flow field would be lost due to time-averaging. Secondly, thin films of water were difficult to examine due to the high x-ray attenuation of steel and the low x-ray attenuation of water. For these reasons, the NIST Center for Neutron Research (NCNR) was chosen for the current work. Neutrons are more readily attenuated by water than steel, and the NCNR is equipped with an area detector/cone beam system which allowed for the entire geometry of interest to be scanned in one column rotation.
At the NCNR, 3 different packings were scanned at 3 different liquid flow rates (2.2, 10.5, and 20.2 GPM/ft2) and 3 different f-factors [0.50, 1.00, 1.35 (ft/s)(lbm/ft3)0.5] were examined. Holdup profiles as a function of elevation were determined using neutron radiography.
Data Collection
A 6 in. OD (5.75 in. ID) aluminum column filled with 18 in. of packing was scanned. Three different packings were tested: Mellapak 250Y, Mellapak 500Y, and MellapakPlus752Y. The Mellapak 250Y has specific surface area of roughly 250 m2/m3. The Mellapak 500Y and MellapakPlus 752Y have a specific surface area of roughly 500 m2/m3. The key difference between the 500Y and 752Y packings involves the geometry at the joint (the interface between two layers of packing). The 752Y packing has a slight bend in the crimp channel at the joint, aligning the flow channel to the vertical direction. This bend allows for a smoother transition between packing elements.
For each liquid flow rate (2.2, 10.5, and 20.2 GPM/ft2) and each f-factor (0.50, 1.00, 1.35 [ft/s][lbm/ft3]0.5) the system was allowed ten minutes to equilibrate. At this point, neutron radiographs were acquired. A parallel neutron beam with an area detector was used to capture a 20x25cm2 area as the base rotated the column from 0° to 180°. Images were also obtained during the return rotation from 180° to 0°. An experimental schematic is shown in figure 1. The packing height (18 in.) was enough to allow for 4 half-elements of 250Y, and 2 full elements of 500Y and 752Y with a quarter-element of 250Y at the top acting to distribute the liquid at the top.
|
Figure 1: Experimental apparatus and process diagram |
Data Analysis
A modified version of the Beer-Lambert Law was used to determine the amount of water in each pixel. Additional quadratic term corrects for beam hardening due to the presence of small amounts of light water (H2O). The linear and quadratic attenuation coefficients are also functions of the concentration of heavy water. This calibration was provided by researchers at the NCNR.
Results
Figure 2 shows the results for average holdup determined using neutron radiography, x-ray CT scans, and traditional experimental methods carried out by the Separations Research Program (SRP). The results are for Mellapak 250Y at a low f-factor (i.e. in the pre-loading regime). In all cases, the holdup determined via neutron radiography is within 15% of other experimental methods. The error bars show a 95% confidence interval.
Figure 2:
Average holdup in Mellapak 250Y determined with various methods.
After determining the validity of this method, it was desired to look at the average holdup at each elevation in the half-elements of Mellapak 250Y packing. This local holdup was determined by summing the amount of water in a row of pixels, and then dividing by the empty column volume of a row of pixels. The results for Mellapak 250Y are presented in Figure 3. The bottom (0 mm) corresponds to the lower limit of the field of view on the detector, and the top (235mm) corresponds to the upper limit of the field of view.
In Figure 3, near elevations 25mm, 130mm, and near the top (235mm), the holdup increases sharply. This increase can be attributed to the joint between layers of packing. The peak of holdup is roughly 2mm above the actual location of the joint, where the largest amount of liquid accumulated. The holdup profile also exhibits an almost periodic profile of liquid holdup. The distance between peaks in the holdup profile corresponds to the distance between perforations in the packing. Using CT scans of the dry packing, it can be determined that the peaks in liquid holdup are at elevations slightly (<1mm) above the perforation elevations.
Figure 3: Local holdup in Mellapak 250Y as a function of column elevation and liquid load.
Figure 4 shows a comparison between Mellapak 500Y and MellapakPlus 752Y at 20GPM/ft2 and f=0.5(ft/s)(lb/ft3)0.5. In both packings, a periodic profile of holdup is again observed, corresponding to the distance between perforations in the packing. When reconstructed CT images are examined, the peaks of the holdup profile are found to be at an elevation slightly above the perforations. The holdup profiles of these structured packings exhibit significantly different behavior in the near-joint region. The enhanced joint geometry of the 752Y packing causes a significant decrease in holdup at this location. In both the 500Y and 752Y packings, the holdup peaks are a similar magnitude and are both observed at roughly 1.5mm above the interface between packing layers. Finally, average holdup in the bulk of the packing is slightly higher for MellapakPlus 752Y.
Figure 4: Comparison of local holdup profiles in Mellapak 500Y and Mellapak 752Y.
Figure 5 shows a comparison between Mellapak 500Y and MellapakPlus 752Y at 20GPM/ft2 and f=1.35(ft/s)(lb/ft3)0.5. The 752Y shows significantly higher holdup in the bottom element; the increased holdup can be explained by maldistribution of the vapor in the bottom element and localized flooding. Because the 500Y packing is significantly less open, the maldistribution of vapor is less noticeable. In the second element, the holdup in the 752Y is observed to be only slightly above the 500Y packing. Finally, at this higher f-factor, there is less holdup at the direct interface between 752Y packing than observed in the 500Y packing.
Figure 5: Comparison of local holdup profiles in Mellapak 500Y and Mellapak 752Y.
1. Green, C. W.; Farone, J.; Briley, J. K.; Eldridge, R. B.; Ketcham, R. A.; Nightingale, B., Novel Application of X-ray Computed Tomography: Determination of Gas/Liquid Contact Area and Liquid Holdup in Structured Packing. Industrial & Engineering Chemistry Research 2007, 46, (17), 5734-5753.