Page 305 - Pressure Swing Adsorption

I
282 PRESSURE SWING ADSORPTION EXTENSIONS OF THE PSA CONCEPT 283

66~----,~--,~-------,---,----,---,
EXPERIMENTAL RESULTS
70 Product 'J>of Period
flowrale Frequency Feeo Valve
(slcm31s1 I cps) Open
62
0 9.14 0.35 Vaned
66 • 9.14
I 6. 14.16 Vaned 50
0.35
Vaned
I • 14.16 Varied 55
58 D 18.90 C.J5 Varied
I I 62 i \ I
--
I I -- z N i 1/1
\\
£' If "' ' '
"'-54/11 z \\
!2 58
\ \
i3 ii / ~ I \
:r
!:: II I ~ \ b
II I , \
I f5 ii
u 54
I . ~
g
' I 0 0
I
0.
i ~ 50
I
I EXPERIMENTAL RESULTS
I Product ~ of Penod Feed
/ flowrate Feed Valve ConipJSition
'!, N2
/ !st.cm3/sl Open
0 9.14 50 32.2
A 9.14 35 32.2 42•~-:---=----:-'::-----:-'---'-~-"---'
D 14.16 55 32.2 0.04 0.06 0.08 0.10 0.12 0.14 0,16
I 0 14.16 50 2&.6 FRACTION Of FEED GAS RATE RECOVERED AS• PROD OCT
I
4
2
0 2 4 6 8 10 Figure 7.13 Purity-recovery profile for N -CH separation by RPSA. (From Kowler
6
and Kadlec, with permission.)
PERIOD, Seconds
Figure 7.12 Effect of cvcie freouencv on product punty for separation of
6
methane-nllrogen mixture by RPSA. (From Kowler and Kadlec, with perm1ssmn.) 7 .3.3 Air Separation by RPSA
A significant improvement m the performance of the RPSA process was
9 11
achieved by Keller and Jones, ' who mtroduced a delay step (inlet and
purity). In a tater study with the same experimental system Kowler and outlet valves closed) pnor to blowdown together with a shortened pressunza~
Kadlec 6 identified cycle freauency, duration of the feed step, and product tion step. The practical feasibility of usmg such a cyCle for air separation (to
flow rate as the ·maJor variables affecting the system performance. The effect produce oxygen) was demonstrated. At small scales of operation where
of cycling frequency on product ourity is shown in Figure 7.12. The optimal power costs are not important, such a process Is competitive with the
frequency of operation appears to be incteoendent of product flow rate. A conventional. two-column system. Doong and, Yang' 12 suc~essfully mocteied
purity-recovery plot showing the effects of other important process variables. the air separation data of Keller and Jones usmg a LDF model and showed
when operating at the optimum frequency, is shown in Figure 7.13. In this that mass transfer resistance may become rmportant under rapid cycling
study it was aiso shown that introduction of a no.flow step (both exhaust and :j. conditions. The effects of feed time, delay, and exhaust time on the punty
product valves closed) did not significantly affect the product purity. How• and recovery of oxygen are shown in Figure 7.14.
ever, such a steo reduces the exhaust flow rate, thereby increasing the overall Very short feed time means inadequate contact time for preferential
recovery of the raffinate product. adsorption of mtrogen and therefore mtrogen appears as ccintammant m the