Page 291 - Pressure Swing Adsorption

: .i.l
I
' EXTENSIONS OF THE PSA CONCEPT 269
268 PRESSURE SWING ADSORPTION
gas phase ( y ).
conventional. PSA system 1s somewhat remote, since independent control of
dq* RT da*
pressure and flow. which is a key feature of the parametnc pump, 1s lacking (7.4)
in PSA. A true pressure swing parametric pump has, however, been demon• de - P dy
strated by Keller and Kuo,' who called their process the molecular gate. The Thus at high pressure da* /de is relatively smaller and we 1s correspondingly
essentiai components of such a system are shown schematically in Figure 7.2. larger. The anaiogy between the thermal parametnc ,pump dnven by temper-
The pistons. which are of unequal displacement, are coupled so that a ature vanation and the gas-phase system driven by a pressure variation ,s
constant phase angle 1s mamtamed. The synchronized movement of the therefore clear.
pistons is adjusted so that the gas flows upwards through the bed at high The maximum vanation m the volume between the 01stons occurs when
Pressure and downwards through the bed at low pressure but, smce the the difference m phase angle is 45° If kinetic effects are unimportant this
displacments are unequal, there is a net flow towards the smaller- piston. condition should give the maximum variation in pressure and therefore the
Fe~d is Introduced near the center of the adsorbent bed (the optimal point best performance. For air separatidn over a zeolite adsorbent the separatron
ctepencts on the feed and oroctuct compositions), while a fraction of the gas 1s factor (aN /Oi) IS about 3.5. Assuming linear isotherms, this means that the
discharged as product from each cylinder at each stroke. The velocity of the capacity of the bed 1s about three and oncwhalf times as great for nitrogen as
concentration wavefront m the adsorbent bed ls governed by Eo. 7.1. The I for oxygen. For any given pressure change the ratio of the nitrogen and
1 sotherm siope dq* /de may be expressed in terms of the mole fraction m the oxygen voJumes reomred to oressunze the bed muSt lie m this ratio. The
expenmental data, shown m Figure 7.3, are m approximate accordance with
l
l
0·...._.9Qo/., N (plus or9on)
2 0· -90°1. N;z(p1t,, orqon)
0~- 90•1., Oz I- 0·-90¾ Oz
c z
w
• 100
i' I ill 100
~ ~
~
<t 0
"'
0
~
" 80
~
0 0 " 80
0
J!N ~
0 I ;;N
" ;J
•
~ 60 8 60
i .!!
N \
z z N '
~ •o ~ I
8 •o I I \
,.:
" " I
I-
~
:E 20 > I
;:
u , ~ u 20
::,
'O 0
e
0
a. "' ..
0
0 0.2 04 06 0.8 lO 0
60' :w o· :w 60'
Shor! •Pis1on Stroke /Lon9 · PiS1on Stfl>M
Log Leod
/a)
PHASE ANGLE,SHORT VS LONG PISTON
Figure 7.3 Performance of the mo1ecuiar gate m atr separation showmg (a) the effect
of piston stoke and (b) the effect of phase angle on productivity. Both products at (b)
~ 90% purity; m (a) the phase cycle 1s 45°. and in (b) the stroke ratio 1s 4: 1. (From
Figure 7.3 (Contmued).
3
Keller and Kuo with oerm1ss1on.)