Download Ion Exchange - Novel Separation Processes - Lecture Notes and more Study notes Learning processes in PDF only on Docsity!
Module :
Chromatographic Separation
and Ion Exchange
Dr. Sirshendu De
Professor, Department of Chemical Engineering
Indian Institute of Technology, Kharagpur
Keywords:
Separation processes, membranes, electric field assisted separation, liquid membrane, cloud point extraction, electrophoretic separation, supercritical fluid extraction
Chromatographic Separation and Ion Exchange
Chromatography is an extremely powerful analytical tool for separating and analyzing complex mixture. The principle of chromatography is elaborated now. It constitutes a bed of particles through which a gas or liquid stream flows. The flowing phase is also known as carrier gas (in case of flowing phase is gas) or solvent (in case of flowing phase is liquid). A feed pulse (a small amount of feed) containing various solutes is introduced into the system. The solutes in the feed pulse are then separated by difference in their velocities. Therefore, the solutes emerging from the chromatographic column (the housing of the packed bed) are detected by refractive index or ultraviolet absorbance and that measurement is directly related to concentration of solutes. Based on this basic principle, chromatographic separation occurs. A typical analytical chromatograph is shown in Fig. 8.1.
Detector mete^ Flow
CO LU M
r
Feed injector
Solvent^ Pump
Typical concentration of feed about 1 mg/ml. The typical column dimension is 10cm×4.6mm, 7-5 μm WCX with 300A^0 pore size. The possible mobile phase composition is 0.05(M) phosphate with NaCl. A typical flow rate is 1.5 ml/min. The detection mode is ultraviolet. For protein solution, the detection is done at wavelength of 280 nm. Typically around 10 μl solution is injected. The peaks correspond to various components, present in the mixture. Area under the curve is then compared with calibration curve of each pure component.
Gas system:
In case of analysis in a gaseous system, a carrier gas, e.g. helium/hydrogen is used and an adsorbent is used as column. Typical adsorbents are zeolite, silica gel, activated alumina, etc.
Gas Liquid Chromatography (GLC):
An inert, porous solid is coated with viscous, high boiling liquid. This stationary liquid phase does the separation. The solid should be inert, porous and inexpensive, like, diatomaceous earth. Solute in feed can dissolve in stationary liquid phase and then vaporize in flowing gas. Separation is based on relative volatility and is essentially adsorption-stripping operation.
Disadvantages of GLC:
Slow vaporization of stationary phase changes (i) Column criteria (ii) Contaminate the products
Capillary gas Chromatography:
A glass/fused silica capillary with a coating of adsorbent or high boiling solvent on the wall is used. Since the amount of stationary phase is small, the capacity is limited. However, open capillary has little resistance to mass transfer and sharp separation is observed.
Liquid-Liquid Chromatography (LLC):
Following are the salient features of a liquid-liquid chromatograph (i) A stationary liquid phase is coated over an inert, porous solid. (ii) Separation is essentially an extraction process. (iii) Useful for separating non-volatile solutes.
Difference of LLC with modern high performance liquid
chromatography (HPLC):
(i) Inert solid is silica. Stationary liquid phase is chemically attached to the solid. Most common stationary phase is C8 or C18 compounds attached to the silica gel. Water is the common solvent. Therefore, there is no loss of stationary phase occurring by the flowing feed. (ii) Short column have very small diameter. Packings are operated at high velocity and high pressure drop. 10-25 cm long, 4-8 mm ID, packing is 3-9 μm and ΔP~1000 psi.
ρ B (^) = (^) ( 1 − ε (^) e )ρ (^) p + ε ρ e f (8.4)
Here, ρ p is particle density and ρ f is fluid density. Particle density is defined as
ρ p = (^) ( 1 − ε (^) p )ρ s (^) + ε ρ p f (8.5)
Where,
ρ s is C rystalline density after crushing, com pressed solid containing no pores
and it is obvious that
ρ f ρ s ,ρ p.
Pores are not uniform in size. Large molecules such as proteins/synthetic polymers may be sterically excluded from some of the pores. Fraction of volume of pores that a molecule can penetrate is Kd. For non-absorbing species,
d e^0 i
K V^ V
V
Where, Ve = elution volume (volume of fluid at which species exit from the column) Ve = elution volume (volume of fluid at which species exit from the column) V 0 (^) = external void volume between particles Vi = internal void volume
For small molecules that can penetrate the entire interparticle volume, Ve = Vi + V 0 (^) and Kd =1.
When molecules are large and can penetrate none of interparticle volume, Ve = V 0 (^) and Kd = 0.
Processes Involved:
Fluid containing solute flows in the void volume outside the particle. The following basic processes are involved: (i) Solute diffuses through an external film to the particle. (ii) Solute may sorb on external surface or diffuse through the stagnant fluid in the pores (most likely). If the pores are small, diffusion is hindered. (iii) Solute finds a vacant site and then sorbs by physical or electrical forces/chemical reactions. (iv) While sorbed, solute may diffuse along the surface, called surface diffusion. (v) Solute desorbs. (vi) Diffuses through the pores and diffuses back to external film and into moving fluid. A given molecule can sorb and desorb number of times. While in moving fluid, solute is carried along the interstitial fluid velocity v , until it diffuses into another particle and whole process is repeated.
Migration of solute is function of f (^) ( ε (^) e , ε (^) p , Kd , v & sorption equilibrium).
Δz Packed Bed
Fig. 8.3: Schematic of the elemental area of the column
(^1 1 1) ( 1 )
s e e e p^ d^ e p^ s
u v K q C
If isotherm is linear,
q = m T ( (^) ) C or (^) Δ^ Δ C q = m (8.8)
If isotherm is nonlinear, q = ACn (8.9)
lim (^ )^1 n^1 C
q q m T C C C n
− →∞
Case 1: For very large molecule, Kd = m T ( ) = 0
Case 2: For small molecule, K (^) d = 1.0, thus smaller molecules move slower.
Case 3: In case of strong adsorption, molecules move slower as velocity decreases. In case of linear adsorption, solute velocity does not depend on solute concentration. In case of non-linear adsorption, solute velocity depends on solute concentration. But at low concentration linear adsorption is valid. z = u t s (8.11)
So, if we plot z vs t , we get a linear with slope of us. By looking into z versus time curve one can interpret the separation efficiency (Fig. 8.4).
A B
C (^) Aor/C (^) A CB/CB
1.
0
C (^) Aor/C (^) A CB/CB time
time
time
1.
0
L Z
Feed Pulse
(a) Feed
(b)
(c) Output
uA^ A B
B A
Fig. 8.4: Solute movement at various points in the column. (a). Feed pulse; (b). Solute movement in column; (c). Product concentration in the outlet
Fig. 8.4 (a) shows the feed pulse containing two components A and B, to be separated. Thus, the non-dimensional concentration in the feed looks like a rectangular pulse over a short duration, as shown in Fig. 8.4 (a). Now, the physico-chemical and adsorption parameters for the two solutes are different. Therefore, the velocities of these two solutes will be different. Thus, they will cover the column length (L) in different periods of time.
Ion Exchange
Ion exchange means exchange of ions from a medium. It has typical application in water softening by exchange of alkaline metal ions like Ca2+, Mg2+^ by Na+. Other common applications are sugar processing, hydrometallurgical application, protein fractionation, biological separation, etc.
Fundamentals:
Ion from a solution is removed when it is passed through a bed of exchangeable ions, called resins.
A +^ + R B −^ +^ + X −^ = R A −^ +^ + B ++ X −
B
In this reaction, R-^ is fixed negative charge on the resin. A+^ and B+^ are called counter- ions and X-^ is called coion in resin phase. An example is addition of KCl to a resin having Na+^ ions. Here, K+^ and Na+^ are counterions and Cl-^ is the coion. Above example is for a monovalent cation exchange process. Anion exchange is similar, except anions are being exchanged. Ion concentration in the liquid and resin are expressed in terms of equiv/m^3 , based on total column volume. In the solution, total ion concentration is CT = CA + C (8.12)
In the resin phase, total ion concentration is C (^) RT = CRA + CRB (8.13)
Eqs. (8.12) and (8.13) are known as the electoneutrality conditions. These hold in both the phases. Thus, in the resin phase as counterion B+^ leaves, equivalent A+^ from the solution should join to maintain electroneutrality. After exchanging the ions, the resin is in the form of R-A+^ from R-B+. To regenerate the resin, a concentrated solution of B+X-
has to be added in the column. Thus, a complete ion exchange cycle consists of three steps, (i) Loading Æ A+^ goes to resin from solution (ii) Regeneration Æ A+^ is removed from resin (iii) Washing Æ washing of excess B+X-^ from the column For a divalent cation with monovalent cation, the reaction becomes D ++^ + 2 R B −^ +^ + 2 X −^ = R D 2 −^ ++^ + 2 B +^ + 2 X − Divalent ion occupies two sites o the resin. Example is removal of Ca2+^ from aqueous solution using a resin having Na+, as counterion. Electroneutrality conditions in resin and solution phase becomes, CT = CB + CD (8.14) C (^) RT = CRB + CRD (8.15)
One can define equivalent fractions of ions in both phases as, in solution phase:
i^ i T x C = (^) C (8.16)
in resin phase:
i^ Ri RT y C = (^) C (8.17)
It may be noted that,
∑ xi^ =∑ yi =^1 (8.18)
Binary ion exchange equilibrium:
Equilibrium equations can be written in terms of ion fractions, using equations (8.16) and (8.17) as
AB^ A^ B B A K y^ x = y x ( ) ( )
A A A A
y x y x
yA can be extracted and can be written in terms of xA and KAB as
(^1) ( 1 ) A^ AB^ A AB A y K^ x = (^) + K − x (8.20) K (^) AB is constant for dilute system.Generally, KAB values are known.
For monovalent-divalent exchange the corresponding equation is,
( 1 ) 2 ( 1 )^2
D (^) DB RT D T D
y (^) K C y C x
= ⎜^ ⎛^ ⎞⎟
x D
Or 1 2^ B^ B^1 RT ( (^12) B ) DB T
y xB y (^) K C x C
0
1
1
KAB > 1 KAB < 1
xA
yA
Fig. 8.16: Equilibrium curve for monovalent - monovalent ion exchange
xD
yD
0
1
1
P >
P <
Fig. 8.17: Equilibrium curve for monovalent - divalent ion exchange P = K (^) DBCCRTT
Ion movement theory:
Result for any ion velocity is
io n 1 1 R T E
e T
u V
C y K
ε C x
+ Δ^ (8.22)
Where, y, x = equivalent ion fraction; KE = a factor to include effects of Donnan exclusion and electroneutrality. If the ion is excluded KE =0, else KE =1. For counterions A+, B+, D++^ KE = For coions X-, KE =
super (^10) 25 cm/min e 0. v v
K (^) CaNa = 1. K (^) CaNaCC^ RTT = 1.3 × (^) 11 10×^2 − 3 =236.4 1.
So, shock wave. After the shock wave, xa Ca (^) , = xF Ca , =0.
y a C a , is calculated from equilibrium relation,
( ) ( )
, 2 , 2 (^1) , 1 ,
a Ca (^) CaNa RT a Ca a Ca T a Ca
y (^) K C x y C x
= ⎜^ ⎛^ ⎞⎟
( )^2
ya Ca (^) , =0.
Resin contains higher concentration of Ca2+^ after the shock wave.
, , , ,
sh 1 1 RT E a Ca b Ca e T a Ca b Ca
u V C (^) K y^ y
ε C x x
yb Ca (^) , = xb Ca , = 0
2 ( )
u sh −
+ ⎜⎝ × ⎟⎠ ⎜⎝^ ⎞⎟⎠
= 0.0113 cm/min
K (^) E = 1.0, Since ions are not excluded. Low velocity of shock wave is due to the resin has a high capacity compared to liquid concentration and it is selective for Ca2+. 200 cm (^) 294.4 hrs. t F = (^) 0.113 cm/min=
(b). Bed is regenerated by 25% (wt.) NaCl at a superficial velocity 0.5 cm/min. Find time required to regenerate the bed? Solution:
super 0.5^ 1.25 cm/min
e 0.
v v
At 25^0 C, ρL =1.194 g/cc
So, 25% NaCl = 0.25^ ×58.4151.194^ × 1000 = 5.311 (^) litermol=5.311 equiv. So, CT =5.311 equiv/liter
xC a ,R eg. = 0 as regeneration liquid is totally NaCl.
C a N a^ R T T
K C
C =^ ×^ =^ <
So, we get a shock wave when material concentrated with in Ca2+^ is removed with regenerate. t = time required of ion wave to go to the column exit 200 cm (^) 160 min. = (^) 1.25 cm/min=
