Citation
Characterization of Vanadium, Copper, and Iron Adsorption Kinetics in Ion Exchange Resins: Part I

Material Information

Title:
Characterization of Vanadium, Copper, and Iron Adsorption Kinetics in Ion Exchange Resins: Part I
Creator:
Brenton, Ryan
Creagan, Jesse
Publication Date:
Language:
English

Subjects

Subjects / Keywords:
Adsorbents ( jstor )
Adsorption ( jstor )
Inlets ( jstor )
Ions ( jstor )
Isotherms ( jstor )
Kinetics ( jstor )
Matrices ( jstor )
Resins ( jstor )
Vanadates ( jstor )
Vanadium ( jstor )
Chemical kinetics
Copper
Iron
Vanadium
Genre:
Undergraduate Honors Thesis

Notes

Abstract:
The main goal of this experimentation is to characterize an ion exchange column that is used to remove Cu, V, and Fe from an acidic stream that contains a mixture of organic acids and develop kinetics models that support previously collected data. In order to do this it is necessary to develop a characteristic differential equation that expresses the concentration as a function both of position and time in the column. This equation, in conjunction with the equilibrium data for the metals with the resin and empirically obtained run data, can be used to develop kinetics models that accurately represent the adsorption and desorption process. To develop these it will also be necessary to obtain the exhaustion data (i.e. complete saturation) of the resins. ( en )
General Note:
Ryan Brenton awarded Bachelor of Science in Chemical Engineering; Graduated May 8, 2012 summa cum laude. Major: Chemical Engineering
General Note:
College/School: College of Engineering
General Note:
Jesse Creagan awarded Bachelor of Science in Chemical Engineering; Graduated May 8, 2012 summa cum laude. Major: Chemical Engineering
General Note:
Legacy honors title: Only abstract available from former Honors Program sponsored database.
General Note:
Advisor: Spyros Svoronos

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Ryan Brenton and Jesse Cregan. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.

Full Text

PAGE 1

Characterization of Vanadium, Copper, and Iron Adsorption Kinetics in Ion Exchange Resins : Part I Authors: Ryan Brenton Coauthor: Jesse Creagan

PAGE 2

Contents 1 Introduction ................................ ................................ ................................ ............................. 4 1.1 Background: ................................ ................................ ................................ ..................... 4 1.2 Purpose: ................................ ................................ ................................ ............................ 4 2 Experimental P rocedure ................................ ................................ ................................ .......... 5 3 Theoretical Basis ................................ ................................ ................................ ..................... 6 3.1 Existence of V(V) Matrices ................................ ................................ .............................. 6 3.2 Kinetic Analysis ................................ ................................ ................................ ............... 7 3.2.1 Non competitive model: ................................ ................................ ........................... 7 3.3 Attachment /Detachment Rates: ................................ ................................ ........................ 7 3.3.1 Data Collection and Approach ................................ ................................ ................ 10 4 Modeling and Characterization ................................ .............. Error! Bookmark not defined. 4.1 Column Characterization ................................ ................. Error! Bookmark not defined. 4.2 Rate Constants ................................ ................................ Erro r! Bookmark not defined. 5 Isotherm analysis using Equilibrium data: ............................. Error! Bookmark not defined. 6 Results : ................................ ................................ ................... Error! Bookmark not defined. 7 Figures: ................................ ................................ ................................ ................................ .. 12 8 SOP ................................ ................................ ................................ ................................ ........ 27 9 Bibliography ................................ ................................ ................................ .......................... 31

PAGE 3

F IGURE 3: V 2 O 7 4 C OMPLEX ................................ ................................ ................................ ................................ ......... 12 F IGURE 4: V 4 O 12 4 C OMPLEX ................................ ................................ ................................ ................................ ....... 12 F IGURE 1: V AN ADATE I ON ................................ ................................ ................................ ................................ ........... 12 F IGURE 2: V ANADATE I ON M M I NTERACTION ................................ ................................ ................................ ........... 12 F IGURE 5: V3O93 C OMPLEX ................................ ................................ ................................ ................................ ....... 13 F IGURE 6: MO V ANADATE ................................ ................................ ................................ ................................ ........... 13 F IGURE 7: P OLYMATH C ODE ................................ ................................ ................................ ................................ ...... 14 F IGURE 8: R ESIN D V ANADIUM I SOTHERM ................................ ................................ ................................ ................... 15 F IGURE 9: R ESIN T V ANADIUM I SOTHERM ................................ ................................ ................................ ................... 15 F IGURE 10: R ESIN D C OPPER I SOTHERM ................................ ................................ ................................ ...................... 16 F IGURE 11: R ESIN T C OPPER I SOTHERM ................................ ................................ ................................ ....................... 16 F IGURE 12: R ESIN D F E I SOTHERM ................................ ................................ ................................ ............................... 17 F IGURE 13: R ESIN T F E I SOTHERM ................................ ................................ ................................ ............................... 17 F IGURE 14: I NVERSE L ANGMUIR I SOTHERM R ESIN D C OPPER ................................ ................................ ...................... 18 F IGURE 15: I NVERSE L ANGMUIR I SOTHERM FOR R ESIN T C OPPER ................................ ................................ ............... 18 F IGURE 16: F REUNDLICH I SOTHERM FOR R ESIN D V ANADIUM ................................ ................................ .................... 19 F IGURE 17: F REUNDLICH I SOTHERM FOR R ESIN T V ANADIUM ................................ ................................ ..................... 19 F IGURE 18: G RAPH D ISPLAYING E XPERIMENTALLY C OLLECTED D AT A WITH F ITTED C URVE FOR THE O UTLET V C ONCENTRATION OF A C OLUMN WITH 100 M L OF R ESIN T WITH I NLET C ONCENTRATION 57 PPM ....................... 21 F IGURE 19: P OLYMATH M ODEL OF C OLUMN DESCRIBED BY F IGURE 18 ................................ ................................ ...... 21 F IGURE 20: G RAPH D ISPLAYING E XPERIMENTALLY C OLLECTED D ATA WITH F ITTED C URVE FO R THE O UTLET V C ONCENTRATION OF A C OLUMN WITH 100 M L OF R ESIN D WITH I NLET C ONCENTRATION 57 PPM ....................... 22 F IGURE 21: P OLYMATH M ODEL OF C OL UMN DESCRIBED BY F IGURE 20 ................................ ................................ ...... 22 F IGURE 22: G RAPH D ISPLAYING E XPERIMENTALLY C OLLECTED D ATA WITH F ITTED C URVE FOR THE O UTLET C U C ONCEN TRATION OF A C OLUMN WITH 30 M L OF R ESIN D WITH I NLET C ONCENTRATION 21 PPM ......................... 23 F IGURE 23: P OLYMATH M ODEL OF C OLUMN DESCRIBED BY F IGUR E 22 ................................ ................................ ...... 23 F IGURE 24: G RAPH D ISPLAYING E XPERIMENTALLY C OLLECTED D ATA WITH F ITTED C URVE FOR THE O UTLET V C ONCENTRATION OF A C OLUMN WITH 30 M L OF R ESIN D WITH I NLET C ONCENTRATION 36 PPM ......................... 24 F IGURE 25: P OLYMATH M ODEL OF C OLUMN DESCRIBED BY F IGURE 24 ................................ ................................ ...... 24 F IGURE 26:G RAPH D ISPLAYING E XPERIMENTALLY C OLLECTED D ATA WITH F ITTED C URVE FOR THE O UTLET C U C ONCENTRATION OF A C OLUMN WITH 30 M L OF R ESIN T WITH I NLET C ONCENTRATION 21 PPM ......................... 25 F IGURE 27: P OLYMATH M ODEL OF C OLUMN DESCRIBED BY F IGURE 26 ................................ ................................ ...... 25 F IGURE 28: G RAPH D ISPLAYING E XPERIMENTALLY C OLLECTED D ATA WITH F ITTED C URVE FOR THE O UTLET V C ONCENTRATION OF A C OLUMN WITH 30 M L OF R ESIN T WITH I NLET C ONCENTRATION 36 PPM ......................... 26 F IGURE 29: P OLYMATH M ODEL OF C OLUMN DESCRIBED BY F IGURE 28 ................................ ................................ ...... 26

PAGE 4

1 Introduction 1.1 Background: The term AMLS refers to Aqueous Mother Liquor Stream, and is a byproduct in the production of Nylon 6 6. Ascend Performance Materials, one of the leaders in U. S. production of Nylon 6 6, has provided samples of AMLS for the analysis of this project. This untreated AMLS contains traces of Copper, Vanadium, and Iron. Ascend Performance Materials tasked an IPPD team under Dr. Svoronos to remove these metal ions whi le retaining the organics in the aqueous solution. After two semesters of experimentation and elimination of possible removal methods, the team focused their efforts on two ion exchan ge resins, labeled D and T These provided the best consistent metal removal. This determination was made on data collected for varying superficial velocities and metal concentrations of the feed stream, as well as for different bed heights of resin. The continuation of this project seeks to further analyze the data collec ted in the previous two semesters and utilize it creat e new models based on the resin s kinetics An extended look into the possibility of anionic complexes, their kinetic effect on adsorption, and the feasibility of column characterization using equilibri um data is explored. 1.2 Purpose: The main goal of this experimentation is to characterize an ion exchange column that is used to remove Cu, V, and Fe from an acidic stream that contains a mixture of organic acids and develop kinetics models that support pre viously collected data. In order to do this it is necessary to develop a characteristic differential equation that expresses the concentration as a function both of position and time in t he column. This equation, in conjunction with the equilibrium data fo r the metals with the resin and empirically obtained run data, can be used to develop kinetics models that accurately represent the adsorption and desorption process. To develop these it will also be necessary to obtain the exhaustion data (i.e. complete s aturation) of the resins. The secondary objective of this experimentation is to examine the possibility of the presence of anionic Vanadium matrices, and to determine whether they can be removed via adsorption to a basic ion exchange resin The objective was chosen because, during ion removal with cationic resins, Vanadium was consistently the hardest metal to remove. It is known that the acidic solution contains the higher oxidation states of Vanadium including Vanadium(V) These higher oxidation states of Vanadium typically form complex matrices in aqueous solutions that may affect the overall rate of adsorption and desorption for Vanadium on the resin. Seeing as this is a preliminary analysis into the possibility of this type of matrix forming it will not assume or propose that any one type of complex forms, but rather empirically determine whether any could exist. The project will therefore reference different complexes and their properties that are known to exist, rather than hypothesize that any one in particular.

PAGE 5

2 Experimental Procedure The experimental procedure for the data collection was similar to that of the IPPD project (see SOP) There were three primary pieces of data that needed to be collected in order to attempt to meet the experiments objectives. The first set of data required exhausting the resin to its fullest saturation capacity. This would be the point at which all of t he solution running through the column would be in equilibrium with the resin for the entire height of the column. This would mean that a mass balance on this system would yield no accumulation of metals on the resin. In order to do this 100mL of both resi ns were run at a rate of 24 bv/h for 75 bed volumes. This was chosen as an overshoot estimate for the saturation limits of the resins, and was based on previous data obtained from the IPPD experiment. The analysis of the data from these runs showed that ev en the Vanadium, the metal which most easily breaks through, was still having marginal removal. The other metals, Copper and Iron, were still having significant removal. Another run was conducted at a resin volume of 30 mL and at a rate o 24 bv/h for 66 b ed volumes. It was intended for both resins to again be run to 75+ bed volumes however due to a lack of solution 66 was the max that could be reached. The analyzed data for this set of runs showed breakthrough and saturation for both the Copper and the Va nadium, however the Iron, was still strongly being adsorbed. The second set of data required obtaining equilibrium data between the resins and the metals. In order to do this 20 mL of resin was added to 5 different concentrations of solution. The differen t concentrations were obtained by dilution of the original solution. The analysis of this data showed that nearly all of the metals were adsorbed in all 10 test samples. In order to obtain proper equilibrium data the samples were spiked with 1000ppm standa rds of Vanadium and 10,000ppm standards of Copper. Unfortunately there was no pure standard of Iron within the ppm range needed to dope the samples. The third set of data needed is that for the anionic resin to determine whether negatively charged complexes of Vanadium can exist in the solution. The anionic resin was run in a similar fashion to the data collected during the screening process of the IPP D project. Seven bed volumes of solution were run at a rate of 10 bv/h and at a resin volume of 100mL. All seven bed volumes were collected for analysis.

PAGE 6

3 Theoretical Basis 3.1 Existence of V(V) Matric es In higher pH ranges, usually greater than pH12, Vanadiu m often forms complexes such as Vanadate, shown in Figure 1 Although Vanadate complexes generally form in ternary geometries it is possible for them to adopt square planar geometries as well when in the presence of other metal ions Figure 2 It was origi nally proposed that Vanadate could be removed via ion exchange because the negatively charged complexes would likely bind to a positively charged exchange site in a basic resin; however the acidity of the solution in the sample is found to have a much lowe r pH than the limit at which Vanadate exists. A search of the literature shows that as pH is lowered Vanadate V(V) forms more complex coordination complexes such as VO 3 (OH) 2 ,V 2 O 7 4 V 4 O 12 4 V 3 O 9 3 VO 2 (OH) 2 2 as well as other more complex configuratio ns Figures 3 5 It is also possible in the lower pH range for Vanadium to form neutral complexes such as Vanadium (III) oxide and Vanadium(V) oxide. In aqueous solutions V(V) is usually the most stable fo rm of Vanadium, and one of the c omplex es listed here is the most likely matrix form possible. The electronic structure of Vanadate is similar to its lower pH relatives in that all of the matrices form using Vanadium(V) ions. Vanadium(V) with its +5 charge is considered a d0 ion. Because of t his it can have only one electronic configuration, and will not exhibit separate high and low spin states based on geometry. This actually contributes to the similarities amongst previously mentioned coordination complexes. Figure 6 shows a molecular orbit al diagram for the Vanadate complex. The importance of the diagram is the ligand orbitals of the Oxygen on the orbitals, and are ordered based on their symmetr orbitals that could contribute to matrix stability only five actually do so for the tetrahedral configuration of Vanadate, the three t2 and two e. This changes based on the overall symmetry assumed by the comp lex and will therefore vary from complex to complex. The fact that all of bonding orbitals will be filled. This means that these bonds are very strong and gives some justification to t he hypothesized existence of matrix complexes in a solution with some part nitric acid. In order to test the existence of these Vanadate like matrices, a basic ion exchange resin will be used in an attempt to purify the AMLS through adsorption. Seven 100 mL samples of AMLS will be tested for metal content before and after being run through 100mL of resin. The specific resin that will be used is Purolite A300, which is a strongly basic gel anion exchange resin. Its structure is that of a polystyrene cross l inked with divinylbenzene, which provides resonance configurations that can contribute to stabilized positively charged regions. A significant reduction in the amount of Vanadium seen will give credence to the existence of anionic forms of Vanadium.

PAGE 7

3.2 Kineti c Analysis The existence of complex metal matrices would likely lead to complex kinetics. Within the limit of time constraints a number of proposed mechanisms and the derivations of their equations are explained. The first of these equations reflect the s implest analysis of the kinetic system within the resin, the case of complete non competitive adso r ption. The second of these is a model for competitive adsorption and the subsequent models are of increasing sophistication and leave a framework in which fu ture more advanced models could be built upon. The hypothesis that Vanadium matrices exist in the AMLS can be accounted for by at least one of these models. 3.2.1 Non competitive model : The following equations represent the rate of attachment and detachment f or each metal in both resin D and T. The rate of attachment of the metals to the resin is similar to that of adsorption of a gas onto a solid. The main difference is that the partial pressure term in gas to solid adsorption, which gives a measure of the nu mber of molecular collisions between the gas and the metal, is replaced here with its equivalent concentration of metal in the solution. The rate of detachment is exactly the same, and is only dependent on the concentration of the metals on the resin. This model is for non competitive adsorption, in which it is assumed that there are three different types of vacant sites on the resin, one for each metal. It also assumes that each metal can only bind to one of these three distinct types of sites, and that no two metals bind to the same site. It is furthermore assumed that each metal binds to only one site, that is that no metal binds to multiple sites regardless of charge. 3.3 Attachment/Detachment Rates:

PAGE 8

Equilibrium Constants: The equilibrium constants express the ratio between the rate constants for adsorption and desorption of the metals on the resin.

PAGE 9

Adsorption Rates: These rat es represent the non competitive rate at which each metal would adsorb to their distinct site. This is a result of the combination of the rate at which metal ions attach to the metal less the rate at which they detach. Total Mass Balances: Because the vacancies are assumed to be specific to each metal, there is a total of three site balances for each resin. The total number of sites for any given metal is therefore fixed and independent of the other sites. This also serves to remind that the metals are only able to bind to a distinct site.

PAGE 10

Rate at Equilibrium (PS SA Pseudo Steady State Approximation ): Concentration of Vacancies (PSSA): By making a pseud o steady state approximation, in which the rate of adsorption and desorption is considered to be constant and equal, and therefore the overall rate is zero, the concentrations of the metals on the resins can be determined. 3.3.1 Data Collection and Ap p roach In addition to the experimentation and data collected for the purposes of the commissioned IPPD project, supplementary experiments were performed and analyzed in order to determine the kinetics data. To obtain equilibrium data, resin samples were exposed to varying metal concentrations for extended periods of time under conditions of continuous stirring. The before and after concentrations of the liquid phase were measured using ICP. Through mass balances, the concentration of each ionic species on a given r esin sample could be determined. This data was subsequently plotted, and fitted with least squares regression lines using Microsoft Excel functions. The data for each resin ion combination proved to follow either a power law or logarithmic expression, as was predicted by the literature ( Geankoplis et al).

PAGE 11

V alues were then extracted as per the methods specified in the Section Error! Reference source not found. to develop Langmuir a nd Freundlich equations Add i tionally, K values were determined using the above PSSA equations. These equations make the assumption that the concentrations of each species have only first order affects on the kinetics of the resin ion exchange process Due to constraints on time and computing power, alt ernate models were not explored. The resulting equilibrium constants ( Table 1 ) were then used as inputs t o the below computational model.

PAGE 12

4 Figures: Figure 3 : V 2 O 7 4 Complex Figure 4 : V 4 O 12 4 Complex Figure 1 : Vanadate Ion Figure 2 : Vanadate Ion M M Interaction

PAGE 13

Figure 5 : V3O93 Complex Figure 6 : MO Vanadate

PAGE 14

Figure 7 : Polymath Code y = 0.0943x 0.3137 R = 0.9367 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0 10 20 30 40 50 q (kg adsorbate/kg adsorbent) Concentration V Resin D Vanadium Isotherm Dowex V Power (Dowex V)

PAGE 15

Figure 8 : Resin D Vanadium Isotherm Figure 9 : Resin T Vanadium Isotherm y = 0.1098x 0.3366 R = 0.9389 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0 10 20 30 40 50 q (kg adsorbate/ kg adsorbent) Concentration V Resin T Vanadium Isotherm Series1 Power (Series1)

PAGE 16

Figure 10 : Resin D Copper Isotherm Figure 11 : Resin T Copper Isotherm y = 0.3754ln(x) + 0.5868 R = 0.9807 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 0 10 20 30 q (kg adsorbate / kg adsorbent) Concentration Cu Resin D Copper Isotherm Dowex Cu Log. (Dowex Cu) y = 0.3642ln(x) + 1.1823 R = 0.8087 0 0.5 1 1.5 2 2.5 3 0 10 20 30 40 q (kg adosrbate/ kg adsorbent) Concentration Cu Resin T Copper Isotherm Series1 Log. (Series1)

PAGE 17

Figure 12 : Resin D Fe Isotherm Figure 13 : Resin T Fe Isotherm 0 0.01 0.02 0.03 0.04 0.05 0.06 0 0.05 0.1 0.15 0.2 0.25 0.3 q (kg adsorbate/kg adsorbent) Concentration Fe Resin D Fe Isotherm Series1 0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0 0.1 0.2 0.3 0.4 0.5 q (kg adsorbate/kg adsorbent) Concentration Fe Resin T Fe Isotherm Series1

PAGE 18

Figure 14 : Inverse Langmuir Isotherm Resin D Copper Figure 15 : Inverse Langmuir Isotherm for Resin T Copper y = 1.9297x + 0.4825 R = 0.98 0 0.2 0.4 0.6 0.8 1 0 0.05 0.1 0.15 0.2 0.25 1/q 1/c Inverse Langmuir Isotherm for Resin D Copper Series1 Linear (Series1) y = 1.4886x + 0.3238 R = 0.975 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0 0.05 0.1 0.15 0.2 1/q 1/c Inverse Langmuir Isotherm for Resin T Copper Series1 Linear (Series1)

PAGE 19

Figure 16 : Freundlich Isotherm for Resin D Vanadium Figure 17 : Freundlich Isotherm for Resin T Vanadium y = 0.0943x 0.3137 R = 0.9367 0.01 0.1 1 0.1 1 10 Log_10 of q Log_10 of c Freundlich Isotherm of q vs c for Resin D Vanadium Series1 Power (Series1) y = 0.1098x 0.3366 R = 0.9389 0.1 1 0.1 1 10 100 Log_10 of q Log_10 of c Freundlich Isotherm of q vs c for Resin T Vanadium Series1 Power (Series1)

PAGE 20

Table 1 : Experimentally Determined Equilibrium Constants K Values Using PSSA (mg resin*L/mol^2) Resin Cu V Fe Dowex 0.003037356 0.000884 N/A Thermax 0.009649013 0.00032 N/A Table 2 : Experimentally Determined Freundlich and Langmuir Coefficients K, q0, n Values Using Freundlich and Langmuir Isotherms Resin Cu V Fe K q0 K n Dowex 4.005 2.072 0.0943 0.3137 N/A Thermax 4.59728 3.088 0.1098 0.3366 N/A

PAGE 21

Figure 18 : Graph Displaying Experimentally Collected Data with Fitted Curve for the Outlet V Concentration of a Column with 100mL of Resin T with Inlet Concentration 57ppm Figure 19 : Polymath Model of Column described by Figure 18 -10 0 10 20 30 40 50 60 0 50 100 150 200 100mL Resin T C_V=57 Thermax T-46ish Fitted Model(30mL)

PAGE 22

Figure 20 : Graph Displaying Experimentally Collected Data with Fitted Curve for the Outlet V Concentration of a Column with 100mL of Resin D with Inlet Concentration 57ppm Figure 21 : Polymath Model of Column described by Figure 20 -10 0 10 20 30 40 50 60 0 50 100 150 200 Outlet Concentration (ppm) Time(min) 100mL Resin D C_V= 57 100mL fitted(100mL)

PAGE 23

Figure 22 : Graph Displaying Experimentally Collected Data with Fitted Curve for the Outlet Cu Concentration of a Column with 30mL of Resin D with Inlet Concent ration 21ppm Figure 23 : Polymath Model of Column described by Figure 22 0 5 10 15 20 25 0 50 100 150 200 Outlet Concentration (ppm) Time(min) 30mL Resin D C_Cu= 21 30mL Fitted Model(30mL)

PAGE 24

Figure 24 : Graph Displaying Experimentally Collected Data with Fitted Curve for the Outlet V Concentration of a Column with 30mL of Resin D with Inlet Concentration 36ppm Figure 25 : Polymath Model of Column described by Figure 24 0 5 10 15 20 25 30 35 40 0 50 100 150 200 Outlet Concentration (ppm) Time(min) 30mL Resin D C_V= 36 30mL Fitted Model(30mL)

PAGE 25

Figure 26 :Graph Displaying Experimentally Collected Data with Fitted Curve for the Outlet Cu Concentration of a Column with 30mL of Resin T with Inlet Concentration 21ppm Figure 27 : Polymath Model of Column described by Figure 26 -5 0 5 10 15 20 25 0 50 100 150 200 Outlet Concentration(ppm) Time(min) 30mL Resin T C_Cu=21 T-46 Fitted Model(30mL)

PAGE 26

Figure 28 : Graph Displaying Experimentally Collected Data with Fitted Curve for the Outlet V Concentration of a Column with 30mL of Resin T with Inlet Concentration 36ppm Figure 29 : Polymath Model of Column describe d by Figure 28 0 10 20 30 40 50 0 50 100 150 200 Outlet Concentration (ppm) Time(min) 30mL Resin T C_V=36 T-46 Fitted Model(30mL)

PAGE 27

5 SOP 1.0. METHOD SUMMARY The capacities of ion exchange resins are tested in the laboratory using a column and variable flow rate pump. The following procedures outline the methods recommended by Dow and Purolite for testing resins using co current regeneration. Regeneration solutions are used to fill ion exchange sites within the resin and are selected based on resin characteristics. One bed volume (BV), chosen as 100 ml, of resin is added to the column. It is th en rinsed in an up flow and down flow manner to both classify the beads and remove any contaminants. These steps are followed by a regeneration step and two final rinse steps. Finally, the testing solution is pumped through the column and collected as 100 ml samples in 150 ml beakers for digestion and filtration. 2.0. EQUIPTMENT AND SUPPLIES 2.1. APARATUS AND MATERIALS 2.1.2. Glass column 2.1.3. Pump 2.1.4. 500 ml graduated cylinder 2.1.5. 500 ml Erlenmeyer flask 2.1.6. 200 ml Griffin beaker 2.1.7. 150 ml Griffin beakers 2.1.8. Parafilm 2.1.9. 100 ml resin 2.2. REAGENTS 2.2.1. Deionized Water 2.2.2. AMLS 2.2.3. Hydrochloric or Sulfuric acid

PAGE 28

3.0. QUALITY CONTROL To ensure good quality control the following will be carried out: Replicate Trials : Multi ple test runs with the same resin will be carried out through the entire analytical process to determine precision. 4.0. PROCEDURES 4.1. PREPARATION 4.1.1. Clean and label all glassware. 4.1.2. Make 200 ml of regeneration solution according to specific resin operating conditions (Use 4% H 2 SO 4 ) 4.1.3. Collect a sample of AMLS from the 5 gal containers using a pump and acid approved tubing. 4.2. RESIN COLUMN TESTING 4.2.1. Fill the 500 ml graduated cylinder with DI water. 4.2.2. Add 100 ml of DI water to a 150 ml beaker. 4.2.3. Add 100 ml of resin to a 150 ml beaker (DO NOT COMPACT THE RESIN) 4.2.4. Pour 50 ml of the DI water into the beaker containing the resin. 4.2.5. Pour the resin slurry into the test column. 4.2.6. Use the remaining 50 ml of DI water to remove any resin left in the beaker and add it to the column. 4.2.7. Pla ce the outlet tubing in the graduated cylinder containing DI water and the outlet tubing in the empty 500 ml flask. 4.2.8. Backwash the resin with DI water in an up flow manner at a rate of 15 BV/h for 12 minutes. 4.2.9. Gradually reduce the flow to zero and allow the r esin to gravity settle.

PAGE 29

4.2.10. Place the inlet tubing in the graduated cylinder containing DI water and the outlet tubing in the 500 ml flask. 4.2.11. Reverse the flow and rinse the resin in a down flow manner with DI water for 2 BV at a rate of 10 BV/h. 4.2.12. Empty the 500 ml flask. 4.2.13. Reverse the flow, and backwash the with regeneration solution at 5 BV/h for 2 BV. 4.2.14. Reverse the flow, and forward rinse the resin with DI water at 5 BV/h for 1 BV. 4.2.15. Fast rinse the resin with DI water at 10 BV/h for 1 BV. 4.2.16. Empty the 500 ml flask into t he appropriate hazardous waste container. 4.2.17. For REGENERATION see section 4.3. 4.2.18. Run AMLS through the resin at a rate of 10 BV/h. 4.2.19. Collect the first sample for testing after 200 ml of liquid has gone through the column. 4.2.20. Collect 1 BV (100 ml) samples in separate 150 ml beakers. 4.2.21. Stop collection and cover samples with parafilm. 4.2.22. Empty waste AMLS into the appropriate hazardous waste container. 4.2.23. If running the resin to EXHAUSTION, flush resin with 3 BV of DI water at 24 BV/h. 4.2.24. Leave the resin soaking in water. 4.2.25. R epeat steps 4.2.7 through 4.2.9 4.2.26. Place the inlet tubing in the regeneration solution and the outlet tubing in the empty 500 ml flask. 4.2.27. Repeat steps 4.2.13 through 4.2.21 4.3. REGENERATION 4.3.1. Forward rinse with 3 BV of DI water at a rate of 24 BV/h. 4.3.2. Backwash with 8% H 2 SO 4 (volume) at a rate of 10 BV/h for 5BV. 4.3.3. Backwash with 1 BV of DI water at a rate of 10 BV/h. 4.3.4. Forward rinse with 3 BV of DI water at a rate of 10 BV/h. Return to step 4.2.16.

PAGE 30

4.4. FILTER 4.4.1. Use folded P8 filter paper in a funnel in a Nalgene bottle. 4.4.2. Pour sample into the filter, (make sure to capture at least 50 mL). 4.4.3. Label should include: Team initials, date the column is run, resin, speed of the pump, if regen: regen 1, 2, or 3. 4.5. CLEAN UP 4.5.1. Dispose of any unused acids or AMLS into their proper hazardo us waste container. 4.5.2. Dispose of resin into the resin hazardous waste container. 4.5.3. Flush empty column and tubing with DI water. 5.0. REFERENCES 5.1. 5. 2. arrying out Laboratory Trials on Ion Exchange Resins and

PAGE 31

6 Bibliography Hu, Bo Yu, Yong Jun Yuan, Chao Jiang, and Qiang Liu. "Rational Oxidation of Cyclohexane to Cyclohexanol, Cyclohexanone and Adipic Acid with Air over Metalloporphyrin and Cobalt Salt." WorldsciNet. Web. 19 Apr. 2012. . Kotrba, Pavel, Martina Mackova, and Tomas Macek. Microbial Biosorption of Metals. Dordrecht: Springer Science+Business Medi a, 2011. Print. 3.3.1.2 pg 25 Geankoplis, Christie J., and Christie J. Geankoplis. "Chapter 12." Transport Processes and Separation Process Principles: (includes Unit Operations). Upper Saddle River, NJ: Prentice Hall Professional Technical Reference, 200 3. 760 68. Print. Bird, R. Byron, Warren E. Stewart, and Edwin N. Lightfoot. Transport Phenomena. New York: J. Wiley, 2002. Print.