@prefix vivo: . @prefix edm: . @prefix ns0: . @prefix dcterms: . @prefix skos: . vivo:departmentOrSchool "Applied Science, Faculty of"@en, "Chemical and Biological Engineering, Department of"@en ; edm:dataProvider "DSpace"@en ; ns0:degreeCampus "UBCV"@en ; dcterms:creator "March, Alan Charles"@en ; dcterms:issued "2010-10-29T15:51:07Z"@en, "1988"@en ; vivo:relatedDegree "Master of Applied Science - MASc"@en ; ns0:degreeGrantor "University of British Columbia"@en ; dcterms:description """A three-column ion-exchange system was designed, fabricated and computer-automated to accommodate a novel 'elution looping' process developed by Dr. Tim Durance (U.B.C. Department of Food Science) during his doctoral studies on the recovery of lysozyme and avidin. This processing technique enhances the simultaneous recovery of these two pharmaceutically important proteins from chicken egg albumen. The processing system prototype was sized to handle throughput rates between approximately five and 300 liters per day of albumen to facilitate both laboratory and small commercial scale work. Very efficient use is made of the ion-exchange resin due to a two-column cascaded feed arrangement. The processing control software was designed to provide flexibility and ease of operation in setting up new and existing method files, allowing for the selection of any column or group of columns to use and providing a 'staged-shutdown' approach toward handling columns fouled with congealed albumen during unattended operation. This approach attempts to maximize the productivity of the system even when one or two of the columns has become fouled with congealed albumen."""@en ; edm:aggregatedCHO "https://circle.library.ubc.ca/rest/handle/2429/29660?expand=metadata"@en ; skos:note "COMPUTER AUTOMATION .OF A NOVEL ION-EXCHANGE PROCESS FOR THE SIMULTANEOUS RECOVERY OF LYSOZYME AND AVIDIN FROM CHICKEN EGG ALBUMEN by ALAN CHARLES MARCH A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTERS IN APPLIED SCIENCE IN THE FACULTY OF GRADUATE STUDIES DEPARTMENT OF BIO-RESOURCE ENGINEERING We accept t h i s thesis as conforming to the required standard The University of B r i t i s h Columbia August, 1988 (£) Alan Charles March, 1988 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of T^JC \"/2^St4/2c£, ^ u r > / / U g H ^ / V ^ -The University of British Columbia Vancouver, Canada DE-6 (2/88) i ABSTRACT A three-column ion-exchange system was designed, fabricated and computer-automated to accommodate a novel 'elution looping' process developed by Dr. Tim Durance (U.B.C. Department of Food Science) during his doctoral studies on the recovery of lysozyme and avidin. This processing technique enhances the simultaneous recovery of these two pharmaceutically important proteins from chicken egg albumen. The processing system prototype was sized to handle throughput rates between approximately five and 300 l i t e r s per day of albumen to f a c i l i t a t e both laboratory and small commercial scale work. Very e f f i c i e n t use i s made of the ion-exchange resin due to a two-column cascaded feed arrangement. The processing control software was designed to provide f l e x i b i l i t y and ease of operation in setting up new and e x i s t i n g method f i l e s , allowing for the selection of any column or group of columns to use and providing a 1 staged-shutdown' approach toward handling columns fouled with congealed albumen during unattended operation. This approach attempts to maximize the productivity of the system even when one or two of the columns has become fouled with congealed albumen. i i TABLE OF CONTENTS Abstract i Table of Contents i i L i s t of Tables i i i L i s t of Figures iv Acknowledgements v Introduction 1 1. BACKGROUND 3 Ion-exchange Chromatography 3 Ion-Exchange Resins 9 Factors Affecting Exchange Dynamics 10 Recovery of Lysozyme and Avidin 12 2. AUTOMATION OF LYSOZYME/AVIDIN RECOVERY 14 Physical Plant 14 Operating Sequence Nomenclature 19 Flow Path Design 21 Flow Handling Equipment 24 Valves 24 Pumps 2 4 Columns 25 Homogenizer 25 Tubing 29 F i l t e r s 29 Pressure Transducers 31 U l t r a v i o l e t Monitor 31 Electronic Flow Control Hardware 32 DC Power Source 32 System Control Interface 34 Analog to D i g i t a l (A/D) Board 36 D i g i t a l Input/Output (I/O) Board 39 Process Control Computer 40 3. PROCESS CONTROL SOFTWARE 4 4 Mandatory Control Functions 44 Control Enhancements 45 Interactive input 46 Visual display 47 Default drive path 50 F i l e handling routines 50 Staged shutdown 52 Column selection 54 Manual process control 54 Alarms 54 Start time options 55 Step time assignments 55 Run completion setpoint 56 Automatic column cleaning 56 Program Structure and Operation 57 4. DISCUSSION AND FUTURE CONSIDERATIONS 66 References 70 Appendix: System control states 72 i l l LJST OF TABLES 1.1 Functional groups used for ion exchangers 9 i v LIST OF FIGURES 1.1 Flowchart for a t y p i c a l ion-exchange procedure 8 2.1 Cascade sequence for columns A, B, and C as they operate in the capacity of primary and secondary feed, and regeneration 17 2.2 Liquid flow path, valve and a n c i l l a r y equipment arrangement 22 2.3 A t y p i c a l laboratory s t y l e chromatography column with temperature control jacket 26 2.4 E l e c t r i c a l control system schematic 33 2.5 Darlington-connected N-P-N s i l i c o n power t r a n s i s t o r s . . 35 2.6 Card layout for DATA Translation DT2814 A/D card 37 2.7 DATA Translation DT2814 A/D card user connections .... 37 2.8 D i g i t a l I/O card backplane connections for DATA Translation DT2817 41 2.9 D i g i t a l I/O card user connections for DATA Translation DT2817 42 3.1 A t y p i c a l APC (Automatic Process Control) screen 48 3.2 General control structure of process software 59 4.1 General system layout including potential future additions for nearly t o t a l automation of control 69 V ACKNOWLEDGEMENTS I would l i k e to express my appreciation to Dr. K.V. Lo, my advisor, for his constant support and interest throughout t h i s project. I would also l i k e to thank Dr. S. Nakai and Dr. A. Lau for serving as members of my committee. Technical assistance was great-f u l l y gleaned from Jurgen Pehlke and Neil Jackson. To my wife: a special debt of gratitude for putting up with the l i f e of a computer widow through the many long nights that I spent eradicating software bugs (or creating mutations). Funding for thi s project was supplied by grants from the B r i t i s h Columbia Ag r i c u l t u r a l Science Coordinating Committee and the Canadian Egg Marketing Agency. 1 INTRODUCTION The primary goal in producing t h i s work has been to take a novel ion-exchange method, developed in the Department of Food Science at the University of B r i t i s h Columbia (Durance, 1987) for the simultaneous recovery of lysozyme and avidin from chicken egg albumen, and to develop a computer-automated continuous-production system to accommodate th i s process. Sub-goals supporting t h i s main goal make up the basic design s p e c i f i c a t i o n s for the system and include the following: 1. The system must be capable of operating continuously i according to the novel operating sequence established by Durance (1987). 2. The system should provide for continuous feeding of albumen to ion-exchange columns. 3. The column arrangement should e f f i c i e n t l y use the r e s i n beds in order to minimize the quantity of r e s i n required for a given production rate. 4. The physical structure of the recovery system must promote hygienic operation and ease of maintenance. 5. The physical plant should be compact and portable for use in research. 6. The operating system must be \"user f r i e n d l y \" , enabling i t to be invoked e a s i l y by an operator with a basic knowledge of the process. 7. F i l e handling routines should allow for the reading/writing of method f i l e s to/from storage media 2 (either floppy or hard disk) as well as providing printed output for record-keeping. 8. There should be some degree of f l e x i b i l i t y in sele c t i n g which and how many columns are to be used in order to accommodate both production and research needs. 9. Columns should be e a s i l y and quickly replaced to accommodate the use of columns of d i f f e r e n t sizes and to allow for rapid in-process cleaning of the columns should congealed albumen be f i l t e r e d out of the feed flow stream by the res i n , r e s t r i c t i n g the flow rate. 10. The pump capacities should cover a broad range of flow rates to accommodate the various sizes of columns that may be used to allow for the v a r i a t i o n of contact times between the process l i q u i d s and the ion-exchange r e s i n . 3 1. BACKGROUND Ion-exchange Chromatography The theory of ion-exchange was i n i t i a l l y studied during e f f o r t s to understand the mechanism of nutrient transport in s o i l s . Clay p a r t i c l e s were observed to reversibly bind inorganic nutrient cations such as Na +, K*, Ca-*\"2 and NH***\" as i l l u s t r a t e d in Equation 1.1 (Salisbury and Ross, 1978): 2NH4* + clay»Ca Ca* 2 + clay»(NrU) a 1.1 It was discovered that d i f f e r e n t cations could be c l a s s i f i e d according the r e l a t i v e strengths of their attractions for the negatively charged bonding s i t e s . The inequality which describes t h i s pattern i s ca l l e d the Hofmeister or lyotropic s e r i e s : Al + 3>H*>Ba*2>Mn*2>Ca +2>Mg*2>(K*=NH^)>Na*>Li-* 1.2 The strongest to weakest attractions are shown from l e f t to ri g h t , with K\"*\" and NrU*\" exhibiting roughly equal attractions to negatively charged s i t e s . A cation residing at a negatively charged s i t e w i l l generally r e l i n q u i s h i t s position to another ion having a stronger e l e c t r o s t a t i c a t t r a c t i o n as indicated by i t s p o sition in the s e r i e s . This i s an equilibrium reaction which can be reversed by increasing the concentration of the less strongly attracted ion. It i s due to this fact that an ion-4 exchange material can be stripped of strongly held ions and \"regenerated\" by their replacement with less strongly held ions so that the exchange process can begin again. Commercial ion-exchange processes involve the c y c l i c reversal of equilibrium equations such as Equation 1.1. The exchange of sodium ions for calcium and magnesium ions in water softening systems i s one of the most wide spread. Of greater importance to the bio-processing industries are exchanges involving proteins in order to e f f e c t their recovery. Ionic proteins often\" exist as sols or c o l l o i d a l dispersions which r e l y upon the a f f i n i t y of the protein for water, rather than upon the e l e c t r o s t a t i c repulsion of l i k e charges, for s t a b i l i t y . For t h i s reason they are c a l l e d hydrophilic c o l l o i d s and include such materials as soaps, soluble starch, soluble proteins and synthetic detergents (Clark et a l . , 1977). Hydrophobic c o l l o i d s , on the other hand, r e l y upon the d i e l e c t r i c property of water to prevent f l o c c u l a t i o n and s e t t l i n g of these ions. Due to the polar nature of water molecules, c o l l o i d s with a negative surface charge become surrounded with a layer of water molecules oriented with the positive pole toward the surface. The negative pole of each water molecule points away from the p a r t i c l e r e s u l t i n g in a negatively charged \" s h e l l \" some distance removed from the actual p a r t i c l e . Other similar p a r t i c l e s are repelled by the l i k e charges before the p a r t i c l e s are in close enough proximity for the a t t r a c t i v e Van der Waals forces to bind them together. Metal oxides, usually p o s i t i v e l y charged, form hydrophobic sols (Clark et a l . , 1977). 5 With a size range of roughly 1-200 nm, these p a r t i c l e s are subject to Brownian motion caused by the uneven d i s t r i b u t i o n of c o l l i s i o n s with molecules of the continuous phase. These random c o l l i s i o n s tend to de s t a b i l i z e sols by forcing p a r t i c l e s into s u f f i c i e n t l y close contact that Van der Waals forces dominate over the e l e c t r o s t a t i c repulsion . De s t a b i l i z a t i o n also occurs when the surface charge on a p a r t i c l e i s neutralized by a l t e r i n g the pH of the aqueous system. Repulsive forces are reduced by coagulation (charge neutralizaton by the attachment of e l e c t r o l y t e counter-ions) 'allowing f l o c c u l a t i o n - (chemical bridging between p a r t i c l e s ) to occur followed by s e t t l i n g (Clark et a l . , 1977). | The pH at which the e l e c t r o s t a t i c charge i s neutralized i s c a l l e d the i s o e l e c t r i c pH or the i s o e l e c t r i c point, p i , of the c o l l o i d , at which point i t s tendency to remain dispersed i s at a minimum. Lysozyme e x i b l t s a pi of 11.0 while avidin has a pi of 10.0. While many proteins Including lysozyme have been recovered by the method of i s o e l e c t r i c p r e c i p i t a t i o n , the major problem encountered with th i s method i s that the feed stock from which the protein was recovered becomes contaminated with the neu t r a l i z i n g e l e c t r o l y t e s , often making i t u n f i t for subsequent use or processing (Durance, 1987). In order to overcome such problems, ion-exchange resins were devloped with charged chemical groups covalently bound to an insoluble porous matrix. Counter-ions could then be attached and r e v e r s i b l y exchanged with other ions of the same charge without a l t e r i n g the matrix (Pharmacia, 1980). The major benefit of t h i s i s that s p e c i f i c 6 molecules, including many pharmaceutical^ important proteins, can be s p e c i f i c a l l y targetted and very e f f e c t i v e l y and e f f i c i e n t l y secured onto the charged s i t e s of the matrix without the need to add chemicals to the feed stock. In the instance of lysozyme and avidin recovery, once these have been secured onto the matrix, displacing the i n i t i a l counter-ions, the spent feed stock can be processed exactly as though the protein extraction had not taken place and with no loss of f u n c t i o n a l i t y in terms of gel strength, whipability or n u t r i t i o n a l value (Li-Chan et a l . , 1986). A t y p i c a l set of steps in the general operation of an ion-exchange column includes: 1. If the r e s i n i s being used for the f i r s t time, i t should be prepared for use following the manufacturer's d i r e c t i o n s . This involves, for the Duolite C-464, washing the r e s i n with acid, with deionized water, with a l k a l i and then water again followed by e q u i l i b r a t i o n buffer (Li-Chan et a l . , 1986). 2. The feed stock i s then fed in downflow through the packed bed u n t i l the predetermined percentage of product i s observed to be passing through the column outlet. At t h i s point, the re s i n bonding s i t e s are e s s e n t i a l l y occupied, except in the lower portion of the bed. The proportion of unexchanged s i t e s remaining at breakthrough depends larg e l y upon the feed flow rate and the d i f f u s i o n rates for the desired product in both the l i q u i d and s o l i d phases. A high feed flow rate gives a low contact time 7 which reduces the p r o b a b i l i t y that a particular molecule within the feed l i q u i d w i l l find an exchange s i t e prior to e x i t i n g the bed. The contact time should be chosen with care to optimize the balance between rate of production and e f f i c i e n c y of recovery. 3. Backwash the column with deionized water or a suitable buffer in order to remove non-adsorbed proteins from the r e s i n . 4. Pass a solution of counter-ions down through the column t'o elute the i o n i c a l l y bonded proteins. As the product e l u t i o n step proceeds, an u l t r a - v i o l e t spectrophotometer i s frequently used to monitor the concentration of protein in eluting buffers by comparing the absorbance of l i g h t at 280 nm wavelength between the protein-free eluting buffer and the process eluant. 5. Re-equilibrate the column for the next feed cycle by passing e q u i l i b r a t i o n buffer through the column to replace the eluting counter-ions with the ion most appropriate for the capture of product ions. The process i s now ready to return to the application of feed (Step 2). This cycle consisting of steps 2 to 5 continues u n t i l the feedstock i s exhausted, or a planned shut-down occurs, or in the long run u n t i l the resin is no longer able to carry out i t s function due to deterioration. Figure 1.1 shows schematically in flow chart form the sequence of procedures for the i n i t i a l preparation and c y c l i c operation. 8 Choose ion exchanger Choose starting buffer Swell gel if necessary and pack in suitable column Set up equipment • UV-monitor • Recorder • Fraction Collector » Pump ® Gradient Mixer Equilibrate (2—3 volumes buffer) Apply sample I Equilibrate sample if necessary Optimize separation Wash away unbound substances 1 Elute bound substances Desalt Regenerate gel Analyze separation FIGURE l . 1: Flowchart for a t y p i c a l ion-. exchange procedure Ion-Exchange Resins 9 The most basic d i s t i n c t i o n between resin types i s whether the counter-ions ( i . e . exchangeable ions) are p o s i t i v e l y or negatively charged. Negatively charged resins having p o s i t i v e l y charged counter-ions and are c a l l e d cation exchangers while p o s i t i v e l y charged resins with negatively charged counter-ions are termed anion exchangers. Within each type of exchanger, the various covalently bound ions can be c l a s s i f i e d according to the strength with which they bind'counter-ions. Strong ion exchangers maintain a state of complete ionization over a wide range of pH while weak exchangers are much more influenced by the e f f e c t of pH on the degree of ionic d i s s o c i a t i o n (Pharmacia, 1980). Table 1.1, below shows a sampling of covalently bonded i o n i c groups used in modern resins to produce strong, moderate and weak exchangers. Table 1.1 Functional groups used for ion exchangers (Pharmacia, 1980) ANION EXCHANGERS Aminoethyl (AE-) Diethylaminoethyl (DEAE-) Quaternary aminoethyl (QAE-) CATION EXCHANGERS FUNCTIONAL GROUP - O C H 2 C H 2 N H 3 -- O C H a C H a N - H ( C H 2 C H 3 ) 2 - O C H 2 C H 2 N * ( C 2 H B ) 2 C H 2 C H ( O H ) C H 3 Carboxymethyl (CM-) Phospho Sulphopropyl (SP-) - O C H 2 C O O --PO-oHa\" - C H 2 C H 2 C H 2 S 0 3 -10 Strong exchangers are formed using sulphonic and quaternary amino groups while the phospho group i s of intermediate strength and the others are considered weak exchangers (Pharmacia, 1980). Other factors to be considered when selecting a res i n type include the porosity of the matrix as determined by the degree of polymer cross-linking, the change in resi n volume during changes in pH, ionic strength and counter-ion, the chemical s t a b i l i t y of the re s i n in the solvent used, the physical d u r a b i l i t y of the res i n , and the mesh size and size d i s t r i b u t i o n of the re s i n . Factors Affecting Exchange Dynamics The degree of divinylbenzene cross-linkage in a sulphonated polystyrene re s i n determines the pore size within. A highly cross-linked matrix provides a large number of charged s i t e s while the small pore diameters prevent the entry of large molecules including most proteins. In order to allow proteins access to internal exchange s i t e s , and thus increase the exchange capacity, resins were developed with an open macroporous structure. These resins are generally produced in the form of small spherical beads which are available in a variety of sizes usually between about 16 and 400 mesh (U.S. standard sieve). The size and shape of the molecule as well as i t s surface charge d i s t r i b u t i o n are of great importance in b i o l o g i c a l ion-exchange processes. The resi n pore diameter can be chosen so that only particular molecules which are smaller than the pore diameter can enter the resin matrix. This allows for separation 11 in part by size exclusion as well as by ion-exchange since large ionic species can access only s u p e r f i c i a l bonding s i t e s on the re s i n , while the myriad internal s i t e s are reserved for smaller ions. Liquid-phase d i f f u s i o n c o e f f i c i e n t s for feedstock constituents can be increased by decreasing the l i q u i d v i s c o s i t y , which i s i n turn lowered by r a i s i n g the temperature. Higher d i f f u s i o n rates promote more rapid exchange kinetics and therefore allow for higher flow rates. Homogenization of b i o l o g i c a l l i q u i d s can reduce the v i s c o s i t y by exposing congealed or coagulated protein to high shear forces which break them apart to make them more free-flowing i Durance (1987) used a Manton-Gaulin high pressure (6.9 MPa), small o r i f i c e homogenizer to reduce the albumen v i s c o s i t y prior to feeding to the column. The major reason for homogenizing the egg albumen was to break up congealed lumps of protein in order to prevent plugging of the packed res i n bed. The contact time between the l i q u i d and resin greatly influences the product recovery e f f i c i e n c y . Since the attachment of any p a r t i c u l a r ion to an exchange s i t e is a stochastic process, some minimum contact time i s required in order to ensure a given l e v e l of product recovery. The presence of extraneous ions competing for bonding s i t e s hinders the e f f i c i e n c y of product recovery in many ion-exchange processes. The recovery of lysozyme and avidin is not p a r t i c u l a r l y hindered in t h i s way since their i s o e l e c t r i c points are markedly higher than those of most other proteins. Recovery of Lysozyme and Avidin 12 Lysozymes, basic proteins having an average size of approximately 15,000 daltons, are produced in animals, plants, and insect and fungal c e l l s (Durance, 1987). They exhibit bacteriocidal c h a r a c t e r i s t i c s since they can lyse certain b a c t e r i a l c e l l walls including those of Micrococcus lysodeikticus upon which a popular assay for lysozymes has been developed. In th i s assay, a l i q u i d suspected of containing lysozyme i s mixed with a s l u r r y of lysodelktlcus and then observed t u r b i d i m e t r i c a l l y . If mixing the feed l i q u i d with the dead c e l l s l u r r y causes an increase in the transmission of v i s i b l e l i g h t compared with a feed containing no lysozyme, then the percent increase in transmission gives a quantitative measure of lysozyme concentration (Joll&s et a l . , 1965). The dominant commercial source of lysozyme i s chicken egg albumen which contains approximately 88% water and 10.2% protein (Wilkinson and Dorrington, 1975). Lysozyme accounts for about 3.5%, while a v i d i n makes up only 0.05% of t h i s t o t a l protein ( L i -Chan et a l . , 1986). The recovery of lysozyme has been shown to be between 90 and 95% using Duolite C-464 cation exchange r e s i n in a packed bed (Li-Chan et a l . , 1986). This high recovery along with desirable physical and chemical q u a l i t i e s including a low degree of swelling/shrinking during c y c l i c changes in the l i q u i d phase ionic strength, good chemical inertness and physical s t a b i l i t y make t h i s resin a good choice. 13 Elution-Looping Technique The novelty of the process for which th i s system was designed and b u i l t i s the way in which both lysozyme and avidin can be recovered simultaneously. As mentioned e a r l i e r in th i s section, the concentration of lysozyme i s 70 times greater than that of avidin. Thus, i f both proteins were eluted after each column feeding, the presence of avidin i s almost t o t a l l y masked by the large lysozyme peak. In order to better separate the two peaks, the fact that lysozyme can be eluted using a lower ionic strength saline solution than that required for avidin elution (although the opposite case would be expected from their r e l a t i v e pi's (Durance, 1987)) i s used to advantage. The avidin i s allowed to remain on the column for a sp e c i f i e d number of feed/regeneration cycles while lysozyme is removed each time. As the quantity of avidin bound to the res i n increases with each cycle, i t s delayed removal produces a peak which i s larger and more pure than that from a single cycle. While the c l a r i t y and purity of t h i s delayed peak increases with the number of cycles between avidin elutions, the optimum has not yet been determined. This modified recovery process which Durance has termed \"elution-looping\" promotes the simultaneous recovery of two commercialy important proteins on a single column with high i n i t i a l p u r i t i e s and recoveries. 14 2. AUTOMATION OF LYSOZYME/AVIDIN RECOVERY Prior to discussing the pa r t i c u l a r approach taken to automating the recovery of these two well known proteins, i t should be noted that the intent of t h i s paper i s to provide a total-system prototype, both hardware and software, for the recovery of lysozyme and avidin using the elution-looping technique. It does not attempt to introduce any r a d i c a l l y new concepts in the general process of ion-exchange or in the equipment that i s used. It does, however> attempt to present, a system package that i s p a r t i c u l a r l y suited for both laboratory experimentation and j commercial production. The complete f l e x i b i l i t y regarding which and how many of the three columns are to be used enhance the experimental c a p a b i l i t i e s of the system. The \"staged shutdown\" c a p a b i l i t y allows for the unattended operation of the system during which time the software w i l l modify the operating mode to exclude any column(s) which become clogged with congealed egg white and continue processing with whatever column(s) remain. The operating l o g i c of the system w i l l be explained in more d e t a i l in the following sections describing the physical plant and the electronic control c i r c u i t s . Physical Plant Lysozyme and avidin have been recovered in both batch and continuous ion-exchange arrangements. While batch methods exploit the least expensive technology in their operation, they tend to be r e l a t i v e l y labour intensive and i n e f f i c i e n t in terms of the optimum use of r e s i n . If s u f f i c i e n t resin is used to ensure an accurate contact time between the re s i n and the feed l i q u i d , the res u l t i s a packed bed. If less r e s i n is used, then some form of agita t i o n i s required in order to insure a uniform contact time over a l l of the feed stock. Ion-exchange resins tend to be e a s i l y damaged during agitation with an impeller, either open, or enclosed as in a pump. Packed column continuous feed systems, on the other hand, are better suited to optimizing operating conditions including contact times. The number of columns used dictates whether the process is continuous or not on the basis of feed input. Multiple-column systems are generally operated as multiple individual columns which are fed in sequence. The feeding of a given column continues u n t i l product molecules begin to break through into the column effluent. At t h i s point, since the adsorption of ions i s a stochastic process, a s i g n i f i c a n t portion of the re s i n bed at the bottom of the column i s unsaturated. The exact proportion of under-used res i n varies d i r e c t l y as the flow rate, or inversely as the contact time. If a l l of the resin could be f u l l y loaded, then the capacity of the column would be increased to accommodate more feed, or the volume of the column could be reduced while treating the same quantity of feed. This end was accomplished by operating two columns in a cascade arrangement wherein the feed is introduced into the top of the f i r s t column, and i t s effluent feeds into the top of the 16 second column. The f i r s t column can then be allowed to approach saturation while the product ions which break through are captured in the second column. This is the same eff e c t as sampling a single t a l l column at i t s midpoint for saturation while not having to worry about losing product due to breakthrough. The benefit of the having the columns separated is that upon saturation of the f i r s t column, the feed i n l e t can be directed to the top of the second column while the f i r s t one is regenerated (eluted and re-equilibrated with the desired form of ion). In order for the thi s sequence to continue, a t h i r d column is required. Provided that the time required to load a column with the product ions i s greater than the time required for regeneration, only three columns are necessary. If the feed time is less than that required for regeneration, then either the feed step would have to be interrupted or an additional column would be needed. Two sets of names are used in r e f e r r i n g to the columns. In order to refer to a par t i c u l a r column regardless of i t s current function, each i s given an absolute name: \"A\", \"B\" and \"C\". In order to describe a given column with respect to i t s current function, r e l a t i v e names are also used: \"Primary\" (1°), \"Secondary\" (2<>) and \"Regenerating\" (R). Figure 2.1 i l l u s t r a t e s the use of three columns which operate in t h i s cascade sequence. The three columns exchange operating duties in a continuous cycle such that the most recently regenerated column becomes the secondary one in the two-columns se r i e s , the column that was the secondary becomes the 17 FEEJUUL A B R FEEQ_JJL BARREN FEED OUT L 1 B a) STAGE 1 R BARREN FEED OUT FF.Rn I N ^ 1 A t b) STAGE 2 £T B R c) STAGE 3 BARREN FEED OUT FIGURE 2.1: Cascade sequence for columns A, B and C as they operate i n the capacity of primary and secondary feed, and regeneration 18 primary, while the primary column begins the regeneration stage. During the cascade feeding of the primary and secondary columns, the remaining exhausted ( f u l l y loaded) column i s exposed to the following regeneration steps: 1. The egg white in the column is removed by backwashing with s t a r t i n g buffer in order to remove any unadsorbed protein prior to stripping off the product ions. 2. A weak saline solution eluant is passed downward through the column to 'remove' the lysozyme while leaving the avidin bound to the resin as per the elution-looping technique mentioned e a r l i e r . 3. If the s p e c i f i e d number of loops have been performed, removing only the lysozyme, then a strong saline eluant i s passed through the bed to remove the enhanced avidin peak. 4. If just the weak saline eluant has been used, then no further e q u i l i b r a t i o n i s necessary, however, i f the strong saline was used, then the resin must be equilibrated prior to returning the column to feed duty. This sequence of regeneration steps is thus applied to the column lab e l l e d 11R\" in each of the three STAGES shown in Figure 2.1. 19 Operating Sequence Nomenclature The various valve operating sequences must be la b e l l e d unambiguously in order to prevent errors in control programming. The following terms describe these sequences, beginning with the most fundamental d i v i s i o n and progressing to the most comprehensive. These terms w i l l be presented in c a p i t a l l e t t e r s wherever they are used to indicate these s p e c i f i c l o g i c a l d i v i s i o n s as defined below: STEP: The term STEP indicates a p a r t i c u l a r control setting for the 23 flow control valves, 3 pumps and the homogenizer. Each of these units can be either ON or OFF, that i s , energized or not. The ON/OFF patterns for the various STEPs are shown in Appendix A. STEP time durations are assigned by the operator either through the uploading of an existing method from diskette or through the d i r e c t interactive method. STAGE: For the column being regenerated, the term STAGE represents the t o t a l number of STEPS involved in the regeneration process. The number of STEPS in a STAGE varies dependent upon whether or not the avidin as well as the lysozyme is to be stripped from the column. For the column(s) being fed, the STAGE i s equivalent to the single feed control setting of the valves, feed pump and homogenizer. The actual duration of the STAGE is the greater of the length of time required for the sequence of 20 regeneration STEPS or the specified feeding time. CYCLE: If three columns are being used, then in one CYCLE each of the columns has occupied the position of \"primary\", \"secondary\" and \"regeneration\". CYCLE and STAGE are synonymous when running fewer than three columns since the control settings are no longer completely independent for feeding and regeneration. In the case of two active columns, the cascade feed can only take place i f the regeneration procedure i s completed for the second column. Since non-cascade feeding can progress while regeneration i s carried out in the second column, the feeding and regeneration can be carried out in p a r a l l e l but only in the forward d i r e c t i o n , as defined by the normal operation of three columns. For a single column, i t i s obvious that only one operation can be performed at a time, and that feeding and regeneration must take place in series rather than in p a r a l l e l . For one or two columns, then, the combined set of STEPs for feeding and regeneration is considered as a single sequence, making redundant the use of both of the terms, STAGE and CYCLE. Since CYCLE appears to better describe the completion of a l l required STEPs for each column, the term STAGE was dropped in the operation of just one or two columns. The incorporation of the elution-looping technique requires that a count be kept of the number of CYCLES run so that the avidin can be stripped from the column(s) a f t e r the appropriate CYCLE number. 21 BLOCK: This counter simply indicates the number of times that a v i d i n has been removed from the column. It increments each time the CYCLE counter reaches the predetermined value and i s not t r i v i a l since the number of CYCLES between subsequent avidin removals can be varied at any time during the process via the routine, MPC (manual process co n t r o l ) . Flow Path Design The cascade approach used to recover the target ions was incorporated into the system plumbing arrangement, with an attempt made to minimize the internal volume of the system while accomodating a wide range of flow rates for d i f f e r e n t size columns. Figure 2.2 i l l u s t r a t e s the l i q u i d flow path and valve arrangement. The arrangement of these valves with respect to the i r ON/OFF states was of c r i t i c a l importance in minimizing the number of valves required. Of the 23 e l e c t r i c solenoid valves used, six are simple two-way valves which either permit l i q u i d to flow through them or block i t s path completely (valves 9-14). The other 17 are three-way valves which can select either of two input streams and allow one of them to exit the valve (mixing valves) or can d i r e c t one input l i n e into either of two output l i n e s (diverting valves). These three-way vlaves cannot prevent l i q u i d s from flowing through them, but can only select the ro FIGURE 2.2: Liq u i d flow path, valve and a n c i l l a r y equipment arrangement appropriate flow path. If one of these valves i s operated to permit flow through one of the two possible pathways, then the alternate path i s blocked o f f . In t h i s way each valve can provide dual functions at any given time. Great care had to be taken to assure that a given physical arrangement of valves and possible flow paths would be able to accommodate every required flow arrangement, allowing for the cascade feed approach and the i s o l a t i o n of the regenerating column from the others through i t s various steps. This flow path i n t e g r i t y had to be maintained throughout the rotation of column duties from STAGE to STAGE. In tracing through the flow system, i t might appear that certain flow paths cannot be blocked off as required for a given sequence, however, the use of positive displacement pumps e f f e c t i v e l y provides the system with three more two-way valves as they are turned on or o f f . Once the valve placements within the system s a t i s f i e d the flow path c r i t e r i a , the matching of i n l e t l i n e s with pa r t i c u l a r input ports on the mixing valves, or outlet l i n e s with output ports on d i v e r t i n g valves was done to minimize the amount of time that each valve was required to be energized. This in turn minimized the heat loading from the solenoids which would be transferred to the product or to the surrounding environment. This might be a s i g n i f i c a n t heat source to be considered i f the system were to be operated in a refrigerated room. 24 Flow Handling Equipment Valves: The valves used in the prototype system are Burkert (West Germany) 24v D.C. e l e c t r i c solenoid valves designed for low pressure operation. Six two-way valves (Model 121-A), two three-way mixing valves (121-E) and 15 diverting valves (121-F) were used in the system with f i t t i n g s sized to accommodate the 1/4 inch (6.4mm) outside diameter, 1/8 inch (3.2mm) inside diameter p l a s t i c tubing used. • Pumps: Three p e r i s t a l t i c , positive displacement pumps were used to c i r c u l a t e the l i q u i d s through the system. These are e s p e c i a l l y suited for low pressure systems and allow i n f i n i t e adjustment of rotor speed over a wide range of l i q u i d flow rates; Since the r o l l e r s on the pump rotor squeeze the l i q u i d through the flow l i n e , no d i r e c t contact with the l i q u i d i s ever made, enhancing the lev e l of hygiene of the system. One pump moves feed l i q u i d (egg albumen) through the system; another provides backwashing buffer, while the th i r d one does double duty by providing either strong or weak eluent as required with the aid of a diverting valve. The pump speeds must be set manually on the con t r o l l e r for each pump, while the ON/OFF setting for each was made controllable by the software. 25 Columns:Quick disconnect f i t t i n g were used on either end of the columns to allow for the rapid replacement of columns of various sizes and aspect r a t i o s (ratio of height to diameter) during process optimization steps (not part of th i s study). The columns fabricated for t h i s system were 12\" (305mm) long, made from l\"i.d.(25.4mm) clear a c r y l i c tube, selected for i t s transparency and chemical inertness and sized roughly for the production of 25 l i t e r s per day of albumen. This processing rate was selected as a compromise between the minimum scale of operations which could provide for both laboratory scale research runs and small i n d u s t r i a l scale processing of up to approximately 250-300 l i t e r s per day at the l i m i t of the pumps. Figure 2.3 shows the construction of a t y p i c a l commercial column. Screens provide the primary support for the resi n column i n both upflow and downflow operation. F i l t e r s at either end serve to prevent r e s i n fines from escaping into the flow lines and to help prevent the re s i n bed from being blocked with congealed albumen. Homogenizer: A laboratory size Manton-Gaulin high pressure o r i f i c e homogenizer used by Durance (1987) was e f f e c t i v e in breaking up congealed egg white and resuspending the protein, however, i t s substantial size and weight made i t unsuitable for use in the portable s t y l e system that was developed. The major c r i t e r i a , besides p o r t a b i l i t y , for 26 Heavy wall _ glass column Acrylic jacket Interchangeable bed supports (Polyethylene, Nylon, Polypropylene, or TFE) TFE o-ring shield provides a leak free seal \\ Threaded acetal jacket cap Threaded acetal end cap TFE end fitting CTFE V4\"-20 to V4 \"-28 thread adapter (TOP AND BOTTOM ENDS) FIGURE 2.3: A typical laboratory style chromatography column with temperature control jacket (Kontes Scientif ic Glassware/Instruments) the selection of a homogenizer were the possible range of flow rates, continuous processing c a p a b i l i t y , and the exclusion of a i r from the product while processing. While the range of flow rates needed to be matched to the desired range of processing rates as defined by the pump capacities, the prevention of a i r bubble entrainment in the feed stream was c r i t i c a l to avoid protein denaturation r e s u l t i n g from the high surface free energy at the gas/liquid interface. Equipment using blendor-s t y l e rotating blades were thus unsuitable due to the large-scale entrainment of a i r and the batch operation s t y l e , since no continuous-flow sealed processing vessel was r e a d i l y a v a i l a b l e . The extremely high shear forces encountered in t h i s type of equipment might well be much greater than i s required tp l i q u i f y the congealed protein, even to the point of denaturing at least the larger individual protein molecules. The most suitable type of homogenizer found, within the above constraints, was a mortar and pestle s t y l e unit in which a Teflon (Reg.TM E.I. duPont de Nemours & Co.) rod rotates within a precision bore b o r o s i l i c a t e glass tube. The albumen can be continuously fed through the narrow gap between the rotor (pestle) and stator (mortar). The amount of shear to which the product i s subjected can be p r e c i s e l y controlled over a wide range by varying the speed of rotation and diameter of the pestle with respect to the fixed tube, and the feed rate, 28 which governs the length of time which the l i q u i d takes to pass through the shear zone. No a i r is entrained in t h i s type of unit and the shear forces can be maintained at a low enough l e v e l that bubbles are not created due to c a v i t a t i o n . Although t h i s was the unit of choice, i t s high cost (approx. U.S. $3000) precluded i t s use in th i s prototype system. The unit chosen was an ultrasonic horn with a t o t a l l y enclosed through-flow processing c e l l with cooling jacket. Although microscopic bubbles are formed and collapse during excursions of the vibrating horn t i p , the in t e n s i t y of the v i b r a t i o n can be controlled, and although the vigorous t i p o s c i l l a t i o n s transmit a considerable amount of heat to the l i q u i d , the cooling jacket can be used to maintain temperatures below that which w i l l cause the p r o t e i n to denature. Again, the flow rate of l i q u i d through the shear zone dictates, in large part, the degree of heat buildup in a given volume of l i q u i d flowing through the unit. Durance (1987) found, when using batch homogenized feed, that the albumen tended to recoagulate over a period of time while waiting to be processed. It was to a l l e v i a t e t h i s problem that a continuous-flow homogenizer was placed i n - l i n e between the refrigerated feed storage and the dual i n l e t feed f i l t e r s . 29 Tubing: The flow tubing used In the prototype Is Tygon (Reg. TM Norton Company) R-3603 formulation, a clear, f l e x i b l e p l a s t i c tubing with a 1/4 inch (6.4mm) outside diameter and a 1/16 inch (1.6mm) wall thickness. This tubing i s excellent for i t s smooth bore, c l a r i t y and a b i l i t y to bend sharply without collapsing, for quick and easy setups. It i s autoclavable at 121 degrees centigrade for 30 minutes at 15 pounds per square inch (psi) (103.4 kPa) pressure, has a useful operating temperature range of ^45 C to +74 C, and i t w i l l not age or oxidize. The maximum working pressure for t h i s tubing i s 44 psi (303.4 I kPa) at an ambient temperature of 21 C (Cole-Parmer Instrument Company 1987-88 catalogue, p558). The length of tubing In the system was kept short by expedient placement of valves on the mounting frame, although an optimization routine could be used to determine the very best possible arrangement. For the purposes of t h i s study, the time required to set up such a routine for computer aided optimization, in l i g h t of the r e l a t i v e l y i n s i g n i f i c a n t reduction in system internal volume, was deemed excessive. F i l t e r s : Since chicken egg albumen has a s l i g h t tendency to recongeal, even at room temperature after homogenization, two f i l t e r s were placed in p a r a l l e l between the homogenizer and the columns. The f i l t e r holders (Sartorlus D-3400, SM16508B), are constructed of clear polycarbonate with a small internal volume. 50mm diameter Whatman GF/D f i l t e r s (Whatman Ltd.) with a p a r t i c l e retention of 2.7 nm were used because of their high flow capacity, adequate f i l t r a t i o n capacity (generally used as a p r e f l i t e r for membrane f i l t r a t i o n ) , and low cost. Either f i l t e r can be clamped off from the system and changed independently while the process continues with the remaining f i l t e r . Secondary f i l t e r s are located at either end of each resin column. These play a dual role in preventing the r e s i n bed from becoming fouled should one of the p r e f i l t e r s become ruptured and in preventing the re s i n fines from escaping the column into the flow l i n e s . The main f i l t e r supports within the columns are discs of 40x60 mesh st a i n l e s s s t e e l screen next to the r e s i n bed. The f i l t e r material used in the columns was a synthetic non-woven clot h of fine but undetermined p a r t i c l e size retention held in place from the other side by a disc of coarse synthetic sponge-like material used in domestic scouring pads for dish washing. The spongy pads held the f i l t e r c l o t h t i g h t l y against the screen to prevent leakage of re s i n and/or congealed albumen around the sides of the f i l t e r . Support from both sides of the f i l t e r s was required since f l u i d flows through the columns in both upflow and downflow as required by the control program. 31 Pressure Transducers: Due to the tendency of the egg white to congeal and foul the f i l t e r s and resin, the operating condition of the f i l t e r s Is monitored to a l e r t the control program and the operator when a flow blockage i s imminent. The d i f f e r e n t i a l pressure is monitored across each of the three columns as well as across the i n l e t f i l t e r s . The transducers (Honeywell) have a nominal capacity of measuring ± 5 psi (34.5 kPa) according to the unit s p e c i f i c a t i o n s , however, a l l four of the transducers were tested and found to provide a useful signal output up to about +7.3 psi (50.3 kPa) and -5.0 psi (34.5 kPa). The re q u i s i t e DC power source was provided from the computer interface while the output voltage from each pressure sensor was monitored by a separate channel of the analog to d i g i t a l (A/D) converter board (described in a following section) v i a the screw terminal connector. These signals, once processed, provide the control program with information which allows i t to modify or terminate the operating routine to s u i t the physical s i t u a t i o n within the chromatography system. U l t r a v i o l e t Monitor: U l t r a v i o l e t (UV) l i g h t at a wavelength of 280nm i s absorbed by certain amino acid groups. The degree of l i g h t absorbance associated with these groups can be used to indicate the concentration of proteins which contain them. For the purpose of observing the concentration of eluting proteins, an i n - l i n e UV monitor unit (UV-1 Single Path Monitor Optical Unit and UV-1 Single Path Monitor Control Unit by Pharmacia Fine Chemicals) was placed between the processing columns and the two solenoid valves which control the flow of eluting l i q u i d to waste or to the appropriate product reservoir, depending upon whether or not the eluant contains protein at a concentration higher than some spec i f i e d threshold value. The analog signal coming from the UV monitor is fed to the A/D board as an input signal for the control software. The control- program monitors t h i s -signal in order to route the eluting l i q u i d to the appropriate vessel. The UV-1 output signal i s in the range 0-10 mv DC. Electronic Flow Control Hardware Figure 2.4 shows the general arrangement of components which perform the e l e c t r i c and electronic control functions required to translate the computer output into a working flow control system. DC Power Source: A 24 vol t (v) DC power source i s required to operate the flow control solenoid valves. Each valve takes an inrush current of approximately 1.1 amperes (A) and a holding current of 0.4 A. The power source used is one that was available from a previous project, a Power-One (Power-One Inc.) 7.2 A unit which proved adequate for the purpose. The output of t h i s unit was used as an input to the ELECTRICAL SOLENOID VALVES DT2817 32 channel DIG I/O INTERFACE (4x8-bit . ports) backplane connector IBM COMPATIBLE COMPUTER backplane connector DT2814 16 ch.(SE) A/D BOARD (12 b i t s ) VOLTAGE AMPLIF. 115v AC 115v AC UV UNIT CONTROLLER UV OPTICAL UNIT 20 conductor ribbon connector FIGURE 2.4: E l e c t r i c a l c o n t r o l system s c h e m a t i c 34 computer/system interface wherein the control signals from the computer d i g i t a l output lines were used to help switch individual valves on and off according to the control program. System Control Interface: The interface i s the power dispatching center of the system. Its inputs consist of the 24v DC output from the power source mentioned above, 115v AC power used to drive the homogenizer and three p e r i s t a l t i c pumps, and the LSTTL (large scale t r a n s i s t o r - t r a n s i s t o r logic) signals from the computer. The 2.4v (minimum) signal supplied by the d i g i t a l input/output (Dig I/O) board mounted in one of the computer card s l o t s arrives at the interface with a maximum 15.0 mA. This miniscule current must be amplified in order to switch on the flow control solenoids. A Darlington t r a n s i s t o r arrangement, as shown in Figure 2.5 for a single channel, i s used to achieve the desired e l e c t r i c a l output. The d i g i t a l output signal from the I/O board supplies the i n i t i a l base current for the TIP120 Darlington-connected s i l i c o n power tr a n s i s t o r s (Texas Instruments, Inc) which then passes 24v DC current at the required amperage through the 4N32 opto-isolator (Motorola) to one of the 23 solenoids. The four 115v AC outputs are triggered in a similar manner, but with the addition of a moving contact relay for each of the four channels. 35 T Y P E S TIP120, TIP121. TIP122 N - P - N D A R L I N G T O N - C O N N E C T E D S I L I C O N P O W E R T R A N S I S T O R S DESIGNED FOR C O M P L E M E N T A R Y USE WITH TIP125, TIP126, TIP127 • 65 W at 25°C Case Temperature • Min hFE of 1000 at 3 V , 3 A • 5 A Rated Collector Current • 50 mJ Reverse Energy Rating device schematic COLLECTOR Q r BASE O -Wv •» 150 n VA— < j 0> -t c < op to -* co ro mechanical data THE COLLECTOR IS IN ELECTRICAL CONTACT WITH THE MOUNTING TAB J_ e M O J A L L DIMENSIONS A R E IN INCHES absolute maximum ratings at 2 5 ° C case temperature (unless otherwise noted) Collector-Base Voltage Collector-Emitter Voltage (See Note )) Emitter-Base Voltage Continuous Collector Current Peak Collector Current (See Note 2) Continuous Base Current Safe Operating Areas at (or below)-25°C Case Temperature . . . Continuous Device Dissipation at (or below) 25°C Case Temperature (See Note 3) Continuous Device Dissipation at (or below) 25°C Free-Air Temperature (See Note 4) Undamped Inductive Load Energy (See Note 5) Operating Collector Junction Temperature Range Storage Temperature Range Lead Temperature 1/8 Inch from Case for 10 Seconds TIP120 60 V 60 V 5 V TIP122 100 V 100 V 5 V TIP121 80 V 80 V 5 V •. 5 A •a 8 A •• 0.1 A •»— See Figures 7 and 8 -\" 65 W •a 2 W • — — — 5 0 mJ -• — - 6 5 ° C to 150° C — * 6 5 ° C t o 150°C — • « 260° C • \"0TES: 1. Thaia valuai appty whan the baaa-amittar dioda i l opancircuilad. 2. Thli valua appllo* for l w < 0,3 mi , duly cycla < 10%. 3. Darata Mnaarlv to 1S0' C caaa tamperatura at tha rata of 0.62 w / * C or rafar to Diulpatlon Ooratl 4. Darata llnaarlv to 150*C fraa-alr tamparatura at tha rata of 16 rriW/°C or ratar to Diulpatlon Oa< 5. Thl i rating Ii tiaiod on tha capability of tha tramlitori to oparata lafaly In tha circuit of Figur V B a 2 - 0 V, Rg - 0.1 f l , V c c - 20 V. Energy » l c ! l - / 2 . nfl Curva, Figure 0, aiing Curva, Figure 10. a 2. L - 100 mH, R r in,2 - 100 f i . TEXAS INSTRUMENTS 5375 I M I I H I ' O H A I I II PO%1 O f F l C f BOH » 0 l l • D A L L A I r t X A t » t l l » FIGURE 2.5: D a r l i n g t o n - c o n n e c t e d N-P-N s i l i c o n power t r a n s i s t o r s 36 Analog to D i g i t a l (A/D) Board: Most system monitoring devices produce an analog signal based on either voltage or current l e v e l s . In order for d i g i t a l computers to perform any operations using t h i s data, i t must be converted to binary format. The A/D board takes a 'snapshot' of the measured parameter and then produces a d i g i t a l representation of i t s value. The A/D board selected for use in t h i s system i s the DT2814 by Data Translation (Marlboro, MA). It is a hal f -size board including 16 single-ended (SE) analog input channels, a 12-bit A/D converter and a combination hardware/software programmable pacer clock to set the sampling frequency. The maximum sample throughput for the board is 25 kHz. Figure 2.6 gives the general layout of the card and i t s connection to the IBM PC sty l e Bus (PC/XT/AT), while Figure 2.7 shows the user pin assignments. The general operation of the board i s carried out by writing to and reading from two registers located in computer memory at the Base address 220H ( i . e . Hexadecimal 220) and Base+1 (221H). A WRITE command to the Base, or control/status register i s used to set the control b i t s . Bits 0-3 are used to inform the multiplexer which one of the 16 input channels i s to be sampled. Bit 4 (ENB) enables or disables the on-board pacer clock for continuous sampling, while b i t s 5-7 specify the decade di v i s o r for the hardware-jumpered base frequency for 37 16 Analog Inputs 16SE Analog Inputs MUX 10O i o n +12V -12V Address Decode Control Address Clock I B M PC BUS Sample & Hold 12-bit A / D Cont ro l & Status l o g i c B Data Line Interrupt FIGURE 2.6: Card layout for DATA T r a n s l a t i o n DT2814 A/D card PIN FUNCTION PIN PIN FUNCTION Channel 0 1 2 Channel 8 Channel 1 3 4 Channel 9 Channel 2 5 6 Channel 10 Channel 3 7 8 Channel 11 Channel 4 9 10 Channel 12 Channel 5 11 12 Channel 13 Channel 6 13 14 Channel 14 Channel 7 15 16 Channel 15 Power Ground 17 18 Analog Ground +12V Out 19 20 -12V Out FIGURE 2.7: DATA Trans la t ion DT2814 A/D card user connections 38 sampling rate. The useable sampling frequency range i s 0.005 Hz to 20 kHz, however, for the purpose of the control software, continuous sampling is not required but rather single values are requested p e r i o d i c a l l y by the control program. Within the program, a channel i s selected and that number along with a zero value for Bit 4 (ENB) are written to the Base address. When thi s register receives the value sent, the on-board multiplexer (MUX) selects the desired channel and the sampling sequence is triggered. To read the d i g i t a l value, a READ command should be sent to the Base address which now returns status information regarding the data conversion. Bits 0-3 s t i l l contain the selected channel number, for v e r i f i c a t i o n , and Bit 4 i s s t i l l the enable/disable f l a g . B i t 5, however, returns a value of 1 i f the A/D conversion i s in progress or 0 i f i t is complete. Bit 6 (ERR) returns a 1 i f an error i s encountered during a conversion as a resul t of too fast a clock speed or a f a i l u r e to clear the data register prior to the next conversion. Too high a clock speed results in a second conversion being i n i t i a t e d before the f i r s t one is complete. B i t 7 (FINISH) i s set (1) i f the conversion i s complete and the data register value is v a l i d , or is cleared (0) i f the conversion i s incomplete and the data register value is in v a l i d . Where samples are taken in the control program, the selected channel is written to Base and then the FINISH b i t i s monitored u n t i l set, at which time the data register (Base+1) double byte i s READ. In reading the data, two separate READ statements to BASE+1 are required: the f i r s t y ields the most s i g n i f i c a n t 8 b i t s while the second retrieves the least s i g n i f i c a n t 4 b i t s . The 4 b i t s occupy the most s i g n i f i c a n t positions of the second byte read at Base+1. In order to arrive at a single meaningful value from the combination of these two values, the f i r s t one must be« multiplied by 16 ( i . e . shifted four binary columns to the l e f t ) while the second must be divided by 16 in order to compensate for the fact that i t occupies the l e f t (most s i g n i f i c a n t ) half of the second word read. Signals get from the sampling instruments to the A/D J)oard v i a a screw terminal (DT757) and 20 conductor ribbon cable. D i g i t a l Input/Output (I/O) Board This Data Translation (Marlboro, MA) board, DT2817, is a h a l f - s i z e card containing 32 channels. These are programmable for either input or output in groups of 8 channels per port for each of 4 ports. The Base address is factory set at 228H, although i t can be changed v i a on-board jumpers, and contains the control r e g i s t e r . Addresses Base+1 to Base+4 contain the ON/OFF status for 40 each of the 4 ports. A WRITE statement containing the I/O status of the 4 ports must f i r s t be sent to the Base address. Only the four least s i g n i f i c a n t b i t s are used in t h i s statement: a BIT 0 value of 0 sets port 0 (channels 0-7) as input, while a value of 1 sets them for output. The same pattern holds for BITS 1-3 for ports 1-3 (channels 8-32). For input signals to the card, a BIT value of 0 ( l o g i c a l low) corresponds to a maximum voltage of 0.8v at -0.2mA, while a value of 1 ( l o g i c a l high) corresponds to a minimum 2. Ov at 20.0U.A. Output signals must a maximum 0. 4v at 24.0mA for a l o g i c a l 0, while a high signal must be a minimum of 2.4v at -15.0 mA. The DT2817 i s connected to the IBM-PC/XT bus with pin assignments as shown in Figure 2.8. The 32 I/O l i n e s are connected to the system interface by a 50 conductor ribbon cable with pin assignments as shown in Figure 2.9. Process Control Computer The computer used to monitor and control the ion-exchange system i s an IBM PC/XT compatible unit. It contains a multi-purpose card from which the clock is extensively used in the control routines. The on-board memory contains 640 kbytes of random access memory (RAM). The computer clock operates at the standard PC/XT 41 PIN MNEMONIC SIGNAL DESCRIPTION Al A2 A3 A4 A5 A6 A7 A8 A9 AIO All A12 A 1 3 A14 A15 A16 A17 A18 A19 A20 A21 A22 A23 A24 A25 A26 A27 A28 A29 A30 A31 BI B2 B3 B4 B5 B6 B7 B8 B9 BIO Bll B12 B13 B14 B15 B16 B17 B18 B19 B20 B21 B22 B23 B24 B25 B26 B27 B28 B29 B30 B31 -I/O CH CK +D7 +D6 +D5 +D4 +D3 +D2 +D1 +DO -I/O CH RDY +AEN +A19 +A18 +A17 +A16 +A15 +A14 +A13 +A12 +A11 +A10 - • +A9 +A8 +A7 +A6 +A5 +A4 +A3 +A2 +A1 +A0 GND +RESET DRV +5V +IRQ2 -5VDC +DRQ2 -12V RESERVED +12V GND -MEMW -MEMR -IOW -IOR -DACK3 +DRQ3 -DACK1 +DRQ1 -DACKO CLOCK +IRQ7 +IRQ6 +IRQ5 +IRQ4 +IRQ3 -DACK2 +T/C +ALE +5V +OSC +GND No Connection Data Bit 7 (MSB) Data Bit 6 Data Bit 5 Data Bit 4 Data Bit 3 Data Bit 2 Data Bit 1 Data Bit 0 (LSB) No Connection Address Enable No Connection (MSB) No Connection No Connection No Connection No Connection No Connection No Connection No Connection No Connection No Connection Address Bit 9 Address Bit 8 Address Bit 7 Address Bit 6 Address Bit 5 Address Bit 4 Address Bit 3 Address Bit 2 Address Bit 1 Address Bit 0 (LSB) Ground Reset Driver +5 Volt Power No Connection No Connection No Connection No Connection No Connection No Connection Ground No Connection No Connection I/O Write Command I/O Read Command No Connection No Connection No Connection No Connection No Connection 6MHz No Connection No Connection No Connection No Connection No Connection No Connection No Connection No Connection +5 Volt Pover 14.318MHz Ground FIGURE 2.8: D i g i t a l J / O card backplane connections for DATA T r a n s l a t i o n DT2817 42 SIGNAL NAME PIN NO. SIGNAL NAME Digital Ground Port 0, bit 0 Port 0 Port 0 Port Port Port Port Port bit bit bit bit bit bit bit +5V Out(lA max) Dig Dig Dig Dig tal Ground tal Ground tal Ground tal Ground Digital Ground +5V 0ut(lA max) Port 2, bit 0 Port Port Port Port Port Port Port bit 1 bit 2 bit bit bit bit bit rorr. u n / Digital Ground 1 3 5 7 \"9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50 Digital Ground Port 1, bit 0 Port 1, bit 1 Port 1, bit 2 Port 1, bit 3 Port 1, bit 4 Port 1, bit 5 Port 1, bit 6 Port 1, bit 7 +5V 0ut(lA max) Digital Ground Digital Ground Digital Ground Digital Ground Digital Ground +5V 0ut(lA max) Port 3, bit 0 Port 3, bit 1 Port 3, bit Port 3, bit Port 3, bit Port 3, bit Port 3, ' \" Port 3 bitbit 2 3 4 5 6 7 r o r i J I u i i / Digital Ground FIGURE 2.9: D i g i t a l I/O card user connections for DATA T r a n s l a t i o n DT2817 frequency of 4.77 MHz. The only additions to the system were the two I/O cards used to interface with the process monitoring and control equipment. The control system was purposely kept as simple as possible so that the f u n c t i o n a l i t y of the software was not overcome by requirements for expensive graphics cards and massive memory upgrades. While a real-time color graphic display of the physical system is often more desirable than text only, the costs involved in developing and accomodating such*a display are high and unwarranted at th i s time. 44 3 . PROCESS CONTROL SOFTWARE Mandatory Control Functions The most fundamental requirements of the control software include monitoring the condition of the system v i a electronic sensors, accounting for the elapsed time of various sequence steps and providing real time control output signals to the physical flow control system. As in any decision-making process, i t helps to have an up-to-the-minute account of pertinent facts. Those most indicative of the state of the current ion-exchange process include the pressure drop across each of the columns and across the i n l e t albumin f i l t e r s , the presence or absence of proteins in the eluted f r a c t i o n of the column undergoing the regeneration procedure, and the temperature and flow rates of the system l i q u i d s . Due in part to the costs involved in securing extra sensors for thermal and volumetric or mass flow measurement and in part to the r e l a t i v e ease with which both the temperature and flow rates can be controlled, only the pressure sensors and u l t r a v i o l e t monitor were used. Elapsed times for the various STEPs involved in the process were used as the basic c r i t e r i a for trig g e r i n g STEP changes.Since during the course of a run, certain parameters are subject to some degree of var i a t i o n due to changes in physical and/or chemical properties of the organic feed l i q u i d and the res i n , an event-controlled sequence i s in theory preferrable to a r i g i d , timed one. Event control allows for optimal use of the system while timed control has the benefit of s i m p l i c i t y in terms of the sensing equipment requirements and software structure. Since no suitable on-line assay has been perfected for the detection of lysozyme in the albumin leaving the primary feed column, a timed procedure becomes the default method of choice. Output control signals are stored in array form on the program diskette and are uploaded as arrays into computer memory during the i n i t i a l i z a t i o n procedure prior to the operator's interactive p a r t i c i p a t i o n in beginning a run. The control values for each STEP are stored as four 8-bit words with each b i t contributing the ON/OFF status of one output l i n e from the d i g i t a l I/O card to the control interface. These array values are fixed according to the physical structure of the flow control system and cannot be altered from within the program. Control Enhancements The basic control of the ion-exchange system presented herein i s r e l a t i v e l y s traight forward and simple as i s indicated by the modest length of the previous section. At t h i s point, the control program would operate in much the same manner as a mechanical clock used to trigger a signal that would increment the STEP number with i t s corresponding control output values. No information would be passed along to the operator, and a l l control values would have to be assigned to fixed arrays by editing the program between runs. However, since few operators want to be bothered with having to modify the software for each change of conditions, an interactive program which provides menus, prompts and v a l i d i t y checks for required input data is a far more useful vehicle through which to accomodate rapid set-up procedures. The following sections outline these and other enhancements which add to the f l e x i b i l i t y and user-friendliness of the control environment. Interactive input: Interactive data entry provides a fast and safe method of allowing operating parameters to be set and modified by the operator without his/her having to modify the control program. It is a quick data entry method since a l l potential changes are presented on the monitor in a l o g i c a l progression and important value assignments are not e a s i l y overlooked. Information i s prompted for as required. The safety of data entry i s augmented since the program can screen the input data against pre-set l i m i t s to prevent untenable s i t u a t i o n s . The use of menu structures allows the operator to select particular parameters for which to assign values or simply to view exi s t i n g ones without having to progress through the entire parameter l i s t each time. Once presented with a menu of options, a selection i s generally made by entering the line number to i t s l e f t . This allows the operator great f l e x i b i l i t y and speed in adressing particular parameters via single key stroke selections. If data input i s called for once the selection has been made, values are prompted for on 47 the screen, and once entered, they are tested to ensure program i n t e g r i t y . Appropriate error messages are displayed in the event of inappropriate entries. Parameters which can be assigned values prior to or during a run are described in the following sections. Visual display: Just as the software requires certain information in order to provide appropriate control sequences for the system, the operator also must remain informed as to i t s state so that he/she can make knowledgeable adjustments to the control parameters as they are needed. System information i s displayed in text format as the simplest case, although a graphical display of the physical system would be more a e s t h e t i c a l l y pleasing. The monitored and calculated parameter values are updated several times per second. Figure 3.1 shows a t y p i c a l screen during automatic process control (APC). At the top of the screen, the t i t l e indicates that the system i s being a c t i v e l y controlled by the program; the alternative to automatic control is manual process control (MPC) which can be accessed by pressing the \"M\" key at any time during the course of the run as noted in the bottom lin e of the screen. The second lin e provides information on the number of columns operating and whether or not a staged shut-down (described in a following section) is in progress, as well as 48 AUTOMATIC PROCESS CONTROL VER 1.0 AUG/88 OPERATING MODE NORMAL TIME:00:35:42 BLOCK LENGTH: 2 c y c l e s OPERATOR: ACM BATCH: 4 BLOCK 1 CYCLE: 2 STAGE: 3 STEP : 2 VALVE STATE VALVE STATE PUMP STATE DIFFERENTIAL PRESSURE (ps'id) 1 OFF 13 OFF 1 OFF Threshold: 7.0 2 OFF 14 ON 2 OFF Column 1 0.2 . 3 OFF 15 OFF 3 ON Column 2 -0.5 4 ! OFF 16 ON Column 3 0.4 5 OFF 17 ON F i l t e r s 0.2 6 OFF 18 OFF Flag 0 7 ON 19 OFF SONICATOR UV PEAK DETECTION 8 OFF 20 OFF OFF Threshold (Au): 2.0 9 OFF 21 ON Reading (Au): 0.0 10 OFF 22 OFF 11 OFF 23 OFF BATCH VOLUME: 25.000 (L) 12 OFF FEED FLOWRATE: 5.000 (L/HR) STEP STAGE CYCLE BLOCK RUN SET DURATION: 0: 0.5 0: 4.4 0:13.2 REMAINING TIME: 0: 0.1 0: 3.3 0: 1.9 ELAPSED TIME: 0: 0.4 0: 1.1 0:11.3 0 21.1 0:21.1 Press M to enter MANUAL PROCESS CONTROL FIGURE 3.1: T y p i c a l APC (Automatic Process Control) screen 49 the current time. Line 3 shows the values of BLOCK length, which is the number of elution cycles per column between removals of avidin (as per the elution-looping concept discussed in section 2 of t h i s paper), the Operator I d e n t i f i c a t i o n l a b e l , and the Batch Number. Line 4 shows the current values of the sequence i n d i c i e s , BLOCK, CYCLE, STAGE and STEP. The left-most two thirds of the screen from l i n e 5 to l i n e 17 are the state indicators for a l l controlled equipment. The accuracy of the displayed values, ON/OFF, is ensured by having them read from the d i g i t a l I/O card output latches rather than simply assigning them the intended values harboured by the software. This feature is i n v i s i b l e to the operator, although i t is a useful trouble-shooting tool when comparing the actual outputs to the intended ones as shown in Appendix A. The right t h i r d of the screen displays the current values of a l l monitored variables as well as the threshold values which when exceeded are used to trigger some form of control event to override the normal timed sequences. Below these are l i s t e d the Batch Volume and Feed Flowrate for general information only since they are neither monitored nor controlled by the program. In a commercial operation i t may well be expedient to monitor the various l i q u i d flow rates for both system logging and control purposes. The next four lines of the display show the a l l o t t e d , 50 remaining and elapsed time for each of the control sequence times. While th i s screen arrangement is not necessarily optimized for the usefulness of a l l data displayed or the speed with which i t can be digested by the operator, i t is functional and adequate for this version of the program. Default drive path: Menu 1 asks the operator to specify what type of disk drive the computer uses, and from th i s the default drive path is assigned Drive A for single floppy drives, Drive B for dual floppy drives, and Drive C for hard disk systems. This path can be changed at any time during the pre-run set up procedure or during the program v i a Menu 2 . Even when the default path is sp e c i f i e d , i t can be overridden during storage and r e t r i e v a l operations on method f i l e s by specifying the method f i l e path in response to the prompt. If no path is s p e c i f i e d , then the default path is used. F i l e handling routines: In most chromatographic work, some attempt i s made to optimize process parameters for the recovery of s p e c i f i c products from a given feed stock. A l i s t i n g of these run parameter values is ca l l e d the \"method\". When a method i s to be used more than once, i t i s very desirable to store i t in a method f i l e on a computer diskette for subsequent r e t r i e v a l . This not only speeds the set-up procedure greatly, but also 51 prevents the incorporation of typographical errors into the method. Three routines have been written for the storage, r e t r i e v a l and output to printer of method f i l e s . These routines are accessed from the th i r d menu in the i n i t i a l i z a t i o n section of the program. Upon invoking either the storage or r e t r i e v a l routines, the option exists to view the method f i l e names prior to entering the name of the selected f i l e . The operator may also decline t h i s option and enter the drive path and f i l e name at the prompt. If the default drive path (as established at the outset of the program) i s to be used, then only the f i l e name needs to be entered. If the view-files option is selected, then a search i s made of the method f i l e disk in the speci f i e d drive path. Only f i l e s having the extension, \".SET\" (indicating \"set-up\" data) are selected in a directory search. The operator does not have to concern him/herself about the use of the extension since the method storage routine automatically attaches i t to any f i l e s to be stored. When the directory search produces f i l e names ending with \".SET\", they are piped to a DOS SORT routine for alphabetizing and are then redirected into a temporary f i l e from which they are recalled for display in blocks of 25 names. Once the desired f i l e has been i d e n t i f i e d , i t s name is keyed in by the operator. If this routine was accessed through the method storage routine, then i t may be that the operator simply wants to avoid overwriting an existing f i l e 52 or to ensure overwriting the correct one. If the method f i l e s are viewed via the r e t r i e v a l routine, then once the f i l e name has been entered, the contents of the f i l e are up-loaded and assigned to the appropriate parameter labels. In order to prevent accidentally overwriting method f i l e s and/or \"crashing\" the program , a wide variety of DOS f i l e error handling interrupts have been accomodated in the program's control structure. Thus, even i f an operator keys in an ex i s t i n g f i l e name when down-loading a new method, the program w i l l a l e r t him/her to the fact and offer the alternatives of confirming the action, assigning a new f i l e name, or aborting the procedure. If an attempt is made to up-load a method from a non-existent f i l e , the operator is alerted to the fact, and the options of re-entering the f i l e name or aborting the procedure are presented for alternate action. The data stored in these method f i l e s includes the STEP time values for a l l system operating modes, and the values of the process constants presented in Menu 4. Staged shutdown: The concept of a system which can continue to operate in the event that one or two of the three columns becomes fouled with congealed albumin was incorporated into the program to promote the highest possible production from an unattended system. Such a system could be l e f t to operate 24 hours per day with only a single work s h i f t to maintain i t . Rather than providing a system with a simple alarm and a t o t a l shutdown when a blockage occurs, the software brings the impending blockage to the attention of the operator when the pressure drop across one of the columns or the i n l e t f i l t e r s exceeds a lower overpressure threshold. If the p a r t i a l blockage i s not removed and the high overpressure threshold i s reached, then the feed pump i s stopped u n t i l the regenerating column has completed i t s regeneration sequence and is ready to have albumen passed through i t . Once the regeneration i s complete, process control is passed from normal 3-column operation to the routine APC which causes the fouled column to be backwashed with phosphate buffer to remove the feed l i q u i d from the resin bed. Process control i s then passed to a routine which uses a modified control sequence to permit operation of the two remaining columns while keeping track of the elapsed number of CYCLES required for the elution-looping procedure. The process continues with these two columns u n t i l the offending column i s cleaned and returned to service, or one of the remaining two becomes fouled or a t o t a l shutdown i s effected by a programmed endpoint or a manual intervention. If one of the two columns i s badly fouled, then a sequence of events occurs which i s similar to that for the t r a n s i t i o n from three columns to two, leaving one column s t i l l active. The staged shutdown method of operation thus provides for a well controlled system response in the event of a column or f i l t e r f a i l u r e due to fouling. This control strategy is optimized for continuous production work with a minimum of operator intervention. Column sel e c t i o n : At the beginning of the program, Menus 2 and 2.1 provide for the selection of the i n i t i a l l y active column(s). A l l three columns, any combination of two, or any single column can be selected. This f l e x i b i l i t y f a c i l i t a t e s both commercial and research work. Manual process control: j This mode i s entered from any of the automatic control routines by depressing the \"M\" key. Several options are available while the status of the flow control hardware remains fixed as i t was when the key was pressed. From the main MPC menu, STEP time durations and process constants can be viewed and/or changed, equipment ON/OFF status can be toggled for any valve, pump or the homogenizer, the UV monitor baseline can be set, and a DOS s h e l l can be set up using the SYSTEM command to allow other DOS a c t i v i t i e s to be run without exiti n g the control program. The f i n a l s e l e c t i o n permits the return of control to the automatic routine from which the MPC c a l l was made. Alarms: Both audible and v i s u a l alarms are provided to a l e r t the operator to overpressure conditions within the system in 55 order to allow time for correction of the problem before a staged shut down takes place. A warning message is also presented at the bottom of the screen. The audible alarm can be shut off by a keystroke, while the message remains u n t i l the si t u a t i o n is remedied. Audible \"beeps\" are also used to indicate incorrect or out of range data entries prior to reprompting for correct data. Start time options: This feature allows the process to be i n i t i a t e d either immediately or at some sp e c i f i e d date and time. The delayed s t a r t option is l i k e l y to be most b e n e f i c i a l to researchers who often find i t d i f f i c u l t to conduct chromatographic runs within the bounds of normal working hours. With th i s option, a run can begin unattended at any time of day or night and on any date so that . the products of the run can be ready for analysis as soon as the s t a f f arrives in the morning. This c a p a b i l i t y could speed up research involving long run times. Step time assignments: As selected from Menu 2, option 3 allows process STEP durations to be assigned to a l l STEPs having the same function in each of the three operating modes (one, two or three columns active) simultaneously. When thi s selection is made, Menu 2.3, STEP descriptions, is presented without reference to any particular operating mode. Once entered, these durations are then written to the STEP time array variables for each of the modes. STEP times for individual modes can be altered without a f f e c t i n g the time duration of similar STEPs in any other mode. The current STEP time durations can also simply be viewed for a l l or any single operating mode without being queued for changes. Run completion setpoint: The endpoint for a run can be prior to beginning the run or at any time within the run by specifying the BLOCK, CYCLE, STAGE and STEP number at which termination should occur. When these values match the corresponding ones from the screen display the process is halted and a message is displayed to show whether the shutdown was planned or occurred due to malfunction. By setting the endpoint prior to beginning a delayed s t a r t run, the process can be made, for instance, to sta r t a run during the night and stop after one CYCLE or one BLOCK, whether or not the apparatus i s attended by an operator, so that analyses can be undertaken immediately as personnel are available. Automatic column cleaning: When any column(s) is (are) r e t i r e d from the process due to fouling, i t (they) is (are) backwashed with phosphate buffer to remove any albumen and thus prevent i t from being stored at room temperature in the re s i n bed for prolonged periods of time. The same treatment i s accorded a l l active columns at the normal termination of a run. 57 Program Structure and Operation The physical operation of the system involves fi v e basic steps: 1. Turn on power to a l l system components including: - computer - printer - DC power source - control interface - pumps (ensure correct pumping speed) - homogenizer (ensure correct intensity) - d i f f e r e n t i a l pressure transducer module - UV monitor detector and controller 2. Place the program diskette in drive A for single and dual floppy drives. 3. (a) For floppy drives, simply type in \"IX\" and then press \"ENTER\" (or \"RETURN\" depending upon the keyboard), and the program, which resides in the diskette f i l e \"IX.EXE\", w i l l begin as soon as i t is read into the computer memory. (b) For hard disk drives having the program in storage, prefix the program name with the appropriate drive path (see DOS manual for d e t a i l s ) . 4. Enter the run data required via the interactive menus 58 5. Start the automatic control portion of the program via the \"Start time options\" selection from the Main Menu. The general control structure of the computer program is shown in Figure 3.2 with arrow heads representing the d i r e c t i o n of control flow. Lines having arrows on both ends indicate subroutine CALL and return sequences. Included in the figure are the most prominent menu- and non-menu subroutines. The menu-, or interactive portions of the program are shown with the available selections as found in the program. The non-menu routines are comprised of the i n i t i a l declaration statements and the assigning of default values to control, STEP duration and process constants arrays. At the st a r t of a new run, the Main Menu i s presented to the operator. The seven options are presented in the order in which they w i l l generally be used in the course of a run. The f i r s t option, \"System Configuration\", allows the user to specify what type of disk drive arrangement i s being used, which and how many columns are to be used, whether or not the columns are to be equilibrated after the stripping of lysozyme with weak saline solution, and how many CYCLES are to be run prior to the removal of avidin from the column(s). The default values provided are: the previous entry for drive type (stored after each run), a l l three columns, no e q u i l i b r a t i o n , and the previous entry for the elution-looping CYCLE number, respectively. Option 2, \" F i l e s (Recall;Save;List)\", provides access to the f i l e handling routines through which method f i l e s can be up-MAIN DECLARATION STATEMENTS INITIALIZE CONTROL AND STEP TIME ARRAYS BEGIN CALL BLOCK. CALL MENU 1 CALL MENU 2 CALL APC CALL QUIT ERROR HANDLING ROUTINES END MENU 1: ESTABLISH DISK DRIVE CONFIG. MENU 2: MAIN MENU 1. PROCESS CONFIG. 2. METHOD FILES 3. STEP DURATION 4. PROCESS CONSTANTS 5: START TIME OPTIONS 6. SHELL (to DOS) 7. QUIT MENU 2.1 MENU 2.2 MENU 2.3 MENU 2.4 MENU 2.5 Interactive portion of program L FIGURE 3.2: General control structure of process software 60 loaded to the program arrays for use in the current run, down-loaded from the run arrays to disk f i l e , or l i s t e d to the printer (LPT1). Option 3, \"STEP durations\", allows the operator to view the current length of STEP times in any or a l l of the operating modes, from normal operation and staged shutdown modes. Values can be changed by d i r e c t keyboard input at thi s time. The new values w i l l be used in the automatic control portion of the program, but in order to save them for future use they must be •i stored on disk using option 2. Option 4, \"Process Constants\", shows a l i s t of parameters and their current values. Some of these, such as 'Operator ID' and 'Run number', are provided s t r i c t l y for the sake of run documentation while others such as the pressure and UV threshold values allow for the s e n s i t i v i t y of the control system to be varied. The A/D and Dig I/O card base addresses can be changed from the factory-set values should i t become necessary. Most users w i l l never change the default settings supplied with the program, however, when the method f i l e i s stored v i a option 2, the current address values are included. Option 5, \"Start Time Options\", provides entry to the automatic control routines. The run can be started immediately or at any time specified using the delayed-start option. Option 6, \"SHELL\", i s not required to be used, however, i t allows the operator to temporarily i n s t a l l a second copy of COMMAND.COM with which to run other DOS software while keeping the control program resident and ready to go. As outlined in the 61 DOS documentation, the resident program can be reentered by typing \"EXIT\" at the DOS prompt for the second l e v e l C0MMAND.COM. Option 7, \"Quit\", brings up another menu which prompts for confirmaton of t h i s action before allowing the program to end, and allows for the l a s t minute saving of the method f i l e before th i s data i s l o s t . When a l l of the run parameters have been set and the run has been i n i t i a t e d , control passes to the subroutine, APC (automatic process c o n t r o l ) , which selects the proper operating mode based upon which column(s) are to be used. For the sake of t h i s process description, the normal three-column configuration w i l l be considered. The other operating modes, for two or one column(s), follow a s i m i l a r procedure. For the operation of three columns, then, control would be passed to the subroutine, \"NM\" (normal mode), which would then provide the following general sequence of events for control: - Elapsed times are i n i t i a l i z e d for RUN, BLOCK, CYCLE, STAGE and STEP variables as a \"DO LOOP\" i s entered for each of the l a t t e r four parameters ( r e c a l l that STAGE is eliminated i f fewer than three columns are used). The STEP loop is the inner-most, while i t contains a STEP time loop in which the basic control of the system is located. - Within the STEP number loop, but prior to the s t a r t of the time loop, the STEP number is monitored to ensure that the avidin removal STEP is not executed unless the current CYCLE number matches the number of CYCLES specified for elution-looping. As well, the t o t a l time for the 62 regeneration STEPs is compared with the cascade feed duration (cascade or non-cascade time, depending upon which column is being fed, for two columns; non-cascade time for one column) to determine the t o t a l STAGE duration (or CYCLE for fewer than three columns). - The flow control output array values are then sent to the Dig I/O board to establish the correct l i q u i d flow paths and pump/homogenizer assignments for the f i r s t STEP of the f i r s t STAGE (column A is primary, B is secondary, C is regenerating). - The s t a t i c portion of the run time v i s u a l display i s established to which the values to be updated with each pass of the control loop w i l l be superimposed. As well, the column/feed-filters overpressure alarms are i n i t i a l i z e d . - The STEP time loop i s entered within which a l l of the automatic control functions are executed for the current STEP. The loop is exited when the STEP time is complete or when the regenerating column is ready to become the secondary column for the next STEP, which ever requires the most time. - Within th i s STEP time loop the visu a l display of the processing system parameter values is updated with each pass through the loop. - The ON/OFF state of the feed pump is monitored as the routine controls i t so that the elapsed feed time, rather than the elapsed clock time, signals the end of the loop 63 and the beginning of the next STEP and i t s corresponding system control output. - The pressure transducers are next tested against the set l i m i t s to check to overpressure conditions. If an error condition e x i s t s , then an audible as well as a v i s u a l alarm is turned on. While the audible alarm can be turned off with a keystroke as per the prompt, the vis u a l message remains on the screen u n t i l the condition is corrected or the end of the STAGE at which time program control would be transferred back to APC from which an alternate operating mode would be selected. . - If the current STEP requires a column to be eluted, then the UV monitor output is tested against the appropriate threshold value to determine the presence or absence of a protein peak. If a peak is present then the l i q u i d containing the protein fraction is routed to the appropriate storage container for either lysozyme or avidin. - The presence of an eluting peak w i l l override the STEP time as the c o n t r o l l i n g factor in the event that the STEP time has elapsed prior to the end of a peak to prevent the loss of product. If an extended STEP time is required, both audible and v i s i b l e alarms are invoked. - Keyboard input is tested to see i f \"M\" has been pressed, transferring control to MPC (Manual Process Control), or in the event of an extended STEP duration due to peak elution, i f any key has been depressed to stop the audible 64 alarm. - A test is made to see, when the current STAGE time i s complete, i f the elapsed feed time equals the cascade feed time or i f feed time was lost due to f i l t e r maintenance down time. - The STEP time loop is tested for completion. Upon completion, the time loop is exited and the STEP number loop increments by one and restarts the STEP time loop. When the STEP number loop is exited, the STAGE number loop is incremented by one. The completion of the STAGE increments the CYCLE loop, which when completed increments the BLOCK loop which i s e s s e n t i a l l y an i n f i n i t e loop ' unless an endpoint has been specified via MPC. - At any point within the process, the run is terminated when a spe c i f i e d end point has been reached or when overpressure conditions force a shutdown, either staged or immediate depending upon whether the source of the problem was the columns or the i n l e t feed f i l t e r s . The operation of modes SI and S2 for one and two columns shut down, respectively, i s very similar to that described above for the NM routine. As each fouled column is removed from operation during a staged shutdown, i t is backwashed with e q u i l i b r a t i n g buffer to remove the egg white from the resin bed. If three columns are being used (NM routine) and an immediate shutdown is forced due to the feed f i l t e r s plugging, then a l l three would be cleaned out in sequence prior to the system 65 shutting down. This gives added protection against the growth of microorganisms even though the lysozyme is an a n t i - b a c t e r i a l agent. Ending a run: When a run is completed or has been terminated prematurely, the APC routine passes control back to the MAIN program segment after clearing the screen and displaying an end-of-run message indicating b r i e f l y the cause of the termination. At thi s point, the next keystroke produces a small end-of-run menu which allows the operator to st a r t a new run, to save the current method to a disk f i l e , and/or to exit the program. 66 4. DISCUSSION AND FUTURE CONSIDERATIONS While no albumen has been run through the system thus far, the various operating modes, alarms and control transfer conditions have been tested using dyed water and found to perform adequately. As in any project of this kind, approaching a solution to the stated objectives often raises the l i d on Pandora's Box of program revisions and \"new and useful\" routines to add to the existing framework. The addition of more and more subroutines that were not considered wnen the program architecture was l a i d out usually leads to a patchwork q u i l t of code, i Such programs may f u l f i l l the immediate processing requirements, but lack the grace and s i m p l i c i t y of a mature program which has been c a r e f u l l y reduced to the kernel of usefullness. Parts of th i s program have been reworked several times while others have not been so c o l s e l y scrutinized. When working with a moderately long program such as t h i s , the constant threat to ever completing the i n i t i a l version is the wellspring of ideas for improvements to what has already been committed to code. The physical plant and control software designs together meet and exceed the basic design sp e c i f i c a t i o n s l i s t e d in the introduction of th i s paper. Not only does the software promote an orderly shutdown in the case of column fouling, but i t also allows for the removal of the offending column while continuing to run what is l e f t of the system. This approach attempts to maximize production with a minimum of supervision. 67 The delayed run/start feature provides a great deal of operating f l e x i b i l i t y , which is es p e c i a l l y desirable for use in reasearch. The MPC routine gives great f l e x i b i l i t y in allowing the operator to modify the values of most parameters which materially a f f e c t the course of events within the process. While a host of smaller operating enhancements have been included in the software, there are always new functions and refinements to be considered. One such addition that would be useful in tracking the behavior of the system over a long unattended period i s a logging routine that notes and time stamps every change of operating parameter during the run and makes is easy to dump to the printer as a run report. While the program provides adequate control in i t s current form, the marketing of such a product for commercial use would necessitate a complete rewrite, taking into account a l l of the enhancements which have been add-ons to t h i s version. The run-time display should be redesigned to provide only the most essential information, with the less important data being relegated to a secondary screen which could be call e d up with a single keystroke. The aesthetics and uniformity of presentation of the interactive and data display screens would also need to be improved. Other future considerations in terms of the enhancement of the process i t s e l f might include the addition of a secondary p u r i f i c a t i o n step for the proteins and an u l t r a f i l t r a t i o n (UF) module to concentrate the protein fractions without concentrating the accompanying s a l t s . The s a l t solutions recovered using UF 68 could be recycled as shown in Figure 4 . 1 . Two more pumps would be required to r e c i r c u l a t e the l i q u i d s , and the conductivity meter and pH meter shown would allow the operator to ensure that the proper levels of s a l t concentration and pH were maintained. A more sophisticated sytem again would have automatic control over concentration and pH by incorporating computer-controlled dispensing pumps for concentrated eluant and acid and base solutions. Since the physical plant and software have been developed expressly to accomodate the recovery of avidin and lysozyme using the elution-looping technique, the general market appeal for such a system i s l i k e l y to be quite limited. The physical plant i t s e l f , however, can be used with v i r t u a l l y any chromatographic procedure i f accompanied by the appropriately modified software. For example, with a very minor modification to the flow control hardware and a moderate restructuring of the software,the system could be performing Immobilized Metal A f f i n i t y Chromatography (IMAC) for the recovery of immunoglobulins. The use of automatic control in the f i e l d of food processing, and bio-processing in general, has the potential to greatly benefit both the producer and the consumer through the enhancement of production process e f f i c i e n c y , cost effectiveness and quality control. S T R O N G turn* D C I O N I Z E D WATER. \" KEY © TO WASTE 0 PERISTALTIC PUMP © F R A C T I O N COLLECTOR ^ HOMOCENIZER 0 F ILTER COND. METER r o n s i d e r a d f o r f u t u r e s t u d y K A T E B . BARREN FEED A V I D I N U F C O N C . FREEZE DRIER W A S T E 1 m/EE ..r~| M A N U A L E • _ 1 l I T O D R A I N FIGURE 4.1: General system layout including potential future additions for nearly total automation of control REFERENCES 70 Clark, J.W., W.Viessman, J r . , and M.J. Hammer. 1977. Water Supply and Pollution Control . 3 Ed. Harper & Row, Publishers. New York. Cole-Parmer Catalogue. 1987-1988. Cole-Parmer Instrument Company. 7425 North Oak Park Avenue. Chicago, I l l i n o i s 60648-3884 U.S.A.. Durance, T.D.. 1987. Isolation of Avidin and Lysozyme from Egg Albumen. Doctoral thesis. University of B r i t i s h Columbia. Ion Exchange Chromatography: p r i c i p l e s and methods. 1980. Pharmacia Fine Chemicals AB. Uppsala, Sweden. J o l l e s ,P., D. Charlemange, J.F. P e t i t , A.C. Maire and J. J o l l e s . 1965. Biochimie comparee des lysozymes. B u l l . Soc. Chim. B i o l . 47:2241. Li-Chan, E., S. Nakai, J. Sim, D.B. Bragg, and K.V. Lo.. 1986. Lysozyme Separation from Egg White by Cation Exchange Column Chromatography. J.of Food Science. Vol 51 No.4, 1032-1036. Salisbury, F.B. and CW. Ross. 1978 . Plant Physiology , 2 ed. Wadsworth Publishing Co., Inc.. Belmont, CA. 71 Wilkinson, B.R. and R.E. Dorrlngton. 1975. Lysozyme (Muraraldase) from Waste Egg White. Process Biochemistry. March, 1975, 24-25. 72 A P P E N D I X OPBRATIOSAL STATES FOB BORMAL AHD DISABLED SYSTEM MODES SSQUKBCB V1L¥B 1UMBEB PUMPS BOMOG-Bo. colaati DESCRIPTION BHllEB A B C 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 1 » 19 20 21 22 23 1 2 3 BOBHAL MODE (3-coluan c y c l e : 2-coluan cascade) a I 1 2 R b a c k - i i n s e C to barren 0 0 0 1 0 0 0 0 0 0 1 1 0 0 0 0 0 1 0 1 1 0 0 O i l t> I b a c k - r i n s e C to waste 0 0 0 1 0 0 0 1 0 0 0 1 0 0 0 0 0 1 0 1 1 0 0 0 1 1 c I apply weak s a l i n e to C 0 1 0 1 0 0 0 0 0 0 0 1 0 0 0 0 0 1 0 1 0 1 0 1 1 0 d I apply strong s a l i n e to C 1 1 0 1 0 0 0 0 0 0 0 1 0 0 0 0 0 1 0 1 0 1 1 1 1 0 « I e q u i l i b r a t e C 0 0 0 1 0 0 0 1 0 0 0 1 0 0 0 0 0 1 0 1 1 1 0 O i l M C i s i d l e 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 0 1 0 a I K 1 2 b a c k - r i n s e A to barren 0 0 0 0 1 0 0 0 1 0 0 0 1 0 1 0 0 0 1 1 0 0 0 O i l i> I b a c k - r i n s e A to waste 0 0 0 0 1 1 0 0 0 0 0 0 1 0 1 0 0 0 1 1 0 0 0 0 1 1 c I apply veak s a l i n e to A 0 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 0 0 1 0 1 1 0 1 1 0 d I apply strong s a l i n e t o A l O O O l O O O O O O O l O O O O O l O l l l 1 1 0 e I e q u i l i b r a t e A 0 0 0 0 1 1 0 0 0 0 0 0 1 0 1 0 0 0 1 1 0 1 0 0 1 1 1 I * i s i d l e 0 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 1 0 a | 2 R 1 b a c k - r i n s e B to barren 0 0 0 0 0 0 0 0 0 1 0 0 0 1 0 1 1 0 0 0 1 0 0 0 1 1 *> I b a c k - r i n s e B to waste 0 0 0 0 0 0 1 0 0 0 0 0 0 1 8 1 1 0 0 0 1 0 0 O i l c I apply weak s a l i n e to B 0 0 1 0 0 0 0 0 0 0 0 0 0 1 0 0 1 0 0 1 1 1 0 1 1 0 d I apply strong s a l i n e to B 1 0 1 0 0 0 0 0 0 0 0 0 0 1 0 0 1 0 0 1 1 1 1 1 1 0 e I e q o i l i b r a t e B 0 0 0 0 0 0 1 0 0 0 0 0 0 1 0 1 1 0 0 0 1 1 0 0 1 1 1 I B i s i d l e 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 1 0 0 0 1 0 0 0 1 0 SBUTOOW MODE: S l ( a ) (Cohan A i s down; 2 - c o l . i n t e r m i t t e n t cascade) a i « 1 2 cascade B-C 0 0 0 0 1 0 0 0 0 0 0 0 1 fl 0 0 0 0 1 0 0 0 0 0 1 0 b I * B 1 backwash B to barren 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 1 0 0 1 0 0 0 0 0 1 1 c I * B 1 backwash B to waste 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 1 0 0 1 0 0 0 0 0 1 1 d I * 8 1 apply weak s a l i n e to B 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 0 1 1 0 e I * B 1 apply strong s a l i n e to B 1 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 0 f I * B 1 e q u i l i b r a t e B 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 0 0 1 1 g I * 1 8 backwash C to barren 0 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 0 1 0 0 1 0 0 0 1 1 h I * 1 R backwash C to waste 0 0 0 0 1 0 0 1 0 0 0 0 0 0 0 0 0 1 0 0 1 0 0 0 1 1 i I * 1 8 apply weak s a l i n e to C 0 1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 1 0 1 1 0 j I * 1 R apply strong s a l i n e to C 1 1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 1 1 1 1 0 k I * 1 R e q u i l i b r a t e C 0 1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 1 0 0 1 1 OPBRATIOBAL STATES FOR BOBHAL ABD DISABLED S I S T E H MODES ( c o n t ' d ) SEQUEHCE B o . c o h a n DESCRIPTIOB A B C VALVB NUMBER 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 SHUTDOWB MODE: S l ( b ) ( C o h a n B i s down; 2 - c o l . i n t e r m i t t e n t c a s c a d e ) 1 c a s c a d e C - A 0 0 0 R backwash C t o b a r r e n 0 0 0 S backwash C t o was te 0 0 0 R a p p l y weak s a l i n e t o C 0 0 0 R a p p l y s t r o n g s a l i n e t o C 1 0 0 R e q u i l i b r a t e C 0 0 0 backwash A t o b a r r e n 0 0 0 0 backwash A t o was te 0 0 0 0 a p p l y weak s a l i n e t o A 0 1 0 0 0 0 0 0 0 1 0 0 0 0 0 0 1 0 1 0 0 0 1 0 0 0 0 1 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 1 0 0 0 1 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 1 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 1 0 0 1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 a p p l y s t r o n g s a l i n e t o A 1 1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 e q u i l i b r a t e A 0 1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 1 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 1 1 0 0 1 0 0 1 1 0 1 1 1 0 1 0 PUMPS 1 2 3 HOMOG-E N I Z E S SHUTDOBB MODE: S l ( c ) ( C o h a n C i s down; 2 - c o l . i n t e r a i t t e n t c a s c a d e ) c a s c a d e A - B backwash A t o b a r r e n backwash A t o was te a p p l y weak s a l i n e to A 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 a p p l y s t r o n g s a l i n e t o A 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 e q o i l i b r a t e A 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 backwash B t o b a r r e n 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 1 backwash B t o was te 0 0 0 1 0 0 1 0 0 0 0 0 0 0 0 1 a p p l y weak s a l i n e t o B 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 a p p l y s t r o n g s a l i n e t o B 1 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 e q u i l i b r a t e B 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 1 0 1 0 1 1 1 1 1 0 0 0 0 1 i 1 1 1 1 0PB8ATI0JAL STATES FOR HOBHAL ADD DISABLED SISTER MODES (cont'd) SEQUENCE Bo. cohin DKSCEIPTIOH A B C VALVB BUMBS8 1 2 3 4 5 6 7 8 9 10 11 12 13 11 15 16 17 18 19 20 21 22 23 S8UTD0W MODE: S2(a-b) (A i B down; C operating alone) 1 feed C backwash C to barren backwash C to waste apply weak saline to C 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 1 8 0 0 0 0 0 0 0 0 0 0 0 1 0 0 8 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 1 0 8 0 0 0 0 0 0 0 0 0 0 0 apply strong saline to C 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 equilibrate C 0 I 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 1 0 0 0 0 1 0 0 0 0 0 0 1 0 0 0 0 0 0 1 0 o o o o o o o o o o o o 0 0 SHUTDOfl MODE: S2(b-c) (B i C down; A operating alone) 8a I 1 8b I i 8c | 1 Id | R 8e | 8 8f | a 0 0 0 0 1 0 0 0 1 0 0 0 0 1 1 0 apply strong saline to A 1 0 0 0 0 0 0 0 0 0 0 I! 0 0 0 0 0 0 0 0 1 1 1 feed A backwash A to barren backwash A to waste apply weak saline to A 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 1 8 0 0 0 0. 0 0 0 0 1 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 8 0 0 0 0 0 0 0 0 0 0 0 0 0 0 equilibrate A 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 SHUTD0S1 NODE: S2(c-a) (C ( A down; B operating alone) feed B backwash B to barren backwash B to waste apply weak saline to B 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 0 0 apply strong saline to B 1 0 1 0 0 0 0 0 0 0 0 0 0 0 equilibrate B 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 1 0 0 o o o o o o o o o o o o o o o o o o o o 0 0 1 1 1 1 1 1 PUMPS 1 2 3 0 1 0 0 0 1 0 0 1 1 0 0 1 0 0 0 0 1 0 1 G 0 0 1 0 0 1 1 0 0 1 0 0 0 0 1 0 1 0 0 0 1 0 0 1 1 0 0 1 0 0 0 0 1 HOROG-BB11B8 TOTAL SBOTDOW NODE: S3 (all coluins down) 9a I 9b | 9c I 9d I 9e I 9f I 59 I 9h I 9i I 3j I backwash A to barien 0 0 0 0 backwash A to waste 0 0 0 0 equilibrate A 0 0 0 0 backwash B to barren 0 0 0 0 backwash B to waste 0 0 0 0 equilibrate B 0 0 1 0 backwash C to barren 0 0 0 0 backwash C to waste 0 0 0 0 equilibrate C 0 1 0 0 all power off 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 1 0 0 0 0 0 1 0 1 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 1 0 0 1 1 0 0 0 0 1 0 0 1 1 0 0 0' 0 1 0 0 1 1 0 0 0 0 0 tn "@en ; edm:hasType "Thesis/Dissertation"@en ; edm:isShownAt "10.14288/1.0058682"@en ; dcterms:language "eng"@en ; ns0:degreeDiscipline "Chemical and Biological Engineering"@en ; edm:provider "Vancouver : University of British Columbia Library"@en ; dcterms:publisher "University of British Columbia"@en ; dcterms:rights "For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use."@en ; ns0:scholarLevel "Graduate"@en ; dcterms:title "Computer automation of a novel ion-exchange process for the simultaneous recovery of lysozyme and avidin from chicken egg albumen"@en ; dcterms:type "Text"@en ; ns0:identifierURI "http://hdl.handle.net/2429/29660"@en .