Advances in Chemical Engineering Volume 19

Visit the link below to download the full version of this book:

https://medidownload.com/product/advances-in-chemical-engineering-volume-19/

Click Download Now

 ADVANCES IN
CHEMICAL
ENGINEERING
Editor-in-Chief
JAMES WE1
School of Engineering and Applied Science
Princeton University
Princeton, New Jersey

Editors
JOHN L. ANDERSON KENNETH B. BISCHOFF
Department of Chemical Engineering Department of Chemical Engineering
Carnegie Mellon University Uniuersity of Delaware
Pittsburgh, Pennsyluania Newark, Delaware

MORTON M. DENN JOHN H. SEINFELD

College of Chemistry Department of Chemical Engineering
Unitlersityof Califonia at Berkeley California Institute of Technology
Berkeley, California Pasadena, California

GEORGE STEPHANOPOULOS
Department of Chemical Engineering
Massachusetts Institute of Technology
Cambridge, Massachusetts

Volume 19

ACADEMIC PRESS
San Diego New York Boston
London Sydney Tokyo Toronto
This book is printed on acid-free paper. @
Copyright 0 1994 by ACADEMIC PRESS. INC.

All Rights Reserved.

No part of thib publication may be reproduced or transmitted in any form or by any
means. electronic or mechanical, including photocopy, recording. or any information
storage and retrieval system, without permission in writing from the publisher.

Academic Press. Inc.

A Division of Harcourt Brace & Company
52.5 B Street. Suite 1900. San Diego, California 92101-4495

U i i l r p d Kingdom Editior~p h / i s h r d h?
Academic Press Limited
24-28 Oval Road. London NW1 7DX

International Standard Serial Number: 0065-2377

International Standard Book Number: 0-12-008519-4

PRUllZD IN THE U N m STATESOF A M W C A

94 95 9 6 9 7 98 9 9 B B 9 8 7 6 5 4 3 2 1
 CONTENTS

CONTRIBUTORS ...................................... vii

PREFACE. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix

Polymer Systems for Controlled Release of Macromolecules.

Immobilized Enzyme Medical Bioreactors.
and Tissue Engineering
ROBERTLANGER
I. Controlled Release Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
I1. Immobilized Enzyme Bioreactors . . . . . . . . . . . . . . . . . . . . . . . . . . 24
111. Tissue Engineering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
IV. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48

Diffusion and Probability in Receptor Binding and Signaling

J . J . LINDERMAN.P . A. MAHAMA.K. E. FORSTEN.
AND D . A. LAUFFENBURGER
I . Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
I1. Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
111. Probabilistic Formulation of Receptor/Ligand Binding . . . . . . . . . . . . . 66
IV. Diffusion Effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
V. Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
VI . Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
Notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120

Transport Phenomena in Tumors

RAKESHK . JAIN
I. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130
I1. Principles of Cancer Detection and Treatment . . . . . . . . . . . . . . . . . . 132
111. Tumor Models and Their Growth Characteristics ................ 139
IV. Physiological and Transport Parameters . . . . . . . . . . . . . . . . . . . . . . 146
V . Mass Transfer in Tumors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163
VI . Heat Transfer in Tumors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179
VII . Future Perspective .................................. 191
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 194

V
vi CONTENTS

A Systems Approach to Multiphase Reactor Selection

R . KRISHNA
I . Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201
11 . Case Study: Recovery of Oil from Oil Shale . . . . . . . . . . . . . . . . . . . . 204
111 . General Selection Methodology for Multiphase Reactors . . . . . . . . . . .. 216
IV . Closing Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244
Notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 246
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241

Pollution Prevention: Engineering Design at

Macro.. Meso.. and Microscales
DAVIDT. ALLEN
I. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251
I1 . Macroscale Pollution Prevention . . . . . . . . . . . . . . . . . . . . . . . . . . 253
111. Mesoscale Pollution Prevention . . . . . . . . . . . . . . . . . . . . . . . . . . . 276
IV . Microscale Pollution Prevention . . . . . . . . . . . . . . . . . . . . . . . . . . 310
V. Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 318
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 319

Tropospheric Chemistry
JOHN H. SEINFELD. JEAN M . ANDINO.FRANKM . BOWMAN.
HALIJ . L . FORSTNER. PANDIS
AND SPYROS
I . Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 326
11. The Earth's Atmosphere . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 321
111. Atmospheric Physical Removal Processes . . . . . . . . . . . . . . . . . . . . . 328
1v. Agents of Chemical Attack in the Troposphere . . . . . . . . . . . . . . . . . . 331
V . Nitrogen Oxides Chemistry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 335
VI . Chemistry of the Background Troposphere:
The Methane Oxidation Cycle . . . . . . . . . . . . . . . . . . . . . . . . . . . 331
VII . Chemistry of the Urban and Regional Atmosphere . . . . . . . . . . . . . . . . 341
VIII . Atmospheric Reactions of Selected Nitrogen and Sulfur Compounds . . . . . 370
1x. Tropospheric Aerosols and Gas-to-Particle Conversion . . . . . . . . . . . . . 373
X . Aqueous-Phase Atmospheric Chemistry . . . . . . . . . . . . . . . . . . . . . . 376
XI . Atmospheric Chemical Mechanisms . . . . . . . . . . . . . . . . . . . . . . . . 394
XI1 . Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 396
Appendix: Supplementary References . . . . . . . . . . . . . . . . . . . . . . . 391
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 398

INDEX ............................................. 409

C'ONTEYTS IN TFiis SERIAL. . . . . . . . . . . . . . . . . . . . . . . . . .
OF VOLUMES 421
 CONTRIBUTORS

Numbers in parentheses indicate the pages on which the authors’ contributions begin.

DAVID T. ALLEN,Department of Chemical Engineering, University of

California at Los Angeles, Los Angeles, California 90024 (251)
JEAN M. ANDINO, Department of Chemical Engineering, California Institute
of Technology, Pasadena, California 91125 (325)
FRANKM. BOWMAN,Department of Chemical Engineering, California
Institute of Technology, Pasadena, California 91125 (325)
K. E. FORSTEN,Department of Chemical Engineering, Virginia Polytechnic
Institute and State University, Blacksburg, Virginia 24061 (5 1)
HALI J. L. FORSTNER,Department of Chemical Engineering, California
Institute of Technology, Pasadena, California 91125 (325)
RAKESHK. JAIN, Department of Radiation Oncology, Harvard Medical
School and Massachusetts General Hospital, Boston, Massachusetts
02114 (129)
R. KRISHNA,Department of Chemical Engineering, University of Amster-
dam, 1018 WVAmsterdam, The Netherlands (201)
ROBERTLANGER,Department of Chemical Engineering, Massachusetts
Institute of Technology, Cambridge, Massachusetts 02139 (1)
D. A. LAUFFENBURGER, Department of Chemical Engineering, University of
Illinois at Urbana-Champaign,Urbana, Illinois 61801 (51)
J. J. LINDERMAN, Department of Chemical Engineering, University of Michi-
gan, Ann Arbor, Michigan 48109 (51)
P. A. MAHAMA,Department of Chemical Engineering, University of Toledo,
Toledo, Ohio 43606 (51)
S w u o s PANDIS,Departments of Chemical Engineering and Engineering and
Public Policy, Carnegie Mellon University, Pittsburgh, Pennsylvania
15213 (325)
JOHN H. SEINFELD,Department of Chemical Engineering, California Insti-
tute of Technology, Pasadena, California 91125 (325)

vii
This Page Intentionally Left Blank
 Volume 19 of Advances in Chemical Engineering features a variety of
articles on chemical engineering, with a special theme on biomedical
engineering. Chemical engineers have worked to apply their science to
biomedicine since World War 11. In the past decade, their impact on the
practice of medicine has been unprecedented and useful. Langer writes
about pioneering work on using polymer systems for the controlled release
of macromolecules, for immobilized enzymes, for medical bioreactors, and
for tissue engineering. Linderman et al. address receptor binding and
signaling. Jain writes about transport phenomena in tumors, a topic that
has been recognized as the key to effective treatment with drugs. These
three chapters prove that chemical engineering in medicine has advanced
from an academic exercise to widespread clinical practice.
Krishna has worked for the Shell Oil Company for many years; his
chapter on the selection of multiphase reactors carries the knowledge of a
skilled practitioner, which complements his theoretical teachings as a
professor at the University of Amsterdam. Allen is one of the early
pioneers in the application of engineering design to pollution prevention,
and his chapter deals with macro-, meso-, and microscales. Seinfeld et al.
write on tropospheric chemistry, the arena where a great many of our air
pollution problems reside.
Together, these six chapters provide an expanding horizon for the
intellectual scope of chemical engineers, in topics from oil refining to
biomedicine, in scale from transport in tumors to the troposphere, and in
approach from scientific analysis to practical design selections.

James Wei

ix
This Page Intentionally Left Blank
 POLYMER SYSTEMS FOR CONTROLLED RELEASE
OF MACROMOLECULES, IMMOBILIZED
ENZYME MEDICAL BIOREACTORS,
AND TISSUE ENGINEERING
Robert Langer

Department of Chemical Engineering

Massachusetts Institute of Technology
Cambridge, Massachusetts 02139

I. Controlled Release Systems 2

A. Porous Delivery Systems for the Release of Proteins
and Macromolecules 2
B. New Biodegradable Polymers 11
C. Pulsatile Release Systems 17
11. Immobilized Enzyme Bioreactors 24
A. Immobilized Heparinase 24
B. Immobilized Bilirubin Oxidase 36
C. Removal of Low-Density Lipoprotein-Cholesterol 38
111. Tissue Engineering 40
A. Introduction 40
B. Materials and Matrix Structures
for Transplant Devices 41
IV. Conclusions 41
References 48

Recent advances in biology and medicine have created new challenges

and opportunities for engineers. With these advances occurring more and
more at a cellular and molecular level, the chemical engineer, in particu-
lar, has a unique training to address these challenges in a creative fashion.
Research in chemical aspects of biomedical engineering is growing rapidly.
Numerous biotechnology and bioengineering companies are being formed;
this creates and should continue to create an increasing demand for
scientists with interdisciplinary training in biology and engineering. There
are numerous efforts involving research in this area. Our research at
M.I.T. provides one of many such examples. In this chapter we discuss the
following areas of our research: controlled release systems; immobilized
1
Copyright 0 1994 by Academic Press, Inc.
ADVANCES IN CHEMICAL ENGINEERING, VOL. 19 All rights of reproduction in any form reserved.
2 ROBERT LANGER

enzyme medical bioreactors; and tissue engineering using degradable

polymers.

1. Controlled Release Systems

Over the past decade there has been increasing attention devoted to
the development of controlled release systems for drugs, pesticides, nutri-
ents, agricultural products, and fragrances. However, nearly all of the
systems that have been developed have not been capable of slowly releas-
ing drugs of large molecular weight (MW > 600). In fact, up until 1976 it
was a fairly common conception in the field of controlled release that
effective systems could not be developed for macromolecules (1). How-
ever, after several years of effort an approach was discovered that permit-
ted the continuous release of biologically active macromolecules as large
as 2,000,000 daltons from normally impermeable, yet biocompatible, poly-
mers for more than 100 days (2). Three areas of our drug delivery research
are reviewed here: systems that release large molecules through porous
polymer matrices, novel biodegradable polymeric delivery systems; and
pulsatile controlled release polymer systems.

A. POROUSDELIVERY
SYSTEMS
FOR THE RELEASEOF PROTEINS
AND MACROMOLECULES

Our interest in creating controlled release systems for polypeptides and

other macromolecules began in 1974 and stemmed from studies on the
growth of solid tumors. Most solid tumors require ingrowth of new blood
vessels from the host for further tumor development, and we were at-
tempting to isolate a drug that prevents the growth of new blood vessels.
This substance is derived from cartilage, a tissue that contains no blood
vessels. The bioassay used for this substance involved placing a tumor in
the cornea of a rabbit and monitoring the growth of new vessels toward
the tumor. It was desired to deliver the drug to the tumor to see if it
decreased the rate of blood vessel growth. The assay takes 30 days.
Purified fractions of the cartilage material were highly soluble, so that
they disappeared quickly after they were added. Therefore, a small sus-
tained-release system was needed to provide steady diffusion into the
tumor. Such a system had to be inert and noninflammatory. In early work
(3), polyacrylamide pellets had been tried for this purpose. The test
 POLYMER SYSTEMS FOR RELEASE OF MACROMOLECULES 3

protein was mixed with acrylamide before polymerization. After polymer-

ization, however, the small pellets were often highly inflammatory. The
inflammation could be reduced by extensive washing, but it could never be
completely eliminated. Furthermore, washing leached out most of the test
protein.
At that time, the only polymer systems reported for administering large
molecules were those described by Davis (41, polyacrylamide or
polyvinylpyrrolidone. However, these systems damaged the cornea and
permitted only brief periods of sustained release (2,5). Therefore, other
polymers and new ways of placing drugs in these polymers were examined.
However, one problem was that large molecules would only diffuse through
highly porous and permeable membranes (e.g., Millipore filters). In these
cases diffusion was too rapid to be of value. A new approach was
developed that permitted sustained release of large molecules from bio-
compatible polymers (2). The polymer was dissolved in an appropriate
solvent, and the macromolecule was added in powder form. The resulting
mixture can be cast in a mold and dried. When the pellets are placed in
water, they release the molecules trapped within the polymer matrix.
A number of polymer systems were tested for tissue biocompatibility
and release kinetics. The best long-term release results were obtained with
hydrophobic polymers. Examples included non-degradable ethylene-vinyl
acetate or biodegradable polylactic acid. Certain hydrogels such as polyhy-
droxyethylmethacrylate or polyvinylalcohol also worked effectively, but
released proteins for shorter time periods. With the hydrophobic poly-
mers, biologically active protein was released for more than 100 days
(2). In other tests, larger molecules (2 million MW), such as polysac-
charides and polynucleotides, were also successfully released for long time
periods (2).
While these initial studies demonstrated the feasibility of releasing
macromolecules from biocompatible polymers, the kinetics were often not
reproducible; controlled release was not achieved. The irreproducibility
results from drug settling and redistribution during casting and drying,
caused by the insolubility of the incorporated macromolecule powder in
the polymer solvent. At room temperature, the drug migrated vertically,
and visible lateral motion was caused by currents (possibly thermal) in the
mixture. A low-temperature casting and drying procedure was developed
to minimize this drug movement during matrix formation. When the
dissolved polymer-solid drug powder mixture was cast in a mold at
-8O"C, the entire matrix froze before any settling could occur. These
matrices were then dried at -20°C for 2 days until almost all the solvent
was gone. Final drying was conducted under vacuum at room temper-
ature (6).
4 ROBERT LANCER

1. Factors Aflecting Release Kinetics

With this reproducible method, factors that regulated release kinetics
could now be accurately assessed. Such factors were found to be drug
powder particle size and drug loading (drug : polymer ratio) (6). Coating
drug-containing polymeric matrices by dropping them into polymer solu-
tions of differing concentrations and drying them, also affected release
kinetics. By combining these simple fabrication parameters-drug particle
size, loading, and coating-release rates for any drug could be changed
several thousandfold (6).
To understand the release mechanism, cryomicrotomy was used to slice
10 pm-thick sections throughout the matrices. Viewed under an optical
microscope, polymer films cast without proteins appeared as nonporous
sheets. Matrices cast with proteins and sectioned prior to release displayed
areas of either polymer or protein. Matrices initially cast with proteins and
released to exhaustion (e.g., greater than 5 months) appeared as porous
films. Pores with diameters as large as 100 p m , the size of the protein
particles, were observed. The structures visualized were also confirmed by
Nomarski (differential interference contrast microscopy). It appeared that
although pure polymer films were impermeable to macromolecules (21,
molecules incorporated in the matrix dissolved once water penetrated the
matrix and were then able to diffuse to the surface through pores created
as the particles of molecules dissolved. Scanning electron microscopy
showed that the pores were interconnected (7).
Next investigated were changes in pore structure over time. Sections
were prepared from matrices in the process of release. It was observed
that (i) the pore structure changes minimally as a function of time; (ii)
after 16-40 h there is no evidence of a receding interface between
dissolved and dispersed drug; and (iii) none of the drug remains undis-
solved at 40 h (30% release). Observations (ii) and (iii) differ from those
reported for less soluble low molecular-weight drugs such as certain
steroids, and are probably due to the high solubilities of many proteins
such as bovine serum albumin (BSA) (solubility > 500 mg/mL). Thus, the
widely used moving zone models developed by Higuchi may not be
applicable to the situation of macromolecules because of observations (ii)
and (iii).
A number of assumptions were made and then verified to develop a
model: ti) The rate-limiting step for transport is drug diffusion through
pores (other steps such as water penetration into the matrix and drug
dissolution occur in less than 40 hours). (ii) The effect of concentration
dependence on the drug diffusion coefficient is not significant. This was
verified by an analysis of diffusion effects at the concentrations in the
 POLYMER SYSTEMS FOR RELEASE OF MACROMOLECULES 5

matrix. (iii) No drug diffusion occurs through the polymer backbone (2).
(iv) The pores are interconnected, the porosity is uniform, and pore size
changes minimally with time. (v) The initial drug distribution is uniform.
This was also verified by cryomicrotomy. (vi) No boundary layer effects
exist. This was verified by stirring, which would have disrupted boundary
layers had they been present. Release rates of matrices stirred in contain-
ers at 2000 rpm were identical to unstirred release rates, indicating the
lack of boundary layer effect. (vii) Infinite sink conditions exist. The
volume of the release medium is approximately 100 times the volume of
the polymer/protein matrix. Increasing the volume of the release medium
does not alter measured release kinetics. (viii) Minimal effects exist as a
result of osmosis due to solutes in the surrounding environment or charge
interaction of the drug with the polymer. Consonant with this assumption,
no effect on release rate was found to result from increasing the ionic
strength of the medium from 0 to 1 M NaCl.
For these assumptions, permitting release from only one side of the
slab, the boundary conditions are those of zero flux at the coated edges,
and C = 0 at the releasing face.
If diffusion through pores occurs, the Fick diffusion equation can be
solved:

where Mt is the cumulative drug mass released, M, is the drug mass

originally incorporated in the matrix, t is time (hours), and L is the
thickness of the slab (centimeters). In addition, D,, the effective diffusion
coefficient (cm2/hour) of the drug in the matrix is set equal to D,F,
where Do is the bulk diffusion coefficient of the same drug in the release
media that has filled the pores, and F is a factor accounting for the
geometric effects of the pore structure of the matrix (i.e., tortuosity,
dead-end pores, and constrictions between pores). F was determined via a
regression analysis for several cases of RSA released from polymer slabs at
several porosities (Fig. 1). A log-log plot of F versus porosity was well
fitted by the function

log,, F = 0.463 + 5.6410g,,~, (3)

where E is the porosity. Knowing this equation for F , it can then be
6 ROBERT LANGER

0.0 I

0.001 -
--W
LL

0.0001 -

0.00001
0.10 0.15 0.20 0.30 0.40
Volume Fraction of Drug (t 1
FIG. 1 . Log-log plot of factor F = D,/D, as a function of porosity for BSA matrices
[from Bawa et al. (7), with permission of Elsevier Science Publishers BV].

written
0, = Do(2 . 9 0 4 ~ ~ . ~ ) , (4)
and this value of 0, can be substituted into Eq. (2).
Both the slab thickness L and the porosity E were measured. For a
given macromolecule, the bulk diffusivity, Do, is measurable or obtainable
from the literature. Thus, a test of the model is to cast slabs using other
proteins, measure the parameters L , E , and Do, and see whether the
release kinetics follow Eq. (2). This has been done for P-lactoglobulin and
lysozyme (Fig. 2). The solid lines are predictions based on Eq. (2) which
show general agreement with the data (7).
An additional check of the model is to determine if it can predict not
only the time-dependent release of the drug, but also the time-dependent
position of the drug within the matrix. If Eq. (2) is valid, then the drug
distribution within the matrix can be described by
+ 1)TX
c(x,t) =
4c,,
-
T
c 2n( - l +) f l exp[
~

1
-(2n + I ) ’ ~ * D , c / ~ L * ] c (2n
~s
2L

where C,, is the initial concentration of drug in the matrix (mg/cm3

matrix), and C ( x , t ) is the concentration (the concentrations C and C,, are
expressed in terms of the volume of the whole matrix, including both pore
and polymer volumes) at time t and distance x (centimeters) into the
matrix from the exposed face. To test Eq. (5), cryomicrotomy was used to