University of California

0 downloads 0 Views 10MB Size Report
and sandcastle worm gluing16, 17 employ coacervation processes in combination ...... result of electrostatic screening or competition between Cl. - and citrate.
University of California Santa Barbara

Polyelectrolytes in the Synthesis of Bio-Inspired Composite Materials

A dissertation submitted in partial satisfaction of the requirements for the degree of Doctor of Philosophy in Chemistry by

Brandon J. McKenna

Committee in charge: Professor Galen D. Stucky, Chair Professor Steven K. Buratto Professor Alison Butler Professor J. Herbert Waite

September 2007

The dissertation of Brandon J. McKenna is approved.

____________________________________________ Steven K. Buratto

____________________________________________ Alison Butler

____________________________________________ J. Herbert Waite

____________________________________________ Galen D. Stucky, Committee Chair

September 2007

Polyelectrolytes in the Synthesis of Composite Materials

Copyright © 2007 by Brandon J. McKenna

iii

This dissertation is dedicated to my parents: Anne-Marie and John J. McKenna.

iv

ACKNOWLEDGEMENTS

I would like to thank my advisor Professor Galen D. Stucky for the opportunities he has given me while working in his laboratory. The scientific atmosphere supported by Galen promotes curiosity driven research

through

intellectual

freedom

and

has

fostered

my

independence and personal development. He has been an exemplary role model of integrity, and his sheer joy in the process of discovering and learning new science, across a wide range of subfields, has inspired us all and kept us grounded in our pursuits. I would also like to Professor J. Herbert Waite, who has effectively served as my unofficial co-advisor.

His expansive knowledge,

principled philosophies, precise and artistic prose, and passion for discovery have guided me to enjoy the purity in science in an age of increasing hype.

Thanks also to my entire advising committee,

including Prof. Alison Butler and Prof. Steve K. Buratto, who provided invaluable direction at a critical time in my studies. The entire Stucky lab, in each of its incarnations that I have experienced, has been critical to my growth as a Ph.D. student. Each member has provided individual insight, and the list of support is long. At large, the group has provided a positive and supportive atmosphere

v

of dedication, hard work, and of course diversified learning. Several members have particularly contributed to my experience. I would like to thank Prof. Henrik Birkedal for laying the groundwork for turning a physicist into a materials chemist, imparting a basis for conducting research and writing scientifically, and earnestly upholding the wonder of discovery. I thank Dr. Peter Stoimenov for guidance, meaningful conversations, and his appreciation for beauty, either in science or his photographs; he is truly the best. I thank Dr. Muhammet Toprak for insightful and extended discussions, for his friendship, and for our fruitful collaborations from which I learned important complementary research practices. Thanks to Dr. Todd A. Ostomel for leading the way and rocking in the lab late into the nights; philosophical discussions of resolving these practices with his initials were priceless. And thanks to my “Taiwanese research brother,” Frank (Chia-Kuang) Tsung. During our parallel pursuits of “real” science, I have appreciated his friendship and look forward to continuing camaraderie in Berkeley. I would also like to thank Dr. Celia Wrathall, who for all her humility has no power to delete her acknowledgement this time. Celia has provided critical support for my time here and beyond. Many many thanks to Dr. Burçin Temel for her motivation, direction, and support. Her companionship has made the last few years

vi

among my best. Thanks also to all the friends who have supplied needed comic relief throughout my years here. Finally, I would like to give thanks to my parents for their emotional support and life guidance.

Thanks to my mother for

fostering a childhood of wonder and for her patience with my inquisitive nature—whether it regarded trains, windmills, or pockets full of rocks! Thanks to my father for instilling the edicts of hard work, the beauty in simplicity, and adaptability.

This work was partially supported by the MRSEC program of the National Science Foundation under award No. DMR00-80034, and made use of MRL Central Facilities granted by NSF under Award No. DMR05-20415. This work was also supported in part by the U.S. Army Research Laboratory and the U.S. Army Research Office under contract number DAAD19-03-D-0004. This work is also supported in part by the NASA University Research, Engineering and Technology Institute on Bio Inspired Materials (BIMat) under award No. NCC-102037. This work was supported in part by the Public Health Service from NIH grant R01 DE 014572, NSF 0233728.

vii

VITA OF BRANDON J. MCKENNA September 2007 EDUCATION •

Doctor of Philosophy, Chemistry University of California Santa Barbara, CA, September 2007 (anticipated) Thesis title: “Polyelectrolytes in the Synthesis of Bio-Inspired Composite Materials”



Artium Baccalaureus, Chemistry & Physics Harvard University, Cambridge, MA June 2002

AWARDS & SCHOLARSHIPS •

Dow Materials Use Prize 8th Annual New Venture Competition Technology Management Program University of California, Santa Barbara, 2007



Certificate in Technology Management Graduate Program in Management Practice University of California, Santa Barbara, 2007



Chemistry and Biochemistry Department Fellowship University of California, Santa Barbara, 2002-2003, 2003-2004



MRL Distinguished Graduate Fellowship University of California, Santa Barbara, 2002-2003



Advanced Standing Award Harvard University, 1997-2001

RESEARCH EXPERIENCE University of California, Santa Barbara Graduate Researcher, Chemistry and Biochemistry Department, 2002-present with Prof. Galen Stucky

viii



Currently studying the ability of polyelectrolytes and nanoparticles to assemble ordered inorganic structures, in two distinct projects: 1) Developing a device for targeted and magnetic drug delivery. 2) Biomimetic mineralization with acidic macromolecules.

Harvard University, Cambridge, MA Undergraduate Researcher, Physics Department and DEAS, Jun. 2000-Apr. 2002 with Prof. Charles Marcus • Studied the lithographical and electrochemical fabrication of nano-electrodes, to produce quantized conductance, e.g. in single atoms. Harvard University, Cambridge, MA Undergraduate Researcher, Harvard-Smithsonian Center for Astrophysics, Apr.-Jun. 2000 with Dr. Jim Phillips • Arranged and tested a picometer-sensitive laser optics system, for testing Einstein’s Equivalence Principle. University of New Hampshire, Durham, NH Undergraduate Researcher, Constraint Computation Center, Jun.Aug. 1999 with Prof. Eugene Freuder • Modeled and programmed solutions to constraint satisfaction problems.

PUBLICATIONS & PRESENTATIONS 1. McKenna, B. J., Waite, J. H., Stucky, G. D. “Complex Coacervates as Intermediates in Non-classical Crystallization,” in preparation 2. McKenna, B. J., Waite, J. H., Stucky, G. D. “Biomimetic Control of Calcite Morphology with Homopolyanions,” submitted

ix

3. Toprak, M. S., McKenna, B. J., Waite, J. H., Stucky, G. D. “Control of Size and Permeability of Nanocomposite Microspheres,” Chem. Mat. 2007, 19, 4263-4269 4. Huang, X., Bronstein, L. M., Retrum, J., Dufort, C., Stein, B., Stucky, G., McKenna, B. J., Dragnea, B. “Self-Assembled Virus-like Particles with Magnetic Cores,” Nano Lett. 2007, 7, 2407-2416 5. McKenna, B. J., Waite, J. H., Stucky, G. D. “Morphological Control of CaCO3 with Anionic Homopolymers,” poster presented at 2006 Symposium on Recent Advances in Nanoscale Materials Research, and 2007 Materials Research Outreach Program, UCSB. 6. Toprak, M., McKenna, B. J., Waite, J. H., Stucky, G. D. “Tailoring Magnetic Microspheres with Controlled Porosity,” MRS Symp. Proc. 2007, 969, W03-11 7. Toprak, M., McKenna, B. J., Mikhaylova, M., Waite, J. H., Stucky, G. D. “Spontaneous Assembly of Magnetic Microspheres,” Adv. Mat. 2007, 19, 1362-1368 8. McKenna, B. J., Waite, J. H., Stucky, G. D. “Biomimetic Materials,” August 2006 talk at US Gypsum Corporation, invited by Creative Realities, Inc. as Techmax Thought Leader 9. Toprak, M., McKenna, B. J., Stoimenov, P., Waite, J. H., Stucky, G. D. “Controlled Assembly of Magnetic Microspheres,” poster presented at Spring 2006 MRS Meeting 10. McKenna, B. J., Waite, J. H., Stucky, G. D. “Complex Coacervate Mineralization,” oral talk at Fall 2005 MRS Meeting 11. McKenna, B. J., Birkedal, H., Bartl, M. H., Deming, T. J., Stucky, G. D. “Micrometer-sized Spherical Assemblies of Polypeptides and Small Molecules by Acid-Base Chemistry,” Angew. Chem. Int. Ed. 2004, 43, 5652-5622 12. McKenna, B. J., Birkedal,H., Bartl, M. H., Deming, T. J., Stucky,G. D., “Self-Assembling Microspheres from Charged Functional Polyelectrolytes and Multivalent Ions,” poster presentation at UC Systemwide Bioengineering Symposium 2004 13. McKenna, B. J., Birkedal, H., Bartl, M. H., Deming, T. J., Stucky,G. D., “Self-Assembling Microspheres from Charged

x

Functional Polyelectrolytes and Small-Molecule Counterions,” MRS Symp. Proc. 2004, 823, 23; also presented poster at Spring Meeting. PROFESSIONAL AFFILIATIONS • •

Materials Research Society student membership American Physical Society student membership

TEACHING EXPERIENCE University of California, Santa Barbara Chemistry and Biochemistry Department •

Teaching Lab. Assistant, Fall 2003, Winter 2004, and Spring 2004 Inorganic Synthesis Taught upper level laboratory techniques, including analysis, organometallic synthesis, and various advanced characterization techniques (HPLC, transient absorption, bulk electrolysis, etc.)



Teaching Lab. Assistant, Fall 2002, Winter 2003, and Spring 2003 General chemistry Taught sections, guided student experiments, prepared quizzes, graded laboratory reports.

SKILLS •

Materials synthesis (sol-gel, nanoparticles, organic); X-ray diffraction (XRD), surface area and pore size analysis (BET), mass spectrometry (MS), X-ray photoelectron spectroscopy (XPS), atomic force microscopy (AFM), UV-vis and IR spectroscopy, Raman spectroscopy, fluorimetry, thermogravimetric analysis and differential scanning calorimetry (TGA/DSC), scanning electron microscopy (SEM), transmission electron microscopy (TEM), nuclear magnetic resonance (NMR), zeta potential measurements, dynamic light scattering (DLS), light optical microscopy (LOM), confocal laser scanning microscopy (CLSM) including FRAP, circular dichroism (CD).

xi



Special Courses: Transmission Electron Microscopy– Principles and Practice



Programming Languages: C++, JAVA, LISP, Mathematica

xii

ABSTRACT

Polyelectrolytes in the Synthesis of Bio-Inspired Composite Materials

by

Brandon J. McKenna

This original research dissertation contains studies on complex coacervation, methods of modifying coacervates to create new materials as particularly applied to targeted drug delivery, and the use of coacervating polyanions for the assembly of intricate structures of calcium carbonate. Complex coacervation is a liquid-liquid phase separation that typically produces microspherical droplets from the combination of a variety

of

oppositely

charged

ions,

including

polymers

and

nanoparticles. The chemical space of coacervating components was found dependent on the number of charged groups and pH. Coacervates were shown to present chemically active surfaces that could be solidified by various methods, some of which also induced hollow interiors. The resulting assemblies were considered for targeted drug delivery by using superparamagnetic magnetite nanoparticles as

xiii

assembling components. Control over microsphere sizes was obtained from variation of several parameters, and porosities were examined as a function of cross-linking extent to determine encapsulation capabilities. Coacervates were further found to direct mineral growth, first in the form of shells, and then in the form of complicated structures that require substrate interaction via a solution-amorphous-crystalline mechanism.

A ternary phase diagram approach revealed a great

diversity of morphologies that could be modulated by the action of coacervating polyanions. Detailed analysis of one particular stacked lamellar structure suggested an assembly mechanism that may have relevance for biomineralization of nacre.

xiv

TABLE OF CONTENTS

Approval page………………………………………………………….ii Dedication……………………………………………………………..iv Vita…………………………………………………………………...viii Abstract………………………………………………………………xiii I. Complex Coacervation for the Synthesis of New Materials……...….1 II. Assembly and Silica Coating of Polypeptide/Small Molecule Coacervates…………………………………………………………...35 III. Superparamagnetic Nanoparticle Coacervates for Targed Drug Delivery…………………………………………………………….....61 IV. Calcium Carbonate Mineralization via Complex Coacervation...110 V. Morphological Ternary Diagram Studies of Non-classical Calcium Carbonate Mineralization with Homopolyanions…...………………134 VI. Towards a Model for Nacre Formation…..……………………...183

xv

Chapter I

Complex Coacervation for the Synthesis of New Materials

1

Abstract: It is presented in this chapter a generalized scheme for the assembly of microspherical devices from complex coacervates. First, a review of complex coacervation and related fields is presented, and several important properties displayed by most coacervates are highlighted. Polyelectrolyte association is generally enhanced by any parameter that increases the degree of attraction between oppositely charge moieties of the coacervating components.

Then I present

several methods for the stabilization of colloidal microspheres, including the formation of various inorganic shells, the use of organic crosslinking, and physical phase changes. It is anticipated that the extended variability of assembling components and stabilization methods will, in turn, extend the variety of potential device applications.

2

Complex coacervation, in its broadest definition, is a liquid phase separation that occurs between multiple solution species that electrostatically but dynamically attract one another and remain highly solvated, particularly around their binding moieties.1 Most commonly, there are two components, at least one of which is a charged polymer and the solvent is water; often, the other component is an oppositely charged polymer. The most commonly cited example is the mixing of gelatin with gum Arabic. Above gelation temperatures, and at pH values (4-5) that render gelatin with a net positive charge and gum Arabic with a net negative charge, the two polymers mutually attract and manifest in the appearance of micron-sized colloidal droplets that make the solution turbid, i.e. with a milky white appearance. Over time or with force, the droplets coalesce and eventually form a single, continuous, lower liquid layer that is dense with the biopolymers. This layer is called the “coacervate phase”, and the other, usually larger supernatant portion is called the “equilibrium phase”, as it is in dynamic equilibrium with the coacervate and containing a low concentration of the biopolymers. The droplets, prior to coalescing into the continuous coacervate phase, are often referred to as “coacervates” (Fig. 1.1A). The phenomenon is perhaps better understood in contrast to flocculation, the solid phase counterpart that usually takes the shape of fractal-like aggregates due to the rapid, diffusion-limited growth 3

mode.2-5

The main difference, of course, is that flocculations, or

“flocs”, contain less labile (more solid-like) bonds; i.e., they are less dynamic on the timescale of diffusion (Fig. 1.1B).

(A)

(B)

100 μm Figure 1.1. A) Light optical micrograph of a coacervate solution containing poly-L-lysine and citric acid. B) Micrograph of flocculation in a solution of poly-L-arginine and citric acid. Complex

coacervation

has

found

use

primarily

in

microencapsulation, where it is used for taste- masking,6-9 drug formulations,10-12 and carbonless copy paper.13 The phenomenon is also receiving newfound attention in the field of biomaterials, having been found more prevalent in nature than previously suspected. For instance, the DNA/histone interaction may be a coacervate-like interaction.14 Waite et al. have suggested that mussel fiber anchoring15 and sandcastle worm gluing16,

17

employ coacervation processes in

combination with high density phase inversion into foams and subsequent covalent and noncovalent crosslinking.

The exquisite

architecture of marine diatoms have been postulated to occur via liquid

4

phase separation processes,18-20 which suitable describes coacervation as it is known to occur for mimetic polymers.21 “Coacervation” has also been used to describe other, similar interactions aside from association of polyelectrolytes. For instance charged polymers and smaller anions can form coacervates, as with polyacrylic acid and Ca2+.

There are also examples of small

counterions forming coacervates, such as Cd(NO3)2 with sodium succinate and SrCl2 with ammonium molybdate. Such combinations tend to dehydrate and crystallize, losing their fluidity. Finally, “simple coacervation” contrasts with complex coacervation in that it occurs with

a

single

species.

AOT

(sodium

bis(ethylhexyl)octylsulfosuccinate) phase separation in solution is a common example of simple coacervation. Although it has been shown to depend on the concentration of Na+,22 sodium’s role as a noncovalent crosslinker is dubious. These descriptions accord with generally accepted definitions of “coacervation,” and especially with those first espoused by Bungenburg de Jong, who conducted some of the first in-depth studies that revealed some of the trends elaborated below.1 Nonetheless, “coacervation” has also been occasionally relegated to more specific instances; some biological texts describe it as a phase separation between a polysaccharide and protein—only a subset of components that may

5

form a coacervate. The confusion probably originates from the most widespread example of coacervation (gelatin/gum arabic), popularized further in the field of biology by the research activities of Oparin.23 Oparin’s experiments incorporated starch-polymerizing enzymes into these coacervates, which were then observed to grow and divide, leading him to conclude that complex coacervation may have supplied the most primitive “protocellular” compartments at the origin of life.24 The components were made more familiar still by patents for carbonless copy paper, for which such coacervates are used to encapsulate a pH-sensitive dye.13 Aside from occasionally receiving a limited definition, sometimes altogether different terms are used to describe coacervation. The most common example is “polyelectrolyte complex,” abbreviated as PEC or sometimes IPEC for “interpolyelectrolyte complex.” This term is less specific in describing the resultant phase, however, as it may also describe flocculation25 or even soluble dimers.26

Some

reports simply refer to “polyelectrolyte association,” either because the research was not focused on the nature of interaction, because such nature was not confirmed, or because the term has not completely penetrated various realms of science. Recently, phenomena related to coacervation have been given other monikers.

Block Ionomer

Complexes (BICs)27-34 occur between polyelecrolytes and oppositely

6

charged micelles, which have also previously been described straightforwardly.35-37 Although coulombic interactions contribute to their assembly, the importance of surfactant amphiphilicity legitimizes this term as a subset of coacervation. Recently, “Polymer Induced Liquid Precipitates” (PILPs) have been observed as precursors in mineralizing solutions; it remains to be shown that their emergence is caused in the same way as coacervation.38 However, the term PILP has also been applied to describe ordinary coacervate-like polyelectrolyte association.39

Finally, as will be described, our own research has

referred to coacervates as “acid-base microsphere assemblies.”20, 40 The term “coacervation” has been applied to systems seemingly dissimilar to those of the classical definition.

For instance, phase

separation has been observed between partially condensed silicic acid, such as polysilicic acid, and various small polar molecules,41 polyethers, and polyvinyl alcohol.42 These so-called coacervates differ in two significant ways: firstly, the components are not charged but rather interact purely via hydrogen bonding, and secondly, the ensuing aggregation and phase separation is due to hydrophobicity of carbon segments rather than electrostatics. In yet further departures from the classical definition, “coacervation” has been used to describe liquid precipitates of single macromolecules in two-component solvents, and phase separations in organic solvents, such as with ethylcellulose in

7

hexanes.43 In these cases, “coacervation” is taken very generally to mean “liquid phase separation of partially soluble components.”

Coacervate Properties All coacervates demonstrate similar dependencies on solution variables such as pH. It can generally be stated that polyelectrolyte association is enhanced by any parameter that increases the degree of attraction between oppositely charge moieties of the coacervating components. The following trends illustrate this point. 1) Concentration dependence roughly follows a solubility product. Being soluble, coacervates dissolve upon dilution, and the component concentrations limiting coacervation roughly fall along a curve.

These curves also describe

concentrations in the equilibrium phase.

Approximate

solubility curves determined by dilution and optical microscopy appear in figure 1.2.

8

Coacervate Visibility, by Dilution 0.009 0.008

0.006 0.005

-

PLD [COO ] (mM)

0.007

0.004 0.003 0.002

11kDa, pH7 11kDa, pH 10 33 kDa, pH 7 33 kDa, pH 10

0.001 0.000 0.005

0.010

0.015

0.020

0.025

0.030

0.035

0.040

0.045

2+

[Ca ] (mM)

Figure 1.2. Solubility curves for PLD/Ca2+ coacervates, determined by incremental dilutions and visibility under an optical microscope.

2) Salt disrupts coulombic attraction and inhibits coacervation. The dynamic bonds are disrupted with significant effect on the effective solubility products. This effect appears to be due to a combination of ionic strength and a counterion substitution effect. PLK/citrate coacervates are eradicated by

~10mM

chloride

salts

of

Na,

K,

and

tetramethylammonium, for ~0.5 wt% plK and citrate. Higher valency spectator salts are more effective than their monovalent

counterparts

at

inhibiting

coacervation,

requiring only 6 mM CaCl2 or less than 1mM MgCl2. 3) There is an optimal ratio of components. At the point of charge balance, coacervates generally repel less and grow

9

larger. The amount of material in the equilibrium phase is minimized. Due to the aforementioned salt and substitution effects, solubility curves appear to ‘bend back’ and reflect working component ratios. Outside of these ratios, both ionic strength and counterion substitution overcome the drive for coacervation. 4) There is an optimal pH range. The charges of amines and carboxylates, for instance, are controlled by their pKa values, so that a pH range usually exists that maximizes the overall number of complementary charges. 5) The optimal component ratio changes predictably with pH. Higher pH values cause deprotonation and create higher net negative charge, and hence require more polycation component to achieve maximal coacervation. 6) Less polar solvents promote polyelectrolyte association by dehydration and by decreasing the solution dielectric constant. Adding small amounts of alcohol to PLK/citrate coacervates causes significant swelling of coacervate spheres, and eventual aggregation into shapeless gels [Fig. 1.3].

10

100 μm

Figure 1.3. Micrograph depicting the effect of ethanol added to a solution of PLK/citrate coacervates. The coacervates initially swell and slowly aggregate into a solid mass. 7) The degree of coacervation correlates with coacervate phase volume, charge neutrality, bulk viscosity, and colloidal turbidity.

Maximal

polymer-polymer

noncovalent

crosslinking, as determined by component concentrations, pH, etc., in turn maximizes coacervate phase volume, which may be quantified by centrifugation.

This usually also

results in maximal solution turbidity if all other variables are constant (mixing order, etc.).

Charge neutrality reflects

balanced Coulombic interactions and maximal crosslinking, and can be determined by zeta potential measurements. Maximal attraction also reduces the number of free polymer

11

in equilibrium mixtures, reducing viscosity of the bulk colloidal solution. 8) Similarly sized components coacervate more effectively. The degree of coacervation is lessened when component molecular weights mismatch. This effect has been observed with more complicated block copolymer systems, in which phase separation is completely inhibited between mixtures containing charge block segments of different sizes.44

The previous section described established coacervation properties. The following are general observations I have noted from engineering several coacervate systems. 1) The effect of temperature is variable.

Increased

temperatures can dissolve some coacervates, as is the case with PLK/cit, at 70 ºC. PLR/cit undergoes a solid-to-liquid transition at 40 ºC; coacervate microspheres can be seen to ‘bud off’ of flocs with temperature increase. In contrast, PLE/Ca2+ solutions are transparent at room temperature, but become turbid with heating. The PLE/Ca2+ structures remain stable and solidified at room temperature, often in the form of dumbbells or other multiplets (Fig. 1.4). 2)

12

Figure 1.4. PLE/Ca2+ multiplet assemblies following heating and cooling steps.

The phase separation is attributed to desolvation of Ca2+ with concomitant change in tertiary structure of PLE from random coil to beta sheet [Fig. 1.5].

0.10

Circular Dichroism, PLD/Ca2+

0.05 0.00

Ellipticity

-0.05 -0.10 -0.15

21C 30C 40C 50C 60C 70C 75C

-0.20 -0.25 -0.30 -0.35 190

200

210

220

230

240

250

Wavelength (nm)

Figure 1.5. Circular Dichroism spectra of a PLD/Ca2+ solution at various temperatures.

13

The desolvation mechanism is supported by observations of PLD/Mg2+ systems, which are soluble at room temperature but can be observed to form coacervates at 80-90 ºC [Fig. 1.6].

(A)

(B)

100 μm

100 μm

Figure 1.6. LOMs of PLD/Mg2+ solutions at (A) 80 ºC and (B) 90 ºC.

3) In polymer/small-counterion mixtures, the larger polymer dictates the net charge (electrophoretic mobility) and coacervation is optimized with the small counterion in excess. In PLD/Ca2+ coacervates, for example, the CaCl2 component may be required in excess of aspartate monomers before coacervation is appreciable.

The

electrophoretic mobility of these coacervates will always correspond to a net negative charge, reflecting the dynamic association of calciums at the surface. 4) The order of component addition affects size distribution. For a given set of final concentrations, coacervates are

14

largest when two concentrated solutions are mixed and then diluted. If one solution is already dilute, coacervates are larger when the smaller component is added to the larger (diluted) one. Therefore, they are smallest when a small volume of the larger polymer is added to a diluted solution of the smaller component. A variety of components can be used to effect coacervation, including: natural8,

45-50

and synthetic,51 organic polyelectrolytes,

inorganic polyelectrolytes,24 block copolymers,27,

28, 52

dendrimers,53

small multivalent organic ions,20, 40 inorganic ions,54 and nanoparticles and quantum dots that are appropriately capped.55-57

However, there

is currently no reliable method for predicting when coacervation will occur rather than flocculation or dissolution.

Towards this effort,

presented in the following chapter, we have found predictive trends with polymer/small counterion combinations of carboxylate/aminecontaining compounds.20 The pH requirements of these systems were demonstrated to be highly dependent on component pKa. Furthermore, polyamine coacervation requires at least tri-carboxylated counterions, and polycarboxylate coacervation usually requires counterions with at least 5 amine groups. However, exceptions exist, and in general it is necessary to test specific systems individually.

15

Results of work done in our laboratory suggested a general mechanism of coacervating components as spherical templates, followed by a process of solidification/stabilization. The coacervates alone remain liquid out of solution, and in solution eventually wet the container as spread-out droplets (Fig. 1.7). However, the spherical shape could be preserved by condensation of inorganic silica, or with temperature changes. In order to extend the variety of potential future applications for various environments, further work was conducted to engineer other kinds of stabilization processes, including inorganic shell

formation,

organic

covalent

crosslinking,

and

physical

crosslinking.

(A)

(B)

100 μm Figure 1.7. Examples of normal coacervate sedimentation and wetting on glass slides, if not stabilized. Materials and Methods. SiO2 Shells. Standard silica treatments have been described elsewhere. For mesoporous silica, a polymer stock solution 1g of P123 (BASF, Pluronic) was dissolved in 37.5 mL pH 3 DI H2O. Sol-gel batches were typically prepared by mixing 10 μL tetramethoxysiloxane

16

(TMOS) in 258 μL of the P123 solution, and hydrolyzing for 20 min. Aqueous coacervate solutions were prepared by mixing 10 μL polyallylamine hydrochloride (PAH) (Sigma-Aldrich) at a monomeric concentration of 25 mM with 10 μL of a 10mM solution of trisodium citrate solution, and diluting to 100 μL at pH7. A 10 μL aliquot of the hydrolyzed TMOS/P123 solution was mixed with the full coacervate solution and allowed to react for 3h. Au shells: In a typical synthesis, poly-L-lysine (70kDa) and trisodium citrate solutions, each of ~0.5 wt.%, were mixed in a 1:9 ratio to form coacervate templates.

A gold salt solution was prepared

separately by dissolving 5 0mg HAuCl4 into 3 mL H2O.

Ten

microliters of this gold solution was pipetted into 100 μL of coacervates. 0.1 M sodium citrate (20 μL) was added after 10 min. to continue gold reduction on coacervate surfaces. Additional 10 μl gold acid was added, followed by addion of another 20 μl portion of citrate. The solutions were centrifuged after 1h, the supernatant was removed, and the gold spheres were redispersed in DI water. Mineral shells: Calcium carbonate shells were typically prepared by adding 0.5 mL of 100 mM by monomer PAA (15kDa) to 5.25 mL CaCl2, and then diluting with 40mL of DI H2O and NaOH solution, to a pH of 8.5. A fresh solution of 75mM Na2CO3 (75mM) was prepared, and three portions of 750 μL of this were successively

17

mixed with the coacervate solution in 5 min. intervals. The colloids were stable and allowed to sediment for 24-48 h, after which the solutions were decanted and the spheres were redispersed and stabilized in pH 10 NaOH to 50mL. Calcium phosphate shells were prepared from solutions of 5 mg/mL poly-L-aspartic acid (PLD) (33kDa) , 180mM CaCl2, and 10mM KH2PO4. Coacervates were formed from 20 μL PLD and 7.5 μL CaCl2, to which 10 μL phosphate solution was added. Products were collected and washed within 1 h. Covalent

cross-linking:

EDC

(1-ethyl-3-(3-

dimethylaminopropyl) carbodiimide hydrochloride) crosslinking was performed on coacervates of poly-L-glutamic acid (PLE) and pentalysine.

EDC solutions (50 mg/mL) were prepared fresh, and

added in aliquots of 20 μL to 100 μL coacervate solutions. Crosslinked spheres were observed without further purification. Glutaraldehyde crosslinking was performed on poly-L-lysine (PLK) and citrate coacervates, as described elsewhere (See Chapter 3). Typically, 20 μL of 2 mg/mL PLK solution was mixed vigorously with 120 μL TSC (1.6 mg/ml) for 15 seconds, using a vortex-mixer. To this, 120 μL of 2.5 wt% glutaraldehyde was added, and excess aldehydes were quenched by the addition of glycine after reacting for between 1 and 5000 min. Samples were purified by successive centrifugations.

18

Flocculation stabilization: Cationic coacervates were prepared from 25 μL FITC-labeled PLK at a monomeric concentration of 2.5 mM and 50 μL TSC at a concentration of 5mM. To this, 10 μL PLE (0.5 wt.%) was added and quickly mixed. Anionic coacervates were prepared from PAA and CaCl2 solutions to final concentrations of 10mM (monomer) and 10.5 mM, respectively, to a total volume of 100 uL. To this, 10 μL diaminoethylene (50 mM) was added. Physical stabilization: Mixtures either of poly-L-Arginine (PLR) or poly-L-Histidine (PLH) were made with TSC, such that both components comprised ~0.5wt% of solution. PLR/TSC solutions were heated to 60ºC, either in an oven, water bath, or microscope heat stage. PLH solutions were adjusted to pH ~5 with diluted HCl, and pH was increased with aqueous NaOH. PLD/Mg coacervates were made from final concentrations of 13mM PLD (33kDa) and 125 mM MgCl2.

Results SiO2: As described in the following chapter, prehydrolyzed TMOS induced a silica layer on the surface of primary aminecontaining polymers.20, 40 Silica nanoparticles could form shells around the coacervates and induce hollow cores in the coacervate interior; the hollow core effect was more evident with large coacervates (3-5 um).20, 40, 55, 56, 58

Polymer surfactants were found to introduce wormhole

19

mesopores using a surfactant/sol-gel strategy (Fig. 1.8); more work is required to determine if pore ordering is possible with these systems. This is an attractive route for manufacturing aqueous reaction vesicles that incorporate enzymes or other components too instable for porous vesicles that require thermal treatment.

100 nm Figure 1.8. (A) LOM of PAH/citrate coacervates stabilized by shells of mesoporous silica at pH 7. (B) TEM image of wormhole structured shells. Au and other metals: using the reduction properties of citrate in solution, Au salt precursor could be

deposited

onto

surfaces (Fig. 1.9).

coacervate The scheme

works with Ag as well, but is not generalized to work with Cu, for instance,

which

deposits

in

metallic rings on microsphere

Figure 1.9. SEM image of PLK/citrate coacervates coated with Au shell and subject to focused ion beam. Thanks to Dr. Peter Stoimenov for permission to use this micrograph

20

surfaces. Salts of divalent metals: Using CaCO3 and Ca/PO4 as model systems, coacervates of Ca2+ with PAA or PLD were used as microspherical templates and Ca2+ sources for the deposition of low Ksp salts. The phosphate spheres (Fig. 1.10) formed more rapidly and could produce well defined spheres with little side product.

(A)

(B)

5 μm

25 μm

Figure 1.10. (A) LOM and (B) SEM of PLD/Ca2+ coacervates stabilized by shells of phosphate mineral. EDC cross-linking: Taking advantage of the innate proximity of carboxylate and primary amine groups, EDC was used to form amide bonds in certain coacervates (Fig. 1.11). The primary obstacle for this procedure is that EDC works optimally at pH values in the range of 4.55, whereas the coacervates are most stable at higher pH. HOBt can be used to improve the amidation yield.

21

50 μm Figure 1.11. LOM of PLE/K5 coacervates stabilized by EDC crosslinking. Glutaraldehyde cross-linking: Glutaraldehyde is commonly used in the fixation of biological samples by the cross-linking of amine groups into Schiff bases. The advantages of this technique over EDC cross-linking are the range of permissible pH values and the stability of glutaraldehyde in solution (EDC decomposes slowly in water). However, glutaraldehyde cross-links are unstable for extended periods and free glutaraldehyde is unsafe for certain biological applications. Furthermore, while the Schiff base bonds can be made irreversible by NaBH4 reduction, the resulting bonds are not easily degradable. Flocculating counter-polyelectrolytes: Some coacervates could be solidified by the addition of oppositely-charged polyelectrolyte. In order to induce stable microspheres, the polymer must have a ‘solidlike’ or flocculation-inducing interaction with one of the components of the original coacervates, as opposed a

22

‘liquid-like,’ coacervation-

inducing interaction. This distinction is important for studies of LbL assembly, for which polyelectrolyte interactions should be essentially irreversible. For instance, at room temperature and neutral pH, PLD forms a coacervate with PLK, whereas PLE forms flocs with PLK. Therefore, coacervates of PLK/citrate could be stabilized by the addition of PLE. These assemblies are also able to induce hollow cores in the coacervates, as was done with silica NPs (Fig. 1.12).

(A)

(B)

50 μm

5 μm

Figure 1.12. Coacervates of FITC-PLK/citrate stabilized by PLE. (A) LOM, demonstrating structural integrity, and (B) demonstrating partial component redistribution. The flocculation interaction can also be used to stabilize coacervates by using smaller counterions. For instance, oxalate added to PLK/citrate spheres can induce solidified plK/oxalate shells, and diaminoethylene can cap PAA/Ca2+ coacervates. pH change: Some coacervates undergo a liquid-solid phase transition upon changes in pH. For instance, PLH/citrate coacervates are liquid between pH [4-5.5], but above pH 6, the imidazole moiety is

23

neutralized and the spheres solidify and aggregate. Similarly, chitosan is insoluble above pH 6, and will solidify at higher pH values in microspherical combinations with gum Arabic. Temperature change: As mentioned above, coacervate phase transitions with temperature are highly dependent on the exact system being used.

Liquefied PLR/citrate mixtures reversibly solidify at

room temperature (See following chapter).

Such behavior is also

known for gelatin/gum arabic coacervates, but in that instance the solidification is due to the tertiary structure of the gelatin component rather than changes in the stability of carboxyl/amine electrostatics. This procedure constitutes a facile microencapsulation method for applications where the physical interactions are stable. From these various interactions and coacervate sources, a generalized scheme is presented in Table 1.1. Several different types of polyanion and polycation sources can be used to form coacervates, although it is not currently possible to predict the nature of interaction between any given two: soluble, liquid, or solid.

And liquid

coacervates can be stabilized by various mechanisms, which depend in part on the nature of the selected polyelectrolytes.

24

Polyanions Polypeptides (polyaspartate) & polysaccharides (gum arabic)

Polycations

Stabilization

Polypeptides (polylysine, Inorganics (silica, gold, gelatin) calcium carbonate)

Synthetic polymers Synthetic polymers (polyacrylate) (polyallylamine) Small organics (citrate, EDTA) Small organics (amino alkanes) Inorganics (dichromate) Metal ions (Ca2+, Fe2+, Zn2+) Citrate-capped Nanoparticles Amine-terminated (Au, CdSe, Fe3O4…) Dendrimers

Silica nanoparticles (also creates hollow core) Solidification (cooling, gelation) Counter-polyelectrolyte (polyglutamate) Covalent crosslinking (glutaraldehyde, EDC)

Table 1.1. Generalized scheme outlining the variety of nano-sized coacervating components and various methods for their stabilization.

Conclusion Complex coacervation is a liquid-liquid phase separation that occurs, somewhat commonly, between oppositely charged species in solution. For polymer-based coacervates, some trends exist to describe the degree of polyelectrolyte association as it depends on pH, component concentrations, and added salts. Complex coacervation is a long-known phenomenon that nevertheless has remained obscure to most sciences, has continued to be rediscovered in various instances and with different names, and yet has been little-developed in the advancement of materials science. There remains much progress to be made to control coacervation, including an understanding of kinetics, a model to predict the nature of phase separation, a method for reducing their typical large size dispersion, a better understanding of encapsulation capabilities of both hydrophobic and hydrophilic agents,

25

and an expanded chemical toolbox for tailoring coacervate microsphere functionalities, such as porosity or environmental robustness. We have sought to expand the types of chemistry and the range of nanocomponents that can be integrated with the process, in order to produce organized core-shell devices that may find use as: drug delivery vehicles and artificial cells, industrial fillers, chemical microreactors, absorbents of toxins for environmental remediation, or colloidal sensors.

REFERENCES

1.

Bungenberg de Jong, H.G., in Crystallisation- coacervationflocculation in colloid science, H.R. Ed. Kruyt, Editor. 1949, Elsevier: Amsterdam. p. 232-258.

2.

Biggs, S., et al., Aggregate structures formed via a bridging flocculation mechanism. Chemical Engineering Journal, 2000. 80(1-3): p. 13-22.

3.

Larsson, A., C. Walldal, and S. Wall, Flocculation of cationic polymers and nanosized particles. Colloids and Surfaces aPhysicochemical and Engineering Aspects, 1999. 159(1): p. 6576.

26

4.

Thomas, D.N., S.J. Judd, and N. Fawcett, Flocculation modelling: A review. Water Research, 1999. 33(7): p. 15791592.

5.

Berlin, A.A., I.M. Solomentseva, and V.N. Kislenko, Suspension flocculation by polyelectrolytes: Experimental verification of a developed mathematical model. Journal of Colloid and Interface Science, 1997. 191(2): p. 273-276.

6.

Gouin, S., Microencapsulation: industrial appraisal of existing technologies and trends. Trends in Food Science & Technology, 2004. 15(7-8): p. 330-347.

7.

Gibbs, B.F., et al., Encapsulation in the food industry: a review. International Journal of Food Sciences and Nutrition, 1999. 50(3): p. 213-224.

8.

Schmitt, C., et al., Structure and technofunctional properties of protein-polysaccharide complexes: A review. Critical Reviews in Food Science and Nutrition, 1998. 38(8): p. 689-753.

9.

Shahidi, F. and X.Q. Han, Encapsulation of Food Ingredients. Critical Reviews in Food Science and Nutrition, 1993. 33(6): p. 501-547.

10.

Barry, B.W., Novel mechanisms and devices to enable successful transdermal drug delivery. European Journal of Pharmaceutical Sciences, 2001. 14(2): p. 101-114.

27

11.

Lim, S.T., et al., Preparation and evaluation of the in vitro drug release properties and mucoadhesion of novel microspheres of hyaluronic acid and chitosan. Journal of Controlled Release, 2000. 66(2-3): p. 281-292.

12.

Madan, P.L., Microencapsulation .1. Phase Separation or Coacervation. Drug Development and Industrial Pharmacy, 1978. 4(1): p. 95-116.

13.

Green, B.K., and Schleicher, Lowell, Oil-containing Microscopic Capsules and Method of Making Them, U.S.P. Office, Editor. 1957: U.S.A.

14.

Takahagi, M. and K. Tatsumi, Aggregative organization enhances the DNA end-joining process that is mediated by DNA-dependent protein kinase. Febs Journal, 2006. 273(13): p. 3063-3075.

15.

Waite, J.H., et al., Mussel adhesion: Finding the tricks worth mimicking. Journal of Adhesion, 2005. 81(3-4): p. 297-317.

16.

Stewart, R.J., et al., The tube cement of Phragmatopoma californica: a solid foam. Journal of Experimental Biology, 2004. 207(26): p. 4727-4734.

17.

Zhao, H., et al., Cement proteins of the tube-building polychaete Phragmatopoma californica. Journal of Biological Chemistry, 2005. 280(52): p. 42938-42944.

28

18.

Sumper, M., A phase separation model for the nanopatterning of diatom biosilica. Science, 2002. 295(5564): p. 2430-2433.

19.

Sumper, M. and E. Brunner, Learning from diatoms: Nature's tools for the production of nanostructured silica. Advanced Functional Materials, 2006. 16(1): p. 17-26.

20.

McKenna, B.J., et al., Micrometer-sized spherical assemblies of polypeptides and small molecules by acid-base chemistry. Angewandte Chemie-International Edition, 2004. 43(42): p. 5652-5655.

21.

Brunner, E., K. Lutz, and M. Sumper, Biomimetic synthesis of silica nanospheres depends on the aggregation and phase separation of polyamines in aqueous solution. Physical Chemistry Chemical Physics, 2004. 6(4): p. 854-857.

22.

Menger, F.M. and B.M. Sykes, Anatomy of a coacervate. Langmuir, 1998. 14(15): p. 4131-4137.

23.

Oparin, A.I., et al., Synthesis and Dissociation of Starch in Coacervate Drops. Doklady Akademii Nauk Sssr, 1962. 143(4): p. 980-&.

24.

Oparin, A.I., et al., Coacervate Drops with Participation of Polyphosphates and Model of Probiont Division. Doklady Akademii Nauk Sssr, 1977. 232(2): p. 485-488.

29

25.

Dautzenberg, H., et al., Stoichiometry and structure of polyelectrolyte complex particles in diluted solutions. Berichte Der Bunsen-Gesellschaft-Physical Chemistry Chemical Physics, 1996. 100(6): p. 1024-1032.

26.

Li, Y.J., et al., Complex-Formation between Polyelectrolyte and Oppositely Charged Mixed Micelles - Soluble Complexes Vs Coacervation. Langmuir, 1995. 11(7): p. 2486-2492.

27.

Lysenko, E.A., et al., Block ionomer complexes with polystyrene core-forming block in selective solvents of various polarities. 2. Solution behavior and self-assembly in nonpolar solvents. Macromolecules, 2002. 35(16): p. 6344-6350.

28.

Lysenko, E.A., et al., Block ionomer complexes from polystyrene-block-polyacrylate anions and N-cetylpyridinium cations. Macromolecules, 1998. 31(14): p. 4511-4515.

29.

Bronich, T.K., et al., Effects of block length and structure of surfactant on self-assembly and solution behavior of block ionomer complexes. Langmuir, 2000. 16(2): p. 481-489.

30.

Kabanov, A.V. and V.A. Kabanov, Interpolyelectrolyte and block ionomer complexes for gene delivery: Physicochemical aspects. Advanced Drug Delivery Reviews, 1998. 30(1-3): p. 49-60.

30

31.

Lysenko, E.A., et al., Block ionomer complexes with polystyrene core-forming block in selective solvents of various polarities. 1. Solution behavior and self-assembly in aqueous media. Macromolecules, 2002. 35(16): p. 6351-6361.

32.

Oh, K.T., et al., Block ionomer complexes as prospective nanocontainers for drug delivery. Journal of Controlled Release, 2006. 115(1): p. 9-17.

33.

Solomatin, S.V., et al., Environmentally responsive nanoparticles from block ionomer complexes: Effects of pH and ionic strength. Langmuir, 2003. 19(19): p. 8069-8076.

34.

Solomatin, S.V., et al., Colloidal stability of aqueous dispersions of block ionomer complexes: Effects of temperature and salt. Langmuir, 2004. 20(6): p. 2066-2068.

35.

Wang, Y.L., et al., Polyelectrolyte-micelle coacervation: Effects of micelle surface charge density, polymer molecular weight, and polymer/surfactant ratio. Macromolecules, 2000. 33(9): p. 3324-3331.

36.

Wang, Y.L., et al., Effects of salt on polyelectrolyte-micelle coacervation. Macromolecules, 1999. 32(21): p. 7128-7134.

37.

Zhang, G.Z., et al., Interpolymer complexes comprising block copolymers due to specific interactions. Materials Science &

31

Engineering C-Biomimetic and Supramolecular Systems, 1999. 10(1-2): p. 155-158. 38.

Gower, L.B. and D.J. Odom, Deposition of calcium carbonate films by a polymer-induced liquid-precursor (PILP) process. Journal of Crystal Growth, 2000. 210(4): p. 719-734.

39.

Wohlrab, S., H. Colfen, and M. Antonietti, Crystalline, porous microspheres made from amino acids by using polymer-induced liquid precursor phases. Angewandte Chemie-International Edition, 2005. 44(26): p. 4087-4092.

40.

McKenna, B.J., Birkedal, H., Bartl, M. H., Deming, T. J., Stucky, G. D. Self-Assembling Microspheres from Charged Functional Polyelectrolytes and Small-Molecule Counterions. in Mater. Res. Soc. Symp. Proc. 2004.

41.

Iler, R.K., Association between Polysilicic Acid and Polar Organic Compounds. Journal of Physical Chemistry, 1952. 56(6): p. 673-677.

42.

Iler, R.K., Coacervates of Polyvinyl-Alcohol and Colloidal Silica. Journal of Colloid and Interface Science, 1975. 51(3): p. 388-393.

43.

Arshady, R., Microspheres and Microcapsules, a Survey of Manufacturing Techniques .2. Coacervation. Polymer Engineering and Science, 1990. 30(15): p. 905-914.

32

44.

Harada, A. and K. Kataoka, Chain length recognition: Coreshell supramolecular assembly from oppositely charged block copolymers. Science, 1999. 283(5398): p. 65-67.

45.

de Kruif, C.G., F. Weinbreck, and R. de Vries, Complex coacervation of proteins and anionic polysaccharides. Current Opinion in Colloid & Interface Science, 2004. 9(5): p. 340-349.

46.

Doublier, J.L., et al., Protein-polysaccharide interactions. Current Opinion in Colloid & Interface Science, 2000. 5(3-4): p. 202-214.

47.

Veis, A. and C. Aranyi, Phase Separation in Polyelectrolyte Systems .1. Complex Coacervates of Gelatin. Journal of Physical Chemistry, 1960. 64(9): p. 1203-1210.

48.

Weinbreck, F., et al., Complex coacervation of whey proteins and gum arabic. Biomacromolecules, 2003. 4(2): p. 293-303.

49.

Weinbreck, F., et al., Diffusivity of whey protein and gum arabic in their coacervates. Langmuir, 2004. 20(15): p. 6389-6395.

50.

Weinbreck, F., R.H. Tromp, and C.G. de Kruif, Composition and structure of whey protein/gum arabic coacervates. Biomacromolecules, 2004. 5(4): p. 1437-1445.

51.

Prokop, A., et al., Water soluble polymers for immunoisolation I: Complex coacervation and cytotoxicity, in Microencapsulation - Microgels - Iniferters. 1998. p. 1-51.

33

52.

Stuart, M.A.C., et al., Assembly of polyelectrolyte-containing block copolymers in aqueous media. Current Opinion in Colloid & Interface Science, 2005. 10(1-2): p. 30-36.

53.

Leisner, D. and T. Imae, Interpolyelectrolyte complex and coacervate formation of poly(glutamic acid) with a dendrimer studied by light scattering and SAXS. Journal of Physical Chemistry B, 2003. 107(32): p. 8078-8087.

54.

Delacruz, M.O., et al., Precipitation of Highly-Charged Polyelectrolyte Solutions in the Presence of Multivalent Salts. Journal of Chemical Physics, 1995. 103(13): p. 5781-5791.

55.

Cha, J.N., et al., Microcavity lasing from block peptide hierarchically assembled quantum dot spherical resonators. Nano Letters, 2003. 3(7): p. 907-911.

56.

Cha, J.N., et al., Spontaneous formation of nanoparticle vesicles from homopolymer polyelectrolytes. Journal of the American Chemical Society, 2003. 125(27): p. 8285-8289.

57.

Toprak, M.S., et al., Spontaneous assembly of magnetic microspheres. Advanced Materials, 2007. 19(10): p. 1362-+.

58.

Wong, M.S., et al., Assembly of nanoparticles into hollow spheres using block copolypeptides. Nano Letters, 2002. 2(6): p. 583-587.

34

Chapter II

Assembly and Silica Coating of Polypeptide/Small Molecule Coacervates

*Related versions of this chapter have been published as:

Brandon J. McKenna, Henrik Birkedal, Michael H. Bartl, Timothy J. Deming, and Galen D. Stucky, “Micrometer-sized Spherical Assemblies of Polypeptides and Small Molecules by Acid-Base Chemistry,” Angew. Chem. Int. Ed. 2004, 43, 5652-5622 Brandon J. McKenna, Henrik Birkedal, Michael H. Bartl, Timothy J. Deming, and Galen D. Stucky, “Self-Assembling Microspheres from Charged Functional Polyelectrolytes and Small-Molecule Counterions,” MRS Symp. Proc. 2004, 823, 23

35

Abstract: The charged polyamino acids were found to assemble into microspheres

in

combination

with

certain

small,

multivalent

counterions. Assembly was found to be highly dependent on the total number of ionizable groups and their pKa values, relative to the polyelectrolyte pKa values and solution pH. Silica shells were induced on the surfaces of the microspherical assemblies, using either silicic acid condensation or silica nanoparticle precursors.

The use of

nanoparticles created a hollow interior whereas silicic acid did not, suggesting that strong coulombic forces drive the redistribution of assembly components.

36

Self-assembled microspheres are important for their potential to contain and protect one material, while displaying the properties of a different one on the exterior. Such systems could find applications in chemical storage and transport, and in particular, biocompatible microspheres are desirable for applications in drug delivery. Aside from polymeric micelles and liposomes,1 other microsphere systems explored for this application require sacrificial templates and/or surfactants for their self-assembly, or otherwise use organic solvents.2 It was previously shown that microspheres can be obtained directly by self-assembly of CysnLysm block copolypeptides with either citratecoated silver and gold nanoparticles3 or CdSe/CdS nanocrystal quantum dots,4 or of poly-L-lysine (PLK) with citrate-coated CdSe quantum dots.5 These assemblies were mechanically stabilized by adding an outer layer of negatively charged colloidal silica, which also yield “hollow spheres”—that is, microspherical assemblies with internal (core) voids of solution and shells of the assembling components, covered with silica. The goal of this work was to determine the nature of the interaction between citrate-capped inorganic nanoparticles and charged block co-polyamino acids. It was first presumed that the two polymer segments had specific interactions with the two different nanoparticles, 37

resulting in macromolecular, composite lyotropic liquid crystals similar in structure to liposomes. In other words, the cysteine blocks of the polymer would bind to Au NPs, and the lysine blocks would bind to silica, and these building blocks would self-assemble into larger structures. However, this conjectured assembly mechanism was not proven, and it will be demonstrated that it is inconsistent with further scientific observations. Herein we report that large spherical assemblies can be obtained without nanoparticles but simply by reaction of one of several polyelectrolytes

and

certain

small,

functionalized

molecular

counterions.6 These assemblies can then be further functionalized, and we show how those based on polyamines can be protected by silica either in the form of colloidal silica or by condensation of silicic acid. Many earlier efforts have concentrated on the ability of multivalent ions to aggregate oppositely charged polymers, and such systems have been described by theory7,

8

and studied experimentally,9,

10

particularly in

the case of DNA.11 However, in none of these studies was sphere formation observed. In other studies, spherical assembly using polyelectrolytes was observed, but these approaches have required either amphiphilic block copolymers,12-14 proteins,15 two different polyelectrolytes,16 hydrophobic molecules,17 or the presence of both components of an insoluble salt (calcium carbonate).18 38

The work

reported herein demonstrates spherical assembly using only a single polyelectrolyte with one small counterion without stabilization by an inorganic species. The directed formation of hierarchically arranged silica seen in diatoms and sponges provides a promising framework for designing synthetic patterned nanoscale materials. Directed biomineralization can provide novel methods for the assembly of highly ordered structural materials, as demonstrated by the recent in vitro utilization of some biological or biomimetic peptides.19-25 Using a biomimetic approach to silica condensation has many benefits, including: room-temperature synthesis, neutral or moderate pH’s, the opportunity for hierarchical ordering, and the ability to vary the resulting structure by tailoring the active components. Stucky and Morse have shown how silicateins from marine sponges act as catalysts for silica condensation and as scaffolds for the directed growth of polysiloxanes.26,

27

Synthetic

poly(amino acid)s have been shown to mimic silicateins and direct the formation of silica structures, such as spheres.19 Kröger, Sumper and co-workers have reported that silaffins and polyamines from diatoms can template organized silica condensation into 0.3-1- m spheres in the presence of inorganic phosphates.21,

24, 25, 28

In an extension of this

work, Brunner, Lutz, and Sumper have very recently shown that sulfate and phosphate induce microscopic phase separation of polyallylamine 39

(PAA) and that this phase-separated state in turn has high silicaprecipitation activity to yield silica spheres.29 Clarson and co-workers have shown that silica microspheres can be obtained from PLK using silicic acid in phosphate or citrate buffers.22, 23 It is shown herein that preformed assemblies also condense silicic acid, and propose that the formation of microspheres in the work of the groups of Sumper, Brunner, and Clarson may be understood by initial formation of spherical templates, like those presented herein, prior to silica condensation. This model is similar to the microscopic phase-separation picture put forth by Brunner, Lutz, and Sumper.29 The silica spheres from these systems are in direct contrast to the disordered precipitates that result when multivalent anions are not used. 20 Materials. Poly-L-lysine hydrochloride (30 kDa), FITC-labeled poly-L-lysine (70 kDa), poly-L-histidine (10 kDa), poly-L-arginine (30 kDa), poly-L-ornithine (50 kDa), poly-L-glutamate (15 kDa) and polyL-aspartate (35 kDa) were obtained from Sigma and used as received. Snowtex 0 colloidal silica was obtained from Nissan Chemicals. Organic amines were purchased from Sigma-Aldrich. In

extension

of

our

work

based

on

citrate-stabilized

nanoparticles,3-5 we obtained assemblies by the reaction of citrate (final concentration 0.5 wt. %) and PLK (final concentration 0.6 wt. %) at pH

40

7.

Therefore, nanoparticles themselves are not necessary entities;

molecular citrate itself suffices, as shown in figure 2.1. After mixing the two components, the solution immediately turned from clear to cloudy; the resulting colloidal assemblies did not sediment. Within a few minutes of mixing, spheres were observed by light microscopy, as illustrated in Figure 2.1a. Thus, a route to assembly has been found that eliminates the need for nanoparticle reactants. After drying, these assemblies cling to the glass slide and lose their shape. However, they can be stabilized by a protective silica shell through the addition of colloidal silica that condenses on the preformed assemblies (Fig. 2.1b). It was shown that positively charged amine-containing groups of the polymers would be able to attract negatively charged colloidal silica. Adding just a small amount of colloidal silica solution, Snowtex 0, 5% by volume, is enough to add a visible silica shell to the preformed spheres. These spheres can be purified by centrifugation and they maintain their shape out of solution, and can thus be imaged by SEM, etc.

41

Figure 2.1. Images of PLK/citrate microspheres. (a) Optical micrograph of assemblies prior to colloidal silica condensation, with dust particles digitally removed. (b) SEM image of spheres coated with colloidal silica.

Figure 2.2. Chemical structures of the reactants tested for microsphere formation. The polypeptides in the second row are shown with protonated side chains.

To investigate how generalizable the sphere formation process is, we explored a host of multivalent organic acids other than citrate (Table 2.1, Fig. 2.2). PLK-containing spheres were obtained in the presence of two other triacids (isocitrate and trimesate), but with neither diacids nor ethylenediaminetetracacetic acid (EDTA). Assuming that the role of the counterion is to bridge polycations, the diacids may not

42

offer the required kinetic or thermodynamic cooperativity. The failure of EDTA to form spheres with PLK at any pH value is puzzling, but may be related to the better hydration of its carboxylate groups. Table 2.1. Microsphere synthesis at room temperature.[*] Acid

nCOOH

pKa(n)

PLK

PLO

PLH PLR

citric

3

6.43

5.5-9.0 5.5-9.5 4.5-6.0 Precipitate

isocitric

3

6.40

5.5-9.0 5.0-9.5 5.0-6.0 Precipitate

trimesic

3

4.7

4.5-8.0 4.5-9.0 4.0-6.0 Precipitate

4 EDTA 2 carbonate alkanedicarboxylic 2 acids, n(CH2)=0-6 2 tartaric 2 malic 2 fumaric

10.26 (6.16)

NO

NO

NO

6-10

10.33

NO

NO

NO

NO

3.85-5.69

NO

NO

NO

NO (precipitate with oxalate)

4.34

NO

NO

NO

NO

5.2

NO

NO

NO

NO

4.54

NO

NO

NO

NO

[*] Entries indicate which combination of small organic acid and polycation yields assemblies or precipitates, and the approximate pH ranges over which assemblies are visible in the optical microscope. NO=no assembly or precipitate.

Having found some variability in one assembly component, the cationic poly(amino acid) was then varied. The triacids created spheres with poly-L-ornithine (PLO) and with poly-L-histidine (PLH). PLO shows roughly the same behavior as PLK, which is mostly expected since their sidechains differ structurally by only one methylene unit. A striking feature of all the spherical assemblies is that they only existed within certain pH ranges; it appears that all the components must carry

43

a minimum charge5. PLH provides a particularly interesting case in that spheres were not obtained at pH 7 but only below the pKa value of the imidazole side chain of free histidine, 6.0. This result reflects an assembly requirement for charged groups. Indeed, in all cases the lower pH boundary for assembly was defined by the highest pKa value of the acid; in other words, all the carboxyl groups must be deprotonated, and the spheres disassemble at pH values roughly one unit below the acid pKa. This behavior is reversible: the spheres reassemble upon increasing the pH value. Similarly, spheres will not form if the polymer chain is under-protonated, and assembly is only seen up to a pH value similar to the formal pKa of the polypeptide side chains. This situation suggests that the spheres form primarily by electrostatic attraction, which is most likely accompanied by COO/NH3+ acid-base hydrogen bonding that provides further stabilization. The PLH spheres show remarkable behavior when the pH is elevated, however. Around pH 7, PLH becomes insoluble, so the spheres do not dissolve. Instead, spheres of PLH brought to basic conditions aggregate but maintain their individual shapes for a few days, before losing their shapes (Fig. 2.3a). This is important because PLH shows promise as an agent for gene delivery30, and for such applications the assemblies must be able to survive physiological conditions.

44

100 μm

(a)

50 μm

(b)

Figure 2.3. a) Poly-L-histidine/citrate spheres following pH increase. b) Poly-L-arginine/EDTA spheres at pH 12. Poly-L-arginine (PLR), with its pKa 12.5 side chain, also showed remarkably different behavior with the triacids, by forming precipitates even at the highest pHs. However, the PLR-triacid systems were coaxed into assembling spheres by heating the solution, which indicates there is an important entropic contribution to both the stability of the spheres and their kinetics of formation. After cooling, these spheres resolidify and retain their shapes so that stabilization by silica was unnecessary (Fig. 2.4). Also, unlike the other cationic poly(amino acid)s, PLR formed spheres with EDTA, at room temperature. This may be because the highest pKa of EDTA is below that of the arginine side chain, so a proton transfers from EDTA to the side chain and creates an additional acid-base bridge. As with PLH microspheres, the PLR/EDTA microspheres solidified at high pHs (10-12) (Fig. 2.3b).

45

(b)

(a)

Figure 2.4. SEM images of PLR/citrate microspheres following heating and cooling stages. Wrinkling in (b) reflects dehydration that is more pronounced in larger spheres. The stability of PLK/citrate spheres were also tested at concentrations

of

three

monovalent

salts:

NaCl,

KCl,

and

tetramethylammonium chloride. There is little variability between the three cases; the spheres shrink and finally cease to be visible after increasing the salt concentration by about 10 mM, or roughly double the final value of [COO-] and [NH3+]. The spheres may dissolve as a -

result of electrostatic screening or competition between Cl and citrate binding; it is likely a combination of these two effects, since adding more citrate can reform spheres, but adding too much citrate (around 15 mM, depending on the PLK concentration) prohibits sphere formation. The size of the assemblies depends on the ratio of [COO-] and [NH3+]. In agreement with the observations of Brunner, Lutz, and Sumper in the PAA/phosphate system,29 we found that the size of the PLK/citrate assemblies reach a maximum as the [COO-]:[NH3+] ratio is increased.

46

Figure 2.5. Chemical components tested in the synthesis of polyanion/small counterion microspheres.

The microsphere assembly was further extended by combining anionic poly(amino acid)s with cationic molecules containing multiple amine groups (Fig. 2.5). Poly-L-aspartate (PLD) and poly-L-glutamate (PLE) formed spheres with pentalysine and tetraethylenepentamine. Tetravalent tris(ethyleneamine)amine also gave spheres with PLD; however, with PLE, assembly required cooling and produced a combination of spheres and precipitate. Spherical assembly for both polymers did not occur with the explored divalent or trivalent cations: 3,3 -diamino-N-

1,4-bis(3-aminopropyl)piperazine, methyldipropylamine,

melamine,

diethylenetriamine,

2,6-

diaminopyridine, N,N,N ,N ,N -pentamethyldiethylenetriamine, 1-(2aminoethyl)piperazine, 1,3-diaminopropane, 1,6-diaminohexane, 1,8diaminooctane, and 1,12-diaminododecane.

47

The surfaces of the assemblies are chemically active, and a shell of colloidal silica can be deposited. The addition of colloidal silica to a solution of these assemblies gives them a protective shell (Fig. 2.1b). Confocal microscopy was performed on spheres made from FITClabeled PLK (FITC=fluorescein isothiocyanate) to determine polymer distribution. Before the addition of silica, the uncoated assemblies were mobile in solution before adhering to the glass slide and losing shape, and were thus difficult to observe; however, all independent cross-sectional snapshots clearly showed that they are full of polymer, with no noticeable uneven distribution.

This is in contrast to the

organization found in spheres coated with colloidal silica; figure 2.6 shows a series of cross-sectional scans, from which it is obvious that the polymer has rearranged and left a cavity in the center.

Figure 2.6. Cross-sectional images taken with confocal microscopy at equal vertical intervals. The sphere was made from FITC-labeled PLK and citrate, and was coated with colloidal silica. It appears that electrostatic attraction to the negatively charged silica nanoparticles draws most of the polymer towards the shell,

48

leaving a void in the center. This same behavior has recently been observed in systems made with citrate-stabilized gold nanoparticles.31 Figure 2.7 shows a superposition of fluorescent and transmission images, demonstrating that the polymer becomes located inside rim of the silica coat.

(a)

(b)

(c)

Figure 2.7. Cross-sectional images of a colloidalsilica-coated sphere made from FITC-labeled PLK and trimesate, a) the polymer fluorescence in a plane through the center of the sphere, b) the optical transmission image at that focal plane, c) an overlay of (a) and (b). Inspired earlier work,22, 23, 29 we also obtained silica spheres by adding prehydrolyzed tetramethylorthosilicate (TMOS) to a solution of preformed polycation/citrate spheres. TMOS was prehydrolyzed at a final concentration of 0.1 M, in HCl (pH 2-3), for 5 min, before adding a small amount of this precursor ( 5 % v/v) to a solution of polycation/citrate spheres. The prehydrolyzed TMOS could also be quickly neutralized before addition to the spheres. Condensation of the

49

silicic acid was evident from the dark, thin shells seen under the optical microscope, and from the stability of the resulting spheres after drying (Fig. 2.8).

Figure 2.8. PLO/isocitrate spheres functionalized by condensed silicic acid: a) an optical image in which the spheres have a thick, dark outline, in contrast to those in figure 2.1a, b) SEM image of a centrifuged sample.

(a)

Confocal microscopy was performed on these spheres, as had

been done with those coated with colloidal silica, to verify the templating action of the assemblies. Cross-sectional images of spheres after silicic acid condensation also show that their interiors are indeed full of polymer. Moreover, there is no central cavity as was observed for the colloidal silica-coated spheres (Figs. 2.9 & 2.10).

50

20 μm Figure 2.9. Confocal micrograph of FITClabeled PLK/citrate spheres stabilized by silicic acid condensation.

(a)

(c)

(b)

Figure 2.10. Cross-sectional images of FITC-labeled PLK/isocitrate spheres functionalized by condensed silicic acid. Galleries of the fluorescence (a), and transmission (b) responses at different sections through a sphere, c) an overlay of the two images at the central focal plane. The scale bars represent 5 m.

Observations presented here contradict the hollow sphere assembly model in which block copolymers exhibit specific interactions with different nanoparticles. The results suggest a more

51

general assembly process, best described by the existing model of complex coacervation. The substitution of various nanoparticles (Ag, CdSe) for Au and the identical assembly by homopolylysine suggest that cysteine-NP specificity is not fundamentally required. It has been shown herein that citrate itself (the nanoparticle ligand) can be used in place of the full inorganic nanoparticle, and hollow spheres are still produced; furthermore, the assembly is observed with the combination of a range of polyelectolyte/counterion components. In the context of complex coacervation, which was unknown to the researchers at the time, the fact of spherical assembly is hardly surprising, but nevertheless requires verification such as the evidence from confocal microscopy presented here. Without asymmetric components, the mechanism of hollow center formation has to occur by some other mechanism, in order to break the spatial symmetry, besides surfactant-like self-assembly. Two alternative mechanisms for hollow center formation were first considered: 1) the chirality of peptides coupled with condensing counterions may induce asymmetric tertiary structure that induces spherical assembly. For instance, extended planar beta sheets could curve and reduce their free energy by curving into spheres. Rotello et al have demonstrated examples of hydrogen bonding that could beget hollow microcapsules.32

2) The assembly could be template by

52

microscopic air bubbles. Air bubbled through commercial fish tanks, for example, is used to remove organic contaminants and preserve the purity of the water. However, we have shown that the hollow space is formed by a third mechanism: previously full (non-hollow) coacervates expand and condense to their surfaces by the action of strongly oppositely charged colloids. The results were corroborated by independent research, which as well found a variety of assembling chemical components. The work of Wong, et al has demonstrated other surface nanoparticles that can yield hollow structures, for instance using ZnO, SnO2, CdSe carboxylated polystyrene, polyacrylic acid, and polystyrene sulfonate.33 Here, we have demonstrated that the production of a hollow cavity is optional, as silicic acid precursors can be used as an alternative to produce spherical shells with interiors that remain “full”.

Conclusion Complex

coacervates

can

self-assemble

from

low

concentrations of a charged polyamino acid and an oppositely charged, multivalent ion. The coacervates have the same properties as earlier described nanoparticle systems, but they do not require the complexities of large nanoparticles in order to achieve highly ordered systems.

We have shown that all naturally occurring charged

53

poly(amino acid)s can assemble form coacervates in combination with small molecules bearing the proper number of charged groups and pKa values. The pH/pKa requirements and the negative influence of salt concentration confirm the key role of electrostatics in coacervate formation.

However, there do appear to be some additional

thermodynamic requirements for specific polycation/counteranion pairs. That is, in some cases the components remain dissolved, and in other cases a precipitate may form.

Importantly, the coacervate

surfaces are chemically active, as shown by the silica condensation reactions. The resulting shells of silica are uniformly spherical, but the distribution of the internal components depends on the silica source used. In the case of silica nanoparticles, the resulting central void may be useful if the full assembly is to be used as a vesicle for transportation, storage, or isolating chemical reactions. Finally, the silica surfaces can be further functionalized for applications in delivery or detection.

REFERENCES

1.

Euliss, L. E.; DuPont, J. A.; Gratton, S.; DeSimone, J.,

Imparting size, shape, and composition control of materials for nanomedicine. Chemical Society Reviews 2006, 35, (11), 1095-1104. 54

2.

Shchukin, D. G.; Sukhorukov, G. B.; Mohwald, H., Smart

inorganic/organic nanocomposite hollow microcapsules. Angewandte Chemie-International Edition 2003, 42, (37), 4472-4475. 3.

Wong, M. S.; Cha, J. N.; Choi, K. S.; Deming, T. J.; Stucky, G.

D., Assembly of nanoparticles into hollow spheres using block copolypeptides. Nano Letters 2002, 2, (6), 583-587. 4.

Cha, J. N.; Bartl, M. H.; Wong, M. S.; Popitsch, A.; Deming, T.

J.; Stucky, G. D., Microcavity lasing from block peptide hierarchically assembled quantum dot spherical resonators. Nano Letters 2003, 3, (7), 907-911. 5.

Cha, J. N.; Birkedal, H.; Euliss, L. E.; Bartl, M. H.; Wong, M.

S.; Deming, T. J.; Stucky, G. D., Spontaneous formation of nanoparticle vesicles from homopolymer polyelectrolytes. Journal of the American Chemical Society 2003, 125, (27), 8285-8289. 6.

McKenna, B. J., Birkedal, H., Bartl, M. H., Deming, T. J.,

Stucky, G. D. In Self-Assembling Microspheres from Charged Functional Polyelectrolytes and Small-Molecule Counterions, Mater. Res. Soc. Symp. Proc., 2004; 2004; pp 4.12.1-6. 7.

Arenzon, J. J.; Stilck, J. F.; Levin, Y., Simple model for

attraction between like-charged polyions. European Physical Journal B 1999, 12, (1), 79-82.

55

8.

Ha, B. Y.; Liu, A. J., Counterion-mediated attraction between

two like-charged rods. Physical Review Letters 1997, 79, (7), 12891292. 9.

Sabbagh, I.; Delsanti, M., Solubility of highly charged anionic

polyelectrolytes in presence of multivalent cations: Specific interaction effect. European Physical Journal E 2000, 1, (1), 75-86. 10. Spalla,

Delacruz, M. O.; Belloni, L.; Delsanti, M.; Dalbiez, J. P.; O.;

Drifford,

M.,

Precipitation

of

Highly-Charged

Polyelectrolyte Solutions in the Presence of Multivalent Salts. Journal of Chemical Physics 1995, 103, (13), 5781-5791. 11.

Raspaud, E.; de la Cruz, M. O.; Sikorav, J. L.; Livolant, F.,

Precipitation of DNA by polyamines: A polyelectrolyte behavior. Biophysical Journal 1998, 74, (1), 381-393. 12.

Buchhammer, H. M.; Mende, M.; Oelmann, M., Formation of

mono-sized polyelectrolyte complex dispersions: effects of polymer structure, concentration and mixing conditions. Colloids and Surfaces a-Physicochemical and Engineering Aspects 2003, 218, (1-3), 151-159. 13.

Otsuka, H.; Nagasaki, Y.; Kataoka, K., PEGylated nanoparticles

for biological and pharmaceutical applications. Advanced Drug Delivery Reviews 2003, 55, (3), 403-419. 14.

Solomatin, S. V.; Bronich, T. K.; Bargar, T. W.; Eisenberg, A.;

Kabanov, V. A.; Kabanov, A. V., Environmentally responsive

56

nanoparticles from block ionomer complexes: Effects of pH and ionic strength. Langmuir 2003, 19, (19), 8069-8076. 15.

Weinbreck, F.; Tromp, R. H.; de Kruif, C. G., Composition and

structure of whey protein/gum arabic coacervates. Biomacromolecules 2004, 5, (4), 1437-1445. 16.

Biesheuvel, P. M.; Stuart, M. A. C., Electrostatic free energy of

weakly charged macromolecules in solution and intermacromolecular complexes consisting of oppositely charged polymers. Langmuir 2004, 20, (7), 2785-2791. 17.

van Bommel, K. J. C.; Jung, J. H.; Shinkai, S., Poly(L-lysine)

aggregates as templates for the formation of hollow silica spheres. Advanced Materials 2001, 13, (19), 1472-+. 18.

Gower, L. B.; Odom, D. J., Deposition of calcium carbonate

films by a polymer-induced liquid-precursor (PILP) process. Journal of Crystal Growth 2000, 210, (4), 719-734. 19.

Cha, J. N.; Stucky, G. D.; Morse, D. E.; Deming, T. J.,

Biomimetic synthesis of ordered silica structures mediated by block copolypeptides. Nature 2000, 403, (6767), 289-292. 20.

Coradin, T.; Durupthy, O.; Livage, J., Interactions of amino-

containing peptides with sodium silicate and colloidal silica: A biomimetic approach of silicification. Langmuir 2002, 18, (6), 23312336.

57

21.

Kroger, N.; Lorenz, S.; Brunner, E.; Sumper, M., Self-assembly

of highly phosphorylated silaffins and their function in biosilica morphogenesis. Science 2002, 298, (5593), 584-586. 22.

Patwardhan,

S.

V.;

Clarson,

S.

J.,

Silicification

and

biosilicification - Part 1. Formation of silica structures utilizing a cationically charged synthetic polymer at neutral pH and under ambient conditions. Polymer Bulletin 2002, 48, (4-5), 367-371. 23.

Patwardhan, S. V.; Mukherjee, N.; Steinitz-Kannan, M.;

Clarson, S. J., Bioinspired synthesis of new silica structures. Chemical Communications 2003, (10), 1122-1123. 24.

Poulsen, N.; Sumper, M.; Kroger, N., Biosilica formation in

diatoms: Characterization of native silaffin-2 and its role in silica morphogenesis. Proceedings of the National Academy of Sciences of the United States of America 2003, 100, (21), 12075-12080. 25.

Sumper, M.; Lorenz, S.; Brunner, E., Biomimetic control of size

in the polyamine-directed formation of silica nanospheres. Angewandte Chemie-International Edition 2003, 42, (42), 5192-5195. 26.

Shimizu, K.; Cha, J.; Stucky, G. D.; Morse, D. E., Silicatein

alpha: Cathepsin L-like protein in sponge biosilica. Proceedings of the National Academy of Sciences of the United States of America 1998, 95, (11), 6234-6238.

58

27.

Zhou, Y.; Shimizu, K.; Cha, J. N.; Stucky, G. D.; Morse, D. E.,

Efficient catalysis of polysiloxane synthesis by silicatein alpha requires specific hydroxy and imidazole functionalities. Angewandte ChemieInternational Edition 1999, 38, (6), 780-782. 28.

Sumper, M., A phase separation model for the nanopatterning of

diatom biosilica. Science 2002, 295, (5564), 2430-2433. 29.

Brunner, E.; Lutz, K.; Sumper, M., Biomimetic synthesis of

silica nanospheres depends on the aggregation and phase separation of polyamines in aqueous solution. Physical Chemistry Chemical Physics 2004, 6, (4), 854-857. 30.

Putnam, D.; Gentry, C. A.; Pack, D. W.; Langer, R., Polymer-

based gene delivery with low cytotoxicity by a unique balance of sidechain termini. Proceedings of the National Academy of Sciences of the United States of America 2001, 98, (3), 1200-+. 31.

Murthy, V. S.; Cha, J. N.; Stucky, G. D.; Wong, M. S., Charge-

driven flocculation of poly(L-lysine)-gold nanoparticle assemblies leading to hollow microspheres. Journal of the American Chemical Society 2004, 126, (16), 5292-5299. 32.

Thibault, R. J.; Hotchkiss, P. J.; Gray, M.; Rotello, V. M.,

Thermally reversible formation of microspheres through non-covalent polymer cross-linking. Journal of the American Chemical Society 2003, 125, (37), 11249-11252.

59

33.

Rana, R. K.; Murthy, V. S.; Yu, J.; Wong, M. S., Nanoparticle

self-assembly of hierarchically ordered microcapsule structures. Advanced Materials 2005, 17, (9), 1145-+.

60

Chapter III

Superparamagnetic Nanoparticle Coacervates for Targeted Drug Delivery

*Related versions of this chapter have been published as:

Muhammet S. Toprak, Brandon J. McKenna, J. Herbert Waite, and Galen D. Stucky, “Spontaneous Assembly of Magnetic Microspheres,” Adv. Mat. 2007, 19, 1362-1368 Muhammet S. Toprak, Brandon J. McKenna, J. Herbert Waite, and Galen D. Stucky, “Tailoring Magnetic Microspheres with Controlled Porosity,” MRS Symp. Proc. 2007, 969, W03-11 Muhammet S. Toprak, Brandon J. McKenna, J. Herbert Waite, and Galen D. Stucky, “Control of Size and Permeability of Nanocomposite Microspheres,” Chem. Mat. 2007, 19, 4263-4269

61

Abstract: We present here a facile bio-inspired method for the synthesis of inorganic/organic hybrid drug delivery devices based on complex coacervation.

Microspheres were spontaneously formed from the

interaction between cationic polyamines and citrate-coated magnetite nanoparticles, without the use of a template or surfactant. Control over the microsphere size distributions were achieved by varying amine/carboxylate ratios, aging times, temperature, polymer molecular weight, salt content, and dilution factors. The hybrid spheres were found to be stabilized by the addition of glutaraldehyde, and the effect of this organic cross-linking was found to affect the microsphere porosities and the diffusion coefficients of dextran molecules of various molecular weights, within the hybrid magnetic microspheres.

62

The ability to assemble materials that are organized over different length scales has a recognized importance for the development of new functional materials. In particular, the potential for application of emerging nanoscale objects can often only be realized by arranging such components into larger scale assemblies, such that the product exhibits the functionalities of each of its constituents. One embodiment of this concept is in the fabrication of microspheres. Microspheres have been widely considered for isolation of chemical reactions, making them an active research topic in recent decades. In particular, biocompatible microspheres are desirable for applications in drug delivery or other bio-related applications. Targeted drug delivery has been sought for at least three decades.1,

2

In the general scheme, drug-encapsulating vesicles

circulate in the bloodstream until they encounter a chosen cell type to which they then attach and release their drug or toxin into the cellular environment. The advantages of this process over systemic delivery are reduced side effects and reduced drug costs, because the drug is not exposed to the rest of the body and a much smaller amount is required for the desired effect. Prohibitively expensive but highly effective (or highly toxic) drugs could find commercial application.3-8

63

The designs of such systems have at least 3 requirements: an encapsulating material, targeting moieties, and a mechanism for drug release.

Practically, however, more functions are required of such

devices: circulation time in the body must be increased because particulates are rapidly removed by the liver, so it has become standard practice to integrate biologically inert surface materials such as PEG.6, 9 The devices must also be biocompatible/biodegradable, they must have proper shape (usually spherical) to interact with cells,10-12 and they must be small enough to circulate through capillary pores and reticular endothelial system (RES), and preferably to endocytose.13 Liposomes are the most common material for encapsulation, as phospholipids are biocompatible and their assembly is well understood.2,

14

The greatest difficulty has been modulating their

stability to not release in the bloodstream and yet selectively release inside of cells. For this reason, a number of new ‘smart’ materials have been investigated that destabilize under altered solution conditions, most commonly pH.15-17 This, in effect, combines the ‘encapsulating’ and ‘trigger’ considerations into one. Cancer cells have slightly lower pHs than other cells,18 so materials with proper pKa’s or which incorporate chemical bonds whose destruction is catalyzed by H+ have been researched.

I.e., the first category comprises materials with

noncovalent interactions such as polymeric micelles,15, 16, 19, 20 and layer

64

by layer (LbL) polyelectrolyte capsules deposited onto sacrificial cores,21, 22 and the second category comprises molecules such as lipids with reversible covalent linkages, such as ketals, vinyl ethers, or ortho esters.23 Complex coacervation presents new opportunities for singlestep syntheses of ordered micron-scale objects that are composed from predefined nano-scale objects. Coacervation is a spontaneous aqueous phase separation, in which liquid-like microspheres are produced from oppositely-charged chemical entities.24 This imparts some advantages over other systems which require sacrificial templates, surfactants, or else the use of organic solvents. A new advantage of coacervation has emerged with the discovery of nanoparticle-incorporating coacervation in our laboratory.25,

26

Nanoparticles have established importance in

materials science for their useful physical properties (fluorescent, magnetic, electrochemical) that emerge at the nanoscale. However, for commercial

application,

nano-entities

now

require

appropriate

organization in order to achieve bulk material properties that take advantage of their functions.

Coacervation is now one method to

obtaining microspherical packages containing dense collections of nanoparticles. We have previously studied such assemblies made from organic ions, metallic nanoparticles, or quantum dots; self assembly of

65

CysnLysm block copolypeptides with citrate coated gold or silver nanoparticles, CdSe/CdS quantum dots, or of poly-L-lysine (PLK) with citrate coated CdSe quantum dots.25-27 We have sought to extend the variability of assembly components in order to expand the variety of potential uses. We have utilized the coacervates’ chemical properties to form inorganic shell structures. Maghemite nanoparticles and block copolypeptides containing polyaspartic acid were used to form~100 nm clusters.28 Some of these assemblies were mechanically stabilized by the addition of an outer layer of negatively charged colloidal silica. A report by Wong et al. demonstrated the formation of supramolecular aggregates between cationic polyamines and multivalent counteranions via ionic cross-linking in a two-step process; and negatively charged nanoparticles deposit around these aggregates to form a multilayerthick shell.29 We have considered coacervation as a potential ‘smart encapsulating material’ for use in targeted drug delivery.

As

mentioned, assembly is facile and fast, and the distinct coacervate entities require no grinding and present smooth spherical surfaces that are available for chemical (covalent) modification, such as the attachment of targeting antbodies or PEG. The range of coacervating components includes several inexpensive biocompatible polymers of varying pKa’s, and the expanded number of stabilization methods

66

provides a platform for tailorable release mechanisms.

Finally,

coacervates readily incorporate various compounds24 including hydrophobic liquids,30-32 which simplifies constraints for drug encapsulation. In particular, there are many benefits to be derived from integrating magnetic components into a coacervate microsphere formulation. Colloidal particles with magnetic properties have gained increasing attention both technologically and for fundamental studies due to the tunable anisotropic interaction they exhibit.33, 34 They find widespread and diverse use in many fields, such as environmental remediation (removal of toxic and radioactive waste), therapy35 (controlled drug targeting,36 hyperthermia37) and diagnostic biomedical applications

(ELISA,

NMR

imaging,

sensing).38-40

Magnetic

nanoparticles have been used extensively in the field of biomagnetics. This field currently consists of a broad range of applications, including drug

delivery,36

cell

separation,41

biosensing,42

separation

of

biochemical products,43 and cell labeling and sorting.44 In the case of drug delivery, magnetic fields are utilized to direct the particles, and hence the encapsulated drug, to specific locations within the body— thus expediting delivery of drugs and minimizing side effects. We have foremost considered magnetic nanoparticles (MNPs) for targeted drug delivery.

For this application, superparamagnetic

67

nanoparticles, which are smaller than the bulk material’s characteristic domain size, are desirable, since upon removal of the magnetic field they lose their magnetic moment and do not aggregate.40, 45 For these studies, superparamagnetic magnetite (Fe3O4) nanoparticles were used, because of their suspected biodegradability and known chemical synthesis.

We consider them for targeted drug delivery for three

functions: 1) Their location in the body can be imaged with magnetic resonance imaging (MRI). Application as MRI ‘negative’ contrast agents is, itself, important, but for targeted drug delivery, this would provide a method for visualizing and confirming proper device delivery prior to release. 2) They may impart directed delivery capability.

While

ordinary magnetic fields could not completely counteract blood flow and permit magnetically guided delivery, a magnetic field applied near the targeted site (especially near the body surface) could increase the residence time of the device, to orient appropriately and hence the likelihood of attaching to the targeted epithelial cells. 3) They may provide a mechanism for ‘triggered release’. Superparamagnetic nanoparticles have been studied for hyperthermia applications, in which an external oscillating

68

magnetic field (in the form of RF) induces NP oscillation and localized heating.45,

46

This ordinarily confers the

advantage of heating and liberating gaseous O2 within cancer cells, which weakens the cell and makes it more susceptible to toxins (systemic or not). In a coacervatebased design, the MNPs could also destabilize or ‘selfdestruct’ the coacervate this way, and release the encapsulated drug on command. In one example a recent synthetic method, Caruso et al. reported the LbL production of composite magnetic core-shell particles; the shells consisted of magnetite nanoparticle/polyelectrolyte multilayers and the colloidal cores were polystyrene latex microspheres.22 For the various applications of such systems, critical parameters to control include shell permeability, biocompatibility, mechanical stability, and size control. Möhwald et al. improved and well tailored the properties of these polyelectrolyte microcapsules, exemplified with PAH/PSS microcapsules templated on MnCO3 and melamine formaldehyde (MF) particles, by glutaraldehyde cross-linking.47 In this study we present a spontaneous, single-step synthesis of MNP spheres using a combination of positively charged homopolymers and negatively charged magnetic nanoparticles. Polyamino acids have attracted considerable interest in recent years due to their non-toxicity,

69

biocompatibility, nutritional function and pharmacological efficacy as drug carriers.48,

49

Polyamines provide one promising means of

stabilizing drug carriers, since they can self-assemble into micelles via specific interactions with polyacids50-52 that are capable of trapping drugs within a core, while their exteriors can be designed to be stable within a wide range of physiological environments. As with our earlier investigations, hybrid coacervates could also be formed using other cationic homopolyamino acids.

Herein, poly-L-lysine (PLK) was

chosen as the polycation to interact with citrate-capped magnetite NPs, because PLK a biodegradable and its coacervation properties were understood from previous studies.50, 51 The application of coacervate devices depends on the control of several other important parameters, including their size, stabilization, and loading capability.

We present studies for each of these

parameters: We demonstrate several solution methods for adjusting microsphere size.

We employ glutaraldehyde crosslinking for

coacervate stabilization; this method was selected because the reaction is well-understood, easily conducted, and easily tracked by the absorbance of newly formed imide bonds, which also render a positive surface charge. We show loading capability by demonstrating the tailorability of critical pore sizes and permeabilities, via cross-link density. Cross-linked hybrid microspheres were furthermore screened

70

for their toxicity and tested for their ability to incorporate/encapsulate functional molecules as potential drug delivery system.

Experimental Methods Chemicals. FeCl2.4H2O, FeCl3.6H2O, NH4OH, HCl, Poly-LLysine, (PLK 14 kDa, 26 kDa, 46 kDa, 67 kDa), trisodium citrate (TSC), and FITC-Dextran with different MW (4, 10, 20, 40, 70, and 250 kDa), NaCl, CaCl2, MgCl2, Phosphate Buffered Saline (PBS), ammonium acetate, hydroxylamine HCl, o-phenanthroline, and methanol were obtained from Sigma-Aldrich.

Glutaraldehyde and

glycine were obtained from Ted Pella Inc., CA, USA. All chemicals were used as received and aqueous solutions were prepared by dissolving the corresponding chemicals in DI water, 18 MΩ. Surface modified magnetic nanoparticles. Superparamagnetic nanoparticles were prepared by using a previously described procedure.53 In a typical process, a mixture of Fe2+ and Fe3+ were hydrolyzed with NH3 solution at pH >10 in an oxygen-free atmosphere. Afterwards, the reaction mixture was heated to 80oC under Ar flow, followed by the addition of TSC.54 Subsequently, the reaction mixture was cooled, and the magnetic nanoparticles (MNPs) were collected with the help of a strong magnet and washed 4 times with hot water to

71

remove the excessive reagents. Collected particles were then dispersed in TSC and stored in a refrigerator for future use. Coacervate formation and cross-linking. Poly-l-lysines, regardless of molecular weight, were dissolved in deionized water at concentrations of 2 mg/mL

A solution of magnetic nanoparticles

(MNP) having a net negative surface charge was mixed with a chosen amount of PLK solution, upon which the solution turned cloudy. A typical sample was fabricated by mixing 20 μL of 2 mg/mL PLK solution with 120 μL MNP (1.6 mg/ml TSC). The reaction was mixed vigorously for 15 seconds using a vortex-mixer. Glutaraldehyde, GA, has been used frequently as a cross-linking reagent, because of its lower cost, nontoxicity, and high solubility in aqueous solution.55 All samples in the following section were crosslinked by the addition of 120 μL of 2.5 wt% glutaraldehyde, excess of which was quenched by the addition of glycine. Cross-linked coacervates are referred to as spheres/microspheres in the following sections.

Size Effects. Particle Size Analysis

Particle size analysis was performed on

several SEM micrographs counting a minimum of 150 cross-linked

72

microspheres. Size distribution plots are presented using the average nanocomposite sphere size with one standard deviation, dave ± SD. [COO-]/[NH3+] ratio, R: The effect of R on microsphere size was investigated in the range R=1-20. A set of MNP solutions were prepared by suspending MNPs in TSC solutions with concentrations in the range 0.77 mM – 15.3 mM while keeping the MNP concentration constant at 3x1013 MNP/mL. The volumes of PLK (20 μL) and MNP (120 μL) were preserved. Temperature: The effect of temperature on the size of hybrid spheres was investigated in the range 30-70oC. Solutions of PLK4 and MNP with R=7 were equilibrated in a water bath for 5 min at the designated temperature. Solutions were then mixed and kept at the target temperature for another 3 min, after which GA was added to initiate cross-linking. Aging: The effect on size by aging coacervate solutions was studied in the range of 2 min to 3 days. PLK4 and MNP solutions were mixed to yield an R factor of 7 and aged for different durations prior to cross-linking. Polymer Molecular Weight: The influence of polymer molecular weight, MW, on size was examined using commercially available grade PLKs, as detailed in Table 3.1. PLK1-PLK6 (2 mg/mL) and MNP solutions were mixed at R=7 and aged for 6 min prior to cross-linking.

73

Table 3.1. Poly-L-Lysines Used for Experiments a

Poly-L-Lysine

MW, kDa

PLK1 PLK2 PLK3 PLK4 PLK5 PLK6-FITC b a

Data provided by the supplier; supplier.

Ionic Strength.

b

Dispersity index a Mw/Mn

14 28 46 67 130 59

1.4 1.3 1.4 1.1 1.6 NA

0.008 moles FITC/mole lysine unit, provided by the

The effect of ionic strength on size was

investigated by using chloride salts with different cations; namely NaCl, CaCl2, and MgCl2. Stock solutions of these salts at 50 mM concentration were prepared and added to PLK solutions prior to mixing with TSC and glutaraldehyde crosslinking. Dilution. The effect of dilution on size was investigated by adding different volumes of DI water to PLK solution, prior to mixing with TSC and glutaraldehyde crosslinking. Scanning Electron Microscopy, SEM. Scanning Electron Microscopy analysis was performed using FEI XL40 Sirion FEG Digital Scanning Microscope. Following cross-linking and quenching steps, samples were centrifuged and the solution decanted, and then the microspheres were re-dispersed in DI water. This process was repeated three times per sample and the resultant suspension was dropped on a

74

Si wafers and dried. Gold coating was performed at 10 mA for 2 min, prior to SEM analysis. Several SEM micrographs were taken and particle size analysis was performed by counting a minimum of 150 coacervates. Size distribution plots are presented as the average coacervate size with one standard deviation, Dave ± SD. Transmission

Electron

Microscopy,

TEM.

Transmission

Electron Microscopy analysis was performed using a FEI Tecnai G2 Sphera Microscope. A drop of sample, from the procedure used in SEM preparation, was diluted in alcohol and a drop was placed onto the TEM grid. Following drying, the sample was loaded into a TEM column for analysis. Particle size distribution was obtained from three micrographs counting a minimum of 150 particles. Electron diffraction (ED) was also performed to identify the crystalline phase. UV-Vis Spectroscopy, UV-Vis. The concentration of magnetite nanoparticles was determined by a colorimetric method, using UV-Vis spectrometry by using phenanthroline complexation.56 A known amount of stable magnetite stock solution was leached in HNO3, reduced by hydroxylamine. Additions were made of ammonium acetate buffer followed by o-phenantroline. The absorption of the resulting iron-phenantroline complexes was observed at 510 nm, as a function of concentration. An estimate value of 5×1014 MNP/mL was calculated using the concentration obtained from UV-vis analysis and the average

75

particle size obtained from TEM analysis. MNP solutions in TSC were prepared from this stock solution to have a final MNP concentration of 3×1013 MNP/mL. Confocal Microscopy. A Leica microscope equipped with an ArKr laser was used for performing laser scanning confocal microscopy. Samples for confocal microscopy were prepared by using fluorescein isothiocyanate (FITC) labeled PLK and FITC-dextran. A drop of sample was sandwiched between a glass slide and a coverslip, which was sealed to avoid evaporation. Infrared Spectroscopy, IR. Infrared spectroscopy was performed using a Nicolet Magna 850 IR/Raman spectrophotometer. Infrared spectra of pure TSC, uncoated and citrate and coated magnetite particles were obtained using KBr pellets. Incubation tests.

After spontaneously assembled composite

microspheres were cross-linked for different durations, they were collected by centrifugation and then re-dispersed in DI water. FITCdextran solutions with different MWs were then added in 20 µL (~2 mg/mL) portions to 100 µL of the microspheres. After 5 min of incubation/equilibration, the samples were transferred to glass-slides and analyzed by confocal laser scanning microscopy (CLSM). Quantification of Diffusion Coefficient/Permeability. The microsphere permeability was quantified by means of fluorescence 76

recovery after photobleaching (FRAP) using FITC-dextran as a molecular probe. To follow the diffusion of FITC-dextran into the microsphere, the microsphere’s interior was photochemically bleached with the CLSM ArKr laser (488 nm), at 100% intensity, for sufficient durations.

Imaging was typically performed at about 4% of the

maximal laser intensity. The interval between image scans varied, depending on recovery rates established in preliminary experiments. Recovery was considered complete when the intensity of the photobleached region plateaued. For quantitative analysis, the fluorescence intensity signals within closed circular areas were averaged to yield intensity values at each interval.

Results and Discussion. As determined by TEM (Fig. 3.1a), the synthesized magnetic iron oxide nanoparticles particle size distribution (Fig. 3.1b), yielded an average size of 10.5±0.3 nm. The electron diffraction pattern (Fig. 3.1a, inset), indicated a polycrystalline sample and was indexed to magnetite. The room temperature magnetization curve, as measured by vibrating sample magnetometry, is shown in figure 3.1c. The absence of remanence magnetization and coercivity confirms that the nanoparticles are superparamagnetic. The mean size of the magnetic core was also

77

calculated to be 10.5±0.5 nm, by assuming a lognormal distribution of particle volumes for the Langevin function.57 [220]

(a)

(b)

40

[311]

30

[422] [511]

% counts

[400]

20

10

[440] 0 8

9

10

11

12

13

Dh [nm]

60 50

Magnetization [emu/g]

40 30 20 10 0 -10 -20 -30 -40 -50 -60

(c)

-20000

-10000

0

10000

20000

Applied Field [Oe]

Figure 3.1. (a) TEM micrograph, and SAED of magnetic nanoparticles; (b) size distribution histogram; (c) Room temperature magnetization curve of iron oxide nanoparticles. The MNP surfaces were modified with citrate in order to effect the desired self-assembly with polycations, and changes in the MNPs with different processing were assessed by IR analysis. Uncoated MNPs only had significant IR absorption bands at 580 cm-1 and 3400 cm-1. In the range 1000-100cm-1, the IR bands of solids usually relate to crystal lattice ion vibrations.58 The band observed around 600-550 cm-1 corresponds to intrinsic stretching vibrations of the tetrahedral metal site.59 The 3400 cm-1 band is due to physically adsorbed water. Surface modification with citrate yielded extra bands at 1395 and 1620 cm-1, which were assigned to asymmetric and symmetric COO- stretching

78

vibrations, from citrate. Surface modification also affected zeta potential (Fig. 3.2b). The uncoated MNPs displayed an isoelectric point around pH 7, whereas the citrate coated MNPs remained negatively charged, due to the COO- groups of trisodium citrate. 40

uncoated magnetite magnetite w/ citrate

30

ζ potential [mV]

% Transmittance

20

(a) 4000

3500

2500

2000

0 -10 -20 -30

--- uncoated magnetite ... trisodium citrate ___ magnetite w/ citrate

3000

10

(b)

-40 1500

1000

50

1

2

3

4

5

6

7

8

9

10

11

12

pH

-1

Wavenumber [cm ]

Figure 3.2. (a) IR spectra of uncoated magnetite, pure trisodium citrate, and citrate coated magnetite nanoparticles; (b) Zeta potential measurements for uncoated and citrate coated magnetite nanoparticles. Hybrid microspherical coacervates of MNP and PLK were assembled by Coulombic interactions between the positively charged amine groups and negatively charged carboxylates, an interaction that has been hypothesized as charge-stabilized hydrogen bonding.25 Light microscopy imaging revealed oil droplet like formations diameters of ~1 µm (Fig. 3.3a). Confocal fluorescence microscopy images and cross-section analysis of the coacervates, prepared using PLK6, show dome-like features (Fig. 3.3b) because the liquid coacervates wet the substrate surface. There is synthetic flexibility for tailoring and the amount of MNP loading within a desired range. Changing R, while

79

keeping [MNP] constant, hybrid spheres with 3-13 wt% MNP loading have been produced. (a)

(b)

Figure 3.3. (a) Light microscopy and (b) Confocal microscopy images of un-cross linked PLK-MNP spheres. Mechanical stability of the assemblies is essential for their successful application as devices. However, coacervation is dynamic, and the interactions are easily disturbed by various factors that alter the degree of ionization, such as as ionic strength, pH, and temperature. For example, a deviation of ±1 unit from the pI of the polyelectrolyte leads to dissolution.

The addition of even a small amount (10 mM) of

indifferent salt disturbs the system by electrostatic screening. Therefore PLK was cross-linked using glutaraldehyde (GA), a known effective cross-linker between amine groups. Cross-linking introduces new C=N- imide bonds that give rise to absorbance in the UV region.47 Therefore, the cross-linking was followed and verified by UV-Vis spectroscopy (Fig 3.4b). The broad absorption in the UV region of 250-

80

300 nm increased with the treatment time, indicating more imide bonds. The absorbance at 266 nm as a function of time is shown in the inset, which indicates that the reaction initially proceeds very quickly and then gradually slows. Cross-linking for extended periods resulted in aggregation; sedimentation can be monitored by decreasing absorption. Hence, the degree of cross-linking can be conveniently tuned by reaction time prior to quenching with glycine. 2

(a)

(b) Absorption [arb. u.]

Absorption @ 266 nm [arb. u.]

1.2

(c)

1.1

1.0 0.9 0.8 0.7

0.6 0

10

20

30

Time [min]

0 250

2 μm

300

350

400

Wavelength [nm]

Figure 3.4. (a) Optical micrograph of as-prepared magnetic coacervates; (b) UV-Vis spectra of PLK/MNP coacervates crosslinked with 2.5 wt% GA for different durations: 0, 1, 4, 6, 9, 13, 25 min., from bottom to top. The inset shows the absorption due to newly formed imide bonds at 266 nm as a function of time; (c) SEM micrograph of cross-linked magnetic microspheres. Size Study Results. Cross-linked hybrid PLK-MNP microspheres were examined with SEM for their size and morphology (Fig. 3.5). Individual MNPs as well as aggregates of MNPs can easily be seen in figure 3.5b. The formation and size of hybrid coacervates was found to

81

be dependent on the [COO-]/[NH3+] ratio R, PLK MW, aging time, temperature, ionic strength, and dilution.

(a)

(b)

1 μm

200 nm

Figure 3.5. SEM micrographs of cross-linked PLK-MNP spheres: (a) General view; (b) Close-up view showing individual nanoparticles and aggregates The effect of R on size was investigated in the range of 1-20 (Fig. 3.6a). Below R=3, no coacervation occurred. This contrasts with an earlier report where excess positive charge was required for vesicle formation.25 Because R=1 indicates a net charge balance of negatively and positively charged participating ions, this result indicates that the MNP/citrate bridges are too mobile and labile to effectively form noncovalent crosslinks. Coacervates appeared at R=3, with an average size of 1.2 μm, which decreases with increasing R, down to 560 nm at R=20.

This trend is likely the result of increasing ionic strength,

although it may also reflect faster nucleation kinetics or a thermodynamic drive to decrease surface tension (by reducing the number of MNPs per total surface area).

82

The microspheres are

positively charged at all R values, revealing the dominant role of PLK in these formations, in a “cloud” of smaller anions.

1.8

1300

(a)

1.6

1200

1.4

(b)

1100

Particle Size [nm]

Dave [μm]

1.2 1.0 0.8 0.6 0.4

1000 900 800 700

0.2 600

0.0 500

-0.2 R

R1

R2

R3

R5

R6

R7 -

R8

R10

R15

R20

--

T30

+

R [CH3COO ]/[NH3 ]

T40

T50

T60

T70

o

Temperature [ C]

Figure 3.6. Dependence of sphere size on: (a) [COO-]/[NH3+] ratio, R; (b) Temperature. The effect of temperature on the size of the hybrid spheres in the temperature range 30-70oC is shown in Figure 3.6b. The average size of the hybrid spheres is around 700 nm at ambient temperature. This value increases to ~1 μm at 70oC, which is expected, since aggregation of like-charged bodies requires thermal energy to overcome Coulombic repulsion. Aging in the range of 2 min–3d also affected the coacervate size (Fig. 3.7a). A distinct size distribution difference was observed, and larger MNP/PLK hybrid spheres were found with longer aging times. The sphere size ranged from an average diameter of 700 nm at 2 min to about 1.7 μm at 24 hrs. Prolonged aging resulted in the disappearance

83

of the coacervates because they adhered to container walls, as was visualized using PLK6.

2.4

2.4

(a)

2.0

2.0

1.6

Dave[μm]

1.6

Dave [μm]

(b)

1.2

0.8

1.2

0.8

0.4

0.4

0.0

0.0 0

10

20

30

40

50

60

70

80

90

2000 3000 4000 5000

13K

28K

Aging time [min]

46K

67K

130K

MW PLL [kDa]

Figure 3.7. (a) Aging time; and (b) Poly-L-Lysine molecular weight dependence of the size of PLK-MNP spheres, R=7. The microsphere size shows a sub-linear square root dependence to the aging time. This is in agreement with earlier findings on polymer-colloidal particles, and indicates a diffusion limited process in which polymer and/or MNP adsorbs onto coacervate surfaces faster than bulk diffusion within the coacervates.60,

61

Other mechanisms

could also contribute to growth, such as entropy-minimizing coalescence and Ostwald ripening, in which larger coacervates grow at the expense of smaller ones. MW dependence of the hybrid spheres size (Fig 3.7b) is summarized in table 3.2 with the average sphere size, Dave, standard deviation, SD, and minimum and maximum sizes observed for each set. No coacervates were observed for the lowest two MWs (13 kDa and 26

84

kDa). Hybrid coacervate size thereafter increased with PLK MW, from 600 nm (at 46 kDa) to 1.9 μm (at 130 kDa). The threshold MW of PLK for hybrid coacervate formation is suggested to be between 26 kDa and 46 kDa, for this particular system.

Table 3.2. Size of hybrid spheres as a function of PLK MW at R=7. Poly-LLysine

MW, kDa

Dave, μm

SD

Min

Max

PLK1 PLK2 PLK3 PLK4 PLK5

14 26 46 67 130

--0.61 0.95 1.64

--0.13 0.29 0.5

--0.34 0.31 0.66

--0.88 1.61 2.95

The effect of ionic strength on the coacervates size was investigated by using salts with cation valencies of +1 and +2 (Figs. 3.8 and 3.9). The average size was observed to decrease with increasing ionic strength. In the case of NaCl addition, the average microsphere size was reduced from 700 nm to 300 nm with 25 mM of added ionic strength (Fig. 3.8). No formation was observed above 25 mM ionic strength for NaCl. In the case of CaCl2, the average size decreased to 350 nm with an added ionic strength of 6 mM, above which coacervates dissolved (Fig. 3.9). The decrease in average size was attributed to electrostatic screening of noncovalent bridges. When Mg2+ was used, no spherical assemblies were observed at any ionic strength conditions.

85

The significant difference in responses to these two different salts is attributed to complexation of Ca2+ with citrate ions.

This greater

activity of Mg2+ is attributed to its much larger radius of hydration (~4X larger than hydrated Ca2+), which causes significant shielding, spatially, of citrate ions and thus even more interference with assembly.62

55 50 45 40

32

dave

0.9

Ionic strength

28

0.8

(a)

30 25 20 15

(b)

0.7 0.6

dave [μm]

% count

35

1.0

50 mM NaCl 0 μL 6 μL 16 μ L 35 μ L 60 μ L 140 μL

0.5

12

0.3 8 4

0.1

5 0

0.0

-5 100

20 16

0.4

0.2

10

24

Ionic Strength [mM]

60

0

-0.1

1000

0

dave [nm]

50

100

150

200

250

VNaCl [μL]

Figure 3.8. The effect of adding aliquots of 50 mM NaCl on the average hybrid microsphere diameter and solution ionic strength. Error bars indicate one standard deviation.

30

20 dave

0.9

18

Ionic Strength

0.8

16

0.7

(a)

25 20 15 10 5 0

14

(b)

0.6

dave [μm]

% count

1.0

50 mM CaCl2 0 μL 3 μL 6 μL

35

0.5

10

0.4

8

0.3

6

0.2

4

0.1

2

0.0

0

-0.1

100

12

1000

-2 0

dave [nm]

Ionic Strength [mM]

40

5

10

15

VCaCl [μL] 2

Figure 3.9. The effect of adding aliquots of 50 mM CaCl2 on the average hybrid microsphere diameter and solution ionic strength. Error bars indicate one standard deviation.

86

The effect of dilution on the average size of hybrid microspheres was similar to that observed for increasing ionic strength (Fig. 3.10).

Dilution is expected to have two effects on size.

Kinetically, rapidly-formed charged coacervate ‘nuclei’ are too spatially separated to overcome repulsion and then coalesce. Thermodynamically, dilution increases the total amount of material in the equilibrium phase. 1.0

70 60

(DI H2O) 0 μL 25 μL 50 μL 100 μL 200 μL 300 μL 500 μL 1000 μL

55 50 45

% count

40 35 30 25

(b)

0.9 0.8 0.7

dave [μm]

(a)

65

0.6 0.5 0.4

20 15

0.3

10

0.2

5 0 100

1000

0.1 0

100

dave [nm]

200

300

400

500

600

700

800

900 1000

VH O [μL] 2

Figure 3.10. Diameter distributions of hybrid microsphere samples at different dilutions, as calculated from SEM micrographs. Error bars indicate one standard deviation. Based on the data obtained, there is synthetic flexibility for tailoring the size of coacervates. Size control can be achieved by changing the ratio R, the temperature of synthesis, MW of the polyamine or by controlling the aging time of the coacervates.

Permeability/Porosity Study Results. Prior to further detailed confocal microscopy studies, the fluorescence emission data were collected for microsphere components

87

and microspheres at different processing stages, using 488 nm excitation (Fig. 3.8).

Fluorescence intensity

MNP w TSC coa w/ MNP coa no MNP cross-linked PLK

FITC-dextran MS + 40 μL FITC-dextran MS + 20 μL FITC-dextran MS w/ MNP

500

550

600

650

(b)

Fluorescence intensity

(a)

500

700

550

600

650

700

λ [nm]

λ [nm]

Figure 3.8. Fluorescence emission of microspheres at various stages. As control experiments, both hybrid coacervates and crosslinked PLK were analyzed (Fig 3.8a). Coacervate solutions with MNPs alone showed an emission band centered at 520 nm, which was also observed for a suspension of MNPs. Cross-linked magnetic microspheres showed an emission band around 550 nm, as did crosslinked PLK; therefore this band was assigned to emission from imide cross-links. Emission of magnetite entrapped within these cross-linked microspheres could not be resolved, perhaps due to exciton transfer to PLK. The emission of FITC-labeled dextran, centered at 520 nm (Fig. 3.8b), slightly blue-shifted to 510 nm at higher concentrations, because of interaction with the reaction medium. Importantly, this analysis indicates the proper wavelengths for differentiating signals of imide bonds and FITC: FITC detection was performed in the range 510-530

88

nm, while the cross-linked microspheres were detected in the range of 540-570 nm.63 The hydrodynamic size distribution of the various FITC-dextran molecules is presented in figure 3.9a; the average hydrodynamic size increased with MW, as expected (Fig. 3.9b). Thirteen samples of microspheres were prepared with cross-linking durations of 1, 2, 3, 5, 8, 12, 16, 24, 33 min, 1 hr, 3 hr, 24 hr, and 48 hr. The results are summarized in Table I under three categories: (i) permeable to FITCdextran, as observed by complete microsphere filling (Fig. 3.11c); (ii) impermeable to dextran, such that negatively charged FITC-dextran molecules are adsorbed only on the microsphere surfaces (Fig. 3.11d); and (iii) critically permeable, for sample sets in which both (i) and (ii) are observed. These observations show that porosity of these composite microspheres could be tailored by an adjustable and measurable crosslinking treatment.

89

40 4 kD a 1 0 kD a 2 0 kD a 4 0 kD a 7 0 kD a 1 5 0 kD a 2 5 0 kD a

35 30

V o lu m e [% ]

25 20 15 10 5 0 -5 1

10

100

S iz e [d .n m ]

(a)

(c)

20 18 16 14

d [n m ]

12 10 8 6 4 2 0

50

100

150

200

250

M w (F IT C -de xtra n) [kD a]

(b) (d) Figure 3.11. (a) Hydrodynamic size distribution of FITCdextran molecules with different MWs; (b) Average size of FITC-dextran molecules with different MWs; Confocal microscopy images of hybrid microspheres that are (c) permeable and (d) impermeable to FITC-dextran.

90

CL Duration t=1 min t=2 min t=3 min t=5 min t=8 min t=12 min t=16 min t=24 min t=33 min

4 kDa

10 kDa

FITC-dextran with different MWs 20 kDa 40 kDa 70 kDa 150 kDa

250 kDa

t=1 hr t=3 hr t=24 hr t=48 hr Permeable

Critical Permeability

Impermeable

Table 3.3. Permeability of hybrid microspheres, cross-linked (CL) for different durations, to FITC-dextran molecules with different MWs. The two shortest durations of cross-linking, 1 and 2 min., resulted in the largest pores and permitted entry of FITC-dextran molecules with an average hydrodynamic radius of 18 nm (250 kDa). Prolonging cross-linking for 3 and 5 minutes reduced the critical pore size and allowed the diffusion of 12 nm molecules. Eight minute crosslinking reduced the critical pore size to under 12 nm, and 12 min crosslinking further reduced it to around 8 nm. Extended periods of crosslinking (between 16 min-48 hrs) resulted in microspheres with similar porosities—all were permeable to 5 nm dextran. As expected, the longest duration of cross-linking in this series resulted in the smallest pore size. Importantly, pore sizes did not change significantly for samples cross-linked beyond 16 min., despite an increasing UV

91

absorption peak. This suggests that some GA molecules coupled to PLK did not function as intermolecular cross-linkers, despite Schiff base formation. Diffusion coefficients were measured by adding excess fluorescein-labeled

dextran

to

cross-linked

microspheres

and

performing fluorescence recovery after photobleaching (FRAP) according to established procedures.47, 64, 65 Diffusion of FITC-dextran molecules into the microspheres was monitored with time to generate a fluorescence recovery curve. The fluorescence recovery as a function of time, in a central planar cross-section of solid microspheres, is given exactly by (Fig. 3.12a):

⎛ ⎜ 8 Cexact (t ) = π R02C0 ⎜1 − 2 ⎜ π ⎜ ⎝



∑ n =1

e

⎛ (2 n −1)π ⎞ − tD ⎜ ⎟ ⎝ R0 ⎠

(2n − 1) 2

2

⎞ ⎟ ⎟ ⎟ ⎟ ⎠

(1)

where Cexact(t) and Co denote the fluorescence/concentration at times t and t = 0, respectively. D is the diffusion constant, and Ro is the radius of the microsphere.

92

1 1

0.8

(b)

(a)

0.8

0.6

0.6

0.4 0.2

C

0.4

0.2

t -1.5

5

10

15

20

25

-1

-0.5

r 0.5

1

1.5

30

Figure 3.12. (a) Characteristic recovery vs. time curve, for diffusion into a solid sphere cross-section. (b) Radial concentration profiles at equal time increments, using 20 summation terms. The equation was derived from the diffusion equation in spherical coordinates, using separation of time and space variables.

The r-

dependence was matched to a Bessel equation or order ½, and boundary conditions and initial conditions were used to describe an initially empty sphere within a medium that maintains constant concentration at the sphere surface (Fig. 3.12b).66

⎛ 2R C (r , t ) = C0 ⎜1 + 0 ⎜ πr ⎝

(−1) e ∑ n n =1 ∞

n

⎛ nπ ⎞ − tD ⎜ ⎟ ⎝ R0 ⎠

2

⎞ ⎛ nπ r ⎞ ⎟ sin ⎜ ⎟ ⎝ R0 ⎠ ⎟ ⎠

(2)

This full space-time (Eq. 2) equation was integrated to obtain (1). For times after 40% recovery, the expression can be approximated by using only the first term of the infinite summation (Fig. 3.13a): ⎛π ⎞ ⎛ − tD ⎜ ⎟ 8 R 2 C40% (t ) = π R0 C0 ⎜1 − 2 e ⎝ 0 ⎠ ⎜ π ⎝

93

2

⎞ ⎟ ⎟ ⎠

(3)

Finally, the effects of partially unbleached dye at t=0 and instrumental time delay between bleaching and measurements can be simultaneously corrected by fitting with an “unbleached” parameter u: ⎛π ⎞ ⎛ − tD ⎜ ⎟ 8 R C (t ) = π R02C0 ⎜1 − (1 − u ) 2 e ⎝ 0 ⎠ ⎜ π ⎝

2

⎞ ⎟ ⎟ ⎠

(4)

The fluorescence recovery curves for different samples of known crosssectional areas were fit to this equation, using least-squared minimization (Fig. 3.13b). This yielded the diffusion coefficients of FITC-dextran molecules into composite microspheres of different cross-linking durations.

110

0.4

(a)

(b) 100

0.3 90

0.2 80

t

0.1

t

20

0.5

1

1.5

40

60

80

100

120

140

2

Figure 3.13. (a) Convergence of recovery curves, using 1, 2, and 50 summation terms. Use of 1 term is accurate by ~40% recovery; 2 terms give accuracy at around ~20% recovery. (b) Representative fits to FRAP data, demonstrating experimental and theoretical agreement. The diffusion coefficients of FITC-dextran molecules of different MWs into cross-linked microspheres were on the order of 1015

m2/s (Fig. 3.14a), which were about four orders of magnitude smaller

than those of the free molecules in water. This decrease may be partly

94

attributed to altered diffusion in the coacervate phase, but it is mainly due to the cross-linking process.

Furthermore, a series of FRAP

experiments were performed on microspheres cross-linked for various durations, using 10 kDa FITC-dextran molecules. Diffusion coefficients calculated for FITC-dextran molecules within these microspheres (Fig. 3.14b) showed a trend of decreasing D with greater cross-linking. Hence, cross-links reduced pore-sizes and restricted free movement of FITC-dextran molecules. These results suggest that diffusion through, and hence release from, the nanocomposite network can be controlled by macromolecule size relative to microsphere pore size.

1E-10

10 kDa FITC-Dextran

4.5

in 8min cross-linked μspheres in H2O

4.0

3.0 2

D [m /s]

D x 10

-15

2

[m /s]

3.5

2.5 2.0 1.5

(a)

1.0 0

5

(b)

1E-15

10

15

20

25

30

2

35

3

4

5

6

7

8

9

10

11

ddextran [nm]

tcross-link [min]

Figure 3.14. Diffusion coefficients calculated from FRAP experiments for (a) 10 kDa FITC-dextran diffusing into microspheres cross-linked for different durations, and (b) Dextran molecules with different MWs diffusing into 8 min cross-linked microspheres.

95

Conclusions This work demonstrates the first example of coacervation that uses magnetic nanoparticles as assembling components, exemplified here as PLK/TSC microspheres. One-step formation of these MNPbased microspheres can be considered as a nanoparticle self-assembly process induced by the presence of polyelectrolytes.

Solid

microspheres can be rapidly generated at ambient temperature, in water, and without the need of a solid template. The described procedure is straightforwardly applicable to most nanoparticles that bear the proper surface functionality. The resulting objects can be tailored by modifying the conditions during and/or after the assembly step, they retain their magnetic properties. Glutaraldehyde cross-linking proved effective in mechanical stabilization of liquid coacervates. The hybrid sphere size was controlled by adjusting the ratio R, temperature, aging time, the molecular weight of the polymer, ionic strength, and the amount of dilution in the synthesis medium. In addition to providing mechanical stability, cross-linking further permitted the tailoring of porosity, the extent of which was monitored

by

UV-Vis

spectroscopy.

Cross-linking

decreased

microsphere permeabilities and resulted in smaller critical pore sizes.

96

FRAP experiments, performed using FITC-dextran molecules of various MW, revealed that cross-linking could control microsphere permeabilities, and that the dextran diffusion coefficients decreased by four orders of magnitude. These results are of critical importance for the design of functional microspheres that can control diffusion based on size selection, or for entrapment of molecules of a chosen size while allowing free transfer of smaller molecules. Based on the microsphere porosities as controlled by glutaraldehyde cross-linking, it is concluded that this design is appropriate for encapsulating agents larger than 5nm (such as some protein formulations or genes).

For smaller drugs, additional

modifications are necessary; for instance the drugs may be covalently bound to the coacervate interior, or else the outermost pores must be blocked by ‘plugging’ with larger molecules or by coating with additional, probably inorganic, material.

Alternatively, other cross-

linking methods might be sought, such as amide bond formation catalyzed by EDC. This method would be expected to further restrict pore size because 1) two coacervating components rather than one contribute to cross-linking and 2) all cross-links would necessarily be intermolecular. Although PLK and MNPs with citrate were featured here as a case study, the design principles and synthetic methods are applicable

97

to a variety of polyamines and nanoparticles that are appropriately surface-functionalized. The incorporation of magnetic nanoparticles into this drug delivery device design provides opportunities for MRI visualization, in vivo magnetic steering, and RF-triggered hyperthermia. The effectiveness of each of these will require further studies, as will the evaluation of a hyperthermic, triggered release mechanism in vitro. In vitro cell separation studies on antibody-coated devices are needed to prove targeting capabilities. In vitro cell viability studies have already been performed, and suggest that the hybrid microspheres have low toxicity.67 Finally, in vivo studies will be required to identify any additional shortcomings of such devices in the complex and unpredictable settings of animal and human bodies. Particulate-based drug delivery systems usually face the problem of short circulation time due to removal by the liver. Even with modern technology’s best developed antibody formulations and delivery methods, all particulatebased devices currently suffer from their inability to infiltrate more than one-third of the targeted cells. Since cancer requires eradication of 100% of cancer cells in order to avoid progression to metathesis, new physical and biological breakthroughs in delivery methods are required to make targeted drug delivery commercially realistic. Meanwhile, a prospective new material has now been developed and lies in wait of

98

improved targeting strategies. The widespread applicability of these nanocomposites as magnetically functionalized drug delivery devices or Magnetic Resonance Imaging contrast agents is supported by their facile assembly, tunability of properties, encapsulation of dextran, and low toxicity.

REFERENCES

1.

Suit, H. D.; Shwayder, M., Hyperthermia - Potential as an

Antitumor Agent. Cancer 1974, 34, (1), 122-129. 2.

Yatvin, M. B.; Weinstein, J. N.; Dennis, W. H.; Blumenthal, R.,

Design of Liposomes for Enhanced Local Release of Drugs by Hyperthermia. Science 1978, 202, (4374), 1290-1293. 3.

Leduc, P. R.; Wong, M. S.; Ferreira, P. M.; Groff, R. E.;

Haslinger, K.; Koonce, M. P.; Lee, W. Y.; Love, J. C.; McCammon, J. A.; Monteiro-Riviere, N. A.; Rotello, V. M.; Rubloff, G. W.; Westervelt, R.; Yoda, M., Towards an in vivo biologically inspired nanofactory. Nature Nanotechnology 2007, 2, (1), 3-7. 4.

Moghimi, S. M.; Hunter, A. C.; Murray, J. C., Nanomedicine:

current status and future prospects. Faseb Journal 2005, 19, (3), 311330. 5.

Rezler, E. M.; Khan, D. R.; Lauer-Fields, J.; Cudic, M.;

Baronas-Lowell, D.; Fields, G. B., Targeted drug delivery utilizing

99

protein-like molecular architecture. Journal of the American Chemical Society 2007, 129, (16), 4961-4972. 6.

Torchilin, V. P., Multifunctional nanocarriers. Advanced Drug

Delivery Reviews 2006, 58, (14), 1532-1555. 7.

Vasir, J. K.; Reddy, M. K.; Labhasetwar, V. D., Nanosystems in

drug targeting: Opportunities and challenges. Current Nanoscience 2005, 1, (1), 47-64. 8.

Wang, M. D.; Shin, D. M.; Simons, J. W.; Nie, S. M.,

Nanotechnology for targeted cancer therapy. Expert Review of Anticancer Therapy 2007, 7, (6), 833-837. 9.

Stolnik, S.; Illum, L.; Davis, S. S., Long Circulating

Microparticulate Drug Carriers. Advanced Drug Delivery Reviews 1995, 16, (2-3), 195-214. 10.

Champion, J. A.; Mitragotri, S., Role of target geometry in

phagocytosis. Proceedings of the National Academy of Sciences of the United States of America 2006, 103, (13), 4930-4934. 11.

Euliss, L. E.; DuPont, J. A.; Gratton, S.; DeSimone, J.,

Imparting size, shape, and composition control of materials for nanomedicine. Chemical Society Reviews 2006, 35, (11), 1095-1104. 12.

Rolland, J. P.; Maynor, B. W.; Euliss, L. E.; Exner, A. E.;

Denison, G. M.; DeSimone, J. M., Direct fabrication and harvesting of

100

monodisperse, shape-specific nanobiomaterials. Journal of the American Chemical Society 2005, 127, (28), 10096-10100. 13.

Vonarbourg, A.; Passirani, C.; Saulnier, P.; Benoit, J. P.,

Parameters influencing the stealthiness of colloidal drug delivery systems. Biomaterials 2006, 27, (24), 4356-4373. 14.

Heath, T. D.; Fraley, R. T.; Papahadjopoulos, D., Antibody

Targeting of Liposomes - Cell Specificity Obtained by Conjugation of F(Ab')2 to Vesicle Surface. Science 1980, 210, (4469), 539-541. 15.

Bae, Y.; Nishiyama, N.; Fukushima, S.; Koyama, H.; Yasuhiro,

M.; Kataoka, K., Preparation and biological characterization of polymeric micelle drug carriers with intracellular pH-triggered drug release property: Tumor permeability, controlled subcellular drug distribution, and enhanced in vivo antitumor efficacy. Bioconjugate Chemistry 2005, 16, (1), 122-130. 16.

Bae, Y.; Fukushima, S.; Harada, A.; Kataoka, K., Design of

environment-sensitive supramolecular assemblies for intracellular drug delivery: Polymeric micelles that are responsive to intracellular pH change. Angewandte Chemie-International Edition 2003, 42, (38), 4640-4643. 17.

Guo, X.; Szoka, F. C., Chemical approaches to triggerable lipid

vesicles for drug and gene delivery. Accounts of Chemical Research 2003, 36, (5), 335-341.

101

18.

Gatenby, R. A.; Gillies, R. J., Why do cancers have high aerobic

glycolysis? Nature Reviews Cancer 2004, 4, (11), 891-899. 19.

Nishiyama, N.; Kataoka, K., Current state, achievements, and

future prospects of polymeric micelles as nanocarriers for drug and gene delivery. Pharmacology & Therapeutics 2006, 112, (3), 630-648. 20.

Nishiyama, N.; Kataoka, K., Nanostructured devices based on

block copolymer assemblies for drug delivery: Designing structures for enhanced drug function. In Polymer Therapeutics Ii: Polymers as Drugs, Conjugates and Gene Delivery Systems, 2006; Vol. 193, pp 67101. 21.

Shchukin, D. G.; Sukhorukov, G. B.; Mohwald, H., Smart

inorganic/organic nanocomposite hollow microcapsules. Angewandte Chemie-International Edition 2003, 42, (37), 4472-4475. 22.

Caruso, F.; Caruso, R. A.; Mohwald, H., Production of hollow

microspheres from nanostructured composite particles. Chemistry of Materials 1999, 11, (11), 3309-3314. 23.

Huang, Z.; Szoka, F. C., Bioresponsive Liposomes and Their

Use for Macromolecular Delivery. In Liposome Technology, Third ed.; Gregoriadis, G., Ed. Informa Healthcare: 2006; pp 165-196. 24.

Bungenberg de Jong, H. G., In Crystallisation- coacervation-

flocculation in colloid science, Ed. Kruyt, H. R., Ed. Elsevier: Amsterdam, 1949; Vol. II, pp 232-258.

102

25.

Cha, J. N.; Birkedal, H.; Euliss, L. E.; Bartl, M. H.; Wong, M.

S.; Deming, T. J.; Stucky, G. D., Spontaneous formation of nanoparticle vesicles from homopolymer polyelectrolytes. Journal of the American Chemical Society 2003, 125, (27), 8285-8289. 26.

Wong, M. S.; Cha, J. N.; Choi, K. S.; Deming, T. J.; Stucky, G.

D., Assembly of nanoparticles into hollow spheres using block copolypeptides. Nano Letters 2002, 2, (6), 583-587. 27.

Cha, J. N.; Bartl, M. H.; Wong, M. S.; Popitsch, A.; Deming, T.

J.; Stucky, G. D., Microcavity lasing from block peptide hierarchically assembled quantum dot spherical resonators. Nano Letters 2003, 3, (7), 907-911. 28.

Euliss, L. E.; Grancharov, S. G.; O'Brien, S.; Deming, T. J.;

Stucky, G. D.; Murray, C. B.; Held, G. A., Cooperative assembly of magnetic nanoparticles and block copolypeptides in aqueous media. Nano Letters 2003, 3, (11), 1489-1493. 29.

Rana, R. K.; Murthy, V. S.; Yu, J.; Wong, M. S., Nanoparticle

self-assembly of hierarchically ordered microcapsule structures. Advanced Materials 2005, 17, (9), 1145-+. 30.

Luzzi, L. A.; Gerraughty, R. J., Effects of Selected Variables on

Extractability of Oils from Coacervate Capsules. Journal of Pharmaceutical Sciences 1964, 53, (4), 429-&.

103

31.

Weinbreck, F.; Minor, M.; De Kruif, C. G., Microencapsulation

of oils using whey protein/gum arabic coacervates. Journal of Microencapsulation 2004, 21, (6), 667-679. 32.

Green, B. K., and Schleicher, Lowell Oil-containing

Microscopic Capsules and Method of Making Them. 2,800,457, 1957. 33.

Promislow, J. H. E.; Gast, A. P., Low-energy suspension

structure of a magnetorheological fluid. Physical Review E 1997, 56, (1), 642-651. 34.

Wirtz, D.; Fermigier, M., One-Dimensional Patterns and

Wavelength Selection in Magnetic Fluids. Physical Review Letters 1994, 72, (14), 2294-2297. 35.

Tartaj, P.; Morales, M. D.; Veintemillas-Verdaguer, S.;

Gonzalez-Carreno, T.; Serna, C. J., The preparation of magnetic nanoparticles for applications in biomedicine. Journal of Physics DApplied Physics 2003, 36, (13), R182-R197. 36.

Hafeli, U.; Pauer, G.; Failing, S.; Tapolsky, G., Radiolabeling of

magnetic particles with rhenium-188 for cancer therapy. Journal of Magnetism and Magnetic Materials 2001, 225, (1-2), 73-78. 37.

Johannsen, M.; Gneveckow, U.; Eckelt, L.; Feussner, A.;

Waldofner, N.; Scholz, R.; Deger, S.; Wust, P.; Loening, S. A.; Jordan, A., Clinical hyperthermia of prostate cancer using magnetic

104

nanoparticles: Presentation of a new interstitial technique. International Journal of Hyperthermia 2005, 21, (7), 637-647. 38.

Hogemann, D.; Josephson, L.; Weissleder, R.; Basilion, J. P.,

Improvement of MRI probes to allow efficient detection of gene expression. Bioconjugate Chemistry 2000, 11, (6), 941-946. 39.

Mornet, S.; Vasseur, S.; Grasset, F.; Duguet, E., Magnetic

nanoparticle design for medical diagnosis and therapy. Journal of Materials Chemistry 2004, 14, (14), 2161-2175. 40.

Neuberger, T.; Schopf, B.; Hofmann, H.; Hofmann, M.; von

Rechenberg, B., Superparamagnetic nanoparticles for biomedical applications: Possibilities and limitations of a new drug delivery system. Journal of Magnetism and Magnetic Materials 2005, 293, (1), 483-496. 41.

Bergemann, C.; Muller-Schulte, D.; Oster, J.; a Brassard, L.;

Lubbe, A. S., Magnetic ion-exchange nano- and microparticles for medical, biochemical and molecular biological applications. Journal of Magnetism and Magnetic Materials 1999, 194, (1-3), 45-52. 42.

Baselt, D. R.; Lee, G. U.; Natesan, M.; Metzger, S. W.;

Sheehan, P. E.; Colton, R. J., A biosensor based on magnetoresistance technology. Biosensors & Bioelectronics 1998, 13, (7-8), 731-739. 43.

Ugelstad, J.; Berge, A.; Ellingsen, T.; Schmid, R.; Nilsen, T. N.;

Mork, P. C.; Stenstad, P.; Hornes, E.; Olsvik, O., Preparation and

105

Application of New Monosized Polymer Particles. Progress in Polymer Science 1992, 17, (1), 87-161. 44.

Chemla, Y. R.; Crossman, H. L.; Poon, Y.; McDermott, R.;

Stevens, R.; Alper, M. D.; Clarke, J., Ultrasensitive magnetic biosensor for homogeneous immunoassay. Proceedings of the National Academy of Sciences of the United States of America 2000, 97, (26), 1426814272. 45.

Jordan, A.; Scholz, R.; Wust, P.; Fahling, H.; Felix, R.,

Magnetic fluid hyperthermia (MFH): Cancer treatment with AC magnetic field induced excitation of biocompatible superparamagnetic nanoparticles. Journal of Magnetism and Magnetic Materials 1999, 201, 413-419. 46.

Pankhurst, Q. A.; Connolly, J.; Jones, S. K.; Dobson, J.,

Applications of magnetic nanoparticles in biomedicine. Journal of Physics D-Applied Physics 2003, 36, (13), R167-R181. 47.

Tong, W. J.; Gao, C. Y.; Mohwald, H., Manipulating the

properties of polyelectrolyte microcapsules by glutaraldehyde crosslinking. Chemistry of Materials 2005, 17, (18), 4610-4616. 48.

Lee, D.; Rubner, M. F.; Cohen, R. E., Formation of

nanoparticle-loaded microcapsules based on hydrogen-bonded multilayers. Chemistry of Materials 2005, 17, (5), 1099-1105.

106

49.

Malachowski, L.; Holcombe, J. A., Immobilized poly-L-

histidine for chelation of metal cations and metal oxyanions. Analytica Chimica Acta 2003, 495, (1-2), 151-163. 50.

McKenna, B. J., Birkedal, H., Bartl, M. H., Deming, T. J.,

Stucky, G. D. In Self-Assembling Microspheres from Charged Functional Polyelectrolytes and Small-Molecule Counterions, Mater. Res. Soc. Symp. Proc., 2004; 2004; pp 4.12.1-6. 51.

McKenna, B. J.; Birkedal, H.; Bartl, M. H.; Deming, T. J.;

Stucky, G. D., Micrometer-sized spherical assemblies of polypeptides and small molecules by acid-base chemistry. Angewandte ChemieInternational Edition 2004, 43, (42), 5652-5655. 52.

Wang, S. B.; Xu, F. H.; He, H. S.; Weng, L. J., Novel alginate-

poly(L-histidine) microcapsules as drug carriers: In vitro protein release and short term stability. Macromolecular Bioscience 2005, 5, (5), 408414. 53.

Tourinho, F. A.; Franck, R.; Massart, R., Aqueous Ferrofluids

Based on Manganese and Cobalt Ferrites. Journal of Materials Science 1990, 25, (7), 3249-3254. 54.

Deng, Y. H.; Yang, W. L.; Wang, C. C.; Fu, S. K., A novel

approach for preparation of thermoresponsive polymer magnetic microspheres with core-shell structure. Advanced Materials 2003, 15, (20), 1729-+.

107

55.

Walt, D. R.; Agayn, V. I., The Chemistry of Enzyme and

Protein Immobilization with Glutaraldehyde. Trac-Trends in Analytical Chemistry 1994, 13, (10), 425-430. 56.

Atkins, R. C., Colorimetric Determination of Iron in Vitamin

Supplement Tablets - General Chemistry Experiment. Journal of Chemical Education 1975, 52, (8), 550-550. 57.

Chantrell, R. W.; Popplewell, J.; Charles, S. W., Measurements

of Particle-Size Distribution Parameters in Ferrofluids. Ieee Transactions on Magnetics 1978, 14, (5), 975-977. 58.

Brabers, V. A. M., Infrared Spectra of Cubic and Tetragonal

Manganese Ferrites. Physica Status Solidi 1969, 33, (2), 563-&. 59.

Ahn, Y.; Choi, E. J.; Kim, S.; Ok, H. N., Magnetization and

Mossbauer study of cobalt ferrite particles from nanophase cobalt iron carbonate. Materials Letters 2001, 50, (1), 47-52. 60.

Murthy, V. S.; Cha, J. N.; Stucky, G. D.; Wong, M. S., Charge-

driven flocculation of poly(L-lysine)-gold nanoparticle assemblies leading to hollow microspheres. Journal of the American Chemical Society 2004, 126, (16), 5292-5299. 61.

Spicer, P. T.; Pratsinis, S. E., Shear-induced flocculation: The

evolution of floc structure and the shape of the size distribution at steady state. Water Research 1996, 30, (5), 1049-1056.

108

62.

Maguire, M. E.; Cowan, J. A., Magnesium chemistry and

biochemistry. Biometals 2002, 15, (3), 203-210. 63.

Toprak, M. S., McKenna, B. J., Waite, J. H., Stucky, G. D. In

Tailoring Magnetic Microspheres with Controlled Porosity, Mater. Res. Soc. Symp. Proc., 2007; Materials Research Society: 2007; pp 0969-W03-11. 64.

Ibarz, G.; Dahne, L.; Donath, E.; Mohwald, H., Controlled

permeability of polyelectrolyte capsules via defined annealing. Chemistry of Materials 2002, 14, (10), 4059-4062. 65.

Ibarz, G.; Dahne, L.; Donath, E.; Mohwald, H., Resealing of

polyelectrolyte capsules after core removal. Macromolecular Rapid Communications 2002, 23, (8), 474-478. 66.

Crank, J., Diffusion in a Sphere. In The Mathemetics of

Diffusion Second ed.; Oxford University Press Inc.: New York, 1999; pp 89-103. 67.

Toprak, M. S.; McKenna, B. J.; Mikhaylova, M.; Waite, J. H.;

Stucky, G. D., Spontaneous assembly of magnetic microspheres. Advanced Materials 2007, 19, (10), 1362-+.

109

Chapter IV

Calcium Carbonate Mineralization via Complex Coacervation

110

Abstract: This chapter marks a transition in research focus from coacervate-based,

microspherical,

“core-shell”

devices

to

the

employment of coacervates for directing the higher-order growth of inorganic structures.

Using calcium carbonate as a model inorganic

component, coacervates were first used as templates to control the mineralization into spheres. The formation of mineral shells further expanded the framework for device design, and the formation of solid spheres presented a facile route to an industrially important product. However, coupled with interaction at substrate surfaces, the mineralizing coacervates were discovered to guide rod-like growths or cones. Growth of rods occurred using pre-existing coacervate funnels with carbonate vapor infusion, whereas cones were produced by first creating a condition of mutual inhibition by direct addition of carbonate salt.

111

Thus far, two primary kinds of interaction have been demonstrated between coacervates and inorganic material:1-6 1) Inorganic material can comprise one of the assembling components of coacervation (various nanoparticles).7, 8 2) Coacervates can behave as templates for the surface deposition of inorganic material (silica shells). In further research, a third type of interaction was considered: 3) The use of the entire coacervate volume as a template for mineralization. Using calcium carbonate mineralization as a model inorganic component, the coacervates used this way will be shown to result in three kinds of phenomena: a) Solidification of mineral preferentially at the coacervate surface (as before) b) Petrification of the entire internal coacervate volume into solid microspheres c) Coalescence of solidifying spheres into larger structures, as controlled by forces in solution d) Mineralization at the coacervate/substrate interface, whereby solidifying coacervates behave as conduits for the mineral components. The interaction was found to depend primarily on three factors: the solution pH (which controls both mineralization and coacervation), the component concentrations, and the order of addition of the various components. Control over mineral morphology has both industrial and fundamental importance. Industrially, calcium carbonate microspheres

112

are employed as fillers for paint and as basic whitening elements in paper manufacturing.

The properties of these materials are partly

determined by the qualities of the CaCO3 particles, most notably size and shape. Therefore, there have been efforts to understand and control the formation of spherical CaCO3 particles,9-18 or generally, to control mineral morphology into a variety of complex structures.19 As will be elaborated, this also has great importance for understanding biomineralization, as morphological control of mineral is inextricably linked to molecular, crystal controlling processes and emergent bulk, structural organization.

Templating Results The formation of CaCO3 shells was sought first, and this has previously been achieved in different ways. For instance, Gower et al report the use of their PILP process,20 combined with the action of an O/W emulsion to produce oil-core microcapsules of CaCO3.15 Xu et al have reported the use of phytic acid in a similar vapor-induced process to form shells of CaCO3.12 However, these previous reports did not consider the possibility of coacervation when in fact, both types of precursor components can couple with Ca2+ at the proper pH to form complex coacervates (Fig. 4.1). Phytic acid has a particularly sharp

113

transition region from dissolution to coacervation to flocculation, occurring between pH 4 and 5.

(A)

(B)

Figure 4.1. Optical micrographs of pH/Ca2+-induced coacervates of (A) PAA2kDa at pH 10 and (B) phytic acid at pH 4.5.

Conditions for the production of CaCO3 shells were primarily sought with PAA (MW 15k) and PLD (MW 33k), which have sufficient molecular weights to associate efficiently with Ca2+. From solutions of 10mM (monomer), optimal coacervation conditions were sought as functions of pH and [Ca2+].

Values of pH at and above 7

were sufficient for coacervation with these molecular weights, and between pH 9-10 the coacervates partially solidified, aggregated, and sedimented. This may be due to enhanced charge interaction with complete ionization of the polyanion, but is probably also due to partial solidification and neutralization due to the formation of calcium hydroxides in solution (Ksp = 5 x 10-6).

A pH above the

bicarbonate/carbonate pKa of 8.3 is desirable, however, so that

114

mineralization occurs faster than HCO3- diffusion to the coacervate interior. Therefore, pHs of 8.5-9.0 were chosen for initial coacervate template solutions. Selection of the proper [Ca2+] was important for localizing mineralization. At the lowest levels, coacervation is too weak to form templates. Slightly elevating these conditions (as to charge neutrality, [COO-] = 2 [Ca2+]) may yield some coacervates but the relatively low [Ca2+] still impedes mineralization and especially hampers its occurrence at the template surface. In contrast, the highest levels of [Ca2+] unfavorably raise levels in the equilibrium solution, such that mineral occurs non-specifically, away from coacervate surfaces. Careful optimization was therefore required to mineralize specifically and produce high yields of CaCO3 shells. A better understanding of [Ca2+] conditions that select for CaCO3 shells was sought for more general application to different polymer concentrations and different polymer types. The extent of coacervation was investigated for this purpose by using turbidimetry. Solutions of constant [polymer] and varying [Ca2+] were measured by UV-vis spectrometry by adding incremental volumes of each component to initial solutions containing only polymer, and the each sample’s absorbance was recorded, using averaged values between 400500nm (Fig. 4.2). The resulting plots demonstrate several expected

115

trends. In the first leg, the low amount of scattering indicates that little or no coacervation occurs, because too few Ca2+ cross-links are available for polymers association. Next, there is a rapid “S-curve”like increase in turbidity, indicating a critical range for coacervation. This reaches a maximum, indicating optimal coacervation for the given polymer concentration. After this point, turbidity decreases because extra CaCl2 salt increases the ionic strength of solution (A)

1.8 1.6

2.0

Absorbance

1.4 1.2

Absorbance

(B)

2.5

.8mMD 4mMD 9mMD

1.0 0.8 0.6

1.5 1.0 0.5

0.4

0.0

0.2

0

0.0 0

20

40

60

80

100

20

40

60

80

100

2+

[Ca ] (mM)

120

[Ca] (mM)

Figure 4.2. Turbidity curves of (A) PLD and (B) PAA with increasing Ca2+ concentrations. For solutions of 5mM and 10mM monomer of both PAA and PLD, it was found that the critical zone for the structure directing capacity of coacervates occurred between the midway point of the Scurve turbidity increase and the maximum turbidity point. In the case of PLD, there was little variation in the optimal [Ca2+] for different polymer concentration; maximum turbidity occurred with ~45mM Ca2+ at [D]’s of 0.8 mM, 3.6 mM, and [10]mM. For 10mM PAA, the maximum of 10.5 mM Ca2+ was used, and the final procedure was described in Chapter 1.

116

(A)

(B)

(C)

Figure 4.3. PAA/Ca2+ coacervates mineralized by carbonate addition. (A) LOM, (B) SEM, and (C) TEM indicating a less dense interior. Finally, the carbonate solution was considered. NaHCO3 at pH 9-9.5 was used in concentrations and pHs that minimized the salt inhibition effect. The maintenance of proper pH was also important because CaCO3 mineralization releases H+, which inhibits coacervation and mineralization. Hence, acidification was prevented by multiple carbonate and NaOH additions. The final procedure was described in Chapter 1. Micrographs are shown in figure 4.3.

50

(A)

350

(B)

100

300

40

90 30

%Mass

80

200

20 70

150

mW/mg

Intensity

250

10 60

100

0 50

50

-10

10

20

30

40

50

60

0

70

100

200

300

400

500

Temperature (deg. C)

2 Theta

Figure 4.4. (A) XRD of hollow spheres indicating amorphous mineral. (B) TGA/DSC scan during calcination of hollow spheres. The resulting shell material was characterized by XRD (Fig. 4a) and TGA/DSC (Fig. 4.4b). Interestingly, the shells appear completely

117

amorphous.

The organic component was determined to be 48%.

Calcined shells demonstrated hollow interiors (Fig. 4.5) and diffracted as calcite, as expected.

(A)

(B)

(C) Figure 4.5. SEM images of calcium carbonate shells following calcination. Internally mineralized coacervates could be prepared in several ways.

Regarding the above procedure for shell formation, it was

apparent from optical microscopy that the “shells” could extend to the spheres’ centers when mineralization was too slow (data not shown). In other words, petrified spheres could be produced with lower [Ca2+], at somewhat lower pHs, or with less carbonate. Alternatively, solid spheres could be obtained by adding carbonate first and calcium last (Fig. 4.6). In a typical preparation, 20 uL of PLD (5 mg/mL) was mixed with 3 μL NaHCO3 (20mM), and to this was added 7.5 μL CaCl2 (180 mM). Small volumes were used for rapid mixing.

The primary

further obstacle in this procedure is the aggregation of microspheres as they

become

charge

neutralized

118

and

sediment.

Preventing

sedimentation by intermittent redispersion can sufficiently maintain segregated colloids.

25 μm Figure 4.6. Internally mineralized spheres prepared by adding Ca2+ last.

Vapor Infusion Results A

peculiar

demonstration

of

slow

intra-coacervate

mineralization occurred when the ammonium carbonate vapor technique was applied to pre-mixed coacervate systems.21 As with shell formation, the resulting products varied as a function of [Ca2+], but the morphological variation was more complex. Using constant 10mM PLD (33k MW), [Ca2+] of 30+mM resulted in microspheres of varying sizes and aggregation degrees. Above 50mM [Ca2+], spheres were large (2-5μm) and highly aggregated (Fig. 4.7A). Many of these larger formations appear highly clustered at the solution/air interface

119

where pH is highest, as a side product of the mineralization that is often unreported. Closer to 40mM, the spheres are smaller (0.5-2 μm) and more distinct (Fig. 4.7B). Again, sedimentation increases aggregation, but dispersion is prolonged if they are disturbed at least every hour.

(A)

(B)

10 μm

(C)

10 μm

30 μm

Figure 4.7. SEM images of spherical CaCO3 aggregates from during vapor-treated PLD coacervates using (A) 50mM Ca2+ and (B) 40mM Ca2+. (C) Worm-like shapes produced from coalescence of spheres using 20mM Ca2+. However, at [Ca2+] under 30mM, the mineralization rate was slow enough for more complex morphologies to emerge (Fig. 4.6C). In addition to the usual microspheres, segmented worm-like growths appeared.

Slightly lower [Ca2+] induced similar and presumably

mechanistically related, tapered spicule-like rods. Between 10-20mM Ca2+, another, distinct “showerhead” morphology grew (Fig. 4.8). At early growth stages, these appeared as broad “sunflowers”. All three of these morphologies appeared to be nucleated by a larger sphere, which in the case of the showerheads was a broader hemisphere.

120

(A)

(B)

(C)

(D)

Figure 4.8. Sunflower (A & B) and showerhead (C & D) morphologies produced by vapor infusion of PLD coacervates using 15-20 mM Ca2+. Scale bars represent 25 μm. Efforts were made to observe the growth mechanism of these new structures. A crystallization cell was specially constructed so that the mineralization process could be observed in the optical microscope. These cells were placed vertically in a 50mL centrifuge tube containing NH4HCO3 as a vapor source. The cell was removed at various times, carefully turned horizontally, and observed under an optical microscope to determine crystal growth orientation. Synthetic adjustments were made to accommodate the new container, and it was confirmed that showerhead structures grow in the expected direction, with the flat end of the domes facing the substrate (Fig. 4.9). Observations of spicules

121

were less conclusive, but they appeared to grow preferentially on the glass, and roughly parallel to the surface.

Figure 4.9. Optical micrograph of vapor-induced growths viewed directly on their growth substrates. Previous reports by Gower et al have demonstrated that markedly similar systems produced rods by a Solution-Liquid-Solid (SLS) mechanism akin to the VLS mechanism used in inorganic syntheses such as fibers SiC. This is thought to occur by the action of “Polymer-Induced

Liquid

Precipitates”

(PILPs)

that

facilitate

mineralization at a growing surface.20 Because of the similarity of the mineralization systems and the similarity of structures, it is suggested that the spicule and showerhead morphologies form by a similar mechanism. Preferential growth of showerheads at the dome periphery is assumed to be due to increased surface tension at those locations; alternatively, free carbonate from solution may be responsible for the extensions, also explaining restriction to peripheral growth.

The

occurrence of these structures has significance in determining the

122

nature of the PILPs, as these experiments demonstrate that the “liquid precursors” may in fact be coacervates, but which contain levels of carbonate/bicarbonate above the supersaturation of CaCO3. The sensitivity of the vapor infusion method was tested with respect to several variables, including initial concentrations (Ca2+ and polymer as described), ammonium carbonate powder mass, solution height, and the extent of solution capping. All were found to have significant effect on the final morphology, which reflects the complexity of the system process (Scheme 4.1). To generalize, high concentrations, large powder content, and less capping (greater vapor accessibility)

promoted

conditions

homogeneous mineralization.

that

favored

more

rapid,

And at their other extremes, these

variables could inhibit mineralization, so each required careful engineering to achieve heterogeneous, coacervate-directed mineral growth mode.

123

Temperature

CO32mass Vapor

Gas solubility

pH

Polymer

Ca Salt Mineral supersaturation

Coacervation Location/ Morphology Coacervate size

Solution height Substrate

Surface Interaction

Mixing order

Scheme 4.1. Complexity of the vapor infusion crystallization process. Conflicting temperature effects are partially ameliorated by direct addition. The vapor infusion process was particularly sensitive to solution height, which when either too low or too high could cause homogeneous nucleation. This suggests an even more complicated process, which may be best explained by the emergence of a pH gradient (Fig. 4.10, green arrow in Scheme 4.1), as dictated by vapor infusion and solution height.

The higher pHs near the

Figure 4.10. An ammonium bicarbonate solution containing thymol blue dye, demonstrating a pH gradient after 24h.

air/water surface promote partial mineralization of spheres. As they

124

sediment, they may solidify, dissolve, or remain liquid-like. Appropriate pH gradients, then, initiate coacervate/substrate-mediated growth. Too-high pHs (low solution height, rapid infusion), induce rapid solidification.

Too uniform pHs, or shallow gradients (large

solution height, slow diffusion) promote homogeneous mineralization before substrate contact.

In this model, appropriate conditions for

complex shape formation are defined by pH gradient that 1) exists over a short enough distance for sedimentation to occur and 2) is sufficiently steep for the acidity near the substrate to induce a liquid phase separation. A parallel study was conducted using PAA rather than PLD. However, because the PAA/Ca2+ (Ksp~10-7) coacervates are less soluble than those of PLD/Ca2+ (Ksp~10-9), as reflected in the turbidity curves, lower levels of Ca2+ were required with PAA to decrease the nonspecific mineralization.

The results with PAA were markedly

different, however, resulting in strange “multipod” legs or the production of umbrella-shaped “cones,” depending on various experimental conditions (Fig. 4.11).

125

Figure 4.11. (A) Conical and (B) multipedal morphologies produced from vapor-infused coacervates using PAA15kDa rather than PLD. The PAA cones were studied further with two modifications to procedure: direct carbonate salt solution addition, and addition of calcium last.

The vapor infusion method was abandoned because

previously turbid coacervate solutions were found to be transparent after several hours (12h) of infusion, and the effect could be replicated by direct carbonate addition, if the carbonate concentration was >150mM. In great contrast, lower levels of carbonate were found to actually increase turbidity and induce flocculation. Furthermore, the same conical shapes could be induced by this method with sufficiently high carbonate concentration, of around 350-1000mM, which implies that the vapor method also introduces these high concentrations. Ca2+ was added last to remove the possibility that either coacervates or CaCO3 seeds would form first and critically affect crystallization behavior.

Both of these modifications do much to simplify the

dynamics of the crystallization system. Analysis of the number of

126

independent and dependent variables in the first situation reveals a complicated framework with much uncertainty (Scheme 4.1).

For

instance, temperature affects carbonate sublimation, carbon dioxide solubility, and supersaturation.

Using direct addition rather than

carbonate vapor provided much greater control of the process and experimental reproducibility. The loss of turbidity in mineralization solutions may at first be surprising,

because

ordinarily

solutions

of

coacervates

are

(immediately) turbid and solutions of CaCO3 also become quite turbid within ~1min. Therefore both mineralization and coacervation must be inhibited under these conditions.

The inhibitory power of

polyelectrolytes in preventing mineralization or “scaling” is well known, however. And the eradication of coacervation is best described as a result of the salt effect: NH4HCO3 initially behaves as an inactive salt of monovalent ions, which are known to strongly inhibit coacervation. Only with CO2 escape and pH increase do bicarbonate ions populate significantly enough to drive mineralization. Indeed, the mineralization does not generally occur in solutions before a couple hours, or after the pH appreciably increases with respect to the 8.3 pKa of bicarbonate/carbonate. Direct mixing is a less commonly reported method in the literature, except in cases where stopped flow is used to quickly introduce higher supersaturation.22 This is probably because of

127

the assumption that such highly supersaturated solutions are too unstable to achieve well controlled crystallization from an equilibrium solution. The Kitano method is another way of direct mixing with lower pHs; however, the procedure is more complex and is not usually employed after the addition of additives.23 Therefore, the resulting supersaturations and ionic strengths are significantly lower from those using the standard direct addition employed here. The results of direct addition are presented in the following Chapter. Briefly, it was found that both PAA and PLD can induce the same conical morphologies using direct addition.

A systematic

exploration of chemical conditions, combined with direct addition, resulted in highly reproducible morphological control. None of the showerhead, sunflower, or spicule morphologies were observed, however, which suggests that these uniquely depend on assembly by pre-existing coacervate entities as in the PILP process. In contrast, the cones appear to grow via a modular, interfacial growth process that does not occur through a liquid funnel.

Conclusions This work demonstrates the capability of complex coacervation to control mineralization processes. By carefully controlling pH and concentrations, conditions were found that preferentially mineralize

128

CaCO3 at coacervate surfaces or within coacervate volumes.

The

mineral phase of coacervate shells was found to be amorphous and to comprise about half of the capsule weight. Ammonium bicarbonate vapor infusion was found to induce new, exotic shapes, which form at the substrate surfaces at appropriate constituent concentrations. The process was found to be highly dependent on a range of physical parameters.

The variation in shapes suggests a mechanism of

sedimentation and dynamic mineralization that is controlled by solution pH gradients. Direct addition could replicate the synthesis of cones with PAA by way of a controlled assembly process that begins with mutual inhibition of mineral and coacervation. In contrast to cones derived via direct carbonate addition, vapor infusion-induced morphologies proceed via a PILP mechanism.

REFERENCES

1.

Cha, J.N., et al., Microcavity lasing from block peptide hierarchically assembled quantum dot spherical resonators. Nano Letters, 2003. 3(7): p. 907-911.

2.

Cha, J.N., et al., Spontaneous formation of nanoparticle vesicles from homopolymer polyelectrolytes. Journal of the American Chemical Society, 2003. 125(27): p. 8285-8289.

129

3.

Wong, M.S., et al., Assembly of nanoparticles into hollow spheres using block copolypeptides. Nano Letters, 2002. 2(6): p. 583-587.

4.

McKenna, B.J., Birkedal, H., Bartl, M. H., Deming, T. J., Stucky, G. D. Self-Assembling Microspheres from Charged Functional Polyelectrolytes and Small-Molecule Counterions. in Mater. Res. Soc. Symp. Proc. 2004.

5.

McKenna, B.J., et al., Micrometer-sized spherical assemblies of polypeptides and small molecules by acid-base chemistry. Angewandte Chemie-International Edition, 2004. 43(42): p. 5652-5655.

6.

Murthy, V.S., et al., Charge-driven flocculation of poly(Llysine)-gold

nanoparticle

assemblies

leading

to

hollow

microspheres. Journal of the American Chemical Society, 2004. 126(16): p. 5292-5299. 7.

Toprak, M.S., McKenna, B. J., Waite, J. H., Stucky, G. D. Tailoring Magnetic Microspheres with Controlled Porosity. in Mater. Res. Soc. Symp. Proc. 2007: Materials Research Society.

8.

Toprak, M.S., et al., Spontaneous assembly of magnetic microspheres. Advanced Materials, 2007. 19(10): p. 1362-+.

9.

Ajikumar, P.K., et al., Synthesis and characterization of monodispersed spheres of amorphous calcium carbonate and

130

calcite spherules. Crystal Growth & Design, 2005. 5(3): p. 1129-1134. 10.

Euliss, L.E., et al., Design of a doubly-hydrophilic block copolypeptide that directs the formation of calcium carbonate microspheres. Chemical Communications, 2004(15): p. 17361737.

11.

Wang, F., et al., A facile pathway to fabricate microcapsules by in situ polyelectrolyte coacervation on poly(styrene sulfonate)doped CaCO3 particles. Journal of Materials Chemistry, 2007. 17(7): p. 670-676.

12.

Xu, A.W., et al., Stable amorphous CaCO3 microparticles with hollow spherical superstructures stabilized by phytic acid. Advanced Materials, 2005. 17(18): p. 2217-2221.

13.

Yu, J.G., M. Lei, and B. Cheng, Facile preparation of monodispersed calcium carbonate spherical particles via a simple precipitation reaction. Materials Chemistry and Physics, 2004. 88(1): p. 1-4.

14.

Yu, J.G., et al., Facile fabrication and characterization of hierarchically

porous

calcium

carbonate

microspheres.

Chemical Communications, 2004(21): p. 2414-2415.

131

15.

Patel, V.M., et al. Synthesis of Calcium Carbonate-Coated Emulsion Droplets for Drug Detoxification. in ACS Symposium. 2002. Orlando, FL: American Chemical Society.

16.

Naka, K., Y. Tanaka, and Y. Chujo, Effect of anionic starburst dendrimers on the crystallization of CaCO3 in aqueous solution: Size control of spherical vaterite particles. Langmuir, 2002. 18(9): p. 3655-3658.

17.

Colfen, H. and M. Antonietti, Crystal design of calcium carbonate microparticles using double-hydrophilic block copolymers. Langmuir, 1998. 14(3): p. 582-589.

18.

Qi, L.M., J. Li, and J.M. Ma, Biomimetic morphogenesis of calcium carbonate in mixed solutions of surfactants and doublehydrophilic block copolymers. Advanced Materials, 2002. 14(4): p. 300-+.

19.

Colfen, H., Precipitation of carbonates: recent progress in controlled production of complex shapes. Current Opinion in Colloid & Interface Science, 2003. 8(1): p. 23-31.

20.

Gower, L.B. and D.J. Odom, Deposition of calcium carbonate films by a polymer-induced liquid-precursor (PILP) process. Journal of Crystal Growth, 2000. 210(4): p. 719-734.

132

21.

Sugawara, A., T. Ishii, and T. Kato, Self-organized calcium carbonate with regular surface-relief structures. Angewandte Chemie-International Edition, 2003. 42(43): p. 5299-5303.

22.

Chen, P.C., et al., Nucleation and morphology of barium carbonate crystals in a semi-batch crystallizer. Journal of Crystal Growth, 2001. 226(4): p. 458-472.

23.

Kitano, Y., D.W. Hood, and K. Park, Pure Aragonite Synthesis. Journal of Geophysical Research, 1962. 67(12): p. 4873-&.

133

Chapter V

Morphological Ternary Diagram Studies of Non-classical Calcium Carbonate Mineralization with Homopolyanions

*A related version of this chapter has been submitted as:

Brandon J. McKenna, J. Herbert Waite, Galen D. Stucky, “Biomimetic Control of Calcite Morphology with Homopolyanions,” submitted.

134

Abstract: Biomineralization is an intricate process that relies on precise physiological control of solution and interface properties.

Despite

much research of the process, mechanistic details of biomineralization are only beginning to be understood, and studies of additives seldom investigate a wide space of chemical conditions in mineralizing solutions. We present a ternary diagram-based method that globally identifies the changing roles and effects of polymer additives in mineralization. Simple polyanions were demonstrated to induce a great variety of morphologies, each of which can be selectively and reproducibly fabricated. This chemical and physical analysis also aided in identifying conditions that selectively promote heterogeneous nucleation and controlled cooperative assembly, manifested here in the form of highly organized cones. Similar complex shapes of CaCO3 have previously been synthesized using double hydrophilic block copolymers. We have found the biomimetic mineralization process to be modular and interfacial, generating large mesocrystals with high dependence on pH and substrate surface.

135

Introduction Biomineralization is the controlled process of inorganic assembly by organisms. The most familiar examples of biominerals include bones, teeth, and seashells.

Mineralized biomolecular

materials, as well as the processive strategies used to create them, are sources of fascination to materials scientists. Particularly impressive is the degree of exquisite morphological control with which organisms manipulate minerals composed of multivalent ions. Despite their high lattice energies, biology physically tunes them to suit a variety of specific needs.1 Bone, abalone nacre, and the brittlestar skeleton are preeminent examples of complex multifunctional materials, whose properties are derived from precise control of morphology.

The

resulting hierarchically ordered composite structures incorporate very small fractions of organics and yet demonstrate greatly enhanced strengths and toughnesses.2 Biominerals, and the controlling physiological processes by which they form, have fundamental importance for general biological understanding, geological interpretations, both of mineral distributions in the earth’s crust and accurate analysis of the fossil record. With regard to these latter motivational aspects, it is interesting that biologically controlled mineralization appears to have begun relatively

136

late in the course of evolution, which may partly explain why scientists have so far been able to understand genetic aspects like the Central Dogma, but are still unable to accurately understand or mimic biomineralization processes.

The interpretations of mineralization

processes here primarily regard a fundamental understanding and the hope of biomimicry. The inspiration for biomimetic engineering of materials derives from the variety of ways that nature is observed to manipulate inorganic minerals for a variety of uses: optical focusing, navigation, gravity sensing, and mechanical structure.3, 4 The lens of the extinct trilobite is composed of calcite, a naturally birefringent material; yet the organism evolved to overcome the double refraction effect by orienting the c-axis perpendicular to the surface of the lens. Magnetobacteria assemble nanoparticles of iron oxide for navigation according the earth’s magnetic field. The otoliths that occur in the inner ears of vertebrates act as weights that are coupled with cellular sensors to sense changes from gravitational force; interestingly, while they are calcite in humans, other species use alternate polymorphs of calcium carbonate or even entirely different minerals. Siliceous sponge spicules are an example of structural materials that are organized into elaborate 3d cage-like arrays that resist forces of ocean currents and may also enhance light harvesting. Their structure is also controlled at microscopic levels into

137

layered composites that better resist fracture propagation. Thus, nature is able to mold a variety of minerals and endow them with otherwise unnatural, enhanced properties, with control from nanoscopic to macroscopic levels. There is thus a drive to learn from various methods of biological control and apply them to manmade materials. It is of practical concern to harness such synthetic capability for the development of inexpensive, strong, lightweight composite “biomimetic” materials (e.g. for biocompatible replacement materials, as for bone repair with scaffolds and/or mineral sources).5 The ordered arrays of the brittlestar lens and butterfly wings could provide routes to photonic crystals. Spider silk and mussel fibers could provide designs for new, tougher or selfhealing polymers. Mussel thread surface contacts and the glue of the sandcastle worm could provide new principles for better underwater adhesives. The structure of organic fibers like those in tree wood could provide new strategies for reinforcement. Understanding the modes of mineralization of extended single crystal calcite such as starfish arms and sea urchin frameworks could impart new fabrication methods for high purity crystalline materials, as required for high efficiency solar cells. Finally, an understanding of composite structural materials could provide methods for the synthesis of tougher and harder impactresistant, structural materials.

In all cases, nature advantageously

138

directs material structure at various scales in order to effect material function. However, there is a prohibitive problem with attempting to directly apply biological methods to inorganic/composite synthesis: most biominerals are formed with an extremely high degree of control by the complex machinery of cells. Without full access to the cellular synthetic approach, scientists must find other methods of structural mimicry.

Cells interact with growing minerals in several different

ways. The coccolithophores, ancient but very prevalent algaes, exhibits exquisitely assembled CaCO3 shells (the “coccolithes”) on the cell surfaces. A high degree of intracellular control is involved: vesicles in close relation to the Golgi apparatus are constructed with a predetermined shape that controls both mineralization and transport to the cell surface where the primary components lock together and are ejected from the cell. The link between cells and minerals is more apparent considering that biomineralization is thought to originate from such species as a way to excrete mineral “waste product”, since Ca2+ is desirable at only low levels within cytoplasm.

The intracellular

mechanism is also effective in the spicules of sponge (silica) and coral (calcite), for which entire specialized cells called sclerocytes are responsible for encapsulating and constructing the mineral until near completion.

139

Cells can also control biominerals extracellularly. One way to do this is within folds of the cell membrane. Within the microcavities, membrane pumps can modulate pH and concentrations. In bone, it is well understood that the mineral component undergoes constant dissolution/reconstruction by osteoclasts/osteoblasts, respectively. And during the initial construction of bone, it has been observed that cells cluster to form pockets of solution spatially isolated from the rest of the organism. This way, cells can create supersaturated conditions within the enclosed volume and precisely define new mineral. Because of such complex control mechanisms, there is interest in examples of extracellular biomineralization in which cells control diffusion of inorganic material and biological molecules, but which otherwise do not directly participate in the fine mineralization process.

The

principles of these processes should translate to those of more cellular mechanisms. Much work towards understanding these principles has focused on the action of the soluble polymers to determine how the crystal growth mechanisms are altered in their presence. crystallization, selection,

supersaturation

crystal

growth

dictates rates,

nucleation,

and

surface

In classical polymorph energetics

(thermodynamics) coupled with kinetic mechanisms (kink-step growth) dictate crystal growth direction and the prominently exposed crystal

140

faces. In contrast to this situation, various additives, particularly acidic macromolecules, can exhibit profound effects and control over each of these parameters in the crystallization mechanism.

Such additives

include inorganic ions (e.g. Mg2+),6 organic molecules,7 LB monolayers,8 SAMs,9 various synthetic polymers, and extracted biopolymers.10,

11

Anionic macromolecules are of particular interest

biomimetically, because the proteins involved in CaCO3 mineralization typically have low PIs and contain large fractions of phosphoserine, aspartate or glutamate residues. Synthetic additives also provide model systems for understanding non-classical crystal growth. The precise role of acidic polymers remains controversial however, and several equally plausible hypotheses have been proposed. As antiscalants, they prevent mineralization by disturbing nucleation events;12, 13 in contrast, as growth initiators they induce local supersaturation via ion sequestration.14, 15 As habit modifiers, usually as part of a matrix, they selectively initiate crystal growth of specific faces, often by providing appropriate spatial periodicity.16, 17 They may also stabilize amorphous calcium

carbonate

(ACC),

sometimes

prior

to

subsequent

mineralization.18 Other more complicated polyanion functions have been suggested. In one mechanism, demonstrated by Gower et al., polymers induce liquid precursor precipitate (PILP) microspheres, which

141

subsequently nucleate and direct crystal growth.19-21

In oriented

attachment, crystallite assembly is mediated by a partially capping polymer.22,

23

Such precise, oriented construction has also been

suggested to yield mesocrystals of CaCO324-27 as well as structured crystals of BaSO4.28, 29 Double hydrophilic block copolymers (DHBCs) have been suggested as important for dynamic stabilization of growing surfaces.30 Lastly, polyanions can also self-associate in the presence of Ca2+ to yield complex coacervates, which can serve as microspherical templates.31, 32 Although various polyanions have demonstrated all of the above-mentioned roles to modulate CaCO3 morphology, in-depth evaluation of the polymer’s roles under varying physical and chemical conditions is often lacking. In many cases, it has been presumed that a specific polymer performs a single important function (and hence directs only one or few potential morphologies). Herein, we show that even simple homopolyanions can induce many CaCO3 morphologies, simply by altering experimental conditions, and in turn effectively change the polymer’s function. The implementation of ternary phase diagrams enables rapid screening of morphology over a wide range of chemical conditions, revealing many possible roles for one polymer. This facilitated the identification of regions of controlled assembly, which are more representative of biological processes. The crystallinity

142

and polymer distribution of the various morphologies are also analyzed, with particular regard to a highly oriented, layered structure.

Experimental Section Polymer solutions. All polymers were dissolved in DI H2O and diluted to 10 mM per monomer. Polyacrylic acids, sodium salts (PAA), were received as follows: PAA 90 kDa (PAA90k) (Aldrich), PAA 15 kDa (PAA15k) (Aldrich), PAA 6 kDa (PAA6k) (Polysciences), and PAA 2 kDa (PAA2k) (Aldrich). Poly-L-aspartate (PLD) (30 kDa) and polyL-glutamate (PLE) (14 kDa) were received from Sigma and dissolved in deionized H2O. Polystyrenesulfonic acid (PSS) (17 kDa, 18 wt.% in H2O) was received from Aldrich and neutralized with NaOH (Fisher). Synthesis of fluorescein-labeled PAA: 6’-Aminofluorescein was received from Fluka, 1-Hydroxybenzotriazole hydrate (HOBT) was received

from

Aldrich,

and

1-ethyl-3-(3-

dimethylaminopropyl)carbodiimide hydrochloride (EDC) was received from Pierce. To a 100 mM (per monomer) polyacrylic acid solution, 6’aminofluorescein was added to saturation (less than 5% per monomer), and the pH was adjusted to 4.5 using 0.01 M HCl (EMD). A separate solution, containing 2 equivalents of HOBt and 5 equivalents of EDC (per 6’-aminofluorescein), was added to the first solution while

143

maintaining a pH of 4.5. Following 12 h of reaction, the product was thrice dialyzed against deionized water for periods of 1d. Mineral Syntheses: CaCl2 dihydrate (EMD Chemicals Inc.), NH4HCO3, (NH4)2CO3, and NaHCO3 (Fisher) were used as received and dissolved in DI H2O freshly prior to synthesis. Typically, 10.0 mM (per monomer) of PAA, 1.00 M of a carbonate salt, and 10.00 mM CaCl2 were micropipetted and mixed in 0.5 dram shell vials (Fisherbrand, Type I Glass) to a total of 240 μL.

The resulting

mineralization solutions were placed in a humidity chamber that was temperature controlled at 30±1ºC for up to 24 h. Solution volumes decreased by less than 1 μL per hour. For analyses requiring dry or pure product, washing steps were performed by removing the supernatant, twice rinsing with pH 9 NaOH solution, and finally rinsing with deionized H2O. pH measurements:

Measurements of pH were obtained of

various mineralizing solution over the course of 2 h, using an ORION* Thermo Electron Micro pH Electrode. Optical Microscopy: Samples were collected with pipet tips, which were sometimes used to mechanically dislodge precipitates from the container, and samples were pipetted onto a microscope slide and observed with a Nikon Eclipse ME600 operating in transmission mode using a 20x lens.

144

SEM: Washed samples were pipetted and dried on a silicon wafer, and then sputtered with Pd/Au at 10 mA for 2 minutes. Samples were loaded into an FEI Sirion Sphera operating at 5 kV. XRD: Larger syntheses of the various products were screened for morphological sameness with optical microscopy.

Washed and

dried products were collected for analysis, and scanned between 20º and 70º using a Philips XPERT Powder Diffractometer. TEM: Washed products were suspended in water or absolute ethanol, and dry cast onto a lacey carbon TEM grid. Samples were analyzed with a T20 FEI Technai G2 Sphera Microscope operating at 200 kV, to record bright field images and diffraction patterns on photographic plates. Confocal Laser Scanning Microscopy (CLSM): A Leica microscope equipped with an ArKr laser was used for performing laser scanning confocal microscopy. A 0.5 mL volume of fluorescein-labeled sample was loaded into a specially prepared glass slide containing a well below a coverslip. Three-dimensional rendering was performed with the VolumeJ package within ImageJ software. Dynamic Light Scattering (DLS): Mineralizing solutions were mixed to 2 mL, with component proportions corresponding to selected morphologies. Samples were loaded into disposable sizing cuvettes and placed in a Malvern Zetasizer Nano ZS. Measurement durations

145

were set to be determined automatically, and data were accumulated over the course of 3h. The correlation function was processed using the absorption,

indices

of

refraction,

and

viscosity

parameters

corresponding to PMMA and H2O.

Results and Discussion In order to analyze the morphological phase space of the major chemical components, morphological maps were obtained in the form of ternary diagrams. Solution sets of constant volume were prepared by varying constituent proportions incrementally from stock solutions of CaCl2, polymer, and carbonate salt.

Proper stock solution

concentrations were determined empirically, such that they yielded a nearly “optimized” ternary diagram—one which is spanned by many morphologies and which also limits the variety of morphologies within individual solutions. E.g., too concentrated solutions are too far beyond equilibrium to controllably product distinct products, and too-dilute solutions are too far below equilibrium to obtain product or reveal the widest range of possible products. This concentration optimization process also maximizes the sampling information obtained; fewer points within the ternary diagram need to be queried to observe the range of morphologies.

146

For the most detailed investigation (Fig. 5.1), PAA15k and NH4HCO3 were used. Components were mixed to a total of 240 μL, using increments of 16 μL, in 0.5 dram flat-bottom glass vials, and the solutions were incubated at 30ºC for 24h. Products formed as the pH increased due to CO2 evolution, typically requiring a lag time of at least 20 min., at which point the pHs were 8.2; however, most growth occurred at pHs of 8.5 and above. Over a dozen distinct morphological regions were identified, and all products were either calcite or amorphous.

Each inset image represents a prominent or distinct

morphology for its region of phase space, and each composition and morphology can be synthesized reproducibly and selectively, although there is overlap across the guideline borders. Hybrid “twinned buds” also appeared, between the “peanut” and “bud” regions (Fig. 5.2). Ordinary calcite rhombohedra, which occur along the rightmost edge of the diagram, are not shown. The diversity of shapes demonstrates that the multiple functions displayed by this simple polyanion strongly depend on solution conditions. It is apparent that the polymer is not acting only as a template, a nucleator, an inhibitor, or a habit modifier, but rather performs all these functions plus new, emergent behavior.

147

148

Figure 5.2. Twinned bud morphology. The ternary diagrams in this study display some general trends. High proportions of Ca2+ associated rapidly with both polymer and carbonate components, resulting in large, poorly defined aggregates. At high [CO32-], crystallization was highly favored as evidenced by increased faceting. High levels of polymer inhibited crystallization and yielded little precipitate, particularly at low [Ca2+]. The absence of polymer yielded the expected calcite rhombohedra.

Low [CO32-]

products were entirely microspherical liquid charge neutralized polymer-Ca2+complexes also known as coacervates. Increasing [Ca2+] and [CO32-] from this coacervate region induced flocculation of submicron precursors into larger sponge-like masses.

This likely

resulted from the interconnection of polymer aggregates, induced by mild electrostatic screening of the added salt, coupled with rapid solidification of ACC.

149

With a further increase of [CO32-], there was a morphological transition from rapidly formed, interconnected masses into discrete microspheres in low yield, reflecting slower nucleation and weakened aggregation. Over extended periods, some of these exhibited PILP-like growth into rods.21 With modestly higher [CO32-], such microspheres nucleated the growth of larger conical morphologies, first with rod-like character and then with planar sides. At yet higher [CO32-], there was an increasing preference for growth of the initial microsphere over cone formation. This distinguished the fibrous cones from the buds, which had larger, faceted spheres and shorter, less defined rod-like portions, and which indeed occupied a distinct region of chemical space. The microspheres showed an increasing tendency to twin with higher [CO32]. At the highest carbonate concentrations, cones and microspheres ceased to form, in favor of smaller, discrete shapes with increasing faceting. The rice, peanut, and hexagonal shapes decreased in size with greater polymer concentration, supporting PAA’s role in nucleation. Many of the morphologies reported here have been described previously, for either CaCO3 or barium salts, but were formed by employing more complex polymers, instead of altering solution conditions. ACC in the form of flocs has been made previously, using polycations;33 anionic PEO-b-PAA has been shown to stabilize ACC, although as nanoparticulates rather than flocs.34 Domes have been

150

reported using polycations, although these were amorphous rather than calcitic.35 Microspheres have been made, e.g., using PLD, although these were characterized as vaterite instead of calcite.14 Cölfen et al reported the use of PEG-b-PMAA to form peanuts (“dumbbells”),36 plotting part of a morphological map as a function of pH and the [polymer]/[CaCO3] ratio.37 PEG-b-PEI-COOH has been used to make BaSO4 peanuts. 38 Rice-like formations have been made using PAA39 or PEO-b-PMAA.40 PEG-b-PEI polymers with hydrophobic moieties of different lengths have been used to fabricate microsphere-, peanut-, quadruple-, hexagon-, and rice-like shapes.41 Morphologies that are similar to the bud and cone structures have also been observed.

“Petunias” were fabricated using

carboxymethyl chitosan,42 and “shuttlecocks” were made using PEG-bPGL and PEG(84)-b-PHEE with varying degrees of phosphorylation43; these have been suggested to form around CO2 microbubble templates.44 Cones were made with the combination of two different PAA MWs, and their oriented architectures were attributed to consecutive controlled growth via oriented attachment. Structures that appear similar to the fibrous cones reported here have also been made using PAA, but for BaCrO4 or BaSO4 rather than CaCO3. 24, 27, 45 This study’s microspheres and the precursors observed at the tips of the conical morphologies are likely related to the PILPs studied by Gower

151

et al because of the precursor proximity to extended growths. Although Gower’s multi-domained films19 and rod-like structures21 are not described in this study’s standard ternary diagram, they are readily accessible using different conditions, such as different substrates or different CO2 diffusion rates. Moreover, direct comparison of our results with those of many other studies is difficult, because different parameters such as these (substrate and CO2 diffusion) can so radically alter morphology. For instance, previous studies have used different or unquantifiable methods of introducing CO32-, different carbonate salts, and/or different polymers. Furthermore, the chemical space has been less completely determined for these other conditions and methods. From the results obtained here, it would be of considerable interest and benefit to conduct a more thorough analysis of morphological trends over chemical space for these systems as well.

Variations. In order to determine important factors for morphological control, several experimental parameters were varied.

Additional

morphological diagrams were constructed to test variations of: temperature, ionic strength, carbonate sources, polymer type, and PAA molecular weight.

It was found that morphology was primarily

152

controlled by [CO32-], as defined by pH, and by the extent of polymer/Ca2+ association, particularly as functions of ionic strength and polymer type. Increments of 24 μL (10.0%) of the stock solutions were used to prepare solutions within the arrays to determine relative effects. Temperature:

Calcium carbonate’s decreased solubility at

higher temperatures was reflected in morphological changes in the ternary diagram. Refrigerated samples (4º C) precipitated very little product, with the exception of buds. Products at 20º C were similar to the 30 ºC control products, except that cones formed at high [Ca2+], where flocs otherwise occur; this is likely because of decreased polymer association at this temperature. Low-[Ca2+] products (rice, peanut shapes) were larger, reflecting lower nucleation rates. Samples at 40 ºC also formed cones rather than flocs at high [Ca2+], and additionally formed cones with high [PAA]; low [Ca2+] products appeared smaller, however. Molecular Weight: Because polymer molecular weight (MW) is known to affect both coacervation (polymer association increases with MW)31 and the antiscaling properties of PAA (with an optimum around 5 kDa)12, morphological variation with MW was anticipated. Ternary diagrams constructed with different molecular weight PAAs, using constant 10mM monomer, and indeed, despite sharing some overall trends, the diagrams have notable differences (Fig. 5.3). PAA2k was

153

less potent, failing to inhibit precipitation at higher polymer concentrations, compared to all the higher MW polymers. High [Ca2+] did not readily yield flocs, instead promoting cones and rough domes, or no product at all for low [CO32-]. PAA6k solutions formed cones under more low-carbonate conditions than with other MWs, suggesting that this MW may be near optimal for controlled assembly. PAA90k behaved contrarily, more readily forming flocs, and inducing cones only at [PAA] 3 mM and below. Many shapes, such as peanuts and rice, became rough and indistinct.

(A)

(B)

(C)

Figure 5.3. Molecular weight variations (a) PAA6k, (b) PAA2k, (c) PAA90k

154

Polymer Type: Polyanions PSS, PLD and PLE were also tested, and although they all clearly affected crystallization, only PLD yielded similarly controlled products to those made with PAA (Figs. 5.4a-c). PLD and PAA were also the only two of this set which self-associated to form coacervates with Ca2+ at room temperature; this may imply the importance of a minimal level of dynamic Ca2+ binding.

(A)

(B)

(C)

Figure 5.4. Representative optical micrographs of morphologies using (a) PLD, (b) PLE, and (c) PSS. Method of carbonate addition: The present study used aqueous carbonate salt stock solutions, which were mixed directly into mineralizing solutions. Compared with other methods of introducing carbonates, such as the sublimation/infiltration method, pre-dissolving carbonate offers several advantages: the concentration is more easily quantified, by a known initial amount; products are produced more quickly, with a typical lag time of only 3 h instead of 8+ h; and results are more reproducible because variables affecting the infiltration rate

155

are eliminated.

Furthermore, infiltration using crushed ammonium

carbonate salts was found to produce some of the same morphologies, indicating the methods operate similarly. This was elaborated in the previous chapter. Because it only offered advantages for these studies, direct addition was chosen. Carbonate Source, and pH: All three carbonate sources were capable of inducing the morphologies reported here, but required different initial concentrations.

The pH values of solutions with

ammonium salts remained buffered near bicarbonate’s pKa, typically starting near pH 8.0 and ending around pH 9. In contrast, using sodium salt encumbered controlled precipitation, because it induced more dramatic pH shifts; however, fine tuning pH and carbonate content was able to produce appropriate conditions, which verifies that ammonium is not essential to the process (Fig. 5.5).

156

Figure 5.5. Fibrous cones fabricated using sodium bicarbonate. Because an aqueous solution of 0.5 M (NH4)2CO3 is equivalent to a 1.0 M NH4HCO3 solution with half of its carbonate content converted to CO2(g), the two salts are directly comparable. In fact, using such stock solutions yielded roughly equivalent diagrams, again suggesting that the initial concentration of carbonates is less important than adequate pH elevation. Ionic Strength: The imbalance of carbonate compared to the other solution species (roughly 100-fold higher) is particularly striking because marine biomineralization occurs with reversed proportions, in which carbonates total to around 3 mM. Since effects on pH change little after about 100 mM, another explanation for such high

157

concentrations is needed. Therefore, the effect of increasing the ionic strength with the carbonate precursors was tested. It was found that flocs are immediately produced when low levels of carbonate (4mM [Ca2+] were altered, and most strikingly, solutions that otherwise would yield flocs instead produced a small number of fibrous cones (Fig. 5.6). This reaffirms the importance of the ionic strength increasing function of carbonate salt for producing complex morphologies with anionic polymers. Container/substrate type.

Full diagrams based on different

containers were not obtained, because even the use of different glass containers caused different morphologies (Fig. 5.7). This is probably not only due to different qualities of glass, but instead largely due to different physical effects. For instance, larger containers provide more surface area for faster CO2 escape and therefore vastly different spatiotemporal pH conditions. The standard glass vials that were used also limited evaporation because their height created reflux conditions. From the conditions explored, complex growth was found to occur on silica, polystyrene, and polyethylene; only random aggregate growth or flocs were observed on substrates of mica, aminopropylsiloxanemodified glass, and the same glass again modified with terephthalic acid. Perhaps charged surfaces disfavor controlled assembly either by too strongly repelling precursor material or over-adsorbing solution polymer. Growth on aragonite, or aragonite seeding, resulted only in

159

the appearance of few spheres on the mineral surfaces. No higher order growth was observed, probably indicating polymorph specificity as controlled by the polymer, which may instead dissolve aragonite crystal.

(A)

(B)

(C)

Figure 5.7. Cones grown on different surfaces: (a) silicon wafer, (b) Petri dish, (c) polystyrene. Additive Mg2+. Various levels of MgCl2 were tested in another examination of morphogenesis across chemical space.

The most

prominent effect was enhanced mineral inhibition. When added to solutions from the standard ternary diagram, the products were most often coacervates.

Higher order crystalline morphologies were

obtained however, using double the amount of Ca2+ and only 10% the amount of polymer, in the case of 5-20 mM MgCl2. This further confirmed the mineral-inhibiting role of polymer.

While control

samples without polymer yielded expected aragonite crystals, only calcite products were found with added PAA or PLD, at the concentrations tested. This affirms the calcite specific role of both of these polymers.46, 47

160

Calcium phosphate. The ternary phase diagram approach was applied to mixtures of polymer (PAA, PSS, PLD, PLE, alginate, polyphosphoserine, and gelatin) at pHs of 7.5 and 5.5 and at 40˚C. However, at all tested concentrations, the products were all amorphous flocs, or mixtures of flocs with crystallites (Fig. 5.8). Adding extra NaCl to disturb flocs only resulted in a sharper hydroxyapatite diffraction peak (in the case of gelatin), but did not induce any higher order assemblies as was observed for calcium carbonate. This may reflect a few important differences between phosphate and carbonate mineralization: 1) the soluble amorphous-phase inducing proteins for hydroxyapatite may require more specific conformations and primary structure, 2) in contrast to nacre-like, flat-surface mineralization, bone is formed within predefined matrices of collagen, and 3) bone formation involves more precise cellular involvement than, e.g. nacre. It may be that calcium phosphates are unable to maintain disordered, metastable structures for as prolonged times because of much lower solubility products.

161

(A)

(B)

4000 3500

Gelatin, High NaCl polyPhosphoserine

Intensity

3000 2500 2000 1500 1000 500 20

30

40

50

60

70

2 Theta

Figure 5.8. (A) SEM images of gelatin/Ca-PO4 aggregates with minimal but poorly dispersed HAP mineral induced by increased ionic strength. (B) XRD patterns indicating predominantly amorphous material.

Characterization:

All products were analyzed by LOM, SEM,

CLSM, and various diffraction methods.

Conical morphologies

(umbrellas, buds, and fibrous cones) were of particular interest because of their complexity and order, which are indicative a controlled assembly process. The umbrella morphology is new and also exhibits particularly interesting qualities, which are highlighted below. The mechanism by which the mineral structures grow was examined with real-time optical microscopy of solutions in Petri dishes or in specially made reaction slides containing shallow wells. A typical lag of >20 min preceded precipitation, except with coacervates and flocs, which form sooner— often upon initial mixing. Peanut shapes and other high-CO32- morphologies formed while suspended in solution, before eventual sedimentation and further growth on the container surface.

162

In contrast to the other shapes, cones depend on directional growth at the container surface, following nucleation. The first step is the settling of small precipitates, which initiate growth at the substrate surface. Conical shapes proceed to grow at their bases, where they interface with the substrate, such that the cones point upwards, as also observed via time-resolved optical microscopy. This provides direct evidence of the growth mechanisms, and is in contrast to previously proposed mechanisms for other similar conical or “flower-like” shapes.44, 48

The growth direction was confirmed by SEMs of cones

grown directly on Si wafer substrates (Fig. 5.9). Although fibrous cones also appeared to begin growth at their bases; most fell on their sides and continue growing in this position. The preferential growth of conical morphologies at the crystal/substrate surface suggests that the growth mode is sensitive to surface energetics. In fact, cones grown on different substrates were quite different in appearance, even when different glass surfaces were used (Figs. 5.7a-c). Similar dependence has been previously observed.48

163

Figure 5.9. SEM image of cones grown directly on a silicon wafer. SEM images of umbrellas highlight the organization involved during their growth process (Fig. 5.10a). The structures are evidently modular, indicating growth by stepwise addition of solution precursors, which appear nearly monodisperse despite the polydispersity in polymer starting materials. Their periodicity is reminiscent of patterns found in self-regulating systems, and could reflect local fluctuations in pH or other chemical concentrations during formation.

Fibrous

structures do not appear faceted and may be formed by similar precursors which orient and smoothen, in a process that is akin to the SLS mechanism put forth by Gower et al.21 Dynamic light scattering measurements confirm that particles on the order of 200-300 nm are present in mineralizing solutions for at least

5h (Fig. 5.11a).

Depending on [Ca2+], these particles appear at or before the bicarbonate pKa of 8.3; if prevented from sedimentation by disturbing solutions, 164

they appear to grow with increasing pH and, as expected, acquire increasingly negative surface charge (Fig. 5.11B). These precursor “seeds” have been observed previously, and are thought to be amorphous.18, 34, 49 They have been identified as the cause of granular superstructures in non-classical crystallization processes.39 (A)

(B)

(C)

(D)

Figure 5.10. Umbrella fragments. (A) SEM, showing modular alignment. (B) TEM image, its diffraction pattern, and an image at an alternate angle revealing 2 connected planar sections. (C) Averaged projection confocal micrograph, showing striped pattern of polymer (scale bar 10μm). (D) Single plane confocal micrograph, showing a different, PILP-like growth and excluded polymer (scale bar 10μm).

165

1200

(B)

1100 1000 900

Size E

800 700

Size (nm)

Low Ca High Ca

0 -5

-10

Zeta Potential (mV)

(A)

600 500 400 300

-15 -20 -25 -30

200 100

-35

0

-40

-100 7.8

8.0

8.2

8.4

8.6

7.7

7.8

7.9

8.0

8.1

8.2

8.3

8.4

8.5

8.6

pH

pH

Figure 5.11. (A) DLS average particle size and (B) Zeta potential of particles of two otherwise cone-producing mixtures, versus pH increase with time. The crystallinity of products was determined from powder XRD, single crystal XRD, IR spectroscopy, or TEM, depending on product size, yield, and morphological purity. All crystalline products were indexed to calcite, the thermodynamically most stable form of calcium carbonate (Fig. 5.12). Of note are the flocs, which indicate one means of achieving conditions that stabilize ACC.

Microspheres

diffracted as calcite and are widespread across the diagram with a gradual transition from flocs to microspheres appears gradual. Umbrella fragments analyzed with TEM surprisingly revealed that the full shards diffracted as single crystals, confirming the precise, oriented arrangement directed by the growth process (Fig 5.10D). Rice shapes were characterized by TEM analysis and appear as single crystals, but are comprised of smaller, ~20nm grains (Fig. 5.13).

166

(A)

(B)

10000

80

Transmittance

8000

Intensity

100

6000

4000

60

40

Umbrella Flocs

20

2000

0

0 20

30

40

50

60

400

70

600

800

1000

1200

1400

1600

1800

-1

Wavelength (cm )

2 Theta

Figure 5.12. (A) Representative X-ray diffraction spectrum, showing only calcite diffraction peaks from cones. (B) Representative IR spectra of calcite cones and amorphous flocs.

(B)

(A)

Figure 5.13. (A) TEM bright field, SADP, and (B) bright field closeup image of rice morphology.

Confocal Laser Scanning Microscopy (CLSM):

In order to

determine the polymer distributions within various morphologies, a fluorescently labeled polymer was used in the syntheses (a morphological ternary diagram constructed with the labeled polymer did not deviate from the control), and confocal microscopy was used to gather cross sectional images; these images and their three-dimensional

167

reconstructions are available in the supplementary information. The polymer was integrated throughout the structures despite their crystalline properties, although some cross-sections revealed internal regions from which the polymer was partially excluded. The polymer distributions in the conical products demonstrated some discontinuity, reflecting the organization seen in SEM images. The nucleating tips found on many cones usually had the highest concentration of polymer, followed by the longitudinal “legs.”

In

agreement with structures observed with electron microscopy, umbrella fragments had a periodic, striped distribution of polymer that defines the border between horizontal layers of what must be crystalline (Fig 5.10B). A similar effect was evident in images of PILP-type growths, which showed dark (crystalline) rods surrounded by a sheath of excluded polymer (Fig. 5.10C). This patterning is likely due to microphase separation, in which the crystallizing mineral partially excludes polymer. A similar mechanism has been hypothesized for hierarchical biomineral structures,50,

51

and importantly represents a new

methodology of achieving periodic, layered composite materials via cooperative organization on the photonic length scale.

168

169

The combined results of the ternary diagrams and the various characterization procedures suggest mechanisms similar to those in Scheme 5.1.

Peanuts might also form according to previously

described rod-to-dumbbell mechanism,51 but the minor presence of similarly sized single spheres and the existence of the quadruple morphology both indicate that twinning may occur. As previously suggested, mineralization originates from precursor aggregates that arise from metastable solutions following an appropriate pH increase. The exact composition of these aggregates remains unknown, but they are probably distinct from PAA/Ca2+ coacervates, since they occur in a different region of the ternary diagram, and because centrifugations of solutions throughout mineralization yielded only solid, not liquid, precipitates. Following their formation, solution conditions (such as pH, chemical concentrations and physical parameters) dictate whether these aggregate, crystallize, or assemble interfacially. Umbrellas are one example of a morphology that forms interfacially, such that the precursor modules successively position themselves into crystallographic alignment at the base. While oriented attachment cannot be ruled out from these experiments, it is reasonable that in these aqueous media of high ionic strength, the precursors are first amorphous and then crystallize once they have fused with the growing structure. Supporting this hypothesis was the observation that

170

a solution of amorphous flocs slowly transformed into several large cones (Fig. 5.14).

Following fusion to the growing structure,

crystallization induces micro-phase separation, partially excluding the polymer into layers.

(A)

(B)

(C)

(D)

Figure 5.14. Photographs of large fluorescent cones, under UV (a) and ambient light (b); and still single crystal x-ray diffraction (c) of a whole 2mm cone, pictured in (d). Conclusions Polyanionic additives are capable of inducing a multitude of morphologies in crystallizing systems, thereby exhibiting various functions in the process.

Polymers behave as antiscalants, ACC

stabilizers, nucleation sites, and inducers of precursor assemblies— often demonstrating multiple functions at once.52 Exploring the phase

171

space of chemical components, using ternary diagrams, is useful for exploring such diversity. With such diagrams, one can rationally study conditions that promote specific shapes, that selectively promote ACC, or that induce controlled assembly processes.

We have thereby

demonstrated fundamental connections among previous studies, having shown that all such properties can be achieved by the same polymer. This is of particular benefit to a field that has lacked standard synthetic methods for analyzing the effects of solution additives.

The

methodology may provide a deeper understanding of biomineralization and indicate new methods for synthesizing high performance composite materials. Given the potential range of possible growth modes, proper characterization of both synthetic and biological polymers will require testing across chemical space. The various morphologies were shown to be dependent on several critical parameters, and each can be controlled to reproducibly select mechanisms of crystal growth. Low temperature inhibits most assemblies, polymer molecular weight affects the inhibition potency and the degree of association, and polyanions that form complex coacervates with Ca2+ appear to have a more dramatic role in directing crystal assembly.

Carbonate salt content affects the pH, both by

buffering and by inducing pH increase; pH, in turn, acts as a synthetic “switch” to assembly—a common method in biological systems.

172

Importantly, carbonate salts also control ionic strength, which is especially critical for suppressing polymer association. Finally, the growth mode is highly sensitive to substrate properties. This systematic investigation has revealed a controlled, interfacial, modular assembly process involving organic/inorganic phase separation, manifested in the form of highly organized umbrella morphologies with periodic polymer/calcite layers of polymer and crystallographic alignment. Since the PAA and PLD used in these studies are simple and polydisperse, the process is quite general, and such a mechanism should also be accessible with more complex and well defined polymers, as those used in biomineralization.

The

amorphous-crystalline transition method is common to many biological processes. Moreover, the resulting structures share several properties with nacre: alternating organic/inorganic layers, high crystallographic alignment with registry between layers, and, as recently characterized, composition of smaller interconnected crystallites into mesocrystals.5355

Despite obvious differences between the two materials, this

controlled growth mechanism may still come to explain common features, and further research has yet to reveal the effects of different substrates for more complex additive systems.

Regardless of its

biological relevance, this controlled assembly represents a novel, easily reproducible method for arranging synthetic materials into periodic

173

structures, which may be engineered to attain desirable mechanical or optical properties.

REFERENCES

1.

Sanchez, C., H. Arribart, and M.M.G. Guille, Biomimetism and bioinspiration as tools for the design of innovative materials and systems. Nature Materials, 2005. 4(4): p. 277-288.

2.

Smith, B.L., et al., Molecular mechanistic origin of the toughness of natural adhesives, fibres and composites. Nature, 1999. 399(6738): p. 761-763.

3.

Mann, S., Biomineralization: Principles and Concepts in Bioinorganic Materials Chemistry. 2001: Oxford University Press, Inc. 198.

4.

Simkiss, K. and K.M. Wilbur, Biomineralization: Cell Biology and Mineral Deposition. 1989, San Diego, California 92101: Academic Press, Inc. 337.

5.

Hartgerink, J.D., E. Beniash, and S.I. Stupp, Self-assembly and mineralization of peptide-amphiphile nanofibers. Science, 2001. 294(5547): p. 1684-1688.

174

6.

Han, Y.J., et al., Template-dependent morphogenesis of oriented calcite crystals in the presence of magnesium ions. Angewandte Chemie-International Edition, 2005. 44(16): p. 2386-2390.

7.

Fu, G., et al., Acceleration of calcite kinetics by abalone nacre proteins. Advanced Materials, 2005. 17(22): p. 2678-+.

8.

Xu, G.F., et al., Biomimetic synthesis of macroscopic-scale calcium carbonate thin films. Evidence for a multistep assembly process. Journal of the American Chemical Society, 1998. 120(46): p. 11977-11985.

9.

Han, Y.J. and J. Aizenberg, Face-selective nucleation of calcite on self-assembled monolayers of alkanethiols: Effect of the parity of the alkyl chain. Angewandte Chemie-International Edition, 2003. 42(31): p. 3668-3670.

10.

Belcher, A.M., et al., Control of crystal phase switching and orientation by soluble mollusc-shell proteins. Nature, 1996. 381(6577): p. 56-58.

11.

Levi, Y., et al., Control over aragonite crystal nucleation and growth: An in vitro study of biomineralization. Chemistry-a European Journal, 1998. 4(3): p. 389-396.

12.

Loy, J.E., J.H. Guo, and S.J. Severtson, Role of adsorption fractionation in determining the CaCO3 scale inhibition

175

performance of polydisperse sodium polyacrylate. Industrial & Engineering Chemistry Research, 2004. 43(8): p. 1882-1887. 13.

Reddy, M.M. and A.R. Hoch, Calcite crystal growth rate inhibition by polycarboxylic acids. Journal of Colloid and Interface Science, 2001. 235(2): p. 365-370.

14.

Roque, J., et al., Crystal size distributions of induced calcium carbonate crystals in polyaspartic acid and Mytilus edulis acidic organic proteins aqueous solutions. Journal of Crystal Growth, 2004. 262(1-4): p. 543-553.

15.

Tsortos, A. and G.H. Nancollas, The role of polycarboxylic acids in calcium phosphate mineralization. Journal of Colloid and Interface Science, 2002. 250(1): p. 159-167.

16.

Addadi, L., et al., A Chemical-Model for the Cooperation of Sulfates and Carboxylates in Calcite Crystal Nucleation Relevance to Biomineralization. Proceedings of the National Academy of Sciences of the United States of America, 1987. 84(9): p. 2732-2736.

17.

Addadi, L. and S. Weiner, Interactions between Acidic Proteins and

Crystals

-

Stereochemical

Requirements

in

Biomineralization. Proceedings of the National Academy of Sciences of the United States of America, 1985. 82(12): p. 4110-4114.

176

18.

Faatz, M., F. Grohn, and G. Wegner, Amorphous calcium carbonate:

Synthesis

and

potential

intermediate

in

biomineralization. Advanced Materials, 2004. 16(12): p. 996-+. 19.

Gower, L.B. and D.J. Odom, Deposition of calcium carbonate films by a polymer-induced liquid-precursor (PILP) process. Journal of Crystal Growth, 2000. 210(4): p. 719-734.

20.

Olszta, M.J., E.P. Douglas, and L.B. Gower, Scanning electron microscopic analysis of the mineralization of type I collagen via a polymer-induced liquid-precursor (PILP) process. Calcified Tissue International, 2003. 72(5): p. 583-591.

21.

Olszta, M.J., et al., Nanofibrous calcite synthesized via a solution-precursor-solid mechanism. Chemistry of Materials, 2004. 16(12): p. 2355-2362.

22.

Navrotsky,

A.,

Energetic

clues

to

pathways

to

biomineralization: Precursors, clusters, and nanoparticles. Proceedings of the National Academy of Sciences of the United States of America, 2004. 101(33): p. 12096-12101. 23.

Niederberger, M. and H. Colfen, Oriented attachment and mesocrystals: Non-classical crystallization mechanisms based on nanoparticle assembly. Physical Chemistry Chemical Physics, 2006. 8(28): p. 3271-3287.

177

24.

Kulak, A.N., et al., Continuous structural evolution of calcium carbonate particles: A unifying model of copolymer-mediated crystallization. Journal of the American Chemical Society, 2007. 129(12): p. 3729-3736.

25.

Miura, T., et al., Emergence of acute morphologies consisting of iso-oriented calcite nanobricks in a binary poly(acrylic acid) system. Crystal Growth & Design, 2006. 6(2): p. 612-615.

26.

Wang, T.P., M. Antonietti, and H. Colfen, Calcite mesocrystals: "Morphing" crystals by a polyelectrolyte. Chemistry-a European Journal, 2006. 12(22): p. 5722-5730.

27.

Xu, A.W., et al., Uniform hexagonal plates of vaterite CaCO3 mesocrystals formed by biomimetic mineralization. Advanced Functional Materials, 2006. 16(7): p. 903-908.

28.

Wang, T.X. and H. Colfen, In situ investigation of complex BaSO4 fiber generation in the presence of sodium polyacrylate. 1. Kinetics and solution analysis. Langmuir, 2006. 22(21): p. 8975-8985.

29.

Wang, T.X., A. Reinecke, and H. Colfen, In situ investigation of complex BaSO4 fiber generation in the presence of sodium polyacrylate. 2. Crystallization mechanisms. Langmuir, 2006. 22(21): p. 8986-8994.

178

30.

Yu, S.H. and H. Colfen, Bio-inspired crystal morphogenesis by hydrophilic polymers. Journal of Materials Chemistry, 2004. 14(14): p. 2124-2147.

31.

Bungenberg de Jong, H.G., in Crystallisation- coacervationflocculation in colloid science, H.R. Ed. Kruyt, Editor. 1949, Elsevier: Amsterdam. p. 232-258.

32.

Delacruz, M.O., et al., Precipitation of Highly-Charged Polyelectrolyte Solutions in the Presence of Multivalent Salts. Journal of Chemical Physics, 1995. 103(13): p. 5781-5791.

33.

Ono, H. and Y.L. Deng, Flocculation and retention of precipitated

calcium

carbonate

by

cationic

polymeric

microparticle flocculants. Journal of Colloid and Interface Science, 1997. 188(1): p. 183-192. 34.

Guillemet, B., et al., Nanosized amorphous calcium carbonate stabilized by poly(ethylene oxide)-b-poly(acrylic acid) block copolymers. Langmuir, 2006. 22(4): p. 1875-1879.

35.

Xu, X.R., J.T. Han, and K.L. Cho, Deposition of amorphous calcium carbonate hemispheres on substrates. Langmuir, 2005. 21(11): p. 4801-4804.

36.

Cölfen, H.Q., L., Progr. Colloid Polym. Sci., 2001. 117: p. 200203.

179

37.

Colfen, H. and L.M. Qi, A systematic examination of the morphogenesis of calcium carbonate in the presence of a double-hydrophilic block copolymer. Chemistry-a European Journal, 2001. 7(1): p. 106-116.

38.

Qi, L.M., H. Colfen, and M. Antonietti, Control of barite morphology

by

double-hydrophilic

block

copolymers.

Chemistry of Materials, 2000. 12(8): p. 2392-2403. 39.

Donnet, M., et al., Use of seeds to control precipitation of calcium carbonate and determination of seed nature. Langmuir, 2005. 21(1): p. 100-108.

40.

Marentette, J.M., et al., Crystallization of CaCO3 in the presence

of

PEO-block-PMAA

copolymers.

Advanced

Materials, 1997. 9(8): p. 647-651. 41.

Sedlak, M. and H. Colfen, Synthesis of double-hydrophilic block copolymers with hydrophobic moieties for the controlled crystallization of minerals. Macromolecular Chemistry and Physics, 2001. 202(4): p. 587-597.

42.

Liang, P., et al., Petunia-shaped superstructures of CaCO3 aggregates modulated by modified chitosan. Langmuir, 2004. 20(24): p. 10444-10448.

43.

Rudloff, J., et al., Double-hydrophilic block copolymers with monophosphate ester moieties as crystal growth modifiers of

180

CaCO3. Macromolecular Chemistry and Physics, 2002. 203(4): p. 627-635. 44.

Rudloff, J. and H. Colfen, Superstructures of temporarily stabilized nanocrystalline CaCO3 particles: Morphological control via water surface tension variation. Langmuir, 2004. 20(3): p. 991-996.

45.

Yu, S.H., et al., Growth and self-assembly of BaCrO4 and BaSO4

nanofibers

toward

hierarchical

and

repetitive

superstructures by polymer-controlled mineralization reactions. Nano Letters, 2003. 3(3): p. 379-382. 46.

Wierzbicki, A., et al., Atomic-Force Microscopy and Molecular Modeling of Protein and Peptide Binding to Calcite. Calcified Tissue International, 1994. 54(2): p. 133-141.

47.

Zhang, S.K. and K.E. Gonsalves, Synthesis of Calcium Carbonate-Chitosan Composites Via Biomimetic Processing. Journal of Applied Polymer Science, 1995. 56(6): p. 687-695.

48.

Yu, S.H., H. Colfen, and M. Antonietti, Control of the morphogenesis of barium chromate by using double-hydrophilic block copolymers (DHBCs) as crystal growth modifiers. Chemistry-a European Journal, 2002. 8(13): p. 2937-2945.

49.

Bolze, J., et al., Time-resolved SAXS study of the effect of a double hydrophilic block-copolymer on the formation of CaCO3

181

from a supersaturated salt solution. Journal of Colloid and Interface Science, 2004. 277(1): p. 84-94. 50.

Sumper, M., A phase separation model for the nanopatterning of diatom biosilica. Science, 2002. 295(5564): p. 2430-2433.

51.

Simon, P., U. Schwarz, and R. Kniep, Hierarchical architecture and real structure in a biomimetic nano-composite of fluorapatite with gelatine: a model system for steps in dentinoand osteogenesis? Journal of Materials Chemistry, 2005. 15(47): p. 4992-4996.

52.

Furuichi, K., Y. Oaki, and H. Imai, Preparation of nanotextured and nanoribrous hydroxyapatite through dicalcium phosphate with gelatin. Chemistry of Materials, 2006. 18(1): p. 229-234.

53.

Oaki, Y., et al., Bridged nanocrystals in biominerals and their biomimetics: Classical yet modern crystal growth on the nanoscale. Advanced Functional Materials, 2006. 16(12): p. 1633-1639.

54.

Rousseau, M., et al., Multiscale structure of sheet nacre. Biomaterials, 2005. 26(31): p. 6254-6262.

55.

Takahashi, K., et al., Highly oriented aragonite nanocrystalbiopolymer composites in an aragonite brick of the nacreous layer of Pinctada fucata. Chemical Communications, 2004(8): p. 996-997.

182

Chapter VI

Towards a Model for Nacre Formation

183

Abstract: Nacre is a much studied biomineral composite with a lamellar, brick-wall architecture. The mechanism of nacre formation remains elusive, despite increasingly detailed research into nacre’s molecular and crystallographic architecture. The basic and detailed features of nacre are reviewed, and the material is compared and contrasted with the umbrella morphology presented in the previous chapter. An analogous assembly mechanism is proposed which accords with recent detailed observations of nacre, and which also explains previously confusing but basic aspects of nacre formation. Remaining questions and directions for further research are presented.

184

Of the structural biominerals, perhaps the most intriguing structure is nacre (or “mother-of-pearl”), known for its pearlescent luster and brick-wall-like architectures.1,

2

Just one of several other

naturally occurring layers in the shells of mollusks, the nacreous layers of Gastropoda and Bivalvia have attracted scientific attention for many reasons.

Firstly, the pearlescent lustre, as particularly observed in

abalone shells, has been a phenomenon of fascination and of mimetic interest for construction of periodic structures on the photonic length scale. Secondly, its brick-wall-like structure is reminiscent of the manmade version and appears spatially optimized against forces normal to the surface. Thirdly, the detailed organizational characteristics highlight many of the principles that are involved in biomineralization, at large. For instance, it is a composite with features of down to several nanometers, the crystal polymorph and habit are both specifically controlled by organics, the crystal orientation and morphology in turn control the overall material structure, and the polymer matrix (between 0.5 and 5 wt. %) that is responsible for the 3000x toughness enhancement of nacre over ordinary aragonite.3 Fourthly, nacre grows in films, a structure that at once appears more tenable than other 3d complex architectures in nature and has immediately recognizable

185

modes of application. That mollusks can control various planar growth modes is encouraging for mimetic pursuits. And fifthly, in contrast to a great variety of other biominerals, nacre is formed extracellularly. The shells, as of abalone, are formed by excretion of components from the organism’s mantle into the extrapallial fluid.

This fluid and the action of the biomolecules it

contains direct the construction of the shell layers by selecting crystal polymorph,

crystal

morphology,

and

the

distribution

of

macromolecules—naturally all highly interrelated. Therefore, although the process remains complex, the mode of nacre synthesis removes the tremendously daunting tasks of mimicking the cellular machinery such as the Golgi apparatus and attaining such precision of the dynamic chemical parameters in highly confined spaces.

Instead, what are

needed to understand nacre formation are: 1) an understanding of diffusional processes in a “bulk” medium, which chemists are capable of, and 2) knowledge of the biological elements, such as protein sequence and other precursor macromolecules. However, there is a third requirement to understand the interactions of biomineral precursors for given conditions of chemical diffusion. These include nucleation and growth of inorganic minerals, the supramolecular association

of

biomacromoleules,

chemical

crosslinking,

and

interactions of the interaction of biomolecules with inorganic material.

186

Nacre is a layered, periodic material with alternating mostlyorganic and mostly-inorganic layers (lamellae). The inorganic layers consist mostly of planar discs (tablets, slabs) of aragonite oriented with their (c-axis) [001] planes expressed along the surfaces, and the discs grow laterally to achieve full coverage. The crystal orientations of slabs within the same plane are not correlated, but crystallinity is conserved vertically between adjacent slabs.4-6 Either side of the discs is in intimate contact with water-soluble proteins, which form the outer parts of the organic layers. Between these soluble elements are sheets of an insoluble chitinous material similar to silk fibroin. The growth of nacre differs between species, for instance in hexagonality of growing slabs or slab thickness, but there is a clear distinction between the growth modes in Bivalvia and Gastropoda. The former prefers lateral growth, finishing one layer before constructing the next, while the latter prefers normal/axial/vertical growth, in which layers grow laterally only subsequent to vertical propagation into conical ‘stack of coins’ structures that have been likened to Christmas trees or cairns. More detailed observations of nacre’s organization and structure have been revealed by a vast amount of research.7 The chitin-like polysaccharide is also arranged periodically in a less common parallelstrand conformation, and may be in some register with the crystal. The soluble, silk-like proteins have been found not strictly confined to the

187

edges of the aragonitic layers, but rather incorporated more dispersed throughout these inorganic layers.8 The slabs have furthermore been found to be composed of small, crystallographically oriented granules rather than being a continuous mineral.9 The surfaces of the crystalline portions have been reported to be coated with ~5nm of amorphous material.10

Finally, the demineralized organic matrix appears

perforated with small channels or pores through which aragonite mineral bridges are able to propagate crystallinity with preserved orientation from layer to layer.4,

11

This latter detail marked an

important paradigm shift from a growth model that previously described organic components with perfect epitaxial alignment and transmittance of crystal orientation to successive layers. Despite the occurrence of mineral bridging, the organic content of nacre nevertheless has extraordinary control over crystallinity.12-14 In particular, the soluble matrix proteins can specify the calcium carbonate polymorph in crystallizing solutions. Belcher et al demonstrated that proteins derived from the organic matrix may spur nucleation of aragonite in solution conditions that would otherwise favor calcite, even in the absence of Mg2+.12 Furthermore, these proteins are able to modulate polymorph selection, causing aragonite to grow on the surface of a calcite substrate. Finally, there was some evidence that

188

these proteins were sufficient to induce the formation of small layered structures, as would otherwise be found in nacre. Despite the cumulative knowledge of nacre structure, the precise mechanism of formation is still lacking, so some basic questions remain (Fig. 6.1). For instance, regardless of whether crystallization occurs over or within matrix sheets, what determines layer thickness and causes it to be so uniform across layers? That is, why do matrix materials not prematurely bind to growing crystalline layers?

If

organics control the termination of aragonite growth, what prevents premature termination or overgrowth and ‘blanketing’ of organic layers over unfinished aragonite tablets?

What causes the horizontally-

preferred growth of bivalves versus the vertically-preferred growth of gastropods? How can one protein induce a layered structure? Why are the slabs of aragonite composed of granules, and how is polymer occluded throughout the mineral? programmed or incidental?

Are matrix sheet perforations

If the matrix layers are laid down

successively over inorganic layers, why/how do they not blanket over entire unfinished mineral layers?

And finally, how do the different

classes select for lateral or perpendicular/vertical growth modes?

189

(A) (B) (C) Figure 6.1. Modes of growth that are not observed in nacre formation, and which still lack a mechanistic explanation. Herein, I suggest that some of the recent observations of nonclassical crystallization can unify into a general phase separation mechanism that begins to explain the broader features of nacre as well as some of the unanswered questions outlined above.

Progress Towards a New Model The observed umbrella morphologies, as identified through the systematic ternary diagram approach in the previous chapter, serve as an embodiment of the generalized mechanism.

It should be

emphasized, however, that the umbrellas are very different from nacre in several important ways. Most notably, the umbrellas are composed of calcite rather than aragonite. Secondly, they appear to grow where the crystal contacts the substrate, conferring a conical rather than lamellar structure.

Thirdly, they lack an insoluble matrix and

reinforcing proteins. Finally, there is no initial habit-selecting organic layer. Therefore, the umbrella obviously fails, in some aspects, as a model system for nacre formation.

190

However, the umbrellas have several commonalities with nacre, and hence the two structures may share mechanistic growth aspects. Firstly, they are both made from calcium carbonate, and therefore involve similar constraints in solubility, pH, etc. Secondly, they both contain alternating layers of organic and inorganic material. Thirdly, the successive layers of each are crystallographically oriented such that the structures diffract as single crystals. This transfer of crystallinity is also visible macroscopically, such that the modules appear periodically both within and across layers; FFTs of images confirm such patterning (Fig. 6.2). Fourthly, it is also likely that polymer is occluded in small portions throughout the inorganic layers, and that the inorganic layers are composed of smaller granules. Both of these features are more easily observed in the smaller rice morphologies that were reported, as the umbrellas are often too thick to observe granules, and polymer in the inorganic layers can not be resolved between the 250nm spacings, due to photonic limitations to spatial resolution. (A)

(C)

(B)

2 μm Figure 6.2. (A) Umbrella microstructure displays higher order morphological control by crystal structure, evinced by (B) an enhanced Fourier transform and (C) the inverse Fourier transform.

191

Therefore, the structural features of umbrellas at least demonstrate the kinds of complexity feasibly achievable by a simple polymer, and therefore it is possible that the simpler system’s formation mechanism generally applies more broadly to other nonclassical crystallization systems involving polyelectrolytes, including nacre. What then is the mechanism of umbrella formation? There are two broad classes of mechanisms for periodic structures of this sort: those in which localized chemical oscillations alternatingly select assembly modes, and those based on building blocks of well defined size. In the first kind of mechanism, an increase in pH would cause crystallization of one layer of calcium carbonate, then creating a localized concentration of H+, perhaps leading to a halt in crystallization and temporarily preferred adsorption of polymer to the crystal. The bulk solution and perhaps some of the polymer would act as proton sinks, in turn raising the pH and inducing the next layer of crystal onto an incompletely covered previous layer. However, such periodic pattern formation has ordinarily only been observed when diffusion is much slower,15 and the modular appearance of various SEM images suggests the latter mechanism. Other evidence indicates that the cones grow by the assembly of discrete packets, rather than by a continuous but periodic growth process. Consider the other various micromorphological variations of

192

different umbrellas (Fig. 6.3). Some of the structures appear smooth, with sharp edges reminiscent of nacreous structure (Fig. 6.3A). But slight variations in concentrations (currently there is no known trend) induce features which appear derived from modular growth modes. In some cases, modules appear disorganized, yet they clearly form sharp steps (Fig. 6.3B). Other cases, as mentioned in the previous chapter, appear composed of more highly aligned modules (Fig. 6.2). Finally, some structures strongly resemble PILP rods, having bulbous heads and demonstrating polymer/crystal phase separation (Fig. 6.3C).

As

described by Gower et al, such rods grow by the action of a liquid droplet that funnels solution precursor material to the crystal growth site, preserving crystallographic alignment. (A)

(B)

(C)

Figure 6.3. Various umbrella microstructures. (A) Sharp and lamellar, (B) Particulate, (C) Intermediate, PILP-like. Scale bars 10μm. The modular process is supported by DLS data of PAA/Ca2+/CO32- solutions: while lower pH solutions contain 5nm scattering entities, solutions around the pH where much of the growth occurs (~8.3-8.5) contain “precursors,” which are roughly the same size as the observed layer spacings (200-300nm) (Fig. 6.4).

193

Given the

structure of the umbrellas and the correlation with precursor size, the most reasonable mechanism is a modular one. Therefore, it becomes important to understand the nature of these precursors. What is their composition? How hydrated are they? What state of matter are they? If they are liquid, what is the diffusion coefficient inside of them? How do such properties change with parameters like pH and temperature? What kinds of polymers are able to induce them, and how do different polymers change their properties? What emergent properties do they have, as in their abilities to control crystallization? Size Distribution by Volume

1200 1100

(A)

(B)

50

1000 900

Size E

800

40

7.8

Volume (%)

Size (nm)

700 600 500 400

8.1

8.2

30

8.3

20

300 200

8.35 8.4

10

100 0

0

-100 7.8

8.0

8.2

8.4

8.6

1

10

100

1000

Size (d.nm)

pH

Figure 6.4. (A) Umbrella precursor sizes versus pH. (B) Size distributions by volume at various pHs. If interfacial growth is prevented, by keeping precursors dispersed as in these experiments, the precursors continue to grow and precipitate in indistinct aggregates. Precursor mechanisms are not new in the field of calcium carbonate mineralization. The above-mentioned PILP process is one example.

Another proposed mechanism for certain precursors is

oriented attachment, in which small crystallites contain polymer that is preferentially adsorbed onto one (or more) specific crystal faces. The crystallites are free to diffuse and eventually orient themselves to one

194

another, perhaps under the control of their local electrostatic forces, and then ‘sinter’ once they are in alignment. Such a process certainly occurs in some systems (for instance TiO2 nanoparticles16); however, the model does not describe the probability of alignment or the effect of crystal defects on two adjoining crystals, especially as would occur on the scale of 100-500 nm. Furthermore, Cölfen has also studied the crystallinity of precursors and found that they are amorphous for an extended period.17 Other attempts to study the precursors, as by TEM or filtration, have revealed that they are inherently unstable and transient. In fact, a sensible mechanism is suggested by considering the umbrella precursors as metastable amorphous modules that proceed to construct highly oriented structures. Such an idea is also in agreement with reports that biology controls the formation of amorphous calcium carbonate, either transiently or in storage, for use as precursor material to intricate crystalline structures.10, 18-21 It is furthermore in agreement with the suggestion that, prior to crystallization, polymers exist as hydrated gels rather than as crystalline substrates.22 In this proposed mechanism, the polymer is responsible for sequestering amorphous nuclei into discrete packages, increasing local supersaturation, and locally inhibiting the drive crystallization. The precursors are free to diffuse until they contact an exposed crystal face,

195

upon which this new substrate induces oriented crystallization within the precursor, similar to the epitaxial process described by the PILP SLS mechanism.

As crystallization proceeds, the precursor phase

separates, excluding many of the polymers to the exterior (Scheme 1). However, some of the polymers naturally become occluded and interrupt the crystallization process such that it occurs in smaller “granular” components.

Meanwhile, both occluded polymers and

surface segregated polymers become oriented according to the crystal lattice, adopting a structure and orientation to conform within it. Polymers that are unable to attain a proper conformation for a given polymorph may inhibit the propagation of the crystal and instead begin to change the crystal structure, as it accords with lattice matching.

(B) (A) (C) Scheme 6.1. Proposed mechanism of nacre formation, based on precursors and phase-separation. In this way, the kinetics of the polymer and crystallization cooperatively control one another. Crystallization proceeds to the other side of the precursor in this fashion, and the excluded polymer then either binds thermodynamically to certain surfaces or else is released back into solution (explaining the excess of polymer observed in

196

umbrella solutions). The process then results in a new exposed crystal face for the next precursor to bind.

Because umbrella layers are

aligned, it can be inferred that adsorbed polymers do not provide complete surface coverage, and that the incidentally exposed crystal surface edges can behave as “bridges” to propagate crystal orientation to the next layer. As described, this interfacial mechanism can describe several details of the composite structures: preserved crystal orientation, periodic/lattice alignment of soluble matrix polymers with crystal interfaces, occluded polymers, and granular crystallites. explains other important observations (Fig. 6.1).

It also

For instance, the

periodic nature and crystal orientation of such structures can be seen as a natural consequence of the growth, as controlled by polymer adsorption and precursor attachment.

The monodispersity layer

thicknesses is the result of two-fold control: by the size of the precursors and by the size of the first, initiating crystal face of each layer. Additionally, of great importance is the emergence of the layered structure by phase separation. This mechanism describes a method of step-by-step growth of an organic matrix as occurring in concert with crystal growth, explaining a method of achieving inorganic/organic layering without explicit sequential control.

197

This suggested model combines several pertinent observations: the amorphous nature of previously observed precursors, the use of transient ACC in nature prior to complex crystallization, the observation of liquid (and therefore amorphous) PILPs, and the proximity of the conical phases to amorphous ones in the ternary diagram. The phase separation/amorphous precursor mechanism also echoes some basic biochemical principles, for instance the preservation of a high energy and metastable state that organizes in response to a bistable switch.

Based on such principles, phase separation in

biomineralization has been previously suggested by Sumper in an effort to explain the patterning achieved by diatoms.23 Similar to the model proposed here, a mix of polyelectrolyte precursors cause a phase separation or organic and inorganic material, followed by successive splittings as surface tension is modulated. However, in contrast to the mechanism proposed here, the diatom model is based on patterning on organic, coacervate-like spherical templates, whereas the CaCO3 model requires no template, per se, and organizes in layers according to crystal orientation.

198

Remaining Questions Because

precursors

are

induced

by

certain

ordinary

polyelectrolytes, it is likely that they occur during nacre formation with at least one of the known soluble proteins. What remains uncertain is whether the formations participate in the process theorized here, or whether such a process dominates biomineralization. If nacre does form according to this suggested mechanism, there are of course further questions: 1. Can these precursors be observed in the extrapallial fluid? In vitro? 2. How do their sizes and compositions correlate with the final product? 3. How do the insoluble matrix proteins participate in the growth process? Sequentially? Prepackaged as precursor stabilizers? 4. Are nacreous precursors stable indefinitely, or do they solidify (less reversibly) as a function of time/size/pH? 5. What specific polymer attributes (e.g., range of binding energies) are required to initiate such a process? 6.

Do multiple polymers affect the process?

7. Do precursors have a measurable or calculable surface tension, and can such properties predict interaction with various biological or synthetic templates? Of course, these questions extend beyond nacre to other biomineralization systems, as well. As observed by the morphogenesis with the ternary phase diagram approach, precursors need not

199

crystallize into lamellar or even period structures. The longitudinal growth of the fibrous cones, and the rods formed in the similar PILP process, represent one way for precursors to create extended structures (such as spicules) in a very different way than would be found in classical crystallization. The precursors may also be endowed with properties that create favorable interaction with biological templates in 3d structures that are less restricted to orientational growth but still preserve crystallographic orientation.

Such patterning around 3d

objects has been observed for CaCO3 systems without complex additives,24 but may be enhanced in biology by the influence of organic additives.

Conclusion I have described a conceptual model for a mineralization mode that explains features of the umbrella structure and possibly of nacre. The model explains fundamentally unanswered questions about the layered construction of nacre, such as the deposition of matrix layers and uniform periodicity.

It furthermore implies answers to more

detailed observations, such as epitaxial adsorption, mineral bridging, and granular crystallites. Because it more simply describes natural features, accounts for various experimental in vitro observations, and leaves fewer critical unanswered questions, it is an altogether better

200

model to scientifically investigate nacre formation, according to Occam’s razor and logical positivism. Finally, this proposal yields well defined scientific questions for future investigations—an often lacking attribute of the biomineralization field—and hence favorably lends itself to testing and scientific progress.

REFERENCES

1.

Mann, S., Biomineralization: Principles and Concepts in

Bioinorganic Materials Chemistry. Oxford University Press, Inc.: 2001; p 198. 2.

Simkiss, K.; Wilbur, K. M., Biomineralization: Cell Biology

and Mineral Deposition. Academic Press, Inc.: San Diego, California 92101, 1989; p 337. 3.

Currey, J. D., Mechanical-Properties of Mother of Pearl in

Tension. Proceedings of the Royal Society of London Series BBiological Sciences 1977, 196, (1125), 443-&. 4.

Feng, Q. L.; Li, H. B.; Cui, F. Z.; Li, H. D.; Kim, T. N., Crystal

orientation domains found in the single lamina in nacre of the Mytilus edulis shell. Journal of Materials Science Letters 1999, 18, (19), 15471549.

201

5.

Metzler, R. A.; Abrecht, M.; Olabisi, R. M.; Ariosa, D.;

Johnson, C. J.; Frazer, B. H.; Coppersmith, S. N.; Gilbert, P., Architecture of columnar nacre, and implications for its formation mechanism. Physical Review Letters 2007, 98, (26). 6.

Checa, A. G.; Rodriguez-Navarro, A. B., Self-organisation of

nacre in the shells of Pterioida (Bivalvia : Mollusca). Biomaterials 2005, 26, (9), 1071-1079. 7.

Addadi, L.; Joester, D.; Nudelman, F.; Weiner, S., Mollusk shell

formation:

A

source

of

new

concepts

for

understanding

biomineralization processes. Chemistry-a European Journal 2006, 12, (4), 981-987. 8.

Nudelman, F.; Gotliv, B. A.; Addadi, L.; Weiner, S., Mollusk

shell formation: Mapping the distribution of organic matrix components underlying a single aragonitic tablet in nacre. Journal of Structural Biology 2006, 153, (2), 176-187. 9.

Li, X. D.; Xu, Z. H.; Wang, R. Z., In situ observation of

nanograin rotation and deformation in nacre. Nano Letters 2006, 6, (10), 2301-2304. 10.

Nassif, N.; Pinna, N.; Gehrke, N.; Antonietti, M.; Jager, C.;

Colfen, H., Amorphous layer around aragonite platelets in nacre. Proceedings of the National Academy of Sciences of the United States of America 2005, 102, (36), 12653-12655.

202

11.

Schaffer, T. E.; IonescuZanetti, C.; Proksch, R.; Fritz, M.;

Walters, D. A.; Almqvist, N.; Zaremba, C. M.; Belcher, A. M.; Smith, B. L.; Stucky, G. D.; Morse, D. E.; Hansma, P. K., Does abalone nacre form by heteroepitaxial nucleation or by growth through mineral bridges? Chemistry of Materials 1997, 9, (8), 1731-1740. 12.

Belcher, A. M.; Wu, X. H.; Christensen, R. J.; Hansma, P. K.;

Stucky, G. D.; Morse, D. E., Control of crystal phase switching and orientation by soluble mollusc-shell proteins. Nature 1996, 381, (6577), 56-58. 13.

Checa, A. G.; Jimenez-Lopez, C.; Rodriguez-Navarro, A.;

Machado, J. P., Precipitation of aragonite by calcitic bivalves in Mgenriched marine waters. Marine Biology 2007, 150, (5), 819-827. 14.

Fu, G.; Qiu, S. R.; Orme, C. A.; Morse, D. E.; De Yoreo, J. J.,

Acceleration of calcite kinetics by abalone nacre proteins. Advanced Materials 2005, 17, (22), 2678-+. 15.

Imai, H.; Tatara, S.; Furuichi, K.; Oaki, Y., Formation of

calcium phosphate having a hierarchically laminated architecture through

periodic

precipitation

in

organic

gel.

Chemical

Communications 2003, (15), 1952-1953. 16.

Penn, R. L.; Banfield, J. F., Morphology development and

crystal growth in nanocrystalline aggregates under hydrothermal

203

conditions: Insights from titania. Geochimica Et Cosmochimica Acta 1999, 63, (10), 1549-1557. 17.

Bolze, J.; Pontoni, D.; Ballauff, M.; Narayanan, T.; Colfen, H.,

Time-resolved SAXS study of the effect of a double hydrophilic blockcopolymer on the formation of CaCO3 from a supersaturated salt solution. Journal of Colloid and Interface Science 2004, 277, (1), 8494. 18.

Amos, F. F.; Sharbaugh, D. M.; Talham, D. R.; Gower, L. B.;

Fricke, M.; Volkmer, D., Formation of single-crystalline aragonite tablets/films via an amorphous precursor. Langmuir 2007, 23, (4), 1988-1994. 19.

Gehrke, N.; Nassif, N.; Pinna, N.; Antonietti, M.; Gupta, H. S.;

Colfen, H., Retrosynthesis of nacre via amorphous precursor particles. Chemistry of Materials 2005, 17, (26), 6514-6516. 20.

Politi, Y.; Arad, T.; Klein, E.; Weiner, S.; Addadi, L., Sea

urchin spine calcite forms via a transient amorphous calcium carbonate phase. Science 2004, 306, (5699), 1161-1164. 21.

Beniash, E.; Aizenberg, J.; Addadi, L.; Weiner, S., Amorphous

calcium carbonate transforms into calcite during sea urchin larval spicule growth. Proceedings of the Royal Society of London Series BBiological Sciences 1997, 264, (1380), 461-465.

204

22.

Levi-Kalisman, Y.; Falini, G.; Addadi, L.; Weiner, S., Structure

of the nacreous organic matrix of a bivalve mollusk shell examined in the hydrated state using Cryo-TEM. Journal of Structural Biology 2001, 135, (1), 8-17. 23.

Sumper, M., A phase separation model for the nanopatterning of

diatom biosilica. Science 2002, 295, (5564), 2430-2433. 24.

Wucher, B.; Yue, W. B.; Kulak, A. N.; Meldrum, F. C.,

Designer crystals: Single crystals with complex morphologies. Chemistry of Materials 2007, 19, (5), 1111-1119.

205