Molecular Manipulation of Microstructures: Biomaterials ...

diffusional rewelding of the damage, thus pro-viding the ultimate biomimetic property ofself healing.

Thermodynamic compatibility of thepair of two-phase alloys requires a four-phase equilibrium at operating temperaturesand a two-phase equilibrium during solu-tion treatment. A preliminary thermody-namic feasibility analysis, including assess-ment of memory alloy stability require-ments, was performed by a team of juniorsin materials design class. Continued evalu-ation (38) has included a test of mechanicalconcepts that uses a TiNi-reinforced Snalloy composite prototype to demonstrateboth macroscopic strain reversal and thedesired crack-clamping behavior (41). Pre-cise multicomponent phase relations for theFe-based system have been evaluated withdiffusion couple experiments, and prototypesteel composites are being fabricated.

The success of these initial designs sug-gests that the integration of computationalmaterials science within a systems engineer-ing framework offers a powerful new ap-proach for the creation of superior materialsthat have sophisticated control of a multi-level dynamic structure, combined with re-duced time and cost of materials develop-ment. These first steps herald a new synergyof the science and engineering of materials.

REFERENCES AND NOTES___________________________

1. G. B. Olson, M. Azrin, E. S. Wright, Eds., Innovationsin Ultrahigh-Strength Steel Technology [GovernmentPrinting Office (GPO), Washington, DC, 1990].

2. C. S. Smith, A Search for Structure (MIT Press, Cam-bridge, MA, 1981).

3. iiii, in Martensite, G. B. Olson and W. S. Owen,Eds. (ASM International, Materials Park, OH, 1992),pp. 21–39.

4. C. Zener, Elasticity and Anelasticity of Metals (Univ.of Chicago Press, Chicago, IL, 1948).

5. M. Cohen, Mat. Sci. Eng. 25, 3 (1976).6. G. B. Olson, in M. E. Fine Symposium, P. K. Liaw, J. R.

Weertman, H. L. Markus, J. S. Santner, Eds. ( The Min-erals, Metals & Materials Society (TMS) of the AmericanInstitute of Mining, Metallurgical, and Petroleum Engi-neers (AIME), Warrendale, PA, 1991), p. 41.

7. G. M. Jenkins, in Systems Behaviour, J. Beishon andG. Peters, Eds. (Harper and Row, London, for OpenUniv. Press, 1972), pp. 56–82.

8. M. F. Ashby, Materials Selection in Mechanical De-sign (Pergamon, Tarrytown, NY, 1992).

9. COSMAT Summary Report, Materials and Man’sNeeds (National Academy of Sciences, NationalAcademy Press, Washington, DC, 1974); MaterialsScience and Engineering for the 1990s (National Re-search Council, National Academy Press, Washing-ton, DC, 1989).

10. N. P. Suh, The Principles of Design (Oxford Univ.Press, New York, 1990); L. D. Albano and N. P. Suh,Res. Eng. Des. 4, 171 (1992).

11. B. Sundman, B. Jansson, J. O. Andersson, CALPHAD9, 153 (1985).

12. G. Ghosh and G. B. Olson, Acta Metall. Mater. 42,3361 (1994).

13. G. R. Speich, in Innovations in Ultrahigh-Strength SteelTechnology, G. B. Olson, M. Azrin, E. S. Wright, Eds.(GPO, Washington, DC, 1990), pp. 89–111.

14. K. C. King, P. W. Voorhees, G. B. Olson, Metall.Trans. 22A, 2199 (1991).

15. A. J. Allen, D. Gavillet, J. R. Weertman, Acta Metall.41, 1869 (1993).

16. G. B. Olson, T. J. Kinkus, J. S. Montgomery, Surf.Sci. 246, 238 (1991).

17. A. Umantsev and G. B. Olson, Scr. Metall. 29, 1135(1993).

18. J. G. Cowie, M. Azrin, G. B. Olson, Metall. Trans. A2A, 143 (1989).

19. A. Needleman, in Innovations in Ultrahigh-StrengthSteel Technology, G. B. Olson, M. Azrin, E. S.Wright, Eds. (GPO, Washington, DC, 1990), pp.331–346.

20. Y. Huang and J. W. Hutchinson, in Modelling ofMaterial Behavior & Design, J. D. Embury, Ed. ( TMS-AIME, Warrendale, PA, 1990), pp. 129–148.

21. M. J. Gore, G. B. Olson, M. Cohen, in Innovations inUltrahigh-Strength Steel Technology, G. B. Olson,M. Azrin, E. S. Wright, Eds. (GPO, Washington, DC,1990), pp. 425–441.

22. G. B. Olson, J. Phys. V, Colloque C1, 407 (1996).23. R. G. Stringfellow, D. M. Parks, G. B. Olson, Acta

Metall. 40, 1703 (1992).24. S. Socrate, thesis, Massachusetts Institute of Tech-

nology (1996).25. C. J. Kuehmann, J. Cho, T. A. Stephenson, G. B.

Olson, in Metallic Materials for Lightweight Applica-tions, E. B. Kula and M. G. H. Wells, Eds. (GPO,Washington, DC, 1994), pp. 337–355.

26. C. J. McMahon Jr., in Innovations in Ultrahigh-Strength Steel Technology, G. B. Olson, M. Azrin,E. S. Wright, Eds. (GPO, Washington, DC, 1990), pp.597–618.

27. J. R. Rice and J.-S. Wang, Mater. Sci. Eng. A 107,23 (1989).

28. R. Wu, A. J. Freeman, G. B. Olson, Science 265, 376(1994).

29. iiii, Phys. Rev. B 53, 7504 (1996).30. L. Zhong, R. Wu, A. J. Freeman, G. B. Olson, ibid.

55, 11 135 (1997).31. iiii, unpublished results32. P. M. Anderson, J. S. Wang, J. R. Rice, in Innova-

tions in Ultrahigh-Strength Steel Technology, G. B.Olson, M. Azrin, E. S. Wright, Eds. (GPO, Washing-ton, DC, 1990), pp. 619–649.

33. J. F. Watton, G. B. Olson, M. Cohen, in Innovationsin Ultrahigh-Strength Steel Technology, G. B. Olson,M. Azrin, E. S. Wright, Eds. (GPO, Washington, DC,1990), pp. 705–737.

34. G. Ghosh and G. B. Olson, in Proceedings of 51stAnnual Meeting of the Microscopy Society of Amer-ica, G. W. Bailey and C. L Rieder, Eds. (San Fran-cisco Press, San Francisco, CA, 1993).

35. T. A. Stephenson, C. E. Campbell, G. B. Olson, Ad-vanced Earth-to-Orbit Propulsion Technology 1992,R. J. Richmond and S. T. Wu, Eds., NASA Conf.Publ. 3174 (1992), vol. 2, pp. 299–307.

36. J. Wise, unpublished results.37. G. B. Olson and H. Hartman, J. Physique 43, C4–

855 (1982).38. B. Files, thesis, Northwestern University (1997).39. G. B. Olson, K. C. Hsieh, H. K. D. H. Bhadeshia, in

Microstructures LCS ’94 (Iron and Steel Institute ofJapan, Tokyo, 1994).

40. The Northwestern component of the SRG programhas been sponsored by the Office of Naval Re-search, the Army Research Office, NSF, NASA, theU.S. Department of Energy, the Electric Power Re-search Institute, and the Air Force Office of ScientificResearch, with industry gifts and fellowship support.

Molecular Manipulation ofMicrostructures: Biomaterials,Ceramics, and Semiconductors

Samuel I. Stupp* and Paul V. Braun

Organic molecules can alter inorganic microstructures, offering a very powerful tool forthe design of novel materials. In biological systems, this tool is often used to createmicrostructures in which the organic manipulators are a minority component. Threegroups of materials—biomaterials, ceramics, and semiconductors—have been select-ed to illustrate this concept as used by nature and by synthetic laboratories exploringits potential in materials technology. In some of nature’s biomaterials, macromoleculessuch as proteins, glycoproteins, and polysaccharides are used to control nucleation andgrowth of mineral phases and thus manipulate microstructure and physical properties.This concept has been used synthetically to generate apatite-based materials that canfunction as artificial bone in humans. Synthetic polymers and surfactants can alsodrastically change the morphology of ceramic particles, impart new functional proper-ties, and provide new processing methods for the formation of useful objects. Interestingopportunities also exist in creating semiconducting materials in which molecular ma-nipulators connect quantum dots or template cavities, which change their electronicproperties and functionality.

The functionality of materials in macro-scopic form is seldom achieved with purechemical compounds that form single crys-

tals. Many of nature’s remarkable materialscontain mixtures of molecules or micro-structures in which inorganic crystals orglasses coexist with organic molecules. Ex-amples include bone, cartilage, shells,leaves, and skin. Here, we address the con-cept of molecular manipulation of micro-structures in inorganic materials, a biologi-cally inspired synthetic tool for the era of

The authors are in the Department of Materials Scienceand Engineering, Department of Chemistry, BeckmanInstitute for Advanced Science and Technology, Materi-als Research Laboratory, University of Illinois at Urbana-Champaign, Urbana, IL 61801, USA.

*To whom correspondence should be addressed.

SCIENCE z VOL. 277 z 29 AUGUST 1997 z www.sciencemag.org1242

materials by design. As discussed here, mo-lecular manipulation of materials impliesthe use of organic molecules, even in verysmall amounts, to control the microstruc-tures of inorganic solids. This article focuseson materials in which organic molecules arethe minority component; examples of thesesystems can be found in nature, but thepotential exists to discover synthetic ana-logs of technological interest. The inversesystems (which are predominantly organic)or those with comparable amounts of bothtypes are certainly interesting but resemblemore the conventional polymer compositesthat have been investigated over the pastfew decades.

Three groups of materials have been se-lected to illustrate the microstructural con-cept of molecular manipulation—biomate-rials, ceramics, and semiconductors. Bioma-terials are defined here as either naturallyoccurring materials in living organisms ormaterials designed to repair humans. Asdiscussed below, organic molecules imparttoughness to otherwise brittle mineralstructures in many organisms (1), and thussynthetic minerals that use molecular ma-nipulators could be excellent candidates forbone replacement in humans. However,there is no reason to limit the role of or-ganic molecules in such systems to tough-ening functions. Organic molecules couldalso be used to synthesize highly functionalminerals, for example, implants that wouldcarry critical therapeutic agents or mole-cules such as growth factors that would beuseful in tissue engineering.

The ceramic group generally encompass-es chemically resistant materials that aredesigned to withstand elevated tempera-tures but are usually brittle in nature. Thus,a minority component of organic materialin ceramics could also play a tougheningrole in brittle microstructures but could re-duce the high thermal resistance of thesematerials. If extremely high temperature useis not required, organic manipulators couldadd a great deal of functionality to ceramicmaterials. Of course, organics have beenused over the past two decades as the pre-cursors of ceramics in the so-called chemi-cal routes to ceramics (such as sol-gel syn-theses) (2, 3). Such chemical routes to ce-ramics have offered a great deal of syntheticand processing flexibility in this group ofmaterials. A different role for organics inceramic microstructures would be as manip-ulators of morphology, leaving an imprint oftheir original presence even after their dis-appearance at high temperatures (4–6).

In the field of semiconductors, micro-structural manipulation with organics is afield that is just emerging but has enormouspotential. One could envision organicsserving a templating role to access a specific

morphology in inorganic semiconductors(7), or one could disperse organic moleculeswith an electronic or photonic function insemiconducting microstructures. Organicmolecules could also help organize semicon-ducting nanocrystals into functional macro-scopic structures (8). We explore here ex-amples in these three areas, which involvework performed in various laboratories, in-cluding our own.

Biomaterials

Among mineralized biological materials,one finds magnificent examples of micro-structures in which only a small content oforganic matter plays a key role in the de-termination of properties. Weiner andAddadi (1) have recently reviewed thisfield and described many examples of theseremarkable materials synthesized at ambi-ent temperatures and pressures by varioustaxonomic groups. Nature’s mineralizers usesmall amounts of organic macromoleculesto manipulate nucleation, growth, micro-structure, and, consequently, the propertiesof their mineral-based materials. Oneexample in mammals is the microstructureof tooth enamel in the incisor of a rat. Themicrostructure, shown in Fig. 1C, containsrods composed of hundreds of spaghetti-

shaped crystals of carbonated apatite. Thismicrostructure resembles the cross-ply con-figuration of some advanced composites for-mulated with carbon fibers in polymer ma-trices. Two other fascinating examples,which are also shown in Fig. 1, are thespongy ventral plate of a starfish, varying intexture at different sites (Fig. 1B), and thesea urchin spine (Fig. 1A), both of whichare composed entirely of a single crystal ofcalcite. The sea urchin single crystal maycontain as little as 0.02 weight % glycopro-tein, but this small content of organic mat-ter remarkably enhances the mineral’s resis-tance to fracture (9, 10). According toWeiner and Addadi (1), the manner inwhich the mineral phase and the organicmaterial are organized is one of the keyfactors contributing to the distinctive me-chanical properties in these biomaterials.They suggest that this could be part of theorganism’s strategy to create more isotropicproperties from inherently anisotropic com-ponents such as one-dimensional macro-molecules and crystalline minerals. This isin fact a very important goal in the designand engineering of composite materials.

The minerals and macromolecules usedby nature vary greatly, as do the structuralmotifs in which these hybrid materials areorganized. The macromolecules used by na-

Fig. 1. Electron micrographs of molecularly manipulated inorganic microstructures observed in nature.Two unusual calcite single crystals are observed in invertebrates: one is a spine with radial texture in thesea urchin (A) and the other is a sponge with different size pores in the arm of the star Ophiocoma wendti(B) (1) (reproduced by permission of the Royal Society of Chemistry). An example from mammals isobserved in the incisor of a rat, which generates crossed elongated crystals of carbonated apatitereminiscent of an advanced composite microstructure (C) (1) (reproduced by permission of the RoyalSociety of Chemistry). In certain plants, macromolecules can stabilize amorphous silica with a specificmorphology (D) [reproduced from (13) with kind permission from Elsevier Science Ltd, the Boulevard,Langford Lane, Kidlington 0X5 1GB, UK], and in the membrane of some bacteria, they form highlysymmetric templates for mineralization (E and F) [reproduced with permission from (17 )].

ARTICLES

www.sciencemag.org z SCIENCE z VOL. 277 z 29 AUGUST 1997 1243

ture in small quantities to manipulate mi-crostructure include proteins, glycoproteins,and polysaccharides (1). The macromole-cules tend to have common structural fea-tures, such as a high content of carboxylategroups (such as glutamic and aspartic acidresidues in proteins), which lead to interac-tions between organic chains and mineralprecursor ions and also help to attach themacromolecules to solid surfaces. With re-gard to minerals, two important ones foundin endoskeletons and exoskeletons are cal-cium carbonate polymorphs and calciumphosphates such as carbonated crystallineapatite. An interesting example of micro-structural manipulation may be found inthe nucleation of either the aragonitic orcalcitic polymorphs of calcium carbonate,which differ significantly in mechanicalproperties (11). In biological systems, mac-romolecules could also serve to control crys-tal growth and thus the object’s shape byadsorbing to specific crystal planes. Theentrapment of these macromolecules in themineral-based microstructure could thenincrease toughness in an otherwise brittlematerial. A possible toughening mechanismwould involve their deviation of cracks andabsorption of energy from the propagatingcrack (12). A remarkable example is thetoughness of the sea urchin spine shown inFig. 1A. Other examples of molecular ma-nipulation in biominerals could be related

to the stabilization of amorphous minerals[for example, silica (13), calcium carbonate(14), and calcium phosphate (15, 16)] andthe formation of microstructures in whichthere is long-range ordering of mineral par-ticles (13). Biogenic amorphous silica foundin branches, leaves, and hairs of certainplants my have microstructures with fi-brous, sheetlike, or globular morphology,and it is intimately associated with macro-molecular components such as proteins andcarbohydrates (13). A micrograph of amor-phous silica fibrils found in plants is shownin Fig. 1D. Another interesting system inthe present context is the mineralizationobserved on the cell membranes of somebacteria that can be equal to or exceed theircellular weight (17, 18). In some bacteria,the cell membrane contains a regular ar-rangement of proteins on its outer surface(S layer) that serves as a template for min-eralization (Fig. 1E). In these systems, theresulting mineral nanostructure has thesame symmetry and dimensions as the Slayer of proteins (see the example in Fig.1F).

Our laboratory explored nature’s use ofmacromolecules as manipulators of mineralmicrostructure to synthesize materials de-signed to serve as human artificial bone(19–21). We have termed these biomateri-als “organoapatites,” which are synthesizedby the nucleation and growth of hydroxy-

apatite mineral in aqueous solutions of or-ganic macromolecules, including ho-mopolymer poly(amino acids), low molarmass peptides, and synthetic polyelectro-lytes. The microstructure of these materialsis envisioned as apatite lattices threaded orsurrounded by very small amounts of mo-lecularly dispersed organics amounting toonly 2 to 3% of the total weight. Thesesmall amounts of organic content manipu-late dramatically their microstructure andphysical properties and in this regard bearsome resemblance to the previously dis-cussed calcium carbonates that are modifiedby occluded proteins. Furthermore, in thecase of organoapatites, the chemical struc-ture of the minority organic component canalso control the observed biological re-sponse when these materials are implantedin bone (21). As illustrated in Fig. 2, B andD, when organoapatites are synthesized indilute aqueous solutions of poly(L-lysine),one obtains two-dimensional single crystals(micrometers in cross section and nanome-ters in thickness), whereas small nanocrys-tals grow in the presence of similar solutionscontaining poly(L-glutamic acid) (Fig. 2, A

A

C D

B

Fig. 2. In synthetic organo-apatites, very small amountsof poly(amino acids) canmanipulate microstructureby forming either polycrys-talline aggregates of apatitenanocrystals, when poly(L-glutamic acid) is used as themanipulator (A and C), orlarge flat single crystals (mi-crometers in cross sectionand nanometers in thick-ness), when poly(L-lysine) ispresent in the mother liquor (B and D). Note the coherence between the apatite crystal lattice and theamino groups of the poly(L-lysine) chain in (B).

0 6.5

0 3.0

A

B

Fig. 3. The brittleness of an apatite-based artificialbone material is revealed by the common frag-mentation of cylindrical objects implanted in bone(A), whereas mechanically toughened implantscan be synthesized by manipulation of the miner-al’s growth and particle sintering with only 2 to 3weight % organic macromolecules (B) [see (20,21) for synthetic details and definition of the frag-mentation index, which is given on the right-handside of (A) and (B)].


and C) (22). We believe that this is ex-plained by the ability of poly(L-lysine) tofavor the growth of the ab planes of thecrystal. Glutamic acid residues, however,are possibly very effective at nucleatingcrystals. An illustration of the manipulationof physical properties is shown in Fig. 3,which reveals microscopic evidence of thehighly brittle behavior of materials ob-tained in the absence of organic contentand the toughened material formed by poly-ionic organoapatites. Whereas we cannotaccess for synthetic systems the capabilitiesof living cells in biogenic microstructures, itis likely that we may discover fascinatingtools for microstructural control in cell-seeded forms of these materials in theemerging biomaterials approach known astissue engineering (23, 24). In this ap-proach, biodegradable synthetic materialsacting as scaffolds are seeded with cells toinduce the regeneration of natural tissues.

An exciting future prospect in the con-text of organoapatite biomaterials and othermolecularly manipulated inorganics is theuse of the organic content to create highlyfunctional minerals. We recently generatedan example by synthesizing an organoapa-tite containing a molecule designed by us

that contains three structural motifs varyingin function: the anti-inflammatory drug in-domethacin; the amino acid tyrosine,which is potentially adhesive toward solidsurfaces; and the precursor monomer to thebiocompatible polymer poly(2-hydroxy-ethyl methacrylate) (25, 26). One couldenvision future examples of designed func-tional minerals containing organic mole-cules, which may include ceramic materialsto be used at low temperatures and evensemiconductors, in which the organic com-ponent contributes an electronic or pho-tonic function in the inorganic lattice.

Ceramics

Many of the metal-nonmetal structuresknown as ceramic materials have tradition-ally been synthesized by fusion or sinteringof complex mixtures of inorganic com-pounds such as metal oxides (27). Over thepast few decades, new possibilities for theirsynthesis have been identified by the use ofchemical routes, including the sol-gel meth-ods for glasses and ceramics (2). In thesemethods, the synthesis of the inorganic net-works begins at low temperatures with con-densation polymerization of multifunc-

tional metal alkoxide monomers and endswith calcination aimed at the removal ofthe organic by-products of polymerization,solvents, residual monomer, and other or-ganics. There is no question that thesemethods offer the possibility of more chem-ically homogeneous ceramics, new forms offluid precursors, and particle shape control,which is important in further processing.Other approaches in which designed inor-ganic clusters are synthesized as buildingblocks for a self-assembling approach to ce-ramics may eventually emerge as importantmethods. The concept considered here,however, is the possibility of using organicmolecules to drastically change the micro-structure of the inorganic lattice of a ceram-ic material. Some examples are describedbelow.

An interesting recent discovery aboutmolecular manipulation in ceramic micro-structures is their sol-gel synthesis mediatedby organic surfactants, which leads to me-soporous materials (4, 28–32). The meso-porous microstructure obtained by thismethod reflects the order parameter of me-sophases formed by the self-organization oforganic and inorganic components duringsynthesis. Over the past few years, manymesoporous ceramics have been synthe-sized, including aluminosilicate (4), titani-um silicate (33), zinc phosphate (30), andmanganese oxide (34), achieving control ofchemical composition, microstructure, andgross morphology (35). These materials canbe grown as particles (4), free-standingfilms (36), films on a variety of substrates(32), and complex morphologies (37). Awide range of chemistries has been used forthe inorganic precursors and the surfac-tants, but many use silica precursors such astetramethylammonium silicate or colloidalsilica (or both) and a cationic surfactant(38). Presumably, the mesoporous micro-structure emerges as supramolecular assem-blies form that contain the cationic surfac-tant and the growing metal oxide mole-cules. It is believed that the nanoscale andmicroscale symmetry and dimensions of theinorganic phase are controlled by molecularpacking constraints and growth rates of theinorganic phase. This method results instructures controlled mostly by kinetics andnot thermodynamics (37). Periodic arrays ofdesigner pores imprinted by the surfactantin important materials such as silica maylead to important functional ceramics.These functional ceramics may include in-teresting catalytic functions and environ-mental waste remediation properties chem-ically engineered into the mesoporous mi-crostructure (33, 34, 39).

Self-assembled monolayers (SAMs) (40)provide the opportunity to molecularly ma-nipulate the microstructure of inorganic

PVA organoceramic

Ceramic

18 Å

7.8 Å

3 15 272U (degrees)

39 51 60

A

C

B

Fig. 4. Electron micrographs revealing the morphological contrast be-tween the lamellar microstructure of the layered double hydroxide[Ca2Al(OH)6]1 [(OH) z 3H2O]2 (A) and the rosette-shaped particles ob-tained for the same system when the polymer poly(vinyl alcohol) isdissolved in the precursor solutions of this liquid phase synthesis (B).Molecular graphics (in color) illustrate the expected differences innanoscale structure. (C) The x-ray scans reveal a new spacing betweeninorganic layers that can be explained by the confinement of polymer inthe microstructure.

ARTICLES


materials when they are used as growthsubstrates (41). Selection of specific mole-cules to create the SAM can yield surfacesdesigned to interact with specific mineralphases, such as sulfonated SAMs for thegrowth of FeOOH. Through careful control

of solution chemistry and supersaturation,mineral phase growth with specific crystal-lographic orientation can be induced toprecipitate on almost any substrate onwhich SAMs can be deposited. By pattern-ing the SAM on which the mineral phase

will grow, it is even possible to write inor-ganic microdomains (42). We describe be-low another concept in which organic mac-romolecules modify inorganic microstruc-tures and form composite materials.

A different concept that has beenexplored in our laboratory is the precip-itation from solution of calcium andaluminum layered double hydroxides, suchas [Ca2Al(OH)6]1 [(OH)z3H2O]2, in thepresence of organic polymers. This pro-cedure generates organoceramics in whichthe organic macromolecules significantlychange the morphology and properties (5,6, 43). Micrographs of the flake-like mor-phology obtained in the absence of organicsand the very different rosette-shaped com-posite particles obtained in the presence ofpoly(vinyl alcohol) macromolecules areshown in Fig. 4, A and B, respectively. Theorganic-inorganic composite rosettes arefairly monodisperse in size, and these ruf-fled-surface particles have an Al/Ca ratiothat is essentially identical to that of thefully inorganic material and an organic con-tent of ;20 weight %. The mechanicalproperties of bulk materials pressed at roomtemperature are significantly improved inthe organoceramic composition (44).

We believe that the molecular manipula-tion of microstructure occurs here through therole of polymer chains in nucleation of theinorganic layers. Furthermore, we believethat the polymer becomes intercalated be-tween the inorganic principal planes toproduce expanded interlayer spacings.These expanded spacings revealed in x-raydiffraction (XRD) scans (see Fig. 4C) can-not be induced by simple diffusion of thesame macromolecules within the nanome-ter-scale spaces in a stage after precipita-tion. This barrier presumably arises throughthe large entropic barriers involved in mac-romolecular diffusion into such tight spac-es. Becze and Xu have argued that thisintercalation does not take place and thatthe polymer is merely involved in nucle-ation (45); however, we found that thedegree of intercalation was highly depen-dent on the synthetic scheme used (44). Onthe basis of XRD studies, we found that themolecularly manipulated microstructurecan expand and contract reversibly withexposure to moisture, suggesting possibili-ties for sensing minerals (44). Interestingly,an imprint of the spheroidal composite ro-settes remained in the microstructure afterorganic molecules had vanished and thematerial had been converted to CaO andcalcium aluminate ceramics by heating to1000°C (6). The rosette skeleton was clear-ly visible when the structure was almostentirely inorganic at 500°C. These trans-formations are illustrated by micrographsshown in Fig. 5.

Room temperature 500°C 1000°C

CalcinationCalcination

sintering

Fig. 5. The morphology of inorganic-organic composite particles is preserved as the system is calcinedto form a mostly inorganic material (500°C), and the particle’s spheroidal contour is still observed as theorganic content vanishes and the material transforms to ceramic by 1000°C.

Dots Antidots

Microstructures

Molecular manipulators

Layers

Fig. 6. Transmission electron micrographs of three molecularly manipulated microstructures of II-VIsemiconductors: a colloidal crystal of CdSe nanocrystals covered by organic surfactants, a colloid ofCd(1–x)ZnxS punctured by a regular array of cavities templated by cylindrical molecular assemblies ofsurfactant molecules, and a lamellar particle with alternating CdS and organic layers templated by anorganic lamellar mesophase. Below the micrographs, models and schematic representations of thehybrid structures and their respective molecular manipulators are shown in color.


Semiconductors

Given the well-known materials function-ality that is possible with solids of inter-mediate conductivity between metals andinsulators, it would be interesting to de-velop the field of molecular manipulationof semiconducting microstructures. Overthe past few years, some systems haveemerged that have begun to explore thisarea. One example is the synthesis of II-VIsemiconductor nanocrystals, which arehighly regular in size and shape, by the useof organic surfactants to control the pro-cess (8, 46–50). The surfactant moleculescoat the nanocrystals, which in turn canself-organize into highly ordered micro-structures analogous to colloidal crystals(8). These crystals of nanostructures havepotential as large arrays of quantum dotswith interesting electronic properties.

Bawendi and co-workers have achieveda very high degree of control over the size ofsemiconductor quantum dots, generatingorganically coated CdSe nanocrystallitesthat are monodisperse within the limit ofatomic roughness and soluble in commonorganic solvents (47). Because the semicon-ductor crystallites are all virtually the samesize, the dots can assemble into facetedcrystals of micrometer dimensions or as aclosely packed layer on an appropriate sub-strate (8). Molecular monolayers aroundthe nanocrystals prevent the formation ofdisordered structures, and self-organizationof the quantum dots occurs in a controlledfashion as the solubilizing power of thesolvent is decreased. This yields the super-lattice of quantum dots shown in Fig. 6.Along with the formation of these two- andthree-dimensional networks, semiconductorquantum dots can also be joined with spe-cific chemical linkers, forming, for example,dimers of CdSe with defined interparticlespacings (50). Langmuir monolayers werefirst used to manipulate the growth of inor-ganic and organic crystals (51) and morerecently as templates for semiconductorgrowth, resulting in crystallites shaped asrods, triangles, and even a continuous net-work (48). The driving force for the forma-tion of the observed structures is molecularrecognition between monolayer headgroups and specific crystallographic faces ofthe incipient semiconductor crystallite(52).

Our laboratory has studied a differenttype of molecular manipulation in II-VIsemiconductors that achieves the inverseof the quantum dot arrays, namely, a poly-crystalline semiconducting continuumwith periodic nanometer-sized cavitiestemplated directly by assemblies of organicmolecules. We initially reported (7) onthe synthesis of colloid-sized particles of

CdS that were punctured periodically bycylindrical cavities 2 to 3 nm in diameterand spaced 8 nm apart in a hexagonalarray. Recently, we observed improved fi-delity between the template and periodicnanometer-sized cavities by adjusting thereaction conditions (53). We achieved themolecular manipulation of the semicon-ducting microstructure by doping a hexag-onal liquid crystal with precursor Cd21

ions and then allowing precipitation tooccur by diffusion of H2S through thehighly ordered metal-doped gel. Weachieved direct templating by the gel,which was seen because the semiconduct-ing medium copied exactly the character-istic symmetry and dimensions of the lyo-tropic liquid crystal. We observed an ex-cellent example of this templating effectin a ternary system of Cd(1–x)ZnxS. Asshown in Fig. 6, hexagonal symmetry ofthe liquid crystal is expressed not only inthe symmetry and dimension of periodicnanocavities interrupting the semicon-ducting lattice but also in the shape of theparticle itself (54). The microstructure isgenerated because semiconductor growthis excluded from the nonpolar regions ofthe liquid crystal, that is, the cylindricalregions in which hydrophobic segmentsare confined.

Molecularly templated cavities in asemiconducting continuum are interestingfeatures of the microstructure because theycould serve many functions. Their presencecould produce a periodic array of antidotsthat could modify electronic properties ofthe semiconductor (55–57). Alternatively,the cavities could be used to selectivelyadsorb or transport molecules and also tochemically transform molecules diffusingthrough the cavities by taking advantage ofelectronic or photonic properties of thesemiconductor. Interestingly, depending onthe chemical nature of semiconductor pre-cursors, it is possible to anchor irreversiblythe templating molecules in the semicon-ducting lattice (58). This concept, illustrat-ed in Fig. 6, suggests that molecules withinteresting electronic or photonic charac-teristics could serve as templates but onceanchored in the inorganic medium couldmodify electronic properties such as elec-tron-hole recombination rates. This wouldrepresent an electronic analogy to the me-chanical function served by occluded pro-teins in biologically occurring crystals ofcalcium carbonate.

Inorganic-organic interactions in semi-conductor synthesis could open up new av-enues to novel microstructures and controlof properties. For example, semiconductor-metal transitions (59, 60), as well as absorp-tion and luminescence spectra (46, 49, 61,62), are molecularly modifiable. To date,

virtually all research on the design of semi-conductor devices has involved high-tem-perature and high-vacuum syntheses fol-lowed by patterning with conventionallithographic techniques. However, controlof dimensions in the 1- to 5-nm range, aswell as novel properties, may emerge withthe use of molecular engineering of organ-ics. Our own recent work on mushroom-shaped organic nanostructures could beadapted to template the synthesis of hugearrays of quantum dots with interesting mo-lecular connectors between them (63). Liq-uid crystals have potential for three-dimen-sional control in templated syntheses ofinorganics, but many challenges still needto be faced to achieve this goal. The poten-tial is based on the possibility of thermody-namic control of dimensions, symmetry,and internal structure through changes inmesogen chemical structure. We illustrate,for example, in Fig. 6 an additional systemstudied in our laboratory, a lamellar onebased on a liquid crystal formed by oligovi-nyl alcohol amphiphiles and water (64)generating alternating layers of inorganicsemiconductor and organic material.

Conclusions

The examples described in this article in-volving biomaterials, ceramics, and semi-conductors are all based on the use of rela-tively simple organic molecules to manipu-late microstructure. A rather extensive ter-ritory could still be explored if oneconsiders the possible use of molecules withcontrolled stereochemistry, block design,and backbone architecture. The potentialoutcome could be a toolbox of inorganicmaterials synthesized directly into specificmacroscopic or microscopic shapes, me-chanically versatile inorganic materialswith combinations of properties not yet of-fered by conventional composites, and anew set of functions that integrate the prop-erties of both components. It is certainlytrue that an era of designed hybrid materialsof the type discussed here has not yettouched technology and it is not at thepresent time a totally established chapter inmaterials science.

REFERENCES AND NOTES___________________________

1. S. Weiner and L. Addadi, J. Mater. Chem. 7, 689(1997).

2. C. J. Brinker and G. W. Scherer, Sol-Gel Science(Academic Press, New York, 1990).

3. J. Livage, M. Henry, C. Sanchez, Prog. Solid StateChem. 18, 259 (1988).

4. C. T. Kresge, M. E. Leonowicz, W. J. Roth, J. C.Vartuli, J. S. Beck, Nature 359, 710 (1992).

5. P. B. Messersmith and S. I. Stupp, J. Mater. Res. 7,2599 (1992).

6. iiii, Chem. Mater. 7, 454 (1995).7. P. V. Braun, P. Osenar, S. I. Stupp, Nature 380, 325

(1996).

ARTICLES


8. C. B. Murray, C. R. Kagan, M. G. Bawendi, Science270, 1335 (1995).

9. A. Berman, L. Addadi, S. Weiner, Nature 331, 546(1988).

10. A. Berman et al., Science 250, 664 (1990).11. G. Falini, S. Albeck, S. Weiner, L. Addadi, ibid. 271,

67 (1996).12. B. Lawn, Fracture of Brittle Solids (Cambridge Univ.

Press, Cambridge, 1993), chap. 6.13. C. C. Harrison, Phytochemistry 41, 37 (1996).14. J. Aizenberg, G. Lambert, L. Addadi, S. Weiner, Adv.

Mater. 8, 222 (1996).15. G. L. Becker, C. H. Chen, J. W. Greenwalt, A. L.

Lehninger, J. Cell Biol. 61, 316 (1974).16. A. S. Posner, F. Betts, N. C. Blumenthal, Metab.

Bone Dis. Relat. Res. 1, 179 (1978).17. S. Schultze-Lam, G. Harauz, T. J. Beveridge, J. Bac-

teriol. 174, 7971 (1992).18. S. Schultze-Lam, D. Fortin, B. S. Davis, T. J. Bever-

idge, Chem. Geol. 132, 171 (1996).19. S. I. Stupp and G. W. Ciegler, J. Biomed. Mater. Res.

26, 169 (1992).20. S. I. Stupp, G. C. Mejicano, J. A. Hanson, ibid. 27,

289 (1993).21. S. I. Stupp, J. A. Hanson, J. A. Eurell, G. W. Ciegler,

A. Johnson, ibid., p. 301.22. S. I. Stupp, L. S. Li, M. Keser, in preparation.23. N. A. Peppas and R. Langer, Science 263, 1715

(1994).24. J. A. Hubbell, Bio/Technology 13, 565 (1995).25. S. I. Stupp, A. Galan, M. Surhbur, S. Son, J. J.

Hwang, in preparation.26. J. J. Hwang, J. Hancock, K. Jaeger, S. I. Stupp, in

preparation.27. W. D. Kingery, H. K. Bowen, D. R. Uhlmann, Intro-

duction to Ceramics ( Wiley, New York, 1976).28. J. S. Beck et al., J. Am. Chem. Soc. 114, 10834

(1992).29. A. Monnier et al., Science 261, 1299 (1993).30. Q. Huo et al., Nature 368, 317 (1994).31. A. Firouzi et al., Science 267, 1138 (1995).32. H. Yang, A. Kuperman, N. Coombs, S. Mamiche-

Afara, G. A. Ozin, Nature 379, 703 (1996).33. P. T. Tanev, M. Chibwe, T. J. Pinnavaia, ibid. 368,

321 (1994).34. Z.-R. Tian et al., Science 276, 926 (1997).35. Many papers on mesoporous materials have been

published recently [for reviews, see J. S. Beck andJ. C. Vartuli, Curr. Opin. Solid State Mater. Sci. 1, 76(1996); J. Liu et al., Adv. Colloid Interface Sci. 69,131 (1996)].

36. H. Yang, N. Coombs, I. Sokolov, G. A. Ozin, Nature381, 589 (1996).

37. H. Yang, N. Coombs, G. A. Ozin, ibid. 386, 692(1997).

38. A typical surfactant used is cetyltrimethylammoniumchloride or other quaternary ammonium surfactantsof structure CnH2n11(CH3)3NX, where X is Cl or Brand n 5 8 to 22.

39. X. Feng et al., Science 276, 923 (1997).40. G. M. Whitesides, Sci. Am. 273, 146 (September

1995).41. B. C. Bunker et al., Science 264, 48 (1994).42. P. Rieke et al., Langmuir 10, 619 (1994).43. P. B. Messersmith and S. I. Stupp, Mater. Res. Soc.

Symp. Proc. 245, 191 (1992)..44. P. Osenar, P. B. Messersmith, S. I. Stupp, in

preparation.45. C. E. Becze and G. Xu, J. Mater. Res. 12, 566

(1997).46. N. Herron, J. C. Calabrese, W. E. Farneth, Y. Wang,

Science 259, 1426 (1993).47. C. B. Murray, D. J. Norris, M. G. Bawendi, J. Am.

Chem. Soc. 115, 8706 (1993).48. J. H. Fendler and F. C. Meldrum, Adv. Mater. 7, 607

(1995).49. T. Vossmeyer et al., Science 267, 1476 (1995).50. X. Peng, T. E. Wilson, A. P. Alivisatos, P. G. Schultz,

Angew. Chem. Int. Ed. Engl. 36, 145 (1997).51. E. M. Landau, Mol. Cryst. Liq. Cryst. 134, 323

(1986).52. Semiconductor growth can also be accomplished in

phase-separated diblock copolymers [see C. C.Cummins, R. R. Schrock, R. E. Cohen, Chem. Ma-ter. 4, 27 (1992)], but in these systems, the organic

matrix represents a significant volume fraction andcannot be removed without disrupting the micro-structure of the semiconductor. It is also possible toprepare them in reverse micelles [see M. P. Pileni, J.Phys. Chem. 97, 6961 (1993)].

53. V. Tohver, P. V. Braun, M. U. Pralle, S. I. Stupp,Chem. Mater. 9, 1495 (1997).

54. Other inorganics for which this approach has workedinclude CdSe and ZnS (P. V. Braun, P. Osenar, V.Tohver, S. B. Kennedy, S. I. Stupp, in preparation).

55. D. Weiss et al., Phys. Rev. Lett. 66, 2790 (1991).56. R. Fleischmann, T. Geisel, R. Ketzmerick, ibid.. 68,

1367 (1992).57. W. Kang, H. L. Stormer, L. N. Pfeiffer, K. W. Baldwin,

K. W. West, ibid. 71, 3850 (1993).58. Generally, the particles obtained in these syntheses

retain a small (#15% by volume) amount of organicmaterial after work-up, and because the particlesconsist of a continuum of inorganic material, they aremorphologically stable even after removal of most ofthe organic. Precipitation in the lamellar system gen-erates a composite structure that is about 50 volume% organic of alternating sheets containing CdS andamphiphile. The strong affinity of the polar oligovinylalcohol segments for the CdS results in a structurethat is stable to repeated sonication in good solvents

for the organic component, always retaining the la-mellar structure.

59. D. B. Mitzi, C. A. Feild, W. T. A. Harrison, A. M. Guloy,Nature 369, 467 (1994).

60. D. B. Mitzi, S. Wang, C. A. Feild, C. A. Chess, A. M.Guloy, Science 267, 1473 (1995).

61. Y. Tian, C. Wu, N. Kotov, J. H. Fendler, Adv. Mater.6, 959 (1994).

62. J. H. Fendler, Membrane-Mimetic Approach to Ad-vanced Materials (Springer-Verlag, Berlin, 1994).

63. S. I. Stupp et al., Science 276, 384 (1997).64. P. Osenar, P. V. Braun, S. I. Stupp, Adv. Mater. 8,

1022 (1996).65. Supported by NIH—National Institute of Dental Re-

search grant DE 05945, the Air Force Office of Sci-entific Research grant 90-0242, and the U.S. De-partment of Energy, Division of Materials Sciencegrant DEFG02-96ER45439 through the University ofIllinois at Urbana-Champaign, Frederick Seitz Mate-rials Research Laboratory. We thank L. Addadi, C. C.Harrison, and S. Schultze-Lam for providing theelectron micrographs in Fig. 1; M. G. Bawendi forproviding one of the micrographs in Fig. 6; and M.Keser of the authors’ laboratory for assistance withmolecular graphics. A Beckman Institute graduateassistantship for P.V.B. is gratefully acknowledged.

Pathways to Macroscale Orderin Nanostructured Block

CopolymersZhong-Ren Chen, Julia A. Kornfield,* Steven D. Smith,

Jeffrey T. Grothaus, Michael M. Satkowski

Polymeric materials undergo dramatic changes in orientational order in response todynamic processes, such as flow. Their rich cascade of dynamics presents opportunitiesto create and combine distinct alignments of polymeric nanostructures through pro-cessing. In situ rheo-optical measurements complemented by ex situ x-ray scatteringreveal the physics of three different trajectories to macroscopic alignment of lamellardiblock copolymers during oscillatory shearing. At the highest frequencies, symmetryarguments explain the transient development of a bimodal texture en route to thealignment of layers parallel to the planes of shear. At lower frequencies, larger-scalerelaxations introduce rearrangements out of the deformation plane that permit the for-mation of lamellae perpendicular to the shear plane. These explain the change in thecharacter of the pathway to parallel alignment and the emergence of perpendicularalignment as the frequency decreases.

Self-assembly of block copolymers (1–7),surfactants (8–11), colloidal suspensions (8,12), and proteins (13) provides a versatilemeans to create nanostructures with poten-tial applications in biomaterials, optics, andmicroelectronics. These materials form or-dered structures on scales from a few tohundreds of nanometers. Monodisperse,charged colloidal suspensions can assemblethree-dimensional lattices (8). Surfactant

systems form a variety of morphologies andcan be used as precursors to prepare nano-structured solid materials (10). Similarly,block copolymers (BCPs) assemble a fasci-nating array of nanostructures. BCPs havethe desirable feature that their morphologycan be systematically controlled by varyingthe number of blocks, their lengths, andtheir chemical compositions. For example,diblock copolymers can form cubic arrays ofspheres, hexagonal arrays of cylinders, bi-continuous cubic phases, or lamellae, de-pending on the relative block lengths (1).Triblock copolymers composed of three dis-tinct blocks (ABC) can assemble evenmore complex structures (14), such as heli-cal strands surrounding cylinders embedded

Z.-R. Chen and J. A. Kornfield, California Institute ofTechnology, Chemical Engineering 210–41, Pasadena,CA 91125, USA.S. D. Smith, J. T. Grothaus, M. M. Satkowski, Procterand Gamble, Miami Valley Laboratories, Cincinnati, OH45239, USA.

*To whom correspondence should be addressed.


Molecular Manipulation of Microstructures: Biomaterials ...

Documents