MNS 102: Techniques for Materials and Nanosciences

Sources and References

Primary textbook — Donald R. Askeland & Wendelin J. Wright, The Science and Engineering of Materials, 7th ed. (Cengage). Supplementary texts — Guozhong Cao & Ying Wang Nanostructures and Nanomaterials; Yang Leng Materials Characterization; Charles Poole & Frank Owens Introduction to Nanotechnology; Bharat Bhushan Springer Handbook of Nanotechnology; B.D. Cullity & S.R. Stock Elements of X-Ray Diffraction; David Williams & C. Barry Carter Transmission Electron Microscopy. Online resources — MIT OpenCourseWare materials science offerings; Cambridge DoITPoMS (doitpoms.ac.uk).

Chapter 1 — Processing of Metals: Strain Hardening, Recovery, Recrystallization, Annealing

Metals derive much of their practical usefulness from the fact that their mechanical response can be tuned by cold work and by heat treatment, without changing bulk composition. When a polycrystalline metal is plastically deformed below roughly one third of its absolute melting temperature, dislocations are generated and forced through the crystal lattice. They multiply, pile up at grain boundaries, and tangle into cell walls. Each new dislocation raises the stress needed to move the next one, because the elastic strain fields of neighbouring dislocations repel. This phenomenon, strain hardening (or work hardening), is quantified by the empirical relation \( \sigma = K\,\varepsilon^{n} \), where \( n \) is the strain-hardening exponent. Copper, brass and austenitic stainless steels harden strongly (\( n \) around 0.3–0.5); highly worked steels and many aluminum alloys harden less. The percent cold work, \( \%\mathrm{CW} = (A_{0}-A_{f})/A_{0} \times 100 \), controls how much yield strength, tensile strength and hardness rise, and how much ductility falls.

Cold work is reversible by heating. Recovery occurs first, at modest temperatures: point defects annihilate, dislocations climb and rearrange into lower-energy sub-grain walls (polygonization), and residual stresses relax. Electrical resistivity drops sharply, but hardness changes only slightly. Recrystallization follows at a higher temperature—the recrystallization temperature, typically \( 0.3\text{–}0.5\,T_{m} \) in kelvin—where entirely new, strain-free grains nucleate in regions of highest stored energy and grow until they consume the deformed matrix. Strength and hardness drop to their annealed values while ductility recovers. The driving force is the stored elastic energy of the dislocation forest; the higher the cold work, the lower the recrystallization temperature and the finer the initial grain size. Holding longer or hotter produces grain growth, in which larger grains consume smaller ones to reduce total grain-boundary area. The three-stage anneal—recovery, recrystallization, grain growth—is used to restore formability between drawing passes, to set final grain size in sheet metal, and to remove residual stresses in welds and castings. Hot working, carried out above the recrystallization temperature, recrystallizes continuously during deformation and therefore allows very large shape changes with no net hardening.

Chapter 2 — Processing of Ceramics

Ceramics are ionic or covalent crystalline (and often glassy) solids that are strong in compression, brittle in tension, and stable to high temperature. Because they cannot be forged like metals, their processing begins with powders and ends with a sintered or fused solid. A typical ceramic fabrication route runs powder synthesis, milling, mixing with binders and plasticizers, forming (pressing, slip casting, tape casting, extrusion, injection moulding), drying and binder burn-out, and finally sintering. Sintering is a diffusion-driven densification that occurs below the melting point: surface, grain-boundary and lattice diffusion transport atoms into the necks between particles, reducing total surface area and shrinking pores. The driving force is the reduction of surface free energy \( \gamma_{s} \); smaller starting particles sinter faster because they offer more curvature. Hot pressing and hot isostatic pressing apply simultaneous heat and stress to close residual porosity and produce near-theoretical density.

Glass processing differs in character because silicates solidify through a viscous liquid rather than by crystallization. The viscosity \( \eta(T) \) defines standard working, softening, annealing and strain points, and glass articles are shaped between the working and softening points by blowing, pressing, drawing or the float process. After shaping, a controlled anneal at the annealing point relaxes residual stresses. Tempered glass is made by quenching the surface while the interior is still hot, leaving the outer layer in compression and the core in tension—surface cracks must overcome this compressive prestress before they can propagate. Traditional clay-based ceramics (whitewares, porcelain, brick) use a mix of clay, flux and filler that partially melts during firing, with the glassy phase bonding the refractory grains. Engineering ceramics such as alumina, zirconia, silicon nitride and silicon carbide instead use high-purity synthesized powders and carefully controlled dopants. Zirconia makes particular use of a martensitic tetragonal-to-monoclinic transformation around a crack tip—transformation toughening—to raise fracture toughness. Regardless of composition, the final microstructure (grain size, porosity, second phases) is fixed during sintering and determines mechanical, electrical and thermal behaviour.

Chapter 3 — Processing of Polymers

Polymers are long-chain molecules assembled by either addition or condensation polymerization. In addition polymerization, a free-radical, cationic or anionic initiator opens the double bond of a vinyl monomer and the chain grows by successive monomer addition until termination. In condensation (step-growth) polymerization, bifunctional monomers react in pairs, releasing a small molecule such as water or methanol, and molecular weight rises only as the reaction approaches completion. The statistics of chain growth produce a molecular-weight distribution characterized by \( \overline{M}_{n} \) (number average) and \( \overline{M}_{w} \) (weight average); the ratio \( \overline{M}_{w}/\overline{M}_{n} \), the polydispersity index, indicates how narrow the distribution is.

Thermoplastics—polyethylene, polypropylene, polystyrene, PVC, nylon, PET—consist of linear or branched chains held by van der Waals forces. They soften reversibly when heated above the glass transition temperature \( T_{g} \) (for amorphous regions) or the melting temperature \( T_{m} \) (for crystalline regions), which makes them suitable for extrusion, injection moulding, blow moulding, thermoforming and fibre spinning. Thermosets such as epoxies, phenolics and unsaturated polyesters cure into a covalently crosslinked network that cannot be re-melted; they are processed by casting, compression moulding or reaction injection moulding. Elastomers—natural rubber, polyisoprene, silicones—have loosely crosslinked chains that allow large reversible strains, usually vulcanized with sulfur. Crystallinity, never complete, depends on chain regularity, and is enhanced by drawing; oriented crystalline lamellae are what give fibres such as Kevlar and drawn polyethylene their high modulus. Mechanical response is strongly time- and temperature-dependent: creep, stress relaxation and the frequency-dependent storage and loss moduli all reflect the same underlying viscoelastic relaxation spectrum. Additives (plasticizers, fillers, antioxidants, flame retardants, pigments) are routinely blended in to tailor processability, cost and durability.

Chapter 4 — Composite Materials

A composite combines two or more constituents with distinct properties into a bulk material whose performance exceeds that of any individual phase. The reinforcement carries most of the load; the matrix transfers stress to the reinforcement, protects it from damage and defines the shape. Composites are classified by reinforcement geometry—dispersoid (very fine hard particles), particulate (larger particles), fibre-reinforced (continuous or chopped fibres), and laminar (sheets or layers)—and by matrix type (polymer, metal, ceramic).

Dispersion-strengthened systems, such as sintered aluminum powder (SAP) with a fine \( \mathrm{Al_{2}O_{3}} \) dispersoid, rely on incoherent particles that pin dislocations at high temperature, giving creep resistance well above the solvus of conventional precipitation-hardened alloys. Larger-particle composites, such as WC–Co cemented carbides and concrete, combine a hard phase for wear or compressive strength with a tougher binder. For fibre-reinforced systems, the iso-strain (rule-of-mixtures) model gives the longitudinal modulus as

\[ E_{c} = V_{f}E_{f} + (1 - V_{f})E_{m}. \]

Transverse loading is better described by the iso-stress (inverse rule-of-mixtures) expression, and the true response of a laminate is computed layer by layer using classical lamination theory. Load transfer requires a critical fibre length \( \ell_{c} = \sigma_{f}d/(2\tau_{i}) \), where \( \tau_{i} \) is the matrix–fibre interfacial shear strength. Short fibres below \( \ell_{c} \) contribute mainly as stiffeners, while continuous fibres above it can approach the full fibre strength. Common systems include glass-, carbon- and aramid-fibre-reinforced epoxies for aerospace and sporting goods; metal-matrix composites such as boron–aluminum; and ceramic-matrix composites such as C/SiC for high-temperature airframe leading edges. Natural composites—wood, bone, nacre—use hierarchical fibre-and-matrix architectures and inspire many modern biomimetic designs.

Chapter 5 — What Makes Nanomaterials Different? Surface Energy and Quantum Confinement

A material becomes “nano” once at least one external dimension lies roughly between 1 and 100 nm. Two effects then dominate. First, the fraction of atoms that sit on a surface grows rapidly. A 30 nm particle has roughly 5% of its atoms at the surface; a 3 nm particle has nearly half. Because surface atoms have fewer neighbours than interior atoms, they carry an excess free energy per unit area \( \gamma_{s} \). Small particles therefore have a large positive contribution \( \gamma_{s}A \) to their free energy, which lowers their melting point (the Gibbs–Thomson effect), raises their solubility and chemical reactivity, and drives processes such as Ostwald ripening and sintering. The same surface energy sets the equilibrium crystal shape through the Wulff construction.

The second signature of the nanoscale is quantum confinement. When the size of a semiconductor crystal becomes comparable to the exciton Bohr radius, the allowed electronic states become discrete rather than a continuous band. A rough estimate from the particle-in-a-box model gives a confinement energy

\[ \Delta E \approx \frac{\hbar^{2}\pi^{2}}{2\mu L^{2}}, \]

where \( \mu \) is the reduced electron–hole effective mass and \( L \) the confinement length, so the effective band gap grows as \( L \) shrinks. This is why CdSe quantum dots fluoresce from red to blue as their diameter decreases from about 6 nm to 2 nm. In metals, plasmonic resonances shift with particle size and shape, producing the bright ruby reds and yellows of gold and silver nanoparticles. Magnetic crystals smaller than a single domain become superparamagnetic. Mechanical properties also change: the Hall–Petch strengthening of grain refinement persists down to roughly 15 nm, beyond which grain-boundary-mediated deformation softens the material. In short, reducing size changes not just quantitative performance but the underlying physics.

Chapter 6 — Synthesis of 0-D Nanomaterials: Nucleation and Growth

Zero-dimensional nanomaterials—nanoparticles, nanocrystals and quantum dots—are usually grown in solution from an initially homogeneous reactive mixture. Classical nucleation theory describes the process. In homogeneous nucleation the free energy change for forming a spherical nucleus of radius \( r \) is

\[ \Delta G(r) = -\tfrac{4}{3}\pi r^{3}\,\Delta G_{v} + 4\pi r^{2}\,\gamma, \]

where \( \Delta G_{v} \) is the volume free-energy change per unit volume and \( \gamma \) the solid–liquid interfacial energy. The balance between a favourable bulk term and an unfavourable surface term creates a barrier \( \Delta G^{*} \) at a critical radius \( r^{*} = 2\gamma/\Delta G_{v} \). Only nuclei larger than \( r^{*} \) grow; smaller ones redissolve. Heterogeneous nucleation on a pre-existing surface or seed lowers the effective barrier through a wetting-angle factor and is much more common in practice.

To produce monodisperse nanoparticles, LaMer’s scheme prescribes a rapid injection of precursor that raises supersaturation above the nucleation threshold, a short burst of nucleation, and then a prolonged diffusion-limited growth stage at a supersaturation too low to form new nuclei. This separates nucleation from growth in time and narrows the size distribution. Strong-binding surfactants such as oleic acid, oleylamine or trioctylphosphine oxide cap the surface, limit coalescence and allow control of shape by selectively blocking certain facets. Representative systems include the Turkevich and Brust–Schiffrin syntheses of gold colloids, the hot-injection synthesis of CdSe dots in TOPO, the sol–gel growth of silica spheres by the Stöber method, and the co-precipitation of magnetite for MRI contrast. Ostwald ripening—dissolution of small particles in favour of larger ones driven by Gibbs–Thomson—broadens the distribution if growth is not quenched. Spatially confined reactions in micelles, microemulsions and porous templates provide an alternative route in which the reactor geometry itself limits final particle size.

Chapter 7 — Synthesis of 1-D Nanomaterials: Nanowires, Nanotubes and Templates

One-dimensional structures—nanowires, nanorods, nanobelts, nanotubes—require growth to be fast along one axis and suppressed in the other two. Spontaneous growth methods exploit intrinsic anisotropy or catalyst-assisted asymmetry. The vapor–liquid–solid (VLS) mechanism, introduced by Wagner and Ellis, uses a small metal droplet (typically gold) that collects vapour-phase feedstock such as silane, forms a eutectic liquid alloy, supersaturates, and precipitates the solid as a wire beneath it. The droplet diameter sets the wire diameter, and wire length grows with time. Related variants include vapor–solid (VS) growth, solution–liquid–solid (SLS) growth in low-temperature solvents, and catalytic CVD growth of carbon nanotubes from Fe, Co or Ni particles, where arc discharge, laser ablation and hydrocarbon pyrolysis all yield single- or multi-walled tubes depending on conditions.

Template-based methods instead use a pre-existing host whose pores or trenches dictate the shape. Anodic aluminum oxide (AAO) membranes provide regular arrays of parallel cylindrical channels 10–200 nm in diameter and tens of micrometres deep; filling them by electrodeposition, sol–gel infiltration or melt impregnation gives dense arrays of metal, oxide or polymer nanowires. Track-etched polycarbonate membranes offer similar geometry at larger pore size. Mesoporous silica templates such as MCM-41 and SBA-15 provide even smaller channels for high-surface-area catalysts. Biological templates—tobacco mosaic virus, peptide nanotubes, DNA origami—offer precise molecular control of dimensions and chemistry and have been used to direct metallization into monodisperse nanorods. Regardless of route, post-growth removal of the template is often needed, and the choice between etch-back and lift-off determines how freestanding the final wires can be.

Chapter 8 — Synthesis of 2-D Films: PVD and CVD

Thin films are the workhorses of semiconductor, optics and coating industries. Physical vapour deposition (PVD) transports source atoms from a target to a substrate through a vacuum, without chemical reactions along the way. In thermal evaporation the source is heated (resistively or by an electron beam) until its vapour pressure is high enough to produce a line-of-sight flux. In sputtering the target is bombarded by energetic argon ions from a plasma; momentum transfer ejects target atoms, which travel through the sputter chamber and condense on the substrate. Magnetrons confine the plasma near the target and raise the deposition rate; reactive sputtering admits \( \mathrm{O_{2}} \) or \( \mathrm{N_{2}} \) into the chamber to form oxide or nitride films. Pulsed laser deposition (PLD) uses a focused excimer laser to ablate a target and transfer complex stoichiometries (YBCO, perovskite oxides) to the substrate.

Chemical vapour deposition (CVD) instead delivers volatile precursor molecules that react or decompose on a heated substrate. The Thornton zone model, or analogous phase diagrams for CVD, relates growth temperature and pressure to microstructure: at low mobility one obtains porous columnar films; at higher temperature dense, large-grained or even single-crystalline layers emerge. Common variants include atmospheric-pressure CVD (APCVD) for silicon oxide and nitride, low-pressure CVD (LPCVD) for polysilicon, plasma-enhanced CVD (PECVD) for low-temperature dielectric coatings, metal-organic CVD (MOCVD) for III–V compound semiconductors such as GaN LEDs, and atomic layer deposition (ALD), a self-limiting variant that alternates two half-reactions to build up films one atomic layer at a time with excellent thickness control and conformality—ideal for high-\( \kappa \) gate dielectrics and modern FinFET geometries. Epitaxial growth, in which the deposited film inherits the substrate crystal orientation, is central to compound-semiconductor devices and is achieved by molecular beam epitaxy (MBE) or MOCVD under carefully controlled flux and temperature.

Chapter 9 — Lithographic Fabrication: Photolithography, E-Beam, Nanoimprint

Lithography transfers a pattern into a thin film of resist on a substrate, which then serves as a mask for etching or deposition. In optical photolithography, ultraviolet light shines through a mask onto a photoresist-coated wafer; positive resists become soluble in the exposed regions, negative resists become insoluble. The resolution is diffraction-limited, \( R = k_{1}\lambda/\mathrm{NA} \), where \( \lambda \) is the exposure wavelength and \( \mathrm{NA} \) the numerical aperture of the projection optics. Advancing lithography has meant shrinking \( \lambda \) (436 nm g-line, 365 nm i-line, 248 nm and 193 nm deep-UV, and finally 13.5 nm extreme ultraviolet, EUV), raising \( \mathrm{NA} \) through immersion lithography, and squeezing \( k_{1} \) with resolution-enhancement techniques such as phase-shift masks, off-axis illumination, optical proximity correction and multiple patterning.

Electron-beam lithography (EBL) replaces photons with a tightly focused electron beam. Because the de Broglie wavelength of a 30–100 keV electron is picometres, EBL is not diffraction-limited in practice; resolutions below 10 nm are routine in PMMA and HSQ resists. The penalty is serial writing, which makes EBL slow and expensive and therefore confined to research, photomask writing and low-volume prototyping. Focused-ion-beam (FIB) tools add direct material removal and deposition at the nanoscale. Nanoimprint lithography (NIL) offers a high-throughput alternative: a rigid master mould, itself made by EBL, is pressed into a thermoplastic or UV-curable resist, deforming it into the mould’s relief. After release and etching of the residual layer, the pattern is transferred to the substrate. Soft lithography, especially microcontact printing with a PDMS elastomer, stamps self-assembled monolayers onto surfaces and is popular for biological patterning. Directed self-assembly of block copolymers complements top-down lithography by using microphase separation to generate sub-20 nm periodic templates on pre-patterned guides.

Chapter 10 — X-Ray Diffraction

X-ray diffraction (XRD) is the standard non-destructive tool for phase identification, lattice-parameter measurement, crystallite size analysis and residual-stress evaluation. It relies on coherent elastic scattering of X-rays from the regular planes of a crystal. Constructive interference requires the path-length difference between neighbouring planes to be an integer number of wavelengths—Bragg’s law:

\[ n\lambda = 2d_{hkl}\sin\theta. \]

Here \( d_{hkl} \) is the interplanar spacing, \( \theta \) the Bragg angle, \( \lambda \) the X-ray wavelength (Cu K\(\alpha\) at 0.154 nm is typical), and \( n \) an integer order. A powder diffractometer collects intensity as a function of \( 2\theta \); each peak corresponds to a family of planes, and the full pattern serves as a fingerprint that can be matched against the ICDD/PDF database. Peak positions give unit-cell constants and phase identification; relative intensities encode atomic positions and site occupancies through the structure factor \( F_{hkl} = \sum_{j}f_{j}\exp(2\pi i(hx_{j}+ky_{j}+lz_{j})) \). Full-profile Rietveld refinement fits calculated patterns to experimental data and extracts quantitative phase fractions.

Peak broadening carries microstructural information. Instrumental broadening aside, the remaining width has contributions from finite crystallite size and from microstrain. The Scherrer equation relates crystallite size \( D \) to peak width:

\[ D = \frac{K\lambda}{\beta\cos\theta}, \]

where \( \beta \) is the full width at half maximum in radians (after instrument deconvolution) and \( K \approx 0.9 \) is a shape factor. Williamson–Hall plots separate size and strain broadening by plotting \( \beta\cos\theta \) against \( \sin\theta \). Specialized geometries extend XRD beyond powders: glancing-incidence diffraction probes thin films by keeping the incidence angle below a degree; pole figures and texture measurements use four-circle goniometers; \( \sin^{2}\psi \) analysis measures residual stress; and small-angle X-ray scattering (SAXS) probes nanometre-scale density fluctuations and particle sizes that are too big for wide-angle Bragg peaks. Modern synchrotron beamlines extend these techniques to millisecond time resolution, micron spatial resolution and in-situ environments.

Chapter 11 — Electron Microscopy: SEM and TEM

Electron microscopes exploit the short de Broglie wavelength of accelerated electrons—around 0.004 nm at 100 keV—to resolve features far beyond the optical diffraction limit. In a scanning electron microscope (SEM), a focused electron probe (0.5–30 keV) is rastered across a bulk specimen. Inelastic scattering produces secondary electrons from the top few nanometres, which give high-resolution topographic images; elastically backscattered electrons come from deeper and yield compositional (Z-contrast) images. Impact ionization generates characteristic X-rays detected by an energy-dispersive spectrometer (EDS/EDX) for elemental analysis, while electron backscatter diffraction (EBSD) maps local crystallographic orientations with angular resolution better than a degree. Environmental SEMs relax the vacuum requirement and allow imaging of wet or insulating samples with minimal coating.

Transmission electron microscopy (TEM) passes a high-energy beam (typically 100–300 keV) through a specimen thinned to less than 100 nm. The image forms in the back focal plane from electrons that scatter elastically through the thin foil. Bright-field images exclude diffracted beams with an objective aperture and show mass-thickness and diffraction contrast; dark-field images select a single diffracted beam to highlight a given crystallographic phase; selected-area electron diffraction gives spot or ring patterns that identify crystal structure and orientation at sub-micron resolution. High-resolution TEM (HRTEM) records phase-contrast lattice-fringe images that resolve individual atomic columns in favourable orientations. Scanning transmission electron microscopy (STEM) rasters a sub-ångström probe across the sample and collects a signal with an annular dark-field detector, producing Z-contrast images whose intensity scales approximately as atomic number squared. Coupled to EDS and electron energy-loss spectroscopy (EELS), STEM provides simultaneous atomic-resolution imaging and local chemistry, and with aberration correctors has reached 0.05 nm point resolution. Sample preparation—ion milling, focused-ion-beam lift-out, ultramicrotomy, cryo-fixation for biological specimens—is as critical as the microscopy itself.

Chapter 12 — Applications: Energy, Catalysis, Sensors, Biology

The techniques surveyed so far converge in real devices. In energy storage and conversion, nanostructured electrodes are now standard: silicon nanowires and nanoparticles accommodate the 400% volume swing of lithiation without pulverizing, high-surface-area porous carbons deliver fast ion transport in supercapacitors, and thin-film layered oxides such as NMC and LFP dominate commercial lithium-ion cathodes. Dye-sensitized and perovskite solar cells rely on mesoporous \( \mathrm{TiO_{2}} \) scaffolds fabricated by sol–gel and screen printing, while CIGS and CdTe photovoltaics use sputtered and evaporated absorber layers. Thermoelectric nanostructures such as \( \mathrm{Bi_{2}Te_{3}} \) superlattices and nanowire arrays decouple electronic and phononic transport to raise the figure of merit \( zT \). Fuel cells use Pt nanoparticles dispersed on carbon supports to maximize catalytic surface area per gram of precious metal.

In heterogeneous catalysis the ratio of surface to bulk atoms is precisely the point. Industrial ammonia synthesis, reforming, hydrogenation and automotive three-way catalysts all depend on maintaining a fine dispersion of active nanoparticles on high-surface-area supports such as zeolites, alumina or ceria. Single-atom catalysts, where isolated metal atoms are anchored on defect sites, push the limit to the ultimate atom economy. Nanostructured sensors exploit how surface adsorption strongly perturbs electronic transport at the nanoscale. Carbon-nanotube and graphene field-effect transistors detect single molecules of \( \mathrm{NO_{2}} \) or biomarkers; metal-oxide nanowire arrays serve as electronic noses; localized surface plasmon resonance of gold colloids and nanorods reports changes in local refractive index with sub-molecular sensitivity, enabling label-free biosensing. Surface-enhanced Raman scattering (SERS) on rough silver and gold substrates raises Raman signals by factors of \( 10^{6}\text{–}10^{10} \) and allows single-molecule chemical identification.

In biology and medicine, superparamagnetic iron oxide nanoparticles act as MRI contrast agents and hyperthermia mediators, dendrimers and liposomes carry drugs across cellular barriers, and quantum dots provide bright, photostable, multiplexed fluorescent labels for cell imaging. mRNA vaccines use lipid nanoparticles to protect and deliver nucleic acid cargo; engineered scaffolds of electrospun polymer fibres template bone and neural tissue regeneration. The common thread across all these applications is that control of structure between one and a hundred nanometres—through the synthesis, deposition, lithography and characterization techniques introduced in this course—translates directly into macroscopic device performance. Mastery of these techniques is therefore the foundation on which modern materials and nanoscience engineering rests.