Diffusion-limited Growth Research Articles

Orientation and morphology of solid-state dewetting holes

We report on the orientation and morphology of solid-state dewetting holes obtained from a kinetic Monte Carlo model with nearest neighbor and next nearest neighbor interactions on a cubic lattice. The morphologies found in simulations share strong similarities with those of diffusion-limited crystal growth: We find compact shapes, isotropic and anisotropic seaweed shapes, and dendritic shapes. Some of these shapes have been observed in solid-state dewetting experiments with various semi-conductors or metals on insulating substrates. The sides of the fingers can exhibit two types of zigzag instabilities with ${90}^{\ensuremath{\circ}}$ or ${45}^{\ensuremath{\circ}}$. Due to fourfold symmetry, dewetting hole shapes are found to exhibit two possible orientations with fingers along the (110) or (100) directions. We find that the rotation transition from one orientation to the other can be observed in our model by varying anisotropy, temperature, or dewetting strength. The two orientations correspond, respectively, to experimental observations of dewetting for silicon and germanium on insulator and to the dewetting of metal films with different reducing gas flow rate.

Extremely low light scattering in nanostructured glasses formed by simultaneous nucleation and diffusion-limited growth of particles: modeling.

A structural model proposed earlier is used to simulate light scattering in nanostructured glass formed by diffusional growth of simultaneously nucleated particles with small volume fraction. Computer simulation of the structure allows one to calculate partial structure factors for polydisperse particles and to estimate light scattering intensity in the interference approximation. Interference effects lead to a significant decrease in scattered intensity compared with independent scattering in a range of long wavelengths. In some spectral range, the dependence of scattered intensity on the inverse wavelength is described by power function with the exponent p=7.1 which is appreciably greater than the Rayleigh value 4.

Growth mechanism of NPD crystal from amorphous thin film by observation of morphology and measuring growth rate

Crystallization in organic amorphous thin films causes degradation of organic light-emitting devices. Crystallization of N′-di-1-naphthyl-N,N′-diphenylbenzidine (α-NPD) thin films occurs by the migration of NPD molecules on a substrate. In this study, we investigated the kinetic process of the crystallization of α-NPD in amorphous thin films by comparing the growth rate with the morphology of NPD crystals. The growth rate of NPD crystals was changed indicating different kinetic processes. The observation of a depression zone formed around a single fibril crystal revealed that the depression zone was formed only at the side of the fibril crystal, not at the tip of the fibril crystal in thinner films. This shows that NPD molecules were transferred directly from the amorphous region to the tip of the fibril crystal. After the crystalline region became dense by the growth and divergence of fibril crystals, the circular crystalline region was expanded by two-dimensional diffusion-limited growth.

Real-Time Observation of CaCO3 Mineralization in Highly Supersaturated Graphene Liquid Cells.

The mineralization dynamics of calcium carbonate is investigated under highly supersaturated conditions using graphene liquid cell transmission electron microscopy. We demonstrate that the mineralization process has three steps: nucleation, diffusion-limited growth, and Ostwald ripening/coalescence. In addition, we show that the polymorphs precipitate in a specific order, from metastable aragonite to stable calcite, thus proving Ostwald’s rule of stages. In highly supersaturated solutions, the aragonite phase crystallizes in a stable manner, in addition to the calcite phase.

Time-Resolved Dynamics of Struvite Crystallization: Insights from the Macroscopic to Molecular Scale.

The crystallization of magnesium ammonium phosphate hexahydrate (struvite) often occurs under conditions of fluid flow, yet the dynamics of struvite growth under these relevant environments has not been previously reported. In this study, we use a microfluidic device to evaluate the anisotropic growth of struvite crystals at variable flow rates and solution supersaturation. We show that bulk crystallization under quiescent conditions yields irreproducible data owing to the propensity of struvite to adopt defects in its crystal lattice, as well as fluctuations in pH that markedly impact crystal growth rates. Studies in microfluidic channels allow for time-resolved analysis of seeded growth along all three principle crystallographic directions and under highly controlled environments. After having first identified flow rates that differentiate diffusion and reaction limited growth regimes, we operated solely in the latter regime to extract the kinetic rates of struvite growth along the [100], [010], and [001] directions. In situ atomic force microscopy was used to obtain molecular level details of surface growth mechanisms. Our findings reveal a classical pathway of crystallization by monomer addition with the expected transition from growth by screw dislocations at low supersaturation to that of two-dimensional layer generation and spreading at high supersaturation. Collectively, these studies present a platform for assessing struvite crystallization under flow conditions and demonstrate how this approach is superior to measurements under quiescent conditions.

The effects of SeO2 additive on Mn electrodeposition on Al substrate in MnSO4-(NH4)2SO4-H2O solution

Selenium dioxide (SeO2) plays critical role in the nucleation and growth of manganese (Mn) on the foreigner substrates, decreasing SeO2 consumption is one of the most challengeable subjects in electrometallurgy of Mn. This work studies the effects of SeO2 on initial stage of Mn electrodeposition on aluminum (Al) substrate. The use of Al instead of the traditional stainless steel cathode to deposit Mn is expected to decrease the SeO2 concentration in electrolyte. A compact layer is obtained on Al with a good quality. The electrochemical measurements demonstrate that the SeO32− remains inactive on Al at cathodic process, and the Se(IV) can be reduced to Se(0) and Se(-II) only when the Mn grains have formed on Al substrate. Mn2+ discharges firstly and forms the spherical gamma-Mn on Al, which follows a three-dimensional instantaneous nucleation and diffusion-limited growth mechanism. The rapid formation of a large amount of Mn grains provides more active sites for the reduction of SeO32−, and the Se(0) will further react with Mn2+ and form the manganese selenide. These processes induce the formation of nanoscale alpha-Mn on gamma-Mn surface, and improve the utilization of SeO2. The growth of the micro-crystals protects the initial layer against the corrosion and guarantees that higher current efficiency is obtained on Al cathode.

Morphological characterization of particles emitted from monopolar electrosurgical pencils

Monopolar electrosurgical pencils are used extensively in surgical operations. With such pencils, electric current passes to the tissue, and as such, electrosurgical pencil operation generates a significant amount of thermal energy, which in turn leads to the generation of electrosurgical smoke (ES). The health risks of ES are dependent on the size distributions as well as the morphologies of the produced particles. To better characterize such particles, in this study we utilized (1) differential mobility analysis with a condensation particle counter (DMA-CPC), (2) an aerodynamic particle spectrometer (APS), (3) DMA-transmission electron microscopy analysis (DMA-TEM), and (4) DMA-aerosol particle mass analysis (DMA-APM) to examine the size distribution and morphologies of particles produced during simulated operation of an electrosurgical pencil (Neptune E-SEP, Stryker Corporation) on bovine, porcine, and ovine tissue. We find that under a variety of operating conditions, ES particles are broadly distributed, with a mode mobility diameter in the 150–200 nm size range, and concentrations well above background levels in the 50 nm–5 μm size range. We also find that the “cut” mode of monopolar electrosurgical pencil operation generates higher particle concentrations than the “coagulate” mode, and that increasing the maximum applied power from 20 W to 50 W also increases ES particle concentrations. TEM images of mobility selected particles reveal both spherical particles and fractal-like agglomerates in ES; these different particle types are produced under the same operation conditions leading to an externally-mixed, morphologically-complex aerosol. Quantitative analysis of the agglomerate images revealed that agglomerates have an average fractal dimension near 1.93 and that they are structurally similar to agglomerates expected from a diffusion limited cluster aggregation growth mechanism. Despite the presence of both spheres and agglomerates, DMA-APM analysis reveals that all particles have effective densities in the 1000–2000 kg m−3 range, suggesting that they likely contain inorganic components. Finally, we determined that the collection efficiency of the ES capture suction unit attached to the electrosurgical pencil was >95% for particles in the 50–400 nm mobility diameter range.

Revisiting the t0.5 Dependence of SEI Growth

SEI growth in lithium-ion batteries is commonly assumed to scale with t0.5, in line with simple models of diffusion-limited surface layer growth. As a result, this model is widely used for empirical predictions of capacity fade in lithium-ion batteries. However, the t0.5 model is generally not theoretically sufficient to describe all of the various SEI growth modes. Furthermore, previous literature has not convincingly demonstrated that this model provides the best fit to measurements of SEI growth. In this work, we discuss the theoretical assumptions of the t0.5 model, evaluate claims of t0.5 dependence in six previously published datasets and one new dataset, and compare the performance of this model to that of other models. We find that few of the purported t0.5 fits in literature are statistically justified, although t0.5 generally describes SEI growth during storage better than SEI growth during cycling. Finally, we evaluate how the fitted exponents in the power-law models vary as a function of time, and we illustrate the limitations of using t0.5 for prediction without validating its applicability to a particular dataset. This work illustrates the theoretical and empirical limitations of the t0.5 model and highlights alternatives for more accurate estimates and predictions of SEI growth.

Evolution of the structure and dynamics of bovine serum albumin induced by thermal denaturation.

Protein denaturation in concentrated solutions consists of the unfolding of the native protein structure, and subsequent cross-linking into clusters or gel networks. While the kinetic evolution of structure has been studied for some cases, the underlying microscopic dynamics of proteins has so far been neglected. However, protein dynamics is essential to understand the specific nature of assembly processes, such as diffusion-limited growth, or vitrification of dense liquids. Here, we present a study on thermal denaturation of concentrated solutions of bovine serum albumin (BSA) in D2O with and without NaCl. Using small-angle scattering, we provide information on structure before, during and after denaturation. Using quasi-elastic neutron scattering, we monitor in real-time the microscopic dynamics and dynamical confinement throughout the entire denaturation process covering protein unfolding and cross-linking. After denaturation, the protein dynamics is slowed down in salty solutions compared to those in pure water, while the stability and dynamics of the native solution appears unaffected by salt. The approach presented here opens opportunities to link microscopic dynamics to emerging structural properties, with implications for assembly processes in soft and biological matter.

Catalytic effects of NH4+ on hydrogen evolution and manganese electrodeposition on stainless steel

Abstract The effects of (NH4)2SO4 concentration (c((NH4)2SO4)) on hydrogen evolution and Mn electrodeposition on stainless steel (SS) in different potential ranges were investigated by linear sweep voltammetry (LSV), electrochemical impedance spectroscopy (EIS), potentiostatic polarization, chronoamperometry, scanning electron microscope (SEM) and X-ray diffraction (XRD) techniques. The results show that the NH4+ discharge reaction (NDR) intensifies the overall hydrogen evolution kinetics, and the NDR is catalyzed by increasing c((NH4)2SO4) and over-potential. The electro-crystallization of Mn on SS follows a three-dimensional progressive nucleation and diffusion-limited growth mechanism. Increasing the over-potential could accelerate the nucleation rate and also cause the decline of the nucleation density. The absorbed Mn2+ preferably discharges at low over-potential. Increasing c((NH4)2SO4) at medium over-potential could improve the current efficiency and produce more block-like grains. The nucleation process is suppressed by increasing c((NH4)2SO4) at high over-potential, at which the formation of columnar grains with higher hydrogen contents becomes prevailing.

Polyamide Chain Growth and Crosslinking: The Controlling Determinants for Properties of Thin Film Composite Membrane

The desired features of a polyamide thin film composite (TFC) membrane for desalination can be tailored through careful control of membrane preparation condition. This work aims to provide more comprehensive understanding on polyamide film formation chemistry in order to correlate the membrane synthesis condition with its characteristics. Three series of TFC membranes were prepared by interfacial polymerization with adjustment of m-phenylenediamine (MPD) and trimesoyl chloride (TMC) monomer concentration as well as reaction duration. The membrane structural properties were evaluated based on glucose permeation then correlated with the transport and charge behaviors of membrane determined from pure water and inorganic salt permeation. During interfacial polymerization, competition occurs between polyamide chain growth and crosslinking. It was found that higher MPD concentration promoted polyamide chain crosslinking while increasing TMC concentration favored polyamide chain growth during diffusion-limited growth stage. Meanwhile, prolonged degree of polyamide chain growth and crosslinking occurred at longer reaction duration, which eventually caused self-limiting membrane growth. The water transport was primarily controlled by polyamide film thickness, porosity and hydrophilicity while size exclusion and Donnan exclusion worked in tandem in governing the salt separation. The TFC membrane synthesized at 3 w/v% MPD concentration and 0.10 w/v% TMC concentration under 60 s reaction duration achieved the best desalination performance with pure water permeability of 0.853 L/m2·h·bar and 81.4 % NaCl rejection.

Polyamide Chain Growth and Crosslinking: The Controlling Determinants for Properties of Thin Film Composite Membrane

The desired features of a polyamide thin film composite (TFC) membrane for desalination can be tailored through careful control of membrane preparation condition. This work aims to provide more comprehensive understanding on polyamide film formation chemistry in order to correlate the membrane synthesis condition with its characteristics. Three series of TFC membranes were prepared by interfacial polymerization with adjustment of m-phenylenediamine (MPD) and trimesoyl chloride (TMC) monomer concentration as well as reaction duration. The membrane structural properties were evaluated based on glucose permeation then correlated with the transport and charge behaviors of membrane determined from pure water and inorganic salt permeation. During interfacial polymerization, competition occurs between polyamide chain growth and crosslinking. It was found that higher MPD concentration promoted polyamide chain crosslinking while increasing TMC concentration favored polyamide chain growth during diffusion-limited growth stage. Meanwhile, prolonged degree of polyamide chain growth and crosslinking occurred at longer reaction duration, which eventually caused self-limiting membrane growth. The water transport was primarily controlled by polyamide film thickness, porosity and hydrophilicity while size exclusion and Donnan exclusion worked in tandem in governing the salt separation. The TFC membrane synthesized at 3 w/v% MPD concentration and 0.10 w/v% TMC concentration under 60 s reaction duration achieved the best desalination performance with pure water permeability of 0.853 L/m2·h·bar and 81.4 % NaCl rejection.

Two Growth Mechanisms of Thiol-Capped Gold Nanoparticles Controlled by Ligand Chemistry.

Thiols play an important role in the synthesis of well-defined nanoparticles (NPs) with tailored properties, but their effects on the formation kinetics of NPs are still under investigation. Here, we used in situ small-angle X-ray scattering (SAXS)/UV-vis spectroscopy and time-dependent transmission electron microscopy (TEM) to elucidate the role of thiols in the formation process of gold NPs (AuNPs) by changing the adding sequence between thiol ligand and reducing agent. Through quantitative analysis of in situ SAXS/UV-vis and TEM, detailed information on size, size distribution, the number of particles, optical properties, and the size evolution was obtained. Two different growth mechanisms of monodisperse AuNPs controlled by thiol ligand are exhibited: (i) thiol plays a dual role as a digestive ripening etchant and as a stabilizing ligand in the presence of a weak phosphine ligand. The digestive ripening mechanism involving the dissolution of bigger particles and subsequent deposition of monomers onto existing small NPs is responsible for producing narrowly dispersed NPs. (ii) Thiol acts as a strong stabilizing agent; in this case, the formation rate constant is quite slow, thus limiting the growth rate of NPs. Therefore, diffusion-limited growth mechanism is proposed for obtaining narrowly dispersed NPs with a diameter of 5.6 nm (12%). Our findings demonstrate that the formation of nearly monodisperse AuNPs with controllable size distribution could be realized by different growth mechanisms in the presence of thiol ligand.

Effects of tungsten and rhenium additions on phase-separation in a model Ni-Al-Cr-W-Re superalloy: A four-dimensional study

The effects of a combined 1.0 at.% W and 1.0 at.% Re addition to a model Ni-10.0Al-8.5Cr at.% superalloy aged at 1073 K (800 °C) for aging times up to 256 h are investigated using atom-probe tomography (APT) and Vickers microhardness measurements. The nanoscale properties, including volume fraction, mean-volume equivalent-radius and number density of the γ′(L12)-precipitates, the chemical compositions of the γ(f.c.c.)-matrix and γ′(L12)-precipitates, the concentration profiles and partitioning ratios between the two phases, γ(f.c.c.)/γ′(L12), hetero-interfacial concentration widths, and microhardness are measured as function of increasing aging time. Localized excesses of the solvent Ni at the γ(f.c.c.)/γ′(L12) interfaces persist at aging times as long as 256 h. Transient diffusional concentration gradients associated with solute-atoms during diffusion-limited growth exist in either the γ(f.c.c.)-matrix or γ′(L12)-precipitates for the as-quenched state; these gradients disappear after 0.25 h for Al and Cr, and after 1 h for W and Re. Tungsten partitions to the γ′(L12)-precipitates and Re partitions to the γ(f.c.c.)-matrix, respectively, and this partitioning behavior enhances the partitioning of Al to the γ′(L12)-precipitates and Cr to the γ(f.c.c.)-matrix. The partitioning behavior indicates that W and Re evolve temporally more slowly than Al and Cr due to their small diffusivities in Ni. The interfacial concentration widths in the Ni-Al-Cr-W-Re alloy decrease with increasing aging time and increasing mean precipitate-radius. Adding W and Re to the base Ni-10.0Al-8.5Cr at.% alloy increases the γ′(L12)-precipitate volume fraction and the microhardness due to the increased volume-fraction and solid-solution strengthening provided by W and Re.

Embedded ZrC-SiC nanocomposites from hydrothermal precursor with temperature-dependent oxidation resistance and high sinterability

ZrC-SiC nanocomposites were synthesized via a facile, green hydrothermal precursor conversion method. The zirconium, silicon and decomposed glucose can be successfully incorporated into a homogeneous framework. In subsequent pyrolysis, the precursor transforms to a core-matrix structure of the nano-ZrO2 and amorphous carbon/silica mixture, resulting in a short diffusion path and limited grain growth. Carbide first appears at a low temperature of 1200 °C, and the conversion is completed at 1500 °C with low oxygen content. The composite consists of grains about 100 nm, exhibiting specific embedded morphology, and has different oxidation resistance in three temperature zones based on the component. The origin mechanism and properties have been elucidated and analyzed. The present work demonstrates the effectiveness of hydrothermal chemistry for the synthesis of carbide composites and their promising application in high temperature protection. Such nanocomposites with controllable morphology and high sinterability beneficial for subsequent densification are also verified. The additive-free ceramic has been obtained with almost fully density (relative density >99%) at a low temperature of 1700 °C by spark plasma sintering (SPS). The sintered specimen possesses fine microstructure with hierarchical grain size distribution (about 1 μm and 100 nm, respectively) and good mechanical properties (fracture toughness of 4.3 ± 0.4 MPa m1/2 and Vickers hardness of 22.8 ± 0.7 GPa).

Combustion of a rapidly initiated fully dense nanocomposite Al–CuO thermite powder

Very short burn times of nanocomposite, fully dense, stoichiometric 2Al·3CuO thermite particles ignited by electro-static discharge (ESD) observed in earlier experiments are interpreted assuming that the reaction occurs heterogeneously at the Al–CuO interfaces while the initial nanostructure is preserved even after the melting points of various phases present in the particle are exceeded. The heating rate for the ESD-ignited particles is very high, reaching 109 K s−1. The reaction model assumes that the rate of reaction is limited by transport of the reacting species across the growing layer of Al2O3 separating Al and CuO. The model includes the redox reaction steps considered earlier to describe ignition of 2Al·3CuO nanocomposite thermites and adds steps expected at higher temperatures, when further polymorphic phase changes may occur in Al2O3. A realistic distribution of CuO inclusion sizes in the Al matrix is obtained from electron microscopy and used in the model. The model accounts for heat transfer of the nanocomposite particles with surrounding gas and radiative heat losses. It predicts reasonably well the burn times observed for such particles in experiments. It is also found that neglecting polymorphic phase changes in the growing Al2O3 layer and treating it as a single phase with the diffusion-limited growth rate similar to that of transition aluminas (activation energy of ca. 210 kJ mol−1) still leads to adequately predicted combustion temperatures and times for the nanocomposite particles rapidly heated by ESD. The model highlights the importance of preparing powders with fine CuO inclusion sizes in the nanocomposite particles necessary to complete the redox reaction; it is also found that the particle combustion temperatures may vary widely depending on their dimensions. Higher combustion temperatures generally lead to greater reaction rates and, respectively, to the more complete combustion.

Crystallization kinetics of clinopyroxene and titanomagnetite growing from a trachybasaltic melt: New insights from isothermal time-series experiments

In order to investigate the role of crystallization kinetics in mafic alkaline systems, textural measurements, mineral compositional changes and diffusion modelling calculations have been carried out on isothermal time-series experiments. The data were obtained at 400 MPa and 1100 °C under anhydrous (nominally 0 wt% H2O) and hydrous (2 wt% H2O added) conditions. A synthetic trachybasaltic melt was first heated up to the superliquidus temperature of 1300 °C and then rapidly cooled at 80 °C/min down to 1100 °C. The final target temperature was kept constant over variable dwell times in the range of 0.5–24 h. Results from textural analysis indicate the attainment of fast crystal growth kinetics at the shortest experimental run duration, with early achievement of stable crystal sizes for clinopyroxene and titanomagnetite. The surface area to volume ratio weakly decreases with increasing dwell time, according to the development of euhedral crystal morphologies. Crystal growth rates are also observed to progressively decrease from 0.5 to 24 h. Due to the effect of fast growth kinetics, the morphological maturation of clinopyroxene progresses by attachment of dendrite branches, infilling and overgrowth phenomena, leading to the formation of well-faced and euhedral crystals. The kinetically-controlled cation exchange (Si + Mg) → (TAl + Fe3+) controls the clinopyroxene compositional variation, expanding the stability of Tschermak component at the expense of diopside. Conversely, titanomagnetite is characterized by an almost constant composition that, however, is enriched in incompatible Al and Mg cations, as typically observed under rapid crystal growth conditions. Titanomagnetite crystals show always euhedral morphology that develops by heterogeneous nucleation on early-formed clinopyroxene dendrites. Overall, the effect of undercooling causes strong supersaturation phenomena in the trachybasaltic melt, resulting in enhanced nucleation kinetics and fast attainment of a high crystallinity. As the dwell time increases, the bulk system tends to minimize the interfacial free energy between crystals and surrounding melt. This results in the progressive replacement of the early dendritic shapes developed in a diffusion-limited growth regime, by the formation of euhedral morphologies typical of interface-limited regimes that still retain the chemical evidences of the dendritic stage as complex zoning patterns in clinopyroxene.

Effect of Lecithin on the Spontaneous Crystallization of Enzymatically Synthesized Short-Chain Amylose Molecules into Spherical Microparticles

Here, we report a facile and effective one-pot approach to prepare uniform amylose-based polymeric microparticles (PMPs) through enzymatic synthesis of short-chain amylose (SCA) followed by spontaneous self-assembly of the SCA in the presence of lecithin. The effect of lecithin on nucleation and growth kinetics of amylose microparticles was investigated by monitoring the turbidity of reaction solution and the size of particles over the course of the self-assembly process. The results suggest that lecithin played a critical role in controlling the self-assembly kinetics to form uniform amylose microparticles through steric stabilization of the growing particles and diffusion-limited growth effect. The crystallinity of amylose microparticles was not affected by lecithin, implying that lecithin did not disrupt the crystal structure within the particle and would mainly be present on the surface of the microparticles. Considering its biodegradable and biocompatible nature, the amylose-based microparticles would find a range of useful applications in the area of food, cosmetics, medicine, chromatography and other related materials sciences.

The Importance of Salt-Enhanced Electrostatic Repulsion in Colloidal Crystal Engineering with DNA.

Realizing functional colloidal single crystals requires precise control over nanoparticles in three dimensions across multiple size regimes. In this regard, colloidal crystallization with programmable atom equivalents (PAEs) composed of DNA-modified nanoparticles allows one to program in a sequence-specific manner crystal symmetry, lattice parameter, and, in certain cases, crystal habit. Here, we explore how salt and the electrostatic properties of DNA regulate the attachment kinetics between PAEs. Counterintuitively, simulations and theory show that at high salt concentrations (1 M NaCl), the energy barrier for crystal growth increases by over an order of magnitude compared to low concentration (0.3 M), resulting in a transition from interface-limited to diffusion-limited crystal growth at larger crystal sizes. Remarkably, at elevated salt concentrations, well-formed rhombic dodecahedron-shaped microcrystals up to 21 μm in size grow, whereas at low salt concentration, the crystal size typically does not exceed 2 μm. Simulations show an increased barrier to hybridization between complementary PAEs at elevated salt concentrations. Therefore, although one might intuitively conclude that higher salt concentration would lead to less electrostatic repulsion and faster PAE-to-PAE hybridization kinetics, the opposite is the case, especially at larger inter-PAE distances. These observations provide important insight into how solution ionic strength can be used to control the attachment kinetics of nanoparticles coated with charged polymeric materials in general and DNA in particular.

Shape-controlled synthesis and in situ characterisation of anisotropic Au nanomaterials using liquid cell transmission electron microscopy.

Understanding the mechanisms behind crystal nucleation and growth is a fundamental requirement for the design and production of bespoke nanomaterials with controlled sizes and morphologies. Herein, we select gold (Au) nanoparticles as the model system for our study due to their representative applications in biology, electronics and optoelectronics. We investigate the radiation-induced in situ growth of gold (Au) particles using liquid cell transmission electron microscopy (LCTEM) and study the growth kinetics of non-spherical Au structures. Under controlled electron fluence, liquid flow rate and Au3+ ion supply, we show the favoured diffusion-limited growth of multi-twinned nascent Au seed particles into branched structures when using thin liquid cells (100 nm and 250 nm) in LCTEM, whereas faceted structures (e.g., spheres, rods, and prisms) formed when using a 1 μm thick liquid cell. In addition, we observed that anisotropic Au growth could be modulated by Au-binding amyloid fibrils, which we ascribe to their capability to regulate Au3+ ion diffusion and mass transfer in solution. We anticipate that this study will provide new perspectives on the shape-controlled synthesis of anisotropic metallic nanomaterials using LCTEM.