A Historical Introduction to Computer Models for Fractal Aggregates

It has been known for a century that small particles dispersed in liquids and gases can form aggregates with extraordinarily low densities, and a variety of studies in the 1960s and 1970s, based on conceptual models and computer models, suggested that anomalous scaling relationships were associated with the structures of these aggregates. However, the lack of a suitable theoretical framework and the limits of computer technology inhibited the development of a coherent understanding of the structures of these aggregates and the kinetics of their formation. In the 1980s, the popularization of fractal geometry and rapid advances in computer technology removed these barriers to progress. In this review, the early work on fractal aggregates is discussed and the basic particle-cluster and cluster-cluster aggregation models introduced in the 1980s are described. During the 1990s, interest has been focused on the subtle relationships between aggregation, gelation and spinodal decomposition and on the physical behavior of systems containing fractal aggregates. The following papers in this special issue of the Journal of Sol-Gel Science and Technology on Computer simulations of aggregation and sol-gel processes describe recent advances in these directions. They are previewed in this introductory contribution.     

Triglyceride nanocrystal aggregation into polycrystalline colloidal networks: Ultra-small angle X-ray scattering,models and computer simulation

Triacylglycerols (TAGs) are the majority molecules present in edible fats and oils. Many of the functional characteristics of fat products depend on the colloidal fat crystal network present. Identifying the hierarchies of these colloidal networks and how they spontaneously self-assemble is important to understand their functionality and the oil binding capacity, and new insights into the nano- to meso-scale structure in these colloidal fat networks have been reported in recent years. Ultra small angle X-ray scattering (USAXS) is a technique new to the study of edible oil structures and, when combined with modelling and computer simulation, has enabled significant advances to be made in understanding the nano- to micro-scale crystalline structures of edible oils. In the four years since crystalline nanoplatelets (CNPs) were characterized, models have been made of these highly anisotropic nanoscale structures in which they were treated as the primary unit. In those models, CNPs were represented as close-packed rigid layers of spheres, so chosen because the van der Waals sphere–sphere interaction is known. The intent of the models was to predict the hierarchy of colloidal fat networks that would self-assemble from the components in edible oils. Initially, CNP aggregation was modelled under the assumption that all CNPs are present before aggregation begins and that their solubility in liquid oil is very low. The models successfully predicted the fractal dimensions subsequently measured using USAXS. This brief review reports on some of the latest models and simulations together with the results of USAXS experiments carried out on binary lipid systems, such as SSS in OOO, as well as certain complex systems that contain many different TAG molecules. The excellent agreement between the two approaches has established that USAXS is a powerful tool in the elucidation of the nano- to meso-length scales in fats and oils.     

Viscosity of liquid suspensions with fractal aggregates: Magnetic nanoparticles in petroleum colloidal structures

New physical model is presented resulting in a simple formula for the dependence of viscosity η of colloidal liquid solution on the shear rate G applicable to a wide variety of systems including complex natural liquids like petroleum. The principal point of the model is the fractal nature of colloid particle aggregates present in the liquid. Such aggregates are experimentally detected now in non-Newtonian liquids. The model is based on calculation of energy loss on colloidal particle aggregate of fractal structure localized in the flow of liquid with shear rate. We have performed the viscosity measurement experiments which confirmed successfully the developed physical model. Also, we demonstrate experimentally that petroleum colloidal particles and magnetic iron oxide nanoparticles can form composite fractal-like aggregates in natural petroleum materials. Our model can explain both the non-Newtonian properties of petroleum and sensitivity of petroleum viscosity to external magnetic fields.     

高磺化度聚苯胺体系中的分形结构研究

通过透射电镜的观察研究发现磺化聚苯胺的胶体聚集体和胶粒内部结构都具有分形体的特征 ,从而将分形的概念及其数学模型引入共轭导电聚合物体系之中 .磺化聚苯胺胶体的聚集体为很不均匀的分支状开放结构 ,其形成过程可用扩散控制集团聚集模型 (DLCA)进行模拟 ,计算机模拟生成的图形及其分形维数都与实验观测结果相当吻合 .胶粒由于是在分散介质所形成的平均化场中生成 ,屏蔽效应减弱 ,是比由它组成的聚集体要致密的球形结构 ,该结构的生成可用随机雨点模型模拟且结果相近 .     

Linking phase behavior and reversible colloidal aggregation at low concentrations: simulations and stochastic mean field theory

We have studied the link between the kinetics of clustering and the phase behavior of dilute colloids with short range attractions of moderate strength. This was done by means of computer simulations and a theoretical kinetic model originally developed to deal with reversible colloidal aggregation. Three different regions of the phase diagram were accessed. For weak attractions, a gas phase of small clusters in equilibrium forms in the system. For intermediate attractions, the system undergoes liquid-gas separation, which is signatured by the formation of a few large droplike aggregates, a gas phase of small clusters, and an overall kinetics where a few seeds succeed in explosively growing at long times, after a lag time. Finally, for very strong attractions, fractal unbreakable clusters form and grow following DLCA-like (diffusion limited cluster aggregation) kinetics; liquid-gas separation is prevented by the strength of the bonds, which do not allow restructuration. Good qualitative and quantitative agreement is found between the dynamic simulations and the kinetic model in all the three regions.     

Controlled food protein aggregation for new functionality

Globular proteins are an important component of many food products. Heat-induced aggregation of globular proteins gives them new properties that can be useful in food products. In order to optimize functionality, the aggregation process needs to be controlled, which in turn requires good understanding of the mechanism. Heating aqueous solutions of globular proteins leads to the formation of aggregates with one of four distinctly different morphologies: spherical particles, flexible strands, semi-flexible fibrils, and fractal clusters. We review recent research in this area focusing on the parameters that control the morphology including the influence of hydrolysis. The aggregation mechanism and the effect of the morphology on the functionality will be addressed. A distinction is made between primary aggregation leading to roughly spherical particles or more or less flexible strands and secondary aggregation leading to fractal clusters, gels or precipitates. We will discuss how the formation of aggregates with different morphologies is related to the formation of either particulate or fine stranded gels.     

Dynamic scaling properties of TeO2-based gels

Homogeneous, transparent, and mechanically rigid gels have been successfully synthesized in the tellurium isopropoxide-isopropanol-citric acid and water system. The sol to gel transition and the gels microstructure have been studied by using small angle X-ray scattering (SAXS) experiments. For any value of the two key synthesis parameters, which are the citric acid ratio and the alkoxide concentration, very small Te-rich elementary particles, about 1-1.5 nm in radius, form immediately when the water is added, leading to colloidal sols. During gelation, these elementary particles stick progressively together to build up fractal aggregates by a pure hierarchical aggregation process which has been identified as a reaction-limited cluster aggregation (RLCA) mechanism. The SAXS curve analysis, based on scaling concepts, shows that the gelling network exhibits a time and length scale invariant structure factor characterized by self-similarity. This self-similarity is also displayed for a wide range of chemical compositions and the gel microstructures only differ in their fractal aggregate size according to the tellurium isopropoxide concentration as well as the citric acid ratio.     

Flocculation of hematite particles by a comparatively large rigid polysaccharide: schizophyllan

We studied the flocculation kinetics and structure of hematite aggregates induced by a large rigid extracellular polysaccharide, schizophyllan. Transmission electron microscopy (TEM), atomic force microscopy (AFM), photon correlation spectroscopy (PCS), and static light scattering (SLS) were used to characterize hematite particles, schizophyllan chains, and their flocs, to follow the time evolution of floc sizes, and to determine floc fractal dimensions. A maximum flocculation rate was found at a certain schizophyllan/hematite ratio. The maximum rate was considerably smaller than the rate of diffusion-limited aggregation (DLA) of hematite particles induced by simple electrolytes. To interpret the experimental results and to reveal various factors affecting the optimal dosage, Monte Carlo simulations were performed on the flocculation of small colloidal particles by relatively long, monodisperse linear polymers. The existence of the maximum flocculation rate was confirmed by computer simulation. However, a higher optimal dosage of schizophyllan was obtained in the experiments. The difference in the optimal dosage can be attributed mostly to the higher adsorption affinity of the hematite on schizophyllan aggregates present in the initial solution and the presence of a large fraction of free polymer chains which do not participate in the flocculation process. Both experiments and computer simulations demonstrated the fractal nature of the schizophyllan-hematite flocs. The fractal dimensions of the flocs at the optimal dosage were determined. A higher fractal dimension was obtained from experiments than from computer simulations, suggesting a reconstruction of the floc structure. Finally, a two-stage flocculation mechanism for hematite particles in the presence of a relatively long schizophyllan polymer was proposed. In the first flocculation stage, the hematite particles are preferentially adsorbed onto the schizophyllan aggregates in solution. The second stage consists of the association of these reactive entities with each other and also with naked chains to form fractal flocs by a bridging mechanism, where the hematite particles play the role of ligands.     

On techniques for the measurement of the mass fractal dimension of aggregates

A review is presented of a number of techniques available for the characterisation of the structure of aggregates formed from suspensions of sub-micron particles. Amongst the experimental techniques that have been commonly used are scattering (light, X-ray or neutron), settling and imaging and these are the focus of this work. The theoretical basis for the application of fractal geometry to characterisation of flocs and aggregates is followed by a discussion of the strengths and limitations of the above techniques. Of the scattering techniques available, light scattering provides the greatest potential for use as a tool for structure characterisation even though interpretation of the scattered intensity pattern is complicated by the strong interaction of light and matter. Restructuring further complicates the analysis. Although settling has long been used to characterise particle behaviour, the absence of an accurate permeability model limits the technique as a means of determining the porosity of fractal aggregates. However, it can be argued that the determination of fractal dimension is relatively unaffected. The strength of image analysis lies in its ability to provide a great deal of information about particle morphology and the weaknesses lie in the difficulties with image processing and sample size as this is a particle counting technique. There are very few papers which compare the fractal dimension measured by more than one technique. Light scattering potentially provides a useful tool for checking settling results. However, further work is required to develop proper models for aggregate permeability and flow-through effects.     

Aggregate formation and collision efficiency in differential settling

A new method of application of Stokesian dynamics, which can efficiently simulate movements of up to 500 particles with interparticle interactions in reasonable computational times, has been developed for the purpose of investigating particle-cluster aggregation in aqueous systems. The method is applied to monodisperse non-Brownian spherical particles aggregating in differential settling, while repulsive colloidal interaction is presumed to be negligible, so that a minimum separation distance can represent the attractive van der Waals force. The final aggregates formed by this algorithm, composed of 300 primary particles, have a common fractal dimension of approximately 2.0. The computed collision efficiency, defined as the product of a global and a capture efficiency, is about 5.77x10(-3). This value is significantly larger than the collision efficiency of primary particles colliding with an impermeable solid sphere of the same size as the aggregate, illustrating the important interplay between the permeability and the formation of aggregates.     

Destabilization of polystyrene latex particles induced by adsorption of polyvinylamine: mass, size and structure characteristics of the growing aggregates

The obviously visible aggregation of suspended colloidal particles resulting from the addition of polyvinylamine to the aqueous dispersion of polystyrene latex particles bearing surface sulfate groups set in with a delay of 24 h. The aggregation mechanisms and the fractal dimension of the aggregates were derived from the variations with time of the weight and number averaged masses of the aggregates as well as of the weight averaged harmonic mean diameter of the size distribution. Since the establishment of starved layers was determined to be relatively fast and to leave the liquid phase free of polymer, the delay for the obvious destabilization was attributed to the reconformation of adsorbed macromolecules that was expected to be extremely slow. This reconformation promoted the emergence of the diffusion-limited aggregation process that accompanies the permanent reaction-limited aggregation process. The fractal dimension of the latex particles/polyvinylamine aggregates was determined to be 2.12.     

FRACTAL STRUCTURE OF COLLOIDAL SILVER AND ITS EFFECTS ON SERS INTENSITIES OF CRYSTAL VIOLET

The surface-enhanced Raman spectra (SERS) of crystal violet on colloidal silver particles of different sizes and shapes were measured and the structural effect of the silver aggregates on the SEPS intensities was discussed in terms of fractal geometry. It is found that the SERS band intensities of crystal violet change significantly with the fractal dimension (D) of colloidal silver at 1.251.8.     

扩散模型和凝聚模型耦合作用下胶体凝聚动力学的Monte Carlo模拟研究

以扩散模型(Ds(γ)=D0×sγ)和凝聚模型(Pij(σ)=P0×(i×j)σ)为基础,对胶体体系随时间的演变、团簇大小分布及其标度关系、团簇的重均大小S(t)的变化规律以及模型对最终分形维数的影响四个角度进行了比较研究,发现扩散指数γ0和凝聚概率指数σ0对胶体的凝聚动力学过程有相似的影响.本文在较宽的γ和σ取值范围内,对胶体的凝聚动力学进行了模拟研究,对慢速凝聚向快速凝聚的转化机理作了定量分析,并进一步分析了在团簇-团簇凝聚(CCA)模型下,得到类似扩散置限凝聚(DLA)模型的凝聚体的物理意义,结果表明:(1)γ0代表了体系中团簇或单粒做"定向运动"而非无规则的布朗运动的情况.这种"定向运动"的推动力可能来自于大团簇产生的强"长程范德华力"、"电场力"等,或来自于体系边界处的外力场的作用.(2)当σ0时,体系成为先快后慢的慢速凝聚,这可能对应大团簇为一排斥中心,即胶体颗粒存在"排斥力场"的现象.(3)证实了团簇的重均大小在凝聚过程的早期按指数规律增长,而后期按幂函数规律增长的实验现象.模拟研究还表明,胶体体系的凝聚动力学过程,在σ0时是一个存在正反馈机制的非线性动力学过程,而在σ0时则体现出负反馈的特征.     

Model of bound water formation in fractal clusters resulting from reversible aggregation of colloidal particles

A model is proposed for the formation of bound water in fractal clusters that arise in the course of the reversible aggregation of colloidal particles. The dependence of the specific content of bound (nonfreezing) water on the concentration of a colloidal solution is theoretically determined. It is found that this value is related to the concentration of the colloidal solution via a power law. It is shown that the exponent is determined by the total surface area and morphology of fractal clusters, as well as by the mechanism of their aggregation and disintegration.     

Structures of DNA-linked nanoparticle aggregates

The room-temperature structure of DNA-linked gold nanoparticle aggregates is investigated using a combination of experiment and theory. The experiments involve extinction spectroscopy measurements and dynamic light scattering measurements of aggregates made using 60 and 80 nm gold particles and 30 base-pair DNA. The theoretical studies use calculated spectra for models of the aggregate structures to determine which structure matches the observations. These models include diffusion-limited cluster-cluster aggregation (DLCA), reaction-limited cluster-cluster aggregation (RLCA), and compact (nonfractal) cluster aggregation. The diameter of the nanoparticles used in the experiments is larger than has been considered previously, and this provides greater sensitivity of spectra to aggregate structure. We show that the best match between experiment and theory occurs for the RLCA fractal structures. This indicates that DNA hybridization takes place under irreversible conditions in the room-temperature aggregation. Some possible structural variations which might influence the result are considered, including the edge-to-edge distance between nanoparticles, variation in the diameter of the nanoparticles, underlying lattice structures of on-lattice compact clusters, and positional disorders in the lattice structures. We find that these variations do not change the conclusion that the room-temperature structure of the aggregates is fractal. We also examine the variation in extinction at 260 nm as temperature is increased, showing that the decrease in extinction at temperatures below the melting temperature is related to a morphological change from fractal toward compact structures.     

The origin of anomalous enhancement of electromagnetic fields in fractal aggregates of metal nanoparticles

It is shown that the anisotropy of the environment of metal nanoparticles with plasmon absorption in fractal aggregates is the most important and universal characteristic underlying their unique electrodynamic properties. It is noted that it is this morphological feature, but not the fractal distribution of particles in aggregates as such, that plays the dominant role in the manifestation of the enhancement of a local field. In this case, fractal aggregates possess the ability to enhance local electromagnetic fields only owing to their inherent local anisotropy; macroscopic characteristics of aggregates do not markedly influence their electrodynamic interactions with the external field. The quantitative characteristic of local anisotropy is introduced. Statistical correlation between the factor of local anisotropy and fractal dimension D of aggregates is established within the range 1.6 < D < 2.8. It is disclosed that the local anisotropy is independent of the fractal dimension within the wide range (1.6 < D < 2.5) except for the range D > 2.5 corresponding to aggregates with close-packed particles where the factor of local anisotropy tends toward zero. Strong correlation in the spatial arrangement of particles with the largest local anisotropy of the environment in aggregates and the strength of the local electromagnetic field is established using aggregates of silver nanoparticles as an example for the spectrum in the visible range; a polarization dependence of this correlation is revealed. It is noted that parameters of local anisotropy can be used to determine the degree of imperfection of colloidal crystals via optical methods.     

Structure and dynamic properties of colloidal asphaltene aggregates

The abundant literature involving asphaltene often contrasts dynamic measurements of asphaltene solutions, highlighting the presence of small particle sizes between 1 and 3 nm, with static scattering measurements, revealing larger aggregates with a radius of gyration around 7 nm. This work demonstrates the complementary use of the two techniques: a homemade dynamic light scattering setup adapted to dark and fluorescent solutions, and small-angle X-ray and neutron scattering. Asphaltene solutions in toluene are prepared by a centrifugation separation to investigate asphaltene polydispersity. These experiments demonstrate that asphaltene solutions are made of Brownian colloidal aggregates. The hydrodynamic radii of asphaltene aggregates are between 5 and 10 nm, while their radii of gyration are roughly comparable, between 3.7 and 7.7 nm. A small fraction of asphaltenes with hydrodynamic and gyration radii around 40 nm is found in the pellet of the centrifugation tube. The fractal character of the largest clusters is observed from small angle scattering nearly on a decade length scale. Previous results on aggregation mechanisms are confirmed ( Eyssautier, J., et al. J. Phys. Chem. B 2011 , 115 , 6827 ): nanoaggregates of 3 nm radius, and with hydrodynamic properties also frequently illustrated in the literature, aggregate to form fractal clusters with a dispersity of aggregation number.     

Colloidal aggregation induced by long range attractions

The structure of colloidal clusters formed by long-range attractive interactions under diluted conditions is studied by means of Monte Carlo simulations. For a not-too-long attraction range, clusters show self-similar internal structure with lower density than that typical for diffusive aggregation. For long-range interactions, low kappa, nonfractal clusters are formed (dense at short scales but open at long ones). The dependence on the volume fraction shows that more-compact clusters are grown the higher the colloidal density for diffusive aggregation and attraction-driven aggregation in the fractal regime. The whole trend is explained in terms of the interpenetration among aggregates. In attraction-driven aggregations, the interpenetration of clusters competes with aggregation in the tips of the clusters, causing low-density clusters.