Colloid Attachment Research Articles

Collision-attachment simulation of membrane fouling by oppositely and similarly charged colloids

The fouling propensity of oppositely charged colloids (OCC) and similarly charged colloids (SCC) on reverse osmosis (RO) and nanofiltration (NF) membranes are systematically investigated using a developed collision-attachment approach. The probability of successful colloidal attachment (i.e., attachment efficiency) is modelled by Boltzmann energy distribution, which captures the critical roles of colloid-colloid/membrane interaction and permeate drag. Our simulations highlight the important effects of ionic strength Is, colloidal size dp and initial flux J0 on combined fouling. In a moderate condition (e.g., Is =10 mM, dp=50 nm and J0= 100 L/m2h), OCC mixtures shows more severe fouling compared to the respective single foulant owing to electrostatic neutralization. In contrast, the flux loss of SCC species falls between those of the two single foulants but more closely resembles that of the single low-charged colloids due to its weak electrostatic repulsion. Increased ionic strength Is leads to less severe fouling for OCC but more severe fouling for SCC, as a result of the suppressed electrostatic attraction/repulsion. At a high Is (e.g., 3–5 M), all the single and mixed systems show the identical pseudo-stable flux Js. Small colloidal size leads to the drag-controlled condition, where severe fouling occurs for both single and mixed foulants. On the contrary, better flux stability appears at greater dp for both individual and mixed species, thanks to the increasingly dominated role of energy barrier and thus lowered attachment efficiency. Furthermore, higher J0 above limiting flux exerts greater permeate drag, leading to elevated attachment efficiency, and thus more flux losses for both OCC and SCC. Our modelling gains deep insights into the role of energy barrier, permeate drag, and attachment efficiency in governing combined fouling, which provides crucial guidelines for fouling reduction in practical engineering.

Colloid Transport and Retention in Constricted Tube Pore Spaces With Diverse Geometries and Orientations

AbstractColloidal contaminants infiltrate and can be attached onto grain surfaces of soils and aquifers, where they may persist. In this study, Lagrangian particle tracking is used to investigate particle trajectories and attachment in pore and fracture spaces modeled as three‐dimensional constricted tubes with diverse geometries and orientations relative to gravity. A comprehensive force balance arising from hydrodynamic drag and lift, gravitational settling, Brownian motion, and attractive DLVO interactions is simulated. Results show that the collection efficiency η is primarily governed by the dimensionless settling number 𝑆, representing the relative dominance of gravitational over hydrodynamic forces experienced by the particles. High‐𝑆 scenarios have larger η and are more sensitive to pore orientation, while low‐𝑆 scenarios are more sensitive to pore geometry. For all scenarios but especially low‐S scenarios, the majority of colloid attachment occurs near pore extremities, where fluid velocities are low, such that mechanical remobilization of particles attached is improbable. In low‐𝑆 scenarios, particles may spread and become immobilized at greater distances from the contamination source owing to lower η, are harder to mechanically remobilize as they attach more disproportionately at pore extremities, and have trajectories more sensitive to minor forces, rendering their environmental fates complex. Characterizing the collection efficiency and deposition morphology for various pore space geometries and orientations is crucial in understanding particle fate and developing continuum‐scale models of colloid transport in real soils, where pore spaces are heterogeneous and advection paths are tortuous.

Various Bacterial Attachment Functions and Modeling of Biomass Distribution in MICP Implementations

Microbial induced calcium carbonate precipitation (MICP) offers a robust technique to improve strength and stiffness properties of subsurface soils supporting infrastructures. Several unknown factors, including the MICP reactive transport parameters, however, limit the ability to predict spatial distribution of calcium carbonate (CaCO3) precipitation within a subsurface area and with depth. As it was shown that calcium carbonate distribution is highly affected by biomass profiles in subdomains, five bacteria attachment models (constant-rate, power-law, exponential, gamma distribution, and “cstr based on colloid attachment theory”) were calibrated here using data from both small- and large-scale testing programs. Out of the five models, colloid attachment theory with modified velocity and straining terms was shown to be the most promising approach in yielding the most fitted CaCO3 distribution compared with the experimental data. A new parameter, cstr, was incorporated to modify straining and the constraint peak value of biomass attachment due to straining at distances larger than a 0.14×sample size. Using the results from the numerical simulations, relationships were developed for velocity and straining coefficients of “the cstr based on colloid attachment theory” (hereafter “colloid attachment cstr”) as a function of bacteria size, soil particle size, sample size, volume of injected bacteria, and soil pore volume.

Review of Colloid (Nano‐ and Micro‐Particle) Transport and Surface Interaction in Groundwater. William P. Johnson and Eddy F. Pazmino. The Groundwater Project. 111 pp. ISBN: 978‐1‐77470‐070‐9. DOI: https://doi.org/10.21083/978‐1‐77470‐070‐9

Review of <i>Colloid (Nano‐ and Micro‐Particle) Transport and Surface Interaction in Groundwater</i>. William P. Johnson and Eddy F. Pazmino. The Groundwater Project. 111 pp. ISBN: 978‐1‐77470‐070‐9. DOI: https://doi.org/10.21083/978‐1‐77470‐070‐9

Application of temporal moment analysis to interpret colloid and colloid-facilitated solute transport under varying size exclusion and attachment coefficient.

Colloidal particles can attach the contaminants like heavy metals and radionuclides and act as contaminant carriers to provide a faster movement of pollutants through the interconnecting pores of the porous medium. The electrostatic repulsive force between the negatively charged colloids and the solid surface restricts the transport of the mobile colloids to the larger pores of the porous medium and initiates the size exclusion mechanism. The temporal moment analysis is treated as an effective tool to interpret the solute breakthrough curves for analyzing the statistical behavior of the contaminants. In past literature, the temporal moments have not been incorporated with the breakthrough curves of colloids and colloid-facilitated contaminants for statistical interpretation. In this research study, the temporal variations of concentrations of mobile colloids, solute attached to the mobile colloids, and the dissolved solute are obtained numerically in a fully saturated one-dimensional column considering a continuous source for varying size exclusion and colloid attachment coefficient. Utilizing the simulated spatially varying breakthrough curves, the temporal moments are estimated to calculate the mass recovery, average residence time, and the spreading of mobile colloids and dissolved solutes. The temporal moment analysis suggests that the velocity enhancement for higher size exclusion reduces the average residence time of the mobile colloids and the solute adsorbed to the mobile colloids significantly. The mass recovery of mobile colloids and the solute attached to the mobile colloids increases at a specific depth for higher size exclusion. The estimated second central moment attributes that the solute spreading follows the nonlinear trend for low size exclusion. The peaks of the relative concentration of mobile colloids and solute attached to mobile colloids drastically decrease with an increase in attachment coefficient. The peak of the relative concentration of dissolved contaminant enhances with attachment coefficient. The high second temporal moment of the dissolved contaminant at a higher attachment coefficient indicates the slow interaction of dissolved solute and porous medium and that enables a greater spreading of solute through the interconnecting porous medium. The study suggests that the faster movement of mobile colloids and the solute attached to the mobile colloids at higher exclusion imparts a potential risk of groundwater contamination and thorough statistical interpretation is needful to analyze the behavior of colloids and colloid-facilitated contaminants. The research work does not consider the transient flow field and the effect of the presence of air phase in the partially saturated soil column in the groundwater system.

Colloid transport through a variable-aperture fracture under unfavorable attachment conditions: Characterization with a continuous time random walk model

Although the effects of fracture aperture variability on colloid advection and dispersion are well understood, how aperture variability affects colloid deposition under unfavorable geochemical conditions remains underexplored. To address this knowledge gap, we simulated 6200 cases of colloid transport in variable-aperture fractures with different magnitudes of fracture aperture variability and correlation lengths. Flows were laminar (Reynolds number < 1) with dispersion-dominated transport (Peclet number < 1) in the simulations. The numerically obtained breakthrough curves (BTCs) reflected non-Fickian transport behavior with power-law tailing where the power-law exponent increased with aperture variability rather than correlation length. Solutions to the advection-dispersion-reaction equation (ADRE) and a continuous time random walk (CTRW) model were fit to the BTCs. The CTRW model faithfully reproduced BTCs that decayed according to a power law, while the ADRE failed to do so. The ensemble mean of the attachment rate (equivalent to the reaction rate in the ADRE) in the CTRW model was correlated to a measurable property (i.e., aperture variability) using a power-law function. This newly established function facilitates accurate prediction of non-Fickian colloid transport using the CTRW model where dynamic attachment and detachment of colloids are considered under unfavorable deposition conditions.

Important Role of Concave Surfaces in Deposition of Colloids under Favorable Conditions as Revealed by Microscale Visualization.

This study conducted saturated column experiments to systematically investigate deposition of 1 μm positively charged polystyrene latex micro-colloids (representing microplastic particles) on negatively charged rough sand, glass beads, and soil with pore water velocities (PWV) from 4.9 × 10-5 to 8.8 × 10-4 m/s. A critical value of PWV was found below which colloidal attachment efficiency (AE) increased with increasing PWV. The increase in AE with PWV was attributed to enhanced delivery of the colloids and subsequent attachment at concave locations of rough collector surfaces. The AE decreased with further increasing PWV beyond the threshold because the convex sites became unavailable for colloid attachment. By simulating the rough surfaces using the Weierstrass-Mandelbrot equation, the extended Derjaguin-Landau-Verwey-Overbeek (XDLVO) interaction energy calculations and torque analysis revealed that the adhesive torques could be reduced to be comparable or smaller than hydrodynamic torques even under the favorable conditions. Interestingly, scanning electron microscopic experiments showed that blocking occurred at convex sites at all ionic strengths (ISs) (e.g., even when the colloid-colloid interaction was attractive), whereas at concave sites, blocking and ripening (i.e., attached colloids favor subsequent attachment) occurred at low and high ISs, respectively. To our knowledge, our work was the first to show coexistence of blocking and ripening at high ISs due to variation of the collector surface morphology.

Predicting the Rate Coefficients of Attachment and Detachment of Colloids in Saturated Porous Media

The transport behavior of pathogenic microorganisms and nanoparticles (NPs) in the subsurface is usually studied by performing laboratory soil column experiments. Parameters describing colloid deposition on grain surfaces are estimated by fitting observed breakthrough curves with an appropriate one-dimensional model. However, predictive tools to estimate colloid deposition parameters, knowing the system properties such as soil type, colloid type, solution chemistry, and flow velocity, are useful in estimating the travel distances of pathogenic microorganisms in the subsurface. Such predictive models are rare, except the colloid attachment rate coefficient predicted by colloid filtration theory (CFT) under favorable conditions. Although a couple of theoretical and empirical predictive models are available for estimating the deposition parameters under unfavorable conditions, they were developed for a small set of data, and their applicability to a wide range of conditions is unexplored. In this study, several sets of column-experimental data from literature, covering a wide range of experimental conditions, were analyzed to understand key factors that control colloid deposition. Empirical relationships were developed for deposition rate coefficients and sticking efficiency of various colloidal types [viruses, bacteria, graphene oxide (GO) NPs, silver (Ag) NPs, titanium dioxide (TiO2) NPs, and carboxylate-modified latex (CML) colloids] vis-à-vis 11 physicochemical parameters such as porosity, mean pore-water velocity, median grain size, colloid radius, solution ionic strength, surface potentials of colloids and grains, Hamaker constant, temperature, viscosity of water, and dielectric constant. While deposition of viruses and CML colloids on grain surfaces was found to be reversible, deposition of bacteria, GO NPs, Ag NPs, and TiO2 NPs was found to be irreversible. The empirical equations proposed in this study can predict deposition rate coefficients more closely (p &amp;lt; .001, R2 = 0.69−0.85) than CFT (p &amp;lt; .7, R2 ≤ 0.41). The performance of CFT in predicting the attachment rate coefficients of viruses, bacteria, GO NPs, TiO2 NPs, and CML colloids was found to improve significantly when estimated rate coefficients were multiplied by the sticking efficiency calculated using the empirical expression proposed in this study (p &amp;lt; .001, R2 = 0.65−0.95).

Application of machine learning in colloids transport in porous media studies: Lattice Boltzmann simulation results as training data

Colloid transport through a porous medium changes geometrical and hydraulic properties of the pore space. The impact of this effect depends on the colloid types and pore space surface properties which determine the likelihood of pore clogging. Colloid attachment and subsequent detachment are key factors in pore clogging. In this study, the impact of four major fluid and colloids properties on the pore surface and hydraulic conductivity alteration during colloids transport were evaluated using machine learning. These four parameters include solution ionic strength, zeta potential, colloid size and fluid flow velocity.A combined lattice Boltzmann-smoothed profile method was used to simulate accurately coupled mechanisms governing colloid transport to evaluate the impact of the four parameters on the resulting pore space properties after colloid transport. The result of several simulations revealed significant changes of pore surface coverage by the attached colloids, and conductivity, void fraction and coordination number of colloid agglomerates created during transport of individual colloids. Since the simulation of the impact of combination of all possible sets of four parameters is very time consuming, an Artificial Neural Network (ANN) was used as a prognostic method to use the results of several simulations to predict the behavior for a wide range of pore, colloidal and fluid properties. Reported results from a set of 162 simulation case studies for different possible combination of solution ionic strength, zeta potentials, colloid size and flow velocity were selected as input parameters for the machine learning. Four output parameters, namely, pore surface coverage, conductivity, void fraction and coordination number of the colloidal particles were selected.To lower the prediction error value, which is targeted to be lower than 10%, networks were trained 50 times using a MATLAB code, and in each training, after at most 10 epochs, networks were trained. A maximum relative error value of 8.95% was obtained, which is very well within the range of training quality criteria. The results show that the ANN can profoundly predict the simulation outcomes for a wide range of ionic strength (IS) and can be directly used to obtain the value of dependent variables through simple calculations using network weights and transfer functions.

Impact of heavy metal cations on deposition and release of clay colloids in saturated porous media

AbstractWhile the facilitated transport of heavy metals by colloids such as clay particles has been widely recognized, the influence of heavy metals on transport of the colloids has received much less attention. This study conducted saturated column experiments to systematically examine influence of multivalent heavy metal cations on transport of natural clay colloids and fullerene nC60 nanoparticles in glass bead porous media. Results showed that the presence of Cd2+ in the solution can cause more deposition of clay colloids in glass beads compared to Ca2+ at a given ionic strength (IS), and the subsequent release by IS reduction was less, demonstrating that the presence of Cd2+ increased the irreversibility of attachment. When the glass beads were initially adsorbed with Ca2+ and Cd2+, the release of clay colloids was significantly reduced due to strong cation bridge between the clay colloids and collector surfaces via the heavy metal cations. The release of clay colloids was significantly increased upon reduction of solution IS if the Na+ was used to exchange for the Ca2+ or Cd2+ before the IS reduction. Additional experiments showed that the nC60 nanoparticles that were deposited in the presence of Fe3+ and Cu2+ cannot be released by reduction of solution IS. However, the nanoparticle release occurred when the Ca2+ was used to exchange for the Fe3+ and Cu2+. Our work was the first to reveal the influence of heavy metal cations on the irreversibility of colloid attachment, and the findings have important implication to fabrication of functional nanomaterials for stabilization of heavy metals in soil and development of mathematic models for accurate prediction of cotransport of heavy metals and colloids in subsurface environments.

Fluid flow and colloid transport experiment in single-porosity sample; tracking of colloid transport behavior in a saturated micromodel

Release of colloids and their subsequent transport into the subsurface environments takes place during a wide range of applications such as industrial, energy storage, and agricultural activities. Therefore, processes contributing to transport, attachment, and re-mobilization of colloids in porous media are attracting attention. A fraction of the released colloids may cross the soil vadose zone to reach the saturated groundwater. In this study, we explored colloid transport in a micromodel with high repulsion energy barrier where colloid retention is assumed to be low. Three major shortcomings were improved: pore space domain size, imaging resolution, and speed of imaging. The flow path of 1357 colloids with a size of 4 µm were tracked and these enabled precise determination of individual colloid transport mechanism as well as the integrated behavior of the system.Our direct observations have shown that even under unfavorable attachment conditions (defined based on the DLVO theory) colloids deposition occurred which was mainly due to the local flow velocity fluctuations and grain surface heterogeneity. Using the information from collective trajectories, we have quantified the contribution of differently behaved colloids in the observed breakthrough curve which show an integrated, macroscopic, behavior of the system and is often the only available information when performing column or field scale experiments to explore colloid transport in porous media. Furthermore, we have shown that attachment and remobilization of colloids increased the dispersion coefficient, and consequently the dispersivity value of the media.

Favorable and unfavorable attachment of colloids in a discrete sandstone fracture

The transport of cationic amine-modified latex (AML) and anionic carboxylate-modified latex (CML) microspheres through a discrete sandstone fracture with mineralogical heterogeneity and roughness was studied. Two microsphere sizes (200 nm and 1000 nm), two ionic strengths (5 mM and 10 mM), and two specific discharges (0.35 mm.s-1 and 0.70 mm.s-1) were tested to observe the impact on transport under favorable and unfavorable conditions. The difference in retention between AML (net favorable) and CML (net unfavorable) microsphere attachment was 25% for the 200 nm microspheres and 13% for the 1000 nm microspheres. Less than 50% of the AML microspheres were retained in the fracture, postulated to be due to the effects of mineralogical heterogeneity and fracture surface roughness. The effect of an increase in ionic strength in increasing retention was significant for unfavorable attachment, but was not significant for favorable attachment conditions. The effect of specific discharge was minor for all but the 200 nm CML microspheres at 10 mM ionic strength. When flushing the fracture first with cationic microspheres, then with anionic microspheres, the recovery of anionic microspheres resembled favorable attachment presumably due to interaction with cationic microspheres that remained attached to the sandstone surface. Colloid breakthrough curves could be fit well with a two site attachment model, with reversible and irreversible sites.

Simulation of Colloid Transport and Retention Using a Pore‐Network Model With Roughness and Chemical Heterogeneity on Pore Surfaces

AbstractColloid transport and retention in porous media is a common phenomenon in both nature and industry. However, many questions remain on how to obtain colloid transport and retention parameters. Previous work usually assumed constant transport parameters in a medium under a given physicochemical condition. In this study, pore‐network modeling is employed to upscale colloid transport and retention from the pore‐scale to the macro‐scale. The pore‐scale transport parameters including the collection efficiency (η), the sticking efficiency (α), and the fraction of the solid‐water interface that contributes to the colloid attachment (Sf) are obtained using numerical simulation and probability analysis for each pore throat. The influence of roughness and charge heterogeneity on the distribution of pore‐scale parameters is discussed. Breakthrough curves and the retention profiles under different roughness and charge heterogeneity conditions are also analyzed. Results show that pore‐scale parameters η, α, and Sf have various distributions in porous media that may not be accurately described using single‐valued effective parameters. The value of η decreases with velocity and exhibits a wide distribution under low‐velocity conditions. The parameter α tends to decrease with the colloid size and the pore water velocity and increased with the charge heterogeneity fraction. Nanoscale roughness alters α in a non‐monotonic fashion but tends to increase for lower roughness fractions and zeta potential. Microscopic roughness increases values of α for colloids that would otherwise be susceptible to hydrodynamic removal. Breakthrough curves and retention profiles show that more retention occurs for smaller particles, which reflects the influence of blocking.

Stochastic Collision-Attachment-Based Monte Carlo Simulation of Colloidal Fouling: Transition from Foulant-Clean-Membrane Interaction to Foulant-Fouled-Membrane Interaction.

The initial behavior of colloidal fouling is governed by foulant-clean-membrane interaction (F-M), and its long-term behavior is determined by foulant-fouled-membrane interaction (F-F). Nevertheless, the transitional fouling behavior from F-M to F-F has not been fully understood. This study reports a novel collision attachment (CA)-Monte Carlo (MC) approach, with the stochastic colloid-membrane collision events modeled by MC and the probability of colloidal attachment to the membrane determined by the interplay of flux and the energy barrier arising from colloid-membrane interaction (Em for F-M and Ef for F-F). The long-term membrane flux remains stable for large Ef, whereas severe fouling occurs when both Em and Ef are small. Our study reveals the existence of a metastable flux behavior for the combination of large Em but small Ef. The time evolution of flux behavior and colloidal deposition patterns shows a nearly constant flux for an extended period, with the high energy barrier Em retarding initial colloidal deposition. However, accidental random deposition of a colloidal particle could reduce the local energy barrier (toward the smaller Ef), seeding for further colloidal deposition in its vicinity. This initiates an uneven patch-wise fouling and eventually leads to a complete transition to F-F-dominated behavior. The metastable period can be effectively extended by increasing the energy barrier (Em or Ef) or lowering flux, which provides important implications to membrane design and operation.

Colloidal fouling in electrodialysis: A neural differential equations model

The attachment of colloids to the ion-exchange membranes in electrodialysis is an important hurdle when processing bio-based process streams. Previous research showed that fouling strongly depends on the crossflow velocity, the current and the salt concentration of the medium. Predicting the influence of these variables on the fouling rate is challenging due to the complex physics at play and optimising the process conditions to reduce fouling remains a challenge. The objective of this study is the development of a model to predict the dynamic behaviour of electrodialysis fouling under varying process settings to facilitate this optimisation. A neural differential equation is fit to experimental data of an electrodialysis pilot undergoing humic acid fouling. We show that this model can predict the fouling rate even when using a limited set of experimental data. The robustness of the model is demonstrated by a simulation study and a sensitivity analysis indicates that the crossflow velocity is the most important variable influencing the fouling rate (≈40%). Both the effect of the current (≈20%), the salt concentration (≈13%) and their interaction effects are considerable. With the model, the evolution of the stack resistance as a result of membrane fouling can be simulated, facilitating process control or decision-making.

A proximity-based image-processing algorithm for colloid assignment in segmented multiphase flow datasets.

Colloidal transport and deposition are of both environmental and engineering importance. Easier access to x-ray microtomography (XMT) coupled with improved imaging resolution has made XMT a unique and viable tool for visualizing and quantifying these processes. Currently, there is scant information in the literature addressing colloid segmentation and analysis in saturated and unsaturated porous media, in particular related to spatial partitioning of colloids. To support this need, an approach to assign segmented colloidal particles and aggregates to different partitioning classes based on their proximity to different phases is presented here. The method uses different markers for each attachment site (e.g. wetting-nonwetting phase interfaces). An example XMT dataset from a drainage experiment is used to demonstrate the efficacy of the image processing algorithms. Flow conditions, and fluid and colloid properties, can thus be compared to the behaviour of colloids within the porous medium. This algorithm can help elucidate colloidal deposition mechanisms and the importance of different attachment sites, explore the importance of fluid properties, as well as the arrangement and shape of the colloids.

Colloids at Fluid Interfaces

Over the last two decades, understanding of the attachment of colloids to fluid interfaces has attracted the interest of researchers from different fields. This is explained by considering the ubiquity of colloidal and interfacial systems in nature and technology. However, to date, the control and tuning of the assembly of colloids at fluid interfaces remain a challenge. This review discusses some of the most fundamental aspects governing the organization of colloidal objects at fluid interfaces, paying special attention to spherical particles. This requires a description of different physicochemical aspects, from the driving force involved in the assembly to its thermodynamic description, and from the interactions involved in the assembly to the dynamics and rheological behavior of particle-laden interfaces.

Factors affecting the transport of petroleum colloids in saturated porous media

Batch and column experiments were performed to evaluate the responses of attachment and transport of petroleum colloids to the changes in solution chemistry conditions. We found that transport of petroleum colloids was inhibited with increasing the ionic strength or decreasing the pH. This observation can be explained by the increase in particle size of petroleum colloids and the reduction in electrostatic repulsion between colloids and grain surfaces. Meanwhile, the extent of transport inhibition by ionic strength depended on the type of porous media, that is, greater mobility of petroleum colloids could be found in sand rather than in soil at the same ionic strength. This can probably be ascribed to surface chemistry of soil minerals. In the case of divalent ion (e.g., Ca2+), the retention of petroleum colloids was more obvious in soil than in sand resulting from the complexing between the oxygen functional groups from both petroleum colloids and grains, i.e., petroleum colloids could bind to the sand/soil grains through Ca2+ bridging. Additionally, the presence of sodium dodecyl sulfate (SDS, selected as a model surfactant) significantly enhanced the mobility of petroleum colloids as a result of enhanced steric hindrance that remarkably inhibited aggregation of petroleum colloids. Also, a two-site transport model was used to fit the transport data to gain further insights on deposition mechanisms. These findings from this research are helpful to understand the transport and fate of petroleum colloids in subsurface systems.

Mechanism of Colloidal Attachment on Textile Fibrous Media

Mechanism of Colloidal Attachment on Textile Fibrous Media

A Unified Force and Torque Balance for Colloid Transport: Predicting Attachment and Mobilization under Favorable and Unfavorable Conditions.

Colloid attachment and detachment behaviors concern a wide range of environmental contexts but have typically been mechanistically predicted exclusive of one another despite their obvious coupling. Furthermore, previous mechanistic prediction often addressed packed column contexts, wherein specific forces and torques on the colloid could not be well-constrained, preventing robust predictions. These weaknesses were addressed through direct observation experiments under conditions where perfect sink assumptions fail and allow calibration of the contact between the colloid and collector. Attachment and flow perturbation experiments in the presence of colloid-collector attraction (favorable conditions) permitted calibration of contact parameters without the complexity that comes with colloid-collector repulsion (unfavorable conditions). Combining calibrated contact parameters with discrete representative nanoscale heterogeneity, developed to predict unfavorable attachment, provided an independent means to predict unfavorable detachment. The result was mechanistic prediction of colloid attachment and detachment that quantitatively agreed with experimental observation for both ionic strength and flow perturbation results, improving significantly upon previous qualitative prediction.