High-speed discrete broadband nanomechanical mapping of soft materials using atomic force microscope
Description
TitleHigh-speed discrete broadband nanomechanical mapping of soft materials using atomic force microscope
Date Created2022
Other Date2022-01 (degree)
Extent187 pages : illustrations
DescriptionNanoscale mechanical properties quantification of soft and live biological materials using an atomic force microscope (AFM) is essential in investigating their mechanics, structures, and functionalities. More trending studies are focused on the dynamic evolution of the nanomechanical property of the sample in a large area. However, current methods are unable to obtain the temporal resolution of the sample based on time-consuming continuous scanning. Moreover, the nanomechanical measurement is limited to only a single frequency. Thus, a rapid discrete broadband nanomechanical mapping (RB-DNM) is proposed to capture the temporal resolution, especially over a large area, by reducing the number of measurements in the spatial dimension. Only discrete sampling locations of most interest are measured in the discrete mapping, which also provides flexibility in choosing the number, spatial-temporal distribution, and frequency spectrum of measurements. However, challenges arise in RB-DNM, where frequent probe engagement, withdrawal, and transition at high speed are necessary to quantify the sample undergoing dynamic processes: (1) Probe engagement and withdrawal are hard to conduct at high speed with limited induced force. Due to the highly nonlinear force-distance relation, large probe-sample interaction force can be generated, especially at high speed, resulting in sample deformation, damage, and measurement errors. (2) Extra difficulties arise during the rapid probe transition along multiple dimensions over an extensive range. The dynamics of the AFM instrument can be excited, and the lateral-vertical cross-axis dynamics coupling effects become pronounced, leading to significant errors in the probe positioning. (3) The highly nonlinear dynamic material behavior can hardly be estimated online, resulting in challenges for effective spatial-temporal mapping across a large area with limited measurement capacity. The variations in material topography and property induce disturbance to the probe-sample relative positions for maintaining the same layer indentation, which is worsened under adverse effects from the piezo actuator drift. Additional challenges, e.g., hydrodynamic force and the thermal drift, arise in characterizing guard cell wall mechanics under the stomatal movements in vivo. Challenges lie in characterizing the guard cell walls with complicated structural features under the influence of the growth stage, biochemical physiology, and genetic mechanisms.
This dissertation presents a set of dynamics and control tools to tackle the above challenges in the nanomechanical mapping of soft materials using AFM. We propose an online-searching-based optimization approach for rapid probe engagement and withdrawal with probe-sample interaction force minimized. First, the force-displacement profile of the probe is partitioned into phases based on the characteristics of probe-sample interaction. Then, each phase is online optimized to minimize the interaction force and further the engagement/withdrawal time. The online optimization is conducted through a Fibonacci-based iterative searching (FIS) process. The optimal transition trajectory design and the modeling-free inversion-based iterative control technique are immersed in the Fibonacci search algorithm. The proposed approach is demonstrated through experimental implementations on a polydimethylsiloxane (PDMS) sample and a dental silicone sample, respectively, compared to the conventional methods. Built upon this rapid probe engagement and withdrawal technique, next the rapid broadband discrete nanomechanical mapping of soft materials is created. An online-searching learning-based technique is utilized at each sampling location to achieve rapid probe engagement and withdrawal with interaction force minimization. Then, a control-based nanoindentation measurement technique is used to acquire its nanomechanical property. Finally, a decomposition-based learning approach is explored to achieve rapid probe transitions between the sampling locations. The method is demonstrated through experiments using a PDMS sample and a PDMS-epoxy sample as examples. To further enhance the spatial and temporal resolution of the discrete nanomechanical mapping under limited capacity, we developed an adaptive discrete nanomechanical mapping (ADNM) based on the online estimation of material property variance. The effectiveness and efficiency of the method are validated by capturing the dynamic crystallization process of a poly(3-hydroxybutyrate-co-3-hydroxyvalerate) (PHBV) polymer. Finally, we optimized rapid broadband discrete nanomechanical mapping (ORB-DNM) to characterize the dynamic evolution of material properties. An online estimation-based feedforward-feedback control was developed to maintain the consistency of spatial-temporal mapping via the same layer indentation and address the adverse effects. Its efficacy is demonstrated by online evaluating the time-varying viscoelastic property of stomatal guard cell walls of Arabidopsis thaliana plants in vivo during stomatal movements for biophysical insights.
NotePh.D.
NoteIncludes bibliographical references
Genretheses
LanguageEnglish
CollectionSchool of Graduate Studies Electronic Theses and Dissertations
Organization NameRutgers, The State University of New Jersey
RightsThe author owns the copyright to this work.