Estimation of local concentration from measurements of stochastic adsorption dynamics using carbon nanotube-based sensors
Name
11814_2015_Article_124.pdf
Size
1.89 MB
Format
Adobe PDF
Checksum (MD5)
0e66056529c3a4657ad12dd8e6769b63
Author(s)
Jang, Hong
Lee, Jay H.
Braatz, Richard D
Date Issued
August 2015
Journal
Korean Journal of Chemical Engineering
Publisher
Springer US
Citation
Jang, Hong, Jay H. Lee, and Richard D. Braatz. “Estimation of Local Concentration from Measurements of Stochastic Adsorption Dynamics Using Carbon Nanotube-Based Sensors.” Korean J. Chem. Eng. 33, no. 1 (August 18, 2015): 33–45.
Version
Author's final manuscript
Abstract
This paper proposes a maximum likelihood estimation (MLE) method for estimating time varying local concentration of the target molecule proximate to the sensor from the time profile of monomolecular adsorption and desorption on the surface of the sensor at nanoscale. Recently, several carbon nanotube sensors have been developed that can selectively detect target molecules at a trace concentration level. These sensors use light intensity changes mediated by adsorption or desorption phenomena on their surfaces. The molecular events occurring at trace concentration levels are inherently stochastic, posing a challenge for optimal estimation. The stochastic behavior is modeled by the chemical master equation (CME), composed of a set of ordinary differential equations describing the time evolution of probabilities for the possible adsorption states. Given the significant stochastic nature of the underlying phenomena, rigorous stochastic estimation based on the CME should lead to an improved accuracy over than deterministic estimation formulated based on the continuum model. Motivated by this expectation, we formulate the MLE based on an analytical solution of the relevant CME, both for the constant and the time-varying local concentrations, with the objective of estimating the analyte concentration field in real time from the adsorption readings of the sensor array. The performances of the MLE and the deterministic least squares are compared using data generated by kinetic Monte Carlo (KMC) simulations of the stochastic process. Some future challenges are described for estimating and controlling the concentration field in a distributed domain using the sensor technology.
MIT Department
Massachusetts Institute of Technology. Department of Chemical Engineering
Terms of Use
Creative Commons Attribution-Noncommercial-Share Alike
Persistent DSpace Link
DOI of Published Version
https://doi.org/10.1007/s11814-015-0124-9
