Various promising concepts exist for improving the performance of productions of pure enantiomers. An efficient approach is developed for the systematic conceptual design of such processes. The proposed three-step procedure aids a fast selection of the optimal process configuration out of many possible candidates and leads to an optimally designed process. The approach is applied in a case study for an industrially relevant compound, considering different process concepts based on simulated moving bed chromatography, enantioselective crystallization, and racemization. It is demonstrated that mixed-integer nonlinear programming is capable of predicting simultaneously the optimal process configuration and optimal design parameters.

Combining an enantioseparation by continuous chromatography with the racemization of the undesired enantiomer in a recycle is a promising approach. Two different cases were considered: limited conversion of the reaction with moderate product purity requirements and high conversion with high product purity requirements. For the first case, a simple plantwide control strategy is presented, manipulating the solvent removal in front of the racemization reactor. It is shown that the various disturbances can be compensated with this strategy without performing the more complex readjustment of the chromatographic unit. Further, it is shown that this strategy is likely to fail in the second scenario and that readjustment of the chromatographic unit is required for a successful control strategy.

The combination of continuous chromatographic processes and subsequent crystallization with a recycle of the mother liquor can improve significantly the efficiency of separation processes for the production of enantiomers if compared to the stand-alone chromatographic process. For the first time, dynamic operability and plantwide control of such a process combination is investigated. In the first step, the effect of disturbances on the open loop dynamics is evaluated. In a second step, a simple plantwide control strategy is proposed in order to ensure a robust process operation. It is demonstrated that the direct control of the simulated moving bed (SMB) unit is not required to stabilize the process combination while maintaining the desired product specifications. Instead, this can be achieved easily by controlling the amount of solvent removed or added to the system.

It is commonly understood that preservation of Lithium-ion battery life comes at the expense of charge time. In this manuscript a novel real time charging framework has been developed that charges the battery in a time as fast as the constant current-constant voltage (CC-CV) protocol while ensuring minimum loss of cycle life. The framework consists of a reduced order electrochemical model integrated with an optimization routine. The charge profile adapts to the battery state of charge and health in real time and is amenable for on-device implementation. Cycle life benefits of the novel adaptive charging profile are compared with traditional CC-CV charging of commercial Lithium-ion batteries cycled until end of life. In addition to the running capacity, battery health is also compared at a low probe C-rate(C/5) to assess the benefits in minimizing irreversible losses. A differential voltage analysis of Probe (C/5 rate) reveals that adaptive charging is able to eliminate the loss of active material (LAM) that is seen typically at advanced cycles. The framework is implemented on a mobile device and shows a substantial benefit in battery life without any penalty on the charge time.

Estimating the state of health (SOH) of batteries powering electronic devices in real-time while in use is a necessity. The applicability of most of the existing methods is limited to the datasets that are used to train the models. In this work, we propose a generic method for SOH estimation with much wider applicability. The key problem is the identification of the right feature set which is derived from measurable voltage signals. In this work, relative rise in voltage drop across cell resistance with aging has been used as the feature. A base artificial neural network (ANN) model has been used to map the generic relation between voltage and SOH. The base ANN model has been trained using limited battery data. Blind testing has been done on long cycle in-house data and publicly available datasets. In-house data included both laboratory and on-device data generated using various charge profiles. Transfer learning has been used for public datasets as those batteries have different physical dimensions and cell chemistry. The mean absolute error in SOH estimation is well within 2% for all test cases. The model is robust across scenarios such as cell variability, charge profile difference, and limited variation in temperature.