MobiDrop Inc is an rapidly-growing hi-tech company affiliated to Weitz Lab, Harvard University. Its main product is all-in-one droplet digital PCR device, which combines all steps in ddPCR experiments into one device by using a deeply designed microfluidic chip.Microfluidic Chip design and fabrication: Accelerate the iteration of microfluidic chip. Use soft lithography method to fabricate chips for proof-of-concept. Then guide the high volume production by injection method. Inspect the channel width and depth by microscope and microfluidic experiment.Droplet dPCR experiments and data analysis: Perform droplet dPCR experiments in Weitz Lab. Generate droplets by our chip. Observe droplet morphology and flow rate by high speed camera. Detect DNA signal with a laser focusing system and photomultiplier(PMT). Use Python code to analyze videos and PMT output voltage. Verify the droplet generation, PCR process and fluorescence detection functions.Design and Set up Prototype Device: Design and purchase parts for prototype device. Assemble it and run as an experiment platform. Finish the testing report for a presentation in medical device conference.Optical system optimizing and innovation: Reduce the number of lens in the laser focusing and fluorescence detection system. Innovate a multi-PMT system which could detect three kinds of DNA simultaneously.

Research Objectives: Design and fabricate textile-based screen-printed biosensor for portable and highly-sensitive glucose detection in human sweat. Combine the sensor and T-shirt to form a real-time monitoring system.Fabrication and Testing: Used screen-printing method to fabricate sensors. Used AutoCAD to design sensor masks and fabricated them by laser cutting. Finished enzyme immobilization process by drop casting. Applied electrochemical method, used GF-reader to measure CV plot and amperometric plot for each sensor. Drew signal-concentration plot to evaluate the performance of sensor. Improved signal to noise ratio for better reliability. Method Optimization: Used graphene nanoplatelets (GnP) to enhance electron transfer between glucose oxidase and the electrode surface. Increased sensitivity for more than 20 times. Reduced the detection limit from 20mM to 1mM. Optimized the enzyme immobilization method to reduce background noise and STDEV of signal. Research Outcome: The sensor could detect glucose concentration from 1 mM to 200 mM with 1 mM detection limit. It could be stored at least 48 hours in room temperature or 7 days at 4 ℃. The response time has been reduced to 30 seconds (current commercial products usually take 1 to 2 minutes).

Project Objectives: Fabricate 5 types of micro-devices (including capacitor, resistor, diodes, solar cells and MOSFET) on a silicon wafer and get trained with cleanroom procedures and thin film process. Fabrication Process: The fabrication steps are as follows, diffusion doping, depositing silicon oxide and aluminum layer (PECVD), creating masks with PR, etching unwanted parts (photolithography). Device Testing and Simulation: Tested the device performance, such as capacitance and photoelectric conversion efficiency. Used SPROCESS tool in Sentaurus to simulate the device fabrication process and performance.Project Outcomes: Finished 5 devices for each kind, at least 80% are functional. There was less than 10% error between testing results and simulation results.

Research Objectives: Research the characteristic of Bacillus spores. Apply water-response material and design evaporation-driven engines for energy harvesting and storage. Literature Research: Summarized the scaling up evaporation-driven engines development and the potential for natural evaporation as a reliable renewable energy resource. Wrote technical progress reports. Material Characteristics Research: Used AFM to measure the work generated by expansion and contraction of spores in response to changing relative humidity (RH). Calculated the energy density of spores and found out that it was 10-100 times higher than some other stimuli-responsive materials (such as thermal expansion). Artificial Muscle Design: Designed artificial muscle by using spinning method (such as melt spinning and wet spinning). Optimized elastic modulus of polymer by combine spores and other polymer materials.

Drafted the Confidentiality and Nondisclosure Agreement between Goldwind Science & Technology Co., Ltd and Sentient Corporation;Translated in the technique sharing group between Goldwind Science & Technology Co., Ltd and Sentient Corporation;Translated and checked DNVGL-SE-0439 “Certification of condition monitoring”;Translated and checked “Guideline for the Certification of Condition Monitoring Systems for Wind Turbines” ;Checked the differences between VDI3834 2009 version and 2015 version;

Project Objectives: Find the optimized recipe to finish single layer CNT growth on stainless steel mesh by APCVD, reach the highest cover area (no less than 75%). Recipe Optimization: Successfully increased cover area from 40% to 95%, by changing reactant ratio, reaction time and adding pre-treatment process. The ratio of single layer CNT is higher than 80% (current commercial products have the ratio of 75% to 85% by using LPCVD). SEM Photos: Took SEM images for each SLCNT sample. Observed CNT morphology and cover area to evaluate the recipe. Academic Report: Finished research progress reports and made presentation in group meetings.

Worked in DPI of Tsinghua University, MicroSatellite Research Group, Intelligent Microsystem LabLapped the circuit of the satellite power subsystem;Did the functional test on communication protocol of TTC and ADCS;Designed energy project for separation of micro-satellite and pico-satellite;Set up the experimental environment for subsystem testing;