Used In The Design And Study Of Bio-chip Cmos Fluorescence Detection System

Fluorescence-based method is one of the most important detecting techniques in various biological experiments.The fluorescent detecting chips based on Si technology attract widely concern in lots of detecting experiments because of their low cost,low power and high sensitivity.This thesis focuses on the study and design of a CMOS fluorescent detector chip,and the improvement of the fluorescent experiment scheme.The main contributions of this thesis are concluded as following:On the design of the fluorescent detecting system,this thesis firstly establishes a photon-electron conversion model of the standard CMOS photodiodes based on some available technology parameters.The mathematic formulation of the conversion efficiency is also deduced and validated by simulation and testing.Furthermore,the comparisons of the three CMOS compatible photodiodes(N+/Psub photodiode,Nwell/Psub photodiode and P+/Nwell photodiode) on the sensitivity,peak response wavelength and dark current performance are given,which provide the theoretical basis for choosing proper photodectors in the detecting experiment.Secondly,a new capacitive trans-impedance amplifier(CTIA) is proposed to convert the photocurrent generated by weak fluorescence to voltage.This architecture has a higher photocurrent-voltage conversion gain and helps the photodiode have a better dark current performance.The proposed T-type reset switch avoids the leakage current influence effectively during the integrating period.Finally,a 12-bit, 10-MS/s pipelined analog-to-digital converter(ADC) is employed to realize the digitization outputs of the detecting chip.In order to achieve low power and small chip area,some special techniques have been adopted in this ADC.These techniques include the elimination of the conventional S/H circuit;the adoption of 2.5-bit/stage according to the noise model and power model in analog chain;the commutated feedback-capacitor switching(CFCS) technique to release the sampling capacitor mismatch requirement in the stage circuit; dynamic comparators with sampling capacitors and pre-amplifiers;gain boosting cascode amplifiers and so on.The proposed fluorescent detecting system is implemented in SMIC 0.18-μm standard CMOS process.This chip occupies an area of 3 mm~2(including PAD area) and consumes 37 roW.Experiments show that,the proposed photon-electron conversion model of the CMOS compatible photodiodes identifies the testing results.The Nwell/Psub photodiode used in the fluorescence detecting experiment has the maximum sensitivity of 0.176 A/W at wavelength of 580 nm.The CTIA has a charge-voltage conversion gain of 1.6μV/e~- and helps the Nwell/Psub photodiode to achieve a dark current of 300 fA(3 nA/cm~2) with 300 mV reverse biased voltage.The maximum DNL and INL of the designed ADC are +0.8 LSB and -3 LSB,respectively.A signal to noise and distortion ratio(SNDR) of 56.2 dB and a spurious free dynamic range(SFDR) of 61.7 dB are achieved at 1 MHz input frequency and 10 MHz sampling rate.On the testing method of the fluorescence detecting experiment,a "contact imaging" detecting method,which puts the reaction vessel onto the surface of the detecting chip directly,is adopted to avoid the fluorescence loss due to the additional optical instruments and paths.Testing results indicate that,the detector system cooperating with the "contact imaging" detecting method can detect the fluorescence generated by fluorescein solution successfully at room temperature.With a solution volume of 2 ml and an integrating time of 50 ms,the detecting chip is sensitive to the fluorescence(3.5 nW/cm~2) excited from the fluorescein solution with a concentration of 20 ng/ml,and can generate a photocurrent of 7 fA within the photodetector.While with a solution volume of 0.5μl and an integrating time of 10 ms,the detecting chip is also sensitive to the fluorescence(18 nW/cm~2) excited from the fluorescein solution with a concentration of 625 ng/ml,and generates a photocurrent of 36fA.