Tsinghua Science and Technology Tsinghua University, China ISSN: 1007-0212 Vol. 6, Num. 3, 2001, pp. 269-272

Tsinghua Science and Technology, Vol. 6, No. 3, August 2001 pp. 269-272

Temperature Control System for Biochemical Reactions in Microchip-Based Devices

JING Gaoshan ¹, ZHANG Jian¹ , ZHU Xiaoshan ¹, FENG Jihong ¹ , TAN Zhimin 1,3 , LIU Litian ^1,3 ,  CHENG Jing ^1,2

¹Beijing National Biochip Research and Engineering Center; Jia 2# Qinghua West Road, Beijing 100084, China;
² State Key Laboratory of Biomembrane and Membrane Biotechnology, School of Life Science and Engineering,  Tsinghua University, Beijing 100084,  China;   
³Institute of Microelectronics, Tsinghua University, Beijing 100084, China   

Received: 2000-11-21

Code Number: ts01080

Abstract:   

A silicon-glass chip based microreactor has been designed and fabricated for biochemical reactions such as polymerase chain reactions (PCR). The chip based microreactor has integrated resistive heating elements. The computer-controlled temperature control system is highly reliable with precise temperature control, excellent temperature uniformity, and rapid heating and cooling capabilities. The development of the microreaction system is an important step towards the construction of a lab-on-a-chip system.

Key  words:  temperature control; microreactor; microchip; polymerase chain reactions (PCR)

Introduction   

In recent years, the development of miniaturized systems for chemical and biochemical reactions has rapidly progressed[1]. Most of the microdevices for conducting these reactions are made of silicon or glass using conventional microfabrication methods used in the microelectronic industry or by modifying the processing technologies for micro electromechanical systems (MEMS). Several research groups have developed polymerase chain reactions (PCR) chips with various features to perform simple PCR[2] , real time PCR[3] , ligase chain reactions[4] , reverse transcription-PCR[5] , and degenerated oligonucleotide primed-PCR (DOP-PCR)[6]. Precise temperature control is very important to perform chemical or biomedical reactions in a miniaturized chamber. There are different ways for heating and cooling[7]. Compared with conventional approaches for performing biochemical reactions, performing biochemical reactions in microchips has several advantages. Mass production will significantly reduce the cost of the silicon based microdevices. Reaction speed in the silicon-based microchips is fast because of small reagent volumes, high surface to volume ratio and excellent thermal properties (thermal conductivity is 157 WDK·m^-1 DK·K^-1 )[8]. The reaction process can be readily controlled and monitored automatically through microsensors and actuators, and contamination can be reduced to minimum. The cost is low because of reduced consumption of reagents. Silicon based microreactors are especially suitable for biochemical reactions such as PCR where accurate temperature control is demanded.

This paper describes the construction of a microreaction system, which will be an important part of a lab-on-a-chip system. The system is composed of a silicon microchip, a platinum thermal sensor for temperature monitoring and a computer-controlled thermoelectric cooler (TEC) for heating and cooling.

Water, ethanol, and dimethylformamide (DMF) have been used as test liquids. All the system thermal requirements have been satisfied including heating and cooling speeds, temperature accuracy, and repeatability of thermal cycles.

1 Materials and Methods  

The microreaction system consists of a silicon based microreactor, a platinum temperature sensor, a TEC (DT 6-2.5, Marlow Industries, Inc., Dallas, Texas), a custom-made power amplifier circuit to drive the TEC and to convert the temperature signal into a voltage signal, a data acquisition (DAQ) card (AT-MIO-16E-10, National Instruments, Austin, Texas) for online monitoring of the microreactor temperature, and a computer with a program generated using LabVIEW (National Instruments, Austin, Texas) to control the whole microreaction system.

The silicon based microreactor was fabricated by wet etching and photolithographic methods. Six reaction chambers were etched on a silicon wafer having diameter of 2 inch. After slicing, each reaction chip had a chamber size of 10.8 mL and outside dimensions of 12 mmx9 mmx100 mm. The reaction chamber must be airtight for the microreaction. The microreaction chamber was formed using an anodic bonding method. First, a Pyrex glass slide with dimensions of 16 mmx11 mm was prepared with two drilled holes having diameters of 0.6 mm. The silicon chip and the Pyrex glass were placed into ethanol (100%) overnight and then rinsed with deionized water. The Pyrex glass was then air-dried and placed on top of the dried silicon chip. The silicon-glass complex was heated to  500 °C  on a hot plate (Model PC-200, Corning Inc., Corning, New York) and then a 1000 V DC voltage was applied using a customized power supply with the silicon chip connected to the positive electrode and the glass connected to the negative electrode. Anodic bonding formed an airtight microreaction chamber in seconds. The two holes in the Pyrex glass cover are used as the inlet and outlet. Two plastic tubes (Ellsworth Adhesive System, Germantown, Wisconsin) were attached to the holes using UV glue (NOA 68, Norland Products, Inc., New Brunswick, New Jersey), Fig.1.

The  temperature change was measured by the change of the platinum sensor resistance. The linear correlation between the temperature and the resistance can be represented by:

R_t=R_o x [1+a(T-T_o)] ,

where Rt is the sensor resistance (W) at temperature T (°C), Ro is the resistance (W) at the reference temperature To (°C) and a is the temperature coefficient of resistance (TCR, (°C)^-1 ) of Pt. For the current design,Ro is 1000 W, To is  0 °C  and a is 0.0039 (°C)^-1 .

The Pt temperature sensor and the microreactor were both placed on top of the TEC, Fig.2. Both were glued to the TEC with thermal glue for good heat transfer. The TEC is a solid state heat pump which operates on the Peltier effect. The TEC can heat or cool the microreactor by changing the polarity of the DC voltage applied to the TEC. A power amplifier was used to drive the TEC. The LabVIEW based program acquired the temperature signal through the DAQ card and controlled the output of the power amplifier circuit. The input signal was amplified and processed through a normal feedback proportional integral (PI) algorithm for improved thermal cycling performance.

2 Results and Discussion

Because the temperature measured by the sensor was not identical to the temperature in the microreactor chamber, the two temperatures had to be calibrated. A series of temperatures from the Pt sensors and the associated temperature in the reaction chamber were measured simultaneously with a precise thermal meter (HH23, Omega Engineering, Inc., Stamford, Connecticut). The thermal meter precision was  ±0.1 °C . The two temperature values were calibrated using the LabVIEW-based program. Water, ethanol and DMF were used as the test liquids to testvarious thermal characteristics of the microreaction system such as temperature precision, heating and cooling speeds, and repeatability of the thermal cycles. The testing was designed to stimulate the thermal cycles of conventional PCR reactions. Each cycle was divided into three temperature zones:  55 °C  for 30 s,  72 °C  for 30 s, and  95 °C  for 30 s.

The temperature was monitored and controlled by a PI algorithm, so the PI pa rameters played a vital role in the temperature response of the microreaction system. To obtain better temperature control, specific PI parameters were chosen for each predetermined temperature value. Ziegler-Nichols Tuning[9]  was used to adjust these parameters. The parameters for the different set temperatures could be adjusted when needed. Important information about the thermal characteristics of the microreaction system was obtained using this way. The system temperature precision is  ±0.1 °C . Considering the precision of the thermal meter, the absolute value of the system precision is no better than  0.2 °C , which just met the prerequisite for a highly temperature-sensitive PCR reaction.

Temperature ramping is another important thermal characteristic of the microreaction system. The ramping speeds were different in different temperature zones. From  72 °C  to  95 °C , the average heating speed was  3 °CDK·s^-1  , whereas from  55 °C  to  72 °C  the average heating speed was  6 °C·s^-1  . The average cooling speed was  3 °CDK·s^-1  . Compared with results reported by other groups, our system ramping rate was relatively low[10-12] , and unlike other systems where the temperature sensor and the heater/cooler were integrated into one part, our system was relatively simple. The cost of the microreactor allows it to be disposable in future practical use.

Overshoot of the set temperature must be controlled in an effective microreaction system. The overshoot values differed for the different set points. For  55 °C  the overshoot value was  2 °C  while for  72 °C  and  95 °C  the overshoot values were less than  1 °C .

Different controlling strategies were used to adjust the parameters of different set points. In adjusting the normal PI parameters, a conflict occured between the temperature response and the overshoot. Increasing Kp or increasing K (here, Ki is the reciprocal of conventional integral parameter) resulted in a faster temperature response but larger overshoot. On the contrary, decreasing Kp or decreasing K resulted in a much slower temperature response but much less overshoot at the set point. For temperatures above  90 °C , the overshoot had to be minimized so that the reagent in the microreaction chamber would not evaporate orboil which would lead to the disastrous result of enzyme inactivation. For temperatures below  80 °C , the parameters were tuned for a high ramping rate with some overshoot.

The results shown in Fig.3 demonstrate the good repeatability of the thermal cycles, while the results in  Fig.4  demonstrate the accurate temperature control and rapid heating and cooling speeds.

3 Conclusions  

A silicon chip-based microreaction system was constructed and the system thermal characteristics were studied by choosing water, ethanol, and DMF as test liquids. The PI parameters were carefully chosen for accurate temperature control. The microreaction system has relatively rapid heating and cooling speeds which could reach 3 to 6 °CDK·s^-1 . The overshoot value was constrained to less than an average of  2 °C . The system precision was high, i.e., less than  0.2 °C , in general. Compared with similar microreaction systems, the microchip was easier to fabricate and the system was compact and easy to control. A real PCR reaction will need additional work such as surface passivation and other effects[2,13,14]. After successful simulation of the PCR reaction conditions, a real PCR reaction will soon be performed in this microreaction system. Once successful, this system will become an important part of a lab-on-a-chip system.

Acknowledgements

We are grateful to Mr. Xu Junquan for his help.

References

1. Cheng J, Fortina P, Surrey S, Kricka L J, Wilding P. Microchip-based devices for molecular diagnosis of genetic diseases. Molecular Diagnosis, 1996, 1: 183-199.
2. Lao A I K, Lee T M H, Carles M C, Hsing I-M. Thermal management and surface passivation of a miniaturized PCR device for traditional Chinese medicine. In: Proceedings of the mTAS 2000 Symposium, held in Enschede. The Netherlands, 14-18 May, 2000. 139-142.
3. Taylor T B, Winn-Deen E S, Picozza E, Woudenberg T M, Albin M. Optimization of the performance of the polymerase chain reaction in silicon-based microstructures. Nucleic Acids Research, 1997, 25: 3164-3168.
4. Cheng J, Shoffner M A, Mitchelson K R, Kricka L J, Wilding P. Analysis of ligase chain reaction products amplified in silicon-glass chip using capillary electrophoresis. Journal of Chromatography A, 1996, 732: 151-158.
5. Chang C, Perkins S, Wittwer C. Measurement of MDRI gene expression by real time quantitative RT-PCR. Modern Pathology, 1999, 12:190A.
6. Cheng J, Waters L C, Fortina P, Hvichia G, Jacobson S C, Ramsey J M, Kricka L J, Wilding P. Degenerate oligonucleotide primed-polymerase chain reaction and capillary electrophoretic analysis of human DNA on microchip-based devices. Analytical Biochemistry, 1998, 257: 101-106.
7. Oda R P, Strausbauch M A, Huhmer A F R, Borson N, Jurrens S R, Craighead J, Wettstein P J, Eckloff B, Kline B, Landers J P. Infrared-mediated thermocycling for ultrafast polymerase chain reaction amplification of DNA. Analytical Chemistry, 1998, 70: 4361-4368.
8. Lao A I K, Lee T M H, Hsing I-M, Ip N Y. Precise temperature control of microfluidic chamber for gas and liquid phase reaction. Sensors and Actuators, 2000, 84: 11-17.
9. Franklin G F, Powell J David, Emami-Naeini Abbas. Feedback Control of Dynamic Systems. Third Edition. Addison-Wesley Publishing Company, 1994. 191-196.
10. Khandurina J, McKnight T E, Jacobson S C, Waters L C, Foote R S, Ramsey J M. Integrated system for rapid PCR-based DNA analysis in microfluidic devices. Analytical Biochemistry, 2000, 72:  2995-3000.
11. Northrup M A, Benett B, Hadley D, Landre P, Lehew S, Richards J, Stration P. A miniature analytical instrument for nucleic acids based on micromachined silicon reaction chambers. Analytical Biochemistry, 1998, 70: 918-922.
12. Poser S, Schulz T, Dilner U, Baier V, Kohler J M, Schimkat D, Mayer G, Siebert A. Chip elements for fast thermocycling. Sensors and Actuators A, 1997, 62: 672-675.
13. Cheng J, Shoffner M A, Hvichia G E, Kricka L J, Wilding P. Chip PCR.II. Investigation of different PCR amplification systems in microfabricated silicon-glass chips. Nucleic Acids Research, 1996, 24:  380-385.
14. Shoffner M A, Cheng J, Hvichia G E, Kricka L J, Wilding P. Chip PCR.I. Surface passivation of microfabricated silicon-glass chips for PCR. Nucleic Acids Research, 1996, 24: 375-379.Copyright 2001 - Tsinghua Science and Technology