Primary Parameters of VNJ-P3HT Diode - Ammonia Sensing Responses of VNJ-P3HT Diode

CHAPTER 2 Performances and Characteristics of Gas Sensors for Ammonia

2.4 Ammonia Sensing Responses of VNJ-P3HT Diode

2.4.1 Primary Parameters of VNJ-P3HT Diode

Ammonia molecules react with these VNJ-P3HT diodes by diﬀusing into the P3HT ﬁlm through the high-density pores, dedoping the P3HT ﬁlm, and reducing the diode’s current.

The more ammonia is absorbed by sensor, the less current of output is produced. The value of current doesn’t change anymore until the current saturates. The current in steady state is called saturation current (I_ss). Finally, the different specific gas concentration can be presented

0 1 2 3 4 5 6

by the ratio of saturation current to initial current (

Iss

I ). This study defines a response as the ratio of the VNJ-P3HT diode’s output current to the initial current (

I ).

2.4.2 Preconditions and Experiment Results

The responses in 10ppb ~ 3ppm of ammonia are shown in Fig. 7. The output current of the VNJ-P3HT diode doesn’t change anymore because the responses are saturated after 200 seconds. The time of the initial current’s unexpected change from time to time is much longer than 200 seconds (as Fig. 6) and therefore it can be ignored. Thus the concentration of ammonia can be determined by the responses in steady state (saturation).

0 2 0 4 0 6 0 8 0 1 0 0 1 2 0 1 4 0 1 6 0 1 8 0 2 0 0 2 2 0

Fig. 7. The original data of response (

I ) in 10ppb ~ 3ppm of ammonia [19].

Fig. 8(a) shows the current density as a function of applied bias voltage (J−V curves) of the VNJ-P3HT diode. The current density is proportional to the square of the applied bias voltage, indicating that the holes in P3HT follow the space-charge-limited conduction (SCLC).

That is to say, the carrier transport in most intrinsic or low-doped conjugated polymers follows the space-charge-limited current. Thus the injected charge carrier density is much higher than the background doping density in most volume of the sample (i.e., in the bulk region). Because the injected charges are unipolar carriers with very low mobility, they are considered as space charges. Finally, the injected charges are closer to the injection interface have a higher charge density.

The J − V curve of the VNJ-P3HT diode is shown by the red dashed line in Fig. 8(a) after injecting the 3 ppm ammonia (with a background of nitrogen) for 200 s. After contact with ammonia, the slightly right shift of the onset voltage indicates a slight increase of hole injection barrier. However, the main reaction is the current decline in the SCLC zone. The

(b) (a)

Fig. 8. (a) J−V curve of VNJ-P3HT diodes before (black solid curve) and after 200 s of 3 ppm ammonia sensing (red dashed curve). Green symbols represent the response (i.e., the current variation ratio) of VNJ-P3HT diodes. (b) The responses (measured at 2 V with a ﬁxed sensing time as 200 s) to carbon dioxide (5%), nitric oxide (3 ppm), ethanol (100 and 1 ppm), acetone (1 ppm), and ammonia (3 ppm and 500 ppb). Blue bars, red bars, and green bars represent responses with backgrounds as pure nitrogen, dry air, and dry air passing through commercial ammonia ﬁlter (NiCl2·6H2O powders) [19].

ratio of saturation current to initial current (

Iss

I ) is used to represent the response of the sensor.

The green symbols in Fig. 8(a) shows the response to the 3ppm ammonia 200 seconds exposure plotted as a function of applied bias. The peak of response is at around 0.8 V. But the current of diode at 0.8 V is too low to provide a good signal-to-noise ratio. A large and stable response as −0.6 is obtained when the applied bias voltage changes from 1.5 to 3 V.

Thus this study chooses 2 V as a ﬁxed applied bias voltage to measure the response in the following works.

The relations are all the same in different concentration of ammonia according to [19].

The responses in steady state are same, which can determine the concentration of ammonia when the VNJ-P3HT diode is biased in 1.5~3V. Thus this study chooses the VNJ-P3HT diode as the sensor device at present. If this study tries to operate the diode sensor accurately, it needs to satisfy:

1. 1.5~3V should be applied to the bias of the sensor.

2. When initial current varies from 10~100μA, this study can perform response.

3. In order to present the specific gas concentration, use response in steady state.

4. 10ppb of ammonia can be detected by the sensor.

The responses of the VNJ-P3HT diode are also analyzed to several kinds of gases existing in human respiratory gas. The responses of the VNJ-P3HT diode (measured at 2 V with a ﬁxed sensing time as 200 seconds) to carbon dioxide (5%), nitric oxide (3 ppm), ethanol (100 and 1 ppm), acetone (1 ppm) and ammonia (3 ppm and 500 ppb) are compared in Fig. 8(b). The responses with backgrounds as pure nitrogen, dry air, and dry air passing through commercial ammonia ﬁlter (NiCl2•6H2O powders) are represented by blue bars, red bars and green bars. Because of the greatly suppressed vapor pressure of ethanol (2 mmHg) and acetone (20 mmHg) at −20 °C, the response to ethanol and acetone in dry air are signiﬁcantly suppressed. Thus the VNJ-P3HT diode has very small responses (0.009 to

−0.017) to ethanol and acetone with dry air as the background. That is to say, cooling the tube at −20 °C before connecting to the sensor device (i.e., the sensor is at room temperature) leads to a suppressed humidity and suppressed concentrations of VOCs. The concentration of ammonia should not be signiﬁcantly changed in our system because the boiling point of ammonia is −33.3 °C. On the other hand, the response to 3 ppm nitric oxide is positive and signiﬁcant. Because of the oxidation (doping) of P3HT, the reaction is known and irreversible.

Fortunately, the concentration of nitric oxide is very low in our breath (<40 ppb).

The responses of a fresh VNJ-P3HT diode and of a 5 hours aged VNJ-P3HT diode to 400 ppb ammonia as a function of time are compared in breath ammonia testing. The aged sample was used to detect breath ammonia of rats for 12 times during 5 hours. The difference between response of the fresh sample and response of the aged sample is within 4% proving that the VNJ-P3HT diode is reliable enough for at least 5 hours of operation.

CHAPTER 3 The Novel Sensing Circuit for Organic Gas Sensors

The instruments are usually used to measure the gas concentration sensed by VNJ-P3HT diode. There are few chip design studies which focus on the back-end read-out circuit because of the property of sensor device. The design of the proposed sensing circuit can provide more benefits for real-time detection of gas concentration compared with the measurement by traditional instruments. In order to detect ammonia concentration based on steady state current, the responses of VNJ-P3HT diode are measured. This study designs the sensing circuit to process current variation of VNJ-P3HT diode for varied ammonia concentrations. Because of using the ratio of saturation current to initial current to determining ammonia concentration, the sensing circuit solves the problem that the initial current of VNJ-P3HT diode are different from time to time. The real-time detection and convenient for diagnosis can also be achieved by the sensing circuit for biomedical usages. A readout circuit is needed to design as interface with the VNJ-P3HT diode at first. The output voltage which is converted by the output current of front-end sensors through the principle of a trans-impedance amplifier (TIA) can be processed by this circuit. The linear or other relationships between output signals of sensor circuit and gas concentration must be achieved in the next step. It is stand for that the gas concentration can be directly detected by output signals from back-end circuit. Thus the accurate and convenient medical measurement can be further achieved.

3.1 Review of Past Sensing Circuits

3.1.1 A Readout Circuit for an OTFT Gas Sensor with a New Preamplifier

Fig. 9 shows that the previous readout circuit consists of a pre-amplifier, a sample and hold circuit and a differential current amplifier [20]. In order to convert the output current of a gas sensor into voltage, a pre-amplifier is composed of trans-impedance. According to distinct designing time instants in a target gas environment, the sample and hold circuit memorizes various voltages. The differential current is attributable to the voltages of sample and hold circuit using current mirror framework. Then a differential current amplifier amplifies the differential current. The differential current is converted into output voltage through the resistance load. In order to determine gas concentration database of pre-measured which are contrast, a micro-processor control unit (MCU) acquires and stores the outputs.

There are two major disadvantages in the previous readout circuit. First, the voltage value is recorded by a capacitor C_hold of the sample and hold circuit. The voltage drop problem is serious because of the leakage current of the capacitor. Second, the value of initial

Fig. 9

.

Readout circuit for OTFT sensor [20].

current changes unexpectedly from time to time in the previous chapter. But the design of previous sensing circuit is based on stable OTFT sensors, which means that the initial current of OTFT sensors doesn’t change from time to time. The output of the differential current amplifier changes and the gas concentration database is unreliable when the value of initial current changes.

3.1.2 A Front-end Readout Circuit Including an Analog Divider for an OTFT Gas Sensor

The previous readout system (shown in Fig. 10) consists of an analog IC, a micro-processor control unit (MCU) and Man-Machine Interface (MMI). The analog IC comprises further a pre-amplifier and an analog divider, which are used to be an interface with the OTFTs sensors and calculate the ratio of saturation current to initial current. The CPU of proposed MCU is LPC1768 by NXP. The AD/DA of MCU holds the input value of the analog divider and the

VDD

Fig. 10. The sensing system for OTFT sensor [21].

ADC of MCU is used to calculate and store the output data of readout circuit. The MMI comprises further a Liquid Crystal Display (LCD) module and buttons. It is applied to be conveniently used by users and displays results of sensing system, which is controlled by the MCU.

The previous readout circuit also shows two major disadvantages. First, the system includes a MCU and an analog circuit, causing complexity to increase. Second, the design is not accurate, because it is limited by the resolution of the AD/DA converter.

3.2 Types of Divider Circuit

The analog signal processing circuits use the analog divider circuit widely, for example the filters, hearing-aid systems and logarithmic function generators. According to different circuit principles, many types of analog dividers are designed.

M1 M2

VDD

M3 M4

IX I_b I_Y

o u t

V1 V₂

Fig. 11. Weak inversion divider circuit [22].

3.2.1 Weak Inversion Divider

The divider and 1/x functions can be performed by a CMOS current-mode circuit in this study.

In order to meet the application of division, the design uses MOSFETs biased in weak inversion. Fig. 11 shows the proposed circuit. The drain current of a PMOS transistor in weak inversion is given by The output current in Fig. 11 is given by

1 2 Eq. (10) implements a current-mode divider circuit when the bias current I keep constant. _b The input current (I , _X I ) and output current (_Y I_OUT) are too small to be used for organic gas sensor because the MOSFETs are biased in weak inversion. The minimum output current of organic gas sensor is at least greater than 10 μA.

3.2.2 Current-to-Voltage Mode Divider

Fig. 12 shows the proposed circuit. If both M1 and M2 are biased in the triode region without body effect, the source currents I and ₁ I can be expressed as ₂ threshold voltages of M1 and M2, respectively. The current mirror, M5 and M6, is used to copy the current I , so that ₁

Fig. 12. Divider circuit for organic sensor [23].

1 3 4

I  I I . (13) The source voltages of M3 and M4 are equal (V_SB₃ V_SB₄ so V_Tn₃ V_Tn₄) and both of them are biased in saturation if they are matched (K_n₃ K_n₄). Because of the square-law characteristics of MOSFETs, the equation can be found as

1 A current-to-voltage-mode divider circuit is implemented in Eq. (16). The input current (I ) _in can be up to 100 μA because the M1 and M2 biased in the triode region. Thus the current-to-voltage-mode divider can be used for OTFT gas sensor.

3.3 The Sensing Circuit for Organic Gas Sensors

Fig. 13 illustrates the entire readout circuit. It can be seen that the circuit consists of six parts, a pre-amplifier, a peak-detect-and-hold circuit, a divider, a saturation detector, an auto-reset circuit and a result-or-zero circuit. The six parts are designed on-chip, which is accomplished by Taiwan Semiconductor Manufacturing Company (TSMC) 0.35μm 2P4M 3.3V mixed‐

signal CMOS process.

The VNJ-P3HT diode sensor component is represented by a block of sensor. The sensed current that reflects the concentration of sensed gas that interacts with surface on the VNJ-P3HT diode is denoted by I . Ammonia is chose as the gas to sense in our study. The _s operation principle of the VNJ-P3HT diode is based on the change of initial current from time to time mentioned in the previous chapter. Compared to the time that initial current change, which can be ignored, the sensing time is 200 seconds, which is very fast. The more ammonia is absorbed by the surface of the VNJ-P3HT diode, the less current of output is produced. The ratio of saturation current to initial current can present the specific gas concentration according to previous chapter. The readout circuit must records the value of initial current to calculate the ratio of saturation current to initial current. As a result, the pre-amplifier is intended in acquire the initial current information in voltage signals and pays the way for the design of a current-to-voltage converter (a trans-impedance converter) in the following. In the

VDD

Circuit Divider Saturation DetectorLogic Gate and Buffer Sensor

Pre-amplifier

Auto-Reset Circuit

Fig. 13. The readout circuit for VNJ-P3HT diode.

previous design, the MCU is used to record the voltage value. Due to adding of MCU, the incremental complexity is serious and the design is not accurate. As result, the peak-detect-and-hold circuit and super capacitor are used. The system integration between the analog IC and MCU can be accomplished, become a chip. Compared to previous current difference amplifier, the current-to-voltage-mode divider circuit can calculate the value of input current divided by input voltage. The input voltage is provided by the output of peak-detect-and-hold circuit and the input current is provided by the current mirror (M9-7). With the Auto-Reset circuit, saturation detector and logic gate, the circuit achieves an automatic easy-to-use readout circuit. The MCU records the output value of analog divider and transforms it into gas concentration. The LCD shows the value of gas concentration.

3.3.1 Pre-Amplifier

Fig. 14 shows that the pre-amplifier comprises further two current copiers (M9-8 and M9-7), M8 M7

a set of clamped op (OP1) and a negative feedback NMOS transistor [24]. The voltage across the VNJ-P3HT diode sensor V is able to be approximately fixed to V_b₀ b1 because of the negative feedback OP1 with the output connected to the gate voltage of M10. V can control _b₁ the bias voltage of VNJ-P3HT diode sensor. The advantage of this circuit structure is that can ignore the output impedance of the sensor. The only acquired information is the experiment data from sensor component. The reason for analyzing equivalent circuit of sensor is the shortage of completed information for parametric analysis and modeling sensor component.

The gate voltage of the transistor M10 is regulated to achieve balance of the negative feedback circuit. When MOSFET operate in saturation, slight variation of gate voltage can control a great quantity of drain current. The drain current has a great acceptance region with least output swing range of OP1 depending on adjusting gate voltage. And in the meanwhile biasing sensor component and ensuring OP1 work in normal region do not affect current path for sensed current only flows through M10. The output current of the VNJ-P3HT diode is inversely proportional to the concentration of the sensed gas and steadily in this way. The two current mirrors (M9-8 and M9-7) next copy the current of VNJ-P3HT diode. This study uses the current mirror (M9-8) to transform initial current (I ) into voltage by resistance ₀ R . ₀ Consequently, the initial current information (I₀ R ) is recorded by super capacitor of the ₀ peak-detect-and-hold circuit. Meanwhile, this study uses the other current mirror (M9-7) to copy the I as _s I_in, which is input current of the analog divider.

In the first and second stages of Fig. 15, the active current mirrors are used as loads for differential pairs and are also gain stages of this amplifier. The gain of the two-stage amplifier is above 70dB.

M1p M2p

M3p M4p

M5p M7p M6p

M8p

VDD

Vin

Vin

Co

Fig. 15. Two-stage operation amplifier.

3.3.2 Peak-Detect-and-Hold Circuit

The peak-detector-and-hold (PDH) circuit is shown in Fig. 16 with the timing diagram of operations in Fig. 17. One cycle of operation consists of three phases including Reset, Sample and Hold time intervals. In the reset phase, the control signal “reset” is a high voltage level

“1” to enable the capacitance “C” discharged to ground voltage. During the sample phase, the control signal reset is a low voltage level “0”. When the input voltage Vin is greater than the

Fig. 17. PDH timing diagram of operations.

output voltage Vout, the output of the amplifier is a low voltage level "0", the PMOS M1 is turned on and the capacitance C is charged until the input voltage is equal to Vout. When Vin is smaller than the output voltage, the amplifier’s output is "1" and M1 turns off. The output voltage Vout is so sustained as the maximum value of input voltage until the next reset phase begins. When sensor device starts to react with gas, the output current will decrease. The maximum value of input voltage always is the voltage transformed from initial current. Thus this study uses the circuit to acquire the initial current information in voltage signals.

3.3.3 Divider Circuit

A current-to-voltage analog divider can be realized because of Eq. (16). In order to operate the proposed current-to-voltage analog divider with a unipolar supply voltage, it can be further

Fig. 18. Current-to-voltage-mode divider circuit [23].

A current-to-voltage-mode divider can be realized and its gain can be adjusted by the reference voltages V and ₁ V because of Eq. (17). The output offset voltage is not generated ₂ by the proposed current-to-voltage analog divider under unipolar supply voltage. The drain voltages of M and ₁ M are lower than their gate voltages by a threshold voltage when ₂ V ₁

K Rn ) is presented by the output voltage of analog divider, which can present the specific gas concentration when the response is done.

3.3.4 Saturation Detector

Fig. 7 shows that after 200 seconds the responses are saturated cause the output current of the VNJ-diode OTFT doesn’t change anymore. The responses in steady state (saturation) can determine the concentration of ammonia. But timing it with a stopwatch is inconvenient.

Compared with using the stopwatch, the design of the proposed saturation detector can provide more convenience for real-time detection of gas concentration.

When the response is saturated, the output current of the VNJ-P3HT diode doesn’t change anymore. This study can make good use of this characteristic. If the output current of the VNJ-P3HT diode doesn’t change anymore, its rate of change is close to zero. After differentiation, the output of the differentiator is also close to zero. So a differentiator and a comparator are used to detect when the response is saturated. The circuit of the saturation detector is shown in Fig. 19.

A differentiator is a circuit that is designed such that the output of the circuit is approximately directly proportional to the rate of change (the time derivative) of the input. A differentiator circuit includes an operational amplifier (shown in Fig.15.), the resistor is used at feedback side and capacitors are used at the input side. The circuit is based on the

在文檔中基於影像辨識之移動式機械臂抓物控制設計 (頁 25-0)