Fabrication and Evaluation of a Graphene Oxide-Based ... - MDPI

sensors Article

Fabrication and Evaluation of a Graphene Oxide-Based Capacitive Humidity Sensor † Jinfeng Feng 1 , Xiaoxu Kang 2, *, Qingyun Zuo 2 , Chao Yuan 2 , Weijun Wang 2 , Yuhang Zhao 2 , Limin Zhu 3 , Hanwei Lu 3 and Juying Chen 3 1 2

3

* †

School of Mechanical Engineering, Shanghai Jiao Tong University, Shanghai 200240, China; [email protected] Process Technology Department, Shanghai IC R&D Center, Shanghai 201210, China; [email protected] (Q.Z.); [email protected] (C.Y.); [email protected] (W.W.); [email protected] (Y.Z.) PIE Department III, Shanghai Huahong Grace Semiconductor Manufacturing Corporation, Shanghai 201206, China; [email protected] (L.Z.); [email protected] (H.L.); [email protected] (J.C.) Correspondence: [email protected]; Tel.: +86-10-6086-0987; Fax: +86-10-6086-0966 This paper is an extension of conference paper “Humidity Sensor with Graphene Oxide as Sensing Material”.

Academic Editor: Vittorio M.N. Passaro Received: 16 December 2015; Accepted: 13 February 2016; Published: 1 March 2016

Abstract: In this study, a CMOS compatible capacitive humidity sensor structure was designed and fabricated on a 200 mm CMOS BEOL Line. A top Al interconnect layer was used as an electrode with a comb/serpent structure, and graphene oxide (GO) was used as sensing material. XRD analysis was done which shows that GO sensing material has a strong and sharp (002) peak at about 10.278˝ , whereas graphite has (002) peak at about 26˝ . Device level CV and IV curves were measured in mini-environments at different relative humidity (RH) level, and saturated salt solutions were used to build these mini-environments. To evaluate the potential value of GO material in humidity sensor applications, a prototype humidity sensor was designed and fabricated by integrating the sensor with a dedicated readout ASIC and display/calibration module. Measurements in different mini-environments show that the GO-based humidity sensor has higher sensitivity, faster recovery time and good linearity performance. Compared with a standard humidity sensor, the measured RH data of our prototype humidity sensor can match well that of the standard product. Keywords: relative humidity; humidity sensor; graphene oxide; capacitive sensing

1. Introduction With the developing of the Internet of Things and sensor networks, MEMS/sensor “consumerization” has begun to emerge, and more and more new applications of MEMS/sensors have begun to be found in consumer products such as mobile phones, PAD, etc. With the consumer market growth, requirements for high performance and low cost MEMS/sensors are increasing, especially for digital multimedia products [1,2]. Humidity sensors is some of the most widely used sensors, found in industry control, automobile defogger, agriculture, and environment monitoring applications, etc. Recently, humidity sensors have been successfully integrated into high-end mobile phone products including the Samsung Galaxy S4, and their market will increase dramatically with the growing consumer electronics demands. Based on their measurement units, humidity sensors can be divided into two types: relative and absolute humidity sensors. According to the operating principles, these are three main types:

Sensors 2016, 16, 314; doi:10.3390/s16030314

www.mdpi.com/journal/sensors

Sensors 2016, 16, 314

2 of 9

Sensors 2016, 16, 314

2 of 9

Capacitive, resistive, and thermal conductivity humidity sensors. Most humidity sensors are capacitive Capacitive, resistive, and thermal conductivity humidity sensors. Most humidity sensors are RH sensors because of their low cost, linearity, minimal long-term drift and hysteresis [3]. capacitive RH sensors because of their low cost, linearity, minimal long-term drift and hysteresis [3]. Graphene is one of the most promising materials developed in this century with its exceptional Graphene is one of the most promising materials developed in this century with its exceptional mechanical, thermal and electrical properties [4–7]. GO is made up of single or several closely-spaced mechanical, thermal and electrical properties [4–7]. GO is made up of single or several closely-spaced graphene sheets, and has sp2 and sp3 hybridized carbon atoms which can be considered as graphene sheets, and has sp2 and sp3 hybridized carbon atoms which can be considered as insulators [8]. insulators [8]. GO is slightly soluble in water, and has plenty of oxygen-containing functional groups, GO is slightly soluble in water, and has plenty of oxygen-containing functional groups, such as epoxy, such as epoxy, hydroxy (–OH), and carboxy (–COOH) groups, covering its surface [9–12]. GO materials hydroxy (–OH), and carboxy (–COOH) groups, covering its surface [9–12]. GO materials have already have already attracted significant interest in various sensing applications because of their ultrasensitive attracted significant interest in various sensing applications because of their ultrasensitive detection detection properties [13–17], but little work has been done on evaluation of its critical properties properties [13–17], but little work has been done on evaluation of its critical properties for sensor for sensor applications, such as leakage current control, CMOS compatible performance, stability, applications, such as leakage current control, CMOS compatible performance, stability, linearity, etc. linearity, etc. In this work, a capacitive humidity sensor was designed and fabricated with GO as sensing In this work, a capacitive humidity sensor was designed and fabricated with GO as sensing material. The capacitance of the sensor device was increased about four times from ~22.5% RH to material. The capacitance of the sensor device was increased about four times from ~22.5% RH to ~85% RH, which shows excellent sensitivity. After integrating the sensor device with a dedicated ~85% RH, which shows excellent sensitivity. After integrating the sensor device with a dedicated readout ASIC, a prototype humidity sensor was built and its measured data was compared with that readout ASIC, a prototype humidity sensor was built and its measured data was compared with that of a standard humidity sensor. of a standard humidity sensor.

2. Experimental 2. ExperimentalDetails Details Figure 11shows showsthe thewhole whole workflow of this study. Firstly, The CMOS compatible capacitive Figure workflow of this study. Firstly, The CMOS compatible capacitive sensor sensor structure was designed and fabricated on a 200 mm CMOS BEOL line. After that, the wafer structure was designed and fabricated on a 200 mm CMOS BEOL line. After that, the wafer dicing dicing was done, and the sensor was fabricated by filling GO material in this structure. Finally the was done, and the sensor was fabricated by filling GO material in this structure. Finally the prototype prototype humidity sensor was designed and built to evaluate its performance. humidity sensor was designed and built to evaluate its performance.

Figure 1. Workflow Workflow of this study. study. Figure

The sensor sensor structure structure was was designed designed and and fabricated fabricated with with aa comb/serpent comb/serpent capacitor The capacitor structure structure by by aa standard 0.35 μm CMOS compatible process, as shown in Figure 2. The top Al interconnect layer standard 0.35 µm CMOS compatible process, as shown in Figure 2. The top Al interconnect layer was was used as the capacitor electrode with a line/space/height = 0.78 μm/0.54 μm/0.8 μm. After the sensor used as the capacitor electrode with a line/space/height = 0.78 µm/0.54 µm/0.8 µm. After the sensor structure fabrication, fabrication, aa full full wafer wafer CV/IV CV/IV mapping structure mapping test test was was done done to to check check the the process process uniformity uniformity and and leakage performance. performance. leakage Isolation/Sensing material

Al with comb/serpent structure

Dielectric Isolation

Sub Figure 2. Schematic structure. Schematic cross-sectional-view of designed humidity sensor structure.

After wafer dicing was finished, and the sensor was fabricated by filling GO material in the structure. In this post CMOS process, a GO-based water dispersion solution was used as filling material, and after the filling process the sensor was thermally treated to remove the water. XRD was

Sensors 2016, 16, 314

3 of 9

After wafer dicing was finished, and the sensor was fabricated by filling GO material in the structure. In 314 this post CMOS process, a GO-based water dispersion solution was used as filling Sensors 2016, 16, 3 of 9 Sensors 2016, 16, 314 of 9 material, and after the filling process the sensor was thermally treated to remove the water. XRD 3was used the the sensing material properties, and CV/IV was measured different environments used totocheck check sensing material properties, anddata CV/IV data wasinmeasured in different used to check the sensing material properties, and CV/IV data was measured in different to check the leakage and capacitance properties. environments to check the leakage and capacitance properties. environments to check the leakage and capacitance properties. A prototype humidity sensor was designed and built by integrating the sensor with a dedicated A prototype humidity sensor was designed and built by integrating the sensor with a dedicated readout ASIC ASIC and and display/calibration display/calibration module at the the PCB PCB level. level. A standard frequency-voltage frequency-voltage readout ASIC and display/calibration module at the PCB level. A standard frequency-voltage convertor (CAV444, (CAV444, Analog Analog Microelectronics, Microelectronics,Mainz, Mainz,Germany) Germany)was was used used as as readout readout circuit circuit to extract convertor (CAV444, Analog Microelectronics, Mainz, Germany) was used as readout circuit to extract the capacitance variation by the RC oscillation mechanism. The prototype sensor was then evaluated the capacitance variation by the RC oscillation mechanism. The prototype sensor was then evaluated in different different environments, environments, and and all the measured data was compared with a standard humidity sensor in different environments, and all the measured data was compared with a standard humidity sensor (HM 34 Humidity & Temperature Meter, VAISALA, Vantaa, Finland) as shown shown in in Figure Figure 3. 3. (HM 34 Humidity & Temperature Meter, VAISALA, Vantaa, Finland) as shown in Figure 3.

Figure 3. Standard temperature and humidity sensor. sensor. Figure 3. Standard temperature temperature and Figure 3. Standard and humidity humidity sensor.

All the measurements were done at a room temperature of 25 ˝°C. CV was measured by a 4284A All the measurements were done at a room temperature temperature of of 25 °C. C. CV was measured by a 4284A instrument (Agilent, Santa Clara, CA, USA) and IV was measured by an Agilent 4156C device. This an Agilent Agilent 4156C 4156C device. device. This instrument (Agilent, Santa Clara, CA, USA) and IV was measured by an study adopted a very convenient way to measure and calibrate humidity sensors by using saturated study adopted a very convenient way to measure and calibrate humidity sensors by using saturated salt solutions. As shown in Figure 4, the saturated salt solution, which is made up as a slushy mixture in Figure 4, the salt solution, whichwhich is made as a up slushy salt solutions. solutions.As Asshown shown in Figure 4, saturated the saturated salt solution, is up made as amixture slushy with distilled water and chemically pure salt, is enclosed in a sealed glass chamber to build miniwith distilled water and chemically pure salt, is salt, enclosed in a sealed glass chamber to build minimixture with distilled water and chemically pure is enclosed in a sealed glass chamber to build environments with different RH values. environments with different RH values. mini-environments with different RH values.

Figure 4. Mini-environment built with a saturated salt solution sealed in glass chamber. Figure salt solution solution sealed sealed in in glass glass chamber. chamber. Figure 4. 4. Mini-environment Mini-environment built built with with aa saturated saturated salt

The RH of the atmosphere above the saturated salt solution in the sealed chamber was fixed at The RH of the atmosphere above the saturated salt solution in the sealed chamber was fixed at a given and it can cover almost the entire salt range of relative humidity, shown in Table 1. Thetemperature, RH of the atmosphere above the saturated solution in the sealed as chamber was fixed a given temperature, and it can cover almost the entire range of relative humidity, as shown in Table 1. Because thetemperature, concentration of it a saturated solution is fixed at given temperature, it is as very easy in to at a given and can coversalt almost the entire range of relative humidity, shown Because the concentration of a saturated salt solution is fixed at given temperature, it is very easy to determine the saturation state when is solidsalt phase salt left in theat solution. As the RH value will Table 1. Because the concentration ofthere a saturated solution is fixed given temperature, it is very determine the saturation state when there is solid phase salt left in the solution. As the RH value will be disturbed during putting device under testis solid (DUT) intosalt theleftmini-environments, all RH the easy to determine the saturation state when there phase in the solution. As the be disturbed during putting device under test (DUT) into the mini-environments, all the measurements will be done after the RHdevice value under of the mini-environment stable which can beall easily value will be disturbed during putting test (DUT) into theismini-environments, the measurements will be done after the RH value of the mini-environment is stable which can be easily determined by will using output of avalue standard humidity sensor. measurements bethe done afterdata the RH of the mini-environment is stable which can be easily determined by using the output data of a standard humidity sensor. determined by using the output data of a standard humidity sensor. Table 1. Relative humidity (RH) values of normally used saturated salt solutions. Table 1. Relative humidity (RH) values of normally used saturated salt solutions. Saturated Salt Solution Temperature (°C) Relative Humidity (%) Saturated Salt Solution Temperature (°C) Relative Humidity (%) LiCl 25 15 LiCl 25 15 CaCl2 25 31 CaCl2 25 31 K2CO3 25 43 K2CO3 25 43 NaCl 25 75 NaCl 25 75 KCl 25 84 KCl 25 84 K2SO4 25 98 K2SO4 25 98

Sensors 2016, 16, 314

4 of 9

Table 1. Relative humidity (RH) values of normally used saturated salt solutions. Saturated Salt Solution

Temperature (˝ C)

Relative Humidity (%)

LiCl CaCl2 K2 CO3 NaCl KCl K2 SO4

25 25 25 25 25 25

15 31 43 75 84 98

Sensors 2016, 16, 314 Sensors 2016, 16, 314

4 of 9 4 of 9

3. Results and Discussion 3. Resultsand andDiscussion Discussion 3. Results A standard 0.35 μm CMOS BEOL process was used used to to fabricate fabricate the comb/serpent comb/serpent capacitor capacitor standard0.35 0.35 μm µm CMOS CMOS BEOL BEOL process process was was AAstandard used to fabricate the the comb/serpent capacitor structure. A BEOL Al metal layer was used as capacitor electrode with line/space/height = 0.78 µm/ μm/ structure. A 0.78 structure. A BEOL BEOL Al Al metal metal layer layerwas wasused usedasascapacitor capacitorelectrode electrodewith withline/space/height line/space/height = = 0.78 μm/ 0.54 µm/0.8 μm/0.8 μm. The structure was designed with consideration of least fringing field and parasitic parasitic 0.54 µm.The Thestructure structurewas wasdesigned designedwith withconsideration considerationof of least least fringing fringing field field and and 0.54 μm/0.8 μm. parasitic capacitance, and PAD was formed by the same metal layer. A top-view photo of the structure can be capacitance, and and PAD PAD was wasformed formedby bythe thesame samemetal metallayer. layer.A A top-view photo of the structure capacitance, top-view photo of the structure cancan be seen in Figure 5, ininwhich the sensor structure was located on the left, with dimensions of about about be seen Figure which thesensor sensorstructure structurewas waslocated locatedon onthe theleft, left, with with dimensions dimensions of of seen in in Figure 5, 5, in which the about 500 µm μm × 500 μm and the bonding PAD was on the the rightwith with dimensionsofof150 150 μmˆ×150 150 μm. 500 500μm µmand andthe the bonding bonding PAD PAD was 500 μm ×ˆ500 was on on the right right withdimensions dimensions of 150µm μm × 150µm. μm.

Figure 5. Top view photo of the fabricated sensor structure. structure. Figure 5. Top view photo of the fabricated sensor structure.

After wafer wafer fabout, fabout, aa CV CV and and IV IV mapping mapping test test was was done done to to check check the the uniformity uniformity and and leakage leakage After After wafer fabout, a CV and IV mapping test was done to check the uniformity and leakage performance of the structure. As can be seen in Figure 6, the capacitance of the sensor structure has performance of performance of the the structure. structure. As As can can be be seen seenin inFigure Figure6,6,the thecapacitance capacitanceofofthe thesensor sensorstructure structurehas hasa −11 ´ 11 a mean value of ~1.25E−11 f,f, and and a very very good good uniformity uniformity can can be be obtained obtained with with aa standard standard deviation deviation of of amean meanvalue valueofof~1.25E ~1.25E f, and aa very good uniformity can be obtained with a standard deviation of −13 ´ 13 ~1.76E f and NU% of ~ 1.41%. From the IV data, the structure shows a very good leakage control ~1.76E andNU% NU%ofof~~1.41%. 1.41%.From Fromthe theIV IVdata, data,the thestructure structureshows shows aa very very good good leakage leakage control control ~1.76E−13 f fand −14 ´ 14A. with aa max max leakage leakage current current of of ~5E ~5E−14 with A. with a max leakage current of ~5E A.

1.20E-11 1.20E-11 1.23E-11 1.23E-11 1.24E-11 1.24E-11 1.24E-11 1.24E-11 1.23E-11 1.23E-11 1.22E-11 1.22E-11 1.21E-11 1.21E-11 1.20E-11 1.20E-11

1.25E-11 1.25E-11 1.26E-11 1.26E-11 1.27E-11 1.27E-11 1.21E-11 1.21E-11 1.22E-11 1.22E-11 1.25E-11 1.25E-11 1.26E-11 1.26E-11 1.26E-11 1.26E-11 1.25E-11 1.25E-11 1.25E-11 1.25E-11 1.25E-11 1.25E-11 1.23E-11 1.23E-11 1.19E-11 1.19E-11

1.26E-11 1.26E-11 1.26E-11 1.26E-11 1.26E-11 1.26E-11 1.24E-11 1.24E-11 1.26E-11 1.26E-11 1.27E-11 1.27E-11 1.27E-11 1.27E-11 1.27E-11 1.27E-11 1.27E-11 1.27E-11 1.26E-11 1.26E-11 1.26E-11 1.26E-11 1.25E-11 1.25E-11 1.26E-11 1.26E-11 1.25E-11 1.25E-11 1.23E-11 1.23E-11

1.25E-11 1.25E-11 1.22E-11 1.22E-11 1.23E-11 1.23E-11 1.23E-11 1.23E-11 1.25E-11 1.25E-11 1.25E-11 1.25E-11 1.27E-11 1.27E-11 1.28E-11 1.28E-11 1.28E-11 1.28E-11 1.28E-11 1.28E-11 1.28E-11 1.28E-11 1.27E-11 1.27E-11 1.26E-11 1.26E-11 1.25E-11 1.25E-11 1.26E-11 1.26E-11 1.25E-11 1.25E-11 1.22E-11 1.22E-11




1.23E-11 1.23E-11 1.24E-11 1.24E-11 1.25E-11 1.25E-11 1.25E-11 1.25E-11 1.26E-11 1.26E-11 1.25E-11 1.25E-11 1.25E-11 1.25E-11 1.25E-11 1.25E-11 1.24E-11 1.24E-11

2.60E-14 2.60E-14 3.20E-14 3.20E-14 3.00E-14 3.00E-14 2.30E-14 2.30E-14 2.20E-14 2.20E-14 2.50E-14 2.50E-14 3.30E-14 3.30E-14 3.10E-14 3.10E-14







2.30E-14 2.30E-14 2.80E-14 2.80E-14 2.10E-14 2.10E-14 2.00E-14 2.00E-14 2.80E-14 2.80E-14 3.30E-14 3.30E-14 3.40E-14 3.40E-14 4.10E-14 4.10E-14 3.30E-14 3.30E-14

Figure 6. Full wafer CV (Left) and IV (Right) mapping test data of the sensor structure. Figure 6. Full wafer CV (Left) and IV (Right) mapping test data of the sensor structure. Figure 6. Full wafer CV (Left) and IV (Right) mapping test data of the sensor structure.

After wafer dicing, the sensor was fabricated by filling GO material in the sensor structure. A After wafer dicing, the sensor was fabricated by filling GO material in the sensor structure. A GO-based water dispersion solution was used as filling material, and after the filling process the GO-based water dispersion solution was used as filling material, and after the filling process the sensor was treated in a baking oven at a temperature of 60 °C for about 50 min to remove the water. sensor was treated in a baking oven at a temperature of 60 °C for about 50 min to remove the water. Cross-sectional SEM was done to check the filling performance, and the GO filling performance can Cross-sectional SEM was done to check the filling performance, and the GO filling performance can meet the sensor requirements as can be seen in Figure 7. meet the sensor requirements as can be seen in Figure 7.

Sensors 2016, 16, 314

5 of 9

After wafer dicing, the sensor was fabricated by filling GO material in the sensor structure. A GO-based water dispersion solution was used as filling material, and after the filling process the sensor was treated in a baking oven at a temperature of 60 ˝ C for about 50 min to remove the water. Cross-sectional SEM was done to check the filling performance, and the GO filling performance can Sensors 2016, 16, 314 5 of 9 meet the sensor requirements as can be seen in Figure 7.

Figure 7. X-SEM photo of the sensor device after GO filling.

GO film was prepared on a blanket substrate to do XRD analysis. Compared with graphite, GO has a variety of reactive oxygen groups, such as carboxylic acids (COOH), carbonyls (C=O), epoxides 3 structure network. Because of the existence 3 structure and hydroxyls hydroxyls(–OH), (–OH),which whichare arelinked linkedtotoa aC-sp C-sp (–O–) and network. Because of the existence of of oxygen functional groups and some other structural defects, the interlayerdistance distanceof of GO GO is greatly oxygen functional groups and some other structural defects, the interlayer [12–15]. As Ascan canbebeseen seen Figure 8, the XRD pattern of shows GO shows a strong and (0 sharp increased [12–15]. in in Figure 8, the XRD pattern of GO a strong and sharp 0 2) (0 0 2)atpeak at 10.278 about ˝10.278°, corresponds to an interlayer distance of about 8.606 . The insert peak about , which which corresponds to an interlayer distance of about 8.606 Å. TheÅinsert is the ˝ corresponding is thepattern XRD pattern of graphite, whichthe shows thepeak (0 0at2)about peak 26.56 at about 26.56° corresponding to an XRD of graphite, which shows (0 0 2) to an interlayer interlayerofdistance distance 3.34 Å. of 3.34 Å .

Figure 8. XRD photo of the GO sensing material.

Leakage is very critical Figure to any 8.sensor application with a capacitive XRD photo of the GO sensing material.sensing mechanism, because more leakage current will make capacitance measurement inaccurate and cause the device to malfunction. Leakage of the sensor device was measured in voltage sweeping mode from 0 V to 5 V. As shown in Figure 9, the sensor leakage can be controlled below 1E−11 A with a comb/serpent electrode length of ~165 mm at about 50% RH level, which can meet well the sensor application requirements.

Sensors 2016, 16, 314

6 of 9

Leakage is very critical to any sensor application with a capacitive sensing mechanism, because more leakage current will make capacitance measurement inaccurate and cause the device to malfunction. Leakage of the sensor device was measured in voltage sweeping mode from 0 V to 5 V. As Sensors 2016, 16, 314 6 of 9 shown in Figure 9, the sensor leakage can be controlled below 1E´11 A with a comb/serpent electrode Sensors 16, 314 6 of 9 length2016, of ~165 mm at about 50% RH level, which can meet well the sensor application requirements. 1E-11

Current (A) (A) Current

1E-11

1E-12 1E-12

1E-13 0

1

2

0

1

2

1E-13

3

Voltage3(V)

4

5

4

5

(V) after GO filling. Voltage Figure 9. IV curve of the sensor device Figure 9. IV curve of the sensor device after GO filling.

Capacitance was measured at frequency of 100 kHz by an Agilent 4284A instrument. C/C0 ratio is plotted in Figure 10, in which C is the capacitance at given RH level and C0 is the capacitance at Capacitance was measured measuredatatfrequency frequencyofof100 100kHz kHzbyby Agilent 4284A instrument. C/C0 ratio Capacitance was anan Agilent 4284A instrument. C/C0 is minimal RH level of ~22.5% RH. With the RH value changing from ~22.5% to ~85%, the ratio sensor is plotted in Figure 10, in which C is the capacitance at given RH level and C0 is the capacitance at plotted in Figure 10, in which C is the capacitance at given RH level and C0 is the capacitance at minimal capacitance has increased around four times, whereas normally used humidity materials, such as minimal level of ~22.5% With the changing RH valuefrom changing from ~22.5% tosensor ~85%, the sensor RH level RH ofonly ~22.5% With RH. the RH value to to ~85%, theof polyimide, haveRH. a capacitance variation of about 20%. In~22.5% addition plenty oxygencapacitance functional capacitance has increased around four times, whereas normally used humidity materials, such as has increased around four times,its whereas normallytoused humidity such as polyimide, only groups on the GO material, high surface volume ratiomaterials, also contributes to this larger polyimide, only have a capacitance variation of about 20%. In addition to plenty of oxygen functional have a capacitance capacitance change.variation of about 20%. In addition to plenty of oxygen functional groups on the groups on GO surface material, highratio surface to volume to ratio this larger GO material,the its high to its volume also contributes this also largercontributes capacitanceto change. capacitance change.

Figure 10. Sensor capacitance variation ratio of C/CO with different RH. Figure 10. Sensor capacitance variation ratio of C/CO with different RH.

Figure 10. Sensor capacitance variation ratio of C/CO with different RH. To To evaluate evaluate the the potential potential value value of of GO GO material material in in humidity humidity sensor sensor applications, applications, aa prototype prototype humidity sensor was designed and built by integrating the sensor with a dedicated readout humidity sensor was designed and built by integrating the sensor with a dedicated readout ASIC ASIC To evaluate the potential value of GO material in humidity sensor applications, a prototype and module on on the the PCB PCB level. level. A A standard standard Analog Analog Microelectronics Microelectronics CAV444 CAV444 and aa display/calibration display/calibration module humidity sensor was designed and built by integrating the sensor with a dedicated readout ASIC frequency-voltage readout circuit the capacitance capacitance variation variation caused caused frequency-voltage convertor convertor was was used used as as aa readout circuit to to extract extract the and a display/calibration module on the PCB level. A standard Analog Microelectronics CAV444 by the RC oscillation mechanism. schematic circuitcircuit block to of extract the prototype sensor and block diagram frequency-voltage convertor was A used as a readout the capacitance variation caused of the frequency-voltage convertor are shown in Figure 11. by the RC oscillation mechanism. A schematic circuit block of the prototype sensor and block diagram of the frequency-voltage convertor are shown in Figure 11.

Sensors 2016, 16, 314

7 of 9

by the RC oscillation mechanism. A schematic circuit block of the prototype sensor and block diagram of the 2016, frequency-voltage convertor are shown in Figure 11. Sensors 16, 314 7 of 9 Sensors Sensors 2016, 2016, 16, 16, 314 314

77 of of 99

Figure 11. Schematic circuit block of the prototype sensor (Left) and block diagram of the CAV444 (Right). Figure 11. Schematic circuit block of the prototype sensor (Left) and block diagram of the Figure Figure 11. 11. Schematic Schematic circuit circuit block block of of the the prototype prototype sensor sensor (Left) (Left) and and block block diagram diagram of of the the CAV444 CAV444 (Right). (Right). CAV444 (Right).

To build the prototype sensor, the sensor device was firstly wire-bonded with the CAV444 on To build prototype sensor, sensorofdevice was firstly wire-bonded theinCAV444 on the PCB level,the which can convert thethe change capacitance to voltage. As can with be seen Figure 12, To build the prototype sensor, the sensor device was firstly wire-bonded with the CAV444 on the the output PCB level, which convert change of capacitance to value, voltage. Asthe can be seen in Figure 12, voltage hascan a very goodthe linear relation with the RH and insert graph shows the PCB level, which can convert the change of capacitance to voltage. As can be seen in Figure 12, the the has a verythe good linear relation the RH value, and the insert graph shows the PCBoutput photo voltage after integrating sensor device withwith the ASIC. output voltage has a very good linear relation with the RH value, and the insert graph shows the PCB PCB photo after integrating the sensor device with the ASIC. photo after integrating the sensor device with the ASIC.

Figure 12. Output voltage vs RH value of the sensor structure. Figure 12. Output voltage vs RH value of the sensor structure. Figure Figure 12. 12. Output Output voltage voltage vs vs RH RH value value of of the the sensor sensor structure. structure.

After that, the prototype humidity sensor was built by further integrating a LCD display and that, the prototype prototype humidity further integrating integrating a LCD LCD display display and and After that, the calibration module, which can humidity be seen insensor Figurewas 13. built by further Figure 13. 13. calibration module, module, which which can can be be seen seen in in Figure calibration

Figure 13. Prototype sensor with LCD display module (Left), PCB integration of sensor device and ASIC and standard humidity sensor Figure 13. Prototype sensor with LCD display module (Left), PCB integration of sensor device and Figure(Middle) 13. Prototype Prototype sensor with with LCD display display(Right). module (Left), (Left), PCB PCB integration integration of of sensor sensor device device and and Figure 13. sensor LCD module ASIC (Middle) and standard humidity sensor (Right). ASIC (Middle) and standard humidity sensor (Right). ASIC (Middle) and standard humidity sensor (Right).

After linear calibration, the prototype sensor was used to measure the RH value in environments After linear calibration, the prototype sensor used to with measure value in environments at different RH levels. The measured data was thenwas compared that the of aRH standard humidity sensor, at levels. measured thenhumidity compareddata withofthat a standardsensor humidity as different shown inRH Figure 14.The It can be seendata thatwas output ourofprototype can sensor, match as shown in Figure 14.standard It can besensor. seen that output humidity data of our prototype sensor can match well the output of the well the output of the standard sensor.

Sensors 2016, 16, 314

8 of 9

After linear calibration, the prototype sensor was used to measure the RH value in environments at different RH levels. The measured data was then compared with that of a standard humidity sensor, as shown in Figure 14. It can be seen that output humidity data of our prototype sensor can match well output Sensorsthe 2016, 16, 314of the standard sensor. 8 of 9

RelativeHumidity Humidity(%) (%) Relative

Sensors 2016, 16, 314 80 80 75 75 70 70 65 65 60 60 55 55 50 50 45 45 40 40 35 35 30 300 0

8 of 9

DUT DUT Reference Reference

1 1

2 2

3 3

Condition Condition

4 4

5 5

Figure Figure 14. 14. the prototype prototype sensor sensor (Left) (Left) and and measurement measurement photo of of the the 14. Measured Measured humidity humidity data data of of the measurement photo the prototype sensor and standard sensor (Right). prototype prototype sensor sensor and and standard standard sensor sensor (Right). (Right).

Response/recovery was defined as to of aahumidity Response/recoverytime time asthe thetime time(τ) (τ)required required toachieve achieve~63% ~63%value value Response/recovery time was was defined defined as the time (τ) required to achieve ~63% value of of a humidity humidity step function. Recovery time was measured by a simple method as shown in Figure 15. The standard step function. Recovery time was measured by a simple method as shown in Figure 15. The standard step function. Recovery time was measured by a simple method as shown in Figure 15. The standard and prototype sensor was put into aa mini-environment of the same saturated salt solution. After and prototype sensor was put into mini-environment of the same saturated salt solution. After and prototype sensor was put into a mini-environment of the same saturated salt solution. After stabilization, the sensor was quickly taken out from the high RH mini-environment to the outside stabilization, the the sensor sensor was was quickly quickly taken taken out out from from the the high high RH RH mini-environment mini-environment to to the the outside outside stabilization, atmosphere, and the recovery time was measured. From ~93% RH (mini-environment) to ~52% RH atmosphere, and the recovery time was measured. From ~93% RH (mini-environment) to ~52% RH atmosphere, and the recovery time was measured. From ~93% RH (mini-environment) to ~52% RH (outside environment), the recovery time of this work is less than 3 s compared with ~8 s for the (outside environment), the recovery time of this work is less than 3 s compared with ~8 s for the (outside environment), the recovery time of this work is less than 3 s compared with ~8 s for the standard humidity sensor, so the GO-based humidity sensor shows about two times faster recovery standard humidity humidity sensor, sensor, so so the the GO-based GO-based humidity humidity sensor sensor shows shows about about two two times times faster faster recovery recovery standard time than the standard sensor, which is due to its special structure with much higher surface to time than the standard sensor, which is due to its special structure with much higher surface to time than the standard sensor, which is due to its special structure with much higher surface to volume ratio. volume ratio. volume ratio.

Figure 15. Schematic recovery time test procedure. procedure. Figure 15. Schematic recovery time test procedure.

4. 4. Conclusions Conclusions 4. Conclusions In In this this work, work, aaa CMOS CMOS compatible compatible capacitive capacitive humidity humidity sensor sensor structure structure was was designed designed and and In this work, CMOS compatible capacitive humidity sensor structure was designed and fabricated on a 200 mm CMOS BEOL Line. BEOL Al with a comb/serpent structure was used as fabricated on a 200 mm CMOS BEOL Line. BEOL Al with a comb/serpent structure was used as fabricated on a 200 mm CMOS BEOL Line. BEOL Al with a comb/serpent structure was used electrode, and GO was used as sensing material. The sensor device shows excellent sensitivity and electrode, and GO was used and as electrode, and GO was usedasassensing sensingmaterial. material.The Thesensor sensordevice deviceshows shows excellent excellent sensitivity sensitivity and faster time than a standard sensor. A prototype humidity sensor was designed and built by faster recovery recovery sensor was designed and built by faster recovery time time than thanaastandard standardsensor. sensor.AAprototype prototypehumidity humidity sensor was designed and built integrating the sensor device with a dedicated ASIC at the PCB level, and the output value of the integrating the sensor device with a dedicated ASIC at the PCB level, and the output value of the by integrating the sensor device with a dedicated ASIC at the PCB level, and the output value of the prototype standard humidity sensor. prototype sensor sensor was was shown shown to to match match well well that that of of aaa standard standardhumidity humiditysensor. sensor. prototype sensor was shown to match well that of Acknowledgments: work was the result close cooperation with ICR&D Center received Acknowledgments: This close cooperation with thethe ICR&D Center andand received greatgreat help Acknowledgments: This Thiswork workwas wasthe theresult resultofof of close cooperation with the ICR&D Center and received great help from HHGRACE. The authors would like to thank the colleagues from the R&D Department of from HHGRACE. The authors would like to thank the colleagues from the R&D Department of Shanghai IC R&D help from HHGRACE. The authors would like to thank the colleagues from the R&D Department of Shanghai Shanghai IC Center PIE of Special thanks due Meng Gao, Center theand PIEthe department III ofIII HHGRACE. Special thanks areare due toto Meng Jiang, ICR&D R&Dand Center and the PIEdepartment department III ofHHGRACE. HHGRACE. Special thanks are due to MengGao, Gao,Liangliang LiangliangJiang, Jiang, Yong Wang, Wang, Shaojian Hu, Ming Li, Bin Jiang and Xiang Li for their strong support of this work. Yong Shaojian Hu, Ming Li, Bin Jiang and Xiang Li for their strong support of this work. Yong Wang, Shaojian Hu, Ming Li, Bin Jiang and Xiang Li for their strong support of this work. Author Contributions: Contributions: Jinfeng JinfengFeng Fengand and Xiaoxu Kang contributed equally the to the whole work. Qingyun Zuo Author Kang equally whole work. Qingyun Zuo Author Contributions: Jinfeng andXiaoxu Xiaoxu Kangcontributed contributed equallyto tothe thetest. whole work. Qingyun Zuoand and and Weijun Wang designed theFeng humidity test environment and performed Chao Yuan performed the Weijun WeijunWang Wangdesigned designedthe thehumidity humiditytest testenvironment environmentand andperformed performedthe thetest. test.Chao ChaoYuan Yuanperformed performedthe thelayout layout design design and and mask mask tapeout. tapeout. Yuhang Yuhang Zhao Zhao performed performed the the CV/IV/XRD CV/IV/XRD measurements measurements and and analyzed analyzed the the data. data. Limin Zhu, Hanwei Lu and Juying Chen performed the process experiments of the sensor structure and PCM Limin Zhu, Hanwei Lu and Juying Chen performed the process experiments of the sensor structure and PCM data dataanalysis. analysis.

Conflicts Conflictsof ofInterest: Interest: The Theauthors authorsdeclare declareno noconflict conflictof ofinterest. interest.

Sensors 2016, 16, 314

9 of 9

layout design and mask tapeout. Yuhang Zhao performed the CV/IV/XRD measurements and analyzed the data. Limin Zhu, Hanwei Lu and Juying Chen performed the process experiments of the sensor structure and PCM data analysis. Conflicts of Interest: The authors declare no conflict of interest.

References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14.

15. 16. 17.

International Technology Roadmap for Semiconductors (ITRS). Available online: http://public.itrs.net (accessed on 30 October 2014). Vigna, B. More than Moore: Micro-machined products enable new applications and open new markets. In Proceedings of IEEE International Electron Devices Meeting, Washington, DC, USA, 1–8 December 2005. Chen, Z.; Lu, C. Humidity sensors: A review of materials and mechanisms. Sens. Lett. 2005, 3, 274–295. [CrossRef] Novoselov, K.S.; Falko, V.I.; Colombo, L.; Gellert, P.R.; Schwab, M.G.; Kim, K. A roadmap for graphene. Nature 2012, 490, 192–200. [CrossRef] [PubMed] Lin, Y.M.; Dimitrakopoulos, C.; Jenkins, K.A.; Farmer, D.B.; Chiu, H.-Y.; Grill, A.; Avouris, P. 100-GHz Transistors from Wafer-Scale Epitaxial Graphene. Science 2010, 327. [CrossRef] [PubMed] Shao, Y.; Wang, J.; Wu, H.; Liu, J.; Aksay, I.A.; Lin, Y. Graphene Based Electrochemical Sensors and Biosensors: A Review. Electroanalysis 2010, 22, 1027–1036. [CrossRef] Mueller, T.; Xia, F.; Avouris, P. Graphene photodetectors for high-speed optical communications. Nat. Photonics 2010, 4, 297–301. [CrossRef] Young, R.J.; Kinloch, I.A.; Gong, L.; Novoselov, K.S. The mechanics of graphene nanocomposites: A review. Compos. Sci. Technol. 2012, 72, 1459–1476. [CrossRef] Zhu, Y.; Murali, S.; Cai, W.; Li, X.; Suk, J.W.; Potts, J.R.; Ruoff, R.S. Graphene and graphene oxide: Synthesis, properties, and applications. Adv. Mater. 2010, 22, 3906–3924. [CrossRef] [PubMed] Thekkekara, L.V.; Jia, B.; Zhang, Y.; Qiu, L.; Li, D.; Gu, M. On-chip energy storage integrated with solar cells using a laser scribed graphene oxide film. Appl. Phys. Lett. 2015, 107, 031105-1–031105-2. [CrossRef] Hasegawa, M.; Hirayama, Y.; Ohno, Y.; Maehashi, K.; Matsumoto, K. Characterization of reduced graphene oxide field-effect transistor and its application to biosensor. Jpn. J. Appl. Phys. 2014, 53, 89–100. [CrossRef] Tran, M.-H.; Jeong, H.K. Synthesis and characterization of silver nanoparticles doped reduced graphene oxide. Chem. Phys. Lett. 2015, 630, 80–85. [CrossRef] Bi, H.; Yin, K.; Xie, X.; Ji, J.; Wan, S.; Sun, L.; Terrones, M.; Dresselhaus, M.S. Ultrahigh Humidity Sensitivity of Graphene Oxide. Sci. Rep. 2013, 3, 1–7. [CrossRef] [PubMed] Guo, L.; Jiang, H.-B.; Shao, R.-Q.; Zhang, Y.-L.; Xie, S.-Y.; Wang, J.-N.; Li, X.-B.; Jiang, F.; Chen, Q.-D.; Zhang, T.; et al. Two-beam-laser interference mediated reduction, patterning and nanostructuring of graphene oxide for the production of a flexible humidity sensing device. Carbon 2012, 50, 1667–1673. [CrossRef] Lipatov, A.; Varezhnikov, A.; Wilson, P.; Sysoev, V.; Kolmakov, A.; Sinitskii, A. Highly selective gas sensor arrays based on thermally reduced graphene oxide. Nanoscale 2013, 5, 5426–5434. [CrossRef] [PubMed] Yao, Y.; Chen, X.; Guo, H.; Wu, Z.; Li, X. Humidity sensing behaviors of graphene oxide-silicon bi-layer flexible structure. Sens. Actuators B Chem. 2012, 161, 1053–1058. [CrossRef] Varghese, S.S.; Lonkar, S.; Singh, K.K.; Swaminathan, S.; Abdala, A. Recent advances in graphene based gas sensors. Sens. Actuators B Chem. 2015, 218, 160–183. [CrossRef] © 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons by Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).