Highly Sensitive Acetone Gas Sensor Based on g-C3N4 ... - MDPI

0 downloads 0 Views 7MB Size Report
Jul 10, 2018 - Run Zhang 1,2, Yan Wang 1,3,* ID , Zhanying Zhang 3 and ..... Zhang, R.; Zhou, T.; Wang, L.; Zhang, T. Metal−Organic Frameworks-Derived ...
sensors Article

Highly Sensitive Acetone Gas Sensor Based on g-C3N4 Decorated MgFe2O4 Porous Microspheres Composites Run Zhang 1,2 , Yan Wang 1,3, * 1 2 3

*

ID

, Zhanying Zhang 3 and Jianliang Cao 1,2, *

ID

The Collaboration Innovation Center of Coal Safety Production of Henan Province, Henan Polytechnic University, Jiaozuo 454000, China; [email protected] School of Chemistry and Chemical Engineering, Henan Polytechnic University, Jiaozuo 454000, China State Key Laboratory Cultivation Bases for Gas Geology and Gas Control, Jiaozuo 454000, China; [email protected] Correspondence: [email protected] (Y.W.); [email protected] (J.C.); Tel.: +86-391-398-7440 (Y.W. & J.C.)

Received: 3 June 2018; Accepted: 6 July 2018; Published: 10 July 2018

 

Abstract: The g-C3 N4 decorated magnesium ferrite (MgFe2 O4 ) porous microspheres composites were successfully obtained via a one-step solvothermal method. The structure and morphology of the as-prepared MgFe2 O4 /g-C3 N4 composites were characterized by the techniques of X-ray diffraction (XRD), field-emission scanning electron microscopy (FESEM), transmission electron microscopy (TEM), thermal gravity and differential scanning calorimeter (TG–DSC) and N2 -sorption. The gas sensing properties of the samples were measured and compared with a pure MgFe2 O4 -based sensor. The maximum response of the sensor based on MgFe2 O4 /g-C3 N4 composites with 10 wt % g-C3 N4 content to acetone is improved by about 145 times, while the optimum temperature was lowered by 60 ◦ C. Moreover, the sensing mechanism and the reason for improving gas sensing performance were also discussed. Keywords: g-C3 N4 nanosheet; MgFe2 O4 porous microspheres; composites; acetone; gas sensing

1. Introduction Acetone, as a highly volatile and flammable organic compound, is widely used in industries or laboratories as a solvent, chemical intermediate and industrial product [1]. Chronic exposure to an acetone atmosphere causes inflammation and may even cause damage to the liver and kidney, while acute poisoning can harm the central nervous system [2]. In addition, acetone is also a widely accepted breath biomarker for type-I and type-II diabetes [3,4]. Thus, fast and timely monitoring of the existence and concentration of acetone is of great importance for human safety and health. Gas sensors based on metal oxide semiconductors (MOS) are attractive candidates due to their low cost, fast response and easy fabrication [5,6]. Thus, MOS-based gas sensors have been regarded as an important method of monitoring flammable and toxic gases. Up to now, several types of MOSs have been developed as gas-sensing materials to detect acetone, such as In2 O3 [7], WO3 [8,9], SnO2 [10], La2 O3 [11], Co3 O4 [12], ZnO [13], α-Fe2 O3 [14] and so on. However, the gas-sensing materials have limited properties, some of which are unable to satisfy the needs of practical applications. The development of high performing and stable acetone-sensing materials remains a challenging task. In recent years, spinel ferrite MFe2 O4 (M = Mg, Co, Ni, Zn, Mn, Cd, etc.) gas-sensing materials have attracted a significant amount of research interest due to their high selectivity and good chemical stability compared to traditional metal oxide semiconductor gas sensing materials [15]. As a magnetic n-type semiconductor, magnesium ferrite (MgFe2 O4 ) has been used in the research field Sensors 2018, 18, 2211; doi:10.3390/s18072211

www.mdpi.com/journal/sensors

Sensors 2018, 18, 2211

2 of 12

of gas sensing [16–18]. For example, Patil et al. successfully prepared two types of spinel MgFe2 O4 thick films via a sol–gel process, which revealed that the best response to acetone vapor occurred at 350 ◦ C and 450 ◦ C, respectively [19]. However, experimental results have shown that MgFe2 O4 gas sensing materials have obvious shortcomings, such as poor electrical characteristics and high working temperatures. In order to improve the properties of gas sensors, compounding MOS with high specific surface area 2D nanomaterials has been proven to be an effective way of creating a synergistic effect through these components [20,21]. Li et al. synthesized Zn2 SnO4 nanoparticles/reduced graphene oxide via a solvothermal route and found that the nanocomposites showed good sensitivity to ethanol [22]. Chen et al. synthesized WO3 microspheres loaded with small-size Pt-decorated graphene composite, which exhibited a high selectivity and sensitivity to low concentration acetone gas at the operating temperature of 200 ◦ C [9]. As a new two-dimensional (2D) semiconductor, graphitic carbon nitride (g-C3 N4 ) possesses several advantages, such as high chemical stability, high specific surface area, unique electronic structure, facile preparation and non-toxicity [23,24]. Several previous studies focusing on g-C3 N4 decorated various metal oxide composites proved that g-C3 N4 plays a very important role in the composites. It not only enlarges the specific surface area and prevents agglomeration of metal oxide nanoparticles, but also forms a heterojunction with MOSs and provides new chemical and structural properties [25–27]. Cao et al. reported that SnO2 /g-C3 N4 nanocomposites show a favorable response to ethanol by a facile calcination method [28]. Hu et al. synthesized the 2D C3 N4 -tin oxide gas sensors by a one-step method for enhanced acetone vapor detection [29]. In this work, we report the synthesis of g-C3 N4 nanosheet decorated MgFe2 O4 porous microspheres composites via a one-step solvothermal method. With comparison of the gas-sensing properties of MgFe2 O4 /g-C3 N4 composites with different g-C3 N4 contents, including the sensitivity, stability and selectivity, we concluded that the properties of MgFe2 O4 -based sensors are remarkably improved due to the introduction of g-C3 N4 . In particular, the sensor based on 10 wt % g-C3 N4 decorated MgFe2 O4 porous microspheres exhibited the best gas-sensing performance. 2. Experimental 2.1. Preparation of the MgFe2 O4 /g-C3 N4 Composites All chemicals were of analytical purity and were used without further purification. Graphitic carbon nitride (g-C3 N4 ) was synthesized by our previous reported method [30,31]. MgFe2 O4 /g-C3 N4 composites were prepared by a facile solvothermal process. Typically, a certain mass of g-C3 N4 was dissolved in 80 mL of ethylene glycol with ultrasonic treatment for 2 h. After this, 1.015 g of magnesium chloride hexahydrate (MgCl2 ·6H2 O, 99.0%), 2.702 g of ferric chloride nonahydrate (FeCl3 ·9H2 O, 99.0%) and 0.54 g of urea (CO(NH2 )2 , 99.0%) were added into the previously dispersed suspension with magnetic stirring for 30 min. Urea is mainly added in order to adjust the pH value of the solution. Finally, the mixture was transferred into a 100-mL stainless-steel Teflon-lined autoclave and kept for 24 h at 200 °C in an oven. The product was collected and washed with DI water and ethanol several times, before being finally dried at 60 °C for 24 h. The ratio of MgFe2 O4 /g-C3 N4 composites was controlled by adjusting the weight of the g-C3 N4 added. According to this process, the MgFe2 O4 /g-C3 N4 composite with the 5 wt %, 10 wt % and 15 wt % contents of g-C3 N4 decorated MgFe2 O4 were prepared and marked as MgFe2 O4 /g-C3 N4 -5, MgFe2 O4 /g-C3 N4 -10 and MgFe2 O4 /g-C3 N4 -15, respectively. For comparison, pure MgFe2 O4 without g-C3 N4 nanosheets was prepared by the same method. 2.2. Sample Characterization Powder X-ray diffraction (XRD, Cu-Kα, Bruker-AXSD8) (Bruker, Madison, WI, USA) was used to examine the purity and crystalline of the samples over a 2θ range of 10–90◦ . Field emission scanning electron microscopy (FESEM, Quanta™250 FEG) (FEI, Eindhoven, The Netherlands) and

Sensors 2018, 18, 2211

3 of 12

Sensors 2018, 18, x FOR PEER REVIEW

3 of 11

transmission electron microscopy (TEM, JEOL JEM-2100) (JEOL, Tokyo, Japan) were used to analyze analyze the morphologies and structures of the as-prepared samples. Thermal gravity and the morphologies and structures of the as-prepared samples. Thermal gravity and differential scanning differential scanning calorimeter (TG–DSC) for the samples of g-C3N4 and MgFe2O4/g-C3N4-10 were calorimeter (TG–DSC) for the samples of g-C3 N4 and MgFe2 O4 /g-C3 N4 -10 were recorded on a TA-SDT recorded on a TA-SDT Q600 (TA Instruments, New Castle, DE, USA) over a temperature range of Q600 (TA Instruments, New Castle, DE, USA) over a temperature range of 30–800 ◦ C at a heating rate -1 30–800 °C at a heating rate of 10 °C min under a flowing air atmosphere. The porous features of the of 10 ◦ C min-1 under a flowing air atmosphere. The porous features of the samples were characterized samples were characterized by the N2 adsorption–desorption measurement (Quantachrome by the N2 adsorption–desorption measurement (Quantachrome Autosorb-iQ2 sorption analyzer) Autosorb-iQ2 sorption analyzer) (Quantachrome, Boynton Beach, FL, USA). Before obtained the (Quantachrome, Boynton Beach, FL, USA). Before obtained the measurement, the samples were measurement, the samples were degassed at 200 °C for more than 12 h. The specific surface area of degassed at 200 ◦ C for more than 12 h. The specific surface area of the sample was estimated by the sample was estimated by using the Brunauer–Emmett–Teller (BET) method and the pore size using the Brunauer–Emmett–Teller (BET) method and the pore size distribution was derived from the distribution was derived from the Density functional theory (DFT) method. Density functional theory (DFT) method.

2.3. Gas Sensing Property Test 2.3. Gas Sensing Property Test The gas-sensing property 4 porous microspheres and MgFe2O4/g-C3N4 The gas-sensing propertytests testsofofthe thepure pureMgFe MgFe2O 2 O4 porous microspheres and MgFe2 O4 /g-C3 N4 composites with different contents of g-C 3N4 were investigated by using an intelligent gas sensing composites with different contents of g-C3 N4 were investigated by using an intelligent gas sensing Ltd. Beijing, China ). The fabrication and test analysis system of of CGS-4TPS (Beijing Elite Tech. Co., analysis system CGS-4TPS (Beijing Elite Tech. Co., Ltd. Beijing, China). The fabrication and test process for the sensors is similar to our previously reported method. Figure 1 shows a simple device process for the sensors is similar to our previously reported method. Figure 1 shows a simple device schematic diagram. The relative humidity is 25% and thethe temperature is is 2525 °C◦ C in in thethe test chamber schematic diagram. The relative humidity is 25% and temperature test chamber during the process of the gas-sensing testing. The response of the gas sensor is defined as follows: during the process of the gas-sensing testing. The response of the gas sensor is defined as follows: Response g (Ra and Rg were the resistances of the sensor measured in air and in test Response= =RaR/R /R (R and R were the resistances of the sensor measured in air and in test gas, a g a g gas, respectively). respectively).

Ag-Pd Electrodes

ceramic substrate

Sensing film

ceramic substrate

(a)

(b)

Figure 1. (a) The CGS-4TPS gas-sensing test system and (b) the gas sensor substrate. Figure 1. (a) The CGS-4TPS gas-sensing test system and (b) the gas sensor substrate.

3. Results and Discussion 3. Results and Discussion 3.1. Sample Characterization 3.1. Sample Characterization Figure showsthethe patterns g-C3N4, MgFe2O4 porous microspheres and Figure22 shows XRDXRD patterns of g-C3 Nof 4 , MgFe2 O4 porous microspheres and MgFe2 O4 /g-C3 N4 MgFe 2O4/g-C3N4 composites with different g-C3N4 contents. We found that the diffraction peaks of composites with different g-C3 N4 contents. We found that the diffraction peaks of MgFe2 O4 were MgFe 2O4 were consistent with the standard pattern of MgFe2O4 (JCPDS card No. 17-0464) [32]. For consistent with the standard pattern of MgFe2 O4 (JCPDS card No. 17-0464) [32]. For g-C3 N4 , a strong g-C 3N4, a strong peak appears at around 27.71° that to (002) diffraction plane card peak appears at around 27.71◦ that corresponds to corresponds (002) diffraction plane (JCPDS card No.(JCPDS 87-1526) [33], No. 87-1526) [33], which is well-known for the melon network. Another peak at 12.81° corresponds to ◦ which is well-known for the melon network. Another peak at 12.81 corresponds to (100) ordering (100) ordering of tri-s-triazine units. However, there is no diffraction of g-C 3N4 observed in the of tri-s-triazine units. However, there is no diffraction peak of peak g-C3 N 4 observed in the curves curves of MgFe 2O4/g-C3N4 composites. This may be due to the relatively small content of g-C3N4 in of MgFe2 O4 /g-C3 N4 composites. This may be due to the relatively small content of g-C3 N4 in thethe composite. composite. The morphologies and structures of of thethe as-prepared samples were verified byby FESEM and TEM. The morphologies and structures as-prepared samples were verified FESEM and TEM. Figure 3a,d display the SEM and TEM images of pure g-C 3N4. It can be observed that the pure Figure 3a,d display the SEM and TEM images of pure g-C3 N4 . It can be observed that the pure g-C3 N4 g-C 3N4 possesses a typical lamellar structure with many wrinkles. The SEM and TEM images of the possesses a typical lamellar structure with many wrinkles. The SEM and TEM images of the MgFe2 O4 MgFe2O4 microspheres are shown in Figure 3b,e. From the images, we can see that the prepared MgFe2O4 consists of very uniform microspheres with a hierarchical structure and a diameter of

Sensors 2018, 18, 2211

4 of 12

microspheres are shown in Figure 3b,e. From the images, we can see that the prepared MgFe O4 Sensors 2018, 18, x FOR PEER REVIEW 4 2of 11 consists of very uniform microspheres with a hierarchical structure and a diameter of 200–250 nm. The TEMnm. image that hundreds ofhundreds nanoparticles form the building blocks of the MgFe O4 200–250 The shows TEM image shows that of nanoparticles form the building blocks of2the microspheres, leading to the formation of a hierarchical structure and porous features. Figure 3c shows MgFe 2O4 microspheres, leading to the formation of a hierarchical structure and porous features. Sensors 2018, 18, x FOR PEER REVIEW 4 of 11 the SEM3cimage MgFe N4MgFe -10 composite. withCompared Figure 3b,with we found 2 Oimage 4 /g-C3of Figure showsofthe SEM 2O4/g-C3N4Compared -10 composite. Figurethat 3b, the we originally O4 microspheres adhered to each other. Figure 3f,From we of confirmed 200–250 nm. TheMgFe TEMdispersed shows that of nanoparticles formFrom the building blocks the 3f, 2image found thatdispersed the originally MgFe 2Ohundreds 4were microspheres were adhered to each other. Figure the existence ofthe g-C we canformation conclude that the association is due to the MgFe2O4 microspheres, leading a hierarchical structure and porous features. 3 N4 and we confirmed existence ofthus, g-C3to N4the and thus, weof can conclude that the phenomenon association phenomenon is Figure 3c shows the SEM image of MgFe 2 O 4 /g-C 3 N 4 -10 composite. Compared with Figure 3b, we introduction of g-C3 N4 . of g-C3N4. due to the introduction found that the originally dispersed MgFe2O4 microspheres were adhered to each other. From Figure 3f, we confirmed the existence of g-C3N4 and thus, we can conclude that the association phenomenon is due to the introduction of g-C3N4.

Figure 2. XRD patterns of g-C3N4, MgFe2O4 porous microspheres and MgFe2O4/g-C3N4 composites Figure 2. XRD patterns of g-C3 N4 , MgFe2 O4 porous microspheres and MgFe2 O4 /g-C3 N4 composites 2. XRD patterns of g-C3N4, MgFe2O4 porous microspheres and MgFe2O4/g-C3N4 composites withFigure different g-C3N N4 contents. with different g-C 3 4 contents. with different g-C3N4 contents.

1μm

2μm

2μm

1μm

2μm

2μm

(a)

(a)

(b)

(c)

(b)

(c)

g-C3N4

g-C3N4

0.2μm

0.2μm

0.1μm

(d)

(e)

0.1μm (f)

Figure 3. Scanning electron microscope (SEM) of N (a) g-C3N4; (b) MgFe2O4microspheres porous Figure 3. Scanning electron microscope (SEM) images of (a) g-C (d) (e) images 3 4 ; (b) MgFe2 O4 porous(f) microspheres and (c) MgFe2O4/g-C3N4 composite; as well as transmission electron microscopy (TEM) and (c) MgFe2 O4 /g-C3 N4 composite; as well as transmission electron microscopy (TEM) images of Figure 3. of Scanning (SEM) images of (a)2O4g-C (b) MgFe2O4 porous 4; (e) MgFemicroscope 2O4 porous microspheres and (f) MgFe /g-C33N N44; composite. images (d) g-C3Nelectron (d) g-C3 N4 ; (e) MgFe2 O4 porous microspheres and (f) MgFe2 O4 /g-C3 N4 composite. microspheres and (c) MgFe2O4/g-C3N4 composite; as well as transmission electron microscopy (TEM) In order to determine the high temperature property of the g-C3N4 and MgFe2O4/g-C3N4-10, 3N4; (e) MgFe2O4 porous microspheres and (f) MgFe2O4/g-C3N4 composite. images of (d) g-C TG–DSC analysis was applied. As shown in Figure 4, there are three stages of weight loss in the

In order to determine the high temperature property of the g-C3N4 and MgFe2O4/g-C3N4-10, TG–DSC analysis was applied. As shown in Figure 4, there are three stages of weight loss in the

Sensors 2018, 18, 2211

5 of 12

Sensors 2018, 18, x FOR PEER REVIEW

5 of 11

In order to determine the high temperature property of the g-C3 N4 and MgFe2 O4 /g-C3 N4 -10,

MgFe2OTG–DSC 4/g-C3N4analysis curve according theshown peaksinof DSC4,curve. Thethree firststages stageofinweight the temperature was applied.toAs Figure there are loss in the range of 100–300 °C2 Ois4 /g-C due3to desorption adsorbed andcurve. trapped water and gastemperature molecules. The reason MgFe N4the curve according toofthe peaks of DSC The first stage in the range ◦ of 100–300 is due to theloss desorption of adsorbed andremoval gas molecules. reason solvent of the second stageC of weight between 300 °C and andtrapped 450 °Cwater is the of theThe primary ◦ C and 450 ◦ C is the removal of the primary solvent of the second stage of weight loss between 300 (ethylene glycol). The total lost weight of these two stages is 1.9%. The third stage corresponds to the (ethylene glycol). The total lost weight of these two stages is 1.9%. The third stage corresponds to the temperature range of 475–635 °C ◦in the TG curve of pure g-C3N4, which is due to the combustion of temperature range of 475–635 C in the TG curve of pure g-C3 N4 , which is due to the combustion of g-C3N4 in air (consistent with Figure 4 inset, the TG–DSC analysis of pure g-C3N4), and the weight g-C 3 N4 in air (consistent with Figure 4 inset, the TG–DSC analysis of pure g-C3 N4 ), and the weight loss loss of this is5.1%. 5.1%. proved that2 OMgFe O44-10 /g-C 3N4-10 composite could work of thisstage stage is WeWe proved that MgFe composite could work normally withoutnormally 4 /g-C32N ◦ C. decomposing below the gas sensing test temperature of 475 without decomposing below the gas sensing test temperature of 475 °C.

Figure 4. Thermogravimetry–differential calorimeter (TG–DSC) profiles of4 g-C Figure 4. Thermogravimetry–differentialscanning scanning calorimeter (TG–DSC) profiles of g-C3 N and 3N4 and N -10 composites. /g-C23O N44/g-C -10 composites. MgFe2O4MgFe 3 4 N2 -sorption measurementsof of the the as-prepared as-prepared MgFe microspheres and 2 O42Oporous N2-sorption measurements MgFe 4 porous microspheres and MgFe2 O4 /g-C3 N4 -10 composite were further performed to investigate their specific surface MgFe2O4/g-C3N4-10 composite were further performed to investigate their specific surface area and area and porous structure. As shown in Figure 5, the nitrogen adsorption–desorption isotherms porous structure. As shown in Figure 5, the nitrogen adsorption–desorption isotherms of the two of the two samples show a type IV adsorption-isotherm according to the International Union of samples Pure show type IV adsorption-isotherm according the International Union of Pure and andaApplied Chemistry (IUPAC) classification, which istoindicative of a mesoporous structure. AppliedThe Chemistry (IUPAC) classification, indicative of a which mesoporous structure. The hysteresis loop of MgFe samplesisbelongs to H3 -type, demonstrates the 2 O4 /g-C3 N4 -10which presence of MgFe a laminated structure with narrowbelongs slits formed MgFe2which O4 microspheres and g-Cthe 3 N4presence hysteresis loop of 2O4/g-C 3N4-10 samples to Hby3-type, demonstrates sheet. Figure 5 (inset) depicts the pore size distribution curves of MgFe O microspheres and 2 4 of a laminated structure with narrow slits formed by MgFe2O4 microspheres and g-C3N4 sheet. MgFe2 O4 /g-C3 N4 -10 composite. It can be seen from Figure 5 that the pore diameter of MgFe2 O4 Figure and 5 (inset) depicts the pore size distribution curves of MgFe2O4 microspheres and MgFe2 O4 /g-C3 N4 -10 composite is concentrated in the ranges of about 30–40 nm and 5–15 nm, MgFe2O4respectively. /g-C3N4-10This composite. It can bethe seen from Figure 5 that the pore diameter of3 NMgFe 2O4 and is in agreement with TEM analysis results. The result illustrates that g-C 4 fills MgFe2O4the /g-C 3N4-10large composite is concentrated in the ranges of surface about areas 30–40 nm 2and 5–15 nm, relatively pores between the MgFe2 O4 microspheres. The BET of MgFe O4 and 2 − 1 2 − 1 MgFe2 OThis N4 -10 were calculated to be 11.0 m ·g and 16.8results. m ·g , respectively. 4 /g-C3is respectively. in samples agreement with the TEM analysis The result illustrates that g-C3N4 fills the relatively large pores between the MgFe2O4 microspheres. The BET surface areas of MgFe2O4 and MgFe2O4/g-C3N4-10 samples were calculated to be 11.0 m2·g−1 and 16.8m2·g−1, respectively.

MgFe2O4/g-C3N4-10 composite is concentrated in the ranges of about 30–40 nm and 5–15 nm, respectively. This is in agreement with the TEM analysis results. The result illustrates that g-C3N4 fills the relatively large pores between the MgFe2O4 microspheres. The BET surface areas of MgFe2O4 and MgFe2O4/g-C3N4-10 samples were calculated to be 11.0 m2·g−1 and Sensors 2018, 18, 2211 6 of 12 16.8m2·g−1, respectively.

isotherms (inset)the the corresponding pore size distribution Figure 5.Figure N2 xadsorption–desorption 5. N2PEER adsorption–desorption isothermsand and (inset) corresponding pore size distribution Sensors 2018, 18, FOR REVIEW 6 of 11 of the 2MgFe microspheresand andMgFe MgFe22O composites. curves ofcurves the MgFe O4 porous microspheres O44/g-C /g-C33NN4 -10 4-10 composites. 2 O4 porous

3.2. Gas Sensing Property 3.2. Gas Sensing Property

It is well known that the working temperature greatly influences the gas-sensing performance It is well known that the working temperature influences gas-sensing performance of MOS-based sensor because the gas adsorption andgreatly desorption and the surface reaction kinetics are all of MOS-based sensor because the gas adsorption and desorption and surface reaction kinetics are all closely related with the working temperature. Figure 6 shows the response of MgFe2O4 porous closely related with the working temperature. Figure 6 shows the response of MgFe2 O4 porous microspheres and MgFe2O4/g-C3N4 composites-based sensors to 500 ppm acetone at different microspheres and MgFe2 O4 /g-C3 N4 composites-based sensors to 500 ppm acetone at different operating temperatures. It can be seen from Figure 6 that all response curves of sensors exhibit operating temperatures. It can be seen from Figure 6 that all response curves of sensors exhibit “increase–maximum–decrease” temperature.Comparing Comparing different curves, “increase–maximum–decrease”trends trendswith with increasing increasing temperature. different curves, the results demonstrate that the introduction of g-C 3 N 4 into MgFe 2 O 4 can greatly enhance the sensor’s the results demonstrate that the introduction of g-C3 N4 into MgFe2 O4 can greatly enhance the sensor’s response to acetone, with thethe best g-Cg-C 3N34N content inin MgFe 2O 4/g-C 3N34N composites being response to acetone, with best MgFe being1010wt wt%. %. The 4 content 2O 4 /g-C 4 composites /g-C3 N4 -10-based exhibits highest response to acetone (which is(which 275 at is MgFeThe 2O4MgFe /g-C3N sensor sensor exhibits thethehighest responsevalue value to acetone 2 O44-10-based ◦ C). Meanwhile, the highest response of the sensor based on pure MgFe O porous microspheres 320320 275 at °C). Meanwhile, the highest response of the sensor based 2on4 pure MgFe2O4 porous is 1.9 at 380is◦ C. the MgFe2with O4 -based sensor,2Othe maximum response of the sensor based of microspheres 1.9Compared at 380 °C.with Compared the MgFe 4-based sensor, the maximum response on MgFe is improved by about 145 times and the optimum temperature is lowered 2 O4 /g-C 3 N4 -10 the sensor based on MgFe 2O4/g-C3N4-10 is improved by about 145 times and the optimum temperature by 60 ◦ C. is lowered by 60 °C.

Figure 6. 6.Response values sensors based basedononpurepure 2O4, MgFe2O4/g-C3N4-5, Figure Response valuesofof the the sensors MgFeMgFe 2 O4 , MgFe2 O4 /g-C3 N4 -5, 2O4/g-C 3 N 4 -10 and MgFe 2 O 4 /g-C 3 N 4 -15 composites to 500 ppm acetone as a function of MgFeMgFe 2 O4 /g-C3 N4 -10 and MgFe2 O4 /g-C3 N4 -15 composites to 500 ppm acetone as a function of operating temperature. operating temperature.

Figure 7a,b illustrate the response value curves of the pure MgFe2O4 porous microspheres and MgFe2O4/g-C3N4-10-based sensors to varied concentrations of acetone under the working temperature of 320 °C. With an increase in acetone concentration, more acetone molecules can adsorb to the materials’ surface, inducing a rise in the response value. According to the difference of

Sensors 2018, 18, 2211

7 of 12

Figure 7a,b illustrate the response value curves of the pure MgFe2 O4 porous microspheres and MgFe2 O4 /g-C3 N4 -10-based sensors to varied concentrations of acetone under the working temperature of 320 ◦ C. With an increase in acetone concentration, more acetone molecules can adsorb to the materials’ surface, inducing a rise in the response value. According to the difference of acetone concentration range, two linear relationships were fitted respectively. The fitting linear relationships between the response value and acetone concentration are shown in Figure 7a,b, which provides a possibility for accurately monitoring acetone concentration. The responses of the sensor based on MgFe2 O4 /g-C3 N4 -10 towards 500, 1000 and 2000 ppm acetone were 271.1, 580 and 832, respectively. Meanwhile, the response values of MgFe2 O4 porous microspheres based sensor were 1.8, 3.2 and 3.0, respectively. The results proved that the degree of gas sensor promotion is more obvious with an Sensorsincrease 2018, 18, in x FOR PEERconcentration. REVIEW acetone According to the IUPAC definition, the limit of detection (LoD) 7 of 11 = 3(Noiserms /slope) [34]. The sensor noise was extracted from the root-mean-square (rms) deviation of indicates that thefluctuation MgFe2O4using /g-C330 N4points composite could be a good candidate forslope the was selective detection the response at the baseline without target gas and the calculated from the linear part in Figure 7. The limit of detection of sensor is determined to be 30 ppb. of acetone.

(a)

(b)

Figure 7. The response value of the sensors based on (a) MgFe2O4/g-C3N4-10 and (b) MgFe2O4 porous Figure 7. The response value of the sensors based on (a) MgFe2 O4 /g-C3 N4 -10 and (b) MgFe2 O4 porous microspheres to the varied concentrations atthe theworking working temperature of◦320 °Ctheir and their microspheres to the varied concentrationsof of acetone acetone at temperature of 320 C and fittingfitting linear. linear. Figure 8 depicts the response values of the MgFe2 O4 /g-C3 N4 -10-based sensor and the MgFe2 O4 porous microspheres-based sensor to different gases with the same concentration (500 ppm) at 320 ◦ C, including ethanol, methanal, methanol, propanediol and acetone. Obviously, it reveals that the sensor based on MgFe2 O4 /g-C3 N4 -10 has admirable selectivity to acetone than to other gases compared to the MgFe2 O4 porous microspheres-based sensor at 320 ◦ C. This selective test results indicates that the MgFe2 O4 /g-C3 N4 composite could be a good candidate for the selective detection of acetone. Response and recovery times are the important parameters of gas sensors, which are defined as the time required to reach 90% of the change of sensor resistance. The curve of MgFe2 O4 /g-C3 N4 -10-based sensor to 500 ppm acetone at a working temperature of 320 ◦ C is shown in Figure 9. The response time and recovery time were calculated to be 49 s and 29 s, respectively. The result indicates that the gas sensing material of MgFe2 O4 /g-C3 N4 -10 possesses good response–recovery characteristics to acetone and could meet the requirements of practical applications.

Figure 8. Response values of the sensors based on MgFe2O4 and MgFe2O4/g-C3N4-10 to 500 ppm of different types of tested gas at working temperature of 320 °C.

Response and recovery times are the important parameters of gas sensors, which are defined as the time required to reach 90% of the change of sensor resistance. The curve of MgFe2O4/g-C3N4-10-based sensor to 500 ppm acetone at a working temperature of 320 °C is shown in Figure 9. The response time and recovery time were calculated to be 49 s and 29 s, respectively.

(a)

(b)

Figure 7. The response value of the sensors based on (a) MgFe2O4/g-C3N4-10 and (b) MgFe2O4 porous microspheres to the varied concentrations of acetone at the working temperature of 320 °C and their Sensors 2018, 18, 2211 8 of 12 fitting linear.

Figure 8.Figure Response values of the sensors based on MgFe2O4 and MgFe2O4/g-C3N4-10 to 500 ppm of 8. Response values of the sensors based on MgFe2 O4 and MgFe2 O4 /g-C3 N4 -10 to 500 ppm of different different types oftypes tested gas at working temperature of320 320◦ C. °C. of tested gas at working temperature of

Sensors 2018, 18, x FOR PEER REVIEW

8 of 11

Response and recovery times are the important parameters of gas sensors, which are defined as the time required to reach 90% of the change of sensor resistance. The curve of MgFe2O4/g-C3N4-10-based sensor to 500 ppm acetone at a working temperature of 320 °C is shown in Figure 9. The response time and recovery time were calculated to be 49 s and 29 s, respectively. The result indicates that the gas sensing material of MgFe2O4/g-C3N4-10 possesses good response–recovery characteristics to acetone and could meet the requirements of practical applications. In order to evaluate the repeatability and stability of the gas sensing properties of MgFe2O4/g-C3N4-10 material over time, the sensor responses to 500 ppm of acetone were measured during a period of 60 days and the sensor structure was prepared by using the same drop coating method. Figure 10 displays the long-term stability test results of MgFe2O4/g-C3N4-10-based sensor in 60 days. There is a small change within a certain range in the response of sensor from this graph, which further confirms that the MgFe2O4/g-C3N4-10-based sensor might have a practical application based on its good long-term stability. Figure 9.Figure Response –recovery time curve O4/g-C /g-C3NN-10-based 4-10-based sensor to 500 ppm acetone at 9. Response–recovery time curveof of MgFe MgFe22O sensor to 500 ppm acetone at 4 3 4 workingworking temperature of 320 temperature of °C. 320 ◦ C. In order to evaluate the repeatability and stability of the gas sensing properties of MgFe2 O4 /g-C3 N4 -10 material over time, the sensor responses to 500 ppm of acetone were measured during a period of 60 days and the sensor structure was prepared by using the same drop coating method. Figure 10 displays the long-term stability test results of MgFe2 O4 /g-C3 N4 -10-based sensor in 60 days. There is a small change within a certain range in the response of sensor from this graph, which further confirms that the MgFe2 O4 /g-C3 N4 -10-based sensor might have a practical application based on its good long-term stability.

Figure 10. Stability measurement of the sensor based on MgFe2O4/g-C3N4-10-10 to 500 ppm acetone

Figure 9. 2018, Response Sensors 18, 2211 –recovery time curve of MgFe2O4/g-C3N4-10-based sensor to 500 ppm acetone 9 of 12 at working temperature of 320 °C.

Figure 10. Stability measurement ofofthe based MgFe 2O4/g-C3N4-10-10 to 500 ppm acetone Figure 10. Stability measurement thesensor sensor based onon MgFe 2 O4 /g-C3 N4 -10-10 to 500 ppm acetone ◦ at 320 °C. at 320 C. 3.3. Gas Sensing Mechanism 3.3. Gas Sensing Mechanism A modulation model based on the electronic depletion layer [35,36] can explain the gas sensing A modulation model based on the electronic depletion layer [35,36] can explain the gas sensing property. It is well known that MgFe2 O4 and g-C3 N4 are n-type semiconductors. The gas sensitivity property. It is well known that MgFe2O4 and g-C3N4 are n-type semiconductors. The gas sensitivity of the sensor based on n-type semiconductor material essentially originates from the change of its of the sensor on when n-type essentially originates change of its electricbased resistance thesemiconductor sensor is exposedmaterial to different gas atmospheres. Takingfrom MgFethe 2 O4 as an example, when the sensor is exposed air, O2 will adsorbed gas on the surface of MgFeTaking electric resistance when the sensor is to exposed tobedifferent atmospheres. MgFe2O4 as an 2 O4 . The oxygen act as electron the O conduction and form absorbed example,molecules when the sensor is acceptors exposed from to air, 2 will be band adsorbed onsurface the surface of oxygen MgFe2O4. The anions, such as O2 − , O− , and O2− (Equations (1)–(3)). This leads to the formation of a relatively oxygen molecules act as electron acceptors from the conduction band and form surface absorbed thick electronic depletion layer, which results in an increase in the width of the potential barrier and oxygen anions, such as O2−, O−, and O2− (Equations (1)–(3)). This leads to the formation of a relatively results in the high resistance of the sensor (Ra ). When the sensors are exposed to acetone, the former thick electronic depletion layer, which results in anwill increase in the thereducing width of potential barrier and oxygen anions absorbed on the surface of material react with gasthe (Equations (4)–(6)) trapped electrons anionsthe aresensors releasedare back to the conduction band. results inand thethehigh resistance ofby theabsorbed sensor oxygen (Ra). When exposed to acetone, the former The resistance of sensors (R ) can be thus decreased, which accompanies the decrease in the width of (4)–(6)) g oxygen anions absorbed on the surface of material will react with the reducing gas (Equations the potential barrier. and the trapped electrons by absorbed oxygen anions are released back to the conduction band. The O2 (ads) + e− → O2− (ads) (1) resistance of sensors (Rg) can be thus decreased, which accompanies the decrease in the width of the O2− (ads) + e− → 2O(−ads) (2) potential barrier. O−

+ e− → O 2−

− ) ) + e− → O (ads O2((ads 2(ads) ads)

(3)

CH3 COCH3 (ads) + 4O2− (ads) → 3CO2 + 3H2 O + 4e−

(4)

− − CH3 COCH3 (ads O)− + 8O+(adse)- → →3CO 2O2−+ 3H2O + 8e

(5)

− CH3 COCH3 (ads) + 8O(2ads → 3CO2 + 3H2 O + 16e− )

(6)

2 ( ads)

( ads)

In this work, we found that MgFe2 O4 /g-C3 N4 -baesd sensors exhibit better acetone sensing properties than pure MgFe2 O4 -based sensors. One of the possible reasons for the improved gas performance may be attributed to the heterojunction between the MgFe2 O4 microspheres and g-C3 N4 nanosheet. When the composite sensor was exposed to air, the process of electrons inflow from a component to another. The existence of heterojunction could lead to a higher potential barrier, which further results in an obvious increase in the sensor resistance (Ra ). In contrast, due to the surface redox reaction between oxygen anions and acetone molecules, the potential barrier is decreased and the

(1) (2)

Sensors 2018, 18, 2211

10 of 12

sensor resistance (Rg ) is decreased correspondingly. Meanwhile, the larger surface contributes to more oxygen molecules and acetone molecules being adsorbed on the surface of sensor, which could further enhance the redox reaction between acetone molecules and absorbed oxygen anions. 4. Conclusions In summary, we reported a high sensitivity acetone sensor based on MgFe2 O4 /g-C3 N4 composite, which was synthesized via a one-step solvothermal method. By analyzing the given results of XRD, FESEM, TEM, TG–DSC and N2 -sorption, we proved the existence of g-C3 N4 nanosheet in the MgFe2 O4 /g-C3 N4 composite. Due to the introduction of g-C3 N4, the gas sensing property of MgFe2 O4 -based sensor is remarkably improved. Among these composites with different g-C3 N4 contents, the MgFe2 O4 /g-C3 N4 -10-based sensor exhibited the eximious sensing performance to acetone, such as high sensitivity and selectivity, quick response and recovery as well as favorable stability. The as-prepared MgFe2 O4 /g-C3 N4 composite could be a promising candidate for practical applications requiring highly sensitive acetone gas sensors. Author Contributions: R.Z. performed the experiments and analyzed the data; Z.Z. analyzed the gas sensing mechanism; Y.W. and J.C. provided the concept and wrote the paper as the corresponding authors. Funding: This research was funded by the National Natural Science Foundation of China (U1704255, U1704146), Program for Science & Technology Innovation Talents in Universities of Henan Province (19HASTIT042), Natural Science Foundation of Henan Province of China (162300410113), the Research Foundation for Youth Scholars of Higher Education of Henan Province (2017GGJS053, 2016GGJS-040), the Fundamental Research Funds for the Universities of Henan Province (NSFRF170201, NSFRF1606), Program for Innovative Research Team in University of Ministry of Education of China (IRT-16R22), Program for Innovative Research Team of Henan Polytechnic University (T2018-2), Foundation for Distinguished Young Scientists of Henan Polytechnic University (J2016-2, J2017-3) for financial support. Conflicts of Interest: The authors declare no conflict of interest.

References 1. 2.

3. 4.

5.

6.

7. 8. 9.

Zhang, K.; Pei, Z.; Wang, D. Organic solvent pretreatment of lignocellulosic biomass for biofuels and biochemicals: A review. Bioresour. Technol. 2016, 199, 21–33. [CrossRef] [PubMed] Wang, X.-F.; Ma, W.; Jiang, F.; Cao, E.-S.; Sun, K.-M.; Cheng, L.; Song, X.-Z. Prussian blue analogue derived porous NiFe2 O4 nanotubes for low-concentration acetone sensing at low working temperature. Chem. Eng. J. 2018, 338, 504–512. [CrossRef] Lourenço, C.; Turner, C. Breath analysis in disease diagnosis: Methodological considerations and applications. Metabolites 2014, 4, 465–498. [CrossRef] [PubMed] Ueta, I.; Saito, Y.; Hosoe, M.; Okamoto, M.; Ohkita, H.; Shirai, S.; Tamura, H.; Jinno, K. Breath acetone analysis with miniaturized sample preparation device: In-needle preconcentration and subsequent determination by gas chromatography–mass spectroscopy. J. Chromatogr. B Anal. Technol. Biomed. Life Sci. 2009, 877, 2551–2556. [CrossRef] [PubMed] Li, Y.; Luo, N.; Sun, G.; Zhang, B.; Ma, G.; Jin, H.; Wang, Y.; Cao, J.; Zhang, Z. Facile synthesis of ZnFe2 O4 /α-Fe2 O3 porous microrods with enhanced tea-sensing performance. J. Alloys Compd. 2018, 737, 255–262. [CrossRef] Zhang, S.; Sun, G.; Li, Y.; Zhang, B.; Lin, L.; Wang, Y.; Cao, J.; Zhang, Z. Continuously improved gas-sensing performance of SnO2 /Zn2 SnO4 porous cubes by structure evolution and further NIO decoration. Sens. Actuators B 2018, 255, 2936–2943. [CrossRef] Ma, R.-J.; Li, G.-D.; Zou, X.; Gao, R.; Chen, H.; Zhao, X. Bimetallic Pt–Au nanocatalysts decorated In2 O3 nests composed of ultrathin nanosheets for Type 1 diabetes diagnosis. Sens. Actuators B 2018, 270, 247–255. Sachdeva, S.; Agarwal, A.; Agarwal, R. Tungsten Oxide Thin Film Characterizations for Acetone Gas Detection. MAPAN 2018, 33, 57–62. [CrossRef] Chen, L.; Huang, L.; Lin, Y.; Sai, L.; Chang, Q.; Shi, W.; Chen, Q. Fully gravure-printed WO3 /Pt-decorated rGO nanosheets composite film for detection of acetone. Sens. Actuators B 2018, 255, 1482–1490. [CrossRef]

Sensors 2018, 18, 2211

10.

11. 12.

13. 14.

15. 16. 17. 18. 19. 20. 21.

22.

23.

24.

25.

26.

27.

28.

29. 30.

11 of 12

Qin, L.; Xu, J.; Dong, X.; Pan, Q.; Cheng, Z.; Xiang, Q.; Li, F. The template-free synthesis of square-shaped SnO2 nanowires: The temperature effect and acetone gas sensors. Nanotechnology 2008, 19, 185705. [CrossRef] [PubMed] Tang, L.; Li, Y.; Xu, K.; Hou, X.; Lv, Y. Sensitive and selective acetone sensor based on its cataluminescence from nano-La2 O3 surface. Sens. Actuators B 2008, 132, 243–249. [CrossRef] Zhang, R.; Zhou, T.; Wang, L.; Zhang, T. Metal−Organic Frameworks-Derived Hierarchical Co3 O4 Structures as Efficient Sensing Materials for Acetone Detection. ACS Appl. Mater. Interfaces 2018, 10, 9765–9773. [CrossRef] [PubMed] Li, S.-M.; Zhang, L.-X.; Zhu, M.-Y.; Ji, G.-J.; Zhao, L.-X.; Yin, J.; Bie, L.-J. Acetone sensing of ZnO nanosheets synthesized using room-temperature precipitation. Sens. Actuators B 2017, 249, 611–623. [CrossRef] Cheng, Y.; He, Y.; Li, S.; Wang, Y.; Zhao, Y.; Li, Y.; Li, H.; Liu, L. Ultra-sensitive and selective acetone gas sensor with fast response at low temperature based on Cu-doped α-Fe2 O3 porous nanotubes. J. Mater. Sci. Mater. Electron. 2018, 29, 11178–11186. [CrossRef] Sickafus, K.E.; Wills, J.M.; Grimes, N.W. Structure of spinel. J. Am. Ceram. Soc. 2010, 82, 3279–3292. [CrossRef] Liu, Y.-L.; Liu, Z.-M.; Yang, Y.; Yang, H.-F.; Shen, G.-L.; Yu, R.-Q. Simple synthesis of MgFe2 O4 nanoparticles as gas sensing materials. Sens. Actuators B 2005, 107, 600–604. [CrossRef] Hankare, P.P.; Jadhav, S.D.; Sankpal, U.B.; Patil, R.P.; Sasikala, R.; Mulla, I.S. Gas sensing properties of magnesium ferrite prepared by co-precipitation method. J. Alloys Compd. 2009, 488, 270–272. [CrossRef] Godbole, R.; Rao, P.; Bhagwat, S. Magnesium ferrite nanoparticles: A rapid gas sensor for alcohol. Mater. Res. Express 2017, 4, 025032. [CrossRef] Patil, J.; Nadargi, D.; Mulla, I.S.; Suryavanshi, S.S. Spinel MgFe2 O4 thick films: A colloidal approach for developing gas sensors. Mater. Lett. 2018, 213, 27–30. [CrossRef] Lu, J.; Jia, N.; Cheng, L.; Liang, K.; Huang, J.; Li, J. Rgo/CoTiO3 nanocomposite with enhanced gas sensing performance at low working temperature. J. Alloys Compd. 2018, 739, 227–234. [CrossRef] Hassan, M.; Wang, Z.H.; Huang, W.R.; Li, M.Q.; Liu, J.W.; Chen, J.F. Ultrathin tungsten oxide nanowires/reduced graphene oxide composites for toluene sensing. Sensors 2017, 17, 2245. [CrossRef] [PubMed] Li, Y.; Luo, N.; Sun, G.; Zhang, B.; Lin, L.; Jin, H.; Wang, Y.; Bala, H.; Cao, J.; Zhang, Z. In situ decoration of Zn2 SnO4 nanoparticles on reduced graphene oxide for high performance ethanol sensor. Ceram. Int. 2018, 44, 6836–6842. [CrossRef] Mamba, G.; Mishra, A.K. Graphitic carbon nitride (g-C3 N4 ) nanocomposites: A new and exciting generation of visible light driven photocatalysts for environmental pollution remediation. Appl. Catal. B 2016, 198, 347–377. [CrossRef] Hu, Y.; Li, L.; Zhang, L.; Lv, Y. Dielectric barrier discharge plasma-assisted fabrication of g-C3 N4 -Mn3 O4 composite for high-performance cataluminescence H2 S gas sensor. Sens. Actuators B 2017, 239, 1177–1184. [CrossRef] Gong, Y.; Wang, Y.; Sun, G.; Jia, T.; Jia, L.; Zhang, F.; Lin, L.; Zhang, B.; Cao, J.; Zhang, Z. Carbon nitride decorated ball-flower like Co3 O4 hybrid composite: Hydrothermal synthesis and ethanol gas sensing application. Nanomaterials 2018, 8, E132. [CrossRef] [PubMed] Dai, H.; Zhang, S.; Xu, G.; Peng, Y.; Gong, L.; Li, X.; Li, Y.; Lin, Y.; Chen, G. Highly photoactive heterojunction based on g-C3 N4 nanosheets decorated with dendritic zinc(ii) phthalocyanine through axial coordination and its ultrasensitive enzyme-free sensing of choline. RSC Adv. 2014, 4, 58226–58230. [CrossRef] Shao, L.; Jiang, D.; Xiao, P.; Zhu, L.; Meng, S.; Chen, M. Enhancement of g-C3 N4 nanosheets photocatalysis by synergistic interaction of ZnS microsphere and rGO inducing multistep charge transfer. Appl. Catal. B 2016, 198, 200–210. [CrossRef] Cao, J.; Qin, C.; Wang, Y.; Zhang, H.; Zhang, B.; Gong, Y.; Wang, X.; Sun, G.; Bala, H.; Zhang, Z. Synthesis of g-C3 N4 nanosheet modified SnO2 composites with improved performance for ethanol gas sensing. RSC Adv. 2017, 7, 25504–25511. [CrossRef] Hu, J.; Zou, C.; Su, Y.; Li, M.; Yang, Z.; Ge, M.; Zhang, Y. One-step synthesis of 2D C3 N4 -tin oxide gas sensors for enhanced acetone vapor detection. Sens. Actuators B 2017, 253, 641–651. [CrossRef] Cao, J.; Gong, Y.; Wang, Y.; Zhang, B.; Zhang, H.; Sun, G.; Bala, H.; Zhang, Z. Cocoon-like ZnO decorated graphitic carbon nitride nanocomposite: Hydrothermal synthesis and ethanol gas sensing application. Mater. Lett. 2017, 198, 76–80. [CrossRef]

Sensors 2018, 18, 2211

31.

32.

33. 34.

35. 36.

12 of 12

Cao, J.; Qin, C.; Wang, Y.; Zhang, B.; Gong, Y.; Zhang, H.; Sun, G.; Bala, H.; Zhang, Z. Calcination method synthesis of SnO2 /g-C3 N4 composites for a high-performance ethanol gas sensing application. Nanomaterials 2017, 7, E98. [CrossRef] [PubMed] Zu, Y.; Zhao, Y.; Xu, K.; Tong, Y.; Zhao, F. Preparation and comparison of catalytic performance for nano MgFe2 O4 , go-loaded MgFe2 O4 and go-coated MgFe2 O4 nanocomposites. Ceram. Int. 2016, 42, 18844–18850. [CrossRef] Ao, Z.M.; Peeters, F.M. High-capacity hydrogen storage in al-adsorbed graphene. Phys. Rev. B 2010, 81, 2498–2502. [CrossRef] Tshabalala, Z.; Shingange, K.; Dhonge, B.; Ntwaeaborwa, O.; Mhlongo, G.; Motaung, D. Fabrication of ultra-high sensitive and selective CH4 , room temperature gas sensing of TiO2 nanorods: Detailed study on the annealing temperature. Sens. Actuators B 2017, 238, 402–419. [CrossRef] Liao, L.; Lu, H.B.; Li, J.C.; He, H.; Wang, D.F.; And, D.J.F.; Liu, C.; Zhang, W.F. Size dependence of gas sensitivity of ZnO nanorods. J. Phys. Chem. C 2007, 111, 1900–1903. [CrossRef] Ghosh, A.; Bhowmick, T.; Majumder, S. Multi-layered zinc oxide-graphene composite thin films for selective nitrogen dioxide sensing. J. Appl. Phys. 2018, 123, 084501. [CrossRef] © 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).