Deriving Optimized PID Parameters of Nano-Ag Colloid Prepared by Electrical Spark Discharge Method

nanomaterials

Article Deriving Optimized PID Parameters of Nano-Ag Colloid Prepared by Electrical Spark Discharge Method

Kuo-Hsiung Tseng 1,* , Yur-Shan Lin 1, Yun-Chung Lin 2, Der-Chi Tien 1 and Leszek Stobinski 3 1 Department of Electrical Engineering, National Taipei University of Technology, Taipei 10608, Taiwan; [email protected] (Y.-S.L.); [email protected] (D.-C.T.) 2 Power Department, Quanta Computer lnc., Taipei 111, Taiwan; [email protected] 3 Materials Chemistry, Warsaw University of Technology, Warynskiego 1, 00-645 Warsaw, Poland; [email protected] * Correspondence: [email protected]

Received: 3 March 2020; Accepted: 27 May 2020; Published: 1 June 2020

Abstract: Using the electrical spark discharge method, this study prepared a nano-Ag colloid using self-developed, microelectrical discharge machining equipment. Requiring no additional surfactant, the approach in question can be used at the ambient temperature and pressure. Moreover, this novel physical method of preparation produced no chemical pollution. This study conducted an in-depth investigation to establish the following electrical discharge conditions: gap electrical discharge, short circuits, and open circuits. Short circuits affect system lifespan and cause electrode consumption, resulting in large, non-nanoscale particles. Accordingly, in this study, research for and design of a new logic judgment circuit set was used to determine the short-circuit rate. The Ziegler–Nichols proportional–integral–derivative (PID) method was then adopted to find optimal PID values for reducing the ratio between short-circuit and discharge rates of the system. The particle size, zeta potential, and ultraviolet spectrum of the nano-Ag colloid prepared using the aforementioned method were also analyzed with nanoanalysis equipment. Lastly, the characteristics of nanosized particles were analyzed with a transmission electron microscope. This study found that the lowest ratio between short-circuit rates was obtained (1.77%) when PID parameters were such that Kp was 0.96, Ki was 5.760576, and Kd was 0.039996. For the nano-Ag colloid prepared using the aforementioned PID parameters, the particle size was 3.409 nm, zeta potential was approximately 46.8 mV, absorbance was approximately 0.26, and surface plasmon resonance was 390 nm. Therefore, − this study demonstrated that reducing the short-circuit rate can substantially enhance the effectiveness of the preparation and produce an optimal nano-Ag colloid.

Keywords: electrical spark discharge method; nano-Ag colloid; Ziegler–Nichols method; electrical discharge condition; short circuits

1. Introduction Nanotechnology entails the science and technology that apply the physical and chemical characteristics of substances smaller than 100 nm [1] to the design and production of new components and systems. Because of the structural characteristics of increased surface area and a lack of periodic regularity, as well as interactions between the shape and surface size of nanostructures [2], nanosized substances display physical, chemical, and biological characteristics that are distinct from those of other substances [3]. The techniques used to prepare nanoparticles can be classiﬁed into chemical methods and physical methods. Nearly all the chemical methods require the addition of surfactants. The physical methods include mechanical milling [4], thermal evaporation [5], the submerged arc

Nanomaterials 2020, 10, 1091; doi:10.3390/nano10061091 www.mdpi.com/journal/nanomaterials Nanomaterials 2020, 10, 1091 2 of 12

Nanomaterialsnanoparticle 2020, 10 synthesis, x FOR PEER system REVIEW (SANSS) [6], and so on. Mechanical milling has the potential2 of 12 for contamination from the balls and the atmosphere. Thermal evaporation has the drawback of containing contaminationthe carrier gas from and catalyticthe balls particle and the in theatmosphere. grown nanostructure. Thermal evaporation The process has of SANSSthe drawback should occur of containwithining a the vacuum carrier chamber. gas and Incatalytic this work, particle silver in nanoparticlesthe grown nanostructure. with good purity The process were prepared of SANSS at a shouldnormal occur temperature within a vacuum and pressure chamber. using In the this electrical work, sparksilver dischargenanoparticles method with (ESDM). good purity were preparedElectrical at a normal discharge temperature machining and pressure (EDM) isusing used the for electrical a wide varietyspark discharge of materials method because (ESDM) it uses. heatElectrical rather discharge than mechanical machining principles. (EDM) is For used example, for a wide the variety electric of spark materials discharge because method it uses (ESDM) heat ratherhas than been mechanical applied on principles. hard materials, For example, small components, the electric spark and discharge components method requiring (ESDM) high-precision has been appliedmachining on hard or amaterials, complex small shape. components, Two primary and targets component of ESDMs requiring are precision high and-precision reﬁnement, machining and the or amajor complex ESDM shape. trends Two are primary powder-mixed targets of EDM,ESDM precisionare precision wire and EDM, refinement, and a hybrid and the of major microelectrical ESDM trendsdischarge are powder machining-mixed (micro-EDM) EDM, precision and polishing wire EDM, [7]. and a hybrid of microelectrical discharge machiningFigure (micro1 depicts-EDM) EDM and [polis8]. Bothhing the[7]. upper and lower electrodes were connected to the metal part ofFigure the workpiece 1 depicts and EDM immersed [8]. Both inthe a highlyupper and insulating lower electrodes dielectric liquid.were connected In nanotechnology to the metal research, part of thedeionized workpiece water and is oftenimmersed used asin thea highly dielectric insulating liquid. dielectric Application liquid. of direct In nanotechnology current voltage research, enables the deionizedupper electrode water is tooften be controlledused as the by dielectric the servo liq controluid. Application system and of to direct move current slowly voltage toward enables the lower theelectrode. upper electrode Because to thebe controlled two ends ofby the the workpieceservo control are system not in directand to contact, move slowly no physical toward force, the lower such as electrode.contact Because or cutting, the is two involved. ends of When the workpiece the distance are betweennot in direct the two contact, electrodes no physical reaches force, approximately such as contact30 µm, or acutting, discharge is involved. column isWhen formed the betweendistance thebetween two electrodes the two electrodes [9], generating reaches the approximately so-called spark. 30 Theμm, electrica discharge arc can column reach is temperatures formed between as high the as two 5000–6000 electrodes K [10 [9],], causinggenerating atoms the at so the-called metal spark. surface Theto electric melt. This arc can method reach is temperatures called the ESDM. as high as 5000–6000 K [10], causing atoms at the metal surface to melt. This method is called the ESDM.

FigureFigure 1. Electrical 1. Electrical discharge discharge machining machining..

ApplyingApplying ESDM ESDM in in combination combination with with EDM, EDM, the the research research team team of of this this study study successfully successfully preparedprepared several several kind kindss of ofnanocolloid, nanocolloid, including including nano nano-Au,-Au, nano nano-TiO-TiO2, nano2, nano-Al,-Al, and and nanographene nanographene colloidscolloids [11[11–14–14]. Because]. Because traditional traditional EDM EDM equipment equipment is is outdated andand didifficultfficult to to maintain, maintain, micro-EDM micro- EDMequipment equipment was was developed. developed. Micro-EDM Micro-EDM worksworks on the same same principle principle as as EDM EDM.. The The analysis analysis of of nanonano-Au-Au and and nano nano-Ag-Ag colloid colloidss prepared prepared with with this this equipment equipment,, using using high high-precision-precision instruments instruments,, showedshowed that that all all of of the the colloidal colloidal particles particles display displayeded nanoparticle nanoparticle characteristics characteristics [15 [15]. ]. Studies Studies of of nanometalnanometal colloid colloidss prepared prepared using using micro micro-EDM-EDM have have taken taken into into account account only only the the discharge discharge success success rate.rate. This This study study observed observed the the conditions conditions for for currents currents flowing flowing between between electrodes, electrodes, including including gap gap electricalelectrical discharge discharge and and electrode electrode short short circuits, circuits, and and explored explored these these phenomena phenomena in depth. in depth. Because Because the the shortshort-circuit-circuit phenomenon phenomenon may may exert exert an an adverse adverse effect effect on equipment,equipment, thisthis study study designed designed a seta set of of logic logicjudgment judgment circuits circuits to identify to identify short short circuits. circuits. The short-circuitThe short-circuit rate wasrate calculatedwas calculated by computer by computer software softwareusing theusing aforementioned the aforementioned circuit circu outputit output signals. signals. In addition, In addition, this study this study used used the Ziegler–Nichols the Ziegler– Nicholsclassic classic proportional–integral–derivative proportional–integral–derivative (PID) method(PID) method to determine to determine optimal optimal PID parameters PID parameters to reduce to the reduce short-circuit the short- rate.circuit Lastly, rate. Last howly short-circuit, how short- ratecircuit is relatedrate is related to particle to particle analysis, analysis, zeta potential, zeta potential,and absorbance and absorbance was explored was explored [16–19]. [16–19].

2. Materials and Methods

2.1. Logic Circuit Design

Nanomaterials 2020, 10, 1091 3 of 12

2. Materials and Methods

2.1. Logic Circuit Design Nanomaterials 2020, 10, x FOR PEER REVIEW 3 of 12 The electrical discharge conditions of pulse width modulation (PWM), gap voltage (Vgap), and gap

currentThe electrical (Igap) during discharge electrical conditions discharge of pulse are shownwidth modulation in Figure2. (PWM), This comprises gap voltage the transistor-on(Vgap), and gaptime current (Ton )(I andgap) during the transistor-o electricalff dischargetime (Toff are). T shownsuc represents in Figure the 2. successful This comprises electrical the transistor discharge-on time. timeT1 (T representson) and the the transistor time period-off time for gap (Toff electrical). Tsuc represents discharge. the The successful dielectric electrical fluid was discharge observed time. to exhibit T1 representsan insulation the time breakdown period for after gap the electrical spark timedischarge. lag (td The); Vgap dielectricalso decreased fluid was to observed Vsuc, and to Igap exhibitincreased an insulationto Isuc. T2 breakdown is the time after period the spark for short time circuits, lag (td); indicating Vgap also decreased the time at to which Vsuc, and an I electrodegap increased short to circuitIsuc. T2 occurredis the time because period the for anodeshort circuits, and the indicating cathode were the tootime close at which to each an other. electrode Vgap shortdecreased circuit to occurred 0, and Igap becauseincreased the anode to Isc. and The the T2 cathode time period were is too for close the short-circuitto each other. condition. Vgap decreased It indicates to 0, and the Igap time increased for open to circuitsIsc. The atT2 which time period the electric is for field the in short the electrode-circuit condition. gap was insu It indicatesfficiently the strong time to for break open the circuits insulation at of whichthe dielectricthe electric fluid. field V gapin theshows electrode an open-circuit gap was conditioninsufficiently when strong the value to break for I gaptheremained insulation at of 0 Athe [20 ]. dielectric fluid. Vgap shows an open-circuit condition when the value for Igap remained at 0 A [20].

FigureFigure 2. 2.WaveformsWaveforms of of(a) ( apulse) pulse width width modulation modulation (PWM (PWM),), (b ()b V)Vgapgap, and, and (c) (Icgap)I. gap.

ThisThis study’s study’s major major contribution contribution is the is the identification identification of ofthree three electrical electrical discharge discharge conditions: conditions: gap gap electricalelectrical discharge, discharge, short short circuits, circuits, and and open open circuits. circuits. The The logic logic circuits circuits were were designed designed to identify to identify all all electricalelectrical discharge discharge conditions. conditions. First, First, a aset set of of l logicogic circuits identifying andand didifferentiatingfferentiating between between gap gapelectrical electrical discharge discharge and and short short circuits circuits was was designed. designed. The The circuits circuits first first acquire acquire Vgap Vandgap and Igap I,ga andp, and these thesesignals signals are are then then processed processed by by the the comparator comparator and and thethe AND gate gate ( (TheThe AND AND gate gate is isa digital a digital logic logic gategate that that implements implements logical logical conjunction conjunction)) to togenerate generate signals signals determining determining successful successful electrode electrode gap gap discharge.discharge. The The successful successful discharge discharge signals signals are are then then compared compared with with the counter pulsepulse signalssignals fromfrom the thecomputer computer end end to to obtain obtain the the number number of instancesof instances of successfulof successful electrical electrical discharge. discharge. The The short-circuit short- circuitrate rate is determined is determined by directlyby directly comparing comparing the the counter counter pulse pulse signal signal and and the the Igap Igapsignal signal output output by by the thecomputer. computer. When When the the system’s system’s electrical electrical discharge discharge is successful,is successful, Vgap Vgapdecreases decreases and and Igap Igapincreases increases [21 ]. [21Details]. Details of of the the simulation simulation process process are are presented presented as as follows: follows:

2.1.1. V High/Low Levels and I Simulation 2.1.1. Vgap gapHigh/Low Levels and Igap Simulationgap FigureFigure 3 shows3 shows the the measurement measurement of ofelectrode electrode gap gap voltage voltage conditions. conditions. T1 T1 is isthe the time time period period of of thethe electrical electrical discharge discharge condition. condition. T2 T2is the is the short short-circuit-circuit condition. condition. T3 T3is the is the open open-circuit-circuit condition. condition. Figure3a illustrates the V waveform. V is the low-level voltage of V for successful electrical Figure 3a illustrates the Vgapgap waveform. VL isL the low-level voltage of Vgapgap for successful electrical discharge, whereas VH is the high-level voltage. Figure 3b is the simulated waveform of the comparison between Vgap and VL by a comparator. When Vgap is greater than VL, the comparator output signal VL(-) indicates a high potential. Figure 3c is the simulated waveform of the comparison between Vgap and VH by a comparator. When Vgap is smaller than VH, the comparator output signal VH(+) indicates a high potential. Figure 3d shows the logic operation result for VL(-) and VH(+) through the AND gate. Output signals for Vgap and Vsuc are 1 only when both VL(-) and VH(+) are 1. Figure 4

Nanomaterials 2020, 10, 1091 4 of 12

discharge, whereas VH is the high-level voltage. Figure3b is the simulated waveform of the comparison between Vgap and VL by a comparator. When Vgap is greater than VL, the comparator output signal VL(-) indicates a high potential. Figure3c is the simulated waveform of the comparison between V gap and VH by a comparator. When Vgap is smaller than VH, the comparator output signal VH(+) indicates a high potential. Figure3d shows the logic operation result for V L(-) and VH(+) through the AND gate. NanomaterialsOutput signals2020, 10, forx FO VR gapPEERand REVIEW Vsuc are 1 only when both VL(-) and VH(+) are 1. Figure4 illustrates4 of 12 the identiﬁcation of the electrode gap current conditions. Figure4a is the I gap waveform. The low-level illustratescurrent ofthe I gapidentificationfor successful of electricalthe electrode discharge gap current is IL. Figure condit4bions. shows Figure the simulated4a is the I waveformgap waveform. of the Thecomparison low-level current between of IIgapgap andfor successful IL by a comparator. electrical discharge When Igap isis IL greater. Figure than4b shows IL, comparator the simulated output waveformsignals Igap of theand comparison Isuc indicate between a high potential. Igap and Figure IL by a5 indicates comparator. the When logic operation Igap is greater result than of I gap,suc IL, comparatorand Vgap,suc outputthrough signals the I ANDgap and gate. Isuc indicate The output a high signal potential. for Vsuc Figureis 1 only 5 indicates when both the I gap,suclogic operationand Vgap,suc resultare of 1. Igap,suc and Vgap,suc through the AND gate. The output signal for Vsuc is 1 only when both Igap,suc and Vgap,suc are 1.

FigureFigure 3. Voltage 3. Voltage waveforms: waveforms: (a) (Va)Vgap,gap VH,V, andH, and VL; V(bL);( VbL)V inputL input into into the thecomparator comparator;; (c) (VcH)V inputH input into into thethe comparator comparator;; and and (d) (gapd) gap electrical electrical discharge. discharge.

Figure 4. Current Igap: (a) Igap and IL, and (b) Igap and IL input into the comparator.

Nanomaterials 2020, 10, x FOR PEER REVIEW 4 of 12

illustrates the identification of the electrode gap current conditions. Figure 4a is the Igap waveform. The low-level current of Igap for successful electrical discharge is IL. Figure 4b shows the simulated waveform of the comparison between Igap and IL by a comparator. When Igap is greater than IL, comparator output signals Igap and Isuc indicate a high potential. Figure 5 indicates the logic operation result of Igap,suc and Vgap,suc through the AND gate. The output signal for Vsuc is 1 only when both Igap,suc and Vgap,suc are 1.

NanomaterialsFigure 3.2020 Voltage, 10, 1091 waveforms: (a) Vgap, VH, and VL; (b) VL input into the comparator; (c) VH input into 5 of 12 the comparator; and (d) gap electrical discharge.

Nanomaterials 2020, 10, x FOR PEER REVIEW 5 of 12 NanomaterialsNanomaterials 20 202020, 10, 10, x, xFO FORR PEER PEER REVIEW REVIEW 5 5of of 12 12 Figure 4. Current Igap:(a)Igap and IL, and (b)Igap and IL input into the comparator. Figure 4. Current Igap: (a) Igap and IL, and (b) Igap and IL input into the comparator.

Figure 5. Gap electrical discharge simulation. Figure 5. Gap electrical discharge simulation. FigureFigure 5. 5. Gap Gap electrical electrical discharge discharge simulation. simulation. 2.1.2.2.1.2. Computation Computation Simulation Simulation of E oflectrical Electrical Discharge Discharge Success Success Rate Rate 2.1.2.2.1.2. Computation Computation S Simulationimulation of of E Electricallectrical D Dischargeischarge S Successuccess R Rateate FigureFigure 6 6shows shows the the counter counter pulse signalsignal VVPWMPWM outputoutput by the computercomputer (with (with a a frequency frequency of of 1 MHz 1 MHzand Figure and aFigure duty a duty6 cycle 6shows shows cycle of 0.5).the the of counter Thiscounter 0.5). signal This pulse pulse was signal signal signal generated was V VPWM PWM generated output by output the RT-DAC4by by by the the the computer computer RT/PCI-DAC4/PCI interface (with (with carda interfaceafrequency frequency controlled card of of 1 by 1 controlledMHzMHzthe software and and by a athe duty VisSim duty software cycle cycle at the of VisSim of computer 0.5). 0.5). at This This the end. signal computer signal This was signal was end. generated generated constitutes This signal by by the the constitutes the time RT RT base-DAC4/PCI-DAC4/PCI ofthe the time hardware interface interfacebase of circuits. the card card hardwarecontrolledcontrolledFigure7 circuits.shows by by the the the Figure software software counter 7 shows VisSim waveformVisSim the at counterat the signalthe computer computer waveform N SUC ofend. end. successful signal This This N signal SUCsignal electrical of successfulconstitutes constitutes discharge. electrical the the time Thistime discharge.base waveform base of of the the is Thishardwarehardwarethe waveform result circuits. circuits. of theis the computation Figure Figure result 7 7ofshows shows the of computation the the counter signals counter forwaveform waveformof Vthesuc signals(Figure signal signal 5for) N and NVSUCSUCsuc Vof (Figureof PWMsuccessful successful(Figure 5) and electrical electrical6 )V passingPWM (Figure discharge. discharge. through 6) passingThisThisthe waveform AND waveform through gate. theis Usingis the theAND result theresult gate. signal of of theUsing the N computationSUC computation the, the signal VisSim Nof ofSUC softwarethe the, the signals signals VisSim at for the for softwareV computer Vsucsuc (Figure (Figure at end the 5) 5) canandcomputer and countV VPWMPWM the (Figure end(Figure number can 6) 6) countpassingpassingof accumulated the throughnumber through theof successful the accumulated AND AND gate. electricalgate. Using successfulUsing discharges. the the signal signalelectrical N NSUCSUC discharges., ,the the VisSim VisSim software software at at the the computer computer end end can can countcount the the number number of of accumulated accumulated successful successful electrical electrical discharges. discharges.

FigureFigureFigure 6. 6. Quantified6. Quantified QuantifiedQuantiﬁed signals. signals. signals. signals.

NSUCNSUCSUC N TT1T1 1 TT2T2 2 TT3T3 3

T TTonTononTToffoffoff

ttt 1 1period1 period period

Figure 7. Counter waveform of discharge success rate. FigureFigureFigure 7. 7. Counter7. Counter Counter waveform waveform waveform of of dischargeof discharge discharge success success success rate. rate. rate.

2.1.3.2.1.3.2.1.3. Computation Computation Computation S imulationS Simulationimulation of of ofS hortS Shorthort-C-Circuit-Circuitircuit R Rate Rateate

FigureFigureFigure 8a 8a 8ais is theis the the counter counter counter waveform waveform waveform of of ofsignal signal signal N N gap&shortNgap&shortgap&short. This. .This This signal signal signal is is theis the the computation computation computation result result result for for for thethethe signals signals signals I gap,sucI gap,sucIgap,suc (Figure (Figure (Figure 4b 4b 4band and and V VPWM VPWMPWM (Figure (Figure (Figure 7) 7) 7)passing passing passing through through through the the the AND AND AND gate. gate. gate. According According According to to tothe the the signalsignalsignal for for for N NSUC NSUCSUC, the, the, the VisSim VisSim VisSim software software software at at theat the the computer computer computer end end end can can can count count count the the the total total total numbers numbers numbers of of gapof gap gap electrical electrical electrical dischargedischargedischarge and and and short short short-circuit-circuit-circuit instances. instances. instances. Figure Figure Figure 8b 8b 8b shows shows shows the the the signal signal signal for for for N Nshort Nshortshort, , which , which which indicates indicates indicates the the the numbernumbernumber of of shortof short short circuits circuits circuits counted. counted. counted. This This This signal signal signal is is theis the the difference difference difference between between between the the the waveforms waveforms waveforms of of Nof NSUC NSUCSUC (Figure (Figure (Figure 7)7) 7) and and and N N gap&short Ngap&shortgap&short (Figure (Figure (Figure 8a). 8a). 8a). According According According to to to the the the aforementioned aforementioned aforementioned logic logic logic judgment judgment judgment simulation simulation simulation and and and the the the conditionconditioncondition judgment judgment judgment mechanism, mechanism, mechanism, a a set a set set of of of optimized optimized optimized logic logic logic judgment judgment judgment circuits circuits circuits was was was designed. designed. designed. The The The functionsfunctionsfunctions of of ofthis this this circuit circuit circuit set set set encompass encompass encompass computation computation computation of of ofthe the the electrical electrical electrical discharge discharge discharge success success success rate, rate, rate, the the the short short short- - - circuitcircuitcircuit rate, rate, rate, and and and the the the open open open-circuit-circuit-circuit rate. rate. rate.

Nanomaterials 2020, 10, 1091 6 of 12

2.1.3. Computation Simulation of Short-Circuit Rate

Figure8a is the counter waveform of signal N gap&short. This signal is the computation result for the signals Igap,suc (Figure4b and V PWM (Figure7) passing through the AND gate. According to the signal for NSUC, the VisSim software at the computer end can count the total numbers of gap electrical discharge and short-circuit instances. Figure8b shows the signal for N short, which indicates the number of short circuits counted. This signal is the diﬀerence between the waveforms of NSUC (Figure7) and Ngap&short (Figure8a). According to the aforementioned logic judgment simulation and the condition judgment mechanism, a set of optimized logic judgment circuits was designed. The functions of this circuitNanomaterials set encompass 2020, 10, x FO computationR PEER REVIEW of the electrical discharge success rate, the short-circuit rate,6 andof 12 the open-circuit rate.

(a) (b)

FigureFigure 8.8.Counter Counter waveformswaveforms ofof ((aa)) signalsignal I Igapgap and ( b) short circuits.

2.2.2.2. Electrical Discharge ProcessProcess OptimizationOptimization ThisThis study study primarily primarily explored explored the the application application of of PID PID closed-circuit closed-circuit controllers controllers in in EDM EDM for for controllingcontrolling electricalelectrical dischargedischarge gapgap voltage.voltage. The key objectiveobjective isis toto determinedetermine thethe short-circuitshort-circuit raterate andand toto useuse thethe Ziegler–NicholsZiegler–Nichols methodmethod PIDPID toto eeffectivelyffectively reducereduce thethe short-circuitshort-circuit raterate [[2222––2424].]. InIn thisthis study,study, an an Ag Ag wire wire of of 99.9% 99.9% purity purity with with 1 mm1 mm and and 2 mm2 mm diameter diameter was was used used as theas the metallic metallic material material for thefor anode the anode and cathode and cathode electrode, electrode, respectively. respectively. The dielectric The dielectric fluid was fluid low-conductivity was low-conductivity deionized µ water.deionized The water electrical. The discharge electrical voltage discharge was voltage 100 V, and was T on100and V, Tandoff were Ton and set atToff 10 weres. Theset at process 10 µs. time The wasprocess 120 s time at ambient was 12 temperature0 s at ambient (25 ◦C)temperature and atmospheric (25 °C pressure) and atmospheric (1atm). The pressure preparation (1atm capacity). The waspreparation 150 mL. Table capacity1 shows was the 150 control mL. Table parameters 1 shows of the the Ziegler–Nichols control parameters method. of the Table Ziegler2 shows–Nichols the processmethod. parameters. Table 2 shows PID the parameters process parameters. were fine-tuned PID parameters to reduce the were short-circuit fine-tuned rate, to reduce and K uthewas short the- criticalcircuit rate, gain and of the Ku PID was controller.the critical Thegain micro-EDM of the PID controller. motor does The not micro function-EDM when motorK udoes< 1.25 not, whereasfunction thewhen periodic Ku < 1.25 waveforms, whereas are the unstable periodic andwaveforms nonsine are waves unstable when and Ku >nonsine2.5, resulting waves when in divergence. Ku > 2.5, Toresulting avoid said in divergence. problems, KTou wasavoid set said to between problems, 1.25 K andu was 2.5. set On-machineto between 1.25 fine-tuning and 2.5. wasOn-machine used, and fine the- short-circuittuning was used, rate was andrecorded the short for-circuit selecting rate was optimal recorded PID parametersfor selecting for optimal the micro-EDM. PID parameters Professional for the instrumentsmicro-EDM. wereProfessional used to instruments analyze the absorbance,were used to particle analyze size, the andabsorbance, zeta potential particle of size, the nano-Agand zeta colloid.potential The of relationshipthe nano-Ag betweencolloid. The the characteristicsrelationship between of the colloid the characteristics and the short-circuit of the colloid rate was and also the examined.short-circu Figureit rate was9 shows also theexamined. fine-tuning Figure process 9 shows for PIDthe fine parameters.-tuning process for PID parameters.

TableTable 1.1. ControlControl parametersparameters ofof thethe Ziegler–NicholsZiegler–Nichols method.method.

Kp TiKp Td Ti Td

ZieglerZiegler–Nichols–Nichols methodmethod 0.6 Ku 0.6Tu/ 2× Ku Tu/8Tu/2 Tu/8 × Ku = critical gain, Tu = period of oscillation, K = Kp/T ,K = Kp T Ku = critical gain, Tu = period of oscillation,i Ki = Kpi/Ti,d Kd = K×p ×d Td

Table 2. Process parameters.

Title parameters Title parameters Ambient temperature 25 °C Atmospheric pressure 1 atm Electrical Discharge voltage 100 V Electrical discharge current 4 A Ton-Toff Duty cycle Electrode material (purity) Ag (99.99%) 10-10 (μs) Process time 120 sec Dielectric fluid Deionized water Anode 1 mm Electrode diameter Preparation capacity 150 mL Cathode 2 mm

Nanomaterials 2020, 10, 1091 7 of 12

Table 2. Process parameters.

Title Parameters Title Parameters

Ambient temperature 25 ◦C Atmospheric pressure 1 atm Electrical Discharge voltage 100 V Electrical discharge current 4 A T -T Duty cycle on oﬀ Electrode material (purity) Ag (99.99%) 10-10 (µs) Process time 120 s Dielectric ﬂuid Deionized water Anode 1 mm Electrode diameter Preparation capacity 150 mL Cathode 2 mm Nanomaterials 2020, 10, x FOR PEER REVIEW 7 of 12

FigureFigure 9. 9. PProportional–integral–derivativeroportional–integral–derivative ( (PID)PID) parameter parameter fine fine-tuning-tuning flow flow chart chart.. 3. Results 3. Results 3.1. PID Fine-Tuning Optimization 3.1. PID Fine-Tuning Optimization This study used the Ziegler–Nichols method for fine-tuning PID parameters to reduce the short-circuitThis study rate. used One the advantage Ziegler–Nichols of the method method for in fine question-tuning is PID that parameters proper tuning to reduce can be the achieved short- circuit rate. One advantage of the method in question is that proper tuning can be achieved according according to changes in the parameters. The PID fine-tuning process parameter Ku was between 1.25 toand changes 2.5, and in tuning the parameters. was performed The inPID an fine interval-tuning of 0.25. process After parameter tuning, the K short-circuitu was between rate, 1.25 absorbance, and 2.5, andparticle tuning size, was zeta performed potential, in and an ratio interval between of 0.25. the After short-circuit tuning, rate the and short the-circuit discharge rate, absorbance, success rate particlewere analyzed size, zeta and potential, compared and using ratio professional between the instruments short-circuit to rate determine and the thedischarge best parameters success rate for were analyzed and compared using professional instruments to determine the best parameters for micro-EDM. Figure 10 shows the curve describing the relationship between Ku and nanoparticle micro-EDM. Figure 10 shows the curve describing the relationship between Ku and nanoparticle characteristics. Figure 10a shows the curve describing the relationship between Ku and the short-circuit characteristics. Figure 10a shows the curve describing the relationship between Ku and the short- rate. Figure 10b shows the curve describing the relationship between Ku and the ratio between the circuitshort-circuit rate. Figure rate and 10b the shows discharge the curve success describing rate. Figure the relationship10c shows the between curve describing Ku and the the ratio relationship between the short-circuit rate and the discharge success rate. Figure 10c shows the curve describing the between Ku and absorbance. Figure 10d shows the curve describing the relationship between Ku and relationship between Ku and absorbance. Figure 10d shows the curve describing the relationship particle size. Figure 10e shows the curve depicting the relationship between Ku and zeta potential. between Ku and particle size. Figure 10 (e) shows the curve depicting the relationship between Ku and zeta potential. The curves demonstrate that when Ku was 1.6, the lowest short-circuit rate (1.77%) was obtained using PID parameters (i.e., Kp = 0.96; Ki = 5.760576; Kd = 0.039996), with the ratio between the short-circuit rate and the discharge success rate being 0.053. Moreover, the nano-Ag colloid had an absorbance of 0.26, a zeta potential of −46.8 mV, and a particle size of 3.41 nm. The values of these nanoparticle characteristics were optimal. The data demonstrated that all nanoparticle-related characteristics were optimal when the short-circuit rate was the lowest. Therefore, effectively reducing the short-circuit rate enhances the overall electrical discharge effect of micro-EDM.

Nanomaterials 2020, 10, 1091 8 of 12

The curves demonstrate that when Ku was 1.6, the lowest short-circuit rate (1.77%) was obtained using PID parameters (i.e., Kp = 0.96; Ki = 5.760576; Kd = 0.039996), with the ratio between the short-circuit rate and the discharge success rate being 0.053. Moreover, the nano-Ag colloid had an absorbance of 0.26, a zeta potential of 46.8 mV, and a particle size of 3.41 nm. The values of these nanoparticle − characteristics were optimal. The data demonstrated that all nanoparticle-related characteristics were optimal when the short-circuit rate was the lowest. Therefore, eﬀectively reducing the short-circuit rateNanomaterials enhances 2020 the, 10 overall, x FOR PEER electrical REVIEW discharge eﬀect of micro-EDM. 8 of 12

(a) (b)

(e)

FigureFigure 10.10. CurvesCurves describing describing the relationship the relationship between between Ku and K nanoparticleu and nanoparticle characteristics: characteristics: (a) short- (acircuit) short-circuit rate, (b) ratio rate, between (b) ratio betweenshort-circuit short-circuit rate and discharge rate and dischargesuccess rate, success (c) absorbance, rate, (c) absorbance, (d) particle (dsize,) particle and (e size,) zeta and potential. (e) zeta potential.

3.2. Nano-Ag Colloid Characteristics Analysis Applying micro-EDM, this study prepared a nano-Ag colloid with a minimum short-circuit rate using optimal PID parameters (Kp = 0.96, Ki = 5.760576, Kd = 0.039996, and therefore Ku = 1.6). By analyzing the colloid with professional instruments, this study demonstrated that the specimen is nanosized. Details are presented in the following.

Nanomaterials 2020, 10, 1091 9 of 12

3.2.1. UV-Vis and Zeta Potential Analysis The nano-Ag colloid was analyzed with a spectrophotometer (UV-Vis), which demonstrates the relationship between absorbance and wavelength. Various studies have deﬁned the characteristic wavelength of nano-Ag colloids to be between 380 and 410 nm (Figure 11a). The analysis here yielded Nanomaterialsa characteristic 2020, 10 wavelength, x FOR PEER REVIEW of 390 nm for the nano-Ag colloid prepared using ESDM, a nanosized9 of 12 characteristic. The absorbance of the nano-Ag colloid was 0.26, and the value of absorbance of the colloidcolloid was was mostly mostly proportional proportional to to the the concentration concentration of of the the colloid. colloid. In In this this study, study, the the zeta zeta potential potential of of nanoparticlesnanoparticles was was measured measured with with a aZetasizer. Zetasizer. When When the the absolute absolute value value of of the the potential potential was was greater greater thanthan 30 30 mV, mV, the the metal metal particles particles in in the the colloid colloid exhibited exhibited high high suspension suspension stability. stability. The The analysis analysis result result forfor zeta zeta potenti potentialal is shown inin FigureFigure 11 11b.b. The The average average value value for for zeta zeta potential potential measured measured was was46.8 −46.8 mV, − mV,indicating indicating that that the nanocolloidthe nanocolloid prepared prepared here here exhibited exhibited desirable desirable suspension suspension stability. stability.

(a) (b)

FigureFigure 11. 11. NanoNano-Ag-Ag colloid colloid analysis analysis result resultss for for (a ()a )UV UV-Vis-Vis and and (b (b) )zeta zeta potential potential..

3.2.2.3.2.2. Transmission Transmission Electron Electron Microscopy Microscopy and and Energy Energy-Dispersive-Dispersive X X-ray-ray Spectroscopy Spectroscopy ForFor thethe nano-Agnano-Ag colloid colloid prepared prepared using ESDM, using nano-Ag ESDM, particles nano-Ag were particles observed were microscopically observed microscopicallywith a transmission with electron a transmission microscope. electron Figure microscope. 12a shows Figure a nano-Ag 12a shows colloidal a nano structure-Ag colloidal at a scale structureof 200 nm. at a Figure scale of12 200b is nm. the enlargedFigure 12 imageb is the (at enlarged a scale image of 20 nm)(at a of scale the imageof 20 nm) inside of the the image box in insideFigure the 12 a.box Figure in Figure 12c is 12 thea. enlargedFigure 12c image is the (at enlarged a scale of image 5 nm) (at of thea scale image of inside5 nm) theof the box image in Figure inside 12b. theThe box spacing in Figure between 12b. lattice The spacing lines (d-spacing) between was lattice 0.219 lines nm. (d Figure-spacing) 12d was is obtained 0.219 nm. from Figure the results 12d is of obtainedTEM (Figure from 12 theb), which results shows of TEM the size(Figure distribution 12b), which of nano-Ag shows participles. the size distribution According of to the nano ﬁgure,-Ag particip0–5 nmles. nano-Ag According particles to the accounted figure, 0for–5 24%nm ofnano all- nano-AgAg particles particles, accounted 6–10 nm for for 24% 34%, of all11–15 nan nmo-Ag for particles,24%, 16–20 6–10 nm nm for for 16%, 34%, and 11 21–15 nm nm or for more 24%, for 16 2%.–20 Thisnm for ﬁnding 16%, showedand 21 nm that or most more of for the 2%. nano-Ag This findingparticles showed had sizes that of most 6–10 of nm. the Energy-dispersivenano-Ag particles X-rayhad sizes spectroscopy of 6–10 nm. (EDS) Energy is a- typedispersive of analytical X-ray spectroscopytechnology which (EDS) uses is a thetype characteristic of analytical X-rays technology of electron which beams uses the to analyze characteristic the chemical X-rays compositionof electron beamsof specimens to analyze when the each chemical element composition has distinct ofspectral specimens characteristics. when each element Figure 13 has shows distinct the spectral graph of characteristics.EDS analysis ofFigure the nano-Ag 13 shows colloid. the graph The of resultEDS analysis indicates of thatthe nano the nano-Ag-Ag colloid. colloid The result prepared indicates using thatESDM the containednano-Ag colloid only two prepared elements: using oxygen ESDM (O) contained and silver only (Ag). two elements: oxygen (O) and silver (Ag).

Nanomaterials 2020, 10, x FOR PEER REVIEW 9 of 12

colloid was mostly proportional to the concentration of the colloid. In this study, the zeta potential of nanoparticles was measured with a Zetasizer. When the absolute value of the potential was greater than 30 mV, the metal particles in the colloid exhibited high suspension stability. The analysis result for zeta potential is shown in Figure 11b. The average value for zeta potential measured was −46.8 mV, indicating that the nanocolloid prepared here exhibited desirable suspension stability.

(a) (b)

Figure 11. Nano-Ag colloid analysis results for (a) UV-Vis and (b) zeta potential.

3.2.2. Transmission Electron Microscopy and Energy-Dispersive X-ray Spectroscopy For the nano-Ag colloid prepared using ESDM, nano-Ag particles were observed microscopically with a transmission electron microscope. Figure 12a shows a nano-Ag colloidal structure at a scale of 200 nm. Figure 12b is the enlarged image (at a scale of 20 nm) of the image inside the box in Figure 12a. Figure 12c is the enlarged image (at a scale of 5 nm) of the image inside the box in Figure 12b. The spacing between lattice lines (d-spacing) was 0.219 nm. Figure 12d is obtained from the results of TEM (Figure 12b), which shows the size distribution of nano-Ag participles. According to the figure, 0–5 nm nano-Ag particles accounted for 24% of all nano-Ag particles, 6–10 nm for 34%, 11–15 nm for 24%, 16–20 nm for 16%, and 21 nm or more for 2%. This finding showed that most of the nano-Ag particles had sizes of 6–10 nm. Energy-dispersive X-ray spectroscopy (EDS) is a type of analytical technology which uses the characteristic X-rays of electron beams to analyze the chemical composition of specimens when each element has distinct spectral characteristics. Figure 13 shows the graph of EDS analysis of the nano-Ag colloid. The result indicates

Nanomaterialsthat the nano2020-, Ag10, 1091colloid prepared using ESDM contained only two elements: oxygen (O) and10 silver of 12 (Ag).

Nanomaterials 2020, 10, x FOR PEER REVIEW 10 of 12 Nanomaterials 2020, 10, x FOR PEER REVIEW 10 of 12 (a) (b) (a) (b)

0.0219.219nmnm

(c)( c) (d()d )

FigureFigure 12. 12. NanoNano-Ag Nano-Ag-Ag TEM TEM images images: images: ( aa: ))( a200 200) 200 nm nm; nm; ( (b; )() b20 20) 20 nm nm; nm; ( (c; )( c5 5) nm nm;5 nm; and ; and ( d )( dsize ) size distribution distribution. distribution. .

FigureFigure 13. 13. NanoNano-Ag Nano-Ag-Ag particle particle e energy-dispersivenergy energy-disp-dispersiveersive X X-ray- Xray-ray spectroscopy spectroscopy (EDS)(EDS (EDS) analysis.analysis) analysis. . 4. Conclusions 4. 4.Conclusions Conclusions Using real-time micro-EDM monitoring, this study designed a logic judgment circuit set to UsingUsing real real-time-time micro micro-EDM-EDM monitoring, monitoring, this this study study designed designed a alogic logic judgment judgment circuit circuit set set to to determine all EDM conditions, including gap electrical discharge, open circuits, and short circuits of determinedetermine all all EDM EDM conditions, conditions, including including gap gap electrical electrical discharge, discharge, open open circuits, circuits, and and short short circuits circuits of of the EDM equipment. Signals were sent to the software end to determine the probability for each of the thethe EDM EDM equipment. equipment. Signals Signals were were sent sent to tothe the software software end end to todetermin determine thee the probability probability for for each each of of aforementioned conditions. The result demonstrated that short circuits may adversely aﬀect equipment, thethe aforementioned aforementioned conditions. conditions. The The result result demonstrated demonstrated that that short short circuits circuits may may adversely adversely affect affect equipment,equipment, and and reducing reducing the the short short-circuit-circuit rate rate by by PID PID tuning tuning therefore therefore improves improves the the overall overall ele electricalctrical dischargedischarge effect. effect. Last Lastly,ly the, the results results of ofan an analysis analysis using using high high-precision-precision instruments, instruments, including including UV UV-Vis-Vis andand the the Zetasizer, Zetasizer, indicated indicated the the existence existence of ofnano nano-Ag-Ag particles particles in inthe the deionized deionized water. water. Moreover, Moreover, whenwhen the the short short-circuit-circuit rate rate was was the the lowest, lowest, all all nanoparticle nanoparticle characteristics characteristics were were most most optimal. optimal. This This studystudy makes makes the the following following contributions: contributions: 1. 1. InIn this this study, study, a anano nano-Ag-Ag colloid colloid was was prepared prepared using using the the electric electric spark spark discharge discharge method method (ESDM)(ESDM) with with electrode electrode material material (Ag (Ag with with 99.9% 99.9% purity) purity) in in150 150 mL mL of ofdei deionizedonized water. water. The The diameterdiameter of ofthe the anode anode electrode electrode was was 1 mm,1 mm, and and the the diameter diameter of ofthe the cathode cathode electrode electrode was was 2 2 mm.mm. With With the the electrical electrical discharge discharge voltage voltage = 100 = 100 V, V, duty duty cycle cycle (Ton (Ton-Toff)-Toff) = 10= 10-10-10 μ s,μ s, and and a a processprocess time time of of120 120 s, silvers,silver nanoparticles nanoparticles with with an an absorb absorbanceance of of0.26 0.26 could could be be prepared prepared. . 2. 2. TheThe study study used used self self-developed-developed micro micro-EDM-EDM and and ESDM ESDM to to prepare prepare a anano nano-Ag-Ag colloid. colloid. This This methodmethod requires requires no no additional additional surfactant surfactant or or other other chemical chemical materials materials and and can can be be used used at atthe the ambientambient temperature temperature and and pressure. pressure. Moreover, Moreover, this this novel novel physical physical method method for for preparing preparing nano nano- - AgAg colloid colloids resultss results in inno no chemical chemical pollution. pollution. 3. 3. WhenWhen using using ESDM ESDM to to prepare prepare a anano nano-Ag-Ag colloid, colloid, the the electrical electrical discharge discharge conditions conditions were were observedobserved to to comprise comprise gap gap electrical electrical discharge, discharge, short short circuits, circuits, and and open open circuits. circuits. The The short short- - circuitcircuit phenomenon phenomenon was was explored explored in indepth depth in inthis this study, study, and and a selfa self-developed-developed logic logic judgment judgment circuit set was applied to identify short circuits. Signals were then sent to computer software

Nanomaterials 2020, 10, 1091 11 of 12 and reducing the short-circuit rate by PID tuning therefore improves the overall electrical discharge effect. Lastly, the results of an analysis using high-precision instruments, including UV-Vis and the Zetasizer, indicated the existence of nano-Ag particles in the deionized water. Moreover, when the short-circuit rate was the lowest, all nanoparticle characteristics were most optimal. This study makes the following contributions: 1. In this study, a nano-Ag colloid was prepared using the electric spark discharge method (ESDM) with electrode material (Ag with 99.9% purity) in 150 mL of deionized water. The diameter of the anode electrode was 1 mm, and the diameter of the cathode electrode was 2 mm. With the electrical discharge voltage = 100 V, duty cycle (Ton-Toff) = 10-10 µs, and a process time of 120 s, silver nanoparticles with an absorbance of 0.26 could be prepared. 2. The study used self-developed micro-EDM and ESDM to prepare a nano-Ag colloid. This method requires no additional surfactant or other chemical materials and can be used at the ambient temperature and pressure. Moreover, this novel physical method for preparing nano-Ag colloids results in no chemical pollution. 3. When using ESDM to prepare a nano-Ag colloid, the electrical discharge conditions were observed to comprise gap electrical discharge, short circuits, and open circuits. The short-circuit phenomenon was explored in depth in this study, and a self-developed logic judgment circuit set was applied to identify short circuits. Signals were then sent to computer software for computation of the rate. Short circuits may have exerted an adverse effect on the equipment.

4. This study showed that PID parameters such that Kp was 0.96, Ki was 5.760576, and Kd was 0.039996 (and, as a result, Ku was 1.6) produced optimal values for absorbance (0.26), surface plasmon resonance (390 nm), zeta potential ( 46.8 mV), particle size (3.41 nm), short-circuit rate − (1.77%), and the ratio between the short-circuit rate and the discharge success rate (0.053). Therefore, this study demonstrated that the lower the short-circuit rate is, the more the nanocharacteristics are optimized.

Author Contributions: Methodology, Y.-S.L. and D.-C.T.; validation, Y.-S.L., and Y.-C.L.; formal analysis, Y.-C.L. and D.-C.T.; resources, K.-H.T.; data curation, Y.-C.L. and D.-C.T.; writing—original draft preparation, Y.-S.L. and Y.-C.L.; writing—review and editing, L.S.; supervision, K.-H.T. and L.S.; project administration, K.-H.T.; funding acquisition, K.-H.T. All authors have read and agreed to the published version of the manuscript. Funding: This research was funded by the Ministry of Science and Technology, grant number MOST 108-2221-E-027-050-. Acknowledgments: The authors would like to thank the Precision Research and Analysis Center, National Taipei University of Technology for technical support for this research. Conﬂicts of Interest: The authors declare no conﬂict of interest.

References

1. Roll, R. A Possible Explanation of the Small Firm Effect. J. Financ. 1981, 36, 879–888. [CrossRef] 2. Bødker, F.; Mørup, S.; Linderoth, S. Surface effects in metallic iron nanoparticles. Phys. Rev. Lett. 1994, 72, 282. [CrossRef][PubMed] 3. Chamarro, M.; Gourdon, C.; Lavallard, P.; Lublinskaya, O.; Ekimov, A.I. Enhancement of electron-hole exchange interaction in CdSe nanocrystals: A quantum confinement effect. Phys. Rev. B 1996, 53, 1336. [CrossRef][PubMed] 4. Koch, C.C. Synthesis of nanostructured materials by mechanical milling: Problems and opportunities. Nanostruct. Mater. 1997, 9, 13–22. [CrossRef] 5. Yuvaraj, D.; Kumar, R.R.; Selvan, V.T.; Sathyanarayanan, M.; Rao, K.N. Growth of ZnSe nano and microstructures at high vacuum by thermal evaporation. Appl. Nanosci. 2014, 4, 469–475. [CrossRef] 6. Lo, C.H.; Tsung, T.T.; Lin, H.M. Preparation of silver nanofluid by the submerged arc nanoparticle synthesis system (SANSS). J. Alloys Compd. 2007, 434, 659–662. [CrossRef] 7. Rajurkar, K.P.; Levy, G.; Malshe, A.; Sundaram, M.M.; McGeough, J.; Hu, X.; DeSilva, A. Micro and nano machining by electro-physical and chemical processes. CIRP Ann. 2006, 55, 643–666. [CrossRef] Nanomaterials 2020, 10, 1091 12 of 12

8. Tseng, K.H.; Chen, Y.C.; Shyue, J.J. Continuous Synthesis of Colloidal Silver Nanoparticles by Electrochemical Discharge in Aqueous Solutions. J. Nanopart. Res. 2011, 13, 1865–1872. [CrossRef] 9. Yeo, S.H.; Tan, P.C.; Kurnia, W. Eﬀects of powder additives suspended in dielectric on crater characteristics for micro electrical discharge machining. J. Micromech. Microeng. 2007, 17, N91. [CrossRef] 10. Lo, C.H.; Tsung, T.T.; Chen, L.C. Shape-controlled synthesis of Cu-based nanoﬂuid using submerged arc nanoparticle synthesis system (SANSS). J. Cryst. Growth 2005, 277, 636–642. [CrossRef] 11. Lung, J.K.; Huang, J.C.; Tien, D.C.; Liao, C.Y.; Tseng, K.H.; Tsung, T.T. Preparation of gold nanoparticles by arc discharge in water. J. Alloys Compd. 2007, 434–435, 655–658. [CrossRef]

12. Tseng, K.-H.; Chang, C.-Y.; Chung, M.-Y.; Cheng, T.-S. Fabricating TiO2 Nanocolloids by Electric Spark Discharge Method at Normal Temperature and Pressure. Nanotechnology 2017, 28, 1–9. [CrossRef][PubMed] 13. Liao, C.-Y.; Tseng, K.-H.; Lin, H.-S. Preparation of Metallic Aluminum Compound Particles by Submerged Arc Discharge Method in Aqueous Media. Metall. Mater. Trans. B 2013, 44, 91–97. [CrossRef] 14. Tseng, K.-H.; Ku, H.-C.; Tien, D.-C.; Stobinski, L. Parameter control and concentration analysis of graphene colloids prepared by electric spark discharge method. Nanotechnol. Rev. 2019, 8, 201–209. [CrossRef] 15. Tseng, K.-H.; Chung, M.-Y.; Chang, C.-Y. Parameters for Fabricating Nano-Au Colloids through the Electric Spark Discharge Method with Micro-Electrical Discharge Machining. Nanomaterials 2017, 7, 133. [CrossRef] 16. Song, K.C.; Lee, S.M.; Park, T.S.; Lee, B.S. Preparation of colloidal silver nanoparticles by chemical reduction method. Korean J. Chem. Eng. 2009, 26, 153–155. [CrossRef] 17. Clogston, J.D.; Patri, A.K. Zeta Potential Measurement. In Characterization of Nanoparticles Intended for Drug Delivery; Humana Press: Totowa, NJ, USA, 2011; pp. 63–70. 18. Browning, N.D.; Chisholm, M.F.; Pennycook, S.J. Atomic-resolution chemical analysis using a scanning transmission electron microscope. Nature 1993, 366, 143–146. [CrossRef] 19. Khanna, P.K.; Singh, N.; Kulkarni, D.; Deshmukh, S.; Charan, S.; Adhyapak, P.V. Water based simple synthesis of re-dispersible silver nano-particles. Mater. Lett. 2007, 61, 3366–3370. [CrossRef] 20. Tseng, K.-H.; Chang, C.-Y.; Chen, M.-J.; Tseng, Y.-K. Novel electrical discharge machining system with real-time control and monitoring for preparing nanoiron Colloid. Adv. Mech. Eng. 2018, 10, 1–10. [CrossRef] 21. Tseng, K.-H.; Kao, Y.-S.; Chang, C.-Y. Development and Implementation of a Micro-electric Discharge Machine: Real-Time Monitoring System of Fabrication of Nanosilver Colloid. J. Clust. Sci. 2016, 27, 763–773. [CrossRef] 22. Foley, M.W.; Julien, R.H.; Copeland, B.R. A comparison of PID controller tuning methods. Can. J. Chem. Eng. 2005, 83, 712–722. [CrossRef] 23. Ang, K.H.; Chong, G.; Li, Y. PID control system analysis, design, and technology. IEEE Trans. Control Syst. Technol. 2005, 13, 559–576. 24. Darwish, N.M. Design of robust PID controllers for ﬁrst-order plus time delay systems based on frequency domain speciﬁcations. J. Eng. Sci. Assiut Univ. Fac. Eng. 2015, 43, 472–489.

© 2020 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/).