A comprehensive study of hydrogen production from ammonia borane via PdCoAg/AC nanoparticles and anodic current in alkaline medium: experimental design with response surface methodology

Abstract

In this paper, the optimization of hydrogen (H₂) production by ammonia borane (NH₃BH₃) over PdCoAg/AC was investigated using the response surface methodology. Besides, the electro-oxidation of NH₃BH₃ was determined and optimized using the same method to measure its potential use in the direct ammonium boran fuel cells. Moreover, the ternary alloyed catalyst was synthesized using the chemical reduction method. The synergistic effect between Pd, Co and Ag plays an important role in enhancement of NH₃BH₃ hydrolysis. In addition, the support effect could also efficiently improve the catalytic performance. Furthermore, the effects of NH₃BH₃ concentration (0.1–50 mmol/5 mL), catalyst amount (1–30 mg) and temperature (20°C–50°C) on the rate of H₂ production and the effects of temperature (20°C–50°C), NH₃BH₃ concentration (0.05–1 mol/L) and catalyst amount (0.5–5 µL) on the electro-oxidation reaction of NH₃BH₃ were investigated using the central composite design experimental design. The implementation of the response surface methodology resulted in the formulation of four models out of which the quadratic model was adjudged to efficiently appropriate the experimental data. A further statistical analysis of the quadratic model demonstrated the significance of the model with a p-value far less than 0.05 for each model and coefficient of determination (R²) of 0.85 and 0.95 for H₂ production rate and NH₃BH₃ electrroxidation peak current, respectively.

Keywords

ammonia borane / hydrogen production / fuel cell / response surface methodology

Introduction

Fossil fuels such as petroleum and coal, which are used as main energy sources, are harmful to the environment. Thus, in recent years novel renewable and environmentally friendly energy sources have been investigated [¹]. Fossil fuels harm the environment as they release greenhouse gases such as dust, ashes, CO_x, SO_x, and NO_x during combustion reactions. An increase in the emission of greenhouse gases worldwide leads to a rise in atmospheric temperature and average temperature on Earth, which, in turn, causes global warming that has negative impacts on the lives of humans and nature as a result of drought, flooding, and severe storms created by climate changes. Therefore, the necessity of finding new energy sources to be used instead of fossil fuels has become crucial. These energy sources must be available to be used in transportation vehicles, highly efficient, safe, economical, and most importantly, not contaminating the environment. Researchers have been working in many areas such as wind energy, thermonuclear energy, solar energy, geothermal energy, ocean currents, waves, lunar tide and hydrogen (H₂) energy to develop energy technologies that meet these specified characteristics. In particular, H₂ energy is considered to be an alternative source of energy to petrol, which is a fossil fuel, and therefore related studies on H₂ energy have recently increased [²,³].

H₂ draws attention as an energy carrier with its properties among secondary energy sources. The main advantages of H₂ are [⁴,⁵] that its energy conversion efficiency is high; it can be produced with the water splitting technique, which does not cause the release of harmful gasses, the main reason for greenhouse effect; it is plentiful in nature; it can be stored in different forms such as gas or liquid; its transportation is easier even over long distances; it can readily be transformed into other energy structures, and its high and low heating values are higher in proportion to most fossil fuels.

However, H₂ is not readily available in nature. It must be manufactured from the main energy sources and then be used in cells or internal combustion engines. Water is released as a by-product of hydrogen combustion. Unfortunately, hydrogen generation techniques are not as developed as desired and are of high output costs and low efficiency [⁴–⁹]. Due to its high power density and environmental friendliness, great importance has been attached to the use of hydrogen energy in the fuel cell field [¹⁰].

The research conducted on this topic have mainly focused on fuel cell technology where hydrogen is used as fuel [¹¹]. Fuel cells are interesting alternative power sources, which are used to provide the necessary clean energy for transportation and personal electronic implementations where low system gravities and portability are significant [¹²]. These fuel cell sturn chemical energy directly into electricity. Due to their high efficiencies and low emissions, fuel cells have found a wide range of applications in many sectors [¹¹].

Direct liquid fuel cells (DLFCs) are a type of fuel cell that has been intensely studied recently. Hydrogen fuel is extremely combustible and poses various shipping and preservation related problems [¹³]. DLFCs seem to be an alternate kind of fuel cell due to their liquid forms and their smooth use of the fuel. In addition, liquid fuels usually have a high energy intensity of around 600 Wh/kg which is higher compared to lithium-ion polymers. This means that DLFCs can ensure longer talk times for mobile phones, extended lifespans for laptop computers, more power to meet purchaser requisition and appropriate faster refueling in contrast to rechargeable batteries, which need long durations of charging when discharged [¹⁴]. In DLFCs, liquid fuel is linked straight to the anode, at which an oxidation reaction takes shape, while the aerial gas or oxygen is sent to the cathode for the reduction process.

In addition, the alkaline medium allows the electro-oxidation of certain chemical compounds with chemical hydrogen storage fuels with a high energy density directly in DLFC systems: boron-based materials (BBMs) such as sodium borohydride or boranes (ammonia borane, dimethylamine borane, and hydrazine borane) are general examples [¹⁵]. In addition, research has shown that some metal borohydrides can be converted to metal borohydride ammonia (MBA) by reacting with ammonia. For instance, Zr(BH₄)₄-8NH₃ has a hydrogen capacity of 14.8 wt% and a low main dehydrogenation temperature of about 130°C. However, the disadvantage of the synthesis process with the necessity of temperatures below 20°C greatly hampers research and practical applications, likewise in LiBH₄ and Mg(BH₄)₄ [¹⁶].

Current research has suggested the use of liquid ammonia-borane as a fuel for DLFCs because of an energy density of 8.4 Wh/g. Besides, they are non-toxic, inexpensive, easily available, and chemically stable, both for living things and the environment [¹⁷]. Moreover, NH₃BH₃, which contains 19.6% of hydrogen, is easily soluble in water and quite consistent against the hydrolysis in the lack of catalysts, which allows it to become a good alternative fuel for liquid-fed fuel cells [¹⁸]. However, the irreversibility of hydrolysis and the high regeneration cost limit its large-scale applications. Simple and effective regeneration methods are being developed for BBMs [¹⁹–²¹].

NH₃BH₃ is electrochemically oxidized in an alkaline medium to make products environmentally safe. (Reaction 1) [²²].

(1){\mathrm{NH}}_{\mathrm{3}}{\mathrm{BH}}_{\mathrm{3}}+{\mathrm{6OH}}^{-}\mathrm{\hspace{0.17em}}\mathrm{\hspace{0.17em}}\mathrm{\hspace{0.17em}}\mathrm{\hspace{0.17em}}\mathrm{\hspace{0.17em}}\mathrm{\hspace{0.17em}}\mathrm{\hspace{0.17em}}{\mathrm{BO}}_{\mathrm{2}}^{-}+{\mathrm{NH}}_{\mathrm{4}}^{+}+{\mathrm{4H}}_{\mathrm{2}}\mathrm{O}+{\mathrm{6e}}^{-}

Hydrogen production from boron compounds, which are considered the hydrogen storage materials of the future, can be performed widespread through dehydrogenation. Due to the ease of use, heterogeneous catalysts are preferred in these reactions [²³].

In the presence of metal catalysts or solid acids, the hydrolysis of NH₃BH₃ occurs under pressure formed by the reaction at room temperature (Reaction 2) [²⁴].

(2){\mathrm{NH}}_{\mathrm{3}}{\mathrm{BH}}_{\mathrm{3}}+{\mathrm{2H}}_{\mathrm{2}}\mathrm{O}\mathrm{\hspace{0.17em}}\mathrm{\hspace{0.17em}}\mathrm{\hspace{0.17em}}\mathrm{\hspace{0.17em}}\mathrm{\hspace{0.17em}}\mathrm{\hspace{0.17em}}\mathrm{\hspace{0.17em}}\mathrm{\hspace{0.17em}}{\mathrm{NH}}_{\mathrm{4}}^{+}+{\mathrm{BO}}_{\mathrm{2}}^{-}+{\mathrm{3H}}_{\mathrm{2}}

The ability to produce hydrogen gas at any temperature and the controllable hydrogen production make them advantageous for industrial use. Researchers have obtained better results by using metal nanoparticles such as ruthenium (Ru), palladium (Pd), titanium (Ti), gold (Au), cobalt (Co), nickel (Ni), iron (Fe), or ,copper (Cu) as the catalyst in hydrogen production with NH₃BH₃ hydrolysis [²⁵–²⁹]. Catalysts, which can be prepared in polymeric particles and silica composites as support materials [³⁰–³²], are used mandatorily in the preparation of many industrial products. Efforts have been made to reduce the economic cost while increasing product quality and productivity by using different catalysts in industry or laboratory. Therefore, the studies conducted in this direction aim to form new catalyst production systems and usage areas [³³,³⁴].

The three-metal catalyst is a method of incorporating inexpensive catalysts with transition metals such as Cr, Mo, W, Cu, Mn, Zn, and Fe to increase activity. Even a very small amount of metals used as an additive blocks the shrinking of the surface area by preventing the catalytic particles from flocculence, thereby ensuring that a catalyst with high catalytic activity is achieved [³⁵].

Most DLFCs still involve the application of an expensive noble catalyst to facilitate cell reaction. For example, in DMFCs, Pt-Ru was shown as the best catalyst for the anode, while Pt was the best for the cathode. In this paper, Pd is selected as the NH₃BH₃ oxidation catalyst. Studies have shown that Pd is similar to Pt in terms of high activity and fast electrode kinetics, and it is also less expensive than Pt. There are several different methods for catalyst preparation such as chemical reduction, sol-gel and wet impregnation [³⁶]. Chemical reduction methods are the most suitable methods to synthesize transition metal nanoparticles of 1–10 nm with well-defined surface compositions and dimension control from transition metal salts in laboratory conditions [³⁷].

The optimization of the operating parameters in a process could both increase the performance of the process and decrease the cost of operation. Moreover, in order to optimize the parameters, an experimental method which is statistically and experimentally analyzed with a minimum number of experiments is also very important. The response surface methodology (RSM) is an example of statistical and mathematical techniques which improves and optimizes the various types of industrial processes. To examine only one parameter in traditional methods, all other parameters affecting the system are kept constant. However, RSM easily overcomes this disadvantage by evaluating both the individual and the interaction effects of the selected independent parameters at the same time. Furthermore, when compared to traditional procedures, RSM is a practical approach due to the fact that it requires fewer experiments, less time, and fewer chemicals [³⁸–⁴⁰].

Herein, the catalytic dehydrogenation was reported from NH₃BH₃ hydrolysis and electro oxidation. PdCoAg/AC was selected as the NH₃BH₃ dehydrogenation and oxidation catalyst in this paper thanks to its high catalytic activity and fast electrode kinetics at room temperature.

The effect of operating parameters such as the NH₃BH₃ initial substrate concentration (X₁, mmol),the catalyst amount (X₂, mg), and the reaction temperature (X₃, °C) of PdCoAg/AC nanocatalysts prepared by using the NaBH₄ reduction method on maximum hydrogen production rate was investigated using RSM. Then, its effect on NH₃BH₃ electro-oxidation was determined and optimized using the same method to measure its potential use in the direct ammonium boron fuel cell (DABFC). For this purpose, the effect of reaction temperature (A, °C), NH₃BH₃ initial substrate concentration (B, M), and catalyst amount (C, µL) on maximum electro-oxidation peak current was investigated. This paper aimed to provide more information with fewer trials. Experimental design based on statistical modeling can be a useful and time-saving tool for evaluating the effects and interactions of different important parameters and will be of great interest in experimental design in hydrogen production in recent years. For the first time, hydrogen production and electrooxidation experiments (performed for fuel cell application practice) have been combined with a statistical design and presented to the literature with very consistent results.

Experimental

Materials

Potassium tetrachloropalladate (II) (K₂PdCl₄99%) Alfa Aesar® and Borane-ammonia complex (NH₃BH₃, 97%) were purchased from Sigma-Aldrich®. Cobalt(II) chloride (CoCl_2.6H₂O) and silver nitrate (AgNO₃) were purchased from Sigma-Aldrich® and sodium borohydride (NaBH₄, 98%) was purchased from Acros Organics®. Ultrapure water was prepared using the Milli-Q water purification system. All glassware and teflon coated magnetic bars were cleaned with acetone, followed by copious rinsing with distilled water before being dried in an oven at 120°C.

Catalyst synthesis and applications

The PdCoAg/AC nanocomposites were synthesized as follows: 100.0 mg of the prepared AC was dispersed in a 5.0 mL aqueous solution, to which 4.0 mL of aqueous solution containing 30.7 mg of K₂PdCl₄, 14 mg of CoCl₂. 6H₂O, and 2 mg of AgNO₃ was added. After being stirred at room temperature for 3 h, a 1.0 mL aqueous solution of 34 mg NaBH₄ was added by dropping to the mixture mentioned above and the resulting solution was stirred for an hour under air at room temperature. After being copiously washed with ultrapure water, the mixture was then filtrated and dried in a vacuum oven at 85°C. The characterization for the PdCoAg/AC nanocomposites was performed previously by using analytical techniques, namely, XRD, SEM-EDX, TEM, and XPS [⁴¹]. Relatively dispersed Pd_0.6Co_0.2Ag_0.2 nanoparticles with a particle size of 6.2 nm on activated carbon were prepared and characterized. Furthermore, the elemental composition of the catalyst was determined by using an inductively coupled plasma optical emission spectrometry (ICP/OES). The ICP results were shown in to demonstrate that these catalysts were successfully prepared at the desired atom ratios (Pd_0.6Co_0.2Ag_0.2). Pd contents of Pd/AC, Co/AC and Ag/AC were determined as 3.94 wt%, 1.19 wt% and 1.17 wt% respectively by using ICP-OES.

The catalytic activity of the synthesized catalysts with respect to NH₃BH₃ hydrolysis and oxidation were investigated. The experimental set-up for hydrogen production by the hydrolysis of NH₃BH₃ is illustrated in Fig. 1.

For the experiments in which hydrogen production was performed by catalytic hydrolysis of NH₃BH₃, a heater-mixer plate, a water bath, a jacketed Schlenk tube, and a thermometer were used to control the ambient temperature in order to keep the reaction temperature constant in the reaction system. To measure the resulting hydrogen, by-products such as a 250 mL wash bath, ammonia gas, a 250 mL gas burette system and a 250 mL volumetric drop funnel were used to adjust the water content of the gas. The volume of hydrogen gas collected in the pipe was read at certain times and the activity of the catalyst was determined graphically. According to these graphs, the activation energy, the degree of reaction and the data of the reactions associated with the reaction kinetics were obtained.

The electrocatalytic activity of the synthesized catalysts with respect to borohydride oxidation was investigated by cyclic voltammetry (CV). A conventional three electrode system with CHI 660 E potentiostat was used for electrochemical analyses. Glassy carbon electrodes, Pt wire, and Ag/AgCl electrodes were used as the working, counter, and reference electrodes, respectively. CV experiments were conducted in 1M NaOH and voltammograms were recorded from –1.05–0.5 V.

Central composite design experimental design

Of the RSM options, central composite design (CCD) is the most popular program of second-order design effective for the experimental system, which does not include a large number of experiments and design points. First, the effect of three independent parameters, i.e., NH₃BH₃ concentration (X₁, mmol/5 mL), catalyst amount (X₂, mg), and temperature (X₃, °C) on H₂ production rate (mL/(g·min)) were investigated by using CCD in RSM. Secondly, the effects of temperature (A, °C), NH₃BH₃ concentration (B, M), and catalyst amount (C, mL) on electro-oxidation (Current, mA) were examined using the same method. Based on CCD, the number of experiments conducted by using 2^k + 2k+ 6= 20 (the number of independent parameters (k)) is attributed to 3 for both designs. Six replicates of the middle points in both design matrixes were performed to determine the experimental error. Three different levels (–1, 0, + 1) were utilized to analyze the selected independent parameters for both designs (Table 1). The second order polynomial equation explaining the experimental system is expressed as

(3){\widehat{y}}_n={\beta }_0+{\displaystyle \sum _{i=\mathrm{1}}^{\mathrm{3}}{\beta }_i}{x}_i+{\displaystyle \sum _{i=\mathrm{1}}^{\mathrm{3}}{\beta }_{ii}}{x}_i^{\mathrm{2}}+{\displaystyle \sum _{i=\mathrm{1}}^{\mathrm{3}}{\displaystyle \sum _{j=i+\mathrm{1}}^{\mathrm{3}}{\beta }_{ij}}}{x}_i{x}_j,

where ŷ_n is the predicted response; X_i represents the independent parameters; and b₀, b_i, b_ii, and b_ij are the intercept, linear, quadratic, and interaction coefficients, respectively [⁴²].

Results and discussion

Evaluation of experimental results based on CCD

The interactive effects of independent parameters selected for both H₂ production rate (mL/(g·min)) and NH₃BH₃ electro-oxidation (Current, mA) were performed by CCD based on RSM. The optimization of three independent parameters and the determination of the H₂ production rate and current were analyzed by using Design-Expert 7.0 software. The data obtained from CCD and the corresponding responses are listed in Table 2. The empirical relationship between the responses and the real (un-coded) parameters have been given by Eqs. (4) and 5, respectively.

(4)$$\begin{gathered} {\text{H}}_{\text{2}}\text{production}\text{rate}\text{(mL/(g}\cdot \text{min))}\text{(uncoded)}=-598.91890+1478.66157[{\text{NH}}_{\text{3}}{\text{BH}}_{\text{3}}\text{conc}\text{.}] \\ +207.70314[\text{catalyst}\text{amount]}+139.08038[\text{temperature}]+27.65873[{\text{NH}}_{\text{3}}{\text{BH}}_{\text{3}}\text{conc}\text{.][catalyst}\text{amount}] \\ +17.00071[{\text{NH}}_{\text{3}}{\text{BH}}_{\text{3}}\text{conc}\text{.][temperature}]-1.27666[\text{catalyst}\text{amount][temperature}] \\ -463.63515{[{\text{NH}}_{\text{3}}{\text{BH}}_{\text{3}}\text{conc}\text{.}]}^2-8.45605{[\text{catalyst}\text{amount}]}^2-0.23522{[\text{temperature}]}^2, \end{gathered}$$

(5)$$\begin{gathered} {\text{Current}}_{}{\text{(mA)}}_{}\text{(uncoded)}=-3.43737+0.20053\text{[temperature]}+6.89986[{\text{NH}}_{\text{3}}{\text{BH}}_{\text{3}}{}_{}\text{conc}\text{.]} \\ +0.47859[{\text{catalyst}}_{}\text{amount}]-0.041596[\text{temperature}][{\text{NH}}_{\text{3}}{\text{BH}}_{\text{3}}{}_{}\text{conc}\text{.]} \\ -0.011019[{\text{temperature][catalyst}}_{}\text{amount}]-0.040351[{\text{NH}}_{\text{3}}{\text{BH}}_{\text{3}}{}_{}\text{conc}{\text{.][catalyst}}_{}\text{amount}] \\ -2.49152E-003{[\text{temperature}]}^2-3.61481{[{\text{NH}}_{\text{3}}{\text{BH}}_{\text{3}}{}_{}\text{conc}\text{.}]}^2-0.010388{[{\text{catalyst}}_{}\text{amount}]}^2. \end{gathered}$$

The analysis of variance (ANOVA) with Design-Expert program was utilized to statistically evaluate the interaction effects between selected independent parameters and their corresponding responses. The ANOVA results of the quadratic models for H₂ production rate and current are tabulated in Table 3. The ANOVA test indicated that the values of determination coefficient (R²) were found to be 0.85 and 0.95 for H₂ production rate and current, respectively. These values suggest that the predicted values obtained by the model could fit the actual values of the responses. If the p-value is equal to or less than 0.05, it delivers better confidence. If the p-values of the quadratic models are lower than 0.05, it indicates that the quadratic model generated by CCD is statistically significant.

The comparisons of the predicted data versus the experimental data and the plots of the normal % probability to internally studentized residuals for H₂ production rate and current, separately, are presented in Figs. 2 and 3. As can be seen in Fig. 2, the actual data obtained experimentally are randomly distributed and substantially close to the predicted data, indicating that the predicted and actual data are in good precision. Besides, the linear trend insinuates the normal distribution of H₂ production rate and current. Moreover, the predicted values are slightly different from the actual values, leading to the conclusion that each quadratic model is satisfactory in estimating responses. The values of R² are 0.85 and 0.95 for H₂ production rate and current, respectively. In RSM studies, R² values in this interval (≥0.8) of the quadratic model are sufficient to model and optimize a system, leading to the conclusion that the quadratic models obtained are satisfactory in predicting the responses in this paper. Furthermore, in Fig. 3, the residuals for both responses are normally distributed in a straight line, leading to the conclusion that the variances are quite sufficient.

3D-plots were drawn to define the basic and interactive effects of the independent variables on H₂ production rate and current, respectively (Fig. 4). The 3D-plot in Fig. 4(a) demonstrates the catalyst amount and NH₃BH₃ concentration effect on H₂ production rate for the hydrolysis of NH₃BH₃. In contrast with the NH₃BH₃ concentration variability (0.1 and 5.0 mmol/5 mL) and amount of catalyst (1–30 mg) keeping the reaction temperature at 35°C, the H₂ production rate increases to its peak with the increasing catalyst amount or NH₃BH₃ concentration, but then decreases with a further increase in catalyst amount or NH₃BH₃. It is observed that the catalyst amount changes a lot with the increase of NH₃BH₃ concentration,which determines the highest H₂ production rate from 2.55 mmol/mL NH₃BH₃ concentration and 15.50 mg catalyst amount, as shown in Fig. 4(a). The increase in H₂ production rate may be ascribed to the fact that increasing the concentration of NH₃BH₃ would promote catalyst reactant contact. However, after the optimum point for both parameters, it is observed that the hydrogen production rate decreases. The increase in NH₃BH₃ correspondingly increases the viscosity of the solution which inhibites effective mass transfer or catalyst and reactant interaction [⁴¹].The relative interaction between the variables can be evaluated by the degree of the ellipticity of the plots [⁴³]. Thus, it is possible to observe that the greatest interaction effect on hydrogen production rate can be seen for catalyst amount and NH₃BH₃ concentration. The 3D-plot in Fig. 4(b) demonstrates the temperature and NH₃BH₃ concentration effect on H₂ production rate for the hydrolysis of NH₃BH₃. The investigation on the effect of temperature and NH₃BH₃ concentration on the catalytic reaction was conducted by ranging the reaction temperature from 293 K to 323 K spectrum and NH₃BH₃ concentration variability (0.1 and 5.0 mmol/5 mL) keeping the catalyst amount at 15.50 mg. It is clear that the increase in temperature is directly proportional to the increase in hydrogen production, as shown in Fig. 4(b). This indicates that the hydrolysis temperature has a positive effect on catalytic activity.

The ANOVA of the process parameters of the independent variables demonstrates that the rate of hydrogen production is most affected by the process parameters of catalyst amount and NH₃BH₃ concentration as well as that of temperature and NH₃BH₃ concentration on H₂ production rate. The p-values of 0.1227 and 0.3089 obtained for the treatment combinations of catalyst amount/NH₃BH₃ concentration and temperature/NH₃BH₃ concentration show that they have statistically significant effects on the hydrogen production rate.

Figures 4(c–d) display the most convenient fitted response surface plots obtained from the CV responses for experimental (A, B and C) factors. As shown in Fig. 4(c), the variation of NH₃BH₃ concentration is more important than that of temperature on NH₃BH₃ electro-oxidation peak current. This could be attributed to the densities of the oxidation of NH₃BH₃, perhaps via hydroxyl trihydroborate anion intermediate (BH₃OH) generated from the reaction of NH₃BH₃ with hydroxide ion [⁴⁴]. However, the electro-oxidation efficiency relatively increases when the temperature increases. This indicates that the increasing temperature can enhance the rate of electro-oxidation of NH₃BH₃ because it can affect the diffusion rate of the ions in the electrolyte.

Figure 4(d) also indicates that the variation of NH₃BH₃ concentration remarkably affects NH₃BH₃ electro-oxidation, while the variation of catalyst amount is less significant. Therefore, it can be concluded that, to the PdCoAg/AC catalyst, the reaction rate is limited by the electron transfer.

It was found in this paper that the best run was with 2.55 mol/5 L of NH₃BH₃ concentration, 15.50 mg catalyst amount, and 50°C for hydrogen production from NH₃BH₃, and 1 M NH₃BH₃ concentration, 5 µL and 20°C for NH₃BH₃ electro-oxidation.

Optimization process by numerical analysis

The optimum points of the independent parameters selected for H₂ production rate and current were performed by using the numerical optimization method in CCD. The maximum H₂ production rate and current were searched at “in range” levels of the parameter sstudied. Of a number of solutions generated by the program, the most suitable was chosen for optimum points. The optimum values for H₂ production rate based on the numerical method were determined to be 2.91 mmol/5 mL NH₃BH₃ concentration, 13.26 mg catalyst amount, and 50°C temperature. Under these optimal conditions, the maximum H₂ production rate was found to be 10105.6 mL/(g·min). It should be noted that the synergistic effect between Pd, Co, and Ag plays an important role in enhancement for NH₃BH₃ hydrolysis. Synergism was explained by the electronic interaction between constituents, especially at the active sites where Co or Ag is deposited on the top and on the edges of Pd islands [⁴⁵]. Likewise, the optimal conditions of the maximum current equal to 2.77984 mA were 28.21°C, 0.78 M, and 2.5 mL for temperature, NH₃BH₃ concentration, and catalyst amount, respectively. To confirm the adequacy of the model for predicting the H₂ production rate and electro-oxidation current, a new experiment was conducted using the optimum levels, as shown in Table 4 and Figs. 5 and 6, which indicated that the experimental and predicted data were close to each other. It can be said that the RSM approach is promising to optimize and model the H₂ production rate and fuel cell performance.

Conclusions

In this paper, for the first time, the central composite design and response surface methodolgy approaches were employed to investigate both hydrogen production and electro-oxidation from NH₃BH₃ by using PdCoAg/AC catalysts.The maximum hydrogen production rate of 11414.4 mL/(g·min) was estimated for 2.55 mmol/5 mL NH₃BH₃ concentration, 15.50 mg catalyst amount, and 50°C temperature. Likewise, electrochemical properties of PdCoAg/AC catalysts at the maximum current equal to 2.77984 mA, which was optimized by the surface response method, were 28.21°C, 0.78 M, and 2.5 mL for temperature, NH₃BH₃ concentration and catalyst amount, respectively. However, the maximum NH₃BH₃ electro-oxidation current of 2.895 mA was estimated for 1 M NH₃BH₃ concentration, 5 µL catalyst amount, and 20°C temperature. The experimental data obtained from hydrogen production by NH₃BH₃ and electro-oxidation reaction over the PdCoAg/AC catalysts were exposed to ANOVA to investigate the importance of the independent variables and recognize the models. Among the evaluated RSM model, the quadratic model with an R² of 0.85 and 0.95 for H₂ production rate and current, respectively, was the most suitable to the experimental data. In the light of the results obtained, it can be concluded that CCD-based RSM can be a useful and encouraging tool to reliably optimize the production of hydrogen and experimental factors in electrochemistry.

References

[1] Andrews J, Shabani B. Re-envisioning the role of hydrogen in a sustainable energy economy. International Journal of Hydrogen Energy, 2012, 37(2): 1184–1203

[2] Tiwari A, Pandey A. Cyanobacterial hydrogen production—a step towards clean environment. International Journal of Hydrogen Energy, 2012, 37(1): 139–150

[3] Moriarty P, Honnery D. Hydrogen’s role in an uncertain energy future. International Journal of Hydrogen Energy, 2009, 34(1): 31–39

[4] Acar C, Dincer I. Comparative assessment of hydrogen production methods from renewable and non-renewable sources. International Journal of Hydrogen Energy, 2014, 39(1): 1–12

[5] Dincer I, Acar C. Review and evaluation of hydrogen production methods for better sustainability. International Journal of Hydrogen Energy, 2015, 40(34): 11094–11111

[6] Ball M, Weeda M. The hydrogen economy–vision or reality? International Journal of Hydrogen Energy, 2015, 40(25): 7903–7919

[7] Meeks N D, Baxley S. Fuel cells and the hydrogen economy. Chem Eng Prog, 2016, 112(7): 34–37

[8] Granovskii M, Dincer I, Rosen M A. Economic and environmental comparison of conventional, hybrid, electric and hydrogen fuel cell vehicles. Journal of Power Sources, 2006, 159(2): 1186–1193

[9] Van Mierlo J, Maggetto G, Lataire P. Which energy source for road transport in the future? A comparison of battery, hybrid and fuel cell vehicles. Energy Conversion Management, 2006, 47(17): 2748–2760

[10] Wang Y, Meng W, Wang D, Wang Z R, Zou K L, Cao Z Q, Zhang K, Wu S W, Li G D. Ultrafine cobalt-molybdenum-boron nanocatalyst for enhanced hydrogen generation property from the hydrolysis of ammonia borane. International Journal of Hydrogen Energy, 2019, 44(41): 23267–23276

[11] Appleby A J. Fuel cell technology: status and future prospects. Energy, 1996, 21(7–8): 521–653

[12] Amendola S C, Sharp-Goldman S L, Janjua M S, Kelly M T, Petillo P J, Binder M. An ultrasafe hydrogen generator: aqueous, alkaline borohydride solutions and Ru catalyst. Journal of Power Sources, 2000, 85(2): 186–189

[13] Miesse C M, Jung W S, Jeong K J, Lee J K, Lee J, Han J, Yoon S P, Nam S W, Lim T H, Hong S A. Direct formic acid fuel cell portable power system for the operation of a laptop computer. Journal of Power Sources, 2006, 162(1): 532–540

[14] Dillon R, Srinivasan S, Arico A S, Antonucci V. International activities in DMFC R&D: status of technologies and potential applications. Journal of Power Sources, 2004, 127(1–2): 112–126

[15] Zadick A, Dubau L, Artyushkova K, Serov A, Atanassov P, Chatenet M. Nickel-based electrocatalysts for ammonia borane oxidation: enabling materials for carbon-free-fuel direct liquid alkaline fuel cell technology. Nano Energy, 2017, 37: 248–259

[16] Wu D F, Ouyang L Z, Huang J M, Liu J W, Wang H, Shao H, Zhu M. Synthesis and hydrogen storage property tuning of fit Zr(BH₄)₄. 8NH₃ via physical vapour deposition and composite formation. International Journal of Hydrogen Energy, 2018, 43(41): 19182–19188

[17] Zhang X B, Yan J M, Han S, Shioyama H, Yasuda K, Kuriyama N, Xu Q. A high performance anion exchange membrane-type ammonia borane fuel cell. Journal of Power Sources, 2008, 182(2): 515–519

[18] Zhang X B, Han S, Yan J M, Chandra M, Shioyama H, Yasuda K, Kuriyama N, Kobayashi T, Xu Q. A new fuel cell using aqueous ammonia-borane as the fuel. Journal of Power Sources, 2007, 168(1): 167–171

[19] Ouyang L Z, Chen W, Liu J W, Felderhoff M, Wang H, Zhu M. Enhancing the regeneration process of consumed NaBH₄ for hydrogen storage. Advanced Energy Mater, 2017, 7(19): 1700299

[20] Chen W, Ouyang L Z, Liu J W, Yao X D, Wang H, Liu Z W, Zhu M. Hydrolysis and regeneration of sodium borohydride (NaBH₄): a combination of hydrogen production and storage. Journal of Power Sources, 2017, 359: 400–407

[21] Hua T Q, Ahluwalia R K. Off-board regeneration of ammonia borane for use as a hydrogen carrier for automotive fuel cells. International Journal of Hydrogen Energy, 2012, 37(19): 14382–14392

[22] Rees N V, Compton R G. Carbon-free energy: a review of ammonia- and hydrazine-based electrochemical fuel cells. Energy & Environmental Science, 2011, 4(4): 1255–1260

[23] Tan Z H, Ouyang L Z, Liu J W, Wang H, Shao H Y, Zhu M. Hydrogen generation by hydrolysis of Mg-Mg₂Si composite and enhanced kinetics performance from introducing of MgCl₂ and Si. International Journal of Hydrogen Energy, 2018, 43(5): 2903–2912

[24] Chandra M, Xu Q. A high-performance hydrogen generation system: transition metal-catalyzed dissociation and hydrolysis of ammonia-borane. Journal of Power Sources, 2006, 156(2): 190–194

[25] Li X G, Liu T, Sato M, Takahashi S. Synthesis and characterization of Fe-Ti nanoparticles by nitrogen plasma metal reaction. Powder Technology, 2006, 163(3): 183–187

[26] Rakap M, Kalu E E, Ozkar S. Polymer-immobilized palladium supported on TiO₂ (Pd-PVB-TiO₂) as highly active and reusable catalyst for hydrogen generation from the hydrolysis of unstirred ammonia-borane solution. International Journal of Hydrogen Energy, 2011, 36(2): 1448–1455

[27] Sahiner N, Ozay O, Aktas N, Inger E, He J B. The on demand generation of hydrogen from Co-Ni bimetallic nano catalyst prepared by dual use of hydrogel: as template and as reactor. International Journal of Hydrogen Energy, 2011, 36(23): 15250–15258

[28] Yuzawa H, Yoshida T, Yoshida H. Gold nanoparticles on titanium oxide effective for photocatalytic hydrogen formation under visible light. Applied Catalysis B: Environmental, 2012, 115–116: 294–302

[29] Turhan T, Avcibasi Y G, Sahiner N. Versatile p(3-sulfopropyl methacrylate) hydrogel reactor for the preparation of Co, Ni nanoparticles and their use in hydrogen production. Journal of Industrial and Engineering Chemistry, 2013, 19(4): 1218–1225

[30] Tong D G, Han X, Chu W, Chen H, Ji X Y. Preparation of mesoporous Co-B catalyst via self-assembled triblock copolymer templates. Materials Letters, 2007, 61(25): 4679–4682

[31] Sahiner N, Sagbas L. The use of poly(vinyl phosphonic acid) microgels for the preparation of inherently magnetic Co metal catalyst particles in hydrogen production. Journal of Power Sources, 2014, 246: 55–62

[32] Yao Q L, Lu Z H, Yang Y W, Chen Y Z, Chen X S, Jiang H L. Facile synthesis of graphene-supported Ni-CeO_x nano-composites as highly efficient catalysts for hydrolytic dehydrogenation of ammonia borane. Nano Research, 2018, 11(8): 4412–4422

[33] Choi S M, Seo M H, Kim H, Kim W B. Synthesis and characterization of graphene-supported metal nanoparticles by impregnation method with heat treatment in H₂ atmosphere. Synthetic Metals, 2011, 161(21–22): 2405–2411

[34] Ozay O, Aktas N, Inger E, Sahiner N. Hydrogel assisted nickel nanoparticle synthesis and their use in hydrogen production from sodium boron hydride. International Journal of Hydrogen Energy, 2011, 36(3): 1998–2006

[35] Fernandes R, Patel N, Miotello A. Hydrogen generation by hydrolysis of alkaline NaBH₄ solution with Cr-promoted Co-B amorphous catalyst. Applied Catalysis B: Environmental, 2009, 92(1–2): 68–74

[36] Wang S, Zhang D, Ma Y Y, Zhang H, Gao J, Nie Y T, Sun X H. Aqueous solution synthesis of Pt-M (M= Fe, Co, Ni) bimetallic nanoparticles and their catalysis for the hydrolytic dehydrogenation of ammonia borane. ACS Applied Materials & Interfaces, 2014, 6(15): 12429–12435

[37] Tano T, Esumi K, Meguro K. Preparation of organopalladium sols by thermal-decomposition of palladium acetate. Journal of Colloid and Interface Science, 1989, 133(2): 530–533

[38] Bezerra M A, Santelli R E, Oliveira E P, Villar L S, Escaleira L A. Response surface methodology (RSM) as a tool for optimization in analytical chemistry. Talanta, 2008, 76(5): 965–977

[39] Ling Z, Cao J, Zhang W, Zhang Z, Fang X, Gao X. Compact liquid cooling strategy with phase change materials for Li-ion batteries optimized using response surface methodology. Applied Energy, 2018, 228: 777–788

[40] Yılmaz Ş, Ecer Ü, Şahan T. Modelling and optimization of As(III) adsorption onto thiol-functionalized bentonite from aqueous solutions using response surface methodology approach. ChemistrySelect, 2018, 3(32): 9326–9335

[41] Çelik Kazici H, Yildiz F, İzgi M S, Ulaş B, Kivrak H. Novel activated carbon supported trimetallic PdCoAg nanoparticles as efficient catalysts for the hydrolytic dehydrogenation of ammonia borane. International Journal of Hydrogen Energy, 2019, 44(21): 10561–10572

[42] Şahan T, Erol F, Yılmaz Ş. Mercury(II) adsorption by a novel adsorbent mercapto-modified bentonite using ICP-OES and use of response surface methodology for optimization. Microchemical Journal, 2018, 138: 360–368

[43] Hitit Z Y, Lazaro C Z, Hallenbeck P C. Hydrogen production by co-cultures of Clostridium butyricum and Rhodospeudomonas palustris: optimization of yield using response surface methodology. International Journal of Hydrogen Energy, 2017, 42(10): 6578–6589

[44] Zhang X B, Han S, Yan J M, Shioyama H, Kuriyama N, Kobayashi T, Xu Q. Electrochemical oxidation of ammonia borane on gold electrode. International Journal of Hydrogen Energy, 2009, 34(1): 174–179

[45] Smiljanić M, Srejic I, Potocnik J, Mitric M, Rakocevic Z, Strbac S. Synergistic electrocatalytic effect of Pd and Rh nanoislands co-deposited on Au(poly) on HER in alkaline solution. International Journal of Hydrogen Energy, 2018, 43(42): 19420–19431

RIGHTS & PERMISSIONS

Higher Education Press