九州大学学術情報リポジトリ

(1)

九州大学学術情報リポジトリ

Kyushu University Institutional Repository

表面および細孔構造最適化による吸着式ヒートポンプ用高性能活性炭の開発

吉, 鉉植

https://doi.org/10.15017/1470611

出版情報：Kyushu University, 2014, 博士（工学）, 課程博士バージョン：

権利関係：Fulltext available.

(2)

表面および細孔構造最適化による

吸着式ヒートポンプ用高性能活性炭の開発

Development of high-performance activated carbons for adsorption heat pump through

surface and pore structure optimization

2014年 9月

九州大学総合理工学府

量子プロセス理工学専攻

吉鉉植

Hyun-Sig Kil

(3)

表面および細孔構造最適化による

吸着式ヒートポンプ用高性能活性炭の開発

Development of high-performance activated carbons for adsorption heat pump through

surface and pore structure optimization

指導教授：宮脇仁

2014年 9月

九州大学総合理工学府

量子プロセス理工学専攻

吉鉉植 Hyun-Sig Kil

論文調査委員会

主査九州大学准教授宮脇仁副査九州大学教授尹聖昊

副査九州大学教授島ノ江憲剛

(4)

Abstract

Recently, adsorption heat pump (AHP) systems are attracting much attention because AHP can effectively use low-temperature waste heat of 373 K or lower. Various adsorbent-adsorbate pairs such as zeolite-water, silica-gel-water, activated carbon (AC)-methanol, and AC-ammonia have been studied to improve cooling power per volume, because the performance is strongly influenced by a selection of adsorbent-adsorbate pairs and operation conditions.

Among them, AC-ethanol pair is considered to be promising for AHP due to the high adsorption capacity and relatively high vapor pressure even at low temperatures. To develop high performance ACs for AHP, there are various modification factors, such as surface area, pore size and its distribution, surface functional groups, pore depth, packing density, and graphitization degree, to be considered and optimized. However, no systematic studies have been reported how these factors influence on the adsorption and heat characteristics of ACs for the AHP application.

The objective of this study is thedevelopment of high-performance ACs for AHP through surface and pore structure optimizations. To achieve this objective, surface functional groups, which are considered to strongly influence on adsorption behaviors of polar molecules like ethanol, and pore structures of ACs were highlighted and investigated how each factor individually influences on the adsorption characteristics, especially for effective adsorption capacity, adsorption/desorption speed, and adsorption state of ethanol. In this study, two series of model ACs, one having comparable pore structure but different content of surface functionalities, and another having similar content of surface functionalities but different pore structure, were carefully and selectively prepared and applied.

The findings in this study are summarized as follows.

In Chapter 1, working principles of AHP systems, properties and preparation methods of ACs, importance of ACs as an adsorbent for AHP, and purpose of this study were described.

i

(5)

In Chapter 2, the influence of surface functionalities on ethanol absorbability was investigated using the model ACs having comparable surface area and pore size but different contents of oxygen- containing surface functionalities. The oxygen content was controlled by heat treatments in H2or with potassium hydroxide (KOH) with maintaining the pore structure using an ultrahigh surface area AC, Maxsorb III, as a parent material. Abundant surface functionalities were found to induce diffusional hindrance of ethanol molecules in carbon micropores, giving rise to a decrease of adsorption amount and shortening of adsorption/desorption equilibrium time of ethanol. The solid- state NMR and electrochemical analyses confirmed the strong and oriented adsorption state of ethanol molecule with the hydroxyl group facing the oxygen-containing surface functional group.

In Chapter 3, the control of pore structures of ACs by careful adjustments of carbonization and KOH activation conditions was studied. Using a spherical phenol resin (BEAPS) as a precursor, spherical ACs with different specific surface area, pore volume, and pore size were selectively prepared by controlling the carbonization and activation temperatures and KOH/carbon ratio. For example, ACs with high specific surface area of about 3,000 m²/g but different pore size and pore volume were successfully obtained.

In Chapter 4, using the model AC samples having similar content of surface functionalities but different pore structure, which were prepared based on the findings in Chapter 3, influence of the pore structures not only on adsorption performances, such as adsorption amounts and kinetics, but also on adsorption states of the ethanol molecules in carbon micropores were investigated by the ethanol adsorption and solid-state NMR measurements. Both effective adsorption amount and adsorption/desorption speed were improved by the optimization of the pore size of ACs. For example, an AC having 1.6 nm of average pore width gave more than 720 mg/g of the effective adsorption amount of ethanol with excellent adsorption/desorption kinetics. Furthermore, the NMR studies revealed that the influence of the pore size on the adsorption state of ethanol molecules in the micropores is not remarkable as compared with that of the surface functionalities.

ii

(6)

In Chapter 5, the findings obtained in this study were summarized.

iii

(7)

Chapter 1 Introduction

Increase energy consumption around the world causes serious environmental problems such as global warming and the depletion of energy resources. In order to solve such problems, the utilization of renewable energy, waste heat recovery and utilization, and energy saving are needed.

Today, a huge quantity of low-temperature renewable energy is released into the atmosphere such as industrial waste heat and solar heat. There are several different types of heat pumps; the most widely used type is the mechanical heat pump. However, there are lots of issues to be improved because the refrigerant that is being used in the mechanical heat pumps causes serious damages to the environment. Commonly used large-scale industrial absorption heat pump systems have problems with high operating and maintenance costs because pressure changes is extremely high between the parts of the system. Therefore, adsorption heat pump system has been proposed to complement the drawbacks especially using activated carbons as adsorbents.

1

(11)

1.1 Adsorption heat pump

1.1.1 Background

Recent researches on heat pump are mainly focused on the development of environmentally friendly new systems that has high primary energy efficiency and can run by used renewable energy.

Adsorption heat pump that has advantages of being environmental friendly and can work by low temperature thermal sources such as solar energy or industrial waste heat. According to the several studies, AHPs have the potential to compete with absorption and vapor compression heat pump. The COP of AHPs is not so high, compare to the other conventional heat pump. However, the crucial drawbacks such as low COP (coefficient of performance) and discontinuous operation mode should be overcome. Many studies have been performed to eliminate these problems and to improve the COP of the cycle for the development of innovative design. In addition, design for the adsorption bed is being developed in order to increase the heat and mass transfers directly for the system efficiency.

1.1.2 Adsorption heat pump cycle

Adsorption chillers or heat pumps, like refrigerators and air conditioners, are a kind of machineries to transport "heat" from low-temperature heat source to high-temperature heat source by the principle of thermodynamic. The structure of the refrigerator and a heat pump is basically same.

For example, the low-temperature heat source, the machine for the purpose of relieving the heat in the refrigerator is called a chiller and the machine for the purpose of applying heat on heating operation of the air conditioner is called a heat pump. In either case, a refrigerator or a heat pump does not work without any energy input because it is against the natural flow of heat from higher temperature to lower temperature. For example, refrigerators and air conditioners are driven by

2

(12)

electricity to move the compressor. On the other hand, adsorption chiller and heat pump have a great advantage which can operate refrigeration/heat pump by heating.

Adsorption heat pump cycle is basically composed with four main parts: an adsorber (which is filled with adsorbent such as silica gel, zeolite, activated carbon, etc.), a condenser, an evaporator and expansion valves. The schematic of adsorption heat pump system is shown in Fig. 1-1. AHP operates by cycling adsorbate between the adsorber, the condenser, and the evaporator [1, 2].

Adsorption phenomena in the AHP work the same role of mechanical power, so that the operation fluid can be circulated in the cycle without mechanical power.

Fig. 1-2 illustrates the thermodynamic cycle of the basic adsorption heat pump [3]. AHP cycle consists of four steps which are isosteric heating, isobaric desorption, isosteric cooling, and isobaric adsorption as shown in Fig.1- 2.

3

(13)

Fig. 1-1. Schematic of adsorption heat pump system.

4

(14)

Fig. 1-2. Thermodynamic cycle of a basic adsorption heat pump.

5

(15)

Isosteric heating (a→b): The valves between adsorbent bed and the condenser and the evaporator are closed. The adsorbent bed temperature increases by heating without desorption. In this step, the amount of heat is given as Qab.

 Isobaric desorption (b→c): After the isosteric heating of adsorbent bed, the heating process is continued. The valve between the adsorbent and the condenser is opened. The desorption step is started and water vapor is condensed in the condenser. The cycle pressure remains steady. The transferred heat to the adsorbent bed increases the temperature of adsorbent- adsorbate pair and adsorbent bed. The amount of heat is given as Qbc in this step.

 Isosteric cooling (c→d): The valve between the condenser and adsorbent bed is closed and the temperature of adsorbent bed, which is the maximum value of this cycle, decreases.

During the process, the pressure and the temperature of adsorbent bed decreases the temperature of evaporator values. In this step, the amount of cooling is given as Qcd.

 Isobaric adsorption (d→a): The valve between adsorbent bed and evaporator is opened then evaporation of the adsorbate in the evaporator is started. During the adsorbing of the adsorbate in the adsorbent, heat is released by heat of adsorption. The generated heat should be removed from the adsorbent bed and the temperature of adsorbent-adsorbate pair and adsorber are decreased. The heat of evaporation is caused cooling effect and the heat of condensation can be used for heating purpose. In this step, the amount of released heat is given Qda.

The cooling effect in this cycle occurs during the isobaric adsorption process (d→a) when the adsorbate is evaporated by gaining heat from the environment. The heating effect occurs during the isobaric desorption process (b→c) when the adsorbate is condensed by releasing heat to surroundings [3].

6

(16)

1.1.3 Adsorbent-adsorbate pairs

The most adsorbent-adsorbate pairs will undergo a certain amount of adsorption. The amount of refrigerant which can be desorbed from the adsorbent for a certain temperature lift directly affects the efficiency of the machine. It is therefore important that the choice of the adsorbent-adsorbate (refrigerant) pair is made with the knowledge of the application for which it is intended. The basic requirements of adsorbents and adsorbates showed Table 1-1 [4-6].

Adsorption-adsorbate working pairs belong in one of three categories: physical, chemical or composite. The physical adsorbents that are used in adsorption heat pumps rely on van der Waal’s forces to contain adsorbate. Commonly, adsorbents include silica gel, zeolite and activated carbon (or activated carbon fibers). Tather et al. state that working pair selection is one of the most important design decisions in engineering an adsorption heat pump [7]. For more on evaluating working pair types [8]. The remainder of this section discusses properties of widely used adsorbents.

7

(17)

Table 1-1. The basic requirements of adsorbents and adsorbates [4-6].

Adsorbents Adsorbates

 have large change in adsorbate content under the design operating conditions

 have adequate refrigerant evaporating temperature (e.g., subzero freezing temperatures for ice making)

 have high latent heat of vaporization

 have saturation pressures slightly above atmospheric for the design operating conditions

 have low specific volume

 are non-toxic, non-corrosive, and non- flammable

 are stable for repeated use

 have high thermal conductivity

 are low intra-particle and inter-particle mass transfer resistances

 can desorb a large amount of refrigerant over a given temperature lift

 are stable with the refrigerant used.

 have a low specific heat.

 have good heat transfer properties.

 are inexpensive and readily available.

8

(18)

Silica gel works well for low-temperature applications (i.e., desorption temperatures between 50 and 90°C with water refrigerant). The reason silica gel-water works for low temperatures is because it has a favorable isotherm and can adsorb/desorb a large amount of water without too great a temperature change. Water is environmentally benign and has a solidification temperature of 0°C, so that frost or ice accumulation on the air-side of the evaporator is not a concern. On the other hand, it cannot be used for ice making. The adsorbents differ in structure; adsorbent surface area, pore size, pore volume, pore distribution, porosities, and skeletal density [9]. Although silica gel-water works well for low-temperature applications, indicated that silica gel cannot be utilized at driving temperatures above 100°C without compromising the integrity of the adsorbent [10, 11].

The zeolite-water working pair is appropriate for higher desorption temperatures, generally above 200°C. A wide variety of natural and synthetic zeolites exist, and among zeolites for adsorption heat pumps applications with water as refrigerant [12].

Activated carbons can be made using local organic material as opposed to other adsorbents such as some zeolites, which must be synthesized. Methanol cannot be used in cycles with source temperatures above 110-150°C, because it is unstable and decomposes into dimethyl-ether and ethanol, especially in the presence of aluminum [5, 13, 14]. Despite its incompatibility with high source temperature, methanol has the advantage of having a sub-zero freezing point, and the activated carbon-methanol pairs can generally be desorbed using low-grade heat. The sub-zero freezing point is useful for ice making, deep freezing, or automotive air conditioning for cars driven in regions with cold winters [15]. El-Sharkawy et al. studied the activated carbon fiber-ethanol working pair. It was determined that ACF performed substantially better than ACF. The two adsorbents have comparable pore diameters, but have higher specific surface area (1.9×106 m²/kg compared with 1.4×10⁶ m²/kg) and higher specific pore volume (10.28×10^-4 m³/kg compared with 7.65×10^-4 m³/kg). El-Sharkawy et al. showed that for their 10 mm thin ACF sample compacted to a bulk density of 100 kg/m³, temperature gradients were significant. Five minutes into their adsorption

9

(19)

test, the difference in surface and internal temperatures was over 20°C for a 25°C heat sink [16].

Saha et al. measured the equilibrium uptake and diffusivity of ethanol in ACFs at temperatures between 27 and 60°C using thermo gravimetric analysis under isothermal conditions [17]. In a second study, Saha et al. compared the pore diameter, volume and surface area of three types of activated carbon fiber with three types of silica gel, and ACF was determined to have the highest pore volume and surface area (10.28×10^-4 m³/kg and 1.9×106 m²/kg, respectively) [18].

1.1.4 Advantages and disadvantages of adsorption heat pump

In recent years, the importance of adsorption heat pumps and adsorption refrigeration systems has increased since these kinds of systems can directly utilize the primary thermal energy sources and additionally the waste heat generated in various industrial processes. The important advantages of the adsorption heat pumps can be described as follows:

10

(20)

Table 1-2. Advantages and disadvantages of adsorption heat pump [2, 19].

Advantages Disadvantages

 can operate with thermal driving energy sources

 can work with low temperature driving energy sources

 do not require moving parts for circulation of working fluid

 have long life time

 operate without noise and vibration have simple principle of working

 do not require frequent maintenance

 are environmental friendly

 can be employed as thermal energy storage device

 have low COP values

 intermittently working principles

 require high technology and special designs to maintain high vacuum

 have large volume and weight relative to traditional mechanical heat pump systems

11

(21)

1.1.5 Applications

Adsorption heat pumps can be driven by heat source temperature ranges between 40 and 500°C and can provide evaporator temperatures from −30 to 150°C for a variety of applications [12, 20].They can be designed for simplicity, efficiency or low temperature lift [21-23]. Because of their flexibility, investigators have proposed using adsorption heat pumps for a wide variety of applications in the areas of air conditioning, solar refrigeration, food storage, electronics cooling and other applications. The best of these applications take advantage of either the simplicity of or low temperature heat sources requirement [24-28].

12

(22)

1.2 Activated carbons

1.2.1 Background

Activated carbons (ACs) are typical porous carbon materials with developed pore structures.

Activated carbons have been used before because ACs have the strongest physical adsorption forces or the highest volume of adsorbing porosity of any material. ACs are used for liquid-phase adsorption such as wastewater treatment, industrial separation, and drinking water purification. ACs are also used for gas-phase adsorption such air purification in air pollution control. In addition, ACs are being applied including food & beverage industry and medical field. The application fields of activated carbons are getting wider as well.

1.2.2 Preparation of activated carbons

Generally activated carbon can be prepared from various raw materials including agricultural and forestry residues. Generally most of the precursors used for the preparation of activated carbon are rich in carbon [29]. Production of AC was achieved typically through two routes, physical activation and chemical activation [30]. Physical activation involves carbonization of raw material followed by the activation at high temperatures (between 800 and 1100°C) in the presence of oxidizing gases like carbon dioxide or steam [31-34], whereas chemical activation mixing of chemical agent with precursor and then followed by pyrolysis at moderate temperatures in the absence of air [35, 36]. Typical preparation of activated carbon involves carbonization of the raw material in the absence of oxygen, and activation of the carbonized product [37]. Chemical activation, on the other hand enjoys the benefit of development of better porous structure in a single process route at low carbonization temperatures as compared to physical activation.

13

(23)

Physical activation is a two-step process. It involves carbonization of raw material followed by activation at elevated temperatures in the presence of suitable oxidizing gases such as carbon dioxide, steam, air or their mixture gases. Carbonization temperature ranges between 400 to 800°C, and activation temperature ranges between 800 to 1100°C. Generally, CO2 is used as activation gas, since it is clean, easy to handle, and it facilitates control of the activation process due to the slow reaction rate at high temperatures.

Preparation of activated carbon by chemical activation is a single step process in which carbonization and activation is carried out simultaneously. Initially the precursor is mixed with chemical activating agent, which acts as dehydrating agent and oxidant. The most commonly used chemical activating agents are H3PO4, ZnCl2, and KOH. Chemical activation offers several advantages over physical activation which mainly include (1) lower activation temperature (<

800°C) compared to the physical activation temperature (800 - 1100°C), (2) single activation step, (3) higher yields, (4) better porous characteristics, and (5) shorter activation times [29, 33, 38, 39].

1.2.3 Structure of activated carbons

The adsorption capacity of AC highly depends on the structure of activated carbon. The high adsorptive capacities of activated carbons are highly related to porous characteristics such as surface area, pore volume, and pore size distribution. All activated carbons have a porous structure, containing up to 15 % of mineral matter in the form of ash content. The porous structure of AC formed during the carbonization process and was developed further during activation, when the spaces between the elementary crystallites are cleared of tar and other carbonaceous material. The structure of pores and pore size distribution largely depends on the nature of the raw material and activation process route.

14

(24)

The activation process removes disorganized carbon by exposing the crystallites to the action of activating agent which leads to the development of porous structure. The pore systems of activated carbon are of different kinds and the individual pores may vary greatly both in size and shape. Active carbons are associated with pores starting from less than a nanometer to several thousand nanometers.

A conventional classification of pores according to their average width (w), which represents the distance between the walls of slit shaped pore or the radius of a cylindrical pore, proposed by Dubinin et al. [40], and officially adopted by the International Union of Pure and Applied Chemistry (IUPAC) is shown Fig. 1-3 and Table 1-3.

Microcrystalline structure of activated carbons starts to develop during the carbonization process.

The Crystalline structure of activated carbons differed from the graphite with respect to the interlayer spacing. The interlayer spacing ranges between 0.34 and 0.35 in active carbons, which is 0.335 in case graphite. The basic structural unit of activated carbon is closely approximated by the structure of graphite as shown Fig. 1-4. Much of the literature suggests a modified graphite structure for activated carbon. During the carbonization process free valences were created due to the regular bonding disruption of micro-crystallites. In addition, process conditions and presence of impurities influence the formation of vacancies (pores) in microcrystalline structure [41].

15

(25)

Fig. 1-3. Graphical representation of pore structure in activated carbon

Table 1-3. Classification of pores according to their width

Type of pores Width (w)

Micropores

Mesopores

Macropores

< 2 nm (20 Å)

2 – 50 nm (20 – 500 Å)

>50 nm (500 Å)

16

(26)

Fig. 1-4. Layered structure of graphite.

17

(27)

1.2.4 Classification of activated carbons

Activated carbons are complex products and the classification is difficult based on their preparation methods, physical properties, and surface characteristics. However, the general classification of activated carbons based on particle size divides them into Powder type of Activated Carbon (PAC), Granular type of Activated Carbon (GAC), and Activated Carbon Fibers (ACF).

Powder type of Activated Carbon (PAC), has a typical particle size of less than 0.1 mm and the common size of the particle ranges from 0.015 to 0.025 mm. Typical applications of PAC are industrial and municipal waste water treatments, sugar decolorization, in food industry, pharmaceutical, and mercury and dioxin removal from a flue gas stream [42].

Granular type of Activated Carbon (GAC) has mean particle size between 0.6 to 4 mm. It is usually used in continuous processes of both liquid and gas phase applications. GAC has an advantage over PAC, of offering a lower pressure drop along with the fact that it can be regenerated and therefore reused more than once. In addition to the proper micropore size distribution, its high apparent density, high hardness, and a low abrasion index made GAC more suitable over PAC for various applications [43].

Activated carbon Fibers (ACFs) are carbonized carbons which are subsequently heat treated in an oxidizing atmosphere. ACF began to be developed in 1970 using the precursor viscose rayon which mainly consists of cellulose [44]. Later thermoset polymer materials like saran and phenolic resins were used as precursors to produce ACF [45]. A good ACF precursor must be non-graphitic and nongraphitizable carbon fiber which was isotropic in nature. From the end of 1980s, interest is still centered on the production of ACFs from various inexpensive precursors [46].

18

(28)

1.2.5 Characterization of activated carbons

Activated carbons are strongly heterogeneous due to the existence of different sizes of pores including micropores, mesopores and macropores. In addition, the surface heterogeneity of activated carbons is often significant because of various oxygen and other groups present on the surface.

Surface and structural properties of the activated carbons can be studied directly by employing various techniques like electron microscopy, X-ray analysis and various spectroscopic methods. In addition, these properties can be investigated by indirect methods such as gas adsorption and thermal analyses. The data obtained from adsorption can be used mainly to extract information about surface heterogeneity and porosity of adsorbents.

1.2.6 Activated carbons for adsorption heat pump

In the adsorption cooling system, it is very important to select an appropriate adsorbent because the adsorptivity such as adsorption amount and adsorption speed is affected by the adsorbate. As adsorbents for adsorption heat pump, such as activated carbon or activated carbon fiber have a lot of micropores, are rapidly increasing amount of adsorption at low relative pressure, while high relative pressure almost no changes. This type of isotherms (Type I, Fig. 1-5.) exhibited by microporous solids having relatively small external surfaces. The adsorptivity for adsorption heat pump was not only driven by the surface area, but also the micropore volume and pore size.

In addition, it is the key points that are the effective adsorption amount of adsorbate in the adsorption heat pump operating pressure range and adsorption/desorption speed. The AHP operation range is fixed by three temperatures, low temperature TL, middle temperature TM and high temperature TH. For example, when AHP is operated under follow conditions; TL = 283 K, TM =

19

(29)

303K, TH = 353K, adsorbate is ethanol. The operating range is fixed as adsorption relative pressure φ1 = 0.3 and desorption pressure is φ2 = 0.1. The relative pressure is calculated as:

φ1 = equilibrium ethanol vapor pressure at TL / equilibrium ethanol vapor pressure at TM

φ2 = equilibrium ethanol vapor pressure at TM / equilibrium ethanol vapor pressure at TH

20

(30)

Fig. 1-5. Classification of the isotherm types by IUPAC.

I : Microporous materials (e.g. Zeolite and Activated carbon) II : Non porous materials (e.g. Nonporous Alumina and Silica)

III : Non porous materials and materials which have the weak interaction between the adsorbate and adsorbent (e.g. Graphite/water)

IV : Mesoporous materials (e.g. Mesoporous Alumina and Silica)

V : Porous materials and materials that have the weak interaction between the adsorbate and adsorbent (e.g. Activated carbon/water)

VI : Homogeneous surface materials (e.g. Graphite/Kr and NaCl/Kr)

21

(31)

Fig. 1-6. Effective adsorption properties for different micropore size (a) and micropore volume (b) at AHP operating pressure range.

(a)

(b)

22

(32)

Fig. 1-6 demonstrates the effective adsorption properties for different micropore size and different micropore volume at AHP operating pressure range. Firstly, I would describe the effect of pore size on the adsorption properties. The important points are Effective adsorption amount of adsorbate in AHP operating pressure range and Adsorption/desorption speed. The AHP operating pressure range is decided from the saturation vapor pressure at the temperatures of the evaporator and condenser when the ethanol adsorption-desorption operation.

For example, the saturated vapor pressure of each becomes 3.1 kPa and 10.5 kPa, when the temperature of the adsorption bed is assumed to be 10°C and the temperature of the evaporator at 30°C, relative pressure will be 0.3. In addition, the saturated vapor pressure of each becomes 10.5 kPa and 102.2 kPa, when the temperature of the adsorption bed is assumed to be 30°C the temperature of the condenser at 80°C, relative pressure will be 0.1.

In the case of narrow micropore, just a limited amount of effective adsorption in AHP operation conditions between from 0.1 to 0.3 of relative pressure, ∆q, will be obtained. On the other hand, for wide micropore, much larger effective adsorption amount is achievable. In a case of much wider pores, like mesopore, the effective adsorption amount will decrease again as shown Fig. 1-6 (a). In addition, in the case of the wider pores, faster adsorption/desorption speed is expected as compared with the narrower ones.

By increasing the pore volume, it is possible to increase the effective adsorption amount as shown Fig. 1-6 (b). Thus, it is likely activated carbons selectively controlling the micropore volume and pore size, it can improve both the adsorption/desorption speed and effective adsorption amount of the adsorbate.

23

(33)

1.3 Objective of the study

The objective of this study is “Development of high-performance activated carbons for adsorption heat pumps through surface and pore structure optimizations.” Fig. 1-7 shows the strategy for development of high performance ACs for AHPs. To achieve this objective, the adsorption and heat characteristics should be considered about the important properties for ACs to influence effective adsorption capacity, adsorption/desorption speed and thermal conductivity etc. In addition, the properties will be influenced to various factors such as surface functional groups, pore structures, packing density and degree of graphitization etc. Thus, I investigated how the each factor influences on the adsorption and heat characteristics individually by preparing suitable model activated carbons.

At the end of this study, I would like to adjust all these factors to obtain high-performance activated carbons for AHP system.

24

(34)

Fig. 1-7. Strategy for development of high performance ACs for AHPs.

25

(35)

References

[1] S. Ülkü, Adsorption heat pumps, Heat Recovery Syst, (1986) 277-84.

[2] F. Meunier, Adsorptive cooling: A clean technology, Clean Prod Processes, 3 (2002) 8-20.

[3] H. Demir, M. Mobedi, S. Ülkü, A review on adsorption heat pump: Problems and solutions, Renewable and Sustainable Energy Reviews, 12 (2008) 2381-2403.

[4] R.E. Critoph, Performance Limitations of adsorption cycles for solar cooling, Solar Energy, 41 (1988) 21-31.

[5] M.A. Alghoul, M.Y. Sulaiman, B.Z. Azmi, K. Sopian, M.A. Wahab, Review of materials for adsorption refrigeration technology, Anti-Corrosion Methods and Materials, 54 (2007) 225-229.

[6] A.A. Askalany, M. Salem, I.M. Ismail, A.H.H. Ali, M.G. Morsy, A review on adsorption cooling systems with adsorbent carbon, Renewable and Sustainable Energy Reviews, 16 (2012) 493-500.

[7] M. Tather, B. Tantekin-Ersolmaz, A. Erdem-Senatalar, A novel approach to Enhance heat and mass transfer in adsorption heat pumps using the zeolite-water pair, Microporous and Mesoporous Materials, 27 (1999) 1-10.

[8] A. Hauer, Evaluation of adsorbent materials for heat pump and thermal energy storage applications in open systems, Adsorption, 13 (2007) 399-405.

[9] H.T. Chua, K.C. Ng, A. Chakraborty, N.M. Oo, M.A. Othman, Adsorption characteristics of silica gel + water systems, Journal of Chemical and Engineering Data, 47 (2002) 1177-1181.

[10]S. Szarzynski, Y. Feng, M. Pons, Study of different internal vapour transports for adsorption cycles with heat regeneration, International Journal of Refrigeration, 20 (1997) 390-401.

[11]H.T. Chua, K.C. Ng, W. Wang, C. Yap, X.L. Wang, Transient modeling of a two-bed silica gel- water adsorption chiller, International Journal of Heat and Mass Transfer, 47 (2004) 659-669.

[12]L.W. Wang, R.Z. Wang, R.G. Oliveira, A review on adsorption working pairs for refrigeration, Renewable and Sustainable Energy Reviews, 13 (2009) 518-534.

26

(36)

[13]N. Douss, F. Meunier, Experimental study of cascading adsorption cycles, Chemical Engineering Science, 44 (1989) 225-235.

[14]E.J. Hu, A study of thermal decomposition of methanol in solar powered adsorption refrigeration systems, Solar Energy, 62 (1998) 325-329.

[15]G. Restuccia, A. Freni, F. Russo, S. Vasta, Experimental investigation of a solid adsorption chiller based on a heat exchanger coated with hydrophobic zeolite, Applied Thermal Engineering, 25 (2005) 1419-1428.

[16]I.I. El-Sharkawy, K. Kuwahara, B.B. Saha, S. Koyama, K.C. Ng, Experimental investigation of activated carbon fibers/ethanol pairs for adsorption cooling system application, Applied Thermal Engineering, 26 (2006) 859-865.

[17]B.B. Saha, I.I. El-Sharkawy, A. Chakraborty, S. Koyama, S.-H. Yoon, K.C. Ng, Adsorption rate of ethanol on activated carbon fiber, Journal of Chemical & Engineering Data, 51 (2006) 1587- 1592.

[18]B.B. Saha, S. Koyama, I.I. El-Sharkawy, K. Kuwahara, K. Kariya, K.C. Ng, Experiments for measuring adsorption characteristics of an activated carbon fiber/ethanol pair using a plate-fin heat exchanger, HVAC and R Research, 12 (2006) 767-782.

[19]W. Wongsuwan, S. Kumar, P. Neveu, F. Meunier, A review of chemical heat pump technology and applications, Applied Thermal Engineering, 21 (2001) 1489-519.

[20]B.B. Saha, E.C. Boelman, T. Kashiwagi, Computational analysis of an advanced adsorption- refrigeration cycle, Energy (Oxford), 20 (1995) 983-994.

[21]L.L. Vasiliev, D.A. Mishkinis, A.A. Antukh, L.L.J. Vasiliev, Solar-gas solid sorption heat pump, Applied Thermal Engineering, 21 (2001) 573-583.

[22]F. Meunier, Second law analysis of a solid adsorption heat pump operating on reversible cascade cycles: Application to the zeolite-water pair, Journal of Heat Recovery Systems, 5 (1985) 133- 141.

27

(37)

[23]B.B. Saha, S. Koyama, K.C. Ng, Y. Hamamoto, A. Akisawa, T. Kashiwagi, Study on a dual- mode, multi-stage, multi-bed regenerative adsorption chiller, Renewable Energy, 31 (2006) 2076-2090.

[24]G.Z. Yang, Z.Z. Xia, R.Z. Wang, D. Keletigui, D.C. Wang, Z.H. Dong, X. Yang, Research on a compact adsorption room air conditioner, Energy Conversion and Management, 47 (2006) 2167- 2177.

[25]J. Di, J.Y. Wu, Z.Z. Xia, R.Z. Wang, Theoretical and experimental study on characteristics of a novel silica gel-water chiller under the conditions of variable heat source temperature, International Journal of Refrigeration, 30 (2007) 515-526.

[26]L.W. Wang, R.Z. Wang, Z.S. Lu, C.J. Chen, K. Wang, J.Y. Wu, The performance of two adsorption ice making test units using activated carbon and a carbon composite as adsorbents, Carbon, 44 (2006) 2671-2680.

[27]Y.J. Dai, R.Z, Wang, Y.X. Xu, Study of a solar powered solid adsorption-desiccant cooling system used for grain storage, Renewable Energy, 25 (2002) 417-430.

[28]K.C. Ng, M.A. Sai, A. Chakraborty, B.B. Saha, S. Koyama, The electro-adsorption chiller:

Performance rating of a novel miniaturized cooling cycle for electronics cooling, Journal of Heat Transfer, 128 (2006) 889-896.

[29]D. Prahas, Y. Kartika, N. Indraswati, S. Ismadji, Activated carbon from jackfruit peel waste by H3PO4 chemical activation: Pore structure and surface chemistry characterization. Chemical Engineering Journal, 140 (2008) 32-42.

[30]R.C. Bansal, J.B. Donnet, H.F. Stoeckli, Active carbon, Marcel Dekker, New York, (1988) [31]A. Aworn, P. Thiravetyan, W. Nakbanpote, Preparation of CO2 activated carbon from corncob

for methylene glycol adsorption, Colloids and Surfaces A: Physicochemical Engineering Aspects, 333 (2009) 19-25.

[32]M. Lopez de Letona Sanchez, A. Macias-Garcia, M.A. Diaz-Diez, E.M. Cuerda-Correa, J.

28

(38)

Ganan-Gomez, A. Nadal-Gisbert, Preparation of activated carbons previously treated with hydrogen peroxide: Study of their porous texture. Applied Surface Science, 252 (2006) 5984- 5987.

[33]G.H. Oh, C.R. Park, Preparation and characteristics of rice-straw-based porous carbons with high adsorption capacity, Fuel, 81 (2002) 327-336.

[34]Y. Zhu, J. Gao, Y. Li, F. Sun, J. Gao, S. Wu, Y. Qin, Preparation of activated carbons for SO2

adsorption by CO2 and steam activation, Journal of the Taiwan Institute of Chemical Engineers, 43 (2012) 112-119.

[35]A. Klijanienko, E. Lorenc-Grabowska, G. Gryglewicz, Development of mesoporosity during phosphoric acid activation of wood in steam atmosphere, Bioresource Technology, 99 (2008) 7208-7214.

[36]J.M. Rosas, J. Bedia, J. Rodriguez-Mirasol, T. Cordero, HEMP – derived activated carbon fibers by chemical activation with phosphoric acid, Fuel, 88 (2009) 19-26.

[37]O. Ioannidou, A. Zabaniotou, Agricultural residues as precursors for activated carbon production – A review, Renewable and Sustainable Energy Reviews, 11 (2007) 1966-2005.

[38]A.A. El-Hendawy, A.J. Alexander, R.J. Andrews, G. Forrest, Effects of activation schemes on porous, surface and thermal properties of activated carbons prepared from cotton stalks, Journal of Analytical and Applied Pyrolysis, 82 (2008) 272-278.

[39]J. Hayashi, T. Horikawa, I. Takeda, K. Muroyama, F.N. Ani, Preparing activated carbon from various nutshells by chemical activation with K2CO3, Carbon, 40 (2002) 2381-2386.

[40]M.M. Dubinin, The potential theory of adsorption of gases and vapors for adsorbents with energetically nonuniform surfaces, Chemical Reviews, 60 (1960) 235-241.

[41]L.J. Kennedy, J.J. Vijaya, G. Sekaran, Effect of two-stage process on the preparation and characterization of porous carbon composite from rice husk by phosphoric acid activation, Industrial Engineering and Chemistry Research, 43 (2004) 1832-1838.

29

(39)

[42]Y. Satyawali, M. Balakrishnan, Performance enhancement with powdered activated carbon (PAC) addition in a membrane bioreactor (MBR) treating distillery effluent, Journal of Hazardous Materials, 170 (2009) 457-465.

[43]N. Zhang, L.S. Lin, D. Gang, Adsorptive selenite removal from water using iron-coated GAC adsorbents, Water Research, 42 (2008) 3809-3816.

[44]E.G. Doying, US patent 3256206, 1966.

[45]I. Martin-Gullon, R. Font, Dynamic pesticide removal with activated carbon fibers, Water Research, 35 (2001) 516-520.

[46]F. Derbyshire, R. Andrews, D. Jaques, M. Jagtoyen, G. Kimber, T. Rantell, Synthesis of isotropic carbon fibers and activated carbon fibers from pitch precursors, Fuel, 80 (2001) 345- 356.

30

(40)

Chapter 2 Influence of surface functionalities on ethanol adsorption characteristics in activated carbons for adsorption heat pumps

2.1 Introduction

Heat pumps are an environmentally friendly system that provides heating and cooling effects using relatively low-level heat sources such as solar energy, geothermal heat, and industrial waste heat [1, 2]. Recently, adsorption heat pumps (AHPs) have been actively studied and developed because they can effectively use low-temperature waste heat at around 373 K or lower. Two approaches have been used to improve the performance of AHP systems: chemical engineering- based and material-based approaches. For the latter approach, various adsorbent–adsorbate pairs, such as silica gel–water [3, 4], zeolite–water [5], activated carbon (AC)–CO2 [6], AC–water [7], AC–methanol [8, 9], and AC–ethanol [1, 9] pairs, have been considered because adsorptive and thermal characteristics depend not only on individual properties of the adsorbent and adsorbate, but also on their combination. Demir et al. [10] discussed problems and solutions associated with AHP systems. For example, the zeolite–water pair showed low adsorbability of approximately 30 wt.% of the adsorbent. On the other hand, the AC–ethanol pair showed much higher adsorption properties of more than 150 wt.% of the adsorbent [11]. Attan et al. [12] also reported that AC fiber, which has a high specific surface area, high total pore volume, and narrow average pore size, showed good adsorptivity for numerous refrigerants, including water, ammonia, acetone, methanol, and ethanol.

However, the adsorption behavior is influenced by pore structure and surface functionalities,

31

(41)

especially for polar molecules such as water and ethanol. In this study, we explored the role of surface functionalities of ACs on the adsorption behavior of ethanol molecules in carbon micropores.

To examine the influence of surface functionalities, model AC samples with comparable pore structures but different quantities of surface functionalities were required. Therefore, we controlled the oxygen contents using heat treatments in H2 or with KOH while maintaining the pore structure using an ultrahigh-surface-area AC, Maxsorb III [13], as the parent material. The model AC samples were used to investigate the influence of surface functionalities on adsorption performance, such as adsorption amounts and kinetics, as well as the adsorption states of the ethanol molecules in carbon micropores using solid-state nuclear magnetic resonance (NMR). NMR spectroscopy is based on physical phenomena of nuclei in a magnetic field absorbing and re-emitting electromagnetic radiation, and gives information about the number of magnetically distinct atoms of interested in it which has own nuclear spin states. However, spin states are not equivalent energy in an applied magnetic field, because the nucleus is a charged particle, and any moving charge can generates its own magnetic field. Thus, the nucleus has a magnetic moment, generated by its charge and spin. For instance, hydrogen nuclei can adopt only one or the other of orientations with respect to the applied field. The spin state +1/2 is lower energy since it is aligned with the field, while the spin state −1/2 is of higher energy since it is opposed to the applied field. The nuclear magnetic resonance phenomenon occurs when nuclei aligned with an applied field are induced to absorb energy and change their spin orientation with respect to the applied field. The energy absorption is a quantized process, and the energy absorbed must equal the energy difference between the two states (+1/2 or

−1/2). The protons in a molecule are surrounded by electrons and exist in slightly different electronic environments from one to another. The protons are shielded by the electrons that surround them.

Each proton in a molecule is shielded from the applied magnetic field to an extent that depends on the electron density surrounding it that means which induce different resonance frequency due to a slightly different chemical environment. These small differences of resonance as a chemical shift (δ,

32

(42)

ppm) are expressed with the amount by which a proton resonance is shifted from TMS (Tetramethylsilane), of the basic operating frequency. NMR can provide detailed information regarding the structure, dynamics, and chemical state of molecules and is a powerful method for analyzing adsorption states of molecules and ions [14]. Moreover, electrochemical techniques, cyclic voltammetry, and electrochemical impedance spectroscopy were applied to obtain diffusional information on the materials (electrolyte ions) in carbon micropores with different amounts of oxygen-containing surface functional groups. Based on these results, we examined the influence of surface functionalities on the adsorption behaviors of molecules in the carbon micropores.

33

(43)

2.2 Experimental

2.2.1. Sample preparation

The sample is chosen Maxsorb III which has the highest surface area (over 3,000 m²/g) and micropore volumes with an excellent adsorption capacity. Three AC samples were used in this study:

Maxsorb III from Kansai Coke and Chemicals Co., Ltd. (designated as MAX in this study), H2- treated Maxsorb III (with a lower content of oxygen-containing functional groups, named H-MAX), and KOH-activated H2-treated Maxsorb III (with higher contents of oxygen-containing functional groups, abbreviated as Ox-MAX). H-MAX was prepared using the high temperature vacuum furnace as shown Fig. 2-1. The heat treatment of MAX under a reducing atmosphere (Ar/H2 = 8/2 (v/v), total flow rate = 250 cm³/min) at 600°C for 24 h. To prepare Ox-MAX, potassium hydroxide (purity

>85.0%, Wako Pure Chemical Industries, Ltd.) was used as an activating agent. KOH treatment was applied to H-MAX at a KOH/carbon weight ratio (wt/wt) of 2. The mixture was placed in nickel crucible (ϕ 48 mm, L 100 mm), and then a nickel crucible was located in a stainless steel tube (ϕ 60 mm, L 500 mm) as shown Fig. 2-2. In the electric furnace, a stainless steel container was located vertically. The mixture was heat-treated up to the activation temperature at the heating rate of 5°C/min and it was maintained the activation temperature for 1 h under N2 flow. The activation temperature was 600°C. After treatment, the remaining KOH and salts formed during the heat treatment were removed by washing with HCl solution for three times and deionized water once to adjust pH to about 7. After washing, the collected samples were dried at 100°C for 3 h in an air oven and dried again at 150°C for 12 h in a vacuum oven.

34

(44)

Fig. 2-1. High temperature vacuum furnace.

Fig. 2-2. Schematic of KOH treatment system

35

(45)

2.2.2. Analysis

Elemental compositions of AC samples were analyzed using a CHN analyzer (MT-5, Yanako, Japan). The assay of O content (Odiff.) was defined by subtracting the sum of the contents of C, H, and N from 100%. N2 adsorption/desorption isotherms at 77 K were measured using volumetric adsorption equipment (Belsorp-Max-S, BEL Japan Inc., Japan) to investigate porosity.

Ethanol adsorption/desorption kinetics and isotherms were gravimetrically measured at 303 K.

Fig. 2-3 shows the schematic diagram of a gravimetric adsorption apparatus. A sample (about 20 mg with quartz wool) was placed in a sample basket (5) and into an adsorption chamber (3). Prior to adsorption measurements, the sample was pretreated at 150°C under vacuum (5 x 10^-4 Pa) for 2 h.

After cooling the sample, isothermal water (7) and air baths (8) were set to maintain the sample temperature at a constant 303 K. By opening valves, ethanol vapor was introduced into the adsorption chamber from a liquid ethanol vessel (2). After closing the valves, the weight change of the sample was recorded until it reached adsorption equilibrium. The ethanol vapor was introduced from 1 Torr by the first stage to 78 Torr by saturated vapor pressure of ethanol at 303 K. The introduced ethanol vapors have adjusted stage by stage. Here, equilibrium was considered to be achieved when the weight remained constant for 15 min. For the following measuring steps, the same valve operations were repeated and adsorption isotherms were obtained.

In solid-state NMR studies, two types of isotope-labeled ethanol (CD3CH2OH and CH3CH2OD) were used to assess adsorption states of ethanol molecules in carbon micropores. The AC sample was pretreated at 150°C under N2 flow for 2 h for degassing. After exposure to saturated CD3CH2OH (CDN Isotopes, 99.3 atom.% D) or CH3CH2OD (CDN Isotopes, 99.8 atom.% D) vapor in a N2-filled grove box for 72 h at room temperature, the sample was placed in a 3.2 mm NMR sample cell with sealing caps and then prepared for ¹H and ²H-NMR measurements. ¹H-NMR measurement was carried out by using an 800 MHz solid-state NMR (JNM-ECA-800, JEOL, Japan) with a cross

36

(46)

polarization/magic angle spinning probe. The basic sequence of the spin-echo pulse is ‘static − 90°

pulse − T1− single pulse’. The ¹H-NMR spectra were acquired at 6 μs of a 90° pulse length, 4 s of a relaxation delay, and 5 kHz of magic angle spinning (MAS) speed. Typically, 512 scans were performed to obtain a good signal-to-noise ratio. ²H single-pulse solid-state NMR spectra were acquired at a 5 kHz MAS speed at room temperature using an 800 MHz solid-state NMR with a cross polarization/MAS probe and same sample tube. The ²H-NMR spectra were acquired at 20 μs of a 90° pulse length, 2 s of a relaxation delay, and 5 kHz of MAS speed. Typically, 8192 scans were performed to obtain a good signal-to-noise ratio. Fig. 2-4 shows a photograph of an 800 MHz solid- state NMR apparatus.

Diffusional behaviors of electrolyte ions in micropores of the Maxsorb III series were studied using cyclic voltammetry and electrochemical impedance spectrometry using the conventional three- electrode system (HZ-3000 automatic polarization system, Hokuto Denko, Japan). Fig. 2-5 shows the schematic diagram of a home-made electrochemical cell. The working electrode was a dried thin film composed of a mixture of one of the Maxsorb III series and 15 wt.% polyvinylidene fluoride (SOLEF of Solvay Chemicals, Belgium) in N-methyl-2-pyrrolidone (Wako Pure Chemical Industries, Ltd., Japan). Pt and Ag/AgCl ([Cl^–] = 1 M) were used as counter and reference electrodes. All potentials were reported against the reference electrode. N2 gas was used to remove oxygen species in a 0.5 M H2SO4 (Wako Pure Chemical Industries, Ltd., Japan) electrolyte solution. Cyclic voltammetry was performed to investigate the capacitances depending on potential sweep rates, which control the kinetic rate of electron exchange between the charged working electrode and electrolyte molecules. In addition, impedance measurements were performed to examine resistive properties between the working electrode and electrolytic diffusion by applying a low-amplitude alternating voltage of 5 mV within the frequency range of 10 kHz to 100 MHz at an open circuit potential. In particular, diffusion behaviors of electrolyte molecules in carbon micropores with different amounts of surface functional groups were examined in this study.

37

(47)

Fig. 2-3. Schematic diagram of the gravimetric adsorption apparatus.

1. Pressure gauge, 2. Liquid ethanol vessel, 3. Adsorption chamber, 4. Quartz spring, 5. Sample basket, 6. Diffusion and rotary pumps, 7. Isothermal water bath, 8. Isothermal air bath.

38

(48)

Fig. 2-4. 800 MHz solid-state NMR (JNM-ECA-800, JEOL, Japan)

Fig. 2-5. Home-made electrochemical cell

39

(49)

2.3. Results and discussion

2.3.1. Analytical results of Maxsorb series

Fig. 2-6 compares the N2 adsorption and desorption isotherms at 77 K for MAX, H-MAX, and Ox-MAX. A steep increase in the nitrogen adsorption amount at very low relative pressure, which is a characteristic feature of microporous materials, was observed for all samples. The N2 adsorption and desorption isotherms showed negligible hysteresis, indicative of an absence of mesopores. Table 1-1 contains the pore structural parameters estimated using the subtracting pore-effect method with the α^S plots of the N2 adsorption isotherms at 77 K [15] and the elemental compositions obtained from the CHN analysis of the prepared model samples. Specific surface areas, micropore volumes, and average pore widths of the three samples were comparable, indicating that they had similar pore structures. On the other hand, oxygen contents differed among the samples. Compared with MAX (Odiff. = 4.35%), the oxygen content of H-MAX was reduced to 1.75% by heat treatment under a reducing atmosphere, but Ox-MAX showed a much higher oxygen content of 10.46%. Thus, these three samples were considered suitable for analyzing the influence of surface functionalities on the adsorption behavior of ethanol molecules in the carbon micropores.

Fig. 2-7 shows SEM images of Maxsorb III series. The morphology of model samples did not observe huge differences. The particle size was from tens to hundreds micrometers and the particle shape was shapeless and non-uniform. And then I measured apparent (pressureless) packing density of model samples by using the pressing mold. The result of measurement was nearly same (0.276 ~ 0.286 g/cm³). The uniform particle and spherical shape are considered to be increasing the packing density. It is necessary to find a particle mixture with a high packing density in a systematic for AHP, it is advantageous to improve the performance of AHP.

40

(50)

Fig. 2-6. (a) N2 adsorption/desorption isotherms and (b) isotherms on a log scale for the Maxsorb III series at 77 K.

(a)

(b)

41

(51)

Table 1-1. Pore structural parameters and elemental compositions of Maxsorb III series.

42

(52)

Fig. 2-7. SEM images of prepared ACs: (a) MAX, (b) H-MAX, and (c) Ox-MAX.

(a)

(b)

(c)

43

(53)

2.3.2 Ethanol adsorption properties of activated carbons

Ethanol adsorption/desorption kinetics and isotherms were gravimetrically measured at 303 K.

The ethanol adsorption measurements were performed twice to confirm repeatability. Fig. 2-8 shows ethanol adsorption/desorption isotherms and adsorption/desorption equilibrium times at different relative pressures (P/P0) at 303 K. The H2 treatment of MAX did not have a significant impact on the adsorption amounts and adsorption equilibrium times of ethanol. For Ox-MAX with an increased amount of oxygen-containing surface functionalities, significant decreases in the adsorption amounts of ethanol were observed (about 15% compared with MAX and H-MAX). Furthermore, the adsorption equilibrium times were shortened to within 40 min, which was significantly shorter than those of MAX and H-MAX, for the whole P/P0 region. These results suggest diffusional hindrance of ethanol molecules in micropores with abundant surface functionalities. That is, polar ethanol molecules were thought to rapidly and strongly adsorb on the surface functionalities located near the pore entrance, and such “trapped” ethanol molecules inhibited further adsorption at deeper pore spaces.

44

(54)

Fig. 2-8. (a) Ethanol adsorption/desorption isotherms and (b) adsorption/desorption equilibrium times of the Maxsorb III series at 303 K.

(a)

(b)

45

(55)

2.3.3 Ethanol adsorption properties of activated carbons

Fig. 2-9 shows ¹H-liquid NMR and ²H-liquid NMR spectra of the CH3CH2OD and the CD3CH2OH, respectively. In this result of ²H-liquid NMR, I confirmed the peaks of the hydroxyl group (OD) and methyl group (CD3) of isotope-labeled ethanol.

Figs. 2-10 and 2-11 show ²H-NMR spectra of CD3CH2OH-asdorbed and CH3CH2OD-asdorbed Maxsorb III series at a 5 kHz MAS speed, respectively. Regardless of the different oxygen contents, the shape of the NMR spectra of the CD3CH2OH-adsorbed samples were similar, although chemical shifts of the peak differed due to the deshielding effect of the oxygen-containing surface functional groups (Fig. 2-10) [16]. However, the CH3CH2OD-adsorbed samples showed different NMR spectra (Fig. 2-11). The ACs with higher oxygen contents yielded broader spectra, i.e., shorter spin-spin relaxation times, T2, which represents the lifetime of the signal in the transverse plane (XY plane);

the line width of an NMR signal depends on T2 (line width at half height = 1/T2). Slower motion of the observed atomic nucleus indicates a smaller T2 value. Therefore, the observed T2 suggests that hydroxyl groups of ethanol molecules interacted with the surface functionalities via hydrogen bonds.

Furthermore, peaks of spinning side bands were clearly observed for the CH3CH2OD-adsorbed samples (Fig. 2-11 (b)), suggesting that the D atoms of the adsorbed CH3CH2OD molecules were at relatively fixed states, at least during the time scale of the NMR measurements. Such peak broadening and appearance of spinning side band peaks were remarkable for Ox-MAX, with the highest oxygen content among the three samples used in this study. The NMR results were confirmed based on triplicate measurements. Taken together, these data indicate that ethanol molecules in micropores of ACs strongly interact with the surface functionalities and are in oriented adsorption states with the hydroxyl group facing the oxygen-containing surface functional groups.

Fig. 2-12 shows ¹H-NMR spectra of CD3CH2OH-asdorbed (a) and CH3CH2OD-asdorbed (b) Maxsorb III series at a 5 kHz MAS speed. The shapes of the 1H-NMR spectra of the CH3CH2OD-

46

(56)

adsorbed samples were similar (Fig. 2-12 (b)). However, the CD3CH2OH-adsorbed samples showed different NMR spectra (Fig. 2-12 (a)). The ACs with higher oxygen contents yielded broader spectra.

The hydroxyl groups of ethanol molecules strongly interacted with the surface functionalities.

Furthermore, peaks of spinning side bands were clearly observed for the both samples.

47

(57)

Fig. 2-9. (a) ¹H-liquid NMR and (b) ²H-liquid NMR spectra of the CH3CH2OD and CD3CH2OH.

(a)

(b)

48

(58)

Fig. 2-10. ²H-NMR spectra of the CD3CH2OH-adsorbed Maxsorb III series.

49

(59)

Fig. 2-11. ²H-NMR spectra of the CH3CH2OD-adsorbed Maxsorb III series for a (a) narrow chemical shift range and (b) wide chemical shift range.

(b) (a)

50

(60)

Fig. 2-12. ¹H-NMR spectra of the CDCH2OH-adsorbed (a) and the CH3CH2OD-adsorbed (b) Maxsorb III series.

(a)

(b)

51

(61)

2.3.4 Electrochemical analysis of activated carbon

Based on these results, it may be possible to study the influence of surface functionalities on the kinetics of molecules in the carbon micropores of the MAX series by using electrochemical analysis to determine the diffusional rates of electrolyte ions on working electrodes [17, 18]. Fig. 2-13 shows cyclic voltammograms (CVs) for the Maxsorb III series in 0.5 M H2SO4. In Fig. 2-13 (a), the capacitive current (non-faradaic current) of MAX showed the typical rectangular shape with a very small amount of pseudo-capacitive current (faradaic current) induced by a redox reaction of the oxygen-containing surface functional groups with electrolytic molecules, as shown in the potential range of 0.3 to 0.35 V [19-21]. In the case of H-MAX with lower oxygen content, the pseudo- capacitive current was negligible, while the rectangular shape of the CV curves was maintained, as shown in Fig. 2-13 (b). On the other hand, Ox-MAX with abundant oxygen functionalities showed a depressed rectangular shape compared to the other samples over the whole potential sweep range of –0.27 ~ 0.69 V, and the capacitance was much smaller than that of MAX or H-MAX (Fig. 2-13 (c)).

In addition, the characteristic pseudo-capacitive current was observed at 0.3 ~ 0.35 V. Fig. 2-13 (d) summarizes the specific capacitances of the Maxsorb III series derived from variable scan rates.

Especially for Ox-MAX, the specific capacitances showed a rapid decrease when the scan rate was increased from 1 to 200 mV/sec. These results suggest that the electrolyte ions do not easily diffuse within carbon micropores, where the surface is rich with surface functionalities.

52

(62)

(b) (a)

53

(63)

Fig. 2-13. Cyclic voltammograms of (a) MAX, (b) H-MAX, and (c) Ox-MAX measured at different scan rates in H2SO4. The specific capacitances for each scan rate were summarized in graph (d).

(c)

(d)

54

(64)

Fig. 2-14. Nyquist plots for the Maxsorb III series.

55

(65)

Fig. 2-14 shows the Nyquist plots for the Maxsorb III series. The equivalent series resistances can be explained by summing the electronic and ionic contributions. The former is associated with an intrinsic electronic resistance of the carbon material itself and the interfacial resistances of carbon particles to carbon particles and carbon particles to the current collector. The latter refers to the ionic (diffusion) resistance of electrolyte ions moving through the electrolyte solution to the narrow pores [22-24]. As shown in Fig. 2-13, the electronic resistances for the model ACs were similar (around 2.9 Ω), indicating that the carbonaceous structures of the Maxsorb III series were essentially comparable, even after the H2 and KOH treatments. However, especially in the low-frequency region associated with the diffusion of electrolyte ions in the pores, the slope for Ox-MAX, with the highest oxygen contents, was gentle compared with those for MAX or H-MAX, indicating that the diffusion of the electrolyte ions was hindered in carbon micropores with abundant surface functionalities.

Based on the electrochemical results of the CVs and electrochemical impedance spectroscopy, it was concluded that the oxygen-containing surface functional groups on the surface of micropores interrupt the diffusion of electrolyte ions in the narrow pores. These results agreed with ethanol adsorption/desorption isotherm measurements and solid-state NMR analysis for the Maxsorb III series, supporting our hypothesis, even though different adsorptive materials were used (ethanol and electrolyte ions in H2SO4).

56

九州大学学術情報リポジトリ