- Email: [email protected]
Study of influence on the surface energy heterogeneity of multiwalled carbon nanotubes after the adsorption of poly(acrylic acid)
Journal of Colloid and Interface Science 278 (2004) 299–303 www.elsevier.com/locate/jcis
Study of influence on the surface energy heterogeneity of multiwalled carbon nanotubes after the adsorption of poly(acrylic acid) Qingfeng Hou a , Xiancai Lu b,∗ , Xiandong Liu b , Baixing Hu a , Jian Shen a a Research Center of Surface and Interface Chemical Engineering Technology, Nanjing University, Nanjing 210093, People’s Republic of China b The State Key Laboratory of Mineral Deposit Research, Department of Earth Sciences, Nanjing University, Nanjing 210093, People’s Republic of China
Received 15 March 2004; accepted 20 May 2004 Available online 3 August 2004
Abstract The heterogeneity of the surface energy of multiwalled carbon nanotubes (MWNTs) before and after the adsorption of poly(acrylic acid) (PAA) at 77 K was investigated by a nitrogen probe adsorption technique in a wide range of pressures. The adsorption energy distributions (AEDs) were calculated from the low-pressure data of isotherms (i.e., the data of submonolayer adsorption) using the regularization method. Based on the AED, two types of dominant energetic surfaces are identified and assigned to the graphite-like carbon and disordered carbon or defects in the pure MWNTs, respectively. While the adsorption amount of PAA is raised, a significant decrease in the contribution of higher-energy surface in AEDs is observed for those PAA-adsorbed MWNTs. It is thus demonstrated that PAA prefers interacting with higher-energy surfaces to lower-energy surfaces in MWNTs. Nitrogen probe adsorption measurements including low-pressure data are shown to be a feasible and effective tool to characterize the heterogeneity of structure and surface properties of porous materials. 2004 Elsevier Inc. All rights reserved. Keywords: Multiwalled carbon nanotubes; Surface energy heterogeneity; Adsorption; Surface properties
1. Introduction The discovery of multiwalled carbon nanotubes (MWNTs) by Iijima [1] opened up a broad scope of research in their synthesis, purification, structure characterization, and application in various fields [2–6]. The unique structure of crystalline tubes with inner pore size 1–20 nm and large surface area is important for potential practical applications such as adsorbent [7,8]. The heterogeneity of the energy and structure of MWNTs due to various surface groups and impurities and irregularities in nanoporous size and shape generally influence their performance in applications [9]. The coating and/or adsorption of some organosilanes, especially polymers, could change the surface properties of porous materials, which has been proved in many adsorbents including mesoporous silicas and microporous carbons * Corresponding author. Fax: +86-25-83686016.
E-mail address: [email protected] (X. Lu). 0021-9797/$ – see front matter 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.jcis.2004.05.047
[10–12]. Accordingly, the surface treatments of some polymers create opportunities to modify the surface heterogeneity of MWNTs, which is of great value in the development and application of MWNTs. Poly(acrylic acid) (PAA) is selected in our investigations; it is soluble in water and can be prepared in a wide range of chain lengths. Moreover, the fact that PAA is a homopolymer simplifies the interpretation of the results. The present work may facilitate our understanding of the similar surface–interface interaction between MWNTs and polymers. The heterogeneity of surface energy is a key point determining many surface–interface phenomena of MWNTs in various applications. It has been proved that MWNTs generally consist of two energy states of carbon, graphite-like carbon and disordered carbon or defects [13,14]. The content, distribution, and chemical structure of the latter affect many surface properties of MWNTs [15,16]. However, few investigations about heterogeneity of surface energy of MWNTs have been reported as yet because there were no proper methods/techniques available to accurately characterize the surface energy in the past.
300
Q. Hou et al. / Journal of Colloid and Interface Science 278 (2004) 299–303
Currently, probing gas adsorption isotherms, especially lowpressure data, have proved to be a reliable technique for characterizing not only the porous structure but also the heterogeneity of surface energy of the MWNTs [17,18]. This paper reports a detailed study on the change in heterogeneity of surface energy of MWNTs before and after the adsorption of PAA. Low-temperature nitrogen adsorption data were used to calculate the specific surface area, pore size distribution (PSD), and adsorption energy distribution (AED).
2. Experimental 2.1. Materials and sample preparation The solid adsorbents were lab-synthesized MWNTs produced by catalytic decomposition of benzene vapor and reported in detail elsewhere [19]. The pristine MWNTs were treated by the following two-step purification process: acid reflux with 5 M HCl at 413 K for 5 h. After the acid treatment, the resulting suspensions were centrifuged and the supernatant acids were decanted off. The sediments were resuspended in deionized water, centrifuged three times, decanted, vacuum-filtered, and finally dried in a vacuum oven at 313 K. PAA was obtained by polymerization of acrylic acid in aqueous solution at 343 K with about 30% (NH4 )2 S2 O2 as initiator. Number-average molar mass (Mn ), weight-average molar mass (Mw ), and polydispersity index (Mw /Mn ) for the PAA samples were 7.073 × 104 , 9.075 × 104, and 1.283, respectively. They were measured on the miniDAWN multiangle laser scatter analyzer (Wyatt, CA) with ASTRA software (version 4.90.07). Five treated MWNTs (MWNTs-1, MWNTs-2, MWNTs-3, MWNTs-4, MWNTs-5) with different adsorption amounts of PAA were obtained by adding five shares of 2.0 g MWNTs into five kinds of PAA solutions with different concentrations. The solid–liquid mixtures were allowed to reach an equilibrium state for about 48 h and subsequently filtered. The isolated MWNTs were dried in a vacuum oven at 313 K for about 48 h. The adsorption amount of PAA was determined from the concentration difference of the solution before and after the treatment. The related parameters of the examined MWNTs, including one pure sample (MWNTs-0) as a reference, are listed in Table 1. 2.2. Measurements Low-temperature nitrogen adsorption–desorption measurements were carried out on an ASAP2010 volumetric adsorption analyzer from Micromeritics (Norcross, GA, USA) at liquid-nitrogen temperature (77 K) in a relative pressure range from about 1 × 106 to 0.995 using about 0.5000g of
Table 1 Basic parameters of the experimental samples Samples
CPAA (mg/ml)
ΓPAA (mg/g)
SBET (m2 /g)
Vp (m3 /g)
MWNTs-0 MWNTs-1 MWNTs-2 MWNTs-3 MWNTs-4 MWNTs-5
0 0.16 0.48 0.96 1.60 8.00
0 13.09 43.35 89.60 146.53 268.98
216.57 192.49 182.43 169.32 152.49 114.85
1.21 0.92 0.86 0.56 0.51 0.43
Note. CPAA denotes the concentration of PAA solution. Γ denotes the adsorption amount of PAA on the examined MWNTs. SBET denotes the specific surface area calculated according to the BET method. Vp denotes maximum pore volume of the examined MWNTs.
experimental samples. Prior to isotherm measurements, each sample was outgassed to a pressure of less than 0.02 Pa at 313 K for about 12 h. 2.3. Calculation methods 2.3.1. PSD and specific surface area The PSD was calculated from the low-temperature nitrogen adsorption isotherm over the entire pressure range using the DFT model proposed by Olivier et al. with the latest software DFT plus Version 2.00, the detailed information on which has been discussed elsewhere [20,21]. The specific surface area was calculated according to the BET equation from the adsorption isotherm branch of the relative pressure range from 0.050 to 0.200. 2.3.2. Calculation of the AED The low-pressure segments of the adsorption isotherms were utilized to calculate the AED of the examined MWNTs with a method called INTEG [22,23], which inverts an integral equation analogous to θ (p) = θt (p, U )F (U ) dU, (1)
where θ (p) denotes the relative adsorption amount in the submonolayer range. U is the adsorption energy, θt (p, U ) is the energy-dependent local adsorption isotherm, is the integration region, and F (U ) dU denotes the fraction of the surface with adsorption energies between U and U + dU . To invert the integral in Eq. (1) with respect to the AED function F (U ), one is forced to assume a model for the local adsorption. In current work, the Fowler–Guggenheim (FG) equation describes the localized monolayer adsorption with lateral interactions of nitrogen molecules on the surfaces with a patchwise distribution of adsorption sites [17,23], which appears to be realistic for the description of adsorption of nitrogen on the surface of carbons [24,25], θ (p, U ) =
Kp exp(zωθt ) , 1 + Kp exp(zωθt )
(2)
where K = K0 (T ) exp(U/kT ),
(3)
Q. Hou et al. / Journal of Colloid and Interface Science 278 (2004) 299–303
K0 =
N0 σ 0 γ 0 , (2πMRT )1/2
(4)
where K stands for the Langmuir constant for adsorption on monoenergetic sites and the preexponential factor K0 (T ) is expressed in terms of the partition functions for an isolated molecule in the gas and surface phases. The above factor is estimated according to the Adamson method. Here σ 0 denotes the actual area per nitrogen molecule, 0.162 nm2 , γ 0 is Frenkel’s characteristic adsorption time, which is assumed to be 10−13 s in the current study, N0 is Avogadro’s constant, and M is the molecular weight of the nitrogen adsorbate, 28. The nitrogen AEDs for MWNTs examined here were calculated by assuming z = 2 and w/k = 95 K [23].
3. Results and discussions 3.1. PSD and specific surface area The low-temperature nitrogen adsorption–desorption isotherms of the samples are shown in Fig. 1 on linear scales, which are identified as type IV isotherms according to the IUPAC classification [8]. The obvious hysteresis loops reflect a rather broad distribution of mesopores in these samples. All of the isotherms exhibit similar shapes only with
Fig. 1. Nitrogen adsorption–desorption isotherms for the examined MWNTs.
301
slight shifts in adsorption capacities. The PSDs of the examined MWNTs are shown in Fig. 2. Two types of mesopores can be classified clearly in the pure MWNTs (MWNTs-0). The peaks in the range of 1–2 nm can be assigned as the intraparticle pores, a kind of typical cylinder-shaped mesopores of hollow MWNTs [26]. The other broad assembled peaks at 3–140 nm can be assigned as the interparticle pores, which result from the aggregation among the carbon nanotubes [26]. The width of interparticle pores decreases as a whole and those pores wider than 100 nm in the untreated MWNTs-0 decrease remarkably after the adsorption of PAA, which can be attributed to the strong aggregation among the carbon nanotubes caused by the PAA treatment, as we observed in the experiment. As the adsorption amount of PAA increases, the intraparticle pores in the pure MWNTs decrease and a few new pores are formed at about 2 nm, which could be attributed to the pore-blocking effect of the adsorbed PAA on both the intraparticle and smaller interparticle pores [24,27]. Accordingly, as shown in Table 1, the maximum pore volume of the examined MWNTs decreases gradually from 1.21 to 0.43 cm3 /g because of the change in the volume of interparticle pores, and the specific surface area decreases from 216.57 to 114.85 m2 /g, while the adsorption amount of PAA increases from 0 to 268.98 mg/g, which can be attributed to surface screening effect and strong aggregation. Similar phenomena for several kinds of porous materials have been reported in other literature [24,27,28].
Fig. 2. DFT pore size distribution for the examined MWNTs.
302
Q. Hou et al. / Journal of Colloid and Interface Science 278 (2004) 299–303
nearly twice as high as the low-energy one in MWNTs-0, which means the active surface may be the dominant surface of pure MWNTs. Remarkable changes in the AEDs of MWNTs are observed in those PAA-adsorbed samples. While the adsorption amount of PAA increases, the high-energy domain declines gradually (MWNTs-1, MWNTs-2, and MWNTs-3) and ultimately disappears (MWNTs-4 and MWNTs-5), whereas the low-energy domain increases slightly. The gradual shielding of active surface in the AEDs for PAAadsorbed MWNTs suggests that PAA molecules prefer mounting on the surface of disordered carbon or defects to that of graphite-like carbon in MWNTs. The variation in AEDs discussed above reflects that the heterogeneity of surface energy and average surface energy of MWNTs decline due to the adsorption of PAA.
4. Conclusion
Fig. 3. Adsorption energy distribution for the examined MWNTs. The functions were calculated with the regularization method using the same regularization parameter, γ = 0.01.
3.2. Heterogeneity of surface energy The low-pressure range of nitrogen adsorption isotherms, especially the data representing the submonolayer adsorption stage, is essential for extracting information about the heterogeneity of surface energy of the examined solid [25]. The AEDs of all the examined MWNTs are shown in Fig. 3. The AED of MWNTs-0 exhibits two peaks: a highenergy domain in the range of 12–14 kJ/mol and a lowenergy domain in the range of 9–10 kJ/mol. The presence of the two domains suggests the existence of two types of energetic surface in the untreated MWNTs, which relates to the graphite-like carbon and disordered carbon or defects presented on the surface of MWNTs respectively. Generally, graphitization is a process of lowering surface energy of carbon materials; i.e., the surface energy of graphite-like carbon is lower than that of disordered carbon or defects. Accordingly, the high-energy domain represents disordered carbon or defects, whereas the low-energy one represents the other one. It should be noted that the high-energy domain is
The surface properties of MWNTs were clearly modified by the adsorption of PAA. Both the specific surface area and the maximum pore volume of the examined MWNTs decrease gradually. Two types of energy surfaces of the examined MWNTs, which represent the graphite-like carbon and disordered carbon or defects, respectively, are identified by the calculation of AEDs exactly. Systematic changes due to the adsorption of PAA are observed in the AEDs of the examined MWNTs. The heterogeneity of surface energy of MWNTs gradually decline while the adsorption amount of PAA increases. The nitrogen adsorption isotherm is proved a suitable method for characterizing the heterogeneity of surface energy of novel nanostructured materials such as MWNTs, and it could facilitate investigations on the surface–interface phenomena occurring on these materials.
Acknowledgments This work is supported by the National Natural Science Foundation of China (Grants 40003002 and 40373024).
References [1] S. Iijima, Nature 354 (1991) 56. [2] Q.H. Yang, S. Bai, T. Fournier, F. Li, G. Wang, H.M. Cheng, J.B. Bai, Chem. Phys. Lett. 370 (2003) 274. [3] P.X. Hou, S. Bai, Q.H. Yang, C. Liu, H.M. Cheng, Carbon 40 (2002) 81. [4] C. Liu, Q.H. Yang, Y. Tong, H.T. Cong, H.M. Cheng, Appl. Phys. Lett. 80 (2002) 2389. [5] P.M. Ajayan, Chem. Rev. 99 (1999) 1787. [6] S. Inoue, N. Ichikuni, T. Suzuki, T. Uematsu, K. Kaneko, J. Phys. Chem. B 102 (1998) 4689. [7] A. Fujiwara, K. Ishii, H. Suematsu, H. Kataura, Y. Maniwa, S. Suzuki, Y. Achiba, Chem. Phys. Lett. 336 (2001) 205.
Q. Hou et al. / Journal of Colloid and Interface Science 278 (2004) 299–303
[8] Q.H. Yang, P.X. Hou, S. Bai, M.Z. Wang, H.M. Cheng, Chem. Phys. Lett. 345 (2001) 18. [9] X. Zhao, Y. Ando, L.C. Qin, H. Kataura, Y. Maniwa, R. Saito, Physica B 323 (2002) 265. [10] A. Bose, R.K. Gilpin, M. Jaroniec, J. Colloid Interface Sci. 240 (2001) 224. [11] M. Kruk, M. Jaroniec, Micropor. Mesopor. Mat. 44 (2001) 725. [12] I. Sondi, O. Milat, V. Pravdic, J. Colloid Interface Sci. 189 (1997) 66. [13] A. Allaoui, S. Bai, H.M. Cheng, J.B. Bai, Compos. Sci. Technol. 62 (2002) 1993. [14] S. Lefrant, J.P. Buisson, J. Schreiber, O. Chauvet, M. Baibarac, I. Baltog, Synth. Methods 139 (2003) 783. [15] S. Lefrant, Curr. Appl. Phys. 2 (2002) 479. [16] P.H. Tan, C.Y. Hu, F. Li, S. Bai, P.X. Hou, H.M. Cheng, Carbon 40 (2002) 1131. [17] J. Choma, M. Jaroniec, Langmuir 13 (1997) 1026. [18] M. Kruk, M. Jaroniec, J. Choma, Carbon 36 (1998) 1447.
303
[19] X.Z. Wang, Z. Hu, Q. Wu, Y. Chen, Chin. Phys. 10 (2001) S76. [20] J.P. Olivier, J. Porous Mater. 2 (1995) 9. [21] M.L. Occelli, J.P. Olivier, J.A. Perdigon-Melon, A. Auroux, Langmuir 18 (2002) 9816. [22] M.V. Szombathely, P. Brauer, M. Jaroniec, J. Comput. Chem. 13 (1992) 17. [23] M. Heuchel, M. Jaroniec, R.K. Gilpin, P. Braeuer, M.V. Szombathely, Langmuir 9 (1993) 2537. [24] M. Jaroniec, R. Madey, Physical Adsorption on Heterogeneous Solid, Elsevier, Amsterdam, 1998. [25] Y. Bereznitski, M. Jaroniec, K.P. Gadkaree, Adsorption 3 (1997) 277. [26] S. Inoue, N. Ichikuni, T. Suzuki, T. Uematsu, K. Kaneko, J. Phys. Chem. B 102 (1998) 4689. [27] J.F. Lee, C.K. Lee, L.C. Juang, J. Colloid Interface Sci. 217 (1999) 172. [28] C.P. Jaroniec, M. Kruk, M. Jaroniec, A. Sayari, J. Phys. Chem. B 102 (1998) 5503.
Study of influence on the surface energy heterogeneity of multiwalled carbon nanotubes after the adsorption of poly(acrylic acid) Qingfeng Hou a , Xiancai Lu b,∗ , Xiandong Liu b , Baixing Hu a , Jian Shen a a Research Center of Surface and Interface Chemical Engineering Technology, Nanjing University, Nanjing 210093, People’s Republic of China b The State Key Laboratory of Mineral Deposit Research, Department of Earth Sciences, Nanjing University, Nanjing 210093, People’s Republic of China
Received 15 March 2004; accepted 20 May 2004 Available online 3 August 2004
Abstract The heterogeneity of the surface energy of multiwalled carbon nanotubes (MWNTs) before and after the adsorption of poly(acrylic acid) (PAA) at 77 K was investigated by a nitrogen probe adsorption technique in a wide range of pressures. The adsorption energy distributions (AEDs) were calculated from the low-pressure data of isotherms (i.e., the data of submonolayer adsorption) using the regularization method. Based on the AED, two types of dominant energetic surfaces are identified and assigned to the graphite-like carbon and disordered carbon or defects in the pure MWNTs, respectively. While the adsorption amount of PAA is raised, a significant decrease in the contribution of higher-energy surface in AEDs is observed for those PAA-adsorbed MWNTs. It is thus demonstrated that PAA prefers interacting with higher-energy surfaces to lower-energy surfaces in MWNTs. Nitrogen probe adsorption measurements including low-pressure data are shown to be a feasible and effective tool to characterize the heterogeneity of structure and surface properties of porous materials. 2004 Elsevier Inc. All rights reserved. Keywords: Multiwalled carbon nanotubes; Surface energy heterogeneity; Adsorption; Surface properties
1. Introduction The discovery of multiwalled carbon nanotubes (MWNTs) by Iijima [1] opened up a broad scope of research in their synthesis, purification, structure characterization, and application in various fields [2–6]. The unique structure of crystalline tubes with inner pore size 1–20 nm and large surface area is important for potential practical applications such as adsorbent [7,8]. The heterogeneity of the energy and structure of MWNTs due to various surface groups and impurities and irregularities in nanoporous size and shape generally influence their performance in applications [9]. The coating and/or adsorption of some organosilanes, especially polymers, could change the surface properties of porous materials, which has been proved in many adsorbents including mesoporous silicas and microporous carbons * Corresponding author. Fax: +86-25-83686016.
E-mail address: [email protected] (X. Lu). 0021-9797/$ – see front matter 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.jcis.2004.05.047
[10–12]. Accordingly, the surface treatments of some polymers create opportunities to modify the surface heterogeneity of MWNTs, which is of great value in the development and application of MWNTs. Poly(acrylic acid) (PAA) is selected in our investigations; it is soluble in water and can be prepared in a wide range of chain lengths. Moreover, the fact that PAA is a homopolymer simplifies the interpretation of the results. The present work may facilitate our understanding of the similar surface–interface interaction between MWNTs and polymers. The heterogeneity of surface energy is a key point determining many surface–interface phenomena of MWNTs in various applications. It has been proved that MWNTs generally consist of two energy states of carbon, graphite-like carbon and disordered carbon or defects [13,14]. The content, distribution, and chemical structure of the latter affect many surface properties of MWNTs [15,16]. However, few investigations about heterogeneity of surface energy of MWNTs have been reported as yet because there were no proper methods/techniques available to accurately characterize the surface energy in the past.
300
Q. Hou et al. / Journal of Colloid and Interface Science 278 (2004) 299–303
Currently, probing gas adsorption isotherms, especially lowpressure data, have proved to be a reliable technique for characterizing not only the porous structure but also the heterogeneity of surface energy of the MWNTs [17,18]. This paper reports a detailed study on the change in heterogeneity of surface energy of MWNTs before and after the adsorption of PAA. Low-temperature nitrogen adsorption data were used to calculate the specific surface area, pore size distribution (PSD), and adsorption energy distribution (AED).
2. Experimental 2.1. Materials and sample preparation The solid adsorbents were lab-synthesized MWNTs produced by catalytic decomposition of benzene vapor and reported in detail elsewhere [19]. The pristine MWNTs were treated by the following two-step purification process: acid reflux with 5 M HCl at 413 K for 5 h. After the acid treatment, the resulting suspensions were centrifuged and the supernatant acids were decanted off. The sediments were resuspended in deionized water, centrifuged three times, decanted, vacuum-filtered, and finally dried in a vacuum oven at 313 K. PAA was obtained by polymerization of acrylic acid in aqueous solution at 343 K with about 30% (NH4 )2 S2 O2 as initiator. Number-average molar mass (Mn ), weight-average molar mass (Mw ), and polydispersity index (Mw /Mn ) for the PAA samples were 7.073 × 104 , 9.075 × 104, and 1.283, respectively. They were measured on the miniDAWN multiangle laser scatter analyzer (Wyatt, CA) with ASTRA software (version 4.90.07). Five treated MWNTs (MWNTs-1, MWNTs-2, MWNTs-3, MWNTs-4, MWNTs-5) with different adsorption amounts of PAA were obtained by adding five shares of 2.0 g MWNTs into five kinds of PAA solutions with different concentrations. The solid–liquid mixtures were allowed to reach an equilibrium state for about 48 h and subsequently filtered. The isolated MWNTs were dried in a vacuum oven at 313 K for about 48 h. The adsorption amount of PAA was determined from the concentration difference of the solution before and after the treatment. The related parameters of the examined MWNTs, including one pure sample (MWNTs-0) as a reference, are listed in Table 1. 2.2. Measurements Low-temperature nitrogen adsorption–desorption measurements were carried out on an ASAP2010 volumetric adsorption analyzer from Micromeritics (Norcross, GA, USA) at liquid-nitrogen temperature (77 K) in a relative pressure range from about 1 × 106 to 0.995 using about 0.5000g of
Table 1 Basic parameters of the experimental samples Samples
CPAA (mg/ml)
ΓPAA (mg/g)
SBET (m2 /g)
Vp (m3 /g)
MWNTs-0 MWNTs-1 MWNTs-2 MWNTs-3 MWNTs-4 MWNTs-5
0 0.16 0.48 0.96 1.60 8.00
0 13.09 43.35 89.60 146.53 268.98
216.57 192.49 182.43 169.32 152.49 114.85
1.21 0.92 0.86 0.56 0.51 0.43
Note. CPAA denotes the concentration of PAA solution. Γ denotes the adsorption amount of PAA on the examined MWNTs. SBET denotes the specific surface area calculated according to the BET method. Vp denotes maximum pore volume of the examined MWNTs.
experimental samples. Prior to isotherm measurements, each sample was outgassed to a pressure of less than 0.02 Pa at 313 K for about 12 h. 2.3. Calculation methods 2.3.1. PSD and specific surface area The PSD was calculated from the low-temperature nitrogen adsorption isotherm over the entire pressure range using the DFT model proposed by Olivier et al. with the latest software DFT plus Version 2.00, the detailed information on which has been discussed elsewhere [20,21]. The specific surface area was calculated according to the BET equation from the adsorption isotherm branch of the relative pressure range from 0.050 to 0.200. 2.3.2. Calculation of the AED The low-pressure segments of the adsorption isotherms were utilized to calculate the AED of the examined MWNTs with a method called INTEG [22,23], which inverts an integral equation analogous to θ (p) = θt (p, U )F (U ) dU, (1)
where θ (p) denotes the relative adsorption amount in the submonolayer range. U is the adsorption energy, θt (p, U ) is the energy-dependent local adsorption isotherm, is the integration region, and F (U ) dU denotes the fraction of the surface with adsorption energies between U and U + dU . To invert the integral in Eq. (1) with respect to the AED function F (U ), one is forced to assume a model for the local adsorption. In current work, the Fowler–Guggenheim (FG) equation describes the localized monolayer adsorption with lateral interactions of nitrogen molecules on the surfaces with a patchwise distribution of adsorption sites [17,23], which appears to be realistic for the description of adsorption of nitrogen on the surface of carbons [24,25], θ (p, U ) =
Kp exp(zωθt ) , 1 + Kp exp(zωθt )
(2)
where K = K0 (T ) exp(U/kT ),
(3)
Q. Hou et al. / Journal of Colloid and Interface Science 278 (2004) 299–303
K0 =
N0 σ 0 γ 0 , (2πMRT )1/2
(4)
where K stands for the Langmuir constant for adsorption on monoenergetic sites and the preexponential factor K0 (T ) is expressed in terms of the partition functions for an isolated molecule in the gas and surface phases. The above factor is estimated according to the Adamson method. Here σ 0 denotes the actual area per nitrogen molecule, 0.162 nm2 , γ 0 is Frenkel’s characteristic adsorption time, which is assumed to be 10−13 s in the current study, N0 is Avogadro’s constant, and M is the molecular weight of the nitrogen adsorbate, 28. The nitrogen AEDs for MWNTs examined here were calculated by assuming z = 2 and w/k = 95 K [23].
3. Results and discussions 3.1. PSD and specific surface area The low-temperature nitrogen adsorption–desorption isotherms of the samples are shown in Fig. 1 on linear scales, which are identified as type IV isotherms according to the IUPAC classification [8]. The obvious hysteresis loops reflect a rather broad distribution of mesopores in these samples. All of the isotherms exhibit similar shapes only with
Fig. 1. Nitrogen adsorption–desorption isotherms for the examined MWNTs.
301
slight shifts in adsorption capacities. The PSDs of the examined MWNTs are shown in Fig. 2. Two types of mesopores can be classified clearly in the pure MWNTs (MWNTs-0). The peaks in the range of 1–2 nm can be assigned as the intraparticle pores, a kind of typical cylinder-shaped mesopores of hollow MWNTs [26]. The other broad assembled peaks at 3–140 nm can be assigned as the interparticle pores, which result from the aggregation among the carbon nanotubes [26]. The width of interparticle pores decreases as a whole and those pores wider than 100 nm in the untreated MWNTs-0 decrease remarkably after the adsorption of PAA, which can be attributed to the strong aggregation among the carbon nanotubes caused by the PAA treatment, as we observed in the experiment. As the adsorption amount of PAA increases, the intraparticle pores in the pure MWNTs decrease and a few new pores are formed at about 2 nm, which could be attributed to the pore-blocking effect of the adsorbed PAA on both the intraparticle and smaller interparticle pores [24,27]. Accordingly, as shown in Table 1, the maximum pore volume of the examined MWNTs decreases gradually from 1.21 to 0.43 cm3 /g because of the change in the volume of interparticle pores, and the specific surface area decreases from 216.57 to 114.85 m2 /g, while the adsorption amount of PAA increases from 0 to 268.98 mg/g, which can be attributed to surface screening effect and strong aggregation. Similar phenomena for several kinds of porous materials have been reported in other literature [24,27,28].
Fig. 2. DFT pore size distribution for the examined MWNTs.
302
Q. Hou et al. / Journal of Colloid and Interface Science 278 (2004) 299–303
nearly twice as high as the low-energy one in MWNTs-0, which means the active surface may be the dominant surface of pure MWNTs. Remarkable changes in the AEDs of MWNTs are observed in those PAA-adsorbed samples. While the adsorption amount of PAA increases, the high-energy domain declines gradually (MWNTs-1, MWNTs-2, and MWNTs-3) and ultimately disappears (MWNTs-4 and MWNTs-5), whereas the low-energy domain increases slightly. The gradual shielding of active surface in the AEDs for PAAadsorbed MWNTs suggests that PAA molecules prefer mounting on the surface of disordered carbon or defects to that of graphite-like carbon in MWNTs. The variation in AEDs discussed above reflects that the heterogeneity of surface energy and average surface energy of MWNTs decline due to the adsorption of PAA.
4. Conclusion
Fig. 3. Adsorption energy distribution for the examined MWNTs. The functions were calculated with the regularization method using the same regularization parameter, γ = 0.01.
3.2. Heterogeneity of surface energy The low-pressure range of nitrogen adsorption isotherms, especially the data representing the submonolayer adsorption stage, is essential for extracting information about the heterogeneity of surface energy of the examined solid [25]. The AEDs of all the examined MWNTs are shown in Fig. 3. The AED of MWNTs-0 exhibits two peaks: a highenergy domain in the range of 12–14 kJ/mol and a lowenergy domain in the range of 9–10 kJ/mol. The presence of the two domains suggests the existence of two types of energetic surface in the untreated MWNTs, which relates to the graphite-like carbon and disordered carbon or defects presented on the surface of MWNTs respectively. Generally, graphitization is a process of lowering surface energy of carbon materials; i.e., the surface energy of graphite-like carbon is lower than that of disordered carbon or defects. Accordingly, the high-energy domain represents disordered carbon or defects, whereas the low-energy one represents the other one. It should be noted that the high-energy domain is
The surface properties of MWNTs were clearly modified by the adsorption of PAA. Both the specific surface area and the maximum pore volume of the examined MWNTs decrease gradually. Two types of energy surfaces of the examined MWNTs, which represent the graphite-like carbon and disordered carbon or defects, respectively, are identified by the calculation of AEDs exactly. Systematic changes due to the adsorption of PAA are observed in the AEDs of the examined MWNTs. The heterogeneity of surface energy of MWNTs gradually decline while the adsorption amount of PAA increases. The nitrogen adsorption isotherm is proved a suitable method for characterizing the heterogeneity of surface energy of novel nanostructured materials such as MWNTs, and it could facilitate investigations on the surface–interface phenomena occurring on these materials.
Acknowledgments This work is supported by the National Natural Science Foundation of China (Grants 40003002 and 40373024).
References [1] S. Iijima, Nature 354 (1991) 56. [2] Q.H. Yang, S. Bai, T. Fournier, F. Li, G. Wang, H.M. Cheng, J.B. Bai, Chem. Phys. Lett. 370 (2003) 274. [3] P.X. Hou, S. Bai, Q.H. Yang, C. Liu, H.M. Cheng, Carbon 40 (2002) 81. [4] C. Liu, Q.H. Yang, Y. Tong, H.T. Cong, H.M. Cheng, Appl. Phys. Lett. 80 (2002) 2389. [5] P.M. Ajayan, Chem. Rev. 99 (1999) 1787. [6] S. Inoue, N. Ichikuni, T. Suzuki, T. Uematsu, K. Kaneko, J. Phys. Chem. B 102 (1998) 4689. [7] A. Fujiwara, K. Ishii, H. Suematsu, H. Kataura, Y. Maniwa, S. Suzuki, Y. Achiba, Chem. Phys. Lett. 336 (2001) 205.
Q. Hou et al. / Journal of Colloid and Interface Science 278 (2004) 299–303
[8] Q.H. Yang, P.X. Hou, S. Bai, M.Z. Wang, H.M. Cheng, Chem. Phys. Lett. 345 (2001) 18. [9] X. Zhao, Y. Ando, L.C. Qin, H. Kataura, Y. Maniwa, R. Saito, Physica B 323 (2002) 265. [10] A. Bose, R.K. Gilpin, M. Jaroniec, J. Colloid Interface Sci. 240 (2001) 224. [11] M. Kruk, M. Jaroniec, Micropor. Mesopor. Mat. 44 (2001) 725. [12] I. Sondi, O. Milat, V. Pravdic, J. Colloid Interface Sci. 189 (1997) 66. [13] A. Allaoui, S. Bai, H.M. Cheng, J.B. Bai, Compos. Sci. Technol. 62 (2002) 1993. [14] S. Lefrant, J.P. Buisson, J. Schreiber, O. Chauvet, M. Baibarac, I. Baltog, Synth. Methods 139 (2003) 783. [15] S. Lefrant, Curr. Appl. Phys. 2 (2002) 479. [16] P.H. Tan, C.Y. Hu, F. Li, S. Bai, P.X. Hou, H.M. Cheng, Carbon 40 (2002) 1131. [17] J. Choma, M. Jaroniec, Langmuir 13 (1997) 1026. [18] M. Kruk, M. Jaroniec, J. Choma, Carbon 36 (1998) 1447.
303
[19] X.Z. Wang, Z. Hu, Q. Wu, Y. Chen, Chin. Phys. 10 (2001) S76. [20] J.P. Olivier, J. Porous Mater. 2 (1995) 9. [21] M.L. Occelli, J.P. Olivier, J.A. Perdigon-Melon, A. Auroux, Langmuir 18 (2002) 9816. [22] M.V. Szombathely, P. Brauer, M. Jaroniec, J. Comput. Chem. 13 (1992) 17. [23] M. Heuchel, M. Jaroniec, R.K. Gilpin, P. Braeuer, M.V. Szombathely, Langmuir 9 (1993) 2537. [24] M. Jaroniec, R. Madey, Physical Adsorption on Heterogeneous Solid, Elsevier, Amsterdam, 1998. [25] Y. Bereznitski, M. Jaroniec, K.P. Gadkaree, Adsorption 3 (1997) 277. [26] S. Inoue, N. Ichikuni, T. Suzuki, T. Uematsu, K. Kaneko, J. Phys. Chem. B 102 (1998) 4689. [27] J.F. Lee, C.K. Lee, L.C. Juang, J. Colloid Interface Sci. 217 (1999) 172. [28] C.P. Jaroniec, M. Kruk, M. Jaroniec, A. Sayari, J. Phys. Chem. B 102 (1998) 5503.