- Email: [email protected]
Methane preconcentration in a microtrap using multiwalled carbon nanotubes as sorbents
Analytica Chimica Acta 677 (2010) 50–54
Contents lists available at ScienceDirect
Analytica Chimica Acta journal homepage: www.elsevier.com/locate/aca
Methane preconcentration in a microtrap using multiwalled carbon nanotubes as sorbents Chutarat Saridara a,b , Smruti Ragunath c , Yong Pu b , Somenath Mitra b,∗ a b c
Dept. of Chemistry, Faculty of Science and Technology, Rajamangala University of Technology Thanyaburi, Pathumtani 12110, Thailand Dept. of Chemistry and Environmental Science, New Jersey Institute of Technology, 138 Warren Street, Newark, NJ 07102, USA Dept. of Chemical, Biological and Pharmaceutical Engineering, New Jersey Institute of Technology, Newark, NJ 07102, USA
a r t i c l e
i n f o
Article history: Received 9 December 2009 Received in revised form 15 January 2010 Accepted 18 January 2010 Available online 25 January 2010 Keywords: Microtrap Multiwalled carbon nanotubes Preconcentration Gas chromatography Methane
a b s t r a c t The GC monitoring of green house gases is a challenging task because the concentration of organic species such as methane are relatively low (ppm to ppb) and their analysis requires some level of preconcentration. Since methane is highly volatile, it is not easily retained on conventional sorbents. In this paper we present multiwalled carbon nanotubes (MWNTs) as an effective sorbent for a microtrap designed for methane preconcentration. Its performance was compared to other commercially available carbon based sorbents, and it was found to be the most effective sorbent in terms of breakthrough volume and enthalpy of adsorption. © 2010 Elsevier B.V. All rights reserved.
1. Introduction Anthropogenic activities have increased the concentration of greenhouse gases (GHGs) such as carbon dioxide (CO2 ), methane (CH4 ), nitrous oxide (N2 O) and chlorofluorocarbons (CFCs). From the standpoint of control strategies, the monitoring of the individual gases is of great importance. Current methodologies for GHG measurements fall under two basic categories, namely optical spectroscopy and gas chromatography (GC). The most popular optical methods are Fourier transform infrared (FTIR) and the cavity ring down (CRD) spectroscopy. An FTIR spectrometer is a Michelson interferometer, and the CRD is based upon the measurement of the rate of absorption of a light pulse confined in a closed optical cavity with a high Q-factor [1–4]. These are inherently complex and expensive instruments where the quantification of multiple species at low concentration is difficult. The advantage of GC analysis of GHGs is that it is a high resolution separation technique for the diverse components, and it is conceivable that a cost-effective analytical platform can be developed that can serve as a field instrument. However, this is a challenging task because while the concentration of CO2 is relatively high, the concentration of organic species such as CH4 are relatively low (ppm to ppb), and some level of preconcentration is necessary
∗ Corresponding author. Tel.: +1 973 596 5611; fax: +1 973 642 7170. E-mail addresses: [email protected], [email protected] (S. Mitra). 0003-2670/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.aca.2010.01.037
for their monitoring, especially for on-line and field monitoring [5–11]. Micro-sorbent traps referred to as a microtrap serve as integrated concentration-injection devices for continuous monitoring of gas streams [12–16]. These have been fabricated from microbore capillary as well as via silicon micromachining. Their small dimension allows rapid desorption of adsorbed compounds with the potential advantages of faster operation, smaller bandwidth and low detection limit. Consequently they have been used in chemical sensing, trace monitoring in GC, mass spectrometry (MS) and non-methane organic carbon (NMOC) analyzer [7,17–21]. Microtraps are fabricated using small capillary tubings (0.25–0.5 mm ID) packed with very small amounts of sorbents. They are designed to have low heat capacity so as to be heated and cooled rapidly. The sorbent selection for microtraps is of great importance because the analytes need to be retained in a small quantity of sorbent, and at the same time desorbed rapidly for fast, quantitative detection [10]. The breakthrough and desorption efficiency are important characteristics of a microtrap. Accurate quantitation requires the sample amount should not exceed its breakthrough volume, which is defined as the volume that can be sampled per unit weight of the sorbent before the analyte is lost [6,22,23]. This can be overcome by increasing the amount of sorbent, but that requires larger diameter tubing with higher thermal mass that, reduces desorption efficiency. Therefore, the selection of suitable sorbent is essential for obtaining large breakthrough and desorption efficiency of a microtrap [11,21,24].The major challenge is to
C. Saridara et al. / Analytica Chimica Acta 677 (2010) 50–54
51
Fig. 1. (a) Photograph of a microtrap as compared to a ball point pen. The microtrap was 0.53 mm ID, and 15 cm long, containing 13 mg adsorbent. (b) Schematic diagram of the experimental system.
find an adsorbent that would provide high specific capacity for CH4 , so that it can be effective as a microtrap. The conventional carbon based sorbents may be sub-classified into activated carbon, carbon molecular sieves and graphitized carbon blacks [11,25,26]. Carbon nanotubes (CNTs) are a novel class of material with potential applications ranging from nanoelectronics to gas separations. The CNTs have excellent mechanical strength, are thermally stable and exhibit high electrical as well as thermal conductivity [27–29]. In recent years carbon nanotubes have attracted great attention with its unique sorption property in various applications [30–32]. They have been used as sorbents for hydrogen storage [33,34], and in the efficient removal of trace contaminants from water [35,36] and air [21,24,37–39]. CNTs have also been used as stationary phases for high resolution gas chromatographic separation for both small and large molecules [40–43]. They have also been used for analytical preconcentration such as solid phase extraction [44,45]. Recently
we have reported the application of CNTs for preconcentration of volatile and semivolatile organics using a microtrap [21,24,38,46]. Due to its high volatility, the preconcentration of methane has been a challenge and our previous efforts with conventional commercial sorbents has shown limited sensitivity [47]. The objective of this project was to study the sorption for methane in a microtrap with potential application in the monitoring of GHGs. The trapping efficiency of different commercially available sorbents is presented on a comparative basis. 2. Experimental 2.1. Materials The adsorbents used for the study were multiwalled carbon nanotubes (MWNTs) and carbon based sorbents namely, CarboxeneTM ,
Table 1 Breakthrough characteristics and enthalpy calculation of various adsorbents. Adsorbents
MWNT
CarboxeneTM
CarbosieveTM
CarbopackTM
Type of adsorbent Surface areaa
Nanotube 150–3000 m2 g−1
Carbon molecular sieve 500 m2 g−1
Carbon molecular sieve 820 m2 g−1
Graphitized carbon black 10 m2 g−1
Temperature 20 10 0 −10 −20
3 4 5 7 8
1 3 3 3.5 4
Enthalpy of adsorption (kJ mol−1 )
5.1
2.3
a
From Refs. [24,25].
Breakthrough time (min) 1 3 4 4 4 1.7
2 2 2 3 4 1.5
52
C. Saridara et al. / Analytica Chimica Acta 677 (2010) 50–54
Fig. 2. SEM images of the sorbents studied as micro concentrator: (a) MWNT, (b) CarbosieveTM , (c) CarboxeneTM , and (d) CarbopackTM .
CarbosieveTM and CarbopackTM . The MWNTs were purchased from Cheap Tubes Inc. (Brattleboro, VT, USA), while other commercial sorbents (each of mesh size 60/80) and the SilicosteelTM tubing for the microtrap were purchased from Supelco Inc. (Bellefonte, PA, USA). Morphology of the sorbents was studied using scanning electron microscopy (SEM) (Leo 1530VP, Carl Zeiss SMT AG Company, Oberkochen, Germany).
2.2. Microtrap fabrication and operation A photograph of the microtrap is shown in Fig. 1(a). The microtraps were fabricated by packing 15 cm long silicosteel tubing of diameter 0.53 mm ID. Each microtrap was packed with 13 mg of the sorbent. A commercially available vibrator was used to ensure uniform distribution of the sorbent through the microtrap. A resister was placed in series with the microtrap to control the current through it. The breakthrough and desorption efficiency are important characteristics of a microtrap. Since it was designed to have small dimensions, it contained small amounts of sorbent, which will have relatively lower absorption capacity. The detail of microtrap fabrication has been presented before [6,10,22,23]. The experimental system used for these studies is as shown in Fig. 1(b). The microtrap was connected using 1/32 Swagelok type fittings to the GC column. A gas standard containing 10 ppm of CH4 in nitrogen was purchased from Matheson Tri Gas Chicago, IL, USA. It flowed continuous through the microtrap while the CH4 was trapped by the sorbent. The microtrap was resistively heated with a 7–10 A pulse of electric current from a power supply (Variac 100/200 Series, Staco Energy Products & Co., Dayton, OH, USA) at regular intervals so as to desorb the trapped organics to be analyzed. The duration of the pulse was between 0.5 and 2.5 s. An electric timer (Gralab Timer Model 451, Dimco-Gray Co., USA) was to control the durations and interval between electrical pulses. A power resistor was put in series to control the current through the microtrap. A gas chromatograph (Hewlett Packard 5890 Series II) equipped with thermal conductivity detector (TCD) was used for analysis using a 0.53 mm ID and 30 m long CarboxeneTM 1010PLOT column (Supelco, USA). Nitrogen served as the carrier gas at a flow
rate of 5 mL min−1 for the microtrap. The microtrap was placed in an insulated chamber, whose temperature was controlled (−20 ◦ C using dry ice to 20 ◦ C). 3. Results and discussion The sorbents used in this study are presented in Table 1 and their SEM images are shown in Fig. 2. It is evident that the morphology of the MWNT was quite different from the other sorbents used for the study. CarboxeneTM , CarbopackTM and CarbosieveTM were porous material with significant internal surface areas. The CNTs themselves are nonporous structures. In some instances, the opening at the tube-ends may generate internal pores inside the hollow tubes. The interstitial voids between the tubes may also serve as sorption sites. However, for all practical purposes the CNTs may be considered nonporous (relatively nonporous). This is one of the major advantages of CNTs, where the solute is held on the surface by Van der Waals type forces, thus eliminating the mass transfer resistance related to the diffusion into elaborate pore structures. The high capacity of the CNTs comes from their large aspect ratio, which attributes to their high surface area as shown in Table 1. 3.1. Breakthrough characteristics The sorption capacity of the microtrap was evaluated by studying the breakthrough time of the microtrap which is defined as the time required by an analyte to eluting through. In the case of a difficulty to retain compound such as CH4 , the breakthrough time is the most important characteristic of the microtrap. This was determined using to methods established before [21,22,24]. The response was monitored as a function of the injection interval. The response increased linearly as the interval between the injections was increased till the breakthrough was reached. After this the response reached a steady value. To compare the sorption capacity of the different sorbents, the breakthrough times of different sorbents namely, MWNT, CarboxeneTM , CarbosieveTM and CarbopackTM were estimated, and are presented in Fig. 3. This is a plot of detector response as a
C. Saridara et al. / Analytica Chimica Acta 677 (2010) 50–54
Fig. 3. (a) Sequence of desorption peaks generated using microtraps containing different sorbents with 10 ppm CH4 as the sample (flow rate was 5 mL min−1 ). Injection were made at every 4 min and the duration of the desorption pulse was 2.5 s. (b) Detector response as a function of adsorption time (injection intervals) for methane using different microtraps at 10 ◦ C. The microtrap was heated using 2.5 s desorption pulse at 15 V.
function of adsorption time (measured as the interval between injections). The time required to reach the maximum point is the measure of breakthrough time. Under similar conditions, the MWNT showed the longest breakthrough time. The stronger sorption also allowed more CH4 to be trapped in the microtrap. The resulting response in terms of peak height was much higher for MWNT than for the other sorbents. This can be seen from both Fig. 3(a) and (b). This clearly established that the MWNT was the most suitable sorbent for the microtrap. The responses from the microtrap fabricated from the different sorbents are shown in Fig. 3(b). CarbopackTM showed nearly negligible response, while the excellent sensitivity of the MWNT is quite evident. The MWNTs microtrap showed no carryover. A second desorption after initial trapping did not generate any peak, indicating that all the CH4 had been desorbed. This was in line with our previous observations which showed that desorption from MWNTs was more rapid than conventional carbon based sorbents due to the elimination of pore diffusion [21]. The MWNTs microtrap also showed linear response in the 1–10 ppm concentration range with an R2 of 0.9986.
53
Fig. 4. (a) Detector response at different of adsorption time (injection intervals) for MWNT microtrap at various temperatures (flow rate was 5 mL min−1 ). (b) Plot of CH4 peak height as a function of temperature (flow rate was 5 mL min−1 ) for a microtrap made of MWNT.
Table 1 shows the breakthrough time of different sorbent as a function of temperature. The temperature range was between −20 ◦ C and 20 ◦ C. As expected, the breakthrough time at −20 ◦ C was the longest. This also led to higher sensitivity in terms of peak height or area (here the peak height was taken as measure of sensitivity). The breakthrough time more than doubled as the temperature was lowered from 20 ◦ to −20 ◦ C, which is shown in Table 1. The results followed the van’t Hoff-type relationship [48–50] as shown in Fig. 5. The plot of Log BTV as a function of 1/T and was found to be linear according to: log(BTV) = k1
1 T
+ k2
3.2. Breakthrough time as a function of temperature One way to enhance sensitivity and increase the breakthrough time was by lowering the temperature of the microtrap [48,49]. When the sorption temperature was decreased, the breakthrough time increased. The results are shown in Fig. 4(a) and (b) and Table 1.
Fig. 5. Breakthrough volume as a funciton of temperature (1/T) for various sorbents.
54
C. Saridara et al. / Analytica Chimica Acta 677 (2010) 50–54
where BTV is the breakthrough volume (BTV is calculated from breakthrough time and is a product of breakthrough time and flow rate) and k1 and k2 are constants. It was interesting to note that the slopes varied for the different sorbents with MWNT showing the highest while CarbopackTM the lowest. The isosteric heat of adsorption (Hs ) is the amount of heat released when an atom adsorbs on a substrate, and is related to the activation energy of sorption for a sorbate–sorbent system. The strength of interaction of compound with the surface of the adsorbent is represented by the enthalpy of adsorption, Hs , given by [49,51]
Hs = −R
ı(ln Vg) ı(1/T )
The Hs was obtained from the slope of plots of ln Vg vs. 1/T, where Vg is the retention volume of the organic compound on the sorbent. A linear dependence indicated a constant value of the isosteric heat of adsorption in the temperature range studied, while relative change in Hs of sorbents with temperature is attributable with the activation of the sorbent surface [42,52–54]. These values for CH4 are presented in Table 1. The maximum Hs was for MWNT, suggesting that it had the strongest interaction with the analyte. This was followed by CarboxeneTM , CarbosieveTM and CarbopackTM . Once again this demonstrated that the mechanisms of adsorption were quite similar in these sorbents. It is interesting to note that the Hs followed a trend similar to that of the breakthrough times for the different analytes. 4. Conclusion Methane sorption capacity of the microtrap was evaluated by studying the breakthrough characteristic. Compared to commercially available carbon based sorbents, namely CarboxeneTM , CarbosieveTM and CarbopackTM , MWNT was significantly more effective in retaining CH4 and serving as a preconcentrator. The MWNTs showed significantly higher sensitivity, breakthrough times that were of the orders of minutes and the enthalpy of adsorption was between two to three times higher. Acknowledgements This work was supported by a grant from a Phase II SBIR grant from the US Department of Energy. Miss Susana Addo Ntim is acknowledged for her help with the scanning electron microscopy. References [1] R.T. Jongma, M.G.H. Boogaarts, I. Holleman, G. Meijer, Rev. Sci. Instrum. 66 (4) (1995) 2821–2828. [2] E.H. Amers, D. Schram, R. Englen, Chem. Phys. Lett. 365 (2002) 237–243. [3] Y. He, B.J. Orr, Appl. Phys. B: Lasers Opt. 85 (2006) 355–364.
[4] E.R. Crosson, Appl. Phys. B: Lasers Opt. 92 (2008) 403–408. [5] H.P. Tuan, H.G. Janssen, C.A. Cramers, P. Mussche, J. Lips, N. Wilson, A. Handley, J. Chromatogr. A 791 (1997) 187–195. [6] C. Feng, S. Mitra, J. Chromatogr. A 805 (1998) 169–176. [7] D.J. Butcher, Microchem. J. 66 (2000) 55–72. [8] S. Mitra, Sample Preparation Techniques in Analytical Chemistry, first ed., Wiley-Interscience, 2003. [9] C. Saridara, R. Brukh, S. Mitra, J. Sep. Sci. 29 (2006) 446–452. [10] J.M. Sanchez, R.D. Sacks, Anal. Chem. 78 (2006) 3046–3054. [11] M.R. Ras, F. Borrull, R.M. Marce, TrAC, Trends Anal. Chem. 28 (3) (2009) 347–361. [12] S. Mitra, C. Yun, J. Chromatogr. 648 (1993) 415–421. [13] E. Matisova, S. Skrabakova, J. Chromatogr. A 707 (1995) 145–179. [14] S. Mitra, N. Zhu, X. Zhang, B. Kebbekus, J. Chromatogr. A 736 (1996) 165–173. [15] N. Zhu, Z. Li, S. Mitra, J. Microcolumn Sep. 10 (5) (1998) 393–399. [16] L. Pillonel, J.O. Bossetw, R. Tabacchi, Lebensm.-Wiss. u.-Technol. 35 (2002) 1–14. [17] J. Whiting, R. Sacks, Anal. Chem. 74 (2002) 246–252. [18] M. Kim, S. Mitra, J. Chromatogr. A 996 (2003) 1–11. [19] H. Sekiguchi, K. Matsushita, S. Yamashiro, Y. Sano, Y. Seto, T. Okuda, A. Sato, Forensic Toxicol. 24 (2006) 17–22. [20] S. Song, A.K. Singh, Anal. Bioanal. Chem. 384 (2006) 41–43. [21] C.M. Hussain, C. Saridara, S. Mitra, Analyst 133 (2008) 1076–1082. [22] C. Feng, S. Mitra, J. Microcolumn Sep. 12 (4) (2000) 267–275. [23] A. Kroupa, J. Dewulf, H.V. Langenhove, I. Viden, J. Chromatogr. A 1038 (2004) 215–223. [24] C.M. Hussain, C. Saridara, S. Mitra, J. Chromatogr. A 1185 (2008) 161–166. [25] K. Dettmer, W. Engewald, Anal. Bioanal. Chem. 373 (2002) 490–500. [26] J. Pollmann, D. Helmig, J. Hueber, D. Tanner, P.P. Tans, J. Chromatogr. A 1134 (2006) 1–15. [27] S. Iijima, Nature 354 (1991) 56–58. [28] T.P. Kumar, R. Ramesh, Y.Y. Lin, G.T.K. Fey, Electrochem. Commun. 6 (2004) 520–525. [29] M. Trojanowicz, TrAC, Trends Anal. Chem. 25 (5) (2006) 480–489. [30] M. Valcarcel, B.M. Simonet, Anal. Bioanal. Chem. 382 (2005) 1783–1790. [31] E. Meena, F.D. Negra, M. Prato, N. Tagmatarchis, A. Ciogli, F. Gasparrini, D. Misiti, C. Villani, Carbon 44 (2006) 1581–1616. [32] M. Valcarcel, S. Cardenas, B.N. Simonet, Y. Moliner- Martinez, R. Lucena, TrAC, Trends Anal. Chem. 27 (1) (2008) 34–43. [33] K. Shen, H. Xu, Y. Jiang, T. PietraB, Carbon 42 (2004) 2315–2322. [34] Y. Yurum, A. Taralp, T.N. Veziroglu, Int. J. Hydrogen Energy 34 (2009) 3784–3798. [35] C. Lu, F. Su, Sep. Purif. Technol. 58 (2007) 113–121. [36] O.S. Khow, S. Mitra, J. Mater. Chem. 19 (2009) 3713–3718. [37] Q.L. Li, D.X. Yuan, Q.M. Lin, J. Chromatogr. A 1026 (2004) 283–288. [38] C. Saridara, R. Brukh, Z. Iqbal, S. Mitra, Anal. Chem. 77 (2005) 1183–1187. [39] Y.H. Shih, M.S. Li, J. Hazard. Mater. 154 (2008) 23–28. [40] Q. Li, D. Yuan, J. Chromatogr. A 1003 (2003) 203–209. [41] C. Saridara, S. Mitra, Anal. Chem. 77 (21) (2005) 7094–7709. [42] M. Karwa, S. Mitra, Anal. Chem. 78 (6) (2006) 2064–2070. [43] L.M. Yuan, C.X. Ren, L. Li, P. Ai, Z.H. Yan, M. Zi, Z.Y. Li, Anal. Chem. 78 (2006) 6384–6390. [44] Q. Zhou, Y. Ding, J. Xiao, Chromatographia 65 (2007) 25–30. [45] O.S. Khow, S. Mitra, J. Chromatogr. A 1216 (2009) 2270–2274. [46] F. Zheng, D.L. Baldwin, L.S. Fifield, N.C. Anheir, C.L. Aardhal, J.W. Grate, Anal. Chem. 78 (2006) 2442–2446. [47] C. Thammakhet, P. Thavarungkul, R. Brukh, S. Mitra, P. Kanatharana, J. Chromatogr. A 1072 (2005) 243–248. [48] P.W. Atkin, Physical Chemistry, fifth ed., Oxford University Press, Oxford, 1994. [49] V. Simon, M.L. Riba, A. Waldhart, L. Torres, J. Chromatogr. A 704 (1995) 465–471. [50] P. Safaeefar, H.M. Ang, H. Kuramochi, Y. Asakuma, K. Maeda, M.O. Tade, K. Fukui, Fluid Phase Equilibr. 238 (2005) 180–185. [51] E. Diaz, S. Ordonez, A. Vega, J. Colloid Interface Sci. 308 (2007) 7–16. [52] K. Yang, L. Zhu, B. Xing, Environ. Sci. Technol. 40 (2006) 1855–1861. [53] D.H. Yoo, G.H. Rue, Y.H. Hwang, H.K. Kim, J. Phys. Chem. B 106 (2002) 3371–3374. [54] C. Bilgic, A. Askin, J. Chromatogr. A 1006 (2003) 281–286.
Contents lists available at ScienceDirect
Analytica Chimica Acta journal homepage: www.elsevier.com/locate/aca
Methane preconcentration in a microtrap using multiwalled carbon nanotubes as sorbents Chutarat Saridara a,b , Smruti Ragunath c , Yong Pu b , Somenath Mitra b,∗ a b c
Dept. of Chemistry, Faculty of Science and Technology, Rajamangala University of Technology Thanyaburi, Pathumtani 12110, Thailand Dept. of Chemistry and Environmental Science, New Jersey Institute of Technology, 138 Warren Street, Newark, NJ 07102, USA Dept. of Chemical, Biological and Pharmaceutical Engineering, New Jersey Institute of Technology, Newark, NJ 07102, USA
a r t i c l e
i n f o
Article history: Received 9 December 2009 Received in revised form 15 January 2010 Accepted 18 January 2010 Available online 25 January 2010 Keywords: Microtrap Multiwalled carbon nanotubes Preconcentration Gas chromatography Methane
a b s t r a c t The GC monitoring of green house gases is a challenging task because the concentration of organic species such as methane are relatively low (ppm to ppb) and their analysis requires some level of preconcentration. Since methane is highly volatile, it is not easily retained on conventional sorbents. In this paper we present multiwalled carbon nanotubes (MWNTs) as an effective sorbent for a microtrap designed for methane preconcentration. Its performance was compared to other commercially available carbon based sorbents, and it was found to be the most effective sorbent in terms of breakthrough volume and enthalpy of adsorption. © 2010 Elsevier B.V. All rights reserved.
1. Introduction Anthropogenic activities have increased the concentration of greenhouse gases (GHGs) such as carbon dioxide (CO2 ), methane (CH4 ), nitrous oxide (N2 O) and chlorofluorocarbons (CFCs). From the standpoint of control strategies, the monitoring of the individual gases is of great importance. Current methodologies for GHG measurements fall under two basic categories, namely optical spectroscopy and gas chromatography (GC). The most popular optical methods are Fourier transform infrared (FTIR) and the cavity ring down (CRD) spectroscopy. An FTIR spectrometer is a Michelson interferometer, and the CRD is based upon the measurement of the rate of absorption of a light pulse confined in a closed optical cavity with a high Q-factor [1–4]. These are inherently complex and expensive instruments where the quantification of multiple species at low concentration is difficult. The advantage of GC analysis of GHGs is that it is a high resolution separation technique for the diverse components, and it is conceivable that a cost-effective analytical platform can be developed that can serve as a field instrument. However, this is a challenging task because while the concentration of CO2 is relatively high, the concentration of organic species such as CH4 are relatively low (ppm to ppb), and some level of preconcentration is necessary
∗ Corresponding author. Tel.: +1 973 596 5611; fax: +1 973 642 7170. E-mail addresses: [email protected], [email protected] (S. Mitra). 0003-2670/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.aca.2010.01.037
for their monitoring, especially for on-line and field monitoring [5–11]. Micro-sorbent traps referred to as a microtrap serve as integrated concentration-injection devices for continuous monitoring of gas streams [12–16]. These have been fabricated from microbore capillary as well as via silicon micromachining. Their small dimension allows rapid desorption of adsorbed compounds with the potential advantages of faster operation, smaller bandwidth and low detection limit. Consequently they have been used in chemical sensing, trace monitoring in GC, mass spectrometry (MS) and non-methane organic carbon (NMOC) analyzer [7,17–21]. Microtraps are fabricated using small capillary tubings (0.25–0.5 mm ID) packed with very small amounts of sorbents. They are designed to have low heat capacity so as to be heated and cooled rapidly. The sorbent selection for microtraps is of great importance because the analytes need to be retained in a small quantity of sorbent, and at the same time desorbed rapidly for fast, quantitative detection [10]. The breakthrough and desorption efficiency are important characteristics of a microtrap. Accurate quantitation requires the sample amount should not exceed its breakthrough volume, which is defined as the volume that can be sampled per unit weight of the sorbent before the analyte is lost [6,22,23]. This can be overcome by increasing the amount of sorbent, but that requires larger diameter tubing with higher thermal mass that, reduces desorption efficiency. Therefore, the selection of suitable sorbent is essential for obtaining large breakthrough and desorption efficiency of a microtrap [11,21,24].The major challenge is to
C. Saridara et al. / Analytica Chimica Acta 677 (2010) 50–54
51
Fig. 1. (a) Photograph of a microtrap as compared to a ball point pen. The microtrap was 0.53 mm ID, and 15 cm long, containing 13 mg adsorbent. (b) Schematic diagram of the experimental system.
find an adsorbent that would provide high specific capacity for CH4 , so that it can be effective as a microtrap. The conventional carbon based sorbents may be sub-classified into activated carbon, carbon molecular sieves and graphitized carbon blacks [11,25,26]. Carbon nanotubes (CNTs) are a novel class of material with potential applications ranging from nanoelectronics to gas separations. The CNTs have excellent mechanical strength, are thermally stable and exhibit high electrical as well as thermal conductivity [27–29]. In recent years carbon nanotubes have attracted great attention with its unique sorption property in various applications [30–32]. They have been used as sorbents for hydrogen storage [33,34], and in the efficient removal of trace contaminants from water [35,36] and air [21,24,37–39]. CNTs have also been used as stationary phases for high resolution gas chromatographic separation for both small and large molecules [40–43]. They have also been used for analytical preconcentration such as solid phase extraction [44,45]. Recently
we have reported the application of CNTs for preconcentration of volatile and semivolatile organics using a microtrap [21,24,38,46]. Due to its high volatility, the preconcentration of methane has been a challenge and our previous efforts with conventional commercial sorbents has shown limited sensitivity [47]. The objective of this project was to study the sorption for methane in a microtrap with potential application in the monitoring of GHGs. The trapping efficiency of different commercially available sorbents is presented on a comparative basis. 2. Experimental 2.1. Materials The adsorbents used for the study were multiwalled carbon nanotubes (MWNTs) and carbon based sorbents namely, CarboxeneTM ,
Table 1 Breakthrough characteristics and enthalpy calculation of various adsorbents. Adsorbents
MWNT
CarboxeneTM
CarbosieveTM
CarbopackTM
Type of adsorbent Surface areaa
Nanotube 150–3000 m2 g−1
Carbon molecular sieve 500 m2 g−1
Carbon molecular sieve 820 m2 g−1
Graphitized carbon black 10 m2 g−1
Temperature 20 10 0 −10 −20
3 4 5 7 8
1 3 3 3.5 4
Enthalpy of adsorption (kJ mol−1 )
5.1
2.3
a
From Refs. [24,25].
Breakthrough time (min) 1 3 4 4 4 1.7
2 2 2 3 4 1.5
52
C. Saridara et al. / Analytica Chimica Acta 677 (2010) 50–54
Fig. 2. SEM images of the sorbents studied as micro concentrator: (a) MWNT, (b) CarbosieveTM , (c) CarboxeneTM , and (d) CarbopackTM .
CarbosieveTM and CarbopackTM . The MWNTs were purchased from Cheap Tubes Inc. (Brattleboro, VT, USA), while other commercial sorbents (each of mesh size 60/80) and the SilicosteelTM tubing for the microtrap were purchased from Supelco Inc. (Bellefonte, PA, USA). Morphology of the sorbents was studied using scanning electron microscopy (SEM) (Leo 1530VP, Carl Zeiss SMT AG Company, Oberkochen, Germany).
2.2. Microtrap fabrication and operation A photograph of the microtrap is shown in Fig. 1(a). The microtraps were fabricated by packing 15 cm long silicosteel tubing of diameter 0.53 mm ID. Each microtrap was packed with 13 mg of the sorbent. A commercially available vibrator was used to ensure uniform distribution of the sorbent through the microtrap. A resister was placed in series with the microtrap to control the current through it. The breakthrough and desorption efficiency are important characteristics of a microtrap. Since it was designed to have small dimensions, it contained small amounts of sorbent, which will have relatively lower absorption capacity. The detail of microtrap fabrication has been presented before [6,10,22,23]. The experimental system used for these studies is as shown in Fig. 1(b). The microtrap was connected using 1/32 Swagelok type fittings to the GC column. A gas standard containing 10 ppm of CH4 in nitrogen was purchased from Matheson Tri Gas Chicago, IL, USA. It flowed continuous through the microtrap while the CH4 was trapped by the sorbent. The microtrap was resistively heated with a 7–10 A pulse of electric current from a power supply (Variac 100/200 Series, Staco Energy Products & Co., Dayton, OH, USA) at regular intervals so as to desorb the trapped organics to be analyzed. The duration of the pulse was between 0.5 and 2.5 s. An electric timer (Gralab Timer Model 451, Dimco-Gray Co., USA) was to control the durations and interval between electrical pulses. A power resistor was put in series to control the current through the microtrap. A gas chromatograph (Hewlett Packard 5890 Series II) equipped with thermal conductivity detector (TCD) was used for analysis using a 0.53 mm ID and 30 m long CarboxeneTM 1010PLOT column (Supelco, USA). Nitrogen served as the carrier gas at a flow
rate of 5 mL min−1 for the microtrap. The microtrap was placed in an insulated chamber, whose temperature was controlled (−20 ◦ C using dry ice to 20 ◦ C). 3. Results and discussion The sorbents used in this study are presented in Table 1 and their SEM images are shown in Fig. 2. It is evident that the morphology of the MWNT was quite different from the other sorbents used for the study. CarboxeneTM , CarbopackTM and CarbosieveTM were porous material with significant internal surface areas. The CNTs themselves are nonporous structures. In some instances, the opening at the tube-ends may generate internal pores inside the hollow tubes. The interstitial voids between the tubes may also serve as sorption sites. However, for all practical purposes the CNTs may be considered nonporous (relatively nonporous). This is one of the major advantages of CNTs, where the solute is held on the surface by Van der Waals type forces, thus eliminating the mass transfer resistance related to the diffusion into elaborate pore structures. The high capacity of the CNTs comes from their large aspect ratio, which attributes to their high surface area as shown in Table 1. 3.1. Breakthrough characteristics The sorption capacity of the microtrap was evaluated by studying the breakthrough time of the microtrap which is defined as the time required by an analyte to eluting through. In the case of a difficulty to retain compound such as CH4 , the breakthrough time is the most important characteristic of the microtrap. This was determined using to methods established before [21,22,24]. The response was monitored as a function of the injection interval. The response increased linearly as the interval between the injections was increased till the breakthrough was reached. After this the response reached a steady value. To compare the sorption capacity of the different sorbents, the breakthrough times of different sorbents namely, MWNT, CarboxeneTM , CarbosieveTM and CarbopackTM were estimated, and are presented in Fig. 3. This is a plot of detector response as a
C. Saridara et al. / Analytica Chimica Acta 677 (2010) 50–54
Fig. 3. (a) Sequence of desorption peaks generated using microtraps containing different sorbents with 10 ppm CH4 as the sample (flow rate was 5 mL min−1 ). Injection were made at every 4 min and the duration of the desorption pulse was 2.5 s. (b) Detector response as a function of adsorption time (injection intervals) for methane using different microtraps at 10 ◦ C. The microtrap was heated using 2.5 s desorption pulse at 15 V.
function of adsorption time (measured as the interval between injections). The time required to reach the maximum point is the measure of breakthrough time. Under similar conditions, the MWNT showed the longest breakthrough time. The stronger sorption also allowed more CH4 to be trapped in the microtrap. The resulting response in terms of peak height was much higher for MWNT than for the other sorbents. This can be seen from both Fig. 3(a) and (b). This clearly established that the MWNT was the most suitable sorbent for the microtrap. The responses from the microtrap fabricated from the different sorbents are shown in Fig. 3(b). CarbopackTM showed nearly negligible response, while the excellent sensitivity of the MWNT is quite evident. The MWNTs microtrap showed no carryover. A second desorption after initial trapping did not generate any peak, indicating that all the CH4 had been desorbed. This was in line with our previous observations which showed that desorption from MWNTs was more rapid than conventional carbon based sorbents due to the elimination of pore diffusion [21]. The MWNTs microtrap also showed linear response in the 1–10 ppm concentration range with an R2 of 0.9986.
53
Fig. 4. (a) Detector response at different of adsorption time (injection intervals) for MWNT microtrap at various temperatures (flow rate was 5 mL min−1 ). (b) Plot of CH4 peak height as a function of temperature (flow rate was 5 mL min−1 ) for a microtrap made of MWNT.
Table 1 shows the breakthrough time of different sorbent as a function of temperature. The temperature range was between −20 ◦ C and 20 ◦ C. As expected, the breakthrough time at −20 ◦ C was the longest. This also led to higher sensitivity in terms of peak height or area (here the peak height was taken as measure of sensitivity). The breakthrough time more than doubled as the temperature was lowered from 20 ◦ to −20 ◦ C, which is shown in Table 1. The results followed the van’t Hoff-type relationship [48–50] as shown in Fig. 5. The plot of Log BTV as a function of 1/T and was found to be linear according to: log(BTV) = k1
1 T
+ k2
3.2. Breakthrough time as a function of temperature One way to enhance sensitivity and increase the breakthrough time was by lowering the temperature of the microtrap [48,49]. When the sorption temperature was decreased, the breakthrough time increased. The results are shown in Fig. 4(a) and (b) and Table 1.
Fig. 5. Breakthrough volume as a funciton of temperature (1/T) for various sorbents.
54
C. Saridara et al. / Analytica Chimica Acta 677 (2010) 50–54
where BTV is the breakthrough volume (BTV is calculated from breakthrough time and is a product of breakthrough time and flow rate) and k1 and k2 are constants. It was interesting to note that the slopes varied for the different sorbents with MWNT showing the highest while CarbopackTM the lowest. The isosteric heat of adsorption (Hs ) is the amount of heat released when an atom adsorbs on a substrate, and is related to the activation energy of sorption for a sorbate–sorbent system. The strength of interaction of compound with the surface of the adsorbent is represented by the enthalpy of adsorption, Hs , given by [49,51]
Hs = −R
ı(ln Vg) ı(1/T )
The Hs was obtained from the slope of plots of ln Vg vs. 1/T, where Vg is the retention volume of the organic compound on the sorbent. A linear dependence indicated a constant value of the isosteric heat of adsorption in the temperature range studied, while relative change in Hs of sorbents with temperature is attributable with the activation of the sorbent surface [42,52–54]. These values for CH4 are presented in Table 1. The maximum Hs was for MWNT, suggesting that it had the strongest interaction with the analyte. This was followed by CarboxeneTM , CarbosieveTM and CarbopackTM . Once again this demonstrated that the mechanisms of adsorption were quite similar in these sorbents. It is interesting to note that the Hs followed a trend similar to that of the breakthrough times for the different analytes. 4. Conclusion Methane sorption capacity of the microtrap was evaluated by studying the breakthrough characteristic. Compared to commercially available carbon based sorbents, namely CarboxeneTM , CarbosieveTM and CarbopackTM , MWNT was significantly more effective in retaining CH4 and serving as a preconcentrator. The MWNTs showed significantly higher sensitivity, breakthrough times that were of the orders of minutes and the enthalpy of adsorption was between two to three times higher. Acknowledgements This work was supported by a grant from a Phase II SBIR grant from the US Department of Energy. Miss Susana Addo Ntim is acknowledged for her help with the scanning electron microscopy. References [1] R.T. Jongma, M.G.H. Boogaarts, I. Holleman, G. Meijer, Rev. Sci. Instrum. 66 (4) (1995) 2821–2828. [2] E.H. Amers, D. Schram, R. Englen, Chem. Phys. Lett. 365 (2002) 237–243. [3] Y. He, B.J. Orr, Appl. Phys. B: Lasers Opt. 85 (2006) 355–364.
[4] E.R. Crosson, Appl. Phys. B: Lasers Opt. 92 (2008) 403–408. [5] H.P. Tuan, H.G. Janssen, C.A. Cramers, P. Mussche, J. Lips, N. Wilson, A. Handley, J. Chromatogr. A 791 (1997) 187–195. [6] C. Feng, S. Mitra, J. Chromatogr. A 805 (1998) 169–176. [7] D.J. Butcher, Microchem. J. 66 (2000) 55–72. [8] S. Mitra, Sample Preparation Techniques in Analytical Chemistry, first ed., Wiley-Interscience, 2003. [9] C. Saridara, R. Brukh, S. Mitra, J. Sep. Sci. 29 (2006) 446–452. [10] J.M. Sanchez, R.D. Sacks, Anal. Chem. 78 (2006) 3046–3054. [11] M.R. Ras, F. Borrull, R.M. Marce, TrAC, Trends Anal. Chem. 28 (3) (2009) 347–361. [12] S. Mitra, C. Yun, J. Chromatogr. 648 (1993) 415–421. [13] E. Matisova, S. Skrabakova, J. Chromatogr. A 707 (1995) 145–179. [14] S. Mitra, N. Zhu, X. Zhang, B. Kebbekus, J. Chromatogr. A 736 (1996) 165–173. [15] N. Zhu, Z. Li, S. Mitra, J. Microcolumn Sep. 10 (5) (1998) 393–399. [16] L. Pillonel, J.O. Bossetw, R. Tabacchi, Lebensm.-Wiss. u.-Technol. 35 (2002) 1–14. [17] J. Whiting, R. Sacks, Anal. Chem. 74 (2002) 246–252. [18] M. Kim, S. Mitra, J. Chromatogr. A 996 (2003) 1–11. [19] H. Sekiguchi, K. Matsushita, S. Yamashiro, Y. Sano, Y. Seto, T. Okuda, A. Sato, Forensic Toxicol. 24 (2006) 17–22. [20] S. Song, A.K. Singh, Anal. Bioanal. Chem. 384 (2006) 41–43. [21] C.M. Hussain, C. Saridara, S. Mitra, Analyst 133 (2008) 1076–1082. [22] C. Feng, S. Mitra, J. Microcolumn Sep. 12 (4) (2000) 267–275. [23] A. Kroupa, J. Dewulf, H.V. Langenhove, I. Viden, J. Chromatogr. A 1038 (2004) 215–223. [24] C.M. Hussain, C. Saridara, S. Mitra, J. Chromatogr. A 1185 (2008) 161–166. [25] K. Dettmer, W. Engewald, Anal. Bioanal. Chem. 373 (2002) 490–500. [26] J. Pollmann, D. Helmig, J. Hueber, D. Tanner, P.P. Tans, J. Chromatogr. A 1134 (2006) 1–15. [27] S. Iijima, Nature 354 (1991) 56–58. [28] T.P. Kumar, R. Ramesh, Y.Y. Lin, G.T.K. Fey, Electrochem. Commun. 6 (2004) 520–525. [29] M. Trojanowicz, TrAC, Trends Anal. Chem. 25 (5) (2006) 480–489. [30] M. Valcarcel, B.M. Simonet, Anal. Bioanal. Chem. 382 (2005) 1783–1790. [31] E. Meena, F.D. Negra, M. Prato, N. Tagmatarchis, A. Ciogli, F. Gasparrini, D. Misiti, C. Villani, Carbon 44 (2006) 1581–1616. [32] M. Valcarcel, S. Cardenas, B.N. Simonet, Y. Moliner- Martinez, R. Lucena, TrAC, Trends Anal. Chem. 27 (1) (2008) 34–43. [33] K. Shen, H. Xu, Y. Jiang, T. PietraB, Carbon 42 (2004) 2315–2322. [34] Y. Yurum, A. Taralp, T.N. Veziroglu, Int. J. Hydrogen Energy 34 (2009) 3784–3798. [35] C. Lu, F. Su, Sep. Purif. Technol. 58 (2007) 113–121. [36] O.S. Khow, S. Mitra, J. Mater. Chem. 19 (2009) 3713–3718. [37] Q.L. Li, D.X. Yuan, Q.M. Lin, J. Chromatogr. A 1026 (2004) 283–288. [38] C. Saridara, R. Brukh, Z. Iqbal, S. Mitra, Anal. Chem. 77 (2005) 1183–1187. [39] Y.H. Shih, M.S. Li, J. Hazard. Mater. 154 (2008) 23–28. [40] Q. Li, D. Yuan, J. Chromatogr. A 1003 (2003) 203–209. [41] C. Saridara, S. Mitra, Anal. Chem. 77 (21) (2005) 7094–7709. [42] M. Karwa, S. Mitra, Anal. Chem. 78 (6) (2006) 2064–2070. [43] L.M. Yuan, C.X. Ren, L. Li, P. Ai, Z.H. Yan, M. Zi, Z.Y. Li, Anal. Chem. 78 (2006) 6384–6390. [44] Q. Zhou, Y. Ding, J. Xiao, Chromatographia 65 (2007) 25–30. [45] O.S. Khow, S. Mitra, J. Chromatogr. A 1216 (2009) 2270–2274. [46] F. Zheng, D.L. Baldwin, L.S. Fifield, N.C. Anheir, C.L. Aardhal, J.W. Grate, Anal. Chem. 78 (2006) 2442–2446. [47] C. Thammakhet, P. Thavarungkul, R. Brukh, S. Mitra, P. Kanatharana, J. Chromatogr. A 1072 (2005) 243–248. [48] P.W. Atkin, Physical Chemistry, fifth ed., Oxford University Press, Oxford, 1994. [49] V. Simon, M.L. Riba, A. Waldhart, L. Torres, J. Chromatogr. A 704 (1995) 465–471. [50] P. Safaeefar, H.M. Ang, H. Kuramochi, Y. Asakuma, K. Maeda, M.O. Tade, K. Fukui, Fluid Phase Equilibr. 238 (2005) 180–185. [51] E. Diaz, S. Ordonez, A. Vega, J. Colloid Interface Sci. 308 (2007) 7–16. [52] K. Yang, L. Zhu, B. Xing, Environ. Sci. Technol. 40 (2006) 1855–1861. [53] D.H. Yoo, G.H. Rue, Y.H. Hwang, H.K. Kim, J. Phys. Chem. B 106 (2002) 3371–3374. [54] C. Bilgic, A. Askin, J. Chromatogr. A 1006 (2003) 281–286.