Analytica Chimica Acta 819 (2014) 78–81
Contents lists available at ScienceDirect
Analytica Chimica Acta journal homepage: www.elsevier.com/locate/aca
Pore size dynamics in interpenetrated metal organic frameworks for selective sensing of aromatic compounds Matthew Myers a,b,∗ , Anna Podolska c , Charles Heath a , Murray V. Baker b , Bobby Pejcic a a b c
CSIRO Earth Science and Resource Engineering, 26 Dick Perry Avenue, Kensington, WA, Australia The University of Western Australia, School of Chemistry and Biochemistry, Crawley, WA, Australia Curtin University, Department of Exploration Geophysics, 26 Dick Perry Avenue, Kensington, WA, Australia
h i g h l i g h t s
g r a p h i c a l
a b s t r a c t
• Unique exciplex emission response to different monoaromatic compounds.
• Metal-organic framework undergoes structural changes with analyte binding. • Fluorescence and sorption experiments show high selectivity for toluene and p-xylene. • Competitive binding selectivity has affects the interpretation of sensor response.
a r t i c l e
i n f o
Article history: Received 18 December 2013 Received in revised form 24 January 2014 Accepted 1 February 2014 Available online 10 February 2014 Keywords: Metal-organic framework Selectivity Fluorescence Sensor Aromatic
a b s t r a c t The two-fold interpenetrated metal-organic framework, [Zn2 (bdc)2 (dpNDI)]n (bdc = 1,4benzenedicarboxylate, dpNDI = N N -di(4-pyridyl)-1,4,5,8-naphthalenediimide) can undergo structural re-arrangement upon adsorption of chemical species changing its pore structure. For a competitive binding process with multiple analytes of different sizes and geometries, the interpenetrated framework will adopt a conformation to maximize the overall binding interactions. In this study, we show for binary mixtures that there is a high selectivity for the larger methylated aromatic compounds, toluene and p-xylene, over the small non-methylated benzene. The dpNDI moiety within [Zn2 (bdc)2 (dpNDI)]n forms an exciplex with these aromatic compounds. The emission wavelength is dependent on the strength of the host-guest CT interaction allowing these compounds to be distinguished. We show that the sorption selectivity characteristics can have a significant impact on the fluorescence sensor response of [Zn2 (bdc)2 (dpNDI)]n towards environmentally important hydrocarbons based contaminants (i.e., BTEX, PAH). Crown Copyright © 2014 Published by Elsevier B.V. All rights reserved.
1. Introduction Metal-organic frameworks (MOFs) typically have very high porosity and surface area, properties that are advantageous and
∗ Corresponding author at: CSIRO Earth Science and Resource Engineering, 26 Dick Perry Avenue, Kensington, WA, Australia. Tel.: +61 8 6436 8708; fax: +61 8 6436 8555. E-mail address: [email protected] (M. Myers). http://dx.doi.org/10.1016/j.aca.2014.02.004 0003-2670/Crown Copyright © 2014 Published by Elsevier B.V. All rights reserved.
enable many of their potential applications [1–3]; however, in comparison, interpenetrated metal-organic frameworks have relatively low porosity and surface area [4]. MOFs have been shown to be a promising material for a wide variety of applications including separation and storage of harmful gas [5], explosives detection/environmental monitoring [6] radioelement capturing/storage [7] and many others [8]. For environmental monitoring using chemical sensors, selectivity and sensitivity are undoubtedly the two most important properties [9–12]. Sensors for the in situ detection of aromatic
M. Myers et al. / Analytica Chimica Acta 819 (2014) 78–81
79
Fig. 1. Schematic of a two-fold interpenetrated metal-organic framework transformation illustrating the change in pore size.
Normalized Emission Intensity
a
100:0 benzene:toluene 99.9:0.1 benzene:toluene 99:1 benzene:toluene 90:10 benzene:toluene 0:100 benzene:toluene
1.0 0.8 0.6 0.4 0.2 0.0 400
2. Experimental 2.1. Materials [Zn2 (bdc)2 (dpNDI)]n was synthesized and combined with dried alumina powder in a 5:95 mass ratio according to the procedures published previously [17]. These solids were vacuum dried at 120 ◦ C to remove any solvents trapped within the MOF matrix. 2.2. Procedures For fluorescence measurements, approximately 10 mg of [Zn2 (bdc)2 (dpNDI)]n /alumina was added individually to 5 mL of varying benzene/toluene and benzene/p-xylene mixtures. For each, the MOF was exposed to the mixture for approximately 3 h until
450
500
550
600
Wavelength (nm)
b Normalized Emission Intensity
hydrocarbons are commonly based on UV absorbance spectroscopy and fluorescence spectroscopy techniques with several such sensors already commercially available [13]. However, with these types of sensors, it is difficult to adequately differentiate between the various BTEX (benzene, toluene, ethylbenzene and xylenes) compounds due to the very limited differences in their UV absorbance and fluorescence emission spectra [14]. Despite having comparatively low porosity and surface area, interpenetrated MOFs have the potential to adopt different pore dimensions by varying the displacement between the different lattices [15,16]. This ability of interpenetrated metal-organic frameworks to adopt differing geometries could lead to enhanced analyte selectivity (see Fig. 1). In this study, we examine the behavior of a fluorescing two-fold interpenetrated MOF towards mixtures of BTEX compounds and determine what impact sorption selectivity has on its sensor response. Recently, Takashima et al. have shown that upon exposure of the two-fold interpenetrated MOF [Zn2 (bdc)2 (dpNDI)]n (bdc = 1,4-benzenedicarboxylate, dpNDI = N N -di(4-pyridyl)1,4,5,8-naphthalenediimide) to individual aromatic analytes (e.g. benzene and toluene) that there are distinct differences in the wavelength of emission upon excitation with 365 nm light [17]. This result is based on the formation of an exciplex that is formed by the association of 1,4,5,8-naphthalenediimide (NDI) derivative with an aromatic compound [18]. By incorporating the dpNDI moiety into an interpenetrated MOF, there is a much higher binding affinity for aromatic compounds, resulting in an intensified excimer emission. However, understanding the fluorescence response of [Zn2 (bdc)2 (dpNDI)]n when exposed to mixtures vs. single chemical constituents is extremely important if potential sensing applications are to be realized. Binary mixtures of benzene/toluene and benzene/p-xylene are examined here with the goal of developing a better understanding of how interpenetrated MOFs selectively respond to mixtures in the context of chemical sensing.
100:0 benzene:p-xylene 99:1 benzene:p-xylene 95:1 benzene:p-xylene 90:10 benzene:p-xylene 0:100 benzene:p-xylene
1.0 0.8 0.6 0.4 0.2 0.0 400
450
500
550
600
Wavelength (nm) Fig. 2. (a) Fluorescence emission spectra (excitation wavelength of 365 nm) when exposed to benzene/toluene mixtures. (b) Fluorescence emission spectra (excitation wavelength of 365 nm) when exposed to benzene/p-xylene mixtures.
equilibrium was established, and then the mixture was thoroughly shaken and immediately subjected to examination by fluorescence spectroscopy (see Supplementary Information for details). Sorption experiments were undertaken by adding 10 mg of pure [Zn2 (bdc)2 (dpNDI)]n (vacuum dried at 120 ◦ C) to varying mixtures of benzene/toluene and benzene/p-xylene. After equilibrating in solution for approximately 24 h, the solvent was removed by filtration and then the remaining solids were vacuum dried at room temperature for 1 h to remove residual solvent. To extract the benzene/toluene or benzene/p-xylene from the MOF matrix, the MOF sample was stirred in 5 mL of methanol for approximately 24 h. The resulting solution was analyzed using gas chromatography/mass spectrometry (see Supporting Information for details). 3. Results and discussion 3.1. Fluorescence response We have confirmed the results of Takashima et al. showing that when [Zn2 (bdc)2 (dpNDI)]n is exposed to pure benzene, pure toluene and pure p-xylene an excimer emission is seen at 441 nm, 476 nm and 511 nm, respectively [17]. During our work, however, we discovered with mixtures of these solvents, that the wavelength of maximum emission intensity is strongly and non-linearly dependent on the relative concentrations of the solvents. For example, Fig. 2 shows that the fluorescence emission spectrum shifts to higher wavelengths with addition of even small amounts of either toluene or p-xylene to benzene. Fig. 3 shows that the shift of wavelength of maximum emission intensity is very abruptly as the proportion of toluene is increased to about 1% vol. or as the
M. Myers et al. / Analytica Chimica Acta 819 (2014) 78–81
Xtoluene / Xbenzene in extract solution
Wavelength at Maximum Emission Intensity
a 480 475 470 465 460 455
80
70
70
60
60
50
50
40
40
30
30
20
20
10
10
0
0 0.4
0.6
0.8
1.0
Xbenzene
450 0.1
1
10
100 Fig. 4. Plots of the mole fraction ratios for benzene/toluene and benzene/p-xylene mixtures vs. mole fraction of benzene.
% Toluene by Volume
b Wavelength at Maximum Emission Intensity
80
Xp-xylene / Xbenzene in extract solution
80
520 510 500 490 480 470 460 450 440 1
10
100
% p-xylene by volume
Fig. 3. (a) Plot of wavelength at maximum emission intensity vs. % toluene by volume in different benzene/toluene mixtures. In pure benzene the maximum intensity is at 441 nm. (b) Plot of wavelength at maximum emission intensity vs. % p-xylene by volume for MOFs in different benzene/toluene mixtures. In pure benzene the maximum intensity is at 441 nm.
proportion of p-xylene in increased to about 2 vol.% and then remains nearly constant with higher concentrations of either toluene or p-xylene in benzene. For example, with a 99:1 v/v ratio of benzene to toluene the emission maximum wavelength has shifted 18 nm relative to pure benzene; whereas, for a 50:50 v/v ratio of benzene to toluene the emission maximum wavelength has shifted only 3 nm relative to pure toluene. This result clearly demonstrates that this MOF is much more sensitive to either toluene or p-xylene compared to benzene. Takashima et al. have reported that the fluorescence quantum yields are 5%, 22% and 19% for benzene, toluene and p-xylene, respectively [17]. Comparing the sets of fluorescence spectra for benzene/toluene (Fig. 2a) and benzene/p-xylene, the curves for benzene/toluene all have similar widths; whereas, for intermediate ratios of benzene/p-xylene there is significant broadening. We initially hypothesized that the variation of fluorescence properties with solvent may just be an artifact of this difference in quantum yield and we conducted a series of sorption experiments using GCMS to test this hypothesis. 3.2. Sorption characteristics With the different mixtures of benzene/toluene and benzene/pxylene, we have analyzed the composition of methanol extracts
for MOFs that were exposed to mixtures of the BTX compounds to determine binary mixture selectivity coefficients. Not surprisingly, there is a high selectivity for toluene over benzene (
Contents lists available at ScienceDirect
Analytica Chimica Acta journal homepage: www.elsevier.com/locate/aca
Pore size dynamics in interpenetrated metal organic frameworks for selective sensing of aromatic compounds Matthew Myers a,b,∗ , Anna Podolska c , Charles Heath a , Murray V. Baker b , Bobby Pejcic a a b c
CSIRO Earth Science and Resource Engineering, 26 Dick Perry Avenue, Kensington, WA, Australia The University of Western Australia, School of Chemistry and Biochemistry, Crawley, WA, Australia Curtin University, Department of Exploration Geophysics, 26 Dick Perry Avenue, Kensington, WA, Australia
h i g h l i g h t s
g r a p h i c a l
a b s t r a c t
• Unique exciplex emission response to different monoaromatic compounds.
• Metal-organic framework undergoes structural changes with analyte binding. • Fluorescence and sorption experiments show high selectivity for toluene and p-xylene. • Competitive binding selectivity has affects the interpretation of sensor response.
a r t i c l e
i n f o
Article history: Received 18 December 2013 Received in revised form 24 January 2014 Accepted 1 February 2014 Available online 10 February 2014 Keywords: Metal-organic framework Selectivity Fluorescence Sensor Aromatic
a b s t r a c t The two-fold interpenetrated metal-organic framework, [Zn2 (bdc)2 (dpNDI)]n (bdc = 1,4benzenedicarboxylate, dpNDI = N N -di(4-pyridyl)-1,4,5,8-naphthalenediimide) can undergo structural re-arrangement upon adsorption of chemical species changing its pore structure. For a competitive binding process with multiple analytes of different sizes and geometries, the interpenetrated framework will adopt a conformation to maximize the overall binding interactions. In this study, we show for binary mixtures that there is a high selectivity for the larger methylated aromatic compounds, toluene and p-xylene, over the small non-methylated benzene. The dpNDI moiety within [Zn2 (bdc)2 (dpNDI)]n forms an exciplex with these aromatic compounds. The emission wavelength is dependent on the strength of the host-guest CT interaction allowing these compounds to be distinguished. We show that the sorption selectivity characteristics can have a significant impact on the fluorescence sensor response of [Zn2 (bdc)2 (dpNDI)]n towards environmentally important hydrocarbons based contaminants (i.e., BTEX, PAH). Crown Copyright © 2014 Published by Elsevier B.V. All rights reserved.
1. Introduction Metal-organic frameworks (MOFs) typically have very high porosity and surface area, properties that are advantageous and
∗ Corresponding author at: CSIRO Earth Science and Resource Engineering, 26 Dick Perry Avenue, Kensington, WA, Australia. Tel.: +61 8 6436 8708; fax: +61 8 6436 8555. E-mail address: [email protected] (M. Myers). http://dx.doi.org/10.1016/j.aca.2014.02.004 0003-2670/Crown Copyright © 2014 Published by Elsevier B.V. All rights reserved.
enable many of their potential applications [1–3]; however, in comparison, interpenetrated metal-organic frameworks have relatively low porosity and surface area [4]. MOFs have been shown to be a promising material for a wide variety of applications including separation and storage of harmful gas [5], explosives detection/environmental monitoring [6] radioelement capturing/storage [7] and many others [8]. For environmental monitoring using chemical sensors, selectivity and sensitivity are undoubtedly the two most important properties [9–12]. Sensors for the in situ detection of aromatic
M. Myers et al. / Analytica Chimica Acta 819 (2014) 78–81
79
Fig. 1. Schematic of a two-fold interpenetrated metal-organic framework transformation illustrating the change in pore size.
Normalized Emission Intensity
a
100:0 benzene:toluene 99.9:0.1 benzene:toluene 99:1 benzene:toluene 90:10 benzene:toluene 0:100 benzene:toluene
1.0 0.8 0.6 0.4 0.2 0.0 400
2. Experimental 2.1. Materials [Zn2 (bdc)2 (dpNDI)]n was synthesized and combined with dried alumina powder in a 5:95 mass ratio according to the procedures published previously [17]. These solids were vacuum dried at 120 ◦ C to remove any solvents trapped within the MOF matrix. 2.2. Procedures For fluorescence measurements, approximately 10 mg of [Zn2 (bdc)2 (dpNDI)]n /alumina was added individually to 5 mL of varying benzene/toluene and benzene/p-xylene mixtures. For each, the MOF was exposed to the mixture for approximately 3 h until
450
500
550
600
Wavelength (nm)
b Normalized Emission Intensity
hydrocarbons are commonly based on UV absorbance spectroscopy and fluorescence spectroscopy techniques with several such sensors already commercially available [13]. However, with these types of sensors, it is difficult to adequately differentiate between the various BTEX (benzene, toluene, ethylbenzene and xylenes) compounds due to the very limited differences in their UV absorbance and fluorescence emission spectra [14]. Despite having comparatively low porosity and surface area, interpenetrated MOFs have the potential to adopt different pore dimensions by varying the displacement between the different lattices [15,16]. This ability of interpenetrated metal-organic frameworks to adopt differing geometries could lead to enhanced analyte selectivity (see Fig. 1). In this study, we examine the behavior of a fluorescing two-fold interpenetrated MOF towards mixtures of BTEX compounds and determine what impact sorption selectivity has on its sensor response. Recently, Takashima et al. have shown that upon exposure of the two-fold interpenetrated MOF [Zn2 (bdc)2 (dpNDI)]n (bdc = 1,4-benzenedicarboxylate, dpNDI = N N -di(4-pyridyl)1,4,5,8-naphthalenediimide) to individual aromatic analytes (e.g. benzene and toluene) that there are distinct differences in the wavelength of emission upon excitation with 365 nm light [17]. This result is based on the formation of an exciplex that is formed by the association of 1,4,5,8-naphthalenediimide (NDI) derivative with an aromatic compound [18]. By incorporating the dpNDI moiety into an interpenetrated MOF, there is a much higher binding affinity for aromatic compounds, resulting in an intensified excimer emission. However, understanding the fluorescence response of [Zn2 (bdc)2 (dpNDI)]n when exposed to mixtures vs. single chemical constituents is extremely important if potential sensing applications are to be realized. Binary mixtures of benzene/toluene and benzene/p-xylene are examined here with the goal of developing a better understanding of how interpenetrated MOFs selectively respond to mixtures in the context of chemical sensing.
100:0 benzene:p-xylene 99:1 benzene:p-xylene 95:1 benzene:p-xylene 90:10 benzene:p-xylene 0:100 benzene:p-xylene
1.0 0.8 0.6 0.4 0.2 0.0 400
450
500
550
600
Wavelength (nm) Fig. 2. (a) Fluorescence emission spectra (excitation wavelength of 365 nm) when exposed to benzene/toluene mixtures. (b) Fluorescence emission spectra (excitation wavelength of 365 nm) when exposed to benzene/p-xylene mixtures.
equilibrium was established, and then the mixture was thoroughly shaken and immediately subjected to examination by fluorescence spectroscopy (see Supplementary Information for details). Sorption experiments were undertaken by adding 10 mg of pure [Zn2 (bdc)2 (dpNDI)]n (vacuum dried at 120 ◦ C) to varying mixtures of benzene/toluene and benzene/p-xylene. After equilibrating in solution for approximately 24 h, the solvent was removed by filtration and then the remaining solids were vacuum dried at room temperature for 1 h to remove residual solvent. To extract the benzene/toluene or benzene/p-xylene from the MOF matrix, the MOF sample was stirred in 5 mL of methanol for approximately 24 h. The resulting solution was analyzed using gas chromatography/mass spectrometry (see Supporting Information for details). 3. Results and discussion 3.1. Fluorescence response We have confirmed the results of Takashima et al. showing that when [Zn2 (bdc)2 (dpNDI)]n is exposed to pure benzene, pure toluene and pure p-xylene an excimer emission is seen at 441 nm, 476 nm and 511 nm, respectively [17]. During our work, however, we discovered with mixtures of these solvents, that the wavelength of maximum emission intensity is strongly and non-linearly dependent on the relative concentrations of the solvents. For example, Fig. 2 shows that the fluorescence emission spectrum shifts to higher wavelengths with addition of even small amounts of either toluene or p-xylene to benzene. Fig. 3 shows that the shift of wavelength of maximum emission intensity is very abruptly as the proportion of toluene is increased to about 1% vol. or as the
M. Myers et al. / Analytica Chimica Acta 819 (2014) 78–81
Xtoluene / Xbenzene in extract solution
Wavelength at Maximum Emission Intensity
a 480 475 470 465 460 455
80
70
70
60
60
50
50
40
40
30
30
20
20
10
10
0
0 0.4
0.6
0.8
1.0
Xbenzene
450 0.1
1
10
100 Fig. 4. Plots of the mole fraction ratios for benzene/toluene and benzene/p-xylene mixtures vs. mole fraction of benzene.
% Toluene by Volume
b Wavelength at Maximum Emission Intensity
80
Xp-xylene / Xbenzene in extract solution
80
520 510 500 490 480 470 460 450 440 1
10
100
% p-xylene by volume
Fig. 3. (a) Plot of wavelength at maximum emission intensity vs. % toluene by volume in different benzene/toluene mixtures. In pure benzene the maximum intensity is at 441 nm. (b) Plot of wavelength at maximum emission intensity vs. % p-xylene by volume for MOFs in different benzene/toluene mixtures. In pure benzene the maximum intensity is at 441 nm.
proportion of p-xylene in increased to about 2 vol.% and then remains nearly constant with higher concentrations of either toluene or p-xylene in benzene. For example, with a 99:1 v/v ratio of benzene to toluene the emission maximum wavelength has shifted 18 nm relative to pure benzene; whereas, for a 50:50 v/v ratio of benzene to toluene the emission maximum wavelength has shifted only 3 nm relative to pure toluene. This result clearly demonstrates that this MOF is much more sensitive to either toluene or p-xylene compared to benzene. Takashima et al. have reported that the fluorescence quantum yields are 5%, 22% and 19% for benzene, toluene and p-xylene, respectively [17]. Comparing the sets of fluorescence spectra for benzene/toluene (Fig. 2a) and benzene/p-xylene, the curves for benzene/toluene all have similar widths; whereas, for intermediate ratios of benzene/p-xylene there is significant broadening. We initially hypothesized that the variation of fluorescence properties with solvent may just be an artifact of this difference in quantum yield and we conducted a series of sorption experiments using GCMS to test this hypothesis. 3.2. Sorption characteristics With the different mixtures of benzene/toluene and benzene/pxylene, we have analyzed the composition of methanol extracts
for MOFs that were exposed to mixtures of the BTX compounds to determine binary mixture selectivity coefficients. Not surprisingly, there is a high selectivity for toluene over benzene (
