Research article Special Issues

Study of a complex environmental mixture by electrospray ionization and laser desorption ionization high resolution mass spectrometry: the cigarette smoke aerosol

  • Aerosols from the mainstream cigarette smoke have been analyzed by electrospray ionization (ESI) coupled to Fourier transform ion cyclotron resonance mass spectrometry (FTICRMS). Measurements have been conducted in positive ion mode. The chemical composition of cigarette smoke aerosol is significant because it gives insights of one complex indoor environmental mixture. Almost 1300 chemical compositions relative to nitrogen species were successfully determined through the accurate mass measurement and the good ion production of the used technique. The most abundant class of compounds corresponds to N2 one (hydrocarbons with two nitrogen atoms). For other classes, the van Krevelen diagrams ensured to define that other nitrogen and oxygen-nitrogen compounds adopted similar behavior in terms of unsaturation and alkylation range. The detailed composition of cigarette smoke aerosol provided a typical chemical fingerprint from the biomass pyrolysis with tobacco-specific compounds. We examined also the contribution of laser desorption ionization (LDI) technique coupled to FTICRMS for the acute analysis of cigarette smoke aerosol. While a part of the chemical composition were found similar to ESI results, LDI achieved a broader range of poly-aromatic compounds and highlight new class compounds as pure hydrocarbon and oxygen-containing species. The combination of ESI and LDI with high resolution mass spectrometry clearly increased significantly the coverage of the “whole composition” of environmental aerosols such as cigarette smoke aerosol.

    Citation: Vincent Carré, Sébastien Schramm, Frédéric Aubriet. Study of a complex environmental mixture by electrospray ionization and laser desorption ionization high resolution mass spectrometry: the cigarette smoke aerosol[J]. AIMS Environmental Science, 2015, 2(3): 547-564. doi: 10.3934/environsci.2015.3.547

    Related Papers:

    [1] Ahmed Z. Abdullah, Adawiya J. Haider, Allaa A. Jabbar . Pure TiO2/PSi and TiO2@Ag/PSi structures as controllable sensor for toxic gases. AIMS Materials Science, 2022, 9(4): 522-533. doi: 10.3934/matersci.2022031
    [2] Angaraj Singh, Ajitanshu Vedrtnam, Kishor Kalauni, Aman Singh, Magdalena Wdowin . Synthesis routes of zeolitic imidazolate framework-8 for CO2 capture: A review. AIMS Materials Science, 2025, 12(1): 118-164. doi: 10.3934/matersci.2025009
    [3] Laís Weber Aguiar, Cleiser Thiago Pereira da Silva, Hugo Henrique Carline de Lima, Murilo Pereira Moises, Andrelson Wellington Rinaldi . Evaluation of the synthetic methods for preparing metal organic frameworks with transition metals. AIMS Materials Science, 2018, 5(3): 467-478. doi: 10.3934/matersci.2018.3.467
    [4] Jamal Alnofiay, Ahmed Al-Shahrie, Elsayed Shalaan . Green synthesis of high-performance gallium oxide supercapacitor: A path to outstanding energy density. AIMS Materials Science, 2024, 11(6): 1065-1082. doi: 10.3934/matersci.2024051
    [5] Akira Nishimura, Tadaki Inoue, Yoshito Sakakibara, Masafumi Hirota, Akira Koshio, Fumio Kokai, Eric Hu . Optimum molar ratio of H2 and H2O to reduce CO2 using Pd/TiO2. AIMS Materials Science, 2019, 6(4): 464-483. doi: 10.3934/matersci.2019.4.464
    [6] Akira Nishimura, Ryuki Toyoda, Daichi Tatematsu, Masafumi Hirota, Akira Koshio, Fumio Kokai, Eric Hu . Optimum reductants ratio for CO2 reduction by overlapped Cu/TiO2. AIMS Materials Science, 2019, 6(2): 214-233. doi: 10.3934/matersci.2019.2.214
    [7] Masatoshi Sakairi, Hirotaka Mizukami, Shuji Hashizume . Effects of solution composition on corrosion behavior of 13 mass% Cr martensitic stainless steel in simulated oil and gas environments. AIMS Materials Science, 2019, 6(2): 288-300. doi: 10.3934/matersci.2019.2.288
    [8] Pravina Kamini G., Kong Fah Tee, Jolius Gimbun, Siew Choo Chin . Biochar in cementitious material—A review on physical, chemical, mechanical, and durability properties. AIMS Materials Science, 2023, 10(3): 405-425. doi: 10.3934/matersci.2023022
    [9] Toni Varila, Henrik Romar, Tero Luukkonen, Ulla Lassi . Physical activation and characterization of tannin-based foams enforced with boric acid and zinc chloride. AIMS Materials Science, 2019, 6(2): 301-314. doi: 10.3934/matersci.2019.2.301
    [10] Endi Suhendi, Andini Eka Putri, Muhamad Taufik Ulhakim, Andhy Setiawan, Syarif Dani Gustaman . Investigation of ZnO doping on LaFeO3/Fe2O3 prepared from yarosite mineral extraction for ethanol gas sensor applications. AIMS Materials Science, 2022, 9(1): 105-118. doi: 10.3934/matersci.2022007
  • Aerosols from the mainstream cigarette smoke have been analyzed by electrospray ionization (ESI) coupled to Fourier transform ion cyclotron resonance mass spectrometry (FTICRMS). Measurements have been conducted in positive ion mode. The chemical composition of cigarette smoke aerosol is significant because it gives insights of one complex indoor environmental mixture. Almost 1300 chemical compositions relative to nitrogen species were successfully determined through the accurate mass measurement and the good ion production of the used technique. The most abundant class of compounds corresponds to N2 one (hydrocarbons with two nitrogen atoms). For other classes, the van Krevelen diagrams ensured to define that other nitrogen and oxygen-nitrogen compounds adopted similar behavior in terms of unsaturation and alkylation range. The detailed composition of cigarette smoke aerosol provided a typical chemical fingerprint from the biomass pyrolysis with tobacco-specific compounds. We examined also the contribution of laser desorption ionization (LDI) technique coupled to FTICRMS for the acute analysis of cigarette smoke aerosol. While a part of the chemical composition were found similar to ESI results, LDI achieved a broader range of poly-aromatic compounds and highlight new class compounds as pure hydrocarbon and oxygen-containing species. The combination of ESI and LDI with high resolution mass spectrometry clearly increased significantly the coverage of the “whole composition” of environmental aerosols such as cigarette smoke aerosol.


    1. Introduction

    Metal--organic materials (MOMs) are a class of synthesized, often porous, and crystalline materials that have comprised the focus of a large amount of experimental and theoretical studies for the past few decades [1,2,3]. The application of MOMs is very diverse and has become essentially ubiquitous in scientific research, ranging from gas sorption [3,4,5,6,7,8,9,10,11,12,13,14,15,16,17], separation [4,18,19], catalysis [20,21,22,23], sensing [24,25], photoinduced electron transfer [26,27,28,29] and biological uses [30,31]. The major reason for this is that there are conceivably infinite variations of MOM structures that can be synthesized by altering their component metal ions, organic linkers, secondary building units (SBUs) [32,33], solvents, and thermodynamic or physical conditions.

    This study focuses on explaining the sorption properties of CO$_2$ and C$_2$H$_2$ in two isostructural porous materials. A method that separates these gases facilely from a mixture is highly sought for since CO$_2$ is one of the main contaminants in C$_2$H$_2$. The latter is an important feedstock chemical that is used for myriad products, such as pharmaceuticals, plastics, and fuels. There is also a growing need for technology that can mitigate the effects that CO$_2$ has on the environment. Indeed, the detrimental results of excess atmospheric CO$_2$ are both already observed and predicted to propogate in human health and the environment if steps are not taken to combat it [34]. A material that can effectively sorb and/or separate these two gases is therefore desirable and an understanding of the mechanism of such properties in candidate materials can pave the way for better, safer, and cheaper porous materials.

    Recent studies involved examining C$_2$H$_2$/CO$_2$ separation in hybrid ultramicroporous materials (HUMs), a subclass of MOMs that combine ultramicropores with strong electrostatics from inorganic anions that line the pore surface [35,36]. The materials of interest in this particular study are classified as molecular porous materials (MPMs), a class of molecular solids that are held together through weak noncovalent interactions (e.g., hydrogen bonding) as opposed to the conventional fully covalently bonded interactions found in most MOMs. The combination of adenine and Cu$^{+2}$ salts of Cl and Br forms self-assembled [Cu$_2$(adenine)$_4$Cl$_2$]Cl$_2$ and [Cu$_2$(adenine)$_4$Br$_2$]Br$_2$, known as MPM-1-Cl and MPM-1-Br, respectively [37]. Orthographic views of the 1 $\times$ 1 $\times$ 2 supercell of MPM-1-Br are shown in Figure 1.

    Figure 1. (a) Orthographic $a$-axis view, (b) $b$-axis view, and (c) $c$-axis view of the 1 $\times$ 1 $\times$ 2 supercell of MPM-1-Br used for the simulations in this study. The crystal structure of MPM-1-Cl is very similar. Atom colors are C = cyan; N = blue; H = white; Cu = brown; Br = red.

    The syntheses of MPM-1-Cl and MPM-1-Br are reported in references [38,39], respectively. Both MPMs feature an interesting hydrogen-bonding network in which (1) four hydrogen atoms from four different adenine linkers are hydrogen-bonded to a single halide ion and (2) an adenine linker from one [Cu$_2$(adenine)$_4$]$^{4+}$ unit exhibits hydrogen-bonding contacts with an adenine linker from another unit. Unlike most MOMs that contain dinuclear copper paddlewheel clusters [3,40], these MPMs contain halogen atoms bonded to the Cu$^{2+}$ ions of the copper paddlewheels at the axial position to create a saturated metal center. The two MPMs have a minor kink in the X--Cu--Cu--X (X = Cl or Br) motif such that the angle is not 180$^\circ$. Previous theoretical studies on MPMs [41] involved comparing the CO$_2$ and H$_2$ sorption mechanism in MPM-1-Cl and MPM-1-TIFSIX [42], with latter containing TiF$_6^{2-}$ ions in place of the halide ions. Overall, MPMs are of interest due to their thermal and water stability as well as their ease of synthesis compared to most MOMs. When these attributes are combined with their decent gas sorption and separation properties, MPMs could be good and interesting candidates for materials research.

    The purpose of this study is to elucidate (with atomistic resolution) the sorption properties of CO$_2$ and C$_2$H$_2$ in the isostructural MPM-1-Cl and MPM-1-Br analogues. Insights into the reason for why both MPMs prefer C$_2$H$_2$ over CO$_2$ will be obtained. Another interest of this work is examining how the electronic environment (and thus MPM--sorbate interaction) is effected by the use of a different halogen in the material. It will be shown that the van der Waals radius of the halogen and its polarizability play a key role in determining the location of the favorable sorption site and the energetics of binding in the material. To our knowledge, this is the first study reporting simulations of C$_2$H$_2$ sorption in these materials as well as being the first theoretical study of MPM-1-Br.


    2. Methods

    The potential energy function for MPM-1-Cl was developed by our group in previous work [41] and utilized herein. The crystal structure for MPM-1-Br was obtained from reference [39]. For all simulations in both MPMs, the sorbent atoms were treated as rigid to accomodate a constant volume ensemble system. This approximation is especially valid when phononic effects are minor [43]. As with previous work on MPM-1-Cl [41], all atoms of MPM-1-Br were given Lennard-Jones 12--6 parameters, point partial charges, and scalar point polarizabilities to model repulsion/dispersion, stationary electrostatic, and explicit polarization, respectively. The Lennard-Jones parameters for all MPM atoms were taken from the Universal Force Field (UFF)[44], while the exponentially-damped polarizabilities for all atoms other than Cu were obtained from van Duijnen et al. [45]. The polarizability parameter for Cu$^{2+}$ was determined in reference [46] and used herein. The point partial charges for the chemically unique atoms were determined by averaging atomic charges computed via a least squares fit [47] to the electrostatic potential surface of representative MPM fragments (see Supporting Information, Figure S1) generated by quantum mechanical calculations using NWChem [48].

    The potentials used for CO$_2$ were the nonpolarizable CO$_2$-PHAST [49] and TraPPE [50] models and the polarizable CO$_2$-PHAST* model [49]. C$_2$H$_2$-PHAST and C$_2$H$_2$-PHAST* [35,36] were used to simulate C$_2$H$_2$ sorption in both MPMs; the * denotes the inclusion of explicit polarization. Simulations of CO$_2$ and C$_2$H$_2$ sorption in MPM-1-Cl and MPM-1-Br were performed using grand canonical Monte Carlo (GCMC) methods [51]. The sorption simulations, theoretical isosteric heat of adsorption ($Q_{st}$) calculation, binding site analysis, etc. were carried out using the codes developed by our group [52,53]. Further details of performing the simulations and parametrizations are presented in the Supporting Information. All relevant simulation parameters for both MPMs and the gases simulated are also provided in the Supporting Information (Tables S1--S4).

    Simulated annealing calculations were performed using the polarizable CO$_2$ and C$_2$H$_2$ potentials in both MPMs to determine the classical binding energy for each sorbate in the respective materials. This is a computational method which involves executing a normal Monte Carlo procedure (using the MPM with a single sorbate molecule) that is tempered by reducing the temperature with each accepted perturbation [54]. As the temperature decreases, the guest molecule settles into an energy minima whose magnitude can be compared with the $Q_{st}$ near zero loading.


    3. Summary of experimental data


    3.1. C$_2$H$_2$ uptake

    C$_2$H$_2$ uptake in MPM-1-Cl and MPM-1-Br was experimentally measured by Xie et al. [37] at 273 and 298 K and pressures up to 1 atm; the resulting isotherms (as estimated from [37]) are shown in Figure 2a. It can be observed that as the pressure increases at 273 K, the Cl variant exhibits mildly greater affinity for C$_2$H$_2$ than the Br analogue. This could be due to the Cl$^-$ ions having a smaller ionic radius than Br$^-$, resulting in a slightly larger pore volume in MPM-1-Cl. This increased accessible volume in MPM-1-Cl relative to MPM-1-Br can allow for greater number of C$_2$H$_2$ molecules to occupy the pore space in the material at conditions approaching saturation.

    Figure 2. Experimental gas sorption isotherms of (a) C$_2$H$_2$ and (b) CO$_2$ for MPM-1-X (X = Br, circles or Cl, triangles). at two different temperatures (blue = 273 K; red = 298 K) and pressures up to 1 atm. All experimental data were estimated from [37].

    Interestingly, the relative uptake trend is reversed at 298 K: MPM-1-Br shows slightly greater affinity for C$_2$H$_2$ compared to MPM-1-Cl for all pressures considered at this temperature. The observed phenomenon may be caused by the increased kinetic energy of the C$_2$H$_2$ molecules and the Br variant at higher temperatures. This permits the C$_2$H$_2$ molecules to diffuse through and collide with MPM-1-Br at a greater rate than at 273 K. Since Br is heavier than Cl, the Br$^-$ ions likely vibrate more slowly than the Cl$^-$ ions, thus making MPM-1-Br more rigid. When the C$_2$H$_2$ molecules are moving more quickly at higher temperatures, the favorable geometry of the MPM-1-Br framework is less flexible, whereas the higher kinetic energy of gas molecules may induce a slightly unfavorable conformational change for the Cl analogue. This is further emphasized from the fact that the primary binding site discovered for C$_2$H$_2$ in MPM-1-Br has a lower potential energy and closer distance to the Cu paddlewheel than the analogous site in MPM-1-Cl; this will be explained below in Section 4.3.


    3.2. CO$_2$ uptake

    Xie et al. also measured CO$_2$ sorption in MPM-1-Cl and MPM-1-Br under the same conditions as C$_2$H$_2$ [37]; these isotherms are presented in Figure 2b. We note that Zaworotko's group also reported CO$_2$ sorption data in the Cl analogue previously [41,42]. It can be seen that MPM-1-Cl sorbs a greater quantity of CO$_2$ for nearly all state points considered. This phenomenon may be explained by Cl exhibiting a higher electronegativity than Br, which allows for greater electrostatic interactions between the halide ion and the CO$_2$ molecule. Although CO$_2$ has no permanent dipole moment in the bulk, it displays high induced dipoles when interacting with the MPM framework [41]. The interaction between the Cl$^-$ ion and the partial positive carbon atom of the CO$_2$ molecule is fortified with a greater electronegativity compared to Br$^-$. Our electronic structure calculations in both MPMs confirmed that the magnitude of the partial negative charges of the Cl$^{-}$ ions in MPM-1-Cl are greater than those for the Br$^{-}$ ions in MPM-1-Br (see Supporting Information, Table S1).


    3.3. Isosteric heats of adsorption

    Xie et al. [37] derived the experimental $Q_{st}$ values for C$_2$H$_2$ and CO$_2$ in MPM-1-Cl and MPM-1-Br by applying the virial method [55] to the corresponding experimental isotherms. These $Q_{st}$ values are plotted as a function of gas loading in Figure 3. For both MPMs, the $Q_{st}$ for C$_2$H$_2$ is higher than that for CO$_2$ at all loadings considered, thus indicating that these two MPMs display greater preference for C$_2$H$_2$ over CO$_2$. Table 1 provides a summary of the zero-loading $Q_{st}$ values ($Q_{st}^0$) for C$_2$H$_2$ and CO$_2$ in both MPMs.

    Figure 3. Experimental isosteric heats of adsorption ($Q_{st}$) for C$_2$H$_2$ (circles) and CO$_2$ (triangles) for MPM-1-X (X = Br, blue or Cl, red). All experimental data were estimated from [37].
    Table 1. Summary of data from key state points for gas sorption in MPM-1-Cl and MPM-1-Br. Experimental values [37] are compared with simulation using various potentials for each sorbate. $Q_{st}^0$ is the $Q_{st}$ value for the lowest reported loading. Units are $^a$kJ mol$^{-1}$, $^b$mmol g$^{-1}$.
    CO$_2$C$_2$H$_2$
    MPM-1-ClExp.SimulationExp.Simulation
    ModelCO$_2$-PHAST*CO$_2$-PHASTTraPPEC$_2$H$_2$-PHAST*C$_2$H$_2$-PHAST
    $Q_{st}^0$$^a$23.7622.7724.3326.2228.5725.4925.65
    0.05 atm loading, 273 K$^b$0.260.230.360.800.801.080.81
    1.0 atm loading, 273 K$^b$3.503.634.755.063.865.055.25
    0.05 atm loading, 298 K$^b$0.120.130.170.350.340.340.25
    1.0 atm loading, 298 K$^b$1.972.072.743.602.784.214.57
    MPM-1-BrExp.SimulationExp.Simulation
    ModelCO$_2$-PHAST*CO$_2$-PHASTTraPPEC$_2$H$_2$-PHAST*C$_2$H$_2$-PHAST
    $Q_{st}^0$$^a$21.6125.0225.5725.4025.0530.6127.55
    0.05 atm loading, 273 K$^b$0.190.230.330.510.811.411.23
    1.0 atm loading, 273 K$^b$2.542.523.203.773.593.693.66
    0.05 atm loading, 298 K$^b$0.120.110.130.200.430.340.22
    1.0 atm loading, 298 K$^b$1.561.521.802.472.823.313.30
     | Show Table
    DownLoad: CSV

    With regards to C$_2$H$_2$, MPM-1-Cl exhibits a $Q_{st}^0$ value of 28.57 kJ mol$^{-1}$, while that for MPM-1-Br is 25.05 kJ mol$^{-1}$. This suggests that MPM-1-Cl displays a more favorable initial loading site for C$_2$H$_2$ than MPM-1-Br according to experiment. The C$_2$H$_2$ $Q_{st}$ plot for MPM-1-Cl exhibits an expected decrease in $Q_{st}$ with each successive C$_2$H$_2$ molecule sorbed in the MPM at low loading; however, after ~1.5 mmol g$^{-1}$ loading is reached, the $Q_{st}$ values begin to increase and eventually surpass the $Q_{st}^0$ value. The analogous plot for MPM-1-Br shows that the $Q_{st}$ continually increases as the loading increases. Overall, for both MPMs, the increase in $Q_{st}$ as a function of loading could be due to a combination of MPM--sorbate and cooperative sorbate--sorbate interactions [41,56]. Interestingly, while the experimentally reported C$_2$H$_2$ $Q_{st}$ for MPM-1-Br is higher than that for MPM-1-Cl at high loadings ($>$1 mmol g$^{-1}$), the reported C$_2$H$_2$ uptake is higher for MPM-1-Cl at 273 K and high pressures. We believe that this could be due to the empirical fitting method employed to obtain the $Q_{st}$ values from experimental isotherms. This is discussed in more detail in Section 4.1.3.

    The $Q_{st}^0$ value for CO$_2$ in MPM-1-Cl and MPM-1-Br is 23.76 and 21.61 kJ mol$^{-1}$, respectively. Further, the CO$_2$ $Q_{st}$ for MPM-1-Cl is greater than that for MPM-1-Br across the considered loading range. This implies that CO$_2$ sorption is more favorable in MPM-1-Cl than MPM-1-Br, presumably due to the greater electronegativity and calculated partial negative charge of the Cl$^{-}$ ion as explained in the previous subsection. Interestingly, the CO$_2$ $Q_{st}$ plot for MPM-1-Cl shows the same behavior as the C$_2$H$_2$ $Q_{st}$ plot for MPM-1-Br. Likewise, the shape of the $Q_{st}$ plot for CO$_2$ in MPM-1-Br is similar to that for the C$_2$H$_2$ $Q_{st}$ plot for MPM-1-Cl. As with C$_2$H$_2$, both MPMs exhibit increasing $Q_{st}$ with increasing loading in their CO$_2$ $Q_{st}$ plots, which indicates that CO$_2$--CO$_2$ interactions are favorable in these materials.


    4. Simulation results and discussion


    4.1. C$_2$H$_2$ sorption


    4.1.1. MPM-1-Cl

    The experimental and simulated C$_2$H$_2$ sorption isotherms in MPM-1-Cl at 273 and 298 K are displayed in Figure 4a. At 273 K, the calculated uptakes for the nonpolarizable C$_2$H$_2$-PHAST and polarizable C$_2$H$_2$-PHAST* models are in good agreement with experiment at pressures of 0.05 atm and lower. At higher pressures, however, the simulated uptakes for both potentials are significantly higher than experiment. At 298 K, both models generated sorption isotherms that show very good agreement with experiment for pressures up to 0.2 atm, while they notably oversorb experiment at greater pressures. The fact that the simulated uptakes for the two potentials are in close agreement with experiment at low pressures for both temperatures suggest proper modeling of the initial C$_2$H$_2$ binding sites in this material. The significant oversorption compared to experiment implies that the simulations produced inflated sorbate--sorbate interactions in the MPM relative to what is physically captured in experiment, which could be attributed to inadequacies in the present C$_2$H$_2$ potentials. It can be observed that the polarizable model produced uptakes that are somewhat closer to experiment at higher pressures.

    Figure 4. Simulated gas sorption isotherms of (a) C$_2$H$_2$ and (b) CO$_2$ for MPM-1-Cl compared with experiment at two different temperatures (blue = 273 K; red = 298 K) and pressures up to 1 atm. Experiment = pure solid line; C$_2$H$_2$/CO$_2$-PHAST model = solid line with triangles; C$_2$H$_2$/CO$_2$-PHAST* model = solid line with stars; TraPPE model = solid line with circles.

    4.1.2. MPM-1-Br

    Figure 5a shows the simulated C$_2$H$_2$ sorption isotherms in MPM-1-Br at 273 and 298 K compared to experiment. As with MPM-1-Cl, simulations using the C$_2$H$_2$-PHAST and C$_2$H$_2$-PHAST* models in MPM-1-Br at both temperatures generally produced uptakes that are in good agreement with experiment at low pressures and those that are remarkably higher than experiment at higher pressures. Both models yielded comparable isotherms in MPM-1-Br at 273 K, with uptakes that are in closer agreement with experiment as the pressure approaches 1 atm. This could be a consequence of the material reaching C$_2$H$_2$ saturation under these conditions. At 298 K, the uptakes for the C$_2$H$_2$-PHAST* model is in better agreement with experiment at pressures lower than 0.1 atm.

    Figure 5. Simulated gas sorption isotherms of (a) C$_2$H$_2$ and (b) CO$_2$ for MPM-1-Br compared with experiment at two different temperatures (blue = 273 K; red = 298 K) and pressures up to 1 atm. Experiment = pure solid line; C$_2$H$_2$/CO$_2$-PHAST model = solid line with triangles; C$_2$H$_2$/CO$_2$-PHAST* model = solid line with stars; TraPPE model = solid line with circles.

    As shown in Table 1, the experimental C$_2$H$_2$ uptakes for MPM-1-Br at both temperatures are higher than those for MPM-1-Cl at low pressures (0.05 atm), whereas such uptakes for the former are lower than those for the latter at higher pressures (1.0 atm). This trend is consistent with what was observed in the simulations when utilizing both C$_2$H$_2$ potentials. Therefore, MPM-1-Br exhibits greater interactions with the C$_2$H$_2$ molecules than MPM-1-Cl at low loadings, which could be explained by the Br$^{-}$ ions having a larger polarizability and van der Waals parameters than Cl$^{-}$ ions (see Supporting Information, Table S2). Nevertheless, at higher pressures, MPM-1-Br displays lower C$_2$H$_2$ uptake than the Cl analogue because less accessible space is available in the material under these conditions due to the larger ionic radius of the Br$^{-}$ ions. Indeed, the shape of the simulated C$_2$H$_2$ sorption isotherms for MPM-1-Br suggests that the material is near C$_2$H$_2$ saturation at ca. 3.5 mmol g$^{-1}$ loading.


    4.1.3. Isosteric heats of adsorption

    The simulated C$_2$H$_2$ $Q_{st}$ values for MPM-1-Cl and MPM-1-Br are compared with their corresponding experimental $Q_{st}$ plots in Figures 6a and 7a, respectively. It can be observed that simulations using the two models captured the apparent increase in $Q_{st}$ as the loading increases for both MPMs. This shape is consistent with the experimental C$_2$H$_2$ $Q_{st}$ plot for MPM-1-Br. Overall, the trend in the simulated $Q_{st}$ values for MPM-1-Cl and MPM-1-Br support the experimental observation that both MPMs tend to facilitate C$_2$H$_2$ sorption such that initially loaded molecules tend to cooperatively interact with each other to lower the potential energy of subsequent guest molecules, thereby leading to enhanced sorbate--sorbate interactions at higher loadings. When considering all loadings, both models produced $Q_{st}$ values that are within the vicinity of experiment for both MPMs.

    Figure 6. Simulated isosteric heat of sorption ($Q_{st}$) for (a) C$_2$H$_2$ and (b) CO$_2$ for MPM-1-Cl compared with experiment. Experiment = solid black line; Simulation = solid blue line with symbols (C$_2$H$_2$/CO$_2$-PHAST = triangles; C$_2$H$_2$/CO$_2$-PHAST* = stars; TraPPE model = circles).
    Figure 7. Simulated isosteric heat of sorption ($Q_{st}$) for (a) C$_2$H$_2$ and (b) CO$_2$ for MPM-1-Br compared with experiment. Experiment = solid black line; Simulation = solid blue line with symbols (C$_2$H$_2$/CO$_2$-PHAST = triangles; C$_2$H$_2$/CO$_2$-PHAST* = stars; TraPPE model = circles).

    The theoretical $Q^0_{st}$ value for the C$_2$H$_2$-PHAST and C$_2$H$_2$-PHAST* models in MPM-1-Cl are 25.49 and 25.65 kJ mol$^{-1}$, respectively, which are lower than the corresponding experimental value of 28.57 kJ mol$^{-1}$ (Table 1). Simulations in MPM-1-Br yielded $Q^0_{st}$ values of 30.61 and 27.55 kJ mol$^{-1}$ for the nonpolarizable and polarizable models, respectively, which are higher than the experimental value for this MPM (25.05 kJ mol$^{-1}$). In general, our simulations suggest that MPM-1-Br exhibits a higher $Q^0_{st}$ value for C$_2$H$_2$ than MPM-1-Cl, whereas the opposite trend is observed according to experiment. From a molecular point of view, it is expected that MPM-1-Br would display a greater affinity for C$_2$H$_2$ due to the higher polarizability of the Br$^{-}$ ions. Such an effect is probably captured in our simulations that include both implicit and explicit polarization. This behavior is not reflected in the reported experimental C$_2$H$_2$ $Q_{st}$ for both MPMs even though the experimental isotherms at both temperatures suggest that MPM-1-Br outperforms MPM-1-Cl for C$_2$H$_2$ uptake at low pressures (see Table 1). It is notable that the experimental $Q_{st}$ values are derived from an empirical fitting procedure on the measured sorption isotherms, whereas the theoretical values are obtained directly from GCMC simulation [57]. We could therefore attribute the discrepancy in the relative $Q^0_{st}$ trend between experiment and simulation to the empirical fitting that was implemented to extract the experimental $Q_{st}$ values.


    4.2. CO$_2$ sorption


    4.2.1. MPM-1-Cl

    The simulated CO$_2$ uptakes for three different potentials (CO$_2$-PHAST, CO$_2$-PHAST*, and TraPPE) in MPM-1-Cl at 273 and 298 K are shown in Figure 4b. The isotherms produced by the CO$_2$-PHAST and TraPPE models oversorb experiment for nearly all state points considered. On the other hand, the isotherms for the CO$_2$-PHAST* model are in excellent agreement with experiment across the considered pressure range at both temperatures. The strong agreement between the experimental and simulated isotherms for the CO$_2$-PHAST* model demonstrates that this potential is highly accurate for simulations in heterogeneous media [49]. The TraPPE model yielded the highest calculated CO$_2$ uptakes in MPM-1-Cl out of all models. Indeed, at 273 K, the calculated uptakes for this model are $>$1.5 mmol g$^{-1}$ higher than experiment at pressures of 0.2 atm and beyond. Although the TraPPE model is widely used, it has been shown to generate isotherms that notably oversorb experiment in many porous materials [3,16,58,59,60].

    The CO$_2$-PHAST* model yielded the lowest CO$_2$ uptakes in MPM-1-Cl out of all models for all considered state points even though it includes explicit many-body polarization. Generally, the theoretical gas uptake in MOMs increases with increasing terms in the potential energy function [46]. This was not the case here as the CO$_2$-PHAST model gives uptakes that are higher than those for its polarizable counterpart. It is predicted that the inclusion of polarization produces a favorable sorbate ordering effect in this MPM, which causes the CO$_2$ uptake to decrease compared to when only stationary electrostatic interactions are considered. A similar phenomenon was observed from theoretical studies in a HUM [56]. We note that CO$_2$ sorption in MPM-1-Cl was investigated previously through GCMC simulations using only the CO$_2$-PHAST* model [41]. Herein, we show the results that were obtained using different CO$_2$ potentials in this material.


    4.2.2. MPM-1-Br

    A comparison of the experimental and simulated CO$_2$ sorption isotherms for the three different potentials in MPM-1-Br at 273 and 298 K is presented in Figure 5b. The results for the three models in this MPM are consistent with that for MPM-1-Cl. Notably, the isotherms for the TraPPE model significantly oversorb experiment at both temperatures. The amount of oversorption for this model is generally proportional to the pressure. The CO$_2$-PHAST model produces the next highest calculated CO$_2$ uptakes for all pressures, followed by the CO$_2$-PHAST* model. As observed in MPM-1-Cl, the CO$_2$-PHAST* model yields uptakes that are in very good agreement with experiment for all pressures at both temperatures, thus further illustrating the accuracy of this potential for simulation CO$_2$ sorption in porous materials. The isotherm for the nonpolarizable CO$_2$-PHAST model is in good agreement with experiment at 298 K, but somewhat oversorbs experiment at 273 K. The increase in temperature may provide enough kinetic energy to decrease electrostatic interactions between the Br$^-$ ions and the CO$_2$ molecule; this could explain why the CO$_2$-PHAST model shows better agreement with experiment at the higher temperature.

    As shown in Table 1, the experimental atmospheric CO$_2$ uptakes in MPM-1-Br at 273 and 298 K are 2.54 and 1.56 mmol g$^{-1}$, respectively. The corresponding values for MPM-1-Cl are 3.50 and 1.97 mmol g$^{-1}$, respectively, which are higher than those for MPM-1-Br. Our simulations confirmed the experimental finding that the CO$_2$ uptakes under these thermodynamic conditions are greater for MPM-1-Cl than MPM-1-Br; this is true regardless of the CO$_2$ potential that was used. It is expected that CO$_2$ sorption in these MPMs is governed by electrostatic interactions between the halide ion and the positively charged carbon atom of the CO$_2$ molecule. MPM-1-Br displays weaker interactions with the CO$_2$ molecules than MPM-1-Cl, probably due to the lower calculated partial negative charges for the Br$^-$ ions compared to the Cl$^-$ ions (see Supporting Information, Table S1). Lower magnitudes for the partial negative charge of the halide ion will result in decreased electrostatic interactions with the CO$_2$ molecule.

    Note, we also performed simulations of CO$_2$ sorption in both MPMs using the EPM2 [61] model and the resulting simulated CO$_2$ sorption isotherms are provided in the Supplementary Information, Figure S3. The calculated isotherms for this model are comparable to those for the CO$_2$-PHAST model; both potentials generated uptakes that oversorb experiment at higher pressures. Overall, the polarizable CO$_2$-PHAST* model produced isotherms that are in closest agreement with experiment. As a result, the binding sites that are discussed later (in Section 4.3) are based on simulations using this model.


    4.2.3. Isosteric heats of adsorption

    The simulated CO$_2$ values for the CO$_2$-PHAST, CO$_2$-PHAST*, and TraPPE models for MPM-1-Cl and MPM-1-Br are displayed in Figures 6b and 7b, respectively. According to our simulations, the relative trend in the magnitude of the CO$_2$ $Q_{st}$ values in both MPMs is: TraPPE $>$ CO$_2$-PHAST $>$ CO$_2$-PHAST*. This is consistent with the pattern that was observed for the simulated uptakes for the three models in both MPMs. The simulated $Q_{st}$ values for the CO$_2$-PHAST and CO$_2$-PHAST* models in MPM-1-Cl are in good agreement with experiment as well as each other for all considered loadings. The TraPPE model $Q_{st}$ values for this MPM are somewhat higher than experiment across the loading range. The $Q_{st}$ plots for all models in MPM-1-Cl show increasing $Q_{st}$ with higher loadings, which is consistent with the shape for the experimental plot. Therefore, our simulations captured the expected cooperative CO$_2$--CO$_2$ interactions in this MPM.

    In MPM-1-Br, the $Q_{st}$ values for all models are consistently higher than experiment for all loadings, with values produced by the polarizable CO$_2$-PHAST* model being closer to experiment. The experimental $Q_{st}$ values in this MPM begins to increase when the material is partially saturated at ~1.0 mmol g$^{-1}$ loading, indicating cooperativity between the CO$_2$ molecules under these conditions. Unlike in MPM-1-Cl, the $Q_{st}$ plot for all three models in MPM-1-Br does not show much signs of sorbate--sorbate cooperativity in the material. Rather, the theoretical $Q_{st}$ values slightly decreases with increasing CO$_2$ uptake, as does the experiment. In addition, the simulated $Q_{st}$ plots level off at a loading of ~0.4 mmol g$^{-1}$, but there is neither a sustained or defined increasing trend in $Q_{st}$ as the uptakes continue to increase.

    The calculated $Q^0_{st}$ values for all three potentials in both MPMs are summarized in Table 1. The $Q^0_{st}$ values produced by the CO$_2$-PHAST, CO$_2$-PHAST*, and TraPPE models in MPM-1-Cl are 24.33, 22.77 and 26.22 kJ mol$^{-1}$, respectively, whereas the analogous quantities for the three potentials in MPM-1-Br are 25.57, 25.02 and 25.40 kJ mol$^{-1}$, respectively. Simulations using the CO$_2$-PHAST and CO$_2$-PHAST* potentials suggest that the MPM--CO$_2$ interaction is greater for MPM-1-Br than MPM-1-Cl. This is in contrast to what was observed for experiment as the empirical $Q^0_{st}$ value for CO$_2$ is higher for MPM-1-Cl than MPM-1-Br. A potential reason for why our simulations generated a higher initial loading CO$_2$ $Q_{st}$ value for MPM-1-Br is explained in next section where the binding sites are discussed.


    4.3. Binding sites

    Pham et al. [41] reported a binding site for CO$_2$ in MPM-1-Cl in which the sorbate molecule aligns between two chlorine atoms, one that is bound to the copper paddlewheel complex and another which forms hydrogen-bonding interactions with four hydrogen atoms of two different adenine linkers (see Supporting Information, Figure S2). Interestingly, this site was minimally observed for simulations in MPM-1-Br. Indeed, the main binding site for CO$_2$ in MPM-1-Br was actually between two Br atoms that are coordinated to the Cu$^{2+}$ ions of two different [Cu$_2$(adenine)$_4$Br$_2$]$^{2+}$ complexes (Figure 8b). Notably, the positively charged carbon atom of the CO$_2$ molecule can interact with two electronegative Br$^{-}$ ions simultaneously. We also observed this site in MPM-1-Cl for the simulations reported herein (Figure 8a). The larger van der Waals radius of the Br$^-$ ions causes this region to be more favorable for the CO$_2$ molecules in MPM-1-Br than in MPM-1-Cl. This is because the larger size of the bromide ions will result in greater close-fitting interactions with the sorbate molecules. The CO$_2$ molecule will exhibit a lower potential energy in the attractive well-depth potential region of both Br atoms as a result of the larger ionic radius. This could explain why the CO$_2$-PHAST and CO$_2$-PHAST* models produced CO$_2$ $Q^0_{st}$ values that are greater for MPM-1-Br.

    Figure 8. Molecular illustration of the primary binding site for CO$_2$ observed in (a) MPM-1-Cl and (b) MPM-1-Br through simulation, which is nestled between two X (X = Cl, Br) atoms bound to two different [Cu$_2$(adenine)$_4$]$^{4+}$ complexes. Atom colors are C = cyan; N = blue; H = white; Cu = brown; Cl = yellow; Br = red.

    The electrostatic and electrodynamic (polarizable) effects serve to attract the positively charged carbon center of the CO$_2$ molecule. The Lennard-Jones mixed $\sigma$ parameter for Br--C used herein is ~3.6 Å, and the measured Br--C$_{\mathrm{CO}_2}$ distance at the primary binding site is ~3.2 Å. The expected repulsion of the CO$_2$ carbon atom is therefore counteracted by the attraction of the oxygen atoms on the molecule, which (being further away from the Br atoms) reside in the well-depth region of the dispersive potential. In addition, the calculated partial charge for this particular Br atom is $-$0.51 $e^-$, which is slightly lower in magnitude than that for the analogous Cl atom in MPM-1-Cl ($-$0.57 $e^-$) (see Supporting Information, Table S1). While the lower partial negative charge results in a weaker attraction between the halogen and the CO$_2$ carbon atom, it also reduces the repulsion experienced by each electronegative oxygen atom on the sorbate. The net effect of all these interactions is a ``balancing act'' between the Br$^-$ ions in MPM-1-Br, which is a unique feature compared to the Cl analog.

    MPM-1-Cl is able to sorb more CO$_2$ overall than MPM-1-Br as demonstrated through both experimental measurements [37] and simulation (Figures 4 and 5). This could be due to the higher magnitude of the partial negative charge of the Cl$^-$ ion as mentioned above. Electrostatic interactions are especially important for CO$_2$ sorption in both MPMs. It appears that the Coulombic attraction between the electropositive carbon atom of the CO$_2$ molecule and the halide ion increases with increasing partial negative charge of the anion. Further, the Cl$^-$ ions tend to attract more electron density from the guest molecules via dispersive forces than the Br$^-$ ions due to having a higher electronegativity.

    The primary binding site for C$_2$H$_2$ in both MPM-1-Cl and MPM-1-Br is located near the copper paddlewheel complex, where the sorbate molecule orients parallel to the X--Cu--Cu--X axis (Figure 9). At this site, the center-of-mass (COM) of the C$_2$H$_2$ molecule is positioned approximately 5.5 and 5.2 Å away from the geometric center of the Cu$^{2+}$ ions in MPM-1-Cl and MPM-1-Br, respectively. This particular alignment of the C$_2$H$_2$ molecule about the copper paddlewheels allows the positively charged H atom on both sides of the sorbate to interact with the electronegative halide ions that are coordinated to the Cu$^{2+}$ ions. The negatively charged COM of the C$_2$H$_2$ molecule can also interact somewhat with the Cu$^{2+}$ ions of the copper paddlewheels.

    Figure 9. Molecular illustration of the primary binding site for C$_2$H$_2$ observed in (a) MPM-1-Cl and (b) MPM-1-Br through simulation, which is aligned with the X--Cu--Cu--X (X = Cl, Br) axis of the dinuclear copper paddlewheel complex. Atom colors are C = cyan; N = blue; H = white; Cu = brown; Cl = yellow; Br = red.

    Overall, there are more concurrent interactions between the C$_2$H$_2$ molecule and the framework about the primary binding site in both MPMs relative to that for CO$_2$. Indeed, at the primary binding site, the C$_2$H$_2$ molecule exhibits two H$\cdot\cdot\cdot$X$^-$ and two COM(C$_2$H$_2$)$\cdot\cdot\cdot$Cu$^{2+}$ interactions. On the other hand, there are only two C(CO$_2$)$\cdot\cdot\cdot$X$^-$ interactions for CO$_2$ at the favorable binding site for this sorbate in both MPMs. The greater simultaneous interactions for C$_2$H$_2$ in MPM-1-X could explain why both MPMs display greater affinity and selectivity for C$_2$H$_2$ than CO$_2$. We note that the longer COM(C$_2$H$_2$)$\cdot\cdot\cdot$Cu$^{2+}$ distance for MPM-1-Cl results in slightly weaker affinity at this site relative to MPM-1-Br, thus potentially explaining why our simulations produced lower initial C$_2$H$_2$ $Q_{st}$ values for the Cl analogue.

    The results for the classical binding energy calculations from simulated annealing are presented in Table 2. It is clear from these simulations that both materials favor C$_2$H$_2$ over CO$_2$ (measured as a single molecule competition), which is consistent with experimental data. The energy minimum binding sites that were discovered through these simulations are the same as those shown in Figures 8 and 9. As observed in the results for the theoretical $Q_{st}$ values (Table 1), the calculated binding energies for MPM-1-Br are higher than those for MPM-1-Cl. This is likely due to the higher polarizability of Br used in these simulations.

    Table 2. Classical binding energies (in kJ mol$^{-1}$) for gases in MPM-1-Cl and MPM-1-Br as calculated by single-particle simulated annealing using the polarizable models for CO$_2$ and C$_2$H$_2$. The starting temperature for all simulations was 500 K, to ensure free and full exploration of the phase space of the system.
    MPM-1-ClBinding energy (kJ mol$^{-1}$)Steps ($\times\ 10^6$)Final Temp. (K)
    CO$_2$$-$26.32.37113
    C$_2$H$_2$$-$29.92.46125
    MPM-1-Br
    CO$_2$$-$28.23.24129
    C$_2$H$_2$$-$34.74.53150
     | Show Table
    DownLoad: CSV

    5. Conclusions

    This study aimed to elucidate the CO$_2$ and C$_2$H$_2$ sorption properties and mechanisms in two isostructural MPMs, MPM-1-Cl and MPM-1-Br. Our simulations revealed that MPM-1-Cl exhibits a better overall sorption capacity for both gases than MPM-1-Br, which is consistent with the experimental measurements reported by Xie et al. [37]. This could be due to a combination of larger accessible space in the Cl analogue at higher pressures (a consequence of the smaller ionic radius for Cl$^-$) and/or the fact that Cl is more electronegative than Br, which affords greater electrostatic interactions with the sorbate. At low loadings, however, the simulations suggest that MPM-1-Br is slightly more favorable toward both sorbates than MPM-1-Cl. We attribute this to the higher polarizablility and van der Waals parameters for the Br$^-$ ions. Notably, repulsion/dispersion interactions could dominate the sorption mechanism at such loadings.

    It was discovered through our simulations that the primary binding site for C$_2$H$_2$ and CO$_2$ in both MPMs are notably different between the two sorbates. Particularly, the C$_2$H$_2$ molecule prefers to align along the X--Cu--Cu--X axis of the copper paddlewheel to allow for favorable H$\cdot\cdot\cdot$X$^-$ interactions on each side of the molecule, while the CO$_2$ molecule seems to favor the region between the X$^{-}$ ions of two adjacent copper paddlewheel units. Both MPMs appear to be more selective toward C$_2$H$_2$ than CO$_2$ because there are a greater number of concurrent interactions about the primary binding site for C$_2$H$_2$.

    Herein, we demonstrated how substitution of the halide ion in two isostructural MPMs with the empirical formula [Cu$_2$(adenine)$_4$X$_2$]X$_2$ affects the gas sorption properties in the material through molecular simulations. We plan to investigate the CO$_2$ and C$_2$H$_2$ sorption performance in MPM-1-TIFSIX [42] through theoretical studies in future work.


    Acknowledgements

    The authors acknowledge the National Science Foundation (Award No. DMR-1607989), including support from the Major Research Instrumentation Program (Award No. CHE-1531590). Computational resources were made available by a XSEDE Grant (No. TG-DMR090028) and by Research Computing at the University of South Florida. B.S. also acknowledges support from an American Chemical Society Petroleum Research Fund grant (ACS PRF 56673-ND6).


    Conflict of interest

    The authors declare no conflict of interest related to the content of this publication.


    [1] Stedman RL (1968) Chemical composition of tobacco and tobacco smoke. Chem Rev 68: 153-207 doi: 10.1021/cr60252a002
    [2] Smith CJ, Perfetti TA, Garg R, et al. (2003) IARC carcinogens reported in cigarette mainstream smoke and their calculated log P values. Food Chem Toxicol 41: 807-817
    [3] Dallüge J, van Stee LLP, Xu X, Williams J, et al. (2002) Unravelling the composition of very complex samples by comprehensive gas chromatography coupled to time-of-flight mass spectrometry: Cigarette smoke. J Chromatogr A 974: 169-184 doi: 10.1016/S0021-9673(02)01384-5
    [4] Takanami Y, Chida M, Hasebe H, et al. (2003) Analysis of Cigarette Smoke by an Online Thermal Desorption System and Multidimensional GC-MS. J Chromatogr Sci 41: 317-322 doi: 10.1093/chromsci/41.6.317
    [5] Borgerding M, Klus H (2005) Analysis of complex mixtures -- Cigarette smoke. Exp Toxicol Pathol 57 Suppl 1: 43-73
    [6] Charles SM, Batterman SA, Jia C (2007) Composition and emissions of VOCs in main- and side-stream smoke of research cigarettes. Atmos Environ 41: 5371-5384 doi: 10.1016/j.atmosenv.2007.02.020
    [7] Adam T, Mitschke S, Baker RR (2009) Investigation of tobacco pyrolysis gases and puff-by-puff resolved cigarette smoke by single photon ionisation (SPI)-time-of-flight mass spectrometry (TOFMS). Beitr Zur Tab Tob Res 23: 203-226
    [8] Wang J, Weng J-J, Jia L-Y, et al. (2012) Study on Gas Phase Components in Mainstream Cigarette Smoke by Synchrotron Radiation Photoionization Mass Spectrometry. Chin J Anal Chem 40: 1048-1052 doi: 10.1016/S1872-2040(11)60559-8
    [9] Counts ME, Hsu FS, Laffoon SW, et al. (2004) Mainstream smoke constituent yields and predicting relationships from a worldwide market sample of cigarette brands: ISO smoking conditions. Regul Toxicol Pharmacol 39: 111-134 doi: 10.1016/j.yrtph.2003.12.005
    [10] Counts ME, Morton MJ, Laffoon SW, et al. (2005) Smoke composition and predicting relationships for international commercial cigarettes smoked with three machine-smoking conditions. Regul Toxicol Pharmacol 41: 185-227 doi: 10.1016/j.yrtph.2004.12.002
    [11] Castro D, Slezakova K, Delerue-Matos C, et al. (2011) Polycyclic aromatic hydrocarbons in gas and particulate phases of indoor environments influenced by tobacco smoke: Levels, phase distributions, and health risks. Atmos Environ 45: 1799-1808 doi: 10.1016/j.atmosenv.2011.01.018
    [12] Li M, Dong J-G, Huang Z-X, et al. (2012) Analysis of Cigarette Smoke Aerosol by Single Particle Aerosol Mass Spectrometer. Chin J Anal Chem 40: 936-939 doi: 10.1016/S1872-2040(11)60555-0
    [13] Wright C (2015) Standardized methods for the regulation of cigarette-smoke constituents. Trends Anal Chem 66: 118-127 doi: 10.1016/j.trac.2014.11.011
    [14] Ding YS, Zhang L, Jain RB, et al. (2008) Levels of Tobacco-Specific Nitrosamines and Polycyclic Aromatic Hydrocarbons in Mainstream Smoke from Different Tobacco Varieties. Cancer Epidemiol Biomarkers Prev 17: 3366-3371 doi: 10.1158/1055-9965.EPI-08-0320
    [15] Ding YS, Trommel JS, Yan XJ, et al. (2005) Determination of 14 Polycyclic Aromatic Hydrocarbons in Mainstream Smoke from Domestic Cigarettes. Environ Sci Technol 39: 471-478 doi: 10.1021/es048690k
    [16] Lu X, Cai J, Kong H, et al. (2003) Analysis of Cigarette Smoke Condensates by Comprehensive Two-Dimensional Gas Chromatography/Time-of-Flight Mass Spectrometry I Acidic Fraction. Anal Chem 75: 4441-4451 doi: 10.1021/ac0264224
    [17] Lu X, Zhao M, Kong H, et al. (2004) Characterization of cigarette smoke condensates by comprehensive two-dimensional gas chromatography/time-of-flight mass spectrometry (GC×GC/TOFMS) Part 2: Basic fraction. J Sep Sci 27: 101-109 doi: 10.1002/jssc.200301659
    [18] Brokl M, Bishop L, Wright CG, et al. (2013) Analysis of mainstream tobacco smoke particulate phase using comprehensive two-dimensional gas chromatography time-of-flight mass spectrometry. J Sep Sci 36: 1037-1044 doi: 10.1002/jssc.201200812
    [19] Hughey CA, Rodgers RP, Marshall AG (2002) Resolution of 11000 Compositionally Distinct Components in a Single Electrospray Ionization Fourier Transform Ion Cyclotron Resonance Mass Spectrum of Crude Oil. Anal Chem 74: 4145-4149 doi: 10.1021/ac020146b
    [20] Rodgers RP, Schaub TM, Marshall AG (2005) Petroleomics: MS Returns to Its Roots. Anal Chem 77: 20-A
    [21] Hertkorn N, Ruecker C, Meringer M, et al. (2007) High-precision frequency measurements: indispensable tools at the core of the molecular-level analysis of complex systems. Anal Bioanal Chem 389: 1311-1327 doi: 10.1007/s00216-007-1577-4
    [22] Kujawinski EB (2002) Electrospray Ionization Fourier Transform Ion Cyclotron Resonance Mass Spectrometry (ESI FT-ICR MS): Characterization of Complex Environmental Mixtures. Environ Forensics 3: 207-216 doi: 10.1080/713848382
    [23] Schmitt-Kopplin P, Gelencsér A, Dabek-Zlotorzynska E, et al. (2010) Analysis of the Unresolved Organic Fraction in Atmospheric Aerosols with Ultrahigh-Resolution Mass Spectrometry and Nuclear Magnetic Resonance Spectroscopy: Organosulfates As Photochemical Smog Constituents. Anal Chem 82: 8017-8026 doi: 10.1021/ac101444r
    [24] Gonsior M, Zwartjes M, Cooper WJ, et al. (2011) Molecular characterization of effluent organic matter identified by ultrahigh resolution mass spectrometry. Water Res 45: 2943-2953 doi: 10.1016/j.watres.2011.03.016
    [25] Cottrell BA, Gonsior M, Isabelle LM, et al. (2013) A regional study of the seasonal variation in the molecular composition of rainwater. Atmos Environ 77: 588-597 doi: 10.1016/j.atmosenv.2013.05.027
    [26] Cortés-Francisco N, Harir M, Lucio M, et al. (2014) High-field FT-ICR mass spectrometry and NMR spectroscopy to characterize DOM removal through a nanofiltration pilot plant. Water Res 67: 154-165 doi: 10.1016/j.watres.2014.08.046
    [27] Gonsior M, Schmitt-Kopplin P, Stavklint H, et al. (2014) Changes in Dissolved Organic Matter during the Treatment Processes of a Drinking Water Plant in Sweden and Formation of Previously Unknown Disinfection Byproducts. Environ Sci Technol 48: 12714-12722 doi: 10.1021/es504349p
    [28] Schramm S, Carré V, Scheffler J-L, et al. (2011) Analysis of Mainstream and Sidestream Cigarette Smoke Particulate Matter by Laser Desorption Mass Spectrometry. Anal Chem 83: 133-142 doi: 10.1021/ac1019842
    [29] Schramm S, Carré V, Scheffler J-L, et al. (2014) Active and passive smoking - New insights on the molecular composition of different cigarette smoke aerosols by LDI-FTICRMS. Atmos Environ 92: 411-420 doi: 10.1016/j.atmosenv.2014.04.052
    [30] Schäfer M, Drayß M, Springer A, et al. (2007) Radical Cations in Electrospray Mass Spectrometry: Formation of Open-Shell Species, Examination of the Fragmentation Behaviour in ESI-MSn and Reaction Mechanism Studies by Detection of Transient Radical Cations. Eur J Org Chem 2007: 5162-5174 doi: 10.1002/ejoc.200700199
    [31] Cole DP, Smith EA, Dalluge D, et al. (2013) Molecular characterization of nitrogen-containing species in switchgrass bio-oils at various harvest times. Fuel 111: 718-726 doi: 10.1016/j.fuel.2013.04.064
    [32] Sudasinghe N, Dungan B, Lammers P, et al. (2014) High resolution FT-ICR mass spectral analysis of bio-oil and residual water soluble organics produced by hydrothermal liquefaction of the marine microalga Nannochloropsis salina. Fuel 119: 47-56 doi: 10.1016/j.fuel.2013.11.019
    [33] Baker RR (1987) A review of pyrolysis studies to unravel reaction steps in burning tobacco. J Anal Appl Pyrolysis 11: 555-573 doi: 10.1016/0165-2370(87)85054-4
    [34] Wu Z, Rodgers RP, Marshall AG (2004) Two- and Three-Dimensional van Krevelen Diagrams: A Graphical Analysis Complementary to the Kendrick Mass Plot for Sorting Elemental Compositions of Complex Organic Mixtures Based on Ultrahigh-Resolution Broadband Fourier Transform Ion Cyclotron Resonance Mass Measurements. Anal Chem 76: 2511-2516 doi: 10.1021/ac0355449
    [35] Cho Y, Ahmed A, Islam A, et al. (2015) Developments in FT-ICR MS instrumentation, ionization techniques, and data interpretation methods for petroleomics. Mass Spectrom Rev 34: 248-263 doi: 10.1002/mas.21438
    [36] Baker RR, Bishop LJ (2004) The pyrolysis of tobacco ingredients. J Anal Appl Pyrolysis 71: 223-311 doi: 10.1016/S0165-2370(03)00090-1
    [37] Mullen CA, Boateng AA (2008) Chemical Composition of Bio-oils Produced by Fast Pyrolysis of Two Energy Crops. Energy Fuels 22: 2104-2109 doi: 10.1021/ef700776w
    [38] Olcese R, Carré V, Aubriet F, et al. (2013) Selectivity of Bio-oils Catalytic Hydrotreatment Assessed by Petroleomic and GC*GC/MS-FID Analysis. Energy Fuels 27: 2135-2145 doi: 10.1021/ef302145g
    [39] Li S, Olegario RM, Banyasz JL, et al. (2003) Gas chromatography-mass spectrometry analysis of polycyclic aromatic hydrocarbons in single puff of cigarette smoke. J Anal Appl Pyrolysis 66: 155-163 doi: 10.1016/S0165-2370(02)00111-0
    [40] Carré V, Aubriet F, Muller J-F (2005) Analysis of cigarette smoke by laser desorption mass spectrometry. Anal Chim Acta 540: 257-268 doi: 10.1016/j.aca.2005.03.034
    [41] Aubriet F, Carré V (2010) Potential of laser mass spectrometry for the analysis of environmental dust particles—A review. Anal Chim Acta 659: 34-54 doi: 10.1016/j.aca.2009.11.047
    [42] Cho Y, Witt M, Kim YH, et al. (2012) Characterization of Crude Oils at the Molecular Level by Use of Laser Desorption Ionization Fourier-Transform Ion Cyclotron Resonance Mass Spectrometry. Anal Chem 84: 8587-8594 doi: 10.1021/ac301615m
    [43] Cho Y, Jin JM, Witt M, et al. (2013) Comparing Laser Desorption Ionization and Atmospheric Pressure Photoionization Coupled to Fourier Transform Ion Cyclotron Resonance Mass Spectrometry To Characterize Shale Oils at the Molecular Level. Energy Fuels 27: 1830-1837 doi: 10.1021/ef3015662
    [44] Le Brech Y, Delmotte L, Raya J, et al. (2015) High Resolution Solid State 2D NMR Analysis of Biomass and Biochar. Anal Chem 87: 843-847 doi: 10.1021/ac504237c
  • This article has been cited by:

    1. Douglas M. Franz, Jonathan L. Belof, Keith McLaughlin, Christian R. Cioce, Brant Tudor, Adam Hogan, Luciano Laratelli, Meagan Mulcair, Matthew Mostrom, Alejandro Navas, Abraham C. Stern, Katherine A. Forrest, Tony Pham, Brian Space, MPMC and MCMD: Free High‐Performance Simulation Software for Atomistic Systems, 2019, 2, 2513-0390, 1900113, 10.1002/adts.201900113
    2. Yu-Yun Lin, Fu-Yu Liu, I-Chia Chen, Hwei-Yan Tsai, Jhen-Wei Huang, Jia-Hao Lin, Chiing-Chang Chen, Photocatalytic reduction of carbon dioxide by BiTeX (X = Cl, Br, I) under visible-light irradiation, 2024, 365, 03014797, 121536, 10.1016/j.jenvman.2024.121536
    3. Yu-Yun Lin, Hong-Han Huang, Shiuh-Tsuen Huang, Fu-Yu Liu, Jia-Hao Lin, Chiing-Chang Chen, Synthesis, characterization, photocatalytic activity of selenium vacancy in BiSeX and BiSeX/GO (X = Cl、Br、I) photocatalysts, 2025, 10106030, 116330, 10.1016/j.jphotochem.2025.116330
  • Reader Comments
  • © 2015 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)
通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索

Metrics

Article views(7270) PDF downloads(1195) Cited by(6)

Article outline

Figures and Tables

Figures(9)

/

DownLoad:  Full-Size Img  PowerPoint
Return
Return

Catalog