Role of Amine Functionality for CO2 Chemisorption ... - ACS Publications

Article pubs.acs.org/JPCB

Role of Amine Functionality for CO2 Chemisorption on Silica Maximilian W. Hahn, Jelena Jelic, Edith Berger, Karsten Reuter, Andreas Jentys,* and Johannes A. Lercher* Department of Chemistry, Catalysis Research Center, Technische Universität München, Lichtenbergstraße 4, 85747 Garching, Germany S Supporting Information *

ABSTRACT: The mechanism of CO2 adsorption on primary, secondary, and bibasic aminosilanes synthetically functionalized in porous SiO2 was qualitatively and quantitatively investigated by a combination of IR spectroscopy, thermogravimetry, and quantum mechanical modeling. The mode of CO2 adsorption depends particularly on the nature of the amine group and the spacing between the aminosilanes. Primary amines bonded CO2 preferentially through the formation of intermolecular ammonium carbamates, whereas CO2 was predominantly stabilized as carbamic acid, when interacting with secondary amines. Ammonium carbamate formation requires the transfer of the carbamic acid proton to a second primary amine group to form the ammonium ion and hence two (primary) amine groups are required to bind one CO2 molecule. The higher base strength of secondary amines enables the stabilization of carbamic acid, which is thereby hindered to interact further with nearby amine functions, because their association with Si−OH groups (either protonation or hydrogen bonding) does not allow further stabilization of carbamic acid as carbamate. Steric hindrance of the formation of intermolecular ammonium carbamates leads to higher uptake capacities for secondary amines functionalized in porous SiO2 at higher amine densities. In aminosilanes possessing a primary and a secondary amine group, the secondary amine group tends to be protonated by Si−OH groups and therefore does not substantially interact with CO2.

1. INTRODUCTION Current strategies to decrease CO2 emissions focus mainly on the expansion of renewable and sustainable energies such as wind, solar energy, or biomass-derived fuels.1−3 Coal, gas, and oil remain however the dominating energy resources for the next decades. Therefore, capture of CO2 will play a crucial role in the reduction of CO2 emissions worldwide.4−6 The present benchmark process for the capture of atmospheric CO2 is the aqueous amine wash, which causes, for example, the effectiveness of coal-fired power plants to decrease by 10−12%.7,8 The main fraction of this energy loss results from the thermal regeneration of the aqueous amine solution.7,8 Amine-functionalized solid sorbents with a lower heat capacity compared to aqueous amine solutions present an opportunity to reduce the energy required for the required regeneration step.9−13 The most common way to obtain amine-functionalized solid sorbents is the impregnation of highly ordered porous supports with low molecular amines or amino polymers.14−16 A wellstructured porosity and large pore volumes are among the most important characteristic of the supports making SBA-15 or MCM-41 often the material of choice.17,18 High molecular weight amines that contain several primary, secondary, or tertiary amine groups, that is, tetraethylenpentamine (TEPA) or polyethylenimine (PEI), are frequently employed in the impregnation.19,20 Although impregnation of amines allows for high CO2 uptake capacities, the resulting adsorbents face several drawbacks associated with the weak, noncoordinated interaction of the amines and the support as well as to high © 2015 American Chemical Society

amine loadings excluding efficient CO2 interaction with the active sites.21 Moreover, thermal degradation and desorption of weakly adsorbed low-molecular amines cannot be neglected for the evaluation of the long-term performance and cycle stability.21 Particularly, sufficient spacing between the amine groups and access to surface hydroxyl groups have been found to be important to obtain high amine efficiencies, that is, the number of CO2 molecules captured per amine group.22,23 However, the impregnation of highly ordered porous supports with low molecular amines or amino polymers does not facilitate an arrangement of the functional groups at a molecular level on the surface but rather leads to a bulky amine conglomerate in the pores of the sorbents.21,24,25 The direct condensation of aminosilanes with surface OH groups (grafting) leads, which is in contrast to immobilized aminosilanes on the surface of the support that also tend to be thermally and chemically stable.21,26 For chemisorption of CO2, Brunelli et al. reported recently that the dominant mechanism for the CO2 adsorption of grafted amines was the cooperative adsorption and formation of ammonium carbamates.23 However, the stabilization of chemisorbed CO2 by a surface Si−OH group significantly enhanced the concentration of adsorbed CO2, when the Special Issue: Bruce C. Garrett Festschrift Received: October 13, 2015 Revised: November 27, 2015 Published: December 23, 2015 1988

DOI: 10.1021/acs.jpcb.5b10012 J. Phys. Chem. B 2016, 120, 1988−1995

Article

The Journal of Physical Chemistry B distance to the surface (linker length) was larger than two C atoms. This allowed for a sufficient flexibility of the hydrocarbon carrying the amine functionality.23 This points to a rather complex surface chemistry, which in turn suggests that the nature of the grafted amines as well as their mutual distance influence the mechanism and strength of CO2 chemisorption. Insight into this surface chemistry is of critical importance to tailor sorbents and to develop novel high capacity materials for CO2 capture. In the present study, we explore the influence of the amine functionality and concentration of in situ immobilized primary (APTMS(1)) and secondary amines (MAPS(2)) with a C3 linker (Scheme 1) on the water-free CO2 adsorption. AmineScheme 1. Aminosilanes Employed in Synthesis

Figure 1. (I) Baseline corrected and (II) difference IR spectra of APTMS(1) (3.31 mmol g−1), at CO2 partial pressures up to 10 mbar and in a range from 1850−1250 cm−1.

The most pronounced peaks in the IR spectra at 0 mbar CO2 are at 1593 and 1430 cm−1, characteristic for an asymmetric NH3+ deformation and a CN stretch of the amine, respectively. The fact that protonated amines are already visible in the IR spectra prior to CO2 adsorption points to a protonation of the grafted amine by adjacent surface silanol groups. During adsorption of CO2 new bands arise, most dominantly at 1561, 1484, and 1431 cm−1 (Figure 1II) that are assigned to the asymmetric COO − stretching vibration in ammonium carbamates, symmetric NH3+ deformation, and the symmetric COO − stretching vibration in ammonium carbamates, respectively. The band at 1317 cm−1 is due to NCOO skeletal vibration. Both bands at 1680 and 1383 cm−1, characteristic for the CO stretching vibration and OH deformation vibration in carbamic acid, are of weak intensity. The IR spectra of CO2 adsorbed on MAPS(2) are shown in (I) and as difference spectra (II) in Figure 2. The most distinct difference to the IR spectrum of APTMS(1) is the strongly defined band at 1472 cm−1, caused by the methyl linker adjacent to the secondary amine group. Upon exposure to CO2, vibrations characteristic for carbamate formation, that is, COO− stretching vibration in ammonium

functionalized, millimeter-scale spherical adsorbents were synthesized in a one-step procedure utilizing the basicity of the amine for the co-condensation of the silane precursors.27,28 The potential of intramolecular CO2 chemisorption was probed by grafting a bibasic amine (AAMS(1,2), Scheme 1). The interaction of amine groups with surface silanol groups during CO2 adsorption at low amine concentrations and the stabilization between two neighboring amines (intermolecular stabilization) at higher amine concentrations were qualitatively and quantitatively investigated in a combined experimental and theoretical study.

2. RESULTS AND DISCUSSION 2.1. Impact of the Amine Functionality on the CO2 Capture Mechanism of Amines on SiO2 Surfaces. The adsorption of CO2 on primary, secondary, and bibasic amines (Scheme 1) immobilized on SiO2 was studied in the absence of molecular water by in situ IR spectroscopy at medium amine concentrations (3.30 mmol N g−1). Such concentrations allow intramolecular interactions between vicinal amine groups as well as sufficient access to surface Si−OH groups (Supporting Information Figure S9). The impact of the amine functionality on the morphology and structure of the sorbents is described in detail in the Supporting Information (see SI section B and C). For further notation, the sorbents are assigned according to the aminosilanes employed, followed by the functionality in brackets, that is, APTMS(1), MAPS(2), AAMS(1,2). The IR spectra of CO2 adsorbed on APTMS(1) are depicted as baseline corrected spectra (I) and difference spectra (II) in Figure 1. In the following description of the spectra of APTMS(1), MAPS(2), and AAMS(1,2) in Figures 1−3 the most relevant bands are explained and differences between the vibration frequencies of the various amines are discussed more in detail. A detailed listing of all relevant bands are presented in Tables S4 and S5 in the Supporting Information.

Figure 2. (I) Baseline corrected and (II) difference IR spectra of MAPS(2) (3.30 mmol g−1) at CO2 partial pressures up to 10 mbar and in a range from 1850−1250 cm−1. 1989

DOI: 10.1021/acs.jpcb.5b10012 J. Phys. Chem. B 2016, 120, 1988−1995

Article

The Journal of Physical Chemistry B carbamates and NH2+ deformation (see Supporting Information Table S5 for vibration frequencies), are less evolved compared to APTMS(1), while the bands characteristic for the formation of carbamic acid (1687 and 1384 cm−1) are more pronounced. The IR spectra of CO2 adsorbed on AAMS(1,2) are depicted in (I) and as difference spectra (II) in Figure 3. The difference

environment, but as the drying conditions of the adsorbents were limited the presence of low concentrations of residual water cannot be fully excluded. DFT calculations indicated that a minimum of two vicinal amine groups is required to stabilize the protonation of one amine group by surface Si−OH groups via hydrogen bonding of the Si−OH with the amine nitrogen, as exemplarily illustrated for MAPS(2) in Figure 4B. It should be noted in passing that the protonated amine and deprotonated surface Si−OH group do not undergo a complete charge separation under water-free conditions and the amine/ ammonium species remain in close proximity to the surface (Figure 4B). The proton affinity (determined by DFT) was 30 and 43 kJ mol−1 higher for secondary amines in MAPS(2) and AAMS(1,2) than for APTMS(1) (Supporting Information Table S2), which is in agreement with the respective gas basicities.30 Thus, protonation was most pronounced for secondary amine groups in AAMS(1,2) linked to a primary amine via a C2 linker (Figure 3). Figure 5 illustrates the arrangement and interaction of vicinal aminosilanes during adsorption of CO2, that is, for primary

Figure 3. (I) Baseline corrected and (II) difference IR spectra of AAMS(1,2) (3.29 mmol g−1), at CO2 partial pressures up to 10 mbar and in a range from 1850−1250 cm−1.

spectra of AAMS(1,2) (Figure 3II) exhibits vibrations characteristic for carbamates formation like APTMS(1), while the protonation of the amine groups (1669 cm−1) is significantly more pronounced for AAMS(1,2). Prior to adsorption of CO2, partial protonation of the amine groups by Si−OH groups was already observed for all sorbents even under H2O free conditions as inferred from the appearance of the band at 1670−1620 cm−1 and 1490−1480 cm−1 characteristic for the asymmetric and symmetric NH3+/ NH2+ deformation vibrations (Figures 1−3).29 Hence, the linker length of 3 C atoms was sufficient to allow primary and secondary amine groups to interact with Si−OH groups.23 DFT calculations were performed to gain a closer insight in the adsorption mechanism of the different amine-functionalized silanes. The arrangement of the aminosilanes on the silica surface according to DFT calculations is illustrated in Figure 4. Please note that the adsorption of CO2 proceeds in a water-free Figure 5. Adsorption of CO2 on (A) APTMS(1), (B) MAPS(2), and (C) AAMS(1,2) via intermolecular stabilization of carbamates (A,C) and formation of carbamic acid (B) on a silica cluster with vicinal aminosilane interaction. Molecular structure is displayed in topview.

amines (Figure 5a), secondary amines (Figure 5b), and bibasic (primary and secondary amines) in Figure 5c. CO2 is transiently bonded as carbamic acid zwitterion on primary amines and then stabilized as an ammonium carbamate between two amines (intermolecular) or as a carbamic acid in absence of adjacent amine groups (Figure 5a, Supporting Information Figure S8). The sorption on APTMS(1) and AAMS(1,2) follows the expected pathway in the formation of ammonium carbamate at sufficient intermolecular amine interactions (Figure 5b,c, eq 1)31

Figure 4. Influence of intermolecular amine interaction and protonation by surface Si−OH groups for secondary aminosilanes. (A) Single site and (B) vicinal amine interaction of MAPS(2). Molecular structure is displayed in topview.

2R1NH 2 + CO2 ⇌ R1NH3+ + R1NHCOO− 1990

(1)

DOI: 10.1021/acs.jpcb.5b10012 J. Phys. Chem. B 2016, 120, 1988−1995

Article

The Journal of Physical Chemistry B R1NH3+ + R 2NH 2 + CO2 ⇌ R1NH3+ + R 2NHCOOH

R1R 2NH 2+ + R1R 2NH + CO2

(2)

⇌ R1R 2NH 2+ + R1R 2NCOOH

R1, R2 alkyl chains The pronounced protonation of the secondary amine group in AAMS(1,2) by surface Si−OH limits, however, intramolecular stabilization of chemisorbed CO2 species (Figure 3). The predominately protonated secondary amine groups in AAMS(1,2) (Figure 3) were essentially inactive for CO2 chemisorption. Thus, the presence of the primary amine function led to essentially the same adsorption mechanism as for APTMS(1), but with a 50% lower overall amine efficiency, that is, the secondary amine groups did not participate in CO2 chemisorption. For primary amines, carbamic acid is only stabilized as such in absence of sufficient amine concentrations, suppressing the formation of the carbamate. This could result from a large spacing between amine groups or from the fact that the primary amine group forming the carbamic acids is surrounded by amines that are either protonated by the surface or are already interacting with CO2 (eq 2). It should be noted that the proton affinity of Si−OH groups (1200 kJ mol−1)32 restricts marked interactions of CO2 (proton affinity of 540 kJ mol−1)33 and already protonated amines (Figure 5). As the binding of CO2 on APTMS(1) and AAMS(1,2) is significantly stronger in the form of ammonium carbamates than in the form of carbamic acid (Table 1, Supporting

2R1R 2NH + CO2 ⇌ R1R 2NCOO− + R1R 2NH 2+

binding energy [kJ mol−1]

carbamate carbamic acid

−66 −43

MAPS(2) −52 −68

AAMS(1,2) −NH2

−NH

−60a −42

nd −28

(4)

R1, R2 alkyl chains As the formation energy of carbamic acid for MAPS(2) is 16 kJ mol−1 less exothermic than the formation of a carbamate (Table 1), it is still conceivable that carbamic acid can also be converted further to carbamates (comparable to APTMS(1) in Supporting Information Figure S8), when the hydrogen bonding and protonation of secondary amine groups is limited by the access to Si−OH groups, that is, when the local density of amines is high (clustering). In summary, two amine groups are required for the adsorption of one CO2 molecule for both APTMS(1) and MAPS(2) on silanol surfaces at medium amine concentrations, resulting in a theoretical maximum efficiency of 50% for each of the amine groups. The inactivity of the secondary amine group in bibasic AAMS(1,2) (Figure 5c) leads to the situation that four amine groups are required to capture one CO2 molecule. Thus, in contrast to amine-impregnated sorbents with higher amine loadings, these type of functionalized or grafted materials are less applicable for CO2 adsorption compared to monofunctional grafted sorbents. 2.2. Effect of the Amine Concentration on the CO2 Adsorption Mechanism and Uptake Capacity. Adsorption isotherms and heats of adsorption (ΔHads) recorded for APTMS(1), MAPS(2), and AAMS(1,2) at medium amine concentrations (3.30 mmol N g−1) and increasing CO2 partial pressures are shown in Figure 6. In line with the mechanisms of adsorption just discussed, the adsorption isotherms of APTMS(1) and MAPS(2) are similar, while the uptake was about 50% higher compared with AAMS(1,2). The calorimetrically determined initial heats of adsorption on APTMS(1) and MAPS(2) were approximately −75 kJ mol−1 and are in the range of heats of adsorption previously determined for grafted aminosilanes at comparable CO2 partial pressures.31 Please note that higher heats of adsorption are likely to be determined for lower CO2 coverages.31 The decrease of the initial heats of adsorption with increasing concentrations of adsorbed CO2 is hypothesized to be the result of steric hindrances between adsorbed species to form carbamates in all individual interactions. The heat of adsorption of CO2 on AAMS(1,2) was −60 kJ mol−1, which is attributed to the higher amine spacing and the lower concentration of primary amines, decreasing in this way the probability that all CO2 forms carbamates intermolecularly and/or lowering the binding energy of intermolecularly adsorbed CO2 by larger separation of the ion pair (Figure 6). During the course of ten adsorption−desorption cycles, a variation in the heats of adsorption for CO2 was not observed (Supporting Information Figure S6), which allows to exclude degradation and changes in the adsorption mechanism with time on stream. A temperature of 75 °C was observed to be sufficient for the full desorption of CO2 from APTMS(1), MAPS(2), and AAMS(1,2) in flowing N2 (Supporting Information Figure S5). The CO2 equilibrium uptake for APTMS(1), MAPS(2), and AAMS(1,2) at amine concentration between 1.3 and 3.6 mmol g−1 determined at 100 mbar under flow conditions is illustrated in Figure 7. At low amine concentrations, the concentration of

Table 1. Binding Energies of CO2 on Vicinal Aminosilanes That Allow Intermolecular Amine Interactions (Determined by DFT) APTMS(1)

(3)

a

Formation of carbamate was determined intermolecular between two primary amine groups.

Information Figure S9), formation of carbamic acid is only plausible under conditions, which exclude the formation of ammonium carbamates for steric reasons, that is, when a high density of amine groups leads to the generation of isolated amines surrounded by carbamates. It should be noted that the difference in binding energy of CO2 in the form of carbamic acid on AAMS(1,2) is mainly a result of the predominantly protonated secondary amine group. This latter situation has been indeed observed, when amine segregation during synthesis led to local patches with higher amine concentrations, which in turn contained some carbamic acids as spectroscopically observed (Supporting Information Figure S9). The higher base strength of secondary amines, such as MAPS(2), limits the stabilization of carbamic acid by adjacent amines (Supporting Information Table S2), as a sizable fraction of them is hydrogen bonding to Si−OH groups or is being protonated. This is corroborated by the intense bands at 1687 and 1384 cm−1 in the IR spectra (Figure 2, eq 3). Moreover, the stronger basicity of MAPS(2) compared to APTMS(1) results in a less facile proton transfer of the carbamic acid to neighboring aminosilanes and thereby weakening the intramolecular stabilization of the adsorbed species (transformation to carbamate)34 1991

DOI: 10.1021/acs.jpcb.5b10012 J. Phys. Chem. B 2016, 120, 1988−1995

Article

The Journal of Physical Chemistry B

Figure 6. (A) Concentration of CO2 adsorbed over the CO2 partial pressure and (B) heats of adsorption versus adsorbed CO2. Values determined at 50 °C.

Figure 7. (A) Concentration of CO2 adsorbed at 100 mbar equilibrium pressure at 50 °C versus the amine concentration of the adsorbents and (B) correlation between the heats of adsorption and the concentration of adsorbed CO2 on the samples with increasing amine concentrations (determined under flow conditions).

adsorbed CO2 significantly decreased to 0.1 mmol g−1 independent of the employed aminosilanes (Figure 7A) and the heat of adsorption (ΔHads) dropped to values as low as −40 kJ mol−1 (Figure 7B). It should be noted that the heats of adsorption were only between −25 and −30 kJ mol−1 for bare SiO2 (physisorption) and the uptake was in the order of 0.04− 0.1 mmol g−1 (Supporting Information Table S3). It appears that a significant amount of amines was inaccessible to CO2. This amount is hypothesized to be constant for all samples prepared (1.5−2 mmol g−1), leading to an increasing fraction of accessible amines at higher overall concentrations of amines. At large distances between the amine groups (that is, for lower amine concentrations), CO2 adsorption via intermolecular amine stabilization is not possible with monofunctional amines. Quantum mechanical modeling suggests that under these conditions CO2 can only be bonded on APTMS(1) and MAPS(2) by the formation of energetically less preferred carbamic acid (Figure 8, Table 2). The DFT calculations indicate that intramolecular ammonium carbamate formation is energetically favored for AAMS(1,2), when intermolecular amine interaction is not possible (Table 2) and the secondary amine is not protonated. However, the majority of the secondary amine groups is protonated by surface silanol groups (Figure 3), and thus formation of carbamic acid is also the predominant adsorption mechanism for AAMS(1,2) (Figure 8C,D). It should, however, be noted in passing that theoretical calculations suggest that some stabilization by Si−OH groups through hydrogen bonding may be beneficial as depicted by the

value calculated for carbamic acid formed on the secondary amine group of AAMS(1,2) (Table 2.). At higher primary amine concentrations, intermolecular amine interactions are possible and the formation of carbamates occurs via deprotonation of the zwitterion/carbamic acid by an adjacent amine group. However, the random chemisorption of CO2 will lead to a significant concentration of primary amine groups in “isolated” positions (that is, amine groups in direct neighborhood to carbamic acid or carbamates are already involved in CO2 bonding), making it difficult to access a second amine group under these conditions (Figure 9A). Thus, at high sorption levels, only the energetically less stable formation of carbamic acid would be possible (Table 1), which results in an overall decrease of adsorption capacity for primary amines (Figure 7A). Note the decrease in the heats of adsorption at the highest APTMS(1) content (Supporting Information Figure S7). A similar transition in the adsorption mechanism is conceptually also proposed for AAMS(1,2) at higher amine concentrations. However, it was not observed experimentally due to limitations in the amine loading that could be achieved in the synthesis of the spherical sorbents (Figure 9C). The CO2 uptake capacity of MAPS(2) was not impaired by a higher concentration of amine functions, as observed for APTMS(1). The steric hindrance and the higher base strength allow stabilization of carbamic acid species for MAPS(2) that may also further be stabilized via hydrogen bonding at sufficient 1992

DOI: 10.1021/acs.jpcb.5b10012 J. Phys. Chem. B 2016, 120, 1988−1995

Article

The Journal of Physical Chemistry B

Figure 8. Adsorption of CO2 on (A) APTMS(1), (B) MAPS(2), and (C,D) AAMS(1,2) via carbamic acid (A,B,C) and carbamate (D) on a silica cluster without intermolecular aminosilane interaction. Molecular structure is displayed in topview.

presence of a sufficient concentration of amine groups, intermolecular ammonium carbamate and carbamic acid are the predominant adsorption species for monofunctional primary and secondary aminosilanes on SiO2, respectively. In comparison to primary amines, the higher base strength of secondary amines makes the proton transfer of the carbamic acid to neighboring aminosilanes less facile. Additionally, these adjacent secondary amines are involved in hydrogen bonding or are protonated by Si−OH groups and, thus, hardly available for interactions with formed carbamic acid species. Thus, the maximum achievable amine efficiency of primary and secondary amine groups with sufficient access to surface Si−OH groups is limited to 50%. Bibasic aminosilanes grafted to a hydroxylated silanol surface preferably form carbamates via their primary amine function, because the secondary amine group tends to be protonated by surface Si−OH groups. By this, intramolecular carbamate formation is hindered and the amine efficiency is reduced to a maximum of 25% by leaving the primary amines as the predominant active species. Steric hindrance of intramolecular ammonium carbamate formation constrains the adsorption of CO2 on primary amines and leads to a marked drop once the level of amine spacing on the surface of the support exceeds an upper limit. Secondary amines that energetically prefer the binding of CO2 via carbamic acid are not substantially impaired in their performance.

Table 2. Binding Energies of CO2 on Single Aminosilanes with No Intermolecular Amine Interaction Determined by Computational Quantum Mechanical Modelling binding energy [kJ mol−1] APTMS(1) carbamate carbamic acid a

−20

MAPS(2)

−25

AAMS(1,2) −NH2

−NH

−39a −23

−28

Formation of carbamate was determined intramolecular.

Figure 9. Adsorption of CO2 on (A) APTMS(1), (B) MAPS(2), and (C) AAMS(1,2) at high amine concentrations and low concentration of accessible surface Si−OH groups.

4. METHODS 4.1. Synthesis. SiO2 spheres were synthesized by a base catalyzed hydrolysis and condensation reaction as reported before.27 The surfactant containing precursor solution was prepared by mixing 5.4 g of the surfactant Pluronic RPE 1740 ((PO)11(EO)27(PO)11)35 and 5.9 g of benzyl alcohol. The Si rich precursor solution contained 6.0 g of tetraethylorthosilicate (TEOS), 4.3 g of phenyltrimethoxysilane (PTMS), and the according concentration of APTMS, MAPS, and AAMS. Both solutions were merged and mixed prior to injection to the water-filled reactor column. The reaction conditions as well as a detailed description of the synthesis steps are presented in our previous work.27 The spheres were directly formed by micelle formation and base-catalyzed condensation.27 Afterward they were aged in deionized (DI) water and washed with copious amounts of DI water and ethanol. In order to remove residual solvents, the SiO2 spheres were dried overnight. Thereafter, Soxhlet extraction was performed for 24 h at 90 °C in ethanol

amine spacing with other carbamic acids and Si−OH groups (Figure 9B, Supporting Information Table S2).

3. CONCLUSIONS The mechanism of CO2 adsorption of amine-functionalized silanes in absence of water, depended particularly on the functionality of the amine as well as the degree of interaction between the amine groups. A higher concentration of grafted amines on the surface of the support enhances the number of CO2 molecules adsorbed per amine group (amine efficiency). This is mainly caused by the decreasing impact of intramolecular stabilization by facilitating intermolecular stabilization of the adsorbed species. Independent of the amine functionality a low density of amine groups on the surface limits CO2 adsorption due to the formation of energetically nonfavored carbamic acid and thereby lowers the adsorption capacity and amine efficiency. Therefore, it is crucial to provide a minimum concentration of adjacent amine groups. In the 1993

DOI: 10.1021/acs.jpcb.5b10012 J. Phys. Chem. B 2016, 120, 1988−1995

Article

The Journal of Physical Chemistry B to decrease the amount of residual surfactants. All sorbents have been classified in a particle size range of 0.2−0.4 mm. 4.2. Characterization. The C, H, and N contents of the SiO2 spheres were determined by combustion analysis with a thermal conductivity detector (TCD). A ThermoFinnigan Sorptomatic automated surface area and pore size analyzer was used to determine surface area, micro- and mesopore volume, as well as the pore size distribution by nitrogen physisorption at −196 °C. After grinding the spheres, the samples were prepared as self-supporting wafers and the IR spectra were recorded on a Bruker VERTEX 70 spectrometer with MCT detector. Before each measurement, the wafer was activated at 100 °C for 1 h under vacuum (p = 10−6 mbar). CO2 was adsorbed on the sample at 50 °C and 1.0, 2.0, 4.0, 8.0, and 10 mbar pressure. A total of 128 scans were collected with a resolution of 4 cm−1 for each spectrum. All spectra were baseline corrected and normalized to the intensity of Si−O overtone resonances in the range of 2095−1755 cm−1. CO2 adsorption isotherms and the heats of adsorption, determined by differential scanning calorimetry (DSC), were obtained under static conditions on a Seteram TG DSC 111 thermogravimetric analyzer (TGA) at 50 °C and up to CO2 partial pressures of 100 mbar. Before each measurement, all samples were heated at 100 °C for 1 h under vacuum to ensure the removal of residual water and solvents. A Setaram Sensys Evo TGA was used to detect the heats of adsorption (DSC) and adsorbed amount of CO2 under atmospheric flow conditions. Twenty milligrams of the sample was placed in a platinum crucible and dried under N2 flow with 16 mL min−1 at 100 °C for 1 h. The specimen was cooled to 50 °C and 10 vol % CO2 was added to the N2 stream, while the total flow remained constant. The adsorption time to reach the adsorption equilibrium was 60 min. Additional information on the adsorption−desorption experiments is illustrated in the Supporting Information. An experimental description of adsorption−desorption and multicycle experiments is also presented in the Supporting Information. 4.3. DFT Calculations. The interactions of CO2 with grafted aminosilanes were determined by density functional theory (DFT) calculations applying the all-electron fullpotential code FHI-aims. 36,37 Electronic exchange and correlation were treated on the level of the generalized gradient approximation (GGA) BLYP functional as well as on the level of the hybrid B3LYP functional.38−40 Dispersive interactions were not considered at these levels of theory and taken into account by applying the dispersion-correction scheme by Tkatchenko and Scheffler (TS).41 Geometry optimization was performed at the BLYP+TS level and using tight tier1 basis sets until residual forces fell below 10−4 eV Å−1. Subsequent, single point calculations on these geometries at hybrid B3LYP+TS level yielded the final values for the energies reported. The surface of the SiO2 spheres was modeled by two clusters of different unit sizes (small cluster 8 SiO2 units, large cluster 12 SiO2 units). The small cluster is large enough to host two APTMS(1) or MAPS(2), as well as one AAMS(1,2). The larger cluster was used to represent higher loadings of AAMS(1,2). During the geometry optimization, all of the atoms in the clusters were allowed to relax. The proton affinity (pa), defined as the negative reaction enthalpy, is calculated for primary, secondary, and bifunctional aminosilanes by eq 5 below

pa = −ΔH = −ΔE + RT = −ΔEel − ΔZPE +

5 RT 2

ΔEel is defined as the change in the electronic energy and ΔZPE as the change of the zero point energy of the normal modes.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jpcb.5b10012. Additional information on experimental techniques as well as several figures and tables illustrating the physiochemical properties of the sorbents are listed. Further IR spectra of the sorbents with the according peak assignment and CO2 adsorption−desorption isotherms are presented. (PDF)

■

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. Phone: +49-(89)-289-13540. *E-mail: [email protected]. Phone: +49-(89)-28913540. Author Contributions

All authors have given approval to the final version of the manuscript. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We would like to thank the Federal Ministry of Education and Research (BMBF) and Clariant for financial support. Financial support from the DFG is gratefully acknowledged. We thank Sebastian Foraita as well as Thomas Przybilla and Benjamin Winter from the group of Erdmann Spiecker in Erlangen for providing high resolution SEM images of SiO2 spheres. We thank WACKER and BASF for providing chemicals. Funded by BMBF Project No. 01RC1106A, DFG project LE 1187/10

■

REFERENCES

(1) Ragauskas, A. J.; Williams, C. K.; Davison, B. H.; Britovsek, G.; Cairney, J.; Eckert, C. A.; Frederick, W. J.; Hallett, J. P.; Leak, D. J.; Liotta, C. L.; Mielenz, J. R.; Murphy, R.; Templer, R.; Tschaplinski. The Path Forward for Biofuels and Biomaterials. Science 2006, 311, 484−489. (2) Hoffert, M. I.; Caldeira, K.; Benford, G.; Criswell, D. R.; Green, C.; Herzog, H.; Jain, A. K.; Kheshgi, H. S.; Lackner, K. S.; Lewis, J. S.; Lightfoot, H. D.; Manheimer, W.; Mankins, J. C.; Mauel, M. E.; Perkins, L. J.; Schlesinger, M. E.; Volk, T.; Wigley, T. M. L. Advanced Technology Paths to Global Climate Stability: Energy for a Greenhouse Planet. Science 2002, 298, 981−987. (3) Dincer, I. Renewable Energy and Sustainable Development: A Crucial Review. Renewable Sustainable Energy Rev. 2000, 4, 157−175. (4) Jenner, S.; Lamadrid, A. J. Shale Gas Vs. Coal: Policy Implications from Environmental Impact Comparisons of Shale Gas, Conventional Gas, and Coal on Air, Water, and Land in the United States. Energy Policy 2013, 53, 442−453. (5) Smil, V. The Long Slow Rise of Solar and Wind. Sci. Am. 2013, 310, 52−57. (6) Hughes, J. D. Energy: A Reality Check on the Shale Revolution. Nature 2013, 494, 307−308. (7) Rochelle, G. T. Amine Scrubbing for Co2 Capture. Science 2009, 325, 1652−1654.

1994

DOI: 10.1021/acs.jpcb.5b10012 J. Phys. Chem. B 2016, 120, 1988−1995

Article

The Journal of Physical Chemistry B

(28) Scholz, S.; Lercher, J. A. Hierarchically Structured MillimeterSized (Organo) Silica Spheres with a Macroporous Shell and a Meso/ Microporous Core. Chem. Mater. 2011, 23, 2091−2099. (29) Knöfel, C.; Martin, C.; Hornebecq, V.; Llewellyn, P. L. Study of Carbon Dioxide Adsorption on Mesoporous AminopropylsilaneFunctionalized Silica and Titania Combining Microcalorimetry and in Situ Infrared Spectroscopy. J. Phys. Chem. C 2009, 113, 21726− 21734. (30) Hunter, E. P.; Lias, S. G. Evaluated Gas Phase Basicities and Proton Affinities of Molecules: An Update. J. Phys. Chem. Ref. Data 1998, 27, 413−656. (31) Alkhabbaz, M. A.; Bollini, P.; Foo, G. S.; Sievers, C.; Jones, C. W. Important Roles of Enthalpic and Entropic Contributions to Co2 Capture from Simulated Flue Gas and Ambient Air Using Mesoporous Silica Grafted Amines. J. Am. Chem. Soc. 2014, 136, 13170−13173. (32) Kramer, G. J.; Van Santen, R. Theoretical Determination of Proton Affinity Differences in Zeolites. J. Am. Chem. Soc. 1993, 115, 2887−2897. (33) Bohme, D. K.; Mackay, G.-I.; Schiff, H. Determination of Proton Affinities from the Kinetics of Proton Transfer Reactions. Vii. The Proton Affinities of O2, H2, Kr, O, N2, Xe, Co2, Ch4, N2o, and Co. J. Chem. Phys. 1980, 73, 4976−4986. (34) Johnson, S.; Morrison, D. L. Kinetics and Mechanism of Decarboxylation of N-Arylcarbamates. Evidence for Kinetically Important Zwitterionic Carbamic Acid Species of Short Lifetime. J. Am. Chem. Soc. 1972, 94, 1323−1334. (35) Scholz, S.; Bare, S. R.; Kelly, S. D.; Lercher, J. A. Controlled One-Step Synthesis of Hierarchically Structured Macroscopic Silica Spheres. Microporous Mesoporous Mater. 2011, 146, 18−27. (36) Blum, V.; Gehrke, R.; Hanke, F.; Havu, P.; Havu, V.; Ren, X.; Reuter, K.; Scheffler, M. Ab Initio Molecular Simulations with Numeric Atom-Centered Orbitals. Comput. Phys. Commun. 2009, 180, 2175−2196. (37) Ren, X.; Rinke, P.; Blum, V.; Wieferink, J.; Tkatchenko, A.; Sanfilippo, A.; Reuter, K.; Scheffler, M. Resolution-of-Identity Approach to Hartree−Fock, Hybrid Density Functionals, Rpa, Mp2 and Gw with Numeric Atom-Centered Orbital Basis Functions. New J. Phys. 2012, 14, 053020. (38) Becke, A. D. Density-Functional Exchange-Energy Approximation with Correct Asymptotic Behavior. Phys. Rev. A: At., Mol., Opt. Phys. 1988, 38, 3098. (39) Becke, A. D. Density-Functional Thermochemistry. Iii. The Role of Exact Exchange. J. Chem. Phys. 1993, 98, 5648−5652. (40) Stephens, P.; Devlin, F.; Chabalowski, C.; Frisch, M. J. Ab Initio Calculation of Vibrational Absorption and Circular Dichroism Spectra Using Density Functional Force Fields. J. Phys. Chem. 1994, 98, 11623−11627. (41) Tkatchenko, A.; Scheffler, M. Accurate Molecular Van Der Waals Interactions from Ground-State Electron Density and FreeAtom Reference Data. Phys. Rev. Lett. 2009, 102, 073005.

(8) Oexmann, J.; Kather, A. Post-Combustion Co2 Capture in CoalFired Power Plants: Comparison of Integrated Chemical Absorption Processes with Piperazine Promoted Potassium Carbonate and Mea. Energy Procedia 2009, 1, 799−806. (9) Chang, F. Y.; Chao, K. J.; Cheng, H. H.; Tan, C. S. Adsorption of Co2 onto Amine-Grafted Mesoporous Silicas. Sep. Purif. Technol. 2009, 70, 87−95. (10) Huang, H. Y.; Yang, R. T.; Chinn, D.; Munson, C. L. AmineGrafted Mcm-48 and Silica Xerogel as Superior Sorbents for Acidic Gas Removal from Natural Gas. Ind. Eng. Chem. Res. 2003, 42, 2427− 2433. (11) Hampe, E. M.; Rudkevich, D. M. Exploring Reversible Reactions between Co2 and Amines. Tetrahedron 2003, 59, 9619−9625. (12) Danon, A.; Stair, P. C.; Weitz, E. Ftir Study of Co2 Adsorption on Amine-Grafted Sba-15: Elucidation of Adsorbed Species. J. Phys. Chem. C 2011, 115, 11540−11549. (13) Drage, T. C.; Arenillas, A.; Smith, K. M.; Snape, C. E. Thermal Stability of Polyethylenimine Based Carbon Dioxide Adsorbents and Its Influence on Selection of Regeneration Strategies. Microporous Mesoporous Mater. 2008, 116, 504−512. (14) Yan, X.; Zhang, L.; Zhang, Y.; Yang, G.; Yan, Z. Amine-Modified Sba-15: Effect of Pore Structure on the Performance for Co2 Capture. Ind. Eng. Chem. Res. 2011, 50, 3220−3226. (15) Kuwahara, Y.; Kang, D. Y.; Copeland, J. R.; Brunelli, N. A.; Didas, S. A.; Bollini, P.; Sievers, C.; Kamegawa, T.; Yamashita, H.; Jones, C. W. Dramatic Enhancement of Co2 Uptake by Poly(Ethyleneimine) Using Zirconosilicate Supports. J. Am. Chem. Soc. 2012, 134, 10757−10760. (16) Wang, D.; Wang, X.; Ma, X.; Fillerup, E.; Song, C. ThreeDimensional Molecular Basket Sorbents for Co2 Capture: Effects of Pore Structure of Supports and Loading Level of Polyethylenimine. Catal. Today 2014, 233, 100. (17) Ma, X. L.; Wang, X. X.; Song, C. S. ″Molecular Basket″ Sorbents for Separation of Co2 and H2s from Various Gas Streams. J. Am. Chem. Soc. 2009, 131, 5777−5783. (18) Calleja, G.; Sanz, R.; Arencibia, A.; Sanz-Pérez, E. S. Influence of Drying Conditions on Amine-Functionalized Sba-15 as Adsorbent of Co2. Top. Catal. 2011, 54, 135−145. (19) Heydari-Gorji, A.; Belmabkhout, Y.; Sayari, A. Polyethylenimine-Impregnated Mesoporous Silica: Effect of Amine Loading and Surface Alkyl Chains on Co2 Adsorption. Langmuir 2011, 27, 12411− 12416. (20) Srikanth, C. S.; Chuang, S. S. C. Spectroscopic Investigation into Oxidative Degradation of Silica-Supported Amine Sorbents for Co2 Capture. ChemSusChem 2012, 5, 1435−1442. (21) Bollini, P.; Didas, S. A.; Jones, C. W. Amine-Oxide Hybrid Materials for Acid Gas Separations. J. Mater. Chem. 2011, 21, 15100− 15120. (22) Bollini, P.; Brunelli, N. A.; Didas, S. A.; Jones, C. W. Dynamics of Co2 Adsorption on Amine Adsorbents. 2. Insights into Adsorbent Design. Ind. Eng. Chem. Res. 2012, 51, 15153−15162. (23) Brunelli, N. A.; Didas, S. A.; Venkatasubbaiah, K.; Jones, C. W. Tuning Cooperativity by Controlling the Linker Length of SilicaSupported Amines in Catalysis and Co2 Capture. J. Am. Chem. Soc. 2012, 134, 13950−13953. (24) Hahn, M. W.; Steib, M.; Jentys, A.; Lercher, J. A. Mechanism and Kinetics of Co2 Adsorption on Surface Bonded Amines. J. Phys. Chem. C 2015, 119, 4126−4135. (25) Heydari-Gorji, A.; Sayari, A. Co2 Capture on PolyethylenimineImpregnated Hydrophobic Mesoporous Silica: Experimental and Kinetic Modeling. Chem. Eng. J. 2011, 173, 72−79. (26) Didas, S. A.; Zhu, R.; Brunelli, N. A.; Sholl, D. S.; Jones, C. W. Thermal, Oxidative and Co2 Iinduced Degradation of Primary Amines Used for Co2 Capture: Effect of Alkyl Linker on Stability. J. Phys. Chem. C 2014, 118, 12302. (27) Hahn, M. W.; Steib, M.; Jentys, A.; Lercher, J. A. Tailoring Hierarchically Structured Sio 2 Spheres for High Pressure Co2 Adsorption. J. Mater. Chem. A 2014, 2, 13624−13634. 1995

DOI: 10.1021/acs.jpcb.5b10012 J. Phys. Chem. B 2016, 120, 1988−1995