Stepwise Pore Size Reduction of Ordered Nanoporous Silica

May 13, 2013 - (c) Pagan-Torres , Y. J.; Gallo , J. M. R.; Wang , D.; Pham , H. N.; Libera , J. A.; Marshall , C. L.; Elam , J. W.; Datye , A. K.; Dum...
0 downloads 0 Views 570KB Size
Download PDF
Subscriber access provided by BRIGHAM YOUNG UNIV

Communication

Stepwise pore size reduction of ordered nanoporous silica materials at angstrom precision Siddharth Jambhrunkar, Meihua Yu, Jie Yang, Jun Zhang, Abhijit Shrotri, Liliana Endo-Munoz, Jo#L J. E. Moreau, Gaoqing (Max) Lu, and Chengzhong Yu J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/ja402463h • Publication Date (Web): 13 May 2013 Downloaded from http://pubs.acs.org on May 14, 2013

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Journal of the American Chemical Society is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 5

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society

Stepwise pore size reduction of ordered nanoporous silica materials at angstrom precision a

a

a

a

a

Siddharth Jambhrunkar , Meihua Yu , Jie Yang , Jun Zhang , Abhijit Shrotri , Liliana EndoMunozb, Joël Moreauc, Gaoqing Lua, Chengzhong Yu*a a

ARC Centre of Excellence for Functional Nanomaterials and Australian Institute for Bioengineering and Nanotechnology, The University of Queensland, Brisbane, QLD 4072, Australia.

b

Diamantina Institute, Princess Alexandra Hospital, Ipswich Rd, Brisbane QLD 4102, Australia

c

Ecole Nationale Supérieure de Chimie de Montpellier, 34296 Montpellier CEDEX 5, France

ABSTRACT: A facile vacuum-assisted vapor deposition process has been developed to control the pore size of ordered mesoporous silica materials in a stepwise manner with angstrom precision, providing an unprecedented paradigm to screen a designer hydrophobic drug nanocarrier with optimised pore diameter to maximise drug solubility.

Since the discovery of MCM-411 and SBA-15,2 ordered mesoporous silica materials have attracted ever increasing attention owing to their potential applications in many fields.3 The uniform mesopores and adjustable pore sizes are important characteristics of mesoporous materials, critically important for the controlled immobilization and release behavior of guest molecules,4 catalytic reaction performance at confined nanospace,5 controlled release of hydrophobic drugs4,6 and as nanocarrier for water insoluble drugs.7 To date, much effort has been made to control the pore size of ordered mesoporous silica materials. Generally, the pore size can be increased using post synthesis methods or by the addition of swelling agents,2 or by varying the structure-directing templates.8 Alternatively, the pore size can be reduced by the grafting method,9 metallorganic chemical vapor deposition technique10 or atomic layer deposition (ALD) technique.11 Of these, ALD technique has been widely used to modify mesoporous silica materials with another oxide phase (e.g. TiO2, HfO2),11b-d showing pore size reduction at atomic scale. However, it is difficult to achieve a constant pore size reduction in each cycle. Moreover, unlike zeolites or metal organic frameworks whose pore sizes can be designed at angstrom precision for size selective applications,12 it is a big challenge to achieve such precise control over the pore size of ordered mesoporous silica materials since their discovery, especially in a wide mesopore size range and in a reproducible manner.

Figure 1. A) A scheme demonstrating the VVD approach to stepwise reduce the pore size of ordered mesoporous silica materials in a precise range using TEOS or TMOS (abbreviated E or M). N is the cycle number, PN and PN+1 are the pore diameters before and after N cycle, respectively. ∆P denotes

the pore size change after one VVD cycle. ∆P ¯ denotes the mean pore size reduction and expressed as mean ± standard deviation after 3 cycles except for MCM-41 with TMOS where only 2 cycles were included in the linear regression. B) Variation of pore size, P, as a function of N for both SBA-15 and 2 MCM-41 materials. R is the regression coefficient.

Up to 40% of new chemical drugs emerging from high throughput screening process are poorly water soluble, posing a challenge to improve their solubility and hence bioavailability.13 Various approaches have been studied to address this issue,14 among which reducing the hydropho-

1

ACS Paragon Plus Environment

Journal of the American Chemical Society

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

bic drug particle size15 is an efficient one as reflected by the Ostwald–Freundlich equation.16 However, drug particle size reduction methods find limitation to downsize the drug particle size below 20 nm, a regime in which a dramatic increase in solubility is expected. Ordered mesoporous silica materials provide an adjustable confined nanospace, hence the size of drug loaded inside can be limited. By confining Indole-3-butyric acid, a hydrophobic drug into MCM-41 with a pore size of 3.31 nm, its solubility was increased by ca. twice compared to the drug itself.7c Surprisingly, this is a rare case showing the influence of nano-confinement towards the solubility of drugs. It is not known whether this concept can be generally applied to other hydrophobic drugs. Importantly, the relationship between the pore size of host mesoporous material and the solubility of loaded drug has not been reported to our knowledge. We hypothesize that there exists an optimized pore size displaying the highest solubility for a given hydrophobic drug (See Supporting Information, SI, Influence of pore size on solubility). This is a theory to be tested, which relies on designing a series of mesoporous silica materials with precisely controlled pore sizes and encapsulating the drug in them.

Page 2 of 5

D refers to the deposition step, C refers to the calcination step (see Figure 1 for N, E and M). The X-ray diffraction (XRD) patterns of SBA-15-CN-E (N=0-3) are shown in Figure S2A. For SBA-15-C0-E before VVD treatment, three well-resolved diffractions appear at 2θ of 0.90, 1.56 and 1.81° with a reciprocal d-spacing ratio close to 1: √3: 2, which can be indexed as 100, 110 and 200 reflections of an ordered 2D hexagonal mesostructure (p6mm). From the intense (100) peak, a d100 spacing of 9.81 nm is calculated, corresponding to a unit cell parameter (a) of 11.33 nm (Table S1). The XRD patterns for SBA15-CN-E (N=1-3) after every VVD cycle are similar to that of SBA-15-C0-E, demonstrating retention of the ordered hexagonal mesostructure. Moreover, the positions of three diffractions are similar for 4 materials. In addition, a gradual increase in the relative intensity of 200 to 110 peaks can be observed, suggesting a slight increase in wall thickness in each VVD cycle18. The N2 adsorption-desorption isotherms for SBA-15-CNE (N=0-3) are shown in Figure S3A, all exhibiting a typical type IV isotherm and a steep capillary condensation step occurring at a relative pressure (P/P0) range of 0.7-0.8, characteristic of ordered mesoporous material with large and uniform mesopores. The pore size variation in SBA15-CN-E can be clearly seen from their pore size distribution curves calculated from the adsorption branch using a BJH model after gaussian fitting (Figure S2B). The pore size is calculated to be 6.67, 6.40, 6.07, 5.71 nm (N= 0, 1, 2, 3, respectively), and the pore size distribution curve is narrow even after 3 VVD cycles. After linear regression analysis, an equation of (P= -0.31N+6.69) can be obtained with a regression coefficient of 0.9960 (Figure 1B), indicating a high statistical significance of the VVD process to achieve a fine control over pore size. The pore sizes as well as surface areas and pore volumes of SBA-15-CN-E are summarized in Table S1 for comparison.

Herein, we report a facile vacuum-assisted vapor deposition (VVD) approach to stepwise and precisely reduce the pore size of ordered mesoporous silica materials (Figure 1). The step size is adjustable at 0.29±0.06 nm using tetraethoxyorthosilicate (TEOS) or at 0.54±0.06 nm using tetramethoxyorthosilicate (TMOS) as the silica precursor. A surface chemistry switch mechanism is proposed as shown in Figure 1A. Taking advantage of the surface silanols at the pore surface, the TEOS or TMOS vapor molecules are anchored on the pore surface in a VVD step I, followed by a calcination step II to complete one cycle. Upon completion of one cycle, the pore diameter is reduced and the silanols are regenerated which can be used for a subsequent cycle. This approach has been tested for two ordered mesoporous silica materials, SBA-152 and MCM-411 with different initial pore sizes (Figure 1B). In consecutive cycles, the pore size reduction is consistent in each cycle for both materials and/or silica precursors. By encapsulating a hydrophobic drug, curcumin in the series of MCM-41 with controlled pore sizes, a pore size – solubility function is obtained which shows an optimized pore size (1.70 nm) where curcumin possesses the highest solubility (4.5 times higher compared to the drug itself).

In order to understand why the pore size can be precisely reduced in the VVD process after every cycle in a stepwise nature, Fourier transform infrared (FTIR) analysis has been carried out for SBA-15 materials treated in each step of 3 cycles. To observe clearly the surface chemistry change, only regions of interest are shown in Figure 2. Before the VVD process, SBA-15-C0-E exhibits one peak at 3745 cm-1 (Figure 2A) assigned to isolated silanols (ν(OH)).19 After the deposition step (SBA-15-D1-E), this peak cannot be observed. Instead, a group of peaks at 2902, 2934 and 2980 cm-1 appear, which can be assigned to ν(CH) of unhydrolysed ethoxy (Si–OCH2CH3) groups.19c After the calcination step (SBA-15-C1-E), the peaks assigned to ethoxy groups disappear while the peak at 3745 cm-1 associated with surface silanols is regenerated. The above results indicate that in one VVD cycle, the surface silanols are "switched-off" and ethoxy groups "on" in the deposition step, then silanols "switched-on" but ethoxy "off" in the calcination step (see Figure 1A). Moreover, the surface chemistry "on-off" is repeatedly observed in consecutive 3 cycles.

SBA-15 and MCM-41 were prepared according to published methods17 with slight variations (see SI, Experimental Section). To adjust the pore size, 0.8 g of calcined SBA-15 or MCM-41 material is placed in a home-designed VVD apparatus to undergo the VVD process (Figure S1). The VVD process was performed by exposing mesoporous silica material to either TEOS or TMOS for 24 h at 60°C during step I of VVD cycle followed by step II of calcination (Figure 1A). The sample treated in one VVD cycle is denoted as sample-DN-E(M) or sample-CN-E(M), where

2

ACS Paragon Plus Environment

Page 3 of 5

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society retained after 3 VVD cycles (Figures S3, S4). FTIR spectra (Figure S5) demonstrate that the mechanism proposed in Figure 1 is reproducible for both SBA-15 and MCM-41 materials. After linear regression, an equation of P = 0.28N+2.37 can be obtained with a regression coefficient of 0.9942 (Figure 1B), confirming that our VVD process can be applied for both SBA-15 and MCM-41 to achieve a stepwise pore size reduction in a precise range of 0.29±0.06 nm per step. The VVD process was repeated on freshly synthesized SBA-15 and MCM-41, further demonstrating its reproducibility (see SI, Reproducibility of VVD process, Figures S6-S8, Table S2). The VVD process was further conducted on SBA-15 and MCM-41 materials using TMOS as the silica precursor. XRD patterns for SBA-15-CN-M (Figure S9A) and MCM41-CN-M (Figure S10A) show that the ordered hexagonal structure is retained after each VVD cycle. N2 adsorptiondesorption isotherms for SBA-15-CN-M (N=0-3) and MCM-41-CN-M (N=0-2) (Figure S11) exhibit typical type IV isotherm similar to the results obtained for SBA-15-CN-E (N = 0-3) and MCM-41-CN-E (N=0-3). The pore sizes as well as surface areas and pore volumes of SBA-15-CN-M (N= 0-3) and MCM-41-CN-M (N = 0-2) are summarized in Table S1. For MCM-41-C3-M, the mesopores are nearly blocked and it is very difficult to detect the pore size using BJH model as the pore size is in microporous range. Hence, the pore size distribution curve for MCM-41 material is calculated only for 2 VVD cycles. The pore size distribution data for SBA-15-CN-M (N = 03) and MCM-41-CN-M (N = 0-2) are shown in Figures S9B and S10B, respectively. After linear regression analysis, an equation of P = -0.53N + 6.39 and P = -0.56N + 2.15 is obtained with regression coefficient of 0.9962 and 1 for SBA15 and MCM-41 group, respectively, indicating high statistical significance (Figure 1B). By using TMOS in VVD process, a stepwise pore size reduction in a range of 0.54±0.06 nm per step is achieved for both SBA-15 and MCM-41. FTIR analysis was performed for SBA-15-CN-M and MCM-41-CN-M. Surface chemistry “on-off” phenomenon similar to SBA-15-CN-E and MCM-41-CN-E is also observed (Figures S12 and S13).

Figure 2. FTIR spectra of SBA-15 sample before and after

treatment with VVD process for 3 cycles using TEOS as the silica precursor. Samples are named as SBA-15-C/D-N-E, where C denotes calcined sample, D denotes TEOS deposited sample, N denotes the cycle number of VVD process and E denotes TEOS.

The surface chemistry switch mechanism is further confirmed by Figures 2B and C. The bands associated with ethoxy groups at 1166 (ρ(CH3)), 1370 (ω(CH2)), 1394 (δs(CH3)), 1445 (δ′a(CH2)) and 1484 cm-1 (δa(CH3)+δa(CH2)) are observed only after each deposition step. The band at 962 cm-1 (Figure 2C) is the characteristic peak for ρ(CH3), which overlaps with the band of silanols (occurred after calcination) with relatively weaker intensity.20 The band observed at 810 cm-1 after each calcination step (Figure 2C) can be indexed to symmetric Si– O–Si stretching (νs(Si-O-Si).19a After deposition, only two peaks at 794 (ν(Si-O)) and 815 cm-1 (δs(Si-O-C)) indexed to the non-reacted (Si-OCH2CH3) groups are observed.21 The intensity of the band associated with physiosorbed water at 1630 cm-1 is higher when Si-OH is "switched on" while lower when Si-OH is "switched off", consistent with the hydrophilicity/hydrophobicity change of pore walls and the surface chemistry switch mechanism.

To understand the difference in pore size reduction step using TEOS (0.29±0.06 nm) or TMOS (0.54±0.06 nm) in our VVD process, the weight change occurring in the deposition step I was carefully monitored for SBA-15 materials using TEOS as the silica precursor. The weight gain observed for SBA-15-DN-E (N=1-3) is listed in Table S3. After each deposition step, the number of TEOS molecule deposited on internal surface of SBA-15 is a constant value of 0.9/nm2. Because TEOS has a molecule size of 0.93 nm,22 the surface coverage constant of 0.9/nm2 is consistent with a monolayer like deposition behavior on the wall surface (Figure S14A). After the calcination step II of converting -Si(OEt)3 to Si-OH, the newly generated Si-OH must be distributed sparsely due to its smaller size compared to -Si(OEt)3. It is noted that a densely packed SiOH monolayer, as shown in Figure S14C, would generate a

The VVD process was further applied to MCM-41 materials using TEOS as the silica precursor. Both XRD and nitrogen sorption results show that the ordered hexagonal mesostructure with narrow pore size distribution can be

3

ACS Paragon Plus Environment

Journal of the American Chemical Society

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

pore size reduction of 0.62 nm theoretically. The sparsely distributed Si-OH in our model (Figure S14B) depicts the observed pore size reduction of 0.29±0.06 nm. Because TMOS molecule is smaller in size compared to TEOS, the number of TMOS molecules deposited on internal surface of SBA-15 during deposition step I should be higher, (see Figure S14A), leading to relatively higher density of newly generated Si-OH after the calcination step II, which results in a larger pore size reduction step of 0.54±0.06 nm.

Page 4 of 5

Corresponding Author * E-mail: [email protected].

ACKNOWLEDGMENT This work was supported by the Australian Research Council.

REFERENCES

(1) C. T. Kresge, M. E. Leonowicz, W. J. Roth, J. C. Vartuli and J. S. Beck, Nature, 1992, 359, 710. (2) D. Y. Zhao, J. L. Feng, Q. S. Huo, N. Melosh, G. H. Fredrickson, B. F. Chmelka and G. D. Stucky, Science, 1998, 279, 548. (3) (a) M. Vallet-Regi, F. Balas and D. Arcos, Angew. Chem., Int. Ed., 2007, 46, 7548. (b) G. A. Somorjai, H. Frei and J. Y. Park, J. Am. Chem. Soc., 2009, 131, 16589. (4) F. Balas, M. Manzano, P. Horcajada and M. Vallet-Regi, J. Am. Chem. Soc., 2006, 128, 8116. (5) (a) J. Y. Ying, C. P. Mehnert and M. S. Wong, Angew. Chem., Int. Ed., 1999, 38, 56. (b) M. Nandi, J. Mondal, K. Sarkar, Y. Yamauchi and A. Bhaumik, Chem. Commun., 2011, 47, 6677. (6) R. Mellaerts, J. A. G. Jammaer, M. Van Speybroeck, H. Chen, J. Van Humbeeck, P. Augustijns, G. Van den Mooter and J. A. Martens, Langmuir, 2008, 24, 8651. (7) (a) J. Lu, M. Liong, J. I. Zink and F. Tamanoi, Small, 2007, 3, 1341. (b) J. Lu, M. Liong, Z. X. Li, J. I. Zink and F. Tamanoi, Small, 2010, 6, 1794. (c) V. Ambrogi, F. Famiani, L. Perioli, F. Marmottini, I. Di Cunzolo and C. Rossi, Micropor. Mesopor. Mat., 2006, 96, 177. (8) C. Z. Yu, J. Fan, B. Z. Tian, G. D. Stucky and D. Y. Zhao, J Phys Chem B, 2003, 107, 13368. (9) X. S. Zhao, G. Q. M. Lu and X. Hu, Chem. Commun., 1999, 1391. (10) Y. Zhang, F. L. Y. Lam, X. J. Hu and Z. F. Yan, Chem. Commun., 2008, 5131. (11) (a) B. S. Berland, I. P. Gartland, A. W. Ott and S. M. George, Chem. Mater., 1998, 10, 3941. (b) S. Mahurin, L. Bao, W. Yan, C. Liang and S. Dai, J. Non-Cryst. Solids, 2006, 352, 3280. (c) Y. J. Pagan-Torres, J. M. R. Gallo, D. Wang, H. N. Pham, J. A. Libera, C. L. Marshall, J. W. Elam, A. K. Datye and J. A. Dumesic, ACS Catal., 2011, 1, 1234. (d) J. Dendooven, B. Goris, K. Devloo-Casier, E. Levrau, E. Biermans, M. R. Baklanov, K. F. Ludwig, P. Van der Voort, S. Bals and C. Detavernier, Chem. Mater., 2012, 24, 1992. (12) J. R. Li, R. J. Kuppler and H. C. Zhou, Chem. Soc. Rev., 2009, 38, 1477. (13) C. Lipinski, Am.pharm.Rev., 2002, 5, 82. (14) R. Chadha, V. K. Kapoor, D. Thakur, R. Kaur, P. Arora and D. V. S. Jain, J. Sci. Ind. Res. India, 2008, 67, 185. (15) B. E. Rabinow, Nat. Rev. Drug Discov., 2004, 3, 785. (16) W. Ostwald, Z. Phys. Chem.-Stoch. Ve, 1900, 34, 495. (17) (a) D. H. Pan, P. Yuan, L. Z. Zhao, N. A. Liu, L. Zhou, G. F. Wei, J. Zhang, Y. C. Ling, Y. Fan, B. Y. Wei, H. Y. Liu, C. Z. Yu and X. J. Bao, Chem. Mater., 2009, 21, 5413. (b) S. Yang, L. Z. Zhao, C. Z. Yu, X. F. Zhou, J. W. Tang, P. Yuan, D. Y. Chen and D. Y. Zhao, J. Am. Chem. Soc., 2006, 128, 10460. (18) S. A. Kozlova and S. D. Kirik, Micropor. Mesopor. Mat., 2010, 133, 124. (19) (a) M. Muroya, Colloid Surface A, 1999, 157, 147. (b) D. H. Pan, L. Z. Zhao, K. Qian, L. Tan, L. Zhou, J. Zhang, X. D. Huang, Y. Fan, H. Y. Liu, C. Z. Yu and X. J. Bao, J. Mater. Res., 2011, 26, 804. (c) M. I. Tejedor-Tejedor, L. Paredes and M. A. Anderson, Chem. Mat., 1998, 10, 3410. (20) J. Zhang, M. H. Yu, P. Yuan, H. N. Wang, K. Qian, L. Tan, Y. H. Wang and C. Z. Yu, J. Mater. Res., 2010, 25, 648. (21) (a) C. Velasco-Santos, A. L. Martinez-Hernandez, W. Brostow and V. M. Castano, J. Nanomater., 2011, 2011. (b) E. Bourgeat-Lami, I. Tissot and F. Lefebvre, Macromolecules, 2002, 35, 6185. (22) A. M. Wrobel, A. WalkiewiczPietrzykowska, M. Stasiak and J. Kulpinski, Chem. Vapor. Depos., 1996, 2, 285.

Figure 3. The correlation between pore size and curcumin solubility in water, after loading curcumin in a series of calcined MCM-41 with precisely controlled pore sizes.

Finally, the series of MCM-41 nanoporous materials with precisely controlled pore sizes were utilized to confine the hydrophobic drug curcumin to study the pore size – solubility relationship. As shown in Figure 3, pure curcumin has a very low solubility in water of 0.31 µg/ml. Curcumin encapsulated in pristine MCM-41 with a pore size of 2.22 nm has improved solubility of 0.53 µg/ml. The curcumin solubility further increases with decreasing pore size, reaching a peak point at 1.70 nm with a maximized solubility of 1.40 µg/ml, 4.5 times of pure curcumin. However, curcumin solubility diminishes significantly when the pore size is further decreased from 1.70 to 1.50 and 1.02 nm. Our experimental observation is in good accordance with the theoretical predictions (See SI, Influence of pore size on solubility). In conclusion, we have developed a novel VVD approach to stepwise reduce the pore size of mesoporous silica materials at angstrom precision. The pore size – solubility relationship has been established to identify a mesoporous material with an optimized pore size displaying maximized drug solubility. With our discovery, the pore sizes of ordered mesoporous materials can be adjusted in a broad regime and at a precision that has not been achieved before, offering new materials for advanced pore size dependent applications.

ASSOCIATED CONTENT Supporting Information: Experimental methods and other characterizations. This material is available free of charge via the Internet at http://pubs.acs.org.

`

4

ACS Paragon Plus Environment

Page 5 of 5

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society

5

ACS Paragon Plus Environment