molecules Article
Efficient Manipulation of Continuous AFI-Type Aluminophosphate Membranes with Distinctive Microstructures on Macroporous α-Al2O3 Substrates Luwei Geng, Hongfeng Dong *, Xiufeng Liu and Baoquan Zhang
ID
School of Chemical Engineering and Technology, Tianjin University, Tianjin 300350, China;
[email protected] (L.G.);
[email protected] (X.L.);
[email protected] (B.Z.) * Correspondence:
[email protected]; Tel.: +86-22-8535-6517 Received: 23 April 2018; Accepted: 8 May 2018; Published: 9 May 2018
Abstract: The availability of continuous and well-defined AFI-type aluminophosphate membranes (AFI membranes) would trigger their applications in innovative materials. A well-designed manipulation strategy is proposed to produce continuous AFI membranes with four different microstructures over porous α-Al2 O3 substrates. A double-layer and highly c-oriented AFI membrane of hexagonal prisms is obtained when a thin layer of medium molecular weight (MMW) chitosan is employed as the structure-directing matrix together with aluminum isopropoxide (AIP) as the Al source. It can be transformed to a single-layer and highly c-oriented AFI membrane of hexagonal prisms if the structure-directing matrix is replaced by a thin layer of low molecular weight (LMW) chitosan. When the Al source is changed to pseudo-boehmite, the single-layer AFI membrane is composed of highly ordered spherical agglomerates of small crystals. Furthermore, the membrane will turn to the double-layer AFI membrane of highly-ordered crystal agglomerates if a thin layer of MMW chitosan is used once again, keeping pseudo-boehmite as the Al source. The manipulation methodology established here is rather reliable with a pretty high reproducibility. Keywords: AFI-type aluminophosphate; oriented membrane; chitosan; microstructure manipulation
1. Introduction Microporous crystalline materials, mainly including zeolites and aluminophosphate molecular sieves, have attracted extensive interests in many fields such as adsorption, catalysis, ion exchange, microelectronics, and chemical sensing due to their large specific surface areas, defined channel systems, high mechanical strength, and hydrothermal stability. Zeolites containing cages and pores of molecular dimensions, which give them their size and shape selectivity, have received intense research for over 60 years [1]. Furthermore, aluminophosphate molecular sieves (AlPO-n), which are built of strictly alternating [AlO4 ] and [PO4 ] tetrahedral units, form a new class of microporous inorganic solids. Following the initial report of aluminophosphates in 1982 [2], the isomorphous substitution of aluminum and phosphorus in the framework with silicon and metals were successful in 1984 and 1986, respectively [3,4]. Aluminophosphates exhibit both compositional and structural diversities. The structural features and unique physicochemical properties of microporous aluminophosphates have expanded the application area of microporous materials [5]. The AFI-type molecular sieve (AlPO4-5 with derivatives), which was composed of one-dimensional channels with a pore diameter of 0.73 nm aligned parallel to its crystallographic c-axis, is one of the most and earliest studied members in the family of aluminophosphates [5,6]. Zeolite and molecular sieve membranes with uniform and molecular-sized pores have excellent thermal, mechanical, and chemical stabilities, and they have acquired potential applications as
Molecules 2018, 23, 1127; doi:10.3390/molecules23051127
www.mdpi.com/journal/molecules
Molecules 2018, 23, 1127
2 of 14
separation membranes and membrane catalysts [7–11]. Besides, new application areas such as low-dielectric constant materials as well as anti-corrosion and anti-microbial coatings have been developed [12,13]. The microstructure of zeolite and molecular sieve membranes including continuity, crystal orientation, and thickness is of great significance in actual applications [8,14]. For instance, the denseness of a molecular sieve membrane is essential for the realization of molecular-sieving function. The zeolite membrane with a preferential crystal orientation shows a greatly improved performance. Ultrathin membranes possess significant advantages that improve permeation performance, while thick membranes prolong the mean residence time of reactants within the catalysis membrane to increase the reaction conversion. However, manipulating the microstructure of zeolite membranes is still a challenging issue. The synthesis of AFI-type aluminophosphate membranes (AFI membranes) has been focused on controlling the crystal orientation to achieve c-oriented channels perpendicular to the substrate surface, leading to the uniform residence time of involved molecules across diffusion channels. Up to the present, there have been several routes for fabricating AFI arrays, films, and membranes with a preferred orientation, including electric field-driven assembly [15–19], geometry-confined growth [20–22], epitaxial growth [23–26], the manipulation of synthesis conditions [19,27–29], microwave-enhanced growth [6,30,31], and induced growth at the substrate surface [32,33]. The highly oriented AFI arrays and films were produced by the electric field-driven assembly or the geometry-confined growth, which could be used as the seed layers in the following seeded growth step to make continuous, dense, and preferentially c-oriented AFI membranes [19,21,22]. Besides, the array or film that was synthesized as such had to be fixed to the substrate surface using other materials before actual applications [16,17]. Basically, the availability of oriented seed layers would be a prerequisite to the epitaxial growth [34]. The manipulation of synthesis conditions was a commonly used method, which required a great deal of experimental investigations [35]. If the growth mechanism and critical influence factors were clear, the cumbersome experiments could be substantially reduced. The manipulation of synthesis conditions could be further enhanced by using microwave heating, which had been applied to either the in situ crystallization [6] or the seeded growth [30,31]. It should be noted that the majority of the AFI membranes that have been reported to date were prepared on dense substrates such as silicon wafers or glass plates, which would not be usable in separation and catalysis processes. By the surface induction, the c-oriented AlPO4-5 crystal grains were attached onto the zirconium phosphonate-modified gold surface because of the strong affinity of phosphonic acid groups to the (00l) planes [32]. In our previous report [33], a precoated chitosan thin layer on a macroporous α-Al2 O3 substrate was used to act as the orientation directing matrix. After the in situ crystallization, the well intergrown and preferentially c-oriented SAPO-5 films over a closely packed layer of SAPO-5 crystals was formed, accompanied with the gradual dissolution of chitosan. Herein, we wish to further manipulate the microstructure of AFI membranes in the in situ crystallization. As such, continuous and dense AFI membranes with single and double-layered structures covered with either highly c-oriented hexagonal prisms or spherical agglomerates could be synthesized at will. There are not relevant reports yet. 2. Results 2.1. Double-Layer and c-Oriented AFI Membranes In our previous work, a thin layer of MMW chitosan was coated on the α-Al2 O3 substrate to induce the formation of a double-layer SAPO-5 membrane, i.e., a highly c-oriented and well intergrown SAPO-5 top layer over a closely packed bottom layer of SAPO-5 crystals (Figure 1). After the c-oriented SAPO-5 crystals on the top layer were completely formed, the precursor solution was unable to penetrate into the bottom layer, leading to the formation of the double-layer SAPO-5 membrane on the α-Al2 O3 substrate. It should be noted that the continuity and thickness of the highly c-oriented
Molecules 2018, 23, 1127
3 of 14
Molecules 2018, 23, x FOR PEER REVIEW
3 of 14
SAPO-5 layer and the packed bottom layer bebottom optimized bycould regulating the ingredients of the highly top c-oriented SAPO-5 top layer and the could packed layer be optimized by regulating synthesis solution and the substrate setting. the ingredients of the synthesis solution and the substrate setting.
(b)
(c) Intensity(a.u.)
(a)
(002)
(100)
(210) (211)
*
* *
*
SAPO-5 membrane
α-Al2O3
10
20
30
40
50
2θ (degree)
Figure Top view and cross-sectional SEM images together with corresponding XRD Figure 1. 1. (a)(a) Top view and (b)(b) cross-sectional SEM images together with (c)(c) thethe corresponding XRD pattern for the double-layer SAPO-5 membrane induced by a medium molecular weight (MMW) pattern for the double-layer SAPO-5 membrane induced by a medium molecular weight (MMW) chitosan layer using aluminum isopropoxide (AIP) as the Al source. Scale correspond 20 in μm chitosan layer using aluminum isopropoxide (AIP) as the Al source. Scale barsbars correspond to 20toµm 2O3 substrate. in SEM images. The asterisks in XRD patterns mark the peaks originating from the α-Al SEM images. The asterisks in XRD patterns mark the peaks originating from the α-Al2 O3 substrate.
In the following studies, the optimized substrate setting (facing downwards) and the SiO2/Al2O3 In the following studies, the optimized substrate setting (facing downwards) and the SiO2 /Al2 O3 ratio in the synthesis solution (0.15) would be adopted unless otherwise stated. Due to the gradual ratio in the synthesis solution (0.15) would be adopted unless otherwise stated. Due to the gradual dissolution of chitosan in the in situ crystallization, both the surface and entire body of the chitosan dissolution of chitosan in the in situ crystallization, both the surface and entire body of the chitosan layer would serve as a three-dimensional structure-directing matrix. The dissolution rate of chitosan layer would serve as a three-dimensional structure-directing matrix. The dissolution rate of chitosan ought to be a key factor to influence the formation of SAPO-5 membranes with a unique ought to be a key factor to influence the formation of SAPO-5 membranes with a unique microstructure microstructure on porous α-Al2O3 substrates. on porous α-Al2 O3 substrates. 2.2. Single-Layer and c-Oriented AFI Membranes 2.2. Single-Layer and c-Oriented AFI Membranes According to a literature report [36], a chitosan tablet usually swelled from the surface to its According to a literature report [36], a chitosan tablet usually swelled from the surface to its inner inner part in acidic or neutral media while the internal glassy region would decline until the glassy part in acidic or neutral media while the internal glassy region would decline until the glassy core core vanished. All swollen gels would dissolve gradually. Compared with low molecular weight vanished. All swollen gels would dissolve gradually. Compared with low molecular weight (LMW) (LMW) chitosan, medium molecular weight (MMW) chitosan swelled to some deeper extent while chitosan, medium molecular weight (MMW) chitosan swelled to some deeper extent while exhibiting exhibiting a low erodible rate. The dissolution rate of both LMW and MMW chitosan in the acidic a low erodible rate. The dissolution rate of both LMW and MMW chitosan in the acidic solution solution was also checked in our experiment, the result of which was in accordance with that of was also checked in our experiment, the result of which was in accordance with that of Huanbutta Huanbutta et al. [36]. As such, we would assume a LMW chitosan coating with a thinner swelling et al. [36]. As such, we would assume a LMW chitosan coating with a thinner swelling layer and layer and increased dissolution rate would direct the crystal growth another way, leading to the increased dissolution rate would direct the crystal growth another way, leading to the formation of formation of an AFI membrane with a novel microstructure. Therefore, LMW chitosan was an AFI membrane with a novel microstructure. Therefore, LMW chitosan was employed to prepare employed to prepare a thin layer on the α-Al2O3 substrate to induce the growth of SAPO-5 films in a thin layer on the α-Al2 O3 substrate to induce the growth of SAPO-5 films in the in situ crystallization. the in situ crystallization. As shown in Figure 2, a single-layer SAPO-5 film could be formed on the α-Al2 O3 substrate due As shown in Figure 2, a single-layer SAPO-5 film could be formed on the α-Al2O3 substrate due to the induction of the LMW chitosan layer. The SAPO-5 film was composed of hexagonal prisms, to the induction of the LMW chitosan layer. The SAPO-5 film was composed of hexagonal prisms, most of which were aligned normally to the substrate surface. The SAPO-5 film exhibited preferred most of which were aligned normally to the substrate surface. The SAPO-5 film exhibited preferred c-orientation, as confirmed by the XRD pattern where the diffraction peak of (002) plane at 21.05◦ c-orientation, as confirmed by the XRD pattern where the diffraction peak of (002) plane at 21.05° was much stronger. However, the compactness of the SAPO-5 film was not satisfactory at the present was much stronger. However, the compactness of the SAPO-5 film was not satisfactory at the stage. The manipulation was further required to increase the continuity and crystal orientation of the present stage. The manipulation was further required to increase the continuity and crystal synthesized SAPO-5 membrane. orientation of the synthesized SAPO-5 membrane. Based on available reports in the literature [27,32,37], the effect of water content on the morphology Based on available reports in the literature [27,32,37], the effect of water content on the and size of AFI-type crystal grains was quite remarkable. A high H2 O/Al2 O3 ratio in the synthesis morphology and size of AFI-type crystal grains was quite remarkable. A high H2O/Al2O3 ratio in the solution favored preferential growth along the c-axis, resulting in AFI-type crystals with a high aspect synthesis solution favored preferential growth along the c-axis, resulting in AFI-type crystals with a ratio (length versus diameter). Decreasing the water content would suppress the one-dimensional high aspect ratio (length versus diameter). Decreasing the water content would suppress the growth along the c-axis to form blocky or even plate-like crystals. In the formation of highly c-oriented one-dimensional growth along the c-axis to form blocky or even plate-like crystals. In the formation AFI membranes, the fast growth along the c-axis (vertical to the substrate surface) would slow down of highly c-oriented AFI membranes, the fast growth along the c-axis (vertical to the substrate the in-plane growth, and restrain the intergrowth of crystal grains. A suitable value of the water surface) would slow down the in-plane growth, and restrain the intergrowth of crystal grains. A content in the synthesis solution would be essential to the manipulation of the microstructure. suitable value of the water content in the synthesis solution would be essential to the manipulation of the microstructure.
Molecules 2018, 23, 1127
4 of 14
Molecules 2018, 23, x FOR PEER REVIEW
4 of 14
(c)
(b)
(a)
(002)
*
(210) (211)
(100)
Intensity(a.u.)
*
*
*
SAPO-5 membrane
Molecules 2018, 23, x FOR PEER REVIEW
4 of 14 α-Al2O3
(c)
(b)
(a)
(002) 20
10
30
40
2θ (degree)
50
* * (100) (211) Figure 2. (a) Top view and (b) cross-sectional SEM images together with (c) the (210) corresponding XRD Intensity(a.u.)
* * Figure 2. (a) Top for view (b) cross-sectional SEM images with (c) the corresponding XRD pattern the and single-layer SAPO-5 membrane induced by a together layer of LMW chitosan using AIP as the pattern for Al thesource. single-layer membrane by aThe layer of LMW chitosan using AIP as the Scale barsSAPO-5 correspond to 20 μm ininduced SEM images. asterisks in XRD patterns mark the SAPO-5 membrane from the α-Al O3 substrate. Al source. peaks Scaleoriginating bars correspond to 220 µm in SEM images. The asterisks in XRD patterns mark the peaks originating from the α-Al2 O3 substrate. α-Al O
When the molar ratio of H2O/Al2O3 in the synthesis solution was 300, the as-synthesized 10 20 30 40 50 θ (degree) crystals were elongated hexagonal prisms, which arrayed upwards in general 2with an inclination and unfilled defects everywhere 2). According to thewith XRD given in Figure When angle the molar ofview H2 and O/Al O3 in(Figure the synthesis was 300, the as-synthesized Figure 2.ratio (a) Top (b) 2cross-sectional SEM imagessolution together (c)pattern the corresponding XRD 2c, crystals the diffraction peaks of (100), (210), and (211) planes exhibited higher intensity. Moreover, the size of for the single-layer by a layer of LMW chitosan AIP as the were elongatedpattern hexagonal prisms,SAPO-5 whichmembrane arrayedinduced upwards in general with using an inclination angle and the crystals was relatively large, owing to the lower density nucleiinusing diluted Al source. Scale bars correspond to 20 μm in SEM images. The of asterisks XRD the patterns marksolution. the unfilled defects everywhere (Figure 2). According to to thetheXRD given in Figure 2c, the diffraction Whenpeaks the originating water content wasα-Al further decreased levelpattern of H2O/Al 2O3 = 200, the membrane from the 2O3 substrate. peaks of (100), (210),ofand (211) exhibited higher intensity. Moreover, the size of the crystals was composed smaller andplanes denser crystals possessed a high c-orientation and improved continuity, as shown in Figure 3a,d.lower A further 2O/Al2the O3 molar ratiowas to 150 inthe a highly When thetomolar ratio of Hdecrease 2O/Al 2Onuclei 3of inthe theH synthesis solution 300,resulted the as-synthesized relatively large, owing the density of using diluted solution. When water content SAPO-5 whereprisms, the hexagonal crystalsupwards were closely intergrown (Figure 3b,e). crystals were elongated hexagonal which arrayed in general with an inclination was furtherc-oriented decreased to themembrane level of H 2 O/Al2 O3 = 200, the membrane composed of smaller and denser The surface became much (Figure smoother without any crystal When a more anglemembrane and unfilled defects everywhere 2). According to the XRDboundaries. pattern given in Figure 2c, crystals possessed a high c-orientation improved continuity, as shown in Figure concentrated synthesis wasand used, theplanes hexagonal crystals could not be observed anymore the diffraction peaks ofsolution (100), (210), and (211) exhibited higher intensity. Moreover, the3a,d. sizeon ofA further decrease ofthe themembrane H2 O/Al Orelatively ratiothe to 150 indensity a highly c-oriented SAPO-5 membrane where although XRD pattern revealed a of preferentially membrane the crystals was large, owing toresulted the lower nuclei usingc-oriented the diluted solution. 2surface, 3 molar (Figure 3c,f). The emergence of further diffraction peaks atto9.5°, 16.0°, in2Othe pattern (Figure When the water content was decreased the3b,e). level and of H20.6° 2O/Al 3 =XRD 200,surface the membrane the hexagonal crystals were closely intergrown (Figure The membrane became much 3f) togetherofwith theand cubic crystals existed on thea high membrane surfaceand (Figure 3c) indicated composed denser crystals possessed c-orientation improved continuity,the smoother without anysmaller crystal boundaries. When a more concentrated synthesis solutionaswas used, structure SAPO-5 to SAPO-34 sieves.ratio Moreover, the enhancement of shown intransformation Figure 3a,d. Afrom further decrease of the Hmolecular 2O/Al2O3 molar to 150 resulted in a highly the hexagonal crystals could be where observed anymore on the membrane surface, although diffraction peaks ofmembrane thenot substrates implied that the continuity of the membrane decreased in3b,e). this the XRD c-oriented SAPO-5 the hexagonal crystals were closely intergrown (Figure pattern revealed a preferentially c-oriented (Figure 3c,f).boundaries. Thefor emergence case. Therefore, the molarbecame ratio ofmuch H2O/Al 2membrane O3 = 150without should be thecrystal appropriate the When synthesis of diffraction a The membrane surface smoother any a of more ◦ , 16.0◦ , SAPO-5 ◦ in membrane single-layer microstructure in thiswith system. concentrated synthesis solution wasaused, the hexagonal crystals could not bethe observed peaks at 9.5c-oriented and 20.6 the with XRD pattern (Figure 3f) together cubicanymore crystalsonexisted on the membrane surface, although the XRD pattern revealed a preferentially c-oriented membrane the membrane surface (Figure 3c) indicated the structure transformation from SAPO-5 to SAPO-34 (Figure The emergence of diffraction peaks at 9.5°, 16.0°, and 20.6° (c) in the XRD pattern (Figure (b) (a) 3c,f). molecular sieves. Moreover, enhancement peaks of the substrates implied 3f) together with the the cubic crystals existedofondiffraction the membrane surface (Figure 3c) indicated the that the continuity structure of the membrane decreased in this case. molecular Therefore, theMoreover, molar ratio of H2 O/Alof2 O3 = 150 transformation from SAPO-5 to SAPO-34 sieves. the enhancement diffraction peaks of for the substrates impliedofthat the continuity of the membrane decreased should be the appropriate the synthesis a c-oriented SAPO-5 membrane with ina this single-layer case. Therefore, the molar ratio of H2O/Al2O3 = 150 should be the appropriate for the synthesis of a microstructure in this system. c-oriented SAPO-5 membrane with a single-layer microstructure in this system. 2
(e)
(100)
(210) (211)
*
*
*
*
SAPO-5 membrane
(f)
(002)
(100)
(002)
(210) (211)
*
* *
*
SAPO-5 membrane
α-Al2O3
10
20
30
40
2θ (degree)
(100)
+
*
*
(210) (211)
+
*
*
+
SAPO-5 membrane
α-Al2O3 50
(002)
(e)
10
20
30
40
2θ (degree)
α-Al2O3 50
(002)
(f)
10
20
30
40
2θ (degree)
50
(002)
(100)
(210) (211)
*
*
*
*
SAPO-5 membrane
(100)
(210) (211)
*
* *
*
SAPO-5 membrane
α-Al2O3
10
20
30
2θ (degree)
40
Intensity(a.u.)
Figure 3. SEM images and the corresponding XRD patterns of SAPO-5 membranes induced by a layer of low molecular weight (LMW) chitosan using AIP as the Al source and the H2O/Al2O3 ratio of Intensity(a.u.)
Intensity(a.u.)
(d)
Intensity(a.u.)
(002)
Intensity(a.u.)
Intensity(a.u.)
(d)
(c)
(b)
(a)
3
(100)
+
+
*
*
(210) (211)
+
*
*
SAPO-5 membrane
α-Al2O3 50
10
20
30
2θ (degree)
40
α-Al2O3 50
10
20
30
2θ (degree)
40
Figure 3. SEM images and the corresponding XRD patterns of SAPO-5 membranes induced by a
50
Figure 3. SEM images and the corresponding XRD patterns of SAPO-5 membranes induced by a layer layer of low molecular weight (LMW) chitosan using AIP as the Al source and the H2O/Al2O3 ratio of of low molecular weight (LMW) chitosan using AIP as the Al source and the H2 O/Al2 O3 ratio of 200 (a,d), 150 (b,e), and 100 (c,f). Scale bars correspond to 20 µm in SEM images. The asterisks and plus signs mark the peaks from the α-Al2 O3 substrate and SAPO-34, respectively.
Molecules 2018, 23, x FOR PEER REVIEW
5 of 14
200 2018, (a,d),23, 150 (b,e), and 100 (c,f). Scale bars correspond to 20 μm in SEM images. The asterisks and5 of 14 Molecules 1127 plus signs mark the peaks from the α-Al2O3 substrate and SAPO-34, respectively. Molecules 2018, 23, x FOR PEER REVIEW
5 of 14
has been been revealed revealed that that the the existence existence of of silicon silicon species species would would be be conducive conducive to to the the formation formation of of ItIt has 200 (a,d), 150 (b,e), and on 100literature (c,f). Scale bars correspond to 20 μm in SEM images. The asterisksof and dense AFI membranes based reports and our experiments. In the absence silicon in dense AFI membranes based on literature reports and our experiments. In the absence of silicon in plus signs mark the peaks from the α-Al2O3 substrate and SAPO-34, respectively. the synthesis synthesis solution, solution, AlPO4-5 AlPO4-5 molecular molecular sieves sieves on on the the substrate substrate surface surface would would grow grow much much faster faster the along the c-axis, as shown in Figure 4, resulting in elongated crystals as long as 40–50 µm. This would along the c-axis, as shown Figure 4, resulting in elongated crystals as longtoas μm. ofThis It has been revealedinthat the existence of silicon species would be conducive the40–50 formation be detrimental the intergrowth of literature crystal with SAPO-5 synthesized in the in same would be detrimental to thebased intergrowth ofgrains. crystal grains. with SAPO-5 in dense AFIto membranes on reportsCompared and our Compared experiments. In the absencesynthesized of silicon conditions, AlPO4-5 molecular sieves were too slender be arranged regularly onmuch the substrate the conditions, synthesis solution, AlPO4-5 molecular sieves on the substrate surface would grow faster the same AlPO4-5 molecular sieves were tooto slender to be arranged regularly on the the c-axis, as shown in to Figure 4,toresulting elongated crystals asthe long 40–50 μm. This surfacealong (Figure 5a,b). According the XRD pattern (Figure 5c), the5c), degree ofasc-orientation of the substrate surface (Figure 5a,b). According the XRDinpattern (Figure degree of c-orientation would be detrimental to the intergrowth of crystal grains. Compared with SAPO-5 synthesized in of crystal layer was lower SAPO-5 membrane, and theand enhancement of the diffraction peaks of the crystal layer was than lowerthe than the SAPO-5 membrane, the enhancement of the diffraction the same conditions, AlPO4-5 molecular sieves were too slender to be arranged regularly on the the substrate suggested the discontinuity of the molecular sieve layer. peaks of the substrate suggested the discontinuity of the molecular sieve layer. substrate surface (Figure 5a,b). According to the XRD pattern (Figure 5c), the degree of c-orientation of the crystal layer was of the diffraction 7 lower than the SAPO-5 membrane, and the enhancement 7 peaks of the substrate suggested the discontinuity of the molecular sieve layer. 6 7
7
6
6
5
5
4
4
2
3
32
1
2
21
0
1
5 4
Aspect Ratio
Aspect Ratio
6
3
5
AlPO4-5
0
AlPO4-5
SAPO-5 SAPO-5
MnAPSO-5 MnAPSO-5
CrAPSO-5 CrAPSO-5
4 3
10 0
Figure 4. The average aspect ratios together with the corresponding size distributions for AFI Figure 4. The average aspect ratios together with the corresponding size distributions for AFI molecular Figure 4. crystals The average ratios heteroatoms. together withScale the corresponding sizetodistributions AFI molecular sieve withaspect or without bars correspond 20-μm SEMfor images. sieve crystals with or without heteroatoms. Scale bars correspond to 20-µm SEM images. molecular sieve crystals with or without heteroatoms. Scale bars correspond to 20-μm SEM images.
(a)
(b)
(b)
(c)
(002)
(c) Intensity(a.u.) Intensity(a.u.)
(a)
(002)
(100)
(100)
(210) (211)
(210) (211)
*
*
*
*
*
**
*
-5 membrane AlPOAlPO -5 membrane 4 4
α-Al2Oα3-Al
2
10 10
(d)
(e) (e)
30 30
40
O3
50
40
50
2θ 2(degree) θ (degree)
(f) (f) Intensity(a.u.) Intensity(a.u.)
(d)
20 20
(002)
(002)
(210) (211)
(100)
*
*
(210) (211)
(100)
*
*
*
*
*
MnAPSO-5 * membrane
MnAPSO-5 membrane α-Al2O3
10
10
20
30
2θ (degree) 20
30
2θ (degree)
40
50
α-Al2O3
40
50
Figure 5. Top view and cross-sectional SEM images together with the corresponding XRD patterns for Figure 5. Top view and cross-sectional SEM images together with the corresponding XRD patterns (a–c) AlPO MnAPSO-5 membranes induced by a layer of LMW chitosan using AIP as the 4 -5 and 4-5 and (d–f) MnAPSO-5 membranes induced by a layer of LMW chitosan using AIP as for (a–c) AlPO(d–f) Figure 5. Top view and cross-sectional SEM images together with the corresponding XRD patterns Al source. Scale bars correspond to 20 µm inμm SEM The The asterisks in the XRD the Al source. Scale bars correspond to 20 in images. SEM images. asterisks in the XRDpatterns patternsmark mark the 4-5 and (d–f) MnAPSO-5 membranes induced by a layer of LMW chitosan using AIP as for (a–c) AlPO peaks originating from the α-Al O substrate. 2 α-Al 3 2O3 substrate. the peaks originating from the the Al source. Scale bars correspond to 20 μm in SEM images. The asterisks in the XRD patterns mark 2O3 substrate. the peaks originating from the α-Al Several research groups worldwide have reported that the addition of silicon species into the
Several research groups worldwide have reported that the addition of silicon species into the reaction mixture decreases the growth rate of AFI molecular sieves along the c-axis, leading to reaction mixture decreases theworldwide growth ratehave of AFI molecular along theofc-axis, crystals Several research groups reported thatsieves the addition siliconleading speciestointo the with low aspect ratios [38–42]. The growth inhibition along the c-axis should be beneficial to increase reaction mixture decreases the growth rate of AFI molecular sieves along the c-axis, leading to the in-plane growth of the AFI membrane. As a typical example, the continuous SAPO-5 membranes
Molecules 2018, 23, 1127 Molecules 2018, 23, x FOR PEER REVIEW
6 of 14 6 of 14
Molecules 2018, x FOR PEERratios REVIEW 6 of to 14 crystals with23, low aspect [38–42]. The growth inhibition along the c-axis should be beneficial
could increase be fabricated via thegrowth in situofcrystallization method, the AlPO could not, until the in-plane the AFI membrane. As abut typical example, the continuous SAPO-5 4 -5 analogues crystals withcould low aspect ratios [38–42]. The inhibition along the c-axis should be4-5beneficial to now. On the other hand, thefabricated isomorphous substitution of framework atoms bythe transitional metals could membranes be via the in growth situ crystallization method, but AlPO analogues increase the in-plane growth of the AFI membrane. As a typical example, the continuous SAPO-5 could not, until now. On the other hand, the isomorphous substitution of framework atoms by provide redox active sites on molecular sieves for heterogeneous catalysis [43]. MnAPSO-5 crystals membranes metals could be fabricated the active in situsites crystallization method, butheterogeneous the AlPO4-5 analogues transitional could providevia redox on molecular sieves catalysiswith possessed almost the same morphology and size as SAPO-5 (Figure 4). Thefor MnAPSO-5 membrane couldMnAPSO-5 not, until crystals now. On the otheralmost hand,the thesame isomorphous substitution ofSAPO-5 framework atoms by [43]. possessed morphology and size as (Figure 4). The the thickness of 20 µm showed excellent continuity as well as a preferential c-orientation (Figure 5d–f). transitional metals couldwith provide active on molecular sieves forcontinuity heterogeneous catalysis MnAPSO-5 membrane the redox thickness of sites 20 μm showed excellent as well as a The incorporation of chromium could further reduce the length of hexagonal prisms along the [43]. MnAPSO-5 crystals possessed almost the same morphology and size as SAPO-5 (Figure 4). The preferential c-orientation (Figure 5d–f). c-axis MnAPSO-5 (Figure 4). membrane Although plate-like CrAPSO-5 could be synthesized withas an aspect ratio the thickness of crystals 20 reduce μm showed excellent continuity The incorporation ofwith chromium could further the length of hexagonal prisms well alongasthea of ca. 0.5, the c-orientation of (Figure the CrAPSO-5 membrane not prominent as an that of the SAPO-5 preferential c-orientation 5d–f). c-axis (Figure 4). Although plate-like CrAPSO-5 crystalswas could beas synthesized with aspect ratio of and MnAPSO-5 membranes, according to the enhancement of diffraction peaks of the (100), (210), The incorporation of chromium could further reduce the length of hexagonal prisms along theand ca. 0.5, the c-orientation of the CrAPSO-5 membrane was not as prominent as that of the SAPO-5 and c-axis (Figure 4). Although plate-liketoCrAPSO-5 crystals be synthesized aspect ratio of the (211) planes (Figure S1). This according phenomenon suggested that could thediffraction aspect ratio of crystals would notand be MnAPSO-5 membranes, the enhancement of peaks ofwith the an (100), (210), ca. 0.5, the c-orientation of the CrAPSO-5 membrane was not as prominent as that of the SAPO-5 and (211) to planes (Figure S1). This phenomenon that the aspect ratio of crystals would not be sole factor influence the c-orientation of AFIsuggested membranes. MnAPSO-5 membranes, to the enhancement of diffraction peaks of the (100), (210), and the sole factor to influenceaccording the c-orientation of AFI membranes. (211) planes (Figure S1). This phenomenon suggested that the aspect ratio of crystals would not be 2.3. Single-Layered AFI Membranes of Crystal Agglomerates the sole factor to influence the c-orientation of AFI membranes. 2.3. Single-Layered AFI Membranes of Crystal Agglomerates
The morphology of AFI crystal grains using AIP as the Al source was mainly hexagonal prisms Theown morphology of AFIresults crystal and grains using AIP as the (Figure Al source6a). wasWhen mainly hexagonal prismswas based on experimental literature reports pseudo-boehmite 2.3.our Single-Layered AFI Membranes of Crystal Agglomerates based on our own experimental results and literature reports (Figure 6a). When pseudo-boehmite used as the Al source, spherical crystal agglomerates would be synthesized as shown in Figure 6b. The morphology of AFI crystal grains AIP as the Al source was mainlyashexagonal was used as the Al source, spherical crystalusing agglomerates would be synthesized shown in prisms Figure These multi-crystal and dense agglomerates were formed through the combination of a great amount based on our own experimental and literature reports (Figure pseudo-boehmite 6b. These multi-crystal and denseresults agglomerates were formed through6a). the When combination of a great of flaky crystals, which be conducive to agglomerates the formation of continuous AFI as membranes. was used the crystals, Al might source, spherical be synthesized shown in Figure amount ofas flaky which mightcrystal be conducive to the would formation of continuous AFI membranes. 6b. These multi-crystal and dense agglomerates were formed through the combination of a great amount (a)of flaky crystals, which might be conducive to the formation of continuous AFI membranes.
(b)
(a)
(b)
SEM images of SAPO-5 crystal grains AIPand (a) pseudo-boehmite and pseudo-boehmite as the Al FigureFigure 6. SEM6.images of SAPO-5 crystal grains usingusing AIP (a) (b) (b) as the Al source. source. Scale bars correspond to 20 μm. Scale bars correspond to 20 µm. Figure 6. SEM images of SAPO-5 crystal grains using AIP (a) and pseudo-boehmite (b) as the Al
source.AIP Scalewas barsreplaced correspond 20 μm. When byto pseudo-boehmite as the Al source, while the other components and When AIP was conditions replaced bywere pseudo-boehmite the 15–20-μm Al source,thick whileand the other components crystallization kept the same,asthe continuous SAPO-5and When AIP was replaced by pseudo-boehmite as the be Althick source, while the otherThe components and membrane composed of intergrown could formed (Figure 7a–c). XRD patterns crystallization conditions were kept the agglomerates same, the 15–20-µm and continuous SAPO-5 membrane crystallization were kept the be same, the 15–20-μm thickwithout andXRD continuous SAPO-5 shown Figureconditions 7c exhibited a highly crystallized AFI-type framework anypatterns impurities. The in composed ofinintergrown agglomerates could formed (Figure 7a–c). The shown membrane composed of intergrown agglomerates could bewas formed (Figure XRD patterns the membrane before template removal checked by7a–c). usingThe theThe leaking test, Figurecompactness 7c exhibitedof a highly crystallized AFI-type framework without any impurities. compactness −9 mol −2 s−1 Pa −1 [44]. shownthe in Figure 7c exhibited a highly crystallized AFI-type framework without any impurities. The where permanence of N 2 was far below 10 m of the membrane before template removal was checked by using the leaking test, where the permanence compactness of the membrane before template removal was checked by using the leaking test, 2 s−1 Pa−1 [44]. of N2 was farthe below 10−9 mol m2 −was where permanence of N far below 10−9 mol m−2 s−1 Pa−1 [44]. (c) (b) (a)
(c)
(b)
Intensity(a.u.) Intensity(a.u.)
(a)
(100) (200) (110) (100) (200) (110)
(210) (211) (002)
*
* *
* (210) (211) (002)
SAPO-5 membrane
*
* *
*
SAPO-5 membrane α -Al2O3 10
20
30
2θ (degree)
40
50 α -Al2O3
10
Figure 7. Cont.
20
30
2θ (degree)
40
50
Molecules 2018, 23, 1127
7 of 14
Molecules 2018, 23, x FOR PEER REVIEW
(d)
7 of 14
(e)
(f)
(100) (210) (211) (002)
Intensity(a.u.)
(200) (110)
*
*
* *
CrAPSO-5 membrane
α -Al2O3
10
(g)
(h)
20
30
40
2θ (degree)
50
(i) (100) (211)
Intensity(a.u.)
(002) (210)
*
* *
*
(200) (110)
CoAPSO-5 membrane
α-Al2O3
10
20
30
2θ (degree)
40
50
Figure 7. Top view and cross-sectional SEM images together with the corresponding XRD patterns
Figure 7. Top view and cross-sectional SEM images together with the corresponding XRD patterns for for (a–c) SAPO-5, (d–f) CoAPSO-5, and (g–i) CrAPSO-5 membranes induced by a layer of LMW (a–c) SAPO-5, (d–f) CoAPSO-5, and (g–i) CrAPSO-5 membranes induced by a layer of LMW chitosan chitosan using pseudo-boehmite as the Al source. Scale bars correspond to 20 μm in SEM images. using pseudo-boehmite thepatterns Al source. Scale bars correspond to 20α-Al µm2O in3 substrate. SEM images. The asterisks The asterisks in theas XRD mark the peaks originating from in the XRD patterns mark the peaks originating from α-Al2 O3 substrate.
Based on a number of experimental investigations, AFI molecular sieves could be synthesized withinon a relatively range of pH using the ingredients this study. The variation would Based a numberwide of experimental investigations, AFIinmolecular sieves could of bepH synthesized cause the morphology change of AFI crystal grains on the one hand. On the other hand, the within a relatively wide range of pH using the ingredients in this study. The variation of pH would dissolution rate of the chitosan layer should be varied under different pH values, resulting in the cause the morphology change of AFI crystal grains on the one hand. On the other hand, the dissolution different microstructure of the synthesized membrane (Figure S2). At pH = 4.5, the agglomerated rate ofSAPO-5 the chitosan layer should be varied under different pH values, resulting in the different crystals were packed loosely on the substrate surface due to a faster dissolution of the microstructure of the synthesized membrane S2). At pH = 4.5, the agglomerated SAPO-5 chitosan layer in the stronger acidic solution,(Figure where the structure-directing function of the chitosan crystals were packed loosely on the substrate surface due to a faster dissolution of the chitosan layer should be very weak (Figure S2a,b). When the pH of the synthesis solution was changed tolayer 6.0, in the stronger acidic solution, where the structure-directing function of the chitosan layer should be very the intergrowth of AFI crystals in the as-synthesized SAPO-5 membrane was significantly improved (Figure S2c,d). the pH the was pH further changed to 7.0,solution a continuous of crystal agglomerates could of weak (Figure S2a,b).If When of the synthesis waslayer changed to 6.0, the intergrowth not be in formed on the substrateSAPO-5 surface. Some chitosan residuals could be observed (Figure on the membrane AFI crystals the as-synthesized membrane was significantly improved S2c,d). If the surface (Figure S2e). It should be noted that the dissolution rate of chitosan was low in the neutral pH was further changed to 7.0, a continuous layer of crystal agglomerates could not be formed on the synthesis solution compared with the above two situations. Actually, this phenomenon acted as substrate surface. Some chitosan residuals could be observed on the membrane surface (Figure S2e). another validation for the relationship between the dissolution rate and the microstructure of AFI It should be noted that the dissolution rate of chitosan was low in the neutral synthesis solution membranes. compared As withaforementioned, the above two asituations. this metal phenomenon acted another validation number ofActually, transitional ions could be as incorporated into the for the relationship between the dissolution rate and the microstructure of AFI membranes. framework of AFI molecular sieves for various potential applications [43,45]. Several metallic As aforementioned, number ofMn transitional metal could be incorporated into the framework elements such as Cr,aCo, Fe, and were added into ions the reaction systems to synthesize MeAPSO-5 of AFImembranes molecular sieves various potential applications [43,45]. elements such as Cr, in this for study. As shown in Figures 7d–i and S3, theSeveral effect ofmetallic incorporated transitional onwere the morphology of crystal agglomerates was very tiny, and single-layered MeAPSO-5 Co, Fe,metals and Mn added into the reaction systems to synthesize MeAPSO-5 membranes in this study. membranes with7d–i satisfactory could be obtained under the structure-directing of As shown in Figures and S3,denseness the effect of incorporated transitional metals on thefunction morphology a LMW chitosan layer. of crystal agglomerates was very tiny, and single-layered MeAPSO-5 membranes with satisfactory Diffuse Reflectance UV-vis spectroscopy could provide an evidence of the incorporation of denseness could be obtained under the structure-directing function of a LMW chitosan layer. cobalt and chromium into the AFI framework. The triplet bands located in the 530–650-nm region Diffuse Reflectance UV-vis spectroscopy could provide an evidence of the incorporation of cobalt were present only in the CoAPSO-5 membrane (Figure 8, line B), which was attributed to the d–d and chromium the AFI framework. triplet bandsThere located thea 530–650-nm regionthe were transitions into of tetrahedrally coordinatedThe Co2+ [28,46,47]. wasinstill controversy about presentattribution only in the membrane (Figure 8, line B), which attributed to theappearing d–d transitions of CoAPSO-5 absorption peaks to the framework substitution of was chromium, the bands at 2+ [28,46,47]. There was still a controversy about the attribution of of tetrahedrally coordinated Co 440 nm and 620 nm for the as-synthesized CrAPSO-5 membrane were generally recognized as the absorption to thein framework substitution of chromium, the bands 440Cr(III) nm and featurepeaks of Cr(III) distorted octahedral coordination (Figure 8, line appearing C), where atthe heteroatoms bonded tetrahedrally to the framework complementarily coordinated H2O of 620 nm for the as-synthesized CrAPSO-5 membranewere were generally recognized as by thetwo feature Cr(III) in distorted octahedral coordination (Figure 8, line C), where the Cr(III) heteroatoms bonded tetrahedrally to the framework were complementarily coordinated by two H2 O molecules as additional
Molecules 2018, 23, x FOR PEER REVIEW Molecules 2018, 23, 1127
8 of 14 8 of 14
molecules as additional ligands [48–50]. In addition, the blue and green color of the synthesized 2+ and Cr3+ had been incorporated into the AFI molecular sieves also indicated ligands [48–50]. In addition, the bluethat and both greenCo color of the synthesized molecular sieves also indicated 2+ 3+ Molecules 2018, 23, x FOR PEER REVIEW 8 of 14 framework, that both Co respectively. and Cr had been incorporated into the AFI framework, respectively.
C
Absorbance (a.u.)
Absorbance (a.u.)
molecules as additional ligands [48–50]. In addition, the blue and green color of the synthesized molecular sieves also indicated that both Co2+ and Cr3+ had been incorporated into the AFI framework, respectively.
CB
BA
300
400
500
600
700
Wavelength (nm)
A
Figure 8. Diffuse-reflectance UV–vis spectra spectra of of500 (A) SAPO-5, (B) CoAPSO-5, CoAPSO-5, and (C) CrAPSO-5 600 (B) 700 Figure 8. Diffuse-reflectance300UV–vis 400 (A) SAPO-5, and (C) CrAPSO-5 Wavelength (nm) membranes corresponding to Figure 7. membranes corresponding to Figure 7. Figure 8. Diffuse-reflectance UV–vis spectra of (A) SAPO-5, (B) CoAPSO-5, and (C) CrAPSO-5
The membranes contents of P, Al, Si, and metals along the cross-section of an AFI membrane were Figure 7. The contents ofcorresponding P, Al, Si, andtometals along the cross-section of an AFI membrane were measured measured by using the electron probe microanalysis (EPMA) technique (Figure S4). In accordance by using the electron probe microanalysis (EPMA) technique (Figure S4). In accordance with the The contentsinofthe P, synthesis Al, Si, and metals the along the cross-section of an AFI were Si with the composition solution, concentration of either P or Al membrane was high, while composition in the synthesis solution, the concentration of either P or Al was high, while Si and metal measured byMn) using thesparsely electron distributed probe microanalysis (EPMA) technique (Figure S4). In accordance and metal (Fe or was within the membrane. (Fe orwith Mn)the was sparsely distributed within the membrane. composition in the synthesis solution, the concentration of either P or Al was high, while Si and metal (Fe or Mn) was sparsely distributed within the membrane. 2.4. Double-Layered AFI Membranes of Crystal Agglomerates 2.4. Double-Layered AFI Membranes of Crystal Agglomerates When a thin layer of MMW chitosan was covered on the substrate surface while 2.4. Double-Layered AFI Membranes of Crystal Agglomerates When a thin layer of MMW chitosan was covered on the substrate surface while pseudo-boehmite pseudo-boehmite was used as the Al source, the double-layered MeAPSO-5 membranes of crystal was used When as the Al the double-layered MeAPSO-5 membranes of crystal agglomerates were a source, thin layer of MMW chitosan was covered on the substrate surface while agglomerates were obtained (Figures 9 and S5). No matter whether Fe (Figure 9), Mn (Figure S5a,b), pseudo-boehmite was the Alwhether source, Fe the(Figure double-layered MeAPSO-5 membranes of crystal obtained (Figures 9 and S5).used No as matter 9), Mn (Figure S5a,b), or Co (Figure S5c,d) or Coagglomerates (Figure S5c,d) was incorporated, 40–50 μm-thick bottom layer consisted of(Figure looselyS5a,b), packed obtained (Figures 9aand S5).consisted No matterof whether (Figure 9), Mn was incorporated, awere 40–50 µm-thick bottom layer loosely Fe packed spherical particles, while spherical while theincorporated, top layer was 20–50 μm-thick andlayer consisted of ofa loosely well-intergrown orlayer Coparticles, (Figure S5c,d) was a 40–50 bottom consisted packed the top was 20–50 µm-thick and consisted of a μm-thick well-intergrown membrane of crystal agglomerates. membrane of crystal agglomerates. The dense MeAPSO-5 top layer was different the c-oriented spherical particles, while top layer was 20–50 μm-thick and of of afrom well-intergrown The dense MeAPSO-5 top layerthe was different from the c-oriented AFIconsisted membrane hexagonal prisms AFI membrane of hexagonal prisms that was synthesized by using AIP as the Al source. The XRD membrane of crystal agglomerates. The dense MeAPSO-5 top layer was different from the c-oriented that was synthesized by using AIP as the Al source. The XRD patterns indicated that the membrane patterns the membrane of AFIbycrystals with crystallinity AFI indicated membranethat of hexagonal prismswas that composed was synthesized using AIP ashigh the Al source. Thewithout XRD was composed of AFI crystals with high crystallinity without any impurities. The attenuation of patterns indicated that the membrane was composed of AFI crystalssuggested with high crystallinity any impurities. The attenuation of diffraction peaks of the substrate the α-Al2O3without was fully diffraction peaks of the substrate suggested the α-Al2 O3 was fully covered by the double-layered AFI anyby impurities. The attenuation diffractionof peaks of the substrate suggested the α-Al2O3 was fully covered the double-layered AFIofmembrane crystal agglomerates. membrane of by crystal agglomerates.AFI membrane of crystal agglomerates. covered the double-layered
(a)
(b)
(b)
(c)
(c)
(002) (211)
(100)
(210) (002) (210) (211)
(100)
Intensity(a.u.) Intensity(a.u.)
(a)
(200) (200) (110) (110)
*
* *
*
* *
*
*
FeAPSO-5 membrane
FeAPSO-5 membrane
O α -Al2α O-Al 2 3 3 10 10
2020
30 30
θ (degree) 2θ 2(degree)
40
40
50
Figure 9.Top (a) Top view and (b) cross-sectional SEM imagestogether together with (c) XRD Figure 9. (a) and (b)(b) cross-sectional images with (c)the the corresponding XRD Figure 9. (a) Topview view and cross-sectional SEM SEM images together with (c)corresponding the corresponding pattern of the double-layer FeAPSO-5 membrane induced by a layer of MMW chitosan using pattern of the double-layer FeAPSO-5 membrane induced by a layer of MMW chitosan using XRD pattern of the double-layer FeAPSO-5 membrane induced by a layer of MMW chitosan using pseudo-boehmite as the Al source. Scale barscorrespond correspond to 20 20 μm. μm. The ininXRD patterns pseudo-boehmite the source. Scale bars Theasterisks asterisks XRD patterns pseudo-boehmite asasthe AlAl source. Scale bars correspond to 20toµm. The asterisks in XRD patterns mark mark the peaks originating from the α-Al2O3 substrate. mark the peaks originating from the α-Al 2O3 substrate. the peaks originating from the α-Al O substrate. 2 3
50
Molecules 2018, 23, 1127
9 of 14
Molecules2018, 2018,23, 23,xxFOR FORPEER PEERREVIEW REVIEW Molecules
of14 14 99of
Obviously, when theAl Alsource source was changed from pseudo-boehmite to and AIP, and the Obviously,when whenthe the Al source was changed from pseudo-boehmite backback toAIP, AIP, andthe the other Obviously, was changed from pseudo-boehmite back to other other components were kept the same as that of Zhang et al. [33], the synthesized product would componentswere werekept keptthe thesame sameas asthat thatof ofZhang Zhanget etal. al.[33], [33],the thesynthesized synthesizedproduct productwould wouldbe bethe the components be the double-layered AFI membrane as shown in Figure 1, where the bottom layer was made double-layered AFI membrane as shown in Figure 1, where the bottom layer was made up of loosely double-layered AFI membrane as shown in Figure 1, where the bottom layer was made up of loosely up of loosely packed spherical particles, andwas the composed top layer was composed of well-intergrown packed spherical particles, and the the top top layer layer was composed of well-intergrown well-intergrown and c-oriented c-oriented packed spherical particles, and of and and c-oriented hexagonal prisms. Thus, a close circuit should be finally formed by changing the hexagonalprisms. prisms.Thus, Thus,aaclose closecircuit circuitshould shouldbe befinally finallyformed formedby bychanging changing thestructure-directing structure-directing hexagonal the structure-directing matrix (a thin layer ofdifferent chitosan with different molecular and the (AIP Al source matrix (a (a thin thin layer layer of chitosan chitosan with different molecular weight) and weight) the Al Al source source (AIP and matrix of with molecular weight) and the and (AIP and pseudo-boehmite) to achieve the microstructure manipulation of AFI membranes with high pseudo-boehmite) to to achieve achieve the the microstructure microstructure manipulation manipulation of of AFI AFI membranes membranes with with high high pseudo-boehmite) reproducibility (Figure 10). reproducibility (Figure 10). reproducibility (Figure 10).
Figure 10. 10. Schematic Schematic diagram diagram of of manipulating manipulating microstructure microstructure of of AFI AFI membranes membranes using using aa thin thin Figure Figure 10. Schematic diagram of manipulating microstructure of AFI membranes using a thin chitosan chitosan layer of different molecular weight and different Al sources. chitosan layer of different molecular weight and different Al sources. layer of different molecular weight and different Al sources.
FormationMechanism Mechanism 3.3.Formation 3. Formation Mechanism randomly oriented oriented and and compact compact SAPO-5 SAPO-5 membrane membrane could could be be fabricated fabricated on on an an unmodified unmodified AA randomly randomly andAIP compact membrane could be fabricated on unmodified α-Al2A 2O substrateoriented by using using as the theSAPO-5 Al source source under microwave microwave heating [6]. [6].an However, the α-Al O 33 substrate by AIP as Al under heating However, the α-Al by film usingofAIP as the Al source under microwave α-Al heating [6]. However, the 2 O3 substrate synthesized SAPO-5 hexagonal prisms on the unmodified 2 O 3 substrate was both synthesized SAPO-5 film of hexagonal prisms on the unmodified α-Al2O3 substrate was both synthesized SAPO-5 film of hexagonal prisms on the unmodified α-Al2(Figure O3 substrate bothwas randomly randomly oriented oriented and non-continuous non-continuous using traditional heating 11a),was which in full full randomly and using traditional heating (Figure 11a), which was in oriented and non-continuous using traditional heating (Figure 11a), which was in full agreement agreement with with our our previous previous report report [33]. [33]. Voids Voids or or defects defects between between hexagonal hexagonal crystals crystals spread spread all all agreement with our previous report [33]. Voids or defects between hexagonal crystals spread allonly oversome the over the substrate surface (circled areas). By using pseudo-boehmite as the Al source, over the substrate surface (circled areas). By using pseudo-boehmite as the Al source, only some substrate surface (circled areas). By using pseudo-boehmite the2O Al substrate source, only biggercrystal crystal agglomerates weredispersed dispersed onthe theunmodified unmodifiedas α-Al withsome muchbigger larger bigger agglomerates were on α-Al 2O33substrate with much larger crystal agglomerates were dispersed on the unmodified α-Al O substrate with much larger voids and 2 Control 3 voids and open areas under traditional heating (Figure 11b). experiments using bare α-Al 2O voids and open areas under traditional heating (Figure 11b). Control experiments using bare α-Al2O 33 open areas under traditional heating (Figure 11b). Control experiments using bare α-Al O substrates 2 3 substrates as as indicated indicated above above confirmed confirmed that that the the chitosan chitosan layer layer should should be be indispensable indispensable for for the the substrates as indicatedand above that of theAFI chitosan layer should be indispensable for the continuity continuity theconfirmed c-orientation membranes on porous porous α-Al2O 2O substrates in the the in in and situ continuity and the c-orientation of AFI membranes on α-Al 33 substrates in situ the c-orientation of AFI membranes on porous α-Al2 O3 substrates in the in situ crystallization. crystallization. crystallization.
(a) (a)
(b) (b)
Figure11. 11.SEM SEMimages imagesofofSAPO-5 SAPO-5layers layerssynthesized synthesizedon onunmodified unmodifiedα-Al α-Al22O 2O substratesusing usingAIP AIP substrates using Figure SAPO-5 layers synthesized on unmodified α-Al 33 3substrates (a)and andpseudo-boehmite pseudo-boehmite(b) (b)as asthe theAl Alsource. source.Scale Scalebars barscorrespond correspondtoto20 20µm. μm. (a) μm.
The research research on on the the formation formation mechanism mechanism of of supported supported zeolite zeolite films films and and membranes membranes dates dates The back to to the the late late 1990s 1990s [51,52]. [51,52]. According According to to available available literature literature reports, reports, there there are are two two formation formation back
Molecules 2018, 23, 1127
10 of 14
The research on the formation mechanism of supported zeolite films and membranes dates back to the late 1990s [51,52]. According to available literature reports, there are two formation mechanisms named the heterogeneous nucleation model (HTN model) and the homogeneous nucleation model (HMN model). In the presence of a chitosan layer, a gel layer is initially formed on the supported chitosan surface in the in situ crystallization. Both nucleation and crystal growth occur inside the gel layer and in the bulk synthesis solution as well. Whether the HTN or HMN model applies to the film formation is directly related to the ratio of the two growth rates [52,53]. Chitosan is insoluble in water and aqueous basic media, but it becomes soluble in aqueous acidic media at pH < 6.5 [54]. Since MMW chitosan heavily swelled with a low erodible rate [38], the entire chitosan layer functioned as a three-dimensional (3-D) structure-directing matrix in the in situ crystallization, leading to a five-step chronological evolution of double-layer AFI membranes, as given in our previous report (Figure S6) [33]. The chitosan layer swelled with dissolution (or was being eroded) at first (A2). In the following, the crystal nucleation occurred in the bulk synthesis solution, in the eroded pores, and on the outer surface of the chitosan layer (A3). The crystals on the top layer at the outer surface and the packed crystals underneath would grow simultaneously (A4 and A5). When the top layer became well intergrown, it would act as a barrier between the precursor solution and the packed crystals, eventually leading to the double-layer AFI membranes, as shown in Figure S6 (A6). Similar to the above, the precoated thin layer of LMW chitosan disappeared after the hydrothermal reaction. Based on the literature report [36] and our own experiment, a layer of LMW chitosan had an increased dissolution rate, but its swelling extent was reduced compared with MMW chitosan. When LMW chitosan was applied, the chitosan layer dissolved to the bulk solution faster, and the precursor solution would contact the chitosan layer all of the time, leading to the formation of the single-layer AFI membranes at last. Therefore, the chronological evolution of singer-layer AFI membranes over porous α-Al2 O3 substrates under the structure-directing effect of the LMW chitosan layer could be generally defined in five steps along the elapsed time of hydrothermal reaction as follows (Figure S7): (1) the onset of chitosan swelling and its dissolution in an increased rate, where the chitosan layer was getting thinner from B1 to B2; (2) the crystal nucleation in the bulk precursor solution, in the eroded pores as well as on the outer surface of the chitosan layer, accompanied by an ever-reducing layer thickness (B3); (3) the growth of nucleus in the eroded pores as well as on the surface of the chitosan layer (B4); (4) the growth of small crystals to be intergrown on the substrate surface due to the disappearance of the chitosan layer (B5); and (5) the oriented growth of the AFI membrane after it has been well intergrown (B6). 4. Materials and Methods The synthesis procedure of AFI membranes on porous α-Al2 O3 substrates is similar to that reported in our previous work [33], which usually included (1) the treatment of substrates; (2) the preparation of chitosan layers; (3) the synthesis of precursor gels; and (4) the hydrothermal crystallization of AFI membranes and sample cleaning. The homemade porous α-Al2 O3 substrates with a thickness of 2 mm and a diameter of 20 mm were polished with sandpaper and sequentially treated with 1 M HCl and 1 M NaOH solution, followed by ultrasonic treatment in acetone and dried in clean atmosphere overnight. The substrates were spin-coated with 2 wt.% chitosan solution, which was made by dissolving 75–85% deacetylated chitosan with medium molecular weight (MMW chitosan, 150–240 kDa, Aldrich, St. Louis, MO, USA) or low molecular weight (LMW chitosan, 50–190 kDa, Aldrich) in a ca. 2 wt.% acetic acid solution. The coating was performed at a spinning rate of 1600 rpm for 60 s using a spin-coater (KW-4A, Microelectronic Institute of Chinese Academy of Sciences, Beijing, China). After being coated three times, the α-Al2 O3 supported chitosan layer was dried at room temperature overnight. Before use, it was refluxed in ethanol for 30 min to remove possible residues, and dried at 393 K for 30 min.
Molecules 2018, 23, 1127
11 of 14
The precursor sol was prepared as follows. Aluminum isopropoxide (AIP, 99.5%, Aldrich) was hydrolyzed in deionized water at 353 K using a water bath heating under stirring for 12 h. After cooling down naturally, orthophosphoric acid (H3 PO4 , 85%) was added into the synthesized alumina sol dropwise under stirring at 298 K. Then, the solution was stirred for 10 h to form a white uniform gelation. The aqueous solution of metal salts was added accordingly if necessary. Triethylamine (TEA, 99%, Tianjin Chemical Reagent Co., Inc., Tianjin, China) and tetraethylorthosilicate (TEOS, 98%, Aldrich) were added sequentially under vigorous stirring, followed by aging for 12 h at room temperature. The final molar ratio of the synthesis solution was Al2 O3 :P2 O5 :0.15SiO2 :2.4TEA:(100–300)H2 O(:0.02 MeOx ). At last, the pH of the sol was adjusted to ca. 6 using acetic acid (99.5%, Tianjin Chemical Reagent Co., Inc.). When pseudo-boehmite was used as the Al source, the hydrolyzation step could be omitted. Pseudo-boehmite was added into the H3 PO4 solution, stirring for 12 h. The steps afterwards were nearly the same as the above except that the pH of the solution was adjusted to 4.0–8.0. The precursor solution (ca. 35 mL) was loaded into a 60-mL Teflon-lined stainless steel autoclave, where the chitosan-modified substrate was placed in the autoclave with the treated side facing down to the bottom. The hydrothermal treatment was performed in the sealed autoclave at 453 K for 72 h. After the autoclave was cooled down, the sample was removed and washed with DI water, and dried at 333 K overnight. The crystal grains at the bottom of the autoclave were collected through centrifugation and washing repeatedly. The compactness of each membrane was tested by gas permeation using a leaking-test setup equipped with a soap film gas flowmeter and a gas chromatograph (GC, Agilent 6890N) [44]. At the feed side, nitrogen was introduced under 0.4 MPa and 298 K. Helium was used as the sweeping gas with a flow rate of ca. 10 mL/min. Scanning electron microscopy (SEM) for membranes and particles was performed using a FEI Nanosem 430 field emission scanning electron microscopy (FE-SEM) at an acceleration voltage of 10~20 kV (Hillsboro, OR, USA). X-ray diffraction (XRD) data were obtained on a Rigaku D-max2500v/pc X-ray diffractometer using Cu Kα radiation in the range of 2θ = 5~50◦ at a scanning rate of 4◦ /min (Tokyo, Japan). UV-vis spectra were recorded by using a Jasco V-550 UV-vis spectrophotometer equipped with an integrated sphere (Easton, MA, USA). The composition of Si, Al, P, and metal along the membrane thickness was measured using electron probe microanalysis (EPMA-1600, Shimadzu, Kyoto, Japan) operated at 15 kV with a secondary electron resolution of 6 nm. 5. Conclusions We have reported a novel strategy for manipulating continuous and well-defined AFI membranes on macroporous α-Al2 O3 substrates in in situ crystallization. Based on our own results and literature data, it revealed that the use of a low molecular weight (LMW) chitosan layer as the structure-directing matrix resulted in the formation of single-layer AFI membranes, while double-layer AFI membranes were formed when a medium molecular weight (MMW) chitosan layer was employed, due to their difference in swelling ability and erodible rate in the acid precursor solution. The AFI membrane was composed of hexagonal prism-shaped crystals when aluminum isopropoxide (AIP) was used as the Al source, whereas spherical agglomerates of small crystals would be the membrane components if AIP is replaced by pseudo-boehmite. The coverage, denseness, and orientation of all of the AFI membranes could be further optimized by adjusting the water content in the precursor solution by fixing both the substrate setting (facing downwards) and the ratio of SiO2 /Al2 O3 (0.15). Besides, the addition of Si or metallic elements into the AFI framework would reduce the aspect ratio of hexagonal prism-shaped AFI crystals, usually leading to the formation of thinner c-oriented AFI membranes. Control experiments confirmed that the chitosan layer should be indispensable for the continuity and orientation of AFI membranes on porous α-Al2 O3 substrates in the in situ crystallization. The single-layer AFI membrane could be induced by using a thin layer of LMW chitosan due to its smaller swelling area and faster dissolution rate, whereas the double-layer AFI membrane would be fabricated
Molecules 2018, 23, 1127
12 of 14
under the structure-directing function of a MMW chitosan layer with a larger swelling area and a lower dissolution rate. The established methodology in this study has provided the platform to fabricate continuous and well-defined AFI membranes with single and double-layers of hexagonal prisms or crystal agglomerates, which will have application potential in the separation of gas/liquid mixtures, membrane catalysis, and microelectronic devices. Supplementary Materials: The supporting Information (Figures S1~S7) are available online. Author Contributions: Conceptualization, X.L. and B.Z.; Methodology, H.D.; Investigation, H.D. and L.G.; Data Curation, H.D.; Writing-Original Draft Preparation, H.D. and L.G.; Writing-Review & Editing, B.Q. and X.L.; Supervision, B.Z.; Project Administration, B.Z.; Funding Acquisition, X.L. Funding: This work was supported by the National Natural Science Foundation of China (Grant No. 21476171). Conflicts of Interest: The authors declare no conflict of interest.
References 1. 2. 3.
4. 5. 6. 7. 8. 9. 10. 11.
12. 13.
14. 15. 16. 17.
Cundy, C.S.; Cox, P.A. The Hydrothermal Synthesis of Zeolites: History and Development from the Earliest Days to the Present Time. Chem. Rev. 2003, 103, 663–702. [CrossRef] [PubMed] Wilson, S.T.; Lok, B.M.; Messina, C.A.; Cannan, T.R.; Flanigen, E.M. Aluminophosphate Molecular Sieves: A New Class of Microporous Crystalline Inorganic Solids. J. Am. Chem. Soc. 1982, 104, 1146–1147. [CrossRef] Lok, B.M.; Messina, C.A.; Patton, R.L.; Gajek, R.T.; Cannan, T.R.; Flanigen, E.M. Silicoaluminophosphate Molecular Sieves: Another New Class of Microporous Crystalline Inorganic Solids. J. Am. Chem. Soc. 1984, 106, 6092–6093. [CrossRef] Flanigen, E.M.; Lok, B.M.; Patton, R.L.; Wilson, S.T. Aluminophosphate Molecular Sieves and the Periodic Table. Pure Appl. Chem. 1986, 58, 1351–1358. [CrossRef] Davis, M.E. Ordered Porous Materials for Emerging Applications. Nature 2002, 417, 813–821. [CrossRef] [PubMed] Yang, W.S.; Zhang, B.Q.; Liu, X.F. Synthesis and Characterization of SAPO-5 Membranes on Porous α-Al2 O3 Substrates. Microporous Mesoporous Mater. 2009, 117, 391–394. [CrossRef] Jiang, H.Y.; Zhang, B.Q.; Lin, Y.S.; Li, Y.D. Synthesis of Zeolite Membranes. Chin. Sci. Bull. 2004, 49, 2547–2554. [CrossRef] Caro, J.; Noack, M. Zeolite Membranes-Recent Developments and Progress. Microporous Mesoporous Mater. 2008, 115, 215–233. [CrossRef] Yu, M.; Noble, R.D.; Falconer, J.L. Zeolite Membranes: Microstructure Characterization and Permeation Mechanisms. Acc. Chem. Res. 2011, 44, 1196–1206. [CrossRef] [PubMed] Pera-Titus, M. Porous Inorganic Membranes for CO2 Capture: Present and Prospects. Chem. Rev. 2014, 114, 1413–1492. [CrossRef] [PubMed] Kim, S.J.; Yang, S.W.; Reddy, G.K.; Smirniotis, P.; Dong, J.H. Zeolite Membrane Reactor for High-Temperature Water-Gas Shift Reaction: Effects of Membrane Properties and Operating Conditions. Energy Fuels 2013, 27, 4471–4480. [CrossRef] Lew, C.M.; Cai, R.; Yan, Y.S. Zeolite Thin Films: From Computer Chips to Space Stations. Acc. Chem. Res. 2010, 43, 210–219. [CrossRef] [PubMed] Gascon, J.; Kapteijn, F.; Zornoza, B.; Sebastián, V.; Casado, C.; Coronas, J. Practical Approach to Zeolitic Membranes and Coatings: State of the Art, Opportunities, Barriers, and Future Perspectives. Chem. Mater. 2012, 24, 2829–2844. [CrossRef] Pina, M.P.; Mallada, R.; Arruebo, M.; Urbiztondo, M.; Navascués, N.; de la Iglesia, O.; Santamaria, J. Zeolite Films and Membranes. Emerging Applications. Microporous Mesoporous Mater. 2011, 144, 19–27. [CrossRef] Caro, J.; Finger, G.; Kornatowski, J.; Richter-Mendau, J.; Werner, L.; Zibrowius, B. Aligned Molecular Sieve Crystals. Adv. Mater. 1992, 4, 273–276. [CrossRef] Noack, M.; Kölsch, P.; Venzke, D.; Toussaint, P.; Caro, J. New One-dimensional Membrane: Aligned AlPO4 -5 Molecular Sieve Crystals in a Nickel Foil. Microporous Mater. 1994, 3, 201–206. [CrossRef] Girnus, I.; Pohl, M.; Richter-Mendau, J.; Schneider, M.; Noack, M.; Venzke, D.; Caro, J. Synthesis of AlPO4 -5 Aluminumphosphate Molecular Sieve Crystals for Membrane Applications by Microwave Heating. Adv. Mater. 1995, 7, 711–714. [CrossRef]
Molecules 2018, 23, 1127
18. 19. 20. 21.
22. 23. 24. 25. 26. 27. 28. 29.
30.
31.
32. 33. 34.
35. 36.
37. 38. 39. 40.
13 of 14
Lin, J.C.; Yates, M.Z.; Petkoska, A.T.; Jacobs, S. Electric-Field-Driven Assembly of Oriented Molecular-Sieve Films. Adv. Mater. 2004, 16, 1944–1948. [CrossRef] Lin, J.C.; Yates, M.Z. Oriented Molecular Sieve Thin Films from Aligned Seed Layers. Chem. Mater. 2006, 18, 4137–4141. [CrossRef] Wu, C.N.; Chao, K.J.; Tsai, T.G.; Chiou, Y.H.; Shih, H.C. Oriented Growth of Molecular Sieves on Inorganic Membranes. Adv. Mater. 1996, 8, 1008–1012. [CrossRef] Tsai, T.G.; Chao, K.J.; Guo, X.J.; Sung, S.L.; Wu, C.N.; Wang, Y.L.; Shih, H.C. Aligned Aluminophosphate Molecular Sieves Crystallized on Floating Anodized Alumina by Hydrothermal Microwave Heating. Adv. Mater. 1997, 9, 1154–1157. [CrossRef] Tsai, T.G.; Shih, H.C.; Liao, S.J.; Chao, K.J. Well-aligned SAPO-5 Membrane: Preparation and Characterization. Microporous Mesoporous Mater. 1998, 22, 333–341. [CrossRef] Xu, R.; Zhu, G.S.; Yin, X.J.; Wan, X.; Qiu, S.L. Epitaxial Growth of Highly-Oriented AlPO4 -5 Molecular Sieve Films for Microlaser Systems. J. Mater. Chem. 2006, 16, 2200–2204. [CrossRef] Hu, E.P.; Huang, Y.; Yan, Q.; Liu, D.P.; Lai, Z.P. Synthesis of Highly c-Oriented AFI Membranes by Epitaxial Growth. Microporous Mesoporous Mater. 2009, 126, 81–86. [CrossRef] Xue, M.; Fan, L.L.; Kang, Z.X.; Zhang, W.T.; Li, H.; Qiu, S.L. Dye-Modified Nanochannels of c-Oriented AFI Film for Heavy Metal Ion Sensing. J. Mater. Chem. 2012, 22, 17644–17648. [CrossRef] Ji, C.; Tian, Y.; Li, Y.; Lin, Y.S. Thin Oriented AFI Zeolite Membranes for Molecular Sieving Separation. Microporous Mesoporous Mater. 2014, 186, 80–83. [CrossRef] Karanikolos, G.N.; Wydra, J.W.; Stoeger, J.A.; Garcia, H.; Corma, A.; Tsapatsis, M. Continuous c-Oriented AlPO4 -5 Films by Tertiary Growth. Chem. Mater. 2007, 19, 792–797. [CrossRef] Karanikolos, G.N.; Garcia, H.; Corma, A.; Tsapatsis, M. Growth of AlPO4 -5 and CoAPO-5 Films from Amorphous Seeds. Microporous Mesoporous Mater. 2008, 115, 11–22. [CrossRef] Veziri, C.M.; Palomino, M.; Karanikolos, G.N.; Corma, A.; Kanellopoulos, N.K.; Tsapatsis, M. Toward Submicrometer c-Oriented Nanoporous Films with Unidimensional Pore Network: AFI Film Morphology Control by Precursor Mixture Manipulation. Chem. Mater. 2010, 22, 1492–1502. [CrossRef] Stoeger, J.A.; Veziri, C.M.; Palomino, M.; Corma, A.; Kanellopoulos, N.K.; Tsapatsis, M.; Karanikolos, G.N. On Stability and Performance of Highly c-Oriented Columnar AlPO4 -5 and CoAPO-5 Membranes. Microporous Mesoporous Mater. 2012, 147, 286–294. [CrossRef] Stoeger, J.A.; Palomino, M.; Agrawal, K.V.; Zhang, X.; Karanikolos, G.N.; Valencia, S.; Corma, A.; Tsapatsis, M. Oriented CoSAPO-5 Membranes by Microwave-Enhanced Growth on TiO2 -Coated Porous Alumina. Angew. Chem. Int. Ed. 2012, 51, 2470–2473. [CrossRef] [PubMed] Feng, S.; Bein, T. Vertical Aluminophosphate Molecular Sieve Crystals Grown at Inorganic-Organic Interfaces. Science 1994, 265, 1839–1841. [CrossRef] [PubMed] Yang, W.S.; Liu, X.F.; Zhang, L.; Zhang, B.Q. In Situ Synthesis and Microstructure Manipulation of SAPO-5 Films over Porous α-Al2 O3 Substrates. Langmuir 2009, 25, 2271–2277. [CrossRef] [PubMed] Wang, C.Z.; Liu, X.F.; Li, J.; Zhang, B.Q. Microwave-Assisted Seeded Growth of the Submicrometer-Thick and Pure b-Oriented MFI zeolite Films Using an Ultra-Dilute Synthesis Solution. CrystEngComm 2013, 15, 6301–6304. [CrossRef] Zhou, M.; Zhang, B.Q.; Liu, X.F. Oriented Growth and Assembly of Zeolite Crystals on Substrates. Chin. Sci. Bull. 2008, 53, 801–816. [CrossRef] Huanbutta, K.; Cheewatanakornkool, K.; Terada, K.; Nunthanid, J.; Sriamornsak, P. Impact of Salt Form and Molecular Weight of Chitosan on Swelling and Drug Release from Chitosan Matrix Tablets. Carbohydr. Polym. 2013, 97, 26–33. [CrossRef] [PubMed] Finger, G.; Richter-Mendau, J.; Bülow, M.; Kornatowski, J. On Synthesis Conditions for Tailoring AlPO4 -5 Crystal Dimensions. Zeolites 1991, 11, 443–448. [CrossRef] Demuth, D.; Stucky, G.D.; Unger, K.K.; Schüth, F. Synthesis of Large Optically Clear Silicoaluminophosphate Crystals with AFI Structure. Microporous Mater. 1995, 3, 473–487. [CrossRef] Schunk, S.A.; Demuth, D.G.; Schulz-Dobrick, B.; Unger, K.K.; Schüth, F. Element Distribution and Growth Mechanism of Large SAPO-5 Crystals. Microporous Mater. 1996, 6, 273–285. [CrossRef] Iwasaki, A.; Sano, T.; Kodaira, T.; Kiyozumi, Y. Growth Behaviors of AFI Type Crystals. Microporous Mesoporous Mater. 2003, 64, 145–153. [CrossRef]
Molecules 2018, 23, 1127
41. 42.
43. 44. 45. 46. 47.
48.
49. 50. 51.
52.
53.
54.
14 of 14
Jhung, S.H.; Chang, J.S.; Hwang, Y.K.; Park, S.E. Crystal Morphology Control of AFI Type Molecular Sieves with Microwave Irradiation. J. Mater. Chem. 2004, 14, 280–285. [CrossRef] Jhung, S.H.; Chang, J.S.; Kim, D.S.; Park, S.E. Effects of Silica on the Synthesis of AFI Molecular Sieve in Acid and Base Conditions under Microwave Irradiation. Microporous Mesoporous Mater. 2004, 71, 135–142. [CrossRef] Liu, D.C.; Zhang, B.Q.; Liu, X.F.; Li, J. Cyclohexane Oxidation over AFI Molecular Sieves: Effects of Cr, Co Incorporation and Crystal Size. Catal. Sci. Technol. 2015, 5, 3394–3402. [CrossRef] Lang, L.; Liu, X.F.; Zhang, B.Q. Synthesis and Characterization of (h0h)-Oriented Silicalite-1 Films on α-Al2 O3 Substrates. Appl. Surf. Sci. 2008, 254, 2353–2358. [CrossRef] Yan, J.; Li, J.; Zhang, B.Q.; Liu, X.F. Synthesis and Characterization of Low Aspect Ratio CrCoAPO-5 Molecular Sieves. Chin. J. Inorg. Chem. 2015, 31, 749–754. [CrossRef] Fan, W.B.; Schoonheydt, R.A.; Weckhuysen, B.M. Hydrothermal Synthesis of Co-Rich CoAPO-5 Molecular Sieves. Phys. Chem. Chem. Phys. 2001, 3, 3240–3246. [CrossRef] Grandjean, D.; Beale, A.M.; Petukhov, A.V.; Weckhuysen, B.M. Unraveling the Crystallization Mechanism of CoAPO-5 Molecular Sieves under Hydrothermal Conditions. J. Am. Chem. Soc. 2005, 127, 14454–14465. [CrossRef] [PubMed] Kornatowski, J.; Zadrozna, G.; Rozwadowski, M.; Zibrowius, B.; Marlow, F.; Lercher, J.A. New Strategy for Chromium Substitution and Crystal Morphology Control-Synthesis and Characteristics of CrAPO-5. Chem. Mater. 2001, 13, 4447–4456. [CrossRef] Laha, S.C.; Kamalakar, G.; Gläser, R. Microwave-Assisted Synthesis of [Cr]APO-5. Microporous Mesoporous Mater. 2006, 90, 45–52. [CrossRef] Kim, J.; Biswas, K.; Jhon, K.W.; Jeong, S.Y.; Ahn, W.S. Synthesis of AlPO4 -5 and CrAPO-5 Using Aluminum Dross. J. Hazard. Mater. 2009, 169, 919–925. [CrossRef] [PubMed] Li, S.; Li, Z.J.; Bozhilov, K.N.; Chen, Z.W.; Yan, Y.S. TEM Investigation of Formation Mechanism of Monocrystal-Thick b-Oriented Pure Silica Zeolite MFI Film. J. Am. Chem. Soc. 2004, 126, 10732–10737. [CrossRef] [PubMed] Lang, L.; Liu, X.F.; Jiang, H.Y.; Lin, J.Y.S.; Zhang, B.Q. Direct Evidence of the Evolutionary mechanism of Zeolite Monolayers on the Substrate Surface in a Hydrothermal Reaction. Langmuir 2010, 26, 5895–5900. [CrossRef] [PubMed] Liu, X.F.; Wang, X.D.; Na, P.; Jiang, H.Y.; Lang, L.; Zhao, H.S.; Zhang, B.Q. Evolutionary Mechanism of b-Oriented TS-1 Film on Porous α-Al2 O3 Supported Chitosan Surface in Hydrothermal Reactions. Chin. Sci. Bull. 2010, 55, 3131–3137. [CrossRef] Krajewska, B. Membrane-Based Processes Performed with Use of Chitin/Chitosan Materials. Sep. Purif. Technol. 2005, 41, 305–312. [CrossRef]
Sample Availability: Samples of the compounds are available on request from the authors. © 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
