二乙胺导向合成 SAPO-34 及与其它模板剂的对比.pdf

Chinese Journal of Catalysis 2012 文章编号: 0253-9837(2012)01-0174-09 国际版 DOI: 10.1016/S1872-2067(11)60325-2 Vol. 33 No. 1 研究论文: 174~182 二乙胺导向合成 SAPO-34 及与其它模板剂的对比 刘广宇 1,2, 田 1 鹏 1, 刘中民 1,* 中国科学院大连化学物理研究所, 辽宁大连 116023 2 河南工业大学化学化工学院, 河南郑州 450001 摘要: 以二乙胺 (DEA) 为模板剂, 合成了 SAPO-34 分子筛, 详细考察了磷酸用量、水用量以及硅源和铝源等因素的影响. 结果表 明, 在 0.7 ≤ n(P2O5)/n(Al2O3) ≤ 1.2 及 25 ≤ n(H2O)/n(Al2O3) ≤ 100 范围内, 可以得到纯相 SAPO-34 分子筛, 铝源对所得样品的组成影 响较大. 同时, 采用三乙胺 (TEA)、吗啉 (MOR) 以及二乙胺-三乙胺 (DEA-TEA) 混合模板剂合成了 SAPO-34, 通过 X 射线衍射、X 射线荧光分析、扫描电镜、热重和 29Si 固体核磁共振等手段对所得样品进行了表征. 结果显示, 不同模板剂促使硅进入 SAPO-34 骨架的能力顺序为 SAPO-34 (DEA) > SAPO-34 (MOR) > SAPO-34 (TEA); SAPO-34 骨架中所能容纳的最大 Si(4Al) 含量顺序为 SAPO-34 (DEA) ≈ SAPO-34 (MOR) > SAPO-34 (TEA). 关键词: SAPO-34; 模板剂; 二乙胺; 三乙胺; 吗啉; 合成 中图分类号: O643 文献标识码: A 收稿日期: 2011-09-03. 接受日期: 2011-10-21. *通讯联系人. 电话/传真: (0411)84685510; 电子信箱: liuzm@dicp.ac.cn 本文的英文电子版(国际版)由Elsevier出版社在ScienceDirect上出版(http://www.sciencedirect.com/science/journal/18722067). Synthesis of SAPO-34 Molecular Sieves Templated with Diethylamine and Comparison with Other Templates LIU Guangyu1,2, TIAN Peng1, LIU Zhongmin1,* 1 2 Dalian Institute of Chemical Physics, Chinese Academy of Sciences, Dalian 116023, Liaoning, China School of Chemistry & Chemical Engineering, Henan University of Technology, Zhengzhou 450001, Henan, China Abstract: SAPO-34 molecular sieves were synthesized using diethylamine (DEA) as a template. Influencing factors, such as H3PO4 content, H2O content, Al source, and Si source, were investigated. Pure SAPO-34 could be obtained under the following synthesis gel conditions: 0.7 ≤ n(P2O5)/n(Al2O3) ≤ 1.2 and 25 ≤ n(H2O)/n(Al2O3) ≤ 100. The Al source has a great influence on the resulting sample composition. SAPO-34 could also be synthesized using triethylamine (TEA), morpholine (MOR), and a DEA-TEA mixture as template, respectively. The products were characterized using X-ray diffraction, X-ray fluorescence, scanning electron microscopy, thermogravimetry, and 29Si MAS NMR. The level of silicon incorporation into the SAPO-34 framework decreased in the following order of SAPO-34 (DEA) > SAPO-34 (MOR) > SAPO-34 (TEA). The Si(4Al) contents in the SAPO-34 frameworks followed the order of SAPO-34 (DEA) ≈ SAPO-34 (MOR) > SAPO-34 (TEA). Key words: SAPO-34; template; diethylamine; triethylamine; morpholine; synthesis Received 3 September 2011. Accepted 21 October 2011. *Corresponding author. Tel/Fax: +86-411-84685510; E-mail: liuzm@dicp.ac.cn English edition available online at Elsevier ScienceDirect (http://www.sciencedirect.com/science/journal/18722067). SAPO-34 分子筛具有适宜的孔道结构、合适的 合成 SAPO-34 分子筛可采用多种模板剂, 如四 酸性及高水热稳定性, 因而在甲醇制烯烃反应 乙基氢氧化铵 [6~8] 、异丙胺 [7] 、二丙胺 [7] 、哌啶 [9] 、 (MTO) 中表现出优良的低碳烯烃选择性, 被认为是 吗 啉 (MOR) [7,10] 、 三 乙 胺 (TEA) [11~13] 、 二 乙 胺 理想的 MTO 催化剂 [1~6], 具有重要的工业应用价值. (DEA) [14~16] 等以及它们的混合物. 一般认为, 模板剂 www.chxb.cn 刘广宇 等: 二乙胺导向合成 SAPO-34 及与其它模板剂的对比 在分子筛合成过程中起结构导向、电荷匹配和空间 填充作用 [17] [18~21] 2 结果与讨论 2.1 DEA 合成 SAPO-34 的影响因素 . 模板剂的种类对合成的分子筛结构和 物化性质等的影响很大. 目前人们对此进行了研究 , 但研究主要集中在模板剂对所得 SAPO-34 晶 粒大小及催化性能的影响. 而有关模板剂对分子筛 酸性质影响的本质原因, 即硅进入分子筛骨架的程 度及分布状况等并未进行细致研究. 近期我们以 DEA 为模板剂, 合成了 SAPO-34 分 子筛, 并考察了其催化 MTO 反应的性能 [15]. 本文详 细考察了该合成过程各因素的影响, 以获得优化的 合成条件; 同时, 比较了 DEA, TEA 和 MOR 等常用的 几种有机胺为模板剂合成样品的特点, 以深入理解 SAPO-34 的合成过程. 1 实验部分 1.1 分子筛的合成 175 磷酸用量 2.1.1 固 定 合 成 原 料 摩 尔 配 比 为 2.0DEA:0.6SiO2: 1.0Al2O3:nP2O5:50H2O, 考察了 H3PO4 (即 P2O5) 用量 对 SAPO-34 合成的影响, 所得样品的 XRD 谱见图 1. 可以看出, 随初始凝胶中 P2O5 用量增加, 晶化前后 pH 值均呈下降趋势. 当 n(P2O5) 为 0.5 时, 产物为无 定形的一水软铝石胶; 至 0.6 时开始有 SAPO-34 生 成; 在 n(P2O5) 为 0.7~1.2 时可得到纯的 SAPO-34; 继 续增大 P2O5 用量, 产物中开始出现具有十元环孔道 的 AFO 型 SAPO-41 分子筛; 当 n(P2O5) 为 1.5 时, 产 物中已经没有 SAPO-34. SAPO-41 boehmite (7) 采 用 水 热 晶 化 法 合 成 SAPO-34 分 子 筛 . 将 铝 (6) 源、硅源 (除特别说明外, 分别为拟薄水铝石、硅溶 剂, 继续搅拌至形成均匀的凝胶, 再移入 200 ml 的聚 (5) Intensity 胶) 、磷酸与去离子水混合, 室温搅拌 1 h 后加入模板 (4) 四氟乙烯内衬的不锈钢合成釜中, 密闭加热到 200 (3) o (2) (1) C, 在 自 身 压 力 下 晶 化 24 h. 典 型 凝 胶 摩 尔 比 为 2.0DEA:0.6SiO2:1.0Al2O3:0.8P2O5:50H2O. 待 晶 化 完 10 全后将固体产物离心分离, 并用去离子水洗涤至中 20 性, 在 120 oC 空气中干燥后, 得到 SAPO-34 分子筛原 粉. 原粉在 550 oC 空气中焙烧 5 h 除去有机模板剂后 即为焙烧型样品. 1.2 分子筛的表征 分子筛样品的 X 射线衍射 (XRD) 测试在日本理 学的 D/max-γb 型 X 射线衍射仪上进行, Cu 靶, Kα 辐 射源 (λ = 0.15406 nm), 电压 40 kV, 电流 40 mA. 样品 形 貌 由 KYKY-AMRAY-1000B 型 扫 描 电 子 显 微 镜 (SEM) 观 测 . 样 品 组 成 采 用 Philips 公 司 的 Magix 2424 X 射线荧光分析 (XRF) 仪测定. 热分析在美国 Perkin-Elmer 公司的 Pyris-1 TGA 型热分析仪上进行, 空气气氛, 升温速率 10 oC/ min. 29 Si 固体核磁共振 (29Si MAS NMR) 谱在 Varian Infinity-plus 400 WB 型固体核磁共振波谱仪上测量, 使用 BBO MAS 探头和 7 mm ZrO2 样品管, 共振频率 为 79.5 MHz, 转速为 4 kHz, 化学位移以三甲基硅丙 磺酸钠 (DSS) 为参考外标. 图1 30 2θ /( o ) 40 50 不同 H3PO4 用量合成样品的 XRD 谱 Fig. 1. XRD patterns for samples synthesized with different H3PO4 amounts in the initial gel (2.0DEA:0.6 SiO2:1.0Al2O3:nP2O5:50H2O). n(P2O5) = (1) 0.5; (2) 0.6; (3) 0.7; (4) 1.2; (5) 1.3; (6) 1.4; (7) 1.5. 2.1.2 水用量 固 定 合 成 原 料 摩 尔 配 比 为 2.0DEA:0.6SiO2: 1.0Al2O3:0.8P2O5:nH2O, 考察了 H2O 用量对 SAPO-34 合成的影响, 结果见表 1. XRD 结果表明, 当 H2O 量 为 15~300 时均可得到 SAPO-34. 由表可见, 晶化前 合成凝胶的 pH 值基本相同, 约为 6.0, 晶化后凝胶 pH 值随 H2O 量的增加而降低. SAPO-34 收率在 n(H2O) > 25 时随水量增多而下降. 当 n(H2O) < 50, 样品相对 结晶度随 H2O 量增多而增大, 至 50 时达最大; 当水 量超过 50, 样品结晶度降低, 至 300 时降到 33%, 有一 些无定形铝物种生成. 化学元素分析显示, 合成样品中硅含量随 H2O 176 催 表1 化 学 Chin. J. Catal., 2012, 33: 174–182 报 H2O 用量对 SAPO-34 合成的影响 Table 1 Effect of H2O content in the initial gel on the synthesis of SAPO-34 pH before pH after Relative crystallinity Relative yield crystallization crystallization (%) (%) 15 5.5 9.0 88 48 25 6.0 9.0 93 100 Si0.154Al0.494P0.352O2 50 6.0 8.0 100 79 Si0.150Al0.494P0.356O2 100 6.0 8.0 78 71 Si0.131Al0.571P0.299O2 300 6.0 6.0 33 44 Si0.096Al0.697P0.207O2 n(H2O) Molar composition Si0.169Al0.496P0.335O2 量增多而下降. H2O 量过少时, 晶化体系的液相离子 SAPO-34 外 , 其 余 的 样 品 均 为 结 晶 度 很 高 的 浓度高, 合成的推动力大, 随着晶化产物的生成, 体 SAPO-34 分子筛纯相. 这是因为 SAPO 分子筛的合 系 pH 值升高, 从而导致合成产物的溶解度增加, 使 成需要弱酸性或中性条件, 而以铝溶胶为铝源时所 得产率下降. 我们前期的研究 [15] 也发现, 增大 DEA 用量 (pH 值升高) , 产物中硅含量增加, 但产率降低. 制凝胶 pH 值较低 (0.5), 酸性较强 (凝胶制备过程没 有另外调节 pH 值), 使合成凝胶变为 α-石英相. H2O 量过多时, 晶化体系的液相离子浓度较低, 合成 由表 2 可见, 铝源对合成样品的组成影响较大. 的 推 动 力 变 小 , 晶 化 产 率 降 低 . 因 此 , n(H2O) 为 以硅溶胶为硅源, 其中以异丙醇铝为铝源时所得样 25~100 比较适宜. 品的硅含量最低, 以拟薄水铝石为铝源时最高. 这可 硅源和铝源 能是因为单体铝源异丙醇铝的活性较高, 与磷酸反 2.1.3 固定合成凝胶摩尔配比, 考察了不同反应活性 应的活性要高于聚合态的铝源, 如拟薄水铝石. 异丙 和聚合度的硅源 (硅溶胶、白碳黑、正硅酸乙酯) 及 醇铝与磷酸混合后, 可以很快形成磷酸铝前驱体. 硅 铝源 (拟薄水铝石、异丙醇铝、氯化铝、铝溶胶) 对 溶胶为聚合态的硅源, 其解聚释放出活性硅物种的 SAPO-34 的影响, 结果见表 2. 可以看出, 除了以合成 速率可能相对较慢, 抑制了硅取代骨架 P 或 P-Al 对 铝溶胶为铝源、硅溶胶为硅源时, 所得样品不是 的过程, 从而导致分子筛样品的硅含量相对较低. 另 表2 Table 2 外, 以白碳黑和正硅酸乙酯为硅源合成的样品中硅 铝源及硅源对 SAPO-34 合成的影响 Effects of the Al source and Si source on the synthesis of 含量略高于硅溶胶的. 图 2 为不同硅源和铝源合成的 SAPO-34 样品的 SAPO-34 products Silica source Alumina source Product Fume silica pseudoboehmite SAPO-34 Relative cry- Molar stallinity (%) composition 108 Si0.157Al0.501P0.342O2 Si(OC2H5)4 pseudoboehmite SAPO-34 94 Si0.157Al0.514P0.329O2 Silica sol pseudoboehmite SAPO-34 100 Si0.150Al0.494P0.356O2 Silica sol Al(i-OC3H7)3 SAPO-34 87 Si0.093Al0.527P0.380O2 Silica sol AlCl3 SAPO-34 90 Si0.106Al0.498P0.396O2 Silica sol alumina sol α-quartz — — (a) (b) 晶体, 其中以白碳黑或正硅酸乙酯为硅源时所得样 品尺寸明显要小于硅溶胶的. 2.2 不同模板剂合成对比研究 本文比较了 DEA, TEA, MOR 以及 DEA-TEA 混 合模板剂所得样品的物化性质, 具体合成原料配比 (c) 图2 SEM 照片. 可以看出, 所得 SAPO-34 样品皆为菱形 (d) (e) 不同铝源、硅源合成样品的 SEM 照片 Fig. 2. SEM images for samples synthesized with different Al and Si sources. (a) pseudoboehmite and fume silica; (b) pseudoboehmite and Si(OC2H5)4; (c) pseudoboehmite and silica sol; (d) Al(i-OC3H7)3 and silica sol; (e) AlCl3 and silica sol. www.chxb.cn 刘广宇 等: 二乙胺导向合成 SAPO-34 及与其它模板剂的对比 表3 不同模板剂合成 SAPO-34 的原料配比及晶化条件 Table 3 Composition of synthesis gels and crystallization conditions for samples synthesized using different templates Sample 共晶体 (含微量的 SAPO-18). 合成样品的元素分析结果见表 4. 可以发现, 单 一模板剂合成样品中, DEA 合成的 SAPO-34 骨架中 Gel composition Time (h) t/oC 硅含量最高, n(Si)/n(Si + Al + P) 为 0.150, MOR 合成 50 24 200 的次之, TEA 合成的最低, 为 0.102, 可见, 以 DEA 为 0.8 50 24 200 模板剂合成样品具有高硅进入的特点. 以 DEA-TEA 1.0 0.8 50 24 200 1.0 0.8 50 24 200 混合模板剂合成的 SAPO-34 骨架硅含量约等于两 1.0 0.8 50 24 200 Template SiO2 Al2O3 P2O5 H2O 2TEA 0.6 1.0 0.8 DT-0.6Si 1DEA+1TEA 0.6 1.0 D-0.6Si 2DEA 0.6 M-0.6Si 2MOR 0.6 D-0.4Si 2DEA 0.4 T-0.6Si 177 者单独合成样品的骨架硅含量平均值, 显示出混合 模板剂的“加和”效应. 图 4 为 合 成 样 品 的 SEM 照 片 . 可 以 看 出 , 以 晶相和元素组成 2.2.1 晶体形貌 2.2.2 和晶化条件见表 3. 图 3 为不同模板剂合成样品的 XRD 谱. 可以看 出, 各样品的衍射峰均与 SAPO-34 的一致, 其中以 TEA 为 模 板剂 合 成 样 品 (T-0.6Si) 的 相对 结 晶 度 较 低, 其余均较高 (见表 4). 值得注意的是, T-0.6Si 样品 中几个主峰强度很高, 但高角度部分的衍射峰很弱 或缺失. 这有可能是模板剂变化引起的晶体骨架扭 MOR 为模板剂合成的 SAPO-34 样品晶粒最大, 达到 10 μm; TEA 的 最 小 , 约 为 2 μm; DEA 合 成 样 品 D-0.6Si 粒径在 2~4 μm 之间, 而以 DEA-TEA 混合模 板剂合成样品粒径介于 DEA 和 TEA 单独合成样品 粒径之间. T- (a) (b) DT- D-0.6 (c) M-0.6 (d) 曲所致, 也有可能是形成了 SAPO-34 和 SAPO-18 的 T-0.6Si Intensity DT-0.6Si D-0.6Si M-0.6Si 10 图3 20 2θ/( o ) 30 40 50 图4 不同模板剂合成样品的 XRD 谱 Fig. 3. XRD patterns for samples synthesized using different templates. 表4 SAPO-34 样品的相对结晶度和元素分析结果 Table 4 Relative crystallinity and chemical composition of SAPO-34 Sample Relative crystallinity (%) Product composition Si incorporation* 不同模板剂合成样品的 SEM 照片 Fig. 4. SEM images for samples synthesized using different templates. (a) T-0.6Si; (b) DT-0.6Si; (c) D-0.6Si; (d) M-0.6Si. 2.2.3 热分析 不同模板剂合成的样品在空气气氛下的 TG 结 果显示, 所有样品的失重分为三个阶段, 分别为< T-0.6Si 63 Si0.102Al0.484P0.414O2 0.71 DT-0.6Si 97 Si0.128Al0.480P0.392O2 0.90 200, 200~490 和 > 490 oC. 第一阶段是分子筛上水的 D-0.6Si 100 Si0.150Al0.494P0.356O2 1.05 脱除; 第二和第三阶段分别对应于分子筛中模板剂 M-0.6Si 91 Si0.134Al0.475P0.391O2 0.94 D-0.4Si — Si0.134Al0.495P0.371O2 1.34 *Molar ratio of (Si/(Si + Al + P))product/(Si/(Si + Al + P))gel. 的燃烧分解 和残留的高 碳氢比的有 机物的燃烧 分 解. 表 5 给出了各阶段的失重结果. 可以看出, 水的 178 催 表5 Table 5 化 学 Chin. J. Catal., 2012, 33: 174–182 报 SAPO-34 的热分析结果 Thermal analysis results of SAPO-34 Percentage Percentage Percentage of template Moles of Sample of H2O of template (%) template (%) (%) 200–490 oC > 490 oC per cage* T-0.6Si 4.01 11.83 79 21 1.02 DT-0.6Si 3.74 13.06 73 27 1.32 D-0.6Si 4.41 13.96 72 28 1.71 M-0.6Si 4.74 16.53 52 48 1.76 T-0.6Si DT-0.6Si D-0.6Si M-0.6Si *The molecular weights for the mixed templates are calculated by averaging those for the individual components. 失重比较接近, 而模板剂的失重相差较大: TEA 最 小, 为 11.83%; MOR 最大, 为 16.53%. 由 TG 结果可 算得分子筛每个 CHA 笼中的平均模板剂数: MOR 最高为 1.76, 与 Prakash 等 [22] 结果相近; DEA 为 1.71, 接近 MOR; TEA 最低, 为 1.02. 使用 DEA-TEA 混合 D-0.4Si -80 图5 -90 -100 δ Fig. 5. 29Si MAS NMR spectra of samples synthesized using different templates. 表6 从 29Si MAS NMR 谱拟合得到的硅环境分布 Table 6 大增加. 我们认为, 不同模板剂在 CHA 笼中数量的 voluted 29Si MAS NMR spectra 差异主要源自它们分子体积的不同, TEA 由于含有 Sample 三个乙基支链, 分子体积明显大于只有两个乙基支 链的 DEA 和成环状的 MOR. 填充作用外, 还具有平衡分子筛骨架电荷的作用. 模 -120 不同模板剂合成样品的 29Si MAS NMR 谱 模板剂, 进入分子筛内的模板剂数量较纯 TEA 有较 在分子筛合成中, 模板剂除了结构导向和空间 -110 Distribution of silicon environments observed in the deconSilicon environments distribution (%) Si(4Al) Si(3Al) Si(2Al) Si(1Al) Si(0Al) T-0.6Si 66.80 25.26 3.45 2.09 2.40 DT-0.6Si 71.09 18.08 3.98 3.50 3.36 D-0.6Si 73.75 14.47 6.21 2.80 2.77 M-0.6Si 86.60 8.62 2.09 0.20 2.49 D-0.4Si 87.00 11.20 0.00 0.00 1.80 板剂所能提供的正电荷数决定了分子筛的骨架负电 荷数, 而这些负电荷是由于硅原子进入骨架取代 P M-0.6Si 中硅环境分布与 D-0.4Si 的接近. DEA-TEA 原子或 P-Al 对形成硅岛造成的, 因此模板剂数量也 混合模板剂样品 DT-0.6Si 的硅环境分布则为两模板 在一定程度上决定了骨架中硅含量. TEA 进入分子 剂样品的中间值. 筛内的量最少, 所以相应的 SAPO-34 样品中骨架硅 根据拟合所得不同硅环境的比例, 结合表 4, 可 含量最低. DEA 进入分子筛内的量与 MOR 接近, 但 算得每个样品的骨架净电荷及相应每个笼中起平衡 合成样品中骨架硅含量却比 MOR 高. 这可能是因为 骨 架 电 荷 作 用 的 模 板 剂 数 量 A1; 结 合 表 5 中 分子筛孔笼内的 MOR 有一部分只起孔道填充作用, SAPO-34 每个笼中的模板剂分子数, 可算得孔笼中 并没有起到电荷补偿作用. 没有起到平衡骨架电荷作用的模板剂数量 A2 , 结果 29 2.2.4 29 Si MAS NMR 分析 见表 7. 对比具有不同硅含量的 D-0.4Si 和 D-0.6Si, Si MAS NMR 可以用来表征 Si 在 SAPO-34 分 尽管分子筛孔笼内的模板剂数量相当, 但随着骨架 子筛骨架上的分布状态. 图 5 给出了各样品的 29Si 中硅含量的增加, 用于平衡骨架电荷的模板剂数量 MAS NMR 谱 . 可 以 看 出 , 几 个 样 品 均 在 δ = –92.5, 也增大. 0.6Si 系列样品中, 以 TEA 和 DEA-TEA 为模 –96.5, –101, –105 和–111 处出现共振峰, 可分别归属 板剂的 T-0.6Si 和 DT-0.6Si 样品的 A2 均为 0, 说明这 为 Si(4Al), Si(3Al), Si(2Al), Si(1Al) 和 Si(0Al) 环 两个样品孔笼中的模板剂全部用于平衡骨架电荷. [11,17] . 将各样品谱峰进行高斯拟合, 结果见表 6. 对 以 DEA 和 MOR 为模板剂的样品中, A2 分别为 0.12 比 D-0.4Si 和 D-0.6Si 样品, 随着骨架硅含量的增加, 和 0.25, 相 应 于 平 衡 骨 架 电 荷 的 模 板 剂 占 总 量 的 骨架中 Si(4Al) 环境所占比例降低, 相应于硅岛形成 93.0% 和 85.8%. 也就是说, 在相同初始凝胶摩尔比 的其它硅环境比例增加. 0.6Si 系列样品中, T-0.6Si 例的情况下, 几种模板剂用于平衡骨架电荷的用量 和 M-0.6Si 分 别 具 有 最 低 和 最 高 的 Si(4Al) 含 量 . 比例顺序为 TEA > DEA > MOR. 境 www.chxb.cn 刘广宇 等: 二乙胺导向合成 SAPO-34 及与其它模板剂的对比 179 DT-0.6Si –0.1120 1.34 0 0.091 D-0.6Si –0.1326 1.59 0.12 0.111 D. J Phys Chem, 1995, 99: 8270 9 Dumitriu E, Azzouz A, Hulea V, Lutic D, Kessler H. Microporous Mater, 1997, 10: 1 10 Vistad Ø B, Akporiaye D E, Taulelle F, Lillerud K P. Chem Mater, 2003, 15: 1639 11 Buchholz A, Wang W, Xu M, Arnold A, Hunger M. Microporous Mesoporous Mater, 2002, 56: 267 12 Wei Y X, Zhang D Z, Xu L, Chang F X, He Y L, Meng S H, Su B L, Liu Z M. Catal Today, 2008, 131: 262 M-0.6Si –0.1262 1.51 0.25 0.116 13 刘中民, 蔡光宇, 何长青, 杨立新, 王作周, 罗静慎, 常 D-0.4Si –0.1280 1.54 0.17 0.116 表 7 不 同 模板 剂 合 成 的 SAPO-34 中 骨 架 电 荷 数 和 最 大 Si(4Al) 量 Table 7 Net charge of SAPO-34 samples synthesized using different templates and the maximum Si(4Al) contents Sample Net charge per mole T-0.6Si a (Al + Si + P) –0.0897 A1a A2b 1.07 0 Maximum Si(4Al) contentc 0.068 Template number per cage used for balancing the net charge; b c Template number per cage – A1; Maximum Si(4Al) content in SAPO-34 per mole (Al + Si + P). 根据各样品硅含量和 Si(4Al) 比例, 算得不同模 板剂合成 SAPO-34 骨架中允许出现的最大 Si(4Al) 量, 即骨架中硅含量大于该值后, 其它硅环境开始出 现 , 结 果 列 于 表 7. DEA 和 MOR 合 成 的 样 品 最 大 Si(4Al) 量接近, 明显大于 TEA 合成的样品. 3 结论 彦君, 石仁敏, 姜增全, 孙承林 (Liu Zh M, Cai G Y, He Ch Q, Yang L X, Wang Z Zh, Luo J Sh, Chang Y J, Shi R M, Jiang Z Q, Sun Ch L). CN 1 088 483. 1994 14 何长青, 刘中民, 蔡光宇, 杨立新, 王作周, 罗静慎, 潘 小珊, 盖增全, 常彦君, 石仁敏 (He Ch Q, Liu Zh M, Cai G Y, Yang L X, Wang Z Zh, Luo J Sh, Pan X Sh, Gai Z Q, Chang Y J, Shi R M). CN 1 096 496. 1994 15 Liu G Y, Tian P, Li J Z, Zhang D Z, Zhou F, Liu Z M. Microporous Mesoporou Mater, 2008, 111: 143 16 Liu G Y, Tian P, Zhang Y, Li J Z, Xu L, Meng S H, Liu Z M. Microporous Mesoporou Mater, 2008, 114: 416 17 Pastore H O, Coluccia S, Marchese L. Ann Rev Mater Res, 2005, 35: 351 18 李军, 张凤美, 李黎声, 舒兴田. 石油炼制与化工 (Li J, 以 DEA 为模板剂合成了 SAPO-34 分子筛, 并与 其 它 模 板 剂 进 行 了 比 较 . 发 现 DEA 导 向 合 成 SAPO-34 的较佳条件为 0.7 ≤ n(P2O5)/n(Al2O3) ≤ 1.2 和 25 ≤ n(H2O)/n(Al2O3) ≤ 100. 铝源对所得 SAPO-34 o 样品组成的影响较大. 在 200 C 晶化 24 h 条件下, DEA 合成的样品结晶度和骨架硅含量最高, TEA 合 成的样品最低, MOR 的介于中间. 这与模板剂促使 硅进入骨架的能力顺序 DEA > MOR > TEA 有关. 以 DEA 和 MOR 为 模 板 剂 合 成 样 品 允 许 出 现 的 最 大 Zhang F M, Li L Sh, Shu X T. Petrol Process Petrochem), 2005, 36(6): 49 19 Lee Y J, Baek S C, Jun K W. Appl Catal A, 2007, 329: 130 20 何 长 青 , 刘 中 民 , 杨 立 新 , 蔡 光 宇 . 催 化 学 报 (He Ch Q, Liu Zh M, Yang L X, Cai G Y. Chin J Catal), 1995, 16: 33 21 刘红星, 谢在库, 张成芳, 陈庆龄. 化学物理学报 (Liu H X, Xie Z K, Zhang Ch F, Chen Q L. Chin J Chem Phys), 2003, 16: 521 22 Prakash A M, Unnikrishnan S. J Chem Soc, Faraday Trans, 1994, 90: 229 Si(4Al) 量接近, 明显大于 TEA 的. DEA-TEA 混合模 板剂合成的样品则体现出两种模板剂的加和效应. 英 译 文 English Text 参 考 文 献 1 Qi G Z, Xie Z K, Yang W M, Zhong S Q, Liu H X, Zhang Ch F, Chen Q L. Fuel Process Technol, 2007, 88: 437 2 Park Y K, Baek S W, Ihm S K. J Ind Eng Chem, 2001, 7: 167 3 Hunger M, Seiler M, Buchholz A. Catal Lett, 2001, 74: 61 4 Arstad B, Kolboe S. Catal Lett, 2001, 71: 209 5 Liang J, Li H Y, Zhao S Q, Guo W G, Wang R H, Ying M L. Appl Catal, 1990, 64: 31 6 Wilson S, Barger P. Microporous Mesoporous Mater, 1999, 29: 117 7 Lok B M, Messina C A, Patton R L, Gajek R T, Cannan T R, Flanigen E M. US 4 440 871. 1984 8 Briend M, Vomscheid R, Peltre M J, Man P P, Barthomeuf Owing to their porous structure, moderate acidity, and good hydrothermal stability, SAPO-34 molecular sieves show a high selectivity towards light olefins in the methanol-to-olefin (MTO) reaction [1–6]. SAPO-34 can be synthesized using organic amines as templates, such as tetraethylammonium hydroxide [6–8], isopropylamine [7], dipropylamine [7], piperidine [9], morpholine (MOR) [7,10], triethylamine (TEA) [11–13], diethylamine (DEA) [14–16], and mixtures thereof. It is generally acknowledged that the template can play structure-directing, charge-compensating, and space-filling roles in the synthesis of molecular sieves [17]. The template exerts a significant influence on both the structures and properties 180 催 化 of the molecular sieves. A number of reports mainly focus on the effects of templates on crystal size and catalytic performance [18–21]. Investigations into the intrinsic reasons of how templates influence SAPO-34 acidity (i.e. how do the templates impact Si incorporation and distribution) are relatively scarce. We have recently investigated the synthesis and catalytic properties of SAPO-34 in the MTO reaction, synthesized using DEA as a template [15]. In the present paper, we probe the influencing factors in the synthesis of SAPO-34, templated with DEA, to obtain the optimal synthesis conditions. Furthermore, a comparative study into several commonly used templates in the synthesis of SAPO-34, including DEA, TEA, and MOR, was performed to reveal the characteristics of samples prepared using different templates. 1 Experimental 1.1 Synthesis SAPO-34 was synthesized using a hydrothermal method. The Si source (unless stated otherwise, it is silica sol), the Al source (unless stated otherwise, it is pseudoboehmite), phosphoric acid, and distilled water were mixed and stirred at room temperature. The template was added 1 h later and the resulting mixture was stirred until homogeneous. The gel was transferred to a 200 ml Teflon-lined stainless-steel autoclave, and heated and crystallized at 200 °C under autogenic pressure for 24 h. The typical molar composition of the gel was 2.0DEA:0.6SiO2:1.0Al2O3:0.8P2O5:50H2O. After crystallization, the as-synthesized sample was obtained after centrifugal separation, washing and drying in air at 120 °C. Calcination was then carried out in air at 550 °C for 5 h to remove the organic template. 1.2 Characterization Powder XRD patterns were recorded on a Rigaku D/MAX-γB X-ray diffractometer using Cu Kα radiation (λ = 0.15406 nm) at 40 kV/40 mA. The crystal morphology was analyzed via scanning electron microscopy (SEM, KYKY-AMRAY-1000B). The chemical composition of the sample was obtained on a Philips Magix-2424 X-Ray Fluorescence (XRF) spectrometer. Thermal analyses were measured on a Perkin Elmer Pyris-1 TGA analyzer, at a temperature-programmed rate of 10 °C/min under an air flow. 29 Si MAS NMR spectra were recorded in 7 mm ZrO2 rotors at 79.5 MHz on a Varian Infinity-plus 400 WB spectrometer, fitted with a BBO probe. The spinning rate of the samples at the magic angle was 4 kHz. The internal standard for chemical shifts was 2,2-dimethyl-2-silapentane-5-sul- 学 报 Chin. J. Catal., 2012, 33: 174–182 fonate sodium salt (DSS). 2 Results and discussion 2.1 Influence factors in the synthesis of SAPO-34 templated by DEA 2.1.1 H3PO4 content The effect of H3PO4 content (i.e. P2O5 content) on the synthesis was investigated with a gel molar composition of 2.0DEA:0.6SiO2:1.0Al2O3:nP2O5:50H2O. The XRD patterns for the obtained samples are shown in Fig. 1. With increasing P2O5 content, the pH values, before and after crystallization, obviously decrease and the products undergo a regular change. When n = 0.5, the product is obtained as an amorphous boehmite. When n = 0.6, SAPO-34 is formed. Pure SAPO-34 crystals can be obtained with 0.7 ≤ n(P2O5) ≤ 1.2. SAPO-41, with an AFO structure and 10-ring channels, is formed with increasing n. When n = 1.5, no SAPO-34 is observed. 2.1.2 H2O content The influence of the H2O level on the synthesis was investigated with gel molar compositions of 2.0DEA:0.6SiO2: 1.0Al2O3:0.8P2O5:nH2O. XRD results show that SAPO-34 can be synthesized with 15 ≤ n(H2O) ≤ 300. Table 1 shows that the pH is almost consistent for all samples (about 6.0). After crystallization, the pH values decrease with increasing H2O. The relative yields of SAPO-34 decrease with increasing H2O content (i.e. n ≥ 25). The relative crystallinity reaches its optimum value when n = 50. This rises with increasing n(H2O) when n < 50, but descends when n > 50. When n = 300, the relative crystallinity decreases to 33% and amorphous aluminum species begin to appear. The chemical analysis given in Table 1 indicates that the silicon content in SAPO-34 declines with increasing H2O content. A reduced H2O content in the gel leads to a higher ion concentration, favoring synthesis. With the formation of products, the pH of the system rises. This increases the product solubility in the gel and reduces the yield. In a former study [15], the silicon content in the SAPO-34 increased with increasing DEA content (with increasing pH), whilst the crystal yield decreases. A larger H2O content in the gel causes a lower ion concentration and a reduced driving force for synthesis. This results in a lower yield. Therefore, the optimal H2O level ranges between 25 ≤ n(H2O) ≤ 100. 2.1.3 Si sources and Al sources The influences of the Si (silica sol, fume silica and tetra- www.chxb.cn 刘广宇 等: 二乙胺导向合成 SAPO-34 及与其它模板剂的对比 ethoxysilane) and Al sources (pseudoboehmite, aluminum isopropoxide, aluminum chloride and alumina sol) on the synthesis were investigated with a gel molar composition of 2.0DEA:0.6SiO2:1.0Al2O3:0.8P2O5:50H2O. The results are shown in Table 2. Except for sample synthesized with alumina sol as the Al source and silica sol as the Si source, all compositions generated pure SAPO-34 with high crystallinity. The preparation of SAPO molecular sieves requires weakly acidic or neutral conditions. The pH value for the system with alumina sol as the Al source is too low (0.5), which leads to the formation of a quartz phase. The chemical analysis given in Table 2 suggests that the Al source has a significant effect on the compositions of the as-synthesized samples. When silica sol is used as the Si source, sample using aluminum isopropoxide as the Al source, gives the lowest silicon content. Meanwhile, sample using pseudoboehmite as the Al source, gives the highest. This may be due to the higher reactivity of the monomeric Al source (aluminum isopropoxide) towards H3PO4, compared with the polymeric Al source (pseudoboehmite). An aluminophosphate precursor can form upon mixing aluminum isopropoxide with H3PO4. However, silica sol is a polymeric Si source, and the rate of depolymerization to give the active Si species is relatively slow. The Si substitution processes, for either P or P-Al pairs within the framework, were inhibited, which leads to the lower silicon content in sample. With pseudoboehmite as the Al source, the silicon contents for samples using fume silica and tetraethoxysilane as the Si source are the same. Both are slightly higher than that for sample using silica sol as the Si source. Figure 2 presents the SEM images for the samples synthesized with the different Al and Si sources. The SAPO-34 crystals are rhombohedra. The crystal sizes in the samples synthesized with fume silica (a) and tetraethoxysilane (b) are clearly smaller than those obtained with silica sol (c, d, e) as the Si source. 2.2 Comparative study for SAPO-34 samples synthesized with different templates The templates used in the comparative study include DEA, TEA, MOR, and a DEA-TEA mixture. The gel compositions and crystallization conditions are shown in Table 3. 2.2.1 Crystal phase and chemical composition Figure 3 shows the XRD patterns for samples synthesized using the different templates. All XRD patterns are in agreement with those of SAPO-34 reported in the literature. With the exception of sample T-0.6Si (templated with TEA), all samples exhibit a relatively high crystallinity (see Table 4). The main peaks in the XRD for sample T-0.6Si are of high 181 intensity, whilst the high-angle peaks are either weak or missing. This may be due to crystal structure distorting or the formation of eutectic SAPO-34 and SAPO-18 (trace amounts of SAPO-18). Of the samples obtained with 0.6Si, that using DEA as the template, lead to the highest Si content (n(Si)/n(Si + Al + P) = 0.150). Meanwhile, that templated with TEA gave the lowest (n(Si)/n(Si + Al + P) = 0.102). This suggests that DEA templated SAPO-34 exhibits the appropriate characteristics for high silicon incorporation. The silicon content in sample DT-0.6Si, which was templated with a DEA-TEA mixture, is approximately equal to the average for samples D-0.6Si and T-0.6Si, consistent with the additive effect. 2.2.2 Crystal morphology Figure 4 presents the SEM images for the samples synthesized with different templates. The crystal size for sample M-0.6Si, templated with MOR, is the largest (up to 10 μm), whilst that for sample T-0.6Si is the smallest (about 2 μm). The crystal size in sample D-0.6Si is in the range of 2–4 μm, which is slightly larger than that of T-0.6Si. The crystal size of DT-0.6Si is between those of samples D-0.6Si and T-0.6Si. 2.2.3 Thermal analysis The TG profiles for the samples synthesized with the different templates reveal three stages of weight loss at < 200 °C, 200–490 °C, and > 490 °C, respectively. The first stage is attributed to water desorption from the sample. The second and third stages are due to the combustion decomposition of the template and the organic residues. The data given in Table 5 indicates that the weight losses associated with water desorption are close for all samples, while those due to the template vary significantly. The weight loss with TEA is the lowest (11.83%), whilst that with MOR is the highest (16.53%). The highest template mole number per cage, calculated based on the template weight loss, was found to be 1.76 for MOR, which is close to the result calculated by Prakash et al. [22]. The mole numbers obtained with TEA and DEA were found to be 1.02 and 1.71, respectively. The mole number for the DEA-TEA mixture was significantly higher than that obtained with TEA. The differences in the mole numbers per cage mainly come from the differences in molecular size. With three ethyl side chains, the molecular size of TEA is obviously larger than that of DEA, which only has two ethyl side chains. In addition to the structure-directing and space-filling roles, the template also plays a charge-compensating role in the synthesis of molecular sieves. The positive charge provided by the template determines the negative charge of the framework. This arises from the Si substitution of the P or 182 催 化 P-Al pairs in framework. Therefore, the template mole number per cage influences the silicon content in framework to a certain degree. Since the TEA mole number per cage is the lowest, the associated silicon content in T-0.6Si is also lowest. Although the mole numbers for DEA and MOR per cage are similar, the silicon content in D-0.6Si is higher than that of M-0.6Si. Portions of the MOR in the cage may not play a charge-compensating role. 2.2.4 29 Si MAS NMR analysis The Si distribution in SAPO-34 framework can be characterized by 29Si MAS NMR. Figure 5 shows the 29Si MAS NMR spectra for the as-synthesized samples. Every sample presents a different Si distribution. According to the literature [11,17], the peaks at δ = –92.5, –96.5, –101, –105, and –111 are ascribed to Si(4Al), Si(3Al), Si(2Al), Si(1Al), and Si(0Al) environments, respectively. Table 6 shows the distribution of the silicon environments, as evaluated from Gaussian fitting. Upon Comparing the D-0.4Si with D-0.6Si samples, the proportion of Si(4Al) environments decreases with increasing silicon content in the framework. Meanwhile, the proportions of the other Si environments increase. From the series of samples with 0.6Si, the proportion of Si(4Al) environments is at its lowest in sample T-0.6Si and at its highest in sample M-0.6Si. The distribution of silicon environments in sample M-0.6Si is similar to that in sample D-0.4Si. The distribution of silicon environments in sample DT-0.6Si is a cross between those for samples D-0.6Si and T-0.6Si. The net charge in framework and the template number per cage (A1) can be calculated based on the proportion of Si environments and the silicon contents in each sample (Table 4). In combination with the template mole number per cage in Table 5, the residual template number per cage (A2), which is not used for balancing the net charge, can be obtained. The results are shown in Table 7. Upon comparing samples D-0.4Si and D-0.6Si, although their template numbers per cage are close, the template number used for balancing the 学 报 Chin. J. Catal., 2012, 33: 174–182 net charge increases with increasing silicon content. The A2 values for samples T-0.6Si and DT-0.6Si are both zero, which means all the templates in the cage are used to balance the framework charge. The A2 values for samples D-0.6Si and M-0.6Si are 0.12 and 0.25, meaning the percentages of template used for balancing the framework charge are 93.0% and 85.8%, respectively. Hence, with the same initial gel, the percentage of template number used for balancing the framework charge follows the order TEA > DEA > MOR. Based on the silicon content and Si(4Al) proportion for the various samples, the maximum Si(4Al) content within the SAPO-34 can be obtained. Other Si environments will appear when the silicon content increases beyond the maximum Si(4Al) level. This indicates that the maximum Si(4Al) contents for samples D-0.6Si and M-0.6Si are similar, and significantly greater than that of T-0.6Si. 3 Conclusions In the present paper we have investigated the synthesis of SAPO-34 molecular sieves templated with DEA, and compared their properties with those synthesized using different templates. The optimal synthesis conditions for the DEA templated SAPO-34 are 0.7 ≤ n(P2O5)/n(Al2O3) ≤ 1.2 and 25 ≤ n(H2O/Al2O3) ≤ 100. The Al source has a great influence on the resulting sample composition. A comparative study revealed the DEA templated sample to have the highest crystallinity and silicon content. MOR had the second highest, and TEA the lowest. This is related to the templates ability to promote Si incorporation into the SAPO-34 framework, which follows the order DEA > MOR > TEA. The maximum Si(4Al) contents for samples templated with DEA and MOR are similar, but both were significantly greater than that observed with TEA. The properties for the samples templated with a DEA-TEA mixture are consistent with the additive effect. Full-text paper available online at Elsevier ScienceDirect http://www.sciencedirect.com/science/journal/18722067