【作者】 蓝建慧；

【导师】 汪文川；

【作者基本信息】 北京化工大学 ， 化学工程与技术， 2009， 博士

【副题名】多尺度计算模拟方法的应用

【摘要】 氢,由于具有众多优异特性如洁净、燃烧后生成物为水、基本实现温室气体和污染物的零排放,被誉为21世纪的能源,许多国家都在加紧部署、实施氢能战略。天然气的主要成分为甲烷,由于具有清洁高效等优势,也成为当前最有前景的新型能源之一。当前,制约氢气、甲烷等新型可替代能源实际利用的主要问题之一即贮存问题,因此开发性能优越、安全性高的氢气和甲烷存储材料迫在眉睫。高压压缩是目前常用的氢气和甲烷存储方式,但该方法存在耗能及设备昂贵等缺点。吸附存储,由于具有可逆吸附等优势,被公认为未来最有前景的能源存储方式之一。二氧化碳是对气候变化影响最大的气体,它产生的增温效应占所有温室气体总增温效应的63%。为了减少二氧化碳的排放,人们必须从工业废气和机动车尾气中捕获和存储二氧化碳。化学吸收法是工业上常用的二氧化碳脱除技术,但该技术必须使用大型设备,且化学吸收剂本身存在腐蚀性。因此,吸附法捕获和存储二氧化碳技术目前受到人们的广泛关注。本文采用“模拟合成”(simulating synthesis)的方法,研究了三维共价有机骨架材料(covalent organic frameworks, COFs)及其Li掺杂复合材料的氢气、甲烷和二氧化碳存储性能,并设计了硅纳米管(silicon nanotube, SiNT)和笼状硅富勒烯氢化物(Li-coated silicon fullerene hydride, Li12Si60H60)两种新型储氢材料。本文的模拟合成方法是指采用结合第一性原理计算和分子模拟技术的多尺度计算模拟方法设计和筛选纳米功能材料,如高效气体吸附剂等。第一性原理方法是基于电子水平的研究方法,在研究纳米材料的电子结构、成键特性、光谱性质(如红外、拉曼和核磁共振谱等)等方面应用广泛；而分子模拟技术是以分子(原子)之间的相互作用为基础,进行统计力学模拟计算从而获得体系的宏观性质。而表述分子(原子)之间相互作用势的分子力场成为联接第一性原理计算和分子模拟两种计算尺度的桥梁。结合第一性原理计算和分子力场技术,当前分子模拟技术在新型纳米材料的设计、纳米多孔材料的气体吸附与分离方面获得广泛应用。本文的模拟合成方法主要分以下四个步骤：(1)通过第一性原理计算设计和搭建新型纳微结构材料；(2)采用第一性原理计算方法获得吸附质与新型纳微结构材料的相互作用；(3)通过力场拟合将两者的相互作用模型化、参数化,从而获得吸附质与主体材料的相互作用力场；(4)将上述工作获得的力场参数植入分子模拟中,运用分子模拟技术预测气体在新型纳微结构材料中的吸附、扩散和分离等性质。上述模拟合成方法可以从理论上预测纳微结构材料的气体吸附特性,筛选高效气体存储、分离材料,为实验工作提供理论指导与依据。三维共价有机骨架材料(covalent organic frameworks, COFs)是一类新型纳微结构多孔材料,由于具有低密度、高比表面等特点成为近年来储氢研究的热门材料。依据上述多尺度计算模拟方法,本文首先采用第一性原理计算获得了氢气与COFs的相互作用势能,然后通过拟合力场构建氢气与COFs的相互作用力场。在此基础上,通过巨正则系综Monte Carlo (grand canonical ensemble Monte Carlo, GCMC)模拟计算了不同热力学条件下COFs的氢气存储容量。研究结果显示,在温度为77时,三维COFs具有较高的氢气负载量,属于目前储氢性能最佳的吸附材料之列。在温度为77K和压力为10MPa时,COF-105和COF-108的氢气总质量吸附容量分别达到18.05wt%和17.80wt%,其最大超额吸附容量分别达到10.31wt%和10.26wt%,上述结果与实验数据吻合较好。由于COFs的氢气负载量在室温下仍未达到美国能源部提出的6wt%的存储标准,本文进一步研究了COFs的表面掺杂改性对其储氢性能的影响。金属Li具有质量轻、易失电子等特点,是目前纳米多孔材料改性的理想候选材料。本文的研究结果表明,当Li原子负载在COFs表面时,由于Li向主体材料的电荷转移作用而易形成阳离子,而负载在表面的Li阳离子有助于提高主体材料的吸氢能力。利用拟合获得的氢气与负载在COFs表面的Li阳离子的相互作用力场,本文通过GCMC模拟预测了室温下掺杂Li的COFs复合材料的储氢性能。研究结果表明,Li掺杂改性显著提高了COFs的氢气存储容量。在温度为298K和压力为10MPa时,掺杂Li的COF-105和COF-108的氢气总质量存储容量分别达到6.84和6.73wt%,达到了美国能源部的标准。综上所述,本文模拟合成的掺杂金属Li的COFs复合材料是非常有应用前景的储氢材料,本工作可为COFs的表面改性提供理论依据。甲烷在纳微结构材料中的吸附是近年来甲烷存储研究的热点。采用上述多尺度计算模拟方法,本文研究了三维COFs及其Li掺杂复合材料的甲烷存储性能。首先通过第一性原理计算和力场拟合,获得甲烷与COFs及其Li掺杂复合材料的相互作用力场,然后通过GCMC模拟预测了COFs及其Li掺杂复合材料的甲烷存储性能。研究结果表明,室温下三维COFs表现出较高的甲烷吸附能力,超过了目前甲烷存储性能最优的MOF材料(PCN-14)。在温度为298K和压力为3.5 MPa时,COF-102和COF-103的超额质量吸附容量分别达到17.72wt%和16.61wt%,且与实验结果一致。研究进一步发现,负载在COFs表面的Li阳离子对甲烷分子表现出以London色散和诱导偶极作用为主的强吸引作用,Li掺杂改性有助于提高COFs的甲烷存储性能。掺杂Li的COF-102和COF-103的超额质量吸附容量分别达到31.35wt%和30.98wt%,其超额体积吸附容量分别达到303v(STP)/v和290v(STP)/v,远超过了美国能源部制定的甲烷存储标准(180v(STP)/v)。上述结果表明,Li掺杂改性显著提高了COFs的甲烷吸附能力,掺杂Li的COFs复合材料在甲烷存储方面应用前景广阔。采用上述多尺度计算模拟方法,本文还研究了硅纳米管(silicon nanotube, SiNT)阵列在室温下的储氢性能。首先通过第一性原理计算和力场拟合得到了氢气与SiNT的相互作用力场,然后采用GCMC模拟预测了不同排列方式的SiNT阵列的储氢性能。为了与碳纳米管(carbon nanotube,CNT)进行比较,本文还模拟了同等条件下氢气在同管径的CNT阵列中的吸附。研究结果显示,由于硅原子的核外电子云密度比碳原子大,SiNT对氢气的色散作用大于同管径的CNT。计算得到,氢气与(9,9)型SiNT的结合能在管内和管外分别达到-2.14和-1.87 kcal/mol,比同管径的(14,14)型CNT分别提高了20%和26%。在温度为298K时,在同等条件下,SiNT阵列的储氢容量明显高于同管径的CNT阵列。在压力为2、6和10MPa时,三角排列的(9,9)型SiNT阵列的氢气质量存储容量分别达到1.30、2.33和2.88wt%,比同管径的(14,14)型CNT阵列分别高106%、65%和52%。上述结果表明,SiNT的储氢性能高于同管径的碳纳米管,是未来比较有前景的储氢候选材料。基于以上工作,本文进一步采用模拟合成的方法设计了具有优良储氢性能的多孔笼状硅富勒烯氢化物(Li-coated silicon fullerene hydride,Li12Si60H60)。首先通过第一性原理计算研究了单个和多个Li原子在Si60H60表面的吸附,搭建了Li12Si60H60的结构并确定其稳定性。然后通过第一性原理计算和力场拟合构建了氢气与Li12Si60H60的相互作用力场,并采用GCMC模拟预测了氢气在Li12Si60H60中的吸附容量。研究结果显示,由于Li-Li原子间相对较低的结合能以及Si-H键的抑制作用,负载在Si60H60笼外的Li原子倾向于分散在五元环位而不是形成团簇。模拟结果显示,修饰在Si60H60笼外的Li阳离子能够显著提高主体材料的储氢容量。在温度为77K时,Li12Si60H60阵列的氢气超额吸附容量的最高值达到7.46wt%。该结果表明,Li12Si60H60阵列是目前非常有前景的储氢材料,该工作可为新型储氢材料的设计提供理论指导。二氧化碳捕获是应对当前全球气候变化的重大课题。本文采用多尺度计算模拟方法,系统地比较了碱金属(Li、Na和K)、碱土金属(Be，Mg和Ca)以及过渡金属(Sc和Ti)在COFs表面的吸附行为,以及上述金属掺杂对二氧化碳吸附捕获的影响。计算结果表明,Li是理想的COFs表面修饰材料。与碱金属Na和K相比,Li与表面的结合最稳定,且具有质量轻、易失电子、在表面的吸附位多等优点,且对二氧化碳分子的吸附介于物理吸附和化学吸附之间。而Na与K不仅质量重,在表面的吸附稳定性差,且与二氧化碳分子的结合能远低于Li。过渡金属Sc和Ti尽管在COFs表面的吸附更稳定,且易形成阳离子,但对二氧化碳分子的结合能高于化学吸附的下限,在室温下可能存在脱附难等问题。通过第一性原理计算和力场拟合,进一步获得了二氧化碳与COFs以及掺杂在表面的Li原子的相互作用力场参数,预测了室温下COFs及其Li掺杂复合材料的二氧化碳负载量。结果表明,掺杂金属Li后,极大的提高了COF-102的二氧化碳存储容量。在T=298K和压力为0.1 MPa时,COF-102的二氧化碳超额吸附容量达到1237.6mg/g,位于当前最优的吸附捕获纳微材料之列。由此可见,掺杂金属Li的三维COFs复合材料在室温下表现出优良的二氧化碳存储性能,是非常有应用前景的二氧化碳捕获与存储材料。更多还原

【Abstract】 Hydrogen is recognized as one of main energy resources in the 21st century, due to its outstanding features, such as clean, zero-emission of greenhouse-gas and pollutants etc. Many countries are deploying and implementing their hydrogen energy strategies urgently. In addition, as a clean and highly-efficient feed stock, natural gas also becomes one of the most promising new energy resources, in which methane is the major component. At present, one of the main problems that impede practical uses of hydrogen and methane is the storage of these two gases. As a result, the development of new materials that can store hydrogen and methane in high capacity and good safety is an important topic. Nowadays, high-pressure compression is still the main storage method of hydrogen and methane, although this method is of some drawbacks, for example, consuming too much energy and requiring expensive equipment. In contrast, owing to the adsorption reversibility and other advantages, adsorption storage is believed as one of the most promising ways for storing hydrogen and methane in future.As is well known, carbon dioxide has the greatest impact on climate changes, and generates about 63% of the total warming effect of green-house gases. In order to reduce the emission of carbon dioxide, people have to capture and store carbon dioxide from industrial gaseous waste and motor vehicle exhaust. Chemical absorption method is the most commonly used technique of capturing carbon dioxide industrially. However, on one hand, this technique needs big equipments generally; on the other hand, the chemical absorbents are corrosive. Therefore, capturing and storing carbon dioxide with adsorption technique have attracted lots of research interest recently.In this work, a "simulating synthesis" approach is adopted to develop several high performance adsorbents for hydrogen, methane and carbon dioxide storage including three-dimensional covalent organic frameworks (3D COFs) and their Li-doped compounds, silicon nanotubes (SiNT) as well as Li-coated silicon fullerene hydride (Li12Si60H60). In this simulating synthesis approach, the design and modification of effective gas adsorbents are carried out by using a multi-scale method, which combines the first principles calculation and molecular simulation method. The first principles calculation is based on the level of electrons, and is widely used for studying electron structures, bonding features, spectral properties (such as infrared, Raman and nuclear magnetic resonance spectroscopy, etc.); while the molecular simulation method is based on the interaction between molecules(atoms), and achieves the macroscopic properties by performing simulations in statistical mechanics. Therefore, a molecular force field that describes the interaction between the molecules of the adsorbate and the atoms of the host material bridges these two levels of theoretical methods. Combining the first principles calculation and molecular force field, the molecular simulation method is then used for obtaining adsorption uptakes. In general, this simulating synthesis approach consists of the following four steps. (1) To design the nano micro-porous materials by using the first principles calculation. (2) To compute the interaction energies between the adsorbate and adsorbent by using the first principles calculation. (3) To fit the interaction energies between the molecules (atoms) of the adsorbate and adsorbent to a force field, and thus the parameters of the force field can be obtained. (4) To obtain the macro-properties by performing molecular simulation. This simulating synthesis approach can be used to predict the adsorption properties of an adsorbent, to screen effective materials for storage and separation of gases, and to provide suggestions for their preparations. To implement the task, this work consists of the following parts.The synthesis of novel crystalline micro-porous materials,3D COFs, has been achieved recently. Due to their unique characteristics, such as extremely low densities and high surface areas,3D COFs have attracted a lot of research interest so far. With the multi-scale method described above, the potential energies between H2 and the COFs were first calculated. Then, the force fields describing the interaction between H2 and the COFs were constructed by the force field fitting. The H2 uptakes of the COFs were predicted by performing grand canonical ensemble Monte Carlo (GCMC) simulation in different thermodynamic conditions. It is found that, at T=77 K, the COFs exhibit high hydrogen uptakes, and belong to the most promising hydrogen adsorbents at present. At T= 77 K and p= 10 MPa, COF-105 and COF-108 show the total hydrogen gravimetric uptakes of 10.05 wt% and 17.80 wt%, and their maximum excess uptakes reach 10.31 wt% and 10.26 wt%, respectively, which are in agreement with those in experiment. Since the hydrogen uptakes of the COFs at room temperature do not meet the storage capacity target of 6 wt% set by the US Department of Energy (DOE) for hydrogen applications, the surface modification of the COFs was carried out to promote their hydrogen uptakes. As is well known, lithium is an ideal element for the surface modification of nano porous materials, because of its unique characteristics, such as light weight and easy to lose its outer valance electron. It is further found that, a single Li atom can be positively charged due to the charge-transfer to the host materials, which can help to improve the affinity of the host materials to hydrogen molecules. In addition, with the fitted force fields that describe the interaction between hydrogen and Li cations doped in the COFs, the hydrogen uptakes of the Li-doped COFs were predicted by performing GCMC simulations. The results show that, at T= 298 K and p= 10 MPa, the Li-doped COF-105 and COF-108 exhibit the total hydrogen uptakes of 6.84 wt% and 6.73 wt%, respectively, which meet the US DOE target for hydrogen storage. Consequently, it is concluded that the Li-doped COFs composites are very promising hydrogen storage materials in future.By using the multi-scale method mentioned above, the storage of methane in the COFs and their Li-doped compounds was then studied. Through the force fields between methane and the non-doped and Li-doped COFs, their methane uptakes were predicted with GCMC simulation. The results show that, the 3D COFs exhibit high methane storage capacities, and exceed the best MOF material reported in the literature, PCN-14. At T= 298 K and p=3.5 MPa, the methane uptakes of COF-102 and COF-103 reach 17.72 wt% and 16.61 wt%, respectively, which are also in good agreement with the experimental results. Moreover, Li atoms loaded in the COFs exhibits strong affinity to methane due to the London dispersion and induced-dipole interaction. The results from GCMC simulation indicate that Li-doping enhances the methane storage performance of the COFs significantly. At T= 298 K and p= 3.5 MPa, the Li-doped COF-102 and COF-103 show the excess methane gravimetric uptakes of 31.35 wt% and 30.98 wt%, while their excess volumetric uptakes are 303 v(STP)/v and 290 v(STP)/v, respectively. Obviously, the volumetric uptakes of methane in the Li-doped COFs significantly exceed the US DOE target of 180 v(STP)/v for methane storage. These results show that the Li-doped COFs in this work are also promising methane storage materials.In addition, novel silicon nanotubes (SiNTs) were synthesized in this work by using the multi-scale method mentioned above. First, the force filed parameters that describe the interaction between hydrogen and the SiNTs were achieved by the first principles calculation and force field fitting. Then the GCMC simulation was performed to predict the hydrogen storage performance of different SiNT arrays at room temperature. For comparison, the adsorption of hydrogen in the carbon nanotubes (CNTs) of the same diameter was also investigated. The results show that, due to the larger density of the extra-nuclear electron cloud of a silicon atom than that of a carbon atom, SiNT shows stronger dispersion interaction to hydrogen than its iso-diameter CNT. The first principles calculation proves that, compared to the (14,14) CNT, the binding energies between a hydrogen molecule and (9,9) SiNT exhibit significant increases of 20% and 26% for inside and outside adsorption, respectively. The results from GCMC simulation show that the hydrogen uptakes of SiNT arrays were significantly higher than the iso-diameter CNT arrays in the same conditions. At p= 2,6 and 10 MPa, the triangular-arranged (9,9) SiNT array shows the hydrogen storage capacities of 1.30,2.33 and 2.88 wt%, respectively, which are 106%,65% and 52% higher than those for the (14,14) CNT array. These results indicate that SiNT exceeds CNT and becomes a more promising candidate for hydrogen storage.Moreover, a novel hydrogen storage material, Li-coated silicon fullerene hydride (Li12Si60H60), was designed. First, the adsorption sites and binding energies of single and multiple Li atoms on Si60H60 were studied with the first principles calculations, so that the reasonable structure of Li12Si60H60 was determined. Then, the force field that describes the interaction between hydrogen and Li12Si60H60 was constructed. Furthermore, the hydrogen uptakes of the Li12Si60H60 array were investigated in different conditions. The results indicate that Li atoms prefer to be adsorbed in the Si60H60 cage by binding with the five-membered rings rather than forming clusters, because of the relatively low Li-Li binding energy and the inhibition of the Si-H bonds. The results from GCMC simulations show that the coated Li atoms outside the Si60H60 cage are positively charged, and can enhance the hydrogen adsorption capacity of Si60H60 cage significantly. At T= 77 K, the maximum excess hydrogen uptake of Li12Si60H60 array reaches 7.46 wt%. This result shows that, Li12Si60H60 is also a very promising adsorbent for hydrogen storage.Carbon dioxide capture is an urgent task for dealing with climate change nowadays. In this work, COFs doped with alkaline metals (Li, Na and K) and alkaline-earth metals (Be, Mg and Ca) as well as transition metals (Sc and Ti) were studied systematically. Their impacts on capturing and storing carbon dioxide were simulated. Our results indicate that Li is an ideal material for modification of COFs due to its light weight, easily losing the valence electron and multiple adsorption sites in COFs. In addition, charged Li can bind with carbon dioxide in a way between physical adsorption and chemical adsorption. On the contrary, Na and K are heavy in weight and unstable when loaded in COFs. Besides, the positively charged Na and K show much weaker affinity to carbon dioxide than Li. It is also found that the transition metals such as Sc and Ti can be adsorbed in COFs very stably, and tend to lose their valence electrons in COFs. However, the binding energies between carbon dioxide and the Sc/Ti cation exceed the lower limt of chemisorption. That is to say, the adsorbed carbon dioxide by Sc/Ti cations has the difficulty of desorption at room temperature. By performing our first-principles calculation and force field fitting, we then obtained the force field parameters between carbon dioxide and the COFs as well as the doped Li cation. Our GCMC simulations show that the Li-doped COF-102 show an excess CO2 uptake of about 1237.6 mg/g at T= 298 K and p= 0.1 MPa, which is among the highest scores of the nanoporous adsorbents before. It indicates that the Li-doped COFs belong to the most promising CO2 adsorbents at present.更多还原