液态Al80Fe20合金的中程有序结构

利用分子动力学模拟（MD）与X射线衍射实验相结合的方法研究了Al80Fe20合金熔体的微观结构，发现在结构因子的小角部分（Q=17.5 nm－1）出现了一个明显的预峰（FSDP），且模拟值与实验值吻合得很好，这被认为是熔体中存在中程有序（MRO）的标志.通过对化学短程序参数(CSRO)α及Bhatis Thornton（BT）结构因子的分析计算，发现熔体中存在较强的化学序，并认为正是这种化学序导致了中程有序结构的产生. Faber Ziman（FZ）偏结构因子的SFe－Fe(Q)和SAl－Fe(Q)在Q=17.5 nm－1处分别存在最大值与最小值，也是熔体中存在着超结构的表征.给出了体系的配位数及代表中程有序的原子团簇的结构模型.     

液态Al80Fe20在快速冷却中的MD模拟

通过Tight binding (TB)势的分子动力学模拟分析了Al80Fe20合金熔体的中程有序结构以及快速凝固过程中体系微观结构的演变规律，发现在倒空间，其结构因子的小角部分都出现了一个预峰.在平衡态（1450 K），模拟结果得到了X射线衍射实验的进一步印证，这被认为是体系中存在中程有序的标志.随着温度的降低，预峰的高度逐渐增大，说明体系中原子团簇尺寸越来越大.通过运用键对分析技术和键取向序参数，发现体系中存在着大量的二十面体短程有序单元.在对平衡态化学短程序参数α的计算过程中，得到了负值的α，证实了熔体中存在着较强的化学序.在FZ偏结构因子中， SAl Fe(Q)在400 K的第二峰较之SAl Al(Q)和SFe Fe(Q)发生了更为明显的劈裂，表明在非晶形成能力方面， Al、Fe元素之间的轨道杂化作用比Al元素或Fe元素单独作用要强.从BT偏结构因子中我们也发现，表征化学序的SCC(Q)在17.5 nm－1左右处出现了第一峰，而这个峰位恰是总结构因子中出现预峰的峰位.因此可推断，正是体系中的化学序导致了中程有序结构的产生.     

Cu-12％Al合金熔体内中程有序原子团簇

通过高温X射线衍射仪研究了Cu 12％Al（质量分数,下同）合金熔体结构,并用纯铜作对比实验.在1250 ℃时,发现Cu 12％Al合金熔体结构因子曲线上18.5 nm－1位置有预峰出现.随着温度的下降,预峰变得更加明锐.预峰的出现是液体中存在中程有序的标志.通过熔态旋淬法获得该合金的快速凝固条带,对条带进行固态X射线衍射分析,其结构是具有有序体心立方晶格的Cu3Al.熔体内的中程有序结构单元尺寸与快凝Cu3Al（111）晶面面间距d111数值一致.由双体分布函数得到的最近邻原子距离、配位数.结合原子团簇结构单元的几何模型,计算得出该体心立方的棱边长(a=3.00 10－10m)与文献中所提供的固态晶格常数（a=2.95 10－10 m）基本吻合.可证明该合金熔体中存在以DO3结构为基本单元的中程有序原子团簇,在液相线以上200 ℃温度范围内这种中程有序都能稳定存在,并随着温度的下降,中程有序的相关尺寸逐渐增大.     

Al-TM合金熔体的中程有序结构及其演化规律

用高温溶体X射线衍射仪对Al-TM（TM = Fe, Ni）合金的熔体结构进行了实验 研究。在Al-1.0at%Fe（原子百分比）合金的结构因子曲线S（Q）上,温度为670 ℃时,存在预峰,即中程有序结构,当温度过热到1400 ℃时,预峰消失。Al- 20at%Ni合金熔体温度在≤1300 ℃范围内,存在预峰;温度高于1400 ℃时,预峰 逐渐消失。这些实验结果表明,Al-Ni, Al-Fe合金熔体中除存在短程有序结构外, 还存在中程有序结构。这种中程有序结构的存在与消失与温度有关。在相似的条件 下,Al-Si合金中没有发现中程有序结构。本文对中程有序结构的演化机理也作了 探讨。     

Fe68Si32合金液态结构与固态组织的相关性

利用液态X射线衍射仪研究了Fe68Si32合金的液态结构，获得了结构因子、径向分布函数、原子间的最近邻距离和配位数.结果表明，在1250～1450 ℃范围内液态合金的最近邻距离为0.259～0.260 nm，配位数为10.3(±0.2)；液态合金的结构因子在Q=15.5 nm－1处有一明显的预峰存在.根据预峰的特性，建立了Fe68Si32熔体的结构模型，即体心立方结构的有序Fe3Si原子团以共面的方式形成Fe3Si面心立方超结构(DO3)；合金在1250 ℃的径向分布函数的Gauss分解结果与合金的面心立方模型吻合较好.预峰的产生是面心立方超结构原子团中Si原子之间相互关联的外在表现. Fe68Si32合金的固态X射线衍射显示合金中含有Fe3Si相，而且其特征峰与合金的结构因子的峰位基本一致，表明Fe68Si32合金的液固态结构之间联系紧密.     

Fe68Si32合金液态结构与固态组织的相关性

利用液态X射线衍射仪研究了Fe68Si32合金的液态结构,获得了结构因子、径向分布函数、原子间的最近邻距离和配位数.结果表明,在1250～1450℃范围内液态合金的最近邻距离为0.259～0.260nm,配位数为10.3(±0.2);液态合金的结构因子在Q=15.5nm-1处有一明显的预峰存在.根据预峰的特性,建立了Fe68Si32熔体的结构模型,即体心立方结构的有序Fe3Si原子团以共面的方式形成Fe3Si面心立方超结构(DO3);合金在1250℃的径向分布函数的Gauss分解结果与合金的面心立方模型吻合较好.预峰的产生是面心立方超结构原子团中Si原子之间相互关联的外在表现.Fe68Si32合金的固态X射线衍射显示合金中含有Fe3Si相,而且其特征峰与合金的结构因子的峰位基本一致,表明Fe68Si32合金的液固态结构之间联系紧密.     

CuAlNi合金的液态结构

利用高温X射线衍射研究了CuAlNi合金液态结构，发现结构因子上有明显的预峰出现，随着温度的升高,Cu75Al25合金熔体结构因子的预峰减弱, 直到1300 ℃预峰消失,这表明中程有序原子团簇可以在高于液相线温度约250 ℃范围内存在,原子团簇的大小和数量都随着Ni的加入而增加，Ni增强原子之间的交互作用,有利于中程有序的形成,根据预峰的特性，提出CuAlNi合金液态结构的原子模型，即由八面体结构共享顶点形成的原子团簇与无规密堆积原子分布区域组成.     

Cu3Au合金熔体的中程有序结构

采用紧束缚(Tight-binding,简称TB)原子间相互作用势,利用分子动力学模拟(Molecular Dynamics,简称MD)的方法研究了Cu₃Au合金熔体的微观结构,发现Cu₃Au合金体系在高于熔点以上的400K～500K的温度范围内,结构因子上存在预峰,且在此范围内,随着温度的升高,预峰的强度逐渐降低。当温度超过1764K时,预峰开始消失。通过Lorentzian近似和Scherrer公式,可得预峰所对应的原子团簇尺寸位于1.35nm～1.85nm之间,属于中程有序结构范畴。Bhatia-Thornton(BT)偏结构因子表明,预峰的起源应归咎于体系中存在的化学序与拓扑序。BT形式的双体分布函数也证实,体系中存在着较强的化学序,且团簇中原子间的配位倾向于同类原子的配位。     

热浸镀过程上浮熔渣形核期原子团簇研究

利用θ-θ X射线衍射研究了Al₃Fe熔体结构, 结果显示熔体中Al和Fe原子之间存在较强的相互作用, 存在着类似化合物结构的原子团簇. 这说明当Zn-55Al热浸镀熔体中Fe含量积累到一定程度, 在熔体中优先形成类似Al₃Fe的原子团簇, 进而形成熔渣. 在Al的有效硬球直径减小而Fe的增大情况下, 利用Percus-Yevick硬球模型可以较好地拟合结构因子实验曲线的第一峰, 从而给出了1550℃ Al₃Fe的密度为3.65 g/cm³, 它小于实际热浸镀Zn-55Al熔体的密度3.7 g/cm³, 可较好地解释“熔渣”上浮的现象.     

Cu-Ni合金结构的分子动力学模拟

采用EAM作用势对Cu-Ni合金的结构特性进行了MD模拟研究.通过FZ结构因子可发现,Cu含量的变化对结构因子的波动影响很小,键取向序参数和键对也表现出相似的变化规律,这表明液态Cu-Ni合金对成份变化不敏感,体系中的化学序较弱.将Cu70Ni30合金熔体的FZ结构因子与Waseda的实验结果进行对比,发现二者吻合得较好,表明EAM势可以很好地描绘Cu-Ni合金的结构特性.在快速冷却过程中,除了Cu20Ni80合金外,其他合金成份的双体分布函数的第二峰都发生了劈裂,标志着体系最终形成了非晶结构,而Cu20Ni80合金的双体分布函数却表现出晶体峰的特征.通过对键取向序参数、键型指数以及铜镍原子的有效扩散系数的分析表明,在快速冷却过程中,Cu20Ni80合金最终形成了hcp晶体结构.     

Experimental evidences for molecular origin of low-Q peak in neutron/x-ray scattering of 1-alkyl-3-methylimidazolium bis(trifluoromethanesulfonyl)amide ionic liquids

Short- and long-range liquid structures of [C(n)mIm(+)][TFSA(-)] with n = 2, 4, 6, 8, 10, and 12 have been studied by high-energy x-ray diffraction (HEXRD) and small-angle neutron scattering (SANS) experiments with the aid of MD simulations. Observed x-ray structure factor, S(Q), for the ionic liquids with the alkyl-chain length n > 6 exhibited a characteristic peak in the low-Q range of 0.2-0.4 A?(-1), indicating the heterogeneity of their ionic liquids. SANS profiles I(H)(Q) and I(D)(Q) for the normal and the alkyl group deuterated ionic liquids, respectively, showed significant peaks for n = 10 and 12 without no form factor component for large spherical or spheroidal aggregates like micelles in solution. The peaks for n = 10 and 12 evidently disappeared in the difference SANS profiles ΔI(Q) [=I(D)(Q) - I(H)(Q)], although that for n = 12 slightly remained. This suggests that the long-range correlations originated from the alkyl groups hardly contribute to the low-Q peak intensity in SANS. To reveal molecular origin of the low-Q peak, we introduce here a new function; x-ray structure factor intensity at a given Q as a function of r, S(Q) (peak)(r). The S(Q) (peak)(r) function suggests that the observed low-Q peak intensity depending on n is originated from liquid structures at two r-region of 5-8 and 8-15 A? for all ionic liquids examined except for n = 12. Atomistic MD simulations are consistent with the HEXRD and SANS experiments, and then we discussed the relationship between both variations of low-Q peak and real-space structure with lengthening the alkyl group of the C(n)mIm.     

Nanoscale heterogeneity in alkyl-methylimidazolium bromide ionic liquids

High-energy x-ray diffraction measurements and atomistic molecular dynamics (AMD) numerical simulations have been carried out on 1-alkyl-3-methylimidazolium bromide ionic liquids, C(n)mimBr, with n = 2, 4, and 6. Excellent agreement between experiment and simulation is obtained, including the region of the low-Q peak that has proved problematic in previous work in the literature. In the partial structure analysis of the AMD results, a distinct peak develops at the leading edge of the ring-ring pair distribution function and shifts to lower r with increasing alkyl chain length, indicating that the preferential parallel and antiparallel alignment of neighboring cation rings plays a larger role with increasing chain length. The ring-ring, anion-anion, and ring-anion partial structure factors are dominated by strong charge-ordering peaks around 1.1 A?(-1), corresponding to a distance between neighboring polar entities of D(2) = 5.7 A?. In contrast, the tail-tail S(Q) is dominated by the low-Q peak that rises and moves to lower Q with increasing chain length; the length scale of this structural heterogeneity D(1) increases from about 10 A? in C(2)mimBr to 14.3 A? in C(4)mimBr and 18.8 A? in C(6)mimBr. Both the length scale of the structural heterogeneity and its anomalous temperature dependence in the C(n)mimBr liquids studied here show considerable similarity to results in the literature for C(n)mimPF(6) liquids, indicating a remarkable insensitivity to the form and size of the anion. Our results are consistent with the concept of nanoscale heterogeneity with small, crystal-like moieties.     

The liquidlike ordering of lipid A-diphosphate colloidal crystals: the influence of Ca2+, Mg2+, Na+, and K+ on the ordering of colloidal suspensions of lipid A-diphosphate in aqueous solutions

A comprehensive study was performed on electrostatically stabilized aqueous dispersion of lipid A-diphosphate in the presence of bound Ca2+, Mg2+, K+, and Na+ ions at low ionic strength (0.10-10.0-mM NaCl, 25 degrees C) over a range of volume fraction of 1.0 x 10(-4)< or =phi< or =4.95 x 10(-4). These suspensions were characterized by light scattering (LS), quasielastic light scattering, small-angle x-ray scattering, transmission electron microscopy, scanning electron microscopy, conductivity measurements, and acid-base titrations. LS and electron microscopy yielded similar values for particle sizes, particle size distributions, and polydispersity. The measured static structure factor, S(Q), of lipid A-diphosphate was seen to be heavily dependent on the nature and concentration of the counterions, e.g., Ca2+ at 5.0 nM, Mg2+ at 15.0 microM, and K+ at 100.0 microM (25 degrees C). The magnitude and position of the S(Q) peaks depend not only on the divalent ion concentration (Ca2+ and Mg2+) but also on the order of addition of the counterions to the lipid A-diphosphate suspension in the presence of 0.1-microM NaCl. Significant changes in the rms radii of gyration (R2G) 1/2 of the lipid A-diphosphate particles were observed in the presence of Ca2+ (24.8+/-0.8 nm), Mg2+ (28.5+/-0.7 nm), and K+ (25.2+/-0.6 nm), whereas the Na+ salt (29.1+/-0.8 nm) has a value similar to the one found for the de-ionized lipid A-diphosphate suspensions (29.2+/-0.8 nm). Effective particle charges were determined by fits of the integral equation calculations of the polydisperse static structure factor, S(Q), to the light-scattering data and they were found to be in the range of Z*=700-750 for the lipid A-diphosphate salts under investigation. The light-scattering data indicated that only a small fraction of the ionizable surface sites (phosphate) of the lipid A-diphosphate was partly dissociated (approximately 30%). It was also discovered that a given amount of Ca2+ (1.0-5.0 nM) or K+ (100 microM) influenced the structure much more than Na+ (0.1-10.0-mM NaCl) or Mg2+ (50 microM). By comparing the heights and positions of the structure factor peaks S(Q) for lipid A-diphosphate-Na+ and lipid A-diphosphate-Ca2+, it was concluded that the structure factor does not depend simply on ionic strength but more importantly on the internal structural arrangements of the lipid A-diphosphate assembly in the presence of the bound cations. The liquidlike interactions revealed a considerable degree of ordering in solution accounting for the primary S(Q) peak and also the secondary minimum at large particle separation. The ordering of lipid A-diphosphate-Ca2+ colloidal crystals in suspension showed six to seven discrete diffraction peaks and revealed a face-centered-cubic (fcc) lattice type (a=56.3 nm) at a volume fraction of 3.2 x 10(-4)< or =phi< or =3.9 x 10(-4). The K+ salt also exhibited a fcc lattice (a=55.92 nm) at the same volume fractions, but reveals a different peak intensity distribution, as seen for the lipid A-diphosphate-Ca2+ salt. However, the Mg2+ and the Na+ salts of lipid A-diphosphate showed body-centered-cubic (bcc) lattices with a=45.50 nm and a=41.50 nm, respectively (3.2 x 10(-4)< or =phi< or =3.9 x 10(-4)), displaying the same intensity distribution with the exception of the (220) diffraction peaks, which differ in intensity for both salts of lipid A-diphosphate.     

Neutron scattering study of the structural change induced by photopolymerization of AOT/D2O/dodecyl acrylate inverse microemulsions

Small-angle and ultrasmall-angle neutron scattering (SANS/USANS) measurements were used to determine the structural changes induced by photopolymerization of AOT/D2O/(dodecyl acrylate) inverse microemulsion systems. Scattering profiles were collected for the initial microemulsions and the films resulting from photopolymerization of the oil phase. The SANS data for the microemulsions were modeled as spherical, core-shell droplets. Upon polymerization, the clear mircoemulsions formed opaque films. From the SANS/USANS data of the films, it was apparent that this morphology was not preserved upon polymerization; however, it was clearly observed that the formulation of the microemulsion had a large impact on the structure within the films. The Guinier region in the USANS data (2.5 x 10(-5) A(-1) < or = Q < or = 5.3 x 10(-3) A(-1)) from the films indicates that very large structures are formed. Simultaneously, a well-defined peak (0.15 A(-1) < or = Q < or = 0.25 A(-1)) in the SANS data indicates that there are also much smaller structures formed. It is proposed that the low-Q scattering arises from aggregation of the nanometer-size water droplets in the microemulsion to form droplets large enough to scatter visible light, while the peak in the high-Q region results from bilayered structures formed by the surfactant.     

Cluster formation in two-Yukawa fluids

We present a different and efficient method for implementing the analytical solution of Ornstein-Zernike equation for two-Yukawa fluids in the mean spherical approximation. We investigate, in particular, the conditions for the formation of an extra low-Q peak in the structure factor, which we interpret as due to cluster formation in the two-Yukawa fluid when the interparticle potential is composed of a short-range attraction and a long-range repulsion. We then apply this model to interpret the small angle neutron scattering data for protein solutions at moderate concentrations and find out that the presence of a peak centered at Q=0 (zero-Q peak) besides the regular interaction peak due to charged proteins implies an existence of long-range attractive interactions besides the charge repulsion.