Initial Stage Of Deposition Research Articles

A novel method used to study growth of ash deposition and in situ measurement of effective thermal conductivity of ash deposit

AbstractAsh deposition always brings boilers some trouble due to fouling or slagging. In this paper, a completely controlled system was developed to study the growth of ash deposit. A novel sampling probe was designed to online measure the heat flux through ash deposit. Additionally, the thickness of ash deposit can be obtained by an online figure collecting system. The results of this research showed that as the thickness of ash deposit increased, the heat flux decreased. It was also found that at the initial stage of ash deposition when the thickness of ash deposit is approximately 1 mm, the heat flux through ash deposit had a sharp reduction. An effective method was attempted to situ measure the effective thermal conductivity of the ash deposit in the simulated combustion flue gas. It was found that temperature of the ash deposit layer had no obvious effect on its value. It was concluded that the structure of ash deposit had no obvious change in a short deposition time of 30 min with varied surface temperatures of the probe head between 400 °C and 600 °C.

Slagging and fouling of Zhundong coal at different air equivalence ratios in circulating fluidized bed

Circulating fluidized bed (CFB) with low operating temperatures may become an appropriate solution to the utilization of high-Na Zhundong coal. In this paper, one kind of Zhundong coal with 3.92% Na2O in coal ash was used as fuel, and its slagging and fouling characteristics at different air equivalence ratios (ERs) were systemically investigated in 0.4t/d CFB test system to determine a better utilization scheme of Zhundong coal. The results show that ash-related problems still occurred when this high-Na coal was used in CFB, which was greatly influenced by ER. At the reducing atmosphere (ER<1.0), slagging obviously occurred in the riser at ∼940°C. Na-based species reacted with the SiO2-rich bed material to form low-melting-temperature sodium silicates (Na2O·nSiO2). The formed sodium silicates covered bed material particles and increased their stickiness. These sticky particles agglomerated and adhered onto the high-temperature metal surfaces by adhesion force or inertial force, resulting in slagging. During slagging process, Ca, Mg and Fe acted as fluxing agents. This type slagging related to the SiO2-rich bed material might be reduced or even avoided via the replacement of the bed material. At the oxidizing atmosphere (ER>1.0), fouling on tail heating surfaces at medium-low temperatures (607–735°C) was the main ash-related problem. The released AAEM-based species were transported towards low-temperature surfaces by the mechanisms of diffusion, condensation and aggregation, or thermophoresis, resulting in a layered deposition. The inner layer of deposits with micron grain sizes, which was mainly composed of Na2SO4 and K2SO4, played a key role at the initial deposition stage. Besides, two kinds of aggregated particles separately rich in alkali silicates and Ca-based species were also important for fouling.

Effect of lead and tin additives on surface morphology evolution of electrodeposited zinc

Effects of Pb and Sn additives on electrodeposition of Zn were investigated for the applications in Zn secondary battery, focused on their roles on morphological evolution. Similarly to well-known effect of Pb addition to smoothly electrodeposit Zn film surface, Sn also exhibited to suppress the formation of mossy structures, which were highly filamentous Zn electrodeposits to cause frequently battery failure. Pb significantly shifted deposition potential of Zn to negative value at concentration less than 1.0mmoldm−3, while Sn showed no substantial effect even at 50mmoldm−3. The morphological evolution analysis demonstrated that the addition of 1.0mmoldm−3 Pb significantly altered nucleation behavior of Zn at initial stages of deposition; stacking of layer-like structures (microsteps) were altered to discrete and uniform nuclei of Zn preferentially oriented to (002) direction. In contrast, with Sn additive, morphological features at the initial stage of deposition were similar to that without additive. However, microsteps evolution was gradually suppressed and relatively strong (002) texture of electrodeposited Zn was mitigated by Sn addition. These results suggested that the beneficial roles of Pb and Sn addition on Zn morphological evolution were ascribed to different reasons; Pb diminished the active growth sites of Zn surface by suppressing the deposition reaction of Zn, while Sn suppressed microsteps evolution by different texture evolution of Zn.

Potentiodynamic Chromium Deposition from Trivalent and Hexavalent Systems on Glassy Carbon Electrodes: Initial Stages and Mechanistic Insights

AbstractDeveloping a functional Cr(III) electrolyte to replace carcinogenic and toxic Cr(VI) electroplating requires a deeper understanding of the deposition mechanisms. Hence, this study investigates and compares the growth of chromium on inert glassy carbon (GC) electrodes from low‐concentration Cr(III) and Cr(VI) electrolytes, using a novel potentiodynamic deposition scheme. This allows the initial stages of deposition to be probed, which are crucial for determining the overall deposition mechanism. Scanning electron microscopy shows that, for both systems, micrometer‐sized hexagonal chromium hydride crystals could be deposited on GC, thus serving to be the first reporting of such structures. In addition, by increasing the deposition time, the changes in a particle structure for both systems were tracked; therefore, possible areas of similarity as well as key differences could be identified, which, when addressed may provide a more viable, environmentally friendly trivalent electrolyte for chromium electroplating.

Simulation of the effect of coke deposition on the diffusion of methane in zeolite ZSM-5

Coke deposition is one of the main reasons for catalyst deactivation, which is often associated with coverage of the active sites and diffusion restrictions. In this article, the self-diffusion coefficient of methane in ZSM-5 (MFI) with different amounts of coke is studied through molecular dynamics simulation combining hard-sphere and pseudo-particle modeling at a gas loading of 4 molecules per unit cell at 723K. The T12 tetrahedral sites are considered as the possible coke deposition sites at which the pore is blocked. It is shown that the self-diffusion coefficient of methane in ZSM-5 without coke (D0) is approximately 7.0×10−9m2/s. Two coke distribution models are proposed to simulate the actual process of coke formation. At the initial stage of coke deposition, the coke with uniform distribution has little influence on the diffusion process. Once the coke deposits are randomly concentrated on some sites and a few pores are therefore completely blocked, the coke has a significant influence on the self-diffusion coefficient, which decreases to 37.0% of D0 almost linearly with increasing coke amount to 20%. The distribution of coke deposition is the result of the coupling of reaction and diffusion processes. These results may be useful for understanding the reaction–diffusion coupling mechanism of the catalyst deactivation.

Structural control of ultra-fine CoPt nanodot arrays via electrodeposition process

CoPt nanodot arrays were fabricated by combining electrodeposition and electron beam lithography (EBL) for the use of bit-patterned media (BPM). To achieve precise control of deposition uniformity and coercivity of the CoPt nanodot arrays, their crystal structure and magnetic properties were controlled by controlling the diffusion state of metal ions from the initial deposition stage with the application of bath agitation. Following bath agitation, the composition gradient of the CoPt alloy with thickness was mitigated to have a near-ideal alloy composition of Co:Pt =80:20, which induces epitaxial-like growth from Ru substrate, thus resulting in the improvement of the crystal orientation of the hcp (002) structure from its initial deposition stages. Furthermore, the cross-sectional transmission electron microscope (TEM) analysis of the nanodots deposited with bath agitation showed CoPt growth along its c-axis oriented in the perpendicular direction, having uniform lattice fringes on the hcp (002) plane from the Ru underlayer interface, which is a significant factor to induce perpendicular magnetic anisotropy. Magnetic characterization of the CoPt nanodot arrays showed increase in the perpendicular coercivity and squareness of the hysteresis loops from 2.0kOe and 0.64 (without agitation) to 4.0kOe and 0.87 with bath agitation. Based on the detailed characterization of nanodot arrays, the precise crystal structure control of the nanodot arrays with ultra-high recording density by electrochemical process was successfully demonstrated.

Simple peptide coacervates adapted for rapid pressure-sensitive wet adhesion.

We report here that a dense liquid formed by spontaneous condensation, also known as simple coacervation, of a single mussel foot protein-3S-mimicking peptide exhibits properties critical for underwater adhesion. A structurally homogeneous coacervate is deposited on underwater surfaces as micrometer-thick layers, and, after compression, displays orders of magnitude higher underwater adhesion at 2 N m-1 than that reported from thin films of the most adhesive mussel-foot-derived peptides or their synthetic mimics. The increase in adhesion efficiency does not require nor rely on post-deposition curing or chemical processing, but rather represents an intrinsic physical property of the single-component coacervate. Its wet adhesive and rheological properties correlate with significant dehydration, tight peptide packing and restriction in peptide mobility. We suggest that such dense coacervate liquids represent an essential adaptation for the initial priming stages of mussel adhesive deposition, and provide a hitherto untapped design principle for synthetic underwater adhesives.

Lithium Deposition from a Piperidinium–based Ionic Liquid: Rapping Dendrites on the Knuckles

AbstractUsing cyclic voltammetry (CV), in‐situ scanning tunneling microscopy (STM) and electrochemical quartz crystal microbalance (EQCM) the initial stages of lithium deposition on Au(111) from a solution of lithium bis(trifluoromethylsulfonyl)imide in the commercially available ionic liquid 1‐methyl‐1‐propylpiperidinium bis(trifluoromethylsulfonyl)imide (MPPipTFSI) were investigated. We could identify three distinct cathodic peaks in the potential range from 0 to −2.5 V (vs. Pt quasi‐reference electrode), corresponding to different lithium deposition modes. While in the potential region of the under‐potential deposition (UPD) (−1.2 to −1.8 V) the growth of monoatomic high islands (300–370 pm) takes place, Li bulk deposition occurs at potentials &lt;−2.3 V. Finally, the third peak at 0 V, which only appears after a previous bulk deposition, is connected to a strand‐like growth of lithium at (111) terraces with a uniform orientation over the whole substrate. Interestingly, once reaching a step‐edge, the one‐dimensional growth continues into the electrolyte, indicating the initial stages of Li dendrite formation.

Fabrication of NPB/Alq3 small‐molecule multilayer structures with suppressed interface mixing by multi‐jet mode electrospray deposition

In this study, we investigated the multi‐jet mode modified electrospray deposition (ESD) production of NPB/Alq3 small‐molecule multilayer structures with suppressed interface mixing, using common solvents. Observations of the initial deposition stage and analysis of the photoluminescence spectra of the multilayer structure were conducted. It was demonstrated that, by optimizing the deposition conditions, surface erosion of the NPB layer and the initiation of interface mixing between the NPB and Alq3 layers could be suppressed. A 55‐nm NPB/60‐nm Alq3 multilayer structure with suppressed intermixing and smooth surface morphology (RMS roughness of 3.9 nm) was obtained at the lowest levels of under‐layer erosion.Schematic of a modified electrospray (nanomist deposition: NMD) system and an NPB/Alq3 multilayer structure deposited by NMD.

Epitaxial GaN layers formed on langasite substrates by the plasma-assisted MBE method

In this publication, the results of development of the technology of the epitaxial growth of GaN on single-crystal langasite substrates La3Ga5SiO14 (0001) by the plasma-assisted molecular-beam epitaxy (PA MBE) method are reported. An investigation of the effect of the growth temperature at the initial stage of deposition on the crystal quality and morphology of the obtained GaN layer is performed. It is demonstrated that the optimal temperature for deposition of the initial GaN layer onto the langasite substrate is about ~520°C. A decrease in the growth temperature to this value allows the suppression of oxygen diffusion from langasite into the growing layer and a decrease in the dislocation density in the main GaN layer upon its subsequent high-temperature deposition (~700°C). Further lowering of the growth temperature of the nucleation layer leads to sharp degradation of the GaN/LGS layer crystal quality. As a result of the performed research, an epitaxial GaN/LGS layer with a dislocation density of ~1011 cm–2 and low surface roughness (<2 nm) is obtained.

The microstructure evolution of hydrogenated microcrystalline germanium promoted by power gradient method

This paper studies the microstructure evolution of hydrogenated microcrystalline germanium (μc-Ge:H) thin films deposited by plasma enhanced chemical vapor deposition (PECVD). There is an amorphous incubation layer formed in the initial deposition stage of μc-Ge:H thin film. It is demonstrated that the thickness of incubation layer can be reduced by high hydrogen dilution and high discharge power method. However, at high hydrogen dilution, the deposition rate of μc-Ge:H appears a sharply decrease. Using a high discharge power can compensate the deposition rate decrease but lead to decrease of average grain size and appearance of micro-void in the μc-Ge:H thin film. In addition, by comparing two thickness groups of μc-Ge:H thin films deposited at different discharge powers, it is noticed that the evolution process relates to the formation of crystal nucleuses. Thus, a power gradient method is proposed to understand the mechanism of nucleation and crystal growth in the initial deposition process of μc-Ge:H films. Finally, by power gradient method, the incubation layer thickness of μc-Ge:H thin films has been decreased to less than 6nm. Moreover, Raman scattering spectra shows a 38nm μc-Ge:H film has a crystal fraction (XC) of 62.4%. Meanwhile, the mobility of TFT devices shows the improved electrical property of μc-Ge:H film deposited by power gradient method.

Is the snail shell repair process really influenced by eggshell membrane as a template of foreign scaffold?

Biominerals are inorganic-organic hybrid composites formed via self-assembled bottom up processes under mild conditions. Biominerals show interesting physical properties, controlled hierarchical structures and robust remodeling or repair mechanisms. Biological processes associated with biominerals remain to be developed into practical engineering processes. Therefore, the formation of biominerals is inspiring for the design of materials, especially those fabricated at ambient temperatures. The study described herein involves the influence of chicken outer eggshell membrane on the type of calcium carbonate (CaCO3) polymorph deposited on the shell of the land snail Helix aspersa during the repair process after an injury. A piece of snail shell was removed by perforating a hole from the largest body whorl. The operated area was left either uncovered or covered with either a thermoplastic flexible polyolefin-based film Parafilm® or a piece of chicken eggshell membrane. The repaired shells of control and experimental animals were analyzed using SEM, EDS, Raman and FTIR spectroscopies. We found that in the presence of eggshell membrane, the polymorph deposited on the substratum during the first hours resembles calcite, the polymorph present in eggshell normal formation, but at 24 and 48h, when snail mantle cells produced their normal organic matrix (mainly β-chitin plus proteins and proteoglycans), the polymorph deposited is aragonite, the characteristic polymorph of Helix shell. Therefore, the eggshell membrane influences the type of polymorph, but only in the initial stages of biomineral deposition, before an organic matrix layer is deposited by the snail.

Influence of MgO underlayers on the structure and magnetic properties of FePt-C nanogranular films for heat-assisted magnetic recording media

In order to optimize the nanogranular structure of FePt-C for heat-assisted magnetic recording media, we investigated the influence of MgO underlayers on the growth of FePt grains in the FePt-C layer. The FePt-C layer was deposited by using the alternating sputtering method, by which FePt and FePt-C layers were alternately deposited. To understand the growth mechanism of the FePt-C layer on the MgO underlayers deposited under various conditions, detailed plan-view and cross sectional transmission electron microscopy observations were made for different film thicknesses. We found that columnar FePt grains grow only when the deposition conditions of the MgO underlayer are optimal. Direct TEM observation of the growth process of the FePt-C layer revealed that the number density of nuclei is sufficient in the initial stage of the film deposition; however, coarsening of the grains after grain impingement causes a substantial decrease in the number density of the FePt grains.

Mathematical modeling and experimental validation of ash deposition in a pulverized-coal boiler

A dynamic mathematical model of the fly ash transport and sticking in a pulverized-coal boiler was developed by a direct simulation Monte Carlo (DSMC) method. In this paper, the critical velocity model was used to calculate ash sticking rate during the initial deposition stage and the viscosity model was used to describe the outer layer deposition. The dynamic process of ash deposition was simulated and the distributions of adhesion particle size were obtained. The simulation results show that the ash deposit thickness increases by 47.52% with the load increases by 20% and the Alkaline Earth Metal (AAEM)-rich enhanced deposits of bulk ash particles are apparently stickier than the other ash particles. The experiments were carried out at two different loads and the morphology of ash deposit samples was analyzed by SEM/EDS. The propotion of particle size in 2μm were most in the deposition inner layer. The mean deposit thickness calculated by the DSMC method agreed with the measured value and the maximum relative error was 5.23%. A parametric study quantified the influence of changed operational conditions and ash material properties on the deposit formation process.

In Situ Electrochemical Measurements of Deposition and Dissolution of Lithium in Li[N(CF3SO2)2]-Glyme Solvate Ionic Liquids

The equimolar mixture of lithium bis(trifluoromethylsulfonyl)amide (LiTFSA)-tetraglyme (G4) has been known as a solvate ionic liquid consisting of [Li(G4)]+ and TFSA–. This solvate ionic liquid contains lithium ions in high concentration (ca. 3 mol dm–3) and has negligible vapor pressure at ambient temperature. It has been reported that the solvate ionic liquid is compatible with sulfur cathode because of the low solubility of lithium polysulfides [1]. To achieve the high energy density of the rechargeable Li-S batteries, it is required to investigate anode materials having high specific capacity such as a lithium anode. In this study, the initial stage of deposition of lithium was investigated in the solvate ionic liquids with different compositions using electrochemical quartz crystal microbalance (EQCM) and optical microscope. LiTFSA-G4 solvate ionic liquids were prepared by mixing LiTFSA and G4 at different molar ratios of LiTFSA (X LiTFSA) in an Ar filled glove box. The potentials of Li(I)|Li were evaluated using a Ag|Ag(I) reference electrode, which consisted of a silver wire immersed in 50.0-50.0 mol% LiTFSA-G4 containing 0.1 mol dm–3 AgCF3SO3. EQCM measurement was performed with an airtight three-electrode cell. A Cu coated quartz crystal electrode was used as a working electrode. Lithium was used as a reference and counter electrode. [Li(G4)]+ is considered a major lithium species in LiTFSA-G4 solvate ionic liquid at X LiTFSA = 50 mol%. On the other hand, both [Li(G4)]+ and [Li(TFSA)2]– are expected to exist in LiTFSA-G4 solvate ionic liquid at X LiTFSA > 50 mol%. It was found that the potential of Li(I)|Li in the LiTFSA-G4 solvate ionic liquid at X LiTFSA > 50 mol% was given by an equilibrium, [Li(TFSA)2]– + e– = Li + 2TFSA–, indicating the active lithium species is [Li(TFSA)2]– rather than [Li(G4)]+. Li was deposited at a current density of –0.01 mA cm–2. The electric charge for deposition was –0.05 C cm–2. In the case of 54.5-45.5 mol% LiTFSA-G4 solvate ionic liquid, Li was deposited sparsely on the Cu electrode surface and grew as a whisker-like shape. On the other hand, whisker-like Li deposit was not observed in 50.0-50.0 mol% LiTFSA-G4 solvate ionic liquid where Li nuclei were uniformly generated on the electrode surface. These results suggest that the nucleation and crystal growth of Li are affected by the composition of the solvate ionic liquids, resulting in the morphology change of Li deposits. Acknowledgment This study was supported by the Advanced Low Carbon Technology Research and Development of Program (ALCA) of the Japan Science and Technology Agency (JST). Reference [1] K. Dokko et al., J. Electrochem. Soc., 160, A1304 (2013).

Numerical simulation of thermal behavior and multicomponent mass transfer in direct laser deposition of Co-base alloy on steel

During direct laser deposition process, rapid melting–solidification and addition of multicomponent powder lead to complex transport phenomena in the melt pool. The thermal behavior and mass transport significantly affect the solidified microstructure and properties of fabricated layer. In this paper, an improved 3D numerical model is proposed to simulate the heat transfer, fluid flow, solidification and multicomponent mass transport in direct laser deposition of Co-base alloy on steel. The solidification characteristics, including temperature gradient (G), solidification growth rate (R) and cooing rate (G×R), can be obtained by transient thermal distribution to predict the morphology and scale of the solidification microstructure. Multicomponent transport equation based on a mixture-averaged approach is combined with other conservation equations. The calculated melt pool geometry and the composition profiles of iron (Fe), carbon (C), cobalt (Co) and chromium (Cr) are compared with the experimental results. The results show that in the initial stage of direct laser deposition, the rapidly mixture of substrate material and added material occur in the melt pool and conduct plays an important role in heat transfer due to the low Peclet number. As the melt pool is developed, the heat and mass transfer in the melt pool are dominated by strong Marangoni convection. An unmixed zone is observed near the bottom of melt pool where the convection is frictionally dissipated due to the presence of solidified dendrites. Since the G/R decreases and G×R increases from the bottom to the top of the solidified track, the morphology of the microstructure changes from planar front to columnar dendrites to equiaxed dendrites and the grain size decreases.

Microstructure evolution and protective properties of TaN multilayer coatings

The protective Tantalum nitride (TaN) multilayer coatings staked by the TaN layers with alternatively structures at various deposition input powers were proposed in this study. The TaN multilayer coatings were sequentially deposited with layers at 75 and 150W by magnetron sputtering. The effects of microstructure and layer thickness for multilayer were investigated. The 20-layer TaN multilayer coating with layer thickness of 50nm exhibited stratified structure with an amorphous/crystalline altering feature, while the 50-layer one with layer thickness of 20nm grew with continuous columnar crystalline structures penetrating through interfaces. Higher hardness and Young's modulus were found for the 50-layer TaN multilayer coating approximately 15.1 and 175.5GPa, respectively. The progressive loading scratch and Rockwell-C adhesion tests were adopted to evaluate the adhesion strength. The anti-corrosion behavior was also analyzed. The amorphous/crystalline structure for the 20-layer TaN multilayer coating showed an elevated critical load for adhesion fracture (Lc2) around 31.6N due to the large volume ratios for the amorphous structure at initial deposition stage. Superior corrosion resistance for the 20-layer TaN multilayer coating was found due to its amorphous structure.

Modeling of ash deposition in a pulverized-coal boiler by direct simulation Monte Carlo method

In this paper, a mathematical model of the fly ash deposition in a 660MW supercritical once-through boiler was developed. The mechanism of particle-particle and particle-wall collisions has been considered. Particle collision is modeled by a direct simulation Monte Carlo (DSMC) method. Three submodels (impaction, thermophoresis, and sticking) were included in the mathematical model. In the ash deposition model, the critical velocity model was used to calculate the ash sticking rate during initial deposition, and the viscosity model was used for the outer layer deposition. An ash deposition probe was inserted into the peephole and ash deposits were collected. The morphology of these samples was analyzed using SEM. The deposit thickness and the distributions of adhesion particle size were obtained by mathematical model. The mean deposit thickness calculated by the DSMC method agreed with the value measured by the ash deposition probe with the relative error within 3.79%. The simulation results show that the deposition inner layer consists of fine particles less than 10μm in size and sticky fine particles were selectively captured during the initial deposition stage. The DSMC method may be a practical alternative for studies of ash deposit formation.

Nanoscale Microstructure and Chemistry of Cu2ZnSnS4/CdS Interface in Kesterite Cu2ZnSnS4 Solar Cells

Sulfurization with various atmosphere and postheat treatments has been reported for earth abundant kesterite Cu2ZnSnS4 (CZTS) preparation as cost‐effective material for next‐generation solar cells. A full understanding of the nanoscale microstructure and chemistry of CZTS/CdS interface obtained from these different fabrication routes is currently lacking, yet is critical to developing optimal processing routes for high‐performance kesterite solar cells. Here, the first detailed investigation of the interfacial microstructure and chemistry of CdS/Cu2ZnSnS4 heterojunctions is presented. For CZTS obtained from sulfurization in a sulfur‐only atmosphere where highly defective surfaces are present, air annealing followed by etching in the initial stage of chemical bath deposition (CBD) process can effectively eliminate interfacial defects and allow the epitaxial growth of CBD‐CdS, improving the minority lifetime, open circuit voltage (VOC), and fill factor (FF) of the devices, while blocking Cd diffusion and deteriorating short circuit current (Jsc). For CZTS from sulfurization in a combined sulfur and SnS atmosphere where CBD‐CdS can directly epitaxially grow on CZTS and Cd‐diffusion is clearly observed, associated devices show the longest lifetime and the highest efficiency of 8.76%. Epitaxial growth of CdS and Cd diffusion into CZTS are found to be two crucial features minimizing interfacial recombination and achieving high‐efficiency devices. This will not only enhance the understanding of the device structure and physics of kesterite based solar cells, but also provide an effective way for designing other chalcogenide heterojunction solar cells.

Two-dimensional Cu deposition on Pt(100) and stepped surfaces of platinum single crystals

The initial stages of Cu deposition at low overpotentials have been studied on the Pt(100) and Pt(111) basal faces and stepped surfaces Pt(610), Pt(410) of platinum single crystal electrodes. The process includes formation of a Cuad adlayer, an epitaxial monolayer of a two dimensional phase and three dimensional copper crystallites on the (100) terraces, while on the Pt(111) face the two dimensional Cu layer is absent as in this case 3-fold coordination of deposited Cu atoms probably cannot provide sufficient decrease in surface energy of the system. At high enough cathodic overpotentials, the 3D-phase of copper is formed on the monolayer of copper adatoms Pt(111)-Cuad. The Stranski-Krastanov mechanism of nucleation and growth is realized on all the studied electrodes.