Synthesis and low temperature fluorination of the alkaline-earth palladates Ba2−xSrxPdO3 (x = 0–2)

Since the discovery of superconductivity in Sr₂CuO₂F_2+δ there has been an increased interest in ternary oxide-fluorides. Sr₂CuO₂F_2+δ is prepared via low temperature (T = 220 °C) reaction routes. Low temperature fluorination induces an interesting structural rearrangement in the parent compound Sr₂CuO₃, which is a one-dimensional material containing linear chains of vertex sharing CuO₄ squares along the crystallographic b axis. Upon fluorination, one oxide is substituted by two fluorides and Cu²⁺ becomes octahedrally coordinated by four oxides and two fluorides. The fluorinated compound Sr₂CuO₂F_2+δ displays the T-type structure (La₂CuO₄). Insertion of excess fluorine, δ, also takes place and this fluorine occupies interstitial sites in the T structure. Although the starting material Ca₂CuO₃ is isostructural to Sr₂CuO₃, Ca₂CuO₂F_2+δ displays the T′ (Nd₂CuO₄) structure due to the smaller radius of Ca²⁺ compared to that of Sr²⁺.

The alkaline-earth palladates with the general formula A₂PdO₃ (A = Ba, Sr) are isostructural with the A₂CuO₃(A = Ca, Sr) materials. We prepared the Ba_2−xSr_xPdO₃ (x = 0–2) series and performed low temperature fluorination, which led to the synthesis of the series Ba_2−xSr_xPdO₂F_2+δ (0 ≤ x ≤ 1.5). All the compounds in the Ba_2−xSr_xPdO₂F_2+δ series show T′ structure (Ca₂CuO₂F_2+δ). Similarities and differences with Sr₂CuO₂F_2+δ and Ca₂CuO₂F_2+δ will be discussed.     

End unit effect on the Tg of differentiated series of perfluoro-poly (oxymethylene-co-oxyethylene) oligomers

Twelve series of linear oligomers of four different degrees of polymerization (x_n = 8.77−41.55), having a common perfluorinated random copolymeric chain as molecular body and two equal foreign end units of one of the types listed in Table 1, have been synthesized by derivatization of base samples of one of them having a diolic---CH₂OH functionality. The glass transition temperature T_g of all the series was measured and thus examined as a function of x_n. A clear end unit effect is observed, dominantly determined in every series by chemical nature and structure of the end units, quantitatively expressed at any x_n by different positive or negative T_g deviations from the common asymptotic T_g∞ value. The results are also discussed in terms of copolymer end effect and of relation between T_g and end copolymeric composition.     

Thermodynamics of (1,4-difluorobenzene + an n-alkane) and of (hexafluorobenzene + an n-alkane)

Excess molar enthalpies H^E and excess molar volumes V^E have been measured, as a function of mole fraction x₁, at 298.15 K and atmospheric pressure for the five liquid mixtures (x₁1,4-C₆H₄F₂ + x₂n-C_lH_2l+2), l = 7, 8, 10, 12 and 16. In addition, H^E and excess molar heat capacities C_P^E at constant pressure have been determined for the two liquid mixtures (x₁C₆F₆ + x₂n-C_lH_2l+2), l = 7 and 14, at the same temperature and pressure. The instruments used were flow microcalorimeters of the Picker design (the H^E version was equipped with separators) and a vibrating-tube densimeter, respectively.

The excess enthalpies of the five difluorobenzene mixtures are all positive and quite large; they increase with increasing chain length l of the n-alkane from H^E(x₁ = 0.5)/(J mol⁻¹) = 1050 for l = 7 to 1359 for l = 16. The corresponding excess volumes V^E are all positive and also increase with increasing l: V^E(x₁ = 0.5)/(cm³ mol⁻¹) = 0.650 for l = 7 and 1.080 for l = 16. Interestingly, the excess enthalphies of the corresponding mixtures with hexafluorobenzene are only about 5% larger, whereas the excess volumes of (x₁C₆F₆ + x₂n-C_lH_2l+2) are roughly twice as large as those of their counterparts in the series containing 1,4-C₆H₄F₂. Specifically, at 298.15 K H^E(x₁ = 0.5)/(J mol⁻¹) = 1119 for (x₁C₆F₆ + x₂n-C₇H₁₆) and 1324 for (x₁C₆F₆ + x₂n-C₁₄H₃₀), and for the same mixtures V^E(x₁ = 0.5)/(cm³ mol⁻¹) = 1.882 and 2.093, respectively. The excess heat capacities for both systems are negative and of about the same magnitude as the excess heat capacities of mixtures of fluorobenzene with the same n-alkanes (Roux et al., 1984): C_P^E(x₁ = 0.5)/(J K⁻¹ mol⁻¹) = −1.18 for (x₁C₆F₆ + x₂n-C₇H₁₆), and −2.25 for (x₁C₆F₆ + x₂n-C₁₄H₃₀). The curve C_P^E vs. (x₁ for x₁C₆F₆ + x₂n-C₁₄H₃₀) shows a sort of “hump” for x₁ 0.5, which is presumed to indicate emerging W-shape composition dependence at lower temperatures.     

Dynamics of electronic polarization in molecular crystals

Dynamics of electronic polarization in the vicinity of charge carriers in molecular crystals is for the first time investigated here in connection with the carrier transport and intramolecular vibronic polarization. According to standard picture it has been assumed that the electronic polarization relaxation time is extremely short, as estimated from the relation τ_c = τ_d1 ≈ h/E_exc, where E_exc is the energy of the first single exciton state. In the case of anthracene (Ac) crystals, the value of τ_e is about 2 × 10⁻¹⁶ s, i.e. by several orders of magnitude shorter than a typical hopping (residence) time of charge carriers τ_h = 10⁻¹⁴ -10⁻¹³ s. It is argued that typical time of full reconstruction of the electronic polarization after individual carrier hops equals, in the slow carrier regime, approximately to t_d2 ≈ h/ΔE_exc is the width of the lowest singlet-exciton band. In Ac, this means t_d2 ≈ 0.73 × 10⁻¹⁴ s. Physical implications of this relatively high value of t_d2 in connection with carrier transport and molecular (vibronic) polarization are discussed.     

Phase equilibria and crystal structure of the solid solution LaFe1−xNixO3−δ (0 ≤ x ≤ 1)

Phase equilibria in the LaFeO₃–“LaNiO₃” were studied at 1100 °C in air. The samples were synthesized by standard ceramic and/or solution route via nitrate or citrate precursors. According to the results of XRD it was found that the homogeneity ranges of LaFe_1−xNi_xO_3−δ solid solution lay within 0.0 ≤ x ≤ 0.4 (sp.gr. Pbnm) and 0.6 ≤ x ≤ 0.8 (sp.gr. ). The structural parameters (bond lengths, atom coordinates) for the single-phase samples were refined using Rietveld analysis. The unit cell parameters versus LaFe_1−xNi_xO_3−δ composition are presented.     

Electron transfer of quinone self-assembled monolayers on a gold electrode

Dialkyl disulfide-linked naphthoquinone, (NQ-C_n-S)₂, and anthraquinone, (AQ-C_n-S)₂, derivatives with different spacer alkyl chains (C_n: n = 2, 6, 12) were synthesized and these quinone derivatives were self-assembled on a gold electrode. The formation of self-assembled monolayers (SAMs) of these derivatives on a gold electrode was confirmed by infrared reflection-absorption spectroscopy (IR-RAS). Electron transfer between the derivatives and the gold electrode was studied by cyclic voltammetry. On the cyclic voltammogram a reversible redox reaction between quinone (Q) and hydroquinone (QH₂) was clearly observed under an aqueous condition. The formal potentials for NQ and AQ derivatives were −0.48 and −0.58 V, respectively, that did not depend on the spacer length. The oxidation and reduction peak currents were strongly dependent on the spacer alkyl chain length. The redox behavior of quinone derivatives depended on the pH condition of the buffer solution. The pH dependence was in agreement with a theoretical value of E_1/2 (mV) = E′ − 59pH for 2H⁺/2e⁻ process in the pH range 3–11. In the range higher than pH 11, the value was estimated with E_1/2 (mV) = E′ − 30pH , which may correspond to H⁺/2e⁻ process. The tunneling barrier coefficients (β) for NQ and AQ SAMs were determined to be 0.12 and 0.73 per methylene group (CH₂), respectively. Comparison of the structures and the alkyl chain length of quinones derivatives on these electron transfers on the electrode is made.     

Electronic absorption spectrum of Ba(MnO4)2·3H2O/Ba(ClO4)2·3H2O

Polarized absorption spectra of Ba(MnO₄)₂·3H₂O/Ba(ClO₄)₂·3H₂O mixed single crystals are reported at 4.2°K. Previous ¹T₂ → ¹A₁ assignments for the 5200 Å and 3000 Å absorption bands of MnO₄⁻ are substantiated; further support is provided for the ¹T₁ → ¹A₁ assignment of the 3600 Å absorption band of MnO₄⁻. The site-splitting of the 5200 Å ¹T₂ state is E(¹E)−E(¹A) ≈ −150 cm⁻¹; that of the 3000 Å ¹T₂ state is E(¹E)−E(¹A) ≈ 300 cm⁻¹. A significant e vibronic intensity component is observed in the 5200 Å ¹T₂ state.     

Structure and redox properties of CexPr1−xO2−δ mixed oxides and their catalytic activities for CO, CH3OH and CH4 combustion

A series of Ce_xPr_1−xO_2−δ mixed oxides were synthesized by a sol–gel method and characterized by Raman, XRD and TPR techniques. The oxidation activity for CO, CH₃OH and CH₄ on these mixed oxides was investigated. When the value x was changed from 1.0 to 0.8, only a cubic phase CeO₂ was observed. The samples were greatly crystallized in the range of the value x from 0.99 to 0.80, which is due to the formation of solid solutions caused by the complete insertion of Pr into the CeO₂ crystal lattices. Raman bands at 465 and 1150 cm⁻¹ in Ce_xPr_1−xO_2−δ samples are attributed to the Raman active F_2g mode of CeO₂. The broad band at around 570 cm⁻¹ in the region of 0.3 ≤ x ≤ 0.99 can be linked to oxygen vacancies. The new band at 195 cm⁻¹ may be ascribed to the asymmetric vibration caused by the formation of oxygen vacancies. The TPR profile of Pr₆O₁₁ shows two reduction peaks and the reduction process is followed: . The reduction temperature of Ce_xPr_1−xO_2−δ mixed oxides is lower than those of Pr₆O₁₁ or CeO₂. TPR results indicate that Ce_xPr_1−xO_2−δ mixed oxides have higher redox properties because of the formation of Ce_xPr_1−xO_2−δ solid solutions. The presence of the oxygen vacancies favors CO and CH₃OH oxidation, while the activity of CH₄ oxidation is mostly related to reduction temperatures and redox properties.     

Homo- and hetero-bridged mixed-valence dinuclear complexes which contain the fragment ‘Rh(C6F5)3’

The reaction of the anionic mononuclear rhodium complex [Rh(C₆F₅)₃Cl(Hpz)]^t- (Hpz = pyrazole, C₃H₄N₂) with methoxo or acetylacetonate complexes of Rh or Ir led to the heterodinuclear anionic compounds [(C₆F₅)₃Rh(μ-Cl)(μ-pz)M(L₂)] [M = Rh, L₂ = cyclo-octa-1,5-diene, COD (1), tetrafluorobenzobarrelene, TFB (2) or (CO)₂ (4); M = Ir, L₂ = COD (3)]. The complex [Rh(C₆F₅)₃(Hbim)]⁻ (5) has been prepared by treating [Rh(C₆F₅)₃(acac)]⁻ with H₂bim (acac = acetylacetonate; H₂bim = 2,2′-biimidazole). Complex 5 also reacts with Rh or Ir methoxo, or with Pd acetylacetonate, complexes affording the heterodinuclear complexes [(C₆F₅)₃Rh(μ-bim)M(L₂)]⁻ [M = Rh, L₂ = COD (6) or TFB (7); M = Ir, L₂ = COD (8); M = Pd, L₂ = η³-C₃H₅ (9)]. With [Rh(acac)(CO)₂], complex 5 yields the tetranuclear complex [{(C₆F₅)₃Rh(μ-bim)Rh(CO)₂}₂]₂₋. Homodinuclear Rh^III derivatives [{Rh(C₆F₅)₃}₂(μ-L)₂]^·- [L₂ = OH, pz (11); OH, S^tBu (12); OH, SPh (13); bim (14)] have been obtained by substitution of one or both hydroxo groups of the dianion [{Rh(C₆F₅)₃(μ-OH)}₂]²⁻ by the corresponding ligands. The reaction of [Rh(C₆F₅)₃(Et₂O)_x] with [PdX₂(COD)] produces neutral heterodinuclear compounds [(C₆F₅)₃Rh(μ-X)₂Pd(COD)] [X = Cl (15); Br (16)]. The anionic complexes 1–14 have been isolated as the benzyltriphenylphosphonium (PBzPh₃⁺) salts.     

Transport phenomenon as a function of counter and co-ions in solution: chronopotentiometric behavior of anion exchange membrane in different aqueous electrolyte solutions

Anion exchange membrane has been investigated in different electrolyte solutions by chronopotentiometry to explore the influence of co-ion and counterion of the exchange group of the membrane, on the transport phenomena. Chloride, nitrate, sulfate and acetate in sodium salts were used as counterions and sodium, potassium, calcium and ammonium in chloride salts were used as co-ions. The membrane showed a potential drop (E₀) in all these electrolytes when a constant current was applied across it, which remained constant for a period less than τ, called the transition time and rose gradually to a maximum (E_max) value. The parameters such as τ, E₀ and E_max and the potential jump (ΔE) and τ and the inflection zone (Δt) along the time axis have been measured and compared at an applied current density (I) of 10 mA cm⁻² in 10 mM solutions. The values of τ^1/2/z_A[A₀] or τ^1/2/z_C[C₀], with or , E₀ and ΔE with or (where r_A and r_C are the ionic radii of counter and co-ions, respectively) have been correlated. Permselectivity (P) and transference number of the membrane with respect to each one of the above electrolytes have been evaluated and discussed.     

Spectroscopic and structural analysis of Ca substituted La orthoferrite

Samples of orthoferrites La_1−xCa_xFeO₃ (0.15≤x≤0.45) were synthesized by double sintering ceramic technique. X-ray diffraction and infrared spectroscopy experiments were carried out and discussed for the investigated samples. The data showed the formation of single-phase orthorhombic structure of space group Pbnm. The FTIR spectra were performed in the region (1200–200 cm⁻¹). Four main absorption bands were present with some side bands and shoulders in the range (1200–400 cm⁻¹). Another four bands were appeared in the range (400–200 cm⁻¹). The positions, intensities and values of the absorptions bands vary depending on the Ca content in the samples. The first absorption band appeared at about 920 cm⁻¹ was assigned to the La–O stretching vibration.     

Ab initio calculations of minimum-energy pathways of the nucleophilic addition of the H anion, LiH molecule and Li/H ion pair to acetylene and methylacetylene

Ab initio calculations were performed for special points of the minimal energy pathways (MEP) of the nucleophilic addition reactions of the isolated H⁻ anion, LiH molecule and Li⁺/H⁻ ion pair to acetylene (A) and methylacetylene (MA) molecules, proceeding in accordance (M) and against (aM) the Markovnikov's rule. All structural parameters were optimized using the restricted Hartree–Fock (RHF) method. For the addition of H⁻, the 6-31++G^* basis set was used and for the reactions of LiH and Li⁺/H⁻ the 6-31G^* basis set with the subsequent recalculation of single point energies, taking into account of electron correlation energy by means of the second-order Möller–Plesset perturbation theory at the MP2/6-31++G^** level. The results of calculations demonstrate, that the energy characteristics of both M- and aM-additions with H⁻ do not differ sufficiently (0.1–1.2 kcal/mol for the activation energies (ΔE_a) and the reaction heats (ΔQ)). The substitution of the H atom by the CH₃ group in A molecule results in practically the same values of ΔQ and ΔE_a. On the contrary, for the LiH molecule and Li⁺/H⁻ ionic pair, the M-addition is favorable (charge control). It is found that the presence of electrophile decreases the activation energy by 3–5 kcal/mol as compared with the addition of the isolated hydride ion H⁻.     

Sub-solidus phase equilibria in CeO2–SrO system

In this communication, we report on the synthesis and characterization of a series of compounds with the general composition Ce_1−xSr_xO_2−x (0.0≤x≤1.0), to establish a detailed phase relation in the CeO₂–SrO system. The X-ray diffraction (XRD) pattern of the each product was refined to determine the solid solubility and the homogeneity range. The solid solubility limit of SrO in CeO₂ lattice, under the slow cooled conditions, is represented as Ce_0.91Sr_0.09O_1.91 (i.e. 9 mol% of SrO). A careful delineation of the phase boundary revealed that the stoichiometric SrCeO₃, in fact, contains a little amount of CeO₂ also. The mono-phasic compound could be obtained at the nominal composition Sr_0.55Ce_0.45O_1.45. The nominal composition Sr₂CeO₄, under the heat treatment used in the present investigation, was a bi-phasic mixture of SrCeO₃ and SrO. No new ordered phases were obtained in this system.     

Equilibrium and structural studies of silicon(IV) and aluminium(III) in aqueous solution—10. A potentiometric study of aluminium(III) pyrocatecholates and aluminium(III) hydroxo pyrocatecholates in 0.6 M Na(Cl)

Equilibria between aluminium(III), pyrocatechol (1,2-dihydroxybenzene, H₂L) and OH⁻ were studied in 0.6 M Na(Cl) medium at 25°C. The measurements were performed as emf titrations (glass electrode) within the limits 1.5 ≤ − log[H⁺] ≤ 9; 0.0005 ≤ B ≤ 0.015 M; 0.006 ≤ C ≤ 0.03 M and 2 ≤ C/B ≤ 30 (B and C stand for the total concentrations of aluminium(III) and pyrocatechol respectively). All data can be explained with a main series of complexes: A1L⁺, log β_−2,1,1 = − 6.337 ± 0.005; A1L₂⁻, log β_−4,1,2 = −15.44 ± 0.017 and A1L₃³⁻, log β_−6,1,3 = − 28.62 ± 0.024 together with two minor species: Al(OH)L₂²⁻, log β_−5,1,2 = − 23.45 ± 0.079 and Al₃(OH)₃L₃, log β_−9,3,3 = − 29.91 ± 0.066. Of the two, the latter probably is a type of average composition complex principally occurring at low C/B quotients. The first acidity constant for pyrocatechol as determined in separate experiments is log β_−1,0,1 = − 9.198 ± 0.001. The standard deviations given are 3σ(log β p,q,r). Data were analyzed with the least squares computer program LETAGROPVRID. In a model calculation using kaolinite as solid phase, we compared the complexation ability of this system with that of the system Al³⁺-OH⁻-salicylic acid, reported earlier in this series.     

I NQR and spectroscopic investigation of impurity-doped and mixed lithium iodate Li1−xHxIO3 crystals

The ¹²⁷I NQR, IR absorption and Raman spectra of impurity-doped and mixed lithium iodate Li_1−xH_xIO₃ crystals grown from water solutions with different LiIO₃/HIO₃ ratios were investigated depending on the content of the impurity hydrogen x. The NQR results suggested that, at small concentration of doping iodic acid x<0.22, the lattice dynamics of the crystal grown from water solution changes significantly though the crystal retains hexagonal symmetry. Spectroscopic studies are compatible with average hexagonal symmetry of the grown doped crystals. From the results of Raman studies at room temperature and 100 K, the concentration range of hydrogen dopant 0.22<x<0.36 was found where disordered solid solution crystals Li_1−xH_xIO₃ are formed.     

Synthesis, structure and oxygen nonstoichiometry of La0.4Sr0.6Co1−yFeyO3−δ

The samples of La_0.4Sr_0.6Co_1−yFe_yO_3−δ (y = 0.2 and 0.4) were prepared using both conventional ceramic technique and nitrate–citrate precursors technique. The phase identification was made by X-ray diffraction method. The refinement of structural parameters from the XRD and neutron diffraction measurements was performed by full profile Rietveld analysis. Neutron diffraction showed that both samples possess distorted perovskite-type structure. Oxygen nonstoichiometry was measured by chemical analysis and thermogravimetry (TG) analysis in the range 20 ≤ T/°C ≤ 900 and 2E-5 ≤ pO₂/atm ≤ 4E-1. TG-experiments indicate a relatively fast and reversible oxygen exchange at pO₂ > 1E-2 atm. Mass saturation occurs at T < 300 °C upon cooling. The absolute value of oxygen nonstoichiometry was determined by iodometric titration measurements. It was found that both samples have practically stoichiometric composition at 300 °C in air and δ increases with increasing temperature and decreasing oxygen partial pressure.     

The gaseous azide ion, N3: Enthalpy of formation, ΔHf(N3(g)) and charge distribution lattice energies of sodium, potassium and rubidium azides

The calculations reported here assign a charge q_N = −0.52 electron units to the terminal nitrogen atoms in the azide ion and a value of 141.9 kJ mole⁻¹ to the enthalpy of formation of the gaseous azide ion, ΔH_f⁰(N₃⁻(g)). The total lattice potential energies are found to be: E_pot(NaN₃) = 725.1 kJ mole⁻¹; E_pot(KN₃) = 650.7 kJ mole⁻¹ and E_pot(RbN₃) = 632.1 kJ mole⁻¹.     

Kinetics study of the gas-phase reactions of C2F5OC(O)H and n-C3F7OC(O)H with OH radicals at 253–328 K

The rate constants, k₁ and k₂ for the reactions of C₂F₅OC(O)H and n-C₃F₇OC(O)H with OH radicals were measured using an FT-IR technique at 253–328 K. k₁ and k₂ were determined as (9.24 ± 1.33) × 10⁻¹³ exp[−(1230 ± 40)/T] and (1.41 ± 0.26) × 10⁻¹² exp[−(1260 ± 50)/T] cm³ molecule⁻¹ s⁻¹. The random errors reported are ±2 σ, and potential systematic errors of 10% could add to the k₁ and k₂. The atmospheric lifetimes of C₂F₅OC(O)H and n-C₃F₇OC(O)H with respect to reaction with OH radicals were estimated at 3.6 and 2.6 years, respectively.     

Preparation of dense, ultra-thin MIEC ceramic membranes by atmospheric spray-pyrolysis technique

Dense ceramic mixed ionic and electronic conducting membranes have been deposited by atmospheric spray-pyrolysis technique onto porous ceramic substrates. Perovskite oxide layers, i.e. manganites La_1−xSr_xMnO₃, ferrites La_1−xSr_xFe_1−y(Co,Ni)_yO₃, gallates La_1−xSr_xGa_1−y(Co,Ni,Fe)_yO₃, cobaltites La_1−xSr_xCoO₃ and related perovskites such as lanthanum nickelate La₂NiO₄ layers have been prepared. The structure, morphology and composition of the layers were characterised by XRD, SEM and WDS, respectively. Density and gas tightness of the layers were studied as a function of deposition process parameters, film thickness (from 0.5 to 3 μm) and preparation procedure. The presence of cracks and defects due to thermo-mechanical stresses applied during or after the preparation process were correlated with the membrane composition and the corresponding thermal expansion coefficient differences between substrates and membranes.     

The interaction of fluorosilanes with hydrogen fluoride clusters: strongly blue-shifted hydrogen bonds

The equilibrium structures, binding energies, and vibrational spectra of the complexes formed between hydrogen fluoride clusters (HF)_n (1≤n≤4) and the fluorosilanes SiHF₃, SiH₂F₂, and SiH₃F are investigated within the second-order Møller–Plesset perturbation theory method applying extended basis sets. It is shown that Si–FH–F halogen–hydrogen bonds are formed in the most stable open dimers, SiHF₃–HF, SiH₂F₂HF, and SiH₃FHF. No Si–HF–H hydrogen bonds occur in these dimers. Nevertheless, blue shifts of Si–H stretching frequencies are calculated. All three trimers, fluorosilane–(HF)₂, all three tetramers, fluorosilane–(HF)₃, and two of the pentamers, fluorosilane–(HF)₄, form cyclic structures with strong Si–FH–F halogen–hydrogen bonds and weak Si–HF–H contacts, the latter displaying, nevertheless, strongly blue-shifted Si–H stretching frequencies. These blue shifts are comparable in size to those of the corresponding fluoromethane–(HF)_n complexes and are with about +50 cm⁻¹ for the case n=3 among the largest ever calculated and definitely the largest for Si–H bonds. In the title complexes, the formation of the Si–FH–F halogen–hydrogen bonds induces a substantial stretching of this Si–F bond, which in turn leads to a significant contraction of the fluorosilane Si–H bond in the Si–HF–H hydrogen bond. This disposition of the fluorosilane monomers is demonstrated with the aid of suitable potential energy surface scans and appears to be a prerequisite for the formation of strongly blue-shifted hydrogen bonds.