Fine recycled aggregates are produced in large quantities when crushing Construction and Demolition Waste (C&DW). Even if coarse recycled aggregates are commonly used for road foundations, fine particles are often rejected as they are considered detrimental for the long-term behaviour of foundations. Physicochemical, mineralogical and mechanical characterizations (through X-ray diffraction, X-ray fluorescence, the chloride and sulphate contents, Los Angeles abrasion, micro-Deval resistance and static plate load tests) were performed on raw and treated fine recycled materials for understanding both the effects of the preparation, the compaction and the freeze–thaw cycles on the properties and the evolution of fine particles. Special attention was provided to the shape analysis of fines by means of image analyser. The results showed that the main characteristic parameters to be considered are the sieving curve and the proportion of grades. The mixes containing the highest quantity of fine particles, specifically lower than 63 µm, usually inducing a higher water demand and a higher capillary rise. This can be explained by specific surface and bluntness parameters which increase with the finer particles, inducing a higher surface roughness and, consequently, a higher potential interaction with water. Compaction did not seem to have a major effect on the production of fines (despite some breakdown occurred during compaction) and on the shape of materials (the bluntness and convexity increased slightly, while the elongation values remained similar after the compaction process). The static plate load tests showed that bearing capacity is slightly lower than the specifications for the road foundation after compaction. However, the studied material could meet the maximum criteria for secondary roads foundation construction on the wear resistance criteria. Recycled aggregates from C&DW without sufficient quality could be blended with other aggregates to enable their usage for upper-level road foundation.

Compounds of the Na 2 (Mn 1-x M 2+x )Fe 2+ Fe 3+ (PO 4 ) 3 (M 2+ = Mg, Cd, Ca, Ni, Zn) alluaudite-type solid solutions were synthesized by hydrothermal techniques at 1 kbar, between 400 and 600 °C, and were investigated by X-ray powder diffraction and infrared spectroscopy. The site occupancy factors of the Na 2 (Mn 1-x Cd x )Fe 2+ Fe 3+ (PO 4 ) 3 compounds, obtained from the Rietveld refinements of the X-ray powder patterns, indicate that the replacement of Mn by Cd mainly takes place on the M1 site. Small amounts of Cd were also detected on the A1 site, compensated by small amounts of Na occurring on M1. The cationic distributions in the Na 2 (Mn 1-x Mg x )Fe 2+ Fe 3+ (PO 4 ) 3 solid solution show a partially disordered distribution of Fe and Mg over the M1 and M2 crystallographic sites. The unit-cell parameters increase significantly when Mn is replaced by Ca or Cd, and decrease significantly when it is replaced by Ni, Mg, or Zn. The infrared spectra show the displacement of two absorption band at ca. 405-445 cm -1 and 595-610 cm -1 , which are assigned to the vibrations of the M 2+ cations localized on the M1 and M2 sites, respectively. This assignment is confirmed by the excellent correlations between the average ionic radius of the cations occurring on the M sites, and the energy of the absorption band.

Alluaudite is a Na-, Mn-, Fe-bearing phosphate mineral, which is known to occur in granitic pegmatites. Moore [1] determined the crystal structure of alluaudite in the C2/c space group and derived the general structural formula, X(2)X(1)M(1)M(2) 2 (PO 4) 3 , with Z = 4. Recently, Hatert et al. [2] proposed a new structural formula for alluaudite, which takes into account the presence of new crystallographic sites in the channels of the structure: [A(2)A(2)'][A(1)A(1)'A(1)'' 2 ]M(1)M(2) 2 (PO 4) 3. In natural alluaudites, divalent cations are localized in the A(1) (Mn, Ca), M(1) (Mn, Fe 2+) , and M(2) (Fe2+, Mn, Mg) crystallographic sites [3]. The exotic cations Cd and Co 2+ have also been inserted experimentally in the alluaudite-type compounds Na 2 Cd 2 M 3+ (PO 4) 3 (M 3+ = Fe 3+ , Ga, Cr) [4] and M + Co 2+ 3 (PO 4)(HPO 4) 2 (M + = Ag, Na) [5] [6]. Since alluaudite is a key mineral in the genetic evolution affecting the Fe-Mn-bearing phosphates, it is of interest to investigate the crystal chemistry of the divalent cation in the alluaudite structure. With this goal in mind, we decided to study the Na 1.5 (Mn 1-x M 2+ x) 1.5 Fe 3+ 1.5 (PO 4) 3 (M 2+ = Cd, Zn) and Na 2 (Mn 1-x M 2+ x)Fe 2+ Fe3+(PO 4) 3 (M2+ = Ca, Cd, Ni, Zn, Mg) solid solutions of alluaudite-type compounds. The phosphates of the Na 1.5 (Mn 1-x M 2+ x) 1.5 Fe 3+ 1.5 (PO 4) 3 solid solutions were synthesized by solid state reactions in air, between 800 and 900°C. The X-ray powder diffraction patterns confirm the presence of pure alluaudite-type compounds for x = 0 to 1. The phosphates of Na 2 (Mn 1-x M 2+ x)Fe 2+ Fe 3+ (PO 4) 3 solid solutions were synthesized hydrothermally between 400 and 600°C at 1 kbar, using horizontally arranged Tuttle-type cold-seal bombs. For x = 1, the X-ray powder diffraction patterns indicate that the Cd-bearing sample is constituted by pure alluaudite, whereas the Ca-, Ni-, Zn-, and Mg-bearing compounds contain small amounts of impurities. The Rietveld refinements of the X-ray powder diffraction patterns of Na 1.5 (Mn 1-x M 2+ x) 1.5 Fe 3+ 1.5 (PO 4) 3 show that Zn is localized in the M(1) and M(2) sites, whereas Cd occurs in the A(1), M(1) and M(2) sites. This feature probably results from the large ionic radius of Cd 2+ (0.95 Å), compared to that of Zn 2+ (0.740 Å) [7]. Several correlations have also been established between the variations of unit-cell parameters and the variations of bond distances.

4 [1]. It exhibits a layered structure. Each layer is composed of [(MS 6)(TlS 6)] n nzigzag chains, which consist of irregular trigonal prisms alternately centred by Tl and M atoms. The layers are stacked perpendicular to the crystallographic a axis and are held together by PS 4 tetrahedra [2]. The bond valence analysis method (BVM) is used to determine where is pointing the Pb 6s 2 lone pair. Extended structure of TlMPS 4 projected along [001].

The binary alkali earth trielides of the A II M III 2 composition exhibit a puzzling multitude of structure types [1] ranging from electron precise Zintl compounds like CaIn 2 [2], to the CeCu 2 type (with still four-bonded M-) and the AlB 2 structure (with graphite analogue M sheets) to the cubic Laves phase e.g. of CaAl 2. The examination of the phase stabilities in mixed compounds AM1 III x M2 III 2-x of two trielides allows to separate the stability ranges in a structure map, taking the electronegativity of M (EN) and the radius ratios (RR=r M /r A) into account: The CaIn 2 type is stable at comparatively large RR, for example over the whole range CaGa 2-CaIn 2 and even up to CaAl 0.6 Ga 1.4 and CaAl 1.2 In 0.8 and in SrIn 2 , in this case also together with a small substitution x in SrM1 x In 2-x by M1=Al (x=0.4) or M1=Ga (x=0.5). Only in a small RR around CaAl 2 the MgCu 2 type is stable, the Laves phase structure type being in accordance with the v.e.c. [3]. At high DEN and low RR values (e.g. Sr/Ba-Ga), the ideal AlB 2 structure type exhibits a distinct stability range. Starting from SrGa 2 up to x=0.8 Ga can be substituted by Al. The CeCu 2 type, which is dominating in AM 2 intermetallics with lower v.e.c., is observed in a region of lower RR and ΔEN values: In BaIn 2 , a substitution of In by 50 % Al and 30 % Ga is possible without a general structure change, in SrAl 2 this holds for a content of up to 50 % In. For the binary trielides, FP-LAPW band structure calculations for the observed and several hypothetical structure types provide insight into the factors favoring the structural stability and also explain the structural changes qualitatively. Starting from the trielides, the influence of a reduced v.e.c. is analysed in a similar way using own and literature data of group 11 and 12 intermetallics (e.g. of the section CaCd 2-CaGa 2 [4] and the ternary system CaAg 2-CaZn 2-CaAl 2 [5]).