Photocatalytic oxidation of propene in gas phase at low concentration by optimized TiO2 nanoparticles

In the present study, nanocrystalline titanium dioxide (TiO2) was prepared by sol–gel method at low temperature from titanium tetraisopropoxide (TTIP) and characterized by different techniques (gas adsorption, XRD, TEM and FTIR). Variables of the synthesis, such as the hydrolyzing agent (acetic acid or isopropanol) and calcination temperatures (300–800 °C), were analyzed to get uniform size TiO2 nanoparticles. The effect that these two variables have on the structure of the resultant TiO2 nanoparticles and on their photocatalytic activity is investigated. The photocatalytic activities of TiO2 nanoparticles were evaluated for propene oxidation at low concentration (100 ppmv) under two different kinds of UV light (UV-A ∼ 365 nm and UV-C ∼ 257.7 nm) and compared with Degussa TiO2 P-25, used as reference sample. The results show that both hydrolyzing agents allow to prepare TiO2 nanoparticles and that the hydrolyzing agent influences the crystalline structure and its change with the thermal treatments. Interestingly, the prepared TiO2 nanoparticles possess anatase phase with small crystalline size, high surface area and higher photocatalytic activity for propene oxidation than commercial TiO2 (Degussa P-25) under UV-light. Curiously, these prepared TiO2 nanoparticles are more active with the 365 nm source than with the 257.7 nm UV-light, which is a remarkable advantage from an application point of view. Additionally, the obtained results are particularly good when acetic acid is the hydrolyzing agent at both wavelengths used, possibly due to the high crystallinity, low anatase phase size and high surface oxygen groups’ content in the nanoparticles prepared with it, in comparison to those prepared using isopropanol.

Synthesis of high surface area TiO2 nanoparticles by mild acid treatment with HCl or HI for photocatalytic propene oxidation

Nanostructured TiO2 photocatalysts with small crystalline sizes have been synthesized by sol-gel using the amphiphilic triblock copolymer Pluronic P123 as template. A new synthesis route, based on the treatment of TiO2 xerogels with acid-ethanol mixtures in two different steps, synthesis and extraction-crystallization, has been investigated, analyzing two acids, hydrochloric and hydriodic acid. As reference, samples have also been prepared by extraction-crystallization in ethanol, being these TiO2 materials amorphous and presenting higher porosities. The prepared materials present different degrees of crystallinity depending on the experimental conditions used. In general, these materials exhibit high surface areas, with an important contribution of microporosity and mesoporosity, and with very small size anatase crystals, ranging from 5 to 7 nm. The activity of the obtained photocatalysts has been assessed in the oxidation of propene in gas phase at low concentration (100 ppmv) under a UVA lamp with 365 nm wavelength. In the conditions studied, these photocatalysts show different activities in the oxidation of propene which do not depend on their surface areas, but on their crystallinity and band gap energies, being sample prepared with HCl both during synthesis and in extraction-crystallizations steps, the most active one, with superior performance than Evonik P25.

Combining visual features and Growing Neural Gas networks for robotic 3D SLAM

The use of 3D data in mobile robotics provides valuable information about the robot’s environment. Traditionally, stereo cameras have been used as a low-cost 3D sensor. However, the lack of precision and texture for some surfaces suggests that the use of other 3D sensors could be more suitable. In this work, we examine the use of two sensors: an infrared SR4000 and a Kinect camera. We use a combination of 3D data obtained by these cameras, along with features obtained from 2D images acquired from these cameras, using a Growing Neural Gas (GNG) network applied to the 3D data. The goal is to obtain a robust egomotion technique. The GNG network is used to reduce the camera error. To calculate the egomotion, we test two methods for 3D registration. One is based on an iterative closest points algorithm, and the other employs random sample consensus. Finally, a simultaneous localization and mapping method is applied to the complete sequence to reduce the global error. The error from each sensor and the mapping results from the proposed method are examined.

Improving 3D Keypoint Detection from Noisy Data Using Growing Neural Gas

3D sensors provides valuable information for mobile robotic tasks like scene classification or object recognition, but these sensors often produce noisy data that makes impossible applying classical keypoint detection and feature extraction techniques. Therefore, noise removal and downsampling have become essential steps in 3D data processing. In this work, we propose the use of a 3D filtering and down-sampling technique based on a Growing Neural Gas (GNG) network. GNG method is able to deal with outliers presents in the input data. These features allows to represent 3D spaces, obtaining an induced Delaunay Triangulation of the input space. Experiments show how the state-of-the-art keypoint detectors improve their performance using GNG output representation as input data. Descriptors extracted on improved keypoints perform better matching in robotics applications as 3D scene registration.

3D model reconstruction using neural gas accelerated on GPU

In this work, we propose the use of the neural gas (NG), a neural network that uses an unsupervised Competitive Hebbian Learning (CHL) rule, to develop a reverse engineering process. This is a simple and accurate method to reconstruct objects from point clouds obtained from multiple overlapping views using low-cost sensors. In contrast to other methods that may need several stages that include downsampling, noise filtering and many other tasks, the NG automatically obtains the 3D model of the scanned objects. To demonstrate the validity of our proposal we tested our method with several models and performed a study of the neural network parameterization computing the quality of representation and also comparing results with other neural methods like growing neural gas and Kohonen maps or classical methods like Voxel Grid. We also reconstructed models acquired by low cost sensors that can be used in virtual and augmented reality environments for redesign or manipulation purposes. Since the NG algorithm has a strong computational cost we propose its acceleration. We have redesigned and implemented the NG learning algorithm to fit it onto Graphics Processing Units using CUDA. A speed-up of 180× faster is obtained compared to the sequential CPU version.