H2 Flows Research Articles

QXAFS法によるUSYゼオライトを担体としたパラジウムの動的挙動解析

Dynamic structural change of palladium loaded on USY zeolites was monitored by means of an in situ Quick-XAFS method inthe H2, O2 atmosphere and solvents. We found that metal Pd clusters were readily generated on exposure to a H2 flow at room temperature. The size of the Pd clusters could be regulated simply by changing the number of times that H2 and O2 flows were alternately introduced into the in situ cell. Then structural change of Pd was monitored during temperature programmed reduction and oxidation. We could observe the coalescence and dispersion behavior of Pd depending on the temperature and oxidation-reduction cycles. The Pd/USY catalysts exhibited prominent activity in the Suzuki-Miyaura coupling when the reaction was performed in o-xylene. Pd K-edge and Pd L3-edge XAFS analyses revealed the formation of atomic Pd with a cationic character in the o-xylene as a solvent.

A novel non-sequential hydrogen-pulsed deep reactive ion etching of silicon

A non-sequential pulsed-mode deep reactive ion etching of silicon is reported that employs continuous etching and passivation based on SF6 and H2 gases. The passivation layer, as an important step for deep vertical etching of silicon, is feasible by hydrogen pulses in proper time-slots. By adjusting the etching parameters such as plasma power, H2 and SF6 flows and hydrogen pulse timing, the process can be controlled for minimum underetch and high etch-rate at the same time. High-aspect-ratio features can be realized with low-density plasma power and by controlling the reaction chemistry. The so-called reactive ion etching lag has been minimized by operating the reactor at higher pressures. X-ray photoelectron spectroscopy and scanning electron microscopy have been used to study the formation of the passivation layer and the passivation mechanism.

Gas composition modeling in a reformed Methanol Fuel Cell system using adaptive Neuro-Fuzzy Inference Systems

This work presents a method for modeling the gas composition in a Reformed Methanol Fuel Cell system. The method is based on Adaptive Neuro-Fuzzy-Inference-Systems which are trained on experimental data. The developed models are of the H2, CO2, CO and CH3OH mass flows of the reformed gas. The ANFIS models are able to predict the mass flows with mean absolute errors for the H2 and CO2 models of less than 1% and 6.37% for the CO model and 4.56% for the CH3OH model.The models have a wide range of applications such as dynamic modeling, stoichiometry observation and control, advanced control algorithms, or fuel cell diagnostics systems.

Enhanced H2 Separation through Mixed Proton–Electron Conducting Membranes Based on La5.5W0.8M0.2O11.25−δ

La(5.5) WO11.25-δ is a proton-conducting oxide that shows high protonic conductivity, sufficient electronic conductivity, and stability in moist CO2 environments. However, the H2 flows achieved to date when using La(5.5) WO11.25-δ membranes are still below the threshold for practical application in industrial processes. With the aim of improving the H2 flow obtained with this material, La(5.5) WO11.25-δ was doped in the W position by using Re and Mo; the chosen stoichiometry was La(5.5) W0.8 M0.2 O11.25-δ . This work presents the electrochemical characterization of these two compounds under reducing conditions, the H2 separation properties, as well as the influence of the H2 concentration in the feed stream, degree of humidification, and operating temperature. Doping with both Re and Mo enabled the magnitude of H2 permeation to be enhanced, reaching unrivaled values of up to 0.095 mL min(-1) cm(-2) at 700 °C for a La(5.5) W0.8 Re0.2 O11.25-δ membrane (760 μm thick). The spent membranes were investigated by using XRD, SEM, and TEM on focused-ion beam lamellas. Furthermore, the stability in CO2 -rich and H2 S-containing atmospheres was evaluated, and the compounds were shown to be stable in the atmospheres studied.

Steam reforming of ethanol over bimetallic RhPt/La2O3: Long-term stability under favorable reaction conditions

Recently, the steam reforming of biofuels has been presented as a potential hydrogen source for fuel cells. Because this scenario represents an interesting opportunity for Colombia (South America), which produces large amounts of bioethanol, the steam reforming of ethanol was studied over a bimetallic RhPt/La2O3 catalyst under bulk mass transfer conditions. The effect of temperature and the initial concentrations of ethanol and water were evaluated at space velocities above 55,000 h−1 to determine the conditions that maximize the H2/CO ratio and reduce CH4 production while maintaining 100% conversion of ethanol. These requirements were accomplished when 21 mol% H2O and 3 mol% C2H5OH (steam/ethanol molar ratio = 7) were reacted at 600 °C. The catalyst stability was assessed under these reaction conditions during 120 h on stream, obtaining ethanol conversions above 99% during the entire test. The effect of both H2 and air flows as catalyst regeneration treatments were evaluated after 44 and 67 h on stream, respectively. The results showed that H2 treatment accelerated catalyst deactivation, and air regeneration increased both the catalyst stability and the H2 selectivity while decreasing CH4 generation. Fresh and spent catalyst samples were characterized by TEM/EDX, XPS, TPR, and TGA. Although the Rh and Pt in the fresh catalyst were completely reduced, the spent samples showed a partial oxidation of Rh and small amounts of carbonaceous residue. A possible Rh–Pt–Rh2O3 structure was proposed as the active site on the catalyst, which was regenerated by air treatment.

Performance of novel bimetallic carbonyl clusters as PEM fuel cell anodes, a comparative study

This work describes the performance of novel bimetallic catalysts, prepared from ruthenium, rhodium and iridium carbonyl clusters by a thermolysis procedure in o-dichlorobenzene. The electrochemical characterization by the rotating disk electrode technique in 0.5 mol L−1 H2SO4 showed that the RuxIry(CO)n, RuxRhy(CO)n and RhxIry(CO)n clusters are able to perform both the hydrogen oxidation reaction (HOR) and oxygen reduction reaction (ORR), even in the presence of fuel cell contaminants such as CO and methanol, respectively. These promising results led us to evaluate the new materials as electrodes in a single fuel cell, using a Fuel Cell Test System designed and built in our group. The performance results of the three bimetallic clusters as anodes in a hydrogen PEM fuel cell are presented in this work. In the tests different H2 and O2 gas flows were fed into the cell to determine the most adequate ratio for maximum power. In the absence of CO, the results showed that although the three bimetallic materials had a similar performance to that of platinum with low flows of both reactants, RuxIry(CO)n showed the best electrocatalytic parameters. When the hydrogen fuel feed was contaminated with 100 ppm and 0.5% CO, the commercial platinum activity decreased considerably or was completely lost. However, while the current density of the novel materials also decreased in the presence of CO, it was significantly above that of platinum nanoparticles, the RhxIry(CO)n and RuxIry(CO)n catalysts showing the best performance in the presence of 100 ppm CO and 0.5% CO respectively. These results are promising in the context of PEM fuel cells using reforming hydrogen.

Systems Modeling of Chemical Hydride Hydrogen Storage Materials for Fuel Cell Applications

A fixed bed reactor was designed, modeled and simulated for hydrogen storage on-board the vehicle for PEM fuel cell applications. Ammonia borane was selected by DOE’s Hydrogen Storage Engineering Center of Excellence as the initial chemical hydride of study because of its high hydrogen storage capacity (up to ∼16% by weight for the release of ∼2.5 molar equivalents of hydrogen gas) and its stability under typical ambient conditions. The design evaluated consisted of a tank with eight thermally isolated sections in which H2 flows freely between sections to provide ballast. Heating elements are used to initiate reactions in each section when pressure drops below a specified level in the tank. Reactor models in Excel and COMSOL were developed to demonstrate the proof-of-concept, which was then used to develop systems models in Matlab/Simulink. Experiments and drive cycle simulations showed that the storage system meets thirteen 2010 DOE targets in entirety and the remaining four at greater than 60% of the target.

H2flows in the Corona Australis cloud and their driving sources

We uncover the H2 flows in the Corona Australis molecular cloud and in particular identify the flows from the Coronet cluster. Near-infrared H2 v=1--0 S(1), 2.12micron-line, narrow-band imaging survey of the R CrA cloud core was carried out. We identify the best candidate-driving source for each outflow by comparing the flow properties, available proper motions, and the known/estimated properties of the driving sources. We also adopted the thumbrule of outflow power as proportional to source luminosity and inversely proportional to the source age to reach a consensus. Results: Continuum-subtracted, narrow-band images reveal several new Molecular Hydrogen emission-line Objects (MHOs). Together with previously known MHOs and Herbig-Haro objects we catalog at least 14 individual flow components of which 11 appear to be driven by the RCrA aggregate members. The flows originating in the Coronet cluster have lengths of ~0.1-0.2 pc. Eight out of nine submillimeter cores mapped in the Coronet cluster region display embedded stars driving an outflow component. Roughly 80% of the youngest objects in the Coronet are associated with outflows. The MHO flows to the west of the Coronet display lobes moving to the west and vice-versa, resulting in nondetections of the counter lobe in our deep imaging. We speculate that these counterflows may be experiencing a stunting effect in penetrating the dense central core. Conclusions:Although this work has reduced the ambiguities for many flows in the Coronet region, one of the brightest H2 feature (MHO2014) and a few fainter features in the region remain unassociated with a clear driving source. The flows from Coronet, therefore, continue to be interesting targets for future studies.

Evaluation of ScSZ-Based Microtubular SOFCs under 3% Humidified CH4 Fuel Flow at Intermediate Temperature

This report summarizes the feasibility studies on the fabrication and evaluation of a Sc2O3-doped zirconia (ScSZ)-based microtubular SOFC under 3 % humidified CH4 fuel flow at intermediate temperature, in addition to those under 3 % humidified H2 fuel flow. The cell was fabricated through co-sintering of a ScSZ electrolyte layer and a NiO-ScSZ anode support, and then a La0.6Sr0.4Co0.2Fe0.8O3-x-Ce0.9Gd0.1O1.95 cathode layer was coated on the electrolyte layer. Evaluation was conducted using a potentio/galvanostat and impedance analyzer under the humidified H2 and CH4 flows. The maximum power densities at 650 and 700 °C were 41.6 and 58.5 mW/cm2, respectively, under the CH4 gas flow. And the cell showed a good stability during one hour current loading test under 3 % humidified CH4 flow.

Microwave Assisted Synthesis of Nano Zeolite Seed for Synthesis Membrane and Investigation of its Permeation Properties for H2 Separation

MFI-type zeolite membranes (ZSM-5) were prepared on α-alumina tubular supports and tested for separation of H2/CO mixtures. The effect of pressure and temperature on H2 and CO flows and selectivity was investigated. The best results in this work were obtained with a ZSM-5 membrane prepared over a porous α-alumina tube seeded with zeolite nanocrystals synthesized with microwave technique. In this case the H2 permeance was obtained 2.8 × 10-6 mol/ (m2.s.Pa) with a H2/CO ideal selectivity of 4.9 at 373 K. The permeation results of four gas (H2, CO, N2, CH4) on the synthesized membrane were also investigated.

Stepwise Growth of Pd Clusters in USY Zeolite at Room Temperature Analyzed by QXAFS

The stepwise growth of Pd in a USY zeolite was followed by in situ quick X-ray absorption fine structure (QXAFS) spectroscopy involving repeated alternating exposures to H2 and O2 flows at room temperature. During the first reduction of 0.4 wt % Pd/USY with 8% H2, Pd clusters with a coordination number (CN) of 5.1 formed within 15 min, probably in the supercage of USY. The growth followed first-order kinetics with respect to the concentration of Pd2+ and the CN of Pd−Pd. The clusters were stable up to 443 K. The clusters were partially oxidized in less than 2 min on exposure to 8% O2, and then quickly reduced with 8% H2 (second reduction) to afford larger Pd clusters with CN = 6.7. The clusters continued to increase in size in a stepwise fashion on further alternate exposures to O2 and H2. The cluster growth proceeded gradually when 0.8 wt % Pd was loaded on H-USY. In contrast, stable Pd clusters were not obtained when NH4-USY or high-silica USY were used as supports, suggesting that the clusters were stabilized on the H+ ions present in H-USY. This demonstrates the potential of QXAFS in following the clustering process of Pd, and indicates the possibility for fine-tuning the sizes of Pd clusters simply by changing the number of O2 and H2 exposure cycles.

Structured bimetallic Pd-Pt/γ-Al2O3 catalysts on FeCrAlloy fibers for total combustion of methane

Bimetallic Pd-Pt/Al2O3 catalysts were deposited by dip coating on FeCrAlloy-type fibers. The influence of different gaseous pretreatments (under O2/H2 or H2 flows) on the physico-chemical properties and the catalytic activity in the total combustion of methane were investigated. Dispersion of Pd and Pt appeared to be a crucial parameter for the catalytic performances. All directly calcined catalysts displayed better catalytic behaviour than the pretreated ones. Long duration tests demonstrated that a bimetallic Pd-Pt catalyst has a better catalytic behaviour than a Pd-based one, which undergoes through sintering as evidenced by the formation of larger agglomerates. After the long duration test, palladium is in a fully oxidized state (PdO) (as well as platinum (PtO)). It is suggested that the better catalytic behaviour of the Pt-Pd catalyst could be due to an active cooperation between PtO, formed during the early stages of the catalytic reaction, and PdO. Two hypotheses can be put forward; in the first case, PtO could supply oxygen atoms to PdO, which actually would play the active role in the catalytic reaction. In the second case, CH4 would be adsorbed and dissociated on Pt atoms, and PdO would supply oxygen atoms leading to the formation of CO2.

Spatial profiling of H(n = 2) atom number densities in a dc arc jet reactor

Spatially resolved measurements of the absorption by H(n = 2) atoms on the Balmer-α transition in the plume of a dc arc jet reactor, operating at an input power of 6.4 kW and with an Ar/H2 feedstock gas mixture, are used to extract radially dependent H(n = 2) atom number densities at flows of H2 of 0.2, 0.5, 0.8 and 1.0 slm (standard litres per minute). The analysis to obtain number densities employs inverse Abel transformation of measured column densities and assumes cylindrical symmetry within the gas plume. The measured number density distributions are compared with the outcomes of a computer model of the arc jet and show quantitative agreement. Annular structure in the H(n = 2) distributions, evident at the lower added H2 flows, is a consequence of Ar+ reaction with H2 molecules diffusing radially into the plume from the cooler periphery of the reactor, followed by a dissociative electron attachment to ArH+ ions. At the higher H2 flows, the H(n ⩾ 2) distribution retracts towards the nozzle through which the plasma enters the reactor and peaks on the central axis, giving a conical structure to the luminous gas plume. These variations of plume structure with added H2 flow are well described by the sophisticated two-dimensional model, which includes heat, mass and radiation transfer and chemical kinetics of the expanding argon–hydrogen plasma, gas-surface processes at the substrate and reactor walls and carefully characterized initial conditions for the gas expansion from the nozzle orifice into the reactor chamber.

Growth of chemical vapor deposition aluminum titanate films at different CO2/H2 and aluminum butoxide inputs

Amorphous aluminum titanate films are prepared on silicon substrates by low-pressure chemical vapor deposition (CVD) using a mixture of aluminum tri-sec-butoxide (ATSB), titanium tetrachloride (TiCl4), CO2, and H2 as the reactants (the ATSB/TiCl4/CO2/H2 system). The effects of the CO2/H2 and ATSB inputs and substrate temperature on the growth, microstructure, and composition of the CVD Al2O3–TiO2 films are discussed. The films have an increased growth rate and an increased Ti content at lower temperatures. The adsorption-controlled reaction is identified, which is attributed to the gas/solid reaction to weaken the film/substrate interface. The growth rates are also higher at higher H2 and ATSB flows. The film thickness is 0.47–1.13 μm for the CO2/H2-varying system and of 0.34–1.37 μm for the ATSB-varying system at deposition temperatures of 350–500 °C. The proposed reactions are presented to explain the film growth. The determined adsorption energy can explain the effect of temperature on composition.

In situ measurements of GaN nucleation layer decompostion

GaN nucleation layer (NL) decomposition was measured using optical reflectance over a wide range of pressure P, temperature T, and H2/NH3 gas mixture. The GaN NLs show measurable decomposition above 800 °C and do not significantly roughen until above 960 °C. The NL decomposition rates increase with increasing P, increasing T, and decreasing NH3 flow. An activation energy EA of 2.68 eV was measured (from 820 to 960 °C) for NL decomposition and an EA of 2.62 eV was measured (from 900 to 1075 °C) for decomposition of thick, high-T bulk GaN films. Depending on P, the pre-exponential factor A0 was four to nine times larger for NL decomposition compared to bulk GaN decomposition. The EA measured for both NL and bulk GaN decomposition in mixed H2 and NH3 flows is similar to the EA for Ga desorption, suggesting that the rate-limiting step for both NL and bulk GaN decomposition is Ga desorption.

Hydrogenotrophic denitrification of drinking water using a hollow fibre membrane bioreactor

Research Article| May 01 2001 Hydrogenotrophic denitrification of drinking water using a hollow fibre membrane bioreactor Sarina J. Ergas; Sarina J. Ergas 1Department of Civil and Environmental Engineering, University of Massachusetts, 18 Marston Hall, Amherst, MA 01003, USATel: 413-545-3424; Fax: 413-545-2202; E-mail: ergas@ecs.umass.edu, http://www.ecs.umass.edu/cee/faculty/ergas.htmlgradient E-mail: ergas@ecs.umass.edu Search for other works by this author on: This Site PubMed Google Scholar Andreas F. Reuss Andreas F. Reuss 2Procter & Gamble GmbH & Co. Manufacturing, OHG Industriepark Am Silberberg 53877, Euskirchen, Germany, 2251-121030 E-mail: reuss.a@pg.com Search for other works by this author on: This Site PubMed Google Scholar Journal of Water Supply: Research and Technology-Aqua (2001) 50 (3): 161–171. https://doi.org/10.2166/aqua.2001.0015 Views Icon Views Article contents Figures & tables Video Audio Supplementary Data Share Icon Share Twitter LinkedIn Tools Icon Tools Cite Icon Cite Permissions Search Site Search Dropdown Menu toolbar search search input Search input auto suggest filter your search All ContentAll JournalsThis Journal Search Advanced Search Citation Sarina J. Ergas, Andreas F. Reuss; Hydrogenotrophic denitrification of drinking water using a hollow fibre membrane bioreactor. Journal of Water Supply: Research and Technology-Aqua 1 May 2001; 50 (3): 161–171. doi: https://doi.org/10.2166/aqua.2001.0015 Download citation file: Ris (Zotero) Reference Manager EasyBib Bookends Mendeley Papers EndNote RefWorks BibTex The objective of this research was to investigate the performance of a hollow fibre membrane bioreactor (HFMB) for hydrogenotrophic denitrification of contaminated drinking water. In the HFMB, H2 flows through the lumen of the hydrophobic hollow fibres and diffuses to an attached H2 oxidizing biofilm. Nitrate in the contaminated water serves as an electron acceptor. A hydrogenotrophic denitrifying culture was enriched from a wastewater seed. Batch culture experiments were conducted to compare heterotrophic (methanol as electron donor) and hydrogenotrophic denitrification rates and to investigate the conditions required for the HFMB studies. The batch cultures demonstrated mixotrophy, with denitrification rates of 30 g NO−3-N m−3 d−1 for heterotrophic and 18 g NO−3-N m−3 d−1 for hydrogenotrophic conditions. A laboratory-scale HFMB was constructed that utilized 2,400 polypropylene hollow fibres with an inner diameter of 200 µm, an outer diameter of 250 µm and a 0.05 µm pore size. After a 70-day start-up period, the NO−3 loading rate was gradually increased over a three-month period. The NO−3 utilization rate reached a maximum of 770 g NO−3-N m−3 d−1 at an influent NO−3 concentration of 145 mg NO−3-N l−1 and a hydraulic residence time of 4.1 hours. Influent NO−3 concentrations of up to 200 mg NO−3-N l−1 were almost completely denitrified. Tests with contaminated water from the Cape Cod aquifer resulted in an increase in product water turbidity and dissolved organic carbon (DOC) concentrations. biological denitrification, bioreactor, drinking water, hydrogen, membranes, nitrate This content is only available as a PDF. © IWA Publishing 2001 You do not currently have access to this content.

Hydrogenated amorphous silicon deposited at very high growth rates by an expanding Ar–H2–SiH4 plasma

The properties of hydrogenated amorphous silicon (a-Si:H) deposited at very high growth rates (6–80 nm/s) by means of a remote Ar–H2–SiH4 plasma have been investigated as a function of the H2 flow in the Ar–H2 operated plasma source. Both the structural and optoelectronic properties of the films improve with increasing H2 flow, and a-Si:H suitable for the application in solar cells has been obtained at deposition rates of 10 nm/s for high H2 flows and a substrate temperature of 400 °C. The “optimized” material has a hole drift mobility which is about a factor of 10 higher than for standard a-Si:H. The electron drift mobility, however, is slightly lower than for standard a-Si:H. Furthermore, preliminary results on solar cells with intrinsic a-Si:H deposited at 7 nm/s are presented. Relating the film properties to the SiH4 dissociation reactions reveals that optimum film quality is obtained for conditions where H from the plasma source governs SiH4 dissociation and where SiH3 contributes dominantly to film growth. Conditions where ion-induced dissociation reactions of SiH4 prevail and where the contribution of SiH3 to film growth is much smaller lead to inferior film properties. A large contribution of very reactive (poly)silane radicals is suggested as the reason for this inferior film quality. Furthermore, a comparison with film properties and process conditions of other a-Si:H deposition techniques is presented.

A Multiwavelength Study of Outflows in OMC-2/3

We present new v = 1–0 S(1) H2, 12CO J = 2 → 1, and 12CO J = 3 → 2 observations of the star-forming clouds OMC-2 and OMC-3, one of the densest known groupings of outflows from low-mass young stellar objects (YSOs) in the sky. High-velocity 12CO J = 2 → 1 gas in this region suggests that previously discovered H2 flows are driving and entraining molecular outflows. However, the large number of sources and flows within the narrow molecular filament means it is difficult to make a firm association of molecular outflow gas with H2 flows, except for in the case of the bipolar east-west H flow. A number of Herbig-Haro (HH) objects, including ones far to the west and east of the main ridge, are identified with H2 knots. High-resolution spectroscopy in the v = 1–0 S(1) line of 10 H2 knots shows line profiles consistent with dual forward and reverse shocks. C-shock modeling suggest that asymmetries seen in suspected bow shocks could be evidence of varying magnetic field orientations throughout the cloud. One of the bow shocks in the H flow, YBD-5, can be successfully modeled by a 100 km s-1 C-shock propagating into a magnetized, 106 cm-3 medium, although the observations and limitations within the computer code itself do not entirely rule out J-shocks. Mass spectra of the H flow are broken power laws, which might be evidence for a jet that has two entrainment mechanisms for accelerating ambient molecular gas into the outflow. The H2 luminosity in this flow is many times smaller than the CO mechanical luminosity, but this fact cannot rule out the possibility that a narrow highly collimated jet drives the molecular outflow, owing to uncertainties in extinction, outflow dynamic times, cooling contributions from other lines, and the wind model used. Outflows from OMC-2/3 are likely to contribute to the turbulent pumping of gas within the molecular ridge north of the Orion Nebula. High-velocity gas clumps north of the sources investigated here may represent evidence of additional undiscovered outflows from young stars.

GaN Decomposition in Ammonia

GaN decomposition is studied as a function of pressure and temperature in mixed NH3 and H2 flows more characteristic of the MOVPE growth environment. As NH3 is substituted for the 6 SLM H2 flow, the GaN decomposition rate at 1000 °C is reduced from 1×1016 cm−2 s−1 (i.e. 9 monolayers/s) in pure H2 to a minimum of 1×1014 cm−2 s−1 at an NH3 density of 1×1019 cm−3. Further increases of the NH3 density above 1×1019 cm−3 result in an increase in the GaN decomposition rate. The measured activation energy, EA, for GaN decomposition in mixed H2 and NH3 flows is less than the EA measured in vacuum and in N2 environments. As the growth pressure is increased under the same H2 and NH3 flow conditions, the decomposition rate increases and the growth rate decreases with the addition of trimethylgallium to the flow. The decomposition in mixed NH3 and H2 and in pure H2 flows behave similarly, suggesting that surface H plays a similar role in the decomposition and growth of GaN in NH3.

Formation of cationic silicon clusters in a remote silane plasma and their contribution to hydrogenated amorphous silicon film growth

The formation of cationic silicon clusters SinHm+ by means of ion–molecule reactions in a remote Ar–H2–SiH4 plasma is studied by a combination of ion mass spectrometry and Langmuir probe measurements. The plasma, used for high growth rate deposition of hydrogenated amorphous silicon (a-Si:H), is based on SiH4 dissociation in a downstream region by a thermal plasma source created Ar–H2 plasma. The electron temperature, ion fluence, and most abundant ion emanating from this plasma source are studied as a function of H2 admixture in the source. The electron temperature obtained is in the range of 0.1–0.3 eV and is too low for electron induced ionization. The formation of silicon containing ions is therefore determined by charge transfer reactions between ions emanating from the plasma source and SiH4. While the ion fluence from the source decreases by about a factor of 40 when a considerable flow of H2 is admixed in the source, the flux of cationic silicon clusters towards the substrate depends only slightly on this H2 flow. This implies a strong dissociative recombination of silicon containing ions with electrons in the downstream region for low H2 flows and it causes the distribution of the cationic silicon clusters with respect to the silicon atoms present in the clusters to be rather independent of H2 admixture. The average cluster size increases, however, strongly with the SiH4 flow for constant plasma source properties. Moreover, it leads to a decrease of the ion beam radius and due to this, to an increase of the ion flux towards the substrate, which is positioned in the center of the beam. Assuming unity sticking probability the contribution of the cationic clusters to the total growth flux of the material is about 6% for the condition in which solar grade a-Si:H is deposited. Although the energy flux towards the film by ion bombardment is limited due to the low electron temperature, the clusters have a very compact structure and very low hydrogen content and can consequently have a considerable impact on film quality. The latter is discussed as well as possible implications for other (remote) SiH4 plasmas.