Recovery of carboxylic acids C1?C3 with organophosphine oxide solvating extractants

This study was aimed at examining the use of the organophosphine oxides Cyanex^®921 and Cyanex^®923 for the extraction of formic, acetic and propionic acids from aqueous solutions. The stripping of monocarboxylic acids with water from the loaded extractants was also examined. The studies were aimed at determining the equilibrium conditions for extraction and stripping. Overall, the effect of the kind of extractant was not significant although Cyanex^®921 extracted carboxylic acids slightly better than Cyanex^®923 with 1:1 complexes being formed by both extractants with the acids during extraction. The efficiency of extraction depended on temperature, acid concentration and solvent, with toluene a better diluent for the extractants than octane or Exxsol^®D 220/230. Extraction efficiency increased as the concentration of acid in the feed decreased and, also, as the temperature increased, the amount of acid extracted decreased. The extraction and stripping isotherms were determined. The apparent enthalpy and entropy of the extraction reaction were determined. Distribution data for the transfer of carboxylic acids from aqueous (NaCl) solutions to organic solvents in the presence of trialkylphosphine oxide were determined at 293 K with the distribution ratios increasing as the concentration of NaCl increased. Copyright © 2005 Society of Chemical Industry     

Extraction and pertraction of phenol through bulk liquid membranes

The extraction and pertraction of phenol through a bulk liquid membrane (BLM) with Cyanex^® 923, Amberlite^® LA‐2 and trioctylamine (TOA) as carriers were studied. Cyanex^® 923 was selected as the best carrier for pertraction. The distribution coefficient of phenol for solvents with carrier and pure n‐alkanes, the individual mass‐transfer coefficient at the extraction interface and the initial flux of phenol through the extraction interface (J_Fo) decreased in the order: Cyanex^® 923 > Amberlite^® LA‐2 > TOA ? pure n‐alkanes. The opposite order was observed for the value of the mass‐transfer coefficient in BLM and the maximum flux of phenol through the stripping interface (J_Rmax). At constant driving forces the maximum fluxes through the extraction and stripping interfaces were similar when amine carriers were used. However, J_Rmax was lower than J_Fo for Cyanex^® 923. Although the kinetics of stripping was the rate‐determining step, the flux of phenol was significantly higher than in pertraction with amine carriers. The adsorption of the carrier at aqueous phase/membrane interfaces was probably responsible for the rapid and slow transfer of phenol through the extraction and stripping interface, respectively. Copyright © 2004 Society of Chemical Industry     

Solvent extraction of titanium from nitrate medium using some organophosphorus extractants

The extraction of Ti(IV) from nitric acid has been carried out using some organophosphorus extractants in kerosene. The Ti(IV) extraction in the investigated systems is endothermic. Oxalic acid was effective for stripping Ti(IV). The maximum loading of Ti(IV) was found to be 6.13 × 10^?3, 6.175 × 10^?3 and 5.005 × 10^?3 M of Ti(IV) per mole extractant for CYANEX 925, 923 and 921, respectively, after five extraction stages. Kerosene is a more effective diluent for extraction of Ti(IV). FT-IR characterization and the separation of Ti(IV) are also discussed.     

Extraction and Separation of Zr(IV) and Hf(IV) from Nitrate Medium by Some CYANEX Extractants

Abstract

A solvent extraction study has been carried out to extract and separate zirconium and hafnium from nitrate medium by using some phosphine oxide extractants (CYANEX 921, CYANEX 923, and CYANEX 925) in kerosene. The influence of the different factors affecting the extraction process was studied in detail. Apparently the rate of extraction of Zr(IV) and Hf(IV) in CYANEX 921, CYANEX 923, and CYANEX 925 is reasonably fast. The extraction increases with increasing temperature, suggesting that the reaction is endothermic. The stripping percent of Zr(IV) and Hf(IV) by 0.5 M HNO₃ from the loaded organic phase after two stages reached 97.5% and 10.2%, respectively, which lead to good separation of the two metals. Under the optimum conditions, the extraction of zirconium was about 90, 87.6, and 91.6% and separation factors equal to 17, 21.4, and 40.7 were obtained for CYANEX 921, CYANEX 923, and CYANEX 925, respectively. The results obtained reveal that 2.0 M nitric acid is the optimum acid concentration for the separation of Zr(IV) and Hf(IV) and 0.4 M CYANEX 925 performs more efficient separation compared with other organophosphorus extractants.     

Separation of As(III) and As(V) by hollow fiber supported liquid membrane based on the mass transfer theory

Separation of As(III) and As(V) ions from sulphate media by hollow fiber supported liquid membrane has been examined. Cyanex 923 was diluted in toluene and used as an extractant. Water was used as a stripping solution. The extractability of As(V) was higher than As(III). When the concentration of sulphuric acid in feed solution and Cyanex 923 in liquid membrane increased, more arsenic ions were extracted into liquid membrane and recovered into the stripping solution. The mathematical model was focused on the extraction side of the liquid membrane system. The mass transfer coefficients of the aqueous phase (k _i ) and organic phase (k _m ) are 7.15×10⁻³ and 3.45×10⁻² cm/s for As(III), and 1.07×10⁻² and 1.79×10⁻² cm/s for As(V). Therefore, the rate-controlling step for As(III) and As(V) in liquid membrane process is the mass transfer in the aqueous film between the feed solution and liquid membrane. The calculated mass transfer coefficients agree with the experimental results.     

Extraction of iron(III) from acidic sulfate solutions with bis(2‐ethylhexyl)phosphoric acid in PENRECO® 170 ES,a new friendly diluent

The extraction of iron(III) from acidic sulfate solutions by bis(2‐ethylhexyl)phosphoric acid (HDEHP) is investigated by using PENRECO® 170 ES as a diluent. PENRECO® 170 ES is new diluent which offers advantages such as improved solvency power, more complete phase disengagement and reduced losses in aqueous streams, with reductions of over 50% in diluent usage after 1 year, compared with conventional paraffinic diluents. The chemical analyses performed in the present work suggest that such properties arise, at least in part, from the presence of a series of hydrophobic branched alcohols in its composition (at least 0.6 mol dm^?3). In spite of the solvation effects due to these alcohols, HDEHP is dimeric in this diluent and, in the presence of an excess of HDEHP, the extraction of iron(III) takes place according to the classical equation: with K_ex = 10^{5.7 ± 0.2} (at I = 1 mol dm^?3). Such a value of K_ex is similar to that reported for pure hexane, which shows that the presence of long chain alcohols in PENRECO® 170 ES has no perceptible influence on the thermodynamics of iron(III) extraction by HDEHP. The extraction of iron(III) by HDEHP in PENRECO® 170 ES is slightly more rapid than in kerosene, which indicates that the molecules of alcohols constituting PENRECO® 170 ES have no negative effect on the kinetics of metal extraction although they compete with the extractant molecules for adsorption at the liquid–liquid interface. Stripping of iron(III) from loaded organic solutions by sulfuric acid is easy and rapid (95% equilibrium reached within 2 min) when HDEHP is used at moderate concentrations (typically 0.1 mol dm^?3). At higher HDEHP concentrations, stripping is difficult and incomplete, as found previously with other diluents. Thus, PENRECO® 170 ES is interesting in its ability to overcome some of the physical problems encountered in liquid–liquid operations, but its use does not modify significantly the chemistry of iron(III) extraction by HDEHP. © 2002 Society of Chemical Industry     

Extraction of Lanthanum and Samarium from Nitrate Medium by some Commercial Organophosphorus Extractants

Abstract

The extraction of lanthanum(III) and samarium(III) from nitrate solutions by some phosphine oxide compounds (Cyanex 921, Cyanex 923, and Cyanex 925) in kerosene was investigated. The influence of the different factors affecting the extraction was studied in detail. The extraction of these metals using the above extractants was compared and the sequence of extraction was found to be Cyanex 921>Cyanex 923?Cyanex 925. The stripping percent of La(III) and Sm(III) by 0.75 M HNO₃ from the loaded organic phase after two stages were 72% and 5.2%, respectively, which could enable a good separation between these two lanthanides.     

Solvent extraction and purification of sugars from hemicellulose hydrolysates using boronic acid carriers

Research was performed to determine whether it was technically feasible to use boronic acid extractants to purify and concentrate the sugars present in hemicellulose hydrolysates. Initially, five types of boronic acids (phenylboronic acid, 3,5‐dimethylphenylboronic acid, 4‐tert‐butylphenylboronic acid, trans‐β‐styreneboronic acid or naphthalene‐2‐boronic acid) dissolved in an organic diluent (Shellsol^® 2046 or Exxal^® 10) containing the quaternary amine Aliquat^® 336 were tested for their ability to extract sugars (fructose, glucose, sucrose and xylose) from a buffered, immiscible aqueous solution. Naphthalene‐2‐boronic acid was found to give the greatest extraction of xylose regardless of which diluent was used. Trials were then conducted to extract xylose and glucose from solutions derived from the dilute acid hydrolysis of sugar cane bagasse and to then strip the loaded organic solutions using an aqueous solution containing hydrochloric acid. This produced a strip solution in which the xylose concentration had been increased over 7× that of the original hydrolysate while reducing the concentration of the undesirable acid‐soluble lignin by over 90%. Hence, this process can be exploited to produce high concentration xylose solutions suitable for direct fermentation. Copyright © 2004 Society of Chemical Industry     

Extraction and Separation of Germanium Using Cyanex 301/Cyanex 923. Its Recovery from Transistor Waste

Abstract

The extraction of Ge(IV) from HCl, HNO₃ and H₂SO₄ media in toluene solution of Cyanex 301 and Cyanex 923 is investigated. It is almost quantitatively extracted (～95%) in Cyanex 301 and Cyanex 923 at 8 molL^?1 HCl but the extractions from H₂SO₄ and HNO₃ are poor in the entire investigated range of acid molarity. Detailed investigations were carried out from HCl medium. Based on the slope analysis data the extracting species is identified as GeCl₄·2R (R=Cyanex 301/Cyanex 923). The extraction of Ge(IV) is higher and comparable in diluents like toluene, n‐hexane and kerosene (160–200°C) and there is no correlation between the dielectric constant and the percent extraction. The extractants are stable towards prolonged acid contact and there is negligible loss in their extraction efficiency even after recycling them for several cycles. The extraction behavior of commonly associated metal ions namely As(V)/(III), Sn(IV), Tl(III), In(III), Ga(III), Fe(III), Al(III), Hg(II), and Cu(II) has also been investigated. Based on the partition data conditions for attaining some binary and ternary separations involving Ge(IV) have been optimized. The separation data have been fused to develop a scheme for the recovery (93%) of pure germanium (～99%) from semi conductor waste.     

Solvent extraction separation of titanium(IV), vanadium(V) and iron(III) from simulated waste chloride liquors of titanium minerals processing industry by the trialkylphosphine oxide Cyanex 923

The solvent extraction of magnesium(II), aluminium(III), titanium(IV), vanadium(V), chromium(III), manganese(II) and iron(III) from hydrochloric acid solutions has been investigated using the trialkylphosphine oxide Cyanex 923 (TRPO) in kerosene as extractant. The results demonstrate that titanium(IV), vanadium(V) and iron(III) are extracted into kerosene as TiOCl₂·2TRPO, VO₂Cl·TRPO and HFeCl₄·2TRPO, respectively. On the other hand magnesium(II), aluminium(III), chromium(III) and manganese(II) are not extracted with TRPO from hydrochloric acid solutions (1.0–4.0 mol dm^?3) under the experimental conditions. IR spectral studies of the extracted complexes were further used to clarify the nature of the extracted complexes. The effect of the diluent on the extraction of titanium(IV), vanadium(V) and iron(III) has been studied and correlated with the dielectric constant. The loading capacity of the TRPO system has been evaluated and the potential for the separation and recovery of titanium(IV), vanadium(V) and iron(III) from simulated waste chloride liquors of the titanium minerals processing industry has been assessed. Copyright © 2004 Society of Chemical Industry     

Reactive extraction of itaconic acid using tri‐n‐butyl phosphate and aliquat 336 in sunflower oil as a non‐toxic diluent

Itaconic acid finds a place in various industrial applications. It can be produced by biocultivation in a clean and environment friendly route but recovery of the acid from the dilute stream of the bioreactor is an economic problem. Reactive extraction is a promising method to recover carboxylic acid but suffers from toxicity problems of the diluent and extractant employed. So there is need for a non‐toxic extractant and diluent or a combination of less toxic extractants in a non‐toxic diluent that can recover acid efficiently. Effect of different extractants: tri‐n‐butylphosphate (TBP) (an organophosporous compound) and Aliquat 336 (a quaternary amine) in sunflower oil was studied to find the best extractant–sunflower oil combination. Equilibrium complexation constant, K_E, values of 1.789 and 2.385 m³ kmol^?1, respectively, were obtained for itaconic acid extraction using TBP and Aliquat 336 in sunflower oil. The problem of toxicity in reactive extraction can be reduced by using a natural non‐toxic diluent (sunflower oil) with the extractant. Copyright © 2010 Society of Chemical Industry     

Equilibrium and dynamic studies of the removal of As(III) and As(V) from contaminated aqueous systems using a functionalized biopolymer

BACKGROUND: A study of the removal of arsenic from a sample of actual groundwater using crosslinked xanthated chitosan is described. RESULTS: Removal of As(III) and As(V) was studied at pH 7.5 under equilibrium and dynamic conditions. The equilibrium data were fitted to Langmuir and Freundlich adsorption models and the various model parameters evaluated. The monolayer adsorption capacity from the Langmuir model for xanthated chitosan flakes (XCF) (As(V) 20.0 ± 0.56 mg g^?1; As(III) 33.0 ± 0.32 mg g^?1) were lower than obtained for xanthated chitosan granules (XCB) (As(V) 36.0 ± 0.52 mg g^?1; As(III) 48.0 ± 0.45 mg g^?1). Adsorption of As (V) was unaffected by the presence of other anions while in the case of As(III) the presence of sulfate and silicate caused a 26.5–50.9% decrease in adsorption. A sample (940 bed volumes) of a groundwater spiked with 200 µg L^?1 As(V) treated with XCF in column experiments reduced the arsenic concentration to < 10 µg L^?1. The adsorbent was also successfully applied for the removal of total inorganic arsenic down to < 10 µg L^?1 from real samples of arsenic‐contaminated groundwater. CONCLUSION: Xanthated chitosan was an efficient adsorbent for the removal of both forms of arsenic from groundwater under near neutral conditions. The presence of sulfur and the amino groups resulted in increased adsorption capacity of the sorbent. Copyright © 2012 Society of Chemical Industry     

Reactive extraction of propionic acid using tri‐n‐octylamine,tri‐n‐butyl phosphate and aliquat 336 in sunflower oil as diluent

BACKGROUND : Propionic acid is widely used in chemical and allied industries and can be produced by biocultivation in a clean and environmentally friendly route. Recovery of the acid from the dilute stream from the bioreactor is an economic problem. Reactive extraction is a promising method of recovering the acid but suffers from toxicity problems of the solvent employed. There is thus a need for a non‐toxic solvent or a combination of less toxic extractants in a non‐toxic diluent that can recover acid efficiently. RESULTS: The effect of different extractants (tri‐n‐butylphosphate (TBP), tri‐n‐octylamine (TOA) and Aliquat 336) and their mixed binary solutions in sunflower oil diluent was studied to find the best extractant‐sunflower oil combination. Equilibrium complexation constant, K_E, values of 4.02, 3.13 and 1.87 m³ kmol^?1 were obtained for propionic acid extraction using Aliquat 336, TOA and TBP, respectively, in sunflower oil. The effect of different modifiers (1‐decanol, methylisobutyl ketone, butyl acetate and dodecanol) on the extraction was also studied and it was found that modifiers enhance extraction, with 1‐decanol found to be the best. CONCLUSION: The problem of toxicity in reactive extraction can be reduced by using a non‐toxic diluent (sunflower oil) or a modifier in a non‐toxic solvent, with the extractant. The addition of modifiers was found to improve the extraction. Copyright © 2008 Society of Chemical Industry     

The Extraction and Separation of Ho,Y, and Er(III) with the Mixtures of Cyanex 302 and Another Organic Extractant

Abstract

The extraction and separation of Ho, Y, and Er(III) with the mixtures of bis(2,4,4‐trimetylpentyl)monothiophosphinic acid (Cyanex 302) and another organic extractant, such as acidic organic extractant (di‐2‐ethylhexyl phosphoric acid P204, 2‐ethylhexyl phosphoric acid mono‐2‐ethylhexyl ester P507, di‐2‐ethylhexyl phosphinic acid P229, and sec‐nonylphenoxy acetic acid CA‐100), neutral organic extractant (tri‐n‐butyl phosphate TBP, di‐(1‐metylheptyl)metyl phosphate P350, and branched trialkylphosphinic oxide Cyanex 925) or primary amine N1923, has been investigated in this paper. The extractability and separation ability for the Ho, Y, and Er with the mixtures of Cyanex 302 and organic extractants has been compared. The synergistic effect of the Ho, Y, and Er extraction with the mixtures of Cyanex 302 and P229, Cyanex 925, CA‐100, or N1923 has been explored and the synergistic enhancement coefficients have been calculated. At last, the Y³⁺ synergistic extraction with the mixtures of Cyanex 302 and CA‐100 has been determined and the extracted complex has been deduced.     

Recovery of rare earth elements from simulated fluorescent powder using bifunctional ionic liquid extractants (Bif‐ILEs)

BACKGROUND: Numerous high purity ammonium‐type ionic liquid extractants have been prepared for engineering purposes. Bifunctional ionic liquid extractants (Bif‐ILEs) have been widely applied to separate and extract rare earths and metal ions with high extraction efficiencies and selectivities. In the present study, new Bif‐ILEs [A336][P204] and [A336][P507] have been used to extract rare earths from a simulated solution of a fluorescent powder in a high concentration of Al(NO₃)₃. RESULTS: Bif‐ILEs were prepared from Aliquat336 (A336) and the commercial organophosphorus acid extractants, P204 and P507. These extractants [A336][P204] and [A336][P507] have similar characteristics to neutral organophosphorus extractants. When these Bif‐ILEs were used to extract RE(III) from a simulated waste fluorescent powder system a third phase appeared which could be eliminated by the addition of 10% isopropanol modifier. The coexisting Al₂O₃ in the fluorescent powder was changed to a salting‐out agent (Al(NO₃)₃) in the extraction process and promoted the extraction efficiency of RE(III). Using a countercurrent extraction process at a phase ratio V_o:V_w = 4:1 and pH = 0.56, the RE(III) recovery reached 95.2% in 5–7 stages. Finally, the extractabilities of these bifunctional extractants were compared with the neutral organophosphorus extractants P350, TBP and Cyanex923 at different concentrations, initial pHs and temperatures. CONCLUSIONS: By comparison with other neutral organophosphorus extractants, Bif‐ILEs [A336][P204] and [A336][P507] can be considered efficient potential extractants for separating and recycling REEs and Al₂O₃ from waste fluorescent powder. Copyright © 2011 Society of Chemical Industry     

Extraction separation of Ir(III) and Rh(III) with Cyanex 923 from chloride media: a possible recovery from spent autocatalysts

Liquid–liquid extraction of Ir(III) and Rh(III) with Cyanex 923 from aqueous hydrochloric acid media has been studied. Quantitative extraction of Ir(III) was observed in the range of 5.0–8.0 mol dm^?3 HCl with 0.1 mol dm^?3 Cyanex 923, while Rh(III) was extracted quantitatively in the range of 1.0–2.0 mol dm^?3 HCl with 0.05 mol dm^?3 Cyanex 923 in toluene along with 0.2 mol dm^?3 SnCl₂. The Ir(III) was back extracted with 4.0 mol dm^?3 HNO₃ quantitatively from the organic phase while Rh(III) was stripped with 3.0 mol dm^?3 HNO₃. The extraction of Rh(III) with Cyanex 923 was not quantitative without use of SnCl₂. However in the extraction of Ir(III) a negative trend was observed in the presence of SnCl₂. Varying the temperature of extraction showed that the extraction reactions of both the metal ions are exothermic in nature, and the stoichiometric ratio of Ir(III)/Rh(III) to Cyanex 923 in organic phase was found to be 1:3. The methods developed were applied to the recovery of these metal ions from a synthetic solution of similar composition to that from leaching of spent autocatalysts in 6.0 mol dm^?3 HCl. © 2002 Society of Chemical Industry     

Treatment of simulated arsenic contaminated groundwater using GAC‐Cu in batch reactor: Optimization of process parameters

This article deals with the experimental investigation related to the removal of arsenic from a simulated contaminated groundwater by the adsorption onto Cu²⁺ impregnated granular activated carbon (GAC‐Cu) in presence of impurities like Fe and Mn. The effects of adsorbent concentration, pH, and temperature on the percentage removal of total arsenic (As(T)), As(III), and As(V) have been discussed. Under the experimental conditions, the optimum adsorbent concentration for GAC‐Cu has been found to be 6 g/L with an agitation time of 24 h, which reduces the As(T) concentration from 188 to 8.5 µg/L. Maximum removal of As(V) and As(III) has been observed in the pH range of 7–9 and 9–11, respectively. Removal of all the above said arsenic species decreases slightly with increase in temperature. Presence of Fe and Mn increases the adsorption of arsenic species. Under the experimental conditions, at 30°C, maximum % removals of As(T), As(III), As(V), Fe, and Mn are found to be 95.5%, 93%, 98%, 100%, and 40%, respectively. It has also been observed that maximum regeneration (～94%) of spent GAC‐Cu is exhibited by a 5NH₂SO₄ solution.     

Vibrating screen printed electrode of gold nanoparticle-modified carbon nanotubes for the determination of arsenic(III)

A linear sweep anodic stripping voltammetric method using a carbon nanotube–gold nanoparticle-modified vibrating screen printed electrode for the determination of arsenic(III) is reported. The experiments were conducted with a 0.1 mol L^?1 solution of H₂SO₄ in order to estimate the electrode area related to gold oxide formation. The results showed a clear reduction peak at approximately +0.85 V corresponding to the reduction of the gold surface oxide with a superficial area of 0.089 cm². A vibrating motor was attached to the screen printed electrode to create a portable and autonomous system with enhanced mass transfer. The repeatability of the measurements was 2.4 % (n = 10) at the level of 0.5 mg L^?1 of arsenic(III) under the best instrumental operating conditions. The peak current was linearly dependent on the arsenic(III) concentration, thus allowing the construction of a linear analytical curve in the range from 10 to 550 μg L^?1 with the equation: ?I_p (μA) = 0.05 + 134.59 [As(III) (μg L^?1)], R² = 0.99. The obtained detection and quantification limits were 0.5 (3 SD) and 1.5 (10 SD) μg L^?1, respectively, using 120 s as the deposition time. It was shown that Cu(II) does not interfere in the detection of As(III) using the proposed method.     

Physicochemical properties,surface active species and formation of reverse micelles in the Cyanex 923‐n‐heptane/cerium(IV)‐H2SO4 extraction system

BACKGROUND: The Cyanex^® 923 (trialkylphosphine oxides, TRPO)‐n‐heptane/cerium(IV)‐H₂SO₄ extraction system has been investigated focusing on the physicochemical properties, surface active species and interfacial phenomena. The effects of H₂SO₄ and Ce(IV) extraction on them were considered. RESULTS: Results showed that the density and refractive index reflect the mass transfer by H₂SO₄ and Ce(IV) extraction and the change of refractive index was more sensitive than density. The interfacial tension decreased on extraction of H₂SO₄ but increased on extraction of Ce(IV). The viscosity of the equilibrium organic phase increased abruptly when the extracted H₂SO₄ concentration in the organic phase reached certain high values. The formation of reversed micelles, with mean diameter of about 10 nm, at high H₂SO₄ concentrations in the organic phase, is suggested by various measurements such as viscosity, interfacial tension and dynamic light‐scattering (DLS). CONCLUSION: It is suggested that TRPO‐H₂SO₄ complexes are more surface‐active than TRPO itself and tend to aggregate into reverse micelles by self‐assembling in the organic phase but the Ce(IV)‐TRPO complexes are neutral, less surface‐active than TRPO and not helpful for reverse micelle formation. Copyright © 2008 Society of Chemical Industry     

Solar‐light assisted removal of arsenic from natural waters: effect of iron and citrate concentrations

The optimal conditions to remove arsenic(III) from a solution were fitted using a factorial experimental design in a reaction catalysed by visible light (black light, 360 nm) and an iron‐citrate complex. Experiments were performed by simultaneously modifying the two variables affecting the removal of arsenic, i.e. iron and citrate concentrations. The single polynomial function obtained with the factorial design methodology indicates that the iron concentration was the most critical parameter in the removal of arsenic by precipitation. Mathematically, it was determined that the optimum molar ratio for arsenic, citrate and iron was 1:4.5:18.7, respectively, over 90% of arsenic being eliminated after 4 h of irradiation. The comparison between the As(III) and As(V) co‐precipitation rates indicates that almost 80% of As(III) was removed after 1 h of irradiation with black light, while As(V) required 4 h of irradiation to reach the same value. When natural water containing approximately 1 mg L⁻¹ of arsenic, only as As(V), was irradiated with solar light under optimised conditions, approximately 95% of the arsenic was removed after 1 h of irradiation. Copyright © 2006 Society of Chemical Industry