Application of a Purge-and-Trap System for Fixed-NH 4 + Determination by Matrix Interference Free Ion Chromatography in Oil Reservoir Rocks

The determination of fixed-NH 4 + in sedimentary rocks has been applied in geochemical studies for exploration purposes, especially in the oil industry. The studies are based on a conceptual model of NH 4 + fixation in oil source rocks during the burial diagenesis. This model explains why larger relative quantities of NH 4 + are fixed in transformed authigenic clays (e.g. illite, illite-smectite or illite-smectitevermiculite), in sandstones and is generally consistent with depth-diagenesis schemes for clay minerals and generation of hydrocarbons from the thermal maturation of organic matter. Relative changes in fixed-NH 4 + concentrations serve as indicators of hydrocarbons formation or may indicate their pathways during the migration from source beds to reservoirs. The wet analytical methods for fixed-NH 4 + determination have been carried out in three steps: (i) previous removal of inorganic exchangeable-N with an extractant (e.g., 2 mol L KCl), followed by the organic-N removal with an oxidant reagent (e.g., KOBr or NaOCl) 2,4,9 or removal of exchangeable-N and organic nitrogen species with 1 mol L KOH; (ii) release fixed-NH 4 + by destruction of the silicates with a HF and HCl mixture and (iii) Kjeldahl steam distillation. The ammonium concentration in the solution obtained after the distillation has been determined by titrimetric; or colorimetric methods. A limit of detection (LOD) of 56 mg kg for fixed-NH 4 + determination in sedimentary rocks was reported. A simplified purge-and-trap system for NH 3 extraction was recently used for total nitrogen determination. This system presented many advantages, among them, the matrix interference elimination, mainly Na, which was necessary for NH 4 + determination by ion chromatography. In the present work, this system was proposed for fixedNH 4 + determination. All results were compared with those obtained with the classical Kjeldahl method. Sandstones from Brazilian oil reservoirs with different fixed-ammonium were analyzed.


Introduction
2][3][4][5][6][7][8] The studies are based on a conceptual model of NH 4  + fixation in oil source rocks during the burial diagenesis.This model explains why larger relative quantities of NH 4 + are fixed in transformed authigenic clays (e.g.illite, illite-smectite or illite-smectitevermiculite), in sandstones and is generally consistent with depth-diagenesis schemes for clay minerals and generation of hydrocarbons from the thermal maturation of organic matter.Relative changes in fixed-NH 4 + concentrations serve as indicators of hydrocarbons formation or may indicate their pathways during the migration from source beds to reservoirs. 2,4he wet analytical methods for fixed-NH 4 + determination have been carried out in three steps: (i) previous removal of inorganic exchangeable-N with an extractant (e.g., 2 mol L -1 KCl), followed by the organic-N removal with an oxidant reagent (e.g., KOBr or NaOCl) 2,4,9 or removal of exchangeable-N and organic nitrogen species with 1 mol L -1 KOH; (ii) release fixed-NH 4 + by destruction of the silicates with a HF and HCl mixture 1,10 and (iii) Kjeldahl steam distillation. 2,4,9,10The ammonium concentration in the solution obtained after the distillation has been determined by titrimetric; 2,4,10 or colorimetric methods. 11A limit of detection (LOD) of 56 mg kg -1 for fixed-NH 4 + determination in sedimentary rocks was reported. 2 simplified purge-and-trap system for NH 3 extraction was recently used for total nitrogen determination. 12This system presented many advantages, among them, the matrix interference elimination, mainly Na + , which was necessary for NH 4 + determination by ion chromatography.In the present work, this system was proposed for fixed-NH 4 + determination.All results were compared with those obtained with the classical Kjeldahl method.Sandstones from Brazilian oil reservoirs with different fixed-ammonium were analyzed.

Apparatus and operating conditions
An US-bath model USC-1400 from Unique (Indaiatuba, SP, Brazil) with a power of 120 W, frequency Vol.21, No. 6, 2010 of 40 kHz and volume of 3 L to convert the released-NH 4 + into NH 3 vapor was used.A water-bath model 304-149 from Quimis Aparelhos Científicos (Diadema, SP, Brazil) was used to remove exchangeable-NH 4 + and organic-N from the samples by the proposed and comparative methods.A CELM centrifuge model LS-3 (Barueri, SP, Brazil) with 3200 rpm of speed was used.A Metrohm chromatography system (Herisau, Switzerland) with a dialysis membrane model 788, a vacuum degasser model 837 and a conductivity detector model 819 were used for NH 4 + determination.Integrated areas were used to quantify ion signals.Ammonium separation was carried out in a METROSEP C2 250/4.0column (Herisau, Switzerland) with 7.0 µm particle diameter.The flow-rate of the eluent (4.0 mmol L -1 tartaric acid/0.75mmol L -1 2,6-pyridinecarboxylic acid) was kept at 1.0 mL min -1 ; the column temperature was 35 °C and the running time was 15 min.The sample loop volume was 20 µL.The mineralogical characterization of all samples was carried out by X-ray diffraction (XRD, powder method) using a Bruker diffractometer model AXS D5005 equipped with Goeble mirror.Working conditions were: CoK a monochromatic radiation, 40 mA and 35 kV.Samples were run at a speed of 0.02 degree s -1 .Qualitative mineralogical determination was estimated with Diffrac Plus software.

Reagents, standards and samples
All chemicals used were of analytical grade.The water was purified with a reverse osmosis system model Elix 5 and deionized Milli-Q from Millipore Indústria e Comércio Ltda.(Barueri, SP, Brazil).The following solutions were used: 1 mol L -1 KOH, 7.5 mol L -1 HF, 1 mol L -1 HCl, 0.1 mol L -1 HCl and 1000 mg L -

Removal of exchangeable-N and organic-N
An aliquot of 25 mL of 1.0 mol L -1 KOH was added to a 50 mL polypropylene centrifuge tube containing 0.2 g of sample, pre-treated as Pontes et al. 12 described.The mixture was then heated at 80 °C for 8 h in a water-bath.After the mixture reached room temperature, the phases were separated by centrifugation.The supernatant was discarded, and the residue was washed three times with 5 mL of 1 mol L -1 KOH and finally with water.This procedure was adopted for the comparative and proposed methods.

Release of fixed-NH 4 +
An aliquot of 10 mL of a mixture containing 7.5 mol L -1 HF and 1.0 mol L -1 HCl was added to the polypropylene centrifuge tube containing the residue.The mixture was heated in a water-bath at 80 °C for 12-16 h, until it was almost dry.This procedure was adopted for both comparative and proposed methods.

Released-NH 4 + determination by the comparative method
After the mixture has reached the room temperature, it was transferred to a polypropylene distillation tube of the Kjeldahl system.Then, NH 3 was distilled, collected as NH 4 + and determined by the indophenol colorimetric method. 12monia extraction by the US-assisted purge-and-trap system (proposed method) The flow diagram for the NH 3 extraction by USassistance was similar to that showed by Pontes et al. 12 with the following modifications: all tubes were made of Teflon and the manifold were of polypropylene; the dimensions of the reaction tube and the NH 3 collection tube were 2.7 cm i.d. and 12 cm length; the dimensions of the U tube were 0.3 cm i.d. and 13 cm length; the manifold (0.6 cm i.d.) had six tips to connect six sets and one more tip to connect the vacuum pump.The polypropylene tube containing the released-NH 4 + in solution was connected to the U tube, and then the procedure was the same as Pontes et al. 12 described.

Analytical results
The working concentration of NH 4 + was in the range of 100 to 800 µg L -1 ; the linear regression of the analytical curve gave a slope of 0.1863 mV s L mg -1 and the correlation coefficient of the calibration curve was higher than 0.99.The blank measurement was 13 ± 2 µg L -1 NH 4 + (n = 10).The LOD was 4.4 µg g -1 NH 4 + .It was calculated from the equation LOD = 3S BL /b, where S BL was the standard deviation of ten blank concentration measurements and b was the slope of the calibration curve.The relative standard deviations (RSD) were lower than 11% (n = 3).Good agreement was found between the results of fixed-NH 4 + obtained by the proposed method and those obtained by the classical Kjeldahl method (Table 1).The sample diffractograms (not shown) indicated the presence of illite, which probably was responsible by the NH 4 + fixation.

Conclusion
The simplified method for ammonia extraction is faster and simpler than the classical Kjeldahl steam distillation method.The limit of detection (LOD) was 4.4 mg g -1 NH 4 + .The relative standard deviations (RSD) were lower than 11% (n = 3).

1 NH 4 +(NH 4
Cl) stock standard solution.The intermediate standard solutions were daily prepared.Sandstones from Brazilian oil reservoirs with different fixed-NH 4 + concentrations were analyzed.

Table 1 .
Concentrations (µg g -1 ) of the fixed-NH 4 + in sandstone samples obtained by the classical and proposed methods (n = 3)