ANaZyst, February, 1968, Vol. 93, j!@. 111-115 111 The Extraction and Determination of Disodium Octaborate in Sitka Spruce BY A. I. WILLIAMS (Forest Products Research Laboratory, Princes Risborough, Aylesbury. Bucks.) A method is described for the rapid determination of the boric acid equivalent of disodium octaborate tetrahydrate in Sitka spruce. The boron- containing compounds are leached from thin sections of wood with sodium hydroxide solution and determined by the spectrophotometric measurement of the red complex formed between boric acid and curcumin. The procedure is particularly useful for the study of the distribution of boron-containing preservatives in wood, THE current interest in the “Timbor” process for preserving timber against attack by wood- destroying fungi and insects has resulted in the need for a rapid, simple and accurate method for the determination of the boric acid equivalent of “Timbor” in treated wood.“Timbor” is the trade name of a water-borne preservative containing a crystallised mixture of boric acid and sodium borate that corresponds approximately to the formula Na2BS0,,.4H2O, disodium octaborate tetrahydrate. It is usual to express the disodium octaborate content as the boric acid equivalent. The “Timbor” process consists in immersing, or spraying, freshly converted green timber in disodium octaborate solution, then closely stacking and covering to prevent drymg, so that diffusion of the salt into the wet wood can take place over a period of time. To ensure that the treatment has been sufficient to meet the specification of the British Wood Preserving Association it is necessary to analyse the treated timber to determine the loading, distribution and penetration of preservative achieved during diffusion.There are available spray techniq~esl1~,~,~16,~ for the determination of loading and penetration, but they are often not accurate enough for research purposes involving diffusion and distribution experiments and weathering tests. Also there are methods based on ashing and leaching followed by titration,297 but although they have the required accuracy they are not practicable when large batches of specimens need to be examined, and cannot be used to determine the distribution of preservative over small areas. The problem posed requires for its solution, first, a simple quantitative procedure for separating the disodium octaborate from the matrix, and secondly, an accurate and precise method for determining 0.01 per cent. upwards of the boric acid equivalent.It was found that disodium octaborate tetrahydrate can be rapidly leached from thin sections of wood. Sitka spruce, Picea sitchertsis, a relatively impermeable species, was used for the experimental work. The low permeability of the wood did not cause any apparent difficulties during the leaching out of disodium octaborate when sections of the order of 200 to 300p in thickness were used. Determination of boron by the spectrophotometric measurement of the rosocyanins complex, formed between boric acid and curcumin, is a well known procedure and has been described by Hayes and Met~alfe.~ Therefore it was decided, by using these techniques, to evolve a method for the determination of the boric acid equivalent.EXPERIMENTAL To examine techniques for determining the boric acid equivalent of disodium octaborate in timber it was necessary to prepare standard samples containing a known amount of preservative. This was achieved by impregnating wood with solutions of known concen- trations of disodium octaborate (boric acid equivalent, 0.014 to 1-370 per cent. w/w) by the full cell process1* and freeze-drying.ll 0 SAC; Crown Copyright Reserved.112 WILLIAMS : EXTR.4CTION AND DETERMINATION [Analyst, VOl. 93 PREPARATION OF STANDARD SAMPLES- Weighed quarter-sawn blocks of Sitka spruce of known moisture content, with over-all dimensions of 3 x 2 x 1 cm, cross-section 3 x 1 cm, radial face 3 x 2 cm and tangential face 2 x 1 cm, were submerged in disodium octaborate solutions of known concentrations and subjected to a vacuum of 71 cm of mercury for 3 hours to evacuate the air from the wood cells.The vacuum was released, and a pressure of 7 kg per cm2 was applied to the solution containing the specimens for half an hour. The pressure was released, the specimens removed from the solution, their surfaces were superficially dried on filter-paper to remove excess of solution, and weighed. The wood absorbed about three times its own weight of solution. To prevent re-distribution of disodium octaborate the specimens were freeze-dried to a moisture content of about 7 per cent. From the observed figures the percentage of disodium octaborate, expressed as the boric acid equivalent based on oven-dry wood, was found by calculation to range from 0.042 to 4-16 per cent.The structure of the wood in the small blocks used for the experiment varies. In “Timborised” timber, concentration gradients of deposited disodium octaborate can occur across the annual rings; more will be found in the spring or early wood, as the void space is greater than in summer or late wood. Therefore, for development work on the procedure, it was decided to use thin cross-sections that are more representative of the preservative in the bulk of the wood. Also, for leaching purposes, the cross-section is far more permeable than the tangential or radial sections. Cross-sections up to 300 p thick, and weighing up to 60 mg, were cut on a microtome.Attempts were made to leach the disodium octaborate out of thin cross-sections of standard samples with cold curcumin - acetic acid reagent solution, but varying recoveries, usually less than 60 per cent. of the calculated preservative content, were observed. Warming did not improve the recovery. Thin sections were leached with 0.5 ml of M sodium hydroxide solution, which greatly improved the recovery, but to obtain complete removal of the di- sodium octaborate from the wood it was found necessary to warm the sodium hydroxide solution at 50” C for 10 minutes. After leaching, curcumin - acetic acid reagent solution was added, followed by a mixture of sulphuric acid and acetic acid to convert the disodium octaborate to boric acid and to dehydrate the solution, so that the reaction between boric acid and curcumin could take place. All the operations were carried out in soda-glass boiling tubes.The complex-forming reaction was complete within 15 minutes; the mixture was then diluted with a solution of acetone and water to a known volume and mixed. The thin sections of wood were left in the solution, which was decanted for spectrophotometric measurement. SPECTROPHOTOMETRIC MEASUREMENT- Measurements of optical densities of the test solutions were made at wavelength 666 m p by using a Unicam SP600 spectrophotometer with 1-cm cells. The spectrophotometer cali- bration graph was constructed in the range 0 to 72 pg of boric acid, and a straight-line relation- ship was obtained.RESULTS The procedure described was applied to thin cross-sections of the standard samples and the results are given in Table I. Three thin cross-sections, up to 300p thick, were taken from each specimen, one from the outside, the second from 0.6 cm in, and the third from the middle of the specimen. The results showed some small variation through the specimen, but the average of the three determinations was in agreement with the calculated over-all boric acid equivalent content. The standard deviation, based on nine determinations, was +O.OlS per cent. at the 0.4 per cent. boric acid level. These results show that the proposed procedure is quantitative and possesses the required speed and precision. As a result of the success of the procedure, thin radial sections up to 200 p thick were examined.The radial sections taken for analysis were narrow strips, so that a representative portion of both spring and summer wood was included in the sample. Three adjacent samples were taken from the surface, and three 06cm from the surface of the standard sample. The results were similar to those obtained for cross-sections. The average of each set of determinations was in agreement with the calculated boric acid equivalent content of the standard samples.February, 19681 OF DISODIUM OCTABORATE IN SITKA SPRUCE TABLE I RESULTS OBTAINED BY USING THIN CROSS-SECTIONS Calculated boric acid equivalent, per cent. found, per cent. Average Boric acid equivalent 4.16 4.08 4.06 3.80 4.30 2-14 2.03 1-29 1-20 0.71 0.76 0-41 0.38 0.19 0.1 8 0.085 0.081 0.044 0.040 2.1 I 2.27 2-16 1.28 1.23 1.24 0.73 0.72 0.73 0.40 0.40 0.40 0.1 8 0.1 6 0.18 0.084 0.076 0.081 0.042 0.041 0.042 113 Tangential sections, up to 200 p thick, of either spring or summer wood were analysed.As expected, the spring wood contained much more disodium octaborate than the summer wood. Therefore, unless a representative sample of both spring and summer wood is taken for analysis, inaccurate results will be obtained with thin tangential sections. It is difficult to sample tangential sections to include a representative amount of spring and summer wood because the gradual change of cell wall thickness cannot be accurately judged during sampling. The difficulty does not arise with cross-sections or radial sections where it is a simple operation to cut thin sections across several annual rings.TABLE I1 The results are given in Table 11. RESULTS OBTAINED BY USING THIS TANGENTIAL SECTIONS Standard Calculated boric acid Boric acid equivalent sample No. equivalent, per cent. found, per cent. Average 126 Spring wood 0.78 Summer wood 0.7s 130 Spring wood 0.35 Summer wood 0.38 0.9 1 1.01 1.10 0.6 1 0.64 0.67 0.4 1 0.4 1 0.4 1 0.25 0.26 0.24 To demonstrate the usefulness of the proposed method the distribution of the boric acid equivalent of disodium octaborate through a 2-inch (51 mm) thick piece of “Timberised” wood, which had been on a weathering test for 18 weeks during which the rainfall was 11+ inches (292 mm), was investigated. In the outer Q inch (3.2 mm) of opposite surfaces of the test piece, five determinations of the boric acid equivalent were made on radial sections at intervals of & inch (0.8 mm).Further determinations were made to complete the distribution pattern through the test piece. The curve obtained by plotting the boric acid equivalent against thickness is shown in Fig. 1. The eighteen determinations, including sample cutting, made to form the distribution curve were completed within 2 hours.114 WILLIAMS : EXTRACTION AND DETERMINATION [Autdyst, VOl. 93 It can be seen from Fig. 1 that it is possible to evaluate rapidly the distribution of boron- This is not possible by previously available containing preservatives over small areas. methods of analysis. I 1 1 I I I I I 0 a t d " t l t ' d Exposed side Inches Underside Fig.1. Distribution of disodium octaborate, expressed as the boric acid equivalent, in a weathered specimen METHOD APPARATUS- A Unicam SP600 spectrophotonaeter. Soda-glass boiling tubes, size 150 x 25 mm. Graduated fEasks, 100-ml capacity. Use high purity reagents when possible. Curcumin solution-Dissolve 0.12 g of curcumin in 100 ml of warm glacial acetic acid, SuZ~FLuric acid - acetic acid solution-Add 50 ml of sulphuric acid, slowly with cooling, Acetone - water solution-Dilute 250 ml of acetone to 500 ml with water. Sodium hydroxide solution, M-Dissolve 4 g of sodium hydroxide pellets in 50 ml of REAGENTS- cool, and store in a polythene bottle. to 50 ml of glacial acetic acid. Store in a stoppered soda-glass bottle. water and dilute to 100ml. Store in a polythene bottle. PROCEDURE- Weigh the sample and place in a soda-glass boiling tube, add 0.5 ml of M sodium hydroxide solution, then warm on a water-bath at 50" C for 10 minutes. Remove the boiling tube from the water-bath and add, from a burette, 3 ml of curcumin solution rinsing the boiling tube wall down with the reagent.Allow to stand for 5 minutes then add, from a burette, 3 ml of sulphuric acid - acetic acid solution, mix and allow to stand for 16 minutes. PourFebruary, 19681 OF DISODIUM OCTABORATE IN SITKA SPRUCE 115 the reaction mixture into a 100-ml graduated flask containing 50ml of acetone- water solution and swirl to mix. Wash the reaction mixture remaining in the boiling tube into the graduated flask with acetone - water solution. Dilute the solution to the mark with the acetone - water solution and mix.Measure the optical density of the red rosocyanin complex against a reagent blank, prepared in a similar way, in l-cm cells at 555 mp on a Unicam SP600 spectrophotometer. To obtain the boric acid content in the test solution compare the spectrophotometer reading with a calibration graph. CALIBRATION PREPARATION OF STANDARD SOLUTION A- Dissolve 2.3140 g of freshly recrystallised sodium borate, Na,B,O,.lOH,O, in water. Transfer the solution to a 250-ml graduated flask, dilute to the mark with water and mix. Immediately store in a polythene bottle. 1-0 ml of standard solution A = 6000 pg of boric acid. PREPARATION OF STANDARD SOLUTION B- the mark with 1.1 M sodium hydroxide solution and mix. bottle. PREPARATION OF STANDARD SOLUTION C- to the mark with M sodium hydroxide solution and mix.bottle. Transfer by pipette, 10 ml of standard solution A into a 100-ml graduated flask, dilute to Immediately store in a polythene 1-0 ml of standard solution B = 600 pg of boric acid. Transfer by pipette, 10 ml of standard solution B into a 100-ml graduated flask, dilute Immediately store in a polythene 1.0 ml of standard solution C = 60 pg of boric acid. Transfer aliquots of 0.1, 0.2, 0-3, 0.4 and 0.5ml of standard solution C to soda-glass boiling tubes and add 0.4, 0.3, 0.2, 0.1 and O-Ornl, respectively, of M sodium hydroxide solution. Transfer aliquots of 0.1 ml of standard solution B and 0.1 ml of standard solution B plus 0-2 ml of standard solution C to boiling tubes and add 0.4 and 0.2 ml, respectively, of M sodium hydroxide solution. The aliquots taken contain 6, 12, 18, 24, 30, 60 and 72 pg of boric acid in 0-5 ml of M sodium hydroxide solution. Continue as described under Pro- cedure. Plot optical densities against micrograms of boric acid to obtain the calibration graph. This paper is published by permission of the Ministry of Technology. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. REFERENCES McMullen, 31. J., Analyst, 1953, 78, 442. Wilson, W. J., Analytica Chinz. Acta, 1958, 19, 516. Cockcroft, R., Holzforschung, 1960, 14, 117. Cummins, N. H. O., “Proceedings of the 6th Annual General Meeting of the New Zealand Wood British Wood Preserving Association, Provisional Standard 105, 12. British Wood Preserving Association, Provisional Standard 105, 14. Spicer, G. S., and Strickland, J. D., J. Chem. Soc., 1952, 4644. Hayes, M. R., and Metcalfe, J., Analyst, 1962, 87, 1041. CartFight, K. St. G., and Findlay, W. P. K., “Decay of Timber and its Prevention,’’ Second Edition, H.M. Stationery Office, London, 1958, p. 286. Smith, D. N. R., and Cockcroft, R., Nature, 1961, 189, 163. Received August 15fh, 1967 -, Ibid., 1959, 21, 2. Preservers’ Association” 1965, p. 39.