Comparative evaluation of boron distribution from ulexite,colemanite and DOT rods in Scots pine wood

Saip Nami Kartal · Evren Terzi ·Aysel Kanturk Figen · Melek Cordan · Seçil Aydin ·Sabriye Pişkin

Abstract This study evaluated boron diffusion from rods made of raw boron minerals, ulexite and colemanite with low water solubility, in comparison with di-sodium octaborate tetrahydrate (DOT). Scots pine (Pinus sylvestris L.)sapwood and heartwood blocks were conditioned to target moisture contents of 30%, 60%, and 90%. The rods were inserted into the blocks through treatment holes and boron diffusion was observed at three assay zones across the blocks after 7, 30, 60 and 90-day-incubation at room temperature.Ethylene glycol was also inserted into the holes to improve boron diffusion. Boron levels increased with increased wood moisture content. With some exceptions, boron in the assay zones did not tend to follow consistent amount gradients with distance from the treatment hole. Boron levels from ulexite rods were higher than those from colemanite rods,with DOT rods with the highest diffusion rates as a result of higher water solubility of DOT than ulexite and colemanite.The results suggest that ulexite-based rods may be useful in the presence of ethylene glycol in sapwood when wood is at high moisture content for extended periods.

Keywords Boron rods · Remedial · Ulexite · Colemanite ·Distribution · DOT

Introduction

Boron compounds are widely used for wood protection and considered as one of the best candidates for the future range of wood preservatives because they have low toxicity to the environment, to humans, and to other non-target organisms.Due to its high diffusion in moist conditions, boron can move to areas in wood not reached by other preservatives or that need additional protection as wood ages in service. Different forms of boron compounds are available for remedial or in-place treatments, including powders, gels, thickened glycol solutions, solid rods, and pastes. Borates in powdered form are generally di-sodium octaborate tetrahydrate or DOT (Na2B8O 13 ·4H2O-di-sodium octaboare tetrahydrate),mixed with water for a 15% concentration. Such DOT solutions are applied by spraying or brushing while holes for internal treatments of wood structures may also be filled by powdered DOT (Lebow and Anthony 2012). However, the most used boron formulation for remedial treatment is boron rods where boric acid is released in the presence of water in wood (Freitag et al. 2011).

Boron compounds are easily absorbed and penetrate into the wood but might show diverse distribution and penetration characteristics due their physical and chemical structures, water solubility, wood moisture content, diffusion direction in wood, presence of refractory heartwood, and temperature (Bhatia 2002; Kartal 2009; Lebow and Anthony 2012; Kartal and Yoshimura 2016).

Using different boron compounds at different degrees of water solubility in boron-based protection systems might result in a varying distribution of boron and long-term protection. Some of the borates are available as raw minerals such as ulexite, colemanite, borax, or tincalconite that have different water solubilities (Birsoy and Özbaş 2012; Özdemir et al. 2014; US Geological Survey Minerals Yearbook 2013;Terzi et al. 2017). On the other hand, the comparison of the properties of these boron minerals to those of commercial compounds such as DOT or boric acid could be useful and the use of raw, unpurified boron minerals could help minimize the cost of boron-based protection systems.Boron minerals such as ulexite and colemanite are available in large quantities in Turkey, South America, and the USA. Colemanite (Ca2B6O 11 ·5H2O-di-calcium hexaborate pentahydrate), is a calcium-including borate mineral having low solubility in water, while ulexite (NaCaB5O 9 ·8H2O-sodium-calcium pentaborate octahydrate), has a solubility of 7.6 g L −1 at 25 °C (Terzi et al. 2017). There is considerable boron distribution data since boron compounds, particularly in solid rod form, have been extensively used in remedial treatments (Dietz and Schmidt 1987; Morrell et al. 1990;Morrell and Freitag 1995; Ra et al. 2001, 2002; Freitag and Morrell 2002, 2005; Cabrera and Morrell 2009; Lebow et al.2010, 2013; Freitag et al. 2011; Koumbi-Mounanga et al.2015). However, there has been relatively little research on boron rods from ulexite and colemanite, and on boron distribution from such rods in wood.

In this study, rods manufactured from ulexite and colemanite without further purification were placed into the holes in the wood blocks, and the diffusion of boron into and from the blocks was determined. The study relied on the production of boron rods by non-purification of boron minerals in order to reduce the costs of production. In our previous studies, we evaluated the potential of such boron minerals in the manufacturing of particleboard and wood plastic composites in powdered form by comparing their biological performance and fire resistance to commercial zinc borate and boric acid/borax mixtures (Terzi et al. 2017,2018). This study however, focused on the potential use of boron rods of raw ulexite and colemanite minerals in remedial treatments. Boron distribution was determined in both heartwood and sapwood blocks of varying initial moisture contents and conditioning times compared with DOT rods.Our objective was to evaluate the potential for moisture absorption by newly developed boron mineral rods to affect diffusion following application.

Materials and methods

Scots pine (Pinus sylvestris L.) sapwood and heartwood blocks (50 × 100 × 150 mm) were oven-dried at 103 ± 2 °C for 24 h, weighed and pressure treated with distilled water.The blocks were weighed prior to air-drying to 30%, 60%,or 90% moisture content (MC). Once each block achieved its target MC, it was dipped in molten paraffin to retard further moisture loss, wrapped with stretch film and stored at 4 °C for four weeks to allow for further equilibration. After conditioning, 9-mm diameter, 35-mm holes were drilled on the narrow face of each block (Fig. 1).

Raw ulexite and colemanite boron minerals and DOT(ETIDOT-67) were purchased from Eti Maden, Ankara, Turkey. The minerals were milled and passed through a 60-mesh sieve while the DOT (di-sodium octaborate tetrahydrate)was used as purchased. Ethylene glycol (HOCH2CH2OH)was purchased from Merck KGaA, Darmstadt, Germany.

The boron rods were manufactured with ulexite, or colemanite or DOT. All 9-mm diameter rods without further purification were manufactured at Remsan Refractory Material Industries Inc., Istanbul, Turkey. As a patent application for manufacturing rods with boron minerals is has been filed,no further details are available. To compare these results with those of our previous study with boron minerals in powder form, the boron rods were prepared by cutting from raw boron bars to obtain average specific weights (ulexite rods: 1.50 g; colemanite rods: 1.27 g; DOT rods: 0.96 g).These specific weights also provide the same amount of elemental boron at a boric acid equivalent (BAE) basis in the rods for comparison with the released boron element from the three boron compounds. For ease of comparison, boron compounds are often compared based on the BAE, which is the amount of boric acid that could be formed from the subject compound. BAE is a standard unit of comparison of efficacy among boron compounds. Considering the chemical formulas of the three boron compounds tested, the BAE of the respective compounds were:

Fig. 1 Test block showing treatment hole and pattern of cutting for zones

The rods were placed in each hole; for the ethylene glycol treatments, 1-mL was introduced into the holes before rod placement. Ethylene glycol was used to improve boron movement from the rods as it allows higher surface loading and increases mineral solubility. The treatment holes were plugged with a plastic cap and the blocks wrapped individually with a stretch film to prevent moisture loss and incubated separately for 7, 30, 60, or 90 days, based on their initial MCs.

At each time period, two blocks conditioned to a given MC were removed, weighed, oven-dried and re-weighed to determine accurate MC. Three sections were sawn adjacent to the treatment hole (0-10 mm; assay zone 1) as well as at 10-20 mm (assay zone 2) and 10 mm from the end of the block (assay zone 3) (Fig. 1).

The wood from the sections was ground, passed through a 20-mesh screen, and the sawdust subjected to a wet ashing procedure based on the AWPA A7-12 standard test method(AWPA 2012) using 70% nitric acid (14 mL, 25.8 molar).A 0.5 g of sawdust was placed into a microwave digestion tube. The vessels with sawdust and acid were placed into a microwave oven (Anton Paar Microwave Digestion System-Multiwave GO, Graz, Austria). Following digestion, the vessels were quantitatively brought up to 50 mL with ultra-pure water.

The digestion extracts were analyzed for boron by an ICP-AES Multitype ICP Emission Spectrometer (Shimadzu ICPE-9000, Tokyo, Japan).

Results and discussion

Block moisture content

Average block moisture content during 7, 30, 60 and 90 days incubation were slightly lower than those of the target moisture levels of 30% and 60% (Table 1). Overall moisture contents dropped gradually over the 90-day period for blocks at all three target moisture levels. This suggests that the treatment holes and wax coating might have resulted in moisture loss even though the blocks were placed in desiccators. At the end of the 90-day incubation period, moisture levels for the 60% and 90% target points were well above the fiber saturation point (FPS); however, for the 30% target moisture level,moisture contents varied between 24 and 34% during the whole incubation period. These results show that free water was available to allow boron to diffuse through the wood.

Boron diffusion in sapwood and heartwood

For the purposes of assessing chemical distribution, the threshold value that was presumed to be effective against internal decay was 0.10% BAE (Freitag and Morrell 2005).Boron levels generally increased with increasing initial moisture content and incubation time from 7 to 90 days(Figs. 2, 3, 4, 5, 6, 7).

Generally boron levels in sapwood blocks with ulexite were higher than those in heartwood blocks. In ulexitetreated sapwood blocks without ethylene glycol, boron levels were lower than the minimum threshold for protection against internal decay (0.10% BAE) (Fahlstrom 1964;Williams and Amburgey 1987; Freitag and Morrell 2005;Cabrera and Morrell 2009) at 7- and 30-day incubation.

As the incubation period extended to 60 and 90 days,boron levels increased at particularly at 90% target moisture contents and exceeded 0.10% BAE threshold value. In heartwood blocks with ulexite rods; however, the blocks with 90%target moisture content only had boron levels exceeding the 0.10% BAE threshold value after incubation for 90 days. At the 90% target MC level, 7- and 30-day incubated sapwood blocks had boron levels exceeding the 0.10% BAE threshold (the minimum threshold for protection against internal decay) in assay zones 1 and 3 with few exceptions. When ethylene glycol was incorporated into the holes along with ulexite, boron levels in both sapwood and heartwood blocks increased compared to ulexite-only blocks. However, the boron contents exceeding the 0.10% BAE threshold were generally found after 60- and 90-day incubation in both sapwood and heartwood blocks. This suggests that ethylene glycol had a role dissolving boron from the rods and a considerable effect in boron distribution as well.

Compared to ulexite rods, colemanite treatments resulted in lower boron levels in both sapwood and heartwood blocks with and without ethylene glycol. Colemanite chemical structure and water-solubility most certainly played an important role in weak distribution compared to ulexitebased rods. The third assay zone in sapwood blocks without ethylene glycol at 30% moisture content for 7-day-incubation showed boron levels higher than the 0.10% BAE threshold value. In ethylene glycol-containing sapwood blocks, the first assay zone at the 30% target moisture content for the 60-day-incubation period and the third assay zone at the 90%target moisture content for the 90-day-incubation period exceeded the 0.10% BAE threshold value.

C.6 (5.2).2 (0.9).2 (2.0).5 (5.5).7 (0.4).8 (2.2).4 (1.2)90% M 86.4 (2.2)76 76 75 84 74 78 76 85.0 (3.1)81.3 (3.4)78.6 (3.4)77.0 (0.8)85.2 (2.2)81.4 (5.0)78.3 (0.1)73.0 (4.2)C 51 55 59 63 61 54 60% M 58.4 (2.8)56.2 (1.4).8 (5.4).8 (0.2).9 (2.2).2 (2.0).0 (2.4).3 (0.8).9 (0.1)57 55.2 (1.2)56.7 (5.0)56.0 (3.1)56.3 (3.0)53.0 (0.4)55.8 (0.1)57.2 (3.1)C OD T R 30% M(2.0(2.6.6 (0.3).7 (5.3)(6.0.8 (4.2)DO 28 26 27 25 33 30 29.0 (2.0).0 (1.4).9 (3.5).7 (2.1)T rod + EG 31.0 (0.0)DO 24.9 (2.0)31.0 (5.7)23.5 (2.1)25.8 (1.1)33.8 (0.9))))25.8 26.0 26.1 C 90% M.4 (8.0).6 (5.1).6 (0.4).4 (5.2).2 (0.3).0 (2.4).9 (0.3).0 (2.0)(1.1 79.6 90.7 (1.9)82.8 (0.2)76.4 (1.1)65.3 (5.2)89.4 (0.9)76.3 (3.1)))(2.6 89 74 69 70 86 77 78 77 76.4 C.4 (4.1).9 (3.0).6 (2.9).6 (5.0).4 (4.6).3 (8.0).8 (3.0)(3.0 60% M 57.4 (2.2)55 47 49 63 61 55 53 60.6 (2.2)53.9 (4.0)47.4 (2.5)40.0 (0.3)57.6 (2.2)55.6 (4.1)56.9 (3.8))51.8 C)% M lemanite rod (3.6(1.3)(3.9).4 (0.5).3 (5.4)29 27 21 20 28 27 30.4 (0.1).6 (3.3)Co.4 (1.2)26.2 26.5 30.7 (3.8).3 (2.5)30.5 (1.0)27.8 (1.0)24.7 (0.4)27.4 26 Colemanite rod + EG.9 (0.2)27.0 (4.0)27.0 (5.3)C))% M(5.0(2.0 72.4 77.7 90(5.6)(2.2)72.9.7 (2.0)79.2 (5.5)88.6 (4.0)67.4 77.2 (8.2)81.4 (6.0)81 77 87 81 88 79 66.3 (4.1).6 (5.5).6 (1.1).2 (5.1).4 (1.2).2 (1.4)71.3 (3.7)ation(1.3 C)))% M 66.1(7.6 53.9 60.7(2.1(3.5)(2.1)(3.3).3 (2.5)41 42 47 61 59 51 52.5 (2.8)51.4.6 (3.7)47.5 62.4.4 (2.0)72.9 (6.1).0 (3.2).7 (5.0)uring incub 60 66.3 (4.8)49.3 (2.2).7 (2.7)tent of test specimens d C % M(1.8 Ulexite rod))))27.6(5.0(6.3 26.8 20.9 26.9(7.2(3.8)(3.6)31.9(3.9)Ulexite rod + EG 30 34.6 38.0 41.0 (3.1)31.5 (7.0)25.9 (3.0).7 (3.0)21 26 35 30 25 28.9 (7.0).4 (4.1).8 (1.4).7 (5.1).4 (3.1)ard deviations; n: 2 ethylene glycol Moisture con cubation In 7 30 time (day)60 90 7 30 60 90 7 30 60 90 7 30 60 90 re content, EG ood d ood d Table 1 oo oo Sapw Heartw Sapw Heartw es in parentheses are stand Valu moistuMC

Fig. 2 Boron levels (% BAE) 1, 2 and 3 mm from the treatment zone in sapwood and heartwood blocks filled with ulexite rod, conditioned to 30%, 60%, and 90% MC and incubated for 7, 30, 60 and 90 days at room temperature

Rods prepared from DOT resulted in a remarkably higher boron content and better diffusion in both sapwood and heartwood blocks when compared to rods from either ulexite or colemanite minerals. The solubility of DOT at 20 °C is 97 g L−1while ulexite and colemanite have solubility values of 7.6 and 0.8 g L −1 at 25 °C, respectively.Almost all assay zones in sapwood blocks treated with DOT had boron levels exceeding the 0.10% BAE threshold level. There were a few exceptions with boron levels lower than 0.10% BAE for the 7-day-incubation period.Heartwood blocks contained slightly less boron compared to sapwood blocks. Greater differences were found for sapwood and heartwood blocks for the 7-day-incubation period. However, these differences disappeared when the incubation period was extended. Somewhat unexpectedly, the addition of ethylene glycol to DOT did not have any considerable effect when compared to the ulexite and colemanite treatments. Uneven moisture distribution in the wood specimens, anatomical characteristics, and the anisotropic nature (different anatomical properties in radial, tangential and axial directions) of wood might be reasons for the unexpected low boron distribution from highly soluble DOT-rods. As the time periods extended up to 90 days,more uniform boron distribution resulted. In addition, the distance from the treatment hole is an important factor for boron movement. For instance, in the specimens with DOT rods without ethylene glycol, the boron content was lower in assay zone 3 (the furthest assay zone), compared to the other zones.

Fig. 3 Boron levels (% BAE) 1, 2 and 3 mm from the treatment zone in sapwood and heartwood blocks filled with ulexite rod + EG, conditioned to 30%, 60%, and 90% MC and incubated for 7, 30, 60 and 90 days at room temperature

In this study, generally but with some exceptions, boron levels did not track consistent gradients with distance from the treatment hole. Boron levels increased at the 90% target moisture content level compared to the 30% and 60%target levels. Because free water is essential for boron diffusion, this suggests that adequate moisture was present in the blocks to allow diffusion to occur at some point in the exposure period at the lowest moisture levels tested.

Fig. 4 Boron levels (% BAE) 1, 2 and 3 mm from the treatment zone in sapwood and heartwood blocks filled with colemanite rod, conditioned to 30%, 60%, and 90% MC and incubated for 7, 30, 60 and 90 days at room temperature

Cabrera and Morrell (2009) studied the effect of moisture content on boron diffusion from fused borate rods in Douglas- fir heartwood blocks. The moisture levels were generally lower than the targeted levels. They also stressed that sealed treatment holes and the wax coatings were the main reasons for moisture loss during incubation periods up to 180 days as suggested in our study. They also showed that boron contents increased with increasing initial moisture of the blocks as well as with incubation time. In the Cabrera and Morrell (2009) study, boron levels tended to follow consistent concentration gradients with distance away from the treatment hole and to be steadily higher at the 60% and 90%target moisture contents. A study by Freitag and Morrell(2002) on the effects of ethylene glycol on boron distribution from fused rods showed that glycol was effective in increasing boron movement; however, its effectiveness was limited to zones near the treatment hole at a moisture content of 15%. When 60% moisture level was reached, glycol effects decreased. In their study, the glycol effect was related to the liquid rather than to a chemical effect because borate plus water solutions resulted in increased boron diffusion.Lebow et al. (2010) indicated that boron diffusion occurred in wood below the fiber saturation point (FSP) after test results showed boron diffusion at 20% moisture content. In another study, Lebow et al. (2013) observed boron diffusion in wood specimens at moisture levels well below the FSP,suggesting that diffusion was not solely attributed to water content. Even small differences in moisture content might result in noteworthy differences in the extent of boron penetration below the FSP. Lebow et al. (2013) also noted that glycol was ineffective in boron diffusion, however, it might be effective for more concentrated solutions. The effect of moisture content of Douglas-f ir heartwood on longitudinal diffusion of boron from fused rods was studied by Morrell et al. (1990). They showed that boron diffusion was effective when moisture content exceeded 40%. Their results also indicated that there was no difference in diffusion rates between 80 and 100% moisture content, while diffusion in the heartwood in a longitudinal direction at 20% moisture content was the least. Considering fibre orientation, Melo et al. (1992) showed that the rate of diffusion of boron salts in water-saturated Eucalyptus heartwood was independent of fibre orientation. They considered that capillary flow was possibly a dominant mechanism for the transfer of a solute into wood well below FSP. However, at moisture levels higher than FSP, molecular diffusion may be a principal mechanism.

Fig. 5 Boron levels (% BAE) 1, 2 and 3 mm from the treatment zone in sapwood and heartwood blocks filled with colemanite rod + EG,conditioned to 30%, 60%, and 90% MC and incubated for 7, 30, 60 and 90 days at room temperature

Conclusions

Fig. 6 Boron levels (% BAE) 1, 2 and 3 mm from the treatment zone in sapwood and heartwood blocks filled with DOT rod, conditioned to 30%, 60%, and 90% MC and incubated for 7, 30, 60 and 90 days at room temperature

Boron levels in wood blocks from ulexite- and colemanite-based rods without ethylene glycol were considerably less than in blocks with DOT rods since mineral solubility was lower. The addition of ethylene glycol with the rods resulted in slightly higher boron levels in both sapwood and heartwood blocks compared to the rods-only treatments. Ulexite rods resulted in enhanced boron diffusion compared to colemanite rods because of better solubility.Generally, the highest boron diffusion levels were with the DOT-rods. The results show that ulexite rods with ethylene glycol may be used for remedial treatments as a cost effective alternative to DOT rods, considering boron levels above the 0.10% BAE threshold value for protection against internal decay in sapwood. For practical applications, the amount of boron (% BAE) in wood and the threshold value of 0.10% BAE, we recommend at least three times more ulexite-based rods over extended periods,considering both heartwood and sapwood and dry conditions in wood. The reservoir effect of boron rods provided acceptable protection for longer periods. Further studies are required to determine boron diffusion from boron mineral rods in the presence of other preservative compounds such as copper and fluoride salts. Evaluation of the times for boron movement will be useful to determine the effectiveness of boron mineral rods to protect wood against fungal decay.

Fig. 7 Boron levels (% BAE) 1, 2 and 3 mm from the treatment zone in sapwood and heartwood blocks filled with DOT rod + EG, conditioned to 30%, 60%, and 90% MC and incubated for 7, 30, 60 and 90 days at room temperature

AcknowledgementsThe authors acknowledge Karaoğluları Forest Products, Construction, Automotive Industry and Trade Incorporated Company, Tuzla, Istanbul, Turkey for Scots pine logs. The boron compounds used were supplied by Eti Maden, Ankarta (Ankara), Turkey.The boron rods were manufactured at Remsan Refractory Material Industries Inc., Istanbul, Turkey. This paper was partly presented at an International Symposium on Boron-BORON 2019 in Nevşehir,Turkey April 17-19, 2019. The work was supported by TUBITAK(The Scientific and Technological Research Council of Turkey) under 1005-National New Ideas and Products R&D Funding Program (Project No: 1160149).