Applikation naturnaher Bindemittel zur Herstellung von Holzfaserdämmstoffen mittels Heißluft-/Heißdampf-Verfahren
by Kolja Ostendorf
Date of Examination:2022-01-21
Date of issue:2022-04-07
Advisor:Dr. Markus PD. Euring
Referee:Dr. Markus PD. Euring
Referee:Prof. Dr. Holger Militz
Persistent Address:
http://dx.doi.org/10.53846/goediss-9152
Files in this item
Name:Dissertation_Kolja Ostendorf_Druckversion_03...pdf
Size:3.06Mb
Format:PDF
The following license files are associated with this item:
Abstract
English
In times of the Anthropocene, a recently proclaimed epoch of world history, human activity is the main factor influencing biological, geological and atmospheric chang-es on planet Earth. Accordingly, only humans themselves can counteract these changes. Sustainable raw materials and technologies are therefore playing an in-creasingly important role. Insulation materials made from renewable raw materials such as wood fibers are a showcase product, as they can protect the climate in sev-eral ways. One problem, however, is the technology-related bonding of the wood fibers with petroleum-based isocyanate binders, e.g. polymeric diphenylmethane diisocyanate (pMDI). Therefore, this work investigated the possibility to manufacture wood fiber insulation boards (WFI) bonded with animal (blood albumin) or vegetable (canola meal) proteins. For this purpose, the novel hot air/hot steam (HA/HS) pro-cess of Euring and Kharazipour (2013) was used, which allows a broader selection of binders, e.g. proteins, due to higher feasible process temperatures. In addition, the aim was to investigate whether renewable additives such as lignin or lignin-containing canola hulls can improve the performance of pMDI and if their addition can reduce the pMDI content by 50%, thus creating a more sustainable product. In the first part of the work, a binder based on blood albumin was investigated. For this purpose, whole porcine blood powder was mixed with water and glycerol as a kind of adhesion promoter. It was shown that it is possible to produce WFI at a resin load of 15% with a panel density of 180 kg/m³ and a thickness of 40 mm, which were comparable regarding board properties (compressive and tensile strength) to those bonded with 4 % pMDI. Only the water absorption failed to meet the requirements according to DIN EN 13171 (2015), claiming that an increase in weight through ab-sorbed water should not exceed 2 kg/m². However, the water absorption could be considerably reduced by adding 2% of a hydrophobing agent. In addition, the albu-min adhesive was chemically analyzed. A comparatively high solids content (> 50 %) for protein adhesives was determined, whilst the viscosity of the albumin adhe-sive was below that of the reference binder pMDI and, thus, easy to process. The pH value was slightly alkaline. The gel time measurement showed that the higher aver-age process temperatures of 156.3 °C using the HA/HS process are crucial to com-pletely cure (denaturate) the albumin-based binder. Cosequently, the lower process temperatures (82.9 °C) generated by conventional hot-steam (HS) technique, was not sufficient to provide testable WFI over the dwell time (60s). In the second part of the work, it was investigated if it is possible to use an adhesive based on canola meal for the production of WFI with various board densities (110, 140, 160, 180 kg/m³) and two thicknesses (40, 60 mm). Canola meal is very rich in protein (> 36 %) and has so far been used mainly in animal feed and shows there-fore a great unexploited potential. The adhesive was produced by mixing canola meal together with water, urea and 1 M sodium hydroxide solution (NaOH) to form a highly viscous but still sprayable dispersion. The pH value of the binder was in the strongly alkaline range, since the NaOH used (pH 14) was indispensable to liberate the proteins from the canola meal. The gelation time was not measurable with the applied measuring method (gel-timer apparatus). The solids content was about 35 %. Industrially produced wood fibers from GUTEX company were then sprayed with 15 % of the canola meal adhesive and cured by means of HA/HS, achieving core temperatures of 156.5 °C. It was possible to produce WFI up to a density of 140 kg/m³ with similar tensile and compressive strengths as comparable boards using 4 % pMDI and cured by HS only. As for the previous work using an albumin adhesive, the water absorption could not meet the required 2 kg/m² from DIN EN 13171 (2015), although the addition of 2 % paraffin emulsion significantly reduced the water ab-sorption. An extractives content analysis of the canola meal-bonded WFI showed that the extractives content was tripled due to the extractives-rich canola meal (espe-cially residual fats) and that the pH of the strongly alkaline adhesive was buffered to a neutral pH of approximately 7 by the acidic wood fibers, which constitutes a not insignificant factor when these are installed together with other materials (e.g. met-als). In the third and final part of this work, two types of kraft lignins (Indulin AT and Bio-Piva 395) and grinded ligneous canola hulls were blended at 1 % each with 2 % and 4 % pMDI, respectively, and together with the equal amount of propylene car-bonate as green solvent. This was to show if the standard pMDI load of 4 % can be reduced to 2 % by adding the aforementioned additives in the production of WFI with a panel density of 160 kg/m³ and a thickness of 40 mm, and whether they can additionally improve the physical-mechanical board properties. The addition of pro-pylene carbonate should prevent pre-curing of the pMDI with the additives and should improve the distribution of the adhesive-mixture on the fibers according to a patent by Thompson et al. (2002). The curing of the boards was carried out using HA/HS-process as well as HS as reference method. Within a chemical material analysis of the canola hulls and the insulation boards produced, it was shown that the canola hulls still contained protein-rich cotyledon tissue as well as residual fats (12.7 %), and therefore the lignin content was only 19.2 %. The board extractives were elevated in the hot water extraction whenever propylene carbonate was in-volved in the binder-system, indicating on a release of this solvent. However, this could not be observed in the evaluation of the cold water extractives and therefore the dissolution seems to be temperature-dependent. Based on the nitrogen content, which was determined by elemental analysis, it was found that the addition of pro-pylene carbonate improved the distribution of the pMDI-mixture on the fiber surface as presumed beforehand. This circumstance therefore also had a positive effect on the board properties, irrespective of the curing method. Principally, satisfactory phys-ical-mechanical properties could be measured when only 2 % of pMDI was used. The addition of, in particular, the Kraft lignin BioPiva 395, however, improved the strength properties, especially the tensile strength, which suggests an interaction or crosslinking of the pMDI with the lignin. In contrast, the addition of the canola hulls mostly had a negative influence on the strength properties as well as on the short-term water absorption. This could be explained by the low lignin content or the fact that the lignin is not present in an isoltated state, as for the kraft lignins, but might be still incorporated in the cell wall structure and thus could not be thoroughly available for the reaction. The interaction of pMDI with additive was more pronounced the higher the pMDI content (4 %). When comparing both curing-processes, a strongly reduced water absorption (< 1 kg/m³) could be measured via HA/HS throughout all variants, indicating on an improved curing reaction of the pMDI, which was also highlighted in previous works.
Keywords: wood fiber insulation boards; hot-air/hot-steam-process; natural binders; protein; canola meal adhesive; albumin adhesive; pMDI; Lignin; ligneous canola hulls
Schlagwörter: Holzfaserdämmstoffe; Heißluft-/Heißdampfverfahren; Natürliche Bindemittel; Rapsmehl; Blutalbumin; pMDI; Lignin; Ligninhaltige Rapssamenschalen; Proteine