Four lignin samples (Alcell, Indulin AT, Curan 27-11P, and Sarkanda) were converted into liquid polyols by a chain extension reaction with propylene oxide (oxypropylation). The reaction was carried out in a batch reactor using potassium hydroxide (KOH) as the catalyst. The formulation was optimized with Alcell lignin by modifying the lignin/propylene oxide ratio [L/PO (w/v)] and catalyst content [C/(C + L) (% w/w)]. L/PO ratios of 10/90, 20/80, 30/70, and 40/60 and catalyst contents from 2% to 5% were used. The influence of the studied variables was examined by following the evolution of the hydroxyl index, viscosity, homopolymer content, and molecular weight distribution. The optimal oxypropylation conditions were established according to the requirements that a given polyol should fill when used in rigid polyurethane foam formulations, i.e., a hydroxyl index between 300 and 800 and a viscosity below 300 Pa·s. The established optimal conditions were applied to three formulations based on the other lignin samples, namely, 20/80/5, 30/70/2, and 35/65/2. The produced polyols were characterized, and the determined values for hydroxyl index and viscosity were found to be in the range of those typical of commercial polyols employed for polyurethane rigid foams.

No abstract

The hydroxyl group contents of four technical lignins [Indulin AT (Meadwestvaco), Alcell (Repap), Curan 27‐11P (Borregaard LignoTech), and Sarkanda (Granit SA)] were investigated in view of their valorization as polyols in polyurethane synthesis. The different hydroxyl group contents were determined by the following methods: titration and 1H‐NMR, 13C‐NMR, and 31P‐NMR spectroscopy. The titration method chosen was on the basis of a standard method commonly used to characterize commercial polyols for polyurethanes synthesis. The values of the total and phenolic hydroxyl contents determined by the different techniques were found to be in good agreement. For the total hydroxyl contents, coefficients of variation of 5.6% (Alcell), 3.2% (Indulin AT), 2.3% (Sarkanda), and 6.2% (Curan 27‐11P) were established. For the phenolic hydroxyl contents, a good correlation was observed between data obtained from 31P‐NMR and 13C‐NMR for all lignin samples, except for the Sarkanda lignin, for which a relatively high coefficient of variation (12.6%) was found. For softwood lignins (Indulin AT and Curan 27‐11P), the phenolic hydroxyl content determined by 1H‐NMR was always lower than that deduced from 31P‐NMR and 13C‐NMR spectroscopy. © 2008 Wiley Periodicals, Inc. J Appl Polym Sci, 2008