Patent Application
20110253326
This invention is directed towards a method for separating lignin from liquid solutions that result from the pretreatment of lignocellulosic materials. The method involves a sequence of steps including acidification, evaporation, and precipitation or centrifugation that are performed under defined conditions, and results in a relatively pure, solid lignin product.
This invention relates to Biofuels production processes that use fermentation to convert lignocellulosic materials (e.g., grass, leaves, wood) into ethanol and other chemicals. Lignocellulosic materials usually need a pretreatment step to facilitate the chemical or enzymatic hydrolysis of cellulose into fermentable sugars. Lignocellulosic materials contain cellulose and hemicellulose bound into a rigid composite structure that is surrounded by lignin. The lignin layers must be disrupted during the pretreatment step to expose the underlying cellulose structure to chemical attack. Lignin is a highly cross-linked polymeric material found in all land-based plants (except bryophytes) that aids in water transport, adds stiffness to plant structures, and helps protect the plant material against disease and attack by pests. The amount of lignin varies according to plant structure, species, and season, but harder lignocellulosic material typically correlates with higher lignin content, and the hardest of the hardwoods, piratinera guianensis or snakewood, may contain up to 39% lignin by weight in the heartwood. Disruption of the lignocellulosic structure may be achieved by physical or chemical means.
If hydrolysis of cellulose into sugars is the only concern in a biofuels production process, then multiple pretreatment methods may be effective for any given lignocellulosic feedstock. In that case, a pretreatment method should be chosen to provide the highest conversion of cellulose into sugars at the least cost for a given feedstock. However, lignin is also a useful material, and the economics of a biofuels plant may be improved if lignin is recovered as a separate product.
Lignin has many current and potential commercial uses. It has a moderate net heat of combustion (lower heating value, or LHV) of approximately 20 MJ/kg when purified, and it may be burned to generate steam for process heating instead of lignite, coal or natural gas, which have heats of combustion of about 11, 29, and 37 MJ/kg, respectively. Lignin may be used as a starting material for the manufacture of useful higher-value chemicals (e.g., lignosulfonates, dispersants, vanillin, dimethylsulfoxide). It may be pyrolyzed to make an oil-like product or a pyrolytic solid which may be further processed using standard petroleum refining techniques to make gasoline, diesel, and jet fuel. Lignin is also being studied and used as a polymer additive for the manufacture of bio-based or “green” polymers, and as a low-cost source material for the manufacture of carbon fibers.
Of the available pretreatment methods, only a sub-set of chemical pretreatment methods—those that operate in part by separating lignin from the lignocellulosic structure rather than destroying it—can be used to generate a relatively pure lignin side stream during the pretreatment stage. Though lignin may also be recovered downstream after cellulose and hemicellulose have been hydrolyzed into sugars, the hydrolysis of cellulose rarely exceeds 90% conversion, and any lignin remaining after hydrolysis is still intermingled with cellulose and other materials.
Lignin removal and recovery by chemical means is not a new concept, and it is performed on an industrial scale in wood pulping mills that use the Sulfite Process (acid-based), from which lignosulfonates can be produced, and the Kraft Process (alkaline-based), from which a purer lignin stream may be isolated, albeit with a relatively high ash content. Because of the use of harsh chemicals and high capital equipment costs, these wood pulping processes have not yet been deployed in biofuels production.
Accordingly, there remains room for improvement and variation within the art.
It is one aspect of at least one of the present embodiments to provide a process of recovering lignin from a biomaterial comprising the steps of providing a lignin-containing cellulosic biomaterial; treating said cellulosic biomaterial with a solution of ammonium hydroxide, thereby extracting lignin from cellulosic biomaterial and providing an ammonium hydroxide lignin solution; separating the treated cellulosic biomass from the solution of ammonium hydroxide lignin solution; evaporating ammonia from the separated ammonium hydroxide lignin solution while maintaining a pH greater than 7.0; acidifying the separated ammonium hydroxide lignin solution to a pH level below 4.0, thereby forming a colloidal suspension of lignin; reducing a volume of said colloidal suspension of lignin by evaporation of water, thereby forming a coagulation of particles of lignin within said reduced volume solution; separating the lignin from the reduced volume solution; optionally washing the separated lignin with dilute sulfuric acid; and, drying said lignin.
It is a further aspect of at least one of the present embodiments of the invention to provide for a process of reducing and recovering lignin content from a bio-feedstock, thereby increasing the conversion efficiency of subsequent hydrolysis steps of the bio-feedstock into a sugar or fuel.
It is yet a further object of at least one embodiment of the present invention to provide for a process of recovering lignin from a bio-feedstock which allows for a greater than 80% hydrolysis of the treated bio-feedstock into a sugar or fuel.
It is a further aspect of at least one of the present embodiments of the present invention to provide for a treatment step of treating a lignin-containing cellulosic material which is treated with a solution of ammonium hydroxide in a volume to mass ratio of between about 1:1 to about 16:1 and at a temperature of about 80° C.
It is a further aspect of at least one embodiment of the present invention to provide for a process of recovering lignin wherein an ammonium hydroxide solution containing an extracted lignin component is heated to a temperature of about 96 to about 98° C. at ambient pressure so as to evaporate ammonia from the solution.
It is yet a further aspect of at least one embodiment of the present invention to provide for a process of separating lignin from a bio-material source following a prior step of evaporating ammonia, the heated lignin containing solution is acidified to form a colloidal suspension of lignin.
These and other features, aspects, and advantages of the present invention will become better understood with reference to the following description and appended claims.
A fully enabling disclosure of the present invention, including the best mode thereof to one of ordinary skill in the art, is set forth more particularly in the remainder of the specification, including reference to the accompanying drawings.
Reference will now be made in detail to the embodiments of the invention, one or more examples of which are set forth below. Each example is provided by way of explanation of the invention, not limitation of the invention. In fact, it will be apparent to those skilled in the art that various modifications and variations can be made in the present invention without departing from the scope or spirit of the invention. For instance, features illustrated or described as part of one embodiment can be used on another embodiment to yield a still further embodiment. Thus, it is intended that the present invention cover such modifications and variations as come within the scope of the appended claims and their equivalents. Other objects, features, and aspects of the present invention are disclosed in the following detailed description. It is to be understood by one of ordinary skill in the art that the present discussion is a description of exemplary embodiments only and is not intended as limiting the broader aspects of the present invention, which broader aspects are embodied in the exemplary constructions.
In describing the various figures herein, the same reference numbers are used throughout to describe the same material, apparatus, or process pathway. To avoid redundancy, detailed descriptions of much of the apparatus once described in relation to a figure is not repeated in the descriptions of subsequent figures, although such apparatus or process is labeled with the same reference numbers.
Biomass Source Material
Samples of switchgrass (Alamo variety) were obtained from Clemson University's Pee Dee Research and Education Center, located near Florence, S.C., USA. The switchgrass was in its second season of growth, and had been harvested in December after the switchgrass tract had gone dormant for the winter. Switchgrass bales consisting of stems and leaves were collected from this tract and chopped using a diesel-powered rotary “bale buster” to reduce the lengths of grass to between approximately 3 and 10 cm. Samples of the chopped switchgrass were milled further to a particle size of between 0.5 and 1.0 mm in length by passing the material multiple times through a Thomas Model 4 Wiley Mill.
Two samples of milled switchgrass taken from random lots of switchgrass were submitted to the Georgia Institute of Technology's Institute of Paper Science and Technology (IPST) for compositional analysis. Analytical tests were performed to identify structural carbohydrate, lignin, and ash content. The IPST used analytical methods that are prescribed by NREL and IPST protocols. The results of these analytical tests are shown in Table 1.
A weighed sample of milled switchgrass was dried in a drying oven at atmospheric pressure and 65° C. overnight and weighed again the following day in order to determine a dry weight. The dried sample weighed 10% less, which indicated that the moisture content of the sample was 10%.
Reagent Preparation
Totally, 15 vol % ammonium hydroxide solution, which is needed to perform pretreatment, was prepared by mixing 50% v/v ammonium hydroxide obtained from the Ricca Chemical Company with the appropriate amount of deionized water. A 10 wt sulfuric acid solution was prepared by mixing 96-98% sulfuric acid from Acros Organics with the appropriate amount of deionized water.
Equipment
Pretreatment operations were performed using laboratory glassware that was heated in a water bath. Boiling of fluids was performed by heating on a hot plate equipped with a magnetic stirrer. Titrations were performed using a graduated glass burette. Measurements of pH were performed using Colorphast pH-indicator strips, range 2.0-14.0, that were manufactured by EMD Chemicals, Inc. Filtrations were performed using a ceramic perforated Buchner funnel either by itself, or with filter paper circles. Two types of filter paper were used, when appropriate: Coarse Fisherbrand Filter Paper Grade P8 (particle retention >20-25 μm), and Fine Fisherbrand Filter Paper Grade G6 (particle retention >1.5 μm). Whenever high-speed centrifugation was performed for the purpose of recovering precipitated lignin, 50-mL conical bottom disposable centrifugation tubes obtained from Cole-Parmer were used. All work was performed in or exhausted to a chemical fume hood to avoid exposure to ammonia vapors.
Experiments
Fifteen trials were performed to study how effectively lignin could be recovered using the described method (see Table 2). As a necessary first step, ammonium hydroxide solutions containing lignin were generated, and this was performed by mixing 50-g batches of milled switchgrass with 15 vol % ammonium hydroxide in 500-mL glass filter flasks at a 6.6:1 or an 8:1 vol (ml):mass (g) ratio. Twelve trials (Trials 1-12) were performed using a 6.6:1 volume:mass pretreatment ratio, and three trials (Trials 13-15) were performed using an 8:1 volume:mass pretreatment ratio. Trials 8-12 were performed to study process scale-up, and used larger amounts of switchgrass than the other trials. For Trials 8-11, three 50-g batches of switchgrass were pretreated for each trial, and the pretreatment solution obtained from each batch was combined for subsequent processing steps. For Trial 12, twenty-two 50-g batches of switchgrass were pretreated, and the pretreatment solution obtained from each batch was combined for subsequent processing steps.
During pretreatment, the flasks were partially submerged in a water bath for 24 h at a bath temperature set point of 80° C. The choice to use a 50-g batch size during the pretreatment step was somewhat arbitrary, but could not be much exceeded due to the height limit of the water bath. The lid of the water bath would not close when 1-L filter flasks were used, and so 500-mL filter flasks were used instead, thus limited the pretreatment fluid volumes to 500 ml or less. As a result, multiple 50-g batches were needed to perform the scale-up trials. Although the flasks were stoppered, they were also vented through the side port of the filter flasks to the fume hood to relieve pressure in the filter flask and to capture escaping ammonia vapors. Occasionally the flasks were shaken by hand to increase mixing. At the end of 24 h, the flasks were removed from the water bath and allowed to cool, and then the material in each flask was filtered using a ceramic Buchner funnel without filter paper to separate the pretreated solids from the pretreatment fluid. In a few cases, the pretreatment solution had to be filtered again using coarse filter paper to capture small particles, and to ensure a solids-free pretreatment fluid.
The lignin recovery method set forth below was then applied to each pretreatment solution. Two variations of the method were tried. In Variation 1, the steps were followed in the order described. In Variation 2, the first step (Ammonia evaporation) and the second step (Acidification) were interchanged in order to determine whether the order of these steps made any difference in the amount of lignin recovered. In the performance of Step 4, either filtration with fine filter paper, or centrifugation followed by decantation was used to separate the coagulated lignin particles from the liquid.
Step One—Ammonia Evaporation
The ammonium hydroxide solution containing lignin is heated in an open vessel (or in a distillation column) to boiling for the purposes of evaporating ammonia. The solution is heated until the boiling point at atmospheric pressure reaches 96-98° C. At this point, nearly all of the ammonia in solution has been removed, while the solution still retains a sufficient amount of residual ammonia to keep the pH above 7, which prevents premature condensation of lignin from solution.
Step Two—Acidification
While the solution is still hot, it is titrated with sulfuric acid until the solution pH falls to below pH 4. The color of the solution changes from black to brown upon crossing the pH neutral point, and a colloidal suspension of lignin forms.
Step Three—Water Evaporation
The solution is allowed to boil and water is evaporated until the volume of the solution is reduced to approximately half of the starting amount, and a “skin” of lignin begins to form on the surface. The formation of a skin indicates that lignin particles have coagulated to form larger, more easily separable particles. In cooking terminology, this step is known as a reduction.
Step Four—Cooling, and Solid/Liquid Separation
Once the solution has cooled, the lignin particles may settle out of solution on their own, in which case the liquid may be directly decanted, or the separation process may be aided by using a vacuum filter or high-speed centrifuge.
Step Five (Optional)—Washing with Dilute Sulfuric Acid
If a very low-ash lignin product is desired (less than 1000 ppm), then the lignin may be washed several times with dilute sulfuric acid, followed by a water rinse. The washing step helps remove salts adhering to the precipitated material, and may help remove entrained ash that was carried along from the pretreatment step.
Step Six—Drying
The lignin may be dried in the open air, or placed in a drying oven. The temperature of the drying oven should not exceed 65° C. to avoid glazing or sintering of the recovered lignin particles.
A sample of milled switchgrass that had been pretreated using a 6.6:1 ratio of 15 vol % ammonium hydroxide was submitted to the IPST for analysis of structural carbohydrate, lignin, and ash content, as described in the Biomass Source Material section. A lignin sample was also drawn from a random trial and was submitted to the IPST for compositional analyses. The analytical results for these samples are shown in Table 3.
The results of the experimental work are shown in Table 2. During the performance of Step Four of the method, it was observed that the cleanest lignin separation could be achieved by using fine filter paper, rather than centrifugation. The filtrate produced was clear and amber-colored, whereas the supernatant liquid produced by centrifugation tended to be a little cloudier, though still amber-colored.
The sample analyses provided in Tables 2 and 3 may be used to calculate two numbers, the percentage of the original lignin content recovered, and the percentage of the extracted lignin recovered from the pretreatment fluid. The percentage of the original lignin content recovered is calculated in the following manner. Since the milled switchgrass was observed to have an approximate moisture content of 10 wt %, each 50-g batch of milled switchgrass has 45 g of dry biomass. Table 1, which was also measured on a dry basis, shows that the milled switchgrass contains approximately 26.6 wt % lignin. Therefore, each 50-g batch contains approximately 12 g of lignin. In Table 3, chemical analysis of a sample of recovered. solid shows that the recovered solid contains ˜84 wt % lignin. Therefore, the calculated percentage of the original lignin content recovered is determined by multiplying the mass of recovered solid by 0.84, dividing by the mass of lignin originally present in the biomass, and then multiplying by 100%. For example, if the amount of recovered solid is 5 g, the amount of lignin in the recovered solid is 5 g×0.84=4.2 g. Then, if the amount of lignin in a 50 g sample of biomass is 12 g, the amount of original lignin recovered is 4.2 g/12 g×100%=35%.
The second number, the calculated percentage of the extracted lignin recovered from the pretreatment solution, is a more precise number, and shows how well the recovery method works in removing dissolved lignin from the pretreatment solution. According to the literature, ammonium hydroxide dissolves lignin and hemicellulose but does not significantly dissolve cellulose (glucan). So the mass of cellulose in the pretreated biomass should be the same before and after pretreatment. According to Table 1, each 50-g batch of switchgrass contains approximately 45 g (dry basis)×0.32=14.3 g glucan. After pretreatment, the glucan content should be the same (˜14 g), but the total mass of pretreated biomass should be less due to removal of lignin and hemicellulose during pretreatment. Table 3 shows that the glucan content has been increased to 40 wt % as a result of pretreatment. If it is assumed that the mass of glucan in the pretreated biomass is the same before and after pretreatment, then the total mass of the biomass after pretreatment is approximately 14.3 g/0.40=35.7 g. From Table 2, it is also known that the pretreated material contains 18.1 wt % lignin (soluble and insoluble). Therefore the mass of lignin remaining in the pretreated material is 35.7 g×0.181=6.5 g. Therefore, pretreatment dissolves ˜12 g (initial)−6.5 g (final)=5.5 g of lignin per 50-g batch. From Table 2 it is also known that the solids collected using this method contain 84% lignin, and so the largest amount of solid that should be recoverable is 5.5 g/0.84=6.5 g. So, if 6.5 g of solids are recovered by this method from 50 g of switchgrass, then 100% lignin of all of the lignin dissolved during pretreatment has been recovered. If only 5 g of solids, for example, are recovered per 50 g batch instead, then the calculated percentage of lignin recovered from the pretreatment solution is (5 g×0.84)/5.5 g×100%=76%.
Statistically, there is no difference between the amount of lignin recovered when Steps 1 and 2 of the lignin separation method are interchanged. Both variations work equally well in recovering lignin from solution. However, it was readily apparent that acidifying the pretreatment solution before the ammonia is evaporated is less efficient because a greater amount of acid is needed to neutralize the ammonia in solution. Also, evaporation of ammonia takes more energy because ammonia is less volatile when it is protonated. Furthermore, the presence of sulfate ion leads to the formation of ammonium sulfate salt, which is not volatile and is retained in the waste liquid.
With regard to the calculated percentage of lignin recovered from the pretreatment solution, there is significant variability, though the method is proficient in all cases at recovering at least 66% all lignin dissolved into solution. The drop in the calculated percentage of lignin recovered for Trials 8-11 may be due to a switch from filtration to centrifugation as a means to separate precipitated lignin from solution. While filtration proved to be effective at achieving a very clean separation of lignin from solution, centrifugation was faster to perform and was more convenient, since no filter media was needed. For Trial 12; filtration was again used as a means of separating lignin particles from solution, and the percentage of lignin recovered increased accordingly. Though a clean separation may be achieved using centrifugation, the technique requires careful decantation of the supernatant liquid after centrifugation to avoid loss of lignin, an the decreased lignin recovery for Trials 8-11 is likely due to variation in the performance of the decantation step.
In Table 2, the percentage of lignin recovered from the pretreatment solution is lower for Trials 8-12, and the cause for this drop is not certain. One possibility is that centrifugation is less effective than filtration in recovering precipitated lignin from solution. Centrifugation was used in these trials instead of filtration in an effort to decrease separation time, and some precipitated solid particles may have been lost when the supernatant liquid was decanted from the centrifugation tubes. Increased solids loss during the solids recovery step would also explain the drop in the percentage of original lignin recovered for these trials.
For Trial 4 and Trials 12-15, the percentage of lignin recovered from the pretreatment solution is significantly greater than 100%, which is not possible unless ammonium hydroxide pretreatment for these particular trials was more successful at extracting lignin from switchgrass than was determined using the data in Table 3. While there isn't a clear explanation as to why this might have occurred for Trial 4, it is suspected that increasing the pretreatment volume:mass ratio in Trials 13-15 from 6.6:1 to 8:1 improved the effectiveness of the lignin extraction, thus making more lignin available in solution for recovery.
As set forth in
While the pre-treatment ratios and trials show efficacy at 6:1 to 16:1 it is believed that much lower ratios, such as 1:1, are also useful depending upon the processing conditions, the type of bio-feedstock utilized, and the degree of mechanical mixing or agitation that might be employed. For instance, one pre-treatment option may involve percolating ammonium hydroxide through a solid bed of bio-mass. In such conditions, the mixing ratio of liquid to solid may be about 1:1, yet the fluid may be saturated with dissolved lignin which can be recovered as described herein. Accordingly, the actual mixing ratio of liquid to solid can be optimized around variables of lignin content of the bio-mass, the physical size dimensions of the bio-mass, extraction processing temperatures and the amount of mixing or agitation which may be employed.
Without being limited by theory, it is believed that the increase in lignin removal when the pre-treatment ratio was increased suggests that the ammonium hydroxide pretreatment step is responsible for the poor lignin recovery from the biomaterials such as sweet sorghum, sweet gum, and loblolly pine. It is believed that the ammonium hydroxide pretreatment process may be affected by a lignin stability limit, the composition of the biomass, or perhaps mass transfer of limitations that are alleviated using a higher liquid to solid ratio for the pretreatment step.
The process described herein has been found useful for resolving between 40-60% of lignin in the feedstock materials. It has also been observed that the treatment methodology will extract less than 20% of any cellulose and less than 5% of cellulose from the pretreated biomass. While other methodologies may remove larger amounts of lignin, it has been found that the current approach provides for a cost-effective removal of a sufficient amount of lignin such that greater than 80% hydrolysis of the pretreated biomass may be converted into useful figures. As a result, the lignin removal process allows for a high percentage of hydrolysis of biomass into useful sugars while generating a easily purified lignin stream that has additional commercial value.
A break-even point was calculated for a lignin separation train for a plant of two different sizes: (1) a pilot plant processing 1 metric ton/day of switchgrass (dry basis), and (2) a production-scale plant processing 350,000 metric tons/year. This amount of switchgrass could be produced in one year by cultivating about 18,000 hectares and collecting it for processing in a regional ethanol production facility. To calculate a break-even point, an existing Aspen Plus flowsheet of a lignocellulosic ethanol production plant was modified to include a representation of the lignin separation process (see
For the larger-scale plant, the marginal utility costs remain the same per unit lignin produced, but determining the cost of equipment becomes more complicated. Once a particular piece of equipment reaches its maximum manufactured size, then multiple units must be deployed to achieve higher throughput, and equipment costs scale linearly with the number of units beyond that point. The point at which a particular piece of equipment reaches its maximum size is not known without consulting specific equipment manufacturers. As a result, only order-of-magnitude marginal costs could be calculated for the larger plant size.
For the 1-ton/day switchgrass plant, which is the size typically discussed for the performance of integrated demonstration facilities, the normalized marginal material and utility cost is $0.56/kg lignin. The individual contributions to this cost are shown in Table 5. Totally, 75% of the marginal this cost is due to operation of the evaporator, which requires the burning of natural gas to heat the evaporator, and the use of an electrically powered chiller to condense the evaporated water vapor. The cost could be reduced if the heat of evaporation were recuperated and used elsewhere in the plant, but no credit was taken for such a step in this analysis. The total estimated marginal cost of equipment for the 1-ton/day switchgrass plant is shown in Table 6. For this plant, the equipment costs are calculated to be $1.80/kg lignin.
For the larger, industrial-scale plant, the marginal utility costs remain the same, since they scale linearly with the throughput of the plant, but the marginal equipment costs are reduced. For this larger-scale plant, the total capital costs of the lignin separation train are estimated to be in the range of $25M, which corresponds to a normalized equipment cost of approximately $0.22/kg lignin.
At these normalized marginal costs, it is not economical to burn the lignin for process heat. Ignoring capital costs altogether, burning the lignin to generate process heat instead of burning natural gas is the equivalent of buying natural gas at a price of $28/GJ, which is more than five times its current market cost. Therefore it is not economical to burn the separated lignin at any scale. The better option would be to find a market for the purified lignin where it could be sold at least for its break-even price as a feed material for the manufacture of other higher-value chemicals or materials. At the 1-ton/day scale, it is not likely that creating a separate lignin stream by this process would be economical. The breakeven point is greater than $2.36/kg, which is a tolerable feedstock price only if the products made from it are not commodity chemicals. At the industrial scale, however, the break-even point is greater than $0.78/kg, and it is possible that lignin produced by this method could be used to make commodity chemicals such as phenol, for example, which currently sells for approximately $1.56-1.74/kg on the commodity markets.
According to the Aspen Plus simulation, the waste stream from the process contains 4% dissolved xylan, 3.5% ammonium sulfate, and 1.7% protein, so there is the potential to further exploit this stream to provide additional revenue for the plant from the generation of value-added coproducts.
Although preferred embodiments of the invention have been described using specific terms, devices, and methods, such description is for illustrative purposes only. The words used are words of description rather than of limitation. It is to be understood that changes and variations may be made by those of ordinary skill in the art without departing from the spirit or the scope of the present invention which is set forth in the following claims. In addition, it should be understood that aspects of the various embodiments may be interchanged, both in whole, or in part. Therefore, the spirit and scope of the appended claims should not be limited to the description of the preferred versions contained therein.
This application claims the benefit of U.S. Application Ser. No. 61/342,751, entitled “Separation of Lignin from Aqueous Ammonia”, filed on Apr. 19, 2010, and which is incorporated herein by reference.
This invention was made with Government support under Contract No. DE-AC09-08SR22470 awarded by the United States Department of Energy. The Government has certain rights in the invention.
| Number | Date | Country | |
|---|---|---|---|
| 61342751 | Apr 2010 | US |
