**1. Introduction**

The World Commission on Environment and Development [1] defined sustainable development as development that meets the needs of the present without compromising the ability of future generations to meet their own needs. With the growing awareness on the need to mitigate greenhouse gas (GHG) emissions and the inevitable depletion of fossil fuel, the world is on the journey of transitioning towards more sustainable and renewable alternatives. The need to minimise fossil fuel use and mitigate its associated GHG emissions drives the ongoing growth in sustainable and renewable alternative energy.

In the world's consumption of fossil fuel (coal, natural gas, and oil), 91% is used for energy applications. In crude oil consumption, 63% is for the global transportation sector and 16% is used to make building-block chemicals and polymers [2]. With transportation demand increasing globally, driven in part by population growth, the challenge to decrease the world's reliance on fossil fuels requires the implementation of cost-effective, large-scale, renewable energy-based transport fuel projects.

Biorefineries are the most promising route to produce biofuel and platform chemicals to support a new bio-based industry [3]. A biorefinery is an industrial facility (or network of facilities) that covers a collection of technologies to sustainably convert biomass into basic building blocks for the production of biofuels, energy, and chemicals [4]. It is analogous to current petroleum refineries. To be a feasible alternative, biorefineries must have a dependable supply of feedstock [5], which usually makes up 40 to 60% of the operating costs [6], and maximise the energy conservation between energy inputs and outputs.

Kraft pulping is a well-established process that can be converted into large-scale biorefineries, producing biofuels as a main product. Kraft mills contain critical components vital for a biorefinery [7], i.e., access to biomass feedstock and supply chains, understanding of biomass refining-type processes, and accessible residual feedstock, such as black liquor. Traditional processing in a chemical pulp mill, like a kraft mill, extracts between 40 to 60% of high-value pulp or paper products from the harvested logs while the remaining dissolved wood in the form of a liquor has relatively low economic potential. To maintain profitability, the kraft pulp sector is facing pressures to expand the range of products produced to more than just pulp, heat, and power production. A pulp mill processes a high volume of biomass feedstock and generates by-product streams like black liquor (a mixture of spent pulping chemicals and lignin), which is partially processed by the pulp production [7]. Biomass components—containing mostly hemicellulose and lignin in black liquor—supplies the energetic demand (heat and power) required by the kraft mill through combustion in a recovery boiler. The recovery boiler also plays a vital role in the inorganic chemical recovery process, which contributes to the overall economy of the kraft process. The organic component in black liquor has the potential to be transformed into bioproducts that have higher value than using black liquor as a fuel for heat and power.

Hydrothermal liquefaction (HTL) is a thermochemical process that depolymerises wet biomass into liquid fuels in a reactor operating at high temperature and pressure and sufficient time to decompose the solid natural polymeric structure to mostly liquid compounds [8]. It is a flexible conversion process due to the variability of bio-based or waste feedstock that have been successfully tested. The key advantage of why the HTL process is successful is because the feedstock of the HTL process does not have to undergo a drying process. Water in the HTL process serves as a reactant and catalyst in the subcritical region as the properties of the water change extremely. In the subcritical region, the dielectric constant of the water decreases significantly, as compared to ambient water [9]. Due to this, the solubility of hydrophobic compounds is higher than at ambient condition. In addition, the subcritical environment of the water increases the rate of acid/base-catalysed reactions due to the higher ionic product of water [10]. Liquid bio-crude is the key product of the process. With upgrading the process (like conventional fuel), this bio-crude can be transformed to the whole distillate range of petroleum-derived equivalent fuel products. When compared to gasification, pyrolysis and HTL have a simpler technical conversion of biomass to a liquid fuel [11]. However, when compared to pyrolysis oils, the lower oxygen content in HTL bio-crude makes it less corrosive and has higher heating value (30–36 MJ/kg). The higher calorific value of HTL bio-crude as compared to pyrolysis (15–22 MJ/kg) is more similar to conventional petroleum (43–46 MJ/kg) [12].

Numerous studies have been focusing on evaluating the technical and economic feasibility of the HTL process of different feedstock and algae, considering the various operating conditions of the reactors. However, techno-economic evaluation of HTL of black liquor still presents a research gap when compared to the other feedstock. Black liquor is a complex organic and inorganic mixture. The organic mixture in black liquor is mostly the remaining cellulosic fibres, lignin and hemicellulose, and the caustic inorganics that are used in the kraft process. The advantages of using black liquor in the HTL process are [13] (1) the organic component serves as feedstock to the process and (2) the inorganic component acts as a caustic catalytic solution, instead of sodium hydroxide, for the HTL process. Huet et al. [14] studied the integration of the HTL process of sulphur-free black liquor with a kraft mill, with the reactor temperature between 270 ◦C and 310 ◦C. HTL processing of black liquor produces both phenolic molecules and bio-crude. A sodium recovery of 97% was reported, which matches with kraft mill inorganics recovery technology. The sodium is recovered in the form of sodium carbonate, which is converted to caustic soda with the current available technology. Kosinkova et al. [15] conducted a study of hydrothermal liquefaction of bagasse, using co-solvents, ethanol, and black liquor. The yield

of the HTL bio-crude increases as the black liquor content increases. This is because black liquor in the co-solvent contains organic residues, which provide additional reactants for conversion and the basicity supports the base-catalysed condensation reaction that leads to oil formation.

Ong et al. [16] analysed the techno-economic feasibility of HTL of radiata pine with black liquor from an existing kraft mill. The estimated minimum fuel selling price (MFSP) of this approach was 1.75 NZD/LGE of fuel (using a conversation rate of 1 NZD = 0.6 USD). Funkenbusch et al. [17] conducted a techno-economic analysis of HTL of lignin, where the lignin are extracted from BL, and determined a MFSP of 1.58 NZD/LGE. Melin et al. [18] experimentally investigated the optimum operating parameters for producing high-quality bio-oil from HTL of BL, using glycerol as the hydrogen donor and sodium hydroxide as the alkali for high heating value fuel. Lappalainen et al. [19] studied the effects of process conditions of the HTL process on the quality of bio-oil from HTL of black liquor. The process parameters studied are residence time and any additives (solvents, catalysts); these conditions are modified to optimise the quantity and quality of the bio-oil production and to minimise the production of secondary products such as biochar and gaseous products.

Prior to today, much of the research on the HTL process studies the process parameters of lab-scale batch reactors. There is, however, a shift towards commercialisation by scaling up the process to a continuous pilot-scale operation [20]. Due to the sub-critical condition required by the process, maximizing the energy efficiency of the HTL process is crucial. Okoro et al. [21] used pinch analysis to conduct heat integration for HTL of meat waste and successfully reduce the heating and cooling demands by approximately 36% and 32%, respectively. Shemfe et al. [22] applied pinch analysis to design the heat exchanger network for upgrading of bio-crude. Anastasaki et al. [20] studied the effectiveness of enhancing a custom-designed heat exchanger design through oscillating of the slurry. Magdeldin et al. [23] conducted a techno-economic assessment of a HTL process that is integrated with downstream combined heat and power (CHP) generation using waste heat and by-products of the process. The introduction of CHP in the HTL process increases the thermal efficiency of the process. Knorr et al. [24] recognized that the maximising the heat integration of the reactor design is a crucial gap. Ong et al. [25] used an iterative process integration and simulation methodology to improve the energy efficiency of integrating HTL with an existing kraft mill.

The aim of this paper is to carry out a techno-economic and carbon emissions assessment of hydrothermal liquefaction of radiata pine and black liquor to assess the techno-economic viability of the process, as compared to fossil fuel feedstock. This paper shows the benefits of integration of the HTL process with a kraft mill with a centralised utility system. The results include a thermo-economic assessment of two other options for reducing marginal fuel use in an existing kraft mill. These options are (1) black liquor evaporators with vapour recompression and (2) replacing the current aged recovery boiler with a high efficiency modern design. The study also explores the trade-off between the GHG emission cost and oil price increase on levelised profit.

#### **2. Materials and Methods**

The current study considers three scenarios and determines the cost-benefit of integrating the kraft mill with the new biorefinery technology by measuring how much the minimum fuel selling price changes for the different scenarios.

#### *2.1. Total Site Heat Integration (TSHI)*

Heat integration reduces heat demand on boilers and the consumption of fuel, e.g., natural gas, residual biomass, wood chips, and/or black liquor. The introduction of a new biorefinery process to an existing kraft mill significantly affects the site's best heat integration design as well as its overall heat and power balance. The methods used throughout all two scenarios are TSHI with the kraft mill and site heat and power utility modelling.

A site utility model of the recovery boiler and supplementary boilers, turbine, and process heat demands has been implemented in an ExcelTM spreadsheet. The boilers assume a constant thermal efficiency of 75%. The turbine model incorporates the extended Willan's line approach of Medina-Flores et al. [26], where historical turbine performance data was used to define turbine model coefficients. Process heat demand from the utility system varies for each of the evaporation system options. The decrease in the low-pressure steam demand adversely impacts on turbine power generation. However, it reduces the required high-pressure steam from the marginal fuel boiler, which is fuelled partially by wood residue (50%) and natural gas (50%). The data is based on an existing kraft mill in the Central North Island of New Zealand.

#### *2.2. Economic Assessment*

The economic assessment is based on the methodology reported by Ong et al. [16]. The data used to estimate the operating costs are as presented in Table 1.


**Table 1.** Estimated material, utility, and carbon emission prices.

#### *2.3. Environmental Impact*

A shortcut life cycle analysis of the GHG calculation uses the method in Martinez Hernandez and Ng [27]. The GHG emission is calculated for the HTL process and its downstream benefits using the GHG factors presented in Table 2.


**Table 2.** Environmental impact coefficients.

The main GHG emissions considered in this study are (1) electricity, (2) natural gas, and (3) wastewater. The Emission Trading Scheme (ETS) in New Zealand excludes emissions from biofuels and covers the conventional liquid fossil fuel emissions [31]. The main products converted from the upgrading of the bio-crude process are (1) gasoline equivalent, (2) diesel equivalent, and (3) heavy fuel oil equivalent. The flue gas emission only considers the emissions from the combustion of natural gas. The combustion of off-gas is considered emission free because the source is from the biomass and this falls outside the ETS. The flue gas emission data is extracted from the HTL process simulation model.

#### *2.4. Process Flowsheet*

The flowsheet for this paper is based on a PhD work that has been carried out [32]. The thesis started out with designing a kraft mill-integrated biorefinery system. The case study applies a process synthesis technique by developing a multi-dimensional, heat and mass integration methodology [33], which combines pinch analysis, total site heat integration [34], and P-graph [35] frameworks, to select a biorefinery option. The novel method takes into account heat and mass integration with an existing kraft mill in Central North Island of New Zealand and also first-order capital costs of the options at the selection process. The result from the paper shows that hydrothermal liquefaction was the optimal choice to be integrated with the existing kraft mill.

Figure 1 shows the final design summarised in the thesis. The biomass slurry consists of radiata pine, black liquor, and water. The biomass slurry is pressurised in multi-stage pumps and reacts to produce bio-crude, aqueous phase with a fraction of organic materials (hydrocarbons, alcohol), non-condensable gaseous products, and organic solid residues (biochar). Black liquor serves as an additional organic feedstock (hemicellulose and lignin) and a catalyst (Na and K) to the process. Kraft pulp residuals, combined with virgin wood chips and sawdust, can be considered to be the primary biomass feedstock, tapping into existing processes, supply chains, and infrastructure. The upgrading process, e.g., hydro-deoxygenation of bio-crude [36], reduces the oxygen content of the bio-crude, producing hydrocarbon fuels equivalent to petroleum products (45 MJ/kg). The process model based on Aspen PlusTM for the HTL process and upgrading of bio-crude is described in Ong et al. [16]. The ultimate analysis of the HTL bio-crude is presented in Table 3. However, the proposed HTL process in Ong et al. [16] does not address the black liquor inorganics recovery. Figure 1 is a process flow diagram of a new proposed HTL process.

**Table 3.** Ultimate analysis of HTL bio-crude, adapted from Rowlands et al. [13].


The inorganic chemicals and sodium and sulphur (Na/S) balance in a kraft mill are crucial for the process economics and environmental viability. The key to that is recycling the Na/S balance at a rate of approximately 97% [37]. Black liquor offers a chemical balance of sodium and sulphur (Na/S balance), which reduces the production cost of pulp and paper. Sodium salts and sulphur anions are primarily in the water and condensates of the HTL process. The concentrations of sodium and sulphur, however, are low, in terms of ppm in the aqueous phase. Evaporating the water from the aqueous phase would incur a high thermal energy cost in an already thermally intensive process. To overcome this issue, supercritical water gasification of the aqueous phase is proposed.

Supercritical water gasification (SCWG) is considered the most appropriate separation approach because:


Additional benefits for implementing this process are the elimination of a hydrogen production plant and the reduction in wastewater treatment requirements.

The HTL, upgrading, and SCWG processes are simulated in Aspen PlusTM v9.0 (AspenTech, Bedford, MA, USA, 2020). The biomass into the HTL process was based on 2000 t/day organic loading, which is converted into bio-crude through the HTL process and upgraded to produce a gasoline and diesel blend.

Hydrothermal liquefaction is an energy-intensive process that operates at high temperature (315–355 ◦C) and pressure (220–250 bar). With these high operating conditions, heat and exergy recovery during cooling and depressurisation of the product flow greatly affects the economic competitiveness of the process. There is still a gap focused on increasing the energy efficiency of the HTL process. In the HTL process, the process conditions affect the product characteristics, yield, and quality. However, the temperature and pressure set points of some streams can be altered without affecting the product characteristics.

Ong et al. [25] established a novel and iterative method to optimise the mass and energy flows and asset of the HTL process, using heat and mass integration simultaneously with process simulation. The iterative procedure uses a combination of existing frameworks and design tools (process simulation tools, pinch analysis, process optimisation, and heat exchanger network design), with the inclusion of the process constraints of the HTL process. The procedure includes the simultaneous need for process simulation to provide in-depth analysis of the multiple impacts of the process modification opportunities. An important section of the analysis and optimisation is the implementation of process constraints based on the best available process knowledge from literature. As a result, the stream parameters and flow sheet design are modified and improved within defined parameters that do not affect the integrity of the process and product and that respect technological limitations. Based on the possible process modification and process constraints of the HTL process that would not affect the yield and quality of the product, the optimised flowsheet of the HTL process is presented in Figure 1.

#### **3. Results**

#### *3.1. Scenario 1: Kraft Mill with Hydrothermal Liquefaction System*

Scenario 1 considers the HTL process of radiata pine and black liquor. Total site heat integration for the HTL process and kraft mill has been used to have an insight into the integration potential between the HTL process and the kraft mill. The utility model is designed to understand the reduction in power generation due to the lower production of steam from the boilers.

#### 3.1.1. Integration of Hydrothermal Liquefaction with Kraft Mill

Figure 2 shows the utility system of the existing kraft mill. The black liquor solids (BLS) are burned in two recovery boilers, RB1 and RB2. PB1 and PB2 are the marginal fuel (MF) boiler that supplies steam to any deficit demands. The current turbine generates 30.4 MW of power, which is about half the power use of the kraft mill.

Figure 3 shows the utility system of integrating the hydrothermal liquefaction system concept with a kraft mill, using 18% of the black liquor. The three processes are internally heat integrated by exchanging mass and heat with the kraft mill, as outlined in Ong et al. [25]. The utility generated from the three processes are delivered to the kraft pulp process through the utility system. In Figure 3, it shows that the heat that is supplied from the HTL process reduces the marginal fuel of PB1 by 34.8%. Due to the lower steam production from the diversion of the black liquor solids and the decrease in the marginal fuel, the power generation reduced by 9.8 MW. The power generated is calculated using a correlated Willan's line based on the current turbine size. As a result, the inherent cost of black liquor as a feedstock includes the power generation lost by using part of the black liquor. The marginal fuel that is reduced from PB1 is sent to the HTL process as feedstock, which is cheaper due to the lower quality.

**Figure 2.** Simplified utility system of the existing kraft mill.
