High-Temperature Heat Pumps for Electrification and Cost-Effective Decarbonization in the Tissue Paper Industry

Abstract

The pulp and paper industry is under increasing pressure to reduce its energy consumption and carbon footprint. This study examines the feasibility of integrating high-temperature heat pumps (HTHP) into tissue paper production to enhance energy efficiency and decarbonization. Focusing on the energy-intensive drying process, the study uses data from a typical tissue paper mill to simulate and optimize an HTHP system producing four tons per hour of nine-bar saturated steam. It also addresses necessary modifications for HTHP integration applicable across the sector. Various refrigerants were analyzed, achieving a maximum coefficient of performance (COP) of 2.01. Results showed that HTHP can reduce energy consumption and emissions by up to 17% and 40%, respectively, based on the European electricity mix. Although steam production costs increase by 55% compared to fossil fuel-based systems, HTHP is more cost-effective than direct electric resistance heating, which raises costs by 196%. With a CO₂ price of EUR 100/t, HTHP offers a 12% cost reduction. However, without public funding, capital expenditures may be unsustainable in many regions, though viable in countries with favorable gas and electricity price differentials. The paper underscores the need for advancements in HTHP technology and cost reductions, emphasizing industry adaptation for seamless HTHP integration.

1. Introduction

In 2022, the industrial sector, predominantly reliant on fossil fuels, emitted 9.0 Gt of CO₂, which accounts for 25% of global energy-related emissions. The International Energy Agency (IEA) reports that these emission levels exceed the benchmarks set for future sustainability; specifically, the Net-Zero Emission targets aim for a reduction to 7 Gt by 2030 [1].

1.1. Energy Demand in the European Pulp and Paper Industry

In 2022, the pulp and paper sector accounted for nearly 2% of industrial emissions, with projections indicating an increase in total paper production by 2030 [2]. This sector relies significantly on biomass, which fulfils 60% of its energy needs, though it still utilizes substantial amounts of fossil fuels, such as natural gas and coal [2]. The European Confederation of Paper Industries (CEPI), which represents 18 European countries and accounts for 92% of Europe’s pulp and paper production, reported that in 2021, its members consumed 1356 PJ of primary energy, with biomass and natural gas providing 728 PJ and 395 PJ, respectively [3]. This extensive use of bioenergy varies significantly across different types of facilities. Biomass is predominantly utilized in pulp and integrated pulp and paper mills, whereas stand-alone paper mills exhibit high fossil fuel intensity due to their lack of access to wood residues [4].

The tissue category, which includes lightweight paper products for personal and kitchen use, is projected to grow at a 6.45% annual rate from 2023 to 2028 [5]. Tissue paper is more energy-intensive than other paper grades due to its low grammage and specialized drying processes [6], which require higher temperatures, complicating the integration with renewable energy sources [7].

Although prioritizing energy conservation is crucial from a decarbonization perspective [8], its potential to reduce fossil CO₂ emissions is limited, typically around 15%, though some estimates suggest it could exceed 20% [9]. However, many of the solutions suggested in the literature [6,10] for energy saving have become widely commercialized and often implemented [11].

1.2. Process Heat Decarbonization through Electrification

A significant opportunity for the decarbonization of the paper industry may reside in electrification, which has the potential to reduce emissions by up to 29% under current conditions, particularly when accomplished with the use of HTHPs [9].

In general, the authors of [12] suggested that electrification of process heat could reduce industrial greenhouse gas emissions by 78% by 2050, assuming an electric carbon intensity of 12 gCO₂/kWh_el. The authors proposed a phased approach, where industrial processes of increasing temperature and complexity are addressed through electric devices. This study estimated a demand of 470 TWh of electricity in the initial phase, necessitating the development of a new fleet of renewable power plants to meet this requirement. A similar analysis conducted in [13] projected that the potential renewable energy production available for industry by 2030 could exceed the 2015 levels by 250 to 375 TWh. This introduces the challenge of energy scarcity, underscoring the importance of prioritizing energy conservation. Heat pumps, which elevate low-temperature thermal energy to supply loads, offer significant electricity savings compared to direct electric heaters. Consequently, overcoming the challenges associated with their application in a broader industrial context is of paramount importance. The paper industry, which requires 97% of its heat at temperatures below 400 °C, was identified [12] as a potential starting point for developing and disseminating HTHP on a systemic level. However, the relative costs of these technologies may present significant barriers. Paper manufacturing ranks second in terms of potential recovery via heat pumps within industrial subsectors, with 10.78% of its heat consumption potentially covered by upgraded waste heat [14].

1.3. High-Temperature Heat Pumps: Technologies, Costs, and Refrigerants

Practical challenges associated with HTHP are examined in [15], which reviews specifications from manufacturers, such as maximum temperatures, coefficients of performance, and refrigerants. Notably, the highest temperature attainable by these systems, such as the 165 °C in the Kobelco device, may not meet the requirements of contemporary tissue paper production systems. The system proposed by Kobelco consists of two stages: a heat pump operating with refrigerant R245fa and a steam compressor performing mechanical vapor recompression (MVR). The authors of [16] focused on HTHPs designed for industrial steam production. These are categorized into closed systems and open systems. The latter includes steam compression, which can be carried out by a compressor (MVR) or an ejector, defined as thermal vapor recompression (TVR). Moreover, this latter technology is widely used in the paper industry [17]. According to the authors of [16], open systems could produce steam up to temperatures of 350 °C.

Recent advancements in the commercialization and development of HTHPs by machinery manufacturers are reported within the International Energy Agency’s Annex 58 program. Several companies have entered the design of industrial heat pumps, focusing on diverse technologies and sizes capable of achieving sink temperatures above 150 °C at technology readiness level 9, such as Olvondo’s Stirling cycle, Heaten’s system with a piston compressor, and Turboden’s centrifugal compressor system [18]. The cost uncertainty of these devices is largely attributable to the lack of standardized technology. This is clearly illustrated by the specific investment costs reported by Heaten and Olvondo for their respective devices, with Heaten’s costs ranging from EUR 280 to 450 per kW_th, and Olvondo’s around EUR 1200 per kW_th.

Due to their operating principles, HTHPs often require specific fluids, commonly referred to as “refrigerants”. These fluids are termed “refrigerants” because they were originally examined for use in cooling devices. The selection of the appropriate refrigerant has become a critical issue for HTHP. Following the Montreal Protocol adoption in 1987, which aimed to phase out chlorofluorocarbons (CFCs), HFCs gained prominence due to their negligible Ozone Depletion Potential (ODP). However, their considerable Global Warming Potential (GWP) prompted regulatory institutions to advocate for their phase-out. Recently, there has been a debate over the definition of per- and poly-fluoroalkyl substances (PFAS) to be banned due to their potential harm. According to some proposals, hydrofluoroolefins (HFOs) and hydrochlorofluoroolefins (HCFOs) could also fall within this definition [19,20]. For this reason, new so-called “natural” refrigerants, such as hydrocarbons, have been evaluated. In this context, the authors of [21] compared eight advanced cycle configurations and nine low-GWP fluids across various temperature lifts. The findings were analyzed through energetic, economic, and environmental criteria. The results indicate that a two-stage device operated more efficiently at higher temperature lifts (above 60 K). Furthermore, despite challenges in making broad comparisons across different cases, the two-stage cascade architecture demonstrated superior performance for the highest temperature lifts. The significance of refrigerant selection is further emphasized in [22], which evaluates the performance of various fluids in heat pumps operating within a temperature range of 50 °C to 150 °C. The study identified benzene and acetone as particularly effective for achieving the highest temperature lift. Additionally, the authors of [23] highlighted the superior performance of natural refrigerants by simulating the performance of 25 different refrigerants in both constant and variable lift heat pumps. However, the authors of [23] also pointed out two significant challenges: low density, as observed with water, and flammability, as seen with hydrocarbons. According to [22], the increased flammability risk may be justified if the performance of these refrigerants surpasses that of other fluids. Another issue addressed in [22] is that the waste heat temperature should ideally allow evaporation at above-atmospheric pressure. If this is not achieved, the evaporator and associated piping must operate under reduced pressure, increasing the risk of air infiltration and necessitating more costly equipment.

1.4. Research and Evaluations on HTHP Implementation in Industries

Despite these uncertainties, several studies have attempted to evaluate the decarbonization potential in industrial applications, including assessments of the associated costs or potential savings. The authors of [24] explored using HTHPs as substitutes for fossil-fuel-fired boilers to produce hot water and steam. The paper highlighted good performance in simulations and test fields, indicating potential energy savings of up to 57% compared to the base case. However, the achievable sink temperatures of HTHPs are below those required for tissue paper production.

If the authors of [14] identified the paper industry as the second sector likely to be significantly impacted by a large-scale process heat electrification campaign, the food and beverage industry is regarded as the primary candidate. This aligns with the results in [12], which, in a multi-step approach to electrification, recognized the food and beverage industries as particularly receptive to early-stage electrification efforts. This likely explains the numerous studies dedicated to evaluating the potential for electrification within this sector. For instance, the authors of [25] reviewed various technologies for the decarbonization of the food industry, which involves multiple processes operating at different temperatures. When producing steam at 120 °C, heat pumps are the only technologies that achieve a levelized cost of heat (LCOH) lower than or comparable to that of natural gas boilers. However, this is highly dependent on the cost of natural gas. The Emission Trading System (ETS) [26] is a carbon pricing mechanism, and it may incentivize large natural gas consumers to adopt these solutions [13]. Still, in relation to the food and beverage industry, the authors of [27] estimated the potential of heat pumps in dairy industries across various regions of the United States. They concluded that this solution is viable in many areas, even in the context of very low natural gas prices and assuming a modest carbon price of USD 20 per ton. A similar analysis was conducted in [28] for the food and beverage industry in Germany. The study found that HTHPs have the potential to cover up to 12 TWh of process heat. However, the actual impact on decarbonization may be limited by the high reliance on coal in the German energy mix. Similarly, the authors emphasized the importance of the carbon tax level, which serves as a critical factor in determining the economic viability of the technology, a factor also highlighted in [29]. Their findings indicated that while HTHPs outperform heat transformers in chemical industry applications, their competitiveness with natural gas boilers is highly dependent on the carbon tax level.

The authors of [30] analyzed the results of an optimization based on the Swiss TIMES Energy System Model (STEM) to evaluate the potential of heat pumps in the food and beverage and paper industries in Switzerland. The study found that up to 100 MW_th in the paper industry and 900 MW_th in the food and beverage industry could potentially be replaced with HTHPs. However, achieving this transition may require carbon tax levels in the range of EUR 100–400 per ton of CO₂ to make the investments economically viable. A similar analysis based on the STEM is presented in [31], which focuses on the Swiss paper industry. The study suggested that emissions in this sector could be reduced by 71% from 2020 to 2050. According to the authors, achieving the Net-Zero Emissions target by 2050 would require the electrification of heat, primarily driven by heat pumps and co-generative systems fueled by biomass.

To the best of our knowledge, the only paper that delves deeply into the practical application of high-temperature heat pumps in the paper industry is [32]. This study presents various solutions, including the innovative replacement of impingement air with superheated steam, and demonstrates that primary energy demand can be reduced by two-thirds and greenhouse gas emissions by over 80% compared to the baseline case, due to the introduction of heat pumps.

1.5. Scope of the Paper

This paper proposes and evaluates replacing the current steam production system based on natural gas with an HTHP in a tissue paper mill. Several technical modifications have been proposed to align the operational conditions of the production system, with the capabilities of an HTHP (maximum temperature around 175 °C) that approaches the potential of systems under commercialization. The three selected refrigerants—representing the categories of HFC, HCFO, and natural HC—were chosen to provide a broader scope for analysis. While they may not represent the optimal selection that could be derived through more rigorous methods, such as those in [22] or [23], they were selected to enhance the general applicability of the study.

Based on the literature, a standard thermodynamic cycle capable of effectively operating with the selected refrigerants was chosen. After optimization, based on the boundary conditions of the actual system, the energy, environmental, and economic performance of the hypothetical HTHP integrated into the production system was evaluated. HTHP performance was then compared with two benchmark systems, one representing the current practice, powered by natural gas, and a hypothetical system based on direct electrification via electric resistors. To consolidate the potential of HTHP technology, its economic viability has been assessed, along with the maximum initial investment cost that could be sustainable.

This paper is motivated by insights from several studies [12,14,30,31] that highlight the significant role HTHPs may play in the future of the paper industry. While those studies primarily focus on exploring the broader policy frameworks necessary to promote the widespread adoption of HTHP technology, this paper aims to delve deeper into the practical potential and limitations of implementing heat pumps within an industrial system. By doing so, it seeks to identify and expose the challenges that researchers and engineers must address to advance the practical application of HTHPs. This paper adopts a similarly functional approach to [32], but it characterizes and optimizes the hypothetical heat pump from the ground up. This approach is intended to highlight key operational features, such as the pressure ratio, maximum temperatures, and heat exchanger transfer curves.

The focus on the tissue paper sector is motivated by its unique characteristics [17], making it an interesting proving ground for HTHP applications that require a sink temperature above the 150 °C threshold. This sector requires steam at higher temperatures than other paper grades and may have a higher temperature waste heat source in the wet exhaust. This paper is prompted by the lack of literature regarding the decarbonization of the tissue paper industry through HTHP. By addressing this gap, the study aims to raise awareness within this fundamental sector and attract the attention of researchers who may propose innovative solutions to enhance the integration of HTHP technology in the tissue paper industry.

2. Materials and Methods

2.1. Cases

This section introduces three plant layouts to investigate the limits and benefits of introducing a high-temperature heat pump in a tissue paper manufacturing process:

• The first represents the current state-of-the-art plants with a high degree of thermal recovery.
• The second is characterized by the replacement of the natural gas boiler with an electric one.
• The third introduces the high-temperature heat pump to enhance waste heat recovery.

This is the primary focus of this analysis. Toscotec SPA, a globally leading company in industrial plants for tissue paper production, provided the specifications related to the plant and manufacturing process. These specifications do not precisely match their commercial offerings but generally represent the most efficient systems currently operating in the examined sector. Although each plant has specific features due to different processing requirements and supply needs, these figures provide general validity to the analysis.

2.1.1. Layout 1: Natural Gas Boiler

This section describes the plant layout of a state-of-the-art tissue paper production plant, as illustrated in Figure 1. Hereafter, this will be referred to as Layout 1. A flow of cellulose pulp diluted in water is sent to the thermal drying system. It has been formed on a wire and has undergone an initial phase of mechanical dehydration by gravity and pressing. A key characteristic of tissue paper production is the absence of a drum battery: the paper is formed by flowing over a single large steel cylinder, known as a Yankee, inside which steam is blown and condensed. Another distinctive feature is the co-primary role of hot air impingement, which involves an additional heat exchange by convection from the heated air inside the hoods to the forming sheet [17]. The analyzed configuration produces about 5 t/hour of paper. The drying process is powered by natural gas, which is used by the burners of the hood section to produce hot air up to 500 °C and the steam generator to produce steam at 17 bar. Steam at 9 bar is needed for paper production, but excess pressure is used to recover the non-condensed steam through an ejector (“thermo-compressor”), as described in the following. This system, found in numerous areas of paper production [17], is an example of TVR technology.

In addition to natural gas, fresh air is introduced into the system, which replaces the air discharged to the environment with the water removed from the paper during the drying process. Finally, a cellulose fiber flow enters, diluted at 45% mass based in water, and exits with a moisture content of 5%, which corresponds to a water evaporation rate of 5.5 t/hour. The fiber flow is not depicted in Figure 1 and subsequent figures, as it is insignificant for the analysis.

At the hood exit, warm (310 °C) and humid air flow is sent to the waste heat exchangers to recover part of the residual heat, which may be used for process or auxiliary functions. Among the various possible uses of thermal recovery that directly impact the process, one can mention the production of 1 ton/hour of process steam. This reduces the flow required from the boiler to 4 tons/hour.

The described setup exemplifies a system with a high degree of thermal recovery. However, only a part of the existing fleet has implemented these systems. Considering the critical role of energy efficiency in plant decarbonization [10], adopting advanced recovery systems is prioritized over any additional energy efficiency measure (e.g., the HTHP introduction). For this reason, only interventions that do not disrupt the energy recovery operation are considered here.

For clarity, the diagram in Figure 1 features an external combustor heating the air directed into the hoods. However, it should be noted that combustion is conducted internally in the actual system operation, leveraging internal recirculation and preheating mechanisms to enhance thermal efficiency. This simplification does not compromise the accuracy of the thermal and mass balances critical to the system’s evaluation and is deemed acceptable for this analysis.

The components considered for replacement with an HTHP are the natural gas steam boiler (“NG steam boiler”) and the thermo-compressor. Although the natural gas air heater (“NG air heater”) serving the hoods plays an essential role in the process economy, the required temperatures are too high for commercial heat pumps or those under imminent development [18].

The TVR system is essential for this setup. In fact, of all the steam sent to the Yankee cylinder, a portion evaporates on the inner surface. However, another part of the steam does not condense and moves out of the cylinder by a pressure differential, carrying the condensate with it, thus evacuating the Yankee’s inner cavity. A phase separator separates the escaping dry-steam fraction (the so-called “blow-through steam”) from the condensate, which is returned to the boiler. The blow-through steam is at around 8 bar due to Yankee’s pressure losses; therefore, it is compressed to 9 bar by an ejector powered by the steam at 17 bar from the boiler. These are illustrative values. The operating pressure in tissue papermaking can vary from 6 to 10 bar, which also influences the pressure of blow-through steam.

Table 1. Relevant data for Layout 1. Flow names refer to Figure 1.

Flow	Specifications
Hot air	ṁ = 11 t/h
Process water	ṁ = 5.5 t/h
Exhaust air 1	ṁ = 16.5 t/h
	T = 310 °C
Exhaust air 2	ṁ = 15 t/h
	T = 75 °C
Recovered water 1	ṁ = 1.5 t/h
Process condensate *	ṁ = 4 t/h
	p = 8 bar
HP saturated steam	p = 17 bar
Blow-through steam	ṁ = 5 t/h
	p = 8 bar
Saturated steam to Yankee *	ṁ = 9 t/h
	p = 9 bar
Fuel 1	Pth = 3300 kWth
Fuel 2	Pth = 2700 kWth

* For the sake of clarity, the additional steam flow produced by the thermal recovery system, referred to in Section 2.1.1, is disregarded.

presents the relevant specifications of Layout 1.

2.1.2. Layout 2: Electric Boiler

In the layout depicted in Figure 2 (referred to as Layout 2), the natural gas boiler for steam production is replaced with an electric boiler operating with an ideal electrical-to-thermal conversion efficiency of 100%, based on indication from equipment suppliers [33].

Apart from the heat production technology, the system configuration remains unchanged, maintaining the capability to generate steam at 17 bar. Therefore, this setup continues to support the operation of the thermo-compressor and enables downstream thermal recovery under analogous conditions.

Table 2. Electric power required by the steam boiler. It refers to Figure 2.

Electric power

Pel = 2400 kWe

only lists the electric power required by the electric boiler, as the other specifications are identical to those in Table 1. The electric power of the boiler replaces what was referred to as Fuel 2 in Table 1.

2.1.3. Layout 3: HTHP

Figure 3 depicts a configuration in which, unlike previous setups, the ‘NG steam boiler’ or ‘electric steam boiler’ and the associated thermo-compressor are replaced by a HTHP system. This configuration will be referred to as Layout 3. The heat pump recovers the residual heat in the flue gas cooled down to 75 °C by pre-existing thermal recovery units. Further cooling of the exhaust flue gas allows the condensation of an additional water stream (termed “Recovered Water 2”), which can be recycled in the process. This heat pump heats the condensate from the system, producing saturated steam at 9 bar. The steam at 17 bar has an evaporation temperature of around 204 °C, representing a significant technological and efficiency challenge for commercial HTHPs [15]. Therefore, in the strategy adopted in this study, steam is produced at 9 bar, to be directly sent to the Yankee cylinder. The saturation temperature at that pressure is 175 °C, which still represents an ambitious target for HTHP, partially mitigated by the fact that the steam is used directly without needing a further intermediate exchanger, so that 175 °C is both the “maximum” temperature of the working fluid and the required sink temperature. However, activating the thermo-compressor described previously will not be possible with steam at 9 bar. Hence, an auxiliary MVR system is required to reintegrate the blow-through steam into the cylinder. It is critical to note that this steam compressor is distinct from the heat pump’s integrated water/steam stage: while the heat pump directly replaces the boiler to produce the required steam at 9 bar, such an external MVR system exclusively manages the blow-through steam recirculation. The characteristics that distinguish this cycle from previous ones, including the technical specifications that the MVR system must meet to replace the thermo-compressor, are listed in

Table 3. Relevant data for Layout 3. They refer to Figure 3.

Saturated steam from HTHP	ṁ = 4 t/h
	p = 9 bar
Saturated steam to Yankee	ṁ = 9 t/h
	p = 9 bar
Electric power MVR	Pel = 40 kWe

. Further optimization will assess the performance of the HTHP. It will then be possible to evaluate the ‘Electric Power HTHP’, the ‘Exhaust Air 3’ flow characteristics, and the ‘Recovered Water 2’ flow.

2.2. Heat Pump Definition

2.2.1. Waste Heat Source

The heat pump recovers the waste heat from the wet flue gas, which carries the water extracted from the drying pulp slurry. At the hood outlet, the flue gas has a mass flow rate of 16,500 kg/h at 165 °C. This corresponds to a dry air flow of 3 kg/s with a moisture content of 0.5 kg_H2O/kg_DA. This flow is channeled to the heat recovery exchangers, which reduce the temperature to 75 °C and condense 1.5 t/h of water to be recycled in the process. The exhaust flue gas can be sent to the heat pump evaporator downstream of the heat recovery system. The flow specifications at the HTHP evaporator inlet are reported in

Table 4. Specifications of exhaust flow at the evaporator inlet.

Flow (dry air)	3.0 kg/s
Humidity	0.37 kgH2O/kgDA
Total flow	15,000 kg/h
Temperature	75 °C

. Figure 4 shows the temperature trend of the flow as a function of the heat transferred.

2.2.2. High-Temperature Heat Pump Layout

In the context of generating steam at temperatures exceeding 100–120 °C, the authors of [15,18] reported that the most suited HTHP solution is a two-stage system. The bottom stage is a vapor compression heat pump operating with a suitable refrigerant, which evaporates the condensate from the Yankee, while the top stage is an open water/steam circuit with a steam compressor, compressing the saturated steam from the bottom condenser (top evaporator) and sending it directly to the Yankee cylinder.

As the refrigerants suited for HTHP operating in the envisioned temperature range usually feature dry expansion, i.e., they show a positive slope of the saturated vapor curve in the T-s diagram, they must operate with a non-negligible degree of superheating before entering the HTHP compressor. To address this requirement effectively, an internal heat exchanger (IHX) is usually introduced in the HTHP [21]. Such a solution increases the temperature of the fluid exiting the evaporator by using residual heat from the flow exiting the condenser. Based on the recommendations of suppliers for comparable systems [18], it is proposed that the heat pump configuration incorporate a multi-stage screw compressor for the bottom stage and a centrifugal steam compressor for the top stage. The efficiencies of the compressors within the heat pump are presented in

Table 5. Performances of the compressors within the heat pump [34].

Screw compressor (bottom cycle)	Adiabatic efficiency: 80%
	Electro-mechanic efficiency: 90%
Centrifugal compressor (top cycle)	Adiabatic efficiency: 85%
	Electro-mechanic efficiency: 95%

. The layout of this heat pump is detailed in Figure 5.

The top cycle compressor elevates the steam pressure to 9 bar, as the process requires. Since the process cannot accept superheated steam, it is necessary to integrate a de-superheating mechanism by injecting liquid water after the steam compressor. The water evaporates, reducing the steam temperature while slightly increasing its mass flow. The water for de-superheating is recirculated from the Yankee condensate by a pump. The power required by this pump is neglected in this analysis.

2.2.3. Refrigerants

Three refrigerants were considered: R245ca, R1336mzz(Z), and n-Pentane. These represent the classes of hydrofluorocarbons (HFCs), hydrofluoroolefins/hydrochlorofluoroolefins (HFOs/HCFOs), and natural refrigerants, in this case, hydrocarbon (HC), respectively. The selection of refrigerants for the first stage is a critical consideration in the design of HTHPs [15]. The fluids used in the model were chosen based on the analyses conducted by [22,23], following the outlined criteria:

• The critical temperature of the refrigerant should be higher than or near the supply temperature to allow the optimization algorithm to explore a reasonable range of feasible solutions. Consequently, several fluids, such as ammonia, were excluded from consideration.
• The evaporation pressure of the refrigerant at the temperature achievable in the evaporator, using waste heat, should be above atmospheric pressure. The expected evaporation temperature is around 45 °C, so it has assumed that refrigerants must have an evaporation pressure higher than 1.01325 bar (1 atm) at 40 °C. Therefore, despite their proven performance [22,23], water, acetone, and benzene were not considered. Additionally, R365MFC, identified in [23] as the best-performing fluid among those studied, was excluded due to its evaporation pressure being slightly below 1 atm at 40 °C.

Some technical specifications for R245ca, R1336mzz(Z), and n-Pentane are reported in

Table 6. Technical specifications for the refrigerants that have been used in the simulations [22].

Name	Class	Critical Point	GWP 100	Flammability	psat,40°C
R245ca	HFC	174.4 °C 39.9 bar	726	1	1.73 bar
R1336mzz(Z)	HFCO	171.3 °C 29.0 bar	2	0	1.28 bar
n-Pentane	HC	196.6 °C 33.7 bar	5	4	1.16 bar

. The objective of this paper was not to identify the optimal refrigerant but to assess the potential of HTHPs in an industrial context. Therefore, more rigorous selection criteria (e.g., [22,23]) could lead to better outcomes. The selection of the three fluids used in this study was intended to align with potential regulatory considerations regarding refrigerants. For instance, R245ca, similar to many other HFCs, has a high GWP, R1336mzz(Z) may pose risks related to PFAS, as is the case with many HFOs and HCFOs, and n-Pentane, similar to many hydrocarbons, is flammable. The exploration of these three HTHPs is not intended to definitively determine the best combination of cycle and refrigerant but rather to generalize the analysis, taking into account potential regulatory directions.

2.2.4. Optimization Process and Analysis of Environmental Sustainability

The analyses presented in this paper are based on a simulated model of a high-temperature heat pump. Although no experimental applications were conducted, the operational requirements of the device were assessed in consultation with experts from Toscotec S.p.A., a leading company in the manufacturing of tissue paper machines. The model simulations were conducted using Aspen HYSYS v.12.1, a widely recognized commercial software extensively used in the simulation of energy and chemical plants. The humid air in the waste heat stream was modeled using the Peng–Robinson equation of state. This model, that is primarily recommended for hydrocarbon simulations, is considered suitable for air modeling as well [35], and it has been employed in previous studies for modeling humid air [36]. The water/steam flow was modeled using the Steam NBS package, which is specifically designed for water and steam [35]. Finally, the refrigerant fluids were simulated using the REFPROP model, which is based on the well-established fluid properties database compiled by the National Institute of Standards and Technology (NIST) [37].

The adiabatic and electromechanical efficiencies for the compressors are provided in Table 5. The pressure drop in the Yankee dryer was set at 1.5 bar, in accordance with the manufacturer’s specifications.

The optimization framework used was the BOX algorithm, primarily based on the “Complex” method of the BOX1 algorithm [35]. The objective of the optimization was to maximize the coefficient of performance (COP), defined in Equation (1):

COP = Useful heat flow/Compression power = Q_tot/(W_HTHP + W_MVR)

(1)

Compression power includes the electric consumption of both bottom- and top-stage compressors, W_HTHP, and the electric consumption of the external steam compressor, W_MVR. As previously mentioned, W_MVR was 40 kW, while the useful heat flow rate Q_tot was the heat transferred from the condensing steam to the Yankee. This quantity amounted to 2300 kW_th, as reported in

Table 7. Main constraints and boundaries for the optimization process of the heat pump.

Min. pinch point at evaporator/boiler heat exchanger	20 °C
Internal heat exchanger pinch point	10 °C
Max. temperature in the bottom cycle	200 °C
Max. temperature in the top cycle	300 °C
Min. pressure in the bottom cycle	1.01325 bar
Heat flow rate demand	2300 kWth

, along with the optimization’s other main constraints. Although more conservative, the selected pinch points for the heat exchanger were consistent with those reported in [38,39].

Since some compressor manufacturers offer solutions with discharge temperatures around 300 °C [40], this temperature was set as the maximum limit for the top compressor. However, it is assumed that in a practical system, there would be stages of intercooling and recovery that significantly reduce the discharge temperature without compromising the overall efficiency of the transformation. Instead, limiting the bottom cycle to a temperature of 200 °C is preferred here, in line with more commercially established technologies.

In the results, the following values will be compared:

• COP, as defined in Equation (1).
• Volumetric heating capacity (VHC), defined as the ratio between the produced heat flow rate and the refrigerant volumetric flow rate at the compressor inlet. This value is interesting from a technological viewpoint, as the higher it is, the smaller the system dimensions will be.
• Compression ratios (β) of the bottom and top cycle, which can be helpful to argue the number of stages required by the actual compressors.
• Water recovery, defined as the additional water condensed—and recovered—compared to the base case. While water recovery is not the focus of this article, paper companies are called upon to make an effort to manage water resources efficiently and with minimal waste. This issue is intrinsically linked to the broader problem of global warming because one of the potential consequences of climate change is water scarcity [41].

Temperature trends in the evaporator and condenser exchange sections will also be reported. Those may be useful for evaluating the performance of these devices.

After conducting optimizations with the three fluids, one of the three cases deemed significant will be selected for potential integration into Layout 3, described in Figure 3. The total steam production system, including the external compressor, will be compared in terms of primary energy consumption and GHG emissions with Layouts 1 and 2.

To obtain a broader picture, the values of the primary energy factor and the GHG emissions factor of the electric production related to some of the major European economies were considered. A European average value was also considered. The values refer to 2019 to rule out the subsequent disturbances the European economies have undergone following the COVID-19 pandemic and the Ukraine crisis. These values are reported in

Table 8. The electrical system’s primary energy factor and GHG emission factors for five major paper producers in Europe, including EU average values.

Country (2019)	Primary Energy Factor [42] (kWhP/kWhel)	GHG Emission Factor [43] (kgCO2/kWh)
EU	1.96	0.255
France	2.43	0.06
Germany	1.75	0.347
Italy	1.69	0.234
Spain	1.97	0.214
Sweden	1.44	0.01

. Five of the main paper-producing countries in Europe were selected. Their annual production exceeds 60% of the European output [3]. Finland, another major producer, was excluded due to difficulties in finding data comparable to that of the other countries.

The primary energy and emission factors for natural gas are reported in

Table 9. Primary energy factor and GHG emission factor for natural gas [44].

Primary energy factor per NG	1.1 [44]
GHG emission factor per NG	0.2 kgCO2/kWh [45]

.

2.2.5. Economic Analysis

This section analyzes some economic indicators expected from the high-temperature heat pump solution compared to Baselines 1 and 2. This comparison will be made for two scenarios, with and without considering a carbon tax.

Regarding the prices of natural gas and electricity, data for 2019 from Eurostat were used [46], considering non-domestic consumers in the same countries as in

Table 10. Energy prices for five major European paper producers, including EU average values [46].

Country (2019)	NG Price (EUR/MWh)	Electric Energy Price (EUR/MWh)
EU	26	86
France	26	68
Germany	26	100
Italy	26	113
Spain	28	83
Sweden	39	53

. A carbon tax value of EUR 100/tCO₂ was considered. This value aligns with expectations related to the carbon permits under the European Union ETS for 2025 [47] and represents a conservative estimation compared to alternative forecasts [48].

In the case of Sweden, it is worth noting that its tissue paper mills extensively utilize liquefied natural gas (LNG) or liquefied petroleum gas (LPG) [49]. Both options are more expensive than simple natural gas and have equal or higher emissions. Therefore, the presented analysis is considered conservative for Sweden.

In the economic evaluation, the burden of emission permits will be attributed only to the combustion of natural gas in the paper mill. Those related, for example, to electricity production from the grid, will be excluded. Although this is formally correct while considering the paper mill alone, larger carbon taxes may also affect future electricity prices.

Where the differences in expenditure highlight a possible saving for a pair of natural gas and electricity prices, the process continues estimating the acceptable capital expenditure (CAPEX) for these systems. A CAPEX is acceptable if it allows a payback period (PBP) of less than five years. The simple PBP used for this evaluation is defined in Equation (2):

PBP = CAPEX/ΔEx

(2)

ΔEx represents the saving in expenses for steam generation compared to Layout 1.

3. Results and Discussion

3.1. Heat Pump Performances

In

Table 11. Optimization results.

Name	COP (−)	VHC (kJ/m3)	β Bottom (−)	β Top (−)	Water Recovery (t/h)
R245ca	1.89	1650	11.4	3.8	1.7
R1336mzz(Z)	1.90	1240	12.3	3.7	1.7
n-Pentane	2.01	1150	10.8	3.8	1.8

, the optimization results for three refrigerants are reported. Regarding the COP, the case with n-Pentane performed slightly better than the other fluids, with a COP 5% higher than that of R245CA and R1336mzz(Z). For this reason, a COP of 1.9 will be considered representative of the proposed application’s performance. As previously mentioned, these fluids may face regulatory limitations due to environmental laws, but they offer lower flammability and a higher volumetric heating coefficient.

Both the screw compressor of the bottom cycle and the centrifugal compressor of the top cycle could consist of two compression stages [17,50].

3.1.1. Water Recovery

The heat pump exchanger operates in a temperature range where the latent heat of the flue gas is significant. This also results in the condensation of a considerable amount of water, which can be recycled directly inside the process. From what has been seen previously, it can be assumed that without any thermal recovery measure, the plant would disperse about 1.1 tons of water into the atmosphere per ton of paper produced. Thermal recovery systems reduce this value to about 0.8 tons. Introducing the heat pump could further reduce this value to below 0.5 tons of water dispersed. This means a reduction of about 37% compared to the case with simple thermal recovery.

3.1.2. Optimization Results

Figure 6 shows the curves describing the coupling between the heat exchanges between the refrigerant and the exhaust flue gas (evaporator) and between the refrigerant and the steam (condenser) in the heat pump’s bottom stage.

Evaporators operate at compromised temperatures between the needs to:

• Maintain the minimum pinch point of 20 °C, imposed as a constraint in the optimization,
• Extract a significant amount of heat flow rate from the flue gas to reduce the demand for electricity,
• Keep the refrigerant pressure high to reduce the compression ratio, thus improving the COP and maintaining evaporating pressure above atmospheric conditions.

The optimizer identified a similar evaporation temperature for all three fluids between 45 and 48 °C. Due to the different properties of the fluids, these temperatures were achieved, respectively, at 2.2 bar, 1.6 bar, and 1.4 bar for R245CA, R1336mzz(Z), and n-Pentane.

On the hot side, the condenser receives the refrigerant exiting the compressor; on the cold one, it enters the condensate from the Yankee. The needs that the optimizer had to balance in this case were:

• Maintaining the maximum temperature of discharge from the bottom compressor below 200 °C. This also involves the superheating that the fluid receives at the IHX.
• Limiting the compression ratio to increase COP.
• Maintaining a pinch point of 20 °C at the condenser.
• Maintaining the maximum temperature of discharge from the top compressor below 300 °C.

Since the maximum temperature of 300 °C at the compressor of the top cycle was reached in all cases, a compression ratio between 3.7 and 3.8 was employed in all cases. This implies that the steam evaporated at a pressure of about 2.4 bar in the top cycles of all the layouts examined.

Hence, the condensation temperatures of the refrigerant were all similar, around 146 °C. The condensation pressures will thus be 24 bar, 19 bar, and 15 bar for R245CA, R1336mzz(Z), and n-Pentane, respectively. Compared to the evaporator, the de-superheating contribution in condensation accounted for a larger share of the heat, roughly between 40% and 45% in all cases. All three fluids approached the maximum temperature imposed by the constraints for the bottom compressor outlet, i.e., 200 °C. This temperature is higher than that corresponding to their condensation pressure and is caused by the superheating due to adiabatic compression. This can lead to considerations regarding introducing an intercooled compression, possibly recovering the compression heat, which could reduce the required work.

3.2. Analysis of Environmental Sustainability

Figure 7 shows the difference in primary energy consumption in the system with the heat pump compared to Layout 1, which operates exclusively with a natural gas boiler. The data refer to the primary energy factors previously reported for various European states. Similarly, the difference in primary energy for Layout 2, i.e., the case where the natural gas boiler is replaced with an electric boiler, is also reported. Introducing heat pumps can lead to savings of about 17% on an average European basis. The mix of sources in the electric system highly influences such a result. In 2019, Sweden had a renewable penetration in the electric sector of 59% [15], and the related savings in terms of primary energy were around 35%. On the other hand, France has a smaller share of renewables and a massive production from nuclear power, which is heavily penalized in terms of the Primary Energy Factor, leading to a potential slight increase in primary energy consumption (+3%). Introducing the electric boiler into the system increases primary energy consumption in all scenarios. At the European level, there could be a potential increase in consumption of 58%. Even in Sweden, there might be an increase of 16%. In France, the related increment would be around 96%. These data are significant considering that the increase in primary energy referred to in the graph effectively involves replacing natural gas with electricity. The production, transportation, and dispatching of low-carbon energy for large systems are significant technological challenges and a social burden. Therefore, the lower demand for primary energy associated with heat pump solutions is relevant, especially from nation- and continent-wide perspectives.

In Figure 8, the reduction in GHG emissions is shown for the case of introducing the heat pump compared to the baseline case. In this scenario, the emission variation is also added in the case of an electric boiler. Both solutions generally contribute to a reduction in GHG emissions. At the European level, with the 2019 electricity generation mix, Layout 2 would result in a 13% increase in emissions, while Layout 3 could reduce them by 40%. Even in Germany, the country with the highest energy CO₂ intensity among those considered, replacing it with an HTHP could significantly reduce emissions (−19%). The difference from the baseline case is generally larger, as the energy mix for electric production is more decarbonized. This applies to both the heat pump case and the electric boiler case. Therefore, in Sweden, where GHG emissions for the electric mix are very low, the level of emission reduction is comparable in both cases.

In conclusion, the heat pump significantly reduces emissions even with partially decarbonized electrical systems. This can be useful in achieving the medium-term objectives set by the European Union and various international institutions.

3.3. Economic Analysis Results

Figure 9 shows the differences in costs to produce process steam for the system with a heat pump (Layout 3) and the one with an electric boiler (Layout 2) compared to the baseline case (Layout 1). In general, an increase in costs can be expected with both alternative solutions. What mainly differs between these two solutions is the extent of the additional energy expense: at the European level, Layout 3 would increase expenses by 55%, while Layout 2 would increase them by 196%. It is important to note that the greater the price differential between electricity and natural gas, the more significant the increase in expenses. In any case, the adoption of heat pumps, compared to direct electrification, helps to reduce the impact of the phase-out of natural gas, at least in terms of operational expenses.

In Figure 10, the difference in expenses is reported for the case where a carbon tax (of the ETS allowances type) on direct carbon dioxide emissions was introduced. With a value of EUR 100/t_CO2eq, there could be a saving by adopting the system with a heat pump compared to the baseline case. At the European level, this could be about 13%. Despite the application of this tax, at the European level, introducing an electric boiler can cause an increase in costs for steam production compared to Layout 1, although this increase is reduced to 66% compared to what is reported in Figure 9. Due to the high cost of electricity, introducing these systems in the Italian and German scenarios may potentially increase expenses.

Table 12. Maximum CAPEX and specific CAPEX to obtain a PBP less than 5 years.

Country	Specific CAPEX (EUR/kW)
EU	280
France	670
Spain	420
Sweden	1600

reports the specific CAPEX below which the system achieves the PBP target in five years. The value for the European Union is around EUR 620,000 in terms of absolute CAPEX and EUR 280/kW in relative CAPEX. In the literature, two systems can be related to the one modeled in the article. These are the Kobelco SGH 165 system, and the system proposed by Turboden [18]. A potential specific CAPEX that fluctuates between EUR 300 and 1000 per kW is reported for the latter. For the Kobelco SGH 165 system, there is no direct data from the supplier, but the authors of [51] estimated a price between EUR 500 and 1500/kW. Generally, with a CO₂ price of EUR 100 per ton, the threshold CAPEX to ensure economic sustainability could still be below the ranges deemed plausible by manufacturers. Furthermore, the authors of [18,51] did not account for the external MVR—the one used to recirculate the blowout steam in the Yankee cylinder. At the same time, they could be significant compared to Layout 1.

It is challenging to definitively assess the potential savings associated with this solution due to significant uncertainties, particularly concerning the cost of the device and operational expenditures, which were not included in the previous analysis. For instance, the need to train boiler technicians to operate these more complex systems could present difficulties. Further uncertainties arise from the volatility of natural gas prices. Although the analysis encompassed several countries to account for price variations, future projections, such as those from the International Energy Agency, indicate that decarbonization policies may reduce natural gas demand [52]. This reduction could result in lower natural gas prices, potentially hindering decarbonization efforts, including the adoption of HTHPs. However, the ETS allowances may counterbalance this trend by driving up prices, as forecasted in [48]. Despite these uncertainties, current conditions in France and Sweden already suggest a favorable combination of electricity and natural gas prices, making this solution viable, particularly if the cost of CO₂ equivalent reaches EUR 100 per ton. For other countries, manufacturers will need to strive to reduce prices while increasing the applicability of these systems. Also, paper-machine manufacturers are asked to contribute to adjusting the operating conditions of their plants in order to facilitate the introduction of high-temperature heat pumps, which adds to the challenges and opportunities brought by the introduction of HTHP and MVR systems in the tissue paper mills.

As previously mentioned, we have also overlooked the possibility that the carbon tax’s weight could impact the electricity price, making the return on these investments more complex. On the other hand, considering the potential reiterated in Section 3.2, a form of public financing could be considered to facilitate the widespread of this emerging technology.

Comparing HTHP and Other Decarbonization Strategies

Introducing heat pumps and direct electrification risks increasing tissue paper’s energy costs (and generally the production costs). Suppose these solutions help extend the scope of the electrical grid decarbonization. In that case, they require investments and innovations in producing, transmitting, and dispatching low-carbon electricity to achieve the broadest levels of availability, stability, and economy.

Two technologies that represent potential alternatives to those studied in the paper are e-fuels and biomass. Regarding e-fuels, the concept involves using electricity to generate products that can replace fossil fuels more or less directly. In any e-fuel production process, from electrolysis onwards, conversion efficiencies are often below 70% [53]. Therefore, a complete replacement of fossil fuels with e-fuels would require production, transmission, and dispatching deployment of electric power, which is even greater than that related to direct electrification. Recent studies [54] suggest how the use of hourly energy surpluses could integrate these processes into efficient electric grid management. On the other hand, if the electric surplus is insufficient, the production of e-fuels and chemicals (hydrogen, ammonia, methanol, and others) might not be enough to cover the existing demand. In this case, paper and chemical industrial sectors, which utilize them directly as feedstock [53], might compete with each other, affecting the price of the e-fuel itself.

The use of biomass, on the other hand, faces availability issues. Due to the low energy density, biomass should be abundant near the plant to reduce the energy, economic, and environmental costs of transporting this fuel. While this solution is feasible in some cases [4], it is unlikely to be considered generalizable across all of Europe.

Further considerations on these two technologies are related to the fact that tissue paper production, in addition to process steam, generally also requires air at 350–500 °C. Since heat pumps do not yet seem capable of reaching these temperatures, it is conceivable that a portion of e-fuel or biomass (appropriately converted) could compete with direct electrification in producing hot air.

4. Conclusions

The electrification of industrial heat is a strategy to promote the decarbonization of goods and services. Some limits and possibilities of using high-temperature heat pumps (HTHPs) have been evaluated in producing tissue paper. This energy-intensive sector, tied to a fast-moving consumer good, has not received much attention recently. The replacement of the natural gas boiler for producing process steam at 9 bar with a two-stage heat pump was evaluated. The first stage uses a refrigerant, while the second operates directly on the steam. This system was optimized to maximize the coefficient of performance (COP) using the process specifications of a real system. With the three refrigerants used in the simulations—R245CA, R1336mzz(Z), and n-Pentane—a maximum COP ranging from 1.89 to 2.01 was achieved.

With the European electric mix of 2019, selected to rule out the effects related to the COVID-19 pandemic and international instabilities, this system could reduce primary energy consumption and emissions by up to 17% and 40%, respectively, compared to steam production with a natural gas boiler, considering the average EU energy mix. The progressive decarbonization of the grid should dramatically improve these results. The comparison with a direct electrification system with resistors was significant. With the average EU energy mix, using an electric boiler would increase primary energy consumption by 58% and CO₂ emissions by 12%, thus worsening the carbon footprint of tissue paper production. This will hold true until the electric generation mix has achieved an adequate level of decarbonization. Even at that point, however, reducing primary energy consumption will be an advantage of HTHPs.

With European energy prices in 2019, the cost of producing process steam would increase by 55% in the case of a heat pump. In the case of an electric boiler, however, this would increase by 196%. The results change if an emission taxation system, such as the EU’s ETS, is considered. With a carbon dioxide equivalent price of EUR 100 per equivalent CO₂ ton, the heat pump system could lead to savings of 12% compared to the baseline case (natural gas boiler). The system with an electric boiler would still increase costs by 66%, making tissue paper production potentially uneconomical and, thus, financially unsustainable.

Replacing a natural gas boiler with a heat pump implies adopting a complex device with higher capital costs than a traditional or electric boiler. The savings achieved when a carbon tax is considered would lead to a return on investment in five years if the specific CAPEX of the plant is below EUR 280, considering the average EU energy prices and energy mix. This figure appears to be lower than short-term market projections for HTHP and MVR systems; however, in some countries, such as France and Sweden, where the differential between electricity and natural gas prices is narrower, adopting such devices could already be economically viable with a PBP lower than five years.

In conclusion, efforts must be directed along two paths to make these solutions feasible in the tissue paper industry and other industrial sectors. On one hand, heat pump manufacturers must strive to increase their systems’ applicability, efficiency, and cost-effectiveness. On the other hand, investment is necessary to ensure that low-carbon electricity is increasingly available, reliable, and affordable. Additionally, it would be prudent for manufacturers of paper production equipment to adapt their systems to integrate heat pump technology.