Investigation of Combustion and Performance Limits of Liquid LPG Direct Injection in a Single-Cylinder Research Engine

Investigation of Combustion and Performance Limits of Liquid LPG Direct Injection in a Single-Cylinder Research Engine | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Investigation of Combustion and Performance Limits of Liquid LPG Direct Injection in a Single-Cylinder Research Engine Aristidis Dafis, Hermann Rottengruber, Bastian Rabe This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8640603/v1 This work is licensed under a CC BY 4.0 License Status: Under Revision Version 1 posted 10 You are reading this latest preprint version Abstract This work investigates the direct injection of Liquefied Petroleum Gas (LPG) in its liquid phase in a single-cylinder research engine, focusing on the primary challenge of maintaining LPG in its liquid state throughout the fuel system to prevent vaporization and gas bubble formation. Due to propane’s low critical temperature (96.8°C), fuel-carrying components exposed to typical engine operating temperatures (80–110°C) are prone to outgassing, which creates pressure drops and disrupts continuous liquid injection. To overcome this, the 200-bar high-pressure fuel system was optimized by combining increased pressure, active cooling, gas separation, and recirculation strategies, ensuring stable liquid conditions upstream of the pump and injector. This robust system allowed for continued stable engine operation and enabled comprehensive comparative measurements—including in-cylinder pressure analysis and FTIR-based emission measurements—to systematically evaluate combustion behavior and environmental impacts. The results demonstrate that LPG’s superior knock resistance (RON 103–111) allowed for earlier ignition timing and higher engine loads compared to E10, with beneficial charge cooling effects further suppressing knock occurrence. In addition, LPG direct injection yielded significant reductions in particulate and CO₂ emissions relative to gasoline, confirming both the performance and environmental advantages of liquid LPG direct injection technology for future engine applications. Combustion Analysis Power Density LPG Direct Injection Knock Limit Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 1 Introduction Motivation The pursuit of cleaner and more efficient combustion processes has led to increased interest in alternative fuels for internal combustion engines. Among these, Liquefied Petroleum Gas (LPG) presents a compelling option. As a byproduct of natural gas processing and crude oil refining, LPG, consisting mainly of propane and butane, is a widely available energy source [ 1 ]. Its composition is regulated by standards such as DIN EN 589 in Germany, which mandates a minimum propane content and a motor octane number of at least 89 [ 2 ]. In the USA, a composition of at least 90% propane and a maximum of 5% propylene is common [ 3 ]. Different seasonal blends are used to maintain the fuel's vapor pressure within specified limits, contrary to this summer blends typically contain more butane, as propane shows a higher vapor pressure at the same temperature [ 2 , 4 ]. The thermodynamic properties of LPG are central to its application in engines. While it is gaseous under standard atmospheric conditions due to the low boiling points of propane (–42.1°C) and butane (–0.5°C), an increase in pressure above its vapor pressure allows it to be stored as a liquid. This phase behavior is highly temperature-dependent, as illustrated by the vapor pressure curves for its primary components in → Fig. 1 . A critical challenge arises from the low critical temperature of propane (96.8°C), a threshold above which it can no longer be liquefied by pressure alone. This is particularly relevant in an engine environment, where coolant and oil temperatures commonly range from 80°C to 110°C. If fuel-carrying components exceed this critical temperature, unintended and uncontrolled vaporization can occur, disrupting the fuel delivery process. Compared to conventional fuels, liquid LPG has a lower density (approx. 507–584 kg/m³) than gasoline (720–775 kg/m³) and diesel (820–845 kg/m³) [ 1 ] and a lower volumetric heating value, which results in higher fuel consumption for the same energy input [ 1 ]. However, its air-fuel requirement of approximately 14–15 kg of air per kg of fuel is comparable to that of gasoline and diesel [ 6 ]. Safety-relevant aspects, governed by regulations like DIN EN ISO 11114-1/-2 [ 7 ] and DGUV Rule 110 − 010 [ 8 ], are also significant [ 1 ]. Historically, LPG has been implemented in vehicles primarily through port fuel injection (PFI) systems. However, the direct injection (DI) of liquid LPG into the combustion chamber remains a largely unexplored field by major manufacturers [ 9 ]. The motivation for investigating LPG-DI is twofold: first, direct injection fundamentally allows for a higher thermal efficiency than PFI [ 10 ], and second, LPG can significantly reduce particulate emissions compared to conventional gasoline [ 11 ], which is a known issue in GDI engines. This research project was therefore established to systematically explore the potential of liquid LPG-DI technology in a single-cylinder research engine (bore 82.5 mm, stroke 89.6 mm, compression ratio ε = 10), which can be operated as a naturally aspirated or externally charged unit and features variable valve timing and compression ratio [ 12 ]. Additionally, LPG's favorable carbon-to-hydrogen ratio and high knock resistance provide potential for improved combustion efficiency and reduced pollutant formation—including total hydrocarbons (THC), nitrogen oxides (NOx), and carbon dioxide (CO₂)—across a broad operating range. However, the environmental benefits of LPG under transient and full-load conditions have not been fully quantified, motivating the comprehensive experimental emissions analysis conducted in this work. Scientific background LPG exhibits gas-phase behavior under standard conditions due to the low boiling points of its primary constituents. Its physical state can be controlled via pressure, but only below the respective critical temperatures—an important consideration in engine environments, where fuel system components may reach temperatures near or above these thresholds. The density and air requirement of LPG are comparable to those of gasoline and diesel, supporting its suitability as an alternative engine fuel. The ignition limits of LPG (typically 1.8–9.4 vol%) are narrow and its ignition temperature is similar to gasoline, highlighting both its flammability and engine compatibility. For engine safety and design, factors such as air requirement, ignition temperature, and storage density must be managed carefully [ 6 , 13 ]. A relevant distinction for combustion system design involves the heating value of LPG, which—while providing a higher gravimetric heating value than gasoline—results in an approximately 25% reduction in energy per liter due to lower density. Consequently, engines require higher LPG fuel volume to provide the same energy as gasoline, which impacts fuel consumption and range [ 13 , 14 ]. Regarding safety and system durability, DIN EN ISO 11114 regulates compatibility of materials for gas-carrying components. LPG and its main constituents are broadly compatible with common metallic materials, but some non-metallic materials—such as silicone seals—may experience swelling and failure upon exposure. Corrosion risk, often due to trace water in LPG, requires careful system design and material selection [ 7 ]. The DGUV 110 − 010 guideline further provides safety measures and operational protocols specifically for handling LPG as a fuel [ 15 ]. Most commercially available automotive LPG systems utilize port fuel injection, delivering the fuel either as a gas or liquid into the intake manifold. Direct liquid LPG injection systems, such as the short-lived Prins DLM, are currently unavailable in commercial markets. Consequently, research into durable and efficient direct LPG injection technology remains crucial, given the lack of viable large-scale solutions [ 9 , 16 , 17 ]. The present project is based on a single-cylinder research engine operated with direct LPG injection and a variable compression ratio, allowing flexible examination of thermodynamic effects and system challenges. High-pressure pumps and robust injector control are utilized to enable liquid-phase fuel delivery, with a test regime designed to systematically analyze combustion characteristics and emission trends [ 18 – 20 ]. 2 Methodical approach On a methodological level, a firm literature review was first conducted to compile fundamentals on the thermodynamic properties of LPG as well as standards and guidelines for safe handling. Building on this foundation, several possible fuel system concepts were developed, whose advantages and disadvantages were analyzed to provide a well-founded recommendation for implementation. The stepwise development of a practically feasible fuel system took place in several iterations, whereby problems were investigated and optimized in each case until stable operation in liquid LPG-DI mode was possible. The central objective of the project was to comprehensively investigate the operation of the single-cylinder research engine with liquid LPG direct injection. The main goal was to achieve a detailed understanding of the combustion and emission characteristics of this configuration in order to evaluate the feasibility of transferring the technology to stationary applications, such as combined heat and power (CHP) units. Meeting this objective required overcoming considerable technical challenges, in particular the ability to generate and maintain a fuel pressure of 200 bar, and above all to guarantee that the fuel would remain fully liquid throughout the entire system up to the injector. In addition, the material compatibility and potential wear effects on all fuel-carrying components were carefully considered. To accomplish these aims, the experimental matrix was carefully developed to systematically probe the engine’s behavior across both varying load and speed regimes. As shown in Fig. 1 (see measurement matrix), the test campaign combined a load sweep at a fixed engine speed of 3000 rpm—covering a number of generator-relevant load points—with a complementary speed sweep in which the engine speed was varied over a broad range while the indicated mean effective pressure (IMEP) was held constant at 10 bar ( → Fig. 2 ). This dual-approach allowed for complete mapping of the operating field and ensured that the influence of both engine load and speed on performance and emissions could be thoroughly assessed. The chosen measurement points provided a clear and structured basis for subsequent analysis. A portion of the recorded measured variables was subsequently evaluated. Not all available data were included in this work, since the emphasis was primarily on the setup as well as the fundamental operation of the fuel system, and a complete evaluation of all measured variables would have exceeded the scope. However, the iterative, rather qualitative approach enabled a deep understanding of the specific requirements for liquid LPG direct injection, which ultimately resulted in the presented recommendation of a suitable fuel system. To ensure the liquid state of LPG throughout the system and minimize the risk of phase changes, it is recommended to provide cooling and/or pressure increase already in the external area of the system. For example, a plate heat exchanger can be installed directly behind the gas cylinder, which is connected to the coolant circuit of the test bench. In this way, the temperature of LPG can be kept above 0°C to prevent freezing of any water residues in the cylinder. An additional pre-feed pump supports the pressure increase, so that a sufficient safety margin exists with regard to the vapour pressure of the fuel. The system is completed, as usual, by a high-pressure pump, which brings LPG to the required pressure level before it is injected into the combustion chamber by means of an injector. 3 Experimental setup Research engine The experimental investigation was conducted on a modular single‑cylinder research engine specifically designed for advanced combustion development and the evaluation of alternative fuels such as LPG. The engine is based on a Hydra crankcase with a configurable cylinder barrel and an N38 cylinder head derived from BMW’s NG engine family with VVT2 valve train technology. In this configuration, geometric parameters relevant to combustion—such as piston crown geometry, combustion chamber layout, valve design, and inlet port configuration—can be adapted via interchangeable components to address different research questions in a cost‑ and time‑efficient manner (see Table 1 ). Table 1 Engine specific parameters and auxiliary information relevant to combustion process Engine data Unit Fired operation Number of cylinders [-] 1 Bore [mm] 82 Stroke [mm] 94.6 Displacement [cm 3 ] 499.6 Combustion system / 4-Valve/ Otto Compression ratios [-] 9.5 to 15 dep. on piston Mixture formation / Mixture formation in central position and external mixture formation in the intake manifold Max.power [kW] 50 Rated speed [min -1 ] 4000 Max. speed, temp. [min -1 ] 4500 Valves [-] 4 Max. cyl. Pressure [bar] 150 Mass balance / 1st order The N38 cylinder head and cylinder liner are conceived as a pre‑assembled module that can be mounted as a unit onto the Hydra crankcase, enabling rapid variant changes on the test bench. A height‑adjustable deck system allows continuous setting of the clearance height and therefore variation of the compression ratio without major mechanical modifications. To maximize flexibility, both the intake and exhaust sides of the cylinder head are equipped with VVT2‑based variable valve timing actuators. The intake system follows a modular concept in which each of the two-cylinder head intake ports contains an exchangeable runner that can be configured with different inserts to influence in‑cylinder charge motion (tumble/swirl) or replaced by dummy tubes without internal structures. Depending on the test configuration, the Y‑piece can be mounted directly to the cylinder head for a very short intake path, or additional modules such as a port fuel injection (PFI) rail section or a swirl‑control valve module can be inserted. The valve train can be operated with either fixed or variable valve timing strategies. Variable timing is realized by axial‑piston phasing units that provide VVT (variable valve timing) functionality with a wide adjustment range of up to 80° crank angle, significantly exceeding the range of typical production vane‑type phasers. For operation with fixed valve timing, dummy elements can be installed in place of the VVT units. This freedom in valve train configuration allows different cam profiles and timing strategies to be applied, supporting a broad spectrum of combustion and knock‑limit investigations. Test bench environment The above-mentioned research engine is braked with an AC 25 OF dynamometer from Froude Consine. Engine control is provided by the FI2RE software together with its FI2RE-VCC control unit from IAV, handling not only fuel pump and injector control, but also ignition and variable camshaft timing. Oil and coolant conditioning systems of type STO 1-4-B10/20-TK6/TKN6 (oil) and STL 150/1-6-B10/20-KN6 (cooling water) by Single are used. The fuel cooler's heat exchanger is operated with a Lauda VC 3000 chiller. When required, intake pressurization is handled by a DLR 300 compressor from Elmo Rietschle, with charge air cooled by a Lauda T 4600 device. All major systems—engine, dynamometer, and measurement equipment—are housed in a dedicated test cell, which is separated by a safety door from the control and monitoring workplaces. Both the test cell and control workplaces are equipped with extensive safety features such as an emergency stop, gas detectors, and shut-off valves. For cylinder-pressure indication, a Kistler KiBox To Go type 2893A with appropriate sensors is used, with cooling provided by the Kistler type 2621 water-cooling system. This setup records a wide variety of engine parameters: pressures, temperatures, and state variables such as current crank angle. Exhaust gas analysis uses the AVL AMA i60 R1 system; particulate emissions can alternatively be measured with a TSI 3776 instrument. All set control values are displayed and directly provided by the FI2RE software interface. Challenges of DI operation The main challenge for direct injection operation of the research engine is the physical state of LPG. The LPG must remain in liquid form throughout the system—from the storage bottle to the combustion chamber—to inject it directly as a liquid. The fuel is extracted from the liquid phase inside the bottle using a riser tube, but the bottle pressure is dependent on vapor pressure, which in turn depends on the temperature of the bottle and thus the environment. As the boiling point of LPG mixtures is below 0°C at atmospheric pressure, the fuel is in gaseous state under normal conditions. Therefore, cooling and/or increasing the system pressure is required to keep LPG in the liquid phase and avoid phase changes within the system. Another significant challenge is wear on components of the fuel system, combustion chamber, and exhaust tract when operating with LPG instead of gasoline [ 4 ]. As already discussed with reference to DIN EN ISO 11114, various statements are available concerning the compatibility of LPG or its constituents with different materials. 4 Implementation on the real test bench Building upon the experimental strategy and component selection described in the previous chapters, this section details the practical realization and incremental optimization of the liquid LPG direct injection system as implemented on the real test bench. In the following chapters the structure of the components implemented in the test bench area are explained and its architecture displayed ( → Fig. 3 and Fig. 4 ). To fully understand the structure a legend is displayed in Table 2 . Table 2 Legend for the test bench components Valves/Component/Sensors Symbol Meaning Pumps P1 Pressure rise pump gas bottle + 1bar P2 Mid pressure pump < = 20bar P3 High pressure pump < = 200bar Manual Valves V1 Bottle valve propane V2 Bottle valve nitrogen V3 Valve manual depressurize V4 Valve calibrate Coriolis in V5 Valve calibrate Coriolis out V6 Throttle valve return Electric Valves MV1 Solenoid valve gas bottle MV2 Solenoid valve depressurize system (emergency, normally open) MV3 Solenoid valve gas chromatography MV4 Solenoid valve gas supply HP-pump Static Valves CV1 Check valve CV2 Automatic PR1 Pressure regulator gas chromatography Pressure sensor p1 Pressure in the bottle p2 Pressure before injector p3 Pressure before HP-pump Temperature sensor T1 Temperature gas in bottle T2 Temperature gas pre injector T3 Temperature mid pressure gas T4 Temperature mid pressure gas + return T5 Temperature gas return \(\:\dot{V}\) ̇ Coriolis Coriolis scale Starting Situation: Initial System Layout The fuel system for liquid LPG direct injection was developed iteratively, focusing on identifying and systematically addressing the root causes of emerging issues. Rather than immediately realizing the optimized topology described in Chap. 4.3, the project began with a deliberately simple initial configuration, which served as a diagnostic baseline. In this initial setup, LPG was extracted via a dip tube from a horizontally stored cylinder, positioned in a secure, external gas bottle cabinet. The dip tube guaranteed the withdrawal of LPG from the liquid phase. Following the bottle valve, a check valve prevented reverse flow, while a safety shut-off valve — integrated with the emergency stop system — immediately cut off fuel flow in case of emergency. The system also offered the option of nitrogen flushing to remove residual LPG from lines prior to recommissioning. A Coriolis mass flow meter measured fuel consumption precisely, and a manual valve allowed for gas chromatographic analysis of the LPG blend. After passing through these monitoring units, the fuel was delivered to a high-pressure pump (BMW N53 type), aimed at generating 200 bar. Pressure and temperature sensors monitored the state of the fuel up to the injector. Initial runs revealed pressure collapse during injection, likely due to the formation of vapor bubbles caused by heat input at the high-pressure pump. It became clear that either increasing the fuel pressure upstream of the pump or improved cooling was necessary to suppress outgassing and ensure system stability. Intermediate System Configurations Subsequent development steps involved targeted modifications and additions to progressively stabilize fuel delivery. Initially, only the cylinder pressure of the LPG storage bottle was used to supply the high‑pressure pump, but this proved insufficient to prevent high‑pressure collapse during injection. To increase the fuel pressure upstream of the high‑pressure pump and provide a greater buffer against vaporization, an automotive low‑pressure gasoline feed pump was added, generating an overpressure of up to approximately 5 bar. After roughly one day of operation, leakage occurred on the pressure side of this pump. The same pump type was deliberately re‑installed to enable further measurements on the high‑pressure side, accepting that it would again develop leaks; however, this approach did not provide a sustainable solution and highlighted the need for a more robust design. As a next step, a water injection pump was tested as a more durable feed unit. Despite its higher structural robustness, this pump also developed leakage at the housing cover after approximately one day of operation. The root cause was traced to the sealing system: the original O‑rings, selected for water operation, were not sufficiently compatible with liquid LPG. Due to its low viscosity and high solubility tendency for certain elastomers, LPG can penetrate and swell standard sealing materials that are suitable for aqueous media but not for liquefied hydrocarbons. As a result, micro‑gaps may form, allowing LPG to migrate through the seal and eventually leak. To address this issue, the two sealing O‑rings were replaced with rings specifically declared as LPG‑compatible, and the pump cover was exchanged for a version using threaded connections instead of quick‑release couplings on the inlet and outlet, thereby reducing potential leak paths. In the final configuration of this stage, this modified unit corresponded to pump P2 in Fig. 4 .​ In parallel to the feed‑pump adaptations, further thermal and hydraulic measures were implemented. An intermediate fuel cooler was installed between the feed pump and the high‑pressure pump, reducing LPG temperature to approximately 10°C. This improved start‑up stability but did not fully eliminate vapor formation and pressure loss during injection. To further mitigate outgassing, a gas separator (“bubble trap”) was integrated; its design allowed free gas to be vented while the high‑pressure pump was supplied from the lower, liquid phase. This configuration enabled stable high‑pressure injection for approximately 30 seconds, but longer‑term operation still resulted in pressure collapse. Replacing the original high‑pressure pump with a modern Bosch HDP 5 unit improved performance and extended stable operation to about one to two minutes before pressure decay recurred, now primarily attributed to heat‑induced gas bubble formation within the pump itself. Subsequent system optimization therefore included a more powerful low‑pressure pump, raising line pressure to approximately 11–12 bar, and the introduction of a return line to recirculate and cool excess fuel. The return circuit allowed hot fuel in the high‑pressure path to be directed back to the low‑pressure side, where it was cooled and re‑pressurized. Additional temperature monitoring points throughout the system improved diagnostic capability and helped localize persistent weaknesses.Despite these targeted improvements, the system remained sensitive to thermal inputs, with occasional pressure spikes and damage to high‑pressure lines revealing the complexity of managing phase behavior in such high‑pressure LPG systems. 4.1.1 Material compatibility of LPG with conventional pump systems The progressive failure of several low‑pressure feed pumps in the development phase effectively created an incidental material compatibility study for LPG in conventional pump systems. The initial approach—using a standard gasoline feed pump—demonstrated that components and seals designed for liquid hydrocarbons at moderate pressures and temperatures may not withstand continuous operation with cold, high‑pressure LPG without modification. Leakage on the pressure side after short operating times indicated that standard elastomer seals and quick‑coupling interfaces were not sufficiently tight or chemically robust for liquefied gas operation.​ The root cause lies in the chemical properties of LPG itself. Propane and butane, the principal constituents of LPG, are low‑molecular‑weight aliphatic hydrocarbons with exceptionally high solubility in many standard elastomeric materials, particularly those based on nitrile rubber (NBR) or silicone compounds. Unlike conventional liquid fuels, which exhibit more moderate interaction with elastomers, LPG's low viscosity, high vapor pressure, and molecular penetrating power allow the fuel to diffuse through and swell standard sealing compounds over time. This process causes dimensional changes, material degradation, and the formation of microscopic gaps—micro‑gaps—that serve as preferential leak paths. Additionally, the high solubility of LPG in certain elastomers can cause the sealing material to become overly soft and lose its ability to maintain contact pressure, further compromising the seal integrity.​ Standard O‑rings selected for aqueous media (as in water injection pump applications) offer even less resistance to LPG than those designed for conventional fuels, since they are formulated to be compatible with polar solvents rather than non‑polar hydrocarbons. The subsequent use of this water‑rated pump confirmed that even structurally robust pump housings do not guarantee LPG compatibility if the sealing concept is not adapted. Only after replacing the original O‑rings with LPG‑rated sealing elements—typically based on fluorocarbon (FKM/Viton®) or fluorosilicone (FVMQ) compounds, which exhibit significantly lower swelling and superior chemical resistance to low‑molecular‑weight hydrocarbons [ 21 , 22 ]—and installing a revised cover with threaded connections instead of quick‑release fittings, could acceptable tightness and operational stability be achieved for pump P2 (Fig. 4 ).​ To ensure reliable and reproducible fuel mass flow measurements with the Coriolis flow meter, an additional pre‑feed stage (pump P1 in Fig. 4 ) was implemented upstream of the measurement section. For the final measurement campaign, a conventional automotive fuel pump was used at a deliberately reduced pressure level to delay the onset of unavoidable long‑term leakage while still providing a sufficiently stable inlet pressure for the Coriolis sensor. This configuration ensured robust fuel‑flow measurements for the conclusive test series and provided a stable hydraulic basis for the final system layout described in the following section. Final concept: stable LPG Operation A stable and robust system configuration was finally achieved by implementing several coordinated measures. The LPG bottle was moved into the test cell, with select lines fitted with visual flow indicators. The return line was actively cooled before rejoining the low-pressure circuit, and a small feed pump was introduced immediately after the storage bottle to provide an initial pressure increase (approx. 0.8 bar). A pressure control valve ensured that the low-pressure circuit did not exceed 12 bar, thereby preventing overpressure and minimizing the risk of vapor bubble nucleation. Through this combination of modest pre-pressure increase, targeted cooling, a gas separator, and recirculation, the fuel system could now reliably maintain 200 bar injection pressure, even during sustained engine operation. As a result, long-term, stable test bench operation was achieved, allowing the planned series of liquid LPG direct injection measurements to be conducted successfully. 5 Measurement procedure The experimental investigation systematically compared liquid LPG direct injection performance against conventional gasoline baselines using two distinct camshaft phasing strategies. Baseline measurements were established using standard E10 gasoline (E10-1 through E10-16) across speed sweep conditions from 1000 to 3800 rpm at constant 10 bar IMEP (E10-1 to E10-7), and load sweep conditions at constant 3000 rpm with IMEP progressively increased from 2 to 18 bar (E10-8 to E10-16). To evaluate the influence of fuel octane rating on knock behavior, high-octane reference measurements were conducted using Aral Ultimate 102 (E10-17 through E10-19) at full-load conditions (16, 18, and 20 bar IMEP at 3000 rpm). Comparative analysis revealed that despite its elevated research octane number of 102, Ultimate 102 exhibited knock behavior nearly identical to standard E10 gasoline (RON 95), showing no significant knock resistance improvement at full load. Consequently, subsequent LPG performance comparisons focus on E10 as the primary baseline, as this more clearly demonstrates LPG's practical advantage through its superior knock resistance (RON 103–111 depending on propane/butane ratio), which enables advanced ignition timing and improved thermal efficiency. The LPG measurement series (LPG-20 through LPG-51) systematically investigated the influence of camshaft phasing on combustion efficiency and knock behavior through two distinct valve timing configurations. The first configuration, designated LPG-SCT (Standard Cam Timing), employed intake valve event spreads of 125° camshaft angle (CA) for IMEP ≤ 10 bar and 120° CA for IMEP > 10 bar, with exhaust spreads of 105° CA for IMEP ≤ 10 bar and 75° CA for IMEP > 10 bar. The second configuration, LPG-RCT (Reduced Cam Timing), utilized symmetrical 80° CA spreads for both intake and exhaust valves across all load conditions. The speed sweep encompassed LPG-20 through LPG-31, with measurements alternating between LPG-SCT and LPG-RCT configurations across the 1000–3800 rpm range at 10 bar IMEP. The load sweep extended from LPG-32 through LPG-51, maintaining the same alternating cam timing pattern with IMEP varied from 2 to 20 bar at constant 3000 rpm engine speed. Operating point LPG-35 was excluded from analysis due to deviations from target parameters and was replaced by LPG-35.1. Operating point LPG-45 was not conducted, as stable operation at 2 bar IMEP could not be reliably achieved under prevailing test conditions. This measurement structure enabled direct quantification of LPG direct injection performance relative to E10 baseline while systematically isolating the influence of valve timing strategy on combustion phasing, efficiency, and knock margin across the full operating envelope. Load sweep results This section presents the results of the load sweep (E10-8 to E10-16, LPG-32 to LPG-51), comparing E10 baseline, LPG-SCT, and LPG-RCT (80/80° camshaft angle configuration) across IMEP values from 2 to 20 bar at constant 3000 rpm engine speed. → Fig. 5 displays eight sub-plots of key performance parameters. Exhaust gas temperature increases progressively with load across all three fuels. LPG-RCT generally exhibits slightly lower values in the mid-load range compared to LPG-SCT and E10, suggesting that the 80/80° valve timing configuration promotes more efficient combustion with reduced thermal burden. At the highest load points, exhaust temperatures converge across all fuels. Injection duration for LPG is substantially longer than for E10, with LPG-SCT showing the most pronounced difference. This corresponds to fuel consumption penalties that decrease from approximately 25% at low loads to below 15% at high loads. LPG-RCT demonstrates more favorable consumption characteristics than LPG-SCT. The 50% mass fraction burned (MFB) reveals critical differences in combustion control. E10 begins to deviate significantly from the desired 8° after top dead center (ATDC) above approximately 12 bar IMEP due to knock-limited ignition timing retard, while LPG maintains proximity to this ideal value up to 16–18 bar IMEP. Knock frequency analysis shows that E10 rapidly enters regimes of elevated knock activity with increasing load, whereas LPG permits significantly earlier ignition timing without proportional knock increase. Ignition timing plots confirm this trend, with LPG consistently utilizing earlier spark advance than E10 across the load range, particularly in the LPG-RCT configuration. Cycle-to-cycle variability of IMEP demonstrates that LPG exhibits superior stability compared to E10 in the low to mid-load range; however, this advantage partially reverses at very high loads. LPG-RCT shows marginally reduced stability at extreme loads, likely attributable to altered in-cylinder flow patterns under these conditions. Brake thermal efficiency reveals that LPG achieves comparable or superior efficiency to E10 at low loads (below approximately 6 bar IMEP), but efficiency declines in the 8–16 bar range before recovering at the highest load points. LPG-RCT demonstrates modest efficiency gains at high loads relative to LPG-SCT. These results indicate that LPG derives clear benefit from its knock resistance at high load, where earlier ignition timing can be utilized, whereas optimization of mixture formation remains necessary in the mid-load regime. Speed sweep results The speed sweep investigation varied engine speed from 1000 to 3500 rpm in 500 rpm increments, with one additional measurement point at 3800 rpm, all at nominally 10 bar IMEP (E10-1 to E10-7, LPG-20 to LPG-31). → Fig. 6 presents four performance parameters across the speed range. Note that LPG-RCT could not be stably operated above 3000 rpm without supercharging; consequently, complete comparative data at higher speeds is unavailable for this configuration. Cycle-to-cycle variability demonstrates that LPG delivers more consistent combustion than E10 across the lower speed range; this stability advantage diminishes at elevated speeds. The combustion center of gravity remains controllable for LPG across a wide speed range, maintaining proximity to the target 8° ATDC, whereas E10 requires repeated ignition timing retard as knock propensity increases. Ignition timing shows the consistent trend observed in the load sweep: LPG permits substantially earlier spark advance throughout, with LPG-RCT achieving the most advanced timing at lower speeds. Brake thermal efficiency exhibits a notable difference from the load sweep: LPG efficiency generally trails E10 across the speed sweep, particularly at elevated engine speeds. This observation suggests that LPG does not derive equivalent efficiency benefit from its knock resistance when operating at constant load across varying speeds. The LPG-RCT configuration offers no pronounced efficiency advantage in this speed-varied regime, indicating that the valve timing optimization benefits observed at high load do not translate uniformly across the speed spectrum at moderate load. Emission characteristics Emission measurements were conducted exclusively during the load sweep (E10-8 to E10-16, LPG-32 to LPG-51; 2 to 20 bar IMEP at 3000 rpm) to characterize transient combustion behavior across the engine's load envelope. → Fig. 7 presents six emission parameters. Note that E10 baseline data at 14 bar IMEP are interpolated from adjacent measurements; these values should be interpreted with appropriate caution. Incomplete Combustion Products Total hydrocarbon (THC) emissions from LPG are occasionally elevated relative to E10 in the very low-load range (below 6 bar IMEP), particularly with the LPG-RCT configuration, but decline substantially above 6 bar IMEP to values often below E10 levels. Combined unburned hydrocarbons and partial oxidation products (CO + H) show occasional elevation in LPG-RCT operation, particularly in certain mid-load points, suggesting incomplete combustion when the interaction of ignition timing, charge motion, and thermal conditions does not ensure homogeneous oxidation throughout the combustion chamber. Thermal NOx Formation and Temperature Effects Nitrogen oxide (NOx) emissions rise monotonically with load for all fuels, reflecting elevated pressure and temperature conditions. LPG-RCT frequently exhibits higher NOx peaks in the mid-load range relative to LPG-SCT, attributable to its earlier ignition timing and associated local temperature elevation; conversely, E10 produces comparatively lower NOx at the highest loads due to knock-mandated ignition timing retard, which reduces peak combustion temperatures. Exhaust gas temperature increases with load for all fuels, with E10 generally achieving the highest temperatures at maximum load. Both LPG-SCT and particularly LPG-RCT operate at lower exhaust temperatures, likely attributable to earlier ignition timing that shifts the bulk heat release event toward the expansion stroke, reducing exhaust gas enthalpy despite potentially higher in-cylinder peak temperatures. Carbon Balance and Particulate Emissions Carbon dioxide (CO₂) concentrations are generally lower for LPG than E10, reflecting the favorable carbon-to-hydrogen ratio of propane and butane. However, this advantage diminishes at high loads where LPG's lower volumetric energy density necessitates higher fuel mass flow. Residual oxygen (O₂) measurements reveal that E10 exhibits lower exhaust oxygen content at very high loads, consistent with extended post-flame oxidation when late ignition timing is imposed. LPG-RCT occasionally shows elevated O₂ levels correlated with higher CO + H concentrations. Particulate matter emissions demonstrate markedly lower soot formation tendency for LPG across the load range compared to E10, despite occasional excursions at very low loads. This advantage aligns with LPG's chemical composition and the improved combustion control enabled by early ignition timing. Summary of Emission Trade-offs Overall, LPG demonstrates principal advantages at high load through superior knock resistance enabling advanced ignition timing, a favorable carbon-to-hydrogen stoichiometry, and dramatically reduced particulate formation. The distinction between LPG-SCT and LPG-RCT valve timing configurations illustrates how camshaft phasing substantially influences NOx and oxidation product formation: earlier spark advance elevates peak in-cylinder temperatures but can modestly reduce exhaust temperature when primary heat release is shifted to the expansion phase. These emission profiles underscore that optimal design of a liquid LPG direct injection system requires careful balancing of efficiency, emission control, and combustion stability across the full operating envelope. 6 Discussion and auxiliary investigations Throughout the testing campaigns, additional investigations were performed to deepen understanding of the fuel system's behavior under varying conditions and to identify optimization potential. For instance, systematic tests of the return cooler examined how altering the temperature of the return line affected system stability. Lowering coolant temperature by 5 K yielded no adverse effects; reduced temperature at a given pressure expectedly minimized outgassing in the high-pressure circuit. In contrast, gradually increasing the temperature up to ambient unveiled a distinct shift: As the return temperature reached room temperature, the high-pressure pump demand rose continuously to maintain the 200 bar threshold. Once maximum actuation was reached, the fuel pressure dropped, the injector could no longer deliver sufficient fuel, and engine operation had to be terminated manually. This highlights that even moderate temperature rises in the return circuit can trigger phase changes and compromise system stability, especially when components are tuned to operate within tight pressure and temperature windows. Another trial evaluated whether the final fuel system design (Version 7) could operate reliably using a Bosch HDP 5 high-pressure pump instead of the prior Audi pump. Because the Bosch pump design does not support a return line, the recirculation loop had to be omitted. In this configuration, the system failed to maintain a stable fuel pressure, underlining the importance of the Audi pump's return system for operational reliability. With recirculation, unused fuel is consistently returned, cooled, and pressurized in the low-pressure circuit, reducing the risk of vaporization in the high-pressure path. Fuel consumption measurements confirmed fundamental expectations: At comparable volumetric flow rates, LPG exhibits lower energy density than gasoline. Depending on load point, observed consumption increases ranged from 25% to 15% with the older valve timing setup, and about 10% with the 80/80 camshaft spread—values consistent with published vehicle trials reporting deviations between 10% and 30%. Knock behavior showed clear LPG advantages: A higher IMEP range could be achieved before knocking combustion occurred. Whereas gasoline required ignition timing retard above approximately 12 bar IMEP to control knock, LPG maintained an ideal combustion centroid of around 8° crank angle after top dead center up to more than 14 bar. Even when combustion phasing had to be shifted, LPG's knock frequency remained lower, allowing consistently earlier ignition timing. These findings underline that liquid LPG direct injection systems offer clear performance benefits, but require stringent demands on fuel system robustness. A minimum pre-pump pressure of 11–12 bar at approximately 20°C is essential to avoid outgassing. The return circuit is critical for recycling and cooling excess fuel, stabilizing the high-pressure region. Despite the increased complexity, the project succeeded in producing stable single-cylinder engine operation on liquid LPG. The results lay a foundation for future system enhancements, including more robust components, optimized return cooling, or further camshaft timing adjustments. Overall, the achieved project status enabled a variety of measurements that reveal both considerable advantages and distinct limitations of LPG in comparison to gasoline. Supplementary tests demonstrate how operational influences such as temperature, pressure, and pump choice impact system reliability, confirming the need for extended durability and materials investigations to assess long-term viability of pumps and sealing components. Declarations Author Contribution A.D. designed and conducted the experiments, performed the data analysis, and wrote the main manuscript text. H.R. supervised the research, guided the scientific concept and methodology, and contributed to the interpretation of the results and the revision of the manuscript. B.R. supported the experimental planning, data evaluation, and technical refinement of the test setup, and contributed to manuscript text. The experimental work on the test bench, including engine modification, fuel system integration, and exhaust gas measurement, was enabled by the technical staff, whose expertise and support were essential for the successful completion of this study. All authors reviewed and approved the final manuscript. References Adolf, J., Balzer, C., Joedicke, A., Schabla, U.: SHELL Flüssiggas-Studie: LPG als Energieträger und Kraftstoff, trends und Perspektiven DIN: EN 589:2022-04, Kraftstoffe_- Flüssiggas_- Anforderungen und Prüfverfahren; Deutsche Fassung EN_589:2018 + A1:2022, Beuth Verlag GmbH Propane Grades and Quality - HD5 HD10 and Commercial Propane: (2025). https://www.propane101.com/propanegradesandquality.htm , January 29 Günther, M.: Comparison of the thermodynamic potential, the octane requirement and the contamination tendency of three LGP concepts in a turbocharged direct injection gasoline SI engine: LPG-DI, LPG-PFI (liquid) and LPG-PFI (gaseous): LPG System Comparison, FVV Project 1069 Clesse, U.K., Solutions for, LPG: Technical Advice - Clesse UK Solutions for LPG, (2025). https://clesse.co.uk/technical-advice/ , November 21 TEGA – Technische Gase und Gasetechnik GmbH: Physikalische Daten und brenntechnische Eigenschaften, https://www.tega.de/fileadmin/Downloads_und_Bilder/FG_Tank/Info3_Physikalische_Daten_und_brenntechnische_Eigenschaften_1.pdf DIN EN ISO 11114-: 1:2020-08, Gasflaschen_- Verträglichkeit von Werkstoffen für Gasflaschen und Ventile mit den in Berührung kommenden Gasen_- Teil_1: Metallische Werkstoffe (ISO_11114-1:2020); Deutsche Fassung EN_ISO_11114-1:2020, Beuth Verlag GmbH Nagel, D., Krieck, M., System Comparison, L.P.G., II, Project No:. 1151, Investigation of the Effect of supercritical LPG on DGUV Regel 110 – 010 Verwendung von Flüssiggas: https://www.bghm.de/fileadmin/user_upload/Arbeitsschuetzer/Gesetze_Vorschriften/Regeln/110-010.pdf Systeme | Prins Autogassystemen BV: (2025). https://www.prinsautogas.com/de/systeme , January 29 Basshuysen, R., Schäfer, F.: Handbuch Verbrennungsmotor: Grundlagen, Komponenten, Systeme, Perspektiven. Springer Vieweg (2014) Seungmook, O.: Combustion and emission characteristics in a direct injection LPG/gasoline spark ignition engine, SAE Paper 2010-01-1461 , (2010) Schrick, G.H.: Nockenwellen-Einbau: Camshaft Installation: https:// (2025). schrick.com/media/d1/7c/3f/1647859651/nockenwelleneinbau__camshaft_installation-berichtigt-25102005.pdf , Zugriff: Jan 29 Jörg Adolf, C., Balzer, B., Gnörich, A., et al.: Joedicke Shell Flüssiggas-Studie: LPG als Energieträger und Kraftstoff Fakten, Trends und Perspektiven,., (2015). 10.13140/RG.2.2.20645.63208 Bender, B., Göhlich, D. (eds.): Dubbel Taschenbuch für den Maschinenbau 2: Anwendungen, 26th ed., Springer Berlin Heidelberg, Berlin, Heidelberg, ISBN 9783662597132, (2020) Unfallversicherung, D.G.: DGUV Regel 110 – 010 „Verwendung von Flüssiggas BRC: Produkte, (2025). https://brc.it/de/produkte/ , November 19 Renzo, L.: LPG | Landi Renzo, (2025). https://landirenzo.com/en/lpg-vehicle-systems , November 19 van Basshuysen, R., Schäfer, F.: Handbuch Verbrennungsmotor, Springer Fachmedien Wiesbaden, Wiesbaden, ISBN 978-3-658-04677-4, (2015) Barrett, S.: European Expert Group reports on future transport fuels. Fuel Cells Bull. 2011 (2), 12–16 (2011). 10.1016/S1464-2859(11)70096-7 Seungmook Oh, S., Lee, Y., Choi, K.-Y., Kang, et al.: Combustion and Emission Characteristics in a Direct Injection LPG/Gasoline Spark Ignition Engine, International Powertrains, Fuels & Lubricants Meeting, SAE International, ISBN 2688-3627, (2010) LIQUEFIED PETROLEUM GAS Resistant Orings: Xrings, Gaskets, Rubber products - Marco Rubber, (2025). https://www.marcorubber.com/chemical-compatibility/LIQUEFIED%20PETROLEUM%20GAS , December 16 O-Ring Compatibility Chemical Resistance Chart: (2025). https://www.allorings.com/o-ring-compatibility , December 16 Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Revision Version 1 posted Editorial decision: Revision requested 07 May, 2026 Reviews received at journal 05 May, 2026 Reviewers agreed at journal 05 May, 2026 Reviewers agreed at journal 18 Mar, 2026 Reviews received at journal 21 Feb, 2026 Reviewers agreed at journal 20 Feb, 2026 Reviewers invited by journal 06 Feb, 2026 Editor assigned by journal 22 Jan, 2026 Submission checks completed at journal 21 Jan, 2026 First submitted to journal 19 Jan, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8640603","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":586882271,"identity":"770299e7-a2ca-40a6-8478-dcd83be694a9","order_by":0,"name":"Aristidis Dafis","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABBklEQVRIie3QMWsCMRTA8VcCd0vANdBivsIToQgW+1WuFLxFQSgcTseB4C2Frkq/REu/wINAXR43FkelXyBjB6GNHhWXoN2K5L8kBH4kLwCh0D+NAIV0K8IYoVmfRUcI/RJGaJ9E3DUCagJwVxwjjefF2thR90oX4tXSKE+fVPqhILvxElUN0D0slUjRgyI0w/lskCmo+v5rWG5nMRJB4qVFGr4w99XF1HiF5ni1I7qQ7S/CPMWafHsJMuCOAMlr9zCR4KJ8d4S8pMVuCt7OYqKs42xrXj6KTlLde0mT40873nRvdTl5W9Im1w0Rr5c26/nH3yf2O/cVyQngsHj1RxAKhUJn3g/bYFch4T/bjQAAAABJRU5ErkJggg==","orcid":"","institution":"Otto-von-Guericke University Magdeburg","correspondingAuthor":true,"prefix":"","firstName":"Aristidis","middleName":"","lastName":"Dafis","suffix":""},{"id":586882272,"identity":"e98fcdc4-719b-4fb0-9dae-29883c40cdf2","order_by":1,"name":"Hermann Rottengruber","email":"","orcid":"","institution":"Otto-von-Guericke University Magdeburg","correspondingAuthor":false,"prefix":"","firstName":"Hermann","middleName":"","lastName":"Rottengruber","suffix":""},{"id":586882273,"identity":"ac3b3a59-624a-4d37-82bd-16a59fe04803","order_by":2,"name":"Bastian Rabe","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Bastian","middleName":"","lastName":"Rabe","suffix":""}],"badges":[],"createdAt":"2026-01-19 14:54:43","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8640603/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8640603/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":104397132,"identity":"54d11b0d-066c-4e85-81b9-99e43d37d86f","added_by":"auto","created_at":"2026-03-11 11:34:54","extension":"jpeg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":60529,"visible":true,"origin":"","legend":"\u003cp\u003eVapour pressure curves of propane and butane [5]\u003c/p\u003e","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8640603/v1/fc8ba340ebfa5e985f13fc59.jpeg"},{"id":102413730,"identity":"04e771f8-bcc8-4fa1-b7c2-8a604c4e8d6a","added_by":"auto","created_at":"2026-02-11 12:30:35","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":1645978,"visible":true,"origin":"","legend":"\u003cp\u003eMeasurement matrix\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8640603/v1/6cecadbd79659bcca8522ed7.png"},{"id":102413729,"identity":"2c689b74-949f-4502-86be-39fdca2c53cf","added_by":"auto","created_at":"2026-02-11 12:30:35","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":103857,"visible":true,"origin":"","legend":"\u003cp\u003eInitial test bench architecture\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-8640603/v1/1c7a0ed0f31296a7f329455c.png"},{"id":102413733,"identity":"4720dfb4-ee38-40d9-820a-63cac18a8f03","added_by":"auto","created_at":"2026-02-11 12:30:35","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":139172,"visible":true,"origin":"","legend":"\u003cp\u003eFinal test bench architecture\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-8640603/v1/adce2a2d5db87bad9a9a7f0c.png"},{"id":102413731,"identity":"ec66f466-abf4-4b17-9c75-a9929a6ed1c9","added_by":"auto","created_at":"2026-02-11 12:30:35","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":253586,"visible":true,"origin":"","legend":"\u003cp\u003eResults load sweep\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-8640603/v1/bd3e98e16bffec96e33ebc39.png"},{"id":103049219,"identity":"3013b972-5609-44f5-9ed1-97966006a421","added_by":"auto","created_at":"2026-02-20 07:38:34","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":177626,"visible":true,"origin":"","legend":"\u003cp\u003eResults speed sweep\u003c/p\u003e","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-8640603/v1/530e3f8d15684e63b84e6c3b.png"},{"id":102745492,"identity":"cb17b24f-88cf-4156-ad6f-3e9baec1cad8","added_by":"auto","created_at":"2026-02-16 08:51:11","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":259720,"visible":true,"origin":"","legend":"\u003cp\u003eEmission results load sweep\u003c/p\u003e","description":"","filename":"floatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-8640603/v1/2928b5b7a11c241447621684.png"},{"id":104779707,"identity":"ef007596-45f6-4c73-83e6-e9c41ad48007","added_by":"auto","created_at":"2026-03-17 07:45:00","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3240072,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8640603/v1/1ba0fc6d-070b-45d9-9fdd-b202dc0148f3.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Investigation of Combustion and Performance Limits of Liquid LPG Direct Injection in a Single-Cylinder Research Engine","fulltext":[{"header":"1 Introduction","content":"\u003cp\u003e \u003cem\u003eMotivation\u003c/em\u003e \u003c/p\u003e \u003cp\u003eThe pursuit of cleaner and more efficient combustion processes has led to increased interest in alternative fuels for internal combustion engines. Among these, Liquefied Petroleum Gas (LPG) presents a compelling option. As a byproduct of natural gas processing and crude oil refining, LPG, consisting mainly of propane and butane, is a widely available energy source [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. Its composition is regulated by standards such as DIN EN 589 in Germany, which mandates a minimum propane content and a motor octane number of at least 89 [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. In the USA, a composition of at least 90% propane and a maximum of 5% propylene is common [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. Different seasonal blends are used to maintain the fuel's vapor pressure within specified limits, contrary to this summer blends typically contain more butane, as propane shows a higher vapor pressure at the same temperature [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eThe thermodynamic properties of LPG are central to its application in engines. While it is gaseous under standard atmospheric conditions due to the low boiling points of propane (\u0026ndash;42.1\u0026deg;C) and butane (\u0026ndash;0.5\u0026deg;C), an increase in pressure above its vapor pressure allows it to be stored as a liquid. This phase behavior is highly temperature-dependent, as illustrated by the vapor pressure curves for its primary components in \u0026rarr; Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eA critical challenge arises from the low critical temperature of propane (96.8\u0026deg;C), a threshold above which it can no longer be liquefied by pressure alone. This is particularly relevant in an engine environment, where coolant and oil temperatures commonly range from 80\u0026deg;C to 110\u0026deg;C. If fuel-carrying components exceed this critical temperature, unintended and uncontrolled vaporization can occur, disrupting the fuel delivery process. Compared to conventional fuels, liquid LPG has a lower density (approx. 507\u0026ndash;584 kg/m\u0026sup3;) than gasoline (720\u0026ndash;775 kg/m\u0026sup3;) and diesel (820\u0026ndash;845 kg/m\u0026sup3;) [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e] and a lower volumetric heating value, which results in higher fuel consumption for the same energy input [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. However, its air-fuel requirement of approximately 14\u0026ndash;15 kg of air per kg of fuel is comparable to that of gasoline and diesel [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. Safety-relevant aspects, governed by regulations like DIN EN ISO 11114-1/-2 [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e] and DGUV Rule 110\u0026thinsp;\u0026minus;\u0026thinsp;010 [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e], are also significant [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eHistorically, LPG has been implemented in vehicles primarily through port fuel injection (PFI) systems. However, the direct injection (DI) of liquid LPG into the combustion chamber remains a largely unexplored field by major manufacturers [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. The motivation for investigating LPG-DI is twofold: first, direct injection fundamentally allows for a higher thermal efficiency than PFI [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e], and second, LPG can significantly reduce particulate emissions compared to conventional gasoline [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e], which is a known issue in GDI engines. This research project was therefore established to systematically explore the potential of liquid LPG-DI technology in a single-cylinder research engine (bore 82.5 mm, stroke 89.6 mm, compression ratio ε\u0026thinsp;=\u0026thinsp;10), which can be operated as a naturally aspirated or externally charged unit and features variable valve timing and compression ratio [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eAdditionally, LPG's favorable carbon-to-hydrogen ratio and high knock resistance provide potential for improved combustion efficiency and reduced pollutant formation\u0026mdash;including total hydrocarbons (THC), nitrogen oxides (NOx), and carbon dioxide (CO₂)\u0026mdash;across a broad operating range. However, the environmental benefits of LPG under transient and full-load conditions have not been fully quantified, motivating the comprehensive experimental emissions analysis conducted in this work.\u003c/p\u003e \u003cp\u003e \u003cem\u003eScientific background\u003c/em\u003e \u003c/p\u003e \u003cp\u003eLPG exhibits gas-phase behavior under standard conditions due to the low boiling points of its primary constituents. Its physical state can be controlled via pressure, but only below the respective critical temperatures\u0026mdash;an important consideration in engine environments, where fuel system components may reach temperatures near or above these thresholds. The density and air requirement of LPG are comparable to those of gasoline and diesel, supporting its suitability as an alternative engine fuel. The ignition limits of LPG (typically 1.8\u0026ndash;9.4 vol%) are narrow and its ignition temperature is similar to gasoline, highlighting both its flammability and engine compatibility. For engine safety and design, factors such as air requirement, ignition temperature, and storage density must be managed carefully [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eA relevant distinction for combustion system design involves the heating value of LPG, which\u0026mdash;while providing a higher gravimetric heating value than gasoline\u0026mdash;results in an approximately 25% reduction in energy per liter due to lower density. Consequently, engines require higher LPG fuel volume to provide the same energy as gasoline, which impacts fuel consumption and range [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eRegarding safety and system durability, DIN EN ISO 11114 regulates compatibility of materials for gas-carrying components. LPG and its main constituents are broadly compatible with common metallic materials, but some non-metallic materials\u0026mdash;such as silicone seals\u0026mdash;may experience swelling and failure upon exposure. Corrosion risk, often due to trace water in LPG, requires careful system design and material selection [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. The DGUV 110\u0026thinsp;\u0026minus;\u0026thinsp;010 guideline further provides safety measures and operational protocols specifically for handling LPG as a fuel [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eMost commercially available automotive LPG systems utilize port fuel injection, delivering the fuel either as a gas or liquid into the intake manifold. Direct liquid LPG injection systems, such as the short-lived Prins DLM, are currently unavailable in commercial markets. Consequently, research into durable and efficient direct LPG injection technology remains crucial, given the lack of viable large-scale solutions [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eThe present project is based on a single-cylinder research engine operated with direct LPG injection and a variable compression ratio, allowing flexible examination of thermodynamic effects and system challenges. High-pressure pumps and robust injector control are utilized to enable liquid-phase fuel delivery, with a test regime designed to systematically analyze combustion characteristics and emission trends [\u003cspan additionalcitationids=\"CR19\" citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e].\u003c/p\u003e"},{"header":"2 Methodical approach","content":"\u003cp\u003eOn a methodological level, a firm literature review was first conducted to compile fundamentals on the thermodynamic properties of LPG as well as standards and guidelines for safe handling. Building on this foundation, several possible fuel system concepts were developed, whose advantages and disadvantages were analyzed to provide a well-founded recommendation for implementation. The stepwise development of a practically feasible fuel system took place in several iterations, whereby problems were investigated and optimized in each case until stable operation in liquid LPG-DI mode was possible.\u003c/p\u003e \u003cp\u003eThe central objective of the project was to comprehensively investigate the operation of the single-cylinder research engine with liquid LPG direct injection. The main goal was to achieve a detailed understanding of the combustion and emission characteristics of this configuration in order to evaluate the feasibility of transferring the technology to stationary applications, such as combined heat and power (CHP) units. Meeting this objective required overcoming considerable technical challenges, in particular the ability to generate and maintain a fuel pressure of 200 bar, and above all to guarantee that the fuel would remain fully liquid throughout the entire system up to the injector. In addition, the material compatibility and potential wear effects on all fuel-carrying components were carefully considered.\u003c/p\u003e \u003cp\u003eTo accomplish these aims, the experimental matrix was carefully developed to systematically probe the engine\u0026rsquo;s behavior across both varying load and speed regimes. As shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e (see measurement matrix), the test campaign combined a load sweep at a fixed engine speed of 3000 rpm\u0026mdash;covering a number of generator-relevant load points\u0026mdash;with a complementary speed sweep in which the engine speed was varied over a broad range while the indicated mean effective pressure (IMEP) was held constant at 10 bar ( \u0026rarr; Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). This dual-approach allowed for complete mapping of the operating field and ensured that the influence of both engine load and speed on performance and emissions could be thoroughly assessed. The chosen measurement points provided a clear and structured basis for subsequent analysis.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eA portion of the recorded measured variables was subsequently evaluated. Not all available data were included in this work, since the emphasis was primarily on the setup as well as the fundamental operation of the fuel system, and a complete evaluation of all measured variables would have exceeded the scope. However, the iterative, rather qualitative approach enabled a deep understanding of the specific requirements for liquid LPG direct injection, which ultimately resulted in the presented recommendation of a suitable fuel system.\u003c/p\u003e \u003cp\u003eTo ensure the liquid state of LPG throughout the system and minimize the risk of phase changes, it is recommended to provide cooling and/or pressure increase already in the external area of the system. For example, a plate heat exchanger can be installed directly behind the gas cylinder, which is connected to the coolant circuit of the test bench. In this way, the temperature of LPG can be kept above 0\u0026deg;C to prevent freezing of any water residues in the cylinder. An additional pre-feed pump supports the pressure increase, so that a sufficient safety margin exists with regard to the vapour pressure of the fuel. The system is completed, as usual, by a high-pressure pump, which brings LPG to the required pressure level before it is injected into the combustion chamber by means of an injector.\u003c/p\u003e"},{"header":"3 Experimental setup","content":"\u003cp\u003e \u003cem\u003eResearch engine\u003c/em\u003e \u003c/p\u003e \u003cp\u003eThe experimental investigation was conducted on a modular single‑cylinder research engine specifically designed for advanced combustion development and the evaluation of alternative fuels such as LPG. The engine is based on a Hydra crankcase with a configurable cylinder barrel and an N38 cylinder head derived from BMW\u0026rsquo;s NG engine family with VVT2 valve train technology. In this configuration, geometric parameters relevant to combustion\u0026mdash;such as piston crown geometry, combustion chamber layout, valve design, and inlet port configuration\u0026mdash;can be adapted via interchangeable components to address different research questions in a cost‑ and time‑efficient manner (see Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eEngine specific parameters and auxiliary information relevant to combustion process\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"3\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEngine data\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eUnit\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eFired operation\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNumber of cylinders\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e[-]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBore\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e[mm]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e82\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eStroke\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e[mm]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e94.6\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDisplacement\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e[cm\u003csup\u003e3\u003c/sup\u003e]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e499.6\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCombustion system\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4-Valve/ Otto\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCompression ratios\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e[-]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e9.5 to 15 dep. on piston\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMixture formation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMixture formation in central position and external mixture formation in the intake manifold\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMax.power\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e[kW]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e50\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRated speed\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e[min\u003csup\u003e-1\u003c/sup\u003e]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMax. speed, temp.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e[min\u003csup\u003e-1\u003c/sup\u003e]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4500\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eValves\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e[-]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMax. cyl. Pressure\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e[bar]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e150\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMass balance\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1st order\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eThe N38 cylinder head and cylinder liner are conceived as a pre‑assembled module that can be mounted as a unit onto the Hydra crankcase, enabling rapid variant changes on the test bench. A height‑adjustable deck system allows continuous setting of the clearance height and therefore variation of the compression ratio without major mechanical modifications. To maximize flexibility, both the intake and exhaust sides of the cylinder head are equipped with VVT2‑based variable valve timing actuators.\u003c/p\u003e \u003cp\u003eThe intake system follows a modular concept in which each of the two-cylinder head intake ports contains an exchangeable runner that can be configured with different inserts to influence in‑cylinder charge motion (tumble/swirl) or replaced by dummy tubes without internal structures. Depending on the test configuration, the Y‑piece can be mounted directly to the cylinder head for a very short intake path, or additional modules such as a port fuel injection (PFI) rail section or a swirl‑control valve module can be inserted.\u003c/p\u003e \u003cp\u003eThe valve train can be operated with either fixed or variable valve timing strategies. Variable timing is realized by axial‑piston phasing units that provide VVT (variable valve timing) functionality with a wide adjustment range of up to 80\u0026deg; crank angle, significantly exceeding the range of typical production vane‑type phasers. For operation with fixed valve timing, dummy elements can be installed in place of the VVT units. This freedom in valve train configuration allows different cam profiles and timing strategies to be applied, supporting a broad spectrum of combustion and knock‑limit investigations.\u003c/p\u003e \u003cp\u003e \u003cem\u003eTest bench environment\u003c/em\u003e \u003c/p\u003e \u003cp\u003eThe above-mentioned research engine is braked with an AC 25 OF dynamometer from Froude Consine. Engine control is provided by the FI2RE software together with its FI2RE-VCC control unit from IAV, handling not only fuel pump and injector control, but also ignition and variable camshaft timing. Oil and coolant conditioning systems of type STO 1-4-B10/20-TK6/TKN6 (oil) and STL 150/1-6-B10/20-KN6 (cooling water) by Single are used. The fuel cooler's heat exchanger is operated with a Lauda VC 3000 chiller. When required, intake pressurization is handled by a DLR 300 compressor from Elmo Rietschle, with charge air cooled by a Lauda T 4600 device.\u003c/p\u003e \u003cp\u003eAll major systems\u0026mdash;engine, dynamometer, and measurement equipment\u0026mdash;are housed in a dedicated test cell, which is separated by a safety door from the control and monitoring workplaces. Both the test cell and control workplaces are equipped with extensive safety features such as an emergency stop, gas detectors, and shut-off valves.\u003c/p\u003e \u003cp\u003eFor cylinder-pressure indication, a Kistler \u003cem\u003eKiBox To Go\u003c/em\u003e type 2893A with appropriate sensors is used, with cooling provided by the Kistler type 2621 water-cooling system. This setup records a wide variety of engine parameters: pressures, temperatures, and state variables such as current crank angle. Exhaust gas analysis uses the AVL AMA i60 R1 system; particulate emissions can alternatively be measured with a TSI 3776 instrument. All set control values are displayed and directly provided by the FI2RE software interface.\u003c/p\u003e \u003cp\u003e \u003cem\u003eChallenges of DI operation\u003c/em\u003e \u003c/p\u003e \u003cp\u003eThe main challenge for direct injection operation of the research engine is the physical state of LPG. The LPG must remain in liquid form throughout the system\u0026mdash;from the storage bottle to the combustion chamber\u0026mdash;to inject it directly as a liquid. The fuel is extracted from the liquid phase inside the bottle using a riser tube, but the bottle pressure is dependent on vapor pressure, which in turn depends on the temperature of the bottle and thus the environment. As the boiling point of LPG mixtures is below 0\u0026deg;C at atmospheric pressure, the fuel is in gaseous state under normal conditions. Therefore, cooling and/or increasing the system pressure is required to keep LPG in the liquid phase and avoid phase changes within the system.\u003c/p\u003e \u003cp\u003eAnother significant challenge is wear on components of the fuel system, combustion chamber, and exhaust tract when operating with LPG instead of gasoline [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. As already discussed with reference to DIN EN ISO 11114, various statements are available concerning the compatibility of LPG or its constituents with different materials.\u003c/p\u003e"},{"header":"4 Implementation on the real test bench","content":"\u003cp\u003eBuilding upon the experimental strategy and component selection described in the previous chapters, this section details the practical realization and incremental optimization of the liquid LPG direct injection system as implemented on the real test bench. In the following chapters the structure of the components implemented in the test bench area are explained and its architecture displayed ( \u0026rarr; Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e and Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). To fully understand the structure a legend is displayed in Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eLegend for the test bench components\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"3\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eValves/Component/Sensors\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSymbol\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMeaning\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003ePumps\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eP1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003ePressure rise pump gas bottle\u0026thinsp;+\u0026thinsp;1bar\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eP2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMid pressure pump\u0026thinsp;\u0026lt;\u0026thinsp;=\u0026thinsp;20bar\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eP3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eHigh pressure pump\u0026thinsp;\u0026lt;\u0026thinsp;=\u0026thinsp;200bar\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"5\" rowspan=\"6\"\u003e \u003cp\u003eManual Valves\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eV1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eBottle valve propane\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eV2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eBottle valve nitrogen\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eV3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eValve manual depressurize\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eV4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eValve calibrate Coriolis in\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eV5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eValve calibrate Coriolis out\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eV6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eThrottle valve return\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"3\" rowspan=\"4\"\u003e \u003cp\u003eElectric Valves\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMV1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eSolenoid valve gas bottle\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMV2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eSolenoid valve depressurize system (emergency, normally open)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMV3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eSolenoid valve gas chromatography\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMV4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eSolenoid valve gas supply HP-pump\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eStatic Valves\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCV1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eCheck valve\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCV2\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAutomatic\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePR1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003ePressure regulator gas chromatography\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003ePressure sensor\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ep1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003ePressure in the bottle\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ep2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003ePressure before injector\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ep3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003ePressure before HP-pump\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"4\" rowspan=\"5\"\u003e \u003cp\u003eTemperature sensor\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eT1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eTemperature gas in bottle\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eT2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eTemperature gas pre injector\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eT3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eTemperature mid pressure gas\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eT4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eTemperature mid pressure gas\u0026thinsp;+\u0026thinsp;return\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eT5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eTemperature gas return\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\dot{V}\\)\u003c/span\u003e\u003c/span\u003ė\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCoriolis\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eCoriolis scale\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"3\"\u003e\u003cem\u003eStarting Situation: Initial System Layout\u003c/em\u003e\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eThe fuel system for liquid LPG direct injection was developed iteratively, focusing on identifying and systematically addressing the root causes of emerging issues. Rather than immediately realizing the optimized topology described in Chap.\u0026nbsp;4.3, the project began with a deliberately simple initial configuration, which served as a diagnostic baseline.\u003c/p\u003e \u003cp\u003eIn this initial setup, LPG was extracted via a dip tube from a horizontally stored cylinder, positioned in a secure, external gas bottle cabinet. The dip tube guaranteed the withdrawal of LPG from the liquid phase. Following the bottle valve, a check valve prevented reverse flow, while a safety shut-off valve \u0026mdash; integrated with the emergency stop system \u0026mdash; immediately cut off fuel flow in case of emergency. The system also offered the option of nitrogen flushing to remove residual LPG from lines prior to recommissioning.\u003c/p\u003e \u003cp\u003eA Coriolis mass flow meter measured fuel consumption precisely, and a manual valve allowed for gas chromatographic analysis of the LPG blend. After passing through these monitoring units, the fuel was delivered to a high-pressure pump (BMW N53 type), aimed at generating 200 bar. Pressure and temperature sensors monitored the state of the fuel up to the injector. Initial runs revealed pressure collapse during injection, likely due to the formation of vapor bubbles caused by heat input at the high-pressure pump. It became clear that either increasing the fuel pressure upstream of the pump or improved cooling was necessary to suppress outgassing and ensure system stability.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cem\u003eIntermediate System Configurations\u003c/em\u003e \u003c/p\u003e \u003cp\u003eSubsequent development steps involved targeted modifications and additions to progressively stabilize fuel delivery. Initially, only the cylinder pressure of the LPG storage bottle was used to supply the high‑pressure pump, but this proved insufficient to prevent high‑pressure collapse during injection. To increase the fuel pressure upstream of the high‑pressure pump and provide a greater buffer against vaporization, an automotive low‑pressure gasoline feed pump was added, generating an overpressure of up to approximately 5 bar. After roughly one day of operation, leakage occurred on the pressure side of this pump. The same pump type was deliberately re‑installed to enable further measurements on the high‑pressure side, accepting that it would again develop leaks; however, this approach did not provide a sustainable solution and highlighted the need for a more robust design.\u003c/p\u003e \u003cp\u003eAs a next step, a water injection pump was tested as a more durable feed unit. Despite its higher structural robustness, this pump also developed leakage at the housing cover after approximately one day of operation. The root cause was traced to the sealing system: the original O‑rings, selected for water operation, were not sufficiently compatible with liquid LPG. Due to its low viscosity and high solubility tendency for certain elastomers, LPG can penetrate and swell standard sealing materials that are suitable for aqueous media but not for liquefied hydrocarbons. As a result, micro‑gaps may form, allowing LPG to migrate through the seal and eventually leak. To address this issue, the two sealing O‑rings were replaced with rings specifically declared as LPG‑compatible, and the pump cover was exchanged for a version using threaded connections instead of quick‑release couplings on the inlet and outlet, thereby reducing potential leak paths. In the final configuration of this stage, this modified unit corresponded to pump P2 in Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e.​\u003c/p\u003e \u003cp\u003eIn parallel to the feed‑pump adaptations, further thermal and hydraulic measures were implemented. An intermediate fuel cooler was installed between the feed pump and the high‑pressure pump, reducing LPG temperature to approximately 10\u0026deg;C. This improved start‑up stability but did not fully eliminate vapor formation and pressure loss during injection. To further mitigate outgassing, a gas separator (\u0026ldquo;bubble trap\u0026rdquo;) was integrated; its design allowed free gas to be vented while the high‑pressure pump was supplied from the lower, liquid phase. This configuration enabled stable high‑pressure injection for approximately 30 seconds, but longer‑term operation still resulted in pressure collapse.\u003c/p\u003e \u003cp\u003eReplacing the original high‑pressure pump with a modern Bosch HDP 5 unit improved performance and extended stable operation to about one to two minutes before pressure decay recurred, now primarily attributed to heat‑induced gas bubble formation within the pump itself. Subsequent system optimization therefore included a more powerful low‑pressure pump, raising line pressure to approximately 11\u0026ndash;12 bar, and the introduction of a return line to recirculate and cool excess fuel. The return circuit allowed hot fuel in the high‑pressure path to be directed back to the low‑pressure side, where it was cooled and re‑pressurized. Additional temperature monitoring points throughout the system improved diagnostic capability and helped localize persistent weaknesses.Despite these targeted improvements, the system remained sensitive to thermal inputs, with occasional pressure spikes and damage to high‑pressure lines revealing the complexity of managing phase behavior in such high‑pressure LPG systems.\u003c/p\u003e \u003cdiv id=\"Sec5\" class=\"Section3\"\u003e \u003cdiv class=\"Heading\"\u003e4.1.1 Material compatibility of LPG with conventional pump systems\u003c/div\u003e \u003cp\u003eThe progressive failure of several low‑pressure feed pumps in the development phase effectively created an incidental material compatibility study for LPG in conventional pump systems. The initial approach\u0026mdash;using a standard gasoline feed pump\u0026mdash;demonstrated that components and seals designed for liquid hydrocarbons at moderate pressures and temperatures may not withstand continuous operation with cold, high‑pressure LPG without modification. Leakage on the pressure side after short operating times indicated that standard elastomer seals and quick‑coupling interfaces were not sufficiently tight or chemically robust for liquefied gas operation.​\u003c/p\u003e \u003cp\u003eThe root cause lies in the chemical properties of LPG itself. Propane and butane, the principal constituents of LPG, are low‑molecular‑weight aliphatic hydrocarbons with exceptionally high solubility in many standard elastomeric materials, particularly those based on nitrile rubber (NBR) or silicone compounds. Unlike conventional liquid fuels, which exhibit more moderate interaction with elastomers, LPG's low viscosity, high vapor pressure, and molecular penetrating power allow the fuel to diffuse through and swell standard sealing compounds over time. This process causes dimensional changes, material degradation, and the formation of microscopic gaps\u0026mdash;micro‑gaps\u0026mdash;that serve as preferential leak paths. Additionally, the high solubility of LPG in certain elastomers can cause the sealing material to become overly soft and lose its ability to maintain contact pressure, further compromising the seal integrity.​\u003c/p\u003e \u003cp\u003eStandard O‑rings selected for aqueous media (as in water injection pump applications) offer even less resistance to LPG than those designed for conventional fuels, since they are formulated to be compatible with polar solvents rather than non‑polar hydrocarbons. The subsequent use of this water‑rated pump confirmed that even structurally robust pump housings do not guarantee LPG compatibility if the sealing concept is not adapted. Only after replacing the original O‑rings with LPG‑rated sealing elements\u0026mdash;typically based on fluorocarbon (FKM/Viton\u0026reg;) or fluorosilicone (FVMQ) compounds, which exhibit significantly lower swelling and superior chemical resistance to low‑molecular‑weight hydrocarbons [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]\u0026mdash;and installing a revised cover with threaded connections instead of quick‑release fittings, could acceptable tightness and operational stability be achieved for pump P2 (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e).​\u003c/p\u003e \u003cp\u003eTo ensure reliable and reproducible fuel mass flow measurements with the Coriolis flow meter, an additional pre‑feed stage (pump P1 in Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e) was implemented upstream of the measurement section. For the final measurement campaign, a conventional automotive fuel pump was used at a deliberately reduced pressure level to delay the onset of unavoidable long‑term leakage while still providing a sufficiently stable inlet pressure for the Coriolis sensor. This configuration ensured robust fuel‑flow measurements for the conclusive test series and provided a stable hydraulic basis for the final system layout described in the following section.\u003c/p\u003e \u003cp\u003e \u003cem\u003eFinal concept: stable LPG Operation\u003c/em\u003e \u003c/p\u003e \u003cp\u003eA stable and robust system configuration was finally achieved by implementing several coordinated measures. The LPG bottle was moved into the test cell, with select lines fitted with visual flow indicators. The return line was actively cooled before rejoining the low-pressure circuit, and a small feed pump was introduced immediately after the storage bottle to provide an initial pressure increase (approx. 0.8 bar). A pressure control valve ensured that the low-pressure circuit did not exceed 12 bar, thereby preventing overpressure and minimizing the risk of vapor bubble nucleation.\u003c/p\u003e \u003cp\u003eThrough this combination of modest pre-pressure increase, targeted cooling, a gas separator, and recirculation, the fuel system could now reliably maintain 200 bar injection pressure, even during sustained engine operation. As a result, long-term, stable test bench operation was achieved, allowing the planned series of liquid LPG direct injection measurements to be conducted successfully.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"5 Measurement procedure","content":"\u003cp\u003eThe experimental investigation systematically compared liquid LPG direct injection performance against conventional gasoline baselines using two distinct camshaft phasing strategies. Baseline measurements were established using standard E10 gasoline (E10-1 through E10-16) across speed sweep conditions from 1000 to 3800 rpm at constant 10 bar IMEP (E10-1 to E10-7), and load sweep conditions at constant 3000 rpm with IMEP progressively increased from 2 to 18 bar (E10-8 to E10-16).\u003c/p\u003e \u003cp\u003eTo evaluate the influence of fuel octane rating on knock behavior, high-octane reference measurements were conducted using Aral Ultimate 102 (E10-17 through E10-19) at full-load conditions (16, 18, and 20 bar IMEP at 3000 rpm). Comparative analysis revealed that despite its elevated research octane number of 102, Ultimate 102 exhibited knock behavior nearly identical to standard E10 gasoline (RON 95), showing no significant knock resistance improvement at full load. Consequently, subsequent LPG performance comparisons focus on E10 as the primary baseline, as this more clearly demonstrates LPG's practical advantage through its superior knock resistance (RON 103\u0026ndash;111 depending on propane/butane ratio), which enables advanced ignition timing and improved thermal efficiency.\u003c/p\u003e \u003cp\u003eThe LPG measurement series (LPG-20 through LPG-51) systematically investigated the influence of camshaft phasing on combustion efficiency and knock behavior through two distinct valve timing configurations. The first configuration, designated LPG-SCT (Standard Cam Timing), employed intake valve event spreads of 125\u0026deg; camshaft angle (CA) for IMEP\u0026thinsp;\u0026le;\u0026thinsp;10 bar and 120\u0026deg; CA for IMEP\u0026thinsp;\u0026gt;\u0026thinsp;10 bar, with exhaust spreads of 105\u0026deg; CA for IMEP\u0026thinsp;\u0026le;\u0026thinsp;10 bar and 75\u0026deg; CA for IMEP\u0026thinsp;\u0026gt;\u0026thinsp;10 bar. The second configuration, LPG-RCT (Reduced Cam Timing), utilized symmetrical 80\u0026deg; CA spreads for both intake and exhaust valves across all load conditions. The speed sweep encompassed LPG-20 through LPG-31, with measurements alternating between LPG-SCT and LPG-RCT configurations across the 1000\u0026ndash;3800 rpm range at 10 bar IMEP. The load sweep extended from LPG-32 through LPG-51, maintaining the same alternating cam timing pattern with IMEP varied from 2 to 20 bar at constant 3000 rpm engine speed. Operating point LPG-35 was excluded from analysis due to deviations from target parameters and was replaced by LPG-35.1. Operating point LPG-45 was not conducted, as stable operation at 2 bar IMEP could not be reliably achieved under prevailing test conditions.\u003c/p\u003e \u003cp\u003eThis measurement structure enabled direct quantification of LPG direct injection performance relative to E10 baseline while systematically isolating the influence of valve timing strategy on combustion phasing, efficiency, and knock margin across the full operating envelope.\u003c/p\u003e \u003cp\u003e \u003cem\u003eLoad sweep results\u003c/em\u003e \u003c/p\u003e \u003cp\u003eThis section presents the results of the load sweep (E10-8 to E10-16, LPG-32 to LPG-51), comparing E10 baseline, LPG-SCT, and LPG-RCT (80/80\u0026deg; camshaft angle configuration) across IMEP values from 2 to 20 bar at constant 3000 rpm engine speed. \u0026rarr; Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e displays eight sub-plots of key performance parameters.\u003c/p\u003e \u003cp\u003eExhaust gas temperature increases progressively with load across all three fuels. LPG-RCT generally exhibits slightly lower values in the mid-load range compared to LPG-SCT and E10, suggesting that the 80/80\u0026deg; valve timing configuration promotes more efficient combustion with reduced thermal burden. At the highest load points, exhaust temperatures converge across all fuels.\u003c/p\u003e \u003cp\u003eInjection duration for LPG is substantially longer than for E10, with LPG-SCT showing the most pronounced difference. This corresponds to fuel consumption penalties that decrease from approximately 25% at low loads to below 15% at high loads. LPG-RCT demonstrates more favorable consumption characteristics than LPG-SCT.\u003c/p\u003e \u003cp\u003eThe 50% mass fraction burned (MFB) reveals critical differences in combustion control. E10 begins to deviate significantly from the desired 8\u0026deg; after top dead center (ATDC) above approximately 12 bar IMEP due to knock-limited ignition timing retard, while LPG maintains proximity to this ideal value up to 16\u0026ndash;18 bar IMEP. Knock frequency analysis shows that E10 rapidly enters regimes of elevated knock activity with increasing load, whereas LPG permits significantly earlier ignition timing without proportional knock increase. Ignition timing plots confirm this trend, with LPG consistently utilizing earlier spark advance than E10 across the load range, particularly in the LPG-RCT configuration.\u003c/p\u003e \u003cp\u003eCycle-to-cycle variability of IMEP demonstrates that LPG exhibits superior stability compared to E10 in the low to mid-load range; however, this advantage partially reverses at very high loads. LPG-RCT shows marginally reduced stability at extreme loads, likely attributable to altered in-cylinder flow patterns under these conditions.\u003c/p\u003e \u003cp\u003eBrake thermal efficiency reveals that LPG achieves comparable or superior efficiency to E10 at low loads (below approximately 6 bar IMEP), but efficiency declines in the 8\u0026ndash;16 bar range before recovering at the highest load points. LPG-RCT demonstrates modest efficiency gains at high loads relative to LPG-SCT. These results indicate that LPG derives clear benefit from its knock resistance at high load, where earlier ignition timing can be utilized, whereas optimization of mixture formation remains necessary in the mid-load regime.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cem\u003eSpeed sweep results\u003c/em\u003e \u003c/p\u003e \u003cp\u003eThe speed sweep investigation varied engine speed from 1000 to 3500 rpm in 500 rpm increments, with one additional measurement point at 3800 rpm, all at nominally 10 bar IMEP (E10-1 to E10-7, LPG-20 to LPG-31). \u0026rarr; Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e presents four performance parameters across the speed range. Note that LPG-RCT could not be stably operated above 3000 rpm without supercharging; consequently, complete comparative data at higher speeds is unavailable for this configuration.\u003c/p\u003e \u003cp\u003eCycle-to-cycle variability demonstrates that LPG delivers more consistent combustion than E10 across the lower speed range; this stability advantage diminishes at elevated speeds. The combustion center of gravity remains controllable for LPG across a wide speed range, maintaining proximity to the target 8\u0026deg; ATDC, whereas E10 requires repeated ignition timing retard as knock propensity increases.\u003c/p\u003e \u003cp\u003eIgnition timing shows the consistent trend observed in the load sweep: LPG permits substantially earlier spark advance throughout, with LPG-RCT achieving the most advanced timing at lower speeds.\u003c/p\u003e \u003cp\u003eBrake thermal efficiency exhibits a notable difference from the load sweep: LPG efficiency generally trails E10 across the speed sweep, particularly at elevated engine speeds. This observation suggests that LPG does not derive equivalent efficiency benefit from its knock resistance when operating at constant load across varying speeds. The LPG-RCT configuration offers no pronounced efficiency advantage in this speed-varied regime, indicating that the valve timing optimization benefits observed at high load do not translate uniformly across the speed spectrum at moderate load.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cem\u003eEmission characteristics\u003c/em\u003e \u003c/p\u003e \u003cp\u003eEmission measurements were conducted exclusively during the load sweep (E10-8 to E10-16, LPG-32 to LPG-51; 2 to 20 bar IMEP at 3000 rpm) to characterize transient combustion behavior across the engine's load envelope. \u0026rarr; Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e presents six emission parameters. Note that E10 baseline data at 14 bar IMEP are interpolated from adjacent measurements; these values should be interpreted with appropriate caution.\u003c/p\u003e \u003cp\u003e \u003cb\u003eIncomplete Combustion Products\u003c/b\u003e \u003c/p\u003e \u003cp\u003eTotal hydrocarbon (THC) emissions from LPG are occasionally elevated relative to E10 in the very low-load range (below 6 bar IMEP), particularly with the LPG-RCT configuration, but decline substantially above 6 bar IMEP to values often below E10 levels. Combined unburned hydrocarbons and partial oxidation products (CO\u0026thinsp;+\u0026thinsp;H) show occasional elevation in LPG-RCT operation, particularly in certain mid-load points, suggesting incomplete combustion when the interaction of ignition timing, charge motion, and thermal conditions does not ensure homogeneous oxidation throughout the combustion chamber.\u003c/p\u003e \u003cp\u003e \u003cb\u003eThermal NOx Formation and Temperature Effects\u003c/b\u003e \u003c/p\u003e \u003cp\u003eNitrogen oxide (NOx) emissions rise monotonically with load for all fuels, reflecting elevated pressure and temperature conditions. LPG-RCT frequently exhibits higher NOx peaks in the mid-load range relative to LPG-SCT, attributable to its earlier ignition timing and associated local temperature elevation; conversely, E10 produces comparatively lower NOx at the highest loads due to knock-mandated ignition timing retard, which reduces peak combustion temperatures. Exhaust gas temperature increases with load for all fuels, with E10 generally achieving the highest temperatures at maximum load. Both LPG-SCT and particularly LPG-RCT operate at lower exhaust temperatures, likely attributable to earlier ignition timing that shifts the bulk heat release event toward the expansion stroke, reducing exhaust gas enthalpy despite potentially higher in-cylinder peak temperatures.\u003c/p\u003e \u003cp\u003e \u003cb\u003eCarbon Balance and Particulate Emissions\u003c/b\u003e \u003c/p\u003e \u003cp\u003eCarbon dioxide (CO₂) concentrations are generally lower for LPG than E10, reflecting the favorable carbon-to-hydrogen ratio of propane and butane. However, this advantage diminishes at high loads where LPG's lower volumetric energy density necessitates higher fuel mass flow. Residual oxygen (O₂) measurements reveal that E10 exhibits lower exhaust oxygen content at very high loads, consistent with extended post-flame oxidation when late ignition timing is imposed. LPG-RCT occasionally shows elevated O₂ levels correlated with higher CO\u0026thinsp;+\u0026thinsp;H concentrations.\u003c/p\u003e \u003cp\u003eParticulate matter emissions demonstrate markedly lower soot formation tendency for LPG across the load range compared to E10, despite occasional excursions at very low loads. This advantage aligns with LPG's chemical composition and the improved combustion control enabled by early ignition timing.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cb\u003eSummary of Emission Trade-offs\u003c/b\u003e \u003c/p\u003e \u003cp\u003eOverall, LPG demonstrates principal advantages at high load through superior knock resistance enabling advanced ignition timing, a favorable carbon-to-hydrogen stoichiometry, and dramatically reduced particulate formation. The distinction between LPG-SCT and LPG-RCT valve timing configurations illustrates how camshaft phasing substantially influences NOx and oxidation product formation: earlier spark advance elevates peak in-cylinder temperatures but can modestly reduce exhaust temperature when primary heat release is shifted to the expansion phase. These emission profiles underscore that optimal design of a liquid LPG direct injection system requires careful balancing of efficiency, emission control, and combustion stability across the full operating envelope.\u003c/p\u003e"},{"header":"6 Discussion and auxiliary investigations","content":"\u003cp\u003eThroughout the testing campaigns, additional investigations were performed to deepen understanding of the fuel system's behavior under varying conditions and to identify optimization potential. For instance, systematic tests of the return cooler examined how altering the temperature of the return line affected system stability. Lowering coolant temperature by 5 K yielded no adverse effects; reduced temperature at a given pressure expectedly minimized outgassing in the high-pressure circuit. In contrast, gradually increasing the temperature up to ambient unveiled a distinct shift: As the return temperature reached room temperature, the high-pressure pump demand rose continuously to maintain the 200 bar threshold. Once maximum actuation was reached, the fuel pressure dropped, the injector could no longer deliver sufficient fuel, and engine operation had to be terminated manually. This highlights that even moderate temperature rises in the return circuit can trigger phase changes and compromise system stability, especially when components are tuned to operate within tight pressure and temperature windows.\u003c/p\u003e \u003cp\u003eAnother trial evaluated whether the final fuel system design (Version 7) could operate reliably using a Bosch HDP 5 high-pressure pump instead of the prior Audi pump. Because the Bosch pump design does not support a return line, the recirculation loop had to be omitted. In this configuration, the system failed to maintain a stable fuel pressure, underlining the importance of the Audi pump's return system for operational reliability. With recirculation, unused fuel is consistently returned, cooled, and pressurized in the low-pressure circuit, reducing the risk of vaporization in the high-pressure path.\u003c/p\u003e \u003cp\u003eFuel consumption measurements confirmed fundamental expectations: At comparable volumetric flow rates, LPG exhibits lower energy density than gasoline. Depending on load point, observed consumption increases ranged from 25% to 15% with the older valve timing setup, and about 10% with the 80/80 camshaft spread\u0026mdash;values consistent with published vehicle trials reporting deviations between 10% and 30%. Knock behavior showed clear LPG advantages: A higher IMEP range could be achieved before knocking combustion occurred. Whereas gasoline required ignition timing retard above approximately 12 bar IMEP to control knock, LPG maintained an ideal combustion centroid of around 8\u0026deg; crank angle after top dead center up to more than 14 bar. Even when combustion phasing had to be shifted, LPG's knock frequency remained lower, allowing consistently earlier ignition timing.\u003c/p\u003e \u003cp\u003eThese findings underline that liquid LPG direct injection systems offer clear performance benefits, but require stringent demands on fuel system robustness. A minimum pre-pump pressure of 11\u0026ndash;12 bar at approximately 20\u0026deg;C is essential to avoid outgassing. The return circuit is critical for recycling and cooling excess fuel, stabilizing the high-pressure region. Despite the increased complexity, the project succeeded in producing stable single-cylinder engine operation on liquid LPG. The results lay a foundation for future system enhancements, including more robust components, optimized return cooling, or further camshaft timing adjustments. Overall, the achieved project status enabled a variety of measurements that reveal both considerable advantages and distinct limitations of LPG in comparison to gasoline. Supplementary tests demonstrate how operational influences such as temperature, pressure, and pump choice impact system reliability, confirming the need for extended durability and materials investigations to assess long-term viability of pumps and sealing components.\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eA.D. designed and conducted the experiments, performed the data analysis, and wrote the main manuscript text. H.R. supervised the research, guided the scientific concept and methodology, and contributed to the interpretation of the results and the revision of the manuscript. B.R. supported the experimental planning, data evaluation, and technical refinement of the test setup, and contributed to manuscript text. The experimental work on the test bench, including engine modification, fuel system integration, and exhaust gas measurement, was enabled by the technical staff, whose expertise and support were essential for the successful completion of this study. All authors reviewed and approved the final manuscript.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAdolf, J., Balzer, C., Joedicke, A., Schabla, U.: SHELL Fl\u0026uuml;ssiggas-Studie: LPG als Energietr\u0026auml;ger und Kraftstoff, trends und Perspektiven\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDIN: EN 589:2022-04, Kraftstoffe_- Fl\u0026uuml;ssiggas_- Anforderungen und Pr\u0026uuml;fverfahren; Deutsche Fassung EN_589:2018\u0026thinsp;+\u0026thinsp;A1:2022, Beuth Verlag GmbH\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePropane Grades and Quality - HD5 HD10 and Commercial Propane: (2025). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.propane101.com/propanegradesandquality.htm\u003c/span\u003e\u003cspan address=\"https://www.propane101.com/propanegradesandquality.htm\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e, January 29\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eG\u0026uuml;nther, M.: Comparison of the thermodynamic potential, the octane requirement and the contamination tendency of three LGP concepts in a turbocharged direct injection gasoline SI engine: LPG-DI, LPG-PFI (liquid) and LPG-PFI (gaseous): LPG System Comparison, FVV Project 1069\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eClesse, U.K., Solutions for, LPG: Technical Advice - Clesse UK Solutions for LPG, (2025). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://clesse.co.uk/technical-advice/\u003c/span\u003e\u003cspan address=\"https://clesse.co.uk/technical-advice/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e, November 21\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTEGA \u0026ndash; Technische Gase und Gasetechnik GmbH: Physikalische Daten und brenntechnische Eigenschaften, \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.tega.de/fileadmin/Downloads_und_Bilder/FG_Tank/Info3_Physikalische_Daten_und_brenntechnische_Eigenschaften_1.pdf\u003c/span\u003e\u003cspan address=\"https://www.tega.de/fileadmin/Downloads_und_Bilder/FG_Tank/Info3_Physikalische_Daten_und_brenntechnische_Eigenschaften_1.pdf\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDIN EN ISO 11114-: 1:2020-08, Gasflaschen_- Vertr\u0026auml;glichkeit von Werkstoffen f\u0026uuml;r Gasflaschen und Ventile mit den in Ber\u0026uuml;hrung kommenden Gasen_- Teil_1: Metallische Werkstoffe (ISO_11114-1:2020); Deutsche Fassung EN_ISO_11114-1:2020, Beuth Verlag GmbH\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNagel, D., Krieck, M., System Comparison, L.P.G., II, Project No:. 1151, Investigation of the Effect of supercritical LPG on DGUV Regel 110\u0026thinsp;\u0026ndash;\u0026thinsp;010 Verwendung von Fl\u0026uuml;ssiggas: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.bghm.de/fileadmin/user_upload/Arbeitsschuetzer/Gesetze_Vorschriften/Regeln/110-010.pdf\u003c/span\u003e\u003cspan address=\"https://www.bghm.de/fileadmin/user_upload/Arbeitsschuetzer/Gesetze_Vorschriften/Regeln/110-010.pdf\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSysteme | Prins Autogassystemen BV: (2025). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.prinsautogas.com/de/systeme\u003c/span\u003e\u003cspan address=\"https://www.prinsautogas.com/de/systeme\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e, January 29\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBasshuysen, R., Sch\u0026auml;fer, F.: Handbuch Verbrennungsmotor: Grundlagen, Komponenten, Systeme, Perspektiven. Springer Vieweg (2014)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSeungmook, O.: Combustion and emission characteristics in a direct injection LPG/gasoline spark ignition engine, \u003cem\u003eSAE Paper 2010-01-1461\u003c/em\u003e, (2010)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchrick, G.H.: Nockenwellen-Einbau: Camshaft Installation: https:// (2025). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003eschrick.com/media/d1/7c/3f/1647859651/nockenwelleneinbau__camshaft_installation-berichtigt-25102005.pdf\u003c/span\u003e\u003cspan address=\"http://schrick.com/media/d1/7c/3f/1647859651/nockenwelleneinbau__camshaft_installation-berichtigt-25102005.pdf\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e, Zugriff: Jan \u003cb\u003e29\u003c/b\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJ\u0026ouml;rg Adolf, C., Balzer, B., Gn\u0026ouml;rich, A., et al.: Joedicke Shell Fl\u0026uuml;ssiggas-Studie: LPG als Energietr\u0026auml;ger und Kraftstoff Fakten, Trends und Perspektiven,., (2015). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.13140/RG.2.2.20645.63208\u003c/span\u003e\u003cspan address=\"10.13140/RG.2.2.20645.63208\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBender, B., G\u0026ouml;hlich, D. (eds.): Dubbel Taschenbuch f\u0026uuml;r den Maschinenbau 2: Anwendungen, 26th ed., Springer Berlin Heidelberg, Berlin, Heidelberg, ISBN 9783662597132, (2020)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eUnfallversicherung, D.G.: DGUV Regel 110\u0026thinsp;\u0026ndash;\u0026thinsp;010 \u0026bdquo;Verwendung von Fl\u0026uuml;ssiggas\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBRC: Produkte, (2025). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://brc.it/de/produkte/\u003c/span\u003e\u003cspan address=\"https://brc.it/de/produkte/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e, November 19\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRenzo, L.: LPG | Landi Renzo, (2025). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://landirenzo.com/en/lpg-vehicle-systems\u003c/span\u003e\u003cspan address=\"https://landirenzo.com/en/lpg-vehicle-systems\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e, November \u003cb\u003e19\u003c/b\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003evan Basshuysen, R., Sch\u0026auml;fer, F.: Handbuch Verbrennungsmotor, Springer Fachmedien Wiesbaden, Wiesbaden, ISBN 978-3-658-04677-4, (2015)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBarrett, S.: European Expert Group reports on future transport fuels. Fuel Cells Bull. \u003cb\u003e2011\u003c/b\u003e(2), 12\u0026ndash;16 (2011). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/S1464-2859(11)70096-7\u003c/span\u003e\u003cspan address=\"10.1016/S1464-2859(11)70096-7\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSeungmook Oh, S., Lee, Y., Choi, K.-Y., Kang, et al.: Combustion and Emission Characteristics in a Direct Injection LPG/Gasoline Spark Ignition Engine, International Powertrains, Fuels \u0026amp; Lubricants Meeting, SAE International, ISBN 2688-3627, (2010)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLIQUEFIED PETROLEUM GAS Resistant Orings: Xrings, Gaskets, Rubber products - Marco Rubber, (2025). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.marcorubber.com/chemical-compatibility/LIQUEFIED%20PETROLEUM%20GAS\u003c/span\u003e\u003cspan address=\"https://www.marcorubber.com/chemical-compatibility/LIQUEFIED%20PETROLEUM%20GAS\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e, December 16\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eO-Ring Compatibility Chemical Resistance Chart: (2025). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.allorings.com/o-ring-compatibility\u003c/span\u003e\u003cspan address=\"https://www.allorings.com/o-ring-compatibility\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e, December \u003cb\u003e16\u003c/b\u003e\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"automotive-and-engine-technology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"aaet","sideBox":"Learn more about [Automotive and Engine Technology](http://link.springer.com/journal/41104)","snPcode":"41104","submissionUrl":"https://submission.nature.com/new-submission/41104/3","title":"Automotive and Engine Technology","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Combustion Analysis, Power Density, LPG Direct Injection, Knock Limit","lastPublishedDoi":"10.21203/rs.3.rs-8640603/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8640603/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThis work investigates the direct injection of Liquefied Petroleum Gas (LPG) in its liquid phase in a single-cylinder research engine, focusing on the primary challenge of maintaining LPG in its liquid state throughout the fuel system to prevent vaporization and gas bubble formation. Due to propane\u0026rsquo;s low critical temperature (96.8\u0026deg;C), fuel-carrying components exposed to typical engine operating temperatures (80\u0026ndash;110\u0026deg;C) are prone to outgassing, which creates pressure drops and disrupts continuous liquid injection. To overcome this, the 200-bar high-pressure fuel system was optimized by combining increased pressure, active cooling, gas separation, and recirculation strategies, ensuring stable liquid conditions upstream of the pump and injector. This robust system allowed for continued stable engine operation and enabled comprehensive comparative measurements\u0026mdash;including in-cylinder pressure analysis and FTIR-based emission measurements\u0026mdash;to systematically evaluate combustion behavior and environmental impacts. The results demonstrate that LPG\u0026rsquo;s superior knock resistance (RON 103\u0026ndash;111) allowed for earlier ignition timing and higher engine loads compared to E10, with beneficial charge cooling effects further suppressing knock occurrence. In addition, LPG direct injection yielded significant reductions in particulate and CO₂ emissions relative to gasoline, confirming both the performance and environmental advantages of liquid LPG direct injection technology for future engine applications.\u003c/p\u003e","manuscriptTitle":"Investigation of Combustion and Performance Limits of Liquid LPG Direct Injection in a Single-Cylinder Research Engine","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-11 12:30:25","doi":"10.21203/rs.3.rs-8640603/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-05-07T07:20:07+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-05T15:13:28+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"316733123537984049193095569619854550891","date":"2026-05-05T15:05:45+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"279016383427605230058881753579321986991","date":"2026-03-18T17:16:30+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-02-21T09:33:32+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"301053582386093061824079614732029929570","date":"2026-02-20T15:23:11+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-02-06T10:02:54+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-01-22T15:59:15+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-01-21T07:08:48+00:00","index":"","fulltext":""},{"type":"submitted","content":"Automotive and Engine Technology","date":"2026-01-19T14:21:45+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"automotive-and-engine-technology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"aaet","sideBox":"Learn more about [Automotive and Engine Technology](http://link.springer.com/journal/41104)","snPcode":"41104","submissionUrl":"https://submission.nature.com/new-submission/41104/3","title":"Automotive and Engine Technology","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"317ae4d0-e335-4114-804f-963f67e7123e","owner":[],"postedDate":"February 11th, 2026","published":true,"recentEditorialEvents":[{"type":"decision","content":"Revision requested","date":"2026-05-07T07:20:07+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-05T15:13:28+00:00","index":14,"fulltext":""},{"type":"reviewerAgreed","content":"316733123537984049193095569619854550891","date":"2026-05-05T15:05:45+00:00","index":13,"fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"in-revision","subjectAreas":[],"tags":[],"updatedAt":"2026-05-07T07:26:09+00:00","versionOfRecord":[],"versionCreatedAt":"2026-02-11 12:30:25","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8640603","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8640603","identity":"rs-8640603","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}