CO₂ Emissions Reduction through Increasing H₂ Participation in Gaseous Combustible—Condensing Boilers Functional Response

Abstract

Considering the imperative reduction in CO₂ emissions, both from household heating and hot water producing facilities, one of the mainstream directions is to reduce hydrocarbons in combustibles by replacing them with hydrogen. The authors analyze condensing boilers operating when hydrogen is mixed with standard gaseous fuel (CH₄). The hydrogen (H₂) volumetric participation in the mixture is considered to vary in the range of 0 to 20%. The operation of the condensing boilers will be numerically modeled by computational programs and prior validated by experimental studies concluded in a European Certified Laboratory. The study concluded that an increase in the combustible flow with 16% will compensate the maximum H₂ concentration situation with no other implications on the boiler’s thermal efficiency, together with a decrease in CO₂ emissions by approximately 7%. By assuming 0.9 (to/year/boiler), the value of CO₂ emissions reduction for the condensing boiler determined in the paper, and extrapolating it for the estimated number of boilers to be sold for the period 2019–2024, a 254,700-ton CO₂/year reduction resulted.

1. Introduction

Nowadays, the environmentally friendly approach of all combustible consuming processes is a must in order to limit atmospheric changes, especially by reducing CO₂ concentrations via lowering CO₂ emissions. One major way of accomplishing this is the use of biogenic fuels that ensure CO₂ recirculation by the burning process corelated with biogenic material growth, resulting in zero overall CO₂ emissions and a recirculation of existing CO₂ in the atmosphere [1]. However, there are two issues regarding this solution. Firstly, there are places and appliances where biomass cannot be used as combustibles; and secondly, there are plenty of already existing gaseous fuel consumers that will continue to own, buy, and use small capacity condensing boilers.

Thus, considering the imperative reduction in CO₂ emissions, especially from household heating and hot water producing facilities, one of the other mainstream directions is to reduce hydrocarbons in the combustibles by replacing them with hydrogen. Whether they work alone as thermal sources for heating and hot water production or are implemented in complex systems along with heat recovery components or unconventional energy sources, the condensing boilers are a key element in household energetics.

If at some point, Romania’s gaseous combustible delivering network will be able to replace part of its hydrocarbon fuels with non-carbonic fuels, such as hydrogen as the most probable choice, it will be of outmost interest to determine how this will change the final users via the operation of condensing boilers. At the same time, an evaluation of the environmental impact would be of great interest.

Within the Centre of the Department of Thermal Sciences, part of the Technical University of Civil Engineering Bucharest, Romania, the authors have consistently conducted testing and experiments on combustion equipment powered by gaseous, liquid, or solid fuel, in order to assess their energy and combustion efficiency. The experimental stand is accredited by the Romanian National Accreditation Body (RENAR), a fact that gives confidence in the results of the tests performed [2].

Until now, the idea approached by the article has preoccupied many researchers. However, the studies carried out so far are more theoretical, and the experimental part is a must in order to validate the results.

Xin Y. et al. [3] conducted a numerical simulation regarding the combustion of natural gas mixed with hydrogen in gas boilers. The main conclusions were that adding hydrogen will increase the theoretical combustion temperature, reduce flue gas and CO₂ emissions, increase the generation of water vapor, and diminish NO_x generation. The authors also determined an optimal hydrogen mixing volume fraction for their study. However, it was approached as a swirl burner modeled in ANSYS 14.0, so an experimental part is further necessary in order to certify the results.

Leicher J. et al. [4] evaluated the impact of hydrogen admixture into natural gas on residential and commercial gas appliances. The authors used COSILAB software suite in order to generate a theoretical study aiming at the possible negative aspects of the hydrogen presence in the combustible gas mixture. It was stated that the laminar combustion velocities are often largely mitigated by a shift towards higher air excess ratios in the absence of combustion control systems. Furthermore, in some cases, the combustion control technologies are hampered by the high concentration of hydrogen.

The impact of hydrogen/natural gas blends on partially premixed combustion equipment has been examined by Glanville P. et al. [5]. One important aspect is that the article conclusions can be applied to natural gas appliances from Canada and the United States. Nevertheless, given the fact that the burning equipment design and natural gas composition from the national grid is not significantly different from European countries, the outcomes represent an important landmark. By adding up to 30% of H₂ by volume, the authors did not experience flashback, flame lift, or CO emissions above 400 ppm. Another conclusion was that the efficiency of the appliances tested only varied by 1–1.5% with 0 to 30% hydrogen added to the mixture. Although this paper has an important experimental part, the burners tested are an atmospheric type and are not suitable for condensing boilers, which makes the separate theoretical and experimental analysis of burning equipment involving the condensation of water vapor in the flue gas the most important.

THyGA Project, carried out by Shaffert J. et al. [6], was one of the most complex projects involving the testing of numerous pieces of combustion equipment on the impact of a hydrogen–natural gas mixture on the performance and emissions. The project members tested gas boilers, room sealed boilers, gas engines, gas turbines, heat pumps, ovens, and fryers, in order to highlight what functional changes occur when replacing natural gas with a mixture containing a percentage of hydrogen. Nonetheless, the experimental results, in terms of condensing boilers, are extremely limited, and thus there is a need for experimental and theoretical research to cover this lack of information. One of the project’s main conclusions was that the burning equipment behavior at high hydrogen participation in combustibles can vary substantially, depending on the system (for example, premixed or non-premixed system). A thermal power decrease of up to a maximum of 12% at 30 vol.% hydrogen admixture, compared to the operation with natural gas, was noted. Regarding CO and NO_x emissions, as well as flame temperature, different results were obtained. In some cases, the concentrations increased; in other cases, decreases were reported. As a direct result, further research is necessary in order to have a clearer and more complete overview on the process.

Suchovsky, C.J. et al. [7] have evaluated the burning appliance and equipment performance fed by hydrogen-enriched natural gas. There were reported no major operable problems and thermal power and CO₂ emissions decreased with increasing hydrogen content. CO and NO_x have not exceeded the permitted limits.

The effect of hydrogen admixture on possible flashbacks, burning velocities and thermal performances related to domestic appliance was investigated by Vries H et al. in [8]. The main conclusion was that 20% hydrogen admixture in comparison with pure methane reduces thermal power by 4.7%, and the end users of the appliance must be aware of this loss, bearing in mind to maximize the decarbonization.

Balanescu et al. [9] conducted a theoretical study in order to determine the effects of hydrogen-enriched methane combustion on latent heat recovery potential and on environmental impact of condensing boilers. For this purpose, the authors used the Ostwald diagram. The authors emphasized the potential rise of NO_x emissions with the increasing H₂ participation in the mixture, which is induced by the increase in the adiabatic flame temperature. At the same time, a CO₂ reduction by 56.15% was found together with an increase in the air excess ratio with 0.1.

Research has also been conducted on the performance of hydrogen-enriched methane combustion in a furnace, by Amaduzzi R. et al. [10]. The authors conducted a CFD numerical simulation of a furnace and concluded that the variation in the injector’s geometry has no significant impact on equipment functionality in the case of high H₂ contents in the mixture.

Evaluation on combustion of hydrogen/natural gas blends in a compression ignition engine has been made by Scrignoli F. et al. [11]. This theoretical study demonstrates that H₂ enrichment accelerates the combustion process and promotes its completion, strongly decreasing CO emissions. CO₂ specific emissions were also reduced (up to about 20% at 50 vol.% of H₂).

Another important aspect that must be considered when discussing hydrogen–natural gas mixtures is that, in this mixture, green hydrogen can be used, as indicated by Ingersoll [12]. By using this type of fuel, designated as RHYME, the effects are: to lower life-cycle carbon dioxide emissions and offer a globally cost-effective approach.

Lastly, the possibility of hydrogen injection into national gas networks must be carefully investigated in order to have a clear view regarding this aspect. Altfeld and Pinchbeck [13] stated that a 10% hydrogen admixture should not imply changes in gas network systems. On the other hand, the authors pointed out that it is not possible to specify a maximum hydrogen value that would generally be valid for all parts of European gas infrastructure and recommended a personalized study for each case.

The present paper intends to simulate the operation of small condensing boilers when hydrogen is mixed with standard gaseous fuel. The gaseous fuel chosen is methane (CH₄), and the hydrogen (H₂) volumetric participation in the mixture is considered to vary in the range of 0 to 20%. The condensing boilers operation will be numerically modeled by computational programs written by the authors in BASIC, and prior validated by comparing the computational results with the experimental results for a range of working regimes. The aim of the study is to highlight the functioning variations due to hydrogen participation in the gaseous combustible and determine if this solution is applicable and what functional changes would imply.

2. Materials and Methods

2.1. Experimental Stand and Measurements

The experimental stand—Figure 1 is located in the Centre of the Department of Thermal Sciences, part of the Technical University of Civil Engineering Bucharest, Romania.

The stand consists of a programmable and completely automated test rig with integrated software, which can assure real-time data acquisition and parameters calculation, measurements and evaluations of all operational parameters, energy and mass flows, environmental impact, and complete flue gases analysis.

In order to monitor the flue gases temperature, the quality of combustion, and also air temperatures and humidity in the endowment of the laboratory, a portable gas analyzer and portable thermo-anemometers (temperature range −20–60 °C, precision ±0.1 °C) are used. The water volumetric flow is measured with propeller flow meters (precision ±1% of the measured value)—Figure 2.

For the current investigation, two water-tube boiler constructive solutions were chosen, one with corrugated stainless-steel pipes and the other with aluminum finned pipes.

The two solutions were chosen based on the fact that they meet all the necessary criteria for carrying out this study:

• Are representative of a wide range of market-available condensing boilers constructive solutions for the household [14,15,16], having a thermal power range between 10 and 35 kW as stand-alone units, or up to 300 kW in packings. It is important to mention that the derived solutions from the studied main ones are very likely to behave the same as the ones analyzed. The only constructive solution that was not approached in the article was the fire-tube boiler. The reason is that it is rarely used in small thermal power condensing boilers, being mostly a solution employed for separate condensing units or medium and high thermal loads condensing boilers.
• The computational programs used for the numerical modelling are already validated [1,16] and so there is no danger of inappropriate modelling of heat and mass transfer phenomena; it is well known that the condensation process of water vapor from the flue gases to the pipe wall is very complex to model, especially when the heat transfer occurs at the same time (mass transfer with temperature variation in the limit layer, and species variation along the process); the temperature differences between the limit layer dew point and condensing surfaces tend to an average of 1 to 5 °C in the last condensing sections of the flue gases, just before exiting the boiler.
• The fact that this equipment is already present in most households in Romania makes the modification of gaseous fuel by H₂ mixing in the CH₄ more important, because the changes in flue gas content and thermal performance can then be extrapolated to obtain an important assessment at a national level.

The testing procedure was carried out according to European Standard 15502-1:2021 for which the Laboratory is accredited. In these types of testing, according to the Standard in question, the data volume is not important. What matters is to perform the measurements in a steady-state regime. Moreover, upon reaching the steady-state operation regime, 2 data sets have to be recorded. Each data set is recorded for a fixed interval of 10 min, during which time the monitored parameters (water and flue gas temperatures and flow, and flue gas concentration) must not fluctuate by more than ±0.5%. Considering the fact that sampling frequency of the measured data is 1 s, a total of 1200 measurement data pairs for each one of the boilers was recorded. The data used in this paper are the average of every measured parameter.

2.2. Numerical Modelling

The numerical modelling was implemented by the authors in the BASIC computational program. The decision to use such computational modelling was driven by the very essence of the BASIC programming method, i.e., a base program for the main calculus algorithm and special specific subroutines for the thermo-physical phenomena and for the physical properties of the thermal agents. Such an approach has two major advantages: first, the separate validation of the subroutines; and secondly, the validation of the main calculus algorithm once the subroutines are validated and the overall computational model is validated.

The authors’ vast experience of modelling the subroutines (for example, convection coefficient determination, radiation calculus, mass transfer module including dew point temperature–pressure correlation, and combustion and flue gases composition and physical properties) has been previously validated by direct comparison with the results obtained in laboratory testing.

By compiling the boilers’ specific geometry from the two thermal agents’ flow point of view and from the thermal heat and mass transfer point of view with the validated subroutines, two modelling programs were written: one for the corrugated pipe boiler (CORCOND) and another one for the finned pipe boiler (FINCOND).

The specific geometry of the corrugated pipes condensing boiler is presented in Figure 3, along with the main flow specific geometries for the flue gases (critical for determining the heat and mass transfer conditions).

It is important to mention that the corrugated pipe coil solution of the condensing boiler is one of the most efficient and demonstrated over the years not only exceptional thermal performances, but also functional durability, and has, along with the other boiler components, an outstanding overall working performance.

The second boiler type, with finned pipes, also has very good thermal performance. It is a broadly used solution, in various geometries, and it can be made from various materials. The constructive geometry selected for this study is presented in Figure 4, along with some important discretization details.

The furnace is represented by the volume between the premix plane surface radiation burner and the main finned pipes heat exchanger body. There are also lateral cooling pipes in the furnace, as presented in Figure 4. For the first convective finned pipes row, there is considered to be only heat transfer, due to the impossibility of draining the formed condensate. Considering its important thermal load, the first row is divided in two calculus sections: the upper part, and the lower part (also due to the fin geometry that differs between the two sections). The first convective body with condensation is the second row of finned pipes, and the second condensing body is the final plane pipes fascicle.

In both cases, the modelling program considered first the convective heat transfer, and after determining a range of parameters as fuel consumption, water temperatures, and flue gases temperatures, it moves to the second step of computation, over-imposing the condensation process. The condensation process will modify the flue gases composition and, consequently, the flue gases heat transfer conditions. The modelling program considers and computes those changes and their thermal implications and reassess the heat and mass transfer conditions (i.e., by modifying the partial pressure of the water vapor in the flue gases, the dew point, the mass transfer parameters, the flow speeds, the convective coefficients, etc.).

The computational program was based on the logical schematics of a verification calculus of a boiler. The main steps followed were:

• Setting the boiler construction details (lengths, diameters, number of pipes, etc.).
• Setting the combustible characteristics (composition, properties from the burning process point of view, thermal properties, physical properties).
• Defining the useful thermal power and the secondary agent temperatures.
• Burning process calculus.
• Furnace heat transfer calculus considering the combined radiation heat transfer from flame and flue gases and convective heat transfer. The furnace output temperature is calculated with Equation (1), [13].

  $${\mathrm{T}}_{\mathrm{F}}=\frac{{\mathrm{T}}_{\mathrm{ad}.}}{{\left(\mathrm{M}·\frac{{\mathrm{C}}_0·\mathsf{\xi }·{\mathrm{T}}_{\mathrm{ad}}^3·{\mathrm{S}}_{\mathrm{R}}}{\left(1-{\mathrm{q}}_{\mathrm{ext}}\right)·\mathrm{B}·{\mathrm{V}}_{\mathrm{g}}·{\mathrm{c}}_{\mathrm{pg}}}\right)}^{0.6}+1} \left[\mathrm{K}\right]$$

  (1)
  Furnace thermal load is determined using Equation (2), [13].

  $${\dot{\mathrm{Q}}}_{\mathrm{F}}=\mathrm{B}·{\mathrm{V}}_{\mathrm{g}}·\left(1-{\mathrm{q}}_{\mathrm{ext}}\right)·\left({\mathrm{c}}_{\mathrm{pg}}·{\mathrm{t}}_{\mathrm{g}}-{\mathrm{c}}_{\mathrm{pg}}·{\mathrm{t}}_{\mathrm{f}}\right) \left[\mathrm{W}\right]$$

  (2)
• Verification computation for heat transfer in each convective section determining the flue gases output temperature and the useful heat load corresponding for each discretization section. In this purpose, Mendeleev convective heat transfer coefficient—Equation (3)—was used.

  $${\mathrm{h}}_{\mathrm{c}}=\mathsf{\epsilon }·0.0263·\mathrm{C}·\frac{\mathrm{k}·{\mathrm{Pr}}^{0.35}}{{\mathrm{d}}_{\mathrm{i}}}·{\mathrm{Re}}^{0.8} \left[\frac{\mathrm{W}}{{\mathrm{m}}^2\mathrm{K}}\right]\hspace{1em}2300<\mathrm{Re}<\mathrm{10,000}$$

  (3)
• Radiation heat transfer coefficient is estimated according to Equation (4).

  $${\mathrm{h}}_{\mathrm{r}}={\mathrm{C}}_0·\frac{{\mathsf{\epsilon }}_{\mathrm{wall}}+1}{2}·{\mathsf{\epsilon }}_{\mathrm{g}}·{\mathrm{T}}_{\mathrm{gm}}{}^3·\left[\frac{1-\left(\frac{{\mathrm{T}}_{\mathrm{wall}}}{{\mathrm{T}}_{\mathrm{gm}}}\right)}{1-{\left(\frac{{\mathrm{T}}_{\mathrm{wall}}}{{\mathrm{T}}_{\mathrm{gm}}}\right)}^{3.6}}\right] \left[\frac{\mathrm{W}}{{\mathrm{m}}^2\mathrm{K}}\right]$$

  (4)

Total heat transfer coefficient to surface is the sum of the convection and radiation coefficients.

The flue gases exit temperature is determined through Equation (5) implementation into the computational program.

$${\mathrm{T}}_{\mathrm{fg},\mathrm{out}}={\mathrm{T}}_{\mathrm{wall}}+\left({\mathrm{T}}_{\mathrm{fg},\mathrm{in}}-{\mathrm{T}}_{\mathrm{wall}}\right)·{\mathrm{e}}^{-\frac{{\mathrm{h}}_{\mathrm{TOT}}·{\mathrm{S}}_{\mathrm{R}}}{\mathrm{B}·{\mathrm{V}}_{\mathrm{g}}·{\mathrm{c}}_{\mathrm{pg}}·\left(1-{\mathrm{q}}_{\mathrm{ext}}\right)}} \left[\mathrm{K}\right]$$

(5)

8.

Numerical determination of the condensing heat transfer:

-

Initial (input) flue gases temperature and composition.

-

Condensing heat transfer computation steps: condensing condition for the wall temperature with regard to flue gases water vapor saturation temperature; dimensionless criterions (invariants) specific for the mass transfer process (Schmidt, Reynolds, and Sherwood) calculation; determination of the condensing coefficient using Equation (6).

$${\mathrm{J}}_{\mathrm{MASS}}=2.157·{10}^{-3}·\frac{\mathrm{Sh}·{\mathrm{C}}_{\mathrm{dif}}}{\mathrm{FL}}·\left({\mathrm{p}}_{\mathrm{H}20-\mathrm{flue} \mathrm{gases}}-{\mathrm{p}}_{\mathrm{H}20-\mathrm{pipewall}}\right)·\frac{1}{{\mathrm{T}}_{\mathrm{gm}}}\left[\frac{\mathrm{kg}}{{\mathrm{m}}^2\mathrm{s}}\right]$$

(6)

-

Convection coefficient determination for the secondary agent.

-

Overall heat exchange coefficient determination considering sensible and latent heat transfer from the flue gases to the “wall”; surface extension geometry (for finned pipes only); surface extension efficiency (for finned pipes only); secondary agent heat transfer resistance.

-

Computation loop closing errors with iteration decision.

-

Final determination for the thermal and physical parameters.

For the finned pipes body boiler, the computational model validation was realized by direct comparison of computational results with experimental results obtained after testing on an experimental stand accredited by the Romanian National Accreditation Body.

The experimental and computational results, along with the calculated errors, for the boiler’s operation in a condensing regime at nominal thermal load, are presented in

Table 1. Experimental vs. computational results for finned pipes boiler.

	Measure Unit	Calculated Results (Modelling Program)	Measured Values (Experimental Stand)	Relative Error %
Inlet water temperature	[°C]	31.2 (input)	31.2
Outlet water temperature	[°C]	51.6 (input)	51.6
Combustible flow	[m3/h]	2.503 (input)	2.503
Excess air at the burner	[-]	1.26 (input)	1.26
Flue gas exit temperature	[C]	55.5	47.9	0.46 *
Total useful thermal load	[kW]	26.34	26.53	−0.72
Sensible useful thermal load	[kW]	24.42	24.56	−0.57
Condensing useful thermal load	[kW]	1.92	1.97	−2.54
Thermal efficiency (by respect to the Net calorific value)	[%]	105.8	106	−0.23

* Error related to the entire temperature span for the flue gases: theoretical burning temperature to chimney temperature.

.

By analyzing Table 1, it may be noticed that there is a good correlation between measured and calculated values, the error ranging between ±1% for the total and sensible heat and for thermal efficiency. The 3% error range for the condensing thermal load is acceptable, consisting of approximately 50 W, which falls in the precision range of the measuring facility.

The thermal efficiency does not include the pump and automation system operation and it is calculated as the ratio between useful thermal load and the fuel heat input, calculated by respect to the net calorific value. By reporting to the net calorific value, the thermal efficiency value increases above one unit.

For the corrugated pipe coil boiler, the computational model validation was also realized by direct comparison between computational results and experimental results, and is plotted in

Table 2. Experimental vs. computational results for corrugated pipes boiler.

	Measure Unit	Calculated Results (Modelling Program)	Measured Values (Experimental Stand)	Relative Error %
Inlet water temperature	[°C]	31.7 (input)	31.7
Outlet water temperature	[°C]	51.5 (input)	51.5
Combustible flow	[m3/h]	2.792 (input)	2.792
Excess air at the burner	[-]	1.17 (input)	1.17
Flue gas exit temperature	[°C]	35.5	37.1	−0.1 *
Total useful thermal load	[kW]	29.48	29.16	1.1
Sensible useful thermal load	[kW]	27.52	27.19	1.4
Condensing useful thermal load	[kW]	1.96	1.97	−0.5
Thermal efficiency (by respect to the Net calorific value)	[%]	105.1	104.9	0.2

* Error related to the entire temperature span for the flue gases: theoretical burning temperature to chimney temperature.

. The results were consistent with those obtained for the finned pipes boiler model validation.

The program calculates every time the boiler efficiency is in a purely convective functioning regime and makes the necessary corrections for the combustible flow. Thus, the combustible flow will be modified until the heat and mass transfer loops are satisfied in the imposed errors, and the boiler’s convective heat transfer load is the one set by input. We considered, mainly due to the fact the condensing process was the aim of this study, that it is best to set as a constant, the convective useful thermal load of the boiler for each case.

3. Results

As previously stated, the computations were performed, for each of the two boilers, for a set of convective heat loads, and kept constant for the various operating situations. The corrugated boiler was set for 27.5 kW and the finned pipe boiler was set for 24.0 kW.

The stoichiometric values correspond to a theoretical complete burning, without any excess air. For each boiler, the operating conditions were computed for three different H₂ concentrations in the gaseous combustible (pure methane—CH₄): 0% H₂ (reference case), 10% H₂ (intermediate case), and 20% H₂ (limit case).

The H₂ concentrations were chosen in this manner because a larger content of H₂ may determine faulty operation of the burner. This is due to the variation in the burning speed of the combustion mixture and the variation in the minimal ignition diameter that can generate backfires in the burner’s mixing fan. The main burning characteristics of the gaseous combustible in the three previously stated cases are presented in

Table 3. Burning process characteristics when hydrogen is infeed in pure methane.

				Parameter Variation (*) (%)
H2 Participation into the Combustible (%)	0	10	20	+20
Stoichiometric flue gases volume [Nm3/Nm3]	10.67	9.90	9.12	−14.5
Stoichiometric air volume [Nm3/Nm3]	9.52	8.81	8.10	−14.9
Net calorific value [kJ/Nm3]	35,800	33,299	30,798	−14.0
Water vapor specific volume [Nm3/Nm3]	2.15	2.04	1.93	−10.2
Carbon dioxide specific volume [Nm3/Nm3]	1	0.9	0.8	−20
Water vapor partial pressure in the flue gases [bar] (excess air 1.2)	0.171	0.175	0.180	+5.26
Carbon dioxide partial pressure in the flue gases [bar] (excess air 1.2)	0.0795	0.0772	0.0745	−6.29
Adiabatic theoretical burning temperature [°C] (excess air 1.2)	1770	1776	1782	<1
Dew point temperature in the flue gases [°C] (excess air 1.2)	56.77	57.24	57.77	+1.76

(*) Calculated for the entire domain of H₂ concentration and referred to the initial value (corresponding to H₂ = 0%).

.

One very important conclusion to be drawn after evaluating Table 3 is that the burning parameters, as air and flue gases stoichiometric volume, decrease along with the H₂ increasing concentration. The same trend is also valid for the net calorific value (or lower calorific power) the decrease reaching 15% for the 20% H₂ concentration.

What is also of great interest is the water vapor volume variation, which is also decreasing, but only by approx. 10%. Considering the decrease in the net calorific value at a higher rate than the water vapor concentration, it is to be expected that, due to fuel flow increasing (under the reserve of constant convective boiler efficiency at H₂ concentration variation), the final water vapor flow will also increase. This will generate a larger available water vapor flow for condensation.

Considering that the water vapor’s partial pressure also increases by approx. 5%, general beneficial condensing conditions are to be expected when H₂ concentration increases in mixture with CH₄. The dew point temperature in the flue gases with a 1.2 excess air coefficient converges to this conclusion, its value is rising with 1 °C.

The adiabatic theoretical burning temperature rises marginally, by less than 1%, having practically no influence over the heat transfer conditions or the burner temperature field (no dangerous high temperatures that might damage the burner’s stabilization surface).

For the computational results, a very large number of parameters were determined and consequently analyzed; the entire process of heat and mass transfer for each separate boiler surface being investigated and presented via the computational modelling of each boiler. A total of 270 functional parameters were investigated for different functioning regimes (different temperatures, velocities, heat fluxes, and heat and mass transfer coefficients) for CORCOND, and 130 parameters for FINCOND modelling program, respectively.

Nevertheless, the interest of the current study being the functional behavior of the boiler from the user’s point of view, and mainly focusing on the condensing process, only a narrow selection of the most defining parameters is presented and further commented on.

The functioning parameters for the corrugated coil and finned pipes condensing boilers are plotted in

Table 4. Functioning parameters of corrugated coil condensing boiler.

Parameter	Boiler with Corrugated Pipe Coil (Thermal Load with No Condensation 27.5 kW)			Parameter Variation (*) (%)
H2 Participation into the Combustible [%]	0	10	20	+20
Flue gas temperature at chimney inlet section [°C]	33.8	33.8	33.7	0
Boiler efficiency when not in condensation regime [%]	99.2	99.2	99.2	0
Gaseous fuel consumption (hourly value) [Nm3/h]	2.787	2.996	3.239	+16.2
Thermal load with no condensation [kW] (purely convective thermal load)	27.50	27.50	27.50	0
Condensing (useful) thermal load [kW]	1.848	1.897	1.954	+5.25
Total (useful) thermal load [kW]	29.35	29.40	29.45	~0
Boiler efficiency in condensation regime [%]	105.1	105.3	105.45	~0
Initial water vapor flow with the flue gases [kg/h]	4.82	4.91	5.00	+3.7
Useful condensate flow [kg/h]	2.81	2.89	2.98	+6.1
Condensing efficiency [%] (condensate from total water vapor inlet)	58.4	58.8	59.5	+1.9
CO2 mass flow in the flue gases [kg/h]	5.47	5.30	5.09	−7.03
Annual CO2 emissions for standard functioning of the boiler [tons] (estimated equivalent of 2800 h/year at nominal output)	15.3	14.8	14.3	−1000 kg/y (**)

(*) Calculated for the entire domain of H₂ concentration and referred to the initial value (Corresponding to H₂ = 0%); (**) variation in absolute value (kg/year).

and

Table 5. Functioning parameters of finned/simple pipes condensing boiler.

Parameter	Boiler with Finned/Simple Pipes (Thermal Load with No Condensation 24.0 kW)			Parameter Variation (*) (%)
H2 Participation into the Combustible [%]	0	10	20	+20
Flue gas temperature at chimney inlet section [°C]	64.8	63.8	63.6	0
Boiler efficiency when not in condensation regime [%]	97.7	97.7	97.7	0
Gaseous fuel consumption (hourly value) [Nm3/h]	2.499	2.686	2.903	+16.2
Thermal load with no condensation [kW] (purely convective thermal load)	24.00	24.00	24.00	0
Condensing (useful) thermal load [kW]	1.746	1.792	1.845	+5.67
Total (useful) thermal load [kW]	25.75	25.79	25.85	~0
Boiler efficiency in condensation regime [%]	103.6	103.8	104.1	~0
Initial water vapor debit with the flue gases [kg/h]	4.32	4.41	4.50	+4.2
Useful debit of condensate [kg/h]	2.65	2.73	2.82	+6.4
Condensing efficiency [%] (condensate from total water vapor inlet)	61.6	62.1	62.6	+1.6
CO2 mass flow in the flue gases [kg/h]	4.91	4.75	4.56	−7.07
Annual CO2 emissions for standard functioning of the boiler [tons] (estimated equivalent of 2800 h/year at nominal output)	13.7	13.3	12.8	−900 kg/y (**)

(*) calculated for the entire domain of H₂ concentration and referred to the initial value (Corresponding to H₂ = 0%); (**) variation in absolute value [kg/year].

, respectively.

4. Discussion

The authors draw important conclusions from the detailed analysis of the data presented in comparative Table 4 and Table 5.

The gaseous fuel consumption must increase as a consequence of constant efficiency and calorific value decrease. This fact may be observed in the above-mentioned tables. The fuel flow increases by approx. 16% over the entire interval in order to maintain useful thermal power.

The useful condensing thermal load increases along with the H₂ concentration; the total effect being around 5%. In spite of this, given that the condensing thermal load (latent thermal load) is less than 10% of the convective thermal load (sensible thermal load), the result is that a general influence over the total thermal load is less than 0.5%, which is negligible.

As a consequence, the boiler total thermal load remains practically constant over the H₂ variation range. This trend also applies to the overall boiler efficiency due to the calculation method, which reports the total useful thermal load (including condensation) to the energy flow brought by the combustible flow (without condensation latent heat flow).

As expected, the inlet water vapor flow and condensed water flow produced by the boilers increase. The condensation efficiency of the boilers remains practically constant, averaging 60%.

CO₂ flow, the main target of the study, decreases by around 7%, and when H₂ is in mixture with CH₄, it reaches 20%. This is in accordance with studies from [3,6,7,8,9,10], which also highlighted the lowering of CO₂ in flue gases.

The CO₂ emission reduction for the modelled boilers was calculated considering an average operation of the boiler for one year.

The operating conditions were determined over a set (historical) functioning diagram describing the way the thermal load of the boiler (in percentage from nominal thermal load) is required over a heating season. We considered that the partial loads were in fact generated by partial time boiler’s operation at a nominal thermal load. This fact is obviously not true, especially for a condensing boiler with a premix burner (achieving down to 15% of the nominal thermal load modulation), but there is no difference for the purpose of the estimation. In fact, what was targeted was an overall thermal demand for a standardized heating season, expressed as hours of nominal operation during the heating season.

The boiler’s domestic hot water production was estimated based on the delivery schedules presented in the European Standards for boiler hot water production energy classification. The total daily hot water production was converted into hours of daily functioning and, by considering the number of days in a year, a total functioning time for the boiler was found for domestic hot water production.

Summing the yearly functioning of the boiler for heating purposes with the yearly functioning of the boiler for hot water production purpose, a total yearly functioning time, at nominal thermal output, was determined.

As an example, considering these premises, and based on an estimation of 700 K boilers functioning in the U.K. in 2020 [17] having thermal power lower than 100 kW, and assuming the working conditions previously presented, a total estimated reduction of: 0.9 (to/year/boiler) × 700 K (boilers) = 630,000-ton CO₂/year can be deducted.

In Romania, the Eco-Design boilers minimum performance standards became mandatory starting in September 2015 [18], and has led to a shift towards condensing boilers starting in 2016. Nonetheless, a lot of domestic water heating devices are still noncondensing boilers fed by natural gas or solid-type fuel. If, for example, in 2014 before the Eco-Design Directive implementation, 90% of the wall-hung condensing boilers sold were of the non-condensing type, a 180-degree change took place leading to 97.5% of condensing type boilers being sold in 2021.

According to [19], the Romanian condensing boiler market is expected to have a slight increase from one year to the next, roughly 2%, starting with the end of the COVID-19 pandemic. Considering the current market reports and future predictions we can deduct the number of gas wall-hung condensing boilers that are going to be sold between the years 2019 and 2024. Thus, 1.415 million units resulted, of which 81% have a thermal power of just about 25 kW and 19% of 70 kW, respectively.

By assuming the same value considered above: 0.9 (to/year/boiler) × 283 K boilers (average value obtained by dividing 1.415 million to 5 years) = a 254,700-ton CO₂/year reduction resulted. It should be noted that this value has only been calculated taking into account new appliances, but it should be borne in mind that the use of hydrogen-enriched gaseous fuel in existing appliances will also contribute to Romania’s decarbonization.

For further expanding the area of knowledge, the authors will test in the next period certain condensing boilers available on the Romanian market, both with natural gas from the national grid and with G222 gas, which is a mixture of 23% hydrogen and 77% methane. The experimental tests are going to be conducted in similar working conditions and will be further compared with the outcome of the numerical modelling program presented in this article. In addition, experimental measurements are going to be conducted on a fire-tube boiler with a separate condensing stage and, subsequently, a particular modelling program for this type will be conducted.

5. Conclusions

The main target of the study was to determine if, from the beneficiary point of view, the modification of fuel composition from 100% CH₄ to a mixture of up to 20% H₂ and 80% CH₄ will imply operational difficulties, when using small condensing boilers (water-tube corrugated coil and finned pipes boilers). This measure may be useful even in the current international context in which the volume of gas available and imported by the European Union is expected to decrease.

The study conclusion is that an increase in the combustible flow with 16% will compensate the maximum H₂ concentration situation with no other implications over the boiler’s thermal efficiency, either from sensible, latent, or overall point of view, while also maintaining the useful boiler’s thermal power output. At the same time, for the previously specified domain, the boilers’ CO₂ emissions will decrease with nearly 7%.

Considering an example of boiler’s conventional yearly functioning of 2800 h at nominal thermal load, an estimate reduction of 900 to 1000 kg of CO₂ emitted within the flue gases is attainable. As a main conclusion, from the boiler’s operating perspective, the reduction in CO₂ emissions by replacing CH₄ with H₂ is perfectly feasible under the conditions outlined in the present study.

The novelty of the study comes from the fact that the literature up to date do not approach thoroughly the effect of mixtures made up from 0 ÷ 20% H₂ and 80% CH₄ on condensing boilers. Particularly, experimental and theoretical studies have been conducted aiming at other type of burning equipment or boilers without condensing. The current paper combines experimental measurement with modelling programs in order to have a much broader and clearer view of the process.

This paper provides for the first time in the available literature important conclusions that can be successfully used in further research. A 254,700-ton CO₂/year reduction may result if all newly purchased wall-hung condensing boilers from Romania between 2019 and 2024 would be fed with a 20% H₂ and 80% CH₄ gaseous mixture.

Moreover, another novelty is represented by the employment of experimentally obtained data, measured in a European Certified Laboratory in order to develop and validate a computational modelling program, which can be further used in order to evaluate the boiler’s response to various changes in operational parameters.