*3.2. Mineralization of CO<sup>2</sup> from BF Top Gas Using Magnesium Hydroxide*

For the assessment of the feasibility of integrated mineralization of CO<sup>2</sup> from BF top gas using Mg(OH)<sup>2</sup> produced from magnesium silicate rock and CO/water shift reaction supplying CO<sup>2</sup> from CO, a BF top gas composed of 20.7 vol% CO2, 20.7vol% CO, 3.8 vol% H2, 6 vol% H2O, and 48.8 vol% N<sup>2</sup> was assumed [32]. The flow sheet for the Aspen Plus model used for the simulations is shown in Figure 13, with SOLIDIN as the Mg(OH)<sup>2</sup> supply that eventually gives solid products COLDCARB and BF top gas stream feed GASIN that is compressed to process conditions eventually giving product gas FINALGAS after expansion. Three gas/solid heat exchangers, a gas/solid (chemical equilibrium) reactor, and a gas/solid separator made up the final process equipment [20]. This model may be readily added to or integrated with Aspen Plus models for BFs or other sections of iron- and steelmaking processes, such as for example given in [33,34]. *Metals* **2020**, *10*, 342 11 of 16 N2 was assumed [32]. The flow sheet for the Aspen Plus model used for the simulations is shown in Figure 13, with SOLIDIN as the Mg(OH)2 supply that eventually gives solid products COLDCARB and BF top gas stream feed GASIN that is compressed to process conditions eventually giving product gas FINALGAS after expansion. Three gas/solid heat exchangers, a gas/solid (chemical equilibrium) reactor, and a gas/solid separator made up the final process equipment [20]. This model may be readily added to or integrated with Aspen Plus models for BFs or other sections of iron- and steelmaking processes, such as for example given in [33,34].

**Figure 13.** The Aspen Plus flow sheet for BF top gas processing.

**Figure 13.** The Aspen Plus flow sheet for BF top gas processing. A gas feed of 1 kmol/s with the abovementioned composition was used with an excess of 17% Mg(OH)2 for the conversion, corresponding to 0.3 mol Mg per mol input gas. Temperature and pressure for the reactor were varied in the ranges of 400–500 °C and 40–100 bar, respectively. This was based on preliminary calculations using Gibbs energy minimization and the earlier work at ÅA on Mg(OH)2 carbonation using CO2-containing exhaust gases from other processes. The simulation results showed that in all cases Mg(OH)2 was converted not only to MgCO3 but also to MgO. At 40 bar, the carbonation efficiency dropped from 93% at 400 °C to 66% at 460 °C. For 100 bar, the carbonation efficiency changed to 97% at 400 °C and 87% at 460 °C. The values for the CO + CO2 conversion efficiency were similar to the values for the carbonation efficiency. Interestingly, there was little effect of process temperature and pressure on the amount of H2 produced, which eventually left the system at approximately six times the amount entering with the feed gas. One feature of the process is that with increasing temperature the CO/water shift reaction equilibrium caused CO to be more stable while the corresponding higher H2O partial pressure could not prevent MgO formation. Experimental work under the preferable process conditions can be a next step, and the required A gas feed of 1 kmol/s with the abovementioned composition was used with an excess of 17% Mg(OH)<sup>2</sup> for the conversion, corresponding to 0.3 mol Mg per mol input gas. Temperature and pressure for the reactor were varied in the ranges of 400–500 ◦C and 40–100 bar, respectively. This was based on preliminary calculations using Gibbs energy minimization and the earlier work at ÅA on Mg(OH)<sup>2</sup> carbonation using CO2-containing exhaust gases from other processes. The simulation results showed that in all cases Mg(OH)<sup>2</sup> was converted not only to MgCO<sup>3</sup> but also to MgO. At 40 bar, the carbonation efficiency dropped from 93% at 400 ◦C to 66% at 460 ◦C. For 100 bar, the carbonation efficiency changed to 97% at 400 ◦C and 87% at 460 ◦C. The values for the CO + CO<sup>2</sup> conversion efficiency were similar to the values for the carbonation efficiency. Interestingly, there was little effect of process temperature and pressure on the amount of H<sup>2</sup> produced, which eventually left the system at approximately six times the amount entering with the feed gas. One feature of the process is that with increasing temperature the CO/water shift reaction equilibrium caused CO to be more stable while the corresponding higher H2O partial pressure could not prevent MgO formation. Experimental work under the preferable process conditions can be a next step, and the required suitable equipment (e.g., a pressurized fluidized bed as in [11]) is available at ÅA.

suitable equipment (e.g., a pressurized fluidized bed as in [11]) is available at ÅA. The study was finalized by making an analysis of energy input/output and exchanger duties, with results as shown below in Figure 14. It can be seen that power requirements may be compensated for by heat that is produced by the overall process with the surroundings temperature T° and the process unit temperatures Ti and the exergies of the combined heat outputs, Qi, calculated by ΣQi∙(1 − T°/Ti), was larger than the required netto power input. See [20] for more details on these results and [30] for the use of exergy analysis, which based on the second law of thermodynamics The study was finalized by making an analysis of energy input/output and exchanger duties, with results as shown below in Figure 14. It can be seen that power requirements may be compensated for by heat that is produced by the overall process with the surroundings temperature T◦ and the process unit temperatures T<sup>i</sup> and the exergies of the combined heat outputs, Q<sup>i</sup> , calculated by ΣQ<sup>i</sup> ·(1 − T ◦ /Ti ), was larger than the required netto power input. See [20] for more details on these results and [30] for the use of exergy analysis, which based on the second law of thermodynamics allows recalculating energy flows of different forms (here are power and heat) into the equal denominator of useful work.

allows recalculating energy flows of different forms (here are power and heat) into the equal

denominator of useful work.

*Metals* **2020**, *10*, 342 12 of 16

**Figure 14.** Energy input and output and heat exchanger duties for process conditions as indicated for the process shown in Figure 13. Power in/out and exergy of reaction heat (**a**) and heat of reaction and heat duties of heat exchangers (**b**). **Figure 14.** Energy input and output and heat exchanger duties for process conditions as indicated for the process shown in Figure 13. Power in/out and exergy of reaction heat (**a**) and heat of reaction and heat duties of heat exchangers (**b**). **(a) (b) Figure 14.** Energy input and output and heat exchanger duties for process conditions as indicated for

the process shown in Figure 13. Power in/out and exergy of reaction heat (**a**) and heat of reaction and

#### *3.3. Metallic NP Production Using an Arc Discharge Route 3.3. Metallic NP Production Using an Arc Discharge Route 3.3. Metallic NP Production Us* heat duties of heat exchangers ( *ing an Arc Discharge Route* **b**).

#### 3.3.1. Specific Electricity Consumption (SEC) 3.3.1. Specific Electricity Consumption (SEC) 3.3.1. Specific Electricity Consumption (SEC)

Analyzing the SEC for the production of metallic NPs leads to a significant penalty resulting from the NP size dependence and the energies (enthalpies) of melting/solidification, evaporation/condensation and the temperatures for these. Melting and evaporation (volatilization) requires heats Qm and Qv at temperatures Tm and Tv, respectively, while later the heat released by condensation Qc is smaller than Qv for condensation temperature Tc < Tv and the heat released by solidification Qs is smaller than Qm for solidification temperature Ts < Tm. Input exergies were significantly larger than output exergies calculated as Q∙(1 **–** T°/T) for the heating and cooling processes involving the material that produces NPs. Analyzing the SEC for the production of metallic NPs leads to a significant penalty resulting from the NP size dependence and the energies (enthalpies) of melting/solidification, evaporation/condensation and the temperatures for these. Melting and evaporation (volatilization) requires heats Q<sup>m</sup> and Q<sup>v</sup> at temperatures T<sup>m</sup> and Tv, respectively, while later the heat released by condensation Q<sup>c</sup> is smaller than Q<sup>v</sup> for condensation temperature T<sup>c</sup> < T<sup>v</sup> and the heat released by solidification Q<sup>s</sup> is smaller than Q<sup>m</sup> for solidification temperature T<sup>s</sup> < Tm. Input exergies were significantly larger than output exergies calculated as Q·(1 – T◦ /T) for the heating and cooling processes involving the material that produces NPs. Analyzing the SEC for the production of metallic NPs leads to a significant penalty resulting from the NP size dependence and the energies (enthalpies) of melting/solidification, evaporation/condensation and the temperatures for these. Melting and evaporation (volatilization) requires heats Qm and Qv at temperatures Tm and Tv, respectively, while later the heat released by condensation Qc is smaller than Qv for condensation temperature Tc < Tv and the heat released by solidification Qs is smaller than Qm for solidification temperature Ts < Tm. Input exergies were significantly larger than output exergies calculated as Q∙(1 – T°/T) for the heating and cooling processes involving the material that produces NPs.

For copper NPs, the results are illustrated by Figure 15, showing the SEC for several NP sizes as a result of exergy input for evaporation not being returned as exergy of condensation and exergy input for melting not being returned as exergy of solidification. This comes on top of the exergy needed for producing increased surface energies of atoms in NPs compared to those of atoms inside the material. It was clearly shown that the evaporation/condensation exergy consumption is by far the most important one of the three processes, which come on top of the SEC that arises from pumping around the carrier gas [14,29]. The values given in Figure 15 are several orders of magnitude lower than the 170 kWh/kg reported for experimental production of 79 nm Cu [29]. **0.3** For copper NPs, the results are illustrated by Figure 15, showing the SEC for several NP sizes as a result of exergy input for evaporation not being returned as exergy of condensation and exergy input for melting not being returned as exergy of solidification. This comes on top of the exergy needed for producing increased surface energies of atoms in NPs compared to those of atoms inside the material. It was clearly shown that the evaporation/condensation exergy consumption is by far the most important one of the three processes, which come on top of the SEC that arises from pumping around the carrier gas [14,29]. The values given in Figure 15 are several orders of magnitude lower than the 170 kWh/kg reported for experimental production of 79 nm Cu [29]. For copper NPs, the results are illustrated by Figure 15, showing the SEC for several NP sizes as a result of exergy input for evaporation not being returned as exergy of condensation and exergy input for melting not being returned as exergy of solidification. This comes on top of the exergy needed for producing increased surface energies of atoms in NPs compared to those of atoms inside the material. It was clearly shown that the evaporation/condensation exergy consumption is by far the most important one of the three processes, which come on top of the SEC that arises from pumping around the carrier gas [14,29]. The values given in Figure 15 are several orders of magnitude lower than the 170 kWh/kg reported for experimental production of 79 nm Cu [29].

**Figure 15.** Specified electricity consumption (SEC) for copper NP production considering surface energy and losses due to melting/solidification and evaporation/condensation. **Figure 15.** Specified electricity consumption (SEC) for copper NP production considering surface energy and losses due to melting/solidification and evaporation/condensation.

#### 3.3.2. LCA of NP Production such as silver NPs in cotton used in hospitals and copper NPs in water, so as to give a nano-cooling

3.3.2. LCA of NP Production

Besides the energy efficiency analysis, a wider analysis of the environmental footprint was made of metallic NP production via the dry, arc discharge process versus more conventional wet, aqueous solution metallic oxide salt reduction. For copper NPs, the results are given in Figure 16, illustrating that the BUONAPART-E concept is preferable only if more of the input metal material is obtained as NP products rather than remaining as deposits inside the production facility. Although NP metal production and LCA results were reported, metal NP yield needs to be improved [24]. fluid. These assessments can be found elsewhere [14,35] and showed great potential and special properties of consumer product materials containing metallic NPs. The increased demand for NP materials obviously calls for scale-up of production routes, while at the same time negative impacts need to be addressed, as was the objective of the work reported here. Note, however, that the LCA studies were cradle-to-gate assessments, i.e., from producing metal from ore to products containing NP particles leaving production. Lack of data on end-of-life product handling makes a full cradle-togate LCA impossible, although a first step in this direction was recently presented [36].

*Metals* **2020**, *10*, 342 13 of 16

Besides the energy efficiency analysis, a wider analysis of the environmental footprint was made of metallic NP production via the dry, arc discharge process versus more conventional wet, aqueous solution metallic oxide salt reduction. For copper NPs, the results are given in Figure 16, illustrating that the BUONAPART-E concept is preferable only if more of the input metal material is obtained as NP products rather than remaining as deposits inside the production facility. Although NP metal

The LCA studies on metal NP production were expanded to products that contain metallic NPs,

**Figure 16.** Life cycle impact (LCI) comparisons of dry, arc discharge NP production and chemical reduction methods for copper particles for four impact categories: Human health (**a**), Ecosystem **Figure 16.** Life cycle impact (LCI) comparisons of dry, arc discharge NP production and chemical reduction methods for copper particles for four impact categories: Human health (**a**), Ecosystem quality (**b**), Climate change (**c**), Resources (**d**). See Figure 12 for the explanations of abbreviations.

quality (**b**), Climate change (**c**), Resources (**d**). See Figure 12 for the explanations of abbreviations.

**4. Conclusions**  The work reported addresses modern trends seen in development of more sustainable process routes for iron- and steelmaking as well as nonferrous materials and products in which these are used. CO2 emissions reduction is obviously on top of many industrial production agendas together with energy efficiency, followed by water use and waste and by-product disposal. This faces the facts of limited resources and an increasing need for more circular economies for materials, resources, and processes that allow producing them. Experimental methods combined with modern theoretical tools such as exergy assessment for energy efficiency analysis and LCA give quantitative information on The LCA studies on metal NP production were expanded to products that contain metallic NPs, such as silver NPs in cotton used in hospitals and copper NPs in water, so as to give a nano-cooling fluid. These assessments can be found elsewhere [14,35] and showed great potential and special properties of consumer product materials containing metallic NPs. The increased demand for NP materials obviously calls for scale-up of production routes, while at the same time negative impacts need to be addressed, as was the objective of the work reported here. Note, however, that the LCA studies were cradle-to-gate assessments, i.e., from producing metal from ore to products containing NP particles leaving production. Lack of data on end-of-life product handling makes a full cradle-to-gate LCA impossible, although a first step in this direction was recently presented [36].

#### whether improved processes or products actually are an improvement vis-à-vis profitability. Three **4. Conclusions**

examples were given in this paper. Slag2PCC is a proven concept developed in Finland for valorization of steelmaking (BOF) slags, binding CO2 while producing high-value PCC. Scale-up and commercialization is ongoing, after the The work reported addresses modern trends seen in development of more sustainable process routes for iron- and steelmaking as well as nonferrous materials and products in which these are used. CO<sup>2</sup> emissions reduction is obviously on top of many industrial production agendas together with energy efficiency, followed by water use and waste and by-product disposal. This faces the facts of limited resources and an increasing need for more circular economies for materials, resources, and processes that allow producing them. Experimental methods combined with modern theoretical tools such as exergy assessment for energy efficiency analysis and LCA give quantitative information on whether improved processes or products actually are an improvement vis-à-vis profitability. Three examples were given in this paper.

Slag2PCC is a proven concept developed in Finland for valorization of steelmaking (BOF) slags, binding CO<sup>2</sup> while producing high-value PCC. Scale-up and commercialization is ongoing, after the lab-scale proof-of-concept was demonstrated followed by successful operation of a pilot plant. The LCA results showed that water use may be a critical factor.

CO<sup>2</sup> (and CO) from/in BF top gas can be converted with Mg(OH)<sup>2</sup> that can be produced from abundant serpentinite rock to MgCO3, hydrogen and steam, integrated with CO/water shift. The process simulation studies showed under the process conditions (pressure: 40 bar; temperature: 400 ◦C) good conversion levels were obtained, yet to be experimentally verified.

The dry production of metallic NPs using high-voltage (arc discharge) evaporation has many benefits, including life cycle impact, compared to with wet methods for NP production, and appears to allow for easier scale-up to production levels of kilograms per hour. The LCA calculations showed the overall benefit from an environmental footprint of metallic NP production as well as the production of products that, with very small amounts of NP, have a variety of beneficial properties.

### **5. Patents**

The slag2PCC concept described in this paper has been patented under Finnish patent 122348 and US patent 8603428.

**Author Contributions:** The author presented this work at the International Process Metallurgy Symposium IPMS2019 event in Espoo Finland 5–6 November 2019 and produced this manuscript based on that presentation.

**Funding:** The development work of the slag2PCC concepts was possible, during 2005–2016 with primarily Tekes funded projects Slag2PCC, Slag2PCC+ and Cleen Oy/Clic Oy CCSP and Finland's Graduate School for Chemical Engineering (2010–2014). The work on metal nanoparticulates was conducted during 2012–2016 funded by EU FP7 project BUONAPART-E (grant agreement: 280765) and Finland's earlier Graduate School of Energy Engineering and Systems (2012–2015).

**Acknowledgments:** Daniel Lindberg of Aalto University, Espoo Finland is acknowledged for inviting the author to present the work at International Process Metallurgy Symposium IPMS2019, and Lauri Holappa of Aalto University is acknowledged for the invitation to produce this manuscript. The author acknowledges his earlier co-workers, currently postdoctoral researchers in most cases, for intensive and productive cooperation, a great deal of which is summarized in this paper.

**Conflicts of Interest:** The author declares no conflicts of interest.
