**2. E**ff**ect of HTC Operating Conditions**

It is well established that the HTC treatment induces some major transformations inside the biomass particles, in accordance with the severity of the process that is performed. From an operative point of view, the three main parameters that can be varied in a HTC process are temperature, residence time and biomass-to-water ratio. It was found that each of these variables can differently influence the intermediate reaction and the output products in terms of yields and composition [33,36]. The evolution of the main biomass components (lignin, cellulose and hemicellulose) in relation to the reaction severity was recently studied and modelled [37,38]. Heidari et al. [37] found almost negligible interactions between lignin, cellulose and hemicellulose on the final hydrochar properties, while cellulose was mostly affected by reaction severity. Authors also conducted an evaluation on the optimal proportion between biomass components to increase carbon content and calorific value of hydrochar, finding that a cellulose, hemicellulose and lignin content respectively of 40%, 35% and 25%, represents the optimal starting conditions. Another main parameter that can significantly influence the composition of the final hydrochar is the origin of the biomass. It is well known that ligno-cellulosic biomass is mainly composed of lignin, cellulose and hemicellulos, but the extent of their proportion in the overall biomass composition is strictly related to the nature of the material considered. At the same time each of these components undergoes degradation processes at different extent in accordance to the process conditions [39,40]. The first component that is subjected to hydrolyzation is hemicellulose, which starts to degrade at temperatures of 180–200 ◦C [41], followed by cellulose depolymerization (above 210 ◦C) [40] and softening of lignin [42]. Complete conversion of hemicellulose for reaction temperature of 180 ◦C and residence time of 4 h was reported by Olszewski et al., when treating brewer's spent grains [43]. Cellulose was found to be the most difficult component to be degraded, as reported

by Titirici [31]. According to the author, cellulose starts to be converted above 210 ◦C according to two different pathways, as depicted in Figure 2 [31]. In their experiments, Olszewski et al. [43] and Volpe et al. [40] reported a complete cellulose conversion at a reaction temperature of 260 ◦C.

**Figure 2.** Cellulose conversion pathways, as proposed in Titirici, 2012.

Due to its very complex structure, lignin undergoes decomposition under a wider temperature range with respect to hemicellulose and cellulose [44]. The deterioration starts at around 200 ◦C, leading to the disruption of highly reactive alkyl-aryl-ether bonds, which further react with other intermediate compounds to form highly stable products [45]. Volpe et al. [19], when performing thermos gravimetric analysis (TGA) on olive mill waste hydrochar, found that while cellulose peak is significantly reduced for HTC temperature of 250 ◦C, the peak associated to lignin increases [44]. Through the same analysis, authors also confirmed the production of lower molecular compounds and/or tarry products coming from the decomposition of lignin. Indeed, a new peak in TGA curve at around 250 ◦C was found for the sample produced at 220 and 250 ◦C. From these considerations, it can be deduced that different reaction temperatures lead to different compositions in terms of cellulose, hemicellulose and lignin. The study conducted by Kim et al. [46] reported the energy retention efficiency (ERE) for each of these components in relation to reaction temperature. Indeed, as previously reported, as HTC temperature is raised, the calorific value increases as well, but the mass yield is reduced. It is therefore of paramount importance to define the best conditions to maximize the ERE, in order to obtain a valuable biochar for energy purposes. The results of this study showed that ERE for lignin and cellulose was maximized, respectively, at 200 ◦C and 220 ◦C, while the best temperature for the conversion of biomass into high energetic medium appeared to be 220 ◦C.

The higher thermal stability of lignocellulosic against non-lignocellulosic biomass was also confirmed by Zhuang et al. [47], testing herb tea waste (lignocellulosic) and penicillin mycelial waste (non-lignocellulosic). The analysis performed on the hydrochars produced showed a superior thermal stability for the tea waste, due to the presence of hemicellulose and lignin which start to degrade above 210 ◦C, while polysaccharide and protein present in the penicillin mycelial waste were hydrolyzed at temperatures lower than 180 ◦C. Finally, some considerations should be addressed regarding the influence of pressure, particle size, and the possibility of water recirculation within the system. With respect to the first parameter, the autogenous pressure that is developed inside the reactor maintains the water in the liquid state, favoring its ability to dissolve polar compounds [48]. Higher pressure levels result in smaller hydrochar dimensions and higher pore volume, together with a mild increase in the HHV. Only when pressure is raised above 100 MPa, significant increases in the HHV can be achieved, but the higher cost of the reactor can limit the application [48]. Regarding particle size and water recirculation, Heidari et al. [49] found that when biomass dimension was raised from <0.25 mm to within the range of 1.25–3 mm, a slight increase in the mass yield was observed,

but the calorific value was lower. This condition is explained by the more severe conversion that biomass endures since water can more easily penetrate and reach the inner core of the biomass particle. Water recirculation affects almost only the mass yield, increasing it after the first recycle of about 12%, while the contribution to HHV is almost negligible, since it was raised by just 2%.

Proximate and ultimate analysis have been widely used to characterize the hydrochars properties. Proximate analysis is used to get an overall picture of the char composition in terms of fixed carbon (FC), volatile matter (VC), ash and moisture content (M), while the ultimate analysis give the weight percentages of their major elements (C, H, N, S, O). Although not all the reactions involved in HTC treatment are well known, the general evolution pathways of the main biomass elements are fairly well understood. As long as reaction severity is increased, most of the volatile compounds are released, increasing carbon content and heating value. The main reactions occurring are dehydration and de-hydrogenation which imply, respectively, the removal of hydroxyl groups and the rupture of the of the long chain carboxylic acids [50]. Different studies evaluated the effect of residence time and reaction temperature on the final hydrochar yields and properties [4,17,51,52]. The results of the studies assert that the main contribution to biomass degradation and increase in the calorific value is reaction temperature rather than residence time [53]. Lucian et al. [51] reported an increase in the heating value of olive trimming from 22.6 to 27.8 MJ kg−<sup>1</sup> when reaction temperature was raised from 180 to 250 ◦C. At the same time, increasing the residence time from 1 to 6 h led to a maximum increase in HHV of +2 MJ kg<sup>−</sup>1, at the temperature of 220 ◦C, passing from 24.7 to 26.7 MJ kg<sup>−</sup>1. The increase in the calorific value is mainly due to the combined effect of the removal of oxygen and volatile compounds and the parallel rise in carbon content. The evolution of the concentration of carbon, oxygen and hydrogen is usually represented through the van Krevelen diagram in H/C and O/C ratios which are plotted. It is fairly stated that an increase in the hydrothermal carbonization temperature boosts dehydration reactions, removing oxygen from the initial biomass structure, resulting in a decrease in the O/C ratio. The evolution of the hydrochar composition is represented in the van Krevelen diagram through a shift toward lower values of O/C, characteristic of highly carbonaceous fuels like lignite or coals. Ulbrich et al. stated that the removal of oxygen for low HTC temperature is only mildly affected by residence time, while in the temperature range of 230–280 ◦C a more pronounced connection between oxygen reduction and residence time was observed [52]. Figure 3 displays the van Krevelen diagrams for three biomass sources reported in the literature: a sugar rich waste, namely rotten apples and coming from the fruit industry, a lignocellulosic residue (olive trimming) and a protein based biomass (Penicillin mycelial waste) [1,51,54]. It can clearly be seen that each waste endures a specific evolution pathway, in relation to its initial structure and composition. Evidences of the different transformations that biomass endures during HTC, in relation to its origin process parameters, are reported in Table 1. Lignocellulosic material, like olive trimming, presents higher thermal stability than proteins and polysaccharides, resulting in higher mass yield (approx. 50%) and fixed carbon (9%) with respect to fruit and protein-based wastes.

Hydrochar produced from fruit wastes resulted in an increase in fixed carbon as long as reaction temperature is raised, due to the conversion of glucose which starts at around 225 ◦C [1]. Different research groups showed that mass yield is mainly affected by reaction temperature, rather than residence time. For low reaction temperature, residence time can even have a positive effect on the final mass yield. Chen et al. [55], as well as Lucian et al. [51], reported an increase in mass yield when reaction temperature was set respectively at 190 and 180 ◦C for a reaction time of 6 h, when compared to mass yields obtained at 1-h reaction time. The increase in hydrochar mass is associated with a re-arrangement and back polymerization of organic compounds from liquid to solid phase [55].

**Figure 3.** van Krevelen diagram of hydrothermal carbonization (HTC) treatment of rotten apples, Penicillin mycelial and olive trimming.

As previously reported, the last parameter that can be varied when performing HTC treatment is the biomass-to-water ratio (B/W). Volpe et al. [17] found that when biomass is increased in the mix up to a B/W ratio equal to 0.2, the share of secondary char production is increased as well, boosting the HHV of the hydrochar. In terms of ash content, HTC presents a huge advantage over other pre-treatment process such as pyrolysis. In fact, although during "dry" pre-treatment all the inorganics are kept inside the biomass structure, increasing their content in the final product due to the mass loss during the heating, in HTC some of the ash forming elements are removed from the biomass and washed away with the process water [24].


240

180

200

220

180

200

220

180

200

220

180

200

220

190

225

260

 0.25

 0.25

 0.25

 37.9

 61.6

 66.8

 5.0

 26.8

 26.0

 39

 0.4

 35.9

 63.9

 64.2

 5.3

 29.4

 25.7

 36

 0.2

 34.2

 65.5

 62.4

 5.4

 30.8

 24.8

 36

 0.3 [1]

 2.0

 2.0

 2.0

 4.0

 4.0

 4.0

 31.9

 64.8

 66.6

 6. 9 21.8

 26.5

 70.3

 63.5

 7.2

 24.9

 28.0

 29.1

 51

 3.3

 60

 3.2

 23.7

 73.1

 59.9

 7.0

 29.1

 26.2

 65

 3.1

 28.7

 68.0

 66.5

 7.3

 21.9

 29.6

 52

 3.2

 25.1

 71.7

 62.3

 7.2

 26.2

 27.5

 62

 3.2

 22.4

 74.2

 60.2

 7.4

 28.3

 26.7

 66

 3.4 [4]

 2.0

 2.0

 2.0

 4.0

 4.0

 4.0

 29.6

 66.2

 67.1

 6.9

 21.1

 29.3

 55

 4.2

 23.0

 70.5

 61.2

 7.1

 27.2

 26.9

 63

 4.1

 26.7

 72.8

 62.9

 7.0

 25.2

 27.6

 67

 4.2

 23.7

 69.0

 65.8

 7.3

 22.1

 29.2

 58

 4.3

 20.8

 71.9

 62.0

 7.0

 26.4

 27.2

 64

 4.3

 17.8

 74.9

 60.3

 7.1

 27.8

 26.5

 68

 4.3 [4]

 0.5

 2.2

 16.8

 71.5

 9.9

 11.0

 7.7

 68

 81

**1.**EffectofHTCreactionparametersonhydrocharcomposition,calorificvalueandmassyield,togetherwithcoalpropertiesas

Brewer spent grains

(80% moisture) Brewer spent grains

(90% moisture)

Rotten apples

14.8 83.6 43.5 6.2 47.5

 17.5

16 76.2 51.2 6.9 36.5

 22.3

16 76.2 51.3 6.9 36.5

 22.3



12
