The heat treatment is part of the barrel making process, being performed by the French technique using fire, or by the American technique using heated steam for bending the staves followed by fire [
161]. In European cooperage, the barrel is heated over a fire of wood shavings with various techniques of spraying or swabbing with water to enable the bending of the staves to the concave shape of a barrel without breaking—the bending phase [
110,
161]. Then, the barrel is placed again over the fire to heat the inner surface and to cause significant toasting in order to modify the structure [
114], the physical properties [
115], and the chemical composition of the wood [
109,
110,
121], which confer a distinct character to the wine or distillate aged in it—the toasting phase. Despite the diversity of toasting protocols, the toasting level is usually classified as light, medium or heavy. In practice, the result mostly depends on the binomial temperature/time applied to each wood botanical species [
109,
110,
143,
162,
163].
4.1. Phenolic Composition
Scientific data about the influence of the wood toasting level on the phenolic composition and related properties of the aged wine spirit are rather scarce. Notwithstanding, older works made on Spanish brandies by Artajona et al. [
96], on Cognacs by Cantagrel et al. [
164] and Viriot et al. [
101], and on French wine spirits by Rabier and Moutounet [
97] and Puech et al. [
165] are noteworthy. Artajona et al. [
96] found increasing contents of phenolic aldehydes in brandies with an increasing of barrel toasting intensity: ca 18 mg/L, 30 mg/L, and 58 mg/L under the influence of light, medium and heavy toasting, respectively. Rabier and Moutounet [
97] observed increasing contents of ellagic acid (ca 15 mg/L and 60 mg/L), gallic acid (ca 4 mg/L and 9 mg/L), and vanillin (ca 0.5 mg/L and 1 mg/L) in a wine spirit aged over a two-year period in new oak barrels with light and heavy toasting levels. Puech et al. [
165] also studied the influence of the toasting level (light, medium and heavy) in a wine spirit aged over a two-year period in new barrels of Limousin oak. They found an increasing content of vanillin with the rise of toasting intensity, which remained below 5 mg/L; a similar behavior was observed for syringaldehyde with ca.1 mg/L, 7 mg/L and 11 mg/L under the effect of light, medium and heavy toasting, respectively; for coniferaldehyde and sinapaldehyde, a sharp increase between light and medium toasting (from ca 3 to ca 13 mg/L, and from ca 2 to ca 22 mg/L, respectively) and a slight decrease under heavy toasting (ca 11 mg/L and 21 mg/L, respectively) were described.
In recent years, a comprehensive investigation was performed [
6,
56,
120]. In that study, the same wine distillate from
Lourinhã region (produced by Adega Cooperativa da Lourinhã) was aged over a four-year period in 250 L barrels. The barrels were made by J. M. Gonçalves cooperage (Palaçoulo, Portugal) using the following kinds of wood: Limousin oak (
Q. robur L.) and Allier oak (
Q. sessiliflora Salisb.) from French forests; American oak (mixture of
Q. alba L./
Q. Stellata Wangenh. and
Q. lyrata Walt./
Q. bicolor Willd.) from Pennsylvania/USA; Portuguese oak (
Q. pyrenaica Willd.) and chestnut (
C. sativa Mill.) from the North of Portugal. These barrels were divided into three groups. Then, each group were submitted to one of the three levels of toasting—light (LT), medium (MT) and heavy (HT)—according to the cooperage protocol: 10 min for light toasting, 20 min for medium toasting and 25 min for heavy toasting [
120]. They were filled with the same wine distillate and kept in the cellar of Adega Cooperativa da Lourinhã in similar environmental conditions.
Analysing the low molecular weight phenolic compounds of the wine spirits aged in them, Canas [
56] showed that the toasting level had a significant effect on the concentration of all phenolic compounds, except for scopoletin (
Table 5), confirming the results of previous studies [
96,
97,
164,
165].
Furthermore, Viriot et al. [
101] and Canas [
56] emphasized a positive relationship between ellagic acid concentration in the wine spirits and the toasting intensity of the barrel. A different pattern is identified for gallic acid; its concentration in the aged wine spirits increases under the influence of medium toasting and slightly decreases under heavy toasting. Recent results obtained for grape marc spirit corroborate it [
149]. This pattern expresses the behavior of gallic acid in the toasted wood, in which it undergoes degradation from the medium toasting as a consequence of higher thermal sensitivity [
97,
166]. In contrast, higher level of ellagic acid is ascribed to its high fusion point and greater accumulation in the toasted wood, as observed by Rabier and Moutounet [
97]. As in the untoasted wood [
108] and toasted wood [
109,
137], ellagic acid and gallic acid still remain the major phenolic acids of the aged wine spirits, being mainly derived from the wood ellagitannins and gallotannins [
118,
129,
130,
167].
It was also demonstrated that the rise of toasting level of the barrel promoted an increase of vanillic acid, syringic acid, ferulic acid and phenolic aldehydes contents in the aged wine spirits, as in the aforementioned studies, except for coniferaldehyde and sinapaldehyde [
165]. It is well-known these compounds resulted from the wood lignin’s decomposition [
117] (
Figure 5). Under mild temperatures, decarboxylation and cleavage of the aryl-alkyl ether bonds of the terminal units of this biopolymer take place, originating the cinnamic aldehydes (coniferaldehyde and sinapaldehyde). At higher temperatures, an oxidative cleavage of double C-C bond of the aliphatic chain of these aldehydes may occur, yielding the corresponding benzoic aldehydes (vanillin and syringaldehyde). The resulting concentrations express the balance between synthesis and degradation reactions. Therefore, the slight decrease of coniferaldehyde and sinapaldehyde contents reported by Puech et al. [
165] for heavy toasting should have resulted from specificity of the toasting protocol, which induced higher degradation of these aldehydes.As the temperature rises, the phenolic aldehydes thus formed give rise, by decarboxylation, to the corresponding phenolic acids. Hence, they accumulate in the toasted wood [
109,
130,
137,
163].
Furthermore, higher permeability of the wood and better access of the wine spirit to wood extraction sites caused by fragmentation of cell structures and reorganization of lignocellulose network [
114,
115] may also facilitate their release into the wine spirit. Likewise, lignin’shydrolysis during the ageing period may contribute to their increase in the beverage [
14,
29,
102,
116,
144]. The presence of oxygen and the mild acidity of the medium, mainly modulated by the increase of acetic acid content over time, favor this pathway [
37].
Regardless the toasting level of the barrel and the ageing time, it has been found [
29,
56,
102] that syringyl-type aldehydes (sinapaldehyde and syringaldehyde) prevailed over those of guaiacyl-type (vanillin and coniferaldehyde) in the aged wine spirits. On the other hand, an increase in the syringyl/guaiacyl ratio with the toasting intensity has been referred [
14,
56,
97]. In the above-mentioned work [
56], mean values of 1.34, 1.82 and 2.41 were obtained for the same wine spirit aged during four years in barrels with light, medium and heavy toasting levels, respectively. This suggests that higher thermal stability of the syringyl compounds and subsequent higher availability in the toasted wood [
117,
118] was the causal effect.
There is also evidence of the increase of umbelliferone content in the wine spirit with the toasting level of the barrel [
56], but the chemical mechanisms underpinning its formation/degradation during the heat treatment of the wood are still unknown.
Concerning the hydrolysable tannins, no significant differences in wine spirits aged in barrels with different toasting level were reported [
122].
4.2. Chromatic Characteristics and Sensory Properties
As for the wood botanical species, data from the literature [
23] show the modulating effect exerted by the toasting level of the barrel on the chromatic characteristics of this beverage (
Figure 6). The higher the toasting level of the barrel the more the evolution of the aged wine spirit color (higher intensity, saturation and red and yellow hues) with significant increments between levels. Acquisition of these chromatic characteristics makes the wine spirit aged in heavy toasting barrels look older than those aged in medium and light toasting barrels (as noticed for the wine spirit aged in chestnut wood when compared with those aged in different kinds of oak wood). These outcomes are correlated with the phenolic compounds extracted from the wood (
Table 5) and are in agreement with those obtained for wine aged in barrels with different toasting levels [
167]. In addition, the oxidative phenomena underlying the ageing process may also be responsible for the color acquired by the aged wine spirit; the higher the toasting level the higher the wood permeability to oxygen [
114,
115], and therefore greater extension of oxidation reactions are expected.
From the sensory point of view, Canas et al. [
23] indicated the predominance of yellow straw and yellow green in the wine spirits aged in light toasting barrels, and the prevalence of golden and topaz in those aged in medium and heavy toasting barrels, respectively (
Figure 7). These results are consistent with those obtained by the CIELab method (
Figure 6), showing a faster ageing of the wine spirit associated with the heavy and medium toasting levels.
Other sensory properties related to the phenolic composition, such as the vanilla aroma and astringency, had higher intensities associated with the heavy toasting barrels [
153]. The wine spirits aged in light toasting barrels and medium toasting barrels revealed opposite and intermediate intensities of these attributes, respectively. Increasing concentrations of vanillin with the toasting intensity (
Table 5) should explain the effect on the vanilla aroma. Regarding astringency, the existing information does not allow the establishment of a reliable relationship with the phenolic composition.