Revisiting the mechanism of β - O -4 bond cleavage during acidolysis of lignin part 10: reactions of C 6 –C 2 -type non-phenolic syringyl model compounds and comparison of the reactions with those of the guaiacyl analogues

A C 6 –C 2 -type non-phenolic syringyl lignin model compound ( I , 2-(2-methoxyphenoxy)-1-(3,4,5-trimethoxyphenyl)ethanol) as well as its derivatives was acidolyzed in aqueous 82vol% 1,4-dioxane containing 0.2 mol/L HBr, HCl, or H 2 SO 4 at 85ºC. The results were compared with those on the guaiacyl analogues ( V as well as its derivatives) obtained in our previous reports. The acidolysis of compound I was slower than that of compound V , which is in accordance with our previous result on the formation rates of the benzyl cation intermediates ( BC ) from a syringyl compound and its guaiacyl analogue. The enol ether-type compound was primarily produced in the acidolysis of compound I , similar to that of compound V . The acidolysis using HBr or HCl was faster than that using H 2 SO 4 , indicating the participation of Br¯ or Cl¯, respectively, in the acidolysis. It was suggested that Br¯ (as well as Cl¯) adds to the cation center of the BC to afford the adduct in the acidolysis of compound I , similar to that of compound V , and hence, this adduct formation is a major bypass of the common route. The bypass activity in the acidolysis of compound I was about half in that of compound V .


Materials
All chemicals other than the synthesized or isolated compounds described below were purchased from FUJIFILM Wako Pure Chemical Industries Co. (Osaka, Japan), Tokyo Chemical Industry Co., Ltd.(Tokyo, Japan), or Sigma-Aldrich Japan K. K. (Tokyo, Japan), and used without further purification except for 4-chlorophenol, which was employed as the internal standard compound (IS) for quantification and purified by recrystallization from ethanol.Water (H 2 O) and 1,4-dioxane were deionized and distilled, respectively, before use followed by degas using sonication under reduced pressure.Hydrochloric acid and sulfuric acid (H 2 SO 4 ) solutions with certified concentrations were purchased.A purchased hydrobromic acid solution was titrated with a purchased sodium hydroxide solution with a certified concentration before use.

Acidolytic treatment
The detailed method was described in our previous report [23], although 4-chlorophenol, the internal standard compound (IS), was always added in reaction solutions before the reactions were started.Typically, acidolysis reactions were conducted in an aqueous 82vol% 1,4-dioxane solution (30 mL) that was prepared to contain a starting compound (I, II (IIc/IIt = 52/48), or III), acid (hydrogen bromide (HBr), hydrogen chloride (HCl), or H 2 SO 4 ), and IS with concentrations of 1.0 mmol/L, 0.20 mol/L, and 2.0 mmol/L, respectively, at 85 °C for specified periods.
To a chloroform-d solution of compound I was added bromotrimethylsilane to prepare the solution containing compound IV.An amount of the chloroform-d solution was added to the above-described acidolytic solution containing HBr or H 2 SO 4 at 85 °C to initiate the reaction, and then the mixture was stirred at the same temperature for 10 s or 3 min.The detailed procedure was described in our previous report [23].The concentration of compound IV in the acidolysis solution was not exactly specified, although compound IV contained in the chloroform-d solution seemed rather pure on the basis of the NMR spectrum.

Quantification
The detailed procedures for quantification were described in our previous report [23].In the acidolyses of compound IV, the procedure was slightly modified as follows: a portion of the reaction solution withdrawn was primarily added to methanol (CH 3 OH) containing sodium methoxide (NaOCH 3 ) in an ice-water bath, and stood for 10 min with stirring.

Prior information for the following contents
Figure 2 illustrates the rationally presumed acidolysis reactions of compounds I and III (syringyl type) under the employed acidolytic conditions on the basis of the knowledge on the acidolyses of compounds V, VI, VII, and VIII (guaiacyl type) obtained in our previous reports [17,18,22,23,25] [18,22].A H 2 O molecule adds to the cation center of Cβ + II to afford the conjugate acid of the hemiacetal compound that readily undergoes the β-O-4 bond cleavage to afford compounds IX and X.
Our previous reports showed the following important results in the acidolyses of compound V [18,22]: (i) the rate-determining step with respect to the disappearance of compound V is the abstraction of either H + from the β-carbon of BC V as well as minor transfer of either H¯ from the β-to α-carbon of BC V ; (ii) pre-equilibria are established between compound V and CA V and between CA V and BC V prior to the rate-determining step.Thus, the acidolysis reaction of a β-O-4-type substructure can be divided into the former and latter halves: the elementary reactions from the substructure to BC; those from BC to the final products including the rate-determining step and β-O-4 bond cleavage.The former half affects the whole reaction rate through regulating the concentration of BC, although it does not include the rate-determining step but consists of only the elementary reactions prior to the step.The fast formation of BC generally results in its high concentration in the pre-equilibria, and hence, being accompanied by the fast rate-determining step.
It was examined in our previous report how the types of aromatic nucleus affect the rate of the former half, viz. the formation of BC, using simple 6 C 6 -C 1 -type model compounds consisting of phenolic and non-phenolic p-hydroxyphenyl, guaiacyl, and syringyl nuclei [31].The obtained results were as follows: (i) the BC formation from the phenolic compound is much faster than that from the corresponding non-phenolic compound regardless of the types of aromatic nucleus; (ii) the BC formation rates are in the order: p-hydroxyphenyl > guaiacyl > syringyl, regardless of being phenolic or non-phenolic; (iii) the BC formation from the phenolic syringyl compound, which is slowest among those from all the  2 Acidolysis reaction of compound I or III rationally presumed on the basis of the knowledge obtained in earlier and our previous reports [2,[7][8][9][10][11][12][13][14][15][16][17][18][19][20][21][22][23], including the characteristic action of HBr as a bypass proposed in our previous reports [23].All the reaction modes of BC I or BC III , the H 2 O addition and H + abstraction as well as the bypass with minor H¯ transfer, are surrounded by solid as well as dotted squares, respectively phenolic compounds, is still faster than that from the non-phenolic p-hydroxyphenyl compound, which is fastest among those from all the non-phenolic compounds.These results suggest that the equilibrated concentration of BC I , which is generated from compound I, the nonphenolic syringyl type, is lower than that of BC V , which is generated from compound V, the non-phenolic guaiacyl type.It can thus be predicted that the disappearance and β-O-4 bond cleavage of compound I are slower than those of compound V.However, because this prediction is based only on the difference in the rates of the former halves between the acidolyses of compounds I and V, the purpose of this study is to quantitatively confirm whether the entire acidolysis reaction of compound I is actually slower than that of compound V.
It has been shown in our previous report that the three aqueous 82vol% 1,4-dioxane solutions containing HBr, HCl, and H 2 SO 4 at a concentration of 0.2 mol/L show their proton activities close to one another [25], using the method employed in another of our previous reports [32].A different method has been employed for measuring their proton activities to suggest the similarity in another of our previous reports [33].Therefore, differences in the rates between acidolyses using the three acids observed in this study do not originate from those between their proton activities but from nucleophilic participation of the counter anions of HBr and HCl, bromide anion (Br¯) and chloride anion (Cl¯), respectively, owing to their high nucleophilicities.In contrast, the counter anion of H 2 SO 4 , hydrogensulfate anion (HSO 4 ¯), is an inert nucleophile in the aqueous solutions employed in this study, and hence, does not participate at all.All the counter anions hardly abstract any H + in the aqueous solutions, because they are bases weaker than the solvents, whose amounts are extremely larger than those of the counter anions.Thus, it is not rational at all to consider that the counter anions directly regulate the rate of the rate-determining step, the H + abstraction from BC I or BC V to generate compound II or VI, respectively.Because proton activity exclusively regulates the rates in acidolyses using H 2 SO 4 without any participation of the counter anion, HSO 4 ¯, the acidolyses are often treated as the reference in discussion on those using HBr and HCl in the following content.
Starting substrates, compounds I, II, and III, and a major reaction product, compound IX, were quantified in this study, although the other starting substrate, compound IV, was not quantified.Compounds X, XI, and XII were not quantified but identified as reaction products.These four compounds were not quantified because of the following facts: (i) compound IV is not stable enough in the post-treatment to be quantitatively analyzed; (ii) isolation of compound X must be accompanied by the difficulty, which is presumed on the basis of that found in the isolation processes of the guaiacyl analogue of compound X, (3,4-dimethoxyphenyl)acetaldehyde (XIII, Fig. 1), in previous reports [16,17]; (iii) the acidolysis rate of compound I, II, or III is independent of the types of final products, because it is confirmed in the following content that elementary reactions after the ratedetermining step determine the types.
Fair amounts of compound XI were produced in the acidolyses of compounds I and II using HBr as well as H 2 SO 4 at least on the basis of the peak areas in the HPLC analyses despite its large absorptivity.Instead, the peaks of compound X seemed rather small.Thus, compound XI may not have formed via the isomerization of compound X but an unknown mechanism, such as the migration of the syringyl nucleus from the α-to β-carbon of Cβ + II .This will be examined in our future work.Table 1 lists the pseudo-first-order reaction rate constants (k obs ) with respect to the disappearances of compounds I, II (total of IIc and IIt), and III when these compounds were individually acidolyzed in this study, squared correlation coefficients (R 2 ) in all their acidolytic runs, and k obs 's with respect to those of compounds V, VI, and VII when they had individually been acidolyzed in our previous report under the same conditions as this study [23].All the k obs values obtained in the acidolyses using HBr and HCl conducted in this study were based on all data points observed before the starting compounds reached a recovery yield of 20 mol%.

Acidolysis of compound I
Figure 3 shows the time-course changes in the recovery yields of compound I and yields of compounds II (sum of IIc and IIt) and IX as well as IIc and IIt in the acidolyses of compound I using HBr, HCl, and H 2 SO 4 .The yield of compound II in the acidolysis using H 2 SO 4 was quite low and did not exceed 0.5% throughout the entire reaction.The sum of the recovery yield of compound I and yields of compounds II and IX was almost 100% throughout the entire reaction in the acidolysis using HBr or HCl.It gradually decreased in the acidolysis using H 2 SO 4 and became lower than 90% at the end (48 h) when the recovery yield of compound I was still about 80%.Thus, some amounts of unknown substances including condensation products were generated in the acidolysis using H 2 SO 4 .The HPLC chromatogram showed peaks at retention times when condensation products should appear.
The disappearance rates of compound I in the acidolyses using the three acids were in the order: HBr > HCl > H 2 SO 4 .This order is the same as that in the acidolyses of compound V using these acids observed in our previous report [23] and suggests that Br¯ or Cl¯ participates in the acidolysis of compound I, similarly to the report.For visualizing the participation activity, the ratio of the k obs values between the acidolyses using HBr and H 2 SO 4 (k I-HBr /k I-H 2 SO 4 ) or between those using HCl and H 2 SO 4 (k I-HCl /k I-H 2 SO 4 ) is calculated to be 9.1 or 3.9, respectively.The participation of Br¯ is thus greater than that of Cl¯.In the acidolyses of compound V, the corresponding ratio is 41. 4 ), respectively.The participation is thus less active in the acidolysis of compound I than in that of compound V.The participation mode is discussed in a later section.
The disappearance of compound I was slower than that of compound V in its acidolysis regardless of the acid types.The slower disappearance was accompanied by the slower β-O-4 bond cleavage, viz. the slower formation of compound IX.This is in accordance with our prediction described in the previous section.For visualizing how much slower the acidolysis of compound I was than that of compound V, the ratio of the k obs values between compounds I and V is calculated to be 1/15.2,1/17.4,or 1/3.4 in the acidolyses using HBr, HCl, or H 2 SO 4 , respectively (k I-HBr /k V-HBr , k I-HCl /k V-HCl , or k I-H 2 SO 4 /k V-H 2 SO 4 , respectively).These ratios indicate that the acidolysis of compound I is essentially 3.4 times slower than that of compound V without any participation and the participation of Br¯ or Cl¯ is less active in the acidolysis of compound I than in that of compound V, as already suggested in the former paragraph.The less active participation is discussed in detail later.
Compound II gradually accumulated, showed the maximum amount, and then decreased regardless of the acid types, although even the maximum yields were only 7%, 4%, and lower than 0.5% in the acidolyses using HBr, HCl, and H 2 SO 4 , respectively.This is the same tendency as the yields of compound VI in the acidolyses of compound V conducted in our previous report, although the maximum yields, 48%, 33%, and 17%, respectively, were much higher [22].These phenomena are discussed in the next section.
Compound IIc formed more preferably than compound IIt at the initial stages in the acidolyses using HBr and HCl, followed by the gradual decreases in the differences between their amounts.The acidolysis using H 2 SO 4 is not covered in this paragraph, because compound II was detected with a yield of no more than 0.5% at any data point.The ratio of the detected amounts between compounds IIc and IIt (IIc/IIt) is calculated to be 1.9 or 2.4 in the acidolysis of compound I using HBr or HCl, respectively, at the first data point (1 h).It gradually decreases during either acidolysis, and reaches 1.3 or 1.5, respectively, at the end (24 h).These ratios show that the formation of compound IIc is kinetically more favorable than that of compound IIt.This is the same tendency as our previous reports which showed that the formation of compound VIc is kinetically more favorable than that of compound VIt in the acidolyses of compound V using the acids [18,22,23].The IIc/IIt ratio in the acidolysis using HBr, 1.9, is smaller than the real kinetic selectivity that should be equal to or slightly larger than 2.4, owing to the fast acidolysis in using HBr.The IIc/IIt ratio would further decrease in an extended acidolysis and reach a bottom that is independent of the acid types and smaller than unity due to the thermodynamic stability of compound IIc lower than that of IIt.Compound IIt should form more preferably than compound IIc when compound II exclusively forms from BC I , which carries two conformers suitable for the formation of compound II.This preferable formation of compound IIt is because it forms from the sterically smaller and thermodynamically more stable conformer.Thus, the kinetically more preferable formation of compound IIc suggests that compound II additionally or exclusively forms from a

Acidolysis of compound II with the results obtained in the previous section
Figure 4 shows the time-course changes in the recovery yields of compound II as well as IIc and IIt and yields of compound IX in the acidolyses of compound II using HBr, HCl, and H 2 SO 4 .The sum of the recovery yield of compound II and yield of compound IX slightly decreased to about 85 − 90% at the end of the reaction (6 h) regardless of the acid types, suggesting the formation of unknown substances including condensation products.Compound I was not detected at all during the entire reactions regardless of the acid types.This confirms that protonation at the double bond of compound II does not generate BC I at all but exclusively Cβ + II .This is the same observation as our previous report in which protonation at the double bond of compound VI does not afford compound V at all but exclusively Cβ + VI [17,18,22,23].The rate-determining step with respect to the disappearance of compound II must be the primary protonation because of the following two confirmations in our previous reports [18,22] does not frequently progress to the reverse deprotonation but further H 2 O addition followed by the β-O-4 bond cleavage; (ii) The relatively fast interconversion between compounds VIc and VIt is a process isolated from the protonation to afford Cβ + VI .The disappearance rates of compound II in the acidolyses using the three acids were in the order: HBr > HCl > H 2 SO 4 .This order is the same as that observed in the acidolyses of compound VI as shown in our previous report [23].These orders suggest that Br¯ or Cl¯ also participates in the acidolysis of compound II or VI.The k II-HBr /k II-H 2 SO 4 or k II-HCl /k II-H 2 SO 4 ratio is calculated to be 2.8 or 2.0, respectively, suggesting that the participation of Br¯ is more active than that of Cl¯.The k VI-HBr /k VI-H 2 SO 4 or k VI-HCl /k VI-H 2 SO 4 ratio is 2.9 or 2.1, respectively, which is close to the k II-HBr /k II-H 2 SO 4 or k II- HCl /k II-H 2 SO 4 ratio, respectively.Thus, Br¯ or Cl¯ participates in the acidolysis of compound II with an activity similar to that in the acidolysis of compound VI.This is contrast to the acidolysis of compound I, in which the participation is less active than that in the acidolysis of compound V.A possible participation mode in the acidolysis of compound II or VI is that protonation at the double bond is concertedly assisted by the anti-attack of Br¯ or Cl¯ to the β-carbon without as well as with affording the discrete Br¯ or Cl¯ adduct.
The disappearance of compound II was slightly slower than that of compound VI in its acidolysis regardless of the acid types.For visualization, the ratio of the k obs values between the acidolyses of compounds II and VI using HBr (k II-HBr /k VI-HBr ), HCl, (k II- HCl /k VI-HCl ), or H 2 SO 4 ( k II-H 2 SO 4 /k VI-H 2 SO 4 ) is calculated to be 1/1.2,1/1.2, or 1/1.1, respectively.This slightly slower disappearance should be owing to the electronwithdrawal effect of the excess methoxy group in the syringyl nucleus of compound II substituting at the meta-position of the side chain (the Hammett substituent σ value: + 0.115).When this relatively small effect is considered together with the above-described exclusive formation of Cβ + II or Cβ + VI in protonation at the double bond of compound II or VI, respectively, the double bond is considered to mesomerize with the lone pairs of the phenoxy oxygen more greatly than with the π-electron system of the 3,4,5-trimethoxyphenyl nucleus in compound II or 3,4-dimethoxyphenyl nucleus in compound VI.The relatively small effect should also be owing to another factor that the stability of Cβ + II or Cβ + VI is not largely affected by the types of aromatic nucleus in the backbone due to the absence of the delocalization of the cation to the aromatic nucleus.The three ratios close to one another suggest that the participation activity of Br¯ or Cl¯ is not different between the acidolyses of compounds II and VI, as suggested in the previous paragraph.This is contrast to largely different participation activities in the acidolyses between compounds I and V (k I-HBr /k V-HBr : 1/15.2, k I- HCl /k V-HCl : 1/17.4,k I-H 2 SO 4 /k V-H 2 SO 4 : 1/3.4).Because the k II-H 2 SO 4 /k VI-H 2 SO 4 ratio (1/1.1) is smaller than unity and larger than the k I-H 2 SO 4 /k V-H 2 SO 4 ratio (1/3.4), the type of aromatic nucleus in the backbone can essentially have an effect on the acidolysis of compound II or VI as well as its analogues less great than that of compound I or V as well as its analogues.
The disappearance of compound II was much faster than that of compound I in its acidolysis regardless of the acid types, which is visualized by calculating the k II-HBr /k I-HBr , k II-HCl /k I-HCl , or k II-H 2 SO 4 /k I-H 2 SO 4 ratio to be 11.2, 18.9, or 36.9, respectively.The ratios also show that the participation of Br¯ or Cl¯ accelerates the disappearance of compound II with an activity less great than that in the disappearance of compound I in its acidolysis, as described above.This ratio is much larger than the k VI-HBr /k V-HBr , k VI-HCl /k V-HCl , or k VI-H 2 SO 4 /k V-H 2 SO 4 ratio (0.9, 1.3, or 12.4, respectively), respectively, regardless of the acid types.Thus, the difference in the disappearance rates between compounds II and I is much larger than that between compounds VI and V.This larger difference results in the observation described in the previous section that the accumulation of compound II in the acidolyses of compound I was smaller than that of compound VI in those of compound V.However, the former small accumulations may have also resulted from another possibility that compound II did not form as the primary product in the acidolyses of compound I, although our previous reports confirmed that almost all molecules of compound V primarily convert to compound VI [17,22,23].This possibility can be evaluated on the basis of the k obs values observed in the acidolyses of compounds I and II.In the acidolyses of compound I, the formation rate of compound II as a primary product is expressed by Eqs. ( 1) and ( 2) owing to the absence of the reconversion to compound I: where [I] t or [II] t shows the concentration of each compound at a reaction time of t, k I or k II is the k obs value of compound I or II, respectively, and k I(i) or k I(ii) corresponds to the rate constant for the reaction route of compound I via compound II or for the sum of all other routes, respectively.The solution of Eq. ( 1) draws a simulational curve for the formation of compound II in the acidolysis of compound I. Figure 5 shows the simulational curves obtained by assuming k I(i) /k I(ii) ratios of 1.0, 0.75, 0.5, and 0.25 in the acidolyses of compound I using HBr and HCl with the experimental curves.The simulational curves are clearly closest to the corresponding experimental curves in a k I(i) /k I(ii) ratio of 1.0 in both acidolyses.Thus, almost all molecules of compound I primarily convert to compound II in the acidolyses of compound I using HBr and HCl.This is in accordance with the result in our previous reports as described above [17,22,23].It is also suggested that the H¯ transfer converting BC I to Cβ + II is just a minor route for BC I in the acidolysis, similarly to the minority of the H¯ transfer converting BC V to Cβ + VI shown in our previous reports [17,18,22,23].The simulational curve even in a ratio of 0.25 for the acidolysis of compound I using H 2 SO 4 seemed to exceed the experimental curve (not shown in Fig. 5), possibly suggesting the existence of reaction routes in which compound II is not generated, although the reliability is quite low due to the tiny yield of compound II.
The ratio of the detected amounts between compounds IIc and IIt (IIc/IIt) decreases from 1.1 at a reaction time of 0 h to about 0.7 at the end of the reaction (6 h) in the acidolysis of compound II using HBr or HCl.This is the same tendency as the decrease of this ratio in the acidolyses of compound I using these acids, as described in the previous section.Thus, the decrease of the IIc/IIt ratio can be discussed similar to the content described in the previous section.The decrease is also the same tendency as our previous report which showed that the same type ratio, VIc/VIt, decreases during the acidolyses of compound V using these acids [23].The IIc/IIt ratio at the initial, 1.1, seems to be maintained during the entire reaction when H 2 SO 4 was used.

Acidolysis of compound III with the results obtained in the previous sections
Figure 6 shows the time-course changes in the recovery yields of compound III and yields of compounds I, II, and IX as well as IIc and IIt in the acidolyses of compound III.The yield of compound II (as well as IIc and IIt) in that using H 2 SO 4 is not shown, because it was absent at most data points and sometimes detected with tiny amounts (≤ 0.1%).The sum of the recovery yield of compound III and yields of compounds I, II, and IX was not largely deviated from 100% during the entire reaction in any acidolysis, suggesting that any acidolysis reaction was mostly quantitatively followed.The rate-determining step with respect to the disappearance of compound III is the CH 3 OH liberation from CA III to generated BC I , because compound III is not naturally reproduced from BC I .
Because BC I primarily forms from compound III, which is not reproduced from BC I , the acidolysis of compound III contributes to analyzing to which reaction mode BC I preponderantly progresses.This cannot be examined in the acidolysis of compound I due to the reconversion of once-produced BC I to compound I.These are reasons for employing the acidolysis of compound III in this study, although native lignin does not have any α-methoxy group.
Compound I was exclusively the major product regardless of the acid types, which shows that BC I progresses much more preponderantly to the H 2 O addition than to the other H + abstraction.All other minor routes are disregarded in the following discussion in this paragraph.The ratio of the initial formation rates between compounds I and II in the acidolysis of compound III can show how preponderant the H 2 O addition is as a reaction mode of BC I over the other H + abstraction.Their initial formation rates correspond to the slopes of the tangential lines to their formation curves drawn in Fig. 6  acidolysis of compound III using HBr is 12.9 or 2.0%/h, respectively, with a ratio of 6.5 (I/II), although the formation curves may not completely follow the experimental data points for the entire reaction.This ratio is 13.2 or almost infinity in the acidolysis using HCl or H 2 SO 4 , respectively.The infinity is because compounds II and IX were not detected at all during the entire reaction in the acidolysis of compound III using H 2 SO 4 except for the final data point.The preponderant degrees of the H 2 O addition are thus in the order of the acidolyses using: HBr < HCl < < H 2 SO 4 .The participation of Br¯ or Cl¯ leads BC I to the H + abstraction mode, which is discussed in detail in the next section.
The ratio of the initial formation rates is 5.6, 11.9, or almost infinity when the same type ratio is calculated between compounds V and VI (V/VI) in the acidolysis of compound VII using HBr, HCl, or H 2 SO 4 , respectively, on the basis of the tangential lines drawn in the figure in our previous report showing the formations of compound V and VI in the acidolyses of compound VII [23].The preponderance of the H 2 O addition is thus greater in BC I than in BC V in the acidolyses using HBr or HCl, whereas the H + abstraction is almost inactive as a reaction mode of both BC I and BC V in those using H 2 SO 4 .Because of this inactivity, the H¯ transfer or another unknown reaction may be a major route in the acidolysis of compound I or V using H 2 SO 4 .In fact, the acidolysis of compound I using H 2 SO 4 afforded compound II via BC I and the successive H + abstraction with the amount that seemed smaller than that even when only 25% of compound I was assumed to convert to compound II as the primary product, as described in the previous section, despite the quite low yield of compound II and hence its low reliability.
The disappearance rates of compound III in its acidolyses using the three acids, which correspond to the formation rates of BC I from it, were in the order: HBr > HCl ≈ H 2 SO 4 .For visualization, the k III- HBr /k III-H 2 SO 4 or k III-HCl /k III-H 2 SO 4 ratio is calculated to be 1.8 or 1.0, respectively, which confirms that Br¯ participates in the CH 3 OH liberation step, but the participation of Cl¯ is not visible.In the acidolyses of compound VII, the k VII-HBr /k VII-H 2 SO 4 or k VII-HCl /k VII-H 2 SO 4 ratio is 1.9 or 1.1, respectively, which shows that the participation activity of Br¯ or Cl¯ is not largely different in the CH 3 OH liberations between the acidolyses of compounds III and VII.A possible participation mode of Br¯ is an assist in the CH 3 OH liberation from CA III or CA VII , following the S N 2-type mechanism, with or without the discrete formation of the Br¯ adduct.This participation should also contribute to the H 2 O liberation from CA I or CA V in the acidolysis of compound I or V, respectively, with a similar activity, so that the k III-HBr /k III-H 2 SO 4 (or k III- HCl /k III-H 2 SO 4 ) ratio can be applied to the ratio of the presumptive rate constants with respect to the former halves (H 2 O liberation from CA I ) in the acidolyses of compound I between HBr and H 2 SO 4 .The k VII-HBr /k VII-H 2 SO 4 (or k VII-HCl /k VII-H 2 SO 4 ) ratio can similarly be applied to  6 Time-course changes in the recovery yield of compound III (green solid line with green contouring white marks) and yields of compound I (black solid line with black contouring white marks), II (gray solid line with gray contouring white marks), and IX (red solid line with red contouring white marks) as well as IIc (gray long broken line with gray marks) and IIt (gray dotted line with gray marks) in the acidolysis of compound III using HBr (left), HCl (middle), or H 2 SO 4 (right).Circles and diamonds represent the 1st trial.Stars and squares represent the 2nd trial.The vertical axes in the lower figures are magnified by a factor of 5 for clearly showing the yields of compound II.The black and gray thin solid lines are tangents at the origin the acidolyses of compound V.When these presumptive ratios are considered with the k I-HBr /k I-H 2 SO 4 (or k I- HCl /k I-H 2 SO 4 ) and k V-HCl /k V-H 2 SO 4 (or k V-HCl /k V-H 2 SO 4 ) ratios, which correspond to the latter halves of the acidolyses of compounds I and V, the participation of Br¯ (or Cl¯) is suggested to be less great in the former halves of the acidolyses of compounds I and V than in the latter halves.Because the participation in the CH 3 OH or H 2 O liberation must catalytically accelerate the formation of BC I or BC V , the equilibrated concentrations of BC I or BC V in the acidolyses of compound I or V, respectively, are not different between the acidolyses using the three acids.
The disappearance of compound III was faster than that of compound I in its acidolysis regardless of the acid types.For visualization, the k III-HBr /k I-HBr , k III-HCl /k I-HCl , or k III-H 2 SO 4 /k I-H 2 SO 4 ratio is calculated to be 3.0, 4.0, or 15.7, respectively.Because it was shown in our previous report that the formation of the BC from a non-phenolic lignin model compound carrying an α-hydroxy group is about 7 times as fast as that from its α-methyl-etherified derivative regardless of solvent types [34], about the sevenfold of the k III-HBr /k I-HBr , k III-HCl /k I-HCl , or k III-H 2 SO 4 /k I-H 2 SO 4 ratio (about 20, 30, or 110, respectively) can correspond to how faster the former half, the formation of BC I , is than the latter half, the rate-determining step, in the acidolysis of compound I.These ratios suggest, similarly to the previous paragraph, that Br¯ or Cl¯ participates in the latter half of the acidolysis of compound I more greatly than in the former half to decrease the ratio from the essential, about 110, to about 20 or 30, respectively.The participation in the latter half is further discussed in detail in the next section.
The disappearance of compound III was slower than that of compound VII in its acidolysis examined in our previous report [23] regardless of the acid types.This shows that the formation of BC I from compound III is slower than that of BC V from compound VII.This is the same tendency as our previous report in which it was shown that the BC formation from a non-phenolic C 6 -C 1 -type syringyl compound, 5-(2-hydroxyethoxymethyl)-1,2,3-trimethoxybenzene, is slower than that from the corresponding guaiacyl compound, 4-(2-hydroxyethoxymethyl)-1,2-dimethoxybenzene [31].This slower degree in the report can be estimated to be 48.5 times at 85 °C.The formation of BC I from compound III slower than that of BC V from compound VII also suggests that the formation of BC I from compound I, the former half of the acidolysis of compound I, is slower than that of BC V from compound V, the former half of the acidolysis of compound V.These observations are rationally explained by the electronwithdrawal effect of the additional methoxy group substituting at the meta-position of the side chain of compound III or I (Hammett substituent σ value: + 0.115) on the α-carbon, the cation center of BC I .For visualization, the k III-HBr /k VII-HBr , k III-HCl /k VII-HCl , or k III-H 2 SO 4 /k VII-H 2 SO 4 is calculated to be 1/22.7,1/24.4,or 1/20.9, respectively.These ratios can also be applied to that of the former halves, H 2 O liberation, between the acidolyses of compounds I and V, similarly to the content of the previous paragraph.These ratios are not largely different, which indicates that the participation activities of Br¯ as well as Cl¯ are not largely different in the acidolyses between compounds III and VII and also in the former halves of those between compounds I and V.No large difference between these ratios is contrast to the large differences between the k I-HBr /k V-HBr , k I-HCl /k V-HCl , and k I-H 2 SO 4 /k V-H 2 SO 4 ratios (1/15.2,1/17.4,and 1/3.4,respec- tively), which are with respect to the latter halves in the acidolyses of compounds I and V.These ratios suggest, similarly to the previous paragraph, that the participation of Br¯ (or Cl¯) in the former half is less great than in the latter half in the acidolyses of I or V.Because the k III- HBr /k VII-HBr , k III-HCl /k VII-HCl , or k III-H 2 SO 4 /k VII-H 2 SO 4 ratio is larger than the above-described estimated ratio, 1/48.5, for the C 6 -C 1 -type non-phenolic syringyl and guaiacyl compounds, the presence of the β-position, guaiacoxymethyl group, in compound III or I counteracts the electron-withdrawal effect from the syringyl nucleus.

Acidolysis of compound IV with the results in previous sections
Because the initial concentration of compound IV in the acidolysis solution was not completely clear due to the procedures described in the "Acidolytic treatment" section, ratios of detected amounts between compounds I and II (I/II) are targeted in this section and listed in Table 2.Because the acidolysis was conducted for the short period (3 min), once-generated compounds I and II remained unreacted.Compound IV seemed to still remain with roughly less than or more than the half of the initial amount in the acidolysis using HBr or H 2 SO 4 , respectively, at a reaction time of 10 s.Compound III was generated from remaining compound IV Table 2 Ratios of the detected amounts of compound I to II in the acidolyses of compound IV using HBr and H 2 SO 4 obtained in this report and those of compound V to VI in the acidolyses of compound VIII using HBr and H 2 SO 4 shown in our previous report [23]  in the post-treatment in the CH 3 OH solution containing NaOCH 3 at 0 °C, and detected in the HPLC analysis instead of compound IV.Because our previous report showed that compound VIII has already disappeared completely at a reaction time of 10 s [23], compound IV is more stable than compound VIII owing to the formation of BC I slower than that of BC V , as frequently discussed above.Although the post-treatment was conducted, compound III was not detected after the reaction for 3 min.
As shown in Fig. 7, compound IV undergoes two reaction modes: "the HBr elimination" to convert to compound II; "the Br¯ liberation" to be BC I .BC I further undergoes three reaction modes: "the H + abstraction" from the β-carbon to convert to compound II; "the H 2 O addition" to convert to compound I via CA I ; "the Br¯ addition" to reconvert to compound IV with repetitively undergoing "the HBr elimination" and "the Br¯ liberation", which is more preferable in the acidolysis using HBr than in that using H 2 SO 4 .The application of HBr directly promotes only "the Br¯ addition" among all the 5 modes when compared with that of H 2 SO 4 , but consequently and indirectly "the HBr elimination" rather than "the Br¯ liberation".The only direct promotion of "the Br¯ addition" in the acidolysis using HBr is because of the following facts: among the other 4 modes, "the Br¯ liberation" and "the H 2 O addition" are competitively suppressed by the application of HBr; the other "the HBr elimination" and "the H + abstraction" are sensitive to the basicity of the acidolysis system; the basicities are not different between the acidolyses using HBr and H 2 SO 4, because the counter anions, Br¯ and HSO 4 ¯, are weaker bases than the solvents, whose amounts are huge in the acidolysis solutions.
The ratio of the detected amounts between compounds I and II (I/II) was 5.7 or 31.9 in the acidolysis of compound IV using HBr or H 2 SO 4 , respectively (Table 2).These ratios indicate that compound IV undergoes both "the HBr elimination" and "the Br¯ liberation", because these ratios would be 0 or infinity when compound IV progresses only to either "the HBr elimination" or "the Br¯ liberation", respectively, on the basis of the discussion described in the previous paragraph.The ratio, 31.9, shows that the most preferable route for compound IV is exclusively "the Br¯ liberation" followed by "the H 2 O addition" to convert to compound I.The presence of the large amount of Br¯ in the acidolysis using HBr enhances "the Br¯ addition", resulting in the indirect activation of "the HBr elimination" to afford compound II and the relatively small ratio, 5.7 (Table 2).Because compound IV certainly progresses to "the HBr elimination", the route from BC I to compound II via compound IV in the acidolysis of compound I using HBr at least partly functions as a bypass of the direct route from BC I to compound II, the rate-determining step of the acidolysis.However, because the V/VI ratio is 4.5 or 10.Two reaction modes of compound IV, "the HBr elimination" and "the Br¯ liberation", and three reaction modes of BC I , "the H + abstraction", "the H 2 O addition", and "the Br¯ addition", in the acidolysis of compound IV the acidolysis of compound VIII using HBr or H 2 SO 4 , respectively (Table 2), as shown in our previous report [23], the bypass functions less actively in the acidolysis of compound I using HBr than in that of compound V using HBr.The bypass activities are further discussed in the next paragraphs.
As shown in the previous section, compound II was not detected at all mostly during the entire reaction in the acidolysis of compound III using H 2 SO 4 .This indicates that the progress of BC I to "the H + abstraction" is invisible due to its relatively extreme slowness.It is thus presumed that BC I generated from compound IV via "the Br¯ liberation" hardly progresses to "the H + abstraction" to be compound II also in the acidolysis of compound IV using H 2 SO 4 .This presumption suggests that the I/II ratio in this acidolysis, 31.9, mostly corresponds to the ratio of the rate constants between "the Br¯ liberation" (k Br¯) and "the HBr elimination" (k HBr ).On the basis of this suggestion, the acidolysis is schematically illustrated in the following formula, in which the "k H2O " shows the rate constant with respect to "the H 2 O addition".Equation (3) is established from the above described I/II ratio, 31.9.
In the acidolysis of compound IV using HBr, "the Br¯ addition" is activated, which is schematically illustrated in the following formula.The I/II ratio in the acidolysis, 5.7, is expressed by Eq. ( 4).Because a steady-state can be presumed for the formation of BC I , Eq. ( 5) with respect to the steady-state can be described.In these equations, k −Br¯, [BC I ], and [IV] show the rate constant with respect to "the Br¯ addition", concentration of BC I , and that of compound IV, respectively.3), (4), and (5) are arranged, Eq. ( 6) is obtained as the correlation between k H 2 O and k −Br¯.
When the same discussion is applied to the acidolysis of compound VIII using H 2 SO 4 and HBr, the correlations between the rate constants are shown as Eqs.( 7) and (8).
On the basis of Eqs.3), ( 6), (7), and ( 8), the following bypass activity can be presented when a specific amount of BC I or BC V is assumed to be the starting substrate in the acidolysis using HBr: 2.5% or 5.0%, respectively, converts to compound II or VI, respectively, via the bypass, while 97.5% or 95.0%, respectively, progresses to "the H 2 O addition" to afford compound I or V, respectively.Thus, the bypass activity in the acidolysis of compound I using HBr is the half of that in the acidolysis of compound V using HBr.
The relative bypass activity corresponds to the ratio of the rate constants in the latter halves between the acidolyses of compounds I and V using HBr without any influence from the former halves.On the other hand, the k I-HBr /k V-HBr ratio, 1/15.2, includes the effects of the former halves on the latter halves, viz. the different equilibrated concentrations between BC I or BC V .Therefore, in considering the relative bypass activity with the k I- HBr /k V-HBr ratio (1/15.2), the equilibrated concentration of BC I in the acidolysis of compound I can be estimated to be 1/7.6 times of that of BC V in the acidolysis of compound V.The k I-HCl /k V-HCl ratio, 1/17.4,possibly suggests that the bypass activity in the acidolysis of compound using HCl is 2.3 times lower than that in the acidolysis of compound V using HCl.Incidentally, the equilibrated concentration of BC I or BC V is independent of the acid types.
Because the progress of BC I or BC V to "the H + abstraction" is invisible in the acidolysis of compound I or V using H 2 SO 4 , respectively, as described in the previous section, the H¯ transfer or another unknown reaction may be a major mode for BC I or BC V .Because the k I-H 2 SO 4 /k V-H 2 SO 4 ratio, 1/3.4, is larger than the estimated ratio of the equilibrated concentrations between the acidolyses of compounds I and V (1/7.6), the H¯ transfer or another unknown mode in the acidolysis of compound I using H 2 SO 4 may possibly be 2.2 times faster than in that of compound I using H 2 SO 4 .
Compound IIc formed more preferably than compound IIt in the acidolysis of compound I using HBr or HCl, as shown and discussed in the "Acidolysis of compound I" section, although the more preferable formation of the latter should be observed when compound II forms directly from BC I via "the H + abstraction".Thus, the more preferable formation of compound IIc suggests, as also described above, that compound II forms mostly via the bypass in which "the HBr elimination" occurs from compound IV following the E2 mechanism, which always prefers the anti-orientation.Two staggered conformers among three with respect to the covalent bond between the α-and β-carbons are suitable for this antielimination, as illustrated in Fig. 8.One of these two is the precursor of compound IIc and thermodynamically more stable owing to a stereoelectronic effect, the gauche effect.This effect originates both from the delocalizations of two electron pairs: the pair present in the bonding orbital of the bond between the β-carbon and hydrogen present at the antiperiplanar of the bromine to the empty anti-bonding orbital of the bond between the α-carbon and bromine; that present in the bonding orbital of the bond between the α-carbon and hydrogen present at the antiperiplanar of the phenoxy group to the empty anti-bonding orbital of the bond between the β-carbon and phenoxy oxygen.Only the former delocalization exists in the other conformation that is the precursor of compound IIt.The existence of the above-described two delocalizations thermodynamically stabilizes the former conformation, although its steric factor is larger than the latter.

Conclusions
Compound I, a non-phenolic syringyl-type compound, as well as its derivatives was acidolyzed for quantitatively analysis.The results were compared with those obtained from the same acidolysis of compound V, the guaiacyl analogue of compound I, as well as its derivatives conducted in our previous reports [17][18][19][20][21][22][23][24][25].
The disappearance of compound I in its acidolysis was slower than that of compound V in its acidolysis regardless of the acid types.It was thus confirmed that syringyl β-O-4-type substructures are acidolyzed more slowly than the guaiacyl analogues.The disappearance rates of compound I were dependent on the acid types and in the order: HBr > HCl > H 2 SO 4 , which shows the participation of Br¯ or Cl¯ in the acidolysis, similarly to the acidolysis of compound V.However, the participation activity was less great than that in the acidolysis of compound V.
The disappearance of compound II in its acidolysis was slower than that of compound VI, the guaiacyl analogue of compound II, in its acidolysis.The difference in the rates between compounds II and VI was smaller than that between compounds I and V, which resulted in the accumulation of compound II in the acidolysis of compound I smaller than that of compound VI in the acidolysis of compound V.It was confirmed that compound II exclusively forms as the primary product in the acidolysis of compound I using HBr or HCl, similarly to the exclusive formation of compound VI as the primary product in the acidolysis of compound V using these acids shown in our previous report.
The disappearance of compound III in its acidolysis, viz. the formation rate of BC I from compound III, was slower than that of compound VII, the guaiacyl analogue of compound III, in its acidolysis, viz. the formation rate of BC V from compound VII.The H 2 O addition was the most preponderant reaction mode for BC I , similarly to BC V regardless of the acid types.Because BC I or BC V did not seem to progress to the route to afford compound II or VI, respectively, in the acidolysis of compound I or V using H 2 SO 4 , respectively, due to the extreme preponderance, the H¯ transfer or an unknown reaction may be a major route.The participation of Br¯ was confirmed in the formation of BC I from compound III, although that of Cl¯ was invisible.8 Three staggered conformations with respect to the covalent bond between the α-and β-carbons of compound IV, among which the leftmost and middle exclusively progress to "the HBr elimination" to convert to compounds IIc and IIt, respectively.Although the steric factor is larger in the leftmost conformation than the middle, the leftmost is thermodynamically more stable owing to a stereoelectronic effect, the gauche effect, described in the text The acidolysis of compound IV using HBr or H 2 SO 4 confirmed that Br¯ adds to the cation center of BC I to be compound IV that progresses to the HBr elimination to be compound II as the bypass of the direct route from BC I to compound II in the acidolysis of compound I using HBr, similarly to the acidolysis of compound VIII conducted in our previous report.The presence of the bypass accelerates the acidolysis of compound I using HBr as well as HCl relatively to that using H 2 SO 4 .It was also suggested that the bypass activity in the acidolysis of compound I is the half in that of compound V.
Acidosis of syringyl lignin has thus been confirmed to be fairly slower than that of guaiacyl lignin.This slower acidolysis should be considered when reaction conditions are specified in acidolysis of hardwood.

Fig. 1
Fig. 1 Chemical structure of model compounds used and discussed in this study.The square shows the carbon numbering used in this study

Fig. 3 Fig. 4
Fig.3Time-course changes in the recovery yield of compound I (black solid line with black contouring white marks) and yields of compounds II (gray solid line with gray contouring white marks) and IX (red solid line with red contouring white marks) as well as IIc (gray long broken line with gray marks) and IIt (gray dotted line with gray marks) in the acidolysis of compound I using HBr (left), HCl (middle), or H 2 SO 4 (right).Circles and diamonds represent the 1st trial.Stars and squares represent the 2nd trial.The vertical axes in the lower figures are magnified by a factor of 5 for clearly showing the yields of compound II

Fig. 5
Figure6shows the time-course changes in the recovery yields of compound III and yields of compounds I, II, and IX as well as IIc and IIt in the acidolyses of compound III.The yield of compound II (as well as IIc and IIt) in that using H 2 SO 4 is not shown, because it was absent at most data points and sometimes detected with tiny amounts (≤ 0.1%).The sum of the recovery yield of compound III and yields of compounds I, II, and IX was not largely deviated from 100% during the entire reaction in any acidolysis, suggesting that any acidolysis reaction was mostly quantitatively followed.The rate-determining step with respect to the disappearance of compound III is the CH 3 OH liberation from CA III to generated BC I , because compound III is not naturally reproduced from BC I .Because BC I primarily forms from compound III, which is not reproduced from BC I , the acidolysis of compound III contributes to analyzing to which reaction mode BC I preponderantly progresses.This cannot be examined in the acidolysis of compound I due to the reconversion of once-produced BC I to compound I.These are reasons for employing the acidolysis of compound III in this study, although native lignin does not have any α-methoxy group.Compound I was exclusively the major product regardless of the acid types, which shows that BC I progresses much more preponderantly to the H 2 O addition than to the other H + abstraction.All other minor routes are disregarded in the following discussion in this paragraph.The ratio of the initial formation rates between compounds I and II in the acidolysis of compound III can show how preponderant the H 2 O addition is as a reaction mode of BC I over the other H + abstraction.Their initial formation rates correspond to the slopes of the tangential lines to their formation curves drawn in Fig.6at the origin.The initial formation rate of compound I or II in the Fig.6 Time-course changes in the recovery yield of compound III (green solid line with green contouring white marks) and yields of compound I (black solid line with black contouring white marks), II (gray solid line with gray contouring white marks), and IX (red solid line with red contouring white marks) as well as IIc (gray long broken line with gray marks) and IIt (gray dotted line with gray marks) in the acidolysis of compound III using HBr (left), HCl (middle), or H 2 SO 4 (right).Circles and diamonds represent the 1st trial.Stars and squares represent the 2nd trial.The vertical axes in the lower figures are magnified by a factor of 5 for clearly showing the yields of compound II.The black and gray thin solid lines are tangents at the origin

2 Fig. 7
Fig.7 Two reaction modes of compound IV, "the HBr elimination" and "the Br¯ liberation", and three reaction modes of BC I , "the H + abstraction", "the H 2 O addition", and "the Br¯ addition", in the acidolysis of compound IV of compound IV using HBr k ¯Br¯( 4) k H 2 O [BC I ]/k HBr [IV] = 5.7 (5) d[BC I ]/dt = k Br − [IV]−k H 2 O [BC I ]−k −Br − [BC I ] = 0 When Eqs. ( Fig.8 Three staggered conformations with respect to the covalent bond between the α-and β-carbons of compound IV, among which the leftmost and middle exclusively progress to "the HBr elimination" to convert to compounds IIc and IIt, respectively.Although the steric factor is larger in the leftmost conformation than the middle, the leftmost is thermodynamically more stable owing to a stereoelectronic effect, the gauche effect, described in the text

Table 1
[23]do-first-order reaction rate constants (k obs ) and squared correlation coefficients (R 2 ) in each run observed in the acidolyses of compounds I, II, and III conducted in this study, and those obtained for compounds V, VI, and VII in our previous reports[23] compound or intermediate other than BC I , which is discussed in a latter section.